Glycoproteins of rat skeletal muscle sarcolemma: Characterization by two-dimensional gel electrophoresis and effect of denervation

EXPERIMENTAL

108,156-161(1990)

NEUROLOGY

Glycoproteins of Rat Skeletal Muscle Sarcolemma: Characterization by Two-Dimensional Gel Electrophoresis and Effect of Denervation R. C. IANNELLO’ Children’s

Medical

Research

Foundation,

AND

P.O. Box 61, Camperdown,

INTRODUCTION In recent years considerable effort has been directed toward determining the relationship among the carbohydrate components of skeletal muscle membrane (sarcolemmal) surface glycoproteins and the processes of muscle differentiation, functioning of the neuromuscu-

0014-4666/90

$3.00

Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

New

South

Wales 2050, Australia

lar junction, and the basis of muscle diseases (6). One approach has been to characterize the carbohydrate moieties of sarcolemmal glycoproteins from various fiber types and the changes occurring following denervation. Quantitatively the monosaccharide sugar components of sarcolemmal glycoproteins are altered during development and following denervation. Sarcolemmal sialic acid, hexosamine, and total hexose increase following denervation (3, 10, 11, 16, 21, 27), whereas the fucose content decreases (21). During postnatal development of rat skeletal muscle sialic acid content decreases and is reflected in reduced sialyltransferase activity (7). The denervation-induced charges in sarcolemmal carbohydrate content were paralleled in glycosyltransferase activities. Elevated carbohydrate monosaccharide content was reflected by increased amounts of glycosyltransferaseenzyme (16,17). Since exoglycosidase activities were found to increase following denervation (16) the carbohydrate content must be the result of an increased synthesis/degradation ratio with elevations in both individual components. Such alterations in carbohydrate metabolism would be expected to reflect as an altered membrane glycoprotein composition. The glycoprotein composition of mixed, fast, and slow muscle membranes has been determined using lz51-labeled lectin binding to intact and solubilized sarcolemma1 membranes (20, 22). Concanavalin A which is specific for internal mannose residues of N-linked carbohydrates was used to detect changes in glycoprotein composition following denervation. Little difference was found between the pattern of distribution of bound lz51-ConA in membranes from normal and denervated mixed sarcolemma. A most surprising feature was that the quantity of ConA bound to the sarcolemmal glycoproteins after solubilization and electrophoresis was very similar in normal and denervated sarcolemma. This is in contrast with the fact that there was a fourfold difference in the binding of ConA to the same membrane when the binding was done to intact membranes. We have interpreted this as an indication that the geometric arrangement of the major glycoproteins in sarcolemma changes following denervation, such that the ConA binding sites on internal mannose residues of N-

Sarcolemmal membrane glycoproteins from rat mixed, fast, and slow muscles were characterized by concanavalin A (ConA) binding following two-dimensional polyacrylamide gel electrophoresis (PAGE). Analysis of electrophoretic profiles revealed that although sarcolemmal membranes prepared from these muscle types contained common glycoprotein species, each had a distinct glycoprotein pattern. In sarcolemma from mixed muscle, three major classes of ConA binding glycoproteins could be distinguished: (i) an acidic species of llO,OOO-120,000 Da, ~15.0 to 5.3 (CG-1, ConA binding glycoproteins-1); (ii) a group of highly charged isomers, ranging from 75,000 to 80,000 Da pl 5.2 to 8.2 (CG-2); and (iii) a group of charged isomers of predominantly acidic nature of approximately 50,000 Dap15.2 to 5.8 (CG-3). ConAbound exclusively to CG-1 in sarcolemma from a fast muscle (extensor digitorum longus muscle, EDL). In soleus muscle sarcolemma (slow fiber type) both CG-1 and CG-3 were readily detected but CG-2 was markedly diminished. ConA binding to slow muscle sarcolemma revealed as well a glycoprotein species of 66,000-70,000 Da, pl 4.3-5.1 (CG-4), which was unique to this fiber type and as such may be a specific marker for slow fiber type. Denervation had no significant effect on the properties of ConA binding to mixed or slow muscle sarcolemma but dramatically altered the ConA binding to fast muscle sarcolemma, specifically increasing binding to CG2. These findings demonstrate that denervation differentially affects the metabolism of ConA binding glycoproteins in these muSCle types. 0 1990 Academic P~ZSS, 1~.

