Download Transactivation of the Epidermal Growth Factor Receptor

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Cell culture wikipedia , lookup

List of types of proteins wikipedia , lookup

Organ-on-a-chip wikipedia , lookup

Tissue engineering wikipedia , lookup

Cellular differentiation wikipedia , lookup

Cell encapsulation wikipedia , lookup

Signal transduction wikipedia , lookup

SULF1 wikipedia , lookup

Chemotaxis wikipedia , lookup

Amitosis wikipedia , lookup

Transcript
Research Article
Transactivation of the Epidermal Growth Factor Receptor by
Formylpeptide Receptor Exacerbates the Malignant
Behavior of Human Glioblastoma Cells
1
3
4
1
2
1
Jian Huang, Jinyue Hu, Xiuwu Bian, Keqiang Chen, Wanghua Gong, Nancy M. Dunlop,
1
1
O.M. Zack Howard, and Ji Ming Wang
1
Laboratory of Molecular Immunoregulation, Cancer and Inflammation Program, Center for Cancer Research and 2Basic Research
Program, Science Applications International Corporation-Frederick, National Cancer Institute at Frederick, Frederick, Maryland;
3
Cancer Research Institute, Xiang-Ya School of Medicine, Central South University, Changsha, P.R. China; and
4
Institute of Pathology, South-West Hospital, Third Military Medical University, Chongqing, P.R. China
Abstract
The G protein-coupled formylpeptide receptor (FPR), which
mediates leukocyte migration in response to bacterial and
host-derived chemotactic peptides, promotes the chemotaxis,
survival, and tumorigenesis of highly malignant human glioblastoma cells. Because glioblastoma cells may also express
other receptors for growth signals, such as the epidermal
growth factor (EGF) receptor (EGFR), we investigated the role
of EGFR in the signaling cascade of FPR and how two
receptors cross-talk to exacerbate tumor growth. We found
that N-formyl-methionyl-leucyl-phenylalanine, an FPR agonist
peptide, rapidly induced EGFR phosphorylation at tyrosine
residue (Tyr) 992, but not residues 846, 1068, or 1173, in glioblastoma cells, whereas all these residues were phosphorylated after only EGF treatment. The FPR agonist-induced
EGFR phosphorylation in tumor cells was dependent on the
presence of FPR as well as GAi proteins, and was controlled by
Src tyrosine kinase. The transactivation of EGFR contributes
to the biological function of FPR in glioblastoma cells because
inhibition of EGFR phosphorylation significantly reduced FPR
agonist-induced tumor cell chemotaxis and proliferation. Furthermore, depletion of both FPR and EGFR by short interference RNA abolished the tumorigenesis of the glioblastoma
cells. Our study indicates that the glioblastoma-promoting
activity of FPR is mediated in part by transactivation of EGFR
and the cross-talk between two receptors exacerbates the
malignant phenotype of tumor cells. Thus, targeting both
receptors may yield antiglioblastoma agents superior to those
targeting one of them. [Cancer Res 2007;67(12):5906–13]
Introduction
Formylpeptide receptor (FPR) is a G protein-coupled receptor
(GPCR), originally identified in phagocytic leukocytes, which
mediates cell chemotaxis and activation in response to the
bacterial chemotactic peptide N-formyl-methionyl-leucyl-phenylalanine ( fMLF). Agonist binding to FPR elicits a cascade of signal
Note: Supplementary data for this article are available at Cancer Research Online
(http://cancerres.aacrjournals.org/).
Requests for reprints: Ji Ming Wang, Laboratory of Molecular Immunoregulation,
Cancer and Inflammation Program, Center for Cancer Research, National Cancer
Institute at Frederick, Building 560, Room 31-76, Frederick, MD 21702-1201. Phone:
301-846-6979; Fax: 301-846-7042; E-mail: [email protected].
I2007 American Association for Cancer Research.
doi:10.1158/0008-5472.CAN-07-0691
Cancer Res 2007; 67: (12). June 15, 2007
transduction pathways that involve phosphatidylinositol 3-kinase
(PI3K), mitogen-activated protein kinases (MAPK), and the
transcription factor nuclear factor-nB ( for review, see ref. 1).
Because of its expression in cells of the immune system and its
interaction with bacterial chemotactic peptides, this receptor was
thought to participate in host defense against microbial infection.
In fact, mice depleted of the FPR analogue FPR1 were more
susceptible to infection by Listeria monocytogenes (2). During
the past few years, a number of novel and host-derived chemotactic agonists of FPR have been identified, including formyl
peptides potentially released by mitochondria of ruptured cells (3),
Annexin I produced by activated epithelia (4), and a neutrophil
granule protein, cathepsin G (5). In addition, functional FPR has
been detected in cells of nonhematopoietic origin, such as lung
epithelial cells (6) and hepatocytes (7). These findings suggest
that FPR may be involved in a broader spectrum of pathophysiologic processes.
Gliomas are the most common tumor type in the brain,
characterized by progressive expansion and resistance to conventional therapy. The capacity of glioma growth and invasion is
closely correlated with the expression of cell surface receptors that
sense the signals present in the tumor microenvironment (8, 9).
Highly malignant human glioblastoma cells express functional FPR
and by responding to potential agonist(s) released by necrotic
tumor cells, FPR promotes the directional migration, survival, and
production of angiogenic vascular endothelial growth factor
(VEGF) by tumor cells (10). Depletion of FPR by short interference
RNA (siRNA) markedly reduced the tumorigenicity of the
glioblastoma cells in immunodeficient mice (10). In addition, FPR
protein was detected in surgical specimens of grade 3 astroblastoma and grade 4 glioblastoma multiforme in human (10). Thus,
both in vitro and in vivo evidence supports the potential role of
FPR in promoting the rapid progression of highly malignant human
glioma.
Human glioma cells may also express the receptor for
epidermal growth factor (EGFR; refs. 11–13), which has been
implicated as one of the most important growth-stimulating
receptors in a great variety of malignant tumors (14) and is a
partner of cross-talk with a variety of cell surface receptors. In this
study, we investigated the capacity of human glioblastoma cells to
concomitantly express both FPR and EGFR and whether these two
receptors share intracellular signaling pathways. We found that
activation of FPR resulted in transactivation of EGFR in
glioblastoma cells and the two receptors interact and synergistically cooperate to exacerbate the malignant behavior of the tumor
cells.
