Report of the 115th ENMC workshop: DM2/PROMM and other myotonic dystrophies:
3rd Workshop, 14–16 February 2003, Naarden, The Netherlands
Article Outline
- 1. Introduction: ‘then and now for myotonic dystrophy type 2 (DM2/PROMM)’
- 2. DM2 mutation: diagnostic procedure and occurrence
- 3. The clinical spectrum of symptoms and signs in patients with DM2 mutation
- 4. Muscle biopsy findings: uncovered differences between DM2 and DM1
- 5. The brain in DM2
- 6. Molecular pathomechanisms of DM2
- 7. Splicing errors in DM2
- 8. Expression profiling results
- 9. Animal models
- 10. Practical management
- 11. PROMM-like disorders without DM1 or DM2 mutations
- 12. Conclusions
- Acknowledgements
- References
- Copyright
1. Introduction: ‘then and now for myotonic dystrophy type 2 (DM2/PROMM)’
Myotonic dystrophy (DM1, Steinert's disease) has been known for about 100 years. Its core features are myotonia, muscle weakness, and cataracts [1]. In 1994 another disorder was published with similar core features but lacking the DM1 mutation [2], [3]. The name given to this disorder was proximal myotonic myopathy (PROMM, OMIM *160900).
The first ENMC workshop on PROMM in 1997 established the clinical criteria for PROMM [4]. Prior to the second ENMC Workshop on PROMM in 2000, a major breakthrough had been accomplished when, in 1998, Ranum and Day [5] described the mapping of a new genetic locus in a large Minnesota family with myotonic dystrophy-like features on chromosome 3q21. This locus was named DM2. As a result, a new nomenclature emerged and the locus for myotonic dystrophy/Steinert's disease was renamed DM1 [6]. Subsequent work showed that most PROMM families also mapped to the DM2 locus, whereas another smaller group of PROMM families appeared not to be linked to this locus. One separately described family with marked proximal atrophies, called proximal myotonic dystrophy [7], also mapped to the DM2 locus [8], and the workshop adopted the term myotonic dystrophy type 2 (DM2) for all the progressive myotonic multiorgan disorders linked to the DM2 locus.
The clinical course of DM2 appeared to be more favourable compared to DM1. Families with DM2 did not have the severe congenital form of illness that occurs in DM1. Abnormalities in the social and cognitive abilities of adults with DM2 were typically mild or absent, and there was no prominent weakness of the facial and bulbar muscles. In DM2 the manual skills largely remained intact, and hypersomnia and mental retardation were not prominent findings. All these characteristic features of DM2 have not changed since the first ENMC Workshop[4]. They are still valid.
In the advent of this 3rd workshop on DM2/PROMM, the definitive milestone of research efforts was achieved in 2001 with the identification of the mutation underlying DM2 by Ranum and Day, in collaboration with Ricker [9]. The mutation is a huge (CCTG)n microsatellite repeat expansion in the first intron of the ZNF9 gene and is thus similar to the (CTG)n repeat causing DM1.
The aims with this 3rd ENMC Workshop on DM2/PROMM were:
2. DM2 mutation: diagnostic procedure and occurrence
Laura Ranum (USA) started the scientific sessions by relating the research history behind finding the DM2 mutation. They studied 23 Minnesota and 52 German families to identify shared markers and refine the DM2 locus. The STR marker CL3N58 on 3q21 showed the greatest degree of linkage disequilibrium. Furthermore, that marker had an unusual segregation pattern in that affected offspring appeared to have inherited only one allele, which always derived from the healthy parent. CL3N58 is in the first intron of the ZNF9gene and has a complex structure: (TG)14–25(TCTG)4–10(CCTG)11–26. When mutated the (CCTG)n portion of the repeat tract is expanded to repeat lengths between 75 and 11,000 repeats, and the mutant allele cannot be amplified by PCR. To date, in their lab, a total of 133 families with 379 affected individuals of northern European background have been shown to carry the DM2 mutation.
Establishing a molecular diagnostic protocol has been difficult. Conventional Southern-blot protocols, routinely used for DM1 mutation analysis, detected the DM2 mutation in only 80% of subjects with known expansions. For molecular diagnosis Ranum and colleagues therefore adopted a three-step procedure: (1) the allele size of the CL3N58 microsatellite serves as a good screening method as all individuals showing two alleles for the marker are excluded from having the DM2 mutation. However, identical allele size on two normal alleles occurs in 12% of the population. (2) All patients appearing to have one allele need further molecular analysis to determine whether or not they carry a DM2 expansion. Because of the incomplete sensitivity of Southern analysis, a DM2 repeat assay (RA) was developed (now commercially available through Athena®). (3) The RA method involves amplifying the CCTG repeat by PCR, and probing the resultant product with an internal probe to assure specificity. The combined use of these methods allows >99% sensitivity and specificity for known expansions. The mutation seems to be fully penetrant in adults who undergo neurological examination in the studied families.
