The study protocol was specifically approved by the Ethics Committee of West China Hospital of Sichuan University (Chengdu, China) and complied with the Declaration of Helsinki. A total of 6 OS tissue samples and 6 osteochondroma tissue samples were collected during surgery from patients between October 1, 2017 and December 31, 2018, and the patient information is presented in Table I. No patients had received radiotherapy or chemotherapy prior to surgery. All patients provided written informed consent and agreed to participate in this research. All collected tissues were frozen instantly in liquid nitrogen following resection and were preserved in an ultralow temperature refrigerator prior to use in the experiments.

The expression of Ski in both OS tissues and osteochondroma tissues was detected by IHC under the standard immunoperoxidase staining procedure. In brief, the tissues were embedded with paraffin and conventionally sliced into 4-µm-thick sections. The sections were baked for 2 h at 60°C before being dewaxed in xylene, and were then rehydrated through graded ethanol, and subsequently placed in sodium citrate buffer (pH 6.0) and heated up to 100°C for 5 min for antigen repair. After cooling down, the sections were incubated with 5% goat serum (Solarbio) at 37°C for 30 min, and the sections were then incubated with rabbit anti-human Ski monoclonal antibodies (1:200; sc-33693, Santa Cruz Biotechnology) at 4°C overnight, and then incubated with the secondary antibody (1:1,000; biotinylated goat anti-mouse lgG, ZB-2305; ZSBIO) at 37°C for 1 h. The sections were developed with diaminobenzidine (DAB; Beyotime Biotechnology) to detect the bound antibody. After staining, the sections were sealed up with balsam before observation using an optical microscope (Olympus Optical). The brownish-yellow presented in cytoplasm or cytomembrane indicated positive results, or otherwise indicated negative results. Five fields were randomly selected for detection to calculate the positive expression. The experiment was performed 3 times.

All the cell lines were purchased from the Cellular Center of Institute of Basic Medical Sciences. Human OS cell lines, including MG63 and U2OS were cultured in Eagle's minimum essential medium containing 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific). The human normal osteoblasts (hFOB1.19) were cultured in Dulbecco's modified Eagle's medium/F12 (DMEM/F12; HyClone) containing 10% FBS (Gibco; Thermo Fisher Scientific) and 1% streptomycin and penicillin (Gibco; Thermo Fisher Scientific). The cells were cultured in a humidified atmosphere in CO₂ cell incubator (37°C, 5% CO₂).

The protein samples were extracted from the tissues or cells using RIPA lysis buffer (990 µl RIPA and 10 µl phenylmethylsulfonyl fluoride; Beyotime) on the ice. The supernatants were then transferred into a new 1.5 ml centrifuge tube and centrifuged in a low temperature environment for 15 min at 17,900 × g, and insoluble material was discarded. The sample concentration was detected using the BCA kit (Beyotime). The protein samples (20 µg) were separated by 10 or 12% SDS-PAGE, and the proteins were then transferred onto PVDF membranes (0.45 µm; Millipore) using a transfer unit (Bio-Rad). Thereafter, the membranes were blocked in TBS solution containing 5% BSA for 2 h at room temperature. The membranes were then incubated with primary antibodies including anti-Ski (mouse, 1:200, sc-33693; Santa Cruz Biotechnology), proliferating cell nuclear antigen (PCNA; 1:1,000, 10205-2-AP, rabbit; Proteintech), CDK4 (sc-23896), cyclin D1 (sc-450) (both mouse, 1:500; Santa Cruz Biotechnology), matrix metalloproteinase (MMP)2 (4022S), MMP9 (3852S) (both rabbit, 1:1,000; Cell Signaling Technology), PI3K (4292S), p-PI3K (4228S), Akt (9272S) (all rabbit, 1:1,000; Cell Signaling Technology), p-Akt (4060S, rabbit, 1:2,000; Cell Signaling Technology) and β-actin (TA09, mouse, 1:1,000; ZSBIO) overnight at 4°C. Following washing with TBST, the membranes were then incubated with secondary antibodies (ZB-2301 and ZB2305, 1:10,000; ZSBIO) for 1 h. After washing with TBST 3 times, the proteins on the membranes were visualized by ECL, and the results were scanned by Quantity One software (Bio-Rad).

