Transforming Growth Factor-beta 1 Alleviates Uterine Bleeding after Medication Abortion in Early Pregnancy by Upregulating the p53/Plasminogen Activator Inhibitor-1 Pathway

Abstract

TGF-β1 plays a significant role in pregnancy outcomes. This research sought to investigate whether TGF-β1 is involved in the bleeding mechanism after medication abortion (MA) in early pregnancy. The MA rat model was established in vivo using mifepristone and misoprostol, and trophoblasts HTR8/SVneo were treated with lipopolysaccharide in vitro. Changes in uterine morphology, weight, and bleeding were assessed. tissue-type plasminogen activator (tPA), urokinase-type plasminogen activator (uPA), estradiol, and progesterone levels were detected by ELISA. HE staining was employed to analyze uterine pathological changes. Apoptosis was assessed by TUNEL staining. Inflammatory cytokine expression was assessed by ELISA and qRT-PCR. Related protein levels were analyzed by Western blot. MA induction led to abnormal uterine morphology, reduced uterine weight, and heavier bleeding. MA rats showed higher tPA, uPA, IL-6, and TNF-α levels, and lower estradiol and progesterone levels compared to controls. Moreover, trophoblast tissue damage with excessive apoptosis was observed in MA rats. TGF-β1, p53, and PAI-1 levels were markedly decreased after MA induction. In HTR8/SVneo cells, lipopolysaccharide treatment significantly inhibited cellular functions, reduced TGF-β1, p53, and PAI-1 levels, and increased IL-6 and TNF-α levels. Notably, these changes were partially reversed by overexpression of TGFB1. In conclusion, TGF-β1 protects trophoblasts and alleviates MA-induced uterine bleeding by upregulating the p53/PAI-1 pathway.

Keywords

early pregnancy medication abortion uterine bleeding trophoblasts TGF-β1/p53/PAI-1 pathway

Introduction

Abortion is a medical procedure used to terminate a pregnancy (Kumwong et al., 2025). Statistics show that the number of abortions in the United States increased by 11% between the second quarters of 2023 and 2024 (Mark et al., 2025). With the development of abortion techniques, both surgical and medication abortion (MA) methods can be provided throughout the entire gestational period (Kapp and Lohr, 2020). The effectiveness of MA based on mifepristone and misoprostol in early pregnancy exceeds 95% (Raymond et al., 2013). Despite the low mortality rate of safe abortion, there are still some complications, such as bleeding and incomplete abortion (Bridwell et al., 2022). MA patients generally experience heavier bleeding lasting for 9 to 12 days (Bridwell et al., 2022). Therefore, it is crucial to explore the mechanisms of bleeding after MA.

Pregnancy is a hypercoagulable state involving the coagulation and fibrinolysis systems, whose interaction may cause gestational complications such as miscarriage (Greer, 2003). Studies have shown that intravenous injection of oxytocin for induced labor can slightly activate the coagulation and fibrinolysis systems, leading to an increase in platelet aggregation (Briel et al., 1979). The fibrinolytic system is the basis for removing fibrin deposits and thrombi, and its excessive activation leads to heavier bleeding (Keragala and Medcalf, 2021). This fibrinolytic pathway is primarily mediated by plasmin, which is formed through its precursor plasminogen (Medcalf and Keragala, 2021). The main serine proteases synthesized include tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA), which convert plasminogen into plasmin and promote fibrinolysis, but their activities are inhibited by plasminogen activator inhibitor 1 (PAI-1) and PAI-2 (Flemmig and Melzig, 2012). Targeting the fibrinolytic system to suppress fibrinolytic enzymes can reduce bleeding. However, the mechanism influencing the fibrinolytic system after MA remains unclear.

The multifunctional cytokine transforming growth factor-beta 1 (TGF-β1) is involved in the regulation of fetal and maternal immune tolerance in healthy and pathological pregnancies (Yang et al., 2021). Lower TGF-β1 levels were found in decidual samples of patients with recurrent spontaneous abortion compared to healthy controls (Zhu et al., 2022). Platelet degranulation at the vascular injury site can prevent bleeding, and TGF-β1 is the main component of platelet α-granules, which accumulate in thrombi (Assoian et al., 1983; Grainger et al., 1995). Additionally, TGF-β1 was found to inhibit plasmin and matrix metalloproteinase activities in flexor tendon cells through PAI-1 (Farhat et al., 2015). Moreover, curcumin targets TGF-β1 to regulate the p53-fibrinolysis system in alveolar epithelial-mesenchymal transition (Shaikh et al., 2021). Chen et al. found that ginsenosides inhibited cerebral hemorrhage in stroke patients treated with recombinant tPA by upregulating TGF-β1 expression (Chen et al., 2016). However, the role of the TGF-β1/p53/PAI-1 pathway in bleeding after MA remains unexplored.

In this study, the role of the TGF-β1/p53/PAI-1 pathway was assessed using MA rats and lipopolysaccharide (LPS)-stimulated HTR8/SVneo cells, focusing on its impact on uterine bleeding. The underlying mechanism was further confirmed through gain-of-function experiments. These findings may provide a theoretical basis for treating bleeding after MA.

