Ethics statement
All participants or their guardians provided signed informed consent before enrollment. The Ethics Committee of The First Affiliated Hospital of Guangxi Medical University provided Ethical Approval for human experiments that conformed to the Declaration of Helsinki.
Bioinformatics analysis
The differentially expressed genes (DEGs) in GC were analyzed through bioinformatics online tool Gene Expression Profiling Interactive Analysis (GEPIA) with |logFC | > 1, p value < 0.05 as the screening criteria. The key word “GC” was searched in GeneCards database to retrieve the genes related to GC. hTFtargets which was a bioinformatics prediction website was utilized to predict the targeting factors of transcription factors and obtain the possible binding sites between the transcription factor and the lncRNA (Supplementary Fig. 1). The overlapping parts of ncRNA related to GC and candidate targeting factors in hTFtargets were found by jvenn online tool. The miRNAs binding to the lncRNA were predicted by the LncBase database. miRNAs related to GC were presumed by MNDR v3.1 and MISIM v2.0. Jvenn online tool was adopted to find the overlap of candidate miRNA of the lncRNA and miRNAs related to GC. The expression of the miRNA in GC samples was analyzed by dbDEMC 2.0 website (http://www.picb.ac.cn/dbDEMC/). To predict the downstream orchestrating factors of the miRNA, TargetScan and miRDB were employed to predict the downstream target genes of the miRNA, and jvenn tool was applied to find the overlaps of target genes, GC-related genes and upregulated genes. We analyzed the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways involved by genes through the Network Analyst website (https://www.networkanalyst.ca/) to further obtain the related pathways of target genes.
Clinical samples
Forty GC and adjacent normal tissues were attained from primary GC patients. All patients had not undergone chemotherapy or radiotherapy. The tissues were harvested during gastroscopy and immediately preserved at −80 °C.
Cell incubation and transfection
Incubation of GC cell lines SNU-1 (CC-Y1478) and NCI-N87 (CC-Y1398) (EK-Bioscience, Shanghai, China; http://www.elisakits.cn/Index/product/ccid/147.html) was conducted in RPMI 1640 complete medium (12633012; Gibco, Invitrogen, USA; https://www.thermofisher.cn/) encompassing 10% fetal bovine serum (FBS; 10099141; Gibco) and 1% penicillin-streptomycin (15070063; Gibco). Incubation of HEK-293T (CC-Y1010; EK-Bioscience) was conducted in Dulbecco’s modified Eagle’s medium (DMEM) (10569044; Gibco) encompassing 10% FBS and 1% penicillin-streptomycin. GC cells were cultured in a 37 °C incubator with 5% CO2.
The cells were transfected with small interfering RNA (si)-negative control (NC), si-c-Myc#1, or si-c-Myc#2; transfected with si-NC, si-c-Myc, or si-c-Myc + overexpression (oe)-KCNQ1OT; transfected with si-NC, si-KCNQ1OT1, or si-KCNQ1OT1 + miR-556-3p inhibitor; or transfected with miR-NC, miR-556-3p mimic, inhibitor NC, miR-556-3p inhibitor, oe-NC, oe-CLIC1, oe-CLIC1 + dimethyl sulfoxide (DMSO), or oe-CLIC1 + LY294002.
The cell models of overexpressing NC and overexpressing CLIC1 were constructed in GC cells which named oe-NC and oe-CLIC1 respectively. GC cells were incubated in a 6-well plate with 2 mL ligand per well. 800 μL fresh virus solution was mixed with 800 μL FBS, followed by supplementing of Polybrene until the final concentration was 6 μg/mL. After 12–24 h of culture, the medium was changed to complete medium to culture sequentially at 37 °C with 5% CO2. After infection for 48 h, the cells were cultured in puromycin (1 μg/mL) medium for 2 weeks to screen stable transfected cell lines. When the cells did not dead in puromycin-encompassing medium, the cell protein was extracted. Then the overexpression of genes or proteins was confirmed by reverse transcription quantitative polymerase chain reaction (RT-qPCR) or Western blot analysis.
Cells were seeded into a 6-well plate and transfected with miR-556-3p mimic, miR-NC, miR-556-3p inhibitor, or inhibitor NC (50 nM, GenePharma, Shanghai, China) by Lipofactamine 2000 reagent (Invitrogen, Carlsbad, California, USA). Then the packaging mixture was transfected into GC cells. si-NC, si-c-Myc, or si-KCNQ1OT1 (30 nmol/L; Invitrogen) were transfected into GC cells (Each gene had at least two siRNA interference sequences, and the siRNA sequence with the best knockdown efficiency was chosen for the following experiments) by Lipofactamine 2000 reagent (Invitrogen). Cells were harvested 48 h later. The knockdown efficiency of si-KCNQ1OT1 and c-Myc was verified by RT-qPCR.
