PLPP1 inhibits Wnt/β-catenin signaling via enhancing Caveolin-1/β-catenin interaction and is negatively regulated by c-Myc in osteosarcoma Authors: Jiangbo Nie, Yang Zhou, Xinsheng Xie, Jinfa Wu, Weilai Tong, Zhiguo Shu, Shijiang Wang, Jiaming Liu, Feng Yang, Zhili Liu Published: 2026 ABSTRACT AbstractBackgroundOsteosarcoma (OS) is the most common primary malignant bone tumor in adolescents, characterized by rapid progression and poor prognosis. The molecular mechanisms underlying OS malignancy remain incompletely understood. Phospholipid phosphatase 1 (PLPP1) has recently been implicated in cancer biology, yet its role in OS is largely unknown.MethodsRNA sequence data from TARGET-OS and GEO cohorts were downloaded and analyzed for assessing the relationship between PLPP1 expression levels and the OS prognosis. PLPP1 expression levels in OS cells and samples were detected using qRT-PCR, WB and immunohistochemistry. The effects of PLPP1 on proliferation, migration, invasion and chemosensitivity of OS cells were evaluated by up- or down-regulated PLLP1 expression. In vivo subcutaneous and tail-vein xenograft models were used to examine tumor growth and metastasis. Then, transcriptome profiling, luciferase reporter assays, co-immunoprecipitation, immunofluorescence and ChIP-PCR were conducted to elucidate the molecular mechanisms.ResultsPLPP1 was markedly down-regulated in OS tissues and cell lines. And low expression level of PLPP1 was correlated with poor prognosis, advanced clinical stage and metastasis. Functionally, PLPP1 overexpression inhibited OS cell proliferation, migration, invasion, and enhanced chemotherapy-induced apoptosis; conversely, silencing PLPP1 produced the opposite effects. In vivo, PLPP1 overexpression suppressed tumor growth and lung metastasis. Mechanistically, overexpression of PLPP1 impaired strength of Wnt/β-catenin signaling without altering β-catenin abundance, but instead inhibited β-catenin nuclear import by enhancing Caveolin-1/β-catenin interaction. Furthermore, PLPP1 was transcriptionally repressed by c-Myc, which directly bound to the PLPP1 promoter. Rescue experiments confirmed that c-Myc promoted OS malignancy by repressing PLPP1, thereby activating Wnt/β-catenin signaling.ConclusionsThis study identified PLPP1 as a prognostic suppressor in OS that inhibited β-catenin nuclear import through enhancing Caveolin-1/β-catenin interaction and was transcriptionally repressed by c-Myc. Targeting the c-Myc/PLPP1/β-catenin axis might represent a promising therapeutic strategy for OS.The translational potential of this articleThis study revealed the potential of PLPP1 as a novel prognostic biomarker of OS, and suggested its potential for therapeutic applications by specifically activating PLPP1 expression using small-molecule activators.Keywords: Caveolin-1, c-Myc, Osteosarcoma, PLPP1, Protein interactions, Wnt/β-catenin Introduction Osteosarcoma (OS) is the most common primary malignant bone tumor, predominantly affecting children and adolescents [1]. It typically arises from the metaphysis of long bones, most frequently in the distal femur, followed by the proximal tibia and proximal humerus [2]. And OS is characterized by atypical clinical manifestations and rapid progression, leading to high disability and mortality rates. Currently, the standard treatment involves surgical resection combined with neoadjuvant and adjuvant chemotherapy, in which doxorubicin (DOX), cisplatin (CDDP) and methotrexate are commonly used first-line agents [3]. With these regimens, the five-year survival rate of patients with localized OS has improved to 65–70% [4]. However, patients with distant metastases have poor outcomes, with survival rates below 20% [5]. Therefore, elucidating the molecular mechanisms underlying OS progression and identifying novel therapeutic targets are urgent. Lipid phosphate phosphatases (LPPs), members of the phospholipid phosphatase (PLPP) family, comprise three isoforms: PLPP1, PLPP2, and PLPP3 [6]. These transmembrane enzymes mainly localize to the plasma membrane, Golgi apparatus and endoplasmic reticulum, where they catalyze the hydrolysis of bioactive lipid phosphates such as phosphatidate, lysophosphatidic acid (LPA), and sphingosine-1-phosphate (S1P) [7]. PLPP1, also known as LPP1 or PPAP2A, is widely expressed across human tissues and plays critical role in lipid metabolism and diverse pathophysiological processes [8]. Previous studies have shown that PLPP1 participates in the degradation of LPA and S1P, thereby attenuating platelet activation, inflammatory responses and cell migration. Increasing evidence suggests that PLPP1 functions as a tumor suppressor by modulating signal transduction in cancer cells [6]. For instance, PLPP1 overexpression promotes LPA catabolism, thereby inhibiting tumor proliferation and metastasis in vivo [9,10]. Furthermore, PLPP1 suppresses transcriptional activity of AP-1 and inhibits the expression of cell cycle proteins, consequently inhibiting breast cancer progression [11]. Nevertheless, the biological role and regulatory mechanisms of PLPP1 in OS remain largely unexplored. Aberrant activation of multiple signaling pathways, including PI3K/AKT, JAK/STAT, Wnt/β-catenin, NOTCH, MAPK, RANK/RANKL, and NF-κB, have been implicated during OS progression [12]. Among them, the Wnt/β-catenin pathway is evolutionarily conserved and plays a pivotal role in regulating cell proliferation, differentiation, and migration, as well as skeletal development and bone remodeling [13]. Dysregulation of Wnt/β-catenin signaling promotes oncogene expression, disrupts osteogenic differentiation and enhances the metastatic potential of OS cells [14]. Besides, pharmacological inhibition of Wnt/β-catenin signaling has been shown to impair proliferation and metastasis of OS cells, highlighting this pathway as a promising therapeutic target [15,16]. In our study, we first found that PLPP1 was expressed at low levels in OS samples and cells. And low expression of PLPP1 was associated with poor prognosis in OS patients. Then, we demonstrated that PLPP1 overexpression suppressed the proliferation and metastasis of OS using in vitro and in vivo experiments. RNA-sequence (RNA-seq) indicated a potential association between PLPP1 and Wnt/β-catenin pathway, which was further confirmed by experimental validation. Mechanistically, PLPP1 inhibited β-catenin nuclear translocation through enhancing its interaction with Caveolin-1. In addition, we identified and validated the upstream transcriptional regulators of PLPP1. Collectively, our study provided novel insights into the tumor-suppressive role of PLPP1 in OS and might uncover potential therapeutic strategies targeting PLPP1/Caveolin-1/β-catenin axis. Method Reagents and chemicals Methyl 3-{[(4-methylphenyl)sulfonyl]amino}benzoate (MSAB), a specific β-catenin inhibitor, and Cell Counting Kit-8 (CCK8) were purchased from MedChemExpress (New Jersey, USA). Matrix (Cat. #40183ES08) was obtained from Yeasen Biotechnology (Shanghai) Co., Ltd. Bioinformatics analysis Clinical and RNA-seq data of 85 OS patients in the TARGET-OS cohort were retrieved from the UCSC Xena platform. Transcriptomic data were transformed