Effects of CDDO‑TFEA on non‑small cell lung cancer growth, apoptosis, and stemness via the MEK/ERK pathway

Abstract

Purpose

Non-small cell lung cancer (NSCLC) remains a leading cause of cancer-related mortality, and therapeutic resistance limits durable responses. We investigated the antitumor activity of CDDO-TFEA, a synthetic trifluoroethyl amide derivative of 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid, in NSCLC models.

Methods

A549 and H1299 cells were treated with CDDO-TFEA and evaluated using MTT assays, Annexin V/7-aminoactinomycin D flow cytometry, tumorsphere and tube-formation assays, immunoblotting, and A549 xenograft experiments.

Results

CDDO-TFEA decreased cell viability in a dose-dependent manner and induced apoptosis, as indicated by increased Annexin V positivity, PARP cleavage, cleaved caspase-3 expression, and TUNEL positivity. At 200 nM, CDDO-TFEA significantly suppressed tumorsphere formation in both cell lines and reduced CD133, octamer-binding transcription factor 4, SRY-box transcription factor 2, and vascular endothelial growth factor expression. Because only A549 cells formed tube-like networks, angiogenesis-related phenotypes were evaluated in A549 cells and were markedly inhibited by CDDO-TFEA. Mechanistically, CDDO-TFEA increased extracellular signal-regulated kinase phosphorylation, whereas MEK inhibition by U0126 or trametinib partially attenuated CDDO-TFEA–induced apoptosis and the suppression of tumorsphere formation, tube-like network formation, and stemness- and angiogenesis-related marker expression. In vivo, systemic CDDO-TFEA administration dose-dependently inhibited xenograft tumor growth.

Conclusion

CDDO-TFEA exerts antitumor activity in NSCLC by inducing apoptosis and suppressing stemness-associated and angiogenesis-related phenotypes, partly through mitogen-activated protein kinase kinase/extracellular signal-regulated kinase signaling. These findings support further preclinical development of CDDO-TFEA in NSCLC.

Supplementary Information

The online version of this article (https://doi.org/10.1007/s00432-026-06544-3) contains supplementary material, which is available to authorized users.

Introduction

Lung cancer is one of the most frequently diagnosed malignancies and remains the leading cause of cancer-related mortality worldwide. Non–small cell lung cancer (NSCLC) accounts for 80–85% of lung cancer cases, whereas small cell lung cancer (SCLC) represents only 10–15% (Bray et al. 2024). NSCLC is characterized by aggressive clinical behavior and a poor overall prognosis and contributes substantially to the global cancer burden (Hendriks et al. 2024). Current therapeutic strategies for NSCLC include surgical resection, radiotherapy, chemotherapy, and molecularly targeted therapies. Surgical resection is primarily suitable for patients with early-stage disease, whereas radiotherapy and chemotherapy constitute the main treatment options for patients with locally advanced or unresectable tumors (Hou et al. 2025). Despite the availability of these multimodal treatments, long-term survival outcomes for patients with NSCLC remain unsatisfactory (Li et al. 2023; The 2024).

The identification of epidermal growth factor receptor (EGFR) gene mutations as critical oncogenic drivers of tumor initiation and progression constitutes a major breakthrough in NSCLC treatment. This discovery led to the development of several EGFR tyrosine kinase inhibitors (EGFR-TKIs), including gefitinib and osimertinib, which have significantly improved clinical outcomes in selected patients with EGFR mutations (Ohashi et al. 2013). However, both intrinsic and acquired resistance to EGFR-TKIs inevitably emerge during treatment, severely limiting long-term therapeutic efficacy and contributing to disease relapse and progression. In parallel, conventional chemotherapy remains one of the most commonly used treatments for NSCLC. Cisplatin—a potent, non–specific cell-cycle blocker—is extensively employed in the treatment of various cancers, including NSCLC (Eljack et al. 2014). However, major limitations of conventional chemotherapeutic agents, such as poor tumor selectivity, low bioavailability, and severe systemic toxicity, together with the activation of complex adaptive responses in cancer cells, enable tumors to evade cytotoxic effects, resume proliferation, and ultimately develop drug resistance (Kryczka and Boncela 2018; Kryczka et al. 2021). These resistance mechanisms represent a critical cause of treatment failure in NSCLC and are closely associated with tumor recurrence and disease progression (He et al. 2020; Sosa Iglesias et al. 2018). Therefore, there is an urgent and unmet clinical need to identify novel therapeutic agents and strategies to overcome drug resistance and improve treatment outcomes for patients with NSCLC (Zhu et al. 2021).

A synthetic trifluoroethyl amide derivative (TFEA) of the oleanane triterpenoid 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid (CDDO), which belongs to a class of highly bioactive small molecules that were originally developed for their potent anti-inflammatory and cytoprotective properties, has been developed. Structurally modified to enhance stability and biological activity, CDDO-based compounds exert their effects primarily through the activation of the nuclear factor erythroid 2–related factor 2 (NRF2) signaling pathway. Mechanistically, CDDO derivatives covalently modify critical cysteine residues of Kelch-like ECH-associated protein 1 (Keap1), and thereby disrupt Keap1-mediated ubiquitination and degradation of NRF2. The stabilized NRF2 subsequently undergoes nuclear translocation and induces the transcription of antioxidant and detoxification genes, including heme oxygenase-1 (HO-1) and NAD(P)H quinone dehydrogenase 1 (NQO1), and thereby confers robust cellular protection against oxidative stress (Ahmad et al. 2006; Casares et al. 2022). Beyond its known cytoprotective role, the accumulating evidence indicates that CDDO-derived compounds exhibit pronounced anti-inflammatory and anticancer activities. These effects are mediated not only through NRF2 activation but also via the suppression of pro-tumorigenic signaling pathways, such as NF-κB, which decreases the expression of inflammatory cytokines, inhibits cell proliferation, and induces apoptosis in multiple cancer models (Ahmad et al. 2006; Tsai et al. 2021). Crucially, CDDO analogs impair tumor cell migration, invasion, and metastatic potential, and this highlights their therapeutic promise as multitarget anticancer agents that integrate redox regulation with growth-suppressive signaling (Casares et al. 2022; Tsai et al. 2021). Although CDDO-derived compounds have been widely studied as NRF2-modulating agents, the present study focused on the antitumor phenotypes and MEK/ERK-associated signaling responses of CDDO-TFEA in NSCLC cells.

