Báo cáo khoa học: Novel suppression mechanism operating in early phase of adipogenesis by positive feedback loop for enhancement of cyclooxygenase-2 expression through prostaglandin F2a receptor mediated activation of MEK⁄ ERK-CREB cascade doc

Novel suppression mechanism operating in early phase of
adipogenesis by positive feedback loop for enhancement
of cyclooxygenase-2 expression through prostaglandin F
2a
receptor mediated activation of MEK
⁄
ERK-CREB cascade
Toshiyuki Ueno and Ko Fujimori
Laboratory of Biodefense and Regulation, Osaka University of Pharmaceutical Sciences, Japan
Keywords
3T3-L1, PGF
2a
; adipocytes; COX-2; CREB;
MEK ⁄ ERK
Correspondence
K. Fujimori, Laboratory of Biodefense
and Regulation, Osaka University of
Pharmaceutical Sciences, 4-20-1 Nasahara,
Takatsuki, Osaka 569-1094, Japan
Fax: +81 72 690 1215
Tel: +81 72 690 1215
E-mail:
(Received 6 April 2011, revised 31 May
2011, accepted 8 June 2011)
doi:10.1111/j.1742-4658.2011.08213.x
Prostaglandin (PG) F
2a
suppresses adipocyte differentiation by inhibiting the
function of peroxisome proliferator-activated receptor c. In this study, we
identified a novel suppression mechanism, operating in the early phase of
adipogenesis, that increased the production of anti-adipogenic PGF

2a
and
PGE
2
by enhancing cyclooxygenase (COX) 2 expression through the PGF
2a
-
activated FP receptor ⁄ extracellular-signal-regulated kinase (ERK) ⁄ cyclic
AMP response element binding protein (CREB) cascade. COX-2 expression
was enhanced with a peak at 1 h for the mRNA level and at 3 h for the pro-
tein level after the addition of Fluprostenol, an FP receptor agonist. The
Fluprostenol-derived elevation of COX-2 expression was suppressed by the
co-treatment with an FP receptor antagonist, AL8810, with a mitogen-acti-
vated protein kinase (MEK; ERK kinase) inhibitor, PD98059. ERK was
phosphorylated within 10 min after the addition of Fluprostenol, and its
phosphorylation was inhibited by the co-treatment with AL8810 or
PD98059. Moreover, FP receptor mediated activation of the MEK ⁄ ERK
cascade and COX-2 expression increased the production of PGF
2a
and
PGE
2
. An FP receptor antagonist and each inhibitor for MEK and COX-2
suppressed the PGF
2a
-derived induction of synthesis of these PGs. Further-
more, promoter-luciferase and chromatin immunoprecipitation assays dem-
onstrated that PGF
2a
-derived COX-2 expression was activated through

binding of CREB to the promoter region of the COX-2 gene in 3T3-L1 cells.
These results indicate that PGF
2a
suppresses the progression of the early
phase of adipogenesis by enhancing the binding of CREB to the COX-2 pro-
moter via FP receptor activated MEK ⁄ ERK cascade. Thus, PGF
2a
forms a
positive feedback loop that coordinately suppresses the early phase of adipo-
genesis through the increased production of anti-adipogenic PGF
2a
and PGE
2
.
Introduction
Adipose tissue plays a critical role as a site for energy
storage [1,2] and has been identified as an endocrine
organ that secretes various biologically active molecules
called adipocytokines [3]. However, excessive lipid
accumulation or enlarged size of adipocytes is asso-
ciated with diseases such as obesity and diabetes [4].
Abbreviations
AKR, aldo-keto reductase; ChIP, chromatin immunoprecipitation; COX, cyclooxygenase; CRE, cyclic AMP responsive element; CREB,
CRE-binding protein; EIA, enzyme immunoassay; ERK, extracellular signal-regulated kinase; MEK, mitogen-activated protein
kinase ⁄ extracellular signal-regulated kinase kinase; pAb, polyclonal antibody; PG, prostaglandin; PLA
2
, phospholipase A
2
; PPAR, peroxisome
proliferator-activated receptor.

FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS 2901
Adipocyte differentiation (adipogenesis) is a complex
process including the coordinated changes in hormone
sensitivity and gene expression in response to various
stimuli including lipid mediators.
Prostaglandins (PGs) are a group of lipid mediators
with numerous functions, and they are synthesized
from membrane lipids by three enzymatic steps. First,
arachidonic acid is liberated by phospholipase A
2
s
(PLA
2
s). Second, it is converted to PGH
2
by either cy-
clooxygenase (COX) 1 or COX-2. Third, PGH
2
, which
is a common precursor of all prostanoids, is metabo-
lized to various PGs by the action of specific PG syn-
thases [5,6]. PGs are involved in the regulation of
adipocyte differentiation. Lipocalin-type PGD-syn-
thase-produced PGD
2
enhances adipocyte differentia-
tion [7], whereas PGE
2
and PGF
2a

suppress the
differentiation of adipocytes through EP4 [8] and FP
[9–12] receptors, respectively, indicating that PGE
2
and
PGF
2a
are anti-adipogenic factors. PGF
2a
plays a vari-
ety of physiological roles in the body and is synthe-
sized by the reduction of either the 9,11-endoperoxide
moiety of PGH
2
or the 9-keto group of PGE
2
in an
NAD(P)H-dependent manner, both of which reactions
are catalyzed by aldo-keto reductases (AKRs) [13]. We
recently showed that AKR1B3 acts as the PGF syn-
thase in adipocytes and that AKR1B3-derived PGF
2a
suppresses the early phase of adipogenesis [14,15]
through a specific cell surface G-protein-coupled recep-
tor, the FP receptor [16,17], which binding leads to the
activation of various kinases [16,18–20].
COX consists of two isozymes, COX-1 and COX-2,
and is the rate-limiting enzyme catalyzing the conver-
sion of arachidonic acid into endoperoxide intermedi-
ates that are ultimately converted by specific synthases

to prostanoids [5,21,22]. COX-1 is constitutively
expressed in most cells including the adipocytes,
whereas COX-2 expression is induced by various stim-
uli and is transiently activated in the early phase of
adipogenesis [15]. The expression of antisense COX-2
mRNA results in the upregulation of adipogenesis,
thus indicating that COX-2 is involved in the produc-
tion of anti-adipogenic prostanoids [23]. In contrast,
selective COX-2 inhibitors impair adipocyte differenti-
ation through inhibition of the clonal expansion phase
[24]. Moreover, COX-2-deficient mice exhibit sup-
pressed adipocyte differentiation [25]. Thus, arachi-
donic acid metabolism during adipogenesis is a process
governed at multiple levels, suggesting a complex role
for PGs during fat cell development [26]. The effects of
COX in adipogenesis are controversial, because COX-
2-derived PGs play suppressive roles in the early phase
of adipogenesis [14,15], while COX-2 and lipocalin-
type PGD-synthase-derived PGD
2
activates the late
phase of adipogenesis [7]. The regulation of adipogene-
sis by PGs is thus not yet fully understood.
In this study, we demonstrate the effect of PGF
2a
–
FP receptor signaling on the expression of COX-2 and
suppression of the early phase of adipogenesis in mouse
adipocytic 3T3-L1 cells. PGF
2a

