KLF5-driven TAZ–FASN signaling reprograms fatty acid metabolism to support Treg differentiation in lung cancer

Abstract

Background

Metabolic reprogramming is a fundamental hallmark of cancer and provides essential biochemical support for malignant progression. In lung cancer, aberrant fatty acid metabolism not only fuels cancer cell growth but also influences regulatory T cell (Treg) differentiation through altered lipid availability. Krüppel-like factor 5 (KLF5) has been implicated in lung cancer progression; however, its role in coordinating cancer fatty acid metabolism and Treg differentiation remains insufficiently defined.

Methods

We combined clinical lung cancer specimens, genetically modified lung cancer cell models, Treg differentiation systems, and mouse tumor models to define the metabolic function of KLF5. Gain- and loss-of-function approaches were used to determine how KLF5 affects lipid storage, fatty acid synthesis, extracellular free fatty acid production, and tumor growth. Conditioned-medium transfer experiments, fatty acid uptake assays, fatty acid oxidation measurements, and flow cytometry were applied to evaluate the impact of cancer cell-derived lipid output on Treg differentiation. Mechanistically, promoter-binding assays, transcriptional reporter analysis, protein-interaction experiments, molecular docking, and TAZ depletion were used to dissect how KLF5 regulates FASN-dependent fatty acid synthesis.

Results

KLF5 was highly expressed in lung cancer tissues and cells and showed positive associations with lipogenic markers and Treg-related indicators. Suppression of KLF5 markedly weakened the fatty acid metabolic phenotype of lung cancer cells, as reflected by reduced lipid droplet accumulation, decreased free fatty acid release, and downregulation of FASN, SCD1, DGAT2, and PLIN5. In vivo, KLF5 knockdown restrained tumor growth and reduced fatty acid synthesis-related molecular features. Mechanistically, KLF5 bound directly to the FASN promoter and cooperated with TAZ to enhance FASN transcription. KLF5 promoted nuclear accumulation of TAZ, whereas TAZ silencing attenuated KLF5-induced FASN expression, lipid accumulation, and free fatty acid production. Functionally, fatty acids released from KLF5-overexpressing lung cancer cells enhanced CD36-associated fatty acid uptake and fatty acid oxidation in Tregs, thereby promoting Treg differentiation. Conversely, KLF5 depletion reduced this lipid-associated differentiation process both in vitro and in tumor-bearing mice.

Conclusion

This study reveals a KLF5-driven fatty acid metabolic program in lung cancer. KLF5 cooperates with TAZ to activate FASN transcription, thereby increasing lipogenesis and free fatty acid release. The resulting lipid output promotes CD36-dependent fatty acid uptake and oxidation in Tregs and supports their differentiation. These findings identify the KLF5–TAZ–FASN axis as a cancer metabolism-centered mechanism linking lung cancer lipogenesis to Treg differentiation, highlighting this pathway as a potential metabolic vulnerability for interfering with fatty acid-supported lung cancer progression.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12967-026-08557-5.

Introduction

Lung cancer ranks as the second most frequent tumor worldwide, contributing to a maximum number of cancer-associated deaths [1]. Immune checkpoint inhibitors are considered a novel therapy that substantially enhances the prognosis of lung cancer patients [2]. Combination strategies based on immunotherapy have also shown clinical potential in advanced lung cancer [3, 4]. However, intrinsic and acquired resistance remains a major clinical challenge [5]. Therefore, identification of novel immune-related targets and signaling pathways remains essential for improving the efficacy of cancer immunotherapy [6]. Regulatory T cells (Tregs), belonged to CD4⁺ T cells, possess an immunosuppressive function, thus leading to immune tolerance [7]. Tregs take part in immune escape of lung cancer cells, which are responsible for the clinical treatment barriers of immunotherapy [8]. Metabolic programming for the synthesis of fatty acids (FA) is one of the major contributors to the fast multiplication of lung cancer cells. Moreover, it helps to create a lipid-rich microenvironment that favors the development of Treg cells and immunosuppression. Recent findings emphasize the possibility of using redox-modulating nanomaterials to interfere with these metabolic interactions. For example, the effectiveness of redox-modulating nanomaterials has been demonstrated by biosynthetic selenium nanoparticles extracted from potato peels, which show high antioxidant and cytotoxic activity [9]. In addition, green ZnS nanoparticles produced from banana peels possess high redox properties [10]. These findings suggest that targeting FA metabolism holds important clinical implications for the suppression of lung cancer progression.

A recent study documents that FA metabolism affects Treg differentiation and autoimmunity [11]. As reported, Tregs are highly dependent on lipid metabolism manifested as up-regulation of CD36 (fatty acid transporter), and high concentration of FA in tumors leads to an increase in Treg infiltration [12]. However, directly inhibiting FA synthesis of Tregs cannot effectively relieve their lipid metabolism remodeling, and consequently reduce the differentiation and immunosuppressive function of Tregs [13, 14]. Literature evidence has documented that increased exoteric FA intake is considered as an important characteristic of Treg metabolic remodeling [15]. For example, intratumoral Tregs exhibited higher level in FA uptake in comparison with peripheral Tregs [16]. Moreover, one study showed that epithelial cell-derived cytokine TSLP facilitated Treg activation via enhancing FA uptake of Tregs [17]. Therefore, uncovering the upstream regulatory mechanism of FA uptake by Treg cells in lung cancer microenvironment is of great clinical significance for overcoming the immune escape of lung cancer.

Mounting evidence has suggested that tumor cells can influence the function of surrounding immune cells in tumor microenvironment. For example, nasopharyngeal carcinoma cells enhanced suppressive activity of Tregs via regulating CD70/CD27 crosstalk-mediated lipid signaling network [18]. Another study documented that SIRT1-mediated glucolipid metabolic reprogramming in colorectal carcinoma cells increased Treg function via enhancing CX3CL1 secretion [19]. Moreover, MK1775 affected lipid metabolism of lung adenocarcinoma cells to influence fatty acid oxidation (FAO) of tumor-associated macrophages and CD8⁺ T cell infiltration within tumor microenvironment [20]. These arouse our speculation that certain mechanism might be responsible for promoting FA synthesis and secretion in lung cancer cells, thereby increasing FA uptake by Tregs within tumor microenvironment.