Despite these encouraging findings, the biological role and therapeutic potential of CDDO-TFEA in NSCLC remain largely unexplored. Considering the critical roles of oxidative stress, chronic inflammation, and adaptive survival signaling in NSCLC progression and treatment resistance, the elucidation of the effects of CDDO-TFEA in the NSCLC context is of particular interest. Therefore, this study aimed to investigate the antitumor efficacy and underlying molecular mechanisms of CDDO-TFEA in NSCLC.

Materials and methods

Cell lines and cell culture

Human NSCLC cell lines A549 (BCRC; 60074; Taiwan) and H1299 (ATCC; CRL-5803; USA) were maintained in a humidified incubator at 37 °C with 5% CO₂. Cells were cultured in RPMI-1640 (Gibco; 31800–022; USA) and DMEM (Gibco; 31,600–034; USA) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich; FC926; USA) and 1% Pen-Strep-Ampho.B Solution (biorion; AACR0003B; Taiwan). Cells were routinely passaged at 70–80% confluence and confirmed to be mycoplasma-free prior to experiments. All experiments were performed using cells within a limited passage range. The identity of A549 and H1299 cell lines was authenticated by short tandem repeat profiling prior to experimentation.

Reagents and antibodies

CDDO-TFEA (MCE; TP-500; USA) was prepared as a 10 mM stock solution in DMSO (Sigma; D2650; USA). The maximum final DMSO concentration introduced by CDDO-TFEA in cell-based assays was 0.02% (v/v), corresponding to the highest CDDO-TFEA concentration tested (2000 nM). Vehicle controls contained an equivalent amount of DMSO. For co-treatment experiments, the total DMSO concentration was matched across all groups, including vehicle, CDDO-TFEA, and CDDO-TFEA plus MEK inhibitor groups. U0126 (MCE; HY-12031A; USA) and trametinib (MCE; JTP-74057; USA) were dissolved in DMSO and used as MEK inhibitors. Based on cell viability screening, both U0126 and trametinib were used at 10 μM in rescue experiments. Primary antibodies used for immunoblotting included PARP (1:500; 13371–1-AP; Proteintech; USA), CD133 (1:200; #64326; Cell Signaling Technology; USA), OCT4 (1:500; 11263–1-AP; Proteintech; USA), SOX2 (1:500; 11064–1-AP; Proteintech; USA), VEGF (1:500; AP6290b; ABGENT; USA), phospho-ERK (1:500; 9101; Cell Signaling Technology; USA), total ERK (1:1000; 9102; Cell Signaling Technology; USA), phospho-JNK (1:500; 4668; Cell Signaling Technology; USA), total JNK (1:1000; 9252; Cell Signaling Technology; USA), phospho-p38 (1:500; 9211; Cell Signaling Technology; USA), total p38 (1:1000; 9212; Cell Signaling Technology; USA), and cleaved caspase-3 (1:200; 9661; Cell Signaling Technology; USA).

Cell viability assay

Cells were seeded in 24-well plates at 3 × 10⁴ cells per well and allowed to attach overnight. Cells were then treated with the indicated concentrations of CDDO-TFEA for 24 h. Cell viability was assessed using an MTT assay (Thermo; L11939.06; USA) according to the manufacturer’s instructions. Absorbance was recorded using a microplate reader, and cell viability was calculated as a percentage relative to the vehicle control (set to 100%).

Flow cytometry-based analysis of apoptosis

Apoptosis was evaluated using the Muse Annexin V & Dead Cell Kit (Cytek; MCH100105; USA) and the Muse Cell Analyzer according to the manufacturer’s instructions. After treatment with CDDO-TFEA (0–1600 nM) or the indicated combinations with MEK inhibitors, cells were harvested, washed with PBS, and resuspended as single-cell suspensions. Equal volumes of cell suspension and Muse Annexin V & Dead Cell Reagent were mixed and incubated for 20 min at room temperature in the dark before analysis. Data were acquired and analyzed using the Muse Annexin V & Dead Cell software module. Cell populations were classified according to the manufacturer’s predefined analysis algorithm as live cells (Annexin V − /7-AAD −), early apoptotic cells (Annexin V + /7-AAD −), late apoptotic/dead cells (Annexin V + /7-AAD +), and dead/debris cells (Annexin V − /7-AAD +). Total apoptosis was calculated as the sum of early and late apoptotic populations. Because the kit uses a combined ready-to-use Annexin V/dead cell reagent and is analyzed using the dedicated Muse software module, separate single-stained Annexin V and 7-AAD compensation controls were not generated. For MEK-inhibition experiments, cells were pretreated with U0126 or trametinib, both at 10 μM, for 60 min before CDDO-TFEA treatment.

Immunoblotting analysis

Following drug treatments, the cells were washed with ice-cold PBS and lysed in RIPA buffer (Millipore; 20–188; USA) containing protease (Roche; 04693159001; USA) and phosphatase inhibitors (Roche; 04906845001; USA). The protein concentrations were determined using a BCA assay (BC03-500; Visual Protein; Taiwan). Equal amounts of protein were resolved by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes (Neobits; NEF1002001PK; USA). The membranes were blocked (5% BSA in TBST) and incubated with primary antibodies at 4 °C overnight, followed by incubation with HRP-conjugated secondary antibodies for 90 min. Next, protein bands were visualized using ECL and imaged using a chemiluminescence-detection system. Densitometric analysis was performed using ImageJ software. For MAPK signaling analysis, phosphorylated ERK, JNK, and p38 were normalized to their corresponding total protein levels. For apoptosis- and stemness-associated immunoblots, protein expression was normalized to total lane protein signals obtained by Ponceau S staining. Briefly, after protein transfer, PVDF membranes were stained with Ponceau S solution, imaged to record total protein loading, and then destained before blocking and antibody incubation.

