PDLIM3 knockdown promotes ferroptosis in endometriosis progression via inducing Gli1 degradation and blocking Hedgehog signaling pathway

Abstract

Aims

Current evidence suggests that there is no completely effective method for endometriosis (EMS) without trauma due to diverse adverse effects. Reliable evidence illustrates that inhibiting ferroptosis is a potential strategy for EMS. We sufficiently verified that the expression of endogenous protein PDZ and LIM domain 3 (PDLIM3) was significantly increased in EMS.

Methods

PDLIM3 knockdown reduced primary ectopic endometrial stromal cells’ (EESCs) viability and migration, and elevated ferroptosis signaling indicators including Fe²⁺, malondialdehyde (MDA), and reactive oxygen species (ROS) in EESCs.

Results

Mechanistic studies revealed that inhibition of PDLIM3 accelerated glioma-associated oncogene-1 (Gli1) degradation and further deactivated Hedgehog signaling. Gli1 inhibitor, GANT61, abrogated the impact of PDLIM3 deletion on EESC growth, migration, and ferroptosis. In vivo experiments suggested that PDLIM3 reduction repressed the growth of endometrial lesions. Likewise, repression of PDLIM3 promoted ferroptosis and attenuated Hedgehog signaling in endometrial lesions.

Conclusions

Collectively, silencing of PDLIM3 facilitates ferroptosis in EMS by inducing Gli1 degradation and blocking Hedgehog signaling. It may provide an alternative strategy for developing therapeutic agents of EMS in the future.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-024-03131-8.

Introduction

Endometriosis (EMS) is a chronic gynecological condition affecting approximately 10% of reproductive-age women [1]. The etiology of EMS is complicated and multifactorial, including local immunity, hormone secretion, environment, and susceptible genes [2]. Therefore, identifying the mechanism implicated in the progression of EMS is essential for developing novel satisfactory therapeutic regimens for EMS.

Ferroptosis, a new form of programmed cell death, is characterized by iron dependence and lethal accumulation of lipid peroxides [3]. In recent years, ferroptosis has aroused great interest in disease treatment research communities because ferroptosis is closely related to the pathophysiological processes of many diseases [4]. Many genes are involved in the ferroptosis process by regulating amino acid metabolism and lipid peroxidation. For instance, acyl-CoA synthetase 4 (ACSL4) encourages the synthesis of phospholipids containing polyunsaturated fatty acids (PUFA), which is the primary substrates for lipid peroxidation [5]. Solute carrier family 7 member 11 (SLC7A11), a component of the cysteine/glutamate transporter family, is responsible for importing extracellular cysteine and exporting intracellular glutamate. Blocking the activity of SLC7A11 results in glutathione depletion and the inactivation of phospholipid hydroperoxide glutathione peroxidase 4 (GPX4) in ferroptosis [5, 6]. Recently, blocking of ferroptosis has emerged as a promising therapeutic strategy for cancers, atherosclerosis, etc. [7, 8].

EMS is characterized by ferroptosis resistance, which assists cells to spread via retrograde menstruation to survive, implant, and form endometriotic lesions within the abdominal cavity [9]. Li and Liang et al. demonstrate that ferroptosis resistance occurs frequently in EMS [6, 10]. Li et al. find that a ferroptosis inducer can trigger ferroptosis in ectopic endometrial stromal cells (EESCs) and alleviate the ectopic lesions in EMS [11]. Recently, bioinformatical analysis has identified that protein PDZ and LIM domain 3 (PDLIM3) is a potential biomarker associated with immune infiltration in patients with EMS [12]. High expression of PDLIM3 in EMS has been reported in several previous bioinformatics reports [13, 14]. Nevertheless, whether PDLIM3 is implicated in ferroptosis in EMS is not clear yet.

For the current report, the aberrant expressions of PDLIM3 in EMS specimens and EESCs were measured. Then, PDLIM3 is implicated in EESC proliferation, migration, and ferroptosis. Mechanistically, we found that PDLIM3 imbibition facilitates ferroptosis in EESCs by regulating Gli1 degradation and Hedgehog signaling.

