Mechanisms of vaginal bacteria-induced fetal membrane damage and tocilizumab’s therapeutic potential in PPROM

Abstract

Background

This study investigates the deleterious effects of vaginal opportunistic pathogens on fetal membranes and evaluates the therapeutic potential of tocilizumab (TCZ), an interleukin-6 (IL-6) receptor antagonist, in attenuating these effects, aiming to improve strategies for the prevention and treatment of preterm prelabor rupture of membranes (PPROM).

Methods

Bioinformatics analyses were conducted to identify inflammatory pathways associated with PPROM. Clinical specimens, including placental tissue and peripheral blood samples, were collected postpartum. Placental pathology was performed to assess the incidence of histological chorioamnionitis. Peripheral blood levels of IL-6 were quantified using enzyme-linked immunosorbent assay (ELISA). The expression of phosphorylated JAK2 and STAT3, key components of the JAK-STAT signaling pathway, in fetal membrane tissues was examined by Western blotting. Immunohistochemistry (IHC) was employed to detect the expression and localization of matrix metalloproteinase-9 (MMP-9) and its tissue inhibitor-1 (TIMP-1) in placental tissues. For in vitro experiments, a co-culture model incorporating bacteria and fetal membrane-derived WISH cells was established to evaluate the effects of bacterial infection on cell viability and apoptosis. Subsequently, IL-6 levels in the cell supernatants were measured. Western blotting was used to detect the activation of the JAK-STAT pathway in bacterially infected WISH cells, and immunocytochemistry (ICC) was performed to assess the expression and localization of MMP-9 and TIMP-1. Finally, the modulatory effects of TCZ on bacterially infected WISH cells were analyzed.

Results

Bioinformatics analysis identified inflammatory pathways, particularly the JAK-STAT pathway, as being associated with PPROM. Clinical samples revealed a higher incidence of clinic chorioamnionitis(CCA) in PPROM cases. Elevated peripheral plasma IL-6 levels were detected in individuals with PPROM, accompanied by increased phosphorylation of JAK2 and STAT3 in fetal membrane tissues, along with upregulated MMP-9 and downregulated TIMP-1 expression. In vitro, infection with Streptococcus agalactiae(S.agalactiae), Enterococcus faecalis(E. faecalis), or Escherichia coli (E.coli) significantly reduced WISH cell viability and induced apoptosis. Bacterial infection also led to markedly increased IL-6 levels in culture supernatants. In infected WISH cells, enhanced p-JAK2 and p-STAT3 were observed, together with elevated MMP-9 and reduced TIMP-1 expression. TCZ treatment reduced IL-6 levels. It inhibited JAK2 and STAT3 phosphorylation in fetal membrane tissues. MMP-9 expression was downregulated, and TIMP-1 expression was restored.

Conclusions

Vaginal opportunistic pathogens contribute to PPROM by compromising fetal membrane integrity through activation of inflammatory pathways such as JAK-STAT, disruption of the MMP-9/TIMP-1 balance, reduction of cell viability, and induction of apoptosis. TCZ, by targeting the IL-6 signaling axis, demonstrates potential to intervene in this pathological process, offering a promising therapeutic strategy for the prevention and treatment of PPROM.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12884-026-09591-3.

Background

Preterm prelabor rupture of membranes (PPROM), defined as rupture of the fetal membranes before 37 weeks of gestation, is a major cause of preterm birth and is strongly associated with significant maternal and neonatal morbidity, PPROM occurs in approximately 2–4% of pregnancies and is characterized by a multifactorial pathogenesis [1]. Although intrauterine infection is widely recognized as a key initiating factor, the precise mechanisms remain to be fully elucidated [2]. The integrity of the fetal membranes is maintained by a balance of multiple factors. however, excessive degradation of the extracellular matrix (ECM) is a direct cause of membrane rupture, primarily driven by an imbalance between matrix metalloproteinases (MMPs) and their endogenous tissue inhibitors (TIMPs) [3].

PPROM is strongly associated with placental inflammation. In this context, Inflammatory responses triggered by pathogen-associated molecular patterns and damage-associated molecular patterns can activate immune pathways, thereby contributing to fetal development [2]. Furthermore, our previous investigations [4] demonstrated that dysbiosis of the vaginal microbiota, characterized by a reduction in lactobacilli and an overgrowth of opportunistic pathogens such as Enterococcus faecalis, Escherichia coli, and Streptococcus agalactiae, is significantly associated with the occurrence of PPROM. Such microbial alterations may contribute to amniotic membrane injury by triggering host inflammatory cascades [5].

Upon ascending infection, bacterial components are recognized by pattern recognition receptors on host cells, triggering the release of pro-inflammatory cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α).These cytokines stimulate the expression and activation of matrix-degrading enzymes in fetal membrane cells. Among the MMP family, specific isoforms play a critical role in this process, MMP-2, MMP-8, and MMP-9 have been well-characterized for their capacity to degrade collagen and other ECM components within the fetal membrane. Inflammatory stimuli have been shown to upregulate the activity of these enzymes [3]. Additionally, Elevated levels of MMP-9 have been detected in patients at high risk of preterm birth, indicating its potential utility as a predictive biomarker [6]. In the context of PPROM, MMP activity is closely associated with chorioamnionitis (CAM), which may arise from either bacterial infection or sterile inflammation [3].

Cellular heterogeneity and intercellular interactions within the fetal membrane contribute significantly to the pathogenesis of PPROM. Single-cell RNA sequencing has revealed distinct cellular subtypes in the amniotic membrane and adjacent placental tissues, highlighting specific interaction patterns among trophoblast subpopulations under various pathological conditions. These findings offer novel insights into the molecular mechanisms underlying PPROM [7]. Collectively, the evidence underscores the central role of MMP/TIMP imbalance in PPROM and identifies promising targets for early prediction and therapeutic intervention.

Given that IL-6 is a well-established mediator of inflammation during infection and signals primarily through the JAK/STAT pathway, we hypothesized that this signaling cascade may represent the critical link between ascending bacterial infection associated PPROM and MMP mediated ECM degradation. It has been reported that elevated levels of pro-inflammatory cytokines, particularly IL-6, have been consistently detected in the amniotic membranes of PPROM patients [3]. IL-6 is known to exert its biological effects by binding to the gp130 receptor, leading to JAK phosphorylation and subsequent STAT3 activation, a pathway that may contribute to inflammation-induced structural degradation of the amniotic membranes. Additionally, TNF-α and IL-6 have been shown to induce apoptosis and inflammatory responses in amniotic membrane cells by upregulating ADAMTS-9, a process involving the JAK/STAT signaling pathway [8, 9].

Recent studies have demonstrated significantly elevated STAT3 expression in fetal membranes from infection-induced preterm births, with levels positively correlated with IL-6, confirming activation of the JAK-STAT pathway in this context [10]. Furthermore, evidence indicates that STAT3 can directly bind to the MMP-9 promoter and enhance its transcription [11], implicating the JAK-STAT pathway in the regulation of MMP-9. Nevertheless, the role of the IL-6/JAK-STAT/MMP-9 axis in ECM degradation and membrane rupture in infection-induced PPROM remains to be elucidated. Therefore, we hypothesized that bacterial infection induces IL-6 overexpression, which in turn activates JAK-STAT signaling, leading to upregulated MMP-9 transcription, disruption of the MMP/TIMP balance, excessive ECM degradation, and ultimately PPROM. This study aims to validate this mechanism through analysis of placental tissues from infection-associated PPROM cases and a co-culture model of human amniotic epithelial cells with pathogenic bacteria, thereby identifying potential therapeutic targets for infection-induced PPROM.

In summary, the pathogenesis of PPROM is multifactorial, involving intrauterine infection, inflammatory responses, and excessive ECM degradation.The imbalance between MMPs and TIMPs represents a direct mechanistic link to membrane rupture, with intrauterine infection serving as a critical upstream trigger. Further investigation into these mechanisms is warranted to inform the development of more effective preventive and therapeutic strategies, ultimately improving maternal and neonatal outcomes.

Methods

Bioinformatics analysis

Gene microarray datasets were retrieved from the NCBI Gene Expression Omnibus (GEO) public database (accession: GSE47619, source: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? acc=GSE47619). Samples were categorized into three groups based on clinical phenotype: term labor (TL), preterm labor (PTL), and PPROM.

All datasets were downloaded in MINiML format, which contains complete GSE records including all platforms and samples. For unnormalized datasets, log₂ transformation was uniformly applied. If a dataset had not been standardized, quantile normalization was performed using the normalize, quantiles function from the preprocessCore package in R. Probe IDs were converted to gene symbols based on corresponding platform annotation information. Probes mapping to multiple genes were excluded, and for genes corresponding to multiple probes, the average expression value was calculated. To correct for batch effects within the same dataset and platform, the removeBatchEffect function from the limma package was employed. For combined analysis involving multiple datasets or different platforms within a single dataset, common gene symbols were first extracted across datasets. Different datasets or platforms were then designated as separate batches, and batch effects were removed using the removeBatchEffect function. Preprocessing outcomes were evaluated using box plots to assess data normalization, and principal component analysis (PCA) was conducted before and after batch effect removal to evaluate the effectiveness of batch correction [12–14].

