Laminin-induced Acetylcholine Receptor Clustering: An Alternative Pathway

The induction of acetylcholine receptor (AChR) clustering by neurally released agrin is a critical, early step in the formation of the neuromuscular junction. Laminin, a component of the muscle fiber basal lamina, also induces AChR clustering. We find that induction of AChR clustering in C2 myotubes is specific for laminin-1; neither laminin-2 (merosin) nor laminin-11 (a synapse-specific isoform) are active. Moreover, laminin-1 induces AChR clustering by a pathway that is independent of that used by neural agrin. The effects of laminin-1 and agrin are strictly additive and occur with different time courses. Most importantly, laminin- 1–induced clustering does not require MuSK, a receptor tyrosine kinase that is part of the receptor complex for agrin. Laminin-1 does not cause tyrosine phosphorylation of MuSK in C2 myotubes and induces AChR clustering in myotubes from MuSK−/− mice that do not respond to agrin. In contrast to agrin, laminin-1 also does not induce tyrosine phosphorylation of the AChR, demonstrating that AChR tyrosine phosphorylation is not required for clustering in myotubes. Laminin-1 thus acts by a mechanism that is independent of that used by agrin and may provide a supplemental pathway for AChR clustering during synaptogenesis.     

Alpha-galactosidase stimulates acetylcholine receptor aggregation in skeletal muscle cells via PNA-binding carbohydrates

Aggregation of nicotinic acetylcholine receptors (AChRs) in skeletal muscle is an essential step in the formation of the mammalian neuromuscular junction. While proteins that bind to myotube receptors such as agrin and laminin can stimulate AChR aggregation in cultured myotubes, removal of cell surface sialic acids stimulates aggregation in a ligand-independent manner. Here, we show that removal of cell surface alpha-galactosides also stimulates AChR aggregation in the absence of added laminin or agrin. AChR aggregation stimulated by alpha-galactosidase was blocked by peanut agglutinin (PNA), which binds to lactosamine-containing disaccharides, but not by the GalNAc-binding lectin Vicia villosa agglutinin (VVA-B4). AChR aggregation stimulated by alpha-galactosidase potentiated AChR clustering induced by either neural agrin or laminin-1 and could be inhibited by muscle agrin. These data suggest that capping of cell surface lactosamines or N-acetyllactosamines with alpha-galactose affects AChR aggregation much as capping with sialic acids does.     

Acetylcholine receptor-aggregating activity of agrin isoforms and mapping of the active site

Agrin is a basal lamina protein that induces aggregation of acetylcholine receptors (AChRs) and other molecules at the developing neuromuscular junction. Alternative splicing of chick agrin mRNA at two sites, A and B, gives rise to eight possible isoforms of which five are expressed in vivo. Motor neurons express high levels of isoforms with inserts at sites A and B, muscle cells synthesize isoforms that lack amino acids at the B-site. To obtain further insights into the mechanism of agrin-induced AChR aggregation, we have determined the EC50 (effective concentration to induce half-maximal AChR clustering) of each agrin isoform and of truncation mutants. On chick myotubes, EC50 of the COOH-terminal, 95-kD fragment of agrinA4B8 was approximately 35 pM, of agrinA4B19 approximately 110 pM and of agrinA4B11 approximately 5 nM. While some AChR clusters were observed with 64 nM of agrinA4B0, no activity was detected for agrinA0B0. Recombinant full-length chick agrin and a 100-kD fragment of ray agrin showed similar EC50 values. A 45-kD, COOH-terminal fragment of agrinA4B8 retained high activity (EC50 approximately equal to 130 pM) and a 21-kD fragment was still active, but required higher concentrations (EC50 approximately equal to 13 nM). Unlike the 45-kD fragment, the 21-kD fragment neither bound to heparin nor did heparin inhibit its capability to induce AChR aggregation. These data show quantitatively that agrinA4B8 and agrinA4B19, expressed in motor neurons, are most active, while no activity is detected in agrinA0B0, the dominant isoform synthesized by muscle cells. Furthermore, our results show that a fragment comprising site B8 and the most COOH- terminal G-like domain is sufficient for this activity, and that agrin domains required for binding to heparin and those for AChR aggregation are distinct from each other.     

