Coordinated Activation of Autophosphorylation Sites in the RET Receptor Tyrosine Kinase
2002; Elsevier BV; Volume: 277; Issue: 3 Linguagem: Inglês
10.1074/jbc.m107992200
ISSN1083-351X
AutoresMuriel Coulpier, Jonas Anders, Carlos F. Ibáñez,
Tópico(s)Axon Guidance and Neuronal Signaling
ResumoThe catalytic and signaling activities of RET, a tyrosine kinase receptor for glial cell line-derived neurotrophic factor (GDNF), are controlled by the autophosphorylation of several tyrosine residues in the RET cytoplasmic domain. To analyze the phosphorylation state of individual tyrosines, we generated antibodies recognizing specific phosphotyrosine sites involved in the catalytic (Tyr905) and downstream signaling (Tyr1015, Tyr1062, and Tyr1096) activities of this receptor. Stimulation with GDNF induced coordinated phosphorylation of the 4 tyrosine residues in neuronal cell lines and in primary cultures of sympathetic neurons isolated from rat superior cervical ganglia. Neurturin and artemin, two other members of the GDNF ligand family, also induced synchronized phosphorylation of RET tyrosines with kinetics comparable to those observed with GDNF. Tyrosine phosphorylation was maximal 15 min after ligand stimulation, decaying thereafter with similar kinetics in all 4 residues. Co-stimulation with a soluble form of the GFRα1 co-receptor potentiated ligand-dependent phosphorylation of different intracellular tyrosines to a similar extent and increased the survival of superior cervical ganglion neurons compared with treatment with GDNF alone. In vivo, high levels of phosphorylated Tyr905, Tyr1015, and Tyr1062 were detected in embryonic mouse dorsal root ganglia, with a sharp decline at early postnatal stages. Protein transduction of anti-Tyr(P)1062 antibodies into cultured cells reduced activation of MAPKs ERK1 and ERK2 and the AKT kinase in response to GDNF and diminished GDNF-dependent neuronal differentiation and survival of embryonic sensory neurons from the nodose ganglion. These results demonstrate synchronized utilization of individual RET tyrosine residues in neurons in vivo and reveal an important role for RET Tyr1062 in mediating neuronal survival by GDNF. The catalytic and signaling activities of RET, a tyrosine kinase receptor for glial cell line-derived neurotrophic factor (GDNF), are controlled by the autophosphorylation of several tyrosine residues in the RET cytoplasmic domain. To analyze the phosphorylation state of individual tyrosines, we generated antibodies recognizing specific phosphotyrosine sites involved in the catalytic (Tyr905) and downstream signaling (Tyr1015, Tyr1062, and Tyr1096) activities of this receptor. Stimulation with GDNF induced coordinated phosphorylation of the 4 tyrosine residues in neuronal cell lines and in primary cultures of sympathetic neurons isolated from rat superior cervical ganglia. Neurturin and artemin, two other members of the GDNF ligand family, also induced synchronized phosphorylation of RET tyrosines with kinetics comparable to those observed with GDNF. Tyrosine phosphorylation was maximal 15 min after ligand stimulation, decaying thereafter with similar kinetics in all 4 residues. Co-stimulation with a soluble form of the GFRα1 co-receptor potentiated ligand-dependent phosphorylation of different intracellular tyrosines to a similar extent and increased the survival of superior cervical ganglion neurons compared with treatment with GDNF alone. In vivo, high levels of phosphorylated Tyr905, Tyr1015, and Tyr1062 were detected in embryonic mouse dorsal root ganglia, with a sharp decline at early postnatal stages. Protein transduction of anti-Tyr(P)1062 antibodies into cultured cells reduced activation of MAPKs ERK1 and ERK2 and the AKT kinase in response to