The EMBO Journal vol.12 no.3 pp.933-941, 1993
Tyrosine 785 is a major determinant of Trk -substrate
Axel Obermeierl, Hartmut Halfter',
Karl-Heinz Wiesmuller2, Gunther Jung2,
Joseph Schlessinger3 and Axel Ulrich1'4
1Department of Molecular Biology, Max-Planck-Institut fiir Biochemie,
Am Klopferspitz 18A, 8033 Martinsried, 2Institut fiir Organische
Chemie, Eberhard-Karls-Universitat Tiibingen, Auf der Morgenstelle
18, 7400 Tiibingen, Germany and 3Department of Pharnacology, New
York University Medical Center, 550 First Avenue, New York, NY
Communicated by A.Ullrich
receptor/Trk with cellular substrates was investigated by
transient co-overexpression in human 293 fibroblasts
using ET-R, a chimeric receptor consisting of the
epidermal growth factor receptor (EGF-R) extraceliular
ligand binding domain and the Trk transmembrane and
intracellular signal-generating sequences. The chimera
fully functional, and associated
phosphorylated phospholipaseC7y(PLC-y), ras GTPase-
activating protein (GAP) and the non-catalytic subunit
dependent manner. Deletion of 15 C-terminal amino
acids, including tyrosine 785 (Y-785) abrogated receptor
and substrate phosphorylation activities. Mutation of
Y-785 to phenylalanine somewhat impaired receptor
phosphorylation activity, which was reflected in reduced
GAP and p85 phosphorylation. In contrast, ET-YF
phosphorylation ofPLCywas significantly reduced, while
the high affinity association potential with this substrate
was abrogated by this point mutation in vitro and in intact
cells. Furthermore, a tyrosine-phosphorylated synthetic
cytoplasmic domain association with PLCy. Thus, the
short C-terminal tail appears to be a crucial structural
of the Trk cytoplasmic domain, and
phosphorylated Y-785 is a major and selective interac-
tion site for PLC-y.
Key words: nerve growth factor/signal transduction/tyrosine
p85, in a ligand-
The missing link between nerve growth factor (NGF) and
cellular signalling mechanisms that lead to the formation of
neurites has been identified by the recent finding that the
proto-oncogene product gpl4Ot* (Trk) (Martin-Zanca et al.,
activity, is a functional receptor for this neurotrophic factor
(Kaplan et al., 1991a,b; Klein et al.,
tyrosine kinases (RTKs) such as epidermal growth factor
receptor (EGF-R), platelet-derived growth factor receptor
a transmembrane receptor with tyrosine kinase
Oxford University Press
(PDGF-R) and macrophage colony stimulating factor
receptor (CSF-lR) are known to function as molecular
switches that transduce extracellular binding to their cognate
ligand into intracellular events, ultimately resulting in DNA
synthesis and cell proliferation or differentiation (reviewed
in Ullrich and Schlessinger, 1990; Cantley et al., 1991).
Ligand binding induces receptor dimerization (Schlessinger,
1988; Ullrich and Schlessinger, 1990) and autophosphory-
lation, followed by association and tyrosine phosphorylation
of a specific subset of cellular protein substrates. Among
these are the RTK substrates phospholipase COy (PLC'y),
GTPase-activating protein (GAP) and the p85 subunit of
phosphatidylinositol 3'-kinase (PI3'-K), which are thought
to be involved in distinct intracellular signal transduction
pathways. Different RTKs appear to utilize distinct sets of
signalling polypeptides to exert their final effects. For
example, PLC&y is a substrate for the tyrosine kinases of
PDGF-R (Meisenhelder et al., 1989; Wahl et al., 1989;
Margolis et al., 1990), fibroblastic growth factor receptors
(FGF-Rs) (Burgess et al., 1990; Mohammadi et al., 1991)
and EGF-R (Margolis et al., 1989; Meisenhelder et al.,
1989; Nishibe et al., 1989), but not for CSF-1R (Downing
et al., 1989) and insulin receptor (Nishibe et al., 1990). In
addition, GAP has been found to be a substrate for EGF-R
and PDGF-R (Kaplan et al., 1990; Kazlauskas et al., 1990),
but not for basic FGF-R (Molloy et al., 1989).
PLC-y, GAP and p85 are thought to bind to phospho-
tyrosine (PY) residues on their respective phosphorylated,
therefore activated, RTKs via their src homology region 2
(SH2) domains (Moran et al., 1990; Cantley et al., 1991;
Koch et al., 1991). Variations in SH2 domains, together with
different amino acid sequences surrounding PY residues, are
likely to be responsible for specific substrate binding to
particular PY residues on distinct receptors (Koch et al.,
1991). For example, Y-992 and Y-1068 on human EGF-R
and Y-766 on chicken FGF-R (Flg) have been identified
as high affinity binding sites for PLC'y SH2 domains
(Mohammadi et al., 1991, 1992; Peters et al., 1992; Rotin
et al., 1992). Similarly, tyrosine residues in the C-tail of
mouse and human ,BPDGF-R appear to be involved in
receptor-PLCy interactions (Ronnstrand et al.,
Seedorf et al., 1992).
Although accumulating evidence points to Trk as the only
component of the high affinity, signal-generating NGF-R
(Klein et al., 1991; Weskamp and Reichardt, 1991; Ibaniiez
et al., 1992; Meakin et al., 1992), it is not yet entirely clear
whether the low affinity NGF-R, p75LNGFR, iS necessary in
addition to Trk to form a high affinity binding site for NGF
and to elicit a full biological response (Hempstead et al.,
1990, 1991; Ragsdale and Woodgett, 1991). To address this
question further and to investigate the significance of
structural domains of Trk signal generation, we utilized a
receptor chimera approach that had been instrumental in
previous studies of RTK function (Riedel et al., 1986;
Lammers et al., 1989; Seedorf et al., 1991). The ET-R
A.Obermeier at al.
Fig. 1. Schematic diagram of receptors. EGF-R extracellular domain
cysteine-rich regions are shown as black boxes; black circles in the
extracellular domain of Trk indicate single cysteine residues. Tyrosine
kinase domains are represented by shaded (EGF-R) and open (Trk)
chimera, consisting of EGF-R extracellular sequences and
Trk transmembrane and cytoplasmic domains, was fully
active in ligand-dependent auto- and substrate phos ho
tion functions, which further confirmed that p75LNGFK
dispensable for NGF signalling through the Trk system. The
demonstrating that PLCy, GAP and p85 are substrates of
the Trk tyrosine kinase. Surprisingly, deletion of the 15
amino acid C-terminal tail rendered the Trk kinase inactive
with respect to autophosphorylation as well as substrate
phosphorylation. Most importantly, our data demonstrate that
the C-terminal-most tyrosine residue of Trk, Y-785, is an
essential part of the PLC-y-binding site.
Expression and autophosphorylation activities of ET-R
To examine the signalling capacity of Trk, we constructed
a chimeric receptor, ET-R, which consists of the EGF-R
extracellular domain fused to the transmembrane and
cytoplasmic sequences of the NGF receptor, Trk (Figure 1).
