Sage Journals: Discover world-class research

Abstract

OBJECTIVE:

This study aimed to explore the correlation of circular RNA ABCB10 (circ-ABCB10) expression with clinicopathological features and survival, as well as its impact on regulating cell proliferation and apoptosis in epithelial ovarian cancer (EOC).

METHODS:

A total of 103 EOC patients were consecutively recruited, then their tumor tissues were obtained for circ-ABCB10 detection using qRT-PCR. Additionally, 53 EOC adjacent tissues were collected as control. Patients’ clinicopathological and survival data were recorded. In vitro, circ-ABCB10 expression was detected in OVCAR3, UWB1.289, SKOV3, CAOV3 and IOSE80 cell lines by RT-qPCR, and the effect of circ-ABCB10 on cell proliferation and apoptosis was detected through circ-ABCB10 overexpression and silencing by plasmids transfection into SKOV3 cells.

RESULTS:

Circ-ABCB10 was upregulated in tumor tissues compared with adjacent tissues, and presented with good value in distinguishing tumor tissues from adjacent tissues (AUC $=$ 0.766, 95% CI: 0.690–0.842). Circ-ABCB10 high expression was correlated with poor differentiation, large tumor size and advanced International Federation of Gynecology and Obstetrics (FIGO) stage in EOC patients. As for survival, circ-ABCB10 was correlated with worse OS. In vitro experiments revealed that circ-ABCB10 was upregulated and promoted cell proliferation but reduced cell apoptosis, and negatively regulated miR-1271, miR-1252 and miR-203 in EOC cells.

CONCLUSIONS:

Circ-ABCB10 correlates with advanced clinicopathological features and unfavorable survival, and promotes proliferation, reduces apoptosis and negatively regulated miR-1271, miR-1252 and miR-203 in EOC.

Keywords

Circ-ABCB10 EOC disease risk OS cell proliferation and apoptosis

1. Introduction

Epithelial ovarian cancer (EOC), accounting for 90% of all ovarian cancer cases, is the second most common gynecological cancers and the leading cause of gynecological cancer-related deaths [1]. EOC patients are commonly treated by standardized therapy including surgical resection and chemotherapy, and are always assisted by targeted therapy, endocrine therapy and radiotherapy [2]. Although therapeutic and diagnostic approaches for EOC have progressed and improved the clinical outcomes for EOC patients during the past decades, it remains a troublesome clinical challenge that EOC is commonly diagnosed at late-stage and is under high risk of recurrence and metastasis, resulting in an unfavorable prognosis in EOC patients [3]. In order to conquer these problems, accurate and reliable biomarkers are critical for prediction of disease risk and prognosis in EOC patients.

Of note, increasing evidences have revealed the close association of non-coding RNAs with the pathogenesis of human diseases [4]. one sort of the abundant and evolutionary conserved non-coding RNAs, circular RNAs (CircRNAs) are characterized by covalent loops formed by back-splicing without 3’ and 5’ ends [5]. They have been discovered to play key roles in various human diseases including cancers, cardiovascular diseases and autoimmune diseases through acting as microRNA (miRNA) sponges, protein sponges, protein scaffolds etc., and particularly in cancer, an increasing number of dysregulated circRNAs and their putative functions have been disclosed [5, 6, 7, 8, 9, 10]. CircRNA ABCB10 (circ-ABCB10) is a novel circRNA initially found to promote breast cancer proliferation and progression by sponging miR-1271, a miRNA that has been reported to participate in the development of various cancers including EOC, breast cancer, pancreatic cancer and so on [10, 11, 12, 13, 14]. And it is also observed to promote non-small-cell lung cancer cell proliferation and migration via regulating miR-1252/Forkhead box 2 (FOXR2) axis, and enhances osteosarcoma progression through sponging miR-203 [15, 16]. These studies indicate the oncogenic role of circ-ABCB10 in cancers, while the expression pattern and regulatory role of circ-ABCB10 in EOC is still unknown. Therefore, this study aimed to explore the correlation of circ-ABCB10 expression with clinicopathological features and survival, as well as its function on regulating cell proliferation and apoptosis in EOC.

2. Methods

2.1 Patients

Between Jan 2014 and Dec 2018, 103 EOC patients who underwent resection at Lishui People’s Hospital were reviewed in this retrospective study. The inclusion criteria were: (1) confirmed EOC by histopathological findings; (2) received resection, and the fresh-frozen tumor tissue was available; (3) with complete clinical and follow-up data; (4) no neoadjuvant therapy. The patients were excluded if they had relapsed or secondary EOC. Informed consents (paper version or electronic edition) were obtained from all patients or their guardians after the study was approval by the Ethics Committee of the Hospital.

2.2 Clinical data collection

Patients’ clinical data, which included age, histological subtype, pathological differentiation, tumor size, the volume of ascites, peritoneal cytology, FIGO stage and CA125 level were collected from electronic medical records. All patients were followed up to 2018/12/31, and the survival data were obtained from follow-up records, which was used to calculate overall survival (OS). The OS was defined as the time interval from resection to death.

