ATF4/SCO2 Inhibits Ferroptosis in Lung Cancer Through Activating the MAPK Signaling Pathway Authors: Liang Yu, Sheng Zhang, Jinxi He, Bo Ma, Xuyang Song, Rui Qiu, Bo Yu Published: 2026 ABSTRACT ABSTRACTFerroptosis is a specific cell death form resulting from the iron‐dependent lipid peroxidation, exhibiting an essential role in cancer progression. The purpose of this study was to explore the underlying mechanism of ferroptosis in lung cancer. The mRNA and protein expression levels were examined via real‐time quantitative PCR and Western blotting. Cell malignant behaviors were assessed using cell counting kit‐8 (CCK‐8) assay, transwell assay, and flow cytometry. Ferroptosis was determined using the indicators lipid reactive oxygen species (ROS), glutathione (GSH), and Fe2+. Dual‐luciferase reporter assay and chromatin immunoprecipitation assay were performed for binding analysis between activating transcription factor 4 (ATF4) and synthesis of cytochrome c oxidase 2 (SCO2). Associated proteins in mitogen‐activated protein kinases (MAPK) signaling pathway were measured by Western blotting. A mouse model was established by implanting lung cancer cells for in vivo research. Lung cancer samples and cells exhibited high expression of SCO2. SCO2 downregulation restrained lung cancer cell growth by inducing ferroptosis. ATF4 was an up‐regulated gene in lung cancer, and it elevated SCO2 expression by binding to the promoter of SCO2. ATF4 knockdown repressed cell proliferation and migration while facilitating apoptosis and ferroptosis by down‐regulating SCO2. Inhibition of ATF4/SCO2 inactivated the MAPK signaling pathway. ATF4‐mediated SCO2 promoted lung tumorigenesis in vivo. This study manifested that ATF4 controlled SCO2 to suppress ferroptosis by activating the MAPK pathway, thus contributing to lung cancer development.Keywords: activating transcription factor 4, ferroptosis, lung cancer, mitogen‐activated protein kinase pathway, synthesis of cytochrome c oxidase 2 Introduction Lung cancer remains the world's primary cause of cancer death, probably due to the prevalence of smoking that accounts for approximately 80%–90% of lung cancer [1, 2]. Although various treatments have been advanced in lung cancer, the survival and prognosis are still unsatisfactory, especially in metastatic patients [3]. The physiological mechanism of lung cancer is complicated, and studying the potential mechanism is of great significance to improve the treatment. Human cancers, including lung cancer, are tightly associated with the regulated cell death (RCD), such as apoptosis, ferroptosis, pyroptosis, and so on [4]. Targeting cell death pathways is pivotal to enhancing cancer therapeutic effectiveness [5]. Ferroptosis is a non‐apoptotic RCD type dependent on iron, and it is mainly evoked by redox imbalance between oxidant production and antioxidant activity [6, 7]. Ferroptosis is closely related to different stages of lung cancer, including origination, proliferation, and development [8]. Induction of ferroptosis can be used to block lung cancer progression [9]. Exploring the underlying mechanisms of ferroptosis is quite important for the application of targeting ferroptosis in lung cancer. Synthesis of cytochrome c oxidase 2 (SCO2) has exhibited increased expression in lung adenocarcinoma (LUAD) [10], and SCO2 is significantly associated with LUAD progression to serve as a potential marker for prognosis [11]. However, the functions of SCO2 in lung cancer cell malignant behaviors and ferroptosis have not been characterized. Activating transcription factor 4 (ATF4) is a stress‐induced transcription factor with crucial regulation in ferroptosis, either promoting or inhibiting [12]. A previous study demonstrated that ATF4 could activate 6‐pyruvoyl‐tetrahydropterin synthase (PTS) that facilitated lung cancer progression [13]. The relation between ATF4 and SCO2 in lung cancer remains to be further investigated. Mitogen‐activated protein kinase (MAPK), a ubiquitous signal transduction pathway, is capable of regulating human diseases, including cancers [14]. Inhibition of the MAPK pathway is responsible for the anti‐tumor role of dictamnine in lung cancer [15]. Protein phosphatase 4 catalytic subunit has been shown to promote lung cancer tumorigenesis through the MAPK signaling activation [16]. Whether the functions of ATF4 and SCO2 in lung cancer are dependent upon the MAPK pathway is unknown. Herein, this study tested the hypothesis that ATF4 was implicated in the signal activation of the MAPK pathway by mediating SCO2 in lung cancer. The effects of ATF4 and SCO2 on ferroptosis were studied, aiming to reveal the ferroptosis mechanism in lung cancer. Materials and Methods Sample Collection A total of 52 patients diagnosed with lung cancer at General Hospital of Ningxia Medical University have been recruited in the current research, based on the signing of the informed consent. These patients did not receive any other specific treatment before surgery. Lung tumor tissues (n = 52, LUAD) were resected, and tumor‐adjacent tissues (n = 52) were used as the normal controls. All