Human lung cancer cells H1299 were obtained from the Chinese Academy of Sciences and were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Corning Inc., Corning, NY, United States). Human monocytic leukemia cells, THP-1 (iCell Bioscience Inc., Shanghai, China), were cultured in Roswell Park Memorial Institute (RPMI 1640 medium (Corning). Cell culture was performed with 10% fetal bovine serum (Corning) and 1% penicillin streptomycin (Corning). Cells were cultured at 37°C, 5% CO₂, and 95% humidity in an incubator (Thermo Fisher Scientific, Waltham, MA, United States).

The cells were used when they reached ∼70% confluence. RNA was isolated using a SteadyPure RNA extraction kit, as per the manufacturer’s instructions (Accurate Biology, Changsha, China). The RNA was reverse transcribed to cDNA using ABScript Ⅲ RT Master Mix with RNA (1 μg) from each sample (ABclonal, Wuhan, China). The specific cDNAs were quantified using quantitative real-time PCR (qPCR) together with EvaGreen 2× qPCR MasterMix (Applied Biological Materials, Richmond, BC, Canada) and measuring the fluorescence intensity. Each reaction system was composed of 2 μL of cDNA, 5 μL of EvaGreen 2× qPCR MasterMix (Applied Biological Materials), 2.4 μL of DEPC (diethypyrocarbonate)-treated water (Sangon Biotech Co., Ltd. Shanghai, China), and 0.3 μL of 10 μM forward and reverse primer mix. Samples were amplified under the following cycling conditions: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 60 s. Cycle thresholds were calculated as the expression fold changes using the formula: 2^-△△CT(Livak and Schmittgen, 2001). Data were analyzed using the Applied Biosystems 7500Real-Time PCR System (Applied Biosystems, Foster City, CA, United States). The Applied Biosystems QuantStudio 7 PRO Real-Time Fluorescence Quantitative PCR System runs were acquired and the results were analyzed using EXCEL (Microsoft Corp., Redmond, WA, United States). The primers used for qPCR, including those for the control GAPDH gene (encoding glyceraldehyde-3-phosphate dehydrogenase), were supplied by Sangon Biotech Co., Ltd. and are shown in the additional file (Supplementary Table S1). For the RNA-seq experiment, total RNA was extracted from H1299^vector and H1299^shEIF4G1 (H1299 cells transfected with the empty vector or the vector expressing an shRNA targeting EIF4G1), followed by Novogene sequencing and data analysis (Novogene, Beijing, China).Gene Set Enrichment Analysis (GSEA) is a computational approach to determine if a pre-defined Gene Set can show a significant consistent difference between two biological states. The genes were ranked according to the degree of differential expression in the two samples, and then the predefined Gene Set were tested to see if they were enriched at the top or bottom of the list. Gene set enrichment analysis can include subtle expression changes. We use the local version of the GSEA analysis tool http://www.broadinstitute.org/gsea/index.jsp, KEGG and DisGeNET data sets were used for GSEA independently.

H1299 cells (1.6× 10⁶ cells per mL) were added to each well of a six-well plate. The medium was changed after 8 h. The supernatant was collected after 24 h and tested using an ELISA kit provided by ABclonal (Wuhan, China), following the manufacturer’s instructions.

Cells were used when they reached ∼70% confluence. Cells were washed twice with Dulbecco’s phosphate-buffered saline (DPBS; Corning), and then incubated for 30 min on ice in Radioimmunoprecipitation assay (RIPA) lysis buffer (Epizyme, Cambridge, MA, United States) with the addition of one percent protease phosphatase inhibitor (NCM Biotech, Newport, RI, United States). The cells were centrifuged, and the supernatant was collected and added to SDS loading buffer (Beyotime, Jiangsu, China), followed by boiling at 100°C for 10 min. After cooling, the samples were subjected to 7.5% SDS-polyacrylamide gel electrophoresis (Epizyme), followed by the electrophoretic transfer of the separated proteins to a polyvinylidene fluoride (PVDF) membrane (Millipore, Billerica, MA, United States). After blocking, the membranes were incubated with the following primary antibodies: Anti-EIF4G1, and anti-β-Actin (Cell Signaling Technologies (CST), Danvers, MA, United States; 1:1,000); anti-mitogen-activated protein kinase 14 (MAPK14 or p38), anti-phospho-p38, anti-rabbit IκBα, anti-phospho-IκBα, anti-JNK1/2/3, anti-phospho-JNK1/2/3, anti-ERK1/ERK2, anti-phospho-ERK1/ERK2, anti-TNFRSF10A, and anti-TNFRSF11B (ABclonal), followed by incubation with secondary antibodies comprising horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG secondary antibody (Epizyme). The immunoreactive protein bands were visualized using a chemiluminescent solution (Thermo Fisher Scientific), and photographs were acquired using a fully-automated chemiluminescent imaging analysis system (Tanon Science & Technology CO., Shanghai, China).

