Portal de Periódicos da CAPES

Neuregulin-Heparan-sulfate Proteoglycan Interactions Produce Sustained erbB Receptor Activation Required for the Induction of Acetylcholine Receptors in Muscle

2001; Elsevier BV; Volume: 276; Issue: 41 Linguagem: Inglês

10.1074/jbc.m104485200

1083-351X

Qunfang Li, Jeffrey A. Loeb,

Cancer Treatment and Pharmacology

Neuregulins bind to and activate members of the EGF receptor family of tyrosine kinases that initiate a signaling cascade that induces acetylcholine receptor synthesis in the postsynaptic membrane of neuromuscular synapses. In addition to an EGF-like domain, sufficient for receptor binding and tyrosine auto-phosphorylation, many spliced forms also have an IG-like domain that binds HSPGs and maintains a high concentration of neuregulin at synapses. Here, we show that the IG-like domain functions to keep the EGF-like domain at sufficiently high concentrations for a sufficiently long period of time necessary to induce acetylcholine receptor gene expression in primary chick myotubes. Using recombinant neuregulins with and without the IG-like domain, we found that IG-like domain binding to endogenous HSPGs produces a 4-fold increase in receptor phosphorylation. This enhancement of activity was blocked by soluble heparin or by pretreatment of muscle cells with heparitinase. We show that at least 12–24 h of neuregulin exposure was required to turn on substantial acetylcholine receptor gene expression and that the erbB receptors need to be kept phosphorylated during this time. The need for sustained erbB receptor activation may be the reason why neuregulins are so highly concentrated in the extracellular matrix of synapses. Neuregulins bind to and activate members of the EGF receptor family of tyrosine kinases that initiate a signaling cascade that induces acetylcholine receptor synthesis in the postsynaptic membrane of neuromuscular synapses. In addition to an EGF-like domain, sufficient for receptor binding and tyrosine auto-phosphorylation, many spliced forms also have an IG-like domain that binds HSPGs and maintains a high concentration of neuregulin at synapses. Here, we show that the IG-like domain functions to keep the EGF-like domain at sufficiently high concentrations for a sufficiently long period of time necessary to induce acetylcholine receptor gene expression in primary chick myotubes. Using recombinant neuregulins with and without the IG-like domain, we found that IG-like domain binding to endogenous HSPGs produces a 4-fold increase in receptor phosphorylation. This enhancement of activity was blocked by soluble heparin or by pretreatment of muscle cells with heparitinase. We show that at least 12–24 h of neuregulin exposure was required to turn on substantial acetylcholine receptor gene expression and that the erbB receptors need to be kept phosphorylated during this time. The need for sustained erbB receptor activation may be the reason why neuregulins are so highly concentrated in the extracellular matrix of synapses. neuregulin acetylcholine receptor heparan-sulfate proteoglycan epidermal growth factor immunoglobulin glyceraldehyde-3-phosphate dehydrogenase fibroblast growth factor An important means by which cells communicate with each other is through the release of growth and differentiation factors from one cell that activates membrane receptors on a nearby cell and ultimately changes its properties through changes in gene expression. Once released, many polypeptide factors have additional binding interactions with heparan-sulfate proteoglycans (HSPGs) situated in the extracellular matrix between cells. The functional consequences of this dual binding interaction are not entirely clear but have been proposed to concentrate these factors at sites where needed, to protect them from proteolysis, and to modulate their interactions with their cell-surface receptors (1Schlessinger J. Lax I. Lemmon M. Cell. 1995; 83: 357-360Abstract Full Text PDF PubMed Scopus (450) Google Scholar). What is even less clear is how these extracellular interactions modulate the intracellular events that ultimately change the properties of the cell. The neuregulins (NRGs)1 are a family of heparin-binding growth and differentiation factors with multiple functions in growth and development of the nervous system and heart and in cancer (2Fischbach G.D. Rosen K.M. Annu. Rev. Neurosci. 