Download Prostaglandin E2 Promotes Lung Cancer Cell Migration via EP4

Molecular
Cancer
Research
Signaling and Regulation
Prostaglandin E2 Promotes Lung Cancer Cell Migration
via EP4-βArrestin1-c-Src Signalsome
Jae Il Kim, Vijayabaskar Lakshmikanthan, Nicole Frilot, and Yehia Daaka
Abstract
Many human cancers express elevated levels of cyclooxygenase-2 (COX-2), an enzyme responsible for the
biosynthesis of prostaglandins. Available clinical data establish the protective effect of COX-2 inhibition on
human cancer progression. However, despite these encouraging outcomes, the appearance of unwanted side
effects remains a major hurdle for the general application of COX-2 inhibitors as effective cancer drugs. Hence,
a better understanding of the molecular signals downstream of COX-2 is needed for the elucidation of drug
targets that may improve cancer therapy. Here, we show that the COX-2 product prostaglandin E2 (PGE2) acts
on cognate receptor EP4 to promote the migration of A549 lung cancer cells. Treatment with PGE2 enhances
tyrosine kinase c-Src activation, and blockade of c-Src activity represses the PGE2-mediated lung cancer cell
migration. PGE2 affects target cells by activating four receptors named EP1 to EP4. Use of EP subtype-selective
ligand agonists suggested that EP4 mediates prostaglandin-induced A549 lung cancer cell migration, and this
conclusion was confirmed using a short hairpin RNA approach to specifically knock down EP4 expression.
Proximal EP4 effectors include heterotrimeric Gs and βArrestin proteins. Knockdown of βArrestin1 expression
with shRNA significantly impaired the PGE2-induced c-Src activation and cell migration. Together, these results
support the idea that increased expression of the COX-2 product PGE2 in the lung tumor microenvironment
may initiate a mitogenic signaling cascade composed of EP4, βArrestin1, and c-Src which mediates cancer cell
migration. Selective targeting of EP4 with a ligand antagonist may provide an efficient approach to better manage patients with advanced lung cancer. Mol Cancer Res; 8(4); 569–77. ©2010 AACR.
Introduction
Cancer diseases claim over half a million lives in the
United States annually, and lung cancer is the number
one cause of death in both men and women. Limited success in the effectiveness of lung cancer treatment is due in
part to the cancer cells' ability to spread and metastasize
very early in the disease course. Accumulating epidemiologic and clinical data provide a strong link between
inflammation and cancer initiation or progression, but
the molecular inflammatory determinants remain to be
established. Nonetheless, the importance of the tumor
microenvironment and inflammation in neoplastic pro-
Authors' Affiliation: Department of Pathology, Medical College of
Georgia, Augusta, Georgia
Note: Present address for V. Lakshmikanthan: Department of Pharmacology, University of Louisiana, Monroe, Louisiana.
Present address for Y. Daaka: Department of Urology and University
of Florida Prostate Disease Center, University of Florida College of
Medicine, Gainesville, Florida.
Corresponding Author: Yehia Daaka, Department of Urology, University of Florida College of Medicine, 2033 Mowry Road, CGRC Room 462,
P.O. Box 103633, Gainesville, FL 32610. Phone: 352-273-8112; Fax:
352-273-8335. E-mail: [email protected]
doi: 10.1158/1541-7786.MCR-09-0511
©2010 American Association for Cancer Research.
www.aacrjournals.org
gression is evident from studies of cancer risk among nonsteroidal anti-inflammatory drug users, who experience
reduced risk for many types of cancers (1).
One of the primary mechanisms underlying the chemopreventive effects of nonsteroidal anti-inflammatory drugs
lies in their ability to inhibit cyclooxygenase enzymes
COX-1 and COX-2 activation. Whereas COX-1 is expressed constitutively, the COX-2 protein is usually not
detected in normal tissues but is instead inducible by cytokines and growth factors at sites of inflammation (2-6).
Indeed, COX-2 protein levels are elevated in several cancer
types, including colorectal, prostate, and lung cancers
(7, 8), and suppression of either COX-2 expression or
COX-2 activation, may be effective in cancer prevention
and therapy as it promotes the repression of a variety of
cancer hallmark traits such as angiogenesis and metastasis
(4, 9, 10). Alas, in spite of hopeful results, the long-term
use of selective COX-2 inhibitors as an effective therapeutic approach to manage cancer progression has been questioned due to unwanted side effects such as increased
cardiovascular risks (8-13).
COX-2 uses plasma membrane–expressed arachidonic
acid as a substrate to generate lipid mediators that are rapidly converted to prostaglandins (PG), i.e., PGD2, PGE2,
PGF2, PGI2, and TxA2. The prostaglandins exert important biological effects in target organs, such as in the regulation of immune function, gastrointestinal homeostasis,
569
Kim et al.
and inflammation (4). It is hypothesized that the cardiovascular risks associated with COX-2–selective inhibition may
result, at least in part, from an imbalance created between
PGI2 and TxA2, both of which possess key physiologic
roles in vasoregulation and platelet aggregation (1, 8,
12). Hence, a better understanding of COX-2 signaling,
and identification of its downstream effector(s), is essential
for developing effective drugs that aim to circumvent the
risk of unwanted cardiovascular events associated with the
selective inhibition of COX-2.
