Quercetin Inhibits the Proliferation and Metastasis of Human Non-Small Cell Lung Cancer Cell Line: The Key Role of Src-Mediated Fibroblast Growth Factor-Inducible 14 (Fn14)/ Nuclear Factor kappa B (NF-κB) pathway.

BACKGROUND Quercetin (Que) is reported to induce apoptosis of lung cancer cells. Src is closely related to the progression of non-small cell lung cancer (NSCLC) and can be modulated by Que in macrophages. In the current study, the interaction between Que and Src signaling in NSCLC cells was explored to explain the anti-NSCLC function of Que. MATERIAL AND METHODS NSCLC cell line HCC827 was subjected to the administrations of Que at different concentrations. The effect of Que on tumor cell proliferation was detected using MTT and colony formation assays. Then the effect on the migration and invasion abilities was assessed using scratch and Transwell assays. At molecular level, the changes in Src/Fn14/NF-kappaB signaling were determined using western blotting assays. The role of Src in the function of Que was further explored by inducing the expression of Src gene in NSCLC cells before Que administration. The results of the in vitro assays were verified using a NSCLC mice model. RESULTS Que inhibited the proliferation and anchorage-independent growth of NSCLC cells. Additionally, Que delayed in the gap closure rate in scratch assays and decreased the membrane-penetrating cell number in Transwell assays. At a molecular level, Que suppressed the expression of Src, which subsequently inhibited Fn14/NF-kappaB signaling. In in vivo assays, Que inhibited the growth of solid tumors. After the overexpression of Src in NSCLC cells, the anti-NSCLC effect of Que was blocked by inducing NSCLC proliferation and metastasis, and by activating Fn14/NF-kappaB signaling. Moreover, the induced level of Src promoted the growth and metastasis potential of solid tumors in mice. CONCLUSIONS Que exerted the anti-NSCLC effect by inhibiting Src-mediated Fn14/NF-kappaB pathway both in vitro and in vivo.

Background

Lung carcinoma accounts for more than 17% of the cancer cases and 20% of the cancer-related death each year, worldwide [1]. Lung carcinoma can be divided into 2 subtypes based on morphological and genetic features: non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) [2]. NSCLC accounts for 80% to 85% of the total lung carcinoma cases [3]. The first-line treatment strategy for NSCLC is chemotherapies using platinum combined with taxanes or other cytotoxic agents [3]. However, many advanced NSCLC patients partially response to these therapies: the response rate is lower than 35% for first-line therapies and it is even lower for the second-line therapies [4,5]. Additionally, the severe side effects associated with chemotherapies renders the application of chemotherapies less satisfactory. Thus, the identification of novel molecular targets or the exploration of novel anti-NSCLC agents is needed.

With increasing attention paid to the treatment of NSCLC, numerous targets for anti-NSCLC therapy development have been identified in the recent years [6], one of which is Src [7]. Src is a member of the Src family kinase (SFK) and is well-characterized by its function in modulating diverse cell signaling transductions [8]. Moreover, its involvement in tumor proliferation, metastasis, and angiogenesis has been previously reported in different cancer types, including prostate cancer [9], gastric cancer [10], and NSCLC [3,11]. The abnormally high level of Src is always associated with the progression of NSCLC, especially for smoking patients [7,12]. Previous studies have indicated that the pro-NSCLC function of Src is exerted through multi-pronged mechanisms: the abnormally high expression of Src contributes to the activation of Fn-14-mediated NF-κB [13] and Ras/PI3K/Akt signaling [14]. Thus, it is reasonable to develop anti-NSCLC therapies by specifically inhibiting the function of Src.

Dietary phytochemicals may have anti-tumor activities through a number of strategies, including inhibiting tumor growth, metastasis, and angiogenesis [15]. As a typical flavonoid abundant in fruits and vegetables, quercetin (Que) (3,30,40,5,7-pentahydroxylflavone) can modulate cell cycle distribution, apoptosis, inflammation, and angiogenesis in multiple cancer types [16-18], including lung cancer [19,20]. In addition, Que can also suppress the activation of Src signaling in macrophages [21]. Moreover, Que is a major extracted from fruits and vegetables, and it has low cell toxicity and is abundant in nature. Thus, Que is a promising candidate for managing cancers when compared with traditional chemo-agents such as cisplatin [22-24]. The aforementioned information indicates that the anti-tumor function of Que might be related to its inhibition on Src.

