Gastric cancer tissues and paired adjacent normal gastric epithelial tissues were obtained from the Third Affiliated Hospital of Soochow University from 2003 to 2015. Tissue specimens were obtained through needle puncture and stored in liquid nitrogen at -80°C. All experiments using the human tissues were conducted under the approval of the Third Affiliated Hospital of Soochow University. All patients gave informed consent.

All cells in this study were obtained from the Chinese Academy of Sciences, Shanghai Institutes for Biological Sciences. HEK293 cells were cultured in DMEM (Gibco, Melbourne, Australia) supplemented with 10% fetal bovine serum (FBS, Gibco, Melbourne, Australia). Gastric epithelial cell GES-1 and gastric cancer cells AGS, SGC-7901, and HGC-27 were cultured in RPMI 1640 (Gibco, Melbourne, Australia) supplemented with 10% FBS.

The full-length human lncRNA GRIK1-AS1 cDNA was cloned into pLVX-IRES-Puro vector (Clontech, CA, USA) to generate GRIK1-AS1 expression plasmids. The siRNA targeting IFIT2 was purchased from GenePharma (Shanghai, China).

siRNA duplexes (100nM), miR-375 mimic, and corresponding negative control (NC) oligonucleotides were transfected into cells mediated by Lipofectamine 3000 (Thermo Fisher, CA, USA) according to manufacturer’s instructions.

To establish individual stable cells, lentivirus was used. Lentiviral particles in cell culture supernatant were harvested at 72h after lentivirus transfection into HEK293 cells. The AGS and SGC-7901 cells were then infected with lentivirus. Stably transfected cells were selected by puromycin (2µg/mL, Sigma-Aldrich, MO, USA) for 1 week at 48h after lentiviral infection.

Cell proliferation assay was performed using CellTiter 96^® Non-Radioactive Cell Proliferation Assay (MTT) kit (Promega) according to the manufacturer’s instructions. Cells (1×10⁴cells/mL) were seeded into a 96-well plate (100 µl/well) and cultured in an incubator with 5% CO₂ at 37°C. Each well was added with 10 µl MTS solution and then incubated at 37°C for 2h. The spectrophotometric absorbance at 590 nm was measured for each sample. All the experiments were repeated 3 times in triplicates.

The cell invasion ability was measured by Matrigel-coated transwell assay, in which transwell chambers (8-μm pore size; Corning) were applied. Cells (1×10⁵) were seeded into the upper chamber with serum-free RPMI 1640 medium, and RPMI 1640 medium supplemented with 20% fetal bovine serum was added into the lower chamber. After 48h incubation, cells on the top surface of filters in the upper chambers were cleaned up gently with a cotton swab. Then, the filters were fixed in 4% PFA for 15 min and stained with 0.1% crystal violet for 10 min. After washed by PBS for 3 times, cells migrated across the filters were imaged and the number of invaded cells was calculated by counting five random views under a microscope. The experiment was performed in triplicates and repeated 3 times.

A wound healing assay was performed to determine the cell migration capability. Cells were seeded in 6-well plates. When the cells were about 90% confluent, the artificial wounds were cut using a 200-ul pipette tube. After 24h, the wound closure was observed and imaged under a microscope. The migration distance was indicated by the fraction of cell coverage across the line. Triplicates were required for each experiment.

Total RNA was extracted using TRIzol according to the manufacturer’s instructions. First-strand cDNA was synthesized using Superscript II (Invitrogen) and 1μg of total RNA was used in each cDNA synthesis reaction. SYBR green Universal Master Mix reagent (Roche) and primer mixtures were used for the real-time qPCR. GAPDH was used as the internal reference for mRNAs.

The miRNAs expression level was measured using the All-in-One™miRNAqRT-PCR Detection Kit (GeneCopoeia)according to the manufacturer’s instructions.U6 small RNA was used as the reference. The primers used for real-time qPCR were as follows: LncRNA GRIK1-AS1, forward: 5’-ATGGAGGATGCAGCAAAAGGG-3’; reverse: 5’-TTCTGTGTCCTGGTTGTTTCTC-3’;IFIT2, forward: 5’-AAGCACCTCAAAGGGCAAAAC-3’; reverse: 5’-TCGGCCCATGTGATAGTAGAC-3’;GAPDH, forward: 5’-CTGGGCTACACTGAGCACC-3’; reverse: 5’-AAGTGGTCGTTGAGGGCAATG-3’;U6, forward: 5’-CTCGCTTCGGCAGCACA-3’; reverse: 5’-AACGCTTCACGAATTTGCGT-3’.