from transcripts per million (TPM) to log2 (TPM + 1) format. Univariate Cox (uni-COX) proportional hazards regression was performed to identify genes associated with OS prognosis, with P < 0.05 considered statistically significant. The list of human sphingolipid metabolism-related genes was obtained from the Molecular Signatures Database (MSigDB) of the Gene Set Enrichment Analysis (GSEA) resource. Overlapping genes between prognostic genes and sphingolipid metabolism-related genes were visualized using a Venn diagram. For external validation, clinical and RNA-seq data of 53 patients from the GSE21257 dataset were downloaded from the Gene Expression Omnibus (GEO). The “surv_cutpoint” function was used to determine the optimal cutoff value of PLPP1 expression. Based on this cutoff, patients in both the TARGET-OS and GSE21257 cohorts were stratified into high- and low-PLPP1 groups, and Kaplan–Meier (K-M) survival curves were generated. Additionally, the Kaplan–Meier Plotter platform (https://www.kmplot.com/) was employed to assess the prognostic value of PLPP1 expression in sarcoma. The RNA-seq data of GSE14359 (2 osteoblast cells and 18 OS cells) and GSE42352 (3 osteoblast, 12 mesenchymal stem cells and 19 OS cells) datasets were used to analysis the expression level of PLPP1 between normal cells (NCs, including osteoblast and mesenchymal stem cells) and OS cells (OSs). Potential upstream transcription factors regulating PLPP1 were obtained from the Cistrome Data Browser, Gene Transcription Regulation Database (GTRD), and Human Transcription Factor Targets (HTFtargets) database. The intersection between predicted transcription factors and prognostic genes was illustrated using the Venn diagram. K-M Plotter was further used to evaluate the prognostic relevance of c-Myc and CEBPA in sarcoma. Cell culture hFOB1.19 cells, Saos-2 cells, and 143B cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). HOS, U2OS and MG63 cells were obtained from Procell (Wuhan, China). Cells were cultured in media supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin in a humidified incubator containing 5% CO2. hFOB1.19 cells were maintained in DMEM/F12 medium supplemented with 10% FBS, 1% penicillin–streptomycin, and 0.3 mg/mL G418 at 33.5 °C. Saos-2 and U2OS cells were cultured in McCoy's 5A medium, whereas 143B, MG63, and HOS cells were cultured in MEM medium at 37 °C. Quantitative real-time polymerase chain reaction (qRT-PCR) Total RNA was extracted using Total RNA Extraction Reagent (Cat. B0004-plus, EZBioscience, the USA). After quantification using Nanodrop, RNA was reversed to cDNA by PrimeScript™ RT Master Mix (Cat. RR036Q, TAKARA, Japan). Then, the genes expressions levels were identified by qRT-PCR analysis according to previous study [17]. The primers used were listed at Table 1. The actin gene was used as the reference gene. Table 1Primers used in the experiment.Table 1GenesSequence (5’ – 3′)PLPP1ForwardCACTCTTCGTTTTCCATGTACTGReverseCGAGAAAGGCCCACATAAATGCTNNB1ForwardGTTCAGTTGCTTGTTCGTGCReverseGTTGTGAACATCCCGAGCTAGc-MycForwardTGGTCTTCCCCTACCCTCTCAReverseTCTTCCTCATCTTCTTGTTCCAXIN2ForwardTGTCCAGCAAAACTCTGAGGReverseGTGCAAAGACATAGCCAGAACPTGS2ForwardACAGGCTTCCATTGACCAGReverseTCACCATAGAGTGCTTCCAACACTBForwardCATGTACGTTGCTATCCAGGCReverseCTCCTTAATGTCACGCACGAT Primers used in the experiment. Western blot (WB) Total proteins were isolated using strong RIPA buffer (Cat. #P0013B, Beyotime, China) and then quantified using the BCA kit (Cat. #A55864, Thermo Scientific™, the USA). A total of 30 μg protein were separated by SDS-PAGE and then transferred to the PVDF membranes. After blocking with 5% skimmed milk for 1 h at room temperature and washing with PBST, the membranes were incubated with primary antibody overnight at 4 °C. Following three 15-min PBST washes, the membrane was incubated with secondary antibody for 1 h at room temperature. Finally, the bands were washed and scanned with Bio-Rad chemiluminescence imaging system. The antibodies used were listed at Table 2. Table 2Antibodies used in the experiment.Table 2AntibodiesCatCorporationApplicationDilution ratios for WBFlag-tag66008-4-IgProteintechWB, IP1:2000MYC-tag16286-1-APProteintechWB, IP1:2000PLPP117967-1-APProteintechIPNot applicablePLPP1ab198280AbcamWB1:500β-cateninA19657ABclonalWB, IP1:1000c-Myc#18583CSTWB1:1000COX227308-1-APProteintechWB1:1000AXIN220540-1-APProteintechWB1:1000Cyclin D160186-1-IgProteintechWB1:1000β-Tublin#T0023Affinity BiosciencesWB1:5000 Antibodies used in the experiment. Immunohistochemistry (IHC) and hematoxylin-eosin (H&E) staining The tumors and lung tissues were fixed with 4% paraformaldehyde (PFA), dehydrated, embedded and sliced. The IHC and H&E staining were performed as our previous studies [18]. The antibodies used were listed at Table 2. For human OS tissues, a total of 35 samples were collected for PLPP1 IHC analysis, and corresponding clinicopathological information was obtained. PLPP1 expression was evaluated independently by an experienced pathologist using the H-score system. Generally, the staining intensity was scored as 0 (negative), 1 (weak), 2 (moderate), or 3 (strong). The H-score was calculated using the following formula: H-score = Σ (percentage of positively stained cells × staining intensity), resulting in a total score ranging from 0 to 300. Based on the median H-score, patients were divided into high-PLPP1 and low-PLPP1 expression groups. The chi-square test was used to analyze the differences in clinicopathological characteristics between the two groups. Lentiviral packaging and transfection Lentiviral plasmids for PLPP1 overexpression and silencing were purchased from MiaoLing Plasmid (China). Plasmids were transformed into Escherichia coli DH5α for amplification and subsequently extracted. The lentiviral transfer plasmid was co-transfected with the packaging plasmids psPAX2 and pMD2.G into HEK293T cells. After 48 h, viral supernatants were collected, filtered, concentrated by centrifugation, aliquoted, and stored at −80 °C. Besides, luciferase-encoding lentivirus were purchased form Shanghai Genechem Co., Ltd. 143B and HOS cells were seeded in 6-well plates and transduced with the viral supernatant. After 48 h post-transduction, the culture medium was replaced, and fluorescence was observed under a fluorescence microscope. Puromycin was then applied to generate stable cell lines, which were then expanded and cryopreserved. CCK8 and cloning colony formation assays The CCK8 and colony formation assays were performed to evaluate the proliferation abilities of cells. Experimental procedures followed previous studies [18]. For the CCK8 assay, the optical density (OD) values of cells in different treatment groups were continuously measured for 7 consecutive days using the CCK8 reagent. Furthermore, to determine the half maximal inhibitory concentration (IC50) values of the drugs, the OD values of the NC and PLPP1-OE OS cells were measured 24 h after treatment with DOX and CDDP. For the colony formation assay, colonies formed by 143B and HOS cells were collected and analyzed on day 14. EdU assay EdU-555 kit (Cat. #CX003, Epizyme Biotech, Shanghai) was used to detect the ability of cell