Tumorsphere formation assay

To assess stemness-associated phenotypes, tumorsphere formation assays were performed in A549 and H1299 cells. Cells were seeded at 100 cells/well in ultra-low attachment plates in serum-free sphere-forming medium including 10 ng/mL EGF (Sigma; SRP3027; USA), 10 ng/mL bFGF (Sigma; F0291; USA), and B-27 (Thermo Fisher; 17,504,001; USA). Cells were treated with vehicle or CDDO-TFEA (200 nM), with or without U0126 or trametinib (10 μM) as indicated. After 10 days of incubation, primary spheres were imaged under an inverted microscope. For secondary tumorsphere formation, primary spheres were dissociated and replated under the same sphere-forming conditions. Sphere formation was quantified using ImageJ by measuring sphere number and sphere diameter.

Tube-formation assay

A tube-formation assay was conducted to evaluate tube-like network formation capacity. Under the experimental conditions that were used, only A549 cells formed tube-like structures; therefore, tube-formation experiments were performed using A549 cells only. Matrigel (ECM; Sigma; E1270; USA) was added to pre-chilled plates at 100 μL/well and allowed to polymerize at 37 °C. A549 cells were seeded onto the solidified matrix at a density of 1 × 10⁴ cells/well in appropriate medium containing vehicle, CDDO-TFEA (200 nM), CDDO-TFEA plus U0126 (10 μM), or CDDO-TFEA plus trametinib (10 μM), as indicated. Cells were incubated for 4 h to allow network formation. Images were captured, and tube-like network formation was quantified using the ImageJ Angiogenesis Analyzer plugin by measuring the number of nodes, total tube length, mesh count, and branching index.

TUNEL fluorescence staining

Apoptotic DNA fragmentation was evaluated using terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) fluorescence staining (Roche; 12156792910; Switzerland). For dose–response experiments, A549 and H1299 cells were treated with CDDO-TFEA (0, 400, 800, 1200, or 1600 nM) for 24 h. For MEK-inhibition rescue experiments, cells were pretreated with U0126 or trametinib, both at 10 μM, for 60 min before treatment with CDDO-TFEA (1200 nM) for 24 h. After treatment, cells were washed with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 15 min at room temperature, and permeabilized with 0.1% Triton X-100 in PBS for 10 min. Cells were then subjected to TUNEL staining using a commercially available TUNEL assay kit according to the manufacturer’s instructions. Nuclei were counterstained with DAPI (Sigma-Aldrich; D9542; USA), and fluorescence images were captured using a fluorescence microscope. TUNEL-positive cells were quantified using ImageJ software and expressed as the percentage of TUNEL-positive cells relative to the total number of DAPI-stained nuclei.

Xenograft tumor model and in vivo treatment

All animal procedures were approved by the Institutional Animal Care and Use Committee of Kaohsiung Medical University (IACUC No. 114069) and were performed in accordance with institutional guidelines. Male NU/NU nude mice aged 8 weeks were used. A549 cells (1 × 10⁶ cells in 100 μL Matrigel) were subcutaneously injected into the right flank to establish xenograft tumors. Tumor growth was monitored using caliper measurements. On day 14 after tumor cell inoculation, when tumors became palpable, mice were randomized into three groups (n = 6 mice per group) and treated with vehicle control (0 mg/kg) or CDDO‑TFEA at 10 or 20 mg/kg. CDDO‑TFEA was administered by intraperitoneal (i.p.) injection once daily from day 14 to day 42. The vehicle group received an equivalent volume of vehicle by the same route and schedule. Tumor length and width were measured every 7 days, and tumor volume was calculated using the following formula:

Mice were euthanized when they reached the humane endpoint defined in the IACUC-approved protocol, including severe distress, impaired mobility, excessive tumor burden, tumor ulceration/necrosis, or other signs requiring euthanasia, or at the experimental endpoint on day 42. At the endpoint, mice were euthanized, and tumors were excised and photographed. Mice were randomized before treatment allocation. The vehicle consisted of saline. Tumor measurements were performed by investigators not blinded to treatment allocation.

Statistical analysis

All in vitro experiments were performed using at least three independent biological replicates. Data are presented as the mean ± SEM as indicated. Statistical analyses were performed using GraphPad Prism 10. Comparisons between two groups were performed using an unpaired two-tailed Student’s t-test. Comparisons among multiple groups were performed using one-way ANOVA followed by Tukey’s post hoc test. Tumor growth curves were analyzed using two-way repeated-measures ANOVA followed by multiple-comparison correction. Normality was assessed before applying parametric tests. A P value < 0.05 was considered statistically significant.

Results

CDDO-TFEA suppresses NSCLC cell viability and induces apoptosis

To determine the antitumor effects of CDDO-TFEA in NSCLC cells, A549 and H1299 cells were treated with increasing concentrations of CDDO-TFEA, followed by assessment of cell viability. CDDO-TFEA markedly reduced cell viability in a concentration-dependent manner in both cell lines, indicating a strong growth-inhibitory effect at higher concentrations (Fig. 1A). To further examine whether the reduction in cell viability was associated with apoptotic signaling, the expression of apoptosis-related proteins was analyzed by immunoblotting. CDDO-TFEA treatment increased the levels of cleaved PARP and cleaved caspase-3 in a dose-dependent manner in both A549 and H1299 cells, supporting activation of caspase-associated apoptotic signaling (Fig. 1B). Consistently, Annexin V/7-AAD flow cytometry revealed a progressive increase in apoptotic populations following CDDO-TFEA treatment, and quantitative analysis confirmed significantly elevated apoptosis in both cell lines at higher CDDO-TFEA concentrations compared with untreated controls (Fig. 1C). In addition, TUNEL fluorescence staining further demonstrated increased DNA fragmentation after CDDO-TFEA exposure, as evidenced by a dose-dependent increase in TUNEL-positive cells in both A549 and H1299 cells (Fig. 1D). Collectively, these results demonstrate that CDDO-TFEA suppresses NSCLC cell viability and induces apoptotic cell death, as supported by PARP and caspase-3 cleavage, Annexin V/7-AAD staining, and TUNEL positivity.

