HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 37, 35078-35086, September 14, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/37/35078    most recent M103843200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by McDearmon, E. L. Articles by Ervasti, J. M. Search for Related Content PubMed PubMed Citation Articles by McDearmon, E. L. Articles by Ervasti, J. M. Social Bookmarking                      What's this?

Differential Vicia villosa Agglutinin Reactivity Identifies Three Distinct Dystroglycan Complexes in Skeletal Muscle^*

Erin L. McDearmon, Ariana C. Combs^§, and James M. Ervasti^§¶

From the  Graduate Program in Molecular and Cellular Pharmacology and the ^§ Department of Physiology, University of Wisconsin-Madison Medical School, Madison, Wisconsin 53706

Received for publication, April 30, 2001, and in revised form, July 5, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We present evidence for the expression of three -dystroglycan glycoforms in skeletal muscle cells, including two minor glycoforms marked by either patent or latent reactivity with the N-acetylgalactosamine-specific lectin Vicia villosa agglutinin. Both minor glycoforms co-isolated with -dystroglycan, but not with other dystrophin/utrophin-glycoprotein complex components, suggesting that they may perform distinct or modified cellular functions. We also confirmed that both patent and latent V. villosa agglutinin-reactive -dystroglycan glycoforms are expressed in C2C12 myotubes. However, we found that the combined effect of saturating concentrations of V. villosa agglutinin and laminin-1 were strictly additive with respect to acetylcholine receptor cluster formation in C2C12 myotubes, which suggests that laminin-1 and V. villosa agglutinin do not compete for the same binding site on the cell surface. Finally, although -N-acetylhexosaminidase digestion dramatically inhibited agrin-, V. villosa agglutinin-, and laminin-1-induced acetylcholine receptor clustering in C2C12 myotubes, treatment with this enzyme had no effect on the amount of -dystroglycan that was bound to V. villosa agglutinin-agarose. We conclude that -dystroglycan is not the V. villosa agglutinin receptor implicated in acetylcholine receptor cluster formation. However, our data provide new support for the hypothesis that different glycoforms of -dystroglycan may perform distinct functions even within the same cell.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The dystroglycan complex was originally identified as a component of the skeletal muscle dystrophin-glycoprotein complex, which spans the sarcolemma of muscle cells and physically couples the actin cytoskeleton with the extracellular matrix (1). The dystroglycan complex consists of -dystroglycan, a highly glycosylated extracellular protein that binds to several extracellular ligands, and -dystroglycan, a single-pass transmembrane protein that links cytoplasmic dystrophin with -dystroglycan (2). Both dystroglycan subunits are encoded by a single highly conserved pro-peptide that is proteolytically processed into - and -dystroglycan, which remain stably associated through noncovalent interactions (2). Like muscle deficient in dystrophin or other core components in the dystrophin-glycoprotein complex (1), deficiency of the dystroglycan complex in skeletal muscle results in compromised sarcolemmal integrity (3). Thus, it is generally thought that the dystroglycan complex in skeletal muscle mainly functions to mechanically protect the sarcolemma against shear stresses imposed during muscle contraction. However, the dystroglycan complex has also been linked with more dynamic developmental or pathological processes (2). In skeletal muscle, several studies (3-5) have implicated the dystroglycan complex in either the formation or maintenance of acetylcholine receptor (AChR)¹-rich folds within the motor end plate of the neuromuscular junction (NMJ). However, it is presently unclear how the single, invariant dystroglycan protein may subserve different functions within a single cell type.

Because -dystroglycan displays tissue-specific differences in glycosylation (6), it is possible that distinct -dystroglycan glycoforms may perform different functions. With regard to a role in AChR cluster formation at the NMJ, we previously confirmed (6) that neural -dystroglycans express exposed (patent) terminal -linked N-acetylgalactosamine (GalNAc) residues, which bound to the GalNAc-specific lectin Vicia villosa agglutinin (VVA). Skeletal muscle -dystroglycan displayed no patent binding to VVA but revealed latent VVA reactivity after digestion with neuraminidase (6). Interestingly, the VVA binding properties of neural and skeletal muscle -dystroglycans were consistent with the characteristics of an unidentified VVA receptor previously implicated in AChR clustering (7, 8). Therefore, we hypothesized the expression of two -dystroglycan glycoforms in skeletal muscle cells: 1) an extrasynaptic form with latent VVA reactivity caused by further modification with terminal sialic acid residues and 2) a motor end plate-specific glycoform marked with exposed -linked GalNAc residues that may specifically participate in AChR cluster formation (6).

Here we show that three -dystroglycan glycoforms are expressed in skeletal muscle, including two minor glycoforms marked by either patent or latent VVA reactivity. Both minor glycoforms co-isolated with -dystroglycan but not other dystrophin/utrophin-glycoprotein complex components, suggesting they may perform distinct cellular functions. Although we confirmed that both patent and latent VVA reactive glycoforms are also expressed in C2C12 myotubes, other experiments indicate that neither glycoform is the receptor that mediates VVA-induced AChR clustering. We conclude that the VVA receptor important in AChR cluster formation remains to be identified.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell Culture-- C2C12 and L6 cell lines were obtained from American Type Culture Collection and used for three to seven passages. Proliferating cells were grown in 10-cm dishes or on poly-L-lysine-coated coverslips with Dulbecco's modified Eagle's medium (Cellgrow; Fisher) containing 10% fetal bovine serum (Hyclone, Logan, UT) plus 1% antibiotic/antimycotic (Sigma) at 37 °C in a humid atmosphere of 5-10% CO₂. After reaching confluency, the medium was switched to Dulbecco's modified Eagle's medium containing 2% equine serum (Hyclone) plus 1% antibiotic/antimycotic. The cells were incubated until full differentiation to multi-nucleate myotubes was observed morphologically (with C2C12, 4 days; with L6, 9-11 days) with fresh medium exchanged every 2 days.

