MOLECULAR AND CELLULAR BIOLOGY, Mar. 1996, p. 977–989
Copyright ? 1996, American Society for Microbiology
Vol. 16, No. 3
Identification of Six Novel Autophosphorylation Sites on Fibroblast
Growth Factor Receptor 1 and Elucidation of Their Importance
in Receptor Activation and Signal Transduction
M. MOHAMMADI,1I. DIKIC,1A. SOROKIN,1W. H. BURGESS,2M. JAYE,3AND J. SCHLESSINGER1*
Department of Pharmacology, New York University Medical Center, New York, New York 100161;
American Red Cross, Rockville, Maryland 208552; and Rhone-Poulenc Rorer
Central Research, Collegeville, Pennsylvania 194263
Received 28 September 1995/Returned for modification 7 November 1995/Accepted 12 December 1995
Fibroblast growth factor receptor (FGFR) activation leads to receptor autophosphorylation and increased
tyrosine phosphorylation of several intracellular proteins. We have previously shown that autophosphorylated
tyrosine 766 in FGFR1 serves as a binding site for one of the SH2 domains of phospholipase C? and couples
FGFR1 to phosphatidylinositol hydrolysis in several cell types. In this report, we describe the identification of
six additional autophosphorylation sites (Y-463, Y-583, Y-585, Y-653, Y-654, and Y-730) on FGFR1. We
demonstrate that autophosphorylation on tyrosines 653 and 654 is important for activation of tyrosine kinase
activity of FGFR1 and is therefore essential for FGFR1-mediated biological responses. In contrast, autophos-
phorylation of the remaining four tyrosines is dispensable for FGFR1-mediated mitogen-activated protein
kinase activation and mitogenic signaling in L-6 cells as well as neuronal differentiation of PC12 cells.
Interestingly, both the wild-type and a mutant FGFR1 (FGFR1-4F) are able to phosphorylate Shc and an
unidentified Grb2-associated phosphoprotein of 90 kDa (pp90). Binding of the Grb2/Sos complex to phos-
phorylated Shc and pp90 may therefore be the key link between FGFR1 and the Ras signaling pathway, mito-
genesis, and neuronal differentiation.
Fibroblast growth factors (FGFs) constitute a large family of
at least nine distinct polypeptide growth factors (7, 31, 34).
FGFs play an important role in the regulation of cell growth,
differentiation, embryogenesis, and angiogenesis (34). Like
other growth factors, FGFs exert their action by binding to and
activating a distinct family of growth factor receptors that has
been previously classified as subclass IV (20, 74). The FGF
receptor family consists of at least four distinct gene products,
each composed of an extracellular ligand-binding domain that
contains three immunoglobulin-like domains, a single trans-
membrane domain, and a cytoplasmic domain that contains
protein tyrosine kinase activity (interrupted by an insertion of
14 amino acids in the kinase domain). One of the characteristic
features of the FGF receptor family is the occurrence of nu-
merous receptor isoforms that are produced from alternatively
spliced transcripts in both the intracellular and extracellular
domains (9, 14, 27, 32, 33). As with other growth factors,
binding of FGF to FGF receptors leads to receptor dimeriza-
tion (3, 70, 73) and subsequent tyrosine autophosphorylation
and phosphorylation of target substrates (6, 13, 23). Autophos-
phorylation on tyrosine is considered to have at least two
functions. One such function is the stimulation of the intrinsic
protein tyrosine kinase activity by an allosteric mechanism, as
seen with the insulin receptor (22, 63, 81, 84, 86). Also, many
autophosphorylation sites serve as binding sites for signaling
proteins that contain Src homology 2 (SH2) domains or the
recently identified phosphotyrosine interaction (also called
phosphotyrosine-binding [PTB]) domains (4, 5, 8, 35, 36, 39,
48, 56, 57). Binding of SH2 or PTB domain-containing proteins
to activated growth factor receptors has been shown to be
important for the activation of downstream signaling mole-
cules. For example, binding of Shc and phospholipase C?
(PLC?) through PTB and SH2 domains, respectively, to the
activated nerve growth factor receptor (Trk) has been shown to
be required for the nerve growth factor-induced activation of
Ras signaling pathways and neuronal differentiation of PC12
cells (16, 55, 72). Several studies have demonstrated that mu-
tation of autophosphorylation sites in platelet-derived growth
factor (PDGF) receptor and in colony-stimulating factor 1
receptor can impair mitogenic signaling in some cell lines (19,
Very little is known about the cellular substrates and target
proteins involved in signaling processes that lead to FGF-
mediated mitogenesis. So far only PLC? has been shown to
associate with the activated FGF receptor 1 (FGFR1) (flg); the
identities of other targets remain unclear. We have previously
identified tyrosine 766 of FGFR1 as the major autophosphor-
ylation site and have shown that this tyrosine and its flanking
sequences represent a high affinity binding site for one of the
SH2 domains of PLC? (52). Mutation of this tyrosine to phe-
nylalanine results in a receptor that is no longer able to stim-
ulate phosphatidylinositol (PI) hydrolysis but that can still in-
duce mitogenesis in L-6 myoblasts and BaF3 cells and neurite
outgrowth in PC12 cells. Moreover, PI hydrolysis seems to be
dispensable for the induction of chemotaxis by acidic FGF
(aFGF) in certain cell types (11). These observations indicate
that activation of target proteins crucial for either mitogenesis,
differentiation, or chemotaxis is not dependent on autophos-
phorylated tyrosine 766 (29, 51, 58, 71). However, autophos-
phorylation on tyrosine 766 of FGFR1 is required for efficient
endocytosis of FGF receptors (69).
As part of our effort to understand FGFR1 signal transduc-
tion, we have mapped six additional autophosphorylation sites
(Y-463, Y-583, Y-585, Y-653, Y-654, and Y-730) on FGFR1
and have studied their roles in various FGFR1-mediated re-
sponses. Two of the identified tyrosines (Y-653 and Y-654) are
* Corresponding author. Mailing address: Department of Pharma-
cology, New York University Medical Center, 550 First Ave., New
York, NY 10016. Phone: (212) 263-7122. Fax: (212) 263-7133.
located in the kinase domain in the region homologous to
tyrosine 416 of pp60c-src. Mutation of these tyrosines reveals
that their autophosphorylation is essential for activation of
FGFR1. Tyrosines Y-653 and Y-654 are conserved among all
known FGF receptors, and we suggest that homologous re-
gions in other FGF receptors are also important in receptor
activation. We also generated mutated FGF receptors in which
the remaining four autophosphorylation sites (Y-463, Y-583,
Y-585, and Y-730) together (FGFR1-4F) or in combination
with the previously identified tyrosine 766 (FGFR1-5F) were
mutated to phenylalanine. Our results indicate that autophos-
phorylation of these tyrosines is dispensable for FGF-mediated
cell proliferation in L-6 myoblasts as well as neuronal differ-
entiation in PC12 cells. We have found that the FGFR1-5F is
still able to phosphorylate Shc and an as yet unidentified,
Grb2-associated phosphoprotein with a molecular mass of 90
MATERIALS AND METHODS
Site-directed mutagenesis and generation of stable cell lines. Site-directed
mutagenesis was performed according to the protocol of the manufacturer (Am-
ersham). The cDNAs encoding the full-length human wild-type FGFR1 (flg) and
the FGFR1 mutant with a Y-to-F mutation at position 766 (Y766F mutant) were
subcloned into the m13MP19 replicative form with BamHI-HindIII cloning sites
(51). Point mutations were introduced with the oligonucleotides 5?-CACATC
GACTTCTATAAAAAGACA-3? (for the Y653F mutant), 5?-CACATCGACT
ACTTTAAAAAGACA-3? (for the Y654F mutant), and 5?-CACATCGACTTC
TTTAAAAAGACA-3? (for the [Y653/654F mutant), 5?-GTCTCTGAGTTT
GAGCTTCCC-3? (for the Y463F mutant), 5?-CCAGGGCTGGAATTCTGCT
TCAACCCCAGCCAC-3? (for the Y583/585F mutant), and 5?-ACCAAC
GAGCTGTTCATGATGATGCGG-3? (for the Y730F), mutant). The triple-
mutant FGF receptor Y463/583/585F was made on the double mutant Y583/
585F as the template. The Y463/583/585F triple mutant ssm13 was then used to
generate the Y463/583/585/730F (FGFR1-4F) mutant FGF receptor. The Y583/
585/766F triple-mutant receptor was generated on the mutant Y766F as the
template. The triple mutant Y583/585/766F was then used to generate the Y463/
583/585/766F mutant receptor. The Y463/583/585/766F ssm13 was used as the
template to generate the Y463/583/585/730/766F (FGFR1-5F) mutant FGF re-
ceptor. The different mutated m13MP19 constructs were digested with BamHI
and HindIII, blunted with Klenow enzyme, and subcloned into pMJ30 under the
control of adenovirus major late promoter and cytomegalovirus enhancer. L-6
myoblasts, which lack endogenous FGF receptors, were transfected with 0.5 ?g
of pSV2neo and 40 ?g of wild-type or mutant FGFR1 in pMJ30-expressing
vectors by calcium phosphate precipitation (10). Clones were isolated after 2 to
3 weeks of selection in G418 (Gibco) and screened for expression of FGFR1 by
binding assay with125I-labeled aFGF as a specific probe. Cell lines expressing
similar levels of FGF receptors were used for further analysis.
