Homozygous mutations in caveolin-3 cause a severe form of rippling muscle disease.код для вставкиСкачать
Homozygous Mutations in Caveolin-3 Cause a Severe Form of Rippling Muscle Disease Christian Kubisch, MD,1 Benedikt G. H. Schoser, MD,2 Monika v. Düring, MD,3 Regina C. Betz, MD,1,4 Hans-Hilmar Goebel, MD,5 Susanne Zahn, MSc,1 Antje Ehrbrecht, MSc,1 Jan Aasly, MD,6 Anja Schroers, MD,7 Nikola Popovic, MSc,7 Hanns Lochmüller, MD,2 J. Michael Schröder, MD,8 Thomas Brüning, MD,9 Jean-Pierre Malin, MD,7 Britta Fricke, MD,3 Hans-Michael Meinck, MD,10 Torberg Torbergsen, MD,11 Hartmut Engels, PhD,1 Bruno Voss, PhD,9 and Matthias Vorgerd, MD7 Heterozygous missense mutations in the caveolin-3 gene (CAV3) cause different muscle disorders. Most patients with CAV3 alterations present with rippling muscle disease (RMD) characterized by signs of increased muscle irritability without muscle weakness. In some patients, CAV3 mutations underlie the progressive limb-girdle muscular dystrophy type 1C (LGMD1C). Here, we report two unrelated patients with novel homozygous mutations (L86P and A92T) in CAV3. Both presented with a more severe clinical phenotype than usually seen in RMD. Immunohistochemical and immunoblot analyses of muscle biopsies showed a strong reduction of caveolin-3 in both homozygous RMD patients similar to the findings in heterozygous RMD. Electron microscopy studies showed a nearly complete absence of caveolae in the sarcolemma in all RMD patients analyzed. Additional plasma membrane irregularities (small plasmalemmal discontinuities, subsarcolemmal vacuoles, abnormal papillary projections) were more pronounced in homozygous than in heterozygous RMD patients. A stronger activation of nitric oxide synthase was observed in both homozygous patients compared with heterozygous RMD. Like in LGMD1C, dysferlin immunoreactivity is reduced in RMD but more pronounced in homozygous as compared with heterozygous RMD. Thus, we further extend the phenotypic variability of muscle caveolinopathies by identification of a severe form of RMD associated with homozygous CAV3 mutations. Ann Neurol 2003;53:512–520 Hereditary rippling muscle disease (RMD, MIM 600332) is a rare autosomal dominant disorder characterized by signs of increased muscle irritability such as percussion/pressure-induced rapid muscle contractions (PIRCs), electrically silent wave-like contractions (rippling muscle), and muscle mounding on percussion. This rather benign myopathy usually is not progressive and not accompanied by dystrophic changes. Recently, we identified missense mutations in the caveolin-3 (CAV3) gene in families with autosomal dominant RMD1 and in one patient with sporadic RMD.2 Mutations in CAV3 also have been described in autosomal dominant limb-girdle muscular dystrophy type 1C (LGMD1C),3,4 in distal myopathy,5 and in children with elevated creatine kinase (hyperCKemia) without neuromuscular symptoms.6 It already has been suggested that CAV3 mutations (G55S, C71W, R125H) also might cause autosomal recessive LGMD.7,8 However, the presence of these alterations in healthy controls and the normal caveolin-3 (Cav-3) immunostaining at the sarcolemma in patients with the G55S alteration show that these alterations are benign polymorphisms.8 Evidence of autosomal recessive inheritance of RMD recently has been reported From the 1Institute of Human Genetics, University of Bonn, Bonn; 2 Friedrich-Baur-Institute, Department of Neurology, LudwigMaximilians-University, Munich; 3Department of Anatomy, RuhrUniversity Bochum, Bochum, Germany; 4Department of Medical Genetics, University of Antwerp, Antwerp, Belgium; 5Department of Neuropathology, University of Mainz, Mainz, Germany; 6Department of Neurology, University of Trondheim, Trondheim, Norway; 7Department of Neurology, Ruhr-University Bochum, Bochum; 8Department of Neuropathology, University Hospital, Aachen; 9Institute for Occupational Medicine, Ruhr-University Bochum, Bochum; 10Department of Neurology, University of Heidel- berg, Germany; and Tromso, Norway. 