Circular RNA circMYBL2 regulates the progression of ovarian cancer through miR-195-5P/BIRC5 axis

Abstract

Ovarian cancer has emerged as the most malignant gynecological tumor, primarily due to the challenges associated with early diagnosis. The majority of patients are diagnosed at an advanced stage, and the disease exhibits a high recurrence rate within three years following comprehensive treatment. Consequently, there is an urgent need to identify reliable prognostic indicators and therapeutic targets. Circular RNA (circRNA), characterized by its stable circular structure, demonstrates greater stability than mRNA within cells. It plays a significant role in regulating the initiation, progression, and chemotherapy resistance of various tumors, making it a promising candidate for novel therapeutic targets. However, a comprehensive understanding of the expression and functional mechanisms of circRNA in ovarian cancer remains elusive. In this study, we identified that circMYBL2 (hsa_circ_0060467) was significantly upregulated in ovarian cancer tissues and cell lines. Functionally, circMYBL2 was found to promote the proliferation, invasion, and migration of OVCAR3 and CAOV3 cells in vitro. In vivo experiments demonstrated that knockdown of circMYBL2 inhibited the growth of CAOV3 cells, an effect that was reversed by the overexpression of BIRC5. Mechanistically, circMYBL2 functions as a molecular sponge for miR-195-5p, upregulating BIRC5 upon binding to miR-195-5p, thereby facilitating the proliferation, invasion, and migration of ovarian cancer cells. These findings indicate that circMYBL2 contributes to ovarian cancer progression through the circMYBL2/miR-195-5p/BIRC5 axis. As such, circMYBL2 may serve as a novel diagnostic marker and a potential therapeutic target for ovarian cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13048-025-01946-2.

Introduction

Ovarian cancer (OC), one of the most common malignant tumors in the female reproductive system, is characterized by complex histopathology, high malignancy, frequent recurrence, and poor prognosis, making it the most lethal gynecological malignancy [1, 2]. In 2020 alone, China reported 55,342 new cases of ovarian cancer and 37,519 related deaths [3]. The absence of noticeable symptoms in the early stages often leads to delayed diagnosis. By the time patients present with apparent symptoms such as abdominal pain, distension, or palpable lower abdominal masses, the disease has typically progressed to an advanced stage. The standard treatment for advanced ovarian cancer involves a combination of surgery and chemotherapy. However, the propensity of advanced ovarian cancer for metastasis and invasion often renders complete surgical resection challenging. Despite comprehensive treatment, approximately 70% of patients experience relapse within three years due to chemotherapy resistance, resulting in a five-year survival rate of only about 40% [4].

Circular RNA (circRNA) is a unique class of non-coding RNA molecules transcribed by RNA polymerase II. Distinguished from linear RNAs, circRNAs lack both 5’ cap structures and 3’ poly(A) tails, instead forming a covalently closed continuous loop that exists naturally in living organisms. Based on their genomic origins, circRNAs are primarily classified into three categories: exonic circRNAs (ecRNAs) derived from one or more exons, exon-intron circRNAs (EIciRNAs) containing both exonic and intronic sequences, and circular intronic RNAs (ciRNAs) originating solely from introns. The unique circular structure confers remarkable stability to circRNAs, making them resistant to RNA exonuclease degradation and resulting in a significantly longer half-life compared to mRNAs in most species [5, 6]. However, circRNA stability in serum is compromised due to the presence of RNase A family ribonucleases. Notably, circRNA expression exhibits remarkable specificity across species, tissues, cells, and developmental stages, with most circRNA sequences demonstrating high evolutionary conservation [7]. These distinctive characteristics position circRNAs as promising biomarkers for ovarian cancer detection and monitoring.

This study revealed that circMYBL2 was significantly overexpressed in both ovarian cancer tissues and cell lines. Functional assays demonstrated that circMYBL2 markedly enhanced the proliferation, invasion, and migration capabilities of CAOV3 and OVCAR3 cells in vitro. Mechanistic investigations further elucidated that circMYBL2 functions as a molecular sponge for miR-195-5p, thereby regulating the downstream expression of BIRC5. These findings suggest that circMYBL2 may serve as an oncogenic regulator promoting ovarian cancer progression by facilitating tumor cell proliferation, invasion, and metastasis. Consequently, circMYBL2 emerges as a promising predictive biomarker and potential therapeutic target for ovarian cancer development and progression.

Materials and methods

Cell culture

The human ovarian cancer cell lines CAOV3, OVCAR3, A2780, and SKOV3, along with the normal human ovarian epithelial cell line IOSE80, were obtained from Fenghui Biological Company, Nanjing Kebai Biotechnology Company, and Changsha Zebra Biological Co., Ltd., respectively. Cells were maintained in appropriate culture media supplemented with 5% fetal bovine serum: McCoy’s 5 A medium for SKOV3 cells, RPMI-1640 medium for A2780 and IOSE80 cells, and DMEM medium for CAOV3 and OVCAR3 cells. All cell lines were cultured under standard conditions at 37 °C in a humidified atmosphere containing 5% CO₂.

Collection of clinical samples

This study was conducted with the approval of the Ethics Committee of the Second Affiliated Hospital of Hainan Medical University (Ethics Approval Number: HYLL-2024-707). A total of 20 paired samples of ovarian cancer tissues and adjacent normal tissues were collected from patients undergoing surgical treatment at our institution between September 2021 and February 2024. All cases were histopathologically confirmed as epithelial ovarian malignancies.Inclusion Criteria:1. Histopathological confirmation of epithelial ovarian malignancy. 2. No prior chemotherapy, radiotherapy, immunotherapy, or other anticancer treatments before surgery. 3. Absence of other primary malignant tumors. 4. Written informed consent obtained from patients or their legal representatives prior to tissue collection. Exclusion Criteria:1. Presence of other primary malignant tumors. 2. Diagnosis of metastatic ovarian cancer. 3. Comorbidities with severe hematological, immunological, or other systemic diseases.

