The natural product Xanth, proteasome inhibitor MG132 and cycloheximide were purchased from Selleck Chemicals. Cell culture medium and supplements, including Dulbecco's modified Eagle's medium (DMEM), RPMI-1640 and fetal bovine serum (FBS), were obtained from Invitrogen (Thermo Fisher Scientific, Inc.). Human NSCLC cell lines, including H520, H358, H1299, H23, HCC827 and H1975, were purchased from the American Type Culture Collection (ATCC). The H520 is an EGFR null cell, HCC827 (E746-A750 deletion) and H1975 (L858R/T790M) are two NSCLC cells with EGFR activation mutation, while H1299 and H23 are wild EGFR harbor cells. As EGFR signaling plays a crucial role in NSCLC, cells with wild-type EGFR or activation mutant were selected in the present study. The immortalized human lung epithelial cell lines HBE and NL20 were obtained from Sigma and ATCC, respectively. All the cells were maintained in an incubator at 37°C in a humidified atmosphere with 5% CO₂ according to the ATCC protocols and subjected to a mycoplasma test every two months. The primary antibodies to cyclin D1, c-Jun, Jun B, Jun D, Fos B, FOS-related antigen 1 (Fra1), phosphorylated (p)-Fra1, c-Fos, p-ERK1/2 and β-actin were obtained from Cell Signaling Technology, Inc. The anti-Ki67 antibody for immunohistochemistry (IHC) was a product of Abcam. Antibody conjugates were visualized by chemiluminescence (Thermo Fisher Scientific, Inc.). The jetPEI (Qbiogene, Inc.) was used for transient transfection following the standard protocol.

The MTS assay was performed as previously described (12). In brief, NSCLC cells were seeded in 96-well plates (3,000 cells/well) and maintained for 24 h. The cells were treated with various doses of Xanth for 72 h. The cell viability was examined using the MTS Cell Proliferation Assay kit (cat. no. G3580; Promega Corp.) following the standard protocol.

The soft agar assay for colony formation was performed as described previously (13). In brief, 3 ml Eagle's basal medium containing 0.6% agar, 10% FBS and various doses of Xanth was used for the agar base in a 6-well plate. NSCLC cells were counted and suspended at a concentration of 8,000 cells/ml using Eagle's basal medium containing 0.3% agar, 10% FBS and Xanth. The re-suspended cells were overlaid into a 6-well plate with a 0.6% agar base and maintained for 2 weeks. The number of colonies was counted under a light microscope (Olympus).

Flow cytometric analysis was performed as described previously (14). In brief, the cells were treated with Xanth or DMSO (control) as indicated. Cells were suspended at a concentration of 1×10⁶ cells/ml with PBS. The propidium iodide staining buffer containing RNase was added to the cell suspension, followed by incubation at room temperature for 15 min in the dark. The cells were washed with PBS and analyzed with a FACSort flow cytometer (BD Biosciences).

The dual reporter assay was performed as described previously (15). In brief, the Renilla luciferase reporter construct pRL-SV40 was co-transfected with the pGL3-Basic control or the pGL3-AP-1 (cat. no. 40342; Addgene) construct which contain three canonical AP-1 binding sites (TGACTCA) into human NSCLC cells using Lipofectamine 2000 (Thermo Fisher). The transfected cells were treated with Xanth for another 24 h. Cell lysates were prepared using the Dual-Luciferase Reporter Assay kit (cat. no. E1910; Promega Corp.) and subjected to luciferase activity analysis. Renilla luciferase activity was used as an internal control for normalization.

NSCLC cells were treated with Xanth for 24 h, total RNA was extracted using the Absolutely RNA^® Purification Kits (Agilent). SYBR^®-Green Quantitative RT-qPCR Kit was used in RT-qPCR. Amplification cycles were 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 55°C for 60 sec). The RT-qPCR results were normalized to β-actin. Cyclin D1 primer sequences used were: Forward, TCTACACCGACAACTCCATCCG; reverse, TCTGGCATTTTGGAGAGGAAGTG.