Of the five prostaglandins produced by COX-2, PGE2 is
the predominant one associated with cancer (14-20). Four
receptor subtypes that belong to the seven transmembrane–spanning G protein–coupled receptor (GPCR) superfamily are known to bind PGE2, and they are named
EP1 to EP4. Upon binding PGE2, each EP subtype transduces signals through distinct heterotrimeric G proteins:
EP1 signals through Gq, EP2 and EP4 signal through
Gs, and EP3 signals through Gi. In addition, stimulation
with PGE2 may potentiate Wnt signaling or transactivate
other types of receptors, including receptor tyrosine kinases
(1), albeit by mechanisms not fully understood. In this
study, we examined the role of PGE2 in lung cancer cell
migration. In A549 lung cancer alveolar cells, COX-1 is
constitutively expressed whereas COX-2 expression is
inducible (5). The A549 cells secrete little PGE2, and
instead, it seems that COX-2 induction is largely responsible for PGE2 production (5, 6). Using A549 cells as a
model, we found that these cells express all four EP subtypes but that only selective activation of EP4 could lead to
increased cell migration. Furthermore, we found that stimulation with PGE2 could activate the non–receptor tyrosine kinase c-Src which is required for PGE2-induced
A549 cell migration. Remarkably, PGE2-mediated c-Src
activation and A549 cell migration failed when the expression of the adaptor protein βArrestin1 was knocked down
with short hairpin RNA (shRNA). These results provide
a rationale for considering the EP4-βArrestin1 signalsome
as an effective drug target to more effectively treat patients
with advanced lung cancer.
Materials and Methods
Cell Culture, Antibodies, and Reagents
A549 and HEK293T cells were obtained from the
American Type Culture Collection. A549 cells were
maintained in F12K medium containing 10% fetal bovine serum, 100 units/mL of penicillin, and 100 μg/
mL of streptomycin at 37°C with 5% CO2. F12K medium supplemented with 0.1% bovine serum albumin
(BSA), 100 units/mL of penicillin, and 100 μg/mL of
streptomycin was used as starvation medium for the
A549 cells. HEK293T cells were cultured in DMEM
supplemented with 10% fetal bovine serum, 100 units/
mL of penicillin, and 100 μg/mL of streptomycin at
37°C with 5% CO2. All cells were maintained at 60%
to 80% confluency, except for cells in the wound-healing
assay (95-100% before wounding). PGE2, butaprost, sul-
570
Mol Cancer Res; 8(4) April 2010
prostone, and PGE1-OH were purchased from Cayman
Chemicals; BSA, polybrene, and puromycin were from
Sigma Aldrich; TRIzol and epidermal growth factor were
from Invitrogen; FuGENE HD transfection reagent was
from Roche Applied Science; F12K and DMEM were
from Cellgro; and fetal bovine serum was from HyClone.
Antibodies used were anti–c-Src sc-18, anti-ERK2 sc-154
(Santa Cruz); anti–p-Tyr 4G10 (Millipore); antiGAPDH, anti–β-actin, and anti-p44/42 (Cell Signaling);
anti–EP1-4 (Cayman Chemicals); and anti-FAK (BD
Biosciences).
Cell Proliferation
Cell proliferation was determined by counting cells
that excluded trypan blue (Invitrogen) with a hemacytometer, and treated under the same conditions described for wound healing assay (described herein).
Briefly, cells were trypsinized and resuspended in growth
medium, aliquots prepared in a 1:1 ratio of 2× trypan
blue, and cells counted using a hemacytometer. An equal
number of cells (5 × 105 cells/well) was resuspended in
medium containing the indicated agents and seeded in
six-well plates. These experiments were done thrice in
duplicate for analysis.
Immunoblotting and Immunoprecipitation Assays
Cells grown on 100 mm dishes were washed with PBS
and harvested in lysis buffer containing 25 mmol/L of Tris
(pH 8.0), 100 mmol/L of NaCl, 1% (v/v) Triton X-100,
10% (v/v) glycerol, 1 mmol/L of EDTA, 1 mmol/L of phenylmethylsulfonyl fluoride, 10 μg/mL of aprotinin, 10 μg/
mL of leupeptin, and 2 μg/mL of pepstatin A. For detection of phosphorylated proteins, phosphatase inhibitor
mixture (P1 and P2, 1:1,000; Sigma) and 1 mmol/L of sodium orthovanadate were included in the lysis buffer. For
immunoprecipitation, the cell lysates were cleared by centrifugation at 10,000 × g for 15 min at 4°C, incubated with
anti–c-Src antibody overnight, and the proteins were precipitated and eluted with 70 μL of a 30% slurry of protein
G plus/protein A-agarose beads (Calbiochem) and rotated
for 4 h at 4°C. Immune complexes were washed thrice
with ice-cold glycerol lysis buffer containing 0.1% (w/v)
SDS, and the samples were then denatured in Laemmli
sample buffer by boiling for 5 min. Before immunoprecipitation, 50 μL of the cell lysate were aliquoted to provide
samples for c-Src expression determined by immunoblotting. Proteins (immunoprecipitated with c-Src antibody
as well as from total cell lysates) were resolved on 10%
SDS-polyacrylamide gels and transferred to nitrocellulose
membranes (Bio-Rad), and immunoblotted with 1:1,000
dilutions of primary antibodies for the protein of interest.
Horseradish peroxidase–conjugated anti-rabbit/mouse secondary antibodies (Jackson ImmunoResearch) were diluted 1:10,000 in blocking buffer (3% BSA-TBST) and blots
incubated for 1 h at room temperature. All antibodies were
diluted in 3% BSA-TBST. Immunoblots were visualized
using Amersham ECL plus chemiluminescence reagent
(GE Healthcare).
