Journal of the Neurological Sciences 163 (1999) 140–152
Secondary reduction of a7B integrin in laminin a2 deficient congenital muscular dystrophy supports an additional transmembrane link in skeletal muscle Ronald D. Cohn a , Ulrike Mayer b , Gesine Saher b , Ralf Herrmann a , Arjan van der Flier c , c d a, Arnoud Sonnenberg , Lydia Sorokin , Thomas Voit * a
Departments of Pediatrics and Pediatric Neurology, University of Essen, Hufelandstr. 55, 45122 Essen, Germany b Max-Planck Institute for Biochemistry, Martinsried, Germany c The Netherland Cancer Institute, Division of Cell Biology, Amsterdam, The Netherlands d Institute of Experimental Medicine, Connective Tissue Research, University of Erlangen-Nuernberg, Erlangen, Germany Received 11 May 1998; received in revised form 12 October 1998; accepted 4 January 1999
Abstract The integrins are a large family of heterodimeric transmembrane cellular receptors which mediate the association between the extracellular matrix (ECM) and cytoskeletal proteins. The a7b1 integrin is a major laminin binding integrin in skeletal and cardiac muscle and is thought to be involved in myogenic differentiation and migration processes. The main binding partners of the a7 integrin are laminin-1 (a1-b1-g1), laminin-2 (a2-b1-g1) and laminin-4 (a2-b2-g1). Targeted deletion of the gene for the a7 integrin subunit (ITGA7) in mice leads to a novel form of muscular dystrophy. In the present study we have investigated the expression of two alternative splice variants, the a7B and b1D integrin subunits, in normal human skeletal muscle, as well as in various forms of muscular dystrophy. In normal human skeletal muscle the expression of the a7 integrin subunit appeared to be developmentally regulated: it was first detected at 2 years of age. In contrast, the b1D integrin could be detected in immature and mature muscle in the sarcolemma of normal fetal skeletal muscle at 18 weeks gestation. The expression of a7B integrin was significantly reduced at the sarcolemma in six patients with laminin a2 chain deficient congenital muscular dystrophy (CMD) (age .2 years). However, this reduction was not correlated with the amount of laminin a2 chain expressed. In contrast, the expression of the laminin a2 chain was not altered in the skeletal muscle of the a7 knock-out mice. These data argue in favor that there is not a tight correlation between the expression of the a7 integrin subunit and that of the laminin a2 chain in either human or murine dystrophic muscle. Interestingly, in dystrophinopathies (Duchenne and Becker muscular dystrophy; DMD/ BMD) expression of a7B was upregulated irrespective of the level of dystrophin expression as shown by a strong sarcolemmal staining pattern even in young boys (age ,2 years). The expression of the b1D integrin subunit was not altered in any of our patients with different types of muscular dystrophy. In contrast, sarcolemmal expression of b1D integrin was significantly reduced in the a7 integrin knock-out mice, whereas the expression of the components of the DGC was not altered. The secondary loss of a7B in laminin a2 chain deficiency defines a biochemical change in the composition of the plasma membrane resulting from a primary protein deficiency in the basal lamina. These findings, in addition to the occurrence of a muscular dystrophy in a7 deficient mice, implies that the a7B integrin is an important laminin receptor within the plasma membrane which plays a significant role in skeletal muscle function and stability. 1999 Elsevier Science B.V. All rights reserved. Keywords: Extracellular matrix; a7 Integrin; Muscle function; Muscular dystrophy
*Corresponding author. Tel.: 149-201-723-5983; fax: 149-201-723-5983. E-mail address:
[email protected]. (T. Voit) 0022-510X / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0022-510X( 99 )00012-X
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
1. Introduction The integrins are a family of heterodimeric transmembrane receptors that mediate cell–extracellular matrix and cell–cell interactions [1] and have been implicated in cell interactions during differentiation, cell migration [2] and morphogenesis [3]. Each integrin is composed of a noncovalently associated paired a and b polypeptide chain with a long extracellular domain, a hydrophobic transmembrane domain and a relatively short cytoplasmic domain. In skeletal muscle, two major types of linkage between the extracellular matrix (ECM) and the cytoskeleton have been identified. One type of linkage involves the b1 integrin family of molecules; they bind proteins associated with the cytoskeleton, e.g. a-actinin and talin, by their cytoplasmic domains [4] and therefore serve as a link between the cytoskeleton and the extracellular matrix at the Z-bands, myotendinous junctions and neuromuscular junctions [5,6]. To date, nine different a-chains (a1, 3–7, 9, and v) are known to form dimers with the b1 integrin subunit in the developing skeletal muscle (for a review, see Ref. [7]). Among these, the a7 integrin subunit has been shown to be localized at the myotendinous junctions [8], the neuromuscular junctions, as well as on the sarcolemma [9] of adult muscle fibers in vivo. The a7 integrin subunit, associated with b1, is a major laminin binding integrin [10] in skeletal and cardiac muscle [11,12] and the a7 subunit is involved in differentiation and migration processes during myogenesis [11,13,14]. The expression of splice variants of the extracellular and cytoplasmic domains of a7 (A,B,C) is differentially regulated during the development of skeletal muscle [9,15–17]. It has been shown that the overall level at which a7 is expressed increases during myoblast fusion and differentiation [18] and a7A is detected at high levels in differentiating myotubes whereas the a7B isoform is present in proliferating myoblasts and fusing myotubes [17,19]. The a7A isoform is restricted to skeletal muscle, while the a7B isoform is exclusively expressed in skeletal muscle, cardiac muscle, the vasculature and the nervous system [20–22]. Interestingly, the expression of a7B in the myocardium of mice is strongly induced postnatally [21]. Velling et al. [19] showed that laminin a2 is present in and around muscle cells at all developmental stages when a7 integrins are present which points to laminin a2 as the major ligand for these integrins in skeletal muscle [21]. Furthermore, in vitro assays demonstrated that a7 integrin may be responsible for the transduction of laminin-induced cell motility [20], and that it also mediates cell adhesive activities on a restricted number of laminin isoforms [10]. Generation of a null allele of the gene for the a7 integrin subunit (ITGA7) in mice results in a muscular dystrophy with onset soon after birth but presenting a typical pattern of skeletal muscle involvement [23]. The histopathological
