Abstract
Although an inflammatory microenvironment is required for successful implantation, an inflammatory overreaction is one of the causes of unexplained recurrent pregnancy losses (uRPL). Prostaglandin E2 (PGE2) plays a pivotal role in regulating immune balance during early pregnancy, and it can stimulate inflammatory reactions via prostaglandin E2 receptor 3 (EP3). However, the role of PGE2 receptor signaling in the uRPL remains unknown. We aimed to investigate whether EP3 signaling is involved in the mechanism of uRPL. Via immunohistochemistry we could show that the expression of cyclooxygenase-2, EP3 and G protein alpha inhibitor 1 (Gi1) was enhanced in the decidua of the uRPL group in comparison to the control group in first-trimester placentas. In vitro, we demonstrated that sulprostone (an EP1/EP3 agonist) inhibited the secretion of beta-hCG and progesterone in JEG-3 cells and the secretion of beta-hCG in HTR-8/SVneo cells while it induced the expression of plasminogen activator inhibitor type 1 in JEG-3 cells. In addition, PGE2/sulprostone was able to stimulate the expression of Gi1, phosphorylated-extracellular signal-regulated kinases 1/2 (p-ERK1/2) and p53. L-798,106 (an EP3-specific antagonist) suppressed the expression of EP3 and p-ERK1/2 without affecting the secretion of beta-hCG. Elevated activation of EP3 signaling in first-trimester placentas plays an important role in regulating the inflammatory microenvironment, the hormone secretion of extravillous trophoblasts and the remodeling of extracellular matrix in the fetal-maternal interface. L-798,106 might be a ‘potential therapeutic candidate’ for the treatment of uRPL.
Introduction
Recurrent pregnancy losses (RPL) are defined as two or more consecutive failed pregnancies before the 20th week of pregnancy according to the Practice Committee of the American Society for Reproductive Medicine (1). RPL affects 1% of all women worldwide (2), and only a minority can be explained by particular causative factors, including genetic abnormalities, structural abnormalities, infection, endocrine abnormalities, immune dysfunction and thrombophilic disorders (1). Up to 50% of RPL cases are classified as unexplained RPL (uRPL). Abnormal inflammation in the fetal-maternal interface and unbalanced extracellular matrix (ECM) remodeling are considered as key factors contributing to uRPL (3).
Prostaglandin E2 (PGE2) has long been known as a potentiator of inflammation (4) and is the major fraction of prostaglandins produced by the placenta (5). PGE2 can regulate the balance between Th1 and Th2 cell responses (6), modulate chemokine production (6) and suppress lymphocyte alloreactivity of decidual cells during early gestation (7). Th1/Th2 cell ratios are higher in women with uRPL compared to controls with healthy pregnancies (8). The prostaglandin E1 analog (Misoprostol) and the PGE2 analog (Sulprostone) have been proven to be effective drugs for pregnancy termination (9, 10). PGE2 has been demonstrated to be upregulated in cervical ovulatory mucus (11) and the endometrium (12, 13) of women with uRPL compared to controls with healthy pregnancies during implantation. Therefore, increased expression of PGE2 might be a disadvantage for implantation, placentation and maintenance of pregnancy.
Arachidonic acid is metabolized by cyclooxygenase-1 and cyclooxygenase-2 (COX-2) into prostaglandin H2 (PGH2), which is converted into multiple prostaglandins including PGE2 (14). As a rate-limiting enzyme that catalyzes the synthesis of PGE2, COX-2 is localized in the endometrium throughout the menstrual cycle and during early pregnancy (15). However, results of COX-2 expression in uRPL patients are contradictory.
PGE2 can combine with both nuclear receptors and membrane receptors. Previous studies of our group analyzed the expression of nuclear receptors in the placenta of spontaneous miscarried and RPL patients (16, 17, 18, 19, 20, 21). However, little is known about the expression of membrane receptors of PGE2 in the placenta of women with uRPL. The effects of PGE2 are mainly facilitated by specific membrane-bound G protein-coupled EP receptors (EP1–EP4) with various signaling pathways: EP1 is coupled to the G protein alpha q (Gq) to mobilize intracellular Ca2+, EP2 and EP4 are coupled to the G protein alpha stimulator (Gs) to stimulate adenylyl cyclase, and EP3 is mainly coupled to the G protein alpha inhibitor (Gi) to inhibit adenylyl cyclase (22). All four membrane receptors of PGE2 are expressed by the placenta and fetal membranes, and only EP3 is intensely localized in the membrane of syncytiotrophoblast in the placenta of women in labor (23). This indicates that EP3 may have an influence on trophoblast invasion and placentation. The expression of COX-2 and EP3 has been proven to be elevated in the placenta of a rat model for ischemia-reperfusion injury (24). PGE2 triggers mast cell activation via an EP3-Gi/o-Ca2+ influx/phosphatidyl-3-kinase signaling pathway, which contributes to vascular permeability and consequent edema formation in mice (4). However, no previous studies exist about the expression of EP3 in the placenta of uRPL patients, as well as the possible signal processes of EP3 in normal pregnancy and uRPL during the first trimester of pregnancy.
In addition, PGE2 binding with EP1/EP3 receptor can induce the production of plasminogen activator inhibitor type 1 (PAI-1) in cardiac fibroblasts (25). PAI-1 levels in the plasma are increased in RPL patients in comparison to women with healthy pregnancies (26). Among all the thrombophilic genes, PAI-1-675 4G/5G polymorphism is one of the most frequently analyzed PAI-1 genetic variants, although the contribution of PAI-1-675 4G/5G to uRPL is conflicting (27).
We hypothesized that EP3 signaling may be an essential pathway in the mechanism of uRPL. Our study group aimed to analyze the main proteins’ expression of EP3 signaling in the placentas of first-trimester pregnancies with uRPL immunohistochemically. Additionally, we studied the effects of PGE2, an EP1/EP3 agonist (sulprostone) and an EP3 antagonist (L-798,106) on the function of trophoblast cells by using the choriocarcinoma cell line JEG-3 and the human trophoblast-derived cells HTR-8/SVneo in vitro.
Materials and methods
Placental tissue samples
The study was approved by the ethical committee of the Medical Faculty, Ludwig-Maximilian-University of Munich (Number of approval: 337-06) and informed consent was obtained from each patient in written form. Samples and clinical information were anonymized and encoded for statistical workup.
We analyzed placenta tissues of 19 patients with a history of more than two consecutive pregnancy losses of unknown causes (uRPL group) and of 19 healthy controls with legal pregnancy termination (control group) from a private practice in Munich, Germany. Exclusion criteria for the uRPL group, described thoroughly in our previous studies (28), include infectious diseases, uterine anomalies, endocrinological dysfunctions, hyperprolactinemia, hyperandrogenemia, thyroidal dysfunctions, autoimmunologic disorders, deficiencies in coagulation factors, as well as fetal and parental chromosomal disorders. All placentas were obtained within the first 24 h after diagnosis without any prior hormonal pretreatment. Samples were obtained by dilatation and evacuation without any prior pharmaceutical induction. These samples were fixed immediately in 4% buffered formalin for 20–24 h and embedded in paraffin for further immunohistochemical and double immunofluorescence analysis (29).
