Prostaglandin E2 Receptor 4 (EP4) Affects Trophoblast Functions via Activating the cAMP-PKA-pCREB Signaling Pathway at the Maternal-Fetal Interface in Unexplained Recurrent Miscarriage

Implantation consists of a complex process based on coordinated crosstalk between the endometrium and trophoblast. Furthermore, it is known that the microenvironment of this fetal–maternal interface plays an important role in the development of extravillous trophoblast cells. This is mainly due to the fact that tissues mediate embryonic signaling biologicals, among other molecules, prostaglandins. Prostaglandins influence tissue through several cell processes including differentiation, proliferation, and promotion of maternal immune tolerance. The aim of this study is to investigate the potential pathological mechanism of the prostaglandin E2 receptor 4 (EP4) in modulating extravillous trophoblast cells (EVTs) in unexplained recurrent marriage (uRM). Our results indicated that the expression of EP4 in EVTs was decreased in women experiencing uRM. Furthermore, silencing of EP4 showed an inhibition of the proliferation and induced apoptosis in vitro. In addition, our results demonstrated reductions in β- human chorionic gonadotropin (hCG), progesterone, and interleukin (IL)-6, which is likely a result from the activation of the cyclic adenosine monophosphate (cAMP)- cAMP-dependent protein kinase A (PKA)-phosphorylating CREB (pCREB) pathway. Our data might provide insight into the mechanisms of EP4 linked to trophoblast function. These findings help build a more comprehensive understanding of the effects of EP4 on the trophoblast at the fetal–maternal interface in the first trimester of pregnancy.


Introduction
The European Society of Human Reproduction and Embryology (ESHRE) guidelines define the concept of recurrent miscarriage (RM) as two or more consecutive failed clin-ical pregnancies confirmed by histopathology or ultrasound before the first 20 weeks of gestation [1]. The classification of RM etiology consists of the following factors: inherited/genetic factors, endocrine, immunological, antiphospholipid syndrome (APS) and environmental factors [1]. RM represents the most challenging and complex area of reproductive research. However, the reasons for approximately 50% of RM cases remain idiopathic, known as unexplained uRM [1].
Trophoblast cells represent the primary cells which play an essential role in implantation and human placental formation, and can compromise embryonic growth and development [2]. Considering the complex situation of the placenta in fetal-maternal communication, it is easy to understand its diverse function at different stages [3]. Within the first weeks of pregnancy, the human placenta generates trophoblasts with diverse biological functions including attachment of the conceptus to the uterine wall, establishment of early nutrition and adaption of the maternal uterine vasculature [2,4]. Trophoblast derived from the trophectoderm (cytotrophoblasts) are progenitors and differentiate into extravillous trophoblasts (EVTs) and syncytiotrophoblasts (STBs), from which multiple functions develop [5]. Current research sustains that disfunction of the extravillous trophoblasts (EVTs) might lead to early and late uRM [6].
Despite the impressive amount of research investigating the role of PGE2 in the female reproductive system, there are still intriguing questions which have not been answered. Our group conducted some primary research to examine the level of PGE2 receptors (EP1, EP2, EP3, and EP4) within the maternal-fetal interface in uRM patients and normal pregnancy. Our previous studies demonstrated that EP3 expression was increased in STBs, which led to upregulation of the inflammatory microenvironment, extracellular matrix remodeling, and hormone production within the fetal-maternal interface of uRM patients [14]. Moreover, it was found that EP2 and EP4 expression was much lower at both syncytium and decidual level in samples obtained from patients who experienced a uRM [15]. In addition, the same work demonstrated that EP2 inhibition significantly impacted the regulation of inflammatory cytokines, proliferation, and secretion of hormone production in HTR-8/SVneo. Furthermore, other research indicated PPARγ decreased within the trophoblasts, and PPARγ is involved in M2 polarization in decidual macrophages in uRM, thus suggesting that downregulation of PPARγ modulated the microenvironment at the maternal-fetal interface in recurrent miscarriage [16]. Our group's previous work found that the expression of EP4 was decreased in the syncytiotrophoblasts (STBs) within the first trimester of pregnancy in the uRM group compared to healthy controls. Nonetheless, its expression and potential function in regulating EVTs remains relevant for future studies, and the its molecular pathological mechanism in trophoblast function is still unknown.
The main aim of our study is to identify the molecular mechanism involved in EP4 signaling and its impact on trophoblast phenotype. Additionally, we will examine the potential effects of EP4 modulation of trophoblast cell functions like secretion, invasion, apoptosis, proliferation, and viability. Finally, we will analyze the expression of the expression of EP4 in the EVTs of the decidua in uRM and healthy pregnancy. We summarize the novelty of this study as follows: (1) as far as we know, this is the first study to investigate the staining of EP4 in the maternal-fetal interface in uRM and healthy pregnancy; (2) we proposed an effective way to identify the functions of EP4 in trophoblast cells in vitro; (3) we found the molecular mechanism of EP4 which modulates the functions of trophoblasts and established the role of effective gene markers for precision treatment for uRM.

