The juxtamembrane region of MuSK has a critical role in agrin-mediated signaling
2000; Springer Nature; Volume: 19; Issue: 1 Linguagem: Inglês
10.1093/emboj/19.1.67
ISSN1460-2075
Autores Tópico(s)Cellular Mechanics and Interactions
ResumoArticle4 January 2000free access The juxtamembrane region of MuSK has a critical role in agrin-mediated signaling Ruth Herbst Ruth Herbst Molecular Neurobiology Program, Skirball Institute, NYU Medical School, 540 First Avenue, New York, NY, 10016 USA Search for more papers by this author Steven J. Burden Corresponding Author Steven J. Burden Molecular Neurobiology Program, Skirball Institute, NYU Medical School, 540 First Avenue, New York, NY, 10016 USA Search for more papers by this author Ruth Herbst Ruth Herbst Molecular Neurobiology Program, Skirball Institute, NYU Medical School, 540 First Avenue, New York, NY, 10016 USA Search for more papers by this author Steven J. Burden Corresponding Author Steven J. Burden Molecular Neurobiology Program, Skirball Institute, NYU Medical School, 540 First Avenue, New York, NY, 10016 USA Search for more papers by this author Author Information Ruth Herbst1 and Steven J. Burden 1 1Molecular Neurobiology Program, Skirball Institute, NYU Medical School, 540 First Avenue, New York, NY, 10016 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (2000)19:67-77https://doi.org/10.1093/emboj/19.1.67 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info MuSK is a receptor tyrosine kinase expressed selectively in skeletal muscle and localized to neuromuscular synapses. Agrin activates MuSK and stimulates phosphorylation and clustering of acetylcholine receptors (AChRs) at synaptic sites. We expressed wild-type or mutant MuSK in MuSK−/− myotubes and identified tyrosine residues in the MuSK cytoplasmic domain that are necessary for agrin-stimulated phosphorylation and clustering of AChRs. The activation loop tyrosines and the single juxtamembrane tyrosine were found to be essential for agrin-stimulated phosphorylation and clustering of AChRs. Further, we show that the juxtamembrane tyrosine, contained within an NPXY motif, is phosphorylated in vivo by agrin stimulation. We constructed chimeras containing extracellular and transmembrane domains from MuSK and cytoplasmic sequences from TrkA and found that inclusion of 13 amino acids from the MuSK juxtamembrane region, including the NPXY motif, is sufficient to convert a phosphorylated but inactive MuSK–TrkA chimera into a phosphorylated active chimera. These data suggest that phosphorylation of the MuSK NPXY site leads to recruitment of a phosphotyrosine-binding domain-containing protein that functions to stimulate phosphorylation and clustering of AChRs. Introduction Contact between the growth cone of a motor neuron and a developing skeletal muscle cell results in an exchange of signals between the neuron and muscle cell, resulting in the induction of a highly specialized postsynaptic membrane and a highly differentiated nerve terminal (Burden, 1998; Sanes and Lichtman, 1999). There is substantial evidence that agrin, an ∼400 kDa protein that is expressed by motor neurons and concentrated in the synaptic basal lamina, has a critical role in inducing postsynaptic differentiation (McMahan, 1990). Importantly, agrin stimulates postsynaptic differentiation in muscle cells grown in cell culture, and mice lacking agrin fail to form neuromuscular synapses (Gautam et al., 1996; Ruegg and Bixby, 1998). Agrin induces postsynaptic differentiation by activating a receptor tyrosine kinase (RTK), termed MuSK, that is expressed selectively in skeletal muscle cells (Jennings et al., 1993; Ganju et al., 1995; Valenzuela et al., 1995; Glass et al., 1996). Although agrin does not bind MuSK directly, it stimulates the rapid phosphorylation of MuSK, which ultimately leads to the redistribution of previously unlocalized proteins, including acetylcholine receptors (AChRs), to synaptic sites (Glass et al., 1996). As expected for a component of an agrin receptor complex, MuSK is required for agrin to stimulate postsynaptic differentiation in muscle cells grown in cell culture, and mice lacking MuSK fail to form neuromuscular synapses (DeChiara et al., 1996). The steps that follow MuSK activation