Dual-targeted exercise mimetic extracellular vesicles regulate the muscle-bone crosstalk to treat osteosarcopenia

Dual-targeted exercise mimetic extracellular vesicles regulate the muscle-bone crosstalk to treat osteosarcopenia | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 26 October 2025 V1 Latest version Share on Dual-targeted exercise mimetic extracellular vesicles regulate the muscle-bone crosstalk to treat osteosarcopenia Authors : Yongzhi Cui 0000-0002-7532-0151 , Qian Xiang , Ke Zhao , Jiao Jiao Li 0000-0002-3584-6765 , Benchi Che , Kaiwen Zheng , Yanchun Gao , Shengrong Guo , and Dehao Fu [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.176150016.60795129/v1 435 views 181 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Clinically, the co-presentation of osteoporosis and sarcopenia often occurs in the same individuals, termed osteosarcopenia (OSP). Although OSP has high incidence and causes significant morbidity, there are no effective treatments concurrently re-establishing bone and muscle mass. Recent researches highlight the existence of inter-organ communication between bone and muscle, inspiring exercise-mimetic therapies that utilizing exercise-inducible myokines and osteokines beyond their original organs. This study proposes the pioneering concept of exercise mimetic extracellular vesicles (EMEVs) for the treatment of OSP. The EMEVs are produced by human induced pluripotent stem cells-derived myotubes and stimulated with mechanical strain to mimic exercise. These EMEVs were found to contain abundant exercise-inducible cargos, especially irisin, with beneficial effects including promoting the proliferation of muscle stem cells, extending the lifespan of myotubes and converting their cytokine profile, and facilitating osteogenic differentiation in vitro . Subsequently, dual-targeted EMEVs (DT-EMEVs) were constructed by modifying EMEVs with bone-targeting and muscle-targeting peptides, which exhibited organ-specific distribution in vivo . In aged OSP murine models, DT-EMEVs effectively recovered both bone and muscle mass, enhanced exercise performance, and improved bone mechanical strength. Beyond targeted and effective treatment of OSP, DT-EMEVs may inspire new paradigms in the managing other musculoskeletal disorders and developing exercise mimetic therapies. 1 Introduction Osteosarcopenia (OSP), the co-occurrence of osteopenia/osteoporosis and sarcopenia, predominantly affects the elderly, obese, and bedridden individuals[1]. As a comorbidity, OSP significantly elevates risks of falls, fractures, immobility, and mortality[1, 2]. Anticipated rises in OSP prevalence, due to global aging, herald significant healthcare and societal burdens[2]. However, effective treatments remain elusive. Current focus is on preventive exercise and nutrition, alongside limited pharmacologic options adapted from single-organ therapies[3]. Developing interventions that concurrently regenerate both bone and muscle represents a major unmet need. To address this, a paradigm shift views bone and muscle as an integrated musculoskeletal system, where exercise-mediated molecular pathways offer a promising therapeutic avenue[4]. Bone provides structural anchorage for muscle, while muscle generates mechanical loads essential for bone formation and metabolism, with bidirectional endocrine crosstalk tightly regulating mutual homeostasis[5, 6]. Crucially, exercise stimulates skeletal muscle to secrete cytokines known as myokines—such as irisin[7, 8], β-aminoisobutyric acid (BAIBA)[9, 10], and apelin—which exert autocrine effects on muscle metabolism (e.g., enhancing mitochondrial function, reducing oxidative stress, and promoting autophagy[7, 11]) and paracrine/endocrine effects on bone, including promoting osteoblast proliferation and differentiation (e.g., irisin and BAIBA[12, 13]) while suppressing inhibitory factors like myostatin (MSTN) that impair osteogenesis and promote bone resorption[14-17]. Similarly, bone-derived osteokines—such as osteocalcin (OCN)[18], prostaglandin E2 (PGE2)[19], and RANK ligand (RANKL)—respond to exercise, with increased OCN and PGE2 enhancing bone formation and muscle glucose/fatty acid utilization[18, 20], while decreased RANKL inhibits osteoclast-mediated resorption and muscle