Influenza a(H1N1) virus infection impairs syncytiotrophoblast functions by promoting ferroptosis through the TNF-α/ISG15/EGR1 axis.

To simulate STB cells in vitro, we established an in vitro syncytiotrophoblast model by treating BeWo cells with 50 µM FSK for 48 h. Our previous research indicated that infecting the cells with the H1N1 (A/Sichuan/1/2009) virus at a multiplicity of infection (MOI) of 1 did not impact the viability of FSK-treated BeWo cells [ 14 ]. Consequently, we employed H1N1 virus at this concentration to infect both the fused BeWo cells and unfused BeWo cells. The findings revealed a significant downregulation of ERVW1 and CGB3 mRNA, accompanied by reduced β-HCG expression and impaired cell fusion capacity. The results demonstrated that the model of H1N1-induced trophoblast cell fusion impairment had been successfully established (Figure S1A-C). To investigate the impact of H1N1 virus infection on STB cells, RNA sequencing was conducted on FSK-treated BeWo cells with and without H1N1 virus infection. Differentially expressed genes (DEGs) were identified based on the criteria of Q value < 0.05 and |logFC|≥ 2.0 (Fig.  1 A). KEGG pathway analysis revealed that the upregulated DEGs were predominantly enriched in the programmed cell death pathway (Fig.  1 B). To acquire a complete understanding of the relative contributions of different types of cell death in FSK-treated BeWo cells infected with H1N1 virus, we analyzed gene markers associated with various forms of programmed cell death, including ferroptosis (GPX4, SLC7A11), necroptosis (MLKL, RIPK3), apoptosis (BAX, BCL2), autophagy (LC3B, p62), and pyroptosis (Caspase-1, NLRP3). Among these, the mRNA levels of GPX4 and SLC7A11 in FSK-treated BeWo cells infected with H1N1 virus were notably lower than in uninfected controls (Fig.  1 C, S2A-H). Given that research on the role of ferroptosis in H1N1-infected trophoblast cells, particularly in the context of syncytiotrophoblast dysfunction, remains limited, we chose to focus our current study on ferroptosis. Fig. 1 Programmed cell death occurs in FSK-treated BeWo cells with H1N1 virus infection. ( A ) Volcano plot showing differentially expressed genes (DEGs) between FSK-treated BeWo cells with and without H1N1 virus infection (n = 3 per group). The adjusted p-value < 0.05 was set as the threshold for screening DEGs. The red, green, and gray dots indicate upregulated, downregulated, and unchanged genes, respectively. ( B ) KEGG pathway enrichment analysis of the DEGs. The top 20 enriched pathways are shown. The vertical axis shows the functional classification, and the horizontal axis shows the rich ratio. The colors of the dots represent the enrichment significance, and dot sizes represent gene counts. Q-value ≤ 0.05 as significantly enriched. The smaller the Q-value, the closer the color is to red. ( C– G ) RT-qPCR assays showing different the expression of different gene markers reflecting various programmed cell death forms, including ferroptosis (GPX4, SLC7A11), necroptosis (MLKL, RIPK3), apoptosis (BAX, BCL2), autophagy (LC3B, p62) and pyroptosis (Caspase-1, NLRP3) in FSK-treated BeWo cells infected with H1N1 virus (n = 3–5). Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. ns, p > 0.05; *, p < 0.05; ****, p < 0.0001 Programmed cell death occurs in FSK-treated BeWo cells with H1N1 virus infection. ( A ) Volcano plot showing differentially expressed genes (DEGs) between FSK-treated BeWo cells with and without H1N1 virus infection (n = 3 per group). The adjusted p-value < 0.05 was set as the threshold for screening DEGs. The red, green, and gray dots indicate upregulated, downregulated, and unchanged genes, respectively. ( B ) KEGG pathway enrichment analysis of the DEGs. The top 20 enriched pathways are shown. The vertical axis shows the functional classification, and the horizontal axis shows the rich ratio. The colors of the dots represent the enrichment significance, and dot sizes represent gene counts. Q-value ≤ 0.05 as significantly enriched. The smaller the Q-value, the closer the color is to red. ( C– G ) RT-qPCR assays showing different the expression of different gene markers reflecting various programmed cell death forms, including ferroptosis (GPX4, SLC7A11), necroptosis (MLKL, RIPK3), apoptosis (BAX, BCL2), autophagy (LC3B, p62) and pyroptosis (Caspase-1, NLRP3) in FSK-treated BeWo cells infected with H1N1 virus (n = 3–5). Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. ns, p > 0.05; *, p < 0.05; ****, p < 0.0001 To experimentally confirm that H1N1 virus infection induces ferroptosis and disrupts trophoblast cell fusion, we evaluated several key indicators of ferroptosis: intracellular iron contents, MDA, and typical ferroptosis biomarkers (GPX4, TFRC, and SLC7A11). We also measured lipid peroxidation levels using C11 BODIPY 581/591 dye. MDA concentrations and intracellular iron contents were significantly higher in the H1N1-infected cells than the FSK-treated BeWo cells (Fig.  2 A–B). Moreover, the C11 BODIPY 581/591 dye indicated a remarkable elevation in the intracellular lipid peroxidation levels (Fig.  2 C). Additionally, we detected decreased GPX4 protein levels and increased TFRC protein levels in the H1N1-infected, FSK-treated BeWo cells (Fig.  2 D). To visualize the effects of ferroptosis, we prepared ultrathin sections of the cells. TEM examinations revealed that H1N1-infected cells showed typical ferroptosis characteristics: shrunken mitochondria with condensed membrane densities, loss of structural integrity, and breakdown of cristae, abruption of the membrane (Fig.  2 E). In contrast, the addition of ferrostatin-1 (Fer-1), a potent inhibitor of ferroptosis, suppressed ferroptosis induced by H1N1 virus infection in the cells (Figure S3A–E). These findings clearly demonstrate that H1N1 virus infection facilitates ferroptosis in FSK-treated BeWo cells. Subsequently, to further explore the impacts of ferroptosis on trophoblast cell fusion, we comprehensively evaluated the protein expression and gene levels associated with syncytiotrophoblast functions using Western blot analysis and RT-qPCR. The findings indicate that the expression levels of ERVW1 and CGB3 mRNA, along with the β-hCG protein levels, were significantly elevated in FSK-treated BeWo cells infected with H1N1 virus and treated with Fer-1 (Fig.  2 F–G). We assessed syncytiotrophoblast (STB) morphology by immunofluorescence staining, using E-cadherin (a cell surface marker of intercellular boundaries) and DAPI for nuclear counterstaining. As shown in Fig.  2 H, immunofluorescence analysis revealed disrupted syncytialization in H1N1-infected cells, and this impairment was significantly mitigated by Fer-1 treatment. Collectively, these data demonstrate that H1N1 infection disrupts trophoblast cell fusion by promoting ferroptosis. Fig. 2 H1N1 virus infection induces ferroptosis and disrupts BeWo cell fusion. MDA concentrations (n = 5) ( A ), intracellular iron contents (n = 7) ( B ), and lipid peroxidation levels (n = 3) ( C ) in FSK-treated BeWo cells infected with H1N1 virus. Scale bars represent 150 µm. ( D ) Western blot examination of ferroptosis-related proteins (TFRC, GPX4). β-Actin served as an internal standard (n = 3–4). ( E ) Representative transmission electron micrographs showing the mitochondrial morphology in FSK-treated BeWo cells infected with H1N1 virus. Mitochondria in H1N1-infected cells appear shrunken, showing condensed membrane densities, compromised structural integrity, and deteriorated cristae. Scale bars represent 2 µm. ( F ) RT-qPCR assays showing ERVW1 and CGB3 expression in FSK-treated BeWo cells infected with H1N1 virus and treated with Fer-1 (n = 3). ( G ) The level of β-hCG was analyzed by Western blotting. β-Actin served as an internal control (n = 3). ( H ) Immunofluorescence staining results showed that Fer-1 alleviated the negative effects of H1N1 virus infection on BeWo cell fusion. E-cadherin was examined by immunofluorescence (green), and nuclei were stained with DAPI (blue). Representative images are shown (20 X images were obtained), and images are shown as merged channels (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. ns, p > 0.05; *, p < 0.05; ***, p < 0.001, or ****, p < 0.0001. H1N1 virus infection induces ferroptosis and disrupts BeWo cell fusion. MDA concentrations (n = 5) ( A ), intracellular iron contents (n = 7) ( B ), and lipid peroxidation levels (n = 3) ( C ) in FSK-treated BeWo cells infected with H1N1 virus. Scale bars represent 150 µm. ( D ) Western blot examination of ferroptosis-related proteins (TFRC, GPX4). β-Actin served as an internal standard (n = 3–4). ( E ) Representative transmission electron micrographs showing the mitochondrial morphology in FSK-treated BeWo cells infected with H1N1 virus. Mitochondria in H1N1-infected cells appear shrunken, showing condensed membrane densities, compromised structural integrity, and deteriorated cristae. Scale bars represent 2 µm. ( F ) RT-qPCR assays showing ERVW1 and CGB3 expression in FSK-treated BeWo cells infected with H1N1 virus and treated with Fer-1 (n = 3). ( G ) The level of β-hCG was analyzed by Western blotting. β-Actin served as an internal control (n = 3). ( H ) Immunofluorescence