1997; 20: 429-458Crossref PubMed Scopus (254) Google Scholar). They are released from motor nerve endings at neuromuscular synapses and activate members of the epidermal growth factor (EGF) family of tyrosine kinase receptors erbB2, erbB3, and erbB4 in the postsynaptic muscle membrane (3Loeb J.A. Khurana T.S. Robbins J.T. Yee A.G. Fischbach G.D. Development (Camb.). 1999; 126: 781-791Crossref PubMed Google Scholar, 4Goodearl A.D. Yee A.G. Sandrock Jr., A.W. Corfas G. Fischbach G.D. J. Cell Biol. 1995; 130: 1423-1434Crossref PubMed Scopus (79) Google Scholar, 5Moscoso L.M. Chu G.C. Gautam M. Noakes P.G. Merlie J.P. Sanes J.R. Dev. Biol. 1995; 172: 158-169Crossref PubMed Scopus (160) Google Scholar, 6Zhu X. Lai C. Thomas S. Burden S.J. EMBO J. 1995; 14: 5842-5848Crossref PubMed Scopus (157) Google Scholar). This transynaptic activation results in a dramatic up-regulation of muscle acetylcholine receptors (AChRs) needed to guarantee proper synaptic transmission. NRG also promotes the transition from embryonic to adult forms of mammalian AChRs by inducing the switch to the ε AChR subunit (7Martinou J.C. Falls D.L. Fischbach G.D. Merlie J.P. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 7669-7673Crossref PubMed Scopus (121) Google Scholar) and the expression of voltage-gated sodium channels (8Corfas G. Fischbach G.D. J. Neurosci. 1993; 13: 2118-2125Crossref PubMed Google Scholar). All of these effects would be expected to increase the efficacy of synaptic transmission as the target muscle fiber increases in size and the neuromuscular junction matures. Consistently, mice in which the Type INRG1 allele is disrupted by homologous recombination exhibited a 50% reduction in the density of AChRs in the postsynaptic membrane, and when challenged by low doses of curare, a reduced safety factor for neuromuscular transmission was demonstrated (9Sandrock Jr., A.W. Dryer S.E. Rosen K.M. Gozani S.N. Kramer R. Theill L.E. Fischbach G.D. Science. 1997; 276: 599-603Crossref PubMed Scopus (245) Google Scholar). A common feature shared by all NRGs is an EGF-like domain. Even when expressed by itself, this domain is sufficient for receptor binding and activation of homo- and heterodimers of erbB2, erbB3, and erbB4 receptors, which are highly concentrated at the neuromuscular synapses in the postsynaptic muscle membrane (5Moscoso L.M. Chu G.C. Gautam M. Noakes P.G. Merlie J.P. Sanes J.R. Dev. Biol. 1995; 172: 158-169Crossref PubMed Scopus (160) Google Scholar, 6Zhu X. Lai C. Thomas S. Burden S.J. EMBO J. 1995; 14: 5842-5848Crossref PubMed Scopus (157) Google Scholar, 10Altiok N. Bessereau J.L. Changeux J.P. EMBO J. 1995; 14: 4258-4266Crossref PubMed Scopus (131) Google Scholar). The rapid autophosphorylation of these receptors on tyrosine residues initiates a signaling cascade that translates this initial binding event into the induction of AChR genes (11Corfas G. Falls D.L. Fischbach G.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1624-1628Crossref PubMed Scopus (71) Google Scholar). This signaling cascade involves a number of signaling pathways, including both the mitogen-activated protein kinase (12Si J. Luo Z. Mei L. J. Biol. Chem. 1996; 271: 19752-19759Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 13Tansey M.G. Chu G.C. Merlie J.P. J. Cell Biol. 1996; 134: 465-476Crossref PubMed Scopus (117) Google Scholar, 14Altiok N. Altiok S. Changeux J.P. EMBO J. 1997; 16: 717-725Crossref PubMed Scopus (111) Google Scholar) and phosphatidylinositol 3-kinase pathways (13Tansey M.G. Chu G.C. Merlie J.P. J. Cell Biol. 1996; 134: 465-476Crossref PubMed Scopus (117) Google Scholar). Most spliced forms of NRG also have an immunoglobulin-like (IG-like) domain NH2-terminal to the EGF-like domain (Fig. 1). We and others (3Loeb J.A. Khurana T.S. Robbins J.T. Yee A.G. Fischbach G.D. Development (Camb.). 