141
phenotype suggests a primary involvement of the myotendinous junctions. A novel b1 integrin subunit splice variant, the b1D integrin subunit, has recently been shown to be exclusively expressed in skeletal and cardiac muscle [24–26]. In skeletal muscle b1D associates with the a7 integrin subunit and is localized predominantly at the costameres, myotendinous and neuromuscular junctions, whereas in cardiac muscle it is localized at the costameres and in the intercalated discs [24,25]. Knock-out mice carrying a targeted deletion of b1D do not have a disease-associated phenotype, probably due to a compensatory replacement by the b1A integrin subunit [27]. The second structural unit known to be important for the integrity of the skeletal muscle membrane is the dystrophin–glycoprotein complex [28]. The dystrophin– glycoprotein complex (DGC) in skeletal muscle is composed of the cytoskeletal proteins dystrophin; adystroglycan, a 156-kDa extracellular glycoprotein that binds the G-domain of various laminins; b-dystroglycan, a 43-kDa transmembrane glycoprotein that binds the cysteine-rich region of dystrophin [29]; a-, b-, g-, and dsarcoglycan, transmembrane glycoproteins of 50, 43, 35, and 35 kDa, respectively [30]; and a 25-kDa transmembrane protein recently identified as sarcospan [31]. The dystroglycan complex has been shown to bind laminin a2 by its extracellular a subunit [29]. These findings plus the common costameric distribution of dystrophin and bdystroglycan [32,33] indicate that the DGC spans the sarcolemma to link the subsarcolemmal actin-cytoskeleton to the extracellular matrix. The disruption of this linkage is presumed to cause severe sarcolemmal instability which, in turn, may render the muscle fibers susceptible to necrosis, the major pathological event in muscular dystrophies [28]. Mutations in the dystrophin gene and in components of the sarcoglycan complex are associated with a secondary reduction of the DGC complex resulting in severe muscular dystrophy. Mutations in the laminin a2 chain (LAMA2) gene responsible for 50% of the congenital muscular dystrophies (CMD) in Caucasian people do not lead to secondary changes in the DGC complex. In contrast, deficiency of laminin a2 leads to a secondary reduction of laminin b2 on the sarcolemma, another extrasynaptic laminin-polypeptide chain of the extracellular matrix in skeletal muscle [34,35]. In the present study we have investigated the expression of the a7B and b1D integrin subunits in normal human skeletal muscle and in different forms of muscular dystrophy. In addition, we studied the expression of various transmembrane and extracellular matrix proteins in the a7 integrin knock-out mouse [23]. The aim of the study was to test whether deficiency for the laminin a2 chain affects the expression of a7B integrin and, conversely, whether deficiency for a7 integrin alters the expression of extracellular ligand, laminin a2. Further-
142
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
more, other laminin or integrin binding proteins of the basal lamina and plasma membrane of skeletal muscle were also studied in human and murine skeletal muscle from different forms of muscular dystrophy.
2. Patients, materials and methods
2.1. Normal and dystrophic muscle specimens Normal control muscle (n525; 4 days to 36 y) was obtained from orthopedic surgery not related to neuromuscular disease or from muscle biopsies without any pathological abnormalities. Normal fetal muscle (18 week gestation) came from abortions carried out for non-musclerelated reasons. We also studied fetal muscle specimens from four DMD (14–20 weeks gestation) and one fetus with the Walker–Warburg syndrome (20 weeks gestation). The expression of the a7B and b1D integrin subunits was analysed in the following forms of muscular dystrophy: laminin a2 deficient congenital muscular dystrophy (CMD) (n56; 3.5–18 y); dy / dy (laminin a2 deficient murine muscular dystrophy n53); laminin a2 chain positive CMD (n56; 3.5–10 y); Duchenne muscular dystrophy (n57; 5 weeks to 9 y); Becker muscular dystrophy (n54; 2–15 y); a-sarcoglycanopathy (n54; 5– 12 y), g-sarcoglycanopathy (n52; 6–9 y); Walker–Warburg syndrome (n52; 1.5–3 y); Emery–Deifuss muscular dystrophy (n53; 14–21 y); and 12 patients with as yet unclassified limb girdle muscular dystrophy (6–37 y). In addition, muscle specimens of patients with genetically confirmed spinal muscular atrophy (n55; 8 months to 7 y) have been studied. Furthermore, expression of various components of the extracellular matrix as well as of the cytoskeleton have been studied in a7 integrin deficient knock-out mice (n5 3).
affinity purified rabbit polyclonal antibody [40]; mAb 1922 Chemicon Corporation against laminin a2 80 kDa (1:500); mAb rat laminin a2 4H8-2 against laminin a2 (1:2) [21]; mAb 2B1 b1D subunit (1:2) [41], mAb slow myosin (1:100), fast myosin (1:100) (NCL MHCS and MHCF, from Novocastra Laboratories, Newcastle upon Tyne, UK), mAb anti caveolin-3 (1:100; Transduction Lab.), mAb anti-fibronectin (1:100; NCL FIB; from Novocastra Laboratories, Newcastle upon Tyne, UK), polyclonal Ab against nidogen (1:1000; 10461); polyclonal Ab against perlecan (1:200; 10301E) and polyclonal Ab against fibulin-1 (1:100); 10311DL. The following biotinylated antibodies have been used: anti rat Ig (1:200); anti rabbit Ig (1:200); anti mouse IgG (1:200); anti IgG1 (1:200); anti IgG2a (1:200); anti IgG2a (1:200), all from Amersham. Immunofluorescence results were subjected to semiquantitative analysis using the Kontron KS 400V2.00 image analysing program in which immunohistochemical intensity signals are transformed into grey values [42]. A defined line with 500 points of measurements was placed across a given stained section which offered a clear immunoreactivity and was free of artefacts. We analysed 15 measurements at three different sites in each specimen. In order to check for membrane integrity we analysed the immunoreactivity of laminin g1 in each patient. Analysis of these measurements revealed maximum intensity values for immunoreactivity at the skeletal muscle basement membrane and minimum intensity values corresponding to background values in the internal part of the muscle fibre. Median values and standard deviation for the maximum and minimum intensity values of all measurements for normal controls and in each patient were determined for each antibody. To obtain comparable parameters of immunoreactivity for the different antibodies in normal controls and in our patients, difference values defined as the delta between the peak values and background signal, were calculated. The difference values were determined as grey values (gv).