Differentially expressed genes (DEGs) between the TL, PTL and PPROM groups were identified using the limma package in R. Genes meeting the thresholds of |log₂ fold change (FC)| ≥ 1.5 and an adjusted P-value (Benjamini–Hochberg) < 0.05 were considered statistically significant DEGs.

To elucidate the functional implications of the identified DEGs, Gene Ontology (GO) enrichment analysis was performed. DEGs were mapped to the GO database (http://www.geneontology.org), a comprehensive and computational model of biological systems that classifies gene functions into three orthogonal domains: cellular components, biological processe, and molecular functions [15]. This analysis aimed to identify functional categories significantly overrepresented among the DEGs, thereby revealing key biological processes and pathways potentially involved in PPROM pathogenesis.

Furthermore, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was conducted using the KEGG database (https://www.genome.jp/kegg) to identify the principal metabolic and signal transduction pathways associated with the DEGs [16]. Pathways are systematically categorized based on their functional roles, including cellular processes, genetic information processing, and environmental information processing. The statistical significance of enrichment for each KEGG pathway was assessed using the hypergeometric distribution test. This test calculates the probability of observing a given number of DEGs within a specific pathway compared to what would be expected by chance, based on the species background gene set. To control for the false discovery rate (FDR) due to multiple comparisons, P-values were adjusted using the Benjamini–Hochberg method. An adjusted P-value (FDR) < 0.05 was considered statistically significant for pathway enrichment. Pathway-level expression values were subsequently ranked based on the magnitude of intergroup differences, and the top 30 enriched pathways were selected for visualization to provide an overview of the most prominent signaling alterations associated with PPROM.

Study design and participant selection

The study cohort consisted of women with singleton pregnancies beyond 14 weeks of gestation (verified via last menstrual period or obstetric assessment) who planned to receive comprehensive prenatal care and delivery at the Department of Obstetrics and Gynecology, West China Second University Hospital, between January 2020 and December 2023. Written informed consent was obtained from all subject.

Exclusion criteria included: (a) documented history of substance abuse, smoking, exposure to toxic chemicals, uterine contractions, preterm delivery, cervical dilation, or premature membrane rupture. (b) Significant concurrent comorbidities (including hypertension, diabetes mellitus, malignant tumors, immune disorders). (c) genital tract infections caused by specific pathogens (fungi, Trichomonas, viruses, Mycoplasma spp., Chlamydia spp.). (d) Antibiotic use within the preceding week or circumstances predisposing to loss to follow-up. (e) Incidents of vaginal bleeding, abdominal trauma, or intercourse during pregnancy. Pregnancy outcomes were prospectively followed. Eligible participants were subsequently divided into two groups: the PPROM and TL group. The diagnosis of PPROM was established based on the presence of amniotic fluid leakage from the vagina or pooling in the posterior fornix containing vernix caseosa or meconium before 37 weeks of gestation, in conjunction with at least one of the following criteria: vaginal fluid pH ≥ 6.5 as determined by pH test paper, identification of ferning patterns upon microscopic examination of dried vaginal fluid, or absence of the amniotic sac upon amnioscopy [1]. According to gestational age at rupture, the PPROM group was further subclassified into late PPROM (L-PPROM; 34–36 weeks of gestation) and preterm PPROM remote from term (P-PPROM; 24–33 weeks of gestation).

Measurement of plasma IL-6

To quantitatively assess systemic or local inflammatory responses, IL-6 levels were measured by enzyme-linked immunosorbent assay (ELISA) (Molecular Devices, USA) in plasma samples obtained from 60 randomly selected cases in the TLgroup and 67 cases in the PPROM group. Random selection was performed using SPSS software to ensure sample representativeness. The ELISA process was summarized as follows, wells pre-coated with anti-IL-6 antibodies were incubated with cell supernatant or serum samples. Standards or samples (50 µL) and HRP-conjugated antibody (100 µL) were added and incubated at 37 °C for 60 min. Following five washes, TMB substrate was added and incubated at 37 °C for 15 min in the dark. The reaction was terminated, and optical density (OD) values were measured at 450 nm. Each sample was assayed in triplicate, and the mean optical density was used to calculate IL-6 concentrations based on a standard curve [17].

Placental histopathologic examination

Placental tissues from mothers in the PPROM and TL groups were collected immediately after delivery. A total of 20 placental tissue samples were randomly selected from each of the TL and PPROM groups for histopathological evaluation. A tissue sample measuring approximately 5 cm² was obtained from the site of membrane rupture. Microscopic (Olympus, Japan) examination was then performed to assess the presence and characteristics of histological chorioamnionitis (HCA) in the placental tissues. Chorioamnionitis scores are presented in Tables 1 [18].

Table 1.

Placenta reaction patterns related to amniotic fluid infection: nomenclature and definitions

Diagnostic categories	Suggested diagnostic terminology	Definitions
Maternal inflammatory response
Stage
1—Early	Acute subchorionitis or chorionitis	PMN in subchorionic fibrin and/or membrane trophoblast
2—Intermediate	Acute chorioamnionitis	Diffuse-patchy PMN in fibrous chorion and/or amnion
3—Advanced	Necrotizing chorioamnionitis	PMN karyorrhexis, amniocyte necrosis, and/or amnion basementmembrane thickening/hypereosinophilia
Grade
1—Mild–moderate	No special terminology required	Not severe as defined below
2—Severe	Severe acute chorioamnionitis or with subchorionic microabscesses	Confluent PMN ( 10  20 cells in extent) between chorion anddecidua;  3 isolated foci or continuous band
Other	Chronic (or subacute) chorioamnionitis	Subamnionic mononuclear cell infiltrate with occasional PMN(meconium and hemosiderin-laden macrophages excluded)
Fetal inflammatory response
Stage
1—Early	With chorionic vasculitis or umbilical phlebitis	Intramural PMN-chorionic vessels and/or umbilical vein
2—Intermediate	With umbilical vasculitis (one or two arteries±vein) or umbilical panvasculitis (all vessels)	Intramural PMN-umbilical artery or arteries ( ±umbilical vein)
3—Advanced	With (subacute) necrotizing funisitis or with concentric umbilical perivasculitis	PMN±associated debris in concentric bands-rings-halos around one or more umbilical vessels
Grade
1—Mild–moderate	No special terminology required	Not severe as defined below
2—Severe	With a severe fetal inflammatory response or with intense chorionic (umbilical) vasculitis	Near confluent intramural PMN-chorionic and/or umbilical vessels with attenuation/degeneration of VSMC
Other	With associated fetal vessel thrombi	Recent thrombosis associated with intramural PMN
Other specific features	Peripheral funisitis	Focal aggregates of PMN at the umbilical cord surface
	Acute villitis	PMN in villous stroma (or between trophoblast and stroma)
	Acute intervillositis with intervillous abscesses	Patchy-diffuse PMN in intervillous space
	Decidual plasma cells	Unequivocal plasma cells in decidua basalis or capsularis

PMN polymorphonuclear leukocyte, VSMC vascular smooth muscle cell

IHC analysis of IL-6 in placental tissues

A total of 3 placental tissue samples were randomly selected from each of the TL and PPROM groups for IHC evaluation, placental tissues were collected from the rupture site, fixed in 4% paraformaldehyde (PFA) (Hunan BKMAM, China), and processed through dehydration, clearing, and paraffin embedding. Sections were cut, mounted, dried, and underwent deparaffinization, rehydration, antigen retrieval, and blocking. Incubation with primary antibodies, IL-6 (Wuhan Sanying, China)1:100, was performed overnight at 4°C, followed by secondary antibody application, 3, 3’-diaminobenzidine (DAB) development, hematoxylin counterstaining, dehydration, clearing, and mounting. Quantitative analysis was performed using the Image-Pro Plus 6.0 image analysis system. In high-power fields, the integrated optical density (IOD) and positively stained area were measured, and the mean density (MD = IOD/area) was calculated. The average MD from three images per sample was used for subsequent analysis. For each sample, three random high-power fields were captured.

Western blot analysis of fetal membrane

After placental delivery, fetal membrane tissue specimens (≥ 5 cm²) from the PPROM and TL groups were immediately collected on ice, rinsed with pre-cooled saline, and stored in a stabilizing solution at 4 °C overnight before transfer to -80 °C for long-term storage. A total of 3 placental tissue samples were randomly selected from the TL and PPROM groups for WB evaluation.

For protein extraction, total protein was isolated from frozen tissue samples by pulverization followed by lysis in RIPA buffer (150–250 µL per 20 mg tissue) on ice for 10 min. Lysates were centrifuged at 14,000×g for 5 min, and the supernatant containing total proteinwas collected. Protein concentrations were determined using a BCA assay kit (Beyotime, China). The expression levels of JAK2(Abcam, UK), p-JAK2(Abcam, UK), STAT3(Abcam, UK), and p-STAT3(Abcam, UK) were assessed by Western blotting.

Primary antibodies were diluted as follows: JAK2 (1:5,000), p-JAK2 (1:5,000), STAT3 (1:2,000), p-STAT3 (1:2,000), and β-Actin (Abclonal, China) (1:100,000). Target protein expression was calculated as the IOD ratio relative to β-Actin, with the control group normalized to 1 for relative quantification.

The steps for performing Western blotting were as follows, proteins were separated by SDS-PAGE, transferred onto PVDF membranes. Membranes were blocked and incubated with specific primary antibodies followed by HRP-conjugated secondary antibodies. Protein bands were detected using enhanced chemiluminescence and quantification by densitometry [19].

