Receptive endometrium cell-derived sEVs enhance potency of trophoblast

Since the sEVs isolated from EECs are uptaken by trophoblast (HTR8/SVneo), we speculated that these sEVs might regulate the function of trophoblast, such as migration and signaling pathway. We examined the amount of the sEV protein, and the same amount of Ishikawa-sEVs or HEC-1-A-sEVs were used to treatment trophoblast cells (Figure S4A). After treatment with PBS, Ishikawa-sEVs (50 μg or 100 μg), or HEC-1-A-sEVs (50 μg or 100 μg), respectively, HTR8/SVneo cells were collected and detected the phosphorylated (P; activated) and total forms of signaling proteins widely implicated in promoting cell migration and adhesion, including focal adhesion kinase (FAK) and c-Jun N-terminal kinase (JNK). Results revealed that whereas 100 μg of Ishikawa-sEVs enhanced P-FAK (Figures 3B and S4B), no P-JNK was affected. There was no significant phosphorylation difference between HEC-1-A-sEV treatment and the control cells, as shown in Figures 3B and 3C. To further interrogate the effect of sEVs during implantation, we performed an injection test to examine the number of implanted embryos. The same amount (also same volume) of sEVs and PBS was injected into each side of the mouse uterine horn (left and right) in preimplantation on D3 of pregnancy, respectively (Figure S3A). We found that Ishikawa-derived sEVs increased the chances of implantation but not significantly compared to the PBS control group (Figures S3B and S3C). Conversely, the injection of HEC-1-A (nonreceptive)-derived sEVs significantly decreased the chances of implantation (Figures 3D–3F). This observation was associated with opposing embryonic adhesion that occurs during the preimplantation period.

Open in a separate window

Figure 3

Endometrial epithelial cell-derived sEVs affect embryo implantation

(A–C) HTR8/SVneo trophoblasts were serum starved and treated with PBS, Ishikawa-sEVs, or HEC-1-A-sEVs. After 48 h, the cells were immunoblotted for phosphorylated FAK (P-FAK) and P-JNK. The blots were also detected for total FAK, JNK, and tubulin or GAPDH. The ratio of phospho protein was measured and calculated using phospho protein/total protein. (D) HTR8/SVneo trophoblasts were incubated with serum-free medium supplemented with PBS, Ishikawa-sEVs, or HEC-1-A-sEVs for 12 h, and then the HTR8/SVneo cells were stained for F-actin using FITC-conjugated phalloidin. The white arrows show leading edges. Bar, 20 μm. (E) The percentages of cells with leading edges were determined. (F) Migration and invasion assay of HTR8/SVneo cells treated with PBS, Ishikawa-sEVs, or HEC-1-A-sEVs using Transwell. Bar, 50 μm. (G and H) Migrated and invaded cells were counted, and representative images were shown. (I) Images of wound-closure assays performed on HTR8/SVneo cells cultured in serum-free medium supplemented with PBS, Ishikawa-sEVs, or HEC-1-A-sEVs for 48 h. The dashed lines indicate the width of the wound. Bar, 500 μm. (J) The relative areas of an open wound in (I) were quantified and plotted. (K) The proliferation assay of HTR8/SVneo cells measured by EdU assay. (L) Quantitative analysis of proliferation in (K) (p < 0.05; p < 0.01; p < 0.001).