1999; 126: 781-791Crossref PubMed Google Scholar, 15Loeb J.A. Fischbach G.D. J. Cell Biol. 1995; 130: 127-135Crossref PubMed Scopus (114) Google Scholar, 16Meier T. Masciulli F. Moore C. Schoumacher F. Eppenberger U. Denzer A.J. Jones G. Brenner H.R. J. Cell Biol. 1998; 141: 715-726Crossref PubMed Scopus (97) Google Scholar) have shown that this domain interacts with HSPGs and may lead to the deposition of NRGs in the extracellular matrix of neuromuscular synapses and within the central nervous system. HSPGs, including agrin, have been identified to play important roles in neuromuscular junction formation (17Sanes J.R. Lichtman J.W. Annu. Rev. Neurosci. 1999; 22: 389-442Crossref PubMed Scopus (1230) Google Scholar). For the NRGs, HSPGs may serve to "direct" the accumulation of NRG forms with the IG-like domain to the basal lamina of developing neuromuscular synapses and other locations in the developing nervous system at key stages of development (3Loeb J.A. Khurana T.S. Robbins J.T. Yee A.G. Fischbach G.D. Development (Camb.). 1999; 126: 781-791Crossref PubMed Google Scholar). The functional consequences of NRG-HSPG interactions on AChR expression, however, are not known. One feature of NRG that distinguishes it from other heparin-binding ligands is that it has distinct domains for heparin and receptor binding that are separated from one another by a glycosylated, spacer region. This gave us the unique ability to determine the direct effects of HSPG binding on receptor and gene activation that would not be readily possible with other heparin-binding ligands. Here, we provide evidence that the IG-like domain functions to keep the EGF-like domain at sufficiently high concentrations near erbB receptors for a sufficiently long period of time necessary to induce AChR gene expression. We have examined how NRG-HSPG interactions affect NRG-erbB receptor binding, erbB receptor autophosphorylation, and downstream activation of AChR genes and newly synthesized proteins in primary chick myotube cultures. Using recombinant NRG β1 isoforms with and without the heparin-binding IG-like domain, we show that IG-like domain potentiates the EGF-like domain on receptor phosphorylation by interaction with endogenous HSPGs. Through HSPG interactions, the IG-like domain induces sustained NRG-erbB receptor phosphorylation for over 8 h that is required to turn on AChR mRNA and protein expression. These results provide a molecular rationale for why NRG is highly concentrated in the extracellular matrix of neuromuscular synapses. Recombinant human NRG polypeptides were generously provided by Amgen (Thousand Oaks, CA). The isolated EGF-like domain corresponds to amino acids 177–246, and the IG-EGF domain corresponds to amino acids 14–246; both of the human β1 forms were expressed in Escherichia coli (Fig. 1) and used previously (15Loeb J.A. Fischbach G.D. J. Cell Biol. 1995; 130: 127-135Crossref PubMed Scopus (114) Google Scholar). Heparin (porcine intestinal mucosa, Mr∼13,000) and bovine serum albumin (Fraction V) were purchased from Sigma. Heparitinase and chondroitinase ABC were purchased from Seikagaku America (Falmouth, MA). The tyrosine kinase inhibitor tyrphostin AG1478 was purchased from Calbiochem. All other tissue culture reagents were purchased from Invitrogen-Life Technologies (Carlsbad, CA). Chick myotube cultures from embryonic day 11 pectoral muscle were cultured on 50 μg/ml collagen type I from rat tail (Collaborative Biomedical Products) as described previously (18Falls D.L. Rosen K.M. Corfas G. Lane W.S. Fischbach G.D. Cell. 1993; 72: 801-815Abstract Full Text PDF PubMed Scopus (552) Google Scholar). 2% chick embryo extract medium was made from 2% chick embryo extract, 10% horse serum, 1% penicillin/streptomysin, 1% glutamine, and 1% pyruvate in minimum essential medium. Cells were plated at 60,000 cells per 48-well tissue culture plate in 300 μl of medium and at 3 × 105 cells/dish in 2 ml of medium in 35-mm tissue culture dishes and kept in a 5% CO2 incubator. Fresh 2% chick embryo extract medium was changed on the 4thday after plating, and experiments were done on the 7th day after plating. Each form of NRG with and without other reagents was applied to 7-day-old chick myotubes for 45 min at 37 °C. The medium was then discarded, and the cells were solubilized in SDS sample buffer and boiled for 5 min as described previously (11Corfas G. Falls D.L. Fischbach G.