Abstract
Objective: Abnormal expression of circular RNA (circRNA) leads to the occurrence and development of endometriosis (EM), but its underlying mechanisms are largely unknown. Methods: Abnormally expressed circRNAs were screened in EM and normal tissues. A series of gain-of-function or loss-of-function experiments were conducted to evaluate the biological behavior of circ_0008927 in EM cells. The role of circ_0008927 in the proliferation, migration, and invasion of EM cells was investigated. The downstream mechanisms of circ_0008927 were studied through bioinformatics analysis and RNA sequencing, and this was confirmed through RNA immunoprecipitation and dual-luciferase reporter assays. Results: Circ_0008927 is highly expressed in EM tissues. From a biological perspective, silencing circ_0008927 can inhibit the proliferation, migration, and invasion of EMS cells in vitro. Mechanistically, circ_0008927 can interact with miR-608 through a competitive endogenous RNA mechanism, upregulating prominin2 (PROM2) and inhibiting ferroptosis in EM, thereby exacerbating the progression of EM. Conclusions: Our research results not only reveal the key role of circ_0008927 in regulating the progression of EM but also advocate for attenuating the circ_0008927/miR-608/PROM2 regulatory axis to combat EM.
Introduction
Endometriosis (EM) is characterized by the abnormal proliferation and infiltration of normal endometrial tissue outside the uterine cavity, with a prevalence rate of approximately 10% [1, 2]. This hormone-dependent condition manifests through symptoms such as pelvic pain, menstrual irregularities, and infertility. It significantly diminishes the quality of life for those affected and may result in long-term fertility impairment and increased cancer risk [3, 4]. Consequently, further fundamental research is essential to elucidate the pathogenesis of EM and to identify innovative diagnostic and treatment strategies.
Noncoding RNAs, including microRNAs (miRNAs), long noncoding RNAs, and circular RNAs (circRNAs), have been extensively studied for their roles in various molecular pathological and physiological processes [5, 6]. Unlike linear RNAs, circRNAs are closed-loop structures that lack both a 5′ cap or 3′ tail. Due to their widespread expression, high stability, and abundant presence in tissues, blood, and exosomes, circRNAs have the potential to serve as biomarkers for the early diagnosis and prognosis of diseases [7, 8]. Furthermore, circRNAs can regulate disease mechanisms by functioning as miRNA sponges. For instance, the knockdown of circ_0007331 suppresses the progression of EM through the miR-200c-3p/hypoxia inducible factor 1 subunit alpha (HIF-1α) axis [9]. Additionally, circ_103470 and circ_101102 have been found to regulate the epithelial-mesenchymal transition of EM through the miR-141-5p sponge [10]. Therefore, the exploration of circRNAs in the context of EM holds significant implications.
Ferroptosis is a regulated form of cell death characterized by iron-dependent lipid peroxidation of the cell membrane [11]. The ultrastructural changes associated with ferroptosis primarily include cell membrane rupture and blebbing, mitochondrial condensation, increased membrane density, reduced or absent mitochondrial cristae, and preserved nuclear morphology without chromatin condensation [12]. Oxidative stress elevates the production of reactive oxygen species (ROS) in the body, which, in the presence of free ferrous iron, leads to the peroxidation of polyunsaturated fatty acids that are abundantly present in the cell membrane [13]. The resulting hydroperoxides from polyunsaturated fatty acids further oxidize the cell membrane, resulting in a loss of membrane integrity and ultimately inducing ferroptosis. Existing studies have demonstrated that ferroptosis is implicated in the occurrence and progression of various diseases, including EM, cancer, immune-related diseases, and ischemia-reperfusion injury [14, 15]. In a mouse model of EM, the administration of erastin reduced the volume of EM-like lesions. Furthermore, the regulation of intracellular iron concentration by ferroportin (FPN) serves as a negative regulatory factor for erastin-induced ferroptosis [16]. Inducers of ferroptosis may represent promising novel therapeutic strategies for EM.
In this study, we identified an aberrantly elevated expression of circ_0008927 in EM. Subsequent in vitro experiments demonstrated the regulatory effect of the circ_0008927/miR-608/(prominin2) PROM2 axis on the proliferation, migration, invasion, and ferroptosis of EM cells. By exploring the functions and molecular mechanisms of circ_0008927, we aim to advance the understanding of the pathogenesis of EM and provide new avenues for its early diagnosis and treatment.
Methods and Materials
Clinical Samples
This study included patients treated for secondary infertility due to EM at Ningbo Women and Children’s Hospital between 2020 and 2022. Fourteen cases, confirmed to have EM through histological examination following laparoscopic surgery, were selected. Fresh eutopic endometrial tissue was obtained from these patients. Additionally, endometrial samples from 22 patients with tubal adhesions, who were matched for age and basal endocrine levels but did not have EM confirmed via laparoscopy, were used as the control group. All endometrial tissues were collected during the proliferative phase of the menstrual cycle. None of the 14 patients with EM had received hormone treatment within 3 months prior to the study. Following collection, the samples were promptly immersed in liquid nitrogen and stored until retrieval for experimental purposes.