samples were saved in liquid nitrogen. This study was approved by the Ethics Committee of General Hospital of Ningxia Medical University. Cell Culture and Transfection Human lung cancer cell lines (A549, NCI‐H1299, and PC‐9) and normal human embryonic lung fibroblast MRC‐5 were acquired from BioVector NTCC Preservation Center (Beijing, China). DMEM with the supplement of 10% fetal bovine serum and 1% penicillin/streptomycin (Sigma, St. Louis, MO, USA) was employed for cell culture. All cells were free from mycoplasma contamination and grown in a humidified atmosphere with 5% CO2 at 37°C. A549 and NCI‐H1299 cells were cultured overnight and transfected with the indicated vectors through Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA). The used vectors were from RIBOBIO (Guangzhou, China), including short hairpin RNA (shRNA) vectors (shSCO2, shATF4) and an overexpression vector oeSCO2, with shNC and oeNC as control groups. Real‐Time Quantitative PCR (RT‐qPCR) Assay Tissues and cells were lysed with TRIzol reagent (Invitrogen), and then total RNA was reverse transcribed into cDNA by using PrimeScript FAST RT reagent Kit with gDNA Eraser (Takara, Beijing, China). The quantitative detection was achieved by using TB Green Premix Ex Taq II reagent (Takara) and the specific primers (Table 1). SCO2 or ATF4 mRNA expression was quantified via the 2‐∆∆Ct method, with β‐actin used as an internal reference. Table 1Primer sequences used for RT‐qPCR.NamePrimers for PCR (5′−3′)ATF4ForwardGTTTTGGATTGGTGGGGTGCReverseTCTTGGTTCCTGCCACGTTTSCO2ForwardGGCGTGGATCCCGGTCTGReverseAAAAGCCAGGACCTCAGATGCβ‐actinForwardCTTCGCGGGCGACGATReverseCCACATAGGAATCCTTCTGACC Primer sequences used for RT‐qPCR. Western Blotting Firstly, total proteins from tissues or cells were extracted through RIPA Buffer (Abcam, Cambridge, UK). Next, an equal amount of proteins per sample was isolated on gels using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS‐PAGE). Subsequently, proteins moved onto polyvinylidene fluoride membranes (Thermo Fisher, Waltham, MA, USA) were blocked in Blocking Solution (Abcam). Then, membranes and the primary antibody were co‐incubated at 4°C overnight. The used primary antibodies were as follows: anti‐SCO2 (Antibodies Online, UK; ABIN7263007), anti‐ATF4 (MedChemExpress, Rahway, NJ, USA; HY‐P80553), anti‐p38 (Abcam, ab170099), anti‐phospho‐p38 (anti‐p‐p38, Abcam, ab195049) anti‐ Extracellular signal‐regulated kinase 1 (ERK1, Abcam, ab32537), anti‐p‐ERK1 (Abcam, ab131438), Glutathione Peroxidase 4 (GPX4, Abcam, ab125066), long‐chain acyl‐CoA synthetase (ACSL4, Abcam, ab155282), and anti‐β‐actin (Abcam, ab213262). Finally, the secondary antibody (Abcam, ab205718) was incubated at room temperature for 1 h, followed by the examination of blots using ECL reagent (Abcam). ImageJ was used for the analysis of the grey value. Cell Proliferation Assay A549 and NCI‐H1299 cells were transfected with indicated group for 0, 1, 2, and 3 d, then cell counting kit‐8 (CCK‐8) solution (Sangon, Shanghai, China) was added with 10 μL/well. The optical value (OD) values were measured at 450 nm under a microplate reader, and the proliferation curve was graphed. Cell Migration Assay Transwell chamber (Corning Inc., Corning, NY, USA) was employed for detection of migration ability. A549 and NCI‐H1299 cells were re‐suspended in serum‐free medium, and then 1 × 104 cells were inoculated into the top chamber. Meanwhile, normal cell medium was pipetted into the bottom chamber. After incubation at 37°C for 24 h, migrated cells following fixation with 4% paraformaldehyde (Sangon) and staining with 0.1% crystal violet (Sangon) were examined by an inverted microscope (Olympus). Cell Apoptosis Assay Apoptotic cells from transfected A549 and NCI‐H1299 were determined via the staining of Annexin V‐FITC and Propidium Iodide (PI) through Annexin V‐FITC/PI Apoptosis Detection Kit (Solarbio, Beijing, China). The operating steps were in accordance with the provided specification of the manufacturer. By using the flow cytometer (BD Biosciences, San Diego, CA, USA), cells stained by Annexin V+/PI−and Annexin V+/PI+ were defined as the apoptotic cells (%). Cell Ferroptosis Examination Lung cancer cells were planted into a 96‐well plate with 2 × 105 cells/well and subjected to different transfections, then lipid reactive oxygen species (ROS) was measured via ROS Assay Kit (Beyotime, Shanghai, China) and a flow cytometer. Glutathione (GSH) and Fe2+ were tested through GSH Assay Kit (Beyotime) and Iron Assay Kit (Abcam) via the microplate reader. In addition, A549 and NCI‐H1299 cells with shSCO2 transfection were treated with ferroptosis inhibitor ferrostatin‐1 (Fer‐1; Sigma), followed by CCK‐8 examination. Dual‐Luciferase Reporter Assay The wild‐type (WT) plasmid SCO2 WT was obtained by cloning the SCO2 promoter sequence into the basic luciferase gene vector pGL4 (Promega, Madison, WI, USA). In addition, the mutant‐type (MUT) plasmid SCO2 MUT was constructed after the SCO2 promoter sequence was mutated. A549 and NCI‐H1299 cells were respectively co‐transfected with shNC/shATF4 and SCO2 WT/SCO2 MUT