The present study collected tumor and paired normal bone tissues from 6 patients with OS at The Second Hospital of Lanzhou University (Lanzhou, China) between October 2021 and December 2023. The cohort consisted of 6 patients (5 men; 1 woman), aged 14–21 years, with pathological diagnoses of conventional OS (n=5) and parosteal OS (n=1). The inclusion criteria were: i) Histopathologically confirmed diagnosis of conventional or parosteal OS; ii) aged between 10 and 25 years; iii) no prior neoadjuvant chemotherapy or radiotherapy before surgery; iv) availability of both tumor and paired normal bone tissue. The exclusion criteria were: i) Recurrent or metastatic OS at diagnosis; ii) history of other malignancies; iii) insufficient tissue quality or quantity for downstream analysis. All samples were immediately frozen in liquid nitrogen after surgery and stored at −80°C. The study was approved by The Second Hospital of Lanzhou University ethics committee (approval no. 2024A-504), and written informed consent was obtained from all patients or their guardians.

The human OS cell lines MG63 and 143B, along with the normal osteoblast cell line hFOB1.19, were purchased from Shanghai Yilei information Technology Co., Ltd. The cells were cultured in high-glucose Dulbecco's Modified Eagle's Medium (HyClone™; Cytiva) supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.). MG63 and 143B cells were incubated at 37°C, whereas hFOB1.19 cells were maintained at 34°C.

FBXL5-targeting short hairpin RNA (shRNA) lentiviral constructs (cat. no. GXDL0418586; FBXL5 Human GV493) were obtained from Shanghai GeneChem Co., Ltd. The target sequences were as follows: FBXL5-RNAi (118431–1): GTACAGGAACAGCTTTAAGAA; FBXL5-RNAi (118432–1): CCTGATGATGAATGGGTGAAA; FBXL5-RNAi (118433–1): GCACTTTACTCATAGCACATT. A second-generation lentiviral packaging system was used, including the shuttle vector GV493 (hU6-MCS-CBh-gcGFP-IRES-puromycin) and packaging plasmids psPAX2 and pMD2.G. Lentiviral particles were produced in 293T cells (Shanghai GeneChem Co., Ltd.) according to the manufacturer's protocol. Plasmids were transfected at a ratio of 20 (GV493), 15 (psPAX2) and 10 µg (pMD2.G), followed by incubation at 37°C with 5% CO₂ for 48 h. Viral supernatants were collected and concentrated by ultracentrifugation (13,680 × g; 2 h; 4°C). For transduction, 143B and MG63 cells were infected with lentivirus at multiplicities of infection (MOI) of 15 and 30, respectively, in the presence of polybrene (8 µg/ml). After 48 h, stable clones were selected using puromycin (2 µg/ml), and selection was maintained for 7 days to establish stable knockdown pools. Knockdown efficiency was confirmed by western blotting (WB) analysis and reverse transcription-quantitative PCR (RT-qPCR). All subsequent experiments were performed 10 days after the initial transduction.

Total RNA was extracted from cells using Trizol (total RNA extraction reagent; cat. no. R0016; Beyotime Biotechnology). RT was performed using the Evo M-MLV One-step RT-PCR kit (with gDNA removal, for qPCR; cat. no. AG11705; AGbio) according to the manufacturer's protocol. The reaction conditions for RT were: 37°C for 15 min (gDNA removal), 42°C for 15 min (RT), followed by 85°C for 5 sec (enzyme inactivation). For qPCR, the reaction was performed using SYBR Green Master Mix (cat. no. AG11732; AGbio), which contains SYBR Green as the fluorophore. The following thermocycling conditions were applied: initial denaturation at 95°C for 30 sec, followed by 40 cycles of 95°C for 5 sec and 60°C for 30 sec. GAPDH was used as the reference gene. The primer sequences used are provided in Table I. Relative expression levels were calculated using the 2^−ΔΔCq method (18).

