Specimen collection and the ethics committee

Forty-seven pairs of human CRC tissues and matched healthy colorectal tissues were collected up from CRC sufferers in the Chengyang People’s Hospital, and the written informed consent was signed by the CRC cases before surgery. Obtained tissues were stored at − 80 °C. The Ethics Committee of the Chengyang People’s Hospital approved this study.

Cell culture and exposure to Sev

Human CRC cell lines (HCT116 and SW620) and normal human colonic epithelial cell line NCM460 were purchased from Procell (Wuhan, China). HCT116 and NCM460 cells were grown in Roswell Park Memorial Institute-1640 (RPMI-1640; Procell, Wuhan, China), and SW620 cells were cultivated in Leibovitz’s L15 media (L15; Procell, Wuhan, China) at 37 °C in humid condition with 5% CO2. Media were supplemented with 10% fetal bovine serum (FBS; Procell, Wuhan, China) and 1% penicillin/streptomycin (Procell, Wuhan, China).

For Sev treatment, HCT116 and SW620 cells were placed in a sealed container with humid atmosphere of 37 °C. Sev (Seebio Biotech, Shanghai) was mixed with 95% air and 5% CO2 using a volatilization tank, and gas monitor was used to adjust concentrations of Sev to 1.7%, 3.4% and 5.1%, respectively. Then, the cells were treated with the various concentrations of Sev for 30 min.

Plasmid construction and oligonucleotide synthesis

The small RNAs against hsa_circ_0000231 (si-hsa_circ_0000231#1, si-hsa_circ_0000231#2 and si-hsa_circ_0000231#3), miR-622 mimics (miR-622), miR-622 inhibitors (anti-miR-622) and control groups (si-NC, NC and anti-NC) were synthesized by GenePharma (Shanghai, China). The overexpression plasmids of hsa_circ_0000231 (hsa_circ_0000231) and control group (circ-NC) were built by Geneseed (Guangzhou, China). Plasmids or oligonucleotides were transfected into cells with TurboFect Reagent (Thermo Fisher, Waltham, MA, USA). The synthesized sequences of oligonucleotides were si-hsa_circ_0000231#1 5′-ACTGAACAGATAAGGGTTTAA-3′, si-hsa_circ_0000231#2 5′-CTGAACAGATAAGGGTTTAAA-3′, si-hsa_circ_0000231#3 5′-CACTGAACAGATAAGGGTTTA-3′, miR-622 5′-ACAGUCUGCUGAGGUUGGAGC-3′, anti-miR-622 5′-GCUCCAACCUCAGCAGACUGU-3′, si-NC 5′-CCTCTACCTGTCGCTGAGCTGTAAT-3′, NC 5′-UUUGUACUACACAAAAGUACUG-3′ and anti-NC 5′-CAGUACUUUUGUGUAGUACAAA-3′.

Quantitative real-time polymerase chain reaction (qRT-PCR)

A miRNeasy Mini Kit (Qiagen, Valencia, CA, USA) was firstly employed to isolate RNA. After that, cDNA was synthesized with a FastKing RT Kit (Tiangen, Beijing, China) or Qiagen reverse transcription kit (Valencia, CA, USA). Then, SuperReal PreMix Color (Tiangen, Beijing, China) was utilized to detect the content of circRNA/miRNA/mRNA. Obtained data were assessed with the 2−∆∆Ct method with U6 or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a reference. The sequences of forward and reverse primers were hsa_circ_0000231 5′-ACTTAGCAGCAGCTCCAC-3′ and 5′-CCACTTCTGTCAGCCATT-3′; miR-622 5′-ACACTCCAGCTGGGACAGTCTGCTGAGGT-3′ and 5′-TGGTGTCGTGGAGTCG-3′; GAPDH 5′-GGTCACCAGGGCTGCTTT-3′ and 5′-GGAAGATGGTGATGGGATT-3′; U6 5′-CTCGCTTCGGCAGCACA-3′ and 5′-AACGCTTCACGAATTTGCGT-3′.

RNase R treatment assay

Cultured HCT116 and SW620 cells were harvested and RNA was isolated according to the method as described above. Then, obtained RNA was incubated with RNase R (RNase R+) (3 U μg−1 RNA; Geneseed, Guangzhou, China) and without RNase R (RNase R-) at 37 °C, respectively. About 30 min later, RNeasy MinElute Cleaning Kit (Qiagen, Valencia, CA, USA) was employed to purify RNA. Hsa_circ_0000231 content was determined by qRT-PCR with linear GAPDH mRNA (linear mRNA) as a reference.

