Human LUAD cell lines (A549, H1299 and PC9) and normal lung epithelial cell line (HBE) were generously provided by Dr. Feng Ma. All cells were routinely cultured in RPMI-1640 (Gibco, Thermo Fisher Scientific, USA), supplemented with 10% fetal bovine serum (FBS, NEST Biotechnology) in culture flasks (SAINING, Biotechnology).

Small interfering RNA (siRNA) targeting BUB1 (si-BUB1), non-targeting control siRNA (si-NC), shRNA targeting BUB1 (sh-BUB1), STAT3 overexpression plasmid (oe-STAT3), GPX4 overexpression plasmid (oe-GPX4) and empty vector were designed and synthesized by GenePharma (Shanghai, China). Cells were seeded into 6-well plates at 5×10⁵ cells/well and transfected with siRNAs or plasmids following the manufacturer’s protocol.

Cells were seeded into 96-well plates at 1×10⁴ cells/well. At 0–5 d post- transfection, 10 μL CCK-8 solution (Life-iLab, Shanghai, China) was added to each well and incubated for 2 h at 37 °C. Absorbance at 450 nm was measured using CYTATION 5 Reader (BioTek, USA).

Confluent cells in 6-well plates were scratched with a sterile 200 μL pipette tip. After washing with PBS twice, cells were cultured in serum-free medium. Wound images were captured at 0 and 24 h using an inverted microscope (Olympus, Tokyo, Japan).

Cells were seeded into 6-well plates at 500 cells/well and cultured for 14 days. Colonies were fixed with 4% paraformaldehyde for 30 min, stained with 0.1% crystal violet (Jiangsu Maige Biotechnology Co., Ltd.) for 15 min.

Transwell chambers (8 μm pore size, Corning, NY, USA) were pre-coated with Matrigel (BD Biosciences, USA) at 37 °C for 30 min. Cells (1×10⁵) in serum-free medium were added to the upper chamber, and medium containing 10% FBS was added to the lower chamber. After 24 h incubation, non-invaded cells on the upper membrane surface were removed with a cotton swab. Invaded cells were fixed, stained and counted in five random fields under a microscope.

Total RNA was extracted from LUAD cells and tumor tissues using TRIzol reagent (Invitrogen, USA). RNA concentration and purity were determined with a NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA). Complementary DNA (cDNA) was synthesized from 1 μg total RNA using the PrimeScript RT Reagent Kit (TaKaRa, Tokyo, Japan). qPCR was conducted on a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) with SYBR Premix Ex Taq II (TaKaRa, Japan). Relative gene expression was calculated via the 2^-ΔΔCt method, with GAPDH as the internal reference.

Total RNA was extracted from si-BUB1 and si-NC transfected A549 cells. RNA sequencing was performed by Novogene (Beijing, China) using the Illumina HiSeq X Ten platform. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were conducted using the DAVID database.

Cells were fixed with 2.5% glutaraldehyde (Beijing Solarbio) overnight, post-fixed with 1% osmium tetroxide for 1 h, dehydrated with gradient ethanol and embedded in epoxy resin. Ultrathin sections (50 nm) were stained with uranyl acetate and lead citrate, then observed under a TEM (Hitachi H-7650, Tokyo, Japan).

Intracellular Fe²⁺ levels were measured using an Iron Assay Kit (Beyotime, Shanghai, China). MDA levels and GSH activity were detected with MDA Assay Kit and GSH Assay Kit (Beyotime, Shanghai, China), respectively. Absorbance was measured with a microplate reader, and concentrations/activities were calculated based on standard curves.

The GPX4​promoter fragment was generated by PCR​and inserted into pGL3-basic. LUAD cells (3 × 10⁴/well, 24-well, triplicate) were transfected with the PGL3-GPX4/promoter reporter plus Renilla, then STAT3 or mock at 24 h. Firefly/Renilla activities were assayed with the Dual-Luciferase reporter kit.

Cell lysates were prepared by resuspending samples in RIPA buffer, and clarified by centrifugation at 13,000 rpm for 20 min at 4 °C. The supernatant and the primary antibody were incubated overnight on a rotary mixer at 4 °C. Then, Protein G magnetic beads (HY-K0202, MCE) were added and the incubation continued. Finally, the binding protein was eluted and analyzed.

