Circular RNA PVT1 promotes the invasion and epithelial-mesenchymal transition of breast cancer cells through serving as a competing endogenous RNA for miR-204-5p.

PURPOSE: Circular RNAs (circRNAs) are a class of covalently closed circular RNA transcripts and have been found to regulate the progression of human malignancies. The objective of this study was to identify the functions of circRNAs in breast cancer (BCa). PATIENTS AND METHODS: BCa and adjacent non-cancerous tissues were collected from 99 patients. Kaplan-Meier analysis was used to analyze the relationship between circPVT1 and prognosis. CircPVT1 expression levels in BCa tissues and cell lines were detected via PCR. Transfection technology was used to silence circPVT1 and overexpress miR-204-5p. Cell biological behavior was analyzed, and epithelial-mesenchymal transition (EMT) related proteins were detected by Western blot. In vivo experiments were performed with the subcutaneous xenograft tumor model.
RESULTS: CircPVT1 was markedly overexpressed in BCa tissues and cell lines. Higher expression of circPVT1 was correlated with poor prognosis of BCa patients. Knockdown circPVT1 significantly suppressed the proliferation, migration and invasion of BCa cells invitro, and suppressed BCa tumor growth invivo. CircPVT1 knockdown upregulated E-cadherin and downregulated N-cadherin, Vimentin, Slug and Twist in BCa cells. Moreover, circPVT1 could serve as a competing endogenous RNA (ceRNA) for miR-204-5p, and restoration of miR-204-5p abrogated the oncogenic role of circPVT1 in BCa cells.
CONCLUSION: CircPVT1 as a potentially valuable biomarker for BCa diagnosis and therapeutic target for BCa treatment. CircPVT1 might promote the invasion and EMT of BCa cells by serving as aceRNA for miR-204-5p.

Introduction

Breast cancer (BCa) remains the most prevalent female malignancy and the leading cause of female mortality worldwide.1 As high as 12.2% of the global BCa cases and 9.6% of the associated mortalities occur in China.2 Although significant developments have been made in the diagnostic and therapeutic methods, the long-term survival of BCa patients still remains largely unfavorable. Therefore, there is of critical importance to elucidate the potential mechanisms underlying BCa progression in order to facilitate identification of novel therapeutic targets.

Circular RNAs (circRNAs), a novel class of endogenous noncoding RNAs, form a covalently closed continuous loop without a 5ʹ cap or a 3ʹ Poly A tail, which make themselves more stable than linear RNAs.3,4 Recently, emerging studies have discovered that thousands of endogenous circRNAs are present in mammalian cells, and they play critical roles in the regulation of multiple human malignancies, including BCa.5 For example, in BCa, increased expression of a circular RNA circDNMT1 was detected,6 and circIRAK3 promoted cell migration, invasion and metastasis.7

circPVT1, one circRNA derived from the PVT1 gene, is upregulated in GC tissues and may promote GC cell proliferation.8 Besides, knockdown of circPVT1 could weaken the resistance to doxorubicin and cisplatin of osteosarcoma cells.9 In this study, we aimed to evaluate the potential role of circPVT1 in BCa, and we believed our findings will extend our understanding of the role of circRNAs in BCa.

Materials and methods

Patients and tissue samples

BCa and adjacent non-cancerous tissues were collected from 99 patients with BCa who underwent surgery at Cangzhou Central Hospital of Hebei Province (Cangzhou, China). No chemotherapy or radiotherapy was conducted in these patients prior to surgery. The clinicopathological information of these patients is listed in Table 1. Overall survival was defined as the time from the date of surgery to mortality or the latest date when censored. Tissues specimens were obtained during operation and immediately frozen at −80°C. The study was approved by the Ethics Committee of Cangzhou Central Hospital and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all the patients or their relatives.

Table 1

Relationship between circPVT1 expression and clinicopathological characteristics of BCa patients

Characteristics	Total number (n=99)	circPVT1 expression		P-value
		Low (n=47)	High (n=52)
Age (years)				0.401
<50	42	22	20
≥50	57	25	32
Tumor size (cm)				0.079
<2.5	65	35	30
≥2.5	34	12	22
Histology grade				0.141
Well-Moderate	71	37	34
Poor	28	10	18
TNM stage				0.012
I–II	61	35	26
III–IV	38	12	26
Lymph nodes metastasis				0.166
Positive	43	17	26
Negative	56	30	26
ER status				0.679
Positive	59	27	32
Negative	40	20	20
PR status				0.508
Positive	45	23	22
Negative	54	24	30
HER2 status				0.339
Positive	54	28	26
Negative	45	19	26

Abbreviations: ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal receptor 2.