GDNF and diminished GDNF-dependent neuronal differentiation and survival of embryonic sensory neurons from the nodose ganglion. These results demonstrate synchronized utilization of individual RET tyrosine residues in neurons in vivo and reveal an important role for RET Tyr1062 in mediating neuronal survival by GDNF. glial cell line-derived neurotrophic factor neurturin artemin GDNF family receptor-α extracellular signal-regulated kinase phosphatidylinositol 3-kinase nerve growth factor dorsal root ganglion/ganglia superior cervical ganglion/ganglia embryonic day postnatal day mitogen-activated protein kinases The GDNF1 ligand family, a group of polypeptides structurally related to the transforming growth factor-β superfamily, is involved in the control of neuronal survival and differentiation, kidney morphogenesis, and spermatogonial cell fate (1Airaksinen M.S. Titievsky A. Saarma M. Mol. Cell. Neurosci. 1999; 13: 313-325Crossref PubMed Scopus (381) Google Scholar, 2Baloh R.H. Enomoto H. Johnson Jr., E.M. Milbrandt J. Curr. Opin. Neurobiol. 2000; 10: 103-110Crossref PubMed Scopus (404) Google Scholar, 3Meng X. Lindahl M. Hyvonen M.E. Parvinen M. de Rooij D.G. Hess M.W. Raatikainen-Ahokas A. Sainio K. Rauvala H. Lakso M. Pichel J.G. Westphal H. Saarma M. Sariola H. Science. 2000; 287: 1489-1493Crossref PubMed Scopus (1055) Google Scholar, 4Sariola H. Saarma M. Int. J. Dev. Biol. 1999; 43: 413-418PubMed Google Scholar). Each of the four members of the GDNF family (i.e.GDNF, NTN, ART, and PSP) binds specifically to different members of a small family of glycosylphosphatidylinositol-anchored receptors, the GDNF family α-receptors, of which four different members (GFRα1–4) are currently known (1Airaksinen M.S. Titievsky A. Saarma M. Mol. Cell. Neurosci. 1999; 13: 313-325Crossref PubMed Scopus (381) Google Scholar, 2Baloh R.H. Enomoto H. Johnson Jr., E.M. Milbrandt J. Curr. Opin. Neurobiol. 2000; 10: 103-110Crossref PubMed Scopus (404) Google Scholar, 5Ibáñez C.F. Trends Neurosci. 1998; 21: 438-444Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar). Intracellular signaling is accomplished by the recruitment of a receptor tyrosine kinase, RET, to the GDNF·GFRα complex. Although all members of the GDNF ligand family utilize RET as a signaling receptor subunit, specificity is achieved by differential binding to individual GFRα molecules. GFRα receptors can mediate activation of RET when expressed on the surface of the same cell (activation in cis) or when presented in soluble form or immobilized on the cell matrix or neighboring cells (activation in trans) (6Yu T. Scully S. Yu Y.B. Fox G.M. Jing S.Q. Zhou R.P. J. Neurosci. 1998; 18: 4684-4696Crossref PubMed Google Scholar, 7Paratcha G. Ledda F. Baars L. Coulpier M. Besset V. Anders J. Scott R. Ibáñez C.F. Neuron. 2001; 29: 171-184Abstract Full Text Full Text PDF PubMed Scopus (237) Google Scholar). Upon ligand binding, RET is thought to form dimers and become phosphorylated at specific cytoplasmic tyrosine residues. Tyrosine autophosphorylation is required for the catalytic activity of RET and for downstream signaling. Thus, tyrosine autophosphorylation constitutes the first intracellular event of the RET signaling cascade activated by members of the GDNF ligand family. Eighteen tyrosine residues, 2 in the juxtamembrane domain, 11 in the kinase domain, and 5 in the carboxyl-terminal tail, are present in the cytoplasmic domain of the long isoform of RET. Tyrosine 905 in the RET kinase domain corresponds to tyrosine 416 in the activation loop of the cytoplasmic tyrosine kinase Src, a conserved residue in many tyrosine kinases known to play a crucial role in kinase activation (8Hanks S.K. Quinn A.M. Hunter T. Science. 