Secondary reduction in calpain 3 expression in patients with limb girdle muscular dystrophy type 2B and Miyoshi myopathy (primary dysferlinopathies)

Neuromuscular Disorders 10 (2000) 553±559

www.elsevier.com/locate/nmd

Secondary reduction in calpain 3 expression in patients with limb girdle muscular dystrophy type 2B and Miyoshi myopathy (primary dysferlinopathies) Louise V.B. Anderson a,*, Ruth M. Harrison b, Robert Pogue a, Elizabeth Va®adaki b, Christine Pollitt b, Keith Davison a, Jennifer A. Moss a, Sharon Keers b, Angela Pyle b, Pamela J. Shaw c, Ibrahim Mahjneh d, Zohar Argov e, Cheryl R. Greenberg f, g, Klaus Wrogemann f, g, Tulio Bertorini h, Hans H. Goebel i, Jacques S. Beckmann j, Rumaisa Bashir b, Kate M.D. Bushby b a

Neurobiology Department, University Medical School, Framlington Place, Newcastle upon Tyne NE2 4HH, UK b School of Biochemistry and Genetics, University of Newcastle upon Tyne, Newcastle upon Tyne, UK c Neurology Department, University Medical School, Newcastle upon Tyne, UK d Division of Neurology, Kainuun Central Hospital, 87140 Kajaani, Finland e Department of Neurology, Hadassah University Hospital, Jerusalem 91120, Israel f Department of Pediatrics and Child Health, University of Mannitoba, Winnipeg, Manitoba R3E 0W3, Canada g Department of Biochemistry and Medical Genetics, University of Mannitoba, Winnipeg, Manitoba R3E 0W3, Canada h Wesley Neurology Clinic, 1211 Union Avenue, Memphis, TN 38104, USA i Department of Neuropathology, Johannes Gutenberg-University Medical Center, 55131 Mainz, Germany j Centre de National GeÂnotypage, 91057 EÂvry, France Received 10 December 1999; received in revised form 6 April 2000; accepted 10 April 2000

Abstract Dysferlin is the protein product of the gene (DYSF) that is defective in patients with limb girdle muscular dystrophy type 2B and Miyoshi myopathy. Calpain 3 is the muscle-speci®c member of the calcium activated neutral protease family and primary mutations in the CAPN3 gene cause limb girdle muscular dystrophy type 2A. The functions of both proteins remain speculative. Here we report a secondary reduction in calpain 3 expression in eight out of 16 patients with a primary dysferlinopathy and clinical features characteristic of limb girdle muscular dystrophy type 2B or Miyoshi myopathy. Previously CAPN3 analysis had been undertaken in three of these patients and two showed seemingly innocuous missense mutations, changing calpain 3 amino acids to those present in the sequences of calpains 1 and 2. These results suggest that there may be an association between dysferlin and calpain 3, and further analysis of both genes may elucidate a novel functional interaction. In addition, an association was found between prominent expression of smaller forms of the 80 kDa fragment of laminin a2 chain (merosin) and dysferlin-de®ciency. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Dysferlin; Calpain 3; Laminin; Merosin; Muscular dystrophy; Muscle proteins; Western blotting

1. Introduction Limb girdle muscular dystrophy (LGMD) may have either autosomal dominant or recessive inheritance. The recessive kinds are in two groups: those involving the sarcoglycan complex of dystrophin-associated proteins (a-, b-, gand d-sarcoglycanopathy, previously also known as LGMD2D, 2E, 2C and 2F), and those that do not. Among the non-sarcoglycan forms, the genes for LGMD2A and 2B * Corresponding author. Tel.: 144-191-222-5728; fax: 144-191-2225227. E-mail address: [email protected] (L.V.B. Anderson).

