Cigarette smoke promotes the progression of non-small cell lung cancer by activating ERK1/2-FOXC1 axis to induce epithelial-mesenchymal transition.

Introduction
Lung cancer is the leading cause of cancer-related deaths and the second most prevalent cancer worldwide. Approximately 1.8 million deaths and 2 million new cases of lung cancer occur annually [1]. Lung cancer is primarily classified into two types: small cell lung cancer and non-small cell lung cancer (NSCLC). NSCLC accounts for > 85% of all lung cancer cases [2]. NSCLC primarily comprises lung adenocarcinoma (LUAD) and lung squamous cell carcinoma [3]. Despite significant advancements in the treatment of NSCLC in recent years, the 5-year survival rate of patients with NSCLC remains below 15% [4, 5]. Cigarette smoke contains various carcinogens, including tobacco-specific nitrosamines, polycyclic aromatic hydrocarbons, and volatile organic compounds, and is a major risk factor for lung cancer [6]. The risk of developing lung cancer in long-term smokers has been indicated to be 20 times greater than that in non-smokers [7]. Approximately 85% of lung cancer cases are attributable to smoking [5, 8]. Retrospective analyses have demonstrated that patients who continue smoking after being diagnosed with NSCLC have poor clinical outcomes. Specifically, compared with patients with NSCLC who do not smoke, those who persist in smoking after diagnosis exhibit a high mortality rate, increased pain levels, and short overall survival [9, 10]. However, the mechanism by which smoking promotes NSCLC progression remains unclear.
Forkhead box protein C1 (FOXC1), located on human chromosome 6p25, is a member of the winged helix-forkhead transcription factor family. It shares structural similarities with other members of this family, all of which possess a conserved forkhead domain that specifically recognizes and binds to target DNA sequences [11, 12]. FOXC1 plays a critical role in sustaining normal epithelial-mesenchymal transition (EMT) throughout development. A deficiency in FOXC1 is linked to Dandy–Walker syndrome, characterized by cerebellar developmental malformation [13], and Axenfeld–Rieger syndrome, which is primarily associated with anterior segment developmental malformations of the eye [14]. Moreover, research has demonstrated that the overexpression of FOXC1 is linked to the progression of several cancers, including pancreatic [15], colorectal [16], liver [17], and breast [18] cancers. Upregulation of FOXC1 expression has been observed in the lung tissues of rats exposed to cigarette smoke [19, 20]. However, it remains uncertain whether smoking facilitates the progression of NSCLC through the upregulation of FOXC1 expression.
Extracellular signal-regulated kinases 1 and 2 (ERK1/2) are components of the mitogen-activated protein kinase signaling pathway. Studies have shown that activation of the ERK1/2 signaling pathway enhances the growth and drug resistance of colorectal cancer [16] and facilitates the proliferation and metastasis of hepatocellular carcinoma by upregulating FOXC1 expression [17]. However, it remains unclear whether the ERK1/2 signaling pathway is associated with FOXC1 expression in NSCLC. Therefore, we used in vivo and in vitro models of smoking-associated NSCLC to investigate the mechanism underlying NSCLC progression.
This study revealed that FOXC1 expression was significantly upregulated in both NSCLC cell lines and tissues. Moreover, cigarette smoke extract (CSE) markedly enhanced FOXC1 expression in NSCLC cells. Subsequent research indicated that CSE-induced FOXC1 expression was correlated with increased proliferation, migration, and invasion capabilities of NSCLC cells. Additional investigations identified the ERK1/2 signaling pathway as an upstream regulator of FOXC1 expression. These findings suggest that the ERK1/2-FOXC1 signaling axis is closely linked to the progression of smoking-related NSCLC, potentially providing new targets for clinical diagnosis and treatment.

Materials and methods

CSE Preparation
A CSE solution was prepared using commercially available Diamond-brand cigarettes. The CSE preparation method was consistent with previously established protocols [21]. Briefly, smoke produced by the combustion of a single Diamond-brand cigarette was bubbled and dissolved in 10 mL of RPMI-1640 medium (pre-warmed to 37 °C) for 5 min using a vacuum pump. Subsequently, a 0.22 μm filtration device was used to remove impurities and microorganisms, and the standardization control was established by measuring absorbance at 320 nm and 540 nm. The resulting solution was designated as 100% CSE and stored at -80 °C.

