1. INTRODUCTION
Lung cancer is one of the most common malignancies and the leading cause of cancer-associated mortality worldwide [1]. Non-small cell lung cancer (NSCLC), accounting for 80% of lung cancers, has poor prognosis, owing to tumor metastasis or drug resistance, thus impeding advances in targeted therapies [2]. The identification of novel targeted drugs that attenuate aberrant proliferation and tumor metastasis remain urgently needed.
The Skp1-Cullin1-F-box protein (SCF) complex, an important E3 ligase in the ubiquitin proteasome system, has emerged as a therapeutic target for various cancers [3]. Numerous anticancer drugs have been developed to target SCF-complex components, particularly the substrate-recognizing F-box protein Skp2. Skp2 interacts with Skp1 and forms two binding pockets around the F-box motif in Skp2, thus facilitating drug targeting [4]. Compounds such as betulinic acid and compound #25 (SZL P1-41), directly bind Skp2 and hinder Skp1-Skp2 interaction [5, 6]. These Skp2 inhibitors abrogate proteolytic ubiquitination of p27 and E-cadherin, as well as non-proteolytic ubiquitination and phosphorylation of Akt, thereby inhibiting cell-cycle progression, tumor metastasis, glycolysis, and cancer stemness [5]. Furthermore, the high expression of Skp2 in multiple human cancers is associated with poor prognosis in patients [7]. Therefore, Skp2 may be a promising therapeutic target for the development of anticancer drugs.
Silybin (SB), also known as silibinin, is the main active component of silymarin, which is isolated from the seeds of Silybum marianum. SB and silymarin are used primarily as hepatoprotective agents to treat hepatitis in clinical settings [8], because of their strong antioxidant and free-radical-scavenging properties. SB, as a natural flavonoid, also has anti-fibrotic, anti-inflammatory, anti-cancer, anti-diabetic, and neuroprotective effects [9]. Recent studies have shown that SB exhibits multiple anticancer activities [10]. SB changes the expression of drug transporters, apoptotic proteins, and cell-cycle regulators, thus exhibiting both chemosensitizing and chemopreventive activities in cancer treatments [11]. SB may also regulate the activity of Stat3, Cdk4/2, MMP-2, PI3K/AKT, and ERK, thereby attenuating cancer cell proliferation and migration [12]. Notably, SB has been found to increase the protein levels of p21/p27 and to induce G1-phase cell-cycle arrest [13, 14], thus contributing to anticancer effects. Therefore, the therapeutic targets and underlying anticancer mechanisms of SB are worthy of further investigation.
Here, we demonstrated that SB binds Skp2 and disrupts the integrity of Skp2-SCF E3 ligase in NSCLC, thus leading to the accumulation of p27 both in vitro and in vivo. SB also was found to promote E-cadherin protein accumulation and attenuate Akt phosphorylation. Our study illustrates the anticancer mechanisms of SB underlying the induction of cell-cycle arrest and suppression of tumor metastasis.
2. MATERIALS AND METHODS
2.1 Reagents and antibodies
SB (98% purity) was obtained from TCI Chemical Co. (Tokyo, Japan). Cycloheximide (CHX) and SZL P1-41 were obtained from MedChemExpress (Monmouth Junction, NJ, USA). Kits used for blood urea nitrogen (BUN), serum creatinine (SCR), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) determination were obtained from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Antibodies against Skp2 (#2652), p27 (#3688), E-cadherin (#14472), p-AKT (Ser473, #12694), and GAPDH (#2118) were obtained from Cell Signaling Technology (Beverly, MA, USA); the antibody against FLAG (F1804) was purchased from Sigma. All secondary antibodies (horseradish-peroxidase-conjugated anti-mouse IgG and anti-rabbit IgG) were purchased from Affinity Biosciences.
2.2 Cell culture
The human lung cancer cell lines H1299 and A549, the mouse Lewis lung carcinoma cell line (LLC), and the human normal bronchial epithelial cell line Beas-2B were purchased from the ATCC (VA, USA). The A549, LLC, and Beas-2B cells were cultured in DMEM, and H1299 cells were cultured in RPMI-1640 medium (Gibco, USA) supplemented with 10% FBS (Gibco, USA), penicillin (100 IU/mL), and streptomycin (100 μg/mL).
