The human osteosarcoma cell lines 143B and U2OS were obtained from ATCC. 143B cells were cultured in α-MEM containing 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and 1% antibiotics (penicillin, 100 U/ml, streptomycin, 100 µg/ml; Gibco; Thermo Fisher Scientific, Inc.). U2OS cells were cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.) containing 10% FBS (Gibco; Thermo Fisher Scientific, Inc.) and penicillin-streptomycin (5,000 U/ml; Gibco; Thermo Fisher Scientific, Inc.). Cells were propagated at 37°C with 5% CO₂ and 100% humidity. The cells were routinely passaged every 2–3 days when they reached ~80% confluence. Cancer stem cells were enriched by serum-free culture. Cells were plated in ultra-low attachment plates at a density of 2,500 cells/ml in RPMI 1640 (Gibco; Thermo Fisher Scientific, Inc.) supplemented with B27 supplement (Invitrogen; Thermo Fisher Scientific, Inc.), 20 ng/ml human epidermal growth factor (EGF; Invitrogen; Thermo Fisher Scientific, Inc.) and 20 ng/ml human basic fibroblast growth factor (bFGF; Invitrogen; Thermo Fisher Scientific, Inc.). Following culture for 2 weeks, colonies with a diameter ≥50 µm were regarded as sarcospheres and quantified by inverted phase contrast microscopy. Spheres were dissociated and re-plated to form the next generation of spheres every 14 days.

Transfection of U2OS and 143B cells with small interfering (si)RNA (100 nM) was performed using Viromer (Lipocalyx GmbH) according to the manufacturer's instructions. siRNA oligonucleotides were purchased from MilliporeSigma. The following siRNA sequences were used: GSK-3β (5′-CUCAAGAACUGUCAAGUAATT-3′), negative control (5′-UAGCGACUAAACACAUCAA-3′). The transfection complex was co-incubated with the cells for 24 h at 37°C. The culture medium was changed after transfection, and subsequent experiments and detection were performed 48 h later.

The 143B and U2OS OS stem cells were treated with 10 µmol/l AZD1080 and the formation and differentiation of spheres observed. The effect of AZD1080 on stemness of OS stem cells was observed by comparing with OS stem cells without AZD1080.

143B and U2OS OS stem cells (3×10³ cells/well) were plated in 96-well plates. At 24 h incubation with AZD1080, CCK-8 reagents were added and incubated for another 2 h. The absorbance at 450 nm using a Multiskan Spectrum microplate reader represented the number of viable cells.

U2OS and 143B OS stem cells were treated with AZD1080. The cells were then washed with PBS and fixed with 70% ice-cold ethanol at 4°C overnight. After fixation, the cells were stained with propidium iodide at room temperature in the dark for 10 min. The cell cycle was analyzed by flow cytometry (FACSCalibur; BD Biosciences). The data were analyzed using CellQuestPro™ v.6.0 software (Becton, Dickinson and Company).

For cell apoptosis detection, the apoptotic rate of the two OS cell lines was analyzed using an Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (Beyotime Institute of Biotechnology). The distribution of viable, early apoptotic, late apoptotic and necrotic cells was detected using a FACSCalibur flow cytometer (BD Biosciences). The data were analyzed using CellQuestPro™ v.6.0 software. The apoptotic rate was calculated as the percentage of early and late apoptotic cells.

Matrigel was rapidly added to the upper chamber of the Transwell insert and this was transferred to a 37°C incubator for 1–2 h. To assess invasion potential, Matrigel-coated Transwell was used. AZD1080-treated cells were seeded in the upper chamber of the Transwell insert. The migration capacity over the Matrigel and the membrane towards the bottom chamber containing 10% FBS was examined. Cells were fixed in 4% paraformaldehyde solution at room temperature for 20 min, and stained with 0.1% crystal violet at room temperature for 10 min and counted.

The colony formation capability of U2OS and 143B OS stem cells in soft agar media was investigated. Briefly, 100 single-cell suspension of U2OS/143B OS stem cells were resuspended in 0.8 ml growth media containing 0.3% low melting temperature agarose and were plated in triplicate on 24-well plate over a base layer of 0.8 ml growth media containing 0.6% low melting temperature agarose. The plates were incubated for 14–15 days until colonies were formed and colonies with a diameter >50 µm were counted.