Materials and methods

Gene expression datasets

Four public RNA sequencing and microarray datasets (GSE7305, GSE5108, GSE25628, and GSE11691) were downloaded from the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). Up-regulated mRNAs were analyzed using the online software GEO2R. Log2 fold change<-2 and P < 0.05 are two screening criteria. The results of the up-regulated genes were presented using a Venn diagram.

Clinical specimens

Forty-eight EMS patients and 46 controls suffering from fibroid or other benign gynecological conditions with infertility were recruited for this study from May 2021 to February 2024 in Songyuan Central Hospital. The inclusion criteria were infertile women aged 22 to 40 years, diagnosed as EMS, with regular menstrual cycles, and ranging from stages I to IV. The stage of EMS was determined based on the revised American Society for Reproductive Medicine (ASRM) classification system [15]. The exclusion criteria were ovarian cancer, pelvic inflammation, primary dysmenorrheal, and other benign ovarian cysts. Normal endometrial tissues (controls) were collected from the normal controls and ectopic endometrial tissues were obtained from patients with EMS. The endometrial samples were collected during the hysteroscopy and determined to be in the proliferative phase of the menstrual cycle. All the patients and controls had regular menstrual cycles and had not received any hormonal medications within 6 months before the specimen collection. The protocol was approved by the Ethics Committee of Songyuan Central Hospital. All the participants had signed the informed consent.

Isolation and culture of primary EESCs

Primary normal endometrial stromal cells (NESCs), eutopic ESCs, and EESCs were isolated from normal endometrial tissues from controls and eutopic or ectopic endometrial tissues in EMS, respectively [16]. After being washed with PBS, endometrial tissues were cut into pieces, and digested with 5% dispase enzyme and collagenase (Sigma-Aldrich, St. Louis, MO, USA) for 60 min at 37 °C. Then, the tissue pieces were filtered through 100-µM and 70-µM nylon cell strainers. After centrifugation (1000 g, 5 min), the precipitate was cultured in red blood cell lysis buffer (Sigma-Aldrich) for 15 min to remove erythrocytes. A 40-µM sieve was used to separate epithelial cells and obtain enriched stromal cells. The human embryonic kidney 293T (HEK293T) cell line was obtained from the American Type Culture Collection (ATCC, Manassas). HEK293T cells were maintained in DMEM. Eutopic ESCs, EESCs, and NESCs were grown in DMEM/F-12 supplemented with 10% FBS (Thermo Fisher Scientific, CA, USA) and 1% penicillin/streptomycin (Beyotime, Shanghai, China) in a humidified 5% CO₂ environment at 37 °C.

Cell transfection and treatment

The whole length of PDLIM3 was amplified by PCR and constructed to pcDNA3.1 vector to construct PDLIM3 overexpression plasmid (pcDNA3.1-PDLIM3). The siRNA targeting PDLIM3 (si-PDLIM3), siRNA negative control (si-NC), empty vector, and pcDNA3.1-PDLIM3 were designed and constructed by Genepharma Corporation (Shanghai, China). EESCs were seeded in 6-well culture plates containing 2 × 10⁶ cells/well. When reaching 70% confluence, transfection was performed in EESCs using Lipofectamine 2000 kits (Thermo Fisher Scientific) following standard protocols. After 6 h, the mixtures were removed, and cells were cultured in the complete medium at 37 °C supplemented with 5% CO₂. Forty-eight hours later, cells were collected for following experiments or treated with several indicated inhibitors, including ferroptosis inhibitor ferrostatin-1 (Fer-1, 1 µM), necroptosis inhibitor necrostatin-1 (10 µM), apoptosis inhibitor ZVAD-FMK (10 µM), pyroptosis inhibitor LDC7559 (5 µM), or Hedgehog signaling agonist, purmorphamine (PM, 1 µM) for another 12 h.

Quantitative real-time PCR (qRT-PCR)

Ectopic endometrial tissues and normal endometrial tissues were taken, cut into pieces, and then homogenated using Trizol reagent (Thermo Fisher Scientific). The tissues and EESCs were dissociated on ice for 30 min to isolate the total RNA. Then, cDNA was synthesized with the reverse transcription kit (TransGen Biotech, Beijing, China). qRT-PCR detection was performed on an ABI 7900 real-time PCR system (Applied Biological Systems, USA). GAPDH was used as an internal control, and relative gene level was analyzed using the 2^-ΔΔCT method. The primer sequences are shown in Supplementary Table 1.