To verify this hypothesis, we selected human NSCLC HCC827 cell line with high expression level of Src to evaluate the anti-NSCLC effect and associated mechanism of Que in vitro and in vivo. We administrated Que in both cell and animal experiments, and we assessed the effects on the survival and metastasis potentials of NSCLC cells. Moreover, the expression of Src was induced in HCC827 cells before Que treatment to confirm that the anti-NSCLC function of Que was associated with the function of Src and its downstream pathways.

Material and Methods

Cell culture and Que administration

Human NSCLC cell lines HCC827 (7-1150) and NCI-H1650 (7-1031) were obtained from Chi Scientific (MA, USA). Both cell lines were commercially available and authenticated using STR method by Chi Scientific. Mycoplasma contamination was excluded using MycAwayTM-Color One-Step Mycoplasma Detection Kit (40611ES25, Yeasen, China). Que (purity >99%) was dissolved in dimethyl sulfoxide (DMSO) at concentrations of 25 μM, 50 μM, and 100 μM [15]. The doses were designated Que L at 25 μM, Que M at 50 μM, and Que H at 100 μM. The cells were incubated with different Que (purity >99%, dissolved in DMSO, Q109798, Aladdin, China) doses for 72 hours and Que of 100 μM was selected for the subsequent assays. Cisplatin of 10 μM was employed as positive control [25,26]. In addition, another NSCLC cell line NCI-H1650 (7-1031, Chi Scientific, USA) was employed to confirm the anti-NSCLC effect of Que; the results are shown in Supplementary Figure 1.

MTT and colony formation assays

NSCLC cells were cultured for 96 hours. Every 24 hours, 5 mg/mL MTT was added into wells and incubated for another 4 hours at 37°C. The cell viability was represented by the optical density (OD)490 value using a microplate reader (ELX-800, BIOTEK, USA).

For the colony formation assay, 200 cells were incubated with 100 μM Que and 0.35% agarose and incubated using a 35 mm plate at 37°C for 2 weeks. Then the colonies were stained with Wright-Giemsa stain and the colony formation rate was calculated.

Scratch assay and Transwell assays

For scratch assay, cells were allowed to grow into a monolayer and reference points were marked on the confluent surface. A cell-free straight line was scratched on the cell layer and cell migration towards the midline was recorded in reference to the marked points in a 48-hour period by taking 3 images at 0, 24, and 48 hours. The migration rate was analyzed using the ImageJ software (US National Institutes of Health).

For Transwell assays, cells were inoculated in the upper chamber of a Transwell system pre-coated with Matrigel. The cells were allowed to penetrate through the for 24 hours. After completely removing the cells on the upper surface, cells on lower surfaces were stained with 0.5% (w/v) crystal violet. The numbers of cells stained dark purple were detected using an Image-Pro Plus 6.0 software (Nikon).

Western blotting assays

Total protein was extracted and subjected to 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The membranes were first incubated with the primary antibodies against Src (1: 500) (D120536, Sango Biotech, Shanghai, China) Fn14 (1: 500) (D121267, Sango Biotech, Shanghai, China), IκBα (1: 500) (bs-1287R, Bioss, Beijing, China), p-IκBα (1: 500) (bs-18129R, Bioss, Beijing, China), IKKβ (1: 500) (bs-4880R, Bioss, Beijing, China), p-IKKβ (1: 500) (bs-3233R, Bioss, Beijing, China), NF-κB p65 (1: 400) (PB0073, Boster, Wuhan, China), E-cadherin (1: 5000) (No. ab76055, Abcam, Cambridge, UK), N-cadherin (1: 5000) (No. ab18203, Abcam, Cambridge, UK), β-actin (1: 1000) (sc-47778, Santa Cruz, CA, USA) or Histone H3 (1: 500) (bs-17422R, Bioss, Beijing, China) at 4°C overnight and then incubated with HRP-conjugated secondary antibodies (A0208, A0216, A0521, Beyotime, Shanghai, China) for 45 minutes. The integrated ODs (IODs) of the blots were calculated by the Gel-Pro-Analyzer (Media Cybernetics, USA).