512 © 2003 Wiley-Liss, Inc. 11 Department of Neurology, University of Received Sep 20, 2002, and in revised form Nov 20. Accepted for publication Nov 21, 2002. Address correspondence to Dr Vorgerd, Department of Neurology, Ruhr-University Bochum, Buerkle-de-la-Camp-Platz 1, D-44789 Bochum, Germany. E-mail: firstname.lastname@example.org in two families from Oman who, in addition to PIRCs and rippling muscle waves, showed cardiac involvement, short stature, and a delayed bone age.9 This phenotype clearly differs from CAV3 mutant RMD where symptoms are restricted to the skeletal muscle. Caveolae are small plasma membrane invaginations that participate in membrane trafficking, transport, and signal transduction.10 –12 Cav-3, the muscle-specific caveolin and major protein of muscle caveolae, forms a complex with the dystrophin-glycoprotein complex. Cav-3 directly interacts with neuronal nitric oxide synthase (nNOS) by negatively regulating the catalytic activity of nNOS.13,14 We recently have shown that cytokine-stimulated NOS activity is increased in C2C12 myotubes transfected with mutant CAV3.1 In agreement with this, transgenic mice expressing the P104L CAV3 mutant in skeletal muscle showed an increase of nNOS activity in skeletal muscle.15 In contrast, a severe reduction of nNOS expression has been shown in LGMD1C4 and in Duchenne muscular dystrophy.16 In this study, we identified homozygous missense mutations in CAV3 in two unrelated patients. Both presented with an unusually severe form of RMD. Immunostaining and immunoblot analyses showed loss of Cav-3. Electron microscopy studies demonstrated an almost complete absence of caveolae in skeletal muscle of both homozygous patients similar to the findings in heterozygous RMD. Moreover, dysferlin immunostaining, studies of NOS activity, and ultrastructural studies of additional plasma membrane irregularities showed more pronounced alterations in homozygous patients compared with RMD patients with heterozygous missense mutations. Subjects and Methods Subjects The study included four unrelated patients with RMD (Table). The clinical phenotypes and mutations of the Norwegian (Patient 3, see Table) and German patient (Patient 4, see Table) with heterozygous RMD have been described earlier.1,17,18 A previously unreported patient (Patient 1, see Table) reported severe muscle stiffness in his legs since the age of 3 years. He was adopted from Colombia at the age of 18 months and his family history is unknown. In childhood, a delayed motor development was recognized, and elevated CK levels suggested a muscular dystrophy. On examination at the age of 20 years, he was athletic with generalized hypertrophic skeletal muscles, especially the shoulder, truncal, and thigh muscles. He had a moderate flexor contracture in the ankle joint and tiptoe walking, but no muscle weakness. He showed generalized PIRCs. He sometimes recognized spontaneous, short-lasting rolling muscle contractions in his thighs after heavy exercise (rippling waves), but this could not be demonstrated on several clinical examinations. Patient 2 reported a slowly progressive muscle weakness, muscle pain, and stiffness in his legs starting not before the age of 26 years. On examination at the age of 29 years, he showed weakness of proximal limb muscles, neck flexors, and ab- Table. Genotype and Clinical Phenotype of Patients with Rippling Muscle Disease Patient No., Gender, Age (yr), Origin Clinical Phenotype Mutation Inheritance Pattern Main Symptoms Ripplinga PIRCb CK (U/L) Others 1, M, 20, Colombian L86P Homozygous Severe permanent muscle stiffness (⫹) ⫹ 1,500–5,000 2, M, 29, Italian A92T Homozygous Muscle weakness, pain and stiffness ⫹ ⫹ 1526 3, M, 20, Norwegian A45T Heterozygous ⫹ ⫹ 919 4, F, 37, German A45T Heterozygous Muscle cramps and intermittent stiffness Muscle cramps and intermittent stiffness dmm; contractures of Achilles tendons; no cm, normal height Weakness of