Cell culture and treatment
Human bronchial epithelial (HBE) and NSCLC (A549 and H1299) cells were obtained from the American Type Culture Collection. All cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. The cells were cultured in an incubator at 37 ℃ with 5% CO₂.
A549 and H1299 cells were treated with different concentrations of CSE for 48 h to establish a cellular model of cigarette smoke exposure. Before CSE treatment, the cells were pre-treated with U0126 (10 nM) for 2 h to inhibit the activation of the ERK1/2 signaling pathway, or the expression of FOXC1 was knocked down by transfecting with FOXC1-siRNA. After treatment, the samples were collected, and functional experiments were performed.

SiRNA transfection of NSCLC cells
According to the manufacturer’s instructions, A549 and H1299 cells were seeded into 6-well plates and cultured overnight. The si-RNA and Lipofectamine 3000 liposome complexes were incubated in serum-free Opti-MEM and added to the cell culture medium for transfection. After 6 h, the medium was replaced with complete medium with or without 2.5% CSE. The sequences of the si-RNAs were as follows:
si-FOXC1#1: 5′-GUCACAGAGGAUCGGCUUGAA-3′,
si-FOXC1#2: 5′-GCGGCGAGCAGAGCUACUA-3′,
si-FOXC1#3: 5′-GAGCUUUCGUCUACGACUGUA-3′,
si-ctrl: 5′-UUCUCCGAACGUGUCACGU-3′.

Cell counting kit-8 assay
Cell cytotoxicity assay: A549 and H1299 cells were cultured overnight in 96-well plates. After that, the cells were treated with different concentrations of CSE (0%, 1%, 2.5%, 5%, 7.5%, and 10%) for 48 h. Then, 10 µL of cell counting kit-8 (BMC, #B6005M) solution was added to each well, and the cells were incubated for 2 h. Subsequently, absorbance was measured at 450 nm.
Cell proliferation assay: The treated A549 and H1299 cells were seeded into 96-well plates. At 0 h, 24 h, 48 h, and 72 h, 10 µL of CCK-8 solution was added to each well, and the cells were incubated for 2 h. Subsequently, absorbance was measured at 450 nm.

5-Ethynyl-2′-deoxyuridine assay
The treated A549 or H1299 cells were seeded into 24-well plates and cultured overnight. The medium was replaced with complete medium containing 10 nM 5-Ethynyl-2′-deoxyuridine (Beyotime, #C0071S), and the cells were further incubated for 2 h. After washing with phosphate-buffered saline to remove dead cells, the cells were fixed and permeabilized. The Click reaction solution was prepared according to the manufacturer’s instructions; 200 µL of the Click reaction solution was added to each well, and the cells were incubated at room temperature for 30 min. DNA was stained with Hoechst 33,342 for 10 min. Finally, the cells were observed under a fluorescence microscope.

Colony formation assay
The treated A549 or H1299 cells were seeded into 6-well plates at a density of 600 cells/well. After 14 days, the cells were fixed with 4% paraformaldehyde. The cells were stained with a crystal violet solution at room temperature for 20 min. After washing and air-drying, cell colonies were counted under an optical microscope, and colonies containing > 50 cells were recorded.

Wound healing assay
The treated A549 or H1299 cells were seeded into 6-well plates at a density of 5 × 10⁵ cells/well and incubated overnight. A uniform scratch was made on the cell surface using a 200 µL pipette tip. The medium was then replaced with a medium containing 1% fetal bovine serum. The scratches were observed and photographed under an optical microscope at 0 and 24 h.