promoted the produc-
tion of anti-adipogenic PGF
2a
and PGE
2
by enhancing
the expression of COX-2 through FP-receptor-activated
mitogen-activated protein kinase⁄ extracellular signal-
regulated kinase kinase (MEK) ⁄ extracellular signal-
regulated kinase (ERK) cascade and the binding of
cyclic AMP response element (CRE) binding protein
(CREB) to the COX-2 promoter. Therefore, PGF
2a
suppresses the early phase of adipogenesis by de novo
synthesis of anti-adipogenic PGF
2a
and PGE
2
through
a positive feedback FP receptor-MEK ⁄ ERK-CREB-
COX-2 loop.
Results
Activation of FP receptor increases PGF
2a
production with enhanced expression of COX-2
During the adipogenesis of mouse 3T3-L1 cells, PGF
2a
is produced in the early phase of adipogenesis with a
peak at 3 h after the initiation of adipogenesis [15].
In this study, we found that when 3T3-L1 cells were

incubated with Fluprostenol, an FP receptor agonist,
for 1 h, PGF
2a
production was significantly increased
(Fig. 1A), indicating that PGF
2a
may enhance PGF
2a
production. To identify the molecular mechanism of
this Fluprostenol-mediated increase in PGF
2a
produc-
tion in adipocytes, at first we investigated the expres-
sion of genes involved in the biosynthesis of PGF
2a
and FP receptor in Fluprostenol-treated 3T3-L1 cells
by performing quantitative PCR analysis. The tran-
scription level of the COX-2 gene was increased  3.7-
fold, whereas that of the FP receptor gene was
decreased about 28%, compared with the levels of the
vehicle control (Fig. 1B). The expression levels of cyto-
solic PLA
2
(cPLA
2
), COX-1 and AKR1B3 genes were
not altered by the treatment of the cells with Flupros-
tenol (Fig. 1B). Moreover, we confirmed that PGF
2a
could activate COX-2 expression in 3T3-L1 cells about

3.1-fold compared with the vehicle control (Fig. 1C).
Next, we examined the time course change in the
COX-2 expression level in the Fluprostenol-treated
3T3-L1 cells. The transcription of COX-2 mRNA was
increased at 1 h after the addition of Fluprostenol and
then quickly decreased to a level lower than that
detected in the vehicle-treated cells (Fig. 1D). The
COX-2 protein level was increased with a peak at 3 h
after the addition of Fluprostenol and then decreased
Suppression of adipogenesis via FP-ERK-CREB-COX-2 T. Ueno and K. Fujimori
2902 FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS
AB
COX-2
0.1
*
FP
0.751.6
*
PGF
2α
(ng·mL
–1
)
0.08
0.06
0.04
0.02
mRNA level
(/TBP mRNA level)
0.5

0.25
**
1.2
0.8
0.4
Fluprostenol
–
+
Fluprostenol
–
+
0
COX-1
10
cPLA
2
10
10
AKR1B3
Fluprostenol
–
+
00
C
0.06
7.5
5
7.5
5
7.5

5
2.5
mRNA level
(/TBP mRNA level)
0.04
0.02
*
Fluprostenol
–
+
0
2.5
Fluprostenol
–
+
0
2.5
0
Fluprostenol
– +
DE
PGF
2α
–
+
0
COX-2mRNA level
(/TBP mRNA level)
COX-2mRNA level
(/TBP mRNA level)

1
1.5
0.1
0.2
0.15
*
COX-2
Actin
*
*
0
0.5
013612
Time after addition of Fluprostenol (h)
Relative band intensity
0
0.05
*
*
*
013612
Time after addition of Fluprostenol (h)
Fig. 1. Enhancement of COX-2 expression by Fluprostenol, an FP receptor agonist. (A) Induction of PGF
2a
production by treatment with
Fluprostenol. Pre-adipocytic 3T3-L1 cells were incubated with Fluprostenol (0.5 n
M; Cayman Chemical) or vehicle for 1 h, followed by treat-
ment with A23187 (5 l
M) for 10 min at 37 ° C, after which the medium was collected to measure the PGF
2a

level by EIA. *P < 0.01, com-
pared with vehicle-treated cells. (B) Expression levels of COX-2, FP receptor, cPLA
2
, COX-1 and AKR1B3 genes in the Fluprostenol-treated
3T3-L1 cells. Cells were incubated with Fluprostenol (0.5 n
M) for 1 h, and the expression level of each gene was measured by quantitative
PCR. The data are presented as the mean ± SD of three independent experiments. *P < 0.01, **P < 0.05, compared with vehicle control.
(C) Transcription level of COX-2 gene in the PGF
2a
-treated 3T3-L1 cells. Cells were cultured with PGF
2a
(100 nM) for 1 h, and the mRNA level
of COX-2 gene was then measured by quantitative PCR. The data are presented as the mean ± SD of three independent experiments.
*P < 0.01, compared with vehicle control. (D) Expression level of COX-2 gene in Fluprostenol-treated 3T3-L1 cells. Cells were incubated with
Fluprostenol (0.5 n
M) for various times. Expression levels of COX-2 gene were measured by quantitative PCR. The data are presented as the
mean ± SD of three independent experiments. *P < 0.01, compared with vehicle-treated cells. (E) Protein level of COX-2 was detected by
western blot analysis. Cells were treated as described in (D). Crude cell extracts (20 lg) were loaded in each lane. *P < 0.01, compared with
vehicle-treated cells.
T. Ueno and K. Fujimori Suppression of adipogenesis via FP-ERK-CREB-COX-2
FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS 2903
to almost the same level as detected in the vehicle-trea-
ted cells, although the actin levels were not altered in
each sample (Fig. 1E). These results, all taken
together, reveal that Fluprostenol-activated FP recep-
tors increased PGF
2a
production and enhanced COX-2
expression in 3T3-L1 cells.
Involvement of MEK signaling in the PGF

