Potential Role of circPVT1 as a proliferative factor and treatment target in esophageal carcinoma

All patients and healthy volunteers signed an informed consent form approved by the institutional review board. We collected 20 patients with esophageal cancer and underwent complete resection at Tianjin Medical University General Hospital from 2017 to 2018. The study was approved by the Ethics Committee of Tianjin Medical University General Hospital (Ethical. NO. IRB2018-XY-034), and obtained written informed consent from all the patients. Besides, we collected blood samples from these patients prior to surgery, and other 20 age- and gender-matched healthy volunteers served as controls.

The cell lines of EC109, CaES-17, TE-1, TE-10 were obtained from China Center for Type Culture Collection (CCTCC) and the Human normal esophageal epithelial cells (HEEC), HepG2, MKN45, SW60, A549 cell lines were purchased from Beijing Beina Science and technology company. The cells were cultured in DMEM (Gibco, USA) medium supplemented with 15% FBS (Gibco, USA), 100 U/mL of penicillin and 100 mg/mL of streptomycin (Hyclone, USA) at 37 °C in a humidified culture chamber (NAPCO5410, USA) supplied with 5% CO₂ atmosphere.

Quantitative real-time PCR (qRT-PCR) analysis was performed to detect the expression of circPVT1 [17]. Total RNA was isolated from tissues or serum using the TRIzol kit (Invitrogen, USA) and from cells using the miRNeasy Mini Kit (Qiagen, China) following to the manufacture’s guide respectively. 2 mg RNA of every sample was then incubated at 37 °C for 20 min with or without RNase R (3 U/mg, Epicentre Technologies, USA) and RNeasy MinElute Cleanup Kit (Qiagen, China) was used for RNA clean and concentration. Quantitative real-time PCR (qRT-PCR) analysis was performed to detect the expression of circPVT1 using the QuantiNova SYBR Green RT-PCR Kit (QIAGEN 208152) on the DSX System (Thermo Lab system) in accordance with the instructions. The expression of circPVT1 was normalized to GAPDH. The PCR primers were shown as follows: circPVT1 forward primers: 5′-GGTTCCACCAGCGTTATTC-3′, reverse primers: 5′-CAACTTCCTTTGGGTCTCC-3′; PVT1 forward primers: 5′-TTCAGCACTCTGGACGGACTT-3′, reverse primers: 5′-TATGGCATGGGCAGGGTAG-3′; GAPDH forward primers: 5′-AGAAGGCTGGGGCTCATTTG-3′, reverse primers: 5′-AGGGGCCATCCACAGTCTTC-3′.

Three kinds of siRNA sequences were designed for the circPVT1 and the sequence of si-circPVT1-1 was 5′-UGGGCUUGAGGCCUGAUCU-3′, si-circPVT1-2 was 5′-CUGUCAGCUGCAUGGAGCUUCGU-3′, si-circPVT1-3 was 5′-GCUUGAGGCCUGAUCUUUU-3′ and the relative si-NC sequence was 5′-AAUUCUCCGAACGUGUCACGU-3′.

The control plasmid and the circPVT1 overexpression plasmid were purchased from Biosyntech (Suzho, China). The sequences of circPVT1 and the control vector were provided in Additional file 1: Figures S1.

Cells were seeded in 12-well plates and the transfection were performed when 60–80% of cell confluency. Lipofectamine 2000 was used for the transfection according to the manufacture’s guide. All reagents were diluted and gently mixed and then incubated for 20 min at room temperature. Thereafter, the cells were added together with the mixture and incubated for another 4–6 h. TE-10 cells were collected 48 h after transfection, and the total RNA was isolated for RT-PCR to detected the expression level of CircPVT1.

In the apoptosis assay, transfection was carried out to TH-10 cells with si-RNA or si-NC and the cells were cultured for another 48 h. Then Hoechst staining was performed using the Hoechst Staining Kit (Beyotime Biotechnology company, China) according to the manufacturer’s instructions. To obtain an unbiased count, cells were blindly scored without knowledge of previous treatment. Uniformly stained nuclei were scored as healthy, viable cells. Condensed or fragmented nuclei were scored as apoptotic.