1 Present address: Department of Anatomy, University San Francisco, San Francisco, CA 94143. * To whom correspondence should be addressed.

P. L. JEFFREY’

of California

156

SARCOLEMMAL

linked carbohydrates, which are cryptic in normal membranes, become exposed (20,21). More ‘251-ConA bound to intact sarcolemmal membranes from normal soleus muscle than from normal EDL or mixed muscles. However, following denervation, the number of binding sites changed little in soleus membranes, whereas they almost doubled in EDL membranes and increased almost fourfold in the mixed muscle sample. These results suggest that the spatial organization of the surface glycoproteins of fast and slow muscle fibers is different, with fast fibers having a large proportion of “cryptic” lectin binding sites (18, 20, 22). In these studies, however, the identification of membrane glycoproteins was limited to characterization by one dimensional SDS-PAGE. Since a more detailed analysis of these membranes components is required before an insight into the cryptic nature of the sarcolemma1 glycoproteins can be achieved, we have chosen to use a high resolving two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) technique to further examine the sarcolemmal glycoprotein composition of normal and denervated mixed, EDL, and soleus muscle. The results of this investigation have allowed the putative identification of muscle fiber-specific glycoproteins and of the differential effects of denervation on sarcolemmal glycoproteins of different muscle types. METHODS

Materials P-Mercaptoethanol and horseradish peroxidase were purchased from Sigma Chemical Co.; Nonidet-P40 (NP40) was purchased from BDH (England); sodium lauryl sulfate was obtained from National Diagnostics; acrylamide, ammonium persulfate, bisacrylamide, and N,N,N’,N’-tetramethylenediamine (TEMED) were purchased from Amresco ARPC (USA); ampholines (preblended), pI3.5-9.5, were obtained from LKB (Sweden); and concanavalin A was purchased from BoehringerMannheim (Germany). Membrane

Preparation

In this study, skeletal muscle membranes were prepared from adult female Sprague-Dawley rats (200-250 g) as described (2). Membranes from three different muscle fiber populations-mixed, fast, and slow-were analyzed. To prepare membranes from mixed muscle fibers, the extensor digitorum longus, anterior tibialis, gastrocnemius, and soleus were pooled. When membrane preparations from fast and slow muscle fibers were required, extensor digitorum longus and soleus muscle were used as the source of fast and slow muscle, respectively. Iodoacetamide (1.0 mM), PMSF (0.1 mM), and pepsatin (0.1 mg/ml) were included to inhibit endog-

157

GLYCOPROTEINS

enous protease o-4°C.

activity.

All steps

were

carried

out at

Muscle Denervation Following anesthesia, rats were denervated by removing a l-cm segment of sciatic nerve from the midthigh. The wound was sutured by Michelle surgical clips (12 mm) and the animals were allowed access to food and water for 7 days before being sacrificed. Sample Preparation Electrophoresis

for Two-Dimensional

Sarcolemmal membranes were prepared for isoelectric focusing using a modification of the method described by Ames and Nikaido (1). To 200 pg of lyophilized membrane protein, 50 ~1 of a solution containing 1.0% (w/v) SDS, 5% (v/v) glycerol, 2.5% (v/v) /3-mercaptoethanol, and 30 mM Tris-HCl, pH 6.8, was added. The sample was heated for 5 min at 100°C and allowed to cool at room temperature before an equal volume of buffer containing 9.5 M urea, 2% (v/v) preblended ampholines (pl 3.5-9.5), and 10% (w/v) NP-40 was added. After allowing the sample to.stand for 30 min, insoluble material was removed by centrifugation for 30 min at 100,OOOg (Beckman Airfuge). Isoelectric