All the surgical specimens were collected in accordance with an Institutional Review Board-approved protocol. NSCLC tissues and matched adjacent non-tumor tissues were collected from 35 patients who provided written informed consent at the Department of Pathology, Hunan Cancer Hospital of Central South University (Changsha, China). The in vivo experiments were approved by the Institutional Animal Care and Use Committee of Central South University (Changsha, China).

The whole-cell extract (WCE) was prepared using the commercial radioimmunoprecipitation assay buffer (cat. no. PI89901; Thermo Fisher Scientific, Inc.) and the protein concentration was determined by the bicinchoninic acid protein assay (cat. no. 23228; Thermo Fisher Scientific, Inc.). For EGF stimulation, NSCLC cells were starved with 0.1% fetal bovine serum in RPMI-1640 medium overnight and pretreated with Xanth for 2 h. EGF (50 ng/ml) was added to the cell culture medium and maintained for 1 h, whole cell lysates were collected for immunoblotting (IB) as described previously (16). In brief, WCE (20 µg) was boiled with loading buffer for 5 min at 95°C and subjected to SDS-PAGE, followed by electrotransfer to the polyvinylidene difluoride membrane. The membrane was blocked with 5% non-fat milk, followed by incubation with primary and secondary antibodies, respectively. The antibodies against cyclin D1 (cat. no. 55506, 1:1,000), p21 (cat. no. 2947, 1:1000), p27 (cat. no. 3686, 1:1,000), c-Jun (cat. no. 9165, 1:1,000), JunB (cat. no. 3753, 1:1,000), JunD (cat. no. 5000, 1:1,000), FosB (cat. no. 2251, 1:1,000), Fra1 (cat. no. 5281, 1:1,000), c-Fos (cat. no. 2250, 1:1,000), p-ERK1/2 (cat. no. 4370, 1:1,000), pP90RSK (cat. no. 8753,1:1,000), p-Fra1 (cat. no. 5841, 1:1,000), Ubiquitin (cat. no. 3936, 1:1,000), Ki67 (cat. no. 9027, 1:1,000), anti-mouse IgG HRP-linked antibody (cat. no. 7076, 1:10,000), and anti-rabbit IgG, HRP-linked antibody (cat. no. 70764, 1:10,000), were purchased from Cell Signaling Technology (Danvers, MA). The target protein was visualized by chemiluminescence.

CRISPR-Cas9-mediated gene knockout was performed as previously described (17). In brief, single-guide (sg)RNAs (#1, GTTCGTGGCCTCTAAGATGA; #2, GAAGCGTGTGAGGCGGTAGT) targeting cyclin D1 were used for the construction of stable cell lines. In brief, NSCLC cells were transfected with cyclin D1 sgRNA and selected by 1 µg/ml puromycin for three weeks. Single colonies were chosen for further study.

The in vivo experiments were approved by the Institutional Animal Care and Use Committee of Central South University (Changsha, China). Mice were kept in colony cages with free access to food and tap water and in standardized housing conditions (natural 12 h light-dark cycle, temperature of 23±1°C, relative humidity of 55±5%). The proper care and use of experimental animals, including efforts to minimize suffering and distress, use of analgesics or anaesthetics, was based on the Guide for the Care and Use of Laboratory Animals (National Academies Press, Washington, DC). The xenograft model was constructed by subcutaneous injection of HCC827 cells (4×10⁶) into the right flank of 6-week-old athymic nude mice (n=5). Compound treatment was initiated when the tumor reached an average volume of 100 mm³. The compound-treated group was administered Xanth (10 mg/kg) every two days by i.p. injection. The health and behaviour of mice were monitored every two days. The tumor volume was determined according to the following formula: Lengthxwidth²/2. The tumor-bearing mouse was euthanized by CO₂ when the tumor volume reached 700 mm³ (32 days). The fill rate of carbon dioxide is 30% of the chamber volume per minute (3 liter/min), and the duration time is 5 min. Death was further confirmed by cervical dislocation.

