Human NSCLC cell lines (H1299 and H460) and human normal lung epithelial cells (BEAS-2B) were purchased from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. The cells were cultured routinely in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) containing 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) in a 5% CO₂ atmosphere at 37°C. Normal human liver cells (MIHA) were purchased from the Cell Bank of Chinese Academy of Sciences, and human umbilical vein endothelial cells (HUVEC) were obtained from the American Type Culture Collection (cat. no. CRL-1730), which were cultured in F-12K medium supplemented with 10% FBS and endothelial cell growth supplement at 37°C in a humidified cell incubator with a supply of 5% CO₂.

Cisplatin was obtained from TargetMol, PD was purchased from Selleck Chemicals (purity: >97%), carboplatin was obtained from MedChemExpress and N-Acetyl-l-Cysteine (NAC) was provided from Sigma Aldrich; Merck KGaA. 3-(4,5-dimethyl thiazol-2-yl) 2,5-diphenyl tetrazolium bromide (MTT) was purchased from Beijing Solarbio Science & Technology Co., Ltd. ROS probe 2',7'-dichlorodihydrofluorescin diacetate (DCFH-DA) and H&E staining kit were supplied from Beyotime Institute of Biotechnology. Antibodies against activating transcription factor 4 (ATF4; cat. no. 11815S), eukaryotic initiation factor 2 (eIF2α; cat. no. 9722S), phosphorylated eIF2α (p-eIF2α; cat. no. 3398S), cleaved-Caspase3 (cat. no. 9664S), Caspase3 (cat. no. 9662S), JNK (cat. no. 9252T), p-JNK (cat. no. 4668S), p38 MAPK (cat. no. 9212S), p-p38 MAPK (cat. no. 9211S), GAPGH (cat. no. 5174S), HRP-linked anti-rat IgG antibody (cat. no. 7077S) and HRP-linked anti-mouse IgG antibody (cat. no. 7076S) were purchased from Cell Signaling Technology, Inc. Ki-67 (cat. no. ab16667) was purchased from Abcam. NOX5 (cat. no. 25,350-1-AP), Vinculin (cat. no. 66,305-1-Ig) and Bcl-2 (cat. no. 60,178-1-Ig) antibodies were purchased from Proteintech Group, Inc.

The MTT assay was employed to assess the effects of PD, cisplatin and their combination on cell viability. In 96-well plates, H1299 and H460 cells were seeded at a density of 3×10³ or 5×10³ cells per well, respectively, and allowed to adhere for 24 h. BEAS-2B, MIHA and HUVEC cells were seeded at a density of 3×10³ per well and incubated for 24 h. The cells were treated with various concentrations of PD (0, 100, 200, 300, 400, 500 and 600 μM) for 72 h. For the combined treatment, the cells were treated with PD, cisplatin or their combination at indicated concentrations for 72 h. After the completion of the incubation period, 25 μl of MTT reagent was added to each well and allowed to incubate for an additional 3 h. Subsequently, 150 μl of DMSO was added to each well to dissolve the formazan product. The absorbance of each sample was measured at a wavelength of 490 nm using a SpectraMax iD3 instrument (Molecular Devices, LLC). The Logit approach was used to evaluate the IC₅₀ values. The CompuSyn 2.0 software (http://www.combosyn.com/index.html) was utilized to compute the combination index (CI) to evaluate drug interactions. It is noteworthy that a CI value of 1 signifies an additive effect when two agents are combined, while a CI >1 or CI <1 indicates an antagonistic or synergistic interaction, respectively.

The cells were seeded at a density of 2×10³ cells per well in 6-well plates and were allowed to adhere in 5% CO₂ atmosphere at 37°C overnight. Subsequently, the cells were subjected to treatment with varying concentrations of PD (0, 25, 50 and 100 μM for H1299 cells and 0, 200, 400, 800 μM for H460 cells) for 72 h. For the combination therapy, the cells were treated with cisplatin (1.5 μM), PD (25 μM for H1299 cells and 500 μM for H460 cells), or their combination with or without 5 mM NAC pre-treatment for 2 h. After 7-10 days incubation, the colonies were subjected to a washing procedure using phosphate-buffered saline (PBS). Subsequently, a fixative solution consisting of 4% paraformaldehyde (PFA) was applied for a duration of 15 min at room temperature. Following fixation, 0.5% Gentian violet staining was performed for a period of 10 min at room temperature (RT). ImageJ software (version 1.53t; National Institutes of Health) was used to quantify colonies (consisting of >50 cells).