Methods

Animal model of MA

Wistar female rats (8 weeks, 180–200 g) were obtained from the Experimental Animal Center of Yangzhou University. Rats were randomly assigned to four groups (n = 6): pregnant control, MA, MA + overexpression negative control (oe-NC), and MA + oe-Tgfb1. Female rats were mated with males (2:1) in the evening. Pregnancy was verified by detecting sperm in a vaginal smear the next morning. On day 7 of pregnancy, rats were gavaged with 8.3 mg/kg mifepristone (MCE, HY-13683) and then given 100 μg/kg misoprostol (MCE, HY-118189) 10 h later to induce abortion (Zhang et al., 2023). For the MA + oe-NC and MA + oe-Tgfb1 groups, oe-NC or oe-Tgfb1 (1 × 10⁹ TU/mL, 100 μL) was intraperitoneally injected into rats. The sequence of Tgfb1 was obtained from the NCBI website and packaged into lentiviruses using a third-generation packaging system with pMDLg/pRRE (packaging plasmid containing Gag, Pol, and Rev response element), pVSV-G (plasmid expressing the vesicular stomatitis virus G glycoprotein), and pRSV-Rev (plasmid expressing Rev under the RSV promoter) mixed at a 5:3:2 ratio. Cotton balls, half-wrapped with plastic film and of the same weight, were placed in the vagina of rats to prevent blood leakage and to collect shed villous and decidual tissues. Orbital blood collection was performed on days 1, 8, and 14 of pregnancy, respectively, after a 12-h fast. The serum was collected for subsequent experiments. After 14 days, rats were euthanized by inhaling excessive CO₂. Their abdominal cavities were opened, and the uteri were removed, weighed, and photographed. Following this, each embryo, along with its placenta and membranes, was carefully separated from the uterus using surgical scissors and forceps. By cutting the umbilical cord between the fetus and placenta, a complete placental unit was obtained for subsequent experiments. This study was approved by the Experimental Animal Welfare Ethics Committee of Yangzhou University (No.202507027).

Assessment of uterine bleeding

On pregnancy day 8, the cotton ball was replaced every 6 h starting at 6:00 until there was no bleeding. Duration and volume of uterine bleeding were recorded, and cotton balls were collected.

On pregnancy day 14, 0.02 mL of blood samples were obtained from the tail vein of each rat and mixed with 4 mL of 5% NaOH solution. The volume of this solution was recorded as V₁. The collected cotton balls were put into a beaker and washed with NaOH solution, with the total volume of the used solution being V₂. The solution was filtered, and absorbance was detected by a UV–visible spectrophotometer. The volume of uterine bleeding was calculated using this formula: V (mL) = 0.02 × V₂ × A₂/(A₁ × V₁), where A₁ is the absorbance of the tail vein blood solution, and A₂ is that of the NaOH wash solution.

Isolation of placental villi tissue

The placenta was placed in a Petri dish containing pre-cooled phosphate-buffered saline (PBS). The maternal side (basal decidua) of the placenta was held with an ophthalmic forceps, and placental villi tissue was gently peeled off from the edge using another ophthalmic forceps. The tissue was washed with PBS to remove any residual blood.

Enzyme-linked immunosorbent assay

The placental villus tissue (20 mg) was mixed with normal saline and homogenized using a tissue grinder, followed by centrifugation to obtain the supernatant. Estradiol, progesterone, tPA, and uPA levels in rat serum, TGF-β1 levels in placental villous tissue, and IL-6 and TNF-α levels in both placental villous tissue and HTR-8/SVneo cells were detected by ELISA. ELISA kits (Mlbio, Shanghai, China) were used following the manufacturer’s instructions. Absorbance was measured using a microplate reader (Wuxi Hiwell Diatek, DR-3518G).

Hematoxylin-eosin staining

The rat uterus was fixed in 4% paraformaldehyde (Beyotime, P0099). After gradient ethanol dehydration, tissues were embedded in paraffin and made into sections with a thickness of 4–7 μm. Sections were dewaxed and stained with the HE staining kit (Beyotime, C0105S), followed by observation under a microscope (Olympus, CKX53).

Terminal deoxynucleotidyl transferase dUTP Nick-End Labeling staining

After conventional dewaxing and rehydration, the placental villus tissue sections were treated with proteinase K working solution and incubated with terminal deoxynucleotidyl transferase reaction solution (Beyotime, C1086) following the instructions. Finally, sections were stained with 4′,6-diamidino-2-phenylindole (Beyotime, C1005) and imaged under a fluorescence microscope. The fluorescence signals were quantified using ImageJ software to calculate the apoptosis index (percentage of TUNEL-positive cells per field). Data analysis was conducted under blind conditions, and the analysts were unaware of the treatment groups.

quantitative reverse transcription polymerase chain reaction

Total RNA was extracted from samples using Trizol (Invitrogen, 15596018). cDNA synthesis was performed using cDNA synthesis premix (Tiangen, KR118-02). Subsequently, PCR was conducted using SYBR Green Master Mix (Lifeint, A4004M). Quantification was achieved through the 2^−ΔΔCt method, and GAPDH levels were employed for data standardization. Primers used are shown in Table 1.