Total proteins were extracted from cells using RIPA lysis buffer (cat. no. P0013B; Beyotime Biotechnology) and quantified by the BCA method. Proteins (30 µg/lane) were separated by 10% SDS-PAGE and electrotransferred to a PVDF membrane (MilliporeSigma). After blocking with 5% non-fat milk (cat. no. MA0406; Dalian Meilun Biology Technology Co., Ltd.cat. no.) at 4°C for 1 h, the membrane was incubated with primary antibodies at 4°C overnight. The primary antibodies and their cat. nos. were as follows: FBXL5 (1:1,000; cat. no. BD-PT6983; Biodragon Immunotechnologies Co., Ltd.), PCNA (1:2,000; cat. no. 10205-2-AP; Proteintech Group, Inc.), CDK4 (1:1,000; cat. no. 11026-1-AP Proteintech Group, Inc.), cyclin D1 (1:1,000; cat. no. 60186-1-Ig; Proteintech Group, Inc.), Bcl-2 (1:1,000; cat. no. AF6139; Affinity Biosciences, Ltd.), Bax (1:1,000; cat. no. AF0120; Affinity Biosciences, Ltd.), cleaved caspase-3 (1:1,000; cat. no. AF7022; Affinity Biosciences, Ltd.), total caspase-3 (1:1,000; cat. no. db12058; Hangzhou Diagbio Technology Co., Ltd.), MMP2 (1:1,000; cat. no. RM8377; Biodragon Immunotechnologies Co., Ltd.), MMP9 (1:1,000; cat. no. RM3763; Biodragon Immunotechnologies Co., Ltd.), MEK1/2 (1:1,000; HUABIO; cat. no. ET1602-3), ERK (1:1,000; cat. no. ET1601-29; HUABIO), Ras (1:1,000; cat. no. RM7373; Biodragon Immunotechnologies Co., Ltd.), Raf (1:1,000; cat. no. RM4242; Biodragon Immunotechnologies Co., Ltd.), and β-actin (1:8,000; cat. no. HA722023; HUABIO). After washing, the membrane was incubated with horseradish peroxidase-conjugated goat anti-rabbit/mouse IgG (1:10,000; cat. no. ZB-2306; Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.) at 4°C for 1 h. Proteins were detected using enhanced chemiluminescence substrate (cat. no. G2020-500ML; Wuhan Servicebio Technology Co., Ltd.). Densitometric analysis was performed using ImageJ software (version 1.52a; National Institutes of Health).

143B and MG63 cells were seeded on coverslips, fixed with 4% paraformaldehyde in PBS at room temperature for 15 min, permeabilized with 0.1% Triton X-100 in PBS for 10 min at room temperature and blocked with 5% normal goat serum (cat. no. SL038; Beijing Solarbio Science & Technology Co., Ltd.) in PBS at room temperature for 1 h. After blocking, cells were incubated overnight at 4°C with the following primary antibodies: FBXL5 (1:100; Santa Cruz Biotechnology, Inc.; cat. no. sc-390102), PCNA (1:100; cat. no. 10205-2-AP; Proteintech Group, Inc.cat. no.), CDK4 (1:100; cat. no. 11026-1-AP; Proteintech Group, Inc.), and cyclin D1 (1:100; cat. no. 60186-1-Ig; Proteintech Group, Inc.cat. no.). After washing, cells were incubated with Alexa Fluor 488-conjugated secondary antibody (1:200; Biodragon Immunotechnologies Co., Ltd.; cat. no. BD9010) at room temperature for 1 h. Nuclei were counterstained with DAPI (1 µg/ml) at room temperature for 5 min. Fluorescence images were captured using a fluorescence microscope.

Proliferation: Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay (cat. no. C0038; Beyotime Biotechnology). Cells were seeded in 96-well plates and cultured for 24, 48 and 72 h. At each time point, CCK-8 reagent was added and incubated for 2 h at 37°C, and the absorbance was measured at 450 nm using a microplate reader. Proliferation rates were also determined using 5-ethynyl-2′-deoxyuridine (EdU) staining with the EdU-488 Cell Proliferation Kit (cat. no. C0071S; Beyotime Biotechnology) according to the manufacturer's instructions. Briefly, cells were incubated with EdU at 37°C for 2 h, then fixed with 4% paraformaldehyde in PBS at room temperature for 15 min, permeabilized with 0.3% Triton X-100 in PBS and stained with the Click-It reaction mixture. Nuclei were counterstained with DAPI (cat. no. C1006; Beyotime Biotechnology). Images were captured using a fluorescence microscope.