5906
www.aacrjournals.org
Transactivation of EGFR by Formylpeptide Receptor
Materials and Methods
Reagents. fMLF was purchased from Sigma-Aldrich. Antibodies against
phosphorylated EGFR (at tyrosine 992, 845, 1068, and 1173) and
extracellular signal-regulated kinases 1/2 (ERK1/2) were from Cell Signaling
Technology. The EGFR-specific tyrosine kinase inhibitor AG1478 was from
Calbiochem. The Src tyrosine kinase inhibitor PP2 [4-amino-5-(4-chlorophenyl)-7-(t-butyl) pyrazolo[3,4-d] pyrimidine] was from Calbiochem.
Neutralizing antibodies against EGF and the extracellular domain of EGFR
were from Santa Cruz Biotechnology. Human glioblastoma cell line U87 was
obtained from the American Type Culture Collection. The cells were grown
in DMEM containing 10% FCS and antibiotics.
Chemotaxis. Chemotaxis assays were done in 48-well chemotaxis
chambers (NeuroProbe; refs. 10, 15). A 27-AL aliquot of chemoattractants
was placed in the wells of the lower compartment, and 50 AL of tumor cells
(at 3 106/mL) were placed in the wells of the upper compartment. The
upper and lower compartments of the chemotaxis chambers were
separated by a 10-Am-pore polycarbonate filter (GE Osmonics Labstore)
coated with collagen type I (BD Biosciences) at 50 Ag/mL. After incubation
at 37jC for 270 min, the filters were removed and stained, and cells that
migrated across the filters were counted under light microscopy. The results
were expressed as the mean number (FSE) of migrated cells in three highpowered fields (400 magnification) in triplicate samples or as chemotaxis
index representing the fold increase in cell migration in response to
stimulants over medium control.
Immunoblot. Tumor cells were lysed in 150 AL of 1 SDS sample buffer
[62.5 mmol/L Tris-HCl (pH 6.8), 2% SDS, 10% glycerol, and 50 mmol/L DTT],
sonicated for 3 s, and then boiled for 5 min. The cell lysate was then
centrifuged at 10,000 g at 4jC for 10 min. Total protein was
electrophoresed on 4% to 12% gradient Tris-glycine precast gels (Invitrogen)
and transferred onto Immunoblotin P membranes (Millipore). The
membranes were blocked by incubation in 3% nonfat dry milk for 1 h at
room temperature and then incubated with primary antibodies in PBS
containing 0.01% Tween 20 overnight at 4jC. After incubation with a
horseradish peroxidase–conjugated secondary antibody, the protein bands
were detected with Super Signal Chemoluminescent Substrate Stable
Peroxide Solution (Pierce) and BIOMAX-MR film (Eastman Kodak). When
necessary, the membranes were stripped with Restore Western Blot
Stripping Buffer (Pierce) and reprobed with antibodies against various
cellular proteins.
Cell proliferation. Tumor cells were cultured in 12-well tissue culture
plates at 2 105 per well in DMEM containing 10% FCS for 12 h. After a
1-h treatment with various inhibitors, the cells were incubated in DMEM
containing 0.5% FCS alone or in the presence of 1 Amol/L fMLF or 10 ng/mL
EGF. After incubation at 37jC for different times, the cells were detached by
0.5% trypsin-EDTA and counted. The results were expressed as mean
number (FSE) of cells in six replicates. 3-(4,5-Dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide (MTT) assays were also done by Cell Growth
Determination kit (Sigma) to assay the proliferation of U87 glioblastoma
cells. Briefly, cells cultured in 96-well plates were incubated with MTT
solution at 100 AL for 3 h, the supernatants were then removed, and 100 AL
MTT solvent were added. After gentle pipetting, absorbance was measured
in a spectrometer at 570 nm wavelength. After subtraction of the
absorbance at 690 nm, the results are expressed as the mean proliferation
units (FSE) obtained from six replicates.
Ca2+ flux. Ca2+ mobilization was measured by incubating 2 107 tumor
cells in 1 mL in loading medium (DMEM containing 10% FCS and 2 mmol/L
glutamine) with 7 Amol/L Fura-2 acetoxymethyl ester (Molecular Probes)
for 45 min at room temperature. The dye-loaded cells were washed and
resuspended in saline buffer [138 mmol/L NaCl, 6 mmol/L KCl, 1 mmol/L
CaCl2, 10 mmol/L HEPES, 5 mmol/L glucose, and 0.1% bovine serum
albumin (pH 7.4)] at a density of 0.5 106/mL. The cells were then
transferred into quartz cuvettes (1 106 cells in 2 mL of saline buffer), and
the cuvettes were placed in a fluorescence spectrometer (Perkin-Elmer).
Stimulants were added to the cuvettes in a volume of 20 AL, and the
intensity of the fluorescence was measured by the use of the ratio of the
absorbance at 340 to 380 nm, calculated with an FL WinLab program
(Perkin-Elmer).
www.aacrjournals.org
Detection of EGFR dimers. After stimulation with EGF or fMLF, U87
cells were washed twice with ice-cold PBS containing 0.33 mmol/L MgCl2
and 0.9 mmol/L CaCl2 [PBS (+)] and chemically cross-linked for 30 min at
room temperature with freshly prepared 5 mmol/L bis(sulfosuccinimidyl)
suberate (Pierce). The reaction was terminated by further incubating the
cells with 200 mmol/L glycine for 5 min. The cells were washed twice
with ice-cold PBS (+) and lysed with 1 SDS. The lysate was centrifuged at
20,000 g for 10 min, then equal amount of proteins were separated by 6%
Tris-glycine SDS-PAGE and subjected to immunoblot to detect EGFR.
Generation of U87 cells stably expressing EGFR siRNA. The
generation of U87 cells transfected by FPRsiRNA was described previously
(10, 16). The FPRsiRNA-transfected cells were further transfected by
retrovirus containing blank or EGFR siRNA expression cassettes. A hairpin
19-nucleotide sequence was targeted to EGFR mRNA at the nucleotides
645 to 663 (5¶-GGAGCTGCCCATGAGAAAT-3¶, Genbank accession no.