Nuclear RNA foci, similar to those reported for DM1, were detected in DM2 cells by FISH studies, suggesting similar RNA pathophysiological mechanisms.
Ralf Krahe (USA) reported the DM2 mutation findings in 14 unrelated families of European descent from different countries (France, Finland, Italy, Spain, Switzerland, UK and USA). In his lab, a modified Southern-blot protocol has been developed using pulsed field/field inversion gel electrophoresis (FIGE). Using this method they were able to detect expansions ranging from 4 to 27 kb in the families examined. Families of diverse ethnic and geographic backgrounds were studied by extended haplotype analysis using both microsatellite markers and SNP markers. With the microsatellites, a limited number of conserved haplotypes, more or less specific for each country, were observed. Many countries, such as Finland, show only one extended haplotype in the families studied. With decreasing distance to the mutation allele sharing of markers across ethnic boundaries was observed. Further fine mapping with SNPs in and around ZNF9 and the DM2 mutation revealed extensive linkage disequilibrium (LD) and a single common haplotype for all families studied. This haplotype was also the most common haplotype in the normal population. The data suggested a single (or a few) founding mutation(s) for DM2 in Caucasians, a situation reminiscent of that in DM1. Krahe hypothesized, since the DM2 mutation, to date, has not been identified in sub-Saharan or east-Asian populations, that the (CCTG)n expansion mutation occurred after the migration out-of-Africa and the divergence of the European and Asian lineages approximately 35,000 years ago.
Wolfram Kress (Germany) reviewed the diagnostic procedures used in their lab for routine mutation detection. In addition to the PCR allele sizing of the CL3N58 marker, they also used the marker CL3N59 to establish linkage disequilibrium, (CA)26 and rarely (CA)27 on the A haplotype that is not found in normal population at all. The haplotype is specific but not very sensitive, as 20% of patients do not have it. Their lab used modified protocols for Southern and the PCR-based repeat assay, including long-range PCR with primers flanking the (CCTG)n repeat. Current tests show high sensitivity in well-studied families. In their experience the DM2 mutation occurs in about 50% of unselected cases with suspected DM2 disease referred for mutation detection.
The consortium agreed on the importance of validating the different diagnostic protocols against each other. Samples will be exchanged between labs, to establish sensitivity and specificity of the methods in use. Since mutation testing has proved to be complicated, it is recommended to keep the routine testing in highly specialized molecular labs.
Anna Vihola (Finland) described new approaches for DM2 mutation detection using chromogenic in situ hybridization techniques (CISH). Similar to DM1, the accumulation of mutant RNA transcript containing (CCUG)n mutation in nuclear foci was previously shown by fluorescent in situ hybridization (FISH) studies. Muscle samples from eight DM2 mutation-confirmed patients were probed with (CAGG)8 anti-sense and (CCTG)8 sense probes. Highly specific, prominent accumulation of RNA in nuclei, both on frozen and paraffin-embedded muscle tissues, was easily encountered. In addition, as a qualitatively new finding with the sense probe, single spot foci in the myonuclei identified directly the (CCTG)n mutation itself on the DNA strand in the 3q21. This labelling was not seen on DM1 or normal control muscle sections. The CISH technique can be applied in any muscle lab and may thus be suitable for routine detection of the DM2 mutation on muscle biopsies.
Riitta Sallinen (Finland) described a methodology using fibre-FISH to establish physical maps and for direct and indirect DM2 mutation detection. The resolution with the technique is 1–2 kb and thus potentially valuable for most DM2 mutations. The method was applied for sizing of the gap between BAC clone 814L21 containing ZNF9 and the DM2 repeat and the closest telomeric clone, for which no data exist in the databases. Detection and sizing by direct hybridization of the DM2 mutation on the DNA strand on stretched chromosome fibres also has the potential to give good estimates of the repeat expansion length. Co-hybridization experiments of (CCTG)8 oligonucleotides with 814L21 have so far been unsuccessful.
