News Physiol Sci 15: 255-259, 2000;
1548-9213/00 $5.00
News in Physiological Sciences, Vol. 15, No. 5, 255-259,
October 2000
© 2000 Int. Union Physiol. Sci./Am. Physiol. Soc.
The Role of Dystroglycan and Its Ligands in Physiology and Disease
Thomas Meier and
Markus A. Ruegg
T. Meier is at MyoContract and M. A. Ruegg is in the Department of Pharmacology/Neurobiology, Biozentrum, University of Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland.
 |
Abstract
|
|---|
Dystroglycan contributes to the formation of basement membrane during embryonic development and enforces cell membrane integrity by bridging cytoskeleton and components of the extracellular matrix. In several forms of muscle disease, dystroglycan is reduced in abundance. Moreover, human viral and bacterial pathogens use dystroglycan as their cellular entry point.
 |
Introduction
|
|---|
About 300,000 cases of Lassa fever infections are reported every year in West African countries, with an overall mortality among hospital cases of ~15%. Food contaminated with excreta of wild rodents carrying Lassa fever virus (LFV) is the principal mode of disease transmission. In addition, there is a high risk of infection with contaminated human body fluids, which is exceptional for arenaviruses. The "cellular receptor" by which arenaviruses, such as LFV and the closely related lymphotic choriomeningitis virus, enter affected cells has now been identified: the peripheral membrane component of the dystroglycan (DG) complex,
-DG (2). Together with the transmembrane counterpart ß-DG, this DG complex has already been associated with human pathology. In several forms of muscular dystrophy, the DG complex is reduced in abundance at the muscle cell surface, where it normally binds to molecules of the extracellular matrix (ECM). The interaction between DG and the ECM component laminin-2 is also the point of entry of another widespread human pathogen, Mycobacterium leprae, the causative organism of leprosy (13). This pathogen invades Schwann cells of the peripheral nervous system, and the resulting damage to the peripheral nerves culminates in incurable disfigurement and physical disabilities, symptoms already known to the oldest human civilizations of China, Egypt, and India. Thus
- and ß-DG seem to be involved in several human diseases. Given this importance, we will summarize the role of
- and ß-DG and their associated molecules in physiology and pathology.
 |
The DG complex
|
|---|
-DG and ß-DG are derived from a common precursor propeptide by posttranslational cleavage. Both
- and ß-DG are part of a large membrane-associated protein complex called the dystrophin glycoprotein complex (DGC; Fig. 1
). This name describes the fact that molecules of the DGC interact with the cytoskeletal component dystrophin in skeletal muscle fibers, where this interaction was first described. The most striking biochemical property of
-DG is its high degree of glycosylation (up to 50% of its molecular mass). There are a few conserved glycosaminoglycan chain attachment sites, as well as potential N-linked glycosylation sites distributed over the entire molecule. In addition, the mucin-like region in the center of the molecule becomes extensively O-glycosylated. Interestingly, the extent of glycosylation on
-DG varies between tissues. For example,
-DG isolated from brain is less glycosylated than
-DG isolated from muscle. There is also evidence that modulation of the binding properties between
-DG and its ligands is a consequence of tissue-specific differences in glycosylation (4, 5). This dependence on carbohydrates for functional binding distinguishes the interaction between
-DG and its ECM ligands from those of other cellular receptors, such as the integrins and their corresponding ECM ligands.

