Bethlem myopathy (BETHLEM) and Ullrich scleroatonic muscular dystrophy: 100th ENMC International Workshop, 23–24 November 2001, Naarden, The Netherlands
Article Outline
- 1. Introduction
- 2. Bethlem myopathy
- 3. Ullrich's disease
- 4. Syndrome of myosclerosis
- Acknowledgements
- Appendix A. Diagnostic criteria (Ullrich CMD due to primary collagen type VI involvement)
- Appendix B. List of workshop participants
- References
- Copyright
1. Introduction
Bethlem myopathy (OMIM # 158810) is an early-onset benign myopathy characterized by proximal muscle weakness and multiple flexion contractures [1], [2], [3]. It is caused by dominant mutations in COL6A1 (OMIM # 120220), COL6A2 (OMIM # 120240) [4], and COL6A3 (OMIM # 120250) [5] genes.
Seventeen active participants attended the International Workshop of the Consortium on Bethlem myopathy from six countries, including France, Italy, Japan, The Netherlands, Turkey, and the United Kingdom.
Since Camacho Vanegas and co-workers [8] demonstrated the association of a second disease with COL6 genes, this second meeting on Bethlem myopathy included sections on Ullrich scleroatonic muscular dystrophy (UCMD) [6], [7] – a subgroup associated with collagen type VI and other clinical entities overlapping with the above disorders in which the molecular bases are still unknown [9].
2. Bethlem myopathy
Cardiac and respiratory muscle involvements are well-known complications in muscular dystrophies. Until now, in Bethlem myopathy, cardiac abnormalities have not been described, and pulmonary involvement is supposed to be rare.
Anneke van der Kooi reported a study by the Dutch group and their collaborators on patients presenting with a phenotype compatible with Bethlem myopathy.
Both frequency and nature of cardiac involvement were investigated in 121 Bethlem myopathy patients (59 males, 62 females) from 61 families. Patients were subdivided into two categories: genetically proven or possible. Cardiac involvement was assessed by electrocardiography (ECG), echocardiography (ECHO) and 24
h ECG (Holter). In addition, pulmonary investigations were performed. ECG was available in 74 out of 121 patients, ECHO in 51, and Holter registration for 24 patients.
Two patients experienced dyspnea, one of them was known to be affected by atrial fibrillation (AF).
ECG recordings in eight patients, Holter recordings in four, and echocardiographies in six patients were abnormal. Three members from one family with an 898G>A (G250S) COL6A2 mutation had intraventricular conduction delay and atrial dilatation. Two possible Bethlem myopathy patients had the following abnormalities; right bandle branch block (RBBB), and pathological Q's in lead V1, anterior vein of the leg (AVL) and V6, an intraventricular conduction defect and diastolic dysfunction in another, and an accelerated atrial rhythm, and one run of 11 ventricular extrasistolic/supraventricular extrasistolic in the third.
Pulmonary investigations were performed in 56 patients. Five patients had a vital capacity below 50% and 11 between 50 and 70%. Two patients were on respiratory support.
In conclusion, cardiac investigations in Bethlem myopathy do reveal abnormalities in 10% of cases. Its relationship with muscle disease remains to be established. Pulmonary involvement is part of the clinical spectrum, and seems to be related to more severe weakness.
The next section reports the clinical features of patients with a known mutation in one of the COL6 genes, the various kinds of mutations so far identified and the first attempt of a genotype/phenotype correlation.
Francesco Muntoni presented the clinical features of several families.
In one family (Family 1), the onset was congenital, with the propositus presenting bilateral hip dislocation and torticollis at birth. Her motor milestones were significantly delayed and serum creatine kinase (CK) normal. A muscle biopsy performed at the age of three led to a diagnosis of congenital muscular dystrophy (CMD). Currently, aged 28, she is able to walk but cannot get up from the floor or climb stairs without a banister. Her weakness is predominantly axial and proximal. She has a striking limitation of arm elevation, and elbow, spine, hip and ankles movements. She also has marked contractures of the long finger flexors. Her general health has remained good despite the mild progression of weakness over the years.
Her daughter, who is also affected, presented a mild clumsiness and tightness of Achilles tendon at the age of four. She has only mild proximal and axial weakness, is capable of walking and running, and has no contractures, except for the long finger flexors. Her CK is normal. Caroline Sewry reported that the muscle biopsy from this child showed reduced immunolabelling of collagen VI.
The second family (Family 2) is a three-generation family with the classic phenotype. Manifestation of the disorder occurred in all affected cases in the first few years of life with mild clumsiness and joint laxity, followed by progression of the weakness and contractures, with difficulties in running during the second decade of life. Difficulty in walking required a wheelchair in the fifth decade of life. Significant respiratory involvement was not a noteworthy feature in this family. Serum CK was normal or only slightly high, muscle biopsy ranged from myopathic to dystrophic. Muscle Magnetic Resonance Imaging in this family showed the generalized and progressive muscle involvement with severe fibroadipose tissue replacement in the limb muscles of an individual in the third generation, moderate changes giving rise to a moth-eaten appearance in an individual of the second generation, while the muscle was almost entirely normal in an affected child aged 7 years.
Characteristic features in the third family (mother and son affected) (Family 3) were moderately high level of serum CK (2–6 times), pes cavus and hypertrophy of the peroneal muscles, while the muscle biopsy was clearly dystrophic in both the child and his mother.