staining results showed that Fer-1 alleviated the negative effects of H1N1 virus infection on BeWo cell fusion. E-cadherin was examined by immunofluorescence (green), and nuclei were stained with DAPI (blue). Representative images are shown (20 X images were obtained), and images are shown as merged channels (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. ns, p > 0.05; *, p < 0.05; ***, p < 0.001, or ****, p < 0.0001. Studies have shown that TNF-α can regulate trophoblast cell differentiation [ 27 ]. Nevertheless, the precise underlying mechanisms remain elusive. In this study, we delved into the alterations in TNF-α expression during the disruption of BeWo cell fusion induced by H1N1 virus infection. Our findings revealed that the mRNA level of TNF-α was elevated in FSK-treated BeWo cells with H1N1 virus infection (Fig.  3 A). This finding led us to hypothesize that H1N1 virus infection may disrupt BeWo cell fusion by enhancing the expression of TNF-α. Given that H1N1 infection induces ferroptosis and is associated with increased TNF-α levels, we hypothesized that H1N1-induced ferroptosis in STB cells could be mediated by TNF-α, leading to disruption of trophoblast cell fusion. To test this hypothesis, we transfected siRNA into FSK-treated BeWo cells infected with H1N1 virus. Among the three siRNAs tested, siTNF-α−1 showed the strongest silencing effect (Figure S4). We eventually selected siTNF-α−1 for further investigation. To clarify the regulatory interaction among TNF-α, ferroptosis, and syncytialization, the key indicators of ferroptosis and markers characteristic of syncytialization were subsequently assessed. Compared to FSK-treated BeWo cells with H1N1 infection, the additional treatment with siTNF-α significantly reduced the MDA concentrations, intracellular iron contents, and lipid peroxidation levels (Fig.  3 B–D). Meanwhile, we observed an increase in the levels of GPX4 and SLC7A11, along with a decrease in the TFRC level (Fig.  3 E, S5A). As expected, in cells treated with siTNF-α, the expression of ERVW1 and CGB3 at the mRNA level and that of β-hCG at the protein levels were elevated (Fig.  3 F, S5B). Immunofluorescence analysis revealed that cell fusion was disrupted in the H1N1-infected group. Strikingly, this disruption was effectively reversed by the siTNF-α treatment (Fig.  3 G). These findings suggest that reducing TNF-α levels inhibits ferroptosis caused by H1N1 virus infection, thereby alleviating the impairment of trophoblast cell fusion. Fig. 3 Knockdown of TNF-α alleviates the damage of ferroptosis to BeWo cell fusion. ( A ) The mRNA level of TNF-α in FSK-treated BeWo cells infected with H1N1 virus was detected by RT-qPCR (n = 3). Analysis of MDA concentrations (n = 5) ( B ), intracellular iron contents (n = 5) ( C ), and lipid peroxidation levels (n = 3) ( D ) following transfection with siTNF-α. Scale bars represent 150 µm. ( E – F ) The protein levels of GPX4, TFRC, and β-hCG were detected by Western blot following transfection with siTNF-α (n = 3–4). ( G ) Immunofluorescence staining revealed the knockdown of TNF-α alleviated the detrimental effects of H1N1 virus infection on BeWo cell fusion (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 Knockdown of TNF-α alleviates the damage of ferroptosis to BeWo cell fusion. ( A ) The mRNA level of TNF-α in FSK-treated BeWo cells infected with H1N1 virus was detected by RT-qPCR (n = 3). Analysis of MDA concentrations (n = 5) ( B ), intracellular iron contents (n = 5) ( C ), and lipid peroxidation levels (n = 3) ( D ) following transfection with siTNF-α. Scale bars represent 150 µm. ( E – F ) The protein levels of GPX4, TFRC, and β-hCG were detected by Western blot following transfection with siTNF-α (n = 3–4). ( G ) Immunofluorescence staining revealed the knockdown of TNF-α alleviated the detrimental effects of H1N1 virus infection on BeWo cell fusion (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 To further investigate how TNF-α disrupts trophoblast cell fusion through ferroptosis, we overexpressed TNF-α and observed significant increases in MDA concentrations, intracellular iron levels, and lipid peroxidation levels (Fig.  4 A–C). The mRNA levels of GPX4 and SLC7A11 declined, the GPX4 protein levels decreased, and the TFRC protein levels increased (Fig.  4 D, S6). Notably, the ferroptosis-related alterations induced by TNF-α overexpression were markedly alleviated by Fer-1 treatment. Similarly, compared to cells treated with TNF-α overexpression alone, cells co-treated with TNF-α overexpression and Fer-1 exhibited enhanced expression of ERVW1 and CGB3, along with an elevated protein level of β-hCG (Fig.  4 E–F). Moreover, the impaired syncytialization caused by TNF-α overexpression was ameliorated by Fer-1 treatment (Fig.  4 G). These findings indicate that TNF-α disrupts trophoblast cell fusion by inducing ferroptosis. Fig. 4 Overexpression of TNF-α promotes ferroptosis to inhibit BeWo cell fusion. MDA concentrations (n = 6) ( A ), intracellular iron contents (n = 5) ( B ), and lipid peroxidation levels (n = 3) ( C ) were measured in FSK-treated BeWo cells transfected with TNF-α overexpression plasmid and treated with Fer-1. Scale bars represent 150 µm. ( D ) The protein levels of TFRC and GPX4 were analyzed by Western blotting in FSK-treated BeWo cells transfected with TNF-α overexpression plasmid and treated with Fer-1 (n = 4). ( E ) RT-qPCR analysis was performed to detect the mRNA levels of ERVW1 and CGB3 in these cells (n = 3). ( F ) The protein level of β-hCG was analyzed by Western blotting (n = 3). ( G ) Immunofluorescence staining showed Fer-1 alleviated the harmful effects of TNF-α overexpression on BeWo cell fusion (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Results are expressed as means ± SEM. ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 Overexpression of TNF-α promotes ferroptosis to inhibit BeWo cell fusion. MDA concentrations (n = 6) ( A ), intracellular iron contents (n = 5) ( B ), and lipid peroxidation levels (n = 3) ( C ) were measured in FSK-treated BeWo cells transfected with TNF-α overexpression plasmid and treated with Fer-1. Scale bars represent 150 µm. ( D ) The protein levels of TFRC and GPX4 were analyzed by Western blotting in FSK-treated BeWo cells transfected with TNF-α overexpression plasmid and treated with Fer-1 (n = 4). ( E ) RT-qPCR analysis was performed to detect the mRNA levels of ERVW1 and CGB3 in these cells (n = 3). ( F ) The protein level of β-hCG was analyzed by Western blotting (n = 3). ( G ) Immunofluorescence staining showed Fer-1 alleviated the harmful effects of TNF-α overexpression on BeWo cell fusion (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Results are expressed as means ± SEM. ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 Given that TNF-α promotes ferroptosis, consequently disrupting trophoblast cell fusion, an in-depth investigation was carried out to clarify its underlying mechanism. Initially, bioinformatics analyses were performed to identify targets interacting with TNF-α. Six potential target genes were then prioritized: ISG15, IL1A, NFKBIZ, NR2F1, BID, and IRF8 (Fig.  5 A). Following, we performed RT-qPCR detection of these six genes in groups treated with overexpression of TNF-α or siTNF-α−1. Among the six genes, ISG15 exhibited the most marked upregulation in FSK-treated cells with overexpression of TNF-α. In line with this, ISG15 was also significantly downregulated in H1N1-infected cells subjected to treatment with siTNF-α−1 (Fig.  5 B). The other five target genes did not exhibit similar alterations (Figure S7A-E). Further verification of the changes in ISG15 at the protein level yielded comparable outcomes (Fig.  5 C). The above results indicate that ISG15 is regulated by TNF-α. As such, ISG15 was selected for further investigation of ferroptosis and trophoblast cell fusion. We transfected three siRNAs into FSK-treated BeWo cells infected with H1N1 virus and found that siISG15-1 showed the strongest silencing effect (Figure S7F). We then used siISG15-1 to downregulate the ISG15 levels in FSK-treated BeWo cells with H1N1 virus infection. We found that the MDA concentration, intracellular iron content, and lipid peroxidation level significantly decreased (Fig.  5 D–F). The mRNA levels of GPX4 and SLC7A11 were markedly elevated (Figure S7G). Specifically, the expression of GPX4 was significantly upregulated, whereas the expression of TFRC was downregulated considerably (Fig.  5 G). These findings imply that the knockdown of ISG15 suppresses ferroptosis triggered by H1N1 virus infection. Additionally, through RT-qPCR and Western blot analysis, it was noted that the knockdown of ISG15 resulted in elevated levels of ERVW1, CGB3, and β-hCG (Fig.  5 H–I). Immunofluorescence staining demonstrated that treatment with siISG15 enhanced the syncytialization of BeWo cells (Fig.  5 J). Collectively, these results suggest that the silencing of ISG15 suppresses ferroptosis, thereby promoting trophoblast cell fusion during H1N1 virus infection. Fig. 5 Knockdown of ISG15 alleviates the damage of ferroptosis to BeWo cell fusion.  (A) Bioinformatics analysis to identify downstream target genes of TNF-α through Protein–Protein Interaction (PPI), including IL1A, NFKBIZ, ISG15, IRF8, BID, and NF2F1. Proteins are the nodes, and their interactions are the edges. (B-C) The mRNA and protein levels of ISG15 were detected by RT-qPCR and Western blotting, respectively, in H1N1-infected cells transfected with siTNF-α or TNF-α overexpression plasmid (n = 3–4). Analysis of MDA concentrations (n = 6) (D) , intracellular iron contents (n = 5) (E) , and lipid peroxidation levels (n = 3) (F) in H1N1-infected cells transfected with siISG15. Scale bars represent 150 µm. (G) Western blot analysis of TFRC and GPX4 proteins in H1N1-infected cells transfected with siISG15 (n = 3). (H) RT-qPCR analysis of ERVW1 and CGB3 expression in H1N1-infected cells transfected with siISG15 (n = 3). (I) Western blot analysis of β-hCG expression in H1N1-infected cells transfected with siISG15 (n = 4). (J) Immunofluorescence staining revealed improved syncytialization in the siISG15-transfected group compared with the H1N1-infected group (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 Knockdown of ISG15 alleviates the damage of ferroptosis to BeWo cell fusion.  (A) Bioinformatics analysis to identify downstream target genes of TNF-α through Protein–Protein Interaction (PPI), including IL1A, NFKBIZ, ISG15, IRF8, BID, and NF2F1. Proteins are the nodes, and their interactions are the edges. (B-C) The mRNA and protein levels of ISG15 were detected by RT-qPCR and Western blotting, respectively, in H1N1-infected cells transfected with siTNF-α or TNF-α overexpression plasmid (n = 3–4). Analysis of MDA concentrations (n = 6) (D) , intracellular iron contents (n = 5) (E) , and lipid peroxidation levels (n = 3) (F) in H1N1-infected cells transfected with siISG15. Scale bars represent 150 µm. (G) Western blot analysis of TFRC and GPX4 proteins in H1N1-infected cells transfected with siISG15 (n = 3). (H) RT-qPCR analysis of ERVW1 and CGB3 expression in H1N1-infected cells transfected with siISG15 (n = 3). (I) Western blot analysis of β-hCG expression in H1N1-infected cells transfected with siISG15 (n = 4). (J) Immunofluorescence staining revealed improved syncytialization in the siISG15-transfected group compared with the H1N1-infected group (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 To determine whether TNF-α regulates ferroptosis and trophoblast cell fusion through targeting ISG15, we set out to perform reverse experiments. As shown in Fig.  6 A–C, overexpression of TNF-α led to a significant increase in MDA concentrations, intracellular iron contents, and lipid peroxidation levels. RT-qPCR and Western blotting revealed that GPX4 and SLC7A11 were downregulated, while TFRC was upregulated (Fig.  6 D–E). The above-mentioned effects were reversed by siISG15 (Fig.  6 A–E). Analogously, the overexpression of TNF-α remarkably suppressed the mRNA expression of ERVW1, CGB3, and the protein levels of β-hCG. Nevertheless, this inhibitory effect was reversed when ISG15 was knocked down (Fig.  6 F–G). The impaired syncytialization caused by TNF-α overexpression was also ameliorated by the knockdown of ISG15 (Fig.  6 H). These data clearly demonstrate that TNF-α promotes ferroptosis and disrupts trophoblast cell fusion through targeting ISG15. Fig. 6 TNF-α promotes ferroptosis through targeting ISG15 to inhibit BeWo cell fusion. Alterations in MDA concentrations (n = 5) ( A ), intracellular iron contents (n = 5) ( B ), and lipid peroxidation levels (n = 3) ( C ) were observed in FSK-treated BeWo cells co-transfected with TNF-α overexpression plasmid and siISG15. Scale bars represent 150 µm. ( D – E ) Alterations in mRNA and protein levels of GPX4, SLC7A11, and TFRC were assessed by RT-qPCR and Western blotting, respectively, under varying conditions (n = 4). ( F – G ) Alterations in mRNA and protein levels of ERVW1, CGB3, and β-hCG were determined by RT-qPCR and Western blotting under varying conditions (n = 3–4). ( H ) Immunofluorescence staining showed the knockdown of ISG15 alleviated the detrimental effects of TNF-α overexpression on BeWo cell fusion (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001. TNF-α promotes ferroptosis through targeting ISG15 to inhibit BeWo cell fusion. Alterations in MDA concentrations (n = 5) ( A ), intracellular iron contents (n = 5) ( B ), and lipid peroxidation levels (n = 3) ( C ) were observed in FSK-treated BeWo cells co-transfected with TNF-α overexpression plasmid and siISG15. Scale bars represent 150 µm. ( D – E ) Alterations in mRNA and protein levels of GPX4, SLC7A11, and TFRC were assessed by RT-qPCR and Western blotting, respectively, under varying conditions (n = 4). ( F – G ) Alterations in mRNA and protein levels of ERVW1, CGB3, and β-hCG were determined by RT-qPCR and Western blotting under varying conditions (n = 3–4). ( H ) Immunofluorescence staining showed the knockdown of ISG15 alleviated the detrimental effects of TNF-α overexpression on BeWo cell fusion (n = 3). Scale bars represent 150 µm. Statistical analysis was performed using one-way ANOVA. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001. To elucidate how ISG15 disrupts trophoblast cell fusion through ferroptosis, we predicted its downstream targets through bioinformatics analyses. As depicted in Fig.  7 A, six potential genes were identified. Notably, EGR1, a pro-ferroptotic transcription factor, was found to interact with ISG15. We further carried out experiments to verify EGR1 as an ISG15 target gene. Our findings demonstrated that EGR1 expression decreased at both the mRNA and protein levels in H1N1-infected cells treated with siISG15, whereas it increased in BeWo cells treated with an ISG15 overexpression plasmid (Fig.  7 B–C). These findings prompted us to investigate the role of EGR1 in mediating the effects of ISG15 on ferroptosis and trophoblast cell fusion. Among the three siRNAs tested, siEGR1-3 showed the highest knockdown efficiency in reducing EGR1 expression and was selected for subsequent studies (Figure S8A). To determine whether ISG15 promotes ferroptosis and trophoblast cell fusion by targeting EGR1, we performed rescue experiments. As shown in Fig.  7 D–F, ISG15 overexpression significantly increased MDA concentrations, intracellular iron contents, and lipid peroxidation levels. At the mRNA level, the expression of both GPX4 and SLC7A11 was downregulated (Figure S8B). At the protein level, GPX4 exhibited decreased expression, whereas TFRC showed upregulated expression (Fig.  7 G). These effects were reversed upon treatment with siEGR1 (Fig.  7 D–G). Similarly, siEGR1 reversed the ISG15-induced downregulation of syncytialization indicators (Fig.  7 H, S8C). Meanwhile, siEGR1 intervention alleviated the impaired syncytialization induced by the overexpression of ISG15 (Fig.  7 I). These findings clearly illustrate that ISG15 is crucial in promoting ferroptosis and disrupting trophoblast cell fusion by specifically targeting EGR1. To explore whether TNF-α exerts its functional effects through the ISG15 or EGR1 signaling pathway, we conducted subsequent experiments. Our results, as evidenced by RT-qPCR, revealed that overexpression of TNF-α led to a significant upregulation of the EGR1 mRNA level. Intriguingly, this effect was effectively reversed by siISG15 (Fig.  7 J). These data suggest that the TNF-α/ISG15/EGR1 axis mediates ferroptosis and impairs syncytialization. Given that EGR1 functions as a transcription factor, we hypothesized that it enhances TNF-α transcription, thereby forming a positive feedback loop. To validate this hypothesis, we transfected siEGR1 into FSK-treated BeWo cells infected with H1N1 virus. Compared with H1N1-infected, FSK-treated BeWo cells, a reduction in TNF-α mRNA expression was observed (Fig.  7 K). Consistent results were also obtained in EGR1 overexpression experiments (Fig.  7 L). Additionally, CHIP assays were performed using EGR1 antibodies to detect the binding level of EGR1 to the TNF-α gene promoter region. The results indicated that EGR1 promotes TNF-α transcription (Fig.  7 M). To figure out the impact of EGR1 knockdown on ferroptosis and syncytialization efficiency in BeWo cells under TNF-α stimulation, we performed rescue experiments. As shown in Figure S9A-C, TNF-α overexpression significantly increased MDA concentration, intracellular iron content, and lipid peroxidation levels. RT-qPCR and Western blotting analyses revealed downregulated expression of GPX4 and SLC7A11, accompanied by upregulated expression of TFRC (Figure S9D-E). Notably, these TNF-α-induced effects were reversed by EGR1 knockdown via siRNA (Figure S9A-E). Similarly, TNF-α overexpression markedly suppressed the mRNA expression of ERVW1 and CGB3 as well as the protein level of β-hCG. However, this inhibitory effect was abrogated by EGR1 knockdown (Figure S9F-G). At the same time, the impairment of syncytialization caused by TNF-α overexpression was ameliorated following EGR1 knockdown (Figure S9H). Taken together, our results reveal that EGR1 is a key target of TNF-α, and the TNF-α/EGR1 axis mediates ferroptosis and subsequent impairment of trophoblast cell fusion. These findings support the hypothesis that TNF-α and EGR1 form a positive feedback loop. Fig. 7 ISG15 promotes ferroptosis by targeting EGR1 to inhibit the BeWo