Immunohistochemistry
Paraffin-embedded slides were dewaxed in xylol and washed in 100% ethanol. For inhibition of the endogen peroxidases, tissue samples were incubated in methanol with 3% H2O2 and rehydrated in a descending alcohol series. Slides were heated in a pressure cooker using sodium citrate buffer (pH = 6.0), containing 0.1 M citric acid and 0.1 M sodium citrate in distilled water. After cooling and washing in PBS, all slides were incubated with blocking solution (Reagent 1, Zytochem-Plus HRP-Polymer-Kit (mouse/rabbit)) for 20 min to avoid non-specific binding of the primary antibodies. The primary antibodies used for the experiment were anti-COX rabbit IgG polyclonal antibody (Sigma, SAB4502491, 1:400 dilution), anti-EP3 rabbit IgG polyclonal antibody (Abcam, ab189131, 1:300 dilution) and anti-Gi1 rabbit IgG polyclonal antibody (Novus Biologicals, Littleton, CO, USA; NBP2-16558, 1:1000 dilution). The slides were incubated with different primary antibodies for 16 h at 4°C. After washing, the secondary antibodies/complexes of the ABC detection kits (Vector Laboratories) were applied following the manufacturer’s protocols to detect reactivity. Immunostaining was visualized with the substrate and the chromogen-3, 3′-diaminobenzidine (DAB; Dako) for 1 min in case of COX-2, 2.5 min for EP3 and 45 sec for Gi1.
Third-trimester placenta was used as a positive control for the immunohistochemical staining of COX-2 and Gi1 to test antibody function and to determine an appropriate dilution of the antibody for staining. Colon was used as a positive control for EP3. Positive cells showed a brownish color and the negative control, as well as unstained cells, appeared blue (30). Negative controls were used with the same control tissues and were performed by replacement of a pre-immune serum at the same concentration as the primary antibody (negative control for super sensitive rabbit antibodies, BioGenex, California, USA).
All slides were analyzed under the microscope by two independent observers using a Leitz photomicroscope. For the light microscopy analysis, a semi-quantitative IRS score was used (31), which is calculated via the multiplication of optical staining intensity (grades: 0 = no, 1 = weak, 2 = moderate and 3 = strong staining) and the percentage range of positive stained cells (0 = no staining, 1 = ≤10% of the cells; 2 = 11–50% of the cells; 3 = 51–80% of the cells and 4 = ≥81% of the cells were stained for the antibody, respectively).
Double immunofluorescence staining
For the characterization of COX-2, EP3 and Gi1-expressing cells in the decidua, we applied the same paraffin-embedded slides with double immunofluorescence staining. The same experimental steps were carried out as for immunohistochemistry until the step of blocking. Slides were blocked with a blocking solution (Ultra V Block, Lab Vision, Fremont, CA, USA) to avoid non-specific staining and then incubated with various primary antibodies overnight at 4°C. Prolactin was used as a specific marker for stromal cells and human leukocyte antigen-G (HLA-G) was used as a specific marker for trophoblast cells. Different primary antibodies were diluted with a diluting medium (Dako) according to the following ratios: 1:400 for rabbit anti-COX-2 polyclonal IgG (Sigma, SAB4502491) and 1:500 for mouse anti-Prolactin polyclonal IgG (Serotec, Kidlington, Oxfordshire, UK; MCA712), 1:300 for rabbit anti-EP3 polyclonal IgG (Abcam, ab189131), 1: 300 for rabbit anti-Gi1 polyclonal IgG (Novus Biologicals, NBP2-16558), 1:50 for mouse anti-HLA-G polyclonal IgG (Serotec, MCA2044). After washing, slides were incubated with Cy2-/Cy3-labeled antibodies (Jackson Dianova, Hamburg, Germany) as fluorescent secondary antibodies for 30 min at room temperature in darkness to avoid fluorescence quenching. Cy2-labeled secondary antibodies were at a dilution of 1:100 and Cy3-labeled antibodies were at a dilution of 1:500. Finally, the slides were embedded in mounting buffer containing 4′,6-diamino-2-phenylindole (DAPI, Vectastain, Vector Laboratories) for blue staining of the nucleus after washing and drying. Digital images were obtained with a digital camera system (Axiocam; Zeiss CF20DXC; KAPPA Messtechnik, Gleichen, Germany) and digitally saved.
Cell culture and cell stimulation
JEG-3 cells (ATCC HTB-36) and HTR-8/SVneo cells (ATCC CRL-3271) were used as trophoblast models. Both cell lines were cultured in RPMI 1640 medium + Gluta MAX (Gibco) supplemented by 10% fetal bovine serum (FCS, Gibco). Cells were seeded into 24-well plates for the determination of beta-hCG, progesterone and PAI-1 and 6-well plates for Western blotting. After 6–8 h, cell culture medium was replaced with fresh RPMI1640 with stimulation chemicals for the remaining 48 h, which included dimethyl sulfoxide (DMSO, 0.5%) as vehicle control, PGE2 (TOCRIS, Bristol, UK), sulprostone (TOCRIS) and L-798,106 (TOCRIS). Both JEG-3 (50,000 cells/ well) and HTR-8/SVneo cells (60,000 cells/well) were incubated with PGE2, sulprostone and L-798,106 at different concentrations of 0.1, 1, 10, 100, 1000 nM and 10 µM in 24-well plates.
Cell viability assay
JEG-3 cells were seeded at the density of 1.5 × 104 cells/well in 96-well plates and were incubated with different concentration of PGE2 and L-798,106 for 48 h. To each well, 20 μg MTT (Sigma) were added for 1.5 h at 37°C. After removing MTT, 200 μL DMSO were added to each well and mixed thoroughly on the shaker for 5 min at room temperature. The optical density was examined at 595 nm using Elx800 universal Microplate Reader.
ADVIA Centaur assay
Supernatants of JEG-3 and HTR-8/SVneo cells in 24-well plates were collected and centrifuged (13,200 g , 10 min) to remove debris. The concentration of beta-hCG and progesterone in the supernatant was detected according to the manufacturer’s instructions on an ADVIA Centaur XP autoanalyzer (Siemens Medical Solution Diagnostics) as described in our previous publications (32, 33). The concentration of total beta-hCG was measured in a two-site sandwich immunoassay (ADVIA Centaur Total hCG Test; Siemens Medical Solutions Diagnostics). The detectable concentration of total beta-hCG ranges from 2.0 IU/mL to 1000 IU/mL. The concentration of progesterone in supernatants was measured by an automated quantitative immunoassay (ADVIA Centaur Progesterone Test; Siemens Medical Solutions Diagnostics) with a sensitivity of 0.2 ng/mL. Intraobserver and interobserver coefficients of variation were 9.8 and 5.8%, respectively.
