Skip Navigation


Human Molecular Genetics Advance Access originally published online on May 11, 2004
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/13/1373    most recent
ddh153v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (37)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Kramerova, I.
Right arrow Articles by Spencer, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kramerova, I.
Right arrow Articles by Spencer, M. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2004, Vol. 13, No. 13 1373-1388
DOI: 10.1093/hmg/ddh153
Human Molecular Genetics, Vol. 13, No. 13 © Oxford University Press 2004; all rights reserved

Null mutation of calpain 3 (p94) in mice causes abnormal sarcomere formation in vivo and in vitro

I. Kramerova1,2,{dagger}, E. Kudryashova1,2,{dagger}, J.G. Tidball2,3 and Melissa J. Spencer1,2,*

1Department of Pediatrics and Mattel Children's Hospital, David Geffen School of Medicine at University of California, Los Angeles, CA 90095-1606, USA, 2Duchenne Muscular Dystrophy Research Center and 3Departments of Physiological Science and Pathology and Laboratory Medicine, University of California at Los Angeles, Los Angeles, CA 90095-1606, USA

Received March 13, 2004; Accepted April 30, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The giant protein titin serves a primary role as a scaffold for sarcomere assembly; however, proteins that mediate this remodeling have not been identified. One potential mediator of this process is the protease calpain 3 (C3), the protein mutated in limb girdle muscular dystrophy type 2A. To test the hypothesis that C3 mediates remodeling during myofibrillogenesis, C3 knockout (C3KO) mice were generated. The C3KO mice were atrophic containing small foci of muscular necrosis. Myogenic cells fused normally in vitro, but lacked well-organized sarcomeres, as visualized by electron microscopy (EM). Titin distribution was normal in longitudinal sections from the C3KO mice; however, EM of muscle fibers showed misaligned A-bands. In vitro studies revealed that C3 can bind and cleave titin and that some mutations that are pathogenic in human muscular dystrophy result in reduced affinity of C3 for titin. These studies suggest a role for C3 in myofibrillogenesis and sarcomere remodeling.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The calpain family of proteases is composed of both ubiquitous and tissue-specific isoforms that share homology in their protease domain (1). Muscles express at least three isoforms called calpains 1, 2 and 3. Calpains 1 and 2 (also known as µ- and m-calpains) are ubiquitous isoforms, whereas calpain 3 (C3 or p94) is predominantly expressed in muscles, although alternatively spliced isoforms have also been identified in other cell types (2–4). Although calpains have diverse substrates and physiological roles, they share several common characteristics including activation by calcium, use of cysteine as the active site residue and limited proteolytic cleavage of substrate (1). This last feature has prompted speculation that calpains function as regulators of other cellular proteins rather than as housekeeping proteases. For example, several studies have implicated calpains in cytoskeletal remodeling in platelet aggregation (5), neurite outgrowth (6) and myoblast fusion (7).

Calpains have also been shown to contribute to the pathophysiology of diseases such as Alzheimer's disease (8), multiple sclerosis (9) and muscular dystrophy (10). In the disease Duchenne muscular dystrophy, cell membrane fragility leads to elevated intracellular calcium and hyperactivation of calpain 1 (µ-calpain) (11). In this setting, dysregulation of calpain activity and promiscuous cleavage of substrate ensue, resulting in promotion of muscle cell death. Although increased calpain 1 activity is a feature of Duchenne dystrophy, decreased C3 activity causes another form of muscular dystrophy called limb girdle muscular dystrophy 2A (LGMD2A) (12). Patients with LGMD2A are characterized by muscle wasting and cell death, characteristics seen in most other muscular dystrophies. The mechanism by which mutations in a proteolytic enzyme might lead to muscle cell death has been elusive so far.

Recent findings have shown that C3 is much more stable in vivo than in vitro, suggesting that C3 must be stabilized through interaction with its ligand(s) (13). Titin is one such protein that might serve a stabilizing role for C3. Titin is a large, muscle-specific protein that is important both in muscle development and passive tension generation in adult muscle, and as a scaffold for the sarcomere (14). Evidence for a relationship between C3 and titin comes from yeast two-hybrid assays in which the binding was mapped to occur at the N2- and M-line regions (15,16). More recently the mutation in the mdm mouse was defined to occur at the N2-line of titin, resulting in profound muscular dystrophy (17). The N2-line fragment carrying the mdm mutation loses its ability to bind C3 in the yeast two-hybrid assays (18), and muscle extracts from mdm mice show reductions in the concentration of C3, supporting the hypothesis that titin serves to stabilize C3 (17). Furthermore, titin mutations in patients with tibial muscular dystrophy (at the M-line) also result in C3 instability in muscle extracts (19). Finally, the yeast two-hybrid assays with the LGMD2A mutants show a reduced binding of C3 to titin (20). These observations taken together lend support for the hypothesis that C3 must be anchored to titin in vivo to protect it from autolytic degradation, and suggest the possibility that loss of C3 may contribute to disease pathogenesis in mdm and tibial muscular dystrophy.

Although titin may play a significant role in determining C3 localization and stability, the question of C3 involvement in regulating titin function has been unexplored. If C3 were to serve this role, perturbations in C3 expression, structure or activity could disrupt vital functions mediated by titin. Because titin is centrally important in myofibrillogenesis where it serves as a template and molecular ruler for thick filament assembly and sarcomere formation, we hypothesized that changes in C3 expression could disrupt normal myofibrillogenesis. In the present investigation, we tested this hypothesis by examining the effects of null mutation of C3 on sarcomere structure in mice and in myotubes in vitro. We further investigated the relationship between C3 and titin by analyzing the effects of pathogenic mutations in C3 on the ability of C3 to bind titin in vitro, and by identifying the titin domains at which C3-mediated cleavage occurs. Collectively, our findings show that C3 expression is required for normal myofibril formation in vivo and in vitro, and that the LGMD2A mutations in C3 influence C3 binding to titin. Furthermore, we found that titin can be cleaved by C3 in vitro, which suggests that C3 may influence titin's role in sarcomere formation through proteolytic cleavage.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Retroviral integration and disruption of the C3 gene
The C3 knockout (C3KO) mice were generated using embryonic stem cells disrupted at the C3 locus using a gene trap retroviral vector (Lexicon Genetics, OST 141731) (21). This construct introduces pre-mature translational stop signals after generation of fusions between the 5' and 3' end of the C3 gene. To determine if C3 mRNA fragments or alternatively spliced products were present in the C3KO muscles, RT–PCR was performed at each end of the mRNA, and no product was detected (data not shown). Recently, new splice variants of the mouse and human C3 mRNA were described (22). To verify that none of these isoforms was present in the KO muscles, RT–PCR was also performed using primers located in exons present in all splice variants of C3 (exons 3–5 and exon 18 through the 3' end; positions of the primers are shown in Fig. 1A). This analysis showed that these exons could not be detected in the C3KO tissues (Fig. 1B). In addition, western blot analysis with epitope-specific antibodies (epitopes are shown in Fig. 1A) along the length of the C3 protein was performed to confirm the absence of full-length 94 kDa C3, and to verify that no fusion proteins of any size were generated by the retroviral integration. This analysis showed complete loss of p94 as well as the absence of any fusion proteins with the vector or products of alternative splicing (Fig. 1C).