Cell transfection

Ovarian cancer cell lines CAOV3 and OVCAR3 were selected for subsequent transfection experiments. Based on the sequences of circMYBL2 and BIRC5, we constructed both overexpression lentiviral vectors (CircMYBL2 and Lv-BIRC5) and knockdown lentiviral vectors (Sh-circMYBL2 and Sh-BIRC5). Corresponding negative control vectors (Vector, Lv-BIRC5-NC, Sh-circMYBL2-NC, and Sh-BIRC5-NC) were also prepared for parallel transfection. Following optimization through preliminary experiments, cells were transfected with an appropriate titer of viral particles supplemented with polybrene as an infection enhancer. After 16 h of incubation, the viral-containing medium was replaced with complete growth medium, and the cells were maintained for an additional 72 h to ensure stable transfection.

Western-blot

Cells were lysed using RIPA buffer (Beyotime Biotechnology, Shanghai, China) supplemented with protease inhibitor cocktail. Total protein extracts were separated by 10% SDS-PAGE and subsequently transferred onto PVDF membranes (Millipore, Billerica, MA, USA). After blocking with 5% non-fat milk in TBST for 1 h at room temperature, membranes were incubated overnight at 4 °C with primary antibodies against BIRC5 (1:1000 dilution; Abcam, Cambridge, UK). Following three washes with TBST, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system (Bio-Rad, Hercules, CA, USA), and images were captured using a chemiluminescence imaging system.

RNA extraction and quantitative real-time PCR (qRT‒PCR)

Total RNA was extracted from tissues and cells using TRIzol reagent (Takara Bio, Shiga, Japan) following the manufacturer’s protocol. All primers used in this study were synthesized by Sangon Biotech (Shanghai, China), with GAPDH serving as an internal control. Detailed primer sequences are provided in Additional file: Table S1. Quantitative real-time PCR was performed using the Tiangen SuperReal PreMix Plus (SYBR Green) kit (Tiangen Biotech, Beijing, China). All reactions were performed in triplicate. Relative gene expression levels were calculated using the 2-ΔΔCt method.

RNase R treated

The RNase R digestion reaction was prepared in a sterile microcentrifuge tube with the following components: 2 µL of 10× RNase R Reaction Buffer, 1 µg of RNA sample, 2–4 U of RNase R enzyme (20 U/µL), and DEPC-treated H2O to a final volume of 20 µL. The reaction mixture was incubated at 37 °C for 20 min to allow for complete digestion, followed by enzyme inactivation at 70 °C for 10 min.

PCR

PCR amplification was performed using the PrimeSTAR^® Max DNA Polymerase kit (Takara Bio, Shiga, Japan) according to the manufacturer’s protocol. The PCR reaction mixture was prepared as follows: 25 µL of PrimeSTAR Max Premix (2×), 1.5 µL each of forward and reverse primers, template DNA (< 200 ng), and nuclease-free water to a final volume of 50 µL. Reactions were carried out in 100 µL sterile PCR tubes using a thermal cycler under the following conditions: initial denaturation at 98 °C for 10 s, followed by 30 cycles of denaturation at 98 °C for 10 s, annealing at 55 °C for 10 s, and extension at 72 °C for 5 s. A final extension step was performed at 68 °C for 30 s.

Dual luciferase reporter assay

To validate the predicted miR-195-5p binding sites, dual luciferase reporter vectors containing either wild-type (WT) or mutant (Mut) circMYBL2 fragments and BIRC5 3’UTR sequences were commercially constructed in psiCHECK2 backbone (Geneseed Biotech, Guangzhou, China). The complete vector sequences are documented in Additional files: Table S2. For functional assays, cells were seeded in 12-well plates at 60% confluency. Using Lipofectamine 3000 transfection reagent (Thermo Fisher Scientific), cells were co-transfected with: (1) WT or Mut reporter plasmids (2) miR-195-5p mimics or negative control (NC) oligonucleotides. Following 48-hour incubation at 37 °C with 5% CO₂, luciferase activities were quantified using the Dual-Luciferase^® Reporter Assay System (Cat# FR201-02; TransGen Biotech, Beijing, China) according to the manufacturer’s protocol.

Fluorescencein situhybridization (FISH) assays

To determine the subcellular localization of circMYBL2, fluorescence in situ hybridization (FISH) was performed according to standard protocols. Briefly, fixed cell preparations were subjected to pre-hybridization in blocking buffer at 55°C for 2 hours. Subsequently, hybridization was carried out overnight at 37°C using a digoxigenin (DIG)-labeled locked nucleic acid (LNA) probe specific to circMYBL2 (5’-DIG-CTGGGGTTACAAACAAGAGAGAG-DIG-3’; 20 nM; Geneseed Biotech, Guangzhou, China). Following three stringent washes in SSC buffer (2×, 1×, 0.5×), nuclei were counterstained with DAPI (1 µg/mL; Sigma-Aldrich) for 10 min at room temperature.

RNA immmuoprecipitation(RIP)

RNA immunoprecipitation (RIP) assays were performed using the Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (Geneseed Biotech, Guangzhou, China) according to the manufacturer’s protocol with modifications. Briefly, CAOV3 and OVCAR3 cells (5 × 10⁶ cells per sample) were lysed in RIP buffer supplemented with RNase inhibitor (40 U/mL; Thermo Fisher Scientific) and protease inhibitor cocktail (1×; Roche). Cell lysates were incubated overnight at 4 °C with Dynabeads™ Protein G (Thermo Fisher Scientific) pre-conjugated with either anti-Ago2 antibody (1:100; ab32381, Abcam) or normal mouse IgG (1:100; sc-2025, Santa Cruz Biotechnology) as negative control. Following immunoprecipitation, RNA-protein complexes were washed six times with RIP wash buffer and eluted. The co-precipitated circMYBL2 was quantified by qRT-PCR using PrimeScript™ RT reagent Kit (Takara) and SYBR Green Master Mix (Roche) on a QuantStudio 6 Flex system (Applied Biosystems). Relative enrichment was calculated using the 2^(-ΔΔCt) method normalized to input controls.

