Abstract

Several types of muscular dystrophy are caused by defective linkage between α-dystroglycan (α-DG) and laminin. Among these, dystroglycanopathy, including Fukuyama-type congenital muscular dystrophy (FCMD), results from abnormal glycosylation of α-DG. Recent studies have shown that like-acetylglucosaminyltransferase (LARGE) strongly enhances the laminin-binding activity of α-DG. Therefore, restoration of the α-DG–laminin linkage by LARGE is considered one of the most promising possible therapies for muscular dystrophy. In this study, we generated transgenic mice that overexpress LARGE (LARGE Tg) and crossed them with dy2J mice and fukutin conditional knockout mice, a model for laminin α2-deficient congenital muscular dystrophy (MDC1A) and FCMD, respectively. Remarkably, in both the strains, the transgenic overexpression of LARGE resulted in an aggravation of muscular dystrophy. Using morphometric analyses, we found that the deterioration of muscle pathology was caused by suppression of muscle regeneration. Overexpression of LARGE in C2C12 cells further demonstrated defects in myotube formation. Interestingly, a decreased expression of insulin-like growth factor 1 (IGF-1) was identified in both LARGE Tg mice and LARGE-overexpressing C2C12 myotubes. Supplementing the C2C12 cells with IGF-1 restored the defective myotube formation. Taken together, our findings indicate that the overexpression of LARGE aggravates muscular dystrophy by suppressing the muscle regeneration and this adverse effect is mediated via reduced expression of IGF-1.

INTRODUCTION

Dystroglycan (DG) is encoded by a single gene, Dag1, located on human chromosome 3p21 and cleaved into two proteins, α- and β-DG, by post-translational processing (1). α-DG is an extracellular peripheral membrane protein and although its molecular mass is predicted to be 72 kDa from its amino acid sequence, its apparent molecular mass in skeletal muscle is 156 kDa owing to extensive glycosylation. α-DG binds to several extracellular matrix proteins including laminin, agrin and perlecan (1) and synaptic proteins such as neurexin and pikachurin (2,3). Transmembrane protein β-DG anchors α-DG at the extracellular surface of the plasma membrane (4). The cytoplasmic domain of β-DG interacts with dystrophin, a large cytoplasmic protein that binds to F-actin (5). Thus, DG plays a central role in stabilizing the plasma membrane by acting as an axis that links the extracellular matrix to the cytoskeleton.

α-DG is composed of distinct three domains: the N-terminal domain, the mucin-like domain and the C-terminal domain. The N-terminal domain is cleaved by the proprotein convertase furin and secreted outside cells (6,7). The mucin-like domain is highly glycosylated by O-linked oligosaccharides. Chiba et al. identified the O-mannosyl glycan, Siaα2-3Galβ1-4GlcNAcβ1-2Man, which is attached to the mucin-like domain of α-DG and is necessary for binding activity with laminin (8). The O-mannose is phosphorylated at the 6th position and further extended on the distal side of the phosphate by a repeating disaccharide -3Xylα1-3GlcUAβ1-. This extension is catalyzed by like-acetylglucosaminyltransferase (LARGE) and regulates the binding activity of α-DG (9,10). Most recently, it was demonstrated that SGK196 phosphorylates the 6th position of O-mannose of GalNAcβ1-3GlcNAcβ1-4Man on α-DG after the mannose had been modified by glycosyltransferase-like domain containing 2 (GTDC2) and β3-N-acetylgalactosaminyltransferase 2 (B3GALNT2) (11).

The linkage between α-DG and laminin is crucial to stabilize the sarcolemma, and disruption of this linkage caused by aberrant glycosylation of α-DG underlies the pathogenesis of several types of muscular dystrophy (12). These disorders are collectively called dystroglycanopathy. Their phenotypes range from severe congenital muscular dystrophy (CMD) to milder adult-onset limb-girdle muscular dystrophy (13). The most severe end of this spectrum includes Walker–Warburg syndrome (OMIM236670), muscle–eye–brain disease (OMIM253280) and Fukuyama-type congenital muscular dystrophy (FCMD) (OMIM253800). These severe forms of muscular dystrophy are often associated with brain anomalies and ocular defects (13). FCMD is the second most common childhood muscular dystrophy in Japan. The founder mutation, a SINE-VNTR-Alu retrotransposal insertion to the 3′ non-coding region, as well as several point mutations of fukutin has been reported in Japan and other countries (14,15). Thus far, mutations in 17 genes encoding known or putative glycosyltransferases have been identified as causative of the dystroglycanopathy. Among their gene products, enzymatic activities of protein O-mannosyltransferase 1 (POMT1) (16,17), protein O-mannosyltransferase 2 (POMT2) (17,18), protein O-linked mannose β-1,2-N-acetylglucosaminyltransferase 1 (POMGnT1) (19), LARGE (10,20), SGK196 (11,21), GTDC2 (19,22) and B3GALNT2 (11,23) have been shown to be involved in the formation of O-mannosyl glycan on α-DG. Defects in dolichol kinase (DOLK) (24), GDP-mannose pyrophosphorylase B (GMPPB) (25), dolichol-P-mannose synthase polypeptide 1 (DMP1) (26), dolichol-P-mannose synthase polypeptide 2 (DMP2) (27) and dolichol-P-mannose synthase polypeptide 3 (DMP3) (28) lead to dystroglycanopathy by inhibiting the biosynthesis of dolichol-P-mannose, which is necessary for the initial step of O-mannosylation. Although the functions of fukutin (14), fukutin-related protein (FKRP) (29), β-1,3-N-acetylglucosaminyltransferase 1 (B3GNT1) (30), isoprenoid synthase domain containing (ISPD) (31,32) and transmembrane protein 5 (TMEM5) (33) have not been elucidated, they are thought to be involved in the formation of O-mannosyl glycan. In addition to these secondary defects in glycosylation, a primary mutation was identified in the N-terminal domain of α-DG in a patient with dystroglycanopathy (34).

Apart from dystroglycanopathy, defects in the ligand of α-DG, laminin-211, also lead to muscular dystrophy. Laminin α2 chain-deficient congenital muscular dystrophy (MDC1A) (OMIM607855) presents as severe CMD associated with dysmyelination of peripheral nerves and white matter abnormalities in the brain. MDC1A is one of the most common forms of CMD, and it accounts for 30–40% of all patients with CMD (35). The causative mutations of MDC1A were identified in the gene encoding laminin α2 chain, LAMA2 (36,37). These facts imply that the axis composed of α-DG and laminin is crucial to protect muscle cells from dystrophic changes.

Barresi et al. reported, using fibroblasts from patients with dystroglycanopathy such as FCMD and muscle–eye–brain disease, that overexpression of LARGE facilitated the glycosylation of α-DG and restored the reduced laminin-binding activity of α-DG irrespective of the gene involved (38). Therefore, the up-regulation of laminin-binding activity of α-DG by the overexpression of LARGE could provide a novel therapeutic strategy for muscular dystrophy, regardless of whether the primary mutation resides in LARGE or not. To test this hypothesis, we generated transgenic mice overexpressing LARGE (LARGE Tg mice) and crossed them with dy2J mice (39) and MCK-fukutin conditional knockout mice (FKTN cKO mice) (40), a model of MDC1A and FCMD, respectively. Remarkably, the muscular dystrophy of both dy2J and FKTN cKO mice was aggravated by the overexpression of LARGE. Morphometric analyses demonstrated that the regeneration of muscle fiber was suppressed in these mice. Furthermore, the transfection of C2C12 cells with LARGE led to deficits in proliferation and fusion of myoblasts. Finally, we found that the expression of insulin-like growth factor (IGF-1) was reduced in both LARGE Tg mice and LARGE-transfected C2C12 cells, and supplementation of IGF-1 rescued the suppressed myotube formation of C2C12 cells. These data indicate that the overexpression of LARGE suppresses the regeneration of skeletal muscle, at least partially, via down-regulation of IGF-1 and worsens the muscular dystrophy of dy2J and FKTN cKO mice.

RESULTS

Generation and characterization of LARGE transgenic mice

First, to test the effect of overexpression of LARGE, we generated transgenic mice that overexpress LARGE (LARGE Tg mice) using CAG promoter (41). Large Tg mice were born, grew normally and exhibited no obvious motor or behavioral abnormalities. The overexpression of LARGE was demonstrated in each tissue, including skeletal muscle, cardiac muscle, brain, peripheral nerve, kidney and liver by western blotting (Fig. 1A). Consistently, the immunoreactivity of IIH6, which recognizes glycosylated form of α-DG, and the laminin-binding activity of α-DG were markedly increased in each tissue of LARGE Tg mice, as compared with the wild type (Fig. 1A and B). In the brain of LARGE Tg mice, the enhanced signal of IIH6 was localized to the brain surface and capillaries (Fig. 1B). Further, in situ overlay assay showed increased binding activity of skeletal and cardiac muscles for both laminin and agrin (Fig. 1C). The expression of laminin α2, β1 and γ1 chains was not significantly changed as confirmed by immunofluorescent analysis (Supplementary Material, Fig. S1A). On western blotting, the expression of each laminin chain was not altered, except for the finding that laminin α1 and β1/γ1 chains were up-regulated in the cardiac muscle and liver (Supplementary Material, Fig. S1B). Hematoxylin and eosin (H-E) staining revealed no morphological changes in each tissue observed (Fig. 1D).

Figure 1.

Overexpression of LARGE and hyperglycosylation of α-DG in LARGE Tg mice. (A) Western blot analysis showed that LARGE was expressed in greater quantities in all tissues tested including skeletal muscle (SKL), cardiac muscle (CRD), brain, peripheral nerve (PN), kidney (KNY) and liver of LARGE Tg mice (Tg) in comparison with wild type (WT). Immunoreactivity of IIH6, an antibody against the glycan structure of α-DG and laminin-binding activity demonstrated by laminin blot overlay assay (Lam O/L) strongly increased in these tissues. Expression of β-DG was unchanged. (B) Immunofluorescent analysis demonstrated that immunoreactivity of IIH6 markedly increased in each tissue of LARGE Tg mice. (C) In situ laminin and agrin overlay assay showed increased binding activity of skeletal (SKL) and cardiac (CRD) muscle for both laminin and agrin in LARGE Tg mice. (D) H-E staining of each tissue of LARGE Tg mice was indistinguishable from control. Scale bar represents 50 μm.

Figure 1.

Overexpression of LARGE and hyperglycosylation of α-DG in LARGE Tg mice. (A) Western blot analysis showed that LARGE was expressed in greater quantities in all tissues tested including skeletal muscle (SKL), cardiac muscle (CRD), brain, peripheral nerve (PN), kidney (KNY) and liver of LARGE Tg mice (Tg) in comparison with wild type (WT). Immunoreactivity of IIH6, an antibody against the glycan structure of α-DG and laminin-binding activity demonstrated by laminin blot overlay assay (Lam O/L) strongly increased in these tissues. Expression of β-DG was unchanged. (B) Immunofluorescent analysis demonstrated that immunoreactivity of IIH6 markedly increased in each tissue of LARGE Tg mice. (C) In situ laminin and agrin overlay assay showed increased binding activity of skeletal (SKL) and cardiac (CRD) muscle for both laminin and agrin in LARGE Tg mice. (D) H-E staining of each tissue of LARGE Tg mice was indistinguishable from control. Scale bar represents 50 μm.