Cell grouping and plasmid transfection information are shown in Supplementary Table 1.
Isolation, purification, and identification of EVs
Isolation and purification of EVs: According to the user’s manual, EVs in supernatant of SNU-1 cell medium were obtained through the EVs extraction kit (ExoQuick, SBI, CA, USA). SNU-1 cells were cultured in a 10 cm dish. When cell confluence reached 90%, the medium was replaced with 1640 medium without FBS. Subsequent to 24-h incubation, supernatant of cell medium was collected from 20 mL medium (1 × 107 cells) and centrifuged at 3000 × g for 15 min. Then 1000 μL ExoQuick Exosome precipitation solution was supplemented to 1000 μL supernatant. After that, the mixture was frozen at 4 °C for 30 min. Subsequent to 30-min centrifugation at 1500 × g, 100 μL sterile 1 × PBS was used to resuspend the EVs precipitate. The supernatant was used as the control without exosome for further experiments.
Identification of EVs: The EV precipitates were identified by a transmission electron microscopy (TEM). Western blot analysis was adopted to identify the protein markers CD9, CD81, CD63, and TSG101 of EVs.
TEM observation
EVs isolated and purified from tumor were taken out and dropped onto the sealing membrane. The side of Formvar membrane from the copper mesh was placed on the suspension. 2–3 copper meshes were prepared for each EVs sample. After the cover was closed, the copper mesh was allowed to absorb EVs in a dry environment for 20 min. Then 100 μL PBS was added to the sealing membrane. The copper mesh (Formvar film face down) was placed on the PBS with tweezers for 5 min and washed twice. For all steps, the surface of Formvar film remained wet with the other side being dry. The copper mesh was positioned on 50 μL glutaraldehyde (1%) for 5 min before 2-min placing with 100 μL double distilled water (washed 8 times). The copper mesh was plated on 50 μL uranyl oxalate droplet for 5 min and then on 50 μL methylcellulose UA solution for 10 min on ice. After the copper mesh was removed by the stainless steel ring, the excess liquid was gently sucked off with a filter paper, leaving a thin layer of methyl cellulose membrane which was dried in the air for 10 min. The image was observed at 100KV by H7650 TEM (Hitachi, Japan).
Particle size analysis of Nanosight EVs
Particle size analysis of nanosight nanoparticles: 20 μg EVs were dissolved in 1 mL PBS and vortex for 1 min to keep the EVs evenly distributed. Then the particle size distribution (PSD) of EVs was measured by NanoSight nanoparticle tracking analyzer (NTA, Malvern Instruments, Malvern, UK).
Nanosight (Merkel Technologies Ltd., Israel, NTA version NTA 3.2 Dev Build 3.2.16) NTA was employed to characterize EV concentration. The size distribution and concentration of EVs were determined by ZetasizerNano ZS90 (Malvern instruments, Zetasizer 7.12). The PSD map was established with the x-axis representing PSD (nm) and the y-axis representing the relative percentage. In addition, PSD map based an intensity was generated according to the particle size (nm) of the x-axis and the intensity (a.u.) of the y-axis.
Tracing experiment of EVs in GC cells
EVs were labeled with membrane marking dye PKH67 green fluorescence (Sigma, St Louis, MO, USA). EVs secreted by SNU-1 cells were labeled with PKH67 dye. The labeled EVs were co-cultured with NCI-N87 cells for 30 min, 2 h, and 24 h, respectively. Then NCI-N87 cells were immobilized with 4% paraformaldehyde. Subsequent to 10-min nuclei staining with 4′,6-diamidino-2-phenylindole (DAPI, C1025, 10 μg/mL, Beyotime, Nantong, China), uptake of labeled EVs by recipient GC cells was observed by Nikon eclipse fluorescence microscope (Nikon, Tokyo, Japan).
Western blot analysis
EV suspension was concentrated or digested with trypsin. The cultured cells were lysed with enhanced Radio-Immunoprecipitation assay cell lysis buffer (BOSTER, Wuhan, Hubei, China) encompassing protease inhibitor at 4 °C for 15 min before 15-min centrifugation at 6000 × g. The protein content in the supernatant was determined through a bicinchoninic acid kit (23227, Thermo Fisher Scientific, Waltham, Massachusetts, USA). Sodium dodecyl sulfate polyacrylamide gel electrophoresis gel was prepared before protein denaturation and electrophoresis. Proteins were separated and electroblotted to a polyvinylidene fluoride membrane that underwent 1-h sealing at ambient temperature with 5% bovine serum albumin (BSA) to block the nonspecific binding, and probed overnight with diluted primary antibodies (Abcam, Cambridge, UK) to CD9 (ab195422, 1:1000), CD81 (ab109201, 1:5000), CD63 (ab134045, 1:1000), Tsg101 (ab83, 1:1000), Calnexin (ab22595, 1:1000), c-Myc (ab32072, 1:1000), CLIC1 (ab219265, 1:1000), pho (Y607)-PI3K (1:1000, ab182651), Total-PI3K (1:1000, ab32089), pho (T308)-AKT (1:1000, ab38449), and Total-AKT (1:1000, ab8805) with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a normalizer at 4 °C. Afterwards, the membrane was reprobed for 1 h with horseradish peroxidase-tagged goat anti-mouse (ab6808, 1: 2000) and goat anti-rabbit (ab6721, 1: 5000) secondary antibodies (Abcam) at ambient temperature. The membrane was put into electrogenerated chemiluminescence (ECL) reaction solution at ambient temperature before 1-min incubation. Then the excess ECL solution was discarded, followed by membrane sealing with plastic film. The X-ray film was placed in the cassette, followed by 5–10 min-exposure. Subsequent to developing and immobilizing, the image J analysis software was applied for quantification of gray level of each band.