atrophy[21-24]. Collectively, exercise regulates this muscle-bone crosstalk through myokines and osteokines, promoting anabolic processes in both tissues, thereby highlighting its potential as a holistic OSP intervention. In addition to influencing the muscle-bone crosstalk, the systemic benefits of long-term exercise—including metabolic regulation, mitochondrial biogenesis, enhanced protein synthesis, and attenuation of chronic inflammation and cellular senescence—are well-established[25-27]. Consequently, exercise is widely prescribed to prevent and manage diseases such as obesity, diabetes, hyperlipidaemia, and OSP[27]. However, patient compliance remains a major barrier due to physical limitations and lack of immediate effects. To address this gap, exercise mimetics—pharmacologic agents that mimic exercise benefits without physical exertion—have emerged as promising therapies for chronic musculoskeletal disorders like OSP. Early AMPK and PPARδ agonists demonstrated oral efficacy in enhancing energy metabolism and exercise endurance[28]. More recently, the synthetic estrogen-related receptor agonist SLU-PP-332 boosted mitochondrial oxidative phosphorylation, energy expenditure, and insulin sensitivity, replicating exercise-induced adaptations[29]. Although the dual osteo-myogenic compound locamidazole (activating Mef2c/PGC-1α to increase muscle/bone mass) showed initial promise for OSP[30], no subsequent reports have documented its clinical translation to date. This underscores the urgent need for new, biocompatible exercise mimetics with potent efficacy and minimal toxicity. In recent decades, extracellular vesicles (EVs) from diverse cellular sources have emerged as a therapeutic focus for musculoskeletal disorders[31]. As natural nanoscale vesicles, EVs carry bioactive molecules and modulate intercellular communication to exert therapeutic effects[32]. Studies confirmed skeletal muscle cells secreted functionally active EVs with context-dependent bioactivity: EVs during myogenic differentiation delivered myogenic growth factors [insulin-like growth factor (IGF), hepatocyte growth factor (HGF), fibroblast growth factor-2 (FGF-2) and platelet-derived growth factor-AA (PDGF-AA)] to enhance muscle regeneration[33]. In addition, post-exercise EVs enriched in lactate dehydrogenase A (LDHA) stimulated glycolysis in bone marrow-derived mesenchymal stem cells (BMSCs), augmenting osteogenesis while inhibiting osteoclast fusion to reverse disuse osteoporosis[34]; endurance-trained muscle EVs transported fibronectin type III domain-containing protein 5 (FNDC5)/irisin to activate AMPK-mediated osteoblast proliferation[12]. Conversely, EVs from disused atrophic muscle impaired BMSC osteogenesis and exacerbated osteoclastogenesis[35]. These findings established that skeletal muscle-derived EV bioactivity was intrinsically linked to muscle functional status, positioning exercise-derived EVs as promising exercise mimetics to concurrently restore muscle/bone mass. Despite EVs derived from skeletal muscle (SkM-EVs) after exercise hold significant therapeutic potential for the treatment of OSP, their translation into clinical therapies faces major challenges. Firstly, SkM-EVs are distributed throughout the body via systemic circulation, making it difficult to effectively purify and isolate. Additionally, similar to most natural EVs[36], SkM-EVs inherently lack specific targeting capability, preventing their selective accumulation in bone and skeletal muscle tissues, thereby limiting effective treatment at low therapeutic doses. Consequently, to develop SkM-EVs as exercise mimetics, it is essential to obtain large quantities of homogeneous skeletal muscle cells, subject them to mechanical stimulation mimicking exercise conditions, and engineer their resultant EVs with functional modifications to enhance tissue-specific targeting and improve therapeutical efficiency[32]. Herein, we introduce a pioneering exercise mimetic therapy based on exercise-induced EVs derived from human induced pluripotent stem cell (iPSC)-differentiated SkMCs. The use of iPSCs as source cells ensures that SkMCs with highly consistent characteristics could be continuously obtained in a non-invasive manner. The iPSCs derived mature SkMCs, also termed myotubes (abbreviated as iMyotubes) were