proliferation. Stably transfected cells were seeded into confocal microscopy plates and cultured for 48 h. EdU reagent was then added, and the cells were incubated for an additional 2 h. The culture medium was subsequently removed, and the cells were fixed with 4% paraformaldehyde for 15 min. EdU staining was performed according to the manufacturer's instructions. After staining, images were acquired using the confocal microscope, and quantitative analysis was conducted by ImageJ software. Wound healing assay Stably transfected cells were seeded into 6-well plates and cultured until confluence. A vertical scratch was introduced using a 200 μL pipette tip, with the bottom of each well marked for reference. After washing with PBS, the cells were incubated in serum-free medium. Microscopic images of the scratch were captured at 0 h and at the indicated time points at the same marked positions. The scratch width was subsequently quantified using ImageJ software. Migration and invasion assays The migration and invasion assays were performed as previous study [19]. Briefly, transwell chambers with or without Matrigel coating were used for the invasion and migration assays, respectively. After pretreatment, medium containing 20% FBS was added to the lower chamber. Cells were digested and resuspended in serum-free medium, and 2 × 104 143B or HOS cells were seeded into the upper chamber. After incubation for 24 h, the chambers were collected. The cells remaining on the upper surface were gently removed, followed by fixation and staining of the migrated or invaded cells. The images were acquired using a microscope and analyzed with ImageJ software. Apoptosis analysis Apoptosis was assessed using Annexin V-APC/PI double staining (Cat. BB-41025, Bestbio, Shanghai, China) according to the manufacturer's instructions. Briefly, stable OS cells were seeded into 6-well plates and treated with CDDP. Then, cells were harvested, washed twice with cold PBS, and resuspended in 1× binding buffer. Subsequently, cells were incubated with 5 μL Annexin V-APC and 5 μL PI solution for 15 min at room temperature in the dark. Stained cells were immediately analyzed using a flow cytometer. Data were processed with FlowJo software. TOP/FOP-flash Cells were seeded into 12-well plates and transfacted with 600 ng TOP-flash, 600 ng FOP-flash and 5 ng pRL-TK plasmids. The culture medium was replaced 6 h after transfection. After 36 h, the luciferase activity was detected by Dual-Luciferase® Reporter Assay System (Cat. #E1910, Promega, the USA). Renilla luciferase chemiluminescence was first measured, followed by the detection of TOP and FOP luciferase activities. TOP and FOP luciferase activities were individually normalized to Renilla luciferase activity as an internal control, and the TOP/FOP ratio was subsequently calculated. Immunofluorescence (IF) Cells were seeded into confocal dishes and cultured overnight. The cells were then fixed with 4% PFA, permeabilized with 0.1% Triton X-100, and blocked with 5% BSA for 1 h. Primary antibody were added and incubated overnight at 4 °C. After washing with PBST, the cells were incubated with fluorescently labeled secondary antibodies at room temperature in the dark for 1 h. Following additional PBST washes, images were acquired using a confocal microscope, and the nuclear protein fluorescence intensity was quantified using ImageJ software. Molecular docking The three-dimensional structures of the target proteins, PLPP1 (UniProt: O14494), Caveolin-1 (UniProt: Q03135), and β-catenin (UniProt: P35222), were retrieved through UniProtKB cross-references to the Protein Data Bank (PDB). When no suitable experimentally resolved structure was available, as in the case of PLPP1, the predicted structure was obtained from the AlphaFold Protein Structure Database (AF-O14494-F1). Structures were prepared in PyMOL 2.6.1 by removing crystallographic water, ions, and co-crystallized heteroatoms (non-protein organic molecules). Each cleaned structure was imported into AutoDockTools 1.5.6 for addition of polar hydrogens, assignment of Gasteiger charges, definition of AutoDock atom types, and export to PDBQT format. AutoDock Vina was used for molecular docking; grid boxes were centered on the binding site or on the predicted interface for PLPP1–Caveolin-1 and Caveolin-1–β-catenin analyses (one partner as receptor, the other as ligand in PDBQT). Docking poses were ranked by the predicted binding affinity (ΔG, kcal·mol−1) reported by Vina, with more negative values indicating stronger predicted binding. Only poses with an affinity ≤ −5.0 kcal mol−1 were considered to represent favorable binding and were retained for further analysis; among these, the pose with the lowest (most negative) score was selected for visualization and interaction analysis in PyMOL 2.6.1. Poses with affinities weaker than −5.0 kcal mol−1 were interpreted as weak or unlikely to form stable interactions under the docking conditions. Co-immunoprecipitation (CO-IP) Cells were lysed using immunoprecipitation lysis buffer (Cat. #PC105, Epizyme Biotech, Shanghai), and the lysates were centrifuged at 15,000 rpm for 15 min at 4 °C. The supernatant was isolated into a 1.5 mL microcentrifuge tube, with a 10% aliquot reserved as the input control. The remaining supernatant was immunoprecipitated with the primary antibody or an isotype-matched IgG control overnight at 4 °C. Subsequently, pre-washed protein A/G magnetic beads were then added, and the mixture was rotated at 4 °C for 2 h. The beads were washed with pre-chilled PBST, resuspended in loading buffer, and boiled to elute the bound proteins, which were subsequently analyzed by WB. Dual-luciferase reporter assay To construct the wild-type PLPP1 promoter plasmid (pPLPP1-WT), the promoter region was cloned into the pGL4.20-basic vector (Promega, USA). Potential c-Myc binding sites within the PLPP1 promoter were predicted using the JASPAR database (http://jaspar.genereg.net). The predicted binding sites were subsequently mutated to generate the mutant plasmid (pPLPP1-Mut). Next, HEK293T cells were co-transfected with either pPLPP1-WT or pPLPP1-Mut, alongside the c-Myc expression vector (pEnCMV-c-Myc) or its corresponding empty vector, together with the pRL-TK plasmid as an internal control. After 36 h of incubation, luciferase activities were determined using the Dual-Luciferase® Reporter Assay System. Chromatin immunoprecipitation-polymerase chain reaction (CHIP-PCR) The ChIP-PCR assay was performed using a commercial ChIP Assay Kit (Cat. #P2078, Beyotime, China), following the manufacturer's instructions. The PCR primers used are listed in Table 1. RNA sequences To further clarify the possible mechanism, RNA-seq was used to detect the differentially expressional RNA between NC or PLPP1 overexpression groups. Cultured stable NC or PLPP1 overexpression 143B cells were rinsed with cold PBS and lysed with TRIZOL buffer. Transcriptome sequencing was performed on the Illumina HiSeq 2000 platform by Tsingke Biotechnology (Hangzhou, China). Differentially expressed genes (DEGs) were identified based on fold change ≥ 1.5 and P value < 0.05. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis and gene set enrichment analysis (GSEA) were conducted and visualized using R software by Tsingke Biotechnology Co., Ltd. For KEGG enrichment analysis, the Benjamini–Hochberg method was applied to control the false discovery rate (FDR), and adjusted P < 0.05 was considered statistically significant. Animal experiments Subcutaneous tumor and tail vein injection of nude mice models were performed to evaluate the abilities of tumor growth and metastasis. All animal experiments were approved by the Ethics Committee of the First Affiliated Hospital of Nanchang University and conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals. A total of 24 female BALB/c nude mice (4 weeks old, approximately 16 g) were used in the animal experiments, with 12 assigned to the subcutaneous xenograft model and 12 to the tail vein injection model. All mice were housed under specific pathogen-free (SPF) conditions, and all experimental procedures complied with the institutional guidelines for animal care and use. In the subcutaneous tumor model, 12 mice were randomly divided into a negative control (NC) group and a PLPP1-overexpression group. NC or PLPP1-overexpressing 143B cells (1 × 106) were subcutaneously injected into the mice. Tumor growth was monitored regularly, and tumor volume was calculated using the formula: volume = (length × width2)/2. On the day 22, tumors were excised, weighed, and fixed for further analysis. In the tail vein metastasis model, another 12 mice were similarly divided into NC and PLPP1-overexpression groups. Luciferase-labeled NC or PLPP1-overexpressing 143B cells were injected into the tail vein. After three weeks, in vivo bioluminescence imaging was performed. Subsequently, lung tissues were collected, fixed, and subjected to hematoxylin–eosin staining. Statistical analysis Each experiment was independently performed in triplicate. Data are presented as mean ± standard deviations (SD) from three independent biological replicates. Data analysis and visualization were conducted using GraphPad Prism software (version 10.0.0). Comparisons between two groups were assessed with an independent samples t-test, while differences among multiple groups were evaluated using one-way ANOVA. During the survival analysis, the “surv_cutpoint” function was used to determine the optimal cut-off value. Patients were grouped according to the cut-off value, and the Log-rank test was used to detect differences in survival. Additionally, Pearson correlation analysis was used to detect the correlation between PLPP1 and MYC (c-MYC, encoding c-Myc) and CCND1 (encoding Cyclin D1). Differences in clinical characteristics between patients in the high-PLPP1 and low-PLPP1 groups were analyzed using the chi-square test. A P value of less than 0.05 was regarded as statistically significant. Result PLPP1 was down-regulated and served as a prognostic indicator in OS To identify potential prognostic markers, we downloaded and analyzed the RNA-seq and clinical data of 85 OS patients from the TARGET-OS dataset. Uni-Cox regression analysis yielded a total of 1948 prognostic related genes (Fig. 1A). Overlap analysis using a Venn diagram highlighted PLPP1 as a prognosis-related gene involved in sphingolipid metabolism (Fig. 1A). Moreover, K-M plots demonstrated that higher PLPP1 expression correlated with favorable patient outcomes (Fig. 1B). Then, we evaluated PLPP1 mRNA levels in NCs and OSs using the GSE14359 and GSE42352 datasets. Consistently, PLPP1 expression was significantly downregulated in OSs compared with NCs (Fig. 1C). These results implied a potential prognostic regulatory role of PLPP1 in OS, highlighting its value for further investigation. Therefore, we examined the mRNA and protein expression levels of PLPP1 in hFOB1.19 cells and five OS cell lines, and found that both mRNA and protein levels of PLPP1 were significantly lower in OS cell lines (Fig. 1D–E). Meanwhile, protein expression levels of PLPP1 were examined in 12 paired OS and adjacent normal tissues. WB analysis revealed that the protein level of PLPP1 was significantly lower in OS tissues compared to adjacent normal tissues (Fig. 1F-G). Furthermore, IHC results demonstrated that protein levels of PLPP1 were reduced in OS patients with distant metastasis relative to those without metastasis (Fig. 1H). To further investigate the clinical significance of PLPP1, we assessed its protein expression in 35 OS patients via IHC and stratified them into high- and low-PLPP1 expression groups based on the median score. Clinicopathological correlation analysis revealed that low PLPP1 expression was significantly associated with a higher incidence of lung metastasis and more advanced clinical stages (Table 3). Taken together, these results identified PLPP1 as a novel downregulated marker in OS with potential tumor-suppressive activity. PLPP1 was down-regulated and served as a prognostic indicator in OS. (A) The Venn plot between 1948 prognosis-related genes and 30 sphingolipid metabolism-related genes. (B) The K-M plotters of patients in High- and Low-PLPP1 expression groups form TARGET-OS, GSE21257 and KM-plotter datasets. (C) The mRNA expression levels of PLPP1 between normal cells (NCs) and OS cells (OSs) in GSE14359 and GSE42352 datasets. (D) The mRNA expression levels of PLPP1 between hFOB1.19 and five OSs (U2OS, HOS, 143B and Saos-2). (E) The protein levels of PLPP1 between hFOB1.19 and five OSs. (F) The protein levels of PLPP1 between OS tumor tissues (T) and paired adjacent normal tissue (N) (n = 12). (G) T Semi-quantitative analysis showed that PLPP1 protein levels were significantly lower in OS tumor tissues than in adjacent normal tissues (n = 12). (H) The representative IHC images of OS tissues from patients with or without metastasis (scale bar = 200 μm). *, P < 0.05; **, P < 0.01; ****, P < 0.0001. Table 3Correlation between PLPP1 expression levels and clinicopathological characteristics in patients.Table 3Clinical characteristicNumberPLPP1 expression levelP valueLowHighGender0.404 Male1679 Female19118Age0.380 <1820911 ≥181596Metastasis0.020 M1871 M0271116Tumor size0.053 <10 cm31138 ≥10 cm324159AJCC staging0.032 I&II251015 III&IV1082Primary tumor site0.735 Limbs281414 Others734 Correlation between PLPP1 expression levels and clinicopathological characteristics in patients. PLPP1 inhibited the proliferation, migration and invasion of OS cells Subsequently, HOS and 143B cells were selected to overexpress or silence PLPP1 for functional assays. The efficiency of PLPP1 overexpression and knockdown in 143B and HOS cells was successfully validated by qRT-PCR and WB analyses (Sup Fig. 1A–D). CCK-8 analysis revealed a suppressive role of PLPP1 in OS cell proliferation, where its overexpression inhibited and silencing stimulated cell growth (Fig. 2A-B). This inhibitory effect was further substantiated by colony formation assays, in which ectopic PLPP1 expression severely restricted the number and size of colonies in 143B and HOS cells, while its depletion noticeably promoted clonogenic survival (Fig. 2C). In addition, the percentage of EdU-positive cells significantly decreased after PLPP1 up-regulation but significantly elevated following its knockdown, further confirming the anti-proliferative role of PLPP1 (Fig. 2D). Furthermore, transwell assays revealed that PLPP1 overexpression significantly suppressed the migratory and invasive potentials of OS cells (Fig. 2E–F), whereas PLPP1 knockdown exerted the opposite effect (Fig. 2G–H). Consistent with these findings, wound healing assays confirmed that PLPP1 overexpression impaired, whereas PLPP1 silencing promoted, the wound closure capacity of the cells (Fig. 2I–J). Moreover, CCK-8 assays demonstrated that PLPP1 overexpression significantly decreased the IC50 values of CDDP and DOX in HOS and 143B cells (Sup Fig. 2A). The flow cytometry and WB analyses demonstrated that PLPP1 overexpression significantly enhanced CDDP-induced apoptosis in OS cells (SupFig. 