Molecular Cancer Research
PGE2 Induces Lung Cancer Cell Migration
Wound Healing Assay
A549 cells were seeded in six-well plates and incubated overnight in starvation medium. Cell monolayers
were wounded with a sterile 200-μL pipette tip, washed
with starvation medium to remove detached cells, and
stimulated with the indicated agonist for 24 h. Where
indicated, antagonists were added to the cells 30 min
before stimulation. Next, cells were fixed in 4% formalin, stained with hematoxylin, counterstained with bluing reagent (Richard-Allan Scientific), washed with
water, and air dried. Cells were photographed using a
digital camera (Nikon) integrated in a bright-field microscope (Zeiss).
Cell Migration
Cell migration assays were done using 8-μm-pore
transwell chambers (Costar). Briefly, A549 cell monolayers were detached using HyQtase (Hyclone), resuspended in starvation medium, and cells added to the
transwell chambers (2 × 105 cells/well). Starvation medium containing the indicated agonists was added to the
lower chambers, and transwells placed in contact with
the agonists at 37°C for 5 h. Where indicated, the antagonists were added to cells 30 min before stimulation.
Cells that migrated to the lower surface of the filter were
fixed with 3.7% (w/v) paraformaldehyde, stained with
0.5% (w/v) crystal violet, and migrated cells counted
through a 400× objective with an Axioskop microscope
(Zeiss).
Knockdown shRNA
For stable knockdown by shRNA, we screened five lentiviral DNA expression vector sets that each contained 21 nucleotide shRNA duplexes that targeted our genes of interest
(Open Biosystems). The shRNA vectors were cotransfected
with equal concentrations of helper plasmids VSVG and delta
8.9 into HEK293T cells for lentivirus production using
FuGENE HD. As a negative control, HEK293T cells were
transfected with the same amount of expression vector
containing the green fluorescent protein targeting sequence,
5′-GCA AGC TGA CCC TGA AGT TCA T-3′. Viruscontaining medium from transfected HEK293T cells were
collected 24 h after transfection and mixed with 5.0 mg of
polybrene for infection of A549 cells. HEK293T cells were
replenished with fresh medium for another 24 h, and
the virus-containing medium was collected for a second infection with polybrene. Cells were infected for 2 h, and
grown in F12K medium containing 10% fetal bovine serum
for 48 h prior to selection with puromycin (5 μg/mL). Uninfected A549 cells were used as a selection control. Cells
were maintained in puromycin (1 μg/mL), and protein expression of target genes was determined by Western blotting.
The cells with optimal knockdown (>50% reduction)
were used for experiments (Open Biosystems: ARRB1
shRNA clone TRCN0000000161, PTGER4 shRNA clone
TRCN0000000204).
PCR Amplification
Total RNA was isolated with TRIzol Reagent, as described by the manufacturer (Invitrogen). The RNA was
FIGURE 1. PGE2 enhances the lung cancer cell
migration. A, effect of PGE2 on wound healing.
A549 cells were incubated in starvation medium
(F12K medium supplemented with 0.1% BSA)
overnight prior to wounding and then incubated
with the same medium containing either vehicle
DMSO (0.2% v/v) or PGE2 (1 and 10 μmol/L).
Cell migration was observed 24 h after
stimulation. B, PGE2 causes an increase in
migratory cells through transwell. Cells were
grown in starvation medium overnight prior
to detachment, washed with PBS, and
resuspended in the same medium. Cells
(2 × 105) were seeded on transwells. The
transwells were then placed onto chambers
containing the indicated concentrations of
PGE2 for 5 h. Experiments were repeated thrice
and five fields were randomly selected and used
to count migrated cells. Data are shown as the
average of three independent experiments.
C, PGE2 does not affect cell proliferation. A549
cells were incubated in starvation medium
overnight prior to treatment with PGE2
(10 μmol/L) or serum (10% v/v) for an additional
24 h. The number of viable cells that
exclude trypan blue was counted with a
hemocytometer. These experiments were
repeated thrice in duplicate and the data are
presented as fold increase over vehicle-treated
control. *, P < 0.05 compared with
nonstimulated control.
www.aacrjournals.org
Mol Cancer Res; 8(4) April 2010
571
Kim et al.
FIGURE 2. Prostaglandins induce EP4-dependent cell migration. A, lung cancer A549 cells express all four EP subtype receptors. Two cell lysates
were taken from independent harvests and analyzed by Western blotting using EP subtype–specific antibodies. B and C, PGE1-OH promotes A549 cell
migration. Cells were treated, or not, with 1 μmol/L of the indicated selective agonists of the cognate EP subtypes for 24 h (B) or 5 h (C), and inspected
as detailed under Fig. 1. Sulprostone (Sulp) is selective for EP1 and EP3, butaprost (Buta) is selective for EP2, and PGE1-OH (POH) is selective for
EP4. D, EP4 mediates prostaglandin-induced lung cancer cell migration. A549 cells were infected with lentivirus-encoding shRNA that specifically targets
the EP4 or green fluorescence (used as control) genes for 48 h, and selected with puromycin. Surviving cells were examined for expression of EP4
protein using Western blot analysis (inset). Cells stably expressing shRNA-GFP or shRNA-EP4 were treated with PGE2 (1 μmol/L) or PGE1-OH (1 μmol/L)
for 5 h and cell migration was analyzed using the transwell assay, as described. These experiments were repeated thrice in duplicate and the data are
presented as fold increase over vehicle-treated control. *, P < 0.05 compared with vehicle control.
