| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Basic Science Research, Department of Obstetrics and Gynecology, University of Louisville Health Sciences Center, Louisville, Kentucky 40292
Address all correspondence and requests for reprints to: Dr. Ch. V. Rao, Department of Obstetrics and Gynecology, 438 MDR Building, University of Louisville Health Sciences Center, Louisville, Kentucky 40292. E-mail: cvrao001{at}gwise.louisville.edu Web site:
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- or ß-subunits
of hCG had no effect on COX-2 protein levels. Studies on signaling revealed that treatment of cells with hCG resulted in an increase in cAMP levels and protein kinase A (PKA) activity. Inhibition of PKA activity by cotreatment with isoquinolinesulfonamide (H-89) prevented hCG from increasing COX-2 protein levels. Treatment with 8-bromo-cAMP mimicked the effect of hCG, and cotreatment with a selective inhibitor of type I PKA, 8-chloro-cAMP, prevented 8-bromo-cAMP and hCG from increasing COX-2 protein levels.
The requirement of receptors for LH/hCG action was investigated by 24-h treatment of human endometrial gland epithelial cells with 21-mer phosphorothioate oligodeoxynucleotides (ODNs) synthesized from human receptor sequence. Treatment with 2 µmol/L antisense, but not sense, ODN resulted in a dramatic reduction in LH/hCG receptor protein levels. hCG was unable to increase COX-2 protein, PGE2, and cAMP levels in an antisense, but not in sense, ODN-treated cells.
In summary, we conclude that hCG and LH treatment can increase expression of the COX-2 gene in human endometrial gland epithelial cells. The effect was time and dose dependent, hormone specific, and mediated by the cAMP/type I protein kinase A signaling pathway. The hCG actions require a normal complement of its receptors in cells. These hCG and LH effects may be another action of these hormones in human endometrium that is important for implantation of the blastocyst and continuation of pregnancy.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following items were obtained as gifts: polyclonal
anti-LH/hCG receptor antibody raised against a synthetic N-terminus
amino acid sequence of 15–38 from Dr. Patrick Roche at the Mayo Clinic
(Rochester, MN); LH/hCG receptor complementary DNA (cDNA) from Dr.
Hugues Loosfelt at Hospital de Bicetre (Paris, France); cDNA of COX-2
from Dr. Harvey Herschman at the University of California at Los
Angeles; cDNA of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) from
Dr. Russell Prough of our institution; highly purified hCG (CR-127;
14,900 IU/mg), human LH (AFP-0264B; 4,015 IU/mg), human FSH
(AFP-87929B; 1,683 IU/mg), human TSH (AFP-4314C; 15 IU/mg), -
(CR-125) and ß- (CR-125; 29 IU/mg) subunits of hCG from the National
Hormone and Pituitary Program supported by the NIDDK, NICHHD, and USDA
(Rockville, MD); and 8-chloro (cl)-cAMP from Drug Synthesis and
Chemistry Branch, NCI (Bethesda, MD). The following items were
purchased: polyclonal anti-COX-2 antibody from Oxford Biochemical
Research (Oxford, MI); enzyme immunoassay kits for PGE2 and
cAMP from Cayman Chemical Co. (Ann Arbor, MI); horseradish
peroxidase-labeled antirabbit IgG, antibiotic-antimycotic solution,
FBS, trypsin, and insulin from Life Technologies, Inc.
(Grand Island, NY); in vitro transcription and random prime
labeling kits, pre-Tag nonradioactive protein kinase A (PKA) and
protein kinase C (PKC) activity measurement kits from Promega Corp. (Madison, WI); Nonidet P-40, Percoll, 10 x HBSS,
DMEM-Ham’s F-12 nutrient mixture (DMEM/F-12 medium),
phenylmethylsulfonylfluoride (PMSF), aprotinin, leupeptin, collagenase
(type IA-S; 320 U/mg solid), monoclonal anti-ß-tubulin antibody, and
8-bromo-cAMP (8-Br-cAMP) from Sigma Chemical Co. (St.