Overexpression of LARGE leads to aggravation of muscular dystrophy in dy2J mice

In dy2J mice, a model for MDC1A, a 57-amino acid deletion and a 1-amino acid substitution in the N-terminal domain of the laminin α2 chain lead to the expression of an N-terminally truncated protein (39). Although the mutant protein exhibits defective polymerization and reduced heparin binding (42), it has not been determined whether its binding activity to α-DG is still preserved. Thus, we first conducted a pH 12 extract overlay assay, in which α-DG on blots were overlaid with pH 12 extracts from the skeletal muscle of dy2J mice that contain native laminin-211. When α-DG from control mouse was overlaid with the pH 12 extract from dy2J mice, the laminin-211 bound to α-DG, which was also detected by IIH6 and anti-α-DG core protein, to the same degree as that from the control extract (Fig. 2A, left). The laminin-211 from dy2J mice also bound to α-DG of LARGE Tg mice as readily as that from control mice (Fig. 2A, right). These results confirm that the mutant laminin-211 in dy2J mice is capable of binding to α-DG.

Figure 2.

Overexpression of LARGE and phenotype of dy2J/LARGE mice. (A) α-DG enriched by WGA chromatography was transferred to blots, overlaid with pH 12 extracts from dy2J and control mice, and the bound laminin was detected by anti-laminin. Both laminin extracted from dy2J and control mice bound equally to α-DG in control mice, which was also visualized by IIH6 antibody and antibody against core protein (left). Laminin extracted from dy2J mice also bound to α-DG in LARGE Tg mice in the same manner as in the control (right). (B) Western blotting analysis of skeletal muscle showed overexpression of LARGE in dy2J/LARGE mice (n = 3 of each type). IIH6 revealed intense and broad bands with molecular mass ranging from 150 to >250 kDa in dy2J/LARGE mice, compared with those around 150 kDa in control and dy2J mice. Signals of β-DG and internal control, α-actinin were not altered. Laminin blot overlay assay (Lam O/L) revealed increased binding activity of α-DG in dy2J/LARGE mice. (C) Immunofluorescent analysis showed increased signal of IIH6 at sarcolemma of dy2J/LARGE mice. Immunostaining of β-DG, laminin α2, β1 and γ1 chains were not significantly changed. Lam = laminin. Scale bar represents 50 μm. (D) Body weight of dy2J/LARGE mice was significantly smaller than dy2J mice both in males and females at 8 weeks (n = 5, 6, 10, 4, 6 and 15 for male control, female control, male dy2J, female dy2J, male dy2J/LARGE and female dy2J/LARGE, respectively). (E) Grip strength of dy2J/LARGE was weaker than dy2J mice in both males and females at 8 weeks (n = 4, 6, 5, 4, 5 and 6 for male control, female control, male dy2J, female dy2J, male dy2J/LARGE and female dy2J/LARGE, respectively). *P < 0.05 and **P < 0.01. (F) Kaplan–Meier estimates of survival probabilities revealed a far shorter life span of dy2J/LARGE mice than dy2J mice. MST = mean survival time (weeks).

Figure 2.

Overexpression of LARGE and phenotype of dy2J/LARGE mice. (A) α-DG enriched by WGA chromatography was transferred to blots, overlaid with pH 12 extracts from dy2J and control mice, and the bound laminin was detected by anti-laminin. Both laminin extracted from dy2J and control mice bound equally to α-DG in control mice, which was also visualized by IIH6 antibody and antibody against core protein (left). Laminin extracted from dy2J mice also bound to α-DG in LARGE Tg mice in the same manner as in the control (right). (B) Western blotting analysis of skeletal muscle showed overexpression of LARGE in dy2J/LARGE mice (n = 3 of each type). IIH6 revealed intense and broad bands with molecular mass ranging from 150 to >250 kDa in dy2J/LARGE mice, compared with those around 150 kDa in control and dy2J mice. Signals of β-DG and internal control, α-actinin were not altered. Laminin blot overlay assay (Lam O/L) revealed increased binding activity of α-DG in dy2J/LARGE mice. (C) Immunofluorescent analysis showed increased signal of IIH6 at sarcolemma of dy2J/LARGE mice. Immunostaining of β-DG, laminin α2, β1 and γ1 chains were not significantly changed. Lam = laminin. Scale bar represents 50 μm. (D) Body weight of dy2J/LARGE mice was significantly smaller than dy2J mice both in males and females at 8 weeks (n = 5, 6, 10, 4, 6 and 15 for male control, female control, male dy2J, female dy2J, male dy2J/LARGE and female dy2J/LARGE, respectively). (E) Grip strength of dy2J/LARGE was weaker than dy2J mice in both males and females at 8 weeks (n = 4, 6, 5, 4, 5 and 6 for male control, female control, male dy2J, female dy2J, male dy2J/LARGE and female dy2J/LARGE, respectively). *P < 0.05 and **P < 0.01. (F) Kaplan–Meier estimates of survival probabilities revealed a far shorter life span of dy2J/LARGE mice than dy2J mice. MST = mean survival time (weeks).

Thereafter, we crossed LARGE Tg mice with dy2J mice to generate dy2J mice overexpressing LARGE (dy2J/LARGE mice). Western blotting analysis showed that LARGE was overexpressed in all dy2J/LARGE mice, and IIH6 recognized intense and broad bands with molecular masses ranging from 150 to >250 kDa, indicating hyperglycosylation of α-DG in dy2J/LARGE mice (Fig. 2B). Signals of β-DG and α-actinin were not altered. Laminin overlay analysis revealed increased binding activity of α-DG in dy2J/LARGE mice (Fig. 2B). Immunofluorescent analysis showed again a remarkable increase in the reactivity of IIH6 at the sarcolemma of dy2J/LARGE mice (Fig. 2C). Immunolabeling of β-DG, laminin α2, β1 and γ1 chains showed no obvious differences (Fig. 2C). However, in spite of the conspicuous hyperglycosylation of α-DG, dy2J/LARGE mice exhibited a small stature and sometimes pronounced inward bend of the hind limbs, suggestive of muscle weakness (Supplementary Material, Fig. S2A). Consistent with this observation, the body weight of dy2J/LARGE mice began to decrease at 4 weeks of age as compared with dy2J mice (data not shown), and it was significantly smaller in both male and female mice at 8 weeks (Fig. 2D). The grip strength was also weaker in both male and female mice at 8 weeks (Fig. 2E). Further, the plot of Kaplan–Meyer estimates of survival probabilities revealed a far shorter life span of dy2J/LARGE mice compared with dy2J mice (Fig. 2F).

On pathological examination at 16 weeks of age using H-E staining, dy2J mice exhibited a moderate degree of muscular dystrophy including occasional myofiber necrosis, interstitial fibrosis and central nucleation. However, the fibrosis and fatty infiltration were rather prominent in all the skeletal muscles tested in dy2J/LARGE mice, including the gastrocnemius, tibialis anterior and quadriceps muscles (Fig. 3A). The same pathological changes were observed at 8 weeks in dy2J/LARGE mice (data not shown). In the following morphometric analyses, we used the gastrocnemius muscles of mice 16–22 weeks of age. The endomysial fibrosis area revealed by anti-mouse IgG (43) was significantly increased in dy2J/LARGE mice (Supplementary Material, Fig. S2B, Fig. 3B). The diameter of muscle fibers of dy2J/LARGE mice tended to be smaller than those of dy2J mice (Fig. 3C), and the percentage of central nuclei was significantly decreased in dy2J/LARGE mice, compared with dy2J mice (Fig. 3D). Because the fibrosis was more prominent in dy2J/LARGE mice than that in dy2J mice, we expected that the regeneration of muscle fibers would be highly activated to compensate for the damaged tissue. However, the myofibers were found to be smaller and central nuclei fewer in dy2J/LARGE mice, both findings implying that the regeneration is suppressed. Next, we performed electron microscopic analysis to observe the basement membrane. In control mice, the thickness of the basement membrane was consistent, and the double layers of lamina densa and lamina lucida were clearly seen. In contrast, the thickness of the basement membrane was variable, and thicker and thinner regions were observed in both dy2J and dy2J/LARGE mice. There was no difference in the basement membrane structure between dy2J and dy2J/LARGE mice (Supplementary Material, Fig. S2C).

Figure 3.

Aggravation of muscular dystrophy in dy2J/LARGE mice. (A) H-E staining of skeletal muscles at 16 weeks of age demonstrated occasional myofiber necrosis, interstitial fibrosis and central nuclei in dy2J mice. Remarkably, fibrosis and fatty infiltration was rather prominent in dy2J/LARGE mice. Abbreviations: GC, gastrocnemius; TA, tibialis anterior; quad, quadriceps. Scale bar represents 50 μm. (B) Quantitative analysis of gastrocnemius muscle revealed that the fibrosis area was significantly increased in dy2J/LARGE mice (n = 3, for all genotypes). (C) Measurements of muscle fiber diameter demonstrated that diameters of dy2J/LARGE mice muscle cells tended smaller than those of dy2J mice (n = 3, for all genotypes). (D) Counting of central nuclei revealed that the percentage was significantly decreased in dy2J/LARGE mice (n = 3, for all genotypes). *P < 0.05, **P < 0.01 and ***P < 0.001.

Figure 3.

Aggravation of muscular dystrophy in dy2J/LARGE mice. (A) H-E staining of skeletal muscles at 16 weeks of age demonstrated occasional myofiber necrosis, interstitial fibrosis and central nuclei in dy2J mice. Remarkably, fibrosis and fatty infiltration was rather prominent in dy2J/LARGE mice. Abbreviations: GC, gastrocnemius; TA, tibialis anterior; quad, quadriceps. Scale bar represents 50 μm. (B) Quantitative analysis of gastrocnemius muscle revealed that the fibrosis area was significantly increased in dy2J/LARGE mice (n = 3, for all genotypes). (C) Measurements of muscle fiber diameter demonstrated that diameters of dy2J/LARGE mice muscle cells tended smaller than those of dy2J mice (n = 3, for all genotypes). (D) Counting of central nuclei revealed that the percentage was significantly decreased in dy2J/LARGE mice (n = 3, for all genotypes). *P < 0.05, **P < 0.01 and ***P < 0.001.

