WDR3 undergoes phase separation to mediate the therapeutic mechanism of Nilotinib against osteosarcoma

WDR3 undergoes phase separation to mediate the therapeutic mechanism of Nilotinib against osteosarcoma | 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 Research Article WDR3 undergoes phase separation to mediate the therapeutic mechanism of Nilotinib against osteosarcoma Minglei Li, Nan Li, Yuying Fan, Zhan Zhang, Long Zhou, Yifan Yu, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6611672/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Jul, 2025 Read the published version in Journal of Experimental & Clinical Cancer Research → Version 1 posted 9 You are reading this latest preprint version Abstract Background Osteosarcoma is highly invasive with a poor prognosis. The phenomenon of liquid-liquid phase separation (LLPS) can promote the formation of biomolecules and participate in the tumor regulation mechanism. Therefore, mining prognostic markers related to LLPS could allow patients to benefit from targeted therapies. Method Microarray analysis was performed to identify LLPS-related biomarkers, followed by validation via molecular docking analysis. Functions of key genes were investigated in U2-OS cells and xenograft mice. LLPS of the key gene were observed by the droplet formation assay and fluorescence recovery after photobleaching. The intrinsically disordered region (IDR) was predicted and mutated to disrupt LLPS, which was rescued by the fusion of hnRNAP1 IDR. Therapeutic mechanism of Nilotinib mediated by LLPS was explored in vitro and in vivo . Results Five LLPS-related biomarkers were screened by bioinformatics analyses to predict the osteosarcoma prognosis. These prognostic signatures were significantly associated with immune cell infiltration, tumor immune escape and drug sensitivity. Among them, WDR3 was a prognostic risk factor for osteosarcoma and stably bound to Nilotinib in molecular docking models. In transfected U2-OS cells and xenograft mice, downregulation of WDR3 significantly inhibited malignant progression of osteosarcoma. More importantly, WDR3 could form droplets in U2-OS cells and restore the fluorescence intensity of WDR3 condensates with liquid-like behavior after photobleaching. The mutation in IDR could disrupt the phase separation ability of WDR3, whereas the fusion of hnRNAP1 IDR rescued the phase separation abnormality caused by WDR3 mutation. Moreover, treatment with Nilotinib improved the progression of osteosarcoma in vivo and in vitro , while inhibiting the production of WDR3 phase separated condensates. Conclusion WDR3 phase separation involves in the therapeutic mechanism of Nilotinib against osteosarcoma, and thus may serve as a potent biomarker to ameliorate adverse events after osteosarcoma treatment. Osteosarcoma liquid-liquid phase separation prognostic biomarker WDR3 Nilotinib IDR mutation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Highlights 1. LLPS-related prognostic signatures including WDR3 can predict survival status in osteosarcoma. 2. WDR3 and Nilotinib exhibited the stable receptor-ligand binding ability in molecular docking. 3. WDR3 formed droplets in U2-OS cells and promoted tumor metastasis. 4. Nilotinib inhibited osteosarcoma progression by disrupting the phase separation of WDR3. 1. Introduction Osteosarcoma is a primary solid bone tumor of mesenchymal cell origin, commonly found in the metaphyses of skeleton (including the femur and tibia) and knee joints [ 1 – 3 ]. Its incidence is bimodal in age, occurring more in adolescents and older adults over the age of 60, with 4.4 cases of osteosarcoma diagnosed annually per million people worldwide [ 4 , 5 ]. The 5-year survival rate for patients with localized tumor can be 70%, but osteosarcoma is highly aggressive, with distal metastases found in about 15–20% of diagnosed cases and most of them involving the lungs [ 6 – 8 ]. Unfortunately, the metastatic rate of osteosarcoma is over 85% and the 5-year overall survival of metastatic patients is less than 20% [ 9 , 10 ]. In addition to traditional treatments such as surgery, chemotherapy and radiotherapy, targeted therapy and immunotherapy are also widely used in the clinical management of osteosarcoma; however, drug resistance may be the most important reason of reduced survival time, especially for patients with metastatic phenotype [ 11 , 12 ]. Genomic instability and aberrations characterize the majority of osteosarcoma cases [ 13 , 14 ], therefore, the development of potential biomarkers from a genetic perspective is an effective strategy to benefit patient with targeted therapies and protect them from drug resistance. Phase separation is a phenomenon that occurs when a mixture of molecules spontaneously separates into two distinct phases with different compositions and concentrations of specific factors [ 15 , 16 ]. In cellular physiology, biomolecules aggregate to form droplet-like structures in the intracellular environment through liquid-liquid phase separation (LLPS) [ 17 , 18 ]. Generally, the intrinsically disordered region (IDR) drives LLPS through interactions between multiple amino acid residues [ 19 ]. This physiological phenomenon creates a specialized microenvironment that regulates various cellular processes, such as DNA replication, RNA processing, ribosome biogenesis, apoptosis, and signal transduction, thereby affecting tumor progression [ 20 – 22 ]. It was reported that the aggregation of core regulatory circuitry and transcriptional machinery proteins on super-enhancers via LLPS can promote the activation of genes associated with osteosarcoma metastasis or drug resistance [ 23 ]. The inhibition of MYC-driven super-enhancers signaling can counteract the elimination of osteosarcoma cell proliferation, migration, and invasion [ 24 ]. Kim et al. demonstrated that ARID1A promotes the oncogenic potential of osteosarcoma through Prion-like domain-mediated LLPS [ 25 ]. To better study the mechanism of LLPS, Sun et al. summarized various bioinformatics databases and tools [ 26 ], but these approaches have not yet been applied in osteosarcoma to identify biomarkers that may affect disease progression and drug sensitivity. Therefore, this study adopted bioinformatics methods to screen LLPS-related biomarkers with prognostic value for osteosarcoma and explored their relationships with immune landscape and drug sensitivity. Molecular docking simulated binding modalities of prognostic signatures and drugs to screen for receptor-ligand complexes with structural stability. Furthermore, the involvement of phase separation of candidate biomarker in the drug therapy against osteosarcoma was also explored by cell- and animal-based experiments. The biomarkers reported in this study and their mediated phase separation mechanism are expected to provide new theories and ideas for the treatment of osteosarcoma. 2. Methods 2.1 Bioinformatic analysis 2.1.1 Data collection and preprocessing The GSE16091, GSE21257, and GSE39055 datasets containing survival information and gene expression profiling data of 124 osteosarcoma patients were collected from the Gene Expression Omnibus (GEO) database, and then merged to remove the batch effect using the ComBat function of sva package in R4.2.2. Furthermore, GSE39058 was screened from GEO as external validation cohort 1; while 85 samples including survival time and expression matrix were collected from the Target database and served as external validation cohort 2. In addition, the GSE126209 dataset containing 12 osteosarcoma samples and 11 normal controls was used to screen for differentially expressed genes (DEGs). Using the limma package in R4.2.2 software, DEGs between osteosarcoma and controls were filtered with a threshold of P 1. A total of 3783 LLPS-related genes were obtained from DrLLPS ( http://llps.biocuckoo.cn/ ), LLPSDB ( http://bio-comp.org .cn/llpsdb/home.html) and PhaSepDB ( http://db.phasep.pro/ ) databases. The intersection of DEGa and LLPS-related gene was visualized by ggplot2 and VennDiagram package in R4.2.2. 2.1.2 Screening of prognostic biomarkers for constructing a predictive model Based on the data in the combined dataset, Cox regression analysis was performed using the coxph function of survival package in R4.2.2 to screen for prognosis-related genes. Further application of least absolute shrinkage and selection operator (LASSO) regression using glmnet package provided the optimal combination of prognosis-related genes with parameter of lambda.min. To avoid overfitting of the model, 50% of samples from the combined dataset were used as the internal training set and the remaining ones were used as the internal validation set. External validation cohorts 1 and 2 were implemented to verify the model stability. The timeROC and survminer packages were used to plot Kaplan-Meier (KM) and receiver operating characteristic (ROC) curves, respectively, to quantify the model prediction results. 2.1.3 Molecular subtype of osteosarcoma ConsensusClusterPlus was used to carry out the concensus clustering analysis, and the optimal number of clusters was selected based on the cumulative distribution function (CDF) curves and principal component analysis (PCA) results. The clustering criteria were as follows: the number of samples in each group was relatively consistent; the CDF curves gradually increased; the samples in each group were aggregated, with obvious differences between groups. Differences between survival status and immune checkpoint expression among subtypes were compared using KM curves and t test. 2.1.4 Multidimensional comparison between prognostic risk groups To predict the pathway scores for each sample in the combined dataset, the GSVA package was applied with c2.cp.kegg.v2023.1.Hs.symbols.gmt in MSigDB database as the background pathway. Subsequently, the limma package was employed to screen pathways with significant differences in scores between high- and low-risk groups, with P < 0.05 as the threshold. The expression matrix of immune cell subtypes was also deconvoluted using CIBERSORT as a means of estimating the infiltration abundance of immune cell for each sample in the combined dataset. To discern the likelihood of tumor immune evasion, TIDE score and microsatellite instability (MSI) score were calculated on TIDE online tool based on the conventional immune checkpoint expression of individual samples in this study. Based on a drug prediction model developed by the GDSC database, this study predicted the sensitivity of samples to different drugs using oncoPredict package. Differences in immune cell infiltration levels, TIDE scores, and drug sensitivities between high- and low-risk groups were assessed using t test, followed by Pearson correlation analyses for prognostic signatures. 2.1.5 Molecular docking To predict effective blinding of drug and prognostic signatures, molecular docking analysis was accomplished in this study. The 3D structures of the drugs were retrieved from PubChem database and exported into PDB format using PyMol software. Following, the charge of the ligand was adjusted and twistable bonds were selected in AutoDockTools-4.2.6 software. For receptors, their gene IDs were retrieved from the Uniprot database. Protein 3D structures were collected from the PDB and AlphaFoldDB databases and imported into PyMol as a PDB file with water molecule sequences removed. Subsequently, a series of processes were performed on the receptor in AutoDockTools-4.2.6, including removal of the original ligand, addition of hydrogens, optimization of amino acids, and calculation of charge. The active region of binding pocket for docking was determined based on the original ligand position in the protein receptor, while the molecular docking was carried out according to receptor name, ligand name, coordinates of the docking centroid, and distances incoming from AutoDock vina. Valid docking results were output if binding energy was less than − 5.0 kcal/mol and hydrogen bonds between receptor-ligand complex could be formed. 2.2 cell culture and treatment The human fetal osteoblast cell line hFOB (CL-233h, SAIOS biotechnology) and the osteosarcoma cell line U2-OS (CL-655h, SAIOS biotechnology) were cultured in DMEM containing 10% fetal bovine serum (FBS, 16140071, Gibco) and 1% penicillin–streptomycin (C11885500BT, Gibco). Another two osteosarcoma cell lines HOS (CL-156h, SAIOS biotechnology) and MG-63 (CL-157h, SAIOS biotechnology) were cultured in MEM (11095080, Gibco) using the same strategy to validate the expression patterns of five prognostic signatures. To observe the effect of WDR3 silencing on osteosarcoma in vitro , sh-WDR3 was transfected into U2-OS cells by lentivirus using Lipofectamine™ 2000 Transfection Reagent (11668030, Invitrogen) in this study, with both control and sh-negative control (NC) groups being set up. Sequences of sh-WDR3 (SS Sequence: GGTTCTCTCTAATCTATAA; AS Sequence: TTATAGATTAGAGAGAACC) were designed in the Designer of Small Interfering RNA website. To explore the pharmacological mechanism of Nilotinib (SC0209, Beyotime), U2-OS cells were treated with different concentrations of Nilotinib (1µM, 2.5µM, 5µM) for 24 h. The optimal concentration (5µM) of Nilotinib was then selected to treat the cells for 24h according to cell viability to explore changes in cell function and WDR3 phase separation level. 2.3 Quantitative real-time polymerase chain reaction (qPCR) The 1 mL of Trizol (15596018, Invitrogen) was added to cells or tissues to release total RNA. After a reverse transcription reaction, cDNA was synthesized and configured with primers for the PCR reaction system. Sequences of primers are detailed in Supplementary Table 1. On a PCR instrument (CFX Connect, BIO-RAD), the PCR reaction was carried out and underwent 40 cycles (95 ℃, 3 min; 95 ℃, 12 s; 62 ℃, 40 s) of amplification. Relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), the mRNA level of target genes was calculated using a 2 −ΔΔCT method. 2.4 Western blotting Cells and tissues were lysed to release total proteins for quantification. Then, samples were loaded and run in sodium dodecyl sulfate-polyacrylamide gel electrophoresis to separate proteins, which were then transferred to polyvinylidene fluoride membranes (FFP24; Beyotime). Membranes were then incubated overnight in diluted primary antibody working solution (anti-WDR3, 1:1000, PA5-144030, Thermo; anti-GAPDH, 1:2500, ab181602, Abcam). After washing thrice, the membranes were placed in a 2000-fold diluted secondary antibody (Goat Anti-Rabbit IgG H&L, ab6721, Abcam) for another incubation of 1 h. Finally, the membranes were developed in enhanced chemiluminescence (ECL) solution (P1000, APPLYGEN), followed by the scanning of exposed films. ImageJ software was used to quantify the gray values of bands, and the protein expression of WDR3 relative to GAPDH was calculated. 2.5 Cell counting kit-8 (CCK-8) Transfected or Nilotinib-treated U2-OS cells were inoculated in 96-well plates (2000 cells per well). After 24 h of routine incubation (37 ℃, 5% CO 2 ), each well was supplemented with 10 µL of CCK-8 reaction solution (C0037, Beyotime). Two hours later, the optical density of each pore at 450 nm was measured by microplate reader (DR-3518G, Wuxi Hiwell Diatek) to evaluate cell viability. 2.6 Transwell assay In this study, Transwell assay was conducted to assess the migration and invasive ability of cells. To observe cell migration, we inoculated digested U2-OS cells into Transwell chambers, which was then cultured in the lower chamber containing medium for 24 h culture under conventional conditions. Later on, Transwell chamber was fetched out to remove medium. After undergoing washing and fixation, crystal violet (C0121, Beyotime) staining was added to the chambers for 20 min to observe unmigrated cells. For invasion detection, we covered the bottom of the Transwell chambers with 50 mg/L Matrigel (354234, Corning) diluted at 1:4 until it polymerized into a gel. The same methods for cell culture, fixation, and staining were repeated to observe the cell invasion in three randomly selected fields under the microscope. 