Human GC cell lines MKN-45 were obtained from BeNa Culture Collection (Shanghai, China), AGS from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). All cell lines were cultured in 1,640 medium (Gibco, United States) supplemented with 10% fetal bovine serum (Clark, Shanghai, China) and penicillin-streptomycin (Gibco, United States), in a humidified environment of 37°C with 5% CO2. To establish radioresistant cell lines, AGS and MKN-45 cells were irradiated at a cell density of 60% with a dose of 4 Gy. Immediately following irradiation, the culture medium was replaced with fresh medium, which was then changed every 2 days. Cells were continuously cultured until they reached 90% confluence, at which point they were enzymatically dissociated using trypsin and transferred to new T25 culture flasks. This process of culturing and irradiation was repeated until a cumulative dose of 60 Gy was achieved (over 15 sessions), resulting in the establishment of the radioresistant cell lines MKN-45/IR and AGS/IR (Li et al., 2014; Petragnano et al., 2020).

Cell viability was assessed using the CCK-8 according to the manufacturer’s instructions. Cells were seeded in 96-well plates at a density of 1,000 cells per well. To prevent edge effects, equal volumes of PBS were added to the peripheral wells. Cells were then subjected to irradiation at doses of 0, 2, 4, 6, and 8 Gy or treated with 100 mg/L β-elemene, or a combination of both. After 48 h, 100 μL of diluted CCK-8 solution was added to each well and incubated for 1–2 h. Absorbance was measured at 450 nm using a microplate reader. For the clonogenic assay, logarithmically growing cells were used. After treatment with trypsin to create a single-cell suspension, cells were seeded in 6-well plates at a density of 1,000 cells per well. Twenty-4 hours later, cells were either irradiated at varying doses of 0, 2, 4, 6, and 8 Gy, treated with 100 mg/L β-elemene, or received both treatments. After 14 days of incubation, colonies were fixed with polyethylene glycol for 20 min and stained with crystal violet. Colonies containing at least 50 cells were counted. Colony survival rates for each treatment were calculated. The EDU assay was performed using the BeyoClick™ EdU-555 Cell Proliferation Assay Kit (Beyotime, Shanghai, China) according to the manufacturer’s instructions. Images were captured using a Leica Thunder fully automated inverted microscope. The experiment was conducted in triplicate.

To explore how β-elemene inhibits the proliferation of radioresistant GC, several specific inhibitors were utilized. These included the ferroptosis inhibitor Ferroptosis Inhibitor-1 (Fer-1), the iron chelator Desferrioxamine (DFO), the autophagy inhibitor 3-Methyladenine (3-MA), the apoptosis inhibitor Z-VAD-FMK (Z-Val-Ala-Asp(OMe)-fluoromethylketone), and the necrosis inhibitor Necrostatin-1 (Nec-1). All reagents were sourced from MedChemExpress (the United States) and stored at −20°C.

RNA was isolated utilizing TRIzol reagent (Invitrogen, United States). The qRT-PCR process was carried out in accordance with methods outlined earlier (Sindhuja et al., 2022). Primer sequences are provided: GPX4:F:5′-GAGGCAAGACCGAAGTAAACTAC-3′,R:5′-CCGAACTGGTTACACGGGAA-3’.

Operation was performed according to the instructions of the MDA assay kit (Beyotime, China). Prepare the working solution by washing the cells to be tested three times with PBS buffer solution. Then, lyse the cells with RIPA lysis buffer containing 1% PMSF for 30 min. Transfer 0.15 mL of the lysate to the sample tube. Add 0.15 mL of the standard solution to the standard tube and 0.15 mL of distilled water to the blank tube. Add 2.5 mL of TBA to each tube, incubate at 100°C for 40 min, and cool with tap water. Centrifuge at 4,000 rpm for 10 min. Transfer the supernatant to a 96-well plate, set the wavelength to 532 nm, and measure the absorbance using a multifunctional microplate reader.

Intracellular ROS levels were measured using 2,7-dichlorofluorescein diacetate (DCFH-DA) (Yeasen, Shanghai, China). Cells were incubated with 10 μM DCFH-DA diluted in RPMI-1640 medium at 37°C in the dark for 30 min. Fluorescence intensity was quantified using ImageJ software. Cells were analyzed using flow cytometry.

Operate according to the instructions of the iron ion colorimetric test kit manual (Beyotime, China).

Total protein separation and Western blot were conducted following the aforementioned methods (Liu et al., 2020; He et al., 2024). The following antibodies were utilized: anti-GPX4, anti-GAPDH, anti-HA and anti-Flag (Proteintech, China), anti-OTUB1(Affinity, China).