Cell culture and transfection

Three BCa cell lines, including MDA-MB-231, MDA-MB-468 and MCF-7, were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT, USA) and antibiotics (100 U/mL penicillin and 100 mg/mL streptomycin) (Invitrogen). The normal human breast epithelial cells (MCF-10A) were cultured in RPMI-1640 medium (Thermo Fisher Scientific, Inc.). All the cell lines were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and cultured at 37°C in an atmosphere containing 5% CO2.

The sequences of short-hairpin RNAs (shRNAs) directed against circPVT1 or negative control were ligated into the pLKO.1-Puro vector (TaKaRa, Dalian, China). Lentivirus was packaged into HEK293 cells using Lipofectamine 2000 (Invitrogen) and collected from the supernatant. The cells were then infected with lentiviral particles and selected with puromycin for four weeks. The synthesized circPVT1 gene fragment was inserted into the pcDNA3.1(+) vector (Invitrogen) to overexpress circPVT1 in BCa cells. An empty vector was used as a negative control. The stably overexpressing or depleting cell lines were further validated by RT-qPCR analysis.

miR-204-5p mimics and the scrambled oligonucleotides (NC) were designed and synthesized by Shanghai GenePharma Co., Ltd. (Shanghai, China). Cells in 6-well dishes were transfected with the oligonucleotides using Lipofectamine 2000.

RNA extraction and RT-qPCR analysis

Total RNA was isolated from tissue samples or cultured cells using TRIzol reagent (Invitrogen). The RNA samples were reverse transcribed to complementary DNA (cDNA) using the PrimeScript™ RT reagent kit (TaKaRa). qPCR analysis was then performed using SYBR Green Master Mix (Applied Biosystems, Foster City, CA, USA) on an ABI PRISM 7500 fast Sequence Detection System (Applied Biosystems). Relative gene expression was calculated using 2−ΔΔCt method,10 with GAPDH or U6 as an internal control.

Western blot analysis

Total protein was extracted using the Whole Protein Extraction kit (Nanjing KeyGen Biotech Co., Ltd., Nanjing, China). The protein concentration was quantified using an Enhanced BCA Protein Assay kit (Beyotime, Shanghai, China). Identical quantities of proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred on a polyvinylidene fluoride membrane (Millipore, Billerica, MA, USA). After blocking with 5% non-fat milk, the membranes were incubated overnight at 4°C with primary antibodies against Bcl-2 (1:1,000; Abcam, Cambridge, MA, USA), Bax (1:1,000; Cell Signaling Technology, Danvers, MA, USA), E-cadherin (1:2,000; Abcam), N-cadherin (1:2,000; Abcam), Vimentin (1:2,000; Abcam), Slug (1:1,000; Cell Signaling Technology), Twist (1:1,000; Abcam) or GAPDH (1:1,000; Cell Signaling Technology). The protein bands were then probed with HRP-conjugated secondary antibodies for 1 hr at room temperature, visualized using an enhanced chemiluminescence reagent (Millipore) and quantified using Image J 3.0 software (). GAPDH was used as an endogenous loading control.

MTT assay

Cell proliferation was evaluated by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. In brief, after transfection, cells were seeded in 96-well plate at a density of 2×103 cells per well. At indicated time points, 20 μL of MTT (5 mg/mL in PBS; Sigma, St. Louis, MO, USA) was added to each well and the cells were incubated for another 4 hrs at 37°C. Then, the medium was removed and 150 µL of DMSO (Sigma) was added to each well to dissolve the formazan crystals. The absorbance was determined at 570 nm using an ELISA reader (Molecular Devices, Sunnyvale, CA, USA).

Colony formation assay

For colony formation assay, cells after transfection were seeded on 6-well plates at a density of 500 cells/well. The culture medium was replaced every 3 days. After 12 days, the colonies were fixed with 4% paraformaldehyde for 30 mins and then stained with 0.5% crystal violet. Then, the number of colonies was counted manually.

Apoptosis analysis by flow cytometry

Cell apoptosis was quantified using Annexin V-FITC/PI Apoptosis Detection kit (Beyotime). Briefly, after transfection, cells were collected by trypsinization and washed with ice-cold PBS. Then, cells were stained with 5 µL of Annexin V-FITC and 10 µL of propidium iodide (PI) in dark at room temperature for 20 mins. The samples were then analyzed using FACScan flow cytometer (BD Biosciences, Franklin lakes, NJ, USA).