Construction of Src expression vector and transfection

Src coding sequence was ligated to pcDNA3.1+ to construct pcDNA3.1-Src vector and transfected into HCC827 cells using Lipofectamine 2000 (Invitrogen) following the instruction. At 48 hours after the transfection, the expression of Src in HCC827 cells were determined by western blotting assays. Then the cells were further administrated with 100 μM Que and the role of Src in the anti-NSCLC effect of Que was assessed with assays.

NSCLC xenograft mice model

Twenty-four BALB/c mice were randomly divided into 4 groups (6 mice in each group). The Control group mice received a subcutaneous injection of 1×107 HCC827 cells (in 0.2 mL volume). The Que group mice received a subcutaneous injection of 1×107 HCC827 cells and a daily intraperitoneally administration of Que of 100 mg/kg body weight. The Src group mice received a subcutaneous injection of 1×107 Src-overexpressed HCC827 cells and a daily intraperitoneally administration of Que of 100 mg/kg body weight. The mice were housed for 3 weeks under the same conditions, and the tumor volumes were measured every 3 days beginning Day 7. After 3 weeks, the mice were sacrificed with an overdose intraperitoneal injection of 150 mg/kg body weight of pentobarbital sodium. Then tumor tissues were collected for the evaluation of histological changes by hematoxylin and eosin (H&E) staining. All the animal experiments were performed following the ethical standards in the 1964 Declaration of Helsinki and its later amendments, with the approval of the Ethics Committee of the Hubei University of Medicine.

H&E staining

H&E staining was performed routinely: tumor tissues were placed into Bouin solution, embedded in paraffin, sectioned, and stained with H&E. The results were captured using a microscope under 200× magnification.

Statistical analysis

Each group was represented by 6 replicates. All the statistical analyses were performed using GraphPad Prism version 6.01 for windows (GraphPad Software). Data were represented by mean±standard deviation (SD) and one-way analysis of variance (ANOVA) and post doc tests with Tukey method were performed using general liner model with a significance level of 0.05 (2-tailed P value).

Results

Que inhibited proliferation potential of HCC827 cells

As shown in Figure 1A, Que of all the 3 doses inhibited the growth potential of NSCLC cells. The effect gradually strengthened with time and the OD values in Que of 50 μM and 100 μM were significantly lower than that of the Control group (P<0.05). Moreover, the effect of Que on cell viability was exerted in a dose-dependent manner: the inhibition effect of Que of 100 μM was stronger than the other lower 2 doses (Figure 1A). Compared with the inhibitive effect of cisplatin, the effects of Que of 25 μM and 50 μM were significantly lower (P<0.05), but the effect of Que of 100 μM was similar to that of cisplatin (Figure 1A). Colony formation assays were performed to assess the influence of Que on the anchorage-independent growth of NSCLC cells. It was shown that Que at 100 μM significantly inhibited the colony formation rate in NSCLC cells (22.17% in the Que H group versus 35.17% in the Control group) (P<0.05) (Figure 1B). The effect of Que of 100 μM was weaker than that of cisplatin (19.94%).

Figure 1

Administration of Que suppressed the proliferation potential of HCC827 cells. For MTT assays, cells were incubated with Que of 25 μM (Que L), 50 μM (Que M), and 100 μM (Que H) for 96 hours. Every 24 hours, cells were subjected to the testing. For colony formation assays, cells were cultured in medium containing 100 μM Que for 2 weeks. (A) Detection results of MTT assays. (B) Detection results of colony formation assays. * P<0.05 versus the Control group.

Que suppressed the inhibited migration and invasion in HCC827 cells

As illustrated by the scratch assays, Que decreased the migration of NSCLC cells in a 48-hour period. The migration rates of the Que H group were significantly lower than those of the Control at the 2 recording points (24.79±3% at 24 hours and 28.95±1.9% at 48 hours versus 33.76±6.32% at 24 hours and 42.40±4.4% at 48 hours) (P<0.05) (Figure 2A). The effect of Que of 100 μM on the migration of HCC827 cells was relatively lower to that of cisplatin (20.50±4.8% at 24 hours versus 23.82±2.76% at 48 hours). Que of 100 μM also resulted in a significantly lower proportion (28.8±2.4%) of HCC827 cells penetrating the Matrigel-coated membrane compared with the Control group (65.2±5.9%) (P<0.05) (Figure 2B).