Transwell assay
Migration and invasion assays were performed using Transwell chambers with 8 μm pore filters (Corning, USA). The treated A549 cells (5 × 10⁴ cells/well) and H1299 cells (4 × 10⁴ cells/well) were suspended in 100 µL of serum-free medium and seeded onto Transwell chambers coated with or without Matrigel (Corning, USA). Then, 600 µL of complete medium was added to the lower chambers. After 24 h of incubation, cotton swabs were used to remove the Matrigel and collect the cells inside the chamber. After fixation with 4% paraformaldehyde and staining with a crystal violet solution for 30 min, the cells were visualized and imaged under a microscope. To count the cells, five fields were randomly selected, and the results were averaged.

Western blotting
Total protein was extracted from cells and tissues using RIPA lysis buffer. Proteins were separated using sodium dodecyl-sulfate polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. After blocking the membrane with 5% non-fat milk powder at room temperature for 1 h, the membranes were incubated overnight at 4 °C with the appropriate antibodies. The following antibodies were used: β-actin (1:50000, ABclonal, #AC026), FOXC1 (1:600, Abcam, #AB227977), CDK2 (1:1000, Proteintech, #10122-1-AP), PCNA (1:1000, Proteintech, #10205-2-AP), Cyclin E1 (1:1000, Proteintech, #11554-1-AP), Snail (1:1000, ABclonal, #A5243), Slug (1:1000, Abmart, #PA3014), Vimentin (1:1000, CST, #5741), MMP2 (1:1000, Abmart, #T57164), MMP9 (1:1000, Proteintech, #10375-2-AP), E-cadherin (1:1000, CST, 3195), N-cadherin (1:1000, CST, #13116), p-ERK1/2 (1:1000, Proteintech, #28733-1-AP), ERK1/2 (1:1000, ABclonal, #A4782), p-c-Jun (1:1000, Proteintech, #28907-1-AP), and p-c-Fos (1:1000, Abmart, #PC12911). Nitrocellulose membranes were incubated with the appropriate secondary antibodies for 1 h, followed by development using an enhanced chemiluminescence working solution (SparkJade, ED0015). Grayscale analysis was performed using the ImageJ software.

Immunofluorescence staining
Cell immunofluorescence: A549 or H1299 cells were seeded into 24-well plates and incubated overnight. The cells were incubated with 0% or 2.5% CSE for 48 h. Following fixation, permeabilization, and blocking, the cells were incubated with a FOXC1 primary antibody (1:50, Abcam, #AB227977) at 4 °C overnight, followed by incubation with a fluorescently -labeled secondary antibody at room temperature for 1 h. The nuclei were counterstained with DAPI. FOXC1 expression was visualized and imaged under a fluorescence microscope.
Tissue immunofluorescence: Xenograft tumor tissues were fixed, paraffin-embedded, and sectioned. After deparaffinization, rehydration, and antigen retrieval, the sections were incubated with a Ki-67 primary antibody (1:100; Servicebio, #GB111499) at 4 °C overnight. This was followed by incubation with a fluorescent secondary antibody at room temperature for 1 h. The nuclei were counterstained with DAPI. Ki-67 expression was examined under a fluorescence microscope.

Construction of stable A549 cells
Control shRNA and shRNA lentiviral vectors targeting FOXC1 were used to generate A549 cells with stable FOXC1 knockdown. Stably transfected cells were selected using 2 µg/mL puromycin, and the knockdown of FOXC1 was verified using western blotting. Subsequently, A549 cells stably transfected with control sh-RNA and sh-FOXC1 were cultured in media containing 0% or 2.5% CSE for 30 days. The cells were harvested for subcutaneous transplantation into nude mice. The sequences of sh-FOXC1 and control shRNA were as follows:
sh-FOXC1: 5′-GAGCUUUCGUCUACGACUGUA-3′,
sh-ctrl: 5′-UUCUCCGAACGUGUCACGU-3′.

Animal experiment
BALB/c nude mice (3–4 weeks, males) were housed under specific pathogen-free conditions with free access to food and water [21]. Twenty mice were randomly divided into four groups. The treated A549 cells (sh-ctrl, sh-FOXC1, CSE + sh-ctrl, and CSE + sh-FOXC1) (5 × 10⁶ cells) were suspended in 100 µL of sterile phosphate-buffered saline and then injected into the right axilla of the mice. The tumor size was measured weekly using a Vernier caliper (tumor volume = 0.5×length×width²). The mice were sacrificed after 5 weeks, and the tumors were harvested for analysis. All mouse experiments were approved by the Animal Welfare and Ethics Committee of Tianjin Medical University.