In addition, three mutations were introduced into the ET-R
background, yielding a kinase-negative point mutant, ET-
KM, a C-terminal truncation mutant, ET-ACT, and a point
mutant, ET-YF, containing a phenylalanine in place of the
C-terminal domain tyrosine residue at position 785 (Figure
Expression plasmids coding for ET-R, ET-KM, ET-ACT
and ET-YF were transiently transfected into a human
fibroblast cell line, 293. Metabolic radiolabelling with
[35S]methionine and immunoprecipitation demonstrated that
all chimeric receptors were synthesized, each appearing as
two bands in SDS-PAGE (Figure 2). The upper bands,
presumably representing the fully processed receptors,
migrated at molecular weights of 142 kDa (ET-R and point
mutants) and 141 kDa (ET-ACT). In each case, the lower
bands migrated at 130 kDa and probably represent high
mannose oligosaccharide-bearing receptors, which fail to exit
the endoplasmic reticulum because they are preferentially
localized inside the cells.
EGF addition to intact transfected cells resulted in a
significant increase in autophosphorylation activity of ET-
Fig.2.Expression andtyrosine phosphorylation of ET-R and mutants.
293 cells transiently expressing ET-R, ET mutants, or parent receptors
(EGF-R andTrk) andmetabolically radiolabelled with [35S]methionine
were stimulated with EGF(EGF-R, ET-R and mutants; +) or NGF
(Trk; +), or left unstimulated (-), lysed and thenimmunoprecipitated
with mAb 108.1 (caEGF-R) (lanes 1-10) or ATC (caTrk) (lanes 11
and 12). Samples weresubjected to 7.5% SDS-PAGE and
electrophoretically transferred to nitrocellulose. Expression levels were
monitoredby autoradiography (upper panel). Phosphotyrosine contents
of the same receptor proteins were analyzed by immunoblotting with
mAb 5E2 (aPY) (lower panel).
R and ET-YF(Figure 2, lowerpanel, lanes 1, 2, 7 and8).
Thephosphorylation activity ofET-YF was somewhat lower
than that ofET-R, indicating that the C-terminaltyrosine
might represent an autophosphorylation site. Asexpected,
ET-KM did notdisplay any kinase activity, yet surprisingly,
phosphorylation of ET-ACT could also not be detected.
Receptor phosphorylation is thought to occur by inter-
molecular transphosphorylation between dimerized
oligomerized receptors in response to ligand stimulation
Transphosphorylation studies employing EGF-R-insulin
receptor chimerae (Ballotti et al., 1989; Lammers etal.,
1990)or an EGF-R-c-kit chimera (Herbst et al., 1991) have
suggestedthat bothhomologous extracellular orcytoplasmic
domainsmay be sufficient to mediate receptor dimerization
demonstrated that transphosphorylation can occur even
between RTKs as distantly related as Trk and EGF-R. As
shown inFigure 3 (lanes 7-10), the kinase-deficient EGF-
R-KA mutant (Honegger et al., 1987) wasphosphorylated
ontyrosines whencoexpressed with the Trk tyrosine kinase
interestingly, transphosphorylation of kinase-negative ET-
KM by EGF-R was not detectable. Thus, Trk cytoplasmic
sequences do not appear to be a substrate for the EGF-R
tyrosine kinase in intact cells, while the reverse ispossible.
Tyrosine phosphorylation of cellular proteins by ET-R
Distinct substrate specificities of different RTKs result in
receptor- and cell type-specific pleiotropic responses (for
OX aM UdKe
Fig. 3. Receptor transphosphorylation. 293 cells transit
the respective receptors or coexpressing combinations (
kinase-negative receptors were treated with EGF (as w
vanadate) (+) or left untreated (-), lysed and then im
precipitated with either mAb 108.1 (aEGF-R) or polyc
ATC (aTrk). Samples were subjected to 6% SDS-PA
electrophoretically transferred to nitrocellulose. Phosph
containing proteins were detected with mAb 5E2 (aPY
by the ECL system.
Fig. 4. Tyrosine phosphorylation of cellular proteins. i
transiently transfected with expression vector pCMV-1
control or expression constructs for EGF-R, ET-R or ]
were stimulated with EGF (+) or not (-) and lysed.
precleared and mixed with sample buffer, and proteins
separated by 9.5% SDS-PAGE. After transfer to nitr
tyrosine-phosphorylated proteins were detected by imm
mAb 5E2 (aPY) and visualized with the ECL system.
p52 is indicated by an arrowhead.
Schlessinger and Ulirich, 1992). To investigate
basis of Trk-specific signal generation, we fli
the total spectrum of substrates phosphorylate
stimulated Trk tyrosine kinase of ET-R.
Lysates of transfected, EGF-stimulated 2
subjected directly to SDS-PAGE and analyze
phosphorylated proteins by immunoblottir
phosphotyrosine mAb 5E2 (aPY) (Figure 4)
of the substrate phosphorylation patterns pri
Trk (lanes 5 and 6) and EGF-R (lanes 3 al
kinases revealed similarities and some clear differences. Four
proteins in the range of 76-95 kDa were much more
strongly phosphorylated by EGF-R
designated p35, p40, p41 and p52, appeared
predominantly phosphorylated by ET-R. These results
characteristics for Trk and EGF-R kinases undercomparable
The phosphorylation pattern of ET-YF was virtually
phosphorylation activity,whichespeciallyaffected thehigher
molecular weight regionbut did notappearto influence the
bands below 50 kDa. This was consistent with the reduced
autophosphorylationlevel ofET-YF relative to ET-R(Figure
2). Surprisingly, the truncation mutant ET-ACT lacked both
auto- and substratephosphorylationfunction,while this was
expected for the ET-KM mutant receptor. Thus, these
experiments indicate that the C-terminal 15 amino acids of
Trk are necessary for the receptor function in terms of
autophosphorylationas well as substrate phosphorylation.
40 4C7 Z
in the range 35-52 kDa,
to that of ET-R, except for
of wild-type and
tell as with Na-
f) and visualized
Receptor interactions withPLC-y,GAP andp85in
To investigate interactions between Trk andPLC'y,we
cotransfected 293 cells withexpression plasmidsfor ET-R
or ET mutants and PLCy. Prior to EGF stimulation,
[35S]methionine-labelled cells were treated with sodium
prevent dephosphorylation by
phosphotyrosine phosphatases (Swarupetal., 1982). As
controls, cells were transfected with thepCMV-1 vectoror
expression constructs for ET-R orPLCy alone,and treated
immunoprecipitated with anti-EGF-R extracellular domain
mAb 108.1 using aquarterof eachlysate.Aquarterof the
same lysate was used in each case forimmunoprecipitation
of PLCOy with polyclonalantiserum CT-PLCy, and half of
each lysate wasimmunoprecipitatedwith aPY mAb 5E2.
