AICAR was obtained from Abcam (Cambridge, UK) (ab120358, Lot: APN12423-1-1;) for cell experiments, and LC Laboratories (Woburn, MA, USA) (A-1098, Lot: ACR-101) for animal experiments. AICAR was dissolved in distilled water and immediately stored at −20°C. This stock solution was diluted in culture medium for in vitro or saline for in vivo experiments, immediately before use.

Two human osteosarcoma cell lines (MG63 and KHOS) were used in this study. MG63 cells were obtained from the RIKEN BRC through the National Bio-Resource Project of the MEXT (Ibaraki, Japan), and KHOS cells were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). Both cell lines were routinely cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) and 100 U/ml penicillin/streptomycin solution (all from Sigma-Aldrich, St. Louis, MO, USA) at 37°C in a humidified 5% CO₂ atmosphere. For all experiments, we used DMEM containing 10% FBS without the antibiotic solution.

All animal experiments were approved by Kobe University Animal Experimentation Regulations (permission no. P-140504). Male BALB/c nude mice, aged 5 weeks, were purchased from CLEA Japan Inc. (Tokyo, Japan) and were maintained in a facility under specific pathogen-free conditions. Animals were fed pathogen-free laboratory chow and allowed free access to autoclaved water in an air-conditioned room with a 12-h light/dark cycle. For in vivo experiments, both MG63 and KHOS osteosarcoma cells were implanted into the dorsal, subcutaneous area of mice (n=12 for each cell line) at a dose of 2.×10⁶ cells in 500 µl phosphate-buffered saline (PBS), as previously described (21), and mice were randomly divided into two groups: the AICAR-treated group (n=6 for each cell line) and the control group (n=6 for each cell line). Treatment with 450 mg/kg AICAR or saline (as a control) by intraperitoneal injection was started after 1 week of cell implantation and was performed every day for 2 weeks. Tumor volume was calculated twice a week, as previously described, according to the formula V=π/6xa²xb, where a and b represent the shorter and longer dimensions of the tumor, respectively (21). The body weight of each mouse was also monitored throughout the experiment. At the end of the experiment, all tumors were excised and immediately stored at −80°C, and mitochondrial proliferation and apoptotic activity in tumor tissues were evaluated by flow cytometry, immunoblotting, and immunofluorescence staining.

Cell lysates were extracted from cells or implanted tumors using whole cell lysis buffer (Mammalian Protein Extraction reagent; Thermo Scientific, Rockford, IL, USA) and protein content was quantified using BCA protein assay reagent (Bio-Rad, Hercules, CA, USA). Samples containing equal amounts of protein were electrophoresed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) on 7.5–15% gradient gels and transferred to polyvinylidene fluoride membranes. After blocking, membranes were incubated overnight at 4°C with primary antibodies in CanGet Signal Solution 1 (Toyobo Co., Ltd., Osaka, Japan). The following primary antibodies were used in this study: anti-human phospho-AMPKα (Thr172) antibody (1:1,000) (2531S, Lot: 12), anti-human AMPKα antibody (1:1,000) (2532S, Lot: 19) (both from Cell Signaling Technology, Danvers, MA, USA), anti-human PGC-1α antibody (1:1,000) (H000110891-M12, Lot: 08162-3G11), anti-human TFAM antibody (1:1,000) (H00007019-B01P, Lot: E1281) (both from Abnova, Walnut, CA, USA), anti-human poly(ADP-ribose) polymerase (PARP) antibody (1:1,000) (9542S, Lot: 13), anti-human cleaved PARP antibody (1:1,000) (5625S, Lot: 8), anti-human caspase-3 antibody (1:1,000) (9668S, Lot: 7), anti-human cleaved caspase-3 antibody (1:1,000) (9664S, Lot: 20), anti-human caspase-9 antibody (1:1,000) (9502S, Lot:8), anti-human cleaved caspase-9 antibody (1:500) (7237S, Lot: 1) (all from Cell Signaling Technology), and anti-human α-tubulin antibody (1:10,000) (T9026, Lot: 092M4792; Sigma-Aldrich). After washing, membranes were incubated with the appropriate secondary antibody conjugated to horseradish peroxidase and were exposed with ECL Plus Western Blot Detection system reagents (GE Healthcare Biosciences, Piscataway, NJ, USA), and protein expression was detected using a Chemilumino Analyzer LAS-3000 mini (Fujifilm, Tokyo, Japan). Membranes were reprobed with anti-human α-tubulin antibody (Sigma-Aldrich) to confirm equal protein loading. Positive bands in immunoblot analyses were semiquantified by densitometrical analyses using the ImageJ software (version 1.47; NIH, Bethesda, MD, USA) (http://rsbweb.nih.gov/ij/download.html). Values were normalized against α-tubulin and presented as a ratio.

