The human NSCLC cell lines A549 (cat. no. 60074) and NCI-H460 (cat. no. 60373), as well as the human normal lung fibroblast cell line IMR-90 (cat. no. 60204), were obtained from the Bioresource Collection and Research Center (BCRC) of the Food Industry Research and Development Institute. All cell lines were authenticated using short tandem repeat profiling and certified as mycoplasma-free by the BCRC prior to distribution. To maintain cell integrity, all experiments were performed using cells within 10 passages after resuscitation. Both A549 and NCI-H460 cells were cultured in Dulbecco's Modified Eagle Medium (HyClone™; Cytiva), whereas IMR-90 cells were maintained in Eagle's Minimum Essential Medium (Corning, Inc.). All culture media were supplemented with 10% fetal bovine serum (HyClone™; Cytiva) and 1% (v/v) penicillin-streptomycin (Gibco^®; Thermo Fisher Scientific, Inc.). All cell lines were maintained at 37°C in a humidified incubator containing 5% CO₂. Cells were subcultured or harvested for subsequent experiments in 0.05% trypsin-0.5 mM EDTA solution (Gibco^®; Thermo Fisher Scientific, Inc.).

TCS was performed using a modified PCR system equipped with waterproof capabilities in the heating area, as described in previous studies (6–8). A minimal volume of water was employed as a thermal conduction medium, enabling the direct placement of cell culture dishes onto the heating area. The modified PCR apparatus controlled both heating and cooling phases. The TCS exposure comprised 15 cycles, each consisting of a 3-min high-temperature phase followed by a 30-sec passive cooling interval (Fig. 1A). This heating-cooling period was established based on prior investigations involving various cancer cell lines (6,7). The low-temperature setting was maintained at 37°C to simulate physiological human body temperature. Actual temperatures experienced by cells at the well bottom were monitored using a needle thermocouple. Temperature cycling reached a steady state after the third cycle, with the recorded temperature cycling between 41.4 and 43.8°C, as shown in Fig. 1B.

EGCG (MedChemExpress; cat. no. HY-13653), 2-aminoethoxydiphenyl borate (2-APB) (MedChemExpress; cat. no. HY-W009724) and N-acetylcysteine (NAC; Sigma-Aldrich; Merck KGaA; cat. no. A9165) were dissolved in sterile distilled water to prepare stock solutions at concentrations of 20, 5 and 500 mM, respectively. By contrast, GSK2606414 [MedChemExpress; cat. no. HY-18072; an inhibitor of the protein kinase R-like ER kinase (PERK)], JNJ-28583113 [MedChemExpress; cat. no. HY-149143; an antagonist of transient receptor potential cation channel subfamily M member 2 (TRPM2)], and SB-705498 [MedChemExpress; cat. no. HY-10633; an antagonist of transient receptor potential cation channel subfamily V member 1 (TRPV1)] were solubilized in dimethyl sulfoxide (DMSO; Sigma-Aldrich; Merck KGaA) at concentrations of 1, 5 and 5 mM, respectively, to make stock solutions. All stock solutions were stored at −20°C until further use. Cells were seeded in 96-well plates at 3,000 cells/well, or in 6-well plates at 100,000 cells/well, and incubated overnight at 37°C to facilitate cell adherence. For the combined treatment of EGCG and TCS, EGCG was administered immediately prior to TCS exposure. Cells were pretreated with NAC or the molecular inhibitors 1 h prior to the administration of EGCG. Unless specifically indicated, cells were maintained at 37°C in a humidified incubator for an additional 24 h after treatment for further analysis.

Cell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma-Aldrich; Merck KGaA) assay. MTT powder was dissolved in distilled water to make a 5 mg/ml stock solution, which was subsequently stored at 4°C. Immediately before use, the MTT stock solution was diluted with culture medium to a final concentration of 0.5 mg/ml. At 24 h after the treatment, the culture medium in the 96-well plates was replaced with 100 µl of the MTT working solution. The cells were then incubated at 37°C in the dark for 4 h, during which time mitochondrial dehydrogenases in viable cells reduced MTT to insoluble formazan crystals. To dissolve the formazan crystals, 100 µl 10% sodium dodecyl sulfate (SDS) (Bioshop Canada, Inc.) solution in 0.01 M hydrochloric acid (HCl) (Echo Chemical Co., Ltd.) was added to each well. After an overnight incubation, the absorbance was measured at 570 nm with background subtraction at 690 nm using a Multiskan GO spectrophotometer (Thermo Fisher Scientific, Inc.). Finally, cell viability was expressed as a ratio based on the relative quantity of formazan compared with the untreated control.