SurePrep™ Nuclear or Cytoplasmic RNA Purification Kit (Fisher BioReagents) was applied to isolate nuclear and cytoplasmic fractions according to the manufacturer’s instructions. RNA levels of lncRNA GRIK1-AS1, the nuclear control RNU6-1, and the cytoplasmic control GAPDH were analyzed by real-time qPCR.

Cells were fixed in 4% PFA for 15 min and then permeabilized by 0.5% TritonX-100 for 15 min at 4°C.0.5% TritonX-100 was used to permeabilize the cells for 15 min at 4°C. Digoxigenin (DIG) labeled GRIK1-AS1 probe or control probe mix haswas performed to incubate cells for 4h at 55°C. After 2xsaline-sodium citrate briefly washing 3 times with 5 min every time, signals were detected by Horseradish peroxidase (HRP)-conjugated anti-DIG secondary antibodies (Jackson). DAPI was used to counterstain nuclear. A con-focal laser scanning microscope was used for images.

Total proteins were extracted and separated using SDA-PAGE gels. IFIT2 antibody (Abcam MA, USA) was used at a concentration of 1:2000. GAPDH antibody (Santa Cruz Biotechnology, 1:2000, CA. USA) was used as a loading control.

GRIK1-AS1-WT luciferase reporter vector was conducted by cloning GRIK1-AS1 cDNA containing predictive miR-375 binding site into the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, WA, USA). GRIK1-AS1-Mut vector was generated by insertion of mutant GRIK1-AS1 containing point mutations in the miR-375 binding site. Likewise, wild-type and mutant IFIT2 3’-UTR fragments were cloned into pmirGLO vector to conduct the IFIT2 3’UTR-WT and IFIT2 3’UTR-Mut luciferase reporter vectors. The miR-375 or miR-NC was co-transfected with the reporter vector into HEK293 cells using Lipofectamine 3000 (Invitrogen). The luciferase activity was measured at 48h after transfection using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s instructions. Triplicates were required for each experiment.

3’-end biotinylated miR-375, miR-375-Mut, or candidate miRNAs were transfected into cells at a final concentration of 20 nmol/L. After 24h, cells were harvested and incubated in the cell lysate with streptavidin-coated magnetic beads (Ambion, Life Technologies). The biotin-coupled RNA complex was pulled down and analyses of the abundance of GRIK1-AS1 were conducted by real-time qPCR.

All experiments were performed using 3 independent repeated experiments with cells. GraphPad Prism 8.0 was applied for statistical analyses. Data in all figures are presented as the mean ± SEM. Statistical significance was determined by Student’s t test, One-way ANOVA and Two-way ANOVA. For all statistical tests, the 0.05 level of confidence (2-sided) was accepted for statistical significance.

To determine the potential disease relevance of GRIK1-AS1 in gastric cancer, the real-time qPCR analysis was used to quantify the expression of GRIK1-AS1 in both gastric cancer tissues and cell lines. As indicated in Figure 1A , GRIK1-AS1 expression was significantly decreased (P<0.005) in tumors compared with adjacent normal tissues ( Figure 1A ). Likewise, the gastric cancer cells (AGS, SGC-7901, HGC-27) showed a lower expression level of GRIK1-AS1 than gastric epithelial cell GES-1 ( Figure 1B ). These results indicate that decreased GRIK1-AS1 expression might be associated with the development of gastric cancer. Next, we aimed to explore the molecular basis of GRIK1-AS1 functioning as a tumor-suppressor in gastric cancer. First, the sub-cellular localization of GRIK1-AS1 in gastric cancer cells (AGS and SGC-7901) was determined by the nuclear/cytoplasm distribution analyses and RNA-FISH assays. Results showed that GRIK1-AS1 was mainly located in the cytoplasm ( Figures 1C, D ).

To investigate the potential functional significance of GRIK1-AS1 in gastric cancer, we ectopically over-expressed GRIK1-AS1 in gastric cancer cells AGS and SGC-7901 ( Figure 2A ). We found that restoration of GRIK1-AS1 significantly attenuated cancer cell proliferation ( Figures 2B, C ). Moreover, the increased GRIK1-AS1 expression significantly suppressed cancer cell invasion and migration abilities ( Figures 2D, E ). These results support the potential tumor suppressor role of lncRNA GRIK1-AS1 in gastric cancer.