for a duration of 48 h. Eventually, relative luciferase activity was tested by using dual‐luciferase reporter detection kit (Promega) as per the user's manual. Chromatin Immunoprecipitation (ChiP) Assay The interaction between the SCO2 promoter and ATF4 was analyzed via ChIP Assay Kit (Beyotime). Briefly, samples were incubated with protein A + G agarose and ATF4/IgG antibody after lysing and ultrasonic processing of A549 and NCI‐H1299 cells. Then, ChIP product was harvested for examination of SCO2 promoter enrichment by RT‐qPCR. Tumor Xenograft Assay Five‐6‐week‐old BALB/c male nude mice (SPF biotech, Beijing, China) were subcutaneously injected with 2 × 106 transfected NCI‐H1299 cells with five mice in each group (shNC+oeNC, shATF4+oeNC, shATF4+oeSCO2). Tumor growth was observed every 7 d through the calculation of tumor volume (length × width2 × 0.5). After xenograft for 28 d, mice were euthanized using CO2 inhalation. The tumors were isolated from mice and weighed, followed by ATF4 and SCO2 protein detection via immunohistochemistry (IHC). The pathological results were evaluated by two members of staff blinded to the study. The present animal research was conducted based on the ratification of the Animal Ethical Committee of General Hospital of Ningxia Medical University. Statistical Analysis Cell experiments were independently carried out three times. Data were indicated as the mean ± SD. Linear correlation between mRNA expression of ATF4 and SCO2 in lung cancer samples was analyzed using Pearson's correlation coefficient and GraphPad Prism software. The survival curves were calculated by the Kaplan‐Meier method and compared with the log‐rank test. The statistical analysis was performed by SPSS (version 19.0, IBM, Armonk, NY, USA), and difference comparison was implemented via Student's t‐test for two groups and ANOVA followed by Tukey's test for more than two groups. p < 0.05 indicated a statistically significant difference. Results SCO2 Was Aberrantly Up‐Regulated in Lung Cancer Online GSCA (https://guolab.wchscu.cn/GSCA/) and TNMPLOT (https://tnmplot.com/analysis/) were used to ensure the expression of SCO2 in lung cancer. As exhibited in Figure 1A,B, SCO2 was highly expressed in LUAD. Ualcan (https://ualcan.path.uab.edu/index.html) also manifested the upregulation of SCO2 in LUAD from TCGA samples (Figure 1C). In the collected lung cancer samples, SCO2 mRNA levels were much higher than in normal samples (Figure 1D). Based on the median value of SCO2, 52 lung cancer patients were divided into the high expression group (n = 26) and the low expression group (n = 26). Then, Clinicopathological characteristics showed that high expression of SCO2 was associated with tumor size, TNM stage, and lymphatic metastasis (Table 2). Moreover, Kaplan‐Meier analysis discovered that lung cancer patients with high SCO2 levels had poor overall survival (p < 0.05) (Figure 1E). Meanwhile, the results of western blot assay displayed that SCO2 protein level was apparently up‐regulated in lung tumor tissues compared to normal controls (Figure 1F). SCO2 was significantly overexpressed in A549, NCI‐H1299, and PC‐9 cell lines by comparison with the normal MRC‐5 cell line (Figure 1G,H). Aberrant expression of SCO2 implied its potential role in lung cancer progression. SCO2 was aberrantly up‐regulated in lung cancer. (A−B) SCO2 expression in lung cancer according to online databases GSCA (A) and TNMPLOT (B) was shown. (C) SCO2 level in LUAD from TCGA samples through Ualcan analysis. (D) RT‐qPCR was carried out for mRNA detection of SCO2 in lung cancer samples. (E) Kaplan‐Meier analysis of the correlation between SCO2 expression and overall survival in 52 lung cancer patients. (F) Western blotting was performed for protein detection of SCO2 in lung cancer samples. (G, H) RT‐qPCR and Western blotting were carried out for mRNA and protein detection of SCO2 in normal human embryonic lung fibroblast MRC‐5 and human lung cancer cell lines (A549 and NCI‐H1299). ***p < 0.001, n = 3. LUAD, lung adenocarcinoma; RT‐qPCR, Real‐time quantitative PCR; SCO2, synthesis of cytochrome c oxidase 2. Table 2Association analysis between SCO2 and clinicopathological characteristics of lung cancer patients (n = 52).ParameterNumber of patients (n = 52)High expression (n = 26)Low expression (n = 26) p valueAges (years)< 602913160.402≥ 60231310Gender0.095Male24159Female281117Tumor size0.000* < 3 cm27720≥ 3 cm25196TNM stage0.012* Ⅰ + Ⅱ23716Ⅲ + Ⅳ291910Lymphatic metastasis0.005* Yes302010No22616* p < 0.05 represents a significant difference. Association analysis between SCO2 and clinicopathological characteristics of lung cancer patients (n = 52). p < 0.05 represents a significant difference. SCO2 Interference Impeded Proliferation and Migration But Promoted Apoptosis and Ferroptosis of Lung Cancer Cells A549 and NCI‐H1299 cells expressing higher SCO2 were applied for further functional research. In A549 and NCI‐H1299 cells, SCO2 protein expression was relatively lower in the shSCO2 group than in the shNC group (Figure 2A). After the successful knockdown of SCO2, a series of cell