Degalactotigonin, a Natural Compound from Solanum nigrum L., Inhibits Growth and Metastasis of Osteosarcoma through GSK3β Inactivation–Mediated Repression of the Hedgehog/Gli1 Pathway

Osteosarcoma is the most common primary malignant bone tumor in childhood and adolescence. Current treatment strategies, which consist of multiagent chemotherapy and aggressive surgery, have significantly improved the 5-year survival rate of patients with osteosarcoma from 10% to 70% over the past 30 years (1). However, despite some advances in the treatment of osteosarcoma, therapies have not changed significantly in the past few years. The survival rate for patients with metastatic osteosarcoma is still less than 20% (2, 3). Therefore, further research is warranted to identify effective agents and develop new therapeutic strategies with less severe side effects for the treatment of this deadly disease.

Over the past three decades, agents derived from natural sources have gained considerable attention from researchers and clinicians because of their safety, efficacy, and immediate availability (4). They have long been used in traditional Chinese medicine and show multiple biological activities, including anticancer, anti-inflammatory, and neuroprotective properties (5, 6). One of the most successful natural agents is artemisinin (qinghaosu), with an antimalarial effect, which is considered a true gift from old Chinese medicine by Youyou Tu (7). Because of the great contribution of artemisinin, the founder, Youyou Tu, received the 2015 Nobel Prize in Physiology or Medicine. However, natural agents have not been popularly accepted, primarily due to their poorly defined molecular mechanisms. One such little investigated agent is degalactotigonin (DGT), which is isolated from Solanum nigrum L. The extractions of Solanum nigrum L have anticancer properties in breast cancer, colon cancer, and prostate cancer while leaving normal cells unharmed (8–11). Thus far, one study has shown growth-inhibitory effects of DGT on a cervical cell line (HeLa) and contraceptive potential but provided no detailed insights into the underlying molecular mechanisms (12). The goal of this study was to determine the range of growth and metastasis repression of DGT against osteosarcoma cells both in vitro and in vivo, and to investigate how the drug mediates these activities.

The Hedgehog (HH) signaling pathway is crucially involved in the development of several cancers, such as glioma, colon cancer, gastric cancer, multiple myeloma, myeloid leukemia, and rhabdomyosarcomas (13–16). The HH signaling is mainly dependent on Gli transcription factor family (Gli1, Gli2, and Gli3), which are downstream effectors. Specifically, Gli1 is the principal transcriptional effector that regulates gene expression in response to HH signaling activation. Moreover, overexpression of Gli1 is a credible indicator of poor prognosis in most solid malignancies, irrespective of intracranial tumors (17). However, its clinical significance and biological function in osteosarcoma growth and metastasis remains unclear.

Here, we investigated the potential of DGT to inhibit growth and metastasis at the cellular level and in animal models to demonstrate that DGT may be a promising natural agent for treating patients with osteosarcoma. Moreover, we propose that one of the mechanisms of action of DGT is through inducing GSK3β inactivation–mediated repression of the HH/Gli1 pathway, which is important in the development of osteosarcoma.

All cell lines were obtained from ATCC except U2OS/MTX (a methotrexate-resistant derivative of the U2OS human osteosarcoma cell line was provided by Dr. M. Serra, Instituti Ortopedici Rizzoli, Bologna, Italy), ZOS, and ZOS-M (from human osteosarcoma patient with primary tumor and metastasis). All of the osteosarcoma cell lines were grown in DMEM (Invitrogen) with 10% FBS (Invitrogen) at 37°C and 5% CO₂. The hMSCs were gifted by Dr. Fangang Meng (The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China) and cultured as shown (18). All cell lines used in this study were authenticated using short tandem repeat profiling when this project was initiated, and the cells have not been in culture for more than 1 month.

