Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
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 (16)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Nawrotzki, R.
Right arrow Articles by Mayer, U.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nawrotzki, R.
Right arrow Articles by Mayer, U.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 5 483-495
© 2003 Oxford University Press

Defective integrin switch and matrix composition at alpha 7-deficient myotendinous junctions precede the onset of muscular dystrophy in mice

Ralph Nawrotzki1,2, Michael Willem1, Nicolai Miosge3, Heinrich Brinkmeier4 and Ulrike Mayer5,*

1Max-Planck-Institute for Biochemistry, 82152 Martinsried, Germany, 2Department of Anatomy and Cellular Neurobiology, University of Ulm, 89069 Ulm, Germany, 3Department of Histology, University of Göttingen, 37075 Göttingen, Germany, 4Institute of Pathophysiology, University of Greifswald, 17495 Karlsburg, Germany and 5Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, University of Manchester, Manchester M13 9PT, UK

Received October 18, 2002; Accepted December 17, 2002


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Force transmission at the myotendinous junction requires a strong link between the muscle cytoskeleton and the extracellular matrix. At the adult junction, two splice variants of the laminin-binding integrins, {alpha}7Aß1D and {alpha}7Bß1D, are highly enriched. The {alpha}7 subunits are critical for the integrity of the junctional sarcolemma because integrin {alpha}7-deficient mice develop muscular dystrophy, primarily affecting this site of the muscle. Here, we report that ß1D integrin coimmunoprecipitates and colocalizes with the {alpha}5 subunit at {alpha}7-deficient junctions, but does not associate with {alpha}3, {alpha}6 or {alpha}v integrins. By immunogold labelling we show that the basement membranes of integrin {alpha}7-deficient muscles recruit abnormally high levels of fibronectin, the ligand of {alpha}5ß1D. Finally, we demonstrate that {alpha}5ß1D is down-regulated at the normal postnatal junction and is displaced by {alpha}7ß1D. These results suggest that the {alpha}7 subunit is implicated in the down-regulation of {alpha}5ß1D and in the removal of fibronectin from the maturing myotendinous junction, thus providing an {alpha}7ß1D-based link to laminin. We propose that the persistence of {alpha}5ß1D in {alpha}7-deficient mice is not compatible with normal muscle function and leads to muscle wasting.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The transduction of force from the muscle fibre to the tendon is a prerequisite for vertebrate motility. It takes place at the myotendinous junction (MTJ), a highly specialized structure at which the sarcolemma interdigitates with the tendon. This arrangement enlarges the interface between muscle and tendon and decreases the force placed on any one area of the terminating myofibre (1–3). A thin muscle basement membrane covers the interdigitations and provides an anchor for the collagen fibres of the tendon.

The molecular composition of the MTJ is not well understood. Two transmembrane receptors are known to be enriched at this site, namely the dystrophin-associated protein complex (DPC) (4) and {alpha}7ß1 integrin (5). The DPC binds to dystrophin and laminin, thereby forming a link between the cytoskeleton and the basement membrane (6,7). Mutations in the dystrophin gene cause Duchenne and Becker muscular dystrophies (8), while the genes encoding components of the DPC are mutated in some limb girdle muscular dystrophies (9). Owing to their lack of dystrophin (10), mdx mice serve as a model for the study of many aspects of muscle wasting. At mdx MTJs, there are elevated levels of dystrophin-related and associated proteins (11) and a reduction of membrane folds. The latter alteration is chiefly detected at the onset of the disease (12).

Integrins are heterodimeric transmembrane proteins of non-covalently linked {alpha} and ß subunits. They are important for diverse cellular functions such as migration, differentiation, tissue repair or programmed cell death (13,14). Among the ß integrins, the ß1 chain is ubiquitously expressed. Its carboxy-terminal splice variant, ß1D, is detected only in skeletal and cardiac muscles (15–17), where it is concentrated at the MTJs, costamers and intercalated discs (17,18). Upon heterologous expression, ß1D is targeted into focal contact sites (17). In ß1-deficient cells, ß1D causes multiple changes in the cell morphology and strengthens the link between the cytoskeleton and the extracellular matrix (19). Although ß1D has been implicated in myogenesis (20), mice depleted of the ß1D-encoding exon have no apparent skeletal muscle phenotype; yet they show a mild disturbance of cardiac function (21). The ß1A and ß1D variants are not functionally equivalent since mice expressing ß1D instead of ß1A in all tissues die during development. On the other hand, the targeted mutation of the ß1D exon seems to be compensated by ß1A (21). Clearly, ß1D integrin has an important, but perhaps not essential, function at the MTJ.

ß1D associates with several {alpha} subunits in the heart (17,22). In skeletal muscle, it binds exclusively to the laminin-binding {alpha}7 chain (17,18,23,24). There are four major splice variants of {alpha}7; two (X1 and X2) are found in the extracellular region of the protein and two (A and B) in the cytoplasmic tail (25,26). In addition, there are less abundant variants of the extracellular part (27,28). All four major variants are present during muscle development, but adult muscles express only X2 and not X1 (29,26). The {alpha}7A and {alpha}7B chains are located at the sarcolemma and also at the neuromuscular junction (30), where {alpha}7 has been implicated in the clustering of acetylcholine receptors (31,32).

The {alpha}7A and {alpha}7B chains are highly enriched at the MTJ where no other {alpha} subunits have been detected so far (5). The exact function of integrin {alpha}7 at this site is not known. We have previously shown that mice carrying a targeted deletion of their {alpha}7 gene (Itga7) develop muscular dystrophy primarily affecting muscle fibres in the vicinity of the MTJ (33). The first signs of abnormal MTJs are detectable around 3 weeks after birth. At adult junctions, the numbers of interdigitations are dramatically reduced and myofilaments appear retracted from the sarcolemma (33,34). Moreover, the muscle basement membranes are widened and do not contain the laminin {alpha}2 chain, which is instead accumulated in the matrix of the tendon (34). Recently, patients suffering from muscular dystrophy/congenital myopathy have been shown to carry mutations in their ITGA7 gene (35,36). All this evidence suggests that the {alpha}7 and ß1D subunits are important components of the MTJ, as they provide a critical link between the junctional cytoskeleton and the extracellular matrix.