NM_005228). Retroviral vector stocks were produced by transient transfection of Phoenix-Ampho cells with the Superfect Transfection Reagent
(Qiagen) and 5 Ag of siRNA expression plasmid (pRNATin-H1.4/Retro,
SD1254, Genscript). The virus was collected from the culture supernatants on day 2 after transfection, and the cells were transfected with the
retroviral vectors in the presence of polybrene at 5 Ag/mL. The stably
transfected U87 cells containing expression cassettes of EGFRsiRNA
(EsiRNA), FPRsiRNA (FsiRNA), or both siRNA (EFsiRNA) were selected
and maintained by incubation with 2 Ag/mL puromycin and 200 ng/mL
hygromycin (BD Biosciences).
Tumor microsphere formation. Anchorage-independent growth of U87
cells containing EsiRNA, FsiRNA, EFsiRNA, or blank cassettes (Mock) were
examined in a semisolid agar. DMEM containing 10% FCS was warmed to
48jC and diluted with Bacto-Agar to make a 0.6% (w/v) solution. The
solution (300 AL) was poured into 24-well culture plates and allowed to
solidify for 15 min. A volume of 200 AL of 0.3% agar solution containing
2,500 tumor cells were then layered over the bottom agar in each well. The
plates were incubated in 5% CO2 at 37jC, with a medium change every
other day. After 3 weeks, tumor colonies were recorded under light
microscopy.
Tumorigenesis. U87 cells containing EsiRNA, FsiRNA, or EFsiRNA were
implanted s.c. into the flank of 4-week-old (20–22 g) female athymic Ncrnu/nu mice [National Cancer Institute (NCI) Animal Production Program,
NCI-Frederick, Frederick, MD] at 5 106 per mouse in 100 AL of PBS. The
growth of implanted tumors was examined every other day. Tumor size was
calculated by the formula lw 2/2, where l is the length of the tumor and w is
the width. Each group contained at least five mice in one experiment.
Nontransfected U87 cells and cells transfected with random siRNA (mock)
were used as controls. The results were expressed as the mean (FSE)
number of colonies counted in triplicates. NCI-Frederick is accredited by
the Association for Assessment and Accreditation of Laboratory Animal
Care International and follows the Public Health Service Policy for the Care
and Use of Laboratory Animals. Animal care was provided in accordance
with the procedures outlined in the Guide for Care and Use of Laboratory
Animals (1996, National Research Council, National Academy Press,
Washington, DC).
Statistical analyses. All experiments were done at least thrice. A
computer-aided t test program SPSS (version 11.0; GraphPad Software, Inc.)
was used to determine the statistical significance of the difference between
cell responses to testing materials versus controls. The statistical
significance of EGF-induced tumor cell chemotaxis was analyzed with
ANOVA. P V 0.05 were considered statistically significant.
Results
Glioblastoma cells express functional EGFR. We first tested
for the presence of functional EGFR on human glioblastoma cell
line U87, which expresses the GPCR FPR. Figure 1 shows that U87
cells migrated in response to EGF at low ng/mL concentration
range (Fig. 1A). At higher ng/mL concentrations, EGF also induced
a robust Ca2+ flux in U87 cells. In addition, EGF induced EGFR
5907
Cancer Res 2007; 67: (12). June 15, 2007
Cancer Research
Figure 1. The expression and function of EGFR and FPR in human glioblastoma cell line U87. A, EGF-induced chemotaxis of U87 cells. Cell chemotaxis was assayed
by 48-well chemotaxis chambers in response to different concentrations of EGF, which were placed in the lower wells of the chemotaxis chamber. Columns, mean chemotaxis
index (CI), representing the fold increase in cell migration in responses to stimulants over medium control; bars, SE. ANOVA was used to analyze the statistical
significance of cell migration induced by various EGF concentrations. *, P < 0.01, significantly increased cell migration compared with medium control (0). w , P < 0.01,
significantly increased cell migration compared with 1 ng/mL EGF-induced chemotaxis. 4, P < 0.01, significantly increased cell migration compared with 10 ng/mL EGF-induced
chemotaxis. B, EGF-induced phosphorylation of EGFR. FCS-starved U87 cells were stimulated with 10 ng/mL EGF at different time points or with different concentrations
of EGF for 2 min. The cell lysates were detected for EGFR phosphorylation by using an antibody against phosphorylated (p-) Tyr992. C, EGF-induced Ca2+ flux was measured
in a fluorescence spectrometer. The fluorescent intensity was expressed as the ratio at wavelength 340/380 calculated by an FLWINLab program. D, fMLF-induced U87
cell chemotaxis (left) and Ca2+ (right). *, P < 0.01, significantly increased cell migration compared with medium control (0).
phosphorylation at tyrosine residue Tyr992 (Fig. 1B and C). In
parallel experiments, we confirmed that U87 cells responded to the
FPR agonist peptide fMLF by chemotaxis and Ca2+ flux (Fig. 1D).
Thus, U87 concomitantly expresses functional FPR and EGFR.
FPR agonist fMLF induces EGFR phosphorylation. We next
examined the capacity of FPR to transactivate EGFR in U87 cells.
Figure 2 shows that stimulation of the tumor cells with EGF
resulted in the phosphorylation of EGFR at Tyr845, Tyr1173, Tyr1068,
in addition to Tyr992 in a time- and concentration-dependent
manner with maximal phosphorylation of all tyrosine residues
occurring at 2 min with 10 ng/mL EGF (Fig. 2A). fMLF also induced
EGFR phosphorylation; however, in contrast to the effect of
EGF, only Tyr992 in EGFR underwent a rapid and persistent
phosphorylation in response to fMLF at an optional concentration
of 100 nmol/L (Fig. 2A–C).
fMLF-induced EGFR Tyr992 phosphorylation is mediated by
FPR. The fMLF-induced EGFR Tyr992 phosphorylation was
mediated by the receptor FPR, because pertussis toxin, an inhibitor of the Gai proteins coupled to FPR, but not cholera toxin,
a Gas protein inhibitor, abolished the effect of fMLF (Fig. 3A).
