| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Articles |
Laboratoire de Physiologie Animale, Institut National de la Recherche Agronomique (G.C., P.R., J-P.T., J.Mar.), 78352 Jouy-en-Josas Cedex; Commissariat à l’Energie Atomique, Service de Pharmacologie et d’Immunologie, Centre d’Etudes Saclay (C.C.), 91191 Gif-sur-Yvette Cedex; I.F.R. Biologie de la Circulation-Lariboisière, INSERM U 348, Hôpital Lariboisière (J.Mac.), 75475 Paris Cedex 10; and Unité Recherche Associée CNRS 1291, laboratoire de Physiologie Animale, Institut National de la Recherche Agronomique (M.G.), 78352 Jouy-en-Josas Cedex, France
Address all correspondence and requests for reprints to: G. Charpigny; laboratoire de Physiologie Animale, Institut National de la Recherche Agronomique, 78352 Jouy-en-Josas cedex, France. E-mail: gch{at}jouy.inra.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
during the luteolytic
phase of the estrous cycle. During luteolysis PGF2 is secreted in a
series of high-amplitude, short duration pulses (1, 2) in response to
peaks in luteal oxytocin secretion (for review see Ref.3). In early
pregnancy the loss of the uterine responsiveness to oxytocin causes the
suppression of the pulsatile pattern of PGF2 secretion and results
in corpus luteum maintenance (reviewed Ref. 4 and 5). However, the
level of basal secretion of PGs is enhanced. The mechanism by which the
oxytocin receptor is suppressed in pregnant ewes is partially
understood. Interferon-
, a 20-kDa protein that is produced by the
trophectoderm of the ovine conceptus between days 10–21 (6, 7),
inhibits the endometrial estrogen receptor and consequently prevents
the estrogen-induced increase in oxytocin receptor number (8). Despite
alterations in the dynamics of PGF2 secretion, the uterus maintains
the same ability to synthesize PGF2 during the luteal phase of
estrous cycle and in comparable stages of early pregnancy.
Other reports have shown that the increase in the ability of the uterus
to synthesize PGs corresponds to changes in the PG synthase enzyme,
which converts arachidonic acid into PGs. In sheep, Huslig et
al. (9) demonstrated that the increase in PGF2 that occurs
around the time of luteolysis corresponds to an 8-fold increase in the
concentration of uterine PG synthase. However, these changes in enzyme
protein were not reflected at the messenger RNA (mRNA) level (10). No
significant differences were seen in PG synthase mRNA concentrations
analyzed by Northern blot throughout the estrous cycle. This
discrepancy could be explained by an increase in the translation rate
of the protein despite the steady state levels of mRNA transcripts. The
extension of the half-life of the enzyme may also account for the
enhanced PG synthase concentration. However, we hypothesize that the
apparent conflicting results between the levels of enzyme protein and
mRNA reflect the existence of two distinct types of PG synthase.
Actually, two isoforms of PG synthase have been recently identified:
cyclooxygenase-1 (cox-1) and cyclooxygenase-2 (cox-2) (for review see
Refs. 11 and 12). The apparent molecular weight of the two isoforms is
nearly 72 kDa on a SDS-PAGE gel. The additional sites of glycosylation
on cox-2 give rise to an second band at 74 kDa (13). Deduced amino acid
sequences showed approximately 60% homology between the two isoforms
within species, whereas 90% homology was observed among the cox-2 of
different species (14). The cox-2 sequence contains an additional
18-amino acid insert in the C-terminal region as compared with the
sequence of cox-1. This short 18-amino acid tail made it possible to
raise specific antibodies against cox-2 (15, 16). Moreover, the
cyclooxygenases differ in their pattern of expression (for review see
Refs. 17 and 18). Cox-1 is a constitutive enzyme that is expressed in
many tissues to ensure the synthesis of PGs for the so-called
housekeeping functions (17). In contrast, cox-2 is an inducible enzyme
that is normally absent from cells and appears as an early responsive
gene triggered by a wide variety of factors [mitogens (19, 20),
interleukin-1 (15, 21), and growth factor (22)]. Cox-2 has been
implicated in reproductive events such as ovulation, embryo
implantation, and parturition. It is induced in rat (23, 24) and bovine
(25) follicles at the time of ovulation under hormonal stimulation.
