A bioorthogonal nanoconcentrisome system for epigenetic and immunometabolic reprogramming of tissue regeneration in aged female rats Authors: Lan Zhao, Zheyuan Zhang, Luyu Zheng, Yiming Liang, Yujun Zhang, Hanshen Yang, Chenchen Zhao, Pengcheng Qiu, Junxin Chen, Wenbin Xu, Shunwu Fan, Yueqi Zhao, Zhaoming Liu, Xiangqian Fang, Sheng Lu, Kaiyu Liang Published: 2026 ABSTRACT Stem cell–based therapies show promise for tissue regeneration but are limited by therapeutic instability, particularly in aged tissues. Here, multi-omics profiling of regenerating bone reveals marked depletion of S-adenosylmethionine (SAM), indicating epigenetic dysregulation associated with impaired repair. To address this, we develop a spatially compartmentalized, fusogenic nanovesicle system (SMNV), termed a “nanoconcentrisome,” composed of SAM-loaded fusogenic liposomes encapsulating mesenchymal stem cell–derived nanovesicles (MNV). This dual-compartment architecture facilitates cytosolic delivery of cargos through membrane fusion with limited lysosomal involvement. In macrophages, SMNV attenuates inflammatory activation and promotes efferocytosis and metabolic reprogramming, while in senescent stem cells it enhances epigenetic integrity, mitochondrial function and osteogenic differentiation. Through coordinated modulation of immune and progenitor cell states, SMNV significantly enhances tissue regeneration and effectively repairs bone defects in aged female rats, highlighting a precision nanomedicine platform that integrates metabolic–epigenetic regulation with cell-free regenerative signaling to counteract ageing-associated regenerative decline. Introduction The aging global population has led to a rising incidence of osteoporosis and fragility fractures, imposing a heavy burden on patients’ mobility and quality of life1. In this context, critical-size bone defects present a particularly formidable clinical challenge. These defects severely impair function and often require complex reconstructive approaches2. Aged bone is especially difficult to regenerate, as age-related vascular regression, chronic inflammation, and metabolic dysregulation collectively compromise healing3. These factors underscore the urgent need for regenerative strategies tailored to the unique challenges of aged bone repair4. Stem cell–based therapies, including stem cells themselves and stem cell–derived nanovesicles, have shown great promise in tissue regeneration5. Among these approaches, stem cell–derived nanovesicles have emerged as a compelling cell-free strategy owing to their low immunogenicity, high biocompatibility, and ease of storage and modification6. Encouragingly, recent studies have demonstrated that stem cell–derived nanovesicles can enhance bone regeneration even in aged microenvironments, underscoring their therapeutic potential in age-related tissue repair7. However, despite these advances, the regenerative efficacy of stem cell–derived nanovesicles remains variable, particularly in pathological or aging-associated tissues. Accumulating evidence suggests that their therapeutic performance is strongly influenced by donor cell status and intrinsic cargo composition, which may not be sufficient to fully overcome the epigenetic and metabolic dysfunction that accumulates within compromised regenerative niches8. These limitations highlight an unmet need for next-generation stem cell–based strategies capable of directly reprogramming the dysfunctional cellular state to achieve robust and reproducible regeneration in complex aging microenvironments. Tissue injury and aging are frequently accompanied by profound metabolic disturbances that compromise the epigenetic regulatory capacity of key supporting cells, including macrophages and mesenchymal stromal cells (MSCs), thereby impairing their regenerative functions7. A growing body of evidence has identified central metabolites—such as SAM, α-ketoglutarate (αKG), and acetyl-CoA—not only as metabolic intermediates but also as essential cofactors for chromatin-modifying enzymes8. Through direct modulation of chromatin states, these metabolites can induce durable transcriptional and phenotypic changes that are difficult to achieve through conventional growth factor–based interventions alone. Among these metabolites, SAM occupies a central position as the universal methyl donor for DNA and histone methylation. Recent studies have demonstrated that SAM-mediated epigenetic regulation enhances macrophage efferocytosis and promotes inflammatory resolution, while also counteracting stem cell aging and functional decline9,10. These findings suggest that metabolic epigenetic insufficiency represents a previously underappreciated barrier to effective regeneration and raise the possibility that integrating epigenetic metabolites into stem cell–derived nanovesicle platforms may provide a rational strategy to overcome the limitations of current regenerative therapies in aged or diseased tissues. In this study, we introduce a conceptually integrated nano-delivery system—a spatially compartmentalized fusogenic nanovesicle, also termed a “nanoconcentrisome”—designed to couple metabolic epigenetic regulation with stem cell–derived regenerative signaling. The system consists of MNVs encapsulated within SAM-loaded fusogenic liposomes and stabilized via click chemistry, forming a dual-compartment architecture that enables hierarchically organized cargo delivery (Fig. 1a). Upon fusion with pro-inflammatory macrophages, SMNVs deliver SAM to suppress inflammation, reprogram cellular metabolism, and enhance efferocytosis, while MNVs provide regenerative cues to facilitate macrophage resolution (Fig. 1b). Similarly, in senescent MSCs, SAM restores epigenetic balance, followed by MNV-mediated enhancement of mitochondrial function and osteogenic differentiation (Fig. 1b). In aged bone defect models, SMNV treatment significantly promoted tissue repair, highlighting the therapeutic potential of integrating metabolic reprogramming with cell-free regenerative signaling to address complex, multicellular disease microenvironments.Fig. 1Overview of nanoconcentrisomes in promoting the repair of aged bone injuries.a Synthesis of nanoconcentrisomes, consisting of dual compartments: the outer compartment containing SAM and the inner compartment comprising MSC-derived vesicles. These compartments are separated by click chemistry linking the glycan chains on the vesicles to the PEG chains on the outer lipid layer. b The regulation of macrophages and stem cells by SMNV, facilitating the construction of a regenerative niche for the rapid repair of aged bone injuries. Overview of nanoconcentrisomes in promoting the repair of aged bone injuries. a Synthesis of nanoconcentrisomes, consisting of dual compartments: the outer compartment containing SAM and the inner compartment comprising MSC-derived vesicles. These compartments are separated by click chemistry linking the glycan chains on the vesicles to the PEG chains on the outer lipid layer. b The regulation of macrophages and stem cells by SMNV, facilitating the construction of a regenerative niche for the rapid repair of aged bone injuries. Results SAM depletion emerges as a key metabolic vulnerability during regenerative responses Aging is associated with widespread epigenetic dysregulation that compromises tissue regeneration, yet the cellular basis and metabolic underpinnings of these defects remain poorly defined11. To address this, we performed single-cell RNA sequencing data analysis of fracture callus tissue collected during bone repair to dissect cell type–specific transcriptional alterations within the aged regenerative microenvironment12. Unsupervised clustering of scRNA-seq data revealed six major cell populations, reflecting the diverse cellular composition of the bone repair site (Fig. 2a, b). These included B lymphocytes, endothelial cells, monocyte–macrophages, neutrophils, erythrocytes, and skeletal stem/progenitor cells (SSPCs) (Fig. 2a, b). Following tissue injury, macrophages act as central immunoregulatory orchestrators within the regenerative niche, coordinating inflammation resolution, efferocytosis, metabolic adaptation, and paracrine communication with osteogenic progenitors13,14. In parallel, SSPCs represent the primary osteogenic progenitor population responsible for new bone formation13,15. Given their well-established and complementary roles in bone regeneration, immune modulation, and osteo–immune coupling, our subsequent single-cell analyses primarily focused on these two cell populations. Notably, genes involved in polyamine biosynthesis—a major route of SAM consumption—were significantly upregulated across monocyte–macrophages and SSPCs within the injured tissue (Fig. 2c). SAM is decarboxylated to form decarboxylated SAM (dcSAM), which serves as an aminopropyl donor in polyamine synthesis, generating 5′-methylthioadenosine (MTA) as a by-product (Fig. 2d). Targeted metabolomic profiling further demonstrated that, compared to young controls, MSCs and macrophages isolated from aged mice exhibited significantly reduced levels of SAM and S-adenosylhomocysteine (SAH), accompanied by higher levels of dcSAM and MTA (Fig. 2e, f). These metabolic alterations were corroborated by quantitative PCR analysis, which confirmed increased expression of polyamine biosynthetic genes in aged MSCs and macrophages (Fig. 2g, h), indicating a broad, cell type–specific depletion of SAM metabolism in the aged microenvironment. Together, our integrative single-cell transcriptomic and metabolomic analyses uncover previously unrecognized molecular defects in the aged bone regenerative niche. The marked and multicellular depletion of SAM suggests an underappreciated epigenetic–metabolic vulnerability associated with aging, and points to SAM metabolism as a potentially tractable target for restoring epigenetic balance and supporting regenerative responses in aged tissues.Fig. 2Multicellular depletion of SAM metabolism in aged bone regenerative microenvironment.a UMAP visualization of single-cell RNA sequencing (scRNA-seq) data from bone repair sites identified six major cell populations, including B lymphocytes, endothelial cells, monocyte–macrophages, neutrophils, erythrocytes, and SSPCs. b Bubble plot showing representative marker gene expression across the six identified cell clusters. Dot size represents the percentage of cells expressing each gene; color intensity reflects average expression level. c Heatmap illustrating the relative expression levels of polyamine biosynthetic genes (Odc1, Srm, Amd1 and Sms) across distinct cell populations. d Schematic diagram of the polyamine biosynthesis pathway. SAM is decarboxylated to form decarboxylated SAM (dcSAM), which donates aminopropyl groups in polyamine synthesis, producing 