and lead to clustering of postsynaptic proteins are not known. In addition to stimulating AChR clustering, agrin stimulates tyrosine phosphorylation of the AChR β-subunit (Wallace et al., 1991; Ferns et al., 1996). A downstream kinase appears to be essential for agrin-mediated signaling, since staurosporine blocks agrin-induced AChR clustering and phosphorylation without inhibiting MuSK phosphorylation (Wallace, 1994; Fuhrer et al., 1997). Clustering and phosphorylation of AChRs, however, appear to be independent, since β-subunit phosphorylation precedes but is not required for AChR clustering (Meyer and Wallace, 1998). These data suggest that MuSK either activates or recruits a kinase(s) that has a role in catalyzing tyrosine phosphorylation of AChRs and stimulating AChR clustering. Pharmacological agents, including those that depolymerize microtubules, destabilize actin filaments, inhibit glycosylation, inhibit methylation, inhibit calmodulin, alter the cyclic nucleotide concentration or increase the activity of GTP-binding proteins, fail to affect agrin-induced AChR clustering (Wallace, 1988). Calcium is required for agrin to induce AChR clustering, but the precise role of calcium in agrin–MuSK signaling is not known (Wallace, 1988; Megeath and Fallon, 1998). Several lines of evidence support the idea that rapsyn, a 43 kDa peripheral membrane protein that is associated with AChRs, is a required intermediate on the pathway that couples MuSK activation to AChR clustering (Gautam et al., 1995). Nevertheless, it is unclear how MuSK activation leads to clustering of rapsyn. Since RTKs initiate signaling by recruiting downstream components to the activated receptor, proteins that are immediately downstream of an activated RTK can be identified by first identifying sequences in the RTK that are necessary to activate downstream signaling (Schlessinger and Ullrich, 1992; Pawson, 1995). We expressed wild-type and mutant forms of MuSK in a MuSK−/− muscle cell line and identified sequences in MuSK that are necessary for agrin to induce phosphorylation and clustering of AChRs. We show that a juxtamembrane tyrosine, Y553, in MuSK is phosphorylated by agrin stimulation and that phosphorylation of this tyrosine residue is required for agrin-mediated signaling. We show that this tyrosine is within a consensus binding site for PTB domain-containing proteins, and our data support the idea that phosphorylation of this juxtamembrane tyrosine residue serves to recruit a PTB domain-containing protein to activated MuSK. Results Forced expression of MuSK in MuSK−/− myotubes restores agrin signaling Ligand-stimulated activation of RTKs leads to phosphorylation of tyrosine residues in the activation loop of the kinase domain as well as phosphorylation of tyrosine residues outside the activation loop (Hubbard et al., 1994). These phosphotyrosine-containing sequences can serve as docking sites for signaling molecules that couple RTK activation to distinct downstream signaling pathways, leading to specific cellular responses (Schlessinger and Ullrich, 1992; Pawson, 1995). In order to learn more about how MuSK activation leads to clustering of synaptic proteins, we sought to identify tyrosine residues in MuSK that are required for agrin to stimulate clustering of AChRs. We established an assay to measure MuSK function by generating MuSK-deficient muscle cell lines from embryos that were homozygous mutant for MuSK and that carried a temperature-sensitive large T oncogene (Jat et al., 1991). Cell lines could be maintained as proliferating myoblasts or induced to differentiate and form myotubes by controlling expression of large T. MuSK−/− myoblasts, like wild-type myoblasts, fuse and differentiate into multinucleated myotubes, and the morphology of wild-type and MuSK−/− myotubes is indistinguishable. Unlike myotubes from wild-type or MuSK heterozygous mice, MuSK−/− myotubes fail to cluster AChRs in response to agrin (Figure 1) (Glass et al., 1996). Myoblasts were infected with a recombinant retrovirus expressing wild-type MuSK, and transiently infected myoblasts were induced to differentiate into myotubes. Figure 1 shows that forced expression of wild-type MuSK restores agrin-stimulated AChR clustering in