2.1 Clinical subjects and specimens

We collected 65 paired GC tissues and adjacent normal tissues from the Department of General Surgery, the First Affiliated Hospital of Nanjing Medical University. No patients were treated with chemotherapy or radiotherapy before surgery. Tissues were immediately frozen after surgical resection and stored in liquid nitrogen. Every subject provided written informed consent.

2.2 RNA extraction and quantitative real-time polymerase chain reaction (qRT-PCR)

TRIzol reagent (Invitrogen, USA) was used to isolate total RNA from GC tissues and cells. A NanoDrop 2000 spectrophotometer (Thermo Scientific, USA) was used to determine the RNA concentration. A PrimeScript RT-PCR kit (Takara, Japan) was used to reverse transcribe the mRNAs, which produced complementary DNAs (cDNAs). The New Poly(A) Tailing 17 Kit (Thermo Scientific, USA) was used for reverse transcription of the miRNAs. QRT-PCR was performed with SYBR Green Master Mix (Roche, USA) on an ABI StepOne Plus system, after which we calculated the relative gene expression levels according to the 2−ΔΔCt method. GAPDH and U6 served as endogenous controls. The sequences of primers used were as follows: DNAH17-AS1 (forward: 5′-GGAGGGTAGAGCACATTCGG-3′, reverse: 5′-GGCACACATGGCTGTATTGG-3′); ONECUT2 (forward: 5′-CCGAACACTCTTCGCCATCT-3′, reverse: 5′-GCTCAGATCGTCTTGCCACT-3′). GAPDH (forward: 5′-CACCCACTCCTCCACCTTTG-3′, reverse: 5′-CCACCACCCTGTTGCTGTAG-3′). miR-202-3p (forward: 5′-AGAGGTATAGGGCATGGGAA-3′). Universal (5′-GCGAGCACAGAATTAATACGAC-3′); U6 (forward: 5′-CTCGCTTCGGCAGCACA-3′, reverse: 5′-AACGCTTCACGAATTTGCGT′).

2.3 Western blot

RIPA lysis buffer (Beyotime, China) was used to extract total protein from GC cells and tissues. The protein concentration was assessed with a BCA protein assay kit. After SDS-PAGE, the proteins were transferred to PVDF membranes. The membranes were subsequently blocked in TBST buffer supplemented with 5% skim milk, followed by incubation with primary antibodies at 4 °C overnight. The next day, the membranes were rinsed three times with TBST and then incubated with the appropriate secondary antibodies for 2 h. An enhanced chemiluminescence (ECL) kit (Servicebio, China) was used to visualize the protein bands after the membranes were washed three times with TBST for 10 min. Finally, the grey values of the proteins were calculated using ImageJ software. The antibodies information was listed in Supplementary Table 1.

2.4 Cell culture and transfection

The five human GC cell lines AGS, HGC-27, MKN-45, MKN-28, and NCI-N87 as well as the normal gastric mucosa epithelial cell line GES-1 were obtained from the Institute of Biochemistry and Cell Biology of the Chinese Academy of Sciences (Shanghai, China). We cultured these cells (with the exception of HGC-27 cells, which were cultured in F12K medium) in RPMI 1640 medium (Wisent, China) supplemented with 10% foetal bovine serum (FBS), streptomycin (100 µg/ml), and penicillin (100 U/ml) in a humidified incubator with 5% CO2 at 37 °C. Small interfering RNAs (siRNAs) targeting DNAH17-AS1 (si-DNAH17-AS1, containing si-DNAH17-AS1#1 and DNAH17-AS1#2) and a negative control (NC), a miR-202-3p mimic, a miR-202-3p inhibitor, and their NCs, as well as siRNAs targeting ONECUT2 and the corresponding NC, were designed and obtained from GenePharma (Shanghai, China). The DNAH17-AS1 sequence was subcloned and inserted into the pcDNA3.1 vector (Invitrogen, USA) (pcDNA3.1-DNAH17-AS1, shortened to DNAH17-AS1) to upregulate DNAH17-AS1 expression, while the empty vector was used as the control. DNAH17-AS1-targeting short hairpin RNA (shRNA, shDNAH17-AS1) and the NC were synthesized by Genechem (Shanghai, China) and cloned and inserted into the pLKO.1 lentiviral vector for in vivo experiments; subsequently, the lentiviruses were injected into MKN-45 cells. Lipofectamine 3000 (Invitrogen, USA) was used for all cell transfection experiments according to the manufacturer’s instructions. We performed qRT-PCR and western blot to determine the transfection efficiency.