Cell Lines and Cell Culture
We purified primary human endometrial stromal cells and obtained immortalized endometrial stromal stem cells (hEM15A) from patients with EM from ATCC. Both cell lines were cultivated in DMEM medium (Gibco, USA), supplemented with 10% fetal bovine serum, and maintained in a CO2 incubator at 37°C. Before use, all cell lines underwent rigorous testing and confirmation to ensure they were free from mycoplasma contamination. For the main instruments and consumables involved in the experiment, please refer to online supplementary Table S3, S4 (for all online suppl. material, see https://doi.org/10.1159/000543000).
Quantitative Real-Time Polymerase Chain Reaction
Following the guidelines, total RNA extraction from plasma of EM patients and normal individuals was performed using TRIzol LS reagent. Additionally, total RNA extraction from the tissues and cells of EM patients was conducted using TRIzol reagent. The extracted total RNA underwent reverse transcription into complementary DNA (cDNA) using a reverse transcription kit, followed by quantitative PCR with NovoStart® SYBR qPCR SuperMix Plus. The cDNA and genomic DNA (gDNA) PCR products were visualized using 2% agarose gel electrophoresis. The expression levels of circ_0008927 and PROM2 were normalized to the reference gene GAPDH. For primer sequences, please refer to online supplementary Table S1.
Cell Transfection
The specific small interfering RNA targeting circ_0008927 (si-circ_0008927) and a nonspecific control (si-NC), as well as the overexpression vector of circ_0008927 (PCD5-circ_0008927) and its control (PCD5), were obtained from Qingke Company. Transfection of the respective vectors or siRNAs into the cells was performed using Lipofectamine 2000 (Invitrogen, USA), and subsequent studies were conducted 24 h after incubation. For the RNA oligonucleotide sequences, please refer to online supplementary Table S2.
Actinomycin D (ActD)
Prepare a stock solution of ActD by dissolving it in an appropriate solvent according to the manufacturer’s guidelines. Create a working solution of 2 mg/mL in DMEM. Replace the regular growth medium with the prepared streptozotocin D-containing medium and incubate the cells for the specified time intervals. At designated time points, harvest the cells for RNA extraction. Subsequently, conduct reverse transcription using a kit to convert RNA into cDNA. Use quantitative real-time polymerase chain reaction (qRT-PCR) to determine the expression levels of circ_0008927 and ATG7 mRNA. Design primers specific to circ_0008927 and ATG7 and then in conjunction with a fluorescent dye or probe tailored for qRT-PCR. Normalize the expression levels to an internal control gene (e.g., GAPDH) to account for variations in RNA input and reverse transcription efficiency.
CCK-8 Assay
Seed well-cultured cells in a 96-well plate at a density of 3 × 103 cells per well, with three replicates each assigned to experimental group. Incubate the cells under standard culture conditions (37°C, 5% CO2) for specified time intervals (e.g., 0, 24, 48, 72, and 96 h). At the designated time points, add 10 μL of prepared CCK-8 solution to each well. Incubate the cells with the CCK-8 solution for an additional 3 h. After the incubation period, the absorbance was measured at a wavelength of 450 nm using an enzyme-linked microplate reader, and the data were analyzed to assess cell proliferation by comparing the absorbance values across different groups and time points.
EdU Assay
Cells cultured in vitro were seeded into a 96-well plate. EdU reagent was added, and the cells were incubated for 4 h to allow for absorption. Subsequently, 4% paraformaldehyde was added to fix the cells. A staining solution containing Azide 555 and Hoechst 33342 was prepared according to the instructions and evenly applied to each well, followed by a 30-min incubation. Cell images were captured using a fluorescence inverted microscope with the appropriate filters. The images were analyzed to calculate the percentage of EdU-positive cells, which is the ratio of the number of EdU-incorporated cells to the total number of Hoechst 33342-stained nuclei.
Wound Healing
Seed the cells in a 6-well plate and culture them until they reach approximately 90% confluence. Use a 200 µL pipette tip to create uniform scratches on the cell monolayer and capture microscopic images to document the initial width. Subsequently, aspirate the culture medium and add serum-free medium containing mitomycin C, then incubate for 48 h to promote cell migration. Capture images of the scratches again to record the healing progress. Analyze the images using ImageJ software to measure the width of the scratches before and after incubation, and calculate the percentage of wound healing.
Transwell
In the Transwell chamber, transfected cells were seeded at a density of 4 × 104 cells per chamber and placed in a 24-well plate. Overall, 200 µL of serum-free medium was added to the upper chamber, while 750 µL of DMEM containing 10% fetal bovine serum was added to the lower chamber. The chambers were incubated for 48 h to allow for cell migration. After incubation, non-migrated cells in the upper chamber were carefully removed using a cotton swab. Subsequently, 4% paraformaldehyde was added to fix the migrated cells, and the chambers were incubated at room temperature for 20 min. The cells were then stained with 0.1% crystal violet solution for 15–30 min. Images of the migrated cells were captured under a microscope, and representative areas were selected to document migration. Finally, the images were analyzed to quantify the number of migrated cells and assess the extent of migration.
