Material and methods

The Cancer Genome Atlas (TCGA) data

Transcript levels of psiTPTE22-HERV were obtained from TSVdb,11 a Web tool for splicing variant analysis in the TCGA dataset (http://www.tsvdb.com). The specific isoforms used in the analysis were isoform_uc010gqq and isoform_uc002zlr, which correspond to the two open-reading-frame-containing transcript variants of psiTPTE22-HERV. In addition, protein data (reverse-phase protein array) for eukaryotic translation initiation factor 4E (EIF4E) binding protein 1 (EIF4EBP1), EIF4EBP1-pT70, and EIF4E were acquired from the Stomach Adenocarcinoma dataset of TCGA (Firehose Legacy) through the cBioPortal for Cancer Genomics (http://www.cbioportal.org/).12,13

Cancer cell lines

Cancer cell lines (MKN45, SGC7901, and MKN1) were grown in 1640 medium supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin at 37 °C in a humidified 5% CO2 atmosphere. MKN45 and SGC7901 cells were obtained from Beyotime (Shanghai, China), with subsequent identification of potential contamination with HeLa cells in SGC7901. MKN1 cells were obtained from RIKEN BioResource Research Center (Tsukuba, Japan). All experiments were performed using mycoplasma-free cells. cDNAs and/or DNA of additional cancer cell lines and the human embryonic kidney HEK293 cell line were obtained from previous studies3,4 or from stocks at the Cancer Institute of Zhejiang University. A complete list of the cell lines used in this study is provided in Table S1.

Primary tissue samples

Gastric cancer tissues were collected during surgical resection at the Second Affiliated Hospital of Zhejiang University in Hangzhou, China. This study adhered to the principles outlined in the Helsinki Declaration. This study was approved by the Clinical Research Ethics Committee of Zhejiang University (IRB# 2016-133). Written informed consent was obtained from all participants. cDNAs of normal human tissues were purchased from Stratagene (La Jolla, California, USA).

RNA extraction and PCR

Total RNA was extracted from cells and tissues using TRIzol reagent (Invitrogen, Carlsbad, California, USA), according to the manufacturer's instructions. cDNA was synthesized using the One Step PrimeScript cDNA Synthesis Kit (TaKaRa, Dalian, China). Reverse transcription PCR was performed using the GoTaq Colorless Master Mix (Promega, Madison, Wisconsin, USA). Quantitative PCR was performed using the AceQ qPCR Probe Master Mix (Vazyme Biotech, Nanjing, China). The nucleotide sequences of PCR primers and probes are listed in Table S2. Each probe carried a 5′ reporter dye, 6-carboxy fluorescein (FAM), and a 3′ quencher dye, 6-carboxytetramethyl-rhodamine (TAMRA). Primers and hydrolysis probes were synthesized by Invitrogen. Quantitative PCR amplifications were conducted in a 20 μL reaction system containing 0.3 μM of each primer and 0.2 μM of probe. The thermal cycler parameters of the ABI PRISM 7500 Fast Sequence Detection System were 50 °C for 2 min, 95 °C for 10 min, and 40 cycles of 95 °C for 15 s and 60 °C for 1 min. PCR amplification specificity was confirmed by Sanger sequencing of the PCR products. The relative expression level of psiTPTE22-HERV was calculated using the ΔCt method relative to that of GAPDH.

Bisulfite genomic sequencing

Bisulfite modification was performed using the EZ DNA methylation kit (Zymo Research, Orange, California, USA) following the manufacturer's instructions. PCR amplification was performed using 2 μL bisulfite-converted DNA. Direct Sanger sequencing of the PCR products was performed to evaluate the methylation levels. The proportion of methylation was calculated as the peak ratio of cytosine to the sum of cytosine and thymine at each site. The nucleotide sequences of the primers used for bisulfite genomic sequencing are listed in Table S2.

Demethylation with 5-Aza agent treatment

Gastric cancer cells were treated with 2 μM DNA demethylation agent, 5-Aza (Sigma–Aldrich, St Louis, Missouri, USA), for 5 days. The culture medium was refreshed every day.