Kenneth Ricker (Germany) estimated the DM2 mutation frequency based on his own experience in Germany and the rest of the world, taking into account that PROMM/DM2 has been known for only 10 years. With incomplete ascertainment, more than 200 DM2 families are currently known, giving a prevalence of ~800 known patients in Germany alone. These figures indicate a minimum prevalence of ~1/100,000 for DM2. However, new patients are diagnosed all the time, so that the numbers will likely increase significantly. During the workshop various participants reported on the first DM2 mutation verified families from non-European populations: one each from Morocco, Algeria, Lebanon, Afghanistan and Sri Lanka.
3. The clinical spectrum of symptoms and signs in patients with DM2 mutation
John Day (USA) presented the findings in 23 families from Minnesota. He found clinical myotonia in 77%, proximal limb weakness in 64%, neck flexion in75%, thumb or deep finger flexor weakness in 55%, muscle pain in 56%, EMG myotonia in 90%, early balding in a minority, hyperhidrosis in 25%, manifest diabetes in 25%, and cataracts that were identical to those in DM1.
The presenting symptom of the disease was myotonia or proximal weakness (40% each) and less often stiffness and pain, or cataracts (12 and 8%, respectively). Ninety percent of patients had elevated CK, 62% elevated gamma-GT, 65% of males had elevated FSH and 75% of subjects had some degree of insulin insensitivity.
Extended EMG studies including single fibre EMG did not detect abnormalities sufficient to suggest a neurogenic etiology of the atrophic fibres seen in muscle biopsies.
Regarding the correlation between genotype and phenotype he found no significant correlation between age of onset and repeat size. There was apparent intergenerational instability with a tendency for contraction of the repeat expansion. The most consistent correlation with repeat size was the age of the patient when the blood sample was drawn, indicating a tendency for repeat length in blood to increase over time.
Ken Ricker and Christiane Schneider (Germany) had looked for catastrophic cardiac events in patients younger than 46 years of age from proven German DM2 families. In five families, there was a sudden death, usually preceded by complaints of syncope and/or ventricular insufficiency. Whether these are related to the DM2 mutation or other genetic background is unclear. In DM1 cardiomyopathy is unusual, whereas conduction abnormalities are more common than in DM2. ECG abnormalities were found in 20% of Minnesota patients (Day). Schneider also showed preliminary results of MR spectroscopy in a few DM2 patients demonstrating decreased pCreatine and ATP.
Giovanni Meola (Italy) reported on 30 patients from three large families with DM2-verified mutation. He found that the majority of patients (28 out of 30) presented with a mild muscle phenotype with MRC megascores almost normal, no grip myotonia, muscle pain, or limitations in everyday life. One patient with disease duration of 20 years, presented with a moderate phenotype. On the other end of the spectrum there was one patient with a severe phenotype: a 41-year-old-female with a progressive myopathy over the last few years and wheelchair-bound since age 37. One patient had mild ECG abnormalities (1st degree AV block). In another patient there was a severe echocardiographic cardiomyopathy, and significant clinical symptoms (palpitations, tachycardia), but she was a heavy alcohol drinker.
Guillaume Bassez (France) had studied 21 patients from 17 DM2 mutation verified families, six of them ethnically originating from France, four from Poland and the rest from Spain, Serbia, Czech, Malta, Algeria, Lebanon and Sri Lanka. Ten patients had cardiac problems, antiarrhythmic medication was prescribed in six patients and there was one instance of sudden death. Peculiar pyloric stenosis, leucoencephalopathy, axonopathy, severe atherosclerosis and immune deficiency were other observations in some of the families.
Joseph Gamez (Spain) described four DM2 mutation verified families of which one originated from Morocco. Severe seborrhoic dermatitis complicated the phenotype in one family.
Mark Rogers (UK) had carried out a study to determine the frequency of the DM2 mutation in 205 new patients referred for clinical genetic evaluation for myotonic dystrophy. One hundred and one had DM1; of the remaining 104 patients 40 were adult cases. In these the DM2 mutation was screened for and was found in three cases (7.5% in that specific subgroup). None of these had the full repertoire of DM2/PROMM phenotype.
Thomas Wieser (Germany) presented the clinical and genetic data from a family previously published in Neuromuscular Disorders as an DM2-unlinked family by genetic linkage analysis. However, the proband and the closest affected relatives had the DM2 mutation, whereas the other branch of the family had an apparently other cause for the segregation of cataracts. Similarly, several other families from Germany, Italy, Norway, and the USA had initially been identified as DM2-unlinked PROMM families, which, subsequently, upon molecular genetic examination for the mutation, proved to be DM2 families. This unusual discrepancy is related to the fact that the clinical phenotype is highly variable and in the mild end of the spectrum shows symptoms and signs that are frequent in the population for other reasons, which makes the clinical classification difficult.