To evaluate mitochondrial numbers in AICAR-treated osteosarcoma cell lines, we examined the relative amount of mtDNA to nuclear DNA (nDNA). Genomic DNA was isolated from the cells using a GenElute Mammalian Genomic DNA Miniprep kit (Sigma-Aldrich). The sequences of the primers designed to amplify a region corresponding to nn 16-408 of a D-loop of human mtDNA were as follows: 5′-GCAGATTTGGGTACCACCCAAGTATTGACTCACCC-3′ (forward) and 5′-GCATGGAGAGCTCCCGTGAGTGGTTAATAGGGTGATAG-3′ (reverse). The PCR conditions were as follows: 1 cycle at 95°C for 15 min, followed by 30 cycles at 95°C for 30 sec, 58°C for 30 sec, and 72°C for 90 sec (22). The relative amount of mtDNA to nDNA was then analysed using the ΔΔCt method.

To evaluate the effects of AICAR on in vitro osteosarcoma cell proliferation, we performed WST-8 cell proliferation assays using a cell counting kit-8 (CCK-8; Dojindo Inc., Kumamoto, Japan). Cells were seeded in 96-well culture plates at a density of 5,000 cells/well in 100 µl medium and incubated with various concentrations of AICAR (0–2,000 µM). At the indicated incubation times, 10 µl of the CCK-8 solution was added into each well and incubated for 1 h. Then, the optical density was measured at a wavelength of 450 nm using a Model 680 Microplate Reader (Bio-Rad), and the relative number of viable cells in each well was calculated.

Immunofluorescence staining was performed to verify the relationship between mitochondrial proliferation and cellular apoptosis in AICAR-treated osteosarcoma cells and implanted tumors using an APO-Direct kit (BD Pharmingen, Franklin Lakes, NJ, USA) and a MitoTracker Deep-Red FM (Invitrogen, Carlsbad, CA, USA) according to the manufacturers' protocols. In vitro, cells were incubated with or without AICAR and then fixed in 4% paraformaldehyde for 30 min at room temperature. Cells were incubated in the prepared DNA Labelling Solution (APO-Direct kit; BD Pharmingen) for 1 h, and were then incubated in the MitoTracker Deep-Red FM (Invitrogen) for 30 min. Nuclear staining was performed using a propidium iodide, and stained cells were assessed using a BZ-8000 confocal microscope (Keyence, Osaka, Japan). In vivo, tumor tissue samples were embedded in an OCT compound (Sakura Finetek Co., Tokyo, Japan), and 10-µm-thick sections were prepared on a cryostat and stored frozen at −80°C. Sections were incubated with anti-actin antibodies (Sigma-Aldrich) diluted in PBS for 30 min at 37°C. After washing, sections were incubated with the APO-Direct kit reagents (BD Pharmingen) and the MitoTracker Deep-Red FM (Invitrogen) in PBS for 30 min in a dark humid chamber at 37°C. The nucleus was stained with the 4′,6-diamidino-2-phenylindole (DAPI). Fluorescence images were obtained using a BZ-8000 confocal microscope (Keyence), and positive staining was semiquantified using ImageJ software.