IHC analysis of MMP-9 and TIMP-1 in placental tissues

A total of 3 placental tissue samples were randomly selected from each of the TL, L-PPROM and P-PPROM groups for IHC evaluation, placental tissues were collected from the rupture site, fixed in 4% PFA (Hunan BKMAM, China), and processed through dehydration, clearing, and paraffin embedding. Sections were cut, mounted, dried, and underwent deparaffinization, rehydration, antigen retrieval, and blocking. Incubation with primary antibodies, MMP-9 (Wuhan Sanying, China) 1:100 and TIMP-1(Wuhan Sanying, China) 1:50 were performed overnight at 4°C, followed by secondary antibody application, 3,3’-DAB development, hematoxylin counterstaining, dehydration, clearing, and mounting. Quantitative analysis was performed using the Image-Pro Plus 6.0 image analysis system. In high-power fields, the IOD and positively stained area were measured, and the MD was calculated. The average MD from three images per sample was used for subsequent analysis. For each sample, three random high-power fields were captured.

Establishment of the bacterial-cell co-culture models

Based on our preliminary findings [4] showing an average ratio of PPROM-associated bacteria to epithelial cells of 16:1 in vaginal secretions and the identification of Streptococcus agalactiae (S. agalactiae), Enterococcus faecalis (E. faecalis), and Escherichia coli (E. coli) as the most prevalent opportunistic pathogens in PPROM cases [22, 23], standard strains of these bacteria were selected for in vitro experiments. Accordingly, S. agalactiae ATCC^® 12,386™, E. faecalis ATCC^® 29,212™, E. coli ATCC^® 25,922™, and L.crispatus ATCC^® 33,197™ (all obtained from BioVector NTCC Collection Center), were co-cultured with the human amniotic epithelial cell line WISH (ATCC CCL-25, Cell Bank, CAS). An (multiplicity of infection) MOI of 20:1 was used as the prime concentration for the bacterial infection assays.

Cell viability assay

WISH cell viability was assessed using the Cell Counting Kit-8 (CCK-8; Dojindo, Japan). Briefly, WISH cells were seeded into 96-well plates at a density of 1 × 10⁴ cells per well and cultured overnight. For bacterial concentration experiments, cells were co-incubated with various bacterial strains at increasing multiplicities of infection (MOIs 2:1, 20:1, and 200:1) for 6 h. For time-course experiments, cells were co-incubated with bacteria at a fixed MOI (20:1) for indicated durations (2, 4, 6, and 24 h). After treatment, 10 µL of CCK-8 solution was added to each well and incubated for an additional 2.5 h at 37 °C. Absorbance was measured at 450 nm using a microplate reader (Bio-Rad, USA). Cell viability was calculated as the percentage of the untreated control group (set at 100%). All experiments were performed in triplicate and repeated three independent times.

SYTO/PI staining fluorescence

Sterile cell culture coverslips (24 mm × 24 mm) were placed into each well of a 6-well plate. WISH cells were seeded at a density of 1 × 10⁶ cells per well and incubated overnight to allow attachment. The culture medium was then removed, and cells were washed three times with phosphate-buffered saline (PBS). Antibiotic-free complete medium (without penicillin–streptomycin) was added to each well.

Except for the control group, experimental wells were co-incubated with S. agalactiae, E. faecalis, or E. coli at a MOI of 20:1 (bacteria: cells) for 10 h. Following co-culture, cells were washed four times with PBS. A volume of 500 µL of ready-to-use propidium iodide (PI) staining solution (KeyGen, China) and 2 µL of SYTO staining solution (KeyGen, China) were added to each well and mixed gently. Fluorescence microscopy was performed immediately. Viable cells exhibited green fluorescence, while necrotic cells showed red fluorescence. All experiments were independently repeated three times.

Annexin V-FITC/PI staining flow cytometry

For apoptosis analysis, WISH cells were seeded at 1 × 10⁶ cells per well in 6-well plates and incubated overnight. After removal of the culture medium, cells were washed three times with PBS, and antibiotic-free complete medium was added. Experimental wells were inoculated with S. agalactiae, E. faecalis, or E. coli at an MOI of 20:1 and co-cultured for 2, 4, 6, or 24 h. Following co-culture, cells were washed four times with PBS, then resuspended in 500 µL of Annexin V Binding Buffer. Annexin V-FITC solution (5 µL; KeyGen, China) was added, and the tube was gently vortexed. Subsequently, 5 µL of PI (KeyGen, China) solution was added, and the tube was vortexed again. The reaction mixture was incubated at room temperature in the dark for 10 min. Samples were analyzed by flow cytometry within 1 h of staining completion. All experiments were independently repeated three times.

Measurement of supernatants IL-6

WISH cells were seeded at 2.5 × 10⁶ cells per flask and cultured overnight. After removal of the culture medium, cells were washed three times with PBS, and antibiotic-free complete medium was added. Experimental flasks were inoculated with S.agalactiae, E.faecalis, or E.coli at an MOI of 20:1 and co-cultured for 6 h. Supernatants were then collected for IL-6 quantification by ELISA. The ELISA assay for IL-6 in co-culture supernatants were conducted following the procedure detailed earlier for plasma samples. After the reaction was terminated, OD values were measured at 450 nm. Each sample was assayed in triplicate, and all experiments were performed in triplicate.

Western blot of infected WISH cell models

WISH cells were seeded at 2.5 × 10⁶ cells per flask and cultured overnight. After removal of the culture medium, cells were washed three times with PBS, and antibiotic-free complete medium was added. Experimental flasks were inoculated with S.agalactiae, E.faecalis, or E.coli at an MOI of 20:1 and co-cultured for 6 h. Cells were collected, and total protein extracts were prepared for Western blot analysis. The procedure for electrophoresis, transfer, and immunodetection was identical to that described for placental tissue samples. All experiments were independently repeated three times.

ICC analysis of infected WISH cell models

For immunocytochemistry (ICC), sterile coverslips were placed in 6-well plates, and WISH cells were seeded at 1 × 10⁶ cells per well. After overnight culture, cells were infected with S.agalactiae, E.faecalis, or E.coli at an MOI of 20:1 for 6 h. Following infection, cells were washed with PBS, fixed with 4% PFA(Hunan BKMAM, China), subjected to antigen retrieval using citrate buffer, and blocked with goat serum. Immunostaining was performed by incubating with primary antibodies against MMP-9 (Wuhan Sanying, China) 1:100 and TIMP-1(Wuhan Sanying, China) 1:50 overnight at 4 °C, followed by incubation with secondary antibodies at 37 °C for 30 min. Detection was achieved using DAB development, followed by hematoxylin counterstaining, dehydration, clearing, and mounting with neutral gum. For each sample, three random high-power fields were captured, and the MD was calculated as IOD/area using Image-Pro Plus 6.0. All experiments were independently repeated three times.

Effect of tocilizumab on bacterially infected WISH cells

WISH cells were seeded at a density of 2.5 × 10⁶ cells per 25 cm² flask and cultured overnight. Following washing with PBS, cells were subjected to the following treatments: the tocilizumab(TCZ) group received 1 µg/mL TCZ (AtaGenix, China) in antibiotic-free medium; the IgG group received 1 µg/mL human IgG isotype control (Invitrogen, USA); and the control group received antibiotic-free medium alone. Except for the control group, flasks were infected with S. agalactiae, E. faecalis, or E. coli at an MOI of 20:1 for 6 h. Culture supernatants were collected and centrifuged at 1000×g for 20 min, and IL-6 levels were measured by ELISA. Adherent cells were lysed with RIPA buffer on ice, and total protein was extracted for BCA quantification and Western blot analysis of JAK2, p-JAK2, STAT3, and p-STAT3. For ICC, cells grown on coverslips underwent identical treatment and infection conditions, followed by detection of MMP-9 and TIMP-1 expression. All experiments were independently repeated three times.

Statistical analysis

Statistical analysis was performed using GraphPad Prism (versions 8.0 and 10.0; GraphPad Software Inc., CA, USA). Continuous variables following a normal distribution were expressed as mean ± standard deviation (SD), and comparisons between two groups were conducted using Student’s t-test. Continuous variables did not follow a normal distribution were expressed as median with interquartile range (IQR), and comparisons between groups were performed using the Mann-Whitney U test. Categorical variables, presented as rates or proportions, were analyzed using the chi-square test. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) was employed. Two-way ANOVA was used to evaluate interactions and main effects in the context of TCZ treatment groups. Pairwise comparisons following ANOVA were performed using appropriate multiple comparison tests. A two-tailed P-value < 0.05 was considered statistically significant.

Results

The upregulation of gene expression within inflammatory signaling pathways

GO enrichment analysis revealed that DEGs between the PPROM and TL groups were primarily involved in biological processes related to immune activation, including leukocyte activation, degranulation, migration, and positive regulation of cytokine production. KEGG pathway analysis further indicated that upregulated genes were significantly enriched in inflammatory signaling pathways, such as the MAPK, PI3K-Akt, and JAK-STAT cascades. Integrated analysis of KEGG and GSEA demonstrated that the differentially expressed genes were most prominently enriched in the cytokine–cytokine receptor interaction pathway. Moreover, protein–protein interaction (PPI) network analysis identified IL-6 as a potential hub gene, suggesting its central role in mediating the associated biological processes (Fig. 1).

Fig. 1.