Finally, Bjarne Udd (Finland) reported on work in progress; a new study based on 52 cases with undetermined myopathies enrolled from three Finnish centres over the last year. Based on PCR allele sizing and CISH studies on muscle biopsy, 16 of these seem to be DM2 patients. Of the 16, only three had a typical PROMM phenotype, however, without family history. Thirteen patients had no clear PROMM phenotype, just non-specific irritative spontaneous activity on EMG, muscle pain, stiffness, cramps and tremors. Preliminary data suggest that the classical or typical PROMM could constitute just the phenotypic ‘tip of the iceberg’ and that many other milder phenotypes may be caused by the same DM2 mutation.
4. Muscle biopsy findings: uncovered differences between DM2 and DM1
A study on DM2 mutation verified patients, performed within the framework of the ENMC consortium as a collaboration between the corresponding groups, was presented. The patient material included three Finnish patients with biopsies taken from vastus lateralis, three French patients with deltoid muscle biopsy and three Italian patients with biopsy specimens from the biceps brachii muscle. Besides routine histochemical stainings, immunohistochemistry with slow and fast MHC antibodies was applied for fibre typing and morphometric analysis. Anna Vihola (Finland) showed the results of myosin immunohistochemistry identifying a type 2 fibre specific atrophy in DM2. Nuclear clump fibres, all of which were type 2 fibres, appear early at the beginning of clinical symptoms. The identification of large numbers of extremely small 5–10-μm type 2 fibres that are not well detected on conventional ATPase staining has largely removed the previously proposed concept of type1 fibre predominance in DM2. However, in some cases of deltoid muscle biopsies, there is still a type 1 fibre predominance after correction with myosin fibre typing. Guillaume Bassez reported on the general histopathological findings showing frequent small group of angulated fibres (5/10) and the occurrence of rimmed vacuolated fibres (3/10). Moth eaten fibres but no sarcoplasmic masses, commonly encountered in DM1, were observed. In addition to the collaborative study the French group conducted further analyses on the internal nuclei. They did not differ between DM1 and DM2 when analysing the total number of fibres. However, the internal myonuclei were significantly more prevalent in type 2 fibres in DM2, whereas they occurred mainly in type 1 fibres in DM1. Giovanni Meola from the Italian part reported on the morphometrical data from the study showing a remarkable fibre type 2 atrophy in a subset of type 2 fibres with other type 2 fibres showing hypertrophy. The identification of highly atrophic type 2 fibres and nuclear clump fibres already in muscles with even normal strength can be used as a screening tool to select patients for molecular diagnostic testing. The paper reporting the collaborative study results was in press in Neurology at the time of the workshop. The reason for rather specific type 2 fibre atrophy in DM2 is not known, but the clear difference to DM1 muscle morphology indicates that the findings on muscle biopsy and the differences between DM2 and DM1 regarding distribution of muscle weakness may have a common denominator.
Benedikt Schoser (Germany) presented a large collection of 60 DM2 verified patients with biopsies from deltoid, biceps, vastus lateralis and tibialis anterior muscles. Nineteen of the samples had been analysed with myosin immunohistochemistry. The findings confirmed the type 2 fibre atrophy and nuclear clump fibre pathology as the major differential finding compared to DM1. An increased number of internal nuclei was found in all samples, angulated atrophic fibres in 96% of the samples were common features whereas targetoid fibres or fibre type grouping as a result of ongoing neurogenic changes were lacking. In the end stage fibrosis and fatty infiltration will occur.
Carmen Navarro (Spain) reported on a series of four DM2 verified patients, and presented the results of electron microscopic studies showing no nuclear inclusions corresponding to the nuclear RNA aggregates shown to accumulate in DM2 in the same way as in DM1.
5. The brain in DM2
Rolf Schröder (Germany) presented a MRI study comparing 10 DM1 patients with seven DM2 patients. All seven DM2 patients had some abnormality; four had considerable white matter lesions more in the posterior regions and five had general cerebral atrophy, although less pronounced in comparison to DM1 patients.