5′-methylthioadenosine (MTA) as a by-product—a major SAM-consuming pathway. Targeted metabolomics analysis of macrophages (e) and MSCs (f) from young and aged bone tissue, quantifying intracellular levels of SAM, dcSAM, MTA, and SAH (n = 6 biologically independent samples per group). qPCR analysis of polyamine metabolic genes (Paox, Sat1, Odc1, Srm, Amd1, and Sms) in macrophages (g) and MSCs (h) isolated from young and aged mice (n = 4 mice per group; mean ± SD). Panels g, h were analyzed using unpaired two-tailed Student’s t-test. Source data are provided as a Source Data file. Multicellular depletion of SAM metabolism in aged bone regenerative microenvironment. a UMAP visualization of single-cell RNA sequencing (scRNA-seq) data from bone repair sites identified six major cell populations, including B lymphocytes, endothelial cells, monocyte–macrophages, neutrophils, erythrocytes, and SSPCs. b Bubble plot showing representative marker gene expression across the six identified cell clusters. Dot size represents the percentage of cells expressing each gene; color intensity reflects average expression level. c Heatmap illustrating the relative expression levels of polyamine biosynthetic genes (Odc1, Srm, Amd1 and Sms) across distinct cell populations. d Schematic diagram of the polyamine biosynthesis pathway. SAM is decarboxylated to form decarboxylated SAM (dcSAM), which donates aminopropyl groups in polyamine synthesis, producing 5′-methylthioadenosine (MTA) as a by-product—a major SAM-consuming pathway. Targeted metabolomics analysis of macrophages (e) and MSCs (f) from young and aged bone tissue, quantifying intracellular levels of SAM, dcSAM, MTA, and SAH (n = 6 biologically independent samples per group). qPCR analysis of polyamine metabolic genes (Paox, Sat1, Odc1, Srm, Amd1, and Sms) in macrophages (g) and MSCs (h) isolated from young and aged mice (n = 4 mice per group; mean ± SD). Panels g, h were analyzed using unpaired two-tailed Student’s t-test. Source data are provided as a Source Data file. Design and characterization of nanoconcentrisomes Bioorthogonal nanoconcentrisomes were rationally designed as a spatially compartmentalized dual-vesicle system to enable coordinated cytoplasmic delivery of both a small-molecule epigenetic metabolite and stem cell–derived nanovesicles while minimizing lysosomal trapping (Fig. 3a). Specifically, SMNV are composed of an outer membrane-fusogenic liposomal shell and an inner compartment consisting of MNV. The outer liposomal layer, formulated with 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), DOPE–PEG2000–DBCO, and 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP), is designed to promote membrane fusion–mediated cellular entry, thereby allowing both cargos to access the cytoplasm directly rather than being trapped within endo–lysosomal pathways16. Meanwhile, the inner MNV compartment preserves the intrinsic bioactivity of stem cell–derived vesicles. To achieve stable and spatially defined assembly of this dual-compartment architecture, azide-modified MNV were covalently linked to the outer liposomal shell through bioorthogonal click chemistry. MNV were generated from MSCs metabolically labeled with the azide-modified sugar derivative N-azidoacetylmannosamine tetraacetate (Ac₄ManNAz). Following cellular uptake, Ac₄ManNAz was processed through metabolic glycoengineering and incorporated into membrane glycoconjugates, resulting in stable presentation of azide groups on the MSC membrane. This was further confirmed by the bioorthogonal fluorescence signal from DBCO–FITC (Supplementary Fig. 1). These azide-functionalized MSCs were subsequently subjected to continuous extrusion through porous membranes to produce MNV bearing surface-exposed azide groups. The incorporation of DOPE–PEG2000–DBCO lipids enabled strain-promoted azide–alkyne cycloaddition (SPAAC) between the liposomal shell and MNV, while the PEG chains provided controlled intermembrane spacing, creating an outer compartment suitable for accommodating SAM and maintaining structural stability. SAM-loaded fusogenic liposomes (SFL) were then co-extruded with azide-functionalized MNV to generate SMNV with a bilayer-compartmentalized structure. Dynamic light scattering (DLS) analysis showed that SMNV exhibited a slightly larger hydrodynamic diameter than MNV alone, consistent with the presence of the PEGylated outer liposomal shell and the formation of a dual-compartment structure (Fig. 3b). Cryo-electron microscopy further confirmed the nanoconcentrisome architecture (Fig. 3c), while zeta potential analysis supported the successful formation of the membrane-enveloped structure (Supplementary Fig. 2). Both MNV and SMNV maintained hydrodynamic stability under serum-containing conditions (Fig. 3d). Western blot analysis demonstrated that SMNV retained key regulatory proteins derived from MSCs, including SIRT1, ANXA1, IL-10, and IDO, indicating preservation of intrinsic vesicle bioactivity (Fig. 3e and Supplementary Fig. 3). We further evaluated SAM encapsulation during SMNV preparation and found that both SFL and SMNV exhibited high encapsulation efficiency, with only a modest reduction observed after incorporation of MNV (Fig. 3f). Cellular uptake of SMNV was confirmed by flow cytometry (Fig. 3g). To optimize membrane fusion–mediated delivery, we systematically varied the DOTAP content within the outer liposomal shell. Fluorescence resonance energy transfer analysis revealed that SMNV lacking DOTAP (0%) displayed limited fusion efficiency (11.8%), whereas increasing DOTAP content progressively enhanced fusion, with SMNV containing 50% DOTAP achieving a maximal fusion efficiency of 79.4% (Fig. 3h, i). This formulation was therefore selected for subsequent studies, as it favors fusion-dominated cellular entry, which is expected to minimize lysosomal sequestration and facilitate direct cytoplasmic delivery of both SAM and MNV cargos. Consistent with this mechanism, pharmacological inhibition of endocytosis did not affect SMNV uptake, whereas low-temperature conditions significantly suppressed internalization, further supporting a membrane fusion–mediated process (Fig. 3j and Supplementary Fig. 4). SMNV exhibited excellent biocompatibility, with more than 90% viability observed in MSCs and macrophages at concentrations up to 200 μg/ml. Moreover, no significant cytotoxicity was detected in osteoblasts, osteoclasts, or HUVECs, nor did SMNV induce an acute pro-inflammatory response in macrophages in vitro (Supplementary Figs. 5 and 6). Collectively, these results demonstrate that the rationally engineered dual-compartment SMNV platform enables stable assembly, efficient cytoplasmic delivery, and coordinated functional activity of dual cargos.Fig. 3Preparation of fusogenic nanoconcentrisomes.a Schematic illustration of the stepwise preparation of SMNVs via encapsulation of mesenchymal stem cell-derived nanovesicles (MNVs) within SAM-loaded fusogenic liposomes using bio-orthogonal click chemistry. b Dynamic light scattering (DLS) analysis showing the hydrodynamic size distribution of MNV and SMNV. c Representative cryogenic transmission electron microscopy (cryo-TEM) images of MNV and SMNV. Scale bars: 50 nm. Representative images from three independent experiments. d Stability analysis of MNV and SMNV based on size measurements in complete culture medium at 37 °C over 24 hours. e Western blot analysis comparing the protein profiles of MSC and SMNV. f Quantification of SAM loading efficiency in SFL and SMNV (n = 3 independent experiments; data are presented as mean ± SD). g Cellular uptake of DiI-labeled SMNVs by macrophages analyzed via flow cytometry. DiI: 1,1’-dioctadecyl-3,3,3’,3’-tetramethylindocarbocyanine perchlorate. h Representative confocal microscopy images showing membrane fusion of SMNVs with macrophages, using varying DOPE/DOTAP ratios. Scale bar: 10 μm. i Quantification of SMNV fusion efficiency with macrophages based on different DOPE/DOTAP compositions (n = 3 biologically independent samples per group; mean ± SD). j Flow cytometry analysis of DiI⁺ macrophages following treatment with SMNV at 37 °C or 4 °C, or after pretreatment with endocytosis inhibitors (n = 3 biologically independent samples per group; mean ± SD). Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test unless otherwise indicated. Source data are provided as a Source Data file. Preparation of fusogenic nanoconcentrisomes. a Schematic illustration of the stepwise preparation of SMNVs via encapsulation of mesenchymal stem cell-derived nanovesicles (MNVs) within SAM-loaded fusogenic liposomes using bio-orthogonal click chemistry. b Dynamic light scattering (DLS) analysis showing the hydrodynamic size distribution of MNV and SMNV. c Representative cryogenic transmission electron microscopy (cryo-TEM) images of MNV and SMNV. Scale bars: 50 nm. Representative images from three independent experiments. d Stability analysis of MNV and SMNV based on size measurements in complete culture medium at 37 °C over 24 hours. e Western blot analysis comparing the protein profiles of MSC and SMNV. f Quantification of SAM loading efficiency in SFL and SMNV (n = 3 independent experiments; data are presented as mean ± SD). g Cellular uptake of DiI-labeled SMNVs by macrophages analyzed via flow cytometry. DiI: 1,1’-dioctadecyl-3,3,3’,3’-tetramethylindocarbocyanine perchlorate. h Representative confocal microscopy images showing membrane fusion of SMNVs with macrophages, using varying DOPE/DOTAP ratios. Scale bar: 10 μm. i Quantification of SMNV fusion efficiency with macrophages based on different DOPE/DOTAP compositions (n = 3 biologically independent samples per group; mean ± SD). j Flow cytometry analysis of DiI⁺ macrophages following treatment with SMNV at 37 °C or 4 °C, or after pretreatment with endocytosis inhibitors (n = 3 biologically independent samples per group; mean ± SD). Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test unless otherwise indicated. Source data are provided as a Source Data file. SMNV suppresses inflammation of macrophages and subsequently mediates a pro-resolving phenotype through enhanced efferocytosis The persistent presence of numerous pro-inflammatory macrophages around damaged tissue can hinder tissue repair, emphasizing the critical importance of initially inhibiting inflammation in these macrophages17. SAM has been shown to exert anti-inflammatory effects, in part by inhibiting the binding of H3K4Me3 to the TNF promoter, thereby reducing TNF transcription18. To model inflammatory conditions, we first stimulated macrophages with LPS, and observed that SMNV treatment markedly suppressed TNF protein levels (Supplementary Fig. 7). In addition to their role in cytokine production, macrophages are essential for clearing apoptotic cells (ACs) and secreting pro-resolving mediators that promote tissue repair during the resolution phase of inflammation19. Impaired macrophage efferocytosis capacity reduces dead cell clearance, leading to inflammation accumulation, which limits the transition to an anti-inflammatory, pro-resolving macrophage phenotype20. We first co-cultured macrophages from different treatment groups with apoptotic cells (Fig. 4a). Flow cytometric analysis showed that both the MNV and SMNV groups exhibited significantly enhanced efferocytosis activity compared with the control group (Fig. 4b). The enhanced efferocytosis observed in the MNV-treated group is consistent with the presence of efferocytosis-related regulatory proteins, such as ANXA1, retained in MNV (Fig. 3e)21. Notably, the further increase observed in the SMNV group suggests an additional contribution from SAM. Although SAM is not a canonical efferocytosis ligand, it may modulate macrophage metabolic or epigenetic states in a manner that favors efferocytotic activity, thereby modestly amplifying this process9. Importantly, efferocytosis is known to actively drive the resolution of inflammation by inducing the production of anti-inflammatory and tissue-repair–associated cytokines22. Based on this, we next assessed the secretion of the anti-inflammatory factors TGF-β1 and IL-10. Compared with MNV, SMNV treatment resulted in significantly higher levels of both TGF-β1 and IL-10 secretion (Fig. 4c, d). Mechanistically, this enhanced anti-inflammatory response may be associated with activation of the SAM–COX-2 (Ptgs2)–TGF-β signaling axis9. Consistent with this notion, Quantitative real-time PCR (qPCR) analysis revealed that SMNV treatment significantly increased the mRNA expression of Ptgs2 and Tgfb1 (Supplementary Fig. 8). Collectively, these results indicate that SMNV not only augments anti-inflammatory cytokine production but also modestly enhances macrophage efferocytosis, together contributing to a pro-resolving inflammatory microenvironment.Fig. 4SMNV-mediated inhibition of inflammation in pro-inflammatory macrophages and metabolic reprogramming induced by enhanced efferocytosis triggers a pro-resolving macrophages phenotype.a Schematic illustrating efferocytosis in lipopolysaccharide (LPS)-stimulated pro-inflammatory macrophages after treatment with MNV or SMNV. MAC, macrophages. Created in BioRender. Liang, K. (2026) https://BioRender.com/2xibut3. b Quantification of efferocytosis capacity in macrophages following treatment with SMNV or MNV (n = 4 biologically independent samples per group; data are presented as mean ± SD). ELISA analysis of anti-inflammatory cytokine secretion, including IL-10 (c) and TGF-β1 (d), in macrophages under different treatments (n = 3 biologically independent samples per group; mean ± SD). e Heatmap showing differentially abundant metabolites in LPS-stimulated macrophages treated with AC alone or AC plus SMNV (n = 5 biologically independent samples per group). f Pathway topology analysis of metabolic alterations induced by SMNV, based on Reactome enrichment of metabolomics data. g, Volcano plot showing significant metabolite changes between the AC and AC + SMNV groups in LPS-stimulated macrophages. Statistical significance was determined using a two-sided unpaired Student’s t-test with Benjamini–Hochberg adjustment. h OCR measurements of mitochondrial oxidative phosphorylation (OXPHOS) during efferocytosis using Seahorse XF (n = 3 biologically independent samples per group; mean ± SD). i Quantification of intracellular NAD⁺ levels (left) and NAD⁺/NADH ratio (right, fold change) in LPS-stimulated macrophages treated with AC or AC + SMNV (n = 5 biologically independent samples per group; mean ± SD). j Quantification of SIRT1 activity in macrophages across treatment groups (n = 5 biologically independent samples per group; mean ± SD). k Representative confocal microscopy images of iNOS and ARG1 expression in macrophages following different treatments (Scale bar: 20 μm). Representative flow cytometry histograms (left) and quantification (right) of iNOS⁺ (l) and ARG1⁺ (m) cell populations (n = 3 biologically independent samples per group; mean ± SD). Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test for panels b–d, l, m For panels h–j statistical significance was determined using an unpaired two-tailed Student’s t-test. Source data are provided as a Source Data file. SMNV-mediated inhibition of inflammation in pro-inflammatory macrophages and metabolic reprogramming induced by enhanced efferocytosis triggers a pro-resolving macrophages phenotype. a Schematic illustrating efferocytosis in lipopolysaccharide (LPS)-stimulated pro-inflammatory macrophages after treatment with MNV or SMNV. MAC, macrophages. Created in BioRender. Liang, K. (2026) https://BioRender.com/2xibut3. b Quantification of efferocytosis capacity in macrophages following treatment with SMNV or MNV (n = 4 biologically independent samples per group; data are presented as mean ± SD). ELISA analysis of anti-inflammatory cytokine secretion, including IL-10 (c) and TGF-β1 (d), in macrophages under different treatments (n = 3 biologically independent samples per group; mean ± SD). e Heatmap showing differentially abundant metabolites in LPS-stimulated macrophages treated with AC alone or AC plus SMNV (n = 5 biologically independent samples per group). f Pathway topology analysis of metabolic alterations induced by SMNV, based on Reactome enrichment of metabolomics data. g, Volcano plot showing significant metabolite changes between the AC and AC + SMNV groups in LPS-stimulated macrophages. Statistical significance was determined using a two-sided unpaired Student’s t-test with Benjamini–Hochberg adjustment. h OCR measurements of mitochondrial oxidative phosphorylation (OXPHOS) during efferocytosis using Seahorse XF (n = 3 biologically independent samples per group; mean ± SD). i Quantification of intracellular NAD⁺ levels (left) and NAD⁺/NADH ratio (right, fold change) in LPS-stimulated macrophages treated with AC or AC + SMNV (n = 5 biologically independent samples per group; mean ± SD). j Quantification of SIRT1 activity in macrophages across treatment groups (n = 5 biologically independent samples per group; mean ± SD). k Representative confocal microscopy images of iNOS and ARG1 expression in macrophages following different treatments (Scale bar: 20 μm). Representative flow cytometry histograms (left) and quantification (right) of iNOS⁺ (l) and ARG1⁺ (m) cell populations (n = 3 biologically independent samples per group; mean ± SD). Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test for panels b–d, l, m For panels h–j statistical significance was determined using an unpaired two-tailed Student’s t-test. Source data are provided as a Source Data file. Given the well-established interplay between metabolism and immune function23, we next investigated whether the increased efferocytosis observed in SMNV-treated macrophages was accompanied by metabolic rewiring. To explore this, we performed untargeted metabolomic profiling on efferocytosing macrophages treated with SMNV (Fig. 4e). A total of 355 metabolites were significantly altered (304 upregulated, 41 downregulated; p < 0.05). Principal component analysis (PCA) revealed tight clustering among biological replicates, indicating that SMNV-treated macrophages formed a metabolically distinct population compared to AC-only controls (Supplementary Fig. 9). Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicated that differentially abundant metabolites were highly enriched in arachidonic acid metabolism and polyamine biosynthesis pathways (Supplementary Fig. 10). Complementary pathway mapping using the Reactome database further revealed strong associations with lipid metabolism, fatty acid metabolism, and specifically, arachidonic acid metabolism (Fig. 4f)24. A volcano plot analysis identified a significant increase in long-chain free fatty acids in the SMNV-treated group (Fig. 4g), which are known substrates for fatty acid oxidation (FAO) and may enhance mitochondrial oxidative metabolism20. This hypothesis was supported by oxygen consumption rate (OCR) measurements, which demonstrated that SMNV significantly increased mitochondrial respiration in efferocytosing macrophages (Fig. 4h). Metabolic reprogramming not only fuels cellular bioenergetics but also alters the pool of immunoregulatory metabolites that govern macrophage polarization25. We found that SMNV treatment elevated intracellular NAD⁺ levels and the NAD⁺/NADH ratio (Fig. 4i), changes that are known to activate the sirtuin family of deacetylases, particularly SIRT1. Consistently, SIRT1 activity was significantly increased in SMNV-treated macrophages (Fig. 4j), suggesting a mechanistic link between enhanced oxidative metabolism and the induction of anti-inflammatory pathways. To further characterize functional consequences of this metabolic shift, we assessed macrophage polarization. SMNV treatment inhibited expression of the pro-inflammatory marker iNOS and upregulated the anti-inflammatory marker ARG1 (Fig. 4k). These findings were corroborated by flow cytometry, which confirmed decreased iNOS and increased ARG1 expression (Fig. 4l, m). qPCR analysis further revealed reduced expression of Il1b and Cd86, and elevated expression of Il10 and Cd206 (Supplementary Fig. 11). In summary, our findings underscore a close link between macrophage functional reprogramming and metabolic remodeling, characterized by shifts in key immunoregulatory metabolites that support a pro-resolving phenotype. Efferocytosing macrophages treated with SMNV provide a regenerative niche We next investigated how SMNV-treated macrophages influence other cell types within the bone microenvironment to facilitate tissue regeneration14. To this end, we collected conditioned medium (CM) from SMNV-treated macrophages and assessed its effects on angiogenesis (Fig. 5a). CM from SMNV-treated macrophages significantly promoted both the migration and tube formation of HUVECs (Fig. 5b, c). Furthermore, CD31 immunostaining demonstrated a marked increase in CD31 expression in HUVECs treated with SMNV-CM (Fig. 5d). In line with this, qPCR analysis revealed significantly elevated mRNA expression of angiogenesis-related genes, including Vegfa, Pecam1, and Vwf, in the SMNV-CM group (Fig. 5e). We next examined the impact of SMNV-treated macrophages on osteogenic and osteoclastic activity (Fig. 5f). SMNV-CM significantly enhanced the osteogenic differentiation potential of MSCs. Alkaline phosphatase (ALP) staining showed markedly increased ALP activity in the AC + SMNV group after 7 days of culture (Fig. 5g, h). After 14 days, Alizarin Red S (ARS) staining revealed abundant calcium nodule formation in the AC + SMNV