View larger version (33K):
[in this window]
[in a new window]
|
FIGURE 1. Diagrammatic representation of dystroglycan (DG) as a transmembrane linker of muscle cells. At extrasynaptic regions of the muscle cell membrane [Non-neuromuscular junction (non-NMJ)], ß-DG is associated with dystrophin and the F-actin network, whereas at the NMJ, utrophin substitutes for dystrophin. Throughout the muscle cell membrane, ß-DG is associated with the sarcoglycans and sarcospan, forming the dystrophin glycoprotein complex (DGC; for clarity sarcoglycans and sarcospan are omitted in the NMJ region). The peripheral membrane protein rapsyn associates with ß-DG, acetylcholine receptors (AChRs), and muscle-specific kinase (MuSK). Perlecan is found in NMJ and non-NMJ regions of the extracellular matrix. ß2-syntrophin is enriched at the NMJ, where it is associated with microtubule-associated serine/threonine kinase (MAST). In extrasynaptic regions, neuronal nitric oxide synthase (nNOS) interacts with 1- and/or ß1-syntrophin. The two carboxy terminal G-like domains of laminin interact with the glycoprotein -DG. Interaction between agrin and laminin is conveyed by the amino terminal NtA domain of agrin. At the carboxy terminal region, the last laminin G-like domain activates the agrin receptor/MuSK complex. The laminin G-like domains 1 and 2 of agrin bind to -DG. ECM, extracellular matrix; CM, cell membrane; CP, cytoplasm; agrin-R, agrin-binding part of the functional agrin receptor (the molecular identity of this molecule is not known); Grb2, GDP/GTP exchanger.
|
|
Binding partners for the transmembrane protein ß-DG range from large cytoplasmic components, such as muscle dystrophin and its close homologue utrophin, to the adapter protein Grb2, which may be involved in intracellular signaling (see below). A list of interacting proteins and their interacting domains is given in Table 1.
 |
Extracellular ligands of -DG
|
|---|
The discovery that the DGC is localized to the membrane of cells, in particular to the membranes of muscle fibers, prompted Campbell and colleagues to postulate that ECM components may serve as ligands for the DGC. Indeed, overlay assays using purified components of the ECM revealed that laminin-1 is a high-affinity ligand for
-DG. Full deglycosylation of
-DG abolishes the binding to laminin-1. Laminin-1 is a member of a growing family of molecules composed of three independent subunits, termed
, ß, and
, each of which is encoded by several genes. Various heterotrimeric protein isoforms are generated by the combination of different chains. Until now, 12 such laminin isoforms have been described. Of those, binding to
-DG was demonstrated for laminin-1 (
1, ß1,
1) and laminin-2 (
2, ß1,
1). Binding of laminin-1 is mediated by the E3 fragment, which consists of the two most carboxy terminal laminin-G-like (LG) domains (15). LG modules have a globular appearance in the electron microscope and are found in all five laminin
-chains as well as in several other ECM molecules, including the two heparan sulfate proteoglycans, perlecan and agrin, and the neurexins, which are transmembrane proteins involved in synaptic neurotransmitter release. In contrast to the LG domains of the laminin
-chains, the homologous domains in agrin, perlecan, and the neurexins are interrupted by another cysteine-rich module, the epidermal growth factor-like (EG) module. EG modules have an extended rod-like structure, suggesting that they serve as spacers or linkers between neighboring modules. Despite this difference in the spacing, both agrin and perlecan do bind to
-DG. As in the laminin
1- and
2-chains, at least two LG domains are required for high-affinity binding. Interestingly, agrin and perlecan bind to
-DG with even higher affinity than laminin. Moreover, binding of all the DG ligands is strictly Ca2+ dependent and is inhibited by high salt and heparin, although the degree of inhibition varies substantially between different ligands. The binding sites on
-DG for laminin, agrin, and perlecan are overlapping because all of these ligands compete with each other in binding assays. The crystal structure of the fifth LG module of the laminin
2-chain shows a 14-stranded ß-sandwich that is structurally related to the pentraxins (7). One Ca2+ is bound to one edge of the sandwich by conserved acidic residues. The binding of a Ca2+ to the LG module may be the basis for the Ca2+ dependence of the binding to
-DG.
