Current Biology, Vol. 15, 1861–1866, October 25, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.cub.2005.09.012
PDGFR?? Signaling Is Regulated
through the Primary Cilium in Fibroblasts
Linda Schneider,1Christian A. Clement,1
Stefan C. Teilmann,2Gregory J. Pazour,3
Else K. Hoffmann,1Peter Satir,4
and Søren T. Christensen1,*
1Department of Biochemistry
Institute for Molecular Biology and Physiology
University of Copenhagen
The August Krogh Building
DK-2100 Copenhagen Ø
2Laboratory of Reproductive Biology
DK-2100 Copenhagen Ø
3University of Massachusetts Medical School
Worcester, Massachusetts 01655
4Department of Anatomy and Structural Biology
Albert Einstein College of Medicine of Yeshiva
Bronx, New York 10461
Recent findings show that cilia are sensory organ-
elles that display specific receptors and ion channels,
which transmit signals from the extracellular environ-
ment via the cilium to the cell to control tissue ho-
meostasis and function [1–6]. Agenesis of primary
cilia or mislocation of ciliary signal components af-
fects human pathologies, such as polycystic kidney
disease  and disorders associated with Bardet-
Biedl syndrome . Primary cilia are essential for
hedgehog ligand-induced signaling cascade regulat-
ing growth and patterning [9,10]. Here, we show that
the primary cilium in fibroblasts  plays a critical
role in growth control via platelet-derived growth fac-
tor receptor ? (PDGFR?), which localizes to the pri-
mary cilium during growth arrest in NIH3T3 cells and
primary cultures of mouse embryonic fibroblasts. Li-
gand-dependent activation of PDGFR?? is followed
by activation of Akt and the Mek1/2-Erk1/2 pathways,
with Mek1/2 being phosphorylated within the cilium
and at the basal body. Fibroblasts derived from
Tg737orpkmutants fail to form normal cilia and to
upregulate the level of PDGFR?; PDGF-AA fails to ac-
tivate PDGFR?? and the Mek1/2-Erk1/2 pathway. Sig-
naling through PDGFR?, which localizes to the
plasma membrane, is maintained at comparable
levels in wild-type and mutant cells. We propose that
ciliary PDGFR?? signaling is linked to tissue homeo-
stasis and to mitogenic signaling pathways.
Results and Discussion
Formation of Primary Cilia in NIH3T3 Fibroblasts
Primary cilia are formed in nearly all growth-arrested
and differentiated cell types in vertebrates [12,13]. Typi-
cally, a single cilium projects from each cell, emerging
from the mother centriole that has become a basal
body. To study ciliary function, NIH3T3 fibroblasts were
grown in DMEM containing 10% fetal calf serum for 2
days to w80% confluency, followed by serum starva-
tion for 24 hr, when the cells enter G0. Prior to star-
vation, cells show a prominent cytoskeleton of acet-
ylated microtubules but no primary cilia (Figures 1A and
1E). After 6 hr serum starvation, <5% of the cells had
primary cilia, the majority of which had lengths of 1–2
?m (Figure 1B). By 12 hr about 40% had cilia, with
lengths up to 6 ?m (Figure 1C). By 24 hr >90% of the
cells possessed cilia, all approximately 6 ?m length;
each axoneme grew above a ciliary basal body (Figures
1D and 1F). These cells now lacked the arrayed cyto-
skeleton containing acetylated microtubules (Figure 1D).
PDGFR? Is Upregulated and Localized to Primary
Cilia during Growth Arrest
Using a gene trap technique to identify genes that are
transcriptionally active during growth arrest of NIH3T3
fibroblasts, Lih et al.  identified platelet-derived
growth factor receptor α (PDGFRα) as a growth arrest-
specific (GAS) gene whose mRNA was preferentially
expressed in serum-deprived cells. The intimate con-
nection between cilia formation and growth arrest led
us to ask if PDGFRα localized to cilia. Using rabbit
polyclonal antibodies, we found that PDGFRα localized
along the primary cilia of NIH3T3 cells (Figures 2A–2C,
upper frames). In contrast, PDGFRβ was primarily dis-
tributed as clusters along the cell surface (Figures 2A–
2C, lower frames). PDGFRβ localization is a control for
how receptor localization appears when distributed in
the plasma membrane; therefore, PDGFRα seems to lo-
calize predominantly in the cilium and not the plasma
membrane. Preincubation of the PDGFRα antibody
with an antibody-specific blocking peptide (P) reduced
the ciliary PDGFRα immunofluorescence. Substitution
of the primary antibody with preimmune rabbit serum
gave no ciliary signal (see Figure S1 in the Supplemental
Data online with this article at http://www.currentbiology.