2a
FP receptor-activated COX-2 expression
To elucidate the regulatory mechanism of the PGF
2a
FP receptor activated COX-2 expression in 3T3-L1
cells, we cultured the cells with AL8810, which is
an FP receptor antagonist, or with an MEK inhibi-
tor, PD98059; a PLC inhibitor, U73122; a PKC
inhibitor, bisindolylmaleimide I; and a general COX
inhibitor, indomethacin, or a selective COX-2 inhibi-
tor, NS-398, in the presence of Fluprostenol for 1 h.
COX-2 expression was enhanced by the treatment with
Fluprostenol compared with that obtained with the
vehicle control (Fig. 2A). Fluprostenol FP receptor-
activated COX-2 expression was abolished by the
co-treatment with AL8810 or PD98059 (Fig. 2A). In
contrast, U73122, bisindolylmaleimide I, indomethacin
and NS-398 had no effect on the expression level of
Fluprostenol FP receptor-activated COX-2 expression
(Fig. 2A). Moreover, the Fluprostenol-derived eleva-
tion of COX-2 mRNA expression was suppressed by
AL8810 or PD98059 in a concentration-dependent
manner (Fig. 2B), results which agree well with the
COX-2 protein level demonstrated by western blot
analysis, although the actin levels were not changed in
each sample (Fig. 2C). These results indicate that
Fluprostenol-derived enhancement of COX-2 expres-
sion occurred through FP receptor and MEK signaling
in 3T3-L1 cells.
Phosphorylation of ERK1 ⁄ 2 by activated PGF

2a
–FP
receptor signaling
The MEK ⁄ ERK signaling pathway is critical in the
determination of the specificity of cellular responses
such as cell proliferation and cell differentiation [27].
Next, we examined whether ERK1 ⁄ 2 was activated by
MEK in the Fluprostenol-treated 3T3-L1 cells. ERK1 ⁄ 2
was expressed constitutively even in the absence of
Fluprostenol, and slightly phosphorylated in the vehi-
cle-treated cells (Fig. 3A). When the cells were cultured
in DMEM containing Fluprostenol, phosphorylation of
ERK1 ⁄ 2 was enhanced with a peak 10 min after the
addition of Fluprostenol and then decreased (Fig. 3A).
To confirm the activation of ERK1 ⁄ 2 via Fluproste-
nol-activated FP receptors, we cultured the cells in the
presence of AL8810 or PD98059 and Fluprostenol,
and assessed the level of phosphorylated ERK1 ⁄ 2 pro-
tein by western blot analysis. Fluprostenol-mediated
phosphorylation of ERK1 ⁄ 2 was diminished by
AL8810 (Fig. 3B). Moreover, PD98059 also decreased
it (Fig. 3B). These results indicate that Fluprostenol
activated the MEK ⁄ ERK signaling pathway in 3T3-L1
cells within 10 min through the FP receptor.
Enhanced de novo synthesis of PGF
2a
and PGE
2
by activation of FP receptor MEK ⁄ ERK cascade
Next we examined whether FP receptor-mediated acti-

vation of the MEK ⁄ ERK cascade would enhance
PGF
2a
production in 3T3-L1 cells. PGF
2a
production
was increased by the treatment with Fluprostenol
(Figs 1 and 4, left panel). Fluprostenol-derived eleva-
tion of PGF
2a
production was decreased by the treat-
ment with AL8810, PD98059, indomethacin or NS-398
(Fig. 4, left panel), whereas U73122 and bisindolylma-
leimide I had no effect on the production (Fig. 4, left
panel). Moreover, when the cells were incubated with
PD98059 or NS-398, PGF
2a
production decreased to
the basal level, indicating that basal and induced
PGF
2a
production are dependent upon COX-2 and
MEK ⁄ ERK cascade. Furthermore, we found that
PGE
2
production was also enhanced by the activation
of the FP receptor MEK ⁄ ERK cascade (Fig. 4, right
panel). The co-treatment with AL8810, PD98059,
indomethacin or NS-398 decreased the Fluprostenol-
induced PGE

2
production (Fig. 4, right panel). In con-
trast, U73122 and bisindolylmaleimide I had no effect
on it (Fig. 4, right panel). Moreover, the levels of
other PGs, PGD
2
, prostacyclin and thromboxane, were
not altered when the cells were incubated with Flupro-
stenol (data not shown). These results reveal that
Fluprostenol-derived activation of the FP receptor
MEK ⁄ ERK cascade enhanced de novo synthesis of
anti-adipogenic PGF
2a
and PGE
2
in 3T3-L1 cells.
Fluprostenol-activated COX-2 expression through
binding of CREB to COX-2 promoter
To examine the transcriptional activation mechanism
of the COX-2 gene triggered by activation of the FP
receptor in 3T3-L1 cells, we conducted luciferase repor-
ter assays using various deleted or mutated promoter-
reporter plasmids (Fig. 5A). The transcription initiation
site of mouse COX-2 gene was determined previously
[28]. When the construct carrying the promoter region
from )500 to +124, named )500 ⁄ +124, was used for
the transfection, efficient reporter activity was detected
(Fig. 5A). Moreover, when the )500 ⁄ +124 construct
Suppression of adipogenesis via FP-ERK-CREB-COX-2 T. Ueno and K. Fujimori
2904 FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS

transfected cells were treated with Fluprostenol, lucifer-
ase reporter activity was  138% (black column) of
that of the vehicle (white column); and this Fluproste-
nol-activated COX-2 promoter activity was suppressed
by the co-treatment with AL8810 (gray column).
Further sequential deletion analysis of the region from
)500 to )140 did not result in any significant change
from that observed when the )500 ⁄ +124 construct
was used for the transfection (Fig. 5A), indicating that
the region from )500 to )140 of COX-2 promoter
0.01
0.0075
#
*
*
0.0025
0.005
COX-2 mRNA level
(/TBP mRNA level)
0.03
*
*
0
V V U Bis
PD
+Fluprostenol
AL
0.03
*
NSI

*
*
0.01
0.02
0.01
0.02
COX-2 mRNA level
(/TBP mRNA level)
COX-2
0
0
COX-2
010500
0555
Fluprostneol (n
M)
AL 8810 (μ
M)
01100
0555
Fluprostenol (n
M)
PD98059 (μ
M)
Actin
Actin
2
3
* *
*

1
1.5
*
**
010500
0555
Fluprostneol (n
M)
AL 8810 (μ
M)01100
0555
Fluprostenol (n
M)
PD98059 (μ
M)
0
1
Relative band intensity
Relative band intensity
0
0.5
A
B
C
Fig. 2. Involvement of MEK signaling in
the enhancement of COX-2 expression. (A)
Inhibition of Fluprostenol-activated COX-2
expression by kinase and COX inhibitors.
3T3-L1 cells were cultured in the presence
of Fluprostenol (0.5 n

M) with or without
AL8810 (50 l
M; Cayman Chemical), which
is an FP receptor antagonist, each of various
protein kinases: the MEK inhibitor, PD98059
(PD) (10 l
M, Calbiochem), the PLC inhibitor
U73122 (U) (1 l
M, Calbiochem), the PKC
inhibitor bisindolylmaleimide I (Bis) (0.1 l
M,
Calbiochem), the general COX inhibitor
indomethacin (I) (2 l
M, Cayman Chemical)
and the COX-2 specific inhibitor NS-398
(NS) (1 l
M, Cayman Chemical) for 5 min.
Cells were further incubated for 1 h with
Fluprostenol (0.5 n
M) in the presence of
each inhibitor. COX-2 mRNA levels were
measured by quantitative PCR. Data are
the mean ± SD of three independent
experiments. *P < 0.01, as indicated by the
brackets. (B) Suppression of Fluprostenol-
activated COX-2 expression by AL8810 or
PD98059. Cells were incubated with
AL8810 (0–50 l
M) or PD98059 (0–10 lM)
together with Fluprostenol (5 n