Laminin-1 redistributes postsynaptic proteins and requires rapsyn,tyrosine phosphorylation,and Src and Fyn to stably cluster acetylcholine receptors

Clustering of acetylcholine receptors (AChRs) is a critical step in neuromuscular synaptogenesis, and is induced by agrin and laminin which are thought to act through different signaling mechanisms. We addressed whether laminin redistributes postsynaptic proteins and requires key elements of the agrin signaling pathway to cause AChR aggregation. In myotubes, laminin-1 rearranged dystroglycans and syntrophins into a laminin-like network, whereas inducing AChR-containing clusters of dystrobrevin, utrophin, and, to a marginal degree, MuSK. Laminin-1 also caused extensive coclustering of rapsyn and phosphotyrosine with AChRs, but none of these clusters were observed in rapsyn -/- myotubes. In parallel with clustering, laminin-1 induced tyrosine phosphorylation of AChR beta and delta subunits. Staurosporine and herbimycin, inhibitors of tyrosine kinases, prevented laminin-induced AChR phosphorylation and AChR and phosphotyrosine clustering, and caused rapid dispersal of clusters previously induced by laminin-1. Finally, laminin-1 caused normal aggregation of AChRs and phosphotyrosine in myotubes lacking both Src and Fyn kinases, but these clusters dispersed rapidly after laminin withdrawal. Thus, laminin-1 redistributes postsynaptic proteins and, like agrin, requires tyrosine kinases for AChR phosphorylation and clustering, and rapsyn for AChR cluster formation, whereas cluster stabilization depends on Src and Fyn. Therefore, the laminin and agrin signaling pathways overlap intracellularly, which may be important for neuromuscular synapse formation.     

Electron microscopic structure of agrin and mapping of its binding site in laminin-1.

Agrin is a large, multidomain heparan sulfate proteoglycan that is associated with basement membranes of several tissues. Particular splice variants of agrin are essential for the formation of synaptic structures at the neuromuscular junction. The binding of agrin to laminin appears to be required for its localization to synaptic basal lamina and other basement membranes. Here, electron microscopy was used to determine the structure of agrin and to localize its binding site in laminin-1. Agrin appears as an approximately 95 nm long particle that consists of a globular, N-terminal laminin-binding domain, a central rod predominantly formed by the follistatin-like domains and three globular, C-terminal laminin G-like domains. In a few cases, heparan sulfate glycosaminoglycan chains were seen emerging from the central portion of the core protein. Moreover, we show that agrin binds to the central region of the three-stranded, coiled-coil oligomerization domain in the long arm of laminin-1, which mediates subunit assembly of the native laminin molecule. In summary, our data show for the first time a protein-protein interaction of the extracellular matrix that involves a coiled-coil domain, and they assign a novel role to this domain of laminin-1. Based on this, we propose that agrin associates with basal lamina in a polarized way.     

Interaction of agrin with laminin requires a coiled-coil conformation of the agrin-binding site within the laminin gamma1 chain

Coiled-coil domains are found in a wide variety of proteins, where they typically specify subunit oligomerization. Recently, we have demonstrated that agrin, a multidomain heparan sulfate proteoglycan with a crucial role in the development of the nerve-muscle synapse, binds to the three-stranded coiled-coil domain of laminin-1. The interaction with laminin mediates the integration of agrin into basement membranes. Here we characterize the binding site within the laminin-1 coiled coil in detail. Binding assays with individual laminin-1 full-length chains and fragments revealed that agrin specifically interacts with the gamma1 subunit of laminin-1, whereas no binding to alpha1 and beta1 chains was detected. By using recombinant gamma1 chain fragments, we mapped the binding site to a sequence of 20 residues. Furthermore, we demonstrate that a coiled-coil conformation of this binding site is required for its interaction with agrin. The finding that recombinant gamma1 fragments bound at least 10-fold less than native laminin-1 indicates that the structure of the three-stranded coiled-coil domain of laminin is required for high-affinity agrin binding. Interestingly, no binding to a chimeric gamma2 fragment was observed, indicating that the interaction of agrin with laminin is isoform specific.     