reverse transcribed (Invitrogen RT-Superscript III Kit) to
prepare the cDNA to be amplified by PCR, using specific
sense and antisense primers for the ARRB1, ARRB2, and
housekeeping gene GAPDH. Primers were synthesized by
Integrated DNA Technologies as follows: βArrestin1 exon
8 (forward GGT AAT AGA TCT CCT TAT CC; reverse
CCA CAA GCG GAA TTC TGT G), βArrestin2 exon 9
(forward GTA CTG TCT CAC AGA GAC TTT; reverse
GAC AAG GAG CTG TAC TAC CA), and GAPDH (forward CAT GGG TGT GAA CCA TGA GAA; reverse GGT
CAT GAG TCC TTC CAC GAT).
Statistics and Quantifications
Data are presented as the mean ± SE of three independent experiments. The data was analyzed for one-way
572
Mol Cancer Res; 8(4) April 2010
ANOVA and followed by the Tukey post-test to determine
the statistical significance. All statistical analyses were done,
and all graphs generated, using GraphPad Prism 5.0 software (GraphPad). The x and y labels of all presented data
were prepared using Adobe Illustrator CS2 (Adobe).
Results
PGE2 Increases Migration of Lung Adenocarcinoma
A549 Cells
PGE2 regulates cancer cell proliferation and survival by
multiple mechanisms, and we asked if it also affects cell
migration. Exposure to PGE2 enhanced the migration of
alveolar cell carcinoma A549 cells in a dose-dependent
manner detectable by both wound healing (Fig. 1A) and
Molecular Cancer Research
PGE2 Induces Lung Cancer Cell Migration
transwell migration (Fig. 1B) assays. Because A549 cells are
capable of migrating under basal conditions and PGE2 can
increase the cancer cell proliferation (21-23), we tested
whether this lipid mediator could actually be increasing
the number of basal migratory cells. Under the conditions
used in our study, stimulation with PGE2 did not seem to
significantly affect the rate of A549 cell proliferation in
comparison with vehicle-treated cells (Fig. 1C). We therefore conclude that PGE2 specifically augments the inherent
migration character of lung cancer A549 cells.
EP4 Mediates the Cell Migration
PGE2 binds to and activates four distinct receptor subtypes named EP1 to EP4. Upon activation by PGE2, EP2
and EP4 couple to heterotrimeric Gs proteins that activate
adenylate cyclase to synthesize cyclic AMP. EP1 and EP3,
on the other hand, transmit signals via Gq and Gi proteins
to transiently increase intracellular Ca2+ levels or inhibit
adenylate cyclase, respectively (16, 24, 25). To begin to
elucidate the mechanisms involved in PGE2-mediated lung
cancer cell migration, we asked which EP subtype(s) are
expressed in the A549 cells and are responsible for mediating the cell migration. Using Western blot analysis, we
were able to show the expression of all four EP subtypes
(Fig. 2A). Next, we used selective EP subtype ligand agonists and show that A549 cells stimulated with EP1/EP3selective sulprostone or EP2-selective butaprost failed to
migrate (Fig. 2B and C). Distinctly, the selective activation
of EP4 with PGE1-OH significantly enhanced the migra-
tion of A549 cells as measured by wound healing (Fig. 2B)
or transwell (Fig. 2C) migration assays.
Because PGE2-mediated activation of EP4 signaling
does not necessarily exclude activation of the other EPs,
we confirmed the selective role of EP4 in the PGE2-induced
cell migration by using shRNA. A549 cells stably expressing shRNA targeting the PTGER4 gene yielded ∼60%
reduction in the EP4 protein expression (Fig. 2D, inset),
and these cells failed to migrate following stimulation with
either PGE2 or PGE1-OH (Fig. 2D). The EP4 knockdown
did not affect basal cell migration, suggesting that EP4
only regulates ligand agonist-mediated A549 cell migration. Because PGE2 treatment of EP4 knockdown cells
does not exclude the availability of EP1 to EP3 for activation but failed to enhance cell migration, we conclude that
selective activation of EP4 is central to PGE2-provoked
migration of the A549 cells.
PGE2-Induced Cell Migration is Independent of
Extracellular Signal-Regulated Kinase Activation
Growth factors and cytokines transmit oncogenic signals
by activating the Ras-extracellular signal regulated kinase 1
and 2 (ERK) signalsome (26). The ERK may be activated
in response to GPCR signaling, and ERK activity is
required for the migration of many cancer cell types
(26-29). In the case of A549 lung cancer cells, PGE2 acts
primarily on EP4 to enhance cell migration, and we tested
whether treatment with PGE2 could affect the ERK activation state. The results show that treatment with either
PGE2 (data not shown) or PGE1-OH (Fig. 3A) leads to
FIGURE 3. ERK activation is not required for the prostaglandin-enhanced cell migration. A, PGE1-OH promotes time-dependent phosphorylation of
ERK in A549 cells. Cells were incubated in starvation medium overnight and treated with PGE1-OH (1 μmol/L) for the indicated times. Cell monolayers were
washed once with ice-cold PBS, lysed with radioimmunoprecipitation assay buffer, and analyzed by Western blotting using anti-phosphorylated ERK
antibodies. The same filter was also analyzed for the expression of total ERK2 to control for equal protein loading in each lane. B, PD98059 inhibits the
PGE2-mediated ERK phosphorylation. Cells were incubated in starvation medium overnight and resuspended in medium containing either vehicle or the
indicated concentrations of PD98059 for 30 min prior to treatment, or not, with PGE2 (1 μmol/L) for 10 min. ERK phosphorylation was determined as
described above. C, activated ERK does not affect PGE2-mediated lung cancer cell migration. A549 cells were pretreated with PD98059 (50 μmol/L) for
30 min and analyzed for cell migration using the transwell assay. These experiments were repeated thrice in duplicate and the data are presented as fold
increase over vehicle-treated control. *, P < 0.05 compared with vehicle control.
www.aacrjournals.org
Mol Cancer Res; 8(4) April 2010
573
Kim et al.
