Down‐regulation of endothelial protein C receptor promotes preeclampsia by affecting actin polymerization

Abstract Preeclampsia is a severe pregnancy‐related disease that is found in 3%–5% of pregnancies worldwide and is primarily related to the decreased proliferation and invasion of trophoblast cells and abnormal uterine spiral artery remodelling. However, studies on the pathogenesis of placental trophoblasts are insufficient, and the aetiology of PE remains unclear. Here, we report that endothelial protein C receptor (EPCR), a transmembrane glycoprotein, was down‐regulated in placentas from preeclamptic patients. Moreover, lack of EPCR significantly reduced the trophoblast cell proliferation, invasion and tube formation capabilities. Microscale thermophoresis analysis showed that EPCR directly bound to protease‐activated receptor 1 (PAR‐1), a G protein‐coupled receptor. This change resulted in a substantial reduction in active Rac1 and caused excessive actin rearrangement. Our findings reveal a previously unidentified role of EPCR in the regulation of trophoblast proliferation, invasion and tube formation through promotion of actin polymerization, which is required for normal placental development.

into EVTs are related to changes in cell behaviour and cytoskeletal organization, 9,10 suggesting critical roles for the integrity and appropriate remodelling of the cell cytoskeleton in trophoblast proliferation and invasion.
Endothelial protein C receptor (EPCR), a 46-kDa transmembrane glycoprotein, is highly expressed in trophoblasts compared to other parts of the maternal-foetal interface during pregnancy, 11 and EPCR expression is down-regulated in preeclamptic placentas compared with healthy placentas. 12 The sequence and structure of EPCR is homologous to that of proteins of the major histocompatibility class 1/cluster of differentiation 1 (CD1) family, 13 and this protein participates in many cytoprotective activities, reducing the damage caused by diverse injuries or diseases from sepsis to stroke. 7 EPCR was reported to participate in the activation of endogenous protease-activated receptor 1 (PAR-1), a G protein-coupled receptor that plays an important role in mediating anti-inflammatory and antiapoptotic activities and protects endothelial barrier functions through regulating cytoskeletal organization. 14 Recently, cancer research has shown that EPCR aberrations are involved in diverse tumour tissues, such as breast cancer tissues, 15 lung cancer tissues 16,17 and ovarian cancer tissues. 18 Overexpression of EPCR is associated with the promotion of tumour growth and infiltration. Many studies have demonstrated striking similarities in the molecular circuits between trophoblasts and cancer cells due to the properties of proliferation and invasion of these two cell types. 19,20 However, whether EPCR is associated with trophoblast function and PE occurrence and development remains unclear. Therefore, we hypothesized that in trophoblast cells, EPCR may participate in the development of PE by affecting cell proliferation and infiltration.
In the current study, we detected EPCR down-regulation in the placentas of patients with PE compared to those from healthy controls. EPCR knockout inhibited cell proliferation, invasion and tube formation. Microscale thermophoresis (MST) analysis showed that EPCR directly bound to PAR-1. The decrease in cleaved PAR-1 caused by EPCR depletion resulted in a substantial reduction in active Rac1, a key factor that regulates the polymerization and depolymerization of actin; these changes resulted in excessive rearrangement of the cytoskeleton. In brief, our study revealed that EPCR down-regulation resulted in decreased proliferation, invasion and tube formation via the rearrangement of actin through the PAR-1/Rac1 signalling pathway in placental trophoblasts in PE.

| Collection of placental tissues
The protocols for placental tissue collection were reported previously. 21 Placental tissues were obtained from PE patients (n = 15) and subjects with healthy pregnancies (n = 15) admitted to Shenzhen Hospital, Peking University. PE was diagnosed by a new onset of systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg after the 20th week of gestation in the presence of proteinuria without preexisting renal diseases or primary hypertension. Patients with other major pregnancy complications were excluded. First-trimester villous tissue (n = 5) was collected from women who underwent legal termination between 6 and 10 weeks of gestational age that was not for medical reasons; these subjects did not have a history of a spontaneous abortion or an ectopic pregnancy. All placental tissues were collected ac-

| Immunohistochemistry
The villous and placental tissues were washed with phosphatebuffered saline (PBS) and fixed overnight with 4% paraformaldehyde (PFA) at room temperature. Then, the samples were dehydrated and embedded in paraffin before they were sectioned into 4-mm-thick sections. For Immunohistochemistry (IHC), the sections were deparaffinized, rehydrated and then microwaved in 10 mmol/L citric sodium (pH 6.0) for 20 minutes to retrieve antigens and blocked with

| Generation of the EPCR knockout (KO) cell line with CRISPR/Cas9
Two days after lentiCRISPRv2 vector transfection, the cells were treated with 2 μg/mL puromycin (Sigma, St. Louis, MO, USA) for three days, and a single cell was isolated. After two weeks, singlecell colonies were picked, and the EPCR knockout was confirmed by IF staining and WB analysis.