Cell counting kit-8(CCK8)

Cell proliferation was assessed using the Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Japan) according to the manufacturer’s protocol with modifications. Lentivirus-transduced stable cell lines were trypsinized and resuspended in complete growth medium. Cells were seeded at a density of 2 × 10⁴ cells/mL (100 µL/well) in 96-well flat-bottom plates (Corning) and allowed to adhere for 6 h at 37 °C in a humidified 5% CO₂ incubator. Following attachment, the medium was replaced with fresh complete medium containing 10% CCK-8 reagent. At 24-hour intervals (days 0–3), absorbance at 450 nm was measured using a SpectraMax M5 microplate reader (Molecular Devices) with a reference wavelength of 650 nm. Three independent experiments were performed with six technical replicates each. Growth curves were generated by plotting the mean absorbance values against time using GraphPad Prism 9.0 software.

Conoly formation

Colony formation assays were performed to assess cellular proliferative capacity. Cells were seeded at a density of 1,000 cells per well in 6-well plates (Corning) with three biological replicates. Each well contained 2 mL of complete growth medium (DMEM supplemented with 10% FBS and 1% penicillin-streptomycin). Plates were gently swirled to ensure even cell distribution and incubated at 37 °C in a humidified 5% CO₂ atmosphere. The medium was replaced every 72 h with fresh complete medium. After 14 days of culture, cells were washed once with phosphate-buffered saline (PBS; pH 7.4) and fixed with 4% paraformaldehyde (1 mL/well) for 30 min at room temperature. Following three PBS washes (5 min each), cells were stained with 1 mL of 0.5% crystal violet solution (Sigma-Aldrich) for 15 min at room temperature protected from light. Excess stain was removed by extensive PBS washing until the background was clear. Colony numbers were quantified using ImageJ software (v1.53) .

Wound-healing assays

The cells were seeded in 6-well plates and allowed to grow to confluence. Three evenly spaced scratches were made perpendicular to the horizontal axis using a 200 µL pipette tip. The wells were then gently washed two to three times with phosphate-buffered saline (PBS) to remove detached cells. Subsequently, serum-free medium was added to each well, and the initial scratch wounds were observed and photographed using an inverted microscope. The plates were transferred to a humidified incubator maintained at 37 °C with 5% CO2. After 24 h of incubation, the cells were re-examined and photographed. The migration distance was measured by calculating the difference in scratch width between 0 h and 24 h, and the results were presented as a bar graph.

EdU assays

The experimental procedures were conducted using the EdU assay kit (C10310-1, Ribo Biotech) following the manufacturer’s instructions. The maximum excitation wavelength was measured at 550 nm, with a corresponding maximum emission wavelength of 565 nm. Additionally, two channels were utilized, exhibiting a maximum excitation wavelength of 350 nm and a maximum emission wavelength of 461 nm.

Immunohistochemistry

The resected tumors from the xenograft mouse models were processed by fixation in 10% formaldehyde in PBS for 24 h, followed by preservation in 70% ethanol and subsequent paraffin embedding. For immunohistochemical detection, the following primary antibody was used: anti-BIRC5 (10508-1-AP, SURVIVIN antibody, 1:400 dilution) purchased from Proteintech (Wuhan Sanying, China). The air-dried sections were evaluated under a microscope to assess the survivin staining intensity in the subcutaneous tumor tissues of nude mice from both the control group and the sh-circMYBL2 expression group.

Animal experiments

All animal experiments were conducted in compliance with national guidelines and were approved by the Animal Research Committee of Hainan Medical University (Approval No. SYXK2022-011). Female BALB/c nude mice (4–6 weeks old) were obtained from Hainan Fen Wei Biological Technology Co., Ltd. To establish subcutaneous xenograft models, 1 × 10^7 CAOV3 cells were implanted into the axillary region of the mice (n = 5–6 per group). Following tumor cell inoculation, Inh-NC or miR-195-5p was administered via subcutaneous injection around the tumor area on days 3, 7, 10, 14, and 17, respectively, for a total of five injections.Five experimental groups were designed as follows: (1) Sh-circMYBL2-NC + Lv-BIRC5-NC + Inh-NC, (2) Sh-circMYBL2 + Lv-BIRC5-NC + Inh-NC, (3) Sh-circMYBL2 + Lv-BIRC5 + Inh-NC, (4) Sh-circMYBL2 + Lv-BIRC5-NC + Inh-miR-195-5p, (5) Sh-circMYBL2-NC + Lv-BIRC5 + Inh-NC. Tumor dimensions and body weights were monitored every five days throughout the experimental period. After 30 days, the mice were humanely euthanized, and the tumors were excised and weighed for further analysis.

HE staining of lung tissues in nude mice

CAOV3 cells with stable sh-circMYBL2 overexpression or control CAOV3 cells (1 × 10⁶ cells per mouse) were injected into nude mice via the tail vein, with 5 mice in each group. The animals were euthanized 28 days after cell injection. Lung tissues were collected, fixed, embedded, and sectioned. Sections were subjected to HE staining following the same dehydration steps as those used for immunohistochemistry. Specifically, the sections were stained with hematoxylin for 2 min and eosin for 5 min. Subsequently, the sections were sequentially immersed in a graded series of alcohol and xylene, followed by mounting with a coverslip. After air-drying, the number of lung metastatic foci in nude mice from the control and sh-circMYBL2 expression groups was quantified under a microscope.

Statistical analysis

Statistical analyses were performed using SPSS 23.0 software, while all images were processed and analyzed with GraphPad Prism 8.0. Data are expressed as mean ± standard deviation (SD). For comparisons between two independent samples, Student’s t-test was employed, whereas one-way analysis of variance (ANOVA) was utilized for multiple group comparisons. A p-value of less than 0.05 (p < 0.05) was considered statistically significant. All experiments were conducted in triplicate to ensure reproducibility.