PAI-1 ELISA
PAI-1 levels in the supernatants of JEG-3 and HTR-8/SVneo were measured utilizing commercially available ELISA kit (R&D system, DSE100). A standard curve was obtained in parallel to each assay and results converted into ng/mL. The intra- and inter-assay variability was respectively 4.6 and 8.7%.
Western blotting
Cell lysates were extracted from JEG-3 and HTR-8/SVneo cells with radioimmunoprecipatation assay buffer (RIPA, Sigma-Aldrich, R0278-50ML). For Western blotting, 20 µg of cell lysates were first separated in 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to a polyvinylidene fluoride membrane (Bio-Rad). The membrane was blocked in 4% skim milk powder and then incubated with the primary antibodies for 16 h at room temperature. Different primary antibodies were used as the following: rabbit polyclonal anti-EP3 antibody (Abcam, ab94496, 1:500), rabbit polyclonal anti-Gi1 antibody (Novus Biologicals, NBP2-16558, 1:500), rabbit polyclonal anti-phosphorylated extracellular signal-regulated kinases (p-ERK1/2) antibody (Abcam, ab47339, 1:500), mouse monoclonal anti-p53 antibody (Abcam, ab26, 1:200). Beta-actin was used as a housekeeping gene and mouse monoclonal anti-beta-actin antibody was diluted 1:1000 in 4% milk powder (Sigma, A5441). Afterward, the membrane was incubated with the goat-anti-rabbit/mouse secondary antibody conjugated with alkaline phosphatase (1:1000 dilution, Jackson Immuno Research), and detected with 5-bromo-4-chloro-3′-indolylphosphate/nitro-blue tetrazolium (BCIP/NBT)-chromogen substrate solution (Promega). Western blots were scanned and quantified using the GelScan V6.0 1D Analysis Software (SERVA, Electrophoresis GmbH, Heidelberg, Germany). Band intensities of EP3, Gi1, p-ERK1/2 and p53 were normalized with band intensities of beta-actin. The blots were repeated three times.
Statistical analysis
All data were analyzed with SPSS Statistics 23 software (IBM Corporation, Armonk, NY, USA) and are expressed as the mean ± s.d. Independent t-test was used to compare the clinical data of uRPL and control group. Mann–Whitney-U test was applied for evaluating IRS scores of EP3, COX-2 and Gi1 expression in placentas of two groups. Spearman’s rank correlation analysis was adopted to evaluate the correlation between two monotonic, nonlinear variables. Wilcoxon test was used for the evaluation of beta-hCG, progesterone and PAI-1 expression levels between vehicle and stimulation groups. Wilcoxon test was also used for analyzing the band intensities of EP3, Gi1, p-ERK1/2 and p53. P-values <0.05 were regarded statistically significant.
Results
Clinical data of the uRPL group and control group
The mean age of the women in the uRPL group was 37.76 ± 4.88 years and of the control group 35.78 ± 5.88 years (P = 0.41). Mean gestational age of the uRPL group was 9.09 ± 2.17 weeks and of the control group 9.71 ± 1.88 weeks (P = 0.66). The gravidity times were decreased in patients with uRPL (3.11 ± 1.08) compared to healthy controls (3.42 ± 1.90, P = 0.002), while the parity times of two groups showed no significant differences (0.94 ± 0.94 vs 1.63 ± 1.12, P = 0.35). Demographic and clinical data are illustrated in Table 1.
Demographic and clinical characteristics of the study population.
| Characteristic | Normal pregnancy (n = 19) | uRPL (n = 19) | P-value (t-test) |
|---|---|---|---|
| Maternal age (years) | 35.78 ± 5.88 (25–46) | 37.76 ± 4.88 (30–44) | 0.41 |
| Gestational age (weeks) | 9.71 ± 1.88 (6–13) | 9.09 ± 2.17 (4–12) | 0.66 |
| Gravidity | 3.42 ± 1.90 (1–7) | 3.11 ± 1.08 (2–5) | 0.002 |
| Parity | 1.63 ± 1.12 (0–4) | 0.94 ± 0.94 (0–3) | 0.35 |
Values are represented as mean ± s.d.; the range is shown in parentheses.
Expression of COX-2, EP3 and Gi1 in the placenta of women with uRPL and controls
Expression of COX-2, EP3 and Gi1 was identified in the cytoplasm of cells in the syncytium and the decidua of first-trimester pregnancies in both the uRPL group and the control group (Fig. 1A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q and R). A significant increase in the expression of COX-2 could be observed in the syncytium of the uRPL group in comparison to the control group (IRS 8.73 vs 5.59; P = 0.007, Fig. 1A, B and E). COX-2 expression in the decidua of the uRPL group was also significantly increased in comparison to the control group (IRS 4.29 vs 2.00; P = 0.004, Fig. 1C, D and F).
Immunohistological analyses of COX-2, EP3 and Gi1 expression in placentas of uRPL patients (B, D, H, J, N and P) and healthy controls (A, C, G, I, M and O) from the first trimester were measured via IRS score. In the syncytium, the stainings of COX-2 (A and B), EP3 (G and H) and Gi1 (M and N) are compared between the uRPL patients and controls, which are represented as box-plots (E, K and Q). In the decidua, the expressions of COX-2 (C and D), EP3 (I and J) and Gi1 (O and P) are increased in uRPL patients compared to controls, which are represented as box-plots (F, L and R). The range between the 25th and 75th percentiles is represented by the boxes with a horizontal line at the median. The bars show the 5th and 95th percentiles. Dots indicate values more than 1.5 box lengths from the 75th percentile. Magnification ×10 lens, scale bar = 200 µm. uRPL = unexplained recurrent pregnancy losses. COX-2, cyclooxygenase-2; EP3, prostaglandin E2 receptor 3; Gi1, G protein alpha inhibitor 1; uRPL, unexplained recurrent pregnancy losses.
Citation: Endocrine Connections 7, 5; 10.1530/EC-18-0106
Expression of both EP3 and Gi1 in the decidua was significantly upregulated in patients with uRPL compared to normal controls (Fig. 1L and R), but no significant changes of EP3 and Gi1 could be observed in the syncytium (Fig. 1K and Q). In the uRPL group, EP3-staining in the decidua was more intense with a mean IRS of 4.15, which is higher than the mean IRS of the control group (IRS = 2.22, P = 0.001, Fig. 1I, J and L). EP3-staining in the syncytium of the uRPL group was similar to the control group (IRS 5.75 vs 6.32; P = 0.350, Fig. 1G, H and K). Gi1-staining was stronger in the decidua of the uRPL group compared to the control group (IRS 4.48 vs 2.89; P = 0.008, Fig. 1O, P and R). There was no significant staining difference for Gi1-staining in the syncytium between the uRPL group and the control group (IRS 5.27 vs 4.47; P = 0.292, Fig. 1M, N and Q). Both positive and negative controls are shown in the Supplementary Fig. 1 (see section on supplementary data given at the end of this article).