Cox-2 is also expressed at the time of blastocyst implantation in the
rat uterus (26, 27) and in human decidua (28). In sheep, cox-2 is
strongly induced in the placenta at the onset of parturition
(29, 30, 31).
This study was designed to determine whether the ovine endometrium
expresses cox-2 during the estrous cycle and early pregnancy and to
characterize the cell types containing the PG synthase isoforms.
Moreover, it has been reported that estradiol and progesterone exert
pronounced effects on the release of endometrial PGF2 from explants
(32) and cultured cells (33), suggesting a direct effect of the ovarian
steroids on PG synthase. We examined the developmental and hormonal
regulation of the two cyclooxygenases using steroid replacement
treatments in ovariectomized ewes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cyclic and pregnant ewes.
Ewes of the Préalpes-du-Sud
breed were used. Estrus was synchronized using intravaginal sponges
containing 60 mg 6-methyl-17-acetoxyprogesterone for 14 days as
previously described by Peterson et al. (34). On the day of
sponge withdrawal, ewes received one injection im of 500 IU PMSG.
Estrus was observed 48 h later (day 0 of the cycle), and those
assigned to be pregnant were mated. On days 6, 9, 12, 15, and 17 of the
estrous cycle or pregnancy, the ewes were slaughtered. Five ewes at
each stage of estrous cycle or pregnancy were used. The uterus was
removed and excised. Conceptuses from mated ewes were removed by
flushing the uterine horns to confirm pregnancy. A piece of the uterine
horn ipsilateral to the corpus luteum was preserved for
immunocytochemical analysis. The endometrium was separated from the
myometrial layer by dissection. Five hundred milligrams (wet wt) of the
endometrium were reserved for explant production. The remaining
endometrium and myometrium were frozen in liquid nitrogen and stored at
-80 C until further analysis.
Steroid-treated ovariectomized ewes.
Twenty-one ewes were
ovariectomized via midventral laparotomy under general anesthesia and
were allowed to recover for 4 weeks. The ewes were then randomly
allocated to three control groups and four experimental groups (3
animals/group) and were given progesterone and estradiol according to a
protocol (35) aimed at mimicking the changes in steroid plasma
concentrations observed during an estrous cycle. Day 0 was the day of
the beginning of the steroid treatment and day -12 was the day of the
beginning of the steroid pretreatment. All ewes were killed on day 12.
Ewes of the three control groups were treated as described below. Group
C, which was the nontreated ovariectomized control group, was exposed
to no treatment from days -12–0, and exposed to im injections of 1 ml
vehicle twice daily from day 1 to slaughter on day 12. Group C-P was
exposed to intravaginal sponges containing 60 mg
6-methyl-17-acetoxyprogesterone from days -12 to -1, followed
by im injections of 1 ml vehicle twice daily from days -1–12. Group
C-E was exposed to intravaginal sponges containing 60 mg
6-methyl-17-acetoxyprogesterone from days -12 to -1, followed
by four im injections of 25 µg estradiol at 12-h intervals on days
-1 and 0, and im injections of 1 ml vehicle twice daily from days
1–12. The animals of the four experimental groups were primed for 12
days (days -12–0) by exposure to intravaginal sponges containing 60
mg 6-methyl-17-acetoxyprogesterone, followed by four im
injections of 25 µg estradiol at 12-h intervals on days -1 and 0.
Then the ewes received various combinations of progesterone and
estradiol injections as described below. Group P10 received im
injections of progesterone at 12-h intervals from days 1–10, then im
injections of 1 ml vehicle twice daily for days 11 and 12. Group P10-E
was treated as group P10 and followed by four im injections of 25 µg
estradiol at 12-h intervals on days 11 and 12. Group P12 was exposed to
im injections of progesterone at 12-h intervals from days 1–12. Group
P12-E was treated as group P12 plus four im injections of 25 µg
estradiol at 12-h intervals on days 11 and 12. Progesterone, when
given, was administrated im, 2 mg/injection on days 1–2, 5
mg/injection on days 3 and 4, and 10 mg/injection from day 5 until the
end of the treatment. Both estradiol and progesterone were
administrated in 1 ml 90% corn oil/10% ethyl alcohol (i.e.
vehicle). All ewes were killed from 4–6 h after the last injection on
day 12. The uteri were excised and each endometrium was separated from
the myometrium and frozen in liquid nitrogen.