Fig. 1.

CDDO-TFEA inhibited NSCLC cell viability and induced apoptotic cell death in A549 and H1299 cells A A549 and H1299 cells were treated with increasing concentrations of CDDO-TFEA (0–2000 nM), and cell viability was assessed and normalized to the untreated control. CDDO-TFEA reduced cell viability in a concentration-dependent manner in both cell lines ( n = 3). B Representative immunoblot analysis of apoptosis-related proteins in A549 and H1299 cells treated with CDDO-TFEA (0, 400, 800, 1200, and 1600 nM). CDDO-TFEA increased the expression of cleaved PARP and cleaved caspase-3 in a concentration-dependent manner. Densitometric quantification of cleaved PARP and cleaved caspase-3 is shown below the blots ( n = 3). C Apoptosis was assessed by Annexin V/7-AAD staining followed by flow cytometry. Representative dot plots are shown for A549 and H1299 cells treated with CDDO-TFEA (0, 400, 800, 1200, and 1600 nM), together with quantification of apoptotic cells ( n = 3). D DNA fragmentation was evaluated by TUNEL fluorescence staining in A549 and H1299 cells treated with CDDO-TFEA (0, 400, 800, 1200, and 1600 nM). Representative fluorescence images show TUNEL-positive cells (red), nuclei counterstained with DAPI (blue), and merged images. Quantification of TUNEL-positive cells is shown in the bar graphs ( n = 3). Data are presented as mean ± SEM. Statistical significance compared with the 0 nM control group is indicated as *P < 0.05, **P < 0.01, and ***P < 0.001. Abbreviations: 7-AAD, 7-aminoactinomycin D; DAPI, 4’,6-diamidino-2-phenylindole; NSCLC, non-small cell lung cancer; PARP, poly(ADP-ribose) polymerase; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling

CDDO-TFEA suppresses stemness-associated phenotypes and tube-like network formation in NSCLC cells

To examine whether CDDO-TFEA affects cancer stemness–associated properties in NSCLC cells, tumorsphere formation assays were performed in A549 and H1299 cells. Treatment with CDDO-TFEA (200 nM) markedly impaired primary tumorsphere formation in both cell lines, as evidenced by significant reductions in sphere number and sphere size compared with the untreated control. To further evaluate the effect of CDDO-TFEA on sphere-forming activity, secondary tumorsphere formation was also assessed. Consistently, CDDO-TFEA significantly suppressed secondary tumorsphere formation in both A549 and H1299 cells, further supporting its inhibitory effect on sphere-forming activity (Fig. 2A). We next evaluated angiogenesis-related tube-like network formation. Under our experimental conditions, only A549 cells formed tube-like structures; therefore, tube-formation analysis was performed exclusively in A549 cells. CDDO-TFEA (200 nM) substantially disrupted tube-like network formation, and quantitative analysis demonstrated significant decreases in the number of nodes, relative total tube length, relative mesh count, and relative branching index compared with the control group (Fig. 2B). These results indicate that CDDO-TFEA suppresses the tube-like network-forming ability of A549 cells. Consistent with these functional findings, immunoblotting showed that CDDO-TFEA treatment at 200 nM reduced the protein expression of stemness-associated markers, including CD133, OCT4, and SOX2, as well as the pro-angiogenic factor VEGF in both A549 and H1299 cells. Densitometric quantification confirmed significant downregulation of these proteins following CDDO-TFEA treatment (Fig. 2C). Collectively, these results suggest that CDDO-TFEA attenuates stemness-associated and angiogenesis-related phenotypes in NSCLC cells, as reflected by reduced tumorsphere formation, impaired tube-like network formation, and decreased expression of stemness- and angiogenesis-associated markers.

Fig. 2.

CDDO-TFEA attenuated stemness-associated phenotypes and reduced tube-like network formation in A549 and H1299 cells A Primary and secondary tumorsphere formation assays were performed in A549 and H1299 cells treated with vehicle control or CDDO-TFEA (200 nM). Representative images of first-generation and second-generation tumorspheres are shown. Quantification of sphere number and sphere size demonstrated that CDDO-TFEA significantly reduced both primary and secondary tumorsphere formation in A549 and H1299 cells. Scale bars are shown in the images ( n = 6). B Tube-formation assay was performed in A549 cells treated with vehicle control or CDDO-TFEA (200 nM). Because H1299 cells did not form tube-like networks under these experimental conditions, tube-formation analysis was conducted in A549 cells only. Representative images are shown, together with quantitative analysis of the number of nodes, relative total tube length, relative mesh count, and relative branching index. CDDO-TFEA markedly impaired tube-like network formation. Scale bars are shown in the images ( n = 3). C Representative immunoblot analysis of stemness-associated markers CD133, OCT4, and SOX2, and the pro-angiogenic factor VEGF in A549 and H1299 cells treated with vehicle control or CDDO-TFEA (200 nM) ( n = 3). Densitometric quantification of protein expression normalized to total protein is shown on the right. Data are presented as mean ± SEM. Statistical significance compared with the control group is indicated as *P < 0.05, **P < 0.01, and ***P < 0.001. Abbreviations: NSCLC, non-small cell lung cancer; OCT4, octamer-binding transcription factor 4; SOX2, SRY-box transcription factor 2; VEGF, vascular endothelial growth factor

CDDO-TFEA prominently activates ERK signaling in NSCLC cells, and this activation is attenuated by MEK inhibition

