Dihydroartemisinin, a potential PTGS1 inhibitor, potentiated cisplatin-induced cell death in non-small cell lung cancer through activating ROS-mediated multiple signaling pathways

Highlights

• •Combined therapy with DHA and cisplatin exerts synergistic anti-NSCLC activity through activating ROS-mediated ER stress, JNK and p38 MAPK signaling pathways both in vitro and vivo.
• •PTGS1 is identified as a potential novel target of DHA. Knockdown of PTGS1 enhanced DHA-induced cell death in NSCLC cells through stimulating ROS-mediated ER stress, JNK and p38 MAPK signaling pathways.

Abstract

Dihydroartemisinin (DHA) exerts an anti-tumor effect in multiple cancers, however, the molecular mechanism of DHA and whether DHA facilitates the anti-tumor efficacy of cisplatin in non-small cell lung cancer (NSCLC) are unclear. Here, we found that DHA potentiated the anti-tumor effects of cisplatin in NSCLC cells by stimulating reactive oxygen species (ROS)-mediated endoplasmic reticulum (ER) stress, C-Jun-amino-terminal kinase (JNK) and p38 MAPK signaling pathways both in vitro and in vivo. Of note, we demonstrated for the first time that DHA inhibits prostaglandin G/H synthase 1 (PTGS1) expression, resulting in enhanced ROS production. Importantly, silencing PTGS1 sensitized DHA-induced cell death by increasing ROS production and activating ER-stress, JNK and p38 MAPK signaling pathways. In summary, our findings provided new experimental basis and therapeutic prospect for the combined therapy with DHA and cisplatin in some NSCLC patients.

Graphical abstract

Introduction

Lung cancer stands as one of the most prevalent malignant tumors, presenting significant health concerns characterized by its poor prognosis, high incidence, and mortality rates [1]. Non-small cell lung cancer (NSCLC) comprises a major subtype, representing approximately 85 % of all lung cancer cases [2]. Currently, chemotherapy remains the standard first-line treatment approach for advanced NSCLC [3,4]. However, considering the severe cytotoxicity and the emergence of drug resistance associated with the chemotherapeutic agents treating NSCLC, there is a compelling and imperative need to develop combined therapeutic strategy with strong anti-tumor effects and minor cytotoxicity.

Research on natural compounds has attracted significant attention recently, especially regarding cancer treatment. In the last four decades, more than 30 % of drugs sanctioned by the Food and Drug Administration (FDA) have originated directly from natural compounds [5]. Dihydroartemisinin (DHA) is purified from the Artemisia annua, a traditional Chinese herbal medicine, and features a unique endoperoxide bridge structure. DHA has been employed in the treatment of malaria over several decades, demonstrating both safety and efficacy. Recent studies have demonstrated anti-proliferative effects of DHA on liver [6], breast [7], colon [8], and cervical cancer cells [9]. The anti-tumor properties of DHA are primary owing to its ability to inhibit angiogenesis, induce apoptosis, and regulate genes associated with tumorigenesis [10,11]. DHA exerts anti-cancer activity by elevating ROS accumulation in cancer cells [12]. DHA triggered DNA damage and endoplasmic reticulum (ER) stress pathways through the induction of ferroptosis, ultimately leading to immunogenic cell death in lung cancer cells [13]. Cisplatin belongs to the platinum class and serves as a first-line adjuvant chemotherapy agent for treating NSCLC. Over the past few decades, the anti-tumor mechanism of cisplatin has been thoroughly studied [14]. Cisplatin acts primarily on DNA molecules and mediates cell death by affecting distinct signal transduction pathways, including MAPK, p53, and Nrf2 [15,16]. Extensive research indicates that cisplatin combined with other herbal compounds exerted superior anti-tumor activity than cisplatin alone [17]. Combined therapy with cisplatin and resveratrol suppressed metastasis, and induced apoptosis by targeting p16/p21 and p38/p53 pathways in gastric cancer [18]. Nevertheless, the therapeutic efficacy and the underlying molecular mechanism of combination treatment of DHA and cisplatin in NSCLC remain largely uncharacterized.

Prostaglandin endoperoxidase (PTGS), alternatively called cyclooxygenase (COX), is a pivotal enzyme in the biosynthesis of prostaglandins [19]. PTGS comprises two isozymes, namely constitutive prostaglandin G/H synthase 1 (PTGS1) and inducible prostaglandin-endoperoxide synthase 2 (PTGS2). PTGS1 is implicated in the progression of arthritic diseases, inflammation, and cancer [20], [21], [22]. Relatively high COX-1 expression was observed in breast, ovarian epithelial, and colon cancer cells [23], [24], [25]. The metabolic reaction network analyses on pan-cancer showed that PTGS1 was an oncogenic component in cancers [26]. Additionally, COX-1 inhibitors reduced the metastatic potential of breast cancer [27], and inhibited alanine aminotransferase (ALT), aspartate aminotransferase (AST), nitric oxide (NO), and myeloperoxidase (MPO) expressions in NSCLC cells [28], suggesting that PTGS1 is a potential chemotherapeutic target for nonsteroidal anti-inflammatory drugs. Increasing evidences suggested that COX pathway activation leads to ROS production by influencing various enzymes responsible for ROS generation [29]. It has been reported that PTGS1 was involved in malondialdehyde (MDA) synthesis, an oxidative stress biomarker [30], indicating that silencing PTGS1 may disrupt the synthesis of MDA, potentially leading to a redox imbalance. Despite the accumulating evidence highlighting the significance of PTGS1 in tumor progression and oxidative stress, the underlying molecular mechanisms and the interaction between DHA and PTGS1 in anti-NSCLC activity remain poorly understood.