Overexpression of LARGE leads to aggravation of muscular dystrophy in FKTN cKO mice

We crossed LARGE Tg mice with FKTN cKO mice to generate FKTN cKO/LARGE mice. Western blotting analysis with IIH6 and laminin overlay assay showed that the defects in glycosylation and laminin binding of α-DG in FKTN cKO mice were restored and they were even enhanced in FKTN cKO/LARGE mice (Fig. 4A). Immunoreactivity of β-DG and α-actinin were unaltered. Immunofluorescent analysis again showed abolished reactivity for IIH6 in FKTN cKO mice, whereas enhanced reactivity in FKTN cKO/LARGE mice (Fig. 4B). Immunolabeling of β-DG, laminin α2, β1 and γ1 chains was not changed (Fig. 4B). The muscular dystrophy of FKTN cKO mice is mild (40). Both FKTN cKO and FKTN cKO/LARGE mice were born, grew normally and exhibited no obvious motor or behavioral abnormalities (Supplementary Material, Fig. S3A). Also, the pathological changes in FKTN cKO mice are mild and characterized by occasional myonecrosis and central nucleation. The fibrosis and infiltration of fat are only rarely observed in the older mice (40). H-E staining of FKTN cKO mice at 24 weeks demonstrated these mild pathological features in the gastrocnemius, tibialis anterior, quadriceps and diaphragm muscles (Fig. 4C). However, the fibrosis and infiltration of fat were often observed in FKTN cKO/LARGE mice (Fig. 4C). In the following morphometric analyses, we used the gastrocnemius muscles of mice 24–27 weeks of age. The fibrosis area revealed by anti-mouse IgG (43) was significantly increased in FKTN cKO/LARGE mice (Supplementary Material, Fig. S3B, Fig. 4D). The diameter of muscle fibers of FKTN cKO/LARGE mice tended to be smaller than those of FKTN cKO mice (Fig. 4E), and the percentage of central nuclei was statistically unchanged between them (Fig. 4F). These results are very similar to those obtained for dy2J/LARGE mice. Thus, in two distinct mouse models, it is clear that the overexpression of LARGE resulted in aggravation of muscular dystrophy most likely via suppression of muscle regeneration.

Figure 4.

Aggravation of muscular dystrophy in FKTN cKO/LARGE mice. (A) Western blotting analysis showed that LARGE was overexpressed in FKTN cKO/LARGE mice (n = 3 of each type). IIH6 antibody failed to detect glycosylated α-DG in FKTN cKO mice, whereas it revealed intense and broad bands of α-DG in FKTN cKO/LARGE mice. Signals of β-DG and the internal control, α-actinin were unaltered. Laminin blot overlay assay (Lam O/L) demonstrated that binding activity of α-DG was abolished in FKTN cKO mice, whereas markedly increased in FKTN cKO/LARGE mice. (B) Immunofluorescent analysis showed that immunoreactivity of IIH6 was abolished in FKTN cKO mice, whereas it was markedly enhanced in FKTN cKO/LARGE mice. Signals of β-DG, laminin α2, β1 and γ1 chains were unchanged among them. Lam = laminin. Scale bar represents 50 μm. (C) H-E staining of skeletal muscle at 24 weeks of age demonstrated mild pathological changes including occasional myonecrosis and central nucleation in FKTN cKO mice. In contrast, in addition to these changes, fibrosis and infiltration of fat were often observed in the skeletal muscle of FKTN cKO/LARGE mice. Abbreviations: GC, gastrocnemius; TA, tibialis anterior; quad, quadriceps; dia, diaphragm. Scale bar represents 50 μm. (D) Quantitative analysis of the gastrocnemius muscle revealed that the fibrosis area was significantly increased in FKTN cKO/LARGE mice (n = 3, for all genotypes). (E) Measurements of muscle fiber diameter demonstrated that diameters of FKTN cKO/LARGE mice muscle cells tended smaller than those of FKTN cKO mice (n = 3, for all genotypes). (F) Counting of central nuclei revealed that the percentage was not significantly altered between FKTN cKO and FKTN cKO/LARGE mice (n = 3, for all genotypes). NS = not statistically significant. *P < 0.05, **P < 0.01 and ***P < 0.001.

Figure 4.

Aggravation of muscular dystrophy in FKTN cKO/LARGE mice. (A) Western blotting analysis showed that LARGE was overexpressed in FKTN cKO/LARGE mice (n = 3 of each type). IIH6 antibody failed to detect glycosylated α-DG in FKTN cKO mice, whereas it revealed intense and broad bands of α-DG in FKTN cKO/LARGE mice. Signals of β-DG and the internal control, α-actinin were unaltered. Laminin blot overlay assay (Lam O/L) demonstrated that binding activity of α-DG was abolished in FKTN cKO mice, whereas markedly increased in FKTN cKO/LARGE mice. (B) Immunofluorescent analysis showed that immunoreactivity of IIH6 was abolished in FKTN cKO mice, whereas it was markedly enhanced in FKTN cKO/LARGE mice. Signals of β-DG, laminin α2, β1 and γ1 chains were unchanged among them. Lam = laminin. Scale bar represents 50 μm. (C) H-E staining of skeletal muscle at 24 weeks of age demonstrated mild pathological changes including occasional myonecrosis and central nucleation in FKTN cKO mice. In contrast, in addition to these changes, fibrosis and infiltration of fat were often observed in the skeletal muscle of FKTN cKO/LARGE mice. Abbreviations: GC, gastrocnemius; TA, tibialis anterior; quad, quadriceps; dia, diaphragm. Scale bar represents 50 μm. (D) Quantitative analysis of the gastrocnemius muscle revealed that the fibrosis area was significantly increased in FKTN cKO/LARGE mice (n = 3, for all genotypes). (E) Measurements of muscle fiber diameter demonstrated that diameters of FKTN cKO/LARGE mice muscle cells tended smaller than those of FKTN cKO mice (n = 3, for all genotypes). (F) Counting of central nuclei revealed that the percentage was not significantly altered between FKTN cKO and FKTN cKO/LARGE mice (n = 3, for all genotypes). NS = not statistically significant. *P < 0.05, **P < 0.01 and ***P < 0.001.

Regeneration is not adequately activated in dystrophic muscles overexpressing LARGE

To further evaluate the muscle regeneration in dy2J/LARGE and FKTN cKO/LARGE mice, we quantitated satellite cells, which are specialized muscle precursor cells that proliferate and fuse into myotubes following muscle damage. Using M-cadherin as a marker, we counted the number of satellite cells located just beneath the basement membrane (Fig. 5A). The satellite cell number in dy2J/LARGE mice tended to decrease as compared with dy2J mice; however, the difference was not statistically significant (Fig. 5B). The satellite cell number in FKTN cKO/LARGE mice was unchanged compared with FKTN cKO mice (Fig. 5C). Because M-cadherin is expressed in both mitotically quiescent satellite cells and actively proliferating myoblasts (44), the number of M-cadherin-positive cells should be increased during muscle regeneration (45). Therefore, these results indicate that the early stage of regeneration is insufficient to compensate for the progression of fibrosis in these mice.

Figure 5.

Suppression of skeletal muscle regeneration in dy2J/LARGE and FKTN cKO/LARGE mice. (A) Satellite cells located just beneath the basement membrane were stained by M-cadherin. Triple immunofluorescent labeling with laminin α2 chain (green), M-cadherin (red) and DAPI (blue) was performed. White arrowheads indicate satellite cells. Scale bar represents 20 μm. (B, C) Satellite cell number was not statistically changed between dy2J and dy2J/LARGE as well as FKTN cKO and FKTN cKO/LARGE mice, although it tended to be decreased in dy2J/LARGE mice compared with dy2J mice (n = 3, for all genotypes). (D) Regenerating fibers were stained by embryonic myosin heavy chain (MyHC). Double immunofluorescent labeling with embryonic MyHC (green) and laminin α2 chain (red) was performed. eMyHC = embryonic myosin heavy chain. Scale bar represents 50 μm. (E, F) The area of regenerating fibers was significantly smaller in dy2J/LARGE mice than in dy2J mice. On the other hand, the area of regenerating fibers was slightly increased in FKTN cKO/LARGE mice (n = 3, for all genotypes). NS = not statistically significant. *P < 0.05 and **P < 0.001.

Figure 5.

Suppression of skeletal muscle regeneration in dy2J/LARGE and FKTN cKO/LARGE mice. (A) Satellite cells located just beneath the basement membrane were stained by M-cadherin. Triple immunofluorescent labeling with laminin α2 chain (green), M-cadherin (red) and DAPI (blue) was performed. White arrowheads indicate satellite cells. Scale bar represents 20 μm. (B, C) Satellite cell number was not statistically changed between dy2J and dy2J/LARGE as well as FKTN cKO and FKTN cKO/LARGE mice, although it tended to be decreased in dy2J/LARGE mice compared with dy2J mice (n = 3, for all genotypes). (D) Regenerating fibers were stained by embryonic myosin heavy chain (MyHC). Double immunofluorescent labeling with embryonic MyHC (green) and laminin α2 chain (red) was performed. eMyHC = embryonic myosin heavy chain. Scale bar represents 50 μm. (E, F) The area of regenerating fibers was significantly smaller in dy2J/LARGE mice than in dy2J mice. On the other hand, the area of regenerating fibers was slightly increased in FKTN cKO/LARGE mice (n = 3, for all genotypes). NS = not statistically significant. *P < 0.05 and **P < 0.001.

Next, to evaluate the late stage of regeneration, we quantified the regenerating myofibers using embryonic myosin heavy chain (MyHC) as a marker (Fig. 5D). The embryonic MyHC-positive fiber area was significantly smaller in dy2J/LARGE than in dy2J mice, revealing that the late stage of regeneration was severely impaired (Fig. 5E), whereas the area was slightly increased in FKTN cKO/LARGE as compared with FKTN cKO mice (Fig. 5F). These results may indicate that under the overexpression of LARGE, the regenerating fibers are recruited to some extent when the dystrophic change is so mild as that found in FKTN cKO mice. However, if the dystrophic change is as severe as that found in dy2J mice and exceeds a certain limit, the overexpression of LARGE renders the damaged skeletal muscle incapable of fully recruiting the necessary regenerating fibers. Together, these results indicate that both the early and late stages of regeneration are suppressed by the overexpression of LARGE.

Myotube formation of C2C12 cell is suppressed by transfection of LARGE

To further clarify the effect of LARGE, we stably transfected C2C12 cells and obtained two clones overexpressing LARGE: clones 1 and 2. On western blotting with antibody against N-terminal polypeptide, LARGE in clone 1 migrated at 75 kDa, whereas in clone 2, it migrated at 90 kDa in both myoblasts and myotubes (Fig. 6A). Using antibody against C-terminus, LARGE in clone 2 was detected at 90 kDa, whereas it was not detected in clone 1 (Fig. 6A). These results indicate that clone 2 expresses full-length LARGE and clone 1 produces C-terminally truncated LARGE proteins, although the reason for this is unclear. Interestingly, IIH6 recognized hyperglycosylated α-DG at 130–250 kDa in clone 2, whereas, in clone 1, IIH6 detected a fainter band at the lower position of 100 kDa than in the control, implying hypoglycosylated α-DG (Fig. 6A). We assume that this might be because the C-terminally truncated LARGE lacks enzymatic activity and competitively inhibits native LARGE. To validate this hypothesis, we generated several constructs of mutant LARGE and transfected them into HEK293 cells (Supplementary Material, Fig. S4A). Thus far, only the three DxD motifs in the catalytic domains 1 and 2 of LARGE were shown indispensable to exert enzymatic activity (46). However, most of our mutants failed to hyperglycosylate α-DG except a chimera in which the N-terminal region of LARGE was replaced with that of POMGnT1 (Supplementary Material, Fig. S4B). These results demonstrate that the entire luminal domain is essential for the function of LARGE and that the C-terminal truncation in clone 1 results in a loss of function.

Figure 6.