Cell and Tissue Solubilization-- Crude surface membranes (CSM) were isolated from rabbit skeletal muscle as described previously (9). For small scale lectin chromatography, 1 mg/ml CSM was solubilized in 1% Triton buffer (50 mM Tris-HCl, pH 7.4, 1 mM CaCl₂, 1 mM MgCl₂, 100 µg/ml benzamidine, 40 µg/ml phenylmethylsulfonyl fluoride, 0.5 µg/ml leupeptin, 0.5 µg/ml aprotinin, 1 mM iodoacetamide, 0.5 µg/ml pepstatin A, 1% Triton X-100) plus 150 mM NaCl, with end-over-end mixing for 1 h at 4 °C. Solubilized proteins were recovered in the supernatant fraction after centrifugation for 30 min at 100,000 × g. For large scale chromatography, 500-600 mg of CSM were solubilized in 1% Triton buffer plus 250 mM NaCl and 0.5 M sucrose and centrifuged as described above.

Fully differentiated C2C12 or L6 myotubes were rinsed twice with 37 °C phosphate-buffered saline and scraped off the dish in 5 ml/dish ice-cold phosphate-buffered saline containing the following protease inhibitors: 100 µg/ml benzamidine, 40 µg/ml phenylmethylsulfonyl fluoride, 0.5 µg/ml leupeptin, 0.5 µg/ml aprotinin, 1 mM iodoacetamide, and 0.5 µg/ml pepstatin A. The cells were pelleted by brief centrifugation at 50-200 × g and solubilized in 0.5 ml/dish 1% Triton buffer containing 150 mM NaCl. The resuspended cell pellet was incubated at 4  °C for 1 h with end-over-end mixing followed by centrifugation as described above.

Lectin Chromatography-- For small scale lectin chromatography, 0.5 ml of Triton solubilate from CSM was incubated overnight at 4 °C with 100 µl of either wheat germ agglutinin (WGA)-agarose or VVA-agarose beads (from Sigma and EY Labs, San Mateo, CA, respectively) that were pre-equilibrated in wash buffer (50 mM Tris-HCl, pH 7.4, 1 mM CaCl₂, 1 mM MgCl₂, 250 mM NaCl, 100 µg/ml benzamidine, 40 µg/ml phenylmethylsulfonyl fluoride, 0.1% Triton X-100). The beads were pelleted, and the void volumes were removed. Lectin-bound proteins were then washed extensively and eluted with the appropriate eluting buffer (for WGA, wash buffer containing 0.3 M N-acetylglucosamine (GlcNAc); for VVA, wash buffer containing 50 mM GalNAc). Small scale chromatography was performed in essentially the same way for Triton-solubilized C2C12 or L6 myotubes except that the wash buffer contained 150 mM NaCl and an extra 100 mM NaCl was added to the VVA eluting buffer. For large scale chromatography, ~100 ml of CSM solubilate was loaded onto a 3-ml VVA-agarose column that was pre-equilibrated with wash buffer containing 250 mM NaCl. The column was washed and eluted with wash buffer containing 50 mM GalNAc. The eluates were concentrated by methanol precipitation.

Enzyme Treatments-- CSM solubilates were incubated with or without 0.1 unit/ml of Clostridium perfringens neuraminidase (Roche Molecular Biochemicals) for 2 h at 37 °C and used in small scale VVA chromatography as described above. Fully differentiated C2C12 or L6 myotubes were incubated with or without 0.1 unit/ml neuraminidase diluted in medium for 2 h at 37 °C. C2C12 myotubes were also treated for 6 h at 37 °C with 1 unit/ml -N-acetylhexosaminidase from beef kidney (Roche Molecular Biochemicals), 1 unit/ml -N-acetylgalactosaminidase from chicken liver (Sigma), or 0.05 unit/ml -N-acetylgalactosaminidase from Acremonium sp. (Calbiochem, San Diego, CA), each diluted in medium. Following incubation, the cells were analyzed for AChR clustering as described below or rinsed and solubilized for lectin chromatography as described above.

SDS-PAGE and Western Blotting-- The samples were separated by SDS-PAGE under reducing or nonreducing conditions and transferred to nitrocellulose as described previously (10). Western blotting was performed with the following antibodies: monoclonal antibody IIH6 to -dystroglycan (10), affinity purified chicken polyclonal antibody against -dystroglycan (6), monoclonal antibody XIXC2 to dystrophin (11), monoclonal antibody SYN1351 to syntrophin (12), anti-dystrobrevin monoclonal antibody 13H1 (13), monoclonal antibody NCL--DG to -dystroglycan, monoclonal antibody NCL--SG to -sarcoglycan (both from Vector Laboratories, Burlingame, CA), Rb 1715 antiserum to -sarcoglycan (14), Rb 56 antiserum to utrophin (15), 7CDB2 antiserum to 7B integrin cytoplasmic domain (16), anti-₁ integrin monoclonal antibody MAB1997 (Chemicon, Temecula, CA), anti-AChR  subunit monoclonal antibody (BD Transduction Laboratories, Franklin Lakes, NJ), and monoclonal antibody HNK-1 (Becton Dickinson, Bedford, MA).

