Search

Coxsackie B viruses are members of the family Picornaviridae which have been associated by retrospective serology with a range of muscle diseases, particularly myocarditis, dilated cardiomyopathy and epidemic pleurodynia (epidemic myalgia or Bornholm disease). It has been proposed that virus-induced myocarditis disposes to the development of idiopathic dilated cardiomyopathy. However, despite many attempts, isolation of infectious virus or immunofluorescent detection of virus-specific antigens in the affected tissue is rare, although virus may be found in faeces early in infection. This discrepancy awaited the development of nucleic acid probes to resolve the problem of whether virus was present consistently in myocardium or other muscle tissues. We report here the synthesis of Coxsackie B virus-specific complementary DNA (cDNA) probes and their use in molecular hybridizations to quantitative slot-blots of RNA prepared from either endomyocardial or skeletal muscle biopsy specimens. Of 50 patients with histologically proven myocarditis or dilated cardiomyopathy, 28 (56%) had an endomyocardial biopsy specimen positive for the presence of Coxsackie B virus-specific RNA. Twenty-two patients with other cardiac diseases of known aetiology, unrelated to virus infection, were all negative. Multiple biopsies were obtained from 20 patients with myocarditis or dilated cardiomyopathy and 15 of these (75%) had at least one biopsy specimen positive, indicating the focal nature of the disease. In analogous investigations, Coxsackie B virus-specific RNA was detected in four out of seven single skeletal muscle biopsy specimens from patients suffering from juvenile dermatomyositis, and one out of two patients with adult polymyositis. Ten muscle controls, either normal or Duchenne muscular dystrophy, were negative for virus RNA.

A patient with systemic lupus erythematosus (SLE) was treated with chloroquine therapy for four years after the onset of her illness. Nine years after cessation of chloroquine, muscle weakness developed as part of the SLE. Four muscle biopsies performed for diagnostic purposes revealed varying degrees of inflammatory change as well as distinctive cytosomes with curvilinear profiles (CCPs). These CCPs were identical to those reported in Batten disease, a degenerative disorder of children which has a clinical course different from SLE. The CCPs seen in this case of SLE are thought to result from the effect of chloroquine on membrane systems within muscle cells. This report calls attention to the fact that CCPs are not unique to Batten disease bu may also occur in muscle of SLE patients treated with chloroquine.

The technique of needle muscle biopsy using the Bergstr öm needle has been in routine use in our muscle clinic since 1978. In an initial feasibility study 24 children had a needle and an open biopsy performed simultaneously through extension of the same incision and 22 had identical interpretation of the needle and open biopsies. Needle biopsies have subsequently been performed in 674 children and have been satisfactory for diagnostic assessment in 656. The samples have been of adequate size and comparable in quality to our previous open biopsies, with good preservation and orientation. Needle muscle biopsy under local anaesthetic is quicker and less traumatic than open biopsy and leaves only a very small scar. Sufficient muscle can be obtained for routine histological, histochemical, and electronmicroscopic diagnosis, as well as for specialised biochemical and research purposes. There seems little justification for the continued use of open biopsy for routine investigation of neuromuscular disease.

Lobulated fibers have been observed in biopsies of 13 patients with various neuromuscular disorders including limb-girdle muscular dystrophy (3), distal myopathy (2), scapuloperoneal muscular dystrophy (2), congenital myopathy, Kugelberg-Welander syndrome, hypothyroidism, steroid myopathy, osteomalacia and systemic lupus erythematosus (on steroids). In all cases there were fibers characterised by small subsarcolemmal triangular aggregates or more diffuse collections extending into the interior of the muscle fiber. These were strongly reactive with oxidative enzymes, acid phosphatase, periodic acid-Schiff (PAS), Verhoeff-van Gieson (VVG) and also stained red with the Gomori trichrome technique. In 5 cases core-like fibers were also seen. Morphometric analysis of the NADH-tetrazolium reductase (NADH-TR) preparations in 11 cases showed atrophy of the lobulated fibers. Ultrastructural studies of lobulated fibers disclosed large peripheral mitochondrial aggregates and focal areas with Z-line streaming and disrupted myofibrils. We consider this structural change of the muscle fiber as a reflection of muscle cell disruption and suggest that they may progress from lobulated fibers to more atrophic core-like fibers.