2.2. Immunohistochemistry Immunohistochemical studies were done as described previously [34]. To control for sarcolemmal integrity, immunohistochemical analysis for laminin g1 [36] and spectrin were performed and we routinely did tests without the first specific antibodies in order to check for nonspecific staining by the secondary antibodies. Sections were viewed and photographed using a Zeiss Axioplan MC 80 microscope. The following antibodies were used in our study: mAb against dys1 (1:50), dys2 (1:100), and dys 3 (1:100) [37], b-dystroglycan (1:100), a- and g-sarcoglycan (1:100) (all from Novocastra Laboratories, Newcastle upon Tyne, UK), a-dystroglycan (1:100), affinity purified goat polyclonal antibody [38], b-sarcoglycan (1:100), affinity purified rabbit polyclonal antibody [39], d-sarcoglycan (1:100),
3. Preparation and characterization of the a7B specific antibodies The antiserum against a7B was obtained by immunizing a rabbit with the synthetic peptide GCLAADWHPELGPDGHPVPATA, corresponding to the last 20 amino acids of the murine a7B integrin. An additional N-terminal glycine–cysteine dipeptide was added for coupling to maleimide activated keyhole limpet hemocyanin (KLH) according to the suppliers protocol (PIERCE). Serum titer was monitored by ELISA titration against the same peptide coupled to maleimide activated bovine serum albumin. The same conjugate was used for affinity purification of the antiserum after coupling to
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
CNBr-activated Sepharose 4B (Pharmacia). Specificity of the antiserum and the purified antibodies was proven by negative immunofluorescent staining of cryosection of alpha7 integrin deficient [23] mice versus wild-type mice and by their failure to precipitate a7 integrin from muscle extracts of a7-deficient mice.
4. Results
4.1. Developmental expression of the a7 B and b1 D integrin subunits in human skeletal muscle We have studied the sarcolemmal expression of a7B in skeletal muscle specimens of 25 children of different ages (see Table 1). In addition, we investigated the expression of a7B in atrophic and hypertrophic muscle fibers in five patients with genetically confirmed spinal muscular atrophy (SMA) (age 8 months to 7 years). Surprisingly, in normal fetal skeletal muscle (18th week of gestation) no sarcolemmal staining for the a7B integrin subunit was observed (Table 1; Fig. 1aA). In contrast, there was a clear immunoreactivity in the blood vessels. The same immunohistochemical staining pattern was seen in four fetuses with DMD (18–24 weeks of gestation) and in one fetus (20 week gestation) diagnosed with the Walker–Warburg syndrome (Table 1). Furthermore, in neonatal muscle (n53; at age of 4 days; 4 weeks and 6 weeks) and in muscle of young children (n54; at age of 8 months; 1 y; 1.3 y and 1.9 y) no sarcolemmal expression of a7B integrin was detected, whereas the blood vessels were positive (see Table 1; Fig. 1aB,C). In three children of 2 years of age a weak a7B integrin sarcolemmal staining [15–19 grey values (gv)] in isolated Table 1 Developmental expression of a7B integrin in human skeletal muscle. Data are expressed as grey values of difference values between the maximum and minimum immunofluorescence intensity Age
Number
Normal fetus 18 wk DMD a fetus 18–24 wk WWS b fetus 20 wk 4 days 4 weeks 6 weeks 8 months 1 year 1.3 years 1.9 years 2 years 3 years 4 years 5–15 years 22–36 years
1 4 1 1 1 1 1 1 1 1 3 3 3 9 2
a b
Duchenne muscular dystrophy. Walker–Warburg syndrome.
Expression of a7B integrin Sarcolemma
Blood vessels
No signal No signal No signal No signal No signal No signal No signal No signal No signal No signal 15–19 20–24 34–38 40–45 40–48
11 11 11 11 11 11 11 11 11 11 11 11 11 11 11
143
muscle fibers was observed (see Table 1). These positive muscle fibers were irregularly distributed, but the pattern was not correlated with fiber types, as determined by staining of serial sections with fast and slow myosin and a actinin-3 (data not shown). The a7B integrin sarcolemmal staining was more regular and intense (20–24 gv) in muscle biopsies of 3 year olds (n53) (Fig. 1aD) and showed a continuous sarcolemmal staining by the age of 4 years (n53) (34–38 gv). Investigation of the skeletal muscle of nine children between the ages of 5 (Fig. 1aE) and 15 years, and of two adults (22 y and 36 y) (Fig. 1aF) revealed a clear continuous sarcolemmal immunoreactivity (40–48 gv) which was only slightly stronger than in the 4 year old children. Interestingly, the sarcolemmal expression of a7B in young patients with SMA was different. In two patients (8 and 11 months), no sarcolemmal staining for a7B could be detected in the atrophic muscle fibers. In contrast, the hypertrophic muscle fibers of three SMA patients (1–7 years) showed a strong, continuous sarcolemmal immunoreactivity (34–38 gv), but the atrophic fibers remained negative for a7B even in the older patients (Fig. 1bB). The sarcolemmal staining was not correlated with fiber type, as demonstrated by staining serial sections for fast and slow myosin (Fig. 1bC,D). The sarcolemmal expression of the b1D integrin subunit in human skeletal muscle appeared not to be developmentally regulated. A strong and continuous staining was detectable in the fetal as well as in the adult muscle specimens (44–48 gv). In contrast, the blood vessels showed no immunoreactivity. In contrast to a7B, the expression of b1D integrin was not altered in the patients with SMA (data not shown).
4.2. Expression of the a7 B and b1 D integrin subunits in various forms of muscular dystrophy The sarcolemmal expression of a7B in six patients (5–18 years) with laminin a2 deficient CMD was significantly reduced (8–14 gv), whereas it appeared to be normally expressed in the blood vessels (Fig. 2). In addition, the expression of a7B was also significantly reduced in the dy / dy mice a murine model for laminin a2 chain deficiency (Table 2, data not shown). Interestingly, the expression of a7B in skeletal muscle was not correlated with the amount of the laminin a2 chain. Three patients with a significantly reduced expression of a7B at the sarcolemma were completely negative for laminin a2 (0 gv, Table 2) while three patients showed only a partial deficiency for laminin a2 chain at the skeletal muscle basement membrane. In some forms of CMD a truncated laminin a2 chain is expressed which lacks portions of the N-terminus [42–44]. Two of the three patients with significantly reduced sarcolemmal a7B expression exhibited such N-terminal deletions revealed by differential staining with mAbs 4H8-2 (N-terminus) (12 and 10 gv)
144
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
Fig. 1. (a) Developmental expression of a7B integrin at the sarcolemma of skeletal muscle. Immunohistochemical labelling with the antibody recognizing the a7B integrin subunit showed no sarcolemmal staining of normal skeletal muscle in fetus (A), newborn (B) and 1 year old infant (C). In contrast, the blood vessels stained positive. At 3 years of age an irregular staining at the sarcolemma could be observed (D), whereas at the age of 5 years (E) and in adult skeletal muscle (F) a strong sarcolemmal staining was observed (3200). (b) Sarcolemmal expression of a7B integrin in spinal muscular atrophy. Immunohistochemical staining of laminin a2 (A), a7B integrin (B), slow myosin I (C) and fast myosin II (D) in serial sections of skeletal muscle of a 1.5 year old child with spinal muscular atrophy (SMA). Note the strong sarcolemmal expression of a7B integrin in the hypertrophic fibres in contrast to the negative atrophic fibres (B). The staining of a7B integrin is not correlated to fibre type (C,D) (3200).