RNA Immunoprecipitation Assay
Transfect hEM15A and HEK293T cells with miR-608 or miR-NC. After the incubation period, collect and lyse the cells using RNA immunoprecipitation (RIP) lysis buffer with ribonuclease and protease inhibitors, incubating on ice. Immunoprecipitate RNA using magnetic beads with either control IgG (negative control) or anti-human AGO2 antibody (positive control). Extract RNA from the immunoprecipitated complexes and reverse transcribe it into cDNA. Perform qRT-PCR with primers specific for circ_0008927, using a fluorescent dye or probe suitable for the assay. Normalize the qRT-PCR data with an internal control gene to account for variations. Compare circ_0008927 enrichment levels between control IgG and anti-AGO2 immunoprecipitations to assess its association with AGO2 and interaction with miR-608. Conduct statistical analysis to evaluate the significance of the findings.
Western Blotting
Determine protein concentrations using a BCA protein assay kit after extracting and lysing cells or tissues with RIPA lysis buffer containing 1% PMSF. Load equal amounts of protein onto a 10% SDS-PAGE gel and transfer to PVDF membranes. Incubate the membranes overnight at 4°C with primary antibodies against E-cadherin, N-cadherin, PROM2, GPX4, and SLC7A11. Block the membranes with skim milk; then incubate with secondary antibodies for 1 h at room temperature after washing. Visualize chemiluminescent signals using an ECL chemiluminescence kit. This method allows for detection and quantification of specific proteins and their relative abundance through a chemiluminescence detection system.
Dual-Luciferase Reporter
Construct dual-luciferase reporter gene vectors to study the interactions between circ_0008927 and miR-608, generating four vectors: circ_0008927-WT (wild type), circ_0008927-MUT (mutant), PROM2-WT, and PROM2-MUT. Transfect circ_0008927-WT and circ_0008927-MUT into cells with miR-608 mimic or miR-NC; similarly, transfect PROM2-WT and PROM2-MUT with miR-608 mimic or miR-NC. After a 48-h incubation, measure luciferase activity using a dual-luciferase reporter assay kit. Normalize firefly luciferase activity to Renilla luciferase activity for comparison across different groups. This assay elucidates the regulatory relationships between circ_0008927, miR-608, and PROM2, with experimental parameters adjusted as necessary.
Measurement of Fe2+ Levels
Treat cells according to kit instructions and measure absorbance at 593 nm using a spectrophotometer, including a blank control for background adjustment. Use the obtained absorbance values to calculate Fe2+ concentration in cells based on a standard curve provided by the assay kit. Perform data analysis to determine and compare Fe2+ concentrations under different experimental conditions.
Measurement of MDA Levels
Analyze 106 stained cells and fully lyse them to release cellular components, including MDA. Centrifuge the lysed samples to separate the supernatant. Prepare the working solution according to the MDA detection kit instructions and incubate it with the supernatant. Measure absorbance at 532 nm using a spectrophotometer. Calculate the MDA content based on the obtained absorbance values compared to a standard curve of known MDA concentrations.
Measurement of GSH Levels
Use a glutathione detection kit to treat transfected EM cells according to the instructions. Measure absorbance at 412 nm using a spectrophotometer to determine the intracellular concentration of GSH.
Cellular ROS Determination
Use an ROS detection kit to measure cellular lipid ROS levels according to the instructions. Add the DCFH-DA fluorescent probe to transfected, well-grown cells at a final concentration of 10 μm and incubate for 20 min in a culture incubator. Measure cell fluorescence using a flow cytometer.
Transcriptome Sequencing
Collect hEM15A cells overexpressing miR-608 mimics and the control group (mimics NC) for total RNA extraction following the provided instructions. Conduct transcriptional sequencing and analysis through OE Biotech (Shanghai, China) to identify differentially expressed genes, applying the criteria of |log2 FC| ≥ 1 and p value <0.05.
Statistical Analysis
Statistical analysis was performed using GraphPad Prism 7. The research results are presented as mean ± SD. A p value of less than 0.05 is considered statistically significant when each experiment is conducted at least 3 times.
Results
The Level of Circ_0008927 Is Elevated in EM Patients
Previous studies utilizing circRNA microarray analyses have identified circ_0008927 as an upregulated circRNA [17], however, its specific function remains largely unexplored. It is posited that circ_0008927 plays a crucial regulatory role in EM. To validate the aberrant expression of circ_0008927 in EM, we assessed its expression levels in endometrial tissues from 14 EM patients and 22 controls with normal endometrium samples using qRT-PCR. The results indicated a notable increase in circ_0008927 levels in ectopic endometrial tissues compared to normal controls (Fig. 1a). Furthermore, circ_0008927 expression was significantly upregulated in the eutopic endometrial stromal cell line (hEM15A) compared to the normal human endometrial stromal cell line (Fig. 1b).