Louis, MO); sodium azide from American Bioanalytical Co. (Natick, MA);
enhanced chemiluminescence Western blot detection kit from
Amersham Pharmacia Biotech (Arlington Heights, IL);
BS3 [bis-(sulfosuccimidy)suberate] from Pierce Chemical Co. (Rockford, IL); [-32P]CTP and
[-32P]deoxy-CTP (800 Ci/mmol) from DuPont-New England
Nuclear Corp. (Boston, MA); the monoclonal antivimentin,
anticytokeratin, anti- smooth muscle actin and anti-CD68 antibodies
from DAKO Corp. (Carpenteria, CA); bisindolylmaleimide
(Bis) and isoquinolinesulfonamide (H-89) from Calbiochem
(San Diego. CA); and peroxidase substrate and avidin-biotin
immunoperoxidase kits from Vector Laboratories, Inc.
(Burlingame, CA). The 21-mer phosphorothioate antisense
(5'-GCCGAGAACCGCTGCTTCATG-3') and sense (5'-CATGAAGCAGCGGTTCTCGGC-3')
oligodeoxynucleotides (ODNs) were synthesized from human LH/hCG
receptor cDNA sequence beginning at the ATG translation initiation
codon.
Collection of tissues
Endometrial specimens were obtained from 22- to 45-yr-old premenopausal women undergoing hysterectomy or diagnostic dilatation and curettage for a variety of medical indications other than endometrial cancer. Only those found to be histologically normal and dated to be in the proliferative or secretory phase were used in the studies. The use of these tissues was approved by our institutional human studies committee. The tissues were immediately brought to the laboratory on ice in sterile tubes containing DMEM/F-12 medium, pH 7.4, with 10% FBS and 1% antibiotic-antimycotic mixture (10,000 U penicillin G sodium, 10 mg/mL streptomycin sulfate, and 25 µg/mL amphotericin and fungizone in 0.85% saline).
Isolation and culture of cells
Endometrial gland epithelial cells were isolated and cultured
with few modifications from previously described procedures (25, 26).
Briefly, endometrial tissues were cut into 1-mm3 pieces and
digested with 0.25% collagenase in DMEM/F-12 medium in capped sterile
tubes in a shaking water bath for 30 min at 37 C. The digest was then
passed first through a 250-µm pore size nylon mesh and then through a
35-µm pore size nylon mesh. The glandular epithelial cells retained
on the mesh were backwashed with 20 mL 1 x HBSS containing 1%
antibiotic-antimycotic mixture. The backwash solution containing
enriched glandular epithelial cells was centrifuged for 5 min at
200 x g. The cell pellet was resuspended in DMEM/F-12
medium containing 10% FBS and incubated for 1 h at 37 C in an
incubator containing a humidified atmosphere of 5%
CO2-95% air. Unattached cells were discarded, and the
purity of the attached cells was determined by immunocytochemistry
using anticytokeratin (epithelial), antivimentin (stromal), anti-
smooth muscle actin, and anti-CD68 (macrophages) antibodies. Cell
viability, which was greater than 95%, was determined by trypan blue
exclusion. After reaching subconfluence in about 4–6 days, the medium
was replaced with fresh DMEM/F-12 containing no FBS and the indicated
concentrations of hCG or other hormones. The number of cells used per
culture well or flask varied among the experiments. However, the same
number of cells was used within the same experiments.
Northern blotting
For this procedure, total ribonucleic acid (RNA) was isolated from endometrial gland epithelial cells, and then 30-µg aliquots were heat denatured at 100 C for 5 min, resolved on formaldehyde-agarose gels, and blotted onto Gene Screen Plus membranes (27). The RNA was cross-linked to the membranes by irradiation for 2.5 min under UV light and baking for 10 min at 70 C. Nonspecific binding was blocked by salmon sperm DNA, and the blots were sequentially hybridized overnight at 65 C with either 1–5 x 107 cpm/mL 32P-labeled LH/hCG receptor or 0.5–1 x 107 cpm/mL 32P-labeled COX-2 riboprobes transcribed from corresponding cDNAs by an in vitro transcription kit. Then the membranes were washed at 65 C twice with 2 x SSC (1 x SSC = 150 mmol/L sodium chloride and 15 mmol/L sodium citrate, pH 7.4) containing 0.1% SDS and, again, twice with 0.1 x SSC-0.1% SDS. The washed membranes were exposed for 48–72 h at -70 C to Kodak XAR-5 film (Eastman Kodak Co., Rochester, NY) with intensifying screens. The membranes were stripped and rehybridized at 42 C with 2–5 x 106 cpm/mL 32P-labeled GAPDH cDNA prepared by using a random priming kit. The membranes were washed as described above at 42 C. The molecular size of the transcripts was determined by running a RNA ladder in an adjacent lane.