Western blot

The harvested endometrial tissues and EESCs were lysed in RIPA lysis. Twenty micrograms of protein sample was further separated using 8% SDS–PAGE and then transferred to PVDF membranes (Millipore, Billerica, USA). The membranes were incubated with specific antibodies against Gli1 (1:1000, Proteintech, Wuhan, China), SMO (Santa Cruz Biotechnology, Texas, USA), PTCH1 (1:1000, Abcam, Cambridge, MA, USA), PTCH2 (1:1000, Abcam), GPX4 (1:1000, Proteintech), SLC7A11 (1:1000, Proteintech), and GAPDH (1:1000, Abcam) for 12 h at 4 °C. After washing three times in TBST, the membranes were incubated with an HRP-labeled secondary antibody (1:5000, Proteintech) for 2 h at room temperature. An enhanced chemiluminescence (Millipore) was used to visualize the bands.

Cell counting kit (CCK)-8 assay

The proliferation of EESCs was measured using a CCK-8 kit (Beyotime). Cells (5 × 10³ cells/well) were seeded into 96-well plates. After 24 h, cells were transfected with si-PDLIM3 or pcDNA3.1-PDLIM3 in the absence or presence of indicated inhibitors. Next, 10 µL of CCK-8 solution was added to the plates. After incubation for 4 h at 37 °C, the absorbance at 450 nm was recorded on a microplate reader (Thermo Fisher Scientific).

5-Ethynyl-2′-deoxyuridine (EdU) assay

EESCs (5 × 10³) were seeded in 12-well plates and cultured 24 h. Then, the cells were incubated with EdU solution (Beyotime) for 30 min. After fixed by 4% paraformaldehyd for 20 min, 0.3% Triton X-100 was applied to permeabilize EESCs for 15 min. Five micrograms per milliliter of DAPI (Beyotime) was applied to stain the cell nucleus at 37 °C for 10 min. Finally, images were captured using a fluorescence microscope.

Transwell migration assay

Cell migration ability was evaluated by Transwell assay using a Transwell chamber (24-well, 8 µM pore size). After specific transfection and treatments, 100-µL cells (5 × 10⁴) were placed into the top chamber, and 600 µL medium supplemented with 10% FBS was added into the lower compartment. Cells were incubated at 37 °C for 24 h. The cells on the bottom surface of the inserts were fixed using 4% paraformaldehyde for 10 min. Then, cells were stained by 1% crystal violet for 15 min. Cells were photographed under an inverted microscope and quantified by counting in five random fields.

Detection of MDA and SOD levels

The endometrial lesions were homogenized, and centrifugated, and the supernatants were collected. EESCs were lysed, and supernatants were collected. The activity of SOD and the level of malondialdehyde (MDA) were examined with ELISA kits (Jiancheng Bioengineering Institute, Nanjing, China).

Reactive oxygen species (ROS) level detection in EESCs

The level of ROS in EESCs was tested by DCFH-DA staining. EESCs were dyed with 5 µM DCFH-DA (Beyotime) at 37 °C for 20 min in the dark. Then, samples were rinsed with PBS three times to remove the unbound dyes, and the DCF fluorescence was detected using CytoFlex S Flow cytometry (Beckman Coulter, CA, USA).

Measurement of ROS production in endometrial lesions

Approximately 100 mg of endometrial lesions was homogenized in PBS and single-cell suspension was prepared using enzymatic digestion [17]. The ROS level was determined using a ROS Fluorometric Assay Kit (Elabscience, Wuhan, China). Fluoresce intensity was determined using a fluorescence microplate reader.

Determination of Fe²⁺ level

Fe²⁺ level in endometrial lesions tissues and EESCs was determined using an Iron Assay Kit (Abcam). Briefly, endometrial lesion tissues were homogenized and centrifugated and the supernatants were collected. EESCs were lysed, and supernatants were collected. Five microliters of iron buffer was added into 100 µL supernatant, followed by incubation for 30 min. Finally, the optical density (OD) value was immediately detected at 593 nm.