Demographic and Clinical Presentation of the Patients
The demographic and clinical data of 38 patients enrolled in our study are illustrated in Table 1 We first assessed PGE2 receptor (EP1, EP2, EP3, and EP4) expression across our two groups of patients via immunohistochemistry [14,15]. The staining of the EP4 was primarily distributed through the cytoplasm of the chorionic villous tissue and decidual cells. Interestingly, our analysis showed a significant decrease in EP4 in the syncytium and decidua of the uRM group compared to the healthy control group.
To further describe these findings, double immunofluorescence staining for human leukocyte antigen G (HLA-G) and EP4 was performed. The analysis showed that EP4 was co-expressed with HLA-G, mainly located in the cytoplasm of EVTs. In addition, intense staining for EP4 was observed in invasive EVTs in the HC group's maternal decidua compared with that of the uRM group ( Figure 1B,E). Subsequently, the percentage of HLA-G+ EP4+/HLA-G+ cells was calculated with nine slides from the uRM group and healthy controls, respectively. The rate of doubled stained cells significantly declined in the uRM group than the HC group (80.11 ± 5.21% vs. 73.53% ± 2.97%, p = 0.014, Figure 1G). These results suggest that the downregulation of EP4 expression in EVTs might affect the invasion of EVTs into a maternal decidual layer in patients with uRM. Furthermore, this decrease might affect trophoblast functions such as proliferation, apoptosis, invasion, and secretion as well. Figure 1. Double immunofluorescence analyses of EP4 and HLAG shown in the decidua of uRM patients and healthy controls. EP4 is co-expressed with HLAG in the cytoplasm of extravillous trophoblast cells (A,B,C,D,E,F). The percentage of double staining cells were counted in HLAG + EP4 + /HLAG + cells over nine slides. The expression of EP4 in EVTs was decreased in uRM patients compared to healthy controls (G). HLAG is specific marker for extravillous trophoblast cells (EVTs). The picture is shown in its original magnification of 40×. * p < 0.05.

Phosphorylation CREB (pCREB) is Involved in the Pathogenesis of uRM
The Genecards website (https://www.genecards.org/cgibin/carddisp.pl?gene=PTGER4#transcripts, accessed on 1 July 2021) was used to predict a transcription factor that could bind with the EP4 promoter. Our search identified several Figure 1. Double immunofluorescence analyses of EP4 and HLAG shown in the decidua of uRM patients and healthy controls. EP4 is co-expressed with HLAG in the cytoplasm of extravillous trophoblast cells (A-F). The percentage of double staining cells were counted in HLAG + EP4 + /HLAG + cells over nine slides. The expression of EP4 in EVTs was decreased in uRM patients compared to healthy controls (G). HLAG is specific marker for extravillous trophoblast cells (EVTs). The picture is shown in its original magnification of 40×. * p < 0.05.