Western blotting
For this procedure, endometrial gland epithelial cells were homogenized in buffer containing 200 mmol/L PMSF and 20 mmol/L leupeptin to inhibit the activity of endogenous proteases. Then, 30-µg aliquots of protein were separated by discontinuous 10% SDS-PAGE under reducing conditions (28). The separated proteins were electroblotted onto Immobilon-P (Millipore Corp., Bedford, MA) membranes (29). After blocking nonspecific binding sites with 5% nonfat dry milk in 5 mmol/L Tris-HCl (pH 7.4), 136 mmol/L NaCl, and 0.1% Tween-20 (TBST buffer), the blots were incubated with a 1:1500 dilution of anti-LH/hCG receptor antibody or a 1:1000 dilution of anti-COX-2 antibody for 2 h at 22 C, washed twice for 10 min each time with TBST buffer, and again washed once with TBST containing 40 mmol/L sodium azide. The washed blots were reincubated for 1 h at 22 C with a 1:2000 dilution of horseradish peroxidase-labeled antirabbit IgG and washed as described above. Then, the binding of anti-LH/hCG receptor and anti-COX-2 antibodies was detected by an enhanced chemiluminescence Western blotting detection kit. To correct the differences in protein loading, after the detection of COX-2 protein, membranes were washed with TBST buffer containing sodium azide for 30 min, then incubated with a 1:2000 dilution of anti-ß-tubulin antibody and washed as before. The molecular sizes of the LH/hCG receptor and COX-2 proteins were determined by running molecular size marker proteins in an adjacent lane. In the procedural control, receptor antibody was preabsorbed with excess LH/hCG receptor peptide, and the primary antibody was omitted in case of COX-2.
Covalent receptor cross-linking
For this procedure, 100-µg aliquots of homogenate protein were incubated for 30 min at 37 C with 1–2 x 106 cpm/mL [125I]hCG in the presence or absence of 5 µg unlabeled hCG (30). The hCG was radioiodinated by a lactoperoxidase technique to a specific activity of 76.1 µCi/µg (31). The receptor-bound [125I]hCG was cross-linked by incubating for 1 h at room temperature with 100 mmol/L BS3 dissolved in 5 mmol/L sodium citrate buffer, pH 5.0. Then the reaction was stopped by the addition of 50 mmol/L Tris-HCl, pH 7.5, and centrifugation for 30 min at 27,000 x g at 4 C. Pellets containing [125I]hCG-receptor complexes were solublized with 1% Triton X-100; diluted with 125 mmol/L Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, and 10% mercaptoethanol; and separated on 8% SDS-PAGE under nonreducing conditions. The gel was fixed, dried, and exposed for 1–3 days to Kodak XAR-5 film with intensify screens at -70 C. Molecular sizes of the [125I]hCG receptor complexes and unbound [125I]hCG were determined by running mol wt marker proteins in an adjacent lane.
Measurement of medium PGE2 levels
PGE2 levels in duplicate 50-µL medium samples were quantified using a commercial kit. Instructions provided in the kits were followed. The specificity of the PGE2 antibody was 100% for PGE2, 43% for PGE3, 18.7% for PGE1, and 1% or less for all other eicosanoids tested. Intra- and interassay coefficients of variation were less than 10%. The detection limit of the assay was 29 pg/mL PGE2.