Proteins were extracted with Protein Lysis Buffer (Mammalian Cell Lysis Kit; MCL1; MilliporeSigma). Lysates were centrifuged at 10,000 × g at 4°C for 10 min and supernatants collected. Protein concentrations were assessed using the BCA Kit. Cell lysates containing 40 µg protein were separated on a 10% SDSPAGE gel and then transferred on polyvinylidene difluoride membranes using a Trans Blot Turbo (Bio-Rad Laboratories, Inc.). Membranes were blocked in a solution of Tris buffered saline containing 0.05% Tween-20 and 5% skimmed milk for 1 h at room temperature. Primary antibodies were incubated overnight at 4°C. Primary antibodies used were anti-Hes1 (ab108937; 1:500; Abcam), anti-Hey1 (ab154077; 1:500; Abcam), anti-GSK3β (ab32391; 1:500; Abcam), anti-PTEN (ab267787; 1:500; Abcam), anti-MMP-2 (ab92536; 1:500; Abcam), anti-MMP-9 (ab76003; 1:500; Abcam), anti-E-cadherin (ab314063; 1:500; Abcam), anti-Vimentin (ab92547; 1:500; Abcam) and anti-GAPDH (ab8245; 1:2,000; Abcam). Horseradish peroxidase-conjugated secondary antibodies (GB23301 and GB23303; 1:3,000; Wuhan Servicebio Technology Co., Ltd.) were incubated for 2 h at room temperature. Finally, membranes were developed using an enhanced chemiluminescence substrate (G2161; Wuhan Servicebio Technology Co., Ltd.). The Bio-Rad Gel Doc XR+ Gel Imaging System Software Image Lab (version 6.1; Bio-Rad Laboratories, Inc.) was used for analysis.

TRIzol^® (Thermo Fisher Scientific, Inc.) was directly added to the culture dish to lyse the cells (U2OS and 143B) after remove of growth media. The cell density for RNA extraction was ~5×10⁶/ml. RNA extraction using the RNeasy Mini Kit (Qiagen, Inc.), generation of cDNA with Superscript III reverse transcription Kit (Invitrogen; Thermo Fisher Scientific, Inc.) and PCR using the SYBR Green PCR Master Mix (Applied Biosystems; Thermo Fisher Scientific, Inc.) and the ABI Prism 7900TM Sequence Detection System (Applied Biosystems; Thermo Fisher Scientific, Inc.). RNA extraction, cDNA synthesis and qPCR were performed according to the manufacturer's protocols. Experiments were performed in triplicate using three independent cDNAs. Cycling conditions were: Denaturation at 94°C for 30 sec, annealing at 60°C for 30 sec and extension at 72°C for 30 sec for 40 cycles followed by 5 min at 72°C. Primer sequences are below: GSK-3β (F:5′-CAAGCGTGAGACTCGTGATCCTTC-3′, R:5′-AGGCTGCTGCACCGGCTTGCAGCT-3′); MMP2 (F:5′-CCCACTGCGGTTTTCTCGAAT-3′, R:5′-CAAAGGGGTATCCATCGCCAT-3′); MMP9 (F:5′-GGGAGACGCCCATTTCG-3′, R:5′-CGCGCCATCTGCGTTT-3′); PTEN (F:5′-TGGATTCGACTTAGACTTGACCT-3′, R:5′-GGTGGGTTATGGTCTTCAAAAGG-3′); GAPDH (F:5′-GAAGGTGAAGGTCGGAGT-3′, R:5′-GAAGATGGTGATGGGATTTC-3′).

The relative expressions (defined as fold change) of the target genes (2^−ΔΔCq) were normalized to the endogenous GADPH references (ΔCq) (21).

All experiments were performed in triplicate with three independent biological replicates. Random design studies were analyzed using an unpaired t-test. For comparisons of more than two groups, one-way analysis of variance with the Bonferroni post hoc test was applied. All tests were two-sided. GraphPad Prism 10.0 (Dotmatics) was used for all statistical analyses. All results are presented as the mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

Considering that AZD1080 exerts an inhibitory effect on tumors, including ovarian cancer (9), its regulatory role on osteosarcoma-derived CSCs was explored. First, U2OS and 143B cells were cultured in serum-free medium for 5–10 days, obtaining non-adherent cell spheres capable of forming spheres (Fig. 1A). It was found that the expression levels of OCT4 and SOX2 were markedly increased in CSCs (Fig. 1B), indicating successful enrichment of CSCs.