have been cloned, while those for LGMD2G, 2H, and the very recent 2I [1], are not yet isolated [2]. Limb girdle MD type 2A is unusual because the gene involved (CAPN3) is that of calpain 3, the muscle-speci®c member of the calcium-activated neutral protease family [3]. Most of the other muscular dystrophy-related proteins are structural rather than enzymatic in nature. The LGMD2B gene is DYSF, which encodes dysferlin, a protein homologous to the nematode spermatogenesis factor fer-1. This gene is also mutated in patients with Miyoshi myopathy (MM) [4,5], although an additional locus for a Miyoshilike phenotype has been localized to chromosome 10 [6]. LGMD2B is characterized by onset of symptomatic

0960-8966/00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0960-896 6(00)00143-7

554

L.V.B. Anderson et al. / Neuromuscular Disorders 10 (2000) 553±559

muscle weakness in the proximal pelvifemoral muscles while in patients with MM the onset of weakness is restricted to the calf muscles. In both diseases the age of onset is usually late teens, with good early athletic prowess and fairly slow progression. A wheelchair may be needed 10±20 years or more after the onset of weakness. Scapular muscle involvement is minor and not present at the onset. Upper limb girdle involvement in LGMD2B follows some years after the onset in lower limbs. In MM, the muscles involved may (rarely) stay restricted to the posterior compartment of the lower leg, or may gradually spread to the proximal muscles and upper limbs [2,7,8]. Some large families may demonstrate both phenotypes in the same sibship [9±11]. Patients with LGMD2B and MM generally have dif®culty standing on their toes, whereas patients with LGMD2A may have dif®culty standing on their heels, due to contractures at the ankles. LGMD2A is generally an atrophic disease with pronounced early scapular winging. The hip abductors are spared, even late in the clinical course, and mild contractures may be present elsewhere in the body. Onset is generally between 8 and 15 years of age, with possible con®nement to a wheelchair 11±29 years later [2,12]. In Newcastle, we are employing a protein-based approach to diagnosis and mutation detection in the limb girdle dystrophies [13]. By labelling all biopsies with all the available antibodies to the known MD gene products we are able to achieve a good indication of the probable diagnosis and an indication of where to start the search for gene mutations. In addition, we are starting to de®ne secondary patterns of altered protein expression. These are important because they can lead to the de®nition of new clinical entities [14] or re¯ect unexpected associations in existing diseases (e.g. laminin B1 chain in autosomal dominant Emery±Dreifuss MD and Bethlem myopathy [15]. Here we have looked at the expression of various proteins in patients with primary dysferlin de®ciency and a clinical diagnosis of either limb girdle muscular dystrophy type 2B or Miyoshi myopathy.

2. Materials and methods 2.1. Patients Patients with LGMD2B or MM were clinically assessed, and the pattern of muscle involvement characterized, by neurologists in the UK, USA, Canada, Italy (Palestinian patients), Israel, Germany and Turkey. All of the clinical presentations were in accord with the consensus descriptions of either LGMD2B with proximal lower girdle onset, or MM with lower distal onset [16]. Clinical details for many of the patients have been published previously [4,7±9,11]. Patients were diagnosed as having a primary dysferlinopathy if linkage to the dysferlin locus was established (in large families only), if a DYSF gene mutation had been detected, or if dysferlin was the most severely de®cient