Expression and characteristics of circ_0008927 in EM. a Relative expression levels of circ_0008927 in tissues from 14 EM patients and 22 normal tissues as detected by qRT-PCR. b Relative expression levels of circ_0008927 in HESCs and hEM15A cells as detected by qRT-PCR. c Genomic location of circ_0008927. Circ_0008927 is formed by reverse splicing of exons 13–17 of ATG7, and the splicing junction was validated by Sanger sequencing. d PCR of gDNA and cDNA using divergent and convergent primers. e Relative expression levels of circ_0008927 and ATG7 in cells after treatment with ActD. f PCR amplification of gDNA and cDNA was performed using divergent and convergent primers after RNase R treatment. *p < 0.05, **p < 0.01, ***p < 0.001.
Expression and characteristics of circ_0008927 in EM. a Relative expression levels of circ_0008927 in tissues from 14 EM patients and 22 normal tissues as detected by qRT-PCR. b Relative expression levels of circ_0008927 in HESCs and hEM15A cells as detected by qRT-PCR. c Genomic location of circ_0008927. Circ_0008927 is formed by reverse splicing of exons 13–17 of ATG7, and the splicing junction was validated by Sanger sequencing. d PCR of gDNA and cDNA using divergent and convergent primers. e Relative expression levels of circ_0008927 and ATG7 in cells after treatment with ActD. f PCR amplification of gDNA and cDNA was performed using divergent and convergent primers after RNase R treatment. *p < 0.05, **p < 0.01, ***p < 0.001.
Circ_0008927 is derived from exons 13 to 17 of the autophagy related 7 (ATG7) gene located on human chromosome 3 (chr3), with a length of 672 nt (Fig. 1c). To ascertain that the circular structure of circ_0008927 is formed via back-splicing of pre-mRNA, we designed both divergent and convergent primers for amplifying circ_0008927 using cDNA extracted from hEM15A cells and gDNA as additional templates. The results indicated that circ_0008927 was exclusively amplified using divergent primers in conjunction with cDNA and not with gDNA (Fig. 1d). Additionally, the stability of circ_0008927 was assessed further. qRT-PCR results demonstrated a significantly extended half-life of circ_0008927 compared to linear ATG7 following treatment with the transcription inhibitor ActD (Fig. 1e).
Circ_0008927 Promotes the Proliferation, Migration, and Invasion of EM Cells
To investigate the biological effects of circ_0008927 in EM, we utilized siRNA to downregulate the expression of circ_0008927 in hEM15A cells and constructed an overexpression vector to enhance circ_0008927 expression in these cells. qRT-PCR analysis demonstrated that si-circ_0008927-1 (si-circ#1) and si-circ_0008927-2 (si-circ#2) reduced circ_0008927 levels (Fig. 2a). The overexpression vector PCD5-circ_0008927 (PCD5-circ) markedly increased circ_0008927 expression (Fig. 2b). Cell Counting Kit-8 (CCK-8) and EDU assays revealed that knockdown of circ_0008927 significantly inhibited the proliferation of hEM15A cells, while the overexpression of circ_0008927 produced the contrary effect (Fig. 2c, d). Wound healing and Transwell assays indicated that circ_0008927 knockdown significantly inhibited the migration capabilities of hEM15A cells, whereas circ_0008927 overexpression significantly notably enhanced these migration abilities (Fig. 2e, f). Furthermore, Western blot analysis showed that circ_0008927 knockdown significantly suppressed decreased N-cadherin protein expression, while E-cadherin expression was significantly increased. Conversely, circ_0008927 overexpression exhibited the opposite trend (Fig. 2g). Collectively, these experiments suggest that circ_0008927 plays a crucial role in promoting the proliferation, migration, and invasion of EM cells.