Phosphorylation CREB (pCREB) Is Involved in the Pathogenesis of uRM
The Genecards website (https://www.genecards.org/cgi-bin/carddisp.pl?gene=PTGER4 #transcripts, accessed on 1 July 2021) was used to predict a transcription factor that could bind with the EP4 promoter. Our search identified several binding sites of the PTGER4 gene promoter: AP-2α, C/EBPα, CREB, PPAR-γ1, and PPAR-γ2. Among those, we selected PPAR-γ, CREB, and pCREB and performed immunohistochemical analysis on samples obtained from first trimester human placentas.
Triple immunofluorescence staining was applied to co-localize EP4 and pCREB expression in decidual EVTs in the first trimester placentas. EP4 was present in the cytoplasm (P, E), while pCREB was positive at both the nuclear and cytoplasm levels (Q, V) in EVTs (O, T) in the decidua of the first trimester placenta in healthy controls and uRM. Both EP4 and pCREB were co-expressed with HLA-G, especially in the healthy group compared to uRM patients (S, X). Together, our results showed that EP4 is downregulated in the EVT of uRM patients, and its expression is positively correlated with pCREB. Figure 2. Immunohistochemical analysis of pCREB and CREB expression in placentae of uRM patients and healthy controls from the first trimester were assessed via IRS score. Expression of pCREB and CREB was identified in the nuclear of cells in the syncytium (A,B,G,H) and in decidual (C,D,I,J) of first-trimester placentas in both healthy control and uRM. The expression of pCREB is decreased in both the syncytium and decidual cells (E,F) were measured via IRS score. EP4 is significantly positively correlated with pCREB in both the syncytium and decidua (M,N). Confocal microscopy imaging showed that EP4 (red) (P, E) co-expresses and co-localizes with p-CREB (green) (Q,V) in extravillous trophoblasts (EVTs) (O,T) of uRM patients (S) and health controls (X). HLAG is specific marker for EVTs. The picture was shown in its original magnification of 40×.

EP4 Downregulation Inhibits Proliferation and Induces Cell Apoptosis
We further investigated the effect of EP4 knockdown on the viability and proliferation of trophoblast cells in vitro. First, we assessed EP4 expression levels in three trophoblast cell lines, HTR-8/SVneo, JEG-3, and BeWo, by Western blot. EP4 protein expression was higher in HTR-8 and JEG-3 than BeWo ( Figure 3A). We subsequently transfected both HTR-8/SVneo and JEG-3 cells with a specific small interfering RNA (siEP4). The efficiency of EP4 siRNA was confirmed by qRT-PCR which showed that the EP4 mRNA level was decreased by 71.7% in HTR-8/SVneo cells with EP4si RNA3 and by 73.4% in JEG-3 cells with EP4si RNA1 compared to the negative control, respectively ( Figure 3B,C). (E) The MTT assay indicated that downregulated EP4 decreased the viability of JEG-3 compared to the negative control. (F) BrdU assay suggested that the proliferation of HTR-8/SVneo is inhibited by EP4 knockdown compared to the negative control. (G) BrdU assay suggested that the proliferation of JGE-3 is inhibited by EP4 knockdown compared to negative control. (H) The apoptosis test showed that the apoptosis of HTR-8/SVneo with EP4 siRNA is improved compared to the negative control and the effect was reversed by 8-Bromo-cAMP. (I) The apoptosis test showed that the apoptosis of JGE-3 with EP4 siRNA is reduced compared to the negative control. * p < 0.05.

EP4 Effects Secretion of Hormones and Cytokines
Human chorionic gonadotropin and progesterone are two essential hormones in early pregnancy secreted from trophoblasts. We measured the secretion levels of these two crucial hormones after culture with different concentration of PGE2, TCS 2510 (a selective EP4 agonist), and L-161,982 (a selective EP4 antagonist) in supernatants of HTR-8/SVneo and JEG-3 cells.

