Introduction

The placenta plays an essential role in sustaining pregnancy and supporting fetal growth. Trophoblast cells differentiate by either fusing into multinucleated syncytiotrophoblasts or developing into invasive extravillous trophoblasts1. Syncytiotrophoblasts facilitate gas and nutrient exchange, secrete a produces a variety of bioactive factors, and act as an immunological barrier between the fetus and the maternal immune system2. Human chorionic gonadotropin (hCG), primarily produced by syncytiotrophoblasts, is crucial for maintaining pregnancy, and contribute to progesterone production, synthesis, angiogenesis, and the differentiation of cytotrophoblast3. Impaired trophoblast syncytialization has been implicated in pregnancy complications such as hypertensive disorders of pregnancy (HDP) and fetal growth restriction (FGR)4.

Ferroptosis is a distinct, non-apoptotic form of cell death characterized by the iron-dependent accumulation of lipid peroxides5. Morphologically, ferroptotic cells exhibit features resembling necrosis, including cytoplasmic swelling and plasma membrane rupture, that distinguish them from apoptosis and necroptosis6. The primary mechanism driving ferroptosis involves dysfunction of the cystine/glutamate-glutathione (GSH)—glutathione peroxidase 4 (GPX4)-dependent antioxidant defense system5,6. Cystine is imported via the cystine/glutamate antiporter (xCT), reduced to cysteine, and subsequently utilized for GSH synthesis, GPX4, a phospholipid hydroperoxidase, reduces lipid peroxides generated by reactive oxygen species (ROS) using GSH as a cofactor7. When this defense pathway is disrupted, Fe2⁺-mediated Fenton reactions accelerate lipid peroxidation cascades, ultimately culminating in ferroptotic cell death8.

The transcription factor nuclear factor erythroid 2-related factor 2 (NRF2) serves as a master regulator of the cellular antioxidant response and plays a pivotal role in the control of ferroptosis9. Under basal conditions, NRF2 is sequestered in the cytoplasm by Kelch-like ECH-associated protein 1 (KEAP1) dimers, which suppress its transcriptional activity10. Heme oxygenase 1 (HO-1), also referred as heat shock protein-32 (Hsp32), is an inducible enzyme involved in iron metabolism and antioxidant defense11. However, excessive expression of HO-1 has paradoxical effects, promoting the Fenton reaction, depleting intracellular GSH, and enhancing lipid peroxidation, leading to ferroptosis12.

The endoplasmic reticulum (ER) is an organelle involved in an lipid and steroid synthesis, protein folding and maturation, calcium storage, and detoxification13. Disruption of ER homeostasis leads to the accumulation of misfolded or unfolded proteins, thereby causing ER stress14. ER stress activates multiple signaling pathways, collectively known as the unfolded protein response (UPR), which can result in cellular dysfunction and death13,14. Recent studies suggest that ER stress plays a crucial role in the process of ferroptosis15. ER stress-induced ferroptosis has been implicated in various diseases, including diabetes, ulcerative colitis, and lung injury16,17,18. In reproductive disorders, ferroptosis has been associated with pregnancy complications such as HDP, preeclampsia, gestational diabetes mellitus, and preterm birth19,20. However, the physiological role of ferroptosis in the fusion of placental trophoblast cells remains unclear. Therefore, the present study aimed to investigate the effects of ferroptosis-related signaling on trophoblast cell fusion, using the BeWo trophoblast cell line.

Results

Ferroptosis inducers potentiate the syncytialization of trophoblasts

The effects of RSL3 and ML-210, which are glutathione peroxidase 4 (GPX4) inhibitors, or erastin, an inhibitor of the cystine/glutamate antiporter, on the viability of trophoblast cells was examined. While low concentrations of the ferroptosis inducers did not affect the viability of BeWo cells, high concentrations of RSL3 (100 nM), ML-210 (200 nM), or erastin (100, 200 μM) significantly reduced cell viability (Fig. 1A). To further investigate their effects on trophoblast syncytialization, BeWo cells were treated with RSL3, ML-210, or erastin in the presence of forskolin (FSK), a known inducer of syncytialization. ML-210 enhanced FSK-induced hCGβ expression, whereas erastin reduced it (Fig. 1B). Furthermore, both RSL3 and ML-210 increased the number of fusogenic cells (Fig. 1C).