cell fusion. ( A ) Bioinformatics analysis identified downstream target genes of ISG15 via the PPI network, including TNF-α, IFNL3, EGR1, BST2, OAS2, and IFIT2. (B-C) The mRNA and protein levels of EGR1 in H1N1-infected cells transfected with siISG15 or ISG15 overexpression plasmid (n = 4), respectively, were detected by RT-qPCR and Western blot. Alterations of MDA concentrations (n = 6) (D) , intracellular iron contents (n = 5) (E) , and lipid peroxidation levels (n = 3) (F) in FSK-treated BeWo cells co-transfected with ISG15 overexpression plasmid and siEGR1. Scale bars represent 150 µm. (G) Detection of TFRC and GPX4 levels by Western blotting under varying conditions (n = 4). (H) Detection of β-hCG protein levels using Western blot under varying conditions (n = 4). (I) Immunofluorescence staining showed siEGR1 alleviated the detrimental effects of ISG15 overexpression on BeWo cell fusion (n = 3). Scale bars represent 150 µm. (J) RT-qPCR analysis of EGR1 mRNA expression in FSK-treated BeWo cells co-transfected with TNF-α overexpression plasmid and siISG15 (n = 3). (K) RT-qPCR analysis of TNF-α mRNA expression in H1N1-infected BeWo cells transfected with siEGR1 (n = 3). (L) The mRNA level of TNF-α was analyzed by RT-qPCR in FSK-treated BeWo cells transfected with EGR1 overexpression plasmid (n = 4). (M) CHIP-qPCR assays were performed to measure EGR1 levels in the TNF-α promoter regions of FSK-treated BeWo cells infected with H1N1 virus (n = 4). Statistical analysis was performed using Student’s t-test and one-way ANOVA. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 ISG15 promotes ferroptosis by targeting EGR1 to inhibit the BeWo cell fusion. ( A ) Bioinformatics analysis identified downstream target genes of ISG15 via the PPI network, including TNF-α, IFNL3, EGR1, BST2, OAS2, and IFIT2. (B-C) The mRNA and protein levels of EGR1 in H1N1-infected cells transfected with siISG15 or ISG15 overexpression plasmid (n = 4), respectively, were detected by RT-qPCR and Western blot. Alterations of MDA concentrations (n = 6) (D) , intracellular iron contents (n = 5) (E) , and lipid peroxidation levels (n = 3) (F) in FSK-treated BeWo cells co-transfected with ISG15 overexpression plasmid and siEGR1. Scale bars represent 150 µm. (G) Detection of TFRC and GPX4 levels by Western blotting under varying conditions (n = 4). (H) Detection of β-hCG protein levels using Western blot under varying conditions (n = 4). (I) Immunofluorescence staining showed siEGR1 alleviated the detrimental effects of ISG15 overexpression on BeWo cell fusion (n = 3). Scale bars represent 150 µm. (J) RT-qPCR analysis of EGR1 mRNA expression in FSK-treated BeWo cells co-transfected with TNF-α overexpression plasmid and siISG15 (n = 3). (K) RT-qPCR analysis of TNF-α mRNA expression in H1N1-infected BeWo cells transfected with siEGR1 (n = 3). (L) The mRNA level of TNF-α was analyzed by RT-qPCR in FSK-treated BeWo cells transfected with EGR1 overexpression plasmid (n = 4). (M) CHIP-qPCR assays were performed to measure EGR1 levels in the TNF-α promoter regions of FSK-treated BeWo cells infected with H1N1 virus (n = 4). Statistical analysis was performed using Student’s t-test and one-way ANOVA. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 To investigate the role of TNF-α in regulating ferroptosis and trophoblast cell fusion in vivo, we established a mouse model of H1N1 virus infection. Pregnant BALB/c mice at embryonic day 13.5 (E13.5) were intranasally infected with 10 6 EID50 H1N1 virus, followed by subcutaneous administration of Etanercept (a TNF-α inhibitor) or saline at E14.5 (Fig.  8 A). Body weight changes were monitored for 7 days post-infection. Mice treated with Etanercept exhibited significantly less weight loss compared to the infected control group, as reflected by a markedly smaller downward trend in the body weight change curve (Fig.  8 B). Viral titers were quantified from the placental lysates of infected mice using the TCID 50 assay. The results confirmed the presence of H1N1 virus in placental tissues of infected mice, validating the successful establishment of the viral infection model (Fig.  8 C). To investigate the impact of TNF-α inhibition on ferroptosis and syncytialization, we next assessed the efficacy of Etanercept. Compared to the H1N1-infected group, the mRNA level of TNF-α was significantly reduced in the H1N1-infected group with Etanercept treatment (Fig.  8 D). Subsequently, the alterations associated with ferroptosis in the mice were observed and documented. IHC staining of mouse placental tissue sections using the TFRC antibody demonstrated a decrease in TFRC expression (Fig.  8 E). In addition, we noted a reduction in both MDA concentrations and intracellular iron contents in the H1N1-infected group with Etanercept treatment compared to the H1N1-infected group without such treatment (Fig.  8 F–G). Furthermore, an elevation in the mRNA expression levels of GPX4 and SLC7A11 was also observed (Fig.  8 H). Western blot analysis indicated that in the H1N1-infected group treated with Etanercept, the protein level of GPX4 was upregulated, whereas that of TFRC was significantly downregulated (Fig.  8 I). These findings indicate that the inhibition of TNF-α alleviates H1N1-induced ferroptosis in pregnant mice. RT-qPCR analysis showed that treatment with Etanercept significantly upregulated the mRNA expression levels of ERVW1 and CGB3 in the H1N1-infected group (Fig.  8 J). In line with these results, Western blot analysis showed a concurrent increase in β-hCG protein levels in the H1N1-infected placental samples treated with Etanercept (Fig.  8 K). Collectively, the data presented herein demonstrate that the inhibition of TNF-α alleviates H1N1-induced ferroptosis and restores trophoblast cell fusion in the placentas of infected pregnant mice. These findings underscore the crucial role of TNF-α in mediating ferroptosis and trophoblast dysfunction during H1N1 infection. Fig. 8 TNF-α inhibitor attenuates H1N1-induced ferroptosis and enhances trophoblast syncytialization in pregnant mice. (A) Schematic diagram of the experimental design. (B) Body weight changes in pregnant mice were tracked during infection and treatment. Weight changes were compared between H1N1-infected and H1N1-infected + Etanercept-treated groups. The TNF-α mRNA level was measured by RT-qPCR in placental tissues of the control, H1N1-infected, and H1N1-infected + Etanercept-treated groups. (C) Viral titers in placentas of infected mice at E20.5 (n = 3). (D) RT-qPCR analysis of TNF-α mRNA levels in placental tissues from different groups (n = 3). (E) Placental sections of different groups were IHC-stained with anti-TFRC antibody. Scale bars represent 50 μm. MDA concentrations (n = 5) (F) and intracellular iron contents (n = 4) ( G) were quantitatively evaluated under different conditions. (H) RT-qPCR analysis of GPX4 and SLC7A11 mRNA levels in placental tissues from different groups (n = 3). (I) Western blot of TFRC and GPX4 protein levels in placental tissues from different groups (n = 4). (J) RT-qPCR analysis of ERVW1 and CGB3 mRNA levels in placental tissues from different groups (n = 3). (K ) Western blot analysis of β-hCG protein expression in placental tissues from different groups (n = 3). Statistical analyses were performed using one-way ANOVA followed by Bonferroni correction and Student’s t test for two means. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001 TNF-α inhibitor attenuates H1N1-induced ferroptosis and enhances trophoblast syncytialization in pregnant mice. (A) Schematic diagram of the experimental design. (B) Body weight changes in pregnant mice were tracked during infection and treatment. Weight changes were compared between H1N1-infected and H1N1-infected + Etanercept-treated groups. The TNF-α mRNA level was measured by RT-qPCR in placental tissues of the control, H1N1-infected, and H1N1-infected + Etanercept-treated groups. (C) Viral titers in placentas of infected mice at E20.5 (n = 3). (D) RT-qPCR analysis of TNF-α mRNA levels in placental tissues from different groups (n = 3). (E) Placental sections of different groups were IHC-stained with anti-TFRC antibody. Scale bars represent 50 μm. MDA concentrations (n = 5) (F) and intracellular iron contents (n = 4) ( G) were quantitatively evaluated under different conditions. (H) RT-qPCR analysis of GPX4 and SLC7A11 mRNA levels in placental tissues from different groups (n = 3). (I) Western blot of TFRC and GPX4 protein levels in placental tissues from different groups (n = 4). (J) RT-qPCR analysis of ERVW1 and CGB3 mRNA levels in placental tissues from different groups (n = 3). (K ) Western blot analysis of β-hCG protein expression in placental tissues from different groups (n = 3). Statistical analyses were performed using one-way ANOVA followed by Bonferroni correction and Student’s t test for two means. Data are expressed as mean ± SEM. *, p < 0.05; **, p < 0.01; ***, p < 0.001, or ****, p < 0.0001

