FABP4 mediates endoplasmic reticulum stress and autophagy to regulate endometrial epithelial cell function during early sheep gestation.

To investigate the involvement of FABP4 in the regulation of endometrial receptivity during implantation of the sheep conceptus, we used RT-qPCR to determine the expression of FABP4 mRNA in the sheep endometria on days 4 and 15 of pregnancy (n = 3). The analysis showed that the transcription level of FABP4 in the sheep endometrium on gestation day 15 was significantly higher than that on gestation day 4 ( Fig. 1a Fig. 1. Expression levels of FABP4 in sheep endometrium. (a) FABP4 mRNA expression in sheep endometrium on day 4 and day 15. (b, c) FABP4 protein expression in sheep endometrium on day 4 and 15. The optical density was normalized to the density of β-actin in the same lane. (d, e) Rabbit IgG group was used as the negative control for analyzing the localization and expression of FABP4 in sheep endometrium. Data are presented as mean ± standard error with significant differences at P < 0.05 and extremely significant differences at P < 0.01. * indicates P < 0.05, ** indicates P < 0.01, and no sign indicates that the difference is not significant. The technique was repeated thrice. ). FABP4 protein levels also showed the same trend when analyzed using WB ( Figs. 1b, c ). Furthermore, immunohistochemical results showed that FABP4 was widely expressed in sheep endometrial epithelial and stromal cells and was significantly upregulated in endometrial epithelial cells on the day 15 of pregnancy ( Figs. 1d, e ). These results suggest that uterine FABP4 expression is elevated during conceptus implantation in sheep. Expression levels of FABP4 in sheep endometrium. (a) FABP4 mRNA expression in sheep endometrium on day 4 and day 15. (b, c) FABP4 protein expression in sheep endometrium on day 4 and 15. The optical density was normalized to the density of β-actin in the same lane. (d, e) Rabbit IgG group was used as the negative control for analyzing the localization and expression of FABP4 in sheep endometrium. Data are presented as mean ± standard error with significant differences at P < 0.05 and extremely significant differences at P < 0.01. * indicates P < 0.05, ** indicates P < 0.01, and no sign indicates that the difference is not significant. The technique was repeated thrice. We successfully isolated and purified SEECs ( Fig. 2a Fig. 2. Expression of FABP4 in SEECs. (a) Isolation and purification of SEECs (250 ×). When the cells reached the sixth passage, they became larger and rounder and began to senesce. (b) Immunofluorescence identification of SEECs (CK18) (100 ×). (c) Localization of FABP4 in SEECs (Yellow arrows are FABP4 in the nuclei) (25 ×). ) and verified their identity using the epithelial cell marker CK18 ( Fig. 2b ). Immunofluorescence confirmed that FABP4 was expressed in the cytoplasm and nucleus of SEECs ( Fig. 2c ). The cells that were cultured in vitro were not spatio-temporally specific, so we constructed a model of SEECs that were stimulated by peri-implantation stage hormones P 4 , E 2 , and IFN-τ to simulate the uterine environment at the peri-implantation stage. The protein expression level of PGR decreased and that of ER increased after hormone treatment ( Fig. 3a Fig. 3. Hormone treatment of SEECs and detection of changes in FABP4 expression. (a) Protein expression levels of PGR and ER after combined hormone treatment. (b) The mRNA expression levels of ISG15, HOXA10, CXCL10, and RSAD2 after hormone treatment. (c) Levels of the prostaglandin secreted from the SEECs after hormone treatment.