Table 1.

Primer Sequences Used in This Study

Gene name	Primer type	Primer sequence
Tgfb1 (rat)	Forward	5′-AGGAGACGGAATACAGGGCT-3′
Tgfb1 (rat)	Reverse	5′-CCACGTAGTAGACGATGGGC-3′
IL-6 (rat)	Forward	5′-CACTTCACAAGTCGGAGGCT-3′
IL-6 (rat)	Reverse	5′-TCTGACAGTGCATCATCGCT-3′
Tnf (rat)	Forward	5′-GATCGGTCCCAACAAGGAGG-3′
Tnf (rat)	Reverse	5′-TCCCTCAGGGGTGTCCTTAG-3′
Gapdh (rat)	Forward	5′-GACATGCCGCCTGGAGAAAC-3′
Gapdh (rat)	Reverse	5′-AGCCCAGGATGCCCTTTAGT-3′
TGFB1 (human)	Forward	5′-TACCTGAACCCGTGTTGCTC-3′
	Reverse	5′-GAAGTTGGCATGGTAGCCCT-3′
GAPDH (human)	Forward	5′-TGTGGGCATCAATGGATTTGG-3′
	Reverse	5′-ACACCATGTATTCCGGGTCAAT-3′

Western blot

Total proteins were extracted from samples using radioimmunoprecipitation assay buffer (Beyotime, P0013B), and protein quantification was performed using the BCA method. Protein samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes (Beyotime, FFP24). The blocked membranes were incubated overnight at 4°C with primary antibodies: TGF-β1 (1:2000; Thermo, PA1-29020), p53 (1:500; Thermo, PA5-27822), PAI-1 (1:1000; Thermo, PA5-35128), Bcl-2 (1:500; Thermo, PA5-27094), Bax (1:2000; Thermo, PA5-11378), pro-caspase-3 (1:2000; Thermo, MA5-54082), cleaved-caspase-3 (1:500; Thermo, PA5-114687), and GAPDH (1:5000; Abcam, ab181602). Membranes were treated with a secondary antibody (1:2000; Abcam, ab6721) for 1 h and visualized using enhanced chemiluminescence (Applygen, P1000).

Cell culture and treatment

Human embryonic trophoblasts HTR8/SVneo (SAIOS, CL-505h) were cultured in Dulbecco’s Modified Eagle Medium-High Glucose (DMEM-H) medium (Gibco, C11995500BT) containing 10% fetal bovine serum (Solarbio, S9030) at 37°C with 5% CO₂. To mimic abnormal inflammation at the uteroplacental interface of MA, cells were treated with different concentrations of LPS (12.5, 25, 50, and 100 ng/mL; Solarbio, L8880) for 24 h.

To evaluate whether TGF-β1 affects bleeding duration after MA, cells were transfected in 6-well plates with oe-TGFB1 or its negative control (oe-NC) for 48 h. Subsequently, cells were treated with LPS (100 ng/mL) for 24 h.

Cell counting kit-8 assay

HTR8/SVneo cells (2000 cells/well) were seeded into 96-well plates. Following grouping treatment, cells were incubated with 10 μL of CCK-8 solution (Beyotime, C0037) for 2 h. Absorbance at 450 nm was measured by a microplate reader.

Transwell assay

HTR8/SVneo cells (2 × 10⁴ cells/well) were seeded into the upper chamber of Transwell, and DMEM-H complete medium was added to the lower chamber. After 24 h, cells were rinsed 1–2 times with PBS (Beyotime, C0221A), followed by fixation with anhydrous methanol (Sinopharm, 10009218) for 30 min. Subsequently, cells were stained with crystal violet (Beyotime, C0121) and imaged under a microscope.

Flow cytometry

Apoptosis was evaluated using the apoptosis detection kit (Beyotime, C1086). HTR8/SVneo cells were resuspended in 195 μL of binding buffer and incubated with 5 μL of Annexin V–fluorescein isothiocyanate and 10 μL of propidium Iodide for 10 min. Cells were detected by a flow cytometer (Beckman, CytoFLEX S) and analyzed by CELL Quest software. Apoptosis rate (%) = Early apoptosis rate + Late apoptosis rate.

Statistical analysis

All statistical analyses were performed using GraphPad Prism 7.0 software. Data were presented as mean ± standard deviation. The unpaired Student’s t-test was used to analyze differences between groups, and one-way ANOVA with Tukey’s post hoc test was employed for multiple-group comparisons. P < 0.05 was considered statistically significant.