In the present study, we observed that the combination of DHA with cisplatin synergistically increased the anti-cancer effects in NSCLC cells via activating ROS-mediated ER stress, JNK and p38 MAPK pathways. Additionally, DHA inhibited PTGS1 expression, and depletion of PTGS1 enhanced anti-NSCLC cells activity of DHA by coordinately elevating ROS production and stimulating ER-stress, suggesting a critical role of PTGS1 in DHA-mediated growth inhibition of NSCLC cells.

Materials and methods

Cell culture

A549, a NSCLC cell line, was obtained from the American Type Culture Collection (ATCC), and H460 was acquired from Shanghai Institute of Biochemistry and Cell Biology. In a humidified incubator, the A549 and H460 cells were incubated in RPMI-1640 medium with 10 % fetal bovine serum and 1 % penicillin/streptomycin (New Cell & Molecular Biotech, Suzhou, China). The incubator was adjusted to 37 °C maintaining 5 % CO₂ atmosphere.

Reagents

DHA was acquired from Selleck Chemicals (Houston, TX, USA), while cisplatin was provided from Sigma (St. Louis, MO, USA). Dimethylsulfoxide (DMSO) (Sigma, St. Louis, USA) was utilized to attain the dissolved form of cisplatin. SP600125 (JNK inhibitor) and SB 203580 (P38 inhibitor) were purchased from TargetMol (Shanghai, China). DHA, cisplatin, SP600125 and SB 203580 were preserved as 20 mM solution, and frozen in separate aliquots at a temperature of −80 °C. N-acetyl-L-cysteine (NAC) was provided by Sigma-Aldrich (St. Louis, MO, USA), and was resolved in phosphate-buffered saline (PBS, 137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄, pH 7.4). The NAC was stored as 1 mM solution and frozen in aliquots at a temperature of −20 °C.

Antibodies against phosphorylated eukaryotic initiation factor 2α (p-eIF2α, 1:1000, cat. no: 3398S), phosphorylated p38 mitogen-activated protein kinase (p-p38 MAPK, 1:1000, cat. no: 9211S), p38 MAPK (1:1000, cat. no: 8690S), phosphorylated c-Jun-amino-terminal kinase (p-JNK, 1:1000, cat. no: 4668S), JNK (1:1000, cat. no: 9252S), activating transcription factor 4 (ATF4, 1:1000, cat. no: 11815S), cleaved caspase 3 (1:1000, cat. no: 9664S), caspase 3 (1:1000, cat. no: 9662S), HRP-linked anti-rat IgG antibody (1:4000, cat. no: 7077S) and HRP-linked anti-mouse IgG antibody (1:4000, cat. no: 7076S) were acquired from Cell Signaling Technology (Danvers, MA USA). The primary antibodies against GAPDH (1:10000, cat. no: 10494-1-AP), eIF2α (1:1000, cat. no: 26520-1-AP), Bcl-2 (1:1000, cat. no: 26593-1-AP), PTGS1 (1:1000, cat. no: 13393-1-AP) and vinculin (1:1000, cat. no: 26520-1-AP) were ordered from Proteintech (Wuhan, China). The Ki67 (1:1000, cat. no: ab15580) antibody was provided by Abcam (Cambridge, MA, USA). 2′, 7′-dichlorodihydrofluorescin diacetate (DCFH-DA) probe for ROS, and Hematoxylin and Eosin (H&E) Staining Kit were bought from Beyotime (Shanghai, China). Diphenyltetrazolium Bromide (MTT) was acquired by Keygene Biotech (Nanjing, China).

Cell viability assay

A549 and H460 cells were plated into 96-well cell culture plates (4–5 × 10³ cells/100 µl per well) and cultured in RPMI1640 medium with 10 % FBS. Twenty-four hours later, DHA was added into A549 cells at various concentrations (0, 0.2, 0.5, 5, 10, 20, 30, 40, 50 µM). Following treatment for 48 h, MTT reagent was introduced (5 mg/mL, 25 µl/well) and left to incubate for another 3 hours. After incubation, the medium was carefully removed. Formazan was dissolved in DMSO (150 µl/well), shaking for 5 min, and absorbance measurement was carried out at 490 nm using a SpectraMax iD3 microplate reader (MD, USA). Likewise, H460 cells underwent treatment with diverse doses (0, 1, 10, 20, 40, 80, 100, 120, 150 µM) of DHA, and the same experimental procedures were performed. The dose-response curves were drawn with GraphPad Prism software, and the half-maximal inhibitory concentrations (IC₅₀) of the two NSCLC cells were obtained from the Logit technique.

Combination treatment was performed on A549 cells and H460 cells. The cells (1.5–2.0 × 10⁵ cells/well) were plated into 6-well cell culture plates and allowed to incubate for 24 h. After incubation, the cells were exposed to 20 µM DHA and various concentrations of cisplatin (10, 20, 30, 40, 50 µM) for 48 h. According to experimental requirement, the cells underwent pretreatment with SP600125 (10 µM), SB203580 (20 µM), or NAC (5 mM) prior to DHA treatment. Then, microscope images were captured, and cell viability was assessed by counting the cell numbers following trypan blue staining. CompuSyn software was utilized to calculate the combination index (CI) for assessing the interaction between DHA and cisplatin. There are three categories for combination effects: additive effect (CI = 1); antagonism effect (CI > 1); synergism effect (CI < 1).