Defects in myotube formation in C2C12 cells caused by overexpression of LARGE. (A) C2C12 cells were transfected with LARGE, and two stable transfectants, clones 1 and 2, were obtained. On western blotting analysis with antibody against the N-terminal domain of LARGE (LARGE-N), clone 1 migrated at 75 kDa (white arrow), whereas clone 2 migrated at 90 kDa (black arrow) in both myoblast and myotube. Using antibody against the C-terminus of LARGE (LARGE-C), clone 2 migrated at 90 kDa (black arrow), whereas clone 1 was not detected. The asterisk represents a nonspecific reaction of antibody. IIH6 antibody recognized hyperglycosylated α-DG at 130–250 kDa in clone 2, whereas, in clone 1, IIH6 detected a fainter band at a lower position than in the control, indicating hypoglycosylated α-DG. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. (B) Cell numbers of clones 1 and 2 of the C2C12 myoblast were significantly less than those of the control 2 days after plating, indicating suppressed proliferation. P < 0.001 is represented by ***. (C) C2C12 myoblasts were allowed to form myotubes by switching the growth medium to differentiation medium. Vigorous myotube formation was observed in both control and clone 1; however, smaller myotubes were only occasionally observed in clone 2. Scale bar represents 100 μm. (D) The fusion index of clone 2 was significantly lower than that of the control and clone 1. NS = not statistically significant. P < 0.001 is represented by ***.

Figure 6.

Defects in myotube formation in C2C12 cells caused by overexpression of LARGE. (A) C2C12 cells were transfected with LARGE, and two stable transfectants, clones 1 and 2, were obtained. On western blotting analysis with antibody against the N-terminal domain of LARGE (LARGE-N), clone 1 migrated at 75 kDa (white arrow), whereas clone 2 migrated at 90 kDa (black arrow) in both myoblast and myotube. Using antibody against the C-terminus of LARGE (LARGE-C), clone 2 migrated at 90 kDa (black arrow), whereas clone 1 was not detected. The asterisk represents a nonspecific reaction of antibody. IIH6 antibody recognized hyperglycosylated α-DG at 130–250 kDa in clone 2, whereas, in clone 1, IIH6 detected a fainter band at a lower position than in the control, indicating hypoglycosylated α-DG. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. (B) Cell numbers of clones 1 and 2 of the C2C12 myoblast were significantly less than those of the control 2 days after plating, indicating suppressed proliferation. P < 0.001 is represented by ***. (C) C2C12 myoblasts were allowed to form myotubes by switching the growth medium to differentiation medium. Vigorous myotube formation was observed in both control and clone 1; however, smaller myotubes were only occasionally observed in clone 2. Scale bar represents 100 μm. (D) The fusion index of clone 2 was significantly lower than that of the control and clone 1. NS = not statistically significant. P < 0.001 is represented by ***.

Then, we tested the proliferation of C2C12 myoblasts using these clones. One and two days after plating, control myoblasts expanded rapidly; however, both clones 1 and 2 proliferated very slowly. This indicates that both the hypo- and hyperglycosylation of α-DG lead to defects in proliferation and that adequate glycosylation is necessary (Fig. 6B). Further, we evaluated the differentiation of myoblasts to myotubes. Five days after switching from the growth to differentiation medium, although vigorous myotube formation was observed in the control and clone 1, smaller myotubes were only occasionally seen in clone 2 (Fig. 6C). The fusion index of clone 2 was significantly lower than those in the control and clone 1, indicating that the hyperglycosylation of α-DG by LARGE suppressed the fusion of C2C12 myoblasts (Fig. 6D). Similar results were obtained by transient transfection of C2C12 cells with LARGE (Supplementary Material, Fig. S5).

Expression of IGF-1 is decreased by overexpression of LARGE and its supplementation restores myotube formation

To elucidate the mechanism by which skeletal muscle regeneration as well as myotube formation of C2C12 is suppressed by the overexpression of LARGE, we performed a DNA microarray analysis using control and LARGE Tg mouse skeletal muscle. In 1209 transcripts that exhibited a significant expression change, we further searched regeneration-related genes using gene ontology with keywords, skeletal muscle regeneration, myoblast and satellite cell (Table 1). Among them, of particular interest was the down-regulation of IGF-1, a well-known regulator of the muscle regeneration (47). Thus, we examined the expression of IGF-1 in skeletal muscle by western blotting. As expected, IGF-1 was decreased in both the quadriceps and gastrocnemius muscles of LARGE Tg mice compared with controls (Fig. 7A). Quantitative measurements of IGF-1 by ELISA further revealed a significant reduction in the skeletal muscles of LARGE Tg mice (Fig. 7B). Western blotting of C2C12 myotubes demonstrated that the expression of IGF-1 was decreased in clone 2, which exhibited defects in myoblast fusion, whereas unchanged in clone 1 in which such defects were not observed (Fig. 7C).

Table 1.

Change in gene expression relating to skeletal muscle regeneration, myoblast and satellite cell

Keyword Symbol Genbank accession Description Z-score Ratio 
Skeletal muscle regeneration Igf1 NM_010512 Insulin-like growth factor 1 (Igf1), transcript variant 1 −2.42 0.33 
Igf1 NM_184052 Insulin-like growth factor 1 (Igf1), transcript variant 2 −2.59 0.31 
Myoblast Thbs4 NM_011582 Thrombospondin 4 (Thbs4) 3.43 4.07 
Itgb1bp3 NM_027120 Integrin beta 1 binding protein 3 (Itgb1bp3) 3.23 3.56 
Thbs4 NM_011582 Thrombospondin 4 (Thbs4) 2.75 2.95 
Plg NM_008877 Plasminogen (Plg) 2.23 15.10 
Neo1 AK052439 13-day embryo lung cDNA, RIKEN full-length enriched library, clone:D430023D05 2.02 11.48 
Casp1 NM_009807 Caspase 1 (Casp1) −2.05 0.31 
Igf1 NM_010512 Insulin-like growth factor 1 (Igf1), transcript variant 1 −2.42 0.33 
Igf1 NM_184052 Insulin-like growth factor 1 (Igf1), transcript variant 2 −2.59 0.31 
Btg1 NM_007569 B-cell translocation gene 1, anti-proliferative (Btg1) −2.76 0.33 
Satellite cell Igf1 NM_010512 Insulin-like growth factor 1 (Igf1), transcript variant 1 −2.42 0.33 
Igf1 NM_184052 Insulin-like growth factor 1 (Igf1), transcript variant 2 −2.59 0.31 
Keyword Symbol Genbank accession Description Z-score Ratio 
Skeletal muscle regeneration Igf1 NM_010512 Insulin-like growth factor 1 (Igf1), transcript variant 1 −2.42 0.33 
Igf1 NM_184052 Insulin-like growth factor 1 (Igf1), transcript variant 2 −2.59 0.31 
Myoblast Thbs4 NM_011582 Thrombospondin 4 (Thbs4) 3.43 4.07 
Itgb1bp3 NM_027120 Integrin beta 1 binding protein 3 (Itgb1bp3) 3.23 3.56 
Thbs4 NM_011582 Thrombospondin 4 (Thbs4) 2.75 2.95 
Plg NM_008877 Plasminogen (Plg) 2.23 15.10 
Neo1 AK052439 13-day embryo lung cDNA, RIKEN full-length enriched library, clone:D430023D05 2.02 11.48 
Casp1 NM_009807 Caspase 1 (Casp1) −2.05 0.31 
Igf1 NM_010512 Insulin-like growth factor 1 (Igf1), transcript variant 1 −2.42 0.33 
Igf1 NM_184052 Insulin-like growth factor 1 (Igf1), transcript variant 2 −2.59 0.31 
Btg1 NM_007569 B-cell translocation gene 1, anti-proliferative (Btg1) −2.76 0.33 
Satellite cell Igf1 NM_010512 Insulin-like growth factor 1 (Igf1), transcript variant 1 −2.42 0.33 
Igf1 NM_184052 Insulin-like growth factor 1 (Igf1), transcript variant 2 −2.59 0.31 
Figure 7.

Reduced expression of IGF-1 by overexpression of LARGE and restoration of C2C12 myoblast fusion by IGF-1 supplementation. (A) Western blotting analysis demonstrated that expression of IGF-1 was reduced in both quadriceps and gastrocnemius muscles of LARGE Tg mice as compared with the controls (n = 3 of each strain). Myosin light chain 2 (MLC2) was used as an internal control. (B) Quantitative measurements by ELISA revealed that IGF-1 was significantly decreased in the skeletal muscle of LARGE Tg mice (n = 3, for all genotypes). P < 0.05 is represented by *. (C) Western blotting showed that expression of IGF-1 in C2C12 myotube was reduced in clone 2 but not in clone 1, indicating that the reduction of IGF-1 is associated with hyperglycosylation of α-DG. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. (D) Clone 2 of C2C12 myoblasts was supplemented with IGF-1 and allowed to differentiate into myotubes. Counting nuclei at 6 days after plating revealed that the number of nuclei in the IGF-1-treated clone 2 was slightly increased as compared with the untreated clone, whereas it was significantly less than that of the control. (E) The fusion index of IGF-1-treated clone 2 was markedly increased and exhibited no significant difference from the control. NS = not statistically significant. *P < 0.05, **P < 0.01 and ***P < 0.001.

Figure 7.

Reduced expression of IGF-1 by overexpression of LARGE and restoration of C2C12 myoblast fusion by IGF-1 supplementation. (A) Western blotting analysis demonstrated that expression of IGF-1 was reduced in both quadriceps and gastrocnemius muscles of LARGE Tg mice as compared with the controls (n = 3 of each strain). Myosin light chain 2 (MLC2) was used as an internal control. (B) Quantitative measurements by ELISA revealed that IGF-1 was significantly decreased in the skeletal muscle of LARGE Tg mice (n = 3, for all genotypes). P < 0.05 is represented by *. (C) Western blotting showed that expression of IGF-1 in C2C12 myotube was reduced in clone 2 but not in clone 1, indicating that the reduction of IGF-1 is associated with hyperglycosylation of α-DG. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. (D) Clone 2 of C2C12 myoblasts was supplemented with IGF-1 and allowed to differentiate into myotubes. Counting nuclei at 6 days after plating revealed that the number of nuclei in the IGF-1-treated clone 2 was slightly increased as compared with the untreated clone, whereas it was significantly less than that of the control. (E) The fusion index of IGF-1-treated clone 2 was markedly increased and exhibited no significant difference from the control. NS = not statistically significant. *P < 0.05, **P < 0.01 and ***P < 0.001.

Finally, to examine whether IGF-1 could restore the fusion of myoblasts in clone 2, we added recombinant IGF-1 to the medium. Six days after plating the cells, the number of nuclei in the IGF-1-treated clone 2 cells increased slightly as compared with the untreated clone 2 cells; however, it was still significantly less than that in the controls (Fig. 7D). In contrast, the fusion index of the treated clone 2 cells markedly increased and showed no significant difference from the controls (Fig. 7E). These results indicate that the supplementation of IGF-1 restored the defective myoblast fusion caused by the overexpression of LARGE.