Additionally, nine dominant families with features evocative of Bethlem myopathy were also reported. For two of them, linkage analysis was consistent with the clinical diagnosis of Bethlem; in the rest of the families, molecular studies are still in progress.
Kate Bushby presented data on three families where the collagen VI mutations had been defined. In one family (Family 4), the diagnosis of autosomal dominant Emery Dreifuss muscular dystrophy had first been considered based on the presence of severe contractures at the elbows and Achilles tendons. A mother and her son were both affected. Guglielmina Pepe reported that a multiexon skipping in COL6A1 was found.
In the second family (Family 5), the diagnosis of congenital myopathy had been made in the mother: she developed progressive contractures of her Achilles tendons, elbows and fingers. Two of her three daughters presented with proximal muscle weakness in the first 3 years. They did not have Achilles tendon contractures at presentation; however, finger contractures were detectable on examination.
The third family (Family 6) represented a sporadic case. Progressive Achilles tendon contractures bound this 8-year-old girl to a wheelchair for mobility: a subsequent release procedure significantly improved her mobility.
Rabah Ben Yaou reported on seven patients, five from France and two from Italy. Two cases were sporadic and five familial. The clinical onset ranged from early childhood to the first decade. The most recurrent symptoms upon examination were weakness in lower limbs (frequent falls, difficulty in running, climbing stairs, jumping, getting up from the floor and walking), rather than axial weakness (difficulties in controlling head movements), which was less frequent, and sometimes weakness in upper limbs.
No symptoms in the neonatal period, no documented delayed motor milestones and no mental retardation were reported. Course was characterized by slowly progressive periods in three patients, stationary periods in five patients, and rapidly progressive periods detected in two patients. All patients are still ambulant.
Clinical examination showed muscle wasting (the most frequent pattern is proximal muscle wasting in the four limbs and distal one in the lower limbs) with wasting of the thigh muscles being very common. Muscle weakness predominantly in the proximal parts (upper and/or lower limbs) is the most frequent pattern (thighs are particularly affected in the lower limbs). Axial participation (flexors of trunk) was detected in three patients. Contractures were observed in six patients (finger flexors and ankles), two patients (elbows), and one patient (spine with rigidity, hip, toe flexors and face). One patient had evidence of hyperlaxity in tarsometatarsian and knee joints. Thorax deformity, prognathism and scoliosis were also found in one patient and pes cavus was present in two patients, and muscle hypertrophy was absent. Severe myopia was observed in one patient. No cardiac abnormalities were observed.
Ancillary examinations: muscle computer tomographic (CT) scan (in three patients) showed fatty infiltration predominantly in the proximal muscles. CK was normal (two patients) or slightly high (<3) in three others. Muscle biopsy from six patients showed a dystrophic pattern in three and myopathic pattern in three. The most frequent histological abnormalities were fiber size variation with slight or moderate fibrosis, internal nuclei and type 1 fiber predominance.
Marianne de Visser presented a family (Family 7) encompassing four affected family members (three males and one female, ranging from 26 to 54 years of age).
The index case is a 32-year-old man who had delayed motor milestones. He could walk independently at the age of 20 months. He fell frequently and was never able to run properly. Weakness was slowly progressive urging him to use a cane. On examination, there was weakness of the neck flexors and generalized weakness of the limbs, proximal more than distal. He had pes cavus and claw toes, and contractures of the biceps brachii, finger flexors, hamstrings and triceps surae. Serum CK activity was markedly high (10 times). A muscle biopsy showed dystrophic changes.
His 40-year-old sister had a similar clinical picture, albeit more severe, because she needed a wheelchair for outdoors transportation. She had torticollis.
A paternal uncle (54 years) had limb–girdle weakness and the same contractures and foot deformities as described above, but he was still able to work. CK was normal.
His eldest son (29 years) had only contractures and atrophy of the sternocleidomastoid muscles but no weakness whatsoever. His CK was six times above the upper limit.
They carry an exon-skipping mutation in COL6A1.
Betti Giusti reported the mutations characterized in a screening of 40 Bethlem patients which is still in progress. We subdivided the mutation screening analysis into two phases. In the first phase of the project, we analyzed the cDNA obtained from fibroblast cell cultures, lymphoblastoid cell lines or muscle biopsies with the following technical approach: total RNA extraction from cells or tissue, reverse transcription (RT)-RNA polymerase chain reaction (PCR) amplification, heteroduplex analysis by conformational sensitive gel electrophoresis and, whenever a heteroduplex was identified, direct sequencing of the fragment were performed. The identified mutations were all confirmed on genomic DNA and investigated in the relatives of the familial cases. Segregation analysis in informative Bethlem families limited the following mutation screening analysis to only one gene in 47% and to two genes in 41% of the families. In the families where COL6A2 and COL6A3 were excluded and where we did not find a mutation in the COL6A1 cDNA we are still carrying out a genomic DNA analysis on each single exon. Genomic DNA analysis represents the second phase of the mutation screening analysis.
