Helicobacter pylori promotes gastric cancer progression by upregulating EPS8L3 to inhibit ferroptosis

Abstract

BACKGROUND

Gastric cancer (GC) represents one of the most prevalent malignancies globally and is ranked as the fourth leading cause of cancer-related mortality. Helicobacter pylori (H. pylori) infection is a major risk factor for GC, however, the molecular mechanisms by which H. pylori promotes gastric carcinogenesis remain unclear.

AIM

To investigate how H. pylori promotes GC development by activating specific signaling pathways.

METHODS

Transcriptome sequencing identified epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) as highly expressed in GC cells co-cultured with H. pylori, which was further validated experimentally. Lentiviral knockdown or overexpression was used to assess its effects on GC cell proliferation, invasion, and migration. Ferroptosis was analyzed by flow cytometry. Nuclear factor kappa-B (NF-κB) was predicted as the upstream transcription factor using bioinformatics, and a nude mouse xenograft model evaluated the role of EPS8L3 in vivo.

RESULTS

H. pylori infection increased EPS8L3 expression in gastric epithelial and GC cells via activating NF-κB. Overexpression EPS8L3 enhanced GC cell proliferation, invasion, migration, and tumorigenicity, whereas EPS8L3 knockdown exerted the opposite effects. Mechanistically, EPS8L3 inhibited ferroptosis in GC cells by regulating glutathione metabolism and maintaining the stability of glutathione peroxidase 4 and solute carrier family 7 member 11. Inhibition of NF-κB reduced EPS8L3 expression and attenuated the malignant phenotype of GC cells.

CONCLUSION

H. pylori promotes gastric carcinogenesis and progression via activating the NF-κB/EPS8L3 pathway, which suppresses ferroptosis in GC cells. EPS8L3 may represent a potential therapeutic target for GC.

Key Words: Gastric cancer; Ferroptosis; Helicobacter pylori; Digestion; Nuclear factor kappa B

Core Tip: In this study, Helicobacter pylori co-cultured with the stomach cancer cell line AGS was subjected to transcriptome sequencing. The findings showed that Helicobacter pylori increase the transcription of epidermal growth factor receptor pathway substrate 8-like protein 3 via activating the nuclear factor kappa B signaling pathway. By participating in glutathione metabolism and interacting with solute carrier family 7 member 11 and glutathione peroxidase 4, epidermal growth factor receptor pathway substrate 8-like protein 3 plays a crucial function in preventing ferroptosis in gastric cancer. This process promotes the growth and spread of stomach cancer.

INTRODUCTION

Worldwide, gastric cancer (GC) continues to be among the most common malignancies and represents a major cause of cancer-associated mortality on a global scale[1]. Nearly 50% of GC cases occur in China, posing a significant public health burden[2]. Although systemic treatments, including chemotherapy, targeted therapy, and immunotherapy, have improved in recent years, the overall survival rate of GC patients remains unsatisfactory[3,4]. Established risk factors for GC include Infection with Helicobacter pylori (H. pylori), advancing age, elevated dietary salt intake, and inadequate consumption of fruits and vegetables[5].

Among these factors, H. pylori infection is considered the most important etiological contributor to gastric carcinogenesis. Approximately 50% of the global population is infected with H. pylori[6]. Epidemiological studies estimate that this bacterium accounts for nearly 75% of GC cases and contributes to approximately 5.5% of all cancers worldwide[7]. Persistent infection with H. pylori induces chronic gastric inflammation and activates multiple oncogenic signaling pathways. In particular, increasing evidence indicates that H. pylori infection promotes phosphorylation and activation of nuclear factor-κB (NF-κB), a key transcription factor involved in inflammatory responses, tumor initiation, and cancer progression[8-12]. However, the downstream molecular mechanisms through which NF-κB signaling promotes gastric tumorigenesis remain incompletely understood.

Regulated cell death plays a crucial role in maintaining tissue homeostasis and preventing malignant transformation[13]. Beyond apoptosis and necroptosis, ferroptosis has recently been identified as a unique modality of regulated cell death, distinguished by iron-dependent lipid peroxidation[14,15]. Ferroptosis is primarily triggered by disruption of intracellular antioxidant systems, particularly inhibition of the cystine/glutamate antiporter system Xc^- (solute carrier family 7 member 11, SLC7A11) or inactivation of glutathione (GSH) peroxidase 4 (GPX4), both of which are essential for maintaining redox balance and preventing lipid reactive oxygen species (ROS) accumulation[16,17]. Increasing evidence suggests that ferroptosis functions as an important tumor-suppressive mechanism and represents a promising therapeutic target for cancer treatment. Interestingly, bacterial infections have been reported to influence ferroptotic processes and contribute to tissue damage and tumor progression[18].

Epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) is a member of the EPS8 family and participates in epidermal growth factor receptor signaling. This protein plays important roles in actin cytoskeleton remodeling, cell morphology regulation, and cell motility[19,20]. Previous studies have implicated EPS8L3 in several biological processes, including hair follicle cycling and Rho GTPase signaling pathways[21,22]. Emerging evidence also suggests that EPS8L3 contributes to tumor progression in multiple malignancies, such as hepatocellular carcinoma and prostate cancer[19,23-26]. Nonetheless, the biological role of EPS8L3 in GC, especially in relation to H. pylori infection, remains unexplored.