In addition, cDNAs encoding FGFR1-Y653F, FGFR1-Y654F, FGFR1-Y653/
654F, FGFR1-4F (Y463/583/585/730F), and FGFR1-5F (Y463/583/585/730/
766F) were subcloned in pRK5 mammalian expression vector with the BamHI
and HindIII cloning sites. PC12 cells were cotransfected with 0.5 ?g of pSV2neo
and 20 ?g of mutant FGFR1 with a Bio-Rad electroporator set at 250 V and 960
?F. Cells were grown for 4 to 5 weeks in medium containing G418, and stable cell
lines expressing approximately equal numbers of mutant receptors per cell were
isolated by a binding assay with125I-labeled aFGF and/or immunoblotting with
anti-FGFR1 antibodies. The construction of the deletion mutant Flg?COOH-
FGFR1 and generation of L-6 myoblasts overexpressing this mutant receptor
have been described previously (69). Flg?COOH-FGFR1 lacks the 56 carboxy-
terminal amino acids of the receptor; therefore, tyrosine 766 is not present in this
Expression of the FGFR cytoplasmic domain in insect cells. The cDNA con-
struct encoding the cytoplasmic domain of FGFR1 was subcloned into baculo-
virus transfer vector pBlueBac HistagB with NcoI and HindIII cloning sites
(Invitrogen). This construct lacks the 56 carboxy-terminal amino acids of the
cytoplasmic domain, since the codon for tyrosine (TAC) was changed to a stop
codon (TAG) by site-directed mutagenesis. Therefore, tyrosine 766 is not
present in this construct. Transfection of insect cells (Sf9) was performed with
the BaculoGold transfection system according to the protocol of the manufac-
turer (Pharmingen). Following identification of the positive plaques, the recom-
binant virus was amplified to high titer (multiplicity of infection, 108). The fusion
protein was first purified over an Ni2?-chelating column and then analyzed by
anion exchange chromatography (Mono Q column). The fractions containing the
fusion protein were concentrated with Centricon 10 (Amicon), and the histidine
tag was removed by digestion with enterokinase (Biozyme) overnight. The
cleaved kinase domain was then separated from the histidine tag by size exclu-
sion (Superose 12 column) and an additional round of anion exchange (Mono Q
column) chromatography. The purity of the kinase was assessed by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie
blue staining and was greater than 95%.
In vivo phosphorylation of FGFR1. L-6 myoblasts expressing wild-type FGFR1
or the deletion mutant Flg?COOH were starved overnight in Dulbecco’s mod-
ified Eagle medium (DMEM) containing 0.1% fetal bovine serum (FBS) and
then were prelabeled with32Pi(1 mCi/ml) for 2 h in phosphate-free medium
(Gibco). Following stimulation with aFGF (100 ng/ml; 5 min at 25?C), the cells
were washed with phosphate-buffered saline (PBS) on ice and lysed in lysis buffer
containing phosphatase inhibitors. The lysates were spun at 13,000 ? g in a
microcentrifuge for 10 min at 4?C, and the supernatants were immunoprecipi-
tated with anti-FGFR1 antibodies (anti-Flg3B) for 2 h at 4?C. The anti-Flg3B
antibodies were raised against a peptide derived from the kinase insert region of
FGFR1 (residues 580 to 596) and were used for immunoprecipitation of both the
wild type and the deletion mutant. The immunoprecipitates were washed several
times with lysis buffer, 3? sample buffer was added, the solution was boiled for
5 min, and the products were analyzed by SDS-PAGE (7% polyacrylamide). The
gel was dried, and the32P-labeled bands corresponding to the FGFR1 were cut
out and digested with trypsin as described previously (52). The eluted32P-labeled
peptides were filtered, soybean trypsin inhibitor (10 ?g/ml) was added, and the
solution was incubated with monoclonal anti-phosphotyrosine antibodies immo-
bilized on beads (Oncogene Science) for 2 h at 4?C. The beads were washed three
times with washing buffer (20 mM HEPES [N-2-hydroxyethylpiperazine-N?-2-
ethanesulfonic acid] [pH 7.5], 50 mM NaCl). The phosphotyrosine-containing
peptides were then eluted with 1 ml of 50 mM phenylphosphate in washing
buffer, filtered, and directly resolved on an Aquapore C18reverse-phase high-
performance liquid chromatography (HPLC) column (4.6 by 250 mm) at a flow
rate of 1 ml/min with an acetonitrile gradient in 0.1% aqueous trifluoroacetic
acid. The following gradient was used: 0 to 10 min, 0% CH3CN; 10 to 70 min,
linear gradient to 60% CH3CN. Fractions of 0.5 ml were collected, and32P
radioactivity was analyzed with a beta counter.
Identification of tyrosine autophosphorylation sites of FGF receptor. The
purified recombinant kinase domain (200 ?g) was subjected to an in vitro kinase
assay in the presence of 10 mM Mg2?, 5 mM ATP, and a trace amount of
[?-32P]ATP (10 ?Ci) for 5 min at room temperature. We used a trace amount of
[?-32P]ATP in order to monitor the tyrosine-phosphorylated peptides during
tryptic digestion and reverse-phase HPLC analysis. The32P-labeled autophos-
phorylated kinase domain was analyzed by SDS-PAGE (12% polyacrylamide) in
order to remove the unincorporated [?-32P]ATP. The gel was dried, and the
radioactive bands corresponding to the kinase domain were excised from the gel
and subjected to tryptic digestion as described previously (52). The eluted pep-
tides were then filtered, and soybean trypsin inhibitor (10 ?g/ml) was added. The
mixture was then incubated with monoclonal anti-phosphotyrosine antibodies
immobilized on beads (Oncogene Science) for 2 h at 4?C. The beads were washed
three times with washing buffer, and the phosphotyrosine-containing peptides
were eluted with phenylphosphate (see above). The32P-labeled peptides were
resolved on an Aquapore C18reverse-phase HPLC column with the same gra-
dient as described above. Fractions of 0.5 ml were collected, and32P-labeled
fractions were analyzed by a pulsed liquid-phase protein Microsequencer (model
477A with an on-line model 120 phosphothiohydantoin analyzer; Applied Bio-
systems Inc., Foster City, Calif.). About 20 to 50 pmol of the32P-labeled phos-
phopeptides was used for microsequencing. All residues were correctly identified
as phosphothiohydantoin-derived amino acids, except tyrosine residues which
were phosphorylated. The inability to detect tyrosine-phosphorylated residues is
expected, since the charged phosphate group prevents the extraction of the
tyrosyl derivative into the organic phase in the sequencer.
Phosphoamino acid analysis. Following tryptic digestion of the gel pieces
containing the32P-labeled autophosphorylated kinase domain, the eluted phos-
phopeptides were dried and the residual ammonium bicarbonate was removed by
evacuation three times with water and later hydrolyzed in 50 ?l of 6 N HCl for
1 h at 110?C. Samples were dried, and the residual HCl was removed by evac-
uation twice with water. The samples were dissolved in electrophoresis buffer
(pH 1.9) containing phosphoamino acid standards and then spotted onto cellu-
lose thin-layer chromatography plates. Phosphoamino acids were resolved by
two-dimensional electrophoresis (pH 1.9 followed by pH 3.5) as described pre-
Immunoprecipitation and immunoblot analysis. Cells expressing wild-type or
mutant receptors were grown in 10-cm-diameter dishes until 80 to 90% conflu-
ency was reached. The cells were starved overnight in DMEM containing 0.1%
FBS and then treated with aFGF (100 ng/ml) for 5 min at 37?C. The cells were
washed briefly with cold PBS, and lysed in 1 ml of lysis buffer (20 mM HEPES,
150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl2, 1 mM EGTA, 1
?g of leupeptin per ml, 1 ?g of aprotinin per ml, and 1 mM phenylmethylsulfonyl
fluoride) containing phosphatase inhibitors (50 mM sodium fluoride, 10 mM
sodium PPi, and 0.2 mM sodium orthovanadate). Lysates were incubated with
appropriate antibodies overnight. Protein A-Sepharose was added, and after 2 h
the immunocomplexes were washed three times with HNTG (20 mM HEPES,
150 mM NaCl, 0.1% Triton X-100, and 10% glycerol). Sample buffer (3?) was
added, and the samples were boiled for 5 min. After analysis by SDS-PAGE,
proteins were transferred electrophoretically onto nitrocellulose membranes and
immunoblotted with different antibodies. Blots were treated with125I-labeled
protein A and analyzed by autoradiography. The following antibodies were used
978MOHAMMADI ET AL.MOL. CELL. BIOL.
throughout these studies: anti-FGFR1 (29); anti-PLC?1 (45); anti-Sos1 (44), and
anti-Grb2 and anti-Shc (2). In the in vivo labeling experiments we used an
anti-Flg3B antiserum. These antibodies were raised against the kinase insert
region of FGFR1 (residues 580 to 596) (52).
In vitro kinase assay. Cells expressing wild-type and FGFR1-Y653F, FGFR1-
Y654F, and FGFR1-Y653/654F mutant receptors were stimulated with aFGF
(100 ng/ml) for 5 min at 37?C, lysed, and immunoprecipitated with anti-FGFR1
receptor antibodies. The immunoprecipitates were subjected to an in vitro kinase
assay (50 ?l) in the presence of [?-32P]ATP (5 ?M), 10 mM Mg2Cl, and an
exogenous substrate, either 5 mM angiotensin or a fusion protein containing the
carboxy-terminal tail of FGFR1 fused to glutathione S-transferase (GST-
FGFR1-CT) (10 ?g per reaction), for 5 min. Sample buffer was added, and the
reaction products were analyzed by SDS-PAGE and autoradiography. Angioten-
sin was separated from [?-32P]ATP by using a gradient gel of 5 to 15% (15 ml)
that was layered on the top of a previously poured 20% gel (10 ml).
PI hydrolysis. L-6 myoblasts expressing wild-type or mutant receptors were
labeled with [3H]myoinositol (2 ?Ci/ml) in DMEM containing 0.5% FBS for 24
h and incubated in DMEM containing 20 mM LiCl for 20 min before addition of
aFGF (100 ng/ml) for an additional 30 min at 37?C. Cells were extracted with
perchloric acid and inositol phosphate formation was measured according to
published procedures (49).
[3H]thymidine incorporation. L-6 cells were seeded in 96-well plates (2 ? 104
cells per well), and 24 h later the medium was changed to DMEM containing
0.1% FBS. After 48 h of serum starvation, the medium was replaced by medium
containing aFGF or, as control, 10% FBS. After stimulation for 24 h, cells were
incubated with [3H]thymidine (final concentration, 0.5 ?Ci/ml) for 16 h at 37?C.
Cells were washed with PBS, trypsinized, and collected with a PHD cell harvester
(Cambridge Technologies), and the amount of incorporated [3H]thymidine was
quantitated by a liquid scintillation counting device (LKB).