Luciferase reporter assay

Human Gli1 promoter (TransGen Biotech) was subcloned into pGL3-Basic vector (Promega) to obtain the pGL3‐Gli1‐Luc vector. HEK-293T cells (5 × 10⁴) were seeded into 6-well plates. si-PDLIM3 or pcDNA3.1-PDLIM3 plasmid combination with pGL3‐Gli1‐Luc vector was co-transfected into HEK-293T cells for 48 h. The Renilla luciferase was used as an internal control. Luciferase activity was assayed using a Dual-Luciferase reporter assay system (Promega, Madison, WI, USA).

Co-immunoprecipitation (Co-IP) assay

Cells were incubated with 300 µL lysis buffer on ice for 5 min. Cell lysates were incubated with primary antibody at 4 °C for 24 h, followed by addition of 50 µL protein G agarose beads. After being washed three times with lysis buffer, the immunoprecipitates were separated by centrifugation at 10,000 rpm for 30 s, and then heated with buffer for western blot assay.

Establishment of EMS model

Seven-week-old female ICR mice were purchased from Shanghai SLAC Animal Center (Shanghai, China) and maintained following the guidelines for the use and care of laboratory animals. The uterine horns from the donor mice were removed and cut into 1-mm³ pieces of fragments. Each section was opened longitudinally and kept in cold PBS until implantation. The uterine biopsies were sutured to the subcutaneous abdominal cavity of recipient mice using 4-0 vicryl sutures. After endometrial-like lesions were established (8 days), the recipient mice were randomly divided into four groups (n = 5 per group): model, Ad-shRNA-negative control (Ad-sh-NC), Ad-shRNA-PDLIM3 (Ad-sh-PDLIM3), and 10 mg/kg GANT61 group. In the GANT61 group, mice received 10 mg/kg GANT61 via intraperitoneal (i.p.) injection every 3 days. In the model group, mice received an equal volume of saline. Adenoviral vectors containing Ad-sh-NC or Ad-sh-PDLIM3 were purchased from Genepharma Co., Ltd (Shanghai, China). 1 × 10⁸ virus particles [vp]/20 µL Ad-sh-NC or Ad-sh-PDLIM3 adenovirus solutions were injected into mice via tail intravenous (i.v.) injection every week. Estrogen 10 mg/kg was administered subcutaneously three times a week throughout the experiment. Lesions were measured with two perpendicular diameters (the length and the width) using a vernier caliper every 3 days. The lesions volume = (length × width²)/2). Fifteen days after treatment, endometrial lesions were collected. The animal experimental procedures were approved by the Animal Experimental Ethics Committee of Songyuan Central Hospital.

Statistical analysis

All statistical analyses were conducted using GraphPad Prism 7.0. The experimental data are presented as mean ± standard deviation (SD). Statistical comparisons were performed using Student’s t-test for two groups or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple groups. P < 0.05 is considered as a statistically significant.

Results

Abnormally high expression of PDLIM3 in EMS

Up-regulated genes associated with EMS (324 in GSE7305, 256 in GSE5108, 104 in GSE25628, and 206 in GSE11691) were identified. There were seven genes (GATA6, PTGIS, ADH1B, MYH11, PDLIM3, FZD7, and WISP2) among the four EMS datasets as shown in the Venn diagram (Fig. 1A). qRT-PCR analysis was performed on 46 eutopic endometrial tissues (controls) from normal endometrium specimens of women without EMS, and 46 ectopic endometrial tissues of the patients with EMS. The results showed that PDLIM3 and GATA6 were significantly up-regulated in EMS compared to the controls (Fig. 1B). Aberrant GATA6 expression has been implicated in the progression of EMS, while the biological significance of PDLIM3 in EMS is still poorly understood [18, 19]. To explore ferroptosis status in EMS, immunoblotting analysis was executed to examine the expressions of ferroptosis-related biomarkers in control and EMS tissues. PDLIM3, GPX4, SLC7A11, and Gli1 expressions were up-regulated in EMS tissues when compared with controls (Fig. 1C). Similarly, the expressions of PDLIM3, SLC7A11, GPX4, and Gli1 were elevated in primary EESCs and eutopic ESCs when compared to NESCs (Fig. 1D). These results suggest that PDLIM3 expression is abnormally up-regulated in EMS.

