Mitochondrial Damage-induced Ferroptosis: The Molecular Mechanism by Which Psoralen Inhibits the Proliferation and Invasion of Non-small-cell Lung Cancer Cells

Abstract

Background/Aim

Ferroptosis, an iron-dependent form of cell death mediated by lipid peroxidation, plays a critical role in non-small-cell lung cancer (NSCLC) progression. Psoralen, a bioactive natural compound, exhibits anticancer properties, but its effects and mechanisms in NSCLC remain unclear. This study explored whether psoralen induces ferroptosis by triggering mitochondrial damage and investigates the underlying molecular mechanisms.

Materials and Methods

Cell Counting Kit-8 was used to assess the impact of psoralen on cell viability, while 5-ethynyl-2'-deoxyuridine incorporation, colony-formation, scratch wound-healing, and Transwell assays evaluated its effects on proliferation, migration, and invasion. FerroOrange and 2′,7′-dichlorodihydrofluorescein diacetate fluorescence probes, Western blot, and kits for malondialdehyde (MDA), lipid peroxidation (LPO), reduced glutathione (GSH), and oxidized glutathione disulfide (GSSG) were used to assess ferroptosis-related markers. JC-1, MitoTracker Green, and MitoSOX Red probes, along with transmission electron microscopy, were used to evaluate mitochondrial damage. Bioinformatics analysis, network pharmacology, and molecular docking were conducted to elucidate potential mechanisms.

Results

Psoralen disrupted mitochondrial structure and function; increased Fe²⁺ accumulation; elevated levels of reactive oxygen species, MDA and LPO; depleted GSH; and downregulated glutathione peroxidase 4 (GPX4) and solute carrier family 7 member 11 (SLC7A11), ultimately inducing ferroptosis and inhibiting NSCLC cell proliferation and invasion. Eleven key target genes (PRKCB, MIF, GPI, AKR1C3, PDE3B, VDR, ALOX5, PTGS2, NQO1, MMP13, and CA9) were identified, with enrichment analysis linking them to arachidonic acid metabolism, vascular endothelial growth factor signaling, lipid metabolism, and oxidative stress. Molecular docking confirmed strong binding affinity of psoralen’ to these targets.

Conclusion

Psoralen induces ferroptosis in NSCLC by disrupting mitochondrial structure and function. These findings highlight its potential as a natural ferroptosis-targeting agent and provide insights for developing psoralen-based anticancer therapeutics.

Introduction

Lung cancer remains one of the most prevalent and deadliest malignancies worldwide. According to statistics from the International Agency for Research on Cancer, lung cancer accounted for 12.4% of all newly diagnosed cancer cases globally in 2022, with lung cancer-related deaths comprising 18.7% of total cancer mortality. The overall five-year survival rate remains below 20% (1). Non-small cell lung cancer (NSCLC), the predominant subtype, constitutes approximately 85% of all lung cancer cases (2). Due to the absence of specific symptoms in the early stages, most patients are diagnosed with disease at locally advanced or metastatic stages, significantly compromising treatment outcomes and survival prognosis (3). Currently, the primary therapeutic strategies for advanced NSCLC include chemotherapy, targeted therapy, and immune checkpoint inhibitors. Targeted therapy, designed for patients harboring driver gene mutations such as of epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), and ROS proto-oncogene 1 (ROS1), has demonstrated a significant extension in survival (4). However, the emergence of drug resistance limits its long-term efficacy. Therapy with immune checkpoint inhibitors, which enhance the immune system’s ability to eliminate tumor cells, has shown promising outcomes in patients with high programmed death-ligand 1 (PD-L1) expression. Nevertheless, the overall response rate remains relatively low, and some patients experience immune-related adverse events (5). For those lacking actionable mutations, chemotherapy remains the mainstay treatment; however, its long-term application is constrained by drug resistance and toxicity (6). Despite the advancements in NSCLC treatment, challenges such as therapeutic resistance, low response rates, and treatment-associated toxicities persist, highlighting the urgent need to explore novel therapeutic strategies.

In recent years, an increasing number of studies have highlighted the critical role of ferroptosis in the pathogenesis, progression, and treatment resistance of NSCLC [reviewed in (7)]. Ferroptosis is a form of non-apoptotic programmed cell death characterized by elevated intracellular reactive oxygen species (ROS) levels, enhanced lipid peroxidation, and depletion of reduced glutathione (GSH) (8). Mitochondrial dysfunction is considered a key inducer of ferroptosis, as mitochondrial membrane depolarization, increased ROS generation, and morphological alterations can accelerate the ferroptotic process in NSCLC cells (9). Studies have shown that the anti-angiogenic drug anlotinib can induce ferroptosis in NSCLC cells by modulating the tumor protein p53 (p53)/solute carrier family 7 member 11 (SLC7A11)/ glutathione peroxidase 4 (GPX4) pathway, thereby exerting therapeutic effects (10). These findings suggest that targeting ferroptosis may represent a promising strategy for overcoming treatment resistance and improving NSCLC management.

Against this backdrop, psoralen has garnered increasing attention for its potential applications in cancer therapy (11). A furanocoumarin compound derived from Psoralea corylifolia L., psoralen (Figure 1A) has a long history of use in traditional Chinese medicine. Studies have shown that flavonoids from P. corylifolia can inhibit epithelial-mesenchymal transition in NSCLC cells, thereby suppressing tumor cell proliferation (12). Psoralen has been demonstrated to induce cell-cycle arrest in breast cancer cells by modulating the WNT/β-catenin pathway, leading to the inhibition of tumor cell proliferation (13). Additionally, it has been reported to suppress hepatocellular carcinoma cell growth and induce apoptosis by triggering endoplasmic reticulum stress (14).

Figure 1.

Psoralen inhibits the proliferation of A549 cells. (A) Chemical structure of psoralen. (B) Cell viability of A549 cells treated with different concentrations of psoralen for 12, 24, and 48 h, measured using the CCK-8 assay. (C) Effects of psoralen on A549 cell proliferation, assessed by the 5-ethynyl-2'-deoxyuridine (EdU) incorporation assay. (D) Quantification of EdU-positive cells. (E) Effect of psoralen on colony-forming ability of A549 cells. (F) Quantification of colony-forming units. Results are presented as the mean±standard deviation, n = 3. **Significantly different at p<0.01.

In recent years, the potential role of ferroptosis in NSCLC treatment has gained increasing attention. Previous studies have demonstrated that certain bioactive compounds from traditional Chinese medicine, such as andrographolide, can promote ferroptosis in NSCLC cells by disrupting mitochondrial function (15). However, whether psoralen exerts similar effects, particularly in regulating lipid peroxidation, oxidative stress and mitochondrial dysfunction, remains to be elucidated. Therefore, this study aims to investigate the effects of psoralen on NSCLC cells and further explore whether its anticancer activity is mediated through the induction of ferroptosis. These findings may provide new insights into potential therapeutic strategies for NSCLC.