group, and quantitative analysis confirmed significantly elevated calcium deposition (Fig. 5g, i). In parallel, qPCR results showed that the expression of osteogenesis-related genes — including Opn, Ocn, Runx2, and Alpl — was significantly upregulated in the AC + SMNV group (Fig. 5j). Finally, we investigated the influence of SMNV-treated macrophages on osteoclast differentiation. qPCR analysis demonstrated that the expression of key osteoclast-related genes — c-Fos, Acp5, Ctsk, and Nfatc1 — was significantly downregulated following culture with SMNV-CM (Fig. 5k). To further elucidate the mechanisms by SMNV-CM promotes angiogenesis, osteogenesis, and osteoclast regulation, we systematically analyzed changes in key soluble mediators in the macrophage-conditioned medium. Compared with the control group, SMNV treatment markedly reduced the levels of the pro-inflammatory cytokines IL-6 and TNF-α, while significantly increasing the levels of the anti-inflammatory and pro-regenerative factors IL-10, TGF-β, and VEGF (Supplementary Fig. 12a). IL-10 and TGF-β are well-established mediators that promote osteogenic differentiation and inhibit osteoclast formation, whereas VEGF plays a central role in endothelial cell proliferation and angiogenesis. To further delineate the functional contributions of individual cytokines within SMNV-CM, cytokine receptor blockade experiments were performed. Inhibition of TGF-β signaling using the TGF-β receptor inhibitor SB431542 significantly attenuated the pro-osteogenic effects of SMNV-CM, as evidenced by decreased expression of the osteogenic marker genes Alpl and Runx2 (Supplementary Fig. 12b). In contrast, TNF-α is a potent stimulator of osteoclastogenesis. Consistent with this, supplementation of SMNV-CM with recombinant TNF-α restored osteoclast activity and significantly upregulated the expression of osteoclast-related genes, including Acp5 and Ctsk (Supplementary Fig. 12c), confirming that reduced TNF-α levels are a key determinant of the inhibitory effect on osteoclastogenesis mediated by SMNV-CM. Conversely, blockade of VEGF signaling with the VEGFR inhibitor Ki8751 suppressed endothelial angiogenic activity, leading to reduced expression of key vascular marker genes Pecam1 and Vwf (Supplementary Fig. 12d). Collectively, these results indicate that SMNV modulates macrophage function in a manner that promotes paracrine-mediated angiogenic and osteogenic responses, while suppressing osteoclastogenesis. These effects support the formation of a regenerative bone microenvironment.Fig. 5Pro-regenerative immune niche produced by efferocytosing macrophages following SMNV treatment.a Schematic illustrating the paracrine regulatory effects of SMNV-induced macrophage efferocytosis on downstream endothelial cells (HUVECs). Created in BioRender. Liang, K. (2026) https://BioRender.com/yyz3j4k. b Representative images (left) and quantification (right) of HUVEC migration assessed by scratch assay (n = 4 biologically independent samples per group; data are presented as mean ± SD). Scale bar: 200 μm. c Representative images (left) and quantification (right) of HUVEC tube formation on Matrigel (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 50 μm. d Representative confocal images (left) and quantification (right) of CD31 expression in HUVECs (n = 3 biologically independent samples per group; mean ± SD). Scale bar: 10 μm. e qPCR analysis of angiogenic genes including Vegfa, Pecam1, and Vwf in HUVECs (n = 3 biologically independent samples per group; mean ± SD). f Schematic illustrating the paracrine effects of efferocytosing macrophages on MSCs and osteoclasts (OCs). Created in BioRender. Liang, K. (2026) https://BioRender.com/mtnnb4w. Representative images (g) and corresponding quantification of ALP staining (h) and ARS staining (i) in MSCs after treatment with conditioned medium (n = 4 biologically independent samples per group; mean ± SD). j RT–qPCR analysis of osteogenic genes Opn, Ocn, Runx2, and Alpl in MSCs (n = 3 biologically independent samples per group; mean ± SD). k RT–qPCR analysis of osteoclast-related genes Acp5, Ctsk, Nfatc1, and c-Fos in OCs (n = 3 biologically independent samples per group; mean ± SD). Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test for panels d, e, h–k. For panels b, c statistical significance was determined using an unpaired two-tailed Student’s t-test. Source data are provided as a Source Data file. Pro-regenerative immune niche produced by efferocytosing macrophages following SMNV treatment. a Schematic illustrating the paracrine regulatory effects of SMNV-induced macrophage efferocytosis on downstream endothelial cells (HUVECs). Created in BioRender. Liang, K. (2026) https://BioRender.com/yyz3j4k. b Representative images (left) and quantification (right) of HUVEC migration assessed by scratch assay (n = 4 biologically independent samples per group; data are presented as mean ± SD). Scale bar: 200 μm. c Representative images (left) and quantification (right) of HUVEC tube formation on Matrigel (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 50 μm. d Representative confocal images (left) and quantification (right) of CD31 expression in HUVECs (n = 3 biologically independent samples per group; mean ± SD). Scale bar: 10 μm. e qPCR analysis of angiogenic genes including Vegfa, Pecam1, and Vwf in HUVECs (n = 3 biologically independent samples per group; mean ± SD). f Schematic illustrating the paracrine effects of efferocytosing macrophages on MSCs and osteoclasts (OCs). Created in BioRender. Liang, K. (2026) https://BioRender.com/mtnnb4w. Representative images (g) and corresponding quantification of ALP staining (h) and ARS staining (i) in MSCs after treatment with conditioned medium (n = 4 biologically independent samples per group; mean ± SD). j RT–qPCR analysis of osteogenic genes Opn, Ocn, Runx2, and Alpl in MSCs (n = 3 biologically independent samples per group; mean ± SD). k RT–qPCR analysis of osteoclast-related genes Acp5, Ctsk, Nfatc1, and c-Fos in OCs (n = 3 biologically independent samples per group; mean ± SD). Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test for panels d, e, h–k. For panels b, c statistical significance was determined using an unpaired two-tailed Student’s t-test. Source data are provided as a Source Data file. SMNV rescues senescent stem cells by restoring heterochromatin marker levels The aged bone microenvironment induces senescence in MSCs, thereby affecting their proliferation and differentiation, ultimately hindering tissue regeneration26. Consequently, we initially investigated whether SMNV could inhibit stem cell senescence. Senescence-associated β-galactosidase assays revealed that SMNV significantly inhibits doxorubicin (Doxo)-induced cellular senescence, whereas MNV shows only a slight inhibitory effect (Fig. 6a). Similarly, staining for the DNA damage marker γ-H2AX and the cell cycle regulatory protein p21 further supports this conclusion (Fig. 6b–d). Additionally, cell viability assays demonstrated that SMNV significantly enhances stem cell viability (Fig. 6e). Recent studies have shown that chromatin structure alterations are closely linked to cellular dysfunction during aging27. Inadequate chromatin compaction often leads to genomic instability, which results in increased DNA damage and cellular senescence28. SAM, the principal methyl donor for histone and DNA methylation, directly influences chromatin modification29. Thus, we investigated whether SMNV could rejuvenate aging stem cells by restoring intracellular SAM levels and promoting heterochromatin formation10. To test this hypothesis, senescent MSCs were treated with SMNV or MNV, and intracellular SAM levels were quantified. SMNV treatment resulted in a significant elevation of intracellular SAM compared with MNV (Fig. 6f), exhibiting a time-dependent increase that reached a plateau at approximately 6 h (Supplementary Fig. 13). We next examined whether restoration of intracellular SAM could reinforce heterochromatin integrity in senescent MSCs. Consistent with this hypothesis, SMNV treatment markedly increased the levels of heterochromatin-associated markers, including H3K9me3 and HP1α, whereas MNV alone exerted minimal effects (Fig. 6g–l). To establish a causal link between SAM availability and heterochromatin remodeling, we performed pharmacological inhibition using UNC0642, a selective inhibitor of the H3K9 methyltransferases G9a/GLP. Despite elevated intracellular SAM levels, inhibition of G9a/GLP largely abolished the SMNV-induced increase in H3K9me3 and significantly attenuated HP1α accumulation (Supplementary Fig. 14), indicating that the epigenetic effects of SMNV are dependent on SAM-driven H3K9 methylation rather than nonspecific metabolic alterations. To further assess whether these histone modifications were accompanied by broader chromatin remodeling, we performed genome-wide chromatin accessibility profiling of aged MSCs following SMNV treatment. Compared with untreated aged MSCs, SMNV-treated cells exhibited a global reduction in accessibility peak density, along with decreased chromatin accessibility around transcription start sites (TSSs), suggesting widespread chromatin compaction (Supplementary Fig. 15a, b). In parallel, the expression of transposable elements (TEs) was significantly reduced in SMNV-treated aged MSCs relative to aged controls (Supplementary Fig. 15c), consistent with improved chromatin stability and suppression of age-associated epigenetic drift. We subsequently examined the expression of pro-inflammatory genes and observed that SMNV treatment significantly reduced their mRNA levels (Supplementary Fig. 16). We propose that SAM supplementation reinforces heterochromatin formation, resulting in a more compact chromatin architecture that restricts the accessibility of inflammatory gene loci and suppresses their transcription18. This interpretation is supported by our genome-wide chromatin accessibility analysis, which revealed a global reduction in accessible chromatin regions and decreased promoter openness in SMNV-treated aged MSCs (Supplementary Fig. 15a, c). Beyond transcriptional repression, enhanced chromatin integrity may also contribute to improved genomic stability. Age-associated heterochromatin erosion has been linked to increased DNA damage and aberrant cytosolic DNA leakage, which can activate innate immune sensing pathways such as cGAS–STING. Thus, by restoring heterochromatin structure, SMNV treatment may indirectly attenuate DNA damage–induced cGAS–STING activation, further contributing to the suppression of pro-inflammatory signaling10,30. Consistent with this notion, ELISA analysis demonstrated that SMNV markedly reduced the secretion of senescence-associated secretory phenotype (SASP) factors, including