To better investigate whether mitogen-activated protein kinase (MAPK) signaling pathways are involved in the cellular response to CDDO‑TFEA, A549 and H1299 cells were treated with increasing concentrations of CDDO‑TFEA for 6 h, and the phosphorylation status of ERK, JNK, and p38 was examined by immunoblotting. CDDO‑TFEA induced a clear, concentration-dependent increase in ERK phosphorylation in both A549 and H1299 cells, and densitometric analysis confirmed significant elevation of the p‑ERK/ERK ratio compared with untreated controls (Fig. 3A). In contrast, the phosphorylation levels of JNK and p38 were not markedly altered following CDDO‑TFEA treatment, as indicated by relatively unchanged p‑JNK/JNK and p‑p38/p38 ratios in both cell lines (Fig. 3A). These findings suggest that ERK activation represents the most prominent early MAPK response to CDDO-TFEA treatment under these experimental conditions. To further determine whether CDDO‑TFEA–induced ERK phosphorylation was mediated through MEK signaling, cells were treated with CDDO‑TFEA in the presence or absence of pharmacological MEK inhibitors. U0126 markedly attenuated CDDO‑TFEA‑induced ERK phosphorylation in both A549 and H1299 cells. To strengthen the pharmacological validation and reduce the possibility of inhibitor-specific off-target effects, trametinib, a structurally distinct MEK inhibitor, was also used. Consistently, trametinib significantly suppressed CDDO‑TFEA‑induced ERK phosphorylation in both cell lines (Fig. 3B). Together, these results indicate that ERK activation is a prominent MAPK response to CDDO-TFEA in NSCLC cells and that this activation can be attenuated by MEK inhibition using two independent pharmacological inhibitors.

Fig. 3.

CDDO‑TFEA activated ERK signaling in A549 and H1299 cells and was attenuated by MEK inhibition A A549 and H1299 cells were treated with increasing concentrations of CDDO‑TFEA (0, 400, 800, and 1200 nM) for 6 h to evaluate early mitogen-activated protein kinase signaling responses. Cell lysates were subjected to immunoblot analysis to detect phosphorylated and total ERK, JNK, and p38. Representative blots are shown, and densitometric quantification was performed by normalizing p‑ERK to total ERK, p‑JNK to total JNK, and p‑p38 to total p38. Quantified values are expressed relative to the 0 nM control group ( n = 3). B A549 and H1299 cells were pretreated with the MEK inhibitors U0126 or trametinib (10 μM) before treatment with CDDO‑TFEA (1200 nM) for 6 h. ERK phosphorylation was examined by immunoblotting, and total ERK was used for normalization. Both U0126 and trametinib attenuated CDDO‑TFEA–induced ERK phosphorylation in A549 and H1299 cells ( n = 3). Data are presented as mean ± SEM. Statistical significance compared with the control group is indicated as *P < 0.05, **P < 0.01, and ***P < 0.001. Statistical significance compared with the CDDO‑TFEA–treated group is indicated as #P < 0.05, ##P < 0.01, and ###P < 0.001. Abbreviations: ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein kinase; MEK, mitogen-activated protein kinase kinase; NSCLC, non-small cell lung cancer

MEK inhibition attenuates CDDO‑TFEA–induced apoptosis in NSCLC cells

To determine whether MEK/ERK signaling contributes to CDDO-TFEA–induced apoptotic cell death, we first evaluated the cytotoxicity of two pharmacological MEK inhibitors, U0126 and trametinib, in A549 and H1299 cells. Cell viability analysis showed that U0126 maintained more than 90% cell viability at concentrations up to 10 μM in both cell lines, whereas higher concentrations reduced cell viability (Supplementary Fig. S1A). Similarly, trametinib showed limited cytotoxicity at concentrations up to 10 μM, with more than 90% cell viability retained in both A549 and H1299 cells, whereas higher concentrations decreased cell viability (Supplementary Fig. S1B). Therefore, 10 μM was selected as the working concentration for both U0126 and trametinib in subsequent rescue experiments. We next examined whether MEK inhibition could alter the effects of CDDO-TFEA on NSCLC cell viability. A549 and H1299 cells were treated with CDDO-TFEA (1200 nM) in the presence or absence of U0126 or trametinib (10 μM). CDDO-TFEA significantly reduced cell viability in both cell lines, whereas co-treatment with either U0126 or trametinib partially attenuated the CDDO-TFEA-mediated reduction in cell viability (Fig. 4A). Consistently, Annexin V/7-AAD flow cytometry demonstrated that CDDO-TFEA (1200 nM) markedly increased apoptotic populations in both A549 and H1299 cells, while co-treatment with U0126 or trametinib significantly reduced CDDO-TFEA–induced apoptosis (Fig. 4B). To further confirm these findings at the molecular level, apoptosis-associated protein expression was assessed by immunoblotting. CDDO-TFEA (1200 nM) increased PARP cleavage and cleaved caspase-3 expression in both A549 and H1299 cells, indicating activation of caspase-associated apoptotic signaling. Notably, both U0126 and trametinib (10 μM) attenuated CDDO-TFEA–induced PARP cleavage and reduced cleaved caspase-3 levels (Fig. 4C). In agreement with the flow cytometry and immunoblotting results, TUNEL fluorescence staining showed a marked increase in TUNEL-positive cells following CDDO-TFEA treatment, whereas co-treatment with either U0126 or trametinib decreased the proportion of TUNEL-positive cells in both cell lines (Fig. 4D). Collectively, these results indicate that pharmacological inhibition of MEK by either U0126 or trametinib partially attenuates CDDO-TFEA–mediated loss of cell viability and apoptotic cell death, supporting the involvement of MEK/ERK signaling, at least in part, in the pro-apoptotic effects of CDDO-TFEA in NSCLC cells.

Fig. 4.

