We have studied the effect of diltiazem on injury and recovery of control and dysferlin-deficient muscle after injury in vitro and in vivo. Using osmotic shock injury (OSI) in vitro to disrupt transverse (T) tubules, we find that muscle fibers from dysferlin-null (A/J) mice are more susceptible to T-tubule disruption than controls, as measured by the diffusion of sulforhodamine B from the tubules. After performing the same OSI in a buffer containing low calcium, the rates of efflux of sulforhodamine B from A/J and control fibers are indistinguishable. Similar to low calcium, pre-treatment with diltiazem, an L-type calcium channel blocker, significantly improved the recovery of A/J fibers from OSI. Following large strain injury of the ankle dorsiflexor muscles of A/J mice treated with diltiazem, contractile function was improved, and necrosis and macrophage infiltration were suppressed, consistent with the protective action of the drug in vitro. Together, these data lead us to conclude that dysregulated calcium entry following muscle injury is detrimental to the recovery of dysferlinopathic muscle and that treatment with calcium channel blockers may be useful in reducing short term pathological changes in murine dysferlinopathies.

Inhibition of p38 MAPK Ameliorates MD in Mouse Models of Duchenne, Limb-Girdle, and Dysferlin Deficiency

Jeffery D Molkentin¹

¹Cincinnati Children's Hospital and Howard Hughes Medical Institute

Neuroendocrine signaling through cytokines and inflammatory factors enhances myofiber ROS generation and deleterious signaling pathways to initiate myofiber death.Here we show that p38a, a kinase in the greater mitogen-activated protein kinase (MAPK) signaling network, serves as a nodal regulator of disease in dystrophic muscle.We deleted p38a in the skeletal muscle of d-sarcoglycan (sgcd) null mice and in mdx mice using the MLC1f-cre knock-in to examine its role in MD. We observed a significant reduction in pathology and an increase in muscle performance. A similar reduction in pathology in skeletal muscle of sgcd-/- and mdx mice was observed when treated with a p38a/ß pharmacologic inhibitor over 9 weeks. To determine if this disease pathway was also functioning in A/J mice we treated 4 month-old mice with p38a/ß inhibitor, which similarly showed protection. Finally, we generated MKK6-transgenic mice to produce constitutive p38 activation in skeletal muscle, which showed severe muscle wasting with the hallmarks of MD.These data suggest a novel therapeutic approach since the class of p38a/ß inhibitor used is known to be safe in humans from numerous clinical trials.

MG53 Protein Modulates Therapeutic Cell Membrane Repair in Skeletal Muscle

Noah Weisleder^1,2,, Bo He³, Norio Takizawa², Peihui Lin¹, Tao Tan², Christopher Ferrante², Hua Zhu¹, Pin-Jung Chen², Rosalie Yan², Matthew Sterling¹, Xiaoli Zhao¹, Moonsun Hwang¹, Chuanxi Cai¹, Heping Cheng⁴, Ruiping Xiao⁴, Hiroshi Takeshima⁵, Xiao Xiao³, Jianjie Ma^1,2

¹Department of Physiology, Robert Wood Johnson Medical School, Piscataway, NJ; ²Division of Protein Therapy, TRIM-edicine, Inc, North Brunswick, NJ; ³Division of Molecular Pharmaceutics, University of North Carolina School of Pharmacy, Chapel Hill, NC; ⁴Institute of Molecular Medicine, Peking University, Beijing, China; ⁵Department of Biological Chemistry, Kyoto University Graduate School of Pharmaceutical Sciences, Kyoto, Japan

In previous studies we established that mitsugumin 53 (MG53) is an essential component of the cell membrane repair machinery in striated muscle fibers. MG53 interacts with dysferlin and caveolin-3 and appears to be required for the translocation of certain vesicles to sites of sarcolemmal membrane disruption. Our recent work shows that in addition to the intracellular action of MG53, acute injury of the cell membrane leads to exposure of lipid signals to the extracellular space that can be detected by MG53, allowing recombinant MG53 protein to repair membrane damage when provided in the extracellular space. In the current study we tested the translational value of recombinant human MG53 protein in treatment of limb girdle muscular dystrophy type 2B (LGMD2B). Application of MG53 to individual flexor digitorum brevis muscle fibers isolated from dysferlin null mice (129-Dysf^tm1Kcam/J) can minimize entry of FM1-43 dye entry into fibers injured by UV laser irradiation. Intravenous injection of MG53 into A/J mice prior to cardiotoxin injury also indicates recombinant MG53 may provide some protective effects. Further studies indicate that genetic overexpression of MG53 can alleviate the symptoms of muscular dystrophy resulting from the absence of delta-sarcoglycan. AAV8-mediated gene transfer of the human MG53 gene can rescue skeletal and cardiac defects in dystrophic TO-2 hamsters. This improvement in dystrophic pathology was accompanied by decreased entry of FM1-43 dye into isolated muscle fibers during laser wounding assays. Furthermore, MG53 gene transfer also enhanced dysferlin expression and facilitated its trafficking to muscle membranes in a caveolin-3 dependent manner. Our data support the concept of targeting cell membrane repair in regenerative medicine, and indicate both recombinant MG53 protein and genetic overexpression of the MG53 gene have potential as therapeutic agents for treatment of LGMD2B.

Mstn/Dysf Double Knockout Mice Gain Muscle Mass but no Strength

Simone Spuler¹, Stephanie Adams¹, Verena Schöwel¹, Christian Herrmann¹, Isabelle Richard², Se-Jin Lee³

¹Charité Medical Faculty, Berlin, Germany; ²Genethon, Evry, France; ³Johns Hopkins School of Medicine, Baltimore, MD, U.S.

Mutations in myostatin lead to a massive increase in muscle mass suggesting potential relevance of myostatin inhibition for therapeutic treatment of muscle wasting. Contradictory data in mdx and laminin a2 deficient mice have been published debating on the effect of myostatin inhibition. We asked how a myostatin knock-down would affect dysferlin deficient BLA/J mice (B6.A/J-Dysfprmd). We obtained Mstn-/- and BLA/J – double knockout mice (DKO) and compared them to the single mutants and wildtype. We obtained individual muscle mass, body composition, treadmill running, single fiber contraction, grip strength and histology from all four groups. Mstn-/- and DKO mice equally gained muscle mass but differed significantly in physiological muscle function. In treadmill analysis BLA/J mice slowly worsened compared to WT and Mstn-/- while DKO showed a significantly reduced running distance and a higher rate of drop outs very early in their exercise program. Histological analysis is in concordance to the physiological tests in that DKO mice displayed increased numbers of central nucleated, necrotic and regenerating fibers. These results indicate that permanent myostatin knock-out aggravates the course of dysferlinopathy and might not be a promising therapy. We soon hope to explain the negative additive effects of myostatin and dysferlin on muscle function.

Increased Sphingosine 1-Phosphate Ameliorates Muscle Degeneration in Drosophila Mutants of Muscular Dystrophy

Hannele Ruohola-Baker², Mario Pantoja², Nicholas Ieronimakis¹, Karin A. Fischer², Junlin Qi², Morayma Reyes¹

¹Department of Pathology, Univ. of Washington, Seattle, WA, USA; ²Department of Biochemistry, Univ. of Washington, Seattle, WA, USA

Since Duchenne Muscular Dystrophy (DMD) is a muscle wasting disease for which there is no cure or effective treatment, new advances in treatment are imperative. Muscular dystrophy models in genetically tractable organisms like Drosophila offer the ability to identify interacting genes. We have shown that deletion of dystrophin in Drosophila results in DMD-like phenotypes; age dependent loss of muscle activity, muscle wasting and defective pattern of Titin-like sarcomere protein, Projectin.We screened for dystrophin (dys) suppressors and identifiedwunen, a lipid phosphate phosphatase, LPP3 that inactivates the bioactive lipid, Sphingosine 1-Phosphate (S1P). Reduction of Drosophila LPP3 rescues the defective Projectin pattern in dysmutant sarcomeres, suppresses muscle wasting and increases climbing and general activity of the dystrophic flies.Furthermore, increasing S1P levels by reducing S1P lyase, Sply or by upregulating lace, a serine palmitoyl-CoA transferase, also leads to suppression of dystrophic muscle degeneration.Oral delivery of 2-acetyl-4(5)-tetrahydroxybutyl imidazole (THI), an S1P lyase inhibitor, to dystrophic flies phenocopies the genetic suppression observed with Splyreduction.Thus, upregulation of the S1P pathway suppresses dystrophic muscles defects in Drosophila. Importantly, suppression of muscular dystrophy defects by S1P is evolutionarily conserved since injections of S1P or THI significantly increase muscle regeneration in mdxmice. S1P based therapy may be beneficial to treat dysferlinopathy as indicated by beneficial effects observed in the AJ/SCID mouse model.