Leading edges are considered to be a marker of polarized cell migration.28 We examined whether sEVs derived from EECs could trigger the formation of leading edges in the HTR8/SVneo trophoblast. Trophoblasts were serum starved and incubated with PBS or 100 μg sEVs for 24 h, and then the filamentous actin (F-actin), located in leading edges of migrating cells, was stained. The fluorescence microscopy images of HTR8/SVneo showed that Ishikawa-sEV-treated trophoblasts exhibited a polarized morphology, with the F-actin localized in the leading edge of the cells (Figure 3D, middle panel). On the contrary, PBS or HEC-1-A-sEV-treated trophoblasts displayed low polarity with a round shape (Figure 3D, left and right panels). Approximately 47% of Ishikawa-sEV-treated trophoblasts showed leading edges, a two-fold increase compared with either PBS or HEC-1-A-sEV-treated HTR8/SVneo (Figure 3E). Formation of leading edges by the sEV-treated trophoblasts indicates that the sEVs derived from receptive EECs promote the migration ability of trophoblasts. To further investigate the effects of sEVs in the motility potential of trophoblasts, a Transwell migration and invasion assay was performed. The results showed that Ishikawa-derived sEVs enhanced migration of trophoblasts (Figures 3F and 3G). However, PBS or HEC-1-A-sEV-treated HTR8/SVneo cells exhibited low motility potential. The same results were shown when migration was measured in wound-healing (migration) assays. The monolayers of HTR8/SVneo trophoblasts were struck to create wounds and then cultured in medium supplemented with PBS, Ishikawa-sEVs, or HEC-1-A-sEVs. After 48 h, trophoblasts with 100 μg Ishikawa-sEV stimulation were capable of migrating into the wound, whereas there was no difference in the migration of trophoblasts stimulated with PBS or HEC-1-A-sEVs or 50 μg sEV stimulation (Figures 3I, 3J, and S4C).

The invasion and proliferation of embryonic trophoblast cells have the ability for successful embryo implantation and placenta formation. These events increase the connection between the embryos and the maternal uterus for implantation. The invasion ability of HTR8/SVneo was measured by the Transwell assay. HTR8/SVneo trophoblasts were pretreated with PBS, Ishikawa-sEVs, or HEC-1-A-sEVs and then cultured in a Transwell plate coated with Matrigel. Our results showed that Ishikawa-sEVs stimulated the invasion ability of HTR8/SVneo more than the HEC-1-A-sEVs or control PBS (Figures 3F–3H). In addition, we used the 5-ethynyl-2′-deoxyuridine incorporation (EdU) assay to evaluate the effects of sEVs on trophoblasts proliferation. The measurements were conducted 24 h after the transfer of sEVs or PBS into the HTR8/SVneo cells. Whereas PBS or HEC-1-A-sEVs had no significant effect on HTR8/SVneo trophoblast proliferation, receptive EEC-derived sEVs (Ishikawa-sEVs) significantly promoted proliferation of HTR8/SVneo (Figures 3K and 3L). Collectively, the results suggest that receptive endometrium cell-derived sEVs enhance the potency of trophoblasts for embryo implantation.

sEV-derived miRNAs mediate embryo implantation

Exosomes consider being miRNA-abundant EVs, which are released by cells and delivered into recipient cells to involve the biological processes.21 In this study, we have proven that sEVs derived from EECs affect the migration and invasion of trophoblast cells. To assess the changes in miRNA expression between Ishikawa-sEVs and HEC-1-A-sEVs during early pregnancy, we conducted the next deep sequencing of small RNAs from nonreceptive and receptive cell sEVs. There were over 300 miRNAs identified in EEC-derived sEVs (Table S1). In addition, we identified a set of miRNAs that were differentially expressed in Ishikawa versus HEC-1-A cell-derived sEVs. Among these differentially expressed miRNAs (DEMs), Figure 4A shows the most significant (with conditions: |log fold change [FC]| > 1.0 and a p value < 0.05) difference between Ishikawa and HEC-1-A cell-derived sEVs. To reveal the potential functions of the differentially expressed sEV miRNAs, we predicted the target genes of the thirteen miRNAs in silico. The results of bioinformatics analysis revealed that 2,405 mRNAs were potentially targeted by these miRNAs. The biological processes and signal pathways of target genes were mainly enriched in cellular process, biological adhesion, extracellular exosome, focal adhesion, and phosphatidylinositol 3-kinase (PI3K)-Akt signaling pathways (Figures S5A–S5D). The results indicate that the sEV miRNAs derived from endometrial cells might regulate the functions of trophoblast.