PC12 cell differentiation assay. To examine the effect of FGF on morphology
of PC12 cells, cells were seeded at a density of 1.5 ? 105cells per 6-cm-diameter
tissue culture dish in DMEM supplemented with 3% horse serum, 2.5% fetal calf
serum, penicillin, streptomycin, and L-glutamine. Cells were stimulated with
aFGF (100 ng/ml) or with medium alone as a control. The kinetics and extent of
growth factor-induced neurite outgrowth were analyzed by randomly scoring 500
cells for each plate and calculation of the percentage of cells with neurites longer
than two cell bodies as previously described (17).
Identification of autophosphorylation sites in FGFR1. The
intracellular domain of FGFR1 was expressed as a histidine-
tagged fusion protein in Sf9 cells with a baculovirus-based
expression system. The recombinant protein was purified to
homogeneity by metal chelating, anion exchange, and size ex-
clusion chromatography. The purified kinase domain was
phosphorylated in vitro in the presence of [?-32P]ATP and
MgCl2and was further analyzed by SDS-PAGE (Fig. 1A).
Phosphoamino acid analysis of the in vitro-phosphorylated ki-
nase domain revealed the presence of only phosphotyrosine in
the autophosphorylated FGFR1 kinase domain (Fig. 1B). Fol-
lowing tryptic digestion, the phosphopeptides were purified on
an anti-phosphotyrosine antibody affinity column and were
separated by reverse-phase HPLC. Seven different peaks were
detected by HPLC (Fig. 1C) and were further analyzed by
microsequencing (Fig. 1D). Peak 1 was shown to be a doubly
phosphorylated peptide with phosphate groups on both ty-
rosines 653 and 654. Peak 2 contained the same peptide as
peak 1 except that only tyrosine 653 was found to be phos-
phorylated. Peak 1 was a partially digested peptide, since tryp-
sin did not cleave on the carboxy-terminal site of the lysine 655,
while peak 2 was cleaved at this site. We did not detect any
peptide that was phosphorylated on tyrosine 654 alone. Peak 3
represented a phosphopeptide derived from the kinase insert
region of the FGF receptor, which contained phosphorylated
tyrosines 583 and 585. Peak 4 contained the same peptide as
peak 3, except that only tyrosine 583 was phosphorylated. Peak
5 was derived from the juxtamembrane region, where tyrosine
463 was found to be phosphorylated. Peaks 6 and 7 each
contained a phosphate group on tyrosine 730, a residue located
in the second half of the kinase domain. Since the expressed
kinase domain lacks 56 carboxy-terminal amino acids, we did
not detect peaks (peaks 8 and 9) corresponding to autophos-
phorylated tyrosine 766 (Fig. 1C).
Next we compared the tryptic phosphopeptide map of the
recombinant kinase domain with the tryptic phosphopeptide
maps of wild-type FGFR1 or a deletion mutant (Flg?COOH)
phosphorylated in living cells. L-6 myoblasts expressing the
wild-type or the deletion mutant (Flg?COOH) receptors were
prelabeled with Pi, treated with aFGF, and lysed, and the
lysates were immunoprecipitated with anti-FGFR1 antibodies.
Following SDS-PAGE analysis, the32P-labeled receptor bands
were digested with trypsin and the eluted phosphopeptides
were purified on an anti-phosphotyrosine affinity column. The
mixture of the tyrosine-phosphorylated
phopeptides was then resolved by reverse-phase HPLC (Fig.
1E). Seven different
through 7) with similar retention times to those of the phos-
phopeptides obtained from in vitro-phosphorylated recombi-
nant kinase domain were detected in the HPLC maps of both
the in vivo-labeled wild-type receptor as well as the deletion
mutant Flg?COOH FGF receptor. Moreover, one additional
phosphopeptide (peak 8) was observed in the HPLC map of
the in vivo-labeled wild-type FGF receptor. This phosphopep-
tide (peak 8) represents the carboxy-terminal autophosphory-
lation site (tyrosine 766), and as expected this phosphopeptide
is absent from the HPLC map of the deletion mutant as well as
from the HPLC map of the in vitro-phosphorylated recombi-
nant kinase domain. The phosphorylation of tyrosines 583 and
585, peaks 3 and 4 (derived from the kinase insert region), in
vivo was weaker than the phosphorylation of the same sites
detected in in vitro-phosphorylated recombinant kinase do-
main. This could be due to rapid dephosphorylation of these
sites by intracellular phosphatases or due to the fact that the
antibodies used for immunoprecipitation are directed against
the kinase insert region of FGFR1. Phosphorylation of these
two tyrosines may affect the ability of the anti-Flg3B antibodies
to recognize the kinase insert region of FGFR1. Furthermore,
phosphopeptide 1, which contains phosphate groups on both
tyrosines 653 and 654, is less abundant in vivo than the phos-
phopeptide derived from the in vitro-phosphorylated kinase
domain. In vitro after 1 min of incubation peak 2 (Tyr-653) is
the major phosphorylation site (Fig. 1F). Peak 2 is also a major
autophosphorylation site in the HPLC maps of the in vivo-
labeled wild type as well as the deletion mutant. We suggest
that peak 2 is a doublet because of the presence of two adja-
cent arginine residues (R-655 and R-656). Trypsin will cleave
either after R-656 or between the two arginines, causing a
slight change in the hydrophobicity of phosphopeptide 2. We
suggest that peak 1 is not a doublet, because it contains a
negatively charged phosphate group on tyrosine 654. The pres-
ence of a negatively charged phosphate group may affect the
hydrolysis of the peptide bond connecting R-655 and R-656.
We have shown that tyrosines 766, 653, and 654 are the major
tyrosine autophosphorylation sites of FGFR1 in vivo. Deter-
mination of the actual stoichiometry of autophosphorylation in
vivo is probably an underestimate, since it is impossible to
completely prevent the action of tyrosine phosphatases in liv-
ing cells. Moreover, following in vivo labeling the FGF recep-
tor was immunoprecipitated with anti-FGFR1 antibodies,
which may preferentially recognize a certain population of
phosphorylated receptors. Furthermore, anti-phosphotyrosine
antibodies that were used for isolation of tyrosine-phosphory-
lated peptides may also select for a certain population of phos-
phopeptides. Finally, hydrophobic phosphopeptides are usu-
ally not well represented in HPLC maps because of their lower
levels of recoveries during purification (because of stickiness).
For these reasons it is difficult to determine with certainty the
32P-labeled phosphopeptides (peaks 1
VOL. 16, 1996 SIGNAL TRANSDUCTION BY FGFR1 979
exact stoichiometry of phosphorylation of tyrosine phosphor-
ylation sites in living cells.
We have analyzed the tyrosine phosphorylation of the cata-
lytic kinase domain as a function of time and found that ty-
rosine autophosphorylation is an ordered event. Tyrosine 653
is autophosphorylated prior to tyrosine 654, and autophos-
phorylation on tyrosine 583 occurs prior to phosphorylation on
tyrosine 585 (Fig. 1F). It is possible that efficient autophos-
phorylation on tyrosines 654 and 585 requires an already-phos-
phorylated tyrosine residue amino terminal to tyrosines 654
and 585. This notion is consistent with the observation that
monophosphorylated peptides with a phosphate group on ty-
rosine 654 or 585 alone have not been found (Fig. 1F).
Autophosphorylation of tyrosines 653 and 654 is essential
for kinase activity of FGFR1 in vitro and in vivo. Two of the
newly identified tyrosine autophosphorylation sites are located
in a region homologous to tyrosine 416 of pp60c-src. It has been
demonstrated in the case of the insulin receptor, the scatter
FIG. 1. Identification of novel autophosphorylation sites on FGFR1. (A) In vitro kinase assay of the purified intracellular domain of the FGFR1 expressed in Sf9
cells. The FGFR1 kinase domain was subjected to in vitro autophosphorylation in the presence of [?-32P]ATP and Mg2?. The reaction product was analyzed by
SDS-PAGE and autoradiography. (B) Phosphoamino acid analysis of the in vitro-phosphorylated kinase domain. The autophosphorylated kinase domain band was
excised from the gel and subjected to tryptic digestion. The tryptic phosphopeptides were then hydrolyzed and analyzed by two dimensional chromatography. (C)
Phosphopeptide map of the FGFR1 cytoplasmic domain analyzed by reverse-phase HPLC. The autophosphorylated FGFR1 kinase domain was digested with trypsin,
and phosphotyrosine-containing phosphopeptides were first purified on an anti-phosphotyrosine affinity column and later analyzed by reverse-phase HPLC as described
in Materials and Methods. Radioactive peaks numbered from 1 to 7 are shown by solid arrows. The expected positions of phosphopeptides corresponding to tyrosine
766 are indicated by open arrows (peaks 8 and 9). (D) Amino acid sequences of the phosphopeptides analyzed by microsequencing. The purified phosphopeptides
isolated from HPLC were sequenced by a pulsed liquid-phase protein Microsequencer as described in Materials and Methods. Note that in peak 6 trypsin cleaves after
arginine 721, even though this arginine is followed by a proline. It was shown that trypsin can catalyze the hydrolysis of lysyl and arginyl peptide bonds followed by
proline residues, although the rate of hydrolysis may be slower (1). (E) Comparison of the tryptic phosphopeptide map of recombinant kinase domain with the tryptic
phosphopeptide maps of wild-type (WT) FGFR1 and the deletion mutant Flg?COOH. L-6 myoblasts expressing the wild-type FGFR1 or the deletion mutant
Flg?COOH were prelabeled with32Pi(1 mCi/ml) for 2 h and stimulated with aFGF (100 ng/ml; 5 min at 25?C). Following cell lysis, the lysates were immunoprecipitated
with anti-FGFR1 antiserum and analyzed by SDS-PAGE. The32P-labeled bands corresponding to FGFR1 were subjected to tryptic digestion, and the phosphotyrosine-
containing peptides were purified on an anti-phosphotyrosine affinity column and further resolved by reverse-phase HPLC. Radioactive peaks are numbered 1 through
8. Peak 8 corresponds to the phosphopeptide containing the autophosphorylated tyrosine 766. This peak is absent both from the HPLC maps of the in vitro-
phosphorylated kinase domain and from the in vivo-labeled deletion mutant Flg?COOH. Peak 2 in the HPLC map of wild-type FGFR1 (in vivo) is split because of
partial digestion at the carboxy terminus of arginine 656. (F) Time course of in vitro autophosphorylation of the recombinant kinase domain. The purified recombinant
kinase domain was subjected to in vitro autophosphorylation as a function of time (1, 2, and 5 min). Following SDS-PAGE, the32P-labeled bands corresponding to
the autophosphorylated kinase domain were excised from the gel and subjected to tryptic digestion. The eluted32P-labeled phosphopeptides were resolved by
reverse-phase HPLC as described in Materials and Methods. Peaks are numbered 1 through 7. The minor peak observed between peak 2 and peak 3 is due to partial
cleavage at the carboxy-terminal site of arginine 656.