The aPY immunoblot of the caEGF-Rimmunoprecipitates
(Figure 5, upper panel)demonstratedclearlythat PLCy was
coprecipitated with andtyrosine-phosphorylated byET-R
(lanes 7 and 8).
immunoblot ofcPY immunoprecipitates (middle panel,lane
7). Receptor phosphorylationwasrequired,since both ET-
KM and ET-ACT had no effect (lanes 9-12), and the basal
tyrosine phosphorylationof ET-R led tovirtualabolition of
PLCOyassociation andphosphorylation(lane 8). Interestingly,
coprecipitation ofPLCy with Trk cytoplasmic sequences was
abrogated by the single amino acid substitution in ET-YF
(Figure 5, lanes 13 and 14). Visualization of 35S-labelled
proteins on the immunoblotby autoradiographyrevealed that
the phosphorylationdifferences were not due tovariability
in theexpressionlevels of eitherPLC'y (Figure 5, lower
panel)or thereceptors (datanot shown).Aslightshift of
theupperbandrepresenting PLCy (lower panel,lane 13)
indicated that a small fraction of the moleculesmayhave
been phosphorylated by ET-YF,which was not detectable
by thecPY antibody. This was confirmed by caPLCy
precipitationand detection withaPY antibody (not shown),
andsuggestedthat additional lowaffinity bindingsitesmay
bepresentin the Trk kinase domain which mediate inter-
action with the substrate sufficientlyto result intyrosine
phosphorylationof the substrate.
7 F-+ -_-I
This wasconfirmed by
d by the EGF-
93 cells were
ng with anti-
oduced by the
nd 4) tyrosine
A.Obermeier et al.
+ _I11 T
Blt. Ab: aPY
Blt. Ab: ciPLCy
Fig. 5. PLC-y tyrosine phosphorylation and coprecipitation. 293 cells
were transfected with pCMV-1 or cDNA constructs for ET-R or
PLC,y as negative controls, or cotransfected with PLC-y and ET-R,
ET-KM, ET-ACT or ET-YF. To monitor expression levels, cells were
metabolically radiolabelled with [35S]methionine. Prior to lysis, cells
were pretreated with Na-vanadate as well as stimulated with EGF
where indicated (+). A quarter of each precleared lysate was
immunoprecipitated with mAb 108.1 (aEGF-R), a quarter with
polyclonal antiserum CT-PLC-y (caPLC-y) and half with mAb 5E2
(caPY). Precipitates were subjected to 6% SDS-PAGE and then
electrophoretically transferred to nitrocellulose. 108.1-precipitated
proteins were immunoblotted with mAb 5E2 (upper panel), and
5E2-precipitated proteins with CT-PLCy (middle panel). An
autoradiograph is shown of CT-PLC-y precipitates (lower panel).
To determine whether GAP is also a substrate for the Trk
tyrosine kinase, we performed an analogous 293 cell
coexpression experiment. Prior to lysis, [35S]methionine-
labelled cells were treated with sodium vanadate, and where
indicated, stimulated with EGF. Lysates were halved and
either immunoprecipitated with anti-EGF-R extracellular
domain mAb 108.1 or the polyclonal antiserum CT-GAP
(aGAP), respectively. Figure 6 shows the ctPY immunoblot
and the autoradiograph of aGAP precipitates. GAP was
tyrosine-phosphorylated by ET-R and to a lesser yet signifi-
cant extent by ET-YF (Figure 6, upper panel, lanes 7, 8,
13 and 14). Weak coprecipitation with GAP was observed
for ET-R (lane 7) but not for ET-YF under our experimental
conditions. Equal expression of GAP and receptors in the
respective lanes (Figure 6, lower panel and data not shown)
rules out the possibility that effects were due to unequal
We next examined the suitability ofp85 as a Trk substrate.
Figure 7 shows the result of a coexpression experiment
performed analogously to that described for PLC&y and GAP.
blot of proteins
polyclonal antiserum CT-p85 (ap85) revealed tyrosine
phosphorylation of p85 by ET-R and ET-YF. In contrast
to PLCy and similar to GAP, phosphorylation of p85 by
ET-YF was not significantly reduced, in comparison to ET-
R. Moreover, comparable amounts of receptors ET-R and
ET-YF were coprecipitated with p85 in addition to other
proteins. It should be noted, however, that endogenous 293
cell p85 apparently associated with and coprecipitated ET-
R (lane 3). Interestingly, p85 tyrosine phosphorylation was
not detected in this lane, which suggests that phosphorylation
Fig. 6. GAP tyrosine phosphorylation and coprecipitation. 293 cells
were transfected with pCMV-1 or cDNA constructs for ET-R or GAP
as negative controls, or cotransfected with GAP and ET-R, ET-KM,
ET-ACT or ET-YF. Prior to lysis, [35S]methionine-labelled cells were
pretreated with Na-vanadate as well as stimulated with EGF where
indicated (+). Precleared lysates were immunoprecipitated with
polyclonal antiserum CT-GAP (cxGAP). Precipitates were subjected to
7.5% SDS-PAGE and electrophoretically transferred to nitrocellulose.
Phosphotyrosine-containing proteins of the CT-GAP precipitates were
detected by immunoblotting with mAb 5E2 (caPY) (upper panel). An
autoradiograph monitoring expression levels is shown for CT-GAP
precipitates (lower panel).
Fig. 7. p85 tyrosine phosphorylation and coprecipitation. 293 cells
were transfected with pCMV-1 or cDNA constructs for ET-R or p85
as negative controls, or cotransfected with p85 and ET-R, ET-KM,
ET-ACT or ET-YF. Prior to lysis, [35S]methionine labelled cells were
pretreated with Na-vanadate as well as stimulated with EGF where
indicated (+). Precleared lysates were immunoprecipitated with
polyclonal antiserum CT-p85 (ap85). Precipitates were subjected to
7.5% SDS-PAGE and electrophoretically transferred to nitrocellulose.
CT-p85 precipitates were immunoblotted with mAb 5E2 (csPY) (upper
panel). An autoradiograph of the CT-p85 precipitates monitoring p85
expression levels is also shown (lower panel).
of this PI3'-K subunit may not be detectable under these
stoichiometric circumstances. Additional phosphotyrosine-
containing proteins that were coprecipitated with p85 are of
unknown identity. Thus, in contrast to PLCy, the Trk C-
terminal-most tyrosine residue appears not to play a major
role in the interaction with p85.
In vitro association of PC12 cell proteins with
To obtain further information about the substrate specificity
of Trk in a more relevant environment, the association of
endogenous proteins from neuron-like PC12 pheochromo-
cytoma cells with ET receptors was examined. For this
purpose, unlabelled lysates of receptor-expressing 293 cells
were mixed with equal amounts of [35S]methionine-labelled
PC12 cell lysates, and the mixtures were immunoprecipitated
with anti-EGF-R mAb 108.1. For comparison, EGF-R was
After separation of proteins by
SDS -PAGE, the gel was processed for autoradiography.