M). The
expression level of COX-2 gene was quanti-
fied by PCR. Data are the mean ± SD of
three independent experiments. *P < 0.01,
as indicated by the brackets. (C) Western
blot analysis. Cells were treated as
described in (B). Crude cell extracts (15 lg)
were loaded in each lane. Data are the
mean ± SD of three independent
experiments. *P < 0.01, as indicated by the
brackets.
T. Ueno and K. Fujimori Suppression of adipogenesis via FP-ERK-CREB-COX-2
FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS 2905
containing the nuclear factor jB element was not
involved in the activation of COX-2 promoter activity.
In contrast, when the region from )140 to )110 was
deleted, the luciferase reporter activity was  24%
decreased compared with that of the )140 ⁄ +124 con-
struct (Fig. 5A). However, the responsiveness to Flupro-
stenol and AL8810 was still observed, suggesting the
existence of a cis-regulatory element for the basal
expression of the COX-2 gene in the region from )140
to )110 of the COX-2 promoter (Fig. 5A). This result
was consistent with previous results showing that the
binding element for CCAAT ⁄ enhancer binding protein
at )135 of the COX-2 promoter acts as a positive
cis-regulatory element in basal COX-2 gene expression
in 3T3-L1 cells [29]. Furthermore, when the region
from )110 to )50 was deleted, the luciferase reporter
activity was significantly decreased, and the responses

to Fluprostenol and AL8810 disappeared (Fig. 5A),
P-ERK1/2
ERK1/2
A
P-ERK1/2
ERK1/2
B
0103060
Time after addition of
Fluprostenol (min)
Fluprostenol
AL8810
PD98059
–
+
–
–
–
–
+
–
+
+
–
+
5
4
3
2
*

*
*
**
*
0103060
Time after addition of
Fluprostenol (min)
Fluprostenol
AL8810
–
+
––
+
+
+
–
0
2
3
1
Relative band
intensity
Relative band
intensity
0
1
*
PD98059
–– –
+

Fig. 3. Time course of PGF
2a
-stimulated ERK1 ⁄ 2 phosphorylation. (A) Fluprostenol-induced phosphorylation of ERK1 ⁄ 2 in 3T3-L1 cells. Cells
were cultured with Fluprostenol (0.5 n
M) for 0–60 min. Cell lysates (20 lgÆlane
)1
) were subjected to SDS ⁄ PAGE and western blot analysis
for phosphorylated ERK1 ⁄ 2 and total ERK1 ⁄ 2 proteins. Band intensity was measured by the use of
MULTI GAUGE software. The data shown
are representative of three independent experiments. (B) The MEK inhibitor, PD98059, prevented Fluprostenol-mediated phosphorylation of
ERK1 ⁄ 2 in 3T3-L1 cells. Cells were cultured with Fluprostenol (0.5 n
M) in the presence or absence of AL8810 (50 lM) or PD98059 (10 lM)
for 10 min. Cell lysates (20 lgÆlane
)1
) were used for SDS ⁄ PAGE and western blot analysis to determine the levels of phospho-ERK1 ⁄ 2 and
total ERK1 ⁄ 2 proteins. Band intensity was measured by using
MULTI GAUGE software. The data shown are representative of three indepen-
dent experiments. *P < 0.01, as indicated by the brackets.
1.2
1.5
1.8
20
25
30
* *
*
*
*
*
*

*
*
*
PGF
2α
(ng·mL
–1
)
PGE
2
(ng·mL
–1
)
0
0.3
0.6
0.9
V V AL PD I NS U Bis
5
10
15
0
V V AL PD I NS U Bis
+ Fluprostenol + Fluprostenol
Fig. 4. Effect of kinase inhibitors or COX inhibitors on Fluprostenol-stimulated PGF
2a
and PGE
2
production. 3T3-L1 cells were incubated in
DMEM containing AL8810 (AL) (10 l

M), PD98059 (PD) (10 lM), U73122 (U) (1 lM), bisindolylmaleimide I (Bis) (0.1 lM), COX inhibitors indo-
methacin (I) (2 l
M) or NS-398 (NS) (1 lM) together with Fluprostenol (5 nM) for 1 h. The medium was then removed and replaced with fresh
medium containing inhibitor and A23187 (5 l
M), and the cells were further incubated for 10 min. The medium was collected for the mea-
surement of PGF
2a
and PGE
2
levels by performing the respective EIAs. Data are expressed as the mean ± SD of three independent experi-
ments. *P < 0.01, as indicated by the brackets.
Suppression of adipogenesis via FP-ERK-CREB-COX-2 T. Ueno and K. Fujimori
2906 FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS
revealing that the region from )110 to )50 contained
a critical cis-regulatory element for the basal and
PGF
2a
-derived activation of COX-2 gene expression in
3T3-L1 cells.
We searched for putative cis-regulatory elements in
the promoter region from )110 to )50 of the COX-2
promoter by using matinspector software [30] and
found one putative CRE at )59 of the COX-2 pro-
moter (Fig. 5A). To confirm the importance of the
CRE at )59 in the PGF
2a
-derived elevation of COX-2
gene expression in 3T3-L1 cells, we introduced a muta-
tion at this CRE at )59 of the COX-2 promoter of the
)500 ⁄ +124 construct: )500 ⁄ +124(mu). When the

cells were transfected with the )500 ⁄ +124(mu) con-
struct, the luciferase reporter activity was significantly
decreased compared with that of the wild-type
)500 ⁄ +124 construct (Fig. 5A). In addition, the
responsiveness to Fluprostenol and AL8810 was lost
+124
–500/+124
A
+1
–
500
CRE
(–59)
CCAAT
(–135)
NF-κB
(–224)
*
*
–110/+124
–50/+124
–300/+124
–140/+124
*
*
*
*
*
*
*

*
*
*
*
*
*
–500/+124
(mu)
pGL4.10
13
56
Relative luminescence
0
WT
mu
CTACGTCA
CagtaggA
Vehicle
Fluprostenol
Fluprostenol + AL8810
24
B
Input
CRE
–59 –52
168-bp
3′5′
Fluprostenol
–
+