Cell transcriptome sequencing
Total RNA was isolated using the Trizol Reagent (Invitrogen Life Technologies), and the concentration, quality, and integrity were determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific). Three micrograms of RNA was used as the input material for RNA sample preparation. Sequencing libraries were generated according to the following steps: First, mRNA was purified from the total RNA using poly(T) oligo-attached magnetic beads. Fragmentation was performed using divalent cations at elevated temperatures in the Illumina proprietary fragmentation buffer. First-strand cDNA was synthesized using random oligonucleotides and SuperScript II. Second-strand cDNA synthesis was subsequently performed using DNA polymerase I and RNase H. The remaining overhangs were converted into blunt ends via exonuclease/polymerase activities, and the enzymes were removed. After adenylation of the 3′ ends of the DNA fragments, Illumina PE adapter oligonucleotides were ligated to prepare them for hybridization. To select cDNA fragments of the preferred 400–500 bp in length, library fragments were purified using the AMPure XP system (Beckman Coulter, Beverly, CA, USA). DNA fragments with ligated adaptor molecules at both ends were selectively enriched using Illumina PCR Primer Cocktail in a 15-cycle polymerase chain reaction. The products were purified (AMPure XP system) and quantified using an Agilent high-sensitivity DNA assay on a Bioanalyzer 2100 system (Agilent Technologies). The library was sequenced using a NovaSeq Xplus platform (Illumina).

Statistical analysis
All experiments were performed in triplicates. Statistical analyse were performed using the statistical tool GraphPad Prism 10.0 (GraphPad Software Inc, San Diego, CA, USA). Student’s t-test and ANOVA were used to detect the differences between two groups and among multiple groups, respectively. A P-value < 0.05 was considered statistically significant.

Result

CSE promoted proliferation of NSCLC cells
To determine the appropriate concentration of CSE, A549 and H1299 cells were treated with various concentrations of CSE (0%, 1%, 2.5%, 5%, 7.5%, and 10%) for 48 h. The results indicated that the viability of A549 and H1299 cells increased at low concentrations of CSE (1% and 2.5%) but decreased when the concentration exceeded 2.5% (Fig. 1A).

Subsequently, the A549 and H1299 cells were treated with 1% and 2.5% CSE to generate cell growth curves. The results indicated that CSE significantly enhanced the proliferation of A549 and H1299 cells in a concentration-dependent manner (Fig. 1B). Western blot results indicated that CSE enhanced the expression of PCNA, CDK2, and Cyclin E1 in A549 and H1299 cells in a concentration-dependent manner (Fig. 1C). To achieve optimal experimental outcomes, 2.5% CSE was used for subsequent experiments. Additionally, CSE significantly enhanced colony formation (Fig. 1D) and the proportion of EdU-positive cells (Fig. 1E) in the A549 and H1299 cell lines.

CSE promoted EMT, migration, and invasion of NSCLC cells
Enhanced migratory and invasive capabilities of cancer cells are critical drivers of cancer progression. EMT is a key process by which tumor cells acquire migratory and invasive properties.
We investigated the effect of CSE on EMT of NSCLC cells. Western blot analysis revealed that CSE increased the expression of Snail, Slug, Vimentin, MMP2, and MMP9, but decreased the expression of E-cadherin in A549 cells (Fig. 2A). In H1299 cells, CSE increased the expression of Snail, Slug, Vimentin, N-cadherin, MMP2, and MMP9 (Fig. 2B). CSE induced the appearance of spindle shaped and long spindle shaped cells in A549 and H1299 cell lines (Fig. 2C). These results indicate that CSE induces EMT of NSCLC cells. Subsequently, we performed wound healing and Transwell assays to evaluate the effects of CSE on the migratory and invasive abilities of A549 and H1299 cells. The results showed that CSE promoted the wound healing rates of A549 and H1299 cells (Fig. 2D) and increased the number of cells passing through the chamber membrane in Transwell migration and invasion assays (Fig. 2E).