Importantly, although NF-κB signaling is known to regulate inflammatory responses and tumor progression, the molecular link between NF-κB activation and ferroptosis resistance in GC remains poorly defined. In the present investigation, we identify EPS8L3 as a novel downstream target of NF-κB that connects inflammatory signaling with ferroptosis regulation. We demonstrate that EPS8L3 suppresses ferroptosis by regulating GSH metabolism and maintaining the expression of SLC7A11 and GPX4, thereby promoting gastric tumorigenesis induced by H. pylori infection. Our findings reveal a previously unrecognized NF-κB/EPS8L3/ferroptosis regulatory axis in GC. Targeting EPS8L3 may therefore represent a potential therapeutic strategy for the treatment of H. pylori-associated GC.

MATERIALS AND METHODS

Cell culture

Normal gastric epithelial cells (GES-1) and GC cell lines, including AGS, SGC-7901, HGC-27, MGC-803, and MKN-45, were sourced from Genechem (Shanghai, China). The authenticity of these cell lines was verified by STR profiling. In a humidified incubator at 37 °C with 5% CO₂, cells cultivated in RPMI1640 medium (10-040-CVR Corning, United States) supplemented with 1% penicillin-streptomycin (SV30010 HyClone, United States) and 10% foetal bovine serum (FB25015 Clark Bioscience, United States).

H. pylori strains

H. pylori strains (ATCC43504 and ATCC26695) were purchased from Fengshou (Shanghai, China) Biotechnology Co., Ltd. The bacteria were cultured on Columbia blood agar plates supplemented with 5% sheep blood and maintained in a tri-gas incubator at 37 °C under microaerophilic conditions (85% N₂, 10% CO₂, and 5% O₂).

For infection experiments, H. pylori were harvested at the logarithmic growth phase and co-cultured with gastric epithelial or GC cells at a multiplicity of infection of 100:1. The co-culture was maintained for 24-48 hour, based on previously reported experimental conditions[9]. To ensure experimental reproducibility, heat-killed H. pylori (65 °C for 30 minutes) were used as an inactive bacterial control, while cells cultured without bacteria served as the negative control (NC).

Lentivirus infection

Lentiviruses, including EPS8L3 overexpression and EPS8L3 knockdown, were acquired from Anhui, China’s General Biological Company. Lentiviral transfection employing small hairpin RNA was conducted on AGS and SGC-7901, whereas lentiviral-mediated overexpression transfection was performed on HGC-27 and MGC-803. Stable knockdown and overexpression cell lines were created by planting around 1 × 10⁶ cells per well in 12-well plates and incubating them overnight. Lentivirus (multiplicity of infection = 10) and a NC were introduced the next day when the cells had achieved 30% to 50% confluence. To create stable cell lines, cells were chosen after 48 hours and treated with 2 mg/L puromycin for 2 weeks. For the tests that followed, 1 mg/L puromycin was used to maintain stable transfected cells. Overexpression and knockdown efficiencies were assessed by immunoblot analysis 72 hours post-infection.

The shEPS8L3 sequences were as follows: ShEPS8L3-1: 5’-GACCAGCAGGAAGAAGAAAUUTT-3’; shEPS8L3-2: 5’-AACCUGCUACAGUCCUGUCUATT-3’.

Organoid culture

Fresh GC tissue samples were transferred to a culture dish and subsequently rinsed two to three times with phosphate buffered saline (PBS) supplemented with 1% penicillin-streptomycin. The tissues were then cut into 2-3 mm³ fragments using surgical scissors and transferred into a 1.5 mL EP tube, where they were further minced into a pasty consistency. The minced tissue fragments were subsequently transferred to a 15 mL centrifuge tube and digested with 25-50 times the tissue volume of tissue digestion solution (OrganPharma Biotech, Hangzhou, China, Cat# NGH030021) for 20 minutes at 37 °C. The digestion was terminated by adding PBS and thorough mixing. The resultant suspension was passed through a 100 μm cell strainer, and the filtrate was subsequently collected and subjected to centrifugation at 200 g for 5 minutes at 4 °C. Following centrifugation, the supernatant was removed, and the cell pellet was resuspended in 1 mL of culture medium (OrganPharma Biotech, Hangzhou, China, Cat# NGH020007). Subsequently, 3 mL of Matrigel (OrganPharma Biotech, Hangzhou, China, Cat# NGH030024) was added and mixed thoroughly with the cell suspension. The mixture was then seeded into pre-warmed 48-well plates and allowed to solidify for 15 minutes in a 37 °C incubator. After solidification, 200 μL of culture medium (NGH020007) was added to each well. The plates were finally placed in a humidified incubator at 37 °C with 5% CO₂ for subsequent culture.

Co-immunoprecipitation

Collect the lysis buffer of the cell samples, add 2 μg of each primary antibody or IgG and mix it with 500 μg of the protein sample. Incubate overnight at 4 °C by rotation. Next, mix the sample with protein A/G agarose and let it sit overnight. Centrifuge and discard the supernatant. Detect the immunoprecipitation complex by western blotting.

Chromatin immunoprecipitation

A chromatin immunoprecipitation assay was carried out using a dedicated kit (#abs53004, absin, China) following the instruction manual. GC cells, transfected with an NF-κB plasmid for 48 hours, were fixed with 1% formaldehyde and subsequently quenched with glycine. The fixed cells were lysed, and the chromatin was sheared to an average size of 50-150 bp by enzymatic digestion (37 °C, 10 minutes). An aliquot of the pre-cleared chromatin lysate was incubated overnight at 4 °C with magnetic beads conjugated to a specific anti-NF-κB antibody. Normal IgG was used as a NC. After a series of washes, the immunoprecipitated protein-DNA complexes were eluted and reverse-cross-linked. The isolated DNA was subsequently analyzed using real-time quantitative polymerase chain reaction (PCR) to quantify the binding affinity of NF-κB to the EPS8L3 promoter region.