View larger version (61K):
[in this window]
[in a new window]
  		Figure 1. The C3KO mice lack both calpain 3 mRNA and protein. (A) Schematic representation of the C3 domain structure and antibody epitopes. C3 consists of four domains (I–IV) and includes several insertion sequences (NS, IS1 and IS2) that are not present in other calpains. Positions of the primers for RT–PCR are shown as arrows above the diagram. Epitopes for the C3 antibodies used in this study are indicated below the diagram by arrows. (B) RT-PCR performed on the C3KO or WT mRNA shows the absence of the PCR product in the C3KO samples. Primers were chosen in exons that are found in all alternatively spliced isoforms of C3 (22). Primers 1up (exon 3) and 1dn (exon 5) were used to generate a 0.3 kb fragment. Primers 2up (exon 18) and 2dn (3'-end of the open reading frame) were used to generate a 0.4 kb fragment. (C) Western blot analysis of the C3KO and WT muscle lysates with C3 antibodies that recognize epitopes along the length of C3 shows the absence of the full-length 94 kDa C3 in the C3KO muscles. No other additional bands specific for C3KO were detected, confirming the absence of fusion proteins generated by the integration of the retrovirus. Arrowheads indicate the position of the full-length C3 (p94) in the WT lysates. This analysis also allowed for the determination of antibody specificity for the various C3 antibodies.

 
Several attempts have been made to determine the tissue localization of C3 in vivo. Using different antibodies, C3 has been documented to be at the N2-line (23) as well as at multiple other sites including the Z disk, M-line, costameres, myotendinous junctions and nuclei (24). All of these studies, however, were done with antibodies which, according to previously published results (23) and our data (Fig. 1C), recognize several additional bands, besides C3, on western blots. In an effort to identify the tissue localization of C3, we tested all available antibodies in immunohistochemistry assays. Unexpectedly, all antibodies that worked for immunohistochemistry were able to stain the C3KO muscles (data not shown). This result questions published C3 immunolocalization data. Generation of highly specific antibodies would be very helpful for further investigation. Thus, the precise location of C3 in muscle remains unclear and this has made the identification of in vivo substrates and its biological function elusive.

The C3KO mice replicate features of the LGMD2A phenotype
The C3KO mice were viable and fertile but were smaller than wild-type (WT) mice and had reduced muscle mass (Fig. 2A and B). Since the mice were on a mixed background, variability was observed from one litter to the next. Occasionally, extremely small mice were observed and some were not viable. These sick mice were not analyzed in the current study. Fiber area measurement of the soleus showed ~28% decrease in fast fiber area and ~21% decrease in slow fiber area in the KO (Fig. 2C). A 15% reduction was also observed in the slow fibers of the C3KO gastrocnemius muscle (Fig. 2D). Thus, fiber area reduction occurred in both fast- and slow-type muscle fibers. The distribution of muscle fiber areas was significantly different between WT and KO. While the WT mouse showed a normal distribution of fiber areas, the KO mouse had a much higher percentage of smaller fibers and a higher level of fiber size variability (data not shown). Similar changes have also been observed in muscle biopsies from the LGMD2A patients with mild or moderate manifestation of the disease (25,26).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. The C3KO mice display a mild atrophic phenotype. (A) Body weight of the young C3KO males is reduced when compared with the WT males. With age this difference decreases. The increase in body weight with aging is most likely due to excessive body fat accumulation in the C3KO animals based on visual analysis of the animals. Each point on the graph represents an average of 6–10 animals. (B) Tibialis anterior (TA) muscles are smaller in the C3KO males, and the difference in the muscle weight between the C3KO and WT animals increases with age. Four to eight animals were analyzed for each group. (C) Fiber area measurement shows a significant decrease in average cross-sectional area for both slow and fast fibers in the soleus muscle of the C3KO animals. (D) Fiber area measurement shows a significant decrease in average cross-sectional area for slow fibers in the gastrocnemius muscle of the C3KO animals. Three to five of both the WT and the C3KO mice were used for cross-sectional area measurements. (* Indicates significant difference at P<0.05 for all data shown.)

 
Patients with mild or pre-clinical LGMD2A show almost normal muscle histology except for small areas of focal necrosis (25,26). The histological appearance of the C3KO mice was examined for similar evidence of muscle pathology. Cross-sections of gastrocnemius, soleus, tibialis anterior (TA) and diaphragm muscles showed rare and small foci of necrosis and regeneration surrounded by primarily healthy looking tissue (Fig. 3). The soleus and diaphragm were the most affected muscles examined. Frequently, only one small necrotic lesion would appear in a cross-section, occupying approximately the space of five to eight fibers. These areas were always accompanied by inflammatory cells (Fig. 3B and D) and were regenerating, as evidenced by the presence of myogenin-positive cells (Fig. 3E) and developmental myosin heavy chain positive (MHC) myotubes (Fig. 3F). The observations of atrophy, fiber area variability and necrosis and regeneration show that the C3KO muscles possess similar histological characteristics as the LGMD2A biopsies from patients in the earliest stages of the disease, or with mild manifestations of the disease (25,26). Thus, the C3KO mouse is an ideal model for studying the biological function of C3 because early pathological events resulting from the absence of C3 can be analyzed without being confounded by secondary pathogenic processes.



View larger version (67K):
[in this window]
[in a new window]
  		Figure 3. The C3KO muscles show focal pathological features but no apoptotic myonuclei. (A) The C3KO diaphragm cross-section stained with hematoxylin. Arrows point to centrally located nuclei in regenerating fibers. (B) The C3KO soleus cross-section stained with CD11b antibodies (a marker of immune cells, stained red) and counterstained with hematoxylin (blue) to show invasion of inflammatory cells at the area of degeneration. Centrally nucleated fibers are also apparent. (C–F) Serial sections of the C3KO soleus stained with hematoxylin (C), CD11b (D), myogenin (E) and developmental MHC (F) to demonstrate the presence of both inflammation and regeneration in lesions of the C3KO muscle. (G, H) Serial sections of the C3KO soleus stained with hematoxylin (G) and double-stained for apoptotic nuclei (in black, labeled TdT) and sarcolemma (red) using anti-dystrophin antibody. Black apoptotic nuclei are always localized outside the sarcolemma in the area of necrosis and regeneration. (I, J) Apoptotic nuclei (in black, labeled TdT) in the C3KO soleus co-localize with an immune cell marker stained with CD11b antibodies (red). (J) is 3x magnification image of the boxed area shown in (I). Bar in (A) and (B), 100 µm. Bar in (C–I), 50 µm.

 
Apoptotic myonuclei are not a primary feature of murine calpainopathy
Previous studies have shown that apoptotic nuclei are present in LGMD2A biopsies and in mice lacking exons 2 and 3 of C3 (27,28). In these studies, the apoptotic nuclei were assumed to be myonuclei, based on their location beneath the plasma membrane. We similarly examined the C3KO mouse for apoptotic nuclei using TUNEL staining and observed the presence of apoptotic nuclei; however, these apoptotic nuclei tended to occur near areas of necrosis and regeneration, and were outside the dystrophin-demarcated, sarcolemmal membrane (Fig. 3H). To ascertain the identity of the apoptotic cells, double staining for apoptosis and for markers of different cell types was performed. This analysis revealed that almost all apoptotic nuclei in the C3KO muscle co-stained for CD11b, an immune cell marker (Fig. 3I and J). Thus, nuclei in the C3KO muscle do not appear to be myonuclei. Since immune cells undergo apoptosis at the resolution of an inflammatory response, this finding suggests that in our model apoptosis is not a primary pathogenic mechanism occurring in murine calpainopathy but is rather a secondary consequence of muscle damage and inflammation (29).