2.3 Detection of circ-ABCB10 relative expression in EOC tumor and adjacent tissues

Circ-ABCB10 relative expression in fresh-frozen tumor tissues from 103 EOC patients was detected by Quantitative Real-time Reverse Transcription (qRT-PCR), meanwhile, 52 EOC adjacent tissues were collected as controls, and the circ-ABCB10 relative expression in the adjacent tissues was detected using qRT-PCR as well. All tumor tissues and adjacent tissues were collected from storage compartment after approval by the Pathology Department of the Hospital.

2.4 Cell source and culture

Human EOC cell lines OVCAR3, UWB1.289, SKOV3 and CAOV3 were purchased from American Type Culture Collection (ATCC) (Manassas, USA), while human normal ovarian epithelial cell line IOSE80 was purchased from BioVector NTCC (Beijing, China). OVCAR3, UWB1.289, SKOV3 and IOSE80 cell lines were cultured in 90% RPMI-1640 medium (Gibco, USA) and 10% fetal bovine serum (FBS) (Gibco, USA) under 95% air and 5% CO ${}_{2}$ condition at 37 ${{}^{\circ}}$ C, CAOV3 cell line was cultured in 90% DMEM medium (Gibco, USA) and 10% FBS (Gibco, USA) under 95% air and 5% CO ${}_{2}$ condition at 37 ${{}^{\circ}}$ C.

2.5 Detection of circ-ABCB10 relative expression in human EOC cell lines and normal ovarian epithelial cell line

Circ-ABCB10 relative expression was detected in OVCAR3, UWB1.289, SKOV3, CAOV3 and IOSE80 cell lines by RT-qPCR.

2.6 Cell transfection

The control overexpression, circ-ABCB10 overexpression, control shRNA and circ-ABCB10 shRNA plasmids were constructed by Shanghai GenePharma Company (Shanghai, China) using pEX-2 and pGPU6. Then these plasmids were transfected into SKOV3 cells and UWB1.289 cells using HilyMax (Dojindo, Japan) and were divided into NC ( $+$ ), ABCB10 ( $+$ ), NC ( $-$ ) and ABCB10 ( $-$ ) groups.

2.7 Cell proliferation and apoptosis

SKOV3 cell and UWB1.289 cell proliferation was detected by OD value at 0 h, 24 h, 48 h and 72 h post transfection using Cell Counting Kit-8 (Sigma, USA) according to the instructions of manufacturer. SKOV3 cell and UWB1.289 cell apoptosis rate was detected at 24 h post transfection using Annexin V-FITC Apoptosis Detection Kit (Sigma, USA) according to the instructions of manufacturer. Besides, in order to further confirm the apoptosis, cleaved caspase 3 (C-Caspase 3) and B-cell lymphoma-2 (Bcl-2) protein expressions were detected at 24 h post transfection using Western blot as well.

2.8 qRT-PCR process

The qRT-PCR was performed as follows: first, TRIzol reagent (Invitrogen, USA) was used to extract total RNA according to the manufacturer’s instructions; then the linear RNA was digested using the RNase R enzyme (Epicentre, USA); subsequently, cDNA was synthesized from the RNA using the reverse transcription kit (Qiagen, German); finally, ABI 7900HT Real-time PCR system (Applied Biosystems, USA) was used for PCR reaction by use of QuantiNova SYBR Green PCR Kit (Qiagen, German). GAPDH or U6 was used as normalized internal reference and the relative expression of circ-ABCB10 or miRNAs was calculated using the 2 ${}^{-\Delta\Delta\text{Ct}}$ method. The primers used in the PCR were as follows: circ-ABCB10: forward: 5’-GTG CTGATGACCCTTCCTCTG-3’, reverse: 5’-GGGAA TCCTGAGTGACTTTGGT-3’; GAPDH: forward: 5’-GAGTCCACTGGCGTCTTCAC-3’, reverse: 5’-ATC TTGAGGCTGTTGTCATACTTCT-3’; miR-1271: forward: 5’-ACACTCCAGCTGGGCTTGGCACCTAGC AAG-3’, reverse: 5’-TGTCGTGGAGTCGGCAATTC -3’; miR-1252: forward: 5’-ACACTCCAGCTGGGAG AAGGAAATTGAATT-3’, reverse: 5’-TGTCGTGGA GTCGGCAATTC-3’; miR-203: forward: 5’-ACACTC CAGCTGGGAGTGGTTCTTAACAGT-3’, reverse: 5’ -TGTCGTGGAGTCGGCAATTC-3’; U6, forward, 5’-CGCTTCGGCAGCACATATACTA-3’, reverse, 5’-AT GGAACGCTTCACGAATTTGC-3’.

2.9 Western blot process

Proteins were extracted from tissue using RIPA Lysis and Extraction Buffer (Thermo, USA) and were then quantified using Pierce ${}^{\text{TM}}$ BCA Protein Assay Kit (Thermo, USA). A fixed amount of protein samples (20 ug) were transferred onto NuPAGE ${}^{\text{TM}}$ 4–12% Bis-Tris Protein Gel (Invitrogen, USA) for electrophoresis, and transferred to Nitrocellulose filter membrane (Millipore, USA) afterwards. The membranes were blocked by BSA (Thermo, USA) for 2 hours and incubated under 4 ${}^{\circ}$ C with primary antibodies overnight. Following that, the membranes were incubated under 37 ${{}^{\circ}}$ C with secondary antibodies for 90 minutes. Finally, the protein bands on membranes were luminized by Pierce ${}^{\text{TM}}$ ECL Plus Western Blotting Substrate (Thermo, USA) and visualized on X-ray film (Kodak, USA). Antibodies used were listed in Table 1.