Analysis of AChR Clustering-- To assay for AChR clustering, C2C12 and L6 myotubes were grown on poly-L-lysine-coated glass coverslips. After treatment with laminin-1 (the gift of Dr. Hynda Kleinman, NIDR), VVA lectin (Sigma or EY Labs), C-Ag4,8 conditioned medium (17) (the gift of Drs. Mark Bowe and Justin Fallon, Brown University), and/or glycosidases, the myotubes were washed with 37 °C phosphate-buffered saline and incubated for 1 h at 37 °C with 500 ng/ml Alexa 488-conjugated bungarotoxin (Molecular Probes, Eugene, OR) diluted in Dulbecco's modified Eagle's medium. The cells were washed and fixed in 4% paraformaldehyde for 10 min at room temperature. Fixed cells were rinsed in phosphate-buffered saline and mounted on slides for immunofluorescence analysis. Images were collected with a Bio-Rad MRC 1000 confocal microscope (Keck Center for Biological Imaging) using a 40× oil immersion objective. AChR clusters were counted from five randomly chosen fields for each experiment.

Immunocytochemistry-- 7-µm cross-sections of control and mdx mouse hind limb muscle were adhered to Plus slides (Fisher), fixed in 4% paraformaldehyde for 10 min at room temperature, rinsed, and blocked in 1% bovine serum albumin for 2 h at room temperature. Blocked sections were rinsed and incubated either with Rab 56 antiserum to utrophin for 2 h at 37 °C followed by 20 µg/ml Alexa 568-conjugated rabbit secondary antibody (Molecular Probes) for 1 h at 37 °C or with 500 ng/ml rhodamine-conjugated bungarotoxin (Molecular Probes) for 1 h at 37 °C. All sections were double-labeled with 50 µg/ml VVA-fluorescein isothiocyanate lectin (ICN, Costa Mesa, CA), which was included during the incubation with secondary antibodies or bungarotoxin. The images were collected as described above.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Identification of a VVA-reactive Dystroglycan Complex in Skeletal Muscle-- To determine whether a VVA-reactive glycoform of -dystroglycan is expressed in adult skeletal muscle, we attempted to isolate -dystroglycan from detergent-solubilized rabbit skeletal muscle CSM by small scale VVA-agarose chromatography. The chromatography protocol was first tested with WGA-agarose, a lectin that has previously been shown to bind -dystroglycan isolated from a wide variety of tissues and species, including rabbit skeletal muscle (11). We observed that all -dystroglycan in 1 mg of solubilized CSM protein quantitatively bound to WGA-agarose and was eluted with buffer containing 0.3 M GlcNAc (Fig. 1a). In addition, several other dystrophin-glycoprotein complex constituents were observed to co-purify with -dystroglycan in the WGA eluate (Fig. 1a). However, no -dystroglycan was detected in the GalNAc eluate from VVA-agarose using the small scale chromatography protocol (Fig. 1b). Because the motor end plate makes up less than 0.1% of the entire muscle cell plasma membrane (18), it seemed probable that a VVA-reactive species of -dystroglycan might be present in such low abundance that we could not detect it by small scale chromatography. Therefore, we performed a large scale VVA chromatography protocol in which 500-600 mg of CSM solubilate was loaded onto a 3-ml VVA-agarose column. The GalNAc eluate from the 3-ml VVA-agarose column was then concentrated 300-400-fold, relative to the CSM solubilate. On this scale of amplification, we readily detected two distinct species of -dystroglycan in the concentrated VVA eluate with molecular weights of ~156,000 and 110,000 (Fig. 2). Both proteins were reactive with the -dystroglycan-specific monoclonal antibody IIH6 and an affinity-purified polyclonal antibody to -dystroglycan (Fig. 2).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Small scale WGA and VVA chromatography of skeletal muscle membranes. In a, 1 mg of Triton-solubilized rabbit skeletal muscle CSM was incubated overnight with 0.1 ml of WGA-agarose beads. The beads were washed and eluted with an equivalent volume of wash buffer containing 300 mM GlcNAc. The samples were separated by SDS-PAGE, transferred to nitrocellulose, and stained with antibodies that recognize the following proteins: -dystroglycan (IIH6), -dystroglycan (-DG), dystrophin (DYS), -sarcoglycan (-SG), syntrophin (Syn), and dystrobrevins 1,2 (Db). In b, 1 mg of Triton-solubilized rabbit skeletal muscle CSM was incubated overnight with 0.1 ml of VVA-agarose beads. The beads were washed and eluted with buffer containing 50 mM GalNAc. The samples were separated by SDS-PAGE, transferred to nitrocellulose, and stained with monoclonal antibody IIH6 against -dystroglycan.

View larger version (37K): [in this window] [in a new window]   Fig. 2.   Amplification of distinct -dystroglycan glycoforms from skeletal muscle membranes by large scale VVA chromatography. 600 mg of Triton-solubilized rabbit skeletal muscle CSM were circulated overnight on a 3-ml VVA-agarose column. After washing, the column was eluted with buffer containing 50 mM GalNAc, and the eluates were concentrated by methanol precipitation. The samples were separated by SDS-PAGE, transferred to nitrocellulose, and stained with antibodies that recognize the following proteins: -dystroglycan (-DG polyclonal antibody (pAb) and IIH6), -dystrobrevins-1 and -2 (Db), -dystroglycan (-DG), dystrophin (DYS), utrophin (UTR), -sarcoglycan (-SG), -sarcoglycan (-SG), and syntrophin (Syn). Note that the eluates were concentrated ~300-fold relative to the CSM solubilate (SUP).