DGT (MW:1035.1; >98% purity) was extracted from Solanum nigrum L. by Q. Jia and purchased from Sigma Aldrich (catalog no.: 39941-51-0). Antibodies against P21 (#2947S), CyclinD1 (#2978S), p-H2AX (#9718S), H2AX (#2595S), Gli1 (#2643S), Akt (#2920S), p-Akt(#4060S), ERK (#4348S), p-ERK (#4370S), GSK-3β (#9315S), phospho-Ser9-GSK-3β (#9323S), PARP (#9532S), and cleaved PARP (#5625S) were purchased from Cell Signaling Technology. Antibodies against CyclinA1 (sc-53233), CyclinB (sc-166152), GAPDH (sc-32233), and tubulin (sc-73242) were purchased from Santa Cruz Biotechnology. Antibodies against Gli1 (ab217326), smo (ab38686), PTCH1 (ab53715), and SHH (ab53281) were purchased from Abcam. The reagents for IHC analyses were obtained from Dako Cytomation. The human Gli1 plasmid was purchased from youbia.

Osteosarcoma cells were seeded in 96-well plates at a density of 4,000 cells per well. They were treated with different concentrations of DGT for the indicated times, and the cell viability was measured by MTT assay as described previously (19).

Briefly, osteosarcoma cells were plated in triplicate at 500 cells per well in 12-well plates in Fig. 1C with or without DGT and 2,000 cells per well in 6-well plates for in Fig. 6D. All the cells were cultured for 12 days. Then, cell clones were washed three times with PBS, fixed in methanol for 10 minutes, and dyed with crystal violet for 10 minutes at room temperature. Afterward, the dye was washed off and colonies that contained more than 50 cells were counted.

Osteosarcoma cells were treated with DGT for 48 hours and were subsequently collected as described previously (20). Cell-cycle analysis was performed by propidium iodide staining (Sigma-Aldrich) for DNA content and followed by flow cytometric analysis. The cells were stained with both propidium iodide and Annexin V (BD Biosciences) and assayed on a LSRII flow cytometer (BD Biosciences) for the apoptosis assay. All flow cytometry data were analyzed using FlowJo software (Tree Star).

Cells were seeded at 50% confluency in 6-well plates, and after overnight incubation, the cells were treated with vehicle or DGT for 48 hours. Then, the standard procedures as previously described were performed for Hoechst 33258 staining and caspase-3 activity assay (19).

To measure the p-H2AX expression in control and DGT-treated osteosarcoma cells, an immunocytochemical analysis was performed. Briefly, U2OS cells were plated on culture slides (Costar) and treated with or without DGT for 48 hours. Then, samples were fixed using 4% paraformaldehyde solution for 15 minutes at room temperature and extracted with buffer containing 0.5% Triton X-100 for 5 minutes. The cells were then incubated with primary antibodies overnight. After three washes in PBS, the samples were incubated with secondary antibody at room temperature for 1 hour. Cells were then counterstained with DAPI to visualize nuclear DNA and examined using an Olympus confocal imaging system (Olympus FV100).

Western blotting was carried out as described earlier (19). Briefly, the cells were lysed in RIPA buffer containing protease inhibitor and phosphatase inhibitor cocktails (Thermo Fisher Scientific). The nuclear protein was isolated according to the protocol provided by the Nuclear Protein Extraction Kit (Thermo Fisher Scientific). Then, equal amounts of protein was resolved on 10% SDS-PAGE and transferred to a PVDF membrane (Millipore). Membranes were incubated with primary antibody overnight at 4°C. Membranes were washed with TBST and incubated with horseradish peroxidase–conjugated secondary antibody. Proteins were visualized using ECL detection reagents (Beyotime Co.).

Small interfering RNAs against the following genes were synthesized by RiboBio.

GSK3β (sense 5′-CUCAAGAACUGUCAAGUAATT-3′, antisense: 5′-UUACUUGACAGUUCUUGAGTT-3′); AKT, (sense 5′-UGCCCUUCUACAACCAGGATT-3′, antisense 5′-UCCUGGUUGUAGAAGGGCATT-3′); β-catenin (sense, 5′-GCAGUUGUAAACUUGAUUATT-3′, antisense, 5′-UAAUCAAGUUUACAACUGCTT-3′). The shRNA targeting Gli1 was constructed as below. Briefly, the targeting sequence (Gli1 shRNA #1, CTTTGATCCTTACCTCCCA; Gli1 shRNA#2, AGCTCCAGTGAACACATAT) was inserted into pLKO.1-puro gifted by Tiebang Kang (Sun Yat-Sen University Cancer Center, Guangzhou, China) following the protocol provided by Addgene. Targeting siRNAs or shRNA were transfected into osteosarcoma cells at 30% to 50% confluence in 6-well plates using Lipofectamine 2000 following the instructions (Invitrogen).