AL8810
+
––
+
+
–
ChIP: anti-CREB
PD98059
–––
+
3
4
6
2
*
5
*
*
0
1
Relative band intensity
Fig. 5. Enhancement of PGF
2a
-mediated COX-2 gene expression through binding of CREB to COX-2 promoter. (A) Deletion and mutation
analyses of mouse COX-2 promoter region in 3T3-L1 cells. Transfected cells were treated with Fluprostenol (0.5 n
M, black column), Flupros-
tenol and AL8810 (10 l
M, gray columns) or incubated without any treatment (white columns) for 6 h; then luciferase reporter activities were
measured (right panel). The data represent the mean ± SD of three independent assays. The putative cis-regulatory elements are indicated
at the top of the diagram, and mutated nucleotides by small characters (left panel). *P < 0.01, as indicated by the brackets. (B) ChIP assay

of the CRE of mouse COX-2 promoter in 3T3-L1 cells. The scheme for the ChIP assay for the mouse COX-2 promoter is shown at the left.
Cells were treated with Fluprostenol (0.5 n
M) with or without AL8810 (10 lM) or PD98059 (10 lM) for 1 h, and the ChIP assay was then car-
ried out. The profile of the amplicon is shown at the right and the input control (input) means that a small aliquot before immunoprecipitation
was used for PCR amplification. The data are representative of three independent experiments. *P < 0.01, as indicated by the brackets.
T. Ueno and K. Fujimori Suppression of adipogenesis via FP-ERK-CREB-COX-2
FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS 2907
(Fig. 5A). These results indicate that PGF
2a
activated
COX-2 gene expression through the CRE at )59 of
the mouse COX-2 promoter in 3T3-L1 cells.
Next, we examined the binding of CREB to the
CRE at )59 of the COX-2 promoter by performing a
chromatin immunoprecipitation (ChIP) assay with
anti-CREB antibody. The expected size (168 bp;
Fig. 5B, left panel) of an amplicon containing the
CRE at )59 was detected in the formaldehyde-medi-
ated DNA–protein complexes immunoprecipitated
with anti-CREB antibody (Fig. 5B, right panel). More-
over, the binding efficiency was enhanced when the
cells were treated with Fluprostenol (Fig. 5B, right
panel), and the Fluprostenol-derived increase in the
efficiency of binding of CREB to the CRE was clearly
suppressed by the co-treatment with either AL8810 or
PD98059 (Fig. 5B, right panel). In contrast, there was
no detectable signal when rabbit normal IgG was
added, although the signals were detected in both
input controls (data not shown). These results, taken
together, indicate that Fluprostenol-mediated COX-2

expression via the FP receptor and MEK ⁄ ERK cas-
cade was enhanced through the binding of CREB to
the CRE at )59 of the COX-2 promoter in 3T3-L1
cells.
Discussion
PGs are lipid mediators involved in the regulation of
cell growth, differentiation and homeostasis [5] as well
as that of adipogenesis. PGD
2
accelerates adipogenesis
[7] and PGD
2
-overexpressing mice become obese when
given a high fat diet [31]. In contrast, PGF
2a
and
PGE
2
suppress adipogenesis through specific receptors,
i.e. FP [9–12] and EP4 [8], respectively, and inhibit the
functions of peroxisome proliferator-activated receptor
(PPAR)c, which is a key transcription factor that regu-
lates adipogenesis and is expressed in the mid-late
phase of adipogenesis [32,33]. A recent study demon-
strated that PGF
2a
was detected in pre-adipocytes and
that its level was enhanced with a peak at 3 h after the
initiation of adipogenesis and then decreased [15], indi-
cating that PGF

2a
-mediated suppression of adipogene-
sis is an early event during adipogenesis. In this study,
we found an increase in the PGF
2a
level along with
enhanced expression of COX-2 when the FP receptor
in adipocytes was activated by a stable PGF
2a
analog,
Fluprostenol (Fig. 1A). Thus, we suspect the existence
of a regulatory loop for the enhancement of PGF
2a
production by activation of FP receptor signaling. Our
present study demonstrated that PGF
2a
enhanced anti-
adipogenic PGF
2a
and PGE
2
production through
the FP receptor-activated COX-2 expression via the
MEK ⁄ ERK-CREB cascade to form a positive feed-
back loop, one that probably plays an important role
in the regulation of the early phase of adipogenesis
(Fig. 6).
PGs have numerous functions, and their activities
are achieved by their binding to their specific receptors.
PGF

2a
binds to the FP receptor, which is a G-protein-
coupled receptor; and this binding activates the
downstream signaling pathways including the Gq
heterotrimeric G protein, and then further increases
the intracellular calcium level and activates various
kinases including MEK [16,18–20]. MEK, also known
as MAPK kinase [34], is an activator of ERK, a
MAPK. The MEK ⁄ ERK pathway is a critical signal-
ing pathway that regulates a number of cellular func-
tions such as cell growth and differentiation [34]. The
upstream component of the MEK ⁄ ERK pathway is
the Ras GTPase, which activates the serine ⁄ threonine
kinase Raf, which in turn phosphorylates ERK1 ⁄ 2
through the Ras ⁄ Raf ⁄ MEK ⁄ ERK signal transduction
pathway [34]. PGF
2a
stimulates Raf ⁄ MEK ⁄ ERK sig-
naling in luteal [35], endometrial [36], pulp cells [37]
and osteoblasts [38,39]. Previously, PGF
2a
was shown
to regulate COX-2 expression in an autocrine ⁄ para-
crine manner to establish a positive feedback system in
Ishikawa endometrial adenocarcinoma cells [36,40].
Here, we demonstrated the PGF
2a
–FP receptor-derived
activation of COX-2 expression via the activation of
the MEK ⁄ ERK cascade. Treatment of 3T3-L1 cells

with the FP receptor antagonist or MEK inhibitor
significantly reduced the Fluprostenol-activated COX-2
expression. To determine whether the enhanced COX-2
PGF
2α
α
FP
MEK
ERK
P
PPARγ
Adipogenesis
COX-2
AKR1B3
PGF
2
α
CREB
PGE
2
Fig. 6. Schematic representation of a novel suppression mecha-
nism operating in the early phase of adipogenesis through a posi-
tive feedback loop by enhancement of PGF
2a
-mediated COX-2
expression via FP receptor-activated MEK ⁄ ERK-CREB cascade.
Suppression of adipogenesis via FP-ERK-CREB-COX-2 T. Ueno and K. Fujimori
2908 FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS
expression induced after the PGF
2a

–FP receptor inter-
action could lead to de novo biosynthesis of prosta-
noids, we treated 3T3-L1 cells with Fluprostenol for
1 h to induce COX-2 expression in the absence or
presence of the selective inhibitors of COX-2 and
MEK or FP receptor antagonist. We found that
PGF
2a
and PGE
2
were de novo synthesized by the acti-
vation of FP receptor and that this effect was abol-
ished by the co-treatment with the FP receptor
antagonist or MEK inhibitor (Fig. 4). The inhibition
of PG biosynthesis by the specific COX-2 inhibitor
NS-398 and the MEK inhibitor PD98059 confirmed
that the increased production of PGF
2a
and PGE
2
was
a consequence of the elevated expression of the COX-2
gene, which was dependent on the phosphorylation of
ERK1 ⁄ 2 (Fig. 3B). Thus, PGF
2a
and PGE
2
produc-
tion through PGF
2a