Apoptosis: Apoptotic cells were identified using the TUNEL staining. Cells were fixed with 4% paraformaldehyde in PBS at room temperature for 15 min, permeabilized with 0.1% Triton X-100 in PBS for 5 min and then incubated with TUNEL reaction mixture (cat. no. C1088; Beyotime Biotechnology) at 37°C for 60 min according to the manufacturer's protocol. Nuclei were counterstained with DAPI (1 µg/ml) at room temperature for 5 min. After rinsing with PBS, the slides were mounted using anti-fade mounting medium (Beyotime Biotechnology China; cat. no. P0126). At least five random fields of view per sample were observed under a fluorescence microscope, and TUNEL-positive cells were counted.

Migration: A wound healing assay was performed to measure cell migration. No substrate coating was used. 143B and MG63 cells were seeded in 6-well plates and cultured until reaching 90–100% confluence. A linear scratch was created across the cell monolayer using a sterile 200 µl pipette tip. Debris was removed by washing with PBS, and cells were then cultured in Gibco (Thermo Fisher Scientific, Inc.) basal medium to exclude the effect of proliferation. Wound closure was monitored and imaged at 0, 24, 48, 72 and 96 h under a light microscope. The wound area was measured using ImageJ software (version 1.52a, National Institutes of Health), and the percentage of wound closure was calculated as: [(area at 0 h-area at time point)/area at 0 h] ×100%. No drugs were used in this assay.

A total of 36 BALB/c nude mice (4–6 weeks old; female; weighing 16–18 g) were obtained from Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences. Mice were housed under specific pathogen-free conditions with temperature 22±2°C, humidity 50±10%, a 12 h light/dark cycle, and ad libitum access to food and water. Mice were randomly assigned to six groups (n=6 per group): Group 1: 143B + control; group 2: 143B + sh-NC; group 3: 143B + sh-FBXL5; group 4: MG63 + control; group 5: MG63 + sh-NC; group 6: MG63 + sh-FBXL5. A xenograft tumor model was established by subcutaneously injecting 2×10⁶ cells into the right axilla of each mouse. Tumor volume was measured every 3 days using a caliper and calculated as volume=(length × width²)/2. The ethical endpoint was defined as tumor volume >1,500 mm³, tumor ulceration, substantial weight loss (>20% of initial body weight) or signs of pain/distress. Tumor volume and body weight were measured every 3 days from day 0 until the endpoint for each group. The 143B xenografts grew more rapidly and reached the ethical endpoint (tumor volume >1,500 mm³) by day 30, whereas the MG63 ×enografts grew more slowly and did not reach the endpoint until day 34. Therefore, mice in the 143B and MG63 groups were euthanized on day 30 and day 34, respectively, to prevent exceeding the predefined ethical endpoints. No mice were euthanized before the scheduled endpoints. Mice in the 143B and MG63 groups were euthanized by CO₂ inhalation (30–50% vol/min) on day 30 and day 34, respectively. The tumor tissues were then processed for subsequent analysis. The animal study was approved by the Animal Ethics Committee of Lanzhou University Second Hospital (approval no. D2024-417).

Total RNA was extracted from cells using TRIzol reagent (cat. no. R0016; Beyotime Biotechnology). RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA). Sequencing libraries were prepared using the TruSeq RNA Sample Prep Kit v2 (Illumina, Inc.; cat. no. RS-122-2001) according to the manufacturer's protocol. The libraries were sequenced on the Illumina NovaSeq 6000 platform (Illumina, Inc.) using the NovaSeq 6000 S4 Reagent Kit v1.5 (300 cycles; Illumina, Inc.; cat. no. 20028312) with 150 bp paired-end sequencing. The final library was loaded at a concentration of 1.8 pM, as measured by Qubit 2.0 Fluorometer (Thermo Fisher Scientific, Inc.). For data analysis, Trimmomatic (version 0.39; http://www.usadellab.org/cms/index.php?page=trimmomatic) was used for quality control, HISAT2 (version 2.2.1; http://ccb.jhu.edu/software/hisat2/) was used for alignment to the reference genome, and featureCounts (version 2.0.1; http://subread.sourceforge.net/) was used to quantify gene expression levels. edgeR (version 3.40.0) was applied to identify differentially expressed genes (DEGs) with criteria of |log2FC|≥1 and FDR ≤0.05. Functional enrichment analysis was conducted using GOATools (version 1.0.0; http://github.com/tanghaibao/GOatools) for Gene Ontology (GO) enrichment and KOBAS (version 3.0; http://bioinfo.org/kobas3/) for KEGG pathway enrichment. DAVID (Database for Annotation, Visualization and Integrated Discovery; http://david.ncifcrf.gov) and GSEA (Gene Set Enrichment Analysis; http://www.gsea-msigdb.org/gsea/index.jsp) were also used for functional annotation and enrichment analysis.