Bioinformatics analysis of PPROM. (a) Microarray datasets of maternal genes from the PPROM and TL groups were obtained from the NCBI GEO public database. GO enrichment analysis revealed that DEGs were significantly enriched in biological processes related to leukocyte activation. (b) KEGG pathway enrichment analysis indicated that upregulated genes in the PPROM group were predominantly associated with inflammatory signaling pathways, including the MAPK, PI3K-Akt, and JAK-STAT signaling cascades. (c) Integrated analysis using KEGG and GSEA further demonstrated that the DEGs were most significantly enriched in the “Cytokine-cytokine receptor interaction” pathway. (d) PPI network analysis of the DEGs identified hub proteins, including IL-6, suggesting their central role in orchestrating the relevant biological processes. Abbreviations: PPROM, preterm premature rupture of membranes; TL, term labor (control group); GEO, Gene Expression Omnibus; GO, Gene Ontology; DEGs, differentially expressed genes; KEGG, Kyoto Encyclopedia of Genes and Genomes; GSEA, Gene Set Enrichment Analysis; PPI, Protein-protein interaction

Clinical information and higher plasma IL-6 concentrations

A total of 1,863 pregnant women were enrolled and followed for pregnancy outcomes. PPROM occurred in 3.6% (67/1,863) of cases, while 60.39% (1,125/1,863) delivered at term (TL). Among the PPROM cases, 8 were classified as P-PPROM (occurring before 34 weeks of gestation) and 59 as L-PPROM (occurring between 34 and 36 weeks of gestation). Peripheral plasma IL-6 concentrations were measured by ELISA. The IL-6 level was significantly elevated in the PPROM group [median (IQR): 29.44 (6.82–78.35) pg/mL] compared with the TL group [3.42 (1.28–6.44) pg/mL] ( P < 0.001), indicating a heightened inflammatory response associated with PPROM (Table 2).

Table 2.

Maternal Characteristics and IL-6 levels in Women with Preterm Premature Rupture of Membranes and Term Labor

Item (Unit)	PPROM group(n = 67)	TL group(n = 1125, #n = 60)	P value
Ages(mean ± SD)	30.42 ± 3.75	30.36 ± 3.79	0.862
BMI ≥ 25(n,%)	45(67.16)	563(50.04)	< 0.001
Cesarean section(n,%)	24(35.82)	356(31.64)	0.397
CCA(n,%)	10(14.93)	39(3.47)	< 0.001
#IL-6(pg/mL)median (IQR)	29.44(6.82–78.35)	3.42(1.28–6.44)	< 0.001

Data are presented as mean ± standard deviation, median (IQR) or number (percentage). #60 randomly selected cases in the TL group

Abbreviations: PPROM preterm premature rupture of membranes, TL term labor, BMI body mass index, CCA clinical chorioamnionitis, IL-6 interleukin-6

Histopathological features and expression of IL-6, JAK2/STAT3, MMP-9/ TIMP-1 in placental tissues of the PPROM and TL groups

Higher incidence of HCA

Histological examination using HE staining of placental tissues from mothers in both the PPROM and TL groups revealed the following, Histological examination of placental tissues revealed marked differences in the incidence of chorioamnionitis between the TL and PPROM groups. In the TL group, mild chorioamnionitis was observed in 1 of 20 randomly selected samples (5%, 1/20). In contrast, chorioamnionitis was detected in 7 of 20 randomly selected samples from the PPROM group (35%, 7/20). Among these PPROM cases, the presence of neutrophils in the subchorionic space, indicative of mild chorioamnionitis, was observed in 4 samples (20%, 4/20). Neutrophil infiltration into the chorionic connective tissue, characteristic of moderate chorioamnionitis, was noted in 2 samples (10%, 2/20). Finally, neutrophil infiltration extending to the amnion, diagnostic of severe chorioamnionitis, was identified in 1 sample (5%, 1/20). (P < 0.001).

Upregulated JAK-STAT pathway in fetal membranes

To determine whether the JAK2/STAT3 signaling pathway is activated in PPROM, we examined the expression and phosphorylation levels of JAK2 and STAT3 in fetal membranes tissues from TL and PPROM groups by Western blot analysis. The results revealed that total JAK2 and total STAT3 protein levels were comparable between the TL and PPROM groups, indicating no significant alterations in overall expression(^ns P > 0.05). In contrast, the levels of p-JAK2 and p-STAT3 were significantly elevated in PPROM tissues compared with TL controls (p-JAK2: 2.7-fold increase, P < 0.001; p-STAT3: 7.6-fold increase, P < 0.001) (Fig. 2C-D). This enhanced phosphorylation status, without changes in total protein expression, indicates activation of the JAK2/STAT3 signaling pathway in PPROM. These findings suggest that JAK2/STAT3 pathway activation may contribute to the inflammatory and degradative processes underlying PPROM pathophysiology at the tissue level.

Fig. 2.

Histopathological features and expression of IL-6, JAK2/STAT3, MMP-9/ TIMP-1 in placental tissues of the PPROM and TL groups. (a) Representative HE Staining images of histological chorioamnionitis in placental tissue. (a1)The TL group did not exhibit significant inflammatory cell infiltration. (×100, Scale bar 200 μm) (a2) Mild chorioamnionitis in the PPROM group, characterized by neutrophil infiltration in the subchorionic space. (×100, Scale bar 200 μm) (a3) Moderate chorioamnionitis in the PPROM group, characterized by neutrophil infiltration in the chorionic connective tissue. (×100, Scale bar 200 μm) (a4) Severe chorioamnionitis in the PPROM group, characterized by neutrophil infiltration in the amnion(×100, Scale bar 200 μm) (indicated by arrows). (b)Representative IHC images of IL-6 expression(× 400, Scale bar 40 μm). (b1)The TL group showed diffuse weak positivity. (b2)The PPROM group exhibited Strong diffuse positivity. (c) Representative Western blot images of β-Actin, JAK2, p-JAK2,STAT3, and p-STAT3 in fetal membranesplacental tissues. (d) Relative quantification of Western blot bands from (c).Results are presented as fold change relative to the TL control group (set as 1.0). Compared with the TL group, p-JAK2: 2.7-fold increase; p-STAT3: 7.6-fold increase in the PPROM group. (e) Representative IHC images of MMP-9 and TIMP-1 in placental tissues from the TL, L-PPROM and P-PPROM group (× 400, Scale bar 40 μm), (e1-3)MMP-9,(e1) Diffuse weak positivity, (e2) Diffuse positivity, (e3)Strong diffuse positivity, (e4-6)TIMP-1, (e4) Strong diffuse positivity, (e5)Diffuse positivity, (e6) Diffuse weak positivity. (f) A semi-quantitative analysis of IHC from(e), for each sample, three random high-power fields were captured, and the MD was calculated as IOD/area. MMP-9 expression (TL: 0.14 vs. L-PPROM: 0.26 vs.P-PPROM: 0.34), TIMP-1 expression (TL: 0.33 vs.L-PPROM: 0.24 vs. P-PPROM: 0.2). 20 placental tissue samples were randomly selected of each group for HE Staining, 3 placental tissue samples were randomly for selected of each other experiments. All data were presented as mean ± SEM, ^nsP > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001 compared to the TL group. Abbreviations: PPROM, preterm premature rupture of membranes; TL, term labor (control group); IHC, immuno histochemical staining; JAK2, Janus kinase 2; p-JAK2, phosphorylated Janus kinase 2; STAT3, signal transducer and activator of transcription 3; p-STAT3, phosphorylated signal transducer and activator of transcription 3; β-Actin, beta-actin (loading control). L-PPROM, late preterm premature rupture of membranes (34–36 weeks of gestation); P-PPROM, preterm premature rupture of membranes remote from term (24–33 weeks of gestation); MMP-9, matrix metalloproteinase-9; TIMP-1, tissue inhibitor of metalloproteinases-1, MD, mean density

Elevated expression of IL6/MMP-9/TIMP-1 in placental tissues

To determine whether MMP/TIMP imbalance occurs in PPROM and to assess the involvement of IL-6, we examined IL-6, MMP-9, and TIMP-1 expression in placental tissues from TL and PPROM groups by IHC. IHC staining revealed that cells with negative staining appeared blue, while positive staining appeared yellow, with positive signals widely distributed in amnion, chorion, and decidual cells. The staining intensity reflected the relative protein expression levels. Semi-quantitative analysis using mean OD values from three randomly selected fields per sample demonstrated that IL-6 expression was significantly elevated in PPROM tissues compared with TL controls (TL: 0.12 vs. PPROM: 0.35, P < 0.01). Similarly, MMP-9 expression was significantly elevated in PPROM tissues (TL: 0.14 vs. PPROM: 0.32, P < 0.01), with a more pronounced increase in the early PPROM subgroup (P-PPROM: 0.34) than in the late PPROM subgroup (L-PPROM: 0.26). Conversely, TIMP-1 expression was markedly reduced in the PPROM group (TL: 0.33 vs. PPROM: 0.22, P < 0.01), with the lowest levels detected in the early PPROM subgroup (P-PPROM: 0.2) compared with the late PPROM subgroup (L-PPROM: 0.24). This reciprocal dysregulation of MMP-9 and TIMP-1, accompanied by elevated IL-6 expression, resulted in a substantially increased MMP-9/TIMP-1 ratio in PPROM, indicating a net shift toward a proteolytic environment that favors ECM degradation. These findings confirm that IL-6-mediated inflammation and MMP/TIMP imbalance are key features of PPROM pathophysiology at the tissue level (Fig. 2b, e and f).