MEK inhibition partially attenuated CDDO-TFEA-induced loss of viability and apoptotic cell death in A549 and H1299 cells A A549 and H1299 cells were treated with CDDO-TFEA (1200 nM) in the presence or absence of the MEK inhibitors U0126 or trametinib (10 μM), and cell viability was assessed. CDDO-TFEA significantly reduced cell viability in both cell lines, whereas co-treatment with U0126 or trametinib partially attenuated the CDDO-TFEA-mediated loss of cell viability ( n = 3). B Apoptosis was assessed by Annexin V/7-AAD staining followed by flow cytometry. Representative dot plots are shown for A549 and H1299 cells treated with vehicle control, CDDO-TFEA (1200 nM), or CDDO-TFEA combined with U0126 or trametinib (10 μM). Quantification of apoptotic cells is shown in the bar graphs ( n = 3). C Representative immunoblot analysis of PARP and cleaved caspase-3 in A549 and H1299 cells treated as indicated. Densitometric quantification of cleaved PARP and cleaved caspase-3 expression normalized to total protein is shown on the right. Both U0126 and trametinib attenuated CDDO-TFEA–induced cleaved caspase-3 expression ( n = 3). D NA fragmentation was evaluated by TUNEL fluorescence staining in A549 and H1299 cells treated with vehicle control, CDDO-TFEA (1200 nM), or CDDO-TFEA combined with U0126 or trametinib (10 μM). Representative fluorescence images show TUNEL-positive cells in red, DAPI-stained nuclei in blue, and merged images. Quantification of TUNEL-positive cells is shown in the bar graphs. Scale bars are shown in the images ( n = 3). Data are presented as mean ± SEM. Statistical significance compared with the control group is indicated as *P < 0.05, **P < 0.01, and ***P < 0.001. Statistical significance compared with the CDDO-TFEA-treated group is indicated as #P < 0.05, ##P < 0.01, and ###P < 0.001. Abbreviations: 7-AAD, 7-aminoactinomycin D; DAPI, 4’,6-diamidino-2-phenylindole; MEK, mitogen-activated protein kinase kinase; PARP, poly(ADP-ribose) polymerase; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling

MEK inhibition partially attenuates CDDO-TFEA–mediated suppression of tumorsphere formation, tube-like network formation, and stemness/angiogenic marker expression in NSCLC cells.

To further determine whether MEK/ERK signaling contributes to the inhibitory effects of CDDO-TFEA on stemness-associated and angiogenesis-related phenotypes, A549 and H1299 cells were treated with CDDO-TFEA (200 nM) in the presence or absence of the MEK inhibitors U0126 or trametinib (10 μM), followed by functional and molecular analyses. In tumorsphere formation assays, CDDO-TFEA markedly reduced both primary and secondary tumorsphere formation in A549 and H1299 cells, as evidenced by decreased sphere number and sphere size compared with the control group. Co-treatment with either U0126 or trametinib partially preserved tumorsphere formation, suggesting that MEK inhibition attenuated, at least in part, the suppressive effect of CDDO-TFEA on tumorsphere-forming activity (Fig. 5A). Because only A549 cells formed tube-like structures under our experimental conditions, tube-formation analysis was performed in A549 cells only. CDDO-TFEA substantially disrupted tube-like network formation, as shown by reductions in the number of nodes, relative total tube length, relative mesh count, and relative branching index. Notably, co-treatment with U0126 or trametinib partially attenuated tube-like network organization and significantly improved these angiogenesis-related parameters compared with CDDO-TFEA treatment alone (Fig. 5B). Consistent with these functional findings, immunoblotting showed that CDDO-TFEA reduced the expression of stemness-associated markers, including CD133, OCT4, and SOX2, as well as the pro-angiogenic factor VEGF in both A549 and H1299 cells. Importantly, co-treatment with either U0126 or trametinib partially attenuated the CDDO-TFEA–mediated downregulation of these proteins, although the extent of restoration varied among markers and cell lines (Fig. 5C). Collectively, these results indicate that MEK/ERK signaling contributes, at least in part, to the suppressive effects of CDDO-TFEA on stemness-associated and angiogenesis-related phenotypes in NSCLC cells.

Fig. 5.

MEK inhibition attenuated CDDO-TFEA-mediated suppression of stemness and tube formation in A549 and H1299 cells A Primary and secondary tumorsphere formation assays were performed in A549 and H1299 cells treated with vehicle control, CDDO‑TFEA (200 nM), or CDDO‑TFEA combined with the MEK inhibitors U0126 or trametinib (10 μM). Representative images of first- and second-generation tumorspheres are shown. Quantification of sphere number and sphere size demonstrated that CDDO‑TFEA significantly suppressed tumorsphere formation in both A549 and H1299 cells. Scale bars are shown in the images ( n = 6). B Tube-formation assays were performed in A549 cells treated with vehicle control, CDDO‑TFEA (200 nM), or CDDO‑TFEA combined with U0126 or trametinib (10 μM). Because H1299 cells did not form tube-like networks under these experimental conditions, tube-formation analysis was conducted in A549 cells only. Representative images are shown, together with quantification of the number of nodes, relative total tube length, relative mesh count, and relative branching index. Co-treatment with either U0126 or trametinib partially attenuated the inhibitory effect of CDDO‑TFEA on tube-like network formation ( n = 3). C Representative immunoblot analysis of stemness-associated markers CD133, OCT4, and SOX2, and the pro-angiogenic factor VEGF in A549 and H1299 cells treated with vehicle control, CDDO‑TFEA (200 nM), or CDDO‑TFEA combined with U0126 or trametinib (10 μM). Densitometric quantification of protein expression normalized to total protein is shown on the right. U0126 and trametinib partially attenuated the CDDO‑TFEA–mediated downregulation of stemness- and angiogenesis-associated proteins ( n = 3). Data are presented as mean ± SEM. Statistical significance compared with the control group is indicated as *P < 0.05, **P < 0.01, and ***P < 0.001. Statistical significance compared with the CDDO‑TFEA-treated group is indicated as #P < 0.05, ##P < 0.01, and ###P < 0.001. Abbreviations: MEK, mitogen-activated protein kinase kinase; NSCLC, non-small cell lung cancer; OCT4, octamer-binding transcription factor 4; SOX2, SRY-box transcription factor 2; VEGF, vascular endothelial growth factor

CDDO‑TFEA suppresses tumor growth in vivo in a dose‑dependent manner

To validate the antitumor activity of CDDO‑TFEA in vivo, an A549 xenograft mouse model was established. Tumor-bearing mice were randomized into three groups and treated with vehicle control (0 mg/kg) or CDDO‑TFEA at 10 or 20 mg/kg beginning on day 14 after tumor cell inoculation, when tumors became palpable. CDDO‑TFEA was administered by intraperitoneal injection once daily until day 42. At the experimental endpoint, excised tumors from CDDO‑TFEA–treated mice were visibly smaller than those from vehicle-treated mice, with the 20 mg/kg group showing the most pronounced reduction in tumor burden (Fig. 6A). Consistently, longitudinal tumor volume measurements demonstrated that CDDO‑TFEA significantly delayed tumor growth in a dose-dependent manner. After treatment initiation on day 14, tumors in the vehicle group continued to grow rapidly, whereas tumor growth was markedly suppressed in both the 10 and 20 mg/kg CDDO‑TFEA groups, with significant differences observed at later time points (Fig. 6B). These findings indicate that CDDO‑TFEA administration by intraperitoneal injection once daily effectively suppresses xenograft tumor growth in vivo.