DISCUSSION

The interaction between α-DG and laminin plays a critical role in stabilizing the sarcolemma against muscle contraction. Disruption of this linkage results in a high susceptibility to contraction-induced damage, eventually followed by muscular cell degeneration. This mechanism is hypothesized to underlie the pathogenesis of dystroglycanopathy (12). In the case of laminin α2 chain-deficient MDC1A, the main cause of muscle cell degeneration seems to be a defect in polymerization of laminin-211; however, in some cases where mutations reside in the G domain of laminin α2, the defective interaction of laminin with α-DG may also play an important role (37,39,42). LARGE is a glycosyltransferase that catalyzes addition of the repeating disaccharide -3Xylα1-3GlcUAβ1- to the O-mannosyl glycan of α-DG (10). Because LARGE facilitates the glycosylation of α-DG and highly enhances its laminin-binding activity irrespective of the gene involved (38), the overexpression of LARGE is considered one of the most promising possible therapies for muscular dystrophies such as dystroglycanopathy and MDC1A. In the present study, we generated transgenic mice that overexpress LARGE to evaluate the effects on skeletal muscles. To date, several LARGE transgenic mouse lines have been reported. First, Brockington et al. reported LARGE Tg mouse lines, in which the CAG promoter drove the expression of LARGE, exhibited hyperglycosylation of α-DG in skeletal and cardiac muscles but not in the brain, kidneys, liver or intestines (48). A second mouse line described by Gumerson et al. expressed LARGE under control of the MCK promoter, and α-DG was hyperglycosylated in the skeletal and cardiac muscles (49). In contrast to these transgenic lines, our LARGE Tg mouse exhibits hyperglycosylation of α-DG widely in most tissues, including skeletal and cardiac muscles, the brain, peripheral nerves, kidney and liver. As dystroglycanopathy and MDC1A are multi-organ disorders involving brain, eye and peripheral nerves as well as skeletal and cardiac muscles, our LARGE Tg mouse represents a powerful tool to test the effect of overexpression of LARGE in these tissues.

We crossed the LARGE Tg mice with two distinct mouse models, dy2J and FKTN cKO mice, the former is a model for MDC1A and the latter for FCMD (39,40). Contrary to expectations, the muscular dystrophy was worsened in both mouse models. The features of the worsened phenotype was common to these two lines and characterized by decreased or unchanged central nucleation and reduced diameter of muscle fibers, suggesting the insufficiently activated regeneration. Because fibrosis is a result of muscle fiber necrosis, regeneration should be activated to replace the damaged tissue. However, in both dy2J/LARGE and FKTN cKO/LARGE mice, the central nucleation was not accelerated and the diameter of myofibers remained small, suggesting that the regeneration was insufficiently activated. Very recently, Whitmore et al. have reported that the muscular dystrophy of FKRPMD mouse, a model of another dystroglycanopathy (LGMD2I), was exacerbated when these mice were crossed with their LARGE Tg mice. However, they did not investigate the cause leading to the worsened muscular dystrophy in these mice (50). The present report is the first to describe that the worsened phenotype is related to the defective regeneration of skeletal muscle.

To gain further insight into the regeneration process in dy2J/LARGE and FKTN cKO/LARGE mice, we assessed satellite cells and regenerating fibers in these mice. Satellite cells are muscle stem cells located in a niche on the surface of myofiber beneath the ensheathing basement membrane (51). Upon activation, they rapidly generate myoblasts. After proliferation and migration, the myoblasts further differentiate and eventually fuse together to form myotubes (51,52). We labeled the satellite cells with M-cadherin and the regenerating fibers with embryonic MyHC. Because M-cadherin stains satellite cells and actively proliferating myoblasts, whereas embryonic MyHC stains newly formed myotubes, immunolabeling of the former reflects an early stage and the latter a late stage of regeneration (44,52). However, neither the M-cadherin-positive cells nor the embryonic MyHC-positive myofibers were consistently increased in either the dy2J/LARGE or the FKTN cKO/LARGE mice. These should increase if the regeneration process proceeds properly (44,45). Therefore, our findings provide strong evidence that both the early and the late stage of regeneration are impaired in dy2J/LARGE and FKTN cKO/LARGE mice. The embryonic MyHC-stained area was slightly increased in FKTN cKO/LARGE mice. This may be accounted for by the existence of a limit of regeneration, which was exceeded in dy2J/LARGE mice but not in FKTN cKO/LARGE mice. In a previous study, we injected cardiotoxin into LARGE Tg mice to induce an acute muscle injury and observed a reduction in muscle fiber diameter as compared with controls, suggesting a defect in myoblast fusion (41). Our current data are consistent with this observation and further support the notion that the overexpression of LARGE results in suppression of the skeletal muscle regeneration, regardless of whether the involved injury is acute or chronic.

Prompted by these results, we further examined whether the overexpression of LARGE affects the myotube formation of C2C12 cells, an established in vitro model for the muscle regeneration. In agreement with the data obtained using model mice, the overexpression of LARGE in C2C12 cells led to defects in both proliferation and fusion of myoblasts. The proliferation and fusion of C2C12 cells represent the early and the late stage of the myotube formation, and the overexpression of LARGE affected these both stages. This recapitulates the defect in the muscle regeneration of mice where both the M-cadherin-positive satellite cells (early stage) and the embryonic MyHC-labeled newly fused myotubes (late stage) were affected. Interestingly, the clone 1 of C2C12 cells, in which α-DG was hypoglycosylated, exhibited suppressed proliferation, but normal fusion of myoblasts. In a previous paper, we showed that the coordinated up-regulation of LARGE and the extension of the repeating disaccharides on α-DG are necessary for adequate muscle regeneration (41). In the present study, we provide the first data showing that both the hypo- and hyperglycosylation of α-DG result in defective regeneration, the former by suppressing myoblast proliferation and the latter by reducing proliferation and fusion of myoblasts.

To dissect the mechanism underlying the suppression of regeneration by LARGE, we conducted a DNA microarray analysis and identified a significant reduction in the expression of IGF-1. This reduction was confirmed at the protein level both in LARGE Tg mice and clone 2 of C2C12 cells. IGF-1 is an autocrine/paracrine peptide growth factor with primary roles in promoting myoblast proliferation, differentiation to myofibers and the hypertrophy of skeletal muscles (47). Muscle injury up-regulates the expression of IGF-1 by satellite cells, and after secretion, it binds to the IGF-1 receptors on the muscle cell surface (47). Transgenic overexpression of IGF-1 in mice demonstrated the maintained regeneration efficacy in aged mice and the reduced muscle pathology in dystrophic mice (53–55). Remarkably, the supplementation of IGF-1 fully restored the fusion of myoblasts in C2C12 cells, implying that the impaired fusion of myoblasts was caused, at least partially, via the reduced expression of IGF-1. In the present study, the number of C2C12 nuclei was not restored after the supplementation of IGF-1. The IGF-1 supplementation favors proliferation of myoblasts in some cases and facilitates myotube formation in others, depending on its concentration and the timing of the supplementation (56,57). This biphasic effect of IGF-1 may explain why the number of nuclei was not fully rescued in the C2C12 cells.

The mechanism leading to the reduction of IGF-1 expression by the overexpression of LARGE is still unclear. One possibility is that alteration of the signal transduction through DG might affect the expression of IGF-1. It is known that a tyrosine residue in the C-terminal PPxY motif of β-DG is phosphorylated in an adhesion-dependent manner (58). This phosphorylation regulates the interaction of β-DG with dystrophin, utrophin and SH2 domain-containing signaling proteins (58–60). Moreover, the interaction of laminin with α-DG propagates signals through the PI3K/AKT and Rac1/JNK pathways (61,62). The increased interaction of laminin with α-DG by LARGE may alter these or still unknown signaling pathway and eventually affect the expression of IGF-1. Another possibility is that unidentified proteins other than α-DG might be modified by LARGE and cause the alteration of the IGF-1 expression (63). Finally, it should be considered that factors or processes unrelated to decreased IGF-1 expression might also contribute to the regeneration defect. It has been shown that the interaction of α-DG with laminin inhibits the migration of cultured cells by attenuating the integrin signal that activates the ERK/AKT pathway (64). As myoblasts must migrate to the damaged site during muscle regeneration (51,52), increased glycosylation by LARGE might suppress regeneration by inactivating this pathway. In addition, mechanisms not involving muscle regeneration might also underlie the worsened muscular dystrophy especially in FKTN cKO/LARGE mice, in which the regeneration defect was not so striking as dy2J/LARGE mice.

In conclusion, we generated LARGE Tg mice and crossed them with dy2J and FKTN cKO mice to investigate the effect of overexpression of LARGE. In both resulting strains, the muscular dystrophy was worsened. We identified suppressed muscle regeneration, which at least partially resulted from the reduced IGF-1 expression, as the cause of the deterioration seen in muscular dystrophy. In our study, as well as others (50), the transgenic overexpression of LARGE led to the exacerbation of muscular dystrophy, whereas adeno-associated virus-mediated transfer of LARGE has been reported to ameliorate the phenotype of muscular dystrophy in LARGEmyd and POMGnT1 KO mice (65). Although the reason for this discrepancy remains unclear, the expression level, timing of expression and/or transduced cell type may be different between these two gene delivery systems. In order to develop a therapeutic strategy using the overexpression of LARGE, its adverse effects on skeletal muscle should be carefully investigated.

MATERIALS AND METHODS

Generation of mice

Generation of LARGE Tg mice and MCK-fukutin conditional knockout mice (FKTN cKO mice) were reported previously (40,41). B6.WK-Lama2dy-2J/J (Dy2J) mice were obtained from Jackson laboratory (Bar Harbor, Maine, USA). Mice heterozygous for the Lama2dy-2J mutation were crossed with mice hemizygous for LARGE Tg, and mice heterozygous for Lama2dy-2J carrying LARGE were further crossed with mice heterozygous for Lama2dy-2J to generate mice homozygous for Lama2dy-2J carrying LARGE hemizygously (Dy2J/LARGE). To generate FKTN cKO/LARGE mice, we first crossed heterozygous FKTNlox/+ mice carrying MCK-Cre hemizygously with mice hemizygous for LARGE Tg. Then, FKTNlox/+ mice carrying both MCK-Cre Tg and LARGE Tg were crossed with homozygous FKTNlox/lox mice to obtain FKTNlox/lox mice carrying both MCK-Cre Tg and LARGE Tg hemizygously (FKTN cKO/LARGE). Genotyping of FKTNlox, MCK-Cre Tg and LARGE Tg was performed using PCR. To identify the single base substitution in the Lama2dy-2J allele, we used DNA sequencing. The experiments were approved by the living modified organism safety committee and animal ethics committee of Teikyo University School of Medicine, and the mice were maintained in accordance with the animal care guideline of Teikyo University School of Medicine.

Antibodies

Rabbit polyclonal antibodies against 40 amino acids in the N-terminal domain (a.a. 61–100) and 31 amino acids in the C-terminal domain (a.a. 726–756) of human LARGE were generated (Supplementary Material, Fig. S4A). Goat polyclonal antibody, GT20ADG, against core protein of α-DG (12) was a kind gift from Dr K. P. Campbell (University of Iowa, Iowa City, Iowa, USA). In addition, the following antibodies were used in this study: mouse monoclonal antibody IIH6 against glycosylated α-DG (Millipore), rabbit polyclonal antibody against C-terminal domain of β-DG (Sigma–Aldrich), rabbit polyclonal antibody against EHS-laminin (Sigma–Aldrich), rat monoclonal antibody against laminin α2 chain (Enzo), rat monoclonal antibody against laminin β1 chain (Millipore), rat monoclonal antibody against laminin γ1 chain (Millipore), rabbit polyclonal antibody against M-cadherin (Life Technologies), goat polyclonal antibody against IGF-1 (R&D systems), mouse monoclonal antibody against FLAG (Sigma–Aldrich), mouse monoclonal antibody against α-actinin (Sigma–Aldrich), rabbit polyclonal antibody against GAPDH (Santa Cruz Biotechnology), mouse monoclonal antibody F1.653 against embryonic MyHC (Developmental Studies Hybridoma Bank), mouse monoclonal antibody MF-20 against MyHC (Developmental Studies Hybridoma Bank), Alexafluor 488- and 594-conjugated secondary antibodies (Life Technologies) and horseradish peroxidase-labeled secondary antibodies (GE Healthcare).