2.5 Cell counting kit-8 (CCK-8) assay

We seeded the treated GC cells (density: 3000 cells per well) into a 96-well plate (Corning, USA) and added 100 µl of DMEM containing 10% FBS. After they were cultured for 1, 2, 3, 4, or 5 days, the cells were incubated for 2 h with 10 µl of CCK-8 reagent (Dojindo, Japan). A microplate reader (Bio-Rad, USA) was then used to measure the absorbance of each well at a wavelength of 450 nm.

2.6 Colony formation assay

The transfected GC cells (density: 500 cells/well) were seeded into a six-well plate. After 14 days of culture, the cells were washed three times with phosphate-buffered saline (PBS). Then, the cells were fixed in 4% paraformaldehyde and stained with 0.1% crystal violet (Kaigen, China) for 30 min. After removal of the dye, the number of colonies was counted.

2.7 Flow cytometry assay

For the cell cycle analysis, we collected the treated GC cells and fixed them in 70% ethanol overnight, followed by centrifugation at 1000 r/min. Then, we added 100 µl of RNAse to the cells and incubated them for 30 min at 37 °C. After that, 100 µl of propidium iodide (PI) was added and the cells were incubated for 30 min at 4 °C in the dark. Finally, we used a flow cytometer to analyse the cell cycle. For the apoptosis analysis, we seeded GC cells in a six-well plate at a density of 2 × 105 cells per well. The cells were then harvested after the incubation period and were washed with PBS. After the cells were centrifuged and resuspended in 100 µl of Annexin-binding buffer, 5 µl of Annexin V-fluorescein isothiocyanate (FITC) and 5 µl of PI were added to the sample tube and the cells were incubated for 15 min in the dark. Next, the cells were centrifuged and resuspended in 400 µl of annexin-binding buffer. Finally, flow cytometry was conducted to analyse the percentage of apoptotic cells.

2.8 Fluorescence in situ hybridization (FISH)

RiboBio Biotech (Guangzhou, China) was commissioned to design and synthesize the specific DNAH17-AS1 cDNA probes. Briefly, we seeded the cells into 24-well plates at a density of 6 × 104 cells per well, after which the cells were fixed in 4% paraformaldehyde once a confluence of 80% was reached. After permeabilization, 250 µl of DNAH17-AS1 hybridization solution containing the probe (0.5 µg/ml) was added to the cells, which were incubated in a humidified chamber overnight. The cells were stained in diluted 4,6-diamidino-2-phenylindole for 5 min. We selected signals and acquired all images with a Leica SP5 confocal microscope (Leica Microsystems, Germany).

2.9 Nuclear and cytoplasmic RNA isolation

In accordance with the manufacturer’s instructions, we used a Cytoplasmic and Nuclear RNA Purification kit (Norgenbiotek Corporation, Canada) to separate the cytoplasmic and nuclear RNA. Briefly, after the cells were lysed in cell fractionation buffer, they were then centrifuged at 500×g for 5 min. We collected the supernatant as the cytoplasmic fraction. Next, cell disruption buffer was used to lyse the remaining nuclear pellets, and the nuclear fraction was obtained. Finally, qRT-PCR was conducted to determine DNAH17-AS1 expression in both fractions. U6 and GAPDH were selected as the nuclear and cytoplasmic controls, respectively.

2.10 Luciferase reporter assay

A Mut Express II Fast Mutagenesis Kit (Vazyme, China) was used to generate mutations in the miRNA binding sites in DNAH17-AS1 and ONECUT2. The 3′-UTR wild-type and mutant sequences of DNAH17-AS1 or ONECUT2 were inserted into the pGL-3 luciferase reporter vector (Realgene, China). After the cells were cultured in 48-well plates overnight, the miR-202-3p mimic and the above reporter plasmids were cotransfected into HEK293T cells. The Dual-Luciferase Reporter Assay System (Promega, USA) was used to detect Firefly and Renilla luciferase activities.

2.11 RNA immunoprecipitation (RIP) assay

A Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Germany) was used to perform the RIP assay, detecting the interaction of DNAH17-AS1 and miR-202-3p. Briefly, anti-Ago2 antibody, the control anti-IgG antibody, and magnetic beads were coupled. Then the corresponding cell lysate was incubated with the beads at 4 °C all night. Following digestion by proteinase K, the coprecipitated RNA was extracted and measured by qRT-PCR.