EP4 Modulates the Axis of cAMP-PKA-pCREB
We found that knockdown of EP4 inhibited the production of β-hCG in the supernatants of both HTR-8/SVneo and JEG-3 cells. Phosphorylated CREB and ERK1/2 are known as the downstream transcription factors of the cAMP and PKA pathways, mediating the transcription of β-hCG and progesterone.

Discussion
The main finding of our study is that the expression of EP4 is downregulated in the EVTs of uRM groups and the expression of EP4 has a positive correlation with p-CREB. Interestingly, EP4 demonstrated effects on proliferation and promoted the apoptosis of trophoblast cells in vitro. Finally, downregulation of EP4 attenuated the secretion of β-hCG, progesterone, and IL-6, and activates cAMP-PKA-pCREB signaling pathway.
In our project, we explored the molecular mechanisms behind the regulation of the PGE2 receptor EP4 in trophoblasts in first trimester placentas. Liu et al. suggested that PGE2 modulates biochemical and morphological placental trophoblast differentiation during implantation and placentation [17]. Previous research demonstrated that placental hypoxia activated cAMP production, proliferation, and invasion by mediating cAMP-PKA-pCREB signaling in trophoblast cells, predominantly through EP1 and EP4 [18,19]. Our findings are consistent with previous studies that showed that knockdown of EP4 decreased the proliferation and viability of HTR-8/SVneo and JEG-3 cells in vitro by inhibiting cAMP levels. However, Biondi et al. reported that PGE2 suppressed proliferation and migration through EP2 and EP4 by increasing intracellular cAMP in HTR-8/SVneo cells. We consider that the actual role of PGE2 either stimulates or inhibits trophoblast cell proliferation depending on the concentration of PGE2 and EP receptor subtypes activated in the first trimester of EVTs. In the adenylate cyclase pathway, increased cAMP levels resulted in PKA activation and a transcriptional factor that binds to CREB transactivates the transcription of specific primary response genes that initiate cell proliferation [20,21]. PGE2 has been shown to increase colon cancer cell proliferation and motility by activating the PI3K/Akt pathway by EP4 receptor activation [22].
We found that EP4 downregulation promotes apoptosis, and PKA agonists can reverse this effect. Consistent with these data, previous studies reported that EP4 receptor blockade hindered proliferation and induced apoptosis in cervical cancer cells via the EP4-PKA-CREB pathway. Moreover, reduced PKA induced apoptosis through Bcl-2/Bax-induced apoptosis [23] and activation of the caspase-8/9/11 pathway and pyroptosis [24,25]. The PKA inhibitor blocked PGE2 through EP2-and EP4-induced apoptosis, indicating that the PKA pathway is mainly responsible for PGE2-mediated inhibition of apoptosis [26].
At the initial stages in the first trimester of pregnancy, invasive extravillous trophoblasts (iEVTs) secrete β-hCG, which stimulates trophoblast invasion, angiogenesis, and placenta immune tolerance [27]. hCG might represent a serum marker of implantation and early trophoblast invasion. Progesterone production by EVTs is crucial to fine-tune vascular remodeling and escape immune attack in the maternal decidua [28]. In primary human trophoblasts, human trophoblast derived choriocarcinoma cell line BeWo, demonstrated that increasing intracellular cAMP levels activates PKA and exchanges protein directly activated by cAMP (EPAC), both of which induce the secretion of hCG and progesterone, resulting in trophoblast syncytialization and cell fusion [29][30][31]. cAMP-dependent protein kinase A (PKA) is associated with activation of the transcription factor cAMP response element-binding protein (CREB) [32]. An essential step for CREB activation and dimerization is phosphorylation of the serine residue at position 133. It recruits the adaptor proteins CBP and p300, which act as co-activates for transcription of αHCG in villous trophoblasts [33].
