Clinical specimens. A total of 56 paired GC and non-cancerous tissues (≥5 cm from tumor margin; 26 males and 30 females; age range: 40-78 years old) were obtained during surgery at the First Affiliated Hospital of Jinzhou Medical University (Jinzhou, China) during July 2010 and June 2013. The tissues were cut into sections and snap‑frozen using liquid nitrogen immediately after surgery, and then stored at ‑80˚C until further use. The patients were sorted into circRNA_100269 high- or low-expression groups according to average circRNA_100269 level. All the biopsies were checked by two independent pathologists, and clinicopathological features of enrolled patients were presented in Table I. Overall survival rates were evaluated using Kaplan-Meier method. Written informed consents were signed by patients, and all the tissues were kept anonymously. The present study was approved by the Medical Ethics Committee of the First Affiliated Hospital of Jinzhou Medical University.
Immunohistochemistry analysis. Immunostaining of phospho-PI3K was performed on paraffin-embedded sections. The biopsies were dewaxed using xylene and rehydrated through a graded ethanol series and water. Antigen retrieval was carried out using microwave treatment in 10mM sodium citrate for 15 min. Tissues were then incubated with FCS at room temperature for 30 min and with primary antibody against phospho-PI3K (1:200; cat. no. ab182651, Abcam) in a humid chamber at 4°C overnight. The following day, the sections were rinsed in TBS for three times and incubated with biotinylated secondary antibody (1:100; Dako, UK). Antigen was then detected using a streptavidin-biotin-peroxidase system (ABC kti; Dako, UK). Staining intensity was quantified using Image J software (version 1.46; NIH, Bethesda, MD, USA).
Cell culture. Four human GC cell lines (AGS, SGC-7901, BGC-823 and MKN-45) as well as one normal human gastric epithelial cell line (GES-1) were obtained from the American Type Culture Collection (Manassas, VA, USA). The cells were maintained within DMEM supplemented with 10% fetal bovine serum (FBS; GE Healthcare Life Science), 100 µg/ml streptomycin and 100 U/ml penicillin (HyClone; GE Healthcare Life Science), and kept at 37˚C in a humid incubator supplied with 5% CO2.
Cell transfection. To establish cell model overexpressing circRNA_100269, wildtype (o/e-circRNA_100269) or mutant (o/e-NC) circRNA_100269 fragment was amplified by PCR, and subsequently integrated into pcDNA3.1 vector (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). In order to produce circRNA_100269 knockdown model, shRNA against circRNA_100269 (sh-circRNA_100269) or negative control (sh-NC) were purchased from Genepharm Co. Ltd. (Shanghai, China). Following annealing, shRNA were inserted into lentiviral pU6-Luc-Puro vector (Genepharm Co. Ltd.). Cells without any shRNA treatment were used as the control group. Up- or down-regulation of circRNA_100269 was further evaluated by RT-qPCR. All the transfections were conducted using Lipofectamine®2000 (Invitrogen; Thermo Fisher Scientific, Inc.). Eight hours following transfection, culture supernatants were replenished by fresh DMEM containing 10% FBS. For the inhibition of PI3K signalling, cells were treated with LY294002 (10mM; Cell Signaling Technology, Beverly, USA).