com/cgi/content/full/15/20/1861/DC1/). In Western blot
analysis, anti-PDGFRα reacted uniquely with the recep-
tor (Figure 2D). Two protein bands at w170 kDa are
seen in the immunoblot, the upper band being the ma-
ture and fully glycosylated form and the lower band the
immature and only partly glycosylated form of the re-
ceptor . Goat polyclonal antibody to PDGFRα also
localized to the primary cilium, whereas neither anti-
EGFR nor anti-IRα became associated with the cilium
Next we studied the time course for appearance of
PDGFRα along the elongating cilium. After 12 hr serum
starvation, PDGFRα was localized around the base and
the lower part of the cilium in about 70% of the cilia
(Figure 2F). By 24 hr serum starvation, the receptor was
distributed along the entire length of the cilium in about
80% of the cilia (Figure 2G). Immunoblots with anti-
PDGFRα showed that the receptor was present in
growing interphase cells (0 hr serum starvation), but
Figure 1. Time Course for Growth of the Primary Cilium in NIH3T3 Fibroblasts
Cilia and microtubules are localized with anti-acetylated α-tubulin (tb, green; arrows). Immunofluorescence microscopy analysis of nonar-
rested, interphase cells (A and E) and after serum starvation for 6 (B), 12 (C), and 24 hr (D and F). Centrioles are localized with anti-γ-tubulin
(γ-Tb, magenta, asterisks). In (F), the mother centriole is marked with a blue asterisk and the daughter centriole with a white asterisk. Nuclei
(nu) are stained with DAPI (blue). Scale bars: 10 ?m.
that the level of receptor increased after serum starva-
tion (Figure 2H). The mature receptor form was upregu-
lated w7-fold from 0 to 24 hr serum starvation (Figure
2I), consistent with specific placement of much of the
upregulated fully glycosylated, mature PDGFRα in the
ciliary membrane. In contrast, the protein level of
PDGFRβ was high in interphase cells and weakly in-
creased upon serum starvation (Figures 2H and 2I).
Activation of PDGFR? in the Primary
Cilium of NIH3T3 Cells
The ligand PDGF-AA binds and activates the homodi-
mer of PDGFRα (PDGFRαα) but not homodimers of
PDGFRβ (PDGFRββ) or heterodimers of PDGFRαβ,
whereras PDGF-BB binds and activates all three recep-
tor dimers . As a control for ligand specificity, acti-
vation of PDGFRα and PDGFRβ was analyzed by im-
munoprecipitation of the receptors in cells stimulated
with either PDGF-AA or PDGF-BB for 3 min in nonar-
rested interphase cells (growth) and in 24 hr serum-
starved, growth-arrested cells (arrest) followed by
Western blotting with anti-phosphotyrosine (Figures 3A
and 3B). PDGFRβ was not activated by phosphorylation
upon stimulation with PDGF-AA, whereas both receptor
forms were phosphorylated by PDGF-BB. Further,
PDGFRβ was activated in both growing and growth-
arrested cells, whereas PDGFRα was heavily phos-
phorylated only during growth arrest because the level
of immunoprecipitated PDGFRα was high only during
growth arrest (Figure 2H). We examined tyrosine phos-
phorylation of the immunoprecipitated receptor after 0,
12, and 24 hr of serum starvation and subsequent
PDGF-AA stimulation. Immunoprecipitated PDGFRα
was loaded on SDS-PAGE at equal concentrations and
corresponding Western blots monitored. After 3 min
of PDGF-AA stimulation, tyrosine phosphorylation of
PDGFRα increased about 17-fold after 24 hr serum
starvation compared to cells in interphase growth (Fig-
ure 3C). These results support the conclusion that the
activation response to PDGF-AA increased in corre-
spondence with development of the primary cilium.