Fig. 1
Fig. 1
Full size image

Ferroptosis inducers enhance trophoblasts syncytialization. (A) BeWo cells were treated with forskolin (FSK, 2.5 µM) and RSL3 (10, 50, or 100 nM), ML-210 (5, 20, 100, or 200 nM), or erastin (5, 10, 25, 50, 100, or 200 µM) for 48 h. CCK-8 assay analysis of cell viability. (B–D) BeWo cells treated with forskolin (FSK, 2.5 µM) and RSL3 (50 nM), ML-210 (100 nM), or erastin (50 µM) for 48 h. (B) Immunoblotting showing hCGβ and GPX4 protein levels in lysates from FSK-treated BeWo cells. β-actin served as the loading control. Representative data from three independent experiments are shown. The graph shows hCGβ and GPX4 levels normalized to β-actin levels from three independent experiments. Data are presented as mean ± SEM. ###P < 0.001 vs. Ctrl, *P < 0.05, **P < 0.01, ***P < 0.001 vs. FSK. (C) Cells were immunostained with anti-E-cadherin antibody (green) and DAPI (blue) to visualize syncytialization. A representative images from three independent experiments and quantification of fused cells (n = 5) are shown; syncytialized cells are outlined with stippled lines. Scale bar = 100 µm. Values are represented as mean ± SEM of three independent experiments. #P < 0.05 vs. Ctrl, ***P < 0.001 vs. FSK.

GPX4 inhibitors promote the syncytialization via the increase in intracellular Fe2⁺

To investigate whether GPX4 inhibitors increase intracellular Fe2⁺ and lipid peroxide levels, which are characteristic of ferroptosis5, during FSK-stimulated syncytialization, we first examined the effects of RSL3 and ML-210. FSK alone increased intracellular Fe2⁺ levels, which were further potentiated by RSL3 or ML-210 (Fig. 2A). While RSL3 or ML-210 tended to increase lipid peroxide levels (Fig. 2B and C). To evaluate the functional significance of Fe2+ accumulation and lipid peroxidation in syncytialization, BeWo cells were treated with deferoxamine (DFO), an iron chelator or ferrostatin-1(Fer-1), an inhibitor of lipid peroxidation. Both DFO and Fer-1 attenuated the FSK/ML-210-induced upregulation of hCGβ (Fig. 2D), and suppressed the increase in the number of fusogenic cells induced by the same treatment (Fig. 2E). Furthermore, N-acetyl-L-cysteine (NAC), an antioxidant, similarly inhibited hCGβ expression and reduced the number of fusogenic cells in the FSK/ML-210 treated cultures (Fig. 2F and G).

Fig. 2
Fig. 2
Full size image

Ferroptosis inducers promote syncytialization by increasing intracellular Fe2⁺ levels. BeWo cells were treated with FSK (2.5 µM) and RSL3 (50 nM) and ML-210 (100 nM) for 48 h. (A) Flow cytometry analysis Fe2⁺ levels. (B) Fluorescence micrographs showing lipid peroxidation. Cells were immunostained with anti-E-cadherin antibody (green), DAPI (blue) and 4-hydroxynonenal (red). Scale bar = 20 µm. (C) Quantification of the level of lipid peroxide determined using Image J (n = 3). (D) BeWo cells were treated with FSK (2.5 µM), ML-210 (100 nM), DFO (250 µM) or Fer-1(0.5 µM) for 48 h. Changes in hCGβ expression were evaluated by immunoblotting. β-actin served as a loading control. Representative data from three independent experiments are shown. Graphs show hCGβ levels normalized to β-actin levels. Values represent mean ± SEM. ###P < 0.001 vs. Ctrl, ***P < 0.001 vs. FSK, ††P < 0.05, †††P < 0.001 vs. ML + FSK. (E) Cells were immunostained with anti-E-cadherin antibody (green) and DAPI (blue) to visualize syncytialization. A representative image from three independent experiments is shown, with syncytialized cells indicated by a stippled line. Scale bar = 100 µm. The number of fused cells was quantified. Values are represented as mean ± SEM of three independent experiments. #P < 0.05 vs. Ctrl, ***P < 0.001 vs. FSK, †††P < 0.001 vs. ML + FSK. (F) BeWo cells were treated with FSK (2.5 µM), ML-210 (100 nM) and NAC (5 mM) for 48 h. Changes in hCGβ expression were assessed by immunoblotting. β-actin was used as a loading control. Representative results from three independent experiments are shown. The graph shows hCGβ levels normalized to β-actin levels from three independent experiments. #P < 0.05 vs. Ctrl, *P < 0.05 vs. FSK, †P < 0.05 vs. ML + FSK. Values represent mean ± SEM. (G) Cells were immunostained with anti-E-cadherin antibody (green) and DAPI (blue) to visualize syncytialization. A representative image from three independent experiments is shown, with syncytialized cells indicated by a stippled line. Scale bar = 100 µm. The number of fused cells was quantified. Values are represented as mean ± SEM of three independent experiments. ##P < 0.01 vs. Ctrl, **P < 0.01 vs. FSK, †††P < 0.001 vs. ML + FSK.

