The C. tinctoria was from the research base of the Uyghur Autonomous Region Institute of Medicinal Plants (Xinjiang, China). Human lung carcinoma (A549 and NCI-H292) cells and BEAS-2B were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). RPMI-1640 medium and fetal bovine serum (FBS) was from Gibco (United States). Penicillin/streptomycin, and dimethyl sulfoxide (DMSO) were purchased from Sigma (Missouri, United States). Cell Counting Kit-8 (CCK-8) and Annexin V-FITC/PI apoptosis detection kits were from BD Biosciences (United States). HPLC grade methanol and acetonitrile were purchased from Merck (Darmstadt, Germany). The antibodies of Caspase-3, Bax, and Bcl-2 were all obtained from Abcam (United States); those of p-PI3K, PI3K, p-Akt, and Akt were from Affinity Biosciences (Cincinnati, United States); and, that of β-actin was from Proteintech Group (Wuhan, China). An enhanced chemiluminescence (ECL) kit was purchased from PerkinElmer (Waltham, United States).

The total flavonoids with a purity of 80% (w/w) were extracted by evaporator method and purified by Sephadex LH-20 column. In brief, the C. tinctoria was dried at 60°C, made into powders, and filtered through a 60-mesh sieve. The dried powder was incubated in 70% ethanol (1:10, w/v) for 24 h, filtered, and extracted twice in a vacuum rotary evaporator at 70°C, each time for 2 h. The extracted solutions were combined and concentrated to yield a dried residue. The residue was then subjected to HPD-100A resin and eluted with H₂O, 20, 40, and 65% ethanol, successively. The fraction eluted with 40% ethanol was subjected to Sephadex LH-20 column and eluted with 10–90% methanol for purification of total flavonoids.

A549, H292, and BEAS-2B cells were cultured in RPMI-1640 medium supplemented with 10% FBS and 1% penicillin/streptomycin in a humidified atmosphere of 95% air and 5% CO₂ at 37°C. The medium was subsequently replaced at 2-day intervals.

The CCK-8 assay was used to determine the effect of CTFs on the cell viability of cells. Briefly, the cells were seeded onto 96-well plates at 1 × 10⁴ cells/well for 24 h and then incubated with various concentrations of CTFs (25, 50, 100, 200, 400, 600, 800 or 1,000 µg/ml) for 24, 48, and 72 h at 37°C. The cells treated by DMSO were used as negative control. Following incubation, CCK-8 was added to each well, followed by incubation for 2 h at 37°C. The optical density (OD) at 490 nm was measured using a microtiter plate reader. Cell viability (%) = OD (experimental group)/OD (control group) × 100%. Each experiment was performed in triplicate.

UHPLC-MS/MS was used to determine the chemical profile of CTFs. The separation was carried out on Waters Acquity UPLC, BEH C18 (1.7 μm, 2.1*150 mm) at a flow rate of 0.3 ml/min. The mobile phase consisted of a combination of A (0.1% formic acid in water, v/v) and B (acetonitrile). AB Sciex QTrap 6500 + mass spectrometer was applied for assay development. Typical ion source parameters were: IonSpray Voltage: +5000/−4500 V, Curtain Gas: 35 psi, Temperature: 50°C, Ion Source Gas 1: 55 psi, Ion Source Gas 2:60 psi.

To evaluate the pharmacokinetic properties of CTFs, all compounds were analyzed using the TCMSP database (http://lsp.nwu.edu.cn/). The active components of CTFs were screened according to the criteria of oral bioavailability (OB) ≥ 30% and drug-likeness (DL) ≥ 0.18. Then, the targets of the active components were predicted by the TCMSP platform and ETCM database (http://www.tcmip.cn/ETCM/index.php). The STRING database (https://string-db.org/) and the “Homo sapiens” species settings were used to identify and standardize the target proteins of the active components of CTFs.

The lung cancer-related protein targets were screened using the following two databases: OMIM (http://www.ncbi.nlm.nih.gov/omim/) and the DisGeNET (http://www.disgenet.org/). We screened the predicted targets using “lung cancer” as the keyword. All high-quality binary protein-protein interactions of human species were downloaded from the high-quality proteomics database HINT (High-quality INTeractomes) to construct a human genome-wide protein-protein interaction network (PPI). All disease targets of lung cancer were mapped to the HINT background network. After extracting the lung cancer network, the CTFs targets were mapped to the lung cancer network. All target nodes and the edges between them were collected, and only the largest connected branch was retained to obtain the target network. Through network topology analysis, potential targets were obtained.