NSCLC cells highly expressed FOXC1
To investigate the biological role of FOXC1 in NSCLC, we evaluated the expression of FOXC1 in NSCLC cell lines (A549 and H1299) and a normal human airway epithelial cell line (HBE). Western blot results showed that the expression of FOXC1 was significantly higher in A549 and H1299 cells than in HBE (Fig. 3A). High FOXC1 mRNA expression in NSCLC tissues was confirmed using the Gene Expression Profiling Interactive Analysis database (Fig. 3B) [22].

CSE promoted the expression of FOXC1 in NSCLC cells
To investigate whether FOXC1 is involved in the smoking-induced promotion of NSCLC progression, we analyzed the effect of CSE on FOXC1 expression in NSCLC cells. Western blot (Fig. 4A) and immunofluorescence staining (Fig. 4B) demonstrated that CSE enhanced the expression of FOXC1 in A549 and H1299 cells. Transcriptome sequencing revealed that CSE promoted the transcription of FOXC1 in A549 cells (Fig. 4C and D).

Knockdown of FOXC1 reversed the promoting effect of CSE on the proliferation of NSCLC cells
To investigate the role of FOXC1 in CSE-induced proliferation of NSCLC cells, we performed a targeted knockdown of FOXC1 using siRNA. The knockdown efficiency was examined using western blot analysis. The results revealed that among the three siRNAs designed, si-FOXC1#3 exhibited the optimal knockdown effect (Fig. 5A). Therefore, si-FOXC1#3 was selected for subsequent experiments. Western blot results demonstrated that the expression of FOXC1 was also downregulated by si-FOXC1 in A549 and H1299 cells treated with CSE (Fig. 5B).

Subsequently, we evaluated the effect of FOXC1 knockdown on the CSE-induced proliferation of NSCLC cells. The results showed that FOXC1 knockdown inhibited the proliferation rate (Fig. 5C), proportion of EdU-positive cells (Fig. 5D), and colony formation (Fig. 5E) of A549 and H1299 cells induced by CSE. Western blot analysis showed that FOXC1 knockdown reduced the expression of PCNA, CDK2, and Cyclin-E1 in CSE induced A549 and H1299 cells (Fig. 5F).

Knockdown of FOXC1 reversed the promoting effect of CSE on the growth of xenograft tumors
To determine whether CSE and FOXC1 affected the in vivo growth of NSCLC cells, A549 cells with stable FOXC1 knockdown (Fig. 6A) and CSE exposure were established. Cells were transfected with sh-FOXC1 and continuously cultured in 2.5% CSE for 30 days. Subsequently, tumor xenografts were generated (Fig. 6B).

Western blot assay was used to analyze the expression of FOXC1 in xenograft tumor tissues. The results showed that the expression of FOXC1 was significantly increased in xenograft tumors formed after CSE exposure. Meanwhile, the expression of FOXC1 in xenograft tumors formed after CSE exposure was inhibited by sh-FOXC1 (Fig. 6C). The results of the xenograft tumor experiment showed that, tumor growth in the CSE-treated group was faster (Fig. 6D), the volume of the xenograft tumors was larger (Fig. 6E), and the expression of Ki- 67 was higher than that in the control group (Fig. 6F). However, these effects were reversed by FOXC1 knockdown. Similarly, pretreatment with CSE significantly increased the expression of PCNA, CDK2, and Cyclin-E1 in xenograft tumor tissues. FOXC1 knockdown reversed the increase in the expression of these proteins (Fig. 6G).

Knockdown of FOXC1 reversed the promoting effects of CSE on the migration, invasion, and EMT of NSCLC cells
The results of the wound healing assay (Fig. 7A) and Transwell assay (Fig. 7B) showed that the knockdown of FOXC1 significantly inhibited the migration and invasion abilities of A549 and H1299 cells that were induced by CSE. Western blot analysis showed that CSE-induced changes in EMT marker protein levels in A549 and H1299 cells were reversed by FOXC1 knockdown (Fig. 7C). Similarly, CSE induced EMT of xenograft tumor tissues, as evidenced by the upregulation of Snail, Slug, Vimentin, MMP2, and MMP9 expression and the downregulation of E-cadherin expression. These alterations were reversed by FOXC1 knockdown (Fig. 7D).