proximal limb muscles MRC grade neck flexors MRC grade 4 and weakness of abdominal muscles; no cm, normal height no cm, normal height (⫹) ⫹ 440 no cm, normal height a (⫹), Present only with stretching maneuvers on examination or reported by the patient; ⫹, easily elicited on examination; ⫹, easily elicited on examination. b PIRC ⫽ percussion/pressure-induced rapid muscle contractions; MRC ⫽ Medical Research Council (grading system of muscle weakness ranging from 0 ⫽ paralysis to 5 ⫽ full strength); dmm ⫽ delayed motor milestones; cm ⫽ cardiomyopathy; CK ⫽ creatine kinase (normal: ⬍80U/L). Kubisch et al: Caveolin-3 Gene 513 dominal muscles. In addition, he could demonstrate rippling in his quadriceps muscle and had generalized PIRCs. His parents came from Sicily, Southern Italy, and both were reported to be asymptomatic. They were not available for clinical examination nor for mutational analysis. Mutational and Linkage Analysis Genomic DNA was extracted from peripheral blood lymphocytes by a standard protocol and the two coding exons of CAV3 were amplified and directly sequenced as previously described.1 Alternative primers for exon 2 amplification were CAV3 Ex2F2: 5⬘-ctt ctg tga gtt gag gct tcc-3⬘; CAV3 Ex2R2: 5⬘-atc atg ggg tat gga gca gtc-3⬘; CAV3 Ex2F3: 5⬘-agg tta acc tga cct cta ggg-3⬘; CAV3 Ex2R3: 5⬘-cat tgt gct tct gtg gct gg-3⬘. Genotyping was performed with highly polymorphic fluorescence marked microsatellite markers. The order of markers was derived from the UCSC Genome Browser (April 2002 freeze; http://genome.ucsc.edu/) and Ensembl (http://www.ensembl.org/). Genotyping was done on an ABI 3100 automated sequencer, and alleles were scored manually using genotyper. Fluorescence In Situ Hybridization The caveolin-3–containing BAC RP11-128A5 (GenBank accession number AC068312, size 170,348bp) was hybridized simultaneously with the hybridization control PAC 196F4 (subtelomeric 3q29). BAC and PAC DNA was isolated according to a standard protocol. Probe DNA was digested with DNAse I (Roche, Mannheim, Germany). DNA fragments were purified using QIAquick PCR purification kits (Qiagen, Chatsworth, CA). Chemical labeling of the probes was performed with the Universal Linkage System (ULS; Kreatech Diagnostics, Amsterdam, The Netherlands). BAC RP11-128A5 DNA was labeled with dGreen-ULS and PAC 196F4 with Cy3-ULS. The labeling reactions were purified using QIAquick PCR purification kits and coprecipitated with ethanol in the presence of ⫻25 excess human c0t1 DNA (Gibco-BRL, Karlsruhe, Germany). The probe mixture was dissolved in 60% deionized formamide, 2 ⫻ standard saline citrate, 50mM sodium phosphate, pH 7.0, 10% dextran sulphate. Fluorescence in situ hybridization (FISH) was performed as described by Tanke and colleagues19 with minor modifications. The chromosomes were counterstained by incubation in a 4,6-diamidino-2-phenylindole-2 HCl solution (10ng/ml). After dehydration in an ethanol series, the slides were embedded in antifade solution (Citifluor AF1; Agar Scientific, Stansted, UK). Imaging and analysis were conducted on an Axioplan 2 imaging fluorescence microscope (Zeiss, Jena, Germany) equipped with a Sensys charged-coupled device camera (Photometrics, Tucson, AZ) and a Cytovision workstation (Applied Imaging, Newcastle upon Tyne, UK). Immunohistochemistry and Western Blotting Skeletal muscle sections were prepared from control tissue and RMD Patients 1, 2, and 3 (see Table) according to standard techniques. A panel of proteins was examined immunohistochemically, including caveolin-1, -2, and -3 (Trans- 514 Annals of Neurology Vol 53 No 4 April 2003 duction Laboratories, Lexington, KY), ␣-dystroglycan (Upstate Biotechnology, Lake Placid, NY), ␤-dystroglycan, dystrophin, dysferlin, laminin ␣2 (Novocastra, Newcastle upon Tyne, UK), and nNOS (Sigma, St. Louis, MO). Proteins from muscle homogenates of RMD Patients 1, 2 and 3 (see Table) were subjected to electrophoresis and Western blotting. Equal amounts of total protein