Identification of RET Autophosphorylation Sites byMass Spectrometry*

Received for publication, November 18, 2003, and in revised form, January 3, 2004Published, JBC Papers in Press, January 6, 2004, DOI 10.1074/jbc.M312600200

Yoshiyuki Kawamoto‡, Kozue Takeda‡, Yusuke Okuno‡, Yoshinori Yamakawa§,Yasutomo Ito§, Ryo Taguchi¶, Masashi Kato�, Haruhiko Suzuki‡, Masahide Takahashi**,and Izumi Nakashima‡ ‡‡

From the ‡Department of Immunology, the §Equipment Center for Research and Education, and the **Department ofPathology, Nagoya University Graduate School of Medicine, Showa-ku, Nagoya, Japan, and the ¶Department ofMetabolome, Graduate School of Medicine, University of Tokyo, Hongo, Bunkyo-ku, Tokyo, Japan, and the�Department of Environmental and Preventive Medicine, Kanazawa University Graduate School of Medical Science,Takara-machi, Kanazawa, Japan

The catalytic and signaling activities of RET, a recep-tor-type tyrosine kinase, are regulated by the autophos-phorylation of several tyrosine residues in the cytoplas-mic region of RET. Some studies have revealed a fewpossible autophosphorylation sites of RET by [32P]phos-phopeptide mapping or by using specific anti-phospho-tyrosine antibodies. To ultimately identify these andother autophosphorylation sites of RET, we performedmass spectrometry analysis of an originally preparedRET recombinant protein. Both the autophosphoryla-tion and kinase activity of myelin basic protein as anexternal substrate of the recombinant RET protein weresubstantially elevated in the presence of ATP withoutstimulation by a glial cell line-derived neurotrophic fac-tor, a natural ligand for RET. Mass spectrometric anal-ysis revealed that RET Tyr806, Tyr809, Tyr900, Tyr905,Tyr981, Tyr1062, Tyr1090, and Tyr1096 were autophospho-rylation sites. Levels of autophosphorylation and kinaseactivity of RET-MEN2A (multiple endocrine neoplasia2A), a constitutively active form of RET with substitu-tion of Tyr900 by phenylalanine (Y900F), were compara-ble with those of original RET-MEN2A, whereas those ofthe mutant Y905F were greatly decreased. Interestingly,those of a double mutant, Y900F/Y905F, were completelyabolished. Both the kinase activity and transforming ac-tivity were impaired in the mutants Y806F and Y809F.These results provide convincing evidence for both pre-viously suggested and new tyrosine autophosphorylationsites of RET as well as for novel functions of Tyr806, Tyr809,and Tyr900 phosphorylation in both catalytic kinase activ-ities and cell growth. The significance of the identifiedautophosphorylation sites in various protein-tyrosine ki-nases registered in a data base is discussed in this paper.

Activation of receptor-type or nonreceptor-type protein tyro-sine kinases (PTKs)1 plays a pivotal role in signal transduction

in cells. The catalytic activity of PTKs is known to be regulatedeither positively or negatively by phosphorylation/dephospho-rylation at specified tyrosine residues. RET is a proto-oncogenethat encodes a receptor-type PTK with a cadherin-related motifand a cysteine-rich domain in the extracellular domain (1–3). Itis known that rearrangement of RET is the major cause ofpapillary thyroid carcinoma (4) and that germ line single-pointmutations of RET cause multiple endocrine neoplasia 2A(MEN2A), MEN2B, familial medullary thyroid carcinoma, andHirschsprung’s disease (5–10). The glial cell line-derived neu-rotrophic factor (GDNF) family ligands (GFLs), includingGDNF, neurturin, artemin, and persephin, are involved in thesurvival and differentiation of neurons via activation of RET(11–14). Unlike most other receptor-type PTKs, RET is notactivated via direct binding of GFLs but is activated by theformation of a multicomponent receptor complex that includesglycosylphosphatidylinositol-anchored cell surface proteinscalled GFR�s, which bind GFLs directly. Four members of thefamily of GFR�s (GFR�1 to -4) have been identified, and it hasbeen shown that they bind GFLs with high affinity. GDNF,neurturin, artemin, and persephin use GFR�1, GFR�2,GFR�3, and GFR�4, respectively, as preferred ligand-bindingreceptors (11–14).

RET can activate various signaling pathways, including ex-tracellular signal-regulated kinase, phosphatidylinositol 3-ki-nase/AKT, p38 mitogen-activated protein kinase, and c-JunN-terminal kinase pathways, responsible for cell survival ordifferentiation (15–18). Upon ligand binding, RET formsdimers and is phosphorylated at a specific tyrosine residue(s) inintracellular domains. RET has been shown to be alternativelyspliced to produce three isoforms (short, middle, and long iso-forms) that differ in C-terminal residues (19, 20). RET shortand middle isoforms contain 16 tyrosine residues in their in-tracellular domains, and RET long isoforms have two addi-tional tyrosines in the C-terminal tail. Among these tyrosines,Tyr905, Tyr1015, Tyr1062, and Tyr1096 are thought to be phos-phorylated to become binding sites for GRB7/GRB10, phospho-lipase C�, SHC/SNT(FRS2)/Enigma, and GRB2, respectively(21–26). For neuronal signal transduction, one model has sug-gested that GFR� functions to mediate localization of RET to alipid raft, a sphingolipid, and cholesterol-rich lipid microdo-

* This work was supported in part by grant-in-aid JSPS Fellows, bygrants-in-aid for Scientific Research B and Scientific Research of Pri-ority Areas from the Ministry of Education, Science, Sports and Cultureof Japan, and by funds for comprehensive research on aging and healthfrom the Ministry of Health and Welfare of Japan. The costs of publi-cation of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

‡‡ To whom correspondence should be addressed: Dept. of Immunol-ogy, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho,Showa-ku, Nagoya 466-8550, Japan. Tel.: 81-52-744-2134; Fax: 81-52-744-2972; E-mail: inakashi@med.nagoya-u.ac.jp.

1 The abbreviations used are: PTK, protein-tyrosine kinase; GDNF,

glial cell line-derived neurotrophic factor; RET-MEN2A, RET with mul-tiple endocrine neoplasia type 2A mutation; MBP, myelin basic protein;GST, glutathione S-transferase; MALDI-TOF, matrix-assisted laser de-sorption ionization time-of-flight; FAMS, full automatic modeling sys-tem; GFL, GDNF family ligand; MS, mass spectrometry.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 279, No. 14, Issue of April 2, pp. 14213–14224, 2004© 2004 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 14213

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

main in the plasma membrane. GFR� stimulation via GDNFinduces not only RET dimerization but also RET/Src associa-tion at the raft, and RET can be phosphorylated by activatedSrc kinase (27, 28).

In addition to endogenous ligands, oxidative agents such asH2O2, NO, HgCl2, and ultraviolet rays are known to increasethe levels of catalytic activity of various PTKs, including Src(29–32), Lck (33–35), and RET (36–38). Recently, we demon-strated that Cys987, which is in the kinase domain of RET,plays a critical role in the up-regulation of kinase activity uponstimulation by an oxidative agent (37, 38). It is known thatsome cysteine residues, including Cys987 of RET or its equiva-lents, are widely conserved among a number of receptor-typeand nonreceptor-type PTKs and some serine/threonine kinases(39). By analyzing amino acid sequences of various PTKs reg-istered in a data base, we found that Cys987 exists in the motifsequence of Met-X-X-Cys-Trp (MXXCW), which is located inthe C-end of �-helix H in the kinase domain (40), and that thismotif initially turns on the Tyr905-dependent local switch ofRET kinase for activation through oxidation of Cys987 in theMXXCW motif (40). Thus, after accumulation of RET proteinson the plasma membrane induced by ligand or potentially byoxidative stimulation, autophosphorylation that leads to ki-nase activation and downstream signaling will occur.