One or both of the VVA-reactive -dystroglycan species could have originated from small amounts of blood vessel or peripheral nerve tissues that may contaminate the muscle membrane preparations used. Indeed, we observed that the 110-kDa species, but not the 156-kDa species, reacted with a monoclonal antibody to the HNK-1 epitope (data not shown). Because HNK-1 antibodies have been observed to react with peripheral nerve but not skeletal muscle -dystroglycan (19)² and because peripheral nerve -dystroglycan has a molecular weight of ~120,000 (19), we concluded that the 110-kDa -dystroglycan species in the large scale VVA eluate likely originated from peripheral nerve tissue. However, -dystroglycan from visceral and lung smooth muscle is not reactive with monoclonal antibody IIH6 (20, 21). In addition, -sarcoglycan, a component of the smooth muscle dystrophin-glycoprotein complex (20), did not co-purify with -dystroglycan in the large scale VVA eluates (Fig. 2). These results suggested that the 156-kDa VVA-reactive -dystroglycan species likely originated from skeletal muscle cells. Based on the densitometric intensities of IIH6 staining in the solubilate and VVA eluate and taking into account the difference in sample load, we estimated that the 156-kDa VVA-reactive species comprised ~0.1% of total skeletal muscle -dystroglycan (Table I).

View this table: [in this window] [in a new window]   Table I -Dystroglycan glycoforms in skeletal muscle cells

We next examined the large scale VVA eluates for co-purification of other components in the dystrophin-glycoprotein complex. As expected, -dystroglycan was observed to co-purify with VVA-reactive -dystroglycan (Fig. 2). However, other components of the dystrophin-glycoprotein complex were not detected in the large scale VVA eluates (Fig. 2). Of particular relevance, we did not detect the co-purification of utrophin with VVA-reactive -dystroglycan, as might be expected if VVA-reactive -dystroglycan was expressed at the neuromuscular junction (15). Although it is possible that Triton X-100 may have dissociated VVA-reactive dystroglycan complex from other dystrophin-associated proteins (22), we observed the same results when large scale VVA chromatography was performed with CSM solubilized in 1% digitonin (data not shown). We also observed the same results when large scale VVA chromatography was performed using Triton-solubilized rat skeletal muscle (data not shown). The presence of AChRs in both solubilized CSM and rat skeletal muscle was verified by Western blotting with an antibody against AChR -subunit (data not shown). These results indicate that a small amount of VVA-reactive -dystroglycan is expressed in skeletal muscle cells in association with -dystroglycan but is not stably associated with other proteins in the dystrophin-utrophin glycoprotein complexes.

VVA Receptors at the NMJ Are Not Associated with Utrophin-- The dystrophin homologue utrophin is normally restricted to the adult NMJ (15). Immunofluorescence analysis of skeletal muscle from the dystrophin-deficient mdx mouse previously revealed an up-regulation in utrophin expression and its redistribution to the extrasynaptic sarcolemma normally occupied by dystrophin (23, 24). More recently, proteins stably associated with utrophin in biochemical assays were also shown to redistribute with utrophin to the extrasynaptic sarcolemma of mdx muscle (25, 26). To determine whether VVA receptors in skeletal muscle associate with utrophin in vivo, we double-labeled control and mdx mouse skeletal muscle cross-sections with VVA lectin and utrophin antibody (Fig. 3). From analysis of several fields with identifiable neuromuscular junctions (n = 7), we observed that VVA receptors do not redistribute with utrophin in mdx mouse skeletal muscle. In addition, bungarotoxin-labeled AChRs remained strictly localized at the neuromuscular junctions in mdx mouse skeletal muscle (data not shown and Refs. 27 and 28). The results of Figs. 2 and 3 suggested that neither VVA receptors at the neuromuscular junction nor VVA-reactive -dystroglycan are specifically associated with utrophin. Because AChRs and residual dystroglycan are both retained at the NMJ of dystrophin/utrophin double knock-out mice (29, 30), it remained possible that the small amount of VVA-reactive -dystroglycan present in skeletal muscle may participate in AChR clustering independent of an interaction with dystrophin or utrophin.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   VVA-reactive glycoconjugates localize to the neuromuscular junction but do not redistribute with utrophin in mdx mouse skeletal muscle. 7-µm cross-sections of control or mdx mouse skeletal muscle were fixed and double-labeled with VVA-fluorescein isothiocyanate and Rab 56 antiserum against utrophin, followed by Alexa 568-conjugated rabbit secondary antibody. The images were collected with a Bio-Rad MRC 1000 confocal microscope using a 40× oil immersion objective. Scale bar, 50 µm.