As reported, transcription factors can drive the lipid metabolic reprogramming by control transcription of FA synthases [21]. As a transcription factor, krüppel-like factor 5 (KLF5) is involved in the modulation of various cellular processes, including cell growth, differentiation, and death [22]. KLF5 has been identified to exert contradictory roles in different cancers, serving as a tumor suppressor or promoter [23]. It has been documented that KLF5 is up-regulated in lung cancer, which accelerates the malignant development of this cancer [24, 25]. Previous studies have indicated that KLF5 in tumor cells can affect the biological functions of surrounding immune cells. For example, up-regulation of KLF5 by NAT10 in gastric cancer cells contributed to M2 macrophage polarization, thus favoring liver metastasis [26]. In particular, KLF5 was found to increase prostaglandin E2 (PGE2) release from breast cancer cells, thereby decreased the number and functionality of anti-tumor CD8⁺ T cells [27]. What’s more, Lee et al. reported that KLF5 facilitated FA synthase in androgen-dependent prostate cancer cells [28], indicating the pivotal role of KLF5 in FA synthesis of tumor cells. So far, the regulation of KLF5 in FA synthesis and secretion in lung cancer cells and its subsequent influence on FA uptake and differentiation of Tregs have not been clarified.

FA synthesis in cancer cells can be regulated by various signaling pathways, such as the Hippo pathway [29]. Transcriptional co-activator with PDZ-binding motif (TAZ) is a downstream effector of the Hippo pathway [30, 31]. In multiple malignancies, the nuclear translocation of TAZ may inactivate Hippo pathway to trigger the expression of oncogenes [32]. As reported, nuclear translocation of TAZ contributed to lung cancer stemness via enhancing FA metabolism [33]. Due to lack of DNA binding domain, the nuclear TAZ controlled downstream gene transcription with the assistance of transcription factors, such as TEADs, EGR-1, TBX5, p73, and so on [34]. Interestingly, BioGRID database and molecular docking analysis predicted a potential interaction between KLF5 and TAZ. In addition, JASPAR database found that KLF5 might bind to the promoter of fatty acid synthase (FASN). Therefore, we speculate that KLF5 might interact with TAZ to regulate FASN transcription in lung cancer cells, and the interaction among them remains unexplored.

In this study, we hypothesized that KLF5 bound to nuclear TAZ to contribute to FASN transcription that favored FA synthesis in lung cancer cells. Subsequently, CD36-mediated uptake of free FA (FFA) from lung cancer cells promoted Treg differentiation to accelerate lung cancer progression. Our findings uncovered the role and regulatory mechanism of KLF5 in FA metabolism in lung cancer cells and subsequent Treg differentiation.

Materials and methods

Clinical samples

The lung cancer tissues and paired normal tissues (n = 50) were collected by surgery at the First Affiliated Hospital of Nanchang University (Nanchang, China) from January 2023 to December 2025 and validated by pathologists. The tissue samples were immediately stored in liquid nitrogen after collection. All patients did not undergo any chemoradiotherapy before surgery and provided their informed consent. Clinical parameters were shown in Supplementary Table 1. The clinical tissue acquisition procedure of this study was approved by the Ethics Committee of the First Affiliated Hospital of Nanchang University (Approval No. (2024) CDYFYYLK (08–056)).

Bioinformatic analyses

GEPIA 3 database (https://gepia3.bioinfoliu.com/) was used to analyze the differential expression of KLF5 in lung cancer and its association with the survival of patients. The binding sites of KLF5 to FASN promoter were predicted by JASPAR database (https://jaspar.elixir.no/). BioGRID database (https://thebiogrid.org/) analyzed the interplay between KLF5 and TAZ proteins.

Cell culture

Normal bronchial epithelial cell line BEAS-2B, lung cancer cells A549, NCI-H1975, NCI-H1299, NCI-H1650, Calu-3, and murine Lewis lung carcinoma cells LLC were all purchased from the American Type Culture Collection (ATCC, Manassas, USA). BEAS-2B cells were maintained with Airway Epithelial Cell Basal Medium (ATCC). A549 cells were maintained with F-12 K (Gibco, New York, USA), NCI-H1975, NCI-H1299, NCI-H1650 cells were cultured with RPMI-1640 (Gibco), and Calu-3 cells were maintained in EMEM (Gibco). LLC cells were cultured with DMEM (Gibco). All cells were maintained with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin at 37 °C with 5% CO₂.

Cell transfection

Short hairpin RNA targeting TAZ (sh-TAZ), shRNA-negative control (sh-NC), overexpression (OE) plasmid for KLF5 (OE-KLF5), and OE-NC were all synthesized by GenePharma (Shanghai, China). Lung cancer cells in 6-well plates (5 × 10⁵ cells per well) were transfected for 48 h with shRNAs or overexpression plasmids using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, USA). Lentiviruses carrying sh-KLF5 or sh-NC were packaged by GenePharma. For lentivirus infection, lung cancer cells were infected with the lentiviruses supplemented with 5 µg/mL polybrene from MedChemExpress (MCE, Monmouth Junction, USA), and then selected by 2.5 µg/mL puromycin (MCE).

Confocal microscopic analysis

A549 and NCI-H1975 cells were seeded into 12-well plates (5 × 10⁴ cells/well). After fixation with 4% paraformaldehyde and permeabilization with 0.5% Triton X-100, the cells were incubated with anti-KLF5 (ab137676, 1:100, Abcam, Cambridge, UK) overnight at 4°C, followed by incubation with goat anti-rabbit IgG (ab150077, 1:200, Abcam) for 1 h and subsequent staining with 4’,6-diamidino-2-phenylindole (DAPI, Solarbio, Beijing, China). Location and expression of KLF5 in lung cancer cells were observed with a confocal microscope (Zeiss, Oberkochen, Germany).

Detection of free fatty acids (FFA)

The level of FFA in lung cancer tissues and cellular supernatant was assessed by the Amplex Red Free Fatty Acid Assay Kit (S0215S, Beyotime, Shanghai, China) following the manufacturer’s specification.

BODIPY 493/503 fluorescence staining

Lipid droplet formation was evaluated by BODIPY 493/503 staining. A549 and NCI-H1975 cells in 12-well plates (5 × 10⁴ cells/well) were fixed with 4% paraformaldehyde, followed by washing with Hank’s solution. Subsequently, the cells were stained with 1 µM BODIPY 493/503 solution (HY-W090090, MCE) at 37 °C for 30 min. The images were observed under a fluorescence microscope.

Conditioned medium (CM) collection

For CM colletion from lung cancer cells, we inoculated the same number of lung cancer cells into culture medium of equal volume. After incubation for the same duration, the supernatants were collected. Cell counting was performed prior to collecting CM, and the CM was diluted with fresh culture medium to achieve an equivalent cell concentration.