The tyrosine residues Tyr1015, Tyr1062, and Tyr1096 in acti-vated RET were suggested to be autophosphorylated in earlierstudies by phosphopeptide mapping (41) or by using a specificantibody recognizing phosphotyrosine (42). Results of studiesusing an anti-phosphotyrosine antibody also suggested thatTyr905 is an autophosphorylation site (43, 44). However, tyro-sine residues in RET protein were not identified as autophos-phorylation sites in earlier studies by reliable methods. In thisstudy, we performed mass spectrometry analysis of purifiedrecombinant RET proteins in order to determine RET auto-phosphorylation sites in the purified RET protein activated byATP, and we found that Tyr806, Tyr809, Tyr900, Tyr905, Tyr981,Tyr1062, Tyr1090, and Tyr1096 are bona fide autophosphorylationsites that are actually autophosphorylated. Phosphorylation ofTyr806, Tyr809, Tyr900, Tyr981, and Tyr1090 in RET is reportedhere for the first time. Furthermore, this study demonstratedfor the first time that Tyr806, Tyr809, and Tyr900, which arerelatively highly conserved in a number of PTKs registered ina data base, play roles additional or supplemental to the role ofTyr905 in catalytic activity.

EXPERIMENTAL PROCEDURES

Cell Culture—Murine NIH3T3 cells were grown in Dulbecco’s mod-ified Eagle’s medium supplemented with 10% fetal bovine serum. InsectSpodoptera frugiperda (Sf21) cells were grown in Grace’s insect me-dium supplemented with 10% fetal bovine serum.

Construction of Recombinant Baculoviruses—A cDNA clone contain-ing the entire sequence of the human c-RET gene inserted into theRc/CMV vector (Invitrogen) was digested by EcoRI and NotI ultimatelyto make glutathione S-transferase (GST)-fused plasma membrane andextracellular regions-deleted RET, and then a 1.6-kb fragment wassubcloned into the EcoRI and NotI sites of a baculovirus transfer vector,pAcGHLTTM (Pharmingen). The transfer vector was then co-trans-fected with linearized BaculGoldTM (Pharmingen) DNA into Sf21 cellsto generate a baculovirus that can produce a primary recombinantGST-fused deletion RET protein, designated GST-�RET. The GST-�RET baculovirus was then amplified and maintained in Grace’s insectmedium at 4 °C until used for infection of Sf21 cells to generate largeamounts of recombinant proteins.

Site-directed Mutagenesis—The cDNA encoding human RET-MEN2A with substitution of cysteine 634 by arginine in c-RET wasprepared as described previously (28). The RET-MEN2A gene wassubcloned into the APtag-1 plasmid vector containing the Moloneymurine leukemia virus long terminal repeat. RET-MEN2A single mu-tants, Y806F, Y809F, Y900F, and Y905F, and double mutants, Y806F/Y809F and Y900F/Y905F, were prepared by using a QuikChangeTM

site-directed mutagenesis kit (Stratagene) according to the manufac-turer’s instructions. To confirm the introductions of mutations, theplasmids were sequenced by a DNA sequencer ABI prism (AppliedBiosystems).

Purification and Characterization of GST-�RET—We purified GST-�RET recombinant protein generated by the baculovirus system byusing GSH-agarose beads. The kinase activity of the purified GST-�RET was measured by using an ELISA-based tyrosine kinase activitykit (Chemicon International) according to the manufacturer’s protocol.Briefly, the purified GST-�RET bound to GSH-agarose beads was in-cubated in the absence or presence of ATP (1 mM) in kinase buffer (10mM MgCl2, 5 mM Tris-HCl (pH 7.2), 30 mM NaCl) with a syntheticbiotinylated peptide substrate (poly(Glu:Tyr), 4:1) at 30 °C for 20 min.The enzyme reaction was terminated by adding 24 mM EDTA, and thepeptide substrate samples in the supernatant were then transferred tostreptavidin-coated wells, followed by incubation at 37 °C for 30 min.After blocking the wells with 3% bovine serum albumin in phosphate-buffered saline, a horseradish peroxidase-conjugated anti-phosphoty-rosine antibody was added to the wells, and then tetramethylbenzidineas an horseradish peroxidase substrate was added. Finally, absorbanceof sample wells was measured at A450. The residual GST-�RET samplesthat bound to GSH-agarose beads after ATP reaction were mixed with2� sample buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol,0.2% bromphenol blue, pH 6.8, 10% �-mercaptoethanol) and subjectedto SDS-PAGE. The levels of tyrosine phosphorylation were determinedby immunoblotting as described below.

Preparation of Phosphopeptides of GST-�RET—GST-�RET bound toGSH-agarose beads (20–30 �g) was washed with kinase buffer (10 mM

Tris-HCl (pH 7.4), 5 mM MgCl2) four times and incubated in the absenceor presence of ATP (1 mM) in kinase buffer for 30 min at 30 °C. TheATP-treated purified protein was washed three times with digestionbuffer (50 mM NH4HCO3, pH 7.8) followed by overnight digestion at24 °C in the presence of Glu-C endoproteinase (8 ng/�l) (Roche AppliedScience) or at 37 °C in the presence of sequencing grade modifiedtrypsin (8 ng/�l) (Promega). The collected phosphopeptides were thenincubated in the presence or absence of phosphatases, leukocyte anti-gen-related phosphatase (30 min at 30 °C, in reaction buffer; 25 mM

Tris-HCl (pH 7.0), 50 mM NaCl, 2 mM Na2EDTA, 5 mM dithiothreitol) orprotein phosphatase 1 (30 min at 30 °C, in reaction buffer; 50 mM

Tris-HCl (pH 7.0), 0.1 mM Na2EDTA, 5 mM dithiothreitol, 0.025%Tween 20, 1 mM MnCl2) (New England BioLabs).

Identification of Tyrosine Autophosphorylation Sites by Mass Spec-trometry—The sample peptides were desalted by using C18 Zip Tip(Millipore Corp.), eluted in �-cyano-4-hydroxycinnamic acid solutioncontaining 50% acetonitrile and 3% folic acid, and then analyzed byusing a Voyager Elite MALDI-TOF mass spectrometer (PerSeptiveBiosystems) operated in the positive ion reflectron mode. For molecularmass determination of GST-�RET peptides, two-point external calibra-tion based on angiotensin II (m/z 1046.54) and ACTH fragment peptide18–39 (m/z 2465.20) was used. For ion trap mass spectrometric analy-sis, a mixture of protease-digested peptides was desalted with C18 ZipTip and eluted in 98% acetonitrile solution containing 0.1% formic acid.The eluate was mixed with the same volume of 2% acetonitrile solutioncontaining 0.1% formic acid, and then the sample was directly infusedat 1 �l/min into an ion trap mass spectrometer, LCQ Advantage(Thermo Electron). The MS/MS spectra were analyzed by using theMS/MS ion searching program of MASCOT (Matrix Science; availableon the World Wide Web at www. matrixscience.com) and Bioworks-Browser (version 3.0) (Thermo Electron) to assign individual masssignals, and then amino acid sequences were determined.

Molecular Modeling—Primary amino acid sequences of human PTKfamily proteins were obtained from the SWISS-PROT protein database. The tertiary structure of RET was simulated and determinedusing FAMS, a homology modeling software program, originally pro-duced by H. Umeyama at Kitasato University, based on the tertiarystructure of fibroblast growth factor receptor 1 elucidated by x-raydiffraction. Subsequently, the structure was visualized with WebLabViewerLite 3.20 software (Molecular Simulations Inc.).