and Chemicon 1922 (C-terminus) (68 and 71 gv, Table 2). One patient with significantly reduced a7B expression showed reduced immunoreactivity for both C-terminal and N-terminal fragments of the laminin a2 chain (48 and 39 gv). This indicates that a relative increase in the expression of the C-terminal or N-terminal proteins in partial laminin a2 deficiency does not determine to which degree the expression of a7B is decreased (Fig. 2). Similar findings
could be observed in the dy / dy mice. Semiquantitative analysis revealed that the laminin a2 chain was not completely absent (17–21 gv), whereas a7B expression was significantly reduced at the sarcolemma. Two patients (4.5 and 5 years) with laminin a2 positive CMD showed an irregular reduction of a7B staining at the sarcolemma (22–28 gv) whereas the blood vessels were clearly positive, suggesting that these phenomena are
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
145
Fig. 2. Expression of the a7B and b1D integrin subunits in congenital muscular dystrophy. Immunolabelling of laminin a2 (C-terminus), a7B and b1D integrin in normal controls (5 years), patients with complete and partial deficiency of laminin a2 and one patient with laminin a2 chain positive CMD. In patients with laminin a2 deficient CMD a significant reduction of a7B integrin at the sarcolemma was observed, whereas the blood vessels stained positive. The reduction was not correlated to the amount of laminin a2 chain expression. The patient with laminin a2 chain positive CMD (5 years old) showed an irregular sarcolemmal expression of a7B, but a strong labelling of the blood vessels. The expression of b1D integrin was not altered in any of the patients (3200).
Table 2 Quantitative assessment of the sarcolemmal immunoreactivity of laminin a2, a7B, b1D integrin, dystrophin a-sarcoglycan and b-dystrogylcan in congenital muscular dystrophy and a7 integrin knock-out mice. Data are expressed as grey values of difference values between the maximum and minimum immunofluorescence intensity Laminin a2
a7B integrin
b1D integrin
Dystrophin
a-Sarcoglycan
b-Dystroglycan
86–90 90 0 12–40 17–21 88
40 0 10–14 8–12 5–7 22 / 28
48 15 / 19 41–48 42–46 41–46 45–48
104 104 102–104 101–104 103–106 103 / 104
105 105 102–104 100–104 101–104 101–103
109 107 104–106 102–107 102–106 102–105
104
90
38–44
43–48
102–104
98–105
102–105
106
88
20 / 24
44 / 46
96 / 102
100 / 104
98 / 104
Disease
Age
Number
80 kDa
300 kDa
Normal Control a7 Integrin knock-out mice Laminin a2 negative CMD a Laminin a2 deficient CMD Dy / dy b Laminin a2 positive CMD (a7 integrin diminished) Laminin a2 positive CMD (a7 integrin normal) WWS c
4–36 y 150 d 3.5–6 y 5–18 y 3 4.5–5 y
14 2 3 3 150 d 2
99–106 n.d. 0 48–78 n.d. 105
6–10 y
4
1.5–3 y
2
a
Congenital muscular dystrophy. Laminin a2 deficient murine muscular dystrophy. c Walker–Warburg syndrome. b
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
146
secondary to a primary defect of a protein interacting directly or indirectly with a7B integrin or due to local effects, e.g. cytokine release or protease activity (Fig. 2). The level of expression of a7B at the sarcolemma of
patients with DMD or BMD was increased and, again, did not show any correlation with the level at which dystrophin was expressed (see Table 3, Fig. 3). Furthermore, the upregulation was not restricted to regenerating
Table 3 Quantitative assessment of the sarcolemmal immunoreactivity of laminin a2, a7B, b1D integrin, dystrophin a-sarcoglycan and b-dystroglycan in other forms of muscular dystrophy. Data are expressed as grey values of difference values between the maximum and minimum immunofluorescence intensity Disease
Age
Number
Normal control DMD a ,4 y DMD .4 y BMD b ,4 y BMD .4 y a-Sarcoglycanopathy g-Sarcoglycanopathy EDMD c Limb girdle MD d
4–36 y 5 wk to 2 y 5–12 y 2y 7–15 y 5–12 y 7–10 y 14–21 y 6–37 y
14 4 3 1 3 4 2 3 12
Laminin a2 80 kDa
300 kDa
99–106 99–104 98–103 99–104 99–105 99–105 97–103 99–105 98–106
86–90 88–90 88–90 89–90 89–90 88–90 88–90 89 90
a7B integrin
b1D integrin
Dystrophin
a-Sarcoglycan
b-Dystroglycan
40 62–65 60–65 62–65 61–64 40–44 40–42 42–45 41–46
48 42–47 42–45 44–47 44–46 42–48 43–46 42–48 46–48
104 0 0 28–54 28–48 44–65 35–41 101–104 101–105
105 45–57 45–52 60–88 62–84 0–5 8–10 101–105 101–105
109 34–42 36–41 62–74 64–71 88–94 86–95 98–105 101–106
a
Duchenne muscular dystrophy. Becker muscular dystrophy. c Emery Dreifuss muscular dystrophy. d Unclassified limb girdle muscular dystrophy. b
Fig. 3. Expression of the a7B and b1D integrin subunits in dystrophinopathy and sarcoglycanopathy. Immunolabelling of laminin a2 (C-terminus), a7B and b1D integrin in two patients with DMD (5 weeks and 8 years), one patient with a-sarcoglycanopathy (LGMD 2D) (12 years) and one patient with g-sarcoglycanopathy (LGMD 2C) 10 years (10 years). In dystrophinopathies an overexpression of a7B integrin (even in the young infant) was observed, whereas no overexpression was seen in the sarcoglycanopathies. The expression of b1D integrin was not altered (3200).