SESSION II: WHAT GOES WRONG WHEN DYSFERLIN IS ABSENT?

Manifestation of Oxidative Stress in Dysferlin Deficient Muscles of A/J Null Mice

Miranda D Grounds¹, Jessica Terrill^1,2, Peter Arthur², Hannah Radley-Crabb¹

¹Schools of Anatomy and Human Biology, and ²Biomedical, Biomolecular & Chemical Sciences,the University of Western Australia, Perth, Australia.

This study addresses the role of oxidative stress in dysferlinopathy, in different muscles of A/J mice that lack dysferlin. Many muscular dystrophies exhibit elevated oxidative stress, although this has not been formally considered for the dysferlinopathies. The effects of oxidative stress were measured in dysferlin-deficient muscles, with a focus on reversible thiol oxidation combined with thiol proteomics and irreversible protein damage (carbonylation), in A/J mice aged 8, 12 and 19 months old.Histological analysis of muscles from the oldest A/J dysferlin-deficient mice (19 months) was used to determine severity of the dystropathology in the different muscles. Control normal A/J mice (at 8 months) were also analysed.Because clinical onset of dysferlinopathy usually occurs soon after cessation of growth, this suggests that different molecular mechanisms may be used for sarcolemmal resealing of growing myofibres (dysferlin independent?) compared with mature adult muscles (dysferlin dependent). The hypothesis that growing myofibres have different properties to mature adult myofibres is discussed with respect to dysferlinopathies. If this hypothesis is true, then the need to use ‘mature’ rather than ‘immature’ (e.g. myotubes) muscle for experimental models is emphasised, to identify the precise role of dysferlin and best targets for potential therapeutic interventions for dysferlinopathies.

A Bio-Mechanichal Approach to Understand the Role of Caveolin-3 and Dysferlin Mutations inHuman Muscular Dystrophy

Gillian Butler-Browne^1,2,3, Cédric Blouin², Bidisha Sinha Sinha¹, Anne Bigot³, Pierre Nassoy¹, Vincent Mouly³, Christophe Lamaze²

¹1Université P. et M. Curie/ CNRS UMR 168, Institut Curie, Centre de Recherche, Laboratoire Physico Chimie Curi; ²CNRS UMR 144, Institut Curie, Centre de Recherche, Laboratoire Trafic, Signalisation et Ciblage Intracellulaires; ³INSERM U974, UMR 7215 CNRS, UM 76 Université Pierre et MarieCurie,Institut de Myologie

Caveolae have been associated with numerous functions that still remain debated.Using a unique combination of cell biology and physical approaches, we established a new function for caveolae as mechanosensors in endothelial cells. We show that upon mechanical stress, caveolae flatten out into the plasma membrane to provide additional membrane, and thereby, buffer acute surges of membrane tension. We have extended our investigations to muscle specific caveolin-3 and asked whether caveolin-3 could play a similar role in muscle cells. We have used human myoblasts expressing either wild type caveolin-3 or mutated forms of caveolin-3 described in limb-girdle muscular dystrophy (LGMD-1C), and the dysferlin mutations described in LGM2-2B. After differentiation, the myotubes, were either exposed to hypo-osmotic shock or stretched on a flexible membrane while measuring the membrane tension with optical tweezers. Our preliminary results show that cells expressing wild type caveolin-3 resist to osmotic shock and uni-axial stretching. In contrast, dysferlin mutants and caveolin-3 mutations show both an increase in membrane tension and a higher sensitivity to membrane rupture upon mechanical stress. These results indicate that caveolae are required for the muscle cell response to mechanical stress by conferring a mechanosensitive membrane protection.

Monitoring the Progression of LGMD2B In Vivo

Katie K Maguire¹, Thomas Rando¹

¹Department of Neurology, Stanford University, Stanford, CA, USA

One of the major limitations to the study of therapeutic agents for the treatment of muscular dystrophies such as LGMD2B is the absence of reliable markers of disease activity in the living mouse.We have bred novel dysferlin-deficient reporter mouse strains in which muscle regeneration can effectively be measured over time in vivo.The muscle regeneration reporter mouse expresses an estrogen-responsive Cre-recombinase under the control of the Pax7 locus and a luciferase reporter gene that is Cre-dependent.There is a statistically significant increase in muscle regeneration in the absence of dysferlin over time by measuring an increase in bioluminescence activity.Intriguingly, we find that this manifestation of LGMD2B occurs very early in the life of the dysferlin-deficient mouse, before histological evidence of a muscular dystrophy appears.We use conventional biomarkers of muscle disease activity to correlate luciferase activity with the disease progression.We find that as luciferase signals increase so does the appearance of both centrally nucleated fibers and newly regenerated myofibers supporting the contention that these mice recapitulate the regenerative facets of LGMD2B and can be used as a model to monitor disease progression as well as response to treatments.

Dysferlin/Myoferlin Double Null Mice Develop Progressive Myopathy

Elizabeth M McNally¹, Alexis R Demonbreun¹, Kieran Deveaux¹, Manuel Alvarez², Peter Pytel²

¹Department of Medicine, The University of Chicago, Chicago, IL USA; ²Department of Pathology, The University of Chicago, Chicago, IL

In mice, loss of dysferlin is associated with features of muscular dystrophy that include myofiber necrosis and increased fibrofatty infiltration.The naturally occurring dysferlin null allele found in the A/J strain was bred to the 129SvJ background through 6 generations to generate mice in a homogeneous background (Dysf129).Dysf129 mice were then bred to mice lacking myoferlin on the similar 129SvJ background to generate mice lacking both myoferlin and dysferlin, herein referred to as Fer mice.Overall comparison of muscle morphology at different ages shows that Dysf129 and Fer animals display progressive muscle damage with myofiber necrosis, internalized nuclei and, at older ages, chronic remodeling.These changes were most prominent in proximal muscles of the hindlimb, paraspinal muscles and abdominal muscles, and these features are generally more severe in Fer mice compared to Dysf129, and the pathological features in both of these mutants was always more severe than what was seen in myoferlin null mice.Marked fatty infiltration was seen in some muscle groups in the hindlimb.These mice may prove a useful tool for the understanding of ferlin proteins.

Dysferlin/Myoferlin Double Null Mice Show Abnormal Formation of Tubular Aggregates

Peter Pytel¹, Alexis R Demonbreun², Kieran Deveaux², Manuel Alvarez¹, Elizabeth M McNally^2,3

¹Department of Pathology, Univ. of Chicago, Chicago, IL, USA; ²Department of Medicine, Univ. of Chicago, Chicago, IL, USA; ³Department of Human Genetics, Univ. of Chicago, IL, USA

The precise role for dysferlin within cells has been debated. Dysferlin is a membrane-associated protein that is found at the plasma membrane of muscle and also associated with vesicular structures within muscle. Dysferlin is also expressed outside of the mature myofiber and is found in myoblasts as well as non-muscle cells such as those of the immune system. Several different mouse models of dysferlin deficiency have been established; consistently the phenotype is comparatively mild with pathological signs including fiber necrosis and centrally nucleated fibers. Tubular aggregates are regions of densely packed membranous tubules believed to derive from sarcoplasmic reticulum. They are associated with cases of "tubular aggregate myopathy" and are a secondary feature of other myopathic conditions. Dysferlin localizes to tubular aggregates. We performed immunohistochemical staining for SERCA and ultrastructural studies in mice lacking dysferlin, ("Dysf mice"), myoferlin ("MKO mice") or both dysferlin and myoferlin ("Fer mice"). Tubular aggregates are found most frequently in Fer mice, are present in lower numbers in MKO mice and are absent in Dysf mice as well as age-matched littermate controls. We hypothesize that the loss of multiple ferlin family members exacerbates the myopathic process leading to tubular aggregation.

