1 INTRODUCTION

Ovarian cancer (OC) is the third most common gynecologic malignancy worldwide but accounts for the highest mortality rate among these cancers.1 OC is a diverse set of diseases, and among the most clinically significant, epithelial ovarian cancers, at least four distinct entities exist: serous ovarian carcinoma (SOC), endometrioid carcinoma, clear cell carcinoma, and mucinous carcinoma. It has a high incidence and high mortality rate and is especially difficult to discover during the early stages.2 SOC typically presents as a large ovarian mass accompanied by widespread peritoneal metastasis, causing ~70% of deaths.3 A high rate of recurrence and poor prognosis due to the metastasis of this devastating disease are the common and serious problems to be conquered in SOC.4 The metastasis of SOC involves complex changes in networks of genes, signaling pathways, and gene regulation that control tumor cell invasion and migration.5 Therefore, more extensive research and a better understanding of the molecular changes that induce the metastasis of SOC cells are urgently needed.

Long noncoding RNAs (lncRNAs) are a subcategory of endogenous noncoding RNAs (ncRNAs), with a length longer than 200 nucleotides.6 Accumulating evidence has demonstrated that lncRNAs exert vital roles in tumorigenesis and cancer progression.7-9 As an endogenous RNA, lncRNAs compete with other RNAs for microRNAs (miRNAs) through miRNAs response elements, affecting the regulation of target genes by miRNAs.10 MiRNAs are a class of 22 nucleotides ncRNAs that regulate protein expression. Previous reports reveal that miRNAs have essential positions in cancer cells proliferation, invasion, and metastasis.11

Considering that SOC is commonly diagnosed in advanced stages with widespread peritoneal metastases, leading to a poor prognosis,12 we thus focused primarily on SOC. Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/) is an essential and powerful tool for the bioinformatics data mining of gene expression profiles. In the current study, we analyzed two microarray platform data sets in the GEO database and showed that SET-binding factor 2 antisense RNA 1 (SBF2-AS1) expression was significantly increased in SOC. Mechanistically, downregulation of SBF2-AS1 suppressed cell proliferation and invasion via serving as a sponge of miR-338-3p to regulate the expression of E26 transformation specific-1 (ETS1) in SOC.

2 MATERIALS AND METHODS 2.1 Cell culture

SOC cells (SKOV3, OVCAR3, and A2780) were purchased from the Chinese Academy of Sciences Cell Bank (Shanghai, China). The immortalized fallopian tube epithelial cell line FT-194 was obtained from American Type Culture Collection (ATCC, Manassas, Virginia). OVCAR3 cell was maintained in RPMI-1640 (Thermo Fisher Scientific, Waltham, Massachusetts) supplemented with 10% fetal bovine serum (FBS; Thermo Fisher Scientific) and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin). SKOV3 cell was cultured using McCoy's 5a medium modified (Thermo Fisher Scientific). A2780 cell was cultured in Dulbecco's modified Eagle medium (DMEM; Thermo Fisher Scientific). FT-194 was cultured using DMEM/F12 50:50 Mix [-] L-glutamine (Corning, New York). The cells were incubated in an atmosphere containing 5% CO2 at 37°C.

2.2 Cell transfection

The miR-338-3p mimic, SBF2-AS1 shRNA (sh-SBF2-AS1), and their negative controls (miR-ctrl, sh-ctrl) were purchased from Shanghai GenePharma Co., Ltd. (Shanghai, China). Transient transfection of shRNA or miRNA mimic was performed with Lipofectamine 2000 (Thermo Fisher Scientific) according to manufacturer's instructions. The pcDNA-SBF2-AS1 expression vector was constructed by GenScript Company (Nanjing, China). Empty vector was taken as negative control. Transfection was carried out using Lipofectamine 2000 and 2.5 μg SBF2-AS1 plasmid or empty vector control for 48 h.

