CircRNAs expression profiles in gastric cancer
Human circRNA microarray analysis was performed by Yunxu Biotechnology (Shanghai). Total RNA was extracted from gastric cancer tissues and from adjacent noncancerous tissues, according to Trizol protocol (Takara, Japan). Total RNA was used for circRNA microarray analysis (Agilent human circRNA). RNA labeling and hybridization were performed in accordance to the manufacturer’s indications. Agilent scanner and the Feature Extraction 10.7.1.1 software (Agilent Technologies) were used to obtain the microarray raw data. Microarray results were analyzed using the GeneSpring GX 12.5 software (Agilent Technologies). Differentially expressed circRNAs were identified by using a moderated t-test and Benjamini-Hochberg correction (adjusted P≤0.05 and fold change≥2). Differentially expressed genes were measured similarity with circRNAs by the Pearson correlation and were analyzed in Gene Ontology (GO) analysis, KEGG pathway analysis, and cluster analysis.
Differential expression analysis of miRNAs
The miRNA expression, clinical, and meta- and manifest data on GC were derive from The Cancer Genome Atlas (TCGA). MiRNA-sequencing data was downloaded, a Bioconductor package based on the R language, to screen differentially expressed miRNAs between GC tissue and adjacent normal tissue. Differentially expressed miRNAs with FDR values <0.05 and |log2FC| >1 were considered significantly.
CircRNA targeting miRNA prediction and prognosis of miRNAs
According to the results of the differential circRNA expression analysis, targeted miRNAs of has_circRNA_0008727 were predicted via Cancer-Specific CircRNA Database (CSCD). We used the keyword “hsa_circ_0008278” in the CSCD and obtained the miRNAs that could bind to the hsa_circ_0008278. Furthermore, we extracted the expression and survival time of miRNAs related GC via TCGA. The overall survival curves were analyzed using the R language survival package. Finally, we took the intersection among the differential expression miRNAs, the predicted targeting miRNAs and the survival-related miRNAs. According to the outcome of intersection, Kaplan-Meier survival curve were constructed to evaluate diagnostic and prognostic values.
Construction of the circRNA-miRNA regulatory network
According to the results of the above analysis. Type and network txt files were designed. The interaction ceRNA network of circRNA–miRNA was constructed using Cytoscape software 3.6.0 .
Human tissue samples
This research was approved by the Institutional Ethics Committee of Nanjing Medical University. All cancerous and adjacent noncancerous tissues were gathered from the Jiangsu Province Hospital (Nanjing, China), which were eventually diagnosed as GC via postoperative pathological analysis. These patients were not undergone the chemotherapy or radiotherapy before resection.
Cell culture and transfection
Normal gastric epithelial cell line GES-1 and GC cell line SGC-7901 were obtained from the Shanghai Institute of Cell Biology (Shanghai, China). Cells were cultured in 10% fetal ox-like serum PRIM-1640 (GIBCO, Rockville, USA) medium supplemented with 1% penicillin at 37 degrees C in a humidified atmosphere of 5%CO2/95% air incubator. Cells cultured on the six-well plate were transfected. Plasmid mediated YY1 or circRNA_0008278 overexpression vector, siRNA targeting YY1 or circRNA_0008278 vector, miR-378 mimic, inhibitor and their respective controls were purchased from Genechem. Cells are transiently transfected with plasmids by utilizing Lipofectamine RNAimax (Invitrogen, Carlsbad, CA, USA). Finally, cells were used in further experiments 48h post-transfection.
RNA fluorescence in situ hybridization (FISH)
AGS cells were grown to the exponential phase and were 80–95% confluent at the time of fixation. After pre-hybridization (1 × PBS/0.5% Triton X-100), the cells were hybridized in hybridization buffer (40% formamide, 10% dextran sulfate, 1 × Denhardt’s solution, 4× SSC, 10 mM DDT, 1 mg ml− 1 yeast transfer RNA, 1 mg ml−1sheared salmon sperm DNA) with DIG-labelled probes specific to circRNA_0008278 and miR-378 at 60 °C overnight. The expression level of circRNA_0008278 and miR-378 was evaluated by FISH. The probe signals were determined with the Fluorescent in Situ Hybridization Kit (RiboBio, Guangzhou, China) according to the manufacturer’s guidelines. Alexa Fluor 555-labeled circRNA_0008278 probes and Alexa Fluor 488-labeled miR-378 probes were designed and synthesized by Life Technologies. Nuclei were counterstained with 4,6-diamidino-2-phenylindole (DAPI). Images were acquired using a fluorescence microscope (Eclipse E600; Nikon Corporation, Tokyo, Japan).