CSE activated the ERK1/2 signaling pathway
To further explore the mechanism by which CSE promotes NSCLC progression by upregulating FOXC1 expression, we analyzed the ERK1/2 signaling pathway, which is closely associated with cancer. Western blot analysis (Fig. 8) showed that CSE significantly upregulated the expression of p-ERK1/2 in A549 and H1299 cells but had no significant effect on the expression of ERK1/2. CSE also increased the expression of p-AP-1 (p-c-Jun and p-c-Fos), which are downstream molecules of the ERK1/2 signaling pathway. These results indicate that CSE activates the ERK1/2 pathway in NSCLC cells.

The ERK1/2 inhibitor U0126 reversed the CSE induced increase of FOXC1 in NSCLC cells
To further investigate the relationship between the activation of the ERK1/2 pathway and FOXC1 in the promotion of malignant biological behaviors of NSCLC cells by CSE, we performed rescue experiments using the specific ERK1/2 pathway inhibitor U0126 (10 nM). Western blot analysis revealed that the CSE-induced upregulation of FOXC1 expression in A549 and H1299 cells was reversed by U0126. These findings implied that CSE promoted FOXC1 expression in NSCLC cells by activating the ERK1/2 pathway (Fig. 9).

The ERK1/2 inhibitor U0126 reversed the promoting effect of CSE on the proliferation of NSCLC cells
The role of the ERK1/2 pathway in CSE-promoted proliferation of NSCLC cells was investigated using rescue experiments. The results of the growth curve (Fig. 10A), EdU proliferation assay (Fig. 10B), and colony formation assay (Fig. 10C) demonstrated that U0126 reversed the promoting effects of CSE on the proliferation of A549 and H1299 cells. U0126 also decreased the CSE-induced expression of PCNA, CDK2, and Cyclin E1 in A549 and H1299 cells (Fig. 10D).

The ERK1/2 inhibitor U0126 reversed the promoting effects of CSE on the migration, invasion, and EMT of NSCLC cells
The role of the ERK1/2 pathway in CSE-promoted migration and invasion of NSCLC cells was investigated using rescue experiments. The results showed that U0126 reversed the promoting effects of CSE on the migration and invasion of A549 and H1299 cells in the wound healing assay (Fig. 11A) and Transwell assay (Fig. 11B). Western blot analysis of EMT marker proteins revealed that the CSE-induced changes in EMT marker protein levels in A549 and H1299 cells were reversed by U0126 (Fig. 11C).