Table 1
Antibodies applied in Western Blot

Antibody	Company	Dilution
Primary antibody
Caspase-3 Rabbit mAb	CST (USA)	1:1000
Cleaved Caspase-3 Rabbit mAb	CST (USA)	1:1000
Bcl-2 Rabbit mAb	CST (USA)	1:1000
GAPDH Rabbit mAb	CST (USA)	1:1000
Secondary antibody
Goat Anti-Rabbit IgG H&L (HRP)	Abcam (UK)	1:5000

Figure 1.

Circ-ABCB10 expression in distinguishing tumor tissue and adjacent tissue in EOC patients. Circ-ABCB10 expression was higher in tumor tissue compared with adjacent tissue (A), and it presented good value in distinguishing EOC tumor tissue from adjacent tissue (B). Comparison of circ-ABCB10 expression between tumor tissue and adjacent tissue was performed using Wilcoxon rank-sum test. ROC curve and AUC were used to assess the value of circ-ABCB10 relative expression in distinguishing tumor tissue from adjacent tissue. $P<$ 0.05 was considered significant. EOC, epithelial ovarian cancer; ROC, receiver operating characteristic; AUC, area under the curve.

2.10 Statistical analysis

All data analyses were performed using SPSS 22.0 software (SPSS Inc., USA), and all figures were made with the use of GraphPad Prism 7.00 software (GraphPad Software, USA). Data were presented as mean value $\pm$ standard deviation, median (25 ${}^{\text{th}}$ –75 ${}^{\text{th}}$ quantiles) or count (percentage). Comparison of difference among groups was determined by One-way ANOVA followed by multiple comparisons test. Comparison of difference between two groups was determined by $t$ test, Chi-square test, Wilcoxon rank-sum test. The receiver operating characteristic (ROC) curve and area under the curve (AUC) were used to assess the value of circ-ABCB10 relative expression in distinguishing tumor tissue from adjacent tissue. The OS difference between two groups was determined by Kaplan-Meier method and Log-rank test. Factors affecting OS were determined by univariate and multivariate Cox’s proportional hazards regression model analyses. $P$ value $<$ 0.05 was considered significant.

3. Results

3.1 Patients’ baseline characteristics

The mean age of 103 EOC patients was 56.6 $\pm$ 13.1 years. Regarding histological subtype, there were 67 (65.0%) patients with serous subtype and 36 (35.0%) patients with other subtypes. As for pathological differentiation, 48 (46.6%) patients had well/moderate differentiation and 55 (53.4%) patients had poor differentiation. Besides, the numbers of patients in FIGO stage I/II and FIGO stage III/IV were 37 (35.9%) and 66 (64.1%) respectively. Other detailed clinicopathological characteristics were listed in Table 2.

Table 2
Baseline characteristics of EOC patients

Characteristics	EOC patients ( $N=$ 103)
Age (years)	56.6 $\pm$ 13.1
Histological subtype ( $n$ /%)
Serous	67 (65.0)
Others	36 (35.0)
Pathological differentiation ( $n$ /%)
Well/moderate	48 (46.6)
Poor	55 (53.4)
Tumor size ( $n$ /%)
$\leqslant$ 10 cm	53 (51.5)
$>$ 10 cm	50 (48.5)
The volume of ascites ( $n$ /%)
$\leqslant$ 100 mL	35 (34.0)
$>$ 100 mL	68 (66.0)
Peritoneal cytology ( $n$ /%)
Negative	38 (36.9)
Positive	46 (44.7)
Unknow	19 (18.4)
FIGO stage ( $n$ /%)
I/II	37 (35.9)
III/IV	66 (64.1)
CA125 level ( $n$ /%)
$<$ 1000 U/mL	71 (68.9)
$\geqslant$ 1000 U/mL	32 (31.1)

Data were presented as mean value $\pm$ standard deviation or count (percentage). EOC: Epithelial ovarian cancer; FIGO: International Federation of Gynecology and Obstetrics; CA125: carbohydrate antigen 125.

3.2 Circ-ABCB10 relative expression between tumor tissue and adjacent tissue

Circ-ABCB10 expression in tumor tissue was 2.017 (1.028–3.377), which was higher compared to adjacent tissue (0.883 (0.249–1.575)) ( $P<$ 0.001) (Fig. 1A). ROC curve illustrated that circ-ABCB10 was of good value in distinguishing tumor tissue from adjacent tissue (AUC $=$ 0.766, 95% CI: 0.690–0.842) (Fig. 1B).

3.3 Correlation of circ-ABCB10 expression with patients’ characteristics

EOC patients were divided into circ-ABCB10 high expression group ( $N=$ 52) and circ-ABCB10 low expression group ( $N=$ 51) according to the median value of circ-ABCB10 expression in EOC tumor tissues, and correlation of circ-ABCB10 relative expression with patients’ characteristics was analyzed and presented in Table 3. We observed that circ-ABCB10 high expression was correlated with poor differentiation ( $P=$ 0.039), large tumor size ( $P=$ 0.002) and advanced FIGO stage ( $P<$ 0.001).

