1 Introduction

Ovarian cancer (OC) is the most deadly malignant tumor in women diagnosed with gynecological cancers [1]. In the United States and Europe, the incidence of OC is 11.7–12.1 cases per 100,000 people, while it is slightly lower in the Middle East and Asia [2]. More than 90% of malignant ovarian tumors originate from the epithelial tissue of ovary and/or fallopian tube. OC is divided into six types by the World Health Organization, including endometrial, mucinous, serous, clear cell, squamous cell and transitional cell carcinomas [3]. Due to its early spread, late detection and high recurrence rate in abdominal cavity, OC has a poor prognosis as well as a five-year survival rate less than 40% [4]. The treatment options of OC include platinum compound combined chemotherapy and cytoreductive surgery. Generally, OC is very sensitive to chemotherapy, but most of patients suffer from recurrent OC (ROC) after cytoreductive surgery and first-line chemotherapy [5].

ROC and its drug resistance to current chemotherapy schemes are global challenges. Notably, drug resistance is a complex phenomenon involving multiple genes and signaling pathways [6]. Therefore, it is of great importance to clarify the potential molecular mechanism of drug resistance, which is helpful for the management of decision-making for treatment and the identification of new effective drug targets. OC stem cells (OCSCs) have high self-renewal ability, play a very important role in tumor initiation, spread, metastasis and drug resistance, and are considered to be highly resistant to chemotherapy and the cause of high recurrence rate of OC [7, 8]. There are many surface molecular markers of OCSCs, such as CD24, CD44, CD117 and CD133 [8]. Among them, CD133, also known as prominin-1, is a glycoprotein with five transmembrane characteristics, which was first discovered in mouse neuroepithelial cells [9]. Also, a previous research has reported that CD133 exists in different adult stem cells and can inhibit cell differentiation [10]. Related studies show that the expression of CD133 is related to the chemosensitivity of OCSCs and can be used to monitor and predict the clinical outcome after chemotherapy. There are many factors leading to the tolerance of OC chemotherapy drugs [11]. The family members of ATP-binding cassette transporters (ATP-binding cassette transporters), including multidrug resistance protein 1 (MDR1), breast cancer resistance protein (BCRP) and MDR-associated protein 1 (MRP1), provide tumors with multiple anticancer drug resistance, such as paclitaxel, topoisomerase inhibitors and tyrosine kinase inhibitors [12]. Although traditional treatment can significantly reduce the tumor size and temporarily improve the symptoms of patients, it does not specifically and highly effectively target tumor subgroups, resulting in poor therapeutic effect of OC [13]. Therefore, it is of great significance to further study the drug resistance mechanism of OCSCs and explore new therapeutic targets.

RAD51-associated protein 1 (RAD51AP1) plays an important role in homologous recombination (HR) by activating RAD51 recombinase [14]. On the one hand, RAD51AP1 interacts with RAD51 or DMC1 recombinase to enhance their recombinase activity and stimulate the formation of D-loop [15]; and such process is the key step of HR-mediated DNA repair in mitosis and meiosis of cells. On the other hand, RAD51AP1 is also involved in promoting growth-related signal transduction. Moreover, some studies have shown that RAD51AP1 highly expresses in OC, cholangiocarcinoma and hepatocellular carcinoma, indicating its potential role in tumor proliferation [16,17,18]. Additionally, Bridges et al. proposed the important function of RAD51AP1 in tumor stem cells. Briefly, down-regulating RAD51AP1 inhibited the self-renewal and related potential of breast cancer stem cell (BCSC), and improved the efficacy of chemotherapy and radiotherapy [19]. Furthermore, Bridges also discovered that deletion of RAD51AP1 could weaken the renewal of colorectal cancer stem cell (CCSC) and enhance the sensitivity to chemotherapy [20]. However, the expression level and potential mechanism of RAD51AP1 in OCSCs are not clear. Therefore, the objective of this study was to expound the function of RAD51AP1 in OCSCs, further understand the occurrence and development mechanism of OC, and provide theoretical guidance for the development of new targeted drugs for OC.