Results

MA triggered uterine bleeding and decreased TGF-β1, p53, and PAI-1 expression in rats

Abortion was triggered in pregnant rats using mifepristone and misoprostol. The uteri of the pregnancy control rats were pink, with well-developed embryonic tissue and normal embryo implantation sites (Fig. 1A). However, the uteri of the MA rats were dark red with bamboo-like changes, and embryo volume was significantly reduced and uneven in size (Fig. 1A). Petechial hemorrhages were observed in the uterine cavity, accompanied by dark red congestion at the embryo implantation sites (Fig. 1A). In addition, the uterine weight of MA rats was markedly decreased compared to pregnancy control rats (Fig. 1A). No bleeding was observed in the uteri of the pregnancy control rats, while MA rats exhibited a significantly increased volume of bleeding and prolonged bleeding duration compared to controls (Fig. 1B). tPA and uPA activity levels in the fasting serum of rats on days 1, 8, and 14 of pregnancy were evaluated by ELISA. The results demonstrated that tPA and uPA levels exhibited no notable differences between MA and pregnancy control rats on days 1 and 8, but on day 14, both tPA and uPA levels in MA rats were significantly higher than those in controls (Fig. 1C). Estradiol and progesterone levels rose from day 1 to 14 in pregnancy control rats, but in MA rats, they decreased from day 8 to 14 (Fig. 1D). Notably, estradiol and progesterone levels in pregnant control rats were markedly higher than those in MA rats on day 14 (Fig. 1D). HE staining revealed obvious trophoblasts in the uterine tissues of the pregnancy control rats. However, MA rats exhibited an incomplete trophoblast structure, a large number of red blood cells, necrotic decidual cells, and uterine cavity congestion (Fig. 1E). TUNEL staining showed that apoptosis significantly increased in placental villus tissue of MA rats compared to controls (Fig. 1F). These findings confirmed the successful establishment of the MA model.

FIG. 1.

Medication abortion (MA) induced uterine bleeding and trophoblast tissue injury in rats. (A) The uterine morphology and weight of rats. Scale bar = 10 mm. (B) The duration and volume of uterine bleeding. (C) The activity levels of tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA) in the fasting serum of rats on days 1, 8, and 14 were measured by ELISA. (D) The levels of estradiol and progesterone in the fasting serum of rats on days 1, 8, and 14 were measured by ELISA. (E) Representative images of HE staining of rat uteri. Yellow arrows indicate decidual cells, and green arrows indicate trophoblasts. Scale bar = 200 μm. (F) Apoptosis in placental villus tissue was evaluated by TUNEL staining. Scale bar = 20 μm. Quantification of TUNEL-positive cells was performed using ImageJ in a blinded manner. Statistical significance was determined via an unpaired Student's t-test. **P < 0.01.

Excessive infiltration of immune cells and severe inflammatory responses lead to abortion, and TGF-β1 regulates the immune system and is involved in healthy and pathological pregnancies (Li et al., 2019; Yang et al., 2021). Both the mRNA and protein levels of IL-6 and TNF-α were markedly increased, while those of TGF-β1 were decreased in MA rats compared to pregnancy control rats (Fig. 2A and B). PAI-1 is the main inhibitor of the fibrinolytic system, and TGF-β promotes p53/Smads complex formation on the PAI-1 promoter and activates its transcription (Kawarada et al., 2016; Shetty et al., 2003). MA rats exhibited lower TGF-β1, p53, and PAI-1 levels compared to controls (Fig. 2C), suggesting that they may be involved in the mechanism of bleeding after MA.

FIG. 2.

MA induced inflammation and decreased the levels of TGF-β1, p53, and PAI-1 in the placental villus tissue of rats. (A) The levels of TGF-β1, IL-6, and TNF-α in the placental villus tissue of rats were detected by ELISA. (B) The levels of Tgfb1, Il6, and Tnf in the placental villus tissue of rats were measured by qRT-PCR. (C) The levels of p53, PAI-1, and TGF-β1 in the placental villus tissue of rats were detected by Western blot. Statistical significance was determined via an unpaired Student's t-test. **P < 0.01.

Overexpression of TGFB1 exerted protective effects on LPS-treated trophoblasts by upregulating p53 and PAI-1 expression

HTR8/SVneo cells were treated with different concentrations of LPS. Cell viability decreased with increasing LPS concentrations, with 50 and 100 ng/mL significantly inhibiting cell viability (Fig. 3A). Following treatment with LPS at 25, 50, and 100 ng/mL, TGF-β1 levels were markedly decreased, while IL-6 and TNF-α levels were notably increased in cells (Fig. 3B–D). LPS at 100 ng/mL was used for subsequent experiments.

FIG. 3.

LPS treatment inhibited cell viability and induced inflammation, while downregulating TGF-β1 levels in trophoblasts. (A) Cell viability was detected by CCK-8 assay. (B) The levels of TGFB1 in HTR8/SVneo cells were detected by qRT-PCR. (C) The levels of TGF-β1 in HTR8/SVneo cells were assessed by Western blot. (D) The levels of IL-6 and TNF-α in HTR8/SVneo cells were detected by ELISA. Statistical significance was determined via one-way ANOVA with Tukey's post hoc test. *P < 0.05, **P < 0.01.