We have investigated 3 children aged 6, 3, and 2 years, from 2 families, with the clinical features of Marinesco-Sjögren syndrome. Muscle biopsy specimens from all 3 were abnormal and showed small vacuoles and slight variation in fiber size. Electron microscopy revealed vacuolation and membranous whorls and, in particular, a unique dense membranous structure associated with nuclei. These cases emphasize the involvement of muscle in Marinesco-Sjögren syndrome and the importance of electron microscopy in differential diagnosis.

Using an electron histochemical technique, we have observed the binding of a D-galactose-specific lectin to the muscle cell plasma membrane in muscle biopsies taken from patients with various neuromuscular disorders. In spinal muscular atrophy, the only neurogenic disease studied, the plasma membrane stained as in normal muscle. However, in the myopathies Becker and limb-girdle muscular dystrophy and in the polymyositis there was a reduction in both the occurrence and the intensity of staining of the plasma membrane. Reduced lectin binding by the plasma membrane probably reflects secondary changes in the composition of glycoproteins and/or glycolipids in the membrane and seems to be common to all these myopathies to varying degrees.

Using a combination of sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting, a high molecular weight Ricinus communis I (RCA I)-binding glycoprotein (approx. Mr 370000) has been identified in human muscle that is consistently altered or absent in muscle from patients with Duchenne muscular dystrophy (DMD). In addition, a Mr 54000 RCA I-binding glycoprotein was identified in 4 out and 8 DMD muscle samples that was not present in normal muscle. The possibility that the Mr 370000 glycoprotein could be a muscle membrane glycoprotein which is altered or absent in DMD is discussed.

Electron-microscopic X-ray microanalysis (EMMA) was used to study calcium and phosphorus concentrations in the muscle fibre nuclei of 3 genetically definite, 5 probable and 2 possible carriers of Duchenne muscular dystrophy. All cases showed elevated mean calcium:phosphorus ratios compared with 4 female controls and, with the exception of one possible carrier, all showed elevated calcium concentrations. A significantly greater variation of values was also found for both parameters. Analyses were made of nuclei from fibres showing no morphological abnormalities and the results suggest elevation of myonuclear calcium may be an early defect in dystrophic muscle.

Needle muscle biopsies from 14 genetically proven carriers of Duchenne muscular dystrophy (DMD), 14 possible carriers with elevated serum creatine kinase activity (SCK) and 37 possible carriers with normal SCK were compared with similar samples from 10 healthy volunteer females with the aim of identifying subtle changes that might aid carrier detection. Qualitative differences were similar to those previously reported and included variation in fibre size, an increase in the number of internal nuclei and fragmentation of the myofibrillar network. By visual assessment, only 11% of the biopsies were classified as unequivocally abnormal but quantitation of fibre type proportions, fibre diameters and the number of internal nuclei showed that 71% of definite carriers, 86% of possible carriers with elevated SCK and 78% of possible carriers with normal SCK had at least one feature that was significantly different from controls. The commonest abnormalities were an increase in the proportion of type 1 fibres, an alteration in the mean diameter ratio of type 1:type 2 fibres, an increase in the fibre diameter variation coefficient and an increase in the number of internal nuclei. The changes in definite carriers were not related to elevated SCK and several with normal SCK showed morphological differences. It was concluded that quantitation of morphological features is essential for a full appraisal of biopsies from DMD carriers and that this approach is a useful aid to carrier detection.