We identified seven pathogenetic mutations in 14 patients over a total of 40 cDNAs and eight genomic DNAs (second screening) investigated. They are all located inside COL6A1. We detected exon skipping, splicing mutations, missense and nonsense mutations in seven patients. Three were common mutations which occurred in unrelated patients of different countries: a 66
bp in-frame deletion due to a splicing mutation that activated a cryptic splice site inside an exon in four unrelated patients from different countries (two English, one French and one Italian). The second was exon skipping due to splice site mutation consisting of single nucleotide substitution found in three unrelated patients (two English and one German). The third mutation was again exon skipping detected in two unrelated French families and one Dutch. The remaining four mutations were private mutations. One was nonsense mutation, a nucleotidic substitution which created premature stop codon (French), another was glycine to arginine substitution in the triple helix (English) and the last one was 252
bp deletion covering six exons (English).
We also identified several missense mutations (but no glycine substitutions) in the three cDNAs, whose pathogenicity is under investigation. Indeed, for each substitution, we analyzed affected and healthy relatives, whenever available, a panel of 100 controls, and other Bethlem patients.
Based on the accurate clinical description of the patients and the characterization of the mutations that they have, Guglielmina Pepe performed a first attempt of genotype/phenotype correlation in Bethlem patients. The kind of mutations so far detected are mostly represented by exon skipping, causing in-frame deletions and haploinsufficiency or missense glycine substitutions, one nonsense mutation and a multiexon deletion. It is important to stress the fact that we still do not know whether there are mutations that exert a dominant negative effect in collagen VI.
We can roughly subdivide the mutations into five degrees of severity. The most severe one seems to be the one with a COL6A1 multiexon out-of-frame deletion. It is the case described by Katie Bushby of a youngster with severe contractures (Family 3). A large group characterized by a high degree of severity was the one featuring exon skipping inside the triple helix of COL6A1 that comprised the patients carrying glycine substitutions in the triple helix of COL6A1 and COL6A2 [4], [10]. Some of these patients have been confined to a wheelchair since the age of 45–50. Some patients reported in the literature cannot be mentioned in this correlation because we do not have accurate clinical information. The third and fourth group had deletion (cryptic splice site activation) and exon skipping both in the NH2 terminal end of COL6A1. The fourth, detected only in two of the new French families reported by Ben Yaou, was even milder than the third. Interestingly, one Italian family, reported by Bertini et al. in the previous workshop on Bethlem myopathy, presented the mildest phenotype so far described associated to a glycine substitution in the 5′ end of the COL6A3 triple helix [11]. In conclusion, it seems that overall large deletions and mutations inside the triple helix exert a more severe phenotype than those associated with mutations in the N-terminal globular region. The characterization of more mutations and the analysis of the mutant mRNA and protein expression in fibroblasts and tissue will help in gathering data for a genotype/phenotype correlation.
The last presentation of the Bethlem myopathy section by Guglielmina Pepe and Enrico Bertini described a patient with all the clinical stigmas necessary to make a diagnosis of both Bethlem myopathy and Hypermobility/Classic mixed type of Ehlers–Danlos. Ehlers–Danlos syndromes are a heterogeneous group of inheritable connective tissue disorders characterized by articular hypermobility, skin extensibility and tissue fragility. They are subdivided into six major types [12]: (a) classic; (b) hypermobile; (c) vascular; (d) kyphoscoliotic; (e) arthrochalasic and (f) dermatosparaxic. The classic type is inherited as an autosomal dominant trait: major diagnostic criteria are represented by skin hyperextensibility, presence of wide atrophic scars as manifestations of tissue fragility and joint hypermobility. Among the minor criteria, we found smooth velvet skin, dislocations/subluxations, hernias and easy bruising. The hypermobility type is also a dominantly inherited disorder with skin involvement (hyperextensibility and/or smooth, velvet skin) and joint hypermobility as major criteria and recurring joint dislocations and chronic joint/limb pains as minor diagnostic criteria. The proposita, a 46-year-old Italian woman, presented joint hypermobility, spontaneous generalized luxations, smooth velvet skin, easy bruising, spontaneous ecchymosis, chronic joint pain as connective abnormalities; progressive weakness, lost walking at the age of 35, normal CK, concentric degeneration of muscle fascicles by CT scan, flexion contractures of hands, elbows and knees as neuromuscular clinical features and pes planus and fatigue common to both groups of disorders. The patients presented sufficient stigmas to be diagnosed as affected by Bethlem myopathy and mixed Ehlers-Danlos Syndrome (EDS) of classic/hyperextensible clinical type. We also examined the daughter of the proposita, 16 years of age, who presently displayed the clinical stigmas of mixed EDS type. The proposita reported that her father and paternal grandfather had contractures and had difficulty in walking, while her mother presented EDS clinical stigmas. The clinical and family history suggested that the proposita may have had two dominant mutations, one inherited from her mother's side, the other inherited from her father's side. We are investigating the molecular defect underlying these pathologies.