The human choriocarcinoma cell line BeWo was purchased from Otwo Biotech Company and maintained in F-12K medium (Thermo Fisher Scientific), containing 1% penicillin and streptomycin (Thermo Fisher Scientific) and 10% fetal bovine serum (HyClone Laboratories, Inc.). Madin‐Darby canine kidney (MDCK) cells were cultured in Dulbeccos modified Eagles medium (DMEM) (Thermo Fisher Scientific) supplemented with 1% penicillin and streptomycin (Thermo Fisher Scientific) and 10% fetal bovine serum (HyClone Laboratories, Inc.). Influenza virus A/Sichuan/1/2009 was grown in MDCK cells cultured in DMEM with 0.3% bovine serum albumin (Sigma-Aldrich Co.) and 0.5 µg/mL TPCK‐treated trypsin (Sigma—Aldrich Co.). The virus was stored at −70 °C until use. Forskolin (FSK; GlpBio) was dissolved in dimethyl sulfoxide (DMSO) and stored at −20 °C until use. Following cell attachment, BeWo cells were treated with FSK for 48 h. To assess gene expression, the cells were seeded into 6-well plates and infected with the virus at a multiplicity of infection (MOI) of 1 for 1 h at 37 °C. Next, the inoculum containing the virus was removed and cells were washed with 1 × PBS and fresh media was added. All cells were maintained in a humidified incubator containing 5% CO 2 at 37 °C until use. Total RNA was extracted from FSK-treated BeWo cells and FSK-treated BeWo cells infected with H1N1 virus at an MOI of 1 for 24 h and used for RNA-seq analysis. Equal amounts of total RNA from three replicates in each group were pooled to construct cDNA libraries (Illumina, San Diego, CA). The cell lysates were sent to the Genomics Institute (Beijing, China) and subjected to RNA-seq via the BGISEQ-500 platform. High-quality reads were aligned to the human reference genome (GRCh38) via Bowtie2. Gene expression was quantified and normalized via the RSEM tool. Differently expressed genes (DEGs) between the control group (FSK-treated BeWo cells) and H1N1-infected group (FSK-treated BeWo cells infected with H1N1 virus) were identified using the limma package, with an adjusted p-value (false discovery rate, FDR) < 0.05 and |log fold change (FC)|≥ 2 as the thresholds. Volcano plots were used to visualize the DEGs described above. The functional enrichment analysis of the DEGs was carried out via phyper in R software (version 2.13.1). With a Q value of ≤ 0.05 as the threshold, candidate genes that met this condition were defined as being significantly enriched. A protein–protein interaction (PPI) network was constructed using the STRING database ( http://string-db.org ) to explore the interactions among DEGs, TNF-α and ISG15. The PPI pairs were screened by confidence score (> 0.40). The PPI network was visualized via Cytoscape software (version 3.9). The nodes represent genes, and the edges represent the interactions between the genes. To quantify the mRNA expression levels of the target genes, biomarkers of programmed cell death, and syncytialization markers in FSK-treated BeWo cells after various treatments, total RNA was extracted from the placental tissue and cells using TRIzol reagent. cDNA was subsequently generated with a ReverTra Ace qPCR RT kit (TOYOBO). The cDNA was mixed with the ChamQ Universal SYBR qPCR master mix (Vazyme, Q711-02), followed by quantitative real-time PCR with an ABI 7500 real-time PCR system (Applied Biosystems, USA). Relative mRNA expression was calculated via the 2- ΔΔ CT method, with β-actin used as a reference gene. The primers used in the study are listed in Table  1 , and the primer concentration used in the reactions was 0.5 μM according to the manufacturer's instructions. The primer sequences are shown in Table  1 . Table 1 The following primers were used in this study GPX4-F AGATCCACGAATGTCCCAAG GPX4-R CCTCCTCCTTAAACGCACAC SLC7A11-F ACGGTGGTGTGTTTGCTGTCTC SLC7A11-R GCTGGTAGAGGAGTGTGCTTGC Mouse SLC7A11-F AGGGCATACTCCAGAACACG Mouse SLC7A11-R GGACCAAAGACCTCCAGAATG MLKL-F AGGAGGCTAATGGGGAGATAGA MLKL-R TGGCTTGCTGTTAGAAACCTG RIPK3-F GAGACTCCCGGCTTAGAAGG RIPK3-R TCCTTTACCGTGGAGACAGC p62-F GACTACGACTTGTGTAGCGTC p62-R AGTGTCCGTGTTTCACCTTCC LC3B-F CCTAGAAGGCGCTTACAGCT LC3B-R GGGACAATTTCATCCCGAAC BAX-F CCTTTTCTACTTTGCCAGCAAAC BAX-R GAGGCCGTCCCAACCAC BCL-2-F GGTGGGGTCATGTGTGTGG BCL-2-R CGGTTCAGGTACTCAGTCATCC Caspase-1-F TTTCCGCAAGGTTCGATTTTCA Caspase-1-R GGCATCTGCGCTCTACCATC NLRP3-F GATCTTCGCTGCGATCAACAG NLRP3-R CGTGCATTATCTGAACCCCAC β-actin-F CATGTACGTTGCTATCCAGGC β-actin-R CTCCTTAATGTCACGCACGAT Mouse β-actin-F TGGGAATGGGTCAGAAGGA Mouse β-actin-R ATTGAGAAAGGGCGTGGC Mouse CGB3‐F ATGAAGCTGCTGGAGATGGA Mouse CGB3‐R CTTCTGGCTGGAGGTGATGA ERVW1‐F GCTGTCTTCTGGGCTGTTTC ERVW1‐R CAGGGTCAGAGTCCAGGTCT Mouse ERVW1‐F ATGGGA ACCTGTGTCTTCTGG Mouse ERVW1‐R TCAGGGTCAGAGTCCAGGTC IL1A-F ATGACACCACCTGAACGTCTC IL1A-R CTCTCCAGAGCAGTGAGTTCT NFKBIZ-F ATGACCAACAAGTGTCTCCTCC NFKBIZ-R GGAATCCAAGCAAGTTGTAGCTC ISG15-F AAGGTCAGCCAGAACAGGTCGT ISG15-R GAGCATCCTGGTGAGGAATAAC NR2F1-F GCTGTCGATATTGGGGCTTG NR2F1-R GGAAACGCGCTGGTTTTCAT BID-F GTCCAACAATGTGACCCAGAT BID-R ACCTCAACGATATGCAGCCG IRF8-F CCATCCCGCCCACTTTCTAC IRF8-R AGCTCAATCCGTTGTTCAGGC EGR1-F CCACGCCGAACACTGACATT EGR1-R GAGGGGTTAGCGAAGGCTG TNF-α-F CATGTTGTAGCAAACCCTCAAG TNF-α-R GACCTGGGAGTAGATGAGGTACA The following primers were used in this study Protein extracts were prepared from placental tissues and cells were lysed with NP‐40 buffer (Beyotime Biotechnology, China), and protease inhibitor (Beyotime Biotechnology) for 30 min on ice. After centrifugation at 12,000 g for 15 min at 4 ℃, the total protein concentration was assessed using a BCA kit (GlpBio). Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS‒PAGE) and electrotransferred onto nitrocellulose membranes (GE Healthcare). The blots were blocked for 1 h at room temperature with 5% skim milk in TBST buffer and incubated overnight at 4 °C. An anti-β-hCG antibody (1:1000, Proteintech, China), anti-GPX4 antibody (1:1000, ABclonal, China), anti-TFRC antibody (1:1000, Abcam, UK), anti-MLKL antibody (1:500, HUABIO, China), anti- RIK3 antibody (1:1000, Proteintech, China), anti-BAX antibody (1:1000, Proteintech, China), anti-BCL2 antibody(1:500, Proteintech, China), anti-LC3B antibody (1:1000, Proteintech, China), anti-p62 antibody (1:1000, Proteintech, China), anti-Caspase-1 antibody (1:1000, Proteintech, China), anti-NLRP3 antibody (1:1000, CST), anti-GAPDH antibody (1:5000, Proteintech, China) and anti-β-actin antibody (1:1000, HUABIO, China) were used. The membranes were subsequently washed three times with TBST for 15 min each time. Afterward, we incubated the membranes for one hour with secondary antibodies 800 M or 800R, then examined them by ODYSSEY. The grayscale values of the protein bands were analyzed via Image Studio, and β-actin was used as the internal reference. The Caspase-1 activity was assayed by using Caspase-1 activity assay kit (Beyotime, China) according to the manufacturer's instructions. The absorbance was measured at a wavelength of 405 nm. MDC staining was used as a tracer of autophagic vesicles for autophagy detection. All acidic vacuoles in the cells were visualized by MDC staining. According to the manufacturer’s instructions, 2 × 10 5 cells were plated in a 6-well plate overnight before group treatments. After different treatments, BeWo cells were stained with MDC (Beyotime, C3018S, Nanjing, China) for 30 min at 37 °C in the dark, washed with assay buffer three times, and visualized using a fluorescence microscope (Olympus Inc., USA). BeWo cells were inoculated on cell crawls, and the cells were observed under the microscope to be completely adhered to the wall. They were incubated for 24 h after drug administration according to the experimental groups and washed twice with PBS. Then 100 μL of 1 × Annexin V binding buffer, 5 μL of Annexin V-AbFluor™ 488, and 2 μL of propidium iodide (PI) were added to each well and incubated at room temperature and protected from light for 15 min. They were washed with 1 × Annexin V binding buffer and with 1 × Annexin V-AbFluor™ 488/PI. They were incubated with Annexin V binding buffer for 15 min at room temperature and protected from light. Then the cells were washed twice with 1 × Annexin V binding buffer and observed and analyzed under a fluorescence microscope using appropriate filters. FSK-treated BeWo cells were cultured in 6-well plates. In accordance with the research objectives, various reagents were administered to FSK-treated BeWo cells. Following these treatments, the cells were fixed with PBS containing 4% paraformaldehyde (Beijing Solarbio Science & Technology Co., Ltd.) for 20 min and then permeabilized with PBS-Tween (PBST) containing 0.1% Triton X-100 (Biosharp) for 10 min. After being blocked with 3% bovine serum albumin (MP Biomedicals) in PBS, the cells were incubated with an anti-E-cadherin antibody (1:200, Proteintech, China) at 4 °C overnight. The cells were washed three times with PBST and incubated for 1 h with a goat anti-rabbit secondary antibody (Alexa Fluor® 488; Abcam Plc) at room temperature. The cells were washed three times with PBST and incubated with DAPI (Thermo Fisher Scientific, USA) for 3 min at room temperature. Images were acquired via fluorescence microscopy. BeWo cells were stained with anti-E-cadherin (green) or DAPI (blue). The scale bar represents 150 µM. The degree of cell fusion was quantified based on the fusion index. The fusion index was calculated by dividing the total number of fused cells (cells with at least 2 nuclei) in each field of view by the total number of cells in the field of view. Intracellular iron levels in