Western blot analysis

Cellular and tissue samples were lysed utilizing MPER (#78501, Thermo Scientific, United States), contained with inhibitors. Protein concentrations were measured with the bicinchoninic acid protein assay kit (#P0012, Beyotime, China). Sodium-dodecyl sulfate gel electrophoresis was used to separate equal quantities of protein, which were then transferred to polyvinylidene fluoride membranes (Millipore, United States). Primary antibodies against EPS8L3 (AF9061, Affinity Biosciences, China), SLC7A11 (ab175186, Abcam, United Kingdom), GPX4 (ab252833, Abcam, United Kingdom), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, D110016, Sangon Biotech, China) were used to block the membranes for one hour at room temperature. The protein was detected using the enhanced chemiluminescence luminescent solution (SQ201, EpiZyme, China) and the chemical imaging system after incubation with secondary antibody.

Quantitative real-time PCR

TRIzol reagent (Invitrogen, United States) was used to extract total RNA from GC cells and tissues. The DNA synthesis solution (11141ES60, Yeasen Biotechnology, China) was used to synthesize complementary DNA. SYBR Green solution (11202ES03, Yeasen Biotechnology, China,) was used for quantitative PCR using a QuantStudio Real-Time PCR equipment (Thermo Scientific, United States). The relative expression levels of the genes were quantified using the 2^-ΔΔCt method, with GAPDH serving as the internal reference gene. General Biosystems (Anhui, China) created the following primer sequences: Forward EPS8L3 5’-CACAGGGTGGAGCACTTGAT-3’; EPS8L3 reverse 5’-GATGCTGTCTAGGCGGTAAGAG-3’; GAPDH forward 5’-ATCAAGAAGGTGGTGAAGCAGG-3’; GAPDH reverse 5’-CGTCAAAGGTGGAGGAGTGG-3’.

RNA sequencing analysis

Seqhealth Technology Co., Ltd. (Wuhan, China) carried out RNA-seq. The Illumina NovaSeq 6000 platform was used for sequencing following the evaluation of RNA quality and library preparation.

5-ethynyl-2’-deoxyuridine assay

The BeyoClick™ Edu Kit (#C0078S, Beyotime, China) was used to assess cell proliferation. After being sown in 24-well plates, infected GC cells were left to incubate for the whole night. After that, cells were exposed to 10 μM EdU for 2 hours at 37 °C with 5% CO₂. EdU-labeled cells were incubated with Click Addictive Solution for 30 minutes at room temperature. After 10 minutes of Hoechst 33342 staining (1:1000), cells were inspected and photographed under a microscope.

Transwell migration and invasion assays

24-well Transwell chambers with an 8-μm hole size were used for cell migration and invasion tests (Corning, United States). For the invasion or migration experiment, GC cells (8 × 10⁴) were planted in the upper chamber either with or without Matrigel (BD Biosciences, United States). There were 650 μL of culture media with 20% foetal bovine serum in the bottom chamber. Following a 24-hour incubation period, cells were stained with 0.1% crystal violet and fixed with 4% paraformaldehyde. A moist cotton swab was then used to remove any non-invading or non-migrating cells. The tagged cells were photographed using a Leica microscope (DMI1; Wetzlar, Germany).

Colony formation assay

For the colony formation assays, about 1000 GC cells were plated in culture plates. Following an incubation period of two weeks at 37 °C, the resulting colonies were fixed with 4% paraformaldehyde for 20 minutes, subsequently stained with 1% crystal violet, and quantified using digital imaging techniques. All experiments were conducted at least in triplicates. The Mann-Whitney U-test was used for statistical difference analysis.

Cell scratch assay

Cells were plated in 6-well culture plates at a suitable density one day prior to experimentation. Upon reaching confluence, the medium was removed, and cell monolayers were scratched using a 10 μL pipette. Detached cells were washed off with PBS, and the remaining cells were cultured in a serum-free medium. Scratch images were captured at 0, 24, and 48 hours using a microscope (IX70-S8F2, Olympus, Japan) and the wound healing ratio was analyzed using ImageJ software.

Malondialdehyde detection and intracellular ROS

DCFH–DA (S0033S, Beyotime, China) was used to measure intracellular ROS. In short, GC cells grown in plates were exposed to 10 μM DCFH–DA for 30 minutes in the dark at 37 °C in a serum-free medium. Following three PBS washes, the cells were resuspended in 500 μL PBS and subjected to flow cytometry analysis (Beckman Coulter, United States). Using a cell malondialdehyde (MDA) test kit (S0131S, Beyotime, China) and the thiobarbituric acid technique, the concentration of MDA in GC cells was determined.

Patients and tissue microarrays

Between 2012 and 2013, 107 GC tissues and 22 nearby normal tissues were obtained from the General Surgery Department at Anhui Medical University’s First Affiliated Hospital. At least five centimeters away from the tumor’s edge, adjacent normal tissues were removed and histologically verified to be tumor cell-free. Patients were included if they had pathologically confirmed primary gastric adenocarcinoma and underwent radical surgical resection. Patients who received preoperative chemotherapy or radiotherapy or had incomplete clinical data were excluded from the study. H. pylori infection status was determined using histopathological examination and/or urea breath test results recorded in the medical records. The duration of the follow-up was between eight and seventy-one months. The American Joint Committee (7^th edition) and the tumor-node-metastasis staging approach were used to analyze and stage all tissues. Table 1 summarizes pathological and clinical findings.