IL-1β and IL-6 (Supplementary Fig. 17). Taken together, these results support a model in which SMNV elevates intracellular SAM levels in senescent MSCs, promotes SAM-dependent H3K9 methylation and heterochromatin formation, stabilizes global chromatin architecture, and suppresses inflammatory signaling, thereby contributing to the functional rejuvenation of aged MSCs.Fig. 6SMNV promotes heterochromatin formation in senescent stem cells to rescue stem cell dysfunction.a Representative SA-β-gal staining images (left) and corresponding quantification (right) in Doxo (200 nM)-induced senescent MSCs treated with SMNV or MNV (n = 3 biologically independent samples per group; data are presented as mean ± SD). Scale bar: 50 μm. b Representative γ-H2AX immunofluorescence images (left) and quantification (right) in Doxo-induced MSCs after indicated treatments (n = 8 biologically independent samples per group; mean ± SD). Scale bar: 5 μm. Representative images (c) and quantification (d) of p21 expression in MSCs (n = 8 biologically independent samples per group; mean ± SD). Scale bar: 5 μm. e Cell proliferation of MSCs assessed by CCK-8 assay following different treatments (n = 4 biologically independent samples per group; mean ± SD). f Quantification of intracellular S-adenosylmethionine (SAM) levels in MSCs by ELISA (n = 3 biologically independent samples per group; mean ± SD). Representative confocal immunofluorescence images of heterochromatin markers H3K9me3 (g), H3K9me2 (h), and HP1α (i), and corresponding quantification of their fluorescence intensities (j–l) in treated MSCs (n = 8 biologically independent samples per group; mean ± SD). Scale bars: 5 μm. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test. Source data are provided as a Source Data file. SMNV promotes heterochromatin formation in senescent stem cells to rescue stem cell dysfunction. a Representative SA-β-gal staining images (left) and corresponding quantification (right) in Doxo (200 nM)-induced senescent MSCs treated with SMNV or MNV (n = 3 biologically independent samples per group; data are presented as mean ± SD). Scale bar: 50 μm. b Representative γ-H2AX immunofluorescence images (left) and quantification (right) in Doxo-induced MSCs after indicated treatments (n = 8 biologically independent samples per group; mean ± SD). Scale bar: 5 μm. Representative images (c) and quantification (d) of p21 expression in MSCs (n = 8 biologically independent samples per group; mean ± SD). Scale bar: 5 μm. e Cell proliferation of MSCs assessed by CCK-8 assay following different treatments (n = 4 biologically independent samples per group; mean ± SD). f Quantification of intracellular S-adenosylmethionine (SAM) levels in MSCs by ELISA (n = 3 biologically independent samples per group; mean ± SD). Representative confocal immunofluorescence images of heterochromatin markers H3K9me3 (g), H3K9me2 (h), and HP1α (i), and corresponding quantification of their fluorescence intensities (j–l) in treated MSCs (n = 8 biologically independent samples per group; mean ± SD). Scale bars: 5 μm. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test. Source data are provided as a Source Data file. SMNV rescues the osteogenic differentiation ability of senescent stem cells by enhancing mitochondrial function To optimize bone regeneration, effective cues that direct stem cell osteogenic differentiation are essential. We first assessed whether SMNV could restore the osteogenic potential of aged stem cells in vitro. ALP staining and ARS staining, representing early- and late-stage osteogenic differentiation, respectively, demonstrated that SMNV markedly enhanced osteogenic differentiation in senescent stem cells31. MNV alone also improved osteogenic capacity, albeit to a lesser extent than SMNV (Fig. 7a, b), suggesting an additive benefit conferred by SAM-mediated rejuvenation. These observations were further corroborated by immunofluorescence staining and qPCR analysis of osteogenic markers (Supplementary Figs. 18, 19), confirming the restoration of osteogenic function in a physiologically relevant environment. Beyond osteogenesis, we next evaluated whether SMNV broadly restored the multilineage differentiation capacity of aged stem cells—a key hallmark of stem cell functionality. Compared with untreated aged controls, SMNV-treated stem cells exhibited enhanced chondrogenic differentiation, as indicated by increased toluidine blue O staining, as well as improved adipogenic differentiation evidenced by oil red O staining (Supplementary Fig. 20). These findings indicate that SMNV treatment promotes a global restoration of lineage potential rather than a lineage-restricted effect. To validate these findings in vivo, aged stem cells were pretreated with SMNV and subsequently transplanted into a bone defect model. Strikingly, defects receiving SMNV-pretreated aged stem cells displayed substantially enhanced bone regeneration and more effective defect closure compared with defects transplanted with untreated aged stem cells, which exhibited limited repair (Supplementary Fig. 21).Fig. 7SMNV rescues osteogenic function by promoting mitochondrial function in senescent stem cells.a Representative images (left) and quantification (right) of ALP staining in MSCs following different treatments (n = 4 biologically independent samples per group; data are presented as mean ± SD). Scale bar: 100 μm. b Representative images (left) and quantification (right) of ARS staining in MSCs (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 100 μm. c Representative images (left) and quantification (right) of MitoSOX staining to assess mitochondrial reactive oxygen species (ROS) levels in MSCs after treatment (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 50 μm. d Representative Tetramethylrhodamine Ethyl Ester (TMRM) staining images (left) and quantification (right) reflecting mitochondrial membrane potential in treated MSCs (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 50 μm. e Quantification of intracellular ATP levels in MSCs (n = 3 biologically independent samples per group; mean ± SD). f Schematic overview of mitochondrial isolation workflow from MSCs. Created in BioRender. Liang, K. (2026) https://BioRender.com/kc214o7. g High-resolution respirometry analysis of mitochondrial complex I–IV (CI–CIV) function in MSCs using the Oroboros system. Oxygen consumption rate (OCR) was normalized to total protein content (n = 3 biologically independent samples per group; mean ± SD). h Quantification of mitochondrial NAD⁺ content in MSCs across treatment groups (n = 3 biologically independent samples per group; mean ± SD). i Heatmap of differentially expressed genes in MSCs between the Doxo + SMNV group and the Doxo group. n = 3 biologically independent samples. j Volcano plot showing gene expression changes in MSCs between the Doxo + SMNV and Doxo groups. k KEGG pathway enrichment analysis of differentially expressed genes between groups. l GSEA comparing gene signatures related to oxidative phosphorylation and collagen formation between the Doxo + SMNV and Doxo groups. Statistical comparisons were performed using one-way ANOVA with Tukey’s multiple comparisons test unless otherwise indicated. Source data are provided as a Source Data file. SMNV rescues osteogenic function by promoting mitochondrial function in senescent stem cells. a Representative images (left) and quantification (right) of ALP staining in MSCs following different treatments (n = 4 biologically independent samples per group; data are presented as mean ± SD). Scale bar: 100 μm. b Representative images (left) and quantification (right) of ARS staining in MSCs (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 100 μm. c Representative images (left) and quantification (right) of MitoSOX staining to assess mitochondrial reactive oxygen species (ROS) levels in MSCs after treatment (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 50 μm. d Representative Tetramethylrhodamine Ethyl Ester (TMRM) staining images (left) and quantification (right) reflecting mitochondrial membrane potential in treated MSCs (n = 4 biologically independent samples per group; mean ± SD). Scale bar: 50 μm. e Quantification of intracellular ATP levels in MSCs (n = 3 biologically independent samples per group; mean ± SD). f Schematic overview of mitochondrial isolation workflow from MSCs. Created in BioRender. Liang, K. (2026) https://BioRender.com/kc214o7. g High-resolution respirometry analysis of mitochondrial complex I–IV (CI–CIV) function in MSCs using the Oroboros system. Oxygen consumption rate (OCR) was normalized to total protein content (n = 3 biologically independent samples per group; mean ± SD). h Quantification of mitochondrial NAD⁺ content in MSCs across treatment groups (n = 3 biologically independent samples per group; mean ± SD). i Heatmap of differentially expressed genes in MSCs between the Doxo + SMNV group and the Doxo group. n = 3 biologically independent samples. j Volcano plot showing gene expression changes in MSCs between the Doxo + SMNV and Doxo groups. k KEGG pathway enrichment analysis of differentially expressed genes between groups. l GSEA comparing gene signatures related to oxidative phosphorylation and collagen formation between the Doxo + SMNV and Doxo groups. Statistical comparisons were performed using one-way ANOVA with Tukey’s multiple comparisons test unless otherwise indicated. Source data are provided as a Source Data file. We next investigated the cellular mechanisms underlying the restored osteogenic capacity. Given that stem cell–derived vesicles have been reported to enhance mitochondrial function32, we examined mitochondrial status following SMNV treatment. Both MNV and SMNV significantly reduced mitochondrial reactive oxygen species (ROS), restored mitochondrial membrane potential, and increased intracellular ATP levels in senescent stem cells (Fig. 7c–e). To further characterize mitochondrial metabolic recovery, mitochondria were isolated from cultured stem cells for functional analysis (Fig. 7f). Both MNV and SMNV treatments improved the activity of mitochondrial complexes I and II, with a trend toward enhanced complex IV activity (Fig. 7g). Consistently, mitochondrial NAD⁺ levels—which are critical for the tricarboxylic acid cycle and oxidative phosphorylation—were significantly increased following MNV or SMNV treatment in Doxo-induced senescent cells33 (Fig. 7h). To comprehensively define transcriptional changes associated with SMNV-mediated rejuvenation, we performed transcriptomic analysis on senescent stem cells treated with SMNV (Fig. 7i). Differential expression analysis revealed downregulation of senescence-associated genes, including Il6 and Cdkn1a, alongside upregulation of genes involved in mitochondrial