Prior to ligand addition, there was little phosphotyro-
sine labeling along the cilium. When PDGF-AA was
added to the cells after 24 hr serum starvation, a
marked increase in tyrosine phosphorylation within cilia
was detected about 3 min after ligand addition,
whereas little new immunofluorescence was detected
in the cytoplasm (Figure 3D). Among the proteins that
became tyrosine phosphorylated after PDGF-AA addi-
tion was the glycosylated PDGFRα itself (Figure 3E),
which we surmise to be primarily PDGFRαα homodimer
activation in the ciliary membrane. Substantial PDGFRα
phosphorylation seen by 3 min, analyzed by immuno-
precipitation of receptor followed by Western blotting
with anti-phosphotyrosine, persisted for at least 10 min;
the immature receptor was unaffected. Addition of
PDGF-BB led to increased tyrosine phosphorylation of
proteins both in the cilia and around the plasma mem-
brane (Figure 3D), suggesting that PDGF-BB may acti-
vate PDGFRββ, residual PDGFRαα homodimers, and
PDGFRαβ heterodimers in the plasma membrane as
well as PDGFRαα homodimers in the ciliary membrane.
Two sites of tyrosine phosphorylation of PDGFRα
(Y720and Y742) can be monitored by phosphospecific
antibodies. Y720acts as a docking site for adaptor pro-
teins with SH2 domains, including Shp . In response
to PDGF in 3T3 fibroblasts, Shp becomes tyrosine
phosphorylated and contributes to Ras and normal Erk
activation . Y742is a docking site for p85, which is
required for the phosphoinositide 3-kinase (PI3K)/Akt
signaling axis . After PDGFRαα activation, both
Mek1/2-Erk1/2 and Akt pathways were activated by
phosphorylation beginning at 3 min of PDGF-AA stimu-
lation (Figure 3F). The increased level of Mek1/2 (by
PDGFRα Signaling in the Primary Cilium
Figure 2. PDGFRα Is Upregulated and Localized to Primary Cilia in Growth-Arrested NIH3T3 Fibroblasts
(A) Localization of rabbit anti-PDGFRα (green) and rabbit anti-PDGFRβ (green) in 24 hr serum-starved cells. Arrows indicate the primary cilia
detected by anti-acetylated α-tubulin (tb, red). Scale bar: 50 ?m.
(B and C) High-magnification images of rabbit anti-PDGFRα (red, top) and rabbit anti-PDGFRβ (red, bottom) localization to 24 hr serum-
starved cells. Bold arrows indicate the cilium (tb, green) and open arrows the cell surface. Scale bar: 10 ?m.
(D) Rabbit anti-PDGFRα crossreactivity to the receptor in whole-cell lysate of 24 hr serum-starved cells in the absence and in the presence
of antibody blocking peptide, P. Molecular mass marker (kDa) from top to bottom: 200, 116.3, 97.4, 66.3, 55.4, 36.5, and 31.
(E) Ciliary localization (Tb, green, arrows) of rabbit anti-IRα, rabbit anti-EGFR and goat anti-PDGFRα in 24 hr serum-starved cells (red).
PDGFRα targeting (rabbit anti-PDGFRα, red) to the elongating cilium (Tb, green, arrows) in cells serum starved for 12 (F) and 24 hr (G). Nuclei
(nu) are stained with DAPI (blue) or visualized with DIC images.
(H) Level of PDGFRα, PDGFRβ and β-actin (control) upon serum starvation for 0, 6, 12, and 24 hr.
(I) Quantification of PDGFRα and PDGFRβ protein levels upon serum starvation for 0, 6, 12, and 24 hr relative to β-actin and receptor levels
at 0 hr of serum starvation.