Figure 2. Phosphorylation of EGFR induced by
EGF and fMLF. A, FCS-starved U87 cells were
stimulated with EGF (10 ng/mL) or fMLF
(100 nmol/L) for different time periods. The cell
lysates were examined for EGFR phosphorylation
by using antibodies against phosphorylated
residues Tyr992, Tyr845, Tyr1068, and Tyr1173.
B and C, time and dose response of fMLF-induced
EGFR Tyr992 phosphorylation.
Cancer Res 2007; 67: (12). June 15, 2007
5908
www.aacrjournals.org
Transactivation of EGFR by Formylpeptide Receptor
Figure 3. Involvement of FPR and Gai
proteins in fMLF-induced EGFR Tyr992
phosphorylation. A, U87 cells were
preincubated with 500 ng/mL pertussis
toxin (PTX , an inhibitor of Gai protein) or
500 ng/mL cholera toxin (CTX, an inhibitor
of Gas protein) at 37jC for 60 min, then
were stimulated with fMLF (100 nmol/L,
10 min) or EGF (10 ng/mL, 2 min).
The cell lysates were measured for EGFR
Tyr992 phosphorylation. Densitometric
measurement of phosphorylated Tyr992
was done using Image J program after
normalization against total EGFR protein
(EGFR ). B, chemotaxis of mock or FPR
siRNA (FsiRNA)–transfected U87 cells in
response to fMLF or EGF. *, P < 0.01,
significantly reduced cell migration
compared with mock-transfected cells
(Mock ). BM , medium. C, phosphorylation
of EGFR Tyr992 induced by EGF
(10 ng/mL, 1 min) or fMLF (100 nmol/L,
10 min) in mock or FsiRNA-transfected
U87 cells.
In contrast, both pertussis toxin and cholera toxin failed to inhibit
EGF-induced EGFR phosphorylation (Fig. 3A). The need for FPR in
fMLF-induced EGFR Tyr992 phosphorylation was further confirmed
by using U87 cells transfected with FPRsiRNA (10), which lost the
capacity to migrate in response to fMLF and also failed to undergo
EGFR Tyr992 phosphorylation (Fig. 3B and C). In contrast, the same
FPRsiRNA containing U87 cells exhibited normal chemotaxis and
EGFR Tyr992 phosphorylation in response to EGF (Fig. 3B and C).
These results show the requirement for FPR to mediate fMLFinduced EGFR Tyr992 phosphorylation in U87 glioblastoma cells.
The mechanisms controlling EGFR transactivation by
activated FPR. In an effort to elucidate the mechanistic basis
for EGFR Tyr992 phosphorylation induced by FPR agonist peptide
fMLF, we investigated whether fMLF-stimulated glioblastoma
cells produced EGF, which, in turn, activated EGFR in an autocrine
loop. Figure 4 shows that a neutralizing anti-EGFR antibody was
Figure 4. The role of EGF and Src kinase in fMLF-induced EGFR phosphorylation and dimerization of EGFR. A, blockage of EGF-induced, but not fMLF-induced,
U87 cell chemotaxis by anti-EGFR antibody. U87 cells were preincubated with anti-EGFR (1 Ag/mL) at 37jC for 30 min, then were measured for chemotaxis in response
to fMLF or EGF. *, P < 0.01, significantly reduced cell migration compared with IgG-treated U87 cells. B, inhibition of EGF-induced, but not fMLF-induced
phosphorylation by anti-EGFR and anti-EGF antibodies. U87 cells were preincubated with 1 Ag/mL anti-EGFR antibody or anti-EGF antibody at 37jC for 30 min;
the cells were then stimulated with fMLF or EGF (10 ng/mL, 2 min) and measured for EGFR Tyr992 phosphorylation. In parallel experiments, EGF (10 ng/mL) or
fMLF (100 nmol/L) was preincubated with an anti-EGF antibody (1 Ag/mL) for 30 min at 37jC. The mixtures were then added to cell culture (EGF/anti-EGF, 2 min;
fMLF/anti-EGF, 10 min) before measurement of EGFR Tyr992 phosphorylation. C, inhibition of fMLF-induced EGFR Tyr992 phosphorylation by the Src kinase
inhibitor PP2. U87 cells were preincubated with 20 Amol/L PP2 at 37jC for 30 min, then were stimulated with EGF (10 ng/mL, 2 min) or fMLF (100 nmol/L, 10 min) and
measured for Tyr992 phosphorylation. D, U87 cells were treated with 10 ng/mL EGF for 2 min or 100 nmol/L fMLF for 10 min. After two washes with ice-cold PBS (+),
the cells were treated with the protein cross-linker bis(sulfosuccinimidyl) suberate (BS3 ). The cell lysates were separated by 6% Tris-glycine SDS-PAGE and subjected
to immunoblot with anti-EGFR antibody. Densitometry denotes the ratio of EGFR dimmers/monomers.
www.aacrjournals.org
5909
Cancer Res 2007; 67: (12). June 15, 2007
Cancer Research
capable of abolishing EGF-induced U87 cell chemotaxis and the
phosphorylation of EGFR Tyr992 (Fig. 4A and B). However, this antiEGFR antibody failed to show any effect on fMLF-induced U87 cell
migration or phosphorylation of EGFR Tyr992 (Fig. 4A and B). In
addition, whereas the anti-EGF antibody completely inhibited
EGF-induced EGFR Tyr992 phosphorylation in U87 cells, it did not
show any effect on fMLF-induced phosphorylation of EGFR Tyr992
(Fig. 4B). These results suggest that phosphorylation of EGFR
Tyr992 in response to fMLF is not dependent on the release of EGF
by FPR agonist-stimulated tumor cells.
We subsequently examined the signal transduction pathways
coupled to FPR that might be responsible for the phosphorylation
of EGFR Tyr992. In addition to the Gai proteins, which were
essential for the effect of FPR as shown by inhibition with the
Gai-specific inhibitor pertussis toxin (Fig. 3A), we found that an
inhibitor of the Src tyrosine kinase PP2 completely inhibited
fMLF-induced, but not EGF-induced, EGFR Tyr992 phosphorylation
(Fig. 4C). In contrast, the inhibitors of protein kinase C or PI3K did
not block the effect of fMLF on EGFR Tyr992 phosphorylation (data
not shown). We therefore conclude that Src kinase plays a key
role in bridging the signal transduction from FPR to EGFR Tyr992.