Table 3
Correlation of circ-ABCB10 expression with clinicopathological characteristics

Characteristics	Circ-ABCB10		$P$ value
	expression ${}^{\text{a}}$
	High	Low
	( $N=$ 52)	( $N=$ 51)
Age ( $n$ /%)			0.711
$<$ 50 years	16 (53.3)	14 (46.7)
$\geqslant$ 50 years	36 (49.3)	37 (50.7)
Histological subtype ( $n$ /%)			0.369
Serous	36 (53.7)	31 (46.3)
Others	16 (44.4)	20 (55.6)
Pathological differentiation ( $n$ /%)			0.039
Well/moderate	19 (39.6)	29 (60.4)
Poor	33 (60.0)	22 (40.0)
Tumor size ( $n$ /%)			0.002
$\leqslant$ 10 cm	19 (39.8)	34 (64.2)
$>$ 10 cm	33 (66.0)	17 (34.0)
The volume of ascites ( $n$ /%)			0.052
$\leqslant$ 100 mL	13 (37.1)	22 (62.9)
$>$ 100 mL	39 (57.4)	29 (42.6)
Peritoneal cytology ( $n$ /%)			0.600
Negative	22 (57.9)	16 (42.1)
Positive	24 (52.2)	22 (47.8)
FIGO stage ( $n$ /%)			$<$ 0.001
I/II	5 (13.5)	32 (86.5)
III/IV	47 (71.2)	19 (28.8)
CA125 level ( $n$ /%)			0.432
$<$ 1000 U/mL	34 (47.9)	37 (52.1)
$\geqslant$ 1000 U/mL	18 (56.2)	14 (43.8)

Data were presented as count (percentage). Comparisons were determined by Chi-square test. $P$ value $<$ 0.05 was considered significant. ${}^{\text{a}}$ : The high or low expression was classified according to the median value of circ-ABCB10 relative expression in EOC tumor tissues. FIGO: International Federation of Gynecology and Obstetrics; CA125: carbohydrate antigen 125; EOC: Epithelial ovarian cancer.

3.4 Correlation of circ-ABCB10 expression with patients’ OS

The accumulating OS in circ-ABCB10 high expression group (mean: 42.6 months, 95% CI: 36.2–49.0 months) was reduced compared with circ-ABCB10 low expression group (mean: 52.5 months, 95% CI: 48.8–56.1 months) ( $P=$ 0.007) in EOC patients (Fig. 2).

Figure 2.

Comparison of OS between circ-ABCB10 high expression and low expression group in EOC patients. OS was lower in circ-ABCB10 high expression group compared with circ-ABCB10 low expression group. The difference of OS between two groups was determined by Kaplan-Meier method and Log-rank test. $P<$ 0.05 was considered significant. OS, overall survival.

3.5 Factors affecting OS

Univariate Cox’s regression model analysis displayed that high circ-ABCB10 expression (HR $=$ 2.994, $P=$ 0.011), poor differentiation (HR $=$ 2.458, $P=$ 0.037) and large volume of ascites (HR $=$ 4.301, $P=$ 0.005) were correlated with shorter OS in EOC patients (Table 4). Further multivariate Cox’s regression model analysis revealed that circ-ABCB10 expression (HR $=$ 2.257, $P=$ 0.152) was not an independent factor for worse OS, while poor differentiation (HR $=$ 9.101, $P=$ 0.002), large tumor size (HR $=$ 4.249, $P=$ 0.009), high volume of ascites (HR $=$ 11.135, $P<$ 0.001), peritoneal cytology test positive (HR $=$ 4.834, $P=$ 0.015) and advanced FIGO stage (HR $=$ 16.305, $P=$ 0.001) were independent predictive factors for reduced OS in EOC patients. These data indicated circ-ABCB10 might affect the OS via the interaction or correlation with other tumor features such as differentiation, tumor size and FIGO stage.

Table 4
Cox’s regression model analysis of factors affecting OS

Items	Cox’s regression model
	$P$ value	HR	95% CI
			Lower	Higher
Univariate Cox’s regression
Circ-ABCB10 (high vs. low)	0.011	2.994	1.290	6.947
Age ( $\geqslant$ 50 years vs. $<$ 50 years)	0.711	0.850	0.360	2.007
Histological subtype (serous vs. others)	0.839	1.094	0.461	2.597
Pathological differentiation (poor vs. moderate/well)	0.037	2.458	1.055	5.723
Tumor size ( $>$ 10 cm vs. $\leqslant$ 10 cm)	0.089	2.053	0.896	4.704
The volume of ascites ( $>$ 100 mL vs. $\leqslant$ 100 mL)	0.005	4.301	1.557	11.881
Peritoneal cytology (positive vs. negative)	0.146	1.956	0.792	4.828
FIGO stage (IV/III vs. II/I)	0.080	2.255	0.908	5.600
CA125 level ( $\geqslant$ 1000 U/mL vs. $<$ 1000 U/mL)	0.255	1.632	0.702	3.791
Multivariate Cox’s regression
Circ-ABCB10 (high vs. low)	0.152	2.257	0.742	6.868
Age ( $\geqslant$ 50 years vs. $<$ 50 years)	0.885	0.922	0.307	2.765
Histological subtype (serous vs. others)	0.746	1.200	0.398	3.616
Pathological differentiation (poor vs. moderate/well)	0.002	9.101	2.318	35.731
Tumor size ( $>$ 10 cm vs. $\leqslant$ 10 cm)	0.009	4.249	1.433	12.604
The volume of ascites ( $>$ 100 mL vs. $\leqslant$ 100 mL)	$<$ 0.001	11.135	3.153	39.331
Peritoneal cytology (positive vs. negative)	0.015	4.834	1.363	17.143
FIGO stage (IV/III vs. II/I)	0.001	16.305	2.983	89.119
CA125 level ( $\geqslant$ 1000 U/mL vs. $<$ 1000 U/mL)	0.980	0.984	0.272	3.555