were separated by electrophoresis through a 7% (for dysferlin) or 12% (for Cav-3) polyacrylamide gel and transferred to nitrocellulose membranes. Blots were incubated with monoclonal antibodies to dysferlin (dilution 1:80; Novocastra) or Cav-3 (dilution 1:5,000; Transduction Laboratories), followed by a horseradish peroxidase–labeled secondary antibody to mouse IgG (dilution 1:1,250; DAKO, Glostrup, Denmark). Blots were developed using the ECL detection system (Amersham Pharmacia Biotech, Uppsala, Sweden). NADPH Diaphorase (NDP) Activity Assay Ten-micrometer frozen muscle sections of RMD Patients 1, 2 and 3 were fixed in 2% paraformaldehyde in phosphatebuffered saline, pH 7.4, at 2 hours. After rinsing in phosphate-buffered saline, the sections were incubated in 0.2% Triton X-100 for 10 minutes at 37°C. The subsequent reaction was performed for 1 hour in a dark, humidified chamber at 37°C in 0.2% Triton X-100, 0.1mM NADPH, and 0.16mg/ml nitro blue tetrazolium solution. The reaction was terminated by washing with phosphate-buffered saline. Electron Microscopy Skeletal muscle samples of RMD Patients 1 to 4 and of controls were fixed in 2.5% glutaraldehyde and 1% paraformaldehyde in 0.1M phosphate buffer. Tissue samples were postfixed with 2% OsO4, dehydrated in graded ethanol, and embedded in Araldite. Alternating series of semithin and ultrathin (gold/silver) sections were cut on a Reichert-Jung Ultracut E. Ultrathin sections were stained with uranyl acetate followed by lead citrate and analyzed in a Philips 400 or a Philips 420 transmission electron microscope, operating at 80kV. Results Novel Homozygous CAV3 Missense Mutations in Patients with a Severe Form of Rippling Muscle Disease Patients 1 and 2 met the diagnostic criteria for RMD (see Table), which prompted us to search for CAV3 mutations. Moreover, Patient 1 had severe muscle stiffness since early childhood and contractures of the Achilles tendons leading to a gait disturbance. Patient 2 had slowly progressive muscle weakness beginning in early adulthood. In both patients, symptoms were restricted to skeletal muscles, and there was no evidence of a pathological involvement of the heart and bone. Direct sequencing of the two coding exons showed ap- Fig 1. (A) Mutations identified in exon 2 of CAV3 in two patients with rippling muscle disease (RMD). Direct sequencing showed Patient 1 carrying a homozygous T3 C base transition at nucleotide position 215 of CAV3 (top chromatograms) and Patient 2 carrying a homozygous G3 A transition at nucleotide 232 of CAV3 (bottom chromatograms). Underlined sequences highlight the wild-type codon in the normal control as well as the homozygous mutation in the patients. The corresponding amino acid change is shown above the codon. (B) Microsatellite analyses for the CAV3 locus in both homozygous RMD patients. In the first column, the marker and gene names are given in telomeric-centromeric order; the second column shows the physical position of the microsatellites and the CAV3 gene according to the UCSC Genome Browser (April 2002 freeze). The right two columns show the marker alleles for both patients. (C) Results of FISH analysis in Patients 1 (left) and 2 (right). Green hybridization signals of BAC RP11-128A5 (harboring the CAV3 gene) in band 3p25 of both chromosomes 3 demonstrate that there is no microdeletion of this region. Control signals in 3q29 are shown in red. parently homozygous missense mutations within exon 2 of CAV3. In Patient 1, the mutation was a T3 C transition at nucleotide position 215 (L86P; Fig 1A). A homozygous G3 A transition at nucleotide position 232 within codon 92 was found in Patient 2 (A92T; see Fig 1A). The mutations were not found in 120 German control chromosomes. These novel mutations were confirmed with two additional nonoverlapping primer pairs (not shown). Both alterations are located in the membrane-associated domain of Cav-3 and af- Kubisch et al: Caveolin-3 Gene 515 fect amino acids that are conserved