Results

The characteristic of circMYBL2 in ovarian cancer tissues and cell lines

To investigate the expression profile of circRNAs in ovarian cancer (OC), we conducted full transcriptome sequencing, which revealed significant variations in circRNA expression levels between tumor tissues and normal ovarian tissues. Differential expression patterns were visualized through heatmaps (Fig. 1A) and volcano plots (Fig. 1B). A total of 232 circRNAs were significantly upregulated (indicated in red), with circMYBL2 being the most prominent, exhibiting a -Log10Padj value of 27.11182051, while 203 circRNAs were downregulated (indicated in green) (fold change > 1.5, p-value < 0.05). Subsequently, we selected the top 10 upregulated circRNAs for validation using RT-qPCR. Among these, the expression level of circMYBL2 was significantly upregulated in 40 ovarian cancer tissues compared to normal ovarian tissues. This consistent trend was also observed in ovarian cancer cell lines, particularly with more pronounced expression in OVCAR3 and CAOV3 cells. (Fig. 1C-D).Analysis of clinical specimens revealed significantly higher circMYBL2 expression in FIGO stage III/IV tissues compared to stage I/II tissues (Fig. 1E).Ovarian cancer patients with elevated circMYBL2 levels exhibited significantly reduced overall survival(Fig. 1F).CircMYBL2, a 993-nucleotide transcript, is generated through backsplicing of the pre-mRNA of the cell cycle checkpoint gene MYBL2, involving exons 10–14, as annotated in the CircBank database (Fig. 1G). The backspliced junction of circMYBL2 was successfully amplified using divergent primers and further validated by Sanger sequencing (Fig. 1H). Notably, circMYBL2 in cDNA could be amplified by both divergent and convergent primers (Fig. 1I). To further confirm the circular nature of circMYBL2, we assessed its stability using RNase R treatment. The results demonstrated that, in contrast to linear MYBL2, circMYBL2 exhibited resistance to RNase R digestion (Fig. 1J). The RNase R tolerance index indicates that circMYBL2 has an index value greater than 0.8, demonstrating circular characteristics(Fig. 1K). Finally, we determined the subcellular localization of circMYBL2 using fluorescence in situ hybridization (FISH), which revealed that circMYBL2 is predominantly localized in the cytoplasm (Fig. 1L).

Fig. 1.

Verification and characterization of circMYBL2 in ovarian cancer cells and clinical samples: (A) Heatmap and (B) volcano plot illustrating differentially expressed genes in three paired ovarian cancer and adjacent normal tissues. Downregulated genes are indicated in green, while upregulated genes are marked in red. The arrow highlights the position of circMYBL2. (C) Expression levels of circMYBL2 in ovarian cancer tissues compared to normal tissues. (D) Expression levels of circMYBL2 in ovarian cancer cell lines. (E) RT-qPCR was performed to detect circMYBL2 expression levels in ovarian cancer tissues stratified by FIGO stage (I/II vs. III/IV). (F) Kaplan-Meier Survival curves of patients with ovarian cancer showing low and high expression of circMYBL2. (G) Validation of the circular structure of circMYBL2. (H) Schematic representation of Sanger sequencing results confirming the backspliced junction of circMYBL2. (I) Agarose gel electrophoresis demonstrating the expression of circMYBL2 in CAOV3 cells. (J) RT-qPCR analysis of circMYBL2, linear MYBL2, and GAPDH following RNase R treatment. Compared to linear MYBL2, circMYBL2 exhibited resistance to RNase R digestion (n = 3). (K) RNase R Tolerance index : Expression level of circRNA in the RNase R treatment group / Expression level of circRNA in the untreated group(n = 3). (L) Subcellular localization of circMYBL2 detected by FISH assay. *P < 0.05, ***P < 0.001

Overexpression of circMYBL2 significantly enhanced the proliferation, migration and invasion ability of ovarian cancer cells

To investigate the functional role of circMYBL2, we designed and constructed lentiviral vectors for overexpression and transfected them into CAOV3 and OVCAR3 cells. The overexpression efficiency of circMYBL2 was confirmed by RT-qPCR, which revealed a significant increase in circMYBL2 expression levels in both CAOV3 and OVCAR3 cells compared to the control group (Fig. 2A). To further explore the impact of circMYBL2 on ovarian cancer cells, we conducted a series of functional assays, including the CCK-8 assay, colony formation assay, and EdU cell proliferation assay, to evaluate its effect on cell proliferation (Fig. 2B-D). The results demonstrated that overexpression of circMYBL2 significantly enhanced the proliferative capacity of both CAOV3 and OVCAR3 ovarian cancer cell lines. Additionally, the migratory ability of ovarian cancer cells was assessed using wound healing and Transwell migration assays (without Matrigel) (Fig. 2E-F). Overexpression of circMYBL2 markedly increased the migratory potential of CAOV3 and OVCAR3 cells. Furthermore, the invasive ability was evaluated using a Transwell invasion assay (with Matrigel), which revealed that circMYBL2 overexpression significantly enhanced the invasiveness of both cell lines (Fig. 2G). Collectively, these findings indicate that circMYBL2 functions as an oncogene in ovarian cancer by promoting cellular proliferation, migration, and invasion.

Fig. 2.

Effects of circMYBL2 overexpression on the biological behavior of ovarian cancer cells: (A) Overexpression efficiency of circMYBL2 in CAOV3 and OVCAR3 cells following lentiviral transfection was confirmed by RT-qPCR (n = 3). (B) CCK-8 assay, (C) colony formation assay, and (D) EdU assay were performed to evaluate the impact of circMYBL2 overexpression on cell proliferation in CAOV3 and OVCAR3 cells. (E) Wound healing assay was conducted to assess the migratory ability of circMYBL2-overexpressing cells. (F) Transwell migration assay (without Matrigel) was used to further evaluate the migration ability of circMYBL2-overexpressing cells. (G) Transwell invasion assay (with Matrigel) was employed to determine the invasive ability of circMYBL2-overexpressing cells. **P < 0.01, ***P < 0.001

Knockdown of circMYBL2 significantly inhibited the proliferation, migration and invasion of ovarian cancer cells

