PDK1-dependent metabolic reprogramming regulates stemness and tumorigenicity of osteosarcoma stem cells through ATF3

PDK1-dependent metabolic reprogramming regulates stemness and tumorigenicity of osteosarcoma stem cells through ATF3 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article PDK1-dependent metabolic reprogramming regulates stemness and tumorigenicity of osteosarcoma stem cells through ATF3 Eiichi Hinoi, Kazuya Tokumura, Kazuya Fukasawa, Manami Hiraiwa This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5372467/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Jul, 2025 Read the published version in Cell Death & Disease → Version 1 posted 7 You are reading this latest preprint version Abstract Osteosarcoma stem cells (OSCs) are characterized by their self-renewal and multilineage differentiation abilities, contributing to osteosarcoma (OS) malignancy. The Warburg effect describes cancer cells’ preference for glycolysis over mitochondrial oxidative phosphorylation (OXPHOS) for energy production. Unlike differentiated cancer cells, cancer stem cells exhibit unique and diverse metabolic properties depending on the context. This study investigated the metabolic reliance of OSCs and related genes through in silico analyses of clinical OS specimens and in vitro and in vivo genetic and pharmacological analyses. Glycolysis and OXPHOS pathways were more active in OSCs than in non-OSCs at single-cell resolution. Pyruvate dehydrogenase kinase 1 (PDK1), a key enzyme balancing glycolysis and OXPHOS, was upregulated in OSCs and correlated with poor prognosis in patients with OS. Genetic inhibition of PDK1 via RNA interference reduced OSC stemness, tumorigenicity, and glycolysis. Pharmacological inhibition of PDK1 mirrored these genetic effects. Activating transcription factor 3 (ATF3) was identified through screening as a downstream factor of PDK1-regulated OSC properties. Silencing ATF3 reduced OSC stemness, while ATF3 overexpression reversed the stemness reduction caused by PDK1 deficiency. ATF3 expression, glycolysis, and stemness were significantly induced by wild-type PDK1 overexpression but not by a kinase-dead PDK1 mutant in OSCs. Pharmacological inhibition of glycolysis counteracted the upregulation of ATF3 expression and increased stemness in OSCs by PDK1 overexpression. These findings indicate that PDK1 fine-tunes metabolic balance to govern OSC stemness and tumorigenicity through ATF3, suggesting a potential therapeutic approach for targeting OSCs in OS. Health sciences/Diseases/Cancer/Bone cancer Biological sciences/Cancer/Bone cancer PDK1 glycolysis osteosarcoma cancer stem cells Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Osteosarcoma (OS), the most common primary malignant bone tumor, poses a significant risk of bone and lung metastases 1 – 3 . OS incidence peaks in adolescents, young adults, and adults aged 65 and above 1 – 3 . It is highly malignant, invasive, progresses rapidly, and has a high mortality rate 1 . The 5-year survival rate is about 70% without metastases but drops to 25% with metastatic disease 2 . Chemotherapy with drugs like cisplatin, doxorubicin, and methotrexate improves survival but is not suitable for all patients 3 . The exact cell origin of OS is unclear, but it likely arises from osteoblast lineage cells derived from mesenchymal stem cells (MSCs) 1 . Osteosarcoma stem cells (OSCs), defined by their self-renewal and multilineage differentiation abilities, significantly contribute to tumor initiation, recurrence, metastases, and chemoresistance, highlighting the need for novel OSC-targeted therapies 4 – 6 . Cancer cells produce energy differently from normal cells. The Warburg effect describes cancer cells’ preference for glycolysis over mitochondrial oxidative phosphorylation (OXPHOS) for energy, even with sufficient oxygen supply 7 – 9 . Unlike differentiated cancer cells, cancer stem cells (CSCs) exhibit unique and diverse metabolic properties based on cancer type, tissue type, and microenvironment 10 . OXPHOS is the primary energy source for many CSC types, including those in leukemia, glioblastoma, and lung, breast, ovarian, and pancreatic cancers 11 – 16 . However, some CSC types prefer glycolysis to maintain their stem cell traits 17 – 19 . Contradictory findings exist for the same tumor types (e.g., glioblastoma and breast cancer) regarding CSCs’ reliance on glycolysis or OXPHOS for sustaining their characteristics and tumorigenicity 12 , 20 , 17 , 21 , 14 , 18 . During glycolysis, glucose is converted to pyruvate, which further undergoes anaerobic or aerobic conversion to lactate or acetyl-coenzyme A (acetyl-CoA) 22 . Pyruvate dehydrogenase kinase 1 (PDK1), a Ser/Thr kinase, serves as a gatekeeper in balance between glycolysis and OXPHOS 23 , 24 . Phosphoglycerate kinase 1 (PGK1) phosphorylates PDK1 at Thr338, activating PDK1 to phosphorylate and inhibit the pyruvate dehydrogenase (PDH) complex, responsible for converting pyruvate to acetyl-CoA 25 . PDK1 is associated with cell proliferation, metastasis, and poor prognosis in breast cancer, glioblastoma, and leukemia, among others 26 – 28 . In addition, it regulates energy metabolism in normal stem cells such as hematopoietic stem cells (HSCs) and is enriched in breast CSCs, playing a key role in their reprogramming 29 , 30 . However, the role of PDK1 in OSCs and its mechanism in maintaining stemness and tumorigenicity remain unclear. Targeting OSCs is a promising strategy for improving OS treatment 5 . Understanding the energy metabolic preferences and underlying molecular mechanisms of OSCs is essential for developing new OS therapies. Although some studies have explored metabolic pathways in OSCs, data on their metabolic dependency at single-cell resolution of clinical specimens and the genes involved in OS malignancy are limited. This study aimed to investigate the energy metabolism preferences and candidate genes in OSCs associated with poor clinical outcomes in OS through bioinformatics analyses of clinical OS specimens at single-cell resolution and in vitro and in vivo genetic and pharmacological analyses. Material and methods Cell culture HEK293T cells (RIKEN Cell Bank, BioResource Center, Tsukuba, Japan) were cultured in DMEM (FUJIFILM Wako Pure Chemical Co., Osaka, Japan) with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA) and 1% penicillin/streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) at 37°C in an atmosphere containing 5% CO 2 . The human OS cell lines 143B (ATCC, Manassas, VA, USA) and MG-63 (RIKEN Cell Bank) were maintained in DMEM supplemented with 10% FBS, 110 µg/mL sodium pyruvate (FUJIFILM Wako Pure Chemical), and 1% penicillin/streptomycin at 37°C in a 5% CO 2 atmosphere. For OSCs, cells were cultured in tumorsphere medium (DMEM/F12 (FUJIFILM Wako Pure Chemical) with B27 supplement without vitamin A (Thermo Fisher Scientific), GlutaMAX (Thermo Fisher Scientific), 20 ng/mL recombinant human EGF (FUJIFILM Wako Pure Chemical), 20 ng/mL recombinant human basic FGF (FUJIFILM Wako Pure Chemical), and 1% penicillin/streptomycin) at 37°C in a 5% CO 2 atmosphere 31 . Lentiviral supernatant preparation and infection HEK293T cells were transfected with lentiviral, packaging, and envelope plasmids using calcium phosphate to produce lentiviral particles 32 . Fourteen hours post-transfection, the medium was replaced with fresh DMEM containing 10% FBS. Forty-eight hours later, the lentiviral supernatant was collected, infecting 143B and MG-63 cells for 24 h. The following plasmids were used: short hairpin (sh) PDK1 #1 (Sigma-Aldrich, St. Louis, MO, USA, TRCN0000006261), sh PDK1 #2 (Sigma-Aldrich, TRCN0000006262), sh Activating transcription factor 3 ( ATF3 )#1 (Sigma-Aldrich, TRCN0000013572), sh ATF3 #2 (Sigma-Aldrich, TRCN0000329689), pLKO.1 puro plasmid (Addgene, #8453), control vector (Vector Builder, Chicago, IL, USA, # VB010000-9389rbj), PDK1 WT (Vector Builder, VB900004-5579neb), and ATF3 (Vector Builder, VB900000-5162ype). The PDK1 T358A mutant was generated from PDK1 WT using the QuikChange II XL site-directed mutagenesis kit (Agilent, Santa Clara, CA, USA). Nucleotide sequences of PDK1 WT and PDK1 T358A are analyzed using FinchTV software (ver. 1.5.0, Geospiza Inc, Seattle, WA, USA). Western blotting Western blotting Proteins were extracted by lysing cells in a buffer containing 1% nonidet P-40 and a protease inhibitor cocktail. Proteins were separated via SDS-PAGE, transferred onto PVDF membranes, and subjected to blotting 33 . The following antibodies were used. Primary antibodies: anti-PDK1 (1:1,000, Cell Signaling Technology, CST, Danvers, MA, USA, #3820), anti-PDH (1:1,000, CST, #2784), anti-p-PDH (1:1,000, CST, #31866), anti-SOX2 (1:1,000, CST, #14962), anti-c-Myc (1:1,000, CST, #5605), anti-ATF3 (1:500, Santa Cruz Biotechnology, SCBT, TX, USA, #sc-188), and anti-β-actin (1:2,000, SCBT, #sc-47778). Secondary antibodies: antirabbit IgG, HRP-linked (1:2,000, CST, #7074), and antimouse IgG, HRP-linked (1:4,000, CST, #7076). Protein levels were quantified using ImageJ. Sphere formation assay and in vitro limiting dilution assay Cells were seeded at 1,000 cells per well in ultralow attachment 96-well plates (Corning Incorporated, Corning, NY, USA) and cultured in tumorsphere medium with 1% methylcellulose (FUJIFILM Wako Pure Chemical) 31 . After 7 days, the number of spheres formed was counted. Sphere formation efficiency was assessed by counting spheres with > 50 µm diameter. For in vitro limiting dilution assays, cells were plated in 96-well plates at densities of 1, 5, 10, 20, 40, or 80 cells/well (10 replicates each). The presence of a sphere was observed on day 7 and was analyzed using the “statmod” package (ver. 1.5.0) in R software (ver. 4.3.0). MTT, apoptosis, and wound healing assays MTT, apoptosis, and wound healing assays were performed to assess cell characteristics 34 . For the MTT assay, 200 µL of 0.5 mg/mL MTT solution in PBS was added to each well on each measurement day and incubated for 4 h. Subsequently, 200 µL of 0.04 mol/L HCl in isopropanol was added to dissolve formazan crystals, and absorbance was measured at 550 nm with a microplate reader. For the wound healing assay, wounds were created using sterile 100 µL pipette tips, and images were taken at 0 and 24 h using a BZ-X810 Analyzer (KEYENCE, Osaka, Japan). The wound closure area-to-wound area ratio (migration rate) was calculated using ImageJ. For the apoptosis assay, cells were stained with PE-labeled Annexin V (1:50, BD Biosciences, BD, San Jose, CA, USA, #560930) and 7-AAD (1:1,000, BD, #559925) for 30 min at 4°C in the dark, followed by analysis using a CytoFLEX S flow cytometer (Beckman Coulter, Brea, CA, USA). Measurement of glucose uptake, lactate production, and the ADP/ATP ratio Glucose uptake, lactic acid production, and the ADP/ATP ratio were assessed using Glucose Assay Kit-WST (Dojindo, Kumamoto, Japan), Lactate Colorimetric/Fluorometric Assay Kit (BioVision, SF, USA), and ADP/ATP Ratio Assay Kit (Sigma-Aldrich), following the manufacturers’ protocols. Chemicals and reagents 2,2-Dichloroacetophenone (DAP) was obtained from Tokyo Chemical Industry (Tokyo, Japan), and FX11 was obtained from Selleck (Houston, TX, USA). In vivo experiments For in vivo studies, 143B cells (5 x 10 6 ) were subcutaneously injected into BALB/cSlc-nu/nu mice (SLC, Hamamatsu, Japan). Tumor volume (calculated as length × width 2 /2) was measured every 5 days with a digital caliper, with an endpoint set at a tumor diameter of 20 mm. For immunohistochemistry, the following antibodies were used: anti-SOX2 (1:300, CST, #14962) as a primary antibody, Alexa Fluor 546 goat antirabbit IgG (1:400, Invitrogen, CA, USA, #A11035) as a secondary antibody, and DAPI (1:1,000, Dojindo, #340–07971) for nuclear staining. Immunofluorescence was visualized using a BZ-X800 Analyzer (KEYENCE). scRNA-seq data analysis The GSE152048 dataset was analyzed using the “Seurat” package (ver. 5.1.0) in R software (ver. 4.3.0) 35 . During preprocessing, cells with mitochondrial RNA content ≥ 10%, expressed genes > 7,500, or < 300 were filtered as low-quality. single-sample Gene Set Enrichment Analysis (ssGSEA) was performed on tumor cell clusters using the “GSVA” package (ver. 1.50.5) 36 , 37 . OSC cells were identified as the top 1% in ssGSEA scores, utilizing the MALTA_CURATED_STEMNESS_MARKERS gene set (OSC; n = 226, non-OSC; n = 22,392). Differentially expressed genes (DEGs; p < 0.05) were identified between groups with the Wilcoxon rank-sum test using the “presto” package (ver 1.0.0). Gene set enrichment analysis (GSEA) was performed with the “clusterProfiler” package (ver 4.8.3), with gene sets downloaded through the “msigdbr” package (ver 7.5.1) 38 , 39 . Results were visualized using the “enrichplot” package (ver 1.20.3) and “ggplot2” (ver 3.4.4). Bulk RNA-seq data analysis For 143B OSCs RNA-seq, cells were lysed and total RNA was extracted using FastGene RNA Premium Kit (NIPPON Genetics, Tokyo, Japan) per the manufacturer’s instructions. Paired-end sequencing was performed on the Illumina NovaSeq 6000 platform. For the RNA-seq analysis of 143B, PRJNA539828, and GSE126209 datasets, trimming was conducted using “Trimmomatic” (ver. 0.39), followed by quality checks with “FASTQC” (ver 0.12.1) 40 . Alignment to the Homo sapiens genome (GRCh38.p14) was performed using “STAR” (ver. 2.7.11a), and expression levels were calculated from the resulting BAM files using “RSEM” (ver. 1.3.3). Survival analysis Survival data for patients with OS were retrieved from the TARGET-OS, TCGA-SARC, and GSE21257 datasets 41 . Survival analysis was carried out using the log-rank test in the “survival” package (ver. 3.5-5), and Kaplan–Meier curves were generated using the “survminer” package (ver. 0.4.9). Statistical analysis Statistical analyses were performed using R software (ver. 4.3.0), and significance was indicated in the results (* p < 0.05, ** p < 0.01, *** p < 0.001, # p < 0.05, ## p < 0.01, ### p < 0.001). Unless otherwise specified, results are presented as mean ± standard deviation (SD). Comparisons between two groups were evaluated using Student’s t- or Wilcoxon test, whereas comparisons among three or more groups were conducted using one-way or two-way ANOVA followed by Tukey–Kramer test, or one-way ANOVA followed by Dunnett’s test. A p -value < 0.05 was considered statistically significant. Results Enrichment of glycolysis and OXPHOS pathways in OSC We analyzed a scRNA-seq dataset (GSE152048) of clinical OS specimens from five conventional and primary patients to profile the energy metabolic properties of OSCs (Fig. 1 A) 35 . Six clusters were identified through t-SNE analysis based on genetic profiles (Fig. 1 B). Canonical markers were used to annotate different cell types: malignant cells ( COL1A1 , IBSP ), myeloid cells ( CD74 , CD14 ), osteoclasts ( CTSK , MMP9 ), pericytes ( RGS5 , ACTA2 ), endothelial cells ( PECAM1 , VWF ), and tumor-infiltrating lymphocytes ( CD3D , NKG7 ) (Fig. 1 C). The OS cell population was divided into OSCs and non-OSCs by ssGSEA (Fig. 1 D). The enrichment of two gene sets associated with “stemness” in OSCs was confirmed by GSEA (Fig. 1 E), defining these cells as the OSC population. We performed GSEA on HALLMARK gene sets to characterize OSCs, finding enrichment in “glycolysis” and “OXPHOS” gene sets in OSCs compared to non-OSCs (Fig. 1 F). Similarly, GSEA in the REACTOME gene sets showed that both “glycolysis” and “OXPHOS” were enriched in these pathways in OSCs (Figs. 1 G and 1 H). Thus, scRNA-seq analysis of clinical OS specimens suggests activation of glycolysis and OXPHOS pathways in OSCs. PDK1 is upregulated in OSC and linked to prognosis in patients with OS Using scRNA-seq data (GSE152048), we identified DEGs in early energy metabolic pathways in OSCs (Fig. 2 A). The expression levels of SLC2A1 , PFKM , PDHA1 , PDK1 , and LDHA were significantly increased in OSCs, while PKM was significantly decreased in OSCs (Fig. 2 B). The association of the expression levels of these metabolic genes with the survival Kaplan–Meier analysis of patients was assessed using the TARGET-OS database. It showed that higher PDK1 expression correlated with shorter survival than those with lower PDK1 expression being the only gene correlating significantly with the prognosis of patients with OS (Fig. 2 C). High PDK1 expression also predicted poorer overall and metastasis-free survival in patients with OS according to the microarray dataset (GSE21257) (Fig. 2 D and Supplementary Fig. 1A) and was linked to poor prognosis of patients with soft tissue sarcoma (TCGA-SARC) (Supplementary Fig. 1B). PDK1 was significantly upregulated in clinical OS tissues compared to nontumor tissues in two bulk RNA-seq datasets (PRJNA539828, GSE126209) (Figs. 2 E and 2 F). Among the early stage of the energy metabolic pathway genes, PDK1 was the only gene that fulfilled the significantly upregulated in OSCs and the consistently correlated with poor prognosis in patients with OS across cohorts (Fig. 2 G), suggesting its potential role in OS progression. OSC favors glycolysis via PDK1 overexpression To validate our bioinformatics analyses of clinical OS specimens (Figs. 1 and 2 ), we cultured 143B cells, a patient-derived OS cell line, in tumorsphere condition (for OSCs; referred to as “143B OSCs”), followed by RNA-seq (Fig. 3 A). We previously demonstrated that 143B OSCs represent stemness properties and tumorigenicity in vitro and in vivo 31 . We identified DEGs related to the early stage of energy metabolic pathways in 143B OSCs (Fig. 3 B). The expression of PDK1 was significantly upregulated in 143B OSCs rather than in 143B non-OSCs (Fig. 3 C), as observed in the OSC population of clinical OS specimens (Fig. 2 B), along with significant upregulation of stem cell markers, such as SOX2 , KLF4 , and ABCG1 (Fig. 3 D) 42 – 46 . Moreover, the protein level of PDK1 was significantly upregulated in both 143B and MG-63 (also patient-derived OS cell lines) OSCs, concomitant with higher protein levels of stem cell markers SOX2 and c-MYC (Fig. 3 E) 42 , 43 , 47 . Additionally, glucose uptake and lactate production were significantly increased in 143B OSCs (Figs. 3 F and 3 G), while the ADP/ATP ratio was significantly decreased (Fig. 3 H), indicating glycolysis activation in OSCs over differentiated OS cells. These findings suggest that OSCs may favor glycolysis through PDK1 overexpression, highlighting PDK1’s role in regulating OSC stem cell phenotypes and tumorigenicity. Genetic inhibition of PDK1 reduces OSC stemness and glycolysis in vitro We targeted PDK1 in 143B and MG-63 OSCs using lentiviral shRNAs (sh PDK1 #1 and sh PDK1 #2). Knockdown of PDK1 significantly reduced SOX2 and c-MYC protein levels in both cell lines, along with PDK1 itself (Fig. 4 A). Sphere formation assays showed that PDK1 silencing markedly decreased tumorsphere formation in both 143B and MG-63 OSCs (Fig. 4 B). An in vitro limiting dilution assay revealed that PDK1 silencing significantly impaired OSC self-renewal (Fig. 4 C). PDK1 targeting reduced cell proliferation (Fig. 4 D) and increased apoptosis in both (Fig. 4 E). Additionally, PDK1 disruption significantly decreased migration potential (Fig. 4 F). PDK1 knockdown also reduced glucose uptake and lactate production while increasing the ADP/ATP ratio in 143B OSCs, indicating reduced glycolysis (Figs. 4 G- 4 I). These findings suggest that PDK1 is crucial for maintaining OSC stem cell properties and metabolic preferences in vitro . Targeting PDK1 reduces OSC tumorigenicity and stemness in vivo Given PDK1’s role in OSC growth, survival, aggressiveness, and self-renewal in vitro , we investigated its effect on tumorigenicity in a xenograft mouse model (Fig. 5 A). Equal numbers of 143B OSCs infected with sh PDK1 or shControl (shCtrl) were transplanted into immunocompromised mice. Mice with sh PDK1 -infected 143B OSCs had significantly lower tumor volume and weight compared to those with shCtrl-infected OSCs (Figs. 5 B and 5 C). Histological analysis showed a significant reduction in SOX2-positive cells in tumors from sh PDK1 -infected OSCs compared to shCtrl-infected OSCs (Fig. 5 D). These results indicate that PDK1 is essential for maintaining tumorigenicity and stemness in vivo . Pharmacological inhibition of PDK1 suppresses OSC stemness and tumorigenicity in vitro and in vivo Our studies indicate that PDK1 in OSCs controls their stemness, tumorigenicity, and energy metabolism, making it a promising OS therapy target. We tested whether DAP, a PDK1 inhibitor, could suppress OSC stemness and tumorigenicity 48 . DAP treatment significantly decreased SOX2 and c-MYC protein levels in 143B and MG63 OSCs without altering PDK1 levels (Fig. 6 A). DAP also reduced tumor sphere formation dose-dependently (Fig. 6 B). Additionally, DAP decreased cell proliferation and increased apoptosis in OSCs (Figs. 6 C and 6 D). In a xenograft mouse model, DAP was intraperitoneally administered to immunocompromised mice, which significantly reduced tumor volume and weight and SOX2-positive cells in tumors (Figs. 6 E- 6 H). These results suggest that pharmacological PDK1 inhibition effectively reduces OSC stemness and tumorigenicity, similar to genetic PDK1 inhibition. PDK1-dependent metabolic adaptation governs OSC properties through ATF3 To explore PDK1’s control mechanisms, we identified DEGs in “ PDK1 -knockdown OSCs over control OSCs” and “OSCs over differentiated OS cells” via RNA-seq. We found 174 and 157 downregulated genes in sh PDK1 #1-infected 143B OSCs and sh PDK1 #2-infected 143B OSCs, respectively, and 78 upregulated genes in OSCs (Fig. 7 A). Six overlapping genes included ATF3 , associated with stemness in glioma stem cells (GSCs) 49 , 50 . ATF3 protein levels were significantly higher in 143B OSCs than in differentiated cells (Fig. 7 B) and decreased in PDK1 -knockdown OSCs (Fig. 7 C). ATF3 mRNA level was elevated in OSCs (GSE152048) (Fig. 7 D), and TARGET-OS analysis showed a positive correlation between PDK1 and ATF3 expression in patients with OS (Fig. 7 E). ATF3 silencing in 143B OSCs significantly reduced tumorsphere formation and self-renewal potential (Figs. 7 F and 7 G), along with the reduction of ATF3 protein levels (Fig. 7 H and supplementary Fig. 2A). Conversely, ATF3 overexpression reversed the reduction in tumorsphere formation caused by PDK1 knockdown (Figs. 7 I and 7 J, and supplementary Fig. 2B). PDK1 is activated by phosphorylation at Thr358 by PGK1, leading to PDH inactivation, which converts pyruvate to acetyl-CoA (Fig. 7 K) 25 . Overexpression of wild-type- PDK1 ( PDK1 WT ) significantly increased tumorsphere formation, self-renewal potential, and lactate production in 143B OSCs (Figs. 7 L- 7 N). Conversely, introducing kinase-dead (KD)- PDK1 ( PDK1 T358A ) did not significantly alter these properties (Figs. 7 L- 7 N, Supplementary Fig. 2C). OSC phenotypes were significantly reduced with PDK1 T358A compared to PDK1 WT (Figs. 7 L and 7 M). Further, ATF3 expression was significantly increased by PDK1 WT but not by PDK1 T358A in 143B OSCs, despite elevated PDK1 levels in both OSCs (Fig. 7 O). Pharmacological inhibition of lactate dehydrogenase A by FX11 reversed ATF3 upregulation and tumorsphere formation in PDK1 WT -overexpressing 143B OSCs (Figs. 7 P and 7 Q). These results suggest that PDK1-mediated metabolic reprogramming governs OSC phenotypes partly through ATF3. Discussion ATF3, a member of the ATF/CREB family of basic-leucine zipper transcription factors, modulates various cellular functions, including proliferation, apoptosis, and glucose metabolism 51 . It acts as both an oncogene and tumor suppressor, depending on the tumor type and context 49 – 54 . In HSCs, ATF3 is activated by tumor-primed bone marrow MSCs, redirecting hematopoiesis toward monocytic cell expansion 55 . In GSCs, ATF3 enhances the stemness and tumorigenicity via TGF-β1/Smad2 signaling and promotes resistance to temozolomide by inducing ABCB4, an ABC transporter 49 , 50 . While ATF3’s role in ferroptosis in OS has been noted, its impact on the stemness and tumorigenicity of OSCs and OS pathophysiology remains unexplored 56 . We identified alternative candidate genes ( BEAN1 , C11orf96 , CCN3 , DUSP8 , and NIM1K ) linked to PDK1-dependent maintenance of OSC stemness. Despite the need for further research, we found ATF3 in OSCs as a crucial downstream factor in PDK1-dependent regulation of OSC properties and OS pathogenesis. Extensive studies have explored direct glycolysis inhibition as a potential cancer treatment, but success rates have been low. For instance, hexokinase 2 inhibition with lonidamine did not significantly improve overall survival in several cancers, including breast and lung, and caused elevated toxicity 57 , 58 . Glucose transport inhibitors like silibinin (silybin) also caused critical side effects without significant response in prostate cancer 59 . Combination therapies targeting multiple metabolic pathways may offer advantages over single-agent glycolysis inhibitors. Our study showed that both glycolysis and OXPHOS are activated in OSCs, as revealed by scRNA-seq analysis of OS clinical specimens. Pharmacological inhibition of PDK1 effectively repressed OSC stemness and tumorigenicity in vitro and in vivo. Further studies should investigate whether combining PDK1 and OXPHOS inhibitors, such as metformin, could effectively disrupt OSC stem cell properties and suppress OS malignancy without serious side effects. The metabolic characteristics of CSCs are highly heterogeneous. Unlike non-CSCs, which mainly utilize glycolysis, CSCs rely on either glycolysis or OXPHOS in a context-dependent manner, with contradictory results reported for the same tumor entity 12 , 17 , 14 , 18 . For example, GSCs rely on glycolysis for energy production and survival through increased glucose consumption by upregulating GLUT3, while also depending on OXPHOS via IMP2, an RNA-binding protein 20 , 21 . This discrepancy is attributed to CSC heterogeneity, tumor microenvironment, and experimental strategies 60 , 61 . Although the mechanisms regulating ATF3 expression downstream of PDK1-glycolysis in OSCs are unknown, we identified the PDK1-glycolysis-ATF3 axis as a critical regulator of OSC properties and OS malignancy. This is the first preclinical study to reveal the link between metabolic reprogramming in OSCs, their stemness and tumorigenesis, and OS malignancy. Despite extensive research efforts, effective treatments for OS have not advanced significantly in the past 4 decades 1 – 3 . Our findings enhance the understanding of OS pathogenesis and OSC properties. They suggest that targeting energy metabolic reprogramming and the associated genes in OSCs could be a novel and effective approach for developing new treatments for OS in humans. Declarations Data availability The GSE21257, GSE126209, and GSE152048 datasets are deposited in the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). The PRJNA539828 dataset is deposited in the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/). The TARGET-OS and TCGA-SARC datasets are deposited in the Genomic Data Commons data portal (https://portal.gdc.cancer.gov/). Acknowledgements We are grateful for the technical support from the members of the Hinoi lab. This work was partially supported by the Japan Society for the Promotion of Science (20H03407 to E.H.). Author contributions K.T., K.F. and E.H. conceived and designed the project. K.T., K.F. and M.H. performed the experiments and analysis. K.T. and E.H. wrote the manuscript. Ethical approval All animal experiments were approved by the Committees on Animal Experimentation of Gifu Pharmaceutical University and Gifu University and performed in accordance with the guidelines for the care and use of laboratory animals. Competing interest The authors declare no competing interests. References Lin Y-H, Jewell BE, Gingold J, Lu L, Zhao R, Wang LL et al. Osteosarcoma: Molecular Pathogenesis and iPSC Modeling. Trends Mol Med 2017; 23: 737–755. Gianferante DM, Mirabello L, Savage SA. Germline and somatic genetics of osteosarcoma - connecting aetiology, biology