2.3 RNA isolation and quantitative real-time polymerase chain reaction (qRT-PCR)

RNAs were isolated using TRIzol (Invitrogen) according to the manufacturer's instructions. Real-time PCRs were performed with SYBR Green qPCR Super Mix-UDG reagent (Invitrogen) on a CFX96 Touch sequence detection system (Bio-Rad, Hercules, California). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and U6 were used as internal controls for mRNA and miRNA, respectively. The sequences of main primers were as follows: SBF2-AS1 (forward): 5′-AGACCATGTGGACCTGTCACTG-3′, SBF2-AS1 (reverse): 5′-GTTTGGAGTGGTAGAAATCTGTC-3′; GAPDH (forward): 5′-CCACATCGCTCAGACACCAT-3′, GAPDH (reverse): 5′-ACCAGGCGCCCAATACG-3′; miR-338-3p (forward): 5′-GGGTCCAGCATCAGTGATT-3′, miR-338-3p (reverse): 5′-GTGCAGGGTCCGAGGT-3′; U6 (forward): 5′-TTATGGGTCCTAGCCTGAC-3′, U6 (reverse): 5′-CACTATTGCGGGCTGC-3′; ETS1 (forward): 5′-GATAGTTGTGATCGCCTCACC-3′, ETS1 (reverse): 5′-GTCCTCTGAGTCGAAGCTGTC-3′. The 2−ΔΔCt method was applied to analyze the data, and each experiment was performed in triplicate.

2.4 Cell counting kit-8 (CCK-8) assay

Transfected SKOV3 or OVCAR3 cells (5 × 103 cells) were seeded into 96-well plates overnight. The medium of each well was subsequently replaced with 100 μl of fresh culture media with 10% CCK-8 (Beyotime, Nanjing, China) at 24, 36, 48, or 72 h. Then, the cells were incubated for 3 h. The absorbance was measured at 450 nm using a microplate reader. Colony formation was performed as described previously.13

2.5 Invasion assay

Transfected SKOV3 or OVCAR3 cells were placed on the top of the Matrigel-coated invasion chambers (BD Biosciences) with serum-free DMEM; 500 μl of DMEM containing 10% FBS was added into the lower chamber. After 24 h, cotton swabs were used to remove the noninvasive cells. The invading cells were fixed with 95% ethanol and stained with 0.1% crystal violet. The number of invaded cells was counted under an inverted microscope.

2.6 Luciferase reporter assay

The wild-type or mutant fragments of SBF2-AS1 was subcloned into pGL3 plasmids (Promega, Madison, Wisconsin) to construct SBF2-AS1-wt or SBF2-AS1-mut, respectively. Likewise, the 3′-UTR sequences of ETS1 containing predicated or mutated miR-338-3p binding sites were used to synthesize ETS1-WT or ETS1-mut vector. SKOV3 or OVCAR3 cells were cotransfected with miR-338-3p mimic and corresponding reporter plasmids using Lipofectamine 2000 reagent (Thermo Fisher Scientific). At 48 h after transfection, luciferase activity was estimated with the luciferase reporter assay system (Promega).

2.7 Western blot

Total proteins were extracted from cells using RIPA buffer (Beyotime). Proteins were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to poly(vinylidene fluoride) (Millipore, Braunschweig, Germany). The membranes were blocked with 5% skim milk for 1 h. Subsequently, membranes were incubated overnight at 4°C with diluted antibodies against ETS1 or GAPDH (1:1000, Abcam, Cambridge, UK), followed by incubation with an HRP-conjugated secondary antibody (1:1000, Santa Cruz). Bands were then visualized with chemiluminescence molecular imaging system (Bio-Rad) and ECL kit (Millipore).

2.8 A xenograft model of ovarian cancer

Lentiviruses were generated in 293 T cells by transfecting lentiviral vector (pLKO-sh-ctrl or pLKO-sh-SBF2-AS1), using Lipofectamine 2000 Reagent (Invitrogen). At 48 h post-transfection, virus-containing supernatants were collected for infection. For viral transductions, SKOV3 cells were incubated with the pLKO-sh-ctrl or pLKO-sh-SBF2-AS1 lentiviruses overnight at 37°C in a humidified cell culture incubator; 24 h postinfection, cells were selected in the presence of 1.5 μg/ml puromycin. A xenograft model of ovarian cancer was established using SKOV3 cells with SBF2-AS1 depletion (sh-SBF2-AS1 group) or not (sh-ctrl group). Briefly, 6-week-old athymic nude mice were randomly assigned to two groups (n = 3 for each group). For each group of mice, transfected SKOV3 cells (2 × 104) were injected into the right flank. Mice were then monitored for the growth of tumors. Tumor length (L) and width (W) were measured twice a week. And tumor volume (TV) was calculated as TV = L × W2/2. By Day 28 after inoculation, all mice were sacrificed, and tumors were dissected. The dissected tumors were weighed and subjected for immunohistochemistry (IHC) staining using ETS1 antibody. Protocols for animal experiments were approved by the Ethics Committee from the Weifang Maternal and Child Health Hospital.