Giovanni Meola (Italy) presented cognitive and functional data in 19 patients (10 females, 9 males) with an age range of 37–66 years. The evaluation of the cognitive and behavioural profile consisted of neuropsychiatric assessment, neuropsychological tests and functional data by SPECT. A specific ’avoidant’ personality, and a significant impairment on frontal lobe functions, which correlated with a decreased cerebral blood flow in frontal and temporal lobes, were the main findings. A follow-up study 3 years later included four patients who repeated the same battery of tests with the same examiner, indicated that there was no difference in those tests results. Longitudinal studies on a larger cohort of patients and over a longer time space are necessary.
6. Molecular pathomechanisms of DM2
Laura Ranum described the mutation in DM2 as a repeat expansion of a complex short tandem repeat microsatellite on chromosome 3q21, in the first intron of the ZNF9 gene. The normal repeat is complex with a structure of (TG)14–25(TCTG)4–10(CCTG)11–26, where the last part (CCTG)n can be interrupted by GCTG/TCTG and/or TCTG/TCTA tetranucleotides. These interruptions were found on normal alleles, but not on several DM2 CCTG repeat expansions that were sequenced. The DM2 mutation has dramatic repeat instability: mutation lengths of 75 repeats to 11,000 repeats have been identified in DNA from blood of affected individuals; two samples taken 3 years apart from one individual showed an increase in DM2 expansion length of 500 repeats; within single samples from many affected individuals there is marked expansion size heterogeneity. twenty-five parent/offspring pairs showed smaller alleles in the offspring. CCUG repeat containing RNA foci that are found in the nuclei of DM2 cells are comparable to the CUG containing nuclear RNA foci seen in DM1. The DM2 mutation does not appear to interfere with transcription or splicing of the ZNF9gene. The normal role of repeat tract is unknown but the observation that portions of the repeat tract are highly conserved suggests a possible functional role for the repeat.
Charles Thornton (USA) started his talk about the pathogenic RNA-transcript model in DM2 (and DM1) by elucidating the emerging new roles for RNA as having structural, catalytic, sensor, and now pathogenic functions. Focal accumulations of expanded (CCUG)n RNAs are visualized in the DM2 muscle nucleus using in situ hybridization techniques. The term ‘ribonuclear inclusions’ was suggested for these nuclear structures, which are distinct from the intranuclear inclusions seen in (CAG)n repeat diseases. The pathogenic impact of the inclusion is not well understood but one possibility is that they sequester RNA binding proteins. The accumulation itself or fractions of the transcript that are not present in the inclusions may have harmful effects. One toxic effect currently known is the abnormal splicing of pre-mRNAs from a selected group of target genes. The occurrence of aberrant splice variants of the ClC-1 chloride channel, the insulin receptor, tau protein, cardiac troponin, etc., may be responsible for the multiorgan effects, such as myotonia and insulin insensitivity. Proteins binding to the inclusions in the nuclei are not specifically known, but previous studies have shown that CUGBP1 and ETR3 (CUGBP2) bind short (CUG)n and MBNL proteins bind long (CUG)n RNAs in vitro. When localization of these proteins were studied, only MBNL showed strong co-localization with the ribonuclear inclusions in the myonuclei. Ribonuclear inclusions are not restricted to skeletal muscle. In brain they were also encountered in cortical neurons, dentate gyrus, thalamus, medulla, substantia nigra, and Purkinje cells, but not in the granular layer of the cerebellar cortex. In a transgenic mouse model of DM1 it seems that disease manifestations are triggered only when the level of mutant RNA has exceeded a minimum level. However, in human myotonic dystrophy it appears that the amount of repeat expansion RNA in the DM2 myonuclei is actually greater than in DM1.
Derrick Wansink (The Netherlands): described a DM1-knock-in mouse model with repeat tracts of either (CTG)11 or (CTG)84. The latter mouse developed age-dependent, tissue-dependent and background-dependent repeat instability, which was highest in stomach and liver (especially in the hepatocyte fraction), but not more pronounced in muscle or tissues with high proliferative rate, such as blood and semen. This would speak against a role for recombination in repeat tract instability.
Berendt Wieringa (The Netherlands) reported on the role of mismatch repair genes in (CTG)n instability based on experiments with the (CTG)84 knock-in mouse. Spontaneous intergenerational expansion in 10 generations was moderate from 84 to 106. Msh6−/− CH3 mice showed increased instability compared to msh6 +/+ or +/−. The reverse was seen for msh3−/−, where msh3 deficiency stabilizes the expansion compared to msh3 +/− or +/+. Msh2 binds msh3 and msh6; more msh6 in the complex leads to stabilization, while more msh3 in the complex leads to instabilisation. Eventual roles for FEN1 exonuclease have been approached experimentally, but FEN-1 defective (CTG)84 mice show no increased instability.