and therapy. Nat Rev Endocrinol 2017; 13: 480–491. Gill J, Gorlick R. Advancing therapy for osteosarcoma. Nat Rev Clin Oncol 2021; 18: 609–624. Martins-Neves SR, Sampaio-Ribeiro G, Gomes CMF. Self-Renewal and Pluripotency in Osteosarcoma Stem Cells’ Chemoresistance: Notch, Hedgehog, and Wnt/β-Catenin Interplay with Embryonic Markers. Int J Mol Sci 2023; 24: 8401. Menéndez ST, Gallego B, Murillo D, Rodríguez A, Rodríguez R. Cancer Stem Cells as a Source of Drug Resistance in Bone Sarcomas. J Clin Med 2021; 10: 2621. Yan G-N, Lv Y-F, Guo Q-N. Advances in osteosarcoma stem cell research and opportunities for novel therapeutic targets. Cancer Lett 2016; 370: 268–274. Warburg O. On the origin of cancer cells. Science 1956; 123: 309–314. Liao M, Yao D, Wu L, Luo C, Wang Z, Zhang J et al. Targeting the Warburg effect: A revisited perspective from molecular mechanisms to traditional and innovative therapeutic strategies in cancer. Acta Pharm Sin B 2024; 14: 953–1008. An F, Chang W, Song J, Zhang J, Li Z, Gao P et al. Reprogramming of glucose metabolism: Metabolic alterations in the progression of osteosarcoma. J Bone Oncol 2024; 44: 100521. Chu X, Tian W, Ning J, Xiao G, Zhou Y, Wang Z et al. Cancer stem cells: advances in knowledge and implications for cancer therapy. Signal Transduct Target Ther 2024; 9: 170. Lagadinou ED, Sach A, Callahan K, Rossi RM, Neering SJ, Minhajuddin M et al. BCL-2 inhibition targets oxidative phosphorylation and selectively eradicates quiescent human leukemia stem cells. Cell Stem Cell 2013; 12: 329–341. Vlashi E, Lagadec C, Vergnes L, Matsutani T, Masui K, Poulou M et al. Metabolic state of glioma stem cells and nontumorigenic cells. Proc Natl Acad Sci USA 2011; 108: 16062–16067. Ye X-Q, Li Q, Wang G-H, Sun F-F, Huang G-J, Bian X-W et al. Mitochondrial and energy metabolism-related properties as novel indicators of lung cancer stem cells. Int J Cancer 2011; 129: 820–831. Vlashi E, Lagadec C, Vergnes L, Reue K, Frohnen P, Chan M et al. Metabolic differences in breast cancer stem cells and differentiated progeny. Breast Cancer Res Treat 2014; 146: 525–534. Pastò A, Bellio C, Pilotto G, Ciminale V, Silic-Benussi M, Guzzo G et al. Cancer stem cells from epithelial ovarian cancer patients privilege oxidative phosphorylation, and resist glucose deprivation. Oncotarget 2014; 5: 4305–4319. Sancho P, Burgos-Ramos E, Tavera A, Bou Kheir T, Jagust P, Schoenhals M et al. MYC/PGC-1α Balance Determines the Metabolic Phenotype and Plasticity of Pancreatic Cancer Stem Cells. Cell Metab 2015; 22: 590–605. Daniele S, Giacomelli C, Zappelli E, Granchi C, Trincavelli ML, Minutolo F et al. Lactate dehydrogenase-A inhibition induces human glioblastoma multiforme stem cell differentiation and death. Sci Rep 2015; 5: 15556. Dong C, Yuan T, Wu Y, Wang Y, Fan TWM, Miriyala S et al. Loss of FBP1 by Snail-mediated repression provides metabolic advantages in basal-like breast cancer. Cancer Cell 2013; 23: 316–331. Shen Y-A, Wang C-Y, Hsieh Y-T, Chen Y-J, Wei Y-H. Metabolic reprogramming orchestrates cancer stem cell properties in nasopharyngeal carcinoma. Cell Cycle 2015; 14: 86–98. Janiszewska M, Suvà ML, Riggi N, Houtkooper RH, Auwerx J, Clément-Schatlo V et al. Imp2 controls oxidative phosphorylation and is crucial for preserving glioblastoma cancer stem cells. Genes Dev 2012; 26: 1926–1944. Flavahan WA, Wu Q, Hitomi M, Rahim N, Kim Y, Sloan AE et al. Brain tumor initiating cells adapt to restricted nutrition through preferential glucose uptake. Nat Neurosci 2013; 16: 1373–1382. Stine ZE, Schug ZT, Salvino JM, Dang CV. Targeting cancer metabolism in the era of precision oncology. Nat Rev Drug Discov 2022; 21: 141–162. Jeoung NH. Pyruvate Dehydrogenase Kinases: Therapeutic Targets for Diabetes and Cancers. Diabetes Metab J 2015; 39: 188–197. Atas E, Oberhuber M, Kenner L. The Implications of PDK1–4 on Tumor Energy Metabolism, Aggressiveness and Therapy Resistance. Front Oncol 2020; 10: 583217. Li X, Jiang Y, Meisenhelder J, Yang W, Hawke DH, Zheng Y et al. Mitochondria-Translocated PGK1 Functions as a Protein Kinase to Coordinate Glycolysis and the TCA Cycle in Tumorigenesis. Mol Cell 2016; 61: 705–719. Wei Y, Zhang D, Shi H, Qian H, Chen H, Zeng Q et al. PDK1 promotes breast cancer progression by enhancing the stability and transcriptional activity of HIF-1α. Genes Dis 2024; 11: 101041. Jha MK, Suk K. Pyruvate dehydrogenase kinase as a potential therapeutic target for malignant gliomas. Brain Tumor Res Treat 2013; 1: 57–63. Erdem A, Marin S, Pereira-Martins DA, Cortés R, Cunningham A, Pruis MG et al. The Glycolytic Gatekeeper PDK1 defines different metabolic states between genetically distinct subtypes of human acute myeloid leukemia. Nat Commun 2022; 13: 1105. Halvarsson C, Eliasson P, Jönsson J-I. Pyruvate dehydrogenase kinase 1 is essential for transplantable mouse bone marrow hematopoietic stem cell and progenitor function. PLoS One 2017; 12: e0171714. Peng F, Wang J-H, Fan W-J, Meng Y-T, Li M-M, Li T-T et al. Glycolysis gatekeeper PDK1 reprograms breast cancer stem cells under hypoxia. Oncogene 2018; 37: 1062–1074. Osumi R, Sugihara K, Yoshimoto M, Tokumura K, Tanaka Y, Hinoi E. Role of proteoglycan synthesis genes in osteosarcoma stem cells. Front Oncol 2024; 14: 1325794. Fukasawa K, Lyu J, Kubo T, Tanaka Y, Suzuki A, Horie T et al. MEK5-ERK5 Axis Promotes Self-renewal and Tumorigenicity of Glioma Stem Cells. Cancer Res Commun 2023; 3: 148–159. Yamada T, Fukasawa K, Horie T, Kadota T, Lyu J, Tokumura K et al. The role of CDK8 in mesenchymal stem cells in controlling osteoclastogenesis and bone homeostasis. Stem Cell Reports 2022; 17: 1576–1588. Sasaki Y, Tokumura K, Yoshimoto M, Hinoi E. Association between Fat-Soluble Vitamin Metabolic Process and Glioma Progression. Biol Pharm Bull 2024; 47: 1682–1689. Zhou Y, Yang D, Yang Q, Lv X, Huang W, Zhou Z et al. Single-cell RNA landscape of intratumoral heterogeneity and immunosuppressive microenvironment in advanced osteosarcoma. Nat Commun 2020; 11: 6322. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013; 14: 7. Barbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009; 462: 108–112. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005; 102: 15545–15550. Mootha VK, Lindgren CM, Eriksson K-F, Subramanian A, Sihag S, Lehar J et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003; 34: 267–273. Wang X, Qin G, Liang X, Wang W, Wang Z, Liao D et al. Targeting the CK1α/CBX4 axis for metastasis in osteosarcoma. Nat Commun 2020; 11: 1141. Buddingh EP, Kuijjer ML, Duim RAJ, Bürger H, Agelopoulos K, Myklebost O et al. Tumor-infiltrating macrophages are associated with metastasis suppression in high-grade osteosarcoma: a rationale for treatment with macrophage activating agents. Clin Cancer Res 2011; 17: 2110–2119. Maurizi G, Verma N, Gadi A, Mansukhani A, Basilico C. Sox2 is required for tumor development and cancer cell proliferation in osteosarcoma. Oncogene 2018; 37: 4626–4632. Basu-Roy U, Bayin NS, Rattanakorn K, Han E, Placantonakis DG, Mansukhani A et al. Sox2 antagonizes the Hippo pathway to maintain stemness in cancer cells. Nat Commun 2015; 6: 6411. Qi X-T, Li Y-L, Zhang Y-Q, Xu T, Lu B, Fang L et al. KLF4 functions as an oncogene in promoting cancer stem cell-like characteristics in osteosarcoma cells. Acta Pharmacol Sin 2019; 40: 546–555. Li Y, Xian M, Yang B, Ying M, He Q. Inhibition of KLF4 by Statins Reverses Adriamycin-Induced Metastasis and Cancer Stemness in Osteosarcoma Cells. Stem Cell Reports 2017; 8: 1617–1629. Roundhill EA, Jabri S, Burchill SA. ABCG1 and Pgp identify drug resistant, self-renewing osteosarcoma cells. Cancer Lett 2019; 453: 142–157. van Schaijik B, Davis PF, Wickremesekera AC, Tan ST, Itinteang T. Subcellular localisation of the stem cell markers OCT4, SOX2, NANOG, KLF4 and c-MYC in cancer: a review. J Clin Pathol 2018; 71: 88–91. Qin L, Tian Y, Yu Z, Shi D, Wang J, Zhang C et al. Targeting PDK1 with dichloroacetophenone to inhibit acute myeloid leukemia (AML) cell growth. Oncotarget 2016; 7: 1395–1407. Xu J, Zhang G, Hu J, Li H, Zhao J, Zong S et al. UPF1/circRPPH1/ATF3 feedback loop promotes the malignant phenotype and stemness of GSCs. Cell Death Dis 2022; 13: 645. Xu X, Zheng Y, Luo L, You Z, Chen H, Wang J et al. Glioblastoma stem cells deliver ABCB4 transcribed by ATF3 via exosomes conferring glioblastoma resistance to temozolomide. Cell Death Dis 2024; 15: 318. Ku H-C, Cheng C-F. Master Regulator Activating Transcription Factor 3 (ATF3) in Metabolic Homeostasis and Cancer. Front Endocrinol (Lausanne) 2020; 11: 556. Yan L, Gaddis S, Coletta LD, Repass J, Powell KL, Simper MS et al. ATF3-Induced Mammary Tumors Exhibit Molecular Features of Human Basal-Like Breast Cancer. Int J Mol Sci 2021; 22: 2353. Jeon T, Ko MJ, Seo Y-R, Jung S-J, Seo D, Park S-Y et al. Silencing CDCA8 Suppresses Hepatocellular Carcinoma Growth and Stemness via Restoration of ATF3 Tumor Suppressor and Inactivation of AKT/β-Catenin Signaling. Cancers (Basel) 2021; 13: 1055. Huang C, Chen R, Zheng F, Tang Y, Wang X, Chen Z et al. Inhibitory role of ATF3 in gastric cancer progression through regulating cell EMT and stemness. Cancer Cell Int 2021; 21: 127. Perrone M, Chiodoni C, Lecchi M, Botti L, Bassani B, Piva A et al. ATF3 Reprograms the Bone Marrow Niche in Response to Early Breast Cancer Transformation. Cancer Res 2023; 83: 117–129. Li Y, Yan J, Zhao Q, Zhang Y, Zhang Y. ATF3 promotes ferroptosis in sorafenib-induced cardiotoxicity by suppressing Slc7a11 expression. Front Pharmacol 2022; 13: 904314. Berruti A, Bitossi R, Gorzegno G, Bottini A, Alquati P, De Matteis A et al. Time to progression in metastatic breast cancer patients treated with epirubicin is not improved by the addition of either cisplatin or lonidamine: final results of a phase III study with a factorial design. J Clin Oncol 2002; 20: 4150–4159. De Marinis F, Rinaldi M, Ardizzoni A, Bruzzi P, Pennucci MC, Portalone L et al. The role of vindesine and lonidamine in the treatment of elderly patients with advanced non-small cell lung cancer: a phase III randomized FONICAP trial. Italian Lung Cancer Task Force. Tumori 1999; 85: 177–182. Flaig TW, Gustafson DL, Su L-J, Zirrolli JA, Crighton F, Harrison GS et al. A phase I and pharmacokinetic study of silybin-phytosome in prostate cancer patients. Invest New Drugs 2007; 25: 139–146. Eun K, Ham SW, Kim H. Cancer stem cell heterogeneity: origin and new perspectives on CSC targeting. BMB Rep 2017; 50: 117–125. de Visser KE, Joyce JA. The evolving tumor microenvironment: From cancer initiation to metastatic outgrowth. Cancer Cell 2023; 41: 374–403. Additional Declarations (Not answered) Supplementary Files Originalwesternblotsdata.pdf SupplementalFig.pdf Cite Share Download PDF Status: Published Journal Publication published 29 Jul, 2025 Read the published version in Cell Death & Disease → Version 1 posted Editorial decision: revise 28 Nov, 2024 Review # 1 received at journal 15 Nov, 2024 Reviewer # 1 agreed at journal 14 Nov, 2024 Reviewers invited by journal 14 Nov, 2024 Submission checks completed at journal 04 Nov, 2024 First submitted to journal 01 Nov, 2024 Editor assigned by journal 01 Nov, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5372467","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":378271187,"identity":"0d357eeb-6d62-47b3-bedd-2151254027ef","order_by":0,"name":"Eiichi Hinoi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6ElEQVRIiWNgGAWjYBACA2YgkcDAIMfAzPgAIUqMFmMGZmaEQvxaoHRiAwMzXoUIYM7Oe/DBAwab9PnuzGwSjDsOM/C3H2AoLsCjxbKZL9kggSEtd+NhkJYzhxkkziQwGM/A57DDPGYSif8O525s5j8m/bftMAPDDaDPeAhpSWD4n27YDLIFqEWeSC0HEuSZoVoMiNBiDPRLsuEGZmZmC8Yz6TyGZxIb8Pvl/BnDhz8Y7OTl+w8z3mDcYS0nd/zwMWN8IYbQewBIMDYwAJ3E2GZMjA4G+QaIFhBgfkyUllEwCkbBKBgpAACDdT+XRorCuAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-1298-8293","institution":"Gifu Pharmaceutical University","correspondingAuthor":true,"prefix":"","firstName":"Eiichi","middleName":"","lastName":"Hinoi","suffix":""},{"id":378271188,"identity":"5488bb9a-02d2-4547-a8c1-408783fd70e5","order_by":1,"name":"Kazuya Tokumura","email":"","orcid":"https://orcid.org/0000-0001-8269-1922","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Kazuya","middleName":"","lastName":"Tokumura","suffix":""},{"id":378271189,"identity":"a54713ae-ee59-4410-a530-1ebdec669ab9","order_by":2,"name":"Kazuya Fukasawa","email":"","orcid":"","institution":"Gifu Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Kazuya","middleName":"","lastName":"Fukasawa","suffix":""},{"id":378271190,"identity":"c43cd22e-918b-4553-b164-fba0bd0e6ff7","order_by":3,"name":"Manami Hiraiwa","email":"","orcid":"","institution":"Gifu Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Manami","middleName":"","lastName":"Hiraiwa","suffix":""}],"badges":[],"createdAt":"2024-11-01 10:05:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5372467/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5372467/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41419-025-07903-7","type":"published","date":"2025-07-29T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":71736537,"identity":"1201351b-95cf-4551-b683-b67a6015a467","added_by":"auto","created_at":"2024-12-18 07:36:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":911390,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnergy metabolic properties of OSCs analyzed via scRNA-seq. A \u003c/strong\u003eSchematic illustrating scRNA-seq analysis performed using the GSE152048 dataset; \u003cstrong\u003eB\u003c/strong\u003et-distributed stochastic neighbor embedding (t-SNE) plot depicting six distinct cell clusters; \u003cstrong\u003eC\u003c/strong\u003e feature plots displaying canonical marker genes for identifying six cell clusters: malignant cells (\u003cem\u003eCOL1A1\u003c/em\u003e, \u003cem\u003eIBSP\u003c/em\u003e), myeloid cells (\u003cem\u003eCD74\u003c/em\u003e, \u003cem\u003eCD14\u003c/em\u003e), osteoclasts (\u003cem\u003eCTSK\u003c/em\u003e, \u003cem\u003eMMP9\u003c/em\u003e), pericytes (\u003cem\u003eRGS5,\u003c/em\u003e \u003cem\u003eACTA2\u003c/em\u003e), endothelial cells (\u003cem\u003ePECAM1\u003c/em\u003e, \u003cem\u003eVWF\u003c/em\u003e), and tumor-infiltrating lymphocytes (\u003cem\u003eCD3D\u003c/em\u003e, \u003cem\u003eNKG7\u003c/em\u003e); \u003cstrong\u003eD\u003c/strong\u003e t-SNE plot distinguishing OSCs (n = 226) and non-OSCs (n = 22,392) classified via ssGSEA; \u003cstrong\u003eE\u003c/strong\u003e GSEA plot confirming the enrichment of stemness-related gene sets in OSCs; \u003cstrong\u003eF\u003c/strong\u003e GSEA results for HALLMARK gene sets identify pathways enriched in OSCs; \u003cstrong\u003eG\u003c/strong\u003e and\u003cstrong\u003e H\u003c/strong\u003eGSEA of REACTOME gene sets reveal significant enrichment of glycolysis (\u003cstrong\u003eG\u003c/strong\u003e) and oxidative phosphorylation (OXPHOS) (\u003cstrong\u003eH\u003c/strong\u003e) pathways in OSCs\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/68744607a01fac352fcb3930.png"},{"id":71736539,"identity":"9632e475-295e-4863-8843-beb0e08b11ed","added_by":"auto","created_at":"2024-12-18 07:36:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":476092,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eUpregulation of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePDK1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e expression in OSCs and its correlation with poor prognosis in patients with OS. A\u003c/strong\u003eSchematic representing genes involved in early energy metabolism stages; \u003cstrong\u003eB\u003c/strong\u003emRNA expression levels of genes related to early energy metabolism pathways in OSCs (n = 226) and non-OSCs (n = 22,392) within the GSE152048 dataset (Wilcoxon test, mean ± standard error); \u003cstrong\u003eC\u003c/strong\u003e and \u003cstrong\u003eD\u003c/strong\u003e. Kaplan–Meier survival curves showing overall survival for patients with OS with high and low expression of early energy metabolism genes; \u003cstrong\u003eC\u003c/strong\u003e TARGET-OS cohort (high; n = 43, low; n = 42, Log-rank test); \u003cstrong\u003eD\u003c/strong\u003eGSE21257 cohort (high; n = 20, low; n = 20, Log-rank test). \u003cstrong\u003eE\u003c/strong\u003e and \u003cstrong\u003eF\u003c/strong\u003e \u003cem\u003ePDK1\u003c/em\u003e mRNA expression levels in OS and nontumor tissues; \u003cstrong\u003eE\u003c/strong\u003e PRJNA539828 dataset (OS tissues; n = 16, nontumor tissues; n = 4, Wilcoxon test); \u003cstrong\u003eF\u003c/strong\u003e GSE126209 dataset (OS tissues; n = 10, nontumor tissues; n = 9, Wilcoxon test); \u003cstrong\u003eG\u003c/strong\u003e summary table illustrating the relationship between early energy metabolism gene expression and survival outcomes; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/d47ff6a7509e764f6d2eee69.png"},{"id":71737957,"identity":"5c696f43-fdad-4cba-a163-f78869d6c816","added_by":"auto","created_at":"2024-12-18 07:44:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":246443,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePreference of OSCs for glycolysis through PDK1 upregulation. A\u003c/strong\u003e–\u003cstrong\u003eD\u003c/strong\u003eBulk RNA-seq analysis comparing OSCs (n = 3) and non-OSCs (n = 3) derived from 143B cells; \u003cstrong\u003eA\u003c/strong\u003e schematic of culture conditions for OSCs and non-OSCs; \u003cstrong\u003eB\u003c/strong\u003e heatmap illustrating expression levels of genes linked to early energy metabolism; \u003cstrong\u003eC\u003c/strong\u003emRNA expression levels of \u003cem\u003ePDK1 \u003c/em\u003e(Student’s t-test); \u003cstrong\u003eD\u003c/strong\u003emRNA levels of stem cell marker genes (Student’s t-test); \u003cstrong\u003eE\u003c/strong\u003e western blot analysis of PDK1 and stem cell markers in OSCs and non-OSCs derived from 143B and MG-63 cells (n = 4, Student’s t-test); \u003cstrong\u003eF\u003c/strong\u003e–\u003cstrong\u003eH\u003c/strong\u003e analysis of glycolytic activity in OSCs and non-OSCs from 143B cells; \u003cstrong\u003eF\u003c/strong\u003e glucose uptake (n = 3, Student’s t-test); \u003cstrong\u003eG\u003c/strong\u003e lactate production (n = 4, Student’s t-test); \u003cstrong\u003eH\u003c/strong\u003e ADP/ATP ratio (n = 4, Student’s t-test); *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/565018eac9b02af908cd42ca.png"},{"id":71737956,"identity":"82ac2468-26f5-4de3-96c4-bfbe5ae92f25","added_by":"auto","created_at":"2024-12-18 07:44:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":698569,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003ePDK1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eknockdown diminishes OSCs stemness and glycolysis \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ein vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. A\u003c/strong\u003e–\u003cstrong\u003eI\u003c/strong\u003e Analysis of \u003cem\u003ePDK1\u003c/em\u003e knockdown in OSCs derived from 143B and MG-63 cells; \u003cstrong\u003eA\u003c/strong\u003e western blot of PDK1\u003cem\u003e \u003c/em\u003eand stem cell markers (n = 3–6, one-way ANOVA followed by Dunnett’s test); \u003cstrong\u003eB\u003c/strong\u003e sphere formation assay (n = 6, one-way ANOVA followed by Dunnett’s test, scale bar: 30 μm); \u003cstrong\u003eC\u003c/strong\u003e \u003cem\u003ein vitro \u003c/em\u003elimiting dilution assay (n = 10); \u003cstrong\u003eD\u003c/strong\u003e MTT assay for cell proliferation (n = 5, two-way ANOVA with Tukey–Kramer test); \u003cstrong\u003eE\u003c/strong\u003e apoptosis assay with PE-Annexin V and 7-AAD staining (n = 3-4, one-way ANOVA followed by Dunnett’s test); \u003cstrong\u003eF\u003c/strong\u003e wound healing assay for migration analysis (n = 3, one-way ANOVA followed by Dunnett’s test, scale bar: 200 μm); \u003cstrong\u003eG\u003c/strong\u003e–\u003cstrong\u003eI\u003c/strong\u003e analysis of glycolytic activity; \u003cstrong\u003eG\u003c/strong\u003e glucose uptake (n = 3, one-way ANOVA followed by Dunnett’s test); \u003cstrong\u003eH\u003c/strong\u003e lactate production (n = 4, one-way ANOVA followed by Dunnett’s test); \u003cstrong\u003eI\u003c/strong\u003e ADP/ATP ratio (n = 4, one-way ANOVA followed by Dunnett’s test); *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/ad661132fab9fdc780710346.png"},{"id":71736536,"identity":"0ba33c43-eb31-4b14-b306-6ec3da14cb20","added_by":"auto","created_at":"2024-12-18 07:36:24","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":461005,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003ePDK1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e knockdown reduces tumorigenicity and stemness \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ein vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. A\u003c/strong\u003e–\u003cstrong\u003eD\u003c/strong\u003e \u003cem\u003ein vivo\u003c/em\u003e analysis using a xenograft model; \u003cstrong\u003eA\u003c/strong\u003e schematic showing subcutaneous transplantation of \u003cem\u003ePDK1\u003c/em\u003e knockdown 143B cells (5 × 10\u003csup\u003e6\u003c/sup\u003e cells/mouse) into 4-week-old female BALB/cSlc-nu/nu mice; \u003cstrong\u003eB\u003c/strong\u003e tumor volume measurement (n = 9–10, two-way ANOVA with Tukey–Kramer test, scale bar: 20 mm); \u003cstrong\u003eC\u003c/strong\u003e tumor weight (n = 9–10, one-way ANOVA followed by Dunnett’s test, scale bar: 10 mm); \u003cstrong\u003eD\u003c/strong\u003e immunohistochemical analysis of SOX2 expression in OS tissues (n = 4, one-way ANOVA followed by Dunnett’s test, scale bar: 30 μm); *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/094227139b2c8ba3cc399511.png"},{"id":71737958,"identity":"e6329a87-80be-467c-946b-53442b603bd4","added_by":"auto","created_at":"2024-12-18 07:44:24","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":608956,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePharmacological inhibition of PDK1 reduces OSC stemness and tumorigenicity. A–I\u003c/strong\u003e Analysis of DAP-treated OSCs derived from 143B and MG-63 cells; \u003cstrong\u003eA\u003c/strong\u003e western blot showing SOX2 and c-MYC protein levels (1 μM DAP, n = 3, Student’s t-test); \u003cstrong\u003eB\u003c/strong\u003e sphere formation assay (n = 6, Student’s t-test, scale bar: 30 μm); \u003cstrong\u003eC\u003c/strong\u003e MTT assay for cell proliferation (1 μM DAP, n = 5, two-way ANOVA with Tukey–Kramer test); \u003cstrong\u003eD\u003c/strong\u003e apoptosis assay with PE-Annexin V and 7-AAD staining (1 μM DAP, n = 3, Student’s t-test); \u003cstrong\u003eE–H\u003c/strong\u003e \u003cem\u003ein vivo\u003c/em\u003e analysis in a xenograft model; \u003cstrong\u003eE\u003c/strong\u003e schematic of subcutaneous transplantation of 143B cells (5 × 10\u003csup\u003e6 \u003c/sup\u003ecells/mouse) and DAP administration in 4-week-old female BALB/cSlc-nu/nu mice; \u003cstrong\u003eF\u003c/strong\u003e tumor volume (n = 10, two-way ANOVA with Tukey–Kramer test, scale bar: 20 mm); \u003cstrong\u003eG\u003c/strong\u003e tumor weight (n = 10, Student’s t-test, scale bar: 10 mm); \u003cstrong\u003eH\u003c/strong\u003e immunohistochemical analysis of SOX2 in tumor tissues (n = 4, Student’s t-test, scale bar: 30 μm); *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/0ec21c287469e042a46b0d47.png"},{"id":71736542,"identity":"5b987abe-5361-4334-91d2-2ebb054d0e44","added_by":"auto","created_at":"2024-12-18 07:36:24","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":545506,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePDK1-dependent metabolic adaptation regulates OSC traits via ATF3. A–J \u003c/strong\u003eand\u003cstrong\u003e L–P \u003c/strong\u003eAnalysis of OSCs derived from 143B cells; \u003cstrong\u003eA\u003c/strong\u003e venn diagram displaying genes upregulated in OSCs and downregulated in \u003cem\u003ePDK1\u003c/em\u003e knockdown OSCs from RNA-seq data; \u003cstrong\u003eB\u003c/strong\u003e western blot analysis comparing OSCs and non-OSCs (n = 3, Student’s t-test); \u003cstrong\u003eC\u003c/strong\u003e western blot analysis in \u003cem\u003ePDK1\u003c/em\u003e knockdown OSCs (n = 3, one-way ANOVA followed by Dunnett’s test); \u003cstrong\u003eD \u003c/strong\u003emRNA expression levels of ATF3 in OSCs (n = 226) and non-OSCs (n = 22,392; Wilcoxn test, mean ± standard error); \u003cstrong\u003eE\u003c/strong\u003e correlation analysis of \u003cem\u003ePDK1\u003c/em\u003e and \u003cem\u003eATF3\u003c/em\u003e expression in the TARGET-OS cohort; \u003cstrong\u003eF \u003c/strong\u003esphere formation assay in \u003cem\u003eATF3 \u003c/em\u003eknockdown OSCs (n = 6, one-way ANOVA followed by Dunnett’s test); \u003cstrong\u003eG \u003c/strong\u003e\u003cem\u003ein vitro\u003c/em\u003e limiting dilution assay for \u003cem\u003eATF3\u003c/em\u003e knockdown OSCs (n = 10). \u003cstrong\u003eH\u003c/strong\u003e western blot analysis in \u003cem\u003eATF3 \u003c/em\u003eknockdown OSCs; \u003cstrong\u003eI \u003c/strong\u003ewestern blot analysis in sh\u003cem\u003ePDK1\u003c/em\u003e/\u003cem\u003eATF3\u003c/em\u003e OSCs; \u003cstrong\u003eJ\u003c/strong\u003e sphere formation assay (n = 6, two-way ANOVA followed by Tukey–Kramer test); \u003cstrong\u003eK \u003c/strong\u003eschematic showing glycolysis regulation by PGK1 and PDK1; \u003cstrong\u003eL\u003c/strong\u003e sphere formation assay in \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003e\u003cem\u003eT358A\u003c/em\u003e\u003c/sup\u003e OSCs (n = 6, one-way ANOVA followed by Tukey–Kramer test); \u003cstrong\u003eM \u003c/strong\u003e\u003cem\u003ein vitro\u003c/em\u003e limiting dilution assay for \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003e\u003cem\u003eT358A\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003eOSCs (n = 10); \u003cstrong\u003eN\u003c/strong\u003e lactate production measurement in \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003e\u003cem\u003eT358A\u003c/em\u003e\u003c/sup\u003e\u003csup\u003e \u003c/sup\u003eOSCs (n = 4, one-way ANOVA followed by Tukey–Kramer test); \u003cstrong\u003eO \u003c/strong\u003ewestern blot analysis in \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eT358A\u003c/sup\u003e OSCs. (n = 4, one-way ANOVA followed by Tukey–Kramer test); \u003cstrong\u003eP \u003c/strong\u003ewestern blot analysis in \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e and FX11-treated OSCs (20 μM FX11, n = 3, Student’s t-test); \u003cstrong\u003eQ \u003c/strong\u003esphere formation assay (20 μM FX11, n = 6, Student’s t-test); scale bar: 30 μm; *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001, \u003csup\u003e#\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, \u003csup\u003e##\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, \u003csup\u003e###\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig.7.png","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/ded4729a1e4aa97662202f02.png"},{"id":89062812,"identity":"cc0ecd99-f8af-4be0-872c-926036f13ddc","added_by":"auto","created_at":"2025-08-14 09:45:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5069783,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/f8843410-7ae4-49ec-b900-dae296133463.pdf"},{"id":71737955,"identity":"e07f7375-89cd-4b3a-88d7-87e54cba0825","added_by":"auto","created_at":"2024-12-18 07:44:24","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":492459,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"Originalwesternblotsdata.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/735dde2b9ddc0cd69bff9898.pdf"},{"id":71736540,"identity":"9a91140d-a4e1-481f-866c-dd17b5ade2b5","added_by":"auto","created_at":"2024-12-18 07:36:24","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":334426,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementalFig.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5372467/v1/7dd7880a387013ad1e8b3e83.pdf"}],"financialInterests":"(Not answered)","formattedTitle":"PDK1-dependent metabolic reprogramming regulates stemness and tumorigenicity of osteosarcoma stem cells through ATF3","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOsteosarcoma (OS), the most common primary malignant bone tumor, poses a significant risk of bone and lung metastases\u003csup\u003e\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. OS incidence peaks in adolescents, young adults, and adults aged 65 and above\u003csup\u003e\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. It is highly malignant, invasive, progresses rapidly, and has a high mortality rate\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. The 5-year survival rate is about 70% without metastases but drops to 25% with metastatic disease\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Chemotherapy with drugs like cisplatin, doxorubicin, and methotrexate improves survival but is not suitable for all patients\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. The exact cell origin of OS is unclear, but it likely arises from osteoblast lineage cells derived from mesenchymal stem cells (MSCs)\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Osteosarcoma stem cells (OSCs), defined by their self-renewal and multilineage differentiation abilities, significantly contribute to tumor initiation, recurrence, metastases, and chemoresistance, highlighting the need for novel OSC-targeted therapies\u003csup\u003e\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCancer cells produce energy differently from normal cells. The Warburg effect describes cancer cells\u0026rsquo; preference for glycolysis over mitochondrial oxidative phosphorylation (OXPHOS) for energy, even with sufficient oxygen supply\u003csup\u003e\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Unlike differentiated cancer cells, cancer stem cells (CSCs) exhibit unique and diverse metabolic properties based on cancer type, tissue type, and microenvironment\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. OXPHOS is the primary energy source for many CSC types, including those in leukemia, glioblastoma, and lung, breast, ovarian, and pancreatic cancers\u003csup\u003e\u003cspan additionalcitationids=\"CR12 CR13 CR14 CR15\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. However, some CSC types prefer glycolysis to maintain their stem cell traits\u003csup\u003e\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Contradictory findings exist for the same tumor types (e.g., glioblastoma and breast cancer) regarding CSCs\u0026rsquo; reliance on glycolysis or OXPHOS for sustaining their characteristics and tumorigenicity\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eDuring glycolysis, glucose is converted to pyruvate, which further undergoes anaerobic or aerobic conversion to lactate or acetyl-coenzyme A (acetyl-CoA)\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Pyruvate dehydrogenase kinase 1 (PDK1), a Ser/Thr kinase, serves as a gatekeeper in balance between glycolysis and OXPHOS\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Phosphoglycerate kinase 1 (PGK1) phosphorylates PDK1 at Thr338, activating PDK1 to phosphorylate and inhibit the pyruvate dehydrogenase (PDH) complex, responsible for converting pyruvate to acetyl-CoA\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. PDK1 is associated with cell proliferation, metastasis, and poor prognosis in breast cancer, glioblastoma, and leukemia, among others\u003csup\u003e\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. In addition, it regulates energy metabolism in normal stem cells such as hematopoietic stem cells (HSCs) and is enriched in breast CSCs, playing a key role in their reprogramming\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. However, the role of PDK1 in OSCs and its mechanism in maintaining stemness and tumorigenicity remain unclear.\u003c/p\u003e \u003cp\u003eTargeting OSCs is a promising strategy for improving OS treatment\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Understanding the energy metabolic preferences and underlying molecular mechanisms of OSCs is essential for developing new OS therapies. Although some studies have explored metabolic pathways in OSCs, data on their metabolic dependency at single-cell resolution of clinical specimens and the genes involved in OS malignancy are limited. This study aimed to investigate the energy metabolism preferences and candidate genes in OSCs associated with poor clinical outcomes in OS through bioinformatics analyses of clinical OS specimens at single-cell resolution and \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e genetic and pharmacological analyses.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCell culture\u003c/h2\u003e \u003cp\u003eHEK293T cells (RIKEN Cell Bank, BioResource Center, Tsukuba, Japan) were cultured in DMEM (FUJIFILM Wako Pure Chemical Co., Osaka, Japan) with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA) and 1% penicillin/streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) at 37\u0026deg;C in an atmosphere containing 5% CO\u003csub\u003e2\u003c/sub\u003e. The human OS cell lines 143B (ATCC, Manassas, VA, USA) and MG-63 (RIKEN Cell Bank) were maintained in DMEM supplemented with 10% FBS, 110 \u0026micro;g/mL sodium pyruvate (FUJIFILM Wako Pure Chemical), and 1% penicillin/streptomycin at 37\u0026deg;C in a 5% CO\u003csub\u003e2\u003c/sub\u003e atmosphere. For OSCs, cells were cultured in tumorsphere medium (DMEM/F12 (FUJIFILM Wako Pure Chemical) with B27 supplement without vitamin A (Thermo Fisher Scientific), GlutaMAX (Thermo Fisher Scientific), 20 ng/mL recombinant human EGF (FUJIFILM Wako Pure Chemical), 20 ng/mL recombinant human basic FGF (FUJIFILM Wako Pure Chemical), and 1% penicillin/streptomycin) at 37\u0026deg;C in a 5% CO\u003csub\u003e2\u003c/sub\u003e atmosphere\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eLentiviral supernatant preparation and infection\u003c/h3\u003e\n\u003cp\u003eHEK293T cells were transfected with lentiviral, packaging, and envelope plasmids using calcium phosphate to produce lentiviral particles\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Fourteen hours post-transfection, the medium was replaced with fresh DMEM containing 10% FBS. Forty-eight hours later, the lentiviral supernatant was collected, infecting 143B and MG-63 cells for 24 h. The following plasmids were used: short hairpin (sh) \u003cem\u003ePDK1\u003c/em\u003e#1 (Sigma-Aldrich, St. Louis, MO, USA, TRCN0000006261), sh\u003cem\u003ePDK1\u003c/em\u003e#2 (Sigma-Aldrich, TRCN0000006262), sh\u003cem\u003eActivating transcription factor 3\u003c/em\u003e (\u003cem\u003eATF3\u003c/em\u003e)#1 (Sigma-Aldrich, TRCN0000013572), sh\u003cem\u003eATF3\u003c/em\u003e#2 (Sigma-Aldrich, TRCN0000329689), pLKO.1 puro plasmid (Addgene, #8453), control vector (Vector Builder, Chicago, IL, USA, # VB010000-9389rbj), \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e (Vector Builder, VB900004-5579neb), and \u003cem\u003eATF3\u003c/em\u003e (Vector Builder, VB900000-5162ype). The \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eT358A\u003c/sup\u003e mutant was generated from \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e using the QuikChange II XL site-directed mutagenesis kit (Agilent, Santa Clara, CA, USA). Nucleotide sequences of \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e and \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eT358A\u003c/sup\u003e are analyzed using FinchTV software (ver. 1.5.0, Geospiza Inc, Seattle, WA, USA).\u003c/p\u003e\n\u003ch3\u003eWestern blotting\u003c/h3\u003e\n\u003cdiv class=\"Heading\"\u003eWestern blotting\u003c/div\u003e \u003cp\u003eProteins were extracted by lysing cells in a buffer containing 1% nonidet P-40 and a protease inhibitor cocktail. Proteins were separated via SDS-PAGE, transferred onto PVDF membranes, and subjected to blotting\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. The following antibodies were used. Primary antibodies: anti-PDK1 (1:1,000, Cell Signaling Technology, CST, Danvers, MA, USA, #3820), anti-PDH (1:1,000, CST, #2784), anti-p-PDH (1:1,000, CST, #31866), anti-SOX2 (1:1,000, CST, #14962), anti-c-Myc (1:1,000, CST, #5605), anti-ATF3 (1:500, Santa Cruz Biotechnology, SCBT, TX, USA, #sc-188), and anti-β-actin (1:2,000, SCBT, #sc-47778). Secondary antibodies: antirabbit IgG, HRP-linked (1:2,000, CST, #7074), and antimouse IgG, HRP-linked (1:4,000, CST, #7076). Protein levels were quantified using ImageJ.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSphere formation assay and in\u003c/b\u003e \u003cb\u003evitro\u003c/b\u003e \u003cb\u003elimiting dilution assay\u003c/b\u003e\u003c/p\u003e \u003cp\u003eCells were seeded at 1,000 cells per well in ultralow attachment 96-well plates (Corning Incorporated, Corning, NY, USA) and cultured in tumorsphere medium with 1% methylcellulose (FUJIFILM Wako Pure Chemical)\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. After 7 days, the number of spheres formed was counted. Sphere formation efficiency was assessed by counting spheres with \u0026gt;\u0026thinsp;50 \u0026micro;m diameter. For \u003cem\u003ein vitro\u003c/em\u003e limiting dilution assays, cells were plated in 96-well plates at densities of 1, 5, 10, 20, 40, or 80 cells/well (10 replicates each). The presence of a sphere was observed on day 7 and was analyzed using the \u0026ldquo;statmod\u0026rdquo; package (ver. 1.5.0) in R software (ver. 4.3.0).\u003c/p\u003e\n\u003ch3\u003eMTT, apoptosis, and wound healing assays\u003c/h3\u003e\n\u003cp\u003eMTT, apoptosis, and wound healing assays were performed to assess cell characteristics\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. For the MTT assay, 200 \u0026micro;L of 0.5 mg/mL MTT solution in PBS was added to each well on each measurement day and incubated for 4 h. Subsequently, 200 \u0026micro;L of 0.04 mol/L HCl in isopropanol was added to dissolve formazan crystals, and absorbance was measured at 550 nm with a microplate reader. For the wound healing assay, wounds were created using sterile 100 \u0026micro;L pipette tips, and images were taken at 0 and 24 h using a BZ-X810 Analyzer (KEYENCE, Osaka, Japan). The wound closure area-to-wound area ratio (migration rate) was calculated using ImageJ. For the apoptosis assay, cells were stained with PE-labeled Annexin V (1:50, BD Biosciences, BD, San Jose, CA, USA, #560930) and 7-AAD (1:1,000, BD, #559925) for 30 min at 4\u0026deg;C in the dark, followed by analysis using a CytoFLEX S flow cytometer (Beckman Coulter, Brea, CA, USA).\u003c/p\u003e\n\u003ch3\u003eMeasurement of glucose uptake, lactate production, and the ADP/ATP ratio\u003c/h3\u003e\n\u003cp\u003eGlucose uptake, lactic acid production, and the ADP/ATP ratio were assessed using Glucose Assay Kit-WST (Dojindo, Kumamoto, Japan), Lactate Colorimetric/Fluorometric Assay Kit (BioVision, SF, USA), and ADP/ATP Ratio Assay Kit (Sigma-Aldrich), following the manufacturers\u0026rsquo; protocols.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eChemicals and reagents\u003c/h2\u003e \u003cp\u003e2,2-Dichloroacetophenone (DAP) was obtained from Tokyo Chemical Industry (Tokyo, Japan), and FX11 was obtained from Selleck (Houston, TX, USA).\u003c/p\u003e \u003cp\u003e \u003cb\u003eIn vivo\u003c/b\u003e \u003cb\u003eexperiments\u003c/b\u003e\u003c/p\u003e \u003cp\u003eFor \u003cem\u003ein vivo\u003c/em\u003e studies, 143B cells (5 x 10\u003csup\u003e6\u003c/sup\u003e) were subcutaneously injected into BALB/cSlc-nu/nu mice (SLC, Hamamatsu, Japan). Tumor volume (calculated as length \u0026times; width\u003csup\u003e2\u003c/sup\u003e/2) was measured every 5 days with a digital caliper, with an endpoint set at a tumor diameter of 20 mm. For immunohistochemistry, the following antibodies were used: anti-SOX2 (1:300, CST, #14962) as a primary antibody, Alexa Fluor 546 goat antirabbit IgG (1:400, Invitrogen, CA, USA, #A11035) as a secondary antibody, and DAPI (1:1,000, Dojindo, #340\u0026ndash;07971) for nuclear staining. Immunofluorescence was visualized using a BZ-X800 Analyzer (KEYENCE).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003escRNA-seq data analysis\u003c/h3\u003e\n\u003cp\u003eThe GSE152048 dataset was analyzed using the \u0026ldquo;Seurat\u0026rdquo; package (ver. 5.1.0) in R software (ver. 4.3.0)\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. During preprocessing, cells with mitochondrial RNA content\u0026thinsp;\u0026ge;\u0026thinsp;10%, expressed genes\u0026thinsp;\u0026gt;\u0026thinsp;7,500, or \u0026lt;\u0026thinsp;300 were filtered as low-quality. single-sample Gene Set Enrichment Analysis (ssGSEA) was performed on tumor cell clusters using the \u0026ldquo;GSVA\u0026rdquo; package (ver. 1.50.5)\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. OSC cells were identified as the top 1% in ssGSEA scores, utilizing the MALTA_CURATED_STEMNESS_MARKERS gene set (OSC; n\u0026thinsp;=\u0026thinsp;226, non-OSC; n\u0026thinsp;=\u0026thinsp;22,392). Differentially expressed genes (DEGs; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were identified between groups with the Wilcoxon rank-sum test using the \u0026ldquo;presto\u0026rdquo; package (ver 1.0.0). Gene set enrichment analysis (GSEA) was performed with the \u0026ldquo;clusterProfiler\u0026rdquo; package (ver 4.8.3), with gene sets downloaded through the \u0026ldquo;msigdbr\u0026rdquo; package (ver 7.5.1)\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Results were visualized using the \u0026ldquo;enrichplot\u0026rdquo; package (ver 1.20.3) and \u0026ldquo;ggplot2\u0026rdquo; (ver 3.4.4).\u003c/p\u003e\n\u003ch3\u003eBulk RNA-seq data analysis\u003c/h3\u003e\n\u003cp\u003eFor 143B OSCs RNA-seq, cells were lysed and total RNA was extracted using FastGene RNA Premium Kit (NIPPON Genetics, Tokyo, Japan) per the manufacturer\u0026rsquo;s instructions. Paired-end sequencing was performed on the Illumina NovaSeq 6000 platform.\u003c/p\u003e \u003cp\u003eFor the RNA-seq analysis of 143B, PRJNA539828, and GSE126209 datasets, trimming was conducted using \u0026ldquo;Trimmomatic\u0026rdquo; (ver. 0.39), followed by quality checks with \u0026ldquo;FASTQC\u0026rdquo; (ver 0.12.1)\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Alignment to the Homo sapiens genome (GRCh38.p14) was performed using \u0026ldquo;STAR\u0026rdquo; (ver. 2.7.11a), and expression levels were calculated from the resulting BAM files using \u0026ldquo;RSEM\u0026rdquo; (ver. 1.3.3).