(d) Expression levels of FABP4 after hormone treatment. All data are presented as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. ). The mRNA expression levels of ISG15, HOXA10, CXCL10, and RSAD2 did not change significantly under E 2 and P 4 treatment but increased after IFN-τ treatment alone and E 2 , P 4 , and IFN-τ treatment ( Fig. 3b ). After stimulation with IFN-τ, the secretion of PGE 2 ( Fig. 3c ) in the culture supernatants significantly increased, whereas the secretion of PGF 2α ( Fig. 3c ) reduced, indicating that the SEECs continued to secrete PGE 2 after hormone treatment. The changes in FABP4 expression were determined using RT-qPCR and WB, and the results showed that FABP4 expression did not change after E 2 and P 4 treatments but was significantly increased after the addition of IFN-τ ( Fig. 3d ). Expression of FABP4 in SEECs. (a) Isolation and purification of SEECs (250 ×). When the cells reached the sixth passage, they became larger and rounder and began to senesce. (b) Immunofluorescence identification of SEECs (CK18) (100 ×). (c) Localization of FABP4 in SEECs (Yellow arrows are FABP4 in the nuclei) (25 ×). Hormone treatment of SEECs and detection of changes in FABP4 expression. (a) Protein expression levels of PGR and ER after combined hormone treatment. (b) The mRNA expression levels of ISG15, HOXA10, CXCL10, and RSAD2 after hormone treatment. (c) Levels of the prostaglandin secreted from the SEECs after hormone treatment.(d) Expression levels of FABP4 after hormone treatment. All data are presented as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. To explore the effect of FABP4 on SEEC function, the cells were treated with the FABP4 inhibitor BMS309403. The results of the scratch-healing experiment showed that the migratory ability in the control and hormone-treated groups was significantly higher than that in the inhibitor group at 96 h ( Figs. 4a, b Fig. 4. FABP4 inhibition impedes SEEC function. (a, b) Mobility of SEECs at 0, 24, 48, 72 and 96 h was measured using a scratch test. Migratory capacity was calculated as a percentage of healing area relative to time 0. (c) CCK-8 viable cell counts quantified the proliferative capacity of SEECs treated with hormone and FABP4 inhibitor BMS309403. (d–g) Expression levels of EMT after hormone and inhibitor treatment (E-cadherin, N-cadherin, Vim, and β-catenin). (h, i) Changes in endoplasmic reticulum stress-related protein CHOP and GRP78 were measured. (j, k) Changes in autophagy-related proteins p-mTOR, LC3B II/I, and P62. Data are expressed as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. ). The results of the CCK-8 assay showed that compared with that of the control group, the proliferative ability of SEECs in the treatment group did not change after combined hormone and inhibitor BMS309403 treatment ( Fig. 4c ). We also determined the expressional changes of the EMT markers E-cadherin, N-cadherin, Vim, and β-catenin after combined hormone and FABP4 inhibitor treatment compared with that in the control group ( Figs. 4d–g ). The expression of E-cadherin and β-catenin increased, whereas that of Vim and N-cadherin decreased, validating the decrease in migratory capability of SEECs after FABP4 inhibition. FABP4 inhibition impedes SEEC function. (a, b) Mobility of SEECs at 0, 24, 48, 72 and 96 h was measured using a scratch test. Migratory capacity was calculated as a percentage of healing area relative to time 0. (c) CCK-8 viable cell counts quantified the proliferative capacity of SEECs treated with hormone and FABP4 inhibitor BMS309403. (d–g) Expression levels of EMT after hormone and inhibitor treatment (E-cadherin, N-cadherin, Vim, and β-catenin). (h, i) Changes in endoplasmic reticulum stress-related protein CHOP and GRP78 were measured. (j, k) Changes in autophagy-related proteins p-mTOR, LC3B II/I, and P62. Data are expressed as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. Endometrial receptivity is reportedly associated with endoplasmic reticulum stress and autophagy [ 28 , 34 ]. To determine whether FABP4 affects the receptivity of SEECs through endoplasmic reticulum stress or autophagy pathways, the markers of endoplasmic reticulum stress and autophagy were detected using WB ( Figs. 4h–k ). The results confirmed that CHOP, GRP78, P62, p-mTOR, and the ratio of LC3BII/I protein levels were upregulated after inhibitor treatment, and the inhibition of FABP4 expression in SEECs reduced the levels of endoplasmic reticulum stress and autophagy in cells. Treatment of SEECs with P 4 , E 2 , and IFN-τ mimics the intrauterine environment during implantation and increases FABP4 expression. After hormone treatment, the expression of endoplasmic reticulum stress-related proteins GRP78 and eIf2α increased significantly, and the protein expression