stimulated with mechanical strain to mimic exercise. EVs produced by iMyotubes under such condition were termed exercise mimetic EVs (EMEVs). These EMEVs, found to carry abundant endogenous irisin, were shown to enhanced the proliferation and mitochondrial function of MuSCs, improve the metabolic function of SkMCs and convert their cytokine profile akin to exercise stimulation, and facilitate in vitro osteogenic differentiation of pre-osteoblast cells. To facilitate tissue-specific in vivo delivery, dual-targeted EMEVs (DT-EMEVs) were constructed by modifying EMEVs with bone-targeting and muscle-targeting peptides. These DT-EMEVs were found to effectively preserve bone and muscle mass in aged OSP mice, while significantly enhancing their exercise performance and bone strength (Scheme 1). The innovative approach of utilizing EMEVs produced by iPSCs-derived SkMCs, and the potential to modify them with targeting peptides for tissue-specific delivery, open new avenues for the treatment of OSP and other chronic musculoskeletal disorders. 2 Experimental Section 2.1 Human iPSC Maintenance and Differentiation into iMyoblast Undifferentiated human iPSC line (DYR0100) was obtained commercially from Chinese Center for Type Culture Collection. The cells were routinely maintained and expanded feeder-freely on Matrigel (ESC-Qualified, BD Biosciences, Franklin Lakes, NJ, USA) coated six-well plates (NEST Biotechnology, Wuxi, China) and in ncTarget hPSC Medium (RP01020, Shownin Biotechnologies, Hefei, China). The induction of human iPSC differentiation into myoblast cells (iMyoblast) was performed in a thermostatic incubator with 5% O 2 /5% CO 2 and divided into two steps[37]. Firstly, human iPSCs were differentiated into myogenic precursors. iPSCs were reseeded at 2,000 cells per cm 2 onto Matrigel-coated plates and cultured for 10 days in myogenic precursor induction medium containing 5% horse serum (Thermo Fisher Scientific, Oakwood Village, OH, USA), 3 μM CHIR99021 (T2310, TargetMol, Boston, MA, USA), 2 μMAlk5 Inhibitor (Sapphire Bioscience, Redfern, Australia), 10 ng/mL human recombinant epidermal growth factor (hrEGF) (Miltenyi Biotec, San Diego, CA, USA), 200 μM ascorbic acid (Shanghai Standard Technology Co., Ltd), 10 μg/mL insulin (Sigma-Aldrich, St. Louis, USA), and 0.4 μg/mL dexamethasone (Sigma-Aldrich). After 10 days, myogenic precursor cells were dissociated using 0.05% trypsin and reseeded at 2,000 cells per cm 2 onto Matrigel-coated plates for myoblast differentiation over the next 8 days. The skeletal myoblast induction medium contained 5% horse serum, 10 μg/mL insulin, 10 ng/mL EGF, 20 ng/mL HGF (Peprotech, Rocky Hill, NJ, USA), 10 ng/ml PDGF (Peprotech), 20 ng/mL bFGF (Novoprotein, Shanghai, China), 20 µg/mL oncostatin (Miltenyi Biotec), 10 ng/mL insulin-like growth factor1 (Miltenyi Biotec), 2 μM SB431542 (Miltenyi Biotec), and 200 μM ascorbic acid. After the two-step differentiation, the iMyoblast cells were obtained for conducting myotube formation assay to confirm their myogenesis potential. In this assay, the cells were cultured for 7 days in myotube medium, containing 10 μg/mL insulin, 20 μg/mL oncostatin, 50 nM necrosulfonamide (Cellagen Technology, San Diego, CA, USA), and 200 μM ascorbic acid. 2.2 Mechanical Stimulation of iMyotubes The iMyotubes were seeded into flexible-bottom 6-well plates and cultured in a myogenic medium to sustain myotube formation[38]. The static group received the same treatment as the mechanically stretched groups, with the exception of mechanical stretching. Cells were exposed to cyclic strain with elongations of 10%, 15%, and 20% at frequencies of 0.25, 0.5, and 1 Hz, for durations of 1, 2, 4, 6, 8, and 12 hours per day, consecutively, over a 2-day period, using a computer-controlled vacuum stretch apparatus (Flexercell 5000, USA). The viability of iMyotubes cells at each stimulation condition was determined using the CCK-8 kit (Life-iLab, Shanghai, China), in line with the manufacturer’s instructions. Each experiment was performed in triplicate. 