RNA extraction and reverse transcription-quantitative polymerase chain reaction (RT-qPCR). Total RNA from clinical specimens or cells was isolated using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), and then reverse transcribed into cDNA by a PrimeScript™ RT kit (Takara Biotechnology Co., Ltd., Dalian, China). The target cDNA was amplified using SYBR Green PCR Master Mix (TaKaRa Biotechnology Co., Ltd.), and the reaction was carried out on an ABI 7500 Real-Time PCR system (Thermo Fisher Scientific, Inc.) Endogenous GAPDH was used as internal control. The sequences of forward and reverse primer were as follows: circRNA_100269, 5’-CTATGTCGTTCAAGGAAGA-3’ and 5’-GTCTTGAGACTGGGTATGT-3’; PI3K, 5’-AACACAGAAGACCAATACTC-3’ and 5’-TTCGCCATCTACCACTAC-3’; E-cad, 5’-AAGAAGCTGGCTGACATGTACGGA-3’ and 5’-CCACCAGCAACGTGATTTCTGCAT-3’; vimentin, 5’-AGAACCTGCAGGAGGCAGAAGAAT-3’ and 5’-TTCCATTTCACGCATCTGGCGTT-3’; Snail, 5’-TTTCTGGTTCTGTGTCCTCTGCCT-3’ and 5’-TGAGTCTGTCAGCCTTTGTCCTGT-3’; GAPDH, 5’-GCAAGAGCACAAGAGGAAGA-3′ and 5’-ACTGTGAGGAGGGGAGATTC-3’. PCR program was 95˚C for 5 min, followed by 45 cycles of 95˚C for 15s, 60˚C for 20s and 72˚C for 10s. Relative expression levels were analyzed using 2-∆∆Cq method.
Western blotting. Total protein was extracted by radioimmunoprecipitation buffer (Beyotime Institute of Biotechnology, Shanghai, China). The concentration of isolated protein was measured by bicinchoninic acid assay (Beyotime Institute of Biotechnology). Equal amount (30 μg) of protein samples were loaded onto SDS-PAGE gel and then transferred onto a PVDF membrane (EMD Millipore, Billerica, MA, USA). Membranes were subsequently blocked with tris-buffered saline (TBS) supplemented with 5% skimmed milk at room temperature for 2 h and then incubated in corresponding primary antibodies as follows: PI3K (1:2000; cat. no. ab140307; Abcam), Akt (1:1000; cat. no. 9272; Cell Signaling Technology), p-AktS473 (1:1000; cat. no. 4058; Cell Signaling Technology), p53 (1:1000, cat. no. 9282; Cell Signaling Technology), Bcl-2 (1:1000, cat. no. 15071; Cell Signaling Technology), cyclin D1 (1:2000, cat. no. 2926; Cell Signaling Technology), E-cad (1:1000, cat. no. 3195; Cell Signaling Technology), vimentin (1:2000; cat. no. 5741; Cell Signaling Technology), snail (1:1000; cat. no. 3879; Cell Signaling Technology), Bax (1:1000; cat. no. 2772; Cell Signaling Technology), cas-9 (1:2000; cat. no. 14697; Cell Signaling Technology), cleaved cas-3 (1:1000; cat. no. 9661; Cell Signaling Technology), cleaved PARP (1:1000; cat. no. 9541; Cell Signaling Technology) or GAPDH (1:1,000; cat. no. sc-47724; Santa Cruz Biotechnology Inc.) at 4˚C overnight. The following day, membranes were incubated in horseradish peroxidase-labelled anti-mouse (1:5,000; cat. no. sc-2371; Santa Cruz Biotechnology Inc.) or -rabbit IgG (1:5000; cat. no. sc-2357; Santa Cruz Biotechnology Inc.) at room temperature for 1h. Protein bands were visualized using ECL protein detection reagent (Pierce Biotechnology; Thermo Fisher Scientific, Inc). Signals were quantified by densitometric method using Image J software (version 1.46; NIH, Bethesda, MD, USA).
Cell proliferation assay. Cells were harvested 24 h post-transfection, and a total of 1x104 cells were placed in each well of a 96-well plate. The proliferation of cells was evaluated by MTT assay (Sigma-Aldrich; Merck KGaA) at day 1, 2, 3 and 4. Briefly, 20 µl of MTT solution was added into the cells, and the plates were incubated at 37˚C for 4 h, the absorbance at 570 nm was measured using a microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).