3. Ullrich's disease
Enrico Bertini presented the clinical features of four patients from three families, affected by Ullrich disease, who had molecular genetic confirmation, recently published [8]. In Family A, the pedigree suggested autosomal recessive inheritance; in Family B, with two affected male children, the parents were not related and in Family C, the affected girl was a sporadic case. Patients had normal or delayed motor milestones and they were all able to walk autonomously. The male patient of Family A showed congenital torticollis and kyphosis at birth, typical distal hyperlaxity of hands and feet, multiple proximal joint contractures of his knees and elbows, and left hip dislocation. He walked autonomously at the age of 13 months and joint contractures improved spontaneously during the first 2 years of life. Tracheostomy was necessary at the age of 8 and now at the age of 11 he is still able to walk. The two children in Family B showed proximal joint contractures, congenital kyphosis, and marked distal hyperlaxity at birth. They both had delayed walking and are now 2 and 3 years old, respectively. As in Family A, proximal joint contractures and congenital kyphosis improved spontaneously. Finally, the patient of Family C, who had typical clinical features at birth, achieved autonomous walking at the age of 3, was unable to walk 1 year later and died at the age of 10 of respiratory failure, 6 months after she had undergone surgery for implantation of Harrington rods to correct scoliosis. During the follow-up, we did not observe hyperhydrosis, but rather dry skin and protruding hair follicles, particularly in the skin of arms and legs. Three of his patients were also submitted to a detailed ophthalmological examination and ECHO and ruled out corneal opacities and heart involvement.
Kate Bushby presented two patients with a classic Ullrich's phenotype. The first, a 13-year-old girl from a consanguineous family, had presented neonatally with kyphosis and hypotonia. She had been independent from the age of around 2–4 years and subsequently required a wheelchair. Her respiratory function deteriorated so that she needed nocturnal ventilation from the age of 11 years, and progressive scololiosis required operative management at around the same age. Her nutritional status is currently causing concern. Collagen VI immunolabelling in her muscle is absent and haplotyping confirmed linkage to COL6A3. The second ‘classic’ case is a baby of 19 months with kyphosis, proximal contractures and distal joint laxity.
Two additional cases were presented which may represent overlapping phenotypes. In the first, in addition to hypotonia and motor delay, a 2-year-old girl had a neonatal torticollis and abnormal skin pigmentation. Collagen VI immunolabelling in muscle was variable and reduced. The second case was of a girl with short stature, mild learning difficulty, prominent contractures and a very ‘pinched nose’ appearance.
Luciano Merlini presented two families, one with the classic severe form of UCMD, and the other with a mild phenotype. The first case was a 11-year-old boy, the only child of a consanguineous family. He was born by caesarean delivery for a breech presentation. He presented with generalized hypotonia from birth and associated fracture of the right femur, bilateral hip dislocation, and talus of the right foot. He had delayed motor milestones and never achieved the ability to walk. Intelligence was normal. Clinical examination showed diffuse muscle weakness that was more severe in the axial (neck flexion 1/5 Medical Research Council (MRC)) and girdle (1–2/5 MRC), moderate in the proximal and distal compartments (2/5 MRC), and mild in the face. He was able to sit and had severe kyphoscoliosis, and marked contractures of elbows, hips and knees. There was striking distal joint laxity (hands and feet). In addition, he had brown teeth, rough skin, mild myopia, and exophthalmos. CK was normal. He had respiratory failure and severely reduced forced vital capacity (FVC) (330
ml, 18% of the predicted value), and profound nocturnal oxygen desaturation (57% of the night time under 90%). Muscle CT showed a marked diffuse involvement of the axial, shoulder and hip girdles muscles. The thigh muscles (quadriceps and hamstrings) presented the same peripheral type of muscle substitution with a preserved central area of muscle density, which is characteristic of Bethlem myopathy. Muscle biopsy showed a myopathic pattern with fiber size variation and fibrosis. Immunohistochemistry showed normal labelling with laminin α2, absence of integrin α7B around fibers, and complete absence of collagen type VI. His parents had a normal clinical examination, normal CK, and a normal muscle CT. The disorder was linked to COL6A1–COLA2 locus. The boy died a few months later because of respiratory insufficiency. The second case was a 16-year-old girl, born from a consanguineous Italian family, who was floppy at birth, slow in walking since the beginning and was never able to run. Muscle weakness was severe in the axial and distal segments and mild in the girdles and proximal limbs. In the fingers, she showed a combination of finger flexors contractures (Bethlem sign) and hyperlaxity (Ullrich sign). In addition, there was laxity of elbows, knees, and toes, and protruded calcaneus. CK was four times normal. FVC was 89% of the predicted value. Muscle CT revealed diffuse muscle attenuation, in the quadriceps, with the ‘peripheral’ type of involvement, which is characteristic of the Bethlem myopathy. Muscle biopsy was dystrophic, laminin α2 was normal, collagen type VI and Laminin β1 were reduced in some fibers. Pascale Guicheney found in this case a nonsense mutation in the N-terminal domain of COL6A3, which induced an aberrant splicing of exon 5 with an in-frame deletion allowing the production of a functional protein.