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay

Cultivated HCT116 and SW620 cells were diluted in RPMI-1640 (Procell, Wuhan, China) and L15 media (Procell, Wuhan, China), respectively, and grown in 96-well plates. Sixteen hours later, cells were treated with Sev (1.7%, 3.4% or 5.1%) (Sigma, St. Louis, MO, USA), si-hsa_circ_0000231#1, si-hsa_circ_0000231#2, hsa_circ_0000231 or anti-miR-622 based on the defined purposes with 0% Sev (Control), si-NC, circ-NC or anti-NC as a reference. At 48 h after treatment, MTT solution (Beyotime, Shanghai, China) was incubated with cells for 4 h. Then, dimethyl sulfoxide (Sigma, St. Louis, MO, USA) was added into plates to dissolve formazan. Cell viability was determined after samples were analyzed with a microplate reader (Thermo Fisher, Waltham, MA, USA) with a wavelength at 490 nm.

Cell colony formation assay

HCT116 and SW620 cells diluted in RPMI-1640 (Procell, Wuhan, China) or L15 media (Procell, Wuhan, China) were grown in 6-well plates. After various treatments, cells were continued to be cultured for about 14 days. During cell culture, media were renewed every 3 days. Then, cell supernatant was removed, and paraformaldehyde (Sigma, St. Louis, MO, USA) and crystal violet (Sigma, St. Louis, MO, USA) were dripped into plates, respectively. Cell colony-forming ability was determined by analyzing the number of colonies. A colony was considered when cell numbers over 50.

DNA content quantitation assay

Cells were collected, and eluted with phosphate buffer solution (PBS; Procell, Wuhan, China). Afterwards, the cells were fixed with 70% ethanol (Millipore, Bradford, MA, USA). RNase A (Solarbio, Beijing, China) was incubated with the cells at 37 °C in water. Cells were stained with propidium iodide (PI; Solarbio, Beijing, China) in dark. Finally, samples were analyzed with a flow cytometry (Thermo Fisher, Waltham, MA, USA).

Annexin V-fluorescein isothiocyanate (Annexin V-FITC) and PI double staining assay

Cell apoptosis was assessed with an Annexin V-FITC/PI apoptosis detection kit (Solarbio, Beijing, China). In brief, harvested cells were suspended in Binding buffer (Solarbio). Cell supernatant was discarded by centrifuging at 300g for 12 min, followed by the incubation with Annexin V-FITC (Solarbio, Beijing, China) and PI (Solarbio, Beijing, China) in dark. Finally, cells were diluted in PBS (Procell, Wuhan, China) and assessed with flow cytometer (Thermo Fisher, Waltham, MA, USA).

Caspase-3 activity assay

A caspase-3 activity detection kit (Beyotime, Shanghai, China) was employed to detect caspase-3 activity. Briefly, cell supernatant was discarded and cells were collected up by centrifuging. Then, lysis buffer (Beyotime, Shanghai, China) was utilized to lyse cells. Supernatant was harvested by centrifugation. Detection buffer (Beyotime, Shanghai, China) and acetyl-Asp-Glu-Val-Asp p-nitroanilide were mixed with the samples, and the output of wavelength at 405 nm was determined with a microplate reader (Thermo Fisher, Waltham, MA, USA). Finally, caspase-3 activity was determined by assessing the output.

Wound-healing assay

Cell migration was demonstrated in this part. In short, cells were grown in 6-well plates after diverse treatments. Cell wounds were created when the confluence of cells reached approximately 100%. Cells were cultured in RPMI-1640 (Procell, Wuhan, China) or L15 media (Procell, Wuhan, China) without FBS (Procell, Wuhan, China). After 24 h, the migratory capacity of cells was determined by analyzing the width of wounds under microscope (Nikon, Tokyo, Japan) at a 40× magnification.

Transwell invasion assay

The invasive ability of CRC cells was revealed with transwell chambers with Matrigel (Corning, Madison, New York, USA). In brief, HCT116 and SW620 cells were cultivated in the upper chambers supplemented with FBS-free RPMI-1640 (Procell, Wuhan, China) and L15 media (Procell, Wuhan, China), respectively. RPMI-1640 (Procell, Wuhan, China) and L15 media (Procell, Wuhan, China) containing 15% FBS (Procell, Wuhan, China) were added into the lower chambers. At 24 h after culture, supernatant was removed, and cells were orderly incubated with paraformaldehyde (Sigma, St. Louis, MO, USA) and crystal violet (Sigma, St. Louis, MO, USA). Results were determined by counting cell numbers in the lower chambers under microscope (Nikon) with a 100× magnification.