Factors affecting OS were determined by univariate and multivariate Cox’s proportional hazards regression model analyses. $P$ value $<$ 0.05 was considered significant. FIGO: International Federation of Gynecology and Obstetrics; CA125: carbohydrate antigen 125; CI: confidence interval; HR: hazard ratio.

3.6 Circ-ABCB10 expression in human EOC cell lines and normal ovarian epithelial cells

In vitro experiments were further carried out to investigate the function of circ-ABCB10 in EOC. It was observed that circ-ABCB10 expression was higher in OVCAR3 ( $P<$ 0.05), SKOV3 ( $P<$ 0.001) and CAOV3 ( $P<$ 0.001) cell lines but not in UWB1.289 ( $P>$ 0.05) cells compared with normal ovarian epithelial cells IOSE80 (Fig. 3).

Figure 3.

Circ-ABCB10 relative expression in human EOC cell lines and normal ovarian epithelial cells. Compared with normal ovarian epithelial cells IOSE80, the expression of circ-ABCB10 was elevated in OVCAR3, SKOV3 and CAOV3 cell lines but not in UWB1.289 cells. Comparison of circ-ABCB10 expression between two cell lines were performed by $t$ test. $P<$ 0.05 was considered significant. ${}^{*}P<$ 0.05, ${}^{***}P<$ 0.001, NS, non-significant. EOC, epithelial ovarian cancer.

3.7 Effect of circ-ABCB10 on cell proliferation and apoptosis in SKOV3 cells and UWB1.289 cells

In SKOV3 cells, circ-ABCB10 expression was increased in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group ( $P<$ 0.001) but decreased in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group ( $P<$ 0.001), suggesting the successful transfection (Fig. 4A). Cell proliferation was similar at 24 h ( $P>$ 0.05), while promoted at 48 h ( $P<$ 0.05) and 72 h ( $P<$ 0.01) after transfection in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group; cell proliferation was similar at 24 h ( $P>$ 0.05), but reduced at 48 h ( $P<$ 0.05) and 72 h ( $P<$ 0.05) after transfection in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group in SKOV3 cells (Fig. 4B). As for cell apoptosis, it was decreased in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group ( $P<$ 0.01) but increased in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group ( $P<$ 0.001) (Fig. 4C and E). Western Blot assay revealed that C-caspase 3 expression was reduced in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group while elevated in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group; conversely, Bcl-2 expression was increased in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group but decreased in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group in SKOV3 cells (Fig. 4D). In addition, the effect of circ-ABCB10 on cell proliferation (Fig. 5A and B) and apoptosis (Fig. 5C and D) was further validated in UWB1.289 cells, which displayed that circ-ABCB10 promoted cell proliferation and suppressed cell apoptosis in UWB1.289 cells as well.

Figure 4.

Circ-ABCB10 expression, cell proliferation and apoptosis after transfection in SKOV3 cells. Circ-ABCB10 relative expression was elevated by circ-ABCB10 overexpression transfection and lowered by circ-ABCB10 silencing transfection (A). Cell proliferation was promoted by circ-ABCB10 overexpression transfection and reduced by circ-ABCB10 silencing transfection (B). Cell apoptosis was reduced by circ-ABCB10 overexpression transfection and increased by circ-ABCB10 silencing transfection (C and E). C-caspase 3 expression was lowered by circ-ABCB10 overexpression transfection and raised by circ-ABCB10 silencing transfection; Bcl-2 expression was increased by circ-ABCB10 overexpression transfection and decreased by circ-ABCB10 silencing transfection. Comparison of circ-ABCB10 expression, cell proliferation and apoptosis between two groups were performed by $t$ test. $P<$ 0.05 was considered significant. ${}^{*}P<$ 0.05, ${}^{**}P<$ 0.01, ${}^{***}P<$ 0.001.

Figure 5.

Circ-ABCB10 expression, cell proliferation and apoptosis after transfection in UWB1.289 cells. Circ-ABCB10 expression was elevated by circ-ABCB10 overexpression transfection and suppressed by circ-ABCB10 silencing transfection (A). Cell proliferation was increased by circ-ABCB10 overexpression transfection and decreased by circ-ABCB10 silencing transfection (B). Cell apoptosis was reduced by circ-ABCB10 overexpression transfection and promoted by circ-ABCB10 silencing transfection (C and D). Comparison of circ-ABCB10 expression, cell proliferation and apoptosis between two groups were performed by t test. $P<$ 0.05 was considered significant. ${}^{*}P<$ 0.05, ${}^{**}P<$ 0.01, ${}^{***}P<$ 0.001.

Figure 6.