Myotubes from C3KO are delayed developmentally
The finding that apoptosis is not a primary mechanism contributing to atrophy and degeneration of the C3KO muscles suggests the existence of alternative mechanisms occurring in the pathogenesis of calpainopathy. Previous results in transgenic mice overexpressing C3 had suggested the possibility that C3 might play a role in muscle maturation (13); however, the limited availability of biopsy material from human patients has prevented an in-depth morphological and developmental assessment of LGMD2A muscles. The generation of the C3KO provided the opportunity to examine developmental processes in C3-deficient muscles.

A role for C3 in myoblast fusion and myofibrillogenesis was assessed in myogenic cells that were isolated from the C3KO and WT mice at 8 days of age. Myogenic cells were extracted from the C3KO muscles and were enriched for myoblasts using the pre-plate technique. Isolated cells were >90% myogenic cells as assessed by desmin staining in conjunction with DAPI (Fig. 4A–D). Myogenic cells were isolated with the same efficiency as WT and were able to fuse to form myotubes (Fig. 4E–H). Occasionally, C3KO myotubes were present that had disorganized architecture with clusters of nuclei in one region of the myotube, while the majority of the myotubes were indistinguishable from WT. Myofibril formation in vitro was examined in myotubes grown on Thermanox plastic coverslips, fused for 3 days and then subjected to electron microscopy (EM). Cells used for EM are shown in Figure 5A and B. At 3 days post-fusion, myotubes from the WT mice showed clear evidence of myofibrillogenesis. In some areas of the cell, thick filaments could be observed aligning to form A-bands (Fig. 5C). Other areas showed periodicities of A-bands separated by Z-discs (Fig. 5C and E). While the C3KO cells were clearly multinucleated, indicating fusion, there was no evidence for organized sarcomeres (Fig. 5D and F). While thick filaments and Z-bodies could be observed in the C3KO myotubes (Fig. 5D, boxed area), these structures were not well organized. Thus, a comparison between the WT and the C3KO myotubes by EM revealed morphological evidence of delayed myofibrillogenesis in C3KO.



View larger version (75K):
[in this window]
[in a new window]
  		Figure 4. The C3KO myotubes do not show any delay in the fusion process. (A–D) Immunostaining for the myogenic cell marker desmin shows that both the WT and the C3KO primary cultures were >90% enriched in the myoblasts. Cells were double-stained with anti-desmin antibodies (A,B) and DAPI (C,D) to show all nuclei. (E, F) In 24 h after switching to differentiation medium, fusion of myotubes is evident in both cultures. (G, H) In 2.5 days, >50% of myotubes in both the WT (G) and the KO (H) cultures have more that five nuclei. One representative picture for each culture is shown. Bar in (A–D), 25 µm. Bar in (E–H), 50 µm.

 


View larger version (172K):
[in this window]
[in a new window]
  		Figure 5. Abnormal sarcomere assembly in the C3KO myotubes in vitro. (A, B) The C3KO primary myoblasts (B) fused to form myotubes with the same efficiency as WT (A). Shown are the myotubes after 3 days in differentiation media. Bar, 50 µm (C–F) EM of 3 day myotubes. While sarcomere assembly is evident in the WT cells (C), very little organization can be seen in the KO cells (D). Bar in (C), 5 µm. Bar in (D), 2 µm. (E, F) Higher magnification (3x) of the areas shown in (C) and (D), respectively.

 
The MHC isoform expression was also examined to assess the progression of myofibrillogenesis in developing myotubes. According to the premyofibril model of myofibrillogenesis, non-muscle myosin II should precede the placement of muscle-specific isoforms in the sarcomere during early myofibrillogenesis (30). To examine the relative concentrations of myosin isoforms, immunoblotting of the primary cultures was carried out for different isoforms of MHC. This analysis showed that while the WT myotubes expressed very little non-muscle MHC, they expressed abundant amounts of embryonic and mature forms of muscle-specific MHCs (Fig. 6). Conversely, the C3KO myotubes expressed more non-muscle myosin II and less muscle-specific forms of MHC. Other proteins such as actin, {alpha}-actinin and desmin were expressed to the same extent as in the WT cells (Fig. 6). Thus, the delay in sarcomere assembly and muscle myosin expression lends both morphological and biochemical support for the hypothesis that the C3KO myotubes have reductions in organized myofibrillogenesis. These findings are also consistent with the pre-myofibril model of myofibrillogenesis (see Discussion).



View larger version (44K):
[in this window]
[in a new window]
  		Figure 6. Western blot analysis shows reduced muscle-specific MHC expression in the C3KO myotubes. Identical western blots were probed with antibodies against different muscle markers. There was no difference in the expression of actin, desmin or {alpha}-actinin. The C3KO myotubes express reduced amount of muscle-specific MHC isoforms. Staining with anti-C3 antibody, pIS2, confirms the absence of the C3 in the C3KO cells. Ponceau red staining of the membrane after western transfer is included to show the relative amount of proteins in the cell lysates. The KO and WT lanes are indicated at the top.

 
The C3KO sarcomeres show misaligned A-bands
The findings of delayed sarcomere formation in the C3KO myotubes suggested a role for C3 in remodeling that occurs during myofibrillogenesis; however, we asked whether this developmental delay might impact the ultrastructure of the adult C3KO muscle. The adult C3KO muscles were examined by EM, to assess ultrastructural features. The EM analysis revealed that while A-bands of the WT muscle were aligned along the edges (Fig. 7A and B), A-bands of the C3KO muscles were frequently out of register, giving the edges of the A-band a ‘ragged’ appearance (Fig. 7C and D). (Scheme of the skeletal muscle sarcomere is presented on Fig. 8A.) Care was taken to examine fibers that were not degenerating and to examine tissue that was not damaged during preparation. Thus, the presence of abnormal A-bands in the C3KO mice suggests a role for C3 in correct formation of sarcomeres or maintenance of sarcomere alignment.



View larger version (162K):
[in this window]
[in a new window]
  		Figure 7. Ultrastructural analysis reveals abnormal sarcomere organization and mitochondria abundance in the C3KO soleus. (A, B) EM of WT soleus. Arrows point to the well-aligned thick filaments (A-band) in (B). (C, D) EM shows misalignment of thick filaments in the C3KO soleus (arrows in D point to uneven edges of A-bands). Scheme of the skeletal muscle sarcomere is shown on Fig. 8(A). (E) Some areas of the C3KO soleus have abundant, disorganized mitochondria (white arrows). (F) Light microscopy image of hematoxilin-stained soleus section of the 16-month-old C3KO mouse. Arrows indicate multiple lobulated fibers. Bar in (A), (C) and (E), 2 µm. Bar in (B) and (D), 0.5 µm. Bar in (F), 50 µm.

 


View larger version (79K):
[in this window]
[in a new window]
  		Figure 8. Titin localization is normal in the C3KO muscles. (A) Scheme of the skeletal muscle sarcomere. One titin molecule spreads for half of the sarcomere length. The amino-terminal regions of titin molecules from adjacent sarcomeres overlap in the Z-disk; the carboxy-terminal regions of titin molecules from the opposite half-sarcomere overlap in the M-line. C3-binding sites as identified by yeast two-hybrid analysis are shown. (B–D) Double immunostaining of longitudinal sections of the C3KO muscle with M-line titin antibody (red) and Z-disk marker desmin [(B), green], N2-line titin [(C), green], I–A junction titin antibody [(D), green]. (E, F) Three-day primary myotubes of the C3KO (E) or WT (F) stained with titin M-line specific antibodies. Bar in (B), (E) and (F), 5 µm. Bar in (C) and (D), 2 µm.