The in vitro induction and detection of Treg cells

Peripheral blood mononuclear cells (PBMCs) were isolated from peripheral blood samples of healthy donors using the Ficoll-Paque PREMIUM (Cytiva, San Diego, USA). Then, human CD4⁺ T cells were extracted from PBMCs with the EasySep Human Naïve CD4⁺ T Cell Isolation Kit (19555, STEMCELL Technologies, Vancouver, Canada), followed by activating with 5 µg/mL anti-CD3 (16-0037-81, Thermo Fisher Scientific) and 1 µg/mL anti-CD28 (16-0289-81, Thermo Fisher Scientific). CD4⁺ T cells isolated by magnetic-activated cell sorting (purity > 95%) were stimulated with IL-2 (20 ng/mL, HY-P7037B, MCE) and TGF-β1 (10 ng/mL, HY-P2294, MCE) for 96 h at 37 °C to induce Treg differentiation [35]. For the detection of Treg proportion, CD4⁺ T cells were stained with FITC anti-human CD4 (566911, BD Biosciences, San Jose, USA) and PE anti-human CD25 (567214, BD Biosciences), followed by permeabilization and incubation with APC anti-human Foxp3 (77-5774-40, eBioscience, Santa Clara, USA) for 30 min in the dark. The stained cells were examined by a flow cytometer (BD Biosciences), and Treg proportion was analyzed by the FlowJo v7.6.5 software (BD Biosciences). The percentage of Foxp3⁺ cells in CD4⁺ T cells typically ranged from 25% to 30%.

The sorted Tregs by flow cytometry were treated with 800 µM FFA (oleic acid/palmitic acid at a ratio of 2:1, Sigma-Aldrich, Saint Louis, USA) supplemented with 1% fatty acid-free bovine serum albumin (BSA, Gibco) [36, 37] or CM from lung cancer cells, and then subjected to the following experiments.

Measurement of FA uptake

Tregs were collected and maintained with serum-free medium containing 2 µM BODIPY FL C16 probe (D3821, Thermo Fisher Scientific) for 30 min at 37 °C. Thereafter, the Tregs were detected by a flow cytometer.

Determination of fatty acid oxidation (FAO)

FAO in Tregs was detected by the FAO Assay Kit (BR00001, AmyJet Scientific, Wuhan, China). Briefly, the lysates of Tregs were added with 50 µL of reaction solution, followed by incubation for 60 min at 37 °C. The results were detected at OD 492 nm using a microplate reader (Thermo Fisher Scientific).

Cell counting kit-8 (CCK-8)

Cells were seeded in 96-well plates (4000 cells per well). Then, each well was added with 20 µL of the CCK-8 reagent (CK04, Dojindo, Kumamoto, Japan). After 2 h of incubation, the optical density at 450 nm was measured using a microplate reader.

Dual luciferase reporter assay

The wild-type (WT) or mutant (MUT) sequences of FASN promoter were inserted into the pGL3-basic vector (Promega, Madison, USA). Lung cancer cells were transfected with pGL3-FASN-WT/MUT plasmid, pRL-TK (Promega), together with OE-NC or OE-KLF5 using the Lipofectamine 2000. After incubation for 48 h, the luciferase activity was assessed by the Dual Luciferase Reporter Gene Assay Kit (KGAF040, Keygen, Nanjing, China).

Chromatin immunoprecipitation (ChIP)

ChIP assay was conducted using the ChIP Assay Kit (17–295, Sigma-Aldrich). Initially, the cell lysates were cross-linked with 1% formaldehyde and the chromatin was broken into small fragments by sonication. Then, the DNA fragments were immunoprecipitated with anti-KLF5 (ab314109, Abcam), anti-TAZ (ab307440, Abcam), or anti-IgG (ab172730, Abcam) antibody. The captured DNAs were eluted, digested with protease K and purified. Finally, the enrichment of FASN promoter region was detected by qPCR. The amplicon is located at chr17:82,101,854–82,102,032 (hg38) and encompasses the predicted KLF5-binding sites (GCCCCGCCCC). The primer sequences used for qPCR were as follows: forward, 5’-AAGCATTCCCATCCCGCACA-3’; reverse, 5’-GGGCTGGGACTGAGGAGC-3’.

Electrophoretic mobility shift assay (EMSA)

The purified KLF5 protein was provided by Sino Biological (Beijing, China). The biotin-labeled FASN probe, competitive and unspecific probes were synthesized by Tsingke Biotechnology (Beijing, China). The purified KLF5 protein was combined with the DNA probes using a chemiluminescent EMSA Kit (GS009, Beyotime). Binding complexes were separated by 4% polyacrylamide gel. After transferring to the nylon membrane and cross-linking by UV light, the biotin-labeled DNAs were stained using streptavidin-horseradish peroxidase and BeyoECL Moon provided in the EMSA kit.

Molecular docking analysis

Protein interaction between KLF5 and TAZ proteins was analyzed using molecular docking. Protein structures for KLF5 (PDB: AF-Q13887-F1) and TAZ (also called WWTR1, PDB: AF-Q9GZV5-F1) were downloaded from the PDB protein database. Flexible docking was conducted using the HDOCK server. Visualization was performed using PyMOL (Version 2.4.0).

Co-immunoprecipitation (Co-IP)

The protein interaction between KLF5 and TAZ was analyzed by Co-IP. The protein supernatants were isolated from lung cancer cells using the Cell lysis buffer for western and IP (P0013, Beyotime). Then, anti-KLF5 (ab314109, Abcam), anti-TAZ (ab307440, Abcam) or anti-IgG (ab172730, Abcam) antibody was cross-linked with protein A/G magnetic beads using the Pierce Classic Magnetic IP/Co-IP Kit (88805, Thermo Fisher Scientific). The protein lysates were immunoprecipitated with the antibody-protein A/G complex overnight at 4 °C. The immunoprecipitated proteins were subjected to western blotting.

Animal experiments

All animal procedures were approved by the Animal Care and Use Committee of the First Affiliated Hospital of Nanchang University (Approval No. CDYFY-IACUC-2023022040), and were conducted in accordance with the institution’s guidelines for the care and use of laboratory animals. All procedures were performed in compliance with the national guideline “Guiding Opinions on the Kind Treatment of Laboratory Animals” and the ARRIVE 2.0 guidelines. All mice (4 weeks old, 16–18 g) were obtained from the SLAC Laboratory Animal Center (Shanghai, China) and randomly divided into control, sh-NC, sh-KLF5 groups (n = 8 mice per group). To investigate the regulation of KLF5 in tumor growth and FA synthesis in vivo, 5 × 10⁶ A549 and NCI-H1975 cells infected with lentiviruses carrying sh-NC or sh-KLF5 were subcutaneously injected into the right flanks of male BALB/c nude mice. Tumor volume (V) was monitored and calculated using the formula: V = 0.5×length×width² every five days.