Macrophage migration assays were performed using Transwell chambers (Labselect, Hefei city, China) with an 8-μm pore size. Human monocytic leukemia cells, THP-1, were induced to differentiate into macrophages by treatment with 100 ng/mL phorbol 12-myristate-13-acetate (PMA; Sigma, St. Louis, MO, United States) in complete medium for 48 h. Then, 2 × 10⁶ cells per mL were added to the upper layer of the chamber. The H1299^vector group migrated for 24 h after the addition of H1299^vector conditioned medium (CM) to the lower chamber and the addition of different concentrations of human recombinant CXCL8 (ABclonal). Macrophages in the upper chamber of the H1299^shEIF4G1 group migrated for 24 h after incubation with different concentrations of the CXCR1/2 inhibitor Navarixin (MK7123, #HY10198, MedChemExpress, Monmouth Junction, NJ, United States) for 24 h after 30 min of incubation at 37°C.

Cells were used for the assay when they reached ∼70% confluence. Cells were fixed using 4% paraformaldehyde and added with 0.25% TritonX-100, followed by 10% goat serum, and incubated at room temperature. After the goat serum was aspirated off, fresh goat serum containing anti-TNFRSF10A or anti-TNFRSF11B primary antibodies (ABclonal) was added to the cells, and incubated at room temperature for 2 h, followed by incubation with fluorescently-labeled secondary antibodies (Thermo Fisher Scientific) at room temperature and protected from light for 1 h. Finally, the cell nuclei were stained using 0.1% 4′,6-diamidino-2-phenylindole (DAPI) (Beyotime) for 10 min at room temperature and protected from light. Images were acquired under a microscope (Leica, Wetzlar, Germany).

To construct EIF4G1 stable knockdown cell lines, cells were transfected with a Dharmacon (Lafayette, CO, United States) EIF4G1-shRNA, or a non-silencing (NS)-shRNA as a control.

Cells were grown in 96-well plates with 10,000 cells per well. Different concentrations of SBI-756 (a water soluble inhibitor of EIF4G1; MedChemExpress) and TNFSF10 were added to the cells, which were incubated at 37°C, 5% CO₂ and 95% humidity in an incubator for 24 h. Thereafter, 10 μL of Cell Counting Kit 8 (CCK-8) reagent (Share-bio, Hangzhou, China) was added and incubated for 2 h, after which the optical density (OD) values were measured at 450 nm in a microplate reader.

NSCLC tissues and their paracancerous control tissues were collected from 22 patients attending Shanghai Eastern Hospital. All experiments were approved by the Human Rights Committee of Shanghai Eastern Hospital.

The experimental protocol was established according to the ethical guidelines of the Helsinki Declaration and was approved by the Ethics Committee of the Shanghai East Hospital (No. 2019052). Written informed consent was obtained from individuals or their guardians.

Concentrations of multiple cytokines in tissues were assayed after extraction of tissue proteins using a Bio-plex Prot Human Cytokine Grp I Panel 27-Plex kit (Bio-Rad, Hercules, CA, United States) following the supplier’s guidelines.