 
By EM, abundant and disorganized mitochondria were observed frequently in the C3KO muscles (Fig. 7E). This ultrastructural feature has been documented previously to occur in the LGMD2A biopsies (31,32) and these structures can manifest as lobulated fibers by light microscopy (31). We did not observe lobulated fibers in muscles of young (2–3 month) KO mice by light microscopy, but they became apparent as the mice aged (16 month, Fig. 7F). Thus, the presence of abundant, disorganized mitochondria and lobulated fibers further validates this mouse as a good model of calpainopathy in humans.

Titin immunohistochemistry of muscle cells in vivo and in vitro indicates that M-line and N2-line epitopes of titin are normally localized in the C3KO muscles
We tested whether the influence of C3 on myofibrillogenesis might be mediated through its interactions with titin, by investigating whether the absence of C3 affects the assembly of titin in fully differentiated skeletal muscle fibers in vivo. Titin's distribution in the C3KO muscles was examined by double staining longitudinal sections using antibodies against three different epitopes of titin (N2A-line, M-line and I–A junctional region) and a Z-disk marker, desmin. A single molecule of titin extends from the Z-disk to the M-line (half of the sarcomere). Titin molecules from the same sarcomere overlap within the M-line, while titin molecules from the adjacent sarcomere overlap within the Z-disk. Therefore, there are two N2A-line and I–A junction epitopes per sarcomere, and one M-line epitope of titin per sarcomere (Fig. 8A). These immunolocalization studies demonstrated that the distribution of all titin epitopes examined in the C3KO muscles were indistinguishable from the WT muscles (Fig. 8B–D).

To examine myofibrillogenesis in vitro, immunohistochemistry using antibodies against different sarcomeric proteins was carried out on primary cultures undergoing differentiation. While proteins such as {alpha}-actinin formed a regular striated pattern in most myotubes, many other sarcomeric proteins were not yet organized (for example, MHC and desmin, data not shown). We found that the resolution of light microscopy was not high enough to detect a striated pattern for all sarcomeric proteins in the primary cultures, and was thus not sensitive enough to provide substantial information on myofibrillogenesis of the C3KO myotubes. These observations also suggest that skeletal myotubes do not differentiate in vitro as do other primary culture systems that have been used previously to study myofibrillogenesis, such as chicken cardiac muscle. In the C3KO myotubes in vitro, both N2-line (data not shown) and M-line specific titin antibodies stained titin with a periodic pattern (Fig. 8E and F). Thus, the assembly of titin into a sarcomeric pattern does not require the presence of C3. These results suggest that defects in myofibrillogenesis in the C3KO primary cultures occur subsequent to titin placement in the sarcomere but prior to thick filament alignment.

Titin interacts with C3 and can serve as its substrate
Since the C3 and titin interaction was shown previously by yeast two-hybrid analysis (15), we sought to confirm this interaction at the protein level by performing further biochemical analyses. Following cloning and sequencing of titin domains, the protein fragments were expressed in insect cells and purified. For these studies, three regions of titin were cloned by PCR from mouse muscle cDNA using primers that were designed based on the human sequence. The regions cloned included N2-line, M-line and PEVK domains of titin. The N2- and M-lines were chosen because yeast two-hybrid analysis had mapped the binding to these regions. The PEVK region was speculated previously to have inhibitory activity for C3 based on its homology to the ubiquitous calpain inhibitor calpastatin (33). Following purification of the proteins, solid phase ELISA binding assays were performed. These assays confirmed yeast two-hybrid analysis of the binding of C3 to the N2-line and M-line of titin. C3 did not show any binding to the PEVK region of titin. This latter finding validates the specificity of the ELISA binding assay (Fig. 9A and B).



View larger version (52K):
[in this window]
[in a new window]
  		Figure 9. C3 interacts with M-line and N2-line fragments of titin and can cleave PEVK and M-line fragments. (A) Increasing amounts of calpain were incubated with 10 µg of the immobilized titin fragments in solid phase binding assays. The amount of bound calpain was determined with anti-calpain 12A2 antibodies. (B) Relative binding of 5 µg of C3 to the different fragments of titin shown as a ratio of the OD reading of the experimental sample to the background reading determined for several controls. The data from three independent experiments are presented. For details see Materials and Methods. (C, D) Titin fragments alone or in combination with either WT or inactive C129S mutant of C3 were expressed in a baculovirus system. Identical western blots were probed with anti-His antibody to detect titin fragments (C) or anti-C3 antibody 12A2 (D). (C) White arrowheads point to the uncleaved titin fragments, black arrowheads indicate cleavage products of PEVK and M-line of titin after co-expression with WT C3. Notice the absence of the cleavage products after co-expression with inactive C129S mutant. (D) Arrows on the left of blot show position of the full-length 94 kDa C3 (p94) or its autoproteolytic 55 kDa product (p55). C129S mutant protein is resistant to autoproteolytic degradation.

 
Previous findings of a secondary deficiency of C3 in patients carrying a titin mutation and in mdm mice have lead to the suggestion that titin may serve to stabilize C3 and prevent it from autolytically degrading (17,19). Titin binding by C3 might also place C3 in the correct proximity of its proper substrates, and we considered the possibility that titin itself might be a substrate for C3. Previous studies have shown that a small region of the N2-line (53 kDa) was not cleaved by C3 (34). In addition, a small portion of the M-line (Mex 5 region) of titin was observed to be cleaved by C3 (24). Our baculoviral co-expression studies, in which both C3 and larger regions of the titin domains were co-expressed, replicated these findings. In addition, we showed for the first time that the PEVK domain is also cleaved. These studies demonstrated cleavage of both the M-line and PEVK regions of titin by C3, but no cleavage of the N2-line region (Fig. 9C and D). Thus, titin can serve as a substrate for C3 in regions adjacent to where it binds. Succinct cleavage of titin by C3 may allow for the exchange of proteins on the titin scaffold that is a necessary occurrence for remodeling during myofibrillogenesis.

Some pathogenic human mutations reduce the affinity of C3 for titin
Previous studies have demonstrated that the majority of mutations that have been identified in the LGMD2A patients result in loss of enzyme activity rather than loss of binding to titin (20,35), although a few mutations have been identified that result in reduced binding to some titin domains by yeast two-hybrid analysis (20). The confirmation of the binding between C3 and titin led us to question whether this interaction might also be important for disease pathogenesis. Therefore, we tested to see if mutations in C3 that are not predicted to affect protease activity (35) might affect binding between titin and C3. Four different C3 mutations that are known to be pathogenic in humans were examined using the assay system we established (Fig. 10). All four mutations tested exhibited reduced binding to titin compared with the WT C3. In contrast, the inactive C129S mutation did not affect binding to titin. No binding was observed with the PEVK region of titin, confirming our previous observations (Fig. 9). Thus, these data support the hypothesis that some mutations that reduce titin/C3 interactions might destabilize and inactivate C3 or place C3 out of proximity of its endogenous substrate(s).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 10. Recombinant C3 proteins carrying pathogenic human LGMD2A mutations show reduced binding to the titin fragments. Four mutations that are not predicted to affect the proteolytic activity of C3 (35), V98I, I162L, R448H and D705G were tested in the solid-phase binding assay. All four mutant proteins demonstrate reduction in binding to the titin M-line (A) and N2-line (B) fragments. In contrast, the proteolytically inactive mutant C129S shows no reduction compared to the WT C3. Data are presented as averages of three experiments, and differences between WT and mutant forms of C3 (except for the C129S) are statistically significant (P<0.05).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The previous, unexpected discovery that mutations in the protease C3 resulted in a muscular dystrophy did not fit easily into the generalized model of muscular dystrophy in which defects of the dystrophin-associated protein complex were expected to account for most of the dystrophic pathologies. However, the more recent discoveries that mutations in titin or titin-binding proteins could also cause severe muscular dystrophies have indicated that defects in the titin-associated protein complex could underlie a second family of muscular dystrophies. Proteins in this complex that have muscular dystrophy associations include myotilin, telethonin, titin and C3 (12,36–38). Because C3 is a member of the titin protein complex, the C3 mutations could relate to this second family of dystrophies, although the mechanism through which the C3 mutations could lead to dystrophy was unknown. Our present findings show that C3 is necessary for normal myofibril formation in vivo and in vitro. In addition, we have shown that null mutation of C3 produces disruptions of normal patterns of expression of MHC, reductions in muscle mass and smaller muscle fibers. Together, these observations indicate that C3-mediated processes play a significant role in muscle differentiation and growth. It is possible that defects in this mechanism may be a common pathway in all dystrophies involving mutations in the titin protein complex; however, this hypothesis has not yet been tested.

