2.1 CircRNA Array analysis on public dataset
The expression profiles of CircRNA in three gastric cancer (GC) patients’tissues and their corresponding begin adjacent tissues were downloaded from Kong et al [12] .at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE131414. We set P < 0.05 as the statistical criteria for selecting differentially expression.
2.2 Clinical samples
Tumor biobank of Fujian Cancer Hospital (Fuzhou, China) provided 66 pairs of gastric cancer and adjacent normal gastric tissues. From June 2018 to June 2020, the patients undergone surgical treatment before any radiotherapy or chemotherapy in Fujian Cancer Hospital (Fuzhou, China). The fresh tissue quickly frozen down in operation room in dry ice and preserved at -80 ℃until the study. The approval was granted from the Ethics Committee of Fujian Cancer Hospital (Fuzhou, China), and the written informed consent of each participant was acquired.
2.3 Cell culture
Cell lines, including human gastric cancer cell lines (AGS, MGC803, MKN45and MKN28) and human normal gastric epithelial cell line (GES-1), were obtained from Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The culture of cells was performed in in RPMI-1640 (GIBCO, Grand Island, NY, USA) containing 100 IU / ml penicillin, 100 ug / ml streptomycin, 10% FBS (GIBCO) and with 5% CO2 at 37℃.
2.4 RNA preparation and real-time quantitative
For examining mRNAs expression, miRNAs, and circRNAs, RT-qPCR was performed with the use of SYBR1 Green Dye identification systems (Roche, Switzerland) following the description of existing research [22]. To carry out quantitative analysis, we employed the 2- ΔΔCt approach. U6 and GAPDH were used as the control, separately. Supplementary Table S1 lists all primer sequences.
2.5 Fluorescence in situ hybridization (FISH)
The culture of cells was performed on the coverslip for the phase of exponential. After fixing with 4% PFA, the gastric cancer cells were permeated with 0.25% Triton X-100, hybridized with Cy3-labeled circ-0001451 FISH probe(RiboBio, Guangzhou), then incubated with FITC-streptavidin (Invitrogen, Carlsbad, USA), dyed in DAPI, sealed with rubber adhesive and imaged with a fluorescence microscope.
2.6 Oligonucleotides, plasmids and transfection
The overexpression and knockdown vectors were constructed byHanheng Biological Technology Company (Beijing, China) [9] and miRNA mimics and inhibitors made by RiboBio (Guangzhou, China). The transfection was done with Lipofectamine2000 reagent (Invitrogen, Carlsbad, USA). In HEK293T cells, lentivirus particles with the above vectors were produced. The infection of cells was achieved using lentivirus a multiple of infection (MOI) of 10, and selected with 2 ug/ml puromycin.
2.7 MTS test and cellular cycle analysis
The gastric cancer cells transfected with hsa_circ_001451 were plated in 96-well plates (5000 cells / well), after 24, 48 and 72 h, the cells were co-cultured with 20 µl MTS in 100 µl of 1640 medium for 2-hour, and the absorbance at 450 nm was examined using a microplate reader (Molecular Devices, Sunnyvale, CA, USA). The cell cycle was analyzed with the Muse™cell cycle kit Muse™Cell Analyzer (Millipore, Burlington, MA, USA) following the manufacturers’instructions [13].
2.8 Colony Formation Assay
Different infected cells (about 500 cells / well) were inoculated into 6-well plates. The cells were cultured for 14 days, washed twice in PBS, stained with 0.2% crystal violet, and photographed with image scanning equipment (GE, Piscataway, NJ, USA). The clone of more than 50 cells was determined as a proliferative colony and calculated their numbers with Image J.
2.9 Wound healing and transwell invasion tests
The wound healing test and Transwell invasion test were carried out following the previous description [13]. The images of cell migrating to the empty line after 48 h or invaded cells on the other side of membrane were taken with microscope (Zeiss, Axio observer A1, GER) and analyzed with Image J. The mean and standard deviation of five regions in the respective in the groups were calculated and compared for statistical significance.
2.10 RNA immunoprecipitation (RIP) test
The RIP test was carried out with EZ-Magna RIPRNA-Binding Protein Immunoprecipitation Kit(Millipore, Billerica, MA, USA) following themanufacturers’ instruction. The cells were lysed in 100 µlRIP lysis buffer, diluted with 900 µl RIPimmunoprecipitation buffer. Next, the cell suspension was mixed with magnetic beads conjugated withcontrol anti-IgG antibody or anti-Argonaute 2 (Ago2). After incubating overnight, IgG antibodies wereacquired and cleaned with the use of RIP washing buffer and digested with Proteinase K. The RNA was extracted with TRIzol reagent (Invitrogen, CA, USA) and analyzed with RT-qPCR.
2.11 Biotinylated RNA pull-down test
The RNA pull-down test was carried out for the evaluation of the interaction between circ_001451 and miR-197-3p following the previous description [8]. Biotinylated miR-197-3p probe was used, and quantitated with RT-qPCR to test circ_0001451/miR-197-3p.
2.12 Dual luciferase reporter test
The reporter vector psi-check2 (Promega, Madison, WI, USA) was introduced with hsa_circ_001451 or 3’UTR sequence of MT3 (the sequence obtained from UCSC) for constructing recombinant psi-check2. HEK293 cells (Thermo Fisher Scientific) transfected with miR-197-3p mimics or miR-control were seeded in 96-well plates and co-transfected with empty psi-check2 or recombinant psi-check2 for 48 h with the use of Lipofectamine™ 2000 (Invitrogen). The recombinant vector containing hsa_circ_001451 wild type was named as WT and the point mutations one as MUT. The luciferase activity was determined with a dual-luciferase reporter test system (Promega,Madison, WI, USA) following the manufacturers’instructions. The activity of Renilla luciferase was used to normalize the activity of firefly luciferase and expressed as the control percentage.
2.13 Western blotting analysis
Protein levels of MT3, MMP-7, E-cadherin , N-cadherin, VIMENTIN, PCNA, CCND1, BCL-2 were determined with western blotting . A set of antibodies was used, including MT3 (1:500, abcam , Cat No.ab224060), MMP-7 (1:1000, CST, Cat No. 3801S), BCL-2 (1:1000, CST, Cat No.5071S) , CCND1(1:1000, CST, Cat No. E3P5S), PCNA(1:1000, CST, Cat No. D3H8P), VIMENTIN(1:1000, CST, Cat No.5741S), N-cadherin (1:1000, CST, Cat No.13116S), E-cadherin(1:1000, CST, Cat No. 14472S).
2.14 Transplantation of zebrafish with circ-0001451 overexpression gastric cancer cells AGS
After the embryo developed to 48 HPF (hours post fertilization), the fry was anesthetized and placed in the groove of agar plate. Then the cells digested by trypsin were transplanted into the center of yolk sac or ventral yolk space of juvenile zebrafish by microinjection instrument. Each zebrafish was injected with 150–200 tumor cells. A group of at least 30 fish were injected to ensure that 10 fish were included in the group. At 2 and 72 hours after transplantation, 10 fish were photographed under the stereofluorescence microscope, and the growth of tumor cells in zebrafish was observed and counted. The fluorescence areas of zebrafish tumors were photographed by capture software with a body fluorescence microscope; Tumor fluorescence area was examined by Image J software.
2.15 Statistical analysis
All data were processed by spss19.0 (America). Pearson correlation study was carried out. For examining the difference between two groups, two tailed t test was used. For the evaluation of the relationship of hsa_ circ_0001451 level with other features, the paired or non-parametric Kruskal Wallis test was carried out. P< 0.05 was determined as statistical significance.