Expressions of COX-2, EP3 and Gi1 were all increased in the decidua of patients with uRPL compared to healthy controls, implying that a correlation could exist between them. We further analyzed the correlation of each two of these three components in the decidua. Unfortunately, there was no statistically significant correlation between COX-2 and EP3 expression (r = −0.059; P = 0.788), neither for EP3 and Gi1 (r = 0.251; P = 0.261) nor for COX-2 and Gi1 (r = 0.158, P = 0.450) in the decidua.
Localization of COX-2, EP3 and Gi1 in the decidua of first-trimester placentas
As the decidua consists of maternal decidual stroma cells and extravillous trophoblasts, double immunofluorescence was used to identify COX-2, EP3 and Gi1-expressing cells. Prolactin was used as a marker for stromal cells and HLA-G as a maker for trophoblasts, respectively. COX-2 was co-expressed with prolactin predominantly in the cytoplasm of stromal cells (Fig. 2A, B and C). Both EP3 (Fig. 2D, E and F) and Gi1 (Fig. 2H, I and G) were co-expressed with HLA-G in the cytoplasm of extravillous trophoblasts. Gi1 was expressed especially beneath the cell membrane of extravillous trophoblasts (Fig. 2G). Since co-expression of COX-2, EP3 and Gi1 was similar in both the uRPL and control group, we only showed pictures of the uRPL group.
Localization of COX-2, EP3 and Gi1 is shown in the decidua of patients with uRPL. COX-2 is co-expressed with prolactin, which is a specific marker for stromal cells (A, B and C). EP3 is co-expressed with HLA-G, which is a specific marker for extravillous trophoblasts (D, E and F). Gi1 is co-expressed with HLA-G in extravillous trophoblasts (G, H and I). Co-expression of COX-2, EP3 and Gi1 is shown with →. Magnification ×63, scale bar = 20 µm. COX-2, cyclooxygenase-2; EP3, prostaglandin E2 receptor 3; Gi1, G protein alpha inhibitor 1; uRPL, unexplained recurrent pregnancy losses.
Citation: Endocrine Connections 7, 5; 10.1530/EC-18-0106
Influence of PGE2, sulprostone and L-798,106 on the expression of EP3 signaling in JEG-3 cells in vitro
In order to investigate the mechanism of the EP3 signaling in extravillous trophoblasts of uRPL in vitro, JEG-3 and HTR-8/SVneo cells were used and stimulated with different concentrations of PGE2, EP1/EP3 agonist (sulprostone) and EP3 antagonist (L-798,106) for 48 h.
Firstly, the MTT assay was used to assess the viability of JEG-3 cells after 48 h of treatment with 0.1, 1, 10, 100, 1000 nM and 10 µM PGE2, L-798,106 or the vehicle control (DMSO, 0.1%). 10 nM PGE2 and 10 nM L-798,106 significantly increased viability of JEG-3 cells compared to control group (P < 0.05, data not shown).
For JEG-3 cells, the production of beta-hCG was suppressed from 16.33 ± 3.30 IU/mL in the vehicle to 12.20 ± 1.76 IU/mL by 10 nM PGE2 (P = 0.046), while its concentration was not significantly influenced by 0.1 nM and 1 nM PGE2 (Fig. 3A). Additionally, beta-hCG expression was inhibited from 16.33 ± 3.30 IU/mL in the vehicle to 10.19 ± 1.79 IU/mL and 10.99 ± 1.15 IU/mL when sulprostone concentration was 1 nM and 10 nM (P = 0.028, P = 0.028, respectively, Fig. 3A). However, beta-hCG levels were not altered by L-798,106 independent of the concentration of 0.1, 1 or 10 nM (Fig. 3A).
Expression of beta-hCG, progesterone, PAI-1, EP3, Gi1, p-ERK1/2 in JEG-3 cells after incubation with PGE2, the EP1/EP3 agonist (sulprostone) and EP3 antagonist (L-798,106). Beta-hCG, progesterone and PAI-1 expression levels were measured via ELISA (each group n = 6). EP3, Gi1, p-ERK1/2 expression extracted from JEG-3 cells were detected by Western blotting, following treatment with PGE2 (10 nM) and L-798,106 (10 nM) for 48 h (n = 3). All data are shown as mean ± s.d. and statistically significant differences (P < 0.05) between individual treatment groups and the vehicle are marked with an asterisk. EP3, prostaglandin E2 receptor 3; Gi1, G protein alpha inhibitor 1; PGE2, prostaglandin E2; PAI-1, plasminogen activator inhibitor; p-ERK1/2, phosphorylated extracellular signal-regulated kinases 1/2.
Citation: Endocrine Connections 7, 5; 10.1530/EC-18-0106
Progesterone expression of JEG-3 cells was downregulated by PGE2, sulprostone and L-798,106, independent of different concentrations (0.1, 1 and 10 nM) in comparison to the vehicle after 48 h (P = 0.028, each) (Fig. 3B).
PAI-1 levels were significantly altered 34% higher through 10 nM sulprostone than the vehicle, while no significant changes of PAI-1 expression were detected through 10 nM PGE2 and 10 nM L-798,106 (Fig. 3C).
Western blotting was utilized to access the expression of EP3, Gi1 and p-ERK1/2 in JEG-3 cells stimulated by PGE2 and L-798,106, 10 nM each (Fig. 3D). The molecular weight of EP3 is 37 kDa and of Gi1 it is 40 kDa. EP3 expression was decreased by 31.8% after treatment with 10 nM L-798,106 compared to the vehicle group (P = 0.008, Fig. 3D). Gi1 expression was increased by 12.4% through 10 nM PGE2 (P = 0.012, Fig. 3E). The molecular weights of p-ERK1/2 are 44 and 42 kDa.
JEG-3 cells were incubated with 0.5% DMSO for 1, 2, 6, 12, 24, 48 h and the results showed there was the strongest expression of p-ERK1/2 after 1 h incubation (Supplementary Fig. 2A). After 1 h incubation, the expression of p-ERK1/2 was stimulated by 18.3% through 10 nM sulprostone compared to the vehicle (P = 0.035), while no significant change was detected in the group treated with 10 nM PGE2 or L-798,106 (Fig. 3F).
Influence of PGE2, sulprostone and L-798,106 on the expression of EP3 signaling in HTR-8/SVneo cells in vitro
HTR-8/SVneo cells were incubated with PGE2, sulprostone and L-798,106 at a concentration of 10 nM and 100 nM, respectively. Beta-hCG levels of HTR-8/SVneo cells were downregulated through PGE2, sulprostone and L-798,106 for all concentrations after 48 h (P < 0.05, each) (Fig. 4A). Progesterone levels could not be detected in the supernatants of HTR-8/SVneo cells. PAI-1 levels were reduced 3.7% through 10 nM L-798,106 than the vehicle, while no significant changes of PAI-1 expression were detected through 10 nM sulprostone (Fig. 4B)
Expression of beta-hCG (A), PAI-1 (B), EP3 (C), Gi1 (D), p-ERK1/2 (E) and p53 (F) in HTR-8/SVneo cells after incubation with PGE2, the EP1/EP3 agonist (sulprostone) and EP3 antagonist (L-798,106). The levels beta-hCG and PAI-1 were measured via ELISA (each group n = 6). EP3, Gi1, p-ERK1/2 and p53 expression extracted from HTR-8/SVneo cells were detected by Western blotting, following treatment with PGE2 (10 nM), sulprostone (10 nM) and L-798,106 (10 nM) for 48 h (n = 3). All data are shown as mean ± s.d. and statistically significant differences (P < 0.05) between individual treatment groups and the vehicle are marked with an asterisk. EP3, prostaglandin E2 receptor 3; PGE2, prostaglandin E2; PAI-1, plasminogen activator inhibitor; p-ERK1/2, phosphorylated extracellular signal-regulated kinases 1/2.