To examine the influence of KLF5 on Treg differentiation in vivo, male C57BL/6J mice were used. LLC cells with or without infection with lentiviruses carrying sh-NC or sh-KLF5 were subcutaneously injected into the mice (1.5 × 10⁵ cells in 100 µL PBS per mouse). All mice were subjected to euthanasia by cervical dislocation after the injection for 25 days, and the tumors were collected for further analyses.

Immunohistochemical staining

The tumor tissues were fixed with 4% paraformaldehyde and embedded in paraffins, followed by cutting into 5 μm sections. After dewaxing and rehydration, antigen retrieval was conducted using citrate buffer and then blocked in 5% goat serum. The sections were incubated with Foxp3 (ab243890, 1:100, Abcam), FASN (ab128870, 1:200, Abcam), KLF5 (ab314109, 1:50, Abcam), and Ki-67 (ab15580, 1:50, Abcam) antibodies overnight at 4 °C. Thereafter, the sections were reacted with the secondary antibody (ab6721, 1:1000, Abcam) at 37 °C for 2 h. After development with DAB (Beyotime), the positive staining was examined under a light microscope (Olympus, Tokyo, Japan) and quantified using Image J software in 3 random fields by a blinded researcher.

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA was extracted using the Trizol reagent (Thermo Fisher Scientific). cDNA was synthesized with the PrimeScript RT reagent Kit (RR037A, Takara, Tokyo, Japan). Then, qPCR was performed with the QuantiNova SYBR Green PCR Kit (208054, QIAGEN, Hilden, Germany) on the 7500 Real-Time PCR system (Applied Biosystems, Beverly, USA). The relative levels of genes normalized to β-actin were analyzed by the 2^−ΔΔCt method.

Western blotting

Protein was extracted using the Cell lysis buffer for western and IP (P0013, Beyotime). The protein concentration was quantified with the BCA Protein Assay Kit (P0012, Beyotime). The protein lysates were separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis and blotted onto the polyvinylidene fluoride membranes. After blocking with 5% skim milk, the membranes were probed with the primary antibodies against KLF5 (ab314109, 1:1000, Abcam), FASN (ab128870, 1:2000, Abcam), SCD1 (ab236868, 1:1500, Abcam), DGAT2 (PA5-103785, 1:5000, Thermo Fisher Scientific), PLIN5 (ab222811, 1:2000, Abcam), Foxp3 (ab215206, 1:1000, Abcam), CD36 (ab252922, 1:2000, Abcam), TAZ (ab307440, 1:1000, Abcam), Lamin B1 (ab16048, 1:1000, Abcam), and β-actin (ab8227, 1:5000, Abcam) overnight at 4 °C. After incubation with the secondary antibody (ab6721, 1:5000, Abcam), the protein bands were detected using the BeyoECL SuperMoon (P0018HM, Beyotime).

Statistical analysis

All cellular experiments were performed with more than three biological replicates. For animal experiments, each group contained eight mice. Data are presented as mean±standard deviation (SD). Statistical analyses were performed using Student’s t-test for two group comparisons or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple group comparisons using GraphPad Prism 8.0. The survival of lung cancer patients was analyzed by Kaplan-Meier method. Pearson correlation analysis was adopted for evaluation of molecular correlation between KLF5 and FA synthesis-related genes. p-value < 0.05 was considered statistically significant.

Results

KLF5 was highly expressed in lung cancer tissues and cells

To analyze the dysregulation of KLF5 in lung cancer, GEPIA 3 database was adopted. We found that KLF5 level was abnormally elevated in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) patients (Fig. 1A). Besides, KLF5 high expression indicated the trend of a lower survival of LUAD patients, not LUSC patients (Fig. 1B). We further validated the higher expression of KLF5 in 50 lung cancer tissues as compared with normal lung tissues (Fig. 1C). Notably, metastatic lung cancer tissues exhibited a higher level of KLF5 in comparison with non-metastatic cases (Fig. 1D-E). Moreover, our clinical data showed that the lung cancer patients with high KLF5 expression exhibited reduced overall survival (Fig. 1F, Hazard Ratio = 2.339; 95% confidence interval = 1.043–5.246). Among the 50 patients, a total of 24 deaths occurred during the 5‑year follow‑up period: 17 in the high KLF5 group and 7 in the low KLF5 group (Fig. 1F). The median follow‑up time was 35 months for the high KLF5 group and 48 months for the low KLF5 group (Fig. 1F). Consistently, KLF5 was up-regulated in multiple lung cancer cells, especially in A549 and NCI-H1975 cells (Fig. 1G). Confocal microscopic analysis indicated that KLF5 was extensively expressed in the cytoplasm and nuclei of A549 and NCI-H1975 cells (Fig. 1H). Collectively, KLF5 was highly expressed in lung cancer.

Fig. 1.

KLF5 was highly expressed in lung cancer tissues and cells. (A) Differential expression of KLF5 in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) was analyzed by GEPIA 3 database (LUAD: n = 515 for tumor, n = 347 for normal; LUSC: n = 498 for tumor, n = 338 for normal). (B) GEPIA 3 database evaluated overall survival of LUAD and LUSC patients with high or low expression of KLF5 (LUAD: n = 252 for low and high KLF5 expression groups; LUSC: n = 246 for low and high KLF5 expression groups). (C) RT-qPCR analysis of KLF5 mRNA level in 50 paired normal and lung cancer tissues. (D) KLF5 mRNA level in non-metastatic (n = 19) or metastatic (n = 31) lung cancer samples was assessed by RT-qPCR. (E) KLF5 expression in normal and lung cancer tissues with or without metastasis was evaluated by immunohistochemical staining. Scale bar = 100 μm. (F) Survival rate of lung cancer patients with high or low expression of KLF5 (n = 25 for low and high KLF5 expression groups partitioning by median). (G) Western blotting analysis of KLF5 expression in various lung cancer cells and normal BEAS-2B cells. (H) Subcellular localization of KLF5 in A549 and NCI-H1975 cells was observed by confocal microscopic analysis. Scale bar = 20 μm. All cellular experiments were performed with more than three biological replicates. *p < 0.05, **p < 0.01, and ***p < 0.001

KLF5 up-regulation indicated high levels of FA metabolism and Treg differentiation in lung cancer tissues