experiments was subsequently performed. Results from CCK‐8 assay and transwell assay demonstrated that the silence of SCO2 curbed cell abilities of proliferation (Figure 2B) and migration (Figure 2C). Apoptotic cells by flow cytometry were increased in shSCO2‐transfected cells relative to shNC transfection (Figure 2D). Ferroptosis is marked by iron overload and ROS accumulation, and a high level of GSH is essential for clearing excessive ROS in cancer cells [17]. In A549 and NCI‐H1299 cells, reduced expression of SCO2 elevated lipid ROS (Figure 2E), decreased GSH (Figure 2F), and increased Fe2+ (Figure 2G), which suggested that SCO2 downregulation promoted ferroptosis in lung cancer. To affirm whether SCO2 regulates lung cancer cell progression via the ferroptosis pathway, SCO2‐silenced cells were treated with ferroptosis inhibitor Fer‐1. As the data of CCK‐8 indicated in Figure 2H, the OD value reduced by shSCO2 was rescued following Fer‐1 treatment, unraveling that SCO2 knockdown inhibited cell growth through inducing ferroptosis. Hence, SCO2 functioned as an oncogene to facilitate lung cancer cell development by inhibiting ferroptosis. SCO2 interference impeded proliferation and migration but promoted apoptosis and ferroptosis of lung cancer cells. (A) SCO2 protein examination was performed in A549 and NCI‐H1299 cells transfected with shNC and shSCO2. (B‐D) CCK‐8 assay for proliferation (B), transwell assay for migration (C) and flow cytometry for apoptosis (D) were conducted in SCO2‐silenced A549 and NCI‐H1299 cells. (E–G) Lipid ROS (E), GSH (F) and Fe2+ (G) were detected by kits to assess ferroptosis in A549 and NCI‐H1299 cells with SCO2 knockdown. (H) OD value was measured by CCK‐8 assay after cell treatment of shNC, shSCO2, shSCO2+Fer‐1. ***p < 0.001, n = 3. SCO2, synthesis of cytochrome c oxidase 2; CCK‐8, cell counting kit‐8; Lipid ROS, lipid reactive oxygen species; GSH, glutathione; Fer‐1, ferrostatin‐1. ATF4 Was Highly Expressed in Lung Cancer Next, a positive relationship (R = 0.19, p < 0.05) between SCO2 and ATF4 was predicted in lung cancer through the TNMPLOT website (Figure 3A). In the TIMER website (https://cistrome.shinyapps.io/timer/), SCO2 was also positively correlated to ATF4 in LUAD (cor = 0.197, P = 7.01e‐06) (Figure 3B). As for ATF4 expression pattern, it was found to be up‐regulated in lung tumor tissues compared with normal controls after implementation of RT‐qPCR (Figure 3C). According to the median value of ATF4, 52 lung cancer patients were divided into a high‐expression group (n = 26) and a low‐expression group (n = 26). Then, clinicopathological characteristics indicated that high expression of ATF4 was significantly related to tumor size, TNM stage, and lymphatic metastasis (Table 3). Then, Kaplan‐Meier analysis revealed that lung cancer patients with high ATF4 levels had poor overall survival (p < 0.05) (Figure 3D). Meanwhile, western blot assay exhibited that the ATF4 protein level was obviously increased in lung tumor tissues relative to normal controls (Figure 3E). The same upregulation of ATF4 mRNA and protein was detected in A549 and NCI‐H1299 cells, when compared to MRC‐5 cells (Figure 3F,G). ATF4 was also overexpressed in lung cancer and exhibited a positive correlation to SCO2. ATF4 was highly expressed in lung cancer. (A, B) The positive correlation between SCO2 and ATF4 was predicted by TNMPLOT (A) and TIMER (B). (C) ATF4 mRNA levels in lung cancer tissues were detected using RT‐qPCR. (D) Kaplan–Meier analysis of the correlation between ATF4 expression and overall survival in 52 lung cancer patients. (E) ATF4 protein levels in lung cancer tissues were measured using Western blotting. (F, G) ATF4 mRNA and protein levels in normal human embryonic lung fibroblast MRC‐5 and human lung cancer cell lines (A549 and NCI‐H1299) were detected using RT‐qPCR and Western blotting. **p < 0.01, ***p < 0.001, n = 3. ATF4, activating transcription factor 4; RT‐qPCR, Real‐time quantitative PCR; SCO2, synthesis of cytochrome c oxidase 2. Table 3Association analysis between ATF4 and clinicopathological characteristics of lung cancer patients (n = 52).ParameterNumber of patients (n = 52)High expression (n = 26)Low expression (n = 26) P valueAges (years)< 602212100.575≥ 60301416Gender0.165Male271611Female251015Tumor size0.005* < 3 cm24717≥ 3 cm28199TNM stage0.001* Ⅰ + Ⅱ24618Ⅲ + Ⅳ28208Lymphatic metastasis0.000* Yes27225No25421* p < 0.05 represents a significant difference. Association analysis between ATF4 and clinicopathological characteristics of lung cancer patients (n = 52). p < 0.05 represents a significant difference. ATF4 Targeted the Promoter of SCO2 to Increase SCO2 Expression Then, the influence of ATF4 on SCO2 was further determined using Western blotting. Transfection of shATF4 (relative to shNC) significantly knocked down the protein expression of ATF4 (Figure 4A) and reduced SCO2 protein expression in A549 and NCI‐H1299 cells (Figure 4B). Thus, ATF4 could upregulate the expression of SCO2 in lung cancer cells. The binding site between ATF4 and the promoter of SCO2 was predicted by Jaspar (http://jaspar.genereg.net/), as