Cell colony-formation assay

A549 and H460 cells (1500 cells per well) suspension using RPMI-1640 medium supplemented with 10 % FBS were seeded into 6-well plates overnight. Divergent concentrations of DHA (0, 5, 10, 20, 30 and 40 µM for A549; 0, 12.5, 25, 50 and 75 µM for H460) were used to expose the cells for 24 h. Additionally, A549 and H460 cells were subjected to 1.5 µM cisplatin, 5 µM DHA or combination treatment with 5 mM NAC (ROS scavenger) exposure for 2 h. As per experimental requirement, cells were pre-treated with SP600125 (1.5 µM) or SB203580 (2.5 µM) prior to DHA treatment. DMSO was added into the cells as solvent control. After incubation, the medium was replaced and cultured at 37  °C for a period of 7 to 10 days. Finally, fixation with 4 % paraformaldehyde was carried out on the cells, and crystal violet (Beyotime, China) stain were applied to the cells. Image J software was utilized to analyze the colony number.

Measurement of intracellular ROS

A549 and H460 cells were seeded into 6-well cell culture plates and left to incubate for 24 h. Subsequently, 20 µM DHA was added into A549 and H460 cells at different time intervals. For combined treatment, the cells were subjected to 20 µM DHA, 30 µM (A549) or 20 µM (H460) cisplatin, or combination treatment with 5 mM NAC exposure for 1 h. In addition, ROS levels were assessed in PTGS1 knockdown A549 cells and negative control cells after treatment with or without DHA. 10 µM ROS-sensitive DCFH-DA (Beyotime Biotech, Nantong, China) was utilized to stain the cells in culture medium without FBS for 0.5 hour at 37 °C. Following two washes with PBS, the fluorescence intensity was detected by a FACS Calibur flow cytometer (BD Biosciences, CA, USA) or fluorescence microscope (Leica, Germany).

RNA isolation and quantitative real-time PCR (qRT-PCR)

The cells were subjected to siRNA or DHA, and total RNA was extracted by using TRIzol (Takara, Japan). To evaluate the quality and integrity of RNA, Nanodrop One (Thermo Fisher Scientific, USA) was utilized. Following the manufacturer's instructions, total RNA (1 µg) was reverse-transcribed into cDNAs using the PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Japan). The qRT-PCR was performed by TB Green Fast qPCR mix (Takara, Japan) following the guidelines provided by the manufacturer. All reactions were done in triplicates. Table S1 contained primer sequences. The data were standardized to GAPDH housekeeping gene.

Transient transfection of small interfering RNA (siRNA)

The siRNA duplexes and Negative Universal Control in this study were purchased from Genepharma (Shanghai, China) and sequences are shown in Table S2. In accordance with the guidelines provided by the manufacturer, the cells (2×10⁵ cells/well) were plated in 6-well cell culture plates and incubated for 24 h, and the siRNAs against human PTGS1 or control siRNA were transfected into the cells by lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA).

Western blotting analysis

Cells or tumor tissues were lysed in protein lysis buffer with protease/phosphatase inhibitor cocktail (Boster, China), and centrifuged at 12,000 rpm for 10 min at 4 °C to remove debris. The protein concentrations were measured by using the Bradford protein assay kit (Bio-Rad, Herculer, CA). Protein samples were subjected to electrophoresis on 10 % or 12 % polyacrylamide gels, and the separated proteins were transferred onto PVDF membranes (Bio-Rad). The membrane was incubated with fresh 5 % skimmed milk in TBST for 1.5 h at room temperature before being incubated overnight at 4 °C with indicated primary antibodies. Following washing the membrane with TBST three times, the membranes were further incubated in HRP-linked secondary antibodies for 1 h at room temperature. Finally, ECL kit was employed to visualize the protein bands (Bio-Rad, Hercules, CA).

Wound healing assay

To further investigate the impact of combination therapy on cell migration, A549 cells (5×10⁵ cell/well) were placed in 6-well cell culture plates for 24 h. The following day, the cells were scratched with a sterilized pipette tip (10 µl) in the cell monolayer, and rinsed with PBS (pH 7.4). The cells were subjected to 2.5 µM DHA, 3 µM cisplatin, or a combination of both for either 48 h or 96 h. RPMI 1640 medium supplied with 2 % FBS was provided along with or without 5 mM NAC pretreatment for 1 h. An inverted microscope (Olympus, Tokyo, Japan) was used to capture the images. Scratch width was measured by using Image J software. Three separated experiments were performed.

Protein - protein intersection network construction

Super-PRED was utilized to forecast the potential targets of DNA [31], whereas NSCLC targets were analyzed by using GeneCards [32]. The overlapped predicted genes between DHA and NSCLC networks were considered as common targeted genes. To explore the interaction among proteins, pathways and co-expression, we used the STRING database (https://cn.string-db.org/) to examine common targeted genes. We selected the genes with confidence score ≥ 0.4, and removed the nodes that disconnected from the network. We visualized the PPI network using Cytoscape software (version 3.9.0). Degree Centrality (DC), Local Average Connectivity-based method (LAC), Closeness Centrality (CC), Eigenvector Centrality (EC), and Betweenness Centrality (BC) scores were detected by cytoNCA, and core network genes were screened.

Gene Ontology (GO) pathway enrichment analyses

The GO pathway enrichment analyses, encompassing GO Biological Process (BP), GO Cellular component (CC) and GO molecular function (MF), were performed by using Metascape [33]. The top 10 significant results were drawn using bioinformatics (www.bioinformatics.com.cn).

Molecular docking and protein - ligand Interaction

AutoDock was utilized to perform the molecular docking of DHA into its predicted target, PTGS1. The quantitative analysis of molecular docking results involved screening out the poses of multiple ligands with low binding scores. The enlisted molecular docking scores denoted the binding energies in kcal/mol. Analysis of the docked poses and ranking of compounds were calculated based on the estimated binding free energy. The lowest binding site of DHA and PTGS1 was visualized using PyMol software [34].

