Background
Osteosarcoma is the most common primary malignant bone tumor accounting for approximately 60% of all bone sarcoma [
1,
2]. With the advance of chemotherapy, although the long-term cure rate after surgery for non-metastatic osteosarcoma has risen from 25% to 60% [
3], the survival rate for osteosarcoma is still rather low. Most osteosarcomas are high grade with part of them were accompanied by lung metastasis [
4]. Metastatic disease is usually not sensitive to conventional chemotherapy with long-term survival rate approximately 20% [
5]. Therefore, the development of chemotherapy for osteosarcoma is urgently needed.
For a long time, apoptosis was regarded as the sole form of programmed cell death, while necrosis was considered as an unregulated and uncontrollable process. In 2004, Zong, WX, et al. found a regulated form of necrotic cell death during the damage of DNA [
6], which was named as “necroptosis” later and suggested that necrosis might not be absolutely unregulated. In 2005, Degterev, A, et al. found that Nec-1 (necrostatin-1) was a specific inhibitor of necroptosis [
7]. The idea of “necroptosis” was demonstrated by a series of subsequent studies in which increasing signal molecules functioning as initiators or effectors of necroptosis such as receptor-interacting protein 1 [
8] (RIP1, RIPK1) and receptor-interacting protein 3 [
9,
10] (RIP3, RIPK3) or inhibitors such as necrostatin-1 (Nec-1), were discovered. Since necroptosis is a pathway separate from apoptosis, all the barriers set up in cancer cells to avoid apoptosis are no longer problems for necroptosis [
11].
Shikonin, an effective constituent, purified from
Lithospermum erythrorhixon, a Chinese medicinal herb, was widely used in anti-inflammatory process [
12,
13]. Shikonin was thought to have anti-tumor effect by inducing apoptosis until people found that shikonin could circumvent cancer drug resistance by inducing necroptosis in 2007 [
11,
14]. Interestingly shikonin also exert two death modes of apoptosis and necroptosis in KL-60 cells depending on its concentrations [
15]. Moreover, shikonin was demonstrated to mediated necrotic cell death via a RIP1-RIP3 complex similar to TNFα-directed necrotic cell death, and this pronecrotic complex was blocked by a reactive oxygen species (ROS) scavenger or Nec-1 concomitantly with protection against cell death [
16]. In 2011, the first molecular target of shikonin was reported in which shikonin played a role in the anti-tumor effect by inhibiting pyruvate kinase-M2(PKM2). PKM2 is universally over expressed in cancer cells and dictated to the last rate-limiting step of glycolysis vital for cancer cell proliferation [
17]. Recently, shikonin was also found to be a cytotoxic DNA-binding agent [
18]. Furthermore, shikonin and its analogs were demonstrated hardly to inducer cancer drug resistance [
19]. The effect of shikonin on bone sarcomas is still unclear. In this study, we tested whether shikonin had anti-tumor effect on osteosarcoma and explored the underlying mechanism.
Methods
Cell Lines and culture
Murine osteosarcoma cell lines K7, K12 and K7M3 cell lines were from Dr. Kleinerman’s lab in MD Anderson Cancer Center which were originally established by Khanna [
20]. Human osteosarcoma cell lines U2OS and 143B cell lines were obtained from American Type Culture Collection (ATCC). All cells were cultured in high glucose Dulbecco’s Modified Eagle’s Medium (DMEM-h; Thermo, America) supplemented with 10% fetal bovine serum (Thermo, America), 100 U/ml penicillin and 100 μg/ml streptomycin (Thermo, America) in a humidified incubator at 37°C in 5% CO
2.