function and oxidative phosphorylation, such as Ndufs8 and Ndufb6 (Fig. 7j). KEGG pathway enrichment analysis highlighted alterations in cell cycle and p53 signaling pathways related to cellular senescence (Fig. 7k), while Gene Set Enrichment Analysis (GSEA) revealed enrichment of oxidative phosphorylation and collagen formation pathways associated with osteogenic mineralization (Fig. 7l). Collectively, these results indicate that SMNV restores the osteogenic and regenerative capacity of aged stem cells by reprogramming mitochondrial metabolism and rejuvenating transcriptional programs associated with cellular senescence. SMNV promotes the repair of critical bone defects in aged rats Given the pronounced anti-inflammatory and anti-senescent effects of SMNV observed in vitro, we further evaluated whether SMNV could promote bone regeneration at defect sites in aged rats (Fig. 8a). To achieve sustained release of SMNV in vivo, we selected a biocompatible and stable photocurable gelatin hydrogel with highly controllable gelation properties to construct the SMNV@Gel hydrogel drug delivery system (Supplementary Fig. 22). Scanning electron microscopy (SEM) revealed a porous and interconnected three-dimensional network structure that facilitates the loading and sustained release of nanovesicles (Supplementary Fig. 23a). To verify SMNV encapsulation, DiD-labeled SMNV was incorporated into the hydrogel and visualized by confocal microscopy, showing a uniform fluorescence distribution throughout the hydrogel matrix, indicative of successful loading and homogeneous dispersion (Supplementary Fig. 23b). Furthermore, swelling and mechanical property analyses demonstrated that the hydrogel possesses excellent hydration capacity and structural stability under physiological conditions (Supplementary Fig. 23c–f). These results collectively confirm the successful construction of the SMNV@Gel system and support its suitability as a stable in vivo delivery platform for sustained SMNV release. Four groups were established for the in vivo study (Control, Gel, MNV@Gel, SMNV@Gel) to assess the therapeutic mechanism and performance of the hydrogel system in vivo. To characterize the in vivo release behavior of the cargo-loaded nanovesicles, ICG-labeled nanovesicles were encapsulated within the hydrogel and implanted into the defect sites as a fluorescent tracer. Longitudinal in vivo imaging revealed that the nanovesicles were gradually and persistently released and retained within the defect region over time, indicating sustained local cargo delivery to support tissue repair (Fig. 8b). The in vivo biocompatibility of the hydrogel system was further evaluated by blood biochemical analysis. No significant alterations were detected in circulating inflammatory cytokines (IL-6, TNF-α), liver function markers (ALT, AST, total bilirubin), or kidney function indicators (blood urea nitrogen, creatinine, creatine kinase) (Supplementary Figs. 24–25). These findings indicate that the SMNV@Gel system does not induce detectable systemic toxicity and exhibits favorable in vivo biocompatibility.Fig. 8Hydrogel reservoirs loaded with SMNV restore aged bone regeneration capacity.a Schematic of the in vivo study design. SMNV-loaded hydrogel (SMNV@Gel) was applied to critical-sized femoral defects in aged rats to assess bone regeneration efficacy. Created in BioRender. Liang, K. (2026) https://BioRender.com/kgv854a. b Representative images and quantification of in vivo fluorescence intensity at the rat defect site at different time points (n = 3 biologically independent samples per group; mean ± SD). Representative flow cytometry plots (c) and corresponding quantification of M1 (CD86⁺) and M2 (CD206⁺) macrophage polarization in callus tissues at day 7 post-treatment. Macrophages were defined as CD45⁺ CD11b⁺ CD68⁺ cells (n = 4 biologically independent samples per group; mean ± SD). d ELISA quantification of pro-inflammatory cytokines (TNF-α, IL-6, IL-1β) and anti-inflammatory cytokines (IL-4, IL-10) in defect tissues at day 7 (n = 4 biologically independent samples per group; mean ± SD). e Representative 3D reconstructed μCT images showing bone regeneration in each treatment group at day 56 (n = 4 biologically independent animals per group). f Quantification of μCT parameters, including bone volume/total volume (BV/TV), trabecular thickness, and trabecular number at day 56 post-treatment (n = 4 per group; mean ± SD). g Representative hematoxylin and eosin (H&E) and Masson’s trichrome staining images of defect regions in different groups. NB, new bone (n = 4 biologically independent samples per group). Scale bar: 400 μm. h Representative immunofluorescence images of Col1a1 and osteocalcin (Ocn) expression in defect sites (n = 4 biologically independent samples per group). Scale bar: 50 μm. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test. Source data are provided as a Source Data file. Hydrogel reservoirs loaded with SMNV restore aged bone regeneration capacity. a Schematic of the in vivo study design. SMNV-loaded hydrogel (SMNV@Gel) was applied to critical-sized femoral defects in aged rats to assess bone regeneration efficacy. Created in BioRender. Liang, K. (2026) https://BioRender.com/kgv854a. b Representative images and quantification of in vivo fluorescence intensity at the rat defect site at different time points (n = 3 biologically independent samples per group; mean ± SD). Representative flow cytometry plots (c) and corresponding quantification of M1 (CD86⁺) and M2 (CD206⁺) macrophage polarization in callus tissues at day 7 post-treatment. Macrophages were defined as CD45⁺ CD11b⁺ CD68⁺ cells (n = 4 biologically independent samples per group; mean ± SD). d ELISA quantification of pro-inflammatory cytokines (TNF-α, IL-6, IL-1β) and anti-inflammatory cytokines (IL-4, IL-10) in defect tissues at day 7 (n = 4 biologically independent samples per group; mean ± SD). e Representative 3D reconstructed μCT images showing bone regeneration in each treatment group at day 56 (n = 4 biologically independent animals per group). f Quantification of μCT parameters, including bone volume/total volume (BV/TV), trabecular thickness, and trabecular number at day 56 post-treatment (n = 4 per group; mean ± SD). g Representative hematoxylin and eosin (H&E) and Masson’s trichrome staining images of defect regions in different groups. NB, new bone (n = 4 biologically independent samples per group). Scale bar: 400 μm. h Representative immunofluorescence images of Col1a1 and osteocalcin (Ocn) expression in defect sites (n = 4 biologically independent samples per group). Scale bar: 50 μm. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test. Source data are provided as a Source Data file. Because chronic inflammation is a major barrier to bone repair in aged tissues, we next examined whether SMNV@Gel could modulate the early inflammatory microenvironment at the defect site. Flow cytometric analysis revealed that SMNV@Gel treatment significantly reduced the proportion of pro-inflammatory M1 macrophages while increasing the proportion of anti-inflammatory M2 macrophages compared with control groups (Fig. 8c), consistent with the in vitro immunomodulatory effects of SMNV. In parallel, ELISA analysis of injured tissues collected during the early inflammatory phase demonstrated that SMNV@Gel markedly suppressed the expression of pro-inflammatory cytokines TNF-α, IL-6, and IL-1β, while significantly elevating the levels of anti-inflammatory cytokines IL-4 and IL-10 (Fig. 8d). These results indicate that sustained SMNV release effectively remodels the local immune microenvironment, facilitating a timely transition from inflammation to regeneration. To assess long-term bone repair outcomes, calvarial defects were analyzed at day 56 post-treatment using micro-computed tomography (µCT) and histological evaluation. µCT analysis revealed minimal bone regeneration in untreated control defects, confirming the limited endogenous repair capacity under osteoporotic conditions (Fig. 8e). In contrast, SMNV@Gel treatment resulted in near-complete defect healing, with significantly increased bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular number (Tb.N) compared with other groups (Fig. 8f). Histological analyses further corroborated these findings. H&E staining showed robust new bone formation surrounding the implants in the SMNV@Gel group, while Masson’s trichrome staining revealed enhanced collagen deposition and matrix organization (Fig. 8g). Moreover, immunofluorescence staining demonstrated significantly higher numbers of OCN- and Col1a1-positive osteoblasts in the SMNV@Gel group compared with all other groups (Fig. 8h). Collectively, these results demonstrate that SMNV@Gel promotes bone regeneration in aged defects by coordinating sustained local delivery, early immunomodulation, and enhanced osteogenic tissue formation across multiple stages of the regenerative process. In vivo cellular mechanisms of SMNV-mediated regeneration To validate the in vivo cellular mechanisms underlying SMNV-mediated tissue regeneration, we performed fluorescence-activated cell sorting (FACS) on callus tissues harvested from aged rats treated with SMNV@Gel or control formulations (Fig. 9a, b). Based on the distinct temporal and functional roles of stem cells and immune cells during regeneration, MSCs and macrophages were isolated at different time points for mechanistic analysis. At day 14 post-treatment, CD90⁺ CD45⁻ MSCs were sorted to evaluate whether SMNV restores stem cell epigenetic integrity in vivo. Immunofluorescence staining revealed that SMNV@Gel treatment markedly enhanced heterochromatin condensation, as evidenced by increased levels of the repressive histone marks H3K9me3 (Fig. 9c and Supplementary Fig. 26) and H3K9me2 (Fig. 9d and Supplementary Fig. 27). Consistently, the mean fluorescence intensity (MFI) of the heterochromatin-associated protein HP1α was significantly elevated (Fig. 9e and Supplementary Fig. 28), while expression of the senescence marker p21 was reduced in SMNV-treated MSCs (Fig. 9f and Supplementary Fig. 29). These results indicate that SMNV promotes epigenetic restoration and alleviates cellular senescence in aged MSCs in vivo. In parallel, to investigate the immunomodulatory effects of SMNV during the early inflammatory phase, CD11b⁺ CD68⁺ CD45⁺ macrophages were isolated at day 7 post-treatment (Fig. 9g). Functional efferocytosis assays demonstrated a pronounced enhancement in apoptotic cell uptake in the SMNV@Gel group (Fig. 9h), suggesting accelerated inflammation resolution. Flow cytometric analysis further showed that SMNV significantly suppressed COX2 expression while upregulating TGF-β1 levels in macrophages (Fig. 9i, j). Notably, these changes are consistent with a SAM–COX2–TGF-β