Furthermore, PKA can also inhibit AMP-activated protein kinase (AMPK) activity, which is responsible for the inactivation of hormone-sensitive lipase, thereby reducing the ability of LH or PGE2 to provide cholesterol for progesterone synthesis [34]. Waclawik et al. reported that estradiol-17β and progesterone increase the production of PGE2 in response to luteinizing hormone (LH) in endometrial stromal cells [35]. These findings suggest that PGE2 may be involved in a positive feedback loop during the progesterone synthesis in cells. Moreover, studies demonstrated that PGE2 improved endometrial receptivity via EP2 and EP4, acts through autocrine or paracrine pathways in vivo, and is regulated by progesterone secretion through the cAMP/PKA pathway [36][37][38]. β-hCG promotes angiogenesis in uterine vasculature, promotes the fusion of cytotrophoblast cells and differentiation to make syncytiotrophoblast cells, and causes the blockage of any immune or macrophage action by the mother to foreign invading placental cells [39]. Generally, β-hCG and progesterone also promoted maternal immune tolerance to control EVT functions and maintain embryonic-maternal communication. According to the above results, we suggest similarity and synergetic effects among PGE2 and β-hCG and progesterone pathways in HTR-8/SVneo and JEG-3 cells, and they trigger the process via activating adenylate cyclase to increase cAMP concentrations. PGE2, through its nuclear receptor PPARγ, effects trophoblast differentiation to activate downstream targets genes, and this regulation process is mainly mediated through chorion-specific transcription factor-1 (GCM-1) and increased expression of β-hCG [40,41].
It is reported that PGE2 induces cytokines and chemokine expression for trophoblast apposition and adhesion to the decidua for implantation [42,43]. IL-6 is a cytokine known to promote the differentiation of Th2 cells and subsequent suppression of Th1 cells [44]. The Th2-derived cytokines, IL-6, induce hCG release from trophoblasts, and hCG stimulates progesterone production from the corpus luteum in pregnancy [45].
There is evidence that PGE2 induces IL-6 production mediated by intracellular cAMP and PKA-pCREB in orbital fibroblasts [49]. Additionally, Hershko et al. reported that the transcriptional regulation of the IL-6 gene is complex, and CREB is involved [50]. IL-6 stimulates primary trophoblast invasion and migration, utilizing the JAK/STAT and AKT signaling pathway [51,52]. The study illustrated trophoblasts secreted IL-6 to facilitate M2 polarization of macrophages by binding to the IL-6 receptor (IL-6R) on macrophages through the STAT3 signaling pathway. In that way, the activated macrophages enhance the migration and invasion of trophoblast cells in a feedback manner at the maternalfetal interface in normal pregnancy [53]. Our data agree with these studies, which also demonstrated that the decline of EP4 decreased secretion of IL-6 and activated the cAMP-PKA-pCREB signaling pathway in HTR-8/SVneo and JGE-3 cells in vitro.
There are still some limitations in our manuscript. The main limitations of cell culture cancer cell lines are phenotypic and molecular mechanisms research. We did not use primary trophoblast cells, but choriocarcinoma tumor cells. The lack of an animal model which is closer to the microenvironment of the maternal-fetal interface is also a shortage of this manuscript. How PGE2 exerts its effects seems highly dependent on its concentration levels and which G protein receptors are activated in trophoblast cells. Previous studies from our group demonstrated that EP2 regulated the proliferation, hormone production and secretion of cytokines in trophoblast cells [15]. EP3 signaling plays a vital role in the regulation of the inflammatory microenvironment, hormone production and extracellular matrix remodeling in the maternal-fetal interface of uRM patients [14]. We deem that more in-depth research is necessary to obtain a more accurate and comprehensive understanding of the effects of PGE2 on trophoblasts.