We additionally found that activation of FPR in U87 cells induced
rapid dimerization of EGFR (Fig. 4D), an evidence further supporting the transactivation of EGFR by FPR agonist peptide fMLF.
FPR and EGFR cooperate to promote the motility and
growth of glioblastoma cells. To investigate the biological
significance of EGFR Tyr992 phosphorylation in FPR-mediated
glioblastoma cell chemotaxis and growth, we used AG1478, a
specific tyrosine kinase inhibitor of activated EGFR. We found that
both fMLF- and EGF-induced EGFR Tyr992 phosphorylation was
inhibited when U87 cells were treated with AG1478 (Fig. 5A).
Although AG1478-treated U87 cells failed to respond to EGF by
phosphorylation of ERK1/2 MAPK, chemotaxis, and proliferation,
AG1478 partially but also significantly inhibited fMLF-induced
ERK1/2 phosphorylation, chemotaxis, and proliferation of the
tumor cells (Fig. 5A–C). Interestingly, although AG1478-treated U87
cells lost the capacity to respond to EGF by Ca2+ flux (Fig. 5D), they
showed normal Ca2+ mobilization to fMLF (Fig. 5D). Thus, EGFR
Tyr992 transactivation participates in FPR-mediated signaling
pathways that promote glioblastoma cell chemotaxis and proliferation. In contrast, the FPR signaling pathways mediating Ca2+
mobilization is independent of EGFR transactivation.
We then explored the effect of cross-talk between FPR and EGFR
on the malignant behavior of glioblastoma cells. In a soft-agar
tumor microsphere formation model, we found that both wild-type
and mock-transfected U87 cells rapidly formed numerous tumor
spheres (Fig. 6A). U87 cells containing either FPRsiRNA or EsiRNA
each showed significantly (P < 0.01) retarded formation of tumor
spheres (Fig. 6B) as measured by the sphere numbers and size.
Knocking down both EFsiRNA further diminished the capacity of
U87 cells to form larger (50–200 Am) spheres. These results were
consistent with in vitro cell growth curves showing the proliferation of U87 cells with either FsiRNA or EsiRNA to be diminished,
whereas combination of both siRNAs further reduced the rate of
cell growth (Fig. 6C).
We then injected human glioblastoma cells s.c. into the flanks of
athymic mice and measured the rate of tumor formation and
growth. Tumor nodules appeared in all mice injected with wildtype U87 or mock-transfected U87 cells by day 5. In contrast, 60% of
the mice implanted U87 cells containing EsiRNA or FsiRNA grew
visible tumors only by day 21 after implantation. None of the mice
implanted with U87 cells containing both EsiRNA and FsiRNAs
Figure 5. Involvement of transactivated EGFR in
fMLF-induced glioblastoma cell proliferation and
chemotaxis. A, U87 cells were preincubated with
500 nmol/L AG1478 (AG ), an EGFR-specific tyrosine
kinase inhibitor, at 37jC for 30 min, followed by
stimulation with fMLF (100 nmol/L, 10 min) or EGF
(10 ng/mL, 2 min) and measurement of EGFR Tyr992
phosphorylation. U87 cells were also preincubated with
1 Amol/L AG1478 for 30 min followed by stimulation
with fMLF (100 nmol/L, 10 min) or EGF (10 ng/mL,
2 min) and measurement of phosphorylated ERK1/2.
B, U87 cells were preincubated with 500 nmol/L
AG1478 (AG+) at 37jC for 60 min, then were
measured for chemotaxis in response to different
concentrations of fMLF or EGF. *, P < 0.01,
significantly reduced cell migration compared with cells
treated with medium alone (AG). C, U87 cells were
cultured in 24-well tissue culture plates at 1.5 105 per
well in DMEM containing 10% FCS for 12 h. The cells
were then incubated with 0.5% FCS DMEM in the
presence or absence of 500 nmol/L AG for 1 h followed
by fMLF (1 Amol/L) or EGF (100 ng/mL). The cells were
harvested at different time intervals by detachment
with trypsin-EDTA and viable cells were counted under
a microscope. Points, mean number of the cells
counted in four replicates; bars, SE. *, P < 0.05,
significantly reduced cell number in the presence of
AG1478 and fMLF or EGF compared with the cell
numbers in the presence of fMLF or EGF alone.
D, Fura-2–loaded U87 cells were incubated in the
presence (AG+) or absence (AG ) of 500 nmol/L
AG1478 for 30 min, followed by measurement of Ca2+
flux in response to different concentrations of EGF
or fMLF. The fluorescent units represent the ratio
measured at 340/380 wavelength and calculated by
FLWINLab program.
Cancer Res 2007; 67: (12). June 15, 2007
5910
www.aacrjournals.org
Transactivation of EGFR by Formylpeptide Receptor
Figure 6. Sphere formation, in vitro proliferation, and in vivo growth of U87 cells containing EGFR or FPRsiRNA. A and B, nontransfected U87 (wild-type, WT),
U87 cells containing random siRNA (Mock ), EsiRNA, FsiRNA, or EFsiRNA suspended in 200 AL 0.3% agar (2,500 cells) were layered on the solidified bottom agar.
The cells were grown in 5% CO2 at 37jC, with medium changes every other day. After 3 wk, tumor colonies were photographed under light microscope. Columns,
mean numbers of colonies detected in triplicate samples; bars, SE. The spheres were counted at the size of 5 to 100 Am and 100 to 200 Am in diameter with the
aid of an Olympus IX71 inverted microscope and Imagepro-Plus 6.0 software. *, P < 0.01, significantly reduced sphere numbers compared with mock cells;
4, further reduced spheres compared with FsiRNA cells (P < 0.05). The composition of spheres by tumor cells was confirmed by 4¶,6-diamidino-2-phenylindole
(DAPI ) staining of a representative sample from wild-type U87 cells. C, U87 cells, mock cells, or siRNA cells (EsiRNA, FsiRNA , and EFsiRNA) were seeded at
3,000 per well in 96-well flat-bottomed plates in the presence of 200 AL DMEM and 5% FCS. The cells were then cultured at 37jC with 5% CO2 and were measured
for MTT uptake at different time points. Points, mean absorbance at 570 nm wavelength after substraction of the absorbance at 690 nm measured in six replicate
samples; bars, SE. *, P < 0.01, significantly reduced absorbance compared with mock cells. E, P < 0.05, significantly reduced absorbance compared with EsiRNA
cells. n, P < 0.05, significantly reduced absorbance compared with FsiRNA cells. D, U87 cells (WT ), mock cells, and cells containing FsiRNA, EsiRNA, or
EFsiRNA at 5 106 in 100 AL of PBS per mouse were injected s.c. into the flanks of athymic mice. Mice were examined for tumor formation and size on the indicated
days. Points, mean volume (in mm3) of the tumors; bars, SE. *, P < 0.01, significantly reduced size compared with mock cells as measured on days 17, 21, 25,
29, and 33 after implantation. Tumors were photographed on day 33 after implantation. Tumor cells containing both EsiRNA and FsiRNA did not form any tumor nodule
in nude mice by day 33 after implantation.