Our finding that null mutation of C3 produces defects in sarcomere structure in vivo and in vitro may reflect disruption of titin's role in sarcomere formation that is influenced by C3. This possibility is supported by the present investigation in which we show that titin is a substrate for C3. If the C3-mediated, structural modifications in titin were an important feature of sarcomere remodeling during muscle growth or regeneration, loss of C3 could be reflected in the structural and developmental defects in sarcomeres reported here. While it is unlikely that cleavage of titin by C3 is responsible for titin turnover, it is interesting to note that the half-life of titin in sarcomeres (2.9 days) (39) is much briefer than other sarcomeric proteins (~6–9 days) (40), despite its tremendous size, which suggests that its relatively rapid turnover is important for normal homeostasis of muscle. Our data showing succinct cleavage of titin by C3 lend support for a role for C3 in the cleavage-dependent release of proteins from titin, rather than for widespread degradation of titin by C3. Future studies will be devoted for determining the C3 targets involved in this process.

The ability of C3 to cleave many different proteins, at least in vitro, suggests that it may be involved in multiple processes and that its removal from muscle may have pleiotropic effects. For example, extensive myonuclear apoptosis was attributed previously to loss of C3 function in a mouse line in which two exons in the catalytic domain of C3 were deleted, but the rest of the protein was preserved (28). It was proposed that apoptosis observed in this mouse was induced by the absence of C3 protease activity and is the primary pathogenic mechanism in LGMD2A. Although we observed some apoptotic nuclei in the C3KO muscles, the majority of these nuclei were found in or near necrotic lesions, and were located outside of the sarcolemma. Most of the apoptotic nuclei were located in immune cells that invaded the area of pathology. Since it is well known that apoptosis is the common mechanism through which immune cells are removed following activation, we conclude that apoptosis in murine calpainopathy is secondary to muscle inflammation (29). Therefore, these studies suggest that apoptosis is not a primary feature of the pathogenesis of calpainopathy.

Although additional, unidentified, functional or developmental defects may exist in the C3-deficient muscle, our current findings show that disruptions of myofibrillogenesis and sarcomere structure are the most prominent pathological features in murine calpainopathy. Our observations are consistent with a role for C3 in the three-step model of myofibrillogenesis: premyofibrils to nascent myofibrils to mature myofibirils (30). According to the model, the premyofibrils consist of {alpha}-actinin (Z-bodies) connected to actin filaments, and non-muscle myosin II filaments. These premyofibrils align and are modified by the addition of titin and muscle myosin filaments to form nascent myofibrils. Mature myofibrils are formed when non-muscle myosin is eliminated and Z-bodies fuse (Fig. 11, modified from 41). The premyofibril model is supported by several immunolocalization studies as well as observations on live embryonic cardiomyocytes transfected with a GFP–{alpha}-actinin fusion protein (30). A recent study confirmed that myofibrils can contain both non-muscle and muscle-specific myosins, and that bands of non-muscle myosin II alternate with bands of sarcomeric isoforms of {alpha}-actinin in early myofibrils in cardiomyocyte cultures (41). Our data are consistent with the premyofibril model by demonstrating that non-muscle myosin is present in primary cultures of skeletal muscle myotubes, but is later replaced by muscle myosin. The absence of C3 caused a delay in the replacement of non-muscle myosin with muscle-specific isoforms that coincided with delayed myofibrillogenesis as revealed by the EM observations of the C3KO cells compared to WT cells.



View larger version (31K):
[in this window]
[in a new window]
  		Figure 11. Premyofibril model of myofibrillogenesis (adapted from 41). Diagram shows three steps of myofibrillogenesis; premyofibrils contain non-muscle myosin which is being replaced with muscle-specific myosin during the nascent myofibril stage. Titin associated with C3 may facilitate the replacement of non-muscle myosin and further sarcomere assembly.

 
Our observations also show that although the absence of C3 leads to dramatic disruption of myofibrillogenesis in cell culture, the effect is much more mild in vivo. We speculate that under physiological conditions in vivo, other calpains may also contribute to this process and lead to partial rescue of the phenotype. This functional redundancy was demonstrated for conventional calpains in other genetically modified mice. For example, a knockout of the small subunit, which is shared by both calpains 1 and 2, leads to embryonic lethality (42), whereas knockout of only the large subunit of calpain 1 results in a mild phenotype of platelet dysfunction (43). Further studies are needed to establish the level of functional specificity and redundancy among the calpain family.

All models of myofibrillogenesis agree on the important role of the giant molecule titin as a molecular ruler that orchestrates alignment of actin and myosin filaments, as well as other proteins associated with them along the length of sarcomere (14). Titin is one of the earliest proteins expressed during myofibrillogenesis, forming a striated pattern in developing myotubes prior to the organization of other sarcomeric proteins. A functional knockout of titin in a myofibroblast line causes lack of thick filament formation and ordered actin–myosin arrays supporting a role for the titin in coordination of myofibrillogenesis (44). Early yeast two-hybrid investigations showed that C3 is one of the numerous binding partners of titin (15,16,45). Moreover, titin may play a role in the regulation of the C3 activity by inhibiting the C3 autolysis, thus stabilizing it, or placing C3 in the correct proximity of its substrates. This assertion is based on the reduction of C3 in muscle extracts from the titin-mutated tibial muscular dystrophy patients (19) as well as from mdm mice carrying a deletion of the titin domain that binds C3 (17). Our findings provide further evidence for a specific interaction between M-line and N2-line fragments of titin with C3 at the protein level. We also examined several C3 proteins carrying mutations that have been identified in patients with pathogenic LGMD2A. All of these mutations are located outside the catalytic domain, and are unlikely to affect the proteolytic function of C3, as expected from computer modeling of various known LGMD2A point mutations on the predicted crystal structure (35). Our findings suggest that these mutations affect the efficiency of C3 binding to titin and support the hypothesis that loss of titin–C3 interaction can lead to pathogenesis. Future studies will be directed towards establishing whether loss of C3 binding to titin is sufficient to cause disease processes.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Antibodies
Anti-C3 antibodies used for western blot analysis included: mouse anti-12A2 and 11B3 (1 : 100, Novocastra), pIS2 goat polyclonal (1 : 1000) and pNS rabbit polyclonal (1 : 1000) from Dr H. Sorimachi, RP4 rabbit polyclonal (1 : 500, Triple Point Biologics), Pabcpn1-3 rabbit polyclonal antibody (1 : 2500) was generous gift of Dr Satoru Noguchi (23). Other antibodies used for western blotting included: mouse anti-fast MHC, anti-slow MHC, and anti-desmin mAb (Novocastra), rabbit anti-{alpha}-actinin and anti-actin (Sigma), mouse anti-His and anti-GST (Amersham Pharmacia). Anti-titin antibodies used for immunohistochemistry included: anti-titin M-line (M8/M9) X246 rabbit polyclonal (1 : 100) and avian anti-titin N2A-line (I72/I73) X105-X106 (1 : 100) were a generous gift from Dr Siegfried Labeit; mouse anti-titin I–A junction was purchased from Novocastra (1 : 100). Other antibodies used for immunohistochemistry included: mouse anti-desmin and anti-developmental MHC (Novocastra), rat anti-CD11b (BD Biosciences), mouse anti-dystrophin (C-terminal-specific) was a kind gift from Dr Louise Anderson. The mouse anti-myogenin (developed by Dr W.E. Wright), anti-non-muscle myosin II mAb (developed by Dr G.W. Conrad and Dr A.H. Conrad) and anti-embryonic MHC mAb (developed by Dr H.M. Blau) were obtained from Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA, USA.

