Circular RNA circCHSY1 silencing inhibits the malignant progression of esophageal squamous cell carcinoma

The study was carried out with the approval of the Ethics Committee of Gaozhou People’s Hospital and was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki 56 pairs of fresh ESCC tissues and adjacent normal tissues were obtained from this hospital. Informed consent was obtained from all participants before sample collection. The collected tissue specimens were stored at − 80 °C.

Normal esophageal epithelial cells (HET-1A) were acquired from ATCC (Manassas, VA, USA). TE-1, Eca-109, KYSE150, and HUVEC cell lines were obtained from Procell (Wuhan, China), while KYSE70 cells were provided by TongPai Biotechnology (Shanghai, China). KYSE70 cells and KYSE150 cells were maintained in RPMI-1640 medium, and other cells were cultivated in a DMEM medium. During cell culture, medium was supplemented with 1% penicillin/streptomycin (Procell) and 10% fetal bovine serum (FBS). The culture conditions were 37 °C, 5% CO₂, and 95% humidity.

SiRNAs for circCHSY1 (si-circCHSY1-1 and si-circCHSY1-2), shRNA against circCHSY1 (sh-circCHSY1), miR-1229-3p mimic and inhibitor, Tectonic-1 (TCTN1) overexpressed plasmid (pcDNA-TCTN1) and their negative controls were all provided by GenePharma (Shanghai, China). When the cell aggregation rate reached 80%, 50 nM of miRNA mimic, different substances (20 pmoL of siRNA, 0.8 µg of plasmids, and 100 nM of miRNA inhibitor) were used for cell transfection by Lipofectamine^™ 3000 (Invitrogen, Carlsbad, CA, USA) depending on the experimental requirements.

In line with the guidebook of GoldenstarTM RT6 cDNA Synthesis Kit (Chinese Indigo), the isolated RNA by TRIzol^™ was used to synthesize complementary DNA. Premix Ex Taq^™ II (Chinese Indigo) was used for real-time PCR. RNA quantification was performed by the 2^−∆∆Ct method. Primer’s sequences used in this paper are listed in Table 1.

Three μg of RNA was incubated with RNase R (Epicentre Biotechnologies, Madison, WI, USA) for 20 min under normal conditions. Expression of circCHSY1 and its linear gene CHSY1 was measured by qRT-PCR.

1 × 10⁴ TE-1 or Eca-109 cells for each assay were respectively planted onto the 96-well plates and transfected with different substances (si-circCHSY1, si-NC, miRNA inhibitors, inhibitor control, miRNA mimics, mimic control, pcDNA-TCTN1 and pcDNA-NC) into the cells. Cell proliferative ability was then analyzed following the suggestion of CCK8 kit (Beyotime, Shanghai, China).

EdU Imaging Detection Kit (KeyGEN, Nanjing, China) was utilized to assess cell proliferation. ESCC cells (5 × 10⁴/well) with different transfections were seeded in 96-well plates and subsequently went through incubation with an EdU working reagent. Subsequently, the cells were fixed with paraformaldehyde (4%) and infiltrated with Triton X-100 solution (0.5%). After the addition of Click-iT EdU reaction buffer (100 µL), images of EdU-positive cells were recorded under a fluorescence microscope.

Cells with different substances (si-circCHSY1, si-NC, miRNA inhibitors, inhibitor control, miRNA mimics, mimic control, pcDNA-TCTN1, and pcDNA-NC) were cultured at 37 °C for 48 h. 1 × 10⁵ cells were harvested and suspended in 5 μL of Annexin V-FITC and 10 μL of PI (Beyotime) for 10 min. Apoptotic cells were detected by flow cytometry (Countstar, Shanghai, China).

Matrigel (Millipore, Billerica, MA, USA) was utilized for cell invasion analysis. In simple terms, 1 × 10⁵ of ESCC cells with different substances (si-circCHSY1, si-NC, miRNA inhibitors, inhibitor control, miRNA mimics, mimic control, pcDNA-TCTN1, and pcDNA-NC) were added to the upper chambers. After 24 h of culture, the invaded cells were finally photographed under a microscope.

HUVEC cells (3 × 10⁴ cells/well) were treated with si-circCHSY1, si-NC, anti-miR-1229-3p, miR-NC, pcDNA-TCTN1, pcDNA-NC, anti-miR-NC or miR-1229-3p, alone or jointly. The cells were suspended into 96-well plates coated with 60 μL Matrigel (Millipore) and cultured with supernatant of ESCC cells for 5 h. Then, a microscope was used to determine the number of capillary-like branches.

