Extracellular Vesicles Derived from Human Umbilical Cord Mesenchymal Stem Cells Promote Trophoblast Cell Proliferation and Migration by Targeting TFPI2 in Preeclampsia

Preeclampsia is a pregnancy disorder characterized by systemic organ damage and high blood pressure. It has been reported that microRNA-195 (miR-195) is associated with preeclampsia. In this study, we discovered the target of miR-195 in regulating human extravillous cytotrophoblast-derived transformed cell proliferation and migration. We analyzed the clinicopathological factors of preeclampsia and normal pregnancies. The messenger ribonucleic acid (mRNA) levels of miR-195 and tissue factor pathway inhibitor 2 (TFPI2) were measured in placental tissues derived from normal and preeclampsia patients by real-time polymerase chain reaction (PCR). Human umbilical cord mesenchymal stem cell (hUC-MSC)-derived extracellular vesicles were verified by western blot. HTR8-S/Vneo cell proliferation was evaluated by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, and cell migration rate was assessed by the transwell assay. Relative luciferase activities were measured in TFPI2 wild-type (WT) and mutant cells. miR-195 expression was negatively correlated with TFPI2 mRNA levels in preeclampsia patients. Extracellular vesicles derived from hUC-MSCs enhanced HTR8-S/Vneo cell proliferation and migration. In addition, miR-195 isolated from hUC-MSCs enhanced HTR8-S/Vneo cell proliferation and migration by targeting TFPI2. Our findings demonstrate that the upregulation of miR-195 in extracellular vesicles derived from hUC-MSCs promotes HTR8-S/Vneo cell proliferation and migration by targeting TFPI2.


Introduction
Preeclampsia is a disorder characterized by a significant amount of proteins in the urine and the occurrence of hypertension. It usually occurs in around 2%-8% of pregnancies [1][2][3] and is correlated with high morbidity and mortality during pregnancies, with over 50,000 deaths reported worldwide. The most high-risk factors for preeclampsia include diabetes mellitus, overweight, and vascular and connective tissue disorders such as antiphospholipid antibodies and systemic lupus erythematosus [4]. Currently, several therapies such as magnesium sulfate, aspirin, calcium, endothelin antagonists, antioxidants, and cell therapy are available for preeclampsia treatment [5][6][7][8]. However, the lack of biomarkers and clinical markers makes it difficult to classify the severity of preeclampsia.
Several studies have reported on the regulatory function of exosomes on trophoblast cell proliferation and migration. For example, exosomal encapsulation of microRNA (miR-125a-5p) was reported to affect HTR8-S/Vneo cell proliferation and migration by regulating vascular endothelial growth factor A [9]. Another study reported that mesenchymal stem cell (MSC)-derived exosomes boosted trophoblast cell proliferation, migration, and invasion by restricting serum glucocorticoid regulated kinase 1 [10]. Additionally, exosomes have been identified as novel biomarkers in multiple diseases [11]. For instance, exosomes derived from umbilical cord MSCs showed a protective effect against chemotherapy-induced apoptosis and ovarian granulosa cell stress [12].
Evidence has shown that miRNAs are abundant in exosomes and can be transported in cell plasma and delivered to recipient cells by exosomes [13]. Given the transport ability of exosomes and the role of miRNAs in various biological processes, exosomes are gaining increasing research attention as a novel biomarker. For instance, miR-195 has been reported to associate with preeclampsia, and overexpression of miR-195 in exosomes alleviates hypoxia-induced damage of trophoblast cells [14]. In this study, we hypothesized that human umbilical cord MSC (hUC-MSC)-derived extracellular vesicles (EVs) could transfer microRNAs and exhibit protective effects against preeclampsia.

hUC-MSCs
Isolation and Identification. The umbilical cords of healthy newborns were obtained and cut into small sections in DMEM medium containing 3% penicillinstreptomycin (Invitrogen). Briefly, tissue separation was performed at the ultraclean workbench. Neonatal umbilical cord tissue was obtained first, then cord blood was separated by squeezing the umbilical cord, and then, the umbilical cord tissue was rinsed in phosphate-buffered saline (PBS) containing 2% penicillin-streptomycin. The umbilical cord tissue after washing was cut into fragments about 1 cm long, and the umbilical vein and umbilical artery were carefully removed. The remaining umbilical tissue was then carefully placed in DMEM medium for further cutting to about 1 mm 3 fragments. The fragments were collected and placed in a 25 cm 2 culture flask with 4 mL medium. After 7 days of culture, there were spindle cells around the tissue blocks, namely hUC-MSCs, observed by optical microscope.

