Identification of Pinostilbene as a natural STING agonist that triggers FTH1 degradation via K48-ubiquitination to induce ferroptosis in non-small cell lung cancer.

1
Introduction
Non-small cell lung cancer (NSCLC) remains a leading cause of cancer-related deaths worldwide. Current treatment modalities often face significant limitations, including drug resistance, severe side effects, and tumor recurrence [1]. Despite advancements, the prognosis for advanced NSCLC patients remains poor, underscoring an urgent need for novel and effective therapeutic strategies [2,3]. Recently, ferroptosis, a distinct form of regulated cell death characterized by iron-dependent lipid peroxidation, has emerged as a promising therapeutic avenue in cancer treatment [4,5]. This unique mechanism offers potential to overcome drug resistance in various cancers, including NSCLC [6]. Numerous studies highlight the susceptibility of NSCLC cells to ferroptosis induction, suggesting its potential to overcome the limitations of conventional therapies [7]. However, the clinical translation of ferroptosis-inducing strategies is hampered by the lack of potent and specific ferroptosis-inducing agents and a comprehensive understanding of the underlying molecular mechanisms [8].
The Stimulator of Interferon Genes (STING) pathway, a crucial component of the innate immune system, is primarily recognized for its role in detecting cytosolic DNA and initiating type I interferon responses vital for antiviral and antitumor immunity [9,10]. Emerging evidence suggests a fascinating interplay between the STING pathway and ferroptosis [11]. STING activation can induce ferroptosis through several proposed mechanisms, including the upregulation of type I interferons and other pro-inflammatory cytokines, which can subsequently influence cellular iron metabolism and lipid peroxidation [12]. For instance, STING activation can lead to the transcriptional activation of genes involved in iron homeostasis, potentially disrupting iron-sulfur cluster biogenesis or increasing labile iron pools. Furthermore, the downstream signaling of STING, particularly through the TBK1-IRF3 axis, can impact key ferroptosis regulators, such as glutathione peroxidase 4 (GPX4) or the cystine-glutamate antiporter system xCT, thereby promoting the accumulation of toxic lipid reactive oxygen species [13]. This connection highlights the exciting possibility of leveraging STING activation to induce programmed cell death in cancer cells [14]. Given this link, targeting the STING pathway represents a highly promising strategy for inducing ferroptosis in lung cancer, potentially triggering a robust ferroptotic response in NSCLC cells, leading to their demise.
While the therapeutic potential of STING activation is evident, the clinical application of existing STING agonists faces several challenges [15]. Many current STING agonists, particularly cyclic dinucleotides, often suffer from poor cellular permeability due to their hydrophilic nature, limiting their ability to reach the cytosol where STING resides [16]. This poor bioavailability often necessitates high doses or specialized delivery methods, which can lead to increased off-target effects and systemic toxicity. Furthermore, some agonists exhibit rapid degradation by phosphodiesterases in biological fluids, resulting in short half-lives and reduced therapeutic efficacy [17]. Finally, non-specific activation of immune cells or widespread inflammatory responses can lead to significant systemic side effects, hindering their clinical translation [18]. These limitations underscore the urgent necessity to identify and develop novel STING agonist compounds with improved pharmacological properties, including enhanced potency, specificity, and favorable safety profiles, to translate the promise of STING-mediated ferroptosis into effective NSCLC therapies [19,20].
In this study, we aimed to identify and characterize novel STING agonist compounds with enhanced therapeutic potential for NSCLC. We successfully identified Pinostilbene as a potent STING agonist and investigated its mechanism of action in inducing ferroptosis in NSCLC cells. Furthermore, we evaluated its anti-tumor efficacy, its ability to sensitize NSCLC cells to the ferroptosis inducer RSL3, and its impact on the tumor microenvironment in an in vivo mouse model. Our findings highlight the therapeutic potential of Pinostilbene as a novel STING agonist for NSCLC treatment through its ability to induce and enhance ferroptosis.

2
Materials and methods
2.1
Reagents and antibodies
The IFNB1 reporter gene plasmid was purchased from Miaoling Plasmid Platform (Wuhan, China). Lipofectamine 3000 transfection reagent was obtained from Thermo Fisher Scientific (Waltham, MA, USA). The Firefly Luciferase Reporter Gene Assay Kit was procured from Beyotime (Shanghai, China). Pinostilbene was purchased from TargetMol (Boston, MA, USA). The ferroptosis inducer RSL3 and ferroptosis probes FerroOrange were acquired from MedChem Express (Monmouth Junction, NJ, USA). The lipid peroxidation probe C11-BODIPY, and mitochondrial dyes MitoTracker Green and MitoTracker Red were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Hoechst 33342 and DAPI nuclear stains were obtained from Beyotime (Shanghai, China). Antibodies against STING, TBK1, and IRF3 were purchased from Proteintech (Wuhan, China). Phospho-STING (p-STING), phospho-TBK1 (p-TBK1), phospho-IRF3 (p-IRF3), LC3B, IRP1, and IRP2 antibodies were obtained from Cell Signaling Technology (CST, Danvers, MA, USA). Anti-Nuclear receptor coactivator 4 (NCOA4) antibody was purchased from Abcam (Cambridge, UK). Anti-4-hydroxynonenal (4-HNE) antibody was purchased from Proteintech (Wuhan, China). Antibodies against GPX4, Tfrc, FTH1, ACSL4, Flag, HA and β-actin were purchased from Huabio (Hangzhou, China).

2.2
Cell culture and transfection
Human NSCLC cell lines, NCI–H1299 and A549, were cultured in DMEM high-glucose medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. Cells were maintained in a humidified incubator at 37 °C with 5% CO2. When cells reached 70-80% confluence, plasmid DNA was mixed with Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) reagent in serum-free medium, and the mixture was added to the cells after a 15-min incubation according to the manufacturer's instructions. After the transfection, the cells were cultured for an additional 24–48 h.