Citation: Endocrine Connections 7, 5; 10.1530/EC-18-0106
In Western blotting, EP3 expression was increased by 11.6% through 10 nM PGE2 (P = 0.011, Fig. 4C) and decreased by 25.9% through 10 nM L-798,106 compared to the vehicle (P = 0.008, Fig. 4C). Gi1 expression was increased by 22.5% through 10 nM sulprostone (P = 0.046, Fig. 4D), while no significant changes were observed through 10 nM PGE2. The expression of p-ERK1/2 decreased when the incubation time with 0.5% DMSO is increased from 1 to 48 h and the strongest expression of p-ERK1/2 was after 1 h incubation (Supplementary Fig. 2A). The expression of p-ERK1/2 was inhibited by 21.8% after treatment with 10 nM L-798,106 for 1 h compared to the vehicle group (P = 0.011, Fig. 4E).The molecular weight of p53 is 53 kDa, and its expression was stimulated by 17.0% through 10 nM PGE2 (P = 0.028, Fig. 4F) and by 24.5% through 10 nM sulprostone (P = 0.028, Fig. 4F).
Discussion
An increased inflammatory reaction is one of the main causes of uRPL (8). PGE2 plays a pivotal role in regulating immune balance and angiogenesis during early pregnancy (8), and it is able to stimulate inflammatory reactions via EP3 (4). The role of PGE2 receptor signaling in the mechanism of uRPL still remains unknown. Our group is the first to investigate EP3 signaling pathway in placentas of patients with uRPL.
EP3 signaling is induced in the extravillous trophoblasts of uRPL patients. Arachidonic acid is metabolized by COX-2 into PGH2, which is converted into multiple prostaglandins, such as PGE2 (14). PGE2 acts through different membrane receptors (EP1–4) (22). In the placenta of uRPL patients, enhanced levels of COX-2 might be the reason for increased PGE2 expression. PGE2 coupling with EP3 induces Gi1 and reduces cAMP, which can eventually lead to a downregulation of beta-hCG and progesterone and an upregulation of PAI-1. The decreased cAMP/PKA signaling inhibits the secretion of progesterone via PKA (50) and the transcription of beta-hCG via CREB (49). The activated EP3/Gi1 can stimulate p-ERK1/2 and p53, which eventually enhance PAI-1 gene transcription (52). These changes could prevent trophoblast implantation and placentation, finally causing recurrent pregnancy losses. COX-2, cyclooxygenase-2; Gi1, G protein alpha inhibitor 1; Gq, Gprotein alpha q; Gs, G protein alpha stimulator; PAI-1, plasminogen activator inhibitor type 1; p-ERK1/2, phosphorylated extracellular signal-regulated kinases 1/2; PGE2, prostaglandin E2; PGH2, prostaglandin H2; PKA, protein kinase A; uRPL, unexplained recurrent pregnancy losses.
Citation: Endocrine Connections 7, 5; 10.1530/EC-18-0106
As the rate-limiting enzyme for PGE2 synthesis, COX-2 is expressed by human endometrial stromal cells (34), and it is involved in the implantation of blastocysts in the early pregnancy phase (35). So far, a small number of publications exist about COX-2 expression levels in uRPL, but the results are conflicting. One research group reported that lower expression levels of COX-2 have been found in the chorionic villi of women with uRPL (35), and another group showed the similar result in the embryo of mice with autoimmune-type recurrent miscarriages (36). Our result is in accordance with a further study group who found that the expression levels of COX-2 are increased in the endometrium of uRPL women in comparison to healthy controls (13). In this study, expression of PGE2 and pro-inflammatory cytokines is also induced in the endometrium of women with uRPL, such as interleukin 1beta, tumor necrosis factor alpha (TNF-alpha), interferon gamma (IFN-gamma) and transforming growth factor beta1 (TGF-beta1) (13). IL-1, TNF-alpha and IFN-gamma belong to Th1 cytokines, and this Th1 predominance has been known to be one of the immunological reasons for uRPL (37). In addition, TNF-alpha can induce COX-2 gene expression in first-trimester trophoblasts (38).
Ryantova et al. (2008) observed that PGE2 levels in the cervical ovulatory mucus are increased in patients with uRPL (11), which is mainly caused by increased levels of COX-2. However, the effects of PGE2 on trophoblast cells in vitro are contradictory. Biondi et al. (2006) proved that PGE2 inhibits proliferation and migration of HTR-8/SVneo cells by enhancing cAMP (39), while Horita et al. (2007) suggested that PGE2 promotes migration of HTR-8/SVneo cells by leukemia inhibitory factor (40). We investigated that PGE2 inhibited the production of beta-hCG and progesterone in JEG-3 cells and the production of beta-hCG in HTR-8/SVneo cells. Low expression levels of beta-hCG and progesterone in the serum have been used to diagnose early pregnancy losses (41, 42). Intravaginal PGE2 can decrease the excretion of hCG in the urine and progesterone in the plasma (43), which is in line with our study that high expression levels of PGE2 in the placenta are detrimental to pregnancy maintenance.
Prostaglandin E2 receptors (EP1–4) are G protein-coupled receptors that both bind and are activated by PGE2. Yamazaki et al. (2006) indicated that mRNA and protein expression of COX-2 and EP3 are increased in the placenta of rats after ischemia-reperfusion, which is associated with intrauterine growth restriction (24). In our study we could demonstrate that EP3 expression was mainly expressed in the extravillous trophoblasts and increased in extravillous trophoblast cells of uRPL patients compared to normal pregnancies in the first trimester of pregnancy. The EP3 gene formed nine distinct mRNAs encoding at least eight EP3 isoforms (44), and the anti-EP3 antibody we used could detect the protein region within the internal sequence amino acids 360–409. It might partly explain different band signals of EP3 in JEG-3 and HTR-8/SVneo cells for in vitro studies. We furthermore investigated that sulprostone (an EP1/EP3 agonist) inhibited the secretion of beta-hCG and progesterone in JEG-3 cells and beta-hCG expression in HTR-8/SVneo cells. Sulprostone, as a PGE2 analog, can be used for medical termination of early pregnancy (10). Sulprostone coupling with EP1 or EP3 receptor may result in a reduced production of beta-hCG and progesterone in extravillous trophoblasts. Both are detrimental to blastocyst growth and pregnancy maintenance (45). However, Biondi et al. (2006) reported that PGE2 inhibited proliferation and migration of HTR-8/SVneo cells via EP2 and EP4 receptors instead of EP3 (39). In order to investigate the exhaustive mechanisms of membrane receptors of PGE2 in the placenta of uRPL, the remaining membrane receptors of PGE2 (EP1, EP2 and EP4) still need to be explored, among which the EP1 receptor represents an important receptor we will focus on in future investigations (46).