cells and placental tissues were assessed using the Iron Content Assay Kit (Solarbio, Beijing, China). The assay was performed in accordance with the instruction. For cell samples, the cells were centrifuged at 1000 rpm for 3 min, and the resulting precipitates were mixed with 1 mL of extracting solution and sonicated on ice. The mixture was then subjected to centrifugation at 8,000 g for 10 min at 4 °C. The supernatants (20 μL) were incubated with the test solution (180 μL), and the optical density (OD value of the supernatants was measured using a microplate reader at 510 nm (Envision Microplate Reader). For placental tissue samples, 1 mL of extraction buffer was added to the tissues, followed by homogenization in an ice bath, centrifuged at 4000 g for 10 min 4℃, and the supernatant was removed. The supernatant (400 μL) was mixed with the test solution (600 μL). The mixture was incubated in a boiling water bath for 5 min and then immediately cooled on ice. Add 200 μL of chloroform to mix well. After centrifugation at 10,000 rpm for 10 min, 800 μL of superfluid was taken, and the OD was measured at 520 nm using a microplate reader. To assess the level of MDA, a metabolite indicative of cellular lipid peroxidation, an MDA Content Assay Kit (Solarbio,Beijing, China) was used following the manufacturer’s instructions. For placental tissue samples, tissues were homogenized in 1 mL of extraction buffer on ice and centrifuged at 8,000 g for 10 min at 4 °C, and the supernatant was collected. For cell samples, cells were centrifuged at 1,000 rpm for 3 min, and the resulting precipitates were mixed with 1 mL of extracting solution and sonicated at 200 W. The mixture was then centrifuged at 8,000 g for 10 min at 4 °C. Subsequently, 100 μL of the supernatant was mixed with 300 μL of test solution and incubated at 100 °C for 60 min. After incubation, the samples were centrifuged at 10,000 g for 10 min. The OD value of the supernatants were measured using a microplate reader (Envision Microplate Reader) at 532 nm and 600 nm, facilitating the calculation of relative MDA content. BeWo cells were inoculated into 6-well plates and exposed to FSK (50 μM) for 48 h after different treatments. After that, the lipophilic fluorescent dye C11 BODIPY 581/591 (Thermo Fisher Scientific, D3861) was added to cells, incubated at 37 °C for 30 min, and observed under a fluorescence microscope. For quantitative analysis, green fluorescent intensity was calculated and normalized using Image J software. FSK-treated BeWo cells were cultured in serum-free medium and subsequently transfected with various constructs, including TNF-α small interfering RNA (siRNA)—siRNA1, 2, 3; ISG15-siRNA—siRNA1, 2, 3; EGR1-siRNA—siRNA1, 2, 3; as well as control sequences obtained from Genecefe Biotechnology Co., Ltd (Jiangsu, China). Additionally, the cells were transfected with TNF-α overexpression plasmid (Vigene, China), ISG15 overexpression plasmid, and EGR1 overexpression plasmid synthesized by Ke Lei Biological Technology Co., along with the empty vector. The transfection was performed using Opti-MEM (Invitrogen, USA), separately mixed with Lipofectamine 2000 (Thermo Scientific, USA), and siRNA or overexpression plasmid reduced serum medium for 5 min. The two mixtures were then combined and incubated at room temperature for 30 min. The mixture was added to the cell culture plates and then incubated at 37 °C for 48 h in an environment with 5% CO 2 supplied for subsequent experiments. Cell samples were fixed with 2.5% glutaraldehyde at 4 °C overnight and then post-fixed with 1% osmium at 4 °C for 2 h. Samples were dehydrated stepwise in a graded ethanol series (50%, 70%, 90%, and 100%) at 4 °C and embedded in 812 Epon resin. After resin polymerization at 70 ℃ for 2 days, sections of 65–70 nm were cut by an UC6 ultra-microtome (Leica Microsystems) and stained with uranyl acetate and lead citrate before being analyzed in an H-7650 (Hitachi, Tokyo, Japan) operated at 80 kV. Chromatin Immunoprecipitation assays were performed with the CHIP assay kit (catalog no. P2078, Beyotime Biotechnology, China). Briefly, FSK-treated BeWo cells were subjected to cross-linking using a 37% formaldehyde solution for 10 min at 37 ℃, followed by quenching with 125 mmol/L glycine. The cells were centrifuged at 4 ºC for 800–1000 ×  g for 2 min to achieve complete cell precipitation. Subsequently, incubate the samples in an ice bath for 10 min to ensure complete cell lysis. The DNA fragments ranging from 400–800 bp were obtained through sonication and incubation at 65 ºC for 4 h to disrupt the protein-DNA cross-linking. After centrifugation at 12,000 ×  g for 10 min at 4 °C, the lysate was subsequently subjected to immunoprecipitation using anti-EGR1 (Abcam, UK) and IgG antibodies, followed by overnight incubation at 4 °C. Protein A/G beads (60 μL) were added to the antibody-lysate mixture and incubated for an additional hour at 4 °C. The beads were washed, and DNA fragments were eluted, purified using the DNA Purification Kit (D0033; Beyotime Biotechnology, China), and analyzed by qPCR. All animal experiments were approved by the Committee on the Ethics of Animal Experiments of the Harbin Veterinary Research Institute (HVRI) of the Chinese Academy of Agricultural Sciences (CAAS). The ethics number is 230,524‐01‐GJ. BALB/c mice were purchased from Beijing Weitong Lihua Company. All animals were adaptively fed for one week with a 12 h light/dark cycle and free access to food and water. Healthy adult mice were selected. The female and the male were caged at a ratio of 2:1 overnight, then they were separately placed before 8 am the next morning and examined the female genital area. Those with vaginal plugs were considered as mated females and recorded as embryonic day (E)0.5. Pregnant mice were housed in separate cages. The pregnancy status of the mice was determined when they gained 20% of their initial body weight. Their weight changes were recorded daily. An H1N1-infected pregnant mouse model was established by intranasal infection of the 50 μL of 10 6 EID50 of the 2009 pandemic H1N1 virus diluted in phosphate-buffered saline (PBS) at E13.5. The control mice were injected with the same amount of PBS. Pregnant mice infected with H1N1 virus were randomly divided into two groups (8 mice in each group) at E14.5: the H1N1-infected group and the H1N1-infected group with Etanercept treatment. The TNF-α inhibitor Etanercept was obtained from Pfizer Inc. Pregnant mice in the H1N1-infected group with Etanercept treatment received Etanercept subcutaneous 2.5 mg/kg twice daily for seven days, and the H1N1-infected group and uninfected control group received saline. Weight was taken daily, and weight changes were recorded. Pregnant mice in each group were euthanized via CO 2 anesthetization at E20.5, and placental tissue was dissected and separated. The placenta tissue was collected within 10 min after extraction of the placenta in pregnant mice, and the area near the insertion of the umbilical cord placenta was selected to avoid potential infection, bleeding, infarction and calcification. Some of the placental tissues was homogenized in PBS, and the mixture centrifuged at 1,500 g for 5 min. Supernatants were then used for virus detection by hemagglutination tests. Some of the placental tissues were fixed with 4% paraformaldehyde (Beijing Solarbio Science & Technology Co., Ltd.) for immunohistochemical analysis. Another part of the tissue samples was cut into 1-cubic-centimeter pieces, rinsed with PBS, further fragmented, and then frozen at −80 ℃ for subsequent molecular biology experiments. Viral titers were determined in MDCK cells. The culture supernatants were serially tenfold diluted in DMEM with 1 mg/mL TPCK-treated trypsin. A 100 µL aliquot of each diluted sample was added to 96-well plates with 80% confluent MDCK cells and then incubated at 37 °C in 5% CO₂ for 48 h. Next, 50 µL of medium per well was used in a hemagglutination assay to detect the virus. Samples were cut into 5 μm paraffin sections. Deparafnization and rehydration of tissue slices were performed using a graduated alcohol series. Sections were immersed in 0.01 M citrate buffer (pH 6.0) and heated continuously for 20 min. 3% H 2 O 2 was used to inactivate endogenous enzymes for 10 min. Sections were dripped with primary antibody and left overnight at 4 °C. A dilution ratio of 1:500 was used for TFRC (Abcam, UK). After that, the sections were incubated with a goat anti-rabbit immunoglobulin IgG polymer incubating at 37 °C for 30 min. 50 μl of diaminobenzidine (DAB) working solution was added, and the reaction was carried out for 3 min. After hematoxylin and PBS treatment, various alcohols (70–100%) were used for dehydration. After treatment with dimethylbenzene and neutral resin, the samples were observed and photographed under a microscope. Images were captured using an Olympus-fluorescence microscope (Olympus Corporation, Japan, Hyper E301). The data were expressed as mean ± SEM and analyzed using GraphPad Prism 8.0. Statistical analysis was performed with one-way ANOVA followed by Bonferroni correction among multiple groups and Student's t test for two means. P  < 0.05 was considered a statistically significant difference.