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1624-1628Crossref PubMed Scopus (71) Google Scholar). Phosphorylated forms of the erbB receptors (p185) were detected by Western blot analysis using the phosphotyrosine monoclonal antibody PY 20 (Transduction Laboratories) after first resolving on overrun 5% reducing SDS-polyacrylamide gels. The filters were then probed with a goat anti-mouse IgG coupled to peroxidase (Roche Molecular Biochemicals Corp.) and exposed to X-blue film (Eastman Kodak Co.) after treatment with chemiluminescence reagents (PerkinElmer Life Sciences). 125I-IG-EGF NRG was prepared (19Hunter W.M. Greenwood F.C. Biochem. J. 1964; 91: 43-56Crossref PubMed Scopus (133) Google Scholar) using the chloramine-T method yielding a specific activity of 9,000 Ci/nmol. Chick myotube cultures were pretreated with either heparitinase (0.04 unit/ml, to remove endogenous heparan sulfates) or chondroitinase (0.04 unit/ml, to remove chondroitin sulfates) for 90 min at 37 °C. 0.1 nm125I-IG-EGF was applied to prechilled myotube cultures in quadruplicates for 2 h on ice without or with 1 μm cold ligand (IG-EGF) or 500 μg/ml soluble heparin. Total counts/min bound was measured in a γ counter (Packard Instruments) after washing three times with cold minimum essential medium containing 0.1% bovine serum albumin, and total counts/min bound was measured after solubilization in 100 μl of 1 n NaOH, 0.5 mg/ml deoxycholate for 10 min at room temperature. Measurement of newly inserted AChR proteins in the myotube membrane was achieved by blocking AChR binding sites with cold α-bungarotoxin, allowing newly synthesized AChRs to appear in the plasma membrane, and then measuring these new AChRs with 125I-α-bungarotoxin (Amersham Pharmacia Biotech) as described previously (18Falls D.L. Rosen K.M. Corfas G. Lane W.S. Fischbach G.D. Cell. 1993; 72: 801-815Abstract Full Text PDF PubMed Scopus (552) Google Scholar). Northern blots were performed on RNA extracted from either 35-mm plates or pooled wells from 48-well plates using Ultraspec (Biotecx Laboratories, Inc., Houston, TX) as described previously (20Loeb J.A. Fischbach G.D. J. Neurosci. 1997; 17: 1416-1424Crossref PubMed Google Scholar). AChR α subunit exon 7 and flanking sequence in vector pBluescript was kindly provided by Dr. Jakob Schmidt at SUNY, and 32P-labeled probes were prepared from this 399-base pair fragment by random priming using Prime It II kit (Stratagene, La Jolla, CA). The membrane was re-probed with a32P-labeled GAPDH probe for normalization purposes. Quantitation was performed either using a PhosphorImager (Molecular Dynamics) or by image analysis of x-ray films on nonsaturated images after scanning with a Power Look (UMAX) flat bed scanner with transparency adapter and then analyzing with the Metamorph Imaging system, version 4.0 (Universal Imaging Corp.). Relative mRNA levels were normalized to GAPDH levels. To assess the contribution the IG-like domain of NRG makes on erbB receptor phosphorylation, we compared the potency of recombinant NRG forms with (IG-EGF) and without (EGF) the IG-like domain on primary chick myotube cultures. These are biologically active, recombinant NRG Type I β1 isoforms that we have used previously (15Loeb J.A. Fischbach G.D. J. Cell Biol. 1995; 130: 127-135Crossref PubMed Scopus (114) Google Scholar) (see Fig. 1). Although there are some additional sequences besides the IG-like domain on the IG-EGF construct, the heparin-binding portion has been localized to the IG-like domain (16Meier T. Masciulli F. Moore C. Schoumacher F. Eppenberger U. Denzer A.J. Jones G. Brenner H.R. J. Cell Biol. 1998; 141: 715-726Crossref PubMed Scopus (97) Google Scholar). After treatment with NRG, the erbB receptor complex appears as a diffuse band at ∼185 KDa called "p185" that includes phosphorylated forms of erbB2, erbB3, and erbB4 on phosphotyrosine Western blot (11Corfas G. Falls D.L. Fischbach G.