The Dual Role of Exercise in Dysferlin Myopathy

¹Olivier BIONDI, ¹Alice MARCHAND,¹Marie VILLEMEUR,^1,2Fabrice CHRETIEN,⁴Nathalie BOURG,^1,3Romain K. GHERARDI, ⁴Isabelle RICHARD, ^1,3François-Jérôme AUTHIER

¹INSERM U955-E 10 Cell Interactions in Neuromuscular System, Paris Est-Créteil University, France;²Human Histopathology and Animal Models, Institut Pasteur, Paris, France; ³ Reference Center for Neuromuscular Diseases Garches-Necker-Mondor-Hendaye, Henri Mondor Hospital, APHP, Créteil, France; ⁴ Généthon-CNRS UMR8587 LAMBE, Evry, France

Dysferlinopathies are characterized by various clinical phenotypes/modes of presentation without clear phenotype-genotype correlation. A high proportion of patients are very active prior to the onset of symptoms, if not top-level sportspeople. We evaluated the impact of physical strains on dysferlinopathy course by using the dysferlin- deficient A/J and B6.A/J-Dysf^prmd (Bl.A/J) mouse models. The myopathy in Bl.A/J mouse resembled to human LGMD2B and the severity of dysferlin myopathy depended on locomotor activity. By subjecting animals to various types of exercise, we showed that exercise is either deleterious or beneficial according to the type of contraction. Indeed, long-term eccentric exercise (forced running) exacerbated both the pattern and the course of dystrophic process, while long-term concentric exercise (swimming) had no deleterious effect and allowed training-induced improvement of muscle strength. One-off forced running exercise protocols showed that eccentric exercise-induced myofibers necrosis was delayed in an age dependent manner in Bl.A/J mice. Electron microscopy study showed that delayed necrosis of myofibers in Bl.A/J mice was due to the accumulation of membrane damages through injuring exercise only. Our results could explain, at least in part, both the phenotypic diversity and the clinical course of dysferlinopathies, and highlighted the potential benefits of concentric exercise in the management of dysferlin deficient patients.

SESSION III: WHAT DOES DYSFERLIN LOOK LIKE?

Progress Towards the 3D Structure of Human Dysferlin

R. Bryan Sutton¹, Kerry Fuson¹, Nathan Quisenberry¹, Sandra T. Cooper²

¹Department of Cell Physiology and Molecular Biophysics, Texas Tech University Health Sciences Center, Lubbock, TX, USA; ²Institute for Neuroscience and Muscle Research, The Children's Hospital at Westmead, Sydney, Australia

We are currently studying the 3D structure and biophysical characteristics of the C2 domains that make up human dysferlin. Primarily, we rely on X-ray crystallographic techniques; however, we are also investigating the biophysical properties of the disease-causing mutations in dysferlin. To that end, we have evidence that the V67D mutation in C2A attains a molten globule-like state, and this specific misfolding event could explain the amyloid fiber formation observed by some labs. We have recently crystallized the C2A_V1 splice variant of human dysferlin, and we are in the process of determining the 3D structure of this alternative domain. In addition, we are adapting relatively new concepts in protein crystallization to the C2 domains of dysferlin. First, we are using a series of surface entropy optimized maltose-binding protein fusions with C2D and C2F. Alone, these domains are difficult to purify; but, using MBP as a crystallization scaffold, the likelihood of determining the structure of these domains is much improved. We are employing nanobodies to facilitate the crystallization of C2E.

Dysferlin Forms a Dimer In Vitro and In Living Cells

Renzhi Han¹, Li Xu¹

¹Department of Cell and Molecular Physiology, Univ. of Loyola, Maywood, IL, USA

Dysferlin plays an important role in muscle membrane repair. The defective dysferlin-gene leads to the development of adult-onset muscular dystrophies including limb girdle muscular dystrophy type 2B and Miyoshi's myopathy. The information of dysferlin structure, particularly in its fulllength, is very limited. Here we demonstrate that endogenous dysferlin forms a dimer in detergent-solubilized muscle homogenate using sucrose gradient fractionation, blue-native PAGE and cross-linking assays. Furthermore, fluorescence resonance energy transfer (FRET) occurs between ECFP-dysferlin and EYFP-dysferlin in living HEK293 cells, which suggested dysferlin forms a self-associated complex. During photobleaching EYFP-dysferlin, a linear relationship between the fluorescence intensities from ECFP-dysferlin and EYFP-dysferlin indicates that they are also present in a dimeric form in live cells. Finally, we demonstrate that the transmembrane domain and at least one of its C2 domains are involved in mediating the dimerization of dysferlin. Interestingly, the first C2 domain does not form a self-associated complex. Our results indicate that dysferlin forms a homodimer in vitro and in living cells, revealing a possible mechanism of dysferlin functional regulation.

SESSION IV: WHAT DOES DYSFERLIN DO?

An In Vitro Injury to Elucidate the Role of Dysferlin

Andrew Ziman¹, Jaclyn P Kerr¹, Joseph Roche¹, Robert J Bloch¹

¹Department of Physiology, University of Maryland, Baltimore

Results of studies from our laboratory suggest that dysferlin in healthy muscle is concentrated in transverse tubules (TT).We have used hypo-osmotic shock to apply uniform membrane stress to single adult FDB myofibers to probe the role of dysferlin under controlled conditions in vitro.We assessed the effects of osmotic shock injury (OSI) on TT structure and function in isolated control and A/J myofibers by measuring: (i) the diffusion of sulforhodamine B, a cell impermeant fluorescent dye, from the TT; and (ii) [Ca2+]i transients.A/J fibers showed slower dye diffusion after OSI than controls.Although electrically triggered Ca2+i transients are similar in control and A/J fibers under basal conditions, and these transients quickly returned to pre-injury levels after OSI of control cells, OSI caused prolonged disruption of [Ca2+]i transients in A/J fibers, as well as increases in baseline [Ca2+]i.Our results are consistent with a role for dysferlin in protecting TT membranes from injury or promoting their rapid recovery from injury in vitro.They further suggest that OSI is a reliable method for studying the role of dysferlin in healthy and dysferlinopathic muscle.

Characterization of the Biochemical and Therapeutic Properties of Recombinant Dysferlin

Kevin J Sonnemann¹, Colin P Johnson², William M Bement¹

¹Department of Zoology, Univ. of Wisconsin, Madison, WI, USA; ²Howard Hughes Medical Institute and Department of Neuroscience, Univ. of Wisconsin, Madison, WI, USA

Although several lines of evidence suggest dysferlin acts as a calcium sensor involved in muscle membrane resealing, there are minimal biochemical data to support such a role.We have expressed and purified full-length recombinant dysferlin protein to better understand how dysferlin binds calcium, lipids, and SNARE proteins to mediate membrane fusion.Using a series of spectroscopy-based assays, we found that recombinant dysferlin can bind membranes both in the presence and absence of calcium depending on the lipid composition of membrane.Dysferlin bound membranes containing phosphatidylserine in a calcium-dependent manner while dysferlin bound phosphatidylinositol (4,5)-bisphosphate–containing membranes independently of calcium.The calcium and lipid-binding properties of dysferlin are analogous to other C2 domain-containing proteins involved in membrane fusion and support a role for dysferlin in the process.We have also begun to test the ability of recombinant TAT-dysferlin to restore proper membrane repair in dysferlin-deficient muscle fibers.Using the standard laser-wound assay, initial experiments indicated fibers incubated with TAT-dysferlin exhibited less FM1-43 dye-uptake upon damage compared to control fibers.These preliminary data support dysferlin replacement as a potential protein-based therapy in the treatment of dysferlinopathies.

Global Characterization of Dysferlin-Deficient Myoblasts and Myotubes: Comparison of Transcriptome, Proteome, and Secretome with Dystrophin-Deficient and Wild-Type Controls

William J Duddy¹, Tatiana Cohen¹, Stephanie Duguez¹, Terence A Partridge¹

¹Center for Genetic Medicine, Children's Research Institute, Children's National Medical Center, Washington, DC, USA

Dysferlin is primarily expressed in cardiac and in skeletal muscle, where it is detected early during embryogenesis.It is present in a proportion of quiescent myoblasts, in all activated myoblasts, and at a greater level following fusion into myotubes.Pathology due to dysferlin-deficiency is poorly understood but may result from altered behavior of myoblasts and/or myotubes, with membrane repair and immune response being implicated.We seek to establish a general view of altered behavior of these cells by characterizing aberrant gene transcript, intracellular protein levels, and extracellular protein secretion, of in vitro cultures.Since dystrophinopathy and dysferlinopathy share certain features, including the implication of membrane fragility, yet are phenotypically distinct, dystrophin-deficient cells can serve as a disease control to identify features specific to dysferlin-deficiency or, conversely, to identify shared aberrant cell behaviors.We have crossed A/J and mdx mice with the immortomouse to generate conditionally immortalized myoblast lines.The chief findings of a microarray gene expression analysis will be discussed, including A/J-specific changes in extracellular matrix and myofibril contractile components and shared A/J and mdx perturbations of chemokine signaling.Proteome and secretome analyses using mass spectrometry with stable isotope labeling will follow, facilitating integration into a more global perspective.