The prepared cells were evenly spread on a 12-well plate and 10 μM BODIPY-581/591C11 (D3861, Thermo Fisher Scientific) was added. Incubation was carried out at 4 °C and 5% CO₂ for 30 minutes. After washing twice with PBS, the ROS level was analyzed using a CytoFLEX cell analyzer.

Cells grown on coverslips or paraffin-embedded tumor sections were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 5% BSA. Samples were incubated overnight at 4 °C with primary antibodies against STAT3 and p-STAT3, then with Alexa Fluor 488 or 594-conjugated secondary antibodies (Invitrogen, USA) for 1 h at room temperature. Nuclei were stained with DAPI for 5 min. Images were captured using a confocal laser scanning microscope (Leica TCS SP8, Wetzlar, Germany), and fluorescence intensity was quantified with ImageJ software.

Total protein was extracted using RIPA lysis buffer (Beyotime, Shanghai, China). Protein concentration was measured with a BCA Protein Assay Kit (Thermo Fisher Scientific, USA). Equal amounts of protein (30 μg) were separated by 10% SDS-PAGE and transferred to PVDF membranes (Millipore, Billerica, MA, USA). Membranes were blocked with 5% non-fat milk for 2 h at room temperature, then incubated overnight at 4 °C with primary antibodies against BUB1 (ab195268; 1:800; Abcam; USA), STAT3(ab68153; 1:1000; Abcam; USA), p-STAT3(ab76315; 1:800; Abcam; USA), GPX4 (ab125066; 1:1000; Abcam; USA) and β-actin (#21338; 1:5000; Signalway Antibody). After washing with TBST, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (Bioss USA) for 1 h. Protein bands were visualized using an ECL Chemiluminescence Kit (Millipore, USA) and quantified with ImageJ software (v1.8.0).

BALB/c nude mice (4–6 weeks old) were purchased from Guangdong Provincial Medical Laboratory Animal Center. A549 cells transfected with sh-BUB1 or sh-NC (1×10⁷ cells/mouse) were subcutaneously injected into the right flank of nude mice. Tumor volume was measured every 3 days using a caliper (volume = (length × width²)/2). Four weeks post-injection, mice were euthanized by dislocating their cervical vertebrae under isoflurane anesthesia (4% isoflurane inhalation, followed by 2% isoflurane inhalation for anesthesia maintenance), and their tissue samples were then collected. Tumors were excised, weighed, and stored at −80 °C for further analysis. Animal experiments were approved by the Ethical Committee of the Laboratory Animal Center of Jinan University, Guangzhou, China.

Data are presented as mean ± standard deviation (SD) from at least three independent experiments. Statistical analyses were performed using SPSS 22.0 software (IBM, Armonk, NY, USA) and GraphPad Prism 8.0 software (GraphPad Software, San Diego, CA, USA). Differences between groups were analyzed by Student’s t-test or one-way ANOVA followed by Tukey’s post-hoc test. P < 0.05 was considered statistically significant.

Bioinformatics analysis using the TCGA database revealed significantly elevated BUB1 expression in LUAD tumor tissues relative to matched normal tissues (Figure 1A). This finding was further validated at the transcript level, as RT-qPCR demonstrated marked upregulation of BUB1 mRNA in LUAD cell lines (A549, H1299 and PC9) compared to the normal bronchial epithelial cell line HBE (Figure 1B). Consistently, western blot analysis confirmed a substantial increase in BUB1 protein abundance in these LUAD cell lines relative to HBE cells (Figure 1C). The association analysis between BUB1 expression and the clinical stage of LUAD showed that BUB1 expression was positively correlated with the malignancy degree of LUAD, and the BUB1 expression was higher in patients with advanced stage (Stage III-IV) (Figures 1D). Survival analysis based on TCGA data further supported the clinical relevance of BUB1 dysregulation. Kaplan-Meier curves indicated that high BUB1 expression correlated with markedly reduced disease-free survival (DFS) and overall survival (OS) in LUAD patients (Figures 1E, F). Collectively, these data demonstrate that BUB1 is aberrantly overexpressed in LUAD and that elevated expression levels are strongly predictive of poor clinical outcomes.

To investigate the impact of BUB1 on LUAD cells, we systematically silenced BUB1 expression in A549 and H1299 cell lines using three distinct siRNA constructs (si-BUB1#1-3). RT-qPCR analysis confirmed significant BUB1 mRNA reduction in all siRNA groups compared to scramble controls (si-NC), with si-BUB1#1 demonstrating optimal suppression (Figures 2A, B). This siRNA was subsequently selected for functional studies based on its superior efficacy. In addition, we further found that BUB1 knockdown did not induce significant cytotoxicity in the normal HBE cells compared to the control group (Supplementary Figure S1).