GPX4 inhibitor induces oxidative stress and ER stress

To investigate the transcriptional effects of GPX4 inhibition on FSK-induced syncytialization, we performed RNA-seq analysis. This identified 2,785 differentially expressed genes (DEGs), of which 1306 were upregulated and 1379 were downregulated (Fig. 3A). Pathway and Gene Ontology (GO) analyses of molecular functions revealed that the upregulated DEGs were enriched in oxidative stress-related terms (Figs. 3B and S1, and Table S1). Additionally, GO analyses of biological processes and cellular components indicated enrichment in ER stress-associated terms (Figs. 3C and S1, and Table S1). ML-210 increased the expression of oxidative stress-related factors: NRF2, HO-1, and KEAP1 (Fig. 4A) and ER stress-related factors: HSPA5, ATF6, sXBP1, and ATF4 in FSK-treated cells (Fig. 4B). To further explore the involvement of ER stress in ferroptosis induction, BeWo cells were co-treated with GSK2606414 (GSK; PERK inhibitor), Kira6 (IRE1α inhibitor), or AEBSF (ATF6 inhibitor). Kira6 and AEBSF reduced both hCGβ expression and the number of fusogenic cells, whereas GSK had no significant effect (Fig. 4C and D).

Fig. 3
Fig. 3
Full size image

Gene expression profiling in ML-210-treated BeWo cells. BeWo cells treated with FSK (2.5 µM) and ML-210 (100 nM) for 48 h. Total RNA was extracted and subjected to RNA-seq. Differential gene was compared between the FSK and FSK and ML-210 treatment groups. (A) Volcano plot showing the expression of transcripts identified by RNA-seq. Transcripts highlighted in red or blue were differentially expressed by more than twofold changes (FDR < 0.05). (B, C) Differentially expressed genes were functionally classified using enriched pathway analysis or GO analysis for molecular function, biological processes, and cellular components.

Fig. 4
Fig. 4
Full size image

Ferroptosis inducer triggers oxidative stress and ER stress. BeWo cells were treated with FSK (2.5 µM) and ML-210 (100 nM) for 48 h. (A) Expression levels of oxidative stress-related genes (NRF2, HO-1, and KEAP1) were measured by qPCR (n = 3). (B) Expression of ER stress-related genes (HSPA5, ATF6, DDIT3, sXBP1, ATF4, and CARL) was measured using qPCR (n = 3), with 18S rRNA used as the reference gene. (C) BeWo cells were treated with FSK (2.5 µM) and ML-210 (100 nM), in the presence or absence of GSK2606414 (GSK, 250 nM), Kira6 (1 μM), or AEBSF (300 μM) for 48 h. Changes in hCGβ and Bip expression were assessed by immunoblotting. β-actin served as the loading control. Representative data from three independent experiments are shown. The graph shows hCGβ and Bip levels normalized to β-actin levels from three independent experiments. ##P < 0.01 vs. Ctrl, *P < 0.05 vs. FSK, †P < 0.05 vs. ML + FSK. Values represent mean ± SEM. (D) Cells were immunostained with anti-E-cadherin antibody (green) and DAPI (blue) to visualize syncytialization. A representative image from three independent experiments is shown, with syncytialized cells indicated by a stippled line. Scale bar = 100 µm. Quantification of the number of fused cells. Values are represented as mean ± SEM of three independent experiments. #P < 0.05 vs. Ctrl, *P < 0.05 vs. FSK, †P < 0.05 vs. ML + FSK.