Generation of the C3 null mice
The C3KO mice were obtained from Lexicon Genetics Incorporated. Mice were generated using embryonic stem cells that were mutagenized by a retroviral vector. This vector generates fusion proteins of neomycin with the 5' end of the gene and a fusion with BTK at the 3' end of the gene. These fusions introduce premature stop signals, preventing translation of the protein product. Sequencing was performed to confirm that the vector had integrated in the intron between exons 1' and 2. Extensive RT–PCR analysis showed that there were no RNAs for any of the C3 isoforms present in the tissue of the KO. Original founder mice were 129 SvEvBrdxC57 BL/6 heterozygotes from the F2 generation. Inbreeding of the F2 mice produced the KO and WT homozygotes that were then interbred until the F5 generation. All data were collected from mice from the F5 generation that were generated in parallel. Similar results were observed in mice from earlier generations.

Muscle histology and immunohistochemistry
Muscles to be used for immunohistochemistry were dissected from the mice and frozen in isopentane cooled in liquid nitrogen. Frozen sections were cut at 10 µm and kept frozen until use. After thawing, sections were treated with 0.3% H2O2 for 5 min (if horseradish peroxidase was used for color reaction) and blocked in phosphate-buffered saline (PBS) with 0.2% gelatin, 0.5% Tween-20 and 3% bovine serum albumin (BSA) for 30 min. Binding to endogenous mouse IgG was blocked with a MOM Kit (Vector Laboratories). After primary and biotin-conjugated secondary antibodies, sections were incubated with avidin-conjugated horseradish peroxidase and stained using AEC or DAB substrate kits (Vector Laboratories).

TUNEL labeling and immunostaining
TUNEL labeling was performed as described previously (46) using biotin-16-dUTP (Ezolife Science) and TdT (Amersham). If combined with immunostaining, the proteinase K treatment was omitted. The TUNEL reaction was detected with Ni-DAB Substrate Kit (Vector Laboratories) to produce black staining. After the TUNEL reaction, sections were washed in PBS, blocked in PBS with 0.2% gelatin, 0.5% Tween-20 and 3% BSA for 30 min, and immunostained. AEC Substrate Kit (Vector Laboratories) was used to produce red staining.

Fiber area measurements
Fiber cross-sectional area was measured for 150–300 individual fibers in cross-section of gastrocnemius muscles from each of three to four WT or C3KO animals. Sections were stained with anti-slow MHC antibodies, and slow and fast twitch muscles were measured separately using a digitized imaging system (Bioquant, Nashville, TN, USA).

Electron microscopy
Tissue and cells for EM were prepared as described previously (46). Briefly, solei were dissected from three WT and three C3KO mice and fixed in 1.4% glutaraldehyde in 0.2 M sodium cacodylate buffer, pH 7.2 for 30 min on ice, followed by buffer rinse and fixation for 30 min in 1% osmium tetroxide. The same protocol was used for fixation of primary cells grown on Thermanox plastic coverslips (Nunc). Sections were prepared from the epoxy resin embedded samples and examined under the electron microscope model JEM-1200EX (JEOL, Japan) equipped with BioScan 600W digital camera (1024x1024 pixels). DigitalMicrograph 1.2 software (Gatan) was used to generate images. Four animals were examined for the EM analysis. For cell culture experiments, two grids containing 50–100 cells were analyzed for two different cultures, thus 200–300 cells were analyzed for each genotype.

Primary muscle cell culture
Myogenic cells were isolated from the C3KO and WT mice in parallel as described previously (47,48). Three, eight-day-old mice were used for each culture. Cells were maintained in Ham's F10 medium supplemented with 20% fetal bovine serum, 5 ng/ml basic fibroblast growth factor, 100 U/ml penicillin and 100 µg/ml on gelatin-coated or E-C-L (Upstate Biotechnology) coated dishes. Myoblast fusion was induced by switching to differentiation media, Dulbecco's modified Eagle's medium supplemented with insulin-transferrin-selenium A (GIBCO) for 48–72 h.

Cloning of titin fragments and mutagenesis of C3
Titin fragments were produced by RT–PCR using degenerate primers designed based on the human titin sequence (NM_133378, GenBank). Titin N2A-line fragment (28 316–29 860 bp) contained the minimal C3-binding site (15). The PEVK region spanned from 29 846 to 36 285 bp. Titin M-line (101 254–103 252 bp) included domains M6–M10 and the is 7 region. All PCR products were verified by sequencing. The inactive stable mutant of C3, C129S, was obtained from Dr Hiroyuki Sorimachi (49). Other mutations (V98I, I162L, R448H, D705G) were generated in the full-length WT C3 cDNA in pGEX2T using the QuickChange site-directed mutagenesis kit (Stratagene). For primers used to generate these cloned regions, see Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Primers used for generation of the C3 and titin PCR fragments
 
Protein expression and purification
For protein expression all titin fragments were subcloned into baculoviral transfer vectors (pAcHLT) with N-terminal 6x His tag. The C3 and C129S mutants were subcloned into pAcG2T baculoviral transfer vectors with GST-tag for expression in a baculovirus system (BD Biosciences). Alternatively, C3 and its mutants were cloned into pGEX2T with a GST-tag for expression in the rare codon strain of Escherichia coli, Rosetta-gammi (Novagen). His-tagged proteins were purified using Ni-NTA agarose according to the manufacturer's recommendation (Qiagen). GST-tagged proteins were purified on GST-microspin columns (Amersham). After purification proteins were dialyzed against 50 mM HEPES-KOH, pH 7.5.

Protein–protein interaction
The ELISA solid phase system was used for binding experiments. Titin fragments (10 µg per well) were used to coat 96-well microtiter plates (ImmunoPlates with MaxiSorp Surface, Nunc) overnight at 4°C. After washing with PBT (PBS and 0.05% Tween-20), the wells were saturated with blocking buffer (2% BSA in PBT) and incubated with different concentrations (0.5–10 µg per well) of C3 or mutant proteins for 2 h at 37°C. After washes, bound C3 was detected by anti-C3 or anti-GST monoclonal primary antibodies followed by biotinylated secondary antibodies and avidin–alkaline phosphatase conjugate. pNPP (p-nitrophenylphosphate) kit (KPL) was used for color reaction. The product formation was recorded by measuring the net change in absorbance at 405 nm.