7. Splicing errors in DM2
John Day reported on the aberrant splicing of the insulin receptor (IR). Insulin insensitivity is seen in 75% of DM2 patients, with ~36% showing manifest diabetes. The splice variant IR-A (exon 11−) is insulin-resistant, and the IR-B variant (exon 11+) is insulin sensitive. The proportion of IR-B in normal controls is 60–80%. In DM1 patients the proportions of IR-B ranged between 0 and 40%, and in eight DM2 patients a similar trend of splicing alterations of the IR were observed, the range was 18–26%.
Charles Thornton described the aberrant splicing of the skeletal muscle chloride channel (CLCN1). Membrane resistance in a Cl− bath in DM1 transgenic mice indicated a defect in chloride channel function. CLCN1 transcript studies showed aberrant splicing of exon7A in intron 6 dependent on the amount of mutant (CUG)n transcript. The consequence was loss of protein due to premature termination and increase of myotonia with higher proportions of splice variants. Quadriceps muscle samples of DM2 patients showed an even higher degree of myotonia compared to DM1 samples. Abnormal splicing simulates the myotonic events in immature fibres, and they can be increased by denervation.
Kenneth Ricker presented two German families with co-segregation of the common R894X CLCN1 mutation in DM2 patients. No grossly detectable difference in phenotype was seen. However, the co-segregation does not seem to be unusual as two other families have been published and Jurkat-Rott [10] showed the co-segregation in 16% of their PROMM families.
Nicolas Sergeant (France) reported on the aberrant splicing of the microtubule-associated tau protein (MAPT). Tau is known to be abnormally present in various forms of dementia. The gene consists of 14 exons; exons 2, 3, and 10 are alternatively spliced, giving rise to seven splice variants—one foetal and six adult isoforms. Exon 10 is important for microtubule binding. In other tauopathies specific patterns of tau-isoforms can be correlated with the disease. In DM1 only one unique isoform, tau60, occurs with loss of higher MW, exon 10 containing, isoforms. In one autopsy DM2 brain specimen (from the Finnish PDM-family [7]) a Ser 396/404 phosphorylated tau antibody recognized neurofibrillar degeneration with tau in various brain regions similar to DM1, and additional axonal degeneration in the spinal cord not shown previously in DM1. Taken together, these data indicate that similar to DM1, DM2 presents with tau pathology and therefore can be considered a tauopathy.
8. Expression profiling results
Ralf Krahe reviewed relative mRNA expression levels of DMPK (DM1) and ZNF9 (DM2) in normal tissues based on various publicly available human and mouse gene expression profiling databases. Both genes are widely expressed. ZNF9 appears to be expressed at slightly higher overall levels but is expressed at relatively lower levels in foetal brain and various other adult brain sub-tissues, which may explain the lack of a congenital DM2 phenotype with mental retardation similar to that seen in DM1. Gene expression profiling with microarrays was used to globally compare expression in skeletal muscle biopsies of normal, DM1 and DM2 individuals. Comparison of such expression profiles showed considerable overlap in the genes down-regulated among DM1 and DM2 patients and dysregulation of several functional gene categories, including muscle and myogenesis. Numerous skeletal muscle-specific genes (e.g. skeletal muscle sodium channel 1, myosin heavy chain, β-tropomyosin, desmin, several troponins, and α-sarcoglycan, but not dystrophin) were specifically affected. Similar changes were seen in vitro with primary myoblast cultures established from skeletal muscle of two DM1 patients for either total RNA or nuclear and cytoplasmic fractions separated. Expression profiling of nuclear and cytoplasmic fractions of DM1 myoblasts that displayed an inability to differentiate into multinucleated myotubes identified sets of genes down-regulated in both fractions (e.g. β-tropomyosin) and down-regulated in only the cytoplasmic fraction (e.g. troponin I, cardiac actin). These expression data suggest a global trans-effect of the transcribed expansion on the DM1 transcriptome.
9. Animal models
Laura Ranum (USA) described the untranslated (CTG)n expansion mutation of spinocerebellar ataxia type 8 (SCA8), the only other disease identified to date with the same genetic mutation as DM1. A new transgenic mouse model of SCA8 developed in her lab has a progressive lethal neurological phenotype. Development of DM2 mouse models is in progress.