Western blotting was performed using standard procedures. Briefly, Cells were collected and rinsed with PBS. The total protein was extracted by addition of 2% sodium dodecyl sulfate (SDS), 125 mM Tris (pH 6.8) buffer. Lysates were sonicated and the protein was quantified by BCA Protein Assay Kit (Beyotime Biotechnology, China). Then the protein was separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore, USA). The membranes were blocked with 5% skim milk powder at room temperature for 1 h and probed with the adequate primary antibodies: anti- paired box protein 4 rabbit polyclonal antibody (Pax-4; 1:1000, ab42450, Abcam), anti- paired box protein 6 rabbit monoclonal antibody (Pax-5; 1:1000, ab109443, Abcam), anti- peroxisome proliferators-activated receptors rabbit polyclonal antibody α(PPARα; 1:1000, ab23673, Abcam), anti- peroxisome proliferators-activated receptors rabbit polyclonal antibody o (PPAR-ab 1:1000, ab45036, Abcam), anti- Glyceraldehyde-3-phosphate dehydrogenase rabbit monoclonal antibody (GAPDH; 1:1000, ab181602, Abcam) overnight at 4 °C. Then they were incubated with the adequate peroxidase-conjugated secondary antibodies (1:5000, ab6721, Abcam) at the dilutions recommended for 1 h. The analysis was performed using the Super-Signal chemiluminescence Western blotting detection system (Pierce, USA).

For transwell invasion experiments, cells were treated as before and then transfected with control plasmid or the circPVT1 overexpression plasmid. TE-10 cells were collected, counted, and plated (1.5 × 105) into the 24-well Boyden chamber with a non-coated 8-mm pore size filter in the insert chamber (BD Falcon, Corning-Costar, USA). Cells were seeded into the insert chamber containing 0.5 ml Dulbecco’s modified Eagle’s medium/F12 media without containing FBS and allowed to migrate into the bottom chamber containing 0.5 ml of Dulbecco’s modified Eagle’s medium/F12 media containing 10% FBS. Transwell invasion assay lasted 24 h in a humidified incubator at 37 °C in 5% CO₂. The 4,6-diamidino-2-phenylindole staining was used to counter the number of TE-10 and the average percentage of migrated cells was calculated.

RegRNA 2.0 (http://​regrna2.​mbc.​nctu.​edu.​tw) were used as a widely used regulatory RNA motifs identification tool for bioinformatics analysis. FASTA format sequence were submitted and the predictive results were presented via a graphical interface. The sequence of first 3000 dp of PVT1 exon and cir-PVT1 were submitted to predict the potential targets miRNA and protein.

We collected blood samples from 20 patients undergoing esophageal cancer surgery and 20 healthy volunteers. The expression level of circPVT1 was detected and there was no significant difference between EC patients and healthy people (Fig. 1a; P > 0.05). Then the esophagus of surgical resection was separated into tumor and para-carcinoma tissue. As shown in Fig. 1b, circPVT1 was markedly up-regulated in EC tissues compared with the adjacent tissues (Fig. 1b; P < 0.01).

In order to verify whether the highly expressed circPVT1 was produced by tumor cells, several cancerous and non-cancerous cell lines were introduced in this experiment. We observed a significant increase in CIRPVT1 levels in cancer cell lines compared with non-cancer cell lines, especially in EC109, TE-1 and TE-10 (Fig. 1c). This phenomenon indicated that the potential role of circPVT1 in cancers progression, especially in esophageal cancer.

Then EC109, TE-1 and TE-10 cell lines were cultured until over-confluence. Different stages of these three cell lines were collected and the total RNA was isolated for detection of circPVT1 levels. All three kinds of cells had high expression of this circRNA, and we did not find any significant difference between the stages before 100% confluence. However, when the cells maintained over-confluence, circPVT1 levels significantly decreased (Fig. 1d–f; P < 0.01). These results suggested that it had potential relationship between the circPVT1 expression and cell proliferation.

To test the possibility that circPVT1 plays an important role in cell proliferation, we designed three kinds of siRNAs. The target of these siRNAs was circPVT1 specifically, which had no effect on linear PVT1 mRNA (Fig. 2A).

