Egg Yolk-Derived Peptide VPF Orchestrates OPG/RANKL/RANK-Mediated Signaling to Suppress Osteoclast Differentiation and Facilitate Bone Development in Zebrafish Authors: Zhongyan Bao, Juan Du, Yu Gao, Sinian Liu, Nanhai Xiao, Hang Wang, Luoyuan Zhao, Ming Chen, Fengqin Feng, Wei Qu, Qianqian Wang Published: 2026 ABSTRACT This study was designed to comprehensively analyze the benefit effect and underlying mechanism of egg yolk-derived peptide VPF (Val-Pro-Phe) in bone development by integrating network pharmacology, molecular docking/dynamics simulation, and zebrafish in vivo experiments. Results demonstrated that VPF exhibited dose-dependent intervention effects within the concentration range of 0.2–10 μg/mL in normal and bone dysplasia zebrafish. Specifically, 0.2 μg/mL of VPF exhibited significant enhancements in total length (TL), mineralization, and swimming distance by 4.01%, 14.00% (P < 0.05), and 7.31% in normal zebrafish model, whereas 1 μg/mL of VPF led to significant increases in the aforementioned indexes by 6.30%, 45.26%, and 51.05% (P < 0.05) in bone dysplasia zebrafish model. Then, network pharmacology revealed that the highest correlation between VPF in improving bone development and osteoclast differentiation pathway, with osteoprotegerin (OPG) being the core node in this pathway. Molecular docking results showed that VPF exhibited the highest binding affinity for OPG, and this interaction was further validated by 200 ns molecular dynamics simulations, which demonstrated the stability of the VPF-OPG complex. Furthermore, VPF upregulated the gene expression of OPG, and downregulated the gene expression of the receptor activator of the NF-κB ligand (RANKL) and TNF receptor-associated factor 6 Gene (TRAF6), and downregulated the RANKL/OPG ratio in zebrafish. These data indicate that VPF alleviates skeletal dysplasia primarily by OPG-mediated suppression of osteoclast differentiation within the OPG/RANKL/RANK axis, thereby supporting its potential value as a functional food ingredient for promoting bone development. Introduction Skeletal development is a highly orchestrated and dynamic biological process that encompasses the proliferation, differentiation, matrix secretion, and mineralization of osteoblastic and osteoclastic lineage cells, ultimately generating a framework that supports the body, protects viscera, enables locomotion, and maintains mineral homeostasis. Disruption of this program results in childhood growth retardation characterized by linear growth lag, deficient mineralization, and reduced bone strength, imposing severe physical and psychosocial burdens. − Current strategies to promote skeletal development in children primarily involve nutritional interventions, scientific exercise regimens, healthy lifestyle habits, and medical treatments. Nutritional interventions focus on supplementing key nutritional factors such as zinc, magnesium, vitamins, and proteins. Exercise recommendations emphasize high-impact weight-bearing activities and strength training. When these methods are insufficient, pharmacological interventions are employed, primarily involving the supplementation of growth hormone and calcium. Although both of them have shown promise in improving skeletal dysplasia, long-term use can result in complications such as secondary hypercalcemia, insulin resistance, and fibrosis at the injection site. Therefore, finding safe and effective natural alternatives to promote skeletal development is essential and urgent. Food-derived bioactive peptides have gained attention for their potential health benefits and safety compared with pharmaceutical agents. − It is a well-accepted paradigm in food-peptide research that trace-level endogenous peptides exert physiological effects through long-term dietary accumulation, while purified/synthetic peptides are used at moderate effective doses for mechanistic validation. − Studies have demonstrated that egg yolk peptides obtained by the enzymatic hydrolysis of degreased egg yolk exert beneficial effects on bone health. For example, studies have shown that water-soluble egg yolk peptides can enhance mineral density, thus exerting their osteoprotective effects. Moreover, researchers have further shown that HEYP can promote osteogenic differentiation through the MAPK/ELK1 signaling pathway, thus playing a role in strengthening bone. Additionally, a recent study demonstrated that egg yolk-derived peptides could mitigate bone loss in an ovariectomized rat model of osteoporosis. Hydrolyzed egg yolk powder promotes bone growth and development in rats by regulating bone metabolism and gut microbiome. Hydrolyzed egg yolk peptide alleviates ovariectomy-induced osteoporosis by regulating lipid metabolism. However, these studies on the osteogenesis mechanism of egg yolk peptides are limited to only osteoblast differentiation signaling pathways, and it is still unclear how egg yolk peptides act on osteoclast differentiation. Therefore, it is important to clarify the relationship between egg yolk peptides and osteoclast differentiation. Studies showed that the osteoprotegerin (OPG)/ receptor activator of NF-κB ligand (RANKL)/receptor activator of NF-κB (RANK) signaling pathway plays a critical role in regulating osteoclast differentiation and maturation. OPG, a secreted glycoprotein of the tumor necrosis factor receptor superfamily, is a master regulator of bone metabolism. It neutralizes RANKL and prevents its binding to RANK, thereby blocking the RANKL-RANK signaling cascade that drives osteoclast