2B–D). Collectively, these findings established PLPP1 as a key suppressor of OS progression and a potential modulator of chemosensitivity. PLPP1 inhibited the proliferation, migration and invasion of OS cells. (A-B) CCK-8 assay showing that PLPP1 overexpression inhibited, whereas PLPP1 knockdown promoted, proliferation of 143B and HOS cells. (C) Colony formation assays showing that PLPP1 overexpression inhibited, while PLPP1 knockdown promoted, clonogenicity of 143B and HOS cells. (D) EdU assay showing the effects of PLPP1 overexpression and knockdown on OS cells (scale bar = 72.7 μm). (E-F) PLPP1 overexpression inhibited migratory and invasive abilities of OS cells (scale bar = 100 μm). (G-H) Silencing of PLPP1 promoted the migration and invasion of OS cells. (I) PLPP1 overexpression inhibited the wound healing capacity of OS cells. (J) Silencing of PLPP1 promoted the wound healing ability of OS cells. *, P < 0.05; **, P < 0.01. Overexpression of PLPP1 suppressed OS growth and metastasis in vivo To further elucidate the role of PLPP1 in OS progression, we utilized a subcutaneous xenograft model to assess tumor growth and a tail vein injection model to evaluate lung metastasis. The results demonstrated that tumors in the PLPP1 overexpression group were significantly smaller than those in the control group (Fig. 3A). Additionally, the growth curve revealed that PLPP1 overexpression markedly slowed tumor proliferation (Fig. 3B). Consistently, tumor weight was substantially reduced in the PLPP1 overexpression group (Fig. 3C). IHC analysis further revealed a decreased Ki67-positive rate in tumors overexpressing PLPP1 (Fig. 3D). To evaluate metastatic potential, an experimental lung metastasis model was established via tail vein injection of a stable luciferase-expressing cell line. Quantitative in vivo bioluminescence imaging was performed to evaluate metastatic dissemination to the lungs (Fig. 3E). Mice in the PLPP1-overexpression cohort exhibited a substantially attenuated pulmonary bioluminescent signal relative to the control counterparts (Fig. 3E–F). H&E staining of lung sections revealed a marked reduction in pulmonary metastatic burden, with both fewer and smaller metastatic lesions in the PLPP1-overexpression group compared with the control group (Fig. 3G–H). These functional assays demonstrated that PLPP1 acted as a potent tumor suppressor in OS, not only inhibiting cell proliferation, motility and invasion but also enhancing chemotherapy sensitivity. Overexpression of PLPP1 suppressed OS growth and metastasis in Vivo.(A) Photographed images of tumor tissues from normal control (NC) and PLPP1 overexpressing (PLPP1-OE) groups (n = 6). (B) Tumor growth curves showing that PLPP1 overexpression impaired the subcutaneous proliferation (n = 6). (C) Comparison of tumor volumes between the NC and PLPP1-OE groups (n = 6). (D) The H&E staining and IHC staining of tumor tissues between NC and PLPP1-OE groups (scale bar = 200 μm). (E) Bioluminescence images showing lung metastases in NC and PLPP1-OE groups (n = 6). (F) Quantification of bioluminescence intensity indicating metastatic burden in the lungs (n = 6). (G) H&E staining showing lung tissue morphology and metastatic nodules in control and PLPP1-overexpressing groups (n = 6, scale bar = 4000 μm). (H) Quantification of metastatic nodules in the lungs. **, P < 0.01. Over-expression of PLPP1 inhibited wnt/β-catenin signaling pathway To investigate the molecular mechanisms underlying the inhibitory effects of PLPP1 on OS proliferation and metastasis, we conducted transcriptome sequencing of 143B cells stably overexpressing PLPP1 and controls (Sup Fig. 3A). A total of 133 DEGs were identified, comprising 62 upregulated and 71 downregulated genes (Fig. 4A–B). KEGG analysis revealed significant enrichment in tumor progression- and immune-related pathways (Sup Fig. 3B). Among these, the Wnt/β-catenin signaling axis was reported as a key regulator of OS progression [20]. And GSEA further supported the involvement of PLPP1 overexpression in modulating the Wnt/β-catenin pathway (Sup Fig. 3C). Over-expression of PLPP1 inhibited Wnt/β-catenin signaling pathway. (A) RNA-seq analysis of DEGs between the control and PLPP1-overexpression groups. (B) Heatmap showing the expression profiles of DEGs. (C) TOP/FOP reporter assays demonstrated the regulatory effects of PLPP1 overexpression or knockdown on Wnt/β-catenin signaling activity. (D-E) qRT-PCR results of wnt/β-catenin target genes (CTNNB1, c-Myc, PTGS2, CCND1 and AXIN2) after overexpression or knockdown of PLPP1. (F-G) WB images for β-catenin, c-Myc, COX2, AXIN2 and CyclinD1 in cells with overexpression or knockdown of PLPP1. (H) IHC analysis of protein expression in subcutaneous tumors from NC and PLPP1-OE mice (scale bar = 200 μm). (I) Correlation analysis of PLPP1 and MYC, CCND1 transcriptional levels in sarcoma samples from the GEPIA2 database. ns, P > 0.05; *, P < 0.05; **, P < 0.01. Given these observations, we next evaluated Wnt/β-catenin transcriptional activity using the TOP/FOP-Flash reporter assay. The TOP/FOP-Flash reporter assay revealed that PLPP1 overexpression markedly reduced the TOP/FOP ratio, whereas PLPP1 knockdown elicited the opposite effect (Fig. 4C). Subsequently, we examined several downstream target genes involved in Wnt/β-catenin transcriptional regulation, including c-MYC, PTGS2 (encoding COX-2), CCND1, and AXIN2 (encoding AXIN2) as well as CTNNB1, the gene encoding β-catenin. In line with these results, PLPP1 overexpression significantly downregulated the mRNA expression of c-MYC, PTGS2, CCND1 and AXIN2, yet with minimal effect on CTNNB1 transcript levels (Fig. 4D). Conversely, PLPP1 silencing significantly increased c-MYC, PTGS2, and CCND1 mRNA expression, while only modestly affecting CTNNB1 (Fig. 4E). Parallel alterations were observed at the protein level, where PLPP1 overexpression markedly reduced c-Myc, COX-2, AXIN2, and Cyclin D1 expression, whereas its ablation yielded reciprocal trends (Fig. 4F–G and Sup Fig. 3D–E). However, β-catenin protein abundance remained unchanged upon either PLPP1 overexpression or silencing (Fig. 4F–G). In line with in vitro findings, the expression levels of these downstream targets (c-Myc, COX2, AXIN2, and Cyclin D1) were consistently repressed in tumor xenografts following PLPP1 overexpression, whereas β-catenin protein abundance remained largely unaltered between the two groups (Fig. 4H). Additionally, mRNA expression levels of MYC and CCND1 in SARC tissues from GEPIA2 database were negatively correlated with those of PLPP1 (Fig. 4I). Taken together, these results demonstrated that PLPP1 suppressed OS progression by modulating the Wnt/β-catenin axis, without altering β-catenin protein abundance. Wnt/β-catenin intervention reversed the tumor phenotypes induced by PLPP1 alteration Given that the Wnt/β-catenin signaling pathway mediates the effects of PLPP1 in inhibiting OS progression, we investigated whether modulating Wnt/β-catenin signaling activity influenced PLPP1-induced cellular phenotypes. To this end, MSAB, a specific inhibitor of Wnt/β-catenin signal, was utilized to treat PLPP1-knockdown OS cells [21]. Immunoblotting demonstrated that β-catenin overexpression rescued the downregulation of c-Myc, COX2, AXIN2 and Cyclin D1 proteins that induced by PLPP1 overexpression (Fig. 5A and Sup Fig. 3F). Conversely, MSAB treatment abrogated the upregulation of these target proteins in shPLPP1#2 OS cells (Fig. 5B and Sup Fig. 3G). Importantly, β-catenin overexpression effectively counteracted the suppressive effects of PLPP1 overexpression on proliferation, migration and invasion of OS cells (Fig. 5C, E, G, I), indicating that the Wnt/β-catenin signaling axis was involved in the regulation of PLPP1 in OS. Moreover, functional assays revealed that MSAB treatment diminished the pro-proliferative effects of PLPP1 silencing (Fig. 5D, F). Wound-healing and Transwell assays further demonstrated that MSAB attenuated the enhanced migratory and invasive capacities induced by PLPP1 depletion (Fig. 5H, J) These findings demonstrated that PLPP1 exerted a tumor-suppressive role in OS by inhibiting cell proliferation, migration and invasion via the inactivation of the Wnt/β-catenin signaling axis. Wnt/β-catenin intervention reversed the tumor phenotypes induced by PLPP1 alteration. (A) WB results of Wnt/β-catenin target proteins after co-overexpression of PLPP1 and β-catenin. (B) WB results of Wnt/β-catenin downstream target proteins in OS cells with PLPP1 knockdown followed by treatment with 1 μM MSAB for 24 h. (C) Effect of β-catenin overexpression on cell viabilities of OS cells with PLPP1 overexpression was detected by CCK8 kit at 7 day. (D) Cell viability was determined after treatment with MSAB (0.5 or 1 μM) for 7 days in shPLPP1-expressing OS cells. (E) The colony forming abilities of cells with co-overexpression of PLPP1 and β-catenin. (F) Colony-forming ability of shPLPP1#2 OS cells treated with MSAB (0.5 μM) for 14 days. (G-H) Wound healing abilities of PLPP1-overexpressing OS cells with β-catenin overexpression, and shPLPP1 OS cells treated with MSAB (0.5 μM). (I-J) Effects of β-catenin overexpression in PLPP1-overexpressing OS cells, and MSAB treatment (0.5 μM for 24 h) in shPLPP1 OS cells on cell migration and invasion (scale bar = 100 μm). *, P < 0.05; **, P < 0.01. PLPP1 inhibited β-catenin nuclear translocation by enhancing caveolin-1/β-catenin interaction As above data revealed, PLPP1 inhibited the expression of downstream target proteins in the Wnt/β-catenin signaling pathway, without altering the total mRNA or protein levels of β-catenin. Given that the activation of canonical Wnt/β-catenin signaling relies on the nuclear translocation of β-catenin to execute its transcriptional regulatory functions [22], we investigated whether PLPP1 influences β-catenin nuclear localization. WB and IF assays demonstrated that overexpression of PLPP1 significantly restricted β-catenin nuclear translocation (Fig. 6A–B). To elucidate the underlying mechanism, the IP-MS analysis was performed to identify potential PLPP1-interacting proteins (Sup Fig. 4A–B). After excluding IgG-bound nonspecific proteins, a total of 140 PLPP1-interacting proteins were identified. Functional enrichment analysis using Metascape (https://metascape.org/) revealed that 11 of these proteins were involved in the Wnt signaling pathway (Sup Fig. 4C). Among them, CAV1 (encoding caveolin-1) emerged as a potential candidate, as it has been previously reported to bind β-catenin and sequester it from nuclear import [23]. We therefore hypothesized that caveolin-1 might mediate the inhibitory effect of PLPP1 on β-catenin nuclear import. PLPP1 inhibited β-Catenin nuclear translocation by enhancing Caveolin-1/β-Catenin interaction. (A) Overexpression of PLPP1 suppressed the nuclear import of β-catenin, as determined by WB. (B) IF detection of the effect of PLPP1 overexpression on the nuclear translocation of β-catenin. (C) Molecular docking analysis predicting the interaction between PLPP1 and caveolin-1. (D) Exogenous CO-IP detecting the interaction between Flag-PLPP1 and MYC-caveolin-1. (E-F) Endogenous CO-IP confirming the interaction between PLPP1 and caveolin-1. (G) WB analysis showing that PLPP1 overexpression did not alter caveolin-1 protein levels. (H) CO-IP and WB analyses evaluating the impact of PLPP1 on the interaction between β-catenin and caveolin-1. (I) The effects of overexpression of PLPP1 and silencing of Caveolin-1 on the nuclear translocation of β-catenin. *, P < 0.05; **, P < 0.01; ****, P < 0.0001. To test this hypothesis, we first determined whether caveolin-1 interacts with β-catenin and inhibits its nuclear import in OS cells. Molecular docking analysis revealed a binding energy of −7.1 kcal/mol between caveolin-1 and β-catenin, indicating a strong likelihood of interaction (Sup Fig. 4D). CO-IP experiments further confirmed the physical interaction between caveolin-1 and β-catenin (Sup Fig. 4E–F). Interestingly, although siRNA-mediated silencing of caveolin-1 did not alter total β-catenin protein levels, it significantly enhanced β-catenin nuclear import (Sup Fig. 4G–H) and increased the TOP/FOP Flash reporter ratio (Sup Fig. 4I). These findings consistently demonstrated that caveolin-1 knockdown activated the Wnt/β-catenin signaling pathway in OS cells. Next, we explored whether PLPP1 modulates β-catenin nuclear import through its interaction with caveolin-1. Molecular docking studies revealed a binding energy of −6.7 kcal/mol between PLPP1 and caveolin-1 (Fig. 6C). Reciprocal Co-IP experiments using exogenously expressed tagged proteins confirmed that Flag-PLPP1 interacted with Myc-caveolin-1, and endogenous CO-IP analysis also demonstrated an interaction between PLPP1 and caveolin-1 (Fig. 6D–F). Notably, overexpression of PLPP1 did not affect caveolin-1 protein levels (Fig. 6G). Given that caveolin-1 inhibited β-catenin nuclear import by physically binding to it, we reasoned that PLPP1 might strengthen the caveolin-1/β-catenin complex. Instructively, CO-IP and WB experiments revealed PLPP1 overexpression markedly increased the amount of caveolin-1 that co-immunoprecipitated with β-catenin (Fig. 6H). Furthermore, silencing caveolin-1 in PLPP1-overexpressing cells reversed the PLPP1-induced reduction in β-catenin nuclear import (Fig. 6I). This suggested that PLPP1 inhibited the nuclear import of β-catenin by enhancing the caveolin-1/β-catenin interaction, thereby compromising canonical Wnt/β-catenin signaling pathway in OS cells. PLPP1 was transcriptionally repressed by c-myc To promote