As shown in Figures 2C, D, in both A549 and H1299 cells, the optical density (OD) values reflecting cell viability were notably lower in si-BUB1 groups than in si-NC groups. These data indicate that BUB1 knockdown suppresses the proliferative capacity of tumor cells. We further investigated the impact of BUB1 on cell migration via the wound healing assay. As depicted in Figures 2E, F, the wound closure in si-BUB1 groups was remarkably slower than that in si-NC groups at 24 hours after scratch wounding in both cell lines. Demonstrating that BUB1 silencing markedly suppressed the migratory capacity of A549 and H1299 cells. Furthermore, both colony formation and cell invasion assays indicated that BUB1 knockdown substantially inhibited the proliferative and invasive capabilities of LUAD cells relative to control cells (Figures 2G–N). These comprehensive findings establish BUB1 as a critical regulator of LUAD cell malignancy, with its silencing effectively targeting three hallmark cancer processes-uncontrolled proliferation, collective migration, and invasive potential.

Subsequently, RNA sequencing was conducted on BUB1-knockdown LUAD cells. This analysis identified a total of 293 differentially expressed genes (DEGs), with 174 being upregulated and 119 downregulated (Figures 3A). Enrichment analysis further indicated that ferroptosis was significantly enriched (Figure 3B). To exclude non-ferroptotic mechanisms, we inhibited necroptosis with Nec-1, autophagy with CQ, and apoptosis with Z-VAD in BUB1-silenced A549/H1299 cells. These treatments did not substantially reduce cell death, supporting a primary role for ferroptosis (Supplementary Figure S2). Based on these observations, we proposed the hypothesis that ferroptosis mediates the regulatory effect of BUB1 on LUAD cell progression.

To explore this, TEM was utilized to assess cellular ultrastructural changes. The BUB1-knockdown group displayed distinct ferroptosis-characteristic morphological features, including notably shrunken mitochondria, increased mitochondrial membrane density, and reduced mitochondrial cristae (Figures 3C, D). For functional validation, the ferroptosis inducer RSL3 and inhibitor Ferrostatin-1 (Fer-1) were used as experimental modulators, and cells were divided into four groups: ctrl, RSL3 treatment (RSL3), BUB1-knockdown (si-BUB1) and BUB1-knockdown plus Fer-1 (si-BUB1ckdownarac Ferroptosis-associated protein levels (FTH1, SLC7A11 and COX2) were assessed, and the results demonstrated that both RSL3 treatment and BUB1 silencing significantly decreased FTH1 and SLC7A11 expression, accompanied by an upregulation of COX2 in A549 and H1299 cells compared to the control group. Notably, these alterations were reversed by Fer-1 co-treatment (Figures 3E–G). Since intracellular iron accumulation is a hallmark of ferroptosis, we next quantified intracellular Fe²⁺ levels across all groups. Results showed that both RSL3 treatment and BUB1 silencing significantly increased intracellular iron concentrations in A549 and H1299 cells, while Fer-1 administration markedly attenuated this iron accumulation (Figure 3H). Malondialdehyde (MDA) and glutathione (GSH) are critical regulators of cellular redox homeostasis, and their levels undergo characteristic changes during ferroptosis. Compared to the control group, MDA levels were significantly elevated in the RSL3 and BUB1-silenced groups, and this increase was effectively abrogated by Fer-1 treatment (Figure 3I). Conversely, GSH activity was significantly reduced in both the RSL3 and BUB1-silenced groups, and this inhibitory effect was reversed by Fer-1 co-treatment (Figure 3J). The level of lipid-ROS was obviously reduced with in both the RSL3 and BUB1-silenced treatment, but restored by Fer-1 co-treatment (Figures 3K, L). Collectively, these findings confirm our hypothesis that BUB1 knockdown promotes ferroptosis in LUAD cells, mimicking the effect of the ferroptosis inducer RSL3.