Circ_0008927 promotes cell proliferation, migration, and invasion in vitro. a Expression of circ_0008927 in hEM15A cells transfected with siRNA as detected by qRT-PCR. b Expression of circ_0008927 in hEM15A cells transfected with PCD5-circ_0008927 as detected by qRT-PCR. c, d Evaluation of proliferation ability of hEM15A cells with circ_0008927 overexpression or knockdown using CCK-8 and EDU assays. e, f Evaluation of migration and invasion ability of hEM15A cells with circ_0008927 overexpression or knockdown using wound healing and Transwell assays. g Expression of E-cadherin and N-cadherin proteins in hEM15A cells with circ_0008927 overexpression or knockdown as detected by Western blot. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Circ_0008927 promotes cell proliferation, migration, and invasion in vitro. a Expression of circ_0008927 in hEM15A cells transfected with siRNA as detected by qRT-PCR. b Expression of circ_0008927 in hEM15A cells transfected with PCD5-circ_0008927 as detected by qRT-PCR. c, d Evaluation of proliferation ability of hEM15A cells with circ_0008927 overexpression or knockdown using CCK-8 and EDU assays. e, f Evaluation of migration and invasion ability of hEM15A cells with circ_0008927 overexpression or knockdown using wound healing and Transwell assays. g Expression of E-cadherin and N-cadherin proteins in hEM15A cells with circ_0008927 overexpression or knockdown as detected by Western blot. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Circ_0008927 Negatively Regulates miR-608 in EM Cells
According to the competitive endogenous RNA (ceRNA) theory, circ_0008927 can interact with miRNA and modulate its expression level in cells [18]. Consequently, we predicted the potential binding miRNAs of circ_0008927 using bioinformatics resources. By integrating multiple datasets, we identified three candidate miRNAs (Fig. 3a). We examined the expression levels of these candidate miRNAs in hEM15A cells transfected with either si-circ_0008927 or an overexpressing vector for circ_0008927. Only miR-608 exhibited increased expression in the si-circ_0008927 group, while a decrease was noted in the circ_0008927 overexpression group (Fig. 3b). To determine the interaction between circ_0008927 and miR-608, we co-transfected luciferase reporter vectors (miRGLO) containing either circ_0008927-WT or circ_0008927-MUT along with miR-608 mimic or mimic NC into hEM15A and 293T cells (Fig. 3c). Compared to the mimic NC group, luciferase activity was significantly reduced in the circ_0008927-WT group, whereas no notable difference was observed in circ_0008927-MUT (Fig. 3d). Additionally, we conducted RIP using an Argonaute RISC catalytic component 2 (AGO2) antibody to validate the binding of circ_0008927 to miR-608. The results indicated that more circ_0008927 was enriched with miR-608 in the AGO2 group compared to the control IgG group (Fig. 3e). These data indicate that miR-608 is a target of circ_0008927. Furthermore, we used qRT-PCR to detect the expression levels of miR-608 in the 14 EM patients and 22 normal endometrial samples. The results showed that compared to the normal control group, the level of miR-608 in ectopic endometrial tissues was significantly decreased (Fig. 3f). Simultaneously, Spearman analysis indicated a negative correlation between circ_0008927 and miR-608 (Fig. 3g).
Circ_0008927 acts as a sponge for miR-608. a Prediction of potential miRNAs binding to circ_0008927 using miRanda, RNAhybrid, and TargetScan databases. b Expression levels of miR-6845-5p, miR-6762-5p, and miR-608 in hEM15A cells with circ_0008927 overexpression or knockdown as detected by qRT-PCR. c Prediction of binding sites between miR-608 and circ_0008927 using the CircInteractome database. d Luciferase reporter gene assay revealing the binding between miR-608 and circ_0008927. e Co-immunoprecipitation of IgG group and Ago2 group, followed by qRT-PCR detection of miR-608 and circ_0008927 expression levels. f Relative expression levels of miR-608 in tissues from 14 EM patients and 22 normal tissues as detected by qRT-PCR. g The Spearman analysis revealed a negative correlation between circ_0008927 and miR-608. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Circ_0008927 acts as a sponge for miR-608. a Prediction of potential miRNAs binding to circ_0008927 using miRanda, RNAhybrid, and TargetScan databases. b Expression levels of miR-6845-5p, miR-6762-5p, and miR-608 in hEM15A cells with circ_0008927 overexpression or knockdown as detected by qRT-PCR. c Prediction of binding sites between miR-608 and circ_0008927 using the CircInteractome database. d Luciferase reporter gene assay revealing the binding between miR-608 and circ_0008927. e Co-immunoprecipitation of IgG group and Ago2 group, followed by qRT-PCR detection of miR-608 and circ_0008927 expression levels. f Relative expression levels of miR-608 in tissues from 14 EM patients and 22 normal tissues as detected by qRT-PCR. g The Spearman analysis revealed a negative correlation between circ_0008927 and miR-608. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Circ_0008927 Promotes Proliferation, Migration, and Invasion of hEM15A Cells through miR-608
To further elucidate the role of circ_0008927 in promoting proliferation, migration, and invasion of hEM15A through miR-608, a series of rescue experiments were conducted. CCK-8, EDU, wound healing, and Transwell assays were performed to assess whether the biological effects of circ_0008927 could be reversed by miR-608 mimics (Fig. 4). The results showed that miR-608 mimics effectively reversed the promotive effects of circ_0008927 overexpression on proliferation, migration, and invasion. These results confirm that circ_0008927 can act as a sponge for miR-608, promoting the progression of EM.