2 Materials and methods

OC tumor and adjacent tissues were selected from patients receiving surgical resection in Hainan West Central Hospital from January 2023 to March 2023. The inclusion criteria were shown as follows: patients (1) were diagnosed as OC by histopathology; (2) had complete clinical and pathological data and follow-up information. The exclusion criteria were listed as follows: patients (1) were complicated with serious heart, brain, blood vessels and other important organ diseases; (2) had a history of tumors derived from other tissues; (3) had infection or pregnancy. This study was approved by the Ethics Committee of Hainan West Central Hospital (No. SH9H-2019-A681-1), and all patients signed written informed consent. All methods were carried out in accordance with relevant guidelines and regulations.

2.1 Cell culture

OVCAR4 (WARNER, WN-C10067) cells were cultured in RPMI1640 medium (Sigma, #51536 C) containing 10% fetal bovine serum (FBS, Gibco, #10099) and placed in a cell incubator containing 5% CO2 at 37 ℃.

2.2 CD133 positive (CD133+) cell sorting

OVCAR4 cells were prepared into single cell suspension, and the suspension density was adjusted to 1 × 107/mL by phosphate buffered saline (PBS). Next, CD133-APC antibody (Biolegend, #397906) was added according to the dilution ratio of 1:100. The mixture was incubated on ice in the dark for 30 min. Subsequently, the cells were washed twice with PBS, resuspended with 500 µL PBS and filtered with a 300-mesh filter. According to the expression level of CD133, the cells were sorted by flow cytometry (BD, LSRII). Finally, CD133 positive (CD133+) OVCAR4 and CD133 negative (CD133−) OVCAR4 cells were collected.

2.3 Plasmid construction and cell transfection

The RAD51AP1 gene fragments were amplified by KOD plus PCR (TOYOBO, # KOD-201) polymerase. Through one-step cloning kit (Vazyme, #C112-01), the obtained RAD51AP1 gene fragments were constructed into pcDNA3.1 to form plasmid pcDNA3.1-RAD51AP1. Next, the pcDNA3.1-RAD51AP1 plasmid formed was sent to BGI Genomics for sequencing and sequence comparison to ensure correctness. Subsequently, Lipofectamin 3000 (ThermoFisher, #L3000001) was used to transfect the constructed clone into CD133+OVCAR4 cells, and after transfection for 24 h, the subsequent experimental operation was carried out. GenePharma company was entrusted to synthesize the siRNA targeting RAD51AP1, with the sequence of 5′-AACCTCATATCTCTAATTGCA-3′. Later, siRNA was transfected into CD133+OVCAR4 cells by Lipofectamin 3000, and after transfection for 24 h, the subsequent tests were performed.

2.4 Sphere-formation assay

OVCAR4 cells in different groups were seeded into 6-well ultra-low adsorption plates (Corning, #3471) at a density of 2 × 104 cells/well using serum-free DMEM/F-12 (Gibco, #21331020) medium. The culture medium was supplemented with B-27 supplement (Gibco, #17504044, concentration: 1×), human epidermal growth factor (Sigma-Aldrich, # GF144, concentration: 10 ng/mL) and human beta fibroblast growth factor (Gibco, #13256-029, concentration: 20 ng/mL) [21]. The cells were cultured in an incubator for 7 days, then the images were collected under an optical microscope, and the number of spheres was statistically analyzed.

2.5 Cell colony forming assay

OVCAR4 cells in different groups were seeded into 6-well plates at a density of 2 × 104 cells/well. The cells were cultured at 37 ℃ and 5% CO2 to form colonies, and the culture medium was changed every five days. After 10 days, the medium was removed, and the cells were fixed with 10% (v/v) methanol for 15 min and then stained with 0.1% crystal violet (Sigma-Aldrich, #548-62-9) solution for 30 min. Subsequently, the cells were washed with PBS for three times, the images were collected under the optical microscope, and the number of cell colonies was recorded.