IAV infection continues to increase the risk of adverse pregnancy outcomes. Despite current advancements in the prevention and control of viral infections, there remains a lack of comprehensive understanding regarding the mechanisms that contribute to increased maternal and fetal morbidity and mortality during IAV infection. Thus, it is of utmost importance to explore the molecular mechanisms underlying the IAV infection-induced disruption of placental barriers. In this study, we utilized BeWo cells to investigate IAV infection and mimicked syncytiotrophoblast (STB) cells via forskolin (FSK) treatment in vitro. Our findings demonstrated that H1N1 virus infection promoted ferroptosis through the TNF-α/ISG15/EGR1 axis, which in turn eventually impaired trophoblast syncytialization. Ferroptosis, a novel mode of cell death, is characterized by the accumulation of iron and lipid peroxidation [ 28 ]. Accumulating evidence indicates that ferroptosis is intricately associated with villous physiology and pathobiology, as well as various conditions such as infertility, endometriosis, and preeclampsia[ 16 , 29 – 31 ]. In order to understand the biological functions of DEGs, pathway enrichment analysis was performed at the KEGG database, and the top 20 enriched pathways are presented in Fig.  1 B, including ferroptosis. Our results demonstrated that H1N1 virus infection in FSK-treated BeWo cells led to alterations related to ferroptosis, and these changes were reversed by Fer-1 treatment. Importantly, Fer-1 also ameliorated H1N1-induced impairment of BeWo cell syncytialization, providing direct evidence that H1N1 virus infection induces ferroptosis in FSK-treated BeWo cells and that this ferroptosis may contribute to syncytialization defects. Additionally, these findings align with previous studies reporting that IAV infection can induce multiple forms of programmed cell death, including apoptosis and necroptosis, in various cell types [ 32 , 33 ]. Apoptosis and necroptosis may play a role in IAV-induced impairment of syncytialization. Notably, ferroptosis was found to be the dominant form of cell death induced by H1N1 in our cellular model. The predominance of ferroptosis may reflect a trophoblast-specific response, as placental cells are known to exhibit high susceptibility to lipid peroxidation due to their unique metabolic profile. Nevertheless, the molecular mechanisms by which ferroptosis impairs syncytialization require further in-depth experimental exploration. TNF-α exerts a variety of physiological functions during pregnancy, including the remodeling of endometrial tissue and the regulation of trophoblast cell growth during their invasion into the endometrium [ 24 ]. One particularly crucial process in which TNF-α is believed to be involved is the formation of placental STB cells [ 27 ]. We investigated the interference of TNF-α with FSK-induced syncytialization in BeWo cells, focusing on its effects on the expression of placental function markers. Previous reports have indicated that TNF-α impairs β-hCG expression and cell fusion in human villous CTBs, reduces syncytial function, increases syncytial shedding, and elevates the release of cytokine factors through SphK1 activity [ 34 , 35 ]. Consistent with these findings, our in vitro data showed that knockdown of TNF-α not only significantly attenuated H1N1-induced disruption of trophoblast cell fusion but also reduced the abnormal upregulation of ferroptosis-related markers. Conversely, overexpression of TNF-α promoted ferroptosis and suppressed trophoblast cell fusion, and these effects were effectively reversed by Fer-1 treatment. Notably, despite the therapeutic potential of TNF-α inhibition in virus-infected pregnancies, its application in normal gestation warrants caution. Studies report that Etanercept in healthy rats did not affect maternal reproductive parameters but caused fetal-placental impairments, including reduced weights and increased abnormalities [ 36 ]. Similarly, anti-TNF-α therapies have been independently linked to adverse outcomes such as intrauterine growth restriction and spontaneous abortion in uninfected pregnancies [ 37 ]. These findings underscore the indispensable physiological role of TNF-α in placental development and collectively highlight its context-dependent function: while TNF-α exerts pro-ferroptotic effects in infected STB cells, it maintains critical physiological roles during healthy pregnancies. To further validate these in vitro observations, we established an H1N1-infected pregnant mouse model. In vivo results revealed that H1N1 viral infection significantly enhances ferroptosis and impairs syncytialization in placental tissues. Importantly, treatment with a TNF-α inhibitor effectively ameliorated H1N1-induced ferroptosis and restored trophoblast cell fusion in the placentas of infected pregnant mice, indicating a potential therapeutic approach for viral-induced placental dysfunction. Mechanistically, our results suggest that H1N1 infection triggers TNF-α upregulation, which in turn induces ferroptosis and impairs trophoblast fusion. However, the precise molecular mechanisms underlying this process remain to be further elucidated in future studies. After demonstrating the negative effect of TNF-α on trophoblast cell fusion, we next explored the downstream genes of TNF-α. ISG15, a member of the interferon-stimulated gene family, is induced by type I interferons (IFNα/β) [ 38 ]. ISG15 proteins are covalently attached to target proteins through a three-step enzymatic process known as ISGylation, which involves a series of enzymatic reactions analogous to those occurring in ubiquitination [ 39 ]. The expression level of ISG15 is remarkably elevated in various inflammatory diseases, cancers, neurodegenerative disorders, and tuberculosis [ 40 ]. Studies have shown that the upregulation of ISG15 is associated with innate immune responses [ 41 , 42 ]. Nevertheless, it remains uncertain whether ISG15 is involved in the mechanism through which IAV disrupts trophoblast syncytium formation. Bioinformatics analysis, combined with experimental validation, identified ISG15 as a downstream target gene of TNF-α. We then demonstrated that the knockdown of ISG15 could mitigate ferroptosis and further alleviate the damage to trophoblast cell fusion induced by H1N1 virus infection. Additionally, we conducted rescue experiments to confirm the specific role of ISG15. Our findings demonstrated that inhibition of ISG15 significantly attenuated the pro-ferroptotic effect and the disruptive effect on trophoblast cell fusion induced by TNF-α overexpression. As is widely recognized, TNF-α is a multifunctional Th1 cytokine that exerts diverse functions by binding to cell surface receptors, namely TNFR1 and TNFR2 [ 43 ]. By contrast, as a cytoplasmic ubiquitin-like molecule, ISG15 regulates target protein functions at the cytoplasmic level via post-translational modification. However, the mechanisms underlying TNF-α-mediated regulation of ISG15, whether through the aforementioned cell surface receptors or other alternative pathways, remain poorly understood and require further investigation. Subsequently, through bioinformatics analysis and a series of experiments, EGR1 was identified as a downstream target of ISG15. EGR1, characterized as a zinc finger transcription factor, is known to target specific genes and enhance their transcriptional activity [ 44 ]. A large number of studies have focused on exploring the role of EGR1 in embryonic implantation and preeclampsia due to its significant involvement in modulating placental trophoblast cell differentiation, angiogenesis, and migration [ 45 , 46 ]. Prior investigations have shown that EGR1 is upregulated during the regulation of placental inflammation [ 47 ]. Moreover, in placental tissues from preeclampsia patients, EGR1 is aberrantly downregulated, and this downregulation leads to a notable reduction in β-hCG concentration in trophoblast cells [ 48 ]. No research to date has directly explored the function of EGR1 in the disruption of the placental barrier by IAV. Partially consistent with this observation, we confirmed that EGR1 reduced β-hCG expression and impaired trophoblast cell fusion. EGR1 is a gene that promotes ferroptosis. Previous studies demonstrated that GPX4 regulates ferroptosis through the upregulation of EGR1 in triple-negative breast cancer cells [ 49 ]. Moreover, it has been extensively reported that EGR1 promotes ferroptosis and intervertebral disc cartilage degeneration through the MAP3K14/NF-κB axis [ 50 ]. Our findings indicated that EGR1 knockdown mitigated the adverse effects of ferroptosis and the impairment of BeWo cell fusion induced by ISG15 overexpression. Additionally, we identified a TNF-α-mediated mechanism through the ISG15/EGR1 axis, which promotes ferroptosis and disrupts trophoblast cell fusion. Significantly, we discovered a positive regulatory loop between TNF-α and EGR1: TNF-α upregulates EGR1 expression in trophoblast cells, whereas EGR1 enhances TNF-α transcription. Once this feedback loop is activated, the downstream expression of ISG15 is triggered, which further exacerbates ferroptosis and impairs trophoblast cell fusion. Nevertheless, the precise molecular mechanisms underlying the interaction between TNF-α and EGR1 have yet to be comprehensively elucidated and require further investigation. This study has certain limitations. Specifically, further in vivo validation of the experimental findings and their clinical translation are necessary. As a ubiquitin-like modifier, ISG15 has been demonstrated to directly modify the transcription factor HIF-2α, resulting in polymorphic modification products. Notably, this modification does not trigger degradation; instead, it elicits changes in subcellular localization and protein–protein interactions. Specifically, ISGylation enhances the translational regulatory activity of HIF-2α [ 51 ]. Given the functional parallels between HIF-2α and EGR1, as well as the broad substrate specificity of ISG15, it is conceivable that ISG15 may similarly modulate EGR1 function via ISGylation, potentially through stabilizing EGR1 or modulating its transcriptional activity. Further experimental investigations are needed to determine whether EGR1 undergoes ISG15-mediated post-translational modification and to assess the functional consequences of this modification. In conclusion, our transcriptomic analysis reveals the impact of H1N1 virus infection on trophoblast cell fusion. At the cellular level, we demonstrate that H1N1 infection triggers ferroptosis and disrupts trophoblast cell fusion through the TNF-α/ISG15/EGR1 axis. Importantly, we present direct evidence for a positive feedback loop between TNF-α and EGR1, which amplifies the downstream effects of ISG15. These findings uncover a novel signaling pathway mediating ferroptosis and advance our understanding of the pathophysiological mechanisms by which IAV impairs the placental barrier function.