2.12 Animal models and immunohistochemistry analysis

For the tumorigenesis assays, after they were suspended in 100 µl of PBS, 2 × 106 stably transfected MKN-45 cells were subcutaneously injected into the flank region of 4-week-old female nude mice. We measured the short diameter (a) and the long diameter (b) of the tumours every 3 days and calculated the volume according to the following formula: volume = 0.5 × a2 × b. Fifteen days later, the mice were sacrificed, and the subcutaneous tumours were excised, photographed, and weighed. Xenografts from nude mice were fixed in a 4% formaldehyde solution and then embedded in paraffin. The sections were incubated with anti-ki67 antibody overnight at four degrees Celsius, followed by incubation with HRP-conjugated secondary antibodies for 30 min. After they were stained with 3,3′-diaminobenzidine solution and counterstained with haematoxylin, the slides were examined under a microscope. The Ethics Committee of Nanjing Medical University requires that the maximum tumour volume should not exceed 2000 mm3. And the maximal tumour size from nude mice in our study was not exceeded.

2.13 Clonogenic survival assay

GC cells were seeded into 6-well plates at densities of 300, 500, 1000, 2000, and 4000 cells per well. After 24 h, the cells were subjected to 0, 2, 4, 6, or 8 Gy X-ray irradiation via a medical linear accelerator (Precise Accelerator, Elekta, Sweden). After incubation for 2 weeks, the cells were fixed in methanol and stained with 1% crystal violet. We counted the colonies that contained more than 50 cells. The plating efficiency (PE) was equal to the number of colonies observed/number of cells plated in the control group. The following formula was used to calculate the surviving fraction (SF): number of colonies formed/(number of cells seeded × PE). The survival curves were fitted by the single-hit multiple-target model: SF = 1 − (1 − EXP(− D/D0))N. Dq = D0 * ln(N). The mean lethal dose (D0) represents the dose required to reduce cell survival to 37%. The quasithreshold dose (Dq) represents the ability to accumulate sublethal damage. A higher D0 or Dq indicates enhanced radioresistance.

2.14 Bioinformatics analysis and statistical analysis

The RNA sequencing data of DNAH17-AS1 and miR-202-3p in GC and normal tissues were downloaded from UCSC (https://xenabrowser.net/datapages/; dataset ID: TCGA-STAD.htseq_fpkm-uq.tsv and TCGA-STAD.mirna.tsv) and the GEO database (https://www.ncbi.nlm.nih.gov/geo/). The expression of ONECUT2 was evaluated via GEPIA online software (http://gepia.cancer-pku.cn/) and the GEO database. Survival data related to DNAH17-AS1 and ONECUT2 for gastric cancer patients were extracted from GEPIA. Sangerbox (http://www.sangerbox.com/login.html) was used to download the RNA sequencing data of lncRNAs in GC and normal tissues from the TCGA and to conduct hierarchical clustering analysis. MiRDB (https://mirdb.org/custom.html) and lncRNASNP2 (http://bioinfo.life.hust.edu.cn/lncRNASNP/#!/) were used to predict potential lncRNA‒miRNA interactions. TargetScan (http://www.targetscan.org/vert_72/), miRDB,

miRWalk (http://mirwalk.umm.uni-heidelberg.de/), miRTarBase (https://mirtarbase.cuhk.edu.cn/~miRTarBase/miRTarBase_2022/php/search.php), and DIANA (https://dianalab.e-ce.uth.gr/html/dianauniverse/index.php?r=microT_CDS) were employed to search for mRNAs that bind to miRNAs. The cytoscape software (version 3.7.1) was used to construct and visualize the lncRNA-miRNA-mRNA regulatory network. SPSS 25.0 was used to conduct all the statistical analyses. We generated the relevant diagrams with GraphPad Prism 6. The results are shown as the mean value ± standard deviation (SD) or the mean value ± standard error (SE). Student’s t test was used to assess significant differences between two independent groups. For CCK-8 assays, considering that the ability of cell proliferation was affected by two factors, namely gene expression levels and time, so the two-away ANOVA was used to compare the results of different treatment groups. The associations of DNAH17-AS1 expression with the clinical features of GC patients were evaluated via Pearson’s χ2 test. The correlations between gene expression levels were determined via linear regression. P < 0.05 (*), P < 0.01 (**), and P < 0.001 (***) were considered statistically significant.