Abnormal activation of FA metabolism has been recognized as a driver of lung cancer [38]. Thus, we evaluated the expression of FA synthesis-related markers in lung cancer samples. As assessed by RT-qPCR, FASN, SCD1, DGAT2, and PLIN5 mRNA levels were aberrantly elevated in lung cancer tissues (Fig. 2A), which were positively correlated with KLF5 level (Fig. 2B). In addition, FFA content was higher in lung cancer tissues than that in normal tissues (Fig. 2C). Furthermore, we observed an elevation in expression of Foxp3, a marker of Tregs [39], in lung cancer samples (Fig. 2D). Besides, the mRNA levels of Foxp3 and CD36 (a transporter of FA [40]) were up-regulated in lung cancer tissues (Fig. 2E). Accordingly, a positive association between KLF5 and Foxp3/CD36 was validated in lung cancer samples (Fig. 2F). These findings indicated that up-regulation of KLF5 had close correlation with FA metabolism and Treg differentiation in lung cancer.

Fig. 2.

KLF5 high expression was correlated with high levels of FA metabolism and Treg differentiation in lung cancer. (A) RT-qPCR analysis of the expression of FA synthesis-related markers FASN, SCD1, DGAT2, and PLIN5 in 50 paired normal and lung cancer tissues. (B) Correlation between KLF5 and FASN/SCD1/DGAT2/PLIN5 in lung cancer tissues was analyzed by Pearson correlation analysis (n = 50). (C) FFA level in normal and lung cancer tissues was determined by a commercial kit (n = 50). (D) Expression of Foxp3 (Tregs marker) in normal and lung cancer tissues was evaluated by immunohistochemical staining. Scale bar = 100 μm. (E) RT-qPCR analysis of Foxp3 and CD36 levels in 50 paired normal and lung cancer tissues. (F) Correlation between KLF5 and Foxp3/CD36 in lung cancer tissues was analyzed by Pearson correlation analysis (n = 50). **p < 0.01 and ***p < 0.001

KLF5 knockdown restrained FA synthesis of lung cancer cells

To explore the influence of KLF5 on FA synthesis, lung cancer cells were infected with lentiviruses containing sh-KLF5. KLF5 expression was efficiently down-regulated in A549 and NCI-H1975 cells of sh-KLF5 group (Fig. 3A-B). As detected by BODIPY 493/503 probe, lipid droplet formation in lung cancer cells was weakened by KLF5 depletion (Fig. 3C). Additionally, the level of FFA was declined in cellular supernatant of KLF5-silenced cells (Fig. 3D). Moreover, KLF5 deficiency remarkably down-regulated FASN, SCD1, DGAT2, and PLIN5 expression in lung cancer cells (Fig. 3E). Therefore, KLF5 silencing effectively repressed FA synthesis of lung cancer cells.

Fig. 3.

KLF5 down-regulation reduced FA synthesis of lung cancer cells. Lung cancer cells were infected with lentiviruses carrying sh-KLF-5 or sh-NC for 48 h. (A-B) The mRNA and protein levels of KLF5 were measured by RT-qPCR (A) and western blotting (B). (C) Lipid droplet formation in lung cancer cells was evaluated by BODIPY 493/503 probe. Scale bar = 100 μm. (D) FFA level in the supernatant of lung cancer cells was assessed by a commercial kit. (E) Western blotting analysis of FASN, SCD1, DGAT2, and PLIN5 protein levels in lung cancer cells. All cellular experiments were performed with more than three biological replicates. *p < 0.05, **p < 0.01, and ***p < 0.001

KLF5 deficiency delayed lung cancer growth and FA synthesis in vivo

To validate whether KLF5 affected lung cancer growth and FA metabolism in vivo, the BALB/c nude mouse model was established. We found that in vivo tumor growth was strikingly delayed by KLF5 silencing (Fig. 4A-C). Similarly, the expression of proliferation index Ki-67 was reduced in KLF5-deficient tumor tissues (Fig. 4D-E). Consistent with the in vitro data, KLF5 silencing evidently reduced the expression of FASN, SCD1, DGAT2, and PLIN5 in tumors (Fig. 4F-G, and I). Besides, the FFA level was lowered in tumor tissues when KLF5 was down-regulated (Fig. 4H). These results implied that KLF5 depletion inhibited in vivo lung cancer growth by repressing FA synthesis.

Fig. 4.

KLF5 knockdown repressed lung cancer growth and FA synthesis in vivo. A549 and NCI-H1975 cells infected with lentiviruses carrying sh-NC or sh-KLF5 were subcutaneously injected into BALB/c nude mice. Experiment groups: control, sh-NC, sh-KLF5. (A) Images for tumors collected from nude mice. (B-C) Tumor volume and weight were monitored. (D-E) Immunohistochemical staining detected Ki-67 expression in tumor tissues. Scale bar = 50 μm. (F-G) FASN expression in tumors was evaluated by immunohistochemical staining. Scale bar = 50 μm. (H) FFA level in tumor tissues was detected by a commercial kit. (I) FASN, SCD1, DGAT2, and PLIN5 protein abundance in tumors was determined by western blotting. For animal experiments, each group contained eight mice. ***p < 0.001

Loss of KLF5 in lung cancer cells repressed FFA-induced Treg differentiation

Given that KLF5 silencing suppressed FA synthesis in lung cancer cells, we further investigated its regulation in Treg differentiation mediated by FFA. For this purpose, naive CD4⁺ T cells were cultured under Treg-inducing condition, which were then treated with FFA or CM derived from lung cancer cells. Tregs were sorted by flow cytometry and results indicated that the proportion of differentiated Tregs was enhanced by FFA or CM treatment, whereas CM from KLF5 depleted-lung cancer cells attenuated this change (Fig. 5A-B). Meanwhile, treatment with FFA or CM distinctly increased FA uptake (Fig. 5C-D) and FAO level (Fig. 5E) in sorted Tregs; however, loss of KLF5 counteracted these alterations (Fig. 5C-E). Foxp3 and CD36 protein levels in sorted Tregs were raised by FFA or CM treatment, which could be attenuated by KLF5 knockdown (Fig. 5F). CCK-8 results demonstrated that Treg proliferation was not affected by FFA or CM from different groups (Fig. 5G). These findings suggested that KLF5 knockdown in lung cancer cells inhibited FFA-induced Treg differentiation in vitro.

Fig. 5.