To evaluate the synergistic interaction between EGCG and TCS, the Bliss independence model was applied. The expected fractional inhibition on cell viability for the combination treatment (E_exp) was calculated using the formula: E_exp=E_EGCG + E_TCS-(E_EGCG × E_TCS), where E_EGCG and E_TCS represent the fractional inhibitions induced by individual treatments of EGCG and TCS, respectively. The Bliss excess value (ΔE) was defined as the difference between the observed fractional inhibition (E_obs) and E_exp (namely, ΔE=E_obs-E_exp). A positive Bliss excess value (ΔE >0) was indicative of synergy (21).

To assess their colony-forming capabilities, A549 cells were seeded into 35-mm culture dishes at a density of 300 cells per dish and allowed to adhere overnight. On the following day, cells were treated with EGCG, TCS, or their combination. The culture medium was replaced after 24 h, followed by an additional 10-day incubation period. Subsequently, colonies were fixed using 4% paraformaldehyde (Sigma-Aldrich; Merck KGaA) for 10 min and stained with 0.1% crystal violet (Sigma-Aldrich; Merck KGaA) for 15 min at room temperature. Aggregates containing 50 or more cells were identified as viable colonies. Colony numbers were quantified manually using an optical microscope, and the results were expressed relative to the untreated control group.

Protein levels in A549 cells were assessed using western blotting. After treatment, cells were washed with phosphate-buffered saline (PBS) and subsequently lysed on ice for 1 h using RIPA lysis buffer (50 mM Tris-HCl, pH 7.4, 0.15 M NaCl, 0.25% deoxycholic acid, 1% NP-40, 1% Triton X-100, 0.1% SDS, 1 mM EDTA) (MilliporeSigma) supplemented with protease and phosphatase inhibitor cocktails (MilliporeSigma). Lysates were centrifuged at 10,000 × g for 30 min at 4°C, and the supernatants were then collected. Protein concentrations were determined using the Bradford protein assay (Bioshop Canada Inc.). For the detection of total ubiquitinated proteins, cells were directly harvested and lysed in pre-heated sample loading buffer at 95°C for 10 min to rapidly deactivate endogenous deubiquitinating enzymes and preserve polyubiquitinated species. Equal amounts of protein samples (20 µg) were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE; 12% acrylamide gels), and then transferred onto polyvinylidene fluoride (PVDF) membranes (MilliporeSigma). To facilitate the detection of total ubiquitinated proteins, PVDF membranes were soaked in PBS and subsequently subjected to autoclaving at 121°C for 30 min prior to the blocking step, thereby improving epitope accessibility. Membranes were blocked with 5% bovine serum albumin (BioShop Canada Inc.) in Tris-buffered saline with Tween-20 (TBST; 20 mM Tris, 150 mM NaCl and 0.1% Tween 20) for 1 h at room temperature, followed by an overnight incubation at 4°C with primary antibodies targeting proteasome subunit alpha type-3 (PSMA3; cat. no. 12446; 1:1,000), 26S proteasome regulatory subunit 6A (PSMC3; cat. no. 13923; 1:1,000), HSP70 (cat. no. 4872; 1:1,000) and phosphorylated c-Jun N-terminal kinase (p-JNK; Thr183/Tyr185; cat. no. 4668; 1:1,000) (all from Cell Signaling Technology, Inc.), and HSP105 (cat. no. GTX03248; 1:1,000), ubiquitin (cat. no. GTX128826; 1:5,000), phosphorylated eukaryotic initiation factor 2 subunit α (p-eIF2α; Ser51; cat. no. GTX130006; 1:1,000), total-eIF2α (t-eIF2α; cat. no. GTX101241; 1:2,000), total-JNK (t-JNK; cat. no. GTX133806; 1:1,000), poly(ADP-ribose) polymerase (PARP; cat. no. GTX132329; 1:1,000) and GAPDH (cat. no. GTX100118; 1:10,000) (all from GeneTex, Inc.). After washing with TBST, the membranes were incubated overnight at 4°C with horseradish peroxidase-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, Inc.; 1:10,000). Finally, the immunoreactive signals were visualized using an enhanced chemiluminescence substrate (Advansta Inc.) and detected in an Amersham Imager 600 imaging system [Cytiva (formerly GE Healthcare Life Sciences)]. Finally, protein bands were quantified using Image Lab software (Bio-Rad Laboratories, Inc.).