The TE-10 cells were transfected with these siRNAs. Forty eight hours later, TE-10 cells were collected and total RNA was isolated for RT-PCR to detect the expression levels of circPVT1. We digested linear PVT1 mRNA with RNAse-R instead of circ-PVT1. Following this, the levels of circPVT1 and linear PVT1 were detected. We found that the expression of circPVT1 was substantially decreased 48 h after treated with siRNA especially siRNA-3 (more than 69 percentage, Fig. 2B, P < 0.01) without affecting the expression of linear PVT1 (Fig. 2B, C).

To examine the phenotypic effects of these siRNAs, apoptosis of TE-10 cells was analyzed by Hoechst staining. The number of cells with apoptotic morphology appearing condensed or fragmented nuclei was counted. As shown in Fig. 2E–G, the positive cells significantly increased after siRNA was transfection to the TE-10 cells (23.67% ± 1.53% for si-RNA-1, 31.33% ± 2.08% for si-RNA-2, 31.66% ± 3.51% for si-RNA-3 respectively) compared to the control group (11.67% ± 2.08%, Fig. 2D). The number of positive cells was up-regulated by siRNAs, indicating that silencing circ-PVT1 subsequently reduced the survival ability of cancer cells (Fig. 2H).

Taken together, these data suggested that circPVT1 knockdown by siRNA might partially impeded the proliferation of EC cells and leaded to apoptosis in vitro.

The invasive ability of tumors is closely related to the degree of malignancy. To determine whether higher levels of circPVT1 expression would enhance tumor aggressiveness, we designed overexpression plasmids and transfected them into TE-10 cells. Then the transwell invasion assay was used to determine the invasiveness of such cancer cells. The results showed that compared with control group, the number of migrating cells was significantly increased (over 3.03-fold) after overexpressing of circPVT1 (Fig. 3A; P < 0.01).

To confirm whether this enhanced invasive ability was caused by over-expressed circPVT1, the overexpression plasmid was transfected wi siRNA and it was found that the invasive ability of tumor cells was inhibited and returned to the level of the control group (Fig. 3B; P < 0.01). In summary, overexpression of circPVT1 can increase the invasiveness of TE-10 cells, and this phenomenon disappeared after knockdown of circ-PVT by siRNA.

CircPVT1 plays an important role in the regulation of cellular proliferation and invasion during tumor progression, but how its regulatory mechanism works is still unclear in esophageal squamous cell carcinoma.

To further explore the regulatory mechanism of circPVT1, bioinformatics analysis was used in this study. The investigation of the potential miRNAs (Additional file 2: Table S1, http://​regrna2.​mbc.​nctu.​edu.​tw/​Results/​1549033793.​all.​result) and binding sites (Additional file 2: Table S2, http://​regrna2.​mbc.​nctu.​edu.​tw/​Results/​1549034097.​all.​result) were performed, which were based on the sequence of the first exon of the PVT1 gene. In addition, potential miRNAs that bind to circPVT1 were also investigated (Additional file 2: Table S3, http://​regrna2.​mbc.​nctu.​edu.​tw/​detection_​output.​ph) and hsa-miR-4633 was selected to construct the over expressional plasmid.

Then protein levels of paired box genes (Pax-4 and Pax-6) were examined by western blotting, which were considered to promote tumor growth. Besides,peroxisome proliferators-activated receptors (PPARs, PPARα and PPAR-γ), which inhibit the growth of tumor were also detected. The results showed that both Pax-4 and Pax-6 were downregulated when miR-4663 was overexpressed in TE-10 cell (Fig. 4a). On the other hand, PPAR-α and PPAR-γ were up-regulated simultaneously (Fig. 4a). These experiments showed that miR-4663 had the inhibitory effect on tumor growth. Furthermore, when circPVT1 was also translated into the miR-4663 overexpression cell line, Paxs increased and PPARs decreased significantly compared to the control or circPVT1 only group, which indicated that the inhibition of tumor by miR-4663 no longer functions after the addition of circPVT1 (Fig. 4a).

Taken together, circPVT1 may affect the malignant degree of tumor by affecting miRNA and regulating the levels of Paxs and PPARs.