The promoting effect of circ_0008927 in EM can be reversed by miR-608. Co-transfection of PCD5+miR-NC, PCD5+miR-608 mimics, and circ_0008927+miR-608 mimics in cells. a, c Evaluation of cell proliferation ability using CCK-8 and EDU assays. b, d Evaluation of cell migration and invasion ability in EM cells using wound healing and Transwell assays. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
The promoting effect of circ_0008927 in EM can be reversed by miR-608. Co-transfection of PCD5+miR-NC, PCD5+miR-608 mimics, and circ_0008927+miR-608 mimics in cells. a, c Evaluation of cell proliferation ability using CCK-8 and EDU assays. b, d Evaluation of cell migration and invasion ability in EM cells using wound healing and Transwell assays. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
miR-608 Directly Targets PROM2 in EM
miRNA can bind to the 3′-untranslated region of target genes [19]. We transfected miR-608 mimic and miRNA-NC control into hEM15A cells and assessed gene expression changes through transcriptome sequencing. The results showed significant alterations in the expression of PROM2, caspase recruitment domain family member 16 (CARD16), MYC binding protein associated protein (MYCBPAP), phospholipid scramblase 2 (PLSCR2), and left-right determination factor 1 (LEFTY1) (Fig. 5a). We examined the expression levels of differentially expressed genes in hEM15A cells transfected with either miR-608 mimic or si-circ_0008927, finding that only PROM2 exhibited significant downregulation (Fig. 5b). Western blot analysis further revealed that miR-608 mimic reduced the protein level of PROM2 in hEM15A cells, while miR-608 inhibitor had the opposite effect (Fig. 5c). Dual-luciferase reporter gene assay provided further evidence of the direct binding between PROM2 and miR-608. The luciferase reporter vectors, miRGLO containing PROM2-WT or PROM2-MUT, were co-transfected with miR-608 mimic or mimic NC into hEM15A and 293T cells to determine the interaction between PROM2 and miR-608. Compared to the mimic NC group, luciferase activity was significantly reduced in the PROM2-WT group, while no notable difference was observed in PROM2-MUT (Fig. 5d, e). These data indicate that PROM2 is a target of miR-608.
PROM2 is a downstream target gene of miR-608. a Differential gene clustering heat map. b Expression levels of PROM2, CARD16, MYCBPAP, PLSCR2, and LEFTY1 in cells transfected with miR-608 mimics or si-circ_0008927 as detected by qRT-PCR. c Expression level of PROM2 protein in cells transfected with miR-608 mimics and miR-608 inhibitor as detected by Western blot. d Prediction of binding sites between miR-608 and PROM2 using the TargetScan database. e Luciferase reporter assay revealing the binding between miR-608 and PROM2. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01.
PROM2 is a downstream target gene of miR-608. a Differential gene clustering heat map. b Expression levels of PROM2, CARD16, MYCBPAP, PLSCR2, and LEFTY1 in cells transfected with miR-608 mimics or si-circ_0008927 as detected by qRT-PCR. c Expression level of PROM2 protein in cells transfected with miR-608 mimics and miR-608 inhibitor as detected by Western blot. d Prediction of binding sites between miR-608 and PROM2 using the TargetScan database. e Luciferase reporter assay revealing the binding between miR-608 and PROM2. Data represent the average ± SD from three independent experiments. *p < 0.05, **p < 0.01.
Overexpression of PROM2 Rescued the Silenced Circ_0008927-Mediated Inhibition of Proliferation, Migration, Invasion, and Ferroptosis in EM Cells in vitro
To elucidate the molecular mechanisms of circ_0008927/miR-608/PROM2 axis in EM, we conducted a series of rescue experiments. CCK-8, EDU, wound healing, and transwell assays were assess to determine whether the biological functions of circ_0008927 could be reversed by PROM2. The results showed that OE-PROM2 could reverse the proliferation and migration inhibition caused by circ_0008927 knockdown in hEM15As (Fig. 6a–d). These results confirmed that circ_0008927 could act as a sponge for miR-608 to regulate PROM2 expression, thereby promoting the progression of EM.
Overexpression of PROM2 rescued the silenced circ_0008927-mediated inhibition of proliferation, migration, invasion, and ferroptosis in hEM15A cells in vitro. After co-transfection of miR-NC+PCD5, si-circ_0008927+PCD5, and si-circ_0008927+PROM2 in cells, CCK8 and EDU assays were performed to evaluate cell proliferation ability (a, d). b, c Wound healing and Transwell assays were conducted to evaluate migration and invasion ability of hEM15A cells. e Cellular Fe2+ levels were measured using an iron assay kit. f Cellular MDA content was detected using the lipid peroxidation assay. g Cellular GSH levels were determined using the glutathione assay kit. h, i Lipid ROS levels were assessed using the DCFH-DA probe, and fluorescence analysis was performed using flow cytometry. j Expression levels of PROM2, GPX4, and SLC7A11 were detected by Western blot. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Overexpression of PROM2 rescued the silenced circ_0008927-mediated inhibition of proliferation, migration, invasion, and ferroptosis in hEM15A cells in vitro. After co-transfection of miR-NC+PCD5, si-circ_0008927+PCD5, and si-circ_0008927+PROM2 in cells, CCK8 and EDU assays were performed to evaluate cell proliferation ability (a, d). b, c Wound healing and Transwell assays were conducted to evaluate migration and invasion ability of hEM15A cells. e Cellular Fe2+ levels were measured using an iron assay kit. f Cellular MDA content was detected using the lipid peroxidation assay. g Cellular GSH levels were determined using the glutathione assay kit. h, i Lipid ROS levels were assessed using the DCFH-DA probe, and fluorescence analysis was performed using flow cytometry. j Expression levels of PROM2, GPX4, and SLC7A11 were detected by Western blot. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Previous studies have highlighted the critical role of PROM2 in regulating iron export mediated by ferritin, which can influence cellular sensitivity to ferroptosis. Therefore, we further investigated whether circ_0008927 mediates ferroptosis in hEM15A cells through the miR-608/PROM2 axis. A series of experiments were conducted, revealing that si-circ_0008927 inhibited cell viability in erastin-treated hEM15A cells, while PROM2 overexpression could rescue this phenotype. Additionally, knockdown of si-circ_0008927 increased the levels of Fe2+, MDA, and ROS, while decreasing glutathione (GSH) levels; however, PROM2 overexpression reversed these effects in hEM15A cells (Fig. 6e–i). Western blot analysis showed that knockdown of circ_0008927 decreased the expression of PROM2, glutathione peroxidase 4 (GPX4), and solute carrier family 7 member 11 (SLC7A11) in hEM15A cells. Notably, the concomitant overexpression of PROM2 reversed the ferroptosis effect result from circ_0008927 knockdown, indicating that circ_0008927 induces PROM2 to inhibit ferroptosis in hEM15A cells (Fig. 6j).