in human, rat, and mouse. Because it was not possible to obtain DNA samples of further family members, we could not directly prove whether the patients are indeed homozygous for these alterations. A possible alternative would be compound heterozygosity for the missense mutations and a larger deletion on the other allele, which because of insufficiency of additional genes may be responsible for the more pronounced phenotype. To investigate this possibility, we performed microsatellite analyses and FISH on human chromosome 3p25. In Patient 1, we demonstrated homozygosity for a region between D3S1515 and D3S1286 spanning at least 9MB of genomic sequence (see Fig 1B). FISH analysis with the 170kb spanning BAC clone harboring the CAV3 gene showed two signals on 3p25 (see Fig 1C). This finding excludes a larger deletion and together with the microsatellite data supports that the identified missense mutation is homozygous by descent. In contrast, in Patient 2 we did not find homozygosity for the microsatellites. The known surrounding markers, that is, D3S1489 (approximately 160kb telomeric of CAV3) and D3S3691 (approximately 52kb centromeric), were heterozygous (see Fig 1B). FISH analysis showed two equally intense signals on 3p25 excluding a larger deletion in Patient 2 (see Fig 1C), although we cannot definitely rule out a smaller deletion harboring CAV3 or a part of it on one allele. Typing of three novel polymorphic microsatellite markers 7kb telomeric, within intron 1, and 5kb centromeric of CAV3 showed homozygosity in Patient 2 and heterozygosity in a control (not shown). Although there is no overt consanguinity of the parents in Patient 2, both come from a small town in Sicily, making it likely that the observed missense mutation is indeed homozygous and identical by descent. By Immunohistochemistry, Cav-3 Is Absent and Dysferlin Reduced in the Sarcolemma in Rippling Muscle Disease Muscle biopsies of Patients 1, 2, and 3 showed variation in fiber size with marked muscle fiber hypertrophy Fig 2. (A) Immunofluorescence analysis of caveolin-3 (Cav-3) and dysferlin in muscle biopsies of patients with homozygous and heterozygous rippling muscle disease (RMD). In all patients there are few fibers with abnormal patchy plasma membrane staining for Cav-3 and diminished but not absent dysferlin staining in the sarcolemma. (B) Immunoblot analysis of muscle biopsies from the homozygous and heterozygous RMD patients. Samples containing identical amounts of protein were subjected to immunoblotting with anti–Cav-3 and anti–dysferlin antibodies. (lane 1) Normal control; (lanes 2 and 3) homozygous RMD patients; (lane 4) heterozygous RMD patient. In all RMD patients dysferlin was normal, whereas Cav-3 was completely absent. 516 Annals of Neurology Vol 53 No 4 April 2003 and atrophic muscle fibers and an increase in central nuclei. These findings were more pronounced in homozygous patients. Signs of degeneration, for example, necrosis and connective tissue proliferation, were not observed. Skeletal muscle fibers of all RMD patients showed an almost complete loss of Cav-3 expression in the sarcolemma (Fig 2A). There was no upregulation of Cav-1 and Cav-2 (not shown). All RMD patients showed a complete loss of Cav-3 by immunoblot analysis (see Fig 2B). Dysferlin immunoreactivity was reduced in the sarcolemma in most fibers of the homozygous RMD patients. We also found a slight reduction of dysferlin in the sarcolemma in the heterozygous RMD patient that was less pronounced than in the homozygous RMD patients (see Fig 2A). In immunoblot analysis, all RMD patients showed normal expression of dysferlin (see Fig 2B). All other membrane proteins tested by immunohistochemistry were unchanged, indicating a preserved dystrophin–glycoprotein complex (not shown). NADPH Diaphorase Activity Is Increased in Rippling Muscle Disease Immunofluorescence analysis showed normal nNOS immunoreactivity in all RMD patients (Fig 3). Like nNOS, NDP activity in skeletal