(Received for publication, October 2, 1996, and in revised form, December 2, 1996)
From the The synthesis of prostaglandins (PGs) is
regulated by the arachidonic acid release by phospholipase
A2 (PLA2) and its conversion to PGs by
cyclooxygenase (COX). In the present study, we examined the regulation
of PG synthesis by interleukin (IL)-1 Prostaglandins (PGs)1 are potent
regulators of bone metabolism that are produced mainly by osteoblasts
in the bone tissue. Among several forms of PG, PGE2 is the
major product of osteoblasts, and its production is regulated by
several growth factors and cytokines including interleukin-1 (IL-1),
parathyroid hormone, basic fibroblast growth factor (bFGF), and
PGE2 itself (1-6). PGE2 is a complicated
regulator of bone metabolism, potently stimulating bone resorption
in vitro (7) and stimulating both bone formation and bone
resorption in vivo (8). PGE2 production is
involved in the mechanism of bone resorption induced by IL-1 in
vitro (7, 9).
PG synthesis is regulated by two successive metabolic steps, the
release of arachidonic acid from membranous phospholipids and its
conversion to prostanoids. Phospholipase A2
(PLA2), the enzyme responsible for arachidonic acid
release, consists of two forms, secretory PLA2
(sPLA2) and cytosolic PLA2 (cPLA2)
(10-13). Secretory PLA2 requires millimolar levels of
calcium for its activation and is divided into several groups, among
which pancreatic (type I) and nonpancreatic (type II) subtypes are well
characterized (10). Type II sPLA2 is widely distributed in
several tissues and inflammatory exudates. On the other hand,
cPLA2 is an intracellular enzyme that is found in various
cells and tissues including macrophages, mast cells, platelets, kidney,
and brain (10, 14, 15). This enzyme preferentially hydrolyzes
arachidonic acid at the Sn2 position of
membranous phospholipids and requires micromolar levels of calcium for
its activation. Both the phosphorylation and
Ca2+-dependent translocation of
cPLA2 to the membranes are essential for its activation
(16-20).
Cyclooxygenase (COX) is a rate-limiting enzyme for the conversion of
arachidonic acid to prostanoids. Two COX genes, constitutive COX
(cox-1) and inducible COX (cox-2), have been
identified (21-24). These glycosylated enzymes have 60% homology in
nucleic acid and amino acid sequences and a similar molecular mass
(70-74 kDa). cox-1 is constitutively expressed in several
mammalian tissues (21). In contrast, cox-2 is undetectable
under physiological conditions but markedly induced by several
cytokines and growth factors (2, 4, 6, 23, 24).
Both cox-1 and cox-2 are expressed in
osteoblastic cells. cox-2 is the main enzyme regulating PG
synthesis in response to several bone-resorbing factors such as IL-1,
bFGF, and PGE2 in osteoblastic cells and mouse calvarial
cultures (2-4, 6). We reported that IL-1 PLA2 in osteoblasts, however, has not been well
established. Kawaguchi et al. (25) have shown that
cPLA2 mRNA is detected in neonatal mouse calvarial
cultures. We reported that cPLA2 mRNA is constitutively
expressed in mouse osteoblastic cells and that it is slightly but
significantly increased by IL-1 In this study, we examined the role and regulation of PLA2
and COX in PG synthesis by primary osteoblastic cells isolated from
mouse calvaria. When osteoblastic cells were incubated with IL-1 Newborn and six-week-old male mice of
the ddy, C57BL/6 and Balb/c strains, were obtained from Shizuoka
Laboratory Animal Center (Shizuoka, Japan). Recombinant human IL-1 Primary
osteoblastic cells were isolated from 1-day-old ddy strain mouse
calvariae after five routine sequential digestions with 0.1%
collagenase (Wako Pure Chemicals, Osaka, Japan) and 0.2% dispase (Godo
Shusei, Tokyo, Japan) as described (2). Osteoblasts isolated from
fractions 3-5 were combined and cultured in The concentration
of PGE2 in the culture medium was determined using an
enzyme immunoassay (EIA) (Amersham, Aylesbury, UK). The antibody had
the following cross-reactivity when calculated by the bound/free ratio:
PGE2, 100%; PGE1, 7.0%;
6-keto-PGF1 Total cellular RNA was extracted
from osteoblastic cells using the acid guanidinium/phenol/chloroform
method (2). For Northern blots, 20 µg of total RNA was resolved by
electrophoresis in a 1.5% agarose-formaldehyde gel and transferred
onto nylon membranes (Hybond N, Amersham Corp.) and then hybridized
with a 32P-labeled cDNA probe as reported (2). The
signals were densitometrically quantified using a BioImage analyzer
(BAS-2000, Fuji Film, Tokyo, Japan). Mouse cox-1 and
cox-2 cDNA probes were purchased from Oxford Biomedical
Research, Inc. (Oxford, MI). The cDNA for cPLA2 was
used as a 2.4-kilobase EcoRI fragment of mouse
cPLA2 cDNA (12). The cDNA for type II
sPLA2 was used as a 375-base pair EcoRI fragment
of mouse type II sPLA2 cDNA (26).
