Circular RNA ciRS-7 triggers the migration and invasion of esophageal squamous cell carcinoma via miR-7/KLF4 and NF-κB signals

ABSTRACT

Esophageal squamous cell carcinoma (ESCC) is one of the most prevalent and deadly cancers worldwide, especially in Eastern Asia. It has been indicated that circular RNAs (circRNA) are the key regulators in the development and progression of human cancers. We therefore evaluated the expression and regulation effects of ciRS-7 on the progression of ESCC, which is a recently identified circRNA and acts as a natural competing endogenous RNA. The expression of ciRS-7 was significantly increased in the ESCC tissues and cells as compared with their corresponding controls. In vitro study showed that ciRS-7 can promote the migration and invasion of ESCC cells. Over expression of miR-7, one of well-known targets of ciRS-7, can attenuate ciRS-7 induced invasion of ESCC cells and over expression of matrix metalloproteinase 2 (MMP2). The expression of stem cell marker Kruppel-like factor-4 (KLF-4), which has been reported as the target of miR7, increased significantly in ciRS-7 transfected ESCC cells. Knockdown of KLF-4 also attenuated over expression of ciRS-7 induced cell invasion. In addition, BAY 11–7082, the inhibitor of NF-κB, partially reversed ciRS-7 induced cell invasion. Mechanically studies indicated that ciRS-7 increased the expression of p65 via increasing the phosphorylation of IKK-α. Collectively, our present study revealed that ciRS-7 can trigger the migration and invasion of ESCC cells via miR-7/KLF4 and NF-κB signals. Targeted inhibition of ciRS-7 might be a potential approach for ESCC treatment.

Introduction

As the sixth leading cause of cancer-related death globally, esophageal squamous cell carcinoma (ESCC) is one of major challenge for human health with increasing incidence in the developing countries.¹ It was reported that drinking hot beverages, low intake of fruits and vegetables, and smoking will contribute to the initiation and progression of ESCC.^2,3 Metastasis is one of the main causes that contribute to the unsuccessful treatment in this disease management.⁴ The 5 year survival rate for ESCC patients with early lymph node metastasis was less than 10%, while patient with early stage of ESCC is higher than 90%.⁵ However, the mechanisms involved in the metastasis of ESCC are far from fully illustrated. The identification of key molecules regulated the cell motility of ESCC will greatly contribute to the improvement of disease diagnosis and therapy.

The improvement of RNA-sequencing technologies revealed that non-coding (nc)RNAs contributed to the tumorigenesis and progression of cancer. circRNAs have become another focus in the field of ncRNA research.⁶ They are formed by back-splicing covalently joined 3′- and 5′-ends and generally do not encode protein.⁷ Recent studies indicated that circRNAs are endogenous, stable, abundant and conserved in eukaryotic cells.^8,9 Furthermore, they can regulate the progression of various cancers including pancreatic, ovarian, bladder and liver cancer.^10-12 Mechanically, circRNA can regulate the gene expression via act as a sponge for miRNA.¹¹ Fox example, a circRNA CDR1as, which contained > 70 conventional binding sites and functioned as a super sponge for miR-7.¹¹ Although emerging evidences suggested that circRNA can regulate proliferation and invasion of cancer cells,^13,14 there are few studies concerning the effects of circRNA on the progression of ESCC. In addition, whether circRNAs harbor miRNAs with regulatory roles in ESCC is also unknown.

The experimental evidence about the functions and related mechanisms of circRNAs are few. ciRS-7, which is transcribed in the antisense orientation with respect to the CDR1 gene, can function as a miRNA sponge.⁶ As one of the most intensely studied circRNAs, ciRS-7 can trigger the proliferation, migration and invasion of colorectal,¹⁵ hepatocellular carcinoma,¹⁴ and cervical¹¹ cancer cells. Our present study found that ciRS-7 is significantly increased in the ESCC cells and tissues as compared with their corresponding controls. Mechanically, ciRS-7 can trigger the migration and invasion of ESCC cells via regulation miR-7/KLF4 and NF-κB signals.