Effects of bacterial concentration and exposure duration on WISH cell viability and apoptosis

Decreased cell viability under various bacterial concentrations

L.crispatus, GBS, E. faecalis, and E. coli were co-cultured with WISH cells at a MOI of 2:1, 20:1, and 200:1 for 6 h. Cell viability was assessed using the CCK8 assay. Compared with the control group, co-culture with L. crispatus at all MOIs resulted in no significant difference in viable cell numbers (P > 0.05), suggesting that this commensal bacterium exerts minimal cytotoxic effects on amniotic epithelial cells. In contrast, exposure to GBS, E. faecalis and E. coli resulted in significant, concentration-dependent reductions in cell viability, with the most pronounced inhibition observed at an MOI of 2:1 (P < 0.01, P < 0.05, and P < 0.001, respectively), MOI of 20:1 (P < 0.001, P < 0.01, and P < 0.001, respectively), MOI of 200:1 (P < 0.001, P < 0.001, and P < 0.001, respectively). Notably, the inhibitory potency varied among pathogenic species, E. coli exhibited the strongest cytotoxic effect, followed by GBS and E. faecalis (Fig. 3a and b). These findings indicate that pathogenic bacteria commonly associated with vaginal dysbiosis can directly impair amniotic cell survival in a dose-dependent manner, whereas L. crispatus appears non-cytotoxic.

Fig. 3.

Effects of bacterial concentration and exposure duration on WISH cell viability and apoptosis. (a) Cell map of each node in the CCK8(× 400, Scale bar 40 μm). (b) Effects of different bacterial species and concentrations on WISH cell viability, Cells were co-incubated with various bacteria at increasing MOIs, and cell viability was determined using the CCK-8 assay. Relative viability was normalized to control (set at 1.0). (c) Time-dependent effects of different bacterial species on WISH cell viability. Cells were co-incubated with bacteria for the indicated durations, and viability was assessed by CCK-8 assay. Relative viability was normalized to control (set at 1.0). (d) Representative fluorescence images of WISH cells co-cultured with different bacterial species, visualized by dual staining with SYTO 9 (green, live cells) and propidium iodide (PI; red, dead cells). (× 100, Scale bar 200 μm). (d1)The control group contained the largest number of green fluorescent cells, (d2-4)a marked decrease in the number of green fluorescent protein-labeled WISH cells was observed, accompanied by a corresponding increase in the number of red fluorescent protein-labeled cells, (d2)GBS, (d3) E. faecalis,(d4) E. coli. (e) Representative flow cytometry dot plots of WISH cell apoptosis assessed by Annexin V-FITC/PI double staining. Cells were categorized into four populations: viable cells (Q3, Annexin V⁻/PI⁻), early apoptotic cells (Q4, Annexin V⁺/PI⁻), late apoptotic cells (Q2, Annexin V⁺/PI⁺), and necrotic cells (Q1, Annexin V⁻/PI⁺). Data are presented as the percentage of cells in each quadrant. (f) Quantitative analysis of total apoptosis rates (Q2 + Q4) in WISH cells from(e). Data were presented as mean ± SD, rates or proportions from three independent experiments. ^nsP > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001 compared to the control group. Abbreviations: WISH, human amnion cell line; MOI, Bacterial multiplicity of infection, CCK8, Cell Counting Kit-8.GBS, Group B Streptococcus (Streptococcus agalactiae); E. faecalis, Enterococcus faecalis; E. coli, Escherichia coli; SYTO, SYTO green fluorescent nucleic acid dye; PI, propidium iodide; Annexin V-FITC, annexin V conjugated to fluorescein isothiocyanate; PI, propidium iodide

Decreased cell viability under different bacterial exposure times

To investigate the temporal dynamics of bacteria-induced cytotoxicity, GBS, E. faecalis, and E. coli were further co-cultured with WISH cells at an MOI of 20:1 for 2, 4, 8, 16, and 24 h, and cell viability was measured by CCK-8 assay. No significant reduction in cell viability was detected at 2 h post-infection for any bacterial strain compared with controls. However, from 4 h onward, all three pathogenic bacteria induced progressive, time-dependent decreases in WISH cell viability (Fig. 3c). These results demonstrate that the cytotoxic effects of these pathogens require sustained exposure and become evident only after an initial lag phase, suggesting that bacteria-induced cell damage involves active cellular responses rather than immediate toxicity.

Decreased cell viability after bacterial treatment

To directly visualize bacteria-induced cell death, WISH cells were co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 10 h and stained with SYTO 9 (live cells, green) and PI (dead cells, red). Fluorescence microscopy revealed abundant green fluorescence with minimal red signal in control cultures, indicating high cell viability. In contrast, all bacterial co-cultures exhibited a marked reduction in green-fluorescent cells and a concomitant increase in red-fluorescent cells, confirming that exposure to these pathogens induces amniotic epithelial cell death (Fig. 3d).

Bacterial infection induces time-dependent apoptosis in WISH cells

To investigate whether pathogenic bacteria induce apoptosis in amniotic epithelial cells, WISH cells were co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 2, 4, 6, and 24 h, and apoptosis was assessed by Annexin V-FITC/PI staining followed by flow cytometry. Apoptotic cells were defined as those in the early apoptotic (Annexin V⁺/PI⁻) and late apoptotic (Annexin V⁺/PI⁺) quadrants.

At 2 h post-infection, minimal apoptosis was observed across all groups. The apoptosis rates were 4.5% in the control group, 10.0% in the GBS group, 8.0% in the E. faecalis group, and 7.0% in the E. coli group. Compared with the control group, only GBS showed a statistically significant increase in apoptosis at this early time point (P < 0.05).

At 4 h, apoptosis rates increased to 6.0% in controls, 15.0% in GBS-treated cells, 12.0% in E. faecalis-treated cells, and 18.0% in E. coli-treated cells. All three bacterial strains induced significantly higher apoptosis rates compared with the control group (P < 0.01 for GBS and E. faecalis; P < 0.001 for E. coli).

At 6 h, a more pronounced increase in apoptosis was observed, with rates of 11.0% in controls, 28.0% in GBS-treated cells, 20.0% in E. faecalis-treated cells, and 55.0% in E. coli-treated cells. All bacterial strains continued to show significantly elevated apoptosis compared with controls (P < 0.001 for all), with E. coli inducing the highest rate.

By 24 h, the majority of WISH cells had undergone apoptosis following bacterial exposure. Apoptosis rates reached 13% in controls, 85% in GBS-treated cells, 75% in E. faecalis-treated cells, and 98% in E. coli-treated cells. All bacterial strains induced significantly higher apoptosis rates compared with the control group (P < 0.001 for all) (Fig. 3e and f).

The apoptosis-inducing potency, ranked from weakest to strongest at each time point, was E. faecalis < GBS < E. coli. These findings demonstrate that pathogenic vaginal bacteria directly induce time-dependent apoptosis in WISH cells, with E. coli exhibiting the most potent pro-apoptotic effect, further supporting the mechanistic link between bacterial infection and amniotic membrane injury in the pathogenesis of PPROM.

Expression of IL-6, JAK2/STAT3, MMP-9/ TIMP-1 in bacteria-treated WISH cells

Elevated IL-6 secretion in WISH cells following bacterial stimulation

To investigate whether pathogenic bacteria trigger inflammatory responses in amniotic epithelial cells, we measured IL-6 levels in culture supernatants of WISH cells co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h by ELISA. The results demonstrated that all three bacterial strains significantly increased IL-6 secretion compared with untreated controls (control: 100.25 ± 7.41pg/mL vs. GBS: 231.5 ± 10.5 pg/mL, E. faecalis: 189.6 ± 31.89 pg/mL, E. coli: 428.34 ± 41.42 pg/mL; P < 0.001 for all) (Fig. 4a). Notably, E. coli induced the highest IL-6 production, followed by GBS and E. faecalis, suggesting that the magnitude of the inflammatory response varies among bacterial species. These findings indicate that exposure to pathogenic vaginal bacteria triggers a robust IL-6-mediated inflammatory response in amniotic epithelial cells, which may contribute to the pathogenesis of infection-associated PPROM.

Fig. 4.