Table 1 Correlation of epidermal growth factor receptor kinase substrate 8-like protein 3 expression with clinicopathologic parameters in gastric cancer patients.

Variables	Cases	EPS8L3 expression		χ2	P value
		Low	High
Gender				2.51	0.113
Male	71	28	43
Female	36	20	16
Age (years)				0.12	0.732
< 61	56	26	30
≥ 61	51	22	29
Tumor location				0.78	0.377
Upper	36	14	22
Middle + lower	71	34	37
Tumor size (cm)				6.39	0.011a
< 6	48	28	20
> 6	59	20	39
Depth of invasion				6.86	0.009a
T1 + T2	29	19	10
T3 + T4	78	29	49
Differentiation				1.09	0.297
Well + moderate	41	21	20
Poor + undifferentiated	66	27	39
Lymph node metastasis				0.63	0.426
Absent	23	12	11
Present	84	36	48
TNM stage				0.10	0.755
I + II	34	16	18
III + IV	73	32	41

^aP < 0.05.

EPS8L3: Epidermal growth factor receptor kinase substrate 8-like protein 3; TNM: Tumor-node-metastasis.

Immunohistochemical staining

As previously indicated, the expression of the EPS8L3 protein was evaluated through immunohistochemical (IHC) performed on a TMA[16]. Two pathologists independently graded the staining on the basis of staining area. The scale is defined as follows: 0 corresponds to none; 1 represents 0% to 25%; 2 indicates 26% to 50%; 3 denotes 51% to 75%; and 4 signifies 76% to 100%, and intensity (0 indicating negative; 1 indicating mild; 2 indicating moderate; 3 indicating strong) after the TMA was incubated with anti-EPS8L3. The staining area multiplied by the intensity yielded the final score. A final score of ≥ 5 indicated high expression, whereas a score of 0-4 indicated poor expression.

Xenograft mouse model

The Anhui Medical University Ethics Committee for Animal Studies authorized the animal trials, which were carried out in accordance with National Institutes of Health guidelines. Male BALB/c nude mice that were four weeks old and kept in specified pathogen-free conditions were acquired from GemPharmatech Animal Center in Jiangsu, China. Using a random number generator, mice were divided into experimental groups at random (n = 6). Each mouse’s axilla received an injection of transfected GC cells (5 × 10² cells in 100 μL). Every three days, the tumor’s development was tracked, and its volume was determined using the formula V = (length × width²)/2. To minimize experimental bias, tumor measurements and data analyses were performed in a blinded manner. Mice had ad libitum access to food and water throughout the experiment. Mice demonstrating a body weight reduction of 20% or greater, or presenting with a tumor size larger than 1.5 cm, were humanely euthanized in accordance with established humane endpoint criteria.

Bioinformatic analysis

The Cancer Genome Atlas (https://tcga-data.nci.nih.gov/tcga/) provided the GC mRNA data, which were then bioinformatically analyzed using Xiangtan (https://www.xiantaozi.com/). Kaplan-Meier Plotter (www.kmplot.com) was used to create survival curves for GC patients with high and low EPS8L3 expression. The GEPIA (http://gepia.cancer-pku.cn/) online database was used to assess EPS8L3 expression across various tumors and predict potential interactions between the EPS8L3 gene and NF-κB.

Statistical analysis

GraphPad Prism 9.5 and SPSS 22.0 (GraphPad Software Inc., United States; SPSS Inc., United States) were used for statistical analyses. The mean ± SD is used to display the data. The Shapiro-Wilk test was used to determine the normality of the data prior to the use of parametric tests. Student’s t-test was used to compare two groups, and one-way analysis of variance followed by Bonferroni’s post hoc test for multiple comparison correction was used to compare multiple groups. Pearson’s χ²d test was used to examine the relationship between EPS8L3 expression and clinicopathological factors. The Kaplan-Meier technique was used for survival analysis, and the log-rank test was used to assess group differences. Statistical significance was defined as a P value < 0.05.

RESULTS

EPS8L3 upregulation in patients with GC indicated a poor prognosis

The GEPIA database was employed to analyze the expression pattern of EPS8L3 within the gastrointestinal tract. The findings indicated that EPS8L3 expression was significantly elevated in multiple tumor types, including GC (Figure 1A). Western blot (Figure 1A-C) and quantitative reverse transcription PCR (Figure 1D and E) further confirmed that EPS8L3 expression was significantly elevated in GC cell lines (SGC7901, AGS, MKN45) and tissue samples. These results suggest that EPS8L3 may play a pivotal role in driving cancer progression. Furthermore, IHC was performed on TMAs to assess EPS8L3 expression. Representative images of varying EPS8L3 expression levels are shown in (Figure 1F). To assess the clinical significance of EPS8L3 in GC, the association between EPS8L3 expression levels and various clinicopathological parameters was examined. The results are summarized in Table 1, increased EPS8L3 expression was significantly associated with tumor size (P = 0.003) and depth of invasion (P = 0.009). Furthermore, the analysis of overall survival demonstrated that patients with GC characterized by high EPS8L3 expression experienced significantly reduced survival durations in comparison to patients exhibiting lower levels of EPS8L3 expression (log-rank P < 0.001). The result is consistent with findings from the Kaplan-Meier plotter database (Figure 1G and H), and tumors in patients with high EPS8L3 expression were found to be larger (Figure 1I).