The signals obtained correspond to proteins coprecipitated
with the respective receptors as a consequence of specific
EGF-R and ET-R displayed distinct association patterns,
demonstrating clear differences in substrate affinity of EGF-
R and Trk cytoplasmic domains (Figure 8A). The strong
bands in lanes 2 and 3 could represent endogenous rat EGF-R
from PC12 cells, which associate more efficiently with the
unlabelled human EGF-R than with ET-R, and are not due
to significant cross reaction ofmAb 108.1 with rat EGF-R
because ofthe absence ofthis band in lanes 4-9. The most
striking signal was a strong band migrating at 145 kDa,
which was exclusively associated with stimulated ET-R
(Figure 8A, lane 4). Very faint 145 kDa bands appeared in
the lanes corresponding to unstimulated ET-R and stimulated
EGF-R only after long exposure times.
Analysis ofthe immunoprecipitates by immunoblotting and
probing with the polyclonal antiserum CT-PLCy identified
the 145 kDa band as PLCy (Figure 8B, lane 4) and
demonstrated that the affinity of PLC-y for Trk cytoplasmic
sequences was remarkably higher (> 100 x) than for EGF-
R, which is well known to phosphorylate this substrate with
high stoichiometry. The extraordinarily high affinity ofTrk
for PLCGy is also obvious from the fact that about the same
amount of PLCy was coprecipitated with ET-R from the
reaction mixture as was precipitated by an excess of
polyclonal ctPLC-y antiserum (CT-PLCy) (data not shown).
Interestingly, replacement of the most C-terminally located
tyrosine residue (Y-785) ofTrk with phenylalanine (ET-YF)
completely abrogated this high affinity PLCOy association.
This, together with the results from the receptor/PLCOy
coexpression experiment (see Figure 5), strongly suggested
that Y-785 of Trk is indispensable for high affinity binding
to PLCOy. Receptor phosphorylation was monitored by
reprobing the immunoblot with anti-PY mAb 5E2 (Figure
8B, lower panel). Coomassie staining of proteins in the gels
ruled out the possibility that the effects described above were
due to different receptor expression levels (data not shown).
Inhibition of PLCy binding to ET-R by a synthetic
To investigate further whether Y-785 of Trk is the binding
site for PLC-y, we performed an in vitro association
[35S]methionine-labelled PC12 cell lysate, and added a
pentadecapeptide identical in sequence to the Trk C-terninus
in which the tyrosine residue corresponding to Trk-Y-785
was phosphorylated. The control experiment included a non-
phosphorylated peptide of the same sequence.
significantly reduced the amount ofPLC-y that coprecipitated
with ET-R (Figure 9, lane 4). Substrate-receptor association
with ET-R and
Fig. 8. In vitro association with PC12 cell proteins. Receptor-
expressing 293 cells were treated with EGF (as well as with Na-
vanadate) where indicated (+) and lysed, and the precleared lysates
were mixed with equal aliquots of precleared lysate from
[35S]methionine-labelled PC12 cells. Receptors were
immunoprecipitated with mAb 108.1, the samples halved and
separately electrophoresed on 7.5% SDS. One gel was processed for
autoradiography (A) to visualize coimmunoprecipitated proteins from
labelled PC12 cell lysate, whereas proteins of the other gel were
electrophoretically transferred to nitrocellulose and probed with
polyclonal antiserum CT-PLC-y (aPLC-y) and with mAb 5E2 (caPY)
(B). Each sample contains receptor from a 10 cm dish of subconfluent
293 cells mixed with labelled lysate of a 15 cm dish of subconfluent
PC12 cells. The control lane differs from ET-R (+) lane only in that
293 cell lysate containing autophosphorylated ET-R was left without
radiolabelled PC12 cell lysate. The difference in the signal intensity of
PLC-y bands between control and ET-R (+) lanes demonstrates that
associated PLC-y is mainly derived from PC12 cells.
concentrations, and was completely abolished to undetectable
amounts at 500 nM (Figure 9, lane 10). In contrast, under
the same conditions the nonphosphorylated peptide had no
capacity to inhibit PLCy binding (Figure 9). Comparable
effects, however, were obtained at concentrations four orders
of magnitude higher than that employed for the phospho-
peptide (data not shown). These results demonstrate that the
C-terminal sequence of Trk, containing phosphorylated
Y-785, is the binding site for PLCy.
Spatial and temporal organization of multicellular organisms
requires a tightly controlled balance between cell growthand
A.Obermeier et al.
1 2 3
Fig. 9. Phosphotyrosine peptide inhibition of PLC-y binding to ET-R.
ET-R-expressing 293 cells were treated with EGF (as well as as with
Na-vanadate) where indicated (+) and lysed, and the precleared lysates
were mixed with precleared lysate from PC12 cells. Then different
amounts of phosphopeptide (PY) QALAQAPPVY(PO3H2)LDVLG or
nonphosphorylated peptide (Y) of the same sequence were added. ET-
R was immunoprecipitated with mAb 108.1, the precipitates were
separated by 7.5% SDS-PAGE and electrophoretically transferred to
nitrocellulose. Amounts of coprecipitated PLCy were detected with
cxPLC-y polyclonal antiserum (CT-PLCy). [Lysates from seven 15 cm
dishes of subconfluent PC12 cells as well as lysates from nine 10 cm
dishes of subconfluent, ET-R expressing and stimulated 293 cells were
pooled and equal aliquots of both pools were mixed. The amount of
ET-R from one 10 cm dish of unstimulated 293 cells (lane 2) was
equal to ET-R amounts in all other lanes as judged by Ponceau
S-staining of transferred proteins (not shown)].
differentiation. Cell surface receptors with tyrosine kinase
activity, such as the receptors for EGF, PDGF, insulin and
CSF-1, are crucial regulatory components in these processes.
Recently, a new RTK subfamily has been described which
consists of the closely related Trk, TrkB and TrkC receptors
(reviewed in Barbacid et al., 1991). These structurally
similar tyrosine kinases are functional receptors for the
neurotrophic factors, NGF, brain-derived neurotrophic factor
(BDNF) and neurotrophin-3 (NT-3), respectively (reviewed
in Meakin and Shooter, 1992). All ofthese receptor-ligand
pairs are implicated in neuronal survival and differentiation.
While there is agreement that high affinity receptor binding
is a prerequisite for the induction of neurotrophic biological
effects (Green et al., 1986; Weskamp and Reichardt, 1991;
Meakin and Shooter, 1992), there is presently no consensus
as to the role of the low affinity NGF-R, p75 NGFR, in NGF
Trk/p75LNGFR-receptor complex comprises the high
affinity NGF binding site (Hempstead et al., 1990, 1991;
Ragsdale and Woodgett, 1991). Another hypothesis is that
Trk alone, possibly in the form of a homodimer, is the only
component required for high affinity NGF binding (Klein
et al., 1991; Weskamp and Reichardt, 1991; Ibafiez et al.,
1992; Meakin et al., 1992).