Figure 2

Administration of Que inhibited the invasion and migration potentials, and the signaling transduction of Src/Fn14/NF-κB pathway in HCC827 cells. For scratch assays, cells were incubated with Que of 100 μM (Que H) for 48 hours. For Transwell assays, cells were incubated with Que of 100 μM (Que H) for 24 hours. For western blotting assays, cells were incubated with Que of 100 μM (Que H) for 24 hours. (A) Detection results of scratch assays. (B) Detection results of Transwell assays. (C) Detection results of western blotting assays. * P<0.05 versus the Control group.

Anti-NSCLC effect of Que was dependent on the inhibition of Src

To verify our hypothesis that the anti-NSCLC effect of Que was exerted through the Src pathway, we detected the changes in Src/Fn14/NF-κB signaling. The results showed that Que of 100 μM suppressed the levels of Src, Fn14, p-IκBα, p-IKKβ, and nuclear NF-κB p65, while it increased the level of IκBα (Figure 2C), indicating that the anti-NSCLC effect of Que was associated with the inhibition of Src/Fn14/NF-κB pathway. Then Src was stably induced in HCC827 cells before Que treatment (Figure 3A). The overexpression of Src counteracted the inhibition effect of Que on HCC827 cells. In cells transfected with Src expression vector, the OD490 value (Figure 3B), colony formation rate (28.94±4.67% versus 39.44±5.09%) (Figure 3C), membrane-penetrating cell number (18.47% ± 2.11% at 24 hours and 22.25% ± 0.67% at 48 hours versus 29.25±1.87% and 35.65±4.82% at 48 hours) (Figure 3D), and migration rate (24.2±1.5% versus 41.2±1.9%) (Figure 3E) were much higher than those in Que H+NC group, indicating the restored viability anchorage-independent growth, migration, and invasion in HCC827 cells.

Figure 3

Overexpression of Src blocked the anti-NSCLC function of Que in HCC827 cells. Cells transfected with negative control vector (NC) or Src expression vector (Src) were treated with Que of 100 μM (Que H) and subjected to MTT assay for 96 hours (cells were collected every 24 hours), and colony formation was assay at 2 weeks, Transwell assay at 24 hours, and scratch assay at 48 hours. (A) Detection results of Src level. (B) Detection results of MTT assays. (C) Detection results of colony formation assay. (D) Detection results of Transwell assay. (E) Detection results of scratch assay. # P<0.05 versus Que H group.

Que inhibited solid tumor growth in NSCLC xenograft mice by inhibiting Src

As shown in Figure 4A, the tumor volumes were solidly inhibited by Que of 100 mg/kg body weight in a 21-day time period when compared with the Control group. However, for mice transplanted with HCC827 cells transfected with Src expression vector, the tumor volumes dramatically increased to a level similar to the Control group (Figure 4A). For H&E staining, cell structure in the Control and Que+Src groups were more integrated than that in the Que group (Figure 4B) (P<0.05). Que also changed levels of epithelial-mesenchymal transition (EMT) indicators by inducing E-cadherin level, while suppressing N-cadherin level in tumor tissues, which was reversed by Src overexpression (Figure 4C). The effects of Que on solid tumor were also compared with cisplatin. As shown in Supplementary Figure 2, the effects of Que of 100 mg/kg body weight were weaker than cisplatin

Figure 4

Que inhibited the growth and metastasis potential of NSCLC in vivo in by inhibiting Src signaling. Mice were injected with different HCC827 cells and administrated with Que of 100 mg/kg body weight for 3 weeks. (A) Detection results of tumor volume. (B) Detection results of hematoxylin and eosin detection of tumor tissue. (C) Detection results of western blotting detection of E-cadherin and N-cadherin. * P<0.05 versus Que H+Src group. Scale bar, 100 μm.

Discussion

Flavonoids are antioxidants abundant in diets and constitute the largest group of natural phenols [27,28]. Increasing evidence proves that these antioxidants also have anti-tumor functions [27,28]. In the current study, we assessed the effects of Que on the growth and metastasis potentials of NSCLC cells. The results showed that Que inhibited the viability, anchorage-independent growth, migration, and invasion of NSCLC cells. The effects of Que NSCLC cells were associated with inhibition of Src/Fn14/NF-κB signaling.