In vivo anti-tumor study

The experiments on animals were conducted following the guideline of the Wenzhou Medical University's Policy on the Care and Use of Laboratory Animals. Animal experimentation methods were authorized by the Wenzhou Medical University Animal Policy and Welfare Committee (approval no. wydw2020-0347). Female nude mice (Vital River Laboratories, Beijing, China), aged five weeks and weighing 15–17 g, were adaptively fed for 1 week. Subsequently, the mouse underwent subcutaneous injection with 4×10⁶ H460 cells in 100 µl of PBS into the right flank. Upon tumor volumes reaching approximately 100–150 mm³, we divided the mice into five group at random: vehicle, DHA, cisplatin, DHA + cisplatin, DHA + cisplatin + NAC (n = 5 per group). DHA (5 mg/kg), cisplatin (4 mg/kg), or their combination were administered by intraperitoneal (i.p.) injection once every alternate day. During the corresponding days, the group receiving the combination treatment was provided with 0.5 g/l NAC in drinking water. Tumor dimensions and body weight were assessed three times per week. The measurements included tumor length and width, and subsequently calculated the tumor volumes (V = 0.5×length×width²) at specified time intervals. At completion of the treatment, the nude mice were euthanized, and the tumors were extracted and weighed.

Immunohistochemistry

The tumor samples acquired from mice were preserved in 10 % formalin at ambient temperature, underwent a dehydration process, and embedded in paraffin. Afterwards, paraffin-embedded tissues were sliced to a thickness of 5 µm and positioned on slides with a positive charge. Indicated antibodies were applied to the tumor tissues and left overnight for incubation at 4  °C. Subsequently, secondary antibodies were employed to evaluate the immunoreactivity, followed by 3,3′-diaminobenzidine (DAB) for color detection. An ortho-microscope (Leica, Germany) was employed to obtain a quantitative assessment of immunochemistry data.

Statistical analysis

The experimental data were analyzed in triplicate (n = 3). Statistical analysis was carried out by the GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA, USA). The data was displayed as means ± standard deviation (SD), and the determination of statistical significance was conducted through the application of a one-way ANOVA multiple comparisons test. Statistically significant results were defined as p-values < 0.05.

Results

DHA inhibited cell growth, induced ROS generation, and subsequently activated the ER stress pathway in NSCLC cells

DHA has demonstrated potent anti-tumor activity in several cancer types, including lung, colorectal and ovarian cancers [35]. The DHA structure was illustrated (Fig. S1A). Firstly, we examined the impact of DHA treatment on NSCLC cells (A549 and H460 cells) viability and colony forming ability. DHA suppressed cell viability in a concentration-dependent manner with an IC₅₀ of 26.37 µM for A549 cells and 36.14 µM for H460 cells, respectively (Fig. 1A, B). Moreover, DHA exhibited a concentration-dependent inhibition of colony-forming ability in both A549 and H460 cells (Fig. 1C, D and S1B, C). Additionally, we investigated the impact of DHA on NSCLC cell apoptosis by investigating apoptosis-related proteins. Western blotting assay showed that DHA evidently decreased Bcl-2 and caspase 3 expressions, while increased cle-caspase 3 expression in a concentration-dependent way in both A549 and H460 cells (Fig. S1D, E), suggesting that DHA induced NSCLC cell apoptosis.

Fig. 1.

DHA exerted anti-NSCLC activity by stimulating ROS-mediated ER stress pathway. (A, B) A549 (0, 0.2, 0.5, 5, 10, 20, 30, 40, 50 µM) and H460 (0, 1, 10, 20, 40, 80, 100, 120, 150 µM) cells were treated with diverse concentrations of DHA for 48 h. MTT assay was carried out to evaluate cytotoxic effects of DHA on A549 and H460 cells. (C, D) The different dosages of DHA (0, 5, 10, 20, 30 and 40 µM for A549, and 0, 12.5, 25, 50 and 75 µM for H460) were used to treat NSCLC cells, and assessed the colony forming ability. The cells were fixed and stained with 1 % crystal violet and relative colony forming area was analyzed by Image J software. ***p < 0.001 and ****p < 0.0001, DHA treatment group vs. control group. (E, F) The relative ROS levels were measured by DCFH-DA fluorescent probe after treatment with 30 µΜ DHA (A549) for 1, 2, 4, 6, 8 h or 40 µΜ DHA (H460) for 0.5, 1, 2, 4, 6 h. ****p < 0.0001, DHA treatment group vs. control group. The ROS levels were assessed by using a FACS Calibur flow cytometer. (G, H) ATF4 and p-eIF2α protein expressions in A549 (30 µM) and H460 (40 µM) cells were detected by using western blotting assay after treatment with DHA at the indicated times. Data were expressed as means ± SD, and p-values were calculated by one-way ANOVA with Tukey's multiple comparisons test.

DHA facilitates ROS accumulation to trigger cell apoptosis [36,37]. Similarly, we observed an increase in ROS production in A549 and H460 cells following DHA treatment (Fig. 1E, F). Excessive intracellular ROS production leads to ER stress-mediated cell death [38]. To ascertain whether DHA stimulates ER stress, we assessed the p-eIF2α and ATF4 protein expression levels, known markers of the ER stress response [39]. DHA time-dependently enhanced the protein expressions of p-eIF2α and ATF4 in both A549 and H460 cells (Fig. 1G, H). Furthermore, NAC pretreatment partly but consistently attenuated inhibitory effects of DHA on cell viability (Fig. 2A) and colony formation (Fig. 2D and S2A). As expected, pretreatment of cells with NAC consistently reversed DHA induced p-eIF2α and ATF4 expressions (Fig. 2G), suggesting that DHA activated the ER stress response by promoting ROS accumulation, ultimately leading to cell apoptosis in NSCLC cells.