2.3 Extraction & Characterization of EVs iMyotubes derived from iPSCs were used for exosome isolation from the conditioned medium according to published protocols[37]. Briefly, iMyotubes were cultured for 48 h under static or exercise mimetic conditions. The conditioned medium underwent a two-step centrifugation process: initially at 300 × g for 10 min to remove nonadherent cells, followed by a second centrifugation at 10 3 × g for 15 min to eliminate dead cells and debris. Subsequently, larger vesicles were removed by centrifugation at 10 4 × g for 60 min. The remaining supernatant was subjected to ultracentrifugation at 10 5 × g for 70 min using a Beckman Coulter ultracentrifuge (Brea, CA, USA). The pellet of small EVs obtained was resuspended in phosphate-buffered saline (PBS) and underwent an additional round of ultracentrifugation under identical conditions to remove contaminating proteins. Accordingly, the small EVs isolated from static and exercise-mimetic conditions were referred to as EVs and exercise-mimetic exosomes (EMEVs), respectively, throughout this study. Protein concentration was quantified using a bicinchoninic acid (BCA) protein assay kit (Beyotime). The size distribution and particle numbers were evaluated through dynamic light scattering (DLS) using a Zetasizer Nano ZS90 (Malvern Instruments, Malvern, UK). The morphology of the EVs was examined by transmission electron microscopy (TEM) using a JEM-1200EX microscope (JEOL Ltd., Tokyo, Japan) at an acceleration voltage of 100 keV. 2.4 Primary MuSCs Isolation & Culture Five-week-old C57BL/6 mice were euthanized and their hindlimb muscles were dissected, which were digested using collagenase type 2 dissolved in CaCl 2 solution (10 mM) at 37 °C in a sterile incubator for 30 min. The digested tissues were then centrifuged and washed using PBS, and further digested with collagenase D (1 mL, 1.5 U/mL) and Dispase II (2.4 U/mL, Roche Applied Science, Indianapolis, IN). The mixture was incubated for 60 min, followed by centrifugation at 500 g for 5 min. The obtained cell pellet was resuspended in Dulbecco’s modified eagle’s medium (DMEM, PM150210B, Procell, Wuhan, China) supplemented with 1% penicillin–streptomycin (PS) and 10% fetal bovine serum (FBS). The resuspended cells were seeded in 6-well plates for 4 h to remove fibroblasts. The unattached cells (MuSCs) were then transferred to Matrigel (BD Biosciences) precoated plates for growth in an incubator at 37 °C. To study the protective effects of EMEVs, the MuSCs were treated with TNF-α (25 ng/mL) dissolved in the culture medium[39]. For myogenic differentiation, DMEM supplemented with 1% PS and 2% horse serum was used as the differentiation medium and used to culture the MuSCs for 3 days. To investigate the pro-myogenic effects exerted by EMEVs, the MuSCs were switched into differentiation media containing EMEVs (10 9 particles/mL) for further differentiation. The cellular morphology and myotube formation were evaluated using a fluorescence microscope. 2.5 Osteogenic Induction & Evaluation MC3T3 cells (2.5 × 10 4 cells/well) were seeded in 12-well plates. Following plate adherence and overnight culture, the cells were grown in commercial mouse osteogenic differentiation medium (Oricell), which was replaced every 72 h. Alizarin Red S (ARS) staining kit for osteogenic mineralization was used to conduct the ARS staining assay in accordance with the manufacturer’s protocols. For staining quantification, the samples were treated using 10% acetic acid overnight, followed by centrifugation and neutralizing the supernatant using 10% ammonium hydroxide. The amount of staining was quantified by reading the absorbance at 405 nm with a microplate reader. 2.6 Cellular Immunofluorescence Staining Cells were fixed using 4% paraformaldehyde (PFA) at room temperature for 15 min and permeabilized using 0.2% Triton X-100 for 7 min, then incubated with the antibodies listed below to visualize the specific protein markers. For the staining of cytoskeleton, the cells were incubated with FBS (UR50100, Umibio Co., Ltd., Shanghai, China) containing phalloidin-FITC (40735ES75, Yeasen Biotechnology Co., Ltd., Shanghai, China) for 35 min and then washed using PBS. The primary antibodies used in this assay were as below: Pax7 (R381466, ZENBIO), MyoD1 (bs-23809R, Bioss, Beijing, China), MyHC (A17427, ABclonal Technology), Runx2 (R381417, ZENBIO) and Sp7 (A6205, ABclonal Technology). These antibodies diluted in antibody diluent (WB100D, New Cell & Molecular Biotech, Suzhou, China) were used to incubate the cells overnight. The matched secondary antibodies with Alexa Fluor fluorescence (1:500) were used in the next incubation, and 4′,6-diamidino-2-phenylindole (DAPI, Solarbio, Beijing, China) was used to visualize the cell nuclei. 