The localisation of the complement components C8 and C9 was studied immunocytochemically in human diseased muscle to determine the role of complement in muscle fibre damage. Monoclonal antibodies to 2 epitopes of C9 and a monoclonal antibody to the alpha subunit of C8 were applied to frozen sections of muscle biopsies from 9 cases of dermatomyositis, 5 cases of polymyositis, 7 cases of Duchenne muscular dystrophy and 4 cases of Becker muscular dystrophy. These were compared with 6 control biopsies which were morphologically normal. In all cases of inflammatory myopathies several non-necrotic fibres showed discrete peripheral patches of C9 and to a lesser extent C8. In the muscular dystrophies peripheral C9 was observed on a few non-necrotic fibres and basophilic fibres showed C9 between the fibres as well as at the periphery. In all cases necrotic fibres labelled intensely with C9 and C8 but intensities varied with the different monoclonal antibodies. This was thought to result from differences in the polymerisation of the C9 molecule in the membrane attack complex. Complement C8 and C9 were also localised to blood vessels in 3 cases of muscular dystrophy, 2 cases of polymyositis and all cases of juvenile dermatomyositis. No complement was observed in the control samples. Our results provide evidence for the sublytic formation of the membrane attack complex (MAC) on non-necrotic fibres in inflammatory myopathies and muscular dystrophy. This sublytic formation of the MAC may induce sublethal metabolic damage, mediated by calcium, and suggests a primary role of complement in muscle damage not only in inflammatory disorders but also muscular dystrophy.

Eight unrelated children with centronuclear (myotubular) myopathy are described, ranging in age at the time of diagnosis from 5 days to 12 years. Six had an intrauterine onset and 5 were severely asphyxiated at birth. All had facial involvement and 6 had ophthalmoplegia. Detailed study of the parents in 7 of the families suggested an autosomal recessive inheritance or sporadic occurrence in 2 and X-linked inheritance in 5. Classification in this very variable disorder should be based on severity and mode of presentation together with the genetic pattern, allowing three subgroups to be defined: a severe neonatal X-linked recessive type, a less severe infantile or juvenile autosomal recessive type and a milder autosomal dominant type. For genetic counselling, available relatives should be examined for mild degrees of clinical involvement and morphological abnormalities on needle muscle biopsy.

Coxsackie-B-virus-specific probes prepared by reverse transcription of purified virus genomic RNA and molecular cloning techniques were used in quantitative slot-blot hybridisations to test for the presence of virus RNA in skeletal muscle biopsy samples. The samples tested were from two patients with adult polymyositis, seven with juvenile dermatomyositis, four with Duchenne muscular dystrophy, and six normal controls. Of the nine patients with inflammatory muscle disease, five were positive for Coxsackie-B-virus RNA; no viral sequences were found in the ten Duchenne muscular dystrophy and control biopsy samples.

Recently, it has become clear that a structural or functional abnormality of the plasma membrane may be involved in the pathogenesis of Duchenne muscular dystrophy (DMD). We have used a D-galactose specific lectin, that binds strongly to the muscle cell surface, to search for differences in sarcolemmal staining in biopsies from 5 normal and 7 DMD individuals. Electron optical studies revealed a dramatic and consistent decrease in lectin binding by the muscle plasma membrane in DMD. Although unable at this stage to identify the specific molecular abnormality responsible for this decreased staining, the results provide good evidence that a basic structural change within the plasma membrane may be intimately associated with DMD.

The binding characteristics of lectins with varying sugar specificities were investigated in muscle biopsies from normal individuals and from patients with neuromuscular disorders. Horseradish peroxidase (HRPA) and fluorescein isothiocyanate (FITC)-conjugated lectins were used for light microscopy and ferritin-conjugated Concanavalin A (Con A) for electron microscopy. In normal and diseased muscle a lectin specific for alpha-D-mannosyl residues (Con A) and a group of lectins specific for beta-D-galactosyl residues were found to bind to the perimysial and endomysial connective tissue, blood vessels and capillaries and clearly demonstrated the perimeter of each muscle fibre. Wheat germ agglutinin (WGA), specific for N-acetylglucosamine and N-acetyl-neuraminic acid residues, had a similar distribution of staining although it appeared to stain the capillaries more strongly. In contrast, lectins specific for alpha-L-fucose and N-acetyl-galactosamine did not stain specifically any structures in normal or diseased muscle. In biopsies from dystrophic patients Con A, WGA and the beta-D-galactose specific lectins were always associated with splits and in biopsies from a variety of disorders discontinuities in staining were observed at the periphery of occasional fibers. These were not found in normal muscle. Electron microscopy showed Con A bound to the basement membrane of muscle fibres and capillaries and the connective tissue. The plasma membranes themselves were unstained. These preliminary investigations of lectin binding in muscle have shown important differences in diseased muscle and demonstrate the application of lectin chemistry to the study of membrane structure.