Focusing

Gel Composition

IEF gel solution was prepared by the sequential addition of 5.5 g urea; 1.0 ml acrylamide stock (38% acrylamide; 2.0% bisacrylamide); 2 ml 10% (w/v) NP-40; 2.5 ml deionized, distilled water; and 0.75 ml of (PI 3.5-9.5) 3% (v/v) preblended ampholines. Polymerization was initiated by the addition of 11~1 TEMED followed by 15 ~1 of 10.0% (w/v) ammonium persulfate. The mixture was poured into glass tubes (3.0-mm i.d.) to a height of 115 mm (0.8 ml per tube) and overlayed with water. After 2 h, the water was removed and replaced with sample overlay buffer consisting of 9.5 M urea, and 2% (v/v) preblended ampholines (~13.5-9.5). Loading and Running of Membrane in the First Dimension

Samples

Following polymerization, IEF gels were prerun as described by O’Farrell (25) to remove ammonium persulfate and to establish a pI gradient. The power was then turned off, and the overlay solution was removed and replaced with equivalent amounts of the prepared membrane samples (usually 30-50 ~1). Samples were overlayed with 10 ~1 of fresh sample overlay buffer and run for 16 h at 350 V followed by 800 V for 1 h. H,PO, (0.1 M) and NaOH (0.02 M) were used as anode and cathode solutions, respectively. The IEF parameters described above were determined from a preliminary study in which different ratios of ampholine ranges and various

158

IANNELLO

FIG. 1. Mixed muscle sarcolemmal lin A-horseradish peroxidase (ConA-HRP) teins l-7 as described in the text.

glycoprotein staining

in the Second Dimension

Prior to two-dimensional separation, IEF gels were equilibrated for 10 min in equilibration buffer containing 2.3% (w/v) SDS, 10% (v/v) glycerol, 5.0% (v/v) pmercaptoethanol, 62.5 mM Tris-HCl, pH 6.8. Separation by SDS-PAGE was performed as described by O’Farrell (25) using 10% polyacrylamide gels and the discontinuous buffer system of Laemmli (19). Glycoprotein

Visualization

Following 2D-PAGE separation, membrane components were transferred to nitrocellulose by Western transfer and ConA binding glycoproteins were visualized by the concanavalin A-horseradish peroxidase method described by Faye and Chrispeels (14). RESULTS

Normal

Sarcolemmal

JEFFREY

profiles. (A) Normal. (B) Denervated. Glycoproteins as described under Methods. CG-1, -2, -3, and CG-5,

focusing times were tested. The IEF conditions used in this study generated pI gradients which remained stable over the focusing period and gave reproducible glycoprotein profiles. Electrophoresis

AND

Profiles

The distribution of sarcolemmal glycoproteins from rat mixed, fast, and slow muscle following BD-PAGE was analyzed using concanavalin A binding. This lectin binds to the outer terminal trimannosyl group of high mannose and hybrid type glycoproteins as well as to the internal mannose residues in complex type glycoproteins (4,5, 23). Since mannose is a common constituent of glycoproteins, we would expect a large variety of these membrane components to be represented in this study. When mixed muscle sarcolemmal glycoproteins were analyzed by BD-PAGE, profiles revealed over 60 ConA binding glycoprotein species. Their apparent molecular size ranged from 30,000 to 200,000 Da with PI’S between 4.0 and 8.5. The majority of these however, fell between pI 4.5 and 6.0 (Fig. 1A). Furthermore, the distribution of