Total cell lysates were incubated with 1 μg of primary antibody or negative control rabbit IgG overnight at 4°C. Subsequently, 20 μL of Protein A+ G Agarose (Bioworld Technology, St. Louis Park, MN, United States) was added, and the mixture was further incubated for 2 h at 4°C. The protein-antibody mixtures were washed four times with PBS and collected by agarose bead spinning. Following this, proteins were identified through SDS-polyacrylamide gel electrophoresis and subsequent Western blot analysis.

First, GC cells were washed three times with cold phosphate-buffered saline (PBS) (Gibco, United States). They were then fixed with 4% paraformaldehyde (Beyotime, China) for 25 min, followed by permeabilization with 0.1% Triton X-100 (Beyotime, China) for 10 min. This was followed by a blocking step in 5% bovine serum albumin (Solarbio, China). Subsequently, the cells were incubated with anti-GPX4 antibody overnight at 4°C, followed by three washes with PBS. Next, the cells were incubated with Alexa Fluor 488 conjugated goat anti-Mouse IgG (ABclonal, China) for 1 h, followed by three washes with PBS. The cells were then incubated with anti-OTUB1 antibody overnight at 4°C, followed by PBS washes. Subsequently, the cells were incubated with Alexa Fluor 647 conjugated goat anti-Rabbit IgG (ABclonal, China) for 1 h and stained with DAPI (Cell Signaling Technology, United States) for 15 min. Imaging of the stained cells was performed using a Zeiss LSM 900 confocal microscope.

IHC were performed according to methods previously described (Zhu et al., 2021; Hu et al., 2020; Yilin Hu et al., 2024). Anti-GPX4, anti-Ki-67, anti-4-HNE antibodies were used. Two experienced pathologists manually scored the staining intensity according to the following criteria: 0 = no staining, 1 = weak staining, 2 = moderate staining, 3 = strong staining. The proportion of positive cells was scored as 1 (0%–10%), 2 (11%–50%), 3 (51%–80%), 4 (81%–100%) (Lu et al., 2020; Zhu et al., 2019; Xian et al., 2014; Jing-Lin Wang et al., 2023). The final IHC score was calculated by multiplying the intensity score by the percentage of positive cells.

Thirty-two 4-week-old male nude mice (purchased from the Model Animal Research Center of Nantong University) were randomly divided into eight groups. Four groups were injected with MKN-45/IR cells (control group, β-elemene group, radiation therapy group, and combination therapy group), and four groups were injected with AGS/IR cells (control group, β-elemene group, radiation therapy group, and combination therapy group), with four mice per group. Each mouse was subcutaneously injected in the right axillary area with 200 μL of MKN-45/IR or AGS/IR cells (2 × 10^6 cells). The growth of the mice was monitored regularly, and once the subcutaneous tumor size reached approximately 150 mm³ (around 15 days), different treatments were administered to the groups. Tumor volume and weight were measured every 3 days. After 36 days, the mice were euthanized, and tumor tissues were harvested, weighed, photographed, recorded, and tumor growth curves were plotted. All animal experiments were approved by the Animal Ethics Committee of Nantong University.

Data were analyzed using GraphPad Prism 8.0 and the R programming language (version 4.2.3). Measurement data with normal distribution were expressed as mean ± SD and statistically analyzed with two-sample t-test. Non-normal distribution measurement data were expressed as median (range) and analyzed by Mann-Whitney U test. For groups of three or more that conform to a normal distribution, one-way ANOVA is utilized, while for those that do not follow a normal distribution, the Kruskal–Wallis test is employed. All statistical analyses were two-sided, and p < 0.05 was used to define statistical significance. All experiments were conducted more than three times.

To investigate whether β-elemene treatment can reverse the radioresistance of GC cells, we first established radioresistant GC cell lines (MKN-45/IR and AGS/IR) (Figure 1A). The radioresistance of MKN-45/IR and AGS/IR cells was assessed using the Cell Counting Kit-8 (CCK-8) assay and clonogenic survival assay. Results showed that MKN-45/IR and AGS/IR cells exhibited higher growth capabilities and survival rates compared to their parental cells (Figures 1B–E). We further determined the mean lethal dose (D0), quasi-threshold dose (Dq), and the sensitization enhancement ratio (SER) of cell survival before and after radiotherapy. The results indicated that the D0 for MKN-45 was 3.130 and the Dq was 2.665, while for MKN-45/IR cells, the D0 was 4.476 and the Dq was 3.052, with an SER of 0.75. For AGS, the D0 was 3.302 and the Dq was 2.376, while for AGS/IR cells, the D0 was 4.441 and the Dq was 2.904, with an SER of 0.74. These findings suggest that the resistant cell lines exhibit stronger radioresistance than their parental cells. To further explore the impact of β-elemene treatment on radioresistance, we validated this effect using a clonogenic formation assay. The results demonstrated that the combination of β-elemene and radiotherapy significantly reduced the colony-forming ability of MKN-45/IR and AGS/IR cells (Figures 1F–I).