Measurement of medium cAMP levels
Levels of cAMP in 50-µL aliquots of medium samples were quantified using a commercial kit. Instructions provided in the kit were followed. The specificity of the cAMP antibody was 100% for acetylated cAMP, 0.3% for cAMP, 0.05% for acetylated cGMP, and 0.01% or less for cGMP, acetylated adenosine, cytidine, guanoisine, and uridine. Intra- and interassay coefficients of variation were less than 10%. The detection limit of the assay was 1.1 pmol/mL.
Immunocytochemistry
The cells were fixed in Bouin’s solution for 30 min and then
immunostained by an avidin-biotin immunoperoxiadase method using a
1:500 dilution of anti-LH/hCG receptor, anticytokeratin, antivimentin,
anti- smooth muscle actin, and anti-CD68 antibodies (1). For the
procedural controls, the primary antibodies were replaced with
nonspecific IgG.
Measurement of PKA and PKC activities
Cells were lysed by sonication and freezing/thawing in 200 µL 25 mmol/L Tris-HCl buffer, pH 7.5, containing 1 mmol/L ethylenediamine tetraacetate, 1 mmol/L dithiothreitol, 20 mmol/L NaCl, 0.5 mmol/L PMSF, 1 µmol/L aprotinin, and 50 µmol/L leupeptin. PK activities were determined by incubating 3–5 µg lysate protein for 30 min at 30 C with fluorescent-labeled A1 peptide for the PKA assay and fluorescent-labeled C1 peptide for the PKC assay. Nonphosphorylated and phosphorylated fluorescent peptides were separated on 0.8% agarose gels. Phosphorylated fluorescent peptide bands were excised and eluted, and the optical density at 570 nm was measured using a 96-well plate reader. PK activities were calculated from the densitometric values using instructions provided by the kit manufacturer. Positive and negative controls supplied in the kits were assayed at the same time with endometrial cell samples.
Culturing cells with antisense and sense ODNs
Cells were cultured in six-well plates (5 x 105 cells/well) for 24 h in serum- and phenol red-free medium containing 2 µmol/L antisense or sense ODNs. After 24 h, immunocytochemistry for LH/hCG receptors was performed on some cells, whereas others were cultured for 4 h in the presence or absence of 100 ng/mL hCG. Media were then removed for the measurement of PGE2 and cAMP levels, and cells were recovered for the measurement of COX-2 protein by Western blotting.
Densitometry
Optical densities of autoradiographic bands of COX-2 and GAPDH messenger RNAs (mRNAs) and COX-2 and ß-tubulin protein bands were measured in a linear range using a Z-gel scanning system (Zaxis, Hudson, OH).
Replication of experiments and statistical analyses
Each experiment was duplicated and repeated at least three times on cells from different endometrial specimens. Data from all of the experiments were pooled for calculation of the means and SE and for one-way ANOVA and Duncan’s multiple range test (32).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Purity was determined by immunocytochemistry using four different
antibodies, three of which are directed against common cell
contaminants. As shown in Fig. 1
, virtually all cells were immunostained with anticytokeratin antibody,
which binds to epithelial cells (Fig. 1A). Antibodies that recognize
fibroblasts (vimentin; Fig. 1B), smooth muscle cells (Fig. 1C), and
macrophages (CD68; Fig. 1D) did not immunostain the cells. The
procedural control showed no immunostaining (Fig 1E).
|
Even though previous studies showed that glands in intact tissue
contained higher LH/hCG receptor levels than the other uterine cells
(1, 2, 3, 7), we needed to reestablish that endometrial gland epithelial
cells in culture retained these receptors. Northern analysis
demonstrated that cells contained a major 4.3-kb and minor 3.6-, 2.4-,
and 1.8-kb transcripts of LH/hCG receptors (Fig. 2, lane 1). Western blotting showed that
cells contained 80-kDa protein (lane 2), which was not detected when
the receptor antibody was preabsorbed with the excess receptor peptide
(lane 3). Covalent receptor cross-linking revealed that
[125I]hCG forms a complex with a molecular size of 125
kDa (lane 4). Formation of this complex was inhibited when excess
unlabeled hCG (lane 5) and LH (lane 6), but not when TSH (lane 7), FSH
(lane 8), and - (lane 9) and ß- (lane 10) subunits of hCG were
present in the incubation medium. The 45-kDa band represents unbound
[125I]hCG, and the 80-kDa difference between 125 and 45
kDa represents the molecular size of free receptor protein.
|
Endometrial gland epithelial cells contained an expected 4.4-kb
transcript of COX-2 (Fig. 3). Transcript
levels significantly increased in a time- and dose-dependent manner
after treatment with highly purified hCG compared with control values.