2.3 Cell viability and colony formation assays
LLC cells (4000 cells/well), Beas-2B cells (3000 cells/well), H1299 cells (3000 cells/well), and A549 cells (3500 cells/well) were plated in 96-well plates and treated with SB for 24 h or 48 h. The cells were incubated with MTT for another 3 h. Then the absorbance was measured at 490 nm after MTT formazan crystals were dissolved. For colony formation assays, H1299 and A549 cells (500 cells/well) were plated in six-well plates and treated with SB for 7 days. Then the cell colonies were stained with crystal violet, and colonies with more than 50 cells were counted.
2.4 Cell-cycle analysis
After SB treatment for 24 h, H1299 and A549 cells were harvested and fixed in 70% ethanol at −20°C overnight. Subsequently, a Cell-cycle Detection kit (Meilunbio, China) was used to examine the cell cycle distribution with flow cytometry (BD Bioscience, USA) and ModFit LT 5.0 software.
2.5 Cellular thermal shift assays
Cellular thermal shift assays were performed to examine whether SB binds Skp2 in NSCLC cells [15]. Briefly, H1299 cells were harvested after treatment with DMSO or SB for 3 h. Then the cell suspensions in PBS were divided among seven PCR tubes and exposed to the indicated temperatures (42, 46, 50, 54, 58 and 62 °C) for 3 min in a PCR thermocycler. Subsequently, these cells were frozen repeatedly three times in liquid nitrogen, and the supernatants were collected after centrifugation (10000 rpm, 10 min) for western blotting.
2.6 Western blotting
The NSCLC cell lysates were harvested with RIPA lysis buffer (Beyotime) and subjected to SDS-PAGE analysis. Then total proteins were transferred to PVDF membranes and incubated with primary antibodies and horseradish-peroxidase-conjugated secondary antibodies. The signal of the indicated protein bands was detected with a ChemiDoc XRS system with Quantity One software (Bio-Rad, CA, USA), and the densitometry of these protein bands was analyzed in ImageJ software (NIH, USA). Each experiment was performed in triplicate.
2.7 Molecular docking
Molecular docking simulations were performed as previously described [6]. In brief, the three-dimensional structure of SB (Pubchem) and the crystal structure of the Skp2-Skp1 complex (Protein Data Bank ID: 2AST) were prepared with the OpenBabel and AutoDock Tools, respectively. Molecular docking was performed with AutoDock Vina.
2.8 Animal experiments
C57/BL6 mice (4 weeks of age) were purchased from the Laboratory Animal Services Center of Guangzhou University of Chinese Medicine. All animal experiments in this study were approved by the Animal Ethics Committee of Guangzhou University of Chinese Medicine. For orthotopic xenografts of lung cancer cells, the mice were injected with LLC cells (1.5 × 106) via the tail vein. The mice were randomly divided into four groups with ten mice each 7 days after cell injection: vehicle solvent (10% DMSO, 70% Cremophor EL/ethanol (3:1), and 20% saline, i.p.), SB 200 mg/kg (daily, i.p.), SB 400 mg/kg (daily, i.p.), and cisplatin 2 mg/kg (once every other day, i.p.) groups. SB was dissolved in vehicle solvent, and cisplatin was dissolved in saline. Because most mice die within 45 days after tumor xenografting, the survival time of the mice was assessed over 45 days. The mice were considered dead when they appeared moribund, and were then euthanized. At the end of the experiment, all surviving mice were sacrificed with CO2 and dissected to count the number of lung tumor nodules. Differences in the distribution of survival percentages among groups were analyzed with the Kaplan–Meier method with log-rank tests.
2.9 Drug adverse effects, animal tissue preparation, hematoxylin-eosin staining, and immunohistochemistry (IHC)
For examination of drug adverse effects, the body weight of each mouse was recorded from day 8 to day 29, because no mice died before day 29. Serum was collected from these mice and used to measure SCR, BUN, AST, and ALT levels with colorimetric assays according to the manufacturer’s instructions. For western blotting, the tumor nodules in the lung were dissected from six mice in each group and homogenized in RIPA lysis buffer to obtain total protein extract. For hematoxylin-eosin (H&E) staining, the lung tissues were fixed in 4% paraformaldehyde, dehydrated in a graded ethanol series (70%, 85%, 95%, and 100%), incubated in xylene, and embedded in paraffin. Tissue sections with 4 μm thickness were stained with H&E for histological examination. For IHC experiments, the tissue sections were prepared by deparaffinization and antigen retrieval, and were followed by incubation with antibody to Skp2 and staining with a DAB IHC kit (Solarbio, China).