| Sequencing
Regions surrounding sgRNA target/off-target sites within the EPCR were amplified by PCR and analysed by BGI (Shenzhen, China) using Sanger sequencing (ACGT, Inc)

| Three-dimensional (3D) modelling
Actin was imaged as a z-series at 1 µm intervals to capture the entire actin structure via 3D confocal z-stacks. The CZI files were imported, and actin status was analysed using the Zen program software.

| Colony formation assay
Cells were seeded in 10 mm plates at a density of 500 cells/plate.
After incubation for 10 days at 37°C, the cells were fixed in 4% PFA for 30 minutes and stained with Giemsa stain solution for 20 minutes.
The number of colonies consisting of more than 50 cells was counted.

| Real-time cell proliferation and invasion assays
Cell proliferation and invasion abilities were examined using a real-time cell analysis (RTCA) system (RTCA DP Instrument; ACEA Biosciences, Inc, USA). For continuous monitoring of cell proliferation, the cells were seeded into RTCA E-plates at a density of 5 × 10 3 cells per well, and then, the electrical impedance in each well was measured continuously for 50 hours. The shift in electrical impedance is expressed as the cell index, which is a parameter of cell proliferation. The invasion experiments were performed in CIM-16 plates, and a layer of Matrigel (BD Biosciences) was added to the upper chamber of the plate for an hour. Next, 10 µg/mL of mitomycin was used to inhibit cell proliferation for 3 hours, and then, 3 × 10 4 cells were seeded and monitored for 80 hours. The sensor impedance following cell invasion was defined as the cell index.

| MST assay
MST was performed according to a previously described protocol. 22

| Cell-based enzyme-linked immunosorbent assay
In total, 2 × 10 4 cells were added to each well of a 96-well culture plate and incubated overnight. The medium was then removed, and the cells were fixed with 4% PFA and blocked with 1% BSA. The fixed cells were incubated with an anti-ATAP2 antibody (

| Rac1 activation measurement
Cells were grown to confluence in 6-well culture plates and starved overnight in serum-free medium. Analysis of Rac1 activation (Rac1-GTP) was performed using commercially available kits (Cytoskeleton, Inc). Briefly, appropriate cellular lysates were incubated with glutathione S-transferase-PAK PBD (Rac1 effector protein, p21-activated kinase 1) beads that bind specifically to the active GTP-bound form of Rac1 at 4°C with rotation for 1 hours.
The beads were washed and resuspended in loading buffer. Total and activated Rac1 levels were detected by WB analysis using an anti-Rac1 monoclonal antibody provided in the kit as described by the manufacturer.

| F-actin/G-actin ratio measurement
The F-actin/G-actin ratio was determined using commercially avail-

| Trophoblastic EPCR is down-regulated in preeclamptic placentas
We first used IHC to evaluate the expression levels of EPCR in firsttrimester human placentas. High expression of EPCR was detected in different subtypes of trophoblasts, including villous cytotrophoblasts (CTB), cell column trophoblasts (CCT) and interstitial EVT (iEVT) cells in the decidua ( Figure 1A). To further compare the differences in EPCR expression between preeclamptic placentas and placentas from healthy age-matched subjects, we performed a WB assay to detect the levels of EPCR in both groups. As shown in Table 1, there were no significant differences in maternal age, body mass index, gestational age at delivery or infant birth weight between healthy pregnant (n = 15) and preeclamptic (n = 15) women. The WB results demonstrated that the EPCR level was significantly lower in preeclamptic placentas than in healthy placentas ( Figure 1B). The IHC results further revealed strong EPCR staining in the villous CTBs and iEVTs in the basal plate of healthy, preterm placentas, whereas EPCR was weakly expressed in preeclamptic placentas ( Figure 1C,D).
Collectively, these results showed that the down-regulation of EPCR in trophoblasts is associated with preeclamptic placentas.

| EPCR knockout inhibits trophoblast proliferation and invasion
To ascertain the causal relationship between trophoblastic EPCR deficiency and PE development, we investigated the role of EPCR the membrane and were detected as the cell index. 23 As shown in Figure 2F, the invasive ability of cells in the EPCR-KO group was much lower than that in the control group. Thus, these data indicate that EPCR is essential for the proliferation and invasion of trophoblasts.

| EPCR knockout inhibits trophoblast tube formation
Failure of spiral artery remodelling can result in placental ischaemia and hypoxia and further leads to the occurrence and development of PE. 24 When the HTR8/SVneo cell line is cultured on Matrigel (with or without endothelial cells), it can spontaneously form endothelial-like tubes, which is believed to reflect the placental angiogenic ability. 25,26 To determine whether EPCR expression influences the angiogenic properties of trophoblasts, we performed tube formation assays. As shown in Figure 3A, the EPCR-KO cells exhibited decreased tube formation compared with the control cells. The quantitative analysis results demonstrated that the number of nodes and meshes, mesh area and total segment length was significantly reduced ( Figure 3B-E). These data demonstrated that EPCR knockout decreases the angiogenic properties of trophoblasts.