1988; 241: 42-52Crossref PubMed Scopus (3775) Google Scholar). Mutation of tyrosine 905 to phenylalanine (Y905F) impairs the kinase activity and abolishes the transforming activity of RET-MEN2A, an oncogenic, constitutively active form of RET identified in patients with multiple endocrine neoplasia type 2A (9Iwashita T. Asai N. Murakami H. Matsuyama M. Takahashi M. Oncogene. 1996; 12: 481-487PubMed Google Scholar). Tyrosine 905 is also involved in the binding of two adaptor proteins containing SH2 domains, Grb7 and Grb10, presumably involved in downstream signaling events (10Pandey A. Liu X. Dixon J.E. Difiore P.P. Dixit V.M. J. Biol. Chem. 1996; 271: 10607-10610Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 11Pandey A. Duan H. Di F.P. Dixit V.M. J. Biol. Chem. 1995; 270: 21461-21463Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar, 12Durick K. Wu R.Y. Gill G.N. Taylor S.S. J. Biol. Chem. 1996; 271: 12691-12694Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). Six additional tyrosines (Tyr687 in the juxtamembrane domain; Tyr826 in the catalytic domain; and Tyr1015, Tyr1029, Tyr1062, and Tyr1096 in the carboxyl-terminal tail) have been shown to be autophosphorylated in various oncogenic forms of RET by site-directed mutagenesis and phosphopeptide mapping experiments (13Liu X. Vega Q.C. Decker R.A. Pandey A. Worby C.A. Dixon J.E. J. Biol. Chem. 1996; 271: 5309-5312Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). The functional importance of the phosphorylation of Tyr687, Tyr826, and Tyr1029 is unknown. On the other hand, phosphorylation of Tyr1015, Tyr1062, and Tyr1096 has been linked to distinct downstream signaling events. Tyrosine 1015 is part of the motif YLXL, a docking site for phospholipase Cγ; and mutation of the corresponding residue in theRET/PTC2 oncogene impairs its ability to activate phospholipase Cγ and reduces drastically its oncogenic activity in NIH 3T3 cells (14Borrello M.G. Alberti L. Arighi E. Bongarzone I. Battisini C. Bardelli A. Pasini B. Piutti C. Rizzetti M.G. Mondellini P. Radice M.T. Pierotti M.A. Mol. Cell. Biol. 1996; 16: 2151-2163Crossref PubMed Google Scholar). Tyrosine 1062 is part of the motif NKXY, which constitutes a docking site for the phosphotyrosine-binding domain of Shc and FRS2 adaptor proteins. Interaction between phosphorylated Tyr1062 and either of these two adaptors leads to activation of the Ras/ERK and PI3K/AKT pathways in oncogenic as well as ligand-activated RET (15Asai N. Murakami H. Iwashita T. Takahashi M. J. Biol. Chem. 1996; 271: 17644-17649Abstract Full Text Full Text PDF PubMed Scopus (159) Google Scholar, 16Arighi E. Alberti L. Torriti F. Ghizzoni S. Rizzetti M.G. Pelicci G. Pasini B. Bongarzone I. Piutti C. Pierotti M.A. Borrello M.G. Oncogene. 1997; 14: 773-782Crossref PubMed Scopus (107) Google Scholar, 17Besset V. Scott R.P. Ibáñez C.F. J. Biol. Chem. 2000; 275: 39159-39166Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar, 18Segouffin-Cariou C. Billaud M. J. Biol. Chem. 2000; 275: 3568-3576Abstract Full Text Full Text PDF PubMed Scopus (186) Google Scholar, 19Kurokawa K. Iwashita T. Murakami H. Hayashi H. Kawai K. Takahashi M. Oncogene. 2001; 20: 1929-1938Crossref PubMed Scopus (79) Google Scholar, 20Melillo R.M. Santoro M. Ong S.H. Billaud M. Fusco A. Hadari Y.R. Schlessinger J. Lax I. Mol. Cell. Biol. 2001; 21: 4177-4187Crossref PubMed Scopus (109) Google Scholar). Interestingly, the splicing event that leads to the generation of the short and long isoforms of RET takes place precisely after Tyr1062 and places this tyrosine residue in a perfect context for binding to SH2 domains in the short (but not the long) RET isoform. Thus, both phosphotyrosine-binding domain-containing and SH2 domain-containing target proteins may bind to this phosphorylated