Basal lamina components are concentrated in premuscle masses and at early acetylcholine receptor clusters in chick embryo hindlimb muscles

As an initial step in characterizing the function of basal lamina components during muscle cell differentiation and innervation in vivo, we have determined immunohistochemically the pattern of expression of three components--laminin, proteins related to agrin (an acetylcholine receptor (AChR)-aggregating protein), and a heparan sulfate proteoglycan--during the development of chick embryo hindlimb muscles. Monoclonal antibodies against agrin were used to purify the protein from the Torpedo ray and to characterize agrin-like proteins from embryonic and adult chicken. In early hindlimb buds (stage 19), antibodies against laminin and agrin stained the ectodermal basement membrane and bound to limb mesenchyme with a generalized, punctate distribution. However, as dorsal and ventral premuscle masses condensed (stage 22-23), mesenchymal immunoreactivity for laminin and agrin-like proteins, but not the proteoglycan, became concentrated in these myogenic regions. Significantly, the preferential accumulation of these molecules in myogenic regions of the limb preceded by 1-2 days the appearance of muscle-specific proteins, myoblast fusion, and muscle innervation. All three basal lamina components were preferentially associated with all AChR clusters from the time we first observed them on newly formed myotubes at stage 26. Localization of these antigens in three-dimensional collagen gel cultures of limb mesenchyme, explanted prior to innervation of the limb, paralleled the staining patterns seen during limb development in the embryo. These results indicate that basal lamina molecules intrinsic to limb mesenchyme are early markers for myogenic and synaptic differentiation, and suggest that these components play important roles during the initial phases of myogenesis and synaptogenesis.     

Conjugation of LG domains of agrins and perlecan to polymerizing laminin-2 promotes acetylcholine receptor clustering

Neuromuscular junction (NMJ) assembly is characterized by the clustering and neuronal alignment of acetylcholine receptors (AChRs). In this study we have addressed post-synaptic contributions to assembly that may arise from the NMJ basement membrane with cultured myotubes. We show that the cell surface-binding LG domains of non-neural (muscle) agrin and perlecan promote AChR clustering in the presence of laminin-2. This type of AChR clustering occurs with a several hour lag, requires muscle-specific kinase (MuSK), and is accompanied by tyrosine phosphorylation of MuSK and betaAChR. It also requires conjugation of the agrin or perlecan to laminin together with laminin polymerization. Furthermore, AChR clustering can be mimicked with antibody binding to non-neural agrin, supporting a mechanism of ligand aggregation. Neural agrin, in addition to its unique ability to cluster AChRs through its B/z sequence insert, also exhibits laminin-dependent AChR clustering, the latter enhancing and stabilizing its activity. Finally, we show that type IV collagen, which lacks clustering activity on its own, stabilizes laminin-dependent AChR clusters. These findings provide evidence for cooperative and partially redundant MuSK-dependent functions of basement membrane in AChR assembly that can enhance neural agrin activity yet operate in its absence. Such interactions may contribute to the assembly of aneural AChR clusters that precede neural agrin release as well as affect later NMJ development.     

Effects of purified recombinant neural and muscle agrin on skeletal muscle fibers in vivo.

Aggregation of acetylcholine receptors (AChRs) in muscle fibers by nerve-derived agrin plays a key role in the formation of neuromuscular junctions. So far, the effects of agrin on muscle fibers have been studied in culture systems, transgenic animals, and in animals injected with agrin--cDNA constructs. We have applied purified recombinant chick neural and muscle agrin to rat soleus muscle in vivo and obtained the following results. Both neural and muscle agrin bind uniformly to the surface of innervated and denervated muscle fibers along their entire length. Neural agrin causes a dose-dependent appearance of AChR aggregates, which persist > or = 7 wk after a single application. Muscle agrin does not cluster AChRs and at 10 times the concentration of neural agrin does not reduce binding or AChR-aggregating activity of neural agrin. Electrical muscle activity affects the stability of agrin binding and the number, size, and spatial distribution of the neural agrin--induced AChR aggregates. Injected agrin is recovered from the muscles together with laminin and both proteins coimmunoprecipitate, indicating that agrin binds to laminin in vivo. Thus, the present approach provides a novel, simple, and efficient method for studying the effects of agrin on muscle under controlled conditions in vivo.     

AChR phosphorylation and aggregation induced by an agrin fragment that lacks the binding domain for alpha-dystroglycan.