Immunoblotting and Antibodies—Western blotting was performedaccording to the method described previously (45). An anti-RET rabbitpolyclonal antibody that recognizes both a 175-kDa form (a matureglycosylated form) and a 155-kDa form (an immature glycosylated form)of RET was produced as described previously (46). The GST-�RETsamples or the lysates (20 �g/lane) of NIH3T3 cells were subjected toSDS-PAGE using 8% polyacrylamide gel and transferred to a polyvi-nylidene difluoride membrane (Millipore Corp.). The membrane wasblocked with 5% nonfat milk for 1 h at room temperature and thenreacted with a first anti-phosphotyrosine or anti-RET antibody for 12 h

Identification of Autophosphorylation Sites of RET14214

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

at 4 °C followed by reaction with a second anti-rabbit horseradishperoxidase-conjugated antibody for 1 h at room temperature. The levelsof the phosphotyrosine or the RET protein signal were detected byWestern blot chemiluminescence reagent (PerkinElmer Life Sciences),and densitometry analysis was performed by using Scion Image soft-ware (Scion Corp.).

Immunoprecipitation and in Vitro Kinase Assay—Immunoprecipita-tion and an in vitro RET kinase assay were performed as describedpreviously (47). Briefly, RET proteins were immunoprecipitated withProtein A beads (Pierce) and washed three times with lysis buffer (30mM Tris-HCl (pH 8.0), 1% Triton X-100, 150 mM NaCl, 1 mM EDTA, 0.5mM Na3VO4). The immunoprecipitated RET proteins were washedthree times with kinase buffer (10 mM Tris-HCl, pH 7.4, 5 mM MgCl2),suspended in the kinase buffer with 2 �g of myelin basic protein (MBP)(Sigma) as an exogenous substrate, and radiolabeled with [�-32P]ATP(370 kBq) (PerkinElmer Life Sciences). The kinase reaction was carriedout for 20 min in a 30 °C water bath and was terminated by adding 2�sample buffer (62.5 mM Tris-HCl (pH 6.8), 2% SDS, 10% glycerin) with2-mercaptoethanol. The immunoprecipitates were then boiled for 5 minand loaded on 12.5% SDS-polyacrylamide gels. The gels were dried andexposed to Imaging Plate (Fuji photo film), and the radioactive signalswere detected by using a Typhoon 8600 imaging system (AmershamBiosciences). Densitometry analysis of signal levels of phosphorylatedMBP was performed by using Scion image software.

Transformation Assay—NIH3T3 cells were transfected with RET-MEN2A and RET-MEN2A mutant cDNAs, and several G418-resistantcell clones were isolated. Trypsinized 1 � 105 cells were mixed with 2 mlof 0.36% soft agar in Dulbecco’s modified Eagle’s medium, poured ontoslightly solid 0.72% hard agar in Dulbecco’s modified Eagle’s medium,and then cultured in a 5% CO2 incubator for 2 weeks. The number ofaggregated cell foci in a dish was counted, and colony-forming efficiency(percentage � (number of colonies)/(number of seeded cells) � 100)was calculated.

RESULTS

Identification of Tyrosine Autophosphorylation Sites—In or-der to investigate the autophosphorylation sites in the intra-cellular domain of RET, we made a deletion mutant RET genein which the plasma membrane and extracellular regions weredeleted (designated as �RET). We then connected glutathioneS-transferase (GST) with the N terminus of �RET (Fig. 1) andconstructed a baculovirus protein expression system. Silverstaining was performed to evaluate the purity of GST-�RET.As shown in Fig. 2A, GST-�RET protein was highly purifiedbut was contaminated by a small amount of endogenous GST,as shown by Western blotting using an anti-GST antibody(data not shown). By Western blotting using an anti-phospho-tyrosine antibody, a tyrosine-phosphorylated protein thatmight be coprecipitated with GST-�RET was observed at about40 kDa. Since there in no known tyrosine kinase of less than 40

kDa, we assumed that non-PTK phosphoproteins were phos-phorylated by GST-�RET and that any phosphorylation thatnewly occurred in vitro with purified GST-�RET resulted fromautophosphorylation. Next, the levels of autophosphorylationof GST-�RET were examined by an in vitro kinase assay. Asshown in Fig. 2B, both levels of autophosphorylation of purifiedGST-�RET and tyrosine phosphorylation of the synthetic sub-strate peptide were greatly increased when ATP was added tothe in vitro kinase system, and these increases were preventedby treating the sample with �-protein phosphatase before theaddition of ATP. These results suggested that the recombinantRET has an enhanced activity in vitro for autophosphorylationin the absence of any ligands and can be used for analyzingtyrosine autophosphorylation sites.

We extracted phosphorylated peptides from in vitro auto-phosphorylation-promoted GST-�RET that had been treatedwith Glu-C endoprotease. The samples were then analyzed byMALDI-TOF mass spectrometry to identify the phosphorylatedpeptides derived from GST-�RET. Because of the negativecharge of phosphate groups, phosphopeptides exhibit low re-sponse to mass spectrometry in the positive ion mode. Never-theless, several substantial intensities of mass spectra corre-sponding to phosphorylated peptides containing tyrosineresidue(s) from the Glu-C endoproteinase-treated GST-�RETwere obtained: Tyr806 � Tyr809 as 1718.53, Tyr900 as 2209.59,Tyr905 as 2343.69, Tyr981 as 3419.05, and Tyr1062 as 1687.41(Fig. 3, A and C). For confirmation of tyrosine phosphorylationof these peptides, mass spectrometry analysis was also per-formed with Glu-C endoproteinase-digested phosphopeptidesfrom GST-�RET that had been treated with leukocyte antigen-related protein-tyrosine phosphatase. As shown in Fig. 3, B andD, the masses of tyrosine-dephosphorylated peptides contain-ing Tyr806 � Tyr809, Tyr900, Tyr905, Tyr981, and Tyr1062 wereshifted to 1559.97, 2130.07, 2263.08, 3339.74, and 1607.80,respectively, confirming that the tyrosines on these peptideswere autophosphorylated. We also performed mass spectrome-try analysis of peptides digested with Glu-C that had beentreated with protein serine/threonine phosphatase 1, whichresulted in development of no significant dephosphorylatedmasses for the peptides containing Tyr806 � Tyr809, Tyr900,Tyr905, Tyr981, and Tyr1062 (data not shown).

MS/MS spectrometry analysis was also performed for endo-protease Glu-C-digested GST-�RET peptides containing Tyr900

or Tyr905. Putative triply charged phosphopeptide signals con-taining Tyr900 and Tyr905 were observed as m/z 737.5 and782.5, respectively, by using the full-scan mode (m/z 150–2000). Subsequently, mass spectra of each parent ion wereobtained, and those of amino acid sequences were determinedusing the MASCOT MS/MS ion searching program and Bio-worksBrowser software. The triply charged peptides were iden-tified mainly by a doubly charged b-type ion in the MS/MSspectra as GRKMKISDFGLSRDVpYEE (where pY representsphosphotyrosine) (Tyr900) (Fig. 4A) and DSpYVKRSQGRIPVK-WMAIE (Tyr905) (Fig. 4B).

We also performed MALDI-TOF mass spectrometry analysisof phosphopeptides that had been digested with another prote-ase, trypsin (Fig. 3, E–H). Corresponding to the Glu-C results,putative phosphorylated fragments containing Tyr900 � Tyr905

(m/z 1406.77), Tyr981 (m/z 1739.04) and Tyr1062 (m/z 2195.32)were observed, although fragments containing Tyr806 or Tyr809

were not detected. In addition, a new putative phosphopeptidecontaining Tyr1090 � Tyr1096 (m/z 2174.22) was observed. Ty-rosine phosphorylation(s) within these phosphopeptides wasconfirmed by treatment with leukocyte antigen-related phos-phatase (Fig. 3, F and H).

Taken together, the results demonstrated that Tyr806,

FIG. 1. Preparation of GST-fused recombinant RET andMEN2A mutants. A c-RET gene fragment for the cytosolic regionobtained by digestion with EcoRI and NotI was subcloned into thebaculovirus expression vector pAcGHLT in which a RET insert wasfused with the GST gene (designated GST-�RET). The RET-MEN2Agene was constructed by replacement of cysteine 634 with arginine ofc-RET. Mutant RET-MEN2A cDNAs in which tyrosine at codon 806,809, 900, or 905 was replaced by phenylalanine were then constructed.TK1, tyrosine kinase domain 1; TK2, tyrosine kinase domain 2.