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSurvival analysis\u003c/h2\u003e \u003cp\u003eSurvival data for patients with OS were retrieved from the TARGET-OS, TCGA-SARC, and GSE21257 datasets\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Survival analysis was carried out using the log-rank test in the \u0026ldquo;survival\u0026rdquo; package (ver. 3.5-5), and Kaplan\u0026ndash;Meier curves were generated using the \u0026ldquo;survminer\u0026rdquo; package (ver. 0.4.9).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses were performed using R software (ver. 4.3.0), and significance was indicated in the results (*\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, \u003csup\u003e#\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u003csup\u003e##\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, \u003csup\u003e###\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Unless otherwise specified, results are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (SD). Comparisons between two groups were evaluated using Student\u0026rsquo;s t- or Wilcoxon test, whereas comparisons among three or more groups were conducted using one-way or two-way ANOVA followed by Tukey\u0026ndash;Kramer test, or one-way ANOVA followed by Dunnett\u0026rsquo;s test. A \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eEnrichment of glycolysis and OXPHOS pathways in OSC\u003c/h2\u003e \u003cp\u003eWe analyzed a scRNA-seq dataset (GSE152048) of clinical OS specimens from five conventional and primary patients to profile the energy metabolic properties of OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA)\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Six clusters were identified through t-SNE analysis based on genetic profiles (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Canonical markers were used to annotate different cell types: malignant cells (\u003cem\u003eCOL1A1\u003c/em\u003e, \u003cem\u003eIBSP\u003c/em\u003e), myeloid cells (\u003cem\u003eCD74\u003c/em\u003e, \u003cem\u003eCD14\u003c/em\u003e), osteoclasts (\u003cem\u003eCTSK\u003c/em\u003e, \u003cem\u003eMMP9\u003c/em\u003e), pericytes (\u003cem\u003eRGS5\u003c/em\u003e, \u003cem\u003eACTA2\u003c/em\u003e), endothelial cells (\u003cem\u003ePECAM1\u003c/em\u003e, \u003cem\u003eVWF\u003c/em\u003e), and tumor-infiltrating lymphocytes (\u003cem\u003eCD3D\u003c/em\u003e, \u003cem\u003eNKG7\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). The OS cell population was divided into OSCs and non-OSCs by ssGSEA (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). The enrichment of two gene sets associated with \u0026ldquo;stemness\u0026rdquo; in OSCs was confirmed by GSEA (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE), defining these cells as the OSC population. We performed GSEA on HALLMARK gene sets to characterize OSCs, finding enrichment in \u0026ldquo;glycolysis\u0026rdquo; and \u0026ldquo;OXPHOS\u0026rdquo; gene sets in OSCs compared to non-OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF). Similarly, GSEA in the REACTOME gene sets showed that both \u0026ldquo;glycolysis\u0026rdquo; and \u0026ldquo;OXPHOS\u0026rdquo; were enriched in these pathways in OSCs (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH). Thus, scRNA-seq analysis of clinical OS specimens suggests activation of glycolysis and OXPHOS pathways in OSCs.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003ePDK1\u003c/b\u003e \u003cb\u003eis upregulated in OSC and linked to prognosis in patients with OS\u003c/b\u003e\u003c/p\u003e \u003cp\u003eUsing scRNA-seq data (GSE152048), we identified DEGs in early energy metabolic pathways in OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The expression levels of \u003cem\u003eSLC2A1\u003c/em\u003e, \u003cem\u003ePFKM\u003c/em\u003e, \u003cem\u003ePDHA1\u003c/em\u003e, \u003cem\u003ePDK1\u003c/em\u003e, and \u003cem\u003eLDHA\u003c/em\u003e were significantly increased in OSCs, while \u003cem\u003ePKM\u003c/em\u003e was significantly decreased in OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). The association of the expression levels of these metabolic genes with the survival Kaplan\u0026ndash;Meier analysis of patients was assessed using the TARGET-OS database. It showed that higher \u003cem\u003ePDK1\u003c/em\u003e expression correlated with shorter survival than those with lower \u003cem\u003ePDK1\u003c/em\u003e expression being the only gene correlating significantly with the prognosis of patients with OS (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). High \u003cem\u003ePDK1\u003c/em\u003e expression also predicted poorer overall and metastasis-free survival in patients with OS according to the microarray dataset (GSE21257) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD and Supplementary Fig.\u0026nbsp;1A) and was linked to poor prognosis of patients with soft tissue sarcoma (TCGA-SARC) (Supplementary Fig.\u0026nbsp;1B). \u003cem\u003ePDK1\u003c/em\u003e was significantly upregulated in clinical OS tissues compared to nontumor tissues in two bulk RNA-seq datasets (PRJNA539828, GSE126209) (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF). Among the early stage of the energy metabolic pathway genes, \u003cem\u003ePDK1\u003c/em\u003e was the only gene that fulfilled the significantly upregulated in OSCs and the consistently correlated with poor prognosis in patients with OS across cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG), suggesting its potential role in OS progression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003eOSC favors glycolysis via PDK1 overexpression\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eTo validate our bioinformatics analyses of clinical OS specimens (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), we cultured 143B cells, a patient-derived OS cell line, in tumorsphere condition (for OSCs; referred to as \u0026ldquo;143B OSCs\u0026rdquo;), followed by RNA-seq (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). We previously demonstrated that 143B OSCs represent stemness properties and tumorigenicity \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. We identified DEGs related to the early stage of energy metabolic pathways in 143B OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). The expression of \u003cem\u003ePDK1\u003c/em\u003e was significantly upregulated in 143B OSCs rather than in 143B non-OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC), as observed in the OSC population of clinical OS specimens (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB), along with significant upregulation of stem cell markers, such as \u003cem\u003eSOX2\u003c/em\u003e, \u003cem\u003eKLF4\u003c/em\u003e, and \u003cem\u003eABCG1\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD)\u003csup\u003e\u003cspan additionalcitationids=\"CR43 CR44 CR45\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. Moreover, the protein level of PDK1 was significantly upregulated in both 143B and MG-63 (also patient-derived OS cell lines) OSCs, concomitant with higher protein levels of stem cell markers SOX2 and c-MYC (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE)\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Additionally, glucose uptake and lactate production were significantly increased in 143B OSCs (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG), while the ADP/ATP ratio was significantly decreased (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH), indicating glycolysis activation in OSCs over differentiated OS cells. These findings suggest that OSCs may favor glycolysis through PDK1 overexpression, highlighting PDK1\u0026rsquo;s role in regulating OSC stem cell phenotypes and tumorigenicity.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eGenetic inhibition of\u003c/b\u003e \u003cb\u003ePDK1\u003c/b\u003e \u003cb\u003ereduces OSC stemness and glycolysis\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e\u003c/p\u003e \u003cp\u003eWe targeted \u003cem\u003ePDK1\u003c/em\u003e in 143B and MG-63 OSCs using lentiviral shRNAs (sh\u003cem\u003ePDK1\u003c/em\u003e#1 and sh\u003cem\u003ePDK1\u003c/em\u003e#2). Knockdown of \u003cem\u003ePDK1\u003c/em\u003e significantly reduced SOX2 and c-MYC protein levels in both cell lines, along with PDK1 itself (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Sphere formation assays showed that \u003cem\u003ePDK1\u003c/em\u003e silencing markedly decreased tumorsphere formation in both 143B and MG-63 OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). An \u003cem\u003ein vitro\u003c/em\u003e limiting dilution assay revealed that \u003cem\u003ePDK1\u003c/em\u003e silencing significantly impaired OSC self-renewal (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). \u003cem\u003ePDK1\u003c/em\u003e targeting reduced cell proliferation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD) and increased apoptosis in both (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). Additionally, \u003cem\u003ePDK1\u003c/em\u003e disruption significantly decreased migration potential (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). \u003cem\u003ePDK1\u003c/em\u003e knockdown also reduced glucose uptake and lactate production while increasing the ADP/ATP ratio in 143B OSCs, indicating reduced glycolysis (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG-\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eI). These findings suggest that PDK1 is crucial for maintaining OSC stem cell properties and metabolic preferences \u003cem\u003ein vitro\u003c/em\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eTargeting\u003c/b\u003e \u003cb\u003ePDK1\u003c/b\u003e \u003cb\u003ereduces OSC tumorigenicity and stemness\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e\u003c/p\u003e \u003cp\u003eGiven PDK1\u0026rsquo;s role in OSC growth, survival, aggressiveness, and self-renewal \u003cem\u003ein vitro\u003c/em\u003e, we investigated its effect on tumorigenicity in a xenograft mouse model (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Equal numbers of 143B OSCs infected with sh\u003cem\u003ePDK1\u003c/em\u003e or shControl (shCtrl) were transplanted into immunocompromised mice. Mice with sh\u003cem\u003ePDK1\u003c/em\u003e-infected 143B OSCs had significantly lower tumor volume and weight compared to those with shCtrl-infected OSCs (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Histological analysis showed a significant reduction in SOX2-positive cells in tumors from sh\u003cem\u003ePDK1\u003c/em\u003e-infected OSCs compared to shCtrl-infected OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). These results indicate that PDK1 is essential for maintaining tumorigenicity and stemness \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003ePharmacological inhibition of PDK1 suppresses OSC stemness and tumorigenicity\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e \u003cb\u003eand\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e\u003c/p\u003e \u003cp\u003eOur studies indicate that PDK1 in OSCs controls their stemness, tumorigenicity, and energy metabolism, making it a promising OS therapy target. We tested whether DAP, a PDK1 inhibitor, could suppress OSC stemness and tumorigenicity\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. DAP treatment significantly decreased SOX2 and c-MYC protein levels in 143B and MG63 OSCs without altering PDK1 levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). DAP also reduced tumor sphere formation dose-dependently (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). Additionally, DAP decreased cell proliferation and increased apoptosis in OSCs (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). In a xenograft mouse model, DAP was intraperitoneally administered to immunocompromised mice, which significantly reduced tumor volume and weight and SOX2-positive cells in tumors (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE-\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH). These results suggest that pharmacological PDK1 inhibition effectively reduces OSC stemness and tumorigenicity, similar to genetic \u003cem\u003ePDK1\u003c/em\u003e inhibition.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003ePDK1-dependent metabolic adaptation governs OSC properties through ATF3\u003c/h2\u003e \u003cp\u003eTo explore PDK1\u0026rsquo;s control mechanisms, we identified DEGs in \u0026ldquo;\u003cem\u003ePDK1\u003c/em\u003e-knockdown OSCs over control OSCs\u0026rdquo; and \u0026ldquo;OSCs over differentiated OS cells\u0026rdquo; via RNA-seq.\u0026nbsp;We found 174 and 157 downregulated genes in sh\u003cem\u003ePDK1\u003c/em\u003e#1-infected 143B OSCs and sh\u003cem\u003ePDK1\u003c/em\u003e#2-infected 143B OSCs, respectively, and 78 upregulated genes in OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Six overlapping genes included \u003cem\u003eATF3\u003c/em\u003e, associated with stemness in glioma stem cells (GSCs)\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. ATF3 protein levels were significantly higher in 143B OSCs than in differentiated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB) and decreased in \u003cem\u003ePDK1\u003c/em\u003e-knockdown OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC). \u003cem\u003eATF3\u003c/em\u003e mRNA level was elevated in OSCs (GSE152048) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD), and TARGET-OS analysis showed a positive correlation between \u003cem\u003ePDK1\u003c/em\u003e and \u003cem\u003eATF3\u003c/em\u003e expression in patients with OS (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE). \u003cem\u003eATF3\u003c/em\u003e silencing in 143B OSCs significantly reduced tumorsphere formation and self-renewal potential (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eF and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eG), along with the reduction of ATF3 protein levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eH and supplementary Fig.