TGF-β1 levels were significantly increased in HTR8/SVneo cells following overexpression of TGFB1 (Fig. 4A and B). TGF-β1, p53, and PAI-1 levels were significantly decreased following LPS induction but increased after overexpression of TGFB1 (Fig. 4C). Moreover, LPS treatment significantly inhibited cell viability, while overexpression of TGFB1 partially restored it (Fig. 4D). Cell migration was evaluated by transwell assay. The results revealed that LPS stimulation markedly inhibited cell migration, which was enhanced by overexpression of TGFB1 (Fig. 4E). Apoptosis was notably increased following LPS treatment but significantly decreased after TGFB1 overexpression (Fig. 4F). In addition, pro-caspase-3 and Bcl-2 levels were markedly decreased following LPS induction but increased after TGFB1 overexpression, while cleaved-caspase-3 and Bax showed opposite trends (Fig. 4G). Furthermore, IL-6 and TNF-α levels were markedly increased after LPS stimulation, while overexpression of TGFB1 significantly inhibited their levels (Fig. 4H). Overexpression of TGFB1 effectively alleviated LPS-induced trophoblast injury, potentially through the regulation of p53 and PAI-1.

FIG. 4.

Overexpression of TGFB1 exerted protective effects on LPS-treated trophoblasts by upregulating p53 and PAI-1 expression. (A) The levels of TGFB1 in HTR8/SVneo cells were detected by qRT-PCR. (B) The levels of TGF-β1 in HTR8/SVneo cells were assessed by Western blot. (C) The levels of p53, PAI-1, and TGF-β1 in HTR8/SVneo cells were detected by Western blot. (D) Cell viability was detected by CCK-8 assay. (E) Cell migration was evaluated by the transwell assay. (F) Apoptosis was detected by flow cytometry (UL = mechanically injured or necrotic cells, UR = late apoptotic cells, LL = live cells, LR = early apoptotic cells). (G) The levels of pro-caspase-3, cleaved-caspase-3, Bcl-2, and Bax in HTR8/SVneo cells were detected by Western blot. (H) The levels of IL-6 and TNF-α in HTR8/SVneo cells were detected by ELISA. Statistical significance was determined via one-way ANOVA with Tukey's post hoc test. *P < 0.05, **P < 0.01.

Overexpression of Tgfb1 reduced uterine bleeding in MA rats by upregulating p53 and PAI-1 expression

The oe-Tgfb1 or oe-NC lentivirus was injected into MA rats. Overexpression of Tgfb1 notably increased TGF-β1, p53, and PAI-1 levels in placental villus tissues of MA rats (Fig. 5A). The uteri in the MA and MA + oe-NC groups were dark red, with petechial hemorrhages observed in the uterine cavity and dark red congestion at reduced embryo implantation sites (Fig. 5B). In the MA + oe-Tgfb1 group, the uteri were light pink, with a few petechial hemorrhages in the uterine cavity and limited dark red congestion at the embryo implantation sites (Fig. 5B). Moreover, overexpression of Tgfb1 significantly reduced the uterine weight of MA rats (Fig. 5B). Notably, overexpression of Tgfb1 markedly reduced the uterine bleeding duration and volume in MA rats (Fig. 5C). On day 14 of pregnancy, tPA and uPA levels were significantly decreased in the fasting serum of MA + oe-Tgfb1 rats compared to MA + oe-NC rats (Fig. 5D). Estradiol and progesterone levels were markedly increased in MA rats after overexpression of Tgfb1 (Fig. 5E). HE staining revealed obvious trophoblast tissue damage accompanied by congestion in the uterine tissues of rats in the MA and MA + oe-NC groups, while MA + oe-Tgfb1 group exhibited no obvious decidual cell damage, and congestion was improved (Fig. 5F).

FIG. 5.

Overexpression of Tgfb1 reduced uterine bleeding and pathological damage in MA rats. (A) The levels of p53, PAI-1, and TGF-β1 in the placental villus tissue of rats were detected by Western blot. (B) The uterine morphology and weight of rats. Scale bar = 10 mm. (C) The duration and volume of uterine bleeding. (D) The activity levels of tPA and uPA in the fasting serum of rats on days 1, 8, and 14 were measured by ELISA. (E) The levels of estradiol and progesterone in the fasting serum of rats on days 1, 8, and 14 were measured by ELISA. (F) Representative images of HE staining of rat uteri. Scale bar = 200 μm. Statistical significance was determined via one-way ANOVA with Tukey's post hoc test. *P < 0.05, **P < 0.01.

Overexpression of Tgfb1 notably inhibited apoptosis in placental villus tissue of MA rats (Fig. 6A). Furthermore, overexpression of Tgfb1 significantly increased Bcl-2 and pro-caspase-3 levels, while reducing Bax and cleaved-caspase-3 levels in trophoblasts of MA rats (Fig. 6B). Compared to the MA + oe-NC group, IL-6 and TNF-α levels were significantly decreased in the MA +oe-Tgfb 1 group (Fig. 6C). Upregulation of the TGF-β1/p53/PAI-1 pathway in placental villus tissue may improve uterine bleeding after MA.

FIG. 6.