MuSK mutant myotubes. Like agrin-stimulated clustering of AChRs in wild-type myotubes, AChR clustering in the rescued myotubes is induced by neural (A4B19) and not non-neural (A4B0) agrin (Figure 1). Figure 1.Forced expression of wild-type MuSK in MuSK−/− myotubes restores agrin-mediated signaling. MuSK−/− myotubes and MuSK−/− myotubes infected with a retroviral vector encoding wild-type MuSK were treated with neural (A4B19) or non-neural (A4B0) agrin, and AChRs were labeled with Texas red-conjugated α-BGT. Download figure Download PowerPoint Identification of tyrosine residues required for MuSK function The cytoplasmic domain of mammalian MuSK contains 17 tyrosine residues within the kinase domain and one tyrosine residue in the juxtamembrane region (Figure 2A) (Valenzuela et al., 1995). We expressed MuSK in insect cells using a baculovirus expression vector and mapped the tyrosines that are phosphorylated in MuSK in an in vitro kinase assay using a MALDI to assign the masses of 32P-labeled tryptic peptides fractionated by HPLC (A.Watty, G.Neubauer, M.Dreger, M.Zimmer, M.Wilm and S.J.Burden, unpublished data). We found that the juxtamembrane tyrosine residue (Y553), two of the three tyrosine residues within the activation loop (Y754 and Y755) and two additional tyrosine residues within the kinase domain (Y576 and Y812) are phosphorylated in MuSK in vitro. To determine whether these tyrosine residues are important for MuSK signaling, we mutated each of these tyrosine residues and used the rescue assay to determine whether these mutated forms of MuSK could restore agrin-stimulated clustering of AChRs. Figure 2.Tyrosine residues within the activation loop and juxtamembrane region of MuSK are required for agrin-stimulated AChR clustering. (A) A cartoon showing the position of the single tyrosine residue in the juxtamembrane region of MuSK and the 17 tyrosine residues in the kinase domain of MuSK; the five tyrosine residues (Y553, Y576, Y754, Y755 and Y812) that are phosphorylated in vitro are indicated. (B) Neural (A4B19) agrin stimulates AChR clustering in MuSK−/− myotubes expressing wild-type MuSK but not MuSK Y553F. (C) The number of AChR clusters induced by agrin in MuSK−/− myotubes expressing wild-type or mutant MuSK. Mutation of the single juxtamembrane tyrosine (MuSK Y553F) or the three activation loop tyrosines (MuSK Y750,754,755F) results in a failure of neural agrin to stimulate AChR clustering. Mutation of a single tyrosine in the activation loop (Y755) or a tyrosine near the beginning of the kinase domain (Y576F) reduces the number of agrin-induced AChR clusters. Neural agrin induces a normal number (∼10 per field) of AChR clusters in MuSK−/− myotubes expressing MuSK Y812F or MuSK Y831F. Download figure Download PowerPoint Figure 2 shows that the juxtamembrane tyrosine residue is essential for downstream signaling, since myotubes expressing MuSK Y553F fail to cluster AChRs in response to agrin (Figure 2B and C). Mutation of a single tyrosine within the activation loop (Y755) reduces AChR clustering by ∼40%, whereas mutation of all three tyrosine residues (Y750, Y754 and Y755) within the activation loop abolishes AChR clustering (Figure 2C), indicating that MuSK kinase activity is required for agrin-stimulated AChR clustering. Other tyrosines within the kinase domain (Y576 and Y812) have little or no role in clustering AChRs, since myotubes expressing MuSK Y812F respond normally to agrin and myotubes expressing MuSK Y576F have 30% fewer AChR clusters than myotubes expressing wild-type MuSK (Figure 2C). Although we did not detect phosphorylation of MuSK Y831 in vitro, there is evidence that phosphorylation of this tyrosine residue is required for the function of certain RTKs (Ming et al., 1999). Figure 2 shows that Y831 is not required for MuSK function, since myotubes expressing MuSK Y831F cluster AChRs normally in response to agrin (Figure 2C). Because the level of MuSK expression from the retroviral vector is less than that of endogenous MuSK in wild-type cells, the ability of MuSK Y831F to restore agrin-mediated signaling is unlikely due to overexpression of defective MuSK. Agrin stimulates phosphorylation of MuSK Y553 To determine whether Y553 is phosphorylated in vivo