The activation of G protein-coupled receptors stimulates the activity of adenylate cyclase (AC), increases intracellular levels of cAMP and activates protein kinase A (PKA), and this cAMP/PKA signaling cascade is a key second messenger pathway for steroid biosynthesis (47). Gi inhibits AC, resulting in decreased levels of cAMP and therefore inhibiting activity of PKA (48). We found that the expression of Gi1 was mainly expressed in extravillous trophoblasts and upregulated in the decidua of women with uRPL. It suggests that the cAMP/PKA signaling pathway may be suppressed in the extravillous trophoblasts of uRPL. cAMP stimulates the transcription of the alpha-hCG gene expression in villous trophoblasts by interacting with the CREB (49). PKA can stimulate progesterone synthesis in the human placenta by phosphorylation of enzymes (50). We investigated that both PGE2 and sulprostone stimulated the expression of Gi1 and suppressed the production of beta-hCG and progesterone, which can be explained by the activated EP3/Gi1 and inhibited cAMP/PKA signaling.
Interestingly, we investigated that only sulprostone was able to upregulate the expression of PAI-1 in JEG-3 cells, which is known as the main inhibitor of fibrinolysis (51). High plasma levels of PAI-1 and high PAI activities are present in women suffering from uRPL in comparison to healthy controls (26). PGE2 utilizes EP1/EP3 to increase PAI-1 levels in cardiac fibroblasts, contributing to elevated fibrin deposition in aortic stenosis (25). TGF-beta1 can induce PAI-1 gene expression by phosphorylation of extracellular signal-regulated kinases (ERK1/2) and p53 (52). The expression of p53 is increased in chorionic villi of RPL patients compared to healthy controls (53). We found that sulprostone stimulated the expression of phosphorylated-ERK1/2 (p-ERK1/2) in JEG-3 cells and the expression of p53 in HTR-8/SV neo cells. Additionally, the expression of p-ERK1/2 was inhibited by L-798,106 in HTR-8/SV neo cells. The activation of EP3 signaling induces increased expression of p-ERK1/2 and p53 in extravillous trophoblasts, which results in increased PAI-1 gene expression and an imbalance of ECM degradation during first-trimester pregnancies. Both the in vivo study of Yamazaki et al. (24) and also our study suggest that under pathological circumstances large differences in EP3 expression in the placenta can be observed. In contrast, only small magnitude changes of EP3 and PAI-1 were detected in our cell culture studies. It implies that trophoblast cells in vitro can mimic parts of the in vivo environment, however it still cannot replace the complex physical milieu of the placenta. Therefore, in vivo investigations of EP3 regulation in animal models or observational studies in humans should be the focus in future.
In conclusion, we suggest that the reason for increased PGE2 expression in the placenta in women with uRPL might due to enhanced levels of COX-2 in stromal cells of the placenta. PGE2 in combination with EP3 receptor of extravillous trophoblasts induces Gi1 and inhibits AC activity, contributing to reduced levels of cAMP and inhibited activity of PKA (Fig. 5). The decreased cAMP/PKA signaling could suppress the production of beta-hCG and progesterone. At the same time, the activated EP3/Gi1 can stimulate p-ERK1/2 and p53, enhancing PAI-1 gene expression (Fig. 5). Upregulation of PAI-1 can inhibit ECM degradation, contributing to intervillous fibrin deposition in the maternal-fetal interface. These changes could prevent trophoblast implantation and placentation, causing failed pregnancy maintenance. It is still unclear whether the observed findings are reflective of the uRPL or causative of it. We explored that the EP3 antagonist (L-798,106) led to a downregulation of EP3 expression without influencing the beta-hCG expression in JEG-3 cells. This suggests that L-798,106 might be a ‘potential therapeutic candidate’ for the treatment of uRPL. To further investigate the mechanism of uRPL, future studies are necessary to understand the role of the additional membrane receptors of PGE2 (EP1, 2 and 4) and other members of Gi family (Gi2 and Gi3) in the placenta.
Supplementary data
This is linked to the online version of the paper at https://doi.org/10.1530/EC-18-0106.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
Yao Ye was a recipient of a scholarship from the China Scholarship Council. The sponsors did not participate in the study design, data analysis or the manuscript writing.
Author contribution statement
Udo Jeschke and Viktoria von Schönfeldt conceived and designed the experiments; Yao Ye, Christina Kuhn and Martina Rahmeh performed the experiments; Myriam Ripphahn and Roland Immler cultured HTR-8/SVneo cells; Yao Ye and Aurelia Vattai analyzed the data and wrote the manuscript; Markus Sperandio and Udo Jeschke revised the manuscript. Sven Mahner, Nina Ditsch and all other authors read and approved the manuscript.
References
- 1↑
Practice Committee of American Society for Reproductive M. Definitions of infertility and recurrent pregnancy loss: a committee opinion. Fertility and Sterility 2013 99 63.
- 2↑
Rai R & Regan L. Recurrent miscarriage. Lancet 2006 368 601–611. (https://doi.org/10.1016/S0140-6736(06)69204-0)
- 3↑
Kwak-Kim J, Yang KM & Gilman-Sachs A. Recurrent pregnancy loss: a disease of inflammation and coagulation. Journal of Obstetrics and Gynaecology Research 2009 35 609–622. (https://doi.org/10.1111/j.1447-0756.2009.01079.x)
- 4↑
Morimoto K, Shirata N, Taketomi Y, Tsuchiya S, Segi-Nishida E, Inazumi T, Kabashima K, Tanaka S, Murakami M, Narumiya S, et al. Prostaglandin E2-EP3 signaling induces inflammatory swelling by mast cell activation. Journal of Immunology 2014 192 1130–1137. (https://doi.org/10.4049/jimmunol.1300290)
- 5↑
Duchesne MJ, Thaler-Dao H & de Paulet AC. Prostaglandin synthesis in human placenta and fetal membranes. Prostaglandins 1978 15 19–42. (https://doi.org/10.1016/S0090-6980(78)80003-3)
- 6↑
Kalinski P. Regulation of immune responses by prostaglandin E2. Journal of Immunology 2012 188 21–28. (https://doi.org/10.4049/jimmunol.1101029)
- 7↑
Lala PK, Kennedy TG & Parhar RS. Suppression of lymphocyte alloreactivity by early gestational human decidua. II. Characterization of the suppressor mechanisms. Cellular Immunology 1988 116 411–422. (https://doi.org/10.1016/0008-8749(88)90241-9)
- 8↑
Kwak-Kim J, Bao S, Lee SK, Kim JW & Gilman-Sachs A. Immunological modes of pregnancy loss: inflammation, immune effectors, and stress. American Journal of Reproductive Immunology 2014 72 129–140. (https://doi.org/10.1111/aji.12234)
- 9↑
Lin CJ, Chien SC & Chen CP. The use of misoprostol in termination of second-trimester pregnancy. Taiwanese Journal of Obstetrics and Gynecology 2011 50 275–282. (https://doi.org/10.1016/j.tjog.2011.07.003)
- 10↑
Mohan AK, Mathew M & Rizvi SG. Use of intravenous sulprostone for the termination of pregnancy with fetal death in second and early third trimester of pregnancy. Sultan Qaboos University Medical Journal 2008 8 306–309.
