Secondary calpain3 deficiency in 2q-linked muscular dystrophy

CME

Secondary calpain3 deficiency in 2q-linked muscular dystrophy Titin is the candidate gene

H. Haravuori*; A. Vihola, MSc*; V. Straub, MD; M. Auranen, MD; I. Richard, PhD; S. Marchand; T. Voit, MD; S. Labeit, MD; H. Somer, MD, PhD; L. Peltonen, MD, PhD; J.S. Beckmann, PhD; and B. Udd, MD, PhD

Article abstract—Background: Tibial muscular dystrophy (TMD), a late-onset dominant distal myopathy, is caused by yet unknown mutations on chromosome 2q, whereas MD with myositis (MDM) is a muscular dystrophy of the mouse, also progressing with age and linked to mouse chromosome 2. For both disorders, linkage studies have implicated titin as a potential candidate gene. Methods: The authors analyzed major candidate regions in the titin gene by sequencing and Southern blot hybridization, and performed titin immunohistochemistry on TMD patient material to identify the underlying mutation. Western blot studies were performed on the known titin ligands in muscle samples of both disorders and controls, and analysis of apoptosis was also performed. Results: The authors identified almost complete loss of calpain3, a ligand of titin, in the patient with limb-girdle MD (LGMD) with a homozygous state of TMD haplotype when primary calpain3 gene defect was excluded. Apoptotic myonuclei with altered distribution of transcription factor NF-kB and its inhibitor IkB␣ were encountered in muscle samples of patients with either heterozygous or homozygous TMD haplotype. Similar findings were confirmed in the MDM mouse. Conclusions: These results imply that titin mutations may be responsible for TMD, and that the pathophysiologic pathway following calpain3 deficiency may overlap with LGMD2A. The loss of calpain3 could be a downstream effect of the deficient TMD gene product. The significance of the secondary calpain3 defect for the pathogenesis of TMD was emphasized by similar calpain3 deficiency in the MDM mouse, which is suggested to be a mouse model for TMD. Homozygous mutation at the 2q locus may thus be capable of producing yet another LGMD. NEUROLOGY 2001;56:869 –877

Tibial muscular dystrophy (TMD; MIM 600334) is an autosomal dominant distal myopathy. Weakness and atrophy are usually confined to the anterior compartment of the lower leg muscles, tibialis anterior, and long toe extensors. First symptoms of impaired dorsiflexion of the ankles occur after the age of 35 years or much later and the disease progression is slow. Patients remain ambulatory, although some may develop mild proximal weakness.1 Morphologically, affected muscles show variable myopathic– dystrophic changes. Rimmed vacuoles are often, but not always, present. Patients with TMD have not been diagnosed with cardiomyopathy, and sarcomere structure appears intact in electron microscopic studies.1 Progressive muscle wasting leads to end-stage pathology with adipose and connective tissue replacement in the target muscles.2

We have assigned linkage in TMD to chromosome 2q by linkage analyses in Finnish families with TMD and later confirmed this assignment in one French family.3,4 All Finnish patients with TMD carry an identical core haplotype for the 2q region, suggesting that they share one ancestral mutation, whereas the patients in the French family carry a distinct haplotype. In the original Finnish TMD pedigree, three patients, affected with a more severe MD of limbgirdle type (LGMD) and childhood onset of the disease, were found to be homozygous for the TMD haplotype.3,5 Additionally, we have linked one family from the United States with late-onset distal myopathy (LODM, Markesbery–Griggs disease) to the 2q locus6,7 and this disease is considered as allelic to TMD. Two other muscle disease loci have been found to overlap the TMD locus: autosomal dominant my-

*These authors contributed equally to this work. From the Department of Human Molecular Genetics (Drs. Auranen and Peltonen and H. Haravuori), National Public Health Institute, and Department of Neurology (Dr. Somer), University of Helsinki; Neurological Department (Dr. Udd and A. Vihola), Vaasa Central Hospital, Finland; Department of Pediatrics (Dr. Voit), University of Essen; Faculty for Clinical Medicine (Dr. Labeit), Mannheim, Germany; Généthon (Drs. Richard and Beckman and S. Marchand), Evry, France; Department of Human Genetics (Dr. Peltonen), UCLA School of Medicine, Los Angeles, CA. Supported by the Association Francaise contre les Myopathies, the Daisy and Yrjö Eskola Fond for the Paulo Foundation, Finska läkaresällskapet rf., Finnish Cultural Foundation, the Finnish Medical Foundation, the Finnish Neurological Society, Muscular Dystrophy Association, Deutsche Forschungsgemeinschaft, and the Ulla Hjelt Foundation for Pediatric Research. Presented at the 52nd annual meeting of the American Academy of Neurology; San Diego, CA; April 29 – May 6, 2000. Received August 17, 2000. Accepted in final form December 10, 2000. Address correspondence and reprint requests to Dr. Bjarne Udd, Department of Neurology, Vaasa Central Hospital, FIN-65130 Vaasa, Finland; e-mail: [email protected] Copyright © 2001 by AAN Enterprises, Inc.