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1624-1628Crossref PubMed Scopus (71) Google Scholar). Employing equal molar amounts, the IG-EGF form was 4-fold more potent than the EGF form alone during the 45 min assay (Fig. 2). We investigated the possibility that this increase in potency of the IG-EGF form is due to HSPG interactions on the myotube surface. We first compared the proportion of NRG that binds to HSPGsversus erbB receptors by selectively removing endogenous HSPGs with heparitinase. A competitive binding study of125I-NRG (IG-EGF) demonstrated that NRG is distributed fairly equally between both erbB receptors and endogenous HSPGs (Fig.3A). Specific binding, assessed by blocking with cold IG-EGF NRG, was about 25% of total binding. Similarly, removing endogenous heparan sulfates reduced total binding by 25%, but did not reduce the specific binding to erbB receptors. Soluble heparin blocked total binding by 50%, similar to the combination of heparitinase and cold NRG. This suggests that a similar proportion of 125I-NRG is associated with its receptor and with endogenous HSPGs. Pretreatment with an enzyme that selectively degrades chondroitin sulfate (chondroitinase) had no effect on binding to either erbB receptors or HSPGs, demonstrating the specificity of NRG interactions with endogenous HSPGs. We examined whether this increased potency of the IG-EGF form on receptor phosphorylation was due directly to interactions with endogenous HSPGs (Fig. 3B). While treatment with heparitinase had no effect on EGF-like domain-induced receptor phosphorylation, it reduced erbB phosphorylation induced by the IG-EGF form to the same level as the EGF-like domain alone. This result suggests that interactions between the IG-like domain of NRG and endogenous HSPGs are directly responsible for the increased potency. This experiment also suggests that the two recombinant ligands are correctly folded, since their activity is the same in the absence of HSPGs. Adding soluble heparin has been shown to have both stimulatory and inhibitory effects on heparin-binding growth factors (1Schlessinger J. Lax I. Lemmon M. Cell. 1995; 83: 357-360Abstract Full Text PDF PubMed Scopus (450) Google Scholar, 15Loeb J.A. Fischbach G.D. J. Cell Biol. 1995; 130: 127-135Crossref PubMed Scopus (114) Google Scholar, 21Krufka A. Guimond S. Rapraeger A.C. Biochemistry. 1996; 35: 11131-11141Crossref PubMed Scopus (59) Google Scholar). Fig. 3A shows that high concentrations of soluble heparin effectively block NRG's ability to bind to its receptor. Here we explored the effects of soluble heparin both in modulating the NRG's ability to stimulate erbB receptor phosphorylation and AChR synthesis in cultured myotubes. Whereas heparin had no effect on receptor phosphorylation induced by the EGF-like domain, as little as 5 μg/ml heparin almost completely blocked receptor phosphorylation (p185) with the NRG form containing the IG-like domain (Fig. 4A). Even though heparin inhibited receptor phosphorylation at all concentrations above 1.0 μg/ml, a biphasic response to heparin was observed for AChR induction. At high concentrations, heparin was inhibitory, while low concentrations of heparin, up to 10 μg/ml, actually stimulated the rate of appearance of newly inserted AChR proteins onto the muscle membrane (Fig. 4B). Given that the AChR induction assay is considerably longer than the phosphorylation assay, these results suggest that the stimulatory effects of low concentrations of heparin may occur as a result of the longer course of this assay. Thus far, our results suggest that the IG-like domain of NRG concentrates NRG along cell surfaces through HSPG interactions and that this concentration effect may be responsible for increasing NRG's potency for receptor binding and phosphorylation. Another potential function of this interaction is to provide sustained signaling. Here we have examined the time of NRG exposure required to induce AChR mRNA and protein expression. We measured AChR induction after treating myotubes for different time periods from 15 min to 24 h with the same concentration of NRG (EGF versus IG-EGF) followed by "washing out" the NRG. We then measured AChR mRNA levels (Fig.5, A and B) and new surface proteins (Fig. 5C) 24 h from the start. The AChR α subunit was measured by Northern blot analysis and compared with the housekeeping gene GAPDH to normalize the data. The α subunit was chosen, because this subunit has been shown previously to have the greatest induction by NRG in chick myotubes (10Altiok N. Bessereau J.L. Changeux J.P. EMBO J. 1995; 14: 4258-4266Crossref PubMed Scopus (131) Google Scholar, 22Harris D.A. Falls D.L. Dill-Devor R.M. Fischbach G.D. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 1983-1987Crossref PubMed Scopus (70) Google Scholar). While maximal activation of AChR mRNA expression by the IG-EGF domain was achieved with a 15-min treatment, 24 h of treatment was needed for the EGF-like domain to induce significant AChR mRNA expression. Similarly, a significant lag of more than 8 h of exposure was required for the EGF-like domain to get a similar response as the IG-EGF form in promoting the insertion of new AChR proteins on the cell surface (Fig. 5C). From our earlier results, it is most likely that the IG-like domain binds to endogenous HSPGs so that the IG-EGF form is not in fact washed out in this assay, but provides a steady source of ligand to activate the erbB receptors. Unexpectedly, the appearance of new AChR proteins on the cell surface preceded the increase in mRNA, suggesting that mechanisms that increase AChR mRNA and protein expression on the cell surface may be distinct. We next examined the phosphorylation state of the erbB receptors for 24 h of NRG exposure (Fig.6A). Consistent with our results in Fig. 2, the initial phosphorylation response was greater with the IG-EGF form than the EGF form for the first 2 h. However, after 2 h, both constructs kept the erbB receptors phosphorylated to a similar, lower level through 24 h. The level of phosphorylation at 24 h was significantly above background levels without NRG treatment as seen in the blank lane. When we washed out these constructs after 45 min of treatment, and then measured p185 phosphorylation at indicated time points (Fig. 6B), the response induced by the EGF-like domain decayed rapidly and was lost within 1 h after the wash. On the other hand, the IG-EGF construct kept the erbB receptors phosphorylated for a full 8 h after the wash. These results suggest that sustained activation of erbB receptors is a consequence of NRG-HSPG interactions and may be required for AChR mRNA and protein induction. Finally, to demonstrate the importance of sustained erbB receptor phosphorylation in AChR induction, we used a specific erbB tyrosine kinase inhibitor AG1478 (23Levitzki A. Gazit A. Science. 1995; 267: 1782-1788Crossref PubMed Scopus (1619) Google Scholar) to block erbB receptor phosphorylation at specific times after adding NRG (Fig. 7). Fig. 7A shows that the concentration of inhibitor required to completely block erbB receptor phosphorylation using the EGF construct was above 1 μm. Using 10 μmAG1478, Fig. 7, B and C, show that 12–24 h of sustained receptor phosphorylation were necessary for the EGF-like domain to increase AChR mRNA levels, suggesting that sustained receptor phosphorylation is needed. Similar results were obtained using the IG-EGF construct of NRG; however, higher concentrations of the inhibitor were required. The results presented in this paper examine how extracellular interactions between the IG-like domain of NRG and HSPGs in the extracellular matrix modulate the intracellular signaling events that regulate the expression of synaptic AChRs. We have identified both short and long term effects of NRG-HSPG interactions. The addition of the IG-like domain to the EGF-like domain of NRG results in a rapid and transient increase in erbB receptor phosphorylation that requires interactions with endogenous HSPGs. Sustained NRG-erbB receptor activation, however, is required to turn on AChR mRNA and protein expression. These results suggest that the IG-like domain is needed to keep sufficiently high concentrations of the EGF-like domain near the erbB receptors for a long enough period of time required to induce AChR gene expression. During embryonic development, HSPGs and NRG become concentrated coincidently in the synaptic basal lamina of neuromuscular junctions and remain there throughout adult life (3Loeb J.A. Khurana T.S. Robbins J.T. Yee A.G. Fischbach G.D. Development (Camb.). 