Wound healing analysis was employed to assess the capability of cell motility. The H1299 NSCLC cells were seeded into 6-well plates at a density of 3×10⁵, cultured with RPMI-1640 medium containing 2% FBS, and allowed to reach confluent monolayers. Then multiple lines were drawn in the cell monolayers with a bacteria-free 10 μl pipette tip to create wound area. The scratched cells were then removed from the plates by washing with PBS. H1299 cells were treated with cisplatin (1.5 μM), PD (20 μM), or a combination of both, with or without 5 mM NAC pretreatment for 1.5 h. Following a 48-h incubation period, the old medium was replaced with fresh RPMI-1640 medium supplemented with 2% FBS. The evaluation of the combined treatment on cell migration was performed by capturing images at the point when the wound in the control group had healed virtually or completely. The results of this assay were analyzed to determine the effects of the combined treatment on cell migration.

In order to measure intracellular ROS generation, the cells were treated with the ROS-sensitive dye (DCFH-DA), and analyzed using flow cytometry. Specifically, the NSCLC cells were seeded into 6-well plates and cultured overnight in standard culture medium. The cells were exposed to PD (500 μM) for 3, 6, 9, 12 and 15 h. For combination treatment, the cells were exposed with cisplatin (60 μM for H1299 and 50 μM for H460), PD (500 μM), or a combination of both for a duration of 12 h. In the case of the combination group, NAC pretreatment was conducted at a concentration of 5 mM for a duration of 1.5 h prior to treatment. Following treatment, the cells were stained with 10 μM DCFH-DA and incubated at 37°C in the absence of light for a period of 30 min. The ROS production (DCF fluorescence) was measured by flow cytometry using FACS Calibur (BD Biosciences), and data was analyzed using the FlowJo software (Tree Star, Inc.).

The homogenization of cells or tumor tissues was accomplished by using protein RIPA lysis buffer (cat. no. AR0103-100, Boster Biological Technology). Subsequently, lysates were subjected to centrifugation at 4°C for 15 min at 13,400 × g to remove any undissolved debris. The Bradford protein assay kit (Bio-Rad Laboratories, Inc.) was used to measure the concentrations of total proteins extracted from cells or tumor tissues. The equal amount of protein (60 μg) and loading buffer were mixed, and separated by using a 10-12% SDS-PAGE. Then the proteins were transferred to PVDF membranes, and blocked with 5% fresh non-fat milk in TBST containing 0.1% Tween 20 for 2 h at RT. The membranes underwent incubation with the primary antibodies specific to the experiments overnight at 4°C, which were ATF4 (1:1,000), eIF2α (1:1,000), p-eIF2α (1:1,000), cleaved-Caspase3 (1:1,000), Caspase3 (1:1,000), JNK (1:1,000), p-JNK (1:1,000), p38 MAPK (1:1,000), p-p38 MAPK (1:1,000), GAPGH (1:10,000), NOX5 (1:1,000), Vinculin (1:5,000) and Bcl-2 (1:2,000). Subsequently, the membranes were subjected to three wash cycles using TBST to remove any unbound antibodies. After washing, the membranes underwent further incubation with corresponding horseradish peroxidase-conjugated secondary antibodies [HRP-linked anti-rat IgG antibody (1:4,000) and HRP-linked anti-mouse IgG antibody (1:4,000)] for 1 h at RT. The ECL detection kit was used to detect immunoreactivity (Bio-Rad Laboratories, Inc.). ImageJ software (version 1.53t; National Institutes of Health) was used to conduct densitometric measurements.

Athymic BALB/c female nude mice (age, 5 weeks-old; weight, 18-20 g; n=25) were purchased from the Vital River Experimental Animal Center. All animal experiments were carried out in accordance with the Wenzhou Medical University's Institutional Animal Care and Use Committee (IACUC) guidelines (approval no. xmsq2022-0602; Wenzhou, China). The mice were maintained under specific pathogen-free conditions with stable RT (20-25°C) and humidity (50-60%), following a 12-h light/dark cycle, and provided with a standard rodent diet and water ad libitum. The 2.5×10⁶ H460 cells per 100 μl of PBS/Matrigel mixture (1:1) were subcutaneously injected into the flanks of nude mice. The mice were divided into five experimental groups (n=5) when the tumor volumes reached to ~100 mm³. The tumor bearing mice were intraperitoneally administered 2 mg/kg cisplatin every three days, 50 mg/kg PD every two days, or their combination for two weeks. The control group was administered vehicle only, and ROS inhibitor (NAC) was administered in drinking water (0.5 g/l) for the combination group. The health and behavior of the mice were monitored daily, and tumor length (L), width (W) and mouse weight were measured every 2 days after treatment. The tumor volume was calculated using the following formula: V=1/2 × L × W². When the tumors reached nearly 15 mm in diameter on day 24, all the mice were sacrificed. No mice reached humane endpoints at the end of the experiments, defined as either losing 20% of body weight or exhibiting moribund behavior. At the experimental endpoint, all mice were euthanized by intraperitoneal injection of pentobarbital sodium (150 mg/kg) followed by cervical dislocation. The confirmation of mouse death was further verified by visually inspecting for cardiac and respiratory arrest. The tumor tissues were excised and weighed for further investigation. Histological examination of certain essential organs, including the heart, liver, kidney and lung were evaluated by using hematoxylin and eosin (H&E) staining.