Distinct -Dystroglycan Glycoforms in Cultured Myotubes-- -Dystroglycan in C2C12 myotubes was recently shown to bind to VVA-agarose (31). However, it remained to be determined what fraction of total C2C12 -dystroglycan was patently reactive with VVA and whether VVA-reactive -dystroglycan may also be expressed in other clonal myotube cell lines. Therefore, we analyzed both C2C12 and rat L6 myotubes for expression of distinct -dystroglycan glycoforms. Both C2C12 and L6 myotubes were shown to express -dystroglycan with similar average molecular weights of 134,000 and 136,000, respectively. Because the polypeptide core of -dystroglycan is 67,000, the high molecular weights of C2C12 and L6 -dystroglycan suggested that these cell lines express glycosylated -dystroglycan. In addition, -dystroglycan expressed in both cell lines reacted with IIH6 (Fig. 4a), which has been shown to recognize carbohydrate residues on -dystroglycan (32). Finally, both C2C12 and L6 myotubes expressed -dystroglycan that bound WGA-agarose (Fig. 4a). These results suggested that -dystroglycan in C2C12 and L6 myotubes is glycosylated in a manner similar to that observed for skeletal muscle CSM -dystroglycan (Fig. 1). Small scale VVA chromatography indicated that ~80% of C2C12 myotube -dystroglycan could bind to VVA-agarose (Fig. 4a). -Dystroglycan also co-purified with VVA-reactive -dystroglycan from C2C12 myotubes, whereas other components of the dystrophin-glycoprotein complex were only detected in the VVA void (Fig. 4b). To determine whether the small fraction of -dystroglycan left in the VVA void (Fig. 4a) was due to saturation of the lectin beads, we incubated the VVA void with a fresh aliquot of VVA-agarose overnight but did not detect any additional binding of -dystroglycan to VVA (data not shown). In contrast to C2C12 myotube -dystroglycan, L6 myotube -dystroglycan showed no detectable binding to VVA-agarose (Fig. 4a). These data confirm that C2C12 myotubes express two distinct -dystroglycan glycoforms, one of which predominates and is marked by its patent reactivity with VVA-agarose. In addition, these results indicate that skeletal muscle cells are capable of expressing three distinct glycoforms of -dystroglycan.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   VVA-reactive -dystroglycan is expressed in C2C12 but not in L6 skeletal muscle myotubes. In a, Triton-solubilized extracts from C2C12 or L6 myotubes were incubated with WGA- or VVA-agarose beads. The beads were washed and eluted with three aliquots of buffer containing 0.3 M GlcNAc or 50 mM GalNAc, respectively. The samples were separated by SDS-PAGE and transferred to nitrocellulose. -Dystroglycan was detected with monoclonal antibody IIH6. In b, equal volumes of Triton-solubilized C2C12 myotubes (TOT), VVA flow-through (VOID), and GalNAc eluate (ELUTE) were assessed for co-purification of dystrophin-associated proteins by Western blot analysis. The molecular weight standards (× 103) are indicated on the left.

Additive AChR Clustering by Laminin-1 and VVA-- Laminin-1 and VVA can each induce AChR clustering in C2C12 myotubes in a manner that is independent of and additive to the well established agrin/MuSK pathway (31, 33, 34) but still linked to agrin/MuSk through dependence on rapsyn to effect clustering (35). Because laminin-1 appears to induce AChR clustering mainly through binding to -dystroglycan (4, 5) and the majority of C2C12 -dystroglycan could also bind VVA-agarose (Fig. 4a), we considered that -dystroglycan may be the cell surface receptor through which laminin-1 and VVA both induce AChR clustering. To test this hypothesis, we compared AChR cluster formation in C2C12 myotubes treated with 120 nM laminin-1, 50 µg/ml VVA, or both laminin-1 and VVA (Fig. 5). Consistent with previous studies (8, 33), these concentrations of laminin-1 and VVA were found to maximally stimulate AChR clustering when each ligand was tested individually over a wide range of concentrations (not shown). We observed that laminin-1 and VVA each significantly increased the number of AChR clusters per field compared with untreated control myotubes (Fig. 5). However, the combined effects of laminin-1 and VVA on AChR cluster frequency were strictly additive (Fig. 5), which suggests that laminin-1 and VVA do not compete for the same cell surface receptor.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Additive AChR clustering by laminin-1 and VVA. a, C2C12 myotubes were incubated with 50 µg/ml VVA, 120 nM laminin-1 (+Lam), or both VVA and laminin-1 for 18 h, and AChR clusters were detected with Alexa 488-labeled bungarotoxin as described under "Materials and Methods." In b, the data represent the average number of AChR clusters (± S.E.) measured in five random fields from each of three independent experiments. AChR clustering induced by VVA (p < 0.05), laminin-1 (p < 0.001), or VVA plus laminin-1 (p < 0.001) were all significantly greater than control as assessed by one-way analysis of variance with Tukey's multiple comparison test. However, the sum of AChR clustering induced by VVA or laminin-1 alone (14 ± 0.4) was not significantly different from clustering induced by simultaneous treatment with VVA and laminin-1 (19 ± 4.0) as determined by Student's t test (p > 0.24).