Wound healing assay

Cells were cultured until they reach 90% confluence in 6-well plates. Wounds were created by scratching cell layer with a sterilized 200 μL pipette tip, and the wells were washed with PBS to remove cell debris. After 48 hrs, the wounds were photographed, and the percentage of wound closure was calculated.

Transwell assay

A total of 5×104 cells were resuspended in 200 μL of serum-free culture medium and seeded in the upper transwell chamber insert (Corning Inc., Corning, NY, USA) coated (for invasion analysis) or uncoated (for migration analysis) with Matrigel (BD Biosciences). The lower chamber was filled with 600 µL medium containing 10% FBS. After 48 hrs of incubation, the cells remaining on the top membrane were removed, and the cells that passed through the filter were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet. The number of stained cells was counted under a microscope.

Dual-luciferase reporter assay

The fragment of circPVT1 containing predicted miR-204-5p binding site was amplified by PCR and constructed into the psiCHECK-2 luciferase reporter vector (Promega, Madison, WI, USA). The binding site was mutated using a Mut Express II Fast Mutagenesis kit (Vazyme, Piscataway, NJ, USA). HEK293T cells were seeded in 96-well plates and co-transfected with the recombinant vectors and miR-204-5p mimics or NC using Lipofectamine 2000. Forty-eight hours after transfection, the luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega).

Tumorigenesis in nude mice

Twenty six-week-old male athymic BALB/c nude mice, purchased from Shanghai SLAC Laboratory Animal, Co., Ltd. (Shanghai, China), were kept under pathogen-free condition. These mice were randomized to two groups (n=10/group). Stably transfected cells were resuspended in serum-free medium at a concentration of 2×106 cells/0.2 mL and subcutaneously injected into a single side of the posterior flank of each mouse. Tumor growth was recorded every three days using digital calipers and tumor volumes were calculated according to the formula: volume=0.5×Length×Width2. All mice were sacrificed four weeks after cell inoculation, and the tumors were weighted. All animal manipulations were approved by the Ethics Committee of Cangzhou Central Hospital, and great efforts were given to reduce the suffering of animals according to the NIH Guide for the Care and Use of Laboratory Animals.

Statistical analysis

All statistical analyses were performed using GraphPad Prism 6.0 software (GraphPad Software, San Diego, CA, USA) and SPSS 20.0 software (IBM, Chicago, IL, USA). All experimental data are expressed as the mean±standard deviation (SD). The differences between groups were analyzed by independent sample t-test or one-way analysis of variance (ANOVA), and χ2 test was used to analyze the differences between categorical variables. The survival curves were plotted using the Kaplan–Meier method and compared using the log-rank test. Two-sided P-values <0.05 were considered statistically significant.

Results

circPVT1 is upregulated in BCa and related to prognosis of BCa patients

In the present study, we first found that circPVT1 was significantly upregulated in BCa tissues compared with non-tumor tissues (Figure 1A). Then, we further observed that its expression levels in three human BCa cell lines (MDA-MB-231, MDA-MB-468 and MCF-7) were remarkably higher than the expression in normal MCF-10A cells (Figure 1B). We further analyzed the association between circPVT1 expression and clinicopathologic characteristics of BCa patients. The 99 BCa patients were classified into 2 groups: the high circPVT1 expression group (>median, n=52) and the low circPVT1 expression group (

Figure 1

circPVT1 is upregulated in BCa and related to prognosis of BCa patients. (A) RT-qPCR analysis of circPVT1 expression in BCa and adjacent non-cancerous tissues. (B) RT-qPCR analysis of circPVT1 expression in three BCa cell lines (MDA-MB-231, MDA-MB-468 and MCF-7). The data represent the mean±SD. *P<0.05 versus MCF-10A cells. (C) Kaplan–Meier analysis of the association between overall survival of BCa patients and circPVT1 expression.

circPVT1 promotes BCa cell proliferation in vitro

To further investigate the biological function of circPVT1 in BCa, we knocked down circPVT1 in MDA-MB-468 cells and overexpressed circPVT1 in MDA-MB-231 cells (Figure 2A). The growth curves determined by MTT assay showed that the proliferation of MDA-MB-468 cells was significantly inhibited after knockdown of circPVT1, and MDA-MB-231 cells exhibiting increased circPVT1 expression showed a higher proliferation rate (Figure 2B). In addition, as demonstrated in Figure 2C, the colony formation ability of MDA-MB-468 cells was remarkably attenuated following circPVT1 knockdown, whereas the number of colonies formed was markedly increased in MDA-MB-231 cells with circPVT1 overexpression. Apoptosis is a critical factor affecting tumor cancer proliferation, and we observed that knockdown of circPVT1 significantly increased the apoptosis rate of MDA-MB-468 cells (Figure 2D).