980MOHAMMADI ET AL.MOL. CELL. BIOL.
factor/hepatocyte growth factor receptor (Met), and the nerve
growth factor receptor (Trk) that autophosphorylation of the
corresponding tyrosines is required for activation of the cata-
lytic domain (18, 21, 22, 24, 40, 41, 46, 50, 54, 62, 72, 81, 86).
We therefore investigated the roles of these two tyrosines in
activation of FGFR1 kinase activity in vivo. The singly and
doubly mutated receptors FGFR1-Y653F, FGFR1-Y654F,
and FGFR1-Y653/654F were constructed. Wild-type FGFR1
and the mutated receptors FGFR1-Y653F, FGFR1-Y654F,
and FGFR1-Y653/654F were expressed in L-6 myoblasts,
which lack endogenous FGF receptors. Several cell lines were
generated and characterized. These cell lines were treated with
aFGF, lysed, and immunoprecipitated with anti-FGF receptor
antibodies, and this was followed by immunoblotting with ei-
ther anti-FGF receptor (Fig. 2A) or anti-phosphotyrosine an-
tibodies (Fig. 2A). In parallel, lysates from aFGF-stimulated or
unstimulated cells were analyzed by immunoblotting with anti-
phosphotyrosine antibodies (Fig. 2B). All receptors except the
lation in vivo. The slight increase in the migration rate of the
FGFR1-Y654F (Fig. 2A; upper part) is probably due to gel
handling for immunoblotting and is not a reproducible finding.
While tyrosine phosphorylation of intracellular substrates was
comparable with that of wild-type receptor in aFGF-treated
cells expressing FGFR1-Y653F, it was diminished in cells ex-
pressing FGFR1-Y654F and was completely abolished in cells
expressing FGFR1-Y653/654F (Fig. 2B). These data strongly
suggest that autophosphorylation of both tyrosines 653 and 654
is required for activation of the kinase domain in vivo. To
confirm this point, we compared the abilities of the wild-type
VOL. 16, 1996SIGNAL TRANSDUCTION BY FGFR1 981
and mutant FGFR1s to phosphorylate PLC?, the best-charac-
terized substrate of FGFR1. Upon treatment of the cells with
aFGF, all of the receptors except the Y653/654F mutant were
able to phosphorylate PLC? (Fig. 2C); however, the extent of
tyrosine phosphorylation of PLC? was weak in cells expressing
FGFR1-Y654F mutant receptors. As anticipated, we detected
no increase in PI hydrolysis in response to aFGF in cells ex-
pressing the Y653/654F double-mutant FGF receptors and
only a moderate increase in PI hydrolysis in cells expressing the
FGFR1-Y654F receptors (Fig. 2D). These experiments dem-
onstrate that mutation of both tyrosines abolishes the ability of
the receptor to undergo autophosphorylation and to phosphor-
ylate intracellular substrates in the context of living cells.
We next compared the in vitro kinase activities of the wild-
type and mutant FGFR1. Immunoprecipitated receptors were
subjected to an in vitro kinase assay in the absence (Fig. 3A) or
presence (Fig. 3B and C) of exogenous substrates. As exoge-
nous substrates we used either a fusion protein containing the
carboxy-terminal tail of FGFR1 (GST-FGFR1-CT) or angio-
tensin. Immunoprecipitates from aFGF-treated cells have
three- to fivefold higher levels of kinase activity than do im-
munoprecipitates from untreated control cells (Fig. 3). While
the FGFR1-Y653F mutant had in vitro kinase activity compa-
rable to that of the wild-type receptor, the FGFR1-Y654F
mutant exhibited a diminished ability to phosphorylate exoge-
nous substrates. Mutation of both tyrosine residues completely
prevented the activation of the kinase domain in response to
aFGF stimulation. Thus, it appears that autophosphorylation
on tyrosines 653 and 654 is required for aFGF-dependent
stimulation of kinase activity in both in vitro and in living cells.
We next measured DNA synthesis in cells expressing wild-
type and mutated forms of FGFR1 (Fig. 4A and B). Mutation
of tyrosine 653 alone did not affect the mitogenic response to
aFGF, while the Y654F mutant receptor had a severely re-
duced ability to stimulate thymidine incorporation. Mutation
of both tyrosines abolished the mitogenic response to aFGF
completely. The small decrease in thymidine incorporation
(less than 50%) in cells expressing FGFR1-Y653F was not a
consistent finding, while DNA synthesis in cell lines expressing
FGFR1-Y654F and FGFR1-Y653/654F was severely dimin-
FIG. 2. Mutation of tyrosines 653 and 654 abolishes kinase activity of FGF receptor in vivo. (A) Anti-FGFR1 immunoprecipitates from aFGF-stimulated (?) or
unstimulated (?) cells expressing wild-type FGFR1 and mutant receptors FGFR1-Y653F, FGFR1-Y654F, and FGFR1-Y653/654F were analyzed by immunoblotting
with either anti-FGFR1 or anti-phosphotyrosine (anti-PTyr) antibodies. (B) Comparison of tyrosine phosphorylation of intracellular proteins induced by aFGF in cell
lines expressing the mutant FGFR1. Lysates from aFGF-treated or untreated cells expressing wild-type FGFR1 or FGFR1-Y653F, FGFR1-Y654F, and FGFR1-
Y653/654F mutant receptors were analyzed by immunoblotting with anti-PTyr antibodies. MAPK1 and MAPK2, MAP kinase 1 and 2, respectively. (C) Tyrosine
phosphorylation of PLC? by wild-type and mutant FGF receptors. Lysates from aFGF-treated or untreated cells expressing wild-type FGFR1 or FGFR1-Y653F,
FGFR1-Y654F, and FGFR1-Y653/654F mutants were immunoprecipitated with anti-PLC? antibodies and analyzed by immunoblotting with either anti-PLC? or
anti-PTyr antibodies. (D) Analysis of PI hydrolysis induced by wild-type and mutant FGFR1. Cells expressing wild-type FGFR1 or FGFR1-Y653F, FGFR1-Y654F, and
FGFR1-Y653/654F mutant receptors were labeled with [3H]myoinositol for 24 h and then stimulated with aFGF for 30 min. As a control, parental cells were stimulated
with PDGF (50 ng/ml) for 30 min. The amount of inositol phosphates was measured as described in Materials and Methods.
982MOHAMMADI ET AL.MOL. CELL. BIOL.
ished in all the cell lines tested. In addition, we have estab-
lished stable PC12 cell lines expressing similar levels of wild-
type FGFR1 or the three mutant receptors FGFR1-Y653F,
FGFR1-Y654F, and FGFR1-Y653/654F. PC12 cells expressing
FGFR1-Y654F had a delayed and less prominent aFGF-in-
duced neurite outgrowth than did wild-type FGFR1 or mutant
FGFR1-Y653F (Fig. 4C). FGF stimulation of PC12 cells over-
expressing FGFR1-Y653/654F did not induce neuronal differ-
entiation (Fig. 4C). These experiments demonstrated that au-
tophosphorylation of Y-653 and Y-654 is essential for the
kinase activity of FGFR1 and hence for mitogenesis in L-6
myoblasts as well as neuronal differentiation in PC12 cells.
Autophosphorylation of four tyrosines in the FGFR1 is not
necessary for the activation of mitogen-activated protein (MAP)
kinase, mitogenesis, and neuronal differentiation. Autophos-
phorylation of growth factor receptors on multiple sites gen-
erates specific binding sites for a variety of intracellular pro-
teins containing SH2 or PTB domains. It is believed that
autophosphorylation sites have a key role in the interaction
between activated tyrosine kinase receptors and downstream
components in the signal transduction pathways. It has been
demonstrated that specificity exists in the interaction between
FIG. 3. Autophosphorylation on tyrosines 653 and 654 is required for stim-
ulation of kinase activity of the FGF receptor in vitro. Cells expressing wild-type
FGFR1 and the FGFR1-Y653F, FGFR1-Y654F, and FGFR1-Y653/654F mu-
tant receptors were stimulated with aFGF (?) or left unstimulated (?). Follow-
ing cell lysis, lysates were immunoprecipitated with anti-FGFR1 antibodies and
subjected to in vitro kinase assay in presence of [?-32P]ATP and Mg2?without
(A) or with an exogenous substrate (GST-FGFR1-CT [B] or angiotensin [C]).
The reaction products were then analyzed by SDS-PAGE and autoradiography.
MW, molecular weight (in thousands).
FIG. 4. Autophosphorylation on tyrosines 653 and 654 is necessary for cell
proliferation in L-6 myoblasts and neuronal differentiation in PC12 cells. (A)
DNA synthesis after aFGF stimulation in cells expressing wild-type (WT) and
mutant FGFR1. Stable cell lines expressing wild-type FGFR1 or FGFR1-Y653F,
FGFR1-Y654F, and FGFR1-Y653/654F mutant receptors were starved for 48 h
in the presence of 0.1% FBS and then treated with different concentrations of
aFGF or, as control, with 10% FBS for 24 h. [3H]thymidine was added, and after
an additional 16 h the amount of incorporated [3H]thymidine was quantitated.
PC, parental PC12. (B) The results obtained in panel A are expressed as per-
centage responses from the treatment with 10% FBS. (C) Quantification of
aFGF-induced neuronal differentiation of PC12 cells expressing wild-type
FGFR1 or FGFR1-Y653F, FGFR1-Y654F, and FGFR1-Y653/654F mutant re-
ceptors. Cells were stimulated with aFGF (100 ng/ml) for 5 days, and neurite
appearance was quantified by scoring 500 cells from randomly selected light
microscope fields of tissue culture dishes. The datum points represent means of
three independent experiments (bars, standard errors).
VOL. 16, 1996SIGNAL TRANSDUCTION BY FGFR1983
these components and the different autophosphorylation sites
in cytoplasmic domains of activated receptor tyrosine kinases
(57). Elimination of certain tyrosine phosphorylation sites
therefore prevents activation of particular signaling pathways.