Total cellular RNA was extracted using the TRIzol reagent (Invitrogen) according to the manufacturer's instructions. RNA was reverse transcribed to produce cDNA and real-time PCR amplification was performed as described previously (19). The primer sequences we used are shown in Supplementary Table S1.

Osteosarcoma cells were seeded in 24-well plate and cotransfected with 200 ng Gli luciferase reporter and 5 ng pRL-TK Renilla luciferase construct (Promega) per well using Lipofectamine 2000 (Invitrogen). After 24 hours, the cells were treated with DGT or siRNA (or Licl). Then, the cells were analyzed after an additional 48 hours according to the Dual-Luciferase Assay System protocol (Promega).

The migration and invasion of osteosarcoma cells were examined using 24-well Boyden chambers with 8-mm inserts coated without (migration) or with Matrigel (invasion) as described previously (21). Generally, the osteosarcoma cells were treated with vehicle or DGT, and then, the cells were collected for migration and invasion assays. A total of 4 × 10⁴ cells per well were plated on the inserts and cultured at 37°C in the upper chambers without serum. After 24 hours, the cells on the lower surface of the filter were fixed, stained, and examined under a microscope. The average number of migrated cells from five random optical fields (×100 magnification) and triplicate filters was determined.

IHC staining was performed as described previously (21). Briefly, the primary antibody was diluted at a ratio of 1:50, and the slides were incubated with primary antibody overnight at 4°C. Detection of the primary antibody was achieved by incubation with a secondary antibody (Envision; Dako) for 1 hour at room temperature. The slides were stained with 3,3-diaminobenzidine after washing in PBS again. Studies of patient sample were approved by the Institutional Review Board of the Sun Yat-Sen University.

The proteome profiler array (R&D Systems) was performed according to the manufacturer's protocol. Briefly, the cell lysates were incubated with activated array membranes overnight at 4°C. Each array membrane was washed 3 times with 1× wash buffer and incubated with the diluted antibody cocktail for 2 hours at room temperature. The array membranes were then washed 3 times with 1× wash buffer and further incubated with streptavidin–horseradish peroxidase for 30 minutes at room temperature. Each array membrane was then washed 3 times with 1× wash buffer, and finally, the membranes were exposed to X-ray film following chemiluminescent detection.

The animal experiments were approved by the Institutional Review Board of Sun Yat-Sen University. Athymic nude (nu/nu) mice, 5 to 6 weeks of age, were purchased from Shanghai Slac Laboratory Animal Company Limited. U2OS/MTX cells (1 × 10⁶ cells in 200 μL of PBS) were injected subcutaneously near the scapula of the nude mice. After 8 days, the mice were randomly separated into three groups. For the osteosarcoma xenograft growth of orthotopic animal model, ZOS cells were used as described previously (22). After 14 days, the mice were randomly separated into the three groups. In both of the models, the first group, the control, was treated with vehicle (DMSO in water). The other two groups were treated with DGT dissolved in the vehicle (either 150 or 300 mg/kg) every day. All the groups received the drug through intraperitoneal injection. The resulting tumors were measured with a caliper every 4 days. The U2OS/MTX tumor volume was calculated using the formula V = 1/2 (width² × length), and the ZOS tumor volume was calculated using the formula V = 4/3π [1/4 (D1 + D2)]². At the end of the experiment, the animals were sacrificed via cervical dislocation, and the tumor weights were measured after careful resection.

For the osteosarcoma metastasis model, mice were injected with 2 × 10⁶ 143B cells (in 100 μL of PBS) into the lateral tail vein (6 mice/group). The mice were treated with DGT (300 mg/kg) every day and monitored 3 times per week for evidence of morbidity associated with pulmonary metastases. After 6 weeks, the mice were sacrificed, and the lungs were harvested, fixed in 4% paraformaldehyde, and embedded in paraffin. To quantify the number of pulmonary metastatic lesions, sequential 3-μm thick sections of whole lungs were obtained. The sections were stained with H&E to identify the metastases using light microscopy as described previously (23).