2.6 MTT assay

MTT assay kit (Beyotime, #C0009S) was used to detect the cell viability. Briefly, OVCAR4 cells in different groups were seeded into 96-well cell culture plates with 5,000 cells per well, and the culture volume was 100 µL. The cells were treated with different concentrations of paclitaxel (0, 1, 3, 7, 15 µg/mL), cisplatin (0, 0.5, 2, 5, 10 µg/mL) and doxorubicin (0, 1, 2.5, 5, 10 µg/mL) [22,23,24]. After 24 h, the culture medium was changed, and the cells were incubated with 10 µL MTT solution for 4 h. Subsequently, the liquid in each well was removed, and 100 µL Formazan solution was added and shaken for 10 min. Lastly, the optical density (OD) at the wavelength of 570 nm was measured by a microplate reader, and the cell viability and IC50 (half maximal inhibitory concentration) of each group were calculated.

2.7 RT-qPCR

Total RNA of cells in each group was extracted with Trizol (Invitrogen, #15596026) reagent. Then, the extracted RNA (1 µg) was reversely transcribed into cDNA with reverse transcription kit (Takara, #639549). Next, the qPCR system was configured using SYBR Green qPCR kit (Bio-Rad, #172–5270), and the reaction was carried out under 7500 Fast real-time fluorescence quantitative PCR system (ThermoFisher, #4351107). Through 2−ΔΔCt method, the relative expression of target genes was calculated taking GAPDH as an internal control. The primer sequences were displayed as follows: SOX2-F: 5′-AGCTCGCAGACCTACATGAA-3′, SOX2-R: 5′-CCGGGGAGATACATGCTGAT-3′; OCT4-F: 5′-CCCGAAAGAGAAAGCGAACC-3′, OCT4-R: 5′-GCAGCCTCAAAATCCTCTCG-3′; NANOG-F: 5′-ACCCAGCTGTGTGTACTCAA-3′, NANOG-R: 5′-CTGCGTCACACCATTGCTAT-3′; KLF4-F: 5′-AGAGACCGAGGAGTTCAACG-3′, KLF4-R: 5′-CGGATCGGATAGGTGAAGCT-3′: GAPDH-F: 5′-GTCTCCTCTGACTTCAACAGCG-3′, GAPDH-R: 5′-ACCACCCTGTTGCTGTAGCCAA-3′.

2.8 Western blot

The total protein of cells in each group was extracted by RIPA lysate (ThermoFisher, #89900), and the protein concentration was determined by BCA method (Vazyme, #E112-01/02). Next, the protein (20 µg) was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) for 120 min, and then transferred to the polyvinylidene fluoride (PVDF) membrane (ThermoFisher, #LC2005) by wet transfer. Subsequently, the PVDF membrane was sealed with 5% skimmed milk for 1 h and incubated with RAD51AP1 (NOVUS, #NBP2-20060), transforming growth factor beta 1 (TGF-β1, Abcam, #ab215715), SMAD4 (Abcam, #ab40759), and GAPDH (ThermoFisher, #MA1-16757) antibodies at 4 ℃ overnight. On the next day, the membrane was washed with TBST for three times and then incubated with anti-rabbit secondary antibody (CST, #14708) at ambient temperature for 1 h. Again, the membrane was washed with TBST for three times. Subsequently, the membrane was exposed and photographed with developers (Merck, #WBKLS0050) under a western blot imaging analyzer (BioRad, #12003153), and the exposure results were quantitatively analyzed with ImageJ software.

2.9 Statistical treatment

The experimental data were statistically analyzed by SPSS 23.0 and plotted by Graph Pad Prism 9.0. Two-sample t-test was used for comparison between two groups, and one-way ANOVA for comparison among multiple groups. All experimental results were expressed as the mean ± standard deviation. P < 0.05 indicated a significant difference.