In this study, we sought to identify molecular abnormalities preceding the onset of muscle wasting in {alpha}7-deficient mice in order to discern the function of integrin {alpha}7 in muscle. We show that ß1D is retained at mutant MTJs and specifically binds to the {alpha}5 chain, but not to {alpha}3, {alpha}6 or {alpha}v-containing integrins. Concomitantly fibronectin, the ligand of {alpha}5ß1, is amply deposited into the basement membranes of mutant muscles. Finally, we show that {alpha}5ß1D is down-regulated at the postnatal MTJs of normal or mdx muscles, but not of {alpha}7-deficient ones. Our data suggest that the {alpha}7 subunit plays an important role in the maturation of the MTJ, that it provides a link to laminin, and that it is involved in displacing {alpha}5ß1D and fibronectin from this site. The persistence of {alpha}5ß1D and fibronectin at {alpha}7-deficient MTJs is a molecular defect that may result in an inferior anchorage of the muscle cytoskeleton and may ultimately lead to muscle wasting.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Integrin {alpha}7-deficient muscles are morphologically abnormal
Integrin {alpha}7-deficient muscles display a dystrophic phenotype only in the vicinity of their MTJs, as we have described on the ultrastructural level (33,34) and is shown by light microscopy in Figure 1A and B. To determine whether mutant muscle fibres showed additional defects indiscernible by histological analysis, we established primary interosseus muscle cultures and examined the fibre morphology. Wild-type fibres (Fig. 1C) were multi-nucleated, cross-striated, and had smooth ends. Fibres derived from mdx mice looked very similar to wild-type ones (Fig. 1E). By contrast, {alpha}7-deficient fibres displayed an altered morphology, their ends being deformed and enlarged and containing vacuoles (Fig. 1D and F). Quantification of this phenotype revealed that, while control cultures contained mostly normal-looking fibres (108/111 and 96/101; Fig. 1C), about half of the {alpha}7 mutant fibres (98/194 and 107/200) had small vacuoles and dented ends (Fig. 1D). One third (61/194 and 73/200) displayed large vacuoles and ends with twice the normal diameter (Fig. 1F). The remaining {alpha}7-deficient fibres (35/194 and 20/200) looked similar to the controls. Notably, despite the absence of {alpha}7 integrin and the severe damage at their ends, mutant fibres were multi-nucleated and cross-striated. Therefore, {alpha}7-deficiency impairs neither myoblast fusion nor the organization of sarcomeres but rather affects the integrity of the ends of differentiated myofibres. These end regions are equivalent to the MTJs in skeletal muscles.



View larger version (159K):
[in this window]
[in a new window]
  		Figure 1. Deformed apical regions of integrin {alpha}7-deficient myofibres. Hematoxylin and eosin staining of cryosections prepared from adult wild-type (A) and {alpha}7-deficient (B) TA muscles. Note that in the region near the tendon (arrows), mutant muscles displayed varying muscle fibre diameters, centralized nuclei and lymphocyte infiltrations. Asterisks mark two necrotic fibers. (C–F) phase contrast images of cultured myofibres isolated from wild-type (C), mdx (E), and {alpha}7-deficient (D and F) interosseus muscles. All fibres (C–F) were multinucleated and showed cross-striation. The distal regions of wild-type and mdx fibres (C, E) were round-shaped, whereas the majority of {alpha}7-deficient fibres showed uneven ends (D), or such that were grossly enlarged and deformed (F). Bars, 50 µm (A, B); 100 µm (C–F).

 
ß1D integrin is retained at {alpha}7-deficient MTJs and associates with the {alpha}5 chain
The molecular mechanisms underpinning integrin {alpha}7-deficient muscular dystrophy are not understood. Since {alpha}7 binds specifically to ß1D integrin at normal MTJs, we expected ß1D to be lost from {alpha}7-depleted muscles as well. However, immunofluorescence staining revealed that the level of ß1D expression at mutant junctions was similar to those of control MTJs. ß1D's extrajunctional distribution was, however, greatly reduced (Fig. 2). These results implied that ß1D associates with a different {alpha} integrin in {alpha}7-deficient muscles. Strong candidates were {alpha}3 and {alpha}6 because of their high sequence similarity to {alpha}7 (26), {alpha}5 due to its regulated expression during myogenesis (37), and {alpha}v because {alpha}v is enriched at MTJ-like regions during development (38).



View larger version (88K):
[in this window]
[in a new window]
  		Figure 2. Localization of ß1D integrin in normal and {alpha}7-/- muscles. Serial cryosections of TA muscles were stained with peptide antibodies against {alpha}7B and ß1D. In the wild-types, {alpha}7B and ß1D immunoreactivities were concentrated at the MTJs (arrows indicate tendons). Weaker staining was detected at the circumference of the muscle fibres. In {alpha}7-deficient muscles (right), ß1D immunoreactivity was retained at the MTJs. Bar, 50 µm.

 
To test these candidates for interaction with ß1D, we generated specific antisera and performed coimmunoprecipitation and immunoblotting studies. As expected, {alpha}7A and {alpha}7B precipitated together with ß1D from wild-type but not from mutant muscle extracts (Fig. 3A). On immunoblots of whole muscle extracts or ß1D immunoprecipitates, we observed pairs of ß1D-reactive bands (Fig. 3A), of which the low molecular weight form corresponded to the incompletely glycosylated ß1D chain (18). Notably, this low molecular weight form did not dimerize {alpha}7 integrin. Moreover, the relative intensity of this band was increased in mutant muscles (Fig. 3A), suggesting that glycosylation of ß1D integrin is less efficient in {alpha}7-deficient muscles. As shown in Figure 3B, the {alpha}3 chain was not detected in adult wild-type or mutant muscle, whereas immunoprecipitation clearly showed the presence of {alpha}6 integrin in these muscles. However, {alpha}6 did not coimmunoprecipitate with ß1D (Fig. 3B). The {alpha}v chain was detected by immunoprecipitation in postnatal day 2 (P2) but not in adult muscles. Yet, {alpha}v also failed to be coimmunoprecipitated with ß1D, nor was there {alpha}v immunostaining at mutant MTJs (Fig. 3C, and data not shown).



View larger version (36K):
[in this window]
[in a new window]
  		Figure 3. Association of integrin ß1D with {alpha}7A and {alpha}7B in normal skeletal muscle. Protein extracts prepared from adult wild-type or {alpha}7-deficient TA muscles were immunoprecipitated with the specific antibodies indicated on the top of the panels, followed by immunoblotting as indicated on the right. (A) Under reducing conditions, the antibody directed against ß1D immunoprecipitated a high molecular weight and a low molecular weight-band. The latter band was more prominent in mutant muscles. In the wild-types, ß1D coimmunoprecipitated with {alpha}7B and {alpha}7A. Note that only the high molecular weight-form of ß1D was coprecipitated with the {alpha}7 antibodies. (B) {alpha}3 integrin was not detected in muscle precipitates, while the {alpha}6 chain was immunoprecipitated by anti-{alpha}6 but not the ß1D antibody. (C) the {alpha}v subunit was absent from adult muscle extracts, although it could be precipitated from P2 muscles. The antibody specificity was demonstrated by loading 20 µg of protein extracts (c) prepared from skeletal muscle (ß1D, {alpha}7A and {alpha}7B) or lung ({alpha}3, {alpha}6 and {alpha}v).