Total RNA was extracted from A549 cells using the miTotal RNA Extraction Miniprep System (Viogene), following the manufacturer's instructions. Reverse transcription (RT) and PCR amplification were performed using a HiScript II One-Step RT-PCR Kit (Vazyme Biotech Co., Ltd.) with gene-specific primers. The primer sequences for human XBP1 were as follows: Forward, 5′-TTACGAGAGAAAACTCATGGCC-3′ and reverse, 5′-GGGTCCAAGTTGTCCAGAATGC-3′; note that the primers flanked the 26-nucleotide intron that was excised by inositol-requiring enzyme 1α (IRE1α)-mediated splicing. GAPDH was amplified as an internal control using the following primers: Forward, 5′-GGAGCGAGATCCCTCCAAAAT-3′, and reverse 5′-GGCTGTTGTCATACTTCTCATGG-3′. To achieve optimal resolution among the unspliced (XBP1u), heteroduplex (XBP1h) and spliced (XBP1s) isoforms, PCR products were separated on a 2.5% agarose gel prepared in TBE buffer (89 mM Tris-borate and 2 mM EDTA, pH 8.3) and electrophoresed at 70 V for 1.5 h. Subsequently, the gel was stained for 45 min using GelRed Nucleic Acid Stain (Biotium, Inc.) diluted at a ratio of 3:10,000 in a 0.1 M NaCl aqueous solution. DNA bands were visualized using the Keta S Standalone Gel Documentation System (Wealtec Bioscience Co., Ltd.). Quantification of band intensities was performed using ImageJ software (version 1.53k; National Institutes of Health), and the ratio of XBP1 mRNA splicing was calculated according to the formula: Splicing ratio=(Intensity of XBP1s + 0.5× Intensity of XBP1h)/(Intensity of XBP1s + Intensity of XBP1h + Intensity of XBP1u) (22).

Cytosolic Ca²⁺ levels were measured using the fluorescent dye Fluo-4 AM (MedChemExpress) to quantify cytosolic Ca²⁺ levels. A 1 mM stock solution of Fluo-4 AM was prepared by dissolving the dye in DMSO, after which it was stored at −20°C until further use. At 24 h after treatments, cells were harvested and washed with Hanks' Balanced Salt Solution (HBSS) and subsequently incubated with 1 µM Fluo-4 AM in HBSS for 1 h at 37°C in the dark to load the dye. Following the incubation, cells were washed with HBSS and finally resuspended in fresh HBSS. Fluorescence intensity of Fluo-4 AM was measured using a FACSLyric flow cytometer (BD Biosciences) in the FITC channel. For each sample, at minimum of 9,000 events were acquired. Cellular debris was excluded based on the forward scatter (FSC) and side scatter (SSC) characteristics. Finally, cytosolic Ca²⁺ levels were quantified by calculating the mean fluorescence intensity and expressed as fold changes relative to the untreated control group.

The dihydroethidium (DHE) (Sigma-Aldrich; Merck KGaA) fluorescent dye was employed to measure ROS levels in A549 cells. At 24 h after treatments, cells were harvested and washed with PBS prior to staining. Cells were subsequently incubated with a 5 µM DHE working solution for 30 min at 37°C in the dark. Fluorescence intensity was measured using a FACSLyric flow cytometer (BD Biosciences) in the PE channel. For each sample, at minimum of 9,000 events were acquired. Cellular debris was excluded based on FSC and SSC characteristics. Finally, ROS levels were determined by calculating the mean fluorescence intensity and expressed as fold changes compared with the control group.