All data are presented as statistical plots generated using GraphPad Prism 5. The differences between two groups were determined using two-tailed t tests. The differences among three or more groups were determined using one-way ANOVA followed by two-tailed t tests. Kaplan–Meier analysis of tumor patients and the log-rank test were performed for comparison of the survival curves according to the Gli1 level. P < 0.05 was considered statistically significant.

DGT (molecular weight, 1,035.1 Da), the major bioactive constituent of Solanum nigrum L., and its chemical structure is shown in Fig. 1A. To explore the effect of DGT on osteosarcoma cells, 9 cell lines were treated with various doses of DGT for 72 hours. The calculated IC₅₀ values ranged from 12.91 to 31.46 μmol/L (Fig. 1B). However, the IC₅₀s of hfob1.19 (239.4 μmol/L) and hMSCs (536.8 and 537.2 μmol/L) are much more than osteosarcoma cell lines (Supplementary Fig. S1). The antiproliferative activity of DGT was further evaluated with a colony formation assay. Colony number and colony size were reduced in U2OS, HOS, and MG63 cells treated with DGT compared with controls (Fig. 1C). Moreover, DGT suppressed the proliferation of U2OS/MTX, HOS, ZOS, and MG63 cells in a dose- and time-dependent manner. At concentration as low as 5 μmol/L, DGT inhibited cell proliferation after 3 days of treatment (Fig. 1D). Significant inhibition of cell growth was observed after 1 day at a concentration of 40 μmol/L. These results indicate that DGT exhibits antiproliferative activity against osteosarcoma cells in vitro.

We examined the effect of DGT on cell-cycle progression in U2OS and MG63 cells using propidium iodide staining. The percentage of U2OS and MG63 cells in G₂–M increased by 8.6% and 15%, respectively, and the percentage of cells in G₀–G₁ decreased by 15.17% and 17.52%, respectively (Fig. 2A). To further investigate the molecules affected by DGT, we examined the levels of CyclinD1, which is related with G₀–G₁ to S-phase transition, and P21, which is a potent inhibitor of cell-cycle progression. The results showed that the expression of CyclinD1 was decreased and P21 was increased at RNA and protein levels, but DGT had little effect on CyclinA1 and CyclinB (Fig. 2B and C). These results indicate that DGT can induce cell-cycle G₀–G₁ decrease and G₂–M arrest in osteosarcoma cell lines. Previous studies have demonstrated that a G₂–M arrest is frequently the result of DNA damage, and phospho-H2AX is a marker of DNA double-stranded breaks and can therefore be used to monitor DNA repair (24). Therefore, we performed a phospho-H2A.X immunostaining analysis of U2OS cells and found that the proportion of phospho-H2A.X⁺ cells and the distribution of phospho-H2A.X⁺ foci per nucleus were obviously increased (Fig. 2D). These results suggest that DGT treatment not only affected cell-cycle progression but also induced DNA damage in osteosarcoma cells.

Increased phospho-H2A.X expression in the nucleus indicates that cells repair the damaged DNA, which usually results in cell death or apoptosis (25). Therefore, we sought to determine whether treatment with DGT could induce osteosarcoma cell apoptosis. U2OS/MTX and ZOS cells were treated with DGT for 48 hours at different concentrations, and Western blotting was used to assess the expression of cleaved PARP, a marker of apoptosis, which was increased after exposure to DGT (Fig. 3A). As shown in Fig. 3B, caspase-3 activity assays were also performed by extracting cellular protein from DGT-treated U2OS/MTX and ZOS cells, and these assays indicated that caspase-3 activity was significantly increased (e.g., caspase-3 activity in U2OS/MTX and ZOS cells treated for 48 hours with 40 μmol/L was approximately 3.44 and 4.12-fold greater compared with untreated control cells). Osteosarcoma cell apoptosis induced by DGT was further analyzed with Hoechst 33258 staining and flow cytometry assays to detect the morphologic changes and analyze the proportion of apoptotic cells. Brighter blue staining and more morphologic changes were found in nuclear chromatin of MG63, U2OS/MTX, and ZOS cells following the DGT treatments for 48 hours. Typical morphologic characteristics of apoptosis, such as reduction in nuclear size, cell pyknosis, and chromatin condensation, were more easily observed in DGT-treated cells than in cells treated with vehicle (Fig. 3C). Annexin V/PI staining of these cells demonstrated a significant increase in apoptotic cells following DGT treatment (Fig. 3D). Conclusively, all the above results indicated that the cytotoxicity of DGT occurs through induction of apoptosis in osteosarcoma cells.