Open in a separate window

Figure 4

The miRNA profiles of endometrial cell-derived sEVs

(A) Differential expressed miRNA profiles between Ishikawa- and HEC-1-A-derived sEVs. (B) Expression level of miR-100-5p in Ishikawa and HEC-1-A cells. (C) miR-100-5p abundance in sEVs secreted by Ishikawa and HEC-1-A. (D) Comparison of the miR-100-5p expression level in Ishikawa and sEVs released by Ishikawa. (E) The fold change of miR-100-5p in mouse endometrium on pregnancy D1, D4, and D5. (F) The relative expression of miR-100-5p in mouse endometrium in the early stage of pregnancy. (G) After treatment with PBS, Ishikawa-sEVs, and HEC-1-A-sEVs for 24 h, respectively, the level of miR-100-5p in HTR8/SVneo was measured by qRT-PCR (p < 0.05; p < 0.01; p < 0.001).

Given that, miR-100-5p was previously reported as being essential for cellular migration and viability. Interestingly, the miR-100-5p was significantly upregulated in villus and increased in serum during pregnancy after embryo transfer.29 Among the 13 DEMs, miR-100-5p was especially noted. In our results, the expression level of miR-100-5p was over 20-fold higher in Ishikawa cells compared to HEC-1-A cells (Figure 4B), which led to the abundance of the miR-100-5p present in Ishikawa-derived sEVs (Figure 4C). We also compared the expression level of miR-100-5p between Ishikawa cells and Ishikawa-sEVs. The results showed that miR-100-5p was enriched in sEVs, showing that the miR-100-5p was secreted from Ishikawa cells and packaged into sEVs (Figure 4D). The high expression of miR-100-5p in receptive EECs indicated that miR-100-5p is involved the process of embryo implantation during the WOI. Our small RNA sequencing results from mice endometrium samples during early pregnancy showed that miR-100-5p increased in the endometrium on D4 pregnancy, the receptive period of the uterus (Figure 4E). Consistent with the sequencing results, the miR-100-5p was markedly increased in D3 or D4 compared to D1, D2, or D5 in mouse endometrium in the early stage of pregnancy (Figure 4F). These results indicate that miR-100-5p plays a critical role during embryo implantation. Incubation of HTR8/SVneo cells with Ishikawa cell-derived sEVs led to a significant increase in miR-100-5p content in HTR8/SVneo, whereas treatment with HEC-1-A-sEVs or PBS led to nonsignificant (ns) effects (Figure 4G). Taken together, these findings indicate that receptive endometrial cell-derived sEVs contain miR-100-5p and might mediate the communication between the embryo and maternal endometrium.

miR-100-5p is essential for embryo implantation

We next want to determine whether the EEC-derived exosomal miRNAs mediate the intercellular communications between uterus and embryos. Besides, we explored whether the exosomal miRNAs affect the function of trophoblasts during embryo implantation. We first determined the effect of miR-100-5p in embryo implantation by injecting miR-100-5p antagomir or diethyl pyrocarbonate (DEPC) water into the female uterine horn on D3 of pregnancy. The implantation sites were checked on D7 of pregnancy. As shown in Figure 5A, inhibition of miR-100-5p during WOI drastically impeded the ability of embryo implantation (Figure 5A, left uterine horns) compared to the DEPC-injected group. miRNAs encapsulated in exosomes are abundant, and the exosomes derived from the uterus play an important role in successful embryo implantation.9 To confirm that this is the case, we transfected miR-100-5p mimics into HTR8/SVneo cells (Figure 5B). After 24 h, we examined the P-FAK and P-JNK levels. Consistent with the HTR8/SVneo trophoblast results after stimulation with exosomes, we found that miR-100-5p dramatically phosphorylated FAK or JNK compared to the control (Figures 5C–5E). Furthermore, we set out to determine whether miR-100-5p promotes the ability of migration of HTR8/SVneo trophoblasts. The results from the Transwell assay confirmed that miR-100-5p enhanced the migration of HTR8/SVneo cells (Figures 5F and 5G). In addition, wound-healing assays also reflected that miR-100-5p mimics promote the edges healing, and the effect increased with prolonged transfection time (Figure 5H). However, there was no significant difference in the wound healing in the control group, 24 h or 48 h after transfection, but the cells significantly migrated after treatment with miR-100-5p mimics (Figure 5G). Collectively, these results indicated that miR-100-5p promotes the migration of HTR8/SVneo trophoblasts. Moreover, we found that miR-100-5p mimics contributed to the proliferation potential of HTR8/SVneo cells (Figure 5J). The miR-100-5p transfection group had high proliferation of HTR8/SVneo trophoblasts compared to the control group (Figure 5K). Additionally, as we expected, the Transwell invasion assay revealed that miR-100-5p mimics promote the invasion of HTR8/SVneo cells (Figures 5L and 5M). Overall, the study highlights the importance of endometrial cell-derived sEV miR-100-5p in activating FAK or JNK. The activation of FAK or JNK signaling mediates the migration and invasion ability of trophoblasts that is required for embryo implantation.