KLF5 silencing inhibited FFA-induced Treg differentiation. Lung cancer cells were infected with lentiviruses carrying sh-KLF-5 or sh-NC. CD4⁺ T cells were maintained with Treg-inducing medium and treatment with FFA or indicated lung cancer cells-derived CM, Tregs were sorted by flow cytometry. (A-B) Percentage of Tregs (CD4⁺CD25⁺Foxp3⁺) was evaluated by flow cytometry. (C-D) The uptake of FA in sorted Tregs was analyzed by BODIPY FL C16 probe and flow cytometry. (E) FAO level in sorted Tregs was detected by a commercial kit. (F) The protein levels of Foxp3 and CD36 in sorted Tregs were determined by western blotting. (G) CCK-8 assay detected cell proliferation of sorted Tregs. (H) Flow cytometry analyzed the proportion of Tregs (CD4⁺CD25⁺Foxp3⁺) in subcutaneous tumors of C57BL/6J mice injected with LLC cells. (I) Western blotting analysis of Foxp3 and CD36 protein levels in sorted Tregs from subcutaneous tumors of C57BL/6J mice injected with LLC cells. (J) FASN, SCD1, DGAT2, and PLIN5 protein abundance in tumor issues was determined by western blotting. All cellular experiments were performed with more than three biological replicates. For animal experiments, each group contained eight mice. *p < 0.05, **p < 0.01, and ***p < 0.001

Next, we investigated whether KLF5 knockdown affected Treg differentiation in C57BL/6J mice in vivo. Notably, the proportion of Tregs in tumor tissues of C57BL/6J mice injected with LLC cells was reduced when KLF5 was silenced (Fig. 5H). Additionally, KLF5 ablation also down-regulated Foxp3 and CD36 levels in sorted Tregs from tumors of C57BL/6J mice (Fig. 5I). Besides, KLF5 deficiency reduced FASN, SCD1, DGAT2, and PLIN5 protein abundance in tumor tissues (Fig. 5J). To sum up, KLF5 deficiency inhibited Treg differentiation via inactivation of FA metabolism in C57BL/6J mice.

KLF5 promoted nuclear translocation of TAZ to trigger FASN transcription

To elucidate the potential modulatory mechanism of KLF5 in FA metabolism, JASPAR database predicted that KLF5 might bind to the promoter of FASN (Fig. 6A). The predicted KLF5-binding motif, the specific binding sequence in FASN promoter, and the MUT sequence used in dual luciferase reporter assay were also shown (Fig. 6A). KLF5 overexpression effectively enhanced the luciferase activity of FASN-WT group, but not FASN-MUT group (Fig. 6B). Additionally, ChIP assay showed that the enrichment of FASN promoter region was strikingly enhanced after precipitation with KLF5 antibody (Fig. 6C). Moreover, EMSA validated the binding of KLF5 protein to biotin-labeled FASN probe (Fig. 6D). These results confirmed the binding of KLF5 to FASN promoter. Furthermore, FASN expression in lung cancer cells was down-regulated by KLF5 silencing, but up-regulated by KLF5 overexpression (Fig. 6E-F). BioGRID database predicted that KLF5 protein might bind to TAZ protein. Molecular docking further proved TAZ as a binding protein with KLF5 protein (Fig. 6G). Interestingly, Co-IP assay proved that KLF5 protein interacted with TAZ protein in lung cancer cells (Fig. 6H-I). In addition, KLF5 overexpression-mediated enhancement in luciferase activity of FASN promoter was partly reversed by TAZ knockdown (Fig. 6J), indicating that KLF5 facilitated FASN transcription in a TAZ-dependent manner. Of note, nuclear expression of TAZ was higher, while cytoplasmic expression of TAZ was lower in lung cancer cells as compared with normal bronchial epithelial cell line BEAS-2B (Fig. 6K). Besides, nuclear translocation of TAZ was restrained in KLF5-silenced lung cancer cells, but intensified in KLF5-overexpressed cells (Fig. 6L). In addition, the binding of TAZ to the FASN promoter was enhanced by KLF5 overexpression, but weakened by KLF5 knockdown (Fig. 6M). Taken together, KLF5 bound to FASN promoter to facilitate its transcription via promoting nuclear translocation of TAZ.

Fig. 6.

KLF5 facilitated nuclear translocation of TAZ to drive transcription of FASN. (A) JASPAR database predicted the binding of KLF5 to FASN promoter, the predicted KLF5-binding motif, the specific binding sequence in FASN promoter, and the MUT sequence used in dual luciferase reporter assay were also shown. (B-D) The binding of KLF5 to FASN promoter was validated by dual luciferase reporter (B), ChIP (C) and EMSA (D) assays. (E-F) FASN expression in lung cancer cells transfected with sh-KLF5 or KLF5 overexpression vector was analyzed by RT-qPCR (E) and western blotting (F). (G) Molecular docking analysis of the binding of KLF5 protein to TAZ protein using HDOCK server. (H-I) The interaction between KLF5 and TAZ proteins was confirmed by Co-IP. (J) Lung cancer cells were transfected with KLF5 overexpression vector together with or without sh-TAZ vector, and transcriptional activity of FASN promoter was detected by dual luciferase reporter assay. (K) The nuclear and cytoplasmic expression of TAZ in BEAS-2B and lung cancer cells was determined by western blotting. (L) The nuclear and cytoplasmic expression of TAZ in KLF5-silenced or overexpressed lung cancer cells was determined by western blotting. (M) The binding of TAZ to FASN promoter was validated by ChIP after KLF5 silencing or overexpression. All cellular experiments were performed with more than three biological replicates. *p < 0.05, **p < 0.01, and ***p < 0.001

KLF5 contributed to FA synthesis with assistance of TAZ

To validate the involvement of KLF5/TAZ axis in FA synthesis, lung cancer cells were transfected with KLF5 overexpression vector together with sh-TAZ vector. We found that KLF5 overexpression evidently increased FASN protein abundance, which was abrogated by TAZ depletion (Fig. 7A). However, KLF5 and TAZ had no influence on each other (Fig. 7A). KLF5 overexpression-mediated up-regulation of FASN mRNA was also reversed by silencing of TAZ (Fig. 7B). Additionally, the promotive effect of KLF5 overexpression on lipid droplet formation in lung cancer cells was attenuated by TAZ silencing (Fig. 7C). Accordingly, TAZ ablation reversed the elevated FFA content in the supernatant of lung cancer cells triggered by KLF5 overexpression (Fig. 7D). Furthermore, up-regulation of SCD1, DGAT2, and PLIN5 in KLF5-overexpressed lung cancer cells was neutralized by TAZ knockdown (Fig. 7E). Thus, KLF5 facilitated FA synthesis of lung cancer cells in a TAZ-dependent manner.