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
147
Fig. 4. Expression of various proteins in wild-type and a7 knock-out mice. Immunohistochemical staining against the a7B and b1D integrin subunit, laminin a2 (300 kDa, N-terminus), dystrophin (dys2, C-terminus), b-dystroglycan (b-DG), a-sarcoglycan (a-SG) and g-sarcoglycan (g-SG) in wild-type (wt) and a7 knock-out mice. The expression of laminin a2 chain, dystrophin, b-dystroglycan and a- and g-sarcoglycan is not altered in the a7 knock-out mice. In contrast, a significant reduction of b1D integrin at the sarcolemma of the a7 knock-out mice was observed (3200).
148
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
muscle fibres as shown by staining with neonatal myosin (data not shown). Four patients with DMD (5 wk to 2 y) and one patient with BMD (2 y) showed a strong and continuous sarcolemmal immunoreactivity with anti a7B with grey values between 62 and 65, in contrast to normal controls who did not show any or only minimal immunoreactivity at the sarcolemma at these ages. In addition, six older patients with DMD and BMD (5–15 y) also showed a slightly enhanced staining for a7B integrin (61–65 gv) on the sarcolemma, as compared to age-matched controls (see Table 3). No upregulation of a7B integrin was observed in DMD fetal muscle. Hence, this secondary phenomenon develops between the 24th week of gestation and the 5th week post-partum. The expression of the a7B integrin subunit in four patients with a primary a-sarcoglycan and two patients with primary g-sarcoglycan defects was not different from that in normal controls (40–44 gv) (see Table 3, Fig. 3). The sarcolemmal expression of a7B was not altered in two patients with Walker–Warburg syndrome (1.5 and 3 y), three patients with Emery Dreifuss muscular dystrophy (14–21 y), in four patients with laminin a2 chain positive CMD (6–10 y) and 12 patients with as yet unclassified limb girdle muscular dystrophy (6–37 y) (see Tables 2 and 3). The expression of b1D integrin in the patients with different forms of muscular dystrophy was not altered as compared to normal controls (41–48 gv) (see Tables 2 and 3, Figs. 2 and 3).
4.3. Altered protein binding activities in the a7 integrin deficient mice The sarcolemmal expression of a7B in normal wild-type mice (5–150 d) was strong, although some single fibers did not react with anti-a7B, without a relation with fibre type. In a7 integrin knock-out mice no staining of a7B on the sarcolemma or in blood vessels could be detected (Fig. 4). Because a7B expression is diminished in patients who are completely negative or deficient for laminin a2 chain, one would expect a reduced expression of laminin a2 chain in a7 integrin knock-out mice. Therefore, we determined the expression of the laminin a2 chain in a7 integrin knock-out mice by using the mAb 4H8-2 which recognizes the N-terminal portion of the protein. In addition, other components of the extracellular matrix and the DGC were studied. Interestingly, the expression of the laminin a2 chain was not altered in the a7 knock-out mice (see Table 2, Fig. 4). In addition, dystrophin, a-, b-, g- and d-sarcoglycan and a- and b-dystroglycan (see Table 2, Fig.4), other components of the extracellular matrix (nidogen, collagen VI, perlecan, fibulin-1 and fibronectin) and caveolin-3 were continuously expressed at the skeletal muscle basement
membrane in normal and a7 knock-out mice (data not shown). In contrast, the expression of b1D integrin was significantly decreased (15 gv) at the sarcolemma of a7 knock-out mice as compared to normal mice (see Table 2, Fig. 4).
5. Discussion In the past various mutations of several components of the dystrophin–glycoprotein complex have been shown to be responsible for the pathogenesis of muscular dystrophies [31]. The current ‘membrane theory’ on the pathogenesis of muscular dystrophy is based on the fact that this oligomeric complex connects the subsarcolemmal cytoskeleton to the extracellular matrix and, if disrupted, results in various forms of muscular dystrophy [28]. However, clinical phenotypes differ widely between dystrophinopathy and laminin a2 chain deficient CMD. Furthermore, it is well established that primary mutations in the dystrophin molecule lead to a secondary reduction of the dystroglycan, as well as of the sarcoglycan complex at the sarcolemma [28]. In addition, the primary absence of one sarcoglycan is usually accompanied by a significant reduction of the other sarcoglycans [45] and, in the case of a primary g-sarcoglycan defect, secondary changes of the sarcolemmal localization of dystrophin have been reported [46]. The laminin a2 chain binds to the dystroglycan complex [29,47] and in contrast its expression is not altered in muscular dystrophies resulting from primary mutations of the DGC. Similarly, the expression of the DGC is not affected in congenital muscular dystrophies with primary mutations in the LAMA2 gene encoding the laminin a2 chain. Moreover, the subcellular distribution of dystrophin / b-dystroglycan (costameric) [32,33] is different from that of the laminin a2 chain (homogeneous) [48] and a secondary reduction of the laminin b2 chain has been observed solely in laminin a2 deficient CMD [34] accompanied by an upregulation of an alternative laminin a chain recently identified as laminin a5 [34,49,50]. Recently, Straub et al. [51] demonstrated that Evans blue, a low molecular weight diazo dye, which does not cross into skeletal muscle fibres in normal mice, showed a distinct accumulation in animal models for DMD (mdx mice) and CMD (dy / dy mice). In the mdx mice, a significant accumulation of Evans blue could be detected, whereas in dy / dy mice only weak accumulation of the dye was observed. These findings also implied that different pathogenetic mechanisms must be responsible for the muscle damage, although the disease causing mutations affect two proteins associated with the dystrophin– glycoprotein complex [31]. All lines of evidence point to the fact that other transmembrane linkages must be in-
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
volved in the pathogenesis of laminin a2 chain deficient CMD. The a7b1 integrin is regarded as the principal receptor for laminin-2 and -4 in skeletal muscle [10], and the importance of the a7 and b1 integrin subunits and their alternative splice variants for skeletal muscle development has been well established [8,15,25,41]. The finding that a targeted deletion of the a7 integrin subunit results in muscular dystrophy in mice offered a new approach to study also in humans a potentially important transmembrane connection between the extracellular matrix and the submembrane cytoskeleton. We therefore chose to determine the expression of a7B and b1D integrin subunits in normal skeletal muscle as well as in various forms of muscular dystrophy.