869

Figure 1. Schematic picture of the sarcomere (A) and titin (B). The layout of the sarcomere shows thin (actin) and thick (myosin) filaments in association with titin and some other sarcomeric proteins. Approximate locations of some antibody-binding sites are indicated with empty arrows (see table 1), and calpain3 binding sites in I- and M-line titin are pointed out with shaded arrows. The titin kinase domain near C-terminus and PEVK-domain are also indicated. PEVK ⫽ region rich in proline, glutamate, valine, and lysine.

opathy with proximal weakness and early respiratory muscle involvement8 and familial dilated cardiomyopathy.9 The linked region on chromosome 2q is 1 cM long. In the preliminary physical mapping studies, we have found that YAC and PAC clones, including the linked markers, harbor the gene encoding titin and most probably the complete 300-kb nucleotide sequence of the titin gene is within the critical region for TMD. Titin, a giant structural protein of the sarcomere (Mr ⱖ 3,000 kd), forms the third filament system in striated muscle in addition to thin and thick filaments formed by actin and myosin. Single titin molecules span half sarcomeres from Z discs to M lines, extending in vivo 2 ␮m (figure 1).10 The gene is expressed in many different splice variants; I-band titin has longer isoforms in skeletal muscle compared to cardiac muscle.10 The central Z-disc region is composed of a variable number of Z-repeats and longer isoforms are found in the heart muscle.11,12 M-line titin has two distinct isoforms.13 Titin has been speculated to act as a molecular ruler, keeping the myosin filament at the center of sarcomere during muscle contraction cycles and to regulate the ultrastructure of the vertebrate thick filament.14 Titin also contributes to sarcomere assembly15,16 and accounts for passive tension of myofibrils.17–20 Several ligand-binding sites exist along the length of the titin molecule.14,16,21,22 A-band titin has multiple myosin and MyBP (myosin binding protein) binding sites along with myomesin binding site.14,23,24 Two of titin’s ligands, the muscle-specific protease calpain (calpain3, CAPN3, p94) and telethonin/T-cap, are of special interest because they are mutated in two types of limb-girdle muscular dystrophy (LGMD2A25,26 and LGMD2G27). Titin has at least two binding sites for 870 NEUROLOGY 56

April (1 of 2) 2001

calpain3: Mex5 in M-line titin, and the three Ig repeats (I81-I83) on N2-A line of I-band titin.22 Telethonin/Tcap binding site along with multiple ␣-actinin binding sites is located in Z-line titin.12,28,29 Further, close to its C-terminus, titin has a Ser/Thr-kinase domain homologous to myosin light-chain kinase (MLCK) family of kinases. This domain is suggested to phosphorylate telethonin in early differentiating myocytes and have role in myofibrillogenesis.30 The titin gene of the mouse has been mapped to the proximal part of chromosome 2, a region syntenic to human chromosome 2.31 A spontaneous mutation in this region causes a recessive degenerative muscle disease, muscular dystrophy with myositis (MDM) of the mouse.32 Phenotypically, the mice show an abnormal gait, are reduced in weight, and have a shortened life span.32,33 Similar to patients with TMD, MDM mice reveal pronounced signs of skeletal muscle fiber degeneration in the tibialis anterior muscle.33 The titin gene Ttn may co-localize with the mdm disease locus and, therefore, is considered as a candidate gene.31,34 We attempted to detect the underlying molecular etiology in TMD, namely the role of titin and calpain3 regulatory functions in this disease. Additional evidence is taken from the examination of the MDM mouse, which is suggested to be an animal model for TMD. Methods. Patients and mouse material. Patients were from the families included in the previous linkage studies.3 Muscle biopsies for immunoblotting, immunohistochemistry studies, RNA extraction, and TUNEL (terminal deoxynucleotidyl transferase–mediated UTP end labeling) staining were obtained from informed and consenting individuals, and some of the previous biopsies also served diag-

Table 1 Titin antibodies in immunohistochemistry Antibody

Epitope

Reference

Type of antibody

x112–x113

Z1–Z2 domains, titin N-terminus

S. Labeit

Rabbit polyclonal

x118–x119

I-band, proximal Ig-tandem repeats

S. Labeit

Rabbit polyclonal

BK283–284

I-band, distal Ig-tandem segment (I20–I22)

S. Labeit

Rabbit polyclonal

x99a–x100

MIR

S. Labeit

Rabbit polyclonal

BD6

A-band, near end of thick filament

J. Trinick

Mouse monoclonal

DB12

Central A-band

J. Trinick

Mouse monoclonal

9D10

I-band, PEVK

M. Greaser

Mouse monoclonal

x105–x106

N2A

S. Labeit

Rabbit polyclonal

McSM1 Mc3B9

I-band

40

S. Kimura

Mouse monoclonal

A/M-bands (six binding sites)

S. Kimura40

Mouse monoclonal

NCL-TITIN

?