Fig. 2.

DHA induced cell apoptosis via ROS-mediated ER stress and MAPK signaling pathways activation. (A-C) The A549 and H460 cells were treated with 30 µM (A549) or 40 µM DHA (H460) with or without 5 mM NAC (A), 10 µM SP600125 (B) or 20 µM SB203580 (C). The images were captured by microscopy and cell viability was assessed by counting the cell numbers following trypan blue staining. ***p < 0.001, ****p < 0.0001, DMSO group vs. DHA treatment group; DHA treatment group vs. DHA treatment with NAC, SP600125 or SB203580 pretreatment. (D-F) A549 and H460 cells were pretreated with 5 mM NAC (D), 1.5 µM SP600125 (E) or 2.5 µM SB203580 (F) before treatment with 5 µM DHA. After 7-10 days incubation, colony forming ability was assessed. *p < 0.05, ***p < 0.001, ****p < 0.0001, DMSO group vs. DHA treatment group; DHA treatment group vs. DHA treatment with NAC, SP600125 or SB203580 pretreatment. (G-I) The 5 mM NAC (G), 10 µM SP600125 (H) or 20 µM SB203580 (I) were administered to the cells, and further treated with 30 µM (A549) or 40 µM DHA (H460). Western blot assay was used to detect the indicated protein expressions.

The JNK and p38 MAPK pathways activation involved in DHA-induced apoptosis

The mitogen-activated protein kinase (MAPK) pathway serves as a crucial regulator of cancer cell behaviors, including proliferation, migration, and apoptosis [40,41]. Specifically, p38 MAPK and JNK signaling pathways activation plays an essential role in inducing apoptosis in cancer cells under oxidative stress conditions [42,43]. Consequently, we examined the impact of JNK and p38 MAPK pathway activation on DHA-induced cell apoptosis. Treatment with the JNK inhibitor (SP600125) or p38 MAPK inhibitor (SB203580) markedly reversed the DHA-induced inhibition of cell viability (Fig. 2B, C) and colony formation (Fig. 2E, F and S2B, C). Furthermore, the increased expressions of p-p38 and p-JNK induced by DHA treatment were mitigated by SP600125 or SB203580, respectively (Fig. 2H, I). These findings suggest that the activation of JNK and p38 MAPK pathways plays a crucial role in mediating DHA-induced apoptosis in NSCLC cells.

Co-treatment with DHA and cisplatin displayed synergistic anti-NSCLC activities

We have found that DHA impeded cell growth by activating the ER stress pathway. Additionally, cisplatin has been reported to stimulate ER stress, leading to apoptosis in lung cancer cells [44]. Hence, we investigated whether the combined treatment of DHA and cisplatin exhibited stronger anti-tumor effects in NSCLC cells than individual treatments alone. The combined treatment with different concentrations of cisplatin, along with 20 µM of DHA, significantly suppressed the viability of NSCLC cells (Fig. 3A, B). Subsequently, we employed CompuSyn software to calculate the CI values between DHA and cisplatin. We found that 20 µΜ DHA and 30 µΜ cisplatin in A549 cells, or 20 µΜ cisplatin in H460 cells, yielded the lowest CI values compared to other combination groups (Fig. 3C, D), indicating stronger synergistic effects of these combinations than other combinations. To further confirm that the synergistic anti-cancer effect of combination treatment is attributed to ROS accumulation, the cells were exposed to NAC. The colony forming ability was greatly suppressed by the combined therapy with DHA and cisplatin, in comparison to individual treatments and this suppressive effect was attenuated by NAC pretreatment (Fig. 3E, F and S3A, B). Importantly, the co-treatment with low concentrations of DHA and cisplatin greatly inhibited the wound healing compared to DHA or cisplatin treatment alone, whilst these effects were mitigated by NAC pretreatment (Fig. S3C, D). The above experiments suggest that combined therapy with DHA and cisplatin exerts synergistic anti-NSCLC effects by increasing ROS generation.

Fig. 3.

DHA facilitated anti-tumor activity of cisplatin in NSCLC cells. (A, B) 20 µM DHA and different concentrations of cisplatin (10, 20, 30, 40 and 50 µM) were added into A549 and H460 cells. The cell numbers were counted by morphological experiments. The cell viability was assessed by counting the cell numbers following trypan blue staining. *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001, DHA treatment group vs. DMSO group. (C, D) CI values were calculated by using the CompuSyn software. CI < 1.0 was considered as synergistic effect. (E, F) A549 and H460 cells were treated with 5 µΜ DHA, 1.5 µΜ cisplatin or their combination of pretreatment with NAC. Image J software was used to quantified relative colony forming area. Results were represented as means ± SD and p-values were calculated by one-way ANOVA with Tukey's multiple comparisons test, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001, cisplatin or DHA treatment group vs. DMSO group; cisplatin or DHA treatment group vs. combination group; combination group vs. combination with NAC pretreatment group.

Combination of DHA and cisplatin markedly induced ROS-mediated ER stress in NSCLC cells

Emerging evidences indicated that increased ROS level is a pivotal characteristic in cancer cells. Thus, elevating ROS levels in cancer cells are an effective strategy for cancer therapy [45]. To further investigate whether DHA in combination with cisplatin significantly induces intracellular ROS generation, we assessed the ROS levels and ER stress-related proteins after combined treatment. As the combination of 20 µΜ DHA and 30 µΜ cisplatin in A549 cells, or 20 µΜ cisplatin in H460 cells, exhibited the lowest CI values, we selected these combinations for further analyses. As anticipated, ROS levels were markedly elevated following the combined treatment, in comparison to monotherapy in both A549 and H460 cells (Fig. 4A, B). Additionally, the combined treatments led to an increase in the p-eIF2α and ATF4 expressions in A549 and H460 cells compared to the DHA or cisplatin monotherapy, while NAC pretreatment partly mitigated these effects (Fig. 4C, D). These results suggest that the concurrent administration of DHA and cisplatin manifests a synergistic effect in inhibiting NSCLC activities, in part through the ROS-mediated ER stress signaling pathway.