5.1. The sarcolemmal expression of a7 B integrin is developmentally regulated In skeletal muscle the expression of two a7 isoforms is developmentally regulated. In the mouse the a7B isoform is predominantly present in fetal muscle, proliferating myoblasts, and later in developing myotubes [19], and the a7A isoform is expressed in adult mouse skeletal muscle and in myotubes [15,19]. In contrast, the expression of a7B in fetal cardiac muscle of mice is restricted to the blood vessels whereas in the myocardium it is only induced after birth [19]. It is therefore surprising that the sarcolemmal localization of a7B integrin seemed to be subject to a distinct developmental regulation in human skeletal muscle. We could not detect any sarcolemmal staining for a7B until the age of 2 years. In contrast, the blood vessels were strongly positive even in fetal skeletal muscle, resembling the expression pattern of a7B in cardiac muscle of mice. Interestingly, in patients with spinal muscular atrophy a sarcolemmal staining for a7B was observed in the hypertrophic fibers even in patients younger than 2 years of age, while the atrophic fibers were negative in all patients. The sarcolemmal immunoreactivity did not correlate with fibre type (specific staining for fast and slow myosin), but exclusively to fiber hypertrophy. These findings indicate that factors other than developmental stage, such as peripheral nerve innervation or functional requirements, must induce the sarcolemmal expression of the a7B integrin subunit in skeletal muscle.
5.2. Role of a7 B integrin for skeletal muscle function and stability In the present study we found a significant secondary decrease of a7B, a specific transmembrane receptor for the laminin isoforms 2 and 4 [10], in the sarcolemma of six patients with CMD who were either completely negative or partially deficient for the laminin a2 chain as well as in the laminin a2 chain deficient dy / dy mice. Interestingly, the
149
sarcolemmal reduction of a7B was not correlated with the degree of laminin a2 chain expression. These findings are in contrast to the secondary reduction of laminin b2 chain observed in laminin a2 deficient CMD. There is a strong correlation between the reduction of laminin b2 chain and the amount of the C-terminal expression of the laminin a2 chain expressed [42]. Taken together these results suggest that the binding between laminin a2 and the a7B integrin is based not only on a structural but also on a functional association in skeletal muscle. This is not surprising because the a7 integrin subunit contains a rich potential for signal transduction, including those mediated by receptor-like protein phosphatases, serine / threonine kinases and interactions with the actin-based cytoskeleton and it therefore may serve as a functional as well as a mechanical link between the ECM and the cytoskeleton [16]. Furthermore, recently, a selective loss of a7B at the neuromuscular junction in laminin b2 knock-out mice has been reported [9], indicating that the laminin a2 chain, the laminin b2 chain and a7B integrin subunits interact functionally and structurally which is essential for the function and stability of muscle cells. Another possible explanation for the reduction of the a7B integrin subunit in the patients with a partial deficiency of the laminin a2 chain is that the mutation in these patients, for example, affects the binding site of a7b1 integrin directly or indirectly via conformational changes. Further studies of the a7B expression in patients whose laminin a2 chains are partially functionally deficient and have different mutations need to be done to clarify this point. Analysis of the expression of the laminin a2 chain in the a7 integrin knock-out mice revealed normal staining at the skeletal muscle basement membrane. Thus, whereas deficiency for or abnormalities of the laminin a2 chain may induce secondary changes in a7B integrin expression, the expression of the laminin a2 chain remains unaffected even when a7 integrin is completely lacking. Previous investigations have shown that the localization of integrins is regulated via cytoskeletal and intracellular signal potentials (inside–outside) [24,52]. Our findings suggest that the influence of the ECM and in particular the laminin a2 chain may also influence the sarcolemmal localization or the stability of a7B integrin. Interestingly, the sarcolemmal expression of the a7B integrin subunit was upregulated in dystrophinopathies. This overexpression of a7B integrin was not restricted to regenerating muscle fibers and was also irrespective of the levels of dystrophin expression. Because the components of the DGC are normally expressed in the a7 knock-out mice, this upregulation probably might be related to functional properties of a7B integrin rather than a structural link to the DGC. In contrast to a7B, the b1D integrin subunit did not show any pathological abnormalities in sarcolemmal expression in any of our patients with muscular dystrophy
150
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
studied, not even in the CMD patients. One possible explanation for this finding is that loss of the laminin a2 chain results in an upregulation of another integrin a subunit which binds to the b1D integrin subunit. In contrast to the human muscular dystrophies studied we found a significant reduction of b1D at the sarcolemma of the a7 knock-out mice. This indicates that expression of a7B is a prerequisite for that of b1D in the sarcolemma. However, a secondary reduction of a7B integrin does not cause destabilization of sarcolemmal b1D expression. It is as yet unclear if the reduction of the integrin system or the preserved expression of the laminin a2 chain can explain these differences in b1D integrin subunit expression. While our work was in progress, in a study by Vachon et al. [53] a vast reduction of b1D was found in laminin a2 deficient CMD as well as in the dy / dy mice which is entirely in contrast to our results. The different observations might be due to the use of different antibodies. In our study we have used a monoclonal b1D antibody which showed complete lack of staining in the b1D knock-out mice but strong staining in wild-type mice [27]. Furthermore, Vachon et al. [53] described a patchy staining for the a7A and a7B integrin subunits in merosin deficient CMD, but a developmental regulation for a7B was not reported.
It is therefore questionable whether the patchy staining for a7B was due to the developmental stage studied or whether it reflected an altered protein expression due to disease. Finally, we have found two patients with laminin a2 chain positive CMD, who showed a reduced sarcolemmal staining for a7B comparable to the mottled staining for dystrophin in patients with BMD. Because the blood vessels stained positively in both patients, and because laminin a2 as well as other known laminin chains and the components of the DGC were normally expressed, we assume that a primary defect of an additional binding or interacting protein for a7B might be responsible for the muscular dystrophy in these patients. In conclusion, our findings suggest that in skeletal muscle the a7Bb1 integrin serves as a significant transmembraneous functional and structural link between the ECM and the cytoskeleton in addition to the DGC (Fig. 5). Furthermore, the secondary changes in the expression of a7B integrin in laminin a2 chain deficient CMD together with the dystrophic phenotype of mice carrying a nullmutation for a7 integrin, make this protein a potential candidate for human muscular dystrophies with as yet uncharacterized primary protein defect.
Fig. 5. Schematic representation of two protein complexes spanning the plasma membrane from the extracellular matrix to the submembrane cytoskeleton.