Potential targets were imported into the DAVID 6.8 database (https://david.ncifcrf.gov/) to perform GO enrichment analysis and KEGG pathway enrichment analysis. “OFFICE_GENE_SYMBOL” and “Homo sapiens” were selected. The top 10 enriched GO entries and top 15 enriched KEGG pathways that met the criteria of p < 0.01 were selected, and uploaded to the OmicShare (http://www.omicshare.com/tools/) cloud platform for data visualization, where the size of the bubble represents the number of genes in the pathway, and the color represents the significance of enrichment.

The integrated network of component-target-pathway was constructed using Cytoscape 3.6.0 to identify the relationships of protein targets with each compound and the involved pathways and diseases. The topology characteristics of the network were evaluated.

To further verify the reliability of the target prediction results, the key targets with degrees greater than the average degree in the “component-target-pathway” network were selected and Auto dock vina (Trott and Olson, 2010). was used for molecular docking to verify the binding activity of key target with components.

A549 and H292 cells were plated into 6-well plates (1 × 10⁶ cells/well) and treated with different concentrations of CTFs (25, 50, 100, and 200 µg/ml) for 24 h. The cells were collected and stained with Annexin V-FITC/PI according to the manufacturer’s protocol for 15 min in the dark. Apoptotic cell ratio was detected using a flow cytometer. Each experiment was done in triplicates.

Mitochondrial mass was observed by the fluorescent probe Mito-tracker Red CMXRos (Beyotime, Shanghai, China). Cells were incubated with different concentrations of CTFs (50, 100, and 200 µg/ml) for 24 h, and then treated with the Mito-tracker Red CMXRos for 30 min at 37°C in the dark. The results of mitochondrial staining was observed by confocal fluorescence microscopy (Leica Microsystems, Heidelberg, Germany).

MMP was detected by the JC-1 (Beyotime, Shanghai, China). Cells were incubated with different concentrations of CTFs (50, 100, 200 µg/ml) for 24 h. Then, cells were stained with JC-1 at 37°C in the dark for 20 min. After washing 2 times with PBS, JC-1 aggregates in cells were analyzed by confocal microscope. The ratio of red/green fluorescence intensity was calculated with the ImageJ software.

After treatment with CTFs (0, 50, 100, and 200 µg/ml) for 24 h, A549 cells were collected and lysed with RIPA lysis buffer (containing a 1% protease inhibitor cocktail). The protein concentration was measured with the BCA Protein Assay Kit. The protein samples were separated by 10% SDS-PAGE and transferred onto PVDF membranes (Millipore, Bedford, MA, United States). After blocking with 5% (v/v) non-fat milk for 2 h, membranes were incubated with primary antibodies overnight at 4°C. The primary antibodies included anti-caspase 3 (1:2,000), anti-Bcl-2 (1:1,000), anti-Bax (1:1,000), anti-p-PI3K (1:1,000), anti-PI3K (1:1,000), anti-p-Akt (1:1,000), anti-Akt (1:1,000), and anti-β-actin (1:500). The membranes were incubated with a horseradish peroxidase-conjugated secondary antibody for 2 h at room temperature. The protein bands were visualized using ECL detection reagents (Bio-Rad, United States) and the band density was evaluated using ImageJ software.

Nude mice (BALB/c null; 6-week-old) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (China). The Institutional Animal Care and Use Committee at Xinjiang Medical University approved all animal experiments in this study. Nude mice were given sterile food and water and kept in pathogen-free conditions. A549 cells (1 × 10⁷) were injected subcutaneously into the left flanks of mice. Tumor length (a) and width (b) were measured using Vernier calipers every 3 days. Tumor volume was calculated using the formula: V (mm³) = 1/2*ab². Following cell injection for 14 days, mice were randomly divided into four groups (n = 6): model group; low dose CTFs group (50 mg/kg); medium dose CTFs group (100 mg/kg); high dose CTFs group (200 mg/kg) and positive control group (5-fluorouracil, 50 mg/kg). Drugs were dissolved in carboxymethyl cellulose sodium salt and administered by gavage. The body weight and tumor volume of mice were measured every 3 days until the 24th day. Then, mice were sacrificed, and tumors were collected and stored at −80°C.