Materials and methods

Cell culture and siRNA transfection

Human ESCC cell KYSE150, KYSE140, KYSE70, Eca9706, EC18, and TE13 and human esophageal epithelial cell line HEEC were obtained from the Cell Bank of the Chinese Academy of Sciences, Shanghai, China. They were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) (HyClone, USA) at 37°C in a humidified atmosphere of 5% CO2. For siRNA transfection, the siRNA negative control or siRNAs for Kruppel-like factor-4 (KLF4), ERK1/2, Akt, p65 or IKK-α were transfected with Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer’s instruction.

Reverse transcription-quantitative real-time PCR (qRT-PCR)

The total RNAs from cells or tissues were extracted by use of TRIzol® Reagent (Invitrogen) according to manufacturer’s instructions. Then 500 ng of total RNA was used to generate the cDNA using GoScript™ Reverse Transcription System (Promega, USA) according to the manufacture’s protocol. The gene expression was measured by use of SsoAdvanced™ Universal SYBR® Green supermix (Biorad) on the CFX Connect Real-Time PCR System (Biorad) with the following primers: GAPDH, 5ʹ-GAC TCA TGA CCA CAG TCC ATG C-3ʹ (forward) and 5ʹ-AGA GGC AGG GAT GAT GTT CTG-3ʹ (reverse); U6, 5ʹ- CTC GCT TCG GCA GCA CA-3ʹ (forward) and 5ʹ-AAC GCT TCA CGA ATT TGC GT-3ʹ (reverse); ciRS-7, 5ʹ-ACG TCT CCA GTG TGC TGA-3ʹ (forward) and 5ʹ- CTT GAC ACA GGT GCC ATC-3ʹ (reverse); miR-7, 5ʹ- CTA GCT AGC TAG AGC ACC AAT AGG GAA GGG-3ʹ (forward) and 5ʹ- GAA GAT CTT CGA GTC TGC CGA TGG GTG T-3ʹ (reverse); p65, 5ʹ-CCC CAC GAG CTT GTA GGA AAG −3ʹ (forward) and 5ʹ- CCA GGT TCT GGA AAC TGT GGA T −3ʹ (reverse); EGFR, 5ʹ- AGG CAC GAG TAA CAA GCT CAC-3ʹ (forward) and 5ʹ-ATG AGG ACA TAA CCA GCC ACC-3ʹ (reverse); XIAP, 5ʹ- GGC CAT CTG AGA CAC ATG CAG-3ʹ (forward) and 5ʹ-GCA TTC ACT AGA TCT GCA ACC-3ʹ (reverse); KLF4, 5ʹ-ACC AGG CAC TAC CGT AAA CAC A-3ʹ (forward) and 5ʹ- GGT CCG ACC TGG AAA ATG CT −3ʹ (reverse); NOTCH1, 5ʹ- CGA CGC ACA AGG TGT CTT CCA-3ʹ (forward) and 5ʹ- CGG ACT TGC CCA GGT CAT CTA C-3ʹ (reverse). GAPDH was set as the reference gene of ciRS‐7, and U6 was set as the reference gene of miR‐7. The gene relative expression was calculated using the 2^−ΔΔCt method. All experiments were performed three times independently and the average was used for comparison.

Human ESCC tissue collection

During July 2014 to Jun 2017, the tumor tissues and matched adjacent normal tissues of 29 ESCC patients with histologically confirmed were collected according to the permission of Ethical Committee in our hospital. The demographic and clinical features of ESCC patients were summarized at Table 1. The samples were store at −80°C immediately after surgery. The RNAs of both tumor and adjacent tissues were extracted by use of Trizol to analyzed the expression of ciRS-7.

Table 1.

The demographic and clinical features of ESCC patients.

Characteristics	N	p value
Tissues
Tumor	29	< 0.001
Adjacent normal	29
Age
≤ 50	8	0.327
＞50	21
Gender
male	19	0.091
female	10
Grade
1	2	0.224
2	17
3–4	10

Over expression of ciRS-7

In order to over expression of ciRS-7, we synthesized the human ciRS-7 cDNA and cloned into pLCDH-ciR vector (Geneseed Biotech Co, Guangzhou, China) to generate pLCDH-ciR/ciRS7 plasmid. Then the vector control and pLCDH-ciR/ciRS7 were transfected by use of Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer’s instruction.