Fifteen lectin-horseradish peroxidase conjugates have been used in a comprehensive histochemical study of human skeletal muscle. The staining patterns of many lectins were found to be coincident with the known distributions of types I, III, IV and V collagen, fibronectin and laminin. One lectin, Bandeiraea simplicifolia (BSA I), selectively stained capillaries in a blood group-specific manner, the significance of which is unknown. The results show that although lectins are useful cytochemical probes for identifying tissue glycoconjugates, lectin binding is not solely determined by monosaccharide specificity as lectins which interact with the same sugars may have completely different staining patterns. Factors such as accessibility, glycan conformation and oligosaccharide sequence also affect lectin binding in tissues. For these reasons, we conclude that a comprehensive histochemical investigation of tissue glycoconjugates should employ a large number of lectins, preferably with overlapping sugar specificities.

An RCA I-lectin binding glycoprotein of Mr = 370 kD is missing from or altered in the plasma membrane of Duchenne muscular dystrophy (DMD) skeletal muscle. In the present study the carbohydrate chain of this glycoprotein was localized to the external face of the plasma membrane in human skeletal muscle, and dystrophin, the protein product of the DMD gene, was localized to the inner (cytoplasmic) face. On double labelled Western blots the two proteins appeared as closely apposed but distinctly separate bands. Comparison of the plasma membrane binding of five lectins with overlapping sugar specificities in skeletal muscle from patients with DMD and the allelic milder disease form, Becker muscular dystrophy (BMD) showed that the RCA I-binding glycoprotein also strongly binds to phytohaemagglutinin, thereby largely characterising the carbohydrate binding site. This glycoprotein was absent or altered in DMD and markedly reduced in clinically manifest BMD but present in preclinical clinical BMD. There was no general depletion of plasma membrane glycoproteins in DMD because consistent plasma membrane binding could be demonstrated by peanut and maclura pomifera lectin. The possible implications of these findings for the pathogenesis of DMD/BMD are discussed.

In previous studies it was shown that a D-galactose-specific lectin, Ricinus communis I (RCA I), does not bind to the plasma membrane of muscle fibres from patients with Duchenne muscular dystrophy (DMD) in contrast to normal muscle. We have now studied RCA I binding to the membranes of developing human fetal muscle in fetuses at 95% risk of DMD (n = 6) and normal controls (n = 5) with a developmental range of 12-20 weeks of gestation. The results were compared to the membrane appearance with conventional ultrastructure. Binding of RCA I to the muscle basement membrane was consistently strong from the early stages of myogenesis, such as in fusing myoblasts/myocytes. RCA I binding to the plasma membrane was weak but detectable in both DMD and normal fetuses at 12-14 weeks of gestation. Both the normal and diseased condition showed an increase of RCA I labelling of the muscle plasma membrane at 15-17 weeks and strong labelling at 18-20 weeks of gestation. No difference was observed in the RCA I localization of normal and diseased human fetal muscle plasma membrane. It is concluded that (a) the plasma membrane in developing fetal muscle undergoes a maturation process between 12 and 20 weeks gestational age leading to an increase in expression of RCA I binding carbohydrate moieties; and (b) that the absence of RCA I binding glycoprotein in mature DMD muscle plasma membrane reflects a change acquired during the course of disease.

Monoclonal antibodies to human complement component C9 were used to localise the membrane attack complex (MAC) of human complement on muscle fibres from patients with autoimmune myositis. A pigeon erythrocyte 'model' system was used to screen the available monoclonal antibodies for their ability to bind C9 after its insertion into the MAC. Three of the five antibodies tested were demonstrated, using radioiodine- or rhodamine-labelled second antibody, to bind. The antibody giving maximum binding in this system was used to demonstrate that MACs were present on histologically normal as well as on necrotic fibres in myositis. No binding was seen in muscle from control subjects. These results suggest a primary role of complement in the pathogenesis of muscle fibre necrosis in myositis.