Osteoblastic cells cultured in 60-mm
dishes were lysed at 4 °C in 70 µl of 50 mM Tris-HCl
(pH 7.4) containing 0.1% SDS, 0.5% Nonidet P-40 (BDH Laboratory, UK),
5 mM Na3VO4 (Wako), 50 µg/ml leupeptin (Sigma), 1.5 µM pepstatin A (Wako), and 1 µM phenylmethylsulfonyl fluoride (Wako). Proteins in the
cell lysate were measured using a BCA protein assay kit (Pierce).
Lysate protein (10 µg) was resolved on an SDS-polyacrylamide gel
(10%; Daiichi Pure Chemicals Co., Tokyo, Japan) and transferred to a
polyvinylidene difluoride membrane (Hybond-PVDF, Amersham Corp.). The
membrane was incubated for 18 h with 5% skim milk in
phosphate-buffered saline without calcium and magnesium containing
0.1% Tween (Polysciences Inc., Warrington, PA) at 4 °C to block
nonspecific binding and further incubated for 2 h with polyclonal
rabbit anti-cox-1 antibody (donated from Dr. W. L. Smith,
Michigan State University, East Lansing, MI), anti-cox-2
antibody (Cayman Chemical Co., Ann Arbor, MI), or
anti-cPLA2 antibody (Santa Cruz Biotechnology, Santa Cruz,
CA) at a dilution of 1:5000, 1:1000, and 1:100, respectively. After
incubation with horseradish peroxidase-conjugated donkey anti-rabbit
IgG (Amersham Corp.) (1:5000 dilution) for 1 h, immunoreactive
bands were stained by an enhanced chemiluminescence Western blot
analysis system (Amersham Corp.).
Osteoblastic cells
were washed once with ice-cold phosphate-buffered saline without
calcium and magnesium and suspended in 1 ml of 10 mM
Tris-HCl (pH 7.4) containing 1 mM EDTA, 250 mM
sucrose, and 1 µM phenylmethylsulfonyl fluoride, and
sonicated to make a cell lysate. Cell lysate protein (2 µg) was
assayed for PLA2 activity using 2 µM
1-palmitoyl-2-[14C]arachidonyl-phosphatidylethanolamine
(DuPont NEN) as a substrate in 250 µl of 50 mM Tris-HCl
(pH 9.0) containing 4 mM CaCl2 (27). In some
experiments,
1-palmitoyl-2-[14C]linoleoyl-phosphatidylethanolamine
(Amersham Corp.) was also used as a substrate. After an
incubation at 37 °C for 20 min, the reaction was stopped by adding
1.25 ml of Dole's reagent, and the amount of hydrolyzed
[14C]arachidonic acid was measured as described (27).
DNA used
in the genomic analysis of mouse type II sPLA2 gene was
prepared from the brains of ddy, Balb/c and C57BL/6 mice. DNA (10 µg)
was digested with BamHI, fractionated on a 0.8% agarose gel, and transferred to a Nylon membrane (Pall BioSupport, East Hill,
NY) and Southern blotted as reported (28). The membrane was hybridized
with 32P-labeled cDNA probe for mouse type II
sPLA2 for 18 h at 42 °C in 40% formamide, 5 × SSC, 10 × Denhardt's solution, 250 µg/ml denatured salmon
testicular DNA, and 1% SDS. The membrane was washed twice at 55 °C
in 0.1 × SSC containing 0.1% SDS, and exposed to an x-ray
film.