following agrin stimulation, we generated antibodies against a phosphopeptide sequence in MuSK containing phosphorylated Y553. We treated C2 myotubes with agrin, immunoprecipitated MuSK and probed Western blots with the Y553 phosphopeptide antibodies. Figure 3 shows that the Y553 phosphopeptide antibodies bind poorly to MuSK from untreated myotubes and intensely to MuSK from agrin-treated myotubes (Figure 3A). Antibody labeling is specific for the Y553 phosphopeptide, since the phosphorylated Y553 peptide competes for antibody binding (Figure 3A). Furthermore, TrkA contains a similar juxtamembrane sequence (Martin-Zanca et al., 1986), but phosphorylated TrkA protein, either expressed in baculovirus and phosphorylated in an in vitro kinase assay or isolated from nerve growth factor (NGF)-stimulated PC12 cells, is not recognized by the Y553 phosphopeptide antibodies (data not shown). To demonstrate further the specificity of the Y553 phosphopeptide antibodies, we treated MuSK−/− myotubes expressing MuSK Y553F with agrin, and probed Western blots of immunoprecipitated MuSK with the Y553 phosphopeptide antibodies. The Y553 phosphopeptide antibodies bind poorly to MuSK Y553F in untreated myotubes and agrin stimulation fails to increase antibody labeling (Figure 3B). Taken together, these data indicate that the Y553 phosphopeptide antibodies are specific for MuSK and that MuSK is phosphorylated in vivo at Y553 (Figure 3). Figure 3.Agrin stimulates phosphorylation of MuSK Y553. (A) MuSK was immunoprecipitated from C2 myotubes treated with neural agrin, and Western blots were probed either with antibodies to phosphotyrosine (anti-PY) or with antibodies to a Y553 phosphopeptide (anti-pY553). Antibodies to phosphotyrosine as well as antibodies to the Y553 phosphopeptide bind to MuSK (arrow) in muscle cells stimulated with neural agrin. Labeling with antibodies to the Y553 phosphopeptide is specific since labeling is blocked by pre- incubation of the antibodies with the phosphopeptide. Western blots were re-probed with antibodies to MuSK to ensure similar loading. (B) Antibodies to the Y553 phosphopeptide (anti-pY553) fail to bind MuSK Y553F and bind poorly to MuSK Y750,754,755F expressed in MuSK−/− myotubes. (C) Neural agrin stimulates tyrosine phosphorylation of the AChR β-subunit in MuSK−/− myotubes expressing wild-type MuSK but not MuSK Y553F or MuSK Y750,754,755F. Download figure Download PowerPoint To determine whether mutation of the activation loop tyrosines results in a failure of agrin to stimulate phosphorylation of Y553, we treated MuSK−/− myotubes expressing MuSK Y750,754,755F with agrin, and we probed Western blots of immunoprecipitated MuSK Y750,754,755F with the Y553 phosphopeptide antibodies. Figure 3 shows that mutation of the activation loop tyrosines severely impairs but does not abolish phosphorylation of Y553 (Figure 3B). Although the activation loop tyrosines may have a separate role in MuSK function (Qian et al., 1998), it is possible that phosphorylation of these tyrosine residues is required to activate the kinase and phosphorylate Y553. These results demonstrate that agrin stimulation results in an increase in phosphorylation of Y553 and suggest that phosphorylation of Y553 is a critical step in MuSK function. MuSK Y553 is required for agrin-stimulated AChR phosphorylation Agrin stimulation of MuSK results in tyrosine phosphorylation of the AChR β-subunit (Wallace et al., 1991; Ferns et al., 1996). To determine whether the AChR β-subunit is tyrosine phosphorylated in myotubes expressing MuSK Y553F, we used α-bungarotoxin (α-BGT) to isolate AChRs from myotubes expressing wild-type or mutant MuSK and probed Western blots with antibodies to phosphotyrosine. Figure 3 shows that tyrosine phosphorylation of the AChR β-subunit is stimulated by agrin in myotubes expressing wild-type MuSK but not in myotubes expressing MuSK Y553F (Figure 3C). These results indicate that MuSK Y553 is required not only to cluster AChRs but also to stimulate AChR tyrosine phosphorylation. MuSK Y553F is catalytically active and expressed on the cell surface Although the failure of MuSK Y553 to rescue MuSK function may be due to a loss of Y553 phosphorylation, we considered the possibility that mutation