2.9 Statistical analysis

All experiments were independently repeated in triplicate. The data are shown as mean ± SD and were analyzed using SPSS 10.0. Statistical evaluation of the data was performed by t test or one-way ANOVA. p value less than 0.05 was statistically significant.

3 RESULTS 3.1 SBF2-AS1 is upregulated in OC

The GEO database (GSE10971 and GSE69428) was used for the bioinformatics data mining of gene expression profiles between SOC tissues and normal fallopian tubes. Volcano plots displayed the differentially expressed genes verified from the two GEO data sets (Figure 1A). The Venn diagram of the differential genes shows the number of common genes in two GEO data sets (Table 1 and Figure 1B). As shown in Figure 1C, the level of SBF2-AS1 was significantly higher in SOC tissues when compared to that in normal fallopian tubes. As longer term follow-up data were not available in the GSE10971 and GSE69428 SOC cohort, we determined the prognostic value of SBF2-AS1 using another GEO data set (GSE18520). As shown in Figure 1D, Kaplan–Meier curve shows that higher SBF2-AS1 expression was associated with poor prognoses in patients with SOC (cutoff value = 81, p = 0.038). Results also indicated that SBF2-AS1 showed increased expression in SOC cells (SKOV3, OVCAR3, and A2780) in comparison with fallopian tube epithelial cell line FT-194 (Figure 1E). Further, we sought to probe the regulatory mechanism of SBF2-AS1 in SOC. The lncATLAS website (http://lncatlas.crg.eu/) predicted that SBF2-AS1 was predominantly located in the cytoplasm (Figure 1F).

SBF2-AS1 is overexpressed in SOC tissues and its clinical significance. (A) Volcano plots of differentially expressed genes in GSE10971 and GSE69428 data set. (B) Venn diagram demonstrates the differentially expressed intersection genes of GEO data sets. (C) Box plot for expression of SBF-AS1 in two GEO data sets. (D) The prognostic value of SBF2-AS1 in SOC was assessed using GSE18520 data. (E) qRT-PCR analysis of SBF2-AS1 expression in fallopian tube epithelial cell line FT-194 and three SOC cells. **p http://lncatlas.crg.eu/). GEO, Gene Expression Omnibus; qRT-PCR, quantitative real-time polymerase chain reaction; SBF2-AS1, SET-binding factor 2 antisense RNA 1; SOC, serous ovarian carcinoma 3.2 SBF2-AS1 enhances the proliferation and invasion of SOC cells

To assess the mechanisms by which SBF2-AS1 promotes the progression of SOC, we knocked down the expression of SBF2-AS1. The knockdown effect of sh-SBF2-AS1 in SKOV3 and OVCAR3 cells was determined by qRT-PCR analysis (Figure 2A). We found that knockdown of SBF2-AS1 reduced the proliferative viability (Figure 2B,C) and invasive potential (Figure 2D) in SKOV3 and OVCAR3 cells. Then, the overexpression effect of SBF2-AS1 plasmid in SKOV3 was assessed by qRT-PCR analysis (Figure 2E). Overexpression of SBF2-AS1 promoted the proliferative viability (Figure F) and invasive potential (Figure 2G) in SKOV3 cells. These data thus corroborate that SBF2-AS1 inhibition is sufficient to block SOC cells growth and invasion.

SBF2-AS1 promotes the proliferation and invasion of SOC cells. (A). qRT-PCR analysis of the knockdown effect of sh-SBF2-AS1 in SKOV3 and OVCAR3 cells. (B,C). CCK-8 analysis of the proliferation viability after the transfection of sh-SBF2-AS1 in SKOV3 and OVCAR3 cells. (D) Transwell analysis of the invasive potential after the transfection of sh-SBF2-AS1 in SKOV3 and OVCAR3 cells. (E) qRT-PCR analysis of the overexpression effect of SBF2-AS1 plasmid in SKOV3 cells. (F) CCK-8 analysis of the proliferation viability after the transfection of SBF2-AS1 plasmid in SKOV3 cells. (G) Transwell analysis of the invasive potential after the transfection of SBF2-AS1 plasmid in SKOV3 cells. Data are the mean ± SD of three experiments. **p < 0.01. qRT-PCR, Quantitative real-time polymerase chain reaction; SBF2-AS1, SET-binding factor 2 antisense RNA 1