were visualized following concanava-6, -7 refer to ConA binding glycopro-

ConA binding was such that we could categorize major glycoprotein species into three major classes: (i) an acidic species of llO,OOO-120,000 Da, ~15.0 and 5.3; (ii) a group of highly charged isomers, ranging from 75,000 to 80,000 Da, pI 5.2 and 8.2; and (iii) a predominantly acidic species of approximately 50,000 Da, pI 5.2 to 5.8 which also appear as highly charged isomers. For convenience we have arbitrarily designated these as CG-1 (ConA binding glycoproteins-1), CG-2, and CG-3, respectively. Minor binding glycoprotein species are found at 45,000 Da, pI 8.2 (CG-5); 49,000 Da, ~16.1-6.6 (CG-6), and 40,000 Da, ~15.5 (CG-7). Surprisingly, we observed that ConA bound predominantly to CG-1 in sarcolemmal membranes prepared from the fast fiber EDL muscle (Fig. 2A). Furthermore, the pattern of ConA binding in these membranes was not altered by increasing the total amount of membrane protein (data not shown). This result is in contrast to previous one-dimensional SDS-PAGE studies in which the distribution of ConA binding to EDL membranes was found to be similar to that observed in mixed muscle sarcolemma (22). It is unlikely that the ConA binding patterns we observed for normal and denervated EDL represent altered membrane glycoprotein contents resulting from differential extraction incurred during either the membrane preparative procedure or during the IEF preparation of the samples. Previous studies have demonstrated that the yield and purity of membranes prepared by the isolation procedure used in this study are similar for both normal and denervated membranes (21, 22). In addition, we examined the efficiency of the detergents used in our IEF sample preparation buffer to extract glycoproteins from sarcolemmal membranes. Using mixed sarcolemmal membranes, no selective differences in glycoprotein extraction between normal and denervated muscles were found (data not shown). Our two-dimensional profiles also revealed that sarcolemma1 membranes prepared from, each of the three muscle fiber types share common glycoproteins. For ex-

SARCOLEMMAL

159

GLYCOPROTEINS

A

200. 116. 92.

66.

45

FIG. staining

2. EDL muscle sarcolemmal as described under Methods.

glycoprotein profiles. CG-1, -2, and -3 refer

(A) Normal. (B) Denervated. to ConA binding glycoproteins

ample, as with mixed and fast muscle EDL sarcolemmal membranes, ConA binding to CG-1 was also observed in soleus muscle sarcolemma (Fig. 3A) and CG-3 appeared as a common constituent in both soleus and mixed muscle sarcolemmal membranes. However, despite these similarities, the ConA binding profiles of mixed, EDL, and soleus sarcolemma were distinctive. Analysis of soleus muscle sarcolemmal membranes revealed a number of differences. CG-2, a major component in mixed sarcolemmal membranes, was markedly diminished in soleus. Instead, soleus membranes were enriched in a group of glycoproteins ranging from 66,000 to 70,000 Da, ~14.3 to 5.1 (CG-4). Interestingly, CG-4

Glycoproteins were visualized following 1,2, and 3 as described in the text.

ConA/HRP

could not be detected in EDL membranes, suggesting that this glycoprotein may be a specific marker for slow fiber muscles. ConA binding to 2D-PAGE did however demonstrate the presence of this species in mixed sarcolemma1 membranes in relatively low abundance. Since the soleus contributes to the fiber population in mixed muscle preparations, this result was not surprising. In addition to the differences observed in the distribution of ConA binding, muscle-specific modifications to common glycoprotein species were also observed. CG-3 is common to both soleus and mixed muscle sarcolemma, however, it differs in each muscle type by the number of isomers present. Soleus membranes lack the ~15.6 and

FIG. 3. Soleus muscle sarcolemmal glycoprotein profiles. (A) Normal. (B) Denervated. HRP staining as described under Methods. CG-1, -3, and -4 refer to ConA binding glycoproteins

Glycoproteins were visualized following 1,3, and 4 as described in the text.

ConA/

160 5.8 isomers which brane profiles.