Fig. 1.

PDLIM3 is up-regulated in ectopic endometrial tissues of patients with EMS and EESCs. A Venn diagram showing the numbers of up-regulated genes in EMS among the four datasets. B qRT-PCR detection of GATA6, PTGIS, ADH1B, MYH11, PDLIM3, FZD7, and WISP2 mRNA levels in ectopic endometrial specimens from patients with EMS and normal endometrial tissues (controls). C Western blot detection of PDLIM3, GPX4, SLC7A11, and Gli1 expressions in ectopic endometrial tissues from patients with EMS and control tissues. D Western blot detection of PDLIM3, GPX4, SLC7A11, and Gli1 expressions in EESCs, eutopic ESCs, and NESCs. ^**P < 0.01 vs. control tissues or NESCs

PDLIM3 knockdown reduces the proliferation and migration of EESCs

We next endeavored to explore the possible biological roles of aberrantly expressed PDLIM3 in EMS. PDLIM3 knockdown was conducted in EESCs using si-PDLIM3 transfection (Fig. 2A). CCK-8, EdU staining, and Transwell analysis revealed that PDLIM3 silencing markedly reduced the proliferation and migration of EESCs when compared to the si-NC transfection group (Fig. 2B–D). EESCs were treated with 10 µM ZVAD-FMK (an inhibitor of apoptosis), 10 µM necrostatin-1 (an inhibitor of necroptosis), 5 µM LDC7559 (an inhibitor of pyroptosis), or 1 µM ferrostatin-1 (Fer-1, an inhibitor of ferroptosis) in the presence of PDLIM3 knockdown. Results of CCK-8 and Transwell assays indicated that PDLIM3 knockdown-mediated EESC proliferation and migration inhibition was significantly reversed by Fer-1 but not ZVAD-FMK, necrostatin-1, or LDC7559 treatment (Fig. 2E, F). Therefore, these data indicate that PDLIM3 regulates EESCs viability and migration probably via regulating ferroptosis.

Fig. 2.

PDLIM3 knockdown reduces EESC proliferation and migration. A si-NC or si-PDLIM3 was transfected into EESCs followed by detection of transfection efficiency. B CCK-8 assay was applied to detect the cell viability of EESCs. C Representative images of Edu staining for EESC proliferation. D Transwell assay was used to determine migration of EESCs. E si-PDLIM3-transfected EESCs were treated with 10 µM ZVAD-FMK, 10 µM necrostatin-1, 5 µM LDC7559, or 1 µM Fer-1. CCK-8 assay was performed to measure cell proliferation. F Transwell assay was conducted to evaluate cell migration. Scar bar, 200 µM. ^**P < 0.01 vs. control

Deletion of PDLIM3 induces EESC ferroptosis

To validate whether PDLIM3 affects EESC ferroptosis, an in vitro assessment of ferroptosis-related signaling was conducted. The down-regulation of PDLIM3 distinctly enhanced the levels of ROS (Fig. 3A), Fe²⁺ (Fig. 3B), and MDA (Fig. 3C), and declined the level of antioxidant enzyme, SOD (Fig. 3D) in EESCs. The Hedgehog signal reception system consists of ligands, including Sonic Hedgehog (SHH), Indian Hedgehog (IHH), and Desert Hedgehog (DHH); transmembrane receptors Patched (PTCH1 and PTCH2); transmembrane protein Smoothened (SMO) that transduces the Hedgehog signal across the plasma membrane; and transcription factor glioma-associated oncogene (Gli) family (Gli1, Gli2, Gli3) [20]. Of these three Gli proteins, Gli1 is the final and key output of Hedgehog signaling [21, 22]. Herein, the result of western blot indicated that inhibition of PDLIM3 remarkedly lessened Gli1 expression in EESCs, while the expressions of SMO, PTCH1, and PTCH2 were not altered by PDLIM3 knockdown (Fig. 3E). Furthermore, PDLIM3 inhibition leads to a reduction in GPX4 and SLC7A11 expression. Altogether, loss of PDLIM3 leads to inactivation of the Hedgehog signaling pathway and ferroptosis in EESCs.