differentiation, activation, and survival. , During skeletal growth, OPG maintains the balance between bone formation and resorption by suppressing osteoclastogenesis. Consequently, the OPG/RANK/RANKL axis is central not only to skeletal homeostasis but also to the pathogenesis of growth retardation, osteoporosis, and rheumatoid arthritis. , This study was designed to comprehensively analyze the beneficial effect and underlying mechanism of egg yolk-derived peptide VPF in bone development. Here, we employed both normal and bone dysplasia zebrafish to comprehensively evaluate the beneficial effect of VPF derived from egg yolks in promoting bone development. Then, we integrated network pharmacology, molecular docking, and molecular dynamics to identify the potential target sites for the promotion of bone development by the VPF. To investigate the mechanisms by which VPF promotes bone development, we further analyzed the expression of key genes in the OPG/RANKL/RANK signaling pathway, a critical regulatory axis for osteoclast differentiation. Our findings provide a scientific basis for developing VPF as a next-generation functional food ingredient or a nutritional supplement for skeletal health. Material and Methods Materials Egg yolk-derived peptide VPF (Val-Pro-Phe, purity 98.0%, desalinated) was purchased from Borun Sida Biotechnology Co., Ltd. (Shenzhen, China). The content of VPF in egg yolk hydrolysate was determined to be 0.5–0.75 mg/g by HPLC (Table S1). The secondary ion mass spectrometry and chemical structural formula is shown in Figure S1. The simulated gastrointestinal enzymatic hydrolysis is shown in Figure S2. Glycerol, KOH, H2O2, and dexamethasone (DEX) were purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Magnesium chloride was purchased from China National Pharmaceutical Group (Shanghai, China). Alizarin red S staining solution and Alcian blue staining solution were purchased from Solarbio Science & Technology Co., Ltd. (Beijing, China). Paraformaldehyde (4%) was purchased from White Shark Biotechnology Co., Ltd. (Hefei, China). PBS was purchased from Meilun Biotechnology Co., Ltd. (Liaoning, China). The total RNA rapid extraction kit was purchased from Eside Biotechnology Co., Ltd. (Hangzhou, China), and the real-time fluorescence quantitative PCR kit was purchased from Novozymes Biotechnology Co., Ltd. (Nanjing, China). Evaluation of the Effects of VPF on Skeletal Development Based on Normal and Bone Dysplasia Zebrafish Adult zebrafish were kept in a constant temperature breeding system of 28 °C, with a 14/10 h light cycle at the same time every day, and fed twice to the mature shrimp at designated locations. Adult zebrafish were placed in a fish tank with a male-to-female ratio of 2:2 on the previous day. The male and female fish were separated using a baffle, and the baffle was removed the next morning. After about 30 min, the fish eggs in the tank were transferred to system water supplemented with methylene blue working solution (80 μL of 0.37 g/100 mL stock solution into 500 mL system water). Then, they were placed in a light incubator at a constant temperature of 28 °C for cultivation. After 3 days of fertilization (dpf), the juveniles were randomly selected for intervention. This experiment has been approved by the Ethics Review Committee for Animal Welfare at Zhejiang University, with ethics number (ZJU20250166). Experimental Design of Normal Zebrafish Model for Bone Development Juvenile zebrafish of 3 dpf were randomly assigned to 6-well plates with 10 fish per well and divided into the following groups: Blank control group: Containing only system water; Different dose of VPF group: System water containing VPF (0.1–10 μg/mL). Intervention was carried out daily with freshly prepared medicated system water until 7 dpf. The standard length, alizarin red S staining of bone, alcian blue staining for cartilage, locomotor activity, and gene expression were measured. Experimental Design of Bone Dysplasia Zebrafish Model for Bone Development Bone Dysplasia Zebrafish Model Establishment Juvenile zebrafish of 3 dpf were randomly assigned to 6-well plates with 10 fish per well and divided into the following groups: Blank control group: Containing only system water; Different dose of DEX group: System water containing DEX (15–35 μmol/L). Intervention was carried out daily with freshly prepared medicated system water until 7 dpf. Total length, alizarin red S staining of bone, alcian blue staining for cartilage, locomotor activity, and gene expression were measured. Experimental Design of DEX-Induced Bone Dysplasia Zebrafish To evaluate the therapeutic effect of VPF on DEX-induced bone dysplasia zebrafish, 3 dpf juvenile zebrafish were randomly assigned to 6-well plates with 10 fish per well and divided into the following groups: Blank control group: Containing only system water; Model control group: Continuously exposed to 30 μmol/L DEX system; DEX + VPF at different doses group: exposed to a system containing DEX and VPF (0.1–10 μg/mL). Intervention was carried out daily with freshly prepared medicated system water until 7 dpf. Total length, alizarin red S staining of bone, alcian blue staining for cartilage, locomotor activity, and gene expression were measured. Total Length Measurement After being cultured to 7 dpf, larvae were anesthetized with 0.008% MS-222 (tricaine methanesulfonate). After confirming the anesthesia of the fish body, it was transferred to a transparent glass plate for total length imaging. A Nikon SMZ18 fluorescence microscope with imaging