Overexpression of Tgfb1 inhibited apoptosis and inflammation in MA rats. (A) Apoptosis in placental villus tissue was evaluated by TUNEL staining. Scale bar = 20 μm. Quantification of TUNEL-positive cells was performed using ImageJ in a blinded manner. (B) The levels of pro-caspase-3, cleaved-caspase-3, Bcl-2, and Bax in trophoblasts were detected by Western blot. (C) The levels of IL-6 and TNF-α in in placental villus tissue were detected by ELISA. Statistical significance was determined via one-way ANOVA with Tukey's post hoc test. *P < 0.05, **P < 0.01.

Discussion

This study revealed that TGF-β1, p53, and PAI-1 were downregulated in MA models. In LPS-treated HTR8/SVneo cells, overexpression of TGFB1 increased p53 and PAI-1 levels, promoted cell proliferation and migration, and inhibited apoptosis and inflammation. Moreover, overexpression of Tgfb1 upregulated p53 and PAI-1 expressions, alleviated the pathological damage of the uterus, reduced uterine bleeding, and suppressed apoptosis and inflammation in MA rats. These findings highlight the role of TGF-β1 in trophoblasts and its regulatory mechanism in post-MA bleeding.

MA is an effective method for terminating early pregnancy (WHO, 2014). Studies have shown that the combined use of mifepristone and misoprostol induces cervical dilation and uterine contractions to expel tissues, and it is a safe alternative to surgical intervention (Austin, 2022). However, a study reported that 77% of pregnant women experienced severe bleeding after MA compared to menstruation (Reynolds-Wright et al., 2022). Consistent with this, we found that MA rats exhibited dark red congestion at embryo implantation sites and reduced uterine weight, with increased volume and duration of uterine bleeding. The fibrinolytic system is the key to maintaining hemostatic balance, and changes in its function can lead to bleeding (Hvas and Larsen, 2023). Studies have revealed that plasma tPA levels are significantly increased in pregnant patients after induced abortion (Yoshimura et al., 1991). Moreover, progesterone-induced estradiol can trigger decidualization of human endometrial stromal cells, thereby increasing PAI-1 expression and inhibiting plasminogen activator levels (Schatz et al., 2016). Similarly, we found that tPA and uPA levels in the fasting serum of rats were significantly higher than those in controls on day 14 after MA, whereas estradiol and progesterone levels showed the opposite trend.

The placenta is a multifunctional organ that connects maternal and fetal structures. It anchors the conceptus to the endometrium, prevents rejection by the maternal immune system, and allows the transfer of nutrients and waste (Maltepe et al., 2010). Trophoblasts are fetal-derived cells that directly interact with maternal immune cells at the maternal-fetal interface (Xu et al., 2021). Notably, subinvolution of placental bed vessels is regarded as a cause of bleeding after MA, likely resulting from abnormal interactions between maternal uterine cells and fetal trophoblast (Kavalar et al., 2012). We observed necrotic decidual cells and incomplete trophoblast tissue, accompanied by excessive apoptosis, in the rat uterus after MA induction. Moreover, LPS treatment inhibited the viability of HTR8/SVneo cells. Trophoblasts interact directly with decidual immune cells or indirectly through soluble factors, contributing to an immune tolerance microenvironment (Xu et al., 2021). A previous study revealed that induced abortion leads to increased IL-6, IL-8, and TNF-α levels in the amniotic fluid of the subjects, suggesting the occurrence of intrauterine inflammation (Roncari et al., 2013). Consistent with these findings, we found that MA induction increased IL-6 and TNF-α levels in the placental villus tissue of rats and LPS-treated HTR8/SVneo cells.

TGF-β1 is an anti-inflammatory cytokine that regulates the implantation and adhesion of trophoblasts, and its promoter gene polymorphism is associated with miscarriage (Dirisipam et al., 2025). Cao et al. revealed that the mechanism of mifepristone as an abortion drug might be partially through macrophages regulating the function of decidua natural killer cells, and TGF-β1 secreted by macrophages might be involved in this process (Cao et al., 2021). Similarly, we found that TGF-β1 levels were downregulated in MA models. PAI-1, a downstream target of TGF-β1, is a main inhibitor of plasminolytic enzyme activity and also a pathogenic factor involved in thrombosis and fibrosis-related diseases (Samarakoon et al., 2013). Studies have shown that p53 is necessary for PAI-1 expression, and TGF-β1 can initiate p53 phosphorylation (Overstreet et al., 2014). Our results indicated that TGF-β1, p53, and PAI-1 levels were markedly decreased in MA models, while overexpression of TGFB1 reversed these changes, suggesting that TGF-β1 regulates the p53/PAI-1 pathway in MA. Recent studies have revealed that THBS4 knockdown inhibits the biological functions of HTR8/SVneo cells by inhibiting TGF-β1 signaling (Shi et al., 2025). Consistent with this, we discovered that overexpression of TGFB1 enhanced cell proliferation and migration, and reduced apoptosis and inflammation in LPS-treated HTR8/SVneo cells. Furthermore, overexpression of Tgfb1 significantly alleviated uterine bleeding and inflammation in MA rats.

Although this investigation confirmed the role of the TGF-β1/p53/PAI-1 pathway in uterine bleeding after MA, some limitations should be acknowledged. Firstly, these findings were obtained from cellular and animal models. Whether this pathway is involved in regulating the mechanism of uterine bleeding after MA in humans still requires further clinical trials for verification. Furthermore, overexpression of TGFB1 can cause thrombosis and fibrosis-related diseases. Therefore, attention should be paid to strictly controlling the dosage when translating these findings into clinical applications.