indicated in Figure 4C. Furthermore, the sequence of SCO2 WT (CGTTACGTCATAG) was mutated to SCO2 MUT (TACCGTACTGCGA) for dual‐luciferase reporter assay (Figure 4D). The results revealed the suppression of luciferase activity after co‐transfection with SCO2 WT and shATF4 (relative to the SCO2 WT+shNC group), but there was no significant difference in the SCO2 MUT+shATF4 group compared to the SCO2 MUT+shNC group (Figure 4E). Moreover, ChIP result demonstrated the high enrichment of the SCO2 promoter by anti‐ATF4, indicating the interaction between ATF4 and SCO2 promoter (Figure 4F). Linear analysis also displayed a positive relation (R = 0.7134, p < 0.001) between expression of ATF4 and SCO2 (Figure 4G). All in all, ATF4 enhanced SCO2 expression by binding to the SCO2 promoter. ATF4 targeted the promoter of SCO2 to increase SCO2 expression. (A) Knockdown effect of shATF4 on ATF4 protein expression was analyzed via Western blotting. (B) SCO2 protein level was measured by Western blotting in A549 and NCI‐H1299 cells with the silencing of ATF4. (C) The binding site between ATF4 and SCO2 promoter was predicted by Jaspar. (D, E) The binding site in SCO2 promoter was mutated and dual‐luciferase reporter assay was employed to validate the interaction between ATF4 and SCO2. (F) ChIP assay was used for detecting the binding between ATF4 and SCO2 promoter. (G) The relationship between expression of ATF4 and SCO2 was analyzed through Pearson's correlation coefficient (R = 0.7134, p < 0.001). ***p < 0.001, ns: no significant difference, n = 3. ATF4, activating transcription factor 4; ChIP, Chromatin immunoprecipitation; SCO2, synthesis of cytochrome c oxidase 2. ATF4 Expedited Cell Proliferation/migration and Repressed Apoptosis/Ferroptosis via Targeting SCO2 In combination with the regulation of ATF4 on SCO2 expression, the ATF4/SCO2 axis in lung cancer progression was investigated. In Western blotting, shATF4‐mediated expression suppression of SCO2 protein was largely reversed following oeSCO2 transfection, showing the effective SCO2 overexpression by oeSCO2 (Figure 5A). Cell proliferation (Figure 5B) and migration (Figure 5C) reduced by the knockdown of ATF4 were recovered when SCO2 was overexpressed. Increased apoptotic cells in ATF4‐silenced A549 and NCI‐H1299 cells were dramatically decreased with the exogenous upregulation of SCO2 (Figure 5D). In addition, shATF4‐induced ferroptosis events, including lipid ROS increase (Figure 5E), GSH reduction (Figure 5F), and Fe2+ aggregation (Figure 5G), were obviously alleviated after manipulative expression increase of SCO2. In addition, the effects of ATF4 knockdown and SCO2 upregulation on the expression of ferroptosis‐related genes (GPX4 and ACSL4) were further analyzed using western blotting. Results displayed that ATF4 deficiency could obviously repress GPX4 protein level and promote ACSL4 expression in A549 and NCI‐H1299 cells, which were partly abolished by SCO2 overexpression (Figure 5H). As expected, the ATF4/SCO2 axis contributed to lung cancer cell progression and impeded ferroptosis. ATF4 expedited cell proliferation/migration and repressed apoptosis/ferroptosis via targeting SCO2. (A) Western blotting for protein detection of SCO2 in A549 and NCI‐H1299 cells transfected with shNC+oeNC, shATF4+oeNC, or shATF4+oeSCO2. (B–D) Cell proliferation (B), migration (C), and apoptosis (D) were estimated using CCK‐8 assay, transwell assay and flow cytometry in transfected cells. (E–G) Ferroptosis was evaluated through lipid ROS (E), GSH (F) and Fe2+ (G) via commercial kits. (H) Western blotting for protein detection of GPX4 and ACSL4 in A549 and NCI‐H1299 cells transfected with shNC+oeNC, shATF4+oeNC, or shATF4+oeSCO2. ***p < 0.001, * denotes shNC+oeNC versus shATF4+oeNC; # p < 0.05, ## p < 0.01, ### p < 0.001, # denotes shATF4+oeNC versus shATF4+oeSCO2; n = 3. ACSL4; SCO2, synthesis of cytochrome c oxidase 2; ATF4, activating transcription factor 4; CCK‐8, cell counting kit‐8; GSH, glutathione, GPX4, Glutathione Peroxidase 4; Lipid ROS, lipid reactive oxygen species; long‐chain acyl‐CoA synthetase. ATF4/SCO2 Activated the MAPK Pathway in Lung Cancer Cells The MAPK signaling pathway consists of four distinct cascades, including ERK1/2, JNK1/2, p38 MAPK, and ERK5 [18]. To study the effect of SCO2 on the MAPK pathway, Western blotting for p38 and ERK1 was carried out. A549 and NCI‐H1299 cells with the silence of ATF4 expressed the down‐regulated p‐p38/p38 and p‐ERK1/ERK1, while the exogenous overexpression of SCO2 increased p‐p38/p38 and p‐ERK1/ERK1 protein levels in ATF4‐knockdown cells (Figure 6A,B). Taken together, ATF4 could activate the MAPK signaling pathway by elevating SCO2 in lung cancer. ATF4/SCO2 activated MAPK pathway in lung cancer cells. (A−B) After shNC+oeNC, shATF4+oeNC, or shATF4+oeSCO2 transfection in A549 (A) and NCI‐H1299 (B) cells, p‐p38/p38 and p‐ERK1/ERK1 protein levels were tested by Western blotting. ***p < 0.001, * denotes shNC+oeNC versus shATF4+oeNC; # p < 0.05, ### p < 0.001, # denotes shATF4+oeNC versus shATF4+oeSCO2; n = 3. ATF4, activating transcription factor 