2.10 Statistical analysis
Data processing and plotting were performed in GraphPad Prism 7.0 and SPSS software 17.0. All data are presented as mean ± standard error of the mean (SEM) from at least three independent experiments. Unpaired Student’s t tests and two-way analysis of variance were used to determine significant differences. *P < 0.05; **P < 0.01; ***P < 0.001.
3. RESULTS
3.1 SB inhibits the proliferation of NSCLC cells by inducing cell-cycle arrest
We first examined the cytotoxicity of SB against NSCLC and Beas-2B cells. SB dose-dependently inhibited the proliferation of A549, H1299, and LLC cells ( Figure 1a–c ); however, the drug did not show any cytotoxicity against Beas-2B cells ( Figure 1d ), in agreement with previous evidence indicating that SB is a safe drug without any toxicity [16]. SB also significantly inhibited colony formation in H1299 and A549 cells ( Figure 1e, f ). Because SB at concentrations less than 150 μmol/L did not induce clear apoptosis in NSCLC cells within 48 h, we examined whether the drug exhibited anti-NSCLC effects by inducing cell-cycle arrest. Similarly to findings in previous studies, SB significantly induced G1-phase cell-cycle arrest in NSCLC cells ( Figure 1g, h ). As expected, SB did not induce clear cell-cycle arrest in Beas-2B cells ( Figure 1i ). Furthermore, we tested the inhibitory effects of SB on NSCLC cell migration. Interestingly, SB also suppressed the migration and invasion of NSCLC cells ( Figure 2a–c ). These results revealed that SB exerts anticancer effects by inducing G1-phase arrest and suppressing cell migration.
SB inhibits proliferation and causes cell-cycle arrest in NSCLC cells.
(a–d) Cell viability assays using MTT reagent in A549, H1299, LLC, and Beas-2B cells after treatment with SB for 24 h and 48 h. (e, f) H1299 and A549 cells were exposed to SB for 7 days, and cell colonies were counted. (g–i) H1299, A549, and Beas-2B cells were treated with SB for 24 h, and the cell-cycle distribution was determined by flow cytometry. The results are expressed as mean ± SEM from three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001 versus control.
SB inhibits migration of NSCLC cells.
(a, b) Wound healing assays in H1299, A549, and LLC cells after SB treatment for 24 h. Cells were photographed under light microscopy, and the relative migration rate was quantified in ImageJ software. (c) Transwell assays in NSCLC cells after treatment with SB; cells were counted to calculate the relative invasion rate. All data are expressed as mean ± SEM from three independent experiments; *P < 0.05, **P < 0.01, and ***P < 0.001 versus control.
3.2 SB potentially inhibits the Skp2/p27 pathway in NSCLC
Considerable evidence has shown that the Skp2-SCF complex promotes cell-cycle progression and migration by regulating the expression of its substrates such as p27 and E-cadherin. Notably, SB markedly increased the protein level of p27 in a dose-dependent manner, and decreased Skp2 protein expression in NSCLC cells ( Figure 3a–c ). Moreover, SB clearly induced the accumulation of E-cadherin, another substrate of Skp2 ( Figure 3d ). To examine whether SB might serve as a potential Skp2 inhibitor, we used a potent specific Skp2 inhibitor, SZL P1-41, as a reference compound [5]. Both SB and SZL P1-41 markedly induced the accumulation of p27, and SB exerted a more profound inhibitory effect on Skp2 expression than SZL P1-41 in NSCLC cells ( Figure 3e, f ), thus demonstrating that SB may target Skp2, and subsequently inhibit the proliferation and migration of NSCLC cells.
SB potentially targets the Skp2/p27 pathway in NSCLC.
(a–c) Skp2 and p27 protein levels, detected by western blotting in H1299 (a), A549 (b), and LLC (c) cells after treatment with SB for 12 h and 24 h. (d) E-cadherin protein levels, detected by western blotting in A549 cells after SB treatment for 12 h and 24 h. (e, f) Skp2 and p27 protein levels, detected in A549 and H1299 cells by western blotting after treatment with SB and SZL (SZL P1-41) for 24 h. The relative protein levels were determined by densitometry analysis from three independent experiments. The results are expressed as mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 versus control group.
3.3 SB inhibits Skp2
Accumulating evidence indicates that Skp2 may serve as a valuable therapeutic target in various cancers [17, 18]. We evaluated the clinical value of Skp2 in NSCLC by using online databases. On the basis of the GEPIA database, Skp2 expression in lung squamous cell carcinoma (LUSC) tumor tissues was found to be significantly higher than that in normal tissues ( Figure 4a and Figure S1 ). Skp2 also had higher expression in NSCLC cells than in normal bronchial epithelial cells ( Figure S2 ). Moreover, Skp2 expression was negatively associated with NSCLC prognosis, on the basis of analysis of data from the Kaplan-Meier plotter database (kmplot, Affymetrix ID: 203625_s_at) ( Figure 4b ). Therefore, Skp2 may serve as a target for lung cancer treatment.