 
As opposed to the other {alpha} subunits, {alpha}5 coimmunoprecipitated with ß1D from mutant muscles, but not from wild-type ones (Fig. 4). On {alpha}5 immunoblots, the ß1D-precipitated signals were equally strong when RIPA extracts were derived from pooled TA or pooled soleus muscles. These results show that ß1D associates specifically with {alpha}5 integrin in {alpha}7-deficient muscles, but not with the {alpha}3, {alpha}6 or {alpha}v subunits.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 4. Association between ß1D and {alpha}5 integrins in {alpha}7-deficient skeletal muscles. In {alpha}7-deficient muscles, ß1D integrin was coimmunoprecipitated by the {alpha}5 antibody (upper panel) and vice versa (lower two panels). There was no apparent difference in the {alpha}5 signals using extracts from either soleus or TA muscles. The {alpha}5 chain precipitated from wild-type extracts was not associated with ß1D integrin. As control (c), 20 µg total protein from lung extracts was loaded.

 
The {alpha}5 and ß1D integrin subunits colocalize in {alpha}7-deficient muscles
To corroborate that {alpha}5ß1D integrin was indeed expressed at {alpha}7-deficient MTJs, we performed colocalization studies. As anticipated, there was no overlapping distribution of {alpha}5 and ß1D in adult wild-type muscles (Fig. 5). {alpha}5 was confined to connective tissue and the endothelium, while ß1D was strongly detected at the MTJs and at the sarcolemma of the soleus. By contrast, mutant MTJs displayed clear colocalization of both immunoreactivities (Fig. 5). Some {alpha}5 signals did not colocalize with ß1D, mostly in fibrotic areas of the muscle and probably resulting from {alpha}5ß1A receptors expressed in fibroblasts. In mutant soleus, the distribution of ß1D and {alpha}5 was discontinuous, a pattern that was not detected in the TA (Fig. 5). All immunofluorescence results were confirmed by confocal laser microscopy (data not shown). Taken together, our double-labelling and coimmunoprecipitation results indicate that {alpha}5ß1D integrin is expressed in place of {alpha}7ß1D at {alpha}7-deficient MTJs.



View larger version (89K):
[in this window]
[in a new window]
  		Figure 5. Colocalization of {alpha}5 and ß1D integrins in {alpha}7-deficient muscles. Cryosections of wild-type and mutant TA or soleus muscles were double-labelled with the {alpha}5 and ß1D antibodies. Extensive colocalization of both immunoreactivities was detected at {alpha}7-/- MTJs (arrows heads) and at the sarcolemma of some mutant soleus fibres. In wild-type muscles, {alpha}5 signals were confined to blood vessels and to the connective tissue of the endomysium. Asterisk indicates the tendon in the TA muscle. Bars, 50 µm.

 
{alpha}5ß1D integrin is down-regulated from wild-type postnatal muscles
To address the question why {alpha}5ß1D was retained at mutant MTJs, we analysed whether the {alpha}5 or {alpha}7 subunits heterodimerized with ß1D during normal muscle maturation. Double-immunofluorescence and coimmunoprecipitation studies revealed that P2 and P5 muscles coexpressed the {alpha}5ß1D and {alpha}7ß1D integrins both at the MTJs and at the extrajunctional sarcolemma (Fig. 6A and B). On {alpha}5 and {alpha}7 immunoblots, there were also bands in the ß1A-precipitated lanes (Fig. 6B). These bands probably resulted from {alpha}5ß1A being expressed in fibroblasts and endothelial cells, and from {alpha}7ß1A being present in smooth muscle cells.



View larger version (70K):
[in this window]
[in a new window]
  		Figure 6. Down-regulation of {alpha}5ß1D integrin from normal postnatal MTJs. (A) Cryosections of mouse hind limbs were double-labelled with {alpha}5 (left panels) and ß1D or {alpha}7B antibodies (right panels). At P2, wild-type muscles showed extensive colocalization of {alpha}5, {alpha}7B and ß1D at the fibre circumference and at the MTJs. At P10, {alpha}5 signals were abundantly detected at intramuscular blood vessels but not at MTJs. In mutant P10 muscles, {alpha}5 immunoreactivity persisted at the MTJ and colocalized with ß1D integrin. Bar, 100 µm. (B) extracts prepared from P2 and P10 wild-type or {alpha}7-/- muscles were immunoprecipitated with the antibodies indicated on the top of the panels, and analysed by immunoblotting using {alpha}5 and {alpha}7B antibodies. At P2, the {alpha}5 chain was coimmunoprecipitated with ß1A and ß1D from wild-type and mutant muscles. At P10, little {alpha}5 coprecipitated with ß1D from wild-type muscles, while mutants displayed strong {alpha}5-reactive bands in the ß1D precipitates. In control experiments, {alpha}7B was immunoprecipitated from wild-type muscles using ß1A and ß1D antibodies.

 
By P10, {alpha}5 immunostaining was found in endothelial cells and was barely detected at the MTJs or the sarcolemma. The {alpha}7 and ß1D chains colocalized with each other, most prominently at the MTJs (Fig. 6A). Coimmunoprecipitation experiments confirmed that most of {alpha}5 was associated with ß1A and little with ß1D, while ß1D dimerized with both {alpha}7A and {alpha}7B (Fig. 6B, and data not shown). On the other hand, mutant P10 MTJs showed strong colocalization of {alpha}5 and ß1D (Fig. 6A), and both subunits coimmunoprecipitated from mutant P10 extracts as they did in the adults (Fig. 6B, see also Fig. 4). These data indicate that, after an initial coexpression of {alpha}5ß1D and {alpha}7ß1D in early postnatal muscles, the {alpha}7ß1D receptor is preserved at the maturing MTJ whereas {alpha}5ß1D becomes down-regulated. This down-regulation of {alpha}5ß1D does not occur at integrin {alpha}7-deficient MTJs.

On northern blots, the levels of {alpha}5 messenger RNA declined dramatically by 3 weeks after birth, irrespective of whether RNA was prepared from normal or from integrin {alpha}7-depleted muscles (data not shown). Thus, a transcriptional up-regulation of {alpha}5 at mutant MTJs was unlikely. On immunoblots of P2 and P10 hind-limb muscle extracts, the {alpha}5 chain was detected with similar intensities, regardless of the animals' genotype (Fig. 7; upper panel). In adult tissues, {alpha}5 was no longer found in the wild-types but was still detected in knock-out muscles. The strong band detected in the mutant soleus reflected the presence of {alpha}5 in muscle fibres and fibrotic tissue (33). On non-reducing immunoblots, ß1D was found at similar levels in all extracts (data not shown). Under reducing conditions, the incompletely glycosylated form of ß1D was more abundant in the knock-outs than in the wild-types (Fig. 7, bottom panel), confirming the coimmunoprecipitation data reported above. Altogether, these results demonstrate that the loss of {alpha}7 integrin disrupts the post-transcriptional down-regulation of {alpha}5ß1D at the postnatal MTJ.