Establishment of stable psiTPTE22-HERV-expressing cells

The full-length open reading frame of psiTPTE22-HERV was generated by reverse transcription PCR and cloned into the mammalian expression vector pcDNA™3.1/myc-His A (Invitrogen). The construct sequence was confirmed by sequencing. MKN45 and SGC7901 cells were transfected with pcDNA3.1 or pcDNA3.1-psiTPTE22-HERV plasmids using Lipofectamine 3000 (Invitrogen). Stable transfectants were selected with G418 antibiotic (Invitrogen) for two weeks.

Knockdown of psiTPTE22-HERV by RNA interference

Knockdown of psiTPTE22-HERV expression in MKN1 cells was performed using RNA interference with a small interfering RNA (siRNA) specifically targeting psiTPTE22-HERV (Table S2). siRNA was transfected using Lipofectamine 3000, with a scrambled sequence as the small interfering negative control RNA (siNC). Knockdown efficiency was evaluated using reverse transcription or quantitative PCR.

Cell viability and colony formation

Cell viability was assessed using the CCK-8 assay (Boster Bio, Pleasanton, California, USA). For colony formation, cells were seeded at a density of 200 cells per well in 6-well culture plates and incubated at 37 °C for at least 14 days until visible colonies formed. The colonies were fixed with methanol for 15 min and stained with 0.5% crystal violet. Colonies consisting of > 50 cells were counted.

Cell cycle analysis

Cells were processed using a cell cycle kit from MultiSciences Biotech (Hangzhou, China) following the manufacturer's instructions. The cells were then sorted using a FACS Calibur Flow Cytometer (BD Biosciences, Franklin Lakes, New Jersey, USA). Cell cycle distribution was analyzed using the ModFitLT software (BD Biosciences).

Apoptosis

Cell apoptosis was assessed using Annexin V-APC/7-AAD staining (MultiSciences Biotech) according to the manufacturer's instructions, followed by flow cytometry. Additionally, in tumor biopsies from nude mice, apoptosis was measured using the terminal deoxynucleotidyl transferase-mediated dUTP-digoxigenin nick-end labelling (TUNEL) assay (Promega). Nuclei with clear staining were regarded as TUNEL-positive apoptotic cells. The apoptosis index was calculated as the percentage of TUNEL-positive nuclei, with at least 1000 cells counted.

Wound-healing and invasion assays

For wound-healing assays, cells were seeded in 6-well culture plates and allowed to reach 80% confluence after 24 h. Wounds were created using 200-μL plastic pipette tips, followed by washing with phosphate-buffered saline (PBS). The wounded cell monolayers were then allowed to heal for 48 h or 72 h in serum-free RPMI 1640 medium. Invasion assays were conducted using Matrigel™ Invasion Chambers (BD Biosciences). Cells suspended in serum-free medium (1 × 105 cells in 100 μL) were added to the upper chamber, whereas medium containing 10% fetal bovine serum was added to the lower chamber. After incubation at 37 °C for 48 h, the number of cells that migrated to the lower side of the upper chamber was determined.

In vivo subcutaneous tumorigenicity and metastasis assays

All animal experiments were conducted in accordance with institutional guidelines and were approved by the Animal Experimental Committee of Zhejiang University (#2016-239). For subcutaneous xenograft models, MKN45 cells (5 × 106 cells in 200 μL PBS) stably transfected with psiTPTE22-HERV expression vector or empty vector were injected subcutaneously into the backs of 5-week-old male Balb/c nude mice (control: n = 10; psiTPTE22-HERV: n = 11). Tumor diameter was measured every 3 days for 30 days, and the volume (mm3) was calculated as follows: (shortest diameter)2 × (longest diameter) × 0.5. Upon sacrifice, the tumors were excised, weighed, and either fixed in 10% neutral-buffered formalin or snap-frozen in liquid nitrogen. For tail vein injection models, SGC7901 cells stably transfected with psiTPTE22-HERV expression vector or empty vector (2 × 106 cells in 100 μL PBS) were injected into the tail veins of 6-week-old female Balb/c nude mice (control: n = 10; psiTPTE22-HERV: n = 11). Four weeks after the injection, the mice were sacrificed and examined. The lungs and livers were dissected, paraffin-embedded, and sectioned. Hematoxylin-eosin staining was performed, and the metastatic tumors were counted in a blinded manner. No animals exhibited significant discomfort, impaired mobility, or compromised well-being during the study period.