Materials and Methods

Materials. Psoralen (>99% purity), erastin (>99% purity), ferrostatin-1 (>99% purity), and 0.25% Trypsin-EDTA were purchased from MedChemExpress (Shanghai, PR China). Dulbecco's modified Eagle medium, fetal bovine serum (FBS), and phosphate-buffered saline (PBS) were purchased from Pricella Biotechnology Co., Ltd (Wuhan, PR China). Hoechst 33342 staining solution, 4% paraformaldehyde fixation solution, crystal violet staining solution, RIPA lysis buffer, phenylmethanesulfonyl fluoride, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample loading buffer, polyvinylidene fluoride membrane, Tris-buffered saline, Tween-20, Cell Counting Kit-8 (CCK-8), lipid peroxidation malondialdehyde (MDA) assay kit, ROS assay kit and 5-ethynyl-2'-deoxyuridine (EdU) cell proliferation detection kit, GSH and GSSG assay kit, Mitochondrial Membrane Potential Assay Kit with JC-1, Mitochondrial Green Fluorescence Staining Kit with Mito-Tracker Green, Mitochondrial Superoxide Assay Kit with MitoSOX Red, and ECL Detection Reagent were purchased from Beyotime Biotechnology Co., Ltd (Shanghai, PR China). Antibodies to GPX4 (ab125066), SLC7A11 (ab307601), β-actin (ab8227), and horseradish peroxidase-conjugated goat anti-rabbit IgG H&L (ab205718) were obtained from Abcam plc (Cambridge, UK). Lipid Peroxidation (LPO) Assay Kit (D799602) was purchased from Sangon Biotech Co., Ltd (Shanghai, PR China). FerroOrange probe (F374) was purchased from Dojindo Chemical Technology Co., Ltd (Kumamoto, Japan).

Cell culture. The human NSCLC cell line A549 (CL-0016) was purchased from Pricella Biotechnology Co., Ltd (Wuhan, PR China). A549 cells were cultured in complete medium (Dulbecco’s modified Eagle’s medium+10% FBS) at 37˚C in an incubator with 5% CO₂.

Cell viability assay. CCK-8 was used to assess the effect of psoralen on the viability of A549 cells. A total of 5×10³ cells per well were seeded into 96-well plates and incubated for 24 h. Subsequently, different concentrations of psoralen (0-160 μg/ml) were added, and cells were incubated at 37˚C with 5% CO2 for 12, 24, and 48 h. Thirty minutes before the end of incubation, 10 μl of CCK-8 reagent was added to each well, and the absorbance was measured at 450 nm. The half-maximal inhibitory concentration (IC₅₀) of psoralen was calculated using GraphPad Prism 9.0 (GraphPad Software, Boston, MA, USA) to evaluate its cytotoxic efficacy.

5-Ethynyl-2'-deoxyuridine (EdU) proliferation assay. A549 cells in the logarithmic growth phase were seeded into 24-well plates at a density of 3×10⁴ cells per well and incubated for 24 h under a 37˚C, 5% CO₂ environment to ensure proper adhesion. Cells were then treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml) for 24 h. Following treatment, cells were incubated with EdU working solution (final concentration: 10 μM) for 3 h according to the manufacturer's instructions. After incubation, fixation and permeabilization were performed as per the kit's protocol. Finally, EdU-positive cells were visualized under a fluorescence microscope. EdU-positive proliferating cells emitted green fluorescence, while total cell nuclei were counterstained with Hoechst 33342 (blue fluorescence). The relative proliferative capacity of the cells was assessed by calculating the ratio of EdU-positive cells to the total number of cells.

Colony-formation assay. A549 cells in the logarithmic growth phase were seeded into 6-well plates at a density of 500 cells per well and incubated at 37˚C in 5% CO₂. Once the cells adhered, the culture medium was replaced with fresh medium containing different concentrations of psoralen (0, 15, 30, and 60 μg/ml). Cells were continuously cultured for 10 days, with the medium refreshed every 2 days to maintain stable culture conditions and promote colony formation. At the end of incubation, cells were fixed with 4% paraformaldehyde for 15 min, followed by staining with 0.1% crystal violet solution for 15 min. After thorough washing with PBS and air drying, images were captured, and colonies consisting of at least 50 cells were counted to evaluate the effect of psoralen on A549 cell colony formation.

Wound-healing and Transwell assays. A549 cells were cultured in 6-well plates at 37˚C in 5% CO₂ until a confluent monolayer was formed. A 1 ml pipette tip was used to create a uniform scratch in the cell monolayer, and cell debris was removed by washing with PBS. Fresh medium containing different concentrations of psoralen (0, 15, 30, and 60 μg/ml) was then added to each well. Images of the wound area were captured at 24 and 48 h using an optical microscope, and the wound-healing rate was quantified using ImageJ software (National Institutes of Health, Bethesda, MA, USA) to assess lateral migration ability.

For the Transwell migration assay, A549 cells were resuspended in serum-free medium, adjusted to a density of 5×10⁴ cells/ml, and seeded into the upper chamber (200 µl per well). The lower chamber was filled with medium containing 10% FBS as a chemoattractant to promote migration. Cells in the chamber were treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml). After 24 h of incubation, the cells which had migrated into the lower chamber were fixed, stained, and counted under a microscope. The number of cells which had migrated was quantified using ImageJ software (National Institutes of Health) to evaluate vertical migration capacity.

Measurement of Fe²⁺ level. A549 cells in the logarithmic growth phase were seeded into 12-well plates at a density of 1×10⁵ cells per well and incubated at 37˚C with 5% CO₂ for 24 h to ensure proper adhesion. Subsequently, the cells were treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml) and erastin (10 μmol/l) for an additional 24 h. Following incubation, the culture medium was removed, and the cells were washed three times with serum-free medium. Then 1 μmol/l FerroOrange working solution (prepared by diluting the FerroOrange stock solution in serum-free medium) was added to each well, and the plates were incubated in the dark for 30 min. Without further washing, Fe²⁺ fluorescence signals were observed under a fluorescence microscope, and fluorescence intensity was quantitatively analyzed using ImageJ software (National Institutes of Health).

Measurement of ROS level. A549 cells were seeded into 6-well plates at a density of 5×10⁴ cells per well and allowed to adhere. After achieving proper attachment, the cells were treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml) and erastin (10 μmol/l) for 24 h. Following treatment, the cells were washed three times with serum-free medium, and 10 μmol/l of 2′,7′-dichlorodihydrofluorescein diacetat (DCFH-DA) working solution (prepared by diluting the DCFH-DA probe in serum-free medium) was added to each well. The plates were incubated at 37˚C in an incubator with 5% CO₂ in the dark for 20 min. After incubation, the ROS level was visualized under a fluorescence microscope, and fluorescence intensity was quantitatively analyzed using ImageJ software (National Institutes of Health).