The hCG effect was specific because mRNA levels of a housekeeping gene,
GAPDH, were unchanged by treatment.
|
shows that endometrial gland
epithelial cells contained 72-kDa COX-2 protein, and its levels also
significantly increased in a time- and dose-dependent manner after
treatment with hCG compared with control values. The increase in COX-2
protein was parallel to, but not identical with, an increase in COX-2
mRNA.
|
- and ß-subunits (Fig. 5).
These data suggest that the hCG effect is hormone specific and requires
the conformation of native hormone.
|
shows that hCG treatment
resulted in a time- and dose-dependent increase in PGE2
levels in medium. The increase in PGE2 levels paralleled
the increase in COX-2 protein.
|
As cAMP is a common signaling molecule involved in hCG and LH
actions, we first investigated the hCG effect on its levels in medium
of endometrial gland epithelial cells. Figure 7 shows that hCG treatment had a
significant time- and dose-dependent stimulatory effect on cAMP levels.
The first increase in cAMP levels preceded an increase in COX-2 mRNA
and protein levels.
|
shows that hCG treatment did indeed
significantly increase PKA activity compared with the control value.
This effect was specific, because hCG treatment had no effect on PKC
activity. These data indicate that PKA, but not PKC, activation is
probably required for the hCG effect. If this is the case, then
inhibition of PKA, but not PKC, activity should prevent hCG from
increasing COX-2 protein levels. When this prediction was tested, we
found that cotreatment with the PKA inhibitor, H-89, but not
cotreatment with the PKC inhibitor, Bis, prevented an increase in COX-2
protein due to hCG treatment (Fig. 9).
The inhibitors alone had no effect, suggesting that these signaling
pathways do not control basal COX-2 protein levels.
|
|
demonstrate that
treatment with 0.1 mmol/L 8-Br-cAMP increased COX-2 protein levels just
as did treatment with hCG.
|
show that it prevented not only
8-Br-cAMP, but also hCG, from increasing COX-2 protein levels. 8-cl-AMP
alone had no effect, suggesting that type I PKA activity does not
control basal levels of COX-2 protein. Requirement of LH/hCG receptors in hCG action
We used antisense ODN approach to determine whether receptors are
required for hCG actions. Antisense and sense ODNs were synthesized
from human LH/hCG receptor sequence. To determine their effectiveness
in inhibiting LH/hCG receptor synthesis, human endometrial gland
epithelial cells were treated for 24 h with 2 µmol/L ODNs, and
then immunocytochemistry was performed. The results presented in Fig. 11 demonstrate that untreated cells
were immunostained for LH/hCG receptors, and this immunostaining was
absent in procedural controls, indicating that it was specific.
Treatment with antisense, but not sense, ODN resulted in a dramatic
reduction in LH/hCG receptor immunostaining.
|
demonstrate that hCG could not increase COX-2 protein,
PGE2, or cAMP (Fig. 12) levels in antisense ODN-treated
cells. Sense ODN-treated cells, which contained a normal complement of
receptors, responded to hCG treatment in the same way as untreated
cells by increases in COX-2 protein, PGE2, and cAMP levels
(Fig. 12). Thus, these data clearly indicate that hCG cannot work in
the absence of a normal complement of its receptors in cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We used primary cultures of endometrial gland epithelial cells to
determine the functional importance of LH/hCG receptors. These cells,
like glands in intact uterus, contain LH/hCG receptors. Northern
analysis detected multiple transcripts, which is a hallmark feature of
the LH/hCG receptor gene transcription in other nongonadal target
tissues just as in gonadal tissues (34). Western blotting detected an
80-kDa receptor protein. Covalent receptor cross-linking revealed that
this protein can bind hCG and LH, but not other hormones in the same
glycoprotein hormone family or the isolated - and ß-subunits of
hCG.