To determine the primary molecular mechanisms by which β-elemene combined with radiotherapy inhibits the growth of radioresistant GC cells, cells were first pre-treated with various cell death inhibitors. Cells were treated with the ferroptosis inhibitor Fer-1, the iron chelator DFO, the autophagy inhibitor 3-methyladenine (3-MA), the apoptosis inhibitor Z-VAD, and the necroptosis inhibitor Necrostatin-1. Cell survival levels were assessed 48 h later. The results indicated that both Fer-1 and DFO significantly reduced the cell death induced by the combination of radiotherapy and β-elemene (Figures 2A,B). This suggests that ferroptosis may play a key role in β-elemene’s ability to reverse radiotherapy resistance. Subsequently, to further validate this hypothesis, we measured levels of reactive oxygen species (ROS), iron ions, and lipid peroxidation in MKN-45/IR and AGS/IR cells. Consistent with previous findings, these indicators were significantly elevated after the combined treatment of β-elemene and radiotherapy (Figures 2C–H). To explore the role of ferroptosis in β-elemene’s reversal of radiotherapy resistance in GC cells, we focused on a key ferroptosis regulatory protein, Glutathione Peroxidase 4 (GPX4). GPX4 is the principal enzyme that prevents the accumulation of peroxidized lipids in cells, and its activity is closely linked to ferroptosis (Liu et al., 2024). Therefore, we assessed the expression of GPX4 in radioresistant GC cells treated with β-elemene via qPCR and Western blot. Although there was no significant change in GPX4 mRNA expression levels, the protein levels decreased significantly, suggesting that β-elemene may regulate the expression of GPX4 by affecting protein stability or degradation pathways, thereby further promoting ferroptosis (Figures 2I–L).

To further clarify how β-elemene affects GPX4 protein levels, we initially treated cells with cycloheximide (CHX) to block new protein synthesis, allowing us to observe the dynamics of protein degradation. Experimental results indicated that in the combination of β-elemene and radiotherapy, GPX4 degraded faster than in the radiotherapy-alone group (Figures 3A–F). This suggests that β-elemene may enhance the degradation of GPX4. Considering that the ubiquitin-proteasome pathway and the autophagy-lysosome pathway are two major routes regulating protein degradation, we used the proteasome inhibitor MG132 and the autophagy inhibitor chloroquine (CQ) to further explore the role of these pathways. Treatment with MG132 significantly inhibited the degradation of GPX4, while treatment with CQ had little effect on the degradation of GPX4 (Figures 3G–J). This result indicates that under the conditions of combined β-elemene and radiotherapy, the degradation of GPX4 primarily occurs through the ubiquitin-proteasome pathway. To verify this, we further examined the ubiquitination levels of GPX4. Compared to the radiotherapy-alone group, the ubiquitination levels of GPX4 were significantly increased in the β-elemene and radiotherapy combination group, supporting our hypothesis that β-elemene may act by promoting the ubiquitination and subsequent proteasome-mediated degradation of GPX4 (Figures 3K, L).

In order to delve deeper into the mechanism of GPX4 ubiquitination and degradation, we utilized the IntAct online database to predict potential interacting proteins of GPX4 (Figure 4A). Among these proteins, OTUB1 has been identified as a deubiquitinase that is intricately involved in the ubiquitin-dependent protein degradation pathway. Previous studies have demonstrated that in GC cell lines, OTUB1 can form a complex with GPX4, thereby reducing the ubiquitination levels of GPX4 and inhibiting its proteasomal degradation (Li et al., 2023). We also found this phenomenon in radioresistant GC cell lines. IF assays confirmed the co-localization of OTUB1 with GPX4 (Figures 4B,C), and COIP assays showed that the interaction between GPX4 and OTUB1 was robust in both the control and the radiotherapy-alone groups. However, this interaction was significantly diminished in the group treated with a combination of β-elemene and radiotherapy (Figures 4D, E).

To explore the therapeutic potential of β-elemene combined with radiotherapy in vivo, we established a subcutaneous tumor model in nude mice by injecting MKN-45/IR or AGS/IR cells subcutaneously (Figure 5A). Once tumors developed, treatments were administered either with β-elemene via gavage (0.2 mL per administration), radiotherapy alone (2 Gy per administration), or a combination of both. Results indicated that tumor growth in the nude mice treated with the combination of β-elemene and radiotherapy was significantly inhibited (Figures 5B–G). Immunohistochemical staining of the tumor tissues showed elevated levels of 4-HNE in the combination therapy group, whereas expressions of Ki-67 and GPX4 were lower compared to the control group (Figures 5H, I).