Identification of Autophosphorylation Sites of RET 14215

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Tyr809, Tyr900, Tyr905, Tyr981, Tyr1062, Tyr1090, and Tyr1096 areautophosphorylation sites. The MALDI-TOF mass spectra areshown in Table I.

The RET protein has 18 tyrosines in the intracellular do-main: 2 in the vicinity of the transmembrane, 11 in the intra-cellular kinase domain, including Tyr806, Tyr809, Tyr900, Tyr905,and Tyr981, and 5 in the carboxyl-terminal tail, includingTyr1062, Tyr1090, and Tyr1096. Tyr905, Tyr1015, Tyr1062, andTyr1096 residues were previously suggested to be autophospho-rylation sites by using phosphospecific antibodies after GDNFstimulation (42–44) and in part by phosphopeptide mapping(41). However, there have been no reports on the phosphoryl-ation of Tyr806, Tyr809, Tyr900, Tyr981, and Tyr1090. For Tyr981,we previously reported that both the levels of autophosphoryl-ation and kinase activity of RET-MEN2A Y981F mutant weresignificantly decreased (28).

Data Base-oriented Evaluation of the Significance of Ty-rosines Determined as Autophosphorylation Sites—Since thetertiary structure of RET has not yet been determined, weperformed in silico analysis by using FAMS computer simula-tion software to predict the structure. As shown in Fig. 5, thepredicted RET structure was constructed by homology molec-ular modeling and visualized. The calculated tertiary structureof RET was found to be similar to that of c-Src (48). RET carriedan N-lobe consisting of five �-sheets and one �-helix and aC-lobe consisting of two �-sheets and seven �-helices. Thetyrosines that we identified as autophosphorylation sites werepositioned in this tertiary structure model. As shown in Fig. 5,Tyr806 and Tyr809 existed in the contralateral side of the acti-vation loop, in which Tyr900 and Tyr905 existed. Tyr981 was veryclose to the MXXCW motif (40), which is on an �-helix of theC-lobe. Tyr1062, Tyr1090, and Tyr1096 were not shown in the

predicted RET structure because of less similarity betweenRET and fibroblast growth factor receptor 1 as the source ofstructural information.

To determine which tyrosine residues are conserved in thehuman PTK family, a search of the SWISS-PROT data basewas made, and human PTKs were manually aligned accordingto previously described methods (49, 50). In the data base, wefound 101 proteins that are registered as human PTK familyproteins. 21 of those 101 PTKs have serine and/or threoninekinase activity in addition to tyrosine kinase activity; hence,these proteins were excluded for the comparison, and 80 pro-teins were examined (Fig. 6). We previously reported that themotif sequence of MXXCW, which is located in the C-end of�-helix H in the kinase domain, is highly conserved in allexcept two (BLK and FGR) of the 80 genes for human PTKs(40). Interestingly, 61 (76.25%) of the 80 human PTK genesencoding the MXXCW motif had a conserved tyrosine residuecorresponding to Tyr905 of RET that is positioned 79 residuesfrom the MXXCW motif and is located in the putative activa-tion segment of the protein kinase domain. In addition, 41genes (51.25%), 15 genes (18.75%), 21 genes (26.25%), and 58genes (72.5%) had conserved tyrosine residues corresponding toTyr806, Tyr809, Tyr900, and Tyr981, respectively. These observa-tions suggest that all of the tyrosines as autophosphorylationsites play critical roles in the regulation of the kinase activationthat are conserved evolutionary.

In Vitro Kinase Assays of RET-MEN2A Mutants—We nextfocused on the functions of Tyr806, Tyr809, Tyr900, and Tyr905

and prepared phenylalanine mutants of them for kinase activ-ity assays. The mutant plasmids were transiently expressed inNIH3T3 cells for performing in vitro kinase assays. As shownin Fig. 7, A and B, the levels of both autophosphorylation and

FIG. 2. Autophosphorylation of pu-rified recombinant RET protein. A,GST-�RET was purified by using GSH-agarose beads and incubated for 30 min at30 °C with or without ATP (1 mM) in ki-nase buffer containing MgCl2. These sam-ples were then subjected to SDS-PAGEfollowed by silver staining for the detec-tion of overall proteins and by Westernblotting (IB) for the detection of tyrosine-phosphorylated proteins. B, in vitro ki-nase assay was performed with ATP-treated or untreated purified GST-�RETfor determination of the levels of PTK ac-tivity to phosphorylate a synthetic sub-strate (B-1) and of autophosphorylation(B-2) according to the method describedunder “Experimental Procedures.” A por-tion of the sample (right lane) was treatedwith �-protein phosphatase (�-PPase)(100 units) for 30 min at 37 °C. Sodiumorthovanadate (0.5 mM) was added to the�-protein phosphatase-treated sample toinhibit the phosphatase activity prior tothe addition of ATP. Quantitation of thetyrosine phosphorylation level of GST-�RET was performed by using Scion Im-age software. Data represent means �S.D. of three independent experiments.

Identification of Autophosphorylation Sites of RET14216

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 3. Identification of RET phosphopeptides by MALDI-TOF mass spectrometry. ATP-treated GST-�RET was digested with V8protease (a Glu-C endoproteinase) (A–D) or trypsin (E–H), and phosphorylated peptides in the samples were determined by MALDI-TOF massspectrometry. To confirm tyrosine phosphorylation, portions of peptide samples were treated with leukocyte antigen-related tyrosine phosphatase(B, D, F, and H) at 30 °C for 30 min as described under “Experimental Procedures.” Results shown are representative of more than threeindependent experiments.

Identification of Autophosphorylation Sites of RET 14217

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 4. Analysis of phosphopeptidesby ion trap mass spectrometry. Phos-phopeptide samples containing Tyr900 orTyr905 were obtained from V8 protease-digested products from GST-�RET as de-scribed under “Experimental Proce-dures.” MS/MS spectra from candidates ofphosphopeptides were acquired by usingan ion trap mass spectrometer, and aminoacid sequences were determined by usingthe MS/MS ion search program of MAS-COT and BioworksBrowser software. Apartial fragment ion spectrum of a triplycharged precursor ion of phosphopeptidesat m/z 737.5 was obtained (A) and identi-fied as a tyrosine-phosphorylated peptideat Tyr900 from MS/MS signal analysis.MS/MS spectra of phosphopeptides at m/z782.5 were analyzed in a similar manner(B), resulting in identification of tyrosinephosphorylation at Tyr905. The resultsshow some characteristic y-type and b-type ion products. (bn)�2, a doublycharged b-ion series product. The peaksdenoted by bn0 and bn* are the results ofwater (�18 Da) and ammonia (�17 Da)losses from the corresponding ion,respectively.

Identification of Autophosphorylation Sites of RET14218

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

MBP phosphorylation of the Y806F mutant were decreased toabout 65% of those of RET-MEN2A, which is a constitutivelyactive form of RET. Those levels of the Y809F mutant werecomparable with those of the Y806F mutant. For the Y806F/Y809F double mutant, the levels of both autophosphorylationand MBP phosphorylation tended to be more decreased thanthose of the Y806F or Y809F single mutant, although thedifference between the levels of the single mutant and doublemutant was not significant. These findings suggest that Tyr806

and Tyr809 contribute to the kinase activity of RET.