Error bars indicate standard errors from three separate experiments.
c-Raf-dependent Mek1/2 phosphorylation on serines
217 and 221 in the activation loop to activate Erk1/2)
preferentially localized along and at the base of the cil-
ium; i.e., the mother centriole (Figures 3G and 3I). No
phospho-Mek1/2 was detected at the cell surface (Fig-
ure 3H and Figure S2). In contrast, PDGF-BB increased
the level of phospho-Mek1/2 in cilia, in the basal body
and partly in the cell cytosol (Figure 3G), supporting the
conclusion that although PDGF-BB acts on receptors
at the cell surface outside the cilium, receptors for
PDGF-AA respond predominently in the cilium. We also
observed ciliary localization of Mek1 phosphorylated
on serine 298 (Figure 3J), produced by p21-activated
protein kinase, PAK, which is a convergence point for
integrating growth factor signaling via the MAPK path-
way . These results support the conclusion that
Raf-dependent activation of Mek1/2 is assisted by
PAK in the cilium and that the primary cilium contains
the signaling machinery from PDGFRαα activation to
Mek1/2 through the Ras-Raf pathway, controlling cell
growth and proliferation.
Ciliary Assembly and Upregulation of PDGFR?
Expression Is Blocked in Tg737orpkMutant Cells
Tg737 encodes the IFT particle protein IFT88/polaris re-
quired for ciliary assembly . No other function of
Tg737 is known, and IFT protein-encoding genes are
found only in organisms that possess cilia. Homozy-
gous Tg737 mutant mice normally die within 2 weeks
of birth with multiple tissue pathologies including cystic
lesions in the pancreas and kidney .
To test the hypothesis that the ciliary-localized
PDGFRα receptor is responsible for the PDGF-AA re-
sponses we observed in NIH3T3 cells more directly, we
compared primary cultures of wt mouse embryonic fibro-
blasts (MEFs) and MEFs derived from Tg737orpkmutant
mice. Wild-type MEFs serum starved for 48 hr devel-
oped primary cilia with lengths of 5–10 ?m; PDGFRα
localized along these cilia (Figure 4A). Mutant MEFs
grown similarly presented no or very short, <1 ?m long,
cilia. PDGFRα localized at the base of these short stubs
(Figure 4B). Importantly, PDGFRα was present at a low
level at the cell surface as in wt cells (Figures 4A and
Figure 3. PDGF-AA Activates PDGFRα in the Primary Cilium of NIH 3T3 Fibroblasts Followed by Mitogenic Signaling through Akt and the
(A and B) Western blot with anti-phosphotyrosine on immunoprecipitated PDGFRα and PDGFRβ in nonarrested, interphase cells (growth)
versus 24 hr serum-starved, growth-arrested cells (arrest) after stimulation with either PDGF-AA or PDGF-BB for 3 min. Molecular mass
markers (kDa) from top to bottom: 200, 150, 120, 100, 85, 70, 60, 50, and 40.
(C) Time course of tyrosine phosphorylation of immunoprecipitated PDGFRα in cells serum starved for 0, 12, and 24 hr and stimulated with
PDGF-AA for 3 min (left) and quantification of the level of tyrosine phosphorylated PDGFRα in cells starved for 24 hr relative to immunoprecipi-
tated PDGFRα and cells starved 0 hr (right). Immunoprecipitated PDGFRα in cells starved for 0 and 12 hr was increased to match the amount
of receptor in cells starved for 24 hr.
(D) Localization of tyrosine phosphorylation (anti-pTyr, red) in the primary cilium (bold arrow; anti-acetylated α-tubulin, tb, green) and at the
plasma membrane (open arrows) before (control) or upon stimulation with PDGF-AA or PDGF-BB for 3 min. Scale bar: 10 ?m.
(E) Time course for tyrosine phosphorylation of immunoprecipitated PDGFRα in cells serum starved for 24 hr and stimulated with PDGF-AA
for 0, 3, and 10 min.
(F) Level of phosphorylation of PDGFRα, Mek1/2, Erk1/2 and Akt upon stimulation with PDGF-AA for 3 and 10 min in cells serum starved for
(G) Shifted overlays of ciliary localization of anti-phospho-Mek1/2 (green) and anti-tb (red, arrows) in cells stimulated with either PDGF-AA or
PDGF-BB for 10 min.