MiR-1271, miR-1252 and miR-203 expression after transfection in SKOV3 cells. MiR-1271 (A), miR-1252 (B) and miR-203 (C) expressions were suppressed by circ-ABCB10 overexpression transfection but promoted by circ-ABCB10 silencing transfection. Comparison of miRNA expression between two groups was performed by $t$ test. $P<$ 0.05 was considered significant. ${}^{**}P<$ 0.01, ${}^{***}P<$ 0.001.

Figure 7.

MiR-1271, miR-1252 and miR-203 expression after transfection in UWB1.289 cells. MiR-1271 (A), miR-1252 (B) and miR-203 (C) expressions were decreased by circ-ABCB10 overexpression transfection but increased by circ-ABCB10 silencing transfection. Comparison of miRNA expression between two groups was performed by $t$ test. $P<$ 0.05 was considered significant. ${}^{*}P<$ 0.05, ${}^{**}P<$ 0.01, ${}^{***}P<$ 0.001.

3.8 Effect of circ-ABCB10 on miR-1271, miR-1252 and miR-203 in SKOV3 cells and UWB1.289 cells

MiR-1271, miR-1252 and miR-203 that were previously shown as target miRNAs of circ-ABCB10 were selected, and the regulatory effect of circ-ABCB10 on these miRNAs in SKOV3 cells and UWB1.289 cells was investigated. In SKOV3 cells, miR-1271 (Fig. 6A), miR-1252 (Fig. 6B) and miR-203 (Fig. 6C) expressions were downregulated in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group, but upregulated in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group (All $P<$ 0.05). And in UWB1.289 cells, miR-1271 (Fig. 7A), miR-1252 (Fig. 7B) and miR-203 (Fig. 7C) levels were reduced in ABCB10 ( $+$ ) group compared with NC ( $+$ ) group but elevated in ABCB10 ( $-$ ) group compared with NC ( $-$ ) group (All $P<$ 0.05).

4. Discussion

Our study disclosed that in EOC patients, (1) circ-ABCB10 was upregulated in tumor tissues and correlated with advanced clinicopathological features; (2) circ-ABCB10 was correlated with unfavorable OS but not an independent predictive factor; (3) Circ-ABCB10 was upregulated, and it promoted proliferation but reduced cell apoptosis as well as negatively regulated miR-1271, miR-1252 and miR-203 in EOC cells.

Different from linear RNAs, circRNAs are considered to function independent of their host genes via multiple means, such as acting as miRNA sponges, protein sponges or decoys etc, and they have been implicated in several hallmarks in cancers due to these unique properties [1]. For instance, circ-SMARCA5 is downregulated in tumor tissues and is negatively correlated with aggressive clinical characteristics in hepatocellular carcinoma (HCC) patients [17]. Conversely, circ-BPTF expression is increased in bladder cancer tissues and it predicts higher tumor grades in bladder cancer patients [18]. In addition, the expression of circ_0067934 is remarkably higher in tumor tissues compared with paired adjacent normal tissues, and is associated with poor disease stage in esophageal squamous cell carcinoma patients [19]. These previous studies have implied the aberrant expression and the close correlation of several circRNAs with human cancers. As for circ-ABCB10, it has been observed to be significantly upregulated in breast cancer tissues and plays oncogenic role in NSCLC as well as osteosarcoma, whereas its role in EOC is still obscure [10, 15, 16]. Therefore, in this study, we investigated the function of circ-ABCB10 in EOC and discovered that it was upregulated in tumor tissue and was correlated with poor pathological differentiation, large tumor size and advanced FIGO stage in EOC patients. This might result from that (1) circ-ABCB might sponge miRNAs such as miR-221-3p and miR-1276 (retrieved from Tissue-Specific CircRNA Database, http://gb.whu.edu.cn/TSCD/) that inhibited cancer cell proliferation and increased cell apoptosis, thereby promoted cancer progression and led to worse clinicopathological features in EOC patients. (2) Circ-ABCB might influence the function of its host gene ABCB10 and alter EOC cell development, which eventually resulted in increased disease risk and advanced disease stage in EOC patients. However, this needed verification by further studies.

Although the aberrant expression of a number of circRNAs and their correlations with clinicopathological status in several human cancers have been revealed, the prognostic roles of circRNAs, especially their predictive value on survival in EOC patients are sparsely reported. One previous study illustrates that circ-ITCH independently predicts favorable OS in EOC patients [20]. Another study performs circRNA profiling and screens out that circ-EXOC6B and circ-N4BP2L2 are correlated with longer OS and progression free survival in EOC patients [21]. As for circ-ABCB10, its prognostic role has not been reported in any cancer researches except for osteosarcoma, in which high circ-ABCB is associated with shortened survival [16]. In our study, we first evaluated the predictive role of circ-ABCB10 on survival in EOC patients and disclosed that circ-ABCB10 was associated with unfavorable OS in EOC patients, but was not an independent predictive factor for OS. This might be attributed to that: (1) circ-ABCB10 might facilitate cancer cell proliferation and inhibit apoptosis, promoted cancer progression and higher disease stage, which brought in poor OS in EOC patients. The effect of circ-ABCB10 on cancer cell proliferation and apoptosis was validated in the following in vitro experiments. (2) Circ-ABCB10 might alter the sensitivity to drugs in EOC cells, reducing the treatment responses and worsening OS in EOC patients. (3) Circ-ABCB10 might affect survival via interacting or correlating with other tumor features such as tumor size, differentiation and FIGO stage, thus, it was not an independent predictive factor for OS. Other than that, patients’ characteristics including poor differentiation, large tumor size, high volume of ascites, peritoneal cytology test positive and advanced FIGO stage were independent predictive factors for shorter OS in EOC patients. The possible reason might be that advanced disease stage might reduce patients’ compliance to treatment, attenuated the treatment efficacy and increase the risk of recurrence, hence worsened the survival in EOC patients.