\u0026nbsp;2A). Conversely, \u003cem\u003eATF3\u003c/em\u003e overexpression reversed the reduction in tumorsphere formation caused by \u003cem\u003ePDK1\u003c/em\u003e knockdown (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eI and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eJ, and supplementary Fig.\u0026nbsp;2B). PDK1 is activated by phosphorylation at Thr358 by PGK1, leading to PDH inactivation, which converts pyruvate to acetyl-CoA (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eK)\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Overexpression of wild-type-\u003cem\u003ePDK1\u003c/em\u003e (\u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e) significantly increased tumorsphere formation, self-renewal potential, and lactate production in 143B OSCs (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eL-\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eN). Conversely, introducing kinase-dead (KD)-\u003cem\u003ePDK1\u003c/em\u003e (\u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eT358A\u003c/sup\u003e) did not significantly alter these properties (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eL-\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eN, Supplementary Fig.\u0026nbsp;2C). OSC phenotypes were significantly reduced with \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eT358A\u003c/sup\u003e compared to \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eL and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eM). Further, ATF3 expression was significantly increased by \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e but not by \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eT358A\u003c/sup\u003e in 143B OSCs, despite elevated PDK1 levels in both OSCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eO). Pharmacological inhibition of lactate dehydrogenase A by FX11 reversed ATF3 upregulation and tumorsphere formation in \u003cem\u003ePDK1\u003c/em\u003e\u003csup\u003eWT\u003c/sup\u003e-overexpressing 143B OSCs (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eP and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eQ). These results suggest that PDK1-mediated metabolic reprogramming governs OSC phenotypes partly through ATF3.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eATF3, a member of the ATF/CREB family of basic-leucine zipper transcription factors, modulates various cellular functions, including proliferation, apoptosis, and glucose metabolism\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. It acts as both an oncogene and tumor suppressor, depending on the tumor type and context\u003csup\u003e\u003cspan additionalcitationids=\"CR50 CR51 CR52 CR53\" citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. In HSCs, ATF3 is activated by tumor-primed bone marrow MSCs, redirecting hematopoiesis toward monocytic cell expansion\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. In GSCs, ATF3 enhances the stemness and tumorigenicity via TGF-β1/Smad2 signaling and promotes resistance to temozolomide by inducing ABCB4, an ABC transporter\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. While ATF3\u0026rsquo;s role in ferroptosis in OS has been noted, its impact on the stemness and tumorigenicity of OSCs and OS pathophysiology remains unexplored\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. We identified alternative candidate genes (\u003cem\u003eBEAN1\u003c/em\u003e, \u003cem\u003eC11orf96\u003c/em\u003e, \u003cem\u003eCCN3\u003c/em\u003e, \u003cem\u003eDUSP8\u003c/em\u003e, and \u003cem\u003eNIM1K\u003c/em\u003e) linked to PDK1-dependent maintenance of OSC stemness. Despite the need for further research, we found ATF3 in OSCs as a crucial downstream factor in PDK1-dependent regulation of OSC properties and OS pathogenesis.\u003c/p\u003e \u003cp\u003eExtensive studies have explored direct glycolysis inhibition as a potential cancer treatment, but success rates have been low. For instance, hexokinase 2 inhibition with lonidamine did not significantly improve overall survival in several cancers, including breast and lung, and caused elevated toxicity\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e,\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. Glucose transport inhibitors like silibinin (silybin) also caused critical side effects without significant response in prostate cancer\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. Combination therapies targeting multiple metabolic pathways may offer advantages over single-agent glycolysis inhibitors. Our study showed that both glycolysis and OXPHOS are activated in OSCs, as revealed by scRNA-seq analysis of OS clinical specimens. Pharmacological inhibition of PDK1 effectively repressed OSC stemness and tumorigenicity \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo.\u003c/em\u003e Further studies should investigate whether combining PDK1 and OXPHOS inhibitors, such as metformin, could effectively disrupt OSC stem cell properties and suppress OS malignancy without serious side effects.\u003c/p\u003e \u003cp\u003eThe metabolic characteristics of CSCs are highly heterogeneous. Unlike non-CSCs, which mainly utilize glycolysis, CSCs rely on either glycolysis or OXPHOS in a context-dependent manner, with contradictory results reported for the same tumor entity\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. For example, GSCs rely on glycolysis for energy production and survival through increased glucose consumption by upregulating GLUT3, while also depending on OXPHOS via IMP2, an RNA-binding protein\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. This discrepancy is attributed to CSC heterogeneity, tumor microenvironment, and experimental strategies\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Although the mechanisms regulating ATF3 expression downstream of PDK1-glycolysis in OSCs are unknown, we identified the PDK1-glycolysis-ATF3 axis as a critical regulator of OSC properties and OS malignancy. This is the first preclinical study to reveal the link between metabolic reprogramming in OSCs, their stemness and tumorigenesis, and OS malignancy. Despite extensive research efforts, effective treatments for OS have not advanced significantly in the past 4 decades\u003csup\u003e\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Our findings enhance the understanding of OS pathogenesis and OSC properties. They suggest that targeting energy metabolic reprogramming and the associated genes in OSCs could be a novel and effective approach for developing new treatments for OS in humans.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe GSE21257, GSE126209, and GSE152048 datasets are deposited in the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). The PRJNA539828 dataset is deposited in the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/). The TARGET-OS and TCGA-SARC datasets are deposited in the Genomic Data Commons data portal (https://portal.gdc.cancer.gov/).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful for the technical support from the members of the Hinoi lab.\u0026nbsp;This work was partially supported by the Japan Society for the Promotion of Science (20H03407 to E.H.).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eK.T., K.F. and E.H. conceived and designed the project. K.T., K.F. and M.H. performed the experiments and analysis. K.T. and E.H. wrote the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll animal experiments were approved by the Committees on Animal Experimentation of Gifu Pharmaceutical University and Gifu University and performed in accordance with the guidelines for the care and use of laboratory animals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLin Y-H, Jewell BE, Gingold J, Lu L, Zhao R, Wang LL \u003cem\u003eet al.\u003c/em\u003e Osteosarcoma: Molecular Pathogenesis and iPSC Modeling. Trends Mol Med 2017; 23: 737\u0026ndash;755.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGianferante DM, Mirabello L, Savage SA. Germline and somatic genetics of osteosarcoma - connecting aetiology, biology and therapy. Nat Rev Endocrinol 2017; 13: 480\u0026ndash;491.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGill J, Gorlick R. Advancing therapy for osteosarcoma. Nat Rev Clin Oncol 2021; 18: 609\u0026ndash;624.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartins-Neves SR, Sampaio-Ribeiro G, Gomes CMF. Self-Renewal and Pluripotency in Osteosarcoma Stem Cells\u0026rsquo; Chemoresistance: Notch, Hedgehog, and Wnt/β-Catenin Interplay with Embryonic Markers. Int J Mol Sci 2023; 24: 8401.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMen\u0026eacute;ndez ST, Gallego B, Murillo D, Rodr\u0026iacute;guez A, Rodr\u0026iacute;guez R. Cancer Stem Cells as a Source of Drug Resistance in Bone Sarcomas. J Clin Med 2021; 10: 2621.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan G-N, Lv Y-F, Guo Q-N. Advances in osteosarcoma stem cell research and opportunities for novel therapeutic targets. Cancer Lett 2016; 370: 268\u0026ndash;274.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarburg O. On the origin of cancer cells. Science 1956; 123: 309\u0026ndash;314.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiao M, Yao D, Wu L, Luo C, Wang Z, Zhang J \u003cem\u003eet al.\u003c/em\u003e Targeting the Warburg effect: A revisited perspective from molecular mechanisms to traditional and innovative therapeutic strategies in cancer. Acta Pharm Sin B 2024; 14: 953\u0026ndash;1008.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAn F, Chang W, Song J, Zhang J, Li Z, Gao P \u003cem\u003eet al.\u003c/em\u003e Reprogramming of glucose metabolism: Metabolic alterations in the progression of osteosarcoma. J Bone Oncol 2024; 44: 100521.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChu X, Tian W, Ning J, Xiao G, Zhou Y, Wang Z \u003cem\u003eet al.\u003c/em\u003e Cancer stem cells: advances in knowledge and implications for cancer therapy. Signal Transduct Target Ther 2024; 9: 170.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLagadinou ED, Sach A, Callahan K, Rossi RM, Neering SJ, Minhajuddin M \u003cem\u003eet al.\u003c/em\u003e BCL-2 inhibition targets oxidative phosphorylation and selectively eradicates quiescent human leukemia stem cells. Cell Stem Cell 2013; 12: 329\u0026ndash;341.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVlashi E, Lagadec C, Vergnes L, Matsutani T, Masui K, Poulou M \u003cem\u003eet al.\u003c/em\u003e Metabolic state of glioma stem cells and nontumorigenic cells. Proc Natl Acad Sci USA 2011; 108: 16062\u0026ndash;16067.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYe X-Q, Li Q, Wang G-H, Sun F-F, Huang G-J, Bian X-W \u003cem\u003eet al.\u003c/em\u003e Mitochondrial and energy metabolism-related properties as novel indicators of lung cancer stem cells. Int J Cancer 2011; 129: 820\u0026ndash;831.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVlashi E, Lagadec C, Vergnes L, Reue K, Frohnen P, Chan M \u003cem\u003eet al.\u003c/em\u003e Metabolic differences in breast cancer stem cells and differentiated progeny. Breast Cancer Res Treat 2014; 146: 525\u0026ndash;534.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePast\u0026ograve; A, Bellio C, Pilotto G, Ciminale V, Silic-Benussi M, Guzzo G \u003cem\u003eet al.\u003c/em\u003e Cancer stem cells from epithelial ovarian cancer patients privilege oxidative phosphorylation, and resist glucose deprivation. Oncotarget 2014; 5: 4305\u0026ndash;4319.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSancho P, Burgos-Ramos E, Tavera A, Bou Kheir T, Jagust P, Schoenhals M \u003cem\u003eet al.\u003c/em\u003e MYC/PGC-1α Balance Determines the Metabolic Phenotype and Plasticity of Pancreatic Cancer Stem Cells. Cell Metab 2015; 22: 590\u0026ndash;605.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaniele S, Giacomelli C, Zappelli E, Granchi C, Trincavelli ML, Minutolo F \u003cem\u003eet al.\u003c/em\u003e Lactate dehydrogenase-A inhibition induces human glioblastoma multiforme stem cell differentiation and death. Sci Rep 2015; 5: 15556.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong C, Yuan T, Wu Y, Wang Y, Fan TWM, Miriyala S \u003cem\u003eet al.\u003c/em\u003e Loss of FBP1 by Snail-mediated repression provides metabolic advantages in basal-like breast cancer. Cancer Cell 2013; 23: 316\u0026ndash;331.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen Y-A, Wang C-Y, Hsieh Y-T, Chen Y-J, Wei Y-H. Metabolic reprogramming orchestrates cancer stem cell properties in nasopharyngeal carcinoma. Cell Cycle 2015; 14: 86\u0026ndash;98.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJaniszewska M, Suv\u0026agrave; ML, Riggi N, Houtkooper RH, Auwerx J, Cl\u0026eacute;ment-Schatlo V \u003cem\u003eet al.\u003c/em\u003e Imp2 controls oxidative phosphorylation and is crucial for preserving glioblastoma cancer stem cells. Genes Dev 2012; 26: 1926\u0026ndash;1944.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlavahan WA, Wu Q, Hitomi M, Rahim N, Kim Y, Sloan AE \u003cem\u003eet al.\u003c/em\u003e Brain tumor initiating cells adapt to restricted nutrition through preferential glucose uptake. Nat Neurosci 2013; 16: 1373\u0026ndash;1382.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStine ZE, Schug ZT, Salvino JM, Dang CV. Targeting cancer metabolism in the era of precision oncology. Nat Rev Drug Discov 2022; 21: 141\u0026ndash;162.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJeoung NH. Pyruvate Dehydrogenase Kinases: Therapeutic Targets for Diabetes and Cancers. Diabetes Metab J 2015; 39: 188\u0026ndash;197.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAtas E, Oberhuber M, Kenner L. The Implications of PDK1\u0026ndash;4 on Tumor Energy Metabolism, Aggressiveness and Therapy Resistance. Front Oncol 2020; 10: 583217.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Jiang Y, Meisenhelder J, Yang W, Hawke DH, Zheng Y \u003cem\u003eet al.