Agrin induces both phosphorylation and aggregation of nicotinic acetylcholine receptors (AChRs) when added to myotubes in culture, apparently by binding to a specific receptor on the myotube surface. One such agrin receptor is alpha-dystroglycan, although binding to alpha-dystroglycan appears not to mediate AChR aggregation. To determine whether agrin-induced AChR phosphorylation is mediated by alpha-dystroglycan or by a different agrin receptor, fragments of recombinant agrin that differ in affinity for alpha-dystroglycan were examined for their ability to induce AChR phosphorylation and aggregation in mouse C2 myotubes. The carboxy-terminal 95 kDa agrin fragment agrin-c95(A0B0), which binds to alpha-dystroglycan with high affinity, failed to induce AChR phosphorylation and aggregation. In contrast, agrin-c95(A4B8) which binds less strongly to alpha-dystroglycan, induced both phosphorylation and aggregation, as did a small 21 kDa fragment of agrin, agrin-c21(B8), that completely lacks the binding domain for alpha-dystroglycan. We conclude that agrin-induced AChR phosphorylation and aggregation are triggered by an agrin receptor that is distinct from alpha-dystroglycan.     

An amino-terminal extension is required for the secretion of chick agrin and its binding to extracellular matrix

Agrin is an extracellular matrix (ECM) protein with a calculated relative molecular mass of more than 200 kD that induces the aggregation of acetylcholine receptors (AChRs) at the neuromuscular junction. This activity has been mapped to its COOH terminus. In an attempt to identify the functions of the NH2-terminal end, we have now characterized full-length chick agrin. We show that chick agrin encoded by a previously described cDNA is not secreted from transfected cells. Secretion is achieved with a construct that includes an additional 350 bp derived from the 5' end of chick agrin mRNA. Recombinant agrin is a heparan sulfate proteoglycan (HSPG) of more than 400 kD with glycosaminoglycan side chains attached only to the NH2-terminal half. Endogenous agrin in tissue homogenates also has an apparent molecular mass of > 400 kD. While the amino acid sequence encoded by the 350-bp extension has no homology to published rat agrin, it includes a stretch of 15 amino acids that is 80% identical to a previously identified bovine HSPG. The extension is required for binding of agrin to ECM. AChR aggregates induced by recombinant agrin that includes the extension are considerably smaller than those induced by agrin fragments, suggesting that binding of agrin to ECM modulates the size of receptor clusters. In addition, we found a site encoding seven amino acids at the NH2-terminal end of agrin that is alternatively spliced. While motor neurons express the splice variant with the seven amino acid long insert, muscle cells mainly synthesize isoforms that lack this insert. In conclusion, the cDNAs described here code for chick agrin that has all the characteristics previously allocated to endogenous agrin.     

Agrin released by motor neurons induces the aggregation of acetylcholine receptors at neuromuscular junctions.

To test the hypothesis that agrin mediates motor neuron-induced aggregation of acetylcholine receptors (AChRs) in skeletal muscle fibers and to determine whether the agrin active in this process is released by motor neurons, we raised polyclonal antibodies to purified ray agrin that blocked its receptor aggregating activity. When the antibodies were applied to chick motor neuron--chick myotube cocultures, they inhibited the formation of AChR aggregates at and near neuromuscular contacts, demonstrating that agrin plays a role in the induction of the aggregates. Rat motor neurons, like chick motor neurons, induce AChR aggregates on chick myotubes. This effect was not inhibited by our antibodies, indicating that, although the antibodies inhibited the activity of chick agrin, they did not have a similar effect on rat agrin. We conclude that agrin released by rat motor neurons induced the chick myotubes to aggregate AChRs.     

Matrix metalloproteinase‐3 removes agrin from synaptic basal lamina

Agrin, a heparin sulfate proteoglycan, is an integral member of the synaptic basal lamina and plays a critical role in the formation and maintenance of the neuromuscular junction. The N‐terminal region of agrin binds tightly to basal lamina, while the C‐terminal region interacts with a muscle‐specific tyrosine kinase (MuSK) to induce the formation of the postsynaptic apparatus. Although the binding of agrin to basal lamina is tight, the binding of agrin to MuSK has yet to be shown; therefore, basal lamina binding is critical for maintaining the presentation of agrin to MuSK. Here we report evidence that supports our hypothesis that matrix metalloproteinase‐3 (MMP‐3) is responsible for the removal of agrin from synaptic basal lamina. Antibodies to the hinge region of human MMP‐3 recognize molecules concentrated at the frog neuromuscular junction in both cross sections and whole mounts. Electron microscopy of neuromuscular junctions stained with antibodies to MMP‐3 reveals that staining is found in the extracellular matrix surrounding the Schwann cell. Treatment of sections from frog anterior tibialis muscle with MMP‐3 results in a clear and reproducible removal of agrin immunoreactivity from synaptic basal lamina. The same MMP‐3 treatment does not alter anti‐laminin staining. These results support our hypothesis that synaptic activity results in the activation of MMP‐3 at the neuromuscular junction and that MMP‐3 specifically removes agrin from synaptic basal lamina. © 2000 John Wiley & Sons, Inc. J Neurobiol 43: 140–149, 2000     