Novocastra

Mouse monoclonal

T11

I-band, near I/A-junction

Sigma41

Mouse monoclonal

41

Ig ⫽ immunoglobulin; MIR ⫽ major immunogenic region of titin; PEVK ⫽ region rich in proline, glutamate, valine, and lysine.

nostic purposes. Whole blood samples for DNA studies were acquired from earlier linkage studies and lymphoblastoid cell lines have been established. Age- and sex-matched normal control (C57BL/6J, C57BL/10), MDX (C57BL/10-mdx; dystrophin deficient) and MDM mice (C57BL/6J-mdm) were studied. Breeder mice of the MDM strain were obtained in 1992 from the Jackson Laboratory. Mice were aged up to 12 weeks. Animals were kept in the animal care unit of the University of Bielefeld and of the University of Essen according to animal care guidelines. The animal care use and review committee of the University of Essen authorized all animal studies. Southern blot hybridization. Total whole blood DNA from two control samples (three patients with TMD, one patient with LGMD homozygous for the TMD haplotype, and one patient with LGMD not carrying the TMD haplotype) was analyzed. DNA was digested with restriction endonucleases EcoRI, HindIII, and PvuII (Amersham), and in Southern blot hybridization membranes were hybridized separately with titin cDNA PCR product pools hh1–9, hh14 –20, hh45–53 recognizing 5' region, A/I junction and 3' regions of titin10,35,36 (URL: http://www.emblheidelberg. de/ExternalInfo/Titin/annotation.html). RNA extraction and RT-reactions. Total RNA was extracted with RNeasy Mini Kit (Qiagen, Hilden, Germany) from either immortalized lymphocyte cell cultures or muscle biopsy samples that were surgically obtained and immediately frozen in liquid nitrogen and stored in ⫺70 °C. Complementary DNA was synthesized from 1 ␮g of total RNA using 15 U avian myeloblastosis virus reverse transcriptase (Amersham, Arlington Heights, IL), 1 U RNAse inhibitor (Rnasin, Promega, Madison, WI), 10 pmol of dT30 primer and the four deoxynucleoside triphosphates at 0.4 mM concentration in 20 ␮L of buffer supplied with the enzyme. Sequencing. To search for the mutation causing TMD, samples from patients heterozygous and homozygous for the TMD haplotype were analyzed in a sequencing effort. Controls were healthy family members, a French patient with TMD, a patient with LODM, and a patient with 2q31linked autosomal dominant myopathy with proximal

weakness and early respiratory muscle involvement. These patients were assumed to carry a different mutation than the Finnish patients with TMD. Specific primer pairs were designed to amplify fragments of titin gene or cDNA. Genomic DNA or reverse transcriptase (RT)-reaction product was PCR amplified. The PCR products were purified enzymatically,37,38 and the products were cycle sequenced using ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit (with AmpliTaq® DNA Polymerase, FS).39 Extension products were ethanol/sodium acetate precipitated and electrophoresed with ABI PRISM 377 DNA sequencer (Applied Biosystems, Foster City, CA). Sequence data were analyzed with Sequencing Analysis 3.3 (Perkin-Elmer, Applied Biosystems, Foster City, CA) and Sequencher 3.0 software. Titin immunohistochemistry. Muscle biopsies from both homozygous and heterozygous patients with TMD were analyzed. Standard immunohistochemistry protocols were used on frozen 6-␮m muscle sections, fixed with ice cold acetone for 6 minutes. A total of 12 antibodies were used against different epitopes of titin (table 1, figure 1) and ␣-actinin (NCL-␣-ACT) and telethonin (cpf3a-4c) antibodies in patients with TMD to detect any altered stainingresulting from fortuitous mutations in the titin sequence (Novocastra Laboratories, UK; Sigma, St. Louis, MO) Antigenic detection was performed according to the manufacturer’s instructions using biotinylated antimouse and antirabbit secondary antibodies followed by ABC-DABdetection (Vectastain Elite Universal Kit, Vector Laboratories, Burlingame, CA), or with fluorescein isothiocyanate (FITC) conjugated secondary antibodies for immunofluorescence detection (DAKO F 0232 and DAKO F 0205). A control section from healthy muscle was placed on every slide. Marker analysis of LGMD loci. Three to four fluorescently labeled polymorphic microsatellite markers on LGMD2B-2G loci and especially markers D15S146, D15S514, D15S779, D15S782, D15S778, D15S781, D15S222, and D15S132 adjacent to CAPN3 LGMD2A locus were used for genotyping members of the family with TMD with patients with LGMD. Loci were genotyped using ABI PRISM system and Genotyper 2.0 software April (1 of 2) 2001