In this report we have investigated the signal generation
capacity of Trk by examining its ability to interact with
cellular substrates. For the activation of the Trk tyrosine
kinase, we replaced the extracellular ligand binding domain
ofTrk with the extracellular domain ofEGF-R. The resulting
chimeric receptor, ET-R, was faithfully synthesized and
transported to the cell surface. Moreover, interaction with
EGF clearly stimulated the tyrosine kinase activity contained
within the Trk cytoplasmic domain (Figure 4). Since a
number of earlier studies had demonstrated that the signalling
potential and substrate specificity of RTKs are determined
exclusively by their cytoplasmic sequences (Ballotti et al.,
1989; Pandiella et al., 1989; Riedel et al., 1989; Lev et al.,
1990; Herbst et al.. 1991), we were able to obtain important
clues regarding the signalling capacity of the wild type (wt)
Trk system. The validity of our findings was substantiated
by experiments with wt Trk, which yielded analogous results
and will be presented elsewhere. These results support the
notion that the Trk kinase domain
neurotrophic factor signalling.
To investigate Trk signalling characteristics, we analyzed
total cell protein tyrosine phosphorylation catalyzed by ET-
R and, for comparsion, that catalyzed by EGF-R in intact
293 cells (Figure 4). Four proteins of currently unknown
identity, p35, p40, p41 and p52, were preferentially
phosphorylated by Trk, in comparison with the EGF-R.
Further comparisons with other RTKs will be necessary to
evaluate the role of these substrates in Trk signal definition.
In addition, 293 cells clearly represent a heterologous system,
and the availability and concentrations of these protein
substrates may differ in Trk-expressing neurons.
PLC&, GAP and the noncatalytic subunit ofPI3'-K, p85,
were the first cellular proteins to be identified as RTK
substrates. As yet, however, none of the more downstream
signalling pathways have been elucidated conclusively, and
the cell responses connected to them are unknown. In spite
of the ubiquitous expression of these signal mediators, it is
possible that the corresponding signalling pathways fulfil
different roles in different cell types. This is the case, for
example, in the Ras system, which triggers mitogenesis in
many cell types and differentiation ofPC12 cells (Bar-Sagi
and Feramisco, 1985; Hagag et al., 1986).
We therefore examined the potential role of PLCy, GAP
and p85 as Trk substrates in a transient expression system.
All three proteins associated with ET-R, as judged by their
coimmunoprecipitation, and were tyrosine-phosphorylated
in a ligand-dependent fashion. Therefore our data indicate
that the NGF-R/Trk is capable ofcoupling to PLCy-, GAP-
and P13'-K-mediated pathways for the transduction of
cellular signals. In the case of PLC-y, our results are
consistent with previous findings of NGF-stimulated PLCy
phosphorylation in PC12 cells (Kim et al., 1991; Vetter
et al., 1991). In addition, a trpE-PLCy fusion protein
containing SH3 and both SH2 domains of PLCy has been
shown to associate with p70trk (Ohmichi et al., 1991a), an
oncogenic version of Trk originally found in a human colon
carcinoma (Martin-Zanca et al., 1986), and with wild type
Trk (Ohmichi et al., 1991b). Similarly, a trpE-GAP fusion
protein containing SH3 and both SH2 domains ofGAP was
found to associate weakly with p7Ot'rk, but no association
with Trk could be detected (Ohmichi et al., 1991a,b).
Recently, Li et al. (1992) demonstrated that NGF causes an
increase in GAP activity towards Ras in PC12 cells. The
increase in GAP activity was even stronger in Trk-
overexpressing PC12 cells, suggesting a connection between
Trk and GAP. Because tyrosine phosphorylation of GAP
and association with activated Trk could not be demonstrated
(Li et al., 1992), these authors proposed serine/threonine
phosphorylation of GAP as an alternative mechanism for
GAP activation. We have shown here that Trk tyrosine
kinase activation leads to tyrosine phosphorylation ofGAP,
which provides a link between NGF stimulation of the Trk
signalling pathway and initiation of a signalling cascade
which involves GAP and Ras.
Carter and Downes (1992) recently demonstrated NGF
activation of P13'-K in PC12 cells. However, no evidence
was obtained for direct association ofthis enzymatic activity
with the Trk receptor. Our findings that p85 associates tightly
is sufficient for
Trk - substrate interaction
with ET-R and is phosphorylated in an EGF-dependent
fashion strongly suggest a direct interaction of P13'-K with
the Trk cytoplasmic domain, and thus regulation of its
enzymatic activity by NGF. These observations are in
accordance with the recently reported findings of Solthoff
et al. (1992).
A great deal of progress has been made in terms of the
identification of RTK interaction
molecules involved in signalling pathway initiation or
regulation. Surprisingly, the factors that have been identified
to date that bind to RTK cytoplasmic sequences with high
affinity all contain src-homologous (SH2) regions. These
SH2 sequences include structural determinants that recognize
phosphorylated tyrosines, which are flanked by sequences
that apparently define both the specificity and affinity ofthe
interaction. Interestingly, the better characterized sites for
RTK-substrate interactions are all located within regions
that were suggested early on to be involved in RTK-specific
activities due to their hydrophilic nature and divergent
sequence characteristics (Yarden and Ullrich, 1988). This
includes the kinase-insertion region of subclass III RTKs,
which in the case of PDGF-R and pl45c-k contain major
binding sites forGAP and p85 (Fantl et al., 1992; Kashishian
et al., 1992), and the C-tail regions of EGF-R, PDGF-R
and FGF-R, which contain phosphotyrosine residues that are
recognized by PLC'y SH2 domain structures (Mohammadi
et al., 1991, 1992; Peters et al., 1992; Rotin et al., 1992;
receptor-SH2 domain interactions and the current lack of
evidence for RTK-specific substrates, further underscores
the question ofhow RTK-specific and signalling parameters
Our experiments and those of others indicate that Trk-
mediated signals involve factors such asPLC7y,GAP and
p85, which are also utilized by other receptors with different
biological roles. Comparative analysis of proteins from PC12
cells that associate with Trk and EGF-R cytoplasmic
sequences revealed clear differences, which may reflect the
distinct roles these receptors play in these cells. Ligand-
induced association with ET-R was observed most strikingly
for PLC-y and with lower intensity for polypeptides of 85,
91, 104 and 114 kDa, while the EGF-R association pattern
included bands at 94, 97, 101, 107 and 111 kDa (Figure
8A). In striking contrast, strong association of EGF-R with
PLC'y, under identical conditions, was not detected even
though PLC'y clearly serves as a substrate for this RTK
(Margolis et al., 1989). This indicates an affinity difference
of > 100x between these two RTKs for PLCy.
To assess the role ofthe Trk C-terminal tail region, which
is unusually short (15 amino acids) and contains a single
tyrosine residue at position 785, we examined mutants
bearing either a deletion of the entire sequence, ET-ACT,
or a replacement of the tyrosine with phenylalanine, ET-
YF. Surprisingly, deletion of 15 amino acids in ET-ACT
resulted in complete loss of the kinase function. This was
similar to recently reported observations with a PDGF-R
deletion mutant of 115 C-terminal residues (Seedorfet al.,
1992), but different from EGF-R mutants, which preserve
their kinase activity even after a loss of over 200 C-terminal
amino acids. These findings suggest conformation differences
in the cytoplasmic domains of different RTK subclasses and
a major role of Trk C-terminal sequences in the stabilization
of a functional three-dimensional structure.
sites with substrate
Especially informative was the ET-YF mutant. Although
293 cell overexpression experiments
revealed that its ability to coprecipitate with PLCy had been
lost. The PC12 cell protein association (Figure 8A) and
confirmed this finding and further emphasized the central
role of the PY-785 determinant in Trk-PLC-y interaction.