2.3
In vivo experiments
Six to eight-week-old female C57BL/6 mice (weighing 18–20g) were acclimatized for one week in an SPF-grade environment. Lung cancer models were established by subcutaneously injecting mouse Lewis lung carcinoma (LLC) cells (1 × 106 cells/100 μL PBS) into the right hind flank region of the mouse. Tumor growth was monitored daily, and treatment was initiated when tumor volumes reached 50-100 mm3 (calculated as V = length × width2 × 0.52). The treatment group (n = 6) received intraperitoneal injections of Pinostilbene (10 mg/kg, every two days), intraperitoneal injections of RSL3 (10 mg/kg, every two days), or the combination treatment, while the model group (n = 6) received an equal volume of vehicle, for a duration of 24 days. Tumor dimensions were measured every two days using a caliper, and body weight changes were recorded. At the experimental endpoint, mice were humanely euthanized by cervical dislocation. Tumor tissues were excised, weighed, and photographed. A portion of fresh tumor tissue was digested into a single-cell suspension for flow cytometry analysis of immune cell populations. The remaining tissue was divided for different purposes: one part was immediately snap-frozen in liquid nitrogen for mRNA or protein detection, while others were fixed in 4% paraformaldehyde for immunofluorescence and immunohistochemistry (IHC). Major organs were also collected for toxicity assessment by evaluating the organ index and performing H&E staining. All experimental procedures strictly adhered to animal welfare principles, and were approved by Animal Ethics Committee of China Pharmaceutical University (Ethical Approval No.: YSL-202506094).

2.4
STING agonist screening
H1299 cells were transfected with the IFNB1 promoter-driven luciferase reporter gene plasmid (IFNB1-luc) and cultured for an additional 24 h to ensure stable luciferase expression. Subsequently, cells were seeded into 96-well white plates at a density of 3 × 103 cells/well. After cell adhesion, natural product library compounds (10 μM) or positive controls (e.g., STING agonist-1) were added, with three technical replicates per treatment group. After 24 h of incubation at 37 °C, the cell culture medium was removed. Following the instructions of the Firefly Luciferase Reporter Gene Assay Kit, 100 μL of reporter gene cell lysis buffer was added to each well and shaken at room temperature for 15 min to ensure complete cell lysis. Then, 100 μL of luciferase substrate was added to each well, and chemiluminescence signals were immediately measured using a microplate reader. For data processing, relative luciferase activity was calculated for each compound-treated group, normalized to the untreated control group.

2.5
CCK-8 assay for cell viability
H1299 cells were seeded into 96-well plates at a density of 3 × 103 cells/well. After 24 h of adhesion, cells were treated with RSL3 (5 μM) and Pinostilbene (100 μM) for 24 h. Subsequently, 10 μL of CCK-8 reagent was added to each well and incubated at 37 °C for 1 h. Absorbance at 450 nm (OD value) was measured using a microplate reader. Relative cell viability (%) was calculated as: (experimental group OD value - blank group OD value)/(control group OD value - blank group OD value) × 100%, with the untreated group set as 100% cell viability. The experiment was repeated three times.

2.6
Annexin V/PI staining for cell death detection
H1299 cells were seeded into 6-well plates at a density of 2 × 105 cells/well. After overnight adhesion, cells were treated with RSL3 (5 μM) and Pinostilbene (100 μM) for 24 h. Both detached and adherent cells were collected, washed twice with pre-chilled PBS, and then stained using the Annexin V-APC/PI method to detect cell death. In brief, cells were resuspended in 100 μL of binding buffer, incubated with 5 μg/mL Annexin V-APC and 2 μg/mL PI in the dark for 20 min at room temperature. The reaction was then stopped by adding 400 μL of binding buffer, and cell death was analyzed by flow cytometry. The Annexin V−/PI− population was defined as live cells, while cells in the other three quadrants were defined as dead cells.

2.7
Lipid peroxidation assay
Lipid peroxidation was assessed using the fluorescent probe C11-BODIPY. Briefly, H1299 cells seeded in 6-well plates were treated with RSL3 (5 μM), Pinostilbene (100 μM), or their combination for 24 h after overnight adhesion. For microscopic analysis, treated cells were co-stained with 5 μM C11-BODIPY and 5 μg/mL Hoechst 33342 for 30 min at 37 °C, washed with PBS, and then imaged. The intensity of the oxidized C11-BODIPY (excitation/emission ∼488/510 nm) was quantified to indicate lipid peroxidation, with Hoechst 33342 identifying nuclei. For flow cytometric analysis, cells were trypsinized, stained with 5 μM C11-BODIPY under the same conditions, washed, and analyzed immediately. The mean fluorescence intensity of the oxidized probe was measured using a flow cytometer for quantification.

2.8
Intracellular Fe2+ measurement
Intracellular ferrous iron (Fe2+) levels were determined using complementary approaches. Following the same treatment scheme as described in section 2.7, cells were analyzed either with the fluorescent probe FerroOrange or a commercial colorimetric assay. For fluorescence detection, treated cells were incubated with 1 μM FerroOrange in culture medium at 37 °C for 30 min. For microscopic imaging, cells were washed with PBS and the Fe2+-bound FerroOrange signal (excitation/emission ∼543/580 nm) was captured. For flow cytometry, cells were trypsinized, stained, washed, and the mean fluorescence intensity was measured using a flow cytometer. In parallel, total intracellular Fe2+ content was absolutely quantified using a Ferrous Ion Assay Kit (Beyotime Biotechnology, Shanghai, China) according to the manufacturer's instructions. Briefly, an equal number of treated cells from parallel samples were lysed, and the lysate was incubated with the chromogenic agent. Absorbance was measured at 593 nm, and Fe2+ concentration was calculated based on a standard curve and normalized to the total protein content.

2.9
Transmission Electron Microscopy (TEM)
Mitochondrial morphology was observed using TEM. In brief, following sample preparation through glutaraldehyde and osmium tetroxide fixation, dehydration, and resin embedding, ultrathin sections (50–90 nm) were cut using an ultramicrotome, stained with uranyl acetate and lead citrate, and subsequently imaged under a transmission electron microscope to assess mitochondrial ultrastructure, including shape, cristae morphology, and overall integrity.

2.10
Evaluation of mitochondrial damage by flow cytometry
H1299 cells were seeded into 6-well plates at a density of 2 × 105 cells/well. After overnight adhesion, cells were treated with RSL3 (5 μM) and Pinostilbene (100 μM) for 24 h. Cells were collected, washed twice with PBS, and then incubated with MitoTracker Green and MitoTracker Red (200 nM) at 37 °C in the dark for 30 min. Mitochondrial damage was assessed by flow cytometry. Percentage of cells with abnormal mitochondrial function was calculated for each group.