LPO detection. A549 cells were seeded into 6-well plates at a density of 5×10⁴ cells per well and allowed to adhere. Cells were then treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml) and erastin (10 μmol/l) for 24 h. Subsequently, cell lysis and sample preparation were performed according to the respective kit protocols, followed by the quantification of MDA and LPO levels.

Measurement of GSH/oxidized glutathione disulfide (GSSG) ratio. A549 cells were seeded into 6-well plates at a density of 1×10⁵ cells per well and incubated until adherence. Cells were then treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml) and erastin (10 μmol/l) for 24 h. Following treatment, the cells were washed twice with PBS, and sample preparation, as well as the quantification of GSH and GSSG, were performed according to the manufacturer's instructions.

Western blot analysis. Western blot analysis was performed to assess protein expression levels, with β-actin used as the internal control. A549 cells were treated with different concentrations of psoralen (0, 15, 30, and 60 μg/ml) or with erastin (10 μmol/l) for 24 h. Following treatment, cells were lysed in RIPA buffer supplemented with phenyl-methanesulfonyl fluoride on ice for 30 min, with brief vortexing (5 s every 10 min) to facilitate lysis. After centrifugation, the supernatant was collected as the protein sample, and protein concentration was quantified using BCA Protein Assay Kit. Protein samples were mixed with 5× SDS-PAGE sample loading buffer and denatured by heating at 100˚C for 10 min in a metal bath. The proteins were then separated via 10% SDS-PAGE gel electrophoresis and transferred onto a polyvinylidene difluoride membrane using the wet-transfer method. After transfer, the membrane was blocked with 5% skim milk at room temperature for 1 h, followed by incubation with primary antibodies against GPX4 (1:5,000), SLC7A11 (1:1,000), and β-actin (1:5,000) at 4˚C overnight. The next day, the membrane was washed three times using TBS with 0.1% Tween-20 for 10 min per wash, then incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG H&L at room temperature for 1 h. After another three washes with TBS with 0.1% Tween-20, the protein bands were visualized using ECL Detection Reagent (P0018M, Beyotime). Band intensities were quantitatively measured using ImageJ software (National Institutes of Health), and the relative expression of GPX4 and SLC7A11 was normalized to β-actin. Data were analyzed from at least three independent experiments and expressed as mean±standard deviation.

Mitochondrial membrane potential. To assess mitochondrial membrane potential, A549 cells were divided into four treatment groups: control (0 μg/ml psoralen), psoralen (60 μg/ml), erastin (10 μmol/l), and psoralen (60 μg/ml) combined with ferrostatin-1 (1 μmol/l). After 24 h of treatment, cells were collected and washed twice with PBS to remove residual culture medium. Subsequently, the cells were incubated with JC-1 dye-containing medium at 37˚C in a 5% CO₂ environment for 30 min, allowing JC-1 to enter the cells and stain the mitochondrial membrane. After incubation, the cells were gently washed three times with PBS, and fresh culture medium was added. Finally, fluorescence signals were observed using a fluorescence microscope. Quantitative analysis of mitochondrial membrane potential was performed by calculating the ratio of red to green fluorescence intensity using ImageJ software (National Institutes of Health). Data from three independent experiments were expressed as the mean±standard deviation.

Mito-Tracker Green staining and mitochondrial superoxide assay. After 24 h of treatment with/without psoralen (60 μg/ml), erastin (10 μmol/l), or psoralen (60 μg/ml) plus ferrostatin-1 (1 μmol/l), the culture medium of A549 cells was replaced with pre-warmed serum-free medium, followed by the addition of Mito-Tracker Green working solution for staining. The cells were then incubated at 37˚C in an incubator with 5% CO₂ for 30 min. After incubation, the cells were washed three times with PBS to remove any unbound dye. Green fluorescence signals were recorded using a fluorescence microscope. Mitochondrial Superoxide Assay was performed using a similar procedure, except that MitoSOX Red working solution was added instead of Mito-Tracker Green after replacing the culture medium. Red fluorescence signals were recorded using a fluorescence microscope. Notably, during the incubation with working solutions, the samples were kept protected from light to prevent probe degradation. The average fluorescence intensity was quantitatively analyzed using ImageJ software (National Institutes of Health).

Transmission electron microscopy (TEM). A549 cells were divided into three treatment groups: psoralen (60 μg/ml), erastin (10 μmol/l), and psoralen (60 μg/ml) plus ferrostatin-1 (1 μmol/l). After 24 h of treatment, cells were fixed with 2.5% glutaraldehyde for 2 h. Following fixation, the cells were washed three times with PBS, then further fixed with 1% osmium tetroxide for 2 h. The samples were then dehydrated in a graded ethanol series (30%, 50%, 70%, 90%, and 100%), with each step lasting 10 min. After dehydration, the cells were infiltrated with acrylic resin and polymerized at 60˚C for 48 h to prepare embedding blocks. Subsequently, ultrathin sections were stained with 2% uranyl acetate for 15 min, followed by 0.1% lead citrate staining for 10 min. Finally, the stained sections were examined and imaged using a TEM. Morphological features of mitochondria, including outer membrane integrity, cristae structure, and swelling, were evaluated to assess mitochondrial damage.

Bioinformatics analysis. In this study, NSCLC-related datasets GSE18842 and GSE27262 were downloaded from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). Analysis of differentially expressed genes was performed using the limma package in R (version 4.3.1) (16). The three-dimensional chemical structure of psoralen was retrieved from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) and used to predict potential intervention targets from PharmMapper (https://www.lilab-ecust.cn/pharmmapper/), SwissTarget Prediction (http://swisstargetprediction.ch/), and TargetNet (http://targetnet.scbdd.com/calcnet/index/). Ferroptosis-related genes were obtained from FerrDb V2 (http://www.zhounan.org/ferrdb/current/), GeneCards (https://www.genecards.org/), and the Molecular Signatures Database (MSigDB) (https://www.gsea-msigdb.org/gsea/msigdb). Gene-expression transcriptomic data from patients with NSCLC were downloaded from The Cancer Genome Atlas (TCGA) (https://portal.gdc.cancer.gov), including the TCGA-LUAD and TCGA-LUSC cohorts. Unpaired sample tests were conducted for ferroptosis-related genes that overlapped with the predicted psoralen targets. Additionally, immuno-histochemical images of potential targets were obtained from the Human Protein Atlas (https://www.proteinatlas.org/) to further examine differential expression.