The cells were cultured on small coverslips and allowed to adhere. Thereafter, the cells were washed with PBS solution for 5 min each, then fixed with 4% paraformaldehyde for 30 min, and washed with PBS for 5 min each. The cells were then permeabilized with 0.3% Triton X-100 for 20 min, and washed in PBS for 5 min each. The cells were blocked for 30 min with 10% goat serum at 37°C, and thereafter, anti-Ski (sc-33693, anti-human 1:300; Santa Cruz Biotechnology); antibodies were added followed by incubation overnight at 4°C. The following day, the cells were washed in PBS 3 times, and then incubated with secondary antibodies (Alexa Fluor 488-conjugated anti-mouse, 1:300; A21202; Invitrogen; Thermo Fisher Scientific) for 2 h at room temperature. Finally, the samples were incubated with DAPI for 10 min at room temperature. The images were acquired with a fluorescence microscope (Zeiss; Carl Zeiss).

Ski specific siRNA (Ski-siRNA) and scrambled sequence siRNA (negative control siRNA, NC-siRNA) were purchased from Thermo Fisher Scientific. The target sequence of Ski-siRNA was 5′-CGGACCTTGGCTGGTTCCTCCAATA-3′, and the sequence of 5′-TTCTCCGAACGTGTCACGT-3′ was considered to be NC-siRNA. The MG-63 and U2OS cells were transfected with Ski-siRNA or NC-siRNA using Lipofectamine^® RNAiMAX Reagent (Thermo Fisher Scientific) in accordance with the manufacture's protocol. Following 48 h of transfection, the cells were collected for the used in the following experiments. Western blot analyses were used to assess the silencing effect of Ski-siRNA. The cells were pre-treated with basic culture medium DMEM/F12 with or without 50 µM LY294002 (a specific PI3K inhibitor, S1737, Beyotime) for 6 h prior to transfection.

Cell Counting kit-8 (CCK-8) assays (Dojindo) were used to assess cell proliferation. In brief, at 48 h following transfectiong with Ski-siRNA and NC-siRNA, the cells were trypsinized, and then replanted into 96-well plates with 3×10³ cells/well for the cell proliferation assay. This was followed by the addition of DMEM (100 µl) containing 10 µl of CCK-8 working solution to each well at corresponding time-points and incubation at 37°C for 3 h in an incubator. Following incubation, cell viability was determined by measuring the absorbance at 450 nm using a microplate reader (Bio-Rad).

The EdU reagent kit (Ribobio) was also used to evaluate cell proliferation based on our previous study (15). Briefly, the OS cells were transfected with siRNA for 48 h and the medium was then changed with culture medium to continue culture for 2 and 4 days. The cells were then incubated with complete medium solution containing 50 M EdU for 1 h at corresponding time-points. The following procedure was carried out in accordance with the manufacture's protocol. Finally, images were acquired with a fluorescence microscope (Zeiss; Carl Zeiss).

The OS cells were planted in a cell culture plate (35 mm) with 3×10⁵ cells and then incubated overnight until 50-70% confluent. Prior to transfection, the cells were synchronized for 6 h in serum-free DMEM culture medium. At 48 h following transfection with siRNA, the cells were trypsinized and transferred to a 10 ml centrifuge tube (centrifugation at 200 × g for 5 min at 4°C, washed with cold PBS solution twice, fixed with 70% pre-cold ethanol overnight at −20°C. Subsequently, 190 µl EDTA and 10 µl Rnase A (1 mg/ml) were added to tubes for 5 min at room temperature, and 18 µl propidium iodide and 97 µl PBS were then added for 10 min at 4°C in the dark, respectively. The volume was adjusted to 1 ml with PBS. Subsequently, the percentage of cells cycle was analyzed using a BD FACSCan (BD Biosciences).

Wound healing assay was performed to estimate the OS cell migratory activities. The transfected OS cells were plated into a 6-well plate at a density of 5×10⁵ cells/well and two horizontal lines were marked on the back of the well plate. When the confluence of the cells reached 100%, the cells were scratched with a sterile 100 µl pipet tip, and the cells debris was then removed using PBS solution. Following the addition of serum-free medium containing 2% FBS, the cells were placed back into the incubator (37°C, 5% CO₂) for 48 h. The migration of the cells was photographed under a microscope with Zen Imaging software at 0, 12, 24 and 48 h after scratching. The percentage of wound healing was determined as described in our previous study (16).