2.11
Measurement of mitochondrial membrane potential
Mitochondrial membrane potential was measured using the cell-permeant cationic dye Tetramethylrhodamine methyl ester (TMRM). Cells were incubated with 100 nM TMRM in complete culture medium for 20 min at 37 °C in the dark. After incubation, cells were washed twice with warm PBS to remove unincorporated dye and then maintained in dye-free medium during image acquisition under a fluorescence microscope.

2.12
Immunofluorescence
H1299 cells (5 × 104 cells) were seeded into 20 mm confocal microscopy dishes and cultured for 24 h until adherent. Cells were then treated with different concentrations of Pinostilbene (0, 20, 50, 100 μM) for 24 h. After treatment, the medium was removed, and cells were gently washed three times with pre-cooled PBS for 5 min each. Cells were fixed with 4% PFA at room temperature for 15 min, followed by three more PBS washes. For enhanced antibody penetration, cells were permeabilized with 0.3% Triton X-100 at room temperature for 30 min, then blocked with 10% goat serum at 37 °C for 1 h. Cells were incubated with primary anti-p-IRF3 rabbit monoclonal antibody overnight at 4 °C. After four washes with PBST (5 min each), cells were incubated with Alexa Fluor 594-conjugated secondary antibody (1:500) in the dark at room temperature for 1.5 h. Cell nuclei were stained with Hoechst33342 (1 μg/mL) in the dark for 5 min.

2.13
H&E staining
Mouse organ tissues fixed in 4% PFA for 24 h were routinely dehydrated, cleared, paraffin-embedded, and sectioned at 4 μm. Sections were baked at 80 °C for 30 min, dewaxed with xylene (two times for 10 min each), and rehydrated through graded ethanol (100%–30%, 5 min each). Sections were then stained with hematoxylin for 5 min, differentiated with 1% hydrochloric acid alcohol for 30 s, blued under running water for 5 min, and stained with eosin for 30 s. After routine dehydration with graded ethanol (70%–100%) and clearing with xylene, sections were mounted with neutral balsam. Stained sections were observed and images acquired using a Leica DMi8 inverted biological microscope, with five non-overlapping fields of view randomly selected for analysis per sample.

2.14
IHC
LLC tissue sections were baked at 80 °C for 30 min, dewaxed with xylene (two times for 10 min each), rehydrated through graded ethanol (100%–30%, 5 min each), and incubated with 3% hydrogen peroxide at room temperature for 15 min to block endogenous peroxidase activity. Antigen retrieval was performed by placing sections in sodium citrate buffer (pH 6.0) at high heat until boiling, followed by continuous heating at medium-low heat for 15 min, and then natural cooling to room temperature. After three washes with PBS, sections were blocked with 10% goat serum at room temperature for 30 min. Sections were incubated with primary antibodies (Anti-4-HNE, FTH1, Ki67, 1:200) overnight at 4 °C in a humidified chamber. After PBS washes, HRP-conjugated secondary antibody (1:500) was added and incubated at room temperature for 1 h. DAB chromogen solution was applied for color development, controlled under a microscope. Sections were counterstained with hematoxylin for 1 min, differentiated with hydrochloric acid alcohol, blued under running water, routinely dehydrated, cleared, and mounted with neutral balsam. Positive signals were identified as brown-yellow granular staining in the nucleus/cytoplasm. Five random fields of view were acquired per section.

2.15
RNA extraction and qPCR detection
H1299 cells (5 × 105 cells/well) were seeded into 6-well plates. After 24 h of treatment with RSL3 (5 μM) and Pinostilbene (100 μM) to induce ferroptosis, total RNA was extracted using TRIzol reagent (specific steps: cell lysis followed by chloroform phase separation, isopropanol precipitation, and 75% ethanol wash). RNA concentration and purity were determined. One μg of RNA was reverse-transcribed using a PrimeScript RT kit (37 °C for 15 min, 85 °C for 5 s). qPCR was performed using the SYBR Green method with the following conditions: pre-denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s. The expression of key ferroptosis-related genes (FPN1, GPX4, Tfrc, FSP1, FTH1, ACSL4) was detected. β-Actin was used as an internal control, and relative expression was calculated using the 2 -△△Ct method. Each group had three technical replicates, and the experiment was repeated three times. A no-template control was included to ensure specificity. The qPCR primer sequences are listed in Table S1.

2.16
Western blot
H1299 cells treated with RSL3 (5 μM), Pinostilbene (100 μM), or their combinations for 24 h were lysed on ice for 1 h with RIPA lysis buffer (containing 1% protease inhibitor). Lysates were centrifuged at 12,000 rpm for 15 min at 4 °C, and the supernatant was collected. Protein concentration was determined using the BCA method. Thirty μg of protein sample was subjected to SDS-PAGE (12% separating gel) and transferred to a PVDF membrane (300 mA for 100 min). The membrane was blocked with 5% skim milk at room temperature for 1 h. Primary antibodies against GPX4 (1:2000), Tfrc (1:2000), FTH1 (1:2000), LC3B (1:1000), NCOA4 (1:1000), IRP1 (1:1000), IRP2 (1:1000), and ACSL4 (1:2000) were incubated overnight at 4 °C. After PBST washes, HRP-conjugated secondary antibody (1:10000) was added and incubated at room temperature for 1 h. Blots were developed using ECL reagent, and band gray values were quantitatively analyzed using Image Lab software. β-Actin (1:5000) was used as a loading control. Protein detection after immunoprecipitation using an anti-Flag antibody was also performed following the same protocol.