Figure 1 Epidermal growth factor receptor kinase substrate 8-like protein 3 upregulation in patients with gastric cancer indicated a poor prognosis. A: GEPIA analysis revealed that epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) is upregulated across various tumor types, including cholangiocarcinoma, pancreatic adenocarcinoma, and gastric cancer (GC); B and C: Western blot analysis confirmed a marked upregulation of EPS8L3 protein in GC cell lines (B) and tissues(C); D and E: Quantitative reverse transcription polymerase chain reaction demonstrated significantly higher EPS8L3 mRNA expression in GC cell lines (D) and patient tissues (E); F: Representative EPS8L3 immunohistochemical staining images in a GC tissue microarray; G: Kaplan-Meier survival curve of EPS8L3 expression obtained from the Kaplan-Meier plotter database; H: Kaplan-Meier survival curve of patients with GC exhibiting low and high EPS8L3 expression in our study; I: Immunohistochemical score indicated that the EPS8L3 expression score correlated with tumor size. Data are shown as the mean ± SD (n = 3). ^aP < 0.05 vs group normal, ^bP < 0.01 vs control, and ^cP < 0.001 vs control. CHOL: Cholangiocarcinoma; LIHC: Liver hepatocellular carcinoma; PAAD: Pancreatic adenocarcinoma; READ: Rectum adenocarcinoma; STAD: Stomach adenocarcinoma; UCEC: Uterine corpus endometrial carcinoma; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

H. pylori infection increases EPS8L3 expression

As previously reported[7-9], H. pylori infection is a well-established contributor to GC. In this study, the potential correlation between H. pylori infection and the elevated expression of EPS8L3 in GC was investigated. AGS were co-cultured with H. pylori (ATCC43504), followed by transcriptome sequencing, which revealed that H. pylori significantly increased the mRNA levels of EPS8L3 (Figure 2A and B). Additionally, Gene Set Enrichment Analysis indicated that H. pylori infection activated the NF-κB signaling pathway (Figure 2C). The GEPIA database was employed to explore the correlation between EPS8L3 and NF-κB (Figure 2D). To validate these findings, two H. pylori strains (ATCC26695, ATCC43504) were co-cultured with GES1 and GC cell lines (AGS, MKN-45), and proteins were extracted for analysis. The results confirmed that H. pylori infection led to an upregulation of both EPS8L3 and p-NF-κB protein expression (Figure 2E-G). Additionally, significantly elevated EPS8L3 expression was observed in GC tissues of H. pylori-positive patients relative to their H. pylori-negative counterparts (Figure 2H).

Figure 2 Helicobacter pylori infection increases epidermal growth factor receptor kinase substrate 8-like protein 3 expression. A and B: Heatmap (A) and volcano plot (B) displaying the overall transcriptomic changes in AGS cells after Helicobacter pylori (H. pylori) infection, with epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) significantly upregulated among the differentially expressed genes; C: Gene Set Enrichment Analysis enrichment analysis of RNA-seq data revealed that H. pylori infection activates the nuclear factor kappa-B (NF-κB) signaling pathway; D: The GEPIA database showed a significant correlation between EPS8L3 and NF-κB; E-G: Protein levels of EPS8L3 and NF-κB increased in GES-1, AGS, and MKN45 cells over time in co-culture with H. pylori; H: Immunohistochemical analysis revealed a significantly higher level of EPS8L3 expression in tumor tissues from H. pylori-positive gastric cancer patients compared to those from H. pylori-negative patients. Data are shown as the mean ± SD (n = 3). ^aP < 0.05 vs control, ^bP < 0.01 vs control, ^cP < 0.001 vs control. NF-κB: Nuclear factor kappa-B; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

EPS8L3 overexpression enhanced the proliferation, migration, and invasion of GC cells

Given that H. pylori is a bacterial pathogen, in vitro co-culture with cells results in reduced cell viability over time. To compensate for this, lentiviral vectors were used to overexpress EPS8L3 in GC cells, restoring the elevated EPS8L3 levels induced by H. pylori infection. Given the low endogenous expression levels of EPS8L3 in HGC-27 and MGC-803 cells (Figure 1B), were selected to overexpression for further investigate the role of EPS8L3 in GC. Western blotting was performed to assess the efficiency of overexpression (Figure 3A). The proliferative capacity of these cells was assessed via clonogenic assays, revealing that EPS8L3 overexpression led to the formation of more clones compared to controls (Figure 3B and C). EdU staining confirmed these results, with a higher proportion of EdU-positive cells in the EPS8L3 overexpression group (Figure 3D and E). Moreover, Transwell assays (Figure 3F) and scratch wound assays (Figure 3G and H) demonstrated that EPS8L3 overexpression significantly enhanced the migration and invasion capacities of GC cells. Collectively, these data highlight the critical role of EPS8L3 in promoting the biological processes of GC cells.