Protein-protein interaction (PPI) analysis of intervention targets was conducted using the STRING database (https://cn.string-db.org/), and the interaction network was visualized using Cytoscape 3.9.1. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed to investigate the biological processes, cellular components, molecular functions, and KEGG pathways associated with the intervention targets, aiming to elucidate the potential ferroptosis-mediated mechanism of psoralen action against NSCLC cells. Molecular docking was conducted to evaluate the binding affinity between psoralen and its target proteins. The X-ray crystal structure files of the intervention targets were retrieved from the Protein Data Bank (https://www.rcsb.org/). Protein structures were preprocessed, including removal of water molecules and addition of hydrogen atoms, using AutoDock 4.2.6 (The Scripps Research Institute, La Jolla, CA, USA). Molecular docking simulations and visualization of binding interactions were performed using AutoDock 4.2.6 and PyMOL 2.5.2 (Schrödinger, Inc., New York, NY, USA).

Statistical analysis. All experiments were independently repeated three times, and data from three biological replicates were analyzed for each experimental condition. All data are presented as the mean±standard deviation, and statistical analyses were performed using SPSS 25.0 (IBM, Armonk, NY, USA). One-way analysis of variance was used to compare differences among multiple groups. When a significant difference was observed, Tukey’s post hoc test was applied for pairwise comparisons. For non-normally distributed data or datasets with unequal variances, the Kruskal-Wallis test was used for non-parametric analysis, followed by Dunn’s post hoc test for pairwise comparisons. Statistical graphs were generated using GraphPad Prism 9.0, with significance threshold of p<0.05.

Results

Psoralen inhibits NSCLC cell proliferation. To investigate the effect of psoralen on NSCLC cell growth, the CCK-8 assay was performed. As shown in Figure 1B, the viability of A549 cells decreased significantly with increasing psoralen concentration and prolonged incubation. The IC₅₀ values at 12, 24 and 48 h were 36.19, 29.97 and 23.16 μg/ml, respectively, indicating that psoralen inhibits A549 cell proliferation in a time- and concentration-dependent manner. Based on the IC₅₀ results, subsequent experiments employed 0, 15, 30, and 60 μg/ml to further investigate the impact of psoralen on NSCLC cellular biological processes.

To assess the impact of psoralen on NSCLC cell proliferation, we conducted an EdU incorporation assay. The results (Figure 1C and D) demonstrated that the number of EdU-positive cells decreased significantly with increasing psoralen concentration, with cells treated with 60 μg/ml exhibiting the lowest number of proliferating cells. Furthermore, psoralen markedly suppressed the colony-forming ability of A549 cells. Colony-formation assay results (Figure 1E and F) indicated that colony numbers decreased progressively as psoralen concentration increased, with the most pronounced inhibition observed at 60 μg/ml. Collectively, these findings suggest that psoralen may exert anti-NSCLC effects by modulating cell proliferation pathways, and this inhibitory effect is concentration-dependent.

Psoralen inhibits NSCLC cell invasion and migration. To evaluate the effect of psoralen on the invasion and migration of NSCLC cells, wound-healing and Transwell migration assays were performed. As shown in Figure 2A and B, the wound-healing assay demonstrated that A549 cell migration ability progressively weakened with increasing psoralen concentration. At both 24 and 48 h, cells in the 60 μg/ml group exhibited significantly shorter migration compared to the control, indicating a pronounced inhibitory effect. In contrast, cells treated with 15 μg/ml did not show a significant reduction in migration. Further Transwell migration assay results confirmed that psoralen effectively suppressed A549 cell invasion (Figure 2C and D). The number of invading cells decreased in a concentration-dependent manner, with the highest psoralen concentration exhibiting the most pronounced inhibitory effect. Collectively, these findings indicate that psoralen significantly inhibits the invasion and migration of A549 cells, further supporting its antitumor potential against NSCLC cells.

Figure 2.

Psoralen inhibits A549 cell migration and invasion. (A) Wound healing assay assessing the effect of different psoralen concentrations on A549 cell migration. Red lines indicate the wound edges, showing the extent of cell migration at different time points across treatment groups. (B) Quantification of relative cell migration after 24 and 48 h of treatment with psoralen at different concentrations. (C) Transwell assay evaluating A549 cell invasion after 24 h of psoralen treatment. (D) Quantification of invading cells following with psoralen at different concentrations. Results are presented as the mean±standard deviation, n=3. **Significantly different at p<0.01.

Psoralen induces ferroptosis in NSCLC cells. Subsequently, the underlying mechanisms by which psoralen inhibits NSCLC cell invasion and migration were explored. Treatment with erastin (10 μmol/l) was specifically included as a positive control, given that erastin, a ferroptosis inducer, promotes GSH depletion by inhibiting the activity of the cystine-glutamate antiporter, thereby triggering ferroptosis (17). To assess Fe²⁺ accumulation in A549 cells, the FerroOrange fluorescence probe was used. As shown in Figure 3A and B, 15 μg/ml psoralen treatment did not significantly alter the intracellular Fe²⁺ level compared to the control. However, with increasing psoralen concentration, Fe²⁺ accumulation became progressively more pronounced, with 60 μg/ml psoralen or erastin treatment inducing the most significant Fe²⁺ accumulation. These results suggest that high-dose psoralen markedly promotes the accumulation of intracellular free iron. Since ferroptosis is often accompanied by intensified oxidative stress, intracellular ROS levels were further examined using DCFH-DA fluorescence probe. As shown in Figure 3C and D, the intracellular ROS level increased significantly in a dose-dependent manner following psoralen treatment, with 60 μg/ml psoralen or erastin inducing the most prominent generation of ROS. This indicates that psoralen may induce ferroptosis in A549 cells. To further validate ferroptosis induction, the expression of key ferroptosis regulatory proteins, GPX4 and SLC7A11, was assessed by western blot. The results demonstrated that psoralen treatment led to a dose-dependent reduction in GPX4 and SLC7A11 protein expression levels (Figure 4A and B), suggesting that psoralen may facilitate ferroptosis by downregulating GPX4 and SLC7A11.

Figure 3.

Psoralen increases intracellular Fe²⁺ and reactive oxygen species levels in A549 cells. FerroOrange staining was used to assess the effect of psoralen on intracellular Fe²⁺ level (A), with quantification of fluorescence intensity (B). 2′,7′-Dichlorodihydrofluorescein diacetate fluorescence probe was used to detect the level of intracellular reactive oxygen species (ROS) following psoralen treatment (C), with corresponding quantification of fluorescence intensity (D). Results are presented as the mean±standard deviation, n=3. Significantly different at ***p<0.001.

Figure 4.

Psoralen disrupts antioxidant defense and enhances lipid peroxidation in A549 cells. Western blot analysis of glutathione peroxidase 4 (GPX4) and solute carrier family 7 member 11 (SLC7A11) protein expression levels in A549 cells treated with psoralen psoralen (A), with quantitative analysis (B). Quantification of malondialdehyde (MDA) (C), lipid peroxidation (LPO) (D) levels, and reduced glutathione (GSH)/oxidized glutathione disulfide (GSSG) ratio (E) using assay kits to evaluate lipid peroxidation and antioxidant capacity in psoralen-treated A549 cells. Erastin was used as a ferroptosis inducer. Results are presented as the mean±standard deviation, n=3. Significantly different at *p<0.05, **p<0.01, or ***p<0.001.