protein among those tested and the patients had the appropriate clinical presentations. Blood samples and muscle biopsies were taken as part of the routine diagnostic procedure and the gene and protein analysis was undertaken in Newcastle-upon-Tyne, UK. 2.2. Protein analysis Multiplex immunoanalysis on western blots from biphasic polyacrylamide gels was performed as described previously [17] using antibodies to dysferlin (1/300 NCLhamlet [18] exon 53) and to two epitopes on calpain 3 (1/10 Calp3d/2C4 exon 1 and 1/10 Calp3c/12A2 exon 8 [19]). The antibodies were raised to synthetic peptides so the precise epitopes are known to within 16±19 amino acids. Other antibodies used included those for dystrophin (1/50 Dy8/ 6C5 C-terminus and 1/50 Dy4/6D3 rod domain), a-sarcoglycan (1/10 Ad1/20A6), b-dystroglycan (1/200 43DAG/ 8D5), g-sarcoglycan (1/2 35DAG/21B5), and 1/1000 Chemicon MAB 1922 which labels the 80 kDa fragment of the laminin a2 chain of merosin. Myosin heavy chain staining on the postblotted gel is used to indicate how much actual muscle protein (as opposed to fat and ®brous connective tissue) is loaded in each lane. Dysferlin and calpain 3 analysis was also undertaken for muscle from SJL mice, a naturally occurring model for human dysferlinopathies, in which strain a 171 bp deletion in the dysferlin gene has been identi®ed [20]. The Chemicon 1922 antibody does not react with mouse laminin a2 chain on blots. 2.3. Genetic analysis Mutation analysis was undertaken on lymphocyte DNA using a combination of single stranded conformational polymorphism and heteroduplex analysis, followed by automated sequencing, as described previously [4,21,22]. 3. Results 3.1. Immunoanalysis Fig. 1 shows a pair of multiplex blots that were labelled with our current cocktail of monoclonal antibodies. The sample in lane 7 shows a clear reduction in all the calpain 3 bands (at 94, 30 and ,60 kDa [19]) in the presence of normal labelling of all the other proteins including dysferlin. This pro®le indicates a diagnosis of LGMD2A for this patient. To date, 15/15 patients with genetically con®rmed LGMD2A have all shown normal dysferlin labelling. Lane 3 illustrates muscle from a patient with LGMD2B where only dysferlin expression is affected. This patient has a homozygous frameshifting mutation in the DYSF gene and most muscle samples from such patients show very faint dysferlin immunolabelling on blots with the NCL-hamlet antibody [18]. In patient lanes 2 and 5, dysferlin labelling is reduced to a mere trace while the calpain 3 bands are all severely

L.V.B. Anderson et al. / Neuromuscular Disorders 10 (2000) 553±559

reduced. Thus the proposed primary dysferlin de®ciency is more marked than the secondary reduction in calpain 3. To date, 8 out of 16 patients studied with a primary dysferlinopathy have also shown a less severe secondary reduction in

555

calpain 3 expression. Two of these eight patients have frameshifting mutations in the DYSF gene and one has a probable multi-exonic deletion. One further patient with a frameshifting DYSF mutation had normal calpain 3 expres-

Fig. 1. Multiplex Western blot analysis in a group of eight patients and a non-disease control (C). Blot A was labelled with a cocktail of monoclonal antibodies to the C-terminus of dystrophin (400 kDa), dysferlin (230 kDa), exon 1 of calpain 3 (94 and 30 kDa), a-sarcoglycan (50 kDa) and b-dystroglycan (43 kDa). Blot B was labelled with a different cocktail comprising antibodies to the rod domain of dystrophin (400 kDa plus metabolites down to about 120 kDa), exon 8 of calpain 3 (94 and ,60 kDa), the laminin a2 chain of merosin (80 kDa, where the arrow indicates a prominent lower doublet), b-dystroglycan (43 kDa) and g-sarcoglycan (35 kDa, plus reactivity at 120 kDa with an unidenti®ed singlet/doublet). Antibodies to b-dystroglycan are used on both blots as an internal control. The lowest panel shows the corresponding myosin heavy chain (MHC) band from one of the post-blotted gels. The diagnoses for the patients in lanes 1, 4, 6 and 8 were not resolved on these blots, but these lanes serve to demonstrate the protein pro®les in disease-controls. Patients in lanes 2, 3 and 5 are thought to have primary dysferlinopathies, while the patient in lane 7 is thought to have a primary calpainopathy.