Discussion
Cigarette smoke generated by burning cigarettes contains a minimum of 93 known carcinogens [6]. When carcinogenic substances enter the human body, they can directly damage the DNA of cells, resulting in gene mutations and abnormal cell proliferation, which, in turn, promote the occurrence and progression of cancer. An assessment of the risk factors associated with lung cancer indicated that smoking is the primary risk factor [23]. Approximately 85% of patients with lung cancer have a history of smoking [24], and 6–83% of patients continue to smoke after their diagnosis of lung cancer [9]. NSCLC is the predominant type of lung cancer. Retrospective studies have indicated that patients who continue smoking after an NSCLC diagnosis experience diminished quality of life and poor clinical outcomes [9, 10]. However, the mechanism by which smoking promotes NSCLC progression remains elusive. In this study, we used in vivo and in vitro models of smoking-associated NSCLC to investigate this mechanism.
Our study showed that low CSE concentrations (1% and 2.5%) significantly enhanced the proliferative capacity of A549 and H1299 cells in vitro. Furthermore, 30-day pretreatment with CSE promoted the growth of subcutaneous xenograft tumors formed by A549 cells in nude mice. Similar findings have been reported previously. Research indicates that CSE accelerates the cell cycle and promotes the proliferation of A549 cells by enhancing the expression of CDK2 and Cyclin E1 [21]. Subsequent analysis of protein expression indicated that CSE enhanced the expression of PCNA, CDK2, and Cyclin E1 in NSCLC cells and xenograft tumor tissues. Additionally, a study on NSCLC has revealed that CSE promotes the proliferation of human LUAD cell lines A549 and H838 by inducing abnormal N6-methyladenosine modification of DAPK2, thereby activating the NF-κB pathway [25]. Research findings indicate that CSE enhances the stemness characteristics of bladder cancer cells and promotes tumor sphere formation in Ej and UM-UC-3 cells by activating the Sonic Hedgehog signaling pathway [26]. Research on gastric cancer has revealed that treatment with a low concentration of CSE significantly increased the relative survival rates of HGC-27 and AGS cells in the CCK-8 assay. This effect was mediated through the circ0000670/Wnt/β-catenin signaling pathway, facilitated by exosomes derived from gastric cancer cells [27]. In endometrial cancer studies, CSE was found to promote proliferation of the Ishikawa cell line by upregulating the expression of Cyclin E and Cyclin D [28]. These findings suggest that the pro-proliferative effects of CSE on cancer cells represent a cross-cancer phenomenon, with distinct mechanisms implicated in different cancer types. Our results are consistent with previous findings. This confirms that cigarette smoke promotes the proliferation of NSCLC cells.
EMT is a biological process in which epithelial cells acquire a mesenchymal phenotype that enhances their invasiveness and motility [29]. EMT is primarily classified into three categories: [1] embryonic and organ development [2], wound healing and tissue regeneration, and [3] tumor progression [30]. Cancer cells initiate EMT in a partial and transient manner and acquire traits that facilitate their dissemination and survival [31]. This transition is associated with metastasis and invasion of cancer cells to distant sites, subsequent colonization, cancer stemness, and resistance to chemotherapy [32, 33]. Smoking exacerbates the progression of various cancers by inducing EMT. A previous study showed that CSE induced EMT of A549 LUAD cells and enhanced their migratory and invasive capabilities in vitro by upregulating the RUNX-2/galectin-3 pathway [34]. Additionally, CSE promotes the migration and invasion of cancer cells through EMT induction in colon cancer [35], ovarian cancer [36], bladder cancer [37], and other malignancies. In our study, we found that CSE induced morphological transformation to spindle-shaped and elongated spindle-shaped cells in A549 and H1299 cell lines. Correspondingly, wound healing and Transwell migration and invasion assays indicated that CSE significantly enhanced the in vitro migration and invasion abilities of NSCLC cells. Western blot analysis revealed that CSE altered the expression of EMT-related proteins in A549 and H1299 cells. Moreover, western blot analysis of xenograft tumor tissues confirmed that CSE promotes EMT. These results demonstrated that CSE promotes the malignant behavior of NSCLC cells by inducing EMT of A549 and H1299 cells. However, the underlying mechanism remains unclear.