View larger version (45K):
[in this window]
[in a new window]
  		Figure 7. Protein expression of {alpha}5 and ß1D integrins in normal and {alpha}7-deficient muscles. Immunoblot analysis of protein extracts derived from postnatal hind limb (P2 and P10) and adult (soleus and TA) muscles. Twenty micrograms of total protein were loaded in each lane. Upper panel, under reducing conditions the {alpha}5 chain was detected as 19 kDa band, the intensities being similar in postnatal muscles of both genotypes. In the adult, {alpha}5 was only detected in the mutants. The asterisk indicates an unspecific band stained by the {alpha}5 antibody. Lower panel, ß1D immunoblot under non-reducing conditions showing that the amount of ß1D did not differ significantly between {alpha}7-deficient and wild-type extracts. The molecular weight is shown on the left in kDa.

 
Fibronectin is recruited to basement membranes of {alpha}7-deficient muscles
As a receptor exclusively for fibronectin, {alpha}5ß1D might, aberrantly, recruit this ligand to the basement membranes of {alpha}7-deficient muscles. To distinguish fibronectin signals at the muscle from those at the adjacent fibrotic tissue, we performed immunogold-histochemistry. As predicted from our immunofluorescence data, wild-type MTJs displayed few {alpha}5-reactive signals, but showed instead strong {alpha}7B and ß1D labelling at the plasma membrane (Fig. 8A and B, and data not shown). At {alpha}7-deficient MTJs, {alpha}5 and ß1D-labeled gold particles were detected at the plasma membrane and, in the case of ß1D, abundantly inside the cells (Fig. 8A and B). These intracellular gold particles possibly reflected the incompletely glycosylated ß1D form (see Figs 3, 4 and 7) that fails to be integrated into the plasma membrane.



View larger version (126K):
[in this window]
[in a new window]
  		Figure 8. Ultrastructural distribution of {alpha}5 and ß1D and fibronectin at the normal and {alpha}7-deficient MTJ. Ultrathin sections prepared from six month-old wild-type or {alpha}7-/- soleus muscles were labelled with the following antibodies: {alpha}5 (A), ß1D (B), and fibronectin (C and D). The regions of the MTJs are shown. The {alpha}5 and ß1D subunits were detected at the plasma membrane of mutant muscles (A, B), while only ß1D but not {alpha}5 was present in wild-type (A, B, insets). ß1D was also abundantly present in the cytoplasm of mutant muscles (B). Fibronectin staining was found in the connective tissue of the wild-types (C). By contrast, at {alpha}7-/- MTJs fibronectin signals were detected at the basement membrane of the muscle fibre (D). The inset in D shows a higher magnification image of the same section. Bars, 0.25 µm; in the insets, (0.30 µm).

 
On wild-type sections, fibronectin was detected in the connective tissue of the tendon but barely detected in the muscle basement membrane (Fig. 8C). By contrast, {alpha}7-deficient MTJs displayed ample fibronectin signals in the basement membrane of the muscle fibre (Fig. 8D), in addition to an overall increased labelling of the tendon. Quantification of gold particles revealed a 2 to 3-fold amount of label in {alpha}7-deficient mice as in controls both in the basement membrane of the MTJ (13:5) and the tendon (65:31), while no differences were noted in the basement membrane (2:3) or the endomysium (17:18) at the lateral side of the muscle fibres. Thus, concomitant with the persistence of {alpha}5ß1D at {alpha}7-deficient MTJs, there is a dramatically increased deposition of fibronectin at the muscle basement membrane.

{alpha}5ß1D integrin does not persist at dystrophin-deficient MTJs
Finally, we determined whether {alpha}5ß1D also persisted at dystrophin-deficient MTJs. The dystrophic changes in adult mdx muscles were confirmed by conventional histology and by strong anti-fibronectin labelling indicating fibrosis (data not shown). On cryosections (Fig. 9A), {alpha}5 immunoreactivity was amply detected in the connective tissue of the tendon and in the endomysium near the tendon. However, there was little or no colocalization of {alpha}5 and ß1D (Fig. 9A). Moreover, using extracts prepared from mdx TA muscles, ß1D did not coimmunoprecipitate with the {alpha}5 chain but was associated with {alpha}7 integrin, as it is in the wild-types (Fig. 9B). Therefore, the persistence of {alpha}5ß1D integrin is not a phenotype of dystrophin-deficient muscles, but is instead a molecular change resulting from integrin {alpha}7 deficiency.



View larger version (54K):
[in this window]
[in a new window]
  		Figure 9. Absence of {alpha}5ß1D from dystrophin-deficient MTJs. (A) Distribution of the {alpha}5 and ß1D chains on cryosections of mdx TA muscles. {alpha}5 immunoreactivity was observed in the fibrotic areas, but there was little overlapping staining with ß1D (Merge). Bar, 50 µm. (B) protein extracts prepared from mdx TA muscles were immunoprecipitated using the ß1D antibody and immunoblotted under reducing conditions. Only {alpha}7B, but not {alpha}5 integrin, was detected in the ß1D precipitates. The molecular weight in kDa is shown on the left.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
In this paper, we provide evidence that the maturation of the postnatal MTJ critically depends on {alpha}7ß1D integrin. At wild-type MTJs, {alpha}7ß1D displaces the coexpressed {alpha}5ß1D integrin by 2 weeks after birth. This switch does not occur in {alpha}7-deficient muscles, but instead {alpha}5ß1D persists into adulthood, and fibronectin accumulates in place of laminin at the junctional basement membrane of the muscle fibre. We propose that this {alpha}5ß1D/fibronectin link is inferior to the {alpha}7ß1D/laminin link at the normal MTJ, and that this gain-of-function phenotype is a molecular abnormality preceding the onset of muscle wasting in {alpha}7-deficient mice.