In conclusion, the binding studies and the structural analysis suggest that all LG-module-containing proteins are potential ligands for
-DG. Consequently, the rather widely distributed DG may be bound by distinct sets of ligands, depending on the tissue. For example, laminin-2 (
2, ß1,
1), which is highly abundant in basal lamina of peripheral nerve and in nonsynaptic regions of muscle fibers, is likely to be the main ligand for
-DG in these tissues. In contrast, laminin-4 (
2, ß2,
1), laminin-9 (
4, ß2,
1), laminin-11 (
5, ß2,
1), and agrin are likely to be the main ligands at the neuromuscular junction (NMJ). Although also present in the muscle basal lamina, perlecan is particularly prominent in the basement membrane of blood vessels and chondrocytes and presumably is an important ligand of
-DG in these nonmuscle tissues.
 |
DG and muscle diseases
|
|---|
The level of DG at the cell membrane is often reduced in certain forms of muscle disease, although its aberrant expression appears not to be the primary cause of muscle pathology (Table 2
). Soon after the discovery of the DGs as transmembrane linkers between dystrophin and basal lamina components, several reports documented the reduced amount of
- and ß-DG in muscle cells of Duchenne muscular dystrophy (DMD) patients, as well as mouse and dog models for DMD. On the other hand, Ozawa and colleagues (12) report cases in which ß-DG appeared to be retained in significant amounts in DMD muscle. Hence, it is currently not clear whether
- and ß-DG are regulated independently in DMD muscle cells or whether the residual amounts of dystrophin in the human muscle biopsies under investigation offer an explanation for such conflicting results. Alternatively, utrophin, which is upregulated in DMD muscle, may interact with ß-DG and thus stabilize the DGC. At any rate, the reduction of DG appears to be secondary to the loss of dystrophin. Interestingly, the sarcoglycans, an independent group of transmembrane proteins that interact loosely with DG, show a more profound reduction in dystrophin-deficient muscle, despite the fact that sarcoglycans are not associated directly with dystrophin.
What is the functional consequence of reduced levels of DG in dystrophic muscle? Analysis of DMD and mdx mouse muscle have shown that the mechanical stability of the muscle cell membrane is weakened during cycles of contraction and relaxation, culminating in the formation of "membrane holes," an early stage in muscle cell deterioration. Therefore, the current data support the interpretation that the reduction of
- and ß-DG in dystrophin-deficient muscle may aggravate the loss of cellular integrity and may contribute to the progression of the disease. A recent attempt to ameliorate the dystrophic phenotype of mdx mice clearly showed that overexpression of full-length utrophin restores the membrane-associated DGC, including DG. This results in the normalization of various disease-relevant parameters, such as muscle force generation, muscle cell turnover, and the level of creatine kinase in the blood serum (17).
Another form of muscular dystrophy is caused by mutations in the laminin
2-chain, the main extracellular ligand for
-DG in the extrasynaptic portion of skeletal muscle. Patients who suffer from this congenital muscular dystrophy (CMD) show distorted organization of the basal lamina surrounding muscle fibers and peripheral nerves. Although several clinical parameters, such as increased creatine kinase activity in blood serum, are similar to DMD, the level of membrane-associated
-DG in CMD muscle appears normal. Consequently, the cell membrane is less fragmented, as indicated by the fact that small dye molecules are excluded from muscle fibers of dy mice, a mouse model of CMD. This is in contrast to mdx muscle fibers, which are strongly stained intracellularly with the same dye. Hence, necrosis of muscle fibers in CMD patients must occur by different mechanisms.
Mislocalization of DG is also observed in several distinct forms of limb girdle muscular dystrophy (LGMD). Here, mutations in one of the sarcoglycan genes causes the absence of the entire sarcoglycan complex and the reduction of DG. For example, in the cardiomyopathic BIO 14.6 hamster, which is an animal model for LGMD_2F, the primary absence of
-sarcoglycan results in the lack of DG. Adenovirus-mediated gene transfer of
-sarcoglycan normalizes the levels of
- and ß-DG and the plasma membrane integrity (8). Another example for the involvement of DG in muscle pathology has been observed in LGMD_2D, in which lack of
-sarcoglycan causes the disruption of the entire sarcoglycan complex and, as a secondary effect, the disruption of the interaction between
-DG and the muscle cell membrane (3). Reduced immunostaining for ß-DG has also been observed in Fukuyama-type congenital muscular dystrophy. There, misexpression of fukutin, a presumptive extracellular protein of unknown function, results in abnormalities of the muscle basal lamina. This disease, which is confined to Japan, is of particular genetic interest because it originated from one single founder event and it is the only human disease known so far that is caused by a retrotransposal insertion into the affected gene.