2.17
Flow cytometric evaluation of the immune microenvironment
Single-cell suspensions were prepared from freshly harvested LLC tumor tissues by mechanical mincing followed by enzymatic digestion in a cocktail of 1 mg/mL Collagenase IV, 0.1 mg/mL Hyaluronidase, and 20 μg/mL DNase I in RPMI-1640 medium at 37 °C for 45 min. The digested tissue was filtered through a 70-μm cell strainer, and erythrocytes were lysed. For flow cytometric analysis, 2 × 106 cells per sample were first incubated with an anti-mouse CD16/32 antibody to block Fc receptors. Cells were then stained with BD Horizon™ Fixable Viability Stain 780 for 15 min at room temperature in the dark. After one wash with cold staining buffer (PBS containing 1% BSA), the cells were aliquoted into two panels for surface marker staining. The T-cell panel included antibodies against CD45-FITC, CD4-APC, CD8-BV510, CD25-PE and PD-1-Percp-Cy5.5. The myeloid cell panel included antibodies against CD45-FITC, CD11b-PE, Ly6G-BV510, F4/80-BV421, CD86-PE-Cy7, and CD206-AF647. All antibody incubations were performed for 30 min at 4 °C in the dark, followed by three washes with PBS. Cells stained with the T-cell panel were subsequently fixed, permeabilized using a commercial fixation/permeabilization kit, and stained intracellularly with CD107a-BV421 and IFN-γ-PE-Cy7. Data acquisition was performed on a BD LSR Fortessa flow cytometer. Analysis was conducted on live, singlet-gated CD45+ leukocytes. Within this population, the frequencies of CD4+ and CD8+ T cell subsets were determined. The CD8+ T cells were further analyzed to determine the frequencies of CD107a+, IFN-γ+, CD25+, and PD-1+ subsets. For myeloid cell analysis, the CD45+CD11b+ population was gated to quantify Ly6G+ neutrophils and F4/80+ macrophages. The F4/80+ macrophages were subsequently characterized based on the expression of CD86 (M1-like) and CD206 (M2-like) markers. All flow cytometry antibodies were purchased from BD Biosciences (San Jose, CA, USA).

2.18
Statistical analysis
All the experiments were independently repeated three times. All statistical analyses were performed using GraphPad Prism 9.0. Quantitative data are presented as mean ± standard deviation (SD). Differences between two groups were compared using unpaired Student's t-tests. Differences among multiple groups were compared using one-way ANOVA followed by Tukey's post hoc test for multiple comparison or two-way ANOVA followed by Sidak's multiple comparisons test. A two-sided P-value less than 0.05 was considered statistically significant.

3
Results
3.1
Pinostilbene effectively activates the STING/TBK1/IRF3 pathway in NSCLC cells
To screen for potential novel STING agonists, we established a high-throughput reporter system based on the cGAS/STING/IFNB1 signaling axis. We transfected H1299 cells with a luciferase reporter plasmid driven by the IFNB1 promoter. Upon activation of the STING pathway, the type I interferon response activates the IFNB1 promoter, thereby inducing luciferase expression (Fig. 1A). Using this system, we screened a natural product compound library. Among the more than 1400 compounds screened, Pinostilbene showed the strongest transcriptional activation of IFNB1 (Fig. 1B). Western blot analysis showed that Pinostilbene dose-dependently upregulated the phosphorylation levels of STING, TBK1, and IRF3 in H1299 and A549 NSCLC cells (Fig. 1C–J). Consistently, immunofluorescence staining revealed that Pinostilbene promoted the nuclear translocation of p-IRF3 in a dose-dependent manner (Fig. 1K). Furthermore, qPCR results demonstrated that Pinostilbene dose-dependently enhanced the expression of key downstream cytokines of the STING pathway, including IFBN1, Interferon-Induced Protein with Tetratricopeptide Repeats 1 (IFIT1), Interferon-Induced Protein with Tetratricopeptide Repeats 2 (IFIT2), Interferon-Stimulated Gene 15 (ISG15), C-X-C Motif Chemokine Ligand 10 (CXCL10), Interferon-Induced Protein 44 (IFI44), and Interferon-Induced Protein 44-Like (IFI44L) (Fig. 1L–R). Collectively, these results confirm that the Pinostilbene was identified to effectively activate the STING/TBK1/IRF3 pathway in lung cancer cells.

3.2
Pinostilbene enhances the sensitivity of NSCLC cells to RSL3-mediated ferroptosis
Ferroptosis is considered to have broad application prospects in cancer therapy. Previous studies have reported that STING pathway activation can promote ferroptosis sensitivity. Thus, the effect of Pinostilbene on the sensitivity of lung cancer cells to ferroptosis was further investigated. The CCK-8 assay showed that Pinostilbene significantly enhanced RSL3-induced reduction of cell viability in H1299 cells (Fig. 2A and B). During ferroptosis, phosphatidylserine is externalized and membrane permeability increases [21]. Annexin V/PI staining showed that Pinostilbene markedly increased the sensitivity of H1299 cells to RSL3-mediated ferroptosis (Fig. 2C and D). Lipid peroxidation, which is mediated by intracellular free Fe2+
via the Fenton reaction, is the most typical biochemical feature of ferroptosis. Using the C11-BODIPY probe, we observed that Pinostilbene significantly enhanced RSL3-mediated lipid peroxidation in H1299 cells via fluorescence microscopy and flow cytometry (Fig. 2E–H). We also observed that Pinostilbene treatment markedly increased the fluorescent signal of the Fe2+ probe FerroOrange in H1299 cells, suggesting a potential rise in intracellular free Fe2+ levels (Fig. 2I–L). In parallel, using a separate set of cell samples, intracellular Fe2+ was quantified with a colorimetric ferrous iron assay kit applied to cell lysates. Following normalization to total protein content, the quantitative results aligned with the trend observed by FerroOrange staining (Fig. 2M). Together, these results confirmed that Pinostilbene, a novel STING agonist, can effectively enhance the sensitivity of NSCLC cells to RSL3-mediated ferroptosis.

3.3
Pinostilbene promotes ferroptosis-like mitochondrial damage in NSCLC cells
Shrinkage of mitochondrial volume and destruction of mitochondrial cristae are the most typical morphological features of ferroptosis. Using TEM, we found that the ferroptosis-like mitochondrial damage in H1299 cells caused by the combined treatment of Pinostilbene and RSL3 was more severe than that of the RSL3 single-agent group (Fig. 3A). By utilizing dual-staining with the fluorescent probes MitoTracker Green and MitoTracker Red, we confirmed that the proportion of cells with mitochondrial damage was significantly higher in Pinostilbene and RSL3 co-treated cells compared to the cells treated with RSL3 alone (Fig. 3B and C). Mitochondrial ROS play a critical role in the occurrence of ferroptosis. Mitochondrial damage induced by RSL3 offen couples elevation of ROS and Fe2+ production [22]. As expected, MitoSOX staining showed that Pinostilbene enhanced RSL3-mediated mitochondrial ROS production (Fig. 3D and E). Mito-FerroGreen (a probe for mitochondrial Fe2+) staining results indicated that Pinostilbene enhanced RSL3-mediated mitochondrial Fe2+ elevation, which was consistent with the trend observed for cytoplasmic Fe2+ (Fig. 3F and G). TMRM is a widely used mitochondrial membrane potential probe, and a decrease in its fluorescence intensity indicates mitochondrial depolarization during ferroptosis [23]. Using TMRM staining, we demonstrated that Pinostilbene further exacerbated the RSL3-mediated decrease in mitochondrial membrane potential (Fig. 3H and I).