tyrosine in the short RET isoform. Tyrosine 1062 has also been implicated in the binding of Enigma to RET (12Durick K. Wu R.Y. Gill G.N. Taylor S.S. J. Biol. Chem. 1996; 271: 12691-12694Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar), although this interaction appears to be independent of tyrosine phosphorylation. Recently, a role for this tyrosine residue in the docking and activation of different members of the Dok family of adaptor molecules has also been demonstrated (21Grimm J. Sachs M. Di Britsch S. Cesare S. Schwarz-Romond T. Alitalo K. Birchmeier W. J. Cell Biol. 2001; 154: 345-354Crossref PubMed Scopus (138) Google Scholar). Mutation of tyrosine 1062 (Y1062F) dramatically impairs the transforming activity of oncogenic RET-MEN2A and RET-MEN2B (22Asai N. Iwashita T. Matsuyama M. Takahashi M. Mol. Cell. Biol. 1995; 15: 1613-1619Crossref PubMed Google Scholar). Despite its high degree of connectivity to multiple intracellular pathways, the biological function of this residue in the wild-type RET receptor has not been investigated. Finally, tyrosine 1096, located in the 51-residue carboxyl-terminal tail that is specific for the long isoform of RET, is part of the sequence PYXNX, a well known binding site for the Grb2 adaptor protein. Grb2 has been found to interact with the long isoform of RET/PTC2 and wild-type RET via this residue (17Besset V. Scott R.P. Ibáñez C.F. J. Biol. Chem. 2000; 275: 39159-39166Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar, 23Alberti L. Borrello M.G. Ghizzoni S. Torriti F. Rizzetti M.G. Pierotti M. Oncogene. 1998; 17: 1079-1087Crossref PubMed Scopus (77) Google Scholar). In this work, we have investigated the phosphorylation of individual tyrosine residues in the cytoplasmic domain of RET in cell lines, cultured primary neurons, and in vivo. We have studied differences in the kinetics of phosphorylation and dephosphorylation of individual residues and whether different members of the GDNF ligand family may be capable of inducing distinct patterns of tyrosine phosphorylation in RET. For this purpose, we have generated antibodies recognizing the specific phosphorylation of tyrosines 905, 1015, 1062, and 1096 in this receptor. We have introduced several of these phosphotyrosine-specific antibodies into cell lines and primary neurons to investigate the functional roles played by individual tyrosine residues of RET in GDNF-mediated downstream signaling and neuronal survival. The MG87-α1/RET and MG87-α3/RET lines were derived from MG87 fibroblasts by stable transfection of GDNF receptor subunits. MG87-α1/RET cells express rat GFRα1 and the long isoform of human RET. MG87-α3/RET cells express mouse GFRα3 and the long isoform of human RET. Neuro2A-α1 cells were generated by stable transfection of the mouse neuroblastoma Neuro2A with rat GFRα1. MN1 is an immortalized mouse motor neuron cell line (24Trupp M. Arenas E. Fainzilber M. Nilsson A.-S. Sieber B.A. Grigoriou M. Kilkenny C. Salazar-Grueso E. Pachnis V. Arumäe U. Sariola H. Saarma M. Ibáñez C.F. Nature. 1996; 381: 785-789Crossref PubMed Scopus (719) Google Scholar). Recombinant rat GDNF was produced in Sf21 insect cells and purified as previously described (25Trupp M. Rydén M. Jörnvall H. Timmusk T. Funakoshi H. Arenas E. Ibáñez C.F. J. Cell Biol. 1995; 130: 137-148Crossref PubMed Scopus (517) Google Scholar). Nerve growth factor (NGF) was purchased from Promega, NTN from PeproTech, and GFRα1-Fc from R&D Systems. Recombinant ART was a generously provided by Bob Gordon (Jannssen Research Foundation, Beerse, Belgium). The Chariot reagent used for protein transduction was from ActiveMotive (Rixensart, Belgium). Phosphorylated 15-mer peptides corresponding to four predicted phosphorylation sites in the