Subsequent to finding that DGT inhibits proliferation and induces apoptosis of osteosarcoma cells in vitro, we further evaluated its effectiveness in inhibiting the growth of osteosarcoma in nude mice. First, U2OS/MTX cells were subcutaneously injected into nude mice until a tumor volume of approximately 200 mm³ was reached. The mice were randomly separated into three groups (control, dose 1, and dose 2). The dose 1 group received 150 mg/kg of DGT every day, and the dose 2 group received 300 mg/kg of DGT every day. At the termination of the study, statistical analysis of tumor growth revealed a significant reduction in tumor size in mice treated with DGT. The mean volumes of the tumors were 1,273.78 mm³ for the control group, 802.24 mm³ for the dose 1 group (P < 0.01), and 540.03 mm³ for the dose 2 group (P < 0.001; Fig. 4A). The average tumor weights were 1.248 g for the control group, 0.754 g for the dose 1 group (P < 0.001), and 0.597 g for the dose 2 group (P < 0.001; Fig. 4B). Next, we investigated whether DGT could inhibit osteosarcoma growth using an in vivo orthotopic osteosarcoma model, as described in the Materials and Methods section. As shown in Fig. 4C, notably, the growth of DGT-treated groups was much slower than that of the control group, consistent with the tumor growth curve. At the 42nd day, the mean volumes of the tumors were 928.37 mm³ for the control group, 731.30 mm³ for the dose 1 group (P < 0.05), and 549.41 mm³ for the dose 2 group (P < 0.01). The data also show that there is a concentration-dependent effect of DGT on tumor volume. In addition, in both of the two models, the average body weights of the mice were not significantly different between the DGT-treated groups and the control group, and no obvious pathologic changes were observed in vital organs (heart, liver, kidney, and lung) as detected by hematoxylin and eosin (H&E) staining (Supplementary Figs. S2 and S3). H&E staining also showed that bone destruction around the tibia was more obvious in the control group (Fig. 4D). We found that DGT could induce apoptosis in osteosarcoma cells in vitro. Here, we analyzed the level of apoptosis in tumor tissue using a TUNEL assay, and the results showed that DGT could induce apoptosis in vivo (Fig. 4E). Collectively, these results demonstrate that DGT possesses antitumor properties and can induce apoptosis in human osteosarcoma cells in vivo.

Tumor growth, metastasis, and invasion are key processes during tumor development and progression, and we found that DGT could inhibit the growth of osteosarcoma. We next wanted to assess whether DGT could reduce the migratory and invasive properties of osteosarcoma cells. Notably, DGT treatment of 143B cells significantly reduced cell invasion (Fig. 5A and B) and motility through Matrigel (Fig. 5C and D). Importantly, suppression of cell migration and invasion was not due to reduced overall cell number, as equal numbers of cells were reseeded into the wells after the pretreatment period, and migration or invasion was assessed within 24 hours. To investigate the ability of DGT to prevent osteosarcoma lung metastasis in vivo, we injected 143B cells into the tail vein of nude mice as described previously (26). In addition, the mice were randomly separated into two groups after 2 weeks; the control group received vehicle every day, and the treatment group received 300 mg/kg of DGT every day. After 5 weeks, the lungs were harvested, and micrometastases were analyzed. Consistent with the in vitro results, DGT treatment dramatically diminished the occurrence of osteosarcoma xenograft metastasis to the lungs, as indicated by the number of metastatic nodules in the lung and the wet lung weight (P < 0.001) (Fig. 5E and F; Supplementary Fig. S4). Collectively, these results demonstrate that DGT inhibits osteosarcoma metastasis in vitro and in vivo.