Cell proliferation test

The proliferation of cells transfected with vector control or pLCDH-ciR/ciRS7 were seeded at a cell density of 5 × 10³ cells/well in 96-well plates. After cultured for 48 h, 10 μl of MTT solution was added to each well and incubated for 2 h at 37 °C. The absorbance of each well was measured at 570 nM.

Cell migration and invasion assay

The effects of ciRS-7 on the migration and invasion of ESCC cells were evaluated by use of wound healing and transwell assay, respectively, according to the previous study.¹⁶ Briefly, cells cultured in six-well plates with 90% confluence were scraped with the fine end of 200 μl pipette tips (time 0 h). The migration was photographed at 0 and 48 h after injury. The migrated distance was measured, normalized to the 0 h control, and expressed as relative migration. For invasion assay, transwell chamber were pre-coated matrigel according to the manufacturer’s protocol (BD Science, Bedford, MA, USA). Single cell suspensions were added to the upper chambers and incubated for 48 h. The invaded cells were quantified by counting cells at least three random fields and normalized to the vector control group.

Western blot analysis

At the end of the experiment, cells were lysed in lysis buffer containing protease and phosphatase inhibitors. A fixed amount of protein was separated by use of 10% SDS–PAGE gel and transferred onto polyvinylidene difluoride (PVDF) membrane (Roche, Mannheim, Germany). The membrane was incubated with primary antibodies diluted in 5% milk in PBS with shaking overnight at 4°C followed by secondary antibody at room temperature for 2 h. GPADH was used as the loading control. The protein expression was detected by using enhanced chemiluminescence (Millipore, Darmstadt, Germany).

Statistical analysis

SPSS software (SPSS, Inc., Chicago, IL, USA) was applied for statistical analysis. Statistical analysis was performed using either Student’s t test (two‐group comparison). All experiments were conducted for at least three times and P < 0.05 was considered statistically significant.

Results

The expression of ciRS-7 in ESCC cells and tissues

Firstly, the expression of ciRS-7 in ESCC cell KYSE150, KYSE140, KYSE70, Eca9706, EC18, and TE13 and human esophageal epithelial cell line HEEC were measured by qRT-PCR according to the previous study.¹⁷ Our data showed that the expression of ciRS-7 in most of ESCC cells including KYSE150, KYSE140, KYSE70, Eca9706, EC18 were greater than that in HEEC cells (Figure 1A). Our then evaluated the expression of ciRS-7 in 29 matched ESCC and its adjacent normal tissues. The results revealed that the expression of ciRS-7 in 72.4% (21/29) ESCC patients were greater than their corresponding matched normal tissues (Figure 1B, Table 1). These data revealed that the expression of ciRS-7 is upregulated in ESCC cells and tissues.

Figure 1.

The expression of ciRS-7 in ESCC cells and tissues. (A) The expression of ciRS-7 in ESCC cell KYSE150, KYSE140, KYSE70, Eca9706, EC18, and TE13 and human esophageal epithelial cell line HEEC was measured by qRT-PCR; (B) The expression of ciRS-7 in 29 pairs of matched ESCC and adjacent normal tissues were measured by qRT-PCR.

ciRS-7 increased the migration and invasion of ESCC cells

We over expressed ciRS-7 in KYSE150 and Eca9706 cells via transfection of pLCDH-ciR/ciRS-7 vector (Figure 2A). Our data showed that over expression of ciRS-7 had no significant effect on the proliferation of either KYSE150 or Eca9706 cells after transfection for 48 h (Figure 2B). We then investigated the effects of ciRS-7 on the migration and invasion of ESCC cells. Wound healing assay showed that ciRS-7 significantly increased the in vitro migration of both KYSE150 (Figure 2C) and Eca9706 (Figure 2D) cells. This was confirmed by transwell analysis that over expression of ciRS-7 also significantly increased the in vitro invasion of both KYSE150 and Eca9706 cells (Figure 2E). In addition, increased expression of MMP-2 and MMP-9 was observed in ciRS-7 over expressed KYSE150 and Eca9706 cells (Figure 2F). Collectively, our data showed that ciRS-7 can promote the migration and invasion of ESCC cells.