3.3 BF2-AS1 is associated with miR-338-3p

Recently, lncRNA has been reported to work as miRNA sponges in tumor progression. The miRNAs that were likely to bind to SBF2-AS1 were predicted in RNA22 and starBase databases, and the results from the comparison of the two databases identified an miRNA in the intersection: hsa-miR-338-3p (Figure 3A). We subsequently constructed luciferase reporter plasmids containing the wild-type SBF2-AS1 (SBF2-AS1-WT) and mutant SBF2-AS1 with mutations of predicted miR-338-3p binding sites (SBF2-AS1-MUT). Luciferase reporter assays showed that miR-338-3p mimic led to a marked decrease in luciferase activity in SBF2-AS1-WT, without changing the luciferase activity of SBF2-AS1-MUT in SOC cells (Figure 3B). Meanwhile, qRT-PCR showed that sh-SBF2-AS1 increased the level of miR-338-3p (Figure 3C) whereas SBF2-AS1 overexpression decreased miR-338-3p expression in SOC cells (Figure 3D). Those data indicate that SBF2-AS1 is able to inhibit the expression of miR-338-3p.

SBF2-AS1 is associated with miR-338-3p. (A) Putative miR-338-3p binding site in the sequence of SBF2-AS1. (B) MiR-338-3p mimic reduced the luciferase activity of SBF2-AS1-WT group. **p < 0.01 compared with miR-ctrl. (C) Downregulation of SBF2-AS1 increased the expression level of miR-338-3p in SOC cells. (D) Upregulation of SBF2-AS1 decreased the expression level of miR-338-3p in SOC cells. **p < .01 compared with sh-ctrl or vector. qRT-PCR, Quantitative real-time polymerase chain reaction; SBF2-AS1, SET-binding factor 2 antisense RNA 1; SOC, serous ovarian carcinoma

3.4 ETS1 is a direct target of miR-338-3p

To further explore the potential mechanism of miR-338-3p. miRTarBase database was used. Results showed that miR-338-3p could target ETS1 3′-UTR (Figure 4A). This direct interaction was confirmed using luciferase reporter assay. Results showed that miR-338-3p mimic reduced the luciferase activity of ETS1-WT group, without changing the luciferase activity of ETS1-MUT in SKOV3 and OVCAR3 cells (Figure 4B). The roles of miR-338-3p on the SOC progression were further evaluated after cells were transfected with miR-338-3p mimic (Figure 4C). Western blot showed that ETS1 protein was reduced in SOC cells transfected with miR-338-3p mimic (Figure 4D). Colony formation and transwell assays showed that miR-338-3p mimic significantly reduced SOC cells proliferation and invasive ability in vitro (Figure 4E,F). Taken together, these results disclose that ETS1 is targeted by miR-338-3p and miR-338-3p acts as a tumor suppressor.

ETS1 is a target gene of miR-338-3p. (A). Putative miR-338-3p binding site in the sequence of ETS1. (B) MiR-338-3p mimic reduced the luciferase activity of ETS1-WT group. **p < 0.01 compared with miR-ctrl. (C) Transfection efficiency of miR-338-3p mimic was determined by qRT-PCR. (D) Upregulation of miR-338-3p decreased the protein expression of ETS1 in SOC cells. (E) The effect of miR-338-3p on the colony formation ability of SOC cells was assessed by the colony formation assay. (F) The effect of miR-338-3p on the invasive capacity of SOC cells was assessed by transwell assay. **p < 0.01 compared with miR-ctrl. ETS1, E26 transformation specific-1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; qRT-PCR, quantitative real-time polymerase chain reaction; SOC, serous ovarian carcinoma

3.5 SBF2-AS1 promotes ETS1 expression by regulating miR-338-3p

Next, we detected the abnormal expression of ETS1 between SOC tissues and normal fallopian tubes using GSE10971 and GSE69428. As shown in Figure 5A, ETS1 was remarkedly upregulated in SOC compared with that in normal fallopian tubes. Correlation analysis between the expression of ETS1 and SBF2-AS1 in The Cancer Genome Atlas cohort of SOC patients exhibited that ETS1 significantly positively correlates with SBF2-AS1 (Figure 5B). Next, we explored the interaction among SBF2-AS1, miR-338-3p, and ETS1 in SOC. qRT-PCR showed that miR-338-3p mimic significantly decreased ETS1 mRNA level in SOC cells, while the upregulation of SBF2-AS1 after SBF2-AS1 transfection rescued the mRNA level of ETS1 (Figure 5C). And the results were further confirmed by western blot (Figure 5D). In addition, colony formation and transwell assay demonstrated that SBF2-AS1 partially rescued the colony formation and invasion ability of SOC cells inhibited with miR-338-3p mimic (Figure 5E,F). Thus, we suggest that SBF2-AS1 promotes SOC progression via sponging miR-338-3p to regulate ETS1.