Western blotting and immunohistochemistry

Total cellular proteins were extracted with a RIPA lysis and extraction buffer. Fifty micrograms of protein from each sample were separated on a 10% SDS-PAGE gel and then transferred onto nitrocellulose membranes (GE Healthcare). After incubation with specific primary antibodies at 4 °C overnight and then with secondary antibodies, protein bands were visualized using ECL Plus Western Blotting Detection Reagents (GE Healthcare). Immunohistochemistry was conducted on 5-μm paraffin-embedded xenograft sections. The staining intensity was evaluated as the percentage of cells that displayed positive staining. The antibodies used in this study are listed in Table S3.

Immunofluorescence assay

The cells were fixed with 4% paraformaldehyde in PBS on slides at room temperature for 30 min. The cells were permeabilized using 0.2% Triton X-100 in PBS (pH 7.2) for 4 min and blocked with 5% bovine serum albumin in PBS at room temperature for 30 min. Cells were stained with FITC-conjugated mouse anti-EIF4E antibody (1:100 dilution in PBS containing 1% bovine serum albumin; BD Biosciences) at 4 °C overnight and then with Dylight 488-conjugated goat anti-mouse polyclonal antibody (Abcam) at room temperature for 1 h. After three PBS washes (pH 7.2), the slides were mounted using Vectashield containing DAPI (Vector Laboratories, Burlingame, California, USA) and sealed. Fluorescence was observed using a Zeiss Confocal Laser Scanning Microscope.

RNA sequencing and differential gene expression

Total RNA (3 μg) with RNA integrity number > 6.8 from cells stably transfected with either psiTPTE22-HERV expression vector or empty vector was subjected to RNA sequencing. The mRNA was isolated using Oligo d(T) Magnetic Beads and fragmented for cDNA synthesis. Libraries were prepared using the NEBNext Ultra™ RNA Library Prep Kit (New England Biolabs, Ipswich, Massachusetts, USA) following the manufacturer's instructions and sequenced on an Illumina HiSeq XTEN platform. Raw sequencing reads were quality filtered to remove adapter sequences, poly-N reads (> 10% N content), and low-quality reads (Q-score < 20). All samples yielded > 3.25 Gb clean bases with a < 0.02% error rate and Q20 of > 96.4%. Clean reads were aligned to the reference genome using HISAT2,14 and gene expression levels were quantified as FPKM (Fragments Per Kilobase of transcript per Million mapped reads) and read counts using FeatureCounts v1.5.0-p3.15 Differential expression analysis was performed using the edgeR R package. The resulting P-values were adjusted using the Benjamini and Hochberg method for multiple testing.

Over-representation analysis and gene set enrichment analysis (GSEA)

Enrichment analysis was performed on differentially expressed genes induced by psiTPTE22-HERV overexpression to identify significant gene ontology (GO) biological processes and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Differentially expressed genes were subjected to over-representation analysis and GSEA using the WebGestalt platform (http://www.webgestalt.org/#).16 GO analysis was performed using the Biological Process noRedundant database with default parameters.

Statistical analysis

Data were shown as mean ± standard deviation, mean ± standard error of the mean, or median (interquartile range), as appropriate. All functional data were obtained from at least three independent experiments. The Mann–Whitney U test was performed for comparisons between two sample groups. Percentages were compared between two groups using Fisher's exact test. Differences between experimental groups were assessed using Student's t-test or one-way ANOVA. Linear trends in mRNA expression and promoter methylation were analyzed using one-way ANOVA with multiple comparison tests. The cutoff value was determined by survival significance analysis using Cutoff Finder (http://molpath.charite.de/cutoff/).17 Kaplan–Meier analysis with the log-rank test was used to compare survival distributions between two groups. Hazard ratios were estimated using univariate and multivariate Cox regression analyses to assess the association between predictor variables and risk of death. The difference in tumor growth between the two groups of nude mice was determined using repeated-measures ANOVA. All statistical tests were conducted using SPSS 19.0 software (SPSS Inc., Chicago, Illinois, USA) and GraphPad Prism V9.4 (GraphPad Software, Inc., San Diego, California, USA). P values < 0.05 were considered statistically significant.