The protein Gli1 functions as a downstream transcription factor of HH signaling and plays a pivotal role in cellular growth and metastasis of many types of tumors (27–29). However, the function of Gli1 in osteosarcoma remains unclear. First, to investigate whether Gli1 is overexpressed in osteosarcoma cell lines, we compared the protein expression in osteosarcoma cell lines and hfob1.19 cells. We found that most of the osteosarcoma cell lines have higher expression of Gli1, but other related genes of HH signaling, such as PTCH, Smo, and SHH, were not overexpressed (Fig. 6A). The increased expression of Gli1 has been reported to be clinically correlated with unfavorable overall prognosis in most solid malignancies (28, 30–32). We also determined whether Gli1 expression was associated with the clinical outcome of patients with osteosarcoma. In total, 71 patient samples were collected for IHC using Gli1 antibody. Among them, Gli1 protein was strongly expressed in 66.2% of the samples (Supplementary Fig. S5A and S5B), and Kaplan–Meier survival analysis indicated that patients with high Gli1 expression levels had worse overall survival than those with low expression levels (Supplementary Fig. S5C).

We subsequently investigated whether Gli1 affected the growth and metastasis of osteosarcoma. We stably knocked down Gli1 (using two different shRNAs against Gli1) in osteosarcoma cell lines and demonstrated that Gli1 silencing in the cells reduced proliferation, sphere formation, migration, and invasion (Fig. 6B–F). In addition, inhibition of Gli1 decreased the expression of HH target genes, including Gli1, Sox2, OCT4, PTCH1, and Myc, and increased the expression of P21 (Fig. 6G). These data suggested that Gli1 expression is both necessary and sufficient to promote osteosarcoma cell proliferation, migration, and invasion.

To determine whether DGT prevents Gli1 transcription and expression in osteosarcoma cells, Western blotting and a Gli1 luciferase activity assay were performed. The protein level of Gli1 was decreased in a dose-dependent manner after treatment with DGT in U2OS and ZOS cells (Fig. 6H). The Gli1 luciferase reporter activity was decreased by 40% after treatment with DGT in osteosarcoma cells (Fig. 6I). In addition, the Gli1 target genes, including Gli1, PTCH1, and N-myc, were decreased in dose-dependent manner after treatment with DGT in U2OS cells (Supplementary Fig. S6). Next, we explored whether the inhibitory effect of DGT on osteosarcoma metastasis and growth depends on HH/Gli1 signaling. Osteosarcoma cells were transfected with Gli1 plasmid or empty vector. These cells were then treated with or without DGT, and growth and metastasis were analyzed. As shown in Fig. 6J, treatment with DGT led to downregulation of endogenous and exogenous Gli1 expression, and overexpression of Gli1 enhanced the metastatic ability of ZOS cells, which was partly recovered when combined with DGT treatment. The abovementioned findings indicated that Gli1 plays a key role in DGT-induced growth and metastasis, which is a good prognosis factor for osteosarcoma.

It has been demonstrated that many kinases play key roles in cancer cell survival and growth, and inhibiting these kinases can reduce cancer growth and metastasis. To examine the effects of DGT on intracellular signaling, we screened the phosphorylation status of multiple cellular kinases in U2OS cells treated with vehicle or DGT using a human phospho-kinase antibody array. As shown in Fig. 7A, the phosphorylation levels of multiple kinases were altered after treatment with DGT, including Akt, ERK, GSK3β, and mTOR, whose expression changed 3 to 15-fold. To confirm the result, we tested the changes in these kinases using Western blotting in osteosarcoma cells, including U2OS, MG63, HOS, ZOS, and 143B cell lines, treated with 20 μmol/L DGT for 48 hours. We found that the expression of phospho-Akt and phospho-ERK was decreased, and phospho-GSK3β was increased (Fig. 7B). To confirm the results observed in the Western blot analysis, we conducted IHC analyses of the selected proteins. The results indicated that the DGT-treated tumor tissues expressed decreased Gli1, phospho-Akt, and phospho-ERK but increased phospho-GSK3β (Supplementary Fig. S7). Our previous studies regarding the role of GSK3β showed that increased phospho-GSK3β induces apoptosis and inhibits the growth of osteosarcoma (22). In addition, the level of β-catenin was also decreased, which is associated with osteosarcoma growth and metastasis (33–35). To determine whether Akt, ERK, GSK3β, and β-catenin repression was driving DGT cytotoxicity, U2OS cells were treated with DGT after siRNA-mediated knockdown of these proteins. The results of MTT assays show that inhibition of β-catenin slightly decreased cell viability. However, knockdown of Akt induced greater osteosarcoma cell death, and similar effects were seen in cells with GSK3β knockdown (Fig. 7C). Next, we induced the siRNA-mediated knockdown of these kinases and then treated these cells with vehicle or DGT. When combined with DGT treatment, we found that the cells with siRNA-mediated knockdown of β-catenin exhibited more cell death compared with cells treated with vehicle (Fig. 7D). This result suggested that Akt and GSK3β play a more important role in the survival of osteosarcoma cells. Together, these data strongly support the conclusion that inhibition of Akt/ERK- and GSK3β-mediated apoptosis is an important mechanism of DGT-induced cellular cytotoxicity.