3.4
Pinostilbene accelerates FTH1 degradation via an NCOA4-ferritinophagy-independent pathway under RSL3 stimulation
Next, we sought to explore the molecular mechanism by which the STING agonist Pinostilbene promotes RSL3-mediated intracellular free Fe2+ accumulation and enhances ferroptosis in NSCLC cells (Fig. 4A–Q). Using qPCR, we found that Pinostilbene significantly increased the mRNA expression of the iron-storage protein FTH1 at baseline levels (Fig. 4B). However, under RSL3 induced-ferroptotic stress, Pinostilbene downregulated the mRNA expression of the iron-uptake protein Tfrc (Fig. 4A) and the lipid peroxidation-promoting protein ACSL4 (Fig. 4D). Meanwhile, Pinostilbene did not affect the transcription of the key ferroptosis-inhibiting gene GPX4 either in absence or in presence of RSL3 treatment (Fig. 4C). Furthermore, we found that although Pinostilbene upregulated the protein expression level of Tfrc at baseline, this effect was not observed under RSL3 stimulation (Fig. 4E and F). FTH1 protein expression was not detected at baseline levels in H1299 cells, but RSL3 stimulation significantly upregulated its expression. Since FTH1 inhibits ferroptosis by sequestering iron and reducing its availability for lipid peroxidation [24,25], we propose that the upregulated FTH1 binds to intracellular Fe2+ and exerts a negative feedback inhibitory effect on RSL3-induced ferroptosis. Importantly, Pinostilbene substantially reversed the RSL3-induced upregulation of FTH1 protein levels, suggesting that Pinostilbene may enhance NSCLC cell ferroptosis sensitivity by downregulating the expression of the iron-storage protein FTH1, thus releasing Fe2+ into the cytoplasm (Fig. 4E–G). Considering that Pinostilbene upregulates the mRNA expression level of FTH1 under RSL3 stimulation (Fig. 4B), we speculate that the downregulation of FTH1 protein by Pinostilbene may primarily be achieved by promoting its degradation rather than acting at the transcriptional level. NCOA4-mediated ferritinophagy can mediate the degradation of FTH1. Despite that Pinostilbene significantly upregulated the expression of the autophagy marker LC3B II at baseline, it had no significant effect on the protein expression levels of NCOA4 and LC3B II under RSL3 stimulation (Fig. 4E–H, I). These results suggest that Pinostilbene, the novel STING agonist we identified, may promote the degradation of FTH1 via an NCOA4-ferritinophagy-independent pathway, releasing free Fe2+ and enhancing the sensitivity of NSCLC cells to ferroptosis. Furthermore, under RSL3 stimulation, Pinostilbene upregulated the expression of the ferroptosis-inhibiting protein GPX4 while having no significant effect on the expression of the lipid peroxidation-promoting protein ACSL4 (Fig. 4J and K).
NRF2 is a key transcription factor regulating FTH1 expression [26]. qPCR analysis revealed that Pinostilbene upregulated the mRNA levels of NRF2 and its downstream targets NQO1 and HO-1 under basal conditions, a trend consistent with the upregulation of FTH1 mRNA (Fig. 4B–L–N). However, in the presence of RSL3, while Pinostilbene significantly downregulated the mRNA expression of NRF2 and HO-1, it showed no significant effect on the mRNA levels of FTH1 and NQO1 (Fig. 4B–L–N). The iron-sensing proteins IRP1 and IRP2 post-transcriptionally regulate FTH1 protein abundance by binding to the iron-responsive element (IRE) in the 5′-UTR of FTH1 mRNA. Under iron-deficient conditions, IRP1/2 binding inhibits FTH1 translation, whereas iron sufficiency alleviates this repression, promoting FTH1 synthesis to sequester excess iron [27]. As shown in Fig. 2I–M, Pinostilbene elevated intracellular free Fe2+ levels irrespective of RSL3. Although Pinostilbene increased IRP2 expression at baseline, it significantly downregulated both IRP1 and IRP2 protein levels in the presence of RSL3 (Fig. 4O–Q). Collectively, these results argue against a major role for NRF2-mediated transcriptional regulation or IRP1/2-mediated translational regulation in the sensitization of ferroptosis by Pinostilbene. Instead, they highlight the critical importance of post-translational FTH1 protein degradation in this process.

3.5
Pinostilbene promotes K48-Linked ubiquitination and proteasomal degradation of FTH1
To test whether Pinostilbene affects FTH1 protein stability, The FTH1 protein levels were examined in H1299 cells treated with CHX, a common-used inhibitor for protein synthesis. Pinostilbene significantly accelerated FTH1 degradation in presence of RSL3 treatment (Fig. 5A and B). Since the proteasome and lysosome pathways are the two major routes for protein degradation [28], we next employed specific inhibitors to dissect the mechanism. The promotion of FTH1 degradation by Pinostilbene under RSL3 stimulation could be reversed by the proteasome inhibitor MG132, but not by the lysosome inhibitors E64/Pepstatin, suggesting that this process is primarily mediated through the proteasome pathway (Fig. 5C). By transfecting Flag-FTH1 and HA-Ub plasmids into H1299 cells and treating with the proteasome inhibitor MG132, we found that Pinostilbene markedly enhanced the polyubiquitination of FTH1 under RSL3 stimulation (Fig. 5D). Consistent with this, blockade of the proteasomal degradation following ubiquitination with MG132 led to an accumulation of Pinostilbene-mediated ubiquitination of exogenous FTH1 upon RSL3 treatment, whereas the lysosome inhibitors E64/Pepstatin showed no such effect (Fig. 5E). These findings further suggest that Pinostilbene may mediate the degradation of FTH1 through the ubiquitin-proteasome pathway to enhance the sensitivity of NSCLC cells to ferroptosis. The ubiquitin molecule contains seven conserved lysine residues: K6, K11, K27, K29, K33, K48, and K63. These lysine residues can serve as linkage sites between ubiquitin molecules, forming polyubiquitin chains with different topological structures via isopeptide bonds. To determine the type of polyubiquitin chain linkage catalyzed by Pinostilbene on FTH1, we constructed ubiquitin mutant plasmids where these distinct lysine residues were mutated to arginine, including K6R, K11R, K27R, K29R, K33R, K48R, and K63R. We then co-transfected H1299 cells with Flag-FTH1 and either wild-type or the different ubiquitin mutant overexpression plasmids. We found that the K48R mutation almost completely abolished the FTH1 ubiquitination mediated by the RSL3 and Pinostilbene co-treatment, indicating that K48 is the major site for ubiquitin linkage to FTH1 and subsequent proteasome-mediated degradation (Fig. 5F).

