Reagents and cell culture

Cephaeline (CPL) was purchased from MCE Biotechnology (HY-N4118, Shanghai, China). According to the manufacturer’s information, CPL had a purity of ≥ 98%. CPL powder was stored at − 20 °C protected from light and was dissolved in dimethyl sulfoxide (DMSO) to prepare a stock solution. The stock solution was aliquoted and stored at − 20 °C to avoid repeated freeze–thaw cycles. For cell treatment, the CPL stock solution was diluted with complete culture medium to the indicated working concentrations, and the final DMSO concentration was kept below 0.1% in all groups, including the vehicle control. Murine 4T1 and human MDA-MB-231 cell lines were used as highly aggressive breast cancer models. 4T1 and MDA-MB-231 cancer Cells acquired from zqxzbio (Shanghai, China) were grown in RPMI-1640 medium enriched with 10% FBS (A5669701, Gibco, USA), 1% penicillin/streptomycin (G6784, Sigma-Aldrich, MO, USA), and incubated under 5% CO₂ and 37 °C with saturated humidity, and culture media were replaced every other day to support cell viability.

Evaluation of cellular proliferation

The proliferative capacity of cells was measured using CCK-8 kit (BS350B, Biosharp, Hefei, China). In short [25], cells were plated into 96-well plates at a density of 2,000 cells per well and allowed to attach overnight under standard culture conditions. Cells were treated with CPL at concentrations ranging from 2.5 to 320 nM for 72 h to generate dose–response curves and determine IC50 values. Subsequently, 10 μL CCK-8 reagent was dispensed into each well, and the plates were maintained in a dark environment for 1 h. A microplate reader (BioRad, Richmond, CA, USA) was employed to assess absorbance at 450 nm. Cell viability was calculated relative to the untreated control group, and IC50 values were determined by nonlinear regression analysis using a four-parameter logistic dose–response model in GraphPad Prism software. The IC50 values were derived from three independent experiments and are presented with 95% confidence intervals. Based on the IC50 values determined from preliminary CCK-8 assays, representative concentrations of CPL corresponding to sub-IC50, near-IC50, and supra-IC50 levels were selected for subsequent experiments.

Cell transfection

For RNA interference targeting TP53, 3 × 105 cells were distributed into each well of a 6-well plate. Transfection was carried out using 20 μM TP53 siRNA (Ribio, Guangzhou, China) and Transfection kit following the manufacturer’s protocol.

Reactive oxygen species (ROS) assay

ROS levels in CPL-treated 4T1 and MDA-MB-231 cells were assessed using dihydroethidium (DHE) (Sigma, MO, USA). The cells were exposed to 40 μM DHE for 1 h, rinsed with PBS, and the fluorescence signal was measured using a confocal microscope (Olympus FV3000, Tokyo, Japan). DHE fluorescence was detected using the red fluorescence channel at 594 nm, while DAPI nuclear staining was detected using the blue fluorescence channel. The relative DHE fluorescence intensity was quantified using ImageJ software.

Iron and Fe 2+analysis

After 72 h of CPL treatment, intracellular levels of ferric iron (Fe3+) and ferrous iron (Fe2+) were quantified using an iron levels were assessed using the Abcam (ab83366, MA, USA) iron detection kit, with procedures carried out per the recommended protocol.

ELISA

Concentrations of MDA (RK09070) and GSH (RK04298) were measured using commercially available assay kits (Abclonal). Briefly, 2.5 × 104 cells were distributed into each well of 96-well plates, followed by treatment with 80 nM CPL for 72 h. The liquid above the sediment was subsequently transferred to a fresh plate, and the reaction mixture was introduced following the manufacturer’s instructions. For each assay, standard curves were generated using serial dilutions of the provided standards, and MDA and GSH concentrations were calculated based on the corresponding standard curves. Absorbance was measured using a microplate reader (Thermo Fisher Scientific, MA, USA). All measurements were performed in triplicate, and each experiment was independently repeated at least three times.

Western blot

RIPA lysis buffer enriched with protease inhibitors (Biosharp, Hefei, China) was used to extract proteins from cells [26]. Separation of proteins was carried out through SDS-PAGE, subsequently, protein samples were transferred to nitrocellulose membranes (Millipore, MA, USA), followed by blocking in 5% skim milk and incubation with specific primary antibodies at 4 °C overnight. Following extensive washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 2 h at room temperature. Visualization of protein bands was accomplished using an ECL detection system. The primary antibodies utilized included GPX4 (1:1000, 52,455, CST, USA), SLC7A11 (1:1000, ab175186, Abcam), p53 (1:500, A19585, Abclonal), and GAPDH (1:5000, A19056, Abclonal). Band intensities were quantified using ImageJ software. The intensity of each target protein band was normalized to the corresponding GAPDH, and the relative protein expression level was calculated by comparison with the control group. Quantitative analysis was performed based on three independent experiments.

qPCR

RNA extraction was performed with TRIzol (15,596,018, Invitrogen, USA). RNA pellets were resuspended in chloroform, and PrimeScript RT kit (Takara, Japan) was utilized to synthesize cDNA from the extracted RNA. PCR amplification was carried out using SYBR Green Master Mix (Takara, Japan). The 2-ΔΔCT approach was applied to determine relative gene expression, with 18S rRNA serving as the internal reference. Primer sequences were in Table 1.

Cell scratch assay

Cells were seeded in a 6-well plate at a density of 2 × 105 cells per well [27]. After confluence, a linear wound was generated using a 10 μL pipette tip to simulate cell migration. After scratching, the cells were cultured in low-serum medium containing 1% FBS for 24 h to minimize the influence of cell proliferation on wound closure while reducing the potential adverse effects of complete serum deprivation on cell viability and migration. Photographs of the initial wound and cells entering the scratched area were taken at 0 and 24 h using a light microscope (Olympus, Tokyo, Japan).

Colony growth assay

A total of 200 cells per well were seeded into 6-well plates and maintained for 14 days, with the culture medium replaced every 3 days [28]. Colonies were fixed in 4% paraformaldehyde after three PBS rinses and then subjected to 1 h Giemsa staining. Residual dye was washed off and plates air-dried. Colonies with > 20 cells were counted, and relative colony-forming efficiency was adjusted and compared to control group.

Statistical analysis

To ensure data reproducibility and reliability, all measurements were performed in triplicate, and each experiment was independently repeated at least three times. Statistical analyses were performed using GraphPad Prism software. Data are presented as mean ± standard deviation. For comparisons between two groups, an unpaired Student’s t-test was used. For comparisons among more than two groups, one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test was applied. For experiments involving multiple predefined comparisons, the corresponding reference groups are specified in the figure legends. A p value < 0.05 was considered statistically significant.