3 Results

3.1 OVCAR4 cells contain CD133 positive (CD133+) cancer stem-like cells

Tumor stem cells, as a small number of cells with self-renewal ability and multi-directional differentiation potential, are considered to play an important role in the origin, development, metastasis and drug resistance of tumors [25]. CD133 is regarded as a tumor stem cell marker [26]. In this study, the level of CD133 in OVCAR4 was analyzed by flow cytometry, and the analysis results revealed that about 10.1% cells showed CD133+ characteristics (Fig. 1A). To study the biological characteristics of CD133+ cell population, OVCAR4 cells with CD133− and CD133+ were sorted by flow cytometry, and the self-renewal and proliferation of the sorted cells were further analyzed. The sphere-formation assay results showed that CD133+OVCAR4 cells had significantly higher sphere-formation capacity than CD133−OVCAR4 cells (Fig. 1B, P < 0.01). The cell colony forming assay showed that the colonies formed by CD133+OVCAR4 cells were much more than those formed by CD133−OVCAR4 cells (Fig. 1C, P < 0.01). The above results indicated that the OVCAR4 cell populations contained CD133+ cell populations, and CD133+OVCAR4 cells had a stronger self-renewal ability and proliferation activity. Therefore, CD133+OVCAR4 cells may play an important role in the development, metastasis and drug resistance of OC.

Fig. 1
figure 1

Evaluation of biological characteristics of CD133 positive (CD133 +) OVCAR4 cells. A Flow cytometry to analysis the expression level of CD133 in OVCAR4 cells; B evaluation of the sphere-formation capacity of CD133 positive (CD133+) OVCAR4 and CD133 negative (CD133−) OVCAR4 cells (left) and statistical analysis of the number of spheres (right); C assessment of the colony forming ability of CD 133+OVCAR4 and CD133−OVCAR4 cells (left) and statistical analysis of the number of cell colonies (right). **P < 0.01

Full size image

3.2 High expression of RAD51-associated protein 1 (RAD51AP1) in ovarian cancer (OC) tissues and CD133 positive (CD133+) OVCAR4 cells

To determine the association of RAD51AP1 with OC, the expression level of RAD51AP1 in OC tissues was detected using western blot. The detection results showed that the expression level of RAD51AP1 was significantly elevated in OC tissues compared with that in adjacent normal tissues (Fig. 2A, P < 0.01). In addition, the expression level of RAD51AP1 in CD133+OVCAR4 cells was significantly higher than that in CD133−OVCAR4 cells (Fig. 2B, P < 0.01). The above results indicated that RAD51AP1 may exert important regulatory functions in OCSCs, which in turn affected the progression of OC.

Fig. 2
figure 2

High expression of RAD51-associated protein 1 (RAD51AP1) in ovarian cancer (OC) tissues and CD133 positive (CD133 +) OVCAR4 cells. Western blot was applied to analyze the expression level of RAD51AP1. A, B Expression levels (top) and statistical analysis (bottom) of RAD51AP1 in OC tissues and adjacent tissues (A) as well as CD133+OVCAR4 and CD133−OVCAR4 cells (B). **P < 0.01. RAD51AP1: RAD51-associated protein 1; OC: ovarian cancer; CD133+: CD133 positive; CD133−: CD133 negative