3.6
Pinostilbene activates the STING/ferroptosis pathway to exert in vivo antitumor effects
We then constructed lewis lung cancer xenograft model to examine the anti-tumor efficacy of Pinostilbene. As shown in Fig. 6A, Pinostilbene (10 mg/kg/day) effectively inhibited tumor growth. We euthanized the mice after 24 days of treatment and found that both tumor volume and weight in the Pinostilbene-treated group were significantly lower than in the control group (Fig. 6B and C). H&E staining revealed no significant pathological damage to the heart, liver, spleen, lungs, or kidneys of the tumor-bearing mice (Fig. 6D), suggesting that the compound has good safety at the therapeutic dose. Consistent with the in vitro results, Pinostilbene potently activated the STING/TBK1/IRF3 pathway in lung cancer tissue in vivo (Fig. 6E–H) and upregulated the expression of downstream cytokines of this pathway, including IFNB1, IFIT1, ISG15, USP18, and CXCL10 (Fig. 6I–M). In our in vitro experiments, we found that the novel STING agonist Pinostilbene promotes the degradation of the iron-binding protein FTH1, thereby enhancing the sensitivity of NSCLC cells to ferroptosis. Consistently, through immunohistochemistry and Western blot, we found that Pinostilbene downregulated FTH1 protein expression in tumor tissue in the LLC mouse model (Fig. 6N–Q) and enhanced the level of the ferroptosis biomarker 4-HNE (Fig. 6R and S). Additionally, Ki67 staining showed that Pinostilbene significantly inhibited the proliferation of tumor cells in vivo (Fig. 6T and U). Activation of the STING pathway and ferroptosis is known to enhance the antitumor immune response. We found that Pinostilbene upregulated the expression of inflammatory cytokines IL-1β, IL-6, and TNF-α in tumor tissue (Fig. 6V–X) and increased the infiltration of tumor-killing CD8+ T cells, while having no obvious effect on CD4+ T cell infiltration (Fig. 6Y). These results demonstrate that the novel STING agonist Pinostilbene we identified exerts its antitumor effect by activating the STING/TBK1/IRF3 pathway, which in turn downregulates FTH1 and activates ferroptosis, thereby enhancing CD8+ T cell-mediated antitumor immunity.

3.7
Pinostilbene enhances the in vivo antitumor efficacy of the ferroptosis inducer RSL3
While the classical ferroptosis inducer RSL3 functions by directly inhibiting GPX4, our findings reveal that Pinostilbene triggers ferroptosis in NSCLC cells by mediating the proteasomal degradation of FTH1. This distinct mechanism underlies the ability of Pinostilbene to synergistically enhance the sensitivity of NSCLC cells to RSL3-induced ferroptosis in vitro. Consistent with this, in the mouse LLC model, the combination of Pinostilbene and RSL3 significantly suppressed tumor growth, with an effect markedly stronger than that of RSL3 alone (Fig. 7A–C). Terminal dissection revealed normal morphology of major organs in all treatment groups (Fig. 7D), and the organ indices showed no significant difference compared to the control group (Fig. 7E). H&E staining detected no microstructural damage in the heart, liver, spleen, lungs, or kidneys (Fig. 7F). These results indicate a favorable safety profile for both the combination treatment and each agent administered singly. Compared to RSL3 monotherapy, the combination treatment significantly downregulated FTH1 protein levels in tumor tissues (Fig. 7G and H), upregulated the ferroptosis marker 4-HNE (Fig. 7I and J), and reduced the level of the tumor cell proliferation marker Ki67 (Fig. 7K and L). Collectively, these data demonstrate that Pinostilbene can safely and effectively downregulate FTH1 protein levels in NSCLC tissues, inhibit tumor cell proliferation, and enhance sensitivity to RSL3-induced ferroptosis in vivo.

3.8
Combination of pinostilbene and RSL3 promotes intratumoral CD8+ T cell infiltration, activation, and macrophage M1 polarization in LLC tumors
Finally, utilizing the mouse LLC model, we further evaluated the impact of the Pinostilbene and RSL3 combination on the tumor immune microenvironment. By immunofluorescence and flow cytometry, we found that both Pinostilbene and RSL3, either alone or in combination, increased the proportion of CD8+ T cells, but not CD4+ T cells, within tumors, with the most pronounced effect observed in the combination group (Fig. 8A, B, G, H). Flow cytometric analysis further revealed that, compared to the control, all treatment groups exhibited an increase in activated CD107a+, CD25+, and IFN-γ+ subsets among intratumoral CD8+ T cells, alongside a decrease in the exhausted PD-1+ subset (Fig. 8C–F, I–L). Notably, the combination treatment was significantly more effective than RSL3 monotherapy in promoting CD8+ T cell activation (Fig. 8C–F, I–L). Although the treatments did not significantly alter the overall proportions of Ly6G+ neutrophils or F4/80+ macrophages within the tumor myeloid compartment (Fig. 8M–P, Q), they promoted the polarization of macrophages towards the CD86+ M1 phenotype (Fig. 8N, O, R, S). Collectively, these results demonstrate that the combination of Pinostilbene and RSL3 exerts enhanced antitumor effects by fostering CD8+ T cell infiltration and activation, as well as promoting macrophage M1 polarization within the tumor microenvironment.