1999; 126: 781-791Crossref PubMed Google Scholar). Our observation that sustained erbB receptor activation is required to induce AChR expression may be the reason why such high concentrations of NRG are maintained in the synaptic basal lamina. Our results demonstrate that more than 8 h of receptor activation is needed to induce AChR mRNA and protein expression and that this receptor activation requires the continuous presence of NRG. Earlier studies have shown that more than 5 h of NRG exposure was required to increase the insertion rate of new AChRs (24Buc-Caron M.H. Nystrom P. Fischbach G.D. Dev. Biol. 1983; 95: 378-386Crossref PubMed Scopus (33) Google Scholar) and up to a 12-h exposure was needed to increase voltage-gated sodium channels in chick muscle cultures (8Corfas G. Fischbach G.D. J. Neurosci. 1993; 13: 2118-2125Crossref PubMed Google Scholar). Our results differ considerably from a recent report where only 1-min exposure of the EGF-like domain of NRG was needed to induce AChR mRNA 24 h after addition to C2C12 cells (25Si J. Wang Q. Mei L. J. Neurosci. 1999; 19: 8498-8508Crossref PubMed Google Scholar). While the difference could be due to the different myotube culture systems used, it is also possible that they did not "wash away" the NRG. We have found that when we washed chick myotube cultures with PBS instead of serum-containing medium, we found similar results in that both the EGF and the IG-EGF constructs could induce AChR mRNA after 24 h with only 15 min of exposure. We also suggest that the reason the ligand is required for over 8 h is because the erbB receptors need to be maintained in a phosphorylated state for this entire period to increase AChR mRNA. We showed this using a specific tyrosine kinase antagonist added at different times after NRG addition. It is not clear why sustained erbB receptor activation is needed. There are, however, examples from the literature where sustained activation of tyrosine kinase signaling pathways such as mitogen-activated protein kinase and cAMP response element-binding protein for 4 h or more are needed to induce cell differentiation, migration, and the expression of specific genes (26Liu F.C. Graybiel A.M. Neuron. 1996; 17: 1133-1144Abstract Full Text Full Text PDF PubMed Scopus (163) Google Scholar, 27Marshall C.J. Cell. 1995; 80: 179-185Abstract Full Text PDF PubMed Scopus (4232) Google Scholar, 28McCawley L.J. Li S. Wattenberg E.V. Hudson L.G. J. Biol. Chem. 1999; 274: 4347-4353Abstract Full Text Full Text PDF PubMed Scopus (192) Google Scholar). Sustained receptor activation may be needed to recruit other kinases and/or transcription factors or perhaps is needed to initiate new protein synthesis. New protein synthesis has been shown recently to be needed to induce AChR gene transcription (25Si J. Wang Q. Mei L. J. Neurosci. 1999; 19: 8498-8508Crossref PubMed Google Scholar). In this report, newly made c-JUN and/or c-FOS mRNA were required to mediate the up-regulation of AChR mRNA by NRG. Pretreatment of C2C12 cells with the protein synthesis inhibitor cycloheximide blocked NRG-induced AChR ε subunit mRNA completely, suggesting that de novoprotein synthesis was needed. There are a growing number of growth and differentiation factors that bind to extracellular matrix HSPGs, including NRG, acidic and basic fibroblast growth factors, transforming growth factor-β, granulocyte-macrophage colony-stimulating factor, interleukin-3, interferon γ, the netrins, and heparin-binding epidermal growth factor (1Schlessinger J. Lax I. Lemmon M. Cell. 1995; 83: 357-360Abstract Full Text PDF PubMed Scopus (450) Google Scholar, 29Ruoslahti E. Yamaguchi Y. Cell. 1991; 64: 867-869Abstract Full Text PDF PubMed Scopus (1167) Google Scholar, 30Serafini T. Kennedy T.E. Galko M.J. Mirzayan C. Jessell T.M. Tessier-Lavigne M. Cell. 1994; 78: 409-424Abstract Full Text PDF PubMed Scopus (1161) Google Scholar). Interactions with extracellular matrix HSPGs are thought to provide a readily accessible store of factors, to protect them from proteolytic degradation, and in some instances, to play important regulatory roles on their functions. Fibroblast growth factor (FGF) is one of the best studied growth factors in which heparin or HSPG binding is necessary for FGF receptor dimerization and activation (1Schlessinger J. Lax I. Lemmon M. Cell. 1995; 83: 357-360Abstract Full Text PDF PubMed Scopus (450) Google Scholar, 31Schlessinger J. Cell. 2000; 103: 211-225Abstract Full Text Full Text PDF PubMed Scopus (3523) Google Scholar). For other growth factors, such as transforming growth factor-β, heparin-binding EGF, and NRG, receptor dimerization can be achieved