The mRNA expression data and methylation data were obtained from the Gene Expression Omnibus (GEO) database with the login numbers GSE33335 and GSE30601.
Tissue sample collection and follow-up
The tissues used for immunohistochemistry (IHC) were composed of 83 pairs of tumor and adjacent non-tumor tissues that were collected from patients with GC who underwent curative gastrectomy at Tianjin Medical University Cancer Hospital (Tianjin, China) between January 2004 and August 2007. All patients underwent standard follow-up after curative surgery. The median follow-up time was 22 months (range 3–70 months). The tissue used for the MassARRAY methylation test consisted of 97 cancerous tissues and 6 paracancer tissues from patients undergoing radical gastrectomy between January 2003 and July 2007 at Tianjin Medical University Cancer Hospital (Tianjin, China), of which 97 cancerous tissues were closely observed. The median follow-up time was 17 months (range 2–85 months). The clinical data included the patient’s age, sex, endoscopy result, chest X-ray, B-mode ultrasound and surgical methods. Between August and November 2019, we collected 30 pairs of GC and adjacent non-tumor tissues for RNA extraction. The study was conducted with patient consent. The experimental study protocol and the use of clinical data were approved by the Ethics Committee of the Cancer Institute of Tianjin Medical University and the Cancer Hospital of Tianjin Medical University (Tianjin).
Cell culture and transfection
The cell lines of MKN45, NCI-N87, BGC-823, MGC-803, SGC-7901, SNU-1, KATO-III and GES-1 were cultured in RPMI 1640 medium (Gibco, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS, HyClone) and 1% antibiotics (penicillin/streptomycin) (Gibco). AGS cells were cultured in F12k (Gibco) with 10% FBS and 1% antibiotics. The HEK293T cell line was cultured in Dulbecco’s modified Eagle medium (DMEM) (Gibco) with 10% FBS and 1% antibiotics. Cells were maintained at 37 °C with a 5% CO2 atmosphere. The expression levels of ZSCAN18 and TP53INP2 were enhanced by plasmids (plVX-IRES-Puro-ZSCAN18) and (plVX-IRES-Puro-TP53INP2), respectively, and an empty vector was transfected into the cells (pLVX-IRES-Puro-vector) as control. The shRNA against TP53INP2 interference sequence was synthesized (shTP53INP2, 5′-TGGACGGCTGGCTCATC-3). Western blot and quantitative real-time polymerase chain reaction (qRT-PCR) were used to verify the transfection efficiency.
Further 83 paired GC and matched adjacent non-tumor tissues were retrieved from Tianjin Medical University Cancer Hospital (Tianjin, China), Xijing Hospital of Air Force Medical University (Xi’an, China) and Renji Hospital of Shanghai Jiao Tong University School of Medicine (Shanghai, China) between August 2004 and December 2007, and sent to Shanghai Outdo Biotech Company (Shanghai, China) for tissue microarray analysis to evaluate the expression of ZSCAN18. The antibody concentrations of ZSCAN18 and TP53INP2 were 1:150 and 1:300, respectively. The staining intensity of ZSCAN18 and TP53INP2 was estimated by three independent pathologists without knowledge of the clinical data. The staining intensity was divided into 0 (negative), 1 (weak), 2 (medium), and 3 (strong) ranks. Zero and 1 were considered negative; 2 and 3, positive.
The western blot experiment was conducted in accordance with standard procedures. The primary antibodies used for this research were as follows: anti-ZSCAN18 (TA505326, OriGene), anti-VINCULIN (ab219649, Abcam), anti-p62/SQSTM1 (#16177S, CST), anti-LC3A/B (#12741S, CST) and anti-TP53INP2 (ab273012, Abcam).
Cell viability assay
Cell viability was detected using a Cell Counting Kit 8 (CCK-8) reagent. After the cell count, cells were placed in a 96-well culture plate; the AGS cell line had 1 × 103 cells per well, and the NCI-N87 cell line had 3 × 103 cells per well. Testing started 2 h after the addition of 10 µl of the CCK-8 reagent. AGS was tested once a day, and NCI-N87 was tested once every other day. Another way to assess cell viability was to measure adenosine triphosphate (ATP) levels using the Cell Titer-Glo assay (Promega, Cat#G7570). Chloroquine (CQ; HY-17589A) and rapamycin (RAP; HY-10219) were purchased from MedChemExpress (MCE, USA) and were treated at a concentration of 10 μM (CQ) and 100 nM (RAP) for 24 h, respectively. The ZSCAN18-overexpressed AGS, ZSCAN18-overexpressed NCI-N87 and control cells were cultured in 96-well plates with or without CQ/RAP.
Colony formation assay
The colony-forming method was used to detect the proliferation of GC cells. We conducted experiments on AGS and NCI-N87 cell lines using 1000 AGS cells per well and 8000 NCI-N87 cells per well. After the cell count, the cells were inoculated into 6-well plates and incubated at 37 °C for 12 to 14 days. Cell fluid was changed periodically until a visible clone formed.
Ten 4-week-old female BALB/C nude mice (Vital River) were purchased to construct a subcutaneous tumor xenograft. A total of 3 × 106 cells were injected into the lateral dorsal subcutaneous flank of nude mice. Xenograft tumor formation was measured every 4 days. Three weeks after injection, the nude mice were euthanized by cervical dislocation killing after intraperitoneal injection of 2% pentobarbital sodium (0.5 mL), and the tumor volume (V) was measured by the formula: V = 1/2 × length × (width)2. Tumor tissues were excised for immunohistochemistry.