NEUROLOGY 56

871

Table 2 Results of the titin sequencing Domains

Reference sequence

Z1–Z3

X90568

Z-repeats

Special features

bp

Results

Telethonin interaction site

1,500

Genomic DNA, exons and splice sites, no nucleotide alterations

X90568

Expressed in variable copy numbers in skeletal and heart muscles

5,800

Genomic DNA, exons and splice sites, Zr5 nucleotide change C24263T leading to aa change T7653I. Found in a healthy control and absent in one TMD patient

I27–I68

X90569

Ig repeats not expressed in heart muscle (I28–I67 soleus muscle and I28–I29, I48–I67 psoas muscle)

16,700

Genomic DNA, exons and splice sites, six polymorphisms in exons (five amino acid changes, controls also carried the polymorphism or heterozygous TMD patient carried polymorphism in homozygous state). Intronic polymorphisms.

N2A-line

X90569, end of I79–beginning of PEVK

Calpain3 interaction site, I81–I83

6,500

Genomic DNA including all introns, two polymorphisms in exon coding I81, no amino acid changes, six intronic polymorphisms

Mex1

X9056810,30 aa 24731–25054 X92412 bp 11500–12780 bp

Titin kinase domain

1,280

Genomic DNA, no nucleotide alterations

Mex5

X9241213 end of Mex4–beginning of Mex6

Calpain3 interaction site

1,000

Genomic DNA, including introns, no nucleotide alterations

List of polymorphisms is available on request. Ig ⫽ immunoglobulin; TMD ⫽ tibial muscular dystrophy; PEVK ⫽ region rich in proline, glutamate, valine, and lysine.

(Perkin-Elmer, Applied Biosystems). Haplotypes were constructed that allowed minimal number of recombination events. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting. Muscle biopsies of

seven heterozygous patients with TMD and one with LGMD homozygous for the TMD haplotype were snap frozen in isopentane chilled in liquid nitrogen to about 150 °C, after which they were weighted and homogenized with 19 volumes of SDS-PAGE sample buffer containing 4 M

Figure 2. Haplotypes around the CAPN3 locus in chromosome 15 for the tibial muscular dystrophy (TMD) family with the patients with limb-girdle type MD (LGMD) homozygous for the chromosome 2q31 haplotype. Patients with LGMD are presented as black figures, patients with TMD as shaded figures. The two sibling patients with LGMD do not share any of the haplotypes from their parents, and their LGMD-affected uncle shares alternative haplotypes with each one of them, and the deceased mother has the same haplotype as the uncle, although without her being affected by LGMD. 872 NEUROLOGY 56

April (1 of 2) 2001

Figure 3. Western blotting with two different calpain3 antibodies in patients with tibial muscular dystrophy (TMD) and limb-girdle type MD (LGMD) (TMD homozygous). Proteins on blots were labeled either with NCL-CALP-2C4 or NCLCALP-12A2, followed by DAB detection. Approximate molecular masses are shown on the left. C ⫽ control (pooled control homogenate); C100 ⫽ 100% control homogenate; C70 ⫽ 70% control homogenate dilution; C50 ⫽ 50% control homogenate dilution; C25 ⫽ 25% control homogenate dilution. urea and 4% SDS. Proteins were denatured by incubation at ⫹100 °C for 5 minutes. Next, 8% gels were prepared, and different dilutions of pooled control homogenate were applied into each gel for quantitation. Both gel electrophoresis42 and Western blotting43 were carried out using standard buffers and protocols, with Bio-Rad Protean II xi Cell and Bio-Rad Trans-Blot Cell (Hercules, CA). After electrophoresis, the proteins were transferred onto nitrocellulose membrane sheets and labeled with polyclonal telethonin/T-cap antibody or monoclonal calpain3-antibodies NCL-CALP-2C4 and NCL-CALP-12A2 (Novocastra Laboratories, UK), which recognize 94 kd fullsize calpain3 and additional 30-kd and 60-kd protein fragments.28,44 DAB-detection was used with horseradish peroxidase– conjugated secondary antibody (DAKO

Figure 4. Western blotting of the muscular dystrophy with myositis (MDM) mouse muscle with Calp3c/11B3 antibody. Control material consisted of healthy human muscle and muscle from a patient with LGMD2A as negative control (on the left) as well as normal mouse and MDX (dystrophin deficient) mouse muscle (on the right). In comparison with normal control and MDX muscle, MDM muscle shows a severe reduction of the 94-kd band of calpain3 with antibodies directed against exon 8 sequences.