Expression of IL-6, JAK2/STAT3, MMP-9/ TIMP-1 in WISH cells following bacterial stimulation. (a) IL-6 levels in supernatants of WISH cells exposed to GBS, E. faecalis, or E. coli were measured by ELISA. control: 100.25 ± 7.41pg/mL vs. GBS: 231.5 ± 10.5 pg/mL, E. faecalis: 189.6 ± 31.89 pg/mL, E. coli: 428.34 ± 41.42 pg/mL (b) Representative Western blot images of β-Actin, JAK2, p-JAK2,STAT3, and p-STAT3 in WISH cells after bacterial exposure. (c) Relative quantification of Western blot bands from (b). Results are presented as fold change relative to the TL control group (set as 1.0). Compared with controls, p-JAK2 elevated in WISH with GBS(1.8 fold increase), E. faecalis (2.5 fold increase), or E. coli (2.4 fold increase); p-STAT3 elevated in WISH with GBS(1.6 fold increase), E. faecalis (2.4 fold increase), or E. coli (2.4 fold increase). (d) Representative ICC images of MMP-9 and TIMP-1 in WISH cells following bacterial exposure. (× 400,Scale bar 40 μm), (d1-4)MMP-9,(d1)Negativity, (d2) Diffuse weak positivity, (d3) Diffuse positivity, (d4)Strong diffuse positivity, (d5-8)TIMP-1,(d5)Strong diffuse positivity, (d6) Diffuse positivity, (d7) Diffuse weak positivity, (d8)Negativity. (e) Semi-quantitative analysis of MMP-9 and TIMP-1 staining intensity from (d). For each sample, three random high-power fields were captured, and the MD was calculated as IOD/area. MMP-9 elevated (control: 0.55 ± 0.01 vs. GBS: 0.72 ± 0.01, E. faecalis: 0.71 ± 0.02, E. coli: 1.05 ± 0.04); TIMP-1 reduced (control: 1.05 ± 0.03 vs. GBS: 0.91 ± 0.01, E. faecalis: 0.87 ± 0.01, E. coli: 0.55 ± 0.01). Data were presented as mean ± SD, rates or proportions from three independent experiments. ^nsP > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001 compared to the control group. Abbreviations: IL-6, interleukin-6; WISH, human amnion cell line; GBS, Group B Streptococcus (Streptococcus agalactiae); E. faecalis, Enterococcus faecalis; E. coli, Escherichia coli; ELISA, enzyme-linked immunosorbent assay; JAK2, Janus kinase 2; p-JAK2, phosphorylated Janus kinase 2; STAT3, signal transducer and activator of transcription 3; p-STAT3, phosphorylated signal transducer and activator of transcription 3; β-Actin, beta-actin (loading control).ICC, immunocytochemistry; MMP-9,matrix metalloproteinase 9;TIMP-1,tissue inhibitor of matrix metalloproteinase

Activation of the JAK2/STAT3 signaling pathway in bacteria-treated WISH cells

To determine whether bacterial stimulation activates the JAK2/STAT3 signaling pathway in amniotic epithelial cells, we examined the expression and phosphorylation levels of JAK2 and STAT3 in WISH cells co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h by Western blot analysis. The results revealed that total JAK2 and total STAT3 protein levels were comparable between the control and all bacterial treatment groups, indicating no significant alterations in overall expression (P > 0.05). In contrast, the levels of p-JAK2 were significantly elevated in cells exposed to GBS(1.8 fold increase, P < 0.01), E. faecalis (2.5 fold increase, P < 0.001), or E. coli (2.4 fold increase, P < 0.001) compared with controls, the levels of p-STAT3 were significantly elevated in cells exposed to GBS(1.6 fold increase, P < 0.01), E. faecalis (2.4 fold increase, P < 0.001), or E. coli (2.4 fold increase, P < 0.001) compared with controls (Fig. 4b and c). This enhanced phosphorylation status, without changes in total protein expression, indicates activation of the JAK2/STAT3 signaling pathway in amniotic epithelial cells following bacterial exposure. These findings suggest that JAK2/STAT3 pathway activation may represent a key mechanistic link between ascending bacterial infection and inflammation-induced amniotic membrane injury in PPROM.(Figure 4b and c).

Dysregulation of MMP-9 and TIMP-1 expression in bacteria-treated WISH cells

To determine whether bacterial exposure alters the balance between matrix-degrading enzymes and their inhibitors in amniotic epithelial cells, we examined MMP-9 and TIMP-1 expression in WISH cells following co-culture with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h by ICC. ICC staining revealed that positive signals (yellow-brown) were primarily localized in the cytoplasm and nucleus, with staining intensity reflecting relative protein expression levels. Semi-quantitative analysis using mean OD values from three randomly selected fields per sample demonstrated that MMP-9 expression was significantly elevated in all bacterial treatment groups compared with controls. control: 0.55 ± 0.01 vs. GBS: 0.72 ± 0.01(P < 0.05), E. faecalis: 0.71 ± 0.02(P < 0.05), E. coli: 1.05 ± 0.04(P < 0.01). Conversely, TIMP-1 expression was markedly reduced in bacteria-treated cells, control: 1.05 ± 0.03 vs. GBS: 0.91 ± 0.01(P < 0.05), E. faecalis: 0.87 ± 0.01(P < 0.05), E. coli: 0.55 ± 0.01(P < 0.01). This reciprocal dysregulation of MMP-9 and TIMP-1 resulted in a substantially increased MMP-9/TIMP-1 ratio in all bacterial treatment groups, indicating a net shift toward a proteolytic environment that favors ECM degradation. Among the tested strains, E. coli exerted the most pronounced effects, followed by GBS and E. faecalis. These findings suggest that bacteria-induced activation of the IL-6/JAK/STAT axis may promote MMP-9 upregulation and TIMP-1 suppression, thereby contributing to the extracellular matrix degradation that underlies fetal membrane rupture in PPROM.(Figure 4d and e).

Effect of TCZ inhibition on bacteria-induced effects on fetal membrane cells

TCZ attenuates bacteria-induced cytotoxicity in WISH cells

To investigate whether blockade of IL-6 signaling protects amniotic epithelial cells from bacteria-induced injury, WISH cells were co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h in the presence of TCZ or an IgG isotype control, and cell viability was assessed using the CCK-8 assay. Compared with the IgG control group, TCZ treatment significantly attenuated the reduction in cell viability induced by all three bacterial strains (control: 1 vs.GBS + IgG: 0.70 ± 0.06 vs. GBS + TCZ: 0.81 ± 0.04 vs. E. faecalis + IgG: 0.77 ± 0.12 vs. E. faecalis + TCZ: 0.84 ± 0.05 vs. E. coli + IgG: 0.56 ± 0.12 vs. E. coli + TCZ: 0.77 ± 0.02; P < 0.01 for all TCZ vs. IgG comparisons, P < 0.01; for control vs. others comparisons, P < 0.001). Compared with the IgG co-cultured group, TCZ treatment significantly attenuated the reduction in cell viability induced by all three bacterial strains (GBS + IgG vs. GBS + TCZ, P < 0.01; E. faecalis + IgG vs. E. faecalis + TCZ. P < 0.05; E. coli + IgG vs. E. coli + TCZ, P < 0.001 ) (Fig. 5a). These findings indicate that IL-6 signaling blockade partially rescues amniotic epithelial cells from bacteria-induced cytotoxicity, suggesting a protective role for TCZ in the context of infection-associated amniotic membrane injury.

Fig. 5.

Effects of TCZ intervention on bacteria-induced cell viability, IL-6 expression, and JAK2/STAT3, MMP-9/TIMP-1 protein levels in WISH cells. a Cell viability assessed by CCK-8 assay following exposure to GBS, E. faecalis, or E. coli in the presence of TCZ or IgG control. Relative viability was normalized to control (set at 1.0).Control: 1 vs.GBS + IgG: 0.70 ± 0.06 vs. GBS + TCZ: 0.81 ± 0.04; E. faecalis + IgG: 0.77 ± 0.12 vs. E. faecalis + TCZ: 0.84 ± 0.05; E. coli + IgG: 0.56 ± 0.12 vs. E. coli + TCZ: 0.77 ± 0.02 (b) IL-6 levels in cell culture supernatants measured by ELISA after bacterial exposure with or without TCZ. (Control: 126.4 ± 5.5 vs. GBS + TCZ: 222.3 ± 17.4 pg/mL vs. GBS + IgG: 286.7 ± 12.9 pg/mL vs. E. faecalis + TCZ: 192.7 ± 12.2 pg/mL vs. E. faecalis + IgG: 236.3 ± 34.8 pg/mLvs. E. coli + TCZ: 581.7 ± 46.0 pg/mL vs. E. coli + IgG: 656.7 ± 38.3 pg/mL (c) Representative Western blot images of β-Actin, JAK2, p-JAK2,STAT3, and p-STAT3 in WISH cells after bacterial exposure with or without TCZ or IgG control. d Relative quantification of Western blot bands from (d). Densitometric quantification of p-JAK2/JAK2 and p-STAT3/STAT3 ratios. Results are presented as fold change relative to the TL control group (set as 1.0). p-JAK2/JAK2 ratio relative to IgG control: GBS + IgG(1.4 fold increase), GBS + TCZ(1.3 fold increase), E. faecalis + IgG(2.3 fold increase), E. faecalis + TCZ(1.5 fold increase), E. coli + IgG(2.4 fold increase), E. coli + TCZ(1.8 fold increase); p-STAT3/STAT3 ratio relative to IgG control: GBS + IgG(1.4 fold increase), GBS + TCZ(1.3 fold increase), E. faecalis + IgG(2.1 fold increase), E. faecalis + TCZ(1.7 fold increase), E. coli + IgG(2.5 fold increase) ,E. coli + TCZ(1.9 fold increase). e Representative ICC images of MMP-9 and TIMP-1 in WISH cells after bacterial exposure with or without TCZ (×400, scale bar = 40 μm). f Semi-quantitative analysis of MMP-9 and TIMP-1 staining intensity from (e). for each sample, three random high-power fields were captured, and the MD was calculated as IOD/area. For MMP-9, IgG controls 0.17 ± 0.02 vs. GBS + TCZ: 0.19 ± 0.01 vs. GBS + IgG: 0.25 ± 0.03 vs. E. faecalis + TCZ: 0.18 ± 0.01 vs. E. faecalis + IgG: 0.23 ± 0.02 vs. E. coli + TCZ: 0.22 ± 0.03 vs. E. coli + IgG: 0.27 ± 0.03. For TIMP-1, IgG controls0.35 ± 0.03 vs.GBS + TCZ: 0.32 ± 0.04 vs. GBS + IgG: 0.23 ± 0.02 vs. E. faecalis + TCZ: 0.31 ± 0.03 vs. E. faecalis + IgG: 0.24 ± 0.02 vs. E. coli + TCZ: 0.28 ± 0.03 vs. E. coli + IgG: 0.18 ± 0.02. Data were presented as mean ± SD, rates or proportions from three independent experiments. Statistical significance was determined by two-way ANOVA followed by Dunnett’s test for comparisons with the control group and TCZ vs. IgG groups, by Student’s t-test for comparisons between TCZ vs. IgG groups. ^nsP > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001. Abbreviations: WISH, human amnion cell line; GBS, Group B Streptococcus (Streptococcus agalactiae); E. faecalis, Enterococcus faecalis; E. coli, Escherichia coli; TCZ, tocilizumab (IL-6 receptor antibody); IgG, immunoglobulin G (isotype control); CCK8, Cell Counting Kit-8; JAK2, Janus kinase 2; p-JAK2, phosphorylated Janus kinase 2; STAT3, signal transducer and activator of transcription 3; p-STAT3, phosphorylated signal transducer and activator of transcription 3; β-Actin, beta-actin (loading control)