PGF2 production by endometrial explants.
Explants (300 mg
wet wt) of endometrium were randomly distributed into three dishes
containing 10 ml Ham’s-F12 culture medium. The explants were rinsed
twice for 15 min with fresh medium and were incubated for 6 h at
38 C under 5% CO2 in humidified air. The amount of PGF2
produced by the explants was measured in the culture medium by RIA as
previously described (36). Briefly, an assay was conducted using
raioiodinated tracers of PGF2-histamine derivative. Concentrations
of PGF2 were determined in duplicate in unextracted medium samples,
using an anti-PGF2 antibody purchased from Pasteur-Productions
(Paris, France). To 5-ml tubes were added 0.1 ml iodinated tracer
(20,000 dpm), 0.1 ml PFG2 standard or culture medium sample, and 0.1
ml of a dilution of the antiserum such that the initial binding in the
absence of standard was 40% of the total radioactivity. Iodinated
tracer, PGF2 standards, and samples were diluted in 0.1
M phosphate-buffer, pH 7.3, containing 0.9% NaCl and 0.3%
bovine 
-globulin. Incubation was conducted overnight at 4 C. Free
and bound radioactivity were separated by adding successively 0.1 ml of
a dilution of antirabbit IgG and 1 ml polytethylene glycol 6000 (10%
wt/vol) containing 0.4% CaCl2, and centrifugating at +4 C for 15 min
at 3000 x g. The pellet was counted for 1 min in a
-counter. Calculation of the standard curve and unknown samples was
performed on computer using the logit method. The sensitivity of the
assay was defined as the quantity of PGF2 necessary to give 50%
displacement of B0 and was 12 pg/tube. The interassay coefficient of
variation was 14%, and intraassay coefficient of variation ranged from
8–12%. The PGF2 production by each endometrium was defined as the
mean production determined from three dishes and was expressed in
nanograms per milligram wet tissue.
Microsome preparation
Frozen tissues were weighed, thawed, and maintained at 4 C
throughout all the manipulations. The tissues (10–15 g wet wt) were
homogenized in 50 ml 50 mM Tris-HCl, pH 8.0, containing 250
mM saccharose, 10 mM EDTA, 1 mM
diethyldithiocarbamate, 1 mM benzamidine-HCl, and 10
µg/ml soybean trypsin inhibitor using a polytron homogenizer
(Kinematica, Switzerland). Homogenates were filtered through gauze and
centrifuged at 8000 x g for 20 min at 4 C. The
supernatant was centrifuged at 100,000 x g for 60 min
at 4 C. The pellet was resuspended in 300 µl 0.25% SDS solution. The
suspended microsomes were stored at -80 C until Western blot analysis.
The protein content of the microsome preparation was determined in
triplicate by the method of Peterson et al. (37) with BSA as
the standard.
Western blot analysis of cox-1 and -2
Microsome samples equivalent to 15 µg total protein were mixed
with Laemmli reagent (vol/vol) under reducing conditions (5%
2-mercaptoethanol) and heated for 10 min at 85 C. Proteins were
separated in a 10% SDS-PAGE using the miniprotean II system from
Bio-Rad (Ivry, France). Separation was carried out under a constant
current (15 mA/gel of 0.75 mm thickness). The Rainbow molecular weight
markers from Amersham (les Ulis, France) were used as molecular weight
standards. Proteins were transferred onto 0.1 µm nitrocellulose
membranes for 2 h at 250 mA, in a 25 mM Tris buffer,
pH 8.2, 192 mM glycine containing 0.1% SDS and 15%
methanol. Blots were stained with 0.5% Ponceau Red in 0.3%
trichloroacetic acid for a loading control for each sample. Blots were
performed in duplicate under the same conditions, one for cox-1
analysis and the other for cox-2 analysis.