Influenza A virus (IAV), an enveloped virus with a segmented negative-sense RNA genome belonging to the Orthomyxoviridae family, has been responsible for four pandemics in human history, leading to millions of fatalities [ 1 ]. Influenza viruses are categorized into subtypes on the basis of the antigenic properties of hemagglutinin (H) and neuraminidase (N) [ 2 ]. Currently, only H1N1 and H3N2 subtypes continue to circulate in the human population. Pregnancy is acknowledged as a crucial risk factor for severe illness subsequent to IAV infection [ 3 ]. Maternal IAV infection has the potential to elevate the risk of neonatal complications and adverse pregnancy outcomes, encompassing spontaneous abortion, preterm birth, and neonatal mortality [ 4 ]. Moreover, IAV infection during pregnancy has been strongly associated with long-term health implications and developmental challenges in the offspring, including chronic immune disorders, cardiovascular diseases, and schizophrenia [ 5 ]. Oster et al. reported that prenatal maternal influenza may be associated with right-sided obstructive lesions in all infants and atrioventricular septal defects in infants with Down syndrome [ 6 ]. Several studies have linked prenatal H1N1 infection to dysregulated myelination-related gene expression in the cerebellum and hippocampus, suggesting that influenza virus exposure during pregnancy may increase the risk of neurodevelopmental disorders in the offspring [ 7 , 8 ]. The placenta serves to establish the maternal–fetal interface and assumes a pivotal role in the exchange of gases, nutrients, and waste products. It is also indispensable for the synthesis of pregnancy hormones and offers immune and mechanical safeguards for the developing fetus [ 9 ]. The chorionic villi are composed of multinucleated syncytiotrophoblast (STB) cells that coat the surface of the placenta, along with mononucleated cytotrophoblast (CTB) cells. During the differentiation of CTB cells into STB cells, CTB cells undergo fusion to form the STB layer, a process denoted as syncytialization [ 10 ]. Aberrant formation of the STB layer has been associated with deficiencies in placental function, which may give rise to preeclampsia, fetal growth restriction, and miscarriage [ 11 ]. Consequently, exploring the mechanisms underlying abnormal STB formation could provide effective strategies for deciphering the pathogenesis of IAV infection during pregnancy. Existing studies have confirmed that influenza viruses do possess the ability to infect the placenta. A study by Gu et al. demonstrated through systematic autopsy of H5N1 fatal cases that the virus was present in placental cytotrophoblasts and Hofbauer cells, with viral detection in multiple fetal organs, providing direct evidence for vertical transmission of influenza viruses [ 12 ]. Another study showed that the H1N1/09 virus can effectively infect placental cell lines in vitro, with HA protein expression detected in fresh placental tissues. These findings indicate that H1N1 virus has the potential to infect the placenta [ 13 ]. In line with these findings, our prior studies have shown that IAV can traverse the placental barrier by compromising the function of STB cells [ 14 ]. Nevertheless, the underlying pathological alterations in STB cells and the mechanisms implicated in this disruption subsequent to IAV infection remain to be clarified. Ferroptosis, a newly identified iron-dependent cell death mode, is characterized by the accumulation of reactive oxygen species (ROS) and lipid peroxides [ 15 ]. This process is intricately linked to key molecular events, including the depletion of glutathione (GSH), glutathione peroxidase 4 (GPX4), and solute carrier family seven member 11 (SLC7A11) [ 16 , 17 ]. Moreover, ferroptosis involves the accumulation of intracellular iron contents, oxidation products such as malondialdehyde (MDA), and the upregulation of acyl-CoA synthetase long-chain family member 4 (ACSL4) and transferrin receptor 1 (TFRC) [ 18 ]. Ferroptosis plays a significant pathological role in virus-associated infections and inflammatory diseases, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), hepatitis B virus (HBV), and human immunodeficiency virus (HIV) [ 19 ]. Viruses may induce ferroptosis by suppressing GPX4 activity, depleting intracellular GSH levels, modulating iron uptake-related proteins such as TFRC, and disrupting the system Xc − /GPX4 axis, thereby promoting the accumulation of intracellular free iron and lipid peroxides. Accumulating evidence suggests that ferroptosis also occurs in the human placenta and trophoblasts. One study revealed that PLA2G6 mitigated ferroptosis in trophoblasts and the placenta [ 20 ]. Another investigation revealed that inhibiting miR-30b-5p expression, in conjunction with the administration of ferroptosis inhibitors, alleviated symptoms of preeclampsia in a rat model [ 21 ]. Furthermore, a recent chicken embryo model study demonstrated that H1N1, a prevalent subtype of IAV, triggers oxidative stress-induced ferroptosis in fetal liver and lung tissues, leading to parenchymal cell death, tissue fibrosis, and impaired fetal development, thereby elucidating a key mechanism of IAV-induced placental injury [ 22 ]. Despite these advances, the specific role of ferroptosis in the impairment of the placental barrier caused by IAV remains largely uninvestigated. The cytokine tumor necrosis factor-alpha (TNF-α) is a pro-inflammatory cytokine secreted by macrophages, adipocytes, and the placenta [ 23 ]. Under normal physiological conditions, TNF-α at appropriate levels plays a crucial role in regulating the fate, proliferation, and differentiation of diverse cell types, including CTB cells [ 24 ]. Nevertheless, elevated levels of TNF-α have been closely associated with various pregnancy complications, such as recurrent spontaneous abortion, preterm birth, and hypertensive disorders during pregnancy [ 25 , 26 ]. Additionally, sequencing analysis demonstrated that the expression of TNF-α was significantly higher in H1N1-infected FSK-treated BeWo cells compared to that in the uninfected FSK-treated control cells. Based on these findings, we hypothesized that IAV infection causes damage to STB cells through the TNF-α signaling pathway. In this study, we aimed to investigate the specific mechanisms by which TNF-α mediates H1N1-induced syncytial disintegration. We tested the hypothesis that H1N1 virus infection upregulates TNF-α expression in FSK-treated BeWo cells, disrupting syncytial formation and compromising placental barrier integrity. Secondly, through bioinformatics analysis of RNA-seq data and functional experiments, we demonstrated that TNF-α triggers ferroptosis, leading to placental damage. Finally, we elucidated the molecular mechanisms by which TNF-α promotes ferroptosis through the ISG15/EGR1 axis. On the other hand, EGR1 promotes TNF-α transcription, establishing a feedback loop for TNF-α production. Consequently, these findings suggest that targeting TNF-α might be a new therapeutic strategy for IAV infection during pregnancy.

Additional file1 Additional file2 Additional file1 Additional file2