Skp2 is a potential target of SB.
(a) Gene expression of Skp2 in LUSC or LUAD and adjacent normal tissues from the GEPIA database. (b) Correlation between Skp2 expression level and the prognosis of patients with NSCLC, according to the kmplot database. (c) A549 was transiently transfected with FLAG-Skp1, and co-immunoprecipitation experiments were performed to detect the effect of SB on Skp2-Skp1 interaction. (d) H1299 cells were exposed to CHX with or without SB for the indicated time points, and Skp2 protein levels were detected by western blotting. (e) Cellular thermal shift assays of Skp2 in H1299 cells. (f) Molecular docking performed to predict the binding site of SB in Skp2 protein. Skp2 and SB are shown in cartoon and stick representation. (g) A549 cells were serum-starved in the absence or presence of SB for 24 h, stimulated with EGF or left unstimulated, and harvested for western blotting. LUSC, lung squamous cell carcinoma; LUAD, lung adenocarcinoma; T, tumor tissue; N, normal tissue. All data are expressed as mean ± SEM from three independent experiments; *P < 0.05 versus control.
Because Skp2 inhibitors decrease Skp2 protein stability by disrupting the integrity of the SCF complex, we examined whether SB might serve as a Skp2 inhibitor. We first performed co-immunoprecipitation experiments and found that SB clearly prevented Skp1-Skp2 protein interaction ( Figure 4c ). As expected, SB significantly decreased Skp2 protein stability after CHX co-treatment ( Figure 4d ). More importantly, cellular thermal shift assays showed that SB markedly increased Skp2 protein thermal stability at the indicated temperatures, thus suggesting that SB may directly bind Skp2 protein ( Figure 4e ). To predict the binding sites of SB in Skp2, we further used molecular docking based on the crystal structure of the Skp2-SCF complex. SB was predicted to form four H-bonds with Pro115, Tyr128, Ser132, and Lys145 near the first leucine-rich repeat and the C terminus of Skp2, with a binding affinity energy value of −6.8 kcal/mol ( Figure 4f ), thus further supporting the binding potential between SB and Skp2. Because Skp2-mediated ubiquitination is required for epidermal growth factor (EGF)-induced Akt phosphorylation [5], we examined the effects of SB on the signaling pathway downstream of Skp2. EGF stimulated Akt phosphorylation within 30 min in A549 cells, whereas SB markedly attenuated this process ( Figure 4g ). Therefore, these results demonstrated that SB may act as a potent Skp2 inhibitor.
3.4 SB has potent antitumor activity without toxicity
To evaluate the therapeutic potential of SB in NSCLC, we used highly metastatic LLC cells to establish orthotopic xenografts. Mouse models quickly develop visible tumors in the lung tissue, which usually lead to mortality within 40 days. LLC cells formed multiple large tumor nodules across the entire lung tissue; however, SB and cisplatin administration significantly decreased the number of tumor nodules on the lung surface ( Figure 5a–c ). Consistently with these findings, nine of ten mice in the control group died within 40 days, whereas 60% of mice in the SB high-dose group survived after 45 days ( Figure 5d ). Cisplatin treatment also significantly prolonged the survival of the mice ( Figure 5d ). As expected, SB markedly attenuated Skp2 expression while increasing p27 expression in tumor tissues ( Figure 5e ). IHC analysis also indicated that Skp2 expression in the lung tumor tissues clearly decreased after SB treatment ( Figure 5f ), thus suggesting that SB may also exert anticancer activity via targeting Skp2 in vivo.
SB suppresses tumor growth and metastasis in vivo.
(a–d) Representative images of lung metastases at the end of the animal experiments (a); counts of metastatic nodules of LLC cells on the lung surface (b); H&E staining of lung tissue sections (c); overall survival analysis of the mice (d). (e) Skp2 and p27 expression levels in mouse lung tumor tissues, detected by western blotting. (f) Skp2 protein expression in lung tumor tissues, detected by IHC. Black arrows: metastatic nodules on the lung surface. CP, cisplatin. Bar, 50 μm. The results are expressed as mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 versus control group.