556

L.V.B. Anderson et al. / Neuromuscular Disorders 10 (2000) 553±559

sion. Genetic analysis in the other patients is ongoing. Three patients with known missense mutations, and more dysferlin protein than those with frameshift mutations, showed normal expression of calpain 3. In SJL mouse muscle, in which the deletion is in-frame, there are low levels of dysferlin [20] and apparently normal levels of calpain 3 using the exon 8 antibody. Furthermore, bands corresponding to the 80 kDa fragment of the laminin a2 chain of merosin are of a lower size than normal and rather prominent (e.g. Fig. 1 patient lane 2 (arrowed), 3, and 5). These multiple bands are commonly seen in patients with many forms of muscle disease (and are therefore considered a non-speci®c observation) but they are usually fainter than normal and it is uncommon to see them expressed in such a conspicuous manner. In two patients with dysferlin mutations and a secondary reduction in calpain 3, the 80 kDa fragment appeared even more intensely labelled, almost overexpressed (see also band 7 in Figure 3 in Ref. [17]). 3.2. Mutation analysis To date, mutations have been sought and found in seven of the sixteen patients showing dysferlin de®ciency, and an eighth patient is from a family with unequivocal evidence of linkage to the chromosome 2p13 dysferlin locus. Three patients, whose calpain 3 reduction was detected prior to identi®cation of dysferlin, have been examined for mutations in the calpain 3 gene and single missense mutations have been found in two. These changes (H774D and A798E) are close together in exons 22/23 and are not thought to represent disease-causing mutations since the amino acids are changed to the corresponding ones in calpains 1 and 2. Signi®cantly, a probable deletion of exons 25±29 in the DYSF gene has been identi®ed in the patient with H774D in exon 22 of the CAPN3 gene. Mutation analysis is ongoing in the rest of the patients, but with 55 exons in the dysferlin gene and 24 exons in the calpain 3 gene, it is an extremely slow task. Table 1 summarizes the results of gene and protein analysis in the patients with dysferlin-de®ciency. 4. Discussion To date, we have only observed a severe reduction in dysferlin expression in the presence of the clinical features that would indicate a diagnosis of LGMD2B or MM [2]. More than 85 patients with other muscular dystrophies or neuromuscular diseases have demonstrated normal dysferlin labelling. Traces of dysferlin labelling on a few ®bres have been observed in most patients with frameshifting mutations and this results in a very faint band on blots at the normal size of ,230 kDa. It is not currently certain whether this very low level of labelling is the result of reading frame restoration (as may occur in Duchenne MD [23]), or if the antibody is crossreacting with a homologous protein [18].

Myoferlin is one such homologue, but this protein is also expressed at the nuclear membrane, and is upregulated in mdx mouse muscle [24]. The dysferlin labelling of the antibody used here, NCL-hamlet, does not include the nuclear membrane [18] and is not up-regulated in mdx muscle [20]. Dysferlin is localized to the inner surface of the plasma membrane in skeletal muscle and the presence of multiple (at least six) C2 domains suggests that it may have functional interactions with several other proteins. The C2 domain was originally de®ned in protein kinase C as a Ca 21-binding motif of about 130 residues, but Ca 21-independent functions have also been described [25]. Single and multiple copies of C2 domains have been identi®ed in a growing number of signalling proteins that interact with cellular membranes and mediate molecular traf®cking, vesicle transportation, lipid modi®cations or protein phosphorylation [25]. Both myoferlin and otoferlin (another member of the human ferlin family of proteins) contain similar C2 domains [24,26]. Calpain 3 is a proteolytic enzyme, which may be stabilized in skeletal muscle by binding to titin [27,28]. Reported sites of immunolabelling include myo®brils and nuclei [29], and a cytosolic distribution with concentration at the plasma membrane is reported for other calpains in muscle [30]. It is not clear what the physiological functions of calpain 3 are, but a role in the IkBa/NFkB apoptotic pathway has been suggested [29]. Protein-based diagnosis of LGMD2A, using immunoblotting for calpain 3, requires a higher degree of interpretative skill than for most of the structural proteins causing a muscular dystrophy. For example, levels of calpain 3 may appear falsely reduced after muscle samples have been brie¯y thawed, or if mountant from their use as histological tissue blocks is still present on muscle samples. In these cases, it is obvious from the appearance of the bands in the sample lane that some external in¯uence is artefactually reducing labelling. The use of multiple antibodies on a blot [17] makes protein degradation much easier to detect because of a downwards shift in the pro®le of bands, indicating an increase in molecular species with lower than normal molecular mass. Furthermore, we have found that a small number of patients with proven defects in other genes may demonstrate reduced levels of calpain 3. After analysis of more than 300 patient samples, reduced calpain 3 levels (without evidence of degradation) have been observed in ®ve patients with Duchenne MD and in three with facioscapulohumeral MD. In the face of these interpretative dif®culties, why might the calpain 3 reduction in LGMD2B/MM be signi®cant? Firstly, the overall pro®le of bands in these patients does not indicate protein degradation as a likely cause. Secondly, the patient biopsies come from different countries, thereby eliminating local factors such as sample handling, thawing, contamination with a particular mountant etc. Thirdly, the reduction in calpain 3 expression was detected with two different antibodies to epitopes in different domains, which makes it unlikely that the reduced expression was