4; MAPK, mitogen‐activated protein kinases; p‐ERK1, Phospho‐Extracellular signal‐regulated kinase 1; SCO2, synthesis of cytochrome c oxidase 2. ATF4 Promoted Lung Tumor Growth by Up‐Regulating SCO2 To explore the role of ATF4/SCO2 in vivo, a mouse model was established by injecting tumor cells. When compared with the mice implanted with shNC+oeNC‐transfected NCI‐H1299 cells, those mice implanted with ATF4‐silenced NCI‐H1299 cells manifested a reduced trend of tumor volume (Figure 7A) and weight (Figure 7B,C), whereas this tumor growth inhibition was restored in mice injected with shATF4+oeSCO2‐transfected NCI‐H1299 cells. IHC analysis showed that ATF4 and SCO2 protein levels suppressed by the shATF4+oeNC group were rescued in tumor tissues of shATF4+oeSCO2 group (Figure 7D). Besides, western blotting results exhibited that ATF4 knockdown could apparently suppress GPX4 protein level and boost ACSL4 expression in tumor tissues, whereas these effects were partially overturned by SCO2 upregulation (Figure 7E). ATF4 promoted lung tumor growth in vivo via the upregulation of SCO2. ATF4 promoted lung tumor growth by up‐regulating SCO2. Tumor models were established in mice through injection with transfected NCI‐H1299 cells of shNC+oeNC, shATF4+oeNC, or shATF4+oeSCO2. (A) Tumor volume curve. (B) Tumor images. (C) Tumor weight. (D) IHC analysis for ATF4 and SCO2 in tumor tissues. (E) Western blotting for protein detection of GPX4 and ACSL4 in tumor tissues. ***p < 0.001, *denotes shNC+oeNC versus shATF4+oeNC; ### p < 0.001, # denotes shATF4+oeNC versus shATF4+oeSCO2; n = 3. ATF4, activating transcription factor 4; GPX4, Glutathione Peroxidase 4; long‐chain acyl‐CoA synthetase, ACSL4; IHC, immunohistochemistry; SCO2, synthesis of cytochrome c oxidase 2. Discussion Ferroptosis has vital effects on cancer therapeutics, including chemotherapy, radiotherapy, and immunotherapy. Thus, activation of ferroptosis may be a potential strategy for improving the therapeutic outcomes of cancer [19]. The current findings have revealed the function and mechanism of ATF4 in regulating ferroptosis, providing a better understanding of ferroptosis in lung cancer. Notably, ferroptosis is intimately linked to disulfidptosis, a recently identified cell death mode driven by disulfide stress and cytoskeletal collapse [20, 21]. Both share metabolic vulnerabilities, including glucose starvation and redox imbalance [22]. Given that ATF4 governs amino acid metabolism and SCO2 regulates oxidative phosphorylation, the ATF4/SCO2 axis may concurrently influence disulfidptosis by modulating NADPH availability and actin dynamics. This study thus holds broader significance for dual‐targeting strategies in cancer therapy. Through the database analysis and sample/cell detection, SCO2 was shown to be expressed at a higher level in lung cancer, which was consistent with that in LUAD [10]. SCO2 was considered to be significantly associated with LUAD progression through analysis of clinicopathologic features [11]. To further explore this association, functional analysis was performed, and it was found that reduced SCO2 expression hindered cell proliferation/migration but triggered apoptosis. Meanwhile, lung cancer patients with high SCO2 levels had poor overall survival. Assuredly, SCO2 functioned as an oncogene that accelerated the malignant phenotypes of lung cancer cells. Phillyrin was reported to up‐regulate SCO2 to induce the increased Fe2+ concentration and oxidative injury, which implied that SCO2 promoted ferroptosis in hepatocellular carcinoma [23]. Herein, it was discovered that SCO2 played an anti‐ferroptosis role in lung cancer cells, suggesting that the effect of SCO2 on ferroptosis may be different in different types of tumors. Moreover, the ferroptosis inhibitor recovered cell growth in SCO2‐silenced cells. After integration with these results, it could be regarded that SCO2 knockdown impeded lung cancer cell development through stimulating ferroptosis. Ferroptosis is largely responsible for tumor‐regulating function in various cancers. For example, KLF2 impeded the invasion and migration of colorectal cancer cells via leading to ferroptosis [24]; Caveolin‐1 expedited malignant progression of head and neck squamous cell carcinoma through blocking ferroptosis [25]. Thus, SCO2 exerted its oncogenic role by inhibiting ferroptosis in lung cancer. Of note, the opposing roles of SCO2 in ferroptosis—pro‐death in HCC but pro‐survival in lung cancer—align with multi‐omics findings that cell death regulators exhibit lineage‐specific functions [26, 27, 28]. Transcriptomic and metabolomic profiling indicate that mitochondrial genes like SCO2 switch between pro‐survival and pro‐death roles depending on baseline metabolic states, iron availability, and antioxidant capacity [26]. Such omics‐based predictive frameworks support the context‐dependent function of SCO2 observed herein and underscore the necessity of tumor‐type‐specific evaluation in therapeutic targeting. ATF4 exhibits promoting/suppressing regulation in a variety of cancers. In colon cancer and hepatocellular carcinoma, ATF4 