Similarly, we designed and constructed a lentiviral vector to knock down circMYBL2. The knockdown efficiency of sh-circMYBL2 in CAOV3 and OVCAR3 cells was confirmed by RT-qPCR, which showed a significant reduction in circMYBL2 expression levels(Fig. 3A). However, the expression of linear MYBL2 was not affected(Fig. 3B). To further investigate the functional consequences of circMYBL2 knockdown, we performed a series of assays, including the CCK-8 assay, colony formation assay, and EdU cell proliferation assay. The results demonstrated that knockdown of circMYBL2 significantly inhibited the proliferative capacity of ovarian cancer cells (Fig. 3C-E). In addition, the migratory ability of ovarian cancer cells was assessed using wound healing and Transwell migration assays (without Matrigel). The findings revealed that circMYBL2 knockdown markedly suppressed the migration of CAOV3 and OVCAR3 cells (Fig. 3F-G). Furthermore, the invasive ability was evaluated using a Transwell invasion assay (with Matrigel), which confirmed that circMYBL2 knockdown significantly reduced the invasiveness of ovarian cancer cells (Fig. 3H). In summary, these results demonstrate that knockdown of circMYBL2 in ovarian cancer inhibits cell proliferation, migration, and invasion.

Fig. 3.

Effects of circMYBL2 knockdown on the biological behavior of ovarian cancer cells: (A) Knockdown efficiency of circMYBL2 in CAOV3 and OVCAR3 cells following lentiviral transfection was confirmed by RT-qPCR (n = 3). (B)The expression level of MYBL2 was detected by RT-qPCR. (C) CCK-8 assay, (D) colony formation assay, and (E) EdU assay were performed to evaluate the impact of circMYBL2 knockdown on cell proliferation in CAOV3 and OVCAR3 cells. (F) Wound healing assay was conducted to assess the effect of circMYBL2 knockdown on cell migration. (G) Transwell migration assay (without Matrigel) was used to further evaluate the migratory ability of circMYBL2-knockdown cells. (H) Transwell invasion assay (with Matrigel) was employed to determine the effect of circMYBL2 knockdown on cell invasion. **P < 0.01, ***P < 0.001

CircMYBL2 directly binds to miR-195-5p

According to the competing endogenous RNA (ceRNA) hypothesis, circRNAs are known to function as miRNA sponges, regulating miRNA activity and the expression of downstream target genes. Given the abundance and cytoplasmic stability of circMYBL2, we investigated whether circMYBL2 promotes the biological behavior of ovarian cancer (OC) by sponging miRNAs. To this end, we utilized miRanda and online databases, including StarBase and CircInteractome and identified potential binding sites between circMYBL2 and miR-195-5p (Fig. 4A). To validate the interaction between circMYBL2 and miR-195-5p, we performed a dual-luciferase reporter assay. Wild-type and mutant luciferase reporter plasmids of circMYBL2 (circMYBL2-wt and circMYBL2-mut) were constructed, along with miR-195-5p mimic and its negative control (miR-195-5p mimic NC). The results demonstrated that the luciferase activity of the circMYBL2-wt group was significantly reduced upon transfection with the miR-195-5p mimic compared to the control group. In contrast, no significant difference in luciferase activity was observed between the miR-195-5p mimic and NC groups in the circMYBL2-mut group (Fig. 4B). These results confirmed that circMYBL2 directly binds to miR-195-5p. It is well established that miRNAs regulate target gene expression by interacting with Argonaute 2 (Ago2), a core component of the RNA-induced silencing complex (RISC). To further validate this interaction, we performed RNA immunoprecipitation (RIP) using an AGO2 antibody. The results demonstrated that significantly more circMYBL2 was pulled down in CAOV3 and OVCAR3 cells compared to the IgG control (Fig. 4C). Additionally, we analyzed the expression of miR-195-5p in ovarian cancer (OC) tissues and observed that miR-195-5p levels were significantly downregulated in OC tissues compared to normal tissues (Fig. 4D). Furthermore, miR-195-5p expression was reduced in ovarian cancer cells overexpressing circMYBL2, whereas its expression was increased in cells with circMYBL2 knockdown (Fig. 4E). Consistent with these findings, miR-195-5p expression was lower in OC cell lines (SKOV3, A2780, OVCAR3, CAOV3) compared to the normal ovarian epithelial cell line IOSE80 (Fig. 4F). Collectively, these findings indicate that circMYBL2 functions as a miR-195-5p sponge in ovarian cancer.

Fig. 4.

BIRC5 is regulated by miR-195-5P and circMYBL2: (A) Prediction of the binding sites between miR-195-5p and circMYBL2. (B) Statistical analysis of the dual-luciferase reporter assay for circMYBL2 and miR-195-5P (n = 3). (C) Detection of circMYBL2 binding to Ago2 using RIP (n = 3). (D) RT-qPCR analysis of miR-195-5P expression levels in ovarian cancer tissues compared to adjacent normal tissues. (E) RT-qPCR analysis of miR-195-5P expression following circMYBL2 knockdown or overexpression in ovarian cancer cells. (F) Comparison of miR-195-5p expression levels between normal ovarian epithelial cell lines and ovarian cancer cell lines. (G-L) RT-qPCR analysis of miR-195-5p and PRS14, BIRC5, CBX4, CALR expression levels in CAOV3 and OVCAR3 cells transfected with miR-195-5p mimic or inhibitor. (M) Potential binding sites of miR-195-5p on BIRC5. (N) Statistical analysis of the dual-luciferase reporter assay for BIRC5 and miR-195-5P (n = 3). (O) RT-qPCR validation of BIRC5 overexpression efficiency in CAOV3 and OVCAR3 cells following lentiviral transfection (n = 3). (P) RT-qPCR validation of BIRC5 knockdown efficiency in CAOV3 and OVCAR3 cells following lentiviral transfection (n = 3). (Q-R) RT-qPCR and Western blot (WB) analysis of BIRC5 expression after circMYBL2 knockdown. (S-T) RT-qPCR and Western blot (WB) analysis of BIRC5 expression after circMYBL2 overexpression. * P < 0.05; ** P < 0.01; ***P < 0.001