In 2013, the impact of FOXC1 on the progression of NSCLC was first reported, indicating a correlation between high FOXC1 expression and poor prognoses of patients with NSCLC [38]. Subsequent research has demonstrated that targeted knockdown of FOXC1 in NSCLC cell lines A549 and NCI-H460 inhibited cell proliferation and migration by suppressing EMT [39]. Further studies have indicated that FOXC1 expression is significantly upregulated in the hypoxic environment, which is characteristic of NSCLC. The gain-of-function of FOXC1 promotes cell proliferation, migration, and invasion, whereas its knockout inhibits these processes [40]. Additionally, during NSCLC progression, FOXC1 may induce stem-like characteristics via the β-catenin signaling pathway [41], bind to the lysyl oxidase promoter to enhance metastasis and invasion [42], and contribute to an immunosuppressive microenvironment [43]. In addition to NSCLC, FOXC1 overexpression is associated with the progression of several cancers, including liver [44], breast [45], pancreatic [46], gastric [47], colorectal [48], and bladder [49] cancers. Our experiments also revealed that FOXC1 expression in NSCLC cells (A549 and H1299) was significantly higher than that in HBE cells. Analysis of the Gene Expression Profiling Interactive Analysis database revealed elevated FOXC1 expression in LUAD and squamous cell carcinoma tissues compared with that in normal lung tissues, although this was not statistically significant. Our findings were consistent with those of previous studies; FOXC1 expression was significantly elevated in the lung tissues of a chronic obstructive pulmonary disease rat model exposed to cigarette smoke [19, 20]. Based on these findings, we hypothesized that the malignant behaviors of CSE-induced NSCLC cells are related to FOXC1. We tested our hypothesis using western blotting, immunofluorescence analysis of both cells and tissues, and gene transcriptome sequencing. Our results demonstrated that CSE upregulated FOXC1 expression in A549 and H1299 cells as well as in xenograft tumors. To further validate the role of FOXC1, we performed a targeted knockdown of FOXC1 in NSCLC cells. These results indicated that the enhanced subcutaneous tumorigenic ability of A549 cells induced by CSE in nude mice, along with the increased proliferation, migration, and invasion capabilities of A549 and H1299 cells induced by CSE, were partially reversed by FOXC1 knockdown. These findings align with our hypothesis that CSE promotes the malignant biological behavior of NSCLC cells by upregulating FOXC1 expression.
Previous studies have shown that FOXC1 is a short-lived protein whose stability depends on the phosphorylation of serine 272. Epidermal growth factor activates the ERK1/2 signaling pathway, which phosphorylates serine 272, thereby reducing the ubiquitination and degradation of FOXC1 and maintaining its stability and activity [50]. Our results revealed that CSE enhanced the expression of phosphorylated ERK1/2 and phosphorylated AP-1 (c-Fos and c-Jun), which are downstream components of the ERK1/2 signaling pathway, in A549 and H1299 cells. This suggests activation of the ERK1/2 signaling pathway. Notably, pretreatment with U0126, the specific ERK1/2 pathway inhibitor, reversed the CSE-induced increase in FOXC1 protein levels in both A549 and H1299 cells, indicating that CSE may upregulate FOXC1 expression by activating the ERK1/2 signaling pathway. This outcome aligns with the findings of studies on liver and colorectal cancers. In liver cancer, high levels of reactive oxygen species upregulate FOXC1 expression by activating the ERK1/2 signaling pathway. FOXC1 subsequently induces DNA hypermethylation of the CTH promoter by upregulating DNMT3B, thereby inhibiting CTH expression and promoting the proliferation and metastasis of hepatocellular carcinoma [17]. Research on colorectal cancer has shown that the ERK1/2 signaling pathway enhances tumor growth and mediates resistance to 5-fluorouracil by upregulating FOXC1 expression [16]. Additionally, we explored the effects of U0126 pretreatment on the malignant behavior of NSCLC cells. The results indicated that U0126 pretreatment reversed the promotional effects of CSE on the proliferation, migration, and invasion of A549 and H1299 cells as well as the CSE-induced changes in EMT-related protein levels. Consistent with these findings, previous studies have shown that inhibiting the activation of the ERK1/2 signaling pathway reverses the EMT phenomenon induced by CSE in human urothelial cells (SV-HUC-1) [51], human bladder cancer cells (T24) [37], and endometrial cancer cells (Ishikawa) [28].
We demonstrated that FOXC1 is a significant intermediate molecule through which smoking facilitates the progression of NSCLC. Additionally, we found that CSE activates ERK1/2, leading to the upregulation of FOXC1 expression (Fig. 12). Our findings suggest that targeting the ERK1/2-FOXC1 pathway is a potential strategy for treating patients with smoking-associated NSCLC. However, some limitations of this study must be acknowledged. First, our research was conducted exclusively using A549 and H1299 cell lines. To enhance the reliability of our findings and validate their applicability, additional NSCLC cell lines and clinical samples should be included. Furthermore, although we have shown that the ERK1/2 pathway mediates CSE-induced upregulation of FOXC1 expression, the precise mechanism by which ERK1/2 regulates FOXC1 expression remains unclear. Therefore, comprehensive and in-depth studies are needed to elucidate these regulatory processes.

Supplementary Information