of Y553 may have inactivated MuSK kinase activity. To determine whether mutation of Y553 reduces the kinase activity of MuSK, we expressed wild-type and mutant MuSK in insect cells and measured MuSK kinase activity in immunoprecipitates of MuSK from the baculovirus-infected cells. We measured the ability of MuSK to phosphorylate itself and to transphosphorylate a substrate peptide (Figure 4A and B). Wild-type MuSK and MuSK Y553F have low basal kinase activities that are stimulated substantially and similarly by ATP (Figure 4A and B). These results indicate that wild-type and Y553F mutant MuSK have similar kinase activities. MuSK Y750,754,755F has a higher basal kinase activity than wild-type MuSK, as expected for a destabilized activation loop (Hubbard et al., 1994), but this mutated kinase, unlike wild-type MuSK or Y553F MuSK, can not be stimulated further by ATP. Figure 4.MuSK Y553F is expressed on the cell surface and is catalytically active in vitro. (A) In vitro kinase assays show that MuSK Y553F and wild-type MuSK are phosphorylated similarly in vitro, and stimulated by pre-incubation with ATP. In contrast, phosphorylation of MuSK Y750,754,755F is not stimulated by pre-incubation with ATP; the basal level of MuSK Y750,754,755F phosphorylation is increased probably due to destabilization of the activation loop. (B) In vitro kinase assays show that the IRS-Y727 and the p(EKY) peptides are phosphorylated similarly by MuSK Y553F and wild-type MuSK but poorly by MuSK Y750,754,755F. (C) Cell surface proteins in MuSK−/− myotubes expressing wild-type MuSK or MuSK Y553F were labeled with biotin-NHS. Proteins labeled with biotin and recovered with streptavidin were considered to be on the cell surface (S), whereas proteins that were not recovered with streptavidin were considered to be cytosolic (C). Western blots show that wild-type MuSK and MuSK Y553F are expressed on the cell surface. Control experiments show that the AChR β-subunit is expressed on the cell surface and that rapsyn and Shc are largely cytosolic. Download figure Download PowerPoint Because Y553 is embedded in an NPXY motif, which can function as an internalization signal in other membrane receptors (Chen et al., 1990), we considered the possibility that the failure of MuSK Y553 to rescue MuSK function might be due to a failure of MuSK to be expressed on the cell surface. We labeled cell surface proteins in intact cells expressing wild-type or MuSK Y553F with a membrane-impermeable, biotinylated N-hydroxysuccinimide ester (NHS-biotin), lysed the cells in detergent and isolated the biotin-labeled proteins with streptavidin. NHS-biotin selectively labels cell surface proteins, since membrane proteins that are exposed on the cell surface, including AChRs, are labeled efficiently by NHS-biotin, whereas intracellular proteins, such as rapsyn and Shc, are labeled poorly by NHS-biotin (Figure 4C). Figure 4 demonstrates that MuSK Y553F, like wild-type MuSK, is expressed on the cell surface (Figure 4C). Thus, MuSK Y553F fails to restore MuSK function although the mutant protein is catalytically active and expressed on the cell surface. Y553 is within an NPXY motif that can bind proteins with PTB domains Because proteins containing phosphotyrosine-binding (PTB) domains bind phosphorylated NPXY motifs (Blaikie et al., 1994; Kavanaugh and Williams, 1994), we considered the possibility that a PTB domain-containing protein might bind the NPMY sequence in MuSK. In addition to the tyrosine residue, PTB domain-containing proteins require an asparagine at −3 and prefer a proline at −2, with respect to the tyrosine at +0 (van der Geer and Pawson, 1995). Therefore, we expressed MuSK N550A, P551A as well as Y553A in MuSK-deficient myotubes and determined whether the sequence requirements for agrin-stimulated signaling correspond to the sequence requirements for binding PTB domain-containing proteins. Figure 5 shows that myotubes expressing MuSK N550A and MuSK Y553A fail to cluster AChRs in response to agrin. Like wild-type MuSK and MuSK Y553F, MuSK N550A and MuSK Y553A are expressed on the cell surface (data not shown). Myotubes expressing MuSK P551A cluster AChRs in response to agrin, but the efficiency of agrin-stimulated AChR clustering