2.7 Animals Thirty male Balb/c nude mice (5–6 weeks old) were purchased from the Experimental Animal Center of Yangzhou University, where they were housed under a 12-hour light/dark cycle and had free access to adequate food and water. Of which, 18 mice were randomly divided into three groups (OS, OS + sh-NC, and OS + sh-WDR3; six mice per group) and injected subcutaneously in the right axilla with 1×10 7 U2-OS, sh-NC-transfected U2-OS, and sh-WDR3-transfected U2-OS cells, respectively, for xenograft tumor formation. Tumor volumes were recorded every seven days over a four-week period. On day 28th, mice were anesthetized with 4% isoflurane (HY-A0134, MCE) and then euthanized for tumor collection. Another 12 mice were randomized into OS and OS + Nilotinib (OS + NIL) groups (n = 6) and injected subcutaneously with 1×10 7 U2-OS cells to constructed the xenograft model of osteosarcoma. One week later, Nilotinib was diluted in dimethyl sulfoxide (D8371, Solarbio) and administered for the gavage treatment of mice in the OS + NIL group (30 mg/kg) via an oral delivery carrier containing 0.5% hydroxypropyl methylcellulose (HPMC, HY-A0104J, MCE) and 0.05% Tween80 (TB360, Solarbio). The OS group was gavaged with an equal dose of 0.5% hydroxypropyl methylcellulose and 0.05% Tween80. After gavage therapy once a day for three weeks, the mice were euthanized to collect tumor samples. All animals were carried out under the approval of the Experimental Animal Ethics Committee of Yangzhou University (No.202312013). 2.8 Immunohistochemical (IHC) Tumor tissues from each group of mice were prepared into sections and incubated overnight with a 1000-fold dilution of WDR3 (PA5-144030, Thermo) and Ki67 (9129S, Cell Signaling) antibodies. The washed sections were incubated with a secondary antibody (Goat Anti-Rabbit IgG H&L, ab6721, Abcam) for 15 min at a dilution of 1:2000. To visualize the target proteins, the sections were first stained in DAB staining solution (P0202, Beyotime) for 30 min, followed by re-staining with hematoxylin (G1080, Solarbio) for 3 min. After cleaning, the slices were dried, dehydrated, sealed, and photographed under a microscope in turn. 2.9 Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labelling (TUNEL) The prepared tumor tissue sections were routinely dewaxed and hydrated. Following the instructions of TUNEL kit (C1091, Beyotime), 50 µL of the configured proteinase K working solution was added dropwise to the sections for digestion of 30 min, followed by the incubation with a mixture of 5µL TdT enzyme, 45µL fluorescent labeling solution, and 50µL TUNEL test solution for 30 min. After washing, the sections were stained with 4',6-diamidino-2-phenylindole (DAPI, C1005, Beyotime) for 10 min. Ultimately, sections were sealed with antifade mounting medium (p0126, Beyotime) and observed under a microscope. 2.10 Recombinant protein expression and purification The WDR3 gene was cloned into the pET-28b (+) expression vector (with GFP and His tags) and transformed into E. coli strain BL21 (DE3). The transformed strains were cultured in LB liquid medium containing 100 µg/mL Ampicillin (ST008, Beyotime) for 37°C at 170 rpm until the OD600 of bacterial fluids reached 0.6–1.0. The addition of isopropyl-beta-D-thiogalactoside (ST098, Beyotime) with a final concentration of 1mM induced the protein expression at 37℃ for 4 h. Afterwards, the bacterial fluids were centrifuged at 15,000g for 1 min at 4°C to collect precipitate. The harvested bacterial precipitate was resuspended and lysed by sonication on ice. Then, the recombinant proteins were purified with BeyoGold™ His-tag Purification Resin (P2233, Beyotime). Finally, the proteins were eluted on an elution column and concentrated by ultrafiltration. To construct protein mutants, the intrinsically disordered region (IDR) of WDR3 was predicted on the IUPred2A website. Predictions guided the construction of three WDR3 IDR mutants (WDR3-MUT1, WDR3-MUT2, and WDR3-MUT3) and the fusion to the IDR of hnRNPA1, known to drive condensate formation, to synthesize a rescue phase-separated WDR3 mutant (MUT-IDR). WDR3-MUT and MUT-IDR were then cloned into pET-28b (+) expression vectors (GFP and His tags) to induce protein expression, respectively. Protein expression and purification procedures were described above. 2.11 Droplet formation assay In this study, 20 µM WDR3-GFP purified protein was added to a buffer containing 10% PEG8000 (81268, Sigma) and NaCl (ST1641, Beyotime) with different concentrations (75 mM, 150 mM, 300 mM). The protein solution was then loaded onto a slide and condensates with liquid-like behavior were observed by fluorescence microscopy. The same method was also applied to observe droplet formation of purified WDR3-GFP protein in 125 mM NaCl and 10%PEG buffer at different concentrations (10 µM, 20 µM, 40 µM). To investigate intracellular droplet formation, lentiviral expression vectors for WDR3-MUT and MUT-IDR were constructed to transfect U2-OS cells. Cells were climbed and cultured overnight until wall attachment. The droplet formation in the cells was observed under a confocal laser scanning microscope (CLSM, TCS SP8, Leica). 2.12 Fluorescence recovery after photobleaching (FRAP) To further verify the recovery of fluorescence activity of WDR3 after photobleaching, 20 µM of WDR3-GFP purified proteins were added in a buffer containing 10% PEG8000 and 125 mM NaCl to form droplets and then subjected to FRAP assay by CLSM. The FRAP was programmed with a 488 nm 100% power laser, a 1.5µm radius area as the target focus, and a bleaching time of 20 s. The recovery of protein condensates after 40 s and 60 s of photobleaching was photographed. The same procedures were applied to detect phase separation of exogenous WDR3 in WDR3-GFP-, WDR3-MUT-, or MUT-IDR-transfected U2-OS cells. 2.13 Immunofluorescence In this study, immunofluorescence was used to examined the formation of endogenous and exogenous WDR3 phase-separated condensates in cells. To facilitate observation, cells were sequentially clamped, immobilized, sealed, and incubated in 200 µL of anti-WDR3 (9129S, Cell Signaling) diluted at 1:200 overnight. The next day, cells were incubated in a 1:500 diluted mixture of IgG H&L and DAPI for 30–60 min. After sealing slices, the fluorescence of each group was detected under a CLSM. 2.14 Statistical analysis All data presented as mean ± standard deviation was processed on GraphPad 10.1.2. Comparisons between two groups were conducted using unpaired t test, while comparisons between multiple groups were performed using one-way analysis of variance (ANOVA) with Tukey's post hoc test. Differences between groups under continuous time were compared using two-way RM ANOVA. A P -value below 0.05 was defined as statistical significance. 3. Results 3.1 Predictive efficiency of LLPS-related signatures in osteosarcoma prognosis Based on the expression profiles of the GSE126209 dataset, this study screened out 2489 DEGs between osteosarcoma and normal samples, of which 1497 DEGs were up-regulated and 992 DEGs were down-regulated in osteosarcoma (Fig. 1 A). Within DEGs, 473 of them belonged to LLPS-related genes (Fig. 1 B). To screen prognostically relevant signatures, the GSE16091, GSE21257, and GSE39055 datasets were merged. After removing the batch effect, the samples in the dataset are evenly distributed with no significantly discrete clusters (Fig. 1 C). In the combined dataset, 15 genes significantly associated with survival time in osteosarcoma (log-rank test, P < 0.05) were screened out from the univariate Cox regression algorithm (Fig. 1 D). Among these 15 genes, only ANXA10 and SMURF2 were prognostic protective factors for osteosarcoma. LASSO was then carried out to screen for models with excellent performance but minimal number of variables (Fig. 1 E). With the optimal λ value, a gene list including ANXA10, MYC, TIMM8A, WASF3, and WDR3 were identified as prognostic signature to establish a predictive system according to the equation as Risk Score = -0.651*ANXA10 + 0.974*MYC + 0.051*TIMM8A + 0.234*WASF3 + 0.347*WDR3. The samples were categorized into high- and low-risk groups for comparison based on the median risk score. In both the internal training and validation sets, patients in the high-risk group were significantly less likely to survive than those in the low-risk group, while the area under curve (AUC) values of 1-, 3-, and 5-year ROCs were all above 0.7 (Fig. 1 F-G). To further assess the predictive efficacy of the model, we conducted validation in two independent external cohorts. The results confirmed that patients in the high-risk group had a significantly lower probability of survival than those in the low-risk group, with the predictive model being highly sensitive and specific (Fig. 1 H-I). 3.2 Characteristic evaluation of prognostic signatures Thereafter, this study performed a multidimensional comparison of differences between high- and low-risk groups. First, molecular subtypes of osteosarcoma were identified using consensus clustering analysis. The results showed that k = 2 appeared to be the best choice for grouping the samples into cluster 1 and 2 (Fig. 2 A). Patients in cluster 1 demonstrated a more favorable prognosis compared to those in cluster 2 (Fig. 2 B), and were more commonly distributed in the low-risk group, which was prone to the status of alive (Fig. 2 C). The Limma package also identified 37 pathways with differences in GSVA scores between high and low risk groups (Fig. 2 D). The majority of these pathways was significantly positively correlated with ANXA10, but markedly negatively correlated with MYC, TIMM8A, and WDR3 (Supplementary Fig. 1A). There were also significant differences in infiltration levels of CD8 T cells and activated mast cells between risk groups (Fig. 2 E). Among them, the infiltration level of CD8 T cells was significantly negatively correlated with MYC expression, while activated mast cells were positively correlated with MYC and WDR3 (Supplementary Fig. 1B). Furthermore, patients in the low-risk group exhibited significantly lower TIDE and MSI scores than those in the high-risk group, suggesting a lower likelihood of tumor immune escape to benefit from immunotherapy (Fig. 2 F). In terms of drug sensitivity, a total of 24 drugs were identified with significant differences in IC50 values between the two groups (Fig. 2 G). These drugs were significantly correlated with the expression of MYC, but had no significant association with WASF3 (Fig. 2 H). Moreover, the expression of WDR3 may affect the sensitivity of osteosarcoma patients to drugs, such as AZD4547 and Nilotinib (Fig. 2 H). To probe into the possible binding of proteins encoded by five prognostic signatures with 24 drugs, molecular docking was conducted and 100 valid docking results were obtained. Among them, the interaction of WDR3 and Nilotinib depicted the lowest binding energy (-9.3kcal/mol), with hydrogen bond on residues such as GLN-154 and LEU-285 contributing to the stability of receptor-ligand complex (Fig. 2 I). Nilotinib was also bound to WDR3 residues by Alkyl, Halogen, Pi-Alkyl, and Pi-Cation (Fig. 2 I). 3.3 Effect of prognostic biomarker WDR3 on the progression of osteosarcoma in vitro and in vivo Compared with normal osteoblastic hFOB, this study validated the expression patterns of five prognostic signatures in three osteosarcoma cell lines (U2-OS, HOS, MG-63). qPCR results suggested that the candidate phase separation-related genes were all up-regulated in osteosarcoma cells (Fig. 3A), which was consistent with the results of Bioinformatic analyses. Considering the binding stability of WDR3 to Nilotinib, this study focused on exploring the therapeutic mechanism of WDR3 phase separation in Nilotinib against osteosarcoma. We used lentiviral transfection to specifically down-regulate WDR3 expression in U2-OS cells in order to observe its effect on osteosarcoma progression in vitro and in vivo . As expected, sh-WDR3 significantly reduced WDR3 mRNA and protein expression levels in U2-OS cells compared to those transfected with sh-NC (Fig. 3B–C). In the meantime, U2-OS cells transfected with sh-WDR3 demonstrated significant inhibition of cell viability (Fig. 3D), as well as migration and invasion (Fig. 3E), compared to their negative controls. In vivo , downregulation of WDR3 also significantly reduced tumor volume and weight in xenografted mice of osteosarcoma (Fig. 3F). In addition, sh-WDR3 transfection significantly reduced the positive area of WDR3 and Ki67 in mice tumor tissues (Fig. 3G), while promoting apoptosis (Fig. 3H), relative to controls. Thus, the silencing of WDR3 can effectively alleviate the malignant progression of osteosarcoma. 3.4 WDR3 exhibited phase-separated condensates with liquid-like behavior in OS cells The droplet formation experiments showed that WDR3 can form circular condensates in the presence of low concentration of salt solution, while the number of droplets significantly reduced with the increase of NaCl concentration (Fig. 4 A). The number of droplets also increased with the raised concentration of WDR3-GFP recombinant protein (Fig. 4 B). After photobleaching, the fluorescence signals in the target field were weakened, but recovered rapidly within 60 s, suggesting the liquid-like behavior inside the droplets (Fig. 4 C). In U2-OS cells, the presence of endogenous WDR3 were observed by immunofluorescence (Fig. 4 D). Furthermore, U2-OS cells exogenously transfected with WDR3-GFP demonstrated stronger fluorescence and larger condensates of WDR3 (Fig. 4 E). After photobleaching of this region, the fluorescence signal of WDR3 condensate can also be gradually recovered (Fig. 4 F), further confirming that WDR3 can form droplets through phase separation. To further differentiate the phase-separation ability of WDR3, we mutated its IDR, where amino acid mutations have been shown to alter the phase-separation properties of proteins thereby impairing their recombination and reprogramming [ 27 ]. Using the IUPred2A online tool, this study predicted that IDR of WDR3 was mainly concentrated in the region of amino acids 231–257, 319–348, and 712–743, which may contribute to the formation of dimers or multimers (Fig. 4 G). Furthermore, all glutamate and aspartate residues within these three regions were substituted with alanine (Supplementary Fig. 2), guiding the construction of three WDR3 IDR mutants, named WDR3-MUT1, WDR3-MUT2, and WDR3-MUT3 (Fig. 4 H). Among these, only WDR-MUT1 significantly reduced the number of intracellular WDR3 droplets (Fig. 4 I) and slowed WDR3 condensates reformation after photobleaching (Fig. 4 J), prompting its selection for subsequent investigations. 3.5 Phase separation of WDR3 expedited OS metastasis in vitro The IDR of hnRNPA1 is conducive to protein assembly and droplet properties [ 28 ], and thus can be fused to WDR3-MUT to rescue the phase separation of WDR3 (Fig. 5 A). Under the microscope, WDR3-MUT was found to disrupt the droplet-forming ability of WDR3, resulting in dispersed fluorescence and inability to aggregate. This dysfunction was then rescued by the fusion of hnRNPA1 IDR, allowing WDR3 to reaggregate and form droplets (Fig. 5 B). After photobleaching the target region of MUT-IDR, the fluorescence intensity of WDR3 was gradually restored and the droplet aggregation ability was rescued (Fig. 5 C). After the transfection of WDR3-WT, WDR3-MUT, and MUT-IDR into U2-OS cells, changes in WDR3 phase-separation characteristics were also investigated in this study. As expected, the fluorescence in the WDR3-MUT group began to disperse, accompanied by a decrease in the number of droplets. However, the droplets in the MUT-IDR group were reconcentrated and the number of droplets recovered significantly (Fig. 5 D). FRAP results indicated that the WDR3-MUT-transfected cells remained unchanged fluorescence in the photobleached regions, which could be recovered by the transfection of MUT-IDR (Fig. 5 E). More importantly, mutation of WDR3 significantly down-regulated the proliferation and metastatic ability of U2-OS cells, but the transfection of MUT-IDR restored the malignant phenotype of tumor cells (Fig. 5 F–G). Thus, the phase separation of WDR3 could promote the proliferation, migration and invasion of osteosarcoma in vitro . 3.6 Nilotinib mitigated osteosarcoma progression via inhibition of WDR3 phase separation To clarify the therapeutic mechanism of Nilotinib on osteosarcoma, we first treated U2-OS cells with different concentrations of Nilotinib in vitro for 24 h. Cells treated by Nilotinib were significantly reduced in proliferation in a dose-dependent manner (Fig. 6A). Nilotinib treatment also significantly reduced the mRNA and protein expression levels of WDR3 in U2-OS cells (Fig. 6B–C). Notably, the fluorescence intensity of WDR3 and punctas per cell were significantly increased in U2-OS cells compared to hFOB cells, whereas Nilotinib treatment significantly suppressed these abnormalities (Fig. 6D). The FRAP results suggested that the fluorescence intensity of WDR3 condensates in untreated U2-OS cells after photobleaching could recover within 60 s, which was not observed in Nilotinib-treated cells (Fig. 6E)). Moreover, the addition of Nilotinib significantly reduced the migration and invasion abilities of U2-OS cells (Fig. 6F). The animal experiments were also carried out to confirm the role of WDR3 phase separation in the treatment of osteosarcoma by Nilotinib. The schedule for animal experiments is shown in Fig. 7 A. Consistent with the above results, treatment of Nilotinib significantly reduced tumor volume and weight in xenografted mice of osteosarcoma (Fig. 7 B). Meanwhile, continuous Nilotinib treatment also significantly down-regulated the mRNA and protein expression levels of WDR3, compared to untreated mice (Fig. 7 C–D). Immunofluorescence showed that with the treatment of Nilotinib, the number of WDR3 condensates decreased significantly (Fig. 7 E), suggesting a downregulation of its phase separation level. Therefore, Nilotinib may alleviate the progression of osteosarcoma in vivo and in vitro by inhibiting the phase separation of WDR3. 