The MMP integrity was assessed using the JC-1 (MedChemExpress) fluorescent dye. At 24 h after treatments, cells were harvested and rinsed with PBS. Subsequently, cells were incubated with 5 µg/ml JC-1 in PBS for 30 min at 37°C in the dark. Fluorescence intensities of JC-1 were measured using the FACSLyric system (BD Biosciences) in both the PE (red fluorescence) and FITC (green fluorescence) channels. For each sample, at minimum of 9,000 events were acquired. Cellular debris was excluded based on the FSC and SSC characteristics, and MMP disruption was quantified through assessing the proportion of cells exhibiting a fluorescence shift from red (JC-1 aggregates) to green (JC-1 monomers).

Apoptotic rates were quantified using an annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI) double staining kit (BD Biosciences). At 24 h after treatments, cells were collected, washed with PBS, and subsequently incubated in a binding buffer containing annexin V-FITC and PI for 30 min at 37°C in the dark, following the manufacturer's instructions. Fluorescence intensities were measured using the FACSLyric system (BD Biosciences) in both the PE and FITC channels. Early and late apoptotic cells were quantified as the percentages of cells identified in the (PE⁻/FITC⁺) and (PE⁺/FITC⁺) quadrants, respectively. For each sample, a minimum of 9,000 events were acquired. Cellular debris was excluded based on the FSC and SSC characteristics, and the percentages of these populations were calculated accordingly.

Three independent biological experiments were performed for each experimental condition using separately prepared cell cultures. Data are presented as the mean ± standard deviation. Statistical analyses were performed using one-way analysis of variance (ANOVA), followed by Tukey's post hoc test, using the OriginPro 2015 software (version 92E; OriginLab Corporation). A P-value <0.05 was considered to indicate a statistically significant difference. For clarity, only selected pairwise comparisons are shown in the figures.

To assess the effect of EGCG and TCS on the viability of A549 cells, MTT assays were performed 24 h after treatment. As shown in Fig. 2A, EGCG induced a significant and dose-dependent reduction in A549 cell viability. By contrast, TCS alone caused only a slight decrease in A549 cell viability, as ~90% of the control levels were maintained. Notably, however, the combined administration of EGCG and TCS produced a more substantial inhibition effect on A549 cell viability. The Bliss excess value (ΔE), as calculated in Table IA, confirmed a synergistic cytotoxic effect of EGCG and TCS, with ΔE >0 at EGCG concentrations ranging from 30 to 60 µM. Consistent with the MTT results, morphological examination of the cells under a light microscope at 24 h post-treatment (Fig. 2B) demonstrated that the combined treatment of EGCG and TCS induced pronounced morphological alterations in A549 cells, including reduced cell density, cellular shrinkage, and an increased proportion of non-adherent cells. To further validate that these effects were more widely applicable, the MTT assay was subsequently performed on a different NSCLC cell line, NCI-H460. The results obtained (as shown in Fig. 2C) confirmed that a synergistic cytotoxic effect also occurred upon treatment with EGCG and TCS in NCI H460 cells, with the corresponding Bliss excess values presented in Table IB. According to the data from Table IA, the combination of 50 µM EGCG and TCS produced the highest Bliss excess value in A549 cells, thereby justifying the selection of 50 µM EGCG for subsequent experiments in the present study. Notably, as shown in Fig. 2D, the combination of EGCG and TCS did not lead to a reduction in the viability of the IMR-90 normal lung fibroblasts, demonstrating that this combined treatment is well tolerated by non-malignant lung fibroblasts. To further evaluate whether this combined treatment could impair long-term reproductive viability, a colony formation assay was performed using A549 cells. As shown in Fig. S1A and B, the combination of EGCG and TCS significantly suppressed the clonogenic capability of the A549 cells, resulting in a significantly lower colony count compared with single treatments. Taken together, these results showed that EGCG and TCS work synergistically to inhibit both the short-term metabolic viability, and the long-term clonogenic survival, of NSCLC cells, while exhibiting minimal cytotoxicity on IMR-90 normal lung fibroblasts in vitro.