Discussion

This study showed that ferroptosis signaling may be involved in the physiological syncytialization of trophoblasts. GPX4 inhibitors RSL3 and ML-210 enhanced FSK-induced hCG expression and cell fusion, while the cystine/glutamate antiporter inhibitor erastin did not affect those. Intracellular Fe2+ and lipid peroxide levels were increased during trophoblast syncytialization, Notably, DFO and Fer-1 inhibited FSK-induced hCG expression. These results indicate that ferroptosis signaling mediated by GPX4 inhibition, but does not depend on the cystine/glutamate antiporter, accelerated the process of trophoblast fusion dependent on the level of intracellular Fe2+ and lipid peroxide. The upregulated DEGs in ML-210 treated cells identified by RNA-seq analysis were involved in oxidative and ER stresses. GPX4 inhibitors increased the expression of oxidative stress-related factors NRF2, HO-1, and KEAP1, and ER stress-related factors HSPA5, ATF6, sXBP1, and ATF4. These data suggest that GPX4 inhibition may promote cell fusion by inducing oxidative stress-related ER stress (Fig. 5).

Fig. 5
Fig. 5
Full size image

Schematic illustration of the effect of GPX4 inhibitors in trophoblasts during syncytialization.

Iron requirements increase during pregnancy to support maternal erythropoiesis and fetal growth and development21. The placenta plays a pivotal role in iron transport between mother and fetus during pregnancy20. In both humans and rodents, maternal circulating iron is transferred directly to the fetus via syncytiotrophoblast cells22. Iron deficiency during pregnancy can cause problems for both maternal health and fetal development, while excess iron generates ROS, resulting in activation of oxidative stress and ferroptosis. Thus, iron homeostasis is essential for both maternal health and fetal development. In this study, we found that FSK increased intracellular Fe2⁺ levels in trophoblasts, indicating the increase in susceptibility to ferroptosis during syncytialization, whereas the involvement of ferroptosis under physiological conditions of placentation has not been identified. It has been noted that iron requirements are greatly increased and syncytiotrophoblast transferrin is increased during pregnancy22. These findings suggest that dependent on intracellular Fe2+ -dependent ferroptosis signaling may contribute to physiological placentation.

The cAMP signaling is the primary pathway that induces syncytialization. The cAMP functions as an intracellular second messenger and activates protein kinase A (PKA) and exchange protein directly activated by cAMP (EPAC), which upregulates hCG expression3,23. However, RNA-seq analysis showed that cAMP-related signaling was not enriched among the upregulated enriched pathway and GO terms in ML-210 treated cells. This indicates that ferroptosis signaling may contribute to cell fusion through a mechanism that does not engage the cAMP pathway. The analysis also showed that several DEGs downregulated by GPX4 inhibitors were genes related to the cell cycle, including p53 (data not shown). The cell cycle arrests at the G1 or G2 phase in mononuclear cytotrophoblasts, which then fuse with neighboring cells to form syncytiotrophoblasts24. Syncytiotrophoblasts irreversibly acquire cell cycle arrest during cell fusion. Thus, the downregulation of cell cycle-related genes induced by GPX4 inhibitors further supports the possibility that ferroptosis signaling may be involved in cell fusion.

The ER stress signaling via the unfolded protein response (UPR) pathway has been reported to play an important role in mammalian reproduction, including placentation25. However, excessive ER stress and overactivation of UPR pathway can lead to apoptosis, ferroptosis, and autophagy, which may adversely affect placentation26. ER stress is associated with pregnancy complications, including preeclampsia and FGR27. Excessive ER stress may lead to a disorder of reproductive functions, but adequate ER stress may contribute to syncytialization25. Our previous study showed that ROS was generated during syncytialization28, suggesting that ER stress-induced ROS generation may contribute to cell fusion. In this study, GPX4 inhibitor increased the expression of ER stress signaling-related genes, and inhibition of either IRE1α or ATF6 decreased the expression of hCGβ induced by ML-210. Based on in silico analysis using the JASPAR database, we hypothesize that the transcription factors ATF6 and XBP1, a downstream factor of IRE1α, may bind to the promoter region of the hCG gene. While this prediction suggests a possible regulatory interaction, further experimental validation is required. These findings raise the possibility that GPX4 inhibitor may regulate hCGβ expression through IRE1α and ATF6 signaling during syncytialization.