muscle fibers also was concentrated in the sarcolemma and colocalized with nNOS.20 In the NDP assay, almost all muscle fibers of the homozygous RMD patients showed a more intense signal in the sarcolemma. Moreover, in both homozygous patients we detected a reticular NDP-positive staining within the sarcoplasm. In the heterozygous RMD patient, the level of NDP staining in large muscle fibers was between those levels of homozygous RMD and controls, and there was no sarcoplasmic staining (see Fig 3). By Electron Microscopy, Caveolae Are Numerically Reduced in Rippling Muscle Disease In controls, skeletal muscle fibers exhibited irregularly distributed caveolae in the sarcolemma (Fig 4). Mutations in CAV3 resulted in an almost complete absence of caveolae in skeletal muscle fibers of all RMD patients. All RMD patients displayed foci of electron dense filamentous material irregularly distributed and located closely attached to the cytoplasmic face of the sarcolemma. In addition, all RMD patients showed plasma membrane irregularities. First, there were small membrane irregularities over which the basal lamina was thickened. Second, the surface area was increased because of irregular folds and ridges or finger-like processes. Third, invaginations of the sarcolemma with either flask-shaped or tubular profiles measuring from 230 to 1,000nm were present (see Fig 4). These plasma membrane irregularities and subsarcolemmal vacuoles appeared more pronounced in homozygous than in heterozygous RMD patients. Discussion This study for the first time identified pathogenic homozygous missense mutations in CAV3 in two unrelated patients presenting with an unusually severe form of RMD. In the homozygous patients, the clinical diagnosis of RMD was based on signs of increased muscle irritability. Yet, both patients presented additional symptoms that are not characteristic for autosomal dominant RMD, which suggested a distinct genetic ba- Fig 3. Immunofluorescence analysis of neuronal nitric oxide synthase and NDP assay in muscle biopsies of the homozygous and heterozygous rippling muscle disease (RMD) patients. Muscle tissue from all RMD patients showed a normal sarcolemmal staining of nNOS (top). NDP assay showed an increased sarcolemmal activity and a reticular sarcoplasmic staining in almost all myofibers of the homozygous RMD patients. In the heterozygous RMD patient, there was a similar but less intense NDP-positive staining in most of the large muscle fibers (bottom). Kubisch et al: Caveolin-3 Gene 517 Fig 4. Electron microscopy analysis of skeletal muscle fibers from normal control and RMD patients. In normal controls, caveolae appear as 50 to 80nm invaginations of the sarcolemma (arrows). In all rippling muscle disease (RMD) patients (Patients 1– 4), CAV3 mutations lead to a loss of caveolae at the sarcolemma, whereas Cav-1 induced caveolae in endothelial cells of the capillaries are present (black arrowheads in Pat. 4 panel). In all four RMD patients, irregular foci of filamentous material attached to the cytoplasmic face of the sarcolemma are present (white arrowheads). Note the subsarcolemmal vacuoles (asterisks) in Patients 1, 2, and 3. Pat. 1, ⫻46,000; Pat. 2, ⫻32,000; control; Pat. 3 and 4, ⫻60,000. 518 Annals of Neurology Vol 53 No 4 April 2003 sis. Indeed, we were able to identify apparently homozygous missense mutations in CAV3, which reflect the more pronounced phenotype. Patient 1 has been adopted in early infancy; therefore, no information about his family is available. Concerning Patient 2, we had to rely on his personal information, which does not show any neuromuscular symptoms in his parents or siblings. We therefore are not able to distinguish whether this severe form of RMD is autosomal recessive RMD or whether it is the more severe presentation of autosomal dominant RMD in homozygous patients. In any way, the RMD patients with homozygous CAV3 mutations are different from those with a recently described autosomal recessive