regulatory axis, in which increased intracellular SAM availability restrains COX2 expression and promotes a shift toward a pro-resolving, TGF-β–dominant macrophage phenotype9. ELISA results further confirmed these findings, demonstrating that SMNV treatment reduced the secretion of pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β) while increasing the production of anti-inflammatory mediators (TGF-β and IL-10) in macrophages (Fig. 9k). Collectively, these results indicate that SMNV coordinates metabolic and transcriptional regulation to simultaneously restore the epigenetic integrity of stem cells and reprogram macrophages toward a pro-regenerative phenotype.Fig. 9SMNV restores epigenetic integrity in stem cells and enhances efferocytosis and immune modulation in macrophages in vivo.a Schematic of in vivo mechanistic analysis. Aged rats with skull defects received local SMNV@Gel treatment. Callus tissues were harvested on day 14 for stem cell analysis and on day 7 for macrophage analysis, followed by FACS-based cell sorting and downstream epigenetic and functional assays. Created in BioRender. Liang, K. (2026) https://BioRender.com/kgv854a. b Representative flow cytometry gating strategy used to isolate CD90⁺ CD45⁻ MSCs and CD11b⁺ CD68⁺ CD45⁺ macrophages from callus tissue. Representative immunofluorescence images showing increased expression of heterochromatin markers H3K9me3 (c) and H3K9me2 (d) in MSCs isolated at day 14. Scale bars, 10 µm. Quantification of mean fluorescence intensity (MFI) of HP1α (e heterochromatin structural protein) and p21 (f senescence marker) in sorted MSCs (n = 4 biologically independent samples). g Schematic of efferocytosis assay. Macrophages sorted from defect sites at day 7 were co-cultured with ACs for 1 hour to assess phagocytic capacity. Created in BioRender. Liang, K. (2026) https://BioRender.com/tide0ex. h Representative flow cytometry plots and quantification of efferocytosis index in macrophages, showing enhanced apoptotic cell clearance in the SMNV@Gel group (n = 4 biologically independent samples). Flow cytometric analysis of COX2 (i pro-inflammatory marker) and TGF-β1 (j pro-resolving factor) expression in sorted macrophages. Bar graphs show MFI quantification (n = 4 biologically independent samples). k Cytokine profiling of sorted macrophages via ELISA, showing reduced pro-inflammatory cytokines (TNF-α, IL-6, IL-1β) and increased anti-inflammatory cytokines (TGF-β, IL-10) in the SMNV@Gel group (n = 4 biologically independent samples). Data are presented as mean ± SD; statistical comparisons were performed using one-way ANOVA with Tukey’s multiple comparisons test. Source data are provided as a Source Data file. SMNV restores epigenetic integrity in stem cells and enhances efferocytosis and immune modulation in macrophages in vivo. a Schematic of in vivo mechanistic analysis. Aged rats with skull defects received local SMNV@Gel treatment. Callus tissues were harvested on day 14 for stem cell analysis and on day 7 for macrophage analysis, followed by FACS-based cell sorting and downstream epigenetic and functional assays. Created in BioRender. Liang, K. (2026) https://BioRender.com/kgv854a. b Representative flow cytometry gating strategy used to isolate CD90⁺ CD45⁻ MSCs and CD11b⁺ CD68⁺ CD45⁺ macrophages from callus tissue. Representative immunofluorescence images showing increased expression of heterochromatin markers H3K9me3 (c) and H3K9me2 (d) in MSCs isolated at day 14. Scale bars, 10 µm. Quantification of mean fluorescence intensity (MFI) of HP1α (e heterochromatin structural protein) and p21 (f senescence marker) in sorted MSCs (n = 4 biologically independent samples). g Schematic of efferocytosis assay. Macrophages sorted from defect sites at day 7 were co-cultured with ACs for 1 hour to assess phagocytic capacity. Created in BioRender. Liang, K. (2026) https://BioRender.com/tide0ex. h Representative flow cytometry plots and quantification of efferocytosis index in macrophages, showing enhanced apoptotic cell clearance in the SMNV@Gel group (n = 4 biologically independent samples). Flow cytometric analysis of COX2 (i pro-inflammatory marker) and TGF-β1 (j pro-resolving factor) expression in sorted macrophages. Bar graphs show MFI quantification (n = 4 biologically independent samples). k Cytokine profiling of sorted macrophages via ELISA, showing reduced pro-inflammatory cytokines (TNF-α, IL-6, IL-1β) and increased anti-inflammatory cytokines (TGF-β, IL-10) in the SMNV@Gel group (n = 4 biologically independent samples). Data are presented as mean ± SD; statistical comparisons were performed using one-way ANOVA with Tukey’s multiple comparisons test. Source data are provided as a Source Data file. Discussion In summary, we developed a compartmentalized nanoconcentrisome platform capable of delivering regulatory signals to rejuvenate the aged regenerative niche and guide precise bone repair. By integrating epigenetic modulation with immune and metabolic reprogramming, this strategy addresses a key bottleneck in aging-related regenerative failure. Looking forward, the modular design of this system offers opportunities for further refinement—such as the incorporation of environment-responsive release mechanisms and the development of more sophisticated nanostructures to enable multi-tiered cellular regulation. With its versatility and precision, this platform holds broad translational potential for treating a range of complex, multicellular diseases by dynamically modulating cellular behavior at both molecular and systems levels. Methods Ethics approval All the animal experiments involved in this work were approved by the Animal Committee of Zhejiang University (approval no. ZJU20250712). The experimental procedures on rats complied with the “Regulations for the Administration of Affairs Concerning Experimental Animals” approved by the State Council of the People’s Republic of China. Preparation of MNV Human umbilical cord–derived mesenchymal stem cells (hMSCs) were purchased from Cyagen Biosciences (HUXUC-01001, Cyagen, China) and then were cultured in the human mesenchymal stem cell growth medium (HUXUC-90011, Cyagen, China) at 37 °C under 5% CO2, according to the manufacturer’s instructions. Then, cells were seeded in 150 mm culture dishes supplemented with 50 µM non-natural mannosamine derivative bearing an azide moiety (Ac4ManNAz). When cell confluence reached 80%, the cells were harvested using a 0.25% trypsin-EDTA solution. The cells were then centrifuged at 200 g for 5 minutes to collect the cell pellet. The pellet was sequentially extruded through filters with pore sizes of 5 µm, 1 µm, 400 nm, and 200 nm to prepare azide-functionalized mesenchymal stem cell-derived nanovesicles (MNV). Finally, MNV were collected by centrifugation at 25,000 g for 60 minutes. Preparation and characterization of SMNV Initially, SFL were prepared using the thin-film hydration method. Specifically, the lipid components purchased from Xian ruixi Biological Technology Co., Ltd (DOPE, DOPE-PEG 2000-DBCO, DOTAP) and DiI were mixed in chloroform at a molar ratio of 0.5:0.5:1:0.1. The chloroform was subsequently removed by rotary evaporation under vacuum for 30 minutes. The resulting lipid film was dispersed in a PBS solution containing 10 mM SAM (New England Biolabs), yielding a total lipid concentration of 2 mg/ml. After homogenization in an ultrasonic bath at 4 °C for 20 minutes, SAM-encapsulated liposomal vesicles were formed. The vesicles were then mixed with an equal concentration of MNV and incubated for 30 minutes, followed by sequential extrusion through membranes with pore sizes of 1 µm, 400 nm, and 200 nm to synthesize SMNV. SMNV was isolated by centrifugation at 25,000 g for 30 minutes and subsequently resuspended in water for further use. The size and zeta potential of SMNV was characterized using cryo-transmission electron microscopy and dynamic light scattering. Western blot analysis was performed to assess the retention of SMNV proteins. Cellular uptake and membrane fusion of SMNV At 37 °C, 50,000 enriched RAW 264.7 (Catalog no.: TIB-71, ATCC) cells were incubated with different concentrations of DiI-labeled SMNV for 30 minutes to assess SMNV binding in vitro. After incubation, the cells were washed three to five times with PBS, and DiI fluorescence was assessed directly by flow cytometry. For internalization studies, the cells were incubated with SMNV containing different lipid ratios, washed three times, and then immediately stained with LysoTracker. For the study of cellular uptake pathways, cells pretreated with various endocytic inhibitors (such as chlorpromazine, filipin III, wortmannin, and cytochalasin D) were incubated with SMNV, washed three times, and the internalization level was evaluated by flow cytometry. In vitro macrophage efferocytosis assay Macrophages (Mϕ) pretreated with LPS (10 ng/ml) were placed in 6-well tissue culture plates and subsequently treated with MNV or SMNV (100 µg/ml) for 6 hours. ACs were labeled with PKH67 according to the manufacturer’s instructions, with an optimized labeling duration of 30 min at room temperature, followed by extensive washing to remove excess dye. Labeled ACs were then added to macrophages at a ratio of 5:1 (ACs/Mϕ). Following 30 minutes of incubation, the unbound ACs were gently washed away, and the uptake of labeled cells was analyzed via flow cytometry. qRT-PCR and ELISA Analyses Total RNA was extracted from cultured cells using TRNzol-A+ (DP421, TIANGEN). RT-qPCR reactions were conducted using the Quant One Step RT-qPCR Kit (FP303, Tiangen), with each reaction performed in triplicate. (Primer sequences are provided in the Supplementary Table S1.) For the ELISA assays, cell supernatants were collected and centrifuged at 500 g for 5 minutes to remove residual cellular debris. The levels of pro-inflammatory cytokines TNF-α and anti-inflammatory cytokines (IL-10 and TGFβ1) were quantified following the manufacturer’s instructions (Biolegend). A SAM ELISA kit (MBS2603023, My Bio Source) was used to quantify the SAM content. Metabolomics analysis The cell samples stored at −80 °C were thawed on ice. A 500 μL solution of methanol and water (4:1 ratio) with an internal standard was added to the sample, then vortexed for 3 minutes. The sample was placed in liquid nitrogen for 5 minutes, then on dry ice for 5 minutes, thawed on ice, and vortexed again for 2 minutes. This freeze-thaw cycle was repeated three times. Afterward, the sample was centrifuged at 12,000 g for 10 minutes at 4 °C. A 300 μL supernatant was collected and kept at −20 °C for 30 minutes, followed by another centrifugation at 12,000 g for 3 minutes at 4 °C. Finally, 200 μL of the supernatant was transferred for LC-MS analysis. The sample extracts were analyzed