parameters of 2× was used, and measured from the snout end (front end of the head) to the tail fin end as the standard. Alizarin Red S Staining for Bone Larvae were fixed in 4% paraformaldehyde (4 °C, 24 h), bleached (3% H2O2 + 0.5% KOH, 30 min), rinsed (25% glycerol +0.1% KOH), and stained with 0.01% Alizarin Red S (2 h, dark). After sequential washes (50% glycerol +0.1% KOH), specimens were mounted in 100% glycerol and imaged. Alcian Blue Staining for Cartilage Fixed larvae (4% PFA, 4 °C, 6 h) were bleached, rinsed in PBST, stained with 0.05% Alcian Blue (overnight, dark), dehydrated (80%, 50%, 25% ethanol, 10 min each), rebleached, cleared (25% glycerol +0.1%KOH), and photographed. Morphometric parameters (Intercranial distance, ceratohyal cartilage length, lower-jaw length, and ceratohyal angle) were quantified using ImageJ. Locomotor Activity At 7 dpf, single larvae were placed in 200 μL system water per well of a 96-well plate, acclimated for 5 min and recorded for 40 min (two 10 min dark/light cycles) using a ZebraBox system. Total distance, mean speed, and activity frequency were analyzed. Biochemical Analysis At 7 dpf, 30 zebrafish larvae per group were randomly sampled and pooled in 200 μL ice-cold PBS (pH 7.4). To ensure complete protein extraction and maintain enzymatic stability, the samples were homogenized on ice for 60 s using a high-speed tissue homogenizer at 60 Hz. The resulting homogenates were centrifuged at 12,000g for 10 min at 4 °C to effectively remove cellular debris and obtain a clear supernatant. The supernatant (three technical replicates per group) was immediately collected and used for tartrate-resistant acid phosphatase (TRAP) activity quantification using a commercial kit (Jiangsu Meimian Biotechnology Co., Ltd., China) according to the manufacturer’s instructions. All procedures were performed strictly at 4 °C to prevent potential thermal degradation of the TRAP enzyme.. Network Pharmacology The potential targets of egg yolk-derived peptide VPF were predicted using the SuperPred platform (probability ≥ 0.5), TargetNet database (score ≥ 0.7), and SwissTargetPrediction (probability ≥ 0.1). Simultaneously, disease-related targets were retrieved from the GeneCards and OMIM databases using the keywords “bone growth and development” and “Skeletal Dysplasia.” For GeneCards, only targets with a relevance score ≥1.0 were selected to ensure high correlation. The intersection of VPF-predicted targets and disease-related targets was identified using a Venn diagram to establish the candidate gene set. The identified targets were then imported into the STRING 12.0 database (minimum required interaction score: high confidence 0.700) to construct the protein–protein interaction (PPI) network. The network was visualized using Cytoscape 3.10.1. To identify hub genes, the MCODE module was employed with the following parameters: degree cutoff = 2, node score cutoff = 0.2, K-core = 2, and max depth = 100. Finally, GO biological process enrichment and KEGG pathway analyses were performed using the DAVID 2021 database. Only terms with a p-value of <0.05 and a false discovery rate (FDR) of <0.05 were considered significantly enriched and subsequently visualized. Molecular Docking and Dynamics Molecular Docking The structural files of core target proteins OPG, AKT1, TRAF6, LCK, and NFKB1 were retrieved from the PDB database (https://www.rcsb.org/). Obtain and then use Pymol software to remove water molecules and small-molecule ligands from the structural file to obtain the protein receptor file. The polypeptide chain was constructed using Maestro Academic Edition, imported into AutoDock Vina 1.1.2 for hydrogenation and charge addition, and exported as a ligand file. The optimal binding region between each pair of small-molecule ligands and protein receptor files was calculated using AutoDock Vina 1.1.2. Then, the affinity values of each small molecule with its target were calculated through semiflexible molecular docking, and the binding conformation diagram of the peptide with the target was drawn using Pymol software. Molecular Dynamics The protein–peptide complexes obtained from molecular docking were subjected to 200 ns molecular dynamics simulations using Gromacs v2024.6 software and the CHARMM36 force field. The simulation was conducted by minimizing the initial energy of protein and small-molecule complexes and optimizing the structure using the L-BFGS algorithm to ensure system stability. Then, the system underwent isothermal constant volume (NVT) and isothermal isobaric (NPT) equilibrium processes, with temperature and pressure maintained at 300 K and 1 bar, respectively. The dynamic behavior of the system was calculated by using the parameters and algorithms provided in the force field, and a productive simulation was conducted within 200 ns. Throughout the simulation process, the particle network Ewald (PME) method and all bond lengths were constrained using the LINCS algorithm. To evaluate the binding stability and conformational dynamics, the root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), radius of gyration (R g), and solvent-accessible surface area (SASA) were calculated using GROMACS built-in tools. Furthermore, the number of intermolecular hydrogen bonds was monitored throughout the 200 ns trajectory to assess the dynamic stability of the VPF-OPG interaction. To further characterize the thermodynamic stability and identify the representative stable conformations, the free energy landscape (FEL) was constructed using the RMSD and R g