Interestingly, the loss of high affinity binding in ET-YF did
not result in complete abrogation ofPLCy phosphorylation,
suggesting the existence of other low affinity binding sites
in the Trk cytoplasmic domain.
At this point, we cannot explain the basis for the
remarkable affinity difference for PLCOy between Trk and
EGF-R, nor the contribution of flanking region sequence
determinants. The Trk Y-785 flanking sequences and
recently identified PLCOy binding sites in the PDGF-R,3
(Ronnstrand et al., 1992) do not confirm V/LXXXXEYL/I
as a consensus sequence, as suggested previously on the basis
of the EGF-R and FGF-R binding
(Mohammadi et al., 1992; Rotin et al., 1992). However,
parameters and structural determinants defining receptor
affinities for a common set of substrates, in conjunction with
possible cell type- and receptor-specific signalling pathways,
form the basis of RTK-characteristic signals. Further work
will be necessary to investigate the biological significance
ofNGF-stimulated PLCy phosphorylation and its role in the
processes of nerve cell differentiation and survival that are
mediated by Trk.
its capacityfor auto- and substrate
sites for PLC-y
Materials and methods
Construction of chimeric ET-R and ET mutants
Plasmid CVNHERc (Riedel et al., 1988), containing the full-length cDNA
of human EGF-R, and pLTRgagactrk (generously given by N.Hynes),
containing the cDNA of a human actin-trk fusion oncogene, were used
as cDNA sources for restriction fragments and as polymerase chain reaction
(PCR) templates for the fusion of the EGF-R extracellular domain to the
transmembrane and intracellular sequences of Trk, thereby generating ET-R.
The coding sequence for the EGF-R extracellular domain was primarily
constituted from a 1680 bp XbaI-ApaI restriction fragment. The remaining
418 bp were generated by PCR technology, ligated to the larger fragment
via the Apal site, and simultaneously cloned into an XbaI-EcoRV-linearized
BluescriptIl KS(+) vector (Stratagene).
The coding sequence for the Trk intracellular domain was mainly
constituted from a 1100 bp NarI-EcoRI restriction fragment. The remaining
364 bp, also containing the transmembrane coding region, were generated
by PCR technology, ligated to the larger fragment via the NarI site, and
simultaneously cloned into a SmaI/EcoRI-linearized pT7T3 18U vector
The complete cDNA for the EGF-R extracellular domain and the Trk
transmembrane and intracellular domains were recovered from the Bluescript
vector by digestion with XbaI and PvuI, and from the pT7T3 vector by
digestion with ScaI and EcoRI, respectively, ligated and cloned into a
XbaI/EcoRI-linearized BluescriptII KS(+) vector, creating chimeric ET-R
cDNA. At the E-T fusion point, the codon for the last EGF-R extracellular
amino acid, serine,
transmembrane amino acid, threonine.
For transient expression studies, the chimeric ET-R cDNA was subcloned
as an XbaI-HindIII fragment into the polylinker of the cytomegalovirus
promoter based pCMV-l expression vector (Eaton et al., 1986). The PCR-
derived sequences, including the E-T fusion point, were subsequently
In order to construct various mutants, the NarI-EcoRI Trk cDNA
restriction fragment was finally cloned into a M13mpl8 vector (Boehringer
In vitro mutagenesis was performed accordingtoTayloret al. (1985)
using the 'oligonucleotide-directed in vitro mutagenesis systemversion 2'
is directly joined by the codon for the first Trk
A.Obermeier et al.
(Amersham) with the following 18mer oligonucleotides: 5'-pTGGCTG-
TCATGGCACTGA-3' for ET-KM, a kinase-negative receptor mutant,
which carries a methionine instead of the conserved lysine in the ATP-binding
site (K-538 of Trk); 5'-pCCCGGCTGTAAGCCCTGG-3' for ET-ACT,
a receptor mutant that lacks the 15 C-terminal amino acids defined as the
CCTGGATG-3' for ET-YF, a receptor mutant in which the only C-
terminally located tyrosine residue (Y-785 of Trk) has been replaced by
After verifying the mutations by sequencing, suitable fragments, each
carrying one of the mutations, were cloned into the ET-R background.
Other expression plasmids
In addition to the chimeric ET receptors, human EGF-R and human Trk
were also expressed in 293 cells in some experiments. The genetic constructs
used in these cases were pCMV-1-HERc and pCMV-1-trk, containing the
full-length EGF-R cDNA and Trk, respectively. The full-length Trk cDNA
the Nar -EcoRI
pLTRgagactrk and an 859 bp BamH-NarI plus an 810 bp BamHI-EcoRI
restriction fragment, both from pLM6 (Martin-Zanca et al., 1989), kindly
provided by Mariano Barbacid. The full-length Trk cDNA was integrated
in an EcoRI-linearized pCMV-1 expression vector. For receptor/substrate
coexpression experiments, cDNA sequences coding for PLCy, GAP and
p85 were subcloned into pCMV-1.
Cell culture, transient expression and in vitro association
Human embryonic kidney fibroblasts (293; ATCC CRL 1573) were grown
in DMEM, 4500 mg glucose/liter (Gibco) containing 9% fetal calf serum
(FCS; Gibco), and 2 mM L-glutamine. 30-35 h prior to transfection, 1.5,
3.5 and 10 x 105 cells were seeded into a well of a six-well dish, into 6 cm
dishes and 10 cm dishes, respectively. Transfections were then carried out
using the calcium phosphate coprecipitation technique according to the
protocol of Chen and Okayama (1987) with a total of 4, 8 and 16jigCsCl
gradient-purified plasmid DNA per well or dish, respectively. 12-18 h
after addition of precipitates, cells were washed once with DMEM, and
DMEM containing 0.5% FCS was then added. To quantify expression levels
directly, MEM with Earle's salt, without L-methionine (Gibco), containing
4500 mg glucose/l, 2 mM L-glutamine and 0.5-1.0% FCS, was used
instead ofDMEM, and cells were metabolically radiolabelled overnight with
40,Ci [35S]methionine/ml (1000 Ci/mmol; Amersham).
Cells were stimulated with 150 ng/ml EGF (or NGF) 10 min prior to
lysis (where mentioned, sodium orthovanadate (Na-vanadate) was added
to a final concentration of 1 mM 1 h before stimulation). After stimulation,
cells were lysed on ice with 0.3-1.5 ml (depending on the size of the well
or dish) 'lysis buffer' (50 mM HEPES pH 7.5, 150 mM NaCl, 1.5 mM
phenylmethylsulfonyl fluoride, 250AMp-nitrophenylphosphate, 100ItM
ATP, 10itg/mlaprotinin and 10 pg/mil leupeptin). [The inclusion of ATP
in the Lysis buffer was compensated
pyrophosphate, and had no effect on the results obtained.] 10 min later,
lysates were transferred to microcentrifuge tubes and precleared by centri-
fugation at 12 500 g for 20 min at 4'C.