rapid proliferation and metastasis, PLPP1 expression is typically maintained at a low level in OS. To investigate the mechanisms underlying this downregulation, we screened for potential upstream transcription factors. An overlap analysis between Cistrome DB, GTRD, hTFtargets databases and 1948 prognosis-associated genes identified CEBPA and c-Myc as potential candidates (Fig. 7A). KM plotters revealed that high c-Myc expression and low CEBPA expression were both linked to poor prognosis in OS patients (Fig. 7B and Sup Fig. 5A). Given that previous studies revealed that CEBPA inhibits OS progression, while c-Myc promotes proliferation and metastasis [24,25], We hypothesized that c-Myc represseed PLPP1 expression, while CEBPA might activate PLPP1 expression to modulate OS progression. PLPP1 was transcriptionally repressed by c-Myc. (A) Venn diagram analysis identifying potential transcription factors regulating PLPP1. (B) K-M survival analysis illustrating the prognostic value of c-Myc expression in patients with OS and sarcoma. (C) qRT-PCR analysis of PLPP1 mRNA levels following c-Myc overexpression. (D) WB analysis showing that overexpression of c-Myc reduced PLPP1 protein levels. (E) Dual-luciferase assays of cells co-transfected with pPLPP1-WT and c-Myc plasmids. (F) Schematic representation of the predicted c-Myc binding motif on the PLPP1 promoter and the construction of the mutant plasmid. (G) Dual-luciferase reporter assays in cells co-transfected with c-Myc and either pPLPP1-WT or pPLPP1-mut plasmids. (H) ChIP-PCR analysis demonstrating the direct binding of c-Myc to the PLPP1 promoter. *, P < 0.05; **, P < 0.01. We firstly tested whether CEBPA regulated PLPP1 transcription. WB analysis indicated lower CEBPA levels in OS cells compared to hFOB1.19 cells (Sup Fig. 5B), suggesting a potential regulatory relationship. However, overexpression of CEBPA in OS cells failed to alter PLPP1 mRNA or protein levels (Sup Fig. 5C–D). Furthermore, CEBPA overexpression did not enhance the luciferase activity of the pPLPP1-WT promoter (Sup Fig. 5E), indicating that CEBPA did not regulate PLPP1 transcription. Then, we investigated the regulatory role of c-Myc on PLPP1. Overexpression of c-Myc in 143B and HOS cells significantly decreased PLPP1 mRNA and protein levels (Fig. 7C–D and Sup Fig. 5F). Dual-luciferase reporter assays demonstrated that c-Myc inhibited luciferase activity of the pPLPP1-WT (Fig. 7E). Since c-Myc requires heterodimerization with the MAX protein to bind DNA and exert its transcriptional regulatory function, we used the c-Myc:MAX complex to predict its binding sites within the PLPP1 promoter. Two potential binding sites were identified using the JASPAR database: Site 1 (ATCATGTGGC) and Site 2 (TATCATGTGGC) (Sup Fig. 5G). These sites were only one nucleotide apart, suggesting they represent a single binding site, CATGTGGC. Based on these predictions, we mutated the wild-type pPLPP1-WT plasmid (Fig. 7F). Dual-luciferase reporter assays confirmed that overexpression of c-Myc significantly inhibited luciferase activity of the pPLPP1-WT promoter, while the mutant plasmid (pPLPP1-Mut) partially reversed this effect (Fig. 7G). Furthermore, ChIP-PCR experiments further validated that c-Myc binds to the CATGTGGC site in the PLPP1 promoter (Fig. 7H). In summary, these data confirm that c-Myc directly binds to the PLPP1 promoter region and suppresses its transcription. PLPP1 abrogated c-myc-driven OS progression and wnt/β-catenin activation via a negative feedback loop Our preceding findings indicated that PLPP1 inhibited Wnt/β-catenin signaling and its downstream effector c-Myc, while c-Myc reciprocally suppressed PLPP1 transcription. To systematically investigate the functional interplay of this negative feedback loop in OS progression, we established four stable OS cell models: a negative control group (NC), c-Myc overexpression (c-Myc-OE), PLPP1 overexpression (PLPP1-OE), and c-Myc/PLPP1 co-overexpression (c-Myc + PLPP1-OE). WB analysis revealed that the expression levels of Wnt/β-catenin downstream targets, including COX2, Cyclin D1, and AXIN2, as well as c-Myc itself, were significantly elevated in the c-Myc-OE group, confirming that c-Myc feedback-amplified Wnt/β-catenin activity (Fig. 8A and Sup Fig. 5H). Conversely, the PLPP1-OE group exhibited the opposite trend. Crucially, co-overexpression of PLPP1 significantly counteracted the c-Myc-induced up-regulation of these downstream targets, effectively neutralizing the c-Myc-mediated activation of the Wnt/β-catenin pathway (Fig. 8A and Sup Fig. 5H). Consistent with the molecular changes, functional characterization demonstrated that c-Myc overexpression significantly accelerated the proliferation, colony formation, and motility of 143B and HOS cells (Fig. 8B–E). However, these c-Myc-driven oncogenic phenotypes were markedly attenuated or reversed upon PLPP1 co-overexpression, as evidenced by reduced cell viability, fewer colonies, and impaired migratory/invasive capacities (Fig. 8B–E). PLPP1 abrogated c-Myc-driven OS progression and Wnt/$\beta$-catenin activation via a negative feedback loop. (A) WB analysis of the effect of overexpression of PLPP1 on downstream proteins of Wnt/β-catenin in cells overexpressing c-Myc. (B-C) The proliferation abilities of stably cells were detected by CCK8 and colony formation assays. (D) Wound healing abilities of stably cells. (E) The migration and invasion abilities of different stably cells. **, P < 0.01. Discussion In this study, we demonstrated that PLPP1 was downregulated in OS cells and samples. And overexpression of PLPP1 suppressed OS proliferation and metastasis both in vitro and in vivo. Mechanistically, PLPP1 promoted the interaction between caveolin-1 and β-catenin, thereby preventing β-catenin nuclear translocation and attenuating canonical Wnt/β-catenin signaling. Moreover, we identified c-Myc as a transcriptional repressor of PLPP1, indicating that PLPP1 overexpression could disrupt the positive feedback loop within the Wnt/β-catenin/c-Myc axis. PLPP1, as a member of LPPs, is significantly downregulated in various malignancies, including melanoma, sarcoma, bladder cancer, breast cancer and colorectal cancer [6]. Consistent with these findings, we observed low PLPP1 expression in OS cells and tissues. Notably, although 143B cells are generally considered more proliferative and metastatic than HOS cells, our data showed relatively higher baseline PLPP1 expression in 143B (Fig. 1E). This phenotypic discrepancy likely reflects the profound genetic and molecular heterogeneity among OS cell lines. In highly aggressive contexts like 143B cells, elevated baseline PLPP1 expression may represent an endogenous, compensatory feedback mechanism triggered by hyperactivated upstream oncogenic signals, which, however, remains insufficient to counteract the potent malignant phenotype driven by other concurrent alterations. Importantly, IHC results revealed that patients with high PLPP1 expression exhibited a markedly lower incidence of lung metastasis compared with those with low PLPP1 expression, highlighting PLPP1 as a promising prognostic biomarker in OS. LPPs exert diverse and context-dependent