as reaction coordinates. The energy bins and clusters were visualized to confirm that the complex resided in a global minimum-energy state during the simulation. Quantification of mRNA Expression by RT-qPCR Analysis Total RNA was extracted from 30 larvae per group, , reverse-transcribed using HiScript III RT SuperMix (+gDNA wiper), and amplified with Taq Pro Universal SYBR qPCR Master Mix on a CFX96 system. Relative expression was calculated by 2∧(−ΔΔCt) with β-actin as reference. Primers were designed by Biotechnology Co., Ltd. (Zhejiang). The Primers sequence is shown in Table S2 of the supplementary file. Statistical Analyses Data are mean ± SEM. Differences were evaluated by one-way ANOVA followed by Tukey’s test (GraphPad Prism 10.1.2); P < 0.05 was considered significant. Results VPF Enhances Larval Growth, Bone Mineralization and Locomotor Performance without Cartilage Anomalies in Normal Zebrafish To evaluate the somatic growth of larvae, total length was measured at 7 dpf. Compared with the control group, 0.2–2 μg/mL VPF significantly increased total length in a dose-dependent manner, reaching 4.01% (P < 0.01) at 0.2 μg/mL (Figure A), indicating a growth-promoting potential of the peptide. For comprehensive skeletal assessment, larvae were subjected to Alizarin red S and Alcian blue staining. Alizarin Red S forms red Ca2+-chelating complexes within mineralized bone , whereas Alcian Blue electrostatically binds cartilage glycosaminoglycans. , After 0.2–10 μg/mL VPF, mineralization intensity and area were markedly elevated; 0.2 μg/mL raised mineralization by 14.00% (P < 0.05), while 0.5–2 μg/mL increased it 22.00% (P < 0.01), 23.10% (P < 0.01), 23.10% (P < 0.05) (Figure B–D), demonstrating enhanced calcium deposition. Importantly, no significant differences in cartilage morphology were detected between control and VPF groups; both exhibited clear and structurally intact cartilage (Figure E–I), confirming that VPF promotes development without inducing cartilage lesions. Furthermore, locomotor activity, which serves as an indirect index of skeletal function, was assessed under light/dark cycles. VPF significantly increased total distance and mean speed; at 5 μg/mL, total distance and mean speed increased by 36.66% (P < 0.05) and 36.63% (P < 0.05), while 10 μg/mL yielded enhancements of 36.86% (P < 0.05) and 36.80% (P < 0.05). (Figure J,K). Active and resting frequencies were also improved at 2, 5, and 10 μg/mL (P < 0.01) (Figure L,M), with pronounced speed changes during dark-to-light transitions (Figure N). Collectively, VPF enhances both skeletal mineralization and locomotor performance in larval zebrafish. VPF promotes skeletal development and enhances locomotor performance in normal zebrafish larvae at 7 dpf. (A) Representative measurements of larval total body length. (B) Typical images of bone mineralization visualized by Alizarin Red S staining. (C–D) Quantitative analysis of relative bone mineralization intensity (IOD) and mineralized area. (E) Representative images of craniofacial cartilage visualized by Alcian Blue staining. (F–I) Morphometric analysis of craniofacial structures, including the ceratohyal angle (F), lower-jaw length (G), intercranial distance (H), and Meckel’s cartilage length (I). (J–N) Assessment of locomotor activity, including total distance traveled (J), average velocity (K), active frequency (L), resting frequency (M), and time-course speed distribution (N). Normal zebrafish were treated with VPF at concentrations of 0.2–10 μg/mL from 3 to 7 dpf. Data are expressed as mean ± SEM (n = 10 per group). Significance levels: *P < 0.05, **P < 0.01, and ***P < 0.001 vs the control group. VPF: Val-Pro-Phe. Establishment of a Dexamethasone-Induced Bone Dysplasia Model Glucocorticoids are widely prescribed for inflammatory and autoimmune disorders; however, excess doses suppress osteoblast differentiation, induce osteoblast apoptosis, and enhance osteoclastogenesis, thereby disrupting the bone formation/resorption balance and leading to osteoporosis and skeletal dysplasia. , DEX, a pure glucocorticoid receptor agonist with minimal mineralocorticoid activity, exhibits high potency, a long half-life, and metabolic stability, , making it suitable for immersion exposure of zebrafish larvae and for generating reproducible bone malformation phenotypes. To determine the optimal DEX concentration, larvae were exposed to 20–35 μmol/L for 4 days. At 30 μmol/L and 35 μmol/L, total length was significantly reduced by 3.32% (P < 0.05) and 3.39% (P < 0.05), respectively (Figure A). Mineralization decreased in a dose-dependent manner within 20–35 μmol/L (P < 0.05), with the 30 μmol/L group showing a 32.31% reduction (P < 0.01) (Figure B–D). Compared with controls, DEX-treated larvae exhibited shorter Meckel’s cartilage, narrower intercranial distance, and reduced lower-jaw length (P < 0.05), resulting in a narrower anterior skull, shortened interocular distance, and decreased cranial length (Figure E–I). Behavioral tracking revealed diminished total distance, mean speed and activity, especially in the 30 μmol/L group (−33.07% distance, −37.43% speed (P < 0.05); Figure J–N). Collectively, 30 μmol/L DEX reproducibly induces skeletal dysplasia in larval zebrafish and was selected as the model concentration for subsequent rescue experiments. Characterization of dexamethasone (DEX)-induced bone dysplasia in zebrafish larvae at 7 dpf. (A) Comparison of total body length between control and DEX-treated larvae, indicating systemic growth retardation. (B) Representative