Intact muscle fibres isolated from {alpha}7-deficient mice display grossly deformed ends; at the same time they are cross-striated and properly attached to the culture dish. These results confirm the observation that myoblast fusion and myofibrillogenesis proceed independently of laminin-binding integrin chains (33,39), and support our previous conclusion that {alpha}7ß1 is critical for the integrity of the junctional but not the lateral regions of the muscle (33,34). At the lateral sarcolemma, {alpha}-dystroglycan of the DPC is known to link laminin with dystrophin and other cytoskeletal proteins (4,40). In fact, somewhat overlapping functions of {alpha}7ß1 integrin and {alpha}-dystroglycan are indicated by the finding that the {alpha}7 chain is up-regulated in dystrophin-deficient muscles (41,42) and in those with little or no dystroglycan (43). Moreover, transgenic overexpression of {alpha}7ß1 integrin relieves many of the dystrophic symptoms in mice which have a combined loss of dystrophin- and utrophin-associated proteins (44).

Despite the presence of the DPC, {alpha}7-deficient MTJs seem to lack a functional laminin receptor, as suggested by the aberrant deposition of laminin into the matrix of the tendon (34). Moreover, the retraction of myofilaments (34) suggests an impairment of the cytoplasmic part of mutant MTJs. Our results, showing that {alpha}5ß1D integrin persists at {alpha}7-/- junctions, point to a selective pressure to retain the ß1D chain at this site. In addition to being enriched at regions of force transmission in muscle (15,17), ß1D is able to enhance the adhesive qualities of transfected non-muscle cells by strengthening the interaction with cytoskeletal and extracellular ligands (19). However, mice depleted of their ß1D-encoding exon show no apparent skeletal muscle defect (21), whereas {alpha}7-deficiency leads to muscle damage precisely at the MTJs. Thus, the combined qualities of {alpha}7 and ß1D seem to be required at the junctional sarcolemma, not only to provide a laminin receptor but also to displace the connection between {alpha}5ß1D and fibronectin from the postnatally maturing MTJ.

The switching of extracellular matrix receptors is likely to effect the formation and/or maintenance of the MTJ. By means of mechanotransduction (45), the {alpha}5ß1D and {alpha}7ß1D integrins may transmit specific signals to the inside of the muscle fibre, thereby activating or deactivating signalling pathways and modulating the assembly of the junctional cytoskeleton. In cell culture, the ligands for both receptors are known to have antagonistic effects on muscle, with laminin promoting myogenesis and fibronectin inhibiting differentiation (46–49). Given our results that the physiological integrin switch occurs toward the end of secondary myogenesis in mice and that myogenesis is unimpaired in the mutants, the establishment of the {alpha}7ß1D/laminin connection appears to be particularly important for the maturing MTJ. It may be that only this link ensures the correct anchoring of junctional myofilaments in the differentiated and contracting muscle. Evidence for this conclusion is provided by our previous observation that the sarcomer had retracted from mutant MTJs (33), suggesting a reduced force transmission from the muscle to the tendon when {alpha}7ß1 is absent. Further physiological studies, however, will be needed to support our hypothesis.

The precise mechanism by which {alpha}5ß1D integrin is down-regulated from the MTJ is unknown. As we detected declining levels of {alpha}5 messenger RNA in postnatal wild-type and mutant muscles, but elevated amounts of {alpha}5 protein at adult {alpha}7 mutant MTJs, the expression of the {alpha}5 chain may be regulated in a complex way. Recent studies revealed that cross-talk between integrins can mediate their functional activation or deactivation. For example, overexpression of {alpha}7 integrin in CHO cells interferes with their adherence to fibronectin and the assembly of a fibronectin matrix, apparently by functionally inactivating endogenous the {alpha}5ß1 integrin (50). Similarly, {alpha}3ß1 is able to inhibit the function of {alpha}2ß1 in mammary carcinoma cells and of other integrins in keratinocytes (51,52). Conversely, {alpha}5 integrin-mediated signalling has been shown to positively influence the binding of {alpha}4-containing integrins to fibronectin (53). According to this evidence, the absent down-regulation of {alpha}5ß1D at {alpha}7-deficient MTJs may be due to the lack of a negative cooperative effect that is usually exerted by the {alpha}7 chain. An important question to address next is whether {alpha}5ß1D and {alpha}7ß1D integrins are present in an active conformation when they are coexpressed in postnatal muscles.

Additional clues as to whether cross-talk between {alpha}5ß1D and {alpha}7ß1D mediates the down-regulation of {alpha}5ß1D may come from proteins that bind to the {alpha}5 and {alpha}7 subunits. Nischarin, a protein that can interfere with the activity of the Rac GTPase, associates with the cytoplasmic part of {alpha}5 (54). Two other proteins, TIP-2/GIPC and DRAL/FHL2, bind to {alpha}5 or {alpha}7, respectively (55–57). These latter proteins may act as molecular adapters for other proteins since TIP-2/GIPC contains a PDZ domain (a motif first described in PSD-95, Dlg and ZO-1) and because DRAL/FHL2 is a member of the LIM domain protein family. For example, signalling complexes associated with {alpha}7 integrin contain p130CAS/Crk, focal adhesion kinase and paxillin (58). Future studies will reveal which of these and other protein interactions take place in situ, and whether signalling-mediated receptor cross-talk indeed governs the integrin switch we have reported here.

Unlike integrin {alpha}5-deficiency, which also leads to muscle wasting in mice (59), murine {alpha}7-deficiency serves as a model for the corresponding human muscle wasting disorder (35,36). It is therefore particularly important to decipher the molecular mechanisms underpinning this disease. The loss of the main laminin receptor from {alpha}7-deficient muscles is likely to hamper the mechanical stability of the MTJ and to lead to muscle wasting. Yet, a more complex pathogenesis may result from the defective integrin switch we have reported here. The presence of {alpha}5ß1D may lead to an abnormally increased deposition of fibronectin at mutant MTJ. In turn, fibronectin may trigger signals inside the muscle fibre that form a cytoskeleton/matrix link, which may be insufficient for contracting muscle. In this case, impaired bi-directional signalling at the MTJ would lie at the core of {alpha}7-deficiency. In conclusion, we propose that the defective integrin switch and the assembly of a fibronectin-based matrix is a molecular abnormality that precedes the onset of muscle wasting in {alpha}7-deficient mice and possibly patients.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Isolation of muscle fibres and morphological assessment
Muscle fibres were isolated from 15-week-old wild-type or {alpha}7-deficient or mdx mice by dissecting the interosseus muscles of the foot and dissociating them enzymatically for 2 h at 37°C in 1.5 mg/ml collagenase (Serva) and 2 mg/ml neutral protease type IX (Sigma). For each type of animals, two independent preparations generated six cultures. Dissociated fibres were maintained on plastic Petri dishes at 37°C and 5% CO2 in Ham's F-12 medium supplemented with 10% horse serum. After 2 days, knock-out and control fibres were classified into three groups according to their morphology under phase contrast microscopy: (1) normal fibres with smooth extremities; (2) mildly damaged fibres showing irregular tips and vacuoles; and (3) fibres displaying grossly irregular tips and large vacuoles (tip diameter with 2-fold the size of the fibre diameter).