RT-qPCR
Total RNA extraction was conducted in GC cells or EVs by TRIzol® reagent (15596-018, Solarbio, Beijing, China) before cDNA synthesis. To detect lncRNA and mRNA expression, a PrimeScript™ RT-qPCR Kit (RR047A, TaKaRa, Tokyo, Japan) was applied for reverse transcription. For miRNA detection, Poly(A) Tailing Kit (B532451, Shanghai Sangon Biotechnology Co., Ltd., Shanghai, China; containing universal PCR primer R) was used for reverse transcription to obtain the cDNA of miRNA with PolyA tail. RT-qPCR was conducted on LightCycler 480 system (Roche Diagnostics, Pleasanton, CA, USA) using SYBR® Premix Ex TaqTM II Kit (DRR081, Takara). Primers for amplification were provided by Shanghai General Biotechnology Co., Ltd. (Shanghai, China; Supplementary Table 2). The 2−△△CT method was employed for calculation of the relative transcription levels of target genes that were normalized to GAPDH (for lncRNA and mRNA) and U6 (for miRNA). All investigations involved at least 3 wells, each repeated in triplicate.
In situ hybridization (ISH)
As per the directions of an ISH kit (BOSTER), paraffin-embedded tissue samples were utilized for ISH to assess KCNQ1OT1 expression. The paraffin slices were dewaxed and hydrated before 10-min immersion in xylene. The slices were soaked in xylene for another 10 min after changing xylene. The slices were positioned in 100%, 75%, and 50% alcohol for 5 min, and then in distilled water for 3 min. Following immobilizing, pre-hybridization, and hybridization, the slices were supplemented with biotinylated rat anti-digoxin and biotinylated peroxidase, and stained with a diaminobenzidine Kit (Solarbio) in the light of the instructions.
Dual-luciferase reporter assay
The fragment sequence encompassing the action site was obtained. The predicted binding site of KCNQ1OT1 was inserted into pMIR-REPORT luciferase vector (Promega, Madison, WI, USA) for the generation of the luciferase reporter vector wild type (WT)-KCNQ1OT1. Based on this vector, the binding site of KCNQ1OT1 was mutated through a Quick-change site-directed mutagenesis Kit (Agilent Technologies, Palo Alto, CA, USA) to establish the mutant vector mutant (Mut)-KCNQ1OT1.
The fragment sequence encompassing the action site was harvested. CLIC1 3′-untranslated region (UTR) sequence encompassing predicted miR-556-3p binding site was inserted into pMIR-REPORT luciferase vector (Promega) to generate luciferase reporter vector CLIC1-WT. Based on this vector, the binding site of CLIC1 was mutated by Quick-change site-directed mutagenesis Kit (Agilent Technologies) to construct the mutant vector CLIC1-Mut.
HEK-293 cells were cultured in a 24-well plate for 24 h. After reached 50–60% confluence, the cells were co-transfected with Lipofectamine 2000 transfection reagent according to the following combinations: miR-NC/miR-556-3p mimic and CLIC1-WT; miR-NC/miR-556-3p mimic and CLIC1-Mut, miR-NC/miR-556-3p mimic and WT-KCNQ1OT1; miR-NC/miR-556-3p mimic and Mut-KCNQ1OT1. Following 48-h transfection, cell lysate was attained to detect luciferase activity. Relative luciferase activity was evaluated as per the manuals of a dual-luciferase reporter kit (E1910; Promega). Next, the 80–100 μL lysis buffer was added. Subsequent to 30-min passive lysing on ice and a shaker, cells were centrifuged at 4 °C and 12,000 × g for 10 min. Then 10 μL supernatant was drawn and positioned on a special 96-well plate for luciferase. The cells were added with 50 μL firefly luciferase buffer (FB) or 50 μL Renilla luciferase buffer (RB) in turn, and then detected by enzyme labeling instrument.