Until now there has been no report on a neuromuscular disease in which misexpression of DG is the primary pathological event. Therefore, the involvement of DG in muscular disorders appears to be secondary to the misexpression or absence of muscle cell membrane proteins (e.g., sarcoglycans) or cytoskeletal components (e.g., dystrophin). The induced absence of DG contributes to the aggravation of muscle cell pathology, and, consequently, the restoration of an intact DGC, including DG, is considered a hallmark for the success of future therapeutic strategies for muscular dystrophies in general and DMD in particular.
 |
DG and the formation of basement membranes
|
|---|
The tight correlation between disease phenotype and alterations in DG in several muscular dystrophies has led to the hypothesis that DG mediates the mechanical linkage between the muscle cytosol and the mature ECM. Analysis of DG-deficient animals has now widened the understanding of the function of
- and ß-DG. Disruption of DG leads to early embryonic death due to lack of progression beyond the early egg cylinder stage. The death of the embryo is due to a structural and functional deficit in the formation of Reichert's membrane, which is one of the earliest basement membranes formed in rodents. As a consequence, the embryo is not separated from maternal tissue and becomes reabsorbed. This failure to form a basement membrane suggested that DG might also be a cellular receptor essential for the assembly of basement membranes. Indeed, detailed analysis of embryoid bodies isolated from DG-deficient animals has now shown that components of basement membranes, including laminin, perlecan, and collagen type IV, cannot bind to the surface of the DG-deficient cells (6). These results strongly suggest that DG is necessary for nucleating the assembly of a primary laminin matrix, which then serves as the scaffold for the assembly of the remaining components of the basement membrane. In accordance with this, muscular dystrophies may also be a consequence of a failure to assemble muscle basal lamina properly. This interpretation is supported by the functional analysis of the second class of receptors for ECM molecules, the integrins. Integrins are heterodimers composed of one
- and one ß-subunit. Variations in
- and ß-subunits give rise to >20 integrins in mammals, where the vast majority of the integrins contain the ß1-subunit. Null mutations of the ß1-integrin subunit result in early embryonic lethality that is, at least in part, due to the failure to assemble a functional basement membrane. Consistent with a similar role of DG and the integrins, mice deficient in the
7- or the
5-subunit of integrin, which are both highly expressed in muscle fibers, display muscular dystrophies (10, 16). Interestingly, the mild dystrophy associated with the loss of the
7-subunit becomes much more severe on a mdx background (U. Mayer, personal communication).
In conclusion, these data strongly suggest that integrins and DG have overlapping functions in mediating attachment of cells to ECM and in serving as the cellular receptors that influence assembly and organization of a structured basement membrane. The widespread expression of these receptors further suggests that lessons learned from the study of muscle fibers may also be valid for several other tissues that express DG and integrins. Of particular interest for future analysis are the brain, kidney, and lung.
 |
Does DG have a role in cell signaling?
|
|---|
So far we have discussed DG as a molecule that contributes to the structural integrity of cells and to the formation of basement membranes. The fact that DG interacts with its ligands via its carbohydrate moiety has left the question of DG-activated signaling open for a long time. Nevertheless, there are several features of the DGC that suggest a role in transmembrane signaling. For instance, ß-DG interacts directly via src homology 2/3-like domains with Grb2, a small GDP/GTP-exchanging molecule known to be involved in signal transduction. Another candidate that might be part of cell signaling processes is neuronal nitric oxide synthase, which is indirectly placed in the vicinity of DG by its interaction with the syntrophins. In addition, syntrophin-ß2, an isoform predominantly found at postsynaptic specializations of the NMJ, binds to the serine/threonine-kinase MAST (Fig. 1
; Ref. 9). Thus it is tempting to speculate that ligand binding to
-DG activates these signaling molecules. This "signaling hypothesis' received great attention when it was found that agrin not only binds to
-DG via its LG domain but is also the "key molecule" responsible for the differentiation of synaptic elements of the NMJ. This prompted speculations that DG may convey transmembrane signaling from the basal lamina to the muscle cell's cytosol, culminating in the aggregation of acetylcholine receptors (AChRs), one of the hallmarks of the mature NMJ. However, it is now clear that binding of agrin to the
-DG is not necessary for agrin-mediated AChR aggregation. The most compelling evidence stems from experiments in which truncated agrin fragments lacking the
-DG binding domain were used. Such small carboxy terminal agrin fragments still induced the formation of AChR aggregates on cultured myotubes and in fully innervated muscle in vivo. This clearly demonstrated that the interaction between agrin and DG is not required for initial agrin-mediated steps in synapse formation.