In conclusion, the present study indicates that GPX4 inhibitors increase intracellular Fe2⁺ levels, which induce ER stress and oxidative stress and stimulate hCGβ expression via the UPR-related transcription factors ATF6 and XBP1 during syncytialization. Thus, ferroptosis signaling may regulate physiological trophoblast fusion.

This study has several limitations. First, although we observed significant increases in hCGβ expression and morphological changes induced by GPX4 inhibitors, the precise molecular mechanisms by which altered Fe2⁺ levels lead to ER and oxidative stress remain to be clarified. Furthermore, the specific contributions of oxidative and ER stress to cell fusion, as well as the underlying mechanisms, have yet to be elucidated remain unclear. Second, the findings are based solely on in vitro experiments using BeWo cells; further validation using primary trophoblasts, placental explants, or in vivo models is warranted. Third, although our focus was on GPX4 signaling, other pathways may also contribute to the observed effects. Further investigations are needed to fully elucidate the role of ferroptosis in trophoblast differentiation and its broader implications for placental biology.

Materials and methods

Cell culture

The human choriocarcinoma BeWo cell lines (JCRB Cell Bank, Osaka, Japan) were grown in 1:1 Ham’s F12/Dulbecco’s modified Eagle’s medium (Fujifilm Wako Pure Chemical Corp., Osaka, Japan) supplemented with 10% fetal bovine serum, antibiotics, and antimycotics, and the cells were grown at 37 °C in humidified air containing 5% CO229. The cells were treated for 48 h with FSK (an adenylate cyclase activator, 2.5 µM, Cayman chemical, Ann Arbor, MI USA) and /or ferroptosis inducers: RSL3 (Selleck Biotech, Tokyo, Japan), ML-210 (Selleck Biotech), or erastin (Selleck Biotech); ferroptosis inducers: DFO (an iron chelator, Cayman Chemical) and Fer-1 (an inhibitor of lipid peroxidation, Cayman Chemical); ER stress inhibitors: GSK (a PERK inhibitor, Selleck Biotech), Kira6 (an IRE1α inhibitor, Selleck Biotech), or AEBSF (an ATF6 inhibitor, Selleck Biotech); or NAC (Sigma-Aldrich, Tokyo, Japan).

Cell viability assay

Cell viability was examined using the WST-8 assay (cell counting kit-8; Dojindo, Kumamoto, Japan). BeWo cells (1.6 × 104) were seeded into 96-well plates and then treated with FSK (2.5 µM), and RSL3 (50 nM), ML-210 (100 nM), or erastin (50 μM) for 48 h. After incubating in WST-8, the staining intensity of the medium was measured by absorbance at 460 nm in a spectrophotometer. Data were normalized to the control value.

Western blotting

BeWo cells were lysed with RIPA buffer (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. The constituent proteins were separated by SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Bio-Rad Laboratories, Hercules, CA, USA) using a Trans-Blot Turbo (Bio-Rad). After blocking with Bullet Blocking One (Nacalai Tesque, Kyoto, Japan), the membranes were incubated with primary antibodies against hCG b (1:5000, BC022796; ProteinTech, Chicago, IL, USA), GPX4 (1:5000, P36969; Cell signaling Technology, Danvers, MA, USA), Bip (encoded by the HSPA5 gene; 1:2000, Cell Signaling Technology) or β-actin (1:5000, ab1801; Abcam, Cambridge, UK). Immunoreactive bands were detected using enhanced chemiluminescence (Merck Millipore, Burlington, MA, USA) after incubation with horseradish peroxidase-labeled goat anti-rabbit or anti-mouse IgG (1:5000; Vector Laboratories, Burlingame, CA, USA). Signals were detected using a C-DiGit Blot Scanner (LI-COR, Lincoln, NE, USA), and the relative band density was quantified using Image Studio DiGit software (version 5.2)30.