4
Discussion
NSCLC remains a formidable challenge in oncology, necessitating the development of novel therapeutic strategies that can bypass existing resistance mechanisms and minimize systemic toxicity [29]. Our study addresses this need by identifying and characterizing Pinostilbene, a natural product, as a potent and promising STING agonist. The findings presented here demonstrate that Pinostilbene exerts its robust antitumor effects by effectively activating the STING/TBK1/IRF3 pathway, which in turn orchestrates a multi-faceted response that includes the induction of ferroptosis and the enhancement of CD8+ T cell-mediated antitumor immunity.
Pinostilbene, a natural stilbenoid derivative found in Pinus species and grapes, serves as both a methylated analog of resveratrol and a major metabolite of pterostilbene [30]. Notably, the methylation of its molecular structure significantly enhances its lipophilicity, resulting in superior bioavailability compared to resveratrol [31]. This compound exhibits a wide spectrum of documented biological activities, including potent antioxidant, anti-inflammatory, and neuroprotective effects. The neuroprotective function is particularly well-established, with demonstrated efficacy in mitigating 6-OHDA-induced dopaminergic neurodegeneration through suppression of the JNK/c-Jun pathway [32]. Beyond neurological protection, Pinostilbene shows promising metabolic regulatory functions via dual inhibition of aldose reductase and α-glucosidase, coupled with activation of SIRT1/SIRT7 signaling pathways [33]. In oncological research, Pinostilbene has been shown to exert antitumor effects through induction of mitochondrial apoptosis via the p53/Bax/caspase-3 cascade and suppression of androgen receptor signaling in prostate cancer models [34,35]. While previous studies have highlighted its potential in cancer prevention and treatment by inducing apoptosis, promoting cell cycle arrest, and inhibiting tumor angiogenesis, its role as a direct STING agonist and a ferroptosis inducer has remained largely unexplored. The therapeutic potential of activating the STING pathway is evident, but the clinical translation of existing STING agonists faces significant challenges. Many current STING agonists, particularly hydrophilic cyclic dinucleotides, suffer from poor cellular permeability, which severely limits their ability to reach the cytosol where STING resides. This often necessitates complex delivery systems or high doses, leading to potential off-target effects and systemic toxicity. Furthermore, CDNs can be rapidly degraded by phosphodiesterases such as ENPP1 in the extracellular space, resulting in short half-lives and reduced therapeutic efficacy [17]. This underscores the urgent need to identify and develop novel, effective STING activators with improved pharmacological properties, such as enhanced bioavailability and a favorable safety profile.
Substantial evidence indicates that STING activation promotes ferroptosis in tumor models through multiple molecular mechanisms. Key processes primarily involve the transcriptional suppression of system Xc‾ cystine/glutamate antiporter activity, which results in glutathione depletion [36]. Concurrently, STING activation induces mitochondrial reactive oxygen species generation and facilitates the accumulation of the labile iron pool [[37], [38], [39]]. These coordinated processes collectively disrupt cellular redox homeostasis and accelerate lethal lipid peroxidation. It should be noted, however, that the regulatory function of STING signaling in ferroptosis exhibits complex context dependency. For example, in diabetic nephropathy models, STING activation has been shown to suppress ferroptosis through stabilization of the iron exporter ferroportin, thereby enhancing cellular iron efflux [40]. These apparently divergent effects demonstrate that the biological consequences of STING activation are determined by specific cellular environments, pathological states, and unique microenvironmental conditions. Our investigation establishes a clear and causal link between Pinostilbene-induced STING activation and the subsequent induction of ferroptosis in NSCLC cells. The high-throughput luciferase reporter assay successfully identified Pinostilbene from a natural product library as a strong activator of the IFNB1 promoter, a hallmark of STING pathway activation. This was mechanistically validated by Western blot analysis, which showed a dose-dependent increase in the phosphorylation of STING, TBK1, and IRF3, confirming the activation of the core STING signaling cascade. Downstream of this activation, Pinostilbene promoted the nuclear translocation of p-IRF3 and the transcriptional upregulation of key STING-responsive genes, including IFIT1 and CXCL10 [29,41]. These results align with the growing body of literature that posits STING as a crucial upstream regulator of ferroptosis, suggesting that the ability of Pinostilbene to trigger this pathway is central to its therapeutic action.
The in vitro data further solidified the role of Pinostilbene as a ferroptosis inducer and enhancer. Pinostilbene significantly enhanced the sensitivity of H1299 cells to RSL3, a classic ferroptosis inducer, as evidenced by enhanced cell viability reduction, increased Annexin V/PI-positive cell death, and augmented lipid peroxidation. The observation that Pinostilbene also increased intracellular free Fe2+ levels underscores its direct influence on cellular iron metabolism, a critical driver of the Fenton reaction and subsequent lipid peroxidation. Furthermore, our ultrastructural and fluorescence-based analyses confirmed that Pinostilbene exacerbated the characteristic mitochondrial damage associated with ferroptosis, including mitochondrial shrinkage, cristae destruction, and heightened mitochondrial ROS production.
The intersection of the STING and ferroptosis pathways is an area of intense research. Prior studies have shown that STING activation can influence ferroptosis sensitivity through various mechanisms. For instance, the induction of type I interferons downstream of STING can lead to changes in cellular iron homeostasis by upregulating genes involved in iron sequestration or export [42]. Additionally, certain STING agonists have been shown to affect lipid metabolism or to indirectly inhibit the key ferroptosis suppressor, GPX4, thereby promoting lipid peroxidation [43]. However, the precise molecular links and upstream regulatory mechanisms remain to be fully elucidated.
A major and novel finding of this study is the elucidation of a non-canonical pathway through which Pinostilbene regulates ferroptosis sensitivity. While the cellular abundance of the iron-storage protein FTH1 is classically governed by NRF2-mediated transcription, IRP1/2-mediated translational repression, and NCOA4-driven ferritinophagy, our data reveal that Pinostilbene markedly reduces FTH1 protein independently of these established mechanisms. While FTH1 mRNA expression was paradoxically increased by Pinostilbene under RSL3 stimulation, Western blot analysis revealed a dramatic decrease in FTH1 protein levels. This suggests a post-transcriptional regulatory mechanism. We further showed that this degradation is independent of NCOA4-mediated ferritinophagy, a well-established pathway for FTH1 degradation [44]. This is a critical distinction, as it reveals a new, non-autophagic route for FTH1 turnover, highlighting the unique mechanism of Pinostilbene and providing a novel therapeutic target for overcoming ferroptosis resistance.
Our investigation into the FTH1 degradation pathway led to the discovery of a K48-linked ubiquitination and proteasome-dependent mechanism. The CHX chase assay confirmed that Pinostilbene significantly reduced the half-life of FTH1 protein in the presence of RSL3. Moreover, the pro-degradation effect of Pinostilbene was specifically reversed by the proteasome inhibitor MG132, but not by lysosomal inhibitors (E64/Pepstatin), leading to an accumulation of polyubiquitinated FTH1. The use of various ubiquitin mutants pinpointed K48 as the critical lysine residue responsible for this polyubiquitin chain formation, which directs FTH1 to the proteasome for degradation. This newly identified pathway provides a comprehensive molecular explanation for how Pinostilbene increases the labile iron pool and enhances ferroptosis sensitivity in NSCLC cells.
The increase in IRP1 and IRP2 protein levels following Pinostilbene or RSL3 treatment was unexpected, given the concomitant rise in intracellular Fe2+, which would typically promote their FBXL5-mediated degradation [45,46]. This paradoxical observation, however, is not without precedent, as similar atypical regulation has been reported in cancer cells such as glioblastoma and melanoma challenged with ferroptosis inducers [47,48]. We consider this atypical IRP regulation a notable, yet secondary, aspect of the cellular response to ferroptotic stress in NSCLC cells, warranting further investigation into its precise biological significance.
The in vitro findings were successfully translated to an in vivo tumor model. Notably, in the LLC mouse model, the combination of Pinostilbene and RSL3 demonstrated significantly superior tumor growth suppression compared to RSL3 monotherapy, highlighting a synergistic antitumor effect in vivo. Pinostilbene treatment significantly inhibited tumor growth in LLC tumor-bearing mice without causing noticeable organ toxicity, a crucial finding for its therapeutic potential. The in vivo mechanistic data corroborated our in vitro observations, showing that Pinostilbene activated the STING pathway and downregulated FTH1 protein in tumor tissues, alongside an increase in the ferroptosis biomarker 4-HNE. This provides direct evidence that the STING-FTH1-ferroptosis axis is active and functionally relevant in a living organism.
Furthermore, we found that the antitumor effect of Pinostilbene extends beyond direct ferroptosis induction. Our investigation into the tumor immune microenvironment revealed a pivotal immunomodulatory dimension of the Pinostilbene and RSL3 combination therapy. Activation of the STING pathway is known to trigger a potent inflammatory response, which is crucial for shaping the tumor microenvironment and activating antitumor immunity [49]. Our results confirmed this by showing increased expression of inflammatory cytokines like IL-1β, IL-6, and TNF-α within the tumor microenvironment. Importantly, Pinostilbene and RSL3 combination reshaped the tumor immune microenvironment by driving macrophage polarization towards an antitumor M1 phenotype and fostering the infiltration and activation of tumor-killing CD8+ T cells. The crosstalk between STING activation, ferroptosis, and T cell immunity has been a focal point of recent research. A growing body of evidence suggests that ferroptosis can act as a form of immunogenic cell death [50]. Unlike apoptosis, which is generally immunologically silent, the necrotic-like morphology and cellular breakdown during ferroptosis can lead to the release of damage-associated molecular patterns (DAMPs), such as HMGB1 and ATP [51]. These DAMPs can subsequently be detected by antigen-presenting cells (APCs) like dendritic cells, leading to their maturation and enhanced presentation of tumor-associated antigens to T cells [52]. In parallel, STING activation in APCs or tumor cells can directly produce chemokines like CXCL10, which actively recruit immune cells, including CD8+ T cells, to the tumor site [53]. Combination of a direct cell death mechanism with a robust immune-enhancing effect makes Pinostilbene a highly attractive candidate for cancer therapy.
While our study provides compelling evidence, several key questions remain. First, It remains to be determined whether the pro-ferroptosis effects of Pinostilbene can be extended to other types of lung cancer beyond NSCLC. Second, given that NSCLC affects both genders while our in vivo study utilized female mice exclusively, the potential for gender-specific effects warrants further investigation. Third, future research should focus on identifying the specific E3 ligase responsible for the Pinostilbene-induced K48-linked ubiquitination of FTH1. Fourth, detailed pharmacokinetic and pharmacodynamic studies are also warranted to optimize its therapeutic window. Lastly, exploring the combined effects of Pinostilbene with other conventional therapies or immune checkpoint inhibitors could reveal new strategies for synergistic treatment, further highlighting the therapeutic potential of this natural compound.