Extraction and Identification of EVs.
EVs were purified and identified as described previously [10]. hUC-MSCs were cultured without serum for 2 days, and the supernatant was collected and stored at 4°C for 1 week. The supernatant was first centrifuged at 300 × g for 10 min and then at 15,000 × g for 20 min. After passing through a 0.22 μm filter, the supernatant was centrifuged at 100,000 × g for 1 hr. The EV suspension was obtained by resuspending the precipitates in PBS. EVs were then identified using a transmission electron microscope. In brief, 15 μL of exosomal suspension was transferred to a copper mesh for a few minutes, and 20 μL of a phosphotungstic acid solution (30 g/L) was added. Samples were then dehydrated under an incandescent lamp for about 10-20 min. Finally, EVs were observed using a transmission electron microscope. The EV suspensions were analyzed using the NanoSight LM 10 instrument (NanoSight Ltd., Amesbury, UK) following the published method [19]. EV markers Alix, cluster of differentiation (CD9), CD81, and CD63 were detected by western blot.
The polymerase chain reaction (PCR) assay was performed using a real-time PCR machine (iQ5, BioRad).
2.8. Cell Viability Assay. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Sigma) was used to measure cell viability. HTR8-S/Vneo cells, TFPI2 knockdown, or overexpressed cell lines were seeded at 8 × 10 2 cells per well in a 96-well plate. After incubation for 2 to 3 days, 50 μg of MTT solution was added to each well and incubated with cells for 4 hr in an incubator. Then, 150 μL of dimethyl sulfoxide was added to each well. The absorbance was detected at 595 nm with a reference wavelength of 650 nm using an ELx-800 University Microplate Reader (BioTek, USA).
2.9. Transwell Assay. Transwell assay was used to measure the migration of HTR8-S/Vneo cells under different conditions. In brief, HTR8-S/Vneo cells, TFPI2 knockdown, or overexpressed cell lines (1 × 10 6 cells/mL) were plated into the Transwell chambers (BD Biosciences, USA) without serum. Meanwhile, the chambers were incubated with medium containing 10% fetal bovine serum for 1 day. After that, cells within the chamber were fixed with PFA, stained with a 0.5% crystal violet solution for half an hour, photographed, and quantified using a standard microplate reader.
2.11. Statistics. Data were represented as mean AE standard deviation (SD) and analyzed using a two-tailed Student's ttest or one-way analysis of variance (ANOVA) with a Tukey post hoc test. Experiments were independently repeated in triplicates. A p value < 0.05 was considered statistically significant.

Result
3.1. miR-195 Was Negatively Correlated with mRNA Level of TFPI2 in Preeclampsia Patients. We first analyzed the clinical characteristics of patients with preeclampsia and normal pregnancies. We found that diastolic blood pressure (DBP), systolic blood pressure (SBP), and proteinuria were significantly higher in preeclampsia patients compared to normal individuals (Table 1). Additionally, the body weight of infants born to preeclampsia patients was lower than that of infants born to the normal group (Table 1). To explore the relationship between TFPI2 and miR-195, we compared the messenger ribonucleic acid (mRNA) expression levels of TFPI2 and miR-195 in the preeclampsia and normal pregnancy cohorts. The results showed a significantly increased TFPI2 mRNA level and decreased expression of miR-195 in preeclampsia patients compared to normal samples (Figures 1(a) and 1(b)).