CircMYBL2 functions as a miR-195-5p sponge to upregulate BIRC5 in OC

To identify potential target genes of miR-195-5p, we utilized the StarBase database to predict binding interactions between miR-195-5p and mRNAs such as PRS14, BIRC5, CBX4, and CALR. To validate the regulatory relationship between miR-195-5p and the predicted targets, miR-195-5p mimics and inhibitors were transfected into ovarian cancer cells (Fig. 4G).Functional validation of miR-195-5p interactions with candidate targets (PRS14, BIRC5, CBX4, CALR) was performed through transfection of miR-195-5p mimics or inhibitors into ovarian carcinoma cell lines. qPCR analysis revealed that BIRC5 alone showed a significant inverse correlation with miR-195-5p levels, with pronounced downregulation in mimic-transfected cells and marked upregulation following inhibitor treatment (Fig. 4H-L). BIRC5, a well-known anti-apoptotic gene, was selected for further investigation based on our previous findings that BIRC5 plays a critical role in ovarian cancer progression and drug resistance. Through integrated bioinformatic screening, potential binding sites for miR-195-5p within the 3’-UTR of BIRC5 were identified (Fig. 4M). Furthermore, we performed a dual-luciferase reporter assay to determine whether miR-195-5p directly binds to BIRC5. The results revealed that the luciferase activity of the reporter vector carrying the wild-type BIRC5 3’UTR (BIRC5 3’UTR-WT) was significantly reduced in the miR-195-5p mimic group and increased in the miR-195-5p inhibitor group. In contrast, no significant changes were observed in the luciferase activity of the mutant BIRC5 3’UTR (BIRC5 3’UTR-MUT) vector (Fig. 4N). We successfully generated stable BIRC5-overexpressing and BIRC5-knockdown ovarian cancer (OC) cell lines (Fig. 4O-P). Consistent with these findings, qRT-PCR and Western blot analyses revealed that circMYBL2 knockdown significantly suppressed BIRC5 expression at both the mRNA and protein levels (Fig. 4Q-R). Conversely, circMYBL2 overexpression led to a marked upregulation of BIRC5 expression (Fig. 4S-T). These results collectively indicate that BIRC5 expression is regulated by miR-195-5p and circMYBL2.

CircMYBL2/miR195-5p/BIRC5 axis regulated OC cell proliferation, migration and invasion

To determine whether circMYBL2 promotes malignant biological behaviors through the circMYBL2/miR-195-5p/BIRC5 axis, we performed rescue experiments using miR-195-5p mimics and inhibitors. To verify that circMYBL2 exerts its oncogenic effects via miR-195-5p, we conducted CCK-8, wound healing, Transwell (Matrigel), and colony formation assays following transfection with a miR-195-5p inhibitor to suppress its expression. We found that miR-195-5p knockdown significantly reversed the inhibitory effects of circMYBL2 knockdown on the proliferation, migration, and invasion of ovarian cancer cells (Fig. 5A-D). Rescue experiments consistently confirmed that overexpression of BIRC5 could counteract the inhibitory effects of miR-195-5p on the proliferation, invasion, and migration of ovarian cancer cells, demonstrating that miR-195-5p suppresses these malignant behaviors by targeting BIRC5(Fig. 5G-H).These results indicate that circMYBL2 regulates BIRC5 expression through miR-195-5p, thereby promoting the proliferation, migration, and invasion of ovarian cancer cells. Next, we investigated whether BIRC5 is essential for the oncogenic effects mediated by the circMYBL2/miR-195-5p axis. Our results demonstrated that BIRC5 knockdown markedly attenuated the enhanced proliferation, migration, and invasion of ovarian cancer cells induced by circMYBL2 overexpression (Fig. 5I-L). In summary, these findings reveal that circMYBL2 functions as a molecular sponge for miR-195-5p, upregulating BIRC5 expression and promoting ovarian cancer cell proliferation, migration, and invasion through a competing endogenous RNA (ceRNA) mechanism.

Fig. 5.

Biological effects of circMYBL2 regulation in ovarian cancer cells through the miR-195-5P/BIRC5 axis: (A) Rescue experiments were performed to evaluate the functional role of circMYBL2. (A) CCK-8 assay, (B) colony formation assay, (C) Transwell assay (with Matrigel), and (D) wound healing assay were conducted to demonstrate that miR-195-5P knockdown in CAOV3 cells rescues the inhibitory effects of circMYBL2 knockdown on ovarian cancer cell proliferation, migration, and invasion. (E) CCK-8 assay, (F) colony formation assay, (G) Transwell assay (with Matrigel), and (H) wound healing assay were performed to show that BIRC5 overexpression could counteract the inhibitory effects of miR-195-5p on ovarian cancer cell proliferation, migration, and invasion. (I) CCK-8 assay, (J) colony formation assay, (K) Transwell assay (with Matrigel), and (L) wound healing assay were performed to show that BIRC5 overexpression in CAOV3 cells counteracts the enhanced proliferation, migration, and invasion induced by circMYBL2 overexpression. **P < 0.01, ***P < 0.001

The CircMYBL2/miR195-5p/BIRC5 axis regulates the proliferation and metastasis of ovarian cancer in vivo

To evaluate the effects of circMYBL2 and BIRC5 on ovarian cancer proliferation in vivo, stably transformed CAOV3 cells were injected into BALB/c nude mice. Following tumor cell inoculation, Inh-NC or miR-195-5p was administered via subcutaneous injection around the tumor area on days 3, 7, 10, 14, and 17, respectively, for a total of five injections. The mice were divided into three groups: (1) Sh-circMYBL2-NC + Lv-BIRC5-NC + Inh-NC, (2) Sh-circMYBL2 + Lv-BIRC5-NC + Inh-NC, (3) Sh-circMYBL2 + Lv-BIRC5 + Inh-NC, (4) Sh-circMYBL2 + Lv-BIRC5-NC + Inh-miR-195-5p, (5) Sh-circMYBL2-NC + Lv-BIRC5 + Inh-NC (Fig. 6A-B). Each group consisted of five mice, and 1 × 10⁷ cells were injected per mouse. Thirty days after tumor model establishment, tumor weight and volume were measured (Fig. 6C-D). Compared to the control group, the circMYBL2 knockdown group exhibited significantly reduced tumor weight and volume. However, overexpression of the BIRC5 gene effectively rescued this suppression. Similarly, the miR-195-5p inhibitor was able to counteract this rescue effect. These findings demonstrate that circMYBL2 regulates ovarian cancer cell proliferation in vivo through the BIRC5 axis. Immunohistochemical analysis revealed that upregulation of circMYBL2 expression significantly increased survivin (BIRC5) levels in subcutaneous tumors of nude mice compared with the control group (Fig. 6E). To further analyze the impact of circMYBL2 on the metastasis of ovarian cancer cells in vivo, control group and circMYBL2-knockdown CAOV3 cells were injected into the circulation of nude mice via the tail vein, respectively. The number of lung metastatic foci in nude mice was assessed using HE staining. The results (Fig. 6F) demonstrated that compared with nude mice injected with control CAOV3 cells, those injected with circMYBL2-knockdown CAOV3 cells exhibited a significant reduction in the number of lung metastatic foci, indicating that downregulation of circMYBL2 expression inhibits the metastasis of ovarian cancer cells in vivo.