AZD1080 is a specific inhibitor of GSK-3β activity (9), so 10 µmol/l AZD1080 was co-incubated with CSCs for 1–3 days and no significant effect on cell viability was found (Fig. 1C); flow cytometry analysis of the cell cycle also showed that AZD1080 treatment did not affect the cell cycle distribution of U2OS CSCs and 143B CSCs (Fig. 1D); meanwhile, AZD1080 treatment had no significant effect on the apoptosis rate (Fig. 1E). Notably, after co-culturing 10 µmol/l AZD1080 with CSC spheres for 1–3 days, the spheres showed disintegration and partial adherent growth phenotype (Fig. 2A) and the levels of OCT4 and SOX2 markedly decreased due to AZD1080 treatment (Fig. 2B). To further identify whether AZD1080 affects the formation process of CSC spheres, 10 µmol/l AZD1080 was added to serum-free medium to culture adherent cells and observed the formation of spheres. As shown in Fig. 3A, without addition of AZD1080, cell spheres gradually formed within 1–4 days; while after adding AZD1080, although cells continued to grow adherently, the formation of spheres was completely inhibited. AZD1080 treatment markedly decreased OCT4 and SOX2 (Fig. 3B). These results indicated that 10 µmol/l AZD1080 specifically inhibited the formation and enrichment of osteosarcoma CSCs without affecting cell proliferation ability, exerting a specific stem cell inhibitory activity.

In addition, AZD1080 disrupted existing spheres (Fig. 2A) by inhibiting pathways critical for CSC adhesion and maintenance (for example by downregulating OCT4/SOX2 and Notch targets). Simultaneously, it prevented new sphere formation (Fig. 3A) by blocking early stages of CSC aggregation and stemness acquisition. The two effects were mediated through GSK-3β inhibition, which destabilized CSC niche signaling.

The malignant behavior of cancer stem-like cells is a critical cause of poor prognosis (22). Although 10 µmol/l AZD1080 showed no effect on proliferation of osteosarcoma cells, its effect on invasion and tumor formation capacity was subsequently examined. After 24 h pretreatment with AZD1080, the invasion of CSCs was assessed using a Transwell assay. As shown in Fig. 4A, compared with the untreated group (mock), the AZD1080-treated group exhibited markedly reduced invasion. The present study then examined the effect of AZD1080 on the tumor-forming capacity of CSCs in vitro. As expected, AZD1080 markedly inhibited the colony-forming ability of CSCs in soft agar (Fig. 4B). These results suggested that AZD1080 may inhibit malignant behavior by suppressing cancer stem cell stemness. Thus, AZD1080 inhibited invasion and colony formation specifically in osteosarcoma CSCs, highlighting its therapeutic potential against CSC-driven malignancy.

GSK3β is an established regulator of the Notch signaling pathway (23,24), presenting essential regulatory role on Notch1 in osteosarcoma cells (25). Thus, the target genes of Notch1, including HES1 and HEY1, were detected. As presented in Fig. 5A and B, AZD1080 treatment markedly decreased phosphorylated GSK3β without affecting GSK3β total protein. HES1 and HEY1 levels were decreased markedly by AZD1080 treatment, indicating AZD1080 inhibited GSK3β and its downstream signaling.

PTEN is reported as a tumor suppressor gene, which is inhibited by GSK3β post-transcriptionally in osteosarcoma cells by inhibiting MMP-2 and MMP-9 (26,27). Based on the results showing that AZD1080 inhibits invasion, the present study further evaluated the effects of AZD1080 on PTEN, MMP-2 and MMP-9. As presented in Fig. 6A and B, AZD1080 treatment markedly decreased MMP2 and MMP9, transcriptionally and post-transcriptionally, meanwhile AZD1080 treatment increased PTEN transcriptionally and post-transcriptionally. To confirm whether GSK-3β is a key regulator of MMP2, MMP9 and PTEN, siRNA targeting GSK-3β mRNA was employed to knockdown GSK-3β expression. After knocking down GSK-3β efficiently, MMP2, MMP9 and PTEN were modified (Fig. 6A and B). All these results indicated that AZD1080 inhibits GSK3β phosphorylation and thus regulates its downstream signaling.

To further confirm whether the GSK3β activity inhibited by AZD1080 is critical for maintaining stemness and invasive ability, a GSK3β inhibitor, AR-A014418 or siGSK-3β, was employed to block GSK3β's regulatory effect and then observed the enrichment of CSCs. The results showed (Fig. 7A) that both AR-A014418 and siGSK-3β significantly inhibited sphere formation; meanwhile, the expression of stemness-related factors markedly decreased (Fig. 7B), indicating that GSK3β's regulatory activity is crucial for stemness. Meanwhile, the changes in invasive ability after treatment with AR-A014418 or siGSK-3β were also examined. As expected, both AR-A014418 and siGSK-3β markedly inhibited the invasive ability (Fig. 8A). The present study also examined changes in EMT-related markers, including E-cadherin and Vimentin. The results showed (Fig. 8B) that AR-A014418 or siGSK-3β markedly increased the expression of E-cadherin and decreased the expression of Vimentin, indicating that the EMT process was inhibited.