Nude mouse tumor tissues were fixed with 10% neutral formaldehyde, routinely dehydrated, embedded in paraffin, sectioned, and stained with HE. The morphological changes of the tumor tissues in each group were observed under an optical microscope.

Cell apoptosis in mice tumor tissues was examined using TUNEL assay (Beyotime Biotech, Wuxi, China) according to the manufacturer’s instructions. The sections were incubated with 20 μL of TUNEL reaction mixture for 30 min in a humid chamber in the dark, and then the sections were washed three times (5 min each) with TBS. Then, 50 μL streptavidin-HRP solution (1:100) was added room temperature, followed by three washes with TBS (5 min each) and DAPI redyeing. The slices were mounted and observed under fluorescence microscope.

The xylene deparaffinized and hydrated tissue sections were incubated with 3% H₂O₂ to block endogenous peroxidase for 20 min at room temperature. After washing with PBS (3 min × 3 times), normal goat serum was added for blocking at 37°C for 30 min. Primary antibodies of Bax, Bcl-2, p-PI3K, and p-Akt were then added for incubation overnight at 4°C. After washing again, the secondary antibody of polyperoxidase-anti-Rabbit IgG was added and incubated at 37°C for 30 min. DAB was used for color development. The counterstaining with hematoxylin was performed. Finally, the sections were dehydrated with gradient alcohol, transparent with xylene, and mounted with neutral gum.

Statistical analysis was performed using the SPSS program 23.0. Data were expressed as mean ± standard deviation (SD). Statistical comparisons were performed using Student’s t-test and one-way analysis. Values of p < 0.05 indicated statistically significant differences.

The components of CTFs were analyzed by UHPLC MS/MS. A total of 68 components were identified. Using the TCSMP database based on ADME parameters, 23 compounds met the screening criteria (OB ≥ 30% and DL ≥ 0.18). The details of 23 compounds are shown in Table 1.

The targets collected in the TCMSP natural product database were matched to “Homo sapiens” in the STRING database. After deduplication, 288 targets were obtained. The lung cancer-related targets collected in the OMIM and DisGeNET databases were de-duplicated and mapped to the established HINT background network. After extracting the network, compound targets were mapped to the lung cancer disease network. The largest connected pieces were extracted, and a common-target network was established (Figure 2). According to the network topology analysis, the average connectivity of the network nodes was 5.3. Thus, the 51 targets with connectivity greater than or equal to six were identified as candidate targets, which may play a pivotal role in the entire network.

GO enrichment analyzed the functional distribution of candidate targets, and 249 GO entries were determined according to p < 0.01. Among them, 184 GO entries were related to biological processes, including positive regulation of transcription from RNA polymerase II promoter, negative regulation of the apoptotic process, positive regulation of transcription, DNA-templated, cell proliferation, and intrinsic apoptotic signaling pathway in response to DNA damage. There were 21 GO entries related to cell components, including cytosol, nucleoplasm, nucleus, cytoplasm, and mitochondrion; and, 44 molecular function entries, including transcription factor binding, protein binding, identical protein binding, enzyme binding, protein heterodimerization activity, etc. The top 10 (according to p < 0.05) in the BP, CC, and MF categories were shown in Figure 3.

KEGG results showed that 51 targets were mainly enriched in 77 signal pathways (p < 0.01), primarily involving human diseases, environmental information processing, organic systems, cellular processes, such as pathways in cancer, small cell lung cancer, viral carcinogenesis, non-small cell lung cancer, PI3K-Akt signaling pathway, Toll-like receptor signaling pathway, apoptosis, cell cycle, etc. The top 15 (according to p < 0.05) pathways were shown in Figure 4.

The 15 main pathways were obtained from the enrichment analysis corresponding to the active components and targets of CTFs, and an “active component-target-pathway” network was constructed (Figure 5). The network consisted of chemical components, protein targets, and pathways, including 85 nodes and 456 edges. Through network topology analysis, 17 protein targets with a degree (average degree of connectivity was 10.7) greater than 11 were obtained. Compared with the 51 protein targets mentioned above, these 17 protein targets were more important and may become key targets for CTFs intervention in lung cancer.