Palestine

Turkey

UK

Pakistan/UK USA USA Germany Canada Canada Canada Palestine Israel

UK UK UK

LGMD2B

LGMD2B

LGMD2B

LGMD2B MM MM MM MM Preclinical LGMD2B LGMD2B LGMD2B

LGMD2B LGMD2B LGMD2B

? ? ? ? Homozygous P791R in exon 24 Homozygous P791R in exon 24 Homozygous P791R in exon 24 Linked to 2p13 Homozygous 5245delG, stop at aa1663 in exon 44 ? ? ?

?

Q611X in exon 20; 6352±6353insA, stop aa1996 in exon 53 Homozygous 543014ins23bp (exon 45) stop at aa1721 Homozygous probable Y838R1058del (exons 25±29)

DYSF gene analysis

? ? ?

? ? ? ? Missense Missense Missense ? Frameshift 1/2 1/2 2

1/2 1/2 (Fig. 1 lane 2) 1/2 1/2 (Fig. 1 lane 5) 11 111 11 1/2 1/2 (Fig. 1 lane 3)

1/2

1/2

Deletion ?

1/2

2

Dysferlin protein analysis

Frameshift

Nonsense/frameshift

Mutation effect

1111 1111 1111

11 1 1 1/2 1111 1111 1111 1111 1111

1

1

1

1

Calpain 3 protein analysis

Missense H774D in exon 22 (heterozygous) Missense A798E in exon 23 (heterozygous) No changes found

CAPN3 gene analysis

" "

" " " " " " " " "

"

" "

" "

Laminin a2 chain (80 kDa)

Key to band labelling scale: 1111, strong; 111, moderate; 11, clearly visible; 1, faint; 1/2, very faint; 2, not visible; " " , overexpressed; " , prominent; ?, unknown, genetic analysis still in progress.

UK

LGMD2B

a

Country

Phenotype

Table 1 Gene and protein expression in patients with primary dysferlin de®ciency and secondary abnormalities in other proteins a