developed tumors by day 33 posttransplantation (Fig. 6D). In
addition, tumors formed by wild-type or mock-transfected cells
grew more rapidly than those formed by cells containing either
EsiRNA or FsiRNA (Fig. 6D; Supplementary Fig. S1). These results
indicate that both FPR and EGFR play important roles in the tumor
cell growth and tumorigenesis, and two receptors cooperate to
potentiate the malignant behavior of the glioblastoma cells.
Discussion
FPR in U87 glioblastoma cells transactivates EGFR. In this
study, we showed that FPR in human glioblastoma cells transduces
signals that selectively induce the phosphorylation of EGFR Tyr992,
which is partially responsible for the chemotactic and growthstimulating effect of FPR agonist peptide fMLF on tumor cells. To
our knowledge, this is the first demonstration that FPR, a GPCR
that is implicated in inflammation and host defense against
microbial infection mediated by cells of the innate immunity (1), is
capable of engaging EGFR in promoting glioblastoma cell growth
and tumorigenesis.
www.aacrjournals.org
EGFR transactivation contributes to the biological activity
of FPR in tumor cells. EGFR is a transmembrane cell surface
receptor and a member of the c-erb-B family of tyrosine kinases,
known to be overexpressed in a variety of human malignant
tumors, including carcinomas of the colon, lung, breast, and head
and neck (17–20). Human glioma cells have also been shown to
express EGFR and, in fact, f40% of clinical cases of glioblastomas
show EGFR gene amplification (13). Following ligand binding,
EGFR undergoes rapid dimerization and activation resulting in
tumor cell proliferation, chemotaxis, angiogenesis, and metastatic
spread. Therefore, EGFR has been considered as a molecular
target for cancer treatment. In support of this notion, treatment of
mice bearing human malignant glioma cells with antibody
against EGFR or inhibitors of EGFR tyrosine kinase, considerably,
albeit incompletely, suppressed tumor growth (21, 22). Our study
confirms that EGFR is also expressed by highly malignant human
glioblastoma cells and its depletion by siRNA reduced the
tumorigenic capacity of the tumor cells. Nevertheless, our study
also showed that depletion of EGFR alone was inadequate to
completely suppress the tumorigenicity of glioblastoma cells. This
5911
Cancer Res 2007; 67: (12). June 15, 2007
Cancer Research
goal was achieved only when both FPR and EGFR were silenced in
tumor cells.
Besides activation by its cognate ligand EGF, EGFR has been
reported to be transactivated by a considerable number of cell
surface molecules to convey growth signals (23). For example,
receptors for prolactin, growth hormones, and adhesion molecules
as well as stress factors such as oxidants and irradiation are
reported to transactivate EGFR. Furthermore, some different
GPCRs, in addition to FPR shown in this study, are also capable
of transactivating EGFR, including the receptors for lysophosphatidic acid (LPA; refs. 24–26), thrombin (27–29), endothelin-1 (30–34),
carbachol (35–37), angiotensin (38–40), bombesin (41, 42), and the
chemokine SDF1 (CXCL12; ref. 43). Thus, EGFR plays an important
role in assisting the transmission of growth signals in cells in
response to a great variety of stimulants.
The mechanisms of EGFR transactivation by GPCRs. The
mechanistic basis for GPCRs to transactivate EGFR has not been
previously well established and varies in different cell types.
However, accumulating evidence suggests that GPCR-induced
EGFR transactivation may involve either EGFR ligand–dependent
and EGFR ligand–independent pathways. Studies with Rat-1 cell
lines (44) have revealed that GPCR for LPA ‘‘transactivates’’ EGFR
by activating metalloproteinases to induce EGF shedding from proHB-EGF on the cell surface, actually resulting in a more direct
ligand activation of the EGFR. This pathway has been ruled out in
our experiments because neutralization of either EGF or EGFR with
antibodies did not prevent EGFR phosphorylation induced by FPR
agonist in glioblastoma cells. Therefore, FPR in glioblastoma cells
seems to mediate an intracellular trans-signal cascade that
phosphorylates and dimerizes EGFR (Supplementary Fig. S2). This
mechanism was also shown in studies of COS-7 cells in which the
tyrosine kinase Src plays a crucial role in EGFR phosphorylation in
response to LPA and a2A-adrenergic GPCRs, or by overexpression
of Ghg subunits, without involvement of the intrinsic kinase
activity of EGFR, which presumably is activated mainly by its
cognate ligand, or induction of EGF shedding (45). Our study also
showed that whereas Gai was essential for FPR transactivation of
EGFR in glioblastoma cells, Src seems to play a key role in linking
FPR to the intracellular domains of EGFR where Tyr992 is
phosphorylated (Supplementary Fig. S2). Our study additionally
revealed that FPR agonist fMLF rapidly induced the dimerization of
References
1. Le Y, Murphy PM, Wang JM. Formyl-peptide receptors
revisited. Trends Immunol 2002;23:541–8.
2. Gao JL, Lee EJ, Murphy PM. Impaired antibacterial
host defense in mice lacking the N -formylpeptide
receptor. J Exp Med 1999;189:657–62.
3. Rabiet MJ, Huet E, Boulay F. Human mitochondriaderived N -formylated peptides are novel agonists
equally active on FPR and FPRL1, while Listeria
monocytogenes -derived peptides preferentially activate
FPR. Eur J Immunol 2005;35:2486–95.