images of bone mineralization visualized by Alizarin Red S staining. (C–D) Quantitative analysis of relative bone mineralization intensity (IOD) and mineralized area. (E) Representative images of craniofacial cartilage visualized by Alcian Blue staining. (F–I) Morphometric analysis of impaired craniofacial structures, including the ceratohyal angle (F), lower-jaw length (G), intercranial distance (H), and Meckel’s cartilage length (I). (J–N) Assessment of impaired locomotor performance, including total distance traveled (J), average velocity (K), active frequency (L), resting frequency (M), and time-course speed distribution (N). To establish the bone dysplasia model, zebrafish larvae were treated with 30 μM DEX from 3 to 7 dpf. Data are expressed as mean ± SEM (n = 10 per group). Significance levels: *P < 0.05, **P < 0.01, and ***P < 0.001 vs the control group. DEX: Dexamethasone. Yolk Peptide VPF Alleviates Dexamethasone-Induced Bone Dysplasia in Zebrafish Following DEX exposure, larvae exhibited markedly reduced mineralization, cartilage shortening, and impaired locomotion-phenotypes consistent with previous reports. Intervention with 0.2–10 μg/mL VPF dose-dependently restored both somatic and skeletal parameters. Compared with the DEX group, total length increased by 3.57% (P < 0.05), 6.38% (P < 0.001), 6.30% (P < 0.001), 6.30% (P < 0.001), 6.00% (P < 0.001), and 5.89% (P < 0.001), while mineralization intensity rose by 41.90% (P < 0.05), 49.92% (P < 0.01), 45.26% (P < 0.01), 38.53% (P < 0.05), 39.44% (P < 0.05), and 35.42% across the tested range (Figure A–D). Meckel’s cartilage length, intercranial width, and lower-jaw length were all significantly increased (P < 0.05) and did not differ from blank controls (Figure E–I), indicating full skeletal rescue without cartilage anomalies. DEX also decreased total swimming distance, mean speed, and activity index. VPF (1–5 μmol/L) significantly reversed these deficits: swimming distance improved by 51.05% (P < 0.01), 52.48% (P < 0.01), and 41.24% (P < 0.01), and speed by 57.07% (P < 0.01), 52.46% (P < 0.01), and 41.18% (P < 0.01), respectively (Figure J–N). Collectively, VPF effectively mitigates dexamethasone-induced skeletal dysplasia in larval zebrafish. VPF alleviates dexamethasone (DEX)-induced bone dysplasia and enhances locomotor performance in zebrafish larvae at 7 dpf. (A) Total body length (TL) measurements across different groups. (B) Representative images of bone mineralization visualized by Alizarin Red S staining. (C–D) Quantitative analysis of relative bone mineralization intensity (IOD) and mineralized area. (E) Representative images of craniofacial cartilage visualized by Alcian Blue staining. (F–I) Morphometric analysis of craniofacial structures, including the ceratohyal angle (F), lower-jaw length (G), intercranial distance (H), and Meckel’s cartilage length (I). (J–N) Assessment of locomotor activity recovery, including total distance traveled (J), average velocity (K), active frequency (L), resting frequency (M), and time-course speed distribution (N). Larvae were cotreated with 30 μM DEX and VPF at indicated concentrations (0.2–10 μg/mL) from 3 to 7 dpf. Data are expressed as mean ± SEM (n = 10 per group). Significance levels: *P < 0.05, **P < 0.01, and ***P < 0.001 vs the control group; #P < 0.05, ##P < 0.01, and ###P < 0.001 vs the DEX-induced model group. VPF: Val-Pro-Phe; DEX: Dexamethasone. Network Pharmacology Reveals That VPF Regulates Bone Development by Targeting Osteoclast Differentiation To decode the molecular basis of VPF-driven bone anabolism, we performed a network pharmacology workflow integrating target prediction, PPI construction, and pathway enrichment. The VPF structure was uploaded to SuperPred, TargetNet, and SwissTargetPrediction; intersecting targets with GeneCards entries for “bone growth and development” and “Skeletal Dysplasia” yielded 194 common genes (Figure A). PPI network construction in STRING and Cytoscape (Figure B–C) followed by MODE analysis (degree ≥ 2, node score ≥ 0.2, k-core ≥ 2) identified ten hub targets: MMPB, PTGS2, CXCR4, TRAF6, LCK, AKT1, STAT3, NFKB1, TNFRSF11B, and CDK4 (Figure D). Network pharmacology analysis identifying the potential targets and signaling pathways of VPF in the context of skeletal development. (A) Venn diagram illustrating the intersection of candidate gene sets among VPF, skeletal dysplasia, and bone growth/development. (B–C) Construction of the protein–protein Interaction (PPI) network based on the STRING database, showing the interconnectedness of identified targets. (D) Identification and display of core targets based on network topology analysis. (E–F) KEGG pathway enrichment mapping and annotation, highlighting the involvement of the osteoclast differentiation and MAPK signaling pathways. (G) Gene Ontology (GO) enrichment analysis, categorized by biological processes, molecular functions, and cellular components. (H) Integrated Target-Pathway network showing the complex associations between core targets and their modulated KEGG pathways. All network visualizations were performed using Cytoscape software. VPF: Val-Pro-Phe tripeptide. KEGG enrichment (Figure E) and GO classification (Figure F,G) revealed that the predicted targets are enriched in osteoclast differentiation, TNF signaling, NF-κB signaling, cAMP signaling, IL-17 signaling, PI3K-Akt signaling, Th17 differentiation, GH synthesis/secretion, and MAPK signaling, primarily within the “Environmental Information Processing-Signal Transduction” superclass. GO terms