For immunoprecipitations, 20/1of protein A-Sepharose (Pharmacia;
prewashed in 20 mM HEPES pH 7.5, 0.1 mg/4l) and the appropriate
antiserum were added to the cleared lysate and incubated for 3 h on a rotor
at 4°C. Precipitates were washed three times with 1.5 ml 'Washing buffer'
(50 mM HEPES pH 7.5, 150 mM NaCl, 1.5 mM MgCl2,
10 mM sodium pyrophosphate, 10% glycerol and 0.1% Triton X-100). SDS
sample buffer was added, and the samples were boiled for 5 min before
loading on SDS-polyacrylamide gels.
For analysis of total cellular proteins, 45 1l of sample buffer was added
directly to 90 1l of precleared lysates (from 5 x
was boiled for 15 min.
After separation by SDS-PAGE, proteins were electrophoretically
transferred to nitrocellulose filters. For immunoblot analysis, filters were
preincubated 1 h with 5% milk powder solution in TBST (20 mM Tris pH
7.5, 150 mM NaCl, 0.02% Tween 20), washed with TBST, incubated
overnight with antibody, washed three times and incubated for
horseradish peroxidase-conjugated goat anti-mouse or goat anti-rabbit
antibody. Immunoblots were developed using the ECL system (Amersham;
Thorpe et al., 1985). In order to reprobe proteins with another antibody,
filters were incubated for 2 h in Strip buffer at 50'C (62.5 mM Tris, pH
6.8, 100 mM ,B-mercaptoethanol, 2% SDS).
For in vitro association experiments, receptor-expressing 293 cells were
treated with EGF and Na-vanadate (where indicated) and lysed, and the
precleared lysates mixed with equal aliquots of precleared lysate from
1 mM EGTA, 10 mM sodium pryophosphate, 10% glycerol, 1%
excess of sodium
1 mM EGTA,
I04 cells) and the mixture
1 h with
immunoprecipitated with anti-receptor antibody as described above, except
that incubation on the rotor was extended to 5 h. Proteins were fractionated
by 7.5% SDS-PAGE, and either transferred
immunoblotting or processed for autoradiography: fixed for 1 h in 40%
methanol/10% acetic acid, stained for 20 min in 0.025% Coomassie G 250,
destained in 10% acetic acid, dried and exposed to film (Kodak X-Omat).
PC12 cells were grown in DMEM with 4500 mg glucose/liter containing
2 mML-glutamine, 9% FCS and 4.5%
The antibodies recognizing the human EGF-R extracellular domain (108.1),
phosphotyrosine (5E2), and the C-termini of PLCy (CT-PLC-y), GAP (CT-
GAP) and p85 (CT-p85) have been described previously (Herbst et al.,
1992). Additionally, ATC, a polyclonal rabbit antibody against a peptide
corresponding to the last 15 amino acids of the Trk C-terminus (see below),
corresponding in sequence to the C-terminus of Trk (QALAQAPPV-
YLDVLG) were synthesized as described by Kitas et al. (1989) and analyzed
by ion spray mass spectrometry.
We are grateful to Reiner Lammers for the preparation of the CT-PLCy,
CT-GAP and CT-p85 polyclonal antisera and expression plasmids for PLCOy,
GAP and p85, to Frank McCormick for his generosity in giving the GAP
cDNA, and to Yves-Alain Barde and Georg Dechant for providing the PC12
cell line and horse serum. We also thank Ronald Herbst, Bahija Jallal-Herbst,
Klaus Seedorf and Ralph A.Bradshaw for helpful discussions, and Jeanne
Arch for expert help in the preparation of this manuscript. This work was
supported by a grant from SUGEN, Inc.
Ballotti,R., Lammers,R., Scimeca-J -C., Dull,T., Schlessinger,J., Ullrich,A.
and Van Obberghen,E (1989) EMBO J., 8, 3303-3309.
Barbacid,M., Lamballe,F., Pulido,D. and Klein,R. (1991) Biochim. Biophys.
Acta, 1072, 115-127.
Bar-Sagi,D. and Feramisco,J.R. (1985) Cell, 42, 841-848.
Zilberstein,A., Schlessinger,J. and Jaye,M. (1990) Mol. Cell. Biol., 10,
Candey,L.C. Auger,K.R., Carpenter,C., Duckworth,B., Graziani,A.,
Kapeller,R. and Soltoff,S. (1991) Cell, 64, 281-302.
Carter,N.A. and Downes,P.C. (1992) J. Biol. Chem., 267, 14563-14567.
Chen,C. and Okayama,H. (1987) Mo. Cell. Biol., 7, 2745-2752.
Downing,J.R., Margolis,B.L., Zilberstein,A., Ashmun,R.A., Ullrich,A.,
Sherr,C.J. and Schlessinger,J. (1989) EMBO J.,8, 3345-3350.
Eaton,D.L., Wood,W.I., Eaton,D., Hass,P.E., Hollingshead,P., Wion,K.,
Mather,J., Lawn,R.M., Vehar,G.H. and Gorman,C. (1986) Biochemistry,
Fantl,W.J., Escobedo,J.A., Martin,G.A., Turck,C.W., del Rosario,M.,
McCormick,F. and Williams,L.T. (1992) Cell, 69, 413-423.
Glenney,J.R., Jr (1992) Biochim. Biophys. Acta, 1134, 113-127.
Green,S.H., Rydel,R.E., Connolly,J.L. and Green,L.A. (1986) J Cell. Biol.,
Hagag,N., Halegoua,S. and Viola,M. (1986) Nature, 319, 680-682.
Hempstead,B.L., Patil,N., Thiel,B. and Chao,M.V. (1990) J. Bio. Chem.,
Chao,M.V. (1991) Nature, 350, 678-683.
Herbst,R., Lammers,R., Schlessinger,J. and Ullrich,A. (1991) J. Biol.
Chem., 266, 19908-19916.
Herbst,R., Shearman,M.S., Obermeier,A., Schlessinger,J. and Ullrich,A.
(1992) J. Biol. Chem., 267, 13210-13216.
Honegger,A.M., Dull,T.J., Felder,S., Van Obberghen,E., Bellot,F.,
Szapary,D., Schmidt,A., Ullrich,A. and Schlessinger,J, (1987) Cell, 51,
Honegger,A.M., Kris,R.M., Ullrich,A. and Schlessinger,J. (1989) Proc.
Natl. Acad. Sci. USA, 86, 925-929.
Ibdiiez,C.F., Ebendal,T., Barbany,G., Murray-Rust,J., Blundell,T.L. and
Persson,H. (1992) Cell, 69, 329-341.
Kaplan,D.R., Parada,L.F. and
Trk- substrate interaction
Kaplan,D.R., Morrison,D.K., Wong,G., McCormick,F. and Williams,L.T.
(1990) Cell, 61, 125-133.