Statistical analysis was performed using SPSS 21.0 software (IBM). All data are expressed as the means ± SD. Statistical significance between two groups were analyzed with the Student's t-test. The significance among multiple groups was using one-way ANOVA followed by Tukey's post hoc test. P-values <0.05 were considered to indicate statistically significant differences. Each experiment was carried out in triplicate.

Compared with the osteochondroma tissues, Ski was more highly expressed in the OS tissues. As shown in Fig. 1A and B, Ski was overexpressed in OS tissues, whereas its expression was hardly detected in the osteochondroma tissues by immunohistochemistry. The results of western blot analysis also demonstrated that Ski (the molecular weight of Ski varies between 90 and 132 kDa) increased significantly in the OS sample tissues compared with the osteochondroma samples (Fig. 1C and D). These results indicated that Ski may play an essential role in the tumorigenesis of OS. Thus, we compared Ski expression level in OS cell lines (MG63 and U2OS) with hFOB1.19 normal osteoblast cells. The results of immunofluorescence staining revealed that Ski expression was more abundant in the OS cell lines compared with the osteoblasts (Fig. 2A). Furthermore, western blot analysis confirmed the results again (Fig. 2B).

In order to examine the role of Ski in OS cells, a specific siRNA (Ski-siRNA) was applied in the OS cell lines, MG63 and U2OS cells. The results indicated that Ski expression was inhibited efficiently in the MG63 and U2OS cells (Fig. 3A and D). Subsequently, CCK-8 assay was performed to investigate the effects of Ski-siRNA on OS cell proliferation. The results revealed that OS cell proliferation was markedly suppressed following transfection with Ski-siRNA at 48 and 72 h in the MG63 cells (P<0.001) compared with the control group and NC-siRNA group (Fig. 3E), and similar results were obtained with the U2-OS cells (Fig. 3F). In addition, EdU kit assay was applied to investigate the OS cell proliferative ability, and the results indicated that the MG63 and U2-OS cell proliferative ability was inhibited significantly following the knockdown of Ski (Fig. 4A-D). Moreover, to further confirm the results mentioned above, the proportions of cell cycle profiles were investigated following transfection with Ski-siRNA by flow cytometry. The results revealed that, the percentage of MG63 cells at the G1 phase increased significantly, while the percentage at the G2/M phase decreased in the Ski-siRNA group compared with the NC-siRNA group (Fig. 4E and G); however, no significant difference was observed in the cells in the S phase. Similar results were obtained for the U2OS cells (Fig. 4F and H). Furthermore, the results of western blot analysis revealed that the expression of proliferation-related proteins, including PCNA, CDK4 and cyclin D1 decreased significantly following transfection with Ski-siRNA in OS cells (Fig. 5). All the above-mentioned results strongly indicate that Ski plays a crucial role in the proliferation of OS cells.

In addition to proliferation, OS cell migration also plays a critical role in tumor metastasis. Thus, in this study, a wound healing assay was performed to assess the effects of Ski knockdown on MG63 and U2OS cell migration. Following the knockdown of Ski, the percentages of MG63 cells after wound healing at 48 h in the NC-siRNA group and Ski-siRNA group were 94.47±5.36 and 52.53±6.57%, respectively, and the rate of wound healing in the Ski-siRNA group was much lower than that of the NC-siRNA group (Fig. 6A and B); similar results were obtained for the U2OS cells as well (Fig. 6A and C). In addition, the results of western blot analysis revealed that the expression of migration-related proteins, including MMP2 and MMP9 markedly decreased following the knockdown of Ski (Fig. 7).

The PI3K/Akt signaling pathway plays an essential role in tumorigenesis, which has been well expounded as an essential pathway for the proliferation and migration of OS cells. Thus, in this study, the effects of Ski on the certain molecules which are involved in the PI3K/Akt signaling pathway in MG63 and U2-OS cells were examined. The results revealed that the knockdown of Ski decreased the PI3K and Akt phosphorylation levels significantly in the MG63 cells compared with the control and NC-siRNA groups (Fig. 8A). Similar results were obtained in the U2OS cells (Fig. 8B). Furthermore, to verify the function of the PI3K/Akt pathway, we examined the effects of the LY294002 on the Ski-mediated proliferation and migration of OS cells. The results revealed that LY294002 significantly enhanced the inhibitory effects of Ski-siRNA on OS cell proliferation and migration, and these results were further verified by CCK8 assay (Fig. 8C and D) and wound healing assay (Fig. 8E and F).