biological functions across various cancers—for instance, PLPP3 accelerates glioblastoma growth by stabilizing β-catenin and Cyclin D1 proteins, and PLPP2 drives breast cancer cell proliferation by upregulating c-Myc [26,27]. In contrast, previous studies have reported that PLPP1 suppresses the growth and metastasis of breast and ovarian cancers [[9], [10], [11]]. Corroborating these findings, our study confirmed that PLPP1 overexpression inhibited OS cell proliferation, migration, invasion, and lung metastasis. Notably, we further observed that PLPP1 sensitized OS cells to chemotherapeutic agents such as DOX and CDDP, suggesting potential clinical value in therapeutic targeting of PLPP1. Furthermore, pharmacological activation of PLPP1 by compounds such as Z-Ligustilide has been reported to enhance cisplatin sensitivity, further supporting the potential pharmacological applicability of PLPP1-targeted therapeutic strategies in OS [28]. The Wnt/β-catenin signaling pathway is an evolutionarily conserved cascade that plays a pivotal role in regulating multiple cellular functions. Upon Wnt stimulation, stabilized β-catenin accumulates in the nucleus and interacts with TCF/LEF transcription factors to drive the expression of downstream oncogenic genes [29]. Aberrant activation of Wnt/β-catenin pathway, largely dependent on nuclear import of β-catenin, is a well-recognized driver of OS initiation and progression [30]. In this study, we found that PLPP1 suppressed OS progression by inhibiting canonical Wnt/β-catenin signaling. Using IP-MS, we identified 11 PLPP1-interacting proteins enriched in the Wnt/β-catenin pathway. Among them, Caveolin-1 was of particular interest due to its established role in preventing its β-catenin accumulation [23,[31], [32], [33]]. Consistent with previous findings, we observed that silencing caveolin-1 in OS cells markedly enhanced β-catenin nuclear import and transcriptional activity. Crucially, our data revealed that PLPP1 promoted the formation of the Caveolin-1/β-catenin complex, thereby attenuating β-catenin nuclear localization and downstream pathway activation. Collectively, these findings demonstrated that beyond its classical lipid phosphatase activity, PLPP1 exerted a non-canonical function by scaffolding Caveolin-1 to regulate β-catenin nuclear trafficking [6]. Besides, we identified c-Myc as a direct transcriptional suppressor of PLPP1 and uncovered a novel Wnt/β-catenin/c-Myc/PLPP1 regulatory loop in OS. c-Myc, a classical downstream effector of Wnt/β-catenin signaling, is frequently overactivated in OS and contributes to tumor progression [25]. In this study, we found that c-Myc directly bound to the PLPP1 promoter and inhibited its transcription. Down-regulation of PLPP1 subsequently weakened the interaction between Caveolin-1 and β-catenin, thereby facilitating β-catenin nuclear translocation and further enhancing c-Myc expression. This positive feedback loop may contribute to sustained activation of Wnt/β-catenin signaling in OS. Disruption of this oncogenic feedback circuits could improve therapeutic efficacy and reduce pathway reactivation during treatment. Therefore, restoration of PLPP1 expression or targeting this regulatory axis may represent a potential therapeutic strategy for aggressive OS. Beyond the mechanistic insights, our findings have several important translational implications for osteosarcoma. First, PLPP1 may serve as a promising prognostic biomarker for risk stratification and a potential therapeutic target. Mechanistically, restoring PLPP1 activity represents a novel strategy to inhibit Wnt/β-catenin signaling by preventing β-catenin nuclear translocation through the Caveolin-1/β-catenin axis, providing an alternative approach to direct Wnt/β-catenin inhibition. Moreover, therapeutic restoration of PLPP1 disrupted the Wnt/β-catenin/c-Myc positive feedback loop and enhanced the sensitivity of OS cells to conventional chemotherapeutic agents, suggesting its potential for combination therapy. Future studies should focus on developing PLPP1-based therapeutic strategies, including gene delivery systems and small-molecule activators, and validating their efficacy in patient-derived models, orthotopic OS models, and prospective clinical cohorts to facilitate clinical translation. In conclusion, our study demonstrated that PLPP1 was significantly downregulated in OS tissues and cells, while its overexpression markedly inhibited tumor proliferation and lung metastasis in vitro and in vivo. Mechanistically, PLPP1 enhanced the interaction between Caveolin-1 and β-catenin, thereby restricting β-catenin nuclear translocation and suppressing Wnt/β-catenin signaling. Moreover, c-Myc transcriptionally repressed PLPP1, forming a positive feedback loop that sustained Wnt/β-catenin pathway activation and promoted malignant progression. Importantly, PLPP1 serve as both a prognostic biomarker and a potential therapeutic target in OS. Restoring PLPP1 expression or pharmacologically activating this axis may provide a promising strategy for the treatment of aggressive OS. Author contributions J.B. Nie, Y. Zhou and X.S. Xie performed experiments and analyzed the data. J.F. Wu, W.L. Tong, Z.G. Shu, S.J. Wang, J.M. Liu, and F. Yang supervised statistical analysis. Z.L. Liu planned and supervised the study. J.B. Nie drafted the manuscript. All authors read and approved the final paper. Availability of data and materials The data from this study are available from the corresponding author upon reasonable request. Ethics approval and consent to participate This study was approved by the Ethics Committee of the First Affiliated Hospital of Jiangxi Medical College, Nanchang University [(2025)CDYFYYLK(04-015) and CDYFY-IACUC-202308QR004] and followed the principles of the NIH Guide for Care and Use of Laboratory Animals. Consent for publication No individual person's data are given. Declaration of generative AI in scientific writing No AI was used in the writing of this article. Funding This work was partially funded by Grants from the Natural Science Foundation of Jiangxi Province (20252BAC240042; 20232ACB206043) and the “Double Thousand Plan” of Jiangxi Province. Declaration of conflicts of interests A conflict of interest occurs when an individual's objectivity is potentially compromised by a desire for financial gain, prominence, professional advancement or a successful outcome. The Editors of the Journal of Orthopaedic Translation strive to ensure that what is published in the Journal is as balanced, objective and evidence-based as possible. Since it can be difficult to distinguish between an actual conflict of interest and a perceived conflict of interest, the Journal requires authors to disclose all and any potential conflicts of interest. Acknowledgements The mechanistic scheme of this study was drawn using Figdraw (www.figdraw.com). Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.jot.2026.101202. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 References Associated Data Supplementary Materials Multimedia component 1 Data Availability Statement The data from this study are available from the corresponding author upon reasonable request.