The count data in this paper were described using examples and percentages, and the measured data were described using the mean ± standard deviation (mean ± SD). The statistical methods mainly used the χ² test, t-test, and analysis of variance (ANOVA). p < 0.05 indicates that the difference is statistically significant, and p > 0.05 indicates that the difference is not statistically significant. Statistical analysis was performed using Stata 16 (Stata Corp., College Station, TX, United States) and GraphPad Prism 7 (GraphPad Inc., La Jolla, CA, United States). GraphPad Prism 7 and Adobe Photoshop 2022 (Adobe, San Jose, CA, United States) were used to plot the graphs.

EIF4G1 is highly expressed in various types of tumors and affects the prognosis of patients with laryngeal squamous cell carcinoma (LSCC). It also regulates LSCC cell cycle protein D levels and cell proliferation through the protein kinase B (AKT)/mechanistic target of rapamycin kinase (mTOR) pathway (Jaiswal et al., 2019). To further explore how EIF4G1 functions in tumors and affects those downstream molecules, we performed RNA-seq analysis of H1299^vector and H1299^shEIF4G1 cells. Quantitative analysis of the RNA-seq data identified 1,836 differentially expressed genes and 13,273 co-expressed genes between the samples from the H1299vector and H1299shEIF4G1 groups (Figure 1A).According to DisGeNET analysis, knockdown of EIF4G1 had the greatest impact on inflammation (Figure 1B), and the KEGG results showed that the expression levels of a variety of inflammation-related cytokines was significantly upregulated in EIF4G1 knockdown cells (Figure 1C). In this study, some common inflammation-related cytokines were selected to test their relative expression using qRT-PCR, and the mRNA levels of 84% of the inflammation-related factors were found to be higher in H1299^shEIF4G1 cells relative to their levels in H1299^vector cells (Supplementary Figure S1A).

Clinical samples identified multiple inflammation-related cytokines that were upregulated in NSCLC tissues compared with those in the corresponding paraneoplastic tissues.

Numerous studies have shown that inflammation is closely related to tumors, and that tumor cells regulate the inflammatory state by secreting various inflammatory mediators; however, inflammation is also involved in regulating cancer development, progression, and response to cancer therapy (Lee et al., 2021). To explore which inflammation-related cytokines are highly expressed in NSCLC, we used Bio-plex to examine 25 inflammation-related factors in cancer tissues from 22 patients with NSCLC and their control tissues. The results revealed that 18 inflammation-related cytokines were highly expressed in cancer tissues (Figure 2A). We investigated the relationship between NSCLC and the levels of various inflammation cytokines and found that interleukin (IL)1B (113.91 ± 21.59 vs. 21.87 ± 3.48), CXCL8 (1985.23 ± 438.51 vs. 430.08 ± 27.02), CXCL10 (2,819.79 ± 654.58 vs. 1,016.33 ± 70.82), and VEGF (981.78 ± 149.74 vs. 328.88 ± 42.79) were significantly elevated, and CCL15 (3,420.97 ± 317.86 vs. 4,672.99 ± 587.21) and IL15 (43.07 ± 2.71 vs. 59.73 ± 11.88) were significantly downregulated (Figure 2B). The results of transcriptome sequencing showed that low expression of EIF4G1 was associated with inflammation. Analysis of the expression of inflammation-related cytokines in lung cancer tissue samples from patients with NSCLC and their paracancerous control tissues, showed that a large number of inflammation-related cytokines were highly expressed in NSCLC tissues, including CXCL8, whose results were in line with those of transcriptome sequencing. Figure 2C shows the expression of CXCL8 in cancer tissues and its paracancerous tissues in each NSCLC patient. The correlation between CXCL8 and patient age, sex, tumor-node-metastasis (TNM) stage, and tumor grade was also analyzed; however, no statistical significance was found (Supplementary Table S2). Analysis using qRT-PCR and ELISA verified that knockdown of EIF4G1 led to significantly high expression of CXCL8 (Figures 2D, E).