To elucidate the molecular mechanism by which BUB1 knockdown may contribute to the induction of ferroptosis, we focused on glutathione peroxidase 4 (GPX4), a critical enzyme maintaining lipid bilayer integrity through lipid peroxidation detoxification (15). Western blot analysis revealed significant downregulation of GPX4 protein levels following BUB1 knockdown in both A549 and H1299 cells (Figures 4A, B), accompanied by reduced GPX4 mRNA expression (Figure 4E). Given that previous studies have reported GPX4 to be regulated by multiple molecular factors, including STAT3, we hypothesized that BUB1 might influence GPX4 function via modulation of STAT3. To test this, we conducted additional western blot experiment and observed that BUB1 silencing potently suppressed STAT3 protein expression and its phosphorylation status (Figures 4A–D). Furthermore, BUB1 suppressed the transcription of STAT3 (Figure 4F) and reduced its levels in both the cytoplasm and nucleus (Figures 4G, H), as shown by RT-qPCR and immunofluorescence staining. To dissect the effect of STAT3 on GPX4, dual luciferase and coimmunoprecipitation (CO-IP) were carried out. A luciferase assay reveled that STAT3 activated the GPX4 promoter in A549/H1299 cells, and BUB1 knockout negated this activation (Figures 4I, J). Co-IP was designed to confirm a high degree of integration between STAT3 and GPX4 (Figure 4K). In summary, these findings suggest that BUB1 knockdown suppresses GPX4 transcription by downregulating STAT3 expression, thereby promoting ferroptosis in LUAD cells.

To investigate whether BUB1 regulates the progression of LUAD through the STAT3/GPX4 axis, we overexpressed STAT3 in A549 and H1299 cells and subsequently subjected them to BUB1 knockdown. An analysis of GPX4, STAT3 and p-STAT3 levels showed that the BUB1 silencing-induced reduction in GPX4, STAT3 and p-STAT3expression was partially rescued by STAT3 overexpression (Figures 5A–D). The same intervention was then used to observe the malignant phenotypes of the two LUAD cells. The results indicated that STAT3 overexpression partially mitigated the migratory capacity induced by BUB1 silencing in both cell lines, as shown in Figures 5E, F. Likewise, STAT3 overexpression partially reversed the decrease in clonogenic potential (Figures 5G, H) and invasive ability (Figures 5I, J) caused by BUB1 knockdown in LUAD cells.

To investigate the involvement of the STAT3/GPX4 axis in ferroptosis, we overexpressed STAT3 and GPX4 in LUAD cells and evaluated its impact on ferroptosis-related responses. Compared with the si-BUB1-treated group, both STAT3 overexpression and GPX4 overexpression could counteract the effects of BUB1 silencing on the expression of key ferroptosis-associated genes, including FTH1 (Figure 6A, Supplementary Figure S3A), SLC7A11(Figure 6B, Supplementary Figure S3B), and COX2 (Figure 6C, Supplementary Figure S3C). We further assessed functional ferroptosis markers, including Fe²⁺, MDA and GSH activity. BUB1 knockdown led to a significant increase in Fe²⁺ levels, an effect that was markedly attenuated upon both STAT3 and GPX4 overexpression (Figure 6D, Supplementary Figure S3D). Similarly, the BUB1 silencing-induced elevation in MDA (Figure 6E, Supplementary Figure S3E) and reduction in GSH activity (Figure 6F, Supplementary Figure S3F) were both reversed by STAT3 and GPX4 overexpression. Collectively, these findings indicate that BUB1 silencing promotes ferroptosis in LUAD cells through modulation of the STAT3/GPX4 signaling pathway.

To further dissect the role of BUB1 in a xenograft tumor mouse model, we found that BUB1 knockdown significantly reduced tumor volume (Figures 7A, B) compared to the control group, and Fer-1 could abolished this effect. Subsequent molecular analysis revealed that the proliferation marker Ki67, GPX4 and STAT3 were all lower in the BUB1 knockdown group than in the control (Figures 7C–G). Additionally, analysis of ferroptosis-related gene expression demonstrated that BUB1 knockdown led to downregulated FTH1 (Figure 7H) and SLC7A11 (Figure 7I) expression while increasing COX2 (Figure 7J) expression, these alterations were reversed by Fer-1 co-treatment, which were consistent with our in vitro results. Analysis of functional ferroptosis markers activity reveled that BUB1 silencing-induced elevation in Fe²⁺ and MDA levels (Figures 7K, L) while reduction in GSH activity (Figure 7M), these alterations were reversed by Fer-1 co-treatment, which were also consistent with our in vitro results. Collectively, these data confirm that BUB1 silencing not only suppresses LUAD growth but also modulates the expression of genes involved in ferroptosis.