In order to test the specificity of the protein–protein interactions in the ELISA assays, negative control proteins BSA or His-nitric oxide synthase were used instead of titin fragments. To ensure that binding was not due to GST–GST interactions, GST alone was used instead of C3 on wells coated with titin fragments, and detection was carried out with anti-GST antibody. Since non-specific binding occurred when higher concentrations of proteins were used in the assay, a range of concentrations was analyzed. The datum point chosen for analysis was the highest concentration of protein in which the negative control gave the lowest binding.

Protein co-expression
To test whether C3 could cleave any of the titin fragments, the WT C3 or the inactive C129S mutant was cloned into pVL1393 baculoviral transfer vector without any tags to ensure proper protease activity. Titin and C3 constructs were co-expressed in a baculovirus system. Insect cells were plated on 10-cm cell culture dishes at 50–70% of confluence, and co-infected with both titin and C3 high titer viral stocks. After incubation for 3 days at 27°C, cells were harvested by lysis in Insect Cell Lysis Buffer (BD Biosciences) with protein inhibitor cocktail (Sigma, 1 : 100). Soluble fractions were analyzed by western blotting using anti-His or anti-C3 antibody to detect titin fragments or C3, respectively.


   ACKNOWLEDGEMENTS
 
The authors acknowledge the excellent technical support of Ms Jane Wen. Antibodies to titin M-line and N2-line were generously provided by Dr Siegfried Labeit. Antibodies to calpain 3 (N terminus and IS2) as well as the C129S construct were generously provided by Dr Hiroyuki Sorimachi. The authors would like to thank Dr Jacques Beckmann for valuable discussions about calpain 3, Dr Sergey Ryazantsev and Ms Birgitta Sjostrand for support and advice about EM, and Drs Grace Pavlath and Thomas Rando for advice on primary cultures. The authors thank Drs Joseph and Jean Sanger for helpful discussions and for permission to use their diagram. This work was supported by the NIH (NIAMS, AR48177 to M.J.S. and AR47721 and AR047855 to J.G.T.) and the Muscular Dystrophy Association (to M.J.S.).


   FOOTNOTES
 
* To whom correspondence should be addressed at: Duchenne Muscular Dystrophy Research Center, University of California at Los Angeles, 621 Young Drive South, Los Angeles, CA 90095-1606, USA. Tel: +1 3107945225; Fax: +1 3108258489; Email: mspencer{at}mednet.ucla.edu

{dagger} The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Goll, D.E., Thompson, V.F., Li, H., Wei, W. and Cong, J. (2003) The calpain system. Physiol. Rev., 83, 731–801.[Abstract/Free Full Text]

  2. Ma, H., Fukiage, C., Azuma, M. and Shearer, T.R. (1998) Cloning and expression of mRNA for calpain Lp82 from rat lens: splice variant of p94. Invest. Opthalmol. Vis. Sci., 39, 454–461.[Abstract/Free Full Text]

  3. Welm, A.L., Timchenko, N.A., Ono, Y., Sorimachi, H., Radomska, H.S., Tenen, D.G., Lekstrom-Himes, J. and Darlington, G.J. (2002) C/EBPalpha is required for proteolytic cleavage of cyclin A by calpain 3 in myeloid precursor cells. J. Biol. Chem., 277, 33 848–33 856.

  4. De Tullio, R., Stifanese, R., Salamino, F., Pontremoli, S. and Melloni, E. (2003) Characterization of a new p94-like calpain form in human lymphocytes. Biochem. J., 375, 689–696.[CrossRef][ISI][Medline]

  5. Fox, J.E.B., Reynolds, C.C. and Austin, C.D. (1991) The role of GP IIb–IIIa, an integrin, in regulating activation of calpain in platelets. J. Cell Biol., 115, 5a.

  6. Shea, T.B., Cressman, C.M., Spencer, M.J., Beermann, M.L. and Nixon, R.A. (1995) Enhancement of neurite outgrowth following calpain inhibition is mediated by protein kinase C. J. Neurochem., 65, 517–527.[ISI][Medline]

  7. Barnoy, S., Zipser, Y., Glaser, T., Grimberg, Y. and Kosower, N.S. (1999) Association of calpain (Ca(2+)-dependent thiol protease) with its endogenous inhibitor calpastatin in myoblasts. J. Cell. Biochem., 74, 522–531.[CrossRef][ISI][Medline]

  8. Nixon, R.A. (2003) The calpains in aging and aging-related diseases. Ageing Res. Rev., 2, 407–418.[CrossRef][ISI][Medline]

  9. Schaecher, K., Rocchini, A., Dinkins, J., Matzelle, D.D. and Banik, N.L. (2002) Calpain expression and infiltration of activated T cells in experimental allergic encephalomyelitis over time: increased calpain activity begins with onset of disease. J. Neuroimmunol., 129, 1–9.[CrossRef][ISI][Medline]

  10. Spencer, M.J. and Mellgren, R.L. (2002) Overexpression of a calpastatin transgene in mdx muscle reduces dystrophic pathology. Hum. Mol. Genet., 11, 2645–2655.[Abstract/Free Full Text]

  11. Spencer, M.J., Croall, D.E. and Tidball, J.G. (1995) Calpains are activated in necrotic fibers from mdx dystrophic mice. J. Biol. Chem., 270, 10909–10914.[Abstract/Free Full Text]

  12. Richard, I., Broux, O., Allamand, V., Fougerousse, F., Chiannilkulchai, N., Bourg, N., Brenguier, L., Devaud, C., Pasturaud, P., Roudaut, C. et al. (1995) Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell, 81, 27–40.[CrossRef][ISI][Medline]

  13. Spencer, M.J., Guyon, J.R., Sorimachi, H., Potts, A., Richard, I., Herasse, M., Chamberlain, J., Dalkilic, I., Kunkel, L.M. and Beckmann, J.S. (2002) Stable expression of calpain 3 from a muscle transgene in vivo: immature muscle in transgenic mice suggests a role for calpain 3 in muscle maturation. Proc. Natl Acad. Sci. USA, 99, 8874–8879.[Abstract/Free Full Text]

  14. Gregorio, C.C., Granzier, H., Sorimachi, H. and Labeit, S. (1999) Muscle assembly: a titanic achievement? Curr. Opin. Cell Biol., 11, 18–25.[CrossRef][ISI][Medline]

  15. Sorimachi, H., Kinbara, K., Kimura, S., Takahashi, M., Ishiura, S., Sasagawa, N., Sorimachi, N., Shimada, H., Tagawa, K., Maruyama, K. et al. (1995) Muscle-specific calpain, p94, responsible for limb girdle muscular dystrophy type 2A, associates with connectin through IS2, a p94-specific sequence. J. Biol. Chem., 270, 31158–31162.[Abstract/Free Full Text]

  16. Kinbara, K., Sorimachi, H., Ishiura, S. and Suzuki, K. (1997) Muscle-specific calpain, p94, interacts with the extreme C-terminal region of connectin, a unique region flanked by two immunoglobulin C2 motifs. Arch. Biochem. Biophys., 342, 99–107.[CrossRef][ISI][Medline]

  17. Garvey, S.M., Rajan, C., Lerner, A.P., Frankel, W.N. and Cox, G.A. (2002) The muscular dystrophy with myositis (mdm) mouse mutation disrupts a skeletal muscle-specific domain of titin. Genomics, 79, 146–149.[CrossRef][ISI][Medline]