ERK phosphorylation in a biphasic manner reminiscent of
similar activation patterns seen with ligand-induced stimulation of other GPCRs (30). The existence of both rapid
and sustained ERK activation has been explained in previous studies to be mediated by G protein–dependent or
βArrestin-dependent pathways, respectively (27, 28).
Based on this knowledge, we conclude that stimulation
with PGE2 promotes signaling through both G proteins
and βArrestin.
To determine if PGE 2 -induced ERK activation is
responsible for the A549 cell migration, we used
PD98059, a small molecule MEK inhibitor, to suppress
ERK phosphorylation. Pretreatment with the PD98059
for 30 minutes inhibited the PGE2-induced ERK phosphorylation (Fig. 3B) and basal level of cell migration
(Fig. 3C), but failed to affect the PGE2-mediated cell
migration (Fig. 3C). PD98059 (10, 25, or 50 μmol/L)
exerted no cytotoxic effect on A549 cells (data not shown).
Together, these results indicate that the activation of ERK
by PGE2 stimulation does not play a significant role in the
A549 lung cancer cell migration.
PGE2 Induces Cell Migration via βArrestin1 and Src
Activation
The cellular Src (c-Src) tyrosine kinase has wellestablished roles in the progression of many different human
cancers (31-34). In particular, an increase in c-Src protein
levels and/or tyrosine kinase activity has been shown to promote tumor cell metastasis, whereas inhibition of c-Src activation leads to decreased tumor cell migration and invasion
(35). Our results show that stimulation with PGE2 promotes the latent ERK phosphorylation (Fig. 3A) that is likely mediated through βArrestin-dependent signals (30), and
we have previously reported that βArrestin1 directly interacts with c-Src and increases the c-Src tyrosine kinase activity (36). We show here that PGE2 treatment similarly causes
an increase in c-Src tyrosine phosphorylation in a timedependent manner in the A549 lung adenocarcinoma cells
(Fig. 4A). Importantly, blockade of c-Src activation by pretreatment with the small molecule c-Src inhibitor PP2
(Fig. 4B) completely abolished the ability of PGE2 to promote A549 cell migration (Fig. 4C and D), but showed no
effect on A549 cell viability (data not shown). These results
FIGURE 4. PGE2 promotes c-Src–dependent lung cancer cell migration. A, PGE2 promotes c-Src tyrosine phosphorylation. A549 cells were incubated in
starvation medium overnight prior to treatment with PGE2 (1 μmol/L) for the indicated times or with epidermal growth factor (EGF; 10 ng/mL) for 5 min.
Cell monolayers were washed once with ice-cold PBS and lysed with radioimmunoprecipitation assay buffer. Protein extracts were quantified by the
Bradford method, and 1 mg of total protein was used for immunoprecipitation with anti–c-Src antibody. Immunoprecipitates were washed, separated on
SDS-PAGE, and analyzed by Western blotting with anti–phosphorylated tyrosine (4G10) antibody (top). Total cell lysates were probed with anti–c-Src
antibody (bottom). B, PP2 inhibits the PGE2-mediated c-Src phosphorylation. Cells were incubated in starvation medium overnight and pretreated with PP2
at the indicated concentrations for 30 min prior to treatment with PGE2 (1 μmol/L) for 5 min. Immunoprecipitates were probed for tyrosine phosphorylation of
c-Src, as described. C and D, PP2 inhibits the PGE2 mediated cell migration. A549 cells were pretreated with PP2 (10 μmol/L) for 30 min prior to treatment
with PGE2 or epidermal growth factor (10 ng/mL) and analyzed for migration using the wound healing (C) and transwell (D) assays. Experiments were
repeated thrice in duplicate and data are presented as fold increase above vehicle-treated samples. *, P < 0.05 compared with vehicle control.
574
Mol Cancer Res; 8(4) April 2010
Molecular Cancer Research
PGE2 Induces Lung Cancer Cell Migration
FIGURE 5. βArrestin1 is required for PGE2-induced activation of c-Src and migration. A and B, βArrestin1 gene and protein is expressed in A549 cells.
Total RNA was isolated using the TRIzol method and reverse-transcribed for PCR amplification using specific primers for βArrestin1, βArrestin2, and
GAPDH (A). A549 cells were infected with lentivirus-encoding shRNA that specifically targets βArrestin1 (shβArr1) or green fluorescence (shGFP) gene as
described under Fig. 2. Cells were examined for expression of βArrestin1 using Western blot analysis (B). C, βArrestin1 mediates the prostaglandin-induced
activation of c-Src. A549-shGFP and A549-shβArr1 cells were incubated in starvation medium and treated with the indicated concentrations of PGE2 or
with epidermal growth factor (EGF; 10 ng/mL) for 5 min (positive control). Immunoprecipitates were probed for tyrosine phosphorylation of c-Src as
described under Fig. 4. D, βArrestin1 mediates PGE2-induced migration in A549 cells. Cells were incubated in starvation medium overnight and treated with
PGE2 (1 μmol/L) for transwell assays (5 h). These experiments were repeated thrice in duplicate and the data are presented as fold over vehicle-treated
control. *, P < 0.05 compared with control.
show that activation of c-Src by PGE2 is critical for the migration of lung cancer A549 cells.