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152
Acknowledgements This study was supported by a grant from the Deutsche Forschungsgemeinschaft (VO 392 / 2-4), in part by the Alfried Krupp von Bohlen und Halbach Stiftung, a Netherland Heart Foundation (NHS 96.006) grant to A. Sonnenberg and a Yamanouchi Research Studentship to Arjan van der Flier. We would like to thank Volker Straub and Kevin P. Campbell for kindly supplying the antibodies against a-dystroglycan, b- and d-sarcoglycan. The authors would like to thank Nicola Franke for excellent technical assistance.
References [1] Hynes RO. Integrins: versatility, modulation, and signaling in cell adhesion. Cell 1992;69:11–25. [2] Ruoslahti E, Pierschbacher MD. New perspectives in cell adhesion: RGD and integrins. Science 1987;238:491–7. [3] Leptin M, Aebersold R, Wilcox M. Drosophila position-specific antigens resemble the vertebrate fibronectin-receptor family. EMBO J 1987;6:1037–43. [4] Clark EA, Brugge JS. Integrins and signal transduction pathways: the road taken. Science 1995;268:233–9. [5] Bozyczko D, Decker C, Muschler J, Horwitz AF. Integrin on developing and adult skeletal muscle. Exp Cell Res 1989;183:72– 91. [6] Terracio L, Gullberg D, Rubin K, Craig S, Borg TK. Expression of collagen adhesion proteins and their association with the cytoskeleton in cardiac myocytes. Anat Rec 1989;223:62–71. [7] Beauvais-Jouneau A, Thiery JP. Multiple roles for integrins during development. Biol Cell 1997;89:5–11. [8] Bao ZZ, Lakonishok M, Kaufman S, Horwitz AF. Alpha 7 beta 1 integrin is a component of the myotendinous junction on skeletal muscle. J Cell Sci 1993;106:579–89. [9] Martin PT, Kaufman SJ, Kramer RH, Sanes JR. Synaptic integrins in developing, adult, and mutant muscle: selective association of alpha1, alpha7A, and alpha7B integrins with the neuromuscular junction. Dev Biol 1996;174:125–39. [10] Yao CC, Ziober BL, Squillace RM, Kramer RH. Alpha7 integrin mediates cell adhesion and migration on specific laminin isoforms. J Biol Chem 1996;271:25598–603. [11] Foster RF, Thompson JM, Kaufman SJ. A laminin substrate promotes myogenesis in rat skeletal muscle cultures: analysis of replication and development using antidesmin and anti-BrdUrd monoclonal antibodies. Dev Biol 1987;122:11–20. [12] Kaufman SJ, Foster RF, Haye KR, Faiman LE. Expression of a developmentally regulated antigen on the surface of skeletal and cardiac muscle cells. J Cell Biol 1985;100:1977–87. [13] Goodman SL, Risse G, von der Mark K. The E8 subfragment of laminin promotes locomotion of myoblasts over extracellular matrix. J Cell Biol 1989;109:799–809. [14] Ocalan M, Goodman SL, Kuhl U, Hauschka SD, von der Mark K. Laminin alters cell shape and stimulates motility and proliferation of murine skeletal myoblasts. Dev Biol 1988;125:158–67. [15] Collo G, Starr L, Quaranta V. A new isoform of the laminin receptor integrin alpha 7 beta 1 is developmentally regulated in skeletal muscle. J Biol Chem 1993;268:19019–24. [16] Song WK, Wang W, Sato H, Bielser DA, Kaufman SJ. 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 1993;106:1139–52.
151
[17] Ziober BL, Vu MP, Waleh N, Crawford J, Lin CS, Kramer RH. Alternative extracellular and cytoplasmic domains of the integrin alpha 7 subunit are differentially expressed during development. J Biol Chem 1993;268:26773–83. [18] Song WK, Wang W, Foster RF, Bielser DA, Kaufman SJ. H36-alpha 7 is a novel integrin alpha chain that is developmentally regulated during skeletal myogenesis [published erratum appears in J Cell Biol 1992;118(1):213]. J Cell Biol 1992;117:643–57. [19] Velling T, Collo G, Sorokin L, Durbeej M, Zhang H, Gullberg D. Distinct alpha 7A beta 1 and alpha 7B beta 1 integrin expression patterns during mouse development: alpha 7A is restricted to skeletal muscle but alpha 7B is expressed in striated muscle, vasculature, and nervous system. Dev Dyn 1996;207:355–71. [20] Echtermeyer F, Schober S, Poschl E, von der Mark H, von der Mark K. Specific induction of cell motility on laminin by alpha 7 integrin. J Biol Chem 1996;271:2071–5. [21] Schuler F, Sorokin LM. Expression of laminin isoforms in mouse myogenic cells in vitro and in vivo. J Cell Sci 1995;108:3795–805. [22] Yao CC, Breuss J, Pytela R, Kramer RH. Functional expression of the alpha 7 integrin receptor in differentiated smooth muscle cells. J Cell Sci 1997;110:1477–87. [23] Mayer U, Saher G, Fassler R, Bornemann A, Echtermeyer F, von der Mark H, Miosge N, Poschl E, von der Mark K. Absence of integrin alpha 7 causes a novel form of muscular dystrophy. Nat Genet 1997;17:318–23. [24] Belkin AM, Zhidkova NI, Balzac F, Altruda F, Tomatis D, Maier A, Tarone G, Koteliansky VE, Burridge K. Beta 1D integrin displaces the beta 1A isoform in striated muscles: localization at junctional structures and signaling potential in nonmuscle cells. J Cell Biol 1996;132:211–26. [25] van der Flier A, Kuikman I, Baudoin C, van der Neut R, Sonnenberg A. A novel beta 1 integrin isoform produced by alternative splicing: unique expression in cardiac and skeletal muscle. FEBS Lett 1995;369:340–4. [26] Zhidkova NI, Belkin AM, Mayne R. Novel isoform of beta 1 integrin expressed in skeletal and cardiac muscle. Biochem Biophys Res Commun 1995;214:279–85. [27] Baudoin C, Goumans MJ, Mummery C, Sonnenberg A. Knockout and knockin of the beta1 exon D define distinct roles for integrin splice variants in heart function and embryonic development. Genes Dev 1998;12:1202–16. [28] Campbell KP. Three muscular dystrophies: loss of cytoskeleton– extracellular matrix