TABLE 1. The intersection of abnormally expressed lncRNAs in GSE10971 and GSE69428 Log FC (fallopian vs. SOC) 12 common elements in “GSE10971” and “GSE69428” GSE10971 GSE69428 SOC vs. fallopian Long noncoding RNA EP300-AS1 2.90131594 3.462541 Downregulated Long noncoding RNA KLF3-AS1 1.9261094 1.577749 Downregulated Long noncoding RNA WDR86-AS1 1.67630031 2.840097 Downregulated Long noncoding RNA HAND2-AS1 3.11419551 1.236381 Downregulated Long noncoding RNA ZBED3-AS1 2.29172133 1.532284 Downregulated Long noncoding RNA SBF2-AS1 −1.81955943 −1.656047 Upregulated Long noncoding RNA C8orf34-AS1 2.6222293 2.00811 Downregulated Long noncoding RNA 01091 3.51332354 1.770893 Downregulated Long noncoding RNA 00893 2.47264128 1.719863 Downregulated Long noncoding RNA 01550 1.95801578 1.026989 Downregulated Long noncoding RNA 01296 −2.03827633 −1.21529 Upregulated Long noncoding RNA 00312 1.37021997 1.624301 Downregulated Abbreviations: SBF2-AS1, SET-binding factor 2 antisense RNA 1; SOC, serous ovarian carcinoma.

The expression of ETS1 in SOC. (A). Box plot for expression of ETS1 in GSE10971 and GSE69428 data set. (B) Correlation analysis for SBF2-A1 and ETS1 from TCGA database. (C,D) miR-338-3p upregulation decreased ETS1 expression both in mRNA and protein levels, which could be rescued by SBF2-AS1. (E) SBF2-AS1 overexpression plasmid rescued the effects of miR-338-3p on SOC cells colony formation. (F) SBF2-AS1 overexpression plasmid rescued the effects of miR-338-3p on SOC cells invasion ability. **p < 0.01 compared with miR-ctrl, ##p < 0.01 compared with miR-338-3p. ETS1, E26 transformation specific-1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; SBF2-AS1, SET-binding factor 2 antisense RNA 1; SOC, serous ovarian carcinoma; TGCA, The Cancer Genome Atlas

3.6 Silencing of SBF2-AS1 impedes xenograft tumor growth

A xenograft tumor model was established to observe the tumor growth after subcutaneous inoculation with sh-SBF2-AS1 stably transfected SKOV3 cells. Transfection efficiency was determined using qRT-PCR analysis (Figure 6A). We found that the proliferative activity of the tumors was lowered in sh-SBF2-AS1 group as compared with the sh-ctrl group (Figure 6B). The average tumor weight and TVs were reduced in sh-SBF2-AS1 group (Figure 6C,D). IHC and western blot analysis indicated the decreased expression level of ETS1 in sh-SBF2-AS1 group (Figure 6E,F). qRT-PCR analysis also showed that the level of miR-338-3p was raised in sh-SBF2-AS1 group as compared with the sh-ctrl group (Figure 6G). Therefore, it was concluded that silencing of SBF2-AS1 blocks the growth of SOC cells in vivo.

Knockdown of SBF2-AS1 inhibits the xenograft tumor growth. (A) qRT-PCR analysis of the expression level of SBF2-AS1 in sh-SBF2-AS1 or sh-ctrl stably transfected SKOV3. (B) Representative photograph of the xenograft tumors after the inoculation with sh-SBF2-AS1- and sh-ctrl-transfected SKOV3 cells. (C) Growth curve analysis of the tumor proliferation activity after treatment with sh-SBF2-AS1 and sh-ctrl groups. (D) Comparison of the average tumor weight between sh-SBF2-AS1 and sh-ctrl groups. (E) IHC analysis of the expression levels of ETS1 in sh-SBF2-AS1 and sh-ctrl group. (F) The expression of ETS1 was determined by western blot. (G) qRT-PCR analysis of the expression level of miR-338-3p in xenograft tissue from sh-SBF2-AS1 and sh-ctrl groups. **p < 0.01 compared with sh-ctrl. ETS1, E26 transformation specific-1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; IHC, immunohistochemistry; SBF2-AS1, SET-binding factor 2 antisense RNA 1