Synaptic localization and axonal targeting of agrin secreted by ventral spinal cord neurons in culture

Agrin secreted by motor neurons is a critical signal for postsynaptic differentiation at the developing neuromuscular junction. We used cultures of chick ventral spinal cord neurons with rat myotubes and immunofluorescence with species-specific antibodies to determine the distribution of agrin secreted by neurons and compare it to the distribution of agrin secreted by myotubes. In addition, we determined the distribution of agrin secreted by isolated chick ventral spinal cord neurons and rat motor neurons grown on a substrate that binds agrin. In cocultures, neuronal agrin was concentrated along axons at sites of axon-induced acetylcholine receptor (AChR) aggregation and was found at every such synaptic site, consistent with its role in synaptogenesis. Smaller amounts of agrin were found on dendrites and cell bodies and rarely were associated with AChR aggregation. Muscle agrin, recognized by an antibody against rat agrin, was found at nonsynaptic sites of AChR aggregation but was not detected at synaptic sites, in contrast to neuronal agrin. In cultures of isolated chick neurons or rat motor neurons, agrin was deposited relatively uniformly around axons and dendrites during the first 2-3 days in culture. In older cultures, agrin immunoreactivity was markedly more intense around axons than dendrites, indicating that motor neurons possess an intrinsic, developmentally regulated program to target agrin secretion to axons.     

LRP4 serves as a coreceptor of agrin

Neuromuscular junction (NMJ) formation requires agrin, a factor released from motoneurons, and MuSK, a transmembrane tyrosine kinase that is activated by agrin. However, how signal is transduced from agrin to MuSK remains unclear. We report that LRP4, a low-density lipoprotein receptor (LDLR)-related protein, is expressed specifically in myotubes and binds to neuronal agrin. Its expression enables agrin binding and MuSK signaling in cells that otherwise do not respond to agrin. Suppression of LRP4 expression in muscle cells attenuates agrin binding, agrin-induced MuSK tyrosine phosphorylation, and AChR clustering. LRP4 also forms a complex with MuSK in a manner that is stimulated by agrin. Finally, we showed that LRP4 becomes tyrosine-phosphorylated in agrin-stimulated muscle cells. These observations indicate that LRP4 is a coreceptor of agrin that is necessary for MuSK signaling and AChR clustering and identify a potential target protein whose mutation and/or autoimmunization may cause muscular dystrophies.     

Lithium inhibits a late step in agrin‐induced AChR aggregation

Agrin activates an intracellular signaling pathway to induce the formation of postsynaptic specializations on muscle fibers. In myotubes in culture, this pathway has been shown to include autophosphorylation of the muscle‐specific kinase MuSK, activation of Src‐family kinases, tyrosine phosphorylation of the acetylcholine receptor (AChR) β subunit, a decrease in receptor detergent extractability, and the accumulation of AChRs into high‐density aggregates. Here we report that treating chick myotubes with lithium prevented any detectable agrin‐induced change in AChR distribution without affecting the number of AChRs or the agrin‐induced change in AChR tyrosine phosphorylation and detergent extractability. Lithium treatment also increased the rate at which AChR aggregates disappeared when agrin was removed. The effects of lithium developed slowly over the course of approximately 12 h. Thus, sensitivity to lithium identifies a late step in the agrin signaling pathway, after agrin‐induced MuSK and AChR phosphorylation, that is necessary for the recruitment of AChRs into visible aggregates. © 2002 Wiley Periodicals, Inc. J Neurobiol 54: 346–357, 2003     

Role of muscle fibroblasts in the deposition of type-IV collagen in the basal lamina of myotubes