It is considered that circRNAs take part in tumorigenesis mainly via altering cancer cell activities by absorbing the functional miRNAs in the cytoplasm [6]. In line with that, circ-ABCB10 promotes NSCLC cell proliferation and migration by sponging miR-1252 and increasing FOXR2 levels [15]. Besides, circ-ABCB10 sponges miR-203 in osteosarcoma, and knockdown of circ-ABCB10 suppresses cell proliferation, migration and invasion but promotes cell apoptosis [16]. n addition, circ-ABCB10 is also discovered to increase cell proliferation and suppress cell apoptosis in breast cancer via directly binding miR-1271 [10]. These data suggest that circ-ABCB10 participate in carcinogenesis of NSCLC, osteosarcoma and breast cancer via regulating cell activities, whereas the cellular influence of circ-ABCB10 in EOC is still obscure. Known from our findings that circ-ABCB10 is correlated with increased disease risk, poor clinicopathological features and unfavorable survival in EOC patients, in vitro experiments were conducted to further validate the function of circ-ABCB10 in EOC. We observed that circ-ABCB10 was upregulated in EOC cell lines, and it promoted cell proliferation but inhibited cell apoptosis. The possible reason was that circ-ABCB10 might serve as sponge to tumor suppressive miRNAs, such as miR-1252, miR-1271, and miR-203, as previously reported, in the cytoplasm and influence the downstream protein expression, which subsequently promoted cell proliferation and inhibited cell apoptosis in EOC. This explanation was further validated by our following experiments, which disclosed that circ-ABCB10 negatively regulated miR-1252, miR-1271, and miR-203 levels in both SKOV3 cells and UWB1.289 cells.

In conclusion, circ-ABCB10 correlates with advanced clinicopathological features and unfavorable survival in EOC patients, and it promotes proliferation, reduces apoptosis and negatively regulates miR-1271, miR-1252 and miR-203 in EOC.

Footnotes

Conflict of interest

No potential conflict of interest was reported by the authors.

References

Jayson

G.C.

Kohn

E.C.

Kitchener

H.C.

and Ledermann

J.A.

, Ovarian cancer, Lancet 384 (2014), 1376–1388.

Bolton

K.L.

Chenevix-Trench

Goh

Sadetzki

Ramus

S.J.

Karlan

B.Y.

Lambrechts

Despierre

Barrowdale

McGuffog

Healey

Easton

D.F.

Sinilnikova

Benitez

Garcia

M.J.

Neuhausen

Gail

M.H.

Hartge

Peock

Frost

Evans

D.G.

Eeles

Godwin

A.K.

Daly

M.B.

Kwong

E.S.

Lazaro

Blanco

Montagna

D’Andrea

Nicoletto

M.O.

Johnatty

S.E.

Kjaer

S.K.

Jensen

Hogdall

Goode

E.L.

Fridley

B.L.

Loud

J.T.

Greene

M.H.

Mai

P.L.

Chetrit

Lubin

Hirsh-Yechezkel

Glendon

Andrulis

I.L.

Toland

A.E.

Senter

Gore

M.E.

Gourley

Michie

C.O.

Song

Tyrer

Whittemore

A.S.

McGuire

Sieh

Kristoffersson

Olsson

Borg

Levine

D.A.

Steele

Beattie

M.S.

Chan

Nussbaum

R.L.

Moysich

K.B.

Gross

Cass

Walsh

A.J.

Leuchter

Gordon

Garcia-Closas

Gayther

S.A.

Chanock

S.J.

Antoniou

A.C.

Pharoah

P.D.

, Embrace kConFab

and N. Cancer Genome Atlas Research, Association between BRCA1 and BRCA2 mutations and survival in women with invasive epithelial ovarian cancer, JAMA 307 (2012), 382–390.

Coleman

R.L.

Monk

B.J.

Sood

A.K.

and Herzog

T.J.

, Latest research and treatment of advanced-stage epithelial ovarian cancer, Nat Rev Clin Oncol 10 (2013), 211–224.

Bachmayr-Heyda

Reiner

A.T.

Auer

Sukhbaatar

Aust

Bachleitner-Hofmann

Mesteri

Grunt

T.W.

Zeillinger

and Pils

, Correlation of circular RNA abundance with proliferation-exemplified with colorectal and ovarian cancer, idiopathic lung fibrosis, and normal human tissues, Sci Rep 5 (2015), 8057.

Zhao

Gao

Jian

Hao

Rao

and Li

, Peripheral blood circular RNA hsa_circ_0124644 can be used as a diagnostic biomarker of coronary artery disease, Sci Rep 7 (2017), 39918.

Kristensen

L.S.

Hansen

T.B.

Veno

M.T.

and Kjems

, Circular RNAs in cancer: opportunities and challenges in the field, Oncogene 37 (2018), 555–565.