2.7 Biodistribution of DT-EMEVs DT-EMEVs and their counterparts were stained with lipophilic DiR fluorescent dye (KGMP0026, KeyGEN BioTECH, Nanjing, China). Each mouse was intravenously administered with 50 μL PBS containing 1.0 × 10 10 particles through the tail. After 4 and 8 h, the different organs were harvested and imaged using a biophotonic system Bruker Xtreme (Bruker Corp., Billerica, MA, USA) and its software Bruker MI SE (Bruker Corp.). 2.8 Animals In this study, the C57/BL6 mice were maintained at the Animal Experimental Center of Shanghai Sixth People’s Hospital. All experiments involving animals were executed in line with the guidelines and standards of the Animal Care and Use Committee (IACUC) of Shanghai Sixth People’s Hospital (animal welfare ethical acceptance No.2024-0446). 3 and 18-month-old mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. Mice were housed at 22 °C, following a 12-hour light and 12-hour dark cycle, with adequate access to food and water. These 18-month-old mice were randomized into 5 groups: old group (PBS treatment), and groups treated with EMEVs, MT-EMEVs, BT-EMEVs, and DT-EMEVs by tail vein injection (1.0 × 10 10 particles in 50 μL PBS) once per week for 8 weeks. The 3-month-old mice (equivalent PBS treatment) were set as young group for comparison. 2.9 Muscle Strength Measurement and Physical Activity Tests Muscle strength was measured according to published methods[40]. Briefly, the mice were trained to grasp a horizontal bar due to their instinct to hold onto an object when suspended. The bar was attached to an electronic dynamometer (Handpi HP-5N, China). While the mouse was grabbing the bar, the animal’s body was gently pulled backwards in a horizontal direction, making it resist the pull. The force used to pull the mouse increased gradually until the animal let go of the bar. At this point, the dynamometer reading was recorded, and three measured values were averaged to obtain the forelimb grip strength. In addition to the muscle strength, an exhaustive running test was conducted to investigate the physical activity of mice. The exhaustive running test was conducted with a treadmill (Beijing Zhongshi Dichuang Technology Development Co., Ltd.). Before the test, the mice were trained twice to perform running exercise (10 m/min, 0° slope angle) on the treadmill. During the test, the running speed started from 12 m/min with no slope angle. The speed was then increased by 2 m/min and slope angle by 1°/min until the values reached 38 m/min and 13°, respectively. When the tested animal has left the track for over 20 s and simultaneously showed lags in response to external stimuli, it was taken away for rest and the corresponding time was recorded as exhaustive running time. 2.10 Muscle Mass Measurement The gastrocnemius muscles were dissected and weighed after euthanasia. After weighing, the muscles were fixed with 4% PFA for subsequent assays. Relative muscle mass was calculated according to the following formula: Relative muscle mass (%) = the weight of specific muscle ÷ total body weight × 100. 2.11 Immunofluorescence Staining of Tissue Cryosections Collected muscle tissues from one side per animal were treated with liquid nitrogen and frozen at -80 °C. Bone tissues were fixed using 4% PFA for 24 h and decalcified with 10% ethylene diamine tetraacetic acid (EDTA) for 3 weeks. Before cryosection, all samples were dehydrated using a PBS solution containing 20% sucrose, rinsed in OCT and quick-frozen in liquid nitrogen, then sectioned into 10-μm slices in a freezing microtome with the assistance of Wuhan Pinuofei Biological Technology Ltd. For immunofluorescence staining, non-specific epitopes were blocked with 4% BSA at 4 °C for 30 min after permeabilization using Triton X-100 for 8 min on ice. Primary antibodies (anti-PAX-7, anti-PINK) were incubated with the muscle tissue sections overnight, as well as with the bone tissue sections, followed by incubation with secondary antibodies (APExBIO, Houston, USA) for 60 min. Mounting medium containing DAPI was used to stain nuclei and mount the sections. 