Ethics Approval and Sample Tissues
The Medical Faculty's registered study got recognition from the ethics board under the Ludwig Maximilian University of Munich (Approval Number: 337-06, 29 December 2006). Clinical sample data collected was utilized for statistical purposes once informed consent was obtained. Tissue samples were taken from 19 participating patients with a history of two or more successive miscarriages due to unexplained factors (uRM group) and 19 healthy patients who had a lawful pregnancy termination (healthy control group) who had surgery in Munich, Germany. Exclusion criteria for the uRM group included infectious diseases, uterine anomalies, endocrinological dysfunctions, hyperprolactinemia, hyperandrogenemia, thyroidal dysfunctions, autoimmunologic disorders, deficiencies in coagulation factors, as well as fetal and parental chromosomal disorders.

Immunohistochemistry
Our team has previously described immunohistochemical (IHC) staining [15]. Baked placentas were deparaffinized in xylol for 20 min, then incubated in methanol with H 2 O 2 for another 20 min to block endogenous peroxidase reaction before being rehydrated in an ethanol gradient. A pressure cooker saturated with sodium citrate (pH = 6.0) was used for storing the slides. Elimination of unspecified indissoluble elements of the main antibodies was carried out, with all slides going through a blocking treatment residue (Reagent 1, Zytochem-Plus HRP-Polymer-Kit (mouse/rabbit), Berlin, Germany) about twenty minutes after being washed in PBS. The main antibody was stored overnight at 4 • C on each slide. Table 1 lists all antibodies utilized. Following the manufacturer's guidelines, the reagent from a detection kit, the Avidin-Biotin Complex (ABC) kit (Vector Lab; Burlingame, CA, USA), was used to measure reactivity after a 1 h immersion. Immunostaining was visualized with the substrate and the chromogen-3, 3 -diaminobenzidine (DAB; Dako; Santa Clara, CA, USA) for chromogen for 1 min and hematoxylin for nuclear staining for 2 min. Distilled water was used to terminate the reaction. A semi-quantitative approach was used for the immunoreactive score (IRS) procedure; all slides were read for interpretation of saturation and allocation using a Leitz (Wetzlar, Germany) microscope. The IRS computation was calculated through addition of the percentage of positively stained cells (zero: zero pigmentation; one: mild pigmentation; two: intermediate pigmentation; three: ranges higher than 50 percent but less than 80 percent pigmentation; four: higher than eighty percent pigmentation by the saturation of cell pigmentation (0: none; 1: mild pigmentation; 2: medium pigmentation; 3: high pigmentation)).

Immunofluorescence Staining
The placenta specimens were filtered in xylol for twenty minutes before rinsing in 100%, 70%, 50%, and finally distilled water. A sodium citrate neutralizer (pH = 6.0) was present in the pressurized pot. Slides were washed in PBS and probed for 15 min through an agent (Ultra V Block, Lab Vision, Fremont, CA, USA). The positive control was incubated for 4 h at 4 • C overnight. Table 2 lists all antibodies utilized. After rinsing, the slides were stored in darkness for 30 min with secondary antibodies at ambient temperature. The nucleus was stained with DAPI after the slides had been thoroughly rinsed. An Axioskop fluorescent photomicroscope (Zeiss; Oberkochen, Germany) was utilized for double immunofluorescence investigation of the extravillous trophoblasts of the placenta, and pictures were captured with the Axiocam camera system (Zeiss CF20DXC). Similar paraffinembedded slides were used together with a triple for immunofluorescence staining of EP4 and pCREB co-expression in the extravillous trophoblasts of the placenta. Microscopic confocally laser scanned photos were captured using a Zeiss LSM 880 connected to an Airyscan quality sample imaging with analysis using ZEN blue software.