4 DISCUSSION

Epithelial ovarian cancer is the main pathological type of ovarian cancer, and it has some histological subtypes, such as ovarian clear cell carcinoma, mucinous ovarian cancer, and SOC. SOC is the commonest subtype of epithelial ovarian cancer and accounts for approximately 80% of OC deaths.14 Exploration of molecular mechanisms of SOC metastasis is valuable for developing novel therapeutic approaches and thus improving patients' survival.15 Emerging studies delineate that aberrant lncRNA expression is also involved in tumor proliferation, invasion, and metastasis process. Inhibiting lncRNA function in vivo may represent an effective strategy to modulate lncRNAs for therapeutic treatments.16 In the present study, we provided a novel insight into a new regulatory network of SBF2-AS1, miR-338-3p, and ETS1 in SOC cell growth and invasion.

Recently, studies revealed that SBF2-AS1 might play critical roles in cancer progression. For example, SBF2-AS1 is upregulated in radiotherapy-resistant lung cancer, and silencing SBF2-AS1 decreases the radioresistance of lung cancer cells via modulating miRNA-302a/MBNL3 axis.17 Moreover, downregulated SBF2-AS1 inhibits tumorigenesis and progression of breast cancer by sponging miRNA-143 and repressing RRS1.18 The biological significance of SBF2-AS1 is yet to be explored in SOC. Similarly, we disclosed that SBF2-AS1 expression is higher in SOC tissues and cells when compared with normal. Through loss- and gain-of-function experiments, we demonstrated that SBF2-AS1 contributes to the growth and invasion abilities of SOC cells in vitro.

Lots of studies reported that lncRNA could serve as a miRNA sponge to restore the expression of target genes by blocking miRNAs' function in tumor progression. In our study, bioinformatics analysis showed that SBF2-AS1 might directly interact with miR-338-3p in SOC. MiR-338-3p has been verified to act as a tumor suppressor that disrupts the growth of gastric cancer cells and prostate cancer.19, 20 We observed that miR-338-3p expression was obviously downregulated in SOC cells transfected with SBF2-AS1 overexpression plasmid. However, miR-338-3p expression was significantly raised in SOC cells transfected with SBF2-AS1-silencing plasmid. Those data suggested that SBF2-AS1 might serve as a competing endogenous RNA to compete for miRNAs, thereby negatively regulating the expression of miR-338-3p.

ETS1 is associated with numerous processes of tumorigenesis and metastasis.21, 22 An increasing number of studies showed that ETS1 is highly expressed in a variety of malignant tumors and participates in cell invasion, metastasis, proliferation, and apoptosis by regulating the expression of a variety of genes, including MMPs.23, 24 Bioinformatics and luciferase reporter assays confirmed that ETS1 acted as a potential target of miR-338-3p in SOC progression. Furthermore, we revealed that SBF2-AS1 regulated ETS1 expression by sponging miR-338-3p in SOC cells. Importantly, the aggressiveness of SOC cells inhibited by miR-338-3p was rescued by SBF2-AS1 overexpression plasmid. In addition to these findings, our results showed that silencing of SBF2-AS1 impeded the xenograft tumor growth SOC cell, SKOV3 in vivo. Consistent with the observations in vitro, silencing of SBF2-AS1 declined the expression of ETS1 and elevated the level of miR-338-3p in xenograft tumor tissue.

Kaplan–Meier curve showed that higher SBF2-AS1 expression was associated with poor prognoses in patients with SOC. Certainly, its clinical value warrants further validations with an in-hospital larger cohort. Further studies, such as bioinformatic analysis, are needed to investigate the prognostic value of SBF2-AS1 in other histological types of OC. We will collect an adequate case number to validate its prognostic significance.

Altogether, our current results supported that downregulation of SBF2-AS1 inhibits the proliferation and invasion of SOC cells through sponging miR-338-3p to regulate ETS1. Thus, SBF2-AS1 may be an important molecular target and is a promising direction for future research on the treatment of SOC.

CONFLICT OF INTEREST

All authors declare no conflict of interest.