Effect of Denervation

IANNELLO

are observed

in mixed

on Sarcolemmal

muscle mem-

Glycoproteins

BD-PAGE profiles revealed no gross qualitative alterations in the glycoprotein composition of mixed muscle sarcolemmal membranes following denervation; however, a number of quantitative changes were observed. These included diminished amounts of ConA binding to the 45,000 Da (pl8.2), CG-5; 49,000 Da (p16.1-6.6), CG6; and 40,000 Da (pl 5.5), CG-7 glycoprotein species (Fig. 1B). In contrast, both the ConA distribution and the total binding to EDL sarcolemmal membranes were dramatically altered following denervation (Fig. 2B). These changes included a noticeable increase in CG-1, but more significant was the appearance of CG-2 and CG-3, neither of which were detectable in normal profiles. Finally, denervation did not significantly alter the distribution of ConA binding in soleus sarcolemmal membranes, although changes in the relative amounts of ConA binding for most glycoprotein species were observed including diminished binding to CG-1 and CG-4 and increased binding to CG-3 (Fig. 3B). Interestingly, the denervation effects on the isomeric components of CG-3 were differential in that ConA binding isomers with prs of 5.6 to 5.8 were proportionally larger. DISCUSSION In previous studies, the compositional characterizations of mixed, EDL, and soleus muscle sarcolemmal glycoproteins were analyzed by ConA binding following one-dimensional SDS-PAGE (21, 22). Electrophoretic profiles obtained from these studies revealed that the distribution of ConA binding to membrane components was similar in each of the muscle fiber types. In addition, neither the pattern of distribution nor the total binding of ConA was significantly altered following denervation. In this investigation, however, a high resolving 2DPAGE technique was used to further examine the sarcolemma1 glycoprotein composition of normal and denervated rat skeletal muscles. Using this technique we have been able to demonstrate major differences between the sarcolemmal glycoprotein composition of mixed, EDL, and soleus muscles which were not previously observed using the more conventional method of SDS-PAGE. Of particular value was the observation that sarcolemma1 membranes prepared from different muscle types could be distinguished from one another on the basis of their glycoprotein composition. Two-dimensional electrophoretic profiles revealed that the pattern of ConA binding in each of the muscle membranes is quite distinct. CG-1, a common glycoprotein species in all of the sarcolemmal membranes analyzed, was the only major

AND

JEFFREY

glycoprotein species detected in normal EDL membranes. By contrast, soleus muscle characteristically expressed CG-4, a group of sarcolemmal glycoproteins we believe to be unique to slow muscle fibers. Another abundant component of soleus membranes is CG-3; however, unlike CG-4, this glycoprotein species was also present in mixed sarcolemmal membranes. The major muscle used in preparing mixed muscle sarcolemma is the gastrocnemius, a muscle predominantly composed of fast fibers (8, 9). Surprisingly, CG-2, a major component of mixed muscle sarcolemma, was not detectable in normal EDL membranes. Although we have no explanation for such a result, it would be appealing to speculate that the fast fibers of EDL and those of the gastrocnemius muscles are under diverse physiological and biochemical constraints and that this is depicted in the glycoprotein composition. Interestingly, CG-2 is markedly diminished in soleus sarcolemmal membranes. As well as the putative identification of muscle fiberspecific membrane glycoproteins, characterization of sarcolemmal components using BD-PAGE has also allowed the differential effects of denervation on membrane glycoproteins from different muscle types to be observed. Denervation had no significant effect on the properties of ConA binding in mixed or slow muscle sarcolemma but dramatically altered the ConA distribution in EDL membranes. Several studies have reported differential effects of denervation on a number of biochemical, morphological, and physiological parameters in EDL without significantly affecting soleus muscle. These included increased membrane glycosyltransferase and neuraminidase activities (10, ll), increased ConA binding to intact membranes (22), preferential affects of atrophy (13,24), an increased number of membrane orthogonal arrays (12,26), and alterations in protein synthesis (15). The data presented here are in good agreement with these earlier reports and, taken together, strongly suggest that (i) EDL and soleus are under differential neural regulation, and (ii) denervation differentially affects the glycoprotein turnover in these fiber types. It is interesting to speculate on the activities of the glycosyl transferases and glycosidases in the normal and denervated state on the glycoprotein p1 values from the fiber types. For example, the decrease seen in CG-5 following denervation (Fig. 1) in mixed sarcolemma may reflect an increased turnover, whereas the pl values for CG-4 and CG-3 in soleus (Fig. 3) may reflect an altered ratio of transferase and glycosidase activities for the terminal sugar residues. In this study, no apparent difference in ConA binding to the membrane components of normal and denervated mixed sarcolemma membranes could be detected following membrane separation by BD-PAGE. However, alterations in the carbohydrate composition and glycosyltransferase activities of these membranes have been observed (3, 10, 11, 16, 21). Furthermore, Leung et al. (20,