Additionally, intracellular MDA, LPO, GSH, and GSSG levels were measured to further verify the proposed mechanism. As shown in Figure 4C and D, MDA and LPO levels were significantly elevated following treatment with 60 μg/ml psoralen or erastin, indicating severe lipid peroxidation. Meanwhile, the GSH/GSSG ratio markedly declined (Figure 4E), suggesting a substantial impairment of cellular antioxidant capacity. Taken together, these findings demonstrate that psoralen induces ferroptosis in A549 cells by promoting Fe²⁺ accumulation, enhancing oxidative stress, downregulating GPX4 and SLC7A11, and exacerbating lipid peroxidation, with higher concentrations of psoralen exerting more pronounced effects.

Psoralen-induced ferroptosis in NSCLC cells is associated with mitochondrial damage. To further investigate whether psoralen-induced ferroptosis in NSCLC cells is associated with mitochondrial dysfunction, this study evaluated the effects of psoralen treatment on mitochondrial membrane potential, mitochondrial morphology, and ROS levels. Ferrostatin-1, a well-established ferroptosis inhibitor, is known to counteract erastin-induced lipid peroxidation damage and protect mitochondrial function (18), thereby serving as a crucial control for validating the reliability of the experimental system.

The JC-1 staining assay (Figure 5A) demonstrated that high-dose psoralen significantly reduced the mitochondrial membrane potential, showing a similar trend to that with erastin treatment. Moreover, this effect was reversed by Ferrostatin-1, suggesting that psoralen induces mito-chondrial dysfunction. Mitochondrial structural integrity was further examined using Mito-Tracker Green staining. The results (Figure 5B) indicated that psoralen treatment led to a significant reduction in mean fluorescence intensity, similar to erastin-treated cells, reflecting mitochondrial structural damage. Ferrostatin-1 administration restored fluorescence intensity, further confirming its protective role against psoralen-induced mitochondrial damage. Additionally, MitoSOX Red staining revealed that intramitochondrial ROS levels were significantly elevated following high-dose psoralen treatment, a trend also observed in the erastin-treated group (Figure 5C). Ferrostatin-1 counteracted this effect, indicating that psoralen-induced oxidative stress is linked to ferroptosis. TEM observations further supported these findings (Figure 6). Psoralen-treated A549 cells exhibited mitochondrial swelling, membrane rupture, and cristae disruption, all characteristic signs of mitochondrial damage. Similar structural abnormalities were observed in the erastin-treated group, reinforcing the link between mitochondrial dysfunction and psoralen-induced ferroptosis in A549 cells. Taken together, these results suggest that mitochondrial damage plays a critical role in psoralen-induced ferroptosis in NSCLC cells.

Figure 5.

Psoralen-induced ferroptosis in A549 cells is associated with mitochondrial damage. (A) JC-1 fluorescence probe was used to assess the effect of psoralen on mitochondrial membrane potential, with quantification of the red/green fluorescence intensity ratio shown on the right. Red fluorescence represents high mitochondrial membrane potential, while green fluorescence indicates low mitochondrial membrane potential. (B) Mito-Tracker Green fluorescence probe was used to observe the effect of psoralen on mitochondrial morphology, with corresponding fluorescence intensity quantification shown on the right. (C) MitoSOX Red fluorescence probe was used to measure mitochondrial reactive oxygen species levels in A549 cells following psoralen treatment, with quantification of fluorescence intensity shown on the right. Erastin was used as a ferroptosis inducer, and ferrostatin-1 as a ferroptosis inhibitor. Results are presented as the mean±standard deviation, n=3. Significantly different at **p<0.01 and ***p<0.001 compared to the control.

Figure 6.

Transmission electron microscopy analysis of mitochondrial ultrastructure in A549 cells treated without/with psoralen; erastin was used as a ferroptosis inducer. Key morphological indicators observed include mitochondrial membrane integrity, cristae structure, and organelle swelling.

Bioinformatics analysis. Based on the above experimental results, this study demonstrated that psoralen effectively inhibits NSCLC cell proliferation, invasion, and migration, with ferroptosis induction playing a crucial role in this process. However, the precise molecular mechanisms underlying this effect remain unclear. To further explore these mechanisms, bioinformatics analyses were performed. Firstly, analysis of differentially expressed genes was conducted on NSCLC-related datasets GSE18842 and GSE27262 from the GEO database, comparing NSCLC tissues with adjacent normal lung tissues. Differential expression was defined using the criteria of p<0.05 and log2 (fold change) in expression >1 (Figure 7A). A total of 2,545 differentially expressed genes were identified in GSE18842, while 1,713 differentially expressed genes were identified in GSE27262. Next, potential psoralen intervention targets were identified by integrating data from PharmMapper, SwissTargetPrediction, and TargetNet, yielding 265 candidate targets. Additionally, a total of 1,378 ferroptosis-related genes were obtained from FerrDb V2, GeneCards, and the Molecular Signatures Database. By intersecting genes differentially expressed in NSCLC, psoralen potential targets, and ferroptosis-related genes, 11 potential psoralen targets in NSCLC were identified: protein kinase C beta (PRKCB), macrophage migration inhibitory factor (MIF), glucose-6-phosphate isomerase (GPI), aldo-keto reductase family 1 member C3 (AKR1C3), phosphodiesterase 3B (PDE3B), vitamin D receptor (VDR), arachidonate 5-lipoxygenase (ALOX5), prostaglandin-endoperoxide synthase 2 (PTGS2), NAD (P)H quinone dehydrogenase 1 (NQO1), matrix metallopeptidase 13 (MMP13), and carbonic anhydrase IX (CA9) (Figure 7B).

Figure 7.

Bioinformatics analysis of psoralen-induced ferroptosis in non-small-cell lung cancer (NSCLC) cells. (A) Volcano plots displaying differential gene expression analysis comparing NSCLC tumor tissues versus adjacent normal tissues in the Cancer Genome Atlas datasets GSE18842 and GSE27262. Up: Upregulated; Down: downregulated; Not sig: not significantly altered. (B) Venn diagram showing the intersection of GSE18842, GSE27262, ferroptosis-related genes, and psoralen intervention targets. (C) Comparison of expression changes in psoralen intervention targets between NSCLC and adjacent normal tissues in the Cancer Genome Atlas database. The targets include: protein kinase C beta (PRKCB), macrophage migration inhibitory factor (MIF), glucose-6-phosphate isomerase (GPI), aldo-keto reductase family 1 member C3 (AKR1C3), phosphodiesterase 3B (PDE3B), vitamin D receptor (VDR), arachidonate 5-lipoxygenase (ALOX5), prostaglandin-endoperoxide synthase 2 (PTGS2), NAD (P)H quinone dehydrogenase 1 (NQO1), matrix metallopeptidase 13 (MMP13), and carbonic anhydrase IX (CA9). (D) Protein–protein interaction network of psoralen intervention targets. (E) Results of Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analysis of psoralen intervention targets. (F) Molecular-docking analysis of psoralen with target proteins (PRKCB, MIF, GPI, PDE3B, VDR, ALOX5, PTGS2, NQO1, MMP13, and CA9). Results are presented as the mean±standard deviation, n=3. Significantly different at *p<0.05, **p<0.01, ***p<0.001. BP: Biological process; CC: cellular component; MF: molecular function; VEGF: vascular endothelial growth factor.