Histology, immunofluorescence and electron microscopy

Histological and immunofluorescent microscopic analysis was performed on Dy2J, Dy2J /LARGE, FKTN cKO, FKTN cKO/LARGE and their age-matched control mice between the ages 8 to 27 weeks. Muscles were removed and frozen in liquid nitrogen-cooled isopentane, and cryosections of 8 μm in thickness were prepared. The standard technique was used for hematoxylin–eosin (H-E) staining. For immunofluorescence analysis, the sections were blocked with 5% bovine serum albumin in phosphate-buffered saline (PBS), followed by incubation with primary antibodies for 1 h and then incubated with Alexafluor 488- or 594-conjugated secondary antibody for 1 h. Subsequently, the sections were mounted with Vectashield (Vector Laboratories) and observed under a FSX100 fluorescence microscope (Olympus) or a confocal laser microscope A1 (Nikon). For immunofluorescent analyses of C2C12 cells, myoblasts and myotubes were fixed in 4% paraformaldehyde, permeabilized with 0.2% Triton X-100 and stained with anti-MyHC for 1 h using a mouse on mouse (M.O.M.) kit (Vector Laboratories). After washing with PBS, the cells were incubated for 30 min with Alexafluor 488-conjugated anti-mouse IgG antibody. The slides were mounted with Vectashield with DAPI (Vector Laboratories), and the fluorescent images were taken using a FSX100 fluorescence microscope (Olympus). The basement membranes of the quadriceps muscles were observed with a transmission electron microscopy H-7650 (Hitachi Hightechnologies) using the standard techniques. The in situ ligand overlay assay was described elsewhere (66).

Cell culture and transfection

C2C12 cells were obtained from ATCC and HEK293 cells from the Human Science Research Resource Bank (Osaka, Japan). C2C12 myoblasts were plated at a density of 1.9 × 106/ml on poly-D-lysine/laminin-coated coverslips (Corning) and grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (ATCC), 100 U/ml penicillin G and 100 μg/ml streptomycin (Life Technologies) (growth medium). Next day, cells were induced to differentiate into myotubes by lowering the serum concentration to 2% horse serum (differentiation medium). Five days after replacing the medium, cells were stained for immunofluorescent analysis or harvested for biochemical assay. In some experiments, 10 ng/ml of mouse recombinant IGF-1 (Cell Signaling Technology) was added to both the growth and the differentiation mediums. HEK293 cells were plated on plastic culture dishes (DB Bioscience) and grown in the growth medium described earlier. All cells were grown in a humidified 37°C incubator with 5% CO2 and 95% air. Human LARGE cDNA was obtained from OriGene and cloned into pCMV-FLAG-MAT-Tag-2 (Sigma–Aldrich) for transient transfection of cells. C2C12 and HEK293 cells were transfected using Effectene (Qiagen). For stable transfection of C2C12 cells, the CAG promoter and human LARGE cDNA were cloned into pLenti6/R4R2/V5-DEST using Gateway BP reaction and lentivirus expression vector was generated according to the standard protocol. Stable transfectants were selected in growth medium supplemented with 2.5 μg/ml of Blastcidin S (Life Technologies). Expression vectors for the deletion and chimeric mutants of human LARGE were constructed using the KOD-Plus-Mutagenesis Kit according to the manufacturer's protocols (Toyobo).

Morphometric analysis

For morphometric analysis of skeletal muscles, 5 × 5 stitching images of gastrocnemius muscles, captured at a magnification of 20× using a FSX100 fluorescence microscope (Olympus), were used. Three mice from each genotype were analyzed (n = 3), at ages of 16–22 weeks for dy2J, dy2J/LARGE mice and their controls and 24–27 weeks for FKTN cKO, FKTN cKO/LARGE mice and their controls. For the evaluation of fibrosis, the area stained by anti-mouse IgG, which nonspecifically labels connective tissues and fibrosis, was quantitatively measured by the ImageJ software. For assessment of size variation in muscle fibers, minimal Feret's diameter of individual muscle fiber stained by anti-laminin α2 was measured using ImageJ software. For quantitative evaluation of centrally located nuclei, the total nuclei number was counted using ImageJ software and internal nuclei were assessed by visual inspection of images double-stained by anti-laminin α2 and DAPI. Satellite cells stained by M-cadherin located just beneath the basement membrane labeled with anti-laminin α2 were counted. Regenerating fibers were immunolabeled by embryonic MyHC, and the area was measured using ImageJ software. For morphometric analysis of C2C12 cells, 5 × 5 stitching images of at least five visual fields were captured at a magnification of 20×. Number of DAPI-stained nuclei was counted with ImageJ software, and the fusion index was calculated by dividing the nuclei number in MyHC-stained cells by total nuclei number. The same experiment was repeated three times. Statistical differences were evaluated by t-test, and P-values < 0.05 were considered statistically significant.

Western blotting, blot overlay assay and ELISA

For western blotting and laminin blot overlay assay, tissues were isolated and disrupted with a polytron followed by Dounce homogenization in 50 mm Tris–HCl, pH 7.4, 150 mm NaCl, 0.6 mg/ml pepstatin A, 0.5 mg/ml leupeptin, 0.5 mg/ml aprotinin, 0.75 mm benzamidine and 0.1 mm PMSF. Proteins were then extracted by boiling in sample buffer (65 mm Tris–HCl, pH 7.4, 0.115 m sucrose, 3% SDS, 1% β-mercaptoethanol) for 3 min. After briefly spinning down debris, the homogenate was applied to 4–15% SDS–PAGE. In some experiments, dissected skeletal muscles were homogenized and then incubated with 1% Triton X-100 to solubilize the membrane proteins, and α-DG was enriched by WGA (wheat germ agglutinin) chromatography. Western blotting and laminin blot overlay assay was performed as described previously (12), and images were captured using LAS-3000 software (Fujifilm). For pH 12 extract overlay, skeletal muscles of dy2J and control mice were homogenized, centrifuged at 35 000g for 20 min, and the pellets were incubated at pH 12 for 1 h and centrifuged at 140 000g for 35 min. The supernatant, which included endogenous laminin, was incubated with blots overnight, and the laminin bound to α-DG was detected with anti-laminin antibody. Mouse gastrocnemius muscles and C2C12 myotubes were homogenized as described earlier, centrifuged at 20 000g for 10 min, and the concentration of IGF-1 in the supernatant was measured using a Quantikine ELISA kit for Mouse/Rat IGF-1 (R&D Systems).

DNA microarray analysis

Gastrocnemius muscles of LARGE Tg mouse and control mouse at 16 weeks of age were dissected and stored in RNAlater (Life Technologies). Total RNA was isolated from the muscles using RNeasy Fibrous Tissue Midi Kit (Qiagen). Microarray analysis was performed using Whole Mouse Genome Oligo Microarray 4 × 44 k (Agilent Technologies). Microarray data were extracted from scanned images and analyzed using GeneSpring software (Agilent Technologies). Gene ontology search was performed within the genes that exhibited a significant change in expression, i.e. z-score ≥ 2 and ratio ≥ 1.5 (increase in LARGE Tg) or z-score ≤ −2 and ratio ≤ 0.66 (decrease in LARGE Tg).

Miscellaneous

Body weights of male and female dy2J, dy2J/LARGE and control mice (n = 4–15) were measured at 4 and 8 weeks of age. Grip strength was measured for 10 consecutive trials for dy2J, dy2J/LARGE and control mice (n = 4–6) at the age of 8 weeks using a grip strength meter (Panlab). Kaplan–Meier estimates of survival probabilities were calculated for dy2J and dy2J/LARGE mice (n = 11 and 15, respectively) using survival analysis add-on software to Excel (NAG).

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

FUNDING

This work was supported by an Intramural Research Grant (23-5, 26-8) for Neurological and Psychiatric Disorders of NCNP (the Ministry of Health, Labor and Welfare of Japan) to F.S. and T.T.; and Grant-in-Aid for Scientific Research (C) (23591256, 26461281, 24501357, 25350634, 25430075), (A) (23249049) and Innovative Areas (Deciphering sugar chain-based signals regulating integrative neuronal functions) (24110508) from MEXT (the Ministry of Education, Culture, Sports, Science and Technology of Japan) to F.S., T.S., H.H., K.M, T.T. and M.K., respectively.

ACKNOWLEDGEMENTS

We thank Drs Yutaka Ohsawa and Yoshihide Sunada for their technical advice and fruitful discussion. The antibody against core protein of α-DG was a generous gift from Dr Kevin P. Campbell.

Conflict of Interest statement. None declared.