4. Discussion The bottlenecks in prolonging survival of patients with metastatic osteosarcoma have prompted the search for a number of prognostic biomarkers. Starting with the LLPS mechanism, this study screened out five LLPS-related genes (ANXA10, MYC, TIMM8A, WASF3, and WDR3) with prognostic values by bioinformatics methods, and the risk models constructed accordingly could predict the survival likelihood of osteosarcoma. These prognostic signatures were significantly associated with immune cell infiltration, tumor immune escape, and drug sensitivity. Among them, WDR3 and Nilotinib demonstrated optimal binding stability in molecular docking models. WDR3 is not only a prognostic risk factor for osteosarcoma, but also highly expressed in U2-OS cells. Functional experiments have shown that the knockdown of WDR3 inhibited the proliferation and metastatic ability of osteosarcoma cells while suppressing tumor growth. More importantly, WDR3 can form condensates with liquid-like behavior in U2-OS cells, and mutation of its IDR can eliminate the phase-separated level of WDR, thus reversing the aggressive phenotype of osteosarcoma cells. WDR3, also known as DIP2 and UTP12, belongs to a family of eukaryotic genes that carry the WD repeat region [ 29 ]. It encodes a 943 amino acid nuclear protein composed of a 10 WD repeat sequence module localized to human chromosome 1p12-p13 [ 30 ]. Su et al. found that overexpression of WDR3 is associated with low survival in cancer and promotes pancreatic cancer proliferation and invasion by interacting with GATA4 to induce Hippo pathway activation [ 31 ]. In thyroid cancer, WDR3 maintains genomic stability in patients, while its location on the 1p12 chromosome contributes to disease susceptibility [ 32 , 33 ]. WDR3 expression is linked to inflammatory mediators and can be reduced following the treatment with anti-inflammatory drugs [ 34 ]. McMahon et al. proposed that WDR3 deficiency leads to ribosome biogenesis defects by affecting 18s rRNA processing, thereby reducing p53-mediated proliferation of cancer cells [ 35 ]. Although no studies have been conducted on WDR3 in osteosarcoma, it has been found to promote cancer stem cell characteristics by inhibiting USF2-mediated RASSF1A transcription [ 36 ]. In a multicenter case-control study, genetic polymorphisms in RASSF1A were found to be associated with the risk of osteosarcoma and metastasis in young Chinese adults [ 37 ]. RASSF1A plays a tumor suppressor role in osteosarcoma and exerts anticancer effects by regulating the Wnt/β-catenin pathway [ 38 ]. These findings further confirmed our research, suggesting that WDR3 may have the potential to promote osteosarcoma metastasis by inhibiting transcriptional regulation of RASSF1A. Furthermore, our results indicated that WDR3 expression was significantly positively correlated with the spliceosome pathway score, but had negative correlations with the lysosome pathway. SNRPB, a core component of the spliceosome, was shown to induce malignant behavior in osteosarcoma [ 39 ]. C1GALT1 enhances the drug resistance and metastatic propensity of osteosarcoma by promoting lysosomal degradation and effluence [ 40 ]. Therefore, we speculated that WDR3 phase separation-induced osteosarcoma progression is associated with SNRPB upregulation and lysosomal degradation. In terms of therapeutic potential, WDR3 bound stably to Nilotinib in this study and mediated the therapeutic mechanism of Nilotinib against osteosarcoma. As a second-generation tyrosine kinase inhibitor that allows for a faster deep molecular response, Nilotinib is approved for the first-line treatment of BCR-ABL-positive chronic granulocytic leukemia [ 41 ]. In addition to blood disorders, Nilotinib has been used to treat gastrointestinal stromal tumors and benefits patients with KIT exon 11 mutant phenotypes [ 42 ]. By targeting the suppression of DDR1, Nilotinib effectively blocked the migration of breast cancer cells [ 43 ]. Nilotinib also improved outcomes in colorectal cancer patients receiving anti-PDL1 therapy by restoring MHC-I expression [ 44 ]. In osteosarcoma, Nilotinib was found to downregulate the expression of prognostic marker MAPK1, thereby promoting apoptosis [ 45 ]. Notably, Wei et al. suggested that Nilotinib may exert therapeutic benefits by targeting phase separation in cells [ 46 ]. The present study demonstrated that Nilotinib significantly inhibited the production of phase-separated condensates of WDR3, thereby suppressing tumor growth and metastasis in vivo and in vitro (as shown in Fig. 8 ). Furthermore, WDR3 expression was positively correlated with the infiltration level of activated mast cells in osteosarcoma. In rat peritoneal mast cells, Nilotinib decreased the expression of pro-inflammatory cytokines and TNF-α, and dose-dependent decreased histamine release from mast cells [ 47 ]. Mast cell infiltration is common in the inflammatory response to malignant osteosarcoma, which occurs mainly at the tumor margin and may lead to osteolysis and tumor invasion, but facilitates immunomodulatory therapy [ 48 ]. It was also reported that mast cell accumulation in osteosarcoma is regulated by the CXCL6-CXCR2 axis [ 49 ]. Therefore, disruption of WDR3 phase separation is closely associated with mast cell-mediated inflammatory responses in osteosarcoma inhibited by Nilotinib, thus affecting the mechanisms of metastasis and drug resistance in osteosarcoma. The regulation of LLPS in cells is a sophisticated mechanism involving not only crosstalk between intracellular components, but also influences by extracellular environmental factors such as temperature, ionic concentration, and pH [ 50 ], which may further affect the phase-separation characteristics of proteins. In addition, the energetic state of cell including APT levels may also affect the kinetic changes of the protein LLPS [ 51 ]. However, these factors that may affect the phase separation level of WDR3 were not discussed, which is one of the limitations of our study. Furthermore, the clinical use of Nilotinib involves cardiovascular adverse events [ 52 ], but the cardiovascular toxicity of Nilotinib was neglected in this study. Therefore, subsequent studies are recommended to focus on exploring the intracellular mechanisms affecting WDR3 phase separation, as well as the resistance and toxicity of Nilotinib in osteosarcoma treatment. 5. Conclusion This study proposed five LLPS-related biomarkers with the potential to predict the survival of osteosarcoma. Among them, WDR3 was a prognostic risk factor for osteosarcoma and bound stably to Nilotinib in molecular docking models. Functionally, WDR3 was significantly overexpressed in osteosarcoma cells, while its downregulation inhibited the malignant progression of osteosarcoma both in vivo and in vitro . In addition, WDR3 was found to form droplets, and its IDR mutation eliminated phase-separated levels of WDR3, thereby ameliorating the aggressive phenotype of osteosarcoma cells. Nilotinib may also reduce WDR3 phase-separated condensates and inhibit tumor growth and metastasis. Abbreviations ANOVA one-way analysis of variance CCK-8 Cell counting kit-8 CDF cumulative distribution function DAPI 4',6-diamidino-2-phenylindole DEGs differentially expressed genes ECL enhanced chemiluminescence FBS fetal bovine serum FRAP Fluorescence recovery after photobleaching GEO Gene Expression Omnibus IDR intrinsically disordered region IHC Immunohistochemical LASSO least absolute shrinkage and selection operator LLPS liquid-liquid phase separation MSI microsatellite instability PCA principal component analysis qPCR Quantitative real-time polymerase chain reaction ROC receiver operating characteristic TdT Terminal deoxynucleotidyl transferase TUNEL TdT-mediated dUTP nick-end labelling Declarations Ethics approval and consent to participate All animals were carried out under the approval of the Experimental Animal Ethics Committee of Yangzhou University (No.202312013). Consent for publication Not applicable. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This study was supported by Shengjing Hospital of China Medical University Intra-Hospital Program (No. M0140). Authors’ contributions Minglei Li : Conceptualization, Investigation, Data curation, Formal analysis, Writing - original draft. Nan Li : Conceptualization, Methodology, Data curation, Visualization. Yuying Fan: Resources, Formal analysis, Software, Visualization, Writing - original draft. Zhan Zhang: Resources, Software. Long Zhou: Investigation, Methodology. Yifan Yu: Resources, Visualization. Man Ni: Data curation, Formal analysis. Mingzi Tan: Conceptualization, Investigation, Project administration, Writing - review and editing. WanJie Huang: Conceptualization, Investigation, Project administration, Writing - review and editing. Tong Zhu: Conceptualization, Investigation, Project administration, Writing - review and editing. All authors read and approved the final manuscript. Acknowledgements We thank Minglei Li and Yuying Fan for conducting the mentionable literature research and subsequent data analysis. Thanks are due to Tong Zhu for their detailed revision guidance on the completed first draft of the writing. We would also like to thank The Shengjing Hospital of China Medical University for providing a research platform, which laid the foundation for the smooth progress of the project. In addition, we would like to thank Yangzhou University for providing the venue and platform for the animal experiments. Besides, we acknowledge partial support of Shengjing Hospital of China Medical University Intra-Hospital Program (No. M0140). References Sun C, Li S, Ding J. Biomaterials-Boosted Immunotherapy for Osteosarcoma. Advanced healthcare materials. 2024;13(23):e2400864. Shi Q, Xu J, Chen C, et al. Direct contact between tumor cells and platelets initiates a FAK-dependent F3/TGF-β positive feedback loop that promotes tumor progression and EMT in osteosarcoma. Cancer letters. 2024;591(216902. Cascini C, Ratti C, Botti L, et al. Rewiring innate and adaptive immunity with TLR9 agonist to treat osteosarcoma. Journal of experimental & clinical cancer research : CR. 2023;42(1):154. Yin C, Chokkakula S, Li J, et al. Unveiling research trends in the prognosis of osteosarcoma: A bibliometric analysis from 2000 to 2022. Heliyon. 2024;10(6):e27566. Mohr A, Marques Da Costa ME, Fromigue O, et al. From biology to personalized medicine: Recent knowledge in osteosarcoma. European journal of medical genetics. 2024;69(104941. Yu S, Yao X. Advances on immunotherapy for osteosarcoma. Molecular cancer. 2024;23(1):192. Nerlakanti N, McGuire JJ, Bishop RT, et al. Histone deacetylase upregulation of neuropilin-1 in osteosarcoma is essential for pulmonary metastasis. Cancer letters. 2024;606(217302. Di Patria L, Habel N, Olaso R, et al. C-terminal binding protein-2 triggers CYR61-induced metastatic dissemination of osteosarcoma in a non-hypoxic microenvironment. Journal of experimental & clinical cancer research : CR. 2025;44(1):83. Mosca N, Alessio N, Di Paola A, et al. Osteosarcoma in a ceRNET perspective. Journal of biomedical science. 2024;31(1):59. Wei X, Feng J, Chen L, et al. METTL3-mediated m6A modification of LINC00520 confers glycolysis and chemoresistance in osteosarcoma via suppressing ubiquitination of ENO1. Cancer letters. 2024;217194. Wang Y, Ma X, Xu E, et al. Identifying squalene epoxidase as a metabolic vulnerability in high-risk osteosarcoma using an artificial intelligence-derived prognostic index. Clinical and translational medicine. 2024;14(2):e1586. Zheng Z, Zeng Y, Bao X, et al. OTULIN confers cisplatin resistance in osteosarcoma by mediating GPX4 protein homeostasis to evade the mitochondrial apoptotic pathway. Journal of experimental & clinical cancer research : CR. 2024;43(1):330. Ji R, Wang Y, Pan D, et al. NUCB2 inhibition antagonizes osteosarcoma progression and promotes anti-tumor immunity through inactivating NUCKS1/CXCL8 axis. Cancer letters. 2024;591(216893. Gao W, Zhou J, Huang J, et al. Up-regulation of RAN by MYBL2 maintains osteosarcoma cancer stem-like cells population during heterogeneous tumor generation. Cancer letters. 2024;586(216708. Wu Y, Ma B, Liu C, Li D, Sui G. Pathological Involvement of Protein Phase Separation and Aggregation in Neurodegenerative Diseases. International journal of molecular sciences. 2024;25(18): Cheng Z, Wang H, Zhang Y, et al. Deciphering the role of liquid-liquid phase separation in sarcoma: Implications for pathogenesis and treatment. Cancer letters. 2025;616(217585. Liu YT, Cao LY, Sun ZJ. The emerging roles of liquid-liquid phase separation in tumor immunity. International immunopharmacology. 2024;143(Pt 1):113212. Pu X, Zhang C, Jin J, et al. Phase separation of EEF1E1 promotes tumor stemness via PTEN/AKT-mediated DNA repair in hepatocellular carcinoma. Cancer letters. 2025;613(217508. Bhowmik D, Du M, Tian Y, et al. Cooperative DNA binding mediated by KicGAS/ORF52 oligomerization allows inhibition of DNA-induced phase separation and activation of cGAS. Nucleic acids research. 2021;49(16):9389-9403. Cable J, Brangwynne C, Seydoux G, et al. Phase separation in biology and disease-a symposium report. Annals of the New York Academy of Sciences. 2019;1452(1):3-11. Banani SF, Lee HO, Hyman AA, Rosen MK. Biomolecular condensates: organizers of cellular biochemistry. Nature reviews Molecular cell biology. 2017;18(5):285-298. Chakraborty S, Nandi P, Mishra J, et al. Molecular mechanisms in regulation of autophagy and apoptosis in view of epigenetic regulation of genes and involvement of liquid-liquid phase separation. Cancer letters. 2024;587(216779. Lu B, Zou C, Yang M, et al. Pharmacological Inhibition of Core Regulatory Circuitry Liquid-liquid Phase Separation Suppresses Metastasis and Chemoresistance in Osteosarcoma. Advanced science (Weinheim, Baden-Wurttemberg, Germany). 2021;8(20):e2101895. Chen D, Zhao Z, Huang Z, et al. Super enhancer inhibitors suppress MYC driven transcriptional amplification and tumor progression in osteosarcoma. Bone research. 2018;6(11. Kim YR, Joo J, Lee HJ, et al. Prion-like domain mediated phase separation of ARID1A promotes oncogenic potential of Ewing's sarcoma. Nature communications. 2024;15(1):6569. Sun J, Chen Y, Bi R, Yuan Y, Yu H. Bioinformatic approaches of liquid-liquid phase separation in human disease. Chinese medical journal. 2024;137(16):1912-1925. Wang J, Yu H, Ma Q, et al. Phase separation of OCT4 controls TAD reorganization to promote cell fate transitions. Cell stem cell. 2021;28(10):1868-1883.e1811. Molliex A, Temirov J, Lee J, et al. Phase separation by low complexity domains promotes stress granule assembly and drives pathological fibrillization. Cell. 2015;163(1):123-133. Kobayashi M, Jitoku D, Iwayama Y, et al. Association studies of WD repeat domain 3 and chitobiosyldiphosphodolichol beta-mannosyltransferase genes with schizophrenia in a Japanese population. PloS one. 2018;13(1):e0190991. Claudio JO, Liew CC, Ma J, et al. Cloning and expression analysis of a novel WD repeat gene, WDR3, mapping to 1p12-p13. Genomics. 1999;59(1):85-89. Su W, Zhu S, Chen K, et al. Overexpressed WDR3 induces the activation of Hippo pathway by interacting with GATA4 in pancreatic cancer. Journal of experimental & clinical cancer research : CR. 2021;40(1):88. García-Quispes WA, Pastor S, Galofré P, et al. Possible role of the WDR3 gene on genome stability in thyroid cancer patients. PloS one. 2012;7(9):e44288. Baida A, Akdi M, González-Flores E, et al. Strong association of chromosome 1p12 loci with thyroid cancer susceptibility. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2008;17(6):1499-1504. Gong L, Yu L, Gong X, et al. Exploration of anti-inflammatory mechanism of forsythiaside A and forsythiaside B in CuSO(4)-induced inflammation in zebrafish by metabolomic and proteomic analyses. Journal of neuroinflammation. 2020;17(1):173. McMahon M, Ayllón V, Panov KI, O'Connor R. Ribosomal 18 S RNA processing by the IGF-I-responsive WDR3 protein is integrated with p53 function in cancer cell proliferation. The Journal of biological chemistry. 2010;285(24):18309-18318. Liu W, Xie A, Xiong J, et al. WDR3 promotes stem cell-like properties in prostate cancer by inhibiting USF2-mediated transcription of RASSF1A. The journal of gene medicine. 