(H) Merged DIC image of a single cell with anti-phospho-Mek1/2 (green) and anti-tb (red, bold arrow) stimulated with PDGF-AA for 10 min.
Open arrows indicate the edge of the cell.
(I) Localization of anti-phospho-Mek1/2 (green) and anti-pericentrin (red) that marks the mother and daughter centrioles (blue and white
(J) Colocalization of anti-phospho-Mek1 S298(green) and the primary cilium (red, tb, arrows). Abbreviation: nu, nucleus detected by DIC.
Error bars indicate standard errors from three separate experiments.
4B), supporting the conclusion that the majority of
PDGFRα in quiescent wt MEFs is ciliary. In wt MEFs,
PDGFRα was upregulated essentially as in NIH3T3
cells, whereas in serum-starved mutant cells PDGFRα
expression remained similar to that of nonarrested, in-
terphase cells (Figures 4C and 4D). In order to validate
that mutant cells had entered growth arrest, we ana-
lyzed the level of cdk4-mediated phosphorylation of tu-
mor suppressor protein retinoblastoma (Rb) at serines
807 and 811, which controls progression through the
late G1restriction point, and is a major regulator of the
G1/S transition that marks cycling cells . Western
blot analysis confirmed that the phospho-Rb in mutant
cells decreased during serum starvation and was at a
low level similar to wt cells after 48 hr starvation. Fur-
ther, mutant cells had neither mitotic spindles nor the
PDGFRα Signaling in the Primary Cilium
Figure 4. Ciliary Formation, Upregulation of PDGFRα- and PDGFRα-Mediated Signal Transduction, and Cell Cycle Entrance Is Impaired in
Tg737 Mutant Mouse Embryonic Fibroblasts
Ciliary formation (tb, red, bold arrows) and localization of rabbit anti-PDGFRα (green) in wild-type (A) and Tg737orpk(B) mutant embryonic
mouse fibroblasts, serum starved for 48 hr. Open arrows indicate the edge of cells. Scale bars: 10 ?m.
(C) Level of PDGFRα, PDGFRβ, Erk1/2, Mek1/2, phospho-Rb and β-actin in wild-type and mutant cells after serum starvation for 0, 12, 24,
and 48 hr.
(D) Quantification of PDGFRα and PDGFRβ upon serum starvation for 48 hr relative to β-actin and receptor at 0 hr of serum starvation.
(E) Time course for phosphorylation of PDGFRα, Mek1/2 and Erk1/2 upon stimulation with PDGF-AA for 3 and 10 min in wild-type and mutant
cells serum starved for 48 hr.
(F) Time course for phosphorylation of PDGFRβ, Mek1/2 and Erk1/2 upon stimulation with PDGF-BB for 3 and 10 min in wild-type and mutant
cells serum starved for 48 hr.
(G) Phospho-Rb and β-actin (control) in wild-type and mutant cells after serum starvation for 48 hr unstimulated or stimulated with PDGF-AA
or serum for 26 hr.
(H) Phospho-cdc2 and β-actin (control) in mutant cells after serum starvation for 48 hr unstimulated or stimulated with PDGF-AA or PDGF-BB.
Error bars indicate standard errors from three separate experiments.
characteristic interphase cell cytoskeleton (Figure 4B).