Francesco Muntoni reported the experience regarding the frequency and clinical spectrum of UCMD secondary to collagen VI involvement in his population. Fifteen patients (age 3–23.6 years) with a clinical diagnosis of UCMD were studied: linkage analysis to the three COL6 genes was performed in all informative families (n=7), while immunohistochemical analysis of collagen VI expression in muscle was performed in the remaining cases. An immunocytochemical reduction of collagen VI was observed in six patients. Three of the six cases belonged to informative families and haplotype analysis clearly suggested linkage to the COL6A1/COL6A2 locus in two cases and to the COL6A3 loci in the third case. In the remaining nine patients, a primary collagen VI involvement was excluded on the basis of either the linkage analysis (four families) or immunohistochemical studies. Age and presentation at onset, the distribution and severity of weakness and contractures and the frequency of non-ambulant patients were similar in the patients with and without collagen VI involvement. Distal laxity, rigidity of the spine, scoliosis, failure to thrive and early and severe respiratory impairments were found in all patients at the end of the first decade of life, irrespective of their maximum motor functional ability or their collagen status. Francesco Muntoni concluded that collagen VI involvement is relatively common in UCMD. However, the role of this molecule was clearly excluded in a number of cases, suggesting genetic heterogeneity of this condition.
Ikuya Nonaka reported his experience on this syndrome. In the past 23 years, we have had 252 patients with various forms of congenital muscular dystrophies. Among them the most common form has been Fukuyama-type CMD prevalent in Japan accounting for 69% of the cases. Since we had nine patients with Ullrich's disease (congenital atonic–sclerotic muscular dystrophy) during the same period, the disease seemed to be rare. Nine patients had the clinical characteristics of the disease including early onset of proximal joint contractures including congenital torticollis in five, limited neck flexion in all, early onset kyphoscoliosis and congenital hip dislocation. All had distal joint hyperlaxity and facial muscle involvement with high-arched palate. Four patients became ambulant for several years.
All had myopathic changes with evidence of necrotic and regenerating process in their muscle biopsies. Type-1 fiber predominance was detected in five and type 1 fiber atrophy was found in two patients. On immunohistochemistry with anti-collagen VI antibody, one had complete deficiency and eight had partial deficiency (see Dr Nishino's abstract). There were no definite clinical differences between patients with complete deficiency and partial deficiency.
Haluk Topaloglu presented the clinical features of eight families within the spectrum of Ullrich's disease. Consanguinity was present in all. The age of the patients varied between 20 months and 12 years. They all invariably had neonatal hypotonia, prominent generalized weakness, distal laxity and multiple-joint contractures. Mental retardation was absent in all. Congenital hip dislocation was seen in five of the cases. Four had congenital torticollis and three had scoliosis/kyphosis. Serum CK was normal in seven, and mildly elevated in one. Two patients were able to walk starting from 2 to 3.5 years of age. The clinical course was static or slowly progressive in seven, whereas in one case there was a gradual improvement. Two of these families were linked to the COL6A3 locus, and subsequently a homozygous nonsense mutation was found in one by Pascale Guicheney. Four others were linked to COL6A1/COL6A2 chromosome 21 cluster. The remaining two families are yet to be studied by linkage. Immunohistochemical analysis of collagen VI in four cases showed an absence of the protein in three, and a partial reduction in one. Interestingly, of the two patients who could walk, one had total absence and the other had partial reduction of collagen VI.
4. Syndrome of myosclerosis
Luciano Merlini reviewed the syndrome of Myosclerosis that was first described by Bradley [13] in two siblings of a non-consanguineous family. Difficulty in walking was noted in early childhood together with toes walking and progressive calf contractures. The clinical picture in the early thirties was characterized by slender muscles with firm ‘woody’ consistence, restriction of movement of many joints due to contractures of the muscles including elbows, hips, knees, and ankles. CK was normal. Muscle biopsy in the index case was compatible with indolent primary myopathy with the fibrosis as the most striking abnormality. Subsequent follow-up of this family (K Bushby) has shown a very slowly progressive muscle weakness in the proband together with progressive respiratory muscle involvement now necessitating nocturnal respiratory support. Two sisters are much less severely affected and would conform to a ‘classical’ Bethlem myopathy phenotype. An Italian consanguineous family with a similar phenotype was reported at the first Bethlem workshop [14]. In this family, a brother and a sister had a childhood onset of calf contractures followed by progressive restriction of movements of all joints including jaws, spine, shoulders, elbows, wrists, fingers, hips, and knees. CK was almost twice normal. FVC was reduced (44% and 32% of the predicted value). Heart was normal. Muscle biopsy was dystrophic and laminin beta 1 was reduced. The most striking abnormality was, however, a mosaic expression of collagen VI. We raised the possibility of a primary or secondary involvement of the protein in this disorder [14]. Indeed, Pascale Guicheney showed that in this family the disease was linked with COL6A1–COL6A2 cluster, while rigid spine muscular dystrophy 1 in 1p35–36 and COL6A3 locus were excluded.
Switching from the clinical description of Ullrich families to the animal model, Paolo Bonaldo presented the characterization of the pathogenetical defects of collagen VI knockout mice as animal model for Bethlem myopathy and Ullrich syndrome.