Treatment with EP4 siRNAs
The method was previously described in full [54]. Cells were cultured in 6-well or 96-well saturations in RPMI 1640 solution with 10% FBS, growth averaging 60% confluence, and EP4 and controlling siRNAs were retrieved at a concentration of 20 nM through OriGene setup (CAT: SR321501, Rehovot, HaMerkaz, Israel), transferred utilizing Lipofectamine 3000 connected at a final concentration of 50 nM with producer protocols (Invitrogen, Waltham, MA, USA). For all subsequent tests, cells were immersed in new solution with or without solamargine for up to another 24 h after being cultured, adding up to 30 h.

Real-Time Quantitative Extraction
Using a RNeasy Mini Kit, total RNA was extracted from cells (Qiagen, Hilden, Germany). RNA was transformed into a first-strand cDNA manufacturing kit following the producer's protocol (Epicenter, Madison, WI, USA). With FastStart Essential DNA Probes Master and gene-specific primers (Applied Biosystems, Hs00259261 CE for EP4), the expression of EP4 mRNA was measured using RT-PCR. Next, 1 µL TaqMan ® Gene Expression Assay (Thermo Fisher Scientific, Waltham, MA, USA) 20×, 1µL cDNA, 8 µL H2O (DEPCtreated DI water), and 10 µL TaqMan ® Fast Universal PCR Master Mix 2× were combined in a 20 µL reaction mixture. Then, a 20 L TaqMan ® Gene Expression Assay and TaqMan ® Fast Universal PCR Master Mix experiment was conducted on an Optical Fast 96-well plate with an optical sticky film over it, 2 µL cDNA template and 8 µL RNase-free water were created in each probe. The temperature regimen was 95 • C for 20 s, 95 • C for 3 s, and 60 • C for 30 s, accompanied by 40 cycles of intensification. The housekeeping genes β-actin (Nr. Hs99999903 mL) were employed as reference controls for expression normalization, and the comparative CT technique was employed for computation.

Western Blot Analysis
The cells were immersed in a 200 L catalyzing solution containing a 1:100 dilution of protease inhibitor (Sigma-Aldrich, Merck Darmstadt, Germany) in RIPA buffer for 30 min (Sigma Aldrich, R0278, Merck, Darmstadt, Germany). The lysate was centrifuged, and the supernatant was subjected to a Bradford protein test. Protein separation occurred at 10% SDS-PAGE, with transference into nitrocellulose cells according to their molecular weights (Bio-Rad, Feldkirchen, Germany). Non-specific binding was prevented by incubating the membrane for 1 h with 4% skim milk powder before incubating for 16 h at 4 • C with the main antibodies (the antibodies utilized are listed in Table 1). The samples and secondary antibodies are stored for 1 h at ambient temperature after being washed three times for 10 min in TBS/Tween. Detection was performed with 5-bromo-4-chloro-3 -indolylphosphate/nitro-blue tetrazolium (BCIP/NBT) chromogen substrate solution (Promega, Walldorf, Germany). Western blots were scanned and quantified by densitometry, using the GelScan V6.0 1D Analysis Software (SERVA, Electrophoresis GmbH, Heidelberg, Germany).

Cell Viability Assays and Cell Proliferation
For 24 h, HTR-8/SVneo and JEG-3 cells were seeded in 96-well plates and transfected with EP4 siRNA or control siRNA. Cell credibility was assessed using the MTT (3-(4, 5-dimethylthiazol-2-yl) 2, 5-diphenyl tetrazolium bromide) test (Roche, Mannheim, Germany). Under cultivation conditions, the MTT-labeled reagent was applied for complete saturation of 0.5 mg/mL about 30 min. The optical density (OD) was measured using 595 nm through an Elx800 universal Microplate Reader. HTR-8/SNveo and JEG-3 microorganisms developed into 96-well sections for the proliferation test. The cells were then treated with 0.1, 1, 10 nM PGE2, TCS 2510, and L-161,982 the following day. As a blank control, DMSO was used. The manufacturer's protocol was followed to measure cell proliferation using the 5-Bromo-2 -Deoxyuridine (BrdU) incorporation assay (11647229001, Roche). An Elx800 universal Microplate Reader was used to test color intensity of 450 nm. Each experiment was carried out three times.