10. Practical management
Peter Harper (UK) pointed out that the clinical management is different in DM2 in comparison to DM1. One of the major concerns in the genetic counselling of DM1 families is the problem of anticipation, which is always a major threat. In DM2, the parents have a minor perception of the disorder, because it has a relatively benign course, anticipation is not a regular phenomenon, and both the severe congenital as well as the childhood onset forms are absent. There is usually late onset of the disease in DM2. For all these reasons prenatal diagnosis is not needed. Careful clinical assessment is usually sufficient for presymptomatic testing in DM1. With the later onset and highly variable phenotype this will not be the case in DM2. Smaller expansions in younger ages are easier to detect by current Southern-blot based methods than the huge expansions at later ages. Fewer CNS symptoms and signs, fewer cardiac conduction problems, fewer anaesthetic risks and no swallowing or respiratory problems make practical management in DM2 less problematic than in DM1. All participants agreed that it is very important to establish and communicate the diagnosis to the patients, because many have had difficult experiences with longstanding pains and stiffness with incorrect explanations.
Ken Ricker found the late stage of DM2 less problematic than DM1 because of absent facial changes, bulbar weakness and manual skill disability, which makes the DM2 patient socially more competent. From a practical point, the myotonia is not very disabling, and drugs are usually of no benefit. A challenging problem is the management of muscle pain, which may be disabling for the patients and limiting everyday activities. Some respond well to carbamazepine, some do not, and controlled clinical trials have not yet been performed. The anaesthetic risk is unsettled but at the moment is advisable to follow the guidelines suggested for myotonic dystrophy type1, and specific obstetric problems have not been reported.
11. PROMM-like disorders without DM1 or DM2 mutations
Ken Ricker reported that when the more than 100 families identified to date in Germany were checked for the DM2 mutation, there were no families with a PROMM phenotype that did not segregate the DM2 mutation. However, there are single individuals and families with symptoms and signs bearing similarity to PROMM but without the full range of diagnostic features previously established for PROMM. In one family the father had cataract surgery and proximal muscle weakness after age 58. The son had cataracts and proximal weakness after age 57. CK was almost normal, gamma-GT barely elevated and EMG showed some ‘myotonia-like’ runs. Muscle biopsy was seen as compatible with PROMM. The other son also had cataracts without a muscle phenotype.
Bjarne Udd showed a few sporadic cases with incomplete PROMM phenotype but some myotonia-like features either on EMG or clinically. One group showed findings on muscle biopsy compatible with DM2, such as frequent internal nuclei, nuclear clump fibres and type 2 fibre atrophy. The other group showed a very different biopsy finding with mildly small but angulated type 2 fibres and normal type 1 fibres. In a third group the explicit biopsy feature was muscle fibres expressing both fast and slow myosins.
Ralf Krahe reported on a collection of 24 families presenting with a PROMM-like phenotype and confirmed exclusion of DM1 and DM2. The set included families from Germany, Italy, Brazil, France, the USA and Spain. Fourteen families had more than one affected. Genome-wide scans for genetic linkage in several of the extended kindreds provided maximum lod scores (1.5) close to the maximal attainable lod scores.
The recommendation was that before more effort is invested in molecular genetic studies of these families the consortium agreed to take a closer look at the clinical data using standardized protocols for clinical assessment and consensus diagnosis. Ken Ricker, Bjarne Udd and Giovanni Meola agreed on evaluating the clinical data for further processing on the genetic side.
Guillaume Bassez reported a new large French family with seven affected among 30 members (primarily described and examined by Dr Leber and Dr Hannequin). The mean age at onset was 44. The autosomal dominant disorder included proximal muscle weakness at onset in five, clinical or electrical myotonia in five, DM1-type cataracts in all, and insulin resistance in one affected. A rapidly progressive fronto-temporal dementia (FTD) occurred later in the course of the disease causing the death at ages around 60 years. CK was mildly elevated or normal and no cardiac abnormalities were encountered. Choledocholithiasis, scapula alata, and dropped head were part of the syndrome. Muscle biopsy showed rimmed vacuoles, small angulated type 1 and type 2 fibres, and marked fatty replacement and fibrosis at late stages. Brain imaging revealed variable frontal lobe cortical atrophy, whereas the SPECT fronto-temporal activity was reduced in all patients. On autopsy the neuropathological findings and the tau protein biochemical profile were consistent with a FTD lacking distinctive histology. Molecular analysis excluded a repeat expansion in DMPK and ZNF9 genes and a linkage at the loci SCN4A and CLC-1. No mutation could be identified in MAP-Tau, APP and PSEN1 genes.
Giovanni Meola reported on a small family with dominant myotonia, no wasting, posterior cataracts in the father but no nuclear clump fibres or type 2 fibre atrophy on muscle biopsy.