acts as an anticancer factor to delay cell progression [29, 30]. Inversely, ATF4 enhances cell invasion and proliferation as an oncogene in lung cancer partially through Wnt/β‐catenin signaling regulation [31]. Consistent with the issued report, ATF4 contributed to lung cancer cell development in this study. Furthermore, the current study revealed that lung cancer patients with high ATF4 levels had poor overall survival. ATF4 can induce transcriptional activation of SLC7A11, which plays an important role in tumorigenesis through controlling ferroptosis [32]. Ji et al. have shown that dihydroartemisinin motivates lipid peroxidation and ferroptosis of hepatocellular carcinoma cells via down‐regulating ATF4 [33]. Also, in lung cancer, LSD1 inhibition decreased the expression of ATF4 and then accelerated ferroptosis with anti‐tumor effect [34]. The detection in lung cancer cells revealed that the interference of ATF4 increased oxidative harm and Fe2+ level. Similar to these past findings, ATF4 overexpression also hampered ferroptosis in lung cancer. Online database analysis preliminarily predicted that ATF4 might have a relatively weak correlation with SCO2 in public databases due to platform heterogeneity and complex sample composition. Meanwhile, ATF4 was found to be strongly associated with SCO2 in clinical lung cancer samples. As expected, the current work verified that ATF4 interacted with the SCO2 promoter in lung cancer cells and regulated SCO2 expression positively. ATF4 was indicated to bind to FGF19, thereby being involved in breast cancer migration and invasion [35]. Ma et al. also discovered the activation of ATF4 on PTS and the implication of ATF4/PTS in LUAD cell behaviors [13]. Previous studies have suggested that several key regulatory factors are involved in the process of ferroptosis. For instance, GPX4 could prevent ferroptosis by using GSH to convert lipid hydroperoxides into non‐toxic lipid alcohols, reducing oxidative stress and inhibiting the occurrence of ferroptosis [36, 37]. On the other hand, ACSL4 could promote the incorporation of PUFAs into phospholipids, making them vulnerable to lipoxygenase (ALOX)‐mediated oxidation, which compromises the lipid bilayer and induces ferroptosis [38, 39]. It has been reported that ACSL4‐driven ferroptosis involves the downstream molecule GPX4. Herein, the current work verified that ATF4 silencing could hinder GPX4 protein level and promote ACSL4 expression in vitro and in vivo, which were partly overturned by SCO2 upregulation, supporting that the ATF4/SCO2 axis could drive ferroptosis by regulating the ACSL4/GPX4 pathway. Furthermore, in the present functional research, SCO2 overexpression not only reversed the promoting effect of ATF4 knockdown on cell ferroptosis but also recovered the anti‐tumor regulation on lung tumorigenesis in mice. ATF4/SCO2 axis was testified to participate in lung cancer growth both in vitro and in vivo. The MAPK pathway is associated with the regulation of various tumor behaviors, such as proliferation, apoptosis, migration, and differentiation [18, 40]. Shi et al. disclosed that ZCCHC14 modulated the proliferation and invasion of lung cancer through the MAPK‐P38 pathway [41]. ATF4 was manifested to down‐regulate apelin via the pro‐apoptotic p38 MAPK pathway [42]. Through the protein marker examination, it was affirmed that ATF4 could activate the MAPK pathway by targeting SCO2 in lung cancer cells. In addition, iron overload and lipid peroxidation accumulation can affect MAPK signaling activity [43]. Orlistat has induced ferroptosis in pancreatic neuroendocrine tumors through suppressing the MAPK Pathway [44]. Therefore, the anti‐ferroptosis effect of ATF4/SCO2 on lung cancer might be attributed to the activation of the MAPK signaling pathway. The associations between elevated SCO2/ATF4 and aggressive clinicopathological features (tumor size, TNM stage, lymphatic metastasis) suggest clinical utility for risk stratification. High ATF4/SCO2 expression may identify patients requiring intensified surveillance or aggressive therapy. Therapeutically, targeting this axis offers multiple strategies: (1) ATF4 inhibitors could transcriptionally suppress SCO2 to induce ferroptosis; (2) combining ferroptosis inducers (e.g., erastin, sorafenib) with ATF4/SCO2 blockade may produce synergistic effects, supported by the Fer‐1 rescue experiment; (3) clinically approved MAPK inhibitors (e.g., trametinib) could be repurposed for high ATF4/SCO2 tumors. Ferroptosis biomarkers (lipid ROS, GSH, Fe2+) may serve as predictive markers for patient stratification. Prospective validation of ATF4/SCO2 as companion diagnostic markers and exploration of combination regimens integrating ferroptosis modulators with MAPK inhibitors warrant future clinical investigation. The ATF4/SCO2 axis further enables prioritization of dual ferroptosis/disulfidptosis modulators for pre‐clinical testing. Candidate agents include: (1) disulfiram derivatives, which induce disulfidptosis via actin disruption and promote ferroptosis through copper‐dependent Fenton reactions [45]; (2) glucose transporter