in the absence of soluble heparin or cell surface HSPGs, but HSPGs may function to enhance the biological activity of these growth factors (32Higashiyama S. Abraham J.A. Klagsbrun M. J. Cell Biol. 1993; 122: 933-940Crossref PubMed Scopus (313) Google Scholar). We found that a similar proportion of NRG associates with its receptors and with endogenous HSPGs on the cell surface. As a result of this interaction, NRG-induced activation of erbB receptor phosphorylation is increased by 4-fold. Removing endogenous HSPGs by heparitinase treatment eliminates this increase in potency. A similar observation has been described for the L6 muscle cell line (33Han B. Fischbach G.D. Soc. Neurosci. Abstr. 2000; 25: 43Google Scholar). There are a number of possible explanations as to why HSPGs increases NRG potency. The most likely is that endogenous HSPGs situated on the cell surface concentrate NRG in a micro-environment nearby their erbB receptors, thus increasing the local ligand concentration. In addition, the HSPGs may present NRG molecules to their higher affinity erbB receptors. Finally, it is possible that heparan sulfate simultaneously binds to several NRG monomers producing a multimer of ligands that could more effectively stimulate multimeric receptor activation. Similar mechanisms have been proposed for FGF, where additional interactions between HSPGs and the FGF receptor have been proposed (31Schlessinger J. Cell. 2000; 103: 211-225Abstract Full Text Full Text PDF PubMed Scopus (3523) Google Scholar). We looked at the effects of soluble heparin on NRG-induced receptor phosphorylation and AChR expression. We found that while soluble heparin blocked NRG binding and NRG-induced erbB receptor phosphorylation, soluble heparin had a biphasic effect on AChR protein expression. Specifically, high concentrations of heparin blocked AChR protein expression, whereas low concentrations actually stimulated new AChR protein expression. Similar biphasic effects of soluble heparin have been described for FGF (21Krufka A. Guimond S. Rapraeger A.C. Biochemistry. 1996; 35: 11131-11141Crossref PubMed Scopus (59) Google Scholar). One possible explanation for our observations is that low concentrations of soluble heparin release NRG from endogenous HSPG binding sites, thus freeing it to interact with the high affinity erbB receptors. High concentrations of heparin, on the other hand, not only free NRG from endogenous HSPGs, but also keep it far from the cell surface erbB receptors. Another possibility may come from the time difference between the two assays. In an earlier study, we showed that the IG-like domain of NRG is preferentially degraded by proteases to produce an active EGF-like domain that is no longer inhibited by soluble heparin (15Loeb J.A. Fischbach G.D. J. Cell Biol. 1995; 130: 127-135Crossref PubMed Scopus (114) Google Scholar). Since the receptor phosphorylation assay is complete in 45 min, whereas the AChR protein insertion assay takes over 24 h, it is quite possible that the IG-like domain of the NRG is preferentially degraded over the longer course of this assay and that higher concentrations of heparin are required to prevent this proteolytic degradation. One of the more surprising results here is that even though both AChR mRNA and protein expression required prolonged NRG exposure, new AChR protein expression preceded the rise in AChR mRNA. This observation suggests that the cell-signaling mechanism that increases the insertion of new AChRs on the cell surface may be distinct from the mechanism that increases AChR mRNA. Further experiments will be required to sort out this apparent dissociation between AChR mRNA and protein expression that may involve different signaling pathways initiated by erbB receptor activation. It is possible that some of the earlier signaling events that do not require protein synthesis result in movement of pre-existing AChRs to the cell surface, whereas to make new AChR mRNA, more prolonged signaling pathways are required such as the requirement for new protein synthesis as has been described recently (25Si J. Wang Q. Mei L. J. Neurosci. 1999; 19: 8498-8508Crossref PubMed Google Scholar). We thank Dr. Jakob Schmidt at SUNY for kindly providing the AChR α subunit vector.