Circ_0008927 Inhibits Ferroptosis in hEM15A Cells through the miR-608/PROM2 Axis
To further verify whether circ_0008927 mediates ferroptosis in hEM15A cells via the miR-608/PROM2 axis, we conducted a series of rescue experiments. The results showed that overexpression of miR-608 can increase the levels of Fe2+, MDA, and ROS, while simultaneously decreasing the levels of GSH. In contrast, overexpression of circ_0008927 reverses those effects in hEM15A cells (Fig. 7a–d). Western blot analysis showed that overexpression of miR-608 decreases the expression of PROM2, GPX4, and SLC7A11 in hEM15A, while overexpression of circ_0008927 reverses the ferroptosis-promoting effect of overexpressing miR-608, which indicates that circ_0008927 inhibits the ferroptosis of hEM15A by inducing miR-608/PROM2 (Fig. 7e). In addition, we analyzed the expression of PROM2, GPX4, and SLC7A11 in ectopic endometrial tissues and normal controls; the results showed that the expression of PROM2, GPX4, and SLC7A11 was upregulated in ectopic endometrial specimens (Fig. 7f–h). Furthermore, correlation analysis showed that the levels of PROM2, GPX4, and SLC7A11 in endometrial ectopic tissue are positively correlated with circ_0008927(Fig. 7i–k).
Circ_0008927 inhibits ferroptosis in hEM15A cells through the miR-608/PROM2 axis. After co-transfection of PCD5-NC+miR-NC, PCD5-NC+miR mimics, and PCD5_circ+miR mimics in cells, cellular Fe2+ levels were measured using an iron assay kit (a). b Cellular MDA content was detected using the lipid peroxidation assay. c Cellular GSH levels were determined using the glutathione assay kit. d Lipid ROS levels were assessed using the DCFH-DA probe. e Expression levels of PROM2, GPX4, and SLC7A11 were detected by Western blot. f–h Relative expression levels of PROM2, GPX4, and SLC7A11 in tissues from 14 EM patients and 22 normal tissues as detected by qRT-PCR. i–k The Spearman analysis indicated that circ_0008927 is positively correlated with PROM2, GPX4, and SLC7A11. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Circ_0008927 inhibits ferroptosis in hEM15A cells through the miR-608/PROM2 axis. After co-transfection of PCD5-NC+miR-NC, PCD5-NC+miR mimics, and PCD5_circ+miR mimics in cells, cellular Fe2+ levels were measured using an iron assay kit (a). b Cellular MDA content was detected using the lipid peroxidation assay. c Cellular GSH levels were determined using the glutathione assay kit. d Lipid ROS levels were assessed using the DCFH-DA probe. e Expression levels of PROM2, GPX4, and SLC7A11 were detected by Western blot. f–h Relative expression levels of PROM2, GPX4, and SLC7A11 in tissues from 14 EM patients and 22 normal tissues as detected by qRT-PCR. i–k The Spearman analysis indicated that circ_0008927 is positively correlated with PROM2, GPX4, and SLC7A11. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.
Discussion
Numerous studies have demonstrated that circRNAs play significant roles in the occurrence and progression of human diseases, highlighting their remarkable potential in disease diagnosis and treatment [20]. However, the functional roles of only a limited number of circRNAs in EM have been elucidated. Xu et al. [21] first reported the aberrant expression profiles of circRNAs in normal endometrial and EM tissues and identified circ_0002198 and circ_0004712 as potential biomarkers for diagnosing EM patients [22, 23]. Moreover, it was found that circATRNL1 expression was significantly upregulated in ectopic tissues compared to eutopic tissues. Overexpression of circATRNL1 promoted proliferation, migration, and invasion of Ishikawa cells in vitro and facilitated the progression of EM through the miR-141-3p/miR-200a-3p-YAP1 axis [24]. In this study, by comparing endometrial specimens from EM patients and healthy individuals, we found that circ_0008927 was significantly upregulated in EM tissues. Furthermore, overexpression of circ_0008927 enhanced cell proliferation, migration, and invasion, while knockdown of circ_0008927 showed the opposite effects. Therefore, we speculate that circ_0008927 may act as an oncogenic inducer in EM. Further research is needed to elucidate the molecular mechanisms by which circ_0008927 regulates EM.