in these myotubes is reduced (Figure 5A). Thus, there is a good correspondence between residues required for MuSK function and for binding PTB domain-containing proteins. Figure 5.The NPXY motif in MuSK is required for agrin-mediated signaling, but neural agrin fails to stimulate phosphorylation of Shc, IRS-1, IRS-2, FRS2 or components of MAP kinase or PI3-kinase signaling pathways. (A) Neural agrin fails to induce AChR clusters in MuSK−/− myotubes expressing MuSK N550A or MuSK Y553A and induces fewer AChR clusters in MuSK−/− myotubes expressing MuSK P551A. (B) Tyrosine-phosphorylated proteins were immunoprecipitated from C2 myotubes, PC12 cells or L6 myoblasts, and Western blots were probed with antibodies to Shc, IRS-1, IRS-2 or FRS2. Phosphorylation of Shc and FRS2 is stimulated in PC12 cells treated with NGF, and phosphorylation of IRS-1 and IRS-2 is stimulated in L6 myoblasts treated with insulin. (C) Western blots of lysates from C2 myotubes show that neural agrin fails to stimulate phosphorylation of p38, Akt or Jnk; agrin induces a modest increase in Erk1/2 phosphorylation. NGF stimulates phosphorylation of Erk 1/2, p38 and Akt in PC12 cells (Hempstead et al., 1992). Western blots were re-probed with antibodies to Jnk to ensure similar loading. Download figure Download PowerPoint Neither Shc, IRS-1, IRS-2 nor FRS2 are activated in agrin-stimulated myotubes Signaling proteins containing PTB domains become tyrosine phosphorylated following their recruitment to activated receptors (van der Geer and Pawson, 1995). To determine whether Shc (Pelicci et al., 1992), IRS-1/2 (Sun et al., 1991) or FRS2 (Kouhara et al., 1997) is tyrosine phosphorylated following agrin stimulation, we treated C2 myotubes with agrin, immunoprecipitated tyrosine-phosphorylated proteins with an antibody against phosphotyrosine and probed Western blots with antibodies against Shc, IRS-1, IRS-2 or FRS2. Agrin fails to stimulate tyrosine phosphorylation of Shc, IRS-1, IRS-2 or FRS2 (Figure 5B). In contrast, tyrosine phosphorylation of Shc and FRS2 is stimulated in PC12 cells treated with NGF, and tyrosine phosphorylation of IRS-1 and IRS-2 is stimulated in L6 myoblasts treated with insulin (Figure 5B). Thus, neither Shc, IRS-1, IRS-2 nor FRS2 is activated by agrin. We used a similar assay to determine whether agrin activates kinases that are often stimulated by other RTKs. We treated C2 myotubes with agrin and probed Western blots with antibodies against phosphorylated Erk1/2, JNK, p38 or AKT. Figure 5 shows that agrin fails to activate these downstream kinases (Figure 5C). Likewise, agrin fails to stimulate phosphorylation of the p85 subunit of phosphatidylinositol 3-kinase (PI3-kinase; data not shown). Consistent with these data, neither PD 09857, a MEK1 inhibitor, nor wortmanin, an inhibitor of PI3-kinase, inhibits agrin-induced AChR clustering (data not shown). The juxtamembrane region of MuSK has a role in MuSK phosphorylation Figure 6 shows that agrin fails to stimulate phosphorylation of MuSK N550A, MuSK Y553F and MuSK Y553A (Figure 6A), indicating that mutation of the NPXY motif is sufficient to impair phosphorylation of the entire kinase (Figure 6A). MuSK P551A is phosphorylated by agrin but less efficiently than wild-type MuSK (Figure 6B), and the reduction in phosphorylation parallels the reduction in AChR clustering (Figure 5A). Likewise, mutation of amino acid residues that flank the NPXY motif (MuSK DRLH[545–548]AAAA) and that have a role in binding PTB domain-containing proteins (van der Geer and Pawson, 1995) results in an ∼2-fold reduction in the level of agrin-induced MuSK phosphorylation and AChR clustering (Figure 6B; data not shown). These results indicate that the asparagine and tyrosine residues within the NPXY motif have a critical role in activating the kinase domain of MuSK and/or preventing dephosphorylation of activated MuSK in vivo, whereas the proline residue and amino acids that are N-terminal to the NPXY motif are less crucial for MuSK phosphorylation and AChR clustering. Figure 6.The NPXY motif in MuSK is required for agrin-stimulated MuSK phosphorylation. (A) Neural agrin fails to stimulate phosphorylation of MuSK in MuSK−/− myotubes expressing MuSK Y553F, MuSK