Figure 2

circPVT1 promotes BCa cell proliferation in vitro. (A) RT-qPCR analysis of circPVT1 expression in MDA-MB-468 cells and MDA-MB-231 cells after transfection. (B) The proliferation of MDA-MB-468 cells and MDA-MB-231 cells after transfection was detected by MTT assay. (C) The colony formation abilities of MDA-MB-468 cells and MDA-MB-231 cells after transfection were detected by colony formation assay. (D) The apoptosis of MDA-MB-468 cells was detected by flow cytometric analysis. The data represent the mean±SD. *P<0.05 versus sh-NC or empty vector-transfected cells.

Knockdown of circPVT1 inhibits BCa tumor growth in vivo

To further study the effects of circPVT1 on BCa tumor growth in vivo, we established a xenograft mouse model. It is obvious that tumors formed in the sh-circPVT1 group grew much slower (Figure 3A). Besides, as shown in Figure 3B, the tumor weight of the sh-circPVT1 group was less than that of the sh-NC group. A remarkably lower expression level of circPVT1 was observed in the xenograft tissues of sh-circPVT1 group (Figure 3C). Moreover, the results of Western blot analysis indicated that circPVT1 knockdown increased Bax expression and decreased Bcl-2 expression in the xenograft tissues (Figure 3D).

Figure 3

Knockdown of circPVT1 inhibits BCa tumor growth in vivo. (A) Tumor volume was measured every three days, and tumor growth curves were plotted. (B) Tumor weight was measured after tumor removal. (C) RT-qPCR analysis of circPVT1 expression in the excised tumor tissues. (D) Western blot analysis of Bcl-2 and Bax expression in the excised tumor tissues. The data represent the mean±SD. *P<0.05 versus sh-NC group.

circPVT1 promotes BCa cell migration and invasion in vitro

Cell migration and invasion are important prerequisites for cancer metastasis. The effects of circPVT1 on BCa cell migration and invasion were next examined. In wound healing assay, MDA-MB-468 cells with circPVT1 knockdown showed significantly impaired capacity to recover the scratched wound, and this capacity of MDA-MB-231 cells was enhanced by circPVT1 overexpression (Figure 4A). Moreover, the results of transwell assay also showed that circPVT1 overexpression enhanced, whereas circPVT1 knockdown inhibited the migration and invasion of BCa cells (Figure 4B).

Figure 4

circPVT1 promotes BCa cell migration and invasion in vitro. (A) The migratory abilities of MDA-MB-468 cells and MDA-MB-231 cells after transfection were detected by wound healing assay. (B) The migration and invasion of MDA-MB-468 cells and MDA-MB-231 cells after transfection were detected by transwell assay. The data represent the mean±SD. *P<0.05 versus sh-NC or empty vector-transfected cells.

circPVT1 induces EMT in BCa cells

BCa cell migration and invasion are closely associated with EMT.11 As shown in Figure 5, circPVT1 knockdown significantly reduced the expression of E-cadherin and increased the expression levels of N-cadherin, Vimentin, Slug and Twist in MDA-MB-468 cells. In addition, the expression of E-cadherin was decreased, whereas the expression levels of N-cadherin, Vimentin, Slug and Twist were increased in MDA-MB-231 cells following circPVT1 overexpression.

Figure 5

circPVT1 induces EMT in BCa cells. The expression levels of EMT-related proteins, including E-cadherin, N-cadherin, Vimentin, Slug and Twist, in MDA-MB-468 cells and MDA-MB-231 cells after transfection were detected by Western blot analysis. The data represent the mean±SD. *P<0.05 versus sh-NC or empty vector-transfected cells.

circPVT1 serves as a ceRNA for miR-204-5p in BCa

Through Starbase database (), the complementary sequences between miR-204-5p and circPVT1 were identified (Figure 6A). To validate this predication, dual-luciferase reporter assay was then performed, and the results demonstrated that co-transfection with miR-204-5p mimics notably reduced the luciferase activity of circPVT1-WT reporter vector in HEK293T cells, but this effect was blocked by mutation of the binding sites (Figure 6B). In addition, we found that circPVT1 overexpression decreased, whereas circPVT1 knockdown increased miR-204-5p expression level in BCa cells (Figure 6C). Moreover, in BCa tissues, a remarkable downregulation of miR-204-5p level was observed (Figure 6D), and it was also negatively associated with circPVT1 level (P=0.044; Figure 6E).