To validate their differential expression, these 11 targets were analyzed in NSCLC tissues and adjacent normal tissues from the TCGA database. The results showed that PRKCB, PDE3B, ALOX5, and PTGS2 were downregulated in NSCLC tissues, while MIF, GPI, AKR1C3, VDR, NQO1, MMP13, and CA9 were upregulated, which was consistent with the prior GEO differential analysis results (Figure 7C). Additionally, immunohistochemical data for these targets were retrieved from the Human Protein Atlas database to verify their protein-level expression in human tissues. As shown in Figure 8, the protein expression of PRKCB, ALOX5, and PTGS2 was lower in NSCLC tissues compared to normal lung tissues, whereas MIF, GPI, AKR1C3, NQO1, MMP13, and CA9 showed elevated expression in tumor tissues. These results were generally consistent with our bioinformatics and transcriptomic analyses. Protein-level data for PDE3B and VDR were not available in the database. Further analysis revealed 26 PPI relationships among these 11 targets, with the strongest interaction observed for PTGS2 (Figure 7D).

Figure 8.

Immunohistochemical expression of psoralen-targeted proteins in non-small cell lung cancer (NSCLC) and normal lung tissues. Representative immunohistochemical staining images from the Human Protein Atlas database showing the expression of protein kinase C beta (PRKCB), macrophage migration inhibitory factor (MIF), glucose-6-phosphate isomerase (GPI), aldo-keto reductase family 1 member C3 (AKR1C3), arachidonate 5-lipoxygenase (ALOX5), prostaglandin-endoperoxide synthase 2 (PTGS2), NAD (P)H quinone dehydrogenase 1 (NQO1), matrix metallopeptidase 13 (MMP13), and carbonic anhydrase IX (CA9) in normal lung and NSCLC tissues. Antibody identifiers from the Human Protein Atlas are indicated below each panel.

Further GO and KEGG enrichment analyses (Figure 7E) demonstrated that these targets were primarily associated with biological processes such as unsaturated fatty acid metabolic process, response to oxidative stress, reactive oxygen species metabolic process, and regulation of lipid metabolism. Their cellular components included ficolin-1-rich granules, secretory granule lumen, cytoplasmic vesicle lumen, and cytosolic region, while their molecular functions were linked to antioxidant activity, oxidoreductase activity acting on NAD (P)H, dioxygenase activity, and nuclear receptor binding. KEGG pathway analysis suggested that psoralen exerts its anti-NSCLC effects through key pathways, including arachidonic acid metabolism, parathyroid hormone synthesis, secretion and action, regulation of lipolysis in adipocytes, and the vascular endothelial growth factor (VEGF) signaling pathway.

Finally, molecular docking analysis was performed to further elucidate how psoralen interacts with its potential target proteins (Figure 7F). The results showed that psoralen directly forms hydrogen bonds with the residues Leu-394, Leu-406, and Lys-399 of PRKCB (−5.75 kcal/mol); Leu-19, Ser-20, and Val-14 of MIF (−7.58 kcal/mol); Asn-507 and Tyr-362 of GPI (−5.69 kcal/mol); Asn-830 and His-825 of PDE3B (−5.45 kcal/mol); Arg-252 of VDR (−5.6 kcal/mol); Thr-364 of ALOX5 (−5.5 kcal/mol); Arg-44, Cys-47, and Gly-45 of PTGS2 (−5.45 kcal/mol); Arg-273 and Asn-268 of NQO1 (−6.0 kcal/mol); Lys-140 of MMP13 (−5.08 kcal/mol); and Tyr-143 of CA9 (−5.34 kcal/mol). Notably, although psoralen did not form hydrogen bonds with ARK1C3, the binding energy (−6.06 kcal/mol) suggests that the interaction may rely on non-hydrogen bond interactions such as hydrophobic interactions and Van der Waals forces.

Taken together, psoralen exhibits strong binding affinity with these 11 targets, further supporting its role in NSCLC ferroptosis induction. These results indicate the therapeutic potential and feasibility of psoralen-induced ferroptosis in NSCLC treatment.

Discussion

As one of the most common cancer types worldwide, NSCLC remains a major public health concern threatening human health. In 2024, approximately 681,124 new cases of lung cancer were diagnosed in the P.R. China, with around 478,014 deaths attributed to the disease (19). However, only about 30% of patients with NSCLC are diagnosed at an early stage, thereby missing the optimal window for surgical intervention. For patients with early-stage localized disease, surgical resection remains the preferred treatment option, yet the annual risk of secondary lung cancer recurrence remains 2-3%, posing a significant challenge (20). In recent years, the rapid advancement of targeted therapy and immunotherapy has provided patients with NSCLC with more treatment options. Targeted drugs designed for mutations of Kirsten rat sarcoma viral oncogene homolog (KRAS), ALK, and EGFR genes have significantly improved survival in certain patient populations, while programmed cell death protein 1 (PD1)/PD-L1 immune checkpoint inhibitors have shown promising clinical benefits in advanced NSCLC treatment (21). However, the emergence of acquired drug resistance, tumor heterogeneity, and immune evasion continues to limit the durability of therapeutic efficacy. Additionally, some patients experience immune-related adverse events, leading to treatment discontinuation (5). Therefore, overcoming the limitations of existing NSCLC treatment strategies and identifying safer and more effective therapeutic approaches have become urgent challenges in the field of lung cancer research.

Among emerging therapeutic approaches, natural compounds derived from traditional Chinese medicine have attracted increasing attention due to their multi-target regulatory effects, ability to overcome or reverse drug resistance, and low toxicity profile (22). Psoralen, the major bioactive component of Psoralea corylifolia, has been demonstrated to induce tumor cell apoptosis, inhibit cell proliferation and migration, and disrupt mitochondrial function via oxidative stress induction (13,14). For instance, previous studies have shown that psoralen effectively suppressed breast cancer cell proliferation and reversed multidrug resistance in certain treatment settings (11).