Desensitization of GPCR signaling occurs in a stepwise
manner that involves phosphorylation of the agonist-bound
receptor by GPCR kinases and subsequent binding of βArrestins to the receptor. With βArrestins bound, the GPCRs
are physically uncoupled from further interaction with G
proteins. In addition to desensitization, the βArrestins
could act as scaffolds for a diverse array of signaling pathways and are increasingly becoming appreciated as critical
regulators of directed cell chemotaxis and migration (37).
Our results show that stimulation with PGE2 activates cSrc (Fig. 4A), and βArrestin1 interacts with c-Src and activates it (36). Hence, we established if βArrestin1 plays a role
in the PGE2-induced migration of lung cancer A549 cells.
First, we documented the expression of the βArrestin1 gene
(Fig. 5A) and protein (Fig. 5B) in these cells. Next, we used
a lentiviral shRNA delivery system to create a stable A549
cell line with reduced βArrestin1 expression (Fig. 5B). Having shown that c-Src activation is required for PGE 2 induced cell migration (Fig. 4D), and that βArrestin1
potentiates c-Src activation (36), we tested the effect of
βArrestin1 knockdown on c-Src activation in A549 cells.
Our results show that the knockdown of βArrestin1 expression attenuates the PGE2-induced tyrosine phosphorylation of c-Src (Fig. 5C) and significantly suppresses cancer
cell migration (Fig. 5D). Collectively, these results indicate
that in A549 cells, βArrestin1 is essential for the PGE2mediated activation of c-Src and increase in cell migration.
Discussion
COX-2 expression is increased in human lung cancer tissues (37) and it is a principal enzyme in the biosynthesis of
www.aacrjournals.org
the prostaglandins, including PGE2 (36). Under physiologic conditions, PGE2 activates four cognate receptors that
are collectively involved in the regulation of cellular homeostasis. However, uncontrolled signaling by the PGE2 receptors has been implicated in the initiation and progression of
several human diseases including cancer, but the mechanisms involved remain to be uncovered. In this study, evidence is presented to show that PGE2 activates its cognate
EP4 receptor subtype to induce the migration of lung
cancer A549 cells by activating c-Src through a mechanism that involves the adaptor protein βArrestin1 (Fig. 6).
The nature and magnitude of receptor-induced signals
are directly linked to the production and availability of ligand agonists. The production of prostanoids is initiated
by cytosolic phospholipase A2–mediated release of arachidonic acid from phospholipids (38). The liberated arachidonic acid can be oxygenated via the cyclic prostaglandin
synthase COX-2 pathway into intermediary lipids that are
converted into individual prostaglandins by respective
synthases (4). Significantly, several types of human cancers
express elevated levels of COX-2 and outcome data show
the protective effects of pharmacologic COX-2 inhibitors
in preventing tumor induction as well as inhibiting tumor
growth, providing a rationale to use COX-2 inhibitors as
cancer therapeutics. However, long-term clinical use of selective COX-2 inhibitors has been questioned due to associated health risks (39). In this report, we show that the
COX-2 product PGE2, the expression of which is elevated
in human cancers (7, 8, 40), is efficient at inducing lung
cancer cell migration. These results suggest that the specific
targeting of PGE2 signals may provide a way to harness the
beneficial effects of inhibiting COX-2 activity (i.e., limit
tumor induction and progression) but circumvent the
COX-2 inhibitor-associated unwanted health risks.
Mol Cancer Res; 8(4) April 2010
575
Kim et al.
FIGURE 6. A model of PGE2-mediated migration in A549 cells. See text
for details.
Over the past couple of decades, it has become evident
that GPCRs, like those activated by PGE2, may elicit mitogenic responses at least in part by activating the ERK
mitogen-activated protein kinase pathway. Elucidation of
the numerous protein interactions that regulate ERK activation has made ERK one of the best-characterized growth
factor–mediated mitogenic signaling pathways, and ERK
remains a key target of cancer treatment (41). Different
classes of G proteins, including Gs, Gi, and Gq, mediate
ERK activation by PGE2 receptors. Our results show that
all four EP subtypes transduce signals to activate ERK in
lung cancer A549 cells (using PGE2, sulprostone, butaprost, or PGE1-OH), but activated ERK does not affect
the PGE2-mediated cell migration. Because activation of
each EP subtype with either its selective agonist or with
PGE2 could increase ERK phosphorylation, and only
EP4 activation could increase cell migration, we deduced
the selective importance of the PGE2-EP4 axis for lung
cancer cell migration. These results provide a rationale
for the blockade of EP4 signaling events with specific ligand antagonists as a therapeutic modality to interfere with
lung cancer progression.