Y750,754,755F, MuSK N550A or MuSK Y553A. (B) Neural agrin stimulates MuSK phosphorylation in MuSK−/− myotubes expressing MuSK DRLH[545–548]AAAA or MuSK P551A. (C) Antibodies to a FLAG epitope tag, introduced in the extracellular region of MuSK, induce phosphorylation of wild-type MuSK but fail to induce phosphorylation of MuSK N550A and MuSK Y553A. Download figure Download PowerPoint Current data support the idea that MuSK is not a direct receptor for agrin and that myotubes express an additional activity that is required for agrin to activate MuSK (Glass et al., 1996). These data have led to the suggestion that myotubes may express a protein that binds agrin, associates with MuSK and functions as a signaling complex with MuSK. Thus, it is possible that the juxtamembrane region of MuSK facilitates association between MuSK and this putative agrin-binding receptor and that mutation of the juxtamembrane region prevents this association and blocks agrin activation of MuSK. To bypass the requirement for an agrin receptor, we activated MuSK with antibodies against a FLAG epitope introduced into the extracellular domain of MuSK (Xie et al., 1997; Hopf and Hoch, 1998). Figure 6 shows that these antibodies induce phosphorylation of wild-type MuSK but not MuSK containing mutations in the NPXY motif (Figure 6C). Thus, the juxtamembrane NPXY motif is required for antibodies against MuSK, as well as agrin, to activate and/or maintain phosphorylation of MuSK. Since mutation of the NPXY motif does not reduce the in vitro kinase activity of MuSK (Figure 4A and B), these results suggest that the NPXY motif is required in vivo either to activate MuSK kinase activity or to recruit a protein that prevents a phosphatase from dephosphorylating MuSK. The following experiments provide evidence that the juxtamembrane region of MuSK has a role in clustering and phosphorylating AChRs that supersedes its role in regulating MuSK phosphorylation. The juxtamembrane NPXY motif in MuSK has a critical role in clustering and phosphorylating AChRs The cytoplasmic domains of MuSK and the Trk neurotrophin receptors are similar in sequence (Jennings et al., 1993). In particular, TrkA contains an NPXY motif in the juxtamembrane region that is critical for TrkA signaling (Stephens et al., 1994). The sequence flanking the NPXY motif in TrkA, however, conforms to a consensus binding site for Shc, and both Shc and FRS2 are recruited to this region of activated TrkA (Stephens et al., 1994; Meakin et al., 1999). To determine whether sequences in the TrkA cytoplasmic domain could substitute for sequences in the cytoplasmic domain in MuSK, we constructed MuSK–TrkA chimeras containing the extracellular and transmembrane domains from MuSK and cytoplasmic sequences from TrkA (Figure 7A). We expressed the MuSK–TrkA chimeras in MuSK−/− myotubes and asked whether the chimeras could restore agrin-mediated signaling. A chimera containing the extracellular and transmembrane domains of MuSK and the entire cytoplasmic domain of TrkA is phosphorylated by agrin stimulation (Figure 7B) but fails to induce clustering or phosphorylation of AChRs (Figure 7C and D). In contrast, the same MuSK–TrkA chimera, but including a substitution of 13 amino acids from the juxtamembrane domain of MuSK with the comparable region of TrkA, is phosphorylated by agrin stimulation (Figure 7B) and induces phosphorylation and clustering of AChRs (Figure 7C and D). Thus, 13 amino acids from the juxtamembrane region of MuSK, including the NPXY motif, are sufficient to convert a phosphorylated but inactive chimera into a phosphorylated and active chimera. These results are consistent with the idea that phosphorylation of the NPXY motif in MuSK leads to the recruitment of an adaptor protein(s) that is required for clustering and phosphorylating AChRs. Figure 7.The juxtamembrane region of MuSK has a critical role in agrin-mediated signaling. (A) The cartoon shows wild-type MuSK (MMM), a MuSK–TrkA chimera (MTT) containing the juxtamembrane (NPXY) and kinase domains (kd) of TrkA and a MuSK–TrkA chimera (MMT) containing 13 amino acids (MuSK NPXY) from the juxtamembrane domain of MuSK and the kinase domain of TrkA. (B) Neural agrin stimulates phosphorylation of MMM, MTT a
Referência(s)