For example, FGF-induced PI hydrolysis is totally eliminated
by a point mutation of FGF receptor at Y-766, which prevents
the association with and tyrosine phosphorylation of PLC?.
It is thought that the activity of a given receptor tyrosine
kinase equals the sum of the activities of the signaling mole-
cules that are activated by a given receptor tyrosine kinase
(64). Furthermore, it seems that receptor tyrosine kinases uti-
lize several parallel pathways in order to achieve their biolog-
ical effects. Since the single mutant FGFR-Y766F is still able
to induce mitogenesis (51, 58), we examined the possibility that
the four autophosphorylated tyrosines (463, 583, 585, and 730)
may couple FGFR1 to downstream signaling pathways that
lead to mitogenesis. Therefore, we generated mutated forms of
FGFR1 in which either four or five of the known autophos-
phorylation sites were replaced by phenylalanine (FGFR1-4F
and -5F, respectively). These mutated receptors were trans-
fected into L-6 myoblasts as well as into PC12 cells, and several
stable cell lines were established. Cell lines expressing wild-
type or mutant (FGFR1-4F and FGFR1-5F) receptors were
treated with aFGF, lysed, and immunoprecipitated with anti-
FGF receptor antibodies. The immunoprecipitates were then
analyzed by immunoblotting with either anti-FGF receptor
(Fig. 5A) or anti-phosphotyrosine antibodies (Fig. 5B). As
shown in Fig. 5B, both mutated receptors were able to undergo
tyrosine autophosphorylation, albeit to a lesser extent than the
wild-type receptor, since four or five of the autophosphoryla-
tion sites have been removed. We next compared the pattern
of total cellular tyrosine-phosphorylated proteins upon stimu-
lation of wild-type or mutant FGFR1 with aFGF. As shown in
Fig. 5C, in both wild-type and mutant cell lines the most prom-
inent tyrosine-phosphorylated bands were MAP kinases, with
apparent molecular masses of 42 and 44 kDa (Fig. 5C). In
addition, more slowly migrating forms of MAP kinases were
noticed in all cells by immunoblotting with anti-Erk-2 antibod-
ies (Fig. 5D). Both experiments showed that mutated receptors
were able to activate MAP kinases to extents similar to that of
the wild-type FGFR1 (Fig. 5C and D).
The adaptor protein Shc was shown to be tyrosine phosphor-
ylated in response to aFGF stimulation of cells expressing
wild-type FGFR1 or the mutant FGFR1-Y766F (71, 75, 78). In
order to determine whether tyrosine phosphorylation of Shc
requires tyrosine autophosphorylation of a specific site on the
FGFR1, L-6 cells expressing wild-type FGFR1 or mutant re-
ceptors (FGFR1-4F and FGFR1-5F) were treated with aFGF,
lysed, immunoprecipitated with anti-Shc antibodies, and im-
munoblotted with anti-phosphotyrosine antibodies (Fig. 6A).
This experiment showed that mutant receptors were able to
tyrosine phosphorylate Shc proteins to the same level as the
wild-type receptor. We could not detect tyrosine-phosphory-
lated FGFR1 in immunoprecipitates of Shc nor vice versa (Fig.
6A and data not shown). We next examined whether the Grb2/
Sos complex associates directly or indirectly with the activated
wild-type or mutant FGFR1. Lysates from aFGF-treated or
untreated cells expressing wild-type or mutant FGFR1 were
immunoprecipitated with anti-Grb2 (Fig. 6B) or anti-Sos (Fig.
6C and D) antibodies and were immunoblotted with anti-phos-
photyrosine (Fig. 6B and D) or anti-Sos (Fig. 6C) antibodies.
Several phosphotyrosine-containing bands, with apparent mo-
lecular masses of 90, 65, and 52 kDa, were detected in the anti-
Grb2 immunoprecipitates in all of the cell lines tested (Fig.
6B). The 65- and 52-kDa bands represent different forms of
Shc, while the identity of the tyrosine-phosphorylated protein
with a molecular mass of 90 kDa (pp90) is unknown. Stimula-
tion of cells expressing the FGFR1 mutants caused a mobility
shift of Sos to a similar extent as that detected in cells express-
ing the wild-type FGF receptor (Fig. 6C). Interestingly, we also
noticed the presence of the 90-kDa phosphoprotein in the anti-
Sos immunoprecipitates from aFGF treated cell lines (Fig. 6D).
These results indicate that wild-type and mutant FGFR1 are
able to induce complex formation between Grb2/Sos and ty-
rosine-phosphorylated Shc as well as a yet unidentified ty-
rosine-phosphorylated protein of 90 kDa.
Finally, we investigated the ability of the wild-type and the
mutant FGFR1 to induce DNA synthesis in L-6 cells and
neuronal differentiation in PC12 cells. Both the FGFR1-4F
and FGFR1-5F mutants were able to induce DNA synthesis
with an aFGF dose dependence similar to the dose response
obtained with wild-type FGFR1 (Fig. 7A). Furthermore, aFGF
induced both accelerated appearance and greater length of
neurites in PC12 cells overexpressing the FGFR1-4F and
FGFR1-5F mutants to an extent similar to that in PC12 cells
overexpressing wild-type FGFR1. These results suggest that
the mutated receptors have a capacity equal to that of the
wild-type FGFR1 to induce neuronal differentiation (Fig. 7B).
It therefore appears that autophosphorylation of none of the
five tyrosines (Y-463, Y-583, Y-585, Y-730, and Y-766) is re-
quired for activation of mitogenic or differentiation pathways
In this report we describe the identification of new tyrosine
autophosphorylation sites in FGFR1 and delineate their roles
in receptor activation and signal transduction. The major au-
tophosphorylation site in FGFR1 has previously been mapped
to tyrosine 766 in the carboxy-terminal region of the receptor,
and this has been shown to be a binding site for the SH2
domain of PLC? (51, 52, 58). We have mapped six new auto-
phosphorylation sites in FGFR1 and have investigated their
importance for the regulation of FGFR1 kinase activity and
FGFR1-mediated mitogenesis or neuronal differentiation.
Two of the newly identified autophosphorylation sites, ty-
rosines 653 and 654, are located in a region homologous to
Y-416 of pp60c-src, the phosphorylation of which stimulates Src
kinase activity (38, 59). A group of tyrosine kinase receptors,
including FGF receptor, insulin receptor, insulin-like growth
factor I receptor, scatter factor/hepatocyte growth factor re-
ceptor (Met), and nerve growth factor receptor (Trk), also
have two tyrosine residues in homologous positions (25, 26). In
the case of insulin receptor, TrkA, TrkB, and Met, autophos-
phorylation occurs on both tyrosines and is important both for
kinase activity and for mediating biological responses (18, 21,
22, 24, 40, 41, 46, 50, 54, 62, 72, 80, 86). The three-dimensional
structure of insulin receptor kinase was determined by X-ray
crystallography, and it has been proposed that the kinase is
autoinhibited by the presence of tyrosine 1162 in the active site
of the catalytic domain. This would interfere with the binding
of exogenous substrates as well as with the binding of Mg-ATP
(30). Tyrosine 653 of FGFR1 is in the position homologous to
1162 of insulin receptor and therefore may play a similar role
in regulation of the kinase activity of FGFR1. However, our
results indicate that mutation of tyrosine 653 alone does not
affect the kinase activity of FGFR1. We demonstrate that the
presence of both autophosphorylated tyrosines 653 and 654 in
FGFR1 is necessary for full kinase activity in vitro and for its
biological responses in vivo. Mutation of Y-654 alone reduces
the kinase activity of FGFR1 in vitro and diminishes FGFR1-
mediated mitogenesis in L-6 cells and neuronal differentiation
984 MOHAMMADI ET AL.MOL. CELL. BIOL.
in PC12 cells. However, mutation of tyrosine 653 to phenylal-
anine is dispensable for both kinase activity and biological
responses of FGFR1. Although autophosphorylation on ty-
rosine 653 of FGFR1 was reported previously (28, 65), phos-
phorylation of tyrosine 654 has not been described. Further-
more, the role of these tyrosines in activation of the kinase
activity was not tested. Tyrosines 653 and 654 are conserved in
all known members of the FGF receptor family, and we pro-
pose that homologous tyrosines in other FGF receptors are
also important for activation of protein tyrosine kinase activity.
In addition, we have identified tyrosine 463 in the juxtamem-
brane region, tyrosines 583 and 585 in the kinase insert region,
FIG. 5. Mutation of tyrosines 463, 583, 585, 730, and 766 on FGFR1 does not affect MAP kinase activation in response to aFGF. Expression (A) and tyrosine
autophosphorylation (B) of wild-type and mutant FGF receptors in L-6 myoblasts. Anti-FGFR1 immunoprecipitates (IP) from aFGF-treated (?) or unstimulated (?)
cells expressing wild-type FGFR1 or the mutant FGFR1-4F and FGFR1-5F receptors were analyzed by immunoblotting with either anti-FGFR1 (A) or anti-
phosphotyrosine (anti-PTyr) (B) antibodies. (C) Comparison of tyrosine phosphorylation of intracellular proteins induced by aFGF in cells expressing mutant receptors.
Lysates from aFGF-treated or untreated cells expressing wild-type FGFR1 and FGFR1-4F and FGFR1-5F mutant receptors were analyzed by SDS-PAGE and
immunoblotting with anti-PTyr antibodies. The positions of the 42- and 44-kDa MAP kinases are indicated by arrows. (D) Activation of MAP kinase in response to
aFGF in cells expressing wild-type or mutant FGFR1. Cells were stimulated with aFGF (100 ng/ml) for 5 min and lysed, and lysates were analyzed by SDS-PAGE and
immunoblotting with anti-MAP kinase 2 (anti-MAPK2) antibodies.
VOL. 16, 1996 SIGNAL TRANSDUCTION BY FGFR1985
and tyrosine 730 in the second half of the conserved kinase
domain as four novel autophosphorylation sites of FGFR1.