The tumor tissue samples were fixed with 4% PFA for 24 h at room temperature and then embedded in paraffin. The paraffin blocks were sliced in 5-μm-thick sections and deparaffinized in xylene (Changshu Chemicals Co., Ltd.) and dehydrated with an ethanol gradient. Endogenous peroxidase activity was blocked by incubation with 3% H₂O₂ in methanol at room temperature for 15 min. The IHC was carried out according to the routine procedure. Specifically, the tumor sections were subjected to overnight incubation at 4°C with a primary antibody targeting Ki-67 (1:100), followed by subsequent incubation with HRP-linked anti-rat IgG antibody (1:1,000) for 1 h at ambient temperature. Detection of the targeted antigen was achieved through employment of 3,3'-diaminobenzidine (DAB) for color detection. The histological investigation and potential toxicity assessment were carried out by H&E staining (hematoxylin for 5 min and 0.5% eosin for 1-3 min) at room temperature with heart, liver, kidney and lung tissues. The immunostained sections were observed using a confocal microscope (magnification, x200; Leica Microsystems GmbH).

H1299 cells (3×10⁵ per well) or H460 cells (5×10⁵ per well) were seeded in 6-well plates and incubated for 24 h at 37°C. The NOX5 small interfering (si)RNA (Shanghai GenePharma Co., Ltd.) or non-targeting control siRNA were transfected into the cells with final concentration of 30 nM using lipofectamine 3000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.), and incubated for 24 h at 37°C. Following incubation, the cells were treated with various concentrations of PD for different time-intervals according to experimental requirement. The effectiveness of knockdown was evaluated via western blotting and reverse transcription-quantitative PCR (RT-qPCR), and the siRNA sequences are provided in Table SI. This methodology is in accordance with established academic protocols for transfecting cells and evaluating knockdown efficacy.

Total RNAs were extracted by TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. For cDNA synthesis, 1 μg of total RNA was reverse transcribed using PrimeScript RT Master Mix (cat. no. RR036A; Takara Biotechnology Co., Ltd.) in accordance with the manufacturer's protocol. RT-qPCR was performed by using TB Green Premix Ex Taq II reagent (cat. no. RR820A; Takara) according to the manufacturer's protocol. Briefly, initial denaturation was 30 sec at 95°C, followed by 40 cycles of denaturation (5 sec at 95°C), annealing (34 sec at 60°C) and extension (10 sec at 95°C). The relative gene expression was determined using the 2^-ΔΔCq method (26), and GAPDH was used as an internal control. The primer sequences utilized for the gene amplification are provided in Table SII.

All experiments were performed in triplicate (n=3). The resulting data was presented as the mean ± standard deviation (SD), and analyzed using GraphPad Prism 8.0 software (Dotmatics). Statistical analysis was conducted using two-sample t-tests for comparison between two groups, or one-way ANOVA with Tukey's multiple comparisons test for comparison between multiple groups. P<0.05 was considered to indicate a statistically significant difference.

Previous studies have reported that PD exerted cytotoxicity in various tumor cells with IC₅₀ values ranging from 200 to 500 μM (27-29). To investigate the function of PD on NSCLC cell proliferation, H1299 and H460 cells were treated with varying dosages of PD for 72 h. The chemical structure of PD (PubChem CID: 5281718) is demonstrated in Fig. S1A. As revealed in Fig. 1A and B, PD significantly suppressed both NSCLC cell proliferation in a concentration-dependent manner with IC₅₀ values of ~500 μM. Additionally, the same experiments were conducted in normal human cells. As expected, PD showed less cytotoxicity against BEAS-2B (Fig. S1B), MIHA (Fig. S1C) and HUVEC (Fig. S1D) cells when compared with NSCLC cells. In detail, 600 μM PD inhibited normal human cell proliferation only ~20% when compared with non-PD treatment group. Colony formation assay further identified that PD significantly reduced colony forming ability of both H1299 (Fig. 1C) and H460 (Fig. 1D) cells.