Immunofluorescence staining
Flag-labeled c-Myc was transfected into GC cells. Following 48-h transfection, EVs were isolated from supernatant of the culture medium and co-cultured with GC cells. After co-culture for 30 min, 2 h, and 24 h severally, cells were immobilized with pre-cooled 4% paraformaldehyde for 20 min. The cells were incubated with Triton X-100 (Sigma) and the membrane was broken for 15 min to increase their permeability. The cells were sealed for 1 h with 5% BSA and probed overnight with anti-FLAG (ab18230, Abcam) primary antibody on the shaking table. The primary antibody was discarded, followed by 1-h cell incubation with conjugated fluorescent secondary antibody at ambient temperature. After the secondary antibody was discarded, cells were stained with DAPI (10 μg/mL, C1002, Beyotime) without light for 15 min. Then anti-fluorescence attenuation sealing agent was dripped onto the glass slide. After the cover glass was covered, the diagonal of the film was sealed with nail polish, and the sample was observed by fluorescence microscope (Leica Biosystems, Shanghai, China).
Cell counting kit (CCK)-8
Cells were incubated in a 96-well plate (1 × 103 cells/well) with 100 μL medium encompassing 10% FBS for 3 days. Then, cell number was evaluated as per the protocols of a CCK-8 kit (K1018, Apexbio, USA). Next, 10 μL CCK-8 solution was supplemented to each well for another 4-h culture at 37 °C. Optical density was assessed at 450 nm on days 1, 2, and 3. Five parallel wells were set up for each experiment. Proliferation rates relative to controls were applied to plot cell growth curves that were examined using one-way analysis of variance (ANOVA).
Flow cytometry
The apoptosis rate of GC cells was detected using Annexin V-allophycocyanin (APC) apoptosis detection kit (556547, BD Pharmingen, San Jose, USA). GC cells were made into a suspension of 1 × 106 cells/mL with 1× Binding Buffer. Falcon tube was supplemented with 100 μL cell suspension and 5 μL Annexin V in the light of the kit instructions, mixed gently, and kept in dark at ambient temperature (20–25 °C) for 15 min. Subsequent to once cell washing with 1× Binding Buffer, the supernatant was removed. Cell precipitate was resuspended with 100 μL of 1× Binding Buffer, and supplemented with 5 μL propidium iodide, followed by 15-min placing in dark at ambient temperature (20–25 °C). 400 μL of 1× Binding Buffer was supplemented to each test tube, and the results were determined using a FACScan flow cytometry system (Becton Dickinson, San Diego, CA, USA) within 1 h.
Scratch test
On the bottom of the 6-well plate, a horizontal line was drawn through a ruler and a marker every 0.5–1 cm with that each well shall pass through at least five lines. GC cells (1 × 106) were added to the 6-well plate and grew to fusion. Cells were incubated overnight in medium with or without EVs. Then, a sterile gun head (200 μL) was applied to scratch horizontal lines perpendicular to the back. Cells were removed and medium without serum was supplemented before cell incubation in a 5% CO2 and 37 °C incubator. After 0, 12, and 24 h of culture, the growth and metastasis of GC were observed and recorded by inverted microscope. The migration closure was calculated: Migration area (%) = (A0 − An)/A0 × 100, where A0 represented the initial wound area and An indicated the wound area at the time of measurement.
Transwell migration and invasion experiment
In order to analyze cell migrative ability, 2 × 105 transfected cells were suspended in DMEM (200 μL) without serum and supplemented to Transwell apical chamber without Matrigel reagent (356234, Becton, Dickinson and Company, NJ, USA).
To analyze cell invasive ability, Matrigel reagent from Becton, Dickinson and Company was diluted in DMEM without serum (1:10). The diluted Matrigel (100 μL) was supplemented to Transwell apical chamber before more than 30 min of culture. Then the cells were seeded into the upper chamber coated with Matrigel (BD Pharmingen). For both assays, DMEM encompassing 10% FBS (600 μL) was supplemented to the basolateral chamber before 24-h incubation at 37 °C. Subsequent to 15-min cell immobilizing and 15-min 0.1% crystal violet staining, the positive cells were observed by inverted light microscope (Carl Zeiss, Jena; Evotec Biosystems, Hamburg, Germany) and photographed. The positive cells were counted by ImageJ software.
Statistical analysis
SPSS 21.0 (IBM Corp. Armonk, NY, USA) was employed in statistical analysis. The Measurement data are summarized as mean ± standard deviation. Paired t-test was used to analyze the data between cancer tissues and adjacent normal tissues. Unpaired t-test was conducted for comparison between the other two groups. One-way ANOVA was utilized to compare among multiple groups, while two-way ANOVA was applied for comparing the data at different time points, followed by Tukey’s post-hoc test. p < 0.05 was considered to be statistically significant difference.