We reported that IL-1
Fig. 2A shows the time course of
change in the rapid synthesis of PGE2 induced by
arachidonic acid. After osteoblastic cells were cultured for 24 h
with or without IL-1
To identify the factor(s) that increases the endogenous level of
arachidonic acid for PG synthesis, we examined the effects of various
growth factors and cytokines that may be involved in the PG synthesis
by osteoblastic cells. As shown in Fig. 2B, PDGF-BB markedly
and specifically stimulated the rapid synthesis of PGE2 within 10 min. PDGF-BB affected the rapid synthesis of PGE2
in a time- (2-10 min) and dose (0.1-100 ng/ml) -dependent
manner (Fig. 2B; data not shown). The potency of 100 ng/ml
PDGF-BB was identical to that of 1 µM arachidonic acid in
PG synthesis. PGE2 synthesis was also enhanced by PDGF-AA
and TGF- PG synthesis is regulated by the release of arachidonic
acid from membranous phospholipids by PLA2 and its
conversion into PGH2, a precursor of PGs, by COX. To
examine the effect of PDGF-BB on PG synthesis, we measured
PLA2 activity by the release of
[14C]arachidonic acid from
1-palmitoyl-2-[14C]arachidonyl-phosphatidylethanolamine
using osteoblastic cell lysates. PLA2 activity was doubled
upon PDGF exposure for 10 min in the control group cultured for 24 h (Fig. 3). When osteoblastic cells were cultured for
24 h with IL-1
Type II sPLA2 and cPLA2 are
involved in the production of eicosanoids in various cells and tissues
(29-32). We determined which PLA2 is the enzyme
responsible for PGE2 synthesis in mouse osteoblastic cells
by means of Northern and Western blot analysis. Fig.
4A shows that cPLA2 mRNA was
constitutively expressed in mouse osteoblastic cells, and it increased
at 6 h in the presence of IL-1
The murine sPLA2 gene is naturally disrupted in some inbred
mouse strains such as C57BL/6 (28, 33). Therefore, we performed genomic
analysis of sPLA2 in ddy mice which was used in this study. C57BL/6 mice had a disrupted sPLA2 genotype but that of
Balb/c mice was normal, as reported (28, 33). Ddy mice had a normal sPLA2 genotype (Fig. 5A). In accordance with the
sPLA2 genotype, sPLA2 mRNAs were expressed
at high levels in the small intestine of Balb/c and ddy mice but not in
that of C57BL/6 (data not shown). We did not detect sPLA2
mRNA in osteoblastic cells from C57BL/6, Balb/c, and ddy mice. In
contrast, cPLA2 mRNA was expressed in osteoblastic
cells from all of the mouse strains tested (Fig. 5B).
Despite differences in the sPLA2 genotype, the osteoblastic cells from all of the strains of mice similarly responded to IL-1
Characterization of PLA2 activity in mouse osteoblastic cells
Department of Biochemistry,
in primary mouse osteoblastic
cells isolated from mouse calvaria. Although IL-1
greatly enhanced
cox-2 mRNA expression and its protein levels, PGE2 was not produced until 24 h. When arachidonic
acid was added to osteoblastic cells precultured with IL-1
for
24 h, PGE2 was produced within 10 min. Of several
growth factors tested, platelet-derived growth factor (PDGF)
specifically initiated the rapid synthesis of PGE2, which
was markedly suppressed by a selective inhibitor of cox-2
(NS-398). In mouse osteoblastic cells, cytosolic PLA2 (cPLA2) mRNA and its protein were constitutively
expressed and increased approximately 2-fold by IL-1
, but secretory
PLA2 mRNA was not detected. PDGF rapidly stimulated
PLA2 activity, which was blocked completely by a
cPLA2 inhibitor (arachidonyltrifluoromethyl ketone). The
PDGF-induced cPLA2 activation was accompanied by phosphorylation of its protein. These results indicate that
cox-2 induction by IL-1
is not sufficient, but
cPLA2 activation by PDGF is crucial for IL-1
-induced
PGE2 synthesis in mouse osteoblasts.
markedly stimulates the
mRNA expression of cox-2, but not of cox-1,
in primary osteoblastic cells and that IL-13 and IL-4 inhibit
IL-1-induced bone resorption by suppressing cox-2-dependent PG synthesis (2). Kawaguchi
et al. (4) reported that bFGF rapidly induces
cox-2 expression in osteoblasts, which is responsible for
bFGF-induced bone resorption.
(2).