2023;25(7):e3498. Xu H, Zhan W, Chen Z. Ras-Association Domain Family 1 Isoform A (RASSF1A) Gene Polymorphism rs1989839 is Associated with Risk and Metastatic Potential of Osteosarcoma in Young Chinese Individuals: A Multi-Center, Case-Control Study. Medical science monitor : international medical journal of experimental and clinical research. 2016;22(4529-4535. Wang WG, Chen SJ, He JS, Li JS, Zang XF. The tumor suppressive role of RASSF1A in osteosarcoma through the Wnt signaling pathway. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine. 2016;37(7):8869-8877. Shi Y, Wang Z, Zhang J, et al. Small Nuclear Ribonucleoprotein Polypeptides B and B1 Promote Osteosarcoma Progression via Activating the Ataxia-Telangiectasia Mutated Signaling Pathway through Ribonucleotide Reductase Subunit M2. The American journal of pathology. 2024;194(11):2163-2178. Liu CW, Huang JH, Chang HH, et al. C1GALT1 expression predicts poor survival in osteosarcoma and is crucial for ABCC1 transporter-mediated doxorubicin resistance. The Journal of pathology. 2025; Jabbour E, Kantarjian H. Chronic myeloid leukemia: 2025 update on diagnosis, therapy, and monitoring. American journal of hematology. 2024;99(11):2191-2212. Zhao Z, Zhang J, Zhang W, et al. Efficacy evaluation of nilotinib treatment in different genomic subtypes of gastrointestinal stromal tumors: A meta-analysis and systematic review. Current problems in cancer. 2021;45(3):100705. Wang S, Xie Y, Bao A, et al. Nilotinib, a Discoidin domain receptor 1 (DDR1) inhibitor, induces apoptosis and inhibits migration in breast cancer. Neoplasma. 2021;68(5):975-982. Dong H, Wen C, He L, et al. Nilotinib boosts the efficacy of anti-PDL1 therapy in colorectal cancer by restoring the expression of MHC-I. Journal of translational medicine. 2024;22(1):769. Wu Z, Yu J, Han T, et al. System analysis based on Anoikis-related genes identifies MAPK1 as a novel therapy target for osteosarcoma with neoadjuvant chemotherapy. BMC musculoskeletal disorders. 2024;25(1):437. Wei C, Li M, Li X, Lyu J, Zhu X. Phase Separation: "The Master Key" to Deciphering the Physiological and Pathological Functions of Cells. Advanced biology. 2022;6(7):e2200006. El-Agamy DS. Anti-allergic effects of nilotinib on mast cell-mediated anaphylaxis like reactions. European journal of pharmacology. 2012;680(1-3):115-121. Inagaki Y, Hookway E, Williams KA, et al. Dendritic and mast cell involvement in the inflammatory response to primary malignant bone tumours. Clinical sarcoma research. 2016;6(13. Wang C, Lei Z, Zhang C, Hu X. CXCL6-CXCR2 axis-mediated PD-L2(+) mast cell accumulation shapes the immunosuppressive microenvironment in osteosarcoma. Heliyon. 2024;10(14):e34290. Xu WX, Qu Q, Zhuang HH, et al. The Burgeoning Significance of Liquid-Liquid Phase Separation in the Pathogenesis and Therapeutics of Cancers. International journal of biological sciences. 2024;20(5):1652-1668. Kang J, Lim L, Song J. ATP enhances at low concentrations but dissolves at high concentrations liquid-liquid phase separation (LLPS) of ALS/FTD-causing FUS. Biochemical and biophysical research communications. 2018;504(2):545-551. Wang Z, Jiang L, Yan H, Xu Z, Luo P. Adverse events associated with nilotinib in chronic myeloid leukemia: mechanisms and management strategies. Expert review of clinical pharmacology. 2021;14(4):445-456. Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterials.docx SupplementaryTable1.docx westernblot.docx Cite Share Download PDF Status: Published Journal Publication published 11 Jul, 2025 Read the published version in Journal of Experimental & Clinical Cancer Research → Version 1 posted Editorial decision: Revision requested 29 May, 2025 Reviews received at journal 28 May, 2025 Reviews received at journal 28 May, 2025 Reviewers agreed at journal 20 May, 2025 Reviewers agreed at journal 18 May, 2025 Reviewers invited by journal 08 May, 2025 Editor assigned by journal 08 May, 2025 Submission checks completed at journal 08 May, 2025 First submitted to journal 07 May, 2025 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-6611672","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":454077014,"identity":"f3f9e5cb-2408-4cf6-960e-05bb7aeda566","order_by":0,"name":"Minglei Li","email":"","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Minglei","middleName":"","lastName":"Li","suffix":""},{"id":454077015,"identity":"b52f3d49-8152-4edd-87cd-cf2609f7fcba","order_by":1,"name":"Nan Li","email":"","orcid":"","institution":"The Fourth Afflicted Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Nan","middleName":"","lastName":"Li","suffix":""},{"id":454077016,"identity":"bbeb9d19-4cb8-4398-b07e-7e25fb0f3908","order_by":2,"name":"Yuying Fan","email":"","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yuying","middleName":"","lastName":"Fan","suffix":""},{"id":454077017,"identity":"7741b0e3-9c53-470f-a42b-ed5c67bb8750","order_by":3,"name":"Zhan Zhang","email":"","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhan","middleName":"","lastName":"Zhang","suffix":""},{"id":454077018,"identity":"9a1b8b50-fa23-445f-b444-c272a5f4965d","order_by":4,"name":"Long Zhou","email":"","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Long","middleName":"","lastName":"Zhou","suffix":""},{"id":454077019,"identity":"486eedb8-cfd0-4ff1-aa7c-6ba15f20cb18","order_by":5,"name":"Yifan Yu","email":"","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yifan","middleName":"","lastName":"Yu","suffix":""},{"id":454077020,"identity":"396df15c-8ff3-449e-a1b5-ffb72d52708b","order_by":6,"name":"Man Ni","email":"","orcid":"","institution":"Yangzhou University","correspondingAuthor":false,"prefix":"","firstName":"Man","middleName":"","lastName":"Ni","suffix":""},{"id":454077021,"identity":"9d7089de-641a-4fd9-b889-e97d1e665056","order_by":7,"name":"Mingzi Tan","email":"","orcid":"","institution":"Liaoning Cancer Hospital \u0026 Istitute","correspondingAuthor":false,"prefix":"","firstName":"Mingzi","middleName":"","lastName":"Tan","suffix":""},{"id":454077022,"identity":"d1ed932a-11a3-48b9-9bed-3359b50b7e7c","order_by":8,"name":"WanJie Huang","email":"","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"WanJie","middleName":"","lastName":"Huang","suffix":""},{"id":454077023,"identity":"afca574c-55b6-4570-8a57-93e5c0d63263","order_by":9,"name":"Tong Zhu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIiWNgGAWjYDACCQjFw8/efOCAhEGNHBt7+wFCWhgbgJSMZM+xxAcWBceM+XjOJBClxcbgRo6yQcUH5sR5Eg4GeHXIz+4xf/BxRy2PwZkzbBI3DNjS2yQYEhh+VGzDqYVxzhnDxplnjvNIHu89JjnDQCa3TbrxAGPPmds4tTBL5Bg287Yd4+E7cy5NWsKALbdN5kACM2Mbbi1sMC0MN3LMpP8YMKezSSQY4NXCA9FSwyNwI8fYQMKAOYGgFgmJtMKZM9sO8IADWcLgmGEbMJAP4vOL/IzkDR8+ttXZQ6LyT428fHv7wQc/KnBrgYLDqNwDhNQDQR0RakbBKBgFo2DEAgD0jVq2fVj0xAAAAABJRU5ErkJggg==","orcid":"","institution":"Shengjing Hospital of China Medical University","correspondingAuthor":true,"prefix":"","firstName":"Tong","middleName":"","lastName":"Zhu","suffix":""}],"badges":[],"createdAt":"2025-05-07 11:38:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6611672/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6611672/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13046-025-03456-x","type":"published","date":"2025-07-11T15:57:03+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":82792495,"identity":"4f501a45-e2ee-4006-85c4-ad86dbc86d97","added_by":"auto","created_at":"2025-05-15 10:23:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":8832138,"visible":true,"origin":"","legend":"\u003cp\u003eScreening of prognostic signatures to predict osteosarcoma survival based on a risk score-based model. (A) The volcano plot (left panel) and heatmap (right panel) showing 1497 DEGs between osteosarcoma and normal controls. (B) The Venn diagram depicts 473 intersected genes between DEGs and LLPS-related genes. (C) PCA plots before (top panel) and after (bottom panel) batch effect removal. (D) The univariate Cox regression analyses screened out 15 genes related to osteosarcoma prognosis. (E) Coefficients of the LASSO analysis (left panel) and partial likelihood deviance analysis (right panel) on 15 genes. (F–I) The KM (left panel) and ROC (right panel) curves of combined internal training set (F), combined internal validation set (G), external validation cohort 1 (H), and external validation cohort 2 (I).\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/5d8ffd5de60516ee323ec471.png"},{"id":82792500,"identity":"10817353-ae1c-4ba0-91b5-165121e899d6","added_by":"auto","created_at":"2025-05-15 10:23:13","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":27386150,"visible":true,"origin":"","legend":"\u003cp\u003eCharacteristic evaluation of prognostic signatures. (A) Consensus matrix heatmap defining two clusters (k = 2) and their correlation areas. (B) KM curves showing the survival difference between high- and low-risk groups. (C) Sankey diagram exhibiting distribution of two clusters in risk groups. (D–G) Differences in GSVA pathway scores (D), immune cell infiltrations (E), TIDE scores (F), and drug IC50 values (G) between high and low risk groups. *\u003cem\u003eP\u003c/em\u003e\u0026lt; 0.05; ****\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001. (H) Correlations between prognostic signatures and drug IC50 values. (I) Molecular docking showing the interaction of WDR3 and Nilotinib\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/21b7576e13ded6938b884f40.png"},{"id":82794776,"identity":"cedabc27-6095-46aa-b335-999706728cad","added_by":"auto","created_at":"2025-05-15 10:31:15","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":124479404,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of WDR3 downregulation onthe progression of osteosarcoma \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e. (A) Expression validation of five prognostic signatures using qPCR. (B–C) The mRNA (B) and protein (C) expression of WDR3 in U2-OS cells transfected with sh-NC or sh-WDR3. (D–E) Effects of WDR3 downregulation on cell proliferation (D), as well as migration and invasion (E). (F) Tumor value and weight of mice injected with sh-NC- or sh-WDR3-transfected U2-OS cells. (G–H) Effects of WDR3 downregulation on Ki67 expression (G) and tumor cell apoptosis (H). *\u003cem\u003eP\u003c/em\u003e\u0026lt; 0.05; **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/16d02f73bfd7f93ce50fc0f0.png"},{"id":82792509,"identity":"879f789d-acfe-49f8-810e-06af5e1dd164","added_by":"auto","created_at":"2025-05-15 10:23:13","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":43357450,"visible":true,"origin":"","legend":"\u003cp\u003eWDR3 exhibited phase-separated condensates with liquid-like behavior in OS cells. (A–B) The droplet formation of WDR3 under different concentrations of NaCl (A) and WDR3-GFP recombinant protein (B). *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01. (C) FRAP of WDR3-GFP droplets. (D) The droplet formation of endogenous WDR3 in U2-OS cells. (E) Phase-separated condensates of exogenous WDR3-GFP protein in U2-OS cells. (F) FRAP of WDR3-GFP protein exogenously transfected in U2-OS cells. (G) IDR of WDR3 was predicted in the IUPred2A online tool. (H) Domain structure of WDR3-WT and three WDR3 IDR mutants. (I) Effect of three WDR3 IDR mutants on the number of intracellular WDR3 droplets. **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01 versus WDR3-WT. (J) Effect of three WDR3 IDR mutants on the recovery of WDR3 condensate after photobleaching.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/4bdea2d1d84a4104bbf8e3a7.png"},{"id":82792551,"identity":"d6a8b73f-1de2-4dbd-ab8e-bf3cb37d05b6","added_by":"auto","created_at":"2025-05-15 10:23:15","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":65109045,"visible":true,"origin":"","legend":"\u003cp\u003ePhase separation of WDR3 expedited OS metastasis\u003cem\u003e in vitro\u003c/em\u003e. (A) Schematic illustration of WDR3 mutation and rescue of phase separation. (B) The droplet formation of WDR3-WT, WDR3-MUT, and MUT-IDR. (C) FRAP of MUT-IDR. (D) The droplet formation of WDR3 in U2-OS cells transfected with WDR3-WT, WDR3-MUT, and MUT-IDR. (E) FRAP of WDR3-MUT and MUT-IDR in U2-OS cells. (F–G) The proliferation (F) and metastatic ability (G) of U2-OS cells transfected with WDR3-WT, WDR3-MUT, and MUT-IDR. *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/7446d2b582681958991fdc7b.png"},{"id":82794766,"identity":"2840d9f3-6746-4ccf-94a8-41e56cf78ef4","added_by":"auto","created_at":"2025-05-15 10:31:13","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":36795089,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Nilotinib treatment on WDR3 phase separation and tumor metastasis \u003cem\u003ein vitro.\u003c/em\u003e (A) Changes of cell proliferation under treatment of Nilotinib with different concentrations. (B–C) Effect of Nilotinib treatment on WDR3 mRNA (B) and protein (C) expression. (D) The droplet formation of WDR3 in U2-OS cells treated with Nilotinib. (E) FRAP of WDR3 droplets in U2-OS cells treated with Nilotinib. (F) Effect of Nilotinib treatment on migration and invasion of U2-OS cells. *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003eP\u003c/em\u003e\u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/72d5752aedf09b88fe45e067.png"},{"id":82794787,"identity":"505acc1c-5554-40eb-acfc-cad3efbbe3db","added_by":"auto","created_at":"2025-05-15 10:31:15","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":54281909,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Nilotinib treatment on WDR3 phase separation and tumor progression \u003cem\u003ein vivo.\u003c/em\u003e (A) The schedule for animal experiments. (B) Tumor volume and weight in xenograft mice of osteosarcoma with or without Nilotinib treatment. (C–D) The protein (C) and mRNA (D) expression of WDR3 in mice treated by Nilotinib. (E) Effect of \u003cem\u003ein vivo\u003c/em\u003etreatment of Nilotinib on the number of WDR3 condensates. *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/5c4fbc939eeb9a012d2bc74b.png"},{"id":82796488,"identity":"e013d203-ec46-4626-b138-39c65863b7be","added_by":"auto","created_at":"2025-05-15 10:39:14","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":17498566,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic illustration for the mediation of WDR3 phase separation in the treatment of Nilotinib against osteosarcoma metastasis.\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/133eeb45bb4bd76c748d6ff2.png"},{"id":82792493,"identity":"1d879137-1bfc-4d6a-b2e8-a6d359d6c8f6","added_by":"auto","created_at":"2025-05-15 10:23:12","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1769514,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/ff9a6226f5ca7f3f205bce93.docx"},{"id":82792507,"identity":"99709b03-414e-42e6-a132-63cb15ec2e18","added_by":"auto","created_at":"2025-05-15 10:23:13","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":16289,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/75a085486fbbe93bad4e1df2.docx"},{"id":82792502,"identity":"29991b00-4f1e-4ae6-a98b-280d40d05f82","added_by":"auto","created_at":"2025-05-15 10:23:13","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":493864,"visible":true,"origin":"","legend":"","description":"","filename":"westernblot.docx","url":"https://assets-eu.researchsquare.com/files/rs-6611672/v1/71b3358e9f2064e5c8c98a99.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"WDR3 undergoes phase separation to mediate the therapeutic mechanism of Nilotinib against osteosarcoma","fulltext":[{"header":"Highlights","content":"\u003cp\u003e1. LLPS-related prognostic signatures including WDR3 can predict survival status in osteosarcoma.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2. WDR3 and Nilotinib exhibited the stable receptor-ligand binding ability in molecular docking.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3. WDR3 formed droplets in U2-OS cells and promoted tumor metastasis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4. Nilotinib inhibited osteosarcoma progression\u003cem\u003e\u0026nbsp;\u003c/em\u003eby disrupting the phase separation of WDR3.