The classic signal transduction from GPCR to G proteins
is rapidly desensitized in a highly conserved process that involves phosphorylation of agonist-occupied receptors by
GPCR kinases. The GPCR kinases phosphorylate residues
in the intracellular loop and carboxyl tail of the receptor
leading to the recruitment of cytosolic βArrestins to the
receptor. The binding of βArrestins to GPCR kinase–
phosphorylated receptors sterically hinders receptorinduced G protein activation. In addition to inhibiting the
receptor-mediated G protein activation, βArrestins may also
mediate G protein–independent signals (26, 27). This can
be explained in part by the knowledge that βArrestins are
multifunctional proteins that, in addition to their classic
roles as GPCR-desensitizing molecules, could act as scaffolds and adaptor proteins that facilitate a growing number
of signaling events. For example, we previously showed that
βArrestin1 mediates the β2 adrenergic receptor–induced
activation of c-Src by forming a receptor-anchored signalsome containing βArrestin1 and c-Src (36). A subsequent
study showed that PGE2 uses a similar mechanism to induce
colon cancer cell migration, but the identity of the EP subtype responsible has not been elucidated (25). EP4 has been
shown to be involved in colon carcinoma cell motility,
growth, and anchorage independence (42-44). In this study,
we showed that PGE2 promotes the c-Src activation via
EP4, providing a rationale for its use as a therapeutic target
to treat patients with advanced lung cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant Support
NIH grants CA129155 and CA131988, and the Georgia Cancer Coalition
Distinguished Scholar fund (Y. Daaka).
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.
Received 11/25/2009; revised 01/22/2010; accepted 02/25/2010; published
OnlineFirst 03/30/2010.
References
1.
2.
3.
4.
5.
576
Cha YI, DuBois RN. NSAIDs and cancer prevention: targets downstream of COX-2. Annu Rev Med 2007;58:239–52.
Vane JR, Bakhle YS, Botting RM. Cyclooxygenases 1 and 2. Annu
Rev Pharmacol Toxicol 1998;38:97–120.
Sheehan KM, Sheahan K, O'Donoghue DP, et al. The relationship
between cyclooxygenase 2 expression and colorectal cancer. JAMA
1999;282:1254–7.
Wang MT, Honn KV, Nie D. Cyclooxygenases, prostanoids, and tumor progression. Cancer Metastasis Rev 2007;26:525–34.
Patel KM, Wright KL, Whittaker P, Chakravarty P, Watson ML, Ward
SG. Differential modulation of COX-2 expression in A549 airway ep-
Mol Cancer Res; 8(4) April 2010
6.
7.
8.
ithelial cells by structurally distinct PPARγ agonists: evidence for disparate functional effects which are independent of NF-κB and
PPARγ. Cell Signal 2005;17:1098–110.
Backhus LM, Petasis NA, Uddin J, et al. Dimethyl celecoxib as a novel non-cyclooxygenase 2 therapy in the treatment of non-small cell
lung cancer. J Thorac Cardiovasc 2005;130:1406–12.
Ristimaki A, Sivula A, Lundin J, et al. Prognostic significance of elevated cyclooxygenase-2 expression in breast cancer. Cancer Res
2002;62:632–5.
Fitzgerald GA. Coxibs and cardiovascular disease. N Engl J Med
2004;351:1709–11.
Molecular Cancer Research
PGE2 Induces Lung Cancer Cell Migration
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Arber N. Cyclooxygenase-2 inhibitors in colorectal cancer prevention. Cancer Epidemiol Biomarkers Prev 2008;17:1852–7.
Arber N, Eagle CJ, Spicak J, et al. Celecoxib for the prevention of
colorectal adenomatous polyps. N Engl J Med 2006;355:885–95.
Lenzer J. FDA advisers warn: COX-2 inhibitors increase risk of heart
attack and stroke. BMJ (Clin Res Ed) 2005;330:440.
Solomon SD, Pfeffer MA, McMurray JJ, et al. Effect of celecoxib on
cardiovascular events and blood pressure in two trials for the prevention of colorectal adenomas. Circulation 2006;114:1028–35.
Zheng Y, Ritzenthaler JD, Sun X, Roman J, Han S. Prostaglandin E2
stimulates human lung carcinoma cell growth through induction of
integrin-linked kinase: the involvement of EP4 and Sp1. Cancer
Res 2009;69:896–904.
Krysan K, Reckamp KL, Dalwadi H, et al. Prostaglandin E2 activates
mitogen-activated protein kinase/Erk pathway signaling and cell proliferation in non-small cell lung cancer cells in an epidermal growth
factor receptor-independent manner. Cancer Res 2005;65:6275–81.
Zhang Y, Liu Q, Zhang M, Yu Y, Liu X, Cao X. Fas signal promotes
lung cancer growth by recruiting myeloid-derived suppressor cells
via cancer cell-derived PGE2. J Immunol 2009;182:3801–8.
Alaa M, Suzuki M, Yoshino M, et al. Prostaglandin E2 receptor 2
overexpression in squamous cell carcinoma of the lung correlates
with p16INK4A methylation and an unfavorable prognosis. Int J Oncol 2009;34:805–12.
Zhu YM, Azahri NS, Yu DC, Woll PJ. Effects of COX-2 inhibition on
expression of vascular endothelial growth factor and interleukin-8 in
lung cancer cells. BMC Cancer 2008;8:218.
Yang L, Huang Y, Porta R, et al. Host and direct antitumor effects
and profound reduction in tumor metastasis with selective EP4 receptor antagonism. Cancer Res 2006;66:9665–72.
Murakami M, Nakatani Y, Tanioka T, Kudo I. Prostaglandin E
synthase. Prostaglandins Other Lipid Mediat 2002;68–69:383–99.