The most well-documented function of circRNAs is their ability to competitively bind to miRNAs, thereby inhibiting the interaction of miRNAs with their downstream target genes [25]. This mechanism is widely recognized as playing a crucial role in diseases. For example, circ_001422 functions as a ceRNA for miR-195-5p, accelerating the onset and metastasis of osteosarcoma by modulating the FGF2/PI3K/Akt axis [26]. In our study, we aimed to elucidate whether circ_0008927 regulates the progression of EM through the ceRNA pathway. Through bioinformatics analysis, we identified potential target miRNAs that could bind to circ_0008927, including miR-143-3p, miR-145-3p, miR-205-5p, miR-654-3p, and miR-3688-3p. Preliminary validation indicated that the expression of miR-608 exhibited the most significant downregulation upon circ_0008927 overexpression. This suggests that miR-608 may be the best promising candidate for further investigation. Luciferase reporter assays and RIP assays confirmed the direct binding relationship between circ_0008927 and miR-608. Furthermore, rescue experiments demonstrated that the promotion of cell proliferation, migration, and invasion by circ_0008927 was mediated by miR-608. We conclude that circ_0008927 directly binds to miR-608, promoting the progression of EM.
MiRNAs regulate various biological processes in EM, including proliferation, migration, invasion, apoptosis, ferroptosis, and angiogenesis. For instance, miR-608 may be upregulated in EM and target ADAM metallopeptidase with thrombospondin type 1 motif 7 (ADAMTS7), leading to cell decidualization [27]. Additionally, miR-608 inhibits proliferation and cell cycle progression in bladder cancer cells via the AKT/FOXO3a signaling pathway [21]. In our study, transcriptome sequencing and bioinformatics prediction identified PROM2 as a potential target of miR-608. Subsequent luciferase reporter assays and RIP verification confirmed that miR-608 directly targets the 3′-untranslated region of PROM2. Moreover, overexpression of miR-608 significantly reduced levels of PROM2 mRNA and protein. PROM2 has been reported to play a crucial regulatory role in iron export mediated by ferritin, which influences the sensitivity of tumor cells to ferroptosis [28, 29]. In breast cancer, PROM2 promotes the formation of iron-containing multivesicular bodies [30]. The release of multivesicular bodies as exosomes decreases cellular iron concentration, contributing to evasion of ferroptosis. Consistent with previous findings, our study observed that the expression of PROM2, GPX4, and SLC7A11 is upregulated in endometrial ectopic specimens, and knockdown of PROM2 promotes cell ferroptosis. Furthermore, correlation analysis showed that the levels of PROM2, GPX4, and SLC7A11 in endometrial ectopic tissue are negatively correlated with miR-608 and positively correlated with circ_0008927. Rescue experiments showed that circ_0008927 regulates the proliferation, migration, invasion, and ferroptosis of cells via the miR-608/PROM2 axis. In conclusion, these results suggest that circ_0008927 functions as a ceRNA by binding to miR-608, promoting the expression of PROM2, thereby advancing the progression of EM.
In conclusion, our novel findings suggest that circ_0008927 promotes cell proliferation, migration, and invasion through the miR-608/PROM2 axis in EM, thereby facilitating disease progression and inhibiting ferroptosis. Current research indicates that circ_0008927 is a promising molecule involved in EM development and may serve as a potential therapeutic target. However, this study has several limitations. First, we included only a small number of 14 EM patients, which may limit the generalizability of the findings. Second, we validated the function of circ_0008927 in vitro, without conducting in vivo experiments, which may impact the credibility of our conclusions. Therefore, further validation in larger cohorts encompassing different EM phenotypes, along with in vivo studies, is necessary to confirm the clinical relevance of circ_0008927 in EM.
Statement of Ethics
Informed consent forms were signed by all participating patients. The research protocol received approval from the Clinical Research Ethics Committee of the Women’s and Children’s Hospital in Ningbo City (EC2022-028). All procedural steps adhered to relevant laws and regulations.
Conflict of Interest Statement
The authors declare that they have no competing interests.
Funding Sources
This research is funded by funds of the Zhejiang Province Medical and Healthcare Technology Program Project (No. 2021PY071; 2022KY1165), the Ningbo Natural Science Foundation (No. 2022J240; No. 2022S035), Medical Clinical Research Center (No. 2024L002), the Ningbo Featured Disciplines (2018PPXK-06), and the Ningbo Top Medical and Health Research Program (No.2022020405).
Author Contributions
Yanping Shen: data curation, methodology, software, and original draft preparation. Lingli Li and Yingyan Yan: data curation and original draft preparation. Yun Chen: resources, investigation, and validation. Yanan Xu: supervision, reviewing and editing, and funding acquisition.
Additional Information
Yanping Shen, Yun Chen, and Yanan Xu contributed equally to this work.
Data Availability Statement
All data generated or analyzed during this study are included in this published article and its online supplementary information files. Further inquiries can be directed to the corresponding author.