Fig. 3.

Deletion of PDLIM3 decreases EESC proliferation and migration via inducing ferroptosis. A–D Levels of Fe²⁺, MDA, ROS, and SOD were detected in EESCs upon si-PDLIM3 transfection. E Expression levels of Gli1, PTCH1, PTCH2, GPX4, and SLC7A11 in EESCs upon si-PDLIM3 transfection were determined using immunoblot. Scar bar, 200 µM. ^**P < 0.01 vs. control

Deletion of PDLIM3 induces EESC ferroptosis partially via inactivating Hedgehog signaling

To investigate the biological roles of Hedgehog signaling in EESC ferroptosis caused by PDLIM3 knockdown, purmorphamine (PM), an agonist of the Hedgehog pathway, was used to treat si-PDLIM3-transfected EESCs. After the EESCs were treated with PM (1 µM), the cell viability was higher than that of the control group [23]. PDLIM3 knockdown-mediated inhibition of EESCs viability was remarkedly reversed by PM treatment (Fig. 4A). EdU staining and Transwell assays revealed that PDLIM3 knockdown-mediated inhibition of proliferation and migration in EESCs was significantly prevented by PM treatment (Fig. 4B and C). As expected, PM reversed the ferroptosis-inducing effect of si-PDLIM3 on EESCs (Fig. 4D and G). Additionally, the down-regulation of GPX4 and SLC7A11 expression levels caused by si-PDLIM3 was reversed by PM (Fig. 4H). These data indicate that PDLIM3 causes ferroptosis via inactivating the Hedgehog signaling axis.

Fig. 4.

si-PDLIM3 induces ferroptosis of EESCs via Hedgehog signaling pathway. A EESCs were treated with si-PDLIM3 or si-PDLIM3 combination with PM. The cell viability of EESCs was detected by CCK-8 assay. B Representative images of Edu staining for EESC proliferation. C Transwell assay was used to determine migration of EESCs. D–G Levels of Fe²⁺, MDA, ROS, and SOD were detected in EESCs. H Expression levels of GPX4 and SLC7A11 in EESCs were measured using western blot method. Scar bar, 200 µM. ^**P < 0.01 vs. control. ^##P < 0.05 vs. si-PDLIM3

Deletion of PDLIM3 induces proteasome-dependent degradation of Gli1

To explore the possible mechanism by which PDLIM3 regulates Gli1 expression, the interaction between Gli1 and PDLIM3 was evaluated by Co-IP (Fig. 5A). Upon altering expression of PDLIM3, the mRNA level of Gli1 quantified by qRT-PCR did not show significant change (Fig. 5B). Overexpression or knockdown of PDLIM3 did not affect Gli1 promoter activity (Fig. 5C). However, the Gli1 protein level was found to be decreased in si-PDLIM3-transfected EESCs (Fig. 3E), while it was raised in PDLIM3 overexpressing EESCs (Fig. 5D). These results indicate a role that PDLIM3 affects Gli1 at a post-transcriptional level rather than transcriptionally. Subsequently, EESCs were treated with the proteasomal inhibitor MG132 (2 µM). As shown in Fig. 5E, the inhibiting effect of si-PDLIM3 on Gli1 was attenuated in EESCs upon treated with MG132. Next, EESCs were treated with a protein synthesis inhibitor, Cyclohexamide (CHX, 1 µM). As shown in Fig. 5F, knockdown of PDLIM3 increased Gli1 degradation rate. Altogether, inhibition of PDLIM3 induces proteasome-dependent degradation of Gli1 in EESCs.

Fig. 5.