levels of CHOP and IRE1α showed an upward trend. Compared with that of the hormone treatment group, the addition of FABP4 inhibitor before hormone treatment inhibited the expression of CHOP, GRP78 and eIf2αa proteins. After the addition of the endoplasmic reticulum stress activator TG, the level of endoplasmic reticulum stress significantly increased ( Figs. 5a, b Fig. 5. TG and 3-MA treatment partially restores SEEC function after BMS30940 suppression of FABP4. (a, b) Expression levels of key proteins in the endoplasmic reticulum stress signaling pathway after combined treatment with hormones, BMS309403, and TG. (c, d) Expression levels of key proteins in autophagy and apoptosis signaling pathways after combined treatment with hormones, BMS309403, and 3-MA. (e) Secreted prostaglandin levels from SEECs after combined treatment with hormones, BMS309403, and TG or 3-MA. Data are expressed as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. ). The quantity of the FABP4 inhibitor PGF2α secreted increased significantly before hormone treatment, whereas that of PGE2 secreted decreased after the addition of TG ( Fig. 5e ). These results demonstrate that FABP4 plays an important role in peri-implantation in sheep by influencing the endoplasmic reticulum stress levels. TG and 3-MA treatment partially restores SEEC function after BMS30940 suppression of FABP4. (a, b) Expression levels of key proteins in the endoplasmic reticulum stress signaling pathway after combined treatment with hormones, BMS309403, and TG. (c, d) Expression levels of key proteins in autophagy and apoptosis signaling pathways after combined treatment with hormones, BMS309403, and 3-MA. (e) Secreted prostaglandin levels from SEECs after combined treatment with hormones, BMS309403, and TG or 3-MA. Data are expressed as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. After hormone treatment, the expression levels of p-mTOR, p-P70S6K, P62, and Beclin1 decreased, and the LC3BII/I ratio increased, indicating that autophagy in SEECs was activated after hormone treatment. The expression levels of the FABP4 inhibitor BMS309403p-mTOR, p-P70S6K, P62, and Beclin1 increased before hormone treatment, and the LC3BII/I ratio decreased, indicating that FABP4 inhibition reverses the expression of autophagy-related proteins after hormone treatment. After the addition of the 3-MA autophagy inhibitor, the expression levels of P62 and Beclin1 proteins increased and the LC3BII/I ratio decreased ( Figs. 5c, d ). Compared with that of the combined hormone and BMS309403 treatment group, the addition of 3-MA reduced the quantity of PGF2α secreted ( Fig. 5e ). These results showed that hormone treatment inhibited autophagy in SEECs and that FABP4 was involved in the regulation of cellular function by affecting autophagy in sheep SEECs. To validate the previous result that FABP4 may affect the function of SEECs through the mTOR and UPR pathways at the cellular level, we determined the expression of endoplasmic reticulum stress and autophagy markers on days 4 and 15 of pregnancy. The results showed that the expression of CHOP and GRP78 increased ( Figs. 6a, b Fig. 6. Endoplasmic reticulum stress and autophagy levels in sheep endometrium. (a, b) Expression of CHOP and GRP78 proteins on the day 4 and 15 of pregnancy. (c, d) Expression of p-mTOR, P62, and LC3B II/I. Data are expressed as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01. ), whereas that of p-mTOR and P62 and the LC3BII/I ratio increased ( Figs. 6c, d ) in the sheep endometrium on day 15 of gestation. This indicates that endoplasmic reticulum stress increased and autophagy was inhibited in the endometrium on day 15 of pregnancy. Thus, these experimental results indirectly confirm our earlier findings on SEECs by indirectly demonstrating the changes in the endometrial epithelium during conceptus implantation. Endoplasmic reticulum stress and autophagy levels in sheep endometrium. (a, b) Expression of CHOP and GRP78 proteins on the day 4 and 15 of pregnancy. (c, d) Expression of p-mTOR, P62, and LC3B II/I. Data are expressed as mean ± standard error. Differences were considered significant at P < 0.05 and extremely significant at P < 0.01.