TCZ differentially modulates bacteria-induced IL-6 secretion in WISH cells

To investigate whether IL-6 receptor blockade with TCZ affects bacteria-induced IL-6 production, WISH cells were co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h in the presence of TCZ or IgG control, and IL-6 concentrations in culture supernatants were measured by ELISA. Basal IL-6 levels were 126.4 ± 5.5 pg/mL in the presence of IgG control. Following bacterial stimulation, all three strains induced robust IL-6 secretion in the IgG control groups (GBS: 286.7 ± 12.9 pg/mL; E. faecalis: 236.3 ± 34.8 pg/mL; E. coli: 656.7 ± 38.3 pg/mL). Two-way ANOVA performed on bacteria-stimulated groups revealed significant main effects of bacterial strain and TCZ intervention on IL-6 levels (P < 0.001 for all). Subsequently, TCZ treatment significantly reduced IL-6 secretion compared with the corresponding IgG control for all three bacterial strains. TCZ treatment resulted in markedly lower IL-6 levels compared with IgG controls across all bacterial strains (GBS + TCZ: 222.3 ± 17.4 pg/mL vs. GBS + IgG: 286.7 ± 12.9 pg/mL, P < 0.05; E. faecalis + TCZ: 192.7 ± 12.2 pg/mL vs. E. faecalis + IgG: 236.3 ± 34.8 pg/mL, P < 0.05; E. coli + TCZ: 581.7 ± 46.0 pg/mL vs. E. coli + IgG: 656.7 ± 38.3 pg/mL, P < 0.05) (Fig. 5b). These findings indicate that TCZ treatment attenuates bacteria-induced IL-6 secretion in amniotic epithelial cells, suggesting that IL-6 receptor blockade may disrupt autocrine or paracrine feedback loops that amplify IL-6 production during bacterial infection.

(Fig. 5b and c).

TCZ inhibits bacteria-induced JAK2/STAT3 phosphorylation in WISH cells

To investigate whether TCZ treatment modulates the JAK2/STAT3 signaling pathway downstream of IL-6, WISH cells were co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h in the presence of TCZ or IgG control, and the phosphorylation status of JAK2 and STAT3 was examined by Western blot analysis. The ratios of phosphorylated to total protein (p-JAK2/JAK2 and p-STAT3/STAT3) were calculated to reflect pathway activation levels. Two-way ANOVA revealed significant main effects of bacterial infection and TCZ intervention on both p-JAK2/JAK2 and p-STAT3/STAT3 ratios (P < 0.001 for all). TCZ treatment markedly attenuated the bacteria-induced increase in p-JAK2/JAK2 and p-STAT3/STAT3 ratios compared with the IgG control group. Subsequently, compared with IgG controls, TCZ significantly reduced the p-JAK2/JAK2 ratio for each bacterial strain: GBS (1.41 ± 0.18 vs. 1.26 ± 0.09, P < 0.05), E. faecalis (2.28 ± 0.11 vs. 1.47 ± 0.09, P < 0.001), and E. coli (2.38 ± 0.16 vs. 1.84 ± 0.06, P < 0.001). Similarly, compared with IgG controls, the p-STAT3/STAT3 ratio was significantly lower in TCZ-treated cells : GBS (1.39 ± 0.07 vs. 1.33 ± 0.10, P < 0.05), E. faecalis (2.13 ± 0.21 vs. 1.65 ± 0.31, P < 0.001), and E. coli (2.47 ± 0.18 vs. 1.84 ± 0.15, P < 0.001) (Fig. 5d). These findings confirm that TCZ effectively blocks IL-6-mediated JAK2/STAT3 pathway activation in amniotic epithelial cells exposed to pathogenic bacteria, as evidenced by the reduced phosphorylation ratios, may provide mechanistic evidence for its anti-inflammatory effects in this cellular model of infection-associated PPROM.

TCZ Reverses Bacteria-Induced MMP-9/TIMP-1 Dysregulation in WISH Cells

To determine whether IL-6 receptor blockade with TCZ restores the balance between matrix-degrading enzymes and their inhibitors in bacteria-exposed amniotic epithelial cells, WISH cells were co-cultured with GBS, E. faecalis, or E. coli at an MOI of 20:1 for 6 h in the presence of TCZ or IgG control, and MMP-9 and TIMP-1 expression was assessed by immunocytochemistry. Semi-quantitative analysis using mean OD values was performed on three randomly selected fields per sample.

In IgG control groups, all three bacterial strains significantly upregulated MMP-9 expression compared with unstimulated controls (control: 0.17 ± 0.02 vs. GBS: 0.25 ± 0.03, E. faecalis: 0.23 ± 0.02, E. coli: 0.27 ± 0.03; P < 0.001 for all). Conversely, TIMP-1 expression was significantly reduced in bacteria-stimulated cells compared with controls (control: 0.35 ± 0.03 vs. GBS: 0.23 ± 0.02, E. faecalis: 0.24 ± 0.02, E. coli: 0.18 ± 0.02; P < 0.001 for all) (Fig. 4d and e).

Two-way ANOVA performed on bacteria-stimulated groups revealed significant main effects of bacterial strain and TCZ intervention on both MMP-9 and TIMP-1 expression levels (P < 0.001 for all). Subsequently, TCZ treatment significantly reversed the bacteria-induced dysregulation of MMP-9 and TIMP-1 compared with the corresponding IgG controls for all three bacterial strains.

For MMP-9, TCZ treatment resulted in significantly reduced expression levels compared with IgG controls (GBS + TCZ: 0.19 ± 0.01 vs. GBS + IgG: 0.25 ± 0.03, P < 0.01; E. faecalis + TCZ: 0.18 ± 0.01 vs. E. faecalis + IgG: 0.23 ± 0.02, P < 0.01; E. coli + TCZ: 0.22 ± 0.03 vs. E. coli + IgG: 0.27 ± 0.03, P < 0.05). For TIMP-1, TCZ treatment significantly enhanced expression levels compared with IgG controls (GBS + TCZ: 0.32 ± 0.04 vs. GBS + IgG: 0.23 ± 0.02, P < 0.01; E. faecalis + TCZ: 0.31 ± 0.03 vs. E. faecalis + IgG: 0.24 ± 0.02, P < 0.01; E. coli + TCZ: 0.28 ± 0.03 vs. E. coli + IgG: 0.18 ± 0.02, P < 0.001) (Fig. 5e and f).

This TCZ-mediated reversal of MMP-9/TIMP-1 dysregulation resulted in normalized MMP-9/TIMP-1 ratios across all bacterial treatment groups, indicating restoration of the proteolytic balance toward a non-degradative state. These findings demonstrate that IL-6 signaling blockade with TCZ protects amniotic epithelial cells from bacteria-induced ECM degradation by suppressing MMP-9 upregulation and restoring TIMP-1 expression, further supporting the central role of the IL-6/JAK/STAT axis in infection-associated PPROM pathophysiology.

Discussion

Our study found that vaginal opportunistic pathogens could undermine fetal membrane integrity through the IL-6-JAK-STAT signaling pathway in cases of PPROM. Furthermore, it demonstrated that the IL-6 receptor antagonist, TCZ, effectively interrupted this pathological sequence.

Several recent studies have identified vaginal microbiota dysbiosis is the core pathogenic driver of PPROM [5, 20]. Lactobacillus pecies dominate the healthy vaginal microbiota and maintain an acidic environment through lactic acid production, thereby inhibiting pathogenic bacterial overgrowth [21]. However, when Lactobacillus abundance decreases, opportunistic pathogens such as GBS, E. faecalis, and E. coli proliferate, triggering inflammatory responses and membrane damage [22, 23]. Our previous investigations demonstrated that such dysbiosis, that characterized by reduced lactobacilli and increased opportunistic pathogens, was significantly correlated with PPROM development. This present study extends these observations by providing direct cellular evidence that these bacteria induce WISH cell apoptosis, upregulate IL-6 secretion, and activate JAK/STAT signaling.