long isoform of the mouse RET receptor (see Table I) were synthesized, coupled to keyhole limpet hemocyanin, and used to immunize rabbits by standard procedures. Antisera were evaluated by enzyme-linked immunosorbent assays against phosphorylated and unphosphorylated versions of the peptides. Peptide synthesis, rabbit immunizations, and antibody collection was done by Research Genetics. Antisera showing high titer in enzyme-linked immunosorbent assays were then screened by immunoblotting with protein extracts from control and GDNF-stimulated MG87-α1/RET cells. Reactive antisera were purified by sequential affinity chromatography steps. Total immunoglobulins were first purified on a protein G column (POROS G, PerSeptive Biosystems) and eluted with 0.1 m glycine (pH 2.7). Antisera to Tyr(P)905, Tyr(P)1015, and Tyr(P)1062 showed little or no cross-reactivity against unphosphorylated RET or irrelevant phosphotyrosines and were then affinity-purified over an Affi-Gel 15 affinity column (Bio-Rad) coupled with the corresponding phosphopeptides. Because the anti-Tyr(P)1096 antisera demonstrated cross-reactivity with unphosphorylated RET as well as other phosphotyrosines, these antibodies were first applied to an Affi-Gel 15 affinity column coupled with the unphosphorylated Tyr1096 peptide, followed by chromatography on a phosphotyrosine affinity column (Sigma). The eluent from these two steps was then applied to an Affi-Gel affinity column coupled with the Tyr(P)1096 phosphopeptide. Affinity-purified antibodies were eluted with 0.1 m glycine (pH 2.7), immediately neutralized with 1 m Tris-HCl (pH 9.0), and dialyzed against Tris-buffered saline.Table ISequences of peptides used for immunization (based on mouse RET) aligned in the corresponding regions of rat, chicken, and human RETSequenceDownstream targetsTyr(P)905 MouseDVYEEDSYVKKSKGR Rat............... Chicken..........R.... Human..........R.Q..Grb7, Grb10Tyr(P)1015 MouseMMVKSRDYLDLAAST Rat............... Chicken............... Human....R..........PLCγTyr(P)1062 MouseTWIENKLYGMSDPNW Rat............... Chicken...........Y... Human...............Shc, FRS2, Dok4/5Tyr(P)1096 MouseRYANDSVYANWMVSP Rat............... Chicken............... Human...............Grb2Dots indicate identical positions in the alignment. Phosphorylated tyrosine residues are shown in boldface. Known downstream targets are indicated. PLCγ, phospholipase Cγ. Open table in a new tab Dots indicate identical positions in the alignment. Phosphorylated tyrosine residues are shown in boldface. Known downstream targets are indicated. PLCγ, phospholipase Cγ. Cell line monolayers in 10-cm plates were changed to serum-free medium 2–4 h prior to stimulation with the indicated factors (10 min unless otherwise indicated). Cells were then lysed in a nonionic ice-cold detergent (lysis buffer: 10 mm Tris-HCl (pH 7.5), 137 mm NaCl, 2 mm EDTA, 10% glycerol, and 1% Nonidet P-40) containing a mixture of protease inhibitors (Roche Molecular Biochemicals) and phosphatase inhibitors (1 mmNaO4Va. 20 mm NaF, and 10 mmβ-glycerophosphate). Cell lysates were cleared by centrifugation at 1500 × g for 10 min and immunoprecipitated by overnight incubation at 4 °C with anti-RET antibodies and 40 μl of protein G-Sepharose beads (Amersham Biosciences, Inc.). The immunoprecipitates were washed three times with lysis buffer, solubilized in sample buffer, run on SDS-polyacrylamide gels, and blotted onto polyvinylidene difluoride membranes (Amersham Biosciences, Inc.). Blots were first probed with anti-phosphopeptide antibodies, followed by alkaline phosphatase-conjugated anti-IgG, and developed with the enhanced chemi fluorescence Western detection system (Amersham Biosciences, Inc.). All blots were scanned in a Storm 840 