FABP4 plays an important role in human endometrial epithelial cell function and trophoblast formation, and inhibition of FABP4 expression leads to reduced proliferation, migration, and invasion of epithelial cells. In this study, we collected sheep uterine caruncles on days 4 and 15 of pregnancy and found that FABP4 mRNA and protein levels increased in the uterine caruncles on day 15 of pregnancy. In combination with previous data, we speculate that FABP4 is functionally important in SEECs during the perinatal phase. FABP4 inhibitor reduced cell proliferation, migration, and invasion of human EEC lines at concentrations of 10, 25, 50, and 100 μM with the most pronounced effect observed at 100 μM [ 17 ]. Our results showed that an inhibitor concentration of 20 μM had no significant effect on SEEC proliferation but it did significantly reduce its migratory ability. EMT directly regulates cell migration, and the factors that regulate the EMT process include growth factors, hormones, and WNT/β-catenin pathway [ 35 ]. The WNT-dependent removal of E-cadherin and β-catenin from junctions during EMT is a complex process. In this study, hormone treatment did not significantly affect the migratory ability of SEECs; however, it changed the gene expression trend of E-cadherin and N-cadherin; the migratory ability of SEECs was inhibited after FABP4 inhibitor treatment, and the expression levels of E-cadherin and β-catenin genes significantly decreased. We speculate that the variability in the results of cell proliferation experiments may be caused by differences in cellular parameters, as reflected in the variation in the proliferative ability of primary cells compared with that of the cell lines. In this study, we investigated the roles and molecular mechanisms underlying the action of FABP4 in hormone-treated SEECs. We simulated hormonal stimulation of SEECs in early pregnancy and created a basis for the establishment of a cell model of early pregnancy in sheep with the aim of exploring more changes during early pregnancy in sheep using similar models in the future. FABP4 was previously thought to primarily occur in adipocytes and macrophages, and studies have mostly focused on the occurrence and development of metabolic diseases. In this study, we found that FABP4 was expressed in the sheep endometrial tissue and was highly expressed in the gestational endometrium. FABP4 is a cytoplasmic fatty acid chaperone that shuttles back and forth between the cytoplasm and the nucleus [ 36 ]. Our research showed that FABP4 is found both in the cytoplasm and the nuclei of some SEECs. Our data indicated that FABP4 expression increased significantly in the endometrium of sheep on day 15 of pregnancy, indicating that the endometrium was undergoing lipid transport. The hormonal changes in the uterus when SEECs were used to simulate early pregnancy showed a similar trend, which further confirmed previous findings and indicated that FABP4 plays an important role in early pregnancy in SEECs. After SEECs were treated with E 2 , P 4 and IFN-τ, the PGR protein expression level decreased significantly and that of ER increased only under E 2 and P 4 treatment. IFN-τ is secreted into the uterine vein after entering the uterine cavity, resulting in increased expression of interferon stimulated genes (ISGs) in peripheral blood and luteal tissue. Microarray technology shows that many ISGs are up-regulated during pregnancy recognition [ 37 , 38 ]. Previous studies have shown that the expression of ISG15, RSAD2, HOXA10, and CXCL10 in the endometrial tissue is significantly up-regulated on day 16 after pregnancy in ruminants [ 39 , 40 , 41 ]. In ruminants, the abundant expression of ISG15 is positively correlated with the IFN-τ produced by embryos [ 42 ]. Therefore, in our study, the ISG15, CXCL10, RSAD2, and HOXA10 were measured after subjecting the SEECs to hormone treatment. The results showed that the mRNA expression levels of ISG15, CXCL10, RSAD2, and HOXA10 significantly increased following IFN-τ stimulation. Previous studies have reported that IFN-τ treatment significantly reduced the amount of PGF 2α released by endometrial stromal cells, suggesting that IFN-τ negatively regulates the secretion of