RNA isolation, RNase R digestion and qRT-PCR
Total RNA was isolated from human tissue samples and cells by use of Trizol reagent (Life Technologies, CA, US), then we measured the concentration by NanoDrop 2000 Spectrophotometer (Thermo Scientific, Wilmington, DE, USA). RNA was reverse transcribed the total RNAs by Reverse Transcription Kit (Takara, Tokyo, Japan) and then to measure the expression by qRT-PCR using the SYBR Green PCR Kit (Takara) on the ABI 7900HT (Applied Biosystems). All tests were rehashed multiple times. U6 was an internal control. Fold changes in expression were determined utilizing 2−ΔCt or 2−ΔΔCt.
Protein extraction and Western blotting
Total proteins were extracted using RIPA buffer, moreover, protein concentrations generally are determined by the Bicinchoninic Acid (BCA) Protein Assay. The Western Blot was conducted as followed: first, protein was isolated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Second, we exchanged proteins to polyvinylidene difluoride membrane (PVDF) (Millipore, Billerica, USA). Third, we use 5% bovine serum albumin (BSA) to block the membrane for 2h. Fourth, antibodies YY1 and β-actin (Abcam, Cambridge, USA) were incubated at 4 °C overnight. Fifth, the membranes were washed multiple times. Sixth, the secondary antibody was incubated at room temperature for 1h. Finally, proteins were envisioned with upgraded chemiluminescence reagents (Pierce, Rockford, IL, USA).
Dual-luciferase reporter assay
Cells were plated in the 24-well plate for 24 hours. Then, we transfected with a mixture of firefly luciferase reporter, pRL-CMV Renilla luciferase reporter, and miR-378 mimic or inhibitor. Firefly and Renilla luciferase activities were estimated with the Dual-Luciferase Report Assay System after 48h incubation (Promega, Madison, WI, USA)
Cell Migration and Invasion Assay
In the upper chamber, we seeded 4x104 cell with serum-free medium. Cell was seeded with 600ul medium containing 10% serum in the lower chamber. After 48h, cells were colored with dying solution, and tallied under 20x amplification.
Grow cells in RPMI-1640 supplemented with 10% FBS. Gently scratch the monolayer with a 1 ml pipette tip over the focal point of the well. Scratch another straight line opposite to the main line to make a cross in each well. In the wake of scratching, delicately wash the well twice with medium to expel the withdrew cells. Renew the well with new medium. Micrographs were taken after a traditional scratch wound assay of SCG-7901 cells 48 hours after wounding.
Colony formation assay
The SCG-7901 cell was trypsinized, and we plated 1 × 103 cells in a 6-well plate, then incubated at 37 °C for 10 days. Cells were dyed with crystal violet staining solution (Beyotime, China), then tallied cell colonies.
Cell proliferation assay
1x 104 cells per ml in RPMI-1640 medium with 10% FBS and penicillin (100 U/ml) were evenly seeded in 96-well plates. Then, mixed 10ul/well CCK-8 solution and incubated with CCK8(Beyotime, Nantong, China) solution at 37 °C for 1h. Finally, we measured the absorbance at 450 nm by use of a microplate reader (Bio-Rad, USA).
GraphPad Prism 7.00 Software (USA) and SPSS version 22.0 (SPSS, USA) were used to conduct the statistical analyses. Expression is presented as mean ± SEM of at least three independent experiments for all results. One-way analysis of variance (ANOVA) or student’s t test was performed to determine statistical differences among two or more groups. Sensitivity, specificity, and area under the curve (AUC), including 95% confidence interval (CI), were computed with the aid of the constructed receiver-operating characteristic (ROC) curves  to calculate the optimum cut-off values. The survival analysis included log-rank tests and Kaplan–Meier analyses. A P values less than 0.05 were considered statistical significance. For all figures: *, P < 0.05.