To elucidate the mechanism underlying the synergistic effect of EGCG and TCS, it was first investigated whether these treatments could influence the regulation of HSPs in A549 cells. Western blot analysis was employed to quantify the expression levels of HSP70 and HSP105. As shown in Fig. 3A and B, administering TCS increased the expression levels of both HSP70 and HSP105 in A549 cells, whereas combined treatment with EGCG significantly reduced this induction of protein expression. In addition to HSPs, the expression levels of the proteasome subunits, PSMA3 and PSMC3, were also evaluated via western blot analysis. It was revealed that TCS alone exerted a negligible effect on PSMA3 expression, and induced only a slight reduction in PSMC3 levels (Fig. 3C and D). By contrast, EGCG was found to significantly decrease the expression of both PSMA3 and PSMC3 in A549 cells treated with TCS. To evaluate the overall status of cellular proteostasis under these conditions, the levels of total ubiquitinated proteins were subsequently measured. As shown in Fig. S2A and B, co-treatment with EGCG and TCS resulted in a significant increase in the accumulation of ubiquitinated proteins compared with either treatment in isolation. Taken together, these findings suggested that EGCG suppresses the expression of critical components of the proteostatic machinery, including HSP70, HSP105, and the proteasome subunits. Moreover, this inhibition is associated with an increased accumulation of ubiquitinated proteins in the combination treatment group, reflecting heightened proteostatic dysfunction in A549 cells.

The intracellular accumulation of unfolded proteins is known to provoke ER stress, which, if sustained or severe, may culminate in cellular damage (23). In the present study, the involvement of the PERK-eIF2α signaling axis, a key ER stress pathway, was therefore explored in mediating cell death induced by EGCG and TCS. As shown in Fig. 4A, administration of the PERK inhibitor GSK2606414 partially mitigated the reduction in A549 cell viability caused by combined EGCG and TCS treatment in a dose-dependent manner. Additionally, western blot analysis of the downstream signaling components revealed that TCS significantly enhanced the phosphorylation of eIF2α, with the combination of EGCG and TCS producing an even more pronounced effect (Fig. 4B). Concurrently, TCS treatment alone induced substantial XBP1 mRNA splicing (Fig. S3A), thereby confirming IRE1α activation. Notably, co-treatment with EGCG and TCS led to a significant enhancement in JNK phosphorylation (Fig. S3B), a pro-apoptotic signaling event that can be triggered either downstream of IRE1α or via ROS-apoptosis signal-regulating kinase 1 (ASK1)-dependent pathways. These findings suggest that the combination of EGCG and TCS activates both the PERK-eIF2α and the IRE1α-JNK pathways of ER stress signaling, which may partially contribute to the observed reduction in viability of A549 cells.

ER stress is frequently associated with intracellular Ca²⁺ dysregulation, in which Ca²⁺ stored in the ER can be released via IP₃R channels (24). Therefore, cytosolic Ca²⁺ levels were further quantitatively measured using the Fluo-4 AM fluorescent dye via flow cytometric analysis. As shown in Fig. 5A and B, treatment with EGCG alone did not significantly affect cytosolic Ca²⁺ levels in A549 cells, whereas TCS alone led to only a slight increase. By contrast, the combination of EGCG and TCS produced a substantial elevation in cytosolic Ca²⁺ levels. Notably, this increase was found to be greatly attenuated through administering the IP₃R inhibitor 2-APB at a concentration of 50 µM, suggesting that the elevation in Ca²⁺ level might be mediated via the activation of IP₃R channels. To further elucidate the role of ER Ca²⁺ release in the synergistic cytotoxicity induced by EGCG and TCS in A549 cells, a rescue experiment employing 2-APB was performed using the MTT viability assay. As revealed in Fig. 5C, 2-APB significantly ameliorated the combined treatment's reduction of A549 cell viability in a dose-dependent manner. In addition to its effect on IP₃R channels, 2-APB is also known to affect thermosensitive transient receptor potential (TRP) channels, such as TRPV1 and TRPM2 (25,26). However, in the experiments of the present study, it was found that selective inhibitors for TRPM2 (JNJ-28583113) and TRPV1 (SB-705498) did not reverse the synergistic cytotoxic effects of EGCG and TCS (Fig. 5D), suggesting that these TRP channels are unlikely to mediate a significant role in this process. Taken together, these pharmacological rescue experiments suggested the potential involvement of IP₃R channels and associated ER Ca²⁺ release in contributing to the synergistic anticancer effects induced by the combined EGCG and TCS treatment.