PGF 2α in goat endometrial stromal cells. The transport of PGE 2 from the uterus to the luteum during the establishment of pregnancy in sheep increased the biosynthesis and signal transduction of PGE 2 in the luteum to resist luteolysis [ 43 ]. Our findings replicated the changes in PGE 2 and PGF 2α at the cellular level. The level of PGF 2α secreted by SEECs decreased and that of PGE 2 increased after hormone treatment, indicating that it improved the ability of SEECs to resist luteolysis and improved the conditions for maintaining pregnancy. Changes in hormonal levels in the endometrium are critical for sheep conceptus implantation, and studies have shown that E 2 , P 4 and IFT-τ treatment of goat EECs induced an increase in endoplasmic reticulum stress level [ 29 ]. Recently, many studies have reported that FABP4 is closely associated with endoplasmic reticulum stress and autophagy. Numerous studies have provided evidence that endoplasmic reticulum stress is associated with reproduction in female mammals and that physiological levels of endoplasmic reticulum stress are beneficial for embryo implantation; however, excessive endoplasmic reticulum stress can lead to implantation failure and early pregnancy loss [ 44 , 45 ]. Our results showed that FABP4 inhibition significantly increased the expression of genes downstream of the UPR and mTOR signaling pathways. endoplasmic reticulum stress triggers the UPR response, and UPR is involved in three signaling pathways: (1) PERK/eIF2α/ATF4, (2) IRE1α/XBP-1, and (3) ATF6α pathways [ 46 ]. Hormone treatment combined with FABP4 inhibition in SEECs reduced hormone-induced endoplasmic reticulum stress levels. The expression of CHOP, GRP78, p-eIf2α, and p-IRE1α reduced to various degrees, and the corresponding levels of PGE2 secreted also decreased. The endoplasmic reticulum stress-mediated unfolded protein response is a cellular protective mechanism. The expression of endoplasmic reticulum stress-related proteins is upregulated during embryo implantation, and excessive endoplasmic reticulum stress occurs in the decidual tissue of the endometrium after implantation failure [ 28 ]. Our results indicate that under certain circumstances, endoplasmic reticulum stress is beneficial for embryonic attachment. However, the inhibition of FABP4 caused the migratory ability of SEECs to weaken and the endoplasmic reticulum to be overstressed, which was not conducive to attachment. The appropriate endoplasmic reticulum stress levels after hormone treatment are conducive for embryo attachment. Here, we demonstrated that under hormone-treatment conditions, inhibition of FABP4 influenced endoplasmic reticulum stress and PGE 2 levels and decreased the ability of SEECs to maintain pregnancy. Therefore, endoplasmic reticulum stress is a double-edged sword in various physiological processes. However, the molecular mechanisms underlying this functional transition of endoplasmic reticulum stress have not yet been elucidated. Sustained or excessive endoplasmic reticulum stress can lead to autophagy. Relief of endoplasmic reticulum stress can lead to the breakdown of misfolded and unfolded proteins through endoplasmic reticulum degradation-related proteins and lysosome-mediated autophagy; three downstream molecules associated with the UPR can regulate autophagy through different molecular mechanisms: (1) IRE1/JNK/Beclin 1, (2) PERK/eIF2α/ATF4/Beclin 1, and (3) ATF6/CHOP/LC3B [ 47 ]. In the present study, hormone treatment of SEECs induced high expression of FABP4, and inhibition of FABP4 with BMS309403 counteracted hormone-mediated autophagy and the expression of m-TOR, p-70S6K, beclin1, P62, and LC3B in the autophagy pathway. After adding the autophagy inhibitor 3-MA, the expression of mTOR and that of the downstream molecules p-70S6K and LC3B was significantly downregulated, and the level of PGE 2 secreted by SEECs increased. These results suggest that under certain circumstances autophagy is conducive to the establishment of SEECs receptivity, and the hormone-induced autophagy response after FABP4 inhibition is negated. On day 15 of pregnancy, endometrial endoplasmic reticulum stress and autophagy levels increased, p-mTOR expression decreased, the