  18. Witt, C.C., Ono, Y., Puschmann, E., McNabb, M., Wu, Y., Gotthardt, M., Witt, S.H., Haak, M., Labeit, D., Gregorio, C.C. et al. (2004) Induction and myofibrillar targeting of CARP, and suppression of the Nkx2.5 pathway in the MDM mouse with impaired titin-based signaling. J. Mol. Biol., 336, 145–154.[CrossRef][ISI][Medline]

  19. Haravuori, H., Vihola, A., Straub, V., Auranen, M., Richard, I., Marchand, S., Voit, T., Labeit, S., Somer, H., Peltonen, L. et al. (2001) Secondary calpain3 deficiency in 2q-linked muscular dystrophy: titin is the candidate gene. Neurology, 56, 869–877.[Abstract/Free Full Text]

  20. Ono, Y., Shimada, H., Sorimachi, H., Richard, I., Saido, T.C., Beckmann, J.S., Ishiura, S. and Suzuki, K. (1998) Functional defects of a muscle-specific calpain, p94, caused by mutations associated with limb-girdle muscular dystrophy type 2A. J. Biol. Chem., 273, 17073–17078.[Abstract/Free Full Text]

  21. Zambrowicz, B.P., Friedrich, G.A., Buxton, E.C., Lilleberg, S.L., Person, C. and Sands, A.T. (1998) Disruption and sequence identification of 2,000 genes in mouse embryonic stem cells. Nature, 392, 608–611.[CrossRef][Medline]

  22. Kawabata, Y., Hata, S., Ono, Y., Ito, Y., Suzuki, K., Abe, K. and Sorimachi, H. (2003) Newly identified exons encoding novel variants of p94/calpain 3 are expressed ubiquitously and overlap the alpha-glucosidase C gene. FEBS Lett., 555, 623–630.[CrossRef][ISI][Medline]

  23. Keira, Y., Noguchi, S., Minami, N., Hayashi, Y.K. and Nishino, I. (2003) Localization of calpain 3 in human skeletal muscle and its alteration in limb-girdle muscular dystrophy 2A muscle. J. Biochem. (Tokyo), 133, 659–664.[Abstract/Free Full Text]

  24. Taveau, M., Bourg, N., Sillon, G., Roudaut, C., Bartoli, M. and Richard, I. (2003) Calpain 3 is activated through autolysis within the active site and lyses sarcomeric and sarcolemmal components. Mol. Cell. Biol., 23, 9127–9135.[Abstract/Free Full Text]

  25. Fanin, M., Nascimbeni, A.C., Fulizio, L., Trevisan, C.P., Meznaric-Petrusa, M. and Angelini, C. (2003) Loss of calpain-3 autocatalytic activity in LGMD2A patients with normal protein expression. Am. J. Pathol., 163, 1929–1936.[Abstract/Free Full Text]

  26. Vainzof, M., de Paula, F., Tsanaclis, A.M. and Zatz, M. (2003) The effect of calpain 3 deficiency on the pattern of muscle degeneration in the earliest stages of LGMD2A. J. Clin. Pathol., 56, 624–626.[Abstract/Free Full Text]

  27. Baghdiguian, S., Martin, M., Richard, I., Pons, F., Astier, C., Bourg, N., Hay, R.T., Chemaly, R., Halaby, G., Loiselet, J. et al. (1999) Calpain 3 deficiency is associated with myonuclear apoptosis and profound perturbation of the IkappaB alpha/NF-kappaB pathway in limb-girdle muscular dystrophy type 2A [Erratum (1999) Nat. Med., 5 (7), 849.] Nat. Med., 5, 503–511.[CrossRef][ISI][Medline]

  28. Richard, I., Roudaut, C., Marchand, S., Baghdiguian, S., Herasse, M., Stockholm, D., Ono, Y., Suel, L., Bourg, N., Sorimachi, H. et al. (2000) Loss of calpain 3 proteolytic activity leads to muscular dystrophy and to apoptosis-associated IKBa/Nuclear Factor KB pathway perturbation in mice. J. Cell Biol., 151, 1–9.[Abstract/Free Full Text]

  29. Tidball, J.G. and St. Pierre, B. (1996) Apoptosis of macrophages during the resolution of muscle inflammation. J. Leukoc. Biol., 59, 380–388.[Abstract]

  30. Sanger, J.W., Chowrashi, P., Shaner, N.C., Spalthoff, S., Wang, J., Freeman, N.L. and Sanger, J.M. (2002) Myofibrillogenesis in skeletal muscle cells. Clin. Orthop., S153–S162.

  31. Chae, J., Minami, N., Jin, Y., Nakagawa, M., Murayama, K., Igarashi, F. and Nonaka, I. (2001) Calpain 3 gene mutations: genetic and clinico-pathologic findings in limb-girdle muscular dystrophy. Neuromuscul. Disord., 11, 547–555.[CrossRef][ISI][Medline]

  32. Kawai, H., Akaike, M., Kunishige, M., Inui, T., Adachi, K., Kimura, C., Kawajiri, M., Nishida, Y., Endo, I., Kashiwagi, S. et al. (1998) Clinical, pathological, and genetic features of limb-girdle muscular dystrophy type 2A with new calpain 3 gene mutations in seven patients from three Japanese families. Muscle Nerve, 21, 1493–1501.[CrossRef][ISI][Medline]

  33. Sorimachi, H., Ono, Y. and Suzuki, K. (2000) Molecular analysis of p94 and its application to diagnosis of limb girdle muscular dystrophy type 2A. Methods Mol. Biol., 144, 75–84.[Medline]

  34. Kinbara, K., Ishiura, S., Tomioka, S., Sorimachi, H., Jeong, S.Y., Amano, S., Kawasaki, H., Kolmerer, B., Kimura, S., Labeit, S. et al. (1998) Purification of native p94, a muscle-specific calpain, and characterization of its autolysis. Biochem. J., 335, 589–596.

  35. Jia, Z., Petrounevitch, V., Wong, A., Moldoveanu, T., Davies, P.L., Elce, J.S. and Beckmann, J.S. (2001) Mutations in calpain 3 associated with limb girdle muscular dystrophy: analysis by molecular modeling and by mutation in m-calpain. Biophys. J., 80, 2590–2596.[Abstract/Free Full Text]

  36. Hauser, M.A., Horrigan, S.K., Salmikangas, P., Torian, U.M., Viles, K.D., Dancel, R., Tim, R.W., Taivainen, A., Bartoloni, L., Gilchrist, J.M. et al. (2000) Myotilin is mutated in limb girdle muscular dystrophy 1A. Hum. Mol. Genet., 9, 2141–2147.[Abstract/Free Full Text]

  37. Moreira, E.S., Wiltshire, T.J., Faulkner, G., Nilforoushan, A., Vainzof, M., Suzuki, O.T., Valle, G., Reeves, R., Zatz, M., Passos-Bueno, M.R. et al. (2000) Limb-girdle muscular dystrophy type 2G is caused by mutations in the gene encoding the sarcomeric protein telethonin. Nat. Genet., 24, 163–166.[CrossRef][ISI][Medline]

  38. Hackman, P., Vihola, A., Haravuori, H., Marchand, S., Sarparanta, J., De Seze, J., Labeit, S., Witt, C., Peltonen, L., Richard, I. et al. (2002) Tibial muscular dystrophy is a titinopathy caused by mutations in TTN, the gene encoding the giant skeletal-muscle protein titin. Am. J. Hum. Genet., 71, 492–500.[CrossRef][ISI][Medline]