Antibody production
The following peptides were synthesised, coupled to maleimide-activated keyhole limpet hemocyanin according to the supplier's protocol (Pierce), and injected into rabbits: {alpha}3A, NH2-GCRTRALYEAKRQKAEMKSQPSETERLTDDY; {alpha}5, NH2-GCKRSLPYGTAMEKAQLKPPATSDA; {alpha}6A, NH2-GCKKDHYDATYHKAEIHAQPSDKERLTSDA; {alpha}7A, NH2-GCGWDSSSGRSTPRPPSPSTTQ; {alpha}v, NH2-GCKRVRPPQEEQERE; ß1A, NH2-GCGENPIYKSAVTTVVNPKYEGK; ß1D, NH2-IYKSPINNFKNPNYGRKAGLCG. All antibodies were affinity-purified on columns using immobilized BSA-coupled peptides and tested for specificity on normal tissues by immunoblotting and indirect immunofluorescence. The anti-integrin {alpha}7B antibody has been characterized (42). The anti-integrin {alpha}5 monoclonal antibody was purchased from BD/Pharmingen, the anti-fibronectin polyclonal antibody was obtained from Gibco/BRL. The anti-integrin {alpha}6 monoclonal antibody (GOH3) was a gift of Dr A. Sonnenberg (The Netherlands Cancer Institute). Horseradish peroxidase-coupled secondary antibodies were purchased from Biorad, and fluorophore-conjugated secondary antibodies from Jackson Laboratories.

Histological analysis
For immunofluorescence analysis, freshly prepared muscle tissue was frozen in liquid nitrogen-cooled isopentane, sectioned at 8 µm in the transverse plane, and collected onto TESPA-coated glass slides. The sections were fixed in 1% paraformaldehyde for 10 min at room temperature, followed by methanol fixation for 8 min at -20°C. Thereafter, all incubations were performed at room temperature. The sections were rehydrated and blocked in PBS supplemented with 0.1% (v/v) Tween-20 and 5% normal goat serum for 1 h. This was followed by incubation with the primary antibodies (5 µg/ml, diluted in PBS plus 2% normal goat serum) for 1 h in a humidified chamber. After three washes with PBS, the sections were incubated with Cy2 or Cy3-conjugated secondary antibodies for 30 min, again washed with PBS, and mounted in gelvatol containing 1.5% (w/v) 1,4-diazabicyclo (2.2.2) octane (DABCO, Sigma). For microscopic analysis, we used an Axiophot microscope (Zeiss) and digital recording of the images. Hematoxilin and eosin staining was performed according to standard procedures.

Immunoprecipitation and immunoblotting analysis
Extracts of pooled mouse muscles were prepared by homogenizing freshly dissected tissue in a modified RIPA buffer [50 mM Tris, pH 7.4, 150 mM NaCl, 2.5 mM MgCl2, 1% (v/v) Nonidet-P40, 0.05% (v/v) Triton X-100, 0.5% (w/v) Na-deoxycholate, 2.5 mM EDTA, 5 mM N-ethylmaleimide, 2 mM PMSF]. Immunoprecipitation analysis was performed as described (60). Briefly, 800 µg of pre-cleared RIPA-soluble protein were incubated with 5 µg of affinity-purified antibodies for 2 h at 4°C on a rotating platform. Immune complexes were captured by adding 50 µl of equilibrated Protein A Sepharose CL-4B (Amersham/Pharmacia) and an overnight incubation at 4°C on a rotating platform. After extensive washing in RIPA buffer and once in PBS, immunoprecipitated proteins were eluted from Protein A Sepharose by boiling in SDS/urea buffer [4 M urea, 3.8% SDS, 20% (v/v) glycerol, 75 mM Tris, pH 6.8, 5% (v/v) 2-mercaptoethanol]. Equal volumes of eluted proteins were separated by gradient (5–20%) SDS–PAGE, transferred onto Immobilon-P membranes (Millipore), and detected with antibodies using standard methods. The antibodies against the respective integrin chains were used at 1 µg/ml, and the horseradish peroxidase-conjugated anti-rabbit IgG at 1:10 000 dilution. Protein bands were visualized with ECL (Amersham/ Pharmacia).

Immunogold-histochemistry
Soleus and tibialis anterior (TA) muscles were dissected from 6-week- and 6-month-old mice and immediately fixed for 30 min at 4°C in buffered 4% paraformaldehyde and 0.5% glutaraldehyde. After immersion for 45 min at 4°C in 10 mM ammonium chloride and dehydration in a graded series of ethanol, the specimens were embedded in the acrylic resin LR-Gold (London Resin Company). The resin was hardened at -25°C with the addition of 0.8% benzil and the light of a halogen lamp. Ultrathin sections were cut with a Reichert's ultramicrotome and collected on formvar-coated nickel grids. All ensuing incubations were performed at room temperature (34). Section-containing grids were blocked for 15 min in PBS containing 1% (w/v) BSA, followed by incubating the primary antibodies (used at 5 µg/ml) for 1 h. After washing in PBS, gold-labelled secondary antibodies (Medac) were applied for 20 min, and again washed in PBS. Finally, the grids were rinsed in water and contrasted. Controls for unspecific binding were performed by incubation with un-coated colloidal gold probes, and by using normal rabbit and rat antisera instead of the primary antibodies. For quantification of gold particles, tissue samples of three homozygous mutant and three control mice were taken and gold particles in a total of 30 randomly selected fields of 4 µm2 counted for each genotype.


   ACKNOWLEDGEMENTS
 
The authors would like to thank H. Alberty-Hornberger, E. Burghart, and I. Jannetti for their excellent technical assistance, and Dr A. Sonnenberg for his gift of the GOH3 antibody. We are particularly grateful to Dr D. Blake and K. Dombkowski for their discussions and critical readings of the manuscript. R.N. was supported by a Max-Planck fellowship and U.M. was supported by the Deutsche Forschungsgemeinschaft (Ma 1707/1-2) and by the Wellcome Trust (no. 060549).