Conclusion

TGF-β1 protects trophoblasts and improves uterine bleeding after MA by upregulating the p53/PAI-1 pathway. This study elucidated the mechanism influencing uterine bleeding after MA in early pregnancy, providing a new perspective for its treatment.

Authors’ Contributions

Q.H.: Conception and design of the research, statistical analysis, and drafting the article. H.L.: Statistical analysis, analysis, and interpretation of data. H.Y.: Acquisition of data. X.F.: Conception and design of the research, statistical analysis, analysis and interpretation of data. Y.C.: Conception and design of the research, acquisition of data, and revision of the article for important intellectual content. All the authors read and approved the final article.

Footnotes

Author Disclosure Statement

All authors declare no conflicting/competing interests.

Funding Information

This study was supported by the Zhejiang Provincial Medical and Health Science and Technology Plan (No. 2024KY1880).

Ethical Approval

This study was approved by the Experimental Animal Welfare Ethics Committee of Yangzhou University (No. 202507027).

Data Availability Statement

All data are available from the corresponding author upon request.

References

Assoian

, Komoriya

, Meyers

, et al. Transforming growth factor-beta in human platelets. Identification of a major storage site, purification, and characterization. J Biol Chem, 1983; 258(11):7155–7160.

Austin

. Medication abortion: A perfect solution? Health Serv Res, 2022; 57(4):717–719; doi: 10.1111/1475-6773.14013

Bridwell

, Long

, Montrief

, et al. Post-abortion complications: A narrative review for emergency clinicians. West J Emerg Med, 2022; 23(6):919–925; doi: 10.5811/westjem.2022.8.57929

Briel

, Kunz

, Kidess

. Platelet function, coagulation and fibrinolysis during termination of missed abortion and missed labor by PGF2 alpha and oxytocin. Acta Obstet Gynecol Scand, 1979; 58(4):361–364; doi: 10.3109/00016347909154596

Cao

, Tang

, Niu

, et al. Mifepristone regulates macrophage-mediated natural killer cells function in decidua. Reprod Biol, 2021; 21(3):100541; doi: 10.1016/j.repbio.2021.100541

Chen

, Bai

, Zhao

, et al. Ginsenoside represses symptomatic intracerebral hemorrhage after recombinant tissue plasminogen activator therapy by promoting transforming growth Factor-β1. J Stroke Cerebrovasc Dis, 2016; 25(3):549–555; doi: 10.1016/j.jstrokecerebrovasdis.2015.11.004

Dirisipam

, Madduru

, Jahan

, et al. TGF-β1 promoter functional gene polymorphism -509 C/T in the maternal susceptibility to recurrent pregnancy loss in South Indian women. Hum Immunol, 2025; 86(1):111182; doi: 10.1016/j.humimm.2024.111182

Farhat

, Al-Maliki

, Easa

, et al. TGF-β1 suppresses plasmin and MMP activity in flexor tendon cells via PAI-1: Implications for scarless flexor tendon repair. J Cell Physiol, 2015; 230(2):318–326; doi: 10.1002/jcp.24707

Flemmig

, Melzig

. Serine-proteases as plasminogen activators in terms of fibrinolysis. J Pharm Pharmacol, 2012; 64(8):1025–1039; doi: 10.1111/j.2042-7158.2012.01457.x

10.

Grainger

, Wakefield

, Bethell

, et al. Release and activation of platelet latent TGF-beta in blood clots during dissolution with plasmin. Nat Med, 1995; 1(9):932–937; doi: 10.1038/nm0995-932

11.

Greer

. Thrombophilia: Implications for pregnancy outcome. Thromb Res, 2003; 109(2–3):73–81; doi: 10.1016/s0049-3848(03)00095-1

12.

Hvas

, Larsen

. The fibrinolytic system and its measurement: History, current uses and future directions for diagnosis and treatment. Int J Mol Sci, 2023; 24(18):14179; doi: 10.3390/ijms241814179

13.

Kapp

, Lohr

. Modern methods to induce abortion: Safety, efficacy and choice. Best Pract Res Clin Obstet Gynaecol, 2020; 63:37–44; doi: 10.1016/j.bpobgyn.2019.11.008

14.

Kavalar

, Arko

, Fokter Dovnik

, et al. Subinvolution of placental bed vessels: Case report and review of the literature. Wien Klin Wochenschr, 2012; 124(19–20):725–730; doi: 10.1007/s00508-012-0219-9

15.

Kawarada

, Inoue

, Kawasaki

, et al. TGF-β induces p53/Smads complex formation in the PAI-1 promoter to activate transcription. Sci Rep, 2016; 6(1):35483; doi: 10.1038/srep35483

16.

Keragala

, Medcalf

. Plasminogen: An enigmatic zymogen. Blood, 2021; 137(21):2881–2889; doi: 10.1182/blood.2020008951

17.