Figure 3 Overexpression of epidermal growth factor receptor kinase substrate 8-like protein 3 enhances proliferation, migration, and invasion of gastric cancer cells in vitro. A: Western blot analysis confirmed the efficiency of epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) overexpression; B and C: Colony formation assays showed enhanced colony formation in gastric cancer cells upon EPS8L3 overexpression; D and E: Overexpression of EPS8L3 significantly increased cell proliferation in MGC-803 and HGC-27 cells; F: Invasion and migration capacities were markedly enhanced in MGC-803 and HGC-27 cells upon EPS8L3 overexpression; G and H: Cell scratch assay revealed that EPS8L3 overexpression significantly increased migration in MGC-803 cells. Data are shown as the mean ± SD (n = 3). ^aP < 0.05 vs control, ^bP < 0.01 vs control, and ^cP < 0.001 vs control. EPS8L3: Epidermal growth factor receptor kinase substrate 8-like protein 3; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; OE: Overexpression; NC: Negative control.

EPS8L3 knockdown suppressed the proliferation, migration, and invasion of GC cells

Considering the high endogenous levels of EPS8L3 in AGS and SGC-7901 cells (Figure 1B), to investigate the function of EPS8L3 in an in vitro setting, AGS and SGC-7901 cells were transduced with two different small hairpin RNAs targeting EPS8L3 to reduce its endogenous expression. Lentiviral-mediated knockdown efficiency was confirmed by western blot prior to proceeding with subsequent experiments (Figure 4A). The effects of stable EPS8L3 knockdown on clonogenic potential were assessed in polyclonal cells. Results demonstrated a significant reduction in clonogenicity following EPS8L3 knockdown (Figure 4B and C). To further explore the impact of EPS8L3 silencing on cell proliferation, EdU assays were performed. These assays revealed a marked decrease in the proliferative capacity of GC cells with EPS8L3 knockdown (Figure 4D and E). Additionally, Transwell assays confirmed that the invasion and migration abilities of GC cells were significantly diminished following EPS8L3 knockdown, as compared to controls (Figure 4F and G). These observations were corroborated by the results of the scratch wound healing assay (Figure 4H and I).

Figure 4 Knockdown of epidermal growth factor receptor kinase substrate 8-like protein 3 inhibits proliferation, migration, and invasion of gastric cancer cells in vitro. A: Western blot analysis confirmed efficient epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) knockdown; sh1 and sh2 demonstrated the knockdown efficacy; B and C: Colony formation assays showed a significant reduction in colony formation upon EPS8L3 knockdown; D and E: Edu assays demonstrated a marked decrease in cell proliferation in AGS and SGC-7901 cells upon EPS8L3 knockdown; F and G: Transwell assays indicated a significant reduction in invasion and migration in AGS (F) and SGC-7901 (G) cells with EPS8L3 knockdown; H and I: Cell scratch assays revealed decreased migration capacity in AGS (H) and SGC-7901 (I) cells following EPS8L3 silencing. Data are shown as the mean ± SD (n = 3). ^bP < 0.01 vs negative control group, ^cP < 0.001 vs negative control group. EPS8L3: Epidermal growth factor receptor kinase substrate 8-like protein 3; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; NC: Negative control.

H. pylori activates NF-κB to regulate EPS8L3, inhibition of NF-κB decreases EPS8L3 expression

To investigate the potential involvement of H. pylori in the upregulation of EPS8L3, it is hypothesized that H. pylori infection could activate the NF-κB transcription factor, leading to increased EPS8L3 transcription, based on prior findings (Figure 2C and D)[10,27-29]. Western blot results supported this hypothesis, confirming that H. pylori infection induces the activation of NF-κB signaling pathways and upregulates EPS8L3 expression (Figure 2E). To further explore the relationship between EPS8L3 and NF-κB, the NF-κB inhibitor BAY 11-7082 was employed. Inhibition of NF-κB led to a reduction in EPS8L3 expression, and the degree of EPS8L3 downregulation was inversely correlated with the duration of BAY 11-7082 treatment (Figure 5A and B). To identify putative NF-κB binding sites, we screened the EPS8L3 promoter sequence using the JASPAR database (Figure 5C). The transcriptional regulation of EPS8L3 by NF-κB was further confirmed by chromatin immunoprecipitation assays (Figure 5D). Rescue experiments demonstrated that the addition of BAY 11-7082 suppressed the phenotypic effects induced by EPS8L3 overexpression (Figure 5E and F). These results suggest that H. pylori infection activates NF-κB phosphorylation, which in turn promotes EPS8L3 transcription.

Figure 5 Helicobacter pylori activates nuclear factor kappa-B to regulate epidermal growth factor receptor kinase substrate 8-like protein 3, inhibition of nuclear factor kappa-B decreases epidermal growth factor receptor kinase substrate 8-like protein 3 expression. A and B: Western blot results showed that in three cell lines (GES-1, AGS, and MKN-45), the nuclear factor kappa-B (NF-κB) inhibitor BAY 11-7082 inhibited the protein levels of NF-κB and p-NF-κB, and the protein levels of epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) were also reduced by NF-κB and p-NF-κB. The inhibitory effect became more pronounced over time; C: The Jaspar website predicts the NF-κB binding site in the EPS8L3 promoter; D: Chromatin immunoprecipitation assays were performed in AGS cells with p65 transient expression; E and F: Western blot showing that treatment with BAY11-7082 attenuates EPS8L3 overexpression. Data are shown as the mean ± SD (n = 3). ^cP < 0.001 vs control. EPS8L3: Epidermal growth factor receptor kinase substrate 8-like protein 3; NF-κB: Nuclear factor kappa-B; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