No major differences were observed between wt and
mutant in the levels of PDGFRβ, Mek1/2 and Erk1/2 in
growth-arrested versus nonarrested, interphase cells
PDGFR??-Mediated Signaling and Cell Cycle
Entrance in Wt and Tg737orpkMutant Cells
To investigate the role of the primary cilium in mitogenic
signaling through PDGFRαα in quiescent wt and mu-
tant MEFs, cells were stimulated with PDGF-AA and
subjected to Western blot analysis. Receptor and
Mek1/2-Erk1/2 activations in wt MEFs occurred as in
NIH3T3 fibroblasts, whereas in mutant cells PDGF-AA
failed to activate PDGFRα and the Mek1/2-Erk1/2 path-
way (Figure 4E). Importantly, receptor activation by ty-
rosine phosphorylation in wt MEFs increased about
8-fold compared to mutant cells, in which the relative
level of receptor activation was nearly unchanged dur-
ing the first 10 min of PDGF-AA stimulation, supporting
the conclusion that activation of PDGFRαα in cellular
proliferation and growth signaling depends on its ciliary
localization. In contrast, PDGF-BB-mediated activation
of PDGFRβ and signaling through the Mek1/2-Erk1/2
pathway were largely unaffected by the inability to form
a primary cilium (Figure 4F). To investigate whether
PDGF-AA and PDGF-BB stimulated wt and mutant
cells to reenter the cell cycle, we monitored phosphory-
lation of Rb on S807/811in cells serum starved for 48 hr
followed by addition of either PDGF-AA or serum as a
Current Biology Download full-text
positive control for 16 and 26 hr. In wt cells, both addi-
tions induced the phosphorylation of Rb, whereas in
mutant cells the PDGF-AA-mediated response was
blocked (Figure 4G). We also followed phosphorylation
of cdc2 on tyrosine 15, which distinguishes G2phase
entrance . In Tg737 cells only PDGF-BB, but not
PDGF-AA, increased cdc2 phosphorylation (Figure 4H),
whereas wt cells responded to both ligands, showing
that signaling through the cilium plays a major role in
PDGF-AA cell cycle regulation.
Our study shows that primary cilium formation corre-
lates with an elevation in PDGFRα expression and that
PDGFRα localizes with the cilium. It seems likely that
in quiescent fibroblasts PDGFRα-mediated signaling is
dependent upon its ciliary localization and that much of
the physiologically important PDGFRα signaling occurs
via the cilium. PDGFRα is widely expressed in human
tissues, controlling migration, proliferation, and apo-
ptosis , and mutations in the receptor play a role in
the generation of human malignancies, notably in the
pathogenesis of gastrointestinal stromal tumors , lung
tumors , and ovarian carcinoma . PDGFRαα sig-
naling through the fibroblast primary cilium or compa-
rable pathways in the primary cilium in other tissues
may be important in tissue homeostasis whereas per-
turbations in this pathway could lead to oncogenesis.
Supplemental Data include Supplemental Experimental Procedures
and two figures and are available with this article online at http://
This work was supported by the National Danish Research Council
[#21-02-0120, #21-01-0507, #21-04-0535] (S.T.C. and E.K.H.), the
Foundation of 1870 (S.T.C.), The NOVO Nordisk Foundation (L.S.),
by funds from the Albert Einstein College of Medicine (P.S.), and
from the US National Institutes of Health [GM-60992] (G.J.P.). Anti-
β-actin was a kind gift from Dr. Rønnov-Jessen (IMBF, The August
Krogh Building, University of Copenhagen, Denmark).
Received: May 6, 2005
Revised: September 1, 2005
Accepted: September 1, 2005
Published: October 25, 2005
1. Rosenbaum, J.L., and Witman, G.B. (2002). Intraflagellar trans-
port. Nat. Rev. Mol. Cell Biol. 3, 813–825.
2. Pazour, G.J., and Witman, G.B. (2003). The vertebrate primary
cilium is a sensory organelle. Curr. Opin. Cell Biol. 15, 105–110.
3. Nauli, S.M., and Zhou, J. (2004). Polycystins and mechanosen-
sation in renal and nodal cilia. Bioessays 26, 844–856.
4. Praetorius, H.A., and Spring, K.R. (2005). Physiological view of
the primary cilium. Annu. Rev. Physiol. 67, 515–529.
5. Teilmann, S.C., and Christensen, S.T. (2005). Ciliary localization
of the angiopoietin receptors Tie-1 and Tie-2 in the female re-
productive organs. Cell Biol. Int. 29, 340–346.
6. Quarmby, L.M., and Parker, J.D.K. (2005). Cilia and the cell cy-
cle? J. Cell Biol. 169, 707–710.
7. Pazour, G.J., San Agustin, J.T., Follit, J.A., Rosenbaum, J.L.,
and Witman, G.B. (2002). Polycystin-2 localizes to kidney cilia
and the ciliary level is elevated in orpk mice with polycystic
kidney disease. Curr. Biol. 12, R378–R380.