covered bone-tissue development, osteogenesis, mineralization and cartilage development (BP), Ca2+/BMP binding and ECM structural constituents (MF), and extracellular matrix, plasma membrane, and cytosol (CC). Intersection of hub genes with KEGG pathways (Figure H) showed the highest connectivity for osteoclast differentiation, involving AKT1, TRAF6, LCK, NFKB1, and TNFRSF11B (OPG). Consequently, osteoclast differentiation was selected as the primary pathway for downstream mechanistic validation. Molecular Docking and Dynamics Unveil High-Affinity Binding of Yolk Peptide VPF to OPG The intersection targets AKT1, TRAF6, LCK, NFKB1, and TNFRSF11B were input into the UniProt database to find the corresponding protein names AKT1, TRAF6, LCK, NFKB1, and OPG, and then the proteins were subjected to molecular docking and the binding energy was calculated. Molecular docking binding energy analysis showed that the binding energies of VPF with TRAF6, NFKB1, and OPG proteins were −5.9, −5.8, and −8.2 kcal/moL, respectively (Figure A). The smaller the binding energy, the less free energy required by the compound, the stronger the docking ability, and the higher the stability after molecular docking. The binding energy of VPF and OPG was the lowest, and the binding was more stable. To further study the dynamic process of the binding of the two, molecular dynamics simulation was used to evaluate the dynamic changes within 200 ns. Molecular docking and molecular dynamics (MD) simulation of the VPF-OPG complex over a 200 ns trajectory. (A) Representative 3D schematic and 2D interaction map of VPF docking into the active pocket of the OPG protein. (B) Root-mean-square deviation (RMSD) profile of the complex backbone, indicating structural convergence and stability. (C) Root-mean-square fluctuation (RMSF) values of OPG amino acid residues, reflecting the local flexibility of the protein upon VPF binding. (D) Solvent-accessible surface area (SASA) analysis over time. (E) Radius of gyration (Rg) evolution, representing the compactness of the protein-peptide complex. (F) Evolution of intermolecular hydrogen bonds, demonstrating the persistent polar interactions throughout the 200 ns simulation. (G–H) 2D and 3D free energy landscape (FEL) analysis, identifying the thermodynamic stability and the primary conformational clusters of the VPF-OPG complex. All MD analyses were conducted to evaluate the binding stability under dynamic physiological conditions. VPF: Val-Pro-Phe; OPG: Osteoprotegerin. The RMSD measures the total deviation of all atoms in a specific conformation relative to the reference conformation and is a key indicator for judging system stability. , As shown in Figure B, the binding of the VPF and OPG proteins was stable. Although there was a small peak around 150 ns, the average RMSD increased from 0.4 to 0.6 nm, suggesting that VPF caused a transient conformational adjustment after entering the binding site, but it quickly fell back to the baseline without destroying the overall stability. The RMSF can reflect the flexibility of amino acid residues within the protein. Residue fluctuation analysis showed that the fluctuation values of most residues were in the range of 0.2–0.4 nm, but some residues had higher fluctuations (>0.5 nm), which may be located at the binding interface or flexible loop regions. The overall fluctuation level indicated that the binding region had certain flexibility, but no abnormal fluctuations occurred, supporting the binding stability (Figure C). The SASA results showed that the solvent-accessible surface area of the complex did not increase significantly during the stable period, indicating that the binding interface formed an effective hydrophobic core and reduced solvent exposure (Figure D). The RG remained at 2.4 ± 0.2 nm, with a small fluctuation coefficient, indicating that the overall conformation was compact and no significant unfolding or compression occurred, further verifying the structural stability of the complex (Figure E). The number of hydrogen bonds fluctuated dynamically within 10, indicating that hydrogen bonds played a key role in the binding (Figure F). Although the number fluctuated, hydrogen bonds persisted, suggesting strong binding affinity. At the same time, the hydrogen bond curve was basically synchronous with RMSD, indicating that the hydrogen bond stability may promote the overall system to enter a steady state. As shown in the free energy low-lying areas (blue) in Figure G,H, the system has the lowest free energy near specific RG and RMSD values, indicating that VPF and OPG protein bind most stably under these conformations. At the same time, the stable conformations corresponding to the free-energy low-lying areas were consistent with the RMSD results. Comprehensive simulation results show that the interaction between VPF and the OPG protein is stable, and this interaction may regulate the biological function of OPG through specific amino acid residues and conformational changes. These findings provide an important structural basis and theoretical guidance for further experimental research and drug design. VPF Orchestrates OPG/RANKL/RANK-Mediated Signaling to Suppress Osteoclast Differentiation and Facilitate Bone Development in Zebrafish On the basis of the results of network pharmacology analysis, this study conducted further validation, focusing on investigating the expression changes of genes and signaling pathways related to osteoclast differentiation. The experimental results showed that genes closely associated with