Full size image

3.3 RAD51-associated protein 1 (RAD51AP1) promotes the self-renewal and proliferation of CD133 positive (CD133+) OVCAR4

To investigate the biological function of RAD51AP1 in OCSCs, RAD51AP1 in CD133+OVCAR4 cells was over-expressed by transfection with pcDNA3.1-RAD51AP1; also, the expression level of RAD51AP1 was knocked down by transfection with RAD51AP1 siRNA. The protein expression level of RAD51AP1 in CD133+OVCAR4 cells was verified using western blot, and the results showed that the protein expression level of RAD51AP1 in the RAD51AP1 group was significantly increased compared with that in the Vector group (Fig. 3A, P < 0.01); compared with the siNC group, the protein expression level of RAD51AP1 in the si-RAD51AP1 group was notably decreased (Fig. 3A, P < 0.01). The above outcomes indicated that the protein level of RAD51AP1 was successfully over-expression and knocked down in CD133+OVCAR4 cells. Additionally, the results of the cell colony forming assay showed that the number of colonies in the RAD51AP1 group was significantly increased compared with that in the Vector group; while the number of colonies in the si-RAD51AP1 group was notably decreased compared with that in the siNC group (Fig. 3B, P < 0.01). Furthermore, the results of sphere-formation assay showed that, compared with the Vector group, the sphere-formation ability and the number of spheroids of cells in the RAD51AP1 group were significantly increased; while in contrast to the siNC group, the sphere-formation ability and the number of spheroids of cells in the si-RAD51AP1 group were remarkably decreased (Fig. 3C, P < 0.01). Moreover, the expression levels of transcription factors that could reflect the characteristics and malignancy of tumour stem cells, including sex-determining region Y-box 2 (SOX2), octamer-binding transcription factor 4 (OCT4), NANOG and Kruppel-like factor 4 (KLF4), were detected by qRT-PCR [27]. The results revealed that, compared with the Vector group, the expression levels of SOX2, OCT4, NANOG and KLF4 were significantly increased in the cells of the RAD51AP1 group; while relative to the siNC group, the expression levels of SOX2, OCT4, NANOG and KLF4 were markedly decreased in the cells of the si-RAD51AP1 group (Fig. 3D–G, P < 0.01). The above findings suggested that the elevated expression level of RAD51AP1 could promote the self-renewal and proliferation ability of OCSCs and increase the malignancy of OC.

Fig. 3
figure 3

RAD51-associated protein 1 (RAD51AP1) promotes the self-renewal and proliferation of CD133 positive (CD133 +) OVCAR4. After over-expression and knock-down of RAD51AP1 in CD133+OVCAR4 cells, the subsequently experiments were carried out. A Western blot analysis (left) and statistical analysis (right) for the protein expression level of RAD51AP1 in cells of each group; B cell colony forming experiment analysis (top) and statistical analysis (bottom) for cell colony forming ability of each group; C sphere-formation assay analysis (left) and statistical analysis (right) for sphere-formation ability in each group; D–G the mRNA levels of SOX2 (D), OCT4 (E), NANOG (F) and KLF4 (G) were analyzed by qRT-PCR. **P < 0.01 vs. Vector GROUP, ##P < 0.01 vs. SiNC group. CD133+, CD133 positive; RAD51AP1: RAD51-associated protein 1; SOX2: sex-determining region Y-box 2; OCT4: octamer-binding transcription factor 4; KLF4: Kruppel-like factor 4

Full size image

3.4 RAD51-associated protein 1 (RAD51AP1) promotes drug resistance in CD133 positive (CD133+) OVCAR4

There is evidence that tumour stem cells have more significant resistance to chemotherapeutic agents [28]. To verify whether the same effect was present in OC, MTT was applied to assess the viability of CD133+OVCAR4 and CD133−OVCAR4 cells under different chemotherapeutic drug treatments, including paclitaxel, cisplatin and doxorubicin. The results showed that the viability of cells under the treatment of paclitaxel, cisplatin and doxorubicin in the CD133+OVCAR4 group was significantly higher than that of cells in the CD133−OVCAR4 group; besides, the IC50 of cells treated by paclitaxel, cisplatin and doxorubicin in the CD133+OVCAR4 group were significantly higher than those in cells of the CD133−OVCAR4 group (Fig. 4A–C, P < 0.01). To further investigate the role of RAD51AP1 in drug resistance in OCSCs, MTT was employed to assess the cell viability of each group after over-expression and knock-down of RAD51AP1 expression levels in CD133+ OVCAR4 cells. The results showed that, under paclitaxel, cisplatin and doxorubicin treatments, the viability of cells was significantly higher in the RAD51AP1 group than in the Vector group, while significantly lower in the si-RAD51AP1 group than in the siNC group (Fig. 4D–F, P < 0.01). Besides, compared with the Vector group, the IC50 of cells treated by paclitaxel, cisplatin, and doxorubicin in the RAD51AP1 group was significantly raised; in contrast to the siNC group, the IC50 of cells treated by paclitaxel, cisplatin, and doxorubicin in the si-RAD51AP1 group was significantly lower (Fig. 4D–F, P < 0.01). The above outcomes indicated that the elevated expression level of RAD51AP1 promoted drug resistance in OCSCs, while the knock-down of RAD51AP1 expression level could significantly enhance the sensitivity of OCSCs to chemotherapeutic drugs.