Dual-luciferase reporter assay

Circular RNA Interactome online database (https://circinteractome.nia.nih.gov/api/v2/mirnasearch?circular_rna_query=hsa_circ_0000231&mirna_query=hsa-miR-622&submit=miRNA+Target+Search) was firstly utilized to predict the binding sites of hsa_circ_0000231 in miR-622. Then, the wild-type (wt) and mutant (mut) plasmids of hsa_circ_0000231 were constructed by Geneseed Co., Ltd. (Guangzhou, China) and named as hsa_circ_0000231-wt and hsa_circ_0000231-mut, respectively. Built plasmids were co-transfected into HCT116 and SW620 cells with miR-622 or NC with TurboFect Reagent (Thermo Fisher, Waltham, MA, USA) based on the instruction of manufacturer. Forty-eight hours later, luciferase activities were detected by a Dual-Lucy Assay Kit (Solarbio, Beijing, China). Renilla luciferase activity was used as a reference.

RNA immunoprecipitation (RIP) assay

Cells were lysed with RIP lysis buffer (Millipore, Bradford, MA, USA). Then, the lysates were incubated with magnetic beads coated with the antibodies against argonaute-2 (Anti-Ago2; Abcam, Cambridge, UK) or immunoglobulin G (Anti-IgG; Abcam, Cambridge, UK) for 24 h. Proteins were digested with proteinase K (Millipore, Bradford, MA, USA), and the contents of hsa_circ_0000231 and miR-622 were detected by qRT-PCR.

In vivo assay

The Vital River Laboratories (Beijing, China) provided male BALB/c nude mice (5 weeks of age), and the mice were fed in a sterile environment. All mice were divided into four groups (n = 5, respectively). SW620 cells were cultivated for 24 h after treatment of 5.1% Sev, and the cells were transfected with hsa_circ_0000231 or circ-NC. After 48 h, the cells were digested and diluted in 200 μL PBS (Procell, Wuhan, China), which were hypodermically inoculated right anterior temporal region of mice. Tumor volume was measured every one day. Five days later, nude mice were sacrificed by intraperitoneal injection of xylazine (10 mg kg−1; Seebio Biotech, Shanghai, China) and the neoplasms were harvested. The weight of every tumor was measured. A part of each tumor was kept for further assessment in RNA or protein level. The Animal Care and Use Committee of the Chengyang People's Hospital approved this research.

Western blot analysis

Tissues were lysed using RIPA buffer (Beyotime, Shanghai, China), and lysates were loaded onto 12% bis–tris-acrylamide gels (Thermo Fisher, Waltham, MA, USA). The separated protein bands were electrotransferred onto polyvinylidene fluoride membranes (Millipore, Bradford, MA, USA), and then immersed in 5% nonfat dry milk (Solarbio, Beijing, China). After that, the membranes were incubated with anti-BCL2-associated x protein (anti-Bax) (1:5000; Abcam, Cambridge, UK), anti-matrix metalloprotein 2 (anti-MMP2) (1:5000; Abcam, Cambridge, UK), anti-MMP9 (1:3000; Abcam, Cambridge, UK), anti-caspase 3 (anti-t-caspase 3) (1:5000; Abcam, Cambridge, UK), anti-cleaved-caspase 3 (anti-C-caspase 3) (1:8000; Abcam, Cambridge, UK), anti-proliferating cell nuclear antigen (anti-PCNA) (1:3000; Abcam, Cambridge, UK), anti-MYC proto-oncogene, bHLH transcription factor (anti-c-Myc; 1:1000; Abcam, Cambridge, UK), anti-KRAS proto-oncogene, GTPase (anti-K-RAS; 1:1000; Abcam, Cambridge, UK), anti-B-Raf proto-oncogene, serine/threonine kinase (anti-BRAF; 1:2000; Abcam, Cambridge, UK) and anti-GAPDH (1:15,000; Abcam, Cambridge, UK), respectively. Then, the membranes were incubated with secondary antibody (1:8000; Abcam, Cambridge, UK). Finally, RapidStep ECL Reagent (Millipore, Bradford, MA, USA) was used to visualize the protein bands. GAPDH was chosen as a control.

Statistical analysis

Data derived from three independent duplicate tests were assessed by GraphPad Prism (GraphPad Software, La Jolla, CA, USA), and expressed as means ± standard deviations (SD). Significant differences in Spearman correlation analysis and overall survival curve were compared with Spearman’s correlation test and log-rank test, respectively. Additionally, significant differences were compared with two-tailed Student’s t-tests or Wilcoxon rank-sum test between the two groups and with one-way analysis of variance (ANOVA) with Tukey’s test or Kruskal–Wallis test between the multiple groups. p value < 0.05 was deemed as statistical significance.