Inhibition of PDLIM3 promotes degradation of Gli1. A The interaction between PDLIM3 and Gli1 was estimated using co-immunoprecipitation. Goat anti-rabbit IgG is used as negative control. B Quantification of Gli1 mRNA level in si-PDLIM3 or pcDNA3.1-PDLIM3 plasmid-transfected EESCs. C si-PDLIM3 or pcDNA3.1-PDLIM3 plasmid was transfected into HEK-293T and lysed after 48 h incubation. The percentage of luciferase activity was calculated. D Protein expression of Gli1 in pcDNA3.1-PDLIM3 plasmid-transfected EESCs was determined by western blot. E si-PDLIM3-transfected EESCs were dealt with 2 µM MG132 for 6 h. Western blotting was performed. F si-PDLIM3-transfected EESCs were dealt with 2 µM CHX. Western blotting was performed

Inhibition of PDLIM3 reduces the growth of endometrial lesions in vivo

To determine whether PDLIM3 is crucial for EMS progression in vivo, we established an EMS mouse model (Fig. 6A). Mice were treated with adenovirus containing sh-PDLIM3 (Ad-sh-PDLIM3) via tail intravenous injection (i.v.) or 10 mg/kg GANT61 via intraperitoneal (i.p.) injection. When compared to the Ad-sh-NC group, the volume of ectopic endometrial cysts was reduced by Ad-sh-PDLIM3 administration (Fig. 6B, C). The body weights of mice across all groups were found to be statistically indistinguishable (Fig. 6D). Furthermore, western blot analysis revealed that the expressions of Gli1, GPX4, and SLC7A11 were decreased in the Ad-sh-PDLIM3- and GANT61-treated groups as compared with the model group (Fig. 6E). Simultaneously, treatment with Ad-sh-PDLIM3 elevated the levels of ferroptosis markers (Fe²⁺, ROS, and MDA) and decreased SOD level (Fig. 6F–H). Therefore, inhibition of PDLIM3 exerts a suppressing effect on the growth of endometrial lesions and promotes ferroptosis.

Fig. 6.

Inhibition of PDLIM3 restrains the growth of endometriotic lesions by blocking Hedgehog signaling pathway in vivo. A A schematic diagram for the mouse EMS therapeutic model. B Representative endometriotic lesions in the mouse model of EMS after Ad-sh-PDLIM3 or 10 mg/kg GANT61 administration. C The volume of ectopic lesions. D Body weights of Ad-sh-PDLIM3 or GANT61 (10 mg/kg) treatment group were measured every day. E The protein expressions of Gli1, GPX4, and SLC7A11 in endometrial lesions from Ad-sh-PDLIM3 or GANT61 group were analyzed by western blot. F–H Levels of Fe²⁺, MDA, ROS, and SOD in endometrial lesions were detected using ELISA method. ^**P < 0.01 vs. model

Discussion

The biological meaning of ferroptosis in EMS is gradually being revealed. Multiple lines of evidence have highlighted the potential of ectopic endometrial tissues to resist ferroptosis and facilitate ectopic lesions growth by attenuating local cellular ferroptosis in peritoneal fluid of patients with EMS [10, 24]. Ferroptosis resistance is correlated with primary EESC growth and proliferation [25]. In the current study, we identified that PDLIM3 was remarkably up-regulated in EMS tissues and EESCs when compared with controls. Moreover, PDLIM3 silencing inhibited EESC growth and migration, and significantly increased ferroptosis in EESCs. Mechanistically, we showed that PDLIM3 facilitates ferroptosis in EESCs through blocking Hedgehog signaling.

EESC growth and invasion are involved in the development of EMS and may be used as a cell model for this disease modeling [26]. Inhibition of EESCs biological characteristics is a potential molecular therapeutic strategy against EMS [27, 28]. The EESC phenotype assay showed that PDLIM3 knockdown inhibited EESC growth and invasion by initiating ferroptosis. Lipid peroxidation and oxidative stress are two critical events in ferroptosis. Upon PDLIM3 silencing, the levels of intrinsic Fe²⁺, ROS, and MDA were overloaded, while the activity of antioxidant enzyme, SOD, declined in EESCs. The core molecular machinery of ferroptosis is primarily regulated by two pathways, extrinsic-dependent pathways (e.g., reduces cysteine or glutamine uptake) and intrinsic-regulated pathways (e.g., inhibition of GPX4 antioxidant system). PDLIM3 inhibition led to decreased expressions of endogenous GPX4 and SLC7A11 in EESCs. The ferroptosis inhibitor, Fer-1, abolished the inhibition of PDLIM3 knockdown in EESCs further corroborating that PDLIM3 specifically induced ferroptosis.