The question remains as to whether or not DG-mediated signaling can occur and to what extent it might contribute to the maturation of the NMJ. Such signaling might be initiated by the
-DG ligand laminin. Indeed, it can be demonstrated that stimulation of cultured muscle cells with laminin-1 not only causes the aggregation of AChRs (14) but also increases muscle cells' AChR content (11). A putative laminin/
-DG-mediated cell signaling is also supported by antibody perturbation experiments in which the interaction between
-DG and laminin is blocked and the spontaneous aggregation of AChRs is reduced (1). In addition, DG-mediated transmembrane signaling might not be limited to laminin but could also be initiated by certain agrin isoforms. This hypothesis stems form experiments in which cultured muscle cells were incubated with recombinant agrin fragments that can still bind to
-DG but cannot activate the normal signaling cascade for AChR aggregation involving the muscle-specific receptor tyrosine kinase MuSK. These agrin isoforms specifically reduced the level of phosphorylation of AChR
- and
-subunits (11), an event that clearly requires cell signaling processes.
There is increasing circumstantial evidence that in muscle cells the
- and ß-DG as part of the larger DGC are well placed to mediate transmembrane cell signaling. The detailed understanding of these cell-signaling events during the formation and maintenance of the NMJ may also help to clarify the contribution of DG in human pathology. The recent discovery that LFV and Mycobacterium leprae, two of the most feared pathogens in Africa and Asia, directly or indirectly use DG as an entry point to the host cell has certainly placed
- and ß-DG into the spotlight of human pathology once again. The recent advancement in techniques allowing the spatiotemporally controlled inactivation or overexpression of genes will help to elucidate the role for this receptor complex in specific physiological and pathological contexts.
 |
Acknowledgments
|
|---|
Our research is supported by the Swiss Foundation for Research on Muscle Diseases, Aktion "Benni & Co" (T. Meier), and the Swiss National Science Foundation (M. A. Ruegg).
 |
References
|
|---|
-
Brown SC, Fassati A, Popplewell L, Page AM, Henry MD, Campbell KP, and Dickson G. Dystrophic phenotype induced in vitro by antibody blockade of muscle
-dystroglycan-laminin interaction. J Cell Sci 112: 209216, 1999.[Abstract]
-
Cao W, Henry MD, Borrow P, Yamada H, Elder JH, Ravkov EV, Nichol ST, Compans RW, Campbell KP, and Oldstone MB. Identification of
-dystroglycan as a receptor for lymphocytic choriomeningitis virus and Lassa fever virus. Science 282: 20792081, 1998.[Abstract/Free Full Text]
-
Duclos F, Straub V, Moore SA, Venzke DP, Hrstka RF, Crosbie RH, Durbeej M, Lebakken CS, Ettinger AJ, van der Meulen J, Holt KH, Lim LE, Sanes JR, Davidson BL, Faulkner JA, Williamson R, and Campbell KP. Progressive muscular dystrophy in
-sarcoglycan-deficient mice. J Cell Biol 142: 14611471, 1998.[Abstract/Free Full Text]
-
Ervasti JM, Burwell AL, and Geissler AL. Tissue-specific heterogeneity in
-dystroglycan sialoglycosylation. Skeletal muscle
-dystroglycan is a latent receptor for Vicia villosa agglutinin b4 masked by sialic acid modification. J Biol Chem 272: 2231522321, 1997.[Abstract/Free Full Text]