,
both the mRNA expression and the protein level of cox-2 were greatly increased, but no PGE2 was produced. When
arachidonic acid or platelet-derived growth factor (PDGF)-BB was added
to osteoblastic cells precultured with IL-1
for 24 h,
PGE2 was produced within 10 min. The rapid synthesis of
PGE2 induced by PDGF was due to the activation and
phosphorylation of cPLA2, indicating that the activation of
cPLA2 by PDGF is essential for
cox-2-dependent PGE2 synthesis in
osteoblasts cultured with IL-1
.
Animals and Reagents
and epidermal growth factor (EGF) were purchased from Genzyme
(Cambridge, MA). Recombinant human PDGF-BB and PDGF-AA and transforming
growth factor (TGF)-
were obtained from Life Technologies, Inc.
Recombinant human TGF-
1 was purchased from R&D systems (Minneapolis,
MN). Recombinant bovine bFGF was obtained from Boehringer Mannheim
Biochemica (Mannheim, Germany). Arachidonic acid was purchased from
Serdary Research Laboratories, Inc. (London, Canada). Aspirin was
obtained from Sigma. Arachidonyltrifluoromethyl ketone
(AACOCF3) and NS398 were purchased from Calbiochem.
-minimal essential
medium (
-MEM) supplemented with 10% fetal calf serum (FCS) at
37 °C in a humidified atmosphere of 5% CO2 in air. In
most experiments, osteoblastic cells were cultured for 24 h in
-MEM supplemented with 1% FCS and then incubated with or without
IL-1
.
, 5.4%; PGF2
, 4.3%; and
PGD2, 1.0%.
cox-2 Induction Is Not Sufficient for the IL-1-induced PG
Synthesis in Mouse Osteoblastic Cells
markedly stimulates cox-2 mRNA expression in primary
mouse osteoblastic cells (2). When mouse osteoblastic cells were
cultured in
-MEM containing 1% FCS for 24 h and then incubated
with 2 ng/ml IL-1
, the cytokine markedly enhanced cox-2 mRNA expression at 3 h and its protein levels at 24 h
(Fig. 1, A and B).
cox-1 mRNA and its protein were weakly expressed in osteoblastic cells, but they were not appreciably changed by IL-1
. To examine the effects of IL-1
on PGE2 production, we
measured the concentration of PGE2 in the conditioned media
collected at 24 h. Despite the marked stimulation of
cox-2 mRNA expression, IL-1
did not stimulate
PGE2 production. However, when 1 µM
arachidonic acid was added to osteoblastic cells precultured with
IL-1
for 24 h, the PGE2 levels were greatly
increased within 10 min (Fig. 1C). Arachidonic acid also
slightly stimulated PGE2 synthesis in the control cultures
without IL-1
. These results suggest that cox-2 induction
is not sufficient and that the simultaneous addition of arachidonic
acid is essential for the IL-1
-induced PGE2
synthesis.
Fig. 1.
cox-2 induction by IL-1 is not
sufficient for PGE2 production in mouse osteoblastic cells.
A, to detect cox-1 and cox-2
mRNAs, mouse osteoblastic cells were cultured for 24 h in
-MEM containing 1% FCS and then with 2 ng/ml IL-1
. At 3 h,
total RNA was extracted and Northern blotted using
32P-labeled cox-1, cox-2, and tubulin
cDNA probes. B, to detect cox-1 and
cox-2 proteins under the same conditions shown in
A, cell lysates were extracted from osteoblastic cells
cultured with IL-1
for 24 h. Lysate protein (10 µg) was
resolved by SDS-polyacrylamide gel electrophoresis and immunoblotted
using anti-cox-1 and anti-cox-2 antibodies as
described under "Materials and Methods." C, to examine PGE2 production, osteoblastic cells were cultured under the
same conditions shown in B. At the end of culture, 1 µM arachidonic acid or vehicle (ethanol) was added to the
medium, and the conditioned media were collected 10 min thereafter.
Concentrations of PGE2 in the conditioned media were
measured by EIA. Data are expressed as the means ± S.D. of 3-6
independent experiments.
[View Larger Version of this Image (25K GIF file)]
in
-MEM containing 1% FCS, they were washed
twice with
-MEM and incubated for 2-10 min in
-MEM containing 1 µM arachidonic acid. PGE2 synthesis was
induced within 10 min. Cells cultured without IL-1
also synthesized PGE2 in the presence of arachidonic acid but to a lesser
extent than that in IL-1
-treated cells. These results suggest that
cox-2 induced by IL-1
rapidly converts exogenous
arachidonic acid into PGE2 in osteoblastic cells.
Fig. 2.