LC3B ratio increased, and P62 expression in tissues increased. Increased levels of P62 protein may indicate that autophagy is blocked, and further studies are needed to determine the role of autophagy in the endometrium. With in-depth studies of early embryo engraftment in ruminants, many new ideas have been proposed—the change in endoplasmic reticulum stress levels in early pregnancy being one of them [ 29 ]. Our research has only begun to clarify the physiological changes that occur in SEECs, and researchers need to work together in subsequent studies to elucidate more of the secrets of early embryo implantation in mammals. In recent years, both endoplasmic reticulum stress and autophagy have been shown to play a role in all aspects of reproduction in female mammals. Studies on the role of FABP4 in female reproduction have focused on metabolic diseases during pregnancy. Research on endometrial receptivity has shown that FABP4 is indispensable in the regulation of EECs function; however, the mechanism underlying the action of FABP4 has not been discussed in-depth. In the present study, we have explored the role of FABP4 in SEECs during embryo implantation. FABP4 inhibition disturbed the level of endoplasmic reticulum stress and autophagy levels in cells; however, induction of endoplasmic reticulum stress or inhibition of autophagy reversed the changes in gene expression caused by FABP4 inhibition. FABP4 is highly related to endoplasmic reticulum stress and autophagy during embryo implantation, but its mechanism of action remains to be clarified. Our results support the idea that FABP4 is involved in the functional regulation of ovine SEECs, and that in the presence of E 2 , P 4 and IFT-τ stimulation, FABP4 and endoplasmic reticulum stress levels are elevated, whereas autophagy and PGF2α levels are inhibited. Hormone-induced changes in both endoplasmic reticulum stress and autophagy were prevented by FABP inhibition, and the same trends as were observed after treatment with TG and 3-MA. These data suggest that FABP4 affects SEEC receptivity via the UPR and mTOR pathways. Our results provide new theories and perspectives for studying the mechanisms underlying the implantation of mammalian conceptuses.

The authors declare that they have no competing interests.

All experimental procedures were approved by the Animal Care Commission of the College of Animal Science and Technology at Shihezi University. In this experiment, twelve 24-month-old healthy nulliparous Kazakh ewes were artificially inseminated after simultaneous estrus treatment. The day of artificial insemination was denoted as day 0. Six sheep were euthanized on day 4 and six more on day 15; the ipsilateral sheep uterine caruncles were collected on days 4 (n = 3) and 15 (n = 3) of pregnancy, and some of the tissues were placed in 4% paraformaldehyde (Solarbio, Beijing, China) for fixation, whereas the rest were stored in liquid nitrogen. Isolation and culture of SEECs was performed as described previously [ 31 ]. The cells used in all experiments were cultured until the third passage, and their purity was greater than 95%. When SEEC growth reached 80–90% confluence, the medium was replaced with fresh DMEM/F12 and treated with 20 μM of the FABP4 small molecule inhibitor BMS309403 (Glpbio, Montclair, NJ, USA, Cas: 300657-03-8) [ 32 ]. When multiple drugs were combined, the cells were first treated with 20 μM of BMS309403 for 1 h and then with P 4 (0.1 μM) (Sigma, Missouri, MI, USA, Cas:57-83-0) and E 2 (1 nM) (Sigma, Cas: 50-28-2) for 12 h. Finally, IFN-τ (20 ng/ml) (Sangon Biotech, Shanghai, China, purity: > 97%) was added for 12 h [ 29 ]. In the case of TG (1 μM) [ 33 ] (Glpbio, Cas: 67526-95-8) or 3-MA (5 mM) (Glpbio, Cas: 5142-23-4) treatment groups, TG and 3-MA were added 1 h prior to the addition of IFN-τ. The total treatment duration for all cells was 24 h. All drugs except IFN-τ were dissolved with DMSO; the amount of DMSO used was maintained at 0.1%, and 0.1% DMSO was used as the control. After cell treatment, the culture medium supernatant was stored at –80°C for ELISA experiments, and the RNA and protein were extracted from the cells as needed for subsequent experiments. Total RNA was extracted