AGS and NCI-N87 cells were laid 12 h in advance, and the cell density was approximately 30% on the next day. Cells were incubated with a concentration of 2 μM of 5-aza-2′-deoxycytidine (DAC, Sigma, St. Louis, MO). The liquid was changed every 24 h. The treatment lasted 96 h.
RNA sequencing and analysis
NCI-N87 cells were stably transfected with PLVX-IRES-Puro-ZSCAN18 or empty vector (PLVX-IRES-Puro). RNA from total samples was isolated and purified using TRIzol (Invitrogen, CA, USA). Then, NanoDrop ND-1000 (NanoDrop, Wilmington, DE, USA) was used to control the quantity and purity of total RNA. The captured mRNA was fragmented at 94 °C for 5 to 7 min using a magnesium ion fragmentation kit (NEBNext® RNA Fragmentation Module, article no. E6150S, USA). cDNA was synthesized from the segmented RNA using Invitrogen SuperScript™ II Reverse Transcriptase (article no. 1896649, CA, USA). Illumina NovaSeq™ 6000 (LC Bio Technology Co., Ltd., Hangzhou, China) was used for double-terminal sequencing according to the standard operation, and the sequencing mode was PE150. Cutadapt (https://cutadapt.readthedocs.io/en/stable, version for Cutadapt 1.9) was used to remove the joint plane raw data processing software. A significant difference was analyzed between samples, and the multiple of the difference (fold change [FC] > 2 or [FC] < 0.5 with a p value < 0.05) defined differentially expressed genes (DEGs), which were analyzed for Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment using R packs.
MassARRAY analysis of the methylation level of ZSCAN18
Instructions to extract cells or tissues using the DNA extraction kit (BioTeKe Corporation) were observed. The DNA samples to be tested were processed using a commercial NaHSO3 kit (ZYMO). The gene fragments to be detected were enriched and amplified by PCR reaction, and the product length was 200 to 700 bp. The PCR product was treated with shrimp alkaline phosphatase to remove free deoxyribonucleoside-5′-triphosphates (dNTPs) in the system. A transcriptase digestion reaction was followed by resin purification. The purified products of the resin were transported to a 384-well SpectroCHIP® bioarray (Agena, Inc.) using an Agena NanodispenserRS1000 dot sampling instrument (Agena, Inc.). EpiTYPER™ software provided advanced and convenient quantitative analysis of DNA methylation.
Binding motif prediction
The chromatin immunoprecipitation next-generation sequencing (ChIP-seq) information of ZSCAN18 was acquired from the Cistrome Data Browser database (http://cistrome.org/db/#) and then visualized using the UCSC browser. The predicted binding sequences at −100 bp to +2000 bp from the reference sequence (RefSeq) transcription start site were extracted from the U.S. National Library of Medicine (https://www.ncbi.nlm.nih.gov) and were then inputted into the ChIP-seq database of ZSCAN18. The DNA sequences were extracted from an obviously higher peak, and primers were designed in sections for the chromatin immunoprecipitation (ChIP) experiment.
ChIP experiments followed previous described protocol
The ChIP assay was performed using the ChIP-IT® Express Enzymatic Magnetic Chromatin Immunoprecipitation Kit & Enzymatic Shearing Kit (Cat No. 53009 & 53035, Active Motif, USA) according to the manufacturer's instructions. Briefly, stable ZSCAN18-expressing AGS or NCI-N87 cells and vector-expressing cells were cross-linked with formaldehyde, which were collected in centrifuge tubes for protein extraction. The chromatin was sheared to 200–1000 bp by enzymatic shearing from extracted cross-linked nuclei. The sheared chromatin was incubated with an antibody of ZSCAN18 (UM500081, OriGene); the antibody-bound DNA complexes were precipitated through the use of magnetic Protein G-coupled beads. The same amount of non-specific immunoglobulin G (IgG) is used as control. The captured chromatin was then eluted, the cross-links were reversed, and the recovered DNA was analyzed by qRT-PCR using the primer of TP53INP2 (5′-GGTGGGAAAGCAGAGTGTGT-3).
The GC cells were transfected with recombinant mRFP‐GFP‐LC3 adenovirus. In green- and red‐merged pictures, autophagosomes were shown as yellow puncta, while autolysosomes were shown as red puncta. Otherwise, autophagic flux was detected using fluorescence microscope (Zeiss).
IBM SPSS Statistics (version 24.0; Armonk, NY, USA) and GraphPad Prism 8 software (GraphPad Software Inc., La Jolla, CA, USA) were used to analyze the data for this study. The t test was used to compare statistical differences between the two groups. The chi-squared test and Fisher’s exact test were used to analyze the relationship between methylation and various clinicopathological parameters. Overall survival was determined by log-rank test and the Kaplan–Meier method. The Spearman correlation test was used to analyze the correlation between gene expression and the methylation level. A receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of methylation sites in cancerous and adjacent tissues. In univariate analysis, survival differences were estimated using the Kaplan–Meier method (log-rank test), and independent prognostic factors were subsequently identified in Cox proportional risk regression models for multivariate analysis. p < 0.05 was considered statistically significant.