Effects of arachidonic acid and growth
factors on the rapid induction of PGE2 synthesis in mouse
osteoblastic cells. A, time course of PGE2
production induced by arachidonic acid in osteoblastic cells cultured
for 24 h in the presence or absence of IL-1 (2 ng/ml). After
culture for 24 h in
-MEM containing 1% FCS with (
) or
without (
) IL-1
, cells were washed twice with
-MEM and then
incubated with 1 µM arachidonic acid for 2-10 min.
B, effects of various growth factors on the rapid induction of PGE2 synthesis were examined under the same conditions
as shown in A. After culture for 24 h with or without
IL-1
, cells were washed and incubated for 10 min with PDGF-BB (100 ng/ml), PDGF-AA (100 ng/ml), EGF (100 ng/ml), TGF-
(100 ng/ml),
TGF-
(10 ng/ml), bFGF (100 ng/ml), or vehicle (
-MEM).
Inset, time course of PGE2 production induced by
PDGF-BB (100 ng/ml) was examined under the same conditions as shown in
A. PGE2 levels in the conditioned media were
measured by EIA. Data are expressed as means ± S.D. of 3-6
cultures.
[View Larger Version of this Image (26K GIF file)]
as well, but the effects were marginal. Other growth
factors such as EGF, TGF-
, and bFGF did not affect the rapid
induction of PG synthesis within 10 min. These results indicate that
PDGF-BB mimics the effect of exogenous arachidonic acid upon the rapid
synthesis of PGE2 by mouse osteoblastic cells.
, PLA2 activity was increased to a
higher level than in the control culture, and it was further stimulated
by PDGF-BB (Fig. 3).
Fig. 3.
Effect of PDGF on PLA2 activity
in mouse osteoblastic cells. After culture for 24 h in
-MEM containing 1% FCS with or without IL-1
(2 ng/ml), cells
were washed twice with
-MEM and then incubated with PDGF-BB (100 ng/ml) or vehicle (
-MEM) for 10 min. PLA2 activity was
assayed in 2 µg of cell lysate protein. Using 2 µM
1-palmitoyl-2-[14C]arachidonyl-phosphatidylethanolamine
as the substrate, the release of [14C]arachidonic acid
was detected as described under "Materials and Methods." Data are
expressed as the means ± S.D. of four independent experiments.
[View Larger Version of this Image (20K GIF file)]
. Concomitantly, cPLA2 protein was detected in osteoblastic cells, and the
level was almost doubled at 24 h in the presence of IL-1
(Fig.
4B). On the other hand, no sPLA2 mRNA was
detected in mouse osteoblastic cells by Northern blot analysis (Fig.
5B) and reverse transcription polymerase
chain reaction (data not shown).
Fig. 4.
Effects of IL-1 on the expression of
cPLA2 mRNA and its protein levels in mouse osteoblastic
cells. Osteoblastic cells were cultured for 24 h in
-MEM
containing 1% FCS with or without IL-1
(2 ng/ml). A, to
detect cPLA2 mRNA, total RNA was extracted at 6 and
24 h and Northern blotted using 32P-labeled cDNA
probe of cPLA2 as described under "Materials and Methods." B, to detect cPLA2 protein, cell
lysates were extracted from osteoblastic cells at 6 and 24 h, and
then 10 µg of cell lysate protein was resolved by SDS-polyacrylamide
gel electrophoresis and immunoblotted using anti-cPLA2
antibody as described under "Materials and Methods."
[View Larger Version of this Image (41K GIF file)]
Fig. 5.
Genomic analysis of mouse sPLA2
and the lack of sPLA2 mRNA expression in mouse
osteoblastic cells. A, genomic analysis of sPLA2
gene was performed using Balb/c, C57BL/6, and ddy strains of mice. DNA
was prepared from the brains of each strain. Genomic DNA (10 µg) was
digested with BamHI and Southern blotted using the
32P-labeled cDNA probe of mouse sPLA2
described under "Materials and Methods." +/+, normal
sPLA2 gene; /
, disrupted sPLA2 gene. B, expression of sPLA2 and cPLA2
mRNAs in mouse osteoblastic cells derived from three strains of
mice. After culture for 24 h with IL-1
(2 ng/ml) in
-MEM
containing 1% FCS, total RNA was extracted from osteoblastic cells of
various mouse strains and Northern blotted using
32P-labeled cDNA probes of sPLA2 and
cPLA2. C, PDGF-dependent rapid synthesis of PGE2 was measured in mouse osteoblastic cells
cultured with IL-1
. After culture for 24 h with IL-1
(2 ng/ml) in
-MEM containing 1% FCS, cells were washed twice with
-MEM and then treated with PDGF-BB (100 ng/ml) or vehicle (
-MEM)
for 10 min. PGE2 levels in the conditioned media were
measured by EIA. Data are expressed as the means ± S.D. of 3-6
cultures.