Drugs and antibodies
Purified shikonin (>98%) was purchased from Shanghai Tauto Biotech Co., Ltd. Stock solution at 50 mM was made in dimethyl sulfoxide (DMSO; Sigma, America) and stored in the dark at −20°C. The final shikonin concentrations used for different experiments were prepared by diluting the stock solution with DMEM-h. The antibodies used for Western blot were as follows: rabbit anti-Actin (Santa Cruz, CA, USA), anti-caspase-3 (Cell Signaling Technology Inc., Danvers, MA), anti-caspase-6 (Cell Signaling Technology Inc., Danvers, MA), anti-PARP (Cell Signaling Technology Inc., Danvers, MA) and mouse anti-RIP1 (BD, CA, USA), anti-RIP3 (Abcam, USA).
MTT assay
Cells were seeded into 96-well plates (cultured overnight for adherent cells) and treated with shikonin at a series of concentrations (0, 1, 3, 5, 7.5, 10, 12.5, 15 μΜ) for 8 hours or treated with shikonin (3 μΜ) for 8, 16 or 24 hours. Cells incubated with DMEM-h were regarded as control group. After 8, 16 or 24-hour incubation, 20 μl MTT (5 mg/ml, Sigma, America) was added into each well for another 4-hour incubation. After that, the supernatant was removed and 150 μl DMSO was added into each well in order to solubilize the blue-purple crystals of formazan. The absorbance was then measured using a model ELX800 Micro Plate Reader (Bio-Tec Instruments, Inc.) at 490 nm. The survival rate was calculated according to the following formula: Survival rate = Absorbance of treatment / Absorbance of control × 100%.
In the inhibition experiment, K7, K12, K7M3 and U2OS cells were treated with shikonin (3 μΜ) while 143B cells were treated with shikonin (6 μΜ) in the absence or presence of necrostatin-1 (Nec-1; Sigma, America) or Z-VAD-FMK (Sigma, America) for 8 hours. The cell survival rate was measured by MTT assay. When added MTT, the supernate in the well with Nec-1 was discarded and added DMEM-h again.
Flow cytometry analysis
Osteosarcoma cells (K7, K12, K7M3, U2OS and 143B) were plated in 6-well plates and synchronized with DMEM-h containing 10% fetal bovine serum. After 8-hour incubation, control cells (incubated with DMEM-h) and shikonin-treated cells (143B cells were treated with 6 μΜ shikonin while other cells were treated with 3 μΜ shikonin) in the presence or absence of Nec-1 (50 μΜ) were collected, washed twice in cold PBS. The cells used in cell cycle were mixed in 300 μl of 1× binding buffer, and incubated at room temperature for 15 min with propidium dide (PI, Sigma, America), NP-40, and RnaseA (BD Biosciences) while the cells used in cell death were mixed in 100 μl of 1× binding buffer, and incubated at room temperature for 15 min with an annexin-V/PI (BD Biosciences) double staining solution. Stained cells were analyzed by flow cytometry. The percentage of cells in the different stages and the percentage of necrosis cells were calculated using ModFit LT software (Verity Software House).
Western blot assay
K7, U2OS and 143B cells were treated with different concentrations of shikonin for 8 hours. Cells were washed twice with PBS solution, lysed with RIPA Lysis Buffer (Beyotime Institune of Biotechnology, Shanghai, China) and protease inhibitor (Thermo scientific). Tumor tissues were retrieved from −80°C storage and immersed rapidly in liquid nitrogen. The resulting powder was lysed with RIPA Lysis Buffer and protease inhibitor. Protein concentrations were determined with Pierce BCA Protein Assay Kit (Thermo Scientific). Equivalent amounts of total protein (50 μg) were boiled and electrophoretically seperated on a 10% polyacrylamide gel at 80 volts. The proteins were transferred to a nitrocellulose filter membrane. Membranes were blocked for 60 min with 5% milk solutions prepared in PBS, incubated overnight at 4°C with 1:1000 dilutions of the primary antibodies (RIP1, RIP3, PARP, caspase-3, caspase-6 and Actin), washed three times for 10 min each time with Tween 20 (1:1000 dilution)-PBS, incubated for 1 hour with the appropriate peroxidase-conjugated secondary antibody (1:1000 dilution). Membranes were washed with Tween 20-PBS three times for 10 min each and were developed using the Odyssey two-color infraed laser imaging system. The signal generated by Action was used as an internal control.
