In the Lambert-Eaton myasthenic syndrome (LEMS), there is a decreased release of acetylcholine quanta from the nerve terminal by nerve impulse. Recently, an autoimmune origin of LEMS was documented by passive transfer of its electrophysiologic features from man to mousewith IgG. Freeze-fracture electron microscopy of LEMS neuromuscular junctions has revealed a paucity of presynaptic membrane active zones. Thus, the active zones might be the targets of the pathogenic autoantibodies in LEMS. To test this assumption, freeze-fracture electron microscopic studies were done in mice injected with 10 mg of IgG daily from each of three LEMS patients and in control mice treated with normal human IgG or no IgG. IgG from patients 1 and 2 impaired neuromuscular transmission in mice, but IgG from patient 3 failed to do so. After 52-69 days of treatment, diaphragm or anterior tibial muscles were removed and coded. Paired muscles from control mice and mice receiving LEMS IgG were studied "blindly." Satisfactory freeze-fracture replicas of 185 presynaptic membrane P-faces were analyzed by stereometric methods. In mice treated with LEMS IgG that was pathogenic by electrophysiologic criteria, there was a selective depletion of active zones and activezone particles but not of other membrane particles and there was a concomitant increase of large membrane particles aggregated into clusters. These findings provide additional evidence that the active zones facilitate quantal transmitter release by nerve impulse, lendiurther support to the assumption that the active-zone particles are Ca2+ channels, and establish mediation of the membrane lesions in LEMS by IgG.

To investigate the nature of the inflammatory response in facioscapulohumeral muscular dystrophy (FSHD), we analyzed mononuclear cells in muscle sections obtained from 18 FSHD patients and 8 controls. Monoclonal antibodies reactive for T cells, T cell subsets, B cells, and NK cells were used for cell typing. Macrophages were identified by acid phosphatase reaction. The localization of perforin, granzyme A, MHC-I and -II, dystrophin, and alpha-actinin antigens was also examined. We found that all FSHD patients, both familiar and sporadic cases, had greater amounts of mononuclear cellular infiltrates in muscle than controls, in whose specimens only few extra vascular mononuclear cells were counted. Seventy-two percent (13 of 18) of the patients had more than 50 inflammatory mononuclear cells per 1000 muscle fibers, and 33% (6 of 18) patients had numerous inflammatory cells exceeding 600 per 1000 muscle fibers (1835 +/- 482 SE). Nonnecrotic fibers invaded by mononuclear cells with either T8+, perforin+, or granzyme A+ were not observed in FSHD, while a few degenerating fibers were superficially invaded by T cells and macrophages. Occasional T cells were observed moving through the blood vessel wall. The increased number of necrotic fibers was paralleled by an increased number of inflammatory cells (r = 0.783, P = 0.0001). Genetic analysis, using the probes p13E-11, pFR-1, D4S139, and D4S163, was done in 6 patients (3 familiar, 3 sporadic) who had numerous inflammatory infiltrates. These 6 patients had small (< 28 kb) EcoRI fragments associated with the disease, and the disease was linked to 4q35. These results suggest that, in chromosome 4-linked FSHD: (1) inflammatory changes in muscle are a common histological feature; (2) mononuclear cellular infiltrates may enhance muscle fiber damage; but (3) T-cell-mediated cytotoxicity directed against muscle fibers is unlikely. We speculate that the immune effector mechanism in FSHD is different from that in previously reported inflammatory myopathies and Duchenne muscular dystrophy.

Presynaptic membrane active zones are related to synaptic vesicle exocytosis, and the large intramembrane particles in these zones may represent voltage-sensitive calcium channels. We tested the hypothesis that an abnormality of the zones is associated with the low probability of quantal release in the Lambert-Eaton myasthenic syndrome (LEMS). Freeze-fractured presynaptic membranes were studied in nine patients with LEMS (227 end-plates) and in 14 controls (148 end-plates). Satisfactory replicas of 94 LEMS and 83 control presynaptic membrane Pfaces were obtained. Presynaptic membrane areas were estimated by stereometric analysis. The LEMS samples showed a marked decrease in active zones and active zone particles per unit area. The average number of particles per active zone was also reduced. Clusters of large particles were observed with increased frequency in the LEMS samples. These may have arisen by aggregation of active zone particles. There was no decrease in the overall density of intramembrane particles not associated with active zones or clusters. The distribution of these particles according to size resembled the control distribution except for a small decrease in the frequency of 5.3-5.8 nm particles. The findings can explain the reduced quantal release in LEMS, and strongly suggest that active zone particles are targets of the pathogenic autoantibodies recently demonstrated in this disease.

Direct infection of muscle fibers by human T-lymphotropic virus type I (HTLV-I) has recently been reported in a patient with polymyositis infected with both HTLV-I and human immunodeficiency virus (HIV). Coinfections of these viruses are frequently found in the United States. In Kagoshima, Japan, patients with polymyositis have a significantly increased incidence of seropositivity to HTLV-I alone, when compared with the general population of Kagoshima. In this study, we examined muscle tissue from 6 HTLV-I-positive patients with polymyositis from Kagoshima. To detect HTLV-I products, sensitive immunohistochemistry and in situ hybridization analysis were performed. These were compared with muscle fibers from a well-characterized transgenic mouse model which expressed HTLV-I tax. No specific signals were detected in the biopsied muscles of patients with polymyositis infected with HTLV-I alone. HIV co-infection may, therefore, augment HTLV-I expression through either immunosuppression or direct viral interactions.

We studied 71 Japanese individuals, 42 patients (30 familial and 12 sporadic) suspected to have facioscapulohumeral muscular dystrophy (FSHD) and 29 family members, clinically and genetically using the chromosome4qter DNAmarker pl3E-ll. Early onset FSHDwas detected in 7patients, tortuosity of retinal arterioles and hearing impairment in 3 patients, progressive respiratory failure in 3 patients and limb-girdle type muscular weakness in 6 patients. Thirty-six (85.7%) FSHDpatients, 3 asymptomatic family members and 1 of 35 healthy volunteers showed EcoRI digestion fragments shorter than 28kb. Newmutations were detected in 25%of the patients with shorter EcoRI fragment. The age of disease onset appeared younger with successive generations in 6 parent-child pairs in FSHDfamilies. Weconfirmed the existence of phenotypic and genetic diversities in Japanese patients with FSHD.It is still difficult to explain the phenotypic diversity merely by the size of the abnormal EcoRI fragment detected with the pl3E-ll probe.