P-0260.) Residual proteins on gels were stained with Coomassie Brilliant Blue (R-250) after Western blotting. Dried blots and gels were used for densitometric analysis with Bio-Rad Quantity One-software. The MDM skeletal muscle was homogenized with 1% SDS. After centrifugation, 100 ␮g of the supernatant was resolved by SDS-PAGE42 on 3% to 15% linear gradient gels and transferred to nitrocellulose membranes.43 Nitrocellulose transfers were blocked in Blotto (50 mM sodium phosphate, pH 7.4, 150 mM sodium chloride, and 5% nonfat dry milk) and subsequently incubated overnight with primary anticalpain antibodies Calp3c/11B3 and Calp3d/12A2.44 Immunoblots were then washed with Blotto and incubated for 1 hour with peroxidase-conjugated secondary antibody (Boehringer Mannheim, Mannheim, Germany) at a dilution of 1:1,000. Apoptosis studies. The TUNEL staining method (in situ cell death detection kit) and anticaspase3 antibodies were applied to detect apoptotic myonuclei from the biopsies of patients with TMD, patients with LGMD homozygous for the TMD haplotype, and samples of MDM mouse. The muscle biopsies were counterstained with antidystrophin and DAPI (4,6-diamidino-2-phenylindole), appropriate secondary fluorescent antibodies were used, and the sections were viewed with indirect immunofluorescence confocal microscopy. The TUNEL staining method was used following to manufacturer’s protocol (Boehringer). Positive controls were obtained by incubating sections with 0.5 mg/mL deoxyribonuclease (DNAse)1 at room temperature before TUNEL staining. Further antibodies against I␬B␣ and p65NF-␬B were used on a set of slides to study I␬B␣/NF-␬B expression in muscle cells. Methods used to detect apoptosis, the study of the I␬B␣/NF-␬B route and indirect immunofluorescence confocal microscopy are described in detail elsewhere.45 Results. Southern blot hybridization and sequencing of titin. The fragment patterns produced by multiple restriction enzymes were analyzed. As expected, no gross rearrangements were evidenced by Southern blot hybridApril (1 of 2) 2001

NEUROLOGY 56

873

Figure 5. Apoptosis in muscle biopsies. All muscle sections were costained with DAPI (blue, stains all nuclei), TUNEL (green), anticaspase3, and antidystrophin (red). Images of different stainings and superimposed pictures are presented. (A to D) Healthy human muscle. No TUNEL-positive myonuclei are present in the sections and anticaspase3 and dystrophin stainings are normal (⫻200 before reduction). (E to H) A heterozygous patient with tibial muscular dystrophy (TMD). Dystrophin staining is normal. A cluster of three green TUNEL-positive myonuclei is found. These nuclei, also anticaspase3 positive, are seen in the superimposed picture as yellow (⫻200 before reduction). (I to L) Apoptotic myonuclei in the patient with limb-girdle MD (LGMD) homozygous for the TMD haplotype are even more pronounced than in the patient with TMD (⫻200 before reduction).

ization. Because of the large size of the titin gene, initial sequencing was targeted to regions that are differentially expressed in heart and skeletal muscles because TMD does not affect heart muscle. No mutations were detected (table 2). Immunohistochemistry. No abnormal staining in immunohistochemistry was seen in muscle biopsies from patients with TMD or the patient with LGMD homozygous for the TMD haplotype compared with normal control muscle sections, with the panel of 12 titin antibodies, telethonin/T-cap and ␣-actinin antibodies (results not shown). Normal expression of dystrophin and sarcoglycans has earlier been confirmed in patients with TMD. Marker analysis of LGMD loci. Having the presentation of LGMD phenotype in the patient homozygous for the TMD haplotype, it was important to look for the possible involvement of the other LGMD loci. This was addressed by genotyping for a number of markers bracketing the LGMD2A-2G loci. Haplotype reconstruction ruled out the involvement of any of these loci (figure 2 for the CAPN3 locus data). Western blotting of titin ligands. Given the reported interaction of calpain3 and telethonin with titin, both of 874 NEUROLOGY 56

April (1 of 2) 2001

which are involved in LGMD, their status was examined. Calpain3 was deficient in the patient with LGMD homozygous for the TMD haplotype (figure 3). Calpain3 amounts varied in the muscle samples of the TMD heterozygotes, some of which showed decreased calpain3 (see figure 3). With calpain3 monoclonal antibody (Mab) 2C4, no calpain3 was detectable in the TMD homozygote sample, but with the Mab 12A2 some residual calpain3 could be detected. Immunodetection with the telethonin/T-cap antibody showed no abnormalities (data not shown). Based on the findings of secondary calpain3 deficiency in patients homozygous for the TMD haplotype, we analyzed MDM skeletal muscle for the expression of calpain3. Dystrophin-deficient MDX muscle served as a muscle disease control, and calpain3-deficient skeletal muscle from a patient affected by LGMD2A served as a negative control for calpain3 staining. MDM muscles also showed a severe reduction of the 94 kd band of calpain3 with antibodies Calp3c/11B3 and Calp3d/12A2 against exon 8 sequences (figure 4). (Results not shown for the latter antibody.) Apoptotic myonuclei and I␬B␣/NF-␬B expression. To find out if patients with 2q-linked TMD and patients with