ELISA of cAMP Assay
The intracellular cAMP levels were determined using an enzyme-linked immunosorbent test kit (R&D system, KGE012B, Minneapolis, MN, USA). The color development was halted, and the absorbate were assessed at 450 nm. The cAMP concentration was calculated using ng/mL. The cAMP measurement for each stimulation section was adjusted to the control group cAMP measurements.

Cell Death Detection ELISA
Using the Cell Death Determination ELISA Kit (Roche, cat# 11544675001), apoptotic cell death was collected from HTR-8/SVneo and JEG-3 cells cultivated in triplicate with four treatment groups. Using the manufacturer's protocol, the kit quantifies cell death using a colorimetric test that measures the proportion of cytoplasmic histone-associated DNA fragments. An Elx800 universal Microplate Reader was utilized for measurement of absorbance at 460 mm.

β-hCG and Progesterone Measurement
Supernatants were collected from 24-well plates of HTR-8/SVneo, JEG-3, and BeWo cells that had been incubated in defined medium for 24 and 36 h, respectively, and refluxed at 10,000× g for 10 min for filtration of cell sections. β-hCG and progesterone were gauged using the recommended guidelines with an ADVIA Centaur XP auto analyzer (Siemens Medical Solution Diagnostics, Erlangen, Germany), described in previous studies [17].

Detection of Cytokines by ELISA
After 24 h of transfection with si-EP4 or si-Control, the cells centrifuged at 1000× g for 10 min. The concentrations of IL-6, IL-8, TNF-α, PAI-1, and IL-1 in culture solution were determined with ELISA kits (R&D Systems, Minneapolis, MN, USA) following the production protocol. The absorbance readings were measured using an Elx800 universal Microplate Reader.

Statistic Analysis
The terms used are the mean ± standard deviation (SD). To conduct analytical assessments, the GraphPad Prism program (version 8.0, San Diego, CA, USA) was utilized. The correlation between the two indicators was examined using Pearson's correlation assessment.
The Student's t-test was used for the comparison of maternal age and gestational age in uRM patients and healthy controls. A Mann-Whitney test was performed for gravidity and parity in two groups. A Mann-Whitney test was applied for evaluating the IRS scores of pCREB, CREB and PPARγ expression and the percentage of double staining cells in the placentas of the two groups. A Wilcoxon test was used for evaluation of the proliferation rate, apoptosis rate, β-hCG, progesterone, IL-6, IL-8, TNF-α, IL-1, cAMP and PAI-1 expression levels between vehicle and stimulation groups. A Wilcoxon test was also used for analyzing the band intensities of EP4, PKA, p-CREB, CREB and β-actin. At p < 0.05, differences were considered statistically significant (* p 0.05; ** p 0.01; *** p 0.005).

Conclusions
In conclusion, our study demonstrated that EP4 is decreased in EVTs in uRM patients. Interestingly, knockdown of EP4 inhibited proliferation, promoted apoptosis, and reduced the secretion of β-hCG, progesterone, and IL-6 in in vitro trophoblast cell culture models, which was associated with the cAMP-PKA-pCREB signaling pathway ( Figure 6). Our future in-depth mechanistic studies will address the issue of the role of PGE2 and its specific EP receptor in immune cells (macrophages, dendritic cells, regulatory T cells, and natural killer cells) and crosstalk between trophoblasts and immune cells in the decidua during the first trimester placentas of uRM patients. Inhibiting the EP4 signaling pathway reduces the activity of the cAMP-PKA-pCREB signaling pathway, which can eventually lead to depressing the production of progesterone, β-hCG and IL-6. Downregulated of EP4 decreased proliferation and increased apoptosis in trophoblasts. These abnormal changes in trophoblasts may contribute to recurrent miscarriages. Data Availability Statement: Data can be made available by the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare that the research was conduct in the absence of any potential conflict of interest.