Figure 2.

ciRS-7 increased the migration and invasion of ESCC cells. (A) KYSE150 and Eca9706 cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 24 h, the expression of ciRS-7 was detected by use of qRT-PCR; After transfected with pLCDH-ciR/ciRS-7 or vector control for 48 h, the cell proliferation was detected by CCK-8 kit (B), the migration (C, KYSE150; D, Eca9706) was measured by wound healing assay, the invasion was measured by transwell assay (E), and the expression of MMP-2, MMP-9, and vimentin was measured by western blot analysis (F). ** p < 0.01.

Mir-7 was involved in cirs-7 induced migration and invasion of ESCC cells

Previous studies indicated that ciRS-7 can act as a ceRNA of miR-7.^14,15,17 We evaluated the effects of ciRS-7 on the expression of miR-7 in ESCC cells. Our data confirmed that over expression of ciRS-7 can significantly inhibit the expression of miR-7 in both KYSE150 and Eca9706 cells (Figure 3A). To verify whether miR-7 is involved in ciRS-7 induced migration and invasion of ESCC cells, we over expressed miR-7 in KYSE150 and Eca9706 cells (Figure 3B). Our data showed that over expression of miR-7 can partially attenuate ciRS-7 induced invasion of both KYSE150 (Figure 3C) and Eca9706 (Figure 3D) cells. This was confirmed by western blot analysis that over expression of miR-7 reversed ciRS-7 induced over expression of MMP-2 and MMP-9 in KYSE150 cells (Figure 3E). These data indicated that miR-7 is involved in ciRS-7 induced migration and invasion of ESCC cells.

Figure 3.

miR-7 was involved in ciRS-7 induced migration and invasion of ESCC cells. (A) KYSE150 and Eca9706 cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 24 h, the expression of miR-7 was detected by use of qRT-PCR; (B) KYSE150 and Eca9706 cells were transfected with miR-7 construct or vector control for 24 h; KYSE150 (C) and Eca9706 (D) cells were transfected with pLCDH-ciR/ciRS-7, miR-7 construct or vector control alone or together for 48 h, the cell invasion was evaluated by use of transwell assay; (E) KYSE150 cells were treated as (C), the protein expression was measured by western blot analysis. ** p < 0.01.

KLF-4 mediated the effects of ciRS-7/miR-7 axis on invasion of ESCC cells

We analyzed the potential targets of miR-7 by use of two widely-used bioinformatics tools, PicTar (http://pictar.mdc-berlin.de) and miRNA.org (http://www.microrna.org/microrna/home.do). Among the identified targets, three interesting genes EGFR, X-linked inhibitor of apoptosis protein (XIAP), KLF4, and NOTCH1 were prioritized because of their roles in cancer progression.¹⁸ We found that over expression of ciRS-7 can significantly increase the mRNA expression of EGFR and KLF4 in KYSE150 cells (Figure 4A). While ciRS-7 only increased the mRNA expression of KLF4 in Eca9706 cells (Figure 4B). Western blot analysis confirmed that ciRS-7 can increase the expression of KLF4 in both KYSE150 and Eca9706 cells (Figure 4C). We therefore knocked down the expression of KLF4 in KYSE150 cells via its specific siRNA (Figure 4D). Our data showed that si-KLF4 can also reverse ciRS-7 induced upregulation of MMP-2 and MMP-9 in KYSE150 cells (Figure 4E). In addition, si-KLF4 attenuated ciRS-7 induced in vitro invasion of KYSE150 cells (Figure 4E). These results suggested that KLF-4 mediated the promotion effects of ciRS-7/miR-7 axis on invasion of ESCC cells.

Figure 4.

KLF-4 mediated the promotion effects of ciRS-7/miR-7 axis on invasion of ESCC cells. KYSE150 (A) and Eca9706 (B) cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 24 h, the mRNA expression was detected by use of qRT-PCR; (C) KYSE150 and Eca9706 cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 24 h, the expression of KLF-4 was measured by western blot analysis; (D) KYSE150 cells were transfected with si-NC or si-KLF4-1 ~ 3 for 24 h, the expression of KLF-4 was measured by western blot analysis; KYSE150 cells were transfected with si-NC, si-KLF4-3, vector control, or pLCDH-ciR/ciRS-7 alone or together for 48 h, the protein expression was measured by western blot analysis (E), the cell invasion was measured by transwell analysis (F). *p < 0.05, ** p < 0.01.