ER stress and intracellular Ca²⁺ dysregulation have previously been shown to coincide with excessive ROS generation (5). In the present study, intracellular ROS levels in A549 cells were measured using the fluorescent dye DHE via flow cytometric analysis. As demonstrated in Fig. 6A and the resultant quantification of the data in Fig. 6B, treatment with either EGCG or TCS alone resulted in only a slight, and statistically non-significant, increase in ROS levels. On the other hand, the combined administration of EGCG and TCS produced a pronounced and statistically significant elevation of ROS levels in A549 cells. To explore the role of ROS in the reduction of cell viability caused by this combination, a rescue experiment employing the antioxidant NAC was performed using the MTT assay. The results shown in Fig. 6C demonstrate that NAC could effectively mitigate the synergistic cytotoxicity induced by the combination of EGCG and TCS in A549 cells, suggesting that the excessive ROS production may contribute to the reduced viability in A549 cells.

To explore the potential association between ER Ca²⁺ release and ROS generation, ROS levels in A549 cells were first assessed by DHE staining in the presence of 50 µM 2-APB, an inhibitor of IP₃R-mediated Ca²⁺ release. As shown in Fig. 7A and B, 2-APB treatment led to a significant attenuation of the ROS elevation triggered by the combined EGCG and TCS treatment, suggesting that ER Ca²⁺ release through IP₃R channels may contribute to the substantial increase in ROS levels. On the other hand, the cytosolic Ca²⁺ levels were measured using Fluo-4 AM staining in the presence of 2 mM NAC (Fig. 7C and D). These experiments revealed that NAC treatment effectively suppressed the rise in cytosolic Ca²⁺ levels that was induced by the combined EGCG and TCS treatment, suggesting that elevated ROS levels may facilitate ER Ca²⁺ release. Collectively, these convergent pharmacological findings support a working model of reciprocal interplay between the ER Ca²⁺ release and ROS generation events, wherein mutual amplification of these signals culminates in enhanced cell death.

Excessive ROS accumulation and mitochondrial Ca²⁺ overload are known to cause mitochondrial damage (27,28). Given that, in these experiments, the combined EGCG and TCS treatment effectively triggered ER Ca²⁺ release and an increase in ROS levels, it was sought to further evaluate changes in the MMP using the fluorescent dye JC-1 via flow cytometric analysis. As revealed in Fig. 8A, the combined EGCG and TCS treatment caused a significant shift in JC-1 fluorescence from red to green, indicative of MMP disruption. The quantitative analysis shown in Fig. 8C demonstrates that the combined treatment caused a more substantial loss of MMP compared with either treatment alone. Mitochondrial damage may eventually activate the intrinsic apoptotic pathway (27,28). Accordingly, the evaluation of apoptosis was performed using annexin V-FITC/PI double staining by flow cytometric analysis. As demonstrated in Fig. 8B and D, the combination of EGCG and TCS caused a significant increase in the proportion of apoptotic cells compared with single treatments. To further confirm the occurrence of apoptosis at the molecular level, the cleavage of PARP, a well-established indicator of caspase activation, was examined. As shown in Fig. 8E, western blot analysis demonstrated that the combined treatment with EGCG and TCS led to a significant enhancement in the cleavage of full-length PARP (116 kDa) into its characteristic 89-kDa inactive fragment. Collectively, these results confirmed that the combined treatment of EGCG with TCS effectively induced mitochondrial apoptosis in A549 cells.