[View Larger Version of this Image (29K GIF file)]
and PDGF in PG synthesis (Fig. 5C), suggesting that
cPLA2 is the enzyme responsible for arachidonic acid
release in mouse osteoblastic cells. To confirm this notion, we used
1-palmitoyl-2-[14C]arachidonyl-phosphatidylethanolamine
and 1-palmitoyl-2-[14C]linoleoyl-phosphatidylethanolamine
as a substrate and compared PLA2 activities in osteoblastic
cells (Table I). It is reported that cPLA2,
but not sPLA2, preferentially hydrolyzes arachidonic acid
at the Sn2 position of phospholipids (12-15).
As shown in Table I, arachidonyl-phosphatidylethanolamine was a
selective substrate for the PLA2 activity in mouse
osteoblastic cells. Furthermore, the PLA2 activity was
completely inhibited by 1 µM AACOCF3, a competitive inhibitor of cPLA2 (Table I) and by EDTA (data
not shown). These results indicate that cPLA2 is
selectively expressed and is the enzyme responsible for arachidonic
acid release in mouse osteoblastic cells.
-MEM containing
1% FCS with or without IL-1
(2 ng/ml), washed twice with
-MEM, and then incubated for 10 min with vehicle (
-MEM) or PDGF-BB (100 ng/ml). Using 2 µM
1-palmitoyl-2-[14C]arachidonyl-phosphatidylethanolamine
(Ara-PE) and
1-palmitoyl-2-[14C]linoleoyl-phosphatidylethanolamine
(Lino-PE) as a substrate, PLA2 activity was measured as
described under "Materials and Methods." To examine the effects of
AACOCF3 on PLA2 activity, [14C]Ara-PE was
used as the substrate and AACOCF3 (1 µM) was
added to the cell lysate 30 min before the substrate. Data are
expressed as the means ± S.D. of four independent experiments.
UD, undetectable.
Treatment
PLA2 activity
24
h
10 min
Ara-PE
Lino-PE
Ara PE + AACOCF3 (1 µM)
pmol/min/mg
protein
Vehicle
Vehicle
103
± 6
11 ± 9
UD
Vehicle
PDGF
203 ± 4
19 ± 9
UD
IL-1
Vehicle
171 ± 19
18 ± 5
UD
IL-1
PDGF
415 ± 30
36 ± 11
UD
The experimental results described above
suggest that PDGF rapidly activates cPLA2 in mouse
osteoblastic cells. It appears that the cPLA2 activation is
associated with increased phosphorylation of the protein, which causes
a decreased mobility in SDS-polyacrylamide gel electrophoresis
(16-18). To determine if the PDGF-induced rapid stimulation of
PLA2 activity is accompanied by the mobility change in the
cPLA2 protein, we performed Western blot analyses. Exposing osteoblastic cells to PDGF for 10 min decreased the mobility of cPLA2 (Fig. 6), suggesting that PDGF
stimulates phosphorylation of this protein. IL-1 did not affect the
phosphorylation of cPLA2 by PDGF (Fig. 6). These results
suggest that PDGF phosphorylates and activates cPLA2, which
in turn triggers arachidonic acid release.
Effects of COX Inhibitors on the PG Synthesis Regulated by PDGF and IL-1
COX is essential for the
conversion of arachidonic acid into prostanoids. As described in Fig.
1, mouse osteoblastic cells expressed both cox-1 and
cox-2 mRNA and proteins, but IL-1 preferentially stimulated cox-2 expression. To study the contribution of
cox-1 and cox-2 to IL-1
-induced
PGE2 synthesis in osteoblasts, we examined the effects of
COX inhibitors on the rapid induction of PGE2 synthesis by
PDGF in IL-1
-treated osteoblastic cells. When cells were
preincubated for 24 h with aspirin to block preexistent
cox-1 and cox-2 activities, washed, and further
cultured for 24 h with or without IL-1
, the PDGF-dependent rapid synthesis of PGE2 was
partially suppressed (Fig. 7). In contrast, NS-398, a
selective inhibitor of cox-2, added 3 h before IL-1
,
completely blocked the PGE2 production (Fig. 7). These
results indicate that the induction of cox-2 by IL-1
is
also critical for the rapid synthesis of PGE2 triggered by
PDGF.