-
Gesemann M, Brancaccio A, Schumacher B, and Ruegg MA. Agrin is a high-affinity binding protein of dystroglycan in non-muscle tissue. J Biol Chem 273: 600605, 1998.[Abstract/Free Full Text]
-
Henry MD and Campbell KP. A role for dystroglycan in basement membrane assembly. Cell 95: 859870, 1998.[ISI][Medline]
-
Hohenester E, Tisi D, Talts JF, and Timpl R. The crystal structure of a laminin G-like module reveals the molecular basis of
-dystroglycan binding to laminins, perlecan, and agrin. Mol Cell 4: 783792, 1999.[ISI][Medline]
-
Holt KH, Lim LE, Straub V, Venzke DP, Duclos F, Anderson RD, Davidson BL, and Campbell KP. Functional rescue of the sarcoglycan complex in the BIO 14.6 hamster using
-sarcoglycan gene transfer. Mol Cell 1: 841848, 1998.[ISI][Medline]
-
Lumeng C, Phelps S, Crawford GE, Walden PD, Barald K, and Chamberlain JS. Interactions between ß2-syntrophin and a family of microtubule-associated serine/threonine kinases. Nat Neurosci 2: 611617, 1999.[ISI][Medline]
-
Mayer U, Saher G, Fassler R, Bornemann A, Echtermeyer F, Von der Mark H, Miosge N, Poschl E, and Von der Mark K. Absence of integrin
7 causes a novel form of muscular dystrophy. Nat Genet 17: 318323, 1997.[ISI][Medline]
-
Meier T, Ruegg MA, and Wallace BG. Muscle-specific agrin isoforms reduce phosphorylation of AChR
and
subunits in cultured muscle cells. Mol Cell Neurosci 11: 206216, 1998.[Medline]
-
Ozowa E, Noguchi S, Mizuno Y, Hagiwara Y, and Yoshida M. From dystrophinopathy to sarcoglycanopathy: evolution of a concept of muscular dystrophy. Muscle Nerve 21: 421438, 1998.[ISI][Medline]
-
Rambukkana A, Yamada H, Zanazzi G, Mathus T, Salzer JL, Yurchenco PD, Campbell KP, and Fischetti VA. Role of
-dystroglycan as a Schwann cell receptor for Mycobacterium leprae. Science 282: 20762079, 1998.[Abstract/Free Full Text]
-
Sugiyama JE, Glass DJ, Yancopoulos GD, and Hall ZW. Laminin-induced acetylcholine receptor clustering: an alternative pathway. J Cell Biol 139: 181191, 1997.[Abstract/Free Full Text]
-
Talts JF, Andac Z, Gohring W, Brancaccio A, and Timpl R. Binding of the G domains of laminin
1 and
2 chains and perlecan to heparin, sulfatides,
-dystroglycan and several extracellular matrix proteins. EMBO J 18: 863870, 1999.[ISI][Medline]
-
Taverna D, Disatnik MH, Rayburn H, Bronson RT, Yang J, Rando TA, and Hynes RO. Dystrophic muscle in mice chimeric for expression of
5 integrin. J Cell Biol 143: 849859, 1998.[Abstract/Free Full Text]
-
Tinsley J, Deconinck N, Fisher R, Kahn D, Phelps S, Gillis JM, and Davies K. Expression of full-length utrophin prevents muscular dystrophy in mdx mice. Nat Med 4: 14411444, 1998.[ISI][Medline]
This article has been cited by other articles:

|
 |

|
 |
 
R. Hallmann, N. Horn, M. Selg, O. Wendler, F. Pausch, and L. M. Sorokin
Expression and Function of Laminins in the Embryonic and Mature Vasculature
Physiol Rev,
July 1, 2005;
85(3):
979 - 1000.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
E. A. Albrecht, A. M. Chinnaiyan, S. Varambally, C. Kumar-Sinha, T. R. Barrette, J. V. Sarma, and P. A. Ward
C5a-Induced Gene Expression in Human Umbilical Vein Endothelial Cells
Am. J. Pathol.,
March 1, 2004;
164(3):
849 - 859.
[Abstract]
[Full Text]
[PDF]
|
 |
|