EPS8L3 affects GSH metabolism and intervenes in GC cell ferroptosis

In vivo experiments have demonstrated that EPS8L3 plays a critical role in the proliferation, migration, and invasion of GC cells. To explore the mechanism by which EPS8L3 affects GC cell development, Gene Set Enrichment Analysis was performed. The results revealed a significant correlation between EPS8L3 and GSH metabolism (Figure 6A). Since GSH is a key cofactor for GPX4 in ferroptosis, this finding is particularly significant. It is hypothesized that EPS8L3 may indirectly influence ferroptosis in GC cells by modulating GSH metabolism through its effect on SLC7A11, a transporter responsible for GSH’s intracellular transport. Subsequently, the levels of SLC7A11 and GPX4 proteins were assessed (Figure 6B and C). Knockdown of EPS8L3 led to a reduction in both SLC7A11 and GPX4 expression, resulting in the accumulation of lipid ROS in GC cells and triggering ferroptosis. In contrast, overexpression of EPS8L3 significantly elevated the levels of SLC7A11 and GPX4 (Figure 6B and C), promoting the degradation of lipid ROS and thereby preventing ferroptosis. MDA, a final product of lipid peroxidation, is positively correlated with ferroptosis[30]. To further investigate the role of EPS8L3 in ferroptosis, oxidative stress markers, including MDA and ROS, were measured. The findings demonstrated a marked elevation in the levels of both MDA and ROS in cells with EPS8L3 knockdown (Figure 6D and E), while EPS8L3 overexpression had the opposite effect (Figure 6D and E). These results suggest that EPS8L3 modulates ferroptosis in GC cells by influencing oxidative stress levels. To validate the protein-protein interactions, we performed co-immunoprecipitation and confirmed that EPS8L3 binds to both SLC7A11 and GPX4 (Figure 6F).

Figure 6 Relationship between the expression level of epidermal growth factor receptor kinase substrate 8-like protein 3 and indicators related to ferroptosis in gastric cancer. A: Kyoto Encyclopedia of Genes and Genomes analysis showed that the expression level of epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) in gastric cancer (GC) was significantly correlated with glutathione metabolism; B and C: Western blot results showed that knockdown of EPS8L3 in GC cells significantly reduced the protein levels of solute carrier family 7 member 11 and glutathione peroxidase 4, and the opposite was true for overexpression of EPS8L3; D: Flow assay of reactive oxygen species levels showed that knockdown of EPS8L3 in GC cells significantly increased intracellular reactive oxygen species levels, and the opposite was true for overexpression of EPS8L3; E: Malondialdehyde assay results indicated that knockdown of EPS8L3 significantly increased malondialdehyde levels in GC cells (AGS and SGC-7901), and the opposite was true for overexpression of EPS8L3; F: Co-immunoprecipitation assay demonstrating a direct interaction between EPS8L3 and both solute carrier family 7 member 11 and glutathione peroxidase 4. Data are shown as the mean ± SD (n = 3). ^aP < 0.05 vs negative control group, ^bP < 0.01 vs negative control group, ^cP < 0.001 vs negative control group. SLC7A11: Solute carrier family 7 member 11; GPX4: Glutathione peroxidase 4; EPS8L3: Epidermal growth factor receptor kinase substrate 8-like protein 3; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; NC: Negative control; MDA: Malondialdehyde; OE: Overexpression.

In vivo and GC organoid experiments confirmed that EPS8L3 promotes GC

In vitro experiments have established that EPS8L3 promotes GC progression. To further assess its role in vivo, xenograft experiments were conducted in nude mice (Figure 7A and B). Each group consisted of 6 mice, with one group as the control and the other group with EPS8L3 knockdown. Tumors in the EPS8L3 knockdown group displayed significantly reduced size and weight compared to the control group (Figure 7C and D). IHC and hematoxylin and eosin staining of tumor sections, along with Ki-67 staining, revealed a notable reduction in proliferative activity in the EPS8L3-deficient tumors relative to controls (Figure 7E). Patient-derived GC organoids were established to better recapitulate the tumor microenvironment. In this model, EPS8L3 knockdown markedly inhibited organoid proliferation. Conversely, co-culture with H. pylori effectively reversed the growth suppression induced by EPS8L3 knockdown (Figure 7F).

Figure 7 Knockdown of epidermal growth factor receptor kinase substrate 8-like protein 3 inhibits tumor growth in in vivo experiments. A and B: Six nude mice per group were injected with equal amounts of control and sh-epidermal growth factor receptor kinase substrate 8-like protein 3 (EPS8L3) SGC-7901 cells; C and D: Tumor weight (C) and volume (D) profiles indicated that knockdown of EPS8L3 inhibited the proliferation of GC cells in vivo; E: Ki67 staining showing higher expression in the negative control group than in the sh-group in vivo, and hematoxylin and eosin staining showing tumor tissue morphology; F: Representative images of patient-derived gastric cancer organoids showing that growth inhibition by EPS8L3 knockdown is attenuated by Helicobacter pylori co-culture. Data are shown as the mean ± SD (n = 3). ^bP < 0.05. NC: Negative control.

DISCUSSION

GC remains one of the most common malignancies worldwide, and H. pylori infection is widely recognized as a major etiological factor in gastric carcinogenesis[1]. In the present study, we investigated the potential role of EPS8L3 in GC progression in the context of H. pylori infection. Our findings indicate that H. pylori infection is associated with increased EPS8L3 expression in both GC cells and normal gastric epithelial cells. Moreover, elevated EPS8L3 expression correlated with poorer clinical outcomes in patients with GC, suggesting that EPS8L3 may contribute to gastric tumor progression.