Membranes were saturated overnight at 4 C with 5% fat-free dry milk in 50 mM Tris-HCl, pH 7.5, containing 250 mM NaCl and 0.1% Tween 20 (TBS-TM buffer). The membranes were then incubated with anticox-1 or anticox-2 antibodies diluted in the TBS-TM buffer. The antibodies used in this study were prepared as previously described (15, 16). The mouse anticox-2 monoclonal antibody (mAb294) was raised against the peptide C-NASSSRSGLDDINPTVLLK (cox-2 peptide), which is present in the carboxyl-terminus of human cox-2 and absent in the cox-1 protein. The rabbit anticox-1 polyclonal antibody (L855) was raised against cox-1 purified from ram seminal vesicles. A more detailed description of anticox-1 and anticox-2 antibodies has been reported by Créminon et al. (16). The mAb294 and L855 antibodies were used at a concentration of 3 µg/ml and 1:3000 dilution in TBS-TM (0.2 ml/cm2), respectively. The incubation was performed for 1 h at room temperature. Blots were washed twice for 10 min in the TBS-TM buffer and three times for 15 min in the TBS-TM buffer without milk (TBS-T buffer). Membranes were further incubated for 1 h at room temperature with anti-IgG antibodies conjugated with horseradish peroxidase at 1:3000 dilution in TBS-T buffer. Excess of the second antibody was eliminated by extensive washes in TBS-T buffer (5 consecutive 10-min washes). Chemiluminescent substrate was used according to the manufacturer’s instructions (ECL Kit, Amersham) and immunoreactive proteins were visualized after 1–5 min exposure to hyperfilm-ECL (Amersham). Immunoreaction signals were quantified on the films by scanning densitometry using an image analysis system (M-Lecphor, Biocom, les Ulis, France). To correct for variations in the intensity of the Western blot staining, one endometrium extract was used as internal standard and put on each gel.
Immunohistochemistry
Part of the uterine horn ipsilateral to the corpus luteum was
cross sectioned into 1-cm long pieces. Tissue pieces were immersed in
ice-cold isopentane, snap frozen in liquid nitrogen, and stored at -80
C. All endometrium pieces were sectioned in 6- to 8-µm slices using a
Frigocut cryostat (Reichert-Jung, Paris, France). Slices were collected
on microscope slides (Super-Frost/Plus, C.M.L., Nemours, France), dried
in air, and stored at -20 C. Frozen tissue slices were fixed for 15
min in a cold 4% paraformaldehyde solution and washed twice for 5 min
with 0.1 M PBS, pH 7.2, 0.1 M glycine for
rehydration. Slices were incubated for 1 h with 100 µl nonimmune
goat serum (1:10 dilution) in 0.1 M PBS containing 0.2%
BSA. Sections were then incubated for 1 h at room temperature in a
humidified atmosphere with the anticox-2 monoclonal antibody mAb294
(2.5 µg/100 µl) or anticox-1 polyclonal antibody L855 (1:20
dilution), which were diluted in 0.1 M PBS containing 2%
BSA. Control sections from each tissue were incubated with the primary
antibodies adsorbed with the respective antigen or were incubated with
nonimmune rabbit serum or with incubation buffer alone. After three
washes of 15 min with 0.1 M PBS containing 0.2% BSA, the
sections were incubated for 1 h at room temperature, in darkness,
and in a humidified atmosphere with anti-IgG antibodies conjugated to
fluorescein isothiocyanate and previously diluted at 1:200 in 0.1
M PBS containing 2% BSA. The slides were washed once for
15 min in 0.1 M PBS containing 0.2% BSA with Hoescht’s
dye (1:500 dilution) for chromatogen coloration. Slides were then
washed twice for 15 min in the same buffer without Hoescht’s dye.
Finally, they were rinsed in 0.1 M Tris-HCl buffer, pH 8.0,
and covered with Mowiol (Hoescht France, Paris, France). All sections
were examined for specific fluorescence using a Reichert microscope
equipped with an epifluorescence illuminating system. Microphotographs
were taken on Fuji film (Sensia, 400 ASA, Photoservice, Velizy,
France). The same exposure time was used for both control and
experimental sections.