NF-κB was involved in cirs-7 induced invasion of ESCC cells

Recent study indicated that ciRS-7 can regulate the activation of various signals involved in cell motility such as ERK1/2, PI3K/Akt and NF-κB.^16,17 We then pre-treated cells with PD98059 (PD, ERK1/2 inhibitor), LY294002 (LY, PI3K/Akt inhibitor) or BAY 11–7082 (BAY, NF-κB inhibitor) before the transfection of ciRS-7. The expression of MMP-2 was measured by use of western blot analysis. Our data showed that BAY, while not LY, can attenuate ciRS-7 induced upregulation of MMP-2 in both KYSE150 (Figure 5A) and Eca9706 (Figure 5B) cells. We further knocked down the expression of ERK1/2, PI3K/Akt and p65 in KYSE150 cells (Figure 5C). Only si-p65 can attenuate ciRS-7 induced upregulation of MMP-2 in KYSE150 cells (Figure 5D). In addition, BAY can also reverse ciRS-7 induced invasion of KYSE150 cells (Figure 5C). The upregulation of p-p65 (S536) and p65, but not the acetylation of p65 (Ac-p65 K310), was observed in ciRS-7 transfected KYSE150 and Eca9706 cells (Figure 5F). These data indicated that NF-κB was involved in ciRS-7 induced invasion of ESCC cells.

Figure 5.

NF-κB was involved in ciRS-7 induced invasion of ESCC cells. KYSE150 (A) or Eca9706 (B) cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 6 h and then further exposed to PD98059 (PD, ERK1/2 inhibitor, 10 μM), LY294002 (LY, PI3K/Akt inhibitor, 10 μM) or BAY 11–7082 (BAY, NF-κB inhibitor, 10 μM) for 24 h, the expression of MMP-2 was measured by western blot analysis; (C) KYSE150 cells were transfected with siNC, or siRNA for ERK1/2, Akt, or p65 for 24 h, the expression of protein checked by western blot analysis; (D) KYSE150 cells were transfected with si-NC, si-RNAs, vector control, and ciRS-7 alone or together for 24 h, the expression of MMP-2 was checked; (E) KYSE150 cells were treated as (A), the cell invasion was measured by transwell analysis; (F) KYSE150 or Eca9706 cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 24 h, the expression of p-p65(S536), Ac-p65 (K310), and p65 was measured by western blot analysis (left), the levels of p-p65(S536) and Ac-p65 (K310) were normalized to p65 (right). *p < 0.05, ** p < 0.01.

CiRS-7 increased the expression of p65 via IKKα signals

We then investigated the mechanisms responsible for ciRS-7 induced expression of p65. Our data showed that over expression of ciRS-7 had no significant effect on mRNA expression of p65 in either KYSE150 or Eca9706 cells (Figure 6A). The canonical IKK/IκB pathway is suggested to be responsible for NF-κB activation in various cancer cells.¹⁹ Western blot analysis showed that the phosphorylation of IKK-α, while not IKK-β or IKKγ, was significantly increased in ciRS-7 transfected with KYSE150 cells (Figure 6B). We then knocked down the expression of si-IKK-α (Figure 6C). Our data showed that siRNA significantly reversed ciRS-7 induced upregulation of p65 (Figure 6D). These results indicated that ciRS-7 increased the expression of p65 via IKKα signals.

Figure 6.

ciRS-7 increased the expression of p65 via IKKα signals. (A) KYSE150 or Eca9706 cells were transfected with pLCDH-ciR/ciRS-7 or vector control for 24 h, the mRNA of p65 was detected by use of qRT-PCR; (B) After transfected pLCDH-ciR/ciRS-7 or vector control for 24 h, the phosphorylation and total expression of IKK-α/β/γ in KYSE150 cells were measured by western blot analysis; (C) KYSE150 cells were transfected with si-NC or si-IKK-α 1 ~ 3 for 24 h, the expression of IKK-α was measured by western blot analysis; (D) After transfected with si-NC, si-IKK-α, vector control, or pLCDH-ciR/ciRS-7 alone or together for 24 h, the protein expression was measured by western blot analysis.