Fig. 4.

DHA in combination with cisplatin synergistically induced ROS-mediated ER stress. (A, B) 20 µΜ DHA, 30 µΜ (A549) or 20 µΜ (H460) cisplatin, or their combination with 5 mM NAC were added into A549 and H460 cells, respectively, to detect relative ROS levels using fluorescent probe DCFH-DA by fluorescent microscopy (200×, 50 µm). (C, D) Western blotting assay was used to analyze the ATF4 and p-eIF2α protein expressions in A549 and H460 cells after combination treatment. Vinculin was used as a reference control.

Combination treatment with DHA and cisplatin synergistically stimulated JNK and p38 MAPK pathways in NSCLC cells

We further investigated whether the combined therapy with DHA and cisplatin increased JNK and p38 MAPK phosphorylation in NSCLC cells. Combined treatments with DHA (20 µΜ) and cisplatin (30 µΜ for A549 and 20 µΜ for H460) consistently enhanced p-JNK and p-p38 MAPK expressions in both A549 and H460 cells, while these enhancements were reduced when cells were pre-treated with NAC (Fig. 5A, B). These findings suggest that JNK and p38 MAPK pathways activation by ROS partly accounts for the anti-NSCLC effects elicited by the combined therapeutic approach.

Fig. 5.

DHA in combination with cisplatin strongly stimulated ROS-mediated p-JNK and p-p38 MAPK signaling pathway. (A, B) A549 (30 µΜ cisplatin) and H460 (20 µΜ cisplatin) cells were treated with 20 µΜ DHA, cisplatin or their combination. The p-JNK, JNK, p-p38 and p38 expressions were detected by western blotting analysis. For the combined treatment, the cells were pretreated with 5 mM NAC for 2 h.

DHA enhanced the anti-tumor properties of cisplatin in mice xenograft model

To further assess the effectiveness of combination treatment of DHA and cisplatin in vivo, we subcutaneously injected H460 cells to the mice to establish xenograft models. DHA (5 mg/kg) or cisplatin (4 mg/kg) monotherapy partly hampered the tumor growth when compared to the control group. Consistent with the in vitro findings, the combined treatment of DHA and cisplatin significantly suppressed tumor growth in vivo when compared with monotherapy alone, whilst NAC administration reversed these effects (Fig. 6A-C). Furthermore, immunohistochemical staining revealed that the ki67 positive cells were significantly decreased in combined treatment group, whilst NAC administration mitigated these effects (Fig. 6D and S4B). No significant alterations were observed from the histological examination of essential organs (heart, lung, liver and kidney) by using H&E staining (Fig. S4A), suggesting that the drug doses employed in our animal study were rational and well-tolerated. Moreover, the combination treatment group consistently upregulated p-eIF2α, ATF4, p-JNK and p-p38 expression levels in mice tumor tissues compared to those in single treatment groups. However, these upregulations were strongly blocked by NAC pretreatment (Fig. 6E). Collectively, these results indicate that DHA and cisplatin combination consistently enhanced anti-NSCLC activities in vivo through promoting ROS generation, and subsequent ER stress, JNK and p38 MAPK pathways activation.

Fig. 6.

DHA potentiated anti-tumor activity of cisplatin in xenograft mice model. (A) The mice harboring tumors were treated with 5 mg/kg DHA, 4 mg/kg cisplatin, or their combination whose water were added 0.5 g/l NAC. By one month after injection, the mice were sacrificed, solid tumor were weighed and photographed. (B) Tumor volumes were measured every two days. (C) Tumor weights were measured from different groups. ***p < 0.001 and ****p < 0.0001, cisplatin or DHA treatment group vs. control group; cisplatin or DHA treatment group vs. combination group; combination group vs. combination with NAC pretreatment group. (D) Ki67 positive cells were counted from mice tumor tissues treated with DHA, cisplatin or their combination after immunohistochemical staining. ***p < 0.001, cisplatin or DHA treatment group vs. control group; cisplatin or DHA treatment group vs. combination group; combination group vs. combination with NAC pretreatment group. (E) Western blotting assays were performed to evaluate the p-eIF2α, ATF4, p-JNK and p-p38 protein expression levels in tumor tissues after treatment. GAPDH was used as the loading control. The data are represented as mean ± SD and analyzed by one-way ANOVA with Tukey's multiple comparisons test.

Identification of the DHA targets in NSCLC

To identify the potential targets of DHA in NSCLC cells, we performed bioinformatics analyses. Firstly, we obtained the information about DHA targets and NSCLC targets from Super-PRED and GeneCards, respectively. A Venn diagram demonstrated that there was an intersection of 114 genes, which were derived from a pool of 161 DHA targets and 5872 NSCLC targets (Fig. S5A). Using the STRING online tool and Cytoscape software, we were able to visualize these common targets. According to analysis of the protein-protein interaction (PPI) network, PTGS1 emerged as the highest-ranking gene among the top 10 hub genes (Fig. S5B). Next, we analyzed the significantly enriched terms using the Gene Ontology (GO) database through Metascape. GO enrichment analyses were consisted of three gene function terms including BP, CC and MF [46]. BP and MF analyses indicated that the terms were closely associated with the oxidative stress, including cellular response to reactive oxygen species, oxidative stress, reactive oxygen species and oxidoreductase activity, respectively (Fig. S6A). Furthermore, the molecular docking result indicated that DHA and PTGS1 had the lowest binding energy of −8.64 kcal/mol, suggesting that best-docked pose of DHA and PTGS1 (Fig. S6B). DHA established a hydrogen bond with GLU-465, a crucial amino acid residue, and two hydrogen bonds with GLN-461 in PTGS1 activated pocket, suggesting that potential direct binding between DHA and PTGS1.