Increasing evidence demonstrated that lncRNAs located in the cytoplasm can function as the competitive endogenous RNAs (ceRNAs) for miRNAs, competitively sponging miRNAs to inhibit its regulatory effect on target mRNAs. We then analyzed the miRNA-recognition sequence of GRIK1-AS1 via bioinformatics tools (Star Base, Target Scan, miRcode) and found a list of miRNAs with the complementary sequence to GRIK1-AS1. Using RNA-pull down assays, we screened out the most interacted miRNA. Results showed that GRIK1-AS1 mainly interacted with miR-375 instead of other miRNA candidates ( Figure 3A ). And the luciferase activity assays were performed to further confirm the regulation of GRIK1-AS1 to miR-375 ( Figures 3B, C ). GRIK1-AS1 cDNA containing predicted miR-375 binding site was cloned into luciferase reporter vectors (GRIK1-AS1-WT) and then co-transfected with miR-NC or miR-375. Results showed that the luciferase activity was significantly decreased when GRIK1-AS1-WT was co-transfected with miR-375. Meanwhile, the GRIK1-AS1-Mut luciferase reporter vector was also conducted with a mutation in its miR-375 binding site. Co-transfection with GRIK1-AS1-Mut and miR-375 did not reduce the luciferase activity ( Figure 3C ). RNA-pull down results confirmed that GRIK1-AS interacted with wild-type miR-375 instead of mutant miR-375 ( Figure 3D ).

To further support the direct regulatory relationship between GRIK1-AS1 and miR-375 in gastric cancer cells. Real-time qPCR analyses over-expression of GRIK1-AS1 in AGS and SGC-7901 cells led to a significant down-regulation of miR-375 ( Figure 3E ), which was exactly the opposite from the expression pattern of GRIK1-AS1 ( Figure 1B ). Meanwhile, pre-miR-375 displayed a markedly higher expression level in gastric cancer cells than in gastric epithelial cells ( Figure 3F ). Moreover, miR-375 mimic reversed the suppressed cell proliferation ( Figures 4A–C ) and invasion ( Figure 4D ) capability caused by the over-expression of GRIK1-AS1 in gastric cancer cells. These results, taken together, demonstrated a negatively regulatory relationship between lncRNA GRIK1-AS1 and miR-375.

Having demonstrated the functional significance of lncRNA GRIK1-AS1 in gastric cancer by impacting miR-375 expression, we began to identify the critical target of miR-375 in gastric cancer. Based on target analysis such as Target Scan, we found several target genes whose 3’-UTR contained miR-375 binding sites. miR-NC and miR-375 were separately transfected into HEK293 cells and real-time qPCR was performed to measure the expression level of these candidate targets. IFIT2 was successfully screened out as the direct downstream target gene of miR-375 ( Figure 5A ). Luciferase activity assays indicated that miR-375 significantly reduced the expression of wild-type IFIT2 3’UTR, while did not affect mutant IFIT2 3’UTR expression ( Figures 5B, C ). The results obtained from luciferase activity assays were further strengthened by RNA-pull down ( Figure 5D ). IFIT2 3’-UTR directly interacted with wild-type miR-375 instead of its mutant.

To better establish the relationship between GRIK1-AS1 and miR-375 target gene IFIT2, the IFIT2 expression level was measured in GRIK1-AS1 over-expressed AGS and SGC-7901 cells. As indicated in Figure 5E , IFIT2 expression was markedly up-regulated by GRIK1-AS1 over-expression, while miR-375 over-expression significantly down-regulated IFIT2 ( Figure 5E ). Real-time qPCR ( Figure 5F ) and western blotting ( Figures 5G, H ) analyses indicate a significant (AGS vs GES-1: P<0.005, SGC-7901 vs GES-1: P<0.005, HGC27 vs GES-1: P<0.01) lower expression level of IFIT2 in gastric cancer cells than in gastric epithelial cells. Additionally, IFIT2 knockdown reversed the suppressed cell proliferation and invasion capability caused by the over-expression of GRIK1-AS1 in AGS and SGC-7901 cells ( Figures 6A–E ). Together, these results corroborated that GRIK1-AS1/miR-375/IFIT2 axis is a novel regulatory mechanism in gastric cancer progression.