REFERENCES

1
Ibraghimov-Beskrovnaya
O.
Ervasti
J.M.
Leveille
C.J.
Slaughter
C.A.
Sernett
S.W.
Campbell
K.P.
Primary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix
Nature
 , 
1992
, vol. 
355
 (pg. 
696
-
702
)
2
Sugita
S.
Saito
F.
Tang
J.
Satz
J.
Campbell
K.P.
Südhof
T.C.
A stoichiometric complex of neurexins and dystroglycan in brain
J. Cell Biol.
 , 
2001
, vol. 
154
 (pg. 
435
-
445
)
3
Sato
S.
Omori
Y.
Katoh
K.
Kondo
M.
Kanagawa
M.
Miyata
K.
Funabiki
K.
Koyasu
T.
Kajimura
N.
Miyoshim
T.
, et al.  . 
Pikachurin, a dystroglycan ligand, is essential for photoreceptor ribbon synapse formation
Nat. Neurosci.
 , 
2008
, vol. 
11
 (pg. 
923
-
931
)
4
Ervasti
J.M.
Campbell
K.P.
A role for the dystrophin–glycoprotein complex as a transmembrane linker between laminin and actin
J. Cell. Biol.
 , 
1993
, vol. 
122
 (pg. 
809
-
823
)
5
Jung
D.
Yang
B.
Meyer
J.
Chamberlain
J.S.
Campbell
K.P.
Identification and characterization of the dystrophin anchoring site on β-dystroglycan
J. Biol. Chem.
 , 
1995
, vol. 
270
 (pg. 
27305
-
27310
)
6
Kanagawa
M.
Saito
F.
Kunz
S.
Yoshida-Moriguchi
T.
Barresi
R.
Kobayashi
Y.M.
Muschler
J.
Dumanski
J.P.
Michele
D.E.
Oldstone
M.B.
, et al.  . 
Molecular recognition by LARGE is essential for expression of functional dystroglycan
Cell
 , 
2004
, vol. 
117
 (pg. 
953
-
964
)
7
Saito
F.
Saito-Arai
Y.
Nakamura
A.
Shimizu
T.
Matsumura
K.
Processing and secretion of the N-terminal domain of α-dystroglycan in cell culture media
FEBS Lett.
 , 
2008
, vol. 
582
 (pg. 
439
-
444
)
8
Chiba
A.
Matsumura
K.
Yamada
H.
Inazu
T.
Shimizu
T.
Kusunoki
S.
Kanazawa
I.
Kobata
A.
Endo
T.
Structures of sialylated O-linked oligosaccharides of bovine peripheral nerve α-dystroglycan. The role of a novel O-mannosyl-type oligosaccharide in the binding of α-dystroglycan with laminin
J. Biol. Chem.
 , 
1997
, vol. 
272
 (pg. 
2156
-
2162
)
9
Yoshida-Moriguchi
T.
Yu
L.
Stalnaker
S.H.
Davis
S.
Kunz
S.
Madson
M.
Oldstone
M.B.
Schachter
H.
Wells
L.
Campbell
K.P.
O-mannosyl phosphorylation of α-dystroglycan is required for laminin binding
Science
 , 
2010
, vol. 
327
 (pg. 
88
-
92
)
10
Inamori
K.
Yoshida-Moriguchi
T.
Hara
Y.
Anderson
M.E.
Yu
L.
Campbell
K.P.
Dystroglycan function requires xylosyl- and glucuronyltransferase activities of LARGE
Science
 , 
2012
, vol. 
335
 (pg. 
93
-
96
)
11
Yoshida-Moriguchi
T.
Willer
T.
Anderson
M.E.
Venzke
D.
Whyte
T.
Muntoni
F.
Lee
H.
Nelson
S.F.
Yu
L.
Campbell
K.P.
SGK196 is a glycosylation-specific O-mannose kinase required for dystroglycan function
Science
 , 
2013
, vol. 
341
 (pg. 
896
-
899
)
12
Michele
D.E.
Barresi
R.
Kanagawa
M.
Saito
F.
Cohn
R.D.
Satz
J.S.
Dollar
J.
Nishino
I.
Kelley
R.I.
Somer
H.
, et al.  . 
Post-translational disruption of dystroglycan-ligand interactions in congenital muscular dystrophies
Nature
 , 
2002
, vol. 
418
 (pg. 
417
-
422
)
13
Brown
S.C.
Winder
S.J.
Dystroglycan and dystroglycanopathies: report of the 187th ENMC Workshop 11–13 November 2011, Naarden, The Netherlands
Neuromuscul. Disord.
 , 
2011
, vol. 
22
 (pg. 
659
-
668
)
14
Kobayashi
K.
Nakahori
Y.
Miyake
M.
Matsumura
K.
Kondo-Iida
E.
Nomura
Y.
Segawa
M.
Yoshioka
M.
Saito
K.
Osawa
M.
, et al.  . 
An ancient retrotransposal insertion causes Fukuyama-type congenital muscular dystrophy
Nature
 , 
1998
, vol. 
394
 (pg. 
388
-
392
)
15
Taniguchi-Ikeda
M.
Kobayashi
K.
Kanagawa
M.
Yu
C.C.
Mori
K.
Oda
T.
Kuga
A.
Kurahashi
H.
Akman
H.O.
DiMauro
S.
, et al.  . 
Pathogenic exon-trapping by SVA retrotransposon and rescue in Fukuyama muscular dystrophy
Nature
 , 
2011
, vol. 
478
 (pg. 
127
-
131
)
16
Beltran-Valero
D.B.
Currier
S.
Steinbrecher
A.
Celli
J.
van Beusekom
E.
van der
Z.B.
Kayserili
H.
Merlini
L.
Chitayat
D.
Dobyns
W.B.
, et al.  . 
Mutations in the O-mannosyltransferase gene POMT1 give rise to the severe neuronal migration disorder Walker-Warburg syndrome
Am. J. Hum. Genet.
 , 
2002
, vol. 
71
 (pg. 
1033
-
1043
)
17
Manya
H.
Chiba
A.
Yoshida
A.
Wang
X.
Chiba
Y.
Jigami
Y.
Margolis
R.U.
Endo
T.
Demonstration of mammalian protein O-mannosyltransferase activity: coexpression of POMT1 and POMT2 required for enzymatic activity
Proc. Natl. Acad. Sci. USA
 , 
2004
, vol. 
101
 (pg. 
500
-
505
)
18
van Reeuwijk
J.
Janssen
M.
van den
E.C.
Beltran-Valero de
B.D.
Sabatelli
P.
Merlini
L.
Boon
M.
Scheffer
H.
Brockington
M.
Muntoni
F.
, et al.  . 
POMT2 mutations cause α-dystroglycan hypoglycosylation and Walker-Warburg syndrome
J. Med. Genet.
 , 
2005
, vol. 
42
 (pg. 
907
-
912
)
19
Yoshida
A.
Kobayashi
K.
Manya
H.
Taniguchi
K.
Kano
H.
Mizuno
M.
Inazu
T.
Mitsuhashi
H.
Takahashi
S.
Takeuchi
M.
, et al.  . 
Muscular dystrophy and neuronal migration disorder caused by mutations in a glycosyltransferase, POMGnT1
Dev. Cell
 , 
2001
, vol. 
1
 (pg. 
717
-
724
)
20
Longman
C.
Brockington
M.
Torelli
S.
Jimenez-Mallebrera
C.
Kennedy
C.
Khalil
N.
Feng
L.
Saran
R.K.
Voit
T.
Merlini
L.
, et al.  . 
Mutations in the human LARGE gene cause MDC1D, a novel form of congenital muscular dystrophy with severe mental retardation and abnormal glycosylation of α-dystroglycan
Hum. Mol. Genet.
 , 
2003
, vol. 
12
 (pg. 
2853
-
2861
)
21
Jae
L.T.
Raaben
M.
Riemersma
M.
van Beusekom
E.
Blomen
V.A.
Velds
A.
Kerkhoven
R.M.
Carette
J.E.
Topaloglu
H.
Meinecke
P.
, et al.  . 
Deciphering the glycosylome of dystroglycanopathies using haploid screens for lassa virus entry
Science
 , 
2013
, vol. 
340
 (pg. 
479
-
483
)
22
Manzini
M.C.
Tambunan
D.E.
Hill
R.S.
Yu
T.W.
Maynard
T.M.
Heinzen
E.L.
Shianna
K.V.
Stevens
C.R.
Partlow
J.N.
Barry
B.J.
, et al.  . 
Exome sequencing and functional validation in zebrafish identify GTDC2 mutations as a cause of Walker-Warburg syndrome
Am. J. Hum. Genet.
 , 
2012
, vol. 
91
 (pg. 
541
-
547
)
23
Stevens
E.
Carss
K.J.
Cirak
S.
Foley
A.R.
Torelli
S.
Willer
T.
Tambunan
D.E.
Yau
S.
Brodd
L.
Sewry
C.A.
, et al.  . 
Mutations in B3GALNT2 cause congenital muscular dystrophy and hypoglycosylation of α-dystroglycan
Am. J. Hum. Genet.
 , 
2013
, vol. 
92
 (pg. 
354
-
365
)
24
Lefeber
D.J.
de Brouwer
A.P.
Morava
E.
Riemersma
M.
Schuurs-Hoeijmakers
J.H.
Absmanner
B.
Verrijp
K.
van den Akker
W.M.
Huijben
K.
Steenbergen
G.
, et al.  . 
Autosomal recessive dilated cardiomyopathy due to DOLK mutations results from abnormal dystroglycan O-mannosylation
PLoS. Genet.
 , 
2011
, vol. 
7
 pg. 
e1002427
 