Open in a separate window

Figure 5

Exosomal miR-100-5p promotes the ability (migration, invasion, and proliferation) of trophoblasts for embryo implantation

(A) Day 7 implantation sites of females after injecting miR-100-5p antagomir or DEPC water in each side of horn and arrows indicate implanted embryos. (B) The expression of miR-100-5p in HTR8/SVneo is detected after transfecting miR-100-5p mimics or negative control for 24 h. (C–E) HTR8/SVneo trophoblasts were transfected with miR-100-5p mimics or negative control. 24 h later, the cells were immunoblotted for P-FAK and P-JNK. The blots were also detected for total FAK, JNK, and GAPDH. The ratio of phospho protein was measured and calculated using phospho protein/total protein. (F and G) Migration assay of HTR8/SVneo cell-transfected miR-100-5p-mimic or negative control. Relative areas of migrated cells were counted, and representative images were shown. Bar, 50 μm. (H) Wound-closure assays were performed on HTR8/SVneo cells of transfection with miR-100-5p mimics or negative control. The images were captured in 24 h and 48 h after transfection. Bar, 500 μm. (I) Relative area of open wound in (H) was quantified and plotted. (J and K) HTR8/SVneo cells were transfected directly with miR-100-5p mimics or negative control. Subsequently, after 24 h, the ratio of the cells of proliferation was counted and plotted. Representative images were captured using fluorescence microscopy and shown. (L and M) Invasion analysis of HTR8/SVneo cells transfected with miR-100-5p mimics or negative control. Bar, 50 μm. The invaded cells were counted, and representative images were shown (p < 0.05; p < 0.01; p < 0.001).

Cell culture

Human EEC lines (Ishikawa and HEC-1-A) were purchased from the cell bank of the Chinese Academy of Science (Shanghai, PR China). HTR8/SVneo trophoblast cells were purchased from ATCC. HUVEC was a gift from Dr. Qin Zhiyuan, College of Pharmaceutical Sciences, Zhejiang University. Ishikawa cells and HUVECs were cultured in DMEM medium (Gibco), supplemented with 10% (v/v) FBS (Gibco). HEC-1-A cells were cultured in McCoy’s 5A (Sigma), supplemented with 10% (v/v) FBS. HTR8/SVneo cells were cultured in RPMI-1640 medium (Gibco), supplemented with 5% (v/v) FBS.

Isolation and purification of sEVs

For sEV purification from cell culture medium, cells were cultured with 10% exosome-free FBS. Exosome-depleted FBS was prepared by overnight ultracentrifugation at 100,000 × g, 4°C. Cells were cultured for 48–36 h, and then supernatants were collected for sEV isolation using a standard centrifugation protocol, which referenced previous research. In brief, cell culture supernatants were centrifuged at 500 × g for 10 min and 2,000 × g for 20 min to remove dead cells and debris. Following, the supernatants were centrifuged at 10,000 × g for 30 min and filtered with 0.22 μm to pellet big microvesicles. Supernatants were then centrifuged at 120,000 × g for 75 min twice. The pelleted sEVs were suspended in PBS for further usages.

The size distribution and concentration of sEVs isolated from cell culture supernatants were determined using a ZetaView PMX 100 (Particle Metrix, Meerbusch, Germany). For each group, the analysis was performed independently at least three times.

Transmission electron microscopy

10 μL of freshly isolated sEVs was transferred onto formvar carbon-coated copper grids to dry, rinsed in double-distilled water, and negatively stained by 2% uranyl acetate at room temperature for 1 min. The image was observed with a Tecnai G2 Spirit 120 kV transmission electron microscope operating at 120 kV (Thermo Fisher Scientific, FEI).