Stem Cells International
Notably, a significant negative correlation (p ¼ 0:0021) was observed between the mRNA level of TFPI2 and miR-195 in preeclampsia patients (Figure 1(c)). These results suggest that TFPI2 might be a potential target of miR-195.
3.2. Isolation of EVs from hUC-MSCs. We next isolated hUC-MSCs and EVs derived from hUC-MSCs. The size of the particles in the isolated EV mixture was analyzed through nanoparticle tracking analysis (Figure 2(a)). In addition, we detected the expression of EV markers. Consistently, high expression of CD63, CD81, and CD9 was observed in EVs but not in hUC-MSC cells (Figure 2(b)). In addition, Alix expression was abundant in EVs, while weak in hUC-MSC cells ( Figure S1). The above results indicate that EVs were successfully obtained from hUC-MSCs.

hUC-MSC-Derived EVs Enhanced HTR8-S/Vneo Cell Proliferation and Migration.
To determine whether hUC-MSC-derived EVs play a role in cell proliferation, we used HTR8-S/Vneo cells, a human extravillous cytotrophoblastderived transformed cell line, and performed an MTT assay. We compared HTR8-S/Vneo cell proliferation with or without hUC-MSC-derived EVs. The result, from the optical density 490 value, indicated an increased cell proliferation rate when HTR8-S/Vneo cells were cultured with hUC-MSC-derived EVs (defined as blank-EVs), compared to HTR8-S/Vneo cells without hUC-MSC-derived EVs (defined as blank; Figure 3(a)). Consistently, the mRNA and protein levels of two proliferative markers, PCNA and Ki67, were upregulated in cells with hUC-MSCs-derived EVs (Figures 3(b) and 3(f )). Moreover, the transwell assay showed that hUC-MSC-derived EVs also elevated the HTR8-S/Vneo cell migration (Figure 3(g)). In conclusion, hUC-MSCderived EVs could enhance HTR8-S/Vneo cell proliferation and migration.

hUC-MSC-Derived EVs Regulated the Expression of miR-195 and TFPI2.
To further explore the role of hUC-MSCderived EVs in regulating TFPI2 and miR-195 expression, we detected the mRNA levels of miR-195 and TFPI2 in HTR8-S/Vneo cells cocultured with hUC-MSC-derived EVs (defined as blank-EVs group) or without EVs (defined as blank group). Compared with the control group, we observed significantly enhanced expression of miR-195, but decreased expression of TFPI2, in the cells cocultured with hUC-MSCderived EVs (Figures 4(a) and 4(b)). Consistently, the protein level of TFPI2 was also decreased in the presence of EVs (Figures 4(c) and 4(d).

TFPI2 Was Targeted and Downregulated by miR-195.
We further examined the knockdown and overexpression of TFPI2 on cell proliferation and migration of HTR8-S/Vneo  Figures S2A and S2D). It was further found that knockdown of TFPI2 promoted the proliferation and migration of HTR8-S/Vneo cells ( Figures S2B and 2C), while overexpression of TFPI2 did not affect the proliferation (Figure S2E) but inhibited the cell migration ( Figure S2F). These results demonstrated a negative correlation between TFPI2 expression and cell proliferation and migration. To evaluate whether miR-195 targeted TFPI2, we analyzed the binding site of hsa-miR-195 with the wild-type (WT) 3'-UTR region of TFPI2 mRNA ( Figure 5(a)). By cotransfecting HTR8-S/Vneo and HEK-293T cells with luciferase reporters containing WT or mutant TFPI2 3'-UTR, as well as miR-195 mimics and negative control, we found that luciferase activities of WT TFPI2-3'-UTR reporter were significantly reduced with miR-195 mimics in both HEK293T and HTR8-S/Vneo cells but showed no change in mutant TFPI2-3'-UTR reporter activities (Figures 5(b) and 5 (c)). We next transfected HTR8-S/Vneo cells with miR-195 mimics or negative control ( Figure 5(d)), and found that TFPI2 protein level was decreased in cells with miR-195 mimics transfection (Figures 5(e) and 5(f)), suggesting that TFPI2 was a target gene of miR-195.