The effects of PD on NSCLC cell apoptosis were further investigated. High doses of PD markedly reduced the expression level of the anti-apoptotic protein Bcl-2, while increased the expression of the pro-apoptotic protein Cleaved-Caspase3 in both H1299 and H460 cells (Fig. 1E and F). These results suggested that the inhibition of cell proliferation caused by PD might partly be due to PD-induced apoptotic pathway in NSCLC.

PD has been revealed to have antitumor properties by generating ROS (30). Abundant ROS production stimulates oxidative stress, resulting in cancer cell death (31). To determine whether inhibition of NSCLC cell proliferation caused by PD is associated with the increased ROS generation, H1299 and H460 cells were treated with PD, followed by incubation with DCFH-DA. Treatment of both cell lines with PD elevated fluorescence intensity in a time-dependent manner, suggesting that PD treatment boosted ROS production (Fig. 2A and B). Increased ROS production triggers the ER stress response and promotes the expression levels of p-eIF2α and ATF4 (32,33). Treatment of both NSCLC cell lines (500 μM) with PD time-dependently elevated p-eIF2α and ATF4 expression levels (Fig. 2C and D). As expected, PD treatment dose-dependently elevated p-eIF2α and ATF4 expression levels in NSCLC cells (Fig. 2E and F), indicating that PD exerts antitumor effects in NSCLC cells by stimulating ROS-mediated ER stress pathway.

To evaluate whether the combined treatment has superior anti-proliferative effects than monotherapy, H1299 and H460 cells were treated with PD, cisplatin, or their combination. Treatment of NSCLC with varying doses of cisplatin inhibited cell viability of both H1299 and H460 cells in a dose-dependent manner. Additionally, 500 μM PD generally boosted cytotoxicity of cisplatin against H1299 cells, and robust synergistic inhibitory impact was observed in combined therapy with 60 μM cisplatin and 500 μM PD (Fig. 3A). In the same way, the application of 500 μM PD was found to augment cisplatin-induced cytotoxicity in H460 cells, and significant synergistic inhibitory effects were observed with combined therapy of 50 μM cisplatin and 500 μM PD (Fig. 3C). CompuSyn 2.0 software was utilized to calculate CI values assessing the interaction between PD and cisplatin. Notably, combinations of 60 μM cisplatin and 500 μM PD in H1299 cells (Fig. 3B) or 50 μM cisplatin and 500 μM PD in H460 cells exhibited the lowest CI values amongst different combinations (Fig. 3D), indicating synergistic effects between PD and cisplatin in NSCLC. Furthermore, colony formation assay revealed that combined treatment of H1299 (25 μM) and H460 (500 μM) cells with PD and cisplatin (1.5 μM) significantly reduced colony forming ability of both cell lines when compared with single treatment alone (Fig. 3E and F). Importantly, combined treatment-induced inhibitory effects of colony forming ability were significantly attenuated by the ROS scavenger, NAC. Wound healing assay also showed similar results as combination of low dose of PD and cisplatin markedly inhibited the migration of NSCLC cells, and NAC pre-treatment effectively reversed this effect (Fig. S2). Similar with cisplatin, combination therapy with carboplatin and PD exhibited stronger inhibitory effects on H1299 and H460 cell viability (Fig. S3A-C) and colony formation (Fig. S3D-F) than single treatment alone. These findings indicated that a low dose of PD could improve antitumor effects of cisplatin and carboplatin, and that ROS is an important mediator in the synergistic effects of PD and cisplatin in NSCLC cells.

Cancer cells have been observed to maintain higher levels of ROS compared with normal cells (34). This aspect provides an interesting therapeutic window since cancer cells might be more sensitive than normal cells to agents that induce ROS generation. In the present study, the intracellular ROS levels were examined in NSCLC cells after treatment with PD (500 μM), cisplatin (60 μM in H1299 cells and 50 μM in H460 cells), or combinations thereof. Combined treatment significantly induced generation of ROS when compared with PD or cisplatin alone, whilst NAC pre-treatment effectively prevented generation of ROS induced by combined treatments (Fig. 4A and B). ER stress-related proteins in cells were further investigated after treatment with PD, cisplatin, or their combination. Combined treatment markedly increased p-eIF2α and ATF4 expression in both H1299 and H460 cell lines, as compared with treatment with cisplatin or PD alone (Fig. 4C and E), whilst pre-treatment with NAC partly attenuated the expression of these ER stress-related proteins (Fig. 4D and F). These findings suggested that ROS-mediated ER stress functioned as a key role in the combined treatment, inducing synergistic antitumor activities in NSCLC cells.