Search for New Dysferlin Partner by Two-Hybrid Screens

Isabelle Richard¹, Gaëlle Blandin¹, Sylvie Marchand¹, Karine Charton¹, Nathalie Danièle¹, Fanny Noulet¹, Jean-Baptiste Boucheteil¹, Azzédine Bentaib¹, Evelyne Gicquel¹, Laetitia Barrault¹, Marc Bartoli^1,2

¹ Généthon, CNRS, 1, rue de l’Internationale, 91000 Evry, France ² Hôpital de la Timone and Inserm UMR910, Université de la Méditerranée, 13000 Marseille, France

Mutations in Dysferlin, a protein of the ferlin family implicated in membrane repair, are the cause of Limb-girdle muscular Dystrophy type 2B and Miyoshi Myopathy. To identify new partners for dysferlin, we undertook a series of yeast two-hybrid experiments. Three fragments of dysferlin were used as baits: one N-ter fragment covering the first three C2 domains, a central fragment covering Fer and Dysf domains and a C-ter fragment covering the last four C2 domains. A high-complexity human skeletal muscle library constructed from fetal and adult RNA was used as prey. The bait interaction was tested against an average of 103 millions prey clones to insure a ten-fold coverage of the prey library. For each identified protein-protein interaction, we computed a confidence score [Predicted Biological Score (PBS)] in different classes by using information from clone coverage and local topology of the interaction network: PBS-A, B and C for the most reliable interactions, PBS-D for interactions with a single bait clone and PBS-E for highly connected bait. Overall, 28 preys with a PBS A, B and C and 180 with a PBS-D were identified. A comparison with the previously published immunoprecipitation study by de Morée identified 22 common proteins including desmin, titin, filamin C and AHNAK.

Investigating the Role of Anoctamin 5 in Muscle

Rumaisa Bashir¹, Usha Ramachandran¹, Gareth Marlow¹, Michelle M Maxwell², Rita Barresi³, Richard Charlton³, Ibrahim Mahjneh³

¹School of Biological and Biomedical Sciences, University of Durham, South Road, Durham, UK.; ²MassGeneral Institute for Neurodegenerative Disease, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, USA; ²Muscle Immunoanalysis Unit, University of Newcastle upon Tyne, Richardson Road, Newcastle upon Tyne, UK; ³Department of Neurology, Oulu University Hospital, Oulu, Finland.

The first non-dysferlin MM gene, MMD3 is anoctamin 5, which is also responsible for a LGMD phenotype, LGMD2L. We have suggested that ANO5 functions in the dysferlin membrane trafficking pathway because of the clinical similarities shared by dysferlin MM and non-dysferlin MM patients and because muscle from MMD3/LGMD2L is characterized by membrane breaks. We have previously reported that MMD3 patient fibroblasts exhibit defective membrane repair.ANO5 belongs to the anoctamin protein family, several members of which have been shown to function as calcium activated chloride channels. We and others have not detected calcium activated chloride currents following heterologous expression of ANO5 plasmids in non-muscle cells. Recently other functions have also been attributed to anoctamin proteins; ANO6 which is most similar to ANO5 at the molecular level, functions as a phospholipid scramblase. We have generated monoclonal antibodies using specific N- and C-terminal ANO5 peptides to analyze ANO5 expression in muscle cells and tissue. Our results so far suggest that several isoforms of ANO5 are expressed in muscle tissue and cultured cells. There is evidence of ANO5 oligomerization. ANO5 shows predominant Z-line localization and minimal co-localization with dysferlin. We will present this data and expression of ANO5 in MMD3 muscle.

Evidence of Dyferlinopathic Blood Vessels in Dysferlinopathies

Pascal Bernatchez¹, Cleo Leung¹

¹Department of Anesthesiology, Pharmacology and Therapeutics, Univ. of British Columbia, St. Paul's Hospital, Vancouver, BC, Canada

Progressive muscle weakness is the predominant clinical manifestation of Dysferlinopathies. However, we have observed that dysferlin is highly expressed in human and rodent blood vessels that perfuse skeletal muscles.Lack of dysferlin in endohelial cells, the protective cell layer that is directly in contact with circulating blood, decreases plasma membrane expression of specific endothelial membrane proteins, decreases endothelial cell adhesion and new blood vessel formation. Although only minor changes in the vasodilatory and constriction functions of dysferlinopathic blood vessels were observed, these data further add to the previously reported concept that the blood vessels of dystrophic muscles behave abnormally, which may lead to vascular permeability, inflammation and exacerbation of muscle dystrophy. Mechanistically, since the loss of dysferlin results in decreased expression of membrane-bound proteins, this suggests that dysferlin is required for other biological activities than simple membrane 'patching' events and that it regulates protein trafficking.To better rationalize dysferlin expression in non-mechanically active cells, GST-dysferlin-pulldown assays were performed and many unexpected proteins were identified as cargo of dysferlin-positive lipid ‘patches’.Together our data suggest that dysferlin may regulate muscle function in unexpected ways, which cannot be ignored if development of a cure for dysferlinopathies is sought.

VCP/UBXD1 Mediates Caveolin Endolysomal Degradation and is Disrupted in VCP Associated Muscle Disease

Conrad C Weihl¹, Hemmo Meyer²

¹Department of Neurology, Washington University School of Medicine, St. Louis, MO, USA;²Centre of Medical Biotechnology, University of Duisberg-Essen, Essen, Germany

The degradation of sarcolemmal proteins such as caveolin-3 (cav3) and dysferlin have principally been studied within the context of the secretory pathway.For example some disease associated mutations in cav3 or dysferlin lead to misfolding within the endoplasmic reticulum and rapid turnover prior to reaching the sarcolemma.In contrast, the endolysosomal degradation of cav3 and dysferlin is poorly understood.Endocytosed caveolins were presumed to reside in a unique non-degradative recycling compartment known as the caveosome.The caveosome was a distinct organelle that sequestered lipids and contained stable caveolin oligomers.However this unique organelle has recently been reexamined.Instead caveolins are ubiquitinated at the plasma membrane or early endosome where they enter the endolysosomal system for disassembly and degradation.This raises the possibility that the endolysosomal degradation of caveolins maybe dysregulated in some diseases.Indeed our studies have identified such a muscle disease.Sarcolemmal cav3 is reduced in mice and patients with inclusion body myopathy (IBM) associated with mutations in valosin containing protein (VCP).Moreover, cav1/3 and associated cargo such as dysferlin accumulate on enlarged late endosomes that fail to acidify.Using an unbiased mass spectrometry approach, we identified a unique complex of VCP/UBXD1 and caveolin.This complex was abrogated by disease causing mutations in VCP.This work identifies the molecular machinery necessary for the endolysosomal trafficking of caveolae.We suggest that VCP and UBXD1 may be pharmacologic targets aimed at stabilizing endocytosed sarcolemmal proteins.

SESSION V: Can we find new ways to study dysferlinopathy?

Histopathological but Not Pathophysiological Changes in Injured Skeletal Muscle are Consistent Across Four Different Mouse Strains with Dysferlinopathy.

Robert Bloch¹, Joseph A. Roche¹, Lisa Ru¹

¹Department of Physiology, University of Maryland School of Medicine

We compared 4 dysferlinopathic mouse strains: A/J, BlaJ, SJL/J and C57Bl/10-SJL-Dysf. The latter two, but not the former, are overtly myopathic at 3 months of age, and so are weaker.We subjected them, and appropriate controls, to repetitive large-strain lengthening contractions (LSI). All strains except BlaJ showed ~40% loss in contractile torque immediately after LSI.Three days later, SJL/J and C57Bl/10-SJL-Dysf and all controls recovered their pre-LSI torque almost completely. A/J did not fully recover. BlaJ showed only ~30% decrease in torque post-LSI and recovered to variable extents.All dysferlinopathic strains had elevated CNFs pre-injury and all but A/J had more inflammatory infiltrates (CD68+ macrophages) compared to controls.At D3, all dysferlinopathic strains showed increased necrosis and inflammation, but not more CNFs; controls were unchanged.Our results suggest that: (1) dysferlin is not required for functional recovery shortly after LSI, consistent with a role other than in sarcolemmal resealing; (2) BlaJ mice recover from LSI erratically, suggesting heterogeneity; (3) SJL/J and C57Bl10-SJL-Dysf muscles recover from injury rapidly, perhaps because they are already regenerating in response to ongoing myopathy; (4) although they recover function to different levels, all 4 dysferlinopathic strains show increased inflammation and necrosis following LSI.