Discussion
Body fluids contain EVs and other secreted proteins, which provide clinical information and improve diagnostic accuracy for treatment selection [22]. For example, human ribonuclease 7 protein, which is secreted from cancer cells, has been reported as a potential biomarker in hepatocellular carcinoma targeted therapy [23]. EVs are small lipid-enclosed particles released by multiple cell types containing components such as noncoding RNAs, DNA fragments, mRNAs, and proteins, and play a critical role in cross talk between cells [24]. Multiple studies have highlighted the utility of tissuespecific EVs in the diagnosis of preeclampsia [25,26]. In presymptomatic women, the concentration of total exosomes and placenta-derived exosomes in maternal plasma was significantly higher than those observed in controls [27].
Placenta-derived exosomes have been reported as potential biomarkers of preeclampsia [28]. In the current study, we have discovered that EVs derived from hUC-MSCs are potential liquid biopsy-based biomarkers for preeclampsia development.
Wang et al. [5,10] reported the connection between multiple myeloma (MM) cells and bone marrow stromal cells (BMSCs). They found that exosomes derived from BMSCs regulated the activation of c-Jun, p53, p38, and Akt signaling,   [29] reported that hUC-MSC-derived exosomes played a critical role in the activation of the Wnt signaling pathway, leading to cell proliferation and wound reepithelialization. In addition, microvesicles derived from human liver stem cells can promote hepatic regeneration through transferring microRNAs and proteins [30]. Consistent with previous studies, we found that hUC-MSC-derived EVs elevated HTR8-S/Vneo cell proliferation and migration abilities in vitro.
To investigate the role of exosomal miR-195 in the development of preeclampsia trophoblast cells, we transfected miR-195 into hUC-MSC-derived EVs and HTR8-S/Vneo cells. We observed that EV-derived miR-195 significantly enhanced HTR8-S/Vneo cell proliferation and migration. Furthermore, we found that EVs upregulated miR-195 expression in HTR8-S/Vneo cells, possibly due to protection from degradation. Luciferase reporter assay data indicated that miR-195 binds to the 3'UTR region of the TFPI2 gene. After coculturing hUC-MSC-derived EVs with HTR8-S/Vneo cells, TFPI2 expression was significantly decreased, suggesting that miR-195 could regulate trophoblast cell proliferation and migration by targeting TFPI2. A previous study reported that TFPI2 is a target of miR-195, involved in the regulation of glioblastoma cell proliferation and apoptosis [11]. In our study, we found a negative correlation between TFPI2 and miR-195 expression in preeclampsia patients and showed TFPI2 to be a target gene of miR-195. Our findings are in line with previous work that studied the role of exosomal miR-195 through an in vitro hypoxia model [14], but our study has more clinical relevance, as we investigated the correlation between TFPI2 and miR-195 in the placenta.
TFPI2 belongs to the Kunitz-type serine proteinase inhibitor family and functions as a tumor suppressor gene in multiple cancer types due to its ability to inhibit tumor cell growth, metastasis, and invasion [31,32]. TFPI2 inhibits various serine proteases, such as plasmin, trypsin, VIIa/tissue factor, chymotrypsin, and plasma kallikrein [33]. Many studies have demonstrated that TFPI2 plays a role in the development of preeclampsia. For example, Zhou et al. [34] reported that upregulation of TFPI2 resulted in decreased cell proliferation and invasion, and induced cell apoptosis. TFPI2 was also found to be involved in the regulation of methylation status on the promoter region in preeclampsia placentas [35]. In the current study, we discovered that TFPI2 mRNA levels were significantly upregulated in preeclampsia tissues and were negatively correlated with miR-195 expression levels. Further luciferase assay indicated that miR-195 could bind to the 3'-UTR region of WT TFPI2 mRNA but not to the region of mutant TFPI2 construct. Additionally, miR-195 mimics significantly decreased TFPI2 protein expression, further suggesting that TFPI2 is a potential target of miR-195.

Conclusion
Our study demonstrates that the upregulation of miR-195 in EVs derived from hUC-MSCs promotes HTR8-S/Vneo cell proliferation and migration by targeting TFPI2. Our findings provide a potential therapeutic target to inhibit human extravillous cytotrophoblast-derived transformed cell proliferation and migration.

Data Availability
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.