\u0026nbsp;\u003c/p\u003e"},{"header":"1. Introduction","content":"\u003cp\u003eOsteosarcoma is a primary solid bone tumor of mesenchymal cell origin, commonly found in the metaphyses of skeleton (including the femur and tibia) and knee joints [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Its incidence is bimodal in age, occurring more in adolescents and older adults over the age of 60, with 4.4 cases of osteosarcoma diagnosed annually per million people worldwide [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The 5-year survival rate for patients with localized tumor can be 70%, but osteosarcoma is highly aggressive, with distal metastases found in about 15\u0026ndash;20% of diagnosed cases and most of them involving the lungs [\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Unfortunately, the metastatic rate of osteosarcoma is over 85% and the 5-year overall survival of metastatic patients is less than 20% [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In addition to traditional treatments such as surgery, chemotherapy and radiotherapy, targeted therapy and immunotherapy are also widely used in the clinical management of osteosarcoma; however, drug resistance may be the most important reason of reduced survival time, especially for patients with metastatic phenotype [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Genomic instability and aberrations characterize the majority of osteosarcoma cases [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], therefore, the development of potential biomarkers from a genetic perspective is an effective strategy to benefit patient with targeted therapies and protect them from drug resistance.\u003c/p\u003e \u003cp\u003ePhase separation is a phenomenon that occurs when a mixture of molecules spontaneously separates into two distinct phases with different compositions and concentrations of specific factors [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. In cellular physiology, biomolecules aggregate to form droplet-like structures in the intracellular environment through liquid-liquid phase separation (LLPS) [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Generally, the intrinsically disordered region (IDR) drives LLPS through interactions between multiple amino acid residues [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. This physiological phenomenon creates a specialized microenvironment that regulates various cellular processes, such as DNA replication, RNA processing, ribosome biogenesis, apoptosis, and signal transduction, thereby affecting tumor progression [\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. It was reported that the aggregation of core regulatory circuitry and transcriptional machinery proteins on super-enhancers via LLPS can promote the activation of genes associated with osteosarcoma metastasis or drug resistance [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. The inhibition of MYC-driven super-enhancers signaling can counteract the elimination of osteosarcoma cell proliferation, migration, and invasion [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Kim et al. demonstrated that ARID1A promotes the oncogenic potential of osteosarcoma through Prion-like domain-mediated LLPS [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. To better study the mechanism of LLPS, Sun et al. summarized various bioinformatics databases and tools [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], but these approaches have not yet been applied in osteosarcoma to identify biomarkers that may affect disease progression and drug sensitivity.\u003c/p\u003e \u003cp\u003eTherefore, this study adopted bioinformatics methods to screen LLPS-related biomarkers with prognostic value for osteosarcoma and explored their relationships with immune landscape and drug sensitivity. Molecular docking simulated binding modalities of prognostic signatures and drugs to screen for receptor-ligand complexes with structural stability. Furthermore, the involvement of phase separation of candidate biomarker in the drug therapy against osteosarcoma was also explored by cell- and animal-based experiments. The biomarkers reported in this study and their mediated phase separation mechanism are expected to provide new theories and ideas for the treatment of osteosarcoma.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Bioinformatic analysis\u003c/h2\u003e \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003e2.1.1 Data collection and preprocessing\u003c/h2\u003e \u003cp\u003eThe GSE16091, GSE21257, and GSE39055 datasets containing survival information and gene expression profiling data of 124 osteosarcoma patients were collected from the Gene Expression Omnibus (GEO) database, and then merged to remove the batch effect using the ComBat function of sva package in R4.2.2. Furthermore, GSE39058 was screened from GEO as external validation cohort 1; while 85 samples including survival time and expression matrix were collected from the Target database and served as external validation cohort 2. In addition, the GSE126209 dataset containing 12 osteosarcoma samples and 11 normal controls was used to screen for differentially expressed genes (DEGs). Using the limma package in R4.2.2 software, DEGs between osteosarcoma and controls were filtered with a threshold of \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log fold_change| \u0026gt; 1. A total of 3783 LLPS-related genes were obtained from DrLLPS (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://llps.biocuckoo.cn/\u003c/span\u003e\u003cspan address=\"http://llps.biocuckoo.cn/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), LLPSDB (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://bio-comp.org\u003c/span\u003e\u003cspan address=\"http://bio-comp.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e .cn/llpsdb/home.html) and PhaSepDB (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://db.phasep.pro/\u003c/span\u003e\u003cspan address=\"http://db.phasep.pro/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) databases. The intersection of DEGa and LLPS-related gene was visualized by ggplot2 and VennDiagram package in R4.2.2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.1.2 Screening of prognostic biomarkers for constructing a predictive model\u003c/h2\u003e \u003cp\u003eBased on the data in the combined dataset, Cox regression analysis was performed using the coxph function of survival package in R4.2.2 to screen for prognosis-related genes. Further application of least absolute shrinkage and selection operator (LASSO) regression using glmnet package provided the optimal combination of prognosis-related genes with parameter of lambda.min. To avoid overfitting of the model, 50% of samples from the combined dataset were used as the internal training set and the remaining ones were used as the internal validation set. External validation cohorts 1 and 2 were implemented to verify the model stability. The timeROC and survminer packages were used to plot Kaplan-Meier (KM) and receiver operating characteristic (ROC) curves, respectively, to quantify the model prediction results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.1.3 Molecular subtype of osteosarcoma\u003c/h2\u003e \u003cp\u003eConsensusClusterPlus was used to carry out the concensus clustering analysis, and the optimal number of clusters was selected based on the cumulative distribution function (CDF) curves and principal component analysis (PCA) results. The clustering criteria were as follows: the number of samples in each group was relatively consistent; the CDF curves gradually increased; the samples in each group were aggregated, with obvious differences between groups. Differences between survival status and immune checkpoint expression among subtypes were compared using KM curves and \u003cem\u003et\u003c/em\u003e test.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.1.4 Multidimensional comparison between prognostic risk groups\u003c/h2\u003e \u003cp\u003eTo predict the pathway scores for each sample in the combined dataset, the GSVA package was applied with c2.cp.kegg.v2023.1.Hs.symbols.gmt in MSigDB database as the background pathway. Subsequently, the limma package was employed to screen pathways with significant differences in scores between high- and low-risk groups, with \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 as the threshold. The expression matrix of immune cell subtypes was also deconvoluted using CIBERSORT as a means of estimating the infiltration abundance of immune cell for each sample in the combined dataset. To discern the likelihood of tumor immune evasion, TIDE score and microsatellite instability (MSI) score were calculated on TIDE online tool based on the conventional immune checkpoint expression of individual samples in this study. Based on a drug prediction model developed by the GDSC database, this study predicted the sensitivity of samples to different drugs using oncoPredict package. Differences in immune cell infiltration levels, TIDE scores, and drug sensitivities between high- and low-risk groups were assessed using \u003cem\u003et\u003c/em\u003e test, followed by Pearson correlation analyses for prognostic signatures.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.1.5 Molecular docking\u003c/h2\u003e \u003cp\u003eTo predict effective blinding of drug and prognostic signatures, molecular docking analysis was accomplished in this study. The 3D structures of the drugs were retrieved from PubChem database and exported into PDB format using PyMol software. Following, the charge of the ligand was adjusted and twistable bonds were selected in AutoDockTools-4.2.6 software. For receptors, their gene IDs were retrieved from the Uniprot database. Protein 3D structures were collected from the PDB and AlphaFoldDB databases and imported into PyMol as a PDB file with water molecule sequences removed. Subsequently, a series of processes were performed on the receptor in AutoDockTools-4.2.6, including removal of the original ligand, addition of hydrogens, optimization of amino acids, and calculation of charge. The active region of binding pocket for docking was determined based on the original ligand position in the protein receptor, while the molecular docking was carried out according to receptor name, ligand name, coordinates of the docking centroid, and distances incoming from AutoDock vina. Valid docking results were output if binding energy was less than \u0026minus;\u0026thinsp;5.0 kcal/mol and hydrogen bonds between receptor-ligand complex could be formed.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.2 cell culture and treatment\u003c/h2\u003e \u003cp\u003eThe human fetal osteoblast cell line hFOB (CL-233h, SAIOS biotechnology) and the osteosarcoma cell line U2-OS (CL-655h, SAIOS biotechnology) were cultured in DMEM containing 10% fetal bovine serum (FBS, 16140071, Gibco) and 1% penicillin\u0026ndash;streptomycin (C11885500BT, Gibco). Another two osteosarcoma cell lines HOS (CL-156h, SAIOS biotechnology) and MG-63 (CL-157h, SAIOS biotechnology) were cultured in MEM (11095080, Gibco) using the same strategy to validate the expression patterns of five prognostic signatures.\u003c/p\u003e \u003cp\u003eTo observe the effect of WDR3 silencing on osteosarcoma \u003cem\u003ein vitro\u003c/em\u003e, sh-WDR3 was transfected into U2-OS cells by lentivirus using Lipofectamine\u0026trade; 2000 Transfection Reagent (11668030, Invitrogen) in this study, with both control and sh-negative control (NC) groups being set up. Sequences of sh-WDR3 (SS Sequence: GGTTCTCTCTAATCTATAA; AS Sequence: TTATAGATTAGAGAGAACC) were designed in the Designer of Small Interfering RNA website.\u003c/p\u003e \u003cp\u003eTo explore the pharmacological mechanism of Nilotinib (SC0209, Beyotime), U2-OS cells were treated with different concentrations of Nilotinib (1\u0026micro;M, 2.5\u0026micro;M, 5\u0026micro;M) for 24 h. The optimal concentration (5\u0026micro;M) of Nilotinib was then selected to treat the cells for 24h according to cell viability to explore changes in cell function and WDR3 phase separation level.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Quantitative real-time polymerase chain reaction (qPCR)\u003c/h2\u003e \u003cp\u003eThe 1 mL of Trizol (15596018, Invitrogen) was added to cells or tissues to release total RNA. After a reverse transcription reaction, cDNA was synthesized and configured with primers for the PCR reaction system. Sequences of primers are detailed in Supplementary Table\u0026nbsp;1. On a PCR instrument (CFX Connect, BIO-RAD), the PCR reaction was carried out and underwent 40 cycles (95 ℃, 3 min; 95 ℃, 12 s; 62 ℃, 40 s) of amplification. Relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), the mRNA level of target genes was calculated using a 2\u003csup\u003e\u0026minus;ΔΔCT\u003c/sup\u003e method.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Western blotting\u003c/h2\u003e \u003cp\u003eCells and tissues were lysed to release total proteins for quantification. Then, samples were loaded and run in sodium dodecyl sulfate-polyacrylamide gel electrophoresis to separate proteins, which were then transferred to polyvinylidene fluoride membranes (FFP24; Beyotime). Membranes were then incubated overnight in diluted primary antibody working solution (anti-WDR3, 1:1000, PA5-144030, Thermo; anti-GAPDH, 1:2500, ab181602, Abcam). After washing thrice, the membranes were placed in a 2000-fold diluted secondary antibody (Goat Anti-Rabbit IgG H\u0026amp;L, ab6721, Abcam) for another incubation of 1 h. Finally, the membranes were developed in enhanced chemiluminescence (ECL) solution (P1000, APPLYGEN), followed by the scanning of exposed films. ImageJ software was used to quantify the gray values of bands, and the protein expression of WDR3 relative to GAPDH was calculated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Cell counting kit-8 (CCK-8)\u003c/h2\u003e \u003cp\u003eTransfected or Nilotinib-treated U2-OS cells were inoculated in 96-well plates (2000 cells per well). After 24 h of routine incubation (37 ℃, 5% CO\u003csub\u003e2\u003c/sub\u003e), each well was supplemented with 10 \u0026micro;L of CCK-8 reaction solution (C0037, Beyotime). Two hours later, the optical density of each pore at 450 nm was measured by microplate reader (DR-3518G, Wuxi Hiwell Diatek) to evaluate cell viability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Transwell assay\u003c/h2\u003e \u003cp\u003eIn this study, Transwell assay was conducted to assess the migration and invasive ability of cells. To observe cell migration, we inoculated digested U2-OS cells into Transwell chambers, which was then cultured in the lower chamber containing medium for 24 h culture under conventional conditions. Later on, Transwell chamber was fetched out to remove medium. After undergoing washing and fixation, crystal violet (C0121, Beyotime) staining was added to the chambers for 20 min to observe unmigrated cells. For invasion detection, we covered the bottom of the Transwell chambers with 50 mg/L Matrigel (354234, Corning) diluted at 1:4 until it polymerized into a gel. The same methods for cell culture, fixation, and staining were repeated to observe the cell invasion in three randomly selected fields under the microscope.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Animals\u003c/h2\u003e \u003cp\u003eThirty male Balb/c nude mice (5\u0026ndash;6 weeks old) were purchased from the Experimental Animal Center of Yangzhou University, where they were housed under a 12-hour light/dark cycle and had free access to adequate food and water. Of which, 18 mice were randomly divided into three groups (OS, OS\u0026thinsp;+\u0026thinsp;sh-NC, and OS\u0026thinsp;+\u0026thinsp;sh-WDR3; six mice per group) and injected subcutaneously in the right axilla with 1\u0026times;10\u003csup\u003e7\u003c/sup\u003e U2-OS, sh-NC-transfected U2-OS, and sh-WDR3-transfected U2-OS cells, respectively, for xenograft tumor formation. Tumor volumes were recorded every seven days over a four-week period. On day 28th, mice were anesthetized with 4% isoflurane (HY-A0134, MCE) and then euthanized for tumor collection.\u003c/p\u003e \u003cp\u003eAnother 12 mice were randomized into OS and OS\u0026thinsp;+\u0026thinsp;Nilotinib (OS\u0026thinsp;+\u0026thinsp;NIL) groups (n\u0026thinsp;=\u0026thinsp;6) and injected subcutaneously with 1\u0026times;10\u003csup\u003e7\u003c/sup\u003e U2-OS cells to constructed the xenograft model of osteosarcoma. One week later, Nilotinib was diluted in dimethyl sulfoxide (D8371, Solarbio) and administered for the gavage treatment of mice in the OS\u0026thinsp;+\u0026thinsp;NIL group (30 mg/kg) via an oral delivery carrier containing 0.5% hydroxypropyl methylcellulose (HPMC, HY-A0104J, MCE) and 0.05% Tween80 (TB360, Solarbio). The OS group was gavaged with an equal dose of 0.5% hydroxypropyl methylcellulose and 0.05% Tween80. After gavage therapy once a day for three weeks, the mice were euthanized to collect tumor samples. All animals were carried out under the approval of the Experimental Animal Ethics Committee of Yangzhou University (No.202312013).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Immunohistochemical (IHC)\u003c/h2\u003e \u003cp\u003eTumor tissues from each group of mice were prepared into sections and incubated overnight with a 1000-fold dilution of WDR3 (PA5-144030, Thermo) and Ki67 (9129S, Cell Signaling) antibodies. The washed sections were incubated with a secondary antibody (Goat Anti-Rabbit IgG H\u0026amp;L, ab6721, Abcam) for 15 min at a dilution of 1:2000. To visualize the target proteins, the sections were first stained in DAB staining solution (P0202, Beyotime) for 30 min, followed by re-staining with hematoxylin (G1080, Solarbio) for 3 min. After cleaning, the slices were dried, dehydrated, sealed, and photographed under a microscope in turn.