using an LC-ESI-MS/MS system (UPLC, ExionLC AD, QTRAP® System). The UPLC conditions were: Waters ACQUITY UPLC HSS T3 C18 column (1.8 µm, 2.1 mm*100 mm) at 40 °C; flow rate 0.4 mL/min; injection volume 2 μL. The solvent system used was water (0.1% formic acid) and acetonitrile (0.1% formic acid), with a gradient program starting at 95:5 (V/V) and adjusting to 10:90 (V/V) over 14 minutes. Mass spectrometry analysis was done using a QTRAP® LC-MS/MS system with an ESI Turbo Ion-Spray interface in both positive and negative ion modes, controlled by Analyst 1.6.3 software. The key parameters were: source temperature 500 °C; ion spray voltage 5500 V (positive) and −4500 V (negative); gases set at 55, 60, and 25 psi; and high collision gas. The system was calibrated with polypropylene glycol solutions. Specific MRM transitions were monitored according to the metabolites. Differential metabolites were determined by VIP (VIP > 1) and P-value (P < 0.05). OCR analysis To analyze the OCR in real-time, macrophages were evaluated using the XF-96 Extracellular Flux Analyzer (Seahorse Bioscience). Briefly, cells were seeded in XF-96 cell culture plates and treated with either AC or AC + SMNV. The following compounds were used to measure OCR: 1.5 µM oligomycin, 0.5 µM carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP), and 0.5 µM rotenone/antimycin A. Immunocytochemistry Cells were rinsed once with PBS and then fixed in 4% paraformaldehyde (PFA) for 20 minutes. After fixation, the cells were permeabilized and blocked with 0.5% BSA at room temperature for 1 hour to prevent nonspecific binding. The cells were subsequently incubated with the primary antibody overnight at 4 °C, followed by a 1-hour incubation with the secondary antibody at room temperature. Nuclei were counterstained with ProLong Gold Antifade Mountant containing DAPI (ThermoFisher) to minimize fluorophore fading. All antibodies and kits were used according to the manufacturer’s instructions. The cells were then examined using a confocal microscope (Nikon). Wound healing and tube formation assays The angiogenic potential of HUVECs (Catalog no.: CRL-1730, ATCC) was assessed using wound healing and tube formation assays. For the wound healing assay, HUVECs were seeded at a density of 5 × 10⁵ cells per well in a six-well plate. Upon reaching 100% confluence, a straight-line scratch was made in the center of each well using a sterile 100 µL pipette tip. The cells were subsequently cultured in various preconditioned macrophage media. Wound closure was observed at 0 and 24 hours under an inverted microscope, and the wound closure rate (%) was calculated. The tube formation assay was performed in 48-well plates pre-coated with 120 µL of Matrigel per well. HUVECs were seeded at a density of 10⁵ cells per well onto the Matrigel surface and incubated at 37 °C with 5% CO₂ for 6 hours in different conditioned media. Tube formation was subsequently observed under a microscope. In vitro osteogenic differentiation and quantification To assess ALP activity, cells were induced for 7 days, then washed with PBS and fixed with 4% paraformaldehyde for 10 minutes. After an additional PBS wash, the cells were incubated with BCIP/NBT staining solution. Bright-field images were captured using a Nikon inverted microscope. ALP activity was quantified using an ALP Assay Kit (Beyotime) following the manufacturer’s instructions. On day 14, cells were fixed with 4% paraformaldehyde, washed twice with ultrapure water, and stained with 1% Alizarin Red S solution (Sigma-Aldrich) at 37 °C for 30 minutes. Following staining, the cells were washed five times with ultrapure water, and the stained areas were imaged under a microscope. For quantification, the ARS staining was eluted with 10% cetylpyridinium chloride for 1 hour, and the absorbance was measured at 548 nm using a microplate reader. In vitro osteoclast differentiation To assess osteoclast differentiation at the transcriptional level, bone marrow–derived macrophages (BMDMs) were cultured in osteoclast induction medium supplemented with macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor κB ligand (RANKL). Cells were induced for 5–7 days, with the medium refreshed every 2 days. Total RNA was extracted from cells using TRIzol reagent according to the manufacturer’s instructions, and reverse transcription was performed to synthesize cDNA. qPCR was carried out using SYBR Green Master Mix on a real-time PCR system. The mRNA expression levels of osteoclast-related genes, including Acp5, Ctsk, Nfatc1, and c-Fos, were quantified. Senescence-associated β-galactosidase (SA-β-gal) assay Briefly, cells were fixed with 2.5% glutaraldehyde and washed three times with PBS. The fixed cells were then stained with SA-β-galactosidase staining solution (Beyotime, C0602) at pH 6.0 and incubated overnight at 37 °C in the dark. After washing with deionized water, the stained cells were examined under a microscope for SA-β-galactosidase activity. Mitochondria high-resolution respiration assay Cultured macrophages were dissociated using trypsin, washed with PBS, and centrifuged at 500 g for 10 minutes at 4 °C. The cell pellet was resuspended in mitochondrial isolation buffer containing 0.25% fatty acid-free bovine serum albumin, 210 mM mannitol, 70 mM sucrose, 1 mM EDTA, and 10 mM HEPES, with the pH adjusted to 7.2. The samples were homogenized 20 times at 12,000 g using an IKA RW20 digital homogenizer in a cold room at 4 °C. The homogenate was centrifuged at 800 g for 5 minutes at 4 °C. The supernatant containing mitochondria was transferred to a new tube and centrifuged at 10,000 g for 10 minutes at 4 °C. The resulting mitochondrial pellet was collected for downstream analyses. The activity of complexes I, II, and IV in renal tubular cell mitochondria was assessed using an Oroboros instrument, and mitochondrial NAD levels were measured using the NAD+/NADH Assay Kit (Beyotime). Transcriptomics study Sequencing was performed by Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. (Shanghai, China). Briefly, total RNA was extracted using TRIzol reagent, and RNA concentration and purity were determined using a Nanodrop ND-2000 (Nanodrop Technologies). RNA integrity was assessed using a Bioanalyzer 5300 (Agilent). mRNA containing PolyA was specifically captured through two rounds of purification using oligo (dT) magnetic beads (25-61005, Thermo Fisher, USA). The purified mRNA was fragmented using a Magnesium RNA Fragmentation Module (E6150S, New England Biolabs, USA) at 94 °C for 5 minutes. The fragmented mRNA was reverse transcribed into cDNA using reverse transcriptase (Thermo Fisher, USA). The second strand synthesis was conducted using DNA polymerase I (M0209, New England Biolabs, USA) and RNase H (M0297, New England Biolabs, USA). The composite double strand of DNA and RNA was then converted to double-stranded DNA. dUTP solution (R0133, Thermo Fisher, USA) was added to the double-stranded DNA to create blunt ends, followed by the addition of an A base to each end for ligation with a T base adapter. The DNA was purified using magnetic beads, and the second strand was digested with UDG enzyme (M0280, New England Biolabs, USA). PCR was performed to construct a library with a fragment size of 300 ± 50 bp. The PCR cycle included a pre-denaturation stage (95 °C for 3 minutes), denaturation stage (98 °C for 15 seconds, 8 cycles), annealing stage (60 °C for 15 seconds), extension stage (72 °C for 30 seconds), and a final extension stage (72 °C for 5 minutes). Sequencing was performed using the Illumina NovaSeq 6000. Aged rat calvarial bone defect model Eighteen-month-old female Sprague–Dawley rats were housed in groups of 4–6 per individually ventilated cage under a 12 h light–dark cycle (06:30–18:30 light; 18:30–06:30 dark), with a constant room temperature (21 ± 1 °C) and relative humidity (40–60%). Animals had free access to food and water ad libitum. Following anesthesia, the skin and muscle were carefully dissected to expose the calvarium. A critical-sized calvarial defect (5 mm in diameter) was created by removing the full thickness of the bone using a dental drill. The rats were randomly assigned to the Blank group, Gel group, MNV@Gel group, or SMNV@Gel group and received injections on the first day post-surgery. Postoperatively, rats were closely monitored daily for general condition, body weight, wound healing, and signs of pain or distress. Analgesics were administered as needed, and any animals showing signs of infection or abnormal behavior were treated promptly in accordance with institutional animal care guidelines. One week after treatment, the rats were euthanized under anesthesia, and the inflammatory response was assessed by quantifying the expression of TNF-α, IL-6, IL-1β, IL-4, and IL-10. Eight weeks post-treatment, the rats were euthanized under anesthesia, and the calvaria were harvested to evaluate bone regeneration efficiency. μCT imaging Eight weeks after treatment, the rats were euthanized, and the calvaria were harvested and fixed in 4% paraformaldehyde. Bone regeneration was subsequently analyzed using Micro-CT (SKYSCAN1176, BRUKER, Massachusetts, USA). The analysis software included with the Micro-CT system was used to assess bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular number (Tb.N). Histological evaluation After μCT analysis, all samples were decalcified in 15% ethylenediaminetetraacetic acid (EDTA) for at least 4 weeks and then embedded in paraffin. Serial sections, 5 μm in thickness, were prepared and deparaffinized through a graded ethanol series (70–100%) followed by xylene. At least three sections from each sample were collected and stained with Hematoxylin and Eosin (HE) and Masson’s Trichrome, then examined using a Zeiss optical microscope. Osteogenesis was further assessed by immunofluorescence (IF) staining for osteogenic markers, including Col1a1 and OCN. Statistics and reproducibility All experiments were performed at least in triplicate. Semi-quantitative analyses were conducted on a minimum of three independent samples, and data are presented as mean ± standard deviation (SD) or mean ± standard error of the mean (SEM), as indicated in the corresponding figure legends. Statistical analyses were performed using GraphPad Prism 9 software (GraphPad Software Inc.). Comparisons between two groups were conducted using an unpaired two-tailed Student’s t-test, while comparisons among multiple groups were analyzed using one-way ANOVA followed by Tukey’s post hoc test. A P < 0.05 was considered statistically significant. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information Reporting Summary Transparent Peer Review file Supplementary Information Reporting Summary Transparent Peer Review file Source data Source Data Source Data