Tyrosines 463, 583, and 585 are conserved among several FGF
receptors of human, mouse, chicken, Xenopus, Drosophila, and
Caenorhabditis elegans origin (15, 31, 37, 66). Autophosphory-
lation sites in the juxtamembrane region of PDGF receptor
were shown to be involved in the activation of Src family
kinases (53). Furthermore, autophosphorylation sites in the
juxtamembrane regions of insulin receptor and Trk are impor-
tant for tyrosine phosphorylation of insulin receptor substrate
1 or Shc, respectively (55, 72, 80). We as well as others have not
been able to demonstrate any direct interaction for activated
FGFR1 with Src or with Shc in vivo (reference 43 and our
unpublished observation), although there is one report de-
scribing association of Src with FGFR1 in vitro (85). Tyrosine
phosphorylation sites in the kinase insert region have been
mapped for several class III receptors, such as the PDGF and
colony-stimulating factor 1 receptors and have been shown to
be responsible for the activation of PI 3-kinase (8, 57). The
autophosphorylation sites in the kinase insert region of the
FGF receptor do not have any homology to known SH2 or
PTB domain binding consensus sequence motifs. Tyrosine 730
is located in the carboxy-terminal part of the second half of the
kinase domain and is conserved in the FGF receptors of all
species so far cloned (31, 34). This tyrosine is also found in
other receptors with tyrosine kinase activity, such as epidermal
growth factor, PDGF, colony-stimulating factor 1, and stem
cell factor receptors but is not present in the insulin receptor
(25). However, autophosphorylation of this tyrosine in these
receptors has not been reported. This tyrosine has a YMXM
motif which represents a consensus binding site for the SH2
domain of regulatory subunit of PI 3-kinase p85 (8). However,
we were not able to coimmunoprecipitate FGFR1 with p85 or
to detect any associated PI 3-kinase activity in anti-FGFR1
FIG. 6. Shc and pp90 are tyrosine phosphorylated and associate with Grb2 in response to aFGF stimulation of FGFR1-5F. Comparison of tyrosine phosphorylation
of Shc (A) and Grb2-associated pp90 (B) in response to aFGF in cells expressing wild-type and mutant FGF receptors. Stable cell lines expressing wild-type FGFR1
and the FGFR1-4F and FGFR1-5F mutant receptors were treated with aFGF (?) or left untreated (?). Lysates were then immunoprecipitated with anti-Shc (A),
anti-Grb-2 (B), or anti-Sos (C and D) antibodies and analyzed by SDS-PAGE and immunoblotting with anti-phosphotyrosine (anti-PTyr) (A, B, and D) or anti-Sos
antibodies (C). The small differences seen in tyrosine phosphorylation of Shc or pp90 are not significant. Comparable tyrosine phosphorylation of Shc or pp90 was
obtained with different cell lines expressing each mutant receptor. IP, immunoprecipitate.
986 MOHAMMADI ET AL.MOL. CELL. BIOL.
immunoprecipitates (reference 79 and our unpublished obser-
Using mutational analysis, we investigated the contribution
of these autophosphorylation sites to signaling via FGFR1.
Several experiments have demonstrated that activation of the
Ras/MAP kinase signaling pathway is crucial for FGFR1-me-
diated biological responses (42, 47, 61, 82, 83). Our results
show that removal of all the identified autophosphorylation
sites in FGFR1, except those involved in the activation of
receptor tyrosine kinase activity (Y-653 and Y-654), does not
perturb the ability of FGFR1 to recruit the Grb2/Sos complex
via tyrosine phosphorylation of Shc and/or pp90 or to activate
the MAP kinase cascade.
This observation contrasts with studies of other growth fac-
tor receptors, such as PDGF receptor, colony-stimulating fac-
tor 1 receptor, Trk, and Met, for which mutation of certain
autophosphorylation sites to phenylalanine impaired the abil-
ity to exert biological effects such as mitogenesis, transforma-
tion, or PC12 cell neuronal differentiation (19, 60, 72, 76, 77).
Thus, certain autophosphorylated tyrosines in these growth
factor receptors are responsible for recruitment of signaling
molecules essential for the activation of the Ras/MAP kinase
signaling pathway. Elimination of these sites abolishes the in-
duction of mitogenesis or differentiation by the respective li-
gands. An alternative mechanism appears to be employed by
other receptor tyrosine kinases, such as the insulin receptor.
The insulin receptor undergoes autophosphorylation on ty-
rosines but does not directly bind SH2 domain-containing pro-
teins in vivo. Instead, the insulin receptor phosphorylates other
cellular proteins such as Shc and insulin receptor substrate 1,
which then serve as docking proteins for SH2 domain-contain-
ing proteins (1a, 67, 68).
Our previous reports and the results presented in this study
suggest that autophosphorylation of FGFR1 has multiple roles
in signaling. Autophosphorylated tyrosines 653 and 654 regu-
late the kinase activity of FGFR1 and therefore are necessary
for all responses mediated by the kinase domain of FGFR1
(Fig. 8). Tyrosine 766 is a binding site for the SH2 domain of
PLC? and couples FGFR1 to PI metabolism and Ca2?mobi-
lization but is not required for the activation of the Ras/MAP
kinase signaling pathway or Ras-dependent biological re-
sponses (Fig. 8). The remaining four autophosphorylated ty-
rosines are not necessary for FGF-induced activation of the
Ras/MAP kinase signaling cascade, mitogenesis in L-6 cells, or
FIG. 7. Autophosphorylation on tyrosines 463, 583, 585, 730, and 766 is
dispensable for aFGF-induced mitogenesis and neuronal differentiation. (A)
Thymidine incorporation in response to aFGF in cells expressing wild-type (WT)
and mutant FGF receptors. Stable cell lines expressing wild-type FGFR1 or the
FGFR1-4F and FGFR1-5F mutant receptors were starved for 48 h in the pres-
ence of 0.1% FBS and then treated with different concentrations of aFGF or, as
a control, with 10% FBS for 24 h. [3H]thymidine was added, and after an
additional 16 h the amount of incorporated [3H]thymidine was determined. (B)
aFGF induces neuronal differentiation of PC12 cells expressing wild-type
FGFR1 or FGFR1-4F and FGFR1-5F mutant receptors. Parental PC12 cells or
PC12 cell lines expressing wild-type (PC12-FGFR1) or mutant (PC12-FGFR1-
4F and PC12-FGFR1-5F) receptors were stimulated with aFGF (100 ng/ml) and
scored for neurite outgrowth as a function of time. The weaker response of
parental PC12 cells is due to the low level of wild-type FGF receptors expressed
in these cells. The datum points represent the means of three independent
experiments (bars, standard errors).
FIG. 8. Schematic representation of all known tyrosine autophosphorylation
sites on FGFR1. The seven tyrosine autophosphorylation sites identified on
FGFR1 (flg) so far are shown. Autophosphorylation on tyrosines 653 and 654
activates the FGF receptor kinase domain. Autophosphorylation on tyrosine 766
couples FGF receptor to PLC? activation, which then leads to influx of Ca2?and
activation of protein kinase C. While this pathway is not required for aFGF-
induced mitogenesis, it is involved in the internalization of FGFR1. The function
of the remaining autophosphorylation sites, 463, 583, 585, and 730, remains
unclear. MAPK, MAP kinase.
VOL. 16, 1996 SIGNAL TRANSDUCTION BY FGFR1987
neuronal differentiation in PC12 cells (Fig. 8). It was recently
demonstrated that a C. elegans homolog of FGFR is required
for normal migrations of the sex myoblasts in C. elegans her-
maphrodites (15). It is possible that these autophosphorylation
sites are involved in plasminogen activation and chemotaxis in
response to aFGF stimulation. We were not able to demon-
strate association of any known SH2 domain-containing pro-
teins, except for PLC?, with the activated FGFR1 in several
cell types. This could be due to a weak or transient interaction
between FGFR1 and its targets, or alternatively this may sug-
gest that the remaining autophosphorylated tyrosines do not
serve as docking sites for the known SH2 domain-containing
proteins. The mechanism of FGFR1-induced activation of the
Ras/MAP kinase signaling cascade is not entirely clear at this
moment. We have noticed tyrosine phosphorylation of Shc and
an as yet unidentified 90-kDa protein (pp90) upon stimulation
of wild-type or mutant FGFR1 (FGFR1-5F). Tyrosine-phos-
phorylated Shc binds the Grb2/Sos complex and could link
FGFR1 to the Ras signaling pathway (71). In addition, we have
demonstrated that tyrosine-phosphorylated pp90 associates
with the Grb2/Sos complex and may alternatively link FGFR1
to the Ras signaling pathway. Therefore, FGFR1 may signal in
a manner somewhat similar to that of the insulin receptor, by
employing docking molecules such as Shc or p90 to couple to
the Ras and other signaling pathways (Fig. 8).
M.M. and I.D. contributed equally to this work.
We thank S. Hubbard and M. Lemmon for critical reading of this
manuscript and helpful suggestions. The secretarial help of Janice
Small is acknowledged.
This work was supported by a grant from Sugen Inc. (to J.S.).
1. Allen, G. 1989. Sequencing of proteins and peptides, 2nd ed. Elsevier, New
1a.Backer, J. M., M. G. Myers, Jr., S. E. Shoelson, D. J. Chin, X. J. Sun, M.
Miralpeix, P. Hu, B. Margolis, E. Y. Skolnik, J. Schlessinger, and M. F.
White. 1992. Phosphatidylinositol 3?-kinase is activated by association with
IRS-1 during insulin stimulation. EMBO J. 11:3469–3479.
2. Batzer, A. G., D. Rotin, J. M. Urena, E. Y. Skolnik, and J. Schlessinger. 1994.
Hierarchy of binding sites for Grb2 and Shc on the epidermal growth factor
receptor. Mol. Cell. Biol. 14:5192–5201.
3. Bellot, F., G. Crumley, J. Kaplow, J. Schlessinger, M. Jaye, and C. A.
Dionne. 1991. Ligand-induced transphosphorylation between different FGF
receptors. EMBO J. 10:2849–2854.
4. Blaikie, P., D. Immanuel, J. Wu, N. Li, V. Yajnik, and B. Margolis. 1994. A
region distinct from the SH2 domain can bind tyrosine-phosphorylated
growth factor receptors. J. Biol. Chem. 269:32031–32034.