Discussion

Increasing evidences demonstrated that circRNAs can regulate the progression of various cancers including ESCC.^20,21 As one of the most studied circRNA, ciRS-7 has been reported to regulate the development of colorectal,¹⁵ hepatocellular carcinoma,¹⁴ gastric,¹⁷ and cervical¹¹ cancer. It can modulate the proliferation, migration, and invasion of cancer cells.²² Clinical data suggested that it might be a prognostic biomarker and a potential therapeutic target in colorectal cancer.¹⁵ Our present study revealed that ciRS-7 can trigger the in vitro migration and invasion of ESCC cells, which is consistent with recent study that ciRS-7 accelerates ESCC progression via induction of migration and invasion.²² However, we did not observe any significant effect of ciRS-7 on the proliferation of KYSE150 and Eca9706 cells, which is not consistent with recent results that ciRS-7 increased the proliferation and colony formation of ESCC TE13 cells.²² Over expression of ciRS-7 also did not affect the growth of gastric cancer HGC-27 cells.¹⁷ It suggested that the effects of ciRS-7 on cell proliferation might be cell line dependent.

Our present study revealed that miR-7/KLF-4 axis mediated ciRS-7 induced progression of ESCC cells. ciRS-7 was proved to act as a competing endogenous RNA of miR‐7 in various studies.¹⁸ It was evidenced by the results that ciRS-7 contains 74 binding sites for miR-7 then can bind densely to the miR-7 effector.⁶ miR-7 has been proved to be a tumor suppressor miRNA.^23-26 It can suppress cell growth and promote apoptosis in cancer cell via down regulation of XIAP or other oncogenic signaling pathways.^27,28 As to ESCC, miR-7 levels are strongly correlated with tumor length and the status of lymph node metastasis.²⁹ KLF4 is a transcription factor and well identified target of miR-7.^30,31 KLF4 can promote the differentiation of ESCC via up regulation of keratin 13.³² Our study revealed that both over expression of miR-7 or knockdown of KLF4 can partially attenuate the promotion effect of ciRS-7 induced malignancy of ESCC cells.

Our present data also suggested that IKKα mediated activation of NF-κB is involved in ciRS-7 induced invasion of ESCC cells. This was confirmed by the results that ciRS-7 can induce the expression of p65 via phosphorylation of IKKα, while the inhibitor of NF-κB can attenuate ciRS-7 induced cell migration and invasion. miR-7, the target of ciRS-7, can regulate the expression of p65 in dopaminergic SH‐SY5Y cells.³³ However, over expression of ciRS-7 had no effect on the mRNA expression of p65 in SH-SY5Y cells³⁴ or ESCC cells in the present study, suggesting that ciRS‐7 might act as a miR‐7 buffer or reservoir in modulating miR‐7 function rather than a negative regulator of miR‐7. Our study revealed that ciRS-7 can regulate the phosphorylation of IKKα, this might be due to ciRS-7 can directly bind to IKKα or it can regulate the expression of factors responsible for the phosphorylation of IKKα. Further studies are needed to illustrate the underline mechanisms.

Collectively, our data showed that ciRS-7 is over expressed in ESCC cells and tissues. It can trigger the migration and invasion of ESCC cells via regulation of miR-7/KLF4 axis and activation of NF-κB p65 signals. Our present data, together with published results, suggested that ciRS-7 might be a potential target for ESCC treatment.

Funding Statement

This research was supported by the Basic research project (Natural science funds) of Jiangsu Province (NO. K20160606).

Authors’ contributions

Conception and design: Zhiyun Xu, Hairong Huang;

Acquisition of data: Hairong Huang, Lei Wei, Tao Qin, Nan Yang;

Analysis and interpretation of data: Hairong, Lei Wei;

Writing, review, and/or revision of the manuscript: Zhiyun Xu, Zhongdong Li.

Disclosure of Potential Conflict of interest

No potential conflicts of interest were disclosed.