The present study indicates that cox-2 induction is not
sufficient but that cPLA2-dependent arachidonic
acid release is essential for the PGE2 synthesis induced by
IL-1. Arachidonic acid release was mediated by cPLA2 but
not sPLA2, and the activation of cPLA2 accompanying its phosphorylation was crucial for its function in mouse
osteoblastic cells. We also found that PDGF-BB markedly stimulated the
phosphorylation and activation of cPLA2, which triggered
the rapid induction of PGE2 synthesis in mouse osteoblastic cells. Raisz and his co-authors (4) have reported that osteoblasts cultured in serum-free medium do not produce PGs although
cox-2 is induced by cytokines such as bFGF. We reproduced
their results. Furthermore, the activation of cPLA2 was
essential for the PGE2 synthesis induced by IL-1
in
mouse osteoblastic cells.
PLA2 is the enzyme responsible for the release of
arachidonic acid, and it consists of low molecular mass (14 kDa)
sPLA2 and high molecular mass (60-110 kDa)
cPLA2 (10-13). The former is linked to prolonged PG
synthesis stimulated by cytokines in various cells such as mesangial
cells and macrophage-like P388 D1 cells (29, 30). In this study, we did
not detect sPLA2 mRNA expression in mouse osteoblastic
cells collected from the ddy strain of mice. The sPLA2 gene
is naturally disrupted in some inbred strains of mice such as C57BL/6,
129/Sv, and A/J (28, 33). Therefore, it is possible that the absence of
sPLA2 mRNA in osteoblastic cells was due to natural
disruption of the gene in ddy mice. However, the sPLA2 gene
was normal and sPLA2 mRNA was expressed at high levels
in the small intestine of ddy mice (Fig. 5). MacPhee et al.
(33) have reported that the sPLA2 gene is a major candidate for the mom1 locus, which is closely associated with the
development of intestinal tumors and characterized in multiple
intestinal neoplasia strain of mice. They also showed that there is
100% concordance between the sPLA2 genotype and tumor
susceptibility. This suggests that the disruption of sPLA2
could develop intestinal polyps and tumors. In this study,
sPLA2 was not detected in mouse osteoblastic cells, but
PGE2 synthesis was similarly induced by IL-1 and PDGF in
all of the strains of mice tested (Balb/c, C57BL/6, and ddy). These
observations suggest that there is a difference in the gene regulation
and biological significance of sPLA2 between intestine and
bone.
Cytosolic PLA2 is an arachidonyl-selective PLA2
(14, 15). It is suggested that calcium
(Ca2+)-dependent translocation to the membrane
is essential for the functional activation of cPLA2. Clark
et al. (12) reported that cPLA2 possesses a
Ca2+-dependent phospholipid-binding domain at
the amino terminus of cPLA2. In addition, Lin et
al. (17) reported that the activation of PLA2 is
associated with the increased phosphorylation of its protein. The
phosphorylation of cPLA2 is thought to be mediated by
mitogen-activated protein kinase and/or protein kinase C (17, 34). The
phosphorylation and translocation of cPLA2 to the membranes are important in the arachidonic acid release and PG synthesis in
several cells, for example, macrophages regulated by colony-stimulating factor 1 (35), platelets by thrombin (18), smooth muscle cells by PDGF
(32), and endothelial cells by bFGF (36). The present study showed that
the activity and phosphorylation of cPLA2 are stimulated by
PDGF in mouse osteoblastic cells. It has been reported that PDGF
stimulates calcium influx and elevates the level of intracellular
calcium in mouse osteoblastic cells (37). When mouse osteoblastic cells
were treated with thapsigargin for 1 h which has the capacity to
deplete intracellular calcium (32), or EDTA for 10 min, PDGF-induced
PGE2 synthesis was significantly suppressed (data not
shown). The activity of cPLA2 induced by PDGF was also
completely abolished by EDTA (data not shown). These results suggest
that PDGF induces functional activation of cPLA2 in mouse
osteoblastic cells and that both the elevation of intracellular calcium
and the phosphorylation of the cPLA2 protein are indeed involved in the process of the activation. IL-1 activates
cPLA2 for the arachidonic acid release and also increases
its protein level in some cells, such as rat mesangial cells and human
fibroblasts (31, 38, 39). We also found that IL-1 increases
cPLA2 protein levels in mouse osteoblastic cells (Fig. 4).