For several reasons, we chose COX-2 to determine the functional importance of LH/hCG receptors in endometrial gland epithelial cells. First and foremost was that LH/hCG can increase the expression of COX-2 in human endometrial stromal cells and fallopian tube epithelial cells (19, 22), which makes it likely that these hormones might do the same in endometrial gland epithelial cells. Second, human endometrial glands contain the highest levels of COX, although whether it is COX-1 and/or COX-2 has not been previously determined (35). Third, eicosanoids that are generated as a result of COX-2 activation regulate a number of endometrial functions that are related to implantation (36).
The results of the present study demonstrated that endometrial gland epithelial cells contain the expected size COX-2 mRNA transcripts and protein, and hCG treatment can increase their levels in a time- and dose-dependent manner. The hCG effect was hormone specific and requires the conformation of native hormone. COX-2 protein increased by hCG treatment was catalytically active, as determined by the response of PGE2 secretion. Although PGE2 was used as a measure of an enzyme activity, the catalytic activity of COX-2 can lead to the formation of other eicosanoids in this pathway (37). The measurement of all of these products was beyond the scope of our study. Moreover, PGE2 among these products has greater relevance to implantation.
The signaling involved in hCG action was first investigated by examining changes in cAMP levels in endometrial gland epithelial cells. The results demonstrated that hCG treatment can increase its levels in a time- and dose-dependent manner. As the increase in cAMP levels suggested that PKA might have been activated, we then measured PKA activity and found that hCG treatment can increase its activity, but not the activity of PKC. If PKA activation is necessary for hCG to increase COX-2, then inhibition of PKA, but not PKC, should prevent hCG from increasing COX-2 protein. The use of H-89 and Bis, PKA and PKC inhibitors, respectively, validated this expectation. If cAMP was a mediator of hCG action, then a cAMP analog, such as 8-Br-cAMP, should mimic hCG action, which it did.
Mammalian cells contain type I and type II forms of PKA (33). Studies with 8-cl-cAMP, a selective inhibitor of type I enzyme (33), revealed that it is the one that is involved in 8-Br cAMP and hCG actions. The extent of the type II enzyme’s contribution could not be determined, because there is no inhibitor of which we are aware for this enzyme.
As endometrial gland epithelial cells are relatively new targets of hCG/LH action, we investigated the requirement of receptors in hCG actions. If receptors are absolutely required, then inhibition of their synthesis should result in a loss of hCG function. We used antisense ODN technology to inhibit receptor synthesis. Immunocytochemical analysis revealed that treatment of endometrial gland epithelial cells with the 21-mer antisense, but not sense, phosphorothioate ODN synthesized from human LH/hCG receptor sequence virtually abolished LH/hCG receptor levels. hCG was not able to increase COX-2 protein, PGE2 secretion, and cAMP production in antisense, but not in sense, ODN-treated cells. Thus, these data clearly demonstrate that hCG cannot work in the absence of a normal complement of its receptors in endometrial gland epithelial cells.
Although the present study cannot vouch for LH/hCG controlling endometrial gland functions in vivo, a recent study demonstrated that hCG can (38). This study has shown that hCG administration in premature ovarian failure patients resulted in morphological changes in endometrial glands that generally reflect functional changes.
Targeted disruption of COX-2, but not COX-1, results in several reproductive defects, one of which is the failure of implantation (21). This finding suggests that COX-2 plays a central role in the implantation process. There must be hormones that up-regulate the expression of COX-2; however, there are no data on what these hormones might be in human endometrial glands. It has long been believed that there is a two-way dialogue between implanting blastocyst and receptive endometrium that is essential for successful implantation. It has been suspected that this dialogue may involve hCG produced by the blastocyst acting on the endometrium (39, 40, 41). This possibility has remained as only a concept, because until recently no one has demonstrated that endometrial glands and stroma contain LH/hCG receptors, and that LH/hCG can increase the expression of COX-2 and, consequently, the much needed eicosanoids that are responsible for all of the other changes that make the endometrium receptive for implantation. This leads to the question of whether there might be other sources of LH/hCG to act on endometrial glands. It is possible that periovulatory blood LH levels may reach concentrations high enough at the endometrial level to begin changes in glands. Endometrial glands produce hCG during secretory phase (42), suggesting that the changes initiated by LH could be continued by hCG derived from glands as well as blastocyst. The present findings also suggest that treatment of woman with hCG to induce ovulation could improve the uterine environment for successful implantation. This benefit could be just as important as the induction of ovulation itself.