Fig. 4
figure 4

MTT analysis for the viability of cells in each group treated with different chemotherapeutic agents. A–C MTT analysis for the viability curves (top) and corresponding IC50 values (bottom) of CD133+OVCAR4 and CD133−OVCAR4 cells after treatment with different concentrations of paclitaxel (A), cisplatin (B), and doxorubicin (C). **P < 0.01. D–F After over-expression and knockdown of RAD51AP1 in CD133+ OVCAR4 cells, MTT assay was performed to analyze the viability curves (top) and corresponding IC50 values (bottom) of cells in each group after treatment with different concentrations of paclitaxel (D), cisplatin (E) and doxorubicin (F). **P < 0.01 vs. Vector group; ##P < 0.01 vs. siNC group. IC50: half maximal inhibitory concentration; CD133+: CD133 positive; CD133−: CD133 negative; RAD51AP1: RAD51-associated protein 1

Full size image

3.5 RAD51-associated protein 1 (RAD51AP1) regulates CD133 positive (CD133+) OVCAR4 activity through the TGF-β1/SMAD4 signaling pathway

The TGF-β1/SMAD4 signaling pathway plays an important regulatory role in tumors [29]. To examine whether RAD51AP1 regulated the biological activity of OCSCs through the TGF-β1/SMAD4 signaling pathway, the protein expression levels of TGF-β1 and SMAD4 in each group were analyzed using western blot after over-expression and knock-down of RAD51AP1 in CD133+OVCAR4 cells. The analysis results showed that, compared with the Vector group, the protein level of TGF-β1 was significantly higher while the expression level of SMAD4 was much lower in the cells of the RAD51AP1 group. However, in comparison with the siNC group, the protein level of TGF-β1 was markedly lower while the expression level of SMAD4 was significantly higher in the cells of the si-RAD51AP1 group (Fig. 5A, B, P < 0.01). Consequently, RAD51AP1 may regulate the biological activity of OCSCs through the TGF-β1/SMAD4 signaling pathway.

Fig. 5
figure 5

RAD51-associated protein 1 (RAD51AP1) regulates CD133 positive (CD133+) OVCAR4 activity through the TGF-β1/SMAD4 signaling pathway. A, B Protein expression levels of TGF-β1 and SMAD4 were detected using western blot (A) and statistically analyzed (B) after over-expression and knock-down of RAD51AP1 in CD133+OVCAR4 cells (A). **P < 0.01 vs. Vector group; ##P < 0.01 vs. siNC group

Full size image

4 Discussion

OC has the highest mortality rate among gynecological malignancies, and chemotherapy is one of the most common treatment options for OC [30]. However, recurrence and drug resistance remain a challenge for clinical treatment of patients with advanced tumors. Therefore, it is necessary to elucidate the molecular biology and mechanisms associated with chemosensitivity to improve the survival of OC patients. OCSCs play an important regulatory role in OC recurrence, metastasis and drug resistance [13]. Although a number of studies have reported the mechanisms of drug resistance in OCSCs [31,32,33], further research and refinement are still needed to provide new directions and theoretical guidance for the treatment of OC.

OCSCs are identified and characterized by specific markers on cell membrane surface and intracellular markers [34, 35]. Ferrandina et al. first identified CD133 in OC [36]. Besides, it is reported that OC cells with CD133+ phenotype have more strong proliferation and colony ability than CD133− cells [37]. In this study, CD133 was adopted to analyze the CD133 expression level in OVCAR4 cell line. The results revealed that about 10% of cells expressed CD133+. Moreover, our experimental results demonstrated that CD133+OVCAR4 cells had greater self-renewal capacity and proliferation activity, consistent with the findings of Ricci et al. [37]. In addition, our drug resistance analysis showed that CD133+OVCAR4 cells had more prominent drug resistance, consistent with the findings of Liu et al. The above results not only demonstrated the role of CD133 in the characterization and recognition of OCSCs in OVCAR4 cell lines, but also verified the important function of OCSCs in OC proliferation and chemotherapeutic drug resistance.

RAD51AP1 not only exerts a central function in HR and gene integrity, but also plays a vital role in promoting signal transduction in the early growth phase of tumors [38, 39]. Through bioinformatics analysis, Filipe et al. found that RAD51AP1 showed a trend of high expression in OC tissues [16]. The experimental data in this study also demonstrated the high expression level of RAD51AP1 in OC. In addition, RAD51AP1 has been reported to play an important regulatory role in tumour stem cells. For example, Bridges et al. claimed that defects in RAD51AP1 played a key role in chemoresistance by inhibiting the self-renewal of colorectal cancer stem cells (CRCSC) and making cancer cells re-sensitive to chemotherapy [20]. Moreover, Bridges et al. also demonstrated that RAD51AP1 played a critical regulatory role in the self-renewal of BCSCs through in vivo limiting dilution assays [37]. RAD51AP1 deletion plays a protective role in spontaneous breast tumor and related lung metastasis in mice and can improve the efficacy of chemotherapy and radiotherapy. However, there are no studies that RAD51AP1 is associated with the self-renewal and chemoresistance of OCSCs. In this study, RAD51AP1 protein expression levels were significantly elevated in OCSCs, suggesting the possibility that RAD51AP1 also played an important regulatory role in OCSCs. The RAD51AP1 expression was over-expressed and knocked down in vitro using the CD133+OVCAR4 cell model. Our experimental results directly demonstrated that over-expression of RAD51AP1 significantly not only promoted the self-renewal and proliferation of CD133+OVCAR4, but also enhanced its tolerance to chemotherapeutic agents; whereas knock-down of RAD51AP1 significantly inhibited the self-renewal and proliferation of CD133+OVCAR4, and made cells re-sensitive to chemotherapeutic agents. Our findings are consistent with those of Bridges et al. in CRCSCs and BCSCs. All these studies have confirmed the important regulatory role of RAD51AP1 in the self-renewal and chemoresistance of tumour stem cells and further expanded our understanding of RAD51AP1 in different tumour models.

Notably, a study by Bridges et al. found that knock-down of RAD51AP1 gene significantly inhibited the tumor growth in a mouse model of colorectal cancer, but did not influence the normal homeostasis of the colonic epithelium, suggesting that RAD51AP1 did not affect the growth of normal cells [20]. This finding provides support for the exploration of RAD51AP1 serving as a potential therapeutic target for colorectal cancer. Considering the over-expression of RAD51AP1 in colorectal cancer patients, therapeutic strategies targeting RAD51AP1 only have a positive impact on cancer treatment without adversely affecting the growth of normal cells. Therefore, functional inactivation of RAD51AP1 has important potential in cancer prevention and treatment. Although we did not use the RAD51AP1 knock-down mouse model to analyse the growth status of ovarian epithelial cells in this study, given the conserved nature of the gene function and the results of this study, we hypothesized great potential of RAD51AP1 as a therapeutic target of OC. However, further experimental studies are required to confirm this inference.

There are four key genes involved in stem cell pluripotency, including SOX2, OCT4, KLF4 and NANOG, of which NANOG is the major regulator [40]. NANOG is a transcription factor involved in maintaining the self-renewal of stem cells. Many studies have proved the role of NANOG in normal and cancer stem cell function. High NANOG expression level is associated with a variety of cancers, including melanoma, glioma, colorectal cancer, and hepatocellular carcinoma. Also, high NANOG expression level is correlated with processes such as over-proliferation, drug resistance, and apoptosis inhibition [41]. After over-expression of RAD51AP1, we found that the expression level of NANOG was significantly increased in CD133+ OVCAR4 cells; whereas after knockdown of RAD51AP1, the expression level of NANOG was significantly decreased. Such result suggested that RAD51AP1 regulated the self-renewal ability of OCSCs by affecting NANOG. Liu et al. stated that NANOG regulated the proliferation of prostate cancer stem cells (PCSCs) through the TGF-β1/SMAD signaling pathway [42]. Besides, Xu et al. discovered that NANOG acted as a direct target of TGF-β/Activin-mediated SMAD signaling in human embryonic stem (ES) cells, and played a key role in maintaining self-renewal of human ES cells [43]. Therefore, we hypothesized that RAD51AP1 may regulate the expression level of NANOG in OCSCs by affecting the TGF-β1/SMAD signaling pathway. To prove this hypothesis, we examined the protein levels of TGF-β1 and SMAD. The results showed that the SMAD4 protein expression level increased significantly after knocking down RAD51AP1. It has been reported that SMAD4 is an oncogene capable of suppressing the expression of genes related to tumour growth [44]. In this study, the elevated expression level of SMAD4 not only suppressed the expression of NANOG, but also down-regulated the expression levels of SOX2, OCT4, and KLF4; the down-regulation of the expression levels of these pluripotent genes ultimately suppressed the self-renewal ability of OCSCs and enhanced the sensitivity of tumor cells to chemotherapeutic drugs. Similarly, Zhao et al. also stated that RAD51AP1 regulated the OC progression through the TGF-β1/SMAD signaling pathway [45]. However, compared with the study of Zhao et al., our experimental data further demonstrated that RAD51AP1 regulated OC progression by affecting the TGF-β1/SMAD signaling pathway in OCSCs. Such finding provides a more precise theoretical basis for our in-depth understanding to the function of OCSCs and the molecular mechanism of OC. Notably, the results of Zhao et al. revealed that the expression level of SMAD4 presented a down-regulation trend after knock-down of RAD51AP1 in serous ovarian carcinoma cell line (SKOV3). Such result may be caused by the variability among cell lines, and the mechanism behind it deserves more in-depth study in the future.

Although our study elucidates the possible mechanism by which RAD51AP1 promotes OC resistance. However, our study only explored single cell lines, and future research will need to investigate the characteristics of RAD51AP1 on other types of tumor stem cells, such as CD133+/CD117+, CD44+/CD177, CD44+/CD24 −. In addition, we have not demonstrated our conclusion in animal bodies, and in the next study, we can conduct experimental studies in nude mice.

5 Conclusion

In this study, our results demonstrated that RAD51AP1 plays an important regulatory role in OCSCs. Briefly, knock-down of RAD51AP1 can inhibit the expression levels of genes associated with pluripotency, including NANOG, SOX2, OCT4, and KLF4, through the TGF-β1/SMAD signaling pathway. Down-regulation of the levels of these genes ultimately inhibits OCSCs’ self renewal, proliferation and resistance to chemotherapeutic drugs. The findings of this study not only help us better understand the mechanism of drug resistance in OCSCs, but also provide new targets for the development of anticancer drugs against OCs. These new targets are expected to overcome the problem of resistance to conventional chemotherapeutic drugs, thereby providing a more effective strategy for the treatment of OC.