- 11↑
Ryantova M, Ulcova-Gallova Z, Micanova Z, Bibkova K & Sediva B. Levels of prostaglandin E2 (PGE2) in cervical ovulatory mucus in women with spontaneous miscarriages. Ceska Gynekologie 2008 73 98–101.
- 12↑
Banerjee P, Ghosh S, Dutta M, Subramani E, Khalpada J, Roychoudhury S, Chakravarty B, Chaudhury K. Identification of key contributory factors responsible for vascular dysfunction in idiopathic recurrent spontaneous miscarriage. PLoS ONE 2013 8 e80940. (https://doi.org/10.1371/journal.pone.0080940)
- 13↑
Banerjee P, Jana SK, Pasricha P, Ghosh S, Chakravarty B & Chaudhury K. Proinflammatory cytokines induced altered expression of cyclooxygenase-2 gene results in unreceptive endometrium in women with idiopathic recurrent spontaneous miscarriage. Fertility and Sterility 2013 99 179–187. (https://doi.org/10.1016/j.fertnstert.2012.08.034)
- 14↑
Simmons DL, Botting RM & Hla T. Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacological Reviews 2004 56 387–437. (https://doi.org/10.1124/pr.56.3.3)
- 15↑
Jones RL, Kelly RW & Critchley HO. Chemokine and cyclooxygenase-2 expression in human endometrium coincides with leukocyte accumulation. Human Reproduction 1997 12 1300–1306. (https://doi.org/10.1093/humrep/12.6.1300)
- 16↑
Toth B, Bastug M, Mylonas I, Scholz C, Makovitzky J, Kunze S, Thaler C, Friese K, Jeschke U. Peroxisome proliferator-activated receptor-gamma in normal human pregnancy and miscarriage. Acta Histochemica 2009 111 372–378. (https://doi.org/10.1016/j.acthis.2008.11.020)
- 17↑
Toth B, Bastug M, Scholz C, Arck P, Schulze S, Kunze S, Friese K, Jeschke U. Leptin and peroxisome proliferator-activated receptors: impact on normal and disturbed first trimester human pregnancy. Histology and Histopathology 2008 23 1465–1475.
- 18↑
Hutter S, Knabl J, Andergassen U & Jeschke U. The role of PPARs in placental immunology: a systematic review of the literature. PPAR Research 2013 2013 970276.
- 19↑
Knabl J, Pestka A, Huttenbrenner R, Plösch T, Ensenauer R, Welbergen L, Hutter S, Günthner-Biller M & Jeschke U. The liver x receptor in correlation with other nuclear receptors in spontaneous and recurrent abortions. PPAR Research 2013 2013 575604.
- 20↑
Pestka A, Fitzgerald JS, Toth B, Markert UR & Jeschke U. Nuclear hormone receptors and female reproduction. Current Molecular Medicine 2013 13 1066–1078. (https://doi.org/10.2174/1566524011313070002)
- 21↑
Pestka A, Toth B, Kuhn C, Hofmann S, Wiest I, Wypior G, Friese K, Jeschke U. Retinoid X receptor alpha and retinoids are key regulators in apoptosis of trophoblasts of patients with recurrent miscarriages. Journal of Molecular Endocrinology 2011 47 145–156. (https://doi.org/10.1530/JME-11-0002)
- 22↑
Sokolowska M, Chen LY, Liu Y, Martinez-Anton A, Qi HY, Logun C, Alsaaty S, Park YH, Kastner DL, Chae JJ, et al. Prostaglandin E2 inhibits NLRP3 inflammasome activation through EP4 receptor and intracellular cyclic AMP in human macrophages. Journal of Immunology 2015 194 5472–5487. (https://doi.org/10.4049/jimmunol.1401343)
- 23↑
Grigsby PL, Sooranna SR, Brockman DE, Johnson MR & Myatt L. Localization and expression of prostaglandin E2 receptors in human placenta and corresponding fetal membranes with labor. American Journal of Obstetrics and Gynecology 2006 195 260–269. (https://doi.org/10.1016/j.ajog.2006.01.082)
- 24↑
Yamazaki K, Endo T, Kitajima Y, Manase K, Nagasawa K, Honnma H, Hayashi T, Kudo R, Saito T. Elevation of both cyclooxygenase-2 and prostaglandin E2 receptor EP3 expressions in rat placenta after uterine artery ischemia-reperfusion. Placenta 2006 27 395–401. (https://doi.org/10.1016/j.placenta.2005.04.007)
- 25↑
Kassem KM, Clevenger MH, Szandzik DL, Peterson E & Harding P. PGE2 reduces MMP-14 and increases plasminogen activator inhibitor-1 in cardiac fibroblasts. Prostaglandins and Other Lipid Mediators 2014 113–115 62–68. (https://doi.org/10.1016/j.prostaglandins.2014.09.002)
- 26↑
Gris JC, Neveu S, Mares P, Biron C, Hedon B & Schved JF. Plasma fibrinolytic activators and their inhibitors in women suffering from early recurrent abortion of unknown etiology. Journal of Laboratory and Clinical Medicine 1993 122 606–615.
- 27↑
Ye Y, Vattai A, Zhang X, Zhu J, Thaler CJ, Mahner S, Jeschke U, von Schönfeldt V. Role of plasminogen activator inhibitor type 1 in pathologies of female reproductive diseases. International Journal of Molecular Sciences 2017 18.
- 28↑
Rogenhofer N, Ochsenkuhn R, von Schonfeldt V, Assef RB & Thaler CJ. Antitrophoblast antibodies are associated with recurrent miscarriages. Fertility and Sterility 2012 97 361–366. (https://doi.org/10.1016/j.fertnstert.2011.11.014)
- 29↑
Jeschke U, Mayr D, Schiessl B, Mylonas I, Schulze S, Kuhn C, Friese K, Walzel H. Expression of galectin-1, -3 (gal-1, gal-3) and the Thomsen-Friedenreich (TF) antigen in normal, IUGR, preeclamptic and HELLP placentas. Placenta 2007 28 1165–1173. (https://doi.org/10.1016/j.placenta.2007.06.006)
- 30↑
Worbs S, Shabani N, Mayr D, Gingelmaier A, Makrigiannakis A, Kuhn C, Jeschke U, Kupka MS, Friese K, Mylonas I. Expression of the inhibin/activin subunits (-alpha, -betaA and -betaB) in normal and carcinogenic endometrial tissue: possible immunohistochemical differentiation markers. Oncology Reports 2007 17 97–104.