In our study, the mechanism of action of psoralen in NSCLC cells was investigated. The CCK-8 assay was employed to evaluate cell viability, as its water-soluble tetrazolium salt is reduced by mitochondrial dehydrogenases, producing a soluble orange-colored formazan proportional to the number of viable cells. The results demonstrated that the IC₅₀ values of psoralen in A549 cells at 12, 24 and 48 h were 36.19 μg/ml, 29.97 μg/ml, and 23.16 μg/ml, respectively, indicating time- and dose-dependent cytotoxicity. Furthermore, the EdU proliferation assay, which incorporates a thymidine analog conjugated to a fluorescent dye into newly synthesized DNA, confirmed that psoralen significantly inhibited A549 cell proliferation in a dose-dependent manner, a finding further supported by the colony-formation assay. The wound-healing assay and Transwell migration assay revealed that psoralen effectively suppressed A549 cell migration and invasion, with higher concentrations exerting more pronounced effects. These findings align with previous studies demonstrating the antitumor potential of psoralen (13,14,23), suggesting that its anticancer effects may be mediated through the induction of cell death and the regulation of pathways involved in cell motility. Collectively, these results provide experimental evidence supporting psoralen as a potential therapeutic agent for NSCLC treatment.

Ferroptosis is a recently discovered form of programmed cell death characterized by iron-dependent lipid peroxidation, ultimately leading to cell demise (24). Unlike traditional forms of cell death such as apoptosis, necrosis, and autophagy, ferroptosis exhibits distinct morphological and biochemical characteristics. Specifically, ferroptotic cells do not display typical apoptotic features such as nuclear condensation and chromatin aggregation; instead, they are characterized by mitochondrial shrinkage and increased membrane density (25). This process is primarily triggered by the accumulation of intracellular Fe²⁺ and ROS, which induce lipid peroxidation, compromise cell membrane integrity, and ultimately lead to cell death. In the field of cancer therapy, inducing ferroptosis in tumor cells has emerged as a novel and promising strategy. Traditional Chinese herbal medicines have shown great potential in promoting ferroptosis in NSCLC cells. Many herbs and their bioactive components can modulate redox homeostasis, iron metabolism, and lipid metabolism, thereby triggering ferroptosis (26). FerroOrange, a specific fluorescent probe, selectively detects intracellular free Fe²⁺, emitting red fluorescence upon binding, which serves as a specific indicator of Fe²⁺ accumulation. In this study, psoralen treatment significantly increased Fe²⁺ accumulation in A549 cells, suggesting that psoralen may induce ferroptosis by promoting intracellular iron accumulation.

Oxidative stress plays a critical role in ferroptosis. During this process, Fe ions participate in Fenton reactions to generate excessive ROS, leading to lipid peroxidation and disruption of membrane integrity (27). MDA and LPO serve as key markers of lipid peroxidation, with elevated levels indicating oxidative damage to cellular membranes (28). GSH, the primary intracellular antioxidant, acts as a protective factor by neutralizing ROS and preventing oxidative damage. During ferroptosis, GSH is oxidized to GSSG, leading to an imbalance in cellular redox homeostasis. The GSH/GSSG ratio is commonly used as an indicator of oxidative stress levels (29). In this study, DCFH-DA, a cell-permeable fluorescent probe, was used to assess intracellular ROS levels, as it emits green fluorescence upon oxidation by ROS. The results revealed that the ROS level increased significantly in a psoralen concentration-dependent manner, and high-dose psoralen treatment led to a marked increase in MDA and LPO levels, along with a significant decrease in the GSH/GSSG ratio. These findings indicate that psoralen may induce ferroptosis by triggering oxidative stress, enhancing lipid peroxidation, and disrupting redox homeostasis, ultimately leading to cell death.

Additionally, GPX4 and SLC7A11 are key regulators of ferroptosis. GPX4, a selenium-dependent antioxidant enzyme, reduces lipid peroxides to lipid alcohols, thereby preventing their accumulation and protecting cells from oxidative stress (30,31). SLC7A11, the light chain subunit of the cystine/glutamate antiporter, facilitates cystine uptake for GSH synthesis, maintaining cellular antioxidant capacity (32). Western blot analysis demonstrated that psoralen treatment led to a dose-dependent reduction in GPX4 and SLC7A11 protein expression, further supporting the hypothesis that psoralen promotes ferroptosis by inhibiting antioxidant defense mechanisms. Notably, the effects of high-dose psoralen treatment closely resembled those observed with erastin, a known ferroptosis inducer (17,33), reinforcing the notion that psoralen may suppress NSCLC cell growth and invasion by inducing ferroptosis. These findings provide new insights into potential therapeutic strategies for NSCLC intervention.

Studies have shown that during ferroptosis, mitochondria undergo morphological changes, and their damage and dysfunction directly influence the onset and progression of ferroptosis. Mitochondrial dysfunction leads to excessive ROS production, which in turn promotes lipid peroxidation and ultimately triggers ferroptosis (34). Therefore, mitochondrial dysfunction is not only a consequence of ferroptosis but also a key factor in its initiation. JC-1 is a mitochondrial membrane potential-sensitive probe that aggregates in the mitochondrial matrix under high membrane potential, emitting red fluorescence. When mitochondrial membrane potential decreases, JC-1 exists mainly in its monomeric form, emitting green fluorescence. Thus, the red-to-green fluorescence ratio serves as an indicator of mitochondrial membrane potential. In this study, JC-1 staining revealed that high-dose psoralen treatment significantly reduced the mitochondrial membrane potential in A549 cells, indicating mitochondrial depolarization and dysfunction.

Mito-Tracker Green is a membrane potential-independent mitochondrial probe that specifically labels the total mitochondrial content regardless of the mitochondrial membrane potential. In contrast, MitoSOX Red is a mitochondria-specific superoxide anion probe that permeates the cell membrane, enters mitochondria, and is oxidized by superoxide anions, emitting red fluorescence. This probe is used to assess mitochondrial oxidative stress levels. Additionally, Mito-Tracker Green and MitoSOX Red staining assays demonstrated that high-dose psoralen treatment disrupted mitochondrial structure and markedly increased mitochondrial ROS levels. These findings suggest that psoralen-induced ferroptosis may be mediated through mitochondrial dysfunction and oxidative stress. Notably, the ferroptosis inhibitor Ferrostatin-1 effectively reversed these changes, partially restoring mitochondrial function and reducing the ROS level, further supporting the notion that psoralen exerts its antitumor effects by inducing ferroptosis. Moreover, TEM observations revealed distinct signs of mitochondrial swelling, membrane rupture, and cristae disruption in psoralen-treated A549 cells, further confirming mitochondrial damage. These morphological alterations are consistent with previous findings on Andrographolide-induced ferroptosis, in which mitochondrial dysfunction played a key role in promoting NSCLC cell death (15). Collectively, these results suggest that psoralen induces ferroptosis in NSCLC cells by triggering mitochondrial dysfunction, providing new insights into its underlying anticancer mechanisms.