Fig. 7.

KLF5/TAZ axis contributed to FA synthesis in lung cancer cells. Lung cancer cells were transfected with KLF5 overexpression vector in combination with or without sh-TAZ vector. (A) The protein levels of KLF5, TAZ, and FASN were analyzed by western blotting. (B) RT-qPCR analysis of FASN mRNA in lung cancer cells. (C) BODIPY 493/503 probe determined lipid droplet formation in lung cancer cells. Scale bar = 100 μm. (D) FFA level in the supernatant of lung cancer cells was measured by a commercial kit. (E) Western blotting analysis of SCD1, DGAT2, and PLIN5 protein levels in lung cancer cells. All cellular experiments were performed with more than three biological replicates. *p < 0.05, **p < 0.01, and ***p < 0.001

KLF5/TAZ axis-mediated FA synthesis favored Treg differentiation

Finally, we investigated whether KLF5/TAZ axis-mediated FA synthesis influenced Treg differentiation. Treg cells were sorted by flow cytometry and we found that CM from KLF5 overexpression group strikingly raised Treg percentages, which was reduced by TAZ deficiency (Fig. 8A). FA uptake by sorted Tregs was enhanced by treatment with CM from KLF5-overexpressed lung cancer cells, which was weakened by TAZ silencing (Fig. 8B). FAO in sorted Tregs was heightened by CM of lung cancer cells with KLF5 overexpression, whereas depletion of TAZ in lung cancer cells counteracted this change (Fig. 8C). Moreover, Foxp3 and CD36 were up-regulated in sorted Tregs after treatment with KLF5-overexpressed CM; however, this action was abrogated by loss of TAZ (Fig. 8D). These observations suggested that activation of KLF5/TAZ axis promoted Treg differentiation via enhancing FA synthesis in lung cancer cells.

Fig. 8.

KLF5 facilitated FA synthesis to induce Treg differentiation in a TAZ-dependent manner. Lung cancer cells were transfected with KLF5 overexpression vector in combination with or without sh-TAZ vector. CD4⁺ T cells were maintained with Treg-inducing medium and treatment with FFA or indicated lung cancer cells-derived CM, Tregs were sorted by flow cytometry. (A) Percentage of Tregs (CD4⁺CD25⁺Foxp3⁺) was measured by flow cytometry. (B) The uptake of FA by sorted Tregs was evaluated by BODIPY FL C16 probe and flow cytometry. (C) FAO level in sorted Tregs was assessed by a commercial kit. (D) The protein levels of Foxp3 and CD36 in sorted Tregs were analyzed by western blotting. All cellular experiments were performed with more than three biological replicates. *p < 0.05, **p < 0.01, and ***p < 0.001

Discussion

In addition to classical oncogenic signaling, metabolism-associated pathways have emerged as important contributors to lung cancer progression [41, 42]. Tregs can accelerate the development of lung cancer, which have been considered as a huge obstacle for treatment [8]. A previous study demonstrated that suppression of lipid metabolism inhibited functional maturation of Tregs, thus boosting immunotherapy and inhibiting tumor growth [15]. However, the molecular mechanism underlying FA metabolism and Treg differentiation in lung cancer remains largely unknown. In this study, we discovered that high expression of KLF5 was positively correlated with FA metabolism and Treg differentiation in lung cancer tissues. Deficiency of KLF5 restrained FA synthesis in lung cancer cells to inhibit Treg differentiation, which delayed lung cancer growth. Mechanistically, KLF5 interplayed with TAZ protein to facilitate its nuclear translocation, and subsequently activated transcription of FASN to trigger FA synthesis in lung cancer cells, which enhanced CD36-mediated FA uptake by Tregs, thereby favoring FAO and Treg differentiation. Our findings indicated that targeting KLF5/TAZ/FASN axis might be a new avenue to fight against lung cancer via repressing FA metabolism and Treg differentiation.

Recent studies have increasingly focused on identifying novel molecular targets in lung cancer, highlighting the therapeutic value of dissecting cancer-specific regulatory pathways [43–45]. KLF5 is a zinc finger transcription factor that has been recognized to be an oncogene for lung cancer [46]. A previous work identified KLF5 as a modulator for cardiac FA metabolism [47]. Notably, Zuo et al. reported that KLF5 facilitated lipid droplet formation to reduce oxaliplatin sensitivity of colorectal cancer cells by promoting FABP6 transcription [48]. In addition, KLF5 was responsible for PLA2G16 transcription that increased FFA production, thus promoting growth in pancreatic adenocarcinoma [49]. These literature reports demonstrate that KLF5 plays a critical regulatory role in FA metabolism in other diseases, however, the involvement of KLF5 in FA metabolism in lung cancer has not been reported. Herein, we discovered that up-regulation of KLF5 was positively correlated with high level of FA synthesis in lung cancer samples. Functionally, KLF5 deletion reduced FA synthesis and production of FFA, and consequently restrained tumor growth in lung cancer cells and BALB/c nude mouse model. Our findings provided evidence for the contribution of KLF5 to FA metabolism in lung cancer cells.

The tumor microenvironment constitutes a complex ecosystem composed of tumor cells, stromal cells, and immune cells [50]. Interaction between these cells plays pivotal roles in promoting tumor development [51]. The influence of tumor cells on the biological functions of immune cells has become a hot topic in recent years. Therefore, we further explored the effect of KLF5-mediated FA synthesis and secretion in lung cancer cells on FA uptake and differentiation of Tregs. Tregs are a major sub-group of CD4⁺ T cells with expression of Foxp3 [52]. It was suggested that FAO not only provided energy, but also facilitated differentiation of Tregs [53]. Emerging evidence proved that FA metabolism had a pivotal impact on Treg differentiation [11]. For instance, Wang et al. documented that CD36, a FA transporter, supported Treg function in tumor microenvironment [12]. Of note, FA synthesis promoted by oncogenes in tumors contributed to Treg-mediated immune escape [54]. Our data demonstrated that KLF5 level was positively correlated with the percentages of Tregs in lung cancer tissues. More importantly, we provided evidence that KLF5 enhanced FA metabolism of lung cancer cells, which favored CD36-mediated FA uptake by Tregs to facilitate FAO and consequent Treg differentiation. It is worth noting that conditioned medium derived from lung cancer cells with KLF5 knockdown did not affect the viability of Treg cells. Therefore, these results uncovered a novel mechanism whereby KLF5 favored FA metabolism to enforce Treg differentiation in lung cancer microenvironment.