Animal experiments
Animal experiments were performed on 4-week-old female mice (BALB/C). Mice were housed in a standard animal laboratory with free access to water and food. They were kept under constant environmental conditions with a 12-hour light–dark cycle. All operations were performed under aseptic conditions. All the animal-related procedures were approved by the Animal Care and Use Committee of The Tenth People's Hospital of Shanghai. This study was also approved by the Science and Technology Commission of Shanghai Municipality (ID: SYXK 2007–0006) with the permit number 2011-RES1.
Mice tibial tumor models and treatment regimen
Balb/c mice (4 weeks of age) were purchased from Shanghai Slac Laboratory Animal Co., Ltd. K7 cells were digested and washed by cold PBS for three times, suspended in cold PBS. The final concentration of K7 cells was 1 × 108/ml. The cell suspension (10 μl) was injected into medullary cavity of tibia. Mice were divided into two groups, shikonin group and control group (each had 10 mice). Three weeks later, when the tumors in the tibia were macroscopic, shikonin group was injected with shikonin (2.0 mg/kg, diluted with 5% DMSO) while control group was injected with 5% DMSO. Both groups were injected intraperitoneally every other day for seven times in all. The mice were euthanized two days after the last injection. The primary tumor size and lung metastasis were observed. Posterior limb with tumors and lungs were weighted. Necrotic degree of primary tumors and lung metastasis was detected by HE stain. The expression levels of RIP1 and RIP3 in primary tumor tissues were determined by Western blot.
Balb/c mice (4 weeks of age) were purchased from Shanghai Slac Laboratory Animal Co., Ltd. K7 cells were digested and washed by cold PBS for three times, suspended in cold PBS. The final concentration of K7 cells was 5 × 106/ml. The cell suspension (100 μl) was injected into the mice from caudal vein. Two weeks later, mice were divided into two groups (each had 10 mice), experimental group was injected with shikonin (2.0 mg/kg, diluted with 5% DMSO), control group was injected with 5% DMSO. Both groups were injected intraperitoneally every other day.
Statistical analysis
Statistical analysis was performed using GraphPad Prism 5 (La Jolla, CA, USA). All measurement data were expressed as mean ± standard deviation (SD), and compared between two groups using Student’s t test. P < 0.05 was considered statistically significant.
Discussion
In this study, we used different cell lines to test the efficacy of shikonin on osteosarcoma. We demonstrated that shikonin had prompt anti-tumor effect on osteosarcoma cells, which had no impact on cell cycle. This indicated that shikonin kills osteosarcoma cells directly rather than inhibiting proliferation. The majority of K7, K12, K7M3 and U2OS cell death induced by shikonin could be rescued by Nec-1, a specific inhibitor of necroptosis, while Z-VAD-FMK, a general inhibitor of apoptosis, had no obvious protective effect. As a potent necroptosis inhibitor, Nec-1 was previously believed not to inhibit apoptosis [
7,
14,
21]. Thus the death of K7, K12, K7M3 and U2OS cells induced by shikonin could be considered as necroptosis. The degree of necrosis and apoptosis were further detected by flow cytometry with Annexin V and PI double staining. The results showed that necrotic cells were almost totally prohibited by pretreatment with Nec-1 before exposure of shikonin in K7 cells. Additionally, K7 cells at late apoptotic stage were also attenuated by Nec-1, indicating that part of those cells might be necrotic cells. This result is similar to previous data [
21]. Therefore, shikonin could kill osteosarcoma cells quickly by inducing necroptosis at least in some osteosarcoma cell lines.