Yoshimatsu K, Altorki NK, Golijanin D, et al. Inducible prostaglandin
E synthase is overexpressed in non-small cell lung cancer. Clin Cancer Res 2001;7:2669–74.
Cherukuri DP, Chen XB, Goulet AC, et al. The EP4 receptor antagonist, L-161,982, blocks prostaglandin E2-induced signal transduction
and cell proliferation in HCA-7 colon cancer cells. Exp Cell Res 2007;
313:2969–79.
Robertson FM, Simeone AM, Mazumdar A, et al. Molecular and pharmacological blockade of the EP4 receptor selectively inhibits both
proliferation and invasion of human inflammatory breast cancer cells.
J Exp Ther Oncol 2008;7:299–312.
Loffler I, Grun M, Bohmer FD, Rubio I. Role of cAMP in the promotion
of colorectal cancer cell growth by prostaglandin E2. BMC Cancer
2008;8:380.
Takahashi T, Uehara H, Bando Y, Izumi K. Soluble EP2 neutralizes
prostaglandin E2-induced cell signaling and inhibits osteolytic tumor
growth. Mol Cancer Ther 2008;7:2807–16.
Buchanan FG, Gorden DL, Matta P, Shi Q, Matrisian LM, DuBois RN.
Role of βArrestin1 in the metastatic progression of colorectal cancer.
Proc Natl Acad Sci U S A 2006;103:1492–7.
Sebolt-Leopold JS, Herrera R. Targeting the mitogen-activated protein kinase cascade to treat cancer. Nat Rev 2004;4:937–47.
www.aacrjournals.org
27. Shenoy SK, Drake MT, Nelson CD, et al. βArrestin-dependent, G
protein-independent ERK1/2 activation by the β2 adrenergic receptor. J Biol Chem 2006;281:1261–73.
28. Ahn S, Shenoy SK, Wei H, Lefkowitz RJ. Differential kinetic and spatial patterns of βArrestin and G protein-mediated ERK activation by
the angiotensin II receptor. J Biol Chem 2004;279:35518–25.
29. Lester RD, Jo M, Campana WM, Gonias SL. Erythropoietin promotes
MCF-7 breast cancer cell migration by an ERK/mitogen-activated
protein kinase-dependent pathway and is primarily responsible for
the increase in migration observed in hypoxia. J Biol Chem 2005;
280:39273–7.
30. Wei H, Ahn S, Shenoy SK, et al. Independent βArrestin2 and G protein-mediated pathways for angiotensin II activation of extracellular
signal-regulated kinases 1 and 2. Proc Natl Acad Sci U S A 2003;
100:10782–7.
31. Resh MD. The ups and downs of Src regulation: tumor suppression
by Cbp. Cancer Cell 2008;13:469–71.
32. Arias-Romero LE, Saha S, Villamar-Cruz O, et al. Activation of Src by
protein tyrosine phosphatase 1B Is required for ErbB2 transformation
of human breast epithelial cells. Cancer Res 2009;69:4582–8.
33. Zhu S, Bjorge JD, Fujita DJ. PTP1B contributes to the oncogenic
properties of colon cancer cells through Src activation. Cancer Res
2007;67:10129–37.
34. Di Florio A, Capurso G, Milione M, et al. Src family kinase activity
regulates adhesion, spreading and migration of pancreatic endocrine
tumour cells. Endocr Relat Cancer 2007;14:111–24.
35. Pichot CS, Hartig SM, Xia L, et al. Dasatinib synergizes with doxorubicin to block growth, migration, and invasion of breast cancer cells.
Br J Cancer 2009;101:38–47.
36. Luttrell LM, Ferguson SS, Daaka Y, et al. βArrestin-dependent formation of β2 adrenergic receptor-Src protein kinase complexes. Science 1999;283:655–61.
37. DeFea KA. Stop that cell! βArrestin-dependent chemotaxis: a tale of
localized actin assembly and receptor desensitization. Annu Rev
Physiol 2007;69:535–60.
38. Wymann MP, Schneiter R. Lipid signalling in disease. Nat Rev Mol
Cell Biol 2008;9:162–76.
39. Graham DJ. COX-2 inhibitors, other NSAIDs, and cardiovascular risk:
the seduction of common sense. JAMA 2006;296:1653–6.
40. Ochiai M, Oguri T, Isobe T, Ishioka S, Yamakido M. Cyclooxygenase2 (COX-2) mRNA expression levels in normal lung tissues and nonsmall cell lung cancers. Jpn J Cancer Res 1999;90:1338–43.
41. Yano T, Zissel G, Muller-Qernheim J, Jae Shin S, Satoh H, Ichikawa
T. Prostaglandin E2 reinforces the activation of Ras signal pathway in
lung adenocarcinoma cells via EP3. FEBS Lett 2002;518:154–8.
42. Chell SD, Witherden IR, Dobson RR, et al. Increased EP4 receptor
expression in colorectal cancer progression promotes cell growth
and anchorage independence. Cancer Res 2006;66:3106–13.
43. Mutoh M, Watanabe K, Kitamura T, et al. Involvement of prostaglandin E receptor subtype EP(4) in colon carcinogenesis. Cancer Res
2002;62:28–32.
44. Sheng H, Shao J, Washington MK, DuBois RN. Prostaglandin E2 increases growth and motility of colorectal carcinoma cells. J Biol
Chem 2001;276:18075–81.
Mol Cancer Res; 8(4) April 2010
577