A previous study showed that TNFSF10 stimulation of pancreatic ductal adenocarcinoma (PDAC) cells induced the secretion of CXCL8, and the expression of CXCL8 correlated with the concentration of TNFSF10 added to the cells (Legler et al., 2018). The results of transcriptome sequencing (Figure 1C) showed that TNFRSF10A and TNFRSF11B, encoding the specific receptors of TNFSF10, were significantly highly expressed at the mRNA level in H1299^shEIF4G1 cells. This was verified using qRT-PCR, and the amplification curves showed that the H1299^vector cells hardly expressed TNFRSF10A and TNFRSF11B (Supplementary Figures S1B–D). Interestingly, when examining the expression of both proteins using Western blotting, we observed that the TNFRSF10A level was upregulated when EIF4G1 was knocked down, whereas TNFRSF11B expression did not change significantly (Figures 3A, B). This was also confirmed by immunofluorescence analysis (Figures 3C–E).

H1299^vector and H1299^shEIF4G1 cells were incubated for 24 h in culture medium with the addition of recombinant TNFSF10 at concentrations of 0, 0.01, 0.1, 1, 10, and 20 nM. The concentration of CXCL8 in the supernatant of the cell culture media was detected using ELISA. It was clear that, with the addition of increasing amounts ofTNFSF10, the H1299^shEIF4G1 cells showed the highest level of CXCL8 secretion at a TNFSF10 concentration of 1 nM, while H1299^vector cells showed the highest level of CXCL8 secretion at a TNFSF10 concentration of 10 nM (Figure 4A). SBI-756 is an inhibitor of EIF4G1, capable of disrupting the EIF4F complex. The concentration of CXCL8 in the supernatant of the cell culture medium gradually increased with the addition of increasing concentrations of SBI-756 in H1299 cells (Figure 4B).TNFSF10 has the ability to promote necrotic cell death in tumor cells; therefore, we compared the effects of adding 0, 1, and 2 μM SBI-756 to H1299 cells while adding 0, 0.01, 0.1, 1, 10, and 20 nM of recombinant TNFSF10 and tested cell survival. We observed that EIF4G1 inhibition or knockdown increased the sensitivity of the cells to TNFSF10 (Figure 4C). Functions of the TNFR superfamily show that their excitation activates related downstream inflammatory pathways and mediates tumor apoptosis (Ashkenazi, 2008). Therefore, we analyzed the phosphorylation levels (activation) of JNK1/2/3, p38, and ERK1/2 of the MAPK pathway and IκBα of the NFκB pathway in H1299^vector, H1299^shEIF4G1, and H1299^{vector SBI−756} cells after incubation for 24 h with 1 nM recombinant TNFSF10, respectively. The levels of phospho-p38, phospho-ERK1/2, and phospho-IκBα increased upon the addition of TNFSF10, and knockdown of EIF4G1 resulted in a significant increase in ERK1/2 phosphorylation, whereas the addition of SBI-756 to the cells caused upregulation of IκBα phosphorylation, with the end result being an increase in the level of CXCL8 (Figure 4D).

Macrophages are usually the most abundant immune population present in the TME and are a major component of the immune infiltration in solid tumors. As a chemokine, CXCL8 has the ability to induce targeted chemotaxis of specific immune cells (Mantovani et al., 2002).To verify that EIF4G1 ultimately regulates CXCL8 secretion by affecting TNFRSF10A expression, we tested the ability of H1299^vector and H1299^shEIF4G1 cells to promote CXCL8-induced macrophage chemotaxis using Transwell assays. We prepared H1299^vector and H1299^shEIF4G1CM by culturing the cells in complete medium for 24 h. We added recombinant CXCL8 n at 0, 0.1, 0.5, 1, 2, and 3 μM to the CM of H1299^vector. To the upper macrophage layer of the CM of H1299^shEIF4G1, we added MK7123 at concentrations of 0, 10, 50, 100, 200, and 300 nM. MK7123 is a small molecule inhibitor of CXCL8 receptors CXCR1/CXCR2 on macrophages. We observed a gradual increase in the number of macrophages that underwent chemotaxis with the addition of recombinant CXCL8 at elevated concentrations in the CM of H1299^vector, and a gradual decrease in the number of macrophages that underwent chemotaxis with the addition of MK7123 at elevated concentrations in the upper macrophage layer of the CM of H1299^shEIF4G1. This suggests that EIF4G1 knockdown in the H1299 cell line affects the chemotaxis of macrophages towards tumor cells by regulating CXCL8 (Figures 5A, B).