5
Conclusions
In summary, this study identifies Pinostilbene as a potent natural STING agonist with a novel mechanism of action in NSCLC. We demonstrate that Pinostilbene enhances the susceptibility of NSCLC cells to ferroptosis by promoting the K48-linked ubiquitination and subsequent proteasomal degradation of the iron-storage protein FTH1, independently of the NCOA4-ferritinophagy pathway. Critically, in a preclinical model, the combination of Pinostilbene and RSL3 achieves synergistic tumor control, effectively downregulates tumoral FTH1, promotes ferroptosis, and simultaneously fosters an antitumor immune microenvironment by enhancing CD8+ T cell function and driving macrophage M1 polarization. The robust in vivo antitumor efficacy, coupled with its favorable safety profile and dual action on cancer cell death and immunity, positions Pinostilbene as a promising new agent for NSCLC treatment.

Funding
This work was supported by 10.13039/501100001809National Natural Science Foundation of China (82372973 and 82574436), Jiangsu Provincial Basic Research Foundation (BK20252066), the “Qinglan Project” in Universities of Jiangsu Province, Taishan Industrial Experts Program, and Chongqing Natural Science Foundation General Project (CSTB2023NSCQ-MSX0502).

CRediT authorship contribution statement
Shuangshuang Song: Formal analysis, Investigation, Writing – original draft. Siqi Hua: Conceptualization, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft. Guo Chen: Conceptualization, Funding acquisition, Writing – review & editing. Xianrui Yin: Validation. Zhengguo Chen: Writing – review & editing. Chong Li: Formal analysis, Funding acquisition. Danyang Zhou: Investigation, Validation. Bo Zhu: Conceptualization, Formal analysis, Investigation, Methodology, Visualization, Writing – review & editing.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.