5. Bork, P., and B. Margolis. 1995. A new phosphotyrosine interaction domain.
6. Burgess, W. H., C. A. Dionne, J. Kaplow, R. Mudd, R. Friesel, A. Zilberstein,
J. Schlessinger, and M. Jaye. 1990. Characterization and cDNA cloning of
phospholipase C-?, a major substrate for heparin-binding growth factor 1
(acidic fibroblast growth factor)-activated tyrosine kinase. Mol. Cell. Biol.
7. Burgess, W., and T. Maciag. 1989. Heparin binding (fibroblast) growth factor
family of proteins. Annu. Rev. Biochem. 58:575–606.
8. Cantley, L. C., K. R. Auger, C. Carpenter, B. Duckworth, A. Graziani, R.
Kapeller, and S. Soltoff. 1991. Oncogenes and signal transduction. Cell 64:
9. Chellaiah, A. T., D. G. McEwen, S. Werner, J. Xu, and D. M. Ornitz. 1994.
Fibroblast growth factor receptor (FGFR3). Alternative splicing in immu-
noglobulin-like domain III creates a receptor highly specific for acidic FGF/
FGF-1. J. Biol. Chem. 269:11620–11627.
10. Chen, C., and H. Okayama. 1987. High-efficiency transformation of mam-
malian cells by plasmid DNA. Mol. Cell. Biol. 7:2745–2752.
11. Clyman, R. I., K. G. Peters, Y. Q. Chen, J. Escobedo, L. T. Williams, H. E.
Ives, and E. Wilson. 1994. Phospholipase C gamma activation, phosphati-
dylinositol hydrolysis, and calcium mobilization are not required for FGF
receptor-mediated chemotaxis. Cell Adhesion Commun. 1:333–342.
12. Cooper, J. A., B. M. Sefton, and T. Hunter. 1983. Detection and quantifica-
tion of phosphotyrosine in proteins. Methods Enzymol. 99:387–402.
13. Coughlin, S. R., P. J. Barr, L. S. Cousens, L. J. Fretto, and L. T. Williams.
1988. Acidic and basic fibroblast growth factors stimulate tyrosine kinase
activity in vivo. J. Biol. Chem. 263:988–993.
14. Crumley, G., F. Bellot, J. M. Kaplow, J. Schlessinger, M. Jaye, and C. A.
Dionne. 1991. High-affinity binding and activation of a truncated FGF re-
ceptor by both aFGF and bFGF. Oncogene 6:2255–2262.
15. DeVore, D. L., H. R. Horvitz, and M. J. Stern. 1995. An FGF receptor
signaling pathway is required for the normal cell migrations of the sex
myoblasts in C. elegans hermaphrodites. Cell 83:611–620.
16. Dikic, I., A. G. Batzer, P. Blaikie, A. Obermeier, A. Ullrich, J. Schlessinger,
and B. Margolis. 1995. Shc binding to NGF receptor is mediated by the PI
domain. J. Biol. Chem. 270:15125–15129.
17. Dikic, I., J. Schlessinger, and I. Lax. 1994. PC12 cells overexpressing the
insulin receptor undergo insulin dependent neuronal differentiation. Curr.
18. Ellis, L., E. Clauser, D. O. Morgan, E. Edery, R. A. Roth, and W. J. Rutter.
1986. Replacement of insulin receptor tyrosine residues 1162 and 1163
compromises insulin-stimulated kinase activity and uptake of 2-deoxyglu-
cose. Cell 45:721–732.
19. Fantl, W. J., J. A. Escobedo, G. A. Martin, C. W. Turck, M. del Rosario, F.
McCormick, and L. T. Williams. 1992. Distinct phosphotyrosines on a
growth factor receptor bind to specific molecules that mediate different
signaling pathways. Cell 69:413–423.
20. Fantl, W. J., D. E. Johnson, and L. T. Williams. 1993. Signaling by receptor
tyrosine kinases. Annu. Rev. Biochem. 62:453–481.
21. Ferracini, R., P. Longati, L. Naldini, E. Vigna, and P. M. Comoglio. 1991.
Identification of the major autophosphorylation site of the Met/hepatocyte
growth factor receptor tyrosine kinase. J. Biol. Chem. 266:19558–19564.
22. Flores-Riveros, J. R., E. Sibley, T. Kastelic, and M. D. Lane. 1989. Substrate
phosphorylation catalyzed by the insulin receptor tyrosine kinase. Kinetic
correlation to autophosphorylation of specific sites in the beta subunit. J.
Biol. Chem. 264:21557–21572.
23. Friesel, R., W. H. Burgess, and T. Maciag. 1989. Heparin-binding growth
factor 1 stimulates tyrosine phosphorylation in NIH 3T3 cells. Mol. Cell.
24. Guiton, M., F. J. Gunn-Moore, T. N. Stitt, G. D. Yancopoulos, and J. M.
Tavare. 1994. Identification of in vivo brain-derived neurotrophic factor-
stimulated autophosphorylation sites on the TrkB receptor tyrosine kinase
by site-directed mutagenesis. J. Biol. Chem. 269:30370–30377.
25. Hanks, S. K. 1991. Eukaryotic protein kinases. Curr. Opin. Struct. Biol. 1:
26. Hanks, S. K., A. M. Quinn, and T. Hunter. 1988. The protein kinase family:
conserved features and deduced phylogeny of the catalytic domains. Science
27. Hou, J., M. Kan, K. McKeehan, G. McBride, P. Adams, and W. L. McKee-
han. 1991. Fibroblast growth factor receptors from liver vary in three struc-
tural domains. Science 251:665–668.
28. Hou, J., K. McKeehan, M. Kan, S. A. Carr, M. J. Huddeston, J. W. Crabb,
and W. L. McKeehan. 1993. Identification of tyrosines 154 and 307 in the
extracellular domain and 653 and 766 in the intracellular domain as phos-
phorylation sites in the heparin-binding fibroblast growth factor receptor
tyrosine kinase (flg). Protein Sci. 2:86–92.
29. Huang, J., M. Mohammadi, G. A. Rodrigues, and J. Schlessinger. 1995.
Reduced activation of RAF-1 and MAP kinase by a fibroblast growth factor
receptor mutant deficient in stimulation of phosphatidylinositol hydrolysis. J.
Biol. Chem. 270:5065–5072.
30. Hubbard, S. R., L. Wei, L. Ellis, and W. A. Hendrickson. 1994. Crystal
structure of the tyrosine kinase domain of the human insulin receptor.
Nature (London) 372:746–754.
31. Jaye, M., J. Schlessinger, and C. A. Dionne. 1992. Fibroblast growth factor
receptor tyrosine kinases: molecular analysis and signal transduction. Bio-
chim. Biophys. Acta 1135:185–199.
32. Johnson, D. E., P. L. Lee, J. Lu, and L. T. Williams. 1990. Diverse forms of
a receptor for acidic and basic fibroblast growth factors. Mol. Cell. Biol. 10:
33. Johnson, D. E., J. Lu, H. Chen, S. Werner, and L. T. Williams. 1991. The
human fibroblast growth factor receptor genes: a common structural ar-
rangement underlies the mechanisms for generating receptor forms that
differ in their third immunoglobulin domain. Mol. Cell. Biol. 11:4627–4634.
34. Johnson, D. E., and L. T. Williams. 1993. Structural and functional diversity
in the FGF receptor multigene family. Adv. Cancer Res. 60:1–41.
35. Kavanaugh, W. M., C. W. Turck, and L. T. Williams. 1995. PTB domain
binding to signaling proteins through a sequence motif containing phospho-
tyrosine. Science 268:1177–1179.
36. Kavanaugh, W. M., and L. T. Williams. 1994. An alternative to SH2 domains
for binding tyrosine-phosphorylated proteins. Science 266:1862–1865.
37. Klambt, C., L. Glazer, and B. C. Shilo. 1992. Breathless, a Drosophila FGF
receptor homolog, is essential for migration of tracheal and specific midline
glial cells. Genes Dev. 6:1668–1678.
38. Kmiecik, T. E., and D. Shalloway. 1987. Activation and suppression of
pp60c-src transforming ability by mutation of its primary sites of tyrosine
phosphorylation. Cell 49:65–73.
39. Koch, C. A., D. Anderson, M. F. Moran, C. Ellis, and T. Pawson. 1991. SH2
988MOHAMMADI ET AL.MOL. CELL. BIOL.
and SH3 domains: elements that control interaction of cytoplasmic signaling Download full-text
proteins. Science 252:668–674.
40. Kohanski, R. A. 1993. Insulin receptor autophosphorylation. II. Determina-
tion of autophosphorylation sites by chemical sequence analysis and identi-
fication of the juxtamembrane sites. Biochemistry 32:5773–5780.
41. Komada, M., and N. Kitamura. 1994. Regulatory role of major tyrosine
autophosphorylation site of kinase domain of c-Met receptor (scatter factor/
hepatocyte growth factor receptor). J. Biol. Chem. 269:16131–16136.
42. Kremer, N. E., G. D’Arcangelo, S. M. Thomas, M. De Marco, J. S. Brugge,
and S. Halegoua. 1991. Signal transduction by nerve growth factor and
fibroblast growth factor in PC12 cells requires a sequence of Src and Ras
actions. J. Cell Biol. 115:809–819.
43. Landgren, E., P. Blume-Jensen, S. A. Courtneidge, and L. Claesson-Welsh.
1995. Fibroblast growth factor receptor-1 regulation of Src family kinases.
44. Li, N., A. Batzer, R. Daly, V. Yajnik, E. Y. Skolnik, P. Chardin, D. Bar-Sagi,
B. Margolis, and J. Schlessinger. 1993. Guanine nucleotide releasing factor
hSos1 binds to Grb2 and links receptor tyrosine kinase to Ras signaling.
Nature (London) 363:85–88.
45. Li, N., J. Schlessinger, and B. Margolis. 1994. Autophosphorylation mutants
of the EGF-receptor signal through auxiliary mechanisms involving SH2
domain proteins. Oncogene 9:3457–3465.
46. Longati, P., A. Bardelli, C. Ponzetto, L. Naldini, and P. M. Comoglio. 1994.
Tyrosines 1234-1235 are critical for activation of the tyrosine kinase encoded
by the MET proto-oncogene (HGF receptor). Oncogene 9:49–57.
47. MacNicol, A. M., A. J. Muslin, and L. T. Williams. 1993. Raf-1 kinase is
essential for early Xenopus development and mediates the induction of
mesoderm by FGF. Cell 73:571–583.