After deparaffinization in xylene and rehydration through a graded ethanol series, antigen retrieval was performed by heating the paraffin-embedded tissue sections in citrate buffer (pH 6.0) at 95°C for 15 min. Sections were incubated with primary antibodies against Raf (1:100; cat. no. RM4242; Suzhou Botron Immunotherapy Technology Co., Ltd.) and MEK1/2 (1:100; HUABIO; cat. no. ET1602-3) at 4°C overnight. After washing, sections were incubated with Alexa Fluor 488-conjugated secondary antibody (1:200; Biodragon Immunotechnologies Co., Ltd.; cat. no. BD9010) at room temperature for 1 h. Nuclei were counterstained with DAPI (1 µg/ml) at room temperature for 10 min. Fluorescence images were captured using a fluorescence microscope.

Paraffin-embedded tissue sections were deparaffinized in xylene and rehydrated through a graded ethanol series (100, 95, 70 and 50%) for 2 min each, followed by rinsing in distilled water. Sections were stained with hematoxylin at room temperature for 5 min, followed by eosin staining at room temperature for 2 min. Sections were then dehydrated through a graded ethanol series (70, 95 and 100%), cleared in xylene, and mounted with a coverslip using a resinous mounting medium. Images were captured using a light microscope (Olympus Corporation) and analyzed using ImageJ software (version 1.52a; National Institutes of Health).

All experimental data were statistically analyzed using GraphPad Prism 8.0.2 software (GraphPad; Dotmatics). A paired Student's t-test was used when comparing two measurements from the same subject, whereas an unpaired Student's t-test was used when comparing data from different groups of subjects. For multiple groups, one-way ANOVA was used, followed by Tukey's multiple comparisons test. Data are expressed as the mean ± standard deviation from six independent repeats. Fluorescence images and WB bands were analyzed and measured using ImageJ 1.52a software (National Institutes of Health). P<0.05 was considered to indicate a statistically significant difference.

qPCR and WB analyses revealed that FBXL5 protein (Fig. 1A) and mRNA (Fig. 1B) expression levels were significantly elevated in OS tissues compared to normal tissues. In OS cell lines 143B and MG63, FBXL5 mRNA (Fig. 1D and F) and protein (Fig. 1C and E) expression was also significantly upregulated compared with the normal osteoblast cell line hFOB1.19.

Lentivirus-mediated RNA interference successfully established stable FBXL5 knockdown cell lines, with infection rates exceeding 99% in 143B (Fig. 1G) and MG63 (Fig. 1L) cells. qPCR results showed significantly lower FBXL5 mRNA levels in the sh-FBXL5 group compared with the control and NC groups in both 143B (Fig. 1H) and MG63 (Fig. 1M) cells. WB analysis further confirmed a marked reduction in FBXL5 protein expression in the sh-FBXL5-1 group of 143B (Fig. 1I and J) and MG63 (Fig. 1N and O) cells compared with the control group. Immunofluorescence analysis also demonstrated significantly decreased FBXL5 expression in the sh-FBXL5-1 group in both 143B (Fig. 1K) and MG63 (Fig. 1P) cells compared with the control and NC groups.

These results demonstrate that FBXL5 is highly expressed in OS tissues and cell lines, providing experimental support for investigating its functional role in OS. Lentivirus-mediated RNA interference successfully knocked down FBXL5 expression, with the sh-FBXL5-1 sequence reducing both mRNA and protein expression levels. Therefore, subsequent experiments were conducted using the sh-FBXL5-1 lentivirus.

WB analysis and immunofluorescence were used to detect changes in expression of the proliferation-related proteins PCNA, CDK4 and cyclin D1 after FBXL5 knockdown, with EdU and CCK-8 assays assessing the impact on cell proliferation.