For observation of MVBs in cells, Ishikawa and HEC-1-A cells were harvested, centrifuged, and washed with PBS. The cell mass was fixed with 2.5% glutaraldehyde overnight at 4°C and rinsed in water. Following, the cells were fixed with 1% osmic acid, stained using 2% uranyl acetate, and dehydrated in gradient alcohol. Finally, the cell mass was embedded in epoxy resin. Ultrathin sections were prepared, stained with uranyl acetate and lead citrate, and examined by a frozen transmission electron microscope.

Confocal microscopy

For the characterization of Ishikawa and HEC-1-A cell lines, MUC1 in cells was detected using immunofluorescence. In brief, 50% confluent cells were washed with PBS and fixed with 4% paraformaldehyde for 20 min at room temperature. Then, the cells were treated with 0.1% Triton X-100 and blocked with 5% bovine serum albumin (BSA). The cells were incubated with MUC1 antibody, washed, and then incubated with A555-conjugated secondary antibody, FITC-conjugated phalloidin, and 4′,6-diamidino-2-phenylindole (DAPI; Beyotime). Cells were rinsed with PBS before being measured by fluorescence microscopy. To visualize leading edges on HTR8/SVneo, cells, which had been stimulated with exosomes, were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, blocked with 5% BSA, and stained with DAPI. Then, cells were visualized by a Zeiss laser-scanning confocal microscope (Germany). Images were captured by ZEN 2010 software.

For uptake tracking analysis, 50 μg sEVs was incubated with DiI dye with 20 μM. sEVs were purified by ultracentrifugation at 10,0000 × g. Purified sEVs were then added to the HTR8/SVneo cell culture and incubated for 2, 4, 6, and 12 h. Cells were washed and fixed. Then, cells were treated with 0.1% Triton X-100 and stained with DAPI. Confocal microscopy was used for visualization.

Western blot

WCLs or sEV proteins were separated using 10%–12% SDS-PAGE gels and transferred to polyvinylidene difluoride membranes (Millipore). The blots were blocked with QuickBlock western buffer (Beyotime) for 20 min at room temperature and incubated with primary antibody at dilutions recommended by the manufacturer at 4°C overnight. Following, the blots were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature for 2 h. Immunodetection was detected using BeyoECL Plus (Beyotime; cat. no. P0018S). CD63, Alix, TSG101, HSP70, and CD9 were used as exosome markers. CANX was used as a negative control. MUC1 was used to distinguish cell lines. GAPDH and alpha tubulin were used as a loading control.

Transwell assay

Ishikawa and HEC-1-A cells were incubated with 10 μM DiI (Beyotime) for 20 min at room temperature. For coculture assay, HTR8/SVneo cells were plated in 12-well Transwell plates with inserts (Corning). DiI-labeled Ishikawa and HEC-1-A cells were seeded into the upper compartment of inserts. Then, HTR8/SVneo cells were cocultured with labeled cells for 12 h. Fluorescence detection was performed using a fluorescence microscope (Nikon).

Cell proliferation assay

Cells were incubated with 10 μM EdU (cat. no. C0075S; Beyotime) in a 12-well plate for 2 h and fixed with 4% paraformaldehyde for 15 min at room temperature. Then, cells were washed with 1 mL PBS, 3 times, and permeabilized using PBS containing 0.3% Triton X-100 for 15 min. Next, a 200-μL click reaction mixture (Beyotime) was added to each well and incubated for 30 min. Hoechst was used to stain the nucleus. Images were captured using a fluorescence microscope (Nikon).

RNA extraction and qRT-PCR assay

The same amount of Caenorhabditis elegans cel-39-3p miRNA was spiked into each sEV sample as an external calibration for RNA extraction, RT, and miRNA amplification. Total RNAs were extracted from cells using the Trizol reagent (Tiangen; cat. no. DP421). cDNAs were synthesized using the first-strand cDNA synthesis kit (Tiangen; cat. no. KR118). The relative expression of mRNA was measured using the SuperReal PreMix Color (SYBR Green) qRT-PCR Kit (Tiangen; cat. no. FP215). miRNAs in exosomes, cells, and tissues were extracted using the miRNeasy Mini Kit (QIAGEN; cat. no. 217184), according to the manufacturer’s protocols. Then, RNA was reverse transcribed to cDNA following the kit protocol (Tiangen; cat. no. KR211). The miRNA level was detected using the miRcute Plus miRNA qPCR Kit (SYBR Green) (Tiangen; cat. no. FP411). Data were normalized to levels of GAPDH (mRNA), U6 (cellular and tissue miRNA), or cel-39 (exosomal miRNAs) and analyzed by the 2^−ΔΔCt method.