IL-6 is produced by diverse cell types, including macrophages, fibroblasts, endothelial cells, and lymphocytes. In chorioamnionitis, IHC studies have demonstrated that IL-6 is expressed in decidual cells, macrophages, and amniotic epithelial cells [24]. While we acknowledge that chorioamnionitis typically begins in the decidua and chorion before extending to the amnion in severe cases, our focus on amniotic epithelial cells because PPROM represents the end-stage of this pathological process. Future studies employing co-culture systems incorporating decidual stromal cells, chorionic trophoblasts, and immune cells will be essential to model the full spatiotemporal progression of ascending infection. A potential limitation of our in vitro model is the assumption that bacteria directly interact with amniotic epithelial cells. While our study did not perform Gram staining on fetal membranes, several studies have demonstrated that bacteria can indeed reach and interact with the amniotic epithelium in PPROM [25, 26],however, future studies should include such analyses to directly correlate bacterial presence with the molecular changes observed in our in vitro model.

This histological evidence supports the biological relevance of direct bacterial–amniotic epithelial cell co-culture. However, we acknowledge that in mild to moderate chorioamnionitis, inflammatory infiltration is confined to the decidua and chorion, and bacterial contact with the amnion may occur only in advanced stages. Future studies using decidual or chorionic cell co-culture models may better recapitulate earlier disease stages. In PPROM management, monitoring maternal and fetal inflammatory status is critical. Traditional markers such as white blood cell count and C-reactive protein have limited predictive value, whereas cervical IL-6 shows greater potential as a local inflammatory biomarker [27].

Consistent with our findings, IL-6 elevation has been documented in various inflammatory conditions, including neonatal sepsis, hepatocellular carcinoma, and colorectal cancerr [28–30]. In PPROM, cervical IL-6 levels correlate strongly with neonatal inflammatory parameters, particularly C-reactive protein and procalcitonin on the first postpartum day [27]. Furthermore, umbilical cord blood IL-6 levels are inversely correlated with fetal inflammation severity, especially in preterm infants with funisitis [31]. In cases of PPROM, IL-6 levels in cervical fluid can serve as a non-invasive biomarker for predicting fetal inflammation, with high sensitivity and specificity [32].Our results reinforce the diagnostic and predictive value of IL-6 in PPROM, supporting its potential utility in guiding clinical decisions. However, further research is needed to validate the application of IL-6 in different populations and to determine its best use in clinical practice [27, 32].

Bacterial infection-induced inflammatory responses play a significant role in the pathological mechanisms of PPROM. Our results demonstrate that bacterial infection elevates IL-6 levels, which signals through the JAK/STAT3 pathway to promote membrane rupture. This mechanism aligns with studies in other inflammatory conditions. For instance, IL-6 induces lung injury in severe pneumonia via the IL-6/sIL-6R/JAK1/STAT3 pathway [33]. Similarly, IL-6 regulates lipid metabolism disorders through JAK/STAT signaling in hypoxic laryngeal carcinoma cells [34]. These studies collectively suggest that IL-6 plays a critical role in multiple pathological states through the JAK-STAT pathway.

In the specific context of PPROM, TNF-α and IL-6 upregulation activates ADAMTS-9 in fetal membrane cells [35]. and bacterial infection-induced amniotic membrane thinning correlates strongly with bacterial presence [36]. Further demonstrated that IL-1β and TGF-β, rather than IL-6 alone, may play pivotal roles in disrupting tight junctions in placental endothelial cells, suggesting that multiple cytokines cooperate to compromise membrane integrity [37]. These findings indicate that the increase in IL-6 induced by bacterial infection plays a significant role in PPROM via the JAK-STAT signaling pathway. However, further research is needed to validate the role and significance of pro-inflammatory cytokines other than IL-6 and other relevant signaling pathways.

Multiple studies have supported the mechanism by which bacterial infection-induced IL-6 elevation exerts its effects in PPROM through the JAK-STAT signaling pathway. This mechanism is not only significant in PPROM but also provides a new perspective for the study of other inflammation-related diseases. Future research could further explore the specific mechanisms of IL-6 and the JAK-STAT pathway in PPROM, with the aim of identifying novel therapeutic targets and strategies for clinical treatment.

The observed MMP-9 upregulation and TIMP-1 downregulation in PPROM tissues and bacteria-treated WISH cells is consistent with previous reports. In Ureaplasma-associated PPROM, MMP-9 elevation drives collagen degradation, releasing the neutrophil chemoattractant PGP and amplifying inflammation [38]. This mechanism is closely associated with the elevated MMP-9/TIMP-1 ratio, suggesting a pivotal role of MMP-9 in PPROM.

TCZ, an anti-IL-6 receptor monoclonal antibody, has demonstrated efficacy in various inflammatory conditions, including COVID-19-associated cytokine storms, where it reduces mortality and mechanical ventilation requirements [39, 40]. This drug exerts its therapeutic effects by inhibiting IL-6 activity, thereby mitigating inflammatory responses and playing a role in multiple inflammation-related diseases [41].

Among pregnancy complications, PB and PPROM are significant causes of neonatal mortality and morbidity. The etiology of PPROM is complex, involving multiple factors such as bacterial infection and inflammatory responses [42].In systemic juvenile idiopathic arthritis, early TCZ administration is safe and effective [39]. Given that PPROM shares inflammatory mechanisms with these conditions, TCZ may hold potential for preventing or attenuating infection-associated membrane rupture [43].

Although TCZ has demonstrated favorable efficacy in various diseases, its specific application in PPROM still requires further investigation. Future research should focus on the specific mechanisms of TCZ in PPROM, optimal timing of administration, and combination with other therapeutic approaches, aiming to provide more effective strategies for the prevention and treatment of PPROM [42, 43]. By integrating multiple studies, the application prospects of TCZ in PPROM are promising.

A critical consideration for any therapeutic agent in pregnancy is its placental transfer and fetal safety profile. TCZ, a humanized monoclonal antibody of the IgG1 subclass with a molecular weight of approximately 148 kDa, is theoretically capable of crossing the placental barrier, although its large size suggests that passive diffusion is minimal.While pregnancy safety data for TCZ remain limited, evidence is gradually accumulating. It wasreported that 25 pregnant women receiving TCZ or sarilumab for severe COVID-19. all pregnancies resulted in live births, with 24 maternal survivors [40]. A narrative review on TCZ in pregnancy identified over 600 exposed cases, including 20 with COVID-19, and reported higher rates of spontaneous abortion and preterm birth compared with the general population [41].Current evidence does not support an absolute contraindication for anti-IL-6 therapy in late pregnancy, particularly in the context of severe maternal inflammatory conditions where benefits may outweigh risks. But the use of TCZ for PPROM prevention should be restricted to well-designed clinical trials with careful ethical oversight and long-term infant follow-up.

Conclusion

Our study offers preliminary evidence suggesting the potential involvement of vaginal opportunistic pathogens in the pathogenesis of PPROM. The findings indicate that bacterial infection may lead to elevated IL-6 levels, which could subsequently activate the JAK-STAT pathway in fetal membranes. This activation seems to be associated with an imbalance in MMP-9/TIMP-1, potentially contributing to ECM degradation, reduced cell viability, and increased apoptosis, those factors that may compromise membrane integrity. In our experimental model, TCZ, an IL-6 receptor antagonist, was observed to mitigate these effects, suggesting a potential modulatory role for IL-6 blockade in this pathway. However, these results should be interpreted with caution, as they are derived from in vitro models and require further validation. We reframe IL-6 as a causal executor, rather than merely a prognostic biomarker, and validate the JAK-STAT axis as a therapeutic target in PPROM, challenging the prevailing antibiotic-centric paradigm. Further research is necessary to evaluate the efficacy, safety, and translational relevance of targeting the IL-6/JAK-STAT axis in the clinical management of PPROM.

Abbreviations

BMI

Body mass Index

BV

Bacterial vaginosis

CCA

Clinic chorioamnionitis

ECM

Extracellular matrix

HCA

Histological chorioamnionitis

ICC

Immunocytochemistry

IHC

Immunohistochemistry

IL

Interleukin

IL-6R

Interleukin 6 receptor

IOD

Integrated optical density

JAK

Janus kinase

MMP

Matrix metalloproteinase

MOI

Multiplicity of infection

mRNA

Messager ribonucleic acid

PBS

Phosphate buffer saline

p-JAK2

Phosphorylated JAK2

PPROM

Preterm prelabor rupture of membranes

p-STAT3

Phosphorylated STAT3

STAT

Signal transducer and activator of transcription

DAB

3, 3’-diaminobenzidine

TCZ

Tocilizumab

TIMP

Tissue inhibitor of matrix metalloproteinase

TL

Normal full term labor

Authors’ contributions

YLL and TYT participated in designing the study, collecting the information, performing the statistical analyses, and writing the initial draft of the manuscript. TYTand JYM contributed to the reviewing and revising of the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by the National Key Research and Development Program of China (2024YFC2707400).

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information, while the datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This investigation received ethical clearance from the ethics committee of West China Second University Hospital of Sichuan University (Medical Research 2020 NO. 050, 2023 NO. 131, 2025 NO.1) and followed the ethical standards of the Declaration of Helsinki. Informed consent to participate was obtained from all of the participants.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.