Figure 6

circPVT1 serves as a ceRNA for miR-204-5p in BCa. (A) Putative miR-204-5p binding sequence in circPVT1 fragment. (B) The luciferase activity of circPVT1-WT or circPVT1-MUT in HEK293T cells after transfection with miR-204-5p mimics or NC. (C) RT-qPCR analysis of miR-204-5p expression in MDA-MB-468 cells and MDA-MB-231 cells after transfection. The data represent the mean±SD. *P<0.05 versus NC-transfected cells; #P<0.05 versus sh-NC or empty vector-transfected cells. (D) RT-qPCR analysis of miR-204-5p expression in BCa and adjacent non-cancerous tissues. (E) The correlation between circPVT1 and miR-204-5p expression levels in BCa tissues.

miR-204-5p blocks the oncogenic role of circPVT1 in BCa cells

We then investigated whether miR-204-5p mediated the regulatory role of circPVT1 in BCa cells. The results demonstrated that co-transfection of miR-204-5p mimics obviously inhibited the EMT of circPVT1-overexpressing MDA-MB-231 cells (Figure 7A). In addition, we also found that restoration of miR-204-5p diminished the effects of circPVT1 on the migration and invasion of MDA-MB-231 cells (Figure 7B and C).

Figure 7

miR-204-5p blocks the oncogenic role of circPVT1 in BCa cells. (A) The expression levels of EMT-related proteins in MDA-MB-231 cells after transfection were detected by Western blot analysis. (B) The migratory abilities of MDA-MB-231 cells after transfection were detected by wound healing assay. (C) The migration and invasion of MDA-MB-231 cells after transfection were detected by transwell assay. The data represent the mean±SD. *P<0.05 versus circPVT1+ NC-transfected cells.

Discussion

BCa is a complex heterogeneous tumor involving accumulation of both genetic and epigenetic changes. A large number of differentially expressed circRNAs were found between BCa and adjacent tissues, and might represent as a promising new class of diagnostic biomarkers.12 In this study, we initially examined the expression profile of circPVT1 and found that circPVT1 was markedly upregulated in BCa tissues and cell lines. Considering its higher expression in BCa tissues, circPVT1 can be a potential diagnostic biomarker for BCa patients. It may also represent a prognostic biomarker due to the different overall survival between BCa patients with circPVT1 high expression and low expression.

The oncogenic role of circPVT1 has been well studied in various types of cancers, including head and neck squamous cell carcinoma13 and acute lymphoblastic leukemia.14 To further elucidate the biological role of circPVT1 in BCa, we then performed loss-of-function and gain-of-function experiments, and the results showed that knockdown of circPVT1 significantly suppressed BCa cell proliferation, migration and invasion, and promoted cell apoptosis; while overexpression of circPVT1 promoted BCa cell proliferation, migration and invasion. In addition, knockdown of circPVT1 also suppressed the growth of BCa xenografts in murine model. These data suggested that circPVT1 also serves as an oncogene in BCa malignant progression.

Metastasis is the leading reason for the mortality of BCa patients.15 Epithelial–to–mesenchymal transition (EMT) has been regarded as an important step for tumor infiltration and metastasis.16 During EMT, epithelial carcinoma cells lose their apical-basal polarity and develop a mesenchymal phenotype, thus increasing their migratory and invasive capacities.17 Understanding the biological intricacies of EMT will help us to design new therapies against BCa metastasis.18 This study showed that circPVT1 overexpression promoted, whereas circPVT1 knockdown inhibited the EMT of BCa cells, indicating that EMT might be one of the mechanisms underlying the oncogenic role of circPVT1 in BCa.

miRNAs are a class of small non-coding RNAs with 19–25 nucleotides in length. An increasing number of studies have shown that with miRNA-binding sites, circRNAs can serve as competitive endogenous RNAs (ceRNAs) and thereby regulate cancer process.19 For example, in non-small cell lung cancer, circPVT1 acts as a ceRNA for miR-497.20 Recently, miR-204-5p has been reported as a tumor suppressor in BCa,21 and in this study, we confirmed that circPVT1 could directly interact with and make a negative regulation on miR-204-5p. Further rescue assays indicated that restoration of miR-204-5p blocked the oncogenic role of circPVT1 in BCa cells.

Collectively, this study provides the first link between circPVT1 expression and BCa development. We found that circPVT1 was upregulated in BCa, and its upregulation might promote the invasion and EMT of BCa cells by serving as a ceRNA for miR-204-5p. Therefore, circPVT1 might be a potential diagnostic biomarker and therapeutic target for BCa patients.