Apoptosis is tightly linked to the activated JNK and p38 MAPK pathways (35). Activation of the JNK and p38 pathways is closely involved in oxidative stress-induced apoptosis (36). Thus, the p-JNK and p38 protein levels were investigated in H1299 and H460 cells following treatment with PD (500 μM), cisplatin (60 μM in H1299 cells or 50 μM in H460 cells), or their combination. Combined treatment of NSCLC cells with PD and cisplatin consistently elevated p-JNK and p38 expression (Fig. 5A and C), while pre-treatment with NAC effectively attenuated p-JNK and p-p38 levels induced by combined treatment (Fig. 5B and D). These results indicated that activation of ROS-mediated JNK and p38 MAPK signaling pathways contributed to the combined treatment, inducing synergistic antitumor activities.

To further investigate the synergistic antitumor effects of combined treatment with PD and cisplatin on NSCLC in vivo, H460 xenografts were established in nude mice. The mice were divided into five experimental groups (n=5), and were administered 2 mg/kg cisplatin every three days, 50 mg/kg PD every two days, or their combination for two weeks. The combined treatment group was further divided into two sub-groups, with one sub-group administered NAC in their drinking water (0.5 g/l). The results showed that the combined treatment significantly inhibited tumor growth in vivo compared with cisplatin or PD treatment alone, while this inhibitory effect was reversed in the NAC treatment sub-group (Figs. 6A and B and S4A). There were no notable differences in body weight changes between the combined treatment group and the other groups (Fig. S4B). Additionally, H&E staining demonstrated no histological alterations in vital organs (heart, lung, kidney and liver) (Fig. S4C), suggesting that the combined treatment was administered at rational drug doses and did not induce any significant cytotoxicity in the experimental subjects. Similar with in vitro results, the expression levels of p-eIF2α, ATF4, p-JNK and p-p38 in tumor tissues exhibited a marked increase in the combined treatment group as compared with single treatment groups. Moreover, pre-treatment with NAC consistently attenuated the increase of these proteins in tumor tissues from the combined treatment group (Fig. 6C). As expected, PD treatment increased NOX5 protein expression, whereas cisplatin did not, suggesting that cisplatin increases ROS levels through a NOX5-independent pathway. Accordingly, the combination of cisplatin and PD did not markedly increase NOX5 protein levels compared with PD treatment alone. Additionally, NAC pre-treatment did not reverse the combination treatment-induced NOX5 protein levels, indicating that ROS may not be upstream of NOX5. Furthermore, the IHC staining analysis demonstrated that the co-administration of cisplatin and PD resulted in a significant reduction of Ki-67 positive cells. Notably, the pre-treatment with NAC reversed the aforementioned effects induced by the combined treatment (Fig. 6D). These findings suggested that PD enhanced the antitumor activities of cisplatin in vivo at least partly by stimulating ROS-mediated ER stress, JNK and p38 MAPK signaling pathways.

NOX5 activation stimulates production of ROS and causes subsequent death of tumor cells (24,37,38). It was found that PD treatment increased generation of ROS in NSCLC cells. Thus, it was hypothesized that PD might promote NOX5 activity to regulate generation of ROS. Treatment of NSCLC cells with PD significantly increased both NOX5 mRNA and protein expression levels (Fig. 7A and B). Similarly, PD treatment consistently upregulated NOX5 protein expression in mouse xenograft tumor tissues (Fig. 7C). To further evaluate the impact of NOX5 in PD-induced generation of ROS, NOX5 was knocked down by two independent siRNAs. Knocking down NOX5 consistently reduced both NOX5 mRNA (Fig. 7D and E) and protein expression levels in H1299 and H460 NSCLC cells (Fig. 7F). Since si-217 exhibited stronger knockdown effect than si-1112, si-217 was used for further analyses. Additionally, knocking down NOX5 attenuated the inhibitory effects of PD on NSCLC colony formation (Figs. 7G and H; S5C and D) and cell viability (Figs. 7I and J; S5A and B). Furthermore, knocking down NOX5 reduced PD-induced production of ROS in NSCLC cells (Fig. 8A and B). Additionally, knocking down NOX5 partly diminished PD-induced activation of ER stress-related proteins (p-eIF2α and ATF4) (Fig. 8C and D), and the JNK and p38 MAPK pathways (Fig. 8E and F). These results indicated that NOX5 is a possible target of PD and an important mediator of PD-induced cell death in NSCLC cells.