Kaplan,D.R., Martin-Zanca,D. and Parada,L.F. (1991a) Nature, 350,
Parada,L.F. (1991b) Science, 252, 554-558.
Kashishian,A., Kazlauskas,A. and Cooper,J.A. (1992) EMBO J., 11,
Kazlauskas,A., Ellis,C., Pawson,T. and Cooper,J.A. (1990) Science, 247,
Kim,U.-H., Fink,D.,Jr, Kim,H.S., Park,D.J., Contreras,M.L., Guroff,G.
and Rhee,S.G. (1991) J. Biol. Chem., 266, 1359-1362.
Kitas,E.A., Perich,J.W., Wade,J.O., Johns,R.B. and Pregear,G.W. (1989)
Tetrahedron Lett., 30, 6229-6232.
Klein,R., Shuqian,J., Nanduri,V., O'Rourke,E. and Barbacid,M. (1991)
Cell, 65, 189-197.
Koch,A.C., Anderson,D., Moran,M.F., Ellis,C. and Pawson,T. (1991)
Science, 252, 668-674.
Lammers,R., Gray,A., Schlessinger,J. and Ullrich,A. (1989) EMBO J.,
Lammers,R., Van Obberghen,E., Ballott,R., Schlessinger,J. and Ullrich,A.
(1990) J. Biol. Chem., 265, 16886-16890.
Lev,S., Yarden,Y. and Givol,D. (1990) Mol. Cell. Biol., 10, 6064-6068.
Li,B.-Q., Kaplan,D., Kung-H.-F. and Kamata,T. (1992) Science, 256,
Margolis,B., Rhee,S.G., Felder,S., Mervic,M., Lyall,R., Levitzki,A.,
Margolis,B., Zilberstein,A., Franks,C., Felder,S., Kremer,S., Ullrich,A.,
Rhee,S.G., Skorecki,K. and Schlessinger,J.
Martin-Zanca,D., Hughes,S.H. and Barbacid,M. (1986) Nature, 319,
Martin-Zanca,D., Oskam,R., Mitra,G., Copeland,T. and Barbacid,M.
(1989) Mol. Cell. Biol., 9, 24-33.
Meakin,S.O. and Shooter,E.M. (1992) Trends Neurosci., 15, 323-331.
Meakin,S.O., Suter,U., Drinkwater,C.C., Welcher,A.A. and Shooter,E.M.
(1992) Proc. Natl. Acad. Sci. USA, 89, 2374-2378.
Meisenhelder,J., Suh,P.-G., Rhee,S.G. and Hunter,T. (1989) Cell, 57,
Mohammadi,M., Honegger,A.M., Rotin,D., Fischer,R., Bellot,F., Li,W.,
Dionne,C.A., Jaye,M., Rubinstein,M. and Schlessinger,J. (1991) Mol.
Cell. Biol., 11, 5068-5078.
Mohammadi,M., Dionne,C.A., Li,W., Li,N., Spivak,T., Honegger,A.M.,
Jaye,M. and Schlessinger,J. (1992) Nature, 358, 681-684.
Molloy,C.J., Bottaro,D.P., Fleming,T.P., Marshall,M.S., Gibbs,J.B. and
Aaronson,S.A. (1989) Nature, 342, 711-714.
Moran,M.F., Koch,A.C., Anderson,D., Ellis,C., England,L., Martin,S.G.
and Pawson,T. (1990) Proc. Natl. Acad. Sci. USA, 87, 8622-8626.
Nishibe,S., Wahl,M.I., Rhee,S.G. and Carpenter,G. (1989) J. Biol. Chem.,
Nishibe,S., Wahl,M.I., Wedegaertner,P.B., Kim,J.J., Rhee,S.G. and
Carpenter,G. (1990) Proc. Natl. Acad. Sci. USA, 87, 424-428.
Ohmichi,M., Decker,S.J., Pang,L. and Saltiel,A.R. (1991a) J. Biol. Chem.,
Ohmichi,M., Decker,S.J., Pang,L. and Saltiel,A.R. (1991b) Biochem.
Biophys. Res. Commun., 179, 217-223.
Pandiella,A., Lehvaslaiho,H., Magni,M., Alitalo,K. and Meldolesi,J. (1989)
Oncogene, 4, 1299- 1305.
Peters,K.G., Marie,J., Wilson,E., Ives,H.E., Escobedo,J., Del Rosario,M.,
Mirda,D. and Wilfiams,L.T. (1992) Nature, 358, 678-681.
Ragsdale,C. and Woodgett,J. (1991) Nature, 350, 660-661.
Riedel,H., Dull,T.J., Schlessinger,J. and Ullrich,A. (1986) Nature, 324,
Riedel,H., Dull,T.J., Honegger,A.M., Schlessinger,J. and Ullrich,A. (1989)
EMBO J., 8, 2943 -2954.
Ronnstrand,L., Mori,S., Arridsson,A.-K., Eriksson,A., Wernstedt,C.,
Hellman,U., Claesson-Welsh,L. and Heldin,C.-H. (1992) EMBOJ., 11,
Rotin,D., Margolis,B., Mohammadi,M., Daly,R.J., Daum,G., Li,N.,
Fischer,E.H., Burgess,W.H. and Ullrich,A. and Schlessinger,J. (1992)
EMBO J., 11, 559-567.
Schlessinger,J. (1988) Trends Biochem. Sci., 13, 443-447.
Schlessinger,J. and Ullrich,A. (1992) Neuron,9, 1-20.
Seedorf,K., Felder,S., Millauer,B. Schlessinger,J. and Ullrich,A. (1991)
J. Biol. Chem., 266, 12424-12431.
Seedorf,K., Millauer,B., Kostka,G., Schlessinger,J. andUllrich,A. (1992)
Mol. Cell. Biol., 12, 4347-4356.
(1990) Science, 248,
Soltoff,S.P., Rabin,S.L., Cantley,L.C. and Kaplan,D.R. (1992) J. Biol.
Chem., 257, 17472-17477.
Swarup,G., Speeg,V.K.,Jr, Cohen,S. and Garbers,D.L. (1982) J. Biol.
Chem., 257, 7298-7301.
Taylor,J.W., Ott,J. and Eckstein,F. (1985) Nucleic Acids Res., 13,
Thorpe,G.H.G., Kricka,L.J., Moseley,S.B. and Whitehead,T.P. (1985)
Clin. Chem., 31, 1335-1341.
Ullrich,A. and Schlessinger,J. (1990) Cell, 61, 203-212.
Vetter,M.L., Martin-Zanca,D., Parada,L.F., Bishop,M.J. and Kaplan,D.R.
(1991) Proc. Natl. Acad. Sci. USA, 88, 5650-5654.
Wahl,M.I., Olashaw,N,E., Nishibe,S., Rhee,S.G., Pledger,W.J. and
Carpenter,G. (1989) Mol. Cell. Biol., 9, 2934-2943.
Weskamp,G. and Reichardt,L.F. (1991) Neuron, 6, 649-663.
Yarden,Y. and Ullrich,A. (1988) Annu. Rev. Biochem., 57, 443-478.
Received on October 2, 1992; revised on December 2, 1992