–FP receptor-MEK ⁄ ERK-activated
COX-2 expression was involved in the regulation of
the early phase of adipogenesis. Hence, adipogenesis
could be regulated through a self-amplifying loop, trig-
gered by PGF
2a
–FP receptor coupling and activation
of the COX-2 gene expression via the MEK ⁄ ERK cas-
cade. However, this PGF
2a
-mediated acceleration of
suppression of the early phase of adipogenesis was
transient, returning to basal levels within 3 h after the
treatment with PGF
2a
, indicating that this signaling
cascade was rapidly desensitized in the presence of
PGF
2a
. The desensitization mechanism to PGF
2a
should be elucidated to fully understand the PGF
2a
-
derived suppression of adipogenesis in its early phase.
The regulatory mechanism governing the transcrip-
tion of the COX-2 gene has been extensively studied
[41], and various transcription factors are now known
to be involved in the regulation of COX-2 gene expres-
sion [41]. CRE was first identified as an element in the

promoter of genes transcribed in response to the eleva-
tion of the cAMP level [42]. CREB is also involved in
the regulation of COX-2 gene expression in various
cells [43] through activation of protein kinase C and
Ca
2+
signaling cascades in ileal epithelial IEC-18 cells
[44] or the protein kinase A pathway in human amnion
fibroblasts [45]. In our present study, we demonstrated
CREB to be a critical transcription factor in the
PGF
2a
-derived enhancement of COX-2 gene expression
through the FP receptor-activated ERK ⁄ MEK cascade
in 3T3-L1 cells (Fig. 5A,B).
In summary, our data provide a novel regulatory
loop of the PGF
2a
-activated FP receptor-MEK ⁄ ERK-
CREB-COX-2 cascade that coordinately suppresses
the early phase of adipogenesis via de novo synthesis of
anti-adipogenic PGF
2a
and PGE
2
. Therefore, PGF
2a
plays a critical role in suppressing the progression
of the early phase of adipogenesis, and this positive
feedback loop may provide a novel therapeutic strat-

egy for the treatment of obesity.
Experimental procedures
Cell culture
Mouse 3T3-L1 cells were obtained from the Human Science
Research Resources Bank (Osaka, Japan), cultured in
DMEM (Sigma, St Louis, MO, USA) containing 10%
(v ⁄ v) fetal bovine serum and antibiotics, and maintained in
a humidified atmosphere of 5% CO
2
at 37 °C [15].
RNA preparation and quantification of RNA level
Total RNA was extracted with Sepasol-RNAI (Nacalai
Tesque, Kyoto, Japan), and was then further purified with
an RNeasy Purification System (Qiagen, Hilden, Germany).
The first-strand cDNAs were synthesized from 1 lg of total
RNA with random hexamer and ReverTra Ace Reverse
Transcriptase (Toyobo, Osaka, Japan) at 42 °C for 60 min
after the initial denaturation at 72 °C for 3 min, followed
by heat inactivation of enzyme at 99 °C for 5 min. The
cDNAs were diluted and further utilized as the templates
for quantitative PCR analysis.
Expression levels were quantified by using a LightCycler
system (Roche Diagnostics, Mannheim, Germany) with
Thunderbird SYBR qPCR Mix (Toyobo) and the primer
sets shown in Table S1. The expression level of the target
genes was estimated by the use of concentration-known
standard DNA, and normalized to that of SASA-binding
protein.
Western blot analysis
Cells were lysed in RIPA buffer containing 50 mm

Tris ⁄ HCl, pH 8.0, 150 mm NaCl, 0.1% (w ⁄ v) SDS, 0.5%
(w ⁄ v) sodium deoxycholate, 1% (v ⁄ v) Nonidet P-40 and
1% (v ⁄ v) Triton X-100 along with a protease inhibitor mix-
ture (Nacalai Tesque) and phosphatase inhibitors, 50 lm
Na
2
MoO
4
,1mm NaF and 1 mm Na
3
VO
4
. After mild soni-
cation, cell extracts were prepared by centrifugation for
20 min at 12 000 g at 4 °C to remove the cell debris. Pro-
tein concentrations were measured with a Pierce BCA Pro-
tein Assay Reagent (Thermo Scientific, Rockford, IL,
USA). Proteins were separated on SDS ⁄ PAGE and then
transferred onto an Immobilon PVDF membrane (Milli-
pore, Bedford, MA, USA). Blots were first incubated with
primary antibodies, i.e. anti-COX-2 polyclonal antibody
(pAb) (1 : 200, C-20; Santa Cruz Biotech., Santa Cruz, CA,
USA), anti-ERK1 ⁄ 2, anti-phospho ERK1 ⁄ 2 pAbs (1 : 500;
Cell Signaling, Danvers, MA, USA) or anti-actin monoclo-
nal antibody (1 : 5000, AC-15; Sigma), followed by incuba-
tion with anti-rabbit, anti-goat or anti-mouse IgG antibody
T. Ueno and K. Fujimori Suppression of adipogenesis via FP-ERK-CREB-COX-2
FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS 2909
conjugated with horseradish peroxidase (Santa Cruz Bio-
tech.). Immunoreactive signals were detected by the use of

an Immobilon Western Detection Reagent (Millipore) and
LAS-3000 Luminoimage Analyzer (Fujifilm, Tokyo, Japan),
and analyzed with multi gauge software (Fujifilm).
Measurement of PGF
2a
and PGE
2
by enzyme
immunoassay (EIA)
Cells were incubated in DMEM containing A23187 (5 lm;
Calbiochem, San Diego, CA, USA), a calcium ionophore,
along with any inhibitors for 10 min at 37 °C; then the
medium was collected and measured for PGF
2a
and PGE
2
levels by using PGF
2a
and PGE
2
EIA Kits (Cayman Chem-
ical, Ann Arbor, MI, USA) according to the manufac-
turer’s instructions.
Construction of luciferase reporter vectors and
luciferase assay
The luciferase reporter vectors carrying the mouse COX-2
promoter were generated as follows. An  600 bp and
sequentially deleted region of the COX-2 promoter was
cloned into the pGL4.10[luc2] vector (Promega, Madison,
WI, USA). Site-directed mutation was introduced by using

a QuikChange Site-directed Mutagenesis Kit (Stratagene,
La Jolla, CA, USA) according to the manufacturer’s
instruction. Nucleotide sequences of the constructs were
determined to verify the correct sequences.
3T3-L1 cells were co-transfected with each construct
(0.9 lg) and pRL-SV40 (0.1 lg, Promega) in six-well plates,
the latter plasmid carrying the Renilla luciferase gene under
the control of the SV40 promoter as the transfection con-
trol, along with FuGENE6 Transfection Reagent (Roche
Diagnostics) according to the manufacturer’s instructions.
The cells were cultured for a further 48 h, and the luciferase
activities were measured by using a Dual-Glo Luciferase
Assay System (Promega). The reporter activity was calcu-
lated relative to that of pGL4.10[luc2] vector, which was
defined as 1. Data were obtained from three independent
experiments, and each experiment was performed in tripli-
cate. The relative promoter activities are presented as the
mean ± SD.
Chromatin immunoprecipitation assay
The ChIP assay was performed as described previously [7]
by the use of CREB pAb (SC-25785X; Santa Cruz Bio-
tech.). Immunoprecipitated DNA–protein complexes were
reverse crosslinked, and the DNAs were purified by using a
MinElute PCR Purification Kit (Qiagen) and used for sub-
sequent PCR amplification with KOD FX DNA Polymer-
ase (Toyobo) and the following specific primer set for CRE
at )59 in the COX-2 promoter: 5¢-CAGAGAGGGGGAA
AAGTTGG-3¢ and 5¢-GAGCAGAGTCCTGACTGACTC-3¢.
PCR was conducted under the following conditions: initial
denaturation at 94 °C for 2 min, followed by 32 cycles of