48. Margolis, B. 1992. Proteins with SH2 domains: transducers in the tyrosine
kinase signaling pathway. Cell Growth Differ. 3:73–80.
49. Margolis, B., A. Zilberstein, C. Franks, S. Felder, S. Kremer, A. Ullrich,
S. G. Rhee, K. Skorecki, and J. Schlessinger. 1990. Effect of phospholipase
C-gamma overexpression on PDGF-induced second messengers and mito-
genesis. Science 248:607–610.
50. Middlemas, D. S. J. Meisenhelder, and T. Hunter. 1994. Identification of
TrkB autophosphorylation sites and evidence that phospholipase C-gamma
1 is a substrate of the TrkB receptor. J. Biol. Chem. 269:5458–5466.
51. Mohammadi, M., C. A. Dionne, W. Li, N. Li, T. Spivak, A. M. Honegger, M.
Jaye, and J. Schlessinger. 1992. Point mutation in FGF receptor eliminates
phosphatidylinositol hydrolysis without affecting mitogenesis. Nature (Lon-
52. Mohammadi, M., A. M. Honegger, D. Rotin, R. Fisher, F. Bellot, W. Li, C. A.
Dionne, M. Jaye, M. Rubinstein, and J. Schlessinger. 1991. A tyrosine-
phosphorylated carboxy-terminal peptide of the fibroblast growth factor
receptor (flg) is a binding site for the SH2 domain of phospholipase C-?1.
Mol. Cell. Biol. 11:5068–5078.
53. Mori, S., L. Ronnstrand, K. Yokote, A. Engstrom, S. A. Courtneidge, L.
Claesson-Welsh, and C. H. Heldin. 1993. Identification of two juxtamem-
brane autophosphorylation sites in the PDGF beta-receptor; involvement in
the interaction with Src family tyrosine kinases. EMBO J. 12:2257–2264.
54. Naldini, L., E. Vigna, R. Ferracini, P. Longati, L. Gandino, M. Prat, and
P. M. Comoglio. 1991. The tyrosine kinase encoded by the MET proto-
oncogene is activated by autophosphorylation. Mol. Cell. Biol. 11:1793–1803.
55. Obermeier, A., R. A. Bradshaw, K. Seedorf, A. Choidas, J. Schlessinger, and
A. Ullrich. 1994. Neuronal differentiation signals are controlled by nerve
growth factor receptor/Trk binding sites for SHC and PLC gamma. EMBO
56. Pawson, T. 1995. Protein modules and signaling networks. Nature (London)
57. Pawson, T., and J. Schlessinger. 1993. SH2 and SH3 domains. Curr. Biol. 3:
58. Peters, K. G., J. Marie, E. Wilson, H. E. Ives, J. Escobedo, M. Del Rosario,
D. Mirda, and L. T. Williams. 1992. Point mutation of an FGF receptor
abolishes phospatidylinositol turnover and Ca2? flux but not mitogenesis.
Nature (London) 358:678–781.
59. Piwnica-Worms, H., K. B. Saunders, T. M. Roberts, A. E. Smith, and S. H.
Cheng. 1987. Tyrosine phosphorylation regulates the biochemical and bio-
logical properties of pp60c-src. Cell 49:75–82.
60. Ponzetto, C., A. Bardelli, Z. Zhen, F. Maina, P. dalla Zonca, S. Giordano, A.
Graziani, G. Panayotou, and P. M. Comoglio. 1994. A multifunctional dock-
ing site mediates signaling and transformation by the hepatocyte growth
factor/scatter factor receptor family. Cell 77:261–271.
61. Reichman-Fried, M., B. Dickson, E. Hafen, and B. Z. Shilo. 1994. Elucida-
tion of the role of breathless, a Drosophila FGF receptor homolog, in
tracheal cell migration. Genes Dev. 8:428–439.
62. Rodrigues, G. A., and M. Park. 1994. Autophosphorylation modulates the
kinase activity and oncogenic potential of the Met receptor tyrosine kinase.
63. Rosen, O. M., R. Herrera, Y. Olowe, L. M. Petruzzelli, and M. H. Cobb. 1983.
Phosphorylation activates the insulin receptor tyrosine protein kinase. Proc.
Natl. Acad. Sci. USA 80:3237–3240.
64. Schlessinger, J., and A. Ullrich. 1992. Growth factor signaling by receptor
tyrosine kinases. Neuron 9:383–391.
65. Shi, E., M. Kan, J. Xu, F. Wang, J. Hou, and W. L. McKeehan. 1993. Control
of fibroblast growth factor receptor kinase signal transduction by het-
erodimerization of combinatorial splice variants. Mol. Cell. Biol. 13:3907–
66. Shishido, E., S. Higashijima, Y. Emori, and K. Saigo. 1993. Two FGF-
receptor homologues of Drosophila: one is expressed in mesodermal pri-
mordium in early embryos. Development 117:751–761.
67. Skolnik, E. Y., A. Batzer, N. Li, C. H. Lee, E. Lowenstein, M. Mohammadi,
B. Margolis, and J. Schlessinger. 1993. The function of Grb2 in linking the
insulin receptor to Ras signaling pathway. Science 260:1953–1955.
68. Skolnik, E. Y., C. H. Lee, A. G. Batzer, L. M. Vicentini, M. Zhou, R. Daly,
M. J. Myers, J. M. Backer, A. Ullrich, M. F. White, and J. Schlessinger. 1993.
The SH2/SH3 domain containing protein Grb2 interacts with tyrosine phos-
phorylated IRS1 and Shc: implications for insulin control of Ras signaling.
EMBO J. 12:1929–1936.
69. Sorokin, A., M. Mohammadi, J. Huang, and J. Schlessinger. 1994. Internal-
ization of fibroblast growth factor is inhibited by a point mutation at tyrosine
766. J. Biol. Chem. 269:17056–17061.
70. Spivak-Kroizman, T., M. A. Lemmon, I. Dikic, J. E. Ladbury, D. Pinchasi, J.
Huang, M. Jaye, G. Crumley, J. Schlessinger, and I. Lax. 1994. Heparin-
induced oligomerization of FGF molecules is responsible for FGF receptor
dimerization, activation, and cell proliferation. Cell 79:1015–1024.
71. Spivak-Kroizman, T., M. Mohammadi, P. Hu, M. Jaye, J. Schlessinger, and
I. Lax. 1994. Point mutation in the fibroblast growth factor receptor elimi-
nates phosphatidylinositol hydrolysis without affecting neuronal differentia-
tion of PC12 cells. J. Biol. Chem. 269:14419–14423.
72. Stephens, R. M., D. M. Loeb, T. D. Copeland, T. Pawson, L. A. Green, and
D. R. Kaplan. 1994. Trk receptors use redundant signal transduction path-
ways involving Shc and PLC-gamma 1 to mediate NGF responses. Neuron
73. Ueno, H., M. Gunn, K. Dell, A. Tseng, and L. T. Williams. 1992. A truncated
form of fibroblast growth factor receptor 1 inhibits signal transduction by
multiple types of fibroblast growth factor receptor. J. Biol. Chem. 267:
74. Ullrich, A., and J. Schlessinger. 1990. Signal transduction by receptors with
tyrosine kinase activity. Cell 61:203–212.
75. Vainikka, S., V. Joukov, S. Wennstrom, M. Bergman, P. G. Pelicci, and K.
Alitalo. 1994. Signal transduction by fibroblast growth factor receptor-4
(FGFR-4). Comparison with FGFR-1. J. Biol. Chem. 269:18320–18326.
76. Valius, M., and A. Kazlauskas. 1993. Phospholipase C-gamma 1 and phos-
phatidylinositol 3 kinase are the downstream mediators of the PDGF recep-
tor’s mitogenic signal. Cell 73:321–334.
77. Van der Geer, P., and T. Hunter. 1993. Mutation of Tyr697, a GRB2-binding
site, and Tyr721, a PI 3-kinase binding site, abrogates signal transduction by
the murine CSF-1 receptor expressed in Rat-2 fibroblasts. EMBO J. 12:
78. Wang, J. K., G. Gao, and M. Goldfarb. 1994. Fibroblast growth factor
receptors have different signaling and mitogenic potentials. Mol. Cell. Biol.
79. Wennstroem, S., E. Landgren, P. Blume-Jensen, and L. Claesson-Welsh.
1992. The platelet-derived growth factor ?-receptor kinase insert confers
specific signaling properties to a chimeric fibroblast growth factor receptor.
J. Biol. Chem. 267:13749–13756.
80. White, M. F., J. N. Livingston, J. M. Backer, V. Lauris, T. J. Dull, A. Ullrich,
and C. R. Kahn. 1988. Mutation of the insulin receptor at tyrosine 960
inhibits signal transmission but does not affect its tyrosine kinase activity.
81. White, M. F., S. E. Shoelson, H. Keutmann, and C. R. Kahn. 1988. A cascade
of tyrosine autophosphorylation in the beta-subunit activates the phospho-
transferase of the insulin receptor. J. Biol. Chem. 263:2969–2980.
82. Whitman, M., and D. A. Melton. 1992. Involvement of p21ras in Xenopus
mesoderm induction. Nature (London) 357:252–254.
83. Wood, K. W., C. Sarnecki, T. M. Roberts, and J. Blenis. 1992. Ras mediates
nerve growth factor receptor modulation of three signal-transducing protein
kinases: MAP kinase, Raf-1, and RSK. Cell 68:1041–1050.
84. Yu, K. T., and M. P. Czech. 1984. Tyrosine phosphorylation of the insulin
receptor beta subunit activates the receptor-associated tyrosine kinase activ-
ity. J. Biol. Chem. 259:5277–5286.
85. Zhan, X., C. Plourde, X. Hu, R. Friesel, and T. Maciag. 1994. Association of
fibroblast growth factor receptor-1 with c-Src correlates with association
between c-Src and cortactin. J. Biol. Chem. 269:20221–20224.
86. Zhang, B., J. M. Tavare, L. Ellis, and R. A. Roth. 1991. The regulatory role
of known tyrosine autophosphorylation sites of the insulin receptor kinase
domain. An assessment by replacement with neutral and negatively charged
amino acids. J. Biol. Chem. 266:990–996.
VOL. 16, 1996SIGNAL TRANSDUCTION BY FGFR1 989