SARCOLEMMAL

21) reported a fourfold increase in the number of ConA binding sites in intact mixed membranes following denervation. This difference in ConA binding between normal and denervated membranes, however, could be abolished if membranes were first perturbed by detergent. They postulated that in the native state, conformational changes in membranes glycoproteins take place following denervation which are not seen if the membrane components are disrupted. Similar cryptic behavior was also observed in EDL membranes but not in soleus (22). From the results presented here, the pattern of ConA binding observed for normal and denervated mixed muscle sarcolemma appears consistent with this hypothesis. Our results, however, do not support a similar conclusion for EDL. The mechanism(s) responsible for this conformational change is not clear, but it is believed to involve a loss of membrane-bound L-fucose residues following denervation (R. C. Iannello and P. L. Jeffrey, unpublished results). ACKNOWLEDGMENTS This work was supported by grants from the Australian Grants Scheme and the Muscular Dystrophy Association South Wales.

11. COTRUFO,

R., AND G. SAVATTIERI. 1977. Effects of denervation on the neuraminidase activity of slow and fast muscles of rabbits. J. Neural. Sci. 34: 233-240.

12.

ELLISMAN, M. H., J. E. RASH, L. A. STAEHALIN, AND K. R. PORTER. 1976. Studies of exitable membranes. II. A comparison of specializations of neuromuscular junctions and non-junctional sarcolemma of mammalian fast and slow twitch muscle fibres. J. Cell Biol. 68: 752-774.

13.

ENGEL, W. K., M. H. BROOKE, AND P. H. NELSON. 1966. Histochemical studies of denervated or tenotomized cat muscle: Illustrating difficulties in relating experimental animal conditions to human neuro-muscular disease. Ann. N. Y. Acad Sci. 138: 160186.

14.

FAYE, L., AND M. J. CHRISPEELS. 1985. Characterization of Nlinked oligosaccharides by affinoblotting with concanavalin Aperoxidase and treatment of blots with glycosidases. Anal. Biothem. 149: 218-224.

15.

GOLDSPINK, D. F. 1978. The influence of passive stretch on the growth and protein turnover of denervated extensor digitorum longus muscle. Biochem. J. 174: 595-602.

16.

JEFFREY, P. L., AND S. H. APPEL. 1978. Denervation in surface membrane glycoprotein glycosyltransferases malian skeletal muscle. Exp. Neural. 61: 432-441.

17.

JEFFREY, P. L., W. N. LEUNG, vation alterations in surface muscle. In Muscle, Nerve, and and J. K. Tomkin, Eds.), pp. dam/Oxford.

18.

JEFFREY, P. L., W. N. LEUNG, AND J. A. P. ROSTAS. 1983. The surface membranes of fast and slow muscle: Differential effects of denervation. In Molecular Aspects of Neurological Disorders (L. Austin and P. L. Jeffrey, Eds.), pp. 83-94. Academic Press, Sydney.

19.

LAEMMLI, V. K. 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (London) 227: 680-685.

20.