fluorimager (Molecular Dynamics, Inc.). For reprobing, blots were stripped for 90 min at room temperature in 0.1 macetic acid and 0.15 m NaCl. Antibodies against phosphotyrosine (used at 1:1000 dilution) and the long (1:1000) and short (1:500) isoforms of human RET were from Santa Cruz Biotechnology. For peptide competition assays, phosphorylated and unphosphorylated peptides at 10 or 100 nm were preincubated with the antibodies for 30 min at room temperature, and the mixture was then used in immunoblotting. For developmental analysis of RET phosphorylation, DRG were collected from C57 mice at embryonic day (E) 15, E17, postnatal day (P) 0, P9, P16, and the adult stage in ice-cold Tris-buffered saline containing 1 mm NaO4Va. DRG from three mice were pooled for each embryonic time point, whereas DRG from one mouse were enough for postnatal stages. Tissues were lysed in 70 μl of 1% lysis buffer, cleared by centrifugation, submitted to SDS-PAGE, and immunoblotted onto polyvinylidene difluoride membranes as described above. Each blot was first probed with an anti-phosphopeptide antibody, stripped, and then reprobed with anti-RET antibodies. P1 rat SCG were dissociated by incubation for 30 min at 37 °C in phosphate-buffered saline containing 0.025% trypsin (Invitrogen) and for an additional 30 min after addition of 5 mg/ml collagenase (Sigma), followed by mechanical trituration. Dissociated cultures from E17 mouse DRG and E9 chick nodose ganglia were prepared by incubation for 10 min at 37 °C in phosphate-buffered saline containing 0.025% trypsin, without collagenase and NGF treatments. Neurons were plated in polyornithine/laminin-coated dishes and maintained in neuronal medium (1:1 Dulbecco's modified Eagle's medium/nutrient mixture F-12, Invitrogen), 2 mm glutamine, and 1 mg/ml bovine serum albumin. SCG and nodose ganglion cultures were supplemented with 10 μm cytosine β-d-arabinofuranoside. SCG neurons were maintained in 20 ng/ml NGF. For biochemical analyses, neurons were maintained for 4 days before a 4-h starvation (i.e. without NGF in the case of SCG) and stimulation with GDNF (100 ng/ml) in the presence or absence of soluble GFRα1-Fc (100 ng/ml) for the indicated times. The cells were then lysed, and lysates were processed by SDS-PAGE and immunoblotting as described above. For survival assays, SCG neurons were first maintained for 2 days in neuronal medium supplemented with NGF; washed; and changed to neuronal medium containing either NGF (100 ng/ml) or anti-NGF antibodies (Roche Molecular Biochemicals) together with GDNF (100 ng/ml) and, where indicated, soluble GFRα1-Fc (100 ng/ml). DRG neurons were plated directly with the indicated factors without NGF preincubation. Phase-bright, neurite-bearing neurons were counted 24 and 48 h after treatment. MN1 cells and chick nodose ganglion neurons were cultured in 24-well plates. Protein transduction using the Chariot reagent was essentially performed according to the manufacturer's instructions. In our hands, the highest efficiency of protein transduction was obtained if performed in serum-free medium with cells still in suspension prior to plating. Because SCG neurons need to be treated for a few days with NGF to develop GDNF responsiveness, antibody transduction was performed in embryonic chick nodose ganglion neurons, which are readily responsive to GDNF immediately after extraction (25Trupp M. Rydén M. Jörnvall H. Timmusk T. Funakoshi H. Arenas E. Ibáñez C.F. J. Cell Biol. 1995; 130: 137-148Crossref PubMed Scopus (517) Google Scholar). Two μg of antibody was used together with 2 μl of Chariot reagent. After Chariot-mediated transduction, MN1 cells were first allowed to adhere to the plastic plate for 4 h and then stimulated for 5 min with 100 ng/ml