Hedgehog signaling is a developmental pathway that is activated in some endometrial cancers and the endometrium of women with EMS [20]. Highly specific antibodies recognize Hedgehog ligands and reduce the number of lesions on mesentery in a mouse model of EMS [29]. At present, most known antagonists (NVP-LDE-225 and GDC-0449) of the Hedgehog pathway target SMO and its upstream and downstream pathways, but SMO is susceptible to mutations and chemical resistance [30, 31]. In addition to SMO inhibitors and antibodies targeting Hedgehog ligands, direct inhibitors of Gli1 involved in the canonical Hedgehog-Gli1 pathway may be required for effective treatment in EMS. Deletion of PDLIM3 enormously compromises cilia formation and interferes with Hedgehog signaling transduction in medulloblastoma cells, implying that PDLIM3 activates Hedgehog signaling by supporting ciliogenesis [32]. In medulloblastoma, PDLIM3 expression is associated with Hedgehog pathway activation status [33]. Previous studies have shown that activation of Hedgehog-Gli1 signaling is a negative factor for ferroptosis [34, 35]. Similarly, we demonstrated that deletion of PDLIM3 promoted ferroptosis in EESCs via blocking Hedgehog signaling.

When Hedgehog ligands bind and inactivate the Hedgehog receptor (PTCH1 and PTCH2), PTCH1/PTCH2 loses its catalytic inhibition of the G-protein-coupled receptor-like signal transducer SMO, which triggers the transcriptional activation of Gli1, a zinc finger transcription factor [20]. SMO and Gli1 are strongly increased with clinical stages in the eutopic endometrium, which suggests that the Hedgehog signaling pathway is abnormally activated in EMS [20]. Gli1 inhibition may block EMS progression and therefore be used as target for drug discovery programs [26]. The endogenous SMO, PTCH1, and PTHC2 expressions within EESCs were not altered by PDLIM3 inhibition, suggesting PDLIM3-affected Hedgehog signaling is not dependent on influencing receptors. Subsequently, Gli1 protein expression, not the mRNA level, was differentially affected by PDLIM3 inhibition in EESCs, suggesting a post-transcription modification of Gli1 by PDLIM3. Furthermore, we proved that repressed PDLIM3 inactivated Hedgehog signaling by promoting Gli1 degradation. Finally, like GANT61, PDLIM3 inhibition induces ferroptosis and shrinks the volumes of ectopic endometriotic lesions in a mice model of EMS [36]. Consistent with the results in vitro, the expressions of Gli1, GPX4, and SLC7A11 were decreased in endometrial lesions in mice after treatment with Ad-sh-PDLIM3.

Ferroptosis shows a bidirectional regulation in the pathological progression of EMS. Multiple studies have confirmed that EESCs suppress the ferroptosis process and further benefit the proliferation and migration of endometrial stromal cells [24, 25]. Besides, the occurrence of ferroptosis by EESCs in contact with cyst fluid triggers the production of angiogenic, inflammatory, and growth cytokines in EMS [37]. Whether there is a threshold of ferroptosis in EMS and whether PDLIM3 can regulate the threshold deserve further study. When Hedgehog binds to the PTCH receptor, SMO released to activate the translocation of Gli1 from cytoplasmic into the nucleus. We will focus on exploring whether PDLIM3 regulates nuclear translocation of Gli1 in addition to protein degradation. Growth of endometriotic cells in EMS is promoted via multiple signal pathways, including PI3K/AKT/mTOR, MAPK/MEK/ERK, and Rho/ROCK [38]. Whether PDLIM3 has an impact on these signaling pathways and the underlying exact molecular mechanism in EMS are worth further investigation. This study did not include a sufficient number of cases to analyze the diagnostic value and prognostic significance of PDLIM3 in EMS.

Conclusion

In summary, we report herein that inhibition of PDLIM3 negatively regulates Hedgehog signaling pathway by inducing Gli1 degradation in EESCs, thus results in decreased EESC proliferation and migration and enhanced cell ferroptosis. The identification of PDLIM3/Gli1 axis improves our understanding of the pathogenesis of EMS.

Data availability

All available data has been reported in the manuscript.

Declarations

Competing interests

The authors declare no competing interests.