In cell cultures of quail, chick, or mouse skeletal muscle, both myogenic and fibrogenic cells synthesize and secrete type-IV collagen, a major structural component of the basal lamina. Type-IV collagen, together with laminin, forms characteristic patches and strands on the surface of developing myotubes, marking the onset of basement-membrane formation. The pattern for type-IV collagen and laminin is unique to these proteins and is not paralleled by other matrix proteins, such as fibronectin or type-I or -III collagen. In the present study, we used species-specific antibodies to either mouse or chick type-IV collagen to demonstrate the ability of fibroblast--derived type-IV collagen to incorporate in the basal lamina of myotubes. In combination cultures of embryonic quail skeletal myoblasts and mouse muscle fibroblasts, antibodies specific for mouse type-IV collagen revealed the deposition of type-IV collagen on the surface of quail myotubes in the pattern typical of the beginning of basement-membrane formation. Control cultures consisting of only quail muscle cells containing myoblasts and fibroblasts demonstrated no such reaction with these antibodies. Deposits of mouse type-IV collagen were also observed on the surface of quail myotubes when conditioned medium from mouse muscle fibroblasts was added to quail myoblast cultures. Similarly, in combination cultures of mouse myoblasts and chick muscle fibroblasts, chick type-IV-collagen deposits were identified on the surface of mouse myotubes. These results indicate that type-IV collagen synthesized by muscle fibroblasts may be incorporated into the basal lamina forming on the plasmalemma of myotubes, and may explain ultrastructural studies by Lipton on the contribution of fibroblasts to the formation of basement membranes in skeletal muscle.     

The dystroglycan complex is necessary for stabilization of acetylcholine receptor clusters at neuromuscular junctions and formation of the synaptic basement membrane

The dystrophin-associated protein (DAP) complex spans the sarcolemmal membrane linking the cytoskeleton to the basement membrane surrounding each myofiber. Defects in the DAP complex have been linked previously to a variety of muscular dystrophies. Other evidence points to a role for the DAP complex in formation of nerve-muscle synapses. We show that myotubes differentiated from dystroglycan-/- embryonic stem cells are responsive to agrin, but produce acetylcholine receptor (AChR) clusters which are two to three times larger in area, about half as dense, and significantly less stable than those on dystroglycan+/+ myotubes. AChRs at neuromuscular junctions are similarly affected in dystroglycan-deficient chimeric mice and there is a coordinate increase in nerve terminal size at these junctions. In culture and in vivo the absence of dystroglycan disrupts the localization to AChR clusters of laminin, perlecan, and acetylcholinesterase (AChE), but not rapsyn or agrin. Treatment of myotubes in culture with laminin induces AChR clusters on dystroglycan+/+, but not -/- myotubes. These results suggest that dystroglycan is essential for the assembly of a synaptic basement membrane, most notably by localizing AChE through its binding to perlecan. In addition, they suggest that dystroglycan functions in the organization and stabilization of AChR clusters, which appear to be mediated through its binding of laminin.     

Muscle-derived agrin in cultured myotubes: expression in the basal lamina and at induced acetylcholine receptor clusters.

The synaptic basal lamina (SBL) directs key aspects of the differentiation of regenerating neuromuscular junctions. A range of experiments indicate that agrin or a closely related molecule is stably associated with the SBL and participates in inducing the formation of the postsynaptic apparatus after damage to adult muscle. The selective concentration of agrin-related molecules in the SBL suggests that agrin is secreted locally by cellular components of the nerve-muscle synapse. In vivo studies on aneural embryonic muscle indicate that the muscle cell is one source of the agrin-like molecules in the SBL. Here we have used cultured chick muscle cells to study the expression of agrin-related molecules in the absence of innervation. Immunofluorescence and immunoelectron microscopy show that myogenic cells in culture express agrin-related molecules on their surfaces, and that at least a subset of these molecules are associated with the basal lamina. Moreover, in short term cultures agrin-like molecules accumulate on the surfaces of myogenic cells grown in unsupplemented basal media. We quantified the expression of agrin-like molecules on the cell surface using a solid-phase radioimmune assay. The expression of these molecules is relatively low during the first 6 days of culture and increases fourfold during the second week. The stimulation of the expression of agrin-related molecules in these long-term cultures requires the presence of chick embryo extract or fetal calf serum. We also characterized the expression of muscle-derived agrin-like molecules at clusters of AChR. These agrin-related molecules are not consistently colocalized at spontaneous AChR aggregates; however, they are selectively concentrated at greater than or equal to 90% of the AChR clusters that are induced by Torpedo agrin. These data, together with previous results from in vivo developmental experiments indicate that the agrin-like molecules in the synaptic basal lamina are derived at least in part from the muscle cell. In addition, the expression of agrin-like molecules can be regulated by soluble factors present in CEE and FBS. Finally, the selective localization of these molecules at induced AChR clusters, taken together with their localization in the basal lamina, suggests that agrin-like molecules secreted by the muscle cell play an important role in the formation and/or the stabilization of the postsynaptic apparatus.