Proteins were extracted from cells and tissues using RIPA lysis solution (Beyotime). The proteins were separated by 5% SDS-PAGE and then transferred onto membranes. The membranes were blocked with non-fat milk for 1.5 h and incubated overnight with primary antibodies including anti-PCNA (ab92552, 1:1000, Abcam, Cambridge, MA, USA), anti-Bax (ab32503, 1:1000, Abcam), anti-Bcl-2 (ab182858, 1:1000, Abcam), anti-TCTN1 (15,004-1-AP, 1:500, proteintech, Chicago, Illinois, USA) and anti-GAPDH (ab9485, 1:1000, Abcam). The protein bands were visualized using ECL substrates (Millipore).

Wild (WT) or mutant (MUT) circCHSY1 (or TCTN1) containing the potential binding sites for miR-1229-3p were inserted into the psi-CHECK vector (Promega, Madison, WI, USA). TE-1 and Eca-109 cells were subsequently co-transfected with the constructed plasmids described above with miR-1229-3p or miR-NC. The cells were divided into groups including circCHSY1^WT + miR-1229-3p group, circCHSY1^WT + miR-NC group, circCHSY1^MUT + miR-1229-3p group, circCHSY1^MUT + miR-NC group, TCTN1 3’UTR^WT + miR-1229-3p, TCTN1 3’UTR^WT + miR-NC, TCTN1 3’UTR^MUT + miR-1229-3p, and TCTN1 3’UTR^MUT + miR-NC group. After 48 h, luciferase activity was analyzed using a dual-luciferase assay system.

Eca109 and TE-1 cells were collected and incubated in a RIPA lysis buffer. The lysate was incubated with biotin-labeled oligonucleotide probes and Streptavidin-coupled Dynabeads (Invitrogen). The magnetic beads were incubated at 4 °C for 3 h and the combined RNA in the complex was purified using TRIzol. MiR-1224-3p, miR-1253, and miR-1229-3p expression were assessed by qRT-PCR.

PARIS^™ kit (Invitrogen) was applied to determine the position of circCHSY1 in cells. The cytoplasmic and nuclear components of ESCC cells were isolated and collected for qRT-PCR analysis of circCHSY1 expression.

Specific pathogen-free BALB/c nude female mice (n = 10) aged 4–5 weeks weighing 18–22 g were purchased from Hunan Slyke Jingda Experimental Animal Co., LTD (Changsha, China) and maintained in the absence of specific pathogens. The experiments were approved by the Administrative Panel on Laboratory Animal Care of Gaozhou People’s Hospital. All animal procedures were carried out following the National guidelines of the Animal Care and Use of laboratory animals, the ARRIVE guidelines and the Basel Declaration. The maximal tumor size permitted by the Ethics Committee of Gaozhou People’s Hospital is not exceed 2000 mm³ in size and 20 mm in diameter. The generated xenograft tumors in this study were not exceeded this range. Two groups of mice were injected subcutaneously with TE-1 cells stably expressing sh-NC or sh-circCHSY1. The density of cell suspension was adjusted to 1 × 10⁶ cells/mL, and each mouse was injected with the cell suspension (200 μL) to establish an in vivo model. Tumor volume was recorded weekly. After 4 weeks, the mice were anesthetized using pentobarbital sodium, and the tumor was dissected for tumor weight and gene expression analysis. Confounders were not controlled in this experiment. Immunohistochemistry (IHC) was conducted to analyze PCNA, Bax, Bcl-2, and Ki67 expression as instructed [18]. Antibodies were obtained from Abcam (Shanghai) Trading Co., LTD.

Data in this study were analyzed by GraphPad Prism version 5.0 and presented as mean ± standard deviation. Distribution normality was evaluated using the Shapiro-Wilk normality test. The Student’s t-test was used to analyze the difference between 2 groups. Analysis of variance was performed to analyze data among three or more groups. P-value less than 0.05 meant a significant difference.

CircCHSY1 was generated through the back-splicing of exon 3, which was located on the CHSY1 gene (Fig. 1A). Compared with normal adjacent tissues, circCHSY1 was significantly increased in tumor tissues (Fig. 1B). Moreover, ESCC cell lines (TE-1, Eca-109, KYSE150 and KYSE79) showed a higher expression of circCHSY1 than HET-1A cells (Fig. 1C). TE-1 and Eca-109 cells were selected for the following study due to higher expression of circCHSY1 in the two types of cells. RNase R treatment assay revealed that circCHSY1 was almost unaffected, but CHSY1 mRNA was reduced (Fig. 1D and E), indicating the circular character of circCHSY1. These data demonstrated that circCHSY1 might be associated with ESCC progression.