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labelling (TUNEL)\u003c/h2\u003e \u003cp\u003eThe prepared tumor tissue sections were routinely dewaxed and hydrated. Following the instructions of TUNEL kit (C1091, Beyotime), 50 \u0026micro;L of the configured proteinase K working solution was added dropwise to the sections for digestion of 30 min, followed by the incubation with a mixture of 5\u0026micro;L TdT enzyme, 45\u0026micro;L fluorescent labeling solution, and 50\u0026micro;L TUNEL test solution for 30 min. After washing, the sections were stained with 4',6-diamidino-2-phenylindole (DAPI, C1005, Beyotime) for 10 min. Ultimately, sections were sealed with antifade mounting medium (p0126, Beyotime) and observed under a microscope.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Recombinant protein expression and purification\u003c/h2\u003e \u003cp\u003eThe WDR3 gene was cloned into the pET-28b (+) expression vector (with GFP and His tags) and transformed into \u003cem\u003eE. coli\u003c/em\u003e strain BL21 (DE3). The transformed strains were cultured in LB liquid medium containing 100 \u0026micro;g/mL Ampicillin (ST008, Beyotime) for 37\u0026deg;C at 170 rpm until the OD600 of bacterial fluids reached 0.6\u0026ndash;1.0. The addition of isopropyl-beta-D-thiogalactoside (ST098, Beyotime) with a final concentration of 1mM induced the protein expression at 37℃ for 4 h. Afterwards, the bacterial fluids were centrifuged at 15,000g for 1 min at 4\u0026deg;C to collect precipitate. The harvested bacterial precipitate was resuspended and lysed by sonication on ice. Then, the recombinant proteins were purified with BeyoGold\u0026trade; His-tag Purification Resin (P2233, Beyotime). Finally, the proteins were eluted on an elution column and concentrated by ultrafiltration.\u003c/p\u003e \u003cp\u003eTo construct protein mutants, the intrinsically disordered region (IDR) of WDR3 was predicted on the IUPred2A website. Predictions guided the construction of three WDR3 IDR mutants (WDR3-MUT1, WDR3-MUT2, and WDR3-MUT3) and the fusion to the IDR of hnRNPA1, known to drive condensate formation, to synthesize a rescue phase-separated WDR3 mutant (MUT-IDR). WDR3-MUT and MUT-IDR were then cloned into pET-28b (+) expression vectors (GFP and His tags) to induce protein expression, respectively. Protein expression and purification procedures were described above.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e2.11 Droplet formation assay\u003c/h2\u003e \u003cp\u003eIn this study, 20 \u0026micro;M WDR3-GFP purified protein was added to a buffer containing 10% PEG8000 (81268, Sigma) and NaCl (ST1641, Beyotime) with different concentrations (75 mM, 150 mM, 300 mM). The protein solution was then loaded onto a slide and condensates with liquid-like behavior were observed by fluorescence microscopy. The same method was also applied to observe droplet formation of purified WDR3-GFP protein in 125 mM NaCl and 10%PEG buffer at different concentrations (10 \u0026micro;M, 20 \u0026micro;M, 40 \u0026micro;M).\u003c/p\u003e \u003cp\u003eTo investigate intracellular droplet formation, lentiviral expression vectors for WDR3-MUT and MUT-IDR were constructed to transfect U2-OS cells. Cells were climbed and cultured overnight until wall attachment. The droplet formation in the cells was observed under a confocal laser scanning microscope (CLSM, TCS SP8, Leica).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e2.12 Fluorescence recovery after photobleaching (FRAP)\u003c/h2\u003e \u003cp\u003eTo further verify the recovery of fluorescence activity of WDR3 after photobleaching, 20 \u0026micro;M of WDR3-GFP purified proteins were added in a buffer containing 10% PEG8000 and 125 mM NaCl to form droplets and then subjected to FRAP assay by CLSM. The FRAP was programmed with a 488 nm 100% power laser, a 1.5\u0026micro;m radius area as the target focus, and a bleaching time of 20 s. The recovery of protein condensates after 40 s and 60 s of photobleaching was photographed. The same procedures were applied to detect phase separation of exogenous WDR3 in WDR3-GFP-, WDR3-MUT-, or MUT-IDR-transfected U2-OS cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e2.13 Immunofluorescence\u003c/h2\u003e \u003cp\u003eIn this study, immunofluorescence was used to examined the formation of endogenous and exogenous WDR3 phase-separated condensates in cells. To facilitate observation, cells were sequentially clamped, immobilized, sealed, and incubated in 200 \u0026micro;L of anti-WDR3 (9129S, Cell Signaling) diluted at 1:200 overnight. The next day, cells were incubated in a 1:500 diluted mixture of IgG H\u0026amp;L and DAPI for 30\u0026ndash;60 min. After sealing slices, the fluorescence of each group was detected under a CLSM.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e2.14 Statistical analysis\u003c/h2\u003e \u003cp\u003eAll data presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation was processed on GraphPad 10.1.2. Comparisons between two groups were conducted using unpaired \u003cem\u003et\u003c/em\u003e test, while comparisons between multiple groups were performed using one-way analysis of variance (ANOVA) with Tukey's post hoc test. Differences between groups under continuous time were compared using two-way RM ANOVA. A \u003cem\u003eP\u003c/em\u003e-value below 0.05 was defined as statistical significance.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Predictive efficiency of LLPS-related signatures in osteosarcoma prognosis\u003c/h2\u003e\n \u003cp\u003eBased on the expression profiles of the GSE126209 dataset, this study screened out 2489 DEGs between osteosarcoma and normal samples, of which 1497 DEGs were up-regulated and 992 DEGs were down-regulated in osteosarcoma (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). Within DEGs, 473 of them belonged to LLPS-related genes (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). To screen prognostically relevant signatures, the GSE16091, GSE21257, and GSE39055 datasets were merged. After removing the batch effect, the samples in the dataset are evenly distributed with no significantly discrete clusters (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eC). In the combined dataset, 15 genes significantly associated with survival time in osteosarcoma (log-rank test, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were screened out from the univariate Cox regression algorithm (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eD). Among these 15 genes, only ANXA10 and SMURF2 were prognostic protective factors for osteosarcoma. LASSO was then carried out to screen for models with excellent performance but minimal number of variables (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eE). With the optimal \u0026lambda; value, a gene list including ANXA10, MYC, TIMM8A, WASF3, and WDR3 were identified as prognostic signature to establish a predictive system according to the equation as Risk Score = -0.651*ANXA10\u0026thinsp;+\u0026thinsp;0.974*MYC\u0026thinsp;+\u0026thinsp;0.051*TIMM8A\u0026thinsp;+\u0026thinsp;0.234*WASF3\u0026thinsp;+\u0026thinsp;0.347*WDR3. The samples were categorized into high- and low-risk groups for comparison based on the median risk score. In both the internal training and validation sets, patients in the high-risk group were significantly less likely to survive than those in the low-risk group, while the area under curve (AUC) values of 1-, 3-, and 5-year ROCs were all above 0.7 (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eF-G). To further assess the predictive efficacy of the model, we conducted validation in two independent external cohorts. The results confirmed that patients in the high-risk group had a significantly lower probability of survival than those in the low-risk group, with the predictive model being highly sensitive and specific (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eH-I).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Characteristic evaluation of prognostic signatures\u003c/h2\u003e\n \u003cp\u003eThereafter, this study performed a multidimensional comparison of differences between high- and low-risk groups. First, molecular subtypes of osteosarcoma were identified using consensus clustering analysis. The results showed that k\u0026thinsp;=\u0026thinsp;2 appeared to be the best choice for grouping the samples into cluster 1 and 2 (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA). Patients in cluster 1 demonstrated a more favorable prognosis compared to those in cluster 2 (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eB), and were more commonly distributed in the low-risk group, which was prone to the status of alive (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC). The Limma package also identified 37 pathways with differences in GSVA scores between high and low risk groups (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eD). The majority of these pathways was significantly positively correlated with ANXA10, but markedly negatively correlated with MYC, TIMM8A, and WDR3 (Supplementary Fig. 1A). There were also significant differences in infiltration levels of CD8 T cells and activated mast cells between risk groups (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eE). Among them, the infiltration level of CD8 T cells was significantly negatively correlated with MYC expression, while activated mast cells were positively correlated with MYC and WDR3 (Supplementary Fig. 1B). Furthermore, patients in the low-risk group exhibited significantly lower TIDE and MSI scores than those in the high-risk group, suggesting a lower likelihood of tumor immune escape to benefit from immunotherapy (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eF).\u003c/p\u003e\n \u003cp\u003eIn terms of drug sensitivity, a total of 24 drugs were identified with significant differences in IC50 values between the two groups (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eG). These drugs were significantly correlated with the expression of MYC, but had no significant association with WASF3 (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eH). Moreover, the expression of WDR3 may affect the sensitivity of osteosarcoma patients to drugs, such as AZD4547 and Nilotinib (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eH). To probe into the possible binding of proteins encoded by five prognostic signatures with 24 drugs, molecular docking was conducted and 100 valid docking results were obtained. Among them, the interaction of WDR3 and Nilotinib depicted the lowest binding energy (-9.3kcal/mol), with hydrogen bond on residues such as GLN-154 and LEU-285 contributing to the stability of receptor-ligand complex (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eI). Nilotinib was also bound to WDR3 residues by Alkyl, Halogen, Pi-Alkyl, and Pi-Cation (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eI).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Effect of prognostic biomarker WDR3 on the progression of osteosarcoma \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e\u003c/h2\u003e\n \u003cp\u003eCompared with normal osteoblastic hFOB, this study validated the expression patterns of five prognostic signatures in three osteosarcoma cell lines (U2-OS, HOS, MG-63). qPCR results suggested that the candidate phase separation-related genes were all up-regulated in osteosarcoma cells (Fig. 3A), which was consistent with the results of Bioinformatic analyses. Considering the binding stability of WDR3 to Nilotinib, this study focused on exploring the therapeutic mechanism of WDR3 phase separation in Nilotinib against osteosarcoma. We used lentiviral transfection to specifically down-regulate WDR3 expression in U2-OS cells in order to observe its effect on osteosarcoma progression \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e. As expected, sh-WDR3 significantly reduced WDR3 mRNA and protein expression levels in U2-OS cells compared to those transfected with sh-NC (Fig. 3B\u0026ndash;C). In the meantime, U2-OS cells transfected with sh-WDR3 demonstrated significant inhibition of cell viability (Fig. 3D), as well as migration and invasion (Fig. 3E), compared to their negative controls. \u003cem\u003eIn vivo\u003c/em\u003e, downregulation of WDR3 also significantly reduced tumor volume and weight in xenografted mice of osteosarcoma (Fig. 3F). In addition, sh-WDR3 transfection significantly reduced the positive area of WDR3 and Ki67 in mice tumor tissues (Fig. 3G), while promoting apoptosis (Fig. 3H), relative to controls. Thus, the silencing of WDR3 can effectively alleviate the malignant progression of osteosarcoma.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec26\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4 WDR3 exhibited phase-separated condensates with liquid-like behavior in OS cells\u003c/h2\u003e\n \u003cp\u003eThe droplet formation experiments showed that WDR3 can form circular condensates in the presence of low concentration of salt solution, while the number of droplets significantly reduced with the increase of NaCl concentration (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA). The number of droplets also increased with the raised concentration of WDR3-GFP recombinant protein (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB). After photobleaching, the fluorescence signals in the target field were weakened, but recovered rapidly within 60 s, suggesting the liquid-like behavior inside the droplets (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eC). In U2-OS cells, the presence of endogenous WDR3 were observed by immunofluorescence (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eD). Furthermore, U2-OS cells exogenously transfected with WDR3-GFP demonstrated stronger fluorescence and larger condensates of WDR3 (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eE). After photobleaching of this region, the fluorescence signal of WDR3 condensate can also be gradually recovered (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eF), further confirming that WDR3 can form droplets through phase separation.\u003c/p\u003e\n \u003cp\u003eTo further differentiate the phase-separation ability of WDR3, we mutated its IDR, where amino acid mutations have been shown to alter the phase-separation properties of proteins thereby impairing their recombination and reprogramming [\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e]. Using the IUPred2A online tool, this study predicted that IDR of WDR3 was mainly concentrated in the region of amino acids 231\u0026ndash;257, 319\u0026ndash;348, and 712\u0026ndash;743, which may contribute to the formation of dimers or multimers (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eG). Furthermore, all glutamate and aspartate residues within these three regions were substituted with alanine (Supplementary Fig. 2), guiding the construction of three WDR3 IDR mutants, named WDR3-MUT1, WDR3-MUT2, and WDR3-MUT3 (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eH). Among these, only WDR-MUT1 significantly reduced the number of intracellular WDR3 droplets (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eI) and slowed WDR3 condensates reformation after photobleaching (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eJ), prompting its selection for subsequent investigations.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec27\" class=\"Section2\"\u003e\n \u003ch2\u003e3.5 Phase separation of WDR3 expedited OS metastasis \u003cem\u003ein vitro\u003c/em\u003e\u003c/h2\u003e\n \u003cp\u003eThe IDR of hnRNPA1 is conducive to protein assembly and droplet properties [\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e], and thus can be fused to WDR3-MUT to rescue the phase separation of WDR3 (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eA). Under the microscope, WDR3-MUT was found to disrupt the droplet-forming ability of WDR3, resulting in dispersed fluorescence and inability to aggregate. This dysfunction was then rescued by the fusion of hnRNPA1 IDR, allowing WDR3 to reaggregate and form droplets (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eB). After photobleaching the target region of MUT-IDR, the fluorescence intensity of WDR3 was gradually restored and the droplet aggregation ability was rescued (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eC).