Exosomal miRNA sequencing (miRNA-seq) and bioinformatics analysis

miRNA components in Ishikawa-sEVs (n = 3) and HEC-1-A-sEVs (n = 3) were profiled by miRNA sequencing analysis (Illumina HiSeq). miRNA sequencing reads were normalized and quality assessed with fastp; then, the clean reads were mapped to reference genomes using miRdeep2 software. Next, the number of miRNA reads in each sample was counted, and RPMs (reads per million) were used to normalize the expression. The limma package was used to estimate precision weights for all observations and then identify DEMs. Visualizations were generated with the ggplot2 and Heatmap R packages. Target genes of DEMs were predicted by miRTarBase (http://mirtarbase.mbc.nctu.edu.tw/php/search.php). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis were performed using GO packages in R basic.

miRNA transfection

Mimics and antagomir of miR-100-5p were purchased from GenePharma (Shanghai, PR China) and diluted using DEPC water. Cells were cultured in a six-well plate and transfected with miRNAs using the HiPerFect Transfection Reagent (QIAGEN; cat. no. 301702), according to the manufacturer’s procedures. In brief, 5 μL of 20 μM miR-100-5p mimics was mixed with transfection reagents and incubated for 10 min at room temperature to allow the formation of transfection complexes. Then, the complexes were added drop-wise onto the cells and incubated for 24 h until analysis.

Cell migration and invasion assay

HTR8/SVneo cells were pretreated with exosomes or miRNA mimics for 24 h before the cell motility assay. Then, a Transwell with polycarbonate membranes (8 μm pore size) (Corning) was employed. For a migration assay, 1 × 10⁵ HTR8/SVneo cells were resuspended with RPMI-1640 medium without FBS and seeded into the inserts of the Transwell. Meanwhile, 500 μL RPMI 1640 containing 10% exosome-free FBS was added to the lower compartment. For invasion assay, the inserts of Transwell were precoated with Corning Matrigel (cat. no. 356234) and incubated at 37°C for 2 h to allow polymerization. Next, the cells were cultured in the upper compartment of Transwell without FBS and placed into the lower chamber with 500 μL RPMI 1640 containing 10% exosome-free FBS. After incubation for 24 h, the cell inserts were fixed and stained with 0.1% crystal violet solution (Solarbio; cat. no. G1064). Representative fields were photographed, and the number of cells was counted.

Wound-closure assays

Cells were plated into six-well plates until 70%–80% cell confluent monolayers. Then, the cells were serum starved and wounded using a pipet tip. After removing shedding cells, cells were treated with EEC-derived sEV or miRNA mimics and allowed to migrate for 24 h. The extent of wound closure was observed under the microscope and analyzed.

Injection test

For implantation site detection, a surgical operation was performed for the female mouse on D3 of pregnancy. Equal quality miR-100-5p antagomir or sEVs and their control were injected into one side of the uterine horn, respectively. 4 days later, the number of implantation sites was checked.

Tube-formation assay of HUVEC

In vitro angiogenesis experiments were determined by performing a tube-formation assay in Matrigel (Corning). 48-well plates were precoated using 200 μL Matrigel and placed in a cell incubator at 37°C for 2 h. Then, HUVECs with treatment (sEV or miRNA mimics) were resuspended in medium, supplemented with 10% FBS, seeded into 48-well plates, and cultured for 8 h. Tube formation was examined by an optical microscope (Nikon), and the branch density and tube length were quantified and plotted.

This research was supported by grants from the Department of Science Technology of Huzhou City (2019GZ14), Zhejiang Team Science and Technology Commissioner Project (Tongxiang) and Agricultural Science and Technology Innovation Team of Huzhou City (2019HN01).