L.V.B. Anderson et al. / Neuromuscular Disorders 10 (2000) 553±559 557

558

L.V.B. Anderson et al. / Neuromuscular Disorders 10 (2000) 553±559

caused by a change in the amino acids that constitute an antibody binding site. Fourthly, we were intrigued to ®nd apparently innocuous missense mutations in the CAPN3 gene in two of three patients for whom the analysis was undertaken. These novel changes, close to each other in domain IV (a region for which no function has yet been assigned) replace the calpain 3 amino acids with the corresponding ones that are present in the sequences of calpains 1 and2. Among 114 reported mutations and polymorphic genetic variants, these are the only two like this [22]. Patients with either the distal onset MM or proximal onset LGMD2B phenotypes demonstrated the secondary reduction in calpain 3 expression, so it is unlikely that calpain 3 plays a role in the different distributions of muscle weakness. It is not clear how the two proteins might interact, or indeed if the laminin a2 chain of merosin is also implicated in a functional relationship with dysferlin. Most of the dysferlinopathy biopsies demonstrated prominent labelling of the laminin a2 chain bands on blots, and apparent overexpression has only been observed in two LGMD2B cases among more than 400 biopsies examined from controls and patients with a wide range of myopathies and muscular dystrophies. In the context of this conspicuous expression in dysferlin-de®cient muscle, it is interesting that the lower laminin a2 chain bands may represent re-expression of smaller embryonic isoforms in regenerating muscle ®bres [17], and the dysferlin-de®cient SJL mouse is renown for the regenerative capacity of its muscle [31]. The existence of multiple C2 domains suggests that dysferlin is highly interactive, and may associate with several other proteins. At this stage of our knowledge it is useful to report any abnormalities in protein expression because the gathering of circumstantial evidence is essential for elucidating novel functional interactions. Acknowledgements We would like to thank the clinicians who have referred additional biopsies to us for analysis. The ®nancial support of the Muscular Dystrophy Campaign, Action Research, the Association FrancËaise contre les Myopathies and the Muscular Dystrophy Association is also gratefully acknowledged. References [1] Driss A, Amouri R, Ben Hamida C, Souilem S, Ben Hamida M, Hentati F. A new locus for limb-girdle muscular dystrophy in a large consanguineous Tunisian family. Neuromusc Disord 1999;9:499. [2] Bushby KMD. Making sense of the limb-girdle muscular dystrophies. Brain 1999;122:1403±1420. [3] Richard I, Broux O, Allamand V, et al. Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell 1995;81:27±40. [4] Bashir R, Britton S, Strachan T, et al. A gene related to Caenorhabditis elegans spermatogenesis factor fer-1 is mutated in limb-girdle muscular dystrophy type 2B. Nat Genet 1998;20:37±42.