Wang

Gong

Yang

Zhang

Xiong

Xiang

Zhou

Liao

Zhang

Zeng

and Xiong

, Circular RNAs in human cancer, Mol Cancer 16 (2017), 25.

Liu

Zhang

Qin

Yang

Xiong

Zhang

Wang

Zhao

Wang

and Liu

, Circular RNA EIF6 (Hsa_circ_0060060) sponges miR-144-3p to promote the cisplatin-resistance of human thyroid carcinoma cells by autophagy regulation, Aging (Albany NY) 10 (2018), 3806–3820.

Zhang

Wang

Chen

Zhang

and Shi

, The down-regulation of hsa_circ_0012919, the sponge for miR-125a-3p, contributes to DNA methylation of CD11a and CD70 in CD4 (+) T cells of systemic lupus erythematous, Clin Sci (Lond) 132 (2018), 2285–2298.

10.

Liang

H.F.

Zhang

X.Z.

Liu

B.G.

Jia

G.T.

and Li

W.L.

, Circular RNA circ-ABCB10 promotes breast cancer proliferation and progression through sponging miR-1271, Am J Cancer Res 7 (2017), 1566–1576.

11.

Liu

Rao

Mao

Xin

and Wang

, MiR-1271 inhibits ovarian cancer growth by targeting cyclin G1, Med Sci Monit 21 (2015), 3152–3158.

12.

Wang

Gao

and Teng

, MiR-1271 inhibits cell growth in prostate cancer by targeting ERG, Pathol Oncol Res 24 (2018), 385–391.

13.

Xie

Huang

Liu

C.H.

Lin

X.S.

Liu

L.L.

Huang

D.W.

and Zhou

H.C.

, MiR-1271 negatively regulates AKT/MTOR signaling and promotes apoptosis via targeting PDK1 in pancreatic cancer, Eur Rev Med Pharmacol Sci 22 (2018), 678–686.

14.

Lim

Kim

H.J.

Heo

Huh

Baek

S.J.

Kim

J.H.

Bae

D.H.

Seo

E.H.

Lee

S.I.

Song

K.S.

Kim

S.Y.

Kim

Y.S.

and Kim

, Epigenetic silencing of miR-1271 enhances MEK1 and TEAD4 expression in gastric cancer, Cancer Med (2018).

15.

Tian

Zhang

Jiao

Chen

Shan

and Yang

, CircABCB10 promotes nonsmall cell lung cancer cell proliferation and migration by regulating the miR-1252/FOXR2 axis, J Cell Biochem 120 (2019), 3765–3772.

16.

Zhou

Natino

Qin

Wang

Tian

Cai

Wang

and He

, Identification and functional characterization of circRNA-0008717 as an oncogene in osteosarcoma through sponging miR-203, Oncotarget 9 (2018), 22288–22300.

17.

Q.G.

Wang

Z.G.

Yang

Zhang

J.Z.

Sun

S.H.

Yang

and Zhou

W.P.

, Circular RNA cSMARCA5 inhibits growth and metastasis in hepatocellular carcinoma, J Hepatol 68 (2018), 1214–1227.

18.

Liu

Cai

Dong

Jiang

Yang

Huang

and Lin

, Circ-BPTF promotes bladder cancer progression and recurrence through the miR-31-5p/RAB27A axis, Aging (Albany NY) 10 (2018), 1964–1976.

19.

Bao

Zhong

and Pang

, Upregulation of circular RNA VPS13C-has-circ-001567 promotes ovarian cancer cell proliferation and invasion, Cancer Biother Radiopharm (2018).

20.

Luo

Gao

and Sun

, Circ-ITCH correlates with small tumor size, decreased FIGO stage and prolonged overall survival, and it inhibits cells proliferation while promotes cells apoptosis in epithelial ovarian cancer, Cancer Biomark 23 (2018), 505–513.

21.

Ning

Long

Zhang

Wang

Cao

Yang

Shen

Huang

and Lang

, Circular RNA profiling reveals circEXOC6B and circN4BP2L2 as novel prognostic biomarkers in epithelial ovarian cancer, Int J Oncol 53 (2018), 2637–2646.

Circular RNA ABCB10 correlates with advanced clinicopathological features and unfavorable survival,and promotes cell proliferation while reduces cell apoptosis in epithelial ovarian cancer

Abstract

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Methods

2.1 Patients

2.2 Clinical data collection

2.3 Detection of circ-ABCB10 relative expression in EOC tumor and adjacent tissues

2.4 Cell source and culture

2.5 Detection of circ-ABCB10 relative expression in human EOC cell lines and normal ovarian epithelial cell line

2.6 Cell transfection

2.7 Cell proliferation and apoptosis

2.8 qRT-PCR process

2.9 Western blot process

Table 1 Antibodies applied in Western Blot

3. Results

3.1 Patients’ baseline characteristics

Table 2 Baseline characteristics of EOC patients

3.3 Correlation of circ-ABCB10 expression with patients’ characteristics

Table 3 Correlation of circ-ABCB10 expression with clinicopathological characteristics

Table 4 Cox’s regression model analysis of factors affecting OS

4. Discussion

Footnotes

Conflict of interest

References

Table 1
Antibodies applied in Western Blot

Table 2
Baseline characteristics of EOC patients

Table 3
Correlation of circ-ABCB10 expression with clinicopathological characteristics

Table 4
Cox’s regression model analysis of factors affecting OS