98 °C for 10 s, 55 °C for 20 s and 60 °C for 20 s. The
amplified PCR products (expected size 168 bp) were ana-
lyzed by performing agarose gel electrophoresis.
Statistical analysis
Differences between two groups were analyzed by the
unpaired t-test or Welch t-test. P-values < 0.05 were con-
sidered significant.
Acknowledgements
We acknowledge Dr Fumio Amano (Osaka University
of Pharmaceutical Sciences) for valuable discussions.
This work was supported in part by a Grant-in-Aid
for Scientific Research and Scientific Research on
Innovative Areas from the Ministry of Education, Cul-
ture, Sports, Science and Technology of Japan, and by
grants from Suzuken Memorial Foundation, the Sumi-
tomo Foundation, Gushinkai Foundation, Japan
Foundation for Applied Enzymology, Takeda Science
Foundation and the Research Foundation for Pharma-
ceutical Sciences (to K.F.).
References
1 Kershaw EE & Flier JS (2004) Adipose tissue as an
endocrine organ. J Clin Endocrinol Metab 89, 2548–
2556.
2 Spiegelman BM & Flier JS (2001) Obesity and the regu-
lation of energy balance. Cell 104, 531–543.
3 Matsuzawa Y (2006) The metabolic syndrome and
adipocytokines. FEBS Lett 580, 2917–2921.
4 Berg AH & Scherer PE (2005) Adipose tissue,
inflammation, and cardiovascular disease. Circ Res 96,
939–949.

5 Helliwell RJ, Adams LF & Mitchell MD (2004) Prosta-
glandin synthases: recent developments and a novel
hypothesis. Prostaglandins Leukot Essent Fatty Acids
70, 101–113.
6 Urade Y & Hayaishi O (2000) Prostaglandin D syn-
thase: structure and function. Vitam Horm 58, 89–120.
7 Fujimori K, Aritake K & Urade Y (2007) A novel path-
way to enhance adipocyte differentiation of 3T3-L1 cells
by up-regulation of lipocalin-type prostaglandin D syn-
thase mediated by liver X receptor-activated sterol regu-
latory element-binding protein-1c. J Biol Chem 282,
18458–18466.
8 Tsuboi H, Sugimoto Y, Kainoh T & Ichikawa A (2004)
Prostanoid EP4 receptor is involved in suppression of
Suppression of adipogenesis via FP-ERK-CREB-COX-2 T. Ueno and K. Fujimori
2910 FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS
3T3-L1 adipocyte differentiation. Biochem Biophys Res
Commun 322, 1066–1072.
9 Casimir DA, Miller CW & Ntambi JM (1996) Prea-
dipocyte differentiation blocked by prostaglandin stimu-
lation of prostanoid FP2 receptor in murine 3T3-L1
cells. Differentiation 60, 203–210.
10 Liu L & Clipstone NA (2007) Prostaglandin F2alpha
inhibits adipocyte differentiation via a G alpha q-cal-
cium-calcineurin-dependent signaling pathway. J Cell
Biochem 100, 161–173.
11 Miller CW, Casimir DA & Ntambi JM (1996) The
mechanism of inhibition of 3T3-L1 preadipocyte differ-
entiation by prostaglandin F2alpha. Endocrinology 137,
5641–5650.

12 Serrero G & Lepak NM (1997) Prostaglandin F2alpha
receptor (FP receptor) agonists are potent adipose dif-
ferentiation inhibitors for primary culture of adipocyte
precursors in defined medium. Biochem Biophys Res
Commun 233, 200–202.
13 Watanabe K, Yoshida R, Shimizu T & Hayaishi O
(1985) Enzymatic formation of prostaglandin F2 alpha
from prostaglandin H2 and D2. Purification and prop-
erties of prostaglandin F synthetase from bovine lung.
J Biol Chem 260, 7035–7041.
14 Fujimori K, Ueno T & Amano F (2010) Prostaglandin
F(2alpha) suppresses early phase of adipogenesis, but is
not associated with osteoblastogenesis in mouse mesen-
chymal stem cells. Prostaglandins Other Lipid Mediat
93, 52–59.
15 Fujimori K, Ueno T, Nagata N, Kashiwagi K, Aritake
K, Amano F & Urade Y (2010) Suppression of adipo-
cyte differentiation by aldo-keto reductase 1B3 acting as
prostaglandin F2alpha synthase. J Biol Chem 285,
8880–8886.
16 Breyer RM, Bagdassarian CK, Myers SA & Breyer MD
(2001) Prostanoid receptors: subtypes and signaling.
Annu Rev Pharmacol Toxicol 41, 661–690.
17 Narumiya S, Sugimoto Y & Ushikubi F (1999) Prosta-
noid receptors: structures, properties, and functions.
Physiol Rev 79, 1193–1226.
18 Fujino H & Regan JW (2001) FP prostanoid receptor
activation of a T-cell factor ⁄ beta-catenin signaling path-
way. J Biol Chem 276, 12489–12492.
19 Pierce KL, Fujino H, Srinivasan D & Regan JW (1999)

Activation of FP prostanoid receptor isoforms leads
to Rho-mediated changes in cell morphology and in the
cell cytoskeleton. J Biol Chem 274, 35944–35949.
20 Watanabe T, Nakao A, Emerling D, Hashimoto Y,
Tsukamoto K, Horie Y, Kinoshita M & Kurokawa K
(1994) Prostaglandin F2 alpha enhances tyrosine
phosphorylation and DNA synthesis through phospho-
lipase C-coupled receptor via Ca(2+)-dependent intra-
cellular pathway in NIH-3T3 cells. J Biol Chem 269,
17619–17625.
21 Smith WL, DeWitt DL & Garavito RM (2000) Cyclo-
oxygenases: structural, cellular, and molecular biology.
Annu Rev Biochem 69, 145–182.
22 Smith WL & Song I (2002) The enzymology of prosta-
glandin endoperoxide H synthases-1 and -2. Prostaglan-
dins Other Lipid Mediat 68–69, 115–128.
23 Chu X, Nishimura K, Jisaka M, Nagaya T, Shono F &
Yokota K (2010) Up-regulation of adipogenesis in
adipocytes expressing stably cyclooxygenase-2 in the
antisense direction. Prostaglandins Other Lipid Mediat
91, 1–9.
24 Fajas L, Miard S, Briggs MR & Auwerx J (2003) Selec-
tive cyclo-oxygenase-2 inhibitors impair adipocyte dif-
ferentiation through inhibition of the clonal expansion
phase. J Lipid Res 44, 1652–1659.
25 Ghoshal S, Trivedi DB, Graf GA & Loftin CD (2011)
Cyclooxygenase-2 deficiency attenuates adipose tissue
differentiation and inflammation in mice. J Biol Chem
286
, 889–898.