To elucidate the function of collagen VI in vivo, a knockout mouse model was previously generated by targeted inactivation of the Col6a1 gene. Collagen VI-deficient mice showed myopathic lesions similar to those described in human Bethlem myopathy [15]. In order to dissect the pathogenetical mechanisms leading to myopathic defects, detailed structural and functional investigations were performed on muscles from collagen VI-deficient and control mice. The force developed by Col6a1−/− muscle fibers was markedly reduced when compared to controls, thus indicating that their contractile properties were severely impaired. Ultrastructural analysis of various Col6a1−/− muscles reveals abnormal mitochondria (with swelling and reduced or altered cristae) and dilatation of the terminal cisternae of sarcoplasmic reticulum. As these data suggest, alteration of the intracellular Ca2+ homeostasis and compromised mitochondrial function in collagen VI deficient muscles was further investigated. It is well established that in the presence of abnormal levels of Ca2+, mitochondria becomes swollen and depolarized as a result of a large non-specific increase in membrane permeability, a process known as ‘mitochondrial permeability transition’ which in turn triggers apoptosis [16]. When measured in vitro with specific probes, Col6a1−/− fibers are much more sensitive to mitochondrial depolarization following exposure to oligomycin, a mitochondrial adenosine triphosphate synthase inhibitor. Oligomycin also causes a marked increase in cytosolic Ca2+ levels in Col6a1−/− fibers relative to control fibers, an event which can be blocked by inhibitors for sarcoplasmic reticulum Ca2+ release channels. This suggests that mitochondria of Col6a1−/− fibers are functionally impaired due to a disregulation of intracellular Ca2+ homeostasis. Both the loss of mitochondrial membrane potential and the increase in cytoplasmic Ca2+ levels are completely abolished when the Col6a1−/− fibers are cultured on collagen-VI-coated plates. Finally, several apoptotic and necrotic fibers are detected in Col6a1−/−, but not in control muscles. These findings indicate that collagen VI plays a key role in the maintenance of the structural and functional integrity of muscle fibers. Moreover, these observations provide an insight into the pathogenetical mechanism(s) involved in the murine and human myopathic disease.
The next few presentations will cover biochemical and morphological studies so far performed on Ullrich syndrome.
Giovanna Lattanzi presented the data of her group on fibronectin interaction in normal and collagen VI deficient fibroblasts. They studied fibronectin–collagen VI interaction in cultured human and mouse fibroblasts, in an attempt to identify extra-cellular matrix (ECM) components interacting with collagen VI and altered in cells with collagen VI defects. Fibronectin is one of the main components of the ECM and associates with a variety of other extracellular matrix molecules including collagens. Interaction of fibronectin with cells through integrin receptors activates a number of signal transduction pathways. By using fluorescence microscopy, they detected a fine network of collagen VI filaments and fibronectin fibrils in the ECM of normal murine and human fibroblasts. The two microfibrillar systems were interconnected at discrete sites which could be revealed by immunoelectron microscopy. Direct interaction between collagen VI and fibronectin was also demonstrated by Western assay. In primary fibroblasts from Col6a1 null mutant mice, collagen VI was not detected in the ECM and a different pattern of fibronectin organization was observed, with fibrils running parallel to the long axis of the cells. Similarly, an abnormal fibronectin deposition was observed in fibroblasts from a patient affected by Bethlem myopathy, where collagen VI secretion was drastically reduced. The same pattern of fibronectin organization was also observed in UCMD fibroblasts which either did not produce any detectable amount of collagen VI in the ECM or secreted a strongly reduced amount of mutated collagen VI. To identify the mechanism affecting fibronectin organization in collagen VI deficient ECM, an in vitro study was performed with soluble collagen VI peptides added to the culture medium. A competition effect at the site of collagen VI binding in the fibronectin molecule, caused the same parallel arrangement of fibronectin fibrils observed in the pathological samples, thus suggesting that a protein–protein interaction failure occurred at the fibronectin polymerization sites in the ECM. These results demonstrate that an altered collagen VI matrix or the complete absence of collagen VI from the ECM can affect the three-dimensional organization of fibronectin and possibly of other ECM components. The altered conformation of the fibronectin matrix could in turn trigger additional pathogenetic mechanisms in Bethlem myopathy and UCMD fibroblasts.
Ichizo Nishino reported the analysis of muscle specimens from nine patients with Ullrich disease. The diagnosis of Ullrich disease was clinically made by Dr Nonaka and muscle biopsy was performed in all patients. They performed immunohistochemical analyses for collagen VI and other collagens. Collagen VI was normally colocalized in sarcolemma with collagens VI and V. It was also distributed in the interstitium when fibrosis was present in muscles such as those from muscular dystrophies. In Ullrich disease muscles, they found two patterns of collagen VI abnormality; complete deficiency in one patient and ‘partial’ deficiency in eight other patients. In ‘partial’ deficiency, collagen VI was deficient specifically in sarcolemma but was present in the interstitium. Double immunostaining for collagens IV and VI clearly demonstrated the absence of collagen VI in the sarcolemma, while collagen IV was present. On electron microscopy, basal lamina itself was intact even in degenerating muscle fibers with disorganized myofibrils in either complete or partial deficiency cases. Collagen fibrils in the interstitium appeared normal with typical periodic pattern of about 60
nm interval. In complete deficiency case, they did not find any microfibril, which seemed to be compatible with the fact that microfibrils are mainly composed of collagen VI. In partial deficiency cases, they did find microfibrils but looked intermingled with collagen fibrils and did not have connection with basal lamina. Overall, morphological findings suggest that the disconnection between basal lamina and interstitium due to the absence of collagen VI may be critical in developing Ullrich disease regardless of the pattern of the collagen VI deficiency. Sequence analysis of COL6A2 gene revealed a compound heterozygous mutation in the patient with complete deficiency. Sequence analysis in partial deficiency cases is underway but they expect that mutations may be found in the globular domain rather than in the triple helical domains since connection to basal lamina appears to be lost in the partial deficiency cases.