RIP1 and RIP3 are regarded as crucial modulators of necroptosis [
8‐
10]. In our study, we found the protein levels of RIP1 and RIP3 were significantly increased after treatment with shikonin in a concentration dependent manner in some osteosarcoma cell lines including K7 and U2OS. These results indicated that shikonin induced cell death in some osteosarcoma cell lines including K7 and U2OS via RIP1 and RIP3 dependent necroptosis pathway. Moreover, other reports have also shown that necroptosis could be induced via modulating RIP1 and RIP3 [
16,
22‐
24]. Also, there were results different from ours. Moujalled, D. M. et al. demonstrated that TNF can activate RIP3 and cause necroptosis in the absence of RIP1 [
25]. From the experiments in vivo, we also found that the protein levels of RIP1 and RIP3 in primary tumor tissues were increased in shikonin group compared with control group. It can be inferred that shikonin had anti-tumor effect in vivo probably by inducing necroptosis, just as the mechanism proved by the previous mentioned cell studies.
Interestingly, we found that in osteosarcoma 143B cells, which were previously detected less sensitive to shikonin compared with K7 and U2OS cells, cell death induced by shikonin could neither be reduced by Nec-1 nor by Z-VAD-FMK. We also found that RIP1 and RIP3 had no obvious change while caspase-3, caspase-6 and PARP were not activated after being treated with shikonin. Different from our results, Chang, et al. found that shikonin induces apoptosis through reactive oxygen species/extracellular signal-regulated kinase pathway and PARP was activated in 143B cells after being treated with shikonin for 24 hours [
26]. It may be because the treatment time was different and need further study. As we know, the 143B cell line is a Ki-ras transformed TE85 [
27,
28] and not sensitive to shikonin. Maybe Ki-ras is a barrier to necroptosis. Interestingly, we also found the cell death of SaoS2 cells induced by shikonin could not be rescued by Nec-1(data not shown). U2OS is a p53-positive cell line while SaoS2 is a p53-null cell line [
29]. We found the protein level of p53 was increased after treated with shikonin for 8 hours(data not shown). Maybe p53 is a regulator of necroptosis. Above-mentioned hypothesis is what our recent works focus on and needs further study.
The drug-resistance of cancer is associated with apoptotic pathway tightly, including overexpression of anti-apoptotic proteins, mutations of pro-apoptotic proteins and the loss of caspase [
30‐
32]. In the clinic single-agent activity of methotrexate, cisplatin, doxorubicin and ifosfamide is approximately 40%, 30%, 40% and 30% respectively [
33]. Combination chemotherapy yields slightly better results, but still approximately 40% of patients are not sensitive. The lung metastatic osteosarcoma also exhibits resistance to conventional chemotherapy. The 5-year survival rate for patients of osteosarcoma with metastasis is 20%, much lower than the corresponding survival rate for patients with localized disease (approximately 60%), and most death associated with osteosarcoma is the result of metastatic diseases. The specific mechanism of drug resistance of osteosarcoma might be associated with the activation of the Src and NF-κB pathway and the overexpression of anti-apoptosis genes [
34‐
36]. Based on the results of this study, shikonin has strong anti-tumor effect on both primary and lung metastatic osteosarcoma by inducing necroptosis. As necroptosis undergo pathway independent of apoptosis, all the barriers set up in cancer cells to avoid apoptosis are no longer problems for necroptosis.
Conclusions
Based on both in vivo and in vitro experiments, this study proved that shikonin had prompt but profound anti-tumor effect on both primary and metastatic osteosarcoma. The main mechanism of this effect might be inducing RIP1 and RIP3 dependent necroptosis. In order to develop its clinical use, in the future more studies are still required to evaluate the proper concentration and safety of shikonin, and shikonin related drug interactions in the treatment of osteosarcoma.
Competing interests
The authors declare that they have no financial conflict of interests.
Authors’ contributions
The work presented here was carried out in collaboration between all authors. ZC and YH defined the research theme, designed methods and experiments. ZF carried out the MTT assay, flow cytometry and Western blot, drafted the manuscript. LS, BD, YL, YH and FY carried out the in vivo experiments. YL, HZ and DZ analyzed the data. ZW and YL proofread the manuscript. All authors read and approved the final manuscript.