Through bioinformatics analysis and network pharmacology, this study explored the potential molecular mechanisms underlying psoralen-induced ferroptosis in NSCLC cells. The integration of these approaches provided new insights into psoralen's mechanism of action and lays the foundation for its potential application in NSCLC treatment. A total of 11 potential targets involved in psoralen-induced ferroptosis in NSCLC cells were identified, including PRKCB, MIF, GPI, AKR1C3, PDE3B, VDR, ALOX5, PTGS2, NQO1, MMP13, and CA9. Differential expression analysis using the TCGA database revealed that PRKCB, PDE3B, ALOX5, and PTGS2 were significantly downregulated in NSCLC tissues compared with adjacent normal tissues, while MIF, GPI, AKR1C3, VDR, NQO1, MMP13, and CA9 were upregulated. These genes play diverse roles in tumor progression, oxidative stress response, and metabolic regulation. For instance, PTGS2, a classical marker and modulator of ferroptosis, has been shown to be involved in inflammation-driven tumorigenesis and ferroptotic signaling (35), and exhibited the strongest PPI connectivity in our network analysis. ALOX5, another key lipid metabolism enzyme, catalyzes the formation of lipid peroxides and has been implicated in ferroptotic cell death in cancer cells (36). The downregulation of both ALOX5 and PTGS2 in NSCLC may reflect tumor adaptation to suppress ferroptosis-related damage. On the other hand, genes such as NQO1 and MIF are associated with redox balance and cancer cell survival (37,38), and their upregulation may represent a compensatory antioxidant response in NSCLC. AKR1C3 and GPI, both involved in metabolic reprogramming, are frequently overexpressed in several cancer types and contribute to tumor proliferation and resistance to cell death (39,40). MMP13 and CA9 have well-established roles in extracellular matrix remodeling and tumor hypoxia adaptation, respectively (41,42). Collectively, these targets not only reflect critical nodes in ferroptosis, but also participate in interconnected oncogenic processes such as inflammation, metabolism, and redox regulation. Their identification highlights potential mechanistic links between psoralen-induced ferroptosis and broader tumor biology. Further validation and functional assays focusing on these genes may help delineate how psoralen manipulates ferroptotic networks to exert its antitumor effects in NSCLC.

Subsequent GO enrichment analysis revealed that these 11 targets were primarily associated with biological processes such as unsaturated fatty acid metabolism and oxidative stress response, both of which are closely linked to ferroptosis. This suggests that psoralen may induce ferroptosis by modulating these pathways. Additionally, several targets were related to specific cellular components, such as ficolin-1-rich granules and secretory granule lumens, implying that psoralen might also influence vesicle-mediated oxidative homeostasis. Further molecular function analysis indicated that these targets are enriched in antioxidant activity and oxidoreductase activity, consistent with their roles in maintaining redox balance and regulating lipid peroxidation—two key aspects of ferroptosis. For example, ALOX5 and PTGS2 are critical enzymes in arachidonic acid metabolism that promote lipid peroxide accumulation, a hallmark of ferroptotic death (43,44). NQO1, MIF, and VDR are involved in ROS scavenging and redox regulation (37,38,45), potentially affecting the cellular oxidative stress threshold under psoralen treatment. Meanwhile, CA9 may influence intracellular pH and iron solubility, indirectly contributing to ferroptotic sensitivity (46). Moreover, KEGG pathway analysis supported these findings, showing that the identified targets were enriched in pathways such as arachidonic acid metabolism and VEGF signaling. Arachidonic acid metabolism is directly tied to ferroptosis through its regulation of lipid peroxides (43), while VEGF signaling, though traditionally linked to angiogenesis, may also influence cellular stress responses and micro-environmental iron dynamics (47). To validate these functional predictions, molecular docking was conducted and revealed strong binding affinities between psoralen and several key targets, including PTGS2, MIF, and AKR1C3. These results offer structural evidence that psoralen may directly modulate ferroptosis-related regulators at the molecular level. Together, these findings not only reinforce the systems biology predictions but also deepen our understanding of how psoralen may exert its ferroptosis-inducing effects by disrupting lipid metabolism, redox equilibrium, and iron homeostasis in NSCLC cells. This integrated evidence provides a compelling rationale for future mechanistic and translational studies.

In this study, the mechanism of psoralen-induced ferroptosis in NSCLC cells was systematically investigated, as far as we are aware, for the first time. This was achieved not only through experimental validation but also by integrating bioinformatics predictions and molecular docking analysis. The findings comprehensively elucidate the core targets and signaling networks through which psoralen may mediate ferroptosis in NSCLC cells, providing a systematic and mechanistic understanding of its potential therapeutic role. This discovery not only expands the application of natural compounds in ferroptosis research but also provides new perspectives for developing psoralen-based anticancer therapies. From a translational medicine perspective, targeting ferroptosis induction offers new hope for overcoming chemoresistance and targeted therapy resistance in NSCLC. In the future, with further pharmacological and preclinical investigations, psoralen and its derivatives may emerge as novel ferroptosis inducers, offering more precise and effective treatment options for patients with NSCLC.

Conclusion

This study demonstrated that psoralen inhibits NSCLC cell proliferation and invasion by inducing ferroptosis. Psoralen promoted Fe²⁺ accumulation, increased ROS levels, reduced the expression of GPX4 and SLC7A11, and induced mitochondrial dysfunction and lipid peroxidation, ultimately leading to ferroptosis. Furthermore, bioinformatics analysis identified 11 potential targets, revealing their involvement in lipid metabolism regulation, oxidative stress response, and iron homeostasis, with molecular docking results further validating their relevance. These findings provide experimental evidence supporting the molecular mechanisms of psoralen-induced ferroptosis in NSCLC cells, offering new insights into the therapeutic potential of natural compounds in NSCLC treatment. Future in vivo studies will be crucial for advancing psoralen as a potential ferroptosis-targeting agent for NSCLC, paving the way for novel therapeutic strategies and treatment optimization.

Conflicts of Interest

The Authors confirm that the research presented in this paper was conducted independently and without any commercial or financial affiliations that could potentially lead to a conflict of interest.

Authors’ Contributions

ZM He, WH Gao and RY Yang conceived and designed the study; HY Deng, JC Tang, Y Xu and L Wu carried out experiments; ZB Wang and JT Zhang collected the data; HY Chen performed molecular docking; HY Deng and JC Tang wrote the article; ZM He, WH Gao and RY Yang reviewed the article and provided comments. All Authors contributed to this article and approved the submitted version.

Acknowledgements

The Authors express their gratitude to the School of Integrated Chinese and Western Medicine, Key Laboratory of Hunan Provincial for Integrated Traditional Chinese and Western Medicine, Hunan University of Chinese Medicine for providing us with molecular and cellular laboratories.

Funding

This study was supported by the National Natural Science Foundation of P.R. China (82205227), the Natural Foundation of Hunan Province (2023JJ30448), and Hunan Provincial Administration of Traditional Chinese Medicine Traditional Chinese Medicine Research Project (D2024017).

Artificial Intelligence (AI) Disclosure

No artificial intelligence (AI) tools, including large language models or machine-learning software, were used in the preparation, analysis, or presentation of this manuscript.