bone resorption processes, such as osteoprotegerin (OPG) (P < 0.05), receptor activator of nuclear factor kappa B ligand (RANKL) (P < 0.05), tumor necrosis factor receptor-associated factor 6 (TRAF6) (P < 0.001), P65 (P < 0.01), JNK (P < 0.001), P38 (P < 0.01), etc. Figure A–H also showed changes in expression levels after VPF intervention. Specifically, in zebrafish with delayed growth and development, treatment with dexamethasone (DEX) significantly increased the ratio of RANKL to OPG, leading to enhanced bone resorption processes; after VPF intervention, the ratio significantly decreased, suggesting that VPF may inhibit bone resorption and alleviate bone dysplasia by regulating the RANKL/OPG ratio. VPF alleviates dexamethasone (DEX)-induced bone dysplasia by modulating the OPG/RANKL/RANK signaling cascade in zebrafish larvae at 7 dpf. (A–K) Relative mRNA expression levels of key osteoclastogenic and osteogenic marker genes determined by qRT-PCR, including bone resorption-related genes such as opg (A), rankl (B), traf6 (C), p65 (D), jnk (E), p38 (F), nfatc1 (G), and trap (H); and bone formation-related genes such as tgf-β1 (I), bmp2 (J), and alp (K). (L) Comprehensive schematic diagram illustrating the regulatory mechanism by which VPF orchestrates the OPG/RANKL axis to suppress osteoclast differentiation and facilitate bone development. Total RNA was extracted from pools of 30 larvae per replicate at 7 dpf. Data are expressed as mean ± SEM (n = 3 biological replicates). Significance levels: *P < 0.05, **P < 0.01, and ***P < 0.001 vs the control group; #P < 0.05, ##p < 0.01, and ###P < 0.001 vs the DEX-induced model group. Further analysis showed that VPF intervention not only inhibited the expression of osteoclast related genes but also significantly upregulated genes closely related to bone formation, including transforming growth factor -β1 (TGF-β1) (P < 0.05), bone morphogenetic protein (BMP) (P < 0.05), alkaline phosphatase (ALP) (P < 0.05), etc. (Figure. I–K). These genes play a crucial role in bone development, and their upregulation suggests that VPF may accelerate bone formation by promoting the expression of these genes. Based on the mechanism diagram (Figure. L), it can be seen that VPF upregulates OPG expression, blocks the binding of RANKL and RANK, thereby inhibiting the TRAF6 mediated NF-κB and MAPK (JNK/P38) signaling pathways, reducing the transcriptional activity of NFATc1 (P < 0.01), and ultimately reducing the expression of osteoclast specific genes (TRAP) (P < 0.01). At the same time, VPF activates the TGF-β/BMP/Smad signaling pathway, promotes the expression of osteogenic associated genes (ALP), and enhances the differentiation and mineralization ability of osteoblasts. This “bidirectional regulation” mechanism enables VPF to promote bone formation while inhibiting bone resorption, effectively restoring bone development balance. In summary, VPF achieves negative regulation of osteoclast differentiation by regulating the OPG/RANKL/RANK signaling pathway, and synergistically activates the TGF-β/BMP/Smad signaling pathway to promote osteoblast differentiation, thus exhibiting significant bone protective effects in both in vitro and in vivo experiments. This discovery not only deepens the understanding of the mechanism of action of VPF but also provides new theoretical basis for the development of functional foods targeting bone dysplasia. Discussion Skeletal dysplasia and its attendant growth retardation constitute a major global health burden. Over the past decade, mining food proteins for osteoactive peptides has emerged as a promising avenue for functional foods and nutraceuticals. Here, we provide systematic evidence that the yolk-derived tripeptide VPF promotes physiological skeletal development and rescues DEX-induced bone dysplasia in zebrafish, and we define its molecular mechanism of action. The zebrafish offers optical transparency, rapid development, genetic tractability, high fecundity and low cost, , and has become a premier vertebrate model for investigating early osteogenesis. Real-time high-resolution imaging with fluorochromes such as Calcein and Alizarin Red S enables direct visualization of endochondral and intramembranous ossification, and allows assessment of skeletal responses to systemic cues (hormones, nutrients) , and environmental insults (drugs, toxicants). With 70% of human disease genes having orthologues in zebrafish and highly conserved metabolic pathways, this species represents an ideal platform for dissecting the molecular mechanisms of skeletal development. In our study, we demonstrated that 0.2–10 μg/mL VPF significantly increases total length, bone mineralization, and locomotor performance in normal larvae without inducing cartilage anomalies. This aligns with recent food-peptide studies: Lv et al. reported that hydrolyzed whole yolk accelerates rat skeletal growth, while Printzi et al. showed that early peptide nutrition shapes zebrafish skeletal morphology. Our work narrows the active moiety to a single, defined tripeptide and provides quantitative phenotypic detail, offering a precise functional ingredient for formulation. To assess intervention potential, we further established a DEX-induced bone dysplasia model that faithfully reproduces the clinical phenotype of glucocorticoid excess-linear growth lag, hypomineralisation, and hypoactivity-consistent with mammalian data. , VPF fully reversed these defects at nutritionally achievable concentrations, an effect comparable to that reported for pinoresinol