\u003c/p\u003e\n \u003cp\u003eAfter the transfection of WDR3-WT, WDR3-MUT, and MUT-IDR into U2-OS cells, changes in WDR3 phase-separation characteristics were also investigated in this study. As expected, the fluorescence in the WDR3-MUT group began to disperse, accompanied by a decrease in the number of droplets. However, the droplets in the MUT-IDR group were reconcentrated and the number of droplets recovered significantly (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eD). FRAP results indicated that the WDR3-MUT-transfected cells remained unchanged fluorescence in the photobleached regions, which could be recovered by the transfection of MUT-IDR (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eE). More importantly, mutation of WDR3 significantly down-regulated the proliferation and metastatic ability of U2-OS cells, but the transfection of MUT-IDR restored the malignant phenotype of tumor cells (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eF\u0026ndash;G). Thus, the phase separation of WDR3 could promote the proliferation, migration and invasion of osteosarcoma \u003cem\u003ein vitro\u003c/em\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec28\" class=\"Section2\"\u003e\n \u003ch2\u003e3.6 Nilotinib mitigated osteosarcoma progression via inhibition of WDR3 phase separation\u003c/h2\u003e\n \u003cp\u003eTo clarify the therapeutic mechanism of Nilotinib on osteosarcoma, we first treated U2-OS cells with different concentrations of Nilotinib \u003cem\u003ein vitro\u003c/em\u003e for 24 h. Cells treated by Nilotinib were significantly reduced in proliferation in a dose-dependent manner (Fig. 6A). Nilotinib treatment also significantly reduced the mRNA and protein expression levels of WDR3 in U2-OS cells (Fig. 6B\u0026ndash;C). Notably, the fluorescence intensity of WDR3 and punctas per cell were significantly increased in U2-OS cells compared to hFOB cells, whereas Nilotinib treatment significantly suppressed these abnormalities (Fig. 6D). The FRAP results suggested that the fluorescence intensity of WDR3 condensates in untreated U2-OS cells after photobleaching could recover within 60 s, which was not observed in Nilotinib-treated cells (Fig. 6E)). Moreover, the addition of Nilotinib significantly reduced the migration and invasion abilities of U2-OS cells (Fig. 6F).\u003c/p\u003e\n \u003cp\u003eThe animal experiments were also carried out to confirm the role of WDR3 phase separation in the treatment of osteosarcoma by Nilotinib. The schedule for animal experiments is shown in Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eA. Consistent with the above results, treatment of Nilotinib significantly reduced tumor volume and weight in xenografted mice of osteosarcoma (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eB). Meanwhile, continuous Nilotinib treatment also significantly down-regulated the mRNA and protein expression levels of WDR3, compared to untreated mice (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eC\u0026ndash;D). Immunofluorescence showed that with the treatment of Nilotinib, the number of WDR3 condensates decreased significantly (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eE), suggesting a downregulation of its phase separation level. Therefore, Nilotinib may alleviate the progression of osteosarcoma \u003cem\u003ein vivo\u003c/em\u003e and \u003cem\u003ein vitro\u003c/em\u003e by inhibiting the phase separation of WDR3.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eThe bottlenecks in prolonging survival of patients with metastatic osteosarcoma have prompted the search for a number of prognostic biomarkers. Starting with the LLPS mechanism, this study screened out five LLPS-related genes (ANXA10, MYC, TIMM8A, WASF3, and WDR3) with prognostic values by bioinformatics methods, and the risk models constructed accordingly could predict the survival likelihood of osteosarcoma. These prognostic signatures were significantly associated with immune cell infiltration, tumor immune escape, and drug sensitivity. Among them, WDR3 and Nilotinib demonstrated optimal binding stability in molecular docking models. WDR3 is not only a prognostic risk factor for osteosarcoma, but also highly expressed in U2-OS cells. Functional experiments have shown that the knockdown of WDR3 inhibited the proliferation and metastatic ability of osteosarcoma cells while suppressing tumor growth. More importantly, WDR3 can form condensates with liquid-like behavior in U2-OS cells, and mutation of its IDR can eliminate the phase-separated level of WDR, thus reversing the aggressive phenotype of osteosarcoma cells.\u003c/p\u003e \u003cp\u003eWDR3, also known as DIP2 and UTP12, belongs to a family of eukaryotic genes that carry the WD repeat region [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. It encodes a 943 amino acid nuclear protein composed of a 10 WD repeat sequence module localized to human chromosome 1p12-p13 [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Su et al. found that overexpression of WDR3 is associated with low survival in cancer and promotes pancreatic cancer proliferation and invasion by interacting with GATA4 to induce Hippo pathway activation [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. In thyroid cancer, WDR3 maintains genomic stability in patients, while its location on the 1p12 chromosome contributes to disease susceptibility [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. WDR3 expression is linked to inflammatory mediators and can be reduced following the treatment with anti-inflammatory drugs [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. McMahon et al. proposed that WDR3 deficiency leads to ribosome biogenesis defects by affecting 18s rRNA processing, thereby reducing p53-mediated proliferation of cancer cells [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Although no studies have been conducted on WDR3 in osteosarcoma, it has been found to promote cancer stem cell characteristics by inhibiting USF2-mediated RASSF1A transcription [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. In a multicenter case-control study, genetic polymorphisms in RASSF1A were found to be associated with the risk of osteosarcoma and metastasis in young Chinese adults [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. RASSF1A plays a tumor suppressor role in osteosarcoma and exerts anticancer effects by regulating the Wnt/β-catenin pathway [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. These findings further confirmed our research, suggesting that WDR3 may have the potential to promote osteosarcoma metastasis by inhibiting transcriptional regulation of RASSF1A. Furthermore, our results indicated that WDR3 expression was significantly positively correlated with the spliceosome pathway score, but had negative correlations with the lysosome pathway. SNRPB, a core component of the spliceosome, was shown to induce malignant behavior in osteosarcoma [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. C1GALT1 enhances the drug resistance and metastatic propensity of osteosarcoma by promoting lysosomal degradation and effluence [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Therefore, we speculated that WDR3 phase separation-induced osteosarcoma progression is associated with SNRPB upregulation and lysosomal degradation.\u003c/p\u003e \u003cp\u003eIn terms of therapeutic potential, WDR3 bound stably to Nilotinib in this study and mediated the therapeutic mechanism of Nilotinib against osteosarcoma. As a second-generation tyrosine kinase inhibitor that allows for a faster deep molecular response, Nilotinib is approved for the first-line treatment of BCR-ABL-positive chronic granulocytic leukemia [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. In addition to blood disorders, Nilotinib has been used to treat gastrointestinal stromal tumors and benefits patients with KIT exon 11 mutant phenotypes [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. By targeting the suppression of DDR1, Nilotinib effectively blocked the migration of breast cancer cells [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Nilotinib also improved outcomes in colorectal cancer patients receiving anti-PDL1 therapy by restoring MHC-I expression [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. In osteosarcoma, Nilotinib was found to downregulate the expression of prognostic marker MAPK1, thereby promoting apoptosis [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Notably, Wei et al. suggested that Nilotinib may exert therapeutic benefits by targeting phase separation in cells [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. The present study demonstrated that Nilotinib significantly inhibited the production of phase-separated condensates of WDR3, thereby suppressing tumor growth and metastasis \u003cem\u003ein vivo\u003c/em\u003e and \u003cem\u003ein vitro\u003c/em\u003e (as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Furthermore, WDR3 expression was positively correlated with the infiltration level of activated mast cells in osteosarcoma. In rat peritoneal mast cells, Nilotinib decreased the expression of pro-inflammatory cytokines and TNF-α, and dose-dependent decreased histamine release from mast cells [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. Mast cell infiltration is common in the inflammatory response to malignant osteosarcoma, which occurs mainly at the tumor margin and may lead to osteolysis and tumor invasion, but facilitates immunomodulatory therapy [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. It was also reported that mast cell accumulation in osteosarcoma is regulated by the CXCL6-CXCR2 axis [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. Therefore, disruption of WDR3 phase separation is closely associated with mast cell-mediated inflammatory responses in osteosarcoma inhibited by Nilotinib, thus affecting the mechanisms of metastasis and drug resistance in osteosarcoma.\u003c/p\u003e \u003cp\u003eThe regulation of LLPS in cells is a sophisticated mechanism involving not only crosstalk between intracellular components, but also influences by extracellular environmental factors such as temperature, ionic concentration, and pH [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e], which may further affect the phase-separation characteristics of proteins. In addition, the energetic state of cell including APT levels may also affect the kinetic changes of the protein LLPS [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. However, these factors that may affect the phase separation level of WDR3 were not discussed, which is one of the limitations of our study. Furthermore, the clinical use of Nilotinib involves cardiovascular adverse events [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e], but the cardiovascular toxicity of Nilotinib was neglected in this study. Therefore, subsequent studies are recommended to focus on exploring the intracellular mechanisms affecting WDR3 phase separation, as well as the resistance and toxicity of Nilotinib in osteosarcoma treatment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study proposed five LLPS-related biomarkers with the potential to predict the survival of osteosarcoma. Among them, WDR3 was a prognostic risk factor for osteosarcoma and bound stably to Nilotinib in molecular docking models. Functionally, WDR3 was significantly overexpressed in osteosarcoma cells, while its downregulation inhibited the malignant progression of osteosarcoma both \u003cem\u003ein vivo\u003c/em\u003e and \u003cem\u003ein vitro\u003c/em\u003e. In addition, WDR3 was found to form droplets, and its IDR mutation eliminated phase-separated levels of WDR3, thereby ameliorating the aggressive phenotype of osteosarcoma cells. Nilotinib may also reduce WDR3 phase-separated condensates and inhibit tumor growth and metastasis.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eANOVA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eone-way analysis of variance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eCCK-8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eCell counting kit-8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eCDF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003ecumulative distribution function\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eDAPI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e4\u0026apos;,6-diamidino-2-phenylindole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eDEGs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003edifferentially expressed genes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eECL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eenhanced chemiluminescence\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eFBS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003efetal bovine serum\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eFRAP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eFluorescence recovery after photobleaching\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eGEO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eGene Expression Omnibus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eIDR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eintrinsically disordered region\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eIHC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eImmunohistochemical\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eLASSO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eleast absolute shrinkage and selection operator\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eLLPS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eliquid-liquid phase separation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eMSI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003emicrosatellite instability\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ePCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eprincipal component analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eqPCR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eQuantitative real-time polymerase chain reaction\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eROC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003ereceiver operating characteristic\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eTdT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eTerminal deoxynucleotidyl transferase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eTUNEL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003eTdT-mediated dUTP nick-end labelling\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll animals were carried out under the approval of the Experimental Animal Ethics Committee of Yangzhou University (No.202312013).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by Shengjing Hospital of China Medical University Intra-Hospital Program (No. M0140).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMinglei Li\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e Conceptualization, Investigation, Data curation, Formal analysis, Writing - original draft. \u003cstrong\u003eNan Li\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e Conceptualization, Methodology, Data curation, Visualization. \u003cstrong\u003eYuying Fan:\u003c/strong\u003e Resources, Formal analysis, Software, Visualization, Writing - original draft. \u003cstrong\u003eZhan Zhang:\u003c/strong\u003e Resources, Software. \u003cstrong\u003eLong Zhou:\u003c/strong\u003e Investigation, Methodology. \u003cstrong\u003eYifan Yu:\u003c/strong\u003e Resources, Visualization. \u003cstrong\u003eMan Ni:\u003c/strong\u003e Data curation, Formal analysis. \u003cstrong\u003eMingzi Tan:\u0026nbsp;\u003c/strong\u003eConceptualization, Investigation, Project administration, Writing - review and editing. \u003cstrong\u003eWanJie Huang:\u0026nbsp;\u003c/strong\u003eConceptualization, Investigation, Project administration, Writing - review and editing. \u003cstrong\u003eTong Zhu:\u003c/strong\u003e Conceptualization, Investigation, Project administration, Writing - review and editing. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Minglei Li and Yuying Fan for conducting the mentionable literature research and subsequent data analysis. Thanks are due to Tong Zhu for their detailed revision guidance on the completed first draft of the writing. We would also like to thank The Shengjing Hospital of China Medical University for providing a research platform, which laid the foundation for the smooth progress of the project. In addition, we would like to thank Yangzhou University for providing the venue and platform for the animal experiments. Besides, we acknowledge partial support of Shengjing Hospital of China Medical University Intra-Hospital Program (No. M0140).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSun C, Li S, Ding J. Biomaterials-Boosted Immunotherapy for Osteosarcoma. Advanced healthcare materials. 2024;13(23):e2400864.