Mechanistically, our data suggest that H. pylori infection may promote EPS8L3 transcription via activating NF-κB signaling. Inhibition of NF-κB activity using BAY 11-7082 suppressed both NF-κB phosphorylation and EPS8L3 expression, indicating that EPS8L3 may function as a downstream effector of NF-κB signaling during H. pylori infection. These findings are consistent with previous studies demonstrating that H. pylori activate NF-κB-dependent inflammatory pathways that contribute to gastric tumorigenesis[10,28-33].

A growing body of evidence indicates that ferroptosis, an iron-dependent regulated form of cell death marked by lipid peroxidation, holds a significant role in the biology of cancer. This form of cell death can be induced by a range of anticancer treatments, including chemotherapy, radiotherapy, immunotherapy, and targeted therapeutic strategies[34-36]. Due to their altered metabolic states and elevated ROS levels, cancer cells may exhibit increased susceptibility to ferroptosis under specific conditions[27,37,38]. However, the potential relationship between H. pylori infection and ferroptosis regulation in GC remains poorly understood. In this study, we observed that EPS8L3 expression was associated with reduced markers of ferroptotic cell death in GC cells, including decreased ROS and MDA accumulation. Furthermore, EPS8L3 expression correlated with increased levels of SLC7A11 and GPX4, two key regulators that maintain cellular redox homeostasis and suppress ferroptosis. Co-immunoprecipitation assays suggested that EPS8L3 may interact with the SLC7A11/GPX4 complex; however, although co-immunoprecipitation experiments confirmed protein interactions between EPS8L3 and SLC7A11/GPX4, such evidence alone is insufficient to definitively establish a functional interaction. Additional biochemical and structural studies are required to determine whether EPS8L3 directly regulates these proteins or influences their stability through indirect mechanisms.

EPS8L3 belongs to the EPS8 protein family and is known to participate in epidermal growth factor receptor signaling and cytoskeletal remodeling[31,32]. Previous studies have implicated EPS8 family proteins in cell migration, invasion, and tumor progression. Consistent with these reports, our functional experiments demonstrated that EPS8L3 knockdown significantly reduced GC cell invasion, whereas its overexpression enhanced invasive capacity. These results suggest that EPS8L3 may contribute to GC progression through multiple mechanisms, including modulation of cellular redox homeostasis and ferroptosis susceptibility.

Despite these findings, several limitations of the present study should be acknowledged. First, the clinical data were derived from a single-center patient cohort, which may limit the generalizability of our observations. Second, although H. pylori-cell co-culture models are widely used to study bacterial-host interactions, they may not fully replicate the complex tumor microenvironment associated with chronic H. pylori infection in vivo. In particular, the absence of a chronic infection animal model represents an important limitation of the current study. Future investigations incorporating long-term infection models or H. pylori challenge experiments in vivo will be necessary to further validate the role of EPS8L3 under physiologically relevant conditions.

Another limitation concerns the use of the NF-κB inhibitor BAY 11-7082. Although this compound is widely used to inhibit NF-κB signaling, previous studies have reported potential off-target effects, including modulation of NLRP3 inflammasome activation and other inflammatory pathways[39-42]. Therefore, the observed phenotypic changes following BAY 11-7082 treatment should be interpreted cautiously. Complementary genetic approaches, such as inducible NF-κB pathway inhibition or CRISPR-mediated gene editing, may help to further clarify the role of NF-κB-dependent regulation of EPS8L3.

Interestingly, recent studies have highlighted the potential of combination therapeutic strategies targeting ferroptosis pathways in cancer treatment. Emerging evidence suggests that ferroptosis-inducing agents may enhance the efficacy of conventional chemotherapy and immune checkpoint inhibitors, thereby overcoming therapeutic resistance in several malignancies[43]. In this context, targeting EPS8L3 could potentially sensitize GC cells to ferroptosis-based therapies. Future studies should explore whether may provide synergistic therapeutic benefits for patients with H. pylori-associated GC.

CONCLUSION

In conclusion, our findings suggest that EPS8L3 expression is upregulated in H. pylori-infected GC cells and may be regulated by NF-κB signaling. Elevated EPS8L3 expression is associated with reduced ferroptosis susceptibility and enhanced tumor progression (Figure 8). Although further studies are required to clarify the precise molecular mechanisms involved, these results highlight EPS8L3 as a potential therapeutic target in H. pylori-associated GC and provide a basis for future investigation of ferroptosis-based combination treatment strategies.

Figure 8 Helicobacter pylori infection promotes gastric cancer progression through epidermal growth factor receptor kinase substrate 8-like protein 3-mediated inhibition of ferroptosis. Helicobacter pylori promotes the activation of p-nuclear factor kappa-B and regulates transcription that promotes epidermal growth factor receptor kinase substrate 8-like protein 3 expression. High expression of epidermal growth factor receptor kinase substrate 8-like protein 3 promotes solute carrier family 7 member 11 and glutathione peroxidase 4 by affecting glutathione metabolism, thereby suppressing the occurrence of ferroptosis in cells. NF-κB: Nuclear factor kappa-B; GSH: Glutathione; EPS8L3: Epidermal growth factor receptor kinase substrate 8-like protein 3; GPX4: Glutathione peroxidase 4; ROS: Reactive oxygen species.

ACKNOWLEDGEMENTS

We thank Jing Li and Yi-Da Lu for their invaluable guidance during the course of experiments.