GDNF. The cells were lysed as described above, and 10 μg of protein lysate was processed by SDS-PAGE and immunoblotting with antibodies against phosphorylated ERK1/2 or phosphorylated AKT (Cell Signaling, New England Biolabs Inc.) at 1:2000 dilution. As a control for loading, the blot was stripped and reprobed with total anti-AKT antibodies (Cell Signaling, New England Biolabs Inc.) used at 1:1000 dilution or anti-tubulin antibodies. Neurite outgrowth assay of MN1 cells was performed as previously described (7Paratcha G. Ledda F. Baars L. Coulpier M. Besset V. Anders J. Scott R. Ibáñez C.F. Neuron. 2001; 29: 171-184Abstract Full Text Full Text PDF PubMed Scopus (237) Google Scholar). For survival assay, neurons were plated after Chariot-mediated transduction and maintained in 100 ng/ml GDNF for 48 h. Phase-bright neurons were counted in the entire well (between 100 and 400 neurons/well); the results presented are averages of three different wells. To produce polyclonal antibodies directed against specific phosphotyrosine sites in RET, we immunized rabbits with 15-mer synthetic phosphopeptides corresponding to four distinct motifs in the cytoplasmic domain of mouse RET. The phosphotyrosine motifs targeted by this approach included Tyr(P)905, Tyr(P)1015, Tyr(P)1062, and Tyr(P)1096. These sequence motifs are highly conserved in RET from other vertebrate species, including rat, chicken, and human (TableI). The antibodies were purified from rabbit sera by sequential affinity chromatography steps as described under "Experimental Procedures." In fibroblast cells stably expressing the GFRα1 co-receptor and the long isoform of human RET (MG87-α1/RET), antibodies against Tyr(P)905, Tyr(P)1015, Tyr(P)1062, and Tyr(P)1096 specifically recognized phosphorylated RET in cells treated with GDNF, but not in untreated cells (Fig.1A). Similar results were obtained in a mouse motor neuron cell line (MN1) endogenously expressing GDNF receptors (data not shown). Competition experiments indicated that each of the antibodies was specific for the phosphorylated form of its cognate peptide, as only the corresponding phosphopeptide (but not the unphosphorylated peptide or other unrelated phosphopeptides) was able to block the detection of activated RET (Fig.1A). The specificity of the antibodies was further tested in fibroblast cells stably expressing mutant forms of human RET carrying specific amino acid replacements of cytoplasmic tyrosines, namely Y1015F, Y1062F, and Y1096F. In each case, detection of ligand-activated RET was abolished by mutation of the corresponding tyrosine residue to phenylalanine (Fig. 1B), whereas replacement of non-cognate tyrosines had no effect (data not shown). Because mutation of Tyr905 affects the kinase activity of the receptor, the specificity of the antibodies against Tyr(P)905 was tested in COS cells transiently overexpressing RET carrying the Y905F mutation. Overexpression in COS cells led to high levels of ligand-independent RET phosphorylation, even in the Y905F mutant, which could be detected with anti-phosphotyrosine antibodies (data not shown) or anti-Tyr(P)1015 antibodies (Fig. 1B), but not with antibodies against Tyr(P)905 (Fig. 1B). Because the peptides used for immunization contain sequence motifs that partially overlap with analogous sites in other tyrosine kinase receptors, we tested the ability of our antibodies to detect tyrosine phosphorylation in the neurotrophin-4 receptor TrkB, which, like RET, also contains phosphotyrosine docking sites for Shc, FRS2, and phospholipase Cγ. However, none of the four antibodies was able to recognize ligand-activated TrkB (Fig. 1C), indicating that they are indeed specific for the phosphorylation status of distinct tyrosine residues in RET. To