Tumor tissues from clinical samples and mouse xenograft tumors were fixed and subjected to IHC analysis as previously described (18). In brief, the tissue slides were deparaffinized and rehydrated by consecutive incubation with xylene and ethanol, followed by submerging in sodium citrate buffer (10 mM, pH 6.0) and boiling for 10 min for antigen retrieval. The activity of endogenous peroxidase was quenched by incubation with 3% H₂O₂ in methanol for 10 min. Tissue slides were washed with PBS and blocked with 50% goat serum albumin. After incubation with the primary and secondary antibodies, the target protein was visualized using 3,3′-diaminobenzidine substrate and samples were counterstained with hematoxylin.

Statistical analysis was performed using GraphPad Prism 5.0 (GraphPad, Inc.). Quantitative data are expressed as the mean ± standard deviation. The significance of differences between groups was evaluated using the Student's t-test or One-way ANOVA. Dunnett's method was used to compare treatment groups to a control group, and Tukey's method was used to compare each group with every other group. P<0.05 was considered to indicate a statistically significant difference.

To determine the oncogenic function of cyclin D1 in NSCLC, the protein levels of cyclin D1 were first examined in tumor tissues. As shown in Fig. 1A, cyclin D1 was significantly upregulated in human NSCLC tissues compared with that in the paired non-tumor adjacent tissues. The IB results suggested that cyclin D1 was overexpressed in all human NSCLC cancer cell lines assessed: H520, H358, H1299, H23, HCC827 and H1975 (Fig. 1B). Furthermore, stable cyclin D1-knockout cell lines were constructed from HCC827 and H1975 cells. The MTS data revealed that the viability of cells with stable expression of sgCyclin D1 was significantly reduced (Fig. 1C). To determine the role of cyclin D1 in anchorage-independent growth of NSCLC cells, colony formation was examined using the soft agar assay. The results suggested that knockout of cyclin D1 inhibited colony formation of H1975 and HCC827 cells, as the colony number was decreased >60% when compared to that of cyclin D1-proficient cells (Fig. 1D). The results of the in vivo tumor growth experiment indicated that depletion of cyclin D1 inhibited in vivo tumor development in the xenograft mouse model. The sgCyclin D1-expressing HCC827 ×enograft tumors exhibited a reduced tumor volume (Fig. 1E) and a smaller tumor size (Fig. 1F and G). These results suggested that cyclin D1 is highly expressed in NSCLC tissues and cell lines, while knockout of cyclin D1 decreased the tumorigenic properties of NSCLC cells.

Accumulating evidence indicates that the natural product Xanth (Fig. 2A) has potent anti-tumor activity against multiple human cancer cell types (19). However, the anti-tumor effect of Xanth on NSCLC cells and the underlying mechanisms have remained largely elusive. The present results indicated that Xanth exhibited only slight cytotoxic effects on the immortalized lung epithelial cell lines HBE and NL20, as the cell viability was not significantly reduced after treatment with Xanth (Fig. 2B). However, the viability of Xanth-treated NSCLC cell lines, including HCC827, H1975 and H23, was decreased by Xanth in a dose-dependent manner (Fig. 2C). Treatment with Xanth for 72 h reduced the cell viability >60% in all treated NSCLC cell lines. Furthermore, the results of the soft agar colony formation assay suggested that the anchorage-independent cell growth of NSCLC cells was significantly reduced with Xanth treatment (Fig. 2D). The Xanth exerted its inhibitory effect on colony formation in a dose-dependent manner. The results revealed that treatment with 10 µM Xanth decreased the colony number >95% in HCC827, H1975 and H23 cells (Fig. 2D). These in vitro results indicated that Xanth inhibits NSCLC cells, but not the immortalized lung epithelial cells, in a dose-dependent manner.