Next, we further explored the mechanism by which KLF5 promoted FA synthesis and secretion in lung cancer cells. Emerging documents proved that KLF5 functioned as a transcription factor that induced the transcription and expression of target genes [55]. It was reported that suppression of FASN-mediated FA metabolism effectively delayed the progression of lung adenocarcinoma [56]. Herein, we identified that KLF5 enhanced FASN transcription and expression via binding to its promoter in lung cancer cells. TAZ has been shown to provoke immunosuppression of non-small cell lung cancer [57]. Interestingly, TAZ nuclear localization has been recognized to promote FA metabolism in cancers [29]. Promotion of nuclear localization and activation of TAZ has been validated to drive tumorigenesis [58]. To date, no studies have reported the roles of TAZ in lipid metabolism and Treg regulation in lung cancer. In this work, KLF5 interacted with TAZ to induce its nuclear localization to activate FASN transcription, thus inducing Treg differentiation via reprogramming FA metabolism. Our findings provide evidence for the role of KLF5/TAZ/FASN axis in lipid metabolism and Treg differentiation in lung cancer. This study did not determine whether overexpression of TAZ could reverse the role of KLF5 knockdown, which needs to be verified in the future. But, we speculate that under KLF5 knockdown condition, overexpression of TAZ may not be able to significantly reverse the phenotypic effects, because KLF5 functions as a transcription factor whereas TAZ acts only as a transcriptional co-activator. Although KLF5 function depends on TAZ, when KLF5 expression is markedly reduced, the co-activator function of TAZ would be limited and thus unlikely to exert a strong rescue effect. Notably, reduced nuclear translocation of TAZ after KLF5 silencing does not necessarily mean that TAZ loses all transcriptional activity or cannot bind other nuclear transcription factors. TAZ may still regulate other lipogenic genes through KLF5-independent partners.

Of note, several YAP/TAZ inhibitors have entered clinical trials for the treatment of human cancers [59], highlighting the clinical translatability of targeting the KLF5/TAZ axis. Furthermore, nanotechnology offers a promising platform for the co‑delivery of therapeutic agents and genetic materials to tumor cells [60], offering new avenues for combinatorial regimens. Accordingly, integrating KLF5/TAZ‑targeted strategies with existing inhibitors may potentiate anti-tumor efficacy and provide added clinical benefit for lung cancer patients.

There are several limitations of our work. First, we demonstrated that KLF5 promoted FASN transcription with assistance of TAZ, which supports one possible KLF5-associated mechanism. Whether other targets such as SCD1, DGAT2, and PLIN5 are involved in the overall phenotypes induced by KLF5 or TAZ remains unclear. Second, molecular docking and Co-IP were performed to validate KLF5 and TAZ protein interaction; however, their function still needs to be validated using mutant constructs. Third, subcutaneously implanted tumor models in nude and C57BL/6J mice were established in this study. Orthotopic lung or metastatic models can better recapitulate the tumor microenvironment. Thus, our results are still required to be validated in orthotopic lung or metastatic models. Fourth, experiments of inhibiting fatty acid synthesis, FAO, or CD36 were not performed due to time constraints and limited Tregs sorted from human peripheral blood, this warrants further in‑depth investigation. Furthermore, Seahorse analysis using palmitate as a substrate and measuring etomoxir‑sensitive oxygen consumption would provide more convincing evidence, and this will be addressed in future studies. Fifth, this study primarily focused on the differentiation mechanisms of Tregs rather than their immunosuppressive capacity. Accordingly, functional suppression assays and analyses of CD8⁺ T cells were not performed, representing a promising direction for future research. Sixth, our findings indicated a positive correlation between KLF5 and Foxp3/CD36 in lung tumor tissues, which is not sufficient to prove tumor cell-Treg crosstalk. These correlations may only reflect tumor purity, stromal content, or immune infiltration. Seventh, Tregs were isolated from peripheral blood samples of healthy donors, but not from lung cancer patients, which is another limitation of our study. Finally, functional roles were assessed through knockdown and overexpression experiments in lung cancer cells; however, the clinical examination of overexpression and knockdown techniques is required for clinical translation of these findings.

Conclusion

Taken together, KLF5 exhibited significantly higher expression level in lung cancer tissues, which was positively correlated with elevated FA metabolism and enhanced Treg differentiation in patients. Functionally, KLF5 knockdown restrained FA synthesis in lung cancer cells to inhibit FA uptake and differentiation of Tregs, thereby slowing tumor growth. Mechanistically, KLF5 promoted TAZ nuclear localization to contribute to FASN-mediated FA synthesis in lung cancer cells. Our findings offer a theoretical foundation for the regulation of KLF5/TAZ/FASN axis in FA metabolism-induced Treg differentiation in lung cancer. However, whether targeting this pathway may protect against immune escape in lung cancer is a future research direction.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Hongxiang Huang: conceptualization, investigation, validation, visualization, writing—original draft. Jinhong Chen: investigation, methodology, validation, formal analysis, writing—original draft. Peiyuan Zhong: formal analysis, software, data curation. Sujuan Peng: investigation, methodology. Qiuxa Zhong & Chenfeiyang Deng: investigation. Fen Wang: resources (providing clinical samples). Zhihui Lu: resources (providing reagents/materials). Yangyang Liu: funding acquisition, project administration, resources, supervision, writing—review & editing. Li Chen: conceptualization, funding acquisition, project administration, supervision, writing—review & editing.

Funding

This work was supported by grant from the Natural Science Fund for Youths of Jiangxi Province (No.20224BAB216067), the Youth Program of Ganzhou Sci-Tech and Medical Joint Program (2025YLCE0054) and the Regional Programs of the National Natural Science Foundation of China (Grant Nos. 82560474, 82560461, and 82360507).

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

This study involving human participants was conducted in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments. The use of clinical lung cancer and paired normal tissues was approved by the Ethics Committee of the First Affiliated Hospital of Nanchang University (Approval No. (2024) CDYFYYLK (08–056)), and written informed consent was obtained from all patients prior to sample collection. All animal experiments were performed in compliance with the National Guidelines for the Care and Use of Laboratory Animals issued by the Ministry of Science and Technology of China (2006) and the Guiding Opinions on the Kind Treatment of Laboratory Animals (MOST, 2006). The study protocol was approved by the Institutional Animal Care and Use Committee of the First Affiliated Hospital of Nanchang University (Approval No. CDYFY-IACUC-2023022040) and reported in accordance with the ARRIVE 2.0 guidelines.

Conflict of interest

The authors declare that they have no conflict of interest.