study the kinetics of phosphorylation of individual tyrosine residues in RET after ligand stimulation, we used fibroblast cell lines expressing GFRα1 or GFRα3 together with the wild-type RET receptor (MG87-α1/RET and MG87-α3/RET, respectively) and MN1 cells expressing endogenous GFRα1, GFRα2, and RET receptors. GDNF stimulation elicited synchronized phosphorylation of tyrosines 905, 1015, 1062, and 1096 in MG87-α1/RET cells, corresponding to the pattern of total tyrosine phosphorylation detected by phosphotyrosine antibodies (Fig.2A). RET phosphorylation was maximal between 10 and 15 min after ligand stimulation and could still be detected after 120 min (Fig. 2A). Dephosphorylation of the 4 tyrosines following maximal activation also proceeded with comparable kinetics in all cases (Fig. 2A). Similar results were obtained in MN1 cells, except that tyrosine phosphorylation decayed more rapidly in these cells compared with fibroblast cells (Fig. 2B). Differences in the kinetics of receptor phosphorylation in different cell types could be due to different levels of receptor expression, as shown for the NGF receptor TrkA (26Hempstead B.L. Rabin S.J. Kaplan L. Reid S. Parada L.F. Kaplan D.R. Neuron. 1992; 9: 883-896Abstract Full Text PDF PubMed Scopus (285) Google Scholar), or to different complements of protein-tyrosine phosphatases. Phosphorylation of Tyr905, Tyr1015, and Tyr1062 was also detected in the short isoform of RET after immunoprecipitation from MN1 cells (data not shown). No differences could be seen between the two RET isoforms regarding activation of individual phosphotyrosine residues. We also investigated whether different ligands of the GDNF family could induce distinct patterns of phosphorylation of individual tyrosine residues in RET. Stimulation of MG87-α1/RET fibroblasts or MN1 cells with NTN resulted in a phosphorylation pattern very similar to that observed with GDNF (Fig. 2, C and D). Also ART, signaling via GFRα3 in MG87-α3/RET cells, induced a pattern of tyrosine phosphorylation comparable to those of GDNF and NTN (Fig.2E). Thus, we conclude that Tyr905, Tyr1015, Tyr1062, and Tyr1096become phosphorylated and dephosphorylated in a synchronized manner after ligand stimulation and that different GDNF family ligands utilizing different GFRα receptors induce comparable patterns of tyrosine phosphorylation. The effects of GDNF stimulation of RET in cis (i.e. GFRα1 expressed in the same cell) versus in trans (i.e. GFRα1 supplied exogenously) on the pattern of phosphorylation of individual tyrosines was examined in Neuro2A cells, a mouse neuroblastoma expressing endogenous RET, but little or no GFRα1. Treatment with GDNF alone produced no detectable RET phosphorylation in parental Neuro2A cells (data not shown). In Neuro2A cells stably transfected with GFRα1 (Neuro2A-α1), GDNF induced rapid and transient RET phosphorylation, which returned to basal levels 60 min after treatment (Fig. 3A). In contrast, stimulation of parental Neuro2A cells with GDNF and a soluble form of GFRα1 (GFRα1-Fc) resulted in a delayed but sustained phosphorylation of RET, which persisted for up to 120 min after treatment (Fig. 3B). Similar to the results observed after stimulation in cis, phosphorylation of Tyr905, Tyr1015, and Tyr1062 was synchronized following stimulation of Neuro2A cells with GDNF plus soluble GFRα1 (Fig.3B). In agreement with the pattern of total RET tyrosine phosphorylation, phosphorylation of individual tyrosines was delayed until ∼10 min and was sustained for up to 120 min (Fig.3B). Finally, we compared the patterns of RET tyrosine phosphorylation in MN1 cells expressing RET and GFRα1 treated with GDNF alone (in cis) versus GDNF plus soluble GFRα1 (in cis + in trans), a situatio
Referência(s)