linkage. Cell 1995;80:675–9. [29] Henry MD, Campbell KP. Dystroglycan: an extracellular matrix receptor linked to the cytoskeleton. Curr Opin Cell Biol 1996;8:625–31. [30] Straub V, Campbell KP. Muscular dystrophies and the dystrophin– glycoprotein complex. Curr Opin Neurol 1997;10:168–75. [31] Crosbie RH, Heighway J, Venzke DP, Lee JC, Campbell KP. Sarcospan, the 25-kDa transmembrane component of the dystrophin–glycoprotein complex. J Biol Chem 1997;272:31221–4. [32] Herrmann R, Anderson LV, Voit T. Costameric distribution of betadystroglycan (43 kDa dystrophin-associated glycoprotein) in normal and dystrophin-deficient human skeletal muscle. Biochem Soc Trans 1996;24:501–6. [33] Straub V, Bittner RE, Leger JJ, Voit T. Direct visualization of the dystrophin network on skeletal muscle fiber membrane. J Cell Biol 1992;119:1183–91. [34] Cohn RD, Herrmann R, Wewer UM, Voit T. Changes of laminin beta 2 chain expression in congenital muscular dystrophy. Neuromuscular Disord 1997;7:373–8. [35] Wewer UM, Durkin ME, Zhang X, Laursen H, Nielsen NH, Towfighi J, Engvall E, Albrechtsen R. Laminin beta 2 chain and adhalin deficiency in the skeletal muscle of Walker–Warburg syndrome (cerebro-ocular dysplasia-muscular dystrophy). Neurology 1995;45:2099–101. [36] Mayer U, Nischt R, Poschl E, Mann K, Fukuda K, Gerl M, Yamada
152
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
R.D. Cohn et al. / Journal of the Neurological Sciences 163 (1999) 140 – 152 Y, Timpl R. A single EGF-like motif of laminin is responsible for high affinity nidogen binding. EMBO J 1993;12:1879–85. Nicholson LV, Davison K, Falkous G, Harwood CEOD, Slater CR, Harris JB. Dystrophin in skeletal muscle. I. Western blot analysis using a monoclonal antibody. J Neurol Sci 1989;94:125–36. Ibraghimov-Beskrovnaya O, Milatovich A, Ozcelik T, Yang B, Koepnick K, Francke U, Campbell KP. Human dystroglycan: skeletal muscle cDNA, genomic structure, origin of tissue specific isoforms and chromosomal localization. Hum Mol Genet 1993;2:1651–7. Lim LE, Duclos F, Broux O, Bourg N, Sunada Y, Allamand V, Meyer J, Richard I, Moomaw C, Slaughter C et al. Beta-sarcoglycan: characterization and role in limb-girdle muscular dystrophy linked to 4q12. Nat Genet 1995;11:266–73. Jung D, Duclos F, Apostol B, Straub V, Lee JC, Allamand V, Venzke DP, Sunada Y, Moomaw CR, Leveille CJ, Slaughter CA, Crawford TO, McPherson JD, Campbell KP. Characterization of delta-sarcoglycan, a novel component of the oligomeric sarcoglycan complex involved in limb-girdle muscular dystrophy. Pediatr Neurol 1997;16:79–82. van der Flier A, Gaspar AC, Thorsteinsdottir S, Baudoin C, Groeneveld E, Mummery CL, Sonnenberg A. Spatial and temporal expression of the beta1D integrin during mouse development. Dev Dyn 1997;210:472–86. Cohn RD, Herrmann R, Sorokin L, Wewer UM, Voit T. Laminin alpha2 chain-deficient congenital muscular dystrophy: variable epitope expression in severe and mild cases. Neurology 1998;51:94– 100. Allamand V, Sunada Y, Salih MA, Straub V, Ozo CO, Al-Turaiki MH, Akbar M, Kolo T, Colognato H, Zhang X, Sorokin LM, Yurchenco PD, Tryggvason K, Campbell KP. Mild congenital muscular dystrophy in two patients with an internally deleted laminin alpha2-chain. Hum Mol Genet 1997;6:747–52. Sewry CA, Naom IMDA, Sorokin L, Bruno S, Wilson LA, Dubowitz V, Muntoni F. Variable clinical phenotype in merosindeficient congenital muscular dystrophy associated with differential immunolabelling of two fragments of the laminin alpha 2 chain. Neuromuscular Disord 1997;7:169–75.
[45] Mizuno Y, Noguchi S, Yamamoto H, Yoshida M, Nonaka I, Hirai S, Ozawa E. Sarcoglycan complex is selectively lost in dystrophic hamster muscle. Nihon Shinkei Seishin Yakurigaku Zasshi 1995;15:289–93. [46] Vainzof M, Passos-Bueno MR, Canovas M, Moreira ES, Pavanello RC, Marie SK, Anderson LV, Bonnemann CG, McNally EM, Nigro V, Kunkel LM, Zatz M. The sarcoglycan complex in the six autosomal recessive limb-girdle muscular dystrophies. Hum Mol Genet 1996;5:2019–22. [47] Cohen MW, Jacobson C, Yurchenco PD, Morris GE, Carbonetto S. Laminin-induced clustering of dystroglycan on embryonic muscle cells: comparison with agrin-induced clustering. J Cell Biol 1997;136:1047–58. [48] Straub V, Herrmann R, Bittner R, Anderson LVB, Leger JJ, Voit T. Distribution and organizytion of utrophin and merosin in normal and dystrophic muscle fibers. In: Fukuyama Y, Osawa M, Saito K, editors, Congenital muscular dystrophies, Elsevier, Amsterdam, 1997. [49] Tiger CF, Champliaud MF, Pedrosa-Domellof F, Thornell LE, Ekblom P, Gullberg D. Presence of laminin alpha5 chain and lack of laminin alpha1 chain during human muscle development and in muscular dystrophies. J Biol Chem 1997;272:28590–5. [50] Tome FM, Evangelista T, Leclerc A, Sunada Y, Manole E, Estournet B, Barois A, Campbell KP, Fardeau M. Congenital muscular dystrophy with merosin deficiency. C R Acad Sci III 1994;317:351– 7. [51] Straub V, Rafael JA, Chamberlain JS, Campbell KP. Animal models for muscular dystrophy show different patterns of sarcolemmal disruption. J Cell Biol 1997;139:375–85. [52] Sastry SK, Lakonishok M, Thomas DA, Muschler J, Horwitz AF. Integrin alpha subunit ratios, cytoplasmic domains, and growth factor synergy regulate muscle proliferation and differentiation. J Cell Biol 1996;133:169–84. [53] Vachon PH, Xu H, Liu L, Loechel F, Hayashi Y, Arahata K, Reed JC, Wewer UM, Engvall E. Integrins (alpha7beta1) in muscle function and survival. Disrupted expression in merosin-deficient congenital muscular dystrophy. J Clin Invest 1997;100:1870–81.