[5] Liu J, Aoki M, Illa I, et al. Dysferlin, a novel skeletal muscle gene, is mutated in Miyoshi myopathy and limb girdle muscular dystrophy. Nat Genet 1998;20:31±36. [6] Linssen WHJP, de Visser M, Notermans NC, et al. Genetic heterogeneity in Miyoshi-type distal muscular dystrophy. Neuromusc Disord 1998;8:317±320. [7] Mahjneh I, Passos-Bueno MR, Zatz M, et al. The phenotype of chromosome 2p-linked limb-girdle muscular dystrophy. Neuromusc Disord 1996;6:483±490. [8] Flachenecker P, Kiefer R, Naumann M, Handwerker M, Reichmann H. Distal muscular dystrophy of Miyoshi type. Report of two cases and review of the literature. J Neurol 1997;244:23±29. [9] Weiler T, Greenberg CR, Nylen E, et al. Limb-girdle muscular dystrophy and Miyoshi myopathy in an aboriginal Canadian kindred map to LGMD2B and segregate with the same haplotype. Am J Hum Genet 1996;59:872±878. [10] Illarioshkin SN, Ivanova-Smolenskaya IA, Tanaka H, et al. Clinical and molecular analysis of a large family with three distinct phenotypes of progressive muscular dystrophy. Brain 1996;119:1895±1909. [11] Argov Z, Sadeh M, Mazor K, et al. Cluster of dysferlin-related muscular dystrophy in Libyan Jews: clinical and genetic features. Brain 2000 in press. [12] Fardeau M, Hillaire D, Mignard C, et al. Juvenile limb-girdle muscular dystrophy. Clinical, histopathological and genetic data from a small community living in the Reunion Island. Brain 1996;119:295±308. [13] Anderson LVB. Optimised protein diagnosis in the autosomal recessive limb-girdle muscular dystrophies. Neuromusc Disord 1996;6:443±446. [14] Bushby KMD, Anderson LVB, Pollitt C, Naom I, Muntoni F, Bindoff L. Abnormal merosin in adults ± a new form of late onset muscular dystrophy not linked to chromosome 6q2. Brain 1998;121:581±588. [15] Taylor J, Muntoni F, Robb S, Dubowitz V, Sewry C. Early onset autosomal dominant myopathy with rigidity of the spine: A possible role for laminin b1? Neuromusc Disord 1997;7:211±216. [16] Beckmann JS, Brown RH, Muntoni F, Urtizberea JA, BoÈnnemann CG, Bushby KMD. Report on the 66th/67th ENMC sponsored international workshop: the limb-girdle muscular dystrophies (26±28 March 1999, Naarden, The Netherlands). Neuromusc Disord 1999;9:436±445. [17] Anderson LVB, Davison K. A multiplex western blotting system for the analysis of muscular dystrophy proteins. Am J Pathol 1999;154:1017±1022. [18] Anderson LVB, Davison K, Moss JA, et al. Dysferlin is a plasma membrane protein and is expressed early in human development. Hum Mol Genet 1999;8:855±861. [19] Anderson LVB, Davison K, Moss JA, et al. Characterization of monoclonal antibodies to calpain 3 and protein expression in muscle from patients with limb-girdle muscular dystrophy type 2A. Am J Pathol 1998;153:1169±1179. [20] Bittner RE, Anderson LVB, Burkhardt E, et al. Dysferlin deletion in SJL mice (SJL-Dysf ) de®nes a natural model for limb girdle muscular dystrophy 2B. Nat Genet 1999;23:141±142. [21] Weiler T, Bashir R, Anderson LVB, et al. Identical mutations in patients with limb girdle muscular dystrophy type 2B or Miyoshi myopathy suggests a role for modi®er gene(s). Hum Mol Genet 1999;8:871±877. [22] Richard I, Roudaut C, SaeÂnz A, et al. Calpainopathy ± a survey of mutations and polymorphisms. Am J Hum Genet 1999;64:1524± 1540. [23] Nicholson LVB. The ªrescueº of dystrophin synthesis in boys with Duchenne muscular dystrophy. Neuromusc Disord 1993;3:525±532. [24] Belt Davis D, Delmonte AJ, Ly CT, McNally EM. Myoferlin, a candidate gene and potential modi®er of muscular dystrophy. Hum Mol Genet 2000;9:217±226. [25] Nalefski EA, Falke JJ. The C2 domain calcium-binding motif: structural and functional diversity. Protein Sci 1996;5:2375±2390. [26] Yasunaga S, Grati M, Cohen-Salmon M, et al. A mutation in OTOF,

L.V.B. Anderson et al. / Neuromuscular Disorders 10 (2000) 553±559 encoding otoferlin, a FER-1-like protein, causes DFNB9, a nonsyndromic form of deafness. Nat Genet 1999;21:363±369. [27] Sorimachi H, Kimura S, Kinbara K, et al. Structure and physiological functions of ubiquitous and tissue-speci®c calpain species ± Musclespeci®c calpain, p94, interacts with connectin/titin. Adv Biophys 1996;33:101±122. [28] Herasse M, Ono Y, Fougerousse F, et al. Expression and functional characteristics of calpain 3 isoforms generated through tissue-speci®c transcriptional and posttranscriptional events. Mol Cell Biol 1999;19:4047±4055.

559

[29] Baghdiguian S, Martin M, Richard I, et al. Calpain 3 de®ciency is associated with myonuclear apoptosis and profound perturbation of the IkBa/NFkB pathway in limb-girdle muscular dystrophy type 2A. Nat Med 1999;5:503±511. [30] Spencer MJ, Tidball JG. Calpain translocation during muscle ®ber necrosis and regeneration in dystrophin-de®cient mice. Exp Cell Res 1996;226:264±272. [31] Mitchell CA, McGeachie JK, Grounds MD. Cellular differences in the regeneration of murine skeletal muscle: a quantitative histological study in SJL/J and BALB/c mice. Cell Tissue Res 1992;269:159±166.