4. Xia SH, Hu LP, Hu H, et al. Three isoforms of annexin I
are preferentially expressed in normal esophageal
epithelia but down-regulated in esophageal squamous
cell carcinomas. Oncogene 2002;21:6641–8.
5. Sun R, Iribarren P, Zhang N, et al. Identification of
neutrophil granule protein cathepsin G as a novel
chemotactic agonist for the G protein-coupled formyl
peptide receptor. J Immunol 2004;173:428–36.
6. Rescher U, Danielczyk A, Markoff A, Gerke V.
Functional activation of the formyl peptide receptor
Cancer Res 2007; 67: (12). June 15, 2007
EGFR in glioblastoma cells, consistent with the observation with
the muscarinic acetylcholine receptor, which, when transfected in
human embryonic kidney epithelial cells, induced EGFR dimerization (46).
The biological significance of FPR-mediated EGFR transactivation in glioblastoma cells. The transactivation of EGFR by
FPR in glioblastoma cells may have important pathophysiologic
implications. Although FPR was originally identified in myeloid
cells of the innate immune host defense system, its aberrant
expression in glioblastoma cells renders tumor cells capable of
responding not only to bacterial chemotactic agonist peptides but
also to agonist activity contained in the environment of necrotic
tumor cells (10). These are likely to be present in tumor tissues
because one of the characteristic features of human glioblastomas
is the early appearance of necrotic foci due to rapid progression,
and ruptured cells are likely to release FPR agonists such as
mitochondria peptides (3). In addition, FPR in glioblastoma cells
may also interact with other host-derived agonists such as the
neutrophil granule protein cathepsin G, produced under inflammatory conditions (5). FPR in glioblastoma cells, although directly
contributing to the tumor cell chemotaxis, growth, and production
of the angiogenic VEGF (10), seems also to take the advantage of
EGFR and by transactivating EGFR, increases the level of tumor cell
responses to FPR agonists. This is best shown by the observations
in which knocking down either FPR or EGFR each considerably
diminished the tumorigenicity of the tumor cells, whereas
concomitant depletion by siRNA of FPR and EGFR completely
abolished the tumorigenic capacity of glioblastoma cells in nude
mice. Thus, our study proposes that FPR and EGFR cooperate to
exacerbate the malignant behavior of human glioblastoma cells
and targeting both receptors may yield superior therapeutic effects
compared with targeting either one receptor.
Acknowledgments
Received 2/20/2007; revised 3/20/2007; accepted 4/9/2007.
Grant support: Federal funds from the National Cancer Institutes, NIH, under
contract no. NO1-CO-12400 and Intramural Research Program of the NCI, NIH.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Joost J. Oppenheim for reviewing the manuscript, and Cheryl N.
Magers and Cheryl F. Lamb for secretarial assistance.
by a new endogenous ligand in human lung A549 cells.
J Immunol 2002;169:1500–4.
7. McCoy R, Haviland DL, Molmenti EP, Ziambaras T,
Wetsel RA, Perlmutter DH. N-formylpeptide and complement C5a receptors are expressed in liver cells and
mediate hepatic acute phase gene regulation. J Exp Med
1995;182:207–17.
8. Liotta LA, Kohn EC. The microenvironment of the
tumour-host interface. Nature 2001;411:375–9.
9. Berens ME, Giese A. ‘‘. . .those left behind.’’ Biology and
oncology of invasive glioma cells. Neoplasia 1999;1:208–19.
10. Zhou Y, Bian X, Le Y, et al. Formylpeptide receptor
FPR and the rapid growth of malignant human gliomas.
J Natl Cancer Inst 2005;97:823–35.
11. Humphrey PA, Wong AJ, Vogelstein B, et al.
Amplification and expression of the epidermal growth
factor receptor gene in human glioma xenografts.
Cancer Res 1988;48:2231–8.
12. Bigner SH, Humphrey PA, Wong AJ, et al. Characterization of the epidermal growth factor receptor in
human glioma cell lines and xenografts. Cancer Res
1990;50:8017–22.
5912
13. Huncharek M, Kupelnick B. Epidermal growth factor
receptor gene amplification as a prognostic marker in
glioblastoma multiforme: results of a meta-analysis.
Oncol Res 2000;12:107–12.
14. Angers-Loustau A, Hering R, Werbowetski TE,
Kaplan DR, Del Maestro RF. SRC regulates actin
dynamics and invasion of malignant glial cells in three
dimensions. Mol Cancer Res 2004;2:595–605.
15. Le Y, Hu J, Gong W, et al. Expression of functional
formyl peptide receptors by human astrocytoma cell
lines. J Neuroimmunol 2000;111:102–8.
16. Le Y, Iribarren P, Zhou Y, et al. Silencing the
formylpeptide receptor FPR by short-interfering RNA.
Mol Pharmacol 2004;66:1022–8.
17. Mischel PS, Shai R, Shi T, et al. Identification of
molecular subtypes of glioblastoma by gene expression
profiling. Oncogene 2003;22:2361–73.
18. Gullick WJ. Prevalence of aberrant expression of the
epidermal growth factor receptor in human cancers. Br
Med Bull 1991;47:87–98.
19. Ozanne B, Richards CS, Hendler F, Burns D,
Gusterson B. Over-expression of the EGF receptor is a
www.aacrjournals.org
Transactivation of EGFR by Formylpeptide Receptor
hallmark of squamous cell carcinomas. J Pathol 1986;
149:9–14.
20. Layfield LJ, Willmore C, Tripp S, Jones C, Jensen RL.
Epidermal growth factor receptor gene amplification
and protein expression in glioblastoma multiforme:
prognostic significance and relationship to other
prognostic factors. Appl Immunohistochem Mol Morphol 2006;14:91–6.
21. Johns TG, Luwor RB, Murone C, et al. Antitumor
efficacy of cytotoxic drugs and the monoclonal antibody
806 is enhanced by the EGF receptor inhibitor AG1478.
Proc Natl Acad Sci U S A 2003;100:15871–6.
22. Rich JN, Reardon DA, Peery T, et al. Phase II trial of
gefitinib in recurrent glioblastoma. J Clin Oncol 2004;22:
133–42.