However, IL-1
induced neither rapid stimulation of cPLA2
activity nor its phosphorylation in mouse osteoblastic cells (data not
included). Furthermore, the increase in cPLA2 protein by
IL-1
was not accompanied by concomitant PGE2 synthesis
until 24 h, despite the marked induction of cox-2.
Therefore, we concluded that IL-1
does not induce PGE2 synthesis without PDGF.
Conversion of arachidonic acid into PGH2 is mediated by
cox-1 and cox-2. Mouse osteoblastic cells
expressed both cox-1 and cox-2, but IL-1
specifically stimulated cox-2 expression. The rapid
induction of PGE2 synthesis induced by PDGF in osteoblastic cells cultured with IL-1
was markedly suppressed by NS-398, a selective inhibitor of cox-2. These results indicate that
the endogenous arachidonic acid released by PDGF-dependent
activation of cPLA2 is essential for IL-1
-induced
PGE2 synthesis. Morita et al. (40) examined the
subcellular location of cox-1 and cox-2 in mouse
3T3 fibroblasts using quantitative confocal immunofluorescence microscopy studies. They found that both cox-1 and
cox-2 were distributed in the endoplasmic reticulum (ER) and
nuclear envelope (NE) but not in the plasma membrane. Glover et
al. (19) and Schievella et al. (20) have shown that
cPLA2 is translocated into ER and NE, but not into the
plasma membrane, through a Ca2+-dependent
mechanism. These results suggest that arachidonic acid is produced and
metabolized at the NE and ER. Murakami et al. (41) reported
that there is a linkage between cPLA2 and cox-1 in c-kit ligand-primed, IgE-dependent PGD2
synthesis in mouse mast cells. A linkage between cPLA2 and
cox-2 has also been suggested in monocytes and fibroblasts
(42, 43). Therefore, it is likely that PG synthesis is mediated by
cPLA2 and COX cooperatively, both of which may be
co-localized in NE and ER.
PG synthesis in osteoblasts is regulated by several factors including
IL-1, parathyroid hormone, bFGF, TGFs, and PGE2 itself (1-6). Studies have indicated that cox-2 is the major
enzyme regulating PG synthesis in osteoblasts. Cytokines such as IL-1 and bFGF greatly stimulated cox-2-dependent
PGE2 production in mouse long bone and calvarial cultures
(2, 4). However, osteoblastic cells cultured in serum-free medium or
medium containing only 1% FCS did not produce PGE2 despite
the marked induction of cox-2 by these cytokines (Fig. 1,
Refs. 2 and 4). Preliminary experiments showed that IL-1 markedly
stimulated PGE2 production, when osteoblastic cells were
cultured in medium containing 10% FCS (data not shown, Refs. 6 and
44). Adding 10% FCS significantly stimulated PLA2 activity
in cultures of mouse osteoblastic cells (data not shown). The
discrepancy in PG synthesis between organ and cell culture may be due
to the presence or absence of a bone marrow-derived growth factor(s)
that activates cPLA2 to release arachidonic acid. To
identify the factor(s) in bone marrow that activates cPLA2
needs further investigation.
Several lines of evidence have suggested that PDGF is an important regulator of bone metabolism. PDGF is stored in bone matrix (45) and stimulates both bone formation and bone resorption (46-48). PDGF is a major growth factor in serum. However, the fact that PDGF is produced by human osteoblasts and mouse bone marrow-derived stromal cells (49, 50) indicates that PDGF is also a local factor produced by bone cells. The present study showed that PDGF is a potent stimulator of PGE2 synthesis in osteoblasts. Bone resorption induced by PDGF is suppressed by indomethacin (48). Thus, we speculate that PDGF is an important local regulator of bone metabolism by activating cPLA2 in bone.
In conclusion, IL-1-induced cox-2 induction is not
sufficient for PGE2 production by osteoblasts. The rapid
activation of cPLA2 by PDGF is essential for
cox-2-dependent PGE2 synthesis. The
two-step regulation of PGE2 synthesis, cPLA2
activation by PDGF, and cox-2 induction by IL-1
will help
understand the role of PGE2 in several metabolic bone
diseases.
We thank Dr. W. L. Smith (Michigan State University, East Lansing, MI) for his generous gift of antibody for cox-1.