Among the 80 PTKs, all human PTK genes except CSK hadone or more tyrosine residue near the segment that contains atyrosine(s) corresponding to RET Tyr900 or Tyr905, suggestingthat such tyrosine residues in the activation loop are essentialfor protein kinase activation as was previously reported (48,49). In fact, autophosphorylation of Tyr416 of Src, which isequivalent to Tyr905 of RET, is likely to work as a local switchto induce a conformational change required for kinase activa-tion (48). We previously showed that the RET-MEN2A-Y905Fmutant had reduced but not completely abolished autophos-

TABLE ISummary of tyrosine-phosphorylated peptides in RET analyzed by MALDI-TOF mass spectrometry

GST-�RET treated with ATP was digested with V8 protease or trypsin, and portions of peptide samples were treated with tyrosine phosphatase(LAR). The peptide samples were analyzed by MALDI-TOF mass spectrometry as described under “Experimental Procedures”. The mass valuesobtained were compared with the theoretical values calculated from known amino acid sequences.x

FIG. 5. Tertiary structure of RETprotein simulated by FAMS com-puter software. Since a structural ho-mologue with the extracellular and trans-membrane region of RET does not exist ina protein structure data base (ProteinData Bank), the intracellular region ofRET was simulated by FAMS. This terti-ary structure was constructed by homol-ogy modeling based on fibroblast growthfactor receptor 1 (Protein Data Bank code1FGI), which shares the highest aminoacid homology with RET. The stereo viewof this figure was made with WebLabViewerLite 3.20 software. The �-sheetand �-helix are colored light blue and red,respectively. Positions of identified auto-phosphorylation sites are shown. The pu-tative ATP-binding site is also indicated.

Identification of Autophosphorylation Sites of RET 14219

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

phorylation and kinase activities. We therefore hypothesizedthat Tyr900 phosphorylation supplements the function of Tyr905

that works as an initial local switch. To test this hypothesis, wemade not only Y900F and Y905F mutants but also a Y900F/Y905F double mutant of the RET-MEN2A construct, whichwere transiently expressed in NIH3T3 cells for performing invitro kinase assays. Corresponding to our previous data, thelevels of both autophosphorylation and phosphorylation of anexogenous substrate (MBP) of the Y905F mutant of RET-MEN2A were significantly reduced, but not completely abol-ished, compared with those of the original RET-MEN2A. Thelevels of RET autophosphorylation and MBP phosphorylationof the Y900F mutant were nearly equal to those of the originalRET-MEN2A, but those of the Y905F/Y900F double mutantwere almost completely abolished (Fig. 7D). These results sug-gest that Tyr905 in RET-MEN2A plays a pivotal role in theactivation of RET-MEN2A for both autophosphorylation andMBP phosphorylation and that phosphorylation of Tyr900 sup-plements the function of Tyr905 as a local switch for increasingcatalytic activity, presumably by stabilizing the conformationof activated RET protein.

Transforming Activity of RET-MEN2A Mutants—We previ-ously reported that the RET-MEN2A-Y905F mutant dramati-

cally reduced the transforming ability of NIH3T3 cells on softagar medium compared with the RET-MEN2A transfectant,whereas the RET-MEN2A-Y900F mutant did not. We per-formed transforming assays to examine the effect on down-stream signaling in MEN2A-Y806F or MEN2A-Y809F mu-tants. As shown in Fig. 8, the colony-forming efficiency of stabletransformants of both RET-MEN2A-Y806F and Y809F mu-tants was about 50% lower than that of the original RET-MEN2A. As expected, the efficiency of Y905F was drasticallyreduced. These results coincide with the kinase activities, sug-gesting that both Tyr806 and Tyr809 are needed for the kinase toexert maximum protein kinase activity that leads to excesscell growth.

DISCUSSION

In this study, we used recombinant RET protein without theplasma membrane and extracellular regions, which containsthe kinase domain, and MALDI-TOF and/or ion trap massspectrometry for identification of RET autophosphorylationsites. We used this strategy because it has at least two advan-tages. First, the purified RET protein is suitable for defining“auto” tyrosine phosphorylation sites in vitro. Physiologically,RET protein is activated by stimulation with a ligand such as

FIG. 6. Comparison of PTKs sequences. Eighty human PTKs were found in the SWISS-PROT data base, and the positions of the tyrosinesin these PTKs equivalent to Tyr806, Tyr809, Tyr900, Tyr905, and Tyr981 in RET from the MXXCW motif are displayed. The sequences were comparedwith a part of RET (G798-L1005). Gaps, represented by dashes, were introduced into the sequences to optimize the alignment. The putativeactivation loop domain of PTKs is displayed in a red box enclosing the centers of sequences. The MXXCW motif is shown as a red box on the rightside of sequences. The alphanumeric characters on the left side of individual sequences represent human protein names registered in the data base.The tyrosine residues equivalent to RET Tyr806, Tyr809, Tyr900, Tyr905, and Tyr981 are colored red.

Identification of Autophosphorylation Sites of RET14220

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

GDNF that prompts a monomeric RET protein to gain access toeach other and to form a dimer. In this event, several otherproteins, including Src family PTKs, would accumulate aroundRET, which should phosphorylate RET as a substrate. In fact,Tansey et al. and other groups (27, 51, 52) demonstrated thatGFR�1 recruits RET to a lipid raft, a detergent-insoluble cho-lesterol-rich lipid microdomain enriched with signaling pro-teins, which results in RET/Src association after GDNF stim-ulation. Recently, we also demonstrated that RET-MEN2Aprotein forms a complex with Src tyrosine kinase in a lipid raft,which phosphorylates Tyr905 in RET in trans (28). In thisstudy, GST-�RET protein was highly purified by GSH-agarosebeads, although there was slight contamination of intrinsicGST and coprecipitated proteins of less than 40 kDa (see Fig.2A). This ruled out the possibility of an effect of trans-actingSrc or other PTKs on GST-�RET. We showed that the recom-binant RET protein on GSH-agarose beads, which allowed pro-tein-protein interaction in the absence of a ligand, had kinaseactivity and was autophosphorylated by itself (Fig. 2). Thissystem was therefore considered to be suitable for analyzingautophosphorylation sites. Second, MALDI-TOF MS andMS/MS analyses directly demonstrate the existence and exactpositions of phosphate groups in protease-digested peptides.For digestion of RET protein, we used two kinds of protease,trypsin and protease V8, to confirm the identification of phos-phopeptides by MALDI-TOF MS. Then the mass identical tothe phosphopeptide was analyzed by ion trap MS/MS and de-fined the sequence of amino acid with a phosphate group.

There have been several attempts to determine autophos-

phorylation sites of RET by [32P]phosphopeptide mapping (41)or by using phosphotyrosine-specific antibodies (42–44), andthe results of those studies have suggested, but not proved,that Tyr905 (43), Tyr1015 (41, 42), Tyr1062 (41, 43), and Tyr1096

(41, 44) are phosphorylated. The former technique includestwo-dimensional [32P]phosphopeptide mapping followed bypeptide sequencing by Edman sequencing. These proceduresare rather tedious, and Edman sequencing is methodologicallyproblematic, especially for determining phosphotyrosine sites,because phosphotyrosine is almost insoluble in the conven-tional transfer solvents used in an Edman sequenator, yieldinga gap in the sequence course (53). Actually, this techniquefailed to demonstrate any autophosphorylation sites inside thekinase domain. The latter technique, which is fully based onspecificity of phosphotyrosine-reactive antibodies, does not pro-vide direct evidence. Neither of these techniques is thereforesuitable for identifying autophosphorylation sites directly andexactly. Our data obtained by mass spectrometry have pro-vided for the first time solid evidence that Tyr806, Tyr809,Tyr900, Tyr905, and Tyr981 in the kinase domain and Tyr1062,Tyr1090, and Tyr1096 in the tail are autophosphorylated.