Fig. 6.

CDDO-TFEA inhibited xenograft tumor growth in vivo in a dose-dependent manner. A Representative images of excised tumors collected at the experimental endpoint from mice treated with vehicle (0 mg/kg) or CDDO-TFEA (10 or 20 mg/kg). A ruler is shown for size reference, and scale calibration is indicated in the image. B Tumor growth curves showing changes in tumor volume over time in mice treated with vehicle control or CDDO‑TFEA. Treatment was initiated on day 14 after tumor cell inoculation, as indicated by the arrow, and CDDO‑TFEA was administered by intraperitoneal injection once daily until day 42. Tumor volume was measured every 7 days and calculated as (length × width²) / 2. Data are presented as mean ± SEM (n = 6 mice per group). Statistical significance compared with the vehicle group is denoted as **P < 0.01 and ***P < 0.001

Discussion

This study demonstrated that CDDO-TFEA exerts multifaceted antitumor effects in NSCLC cells, as evidenced by induction of apoptotic cell death, suppression of stemness-associated and angiogenesis-related phenotypes, and activation of ERK signaling. First, CDDO-TFEA triggered pronounced apoptotic cell death in A549 and H1299 cells, as evidenced by Annexin V/7-AAD staining, PARP cleavage, cleaved caspase-3 expression, and TUNEL positivity (Fig. 1). Although cleaved caspase-3, PARP cleavage, Annexin V positivity, and TUNEL staining support activation of apoptotic cell death, an increase in Annexin V-negative/7-AAD-positive dead cells was also observed, particularly in H1299 cells. Therefore, we cannot exclude the possibility that non-apoptotic forms of regulated cell death may also contribute to the response to CDDO-TFEA. Ferroptosis is an iron-dependent form of regulated non-apoptotic cell death driven by lipid peroxidation, and it has been implicated in cancer therapy responses (Dixon et al. 2012; Stockwell et al. 2017). Future studies assessing ferroptosis-related markers, lipid peroxidation, GPX4 expression, and ferrostatin-1 rescue will be required to clarify this possibility. These findings are consistent with the results of previous studies wherein oleanane triterpenoids, such as CDDO, and their derivatives provoke apoptosis in diverse cancer types by both extrinsic and intrinsic pathways (Ikeda et al. 2003). For example, CDDO compounds have been shown to induce mitochondrial cytochrome c release and caspase activation in leukemia and prostate cancer models (frequently via ROS generation and stress kinase signaling) (Ikeda et al. 2003). Indeed, CDDO-Me (bardoxolone methyl)—a closely related analog—exhibits broad-spectrum cytotoxicity against ovarian, prostate, leukemia, breast, lung, and pancreatic cancer cells in vitro while sparing normal cells (and thereby demonstrating a favorable therapeutic index) (Jiao et al. 2011). Concordantly, CDDO-Me and related triterpenoids have demonstrated in vivo efficacy in tumor models; in one study CDDO-Me significantly inhibited the growth of liver, lung, and pancreatic tumor xenografts and even prevented postoperative tumor recurrence by eradicating residual disease (Gao et al. 2015). We observed a similar antitumor effect in vivo, as CDDO-TFEA treatment dramatically suppressed NSCLC xenograft growth (Fig. 6). This study demonstrated that CDDO-TFEA suppresses NSCLC cell viability, induces apoptotic cell death, inhibits stemness-associated and angiogenesis-related phenotypes, activates ERK signaling, and suppresses xenograft tumor growth.