12. Conclusions
12.1. DM2-diagnostic criteria
The identification of the DM2 mutation has completely changed the diagnostic procedures. Verification of the DM2 mutation is the sole relevant method for final diagnosis. In the selection of patients for molecular diagnostics, previously established clinical diagnostic criteria for PROMM are still useful, even if the DM2 mutation may cause a wider spectrum of phenotypes than classical PROMM and may thus occur more frequently than the core phenotype. The new finding of extremely small type 2 fibres and nuclear clump fibres that occur very early, especially in the vastus lateralis muscle, may serve as additional criteria for considering DM2 mutation analysis.
12.2. Genetic counselling
At present, there is no way to correlate the size of the (CCTG)n expansion with the severity of the clinical phenotype. Although anticipation has been reported, it is not considered a regular and general feature of the disease, even if it may occur in selected families. Moreover, no congenital manifestations of the DM2 mutation have been identified. There appears to be nearly 100% penetrance in adults with DM2 expansions who have been neurologically evaluated in families with typical PROMM/DM2 phenotype.
12.3. Nomenclature
The myotonic dystrophies are now divided in two genetically recognized disease entities:
The presenting phenotype in myotonic dystrophy type 2 may be either the classic PROMM phenotype, a phenotype resembling Steinert's disease, or patients presenting with other phenotypic expressions. For patients and clinicians the term myotonic dystrophy type 1 or 2, instead of the abbreviations DM1 or DM2, are preferred when the diagnosis is established. PROMM may still be used as a clinical phenotype description, but once the mutation is verified, type 2 myotonic dystrophy is preferred.
The gene locus in myotonic dystrophy type 2 is marked with the symbol DM2 and the mutation with the symbol DM2, corresponding to DM1 and DM1 in myotonic dystrophy type 1.
12.4. Further consortium procedures
(1) The working group: Udd, Ricker, Meola will evaluate families with PROMM-like complex myotonic disorders lacking both DM1 and DM2 mutations. (2) The next DM2 workshop will take place in 2–3 years time. (3) A newsletter using ENMC Web sites will be considered for communicating new important information on myotonic dystrophy type 2.Workshop participants:
Acknowledgements
This Workshop was made possible thanks to the financial support of the European Neuromuscular Centre (ENMC) and ENMC main sponsors: Association Française contre les Myopathies (France), Deutsche Gesellschaft für Muskelkranke (Germany), Telethon Foundation (Italy), Muscular Dystrophy Campaign (UK), Muskelsvindfonden (Denmark), Prinses Beatrix Fonds (The Netherlands), Schweizerische Stiftung für die Erforschung der Muskelkrankheiten (Switzerland), Österreichische Muskelforschung (Austria), and Vereniging Spierziekten Nederland (The Netherlands).
References
- . Myotonic Dystrophy. 2nd ed.. London: W.B. Saunders Company; 1989;
- . Myotonic dystrophy with no trinucleotide repeat expansion [see comments]. Ann Neurol. 1994;35(3):269–272
- Proximal myotonic myopathy: a new dominant disorder with myotonia, muscle weakness, and cataracts. Neurology. 1994;44(8):1448–1452
- . Proximal myotonic myopathy (PROMM) and other proximal myotonic syndromes. Neuromuscul Disord. 1998;8(7):519–520
- . Genetic mapping of a second myotonic dystrophy locus. Nat Genet. 1998;19(2):196–198
- . New nomenclature and DNA testing guidelines for myotonic dystrophy type 1. Neurology. 2000;54:1218–1221
- . Proximal myotonic dystrophy— a family with autosomal dominant muscular dystrophy, cataracts, hearing loss and hypogonadism: heterogeneity of proximal myotonic syndromes?. Neuromuscul Disord. 1997;7(4):217–228
- . Dominant multi-system proximal myotonic myopathic syndromes: clinical and genetic heterogeneity in three families (Abstract). Neurology. 1999;52(Suppl. 2):A95
- Myotonic dystrophy type 2 caused by a CCTG expansion in intron 1 of ZNF9. Science. 2001;293(5531):864–867
- . Report of the 84th ENMC workshop: PROMM (proximal myotonic myopathy) and other myotonic dystrophy-like syndromes: 2nd workshop. Neuromuscul Disord. 2002;12:306–317
PII: S0960-8966(03)00092-0
doi:10.1016/S0960-8966(03)00092-0
© 2003 Elsevier B.V. All rights reserved.