Wound healing assay. Cells were seeded on a 6-well plate at a density of 4x105 cells/well and transfected with corresponding plasmids. After the cells reached a confluency of ~80-100%, they were pre-treated with 10ug/mL mitomycin C (Thermo Fisher Scientific) for two hours prior to wound healing assay. Then, cell monolayer was scratched by a straight line with a sterile micropipette tip and rinsed three times with PBS, which was replaced with fresh DMEM. Subsequently, the scratch width changes were monitored immediately following the scratch was formed and after 6, 12 and 24 h. The images were captured using a fluorescence microscope (magnificationx100, Olympus Corporation, Tokyo, Japan). The migration of cells was analysed by ImageJ using the following formula: Migration area ratio = proportion of closed wound area/entire field of view area.
Transwell assay. Cells were pre-treated with 10ug/mL mitomycin C (Thermo Fisher Scientific) for two hours prior to assay. A total of 1x105 cells were diluted in FBS-free culture media and seeded onto the upper chamber (BD Biosciences, Franklin Lakes, New Jersey, USA) that is pre-coated with Matrigel® (Sigma-Aldrich, St. Louis, MO, USA). Subsequently, 500 µl culture media containing 10% FBS was added into the lower chamber. Following overnight incubation, non-invasive cells were detached by a cotton swab, whereas invaded cells in the lower chamber were fixed using 4% paraformaldehyde and stained in 0.5% crystal violet. The numbers of invasive cells were counted in five randomly selected fields using an inverted light microscope (magnificationx200, Olympus Corporation, Tokyo, Japan).
Cell cycle and apoptotic rate analysis. Cells were inoculated onto a 6-well plate with a density of 4x105 cells/well following the treatments with o/e-circRNA_100269 or o/e-NC, respectively. Subsequently, cells were collected using low-speed centrifugation (1000rpm) at 4˚C for 5 mins. Cell pellets were rinsed and re-suspended using PBS, subsequently fixed in 70% pre-chilled ethanol and kept at 4˚C for two days. Cells were lysed prior to flow cytometry, an then centrifuged and re-suspended using propidium iodide (PI, Sigma-Aldrich, USA) staining buffer containing 50 µl/ml of PI as well as 250 µl/ml RNase A. The distribution of cell cycle was determined by a flow cytometer (BD Biosciences, USA) and then analysed by Flowjo version 7.6 software (Flowjo LLC, USA). In order to evaluate cell apoptosis, cell suspension was incubated with PI in dark at 4˚C for 30 mins and then stained by 5 µl annexin V-FITC (JingMei Biotech, Beijing, China). Apoptotic rate was examined using flow cytometer (BD Biosciences, USA) and then interpreted by Flowjo version 7.6 software (Flowjo LLC, USA).
In vivo nude mouse xenograft. Female BALB/C nude mice (~5 weeks old) with the weight of ~19g were obtained from The Laboratory Animal Research Centre of Nantong University (Nantong, China). The mice were routinely housed in a temperature- (22±2˚C) and humidity-controlled (~60%) environment, under a 12-h dark/light cycle with libitum access to food and water for more than three days before the experiments. Mice were randomly grouped (n=5 in each group) and injected with AGS cells transfected with sh-NC or sh-circRNA_100269. Briefly, a total of 1x107 cells were diluted in 200μl PBS and then injected into the back of mice subcutaneously. Mice with developing tumors were closely monitored four times a week. Six weeks following injection, the mice were sacrificed. Tumor tissues were removed and evaluated. Tumor volume was calculated as follows: V (mm3) =(length x width2)/2. To trigger the metastasis, 1x105 of cells were suspended using 20 μl PBS and subsequently injected into the lateral tail vein of mice. Experiments contained 10 mice inoculated with tumor cells on day 0. Post-injection, mice were randomly sorted into experimental groups and then evaluated 42 days later. The protocols of animal experiments were approved by the Ethics Committee of our Hospital. The animal work was carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
Statistical analysis. Data were presented as means ± standard error of mean and interpreted using SPSS 17.0 (SPSS, Inc., Chicago, IL, USA). The significance of differences was determined by one-way analysis of variance (ANOVA) or the Student’s t-test. A student-Newman-Keuls test was conducted following ANOVA. The associations between RNA expression levels were evaluated by Pearson’s correlation analysis. P<0.05 was considered to indicate a statistically significant difference.