Figure 6. I␬B␣, NF-␬B, and TUNEL labeling of muscle biopsies. (A to D) Healthy human control. No TUNEL-positive myonuclei are found. I␬B␣ labeling and NF-␬B labeling are evenly distributed in the cytoplasm (⫻200 before reduction). (E to H) A patient with tibial muscular dystrophy (TMD). Some rare TUNEL-positive myonuclei are seen. An accumulation of I␬B␣ is seen in the same nuclei (⫻400 before reduction). (I to L) The patient with limb-girdle MD (LGMD) homozygous for the TMD haplotype. TUNEL-positive apoptotic myonuclei are present, also showing an aberrant label of I␬B␣ protein. NF-␬B protein is seen in subsarcolemma (⫻400 before reduction). LGMD with secondary calpain3 deficiency shared similar cellular pathomechanisms with primarily calpain3deficient patients with LGMD2A, we used TUNEL staining method and anticaspase3 immunolabeling on muscle biopsies from heterozygous patients with TMD as well as from MDM mouse. Apoptotic mononuclei were detected in the muscle of the patient with TMD and more frequently in the muscle of the homozygous patient with LGMD, often in small clusters and in centralized myonuclei (figure 5f). TUNEL-positive myonuclei were also anticaspase3 positive, but in addition anticaspase3 staining positive fibers were more widely distributed (see figure 5). On control muscle samples, no TUNEL positive or anticaspase3 staining positive myonuclei were detected. In samples from patients heterozygous or homozygous

for the TMD haplotype, I␬B␣ was found to be aberrantly relocated in TUNEL-positive myonuclei, more evidently in homozygous LGMD muscle. In the patients, expression of NF-␬B was located mainly in subsarcolemma (figure 6). TUNEL, I␬B␣, and NF-␬B immunostaining of MDM muscle samples showed similar results, although less pronounced (data not shown).

Discussion. Titin is considered as a positional and functional candidate gene for TMD. The results presented here, though they cannot definitely prove titin to be defective in TMD, further strengthen this hypothesis. The exact molecular pathomechanisms of various muscular dystrophies remain unknown. AbApril (1 of 2) 2001

NEUROLOGY 56

875

sent calpain3 in LGMD2A has been an exception among the dystrophies caused by defective membrane proteins. The stimulus to analyze calpain3 and apoptosis in TMD came from the identification of three patients with LGMD homozygous for chromosome 2q haplotype in a large TMD pedigree. Because calpain3 is a ligand of titin, and the patients who had LGMD showed similarities in their clinical picture to the LGMD2A phenotype, the hypothesis of a secondary calpain3 defect was postulated. Using haplotype analyses we excluded the known LGMD2A-2G loci. The major finding on Western blotting studies was calpain3 deficiency in the patient with LGMD homozygous for the TMD haplotype, and similar calpain3 deficiency in the MDM mouse, an animal model considered to be caused by a Ttn gene mutation. Some of the heterozygous patients with TMD also showed decreased calpain3. This finding might represent some kind of a temporal phenomenon in TMD pathogenesis, but at this point we cannot offer a valid explanation for it. Despite that no sequence anomaly was detected in the known calpain3 binding sites of titin, calpain3 is affected in these patients. This might be explained by a new, uncharacterized calpain3-binding site in titin, or mutations beyond the actual binding sites, which could cause conformational changes in calpain3-binding structures. Even involvement of new titin binding sites for other molecules (e.g., affecting calpain3 binding indirectly) cannot be ruled out at this point. However, in case the observed calpain3 defect is caused by defective titin, this is the first physiologic evidence of a link between these two proteins. Furthermore, it would provide a powerful pathophysiologic paradigm to explain the phenotypic consequences of bearing the mutation in single or double dose. The gene product of the normal allele in heterozygous patients can presumably still trap and bind calpain3, thereby stabilizing and protecting it from rapid autolysis, although the impact of the mutation can already be felt on the residual amounts of calpain3. Homozygous carriers of the Finnish mutation presumably no longer bind and protect efficiently calpain3 from degradation. Hence, this protease is eventually lost, and the situation in the muscle fibers resembles that of a primary calpainopathy, explaining the appearance of the proximal LGMD phenotype. On the other hand, the observed calpain3 defect may be caused by other, yet unknown gene defects. However, no other plausible molecular context is known for calpain3 interaction. I␬B␣ masks the nuclear localization signal of the transcription factor NF-␬B, rendering the complex cytoplasmic and inactive. Once IkB␣ is degraded, NF-␬B enters the nucleus and activates genes involved in cell survival and inflammatory response.46 It has been proposed that in LGMD2A, the absence of functional calpain3 in muscle cells results in perturbation of the I␬B␣/NF-␬B pathway, leading to myonuclear apoptosis.45 Furthermore, indications point to I␬B␣ serving as an in vitro substrate for 876 NEUROLOGY 56