We further evaluated the safety of SB in in vivo treatment. SB administration did not affect the body weights of the mice ( Figure 6a ). Furthermore, SB had no clear adverse effects toward the liver and kidneys, as evidenced by normal histological features in H&E-stained liver and kidney tissues, and the normal serum levels of SCR, BUN, AST, and ALT ( Figure 6b–e ). However, CDDP induced moderate body-weight loss ( Figure 6a ), caused dilatation of the renal tubules and hepatocellular enlargement ( Figure 6b, d ), and increased serum levels of SCR, BUN, and AST ( Figure 6c, e ). These results demonstrated that SB exerts promising anti-NSCLC activity without inducing clear adverse effects.
SB has a good safety profile in vivo.
(a) Body weights of the mice, recorded every 3 days after treatment with vehicle or SB. (b–e) At the end of the experiments, H&E staining of kidney (b) and liver (d) tissue was performed, and representative images are shown; the serum levels of BUN, SCR, ALT, and AST were measured (c, e). CP, cisplatin. The results are expressed as mean ± SEM. ***P < 0.001 versus control group.
4. DISSCUSION
Oncogenic E3 ligases, particularly the Skp2-SCF complex, are widely considered promising therapeutic targets in various cancer types [18]. Here, we demonstrated that SB may directly target Skp2 and inhibit Skp2-SCF E3 ligase in NSCLC cells, and consequently upregulate p27 protein and cell-cycle arrest. These results indicate novel mechanisms of action of SB in cancer treatment.
Skp2-SCF E3 ligase mediates the proteolytic ubiquitination of a variety of substrates such as p27, p21, E-cadherin, and FOXO1, as well as the non-proteolytic ubiquitination of AKT, LKB1, and Twist [19]. The oncogenic function of Skp2-SCF E3 ligase has prompted great interest in exploring novel Skp2 inhibitors. Skp2 interacts with Skp1 and its substrates through the F-box motif and the variable protein-protein interaction domain, respectively [4]. Two types of Skp2 inhibitors have been identified on the basis of the Skp2-SCF protein complex structure: the first type, such as compound C1/2, specifically disrupts the interaction between p27 and the Skp2/Cks1 complex [20]; the second type, such as compound #25, directly binds residues around the F-box motif, and inhibits Skp1-Skp2 interaction [5]. Therefore, the second type of Skp2 inhibitors, which disrupt SCF-complex integrity, completely inhibit SCF E3 ligase, thus exerting profound anticancer activities by inhibiting cell proliferation, migration, glycolysis, and even stemness. These inhibitors may provide a promising anticancer strategy. Herein, we found that SB binds Skp2 and disrupts Skp1-Skp2 interaction, thus leading to the accumulation of p27 and E-cadherin protein; therefore, SB may bind the F-box motif of Skp2. However, the potential binding residues of BA in Skp2 require further investigation. Furthermore, SB significantly decreases the protein level of Skp2 in LLC xenografts, thus inhibiting the growth and metastasis of NSCLC in vivo. These results demonstrate that SB displays favorable antiproliferative and antimetastatic potential by targeting Skp2-SCF E3 ligase. However, SB may also target other proteins beyond Skp2, because Skp2 expression was weakly negatively correlated with SB sensitivity in NSCLC cells (R = −0.577, Figure S3 ). The main therapeutic targets of SB and its selective properties against cancer cells require further investigation.
Silymarin is a relatively safe drug in humans and is well tolerated at a therapeutic dose of 700 mg three times per day for 24 weeks [16]. Therefore, silymarin has been used as a hepatoprotective medicine and food supplement [21]. SB, a major constituent of silymarin, is also non-cytotoxic and non-genotoxic at a concentration of 100 μM [16]. In our study, we demonstrated that SB at concentrations of 25–200 μM did not cause cytotoxicity in Beas-2B bronchial epithelial cells. In contrast, SB increased the cell viability of Beas-2B cells at concentrations from 50 μM to 150 μM. Moreover, SB did not show clear adverse effects in mice treated with SB at a high dose of 400 mg/kg for 1 month. To date, multiple clinical studies have been performed to test the effects of SB on liver diseases and other diseases [22, 23]. Novel therapeutic strategies that enhance the water solubility and bioavailability of SB may promote its clinical application [21].
In conclusion, we demonstrated that pharmacological Skp2 inhibition by SB stabilizes p27 and E-cadherin expression, and inhibits EGF-mediated Akt activation, thereby inhibiting NSCLC cell growth and metastasis. Our study uncovered novel anticancer mechanisms of SB, thus further supporting its safety and therapeutic potential in human cancers.