23. Edwin F, Wiepz GJ, Singh R, et al. A historical
perspective of the EGF receptor and related systems.
Methods Mol Biol 2006;327:1–24.
24. Casas-Gonzalez P, Ruiz-Martinez A, Garcia-Sainz JA.
Lysophosphatidic acid induces a1B-adrenergic receptor
phosphorylation through Ghg, phosphoinositide 3kinase, protein kinase C and epidermal growth factor
receptor transactivation. Biochim Biophys Acta 2003;
1633:75–83.
25. Mori K, Kitayama J, Shida D, Yamashita H, Watanabe
T, Nagawa H. Lysophosphatidic acid-induced effects in
human colon carcinoma DLD1 cells are partially
dependent on transactivation of epidermal growth
factor receptor. J Surg Res 2006;132:56–61.
26. Zhao Y, He D, Saatian B, et al. Regulation of
lysophosphatidic acid-induced epidermal growth factor
receptor transactivation and interleukin-8 secretion in
human bronchial epithelial cells by protein kinase Cy,
Lyn kinase, and matrix metalloproteinases. J Biol Chem
2006;281:19501–11.
27. Vaingankar SM, Martins-Green M. Thrombin aivation
of the 9E3/CEF4 chemokine involves tyrosine kinases
including c-src and the epidermal growth factor
receptor. J Biol Chem 1998;273:5226–34.
28. Neaud V, Duplantier JG, Mazzocco C, Kisiel W,
Rosenbaum J. Thrombin up-regulates tissue factor
www.aacrjournals.org
pathway inhibitor-2 synthesis through a cyclooxygenase-2-dependent, epidermal growth factor receptorindependent mechanism. J Biol Chem 2004;279:5200–6.
29. Kanda Y, Mizuno K, Kuroki Y, Watanabe Y.
Thrombin-induced p38 mitogen-activated protein kinase activation is mediated by epidermal growth factor
receptor transactivation pathway. Br J Pharmacol 2001;
132:1657–64.
30. Beaucage P, Moreau P. EGF receptor transactivation
in angiotensin II and endothelin control of vascular
protein synthesis in vivo . J Cardiovasc Pharmacol 2004;
44 Suppl 1:S20–3.
31. Spinella F, Rosano L, Elia G, Di Castro V, Natali PG,
Bagnato A. Endothelin-1 stimulates cyclooxygenase-2
expression in ovarian cancer cells through multiple
signaling pathways: evidence for involvement of transactivation of the epidermal growth factor receptor.
J Cardiovasc Pharmacol 2004;44:S140–3.
32. Iwasaki H, Eguchi S, Ueno H, Marumo F, Hirata Y.
Endothelin-mediated vascular growth requires p42/p44
mitogen-activated protein kinase and p70 S6 kinase
cascades via transactivation of epidermal growth factor
receptor. Endocrinology 1999;140:4659–68.
33. Kawanabe Y, Masaki T, Hashimoto N. Involvement of
epidermal growth factor receptor-protein tyrosine
kinase transactivation in endothelin-1-induced vascular
contraction. J Neurosurg 2004;100:1066–71.
34. Vacca F, Bagnato A, Catt KJ, Tecce R. Transactivation
of the epidermal growth factor receptor in endothelin-1induced mitogenic signaling in human ovarian carcinoma cells. Cancer Res 2000;60:5310–7.
35. Keely SJ, Uribe JM, Barrett KE. Carbachol stimulates
transactivation of epidermal growth factor receptor and
mitogen-activated protein kinase in T84 cells. Implications for carbachol-stimulated chloride secretion. J Biol
Chem 1998;273:27111–7.
36. Keely SJ, Calandrella SO, Barrett KE. Carbacholstimulated transactivation of epidermal growth factor
receptor and mitogen-activated protein kinase in T(84)
cells is mediated by intracellular Ca(2+), PYK-2, and
p60(src). J Biol Chem 2000;275:12619–25.
5913
37. McCole DF, Keely SJ, Coffey RJ, Barrett KE. Transactivation of the epidermal growth factor receptor in
colonic epithelial cells by carbachol requires extracellular release of transforming growth factor-a. J Biol Chem
2002;277:42603–12.
38. Newton JS. Angiotensin and epidermal growth factor
receptor cross talk goes up and down. J Hypertens 2002;
20:597–8.
39. Moriguchi Y, Matsubara H, Mori Y, et al. Angiotensin
II-induced transactivation of epidermal growth factor
receptor regulates fibronectin and transforming growth
factor-h synthesis via transcriptional and posttranscriptional mechanisms. Circ Res 1999;84:1073–84.
40. De Paolis P, Porcellini A, Savoia C, et al. Functional
cross-talk between angiotensin II and epidermal growth
factor receptors in NIH3T3 fibroblasts. J Hypertens 2002;
20:693–9.
41. Zachary I, Rozengurt E. Modulation of the epidermal
growth factor receptor by mitogenic ligands: effects of
bombesin and role of protein kinase C. Cancer Surv
1985;4:729–65.
42. Madarame J, Higashiyama S, Kiyota H, et al. Transactivation of epidermal growth factor receptor after
heparin-binding epidermal growth factor-like growth
factor shedding in the migration of prostate cancer cells
promoted by bombesin. Prostate 2003;57:187–95.
43. Porcile C, Bajetto A, Barbieri F, et al. Stromal cellderived factor-1a (SDF-1a/CXCL12) stimulates ovarian
cancer cell growth through the EGF receptor transactivation. Exp Cell Res 2005;308:241–53.
44. Prenzel N, Zwick E, Daub H, et al. EGF receptor
transactivation by G-protein-coupled receptors requires
metalloproteinase cleavage of proHB-EGF. Nature 1999;
402:884–8.
45. Luttrell DK, Lee A, Lansing TJ, et al. Involvement of
pp60c-src with two major signaling pathways in human
breast cancer. Proc Natl Acad Sci U S A 1994;91:83–7.
46. Tsai W, Morielli AD, Peralta EG. The m1 muscarinic
acetylcholine receptor transactivates the EGF receptor
to modulate ion channel activity. EMBO J 1997;16:
4597–605.
Cancer Res 2007; 67: (12). June 15, 2007