inhibitors (e.g., BAY‐876), triggering disulfidptosis by glucose starvation while depleting NADPH and GSH to sensitize ferroptosis [46]; and (3) thiol‐reactive agents (e.g., elesclomol, piperlongumine) that deplete GSH and disrupt cytoskeletal disulfide homeostasis [47, 48]. High ATF4/SCO2 expression may identify patients resistant to conventional ferroptosis inducers but amenable to such dual‐targeting approaches. However, there are some limitations due to the research resources in the animal model. Firstly, animal models have been established using only H1299 cells, while animal models derived from A549 cells have not been validated. Secondly, mitochondrial damage and the role of the MAPK signaling pathway activation have not been investigated in an animal model. In the future, an animal model of A549 cells, mitochondrial damage, and the role of the MAPK pathway in vivo will be used as the further study direction. Besides, although lipid ROS, GSH, Fe2+, GPX4, and ACSL4 are well‐established ferroptosis markers, they are not absolutely specific to ferroptosis and may be influenced by other oxidative stress‐related processes. Additional orthogonal evidence, such as mitochondrial ultrastructural analysis by TEM and rescue experiments with liproxstatin‐1, would have strengthened our conclusions but were not performed due to resource limitations. Thus, the potential involvement of other cell death pathways cannot be entirely excluded. These aspects warrant further investigation in future studies. Additionally, this study lacked multi‐omics integration to systematically evaluate how transcriptomic, metabolomic, and proteomic layers collectively predict ferroptosis and disulfidptosis regulators in lung cancer. Disulfidptosis‐specific assays (e.g., cytoskeletal disulfide bond detection, SLC7A11‐independent cell death characterization) were not performed, limiting assessment of whether the ATF4/SCO2 axis concurrently regulates disulfidptosis. These represent important future directions. In summary, functional and mechanical findings suggested that ATF4 impeded ferroptosis in lung cancer by inducing transcriptional regulation of SCO2 to promote the MAPK pathway. This study uncovered a molecular mechanism of the ATF4/SCO2/MAPK pathway in inhibiting ferroptosis of lung cancer. ATF4 and SCO2 may have important applications in ferroptosis‐targeted therapy by acting as controllable markers. Author Contributions Liang Yu: writing – original draft, investigation, conceptualization, writing – review and editing, visualization, validation, methodology, project administration, supervision, data curation. Sheng Zhang: writing – review and editing, visualization, validation, methodology, conceptualization, investigation, writing – original draft, formal analysis, supervision. Jinxi He: formal analysis, visualization, writing – review and editing, writing – original draft, investigation, conceptualization, methodology, validation, data curation. Bo Ma: writing – review and editing, investigation, conceptualization, validation, methodology, software, supervision, resources. Xuyang Song: data curation, supervision, formal analysis, software, resources, validation, methodology, investigation, funding acquisition. Rui Qiu: data curation, supervision, resources, software, formal analysis, validation, visualization, funding acquisition, conceptualization, writing – original draft. Bo Yu: investigation, funding acquisition, methodology, validation, software, formal analysis, supervision, data curation, resources. Ethics Statement and Consent Human study: The present study was approved by the ethical review committee of General Hospital of Ningxia Medical University with approval No. KYLL‐2025‐0609. All procedures involving human disease tissue samples complied with the Declaration of Helsinki and its subsequent amendments. Written informed consent was obtained from all enrolled patients. All tissue specimens were fully anonymized to protect patient privacy and were only used for the basic scientific research of this study, not for other purposes. Animal study: All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of General Hospital of Ningxia Medical University with approval No. KYLL‐2024‐0822. The animal study strictly followed the ARRIVE guidelines and the 3Rs principle (Replacement, Reduction, Refinement). All surgical procedures were performed under standardized anesthesia, and every effort was made to minimize animal pain and stress. Humane euthanasia was conducted at the end of experiments in accordance with the national standard “Guidelines for Ethical Review of Laboratory Animal Welfare” (GB/T 35892‐2018). Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting File Yu L., Zhang S., He J., et al., “ATF4/SCO2 Inhibits Ferroptosis in Lung Cancer Through Activating the MAPK Signaling Pathway,” Journal of Biochemical and Molecular Toxicology 40 (2026): e71054. 10.1002/jbt.71054. Liang Yu and Sheng Zhang contributed equally to this work. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Supporting File Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.