Neuraminidase Stimulates AChR Clustering and -Dystroglycan VVA Reactivity in C2C12 but Not L6 Myotubes-- Treatment of C2C12 myotubes with neuraminidase has been shown to dramatically increase cell surface labeling by VVA (8), stimulate AChR clustering in the absence of agrin (8, 36), and increase the binding of C2C12 -dystroglycan to VVA-agarose (31). Furthermore, we previously demonstrated that purified skeletal muscle -dystroglycan becomes VVA-reactive after treatment with neuraminidase (6). Therefore, we examined the effect of neuraminidase treatment on C2C12 and L6 myotube -dystroglycan binding to VVA-agarose. We first confirmed that neuraminidase treatment could stimulate AChR clustering in C2C12 myotubes. A 2-h incubation of C2C12 myotubes with 0.1 unit/ml neuraminidase resulted in a 5.9-fold increase in AChR clusters/field (Fig. 6a), similar to previous results (8, 36). In small scale chromatography experiments, we observed that 100% of -dystroglycan in neuraminidase-treated C2C12 myotubes bound to VVA-agarose as compared with the 80% of -dystroglycan bound to VVA from untreated C2C12 myotubes (Fig. 6b). These data demonstrate that a small fraction (20%) of C2C12 -dystroglycan expresses latent VVA epitopes that can be exposed by digestion with neuraminidase.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Neuraminidase treatment stimulates AChR clustering in C2C12 myotubes and identifies a subpopulation of -dystroglycan with latent VVA reactivity. C2C12 myotubes were incubated in the presence (+C.N.) or absence (Control) of 0.1 unit/ml C. perfringens neuraminidase for 2 h at 37 °C, and AChR clusters were detected with Alexa 488-conjugated bungarotoxin as described under "Materials and Methods." The graph in a indicates the average number of AChR clusters (± S.E.) measured in five random fields from each of three independent experiments. AChR clustering was significantly greater in neuraminidase-treated cultures (p < 0.0001) as assessed by Student's t test. In b, C2C12 or L6 myotubes were incubated with (+CN) or without (CON) 0.1 unit/ml C. perfringens neuraminidase at 37 (0)C for 2 h and then solubilized. Rabbit skeletal CSM were first solubilized and then incubated with or without 0.1 unit/ml neuraminidase at 37 °C for 2 h. Small scale VVA-agarose chromatography was then performed as described under "Materials and Methods." Equivalent volumes of solubilates (SUP), VVA flow-through (VOID), and GalNAc eluates (ELUTE) were separated by SDS-PAGE, transferred to nitrocellulose, and stained with monoclonal antibody IIH6 against -dystroglycan.

Although L6 myotubes express AChRs (37), they exhibit greatly reduced spontaneous AChR clustering activity compared with C2C12 myotubes (38). We hypothesized that the low spontaneous AChR clustering activity of L6 might in part be due to sialic acid capping of VVA-reactive epitopes on L6 myotube -dystroglycan. However, we observed no AChR clustering in L6 myotubes treated with 0.1 unit/ml neuraminidase (not shown). Furthermore, we performed small scale VVA chromatography using solubilates from control or neuraminidase-treated L6 myotubes and again observed no L6 -dystroglycan binding to VVA-agarose (Fig. 6b). Neuraminidase treatment did cause an ~10-kDa decrease in the apparent molecular mass of L6 -dystroglycan, indicating that -dystroglycan was predominantly exposed on the surface of L6 myotubes and was modified with sialic acid residues (Fig. 6b). These results indicate that L6 myotube -dystroglycan lacks both patent and latent reactivity with VVA.

Evidence for Three -Dystroglycan Glycoforms in Skeletal Muscle Cells-- Our lectin chromatography analysis of detergent-solubilized C2C12 and L6 myotubes suggested that skeletal muscle cells can express three distinct glycoforms of -dystroglycan. L6 myotubes appeared to express a single -dystroglycan glycoform that was not reactive with VVA (Figs. 4 and 6 and Table I). C2C12 myotubes appeared to express two distinct glycoforms: a patent form that is endogenously reactive with VVA (Figs. 4 and 6 and Table I) and a latent form that became VVA-reactive following digestion with neuraminidase (Fig. 6 and Table I). Because these data were obtained from two different clonal cell lines, we sought to confirm the possible co-expression of these three -dystroglycan glycoforms and determine their relative abundance in adult skeletal muscle. Therefore, we performed small scale VVA-agarose chromatography using detergent-solubilized rabbit CSM that was either untreated or digested with neuraminidase (Fig. 6b). Consistent with the results presented in Fig. 1b, no - or -dystroglycan was recovered in the GalNAc eluates of untreated CSM solubilates. In contrast, we readily detected a small fraction of - and -dystroglycan binding to VVA-agarose following neuraminidase treatment (Fig. 6b). Based on densitometric intensities of IIH6 immune signal in the CSM solubilate and VVA eluate (Fig. 6b), we estimated that the latent VVA-reactive species comprised 1.7% of total CSM -dystroglycan (Table I). Western blot analysis indicated that both dystrophin and utrophin were absent from the GalNAc eluate containing -dystroglycan after neuraminidase digestion (data not shown). We conclude that at least three distinct glycoforms of -dystroglycan are expressed in cultured myotubes and adult skeletal muscle. In adult skeletal muscle, the predominant glycoform was not reactive with VVA and was stably associated with the canonical dystrophin-glycoprotein complex of skeletal muscle (Fig. 1). The two additional glycoforms were present in substantially lower amounts (Figs. 2 and 6b), exhibited no stable interaction with other dystrophin-associated proteins besides -dystroglycan (Fig. 2), and were distinguishable by either patent (Fig. 2) or latent (Fig. 6) VVA reactivity.