Patrizia Sabatelli and her co-workers investigated the expression, the distribution and organization of collagen VI in muscle, skin and cultured fibroblasts derived from three patients with recessive mutations in COL6A2 or COL6A3 by immunofluorescence and immunoelectronmicroscopy.
The first patient (carrying a homozygous mutation in COL6A3 gene) was affected by a merosin positive CMD with distal laxity and fingers contractures. Muscle biopsy showed a reduced expression of collagen VI at the endomysium, while the areas affected by fibrosis and vessels showed a labelling similar to that of the normal controls. The altered expression and distribution of collagen VI were demonstrated even in two skin biopsies. In particular, the papillary dermis and the hair follicles showed a reduced amount of protein, while vessels, peripheral nerves, smooth muscle and sweat glands showed a labelling comparable with normal controls. Cultured fibroblasts from the skin showed a reduced expression of collagen VI with a dot-like appearance. They performed replicas from in vivo labelled whole-mounted cultured fibroblasts for EM study: the collagen VI microfilaments were secreted and assembled in the ECM with a reduced ability to form fibrils and networks.
The second patient (carrying mutations in COL6A2 gene) showed a typical Ullrich phenotype. The collagen VI was strongly reduced in the muscle. Skin cultured fibroblasts showed a decreased expression of collagen VI with a spot-like appearance. The characteristic collagen VI network was not detectable. The electron microscope examination of replicas obtained from in-vivo-labelled cultured fibroblasts showed the presence of microfilaments in the ECM arranged in abnormal circular arrays. The third patient, with homozygous mutations in COL6A3, also had Ullrich syndrome. Collagen VI was absent both in muscle biopsy and in cultured fibroblasts. These findings demonstrate that the expression, the distribution and organization of collagen VI are altered in the muscle, skin and cultured fibroblasts of patients carrying recessive mutations in COL6A2/COLA3 genes. The immunocytochemical analysis shows different levels of collagen VI expression ranging from mildly reduced to being absent and it was comparable in tissue and fibroblasts from each patients examined. The EM study of the microfibril structure showed an altered organization and ability to form fibrils and networks. The skin biopsy can be considered a useful diagnostic tool even in cases with collagen VI partial deficiency.
Caroline Sewry summarized the data of immunolabelling of collagen VI in muscle biopsies, performed in collaboration with Dr S. Brown (London), from the cases of Ullrich's disease presented by Francesco Muntoni. In control muscle, the basal lamina and endomysial connective tissue are labelled with collagen VI antibodies and in Ullrich the labelling of the basal lamina shows the radical change. It is, therefore, important to check the preservation of the basal lamina with another antibody to a basal lamina protein, such as collagen IV or perlecan. Two cases of Ullrich showed an unequivocal abnormality with virtually no labelling of the basal lamina. Four additional cases showed only a reduction and interpretation was more difficult as the samples showed some damage and uneven labelling of the basal lamina. A chorionic villus sample (CVS) from the fetal sibling of one affected child, with a mutation and very little labelling of collagen VI in the muscle biopsy also showed a virtual absence of collagen VI in the CVS, indicating that immunolabelling has a role in prenatal diagnosis of Ullrich's disease, particularly in cases with an absence of collagen VI. Controlling for preservation of the basal lamina should be perfomed (but not with laminin alpha 2 as this localizes to a different region in CVS) and information on the immunolabelling in the proband is advisable.
Molecular studies were reported by the following.
Pascale Guicheney reported a genome screening in a consanguineous merosine positive CMD family from Morocco with three affected children presenting proximal joint contractures and distal laxity. After identification of three potential loci for this family, analysis of other consanguineous merosin-positive CMD families led to identification of a common locus on chromosome 2q37 for five families with significant logarithm of odds (LOD) scores.
The patients of these families presented some similar phenotypic characteristics, especially joint contractures coexisting with distal laxity. Two recombination events in a patient between D2S336 and GATA5HO2 and between D2S2968 and D2S2253 defined a 13
cm interval containing a candidate gene, COL6A3, which encodes the α3 chain of collagen VI. As expected, given the autosomal recessive mode of transmission, all the affected individuals had both at-risk haplotypes, whereas several unaffected siblings had inherited one at-risk haplotype only.
Immunocytochemical analysis of available muscle biopsies with antibodies raised against collagen VI showed the absence of labelling in several patients or only a mild reduction in others. Three mutations were identified in COL6A3. A nonsense mutation was found in the triple helical domain of a Turkish patient presenting a severe phenotype and an absence of collagen by immunostaining in the muscle biopsy (H. Topaloglu). The two other mutations, a nonsense in the N-terminal domain in an Italian case (L. Merlini) and a splice site mutation inducing an in-frame deletion in the Moroccan family, were associated with much milder phenotypes. In both cases, the synthesis of abnormal alpha3 chains occurs, which is still able to bind alpha1 and alpha2 chains, and gets secreted in the extracellular compartment [17].
Analysis of other consanguineous families not linked to COL6A3, but presenting similar phenotypic features, revealed linkage to COL6A2 or COL6A1 in 21q22.3.