diglucoside in zebrafish but achieved with a smaller, food-grade peptide amenable to oral delivery. In recent years, the combination of network pharmacology, molecular docking, and molecular dynamics simulation has become a good tool for rapidly screening bioactive peptide targets. For example, lactoferrin peptide ENLPEKADRDQYEL has high molecular weight binding affinity with epidermal growth factor receptor (EGFR), thereby promoting osteoblast proliferation activity. The CDOCKER energy value of the ocean peptide YPRKDETGAERT docked to integrin was 204.482 kcal/moL and promoted MC3T3-E1 cell proliferation. The tripeptide Leu-Pro-Lys from Sipunculus nudus protein was used to predict the interaction between Leu-Pro-Lys and bone development through network pharmacology. It was found that Leu-Pro-Lys promotes bone development by regulating endocrine resistance, relaxin signaling pathway, and estrogen signaling pathway. Similarly, our study applied the aforementioned triple strategy for the first time to egg yolk-derived peptide VPF, and predicted its optimal target to be OPG, suggesting that VPF may reshape bone balance through “osteoclast inhibition” rather than simply “promoting bone.” OPG is a natural bait receptor for RANKL; When it binds to RANKL, it can block the RANK-TRAF6-NF-κB-NFATc1 cascade, thereby inhibiting osteoclast differentiation and bone resorption. Our study showed that the RMSD of VPF-OPG complex was 0.4–0.6 nm at 200 ns MD, and hydrogen bonding persisted, confirming the structural stability. Functionally, VPF reduces the RANKL/OPG ratio, downregulates NFATc1 and TRAP expression, directly replicates the phenotype of OPG overexpression, and verifies its role as an “OPG enhancer” from both structural and functional perspectives. Interestingly, current research on egg yolk peptides focuses on promoting bone signaling: Egg yolk peptides stimulate osteogenic differentiation through the MAPK/ELK1 signaling pathway; Egg yolk peptides activate Wnt/β-catenin signaling to induce RUNX2 expression, promoting osteogenesis and osteoblast differentiation to improve osteoporosis and serum metabolites related to bone growth; Egg yolk peptides inhibit fat production and promote bone formation by activating the BMP2/RUNX2 pathway and inhibiting the Akt/PPARγ pathway. In contrast, this study extends the function of egg yolk peptides to “osteoclast inhibition” for the first time, by directly targeting OPG to achieve negative regulation of osteoclast differentiation, completing the last piece of the puzzle of “osteogenic osteoclast” bidirectional regulation. Consistent with the existing peptide gene profile, cucumber seed peptide downregulates RANKL/TRAF6 and inhibits the expression of NFATc1. Wheat germ peptides inhibit the expression of osteoclast differentiation specific genes c-fos, NFATc1, TRAP, and CTSK. Desalinated duck egg white peptides reduce the expression of RANK by regulating the upregulation of Runx2 and OPG. Water-soluble egg yolk fractions FC suppressed the expression of key osteoclastic markers TRAF6, NFATc1, and cathepsin K, thereby promoting osteoclast apoptosis. In our study, VPF simultaneously downregulated RANKL, TRAF6, JNK, P38, P65, NFATc1, and TRAP while upregulating OPG, TGF-β1, and BMP2. For the first time, “OPG centric” bidirectional transcriptional regulation was achieved in food-derived peptides, providing molecular labels for precise nutritional interventions. Through the quadruple evidence of “network docking dynamics gene,” we demonstrate that VPF is a natural OPG enhancer and stably exists after simulated gastrointestinal enzymatic hydrolysis. It can inhibit osteoclast differentiation by mediating the OPG/RANKL/RANK pathway and synergistically activate osteogenic differentiation by mediating the TGF-β/BMP pathway, thereby bidirectionally reshaping bone stability. Future research will focus on bridging the current in vivo findings with high-resolution mechanistic studies and broader physiological validations. Specifically, osteoclast cell models (e.g., RANKL-induced RAW264.7 or BMMs) and whole-mount in situ hybridization (WISH) will be employed to further clarify the cell-specific and spatial–temporal mechanisms of VPF. Additionally, the oral bioavailability and long-term safety of VPF will be validated in ovariectomized or glucocorticoid-treated rat models, thereby laying a scientific foundation for developing mild dietary intervention strategies for bone dysplasia in children. Conclusions On the basis of integrated approaches of network pharmacology, molecular simulations, and in vivo validation, this study demonstrates that egg yolk-derived peptide VPF promotes bone development primarily by modulating the OPG/RANKL/RANK signaling pathway to inhibit osteoclast differentiation. VPF administration significantly enhanced the total length, bone mineralization, and locomotor activity in both normal and bone dysplasia zebrafish models in a dose-dependent manner. Molecular analyses confirmed that VPF upregulates the level of OPG expression while downregulating RANKL, effectively reducing the RANKL/OPG ratio and suppressing bone resorption. These findings not only clarify the mechanism of action of VPF in skeletal development but also provide a new candidate for functional foods or nutraceuticals aimed at bone health promotion. Future work will evaluate the safety, bioavailability, and clinical efficacy of VPF in humans, with the goal of translating this food-derived peptide into a practical intervention for skeletal disorders. Supplementary Material