- 31↑
Remmele W & Stegner HE. Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue. Der Pathologe 1987 8 138–140.
- 32↑
Jeschke U, Karsten U, Reimer T, Richter DU, Bergemann C, Briese V, Mylonas I, Friese K. Stimulation of hCG protein and mRNA in first trimester villous cytotrophoblast cells in vitro by glycodelin A. Journal of Perinatal Medicine 2005 33 212–218.
- 33↑
Jeschke U, Richter DU, Mobius BM, Briese V, Mylonas I & Friese K. Stimulation of progesterone, estradiol and cortisol in trophoblast tumor bewo cells by glycodelin A N-glycans. Anticancer Research 2007 27 2101–2108.
- 34↑
Tamura M, Sebastian S, Yang S, Gurates B, Fang Z & Bulun SE. Interleukin-1beta elevates cyclooxygenase-2 protein level and enzyme activity via increasing its mRNA stability in human endometrial stromal cells: an effect mediated by extracellularly regulated kinases 1 and 2. Journal of Clinical Endocrinology and Metabolism 2002 87 3263–3273.
- 35↑
Wang Y, Zhao AM & Lin QD. Role of cyclooxygenase-2 signaling pathway dysfunction in unexplained recurrent spontaneous abortion. Chinese Medical Journal 2010 123 1543–1547.
- 36↑
Hua F, Li CH, Wang H & Xu HG. Relationship between expression of COX-2, TNF-alpha, IL-6 and autoimmune-type recurrent miscarriage. Asian Pacific Journal of Tropical Medicine 2013 6 990–994. (https://doi.org/10.1016/S1995-7645(13)60178-9)
- 37↑
Challis JR, Lockwood CJ, Myatt L, Norman JE, Strauss JF 3rd & Petraglia F. Inflammation and pregnancy. Reproductive Sciences 2009 16 206–215. (https://doi.org/10.1177/1933719108329095)
- 38↑
Imseis HM, Zimmerman PD, Samuels P & Kniss DA. Tumour necrosis factor-alpha induces cyclo-oxygenase-2 gene expression in first trimester trophoblasts: suppression by glucocorticoids and NSAIDs. Placenta 1997 18 521–526. (https://doi.org/10.1016/0143-4004(77)90005-4)
- 39↑
Biondi C, Ferretti ME, Pavan B, Lunghi L, Gravina B, Nicoloso MS, Vesce F, Baldassarre G. Prostaglandin E2 inhibits proliferation and migration of HTR-8/SVneo cells, a human trophoblast-derived cell line. Placenta 2006 27 592–601. (https://doi.org/10.1016/j.placenta.2005.07.009)
- 40↑
Horita H, Kuroda E, Hachisuga T, Kashimura M & Yamashita U. Induction of prostaglandin E2 production by leukemia inhibitory factor promotes migration of first trimester extravillous trophoblast cell line, HTR-8/SVneo. Human Reproduction 2007 22 1801–1809. (https://doi.org/10.1093/humrep/dem125)
- 41↑
al-Sebai MA, Diver M & Hipkin LJ. The role of a single free beta-human chorionic gonadotrophin measurement in the diagnosis of early pregnancy failure and the prognosis of fetal viability. Human Reproduction 1996 11 881–888. (https://doi.org/10.1093/oxfordjournals.humrep.a019271)
- 42↑
al-Sebai MA, Kingsland CR, Diver M, Hipkin L & McFadyen IR. The role of a single progesterone measurement in the diagnosis of early pregnancy failure and the prognosis of fetal viability. British Journal of Obstetrics and Gynaecology 1995 102 364–369. (https://doi.org/10.1111/j.1471-0528.1995.tb11286.x)
- 43↑
Jones JR & Perez RJ. Intravaginal PGE2 in early abortion. Prostaglandins 1974 7 149–163. (https://doi.org/10.1016/0090-6980(74)90135-X)
- 44↑
Kotani M, Tanaka I, Ogawa Y, Usui T, Tamura N, Mori K, Narumiya S, Yoshimi T & Nakao K. Structural organization of the human prostaglandin EP3 receptor subtype gene (PTGER3). Genomics 1997 40 425–434. (https://doi.org/10.1006/geno.1996.4585)
- 45↑
Kumar P & Magon N. Hormones in pregnancy. Nigerian Medical Journal 2012 53 179–183. (https://doi.org/10.4103/0300-1652.107549)
- 46↑
Zhu J, Mayr D, Kuhn C, Mahner S, Jeschke U & von Schonfeldt V. Prostaglandin E2 receptor EP1 in healthy and diseased human endometrium. Histochemistry and Cell Biology 2018 149 153–160. (https://doi.org/10.1007/s00418-017-1616-y)
- 47↑
Stocco DM, Wang X, Jo Y & Manna PR. Multiple signaling pathways regulating steroidogenesis and steroidogenic acute regulatory protein expression: more complicated than we thought. Molecular Endocrinology 2005 19 2647–2659. (https://doi.org/10.1210/me.2004-0532)
- 48↑
Birnbaumer L. Expansion of signal transduction by G proteins. The second 15 years or so: from 3 to 16 alpha subunits plus betagamma dimers. Biochimica et Biophysica Acta 2007 1768 772–793. (https://doi.org/10.1016/j.bbamem.2006.12.002)
- 49↑
Knofler M, Saleh L, Strohmer H, Husslein P, Wolschek MF. Cyclic AMP- and differentiation-dependent regulation of the proximal alphaHCG gene promoter in term villous trophoblasts. Molecular Human Reproduction 1999 5 573–580. (https://doi.org/10.1093/molehr/5.6.573)
- 50↑
Gomez-Chang E, Espinosa-Garcia MT, Olvera-Sanchez S, Flores-Herrera O & Martinez F. PKA tightly bound to human placental mitochondria participates in steroidogenesis and is not modified by cAMP. Placenta 2014 35 748–762. (https://doi.org/10.1016/j.placenta.2014.06.005)
- 51↑
Hellgren M. Hemostasis during normal pregnancy and puerperium. Seminars in Thrombosis and Hemostasis 2003 29 125–130. (https://doi.org/10.1055/s-2003-38897)
- 52↑
Samarakoon R, Dobberfuhl AD, Cooley C, Overstreet JM, Patel S, Goldschmeding R, Meldrum KK, Higgins PJ. Induction of renal fibrotic genes by TGF-beta1 requires EGFR activation, p53 and reactive oxygen species. Cell Signaling 2013 25 2198–2209. (https://doi.org/10.1016/j.cellsig.2013.07.007)
- 53↑
Shang W, Shu MM, Liu M, Wang AM, Lv LB, Zhao Y, Li M, Yan L. Elevated expressions of p53, CDKNA1, and Bax in placental villi from patients with recurrent spontaneous abortion. European Review for Medical and Pharmacological Sciences 2013 17 3376–3380.