Fig. 6.

The effect of CircMYBL2 on the proliferation and metastasis ability of ovarian cancer cells in nude mice: (A) Representative images of nude mice and xenograft tumors from each experimental group. (B) Images of excised tumors from xenograft nude mice. (C) Tumor growth curves for each group. (D) Tumor weights measured for each group. (E) experiment was used to detect the expression of surviving in subcutaneous tumor tissues of nude mice. (F) HE Staining Analysis of the Effect of Downregulated CircMYBL2 Expression on Lung Metastasis in Nude Mice. * P < 0.05; ** P < 0.01

Discussion

Ovarian cancer remains a significant challenge in gynecological oncology due to its poor prognosis, characterized by a low 5-year survival rate, frequent development of chemotherapy resistance, lack of early symptoms, and absence of effective early detection and prevention strategies. Consequently, the identification of early diagnostic biomarkers and the development of targeted therapies have become critical areas of focus. Among emerging research topics, circular RNA (circRNA) has garnered considerable attention. A growing body of evidence suggests that circRNAs play pivotal roles in cancer initiation and progression through diverse functional mechanisms. Recent studies have demonstrated that circRNAs are involved in regulating key cellular processes, including proliferation, invasion, apoptosis, and chemotherapy resistance, in various malignancies such as ovarian cancer [8], cervical cancer [9], hepatocellular carcinoma [10], lung cancer [11], bladder cancer [12], and colorectal cancer [13].

In this study, we identified elevated expression of circMYBL2 in ovarian cancer tissues and cell lines. For the first time, we elucidated a novel mechanism through which the circMYBL2/miR-195-5P/BIRC5 signaling axis regulates ovarian cancer progression. Our findings demonstrated that upregulation of circMYBL2 significantly enhanced the proliferation, invasion, and migration of CAOV3 and OVCAR3 cells in vitro, as well as tumor growth in vivo. Conversely, downregulation of circMYBL2 suppressed these oncogenic behaviors. To the best of our knowledge, this is the first study to comprehensively investigate the expression and functional role of circMYBL2 in ovarian cancer. Our results suggest that circMYBL2 may serve as both a predictive biomarker and a potential therapeutic target for ovarian cancer.The oncogenic role of circMYBL2 is further supported by studies in other malignancies. For instance, Sun et al. reported that circMYBL2 promotes FLT3/FLT3-ITD protein translation via the circMYBL2/PTBP1 axis in FLT3-ITD acute myeloid leukemia (AML), highlighting its potential as a therapeutic target for FLT3-ITD-mutated AML subtypes [14]. Similarly, in cervical cancer, Wang et al. demonstrated that circMYBL2 facilitates tumor progression by modulating miR-361-3p expression, providing a novel therapeutic target for cervical cancer treatment [15]. These findings collectively underscore the critical role of circMYBL2 in cancer pathogenesis and its potential as a therapeutic target across multiple cancer types.Yi et al. demonstrated that the circMYBL2/miR-1205/E2F1 axis promotes hepatocellular carcinoma (HCC) progression. These findings collectively suggest that circMYBL2 may function as an oncogene and represent a promising therapeutic target for ovarian cancer [16]. CircRNAs are known to exert their biological functions through multiple mechanisms, including acting as miRNA sponges, directly interacting with mRNAs, binding to proteins, and regulating gene transcription and splicing. Among these mechanisms, the role of circRNAs as miRNA sponges has garnered significant research attention. By binding to mRNAs, circRNAs can induce gene silencing. Furthermore, circRNAs containing multiple miRNA binding sites function as competitive endogenous RNAs (ceRNAs). Through microRNA response elements (MREs), circRNAs can sequester miRNAs, preventing them from binding to their target mRNAs. This mechanism effectively abolishes the regulatory effects of miRNAs on their target genes, thereby modulating the expression of downstream genes [17].In this study, bioinformatics analysis and experimental validation confirmed that circMYBL2 contains a binding site for miR-195-5P, suggesting that circMYBL2 may act as a miR-195-5P sponge to promote the proliferation and migration of CAOV3 and OVCAR3 cells. Recent studies have also revealed that certain circRNAs possess protein-coding potential. For instance, Zhao et al. reported that circMYBL2 can function as a tumor suppressor by encoding the p185 protein, which reduces serine biosynthesis through enhanced degradation of PHGDH, thereby inhibiting colorectal cancer progression [18]. However, whether circMYBL2 exhibits protein-coding activity in ovarian cancer and how it influences ovarian cancer progression remain to be elucidated.