Statistical analysis
ANOVA was used to tested effect of day, status, and steroid
treatments on relative amounts of immunoreactive cyclooxygenase in the
endometrium. Differences between individual means were assessed by the
Student-Newman-Keuls multiple range test. Correlations between the
PGF2 concentration and cox-2 or -1 contents were analyzed by the
Pearson product moment test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
revealed cox-1 in
the endometrium and in the myometrium of the ovine uterus. However, as
indicated by the intensity of the chemiluminescent signal, higher
concentrations of the enzyme were present in the myometrium than in the
endometrium. In contrast, the mAb 294 revealed cox-2 in the endometrium
but not in the myometrium (Fig. 2).
|
|
). This
occurred for both the cox-2 standard preparations and the uterine
extracts, confirming the identity of the protein. The additional
protein that was revealed at 60 kDa did not correspond to the binding
of the primary antibody. This band was displayed when the second
antibody was added to the nitrocellulose membrane without previous
incubation with primary antibody (data not shown).
The immunoadsorption of anticox-1 polyclonal antibody with PG synthase
from ram seminal vesicles led to the disappearance of the protein band
for standard cox-1 and ovine tissues (Fig. 2B). A weak signal did
remain, however. Higher concentrations of the PG synthase protein were
required to completely neutralize the polyclonal antibody L855. The
inconvenience of such a protocol is that it induced high background
levels, making it impossible to analyze the immunoblot.
As determined by denaturating electrophoresis, the molecular mass of cox-2 (70 kDa) from the ovine endometrium was consistent with previously published data. In some blots ovine endometrial cox-2 may be resolved as a doublet similar to that observed in the sheep placenta (29), in the bovine ovary (25), and in the rat preovulatory follicles (24), as well as in other mammal cells.
Cox-1 and -2 during the estrous cycle and early pregnancy
The cyclooxygenases were analyzed in ovine endometrium by Western
blots on days 6, 9, 12, 15, and 17 of the estrous cycle (Fig. 3) and on days 9, 12, 15, and 17 of pregnancy (Fig. 4). Endometrial extracts of five ewes were analyzed. The
blots from two are shown by day of collection.
|
|
In nonpregnant ewes, cox-2 was detected by day 12, remained
highly expressed through day 15, and became barely detectable on day
17. Similarly, in pregnant ewes, cox-2 was first detected on day 12 but
was still expressed on day 17. Thereafter the amount of cox-2 decreased
progressively up to day 25 of pregnancy (data not shown). The
quantification of protein bands by scanning densitometry allowed the
patterns of the cox-2 expression during the estrous cycle and early
pregnancy to be compared (Fig. 5). In both cases the
cox-2 concentration was maximal on day 12 and then decreased. There was
a statistically significant difference in the mean values among the
days of the estrous cycle or pregnancy (P < 0.001). No
overall difference was demonstrated between the two endocrine statuses
(P = 0.128). However, on day 17, the cox-2 content of
the pregnant endometrium was statistically different from that of the
nonpregnant tissue (P = 0.025).
|
, A-D. On day 12,
immunoreactivity was demonstrated in the entire caruncular
epithelium, whereas in the intercaruncular epithelium cox-2 was not
present (Fig. 6A) or only sparsely distributed. On day 15 both the
caruncular and intercaruncular epithelia were stained (Fig. 6B). The
uterine glands displayed different degrees of staining according to
their location within the endometrium (Fig. 6C). Only the glands close
to the uterine lumen showed detectable cox-2 levels. No
immunoreactivity was seen in the deeper glands (Fig. 6D). There was no
detectable staining in the stromal endometrium nor in the myometrium.
No positive immunoreactivity was observed in control sections (not
shown). No difference in immunostaining was observed between pregnant
and nonpregnant endometria. To ensure that the polyclonal antibody
against cox-1 did not cross-react with cox-2, the localization of cox-1
was performed on day 9 endometrium only, because cox-2 was not
expressed at this stage (Fig. 3). The localization of immunopositive
cells for cox-1 is shown in Fig. 6, E and F. Cox-1 was observed in both
the luminal and the glandular epithelium, and a weak staining of the
stromal cells underlying the epithelium was also seen (Fig. 6F).