The tertiary structure of RET, which was simulated byFAMS software (Fig. 5), was found to be similar to that of Src(48) or insulin receptor tyrosine kinase (54). The molecularmechanism of the kinase activation in relation to the functionof tyrosine residues of RET might therefore be comparable withthat of Src or insulin receptor tyrosine kinase. Like Src, RETcarries an N-lobe and a C-lobe with an activation loop thatcarries Tyr905. Tyr905 in RET, which was convincingly demon-

FIG. 6—continued

Identification of Autophosphorylation Sites of RET 14221

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

strated to be an autophosphorylation site in this study, isequivalent in position to Tyr416 in Src, which is known as themajor autophosphorylation site principally regulating auto-phosphorylation and catalytic activity of Src kinase. Phospho-rylation of Tyr416 in Src is thought to induce an alteration inthe three-dimensional structure of the activation segment ofthe kinase domain to make it suitable for transferring ATPfrom the ATP-binding site to tyrosine residue(s) on the sub-strate. In fact, it was previously shown that the Y905F mutant

of RET-MEN2A has greatly reduced catalytic and cell-trans-forming activities (28). Taken together, the results suggest thatTyr905 is the major autophosphorylation site of RET in terms ofposition, function, and ability to bind phosphorus.

In this study, we demonstrated that Tyr806, Tyr809, Tyr900,Tyr981, and Tyr1090 are new autophosphorylation sites of RET.We first focused our functional study on Tyr900 as a newlyidentified autophosphorylation site in the activation loop inrelation to Tyr905 as the major autophosphorylation site. Therehas been no report on the role of Tyr900 or phosphorylatedTyr900 of RET. Based on the fact that Tyr900 is close to Tyr905

in the activation loop and the fact that the mutant Y905Fretains low levels of autophosphorylation and kinase activity(Fig. 7) (28), we speculated that Tyr900 can partially replace thefunction of Tyr905 as a local switch for kinase activation. Asexpected, both the autophosphorylation and kinase activitywere almost completely abolished in the Y900F/Y905F doublemutant of RET-MEN2A (Fig. 7D), although the Y900F mutantitself showed activities comparable with those of wild-typeRET-MEN2A.

Tyr1158 of insulin receptor tyrosine kinase, which is equiva-lent in position to Tyr900 of RET, was previously reported to bean autophosphorylation site on the basis of results of analysisby Edman sequencing (55) in addition to Tyr1162 and Tyr1163

(Tyr1163 is equivalent to Tyr905 in RET) as other autophospho-rylation sites in the activation loop (55–57). Similarly, the TRKnerve growth factor receptor was suggested by results of anal-ysis using the Edman method to be autophosphorylated atTyr670 (corresponding to Tyr900 of RET), Tyr674 and Tyr675

(Tyr905 of RET) (58). Hubbard et al. (54, 59) demonstrated thatautophosphorylation of tyrosine residues in the activation loopof insulin receptor tyrosine kinase is the initial event leading totyrosine kinase activation and induces a drastic conformationalchange in the activation loop. After autophosphorylation oftyrosine in the activation loop, the DFG sequence, which isconserved in the beginning of the activation loop of the kinasefor occupying the ATP binding site, shifts away from the ATPbinding site, and the loop is stabilized by the interaction ofphospho-Tyr1163 with a side chain of arginine 1155 and glycine1166 via hydrogen bonds. The phosphate group of phospho-Tyr1158, which corresponds to RET phospho-Tyr900, however,

FIG. 7. Levels of kinase activity of RET-MEN2A and its mu-tants measured by in vitro kinase assays. RET-MEN2A or RET-MEN2A mutant cDNA was cloned into a mammalian expression vector,Rc/CMV vector, and RET protein was transiently expressed in NIH3T3cells. Cell lysates from non-RET-transfected NIH3T3 cells or mutantRET-transfected NIH3T3 cells were subjected to immunoprecipitationwith an anti-RET antibody, and the samples were analyzed by both invitro kinase assays (A) and immunoblotting with the anti-RET antibody(B). 175- and 155-kDa isoforms of RET were detected, and measure-ments of these bands were integrated for the determination of overallphosphorylation level. Quantitation of phospho-RET (pRET) levels andphospho-MBP (pMBP) levels was performed by using Scion Image soft-ware. The calculated values for pRET and pMBP of an NIH3T3 samplewere used as background, and the levels of experimental samples overthe background were measured and then divided by those for pRET (C)and pMBP (D) of parental RET-MEN2A, respectively. The bars repre-sent means � S.D. of three independent experiments. The value foreach of the tyrosine mutants (*) is significantly (p � 0.05) different fromthat for RET-MEN2A. The value for the Y900F/Y905F mutant (**) issignificantly (p � 0.05) different from that for the Y905F mutant.

FIG. 8. Transforming activities of the RET-MEN2A and RET-MEN2A mutant genes with mutations of tyrosine residues. RET-MEN2A and its mutant genes were transfected to NIH3T3 cells, andtransforming assays were performed as described under “ExperimentalProcedures.” The bars represent means � S.D. of four independentexperiments. The values for each of the tyrosine mutants (**) aresignificantly (p � 0.001) different from that for RET-MEN2A.

Identification of Autophosphorylation Sites of RET14222

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

makes no protein contacts and is completely solvent-exposed. Itshould be noted that the mutant Y1158F in insulin receptortyrosine kinase in one case was reported to have no effect on invitro kinase activity (60), whereas the same mutation wasfound in several other studies to reduce in vitro activity bymore than 80% (61, 62). Cunningham et al. (63) demonstratedthat mutations of the activation loop of TRK yielded a hierar-chical loss of nerve growth factor-induced neurite outgrowth:wild type � Y670F �� Y675F � Y670F/Y675F �� Y674F/Y675F. The function we demonstrated in this study with Tyr900

of RET, supplementing the role of Tyr905 as the major auto-phosphorylation site, thus differs from the reported functions ofTyr900 of RET-equivalent Tyr1158 of insulin receptor tyrosinekinase and Tyr670 of TRK and is novel. Collectively, thesefindings and our results confirm a general role of phosphoryl-ated Tyr905 of RET and its equivalents in other PTKs and somediverse additional roles of phosphorylated Tyr900 of RET andits equivalents in kinase activation and downstream signaling.

We also demonstrated that Tyr806 and Tyr809, both of whichexist in the juxtamembrane of RET, were autophosphorylated.There have been very few studies in which the function ofTyr806 or Tyr809 of RET was examined. According to results ofamino acid alignment analysis of PTKs, Tyr806 and Tyr809 ofRET corresponded to tyrosines of 41 and 15 PTKs registered inthe SWISS-PROT data base, respectively. In vitro kinase as-says revealed that the kinase activity was impaired by substi-tution of phenylalanine for tyrosine 806 or 809. The transform-ing activities in both the Y806F and Y809F mutants were alsoimpaired. This is also the first report to present evidence of theexistence of tyrosines as autophosphorylation sites in the ki-nase domain outside the activation loop that are involved inboth kinase activation and downstream signaling. There hasbeen no earlier report on functions of tyrosines equivalent tothose of Tyr806 or Tyr809 in other PTKs. According to thetertiary structure model, both Tyr806 and Tyr809 exist in thecontralateral side of the activation loop and in a very shortregion between the end of the C-terminal �-sheet of the N-lobeand the beginning of the N-terminal �-helix of the C-lobe.These tyrosines lie adjacent to the putative ATP-bindingpocket, indicating that they can affect the efficiency of ATPbinding (see Fig. 5). These facts suggest that Tyr806 and Tyr809,located in this unique position, play a novel supplemental rolefor the activation loop upon phosphorylation. Interestingly, aV804M/Y806C double mutation has been identified in a Japa-nese female MEN2B patient (64). This finding suggests thatthe Tyr806-dependent regulation of kinase activity is physiolog-ically significant.

Phosphorylation of Tyr981 of RET was another novel findingin this study. This tyrosine is also well conserved among PTKs(in 58 (72.5%) of the 80 PTKs). Mohammadi et al. (65) reportedthat Tyr730, which is equivalent to RET Tyr981, is one of theautophosphorylation sites of the fibroblast growth factor recep-tor, although the biological function of Tyr730 was not deter-mined. Our previous study showed that mutation of MEN2A-Y981F abolished the autophosphorylation and kinase activitiesof RET-MEN2A measured in vitro (28). Interestingly, the de-fect in intrinsic kinase activity of the Y981F mutant was able tobe repaired by Src kinase, which promoted Tyr905 phosphoryl-ation in trans. Taken together, the results suggest that phos-phorylation of Tyr981 is not obligatorily required for the cata-lytic activity but plays a supplementary role in initiatingautophosphorylation of Tyr905, which brings about the overallkinase activity.