Kumwong

, Sunder

, Akinbinu

18.

, Zhang

, Xu

, et al. Cell-cell contact with proinflammatory macrophages enhances the immunotherapeutic effect of mesenchymal stem cells in two abortion models. Cell Mol Immunol, 2019; 16(12):908–920; doi: 10.1038/s41423-019-0204-6

19.

Maltepe

, Bakardjiev

, Fisher

. The placenta: Transcriptional, epigenetic, and physiological integration during development. J Clin Invest, 2010; 120(4):1016–1025; doi: 10.1172/jci41211

20.

Mark

, Foster

, Aguilar

, et al. Documenting a new world: Research submitted to the national abortion federation’s 48th annual meeting (May 2–5, 2025, California). Perspect Sexual Reproductive, 2025; 57(2):241–278; doi: 10.1111/psrh.70010

21.

Medcalf

, Keragala

. Fibrinolysis: A primordial system linked to the immune response. Int J Mol Sci, 2021; 22(7):3406; doi: 10.3390/ijms22073406

22.

Overstreet

, Samarakoon

, Meldrum

, et al. Redox control of p53 in the transcriptional regulation of TGF-β1 target genes through SMAD cooperativity. Cell Signal, 2014; 26(7):1427–1436; doi: 10.1016/j.cellsig.2014.02.017

23.

Raymond

, Shannon

, Weaver

, et al. First-trimester medical abortion with mifepristone 200 mg and misoprostol: A systematic review. Contraception, 2013; 87(1):26–37; doi: 10.1016/j.contraception.2012.06.011

24.

Reynolds-Wright

, Johnstone

, McCabe

, et al. Adherence to treatment and prevalence of side effects when medical abortion is delivered via telemedicine: A prospective observational cohort study during COVID-19. BMJ Sex Reprod Health, 2022; 48(3):185–192; doi: 10.1136/bmjsrh-2021-201263

25.

Roncari

, Politch

, Sonalkar

, et al. Inflammation or infection at the time of second trimester induced abortion. Contraception, 2013; 87(1):67–70; doi: 10.1016/j.contraception.2012.09.016

26.

Samarakoon

, Overstreet

, Higgins

. TGF-β signaling in tissue fibrosis: Redox controls, target genes and therapeutic opportunities. Cell Signal, 2013; 25(1):264–268; doi: 10.1016/j.cellsig.2012.10.003

27.

Schatz

, Guzeloglu-Kayisli

, Arlier

, et al. The role of decidual cells in uterine hemostasis, menstruation, inflammation, adverse pregnancy outcomes and abnormal uterine bleeding. Hum Reprod Update, 2016; 22(4):497–515; doi: 10.1093/humupd/dmw004

28.

Shaikh

, Prabhu

, Bhandary

. Curcumin targets p53-fibrinolytic system in TGF-β1 mediated alveolar epithelial mesenchymal transition in alveolar epithelial cells. Endocr Metab Immune Disord Drug Targets, 2021; 21(8):1441–1452; doi: 10.2174/1871530320666200929142503

29.

Shetty

, Bdeir

, Cines

, et al. Induction of plasminogen activator inhibitor-1 by urokinase in lung epithelial cells. J Biol Chem, 2003; 278(20):18124–18131; doi: 10.1074/jbc.M207445200

30.

Shi

, Mao

, Cui

, et al. THBS4 downregulation alters trophoblast function in preeclampsia via the TGF-β1/Smad signaling cascade. Am J Physiol Cell Physiol, 2025; 329(1):C170–C182; doi: 10.1152/ajpcell.00387.2024

31.

WHO. WHO Guidelines Approved by the Guidelines Review Committee. In: Clinical Practice Handbook for Safe Abortion. World Health Organization: Geneva; 2014.

32.

, Li

, Sang

, et al. Crosstalk between trophoblasts and decidual immune cells: The cornerstone of maternal-fetal immunotolerance. Front Immunol, 2021; 12:642392; doi: 10.3389/fimmu.2021.642392

33.

Yang

, Dai

, Yuan

, et al. Role of transforming growth factor-β1 in regulating fetal-maternal immune tolerance in normal and pathological pregnancy. Front Immunol, 2021; 12:689181; doi: 10.3389/fimmu.2021.689181

34.

Yoshimura

, Nakamura

, Ito

, et al. Increase of plasma tissue-plasminogen activator antigen levels after induced abortion. Ric Clin Lab, 1991; 21(1):91–93; doi: 10.1007/bf02919117

35.

Zhang

, Sheng

, Yang

, et al. Integrative transcriptomics and proteomics analyses to reveal the therapeutic effect and mechanism of Buxue Yimu Pills in medical-induced incomplete abortion rats. J Ethnopharmacol, 2023; 305:116113; doi: 10.1016/j.jep.2022.116113

36.

Zhu

, Dai

, Yang

, et al. The mechanism of insulin-like growth factor II mRNA-binging protein 3 induce decidualization and maternal-fetal interface cross talk by TGF-β1 in recurrent spontaneous abortion. Front Cell Dev Biol, 2022; 10:862180; doi: 10.3389/fcell.2022.862180