Gli1 can be regulated by specific serine/threonine kinases (34, 36–38), and we have demonstrated that DGT inhibits GSK3β activity, which are known to be associated with osteosarcoma progression and poor patient outcome. Therefore, we hypothesized that GSK3β activity is involved in DGT-induced Gli1 repression in osteosarcoma. Consistent with this hypothesis, we treated osteosarcoma cells with siRNA and Licl (a specific inhibitor targeting GSK3β), and the results showed that blocking GSK3β downregulates Gli1 expression but not smo and inhibits Gli1 luciferase activity in ZOS cells (Fig. 8A and B). Furthermore, we investigated whether Gli1 also plays a key role in GSK3β-mediated osteosarcoma malignant behaviors, because we have demonstrated that GSK3β prompts the growth of osteosarcoma by activating the NF-κB pathway. We next transfected GSK3β knockdown ZOS cells with vector or ectopic Gli1 expression plasmid and assessed the changes in malignant behaviors, including the cell viability, metastasis, and sphere formation. Our results showed that ectopic Gli1 expression could partially increase cell viability, invasion, and sphere formation impaired by knockdown of GSK3β in ZOS cells (Fig. 8C–H). In addition, DGT impaired the increased cell viability induced by ectopic Gli1 and GSK3β expression (Fig. 8I). Thus, the inhibition of the Gli1 pathway induced by DGT primarily involves repression of GSK3β activity.

In this study, we report the effect of the natural compound DGT from Solanum nigrum L. on osteosarcoma, which is the first step toward new drug development according to the current challenge of osteosarcoma treatment. Numerous researchers have focused on natural extracts owing to the success of artemisinin (qinghaosu) and arsenic (III) oxide (As₂O₃) in the clinic. Owing to their safety, long-term use, and their ability to target multiple pathways, there is a renewed interest in understanding the molecular mechanisms underlying their activity. Our previous studies have also provided evidence of some natural agents having potential anti-osteosarcoma activity, such as dihydromyricetin, cinobufagin, and bufalin (19, 39, 40). However, there is no evidence at the cellular level or in animal models for such an effect of DGT on osteosarcoma progression. The study presented here indicated that DGT diminishes the growth and metastasis of osteosarcoma, which may be a promising therapeutic strategy against osteosarcoma in vitro and in vivo without obvious side effects.

In this study, all the osteosarcoma cells were susceptible to the cytotoxicity of DGT, and the ability of DGT to prevent colony formation and proliferation was remarkable and dose dependent. The results of the cell-cycle assay showed that DGT could induce G₂–M cell-cycle arrest and increase p21 levels in osteosarcoma cells. The histone H2A.X, which is the gold standard for early detection of DNA damage, results in cell-cycle arrest and/or apoptosis (41). We found that DGT treatment not only caused significant DNA damage, as detected by the change in thep-H2A.X level, but also induced apoptosis of osteosarcoma cells in dose-dependent manner, as detected by PARP, caspase-3 activity, and Annexin V/PI staining. We also reported that DGT had antitumor potential in nude mice, including an orthotopic model, without severe side effects. Pulmonary metastasis is the primary cause of medical therapy failure and death in osteosarcoma patients. Given that controlling metastasis is the key to improving survival, there is an urgent need to develop more effective approaches to suppress lung metastasis. Therefore, we investigated the effect of DGT on lung metastasis of osteosarcoma. In vitro and in vivo, we found that DGT inhibited the migration, invasion, and metastasis of osteosarcoma cells. These results encouraged us to investigate the potential anti-osteosarcoma mechanism of DGT, which will be beneficial for developing clinical trials in the future.