form of RMD with additional cardiac and skeletal symptoms,9 which are not present in our patients. In this autosomal recessive form of RMD, no mutational analysis of the CAV3 gene has been published. However, the multisystemic involvement suggests a different genetic origin. We also asked whether the unique clinical and genetic findings are reflected by morphological and/or functional differences. The strong reduction of Cav-3 in immunostaining and immunoblotting was similar in homozygous and heterozygous RMD patients. These results are consistent with electron microscopy findings of nearly complete absence of caveolae in all RMD patients. This is in agreement with previous observations that in Cav-3 knockout mice, and in patients with LGMD1C caveolae are nearly absent at the sarcolemma.21–23 Therefore, pathogenic CAV3 mutations seem to cause a reduction of the Cav-3 protein and of caveolae irrespective of the degree of neuromuscular symptoms. In addition, we detected a reduction of dysferlin by immunostaining more pronounced in the homozygous RMD patients. It was shown that cav-3 coimmunoprecipitates with dysferlin and that in LGMD1C caused by CAV3 mutations (R26Q, T63P) dysferlin was severely reduced in the sarcolemma.24 Our data are consistent with those findings and further indicate that CAV3 mutations can affect anchoring of dysferlin to the sarcolemma. Quantitative electron microscopy of nonnecrotic muscle fibers from LGMD2B patients with primary deficiency of dysferlin have shown plasmalemmal defects, sometimes covered by thickened basal lamina, papillary projections of fibers, and small subsarcolemmal vacuoles.25 Interestingly, the plasma membrane irregularities of RMD patients we observed are very similar and more pronounced in homozygous than heterozygous subjects. Thus, the secondary alteration of dysferlin may be of functional significance and may contribute to the plasma membrane defects in RMD. We demonstrated a normal expression of nNOS at the sarcolemma in all RMD patients. This is in agreement with recent data from Cav-3 knockout mice,11 P104L transgenic mice,15 and previous studies in various other RMD muscle biopsies.1,2 Moreover, an increased NDP activity in RMD muscle was found that was more pronounced in homozygous RMD patients. In contrast, in LGMD1C and Duchenne muscular dystrophy a significant reduction of nNOS expression has been described.4,16 Interestingly, in mdx mice that expressed normal levels of NO in muscle by an nNOS transgene, the muscular dystrophy was ameliorated.26 Therefore, it can be speculated that normal nNOS expression and increased NO production may be functionally important in modifying the phenotype of caveolinopathies. All patients with autosomal dominant RMD and CAV3 mutations showed signs of increased muscle irritability, and the clinical course is usually benign. Both homozygous patients reported here were clinically more severely affected than the hitherto described heterozygous RMD patients.17,18,27 This severe form of RMD overlaps with the clinical symptoms of slowly progressive weakness in LGMD1C.3,4 We suggest that caveolinopathies present as a clinical continuum where signs of increased muscle irritability and muscle weakness are the main findings. These symptoms may either occur alone or appear simultaneously, possibly depending on the type of CAV3 mutation and modifying factors such as dysferlin and nNOS activity. This work was supported by the Deutsche Forschungsgemeinschaft, BONFOR, and the WiMed Bergmannsheil, Bochum. R.C.B. holds a postdoctoral position at the Flemish Fund for Scientific Research (FWO-Vlaanderen). We thank the patients for participation. We are very grateful to S. Galuschka, S. Böhm, L. Augustinowski, K. Knippschild, I. Goebel, A. Stiller, M. Bousfia, G. Reifenberg, M. Schlie, and I. Warlo for excellent technical assistance. We also thank Dr K. Ricker for critical reading of the manuscript. References 1. Betz RC, Schoser BGH, Kasper D, et al. Mutations in Cav3 cause mechanical hyperirritability of skeletal muscle in rippling muscle disease. Nat Genet 2001;28:218 –219. 