   FOOTNOTES
 
* To whom correspondence should be addressed at: Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, University of Manchester, 3.239 Stopford Building, Oxford Road, Manchester M13 9PT, UK. Tel: +44 1612755246; Fax: +44 1612753915; Email: ulrike.mayer{at}man.ac.uk Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Tidball, J.G. and Daniel, T.L. (1986) Myotendinous junctions of tonic muscle cells: structure and loading. Cell Tissue Res., 245, 315–322.[ISI][Medline]

  2. Trotter, J.A. and Baca, J.M. (1987) The muscle-tendon junctions of fast and slow fibres in the garter snake: ultrastructural and stereological analysis and comparison with other species. J. Muscle Res. Cell Motil., 8, 517–526.[CrossRef][ISI][Medline]

  3. Trotter, J.A. (1993) Functional morphology of force transmission in skeletal muscle. A brief review. Acta Anat., 146, 205–522.[ISI][Medline]

  4. Ibraghimov-Beskrovnaya, O., Ervasti, J.M., Leveille, C.J., Slaughter, C.A., Sernett, S.W. and Campbell, K.P. (1992) Primary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix. Nature, 355, 696–702.[CrossRef][Medline]

  5. Bao, Z.Z., Lakonishok, M., Kaufman, S. and Horwitz, A.F. (1993) {alpha}7ß1 integrin is a component of the myotendinous junction on skeletal muscle. J. Cell Sci., 106, 579–589.[Abstract]

  6. Ervasti, J.M. and Campbell, K.P. (1993) A role for the dystrophin–glycoprotein complex as a transmembrane linker between laminin and actin. J. Cell Biol., 122, 809–823.[Abstract/Free Full Text]

  7. Cohn, R.D. and Campbell, K.P. (2000) Molecular basis of muscular dystrophies. Muscle Nerve, 23, 1456–1471.[CrossRef][ISI][Medline]

  8. Blake, D.J., Weir, A., Newey, S.E. and Davies, K.E. (2002) Function and genetics of dystrophin and dystrophin-related proteins in muscle. Physiol. Rev., 82, 291–329.[Abstract/Free Full Text]

  9. Bushby, K.M. (1999) The limb-girdle muscular dystrophies-multiple genes, multiple mechanisms. Hum. Mol. Genet., 8, 1875–1882.[Abstract/Free Full Text]

  10. Sicinski, P., Geng, Y., Ryder-Cook, A.S., Barnard, E.A., Darlison, M.G. and Barnard, P.J. (1989) The molecular basis of muscular dystrophy in the mdx mouse: a point mutation. Science, 244, 1578–1580.[Abstract/Free Full Text]

  11. Law, D.J. and Tidball, J.G. (1993) Dystrophin deficiency is associated with myotendinous junction defects in prenecrotic and fully regenerated skeletal muscle. Am. J. Pathol., 142, 1513–1523.[Abstract]

  12. Law, D.J., Caputo, A. and Tidball, J.G. (1995) Site and mechanics of failure in normal and dystrophin-deficient skeletal muscle. Muscle Nerve, 18, 216–223.[CrossRef][ISI][Medline]

  13. Hynes, R.O. (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell, 69, 11–25.[CrossRef][ISI][Medline]

  14. van der Flier, A. and Sonnenberg, A. (2001) Function and interactions of integrins. Cell Tissue Res., 305, 285–298.[CrossRef][ISI][Medline]

  15. van der Flier, A., Kuikman, I., Baudoin, C., van der Neut, R. and Sonnenberg, A. (1995) A novel ß1 integrin isoform produced by alternative splicing: unique expression in cardiac and skeletal muscle. FEBS Lett., 369, 340–344.[CrossRef][ISI][Medline]

  16. Zhidkova, N.I., Belkin, A.M. and Mayne, R. (1995) Novel isoform of ß1 integrin expressed in skeletal and cardiac muscle. Biochem. Biophys. Res. Commun., 214, 279–285.[CrossRef][ISI][Medline]

  17. Belkin, A.M., Zhidkova, N.I., Balzac, F., Altruda, F., Tomatis, D., Maier, A., Tarone, G., Koteliansky, V.E. and Burridge, K. (1996) ß1D integrin displaces the ß1A isoform in striated muscles: localization at junctional structures and signaling potential in nonmuscle cells. J. Cell Biol., 132, 211–226.[Abstract/Free Full Text]

  18. van der Flier, A., Gaspar, A.C., Thorsteinsdottir, S., Baudoin, C., Groeneveld, E., Mummery, C.L. and Sonnenberg, A. (1997) Spatial and temporal expression of the ß1D integrin during mouse development. Dev. Dyn., 210, 472–486.[CrossRef][ISI][Medline]

  19. Belkin, A.M., Retta, S.F., Pletjushkina, O.Y., Balzac, F., Silengo, L., Fässler, R., Koteliansky, V.E., Burridge, K. and Tarone, G. (1997) Muscle ß1D integrin reinforces the cytoskeleton-matrix link: modulation of integrin adhesive function by alternative splicing. J. Cell Biol., 139, 1583–1595.[Abstract/Free Full Text]

  20. Belkin, A.M. and Retta, S.F. (1998) ß1D integrin inhibits cell cycle progression in normal myoblasts and fibroblasts. J. Biol. Chem., 273, 15234–15240.[Abstract/Free Full Text]

  21. Baudoin, C., Goumans, M.J., Mummery, C. and Sonnenberg, A. (1998) Knockout and knockin of the ß1 exon D define distinct roles for integrin splice variants in heart function and embryonic development. Genes Dev., 12, 1202–1216.[Abstract/Free Full Text]

  22. Brancaccio, M., Cabodi, S., Belkin, A.M., Collo, G., Koteliansky, V.E., Tomatis, D., Altruda, F., Silengo, L. and Tarone, G. (1998) Differential onset of expression of {alpha}7 and ß1D integrins during mouse heart and skeletal muscle development. Cell Adhes. Commun., 5, 193–205.[ISI][Medline]

  23. Schöber, S., Mielenz, D., Echtermeyer, F., Hapke, S., Pöschl, E., von der Mark, H., Moch, H. and von der Mark, K. (2000) The role of extracellular and cytoplasmic splice domains of {alpha}7-integrin in cell adhesion and migration on laminins. Exp. Cell Res., 255, 303–313.[CrossRef][ISI][Medline]

  24. Burkin, D.J. and Kaufman, S.J. (1999) The {alpha}7ß1 integrin in muscle development and disease. Cell Tissue Res., 296, 183–190.[CrossRef][ISI][Medline]

  25. Collo, G., Starr, L. and Quaranta, V. (1993) A new isoform of the laminin receptor integrin {alpha}7ß1 is developmentally regulated in skeletal muscle. J. Biol. Chem., 268, 19019–19024.[Abstract/Free Full Text]

  26. Ziober, B.L., Vu, M.P., Waleh, N., Crawford, J., Lin, C.S. and Kramer, R.H. (1993) Alternative extracellular and cytoplasmic domains of the integrin {alpha}7 subunit are differentially expressed during development. J. Biol. Chem., 268, 26773–26783.[Abstract/Free Full Text]

  27. Leung, E., Lim, S.P., Berg, R., Yang, Y., Ni, J., Wang, S.X. and Krissansen, G.W. (1998) A novel extracellular domain variant of the human integrin {alpha}7 subunit generated by alternative intron splicing. Biochem. Biophys. Res. Commun., 243, 317–325.[CrossRef][ISI][Medline]

  28. Vignier, N., Moghadaszadeh, B., Gary, F., Beckmann, J., Mayer, U. and Guicheney, P. (1999) Structure, genetic localization, and identification of the cardiac and skeletal muscle transcripts of the human integrin {alpha}7 gene (ITGA7). Biochem. Biophys. Res. Commun., 260, 357–364.[CrossRef][ISI][Medline]

  29. Song, W.K., Wang, W., Sato, H., Bielser, D.A. and Kaufman, S.J. (1993) Expression of {alpha}7 integrin cytoplasmic domains during skeletal muscle development: alternate forms, conformational change, and homologies with serine/threonine kinases and tyrosine phosphatases. J. Cell Sci., 106, 1139–1152.[Abstract]

  30. Martin, P.T., Kaufman, S.J., Kramer, R.H. and Sanes, J.R. (1996) Synaptic integrins in developing, adult, and mutant muscle: selective association of {alpha}1, {alpha}7A, and {alpha}7B integrins with the neuromuscular junction. Dev. Biol., 174, 125–139.[CrossRef][ISI][Medline]

  31. Burkin, D.J., Gu, M., Hodges, B.L., Campanelli, J.T. and Kaufman, S.J. (1998) A functional role for specific spliced variants of the {alpha}7ß1 integrin in acetylcholine receptor clustering. J. Cell Biol., 143, 1067–1075.[Abstract/Free Full Text]

  32. Burkin, D.J., Kim, J.E., Gu, M. and Kaufman, S.J. (2000) Laminin and {alpha}7ß1 integrin regulate agrin-induced clustering of acetylcholine receptors. J. Cell Sci., 113, 2877–2886.[Abstract]

  33. Mayer, U., Saher, G., Fässler, R., Bornemann, A., Echtermeyer, F., von der Mark, H., Miosge, N., Pöschl, E. and von der Mark, K. (1997) Absence of integrin {alpha}7 causes a novel form of muscular dystrophy. Nat. Genet., 17, 318–323.[ISI][Medline]

  34. Miosge, N., Klenczar, C., Herken, R., Willem, M. and Mayer, U. (1999) Organization of the myotendinous junction is dependent on the presence of {alpha}7ß1 integrin. Lab. Invest., 79, 1591–1599.[ISI][Medline]

  35. Hayashi, Y.K., Chou, F.L., Engvall, E., Ogawa, M., Matsuda, C., Hirabayashi, S., Yokochi, K., Ziober, B.L., Kramer, R.H., Kaufman, S.J. et al. (1998) Mutations in the integrin {alpha}7 gene cause congenital myopathy. Nat. Genet., 19, 94–97.[CrossRef][ISI][Medline]

  36. Pegoraro, E., Cepollaro, F., Prandini, P., Marin, A., Fanin, M., Trevisan, C.P., El-Messlemani, A.H., Tarone, G., Engvall, E., Hoffman, E.P. and Angelini, C. (2002) Integrin {alpha}7ß1 in muscular dystrophy/myopathy of unknown etiology. Am. J. Pathol., 160, 2135–2143.[Abstract/Free Full Text]

  37. Boettiger, D., Enomoto-Iwamoto, M., Yoon, H.Y., Hofer, U., Menko, A.S. and Chiquet-Ehrismann, R. (1995) Regulation of integrin {alpha}5ß1 affinity during myogenic differentiation. Dev. Biol., 169, 261–272.[CrossRef][ISI][Medline]

  38. Hirsch, E., Gullberg, D., Balzac, F., Altruda, F., Silengo, L. and Tarone, G. (1994) {alpha}v integrin subunit is predominantly located in nervous tissue and skeletal muscle during mouse development. Dev. Dyn., 201, 108–120.[ISI][Medline]

  39. Hirsch, E., Lohikangas, L., Gullberg, D., Johansson, S. and Fässler, R. (1998) Mouse myoblasts can fuse and form a normal sarcomere in the absence of ß1 integrin expression. J. Cell Sci., 111, 2397–2409.[Abstract]

  40. Dickson, G., Azad, A., Morris, G.E., Simon, H., Noursadeghi, M. and Walsh, F.S. (1992) Co-localization and molecular association of dystrophin with laminin at the surface of mouse and human myotubes. J. Cell Sci., 103, 1223–1233.[Abstract/Free Full Text]

  41. Hodges, B.L., Hayashi, Y.K., Nonaka, I., Wang, W., Arahata, K. and Kaufman, S.J. (1997) Altered expression of the {alpha}7ß1 integrin in human and murine muscular dystrophies. J. Cell Sci., 110, 2873–2881.[Abstract]

  42. Cohn, R.D., Mayer, U., Saher, G., Herrmann, R., van der Flier, A., Sonnenberg, A., Sorokin, L. and Voit, T. (1999) Secondary reduction of {alpha}7B integrin in laminin {alpha}2 deficient congenital muscular dystrophy supports an additional transmembrane link in skeletal muscle. J. Neurol. Sci., 163, 140–152.[CrossRef][ISI][Medline]

  43. Cote, P.D., Moukhles, H. and Carbonetto, S. (2002) Dystroglycan is not required for localization of dystrophin, syntrophin, and neuronal nitric-oxide synthase at the sarcolemma but regulates integrin {alpha}7B expression and caveolin-3 distribution. J. Biol. Chem., 277, 4672–4679.[Abstract/Free Full Text]

  44. Burkin, D.J., Wallace, G.Q., Nicol, K.J., Kaufman, D.J. and Kaufman, S.J. (2001) Enhanced expression of the {alpha}7ß1 integrin reduces muscular dystrophy and restores viability in dystrophic mice. J. Cell Biol., 152, 1207–1218.[Abstract/Free Full Text]

  45. Alenghat, F.J. and Ingber, D.E. (2002) Mechanotransduction: all signals point to cytoskeleton, matrix, and integrins. Science STKE, PE6.

  46. Kühl, U., Öcalan, M., Timpl, R. and von der Mark, K. (1986) Role of laminin and fibronectin in selecting myogenic versus fibrogenic cells from skeletal muscle cells in vitro. Dev. Biol., 117, 628–635.[CrossRef][ISI][Medline]

  47. Öcalan, M., Goodman, S.L., Kühl, U., Hauschka, S.D. and von der Mark, K. (1988) Laminin alters cell shape and stimulates motility and proliferation of murine skeletal myoblasts. Dev. Biol., 125, 158–167.[CrossRef][ISI]