It is likely that LH/hCG may also directly and indirectly regulate other endometrial gland functions related to the implantation process. These are, for example, the synthesis and/or secretion of glycodelin, insulin-like growth factor binding protein-1 (43), GnRH (44), and vascular endothelial growth factor (45); the secretion of fluids and electrolytes (46, 47); etc. There is evidence for PGE2 and cAMP regulating some of these functions (46, 47). By increasing PGE2 and cAMP levels, hCG may also regulate these and other endometrial functions.
The present findings may have further implications. For example, coculture of embryos with autologous endometrial epithelial cells has recently been reported to increase the number of blastomeres and decrease the fragmentation rate in preembryos (48). More importantly, coculturing has been found to improve the implantation rate (49), especially in patients who failed to become pregnant in more than three previous cycles in which three or four good quality embryos were transferred (50). The fact that LH/hCG can act on endometrial epithelial cells suggests that hCG treatment may further improve the benefits of coculturing. hCG treatment of women with signs of threatened abortion during early gestation seems to result in less frequent spontaneous abortions, premature deliveries, and intrauterine growth retardation (13), suggesting that hCG improves pregnancy outcome through multiple sites of action, including implantation.
The present findings may also be important for better understanding of endometrial hyperplasias and carcinomas. For example, these tissues overexpress LH/hCG receptors (7, 51), LH and hCG are mitogenic in human endometrial cancer cells (52, 53, 54), and bioactive LH levels are elevated in postmenopausal obese women with endometrial carcinoma compared with those women without this disease (55). If COX-2 is also overexpressed in endometrial hyperplasias and cancers, as in some of the other cancers (56), then further constant stimulation of the enzyme by LH may predispose normal endometrium to become neoplastic. The high PG levels produced from constant COX-2 stimulation can increase cell proliferation, decrease apoptosis, and decrease adhesion to extracellular matrix proteins (56, 57).
In summary, treatment of human uterine endometrial gland epithelial cells with hCG results in a time- and a dose-dependent and hormone-specific increase in COX-2 gene expression. The hCG effect is mediated by the cAMP/type I PKA signaling pathway and requires a normal complement of its receptors in cells. Thus, the effects of hCG and LH may be yet another action of these hormones in human endometrium that is important for implantation of the blastocyst and continuation of pregnancy.
Received February 5, 1999.
Revised April 26, 1999.
Accepted May 20, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. Hudelist, A. Huber, M. Knoefler, S. Haider, A. Kolbus, K. Czerwenka, S. Helmy, E. Kubista, and C. F. Singer {beta}-HCG/LH Receptor ({beta}-HCG/LH-R) Expression in Eutopic Endometrium and Endometriotic Implants: Evidence for {beta}-HCG Sensitivity of Endometriosis Reproductive Sciences, July 1, 2008; 15(6): 543 - 551. [Abstract] [PDF]
|
|
|
|
 |
|
 Z Cheng, E L Sheldrick, E Marshall, D C Wathes, D R E Abayasekara, and A P F Flint Control of cyclic AMP concentration in bovine endometrial stromal cells by arachidonic acid Reproduction, May 1, 2007; 133(5): 1017 - 1026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.M. Lei, M. Yang, X. Li, O. Takikawa, and C.V. Rao Upregulation of Placental Indoleamine 2,3-Dioxygenase by Human Chorionic Gonadotropin Biol Reprod, April 1, 2007; 76(4): 639 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Cameo, M. Szmidt, Z. Strakova, P. Mavrogianis, K. L. Sharpe-Timms, and A. T. Fazleabas Decidualization Regulates the Expression of the Endometrial Chorionic Gonadotropin Receptor in the Primate Biol Reprod, November 1, 2006; 75(5): 681 - 689. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Strakova, P. Mavrogianis, X. Meng, J. M. Hastings, K. S. Jackson, P. Cameo, A. Brudney, O. Knight, and A. T. Fazleabas In Vivo Infusion of Interleukin-1{beta} and Chorionic Gonadotropin Induces Endometrial Changes that Mimic Early Pregnancy Events in the Baboon Endocrinology, September 1, 2005; 146(9): 4097 - 4104. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D.X. Lin, Z.M. Lei, and Ch.V. Rao Dependence of Uterine Cyclooxygenase2 Expression on Luteinizing Hormone Signaling Biol Reprod, August 1, 2005; 73(2): 256 - 260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kang, A. Akoum, P. Chapdelaine, P. Laberge, P. E. Poubelle, and M. A. Fortier Independent regulation of prostaglandins and monocyte chemoattractant protein-1 by interleukin-1{beta} and hCG in human endometrial cells Hum. Reprod., November 1, 2004; 19(11): 2465 - 2473. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Perrier d'Hauterive, C. Charlet-Renard, S. Berndt, M. Dubois, C. Munaut, F. Goffin, M.-T. Hagelstein, A. Noel, A. Hazout, J.-M. Foidart, et al. Human chorionic gonadotropin and growth factors at the embryonic-endometrial interface control leukemia inhibitory factor (LIF) and interleukin 6 (IL-6) secretion by human endometrial epithelium Hum. Reprod., November 1, 2004; 19(11): 2633 - 2643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Fields and M. Shemesh Extragonadal Luteinizing Hormone Receptors in the Reproductive Tract of Domestic Animals Biol Reprod, November 1, 2004; 71(5): 1412 - 1418. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. El-Toukhy, A. Taylor, Y. Khalaf, K. Al-Darazi, P. Rowell, P. Seed, and P. Braude Pituitary suppression in ultrasound-monitored frozen embryo replacement cycles. A randomised study Hum. Reprod., April 1, 2004; 19(4): 874 - 879. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. Lin, X. Li, Z. M. Lei, and Ch. V. Rao Human Cervix Contains Functional Luteinizing Hormone/Human Chorionic Gonadotropin Receptors J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3409 - 3414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Srisuparp, Z. Strakova, A. Brudney, S. Mukherjee, S. Reierstad, M. Hunzicker-Dunn, and A. T. Fazleabas Signal Transduction Pathways Activated by Chorionic Gonadotropin in the Primate Endometrial Epithelial Cells Biol Reprod, February 1, 2003; 68(2): 457 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Albert, N. Garrido, A. Mercader, C.V. Rao, J. Remohi, C. Simon, and A. Pellicer The role of endothelial cells in the pathogenesis of ovarian hyperstimulation syndrome Mol. Hum. Reprod., May 1, 2002; 8(5): 409 - 418. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Kiemer, M. D. Lehner, T. Hartung, and A. M. Vollmar Inhibition of Cyclooxygenase-2 by Natriuretic Peptides Endocrinology, March 1, 2002; 143(3): 846 - 852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.J.P. Jones and A.T. Fazleabas Ultrastructure of epithelial plaque formation and stromal cell transformation by post-ovulatory chorionic gonadotrophin treatment in the baboon (Papio anubis) Hum. Reprod., December 1, 2001; 16(12): 2680 - 2690. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Syed, G. Ulinski, S. C. Mok, G. K. Yiu, and S.-M. Ho Expression of Gonadotropin Receptor and Growth Responses to Key Reproductive Hormones in Normal and Malignant Human Ovarian Surface Epithelial Cells Cancer Res., September 1, 2001; 61(18): 6768 - 6776. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Munir, K. Fukunaga, H. Kanasaki, K. Miyazaki, T. Ohba, H. Okamura, and E. Miyamoto Expression of Cyclooxygenase 2 by Prostaglandin E2 in Human Endometrial Adenocarcinoma Cell Line HEC-1B Biol Reprod, September 1, 2000; 63(3): 933 - 941. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