2. Vorgerd M, Ricker K, Ziemssen F, et al. A sporadic case of rippling muscle disease caused by a de novo caveolin-3 mutation. Neurology 2001;57:2273–2277. 3. Minetti C, Sotgia F, Bruno C, et al. Mutations in the caveolin-3 gene cause autosomal dominant limb-girdle muscular dystrophy. Nat Genet 1998;18:365–368. 4. Herrmann R, Straub V, Blank M, et al. Dissociation of the dystroglycan complex in caveolin-3 deficient limb girdle muscular dystrophy. Hum Mol Genet 2000;9:2335–2340. Kubisch et al: Caveolin-3 Gene 519 5. Tateyama M, Aoki M, Nishino I, et al. Mutation in the caveolin-3 gene causes a peculiar form of distal myopathy. Neurology 2002;58:323–325. 6. Carbone I, Bruno C, Sotgia F, et al. Mutation in the CAV3 gene causes partial caveolin-3 deficiency and hyper-CKemia. Neurology 2000;54:1373–1376. 7. McNally EM, de Sa Moreira E, Duggan DJ, et al. Caveolin-3 in muscular dystrophy. Hum Mol Genet 1998;7:871– 877. 8. DePaula F, Vainzof M, Bernardino ALF, et al. Mutations in the caveolin-3 gene: when are they pathogenic? Am J Med Genet 2001;99:303–307. 9. Koul RL, Chand RP, Chacko A, et al. Severe autosomal recessive rippling muscle disease. Muscle Nerve 2001;24: 1542–1547. 10. Kurzchalia TV, Parton RG. Membrane microdomains and caveolae. Curr Opin Cell Biol 1999;11:424 – 431. 11. Ranzani B, Lisanti MP. Caveolin-deficient mice: insights into caveolar function and human disease. J Clin Invest 2001;108: 1553–1561. 12. Galbiati F, Razani B, Lisanti MP. Caveolae and caveolin-3 in muscular dystrophy. Trends Mol Med 2001;7:435– 441. 13. Michel T, Feron O. Nitric oxide synthases: which, where, how, and why? J Clin Invest 1997;100:2146 –2152. 14. Stamler JS, Meissner G. Physiology of nitric oxide in skeletal muscle. Physiol Rev 2001;81:209 –237. 15. Sunada Y, Ohi H, Hase A, et al. Transgenic mice expressing mutant caveolin-3 show severe myopathy associated with increased nNOS activity. Hum Mol Genet 2001;10:173–178. 16. Brenman JE, Chao DS, Xia H, et al. Nitric oxide synthase complexed with dystrophin and absent from skeletal muscle sarcolemma in Duchenne muscular dystrophy. Cell 1995;82: 743–752. 520 Annals of Neurology Vol 53 No 4 April 2003 17. Torbergsen T. A family with dominant hereditary myotonia, muscular hypertrophy, and increased muscular irritability, distinct from myotonia congenita Thomsen. Acta Neurol Scand 1975;1:225–232. 18. Vorgerd M, Bolz H, Patzold T, et al. Phenotypic variability in rippling muscle disease. Neurology 1999;52:1453–1459. 19. Tanke HJ, Wiegant J, van Gijlswijk RP, et al. New strategy for multi-color fluorescence in situ hybridization: COBRA: COmbined Binary RAtio labelling. Eur J Hum Genet 1999;7:2–11. 20. Kusner LL, Kaminski HJ. Nitric oxide synthase is concentrated at the skeletal muscle endplate. Brain Res 1996;730:238 –242. 21. Hagiwara Y, Sasaoka T, Araishi K, et al. Caveolin-3 deficiency causes muscle degeneration in mice. Hum Mol Genet 2000;9: 3047–3054. 22. Galbiati F, Engelman JA, Volonte D, et al. Caveolin-3 null mice show a loss of caveolae, changes in the microdomain distribution of the dystrophin-glycoprotein complex, and T-tubule abnormalities. J Biol Chem 2001;276:21425–21433. 23. Minetti C, Bado M, Broda P, et al. Impairment of caveolae formation and T-system disorganization in human muscular dystrophy with caveolin-3 deficiency. Am J Pathol 2002;160: 265–270. 24. Matsuda C, Hayashi YK, Ogawa M, et al. The sarcolemmal proteins dysferlin and caveolin-3 interact in skeletal muscle. Hum Mol Genet 2001;10:1761–1766. 25. Selcen D, Stilling G, Engel AG. The earliest pathologic alterations in dysferlinopathy. Neurology 2001;56:1472–1481. 26. Wehling M, Spencer MJ, Tidball JG. A nitric oxide synthase transgene ameliorates muscular dystrophy in mdx mice. J Cell Biol 2001;155:123–131. 27. Ricker K, Moxley RT, Rohkamm R. Rippling muscle disease. Arch Neurol 1989;48:405– 408.