\u003c/em\u003e Mitochondria-Translocated PGK1 Functions as a Protein Kinase to Coordinate Glycolysis and the TCA Cycle in Tumorigenesis. Mol Cell 2016; 61: 705\u0026ndash;719.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei Y, Zhang D, Shi H, Qian H, Chen H, Zeng Q \u003cem\u003eet al.\u003c/em\u003e PDK1 promotes breast cancer progression by enhancing the stability and transcriptional activity of HIF-1α. Genes Dis 2024; 11: 101041.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJha MK, Suk K. Pyruvate dehydrogenase kinase as a potential therapeutic target for malignant gliomas. Brain Tumor Res Treat 2013; 1: 57\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErdem A, Marin S, Pereira-Martins DA, Cort\u0026eacute;s R, Cunningham A, Pruis MG \u003cem\u003eet al.\u003c/em\u003e The Glycolytic Gatekeeper PDK1 defines different metabolic states between genetically distinct subtypes of human acute myeloid leukemia. Nat Commun 2022; 13: 1105.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalvarsson C, Eliasson P, J\u0026ouml;nsson J-I. Pyruvate dehydrogenase kinase 1 is essential for transplantable mouse bone marrow hematopoietic stem cell and progenitor function. PLoS One 2017; 12: e0171714.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng F, Wang J-H, Fan W-J, Meng Y-T, Li M-M, Li T-T \u003cem\u003eet al.\u003c/em\u003e Glycolysis gatekeeper PDK1 reprograms breast cancer stem cells under hypoxia. Oncogene 2018; 37: 1062\u0026ndash;1074.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOsumi R, Sugihara K, Yoshimoto M, Tokumura K, Tanaka Y, Hinoi E. Role of proteoglycan synthesis genes in osteosarcoma stem cells. Front Oncol 2024; 14: 1325794.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFukasawa K, Lyu J, Kubo T, Tanaka Y, Suzuki A, Horie T \u003cem\u003eet al.\u003c/em\u003e MEK5-ERK5 Axis Promotes Self-renewal and Tumorigenicity of Glioma Stem Cells. Cancer Res Commun 2023; 3: 148\u0026ndash;159.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYamada T, Fukasawa K, Horie T, Kadota T, Lyu J, Tokumura K \u003cem\u003eet al.\u003c/em\u003e The role of CDK8 in mesenchymal stem cells in controlling osteoclastogenesis and bone homeostasis. Stem Cell Reports 2022; 17: 1576\u0026ndash;1588.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSasaki Y, Tokumura K, Yoshimoto M, Hinoi E. Association between Fat-Soluble Vitamin Metabolic Process and Glioma Progression. Biol Pharm Bull 2024; 47: 1682\u0026ndash;1689.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou Y, Yang D, Yang Q, Lv X, Huang W, Zhou Z \u003cem\u003eet al.\u003c/em\u003e Single-cell RNA landscape of intratumoral heterogeneity and immunosuppressive microenvironment in advanced osteosarcoma. Nat Commun 2020; 11: 6322.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026auml;nzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013; 14: 7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF \u003cem\u003eet al.\u003c/em\u003e Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009; 462: 108\u0026ndash;112.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA \u003cem\u003eet al.\u003c/em\u003e Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005; 102: 15545\u0026ndash;15550.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMootha VK, Lindgren CM, Eriksson K-F, Subramanian A, Sihag S, Lehar J \u003cem\u003eet al.\u003c/em\u003e PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003; 34: 267\u0026ndash;273.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang X, Qin G, Liang X, Wang W, Wang Z, Liao D \u003cem\u003eet al.\u003c/em\u003e Targeting the CK1α/CBX4 axis for metastasis in osteosarcoma. Nat Commun 2020; 11: 1141.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuddingh EP, Kuijjer ML, Duim RAJ, B\u0026uuml;rger H, Agelopoulos K, Myklebost O \u003cem\u003eet al.\u003c/em\u003e Tumor-infiltrating macrophages are associated with metastasis suppression in high-grade osteosarcoma: a rationale for treatment with macrophage activating agents. Clin Cancer Res 2011; 17: 2110\u0026ndash;2119.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaurizi G, Verma N, Gadi A, Mansukhani A, Basilico C. Sox2 is required for tumor development and cancer cell proliferation in osteosarcoma. Oncogene 2018; 37: 4626\u0026ndash;4632.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBasu-Roy U, Bayin NS, Rattanakorn K, Han E, Placantonakis DG, Mansukhani A \u003cem\u003eet al.\u003c/em\u003e Sox2 antagonizes the Hippo pathway to maintain stemness in cancer cells. Nat Commun 2015; 6: 6411.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQi X-T, Li Y-L, Zhang Y-Q, Xu T, Lu B, Fang L \u003cem\u003eet al.\u003c/em\u003e KLF4 functions as an oncogene in promoting cancer stem cell-like characteristics in osteosarcoma cells. Acta Pharmacol Sin 2019; 40: 546\u0026ndash;555.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Xian M, Yang B, Ying M, He Q. Inhibition of KLF4 by Statins Reverses Adriamycin-Induced Metastasis and Cancer Stemness in Osteosarcoma Cells. Stem Cell Reports 2017; 8: 1617\u0026ndash;1629.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoundhill EA, Jabri S, Burchill SA. ABCG1 and Pgp identify drug resistant, self-renewing osteosarcoma cells. Cancer Lett 2019; 453: 142\u0026ndash;157.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Schaijik B, Davis PF, Wickremesekera AC, Tan ST, Itinteang T. Subcellular localisation of the stem cell markers OCT4, SOX2, NANOG, KLF4 and c-MYC in cancer: a review. J Clin Pathol 2018; 71: 88\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQin L, Tian Y, Yu Z, Shi D, Wang J, Zhang C \u003cem\u003eet al.\u003c/em\u003e Targeting PDK1 with dichloroacetophenone to inhibit acute myeloid leukemia (AML) cell growth. Oncotarget 2016; 7: 1395\u0026ndash;1407.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu J, Zhang G, Hu J, Li H, Zhao J, Zong S \u003cem\u003eet al.\u003c/em\u003e UPF1/circRPPH1/ATF3 feedback loop promotes the malignant phenotype and stemness of GSCs. Cell Death Dis 2022; 13: 645.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu X, Zheng Y, Luo L, You Z, Chen H, Wang J \u003cem\u003eet al.\u003c/em\u003e Glioblastoma stem cells deliver ABCB4 transcribed by ATF3 via exosomes conferring glioblastoma resistance to temozolomide. Cell Death Dis 2024; 15: 318.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKu H-C, Cheng C-F. Master Regulator Activating Transcription Factor 3 (ATF3) in Metabolic Homeostasis and Cancer. Front Endocrinol (Lausanne) 2020; 11: 556.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan L, Gaddis S, Coletta LD, Repass J, Powell KL, Simper MS \u003cem\u003eet al.\u003c/em\u003e ATF3-Induced Mammary Tumors Exhibit Molecular Features of Human Basal-Like Breast Cancer. Int J Mol Sci 2021; 22: 2353.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJeon T, Ko MJ, Seo Y-R, Jung S-J, Seo D, Park S-Y \u003cem\u003eet al.\u003c/em\u003e Silencing CDCA8 Suppresses Hepatocellular Carcinoma Growth and Stemness via Restoration of ATF3 Tumor Suppressor and Inactivation of AKT/β-Catenin Signaling. Cancers (Basel) 2021; 13: 1055.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang C, Chen R, Zheng F, Tang Y, Wang X, Chen Z \u003cem\u003eet al.\u003c/em\u003e Inhibitory role of ATF3 in gastric cancer progression through regulating cell EMT and stemness. Cancer Cell Int 2021; 21: 127.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerrone M, Chiodoni C, Lecchi M, Botti L, Bassani B, Piva A \u003cem\u003eet al.\u003c/em\u003e ATF3 Reprograms the Bone Marrow Niche in Response to Early Breast Cancer Transformation. Cancer Res 2023; 83: 117\u0026ndash;129.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Yan J, Zhao Q, Zhang Y, Zhang Y. ATF3 promotes ferroptosis in sorafenib-induced cardiotoxicity by suppressing Slc7a11 expression. Front Pharmacol 2022; 13: 904314.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerruti A, Bitossi R, Gorzegno G, Bottini A, Alquati P, De Matteis A \u003cem\u003eet al.\u003c/em\u003e Time to progression in metastatic breast cancer patients treated with epirubicin is not improved by the addition of either cisplatin or lonidamine: final results of a phase III study with a factorial design. J Clin Oncol 2002; 20: 4150\u0026ndash;4159.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Marinis F, Rinaldi M, Ardizzoni A, Bruzzi P, Pennucci MC, Portalone L \u003cem\u003eet al.\u003c/em\u003e The role of vindesine and lonidamine in the treatment of elderly patients with advanced non-small cell lung cancer: a phase III randomized FONICAP trial. Italian Lung Cancer Task Force. Tumori 1999; 85: 177\u0026ndash;182.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlaig TW, Gustafson DL, Su L-J, Zirrolli JA, Crighton F, Harrison GS \u003cem\u003eet al.\u003c/em\u003e A phase I and pharmacokinetic study of silybin-phytosome in prostate cancer patients. Invest New Drugs 2007; 25: 139\u0026ndash;146.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEun K, Ham SW, Kim H. Cancer stem cell heterogeneity: origin and new perspectives on CSC targeting. BMB Rep 2017; 50: 117\u0026ndash;125.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Visser KE, Joyce JA. The evolving tumor microenvironment: From cancer initiation to metastatic outgrowth. Cancer Cell 2023; 41: 374\u0026ndash;403.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-disease","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cddis","sideBox":"Learn more about [Cell Death \u0026 Disease](http://www.nature.com/cddis/)","snPcode":"41419","submissionUrl":"https://mts-cddis.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Disease","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"PDK1, glycolysis, osteosarcoma, cancer stem cells","lastPublishedDoi":"10.21203/rs.3.rs-5372467/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5372467/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eOsteosarcoma stem cells (OSCs) are characterized by their self-renewal and multilineage differentiation abilities, contributing to osteosarcoma (OS) malignancy. The Warburg effect describes cancer cells\u0026rsquo; preference for glycolysis over mitochondrial oxidative phosphorylation (OXPHOS) for energy production. Unlike differentiated cancer cells, cancer stem cells exhibit unique and diverse metabolic properties depending on the context. This study investigated the metabolic reliance of OSCs and related genes through \u003cem\u003ein silico\u003c/em\u003e analyses of clinical OS specimens and \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e genetic and pharmacological analyses. Glycolysis and OXPHOS pathways were more active in OSCs than in non-OSCs at single-cell resolution. Pyruvate dehydrogenase kinase 1 (PDK1), a key enzyme balancing glycolysis and OXPHOS, was upregulated in OSCs and correlated with poor prognosis in patients with OS. Genetic inhibition of \u003cem\u003ePDK1\u003c/em\u003e via RNA interference reduced OSC stemness, tumorigenicity, and glycolysis. Pharmacological inhibition of PDK1 mirrored these genetic effects. Activating transcription factor 3 (ATF3) was identified through screening as a downstream factor of PDK1-regulated OSC properties. Silencing \u003cem\u003eATF3\u003c/em\u003e reduced OSC stemness, while \u003cem\u003eATF3\u003c/em\u003e overexpression reversed the stemness reduction caused by \u003cem\u003ePDK1\u003c/em\u003e deficiency. ATF3 expression, glycolysis, and stemness were significantly induced by wild-type \u003cem\u003ePDK1\u003c/em\u003e overexpression but not by a kinase-dead \u003cem\u003ePDK1\u003c/em\u003e mutant in OSCs. Pharmacological inhibition of glycolysis counteracted the upregulation of ATF3 expression and increased stemness in OSCs by \u003cem\u003ePDK1\u003c/em\u003e overexpression. These findings indicate that PDK1 fine-tunes metabolic balance to govern OSC stemness and tumorigenicity through ATF3, suggesting a potential therapeutic approach for targeting OSCs in OS.\u003c/p\u003e","manuscriptTitle":"PDK1-dependent metabolic reprogramming regulates stemness and tumorigenicity of osteosarcoma stem cells through ATF3","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-18 07:36:19","doi":"10.21203/rs.3.rs-5372467/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2024-11-28T15:49:31+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-11-15T13:54:13+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-11-14T18:09:56+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2024-11-14T18:08:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-11-04T11:10:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cell Death \u0026 Disease","date":"2024-11-01T10:01:38+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-11-01T10:01:38+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-disease","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cddis","sideBox":"Learn more about [Cell Death \u0026 Disease](http://www.nature.com/cddis/)","snPcode":"41419","submissionUrl":"https://mts-cddis.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Disease","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"50682d78-3aa5-4b1f-acef-cf8f0511966b","owner":[],"postedDate":"December 18th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":40282556,"name":"Health sciences/Diseases/Cancer/Bone cancer"},{"id":40282557,"name":"Biological sciences/Cancer/Bone cancer"}],"tags":[],"updatedAt":"2025-08-07T07:36:26+00:00","versionOfRecord":{"articleIdentity":"rs-5372467","link":"https://doi.org/10.1038/s41419-025-07903-7","journal":{"identity":"cell-death-and-disease","isVorOnly":false,"title":"Cell Death \u0026 Disease"},"publishedOn":"2025-07-29 04:00:00","publishedOnDateReadable":"July 29th, 2025"},"versionCreatedAt":"2024-12-18 07:36:19","video":"","vorDoi":"10.1038/s41419-025-07903-7","vorDoiUrl":"https://doi.org/10.1038/s41419-025-07903-7","workflowStages":[]},"version":"v1","identity":"rs-5372467","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5372467","identity":"rs-5372467","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}