In this study, we obtained direct evidence of not only tyrosineautophosphorylation sites in the kinase domain of RET but alsoof intriguing functions of Tyr806, Tyr809, and Tyr900 by mass

spectrometry and mutagenesis analysis. It is likely that tyro-sine residues of other PTKs that correspond to the RET auto-phosphorylation sites revealed in this study are also autophos-phorylation sites because these tyrosine residues are wellconserved among both receptor-type and nonreceptor-typePTKs. It was, however, difficult to determine all of the RETautophosphorylation sites in our system, because there aresome methodological limitations in the use of mass spectrom-etry for analysis of phosphopeptides; e.g. ionization efficienciesof some phosphopeptides under a MALDI or ion trap conditionare sometimes too low to detect signals. Therefore, the charac-terization of all autophosphorylation sites requires the devel-opment of improved methods. Autophosphorylation and thedownstream signaling effects of these phosphorylated tyrosinesin cells under physiological conditions remain to be analyzedbecause we used an originally prepared recombinant GST-�RET protein without any ligand stimulation, and we did notuse cells that normally express RET. However, mass spectrom-etry proved to be a useful technique for directly identifyingsites of post-translational phosphomodification of RET, and thedata we obtained by using this technique have contributed toan understanding of the mechanisms of activation of not onlyRET but also any other PTK proteins and to an understandingof downstream signaling under the condition of activatedPTKs.

Acknowledgments—We thank Y. Kojima, H. Saeki, and N. Taniguchifor technical assistance.

REFERENCES

1. Takahashi, M., Buma, Y., Iwamoto, T., Inaguma, Y., Ikeda, H., and Hiai, H.(1988) Oncogene 5, 571–578

2. Takahashi, M., Buma, Y., and Hiai, H. (1989) Oncogene 3, 805–8063. Iwamoto, T., Taniguchi, M., Asai, N., Ohkusu, K., Nakashima, I., and Taka-

hashi, M. (1993) Oncogene 8, 1087–10914. Grieco, M., Santoro, M., Berlingieri, M. T., Melillo, R. M., Donghi, R., Bongar-

zone, I, Pierotti, M. A., Della Porta, G., Fusco, A., and Vecchio, G. (1990)Cell 60, 557–563

5. Mulligan, L. M., Kwok, J. B., Healey, C. S., Elsdon, M. J., Eng, C., Gardner, E.,Love, D. R., Mole, S. E., Moore, J. K., Papi, L., Ponder, M. A., Telenius, H.,Tunnacliffe, A., and Ponder, B. A. (1993) Nature 363, 458–460

6. Donis-Keller, H., Dou, S., Chi, D., Carlson, K. M., Toshima, K., Lairmore, T. C.,Howe, J. R., Moley, J. F., Goodfellow, P., and Wells, S. A., Jr. (1993) Hum.Mol. Genet. 2, 851–886

7. Hofstra, R. M., Landsvater, R. M., Ceccherini, I., Stulp, R. P., Stelwagen, T.,Luo, Y., Pasini, B., Hoppener, J. W., van Amstel, H. K., Romeo, G., Lips,C. J., and Buys, C. H. (1994) Nature 367, 375–376

8. Carlson, K. M., Dou, S., Chi, D., Scavarda, N., Toshima, K., Jackson, C. E.,Wells, S. A., Jr., Goodfellow, P. J., and Donis-Keller, H. (1994) Proc. Natl.Acad. Sci. U. S. A. 91, 1579–1583

9. Romeo, G., Ronchetto, P., Luo, Y., Barone, V., Seri, M., Ceccherini, I., Pasini,B., Bocciardi, R., Lerone, M., Kaariainen, H., and Martucciello, G. (1994)Nature 367, 377–378

10. Edery, P., Lyonnet, S., Mulligan, L. M., Pelet, A., Dow, E., Abel, L., Holder, S.,Nihoul-Fekete, C., Ponder, B. A., and Munnich, A. (1994) Nature 367,378–380

11. Lin, L. F., Doherty, D. H., Lile, J. D., Bektesh, S., and Collins, F. (1993) Science260, 1130–1132

12. Kotzbauer, P. T., Lampe, P. A., Heuckeroth, R. O., Golden, J. P., Creedon,D. J., Johnson, E. M., Jr., and Milbrandt, J. (1996) Nature 384, 467–470

13. Baloh, R. H., Tansey, M. G., Lampe, P. A., Fahrner, T. J., Enomoto, H.,Simburger, K. S., Leitner, M. L., Araki, T., Johnson, E. M., Jr., and Mil-brandt, J. (1998) Neuron 21, 1291–1302

14. Milbrandt, J., de Sauvage, F. J., Fahrner, T. J., Baloh, R. H., Leitner, M. L.,Tansey, M. G., Lampe, P. A., Heuckeroth, R. O., Kotzbauer, P. T., Sim-burger, K. S., Golden, J. P., Davies, J. A., Vejsada, R., Kato, A. C., Hynes,M., Sherman, D., Nishimura, M., Wang, L. C., Vandlen, R., Moffat, B.,Klein, R. D., Poulsen, K., Gray, C., Garces, A., Henderson, C. E., Phillips,H. S., and Johnson, E. M., Jr. (1998) Neuron 20, 245–253

15. Besset, V., Scott, R. P., and Ibanez, C. F. (2000) J. Biol. Chem. 275,39159–39166

16. Hayashi, H., Ichihara, M., Iwashita, T., Murakami, H., Shimono, Y., Kawai,K., Kurokawa, K., Murakumo, Y., Imai, T., Funahashi, H., Nakao, A., andTakahashi, M. (2000) Oncogene 19, 4469–4475

17. Hayashi, Y., Iwashita, T., Murakamai, H., Kato, Y., Kawai, K., Kurokawa, K.,Tohnai, I., Ueda, M., and Takahashi, M. (2001) Biochem. Biophys. Res.Commun. 281, 682–689

18. Segouffin-Cariou, C., and Billaud, M. (2000) J. Biol. Chem. 275, 3568–357619. Tahira, T., Ishizaka, Y., Itoh, F., Sugimura, T., and Nagao, M. (1990) Oncogene

5, 97–10220. Myers, S. M., Eng, C., Ponder, B. A., and Mulligan, L. M. (1995) Oncogene 11,

2039–204521. Pandey, A., Liu, X., Dixon, J. E., Di Fiore, P. P., and Dixit, V. M. (1996) J. Biol.

Identification of Autophosphorylation Sites of RET 14223

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Chem. 271, 10607–1061022. Borrello, M. G., Alberti, L., Arighi, E., Bongarzone, I., Battistini, C., Bardelli,

A., Pasini, B., Piutti, C., Rizzetti, M. G., Mondellini, P., Radice, M. T., andPierotti, M. A. (1996) Mol. Cell. Biol. 16, 2151–2163

23. Arighi, E., Alberti, L., Torriti, F., Ghizzoni, S., Rizzetti, M. G., Pelicci, G.,Pasini, B., Bongarzone, I., Piutti, C., Pierotti, M. A., and Borrello, M. G.(1997) Oncogene 14, 773–782

24. Kurokawa, K., Iwashita, T., Murakami, H., Hayashi, H., Kawai, K., andTakahashi, M. (2001) Oncogene 20, 1929–1938

25. Durick, K., Wu, R. Y., Gill, G. N., and Taylor, S. S. (1996) J. Biol. Chem. 271,12691–12694

26. Alberti, L., Borrello, M. G., Ghizzoni, S., Torriti, F., Rizzetti, M. G., andPierotti, M. A. (1998) Oncogene 17, 1079–1087