April (1 of 2) 2001

calpain3, and that the defect in I␬B␣ turnover results in cytoplasmic sequestration of NF-␬B, thus, indirectly preventing the NF-␬B-dependent expression of survival genes. We have explored whether these features are also present in TMD/LGMD and MDM with secondary calpain3 deficiency, and could observe both apoptosis and altered I␬B␣/NF-␬B distribution. However, their significance for pathogenesis remains to be clarified. Identification of the mutations causing both TMD and MDM will show whether MDM is a true animal model for TMD. The secondary calpain3 defect in 2q-linked LGMD seems to be a separate category of recessive LGMD. The calpain3 deficiency shown in patients with LGMD homozygous for the TMD haplotype and MDM mouse in this study might represent specific secondary events to the primary gene and protein defects. Conversely, if these changes are just concomitant secondary changes with unknown pathophysiologic context, secondary calpain3 deficiency may be part of a more general response in dystrophic muscle cells, not necessarily of less importance. . Acknowledgment The authors thank Dr. Louise V. B. Anderson, UK, who provided valuable advice for this study. Some antibodies were a gift from Dr. Kimura, Japan.

References 1. Udd B, Partanen J, Halonen P, et al. Tibial muscular dystrophy—late adult onset distal myopathy in 66 Finnish patients. Arch Neurol 1993;6:604 – 608. 2. Udd B, Rapola J, Nokelainen P, et al. Nonvacuolar myopathy in a large family with both late adult onset distal myopathy and severe proximal muscular dystrophy. J Neurol Sci 1992;2: 214 –221. 3. Haravuori H, Mäkelä–Bengs P, Udd B, et al. Assignment of the tibial muscular dystrophy locus to chromosome 2q31. Am J Hum Genet 1998;62:620 – 626. 4. de Seze J, Udd B, Haravuori H, et al. The first European family with tibial muscular dystrophy outside the Finnish population. Neurology 1998;51:1746 –1748. 5. Udd B, Kääriäinen H, Somer H. Muscular dystrophy with separate phenotypes in a large family. Muscle Nerve 1991;14: 1050 –1058. 6. Markesbery WR, Griggs RC, Leach RP, et al. Late onset hereditary distal myopathy. Neurology 1974;24:127–134. 7. Haravuori H, Mäkelä–Bengs P, Figlewicz DA, et al. Tibial muscular dystrophy and late-onset distal myopathy are linked to the same locus on chromosome 2q. Neurology 1998;50(suppl 4):A186. Abstract. 8. Nicolao P, Xiang F, Gunnarsson L-G, et al. Autosomal dominant myopathy with proximal weakness and early respiratory muscle involvement maps to chromosome 2q. Am J Hum Genet 1999;64:788 –792. 9. Siu BL, Niimura H, Osborne JA, et al. Familial dilated cardiomyopathy locus maps to chromosome 2q31. Circulation 1999; 99:1022–1026. 10. Labeit S, Kolmerer B. Titins. Giant proteins in charge of muscle ultrastructure and elasticity. Science 1995;270:293–296. 11. Gautel M, Goulding D, Bullard B, et al. The central Z-disc region of titin is assembled from a novel repeat in variable copy numbers. J Cell Sci 1996;109:2747–2754. 12. Sorimachi H, Freiburg A, Kolmerer B, et al. Tissue-specific expression and ␣-actinin binding properties of the Z-disc titin: implications for the nature of vertebrate Z-discs. J Mol Biol 1997;270:688 – 695.