Muscle Growth and Regeneration in Mouse and Man

Terence A Partridge¹, Bill Duddy¹, Helen Johnston¹, Tatiana Cohen¹, Aditi Phadke¹

¹Center for Genetic Medicine, CNMC, Washington, DC, USA

Onset of clinical disease in dysferlinopathies appears to be related to age, occurring most frequently at the end of the juvenile growth period.In the mouse, our commonest experimental model, growth occurs in two sharply divided phases, an initial period involving satellite cell proliferation followed by growth entirely by expansion of the territory around each nucleus. The change between these two corresponds precisely with the onset of disease in the mdx, but not of the dysferlin-deficient myopathies.In man, we have little reliable information, but what information we have suggests that there are not two sharply divided phases of cell proliferation and increase in cell size. Moreover, in man, unlike in any experimental animal model, muscle growth is not a continuous process, but occurs more rapidly in the perinatal and adolescent periods with a relatively slow growth period between the two.These differences should be borne in mind when transferring lessons learned from experimental animals to man.

Creating a Zebrafish Model of Dysferlinopathy

Louis M. Kunkel¹,Peter Serafini¹, Genri Kawahara¹, David Langenau²

¹Program in Genomics, Children’s Hospital Boston, Boston, MA, ²Molecular Pathology Unit and Center for Cancer Research, Massachusetts General Hospital, Charlestown, MA

Zebrafish are ideal lower vertebrate model for studying muscle disorders such as LGMD-2B because they are composed predominantly of skeletal muscle. We have knocked down dysferlin expression using three different antisense oligonucleotide morpholinos targeted to zebrafish dysferlin mRNA. The 4-day-old morpholino-injected fish showed phenotypes with severe alterations in somitic structure and muscle fiber organization as detected by birefringence assay, a technique used to analyze muscle organization at the myosepta. Indysferlin knocked down morphants, expression of dystrophin, laminin, and beta-dystroglycan were all normal, suggesting our morpholinos were specific for fish dysferlin.In our laboratory,the project to create a knockout model of LGMD-2B in zebrafish has been startedusing zinc finger nuclease technology. Zinc finger nucleases (ZFN) are artificial restriction enzymes that make targeted double strand DNA breaks at specific genomic loci. We have designed zinc finger arrays that have target specificity for zebrafish dysferlin. We will clone the zinc finger arrays into left and right expression vectors and make mRNA to be used for injecting into one-cell-stage fish embryos. The left and right zinc finger arrays will dimerize and bind the dysferlin targeted sequence, thereby activating the fok1 nuclease for double-strand DNA cleavage. For raising the dysferlin-null fish line to generate homozygous null fish, we will follow a breeding strategy previously described by Jeff Guyon for a splicing mutant of zebrafish dystrophin (Guyon, et al., 2009).We will characterize the dysferlin mutants by confirming the nature of the mutation via gene sequencing as well as immunohistochemistry and immunoblot analysis of dysferlin expression. We will also observe the muscle quality using motility and birefringence assays. Once characterized, we will use our dysferlin-null fish for small molecule screens.

Running and LSI protocols for Analyzing Dysferlin Deficiency

Isabelle Richard^1,Nathalie Bourg¹, Marina Pryadkina¹,François Monjarret¹, William Lostal¹, Carinne Roudaut¹, Daniel Stockholm¹, Perinne Borel¹, Joseph Roche², Robert Bloch²,

¹Généthon, CNRS, 1, rue de l’Internationale, 91000 Evry, France; ² University of Maryland Baltimore, Baltimore, MD, USA

The objective of this project is to define a test that could be predictive of a therapeutic efficiency. We undertook a comparison in wild-type and dysferlin deficient mice (B6.A/J-Dysfprmd) of different assays, including running protocols in treadmill and the large-strain injury (LSI) assay involving 15 repetitive lengthening contractions (Lovering et al. 2007). All these tests were associated with Evans Blue injection to evaluate the muscle impairment three days after the assay. The difference in the number of Evans Blue positive cells obtained with the LSI was more significant than with the running protocol. Therefore, LSI was applied to dysferlin deficient mice that were injected by dual AAV vectors to investigate whether a restoration to the normal phenotype could be obtained. The results showed that expression of dysferlin was associated with an improved phenotype after LSI, indicating that the AAV-mediated transfer of the dysferlin gene has protected the muscle from damage resulting from eccentric contraction. These results validate the suitability of this test to discriminate between treated and untreated animals.

Measurement of Membrane Resealing Kinetics

Joshua Zimmerberg¹, Glen W. Humphrey¹, Alexandr Chaturiya¹, Paul S. Blank¹

¹Eunice Kennedy Schriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda MD

Passive cable properties of muscle membranes are well described by equations for either the temporal and/or spatial response to a current pulse. We had been exploring the changes in membrane conductance of muscle membranes in response to injury using electrophysiological technique, with the goal of evaluating the membrane properties of dysferlin deficient and control myotubes and isolated FDB fibers as a function of PUFA supplementation. If the repair process occurs via a patch mechanism, then we expect the recovery following wounding to have a weak dependence upon the size of the wound; the expectation is a fast change in recovery at the time of patch insertion. However, if the repair process involves a resealing growth mechanism that occurs over time, then the recovery kinetics of membrane resistance are expected to reflect this growth process. By correlating both the recovery of membrane properties with calcium changes induced by wounding, we will address a fundamental problem in the use of laser wounding; the inability to separate the potential for wounding membrane from the potential to repair membrane wounds. Understanding the membrane recovery kinetics derived from high time resolution electrophysiological measurements coupled with variation in the laser wounding parameters (energy wounding level) may allow separation of the wounding and repair potential and the mechanism by which poly-unsaturated fatty acid membrane modification improves membrane repair.

SESSION VI: What is the role of inflammation in dysferlinopathy?

Beneficial Effect of Wound Healing Macrophages on Myogenesis in Dysferlin-Deficient Muscle Cells

Tatiana V Cohen¹,Gina Many^1,2, Bryan D Fleming³, Eva R Chin², David M Mosser³

¹Center for Genetic Medicine Research, Children's Research Institute, Washington, DC 20010;²Department of Kinesiology, University of Maryland, College Park, MD 20742; ³Department of Cell Biology and Molecular Genetics, University of Maryland, College Park, MD 20742

Dysferlin-deficient skeletal muscle in Limb-Girdle Muscular Dystrophy 2B and A/J mice is marked by myonecrosis and inflammatory foci.The predominant infiltrating cells are macrophages (Mf), which are recruited in response to signals arising from the necrotic fibers.Classically activated Mf (M1) express pro-inflammatory cytokines and are important in host defense, whereas alternatively activated Mf (M2) may perform diverse functions, including the resolution of pro-inflammatory immune responses (M2b) or wound healing (M2a).We hypothesized that Mf may influence ongoing muscle regeneration, and sought to identify whether M1 and M2 phenotypes evoke different myogenic responses.To address this, we implemented a macrophage-H2K myoblast co-culture system using A/J dysferlin-deficient and sufficient cells. Bone marrow Mf were stimulated with either: A) unstimulated, B) LPS (M1), C) LPS and Ova-immune complex (M2b), or D) IL-4 (M2a), and co-cultured in Transwells on top of differentiating myoblasts.We found that A/J dysferlin-deficient myotubes co-cultured with wound-healing Mf resulted in better myotube formation and expressed higher levels of myogenic genes than those co-cultured with unstimulated or classically activated Mf. The data suggest that wound-healing Mf active during inflammatory disease change the cytokine environment and facilitate a pro-myogenic differentiation pathway.