Ichizo Nishino reported on the molecular characterization of the mutations in the only Japanese patient with total absence of collagen VI. By direct sequencing of COL6 cDNAs obtained from muscle biopsy they detected a homozygous mutation in COL6A2 causing an exon skipping at the 3′ end of the triple helix.
Guglielmina Pepe showed the mutations detected in COL6A2 gene in three unrelated Italian patients [8]: one homozygous mutation in a family with consanguineous parents, a single nucleotide insertion in the triple helix causing a premature stop codon; one heterozygous compound, single nucleotide substitutions causing exon skipping one in the triple helix and the other in the COOH-globular domain. The third patient carries the same exon skipping in the triple helix found in the second patient. Only recently, the second mutation of the third patient has been characterized. It is a missense mutation in the triple helix of COL6A2 that is absent in 100 controls. Her group has also detected the first mutation in COL6A1 in a new Ullrich family under investigation. Experiments are in progress to confirm this new mutation.
Pascale Guicheney, Ichizo Nishino and Guglielmina Pepe discussed the severity of their Ullrich patients in relation to the mutations which have been detected. Two of the patients carrying mutations in COL6A3 had a very mild phenotype but presented with only a partial collagen VI deficiency, while the Italian and Japanese patients display a severe phenotype and no collagen VI. At present, it is not clear why the Ullrich mutations are recessive and the ones detected in Bethlem patients are dominant. Since recessive mutations have been found in all three genes, the possibility of a different expression that depends on the chain involved, as suspected by Camacho Vanegas and co-workers, is excluded. Therefore, the alternatives are that the recessive/dominant expression can be related to different domains inside each chain and to differential tissue expression, two hypotheses that are under investigation. The characterization of a large number of mutations together with the studies of mRNA/protein-mutant expression and an investigation of the biochemical and cellular phenotypes hopefully will help to clarify this topic.
The meeting ended with the establishment of the diagnostic clinical and molecular criteria for Ullrich scleroatonic muscular dystrophy due to mutations in COL6 genes reported below.
Acknowledgements
This workshop was made possible thanks to European Community grant (QLG1-CT-1999-00870) and the logistic support of the European Neuromuscular Centre (ENMC).
This work was also supported by the Association Française Contre les Maladies Musculaire (AFM) and by Fondazione Carisbo Diagnosi Riceru e Trattamento nelle Distrofie Muscolari.
Appendix A. Diagnostic criteria (Ullrich CMD due to primary collagen type VI involvement)
Appendix B. List of workshop participants
References
- . Benign myopathy, with autosomal dominant inheritance. A report on three pedigrees. Brain. 1976;99(1):91–100
- . Bethlem myopathy: a slowly progressive congenital muscular dystrophy with contractures. Brain. 1999;122(4):649–655
- . Bethlem myopathy: early-onset benign autosomal dominant myopathy with contractures. Description of two new families. Neuromuscul Disord. 1994;4:503–511
- Type VI collagen mutations in Bethlem myopathy, an autosomal dominant myopathy with contractures. Nat Genet. 1996;14(1):113–115
- Missense mutation in a von Willebrand factor type A domain of the α?3 (VI) collagen gene (COL6A3) in a family with Bethlem myopathy. Hum Mol Genet. 1998;7:807–812
- . Kongenitale atonisch-sklerotische Muskeldystrophie. Monatsschr Kinderheilkd. 1930;47:502–510
- . Kongenitale atonisch-sklerotische Muskeldystrophie, einweiterer Typus der heredodegeneration Erkrankungen des neuromuskularen Systems. Z Ges Neurol Psychiat. 1930;126:171–201
- Ullrich scleroatonic muscular dystrophy is caused by recessive mutations in collagen type VI. Proc Natl Acad Sci USA. 2001;98:7516–7521
- . Congenital muscular dystrophies: 1997 update. Brain Dev. 1998;20:65–74
- A heterozygous splice site mutation in COL6A1 leading to an in-frame deletion of the alpha;1(VI) collagen chain in an Italian family affected by Bethlem myopathy. Biochem Biophys Res Commun. 1999;258:802–807
- A novel de novo mutation in the triple helix of the COL6A3 gene in a two-generation Italian family affected by Bethlem myopathy. A diagnostic approach in the mutations' screening of type VI collagen. Neuromuscul Disord. 1999;9:264–271
- . Ehlers–Danlos syndromes: revised nosology, Villefranche, 1997. Am J Med Genet. 1998;77:31–37
- . The syndrome of myosclerosis. J Neurol Neurosurg Psychiatry. 1973;36(4):651–660
- Bethlem myopathy (BETHLEM) 86th ENMC International Workshop, 10–11 November 2000, Naarden, The Netherlands. Neuromuscul Disord. 2000;12(3):296–305
- . Collagen VI deficiency induces early onset myopathy in the mouse: an animal model for Bethlem myopathy. Hum Mol Genet. 1998;7:2135–2140
- . A mitochondrial perspective on cell death. Trends Biochem Sci. 2001;26:112–117
- Demir E, Sabatelli P, Allamand V, et al. Mutations in COL6A3 cause severe and mild phenotypes of Ullrich congenital muscular dystrophy. Am J Hum Genet 2002;70(6):1446–1458.
PII: S0960-8966(02)00139-6
© 2002 Elsevier Science B.V. All rights reserved.