The functional experiments were performed using siRNA of circCHSY1 due to its high expression in ESCC cells. Firstly, the interference efficiency of circCHSY1 siRNAs was validated, and these two siRNAs significantly reduced circCHSY1 expression without affecting linear genes (Fig. 2A and B). si-circCHSY1-1 (named as si-circCHSY1 in the following studies) was chosen for subsequent investigation due to its stronger inhibitory effect on circCHSY1 expression. CircCHSY1 knockdown significantly reduced cell viability, as evidenced by CCK-8 analysis (Fig. 2C and D). Similar results were obtained in the EdU test, which showed a decrease in cell proliferation after circCHSY1 silencing (Fig. 2E). The number of apoptotic cells was increased after transfection with si-circCHSY1 (Fig. 2F and G). When circCHSY1 expression was down-regulated, Bax was increased significantly, while PCNA and Bcl-2 levels were decreased significantly (Fig. 2H and I). In addition, circCHSY1 silence could inhibit cell migration, invasion, and tube formation (Fig. 2J–L). These findings suggest that inhibition of circCHSY1 has anti-tumor effects in ESCC cells.

CircCHSY1 was mainly presented in the cytoplasm (Fig. 3A and B). Through the prediction of the circinteractome online database, we found that circCHSY1 is potentially bound to miR-1224-3p, miR-1253, and miR-1229-3p. As shown in Fig. 3C and D, only miR-1229-3p expression was significantly upregulated in the circCHSY1 probe group among the three miRNAs. Thus, miR-1229-3p was chosen for the present study. MiR-1229-3p was significantly reduced in ESCC tissues and cells (Fig. 3E and F), so we speculated whether circCHSY1 could target to miR-1229-3p. The binding sites of circCHSY1 for miR-1229-3p are shown in Fig. 3G. The success of miR-1229-3p overexpression is presented in Fig. 3H. As expected, miR-1229-3p overexpression decreased the luciferase activity of circCHSY1^WT, which further proved the combination of the two RNAs (Fig. 3I and J).

The data of CCK-8 (Fig. 4A and B) and EdU analysis (Fig. 4C) showed that circCHSY1 absence inhibited cell proliferation, but transfection of miR-1229-3p inhibitor reversed this inhibition. As shown in Fig. 4D, the cell apoptosis rate was significantly enhanced by decreasing the expression of circCHSY1, while anti-miR-1229-3p reduced the effect. At the protein level, the circCHSY1 deficiency increased Bax protein expression and decreased PCNA and Bcl-2 protein levels, while anti-miR-1229-3p attenuated these effects (Fig. 4E and F). In addition, knockdown circCHSY1 inhibited cell migration, invasion, and angiogenesis. However, these effects were mitigated by co-transfection with miR-1229-3p inhibitors (Fig. 4G–I).

Subsequently, TCTN1 expression was significantly elevated in ESCC tissues and cells (Fig. 5A–C). miR-1229-3p and TCTN1 expression had a negative correlation (Fig. 5D), while TCTN1 and circCHSY1 showed a positive correlation (Fig. 5E). Next, miR-1229-3p was found to target the TCTN1 3’-UTR regions (Fig. 5F). Luciferase reporter activity was repressed in TE-1 and Eca-109 cells co-transfected with miR-1229-3p and WT vector but not with the MUT vector (Fig. 5G and H). In addition, miR-1229-3p decreased TCTN1 expression, and miR-1229-3p downregulation enhanced TCTN1 expression (Fig. 5I and J). In summary, miR-1229-3p was an upstream molecule that targeted TCTN1 and negatively regulated TCTN1 expression.

The function of the miR-1229-3p/TCTN1 axis in ESCC cells was further explored through rescue experiments. The protein level of TCTN1 was significantly increased after transfection with TCTN1 overexpressed plasmid (Fig. 6A). CCK-8, EdU, and flow cytometry experiments showed that TCTN1 overexpression plasmid reversed the effects of miR-1229-3p on cell proliferation and apoptosis (Fig. 6B–G). In addition, miR-1229-3p resulted in significant suppression of migration, invasion, and angiogenesis in both ESCC cell lines, while overexpression of TCTN1 neutralized these effects (Fig. 6H–J).

As shown in Fig. 7A and B, the deletion of circCHSY1 led to a decrease in TCTN1 expression, which was reversed by the miR-1229-3p inhibitor.

Tumor size and tumor weight were suppressed by decreasing circCHSY1 expression (Fig. 8A–C). Besides, circCHSY1 and TCTN1 expression were decreased in the sh-circCHSY1 group (Fig. 8D and E). sh-circCHSY1 treatment decreased Ki-67, PCNA, and Bcl-2 levels and increased Bax level in tumors (Fig. 8F).