Moreover, cox-1 was present in the deeper glands of the endometrium
(Fig. 6E). No differences were seen between pregnant and nonpregnant
ewes.
|
release by endometrium explants shows the PGF2-
production levels in the culture medium by endometrial explants in
relation to concentration of cyclooxygenases. There was no detectable
relationship between the cox-1 concentration and the amount of PGF2
released (R = 0.002; P = 0.995). However, the
PGF2 accumulation in the culture medium increased as the cox-2 level
increased in the endometrium (R = 0.793; P =
0.011). These data did not provide information concerning the
respective contribution of each of the two isoforms to PG production.
However, this result suggested that the ability of endometrium tissue
to produce different amounts of PGs was due to modifications in the
concentration of cox-2.
|
. Cox-1 was observed in all the
control and treated groups, although the concentration of the protein
differed between groups (P = 0.012). There was no
difference between the 10-day progesterone treatment (group P10) and
the three control groups (C, C-P, C-E). In contrast, administration of
progesterone for 12 days (group P12) resulted in a significant increase
in cox-1 level (P < 0.001). Estradiol never caused any
modifications in the cox-1 concentration. Cox-2 was not detected in the
endometrium from control ovariectomized ewes (group C) nor in the
steroid pretreatment groups (groups C-P, C-E). However cox-2 was
strongly induced in all the steroid treatment groups. There was a
significant increase of the enzyme by the 12-day treatment (group P12)
vs. the 10-day treatment (group P10) (P =
0.001). Administration of estradiol for 2 days after the 10-day
progesterone (group P10-E) treatment did not alter the cox-2
concentration. When estradiol was given in combination with
progesterone on days 11 and 12 (group P12-E), an additional increase in
cox-2 seemed to result, but the increase was not statistically
significant as indicated by examination of the densitometric data.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The observation that cox-2 showed marked cyclical changes throughout the estrous cycle led us to examine the role of steroids on the expression of the two cyclooxygenases. In this study we have shown that endometrial cox-2 was primarily controlled by ovarian steroids because cox-2 was not expressed in the absence of steroids. Progesterone alone elicited high levels of endometrial cox-2, whereas endometria from ovariectomized ewes did not exhibit any detectable immunoreactive protein. We did not observe any significant effect of estrogen on cox-2 concentration. It could be assumed that in intact ewes, the exposure of the uterus to increasing concentrations of progesterone over 10 days is primarily responsible for the expression of cox-2. Other evidence for the role of progesterone in stimulating the expression of endometrial PG synthase comes from the in situ hybridization study of Eggleston et al. (41). These authors reported an 8-fold increase in the level of hybridization signal of ewe uterine tissue sections following progesterone administration. They used a complementary RNA probe directed against the mRNA corresponding to the 438- to 455-amino acid region of mature ovine cox-1, which is 33% identical to the equivalent region of human cox-2. Such a riboprobe could hybridize with both mRNA for cox-1 and -2. Consequently it could be assumed that the increase in PG synthase signal observed by these authors after progesterone treatment was due more to an increase in cox-2 mRNA than to cox-1 mRNA.
If the induction of cox-2 expression seemed clearly dependent on progesterone, the regulation of the end of cox-2 expression was not clarified. The fact that on day 17 of pregnancy the expression of cox-2 was maintained, whereas it stopped in nonpregnant ewes, suggests that the drop in progesterone and the increase in estrogen that occurs at the end of the estrous cycle arrests cox-2 expression. However our results did not support this hypothesis. Actually, cox-2 was neither reduced when the progesterone treatment was stopped nor during the estradiol treatment that followed. In both cases the protein levels remained high. It is likely that other factors, which remained to be determined, are responsible for the cessation of cox-2 expression. In other species, recent reports have examined the relationship between the expression of uterine cox-2 and steroids. In the rat endometrium, Shoda et al. (42) observed that the induction of cox-2 coincided with the peak in serum estradiol levels, suggesting the estradiol-dependent induction of cox-2. However, in the mouse, Chakraborty et al. (27) failed to demonstrate an effect of steroid treatment on the induction of uterine cox-2.