8. Ou, G., Blacque, O.E., Snow, J.J., Leroux, M.R., and Scholey,
J.M. (2005). Functional coordination of intraflagellar transport
motors. Nature 436, 583–587.
9. Huangfu, D., and Anderson, K.V. (2005). Cilia and Hedgehog
responsiveness in the mouse. Proc. Natl. Acad. Sci. USA 102,
10. Liu, A., Wang, B., and Niswander, L.A. (2005). Mouse intrafla-
gellar transport proteins regulate both the activator and repres-
sor functions of Gli transcription factors. Development 132,
11. Tucker, R.W., Pardee, A.B., and Fujiwara, K. (1979). Centriole
ciliation is related to quiescence and DNA synthesis in 3T3
cells. Cell 17, 527–535.
12. Wheatley, D.N. (1995). Primary cilia in normal and pathological
tissues. Pathobiology 63, 222–238.
13. Wheatley, D.N., Wang, A.M., and Strugnell, G.E. (1996). Expres-
sion of primary cilia in mammalian cells. Cell Biol. Int. 20,
14. Lih, C.-J., Cohen, S.N., Wang, C., and Lin-Chao, S. (1996). The
platelet-derived growth factor α-receptor is encoded by a
growth-arrest-specific (gas) gene. Proc. Natl. Acad. Sci. USA
15. Heldin, C.H., and Westermark, B. (1999). Mechanism of action
and in vivo role of platelet-derived growth factor. Physiol. Rev.
16. Bazenet, C.E., Gelderloos, J.A., and Kazlauskas, A. (1996).
Phosphorylation of tyrosine 720 in the platelet-derived growth
factor alpha receptor is required for binding of Grb2 and SHP-2
but not for activation of Ras or cell proliferation. Mol. Cell. Biol.
17. Araki, T., Nawa, H., and Neel, B.G. (2003). Tyrosyl phosphoryla-
tion of Shp2 is required for normal ERK activation in response
to some, but not all, growth factors. J. Biol. Chem. 278,
18. Slack-Davis, J.K., Eblen, S.T., Zecevic, M., Boerner, S.A., Tarc-
safalvi, A., Diaz, H.B., Marshall, M.S., Weber, M.J., Parsons,
J.T., and Catling, A.D. (2003). PAK1 phosphorylation of MEK1
regulates fibronectin-stimulated MAPK activation. J. Cell Biol.
19. Moyer, J.H., Lee-Tischler, M.J., Kwon, H.Y., Schrick, J.J., Avner,
E.D., Sweeney, W.E., Godfrey, V.L., Cacheiro, N.L., Wilkinson,
J.E., and Woychik, R.P. (1994). Candidate gene associated with
a mutation causing recessive polycystic kidney disease in
mice. Science 264, 1329–1333.
20. Knudsen, E.S., and Wang, J.Y. (1997). Dual mechanisms for the
inhibition of E2F binding to RB by cyclin-dependent kinase-
mediated RB phosphorylation. Mol. Cell. Biol. 17, 5771–5783.
21. Watanabe, N., Broome, M., and Hunter, T. (1995). Regulation of
the human WEE1Hu CDK tyrosine 15-kinase during the cell cy-
cle. EMBO J. 14, 1878–1891.
22. Heinrich, M.C., Corless, C.L., Duensing, A., McGreevey, L.,
Chen, C.J., Joseph, N., Singer, S., Griffith, D.J., Haley, A., Town,
A., et al. (2003). PDGFRA activating mutations in gastrointesti-
nal stromal tumors. Science 299, 708–710.
23. Fitzer-Attas, C., Feldman, M., and Eisenbach, L. (1993). Expres-
sion of functionally intact PDGF-alpha receptors in highly met-
astatic 3LL Lewis lung carcinoma cells. Int. J. Cancer 53,
24. Henriksen, R., Funa, K., Wilander, E., Backstrom, T., Ridder-
heim, M., and Oberg, K. (1993). Expression and prognostic sig-
nificance of platelet-derived growth factor and its receptors in
epithelial ovarian neoplasms. Cancer Res. 53, 4550–4554.