| EPCR regulates actin polymerization
Actin is crucial in determining cell proliferation and invasion, mainly by F-actin polymerization and rearrangement of the actin cytoskeleton. 27 Recent studies suggest that PAR-1 activation is regulated by EPCR, 28 which mediates the polymerization of F-actin and cortical actin distribution. 29 Therefore, we next investigated whether EPCR knockout affects actin polymerization by activating PAR-1-mediated cell signalling.
First, we performed an MST assay to examine the relationship between EPCR and PAR-1, and we found that EPCR directly bound to PAR-1 ( Figure 4A). Subsequently, the activation of PAR-1 following EPCR knockout in cells was investigated by a cell-based enzyme-linked immunosorbent assay (ELISA). We used a PAR-1 antibody (ATAP2) that specifically binds to intact PAR-1 and a non-cleavage-sensitive antibody (WEDE15) that detects total PAR1 on the cell surface. 30 We found that the binding rate of the ATAP2 antibody was increased in the EPCR-KO group compared with the control group, whereas the total PAR-1 level did not change ( Figure 4C). This finding indicated that EPCR knockout inhibited EPCR-mediated PAR-1 cleavage.

Cortical actin distribution induced by PAR-1 is likely mediated
by activation of the small Rho family GTPase, Rac1, 31 which induces rapid actin polymerization in ruffles near the plasma membrane. 32,33 Next, we evaluated the rate of Rac1 activation in the two groups.
As shown in Figure 4C were present and randomly distributed. (Figure 4D,E). To support our morphological observations, we assessed the actin cytoskeleton reorganization of EPCR-KO cells by WB analysis, and the G/F-actin ratio was determined ( Figure 4F). The F-actin level in EPCR-KO cells was lower than that in the control group. These data suggest that EPCR knockout results in abnormal actin polymerization.

| PAR-1 activation rescues the abnormal actin organization, decreased proliferation and invasion and inhibited tube formation of EPCR-KO trophoblastic cells
To verify that the effect of EPCR is mediated via PAR-1/Rac1, we All experiments were performed at least three times. All data are represented as the mean ± SEM. *P < .05, **P < .01, ***P < .001 the effects on actin and cell function. We selected the concentration based on previous reports 34 and first verified the activation of PAR-1/Rac1. Using cell-based ELISAs and WB analysis, we found that the activated PAR-1 and Rac1 levels were increased after addition of TFLLR-NH2 ( Figure 5A,B).
Then, we explored the effects on actin morphology and cell behaviours. As shown in Figure 5C,D, the actin morphology was restored, and the inhibitory effects on proliferation and invasion of EPCR-KO cells were reversed ( Figure 5E,F). The tube formation ability was also increased ( Figure 5G-K).

| Exogenous EPCR rescues the abnormal actin organization, decreased proliferation and invasion and inhibited tube formation of EPCR-KO trophoblastic cells
To confirm the role of EPCR in actin organization, trophoblast proliferation and invasion and tube formation, we transfected an EPCR-overexpressing vector into EPCR-KO cells as a rescue experiment. The WB results confirmed the successful transfection, as the protein level of EPCR was restored in the EPCR-transfected group ( Figure 6A). After transfection, the active PAR-1 and Rac1 levels were altered ( Figure 6B and C), the actin morphology was nearly completely restored ( Figure 6D and E), and the F-actin level was increased to a level similar to that in the wild-type (WT) cells ( Figure 6F).

The inhibited proliferation and invasion of EPCR-KO cells was
rescued by exogenous EPCR (Figure 7A-C). Tube formation was also increased in the exogenous EPCR group ( Figure 7D). The data demonstrated that the number of nodes and meshes, mesh area and total segment length was significantly increased ( Figure 7E-H).
These results collectively indicate that exogenous EPCR expression can rescue the EPCR knockout phenotype, which further proves that EPCR plays an important role in trophoblastic proliferation, invasion and angiogenesis.

| D ISCUSS I ON
The development of PE can be divided into two stages: in the first stage, the reduced proliferation and shallow invasion of such as choriocarcinoma, 36,37 whereas shallow invasion is characteristic of PE. 35 Many studies have suggested that trophoblasts are very similar in nature to tumour cells primarily based on the proliferative and invasive properties of these two cell types, 19,20 and EPCR is highly expressed in various malignancies and is involved in tumour cell proliferation, invasion, metastasis and apoptosis. 38 proliferation and blood vessel development. 44 In this study, we  This may be due to excessive placental ECPR shedding or to other pathways leading to maternal vascular endothelial cell stimulation. Studies have also shown that the expression level of sEPCR is related to gene polymorphisms and specific site mutations. 53,54 In addition, we did not perform in vivo experiments in this study, and we should further validate our findings in vivo using animal models such as mice.
In summary, for the first time, we elucidated the function of EPCR in the proliferation, invasion and tube formation of human placental trophoblasts and suggested possible pathological mechanisms in PE.

ACK N OWLED G EM ENTS
The

CO N FLI C T O F I NTE R E S T
All authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.