Unlike cox-2, cox-1 was expressed in the absence of steroid treatment. Our results were consistent with those of Salamonsen et al. (10) who showed constitutive expression of PG synthase mRNA in both ovariectomized ewes and steroid treated animals. Our data indicated that the cox-1 levels were increased to a limited but significant extent by progesterone. However we did not observed any suppressive effect of estrogen on cox-1 expression as reported by Salamonsen et al. (10).
Apart from their different modes of expression, the two isoforms also differed in their cellular localization. We showed that the cox-2 enzyme was strictly localized in the luminal cells and in the outer glands of the endometrium. However, it was primarily expressed in the caruncular epithelium, then in the epithelium of the intercaruncular area. As the estrous cycle advanced, the outer glands became increasingly stained but the deeper glands were always negative. In contrast, we did not observe any changes in the staining of cox-1, which was localized in both epithelial and stromal cells as well as in outer and deeper glands. Our results for the cox-2 staining agreed with those of Salamonsen et al. (43) who reported an equivalent continuum of change in PG synthase staining. It could be proposed that the mAb [cyo-1 from Dewitt et al. (44)] that these authors used recognized both cox-1 and -2.
The presence of cox-2 in uterine tissues has been previously reported in rodents (26, 27, 42) and human decidual cells (45). In the mouse, expression of cox-2 is related to the implantation of the blastocyst, because it is exclusively observed in the stromal and epithelial cells adjacent to the site of embryo attachment (26, 27). In mice, the blastocyst appears to be directly responsible for the induction of cox-2 in the uterus. In sheep, our results indicated that the expression of cox-2 differed from that of the mouse. The induction of the enzyme did not require the presence of the blastocyst because cox-2 expression occurred in the uteri of cyclic ewes. Moreover, we did not observe cox-2 in the stromal compartment of the ovine endometrium. This difference in endometrial localization of cox-2 could be related to the mode of implantation of the trophoblast to the uterus. In rodents, the invasive trophoblast comes in close contact with the uterine stroma, whereas in ruminants, the trophoblast does not penetrate the stroma and is apposed to the maternal epithelium, which expresses cox-2. Thus endometrial cox-2 could be involved in the important changes in cellular adhesiveness at the time of implantation (reviewed in Ref.46). Indeed, it has been recently demonstrated that strong expression of cox-2 increases adhesion to extracellular matrix proteins (47).
We have shown that the release of PGF2 by endometrial explants is
correlated with their cox-2 contents, suggesting that the increase in
the ability of the uterus to produce PGs during the luteal phase of the
estrous cycle is due to the increase in cox-2 levels. The ability of
the enzyme to convert arachidonic acid seems to be unaffected by
pregnancy because PGF2 was released in the culture medium at the
same rate from pregnant and nonpregnant endometrial explants. These
findings are consistent with the current agreement that pregnant and
nonpregnant endometria have similar abilities to synthesize PGs (48, 2). It has been well demonstrated that in vivo the patterns
of secretion differ but there is no difference in the mean
concentrations of PGF2 (2). It is of interest to note that cox-2 was
expressed in high concentrations close to the time when the luteolysis
process occurs. It could be suggested that cox-2 function is to induce
the high burst of PGF2 required for the completion of luteolysis.
However, the persistence of high concentrations of cox-2 in early
pregnancy brings into question the physiological significance of the
prostanoids formed. It could be suggested that the generation of a
PG-generating system is not only relevant to the luteolysis process but
argues in favor of a role for PGs in the establishment of
pregnancy.
In conclusion, to appreciate the biological relevance of cox-2 expression it should be noted that, in the ewe, there are two physiological situations in which the synthesis of PGs needs to be strictly regulated, i.e. luteolysis and parturition, and that these two require require the induction of cox-2. In addition, because the endometrium and placenta are two tissues particularly sensitive to steroids, the ewe appears to be a suitable model for the hormonal regulation of cox-2.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received November 21, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and the timing of events
associated with luteolysis in ewes with oestrous cycles of different
length. J Reprod Fertil 83:517–526 during luteolysis in ruminants. Biol Reprod 45:655–663[Abstract]
-interferons. FEBS Lett 228:12–16[CrossRef][Medline]
.