As Xanth exhibited a significant inhibitory effect on NSCLC cell growth, it was then determined whether Xanth affected cell cycle-associated proteins. The IB data revealed that Xanth reduced the protein levels of cyclin D1 in HCC827, H1975 and H23 cells (Fig. 3A). Furthermore, flow cytometric analysis revealed that Xanth induced cell cycle arrest at the G0/G1 phase in HCC827 cells (Fig. 3B). Consistently with this, the IB results revealed that Xanth increased the protein levels of p21 and p27 (Fig. 3C). To investigate the underlying mechanisms of how Xanth reduced cyclin D1 at the protein level, Xanth-treated NSCLC cells were subjected to RT-qPCR analysis of cyclin D1 expression. As shown in Fig. 3D, Xanth decreased the mRNA levels of cyclin D1 in a dose-dependent manner, indicating that Xanth suppressed cyclin D1 transcription. The AP-1 protein is one of the most important transcription factors required for cyclin D1 transcription (20). The present results suggested that Xanth slightly decreased the protein levels of JunD but robustly reduced the expression of Fra1 in HCC827 and H1975 (Fig. 3E) cells. Furthermore, the reporter assay indicated that Xanth significantly reduced the luciferase activity of AP-1 (Fig. 3F and G). These results suggested that Fra1 is a critical transcription factor for cyclin D1 expression in NSCLC cells. Indeed, ectopic overexpression of Fra1 in HCC827 (Fig. 3H) and H1975 (Fig. 3I) cells compromised Xanth-induced cyclin D1 downregulation. Consistently, the RT-qPCR results revealed that overexpression of Fra1 restored cyclin D1 mRNA levels in Xanth-treated NSCLC cells (Fig. 3J). Overall, these results suggest that Xanth reduces the protein levels of cyclin D1 in a Fra1-dependent manner in NSCLC cells.

To determine the underlying mechanisms of how Xanth decreases the protein levels of Fra1, the signaling transduction in Xanth-treated NSCLC cells was examined. Of note, the IB data suggested that Xanth inhibited the activation of ERK1/2 signaling in a dose-dependent manner. The phosphorylation of ERK1/2 and the downstream target kinase P90RSK was robustly decreased after Xanth treatment (Fig. 4A). Furthermore, Xanth attenuated EGF-induced ERK1/2 and P90RSK phosphorylation in NSCLC cells (Fig. 4B). Treatment with the ERK1/2 signaling inhibitor PD98059 suppressed the protein levels of Fra1 and cyclin D1 in HCC827, H1975 and H23 cells. In addition, the phosphorylation of Fra1 on S265, a phosphorylation residue that is catalyzed by ERK1/2, was inhibited in response to PD98059 treatment (Fig. 4C). A previous study suggested that Fra1 S265 phosphorylation promotes Fra1 stability and inhibits protein degradation (21). Indeed, the IB results suggested that Xanth reduced the half-life of Fra1 from nearly 2 h to 40 min (Fig. 4D). Of note, the proteasome inhibitor MG132 restored Fra1 protein levels in Xanth-treated NSCLC cells (Fig. 4E), indicating that Xanth promoted the ubiquitination-dependent protein degradation of Fra1 in NSCLC cells. The ubiquitination analysis revealed that Xanth promoted Fra1 ubiquitination in HCC827 cells (Fig. 4F). The S265D mutant was then constructed, in which Ser265 was mutated to Asp to mimic the constitutive phosphorylation of Fra1. The results indicated that Xanth promoted the ubiquitination of wild-type Fra1 but not that of the Fra1 S265D mutant (Fig. 4G). Overall, these results suggested that inhibition of ERK1/2-mediated Fra1 S265 phosphorylation is required for Xanth-induced Fra1 destruction in NSCLC cells.

To determine the in vivo anti-tumor effect of Xanth, it was tested in a xenograft mouse model established using HCC827 cells. The results revealed that the average tumor volume of the vehicle-treated group of HCC827-derived tumors was 522±121 mm³. By contrast, treatment with Xanth significantly suppressed xenograft tumor growth, as the tumor volume was only 209±58 mm³ (Fig. 5A). Furthermore, Xanth reduced the weight of the tumors >50% when compared with that of the vehicle-treated group (Fig. 5B and C). Xanth did not cause a significant body weight loss in the animals (Fig. 5D), suggesting that Xanth was well-tolerated in them. The IHC staining result indicates that Xanth reduced the population of Ki67-positive cells. Similarly, the protein levels of Fra1 and cyclin D1 were significantly downregulated in Xanth-treated xenograft tumors (Fig. 5E and F), which was consistent with the in vitro results suggesting that Xanth reduced the protein levels of Fra1 and cyclin D1. In conclusion, it was indicated that Xanth inhibited the in vivo tumor growth of NSCLC cells in a xenograft mouse model.