Importantly, CDDO-TFEA reduced tumorsphere formation and decreased the expression of stemness-associated markers, including CD133, OCT4, and SOX2, in NSCLC cells. Tumorsphere formation is commonly used as an in vitro functional readout associated with stemness-like properties, although it does not by itself define a specific cancer stem cell population. In parallel, CDDO-TFEA impaired tube-like network formation in A549 cells and reduced VEGF expression, suggesting suppression of angiogenesis-related or vasculogenic mimicry-like phenotypes. These findings indicate that CDDO-TFEA suppresses malignant properties associated with sphere-forming activity and tube-like network formation. Because CD133 detection can be influenced by glycosylation status and OCT4 antibodies may recognize different isoforms or related epitopes, these immunoblot data were interpreted as changes in stemness-associated marker expression rather than definitive identification of a specific CSC population. Additional assays such as ELDA, flow-based CSC marker analysis, or in vivo limiting dilution would be required to define the affected subpopulation (Hu and Smyth 2009). CDDO-TFEA activates MEK/ERK signaling in NSCLC cells, and pharmacological inhibition using U0126 or trametinib suggests that this pathway contributes, at least in part, to its antitumor effects. Phospho-ERK levels sharply increased following CDDO-TFEA treatment, and co-treatment with the MEK1/2 inhibitor, U0126 or trametinib, significantly attenuated apoptosis, caspase activation, and the loss of stemness markers (Figs. 4 and 5). This requirement for ERK signaling in cell death seemingly is counterintuitive as ERK1/2 (extracellular signal–regulated kinases) are classically associated with mitogenic and pro-survival signaling in cancer. Oncogenic mutations in the RAS–RAF–MEK–ERK cascade drive uncontrolled proliferation in many cancers, and persistent ERK activation is frequently linked to tumor growth and therapy resistance (Wu et al. 2019). However, our results underscore the well-known context-dependent nature of ERK signaling in tumor biology. Accumulating evidence indicates that sustained or hyperacute ERK activation can trigger antiproliferative outcomes—including apoptosis, autophagy, and cellular senescence—especially in the presence of cellular stress or in the context of oncogene-induced senescence (Cagnol and Chambard 2010; Deschênes-Simard et al. 2014). Prolonged ERK activation promoted apoptosis in breast cancer cells that were treated with the flavonoid kaempferol, particularly under 3D culture conditions that better mimicked the tumor microenvironment (Kim et al. 2008). Furthermore, supraphysiologic ERK activity can overwhelm pro-survival signals and lead to cell death (“ERK- associated apoptosis”)—a phenomenon that has prompted the exploration of ERK pathway agonism as a therapeutic strategy (Sugiura et al. 2021). In our study, CDDO-TFEA may induce a level of ERK signaling that reverses its output from a pro-growth to pro-death effect. Mechanistically, ERK activation could facilitate apoptosis via several routes: ERK can phosphorylate and inactivate anti-apoptotic proteins, upregulate pro-apoptotic BH3-only proteins, and/or engage transcription factors, such as FOXO3a or p53, that drive apoptosis in stressed cells (context-dependent mechanisms reported in other systems). In the case of CDDO-TFEA, the ERK pathway seems to intersect with the regulation of stemness markers. We observed that U0126 or trametinib not only rescued cell viability but partly attenuated the expression of SOX2 and other stemness-related proteins in CDDO-TFEA–treated cells. This suggests ERK activation contributes to the attenuation of stemness-associated phenotypes, as reflected by reduced tumorsphere formation and decreased CD133, OCT4, and SOX2 expression. Although MEK inhibition partially attenuated these changes, the present data do not determine whether this effect reflects altered stemness-marker regulation, reduced survival of sphere-forming cells, or other cellular responses. Taken together, our data support a model in which CDDO-TFEA-induced activation contributes, at least in part, to its pro-apoptotic and anti-stemness-associated effects. Future studies to delineate the downstream effectors of ERK in this context (e.g., BIM, p21^CIP1^, or others that execute the apoptotic program and erode stemness) are important. Moreover, U0126 did not fully prevent CDDO-TFEA–induced cell death (especially at later timepoints), which indicates the possible involvement of ERK-independent mechanisms.

CDDO-TFEA has been reported as an NRF2-activating triterpenoid in previous studies, NRF2 activation, downstream antioxidant targets such as HO-1 and NQO1, and redox changes were not directly examined in the present study. Therefore, the contribution of NRF2/redox regulation to the effects observed here remains to be clarified in future work. By suppressing stemness-associated phenotypes while simultaneously impairing tumor vascular mimicry, such agents may enable deeper and more durable antitumor responses. Ongoing and future studies would focus on validating these effects in vivo, optimizing dosing to maximize the therapeutic window, and potentially structurally refine CDDO-TFEA to enhance its anticancer potency while minimizing any off-target toxicity. Overall, our findings position CDDO-TFEA as a promising prototype in the quest for therapies that collapse tumor self-defense mechanisms (Rojo de la Vega et al. 2018), and thereby convert a one-time survival advantage into a lethal liability. Further exploration of this compound and its analogs could enrich the armamentarium against NSCLC—a disease that still remains in need of more effective treatment options, especially in the face of drug resistance and metastasis.

Conclusion

Our study identifies CDDO-TFEA as a promising antitumor candidate against NSCLC. CDDO-TFEA reduced NSCLC cell viability and induced apoptotic cell death in A549 and H1299 cells, as supported by Annexin V/7-AAD staining, PARP cleavage, cleaved caspase-3 expression, and TUNEL positivity. Beyond its cytotoxic effects, CDDO-TFEA suppressed tumorsphere formation, reduced CD133, OCT4, SOX2, and VEGF expression, and impaired tube-like network formation in A549 cells. Mechanistically, CDDO-TFEA activated ERK signaling, and pharmacological MEK inhibition with U0126 or trametinib partially attenuated CDDO-TFEA-induced apoptosis and suppression of stemness-associated and angiogenesis-related phenotypes. Importantly, CDDO-TFEA exerted dose-dependent tumor growth suppression in vivo. Collectively, these findings suggest that CDDO-TFEA induces apoptotic stress and suppresses malignant phenotypes associated with stemness and tube-like network formation in NSCLC. Future studies should clarify the contribution of NRF2/redox regulation, evaluate possible non-apoptotic cell death mechanisms, validate the MEK/ERK axis using genetic approaches, and further assess safety and efficacy across additional NSCLC models.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Jui-Ying Lee: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Visualization, Writing—original draft. Chia-Li Chung: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Visualization, Writing—original draft. Tzu-Ting Tseng: Methodology, Investigation. Shih-Hsun Kuo: Methodology, Formal analysis. Yu-Peng Liu: Formal analysis. Ming-Tsang Wu: Formal analysis. Hung-Pei Tsai: Resources, Supervision, Project administration, Writing—review and editing. Aij-Lie Kwan: Conceptualization, Resources, Supervision, Project administration, Funding acquisition, Writing—review and editing. All authors read and approved the final manuscript. All named authors meet the International Committee of Medical Journal Editors criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.

Funding

This work was supported by Kaohsiung Medical University Hospital, Kaohsiung City, Taiwan (KMUH 99-9M62, KMUH 100-0M10, KMUH113-3M17, and KMUH114-4M19). No separate funding or sponsorship was received for the publication of this article.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of Interest

The authors declare no competing interests.

Ethical approval

All animal experiments were approved by the Institutional Animal Care and Use Committee of Kaohsiung Medical University (IACUC No. 114069) and were performed in accordance with institutional guidelines. This study did not involve human participants.

Medical writing, editorial, and other assistance

The authors received no medical writing, editorial, or other assistance in the preparation of this manuscript.