WB analysis showed a significant downregulation of PCNA, CDK4 and cyclin D1 protein expression levels in the sh-FBXL5 group compared with the control and NC groups in 143B (Fig. 2A-C) and MG63 (Fig. 2I-K) cells. Immunofluorescence analysis further confirmed a significant reduction in the mean fluorescence intensity of these proteins in the sh-FBXL5 group in both 143B (Fig. 2D-F) and MG63 (Fig. 2L-N) cells.

The EdU assay revealed a significant decrease in the number and proportion of EdU-positive cells in the sh-FBXL5 group in 143B (Fig. 2G) and MG63 (Fig. 2O) cells compared to the control and NC groups.

CCK-8 proliferation assays showed that sh-FBXL5 significantly inhibited the proliferation of 143B (Fig. 2H) and MG63 (Fig. 2P) cells, with notably lower optical density values compared to controls starting at 24 h; no significant differences were observed between the control and NC groups. These findings suggest that FBXL5 serves a key role in promoting OS cell proliferation, and its depletion suppresses tumor cell growth through downregulation of PCNA, CDK4 and cyclin D1.

WB analysis was used to analyze the expression changes of apoptosis-related proteins following FBXL5 knockdown, while the TUNEL assay assessed cell apoptosis.

The results showed that in the sh-FBXL5 group, expression of the anti-apoptotic protein Bcl-2 was significantly reduced, while the pro-apoptotic proteins BAX and caspase-3 were significantly increased in both 143B (Fig. 3A-C) and MG63 (Fig. 3E-G) cells. TUNEL staining revealed nuclear fragmentation and morphological abnormalities in the sh-FBXL5 group, with intense red fluorescence signals (Fig. 3D and H). The proportion of TUNEL-positive cells was significantly higher in the sh-FBXL5 group compared with the control and NC groups, whereas there was no statistical difference observed between the control and NC groups.

This indicates that FBXL5 knockdown induces apoptosis in OS cells, suggesting that FBXL5 may function as an anti-apoptotic factor in osteosarcoma by upregulating Bcl-2 and downregulating BAX and caspase-3.

WB analysis was performed to detect changes in the expression of migration and invasion markers MMP2 and MMP9 following FBXL5 knockdown, and a wound healing assay was used to evaluate cell migration ability.

The results showed that MMP2 and MMP9 protein expression were significantly reduced in both 143B (Fig. 4A and B) and MG63 (Fig. 4E and F) cells in the sh-FBXL5 group compared with both the control and NC groups. The wound healing assay demonstrated that 143B (Fig. 4C) and MG63 (Fig. 4G) cells in the control and NC groups migrated markedly more quickly when compared with the sh-FBXL5 group; wound healing was significantly delayed at all time points in the sh-FBXL5 group, with incomplete wound closure at 96 h. The migration distance curve revealed that the migration distance of 143B (Fig. 4D) and MG63 (Fig. 4H) cells in the sh-FBXL5 group was significantly lower than that of the control group at all time points from 24 to 96 h.

These results imply that FBXL5 is involved in regulating OS cell migration and invasion, likely through the downregulation of MMP2 and MMP9, and targeting FBXL5 may limit metastatic potential.

A subcutaneous tumor implantation model in nude mice was used to systematically assess the impact of FBXL5 knockdown on tumor growth, thereby investigating the role of FBXL5 in OS progression.

i) Body weight changes. In the sh-FBXL5 group, the body weight of nude mice in 143B (Fig. 5B) and MG63 (Fig. 5G) xenograft models remained stable, with a slight increase. By contrast, mice in the control and NC groups showed a slight weight increase during the first 10 days, followed by a continuous decline from day 14 onward.

ii) Tumor volume. Tumor formation was observed in all groups by day 10 post-implantation. Tumor volumes of 143B (Fig. 5D) and MG63 (Fig. 5I) in the sh-FBXL5 group were significantly smaller than those in the control group, with a marked reduction in growth rate at each time point.

iii) Tumor weight and morphology. At the study endpoint, gross tumor images showed that tumors in both the control and NC groups exhibited tight adhesion to the ribcage, whereas the sh-FBXL5 group displayed well-encapsulated tumors with no apparent invasion in both the 143B and MG63 ×enograft models (Fig. 5A and F). Tumor weights and volumes of 143B (Fig. 5C and E) and MG63 (Fig. 5H and J) in the sh-FBXL5 group were significantly lower than those in the control group.