DHA inhibited NSCLC cell growth by targeting PTGS1

PTGS1, which is considered as an antioxidant gene, has been reported to be involved in oxidative stress [47]. PTGS1 was constitutively expressed in various tumor tissues, including human testis, ovarian and lung cancers [48], [49], [50], [51]. Our results showed that DHA induced ROS accumulation and PTGS1 is a possible target of DHA in NSCLC cells. Therefore, we speculate that DHA may impact on the function of PTGS1 to promote ROS generation. As expected, DHA treatment significantly decreased the PTGS1 mRNA and protein expressions (Fig. 7A, C). To confirm the role of PTGS1 in DHA-induced ROS generation, siRNA technique was utilized to silence PTGS1 expression in A549 cells. Transfection of siRNAs (siPTGS1-1 and siPTGS1-2) to A549 cells significantly down-regulated both PTGS1 mRNA and protein levels, and the stronger silencing effect was observed in siPTGS1 than siPTGS2 (Fig. 7B, C). Knocking down PTGS1 partly inhibited NSCLC cell proliferative and colony forming abilities, and potentiated the suppressive effects of DHA on cell viability (Fig. 7D, E and S7A, B) and the colony formation ((Fig. 7F, G and S7D, E) in A549 cells. Similarly, silencing PTGS1 intensified the suppressive effects of cisplatin on cell viability (Fig. 7H) and cisplatin-mediated ROS accumulation in NSCLC cells (Fig. 7I). Furthermore, knocking down PTGS1 enhanced DHA-mediated ROS accumulation (Fig. 8A). Additionally, knocking down PTGS1 enhanced DHA-induced ATF4, p-eIF2α, p-JNK and p-p38 expressions in A459 cells. Of note, co-treatment of A549 cells with DHA and siPTGS1 markedly inhibited PTGS1 expression when compared to single treatment alone (Fig. 8B, C). Our results indicate that PTGS1 is a critical mediator in DHA-induced ER stress, JNK and p38 MAPK pathways activation through regulating ROS generation.

Fig. 7.

DHA suppressed NSCLC cell growth by inhibiting PTGS1. (A) qRT-PCR was used to assess the PTGS1 mRNA expression levels after treatment of cells with 20 µM DHA. GAPDH was used as housekeeping gene to normalize mRNA expression. **p < 0.01, ***p < 0.001 and ****p < 0.0001, DHA treatment group vs. control group. (B) Small interfering RNA was used to silence PTGS1 expression in A549 cells. The PTGS1 mRNA expression was detected by qRT-PCR. **p < 0.01 and ***p < 0.001, siPTGS1 group vs. NC group. (C) The PTGS1 protein expression in A549 cells was detected by western blot analysis after 20 µM DHA treatment (upper panel) or transfection with two independent siRNAs against PTGS1 (lower panel). (D, E) The A549 cells were treated with 20 µM DHA, siPTGS1 or their combination for 24 h. The cell viability was analyzed by morphological experiments. The cell viability was assessed by counting the cell numbers following trypan blue staining. Values were presented as means ± SD, **p < 0.01 and ***p < 0.001, DHA treatment group or siPTGS1 group vs. NC group or combination group, respectively. (F, G) Colony formation assay was used to investigate colony forming ability after treatment of cells with 5 µM DHA, siPTGS1 or their combination. The colony forming area was measured by Image J software. **p < 0.01 and ***p < 0.001, DHA treatment group or siPTGS1 group vs. NC group or combination group, respectively. (H) 20 µM cisplatin was administered into PTGS1 knockdown A549 cells. The cell viability was assessed by counting the cell numbers following trypan blue staining. ***p < 0.001, cisplatin treatment group or siPTGS1 group vs. NC group or combination group, respectively. The one-way ANOVA with Tukey's multiple comparisons test was used to calculate the p-values. (I) A549 cells were treated with 20 µM cisplatin, siPTGS1 or combination thereof. Fluorescence microscope was used to detect the cellular ROS levels (200×, 50 µm).

Fig. 8.

Knocking down PTGS1 potentiated DHA-induced ER stress, JNK and p38 MAPK signaling pathways. (A) PTGS1 knockdown or control cells were treated with or without 20 µM DHA for 6 h, respectively, and intracellular ROS levels were assessed by fluorescence microscope (200×, 50 µm). (B, C) siPTGS1 or control siRNA were transfected into A549 cells. After 24 h, the cells were treated with or without 20 µM DHA. The protein expressions of PTGS1, p-eIF2α and ATF4, p-JNK, p-p38, JNK, and p38 were detected by western blotting analysis. GAPDH was served as internal control. (D) Summarized schematic figure of this study. DHA inhibited PTGS1 and enhanced anti-cancer effects of cisplatin by stimulating ER stress, JNK and p38 MAPK signaling pathways.

Discussion

NSCLC demonstrated high mortality and incidence rates worldwide. Despite significant advancements in surgery, immunotherapy, and targeted therapy, chemotherapy remains one of the primary approaches for the treatment of NSCLC [52]. Nevertheless, severe side effects and the development of drug resistance have compromised the clinical efficacy of chemotherapeutic agents [53]. Hence, there is an urgent need for novel treatment approaches to magnify the therapeutic effectiveness of chemotherapeutic drugs.