MicroRNAs (miRNAs) are a class of non-coding RNAs (ncRNAs), typically 20–25 nucleotides in length, that play a pivotal role in the post-transcriptional regulation of messenger RNA (mRNA). Among these, miR-195-5p is a miRNA derived from the 5’ arm of the precursor miR-195. The expression levels of miR-195-5p vary significantly across different types of tumors and are closely associated with key cancer-related processes, including proliferation, invasion, migration, and apoptosis. Notably, miR-195-5p has been implicated in nearly all major cancer types. It is generally downregulated in most cancers, such as non-small cell lung cancer [19], colorectal cancer [20], endometrial cancer [21], breast cancer [22], and hepatocellular carcinoma [23]. Conversely, it is upregulated in certain malignancies, including leukemia and adrenocortical carcinoma. In our study, BIRC5 was identified as a potential target gene of miR-195-5p, a prediction subsequently validated by dual-luciferase reporter assays. Furthermore, rescue experiments demonstrated that circMYBL2 facilitates the proliferation, invasion, and migration of CAOV3 and OVCAR3 cells via the circMYBL2/miR-195-5p/BIRC5 regulatory axis. However, it is noteworthy that existing research has shown miR-195-5p to be a pleiotropic molecule capable of targeting multiple key genes, including OTX1, YAP1, FGF2, and others. BIRC5 is not the sole target of miR-195-5p; it merely serves as an important and functionally relevant downstream mediator in tumor development.

BIRC5 (Baculoviral IAP Repeat Containing 5), also known as Survivin, is a member of the anti-apoptotic protein family. It is highly expressed in various types of tumors and early embryonic tissues but is minimally expressed in fully differentiated adult tissues. The overexpression of BIRC5 is closely associated with enhanced resistance to chemotherapy, reduced sensitivity to radiotherapy, tumor progression, and poor prognosis. Studies have demonstrated that BIRC5 is significantly upregulated in multiple cancers, including ovarian cancer, breast cancer [24], oral cancer [25], and prostate cancer [26]. In oral squamous cell carcinoma, BIRC5 expression exhibits a strong positive correlation with tumor differentiation, clinical stage, and lymph node metastasis. Moreover, high BIRC5 expression is linked to a decreased overall survival rate. Knockdown of BIRC5 has been shown to increase cellular sensitivity to cisplatin and 5-FU, further supporting its role in chemotherapy resistance and poor patient outcomes [25]. Additionally, miR-485-5p has been reported to inhibit breast cancer cell invasion and enhance the sensitivity of breast cancer cells to doxorubicin and paclitaxel by targeting the 3’-UTR of BIRC5²⁴. Furthermore, knockdown or inhibition of BIRC5 expression can enhance the sensitivity of rectal cancer and lung cancer to radiotherapy [27]. Notably, low-dose radiotherapy has been shown to reverse cisplatin resistance in ovarian cancer by downregulating BIRC5 expression [28]. Seon Min Woo et al. demonstrated that USP1 enhances the stability of BIRC5 through ubiquitination, thereby inhibiting ML323 and TRAIL-induced apoptosis [29]. Furthermore, combined inhibition of Mcl-1 and BIRC5 has been shown to significantly enhance etoposide-induced cytotoxicity and apoptosis in acute myeloid leukemia [30]. Notably, BIRC5 can be detected in the blood and body fluids of cancer patients, highlighting its potential as a novel biomarker. For instance, BIRC5 is detectable in the urine of bladder cancer patients, suggesting that this non-invasive detection method could serve as a promising approach for identifying both de novo and recurrent tumors [31]. Additionally, the expression levels of BIRC5 in malignant ascites are significantly higher than those in ascites caused by benign diseases. Among malignant ascites, ovarian cancer exhibits higher BIRC5 levels compared to pancreatic cancer, hepatocellular carcinoma, and gastric cancer. These findings collectively indicate that BIRC5 may serve as a potential diagnostic biomarker for malignant tumors.

In summary, our study demonstrated that circMYBL2 promotes the proliferation, invasion, and migration of CAOV3 and OVCAR3 cells in vitro. Mechanistically, circMYBL2 functions as a miR-195-5p sponge, thereby upregulating BIRC5 expression. Notably, circMYBL2 is significantly upregulated in ovarian cancer tissues. Collectively, our findings reveal that circMYBL2 facilitates the proliferation, invasion, and migration of CAOV3 and OVCAR3 cells through the circMYBL2/miR-195-5p/BIRC5 axis. These results provide a novel therapeutic target for the prediction and treatment of ovarian cancer. CircMYBL2, an oncogene highly expressed in ovarian cancer, holds significant potential as both a diagnostic biomarker and a therapeutic target. However, translating this basic research discovery into clinical practice remains challenging.

CRISPR/Cas13 is a system that targets RNA. It utilizes a guide RNA (crRNA) to locate and cleave specific mRNA or other RNA molecules. This process does not alter the DNA sequence of the gene itself but instead regulates gene expression by preventing its translation into protein. A crRNA can be designed to be perfectly complementary to the unique back-splice junction site of circMYBL2. This would allow the Cas13/crRNA complex to specifically recognize and degrade circMYBL2. However, although studies indicate that circMYBL2 primarily exerts its oncogenic function by sequestering miR-195-5p, other unknown mechanisms of action likely exist (such as interactions with proteins or regulation of transcription and translation). Indiscriminate targeting could fail due to compensatory network effects or even trigger unintended consequences.Furthermore, significant challenges remain, including how to deliver the system efficiently and stably in vivo, as well as concerns regarding off-target effects and overall safety.

Authors’ contributions

Binxin Liu, Yadan Fan, and Qingchun Deng co-authored the study. Binxin Liu, Yadan Fan, and Chunyan Lv performed the experiments. Yadan Fan and Chunyan Lv performed the statistical analysis. Miaomiao Cai collected the key background information. The manuscript was drafted by Binxin Liu and Chunyan Lv and examined by Yinglian Pan. Qingchun Deng and Zeng Gen contributed to study guidance. All authors read and approved the final manuscript.

Funding

This study was supported by the National Natural Science Foundation of China. Project Fund (No. 82360595), Natural Science Foundation of Hainan Province: 825QN545, the collabrative project of Health Science and Technology Innovation of Hainan province: WSJK2025ZD213.

Data availability

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All studies were approved by the Ethics Committee of the Second Affiliated Hospital of Hainan Medical University, and informed consent for clinical research was obtained from patients, with whom they signed the informed consent form. All animal experiments were approved by the Ethics Committee of Hainan Medical University, No. HYLL-2024-707.

Competing interests

The authors declare no competing interests.