-N-Acetylhexosaminidase Treatment Inhibits Agrin-, Laminin-, and VVA-induced Clustering but Has No Effect on the VVA Reactivity of C2C12 -Dystroglycan-- The exoglycosidase -N-acetylhexosaminidase was previously shown to cause a dramatic decrease in cell surface labeling by VVA and also to completely inhibit agrin-induced AChR clustering in C2C12 myotubes (8). Therefore, we made use of -N-acetylhexosaminidase as a tool to further assess the putative role of -dystroglycan in VVA-induced AChR clustering. First, we confirmed that treatment of C2C12 myotubes with -N-acetylhexosaminidase completely blocked agrin-induced AChR clustering (Fig. 7a). Second, we found that -N-acetylhexosaminidase treatment also completely inhibited laminin-induced AChR clustering (Fig. 7b), whereas treatment of C2C12 myotubes with two different sources of -N-acetylgalactosaminidase did not inhibit laminin-1-induced AChR clustering (not shown). Third, VVA-induced AChR clustering was completely inhibited in C2C12 myotubes treated with -N-acetylhexosaminidase (Fig. 7c). Taken together, these data indicate that agrin-, laminin-1-, and VVA-mediated AChR clustering pathways are mutually dependent on one or more -GalNAc-modified glycoconjugates that are sensitive to digestion with -N-acetylhexosaminidase. However, -N-acetylhexosaminidase treatment of C2C12 myotubes had no discernable effect on the amount of -dystroglycan that was bound to VVA-agarose or eluted with GalNAc (Fig. 7d). Thus, although -N-acetylhexosaminidase digestion clearly inhibited agrin-, laminin-1-, and VVA-induced AChR clustering in C2C12 myotubes, the enzyme did not appear to inhibit clustering through a direct effect on -dystroglycan glycosylation.

View larger version (21K): [in this window] [in a new window]   Fig. 7.   -N-Acetylhexosaminidase treatment blocks agrin-, laminin-1-, and VVA-induced AChR clustering in C2C12 myotubes but does not inhibit -dystroglycan binding to VVA-agarose. In a, C2C12 myotubes were either left untreated (Control) or treated with a maximally stimulating dilution of C-Ag4,8-conditioned medium (Agrin), 1 unit/ml -N-acetylhexosaminidase (HEXase), or both (HEXase + Agrin) for 6 h at 37 °C. In b, C2C12 myotubes were either left untreated (Control) or treated with 120 nM laminin (Lam), 1 unit/ml hexosaminidase (HEXase), or both (HEXase + Lam) for 6 h at 37 °C. In c, C2C12 myotubes were either left untreated (Control) or treated with 50 µg/ml VVA (VVA), 1 unit/ml hexosaminidase (HEXase), or both (HEXase + VVA) for 6 h at 37 °C. The graphs in a-c represent the average number of AChR clusters measured in five random fields from each of two independent experiments. In d, control (CON) or hexosaminidase-treated (+HEXase) C2C12 myotubes were solubilized and subjected to VVA small scale chromatography as described under "Materials and Methods." Western blot analysis was performed on equivalent volumes of solubilates (SUP), VVA flow-through (VOID), or GalNAc eluates (ELUTE) using IIH6 monoclonal antibody (DG), antiserum specific for the 7B integrin isoform (7B), or a monoclonal antibody specific for 1 integrin (1).

₇₁ Integrin Is Not a VVA Receptor-- ₇₁ integrin has also been implicated as a receptor that may also mediate laminin-induced AChR clustering (39, 40). Therefore, we examined whether ₇₁ integrin in C2C12 myotubes could bind to VVA-agarose and, if so, whether the binding of this integrin to VVA-agarose was sensitive to pretreatment of myotubes with -N-acetylhexosaminidase. However, Western blot analysis of GalNAc eluates from VVA-agarose with antibodies specific for ₇ or ₁ integrin subunits demonstrated that neither protein bound to VVA-agarose (Fig. 7d). In addition, neither ₇ nor ₁ integrin subunits from C2C12 myotubes became VVA-reactive when C2C12 myotubes were pretreated with neuraminidase (data not shown). Thus, we conclude that neither -dystroglycan nor ₇₁ integrin are solely responsible for mediating VVA-induced AChR clustering.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We have presented evidence indicating that three distinct glycoforms of -dystroglycan are expressed in adult skeletal muscle, including two minor species that display either patent or latent reactivity with the GalNAc-specific plant lectin VVA. These VVA-reactive -dystroglycan glycoforms differ from the predominant -dystroglycan of skeletal muscle in two respects. Besides displaying no reactivity with VVA (Figs. 1 and 6b), the predominant -dystroglycan glycoform also exhibited a biochemically stable association with several other components in the dystrophin-glycoprotein complex (Fig. 1). As noted in the introduction, evidence from a variety of studies indicates that the dystrophin-glycoprotein complex performs a structural role in stabilizing the sarcolemma against mechanical stress. In contrast, neither the patent nor the latent VVA-reactive glycoforms of -dystroglycan exhibited a stable association with any other component in the dystrophin- or utrophin-glycoprotein complexes (Fig. 2 and 6b). Thus, VVA reactivity identifies "free agent" glycoforms of -dystroglycan that may function in ways that are modified or distinct from that of the predominant glycoform in maintaining membrane stability. Other studies have reported evidence for expression of distinct molecular weight species of -dystroglycan within skeletal muscle (41), smooth muscle (20), or nonmuscle tissues (6, 21). Because the -dystroglycan core protein has a predicted molecular weight of 67,000 (42), and chemical deglycosylation reduced skeletal muscle -dystroglycan from an apparent molecular weight of 156,000 to 67,000 (32), the observed differences in -dystroglycan electrophoretic mobility likely arose from differential glycosylation. In one study (41), the expression of distinct -dystroglycan glycoforms in skeletal muscle was shown to vary with developmental state and under neural control. In two studies (20, 21), -dystroglycans of distinct molecular weight were shown either to display differential protein associations or to exist in free agent forms as we have shown for VVA-reactive glycoforms in skeletal muscle. Although past studies have relied on dramatic differences in -dystroglycan electrophoretic mobility as evidence for differential glycosylation, our current results demonstrate that -dystroglycan molecules of identical molecular weight can vary with respect to their expression of distinct glyco-epitopes as well as in their protein-protein interactions.