Complement System Activation in Dysferlinopathy

Francois-Jerome AUTHIER¹, Olivier BIONDI^1,2, Marie VILLEMEUR¹, Louise DELTOUR-FOGLIO¹, Nathalie BOURG³, Isabelle RICHARD³, Romain K GHERARDI¹

¹INSERM U955-Team 10, Paris Est-Creteil Univ., Creteil, FRANCE; ²CNRS UMR 8194, Paris V Univ., Paris, FRANCE; ³CNRS UMR8587 LAMBE, Genethon, Evry, France

The local non-immune activation of complement system leading to the formation of membrane attack complex (C5b-9) at the surface of myocytes, is a peculiar feature of dysferlin myopathy, but its significance remains unknown. We used two distinct dysferlin-deficient mouse strains: A/J (dysferlin deficiency [Dysfprmd] + C5 deficiency [Hc0]), and BlAJ (Dysfprmd backcrossed onto C57/BL6 background [normal C5 expression]). C5b-9 activation correlated with the degree of myofiber necrosis. Complement attack was modeled in vitro by subjecting murine myogenic cells to the action of human serum. Dysferlin-deficient isolated mature myofibers displayed increased sensitivity towards complement attack, a feature not related to a decreased muscle expression of membrane-bound regulators. BlAJ mice-derived myofibers were more responsive to complement attack that those derived from A/J mice. Human polyvalent immunoglobulins fully prevented complement-induced myogenic cell necrosis. Our data suggest that lytic complement activation could perpetuate muscle injuries in dysferlin myopathy and that polyvalent immunoglobulins may represent a promising therapeutic approach in dysferlinopathy.

Role of Toll-Like Receptors (TLRs) in the Pathogenesis of Dysferlin Deficiency

Kanneboyina Nagaraju¹, Beryl Ampong¹, Kitipong Uaesoontrachoon¹, Hee-Jae Cha¹, Andrea Pons¹, Jack VanderMeulen¹

¹Research Center for Genetic Medicine, Children’s National Medical Center, DC, USA

We have previously shown that functionally active toll-like receptors (TLR) are expressed in dysferlin deficient skeletal muscle. To investigate the effect of TLRs on disease phenotype, we have crossed dysferlin deficient AJ mice to myeloid differentiation primary response gene 88 (MyD88) deficient mice. MyD88 is an adapator protein of several TLRs and absence of this protein abrogates TLR signaling. We found that mice double deficient for dysferlin and MyD88 showed an improved body weight, grip strength and maximum force as compared to dysferlin deficient and MyD88 sufficient mice at 8 months of age. We next tested the hypothesis that endogenous ligands such as single stranded ribonucleic acids (ssRNA) released from damaged muscle cells bind to TLRs and perpetuate the disease progression. We found that injection of ssRNA into the skeletal muscle of presymptomatic mice (3 months old) resulted in significant inflammation in dysferlin deficient and MyD88 sufficient mice, whereas inflammation is significantly decreased in double deficient mice. These data suggest a clear role for this pathway in the pathogenesis of dysferlin deficiency and TLR antagonists may have therapeutic value in this disease.

SESSION VII: WHAT IS DYSFERLIN’S ROLE IN MEMBRANE REPAIR?

Dysferlin Deficient FDB Muscle Fibers are Repair Competent but Have a Lower Damage Threshold

Paul S. Blank¹, Glen W. Humphrey¹, Elena Mekhedov¹, Joshua Zimmerberg¹

¹Program in Physical Biology, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Bethesda, MD, USA

Both dysferlin positive (C2C12) and negative (GREG) myotubes reseal IR laser-induced membrane damage in calcium.We now report on freshly isolated and cultured (1 to 2 days) FDB muscle fibers.FDB muscle fibers were labeled with the calcium indicators Fluo-4 AM and Fura Red AM in a 1:5 molar ratio (2 mM:10 mM).The fluorescence emission of both indicators was measured, in time, following IR laser induced membrane wounding.Ratio imaging was used to evaluate the free calcium concentration; the return to baseline calcium levels was taken as wound repair.The loss of calcium homeostasis, identified as a persistent calcium response above baseline, was defined as a failure to repair.Membrane resealing occurs in both dysferlin-positive A/HeJ and dysferlin-negative A/J mouse muscle fibers, but A/HeJ fibers can reseal following wounding at higher laser wounding levels than A/J fibers. Laser wounding at a given energy level may lead to greater damage in A/J membranes, resulting in a larger inward flux of Ca2+, and possibly inhibiting resealing by exceeding a critical threshold for maintaining calcium homeostasis. We plan to evaluate potential therapies, such as lipid modification, aimed at improving membrane repair in dysferlin-deficient muscle fibers using both calcium homeostasis and dye leakage assays.

Equal Force Recovery in Dysferlin-Deficient and Wild-Type Muscles Following Saponin Exposure

Renzhi Han¹, Piming Zhao¹, Li Xu¹, Younss Ait-Mou¹, Pieter P. de Tombe¹

¹Department of Cell and Molecular Physiology, Loyola University Medical Center, Maywood, IL, USA

Plasma membrane damage occurs frequently in the life of a cell. The plasma membrane integrity can be disrupted by mechanical stress, pore-forming toxins derived from pathogens and pore-forming proteins from the host’s immune system. The cell membrane damage must be repaired to allow survival. Dysferlin plays an important role in repairing membrane damage elicited by laser irradiation, and dysferlin deficiency causes muscular dystrophy and associated cardiomyopathy. However, it is not clear whether dysferlin plays a role in repairing membrane damage induced by pore-forming reagents such as the natural detergent saponin. In this study, we found that dysferlin-deficient muscles recovered the tetanic force production to the same extent as their WT counterparts following a 5-min saponin exposure (50 µg/ml), and that the soleus muscle recovered significantly better than the extensor digitorum longus (EDL) muscle. In addition, repairing saponin-induced membrane injury is temperature dependent. Our data suggest that dysferlin unlikely participates in repairing saponin-induced membrane damage, and that the slow muscle may be more efficient than the fast muscle to repair such damage.

Annexin-A5 Two-dimensional Arrays Promote Cell Membrane Repair

Alain R. Brisson¹, Anthony Bouter¹, Céline Gounou¹, Sisareuth Tan¹, Ernst Pöschl,² Bent Brachvogel,³

¹Molecular Imaging and NanoBioTechnology, IECB, UMR-5248 CBMN CNRS-University Bordeaux-1, Talence, F-33402, France; ²School of Biological Sciences, University of East Anglia, Norwich, NR4 7TJ, UK; ³Center for Biochemistry, University of Cologne, Cologne, D-50931, Germany

Eukaryotic cells possess a repair mechanism that ensures rapid resealing of plasma membrane disruptions via an active process that requires extracellular Ca²⁺, intracellular vesicles and a multi-component protein machinery. Dysferlin has been shown to participate in the membrane fusion step of the process in skeletal muscle cells, and it is considered that, in dysferlin-linked muscular dystrophies, disrupted membranes have impaired resealing, due to a lack of dysferlin. Several dysferlin-binding proteins, including annexin-A1 and A2, have also been implicated in membrane repair. We have investigated whether annexin-A5 (AnxA5), an annexin that forms two-dimensional arrays upon membrane binding, was involved in membrane repair. We have compared the responses of wild-type mouse perivascular cells and AnxA5-null cells to membrane injury. AnxA5-null cells exhibit a severe membrane repair defect which can be restored by the addition of AnxA5. We show that AnxA5 promotes membrane repair through the formation of two-dimensional arrays at the edges of damaged membranes. This finding introduces a new step in the membrane repair process. Further work is required to elucidate the exact roles of AnxA5 and the other annexins in this mechanism and their relationship with dysferlin in membrane aggregation and fusion events.

A Role of Annexin A2 in Healing Muscle Injury

Jyoti Jaiswal¹, Sushma Medikayala¹, Arun B Deora², Manohar S Tirunagari³, Aurelia Defour¹, Luana Scheffer¹, JackVandermeulen¹, Kanneboyina Nagaraju¹

¹Center for Genetic Medicine Research, Children’s National Medical Center, Washington DC 20010; ²Department of cell and developmental biology, Weill Medical College of Cornell University, New York, NY 10065; ³Laboratory of Cellular Biophysics, The Rockefeller University, New York, NY 10065

Dysferlin is an integral membrane protein, mutations in which are associated with poor healing of injured muscle cells and dysferlinopathy. While sarcolemmal injury triggers accumulation of dysferlin at the wound site, there is a poor understanding of the cellular mechanism involved in this process. We have identified annexin A2 as a regulator of calcium-triggered dysferlin trafficking to the cell surface. Within a few seconds of cell injury, annexin A2 translocates from the cytosol to the wounded sarcolemma. Lack of annexin A2 compromises the injury-induced cell surface trafficking of dysferlin and prevents efficient healing of wounded muscle cells. These effects of annexin A2 deficiency can be mimicked by transiently inhibiting calcium-triggered cell surface trafficking of annexin A2. This demonstrates the need of calcium-triggered cell surface trafficking of annexin A2 for the localization of dysferlin to the wound site as well as for healing injured muscle cells. Knockout of annexin A2 causes increased myofiber degeneration and fibrosis and reduced muscle force, as measured in vivo and ex vivo. Despite the poor healing, annexin A2 deficient muscles do not exhibit increased inflammation, providing an opportunity to study the link between muscle injury and inflammation in dysferlinopathy.

PTRF Anchors MG53 to Cell Injury Site for Initiation of Membrane Repair

Hua Zhu¹, Pei-hui Lin¹, Gejing De¹, Kyoung-han Choi¹, Hiroshi Takeshima², Noah Weisleder¹, Jianjie Ma¹

¹Department of Physiology and Biophysics, UMDNJ, Piscataway, NJ, USA; ²Graduate School of Pharmaceutical Sciences, Kyoto Universtiy, Japan

Membrane resealing is an elemental process of cell biology and disruption of this process can lead to degenerative human diseases. Mutations in dysferlin are associated with membrane repair defects in limb girdle muscular dystrophy (LGMD) patients. Our recent study showed that MG53 can interact with dysferlin to facilitate muscle membrane repair, and defects in MG53-mediated membrane repair are linked to muscular dystrophy and cardiac dysfunction.Here, we report PTRF (polymerase I and transcript release factor), a gene previously identified in regulation of caveolae membrane structure, is an indispensible component of the membrane repair machinery. PTRF acts as a docking protein for MG53-mediated membrane repair through binding exposed membrane cholesterol at the injury site. Cells lacking endogenous expression of PTRF show defective membrane resealing. Mutation in PTRF associated with human disease alters PTRF localization in the nucleus and disrupts MG53 function in membrane resealing. While RNAi-silencing of PTRF leads to defective muscle membrane repair, overexpression of PTRF can rescue membrane repair defects indysferlin-/- muscles. Our data suggest that membrane-delimited interaction between MG53 and PTRF contributes to initiation of the cell membrane repair response, which can be an attractive target for treatment or prevention of tissue injury in LGMD. Further understanding the functional interaction between MG53, dysferlin and PTRF should provide new insights into the basic biology of membrane resealing as well as for therapeutic interventions for restoration of membrane repair defects in human diseases.

MICAL an F-Actin-Disassembly Factor Links Membrane Repair Including MG53 and Dysferlin

Katsuya Miyake¹, Youhei Egami¹, Moe Hamada¹, Yukiko K. Hayashi³, Chie Matsuda², Nobukazu Araki¹

¹Dep.Hist.& Cell Biol., Medicine, Kagawa University, Japan; ²AIST, Tsukuba, Japan; ³Dep. Neuromuscular Research, National Institute of Neuroscience, NCNP, Tokyo, Japan

Plasma membrane disruption is a common form of cell injury in mammalian tissues under physiological conditions. Cell survival depends on the initiation of a rapid (second time-scale) resealing response that is mounted only in the presence of physiological levels of extracellular Ca2+. Vesicle-vesicle and vesicle-plasma membrane fusion events occurring in cortical cytoplasm surrounding the defect are thought to be a crucial element of the resealing mechanism. We have previously shown that a requisite prelude to this fusion is the disassembly in local cell cortex of the physical barrier constituted by filamentous actin. However, the identity of the proteins involved in this disassembly event remained unknown. Recently, the mical family of enzymes have been found to associate with the cytoplasmic domain of plexins, which are large cell-surface semaphorin receptors. Mical thereby functions as an F-actin-disassembly factor mediating actin remodeling in a variety of cell responses. We expressed MICAL, MG53, dysferlin-GFPs (EGFP, mCherry or RFP) in culture cells (A431, BS-C-1, C2C12, HeLa, etc.) and then subjected them to a plasma membrane disruption created by a two-photon laser or a mechanical wounding of Piezo Micro Manipulator (PMM). Subsequent confocal imaging revealed more striking and faster (second time-scale) accumulation of MICAL-GFP at the disruption site, compared to MG53 or dysferlin-GFPs, followed by actin-GFP depolymerization (second time-scale). We also observed, for the first time in living skeletal muscle cells responding to a membrane disruption, striking accumulation of MICAL, MG53 and dysferlin GFPs at the disruption site. We hypothesize that localized filamentous actin disassembly mediated by MICAL removes a cortical barrier standing in the way of the membrane-membrane contacts leading to the homotypic and exocytotic fusion events required for MG53 and dysferlin mediated repair.

SESSION VIII: HOW DO WE PUT DYSFERLIN BACK?

A Quantitative and Mechanistic Comparison of AAV-Large Gene Delivery Strategies for the Treatment of Dysferlin Deficiency

Matthew L Hirsch¹, Chaoying Yin¹, R J Samulski¹

¹Gene Therapy Center, University of North Carolina, Chapel Hill, NC, USA

Adeno-associated viral vectors are the most promising method of gene delivery currently used in human trials; however, a major limitation of these vectors is that the AAV capsid cannot package DNA > 5 kb.To overcome this, research has demonstrated AAV-mediated delivery of gene segments which are reconstructed into a large transgene (>5kb) by host DNA repair enzymes post-transduction. Traditionally, the most efficient AAV large gene transfer strategy relies on splitting the cDNA into 2 vectors and then performing a co-transduction experiment relying on viral genome concatemerization to reconstruct the large transgene. In addition to the split gene vector technique, another AAV large gene context has been elucidated termed “fragment AAV”. Therefore, we performed a comparison of split gene versus fragment AAV vectors and demonstrate that fragment AAV is more efficient for large gene delivery, especially at lower doses in mouse muscle. In addition, it was found that the large transgene reconstruction of these viral substrates relies on distinct DNA repair factors. Large gene transfer experiments using both of the described AAV strategies are underway in dysferlin deficient mice and insights from the results will be presented as well.

1, 25(OH)2-Vitamin D3 Increases Dysferlin Expression in Human Peripheral Blood Monocytes In Vivo

Jordi Díaz-Manera^1,3, NoemiDe Luna^1,3, Ricard Rojas-García^1,3, Carmen Paradas^2,3, JosefaAraque^1,3, Mireia Genebriera^1,3, Ignasi Gich⁴, Isabel Illa^1,3, Eduard Gallardo^1,3

¹Neuromuscular Disorders Unit. Neurology Department. Hospital de Sant Payu. Barcelona;²Neurology Department, Hospital Universitario Virgen del Rocío, Sevilla, Spain; ³2Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED);⁴Epidemiology department.Hospital de Sant Pau. Barcelona

Experiments in our laboratory showed that Vitamin D3 (D3) increases the expression of dysferlin in human peripheral blood monocytes (PBM) and myotubes. Dysferlin myopathy carriers have reduced levels of dysferlin in PBM. We performed an observational study with calcifediol (vitamin D3) in a group of asymptomatic and one symptomatic carriers of one mutation in DYSF. Twenty-one relatives of patients with dysferlinopathy molecularly confirmed were included in the study. Sixteen subjects received 32.000 IU of D3 once a week for one year, whereas the remaining five did not receive any treatment. Treatment was well tolerated and any patients developed relevant side effects. Three patients gave up the study. We detected a significant increase of dysferlin expression (mean 42%) in PBM by western-blot at the end of the study (Greenhouse-Geisser test, p<0,05) only in the treated patients. However, in spite of the positive effect on dysferline expression, one year of treatment with D3 was not enough to observe changes in the muscle MRI of the single symptomatic carrier included in the study. Further studies in a higher number of patients are needed to demonstrate that early and long standing treatment can promote an effect also at the clinical level.

1 Alpha, 25(OH)2-Vitamin D3 Increases Dysferlin Expression in Human Skeletal Muscle and Monocytes In Vitro.

Eduard Gallardo^1,2, Jordi Diaz-Manera^1,2, Ricardo Rojas-García^1,2, Carmen Paradas^2,3, Josefa Araque^1,2, Mireia Genebriera^1,2, Ignasi Gich⁴, Isabel Illa^1,2, Noemí De Luna^1,2

¹Servei de Neurologia i Laboratori de Neurologia Experimental, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain; ²Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED); ³Servicio de Neurología, Hospital Universitario Virgen del Rocío, Sevilla, Spain; ⁴Servei de Epidemiologia Clínica, Hospital de la Santa Creu i Sant Pau i Institut de Recerca de HSCSP, Universitat Autònoma de Barcelona, Barcelona, Spain

We have studied the effects of 1alpha, 25(OH)2-Vitamin D3 (D3) on dysferlin expression in peripheral blood monocytes (PBM) and myotubes from controls and carriers of dysferlinopathy. The first experiments were performed on HL60 cells (ATCC, CCL-240), because it has been reported that they differentiate into mature CD14+ monocytes upon treatment with vitamin D3. The effect of D3 on cells can be mediated through two different pathways: 1) internalization of the D3 ligand/receptor complex which activates gene expression and 2) a non-genomic pathway through MAPK kinases. We observed increased expression of dysferlin by HL60 cells after 72 hours of treatment with D3. This expression was independent of cell differentiation, because treatment with UO126, a MAPK blocker, decreased the expression of dysferlin but did not modify their differentiation into CD14+ monocytes. D3 increased the expression of dysferlin in myotubes of controls and one symptomatic carrier of dysferlinopathy in a dose dependent manner. Similar results were observed in PBM both in control and carriers of one mutation in DYSF. Our study demonstrates that D3 promotes an increase of dysferlin expression in human skeletal muscle and PBM. These findings may have important implications in future therapies combining D3 and different molecular strategies.

Evaluating the Utility of Adipose- Derived Mesenchymal Stem Cells for Cell Therapy of Muscular Dystrophy.

Marisa Karow¹, W. Edward Jung¹, Christopher L. Chavez¹, Asa Flanigan¹, Michele P. Calos¹

¹Department of Genetics, Stanford University, Stanford, CA, USA

An appealing approach to address the muscle degeneration present in limb girdle muscular dystrophy 2B is to repair the affected muscle with corrected stem cells derived from the patient.Our strategy involves using recombinase genome engineering methods to carry out dysferlin gene addition in candidate stem cells.In the process of determining which type of stem cell to use, we examined the potential of adipose-derived mesenchymal stem cells (AD-MSC).These cells are abundant, expendable, relatively easily accessed, and appeared to have the ability to differentiate into muscle fibers.To test their applicability for a stem cell therapy, we isolated AD-MSC from the inguinal fat pads of mice and used FACS analysis to verify that the cells carried the expected cell surface markers.The cells were nucleofected with plasmid DNA, to introduce the dysferlin-expressing therapeutic plasmid, along with a plasmid encoding phiC31 integrase..While we obtained good transfection efficiency (30 – 60%), the cells did not grow well after nucleofection, so it was not possible to obtain sufficient corrected cells for transplantation. Moreover, since AD-MSC are primary cells, they have a limited lifespan in culture.This feature makes it impossible to work with a clone of cells having one, characterized integration site.Therefore, a population of cells having a mixture of integration sites would need to be used, which could create an insertional mutagenesis risk.An additional roadblock with AD-MSC was their poor ability to differentiate into muscle fibers, in vitro or in vivo.We were unable to demonstrate robust, reproducible differentiation of AD-MSC into muscle fibers using conventional differentiation conditions, such as co-culture with C2C12 cells.Similarly, after injection into the TA muscle of cardiotoxin-treated mice, AD-MSC showed a poor ability to engraft, unlike the myoblast positive controls.For these reasons, we believe that AD-MSCs are not a stem cell choice for use in muscle.

Dose Effect Toxicity of Minidysferlin

Isabelle Richard¹, William Lostal¹, François Monjarret¹, Nathalie Bourg¹, Carinne Roudaut¹, Marina Pryadkina¹, Marc Bartoli², Martin Krahn², Nicolas Levy²

¹Généthon, CNRS, 1, rue de l’Internationale, 91000 Evry, France; ²Hôpital de la Timone and Inserm UMR910, Université de la Méditerranée, 13000 Marseille, France

A form of minidysferlin, cloned from patient with a moderately severe clinical presentation and expressed in muscle via an AAV vector, was previously shown to be efficient in rescuing sarcolemmal repair in vitro. Considering the previously reported toxicity of dysferlin overexpression, we pergformed a dose effect study in WT mice to evaluated a possible toxic effect of minidysferlin overexpression.Morphometrical analysis of HPS-stained muscle sections showed that the muscles injected with the lowest doses were completely normal and muscle fibers injected with doses of minidysferlin which result in expression higher than endogenous dysferlin levels result in an increase in centronucleated fibers (CNFs).This observation suggests that minidysferlin can cause similar toxicity to dysferlin overexpression.

AN INTERNATIONAL STUDY OF NATURAL HISTORY IN DYSFERLINOPATHY

Kate Bushby¹, Brigitta von Rekowski¹

¹TREAT-NMD, Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, United Kingdom

We have been working with the Jain Foundation on a protocol for a natural history study of dyferlinopathy to take place over 5 years including 17 centres in Europe (UK, Spain, Germany, Italy, France, Portugal), Israel, USA, Japan and Australia who have contact with 300-400 confirmed dysferlinopathy patients. We aim to recruit a minimum of 150 patients initially (100 ambulant, 50 non-ambulant). These patients will be assessed at 6 visits over 3 years with physio and medical assessments. MRI will be performed in all patients and MRS in a subset. Blood samples will be collected from all patients, and where ethical approval and consent is obtained, skin samples will also be collected (>100 patients). These will form a valuable resource for research into the pathophysiology of dysferlinopathy, as well as potentially a source for biomarker discovery. The aim of the study is to characterize the progression of dysferlinopathy and to delineate the clinical and investigational measures appropriate for clinical trials. In addition, the study allows for the harmonization of the Jain dysferlinopathy registry with the International UMD dysferlinopathy registry, under the TREAT-NMD registries initiative, which will also form the platform for the natural history study data collection.

USING FRET AND FCS TO STUDY PROTEIN-PROTEIN INTERACTIONS IN LIVING CELLS

Steven S. Vogel¹

¹NIAAA, NIH, Rockville, MD

I will present a tutorial describing the basis of using fluorescence microscopy, specifically FRET (Förster Resonance Energy Transfer) and FCS (Fluorescence Correlation Spectroscopy), for studying protein-protein interactions in living cells. My presentation will focus on the theory behind these methods, problems encountered when implementing these techniques, and the challenge to correctly interpret FRET and FCS data. FRET measures proximity between fluorophores on a 1-10 nm scale, not 'binding'. Thus, interpreting FRET measurements as 'binding' is problematic because "False" FRET positives, due to molecular crowding, are common. This is particularly true for membrane proteins, such as Dysferlin, because they are constrained to a two-dimensional space. False negatives are also possible with FRET. False positives and negatives are seen for both Hetero-FRET (e.g. FRET between a Blue donor fluorophore and a yellow acceptor fluorophore) and for Homo-FRET (e.g. FRET between 2 yellow fluorophores), and it is very difficult to control for these 'false' FRET signals. Fluorescence correlation spectroscopy (FCS) can also be used to study protein complexes in living cells. FCS also does not measure 'binding' per se, FCS measures fluctuation in the fluorescence intensity of a protein or protein complex tagged with fluorophores. Fluctuation analysis can be used to measure the number of fluorophores in a complex (e.g. a monomer would have 1 and a dimer would have 2). FCS can also measure the diffusion coefficient (a function of the mass, shape of a protein complex, as well as the viscosity of the media) of a protein or protein complex. For soluble proteins FCS works very well for differentiating between monomers and dimers, etc. Unfortunately, if two membrane proteins are independently located on the same membrane, FCS analysis might interpret them as being dimers. Again, this is a false positive. I will describe a new method developed in my laboratory, Polarized Fluorescence Correlation Spectroscopy (pFCS) that overcomes many of these problems by combining Homo-FRET microscopy with FCS, and if time permits, I will describe how my laboratory, in collaboration with Dr. Rumaisa Bashir plan to use FRET and FCS to determine if Dysferlin forms a complex with ANO5, and if Dysferlin forms either static or dynamic homo-multimers.

Main menu

Oral Abstracts

SESSION 1: HOW DO WE COMPENSATE FOR DYSFERLIN’S ABSENCE?

Diltiazem Treatment of Dysferlin-Null Muscle Improves Recovery from Injury