Our current study demonstrated that ER stress, JNK and p38 MAPK signaling pathways activation by ROS accumulation contributed to the synergistic anti-NSCLC activity of combination treatment with DHA and cisplatin both in vitro and in vivo. Additionally, our study indicated that PTGS1 is a potential and novel target of DHA. Silencing PTGS1 expression potentiated anti-NSCLC activity of DHA by triggering ROS-mediated ER stress, and activating JNK and p38 MAPK signaling pathways (Fig. 8D).

The unfolded or misfolded proteins accumulation in the ER initiates unfolded protein response (UPR). The PERK (protein kinase RNA (PKR)-like ER kinase) -eIF2α, IRE1α (inositol-requiring protein-1), and ATF6 pathways are most commonly described UPR pathway in cancer [54]. Among these pathways, the eIF2α-ATF4 pathway activation is of great importance in inducing apoptosis triggered by ER stress in cancer cells, including those observed in colon cancer and brain glioma [55], [56], [57], [58].

During ER stress conditions, the JNK and p38 MAPK pathways were promptly activated, and contributed to cancer cell apoptosis [59,60]. DHA exhibited robust anti-tumor activity in cisplatin-resistant oral squamous cell carcinoma by regulating the ROS-MAPK signaling pathway [61]. In the pancreatic ductal adenocarcinoma, DHA enhanced the anti-tumor properties of cisplatin by stimulating ferroptosis [62]. A few studies have reported that effectiveness of combined treatment of DHA and cisplatin in lung cancer. For example, the combination of DHA and cisplatin strongly suppressed lung cancer cell growth through down-regulating vascular endothelial growth factor receptor (VEGFR) expression [63]. In addition, DHA increased the anti-tumor effects of cisplatin by inhibiting vascularization-associated proteins, HIF-1α and VEGF expressions [64], indicating that DHA in combination with cisplatin exerted anti-lung cancer activity by suppressing angiogenesis. Nevertheless, it remains to be explored whether ER stress, JNK and p38 MAPK signaling pathways activation through ROS accumulation contributed to anti-NSCLC effects of combination therapy with DHA and cisplatin. We demonstrated that the combination therapy exhibited synergistic anti-NSCLC effects by activating ER stress, JNK and p38 MAPK signaling pathways through ROS accumulation, both in vitro and in vivo.

The PTGS1 function in cancer progression is controversial. It has been reported that the activation of COX-1 by tobacco carcinogen induced ROS accumulation and DNA damage in lung cancer [65]. Conversely, PTGS1 has been shown to act as an anti-oxidative gene, preventing cells from oxidative stress by blocking ROS accumulation [47]. Oleocanthal (OC), a COX-1 inhibitor, induced ROS production to promote liver cancer cell apoptosis [66], suggesting that critical function of PTGS1 in ROS-mediated cell death. PTGS1 is found to be overexpressed in lung and colorectal cancers, indicating its potential oncogenic characteristics in these malignancies [67]. Additionally, PTGS1 promoted tumorigenesis of cervical cancer by elevating inflammatory and angiogenic factors [68]. PTGS1 also facilitated ovarian cancer progression by promoting neovascularization [69]. Nevertheless, no studies have been reported regarding the function of PTGS1 in DHA-mediated anti-NSCLC activities. In this study, we made a novel discovery, revealing for the first time that DHA treatment significantly suppressed PTGS1 expression in NSCLC cells. Additionally, the silencing of PTGS1 potentiated anti-NSCLC efficacy of DHA through stimulating ROS production, and subsequent ER stress and MAPK signaling pathways activation. Our study provides clear evidence that PTGS1 is a crucial mediator in the DHA-mediated anti-NSCLC activity, and PTGS1 has the potential to be an effective therapeutic target for patients with NSCLC. Although the mechanistic basis for DHA-mediated decrease in PTGS1 expression is currently unknown, we hypothesize that DHA inhibits PTGS1 expression by directly binding to it, thereby potentially affecting PTGS1 mRNA stability and/or its enzymatic activity. Further studies are required to validate this hypothesis.

Conclusions

DHA enhanced the anti-NSCLC effects of cisplatin by targeting PTGS1-ROS-mediated ER stress, JNK, and p38 MAPK signaling pathways. The current study provided a robust experimental foundation for combined treatment of DHA and cisplatin in NSCLC and suggested that targeting the PTGS1-ROS-mediated signaling pathway could be an effective therapeutic strategy for certain NSCLC patients.

Data availability statement

The data presented in this study are available on reasonable request from the corresponding author.

Funding

This work was supported by National Natural Science Foundation of China (81672305), Health Commission of Zhejiang Province (2022RC292) and Natural Science Foundation of Zhejiang Province (LZ22H160006).

Ethics approval and consent to participate

The animal study was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) guidelines, Wenzhou Medical University.

CRediT authorship contribution statement

Lianli Ni: Writing – original draft, Methodology, Investigation, Formal analysis. Xinping Zhu: Methodology, Investigation, Formal analysis. Qi Zhao: Writing – original draft, Investigation, Formal analysis. Yiwei Shen: Formal analysis. Lu Tao: Formal analysis. Ji Zhang: Formal analysis. Han Lin: Formal analysis. Weishan Zhuge: Formal analysis. Young-Chang Cho: Supervision, Writing – review & editing. Ri Cui: Writing – review & editing, Writing – original draft, Supervision, Funding acquisition, Conceptualization. Wangyu Zhu: Writing – review & editing, Writing – original draft, Supervision, Conceptualization.

Declaration of competing interest

The authors declare no conflict of interest regarding this manuscript.