Immunofluorescence analysis confirmed that in both the 143B (Fig. 6A and B) and MG63 (Fig. 6D and E) tumor models, FBXL5 expression was significantly reduced in the sh-FBXL5 group compared to the control and NC groups, confirming that lentivirus-mediated FBXL5 knockdown remained stable in vivo.

Hematoxylin and eosin staining showed that 143B (Fig. 6C) and MG63 (Fig. 6F) tumors in the control and NC groups exhibited characteristic malignant features, including disorganized cell arrangement, nest-like or diffuse distribution, variable morphology, uneven nuclear staining, frequent mitotic figures, marked atypia, unclear tissue boundaries, and extensive areas of necrosis and neovascularization. By contrast, tumors in the sh-FBXL5 group displayed more organized cell arrangement, relatively uniform morphology, even nuclear staining, fewer mitotic figures, reduced atypia, clear tissue boundaries, and significantly less necrosis and neovascularization.

Histological analysis confirmed that FBXL5 knockdown reduces malignant features of OS tumors in vivo, including disorganized cell arrangement, necrosis and neovascularization, further supporting its tumor-suppressive role.

RNA-seq was employed to analyze the transcriptomic changes in 143B and MG63 cell lines following FBXL5 knockdown to further investigate the molecular mechanisms through which FBXL5 knockdown inhibits OS progression.

Following transcripts per million normalization, the gene expression distributions in 143B (Fig. 7A) and MG63 (Fig. 7G) cells exhibited normal distributions with high overlap, indicating high data quality. Volcano plot analysis revealed that, following FBXL5 knockdown, 3,039 genes were upregulated and 2,472 genes were downregulated in 143B cells (Fig. 7B), while in MG63 cells, 2,740 genes were upregulated and 1,776 genes were downregulated (Fig. 7H) (P<0.05, |log₂FC|>1). These GO enrichment results indicate that FBXL5 primarily affects metabolic and proliferation-related biological processes in OS cells.

GO enrichment analysis revealed that the differentially expressed genes were primarily enriched in biological processes such as ‘metabolic process’, ‘cellular process’, ‘immune system process’ and ‘growth’. In the cellular component category, the genes were mainly associated with ‘protein complex’ and ‘intracellular anatomical structure’, while in the molecular function category, they were enriched in ‘catalytic activity’ and ‘binding’ (Fig. 7C and I). Enrichment ratio analysis indicated that metabolism- and proliferation-related pathways were the most prominent (Fig. 7D and J), suggesting that FBXL5 regulates OS progression by modulating these pathways.

KEGG enrichment analysis revealed that the differentially expressed genes were significantly enriched in the ‘MAPK signaling pathway-plant’ (Fig. 7E and K). In the sh-FBXL5 cells, the expression levels of proliferation-related genes (PCNA, cyclin D1 and CDK4) were significantly downregulated, and alterations in the expression of MAPK pathway members [MAPK8, dual specificity phosphatase 2 (DUSP2)] were observed. Notably, the downregulation of DUSP2 may be associated with altered ERK activity and suppressed proliferative signaling; however, the direct regulatory relationship requires further investigation (Fig. 7F and L).

KEGG pathway analysis identified the MAPK signaling pathway as significantly enriched following FBXL5 knockdown, highlighting this pathway as a key mediator of FBXL5′s oncogenic functions in OS (Fig. 7E and K).

WB analysis showed that after FBXL5 knockdown, the expression levels of Ras, Raf, MEK1/2 and ERK1/2 proteins were significantly reduced in 143B (Fig. 8A-D) and MG63 (Fig. 8I-L) cells. Immunofluorescence further confirmed that the expression of MEK1/2 and ERK1/2 was markedly decreased in 143B (Fig. 8E and F) and MG63 (Fig. 8M and N) cells. Immunohistochemistry of xenograft tumors revealed that Raf and MEK1/2 expression levels were significantly reduced in the sh-FBXL5 group in both 143B (Fig. 8G and H) and MG63 (Fig. 8O and P) models.

Collectively, these data demonstrate that FBXL5 knockdown suppresses OS progression by inhibiting the MAPK/ERK signaling pathway at multiple levels (Ras, Raf, MEK and ERK), providing a mechanistic basis for its tumor-suppressive effects.