The evidence for expression of distinct -dystroglycan glycoforms combined with the growing list of extracellular molecules that bind -dystroglycan (2, 43) has tempted speculation that differential glycosylation may specify or modulate the ligand binding properties of -dystroglycan. However, no clear support for this hypothesis has been reported. In the current study, we examined whether the presence of distinct glyco-epitopes may instead serve as useful markers to identify -dystroglycan glycoforms that perform unique functions. We focused our effort on testing a hypothesized role for -linked, terminal GalNAc residues in marking a unique -dystroglycan glycoform that specifically participates in the formation of dense AChR clusters at the NMJ. Several previous results made this hypothesis particularly attractive. First, VVA was shown to specifically label the NMJ in skeletal muscle from a wide variety of species (7). -Dystroglycan purified from nervous tissue was distinguished from the predominant -dystroglycan in skeletal muscle by its patent reactivity with VVA (6). Interestingly, purified skeletal muscle -dystroglycan displayed latent VVA reactivity after digestion with C. perfringens neuraminidase (6), a treatment that also induced agrin-independent AChR clustering in C2C12 myotubes (8, 36). VVA was shown to modestly induce AChR clustering by itself and to potentiate agrin-induced AChR clustering, whereas -N-acetylhexosaminidase digestion dramatically inhibited agrin-induced AChR clustering in C2C12 myotubes (8). Like laminin-1-induced AChR clustering (33, 34), VVA-induced clustering was shown to occur in MuSK / myotubes (35) and in the absence of MuSK or AChR -subunit phosphorylation (31). Furthermore, -dystroglycan in C2C12 myotubes displayed both patent VVA reactivity and also latent reactivity after neuraminidase treatment (31). Finally, VVA-induced AChR clustering was nearly abolished in myotubes from mice in which the dystrophin/utrophin-glycoprotein complex protein -dystrobrevin was ablated (4). The sum of these results provided an attractive but circumstantial argument that the unidentified VVA receptor involved in AChR clustering and -dystroglycan may be one and the same.

In the present study, we confirmed several of the previous results linking VVA with -dystroglycan, and we further demonstrated the expression of two minor -dystroglycan glycoforms in adult skeletal muscle that displayed either patent or latent reactivity with VVA. However, we also performed two additional experiments that lead us to conclude that -dystroglycan is not the receptor mediating the effect of VVA on AChR cluster formation in C2C12 myotubes. First, we found that saturating concentrations of VVA and laminin-1 were strictly additive with respect to AChR cluster formation (Fig. 5), which suggests that laminin-1 and VVA do not compete for the same cell surface receptor on C2C12 myotubes. Second, we found that -N-acetylhexosaminidase digestion dramatically inhibited agrin-, VVA-, and laminin-1-induced AChR clustering in C2C12 myotubes; however, this treatment had no effect on the amount of -dystroglycan that was bound to VVA-agarose (Fig. 7). Rather, our results argue for the existence of a "secret agent" VVA receptor that participates in AChR cluster formation (Fig. 8). Our finding that both agrin- and laminin-1-induced AChR clustering are inhibited by treatment of C2C12 myotubes with -N-acetylhexosaminidase suggests that this unidentified VVA receptor is an important downstream component of both clustering pathways. Future effort will focus on the identification of NMJ constituents that are both reactive with VVA and sensitive to digestion by -N-acetylhexosaminidase.

View larger version (16K): [in this window] [in a new window]   Fig. 8.   V. villosa agglutinin and -N-acetylhexosaminidase appear to act on a novel component of the AChR clustering pathway. Based on our data and other studies (8, 35), we propose the existence of a novel VVA receptor that lies downstream and is common to both the agrin- and laminin-mediated AChR clustering pathways. -N-Acetylhexosaminidase presumably inhibits agrin-, laminin-, and VVA-induced AChR clustering by cleaving GalNAc residues from the novel receptor.

	ACKNOWLEDGEMENTS

We thank Drs. Kevin Campbell and Steven Kaufman for providing antibodies to utrophin and ₇ integrin, respectively and also for insightful discussions of the current study. We are also grateful to Drs. Mark Bowe and Justin Fallon for the C-Ag4,8-conditioned medium, Dr. Hynda Kleinman for purified laminin-1, and Drs. Jonathan Cohen, Stanley Froehner, and Joshua Sanes for providing antibodies used in this study.

	FOOTNOTES

^* This work was supported by a grant from the Muscular Dystrophy Association, National Institutes of Health Grant ARO1985 to (J. M. E.), and an American Heart Association-Northland Affiliate Predoctoral Fellowship (to E. L. M.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Physiology, University of Wisconsin-Madison, 127 Service Memorial Inst., 1300 University Ave., Madison, WI 53706. Tel.: 608-265-3419; Fax: 608-265-5512; E-mail: ervasti@physiology.wisc.edu.

Published, JBC Papers in Press, July 17, 2001, DOI 10.1074/jbc.M103843200

² A. Combs and J. Ervasti, unpublished results.

	ABBREVIATIONS

The abbreviations used are: AChR, acetylcholine receptor; NMJ, neuromuscular junction; GalNac, N-acetylgalactosamine; VVA, V. villosa agglutinin; CSM, crude surface membrane(s); WGA, wheat germ agglutinin; PAGE, polyacrylamide gel electrophoresis.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles: