miR‐744‐5p inhibits cellular proliferation and invasion via targeting ARF1 in epithelial ovarian cancer

Abstract

miR‐744‐5p has been demonstrated to play an important role in cancer progression. However, the functions of miR‐744‐5p in epithelial ovarian cancer (EOC) are not well clarified. In this study, our aim was to investigate the role of miR‐744‐5p and its underlying molecular mechanism in cell progression of EOC. EOC clinical tissues and matched adjacent ovarian epithelial tissues were collected from 18 patients. Tissues and cell lines were analyzed by qPCR or Western blot to investigate the expression of miR‐744‐5p and ARF1 in EOC. Cell proliferative capacity was assessed by CCK8 and colony formation assays. Wound healing and transwell assays were performed to evaluate cell migration and invasion. The potential binding relation between miR‐744‐5p and IRF1 was demonstrated by dual luciferase report assay. The results showed that expression of miR‐744‐5p was low in EOC clinical tissues and cells. Overexpression of miR‐744‐5p inhibited proliferation, migration, and invasion of EOC cells. Further mechanistic study identified that ARF1 is a target of miR‐744‐5p, which is negatively correlated with the expression of miR‐744‐5p, and overexpression of ARF1 could reverse the inhibition of miR‐744‐5p on the proliferation, migration, and invasion of EOC cells. Taken together, our results indicated that miR‐744‐5p attenuated EOC progression via targeting IRF1, providing potential guidance for the clinical treatment of ovarian cancer.

1. INTRODUCTION

Epithelial ovarian cancer (EOC) is one of the most common gynecological malignant tumors of women, exhibiting extremely high mortality, rapid metastasis, strong invasion ability, and high recurrence. ¹ , ² , ³ Due to the absence of obvious symptoms in the early stage of EOC, and the lack of specific and sensitive diagnostic methods, most patients are diagnosed at the advanced stages, and the five‐year survival rate is less than 30%. ⁴ , ⁵ In recent years, although great strides have been made in diagnosis, surgery, chemotherapy, targeted drugs, and biological therapy, the treatment effect and prognosis of patients with advanced EOC are still terrible. Invasion and migration are the most important biological characteristics of EOC malignancy. ⁶ , ⁷ , ⁸ Therefore, it is of great significance to explore biomarkers and mechanisms of EOC occurrence and development.

With the continuous deepening of RNA research, it has been found that microRNAs (miRNAs) mutation or abnormal expression are closely related to the occurrence of cancers, including EOC. ⁹ , ¹⁰ , ¹¹ And miRNAs as oncogene or anti‐oncogene have attracted more and more researchers' attention. miR‐744‐5p is differentially expressed in various cancers and is involved in tumor progression. ¹² , ¹³ , ¹⁴ Studies have indicated that miR‐744‐5p is lowly expressed in multiple cancers, like colorectal cancer, cervical cancer and glioblastoma, and overexpression of miR‐744‐5p can inhibit cell proliferation, migration, and invasion. ¹⁵ , ¹⁶ , ¹⁷ Moreover, miR‐744‐5p is also down‐regulated in EOC cells, its upregulation could induce cell apoptosis. ¹² However, the function of miR‐744‐5p in EOC cell proliferation and metastasis still remain largely unclear. In addition, bioinformatics software analysis suggested that many molecules might be the target genes of miR‐744‐5p. It has been found that defects in ARFs can cause abnormal membrane transport, and ARFs play a significant role in the physiological process of tumor formation, including signal transduction, cell cycle, and apoptosis. ¹⁸ , ¹⁹ Furthermore, high expression of ARF1 in EOC cells can promote cell proliferation and migration. ²⁰ Among selected molecules (from PITA and miRmap databases), ARF1 (ADP‐ribosylation factor 1) is rarely reported in EOC, and its function is closely related to the research phenotype. Nevertheless, it is not clear whether miRNA‐744‐5p could target ARF1 and participate in the EOC process.

In this study, we explored the functional roles of miR‐744‐5p in cell proliferation, migration, and invasion, and the correlation with ARF1 in EOC. Collectively, miR‐744‐5p may be a potential biomarker and inhibit EOC progression, so as to provide guidance for the clinical treatment of EOC.

2. MATERIAL AND METHODS

2.1. Clinical tissue samples

All EOC tumor tissue specimens (EOCT) and surrounding ovarian epithelial tissues (SOET, 2 cm away from the tumor margin) were collected from 18 patients, confirmed as EOC by histopathological analysis and undergone surgical resection in the Affiliated Hospital of Guizhou Medical University. Before surgery, none of the enrolled patients received radiotherapy, chemotherapy, or hormone therapy. The clinicopathological features of patients were listed in Table 1. All tissue samples were removed from the body and quickly frozen in the liquid nitrogen and stored under −80°C for the following researches. This study was approved by the Ethics Committee of the Affiliated Hospital of Guizhou Medical University and the informed consent was obtained from all patients and their families.

TABLE 1.

Clinicopathological features of patients with EOC

Characteristics	Number of cases
Age (years)
≤50	7
>50	11
Tumor size (cm)
≤5	8
>5	10
Clinical stage
I‐II	6
III‐IV	12
Histology type
Serous	13
Non‐serous	5
Metastasis
No	12
Yes	6

2.2. Cell culture

IOSE80 (Shanghai Huiying biotechnology Co., Ltd, Shanghai, China), the human normal ovarian epithelial cell line, was cultured in RPMI‐1640 medium (Gibco) and 10% fetal bovine serum (FBS; Gibco). A2870, SKOV3, CAOV‐3, OVCAR‐3, and ES‐2 cells (ATCC, Manassas, Virginia) were cultured in Dulbecco's Modified Eagle Medium (DMEM medium; Gibco) supplemented with 10% FBS. All cells were incubated under 37°C in a 5% CO₂ atmosphere incubator during the whole study.

2.3. Quantitative real‐time PCR (qPCR)

Total RNA was isolated from EOC cells or tissues using the total RNA isolation system (Roboklon, Berlin, Germany) according to the manufacture's protocol. Then cDNA was reverse transcribed by using a High‐Capacity cDNA Archive Kit (BioRad, Hercules, California). qPCR assay was performed using SYBRPremix ExTaq (TaKaRa, Otsu, Shiga, Japan) and Eppendorf Mastercycler ep realplex detection system (Eppendorf, Hamburg, Germany). The primer sequences were used as follows:

U6‐Forward: 5′‐CTCGCTTCGGCAGCACA‐3′,

U6‐Reverse: 5′‐AACGCTTCACGAATTTGCGT‐3′;

miR‐744‐5p‐Forward: 5′‐TGCGGGGCTAGGGCTA‐3′,

miR‐744‐5p‐Reverse: 5′‐CGGCCCAGTGTTCAGACTAC‐3′;

GAPDH‐Forward: 5′‐CCAGGTGGTCTCCTCTGA‐3′,

GAPDH‐Reverse: 5′‐GCTGTAGCCAAATCGTTGT‐3′;

ARF1‐Forward: 5′‐GACCACCATTCCCACCATAG‐3′,

ARF1‐Reverse: 5′‐GAACACCAGGAGGACAGCAT‐3′.

The relative expression levels of ARF1 and miR‐744‐5p were normalized to GAPDH and U6, respectively.

2.4. Plasmids and transfection

miR‐744‐5p mimics, which correspond to negative control miRNA mimics (mimics NC) were purchased from GenePharma (Shanghai, China). In order to induce overexpression of ARF1, ARF1 overexpression plasmid (pcDNA3.1‐ARF1) and empty plasmid (pcDNA3.1‐NC) were synthesized and purified by RiboBio (Guangzhou, China). Cell transfection was performed using Lipofectamine 3000 (Invitrogen), and the culture medium was discarded and replaced with fresh DMEM supplemented with 10% FBS at 6 hours post‐transfection, according to the manufacturer's protocol.

2.5. Cell counting kit‐8 assay

Cell viability was detected by using Cell Counting kit‐8 assay (CCK‐8 solution; Dojindo, Japan). In brief, transfected cells (3000 cells/well) were seeded into 96‐well plates in carbon dioxide incubator for 0, 24, 48, and 72 hours incubation at 37°C, respectively. And then, 10 μL of CCK‐8 solution was added to each well and incubated for 4 hours at 37°C. The optical density (OD) value was conducted under a wavelength of 450 nm through a microplate reader.

2.6. Colony formation assay

Cells treated with miR‐744‐5p mimics were seeded into 6‐well plates with 500 cells per well and cultured with DMEM medium containing 10% FBS. After 14 days, cells were allowed to grow until the colonies were visible. Then cells were stained with crystal violet solution (Sigma Aldrich, Missouri) and the colonies were taken photos manually.

2.7. Wound healing assay

The ability of cell migration was examined by the wound healing assay. Cells (2 × 10⁶ per well) were inoculated into 6‐well plates and cultured. After transfection, cells were incubated in 5% carbon dioxide incubator at 37°C for 1 day, and 90% confluent cells were used for wound healing assay. Afterward, wound was made with a 10 μL pipette tip and washed with PBS followed by scratches capture using microscope. After another 24 hours incubation in DMEM containing 2% FBS, the scratches were taken photos again. Finally, the migration distances were analyzed as the formula (S_0 h − S_24 h)/S_0 h × 100%.

2.8. Transwell assay

In this study, the ability of cell invasion was assessed by Transwell assay. These transfected cells were cultured at 37°C in 5% carbon dioxide incubator. After 48 hours, cell invasion ability was evaluated with a 8 μm pore size Corning chambers (Corning Costar, Corning, New York) coated with Matrigel (BD Biosciences, Franklin Lakes, New Jersey). Then 2 × 10⁵ cells were implanted into upper Transwell chambers which was loaded with DMEM medium containing 10% FBS in the lower chambers as a chemoattractant. Following incubation at 37°C for 24 hours, the noninvaded cells on the surface of Transwell chambers were gently removed. The invaded cells were fixed with 100% methanol, stained with 0.5% crystal violet solution, photographed, and then counted by an IX71 inverted microscope (Olympus Corporation, Tokyo, Japan) in five randomly selected visual fields per chamber.

2.9. Western blotting analysis

The cells and tissues were harvested and lysed by using RIPA lysate (Beyotime Biotech, China). Then, protein concentration was quantified by BCA Protein Assay Kit (Beyotime Biotech, China). Therewith, the protein samples were separated by sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) and transferred to polyvinylidenedifluoride filter (PVDF) membranes (Millipore, Bedford, Massachusetts). After blocking in 5% non‐fat milk for 1 hour, the membranes were incubated with appropriate primary and secondary antibodies. The membranes were detected using an enhanced chemiluminescence system (Pierce Biotechnology, Inc., Rockford, Illinois). The experiments were carried out three separate times.

2.10. Luciferase reporter assay

ARF1 is predicted as a target of miR‐744‐5p by starBase. The wild‐type (WT) and mutant (MUT) DNA sequences of ARF1 were cloned into the pGL3‐control vector (Promega, Madison, Wisconsin). Then, cells were co‐transfected with plasmids and miR‐744‐5p mimics or the control (mimics NC) using Lipofectamine 3000 (Invitrogen). After 48 hours transfection, EOC cells were collected and then analyzed the luciferase activity via Dual‐Luciferase Reporter Assay System (Promega).

2.11. Statistical analysis

The data are presented as mean ± SD, and all experiments were assayed in triplicates (n = 3). All statistical analyses were performed using the GraphPad Prism 5.0 software (GraphPad, San Diego, California). The differences between the treatment groups were analyzed by student's t‐test (two groups) or one‐way ANOVA followed by Tukey's post hoc test (multiple groups). P‐value less than .05 were considered statistically significant, and spearman correlation coefficient was use to analysis correlation between the two groups.

3. RESULTS

3.1. Differential expression of miR‐744‐5p in clinical tissues and cell lines of EOC

In this study, qPCR was used to analyze the expression of miR‐744‐5p in EOC tissues and cells. Compared with SOET group (the non‐tumor tissues), the expression of miR‐744‐5p in EOC tissues was significantly decreased (Figure 1A). In addition, compare with the IOSE80 (human normal ovarian epithelial cell line), the expression of miR‐744‐5p was strongly down‐regulated in EOC cell lines, among which SKOV3 and OVCAR‐3 cells had the lowest expression (Figure 1B). Therefore, SKOV3 and OVCAR‐3 cells were selected for further study. The above results suggest that lower expression of miR‐744‐5p may play a key role in the development and progression of EOC, which needs to be further validated.

FIGURE 1.

Differential expression of miR‐744‐5p in clinical tissues and cell lines of EOC. The expression of miR‐744‐5p in EOC tissues and cell lines was analyzed by qPCR. A, Compared with SOET group, the expression of miR‐744‐5p in EOC tissues was lower. *P < .05 for EOCT vs SOET. B, Compared with human normal ovarian epithelial cell line IOSE80, the expression of miR‐744‐5p was significantly downregulated in all EOC cell lines. The data are presented as mean ± SD. *P < .05 for A2870, CAOV‐3, and ES‐2 cells vs IOSE80 cells; **P < .01 for SKOV3 and OVCAR‐3 cells vs IOSE80 cells

3.2. Overexpression of miR‐744‐5p inhibits proliferation, migration, and invasion in EOC cells in vitro

To investigate the potential biological function of miR‐744‐5p in EOC progression, the miR‐744‐5p mimics were introduced into SKOV3 and OVCAR‐3 cells. The results showed that miR‐744‐5p mimics produced a great upregulation of miR‐744‐5p expression level (Figure 2A). According to CCK‐8 assay, overexpression of miR‐744‐5p significantly reduced cell viability at different time points (24, 48, and 72 hours), compared with the control group (Figure 2B). From colony formation assay, we found that increased miR‐744‐5p level caused a decrease in the clonogenic SKOV3 and OVCAR‐3 cells compared with negative control (NC; Figure 2C). Wound healing assay and transwell assay revealed the differences in cell migration and invasion. The results showed that the migration and invasion capabilities of tumor cells transfected with miR‐744‐5p mimics decreased significantly (Figure 2D,E). In conclusion, these data strongly indicate that miR‐744‐5p inhibits the cell proliferation, migration, and invasion in EOC in vitro.

FIGURE 2.

Overexpression of miR‐744‐5p inhibits proliferation, migration and invasion in EOC cells in vitro. A, The expression levels of miR‐744‐5p in SKOV3 and OVCAR‐3 cells were significantly upregulated via qPCR analysis. **P < .01 for miR‐744‐5p mimics group vs mimics NC group in both cells. B, Overexpression of miR‐744‐5p reduced cell viability at different time points by CCK‐8 assay. *P < .05 for miR‐744‐5p mimics group vs mimics NC group in OVCAR‐3 cells (24, 48, and 72 hours) and SKOV3 cells (24 and 48 hours); **P < .01 for miR‐744‐5p mimics group vs mimics NC group in SKOV3 cells (72 hours). C, miR‐744‐5p overexpression inhibited the formation of colonies by colony formation assay. **P < .01 for miR‐744‐5p mimics group vs mimics NC group in both cells. D, miR‐744‐5p effectively suppressed the migration of both cell lines based on wound healing assay. *P < .05 for miR‐744‐5p mimics group vs mimics NC group in both cells. E, Upregulation of miR‐744‐5p remarkably inhibited the invasion ability of tumor cells by transwell assay. *P < .05 for miR‐744‐5p mimics group vs mimics NC group in OVCAR‐3 cells; **P < .01 for miR‐744‐5p mimics group vs mimics NC group in SKOV3 cells

3.3. ARF1 is a direct target of miR‐744‐5p

Bioinformatics prediction and luciferase reporter assay were used to verify the targeted relationship between miR‐744‐5p and ARF1. The results showed that miR‐744‐5p contains potential binding sites for ARF1 3′‐UTR, and co‐transfection cells with ARF1‐WT vector and miR‐744‐5p mimics significantly inhibited luciferase reporter activity, but no significant effect on luciferase activity in ARF1‐MUT vector and miR‐744‐5p mimics co‐transfected cells (Figure 3A,B). Subsequently, the regulation of miR‐744‐5p on ARF1 was analyzed by qPCR and Western blot assay. Overexpression of miR‐744‐5p could significantly inhibit mRNA and protein expression levels of ARF1 in SKOV3 and OVCAR‐3 cells (Figure 3C,D). Therefore, the above results suggest that ARF1 is a direct target of miR‐744‐5p.

FIGURE 3.

ARF1 is a direct target of miR‐744‐5p. A, The wild‐type (WT) and mutant (MUT) of putative miR‐744‐5p targeting sequences in ARF1 mRNA. The mutation sequence is underlined. B, The measurement of luciferase activity in SKOV3 and OVCAR‐3 cells. **P < .01 for miR‐744‐5p mimics group vs mimics NC group in ARF1‐WT transfected cells. C,D, miR‐744‐5p mimics significantly inhibited the expressions of ARF1 in SKOV3 and OVCAR‐3 cells by qPCR and Western blot assays. **P < .01 for miR‐744‐5p mimics group vs mimics NC group in both cells

3.4. Negative correlation between ARF1 and miR‐744‐5p expression in EOC tissues

In order to further explore the relationship between miR‐744‐5p and ARF1 in EOC tissues, the expression of ARF1 was first detected by qPCR and Western blot. The results showed that the mRNA and protein expression of ARF1 was significantly increased in EOC tissues (Figure 4A,B). After that, Spearman's correlation analysis found that miR‐744‐5p expression was negatively correlated with ARF1 mRNA level (Figure 4C). Consistent with the above data, the expression of ARF1 were significantly up‐regulated in SKOV3 and OVCAR‐3 cells compared with normal ovarian epithelial cell line (Figure 4D,E). In general, we provide an evidence that the miR‐744‐5p/ARF1 axis exist in EOC progression.

FIGURE 4.

Negative correlation between ARF1 and miR‐744‐5p expression in EOC tissues. A,B, The expression of ARF1 was significantly increased in EOC tissues by qPCR (*P < .05 for EOCT vs SOET) and Westren blot (**P < .01 for EOCT vs SOET). C, Spearman's correlation analysis found that miR‐744‐5p expression was negatively correlated with ARF1 mRNA level. D,E, The expression of ARF1 in SKOV3 and OVCAR‐3 cells were significantly upregulated via qPCR (**P < .01 for SKOV3 and OVCAR‐3 cells vs IOSE80 cells) and Western blot (*P < .05 for OVCAR‐3 cells vs IOSE80 cells; **P < .01 for SKOV3 cells vs IOSE80 cells) analysis

3.5. Overexpression of ARF1 reversed the inhibition of miR‐744‐5p on proliferation, migration, and invasion in EOC cells

In order to prove whether miR‐744‐5p regulates the proliferation, migration, and invasion of EOC cells by suppressing ARF1, miR‐744‐5p, and ARF1 were respectively overexpressed for investigation. Western blot and qPCR assays showed that ARF1 expression was significantly up‐regulated in SKOV3 and OVCAR‐3 cells transfected with pcDNA3.1‐ARF1 (Figure 5A,B). CCK8 assay showed that miR‐744‐5p could significantly decrease cell viability, while ARF1 overexpression could partially restore the function of miR‐744‐5p (Figure 5C). Furthermore, pcDNA3.1‐ARF1 partially reversed the functional role of miR‐744‐5p upregulation in cell proliferation (Figure 5D). The results of wound healing assay and transwell assay showed that overexpression of miR‐744‐5p markedly inhibited cell migration and invasion, while pcDNA3.1‐ARF1 could partially restored the inhibition of miR‐744‐5p mimics (Figure 5E,F). Overall, these results make clear that miR‐744‐5p inhibits cell proliferation, migration, and invasion of EOC cells by targeting ARF1.

FIGURE 5.

Overexpression of ARF1 reversed the inhibition of miR‐744‐5p on proliferation, migration and invasion in EOC cells. A,B, pcDNA3.1‐ARF1 significantly up‐regulated ARF1 level in SKOV3 and OVCAR‐3 cells by qPCR (**P < .01 for pcDNA3.1‐ARF1 group vs pcDNA3.1‐NC group) and Western blot analysis (*P < .05, **P < .01 for pcDNA3.1‐ARF1 group vs pcDNA3.1‐NC group in OVCAR‐3 and SKOV3 cells, respectively). C, miR‐744‐5p mimics could significantly inhibit cell proliferation, while pcDNA3.1‐ARF1 could partially reverse the function of miR‐744‐5p mimics by CCK8 assay. *P < .05 for miR‐744‐5p mimics group vs mimics NC group; **P < .01 for miR‐744‐5p mimics + pcDNA3.1‐ARF1 group vs miR‐744‐5p mimics group. D, miR‐744‐5p overexpression inhibited the cloning formation ability, while pcDNA3.1‐ARF1 partially restore it via colony formation assay (*P < 0.05 for miR‐744‐5p mimics group vs mimics NC group in both cells; *P < 0.05, **P < 0.01 for miR‐744‐5p mimics + pcDNA3.1‐ARF1 group vs miR‐744‐5p mimics group in OVCAR‐3 and SKOV3 cells, respectively). (E and F) ARF1 overexpression partially reversed the inhibition of miR‐744‐5p mimics on EOC cells migration (*P < .05 for miR‐744‐5p mimics group vs mimics NC group in both cells; **P < .01, *P < .05 for miR‐744‐5p mimics + pcDNA3.1‐ARF1 group vs miR‐744‐5p mimics group in OVCAR‐3 and SKOV3 cells, respectively) and invasion (**P < .01 for miR‐744‐5p mimics group vs mimics NC group and miR‐744‐5p mimics + pcDNA3.1‐ARF1 group vs miR‐744‐5p mimics group in both cells) by wound healing assay and transwell assay

4. DISCUSSION

The pathogenesis of EOC is still unclear, the lack of effective diagnosis markers, and the high mortality rate have become the urgent problems to be solved in the treatment of gynecological tumors. Currently, miRNAs have been gradually considered as a potential biological target for tumor diagnosis and treatment with the deepening of research. Some miRNAs, such as miR‐200c, miR‐30, and miR‐494, have been proved to have differential expression in EOC and are closely related to tumor prognosis, which may become potential diagnostic and therapeutic targets. ²¹ , ²² , ²³

Some studies have shown that disordered expression of miR‐744‐5p in a variety of tumors and could be used as biomarkers in the field of cancer to participate in the development of diseases. ²⁴ , ²⁵ , ²⁶ For example, miR‐744‐5p is significantly up‐regulated in prostate cancer, head and neck tumor tissues. ²⁷ , ²⁸ While, it is reported that miR‐744‐5p is downregulated in ovarian cell lines, which is consistent with our results. ¹² Furthermore, miR‐744‐5p behaves as oncogene or tumor suppressor gene in colorectal cancer, glioblastoma, and prostate cancer, and may play an essential role in the tumorigenesis and progression of cancers. ¹⁵ , ¹⁷ , ²⁸ In our study, we found that the overexpression of miR‐744‐5p inhibited proliferation, migration, and invasion in EOC cells. These results implied that miR‐744‐5p might function as a tumor suppressor in EOC.

It was well‐known that miRNAs perform its biological roles by depending on the cellular function of their targets. Therefore, identifying and studying the target of miRNAs may help to elucidate the molecular mechanism of cancers progression. In previous studies, miR‐744‐5p could serve as a potential biomarker and therapeutic target for cancers by targeting RING1, ²⁴ NOB1, ¹⁷ Bcl‐2, ¹⁶ and so on. In this study, starBase predicted that ARF1 may be a target of miR‐744‐5p. Then, luciferase reporter assay and qPCR and Western blot assays revealed miR‐744‐5p could directly target the 3′‐UTR of ARF1, and suppressed ARF1 expression at the mRNA and protein levels in EOC cells. These results indicated that ARF1 is a novel direct target of miR‐744‐5p, and miR‐744‐5p/ARF1 axis may mediate tumor occurrence and development in EOC.

In summary, the expression of miR‐744‐5p was confirmed to be significantly down‐regulated in EOC tissues and cell lines. Upregulation of miR‐744‐5p plays a promotive role in EOC cell proliferation, migration, and invasion by directly targeting ARF1. Nevertheless, the effects of miR‐744‐5p/ARF1 pathway on tumor growth and metastasis in vivo also need to be investigated in the future. Overall, this finding improved understanding of mechanism involved in EOC progression, and provided a potential therapeutic target for EOC patients.

CONFLICT OF INTEREST

All authors declare no conflict of interest.

Zhao L‐G, Wang J, Li J, Li Q‐F. miR‐744‐5p inhibits cellular proliferation and invasion via targeting ARF1 in epithelial ovarian cancer. Kaohsiung J Med Sci. 2020;36:799–807. 10.1002/kjm2.12253