25
Carss
K.J.
Stevens
E.
Foley
A.R.
Cirak
S.
Riemersma
M.
Torelli
S.
Hoischen
A.
Willer
T.
van
S.M.
Moore
S.A.
, et al.  . 
Mutations in GDP-mannose pyrophosphorylase B cause congenital and limb-girdle muscular dystrophies associated with hypoglycosylation of α-dystroglycan
Am. J. Hum. Genet.
 , 
2013
, vol. 
93
 (pg. 
29
-
41
)
26
Yang
A.C.
Ng
B.G.
Moore
S.A.
Rush
J.
Waechter
C.J.
Raymond
K.M.
Willer
T.
Campbell
K.P.
Freeze
H.H.
Mehta
L.
Congenital disorder of glycosylation due to DPM1 mutations presenting with dystroglycanopathy-type congenital muscular dystrophy
Mol. Genet. Metab.
 , 
2013
, vol. 
10
 (pg. 
345
-
351
)
27
Barone
R.
Aiello
C.
Race
V.
Morava
E.
Foulquier
F.
Riemersma
M.
Passarelli
C.
Concolino
D.
Carella
M.
Santorelli
F.
, et al.  . 
DPM2-CDG: a muscular dystrophy-dystroglycanopathy syndrome with severe epilepsy
Ann. Neurol.
 , 
2012
, vol. 
72
 (pg. 
550
-
558
)
28
Lefeber
D.J.
Schonberger
J.
Morava
E.
Guillard
M.
Huyben
K.M.
Verrijp
K.
Grafakou
O.
Evangeliou
A.
Preijers
F.W.
Manta
P.
, et al.  . 
Deficiency of Dol-P-Man synthase subunit DPM3 bridges the congenital disorders of glycosylation with the dystroglycanopathies
Am. J. Hum. Genet.
 , 
2009
, vol. 
85
 (pg. 
76
-
86
)
29
Brockington
M.
Blake
D.J.
Prandini
P.
Brown
S.C.
Torelli
S.
Benson
M.A.
Ponting
C.P.
Estournet
B.
Romero
N.B.
Mercuri
E.
, et al.  . 
Mutations in the fukutin-related protein gene (FKRP) cause a form of congenital muscular dystrophy with secondary laminin α2 deficiency and abnormal glycosylation of α-dystroglycan
Am. J. Hum. Genet.
 , 
2001
, vol. 
69
 (pg. 
1198
-
1209
)
30
Buysse
K.
Riemersma
M.
Powell
G.
van
R.J.
Chitayat
D.
Roscioli
T.
Kamsteeg
E.J.
van den Elzen
C.
van
B.E.
Blaser
S.
, et al.  . 
Missense mutations in β-1,3-N-acetylglucosaminyltransferase 1 (B3GNT1) cause Walker-Warburg syndrome
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
1746
-
1754
)
31
Willer
T.
Lee
H.
Lommel
M.
Yoshida-Moriguchi
T.
de Bernabe
D.B.
Venzke
D.
Cirak
S.
Schachter
H.
Vajsar
J.
Voit
T.
, et al.  . 
ISPD loss-of-function mutations disrupt dystroglycan O-mannosylation and cause Walker-Warburg syndrome
Nat. Genet.
 , 
2012
, vol. 
44
 (pg. 
575
-
580
)
32
Roscioli
T.
Kamsteeg
E.J.
Buysse
K.
Maystadt
I.
van Reeuwijk
J.
van den Elzen
C.
van Beusekom
E.
Riemersma
M.
Pfundt
R.
Vissers
L.E.
, et al.  . 
Mutations in ISPD cause Walker-Warburg syndrome and defective glycosylation of α-dystroglycan
Nat. Genet.
 , 
2012
, vol. 
44
 (pg. 
581
-
585
)
33
Vuillaumier-Barrot
S.
Bouchet-Seraphin
C.
Chelbi
M.
Devisme
L.
Quentin
S.
Gazal
S.
Laquerriere
A.
Fallet-Bianco
C.
Loget
P.
Odent
S.
, et al.  . 
Identification of mutations in TMEM5 and ISPD as a cause of severe cobblestone lissencephaly
Am. J. Hum. Genet.
 , 
2012
, vol. 
91
 (pg. 
1135
-
1143
)
34
Hara
Y.
Balci-Hayta
B.
Yoshida-Moriguchi
T.
Kanagawa
M.
Beltrán-Valero de Bernabé
D.
Gündeşli
H.
Willer
T.
Satz
J.S.
Crawford
R.W.
Burden
S.J.
, et al.  . 
A dystroglycan mutation associated with limb-girdle muscular dystrophy
N. Engl. J. Med.
 , 
2011
, vol. 
364
 (pg. 
939
-
946
)
35
Muntoni
F.
Voit
T.
The congenital muscular dystrophies in 2004: a century of exciting progress
Neuromuscul. Disord.
 , 
2004
, vol. 
14
 (pg. 
635
-
649
)
36
Helbling-Leclerc
A.
Zhang
X.
Topaloglu
H.
Cruaud
C.
Tesson
F.
Weissenbach
J.
Tomé
F.M.
Schwartz
K.
Fardeau
M.
Tryggvason
K.
, et al.  . 
Mutations in the laminin α2-chain gene (LAMA2) cause merosin-deficient congenital muscular dystrophy
Nat. Genet.
 , 
1995
, vol. 
11
 (pg. 
216
-
218
)
37
Geranmayeh
F.
Clement
E.
Feng
L.H.
Sewry
C.
Pagan
J.
Mein
R.
Abbs
S.
Brueton
L.
Childs
A.M.
Jungbluth
H.
, et al.  . 
Genotype-phenotype correlation in a large population of muscular dystrophy patients with LAMA2 mutations
Neuromuscul. Disord.
 , 
2010
, vol. 
20
 (pg. 
241
-
250
)
38
Barresi
R.
Michele
D.E.
Kanagawa
M.
Harper
H.A.
Dovico
S.A.
Satz
J.S.
Moore
S.A.
Zhang
W.
Schachter
H.
Dumanski
J.P.
, et al.  . 
LARGE can functionally bypass α-dystroglycan glycosylation defects in distinct congenital muscular dystrophies
Nat. Med.
 , 
2004
, vol. 
10
 (pg. 
696
-
703
)
39
Sunada
Y.
Bernier
S.M.
Utani
A.
Yamada
Y.
Campbell
K.P.
Identification of a novel mutant transcript of laminin α2 chain gene responsible for muscular dystrophy and dysmyelination in dy2J mice
Hum. Mol. Genet.
 , 
1995
, vol. 
4
 (pg. 
1055
-
1061
)
40
Kanagawa
M.
Yu
C.C.
Ito
C.
Fukada
S.
Hozoji-Inada
M.
Chiyo
T.
Kuga
A.
Matsuo
M.
Sato
K.
Yamaguchi
M.
, et al.  . 
Impaired viability of muscle precursor cells in muscular dystrophy with glycosylation defects and amelioration of its severe phenotype by limited gene expression
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
3003
-
3015
)
41
Goddeeris
M.M.
Wu
B.
Venzke
D.
Yoshida-Moriguchi
T.
Saito
F.
Matsumura
K.
Moore
S.A.
Campbell
K.P.
LARGE glycans on dystroglycan function as a tunable matrix scaffold to prevent dystrophy
Nature
 , 
2013
, vol. 
503
 (pg. 
136
-
140
)
42
Colognato
H.
Yurchenco
P.D.
The laminin α2 expressed by dystrophic dy(2J) mice is defective in its ability to form polymers
Curr. Biol.
 , 
1999
, vol. 
9
 (pg. 
1327
-
1330
)
43
Straub
V.
Rafael
J.A.
Chamberlain
J.S.
Campbell
K.P.
Animal models for muscular dystrophy show different patterns of sarcolemmal disruption
J. Cell. Biol.
 , 
1997
, vol. 
139
 (pg. 
375
-
385
)
44
Irintchev
A.
Zeschnigk
M.
Starzinski-Powitz
A.
Wernig
A.
Expression pattern of M-cadherin in normal, denervated, and regenerating mouse muscles
Dev. Dyn.
 , 
1994
, vol. 
199
 (pg. 
326
-
337
)
45
Kottlors
M.
Kirschner
J.
Elevated satellite cell number in Duchenne muscular dystrophy
Cell Tissue Res.
 , 
2010
, vol. 
340
 (pg. 
541
-
548
)
46
Brockington
M.
Torelli
S.
Prandini
P.
Boito
C.
Dolatshad
N.F.
Longman
C.
Brown
S.C.
Muntoni
F.
Localization and functional analysis of the LARGE family of glycosyltransferases: significance for muscular dystrophy
Hum. Mol. Genet.
 , 
2005
, vol. 
14
 (pg. 
657
-
665
)
47
Clemmons
D.R.
Role of IGF-I in skeletal muscle mass maintenance
Trends Endocrinol. Metab.
 , 
2009
, vol. 
20
 (pg. 
349
-
356
)
48
Brockington
M.
Torelli
S.
Sharp
P.S.
Liu
K.
Cirak
S.
Brown
S.C.
Wells
D.J.
Muntoni
F.
Transgenic overexpression of LARGE induces α-dystroglycan hyperglycosylation in skeletal and cardiac muscle
PLoS One
 , 
2010
, vol. 
5
 pg. 
e14434
 
49
Gumerson
J.D.
Davis
C.S.
Kabaeva
Z.T.
Hayes
J.M.
Brooks
S.V.
Michele
D.E.
Muscle-specific expression of LARGE restores neuromuscular transmission deficits in dystrophic LARGEmyd mice
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
757
-
768
)
50
Whitmore
C.
Fernandez-Fuente
M.
Booler
H.
Parr
C.
Kavishwar
M.
Ashraf
A.
Lacey
E.
Kim
J.
Terry
R.
Ackroyd
M.R.
, et al.  . 
The transgenic expression of LARGE exacerbates the muscle phenotype of dystroglycanopathy mice
Hum. Mol. Genet.
 , 
2014
, vol. 
23
 (pg. 
1842
-
1855
)
51
Relaix
F.
Zammit
P.S.
Satellite cells are essential for skeletal muscle regeneration: the cell on the edge returns centre stage
Development
 , 
2012
, vol. 
139
 (pg. 
2845
-
2856
)
52
Ciciliot
S.
Schiaffino
S.
Regeneration of mammalian skeletal muscle. Basic mechanisms and clinical implications
Curr. Pharm. Des.
 , 
2010
, vol. 
16
 (pg. 
906
-
914
)
53
Musarò
A.
McCullagh
K.
Paul
A.
Houghton
L.
Dobrowolny
G.
Molinaro
M.
Barton
E.R.
Sweeney
H.L.
Rosenthal
N.
Localized Igf-1 transgene expression sustains hypertrophy and regeneration in senescent skeletal muscle
Nat. Genet.
 , 
2001
, vol. 
27
 (pg. 
195
-
200
)
54
Barton
E.R.
Morris
L.
Musaro
A.
Rosenthal
N.
Sweeney
H.L.
Muscle-specific expression of insulin-like growth factor I counters muscle decline in mdx mice
J. Cell. Biol.
 , 
2002
, vol. 
157
 (pg. 
137
-
148
)
55
Kumar
A.
Yamauchi
J.
Girgenrath
T.
Girgenrath
M.
Muscle-specific expression of insulin-like growth factor 1 improves outcome in Lama2Dy-w mice, a model for congenital muscular dystrophy type 1A
Hum. Mol. Genet.
 , 
2011
, vol. 
20
 (pg. 
2333
-
2343
)
56
Coolican
S.A.
Samuel
D.S.
Ewton
D.Z.
McWade
F.J.
Florini
J.R.
The mitogenic and myogenic actions of insulin-like growth factors utilize distinct signaling pathways
J. Biol. Chem.
 , 
1997
, vol. 
272
 (pg. 
6653
-
6662
)
57
Tureckova
J.
Wilson
E.M.
Cappalonga
J.L.
Rotwein
P.
Insulin-like growth factor-mediated muscle differentiation: collaboration between phosphatidylinositol 3-kinase-Akt-signaling pathways and myogenin
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
39264
-
39270
)
58
James
M.
Nuttall
A.
Ilsley
J.L.
Ottersbach
K.
Tinsley
J.M.
Sudol
M.
Winder
S.J.
Adhesion-dependent tyrosine phosphorylation of β-dystroglycan regulates its interaction with utrophin
J. Cell. Sci.
 , 
2000
, vol. 
113
 (pg. 
1717
-
1726
)
59
Ilsley
J.L.
Sudol
M.
Winder
S.J.
The interaction of dystrophin with β-dystroglycan is regulated by tyrosine phosphorylation
Cell Signal.
 , 
2001
, vol. 
13
 (pg. 
625
-
632
)
60
Sotgia
F.
Lee
H.
Bedford
M.T.
Petrucci
T.
Sudol
M.
Lisanti
M.P.
Tyrosine phosphorylation of β-dystroglycan at its WW domain binding motif, PPxY, recruits SH2 domain containing proteins
Biochemistry
 , 
2001
, vol. 
40
 (pg. 
14585
-
14592
)
61
Langenbach
K.J.
Rando
T.A.
Inhibition of dystroglycan binding to laminin disrupts the PI3K/AKT pathway and survival signaling in muscle cells
Muscle Nerve.
 , 
2002
, vol. 
26
 (pg. 
644
-
653
)
62
Zhou
Y.
Jiang
D.
Thomason
D.B.
Jarrett
H.W.
Laminin-induced activation of Rac1 and JNKp46 is initiated by Src family kinases and mimics the effects of skeletal muscle contraction
Biochemistry
 , 
2007
, vol. 
46
 (pg. 
14907
-
14916
)
63
Zhang
P.
Hu
H.
Differential glycosylation of α-dystroglycan and proteins other than α-dystroglycan by like-glycosyltransferase
Glycobiology
 , 
2012
, vol. 
22
 (pg. 
235
-
247
)
64
Bao
X.
Kobayashi
M.
Hatakeyama
S.
Angata
K.
Gullberg
D.
Nakayama
J.
Fukuda
M.N.
Fukuda
M.
Tumor suppressor function of laminin-binding α-dystroglycan requires a distinct β3-N-acetylglucosaminyltransferase
Proc. Natl. Acad. Sci. USA
 , 
2009
, vol. 
106
 (pg. 
12109
-
12114
)
65
Yu
M.
He
Y.
Wang
K.
Zhang
P.
Zhang
S.
Hu
H.
Adeno-associated viral-mediated LARGE gene therapy rescues the muscular dystrophic phenotype in mouse models of dystroglycanopathy
Hum. Gene. Ther.
 , 
2013
, vol. 
24
 (pg. 
317
-
330
)
66
Saito
F.
Masaki
T.
Saito
Y.
Nakamura
A.
Takeda
S.
Shimizu
T.
Toda
T.
Matsumura
K.
Defective peripheral nerve myelination and neuromuscular junction formation in fukutin-deficient chimeric mice
J. Neurochem.
 , 
2007
, vol. 
101
 (pg. 
1712
-
1722
)

Supplementary data