Cell fusion assay

BeWo cells were fixed with methanol and incubated with anti-E-cadherin antibody (1:200, Cell signaling Technology) and AlexaFluor 594-conjugated goat anti-mouse antibody (Thermo Fisher Scientific) to distinguish cell surfaces. The nuclei were counterstained with 4′,6-diamino-2-phenylindole 2HCl (DAPI). Multinucleated cells were counted in five randomly selected microscopic fields per sample across three independent experiments 54. In the same fields, the number of nuclei within syncytiotrophoblasts and the total number of nuclei were recorded, and the fusion index was calculated as follows:31

$${\text{Fusion index }}\left( \% \right) \, = \, \left( {{\text{number of nuclei in syncytia}}/{\text{ total number of nuclei}}} \right)\, \times \,{1}00$$

The data are presented as ratios relative to the control and shown as mean ± SEM from three independent experiments.

Measurement of Fe2⁺ content

BeWo cells (1.6 × 104) were treated with RSL3 (50 nM) and ML-210 (100 nM) in the presence of FSK (2.5 µM) for 48 h. The cells were incubated with PBS for 20 min and then were washed using HBSS. The suspended cells were added 300 μL of 1 μM FerroOrange (Dojindo) diluted with PBS. After the cells were incubated at 37 °C for 30 min, the changes in the content of Fe2⁺ were detected using flow cytometry32.

Lipid peroxide assay

BeWo cells were fixed with methanol and incubated with anti-E-cadherin antibody (1:200, Cell signaling Technology), anti-4-hydroxy-2-nonenal (4-HNE) monoclonal antibody (1:100, MHN-100P, JalCA, Shizuoka, Japan), AlexaFluor 488-conjugated goat anti-rabbit IgG antibody (Thermo Fisher Scientific) and AlexaFluor 594-conjugated goat anti-mouse IgG antibody (Thermo Fisher Scientific) to distinguish cell surfaces. The nuclei were counterstained with 4′,6-diamino-2-phenylindole 2HCl (DAPI). The levels of lipid peroxide were detected using BZX800 fluorescence microscope (Keyence). The levels of lipid peroxide in three randomly selected microscopic areas per sample were quantified with ImageJ.

RNA extraction and quantitative RT-PCR

RNA was extracted using the RNeasy Mini Kit (Qiagen, Tokyo, Japan), according to manufacturer’s instructions. The mRNA was reverse transcribed to cDNA using a ReverTra Ace qPCR RT Kit (Toyobo, Osaka, Japan), and the cDNA produced was subjected to qPCR amplification (PowerUP SYBR Green Master Mix, Thermo Fisher Scientific). The primers used are listed in Table S2. Calibration curves were used to determine the amplification of each target gene with respect to the expression of a reference genes, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and 18S. The mean crossing threshold (Ct) values for each target were calculated using Sequence Detection System software v2.3 (Thermo Fisher Scientific)33.

RNA sequencing (RNA-seq), gene ontology (GO), and pathway analyses

RNA-seq analysis was performed with RNA extracted from cultured BeWo cells. High-throughput sequencing libraries were prepared using a TruSeq Stranded mRNA LT Sample Prep Kit (Illumina, San Diego, CA, USA), according to the manufacturer’s instructions, and data analysis was performed by BGI Genomics (Shenzhen, China). Primary sequencing data were deposited to the DNA Data Bank of Japan (DDBJ) Sequence Read Archive (https://www.ddbj.nig.ac.jp/dra/index-e.html) (accession numbers DRA021539). Data analysis was performed as described previously8. Briefly, trimmed sequences were analyzed using the STAR/RSEM/edgeR pipeline, the human genome (hg38), and reference annotations obtained from the UCSC genome browser (https://genome.ucsc.edu). Significantly differentially expressed genes (DEGs) were identified based on CPM (count per million) levels. GO and enriched signaling pathway analyses were performed using the Enrichr tool (http://amp.pharm.mssm.edu/Enrichr/).

Statistical analysis

Data are expressed as mean ± SEM, and were compared using the Dunnett’s test. A P-value < 0.05 was considered to be statistically significant. Statistical testing was performed using the R software (ver.4.0.5; www.r-project.org). In RNA-seq analysis, a false discovery rate-adjusted P-value (q-value) < 0.05 was considered statistically significant.