The HH signal pathway is also considered to be crucially involved in the development and progression of many cancers because it is overactivated and correlated with growth and metastasis (36, 42, 43). Activated Gli proteins, primarily Gli1, translocate into the nucleus and stimulate the transcription of HH pathway target genes, including Gli1, PTCH1, and many survival-promoting molecules (44–46). Here, we demonstrated that Gli1 is overexpressed in osteosarcoma cell lines and is positively associated with cell survival and metastasis in osteosarcoma and thus presents a promising target. In addition, the Gli1 protein is strongly expressed in 66.2% (47/71) of clinical samples and is correlated with the prognosis of osteosarcoma patients. Most importantly, we found that DGT prevents the expression of Gli1 in osteosarcoma cells in a dose-dependent manner, and Gli1 luciferase reporter activity was decreased after treatment with DGT. When combining DGT with Gli1-overexpressing osteosarcoma cells, the growth and metastatic abilities prompted by Gli1 were partially reversed. These findings suggest that Gli1 is a significant prognostic marker and that DGT inhibits growth and metastasis by inactivating the HH/Gli1 pathway.

Because many kinases play key roles in cancer cell survival and growth, to further determine the potential anti-osteosarcoma molecular mechanism of DGT, we assessed the phosphorylation levels of multiple kinases using a proteome profiler array in DGT-treated osteosarcoma cells, especially the change in GSK3β. Our previous study showed that GSK3β activity may promote osteosarcoma tumor growth, and therapeutic targeting of GSK3β may be an effective way to enhance the therapeutic activity of anticancer drugs against osteosarcoma (22). Our results show that DGT inhibits activation of GSK3β by increasing the phosphorylation level. Gli1 has been reported to be activated by many kinases, such as AKT, MAPK/ERK, and mTOR/S6K1 (42, 47). In this study, we were able to show a link between GSK3β and Gli1. Targeting GSK3β with siRNA or Licl downregulated Gli1 expression but not smo, and it inhibited Gli1 luciferase activity in osteosarcoma cells. In addition, Gli1 plays a key role in GSK3β-mediated osteosarcoma malignant behaviors, because ectopic Gli1 partially increased the viability, invasion, and sphere formation impaired by knockdown of GSK3β in ZOS cells. On the basis of our results, we suggest that DGT possesses antitumor activity due to its ability to affect Gli1 expression and activation by blocking GSK3β. However, how DGT inhibits GSK3β activity and represses the HH/Gli1 pathway requires further research.

In summary, the results from our cell-based and in vivo studies support the apoptosis-inducing, antiproliferative, anti-invasive, and antimetastatic activities of DGT. The underlying mechanism by which DGT exhibits anti-osteosarcoma activity seems to be through inhibition of GSK3β/Gli1 activation. The efficacy of the inhibition of growth and metastasis of the osteosarcoma xenograft at relatively low concentrations strengthens the therapeutic value of DGT. Because osteosarcoma is an extremely aggressive type of cancer that lacks any targeted therapies, this study provides strong evidence for evaluating the safety and efficacy of DGT in clinical studies, and it may be an excellent auxiliary drug for treating patients with osteosarcoma.

No potential conflicts of interest were disclosed.

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conception and design: Z. Zhao, Q. Jia, M.-S. Wu, J.-Q. Yin, J. Shen

Development of methodology: M.-S. Wu, J.-Q. Yin

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Z. Zhao, Q. Jia, M.-S. Wu, X. Xie, Y. Wang, D.-C. Lin, J. Shen

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Z. Zhao, G. Song, C.-Y. Zou, Q. Tang, J. Lu, J. Shen

Writing, review, and/or revision of the manuscript: Z. Zhao, M.-S. Wu, J. Shen

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Q. Jia, G. Huang, J. Wang, J. Shen

Study supervision: H.P. Koeffler, J.-Q. Yin, J. Shen

This work was supported by grants from National Natural Science Foundation of China (no. 81602356, 81560603, and no. 81472506); Guangdong Natural Science Foundation (S2013010016847), Sun Yat-Sen University Clinical Research 5010 Program (no. 200709), and Young teachers cultivating project of Sun Yat-Sen University (no. 14ykpy16).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.