Currently, the most used strategy for handling NSCLC is chemotherapy, such as cisplatin which can induce apoptosis and suppress metastasis in NSCLC cells [3,11]. Despite its high efficiency, chemotherapies are less satisfactory due to side effects. As a natural compound that is abundant in fruits and vegetables, Que is characterized by low cell toxicity and easy availability, which are basic advantages of Que over chemo-agents. A previous study by Zhu et al. shows that Que suppressed the growth of lung cancer by targeting Aurora B kinase [20]. Regarding its effect on NSCLC, the compound has been reported to not only antagonized NSCLC cells through its effect on the cytoskeleton [15], but also induced apoptosis in NSCLC cells [29]. In our study, Que suppressed the survival and metastasis potentials of NSCLC cells. Moreover, the effect of Que on NSCLC cell viability was exerted in a dose-dependent manner, which was consistent with the results of Klimaszewska-Wisniewska et al. [15].

In a study by Yang et al., it was found that Que suppressed the activity of Src, which is known pro-tumor factor for multiple cancer types including NSCLC [3,21]. Our current study detected the changes in Src/Fn14/NF-κB signaling in NSCLC cells under treatment with Que of 100 μM. The suppressive effect of Que on the Src/Fn14/NF-κB was verified by the inhibited expression levels of Src, Fn14, and NF-κB p65. However, such results only prove the association between Que and Src-mediated signaling. To confirm the key role of Src inhibition in the anti-NSCLC function of Que, we induced the level of Src in NSCLC cells. The results showed that in NSCLC cells transfected with Src expression vector, the suppressive effects of Que on viability, anchorage-independent growth, migration, and invasion of NSCLC cells was weakened. These results together inferred that the anti-NSCLC function of Que depended on the inhibition of Src. Generally, the activation of Src is associated with the oncogenesis of different organs, including lung, prostate, pancreas, breast, and colon [7]. Regarding NSCLC, the high expression of Src has been detected in 50% of squamous cell carcinoma samples isolated from NSCLC patients [30]. The factor generally induces Fn14 expression in NSCLC. Moreover, in the study by Funakoshi-Tago et al., Src is key to the interleukin (IL)-1-induced IKKβ activity, which then initiates the activation of the NF-κB pathway [31,32]. Based on the study by Wang et al., Fn14 mediates the effect of Src on NF-κB pathway [13]. Taken together, Que first inhibits the function of Src in NSCLC cells, which subsequently suppresses the activity of Fn14/NF-κB. The changes in Src/Fn14/NF-κB signaling negative influences the proliferation and metastasis potentials of NSCLC.

The anti-NSCLC effects of Que were preliminarily reported by several previous studies [19,20,29], but those studies majorly focused on its effect to induce tumor cell apoptosis. The current study performed a comprehensive assessment on the anti-NSCLC effects of Que using a series of in vitro and in vivo assays. In addition, the current study was the first to explore the mechanism driving the anti-NSCLC effects of Que by focusing on its interaction with Src, which provided supplementary information for the development of Que-based anti-NSCLC therapies.

Conclusions

The findings outlined in the current study provided additional information regarding the mechanism driving the anti-NSCLC function of Que. The compound contributed to the suppression of the proliferation and metastasis potentials of NSCLC cells by inhibiting Src/Fn14/NF-κB signaling. The low cytotoxicity and high abundance of Que represent the promising application prospect of the agent in the clinical management of NSCLC.

Que suppressed the proliferation and metastasis potential of NCI-H1650 cells. For MTT assays, cells were incubated with 100 μM (Que H) for 96 h and every 24 h, cells were subjected to the tests. For colony formation assays, cells were cultured in medium containing 100 μM Que for two weeks. For scratch assays, cells were incubated with Que of 100 μM (Que H) for 28 h. For transwell assays, cells were incubated with Que of 100 μM (Que H) for 24 h. (A) Quantitative analysis results of MTT assays. (B) Representative images and quantitative analysis results of colony formation assays. (C) Representative images and quantitative analysis results of scratch assays. (D) Representative images and quantitative analysis results of transwell assays. * P<0.05 vs. Control group.

Comparison between Que and cisplatin effects in xenograft mice model. (A) Detection results of tumor volume. (B) Detection results of H&E detection of tumor tissue. * P<0.05 vs. Que group. Scale bar, 100 μm.