from tissues and cells using TRIzol reagent according to the manufacturer’s instructions, and the RNA concentration was determined using a microplate spectrophotometer (Thermo, Waltham, MA, USA). RT-qPCR primers were designed using Primer 5.0. Total RNA was transcribed into cDNA, the concentration of which was determined using a spectrophotometer. The RT-qPCR reaction mixture was prepared using SYBR Green (TaKaRa, Dalian, China) according to the manufacturer’s instructions. The experiments were performed using a Roche LightCycler 96 real-time fluorescence qPCR instrument, and the 2 ΔΔCt method was used to analyze gene expression data. The primers were synthesized by the Shanghai Sangon Group. Each experiment was performed in triplicate. The primers used for RT-qPCR are listed in Supplementary Table 1 . The tissue samples were fixed in 4% paraformaldehyde, paraffin-embedded, and serially sectioned at 5 μm thickness. The tissue sections were stained with hematoxylin & eosin (H&E; Solarbio). Using an immunohistochemical staining kit (Boster, Wuhan, China) and FABP4 rabbit polyclonal antibody (Proteintech, Chicago, IL, USA, 15872-1-AP, 1:200), the FABP4 expression in the tissue samples was detected on days 4 and 15 of gestation. Purified rabbit IgG was used as the negative control (SAB5500149, 1:200; Sigma). The signal intensity was determined using ImageJ software. SEECs were fixed with 4% paraformaldehyde at room temperature, blocked with BSA for 30 min, and FABP4 rabbit polyclonal antibody (Boster, BM4029, 1:50) and CK18 mouse monoclonal antibody (Abcam, Cambridge, UK, ab668, 1:50) were added overnight at 4°C. After washing, the secondary antibodies, namely, Alexa Fluor® Plus 594 anti-rabbit IgG (Thermo) and GFP mouse monoclonal antibody (Beyotime, Shanghai, China, 1:200) were added, followed by incubation at 37ºC for 30 min. Nuclear DNA was stained with DAPI for 3 min. The migratory ability of SEECs was determined using a wound-healing assay. Confluent cell monolayers in a 24-well plate were scratched with a 100 μl pipette tip. The cells were successively imaged under an inverted microscope and the healed area was quantified using ImageJ software. The migratory capacity was calculated as the percentage of the scratched area healed at successive times relative to time 0. Cell proliferation was assessed using a CCK-8 viable cell counting kit (Biosharp, Shanghai, China). SEECs were seeded in 96-well plates (~2000 cells/well), and 10 μl of CCK-8 reagent was added to each well after drug treatment. After incubation for 2 h at 37ºC, the absorbance was measured at 450 nm using a spectrophotometer. Six replicates were performed for each treatment group, and each experiment was repeated thrice. SEECs were inoculated into 24-well plates and treated as described in Section 2.1. ELISA kits (Jianglai, Shanghai, China) were used to determine the concentrations of PGF 2α and PGE 2 in the collected media. Data were processed according to the manufacturer’s instructions. Tissue and cell samples were washed thrice with PBS. RIPA lysis buffer (Beyotime, Shanghai, China), PMSF (Solarbio, 1:100), and phosphatase inhibitors (Beyotime, 1:50) were added, and samples were disrupted on ice. Protein concentrations were determined using a BCA kit, and the proteins were separated using SDS-PAGE (Bio-Rad, Hercules, CA, USA). Semi-dry blotting was used to transfer the proteins to PVDF membranes (Solarbio). The blots were blocked with 5% skim milk for 1 h and incubated overnight at 4ºC with the primary antibody. After washing, the secondary antibody was added, and the blots were placed on a shaking table at room temperature for 1 h. The antibodies used in these experiments are listed in Supplementary Table 2 . Blot documentation and analysis were performed using the Odyssey infrared imaging system (Li COR, Nebraska, NE, USA). All data were analyzed using GraphPad Prism 8 or SPSS17.0 (SPSS Co., Chicago, IL, USA). Differences between the means were determined using a t -test or one-way ANOVA, followed by a Bonferroni’s post-hoc test for multiple comparisons. All data are expressed as mean ± standard error. The differences were considered significant at P < 0.05, and extremely significant at P < 0.01.