In the “component-target-pathway” network, there were 17 targets with degree ≥11, and the CTFs potential active components corresponding to these targets were searched backward for molecular docking. Binding energy (affinity) < −7.0  kcal mol⁻¹ indicates better binding activity (Saraswati et al., 2019), and the lower the binding energy, the better the docking effect. The results of molecular docking (Figures 6, 7) demonstrated that the active components of CTFs had good binding activity with the target proteins of lung cancer intersection. Meanwhile, these results verify the reliability of the predicted targets.

To detect the effect of CTFs on proliferation of lung cancer cells, CCK-8 assay was performed. After treatment with CTFs (0, 25, 50, 100, 200, 400, 600, and 800 µg/ml) for 24, 48, 72 h, the semi-inhibitory concentration (IC50) was 152.8,79.52, and 31.23 µg/ml, respectively. The IC50 of H292 was 73.58,16.10, and 13.19 µg/ml, respectively. The results showed that CTFs significantly inhibited the proliferation of A549 and H292 cells in a time- and dose-dependent manner in comparison with control group (Tables 2 and 3). However, CTFs did not suppress BEAS-2B cell viability until its concentration was over 200 µg/ml (Figure 8). Collectively, these results demonstrate that CTFs suppress the survival of lung cancer cells in vitro.

The flow cytometry was used to detect apoptosis. After the treatment with 50, 100, and 200 µg/ml of CTFs for 24 h, the apoptosis rate of A549 cells was 18.87 ± 1.27, 32.57 ± 0.38, 60.17 ± 5.60 vs/ 15.7 ± 2.66, respectively, compared to the control group. The apoptosis rate of H292 cells was 16.27 ± 0.55, 20.33 ± 0.15, 47.03 ± 0.64 vs/ 9.73 ± 2.16, respectively, when cells were treated with CTFs (25, 50, and 100 µg/ml). This suggested that CTFs induced apoptosis of A549 and H292 cells in a dose-dependent manner (Figure 9A). The numbers of cells in both early and late-stage apoptotic proportions were remarkably increased in cells treated with CTFs.

Mito-tracker Red CMXRos can exhibit good photo stability to respond to changes in MMP. As shown in Figure 9B, there was a significant decrease of red fluorescence intensity in the CTFs group compared with control group (p < 0.01), suggesting a decrease of MMP with the presence of CTFs. To further verify this results, the MMP was also monitored by JC-1 fluorescent probe. The relative ratio of red and green fluorescence was used to evaluate the change of MMP. The results in Figure 9C showed that compared with the blank control cells, as the increase of CTFs concentration, the ratio of green fluorescence to red fluorescence increased, suggesting that CTFs can collapse the membrane potential by increasing the permeability of the mitochondrial membrane of lung cancer cells.

To evaluate the effect of CTFs on tumor growth in vivo, a nude mouse model of the transplanted tumor was established. When the tumor volume reached 40 mm³, mice were given intervention with low, medium, and high doses of CTFs for 4 weeks. Compared with the model control group, the high-dose CTFs group significantly reduced the tumor weight and tumor volume of nude mice (0.65 ± 0.25 vs. 0.36 ± 0.14 g, 334.38 ± 26.18 vs. 204.47 ± 79.45 mm3). The tumor weight and volume of low and medium-dose group tended to decrease, but there was no statistical difference compared with the control group (Figure 10A). HE staining results showed that compared with the model group, after CTFs intervention, there were obvious damage and necrosis in the tumor tissue (Figure 10B). The results of TUNEL staining suggested that the intervention of CTFs significantly promoted the apoptosis of cells in tumor tissues (Figure 10C).

Western blot was used to detect changes in the expression of apoptosis-related proteins. The results were shown in Figure 11. Compared with the control cells, CTFs intervention significantly increased the ratio of Bax/Bcl-2 and the expression of Caspase-3 (p < 0.05). The levels of p-PI3K and p-Akt were significantly down-regulated (p < 0.05). However, PI3K and Akt protein expression levels did not change significantly (p > 0.05) in A549 cells (Figure 11A). Immunohistochemical results showed that Bax protein expression increased significantly after CTFs intervention, and Bcl-2, p-PI3K, and p-Akt protein expression decreased significantly (p < 0.05) in vivo (Figure 11B). The above results suggest that CTFs may regulate the expression of Bax, Bcl-2, and Caspase-3 to induce cell apoptosis through PI3K/Akt signaling pathway.