We also assessed apoptotic activity in AICAR-treated cells or implanted tumors by flow cytometry. Briefly, cells were collected from cultured cells or implanted tumors, suspended in 1% paraformaldehyde in PBS, and resuspended in ice-cold ethanol at a concentration of 1×10⁶ cells/ml. Each cell pellet was labelled using the APO-Direct kit (BD Pharmingen) according to the manufacturers' protocols. Fluorescent intensity was analysed using a BD FACSVerse (BD Biosciences, Franklin Lakes, NJ, USA).

Each experiment was performed independently at least three times. All values are expressed as the mean ± standard error of the mean (SEM). Comparisons between two continuous values were made using one-way analysis of variance. Post-hoc analysis was performed by Fisher's protected least significant difference test. For distributed data, p-values <0.05 were considered significant.

We first examined the effects of AICAR on AMPK phosphorylation and mitochondrial biogenesis in osteosarcoma cells in vitro. As shown in Fig. 1, the expression of phosphorylated AMPKα (Thr172) was strongly increased immediately after AICAR treatment in both MG63 and KHOS osteosarcoma cells (Fig. 1A and B). Additionally, AICAR significantly increased the relative number of mtDNA copies in both cell lines (Fig. 1C). Consistent with these results, the expression levels of the mitochondria-related genes, PGC-1α and TFAM, were increased in AICAR-treated osteosarcoma cells compared with those in control cells (Fig. 1D). These findings strongly suggested that AICAR could induce AMPK activation and increase mitochondrial biogenesis through the PGC-1α/TFAM pathway in osteosarcoma cells.

To test the effects of mitochondrial proliferation by AICAR on osteosarcoma cell growth and apoptosis, we assessed cell viability and apoptotic activity in osteosarcoma cells after AICAR treatment. AICAR showed concentration-dependent inhibitory effects on cell viability in both osteosarcoma cell lines (Fig. 2A). Flow cytometric analyses revealed that AICAR strongly increased the number of apoptotic cells (Fig. 2B), and immunoblot analyses demonstrated that cleaved forms of caspase-9, caspase-3 and PARP were strongly increased in osteosarcoma cells after 72 h of treatment with 1,000 µM AICAR (Fig. 2C). Additionally, immunofluorescence staining revealed that AICAR treatment increased the apoptotic cell numbers along with increased mitochondrial proliferation in both osteosarcoma cells (Fig. 3). Semi-quantitative analyses of immunofluorescence positive staining showed that apoptotic cells significantly increased in AICAR-treated cells, and the relative positive staining to control was a 6.87- and a 24.6-fold increase in MG63 and KHOS cells, respectively (p<0.05). Positive staining of mitochondria also significantly increased after AICAR treatment, and the relative positive staining to control was a 3.43- and a 10.7-fold increase in MG63 and KHOS cells, respectively (p<0.05). These observations suggested that AICAR-dependent mitochondrial proliferation induced mitochondrial apoptosis in human osteosarcoma cells.

We evaluated the in vivo antitumor activity of AICAR using human osteosarcoma xenografts. In both osteosarcoma cell implanted models, AICAR significantly suppressed in vivo osteosarcoma tumor growth compared with the growth of control tumors (Fig. 4). At the end of the experiment, tumor volume in the AICAR-treated group was 85.9% in MG63 (Fig. 4A) and 64.6% in KHOS (Fig. 4B) of the volume in each control group (p<0.05). No significant loss in body weight was observed during the experimental period in either osteosarcoma cell model (Fig. 4). In flow cytometry (Fig. 5A) and immunoblot analyses (Fig. 5B), increased apoptotic activity was observed in AICAR-treated KHOS tumor tissues, and the expression levels of the mitochondria-related genes, PGC-1α and TFAM, were also increased in AICAR-treated KHOS tumors (Fig. 5C). Additionally, in immunofluorescence staining, apoptotic cells and mitochondrial proliferation were observed in AICAR-treated both osteosarcoma tumor tissues, consistent with the results from in vitro experiments (Fig. 5D and E). These findings suggested that AICAR could induce apoptosis via mitochondrial proliferation in human osteosarcoma in vivo.