Our laboratory studies the neuromuscular junction as a model of
synaptic structure and function in the nervous
system. Our recent work has focused on two areas: the mechanisms of assembly of the acetylcholine receptor (AChR) and the role in synaptogenesis of agrin, a neurally-secreted protein that causes AChRs to cluster.
The AChR is an ion channel composed of four different highly homologous polypeptide subunits (alpha subscript 2 beta delta gamma) that surround a central aqueous pore. We have investigated the assembly of the subunits in muscle cells and in COS cells and have found that assembly occurs according to a defined pathway in which the first step is the formation of the two heterodimers, alpha delta and alpha gamma. Each subunit has a stereotyped transmembrane orientation with a long, extracellular N-terminal domain, four transmembrane domains, and a large cytoplasmic loop between the third and fourth transmembrane regions. Experiments with dominant-negatives and with chimeric subunits indicate that the identify of each subunit during assembly is determined by the N-terminal domain in the lumen of the endoplasmic reticulum (ER), where AChR assembly takes place. These domains mediate the interactions leading to heterodimer formation; interactions in later steps in the pathway appear to also involve amino acid sequences in the long, cytoplasmic loops of the subunits. Before the subunits assemble, their lumenal domains must fold. Although we have found that the unfolded subunits are associated with BiP, a chaperone protein resident in the ER, the kinetics of formation of the complex that we detect is incompatible with the protein having a role in subunit folding. We have recently begun to study the folding process in an in-vitro translation system.
An early event during the formation of the neuromuscular junction is the clustering of AChRs in the postsynaptic membrane underneath the nerve terminal. The clustering occurs, at least in part, in response to a protein, agrin, that is released from motor nerve terminals. Neurons and muscle cells express different forms of agrin, which are generated by alternative RNA splicing. How agrin causes AChR clustering is unknown. We have investigated the action of agrin on cultured muscle cells in several ways. First, In collaboration with the laboratory of Richard Scheller of Stanford University, we have examined the ability of the various forms of agrin to cluster the AChRs of cultured muscle cells. The most active forms, found exclusively in neurons, are those containing an eight-amino acid insert at one of the splicing sites. Other forms of agrin, found in muscle and other tissues, are significantly less active. Second, we found that a variant of the C2 muscle cell line that is defective in the synthesis of proteoglycans is much less sensitive to agrin than is the parental line, suggesting that proteoglycans are involved in the action of agrin. Finally, we have looked for agrin-binding proteins in muscle that might mediate the action of agrin in muscle cells. Our results and those from several other labs show that the major agrin-binding protein in muscle cells is alpha-dystroglycan, an extracellular matrix protein previously identified as part of a complex of proteins associated with dystrophin. Because dystrophin is associated with actin and alpha-dystroglycan binds laminin, the complex is thought to form a transmembrane linker between the extracellular matrix and cytoskeleton. We are currently investigating whether the interaction of agrin with alpha-dystroglycan is responsible for the formation of AChR clusters, or whether it plays some other role.