LEUNG, W. N., P. L. JEFFREY, AND J. A. P. ROSTAS. 1982. Denervation exposes cryptic ConA binding sites in skeletal muscle membranes. Neurosci. Lett. 30: 31-36.

21.

LEUNG, W. N., P. L. JEFFREY, AND J. A. P. ROSTAS. 1984. The effects of denervation on mammalian sarcolemmal proteins and glycoproteins. Muscle Nerve 7: 35-49.

22.

LEUNG, W. N., P. L. JEFFREY, AND J. A. P. ROSTAS. 1986. Effect of denervation on sarcolemmal proteins and glycoproteins of fast and slow mammalian skeletal muscle. Exp. Neurol. 91: 229-245.

23.

LOTAN, R., AND G. L. NICHOLSON. 1979. Purification brane glycoproteins by lectin affinity chromatography. Biophys. Acta 559: 329-376.

24.

NIEDERLE, B., AND R. MAYR. 1978. Course of denervation phy in type I and type II fibres of rat extensor digitorum muscle. Anat. Embryol. 153: 9-21.

25.

O’FARRELL, P. H. 1975. High resolution two dimensional phoresis of proteins. J. Biol. Chem. 250: 4007-4021.

26.

RASH, J. E., AND M. H. ELLISMAN. 1974. Studies of excitable membranes. I. Macromolecular specializations of the neuromuscular junction and the non-junctional sarcolemma. J. Cell Biol. 63: 567-586.

27.

SMITH, P. B., AND S. H. APPEL. 1977. Development of denervation alterations in surface membranes of mammalian skeletal muscle. Exp. Neural. 56: 102-114.

Research of New

REFERENCES 1.

AMES, G. F. L., AND K. NIKAIDO. 1976. Two-dimensional gel electrophoresis of membrane proteins. Bioc&mistry 15: 616-623.

2.

ANDREW, terization

3.

APPEL, S. H., A. D. ROSES, R. R. SMITH, J. 0. MCNAMARA, AND D. brane biochemical approaches to function. In The Nervous System, 443-454. Raven Press, New York.

C. G., AND S. H. APPEL. 1973. Macromolecular of muscle membranes. J. Biol. Chem. 248:

charac5156-5163.

ALMON, C. G. ANDREW, P. B. A. BUTTERFIELD. 1975. Memaltered muscle structure and (D. B. Tower, Ed.), Vol. 1, pp.

4.

BAENZIGER, J. U., AND D. FIETE. 1979. Structure of the complex oligosaccharides of fetuin. J. Biol. Chem. 254: 789-795.

5.

BHATTACHARYYA, L., C. CECCARINI, P. LORENZONI, AND C. F. BREWER. 1987. Concanavalin A interection with asparaginelinked glycopeptides: Bivalency of high mannose and bisected hybrid type glycopeptides. J. Biol. Chem. 262: 1288-1293.

6.

7.

8. 9. 10.

CAMPBELL, K. P., AND S. D. KAHL. 1989. Association of dystrophin and an integral membrane glycoprotein. Nature (Lendon) 338: 259-262. CLARK, G. F., AND P. B. SMITH. 1983. Studies on glycoconjugate metabolism in developing skeletal muscle membranes. Biochim. Biophys. Actu 755: 56-64. CLOSE, R. I. 1967. Properties of motor units in fast and slow skeletal muscles of rat. J. Physiol. 193: 45-52. CLOSE, R. I. 1972. Dynamic properties muscles. Physiol. Rev. 52: 129-197.

of mammalian

161

GLYCOPROTEINS

skeletal

COTRUFO, R., AND S. H. APPEL. 1973. Effects of denervation glycoproteins of rabbit gastrocnemius and soleus muscles. Neurol. 39: 58-69.

on Exp.

alterations of mam-

AND J. A. P. ROSTAS. 1979. membranes of mammalian Brain Degeneration (A. D. 32-44, Excerpta Medica,

Denerskeletal Kidman Amster-

of cell memBiochim. atrolongus electro-