13. Kolmerer B, Olivieri N, Witt CC, et al. Genomic organization of M line titin and its tissue-specific expression in two distinct isoforms. J Mol Biol 1996;256:556 –563. 14. Labeit S, Gautel M, Lakey A, et al. Towards a molecular understanding of titin. EMBO J 1992;11:1711–1716. 15. Gautel M, Mues A, Young P. Control of sarcomeric assembly: the flow of information on titin. Rev Physiol Biochem Pharmacol 1999;138:97–137. 16. Gregorio C, Granzier H, Sorimachi H, et al. Muscle assembly: a titanic achievement? Curr Opin Cell Biol 1999;11:18 –25. 17. Linke WA, Stockmeier MR, Ivemyer M, et al. Characterizing titin’s I-band Ig domain region as an entropic spring. J Cell Sci 1998;111:1567–1574. 18. Linke WA, Ivemyer M, Mundel P, et al. Nature of PEVK-titin elasticity in skeletal muscle. Proc Natl Acad Sci USA 1998;95: 8052– 8057. 19. Linke WA, Granzier H. A spring tale: new facts on titin elasticity. Biophys J 1998;75:2613–2614. 20. Improta S, Krueger J, Gautel M, et al. The assembly of immunoglobulin-like modules in titin: implications for muscle elasticity. J Mol Biol 1998;284:761–777. 21. Young P, Ferguson C, Bañuelos S, et al. Molecular structure of the sarcomeric Z-disc: two types of titin interactions lead to asymmetrical sorting of ␣-actinin. EMBO J 1998;17:1614 – 1624. 22. Sorimachi H, Kinbara K, Kimura S, et al. Muscle-specific calpain, p94, responsible for limb girdle muscular dystrophy type 2A, associates with connectin through IS2, a p94 specific sequence. J Biol Chem 1995;270:31158 –31162. 23. Trinick J. Titin and nebulin: protein rulers in muscle? Trends Biochem Sci 1994;19:405– 409. 24. Freiburg A, Gautel M. A molecular map of the interactions between titin and myosin-binding protein C—implications for sarcomeric assembly in familial hypertrophic cardiomyopathy. Eur J Biochem 1996;235:317–323. 25. Richard I, Broux O, Allamand V, et al. Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy 2A. Cell 1995;81:27– 40. 26. Ono Y, Shimada H, Sorimachi H, et al. Functional defects of a muscle-specific calpain, p94, caused by mutations associated with limb-girdle muscular dystrophy type 2A. J Biol Chem 1998;273:17073–17078. 27. Moreira ES, Wiltshire TJ, Faulkner G, et al. Limb-girdle muscular dystrophy type 2G is caused by mutations in the gene encoding the sarcomeric protein telethonin. Nat Genet 2000; 24:163–166. 28. Valle G, Faulkner G, De Antoni A, et al. Telethonin, a novel sarcomeric protein of heart and skeletal muscle. FEBS Lett 1997;415:163–168. 29. Mues A, van der Ven FM, Young P, et al. Two immunoglobulin-like domains of the Z-disc portion of titin interact in a conformation-dependent way with telethonin. FEBS Lett 1998;428:111–114. 30. Mayans O, van der Ven PFM, Wilm M, et al. Structural basis for activation of the titin kinase domain during myofibrillogenesis. Nature 1998;395:863– 847. 31. Müller–Seitz M, Kaupmann K, Labeit S, et al. Chromosomal localization of the mouse titin gene and its relation to “muscu-

32. 33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

lar dystrophy with myositis” and nebulin genes on chromosome 2. Genomics 1993;18:559 –561. Lane PW. Muscular dystrophy with myositis (mdm). Mouse News Lett 1985;73:18. Heimann P, Menke A, Rothkegel B, et al. Overshooting production of satellite cells in murine skeletal muscle affected by the mutation “muscular dystrophy with myositis” (mdm, Chr 2). Cell Tissue Res 1996;283:435– 441. Siracusa LD, Silan CM, Justice MJ, et al. A molecular genetic linkage map of mouse chromosome 2. Genomics 1990;6:491– 504. Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 1975; 98:503–515. Feinberg AP, Vogelstein B. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 1984;137:266 –267. Hanke M, Wink M. Direct DNA sequencing of PCR-amplified vector inserts following enzymatic degradation of primer and dNTPs. BioTechniques 1994;17:858 – 860. Werle E, Schneider C, Renner M, et al. Convenient singlestep, one tube purification of PCR products for direct sequencing. Nucleic Acids Res 1994;22:4354 – 4355. Chadwick RB, Conrad MP, McGinnis MD, et al. Heterozygote and mutation detection by direct automated fluorescent DNA sequencing using a mutant Tag DNA polymerase. BioTechniques 1996;20:676 – 683. Itoh Y, Suzuki T, Kimura S, et al. Extensible and lessextensible domains of connectin filaments in stretched vertebrate skeletal muscle sarcomeres as detected by immunofluorescence and immunoelectron microscopy using monoclonal antibodies. J Biochem 1988;104:504 –508. Fürst DO, Osborn M, Nave R, et al. The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line. J Cell Biol 1988;106:1563–1572. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227: 680 – 685. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 1979;76:4350 – 4354. Anderson LVB, Davison K, Moss JA, et al. Characterization of monoclonal antibodies to calpain3 and protein expression in muscle from patients with limb-girdle muscular dystrophy type 2A. Am J Pathol 1998;153:1169 –1179. Baghdiguian S, Martin M, Richard I, et al. Calpain 3 deficiency is associated with myonuclear apoptosis and profound perturbation of the I␬B␣/NF-␬B pathway in limb-girdle muscular dystrophy type 2A. Nat Med 1999;5:503–511. Karin M. The NF-␬B activation pathway: its regulation and role in inflammation and cell survival. Cancer J Sci Am 1998; 4(suppl 1):92–99.

April (1 of 2) 2001

NEUROLOGY 56

877