26 Xie Y, Kang X, Ackerman WET, Belury MA, Koster
C, Rovin BH, Landon MB & Kniss DA (2006)
Differentiation-dependent regulation of the cyclooxy-
genase cascade during adipogenesis suggests a com-
plex role for prostaglandins. Diabetes Obes Metab 8,
83–93.
27 Gehart H, Kumpf S, Ittner A & Ricci R (2010) MAPK
signalling in cellular metabolism: stress or wellness?
EMBO Rep 11, 834–840.
28 Fletcher BS, Kujubu DA, Perrin DM & Herschman
HR (1992) Structure of the mitogen-inducible TIS10
gene and demonstration that the TIS10-encoded protein
is a functional prostaglandin G ⁄ H synthase. J Biol
Chem 267, 4338–4344.
29 Fujimori K & Amano F (2011) Niacin promotes adipo-
genesis by reducing production of anti-adipogenic
PGF(2alpha) through suppression of C ⁄ EBPbeta-acti-
vated COX-2 expression. Prostaglandins Other Lipid
Mediat 94, 96–103.
30 Cartharius K, Frech K, Grote K, Klocke B, Haltmeier
M, Klingenhoff A, Frisch M, Bayerlein M & Werner T
(2005) MatInspector and beyond: promoter analysis
based on transcription factor binding sites. Bioinformat-
ics 21, 2933–2942.
31 Fujitani Y, Aritake K, Kanaoka Y, Goto T, Takahashi
N, Fujimori K & Kawada T (2010) Pronounced adipo-
genesis and increased insulin sensitivity caused by over-
production of prostaglandin D in vivo. FEBS J 277,
1410–1419.
32 Lefterova MI & Lazar MA (2009) New developments

in adipogenesis. Trends Endocrinol Metab 20, 107–114.
33 Siersbaek R, Nielsen R & Mandrup S (2010) PPAR-
gamma in adipocyte differentiation and metabolism –
novel insights from genome-wide studies. FEBS Lett
584, 3242–3249.
T. Ueno and K. Fujimori Suppression of adipogenesis via FP-ERK-CREB-COX-2
FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS 2911
34 Craig EA, Stevens MV, Vaillancourt RR & Camenisch
TD (2008) MAP3Ks as central regulators of cell fate
during development. Dev Dyn 237, 3102–3114.
35 Tai CJ, Kang SK, Choi KC, Tzeng CR & Leung PC
(2001) Role of mitogen-activated protein kinase in pros-
taglandin f(2alpha) action in human granulosa-luteal
cells. J Clin Endocrinol Metab 86, 375–380.
36 Jabbour HN, Sales KJ, Boddy SC, Anderson RA &
Williams AR (2005) A positive feedback loop that regu-
lates cyclooxygenase-2 expression and prostaglandin
F2alpha synthesis via the F-series-prostanoid receptor
and extracellular signal-regulated kinase 1 ⁄ 2 signaling
pathway. Endocrinology 146, 4657–4664.
37 Chang MC, Chen YJ, Lee MY, Lin LD, Wang TM,
Chan CP, Tsai YL, Wang CY, Lin BR & Jeng JH
(2010) Prostaglandin F(2alpha) stimulates MEK-ERK
signalling but decreases the expression of alkaline
phosphatase in dental pulp cells. Int Endod J 43,
461–468.
38 Ahmed I, Gesty-Palmer D, Drezner MK & Luttrell LM
(2003) Transactivation of the epidermal growth factor
receptor mediates parathyroid hormone and prostaglan-
din F2 alpha-stimulated mitogen-activated protein

kinase activation in cultured transgenic murine osteo-
blasts. Mol Endocrinol 17, 1607–1621.
39 Tokuda H, Harada A, Hirade K, Matsuno H, Ito H,
Kato K, Oiso Y & Kozawa O (2003) Incadronate
amplifies prostaglandin F2 alpha-induced vascular
endothelial growth factor synthesis in osteoblasts.
Enhancement of MAPK activity. J Biol Chem 278,
18930–18937.
40 Sales KJ, Grant V & Jabbour HN (2008) Prostaglandin
E2 and F2alpha activate the FP receptor and up-regu-
late cyclooxygenase-2 expression via the cyclic AMP
response element. Mol Cell Endocrinol 285, 51–61.
41 Kang YJ, Mbonye UR, DeLong CJ, Wada M & Smith
WL (2007) Regulation of intracellular cyclooxygenase
levels by gene transcription and protein degradation.
Prog Lipid Res 46, 108–125.
42 Comb M, Birnberg NC, Seasholtz A, Herbert E &
Goodman HM (1986) A cyclic AMP- and phorbol
ester-inducible DNA element. Nature 323, 353–356.
43 Klein T, Shephard P, Kleinert H & Komhoff M (2007)
Regulation of cyclooxygenase-2 expression by cyclic
AMP. Biochim Biophys Acta 1773, 1605–1618.
44 Pham H, Shafer LM & Slice LW (2006) CREB-depen-
dent cyclooxygenase-2 and microsomal prostaglandin E
synthase-1 expression is mediated by protein kinase C
and calcium. J Cell Biochem 98, 1653–1666.
45 Zhu XO, Yang Z, Guo CM, Ni XT, Li JN, Ge YC,
Myatt L & Sun K (2009) Paradoxical stimulation of
cyclooxygenase-2 expression by glucocorticoids via a
cyclic AMP response element in human amnion fibro-

blasts. Mol Endocrinol 23, 1839–1849.
Supporting information
The following supplementary material is available:
Table S1. Nucleotide sequences of primers used in this
study.
This supplementary material can be found in the
online version of this article.
Please note: As a service to our authors and readers,
this journal provides supporting information supplied
by the authors. Such materials are peer-reviewed and
may be reorganized for online delivery, but are not
copy-edited or typeset. Technical support issues arising
from supporting information (other than missing files)
should be addressed to the authors.
Suppression of adipogenesis via FP-ERK-CREB-COX-2 T. Ueno and K. Fujimori
2912 FEBS Journal 278 (2011) 2901–2912 ª 2011 The Authors Journal compilation ª 2011 FEBS