27. Tansey, M. G., Baloh, R. H., Milbrandt, J., and Johnson, E. M., Jr. (2000)Neuron 25, 611–623

28. Kato, M., Takeda, K., Kawamoto, Y., Iwashita, T., Akhand, A. A., Senga, T.,Yamamoto, M., Sobue, G., Hamaguchi, M., Takahashi, M., and Nakashima,I. (2002) Cancer Res. 62, 2414–2422

29. Akhand, A. A., Pu, M., Senga, T., Kato, M., Suzuki, H., Miyata, T., Hamaguchi,M., and Nakashima, I. (1999) J. Biol. Chem. 274, 25821–25826

30. Devary, Y., Gottlieb, R. A., Smeal, T., and Karin, M. (1992) Cell 71, 1081–109131. Pu, M., Akhand, A. A., Kato, M., Hamaguchi, M., Koike, T., Iwata, H., Sabe, H.,

Suzuki, H., and Nakashima, I. (1996) Oncogene 13, 2615–262232. Pu, M., Akhand, A. A., Kato, M., Koike, T., Hamaguchi, M., Suzuki, H., and

Nakashima, I. (1996) J. Cell. Biochem. 63, 104–11433. Hardwick, J. S., and Sefton, B. M. (1995) Proc. Natl. Acad. Sci. U. S. A. 92,

4527–453134. Nakamura, K., Hori, T., Sato, N., Sugie, K., Kawakami, T., and Yodoi, J. (1993)

Oncogene 8, 3133–313935. Nakashima, I., Pu, M. Y., Nishizaki, A., Rosila, I., Ma, L., Katano, Y., Ohkusu,

K., Rahman, S. M., Isobe, K., Hamaguchi, M., and Saga, K. (1994) J. Im-munol. 152, 1064–1071

36. Kato, M., Iwashita, T., Akhand, A. A., Liu, W., Takeda, K., Takeuchi, K.,Yoshihara, M., Hossain, K., Wu, J., Du, J., Oh, C., Kawamoto, Y., Suzuki,H., Takahashi, M., and Nakashima, I. (2000) Antioxid. Redox Signal 2,841–849

37. Kato, M., Iwashita, T., Takeda, K., Akhand, A. A., Liu, W., Yoshihara, M.,Asai, N., Suzuki, H., Takahashi, M., and Nakashima, I. (2000) Mol. Biol.Cell 11, 93–101

38. Takeda, K., Kato, M., Wu, J., Iwashita, T., Suzuki, H., Takahashi, M., andNakashima, I. (2001) Antioxid. Redox Signal 3, 473–482

39. Veillette, A., Dumont, S., and Fournel, M. (1993) J. Biol. Chem. 268,17547–17553

40. Nakashima, I., Takeda, K., Kawamoto, Y., Okuno, Y., Kato, M., Akhand, A.,and Suzuki, H. (2002) The Proceedings of XI Biennial Meeting of the Societyfor Free Radical Research International, pp. 229–234, Monduzzi EditoreS.p.A.-Medimond Inc., Bologna, Italy

41. Liu, X., Vega, Q. C., Decker, R. A., Pandey, A., Worby, C. A., and Dixon, J. E.

(1996) J. Biol. Chem. 271, 5309–531242. Coulpier, M., Anders, J., and Ibanez, C. F. (2002) J. Biol. Chem. 277,

1991–199943. Salvatore, D., Barone, M. V., Salvatore, G., Melillo, R. M., Chiappetta, G.,

Mineo, A., Fenzi, G., Vecchio, G., Fusco, A., and Santoro, M. (2000) J. Clin.Endocrinol. Metab. 85, 3898–3907

44. Tsui-Pierchala, B. A., Ahrens, R. C., Crowder, R. J., Milbrandt, J., and John-son, E. M., Jr. (2002) J. Biol. Chem. 277, 34618–34625

45. Kato, M., Takeda, K., Kawamoto, Y., Tsuzuki, T., Dai, Y., Nakayama, S.,Toriyama, K., Tamada, Y., Takahashi, M., and Nakashima, I. (2001) Onco-gene 20, 7536–7541

46. Takahashi, M., Asai, N., Iwashita, T., Isomura, T., Miyazaki, K., and Mat-suyama, M. (1993) Oncogene 8, 2925–2929

47. Kato, M., Liu, W., Akhand, A. A., Dai, Y., Ohbayashi, M., Tuzuki, T., Suzuki,H., Isobe, K., Takahashi, M., and Nakashima, I. (1999) Oncogene 18,837–842

48. Xu, W., Harrison, S. C., and Eck, M. J. (1997) Nature 385, 595–60249. Hanks, S. K., Quinn, A. M., and Hunter, T. (1988) Science 241, 42–5250. Hanks, S. K., and Quinn, A. M. (1991) Methods Enzymol. 200, 38–6251. Paratcha, G., Ledda, F., Baars, L., Coulpier, M., Besset, V., Anders, J., Scott,

R., and Ibanez, C. F. (2001) Neuron 29, 171–18452. Encinas, M., Tansey, M. G., Tsui-Pierchala, B. A., Comella, J. X., Milbrandt, J.,

and Johnson, E. M., Jr. (2001) J. Neurosci. 21, 1464–147253. Mann, M., Ong, S. E., Gronborg, M., Steen, H., Jensen, O. N., and Pandey, A.

(2002) Trends Biotechnol. 20, 261–26854. Hubbard, S. R. (1997) EMBO J. 16, 5572–558155. Tornqvist, H. E., Pierce, M. W., Frackelton, A. R., Nemenoff, R. A., and Avruch,

J. (1987) J. Biol. Chem. 262, 10212–1021956. Rosen, O. M., Herrera, R., Olowe, Y., Petruzzelli, L. M., and Cobb, M. H. (1983)

Proc. Natl. Acad. Sci. U. S. A. 80, 3237–324057. Ellis, L., Clauser, E., Morgan, D. O., Edery, M., Roth, R. A., and Rutter, W. J.

(1986) Cell 45, 721–73258. Stephens, R. M., Loeb, D. M., Copeland, T. D., Pawson, T., Greene, L. A., and

Kaplan, D. R. (1994) Neuron 12, 691–70559. Hubbard, S. R., Wei, L., Ellis, L., and Hendrickson, W. A. (1994) Nature 372,

746–75460. Zhang, B., Tavare, J. M., Ellis, L., and Roth, R. A. (1991) J. Biol. Chem. 266,

990–99661. Wilden, P. A., Kahn, C. R., Siddle, K., and White, M. F. (1992) J. Biol. Chem.

267, 16660–1666862. Wilden, P. A., Backer, J. M., Kahn, C. R., Cahill, D. A., Schroeder, G. J., and

White, M. F. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 3358–336263. Cunningham, M. E., Stephens, R. M., Kaplan, D. R., and Greene, L. A. (1997)

J. Biol. Chem. 272, 10957–1096764. Miyauchi, A., Futami, H., Hai, N., Yokozawa, T., Kuma, K., Aoki, N., Kosugi,

S., Sugano, K., and Yamaguchi, K. (1999) Jpn. J. Cancer Res. 90, 1–565. Mohammadi, M., Dikic, I., Sorokin, A., Burgess, W. H., Jaye, M., and Schless-

inger, J. (1996) Mol. Cell. Biol. 16, 977–989

Identification of Autophosphorylation Sites of RET14224

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from

NakashimaIto, Ryo Taguchi, Masashi Kato, Haruhiko Suzuki, Masahide Takahashi and Izumi

Yoshiyuki Kawamoto, Kozue Takeda, Yusuke Okuno, Yoshinori Yamakawa, YasutomoIdentification of RET Autophosphorylation Sites by Mass Spectrometry

doi: 10.1074/jbc.M312600200 originally published online January 6, 20042004, 279:14213-14224.J. Biol. Chem. 

  10.1074/jbc.M312600200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/279/14/14213.full.html#ref-list-1

This article cites 64 references, 26 of which can be accessed free at

by guest on April 18, 2018

http://ww

w.jbc.org/

Dow

nloaded from