J Biol Chem 267:10816–10822 by endometrial tissue from
ovariectomized ewes receiving progesterone and estradiol. Biol Reprod 44:404–412[Abstract]
in ewes after
withdrawal of exogenous progesterone. J Reprod Fertil 95:885–893 and 6-keto-prostaglandin F1
in the utero-ovarian venous plasma of non-pregnant and early pregnant
ewes. Biol Reprod 30:936–944[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  F. Nuttinck, B. M.-L. Guienne, L. Clement, P. Reinaud, G. Charpigny, and B. Grimard Expression of genes involved in prostaglandin E2 and progesterone production in bovine cumulus-oocyte complexes during in vitro maturation and fertilization Reproduction, May 1, 2008; 135(5): 593 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. E Spencer, O. Sandra, and E. Wolf Genes involved in conceptus-endometrial interactions in ruminants: insights from reductionism and thoughts on holistic approaches Reproduction, February 1, 2008; 135(2): 165 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. G Kennedy, C. Gillio-Meina, and S. H. Phang Prostaglandins and the initiation of blastocyst implantation and decidualization Reproduction, November 1, 2007; 134(5): 635 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. G Lea and O. Sandra Immunoendocrine aspects of endometrial function and implantation Reproduction, September 1, 2007; 134(3): 389 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Zhang, V. Collins, K. Chakrabarty, J. C Rose, and W. X. Wu Regulation of the prostaglandin enzymatic system by estradiol and progesterone in nonpregnant sheep cervix Reproduction, May 1, 2007; 133(5): 1027 - 1034. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. R. Bilby, A. Guzeloglu, L. A. MacLaren, C. R. Staples, and W. W. Thatcher Pregnancy, Bovine Somatotropin, and Dietary n-3 Fatty Acids in Lactating Dairy Cows: II. Endometrial Gene Expression Related to Maintenance of Pregnancy. J Dairy Sci, September 1, 2006; 89(9): 3375 - 3385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L Cammas, P Reinaud, N Bordas, O Dubois, G Germain, and G Charpigny Developmental regulation of prostacyclin synthase and prostacyclin receptors in the ovine uterus and conceptus during the peri-implantation period. Reproduction, May 1, 2006; 131(5): 917 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Chen, J. A. Green, E. Antoniou, A. D. Ealy, N. Mathialagan, A. M. Walker, M. P. Avalle, C. S. Rosenfeld, L. B. Hearne, and R. M. Roberts Effect of Interferon-{tau} Administration on Endometrium of Nonpregnant Ewes: A Comparison with Pregnant Ewes Endocrinology, May 1, 2006; 147(5): 2127 - 2137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G. W. Fleming, T. E. Spencer, S. H. Safe, and F. W. Bazer Estrogen Regulates Transcription of the Ovine Oxytocin Receptor Gene through GC-Rich SP1 Promoter Elements Endocrinology, February 1, 2006; 147(2): 899 - 911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Gray, C. A. Abbey, P. D. Beremand, Y. Choi, J. L. Farmer, D. L. Adelson, T. L. Thomas, F. W. Bazer, and T. E. Spencer Identification of Endometrial Genes Regulated by Early Pregnancy, Progesterone, and Interferon Tau in the Ovine Uterus Biol Reprod, February 1, 2006; 74(2): 383 - 394. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. X. Wu, T. Coksaygan, K. Chakrabarty, V. Collins, J. C. Rose, and P. W. Nathanielsz Sufficient Progesterone-Priming Prior to Estradiol Stimulation Is Required for Optimal Induction of the Cervical Prostaglandin System in Pregnant Sheep at 0.7 Gestations Biol Reprod, August 1, 2005; 73(2): 343 - 350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Y Lee and F. J DeMayo Animal models of implantation Reproduction, December 1, 2004; 128(6): 679 - 695. [Abstract] [Full Text] [PDF]
|
|