\u003c/li\u003e\n\u003cli\u003eShi Q, Xu J, Chen C, et al. Direct contact between tumor cells and platelets initiates a FAK-dependent F3/TGF-\u0026beta; positive feedback loop that promotes tumor progression and EMT in osteosarcoma. Cancer letters. 2024;591(216902.\u003c/li\u003e\n\u003cli\u003eCascini C, Ratti C, Botti L, et al. Rewiring innate and adaptive immunity with TLR9 agonist to treat osteosarcoma. Journal of experimental \u0026amp; clinical cancer research : CR. 2023;42(1):154.\u003c/li\u003e\n\u003cli\u003eYin C, Chokkakula S, Li J, et al. Unveiling research trends in the prognosis of osteosarcoma: A bibliometric analysis from 2000 to 2022. Heliyon. 2024;10(6):e27566.\u003c/li\u003e\n\u003cli\u003eMohr A, Marques Da Costa ME, Fromigue O, et al. From biology to personalized medicine: Recent knowledge in osteosarcoma. European journal of medical genetics. 2024;69(104941.\u003c/li\u003e\n\u003cli\u003eYu S, Yao X. Advances on immunotherapy for osteosarcoma. Molecular cancer. 2024;23(1):192.\u003c/li\u003e\n\u003cli\u003eNerlakanti N, McGuire JJ, Bishop RT, et al. Histone deacetylase upregulation of neuropilin-1 in osteosarcoma is essential for pulmonary metastasis. Cancer letters. 2024;606(217302.\u003c/li\u003e\n\u003cli\u003eDi Patria L, Habel N, Olaso R, et al. C-terminal binding protein-2 triggers CYR61-induced metastatic dissemination of osteosarcoma in a non-hypoxic microenvironment. Journal of experimental \u0026amp; clinical cancer research : CR. 2025;44(1):83.\u003c/li\u003e\n\u003cli\u003eMosca N, Alessio N, Di Paola A, et al. Osteosarcoma in a ceRNET perspective. Journal of biomedical science. 2024;31(1):59.\u003c/li\u003e\n\u003cli\u003eWei X, Feng J, Chen L, et al. METTL3-mediated m6A modification of LINC00520 confers glycolysis and chemoresistance in osteosarcoma via suppressing ubiquitination of ENO1. Cancer letters. 2024;217194.\u003c/li\u003e\n\u003cli\u003eWang Y, Ma X, Xu E, et al. Identifying squalene epoxidase as a metabolic vulnerability in high-risk osteosarcoma using an artificial intelligence-derived prognostic index. Clinical and translational medicine. 2024;14(2):e1586.\u003c/li\u003e\n\u003cli\u003eZheng Z, Zeng Y, Bao X, et al. OTULIN confers cisplatin resistance in osteosarcoma by mediating GPX4 protein homeostasis to evade the mitochondrial apoptotic pathway. Journal of experimental \u0026amp; clinical cancer research : CR. 2024;43(1):330.\u003c/li\u003e\n\u003cli\u003eJi R, Wang Y, Pan D, et al. NUCB2 inhibition antagonizes osteosarcoma progression and promotes anti-tumor immunity through inactivating NUCKS1/CXCL8 axis. Cancer letters. 2024;591(216893.\u003c/li\u003e\n\u003cli\u003eGao W, Zhou J, Huang J, et al. Up-regulation of RAN by MYBL2 maintains osteosarcoma cancer stem-like cells population during heterogeneous tumor generation. Cancer letters. 2024;586(216708.\u003c/li\u003e\n\u003cli\u003eWu Y, Ma B, Liu C, Li D, Sui G. Pathological Involvement of Protein Phase Separation and Aggregation in Neurodegenerative Diseases. International journal of molecular sciences. 2024;25(18):\u003c/li\u003e\n\u003cli\u003eCheng Z, Wang H, Zhang Y, et al. Deciphering the role of liquid-liquid phase separation in sarcoma: Implications for pathogenesis and treatment. Cancer letters. 2025;616(217585.\u003c/li\u003e\n\u003cli\u003eLiu YT, Cao LY, Sun ZJ. The emerging roles of liquid-liquid phase separation in tumor immunity. International immunopharmacology. 2024;143(Pt 1):113212.\u003c/li\u003e\n\u003cli\u003ePu X, Zhang C, Jin J, et al. Phase separation of EEF1E1 promotes tumor stemness via PTEN/AKT-mediated DNA repair in hepatocellular carcinoma. Cancer letters. 2025;613(217508.\u003c/li\u003e\n\u003cli\u003eBhowmik D, Du M, Tian Y, et al. Cooperative DNA binding mediated by KicGAS/ORF52 oligomerization allows inhibition of DNA-induced phase separation and activation of cGAS. Nucleic acids research. 2021;49(16):9389-9403.\u003c/li\u003e\n\u003cli\u003eCable J, Brangwynne C, Seydoux G, et al. Phase separation in biology and disease-a symposium report. Annals of the New York Academy of Sciences. 2019;1452(1):3-11.\u003c/li\u003e\n\u003cli\u003eBanani SF, Lee HO, Hyman AA, Rosen MK. Biomolecular condensates: organizers of cellular biochemistry. Nature reviews Molecular cell biology. 2017;18(5):285-298.\u003c/li\u003e\n\u003cli\u003eChakraborty S, Nandi P, Mishra J, et al. Molecular mechanisms in regulation of autophagy and apoptosis in view of epigenetic regulation of genes and involvement of liquid-liquid phase separation. Cancer letters. 2024;587(216779.\u003c/li\u003e\n\u003cli\u003eLu B, Zou C, Yang M, et al. Pharmacological Inhibition of Core Regulatory Circuitry Liquid-liquid Phase Separation Suppresses Metastasis and Chemoresistance in Osteosarcoma. Advanced science (Weinheim, Baden-Wurttemberg, Germany). 2021;8(20):e2101895.\u003c/li\u003e\n\u003cli\u003eChen D, Zhao Z, Huang Z, et al. Super enhancer inhibitors suppress MYC driven transcriptional amplification and tumor progression in osteosarcoma. Bone research. 2018;6(11.\u003c/li\u003e\n\u003cli\u003eKim YR, Joo J, Lee HJ, et al. Prion-like domain mediated phase separation of ARID1A promotes oncogenic potential of Ewing\u0026apos;s sarcoma. Nature communications. 2024;15(1):6569.\u003c/li\u003e\n\u003cli\u003eSun J, Chen Y, Bi R, Yuan Y, Yu H. Bioinformatic approaches of liquid-liquid phase separation in human disease. Chinese medical journal. 2024;137(16):1912-1925.\u003c/li\u003e\n\u003cli\u003eWang J, Yu H, Ma Q, et al. Phase separation of OCT4 controls TAD reorganization to promote cell fate transitions. Cell stem cell. 2021;28(10):1868-1883.e1811.\u003c/li\u003e\n\u003cli\u003eMolliex A, Temirov J, Lee J, et al. Phase separation by low complexity domains promotes stress granule assembly and drives pathological fibrillization. Cell. 2015;163(1):123-133.\u003c/li\u003e\n\u003cli\u003eKobayashi M, Jitoku D, Iwayama Y, et al. Association studies of WD repeat domain 3 and chitobiosyldiphosphodolichol beta-mannosyltransferase genes with schizophrenia in a Japanese population. PloS one. 2018;13(1):e0190991.\u003c/li\u003e\n\u003cli\u003eClaudio JO, Liew CC, Ma J, et al. Cloning and expression analysis of a novel WD repeat gene, WDR3, mapping to 1p12-p13. Genomics. 1999;59(1):85-89.\u003c/li\u003e\n\u003cli\u003eSu W, Zhu S, Chen K, et al. Overexpressed WDR3 induces the activation of Hippo pathway by interacting with GATA4 in pancreatic cancer. Journal of experimental \u0026amp; clinical cancer research : CR. 2021;40(1):88.\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-Quispes WA, Pastor S, Galofr\u0026eacute; P, et al. Possible role of the WDR3 gene on genome stability in thyroid cancer patients. PloS one. 2012;7(9):e44288.\u003c/li\u003e\n\u003cli\u003eBaida A, Akdi M, Gonz\u0026aacute;lez-Flores E, et al. Strong association of chromosome 1p12 loci with thyroid cancer susceptibility. Cancer epidemiology, biomarkers \u0026amp; prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2008;17(6):1499-1504.\u003c/li\u003e\n\u003cli\u003eGong L, Yu L, Gong X, et al. Exploration of anti-inflammatory mechanism of forsythiaside A and forsythiaside B in CuSO(4)-induced inflammation in zebrafish by metabolomic and proteomic analyses. Journal of neuroinflammation. 2020;17(1):173.\u003c/li\u003e\n\u003cli\u003eMcMahon M, Ayll\u0026oacute;n V, Panov KI, O\u0026apos;Connor R. Ribosomal 18 S RNA processing by the IGF-I-responsive WDR3 protein is integrated with p53 function in cancer cell proliferation. The Journal of biological chemistry. 2010;285(24):18309-18318.\u003c/li\u003e\n\u003cli\u003eLiu W, Xie A, Xiong J, et al. WDR3 promotes stem cell-like properties in prostate cancer by inhibiting USF2-mediated transcription of RASSF1A. The journal of gene medicine. 2023;25(7):e3498.\u003c/li\u003e\n\u003cli\u003eXu H, Zhan W, Chen Z. Ras-Association Domain Family 1 Isoform A (RASSF1A) Gene Polymorphism rs1989839 is Associated with Risk and Metastatic Potential of Osteosarcoma in Young Chinese Individuals: A Multi-Center, Case-Control Study. Medical science monitor : international medical journal of experimental and clinical research. 2016;22(4529-4535.\u003c/li\u003e\n\u003cli\u003eWang WG, Chen SJ, He JS, Li JS, Zang XF. The tumor suppressive role of RASSF1A in osteosarcoma through the Wnt signaling pathway. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine. 2016;37(7):8869-8877.\u003c/li\u003e\n\u003cli\u003eShi Y, Wang Z, Zhang J, et al. Small Nuclear Ribonucleoprotein Polypeptides B and B1 Promote Osteosarcoma Progression via Activating the Ataxia-Telangiectasia Mutated Signaling Pathway through Ribonucleotide Reductase Subunit M2. The American journal of pathology. 2024;194(11):2163-2178.\u003c/li\u003e\n\u003cli\u003eLiu CW, Huang JH, Chang HH, et al. C1GALT1 expression predicts poor survival in osteosarcoma and is crucial for ABCC1 transporter-mediated doxorubicin resistance. The Journal of pathology. 2025;\u003c/li\u003e\n\u003cli\u003eJabbour E, Kantarjian H. Chronic myeloid leukemia: 2025 update on diagnosis, therapy, and monitoring. American journal of hematology. 2024;99(11):2191-2212.\u003c/li\u003e\n\u003cli\u003eZhao Z, Zhang J, Zhang W, et al. Efficacy evaluation of nilotinib treatment in different genomic subtypes of gastrointestinal stromal tumors: A meta-analysis and systematic review. Current problems in cancer. 2021;45(3):100705.\u003c/li\u003e\n\u003cli\u003eWang S, Xie Y, Bao A, et al. Nilotinib, a Discoidin domain receptor 1 (DDR1) inhibitor, induces apoptosis and inhibits migration in breast cancer. Neoplasma. 2021;68(5):975-982.\u003c/li\u003e\n\u003cli\u003eDong H, Wen C, He L, et al. Nilotinib boosts the efficacy of anti-PDL1 therapy in colorectal cancer by restoring the expression of MHC-I. Journal of translational medicine. 2024;22(1):769.\u003c/li\u003e\n\u003cli\u003eWu Z, Yu J, Han T, et al. System analysis based on Anoikis-related genes identifies MAPK1 as a novel therapy target for osteosarcoma with neoadjuvant chemotherapy. BMC musculoskeletal disorders. 2024;25(1):437.\u003c/li\u003e\n\u003cli\u003eWei C, Li M, Li X, Lyu J, Zhu X. Phase Separation: \u0026quot;The Master Key\u0026quot; to Deciphering the Physiological and Pathological Functions of Cells. Advanced biology. 2022;6(7):e2200006.\u003c/li\u003e\n\u003cli\u003eEl-Agamy DS. Anti-allergic effects of nilotinib on mast cell-mediated anaphylaxis like reactions. European journal of pharmacology. 2012;680(1-3):115-121.\u003c/li\u003e\n\u003cli\u003eInagaki Y, Hookway E, Williams KA, et al. Dendritic and mast cell involvement in the inflammatory response to primary malignant bone tumours. Clinical sarcoma research. 2016;6(13.\u003c/li\u003e\n\u003cli\u003eWang C, Lei Z, Zhang C, Hu X. CXCL6-CXCR2 axis-mediated PD-L2(+) mast cell accumulation shapes the immunosuppressive microenvironment in osteosarcoma. Heliyon. 2024;10(14):e34290.\u003c/li\u003e\n\u003cli\u003eXu WX, Qu Q, Zhuang HH, et al. The Burgeoning Significance of Liquid-Liquid Phase Separation in the Pathogenesis and Therapeutics of Cancers. International journal of biological sciences. 2024;20(5):1652-1668.\u003c/li\u003e\n\u003cli\u003eKang J, Lim L, Song J. ATP enhances at low concentrations but dissolves at high concentrations liquid-liquid phase separation (LLPS) of ALS/FTD-causing FUS. Biochemical and biophysical research communications. 2018;504(2):545-551.\u003c/li\u003e\n\u003cli\u003eWang Z, Jiang L, Yan H, Xu Z, Luo P. Adverse events associated with nilotinib in chronic myeloid leukemia: mechanisms and management strategies. Expert review of clinical pharmacology. 2021;14(4):445-456.\u003c/li\u003e\n\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":"journal-of-experimental-and-clinical-cancer-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jecc","sideBox":"Learn more about [Journal of Experimental \u0026 Clinical Cancer Research](http://jeccr.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jecc/default.aspx","title":"Journal of Experimental \u0026 Clinical Cancer Research","twitterHandle":"@OncoBioMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Osteosarcoma, liquid-liquid phase separation, prognostic biomarker, WDR3, Nilotinib, IDR mutation","lastPublishedDoi":"10.21203/rs.3.rs-6611672/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6611672/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eOsteosarcoma is highly invasive with a poor prognosis. The phenomenon of liquid-liquid phase separation (LLPS) can promote the formation of biomolecules and participate in the tumor regulation mechanism. Therefore, mining prognostic markers related to LLPS could allow patients to benefit from targeted therapies.\u003c/p\u003e\u003ch2\u003eMethod\u003c/h2\u003e \u003cp\u003eMicroarray analysis was performed to identify LLPS-related biomarkers, followed by validation via molecular docking analysis. Functions of key genes were investigated in U2-OS cells and xenograft mice. LLPS of the key gene were observed by the droplet formation assay and fluorescence recovery after photobleaching. The intrinsically disordered region (IDR) was predicted and mutated to disrupt LLPS, which was rescued by the fusion of hnRNAP1 IDR. Therapeutic mechanism of Nilotinib mediated by LLPS was explored \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eFive LLPS-related biomarkers were screened by bioinformatics analyses to predict the osteosarcoma prognosis. These prognostic signatures were significantly associated with immune cell infiltration, tumor immune escape and drug sensitivity. Among them, WDR3 was a prognostic risk factor for osteosarcoma and stably bound to Nilotinib in molecular docking models. In transfected U2-OS cells and xenograft mice, downregulation of WDR3 significantly inhibited malignant progression of osteosarcoma. More importantly, WDR3 could form droplets in U2-OS cells and restore the fluorescence intensity of WDR3 condensates with liquid-like behavior after photobleaching. The mutation in IDR could disrupt the phase separation ability of WDR3, whereas the fusion of hnRNAP1 IDR rescued the phase separation abnormality caused by WDR3 mutation. Moreover, treatment with Nilotinib improved the progression of osteosarcoma \u003cem\u003ein vivo\u003c/em\u003e and \u003cem\u003ein vitro\u003c/em\u003e, while inhibiting the production of WDR3 phase separated condensates.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eWDR3 phase separation involves in the therapeutic mechanism of Nilotinib against osteosarcoma, and thus may serve as a potent biomarker to ameliorate adverse events after osteosarcoma treatment.\u003c/p\u003e","manuscriptTitle":"WDR3 undergoes phase separation to mediate the therapeutic mechanism of Nilotinib against osteosarcoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-15 10:23:05","doi":"10.21203/rs.3.rs-6611672/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-05-29T04:40:42+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-05-29T00:45:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-05-28T07:31:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"328496030306840856642955391809937721838","date":"2025-05-20T06:15:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"84706535270687085074013965383955276722","date":"2025-05-18T21:37:28+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-05-08T16:55:52+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-08T16:35:35+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-08T16:34:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Experimental \u0026 Clinical Cancer Research","date":"2025-05-07T11:36:27+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-experimental-and-clinical-cancer-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jecc","sideBox":"Learn more about [Journal of Experimental \u0026 Clinical Cancer Research](http://jeccr.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jecc/default.aspx","title":"Journal of Experimental \u0026 Clinical Cancer Research","twitterHandle":"@OncoBioMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5912f796-a785-49ca-b1ac-0c6b87fb4932","owner":[],"postedDate":"May 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-07-14T16:06:33+00:00","versionOfRecord":{"articleIdentity":"rs-6611672","link":"https://doi.org/10.1186/s13046-025-03456-x","journal":{"identity":"journal-of-experimental-and-clinical-cancer-research","isVorOnly":false,"title":"Journal of Experimental \u0026 Clinical Cancer Research"},"publishedOn":"2025-07-11 15:57:03","publishedOnDateReadable":"July 11th, 2025"},"versionCreatedAt":"2025-05-15 10:23:05","video":"","vorDoi":"10.1186/s13046-025-03456-x","vorDoiUrl":"https://doi.org/10.1186/s13046-025-03456-x","workflowStages":[]},"version":"v1","identity":"rs-6611672","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6611672","identity":"rs-6611672","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}