2.12 Blood Biochemistry Examination Mouse blood collected from the ophthalmic venous plexus after anesthesia was preserved in heparin anticoagulant tubes. The blood was centrifuged at 1.0 × 10 3 g for 30 min to acquire the serum. Blood biochemistry indicators including alanine aminotransferase (ALT), aspartate aminotransferase (AST), serum creatinine (Scr), blood urea nitrogen (BUN), and total bilirubin (TBil) were measured using an automated chemistry analyzer Rayto Chemray-240 (Rayto Life and Analytical Sciences Co., Ltd., Shenzhen, China). 2.13 Enzyme-linked Immunosorbent Assay (ELISA) The concentrations of bone metabolic markers in the serum were determined using ELISA kits following the manufacturers’ protocols, including bone alkaline phosphatase (BALP), procollagen type-1 N-terminal propeptide (P1NP), tartrate resistant acid phosphatase 5b (TRAP5b), β-isomerized carboxy-terminal telopeptides (β-CTX) (BALP, MU30270; P1NP, MU30602; TRAP5b, MU30923; β-CTX, MU30091, BioSwamp, Wuhan, China), and OCN (JL20366-48T, Shanghai Hengyuan Biological Technology Co., Ltd., Shanghai, China). 2.14 Calcein Labelling These mice were injected with calcein (25 mg/kg, Aladdin, Shanghai, China, dissolved in 2% sodium bicarbonate) subcutaneously at day 9 and day 2 before euthanasia. The femurs were dissected, fixed with 4% PFA and dehydrated with 20% sucrose solution. After embedding in methyl methacrylate, longitudinal sections from the coronal plane were cut into 5-μm thickness and observed using a fluorescence microscope. 2.15 Micro-computed Tomography (μCT) μCT analyses were conducted as reported previously using a μCT scanner with high-resolution (Skyscan 1276, Bruker).[19] NRecon software (SkyScan) was used for data reconstruction; CTAn software (SkyScan) was used for data analysis; and CTVol software (SkyScan) was used for space visualization. The trabecular bone volume/total volume (Tb. BV/TV), trabecular number (Tb. N), trabecular separation (Tb. Sp) and trabecular thickness (Tb. Th) were determined as the parameters for trabecular bone. 2.16 Quantitative Polymerase Chain Reaction (qPCR) The cell or tissue samples for qPCR were harvested and kept in RNA preservation solution. For the isolation of total RNA, TRIzol reagent was used by following the manufacturer’s protocol. RNA quantification was performed with a Thermo Fisher Scientific NanoDrop-2000 spectrophotometer. The qPCR assays were conducted in triplicate using the StarLighter HP SYBR Green qPCR Mix Kit (FS-Q1002, Beijing Foreverstar Biotech) in a BioRad CFX96 instrument (BioRad, California, USA). The relative expression levels of genes were determined using the comparative Ct (2 −ΔΔCt ) method. Primer information was shown in Table S1. 2.17 Western Blotting Whole cell extracts and EV extracts were prepared using standard techniques for western blotting. Cells were collected, re-suspended in radioimmunoprecipitation assay lysis buffer, and centrifuged. The resulting pellet was re-suspended in ProtLytic Protease and Phosphatase Inhibitor Cocktail (P001, New Cell & Molecular Biotech, Suzhou, China) and frozen in liquid nitrogen. Samples were loaded into wells in Tris-HCl gels and initially run at 60 V for 10 min, then switched to 120V for the remainder of the run. Proteins were transferred from the gel to a polyvinylidene difluoride (PVDF) membrane. The membrane was soaked in methanol and equilibrated in transfer buffer for over 30 min before the transfer process. The membrane was blocked to prevent non-specific binding and then incubated with primary antibodies specific to the target proteins as follow: TBK1 (HA601045, HUABIO, Hangzhou, China); p-TBK1 (BD-PB4629, Biodragon, Suzhou, China); IRF3 (240072, ZENBIO), p-IRF3 (530857, ZENBIO), CD63 (R23327, ZENBIO), TSG101 (R25999, ZENBIO), STING (300415, ZENBIO) and p-STING (FNab11024, Wuhan Fine Biotech, Wuhan, China). Detection of protein bands was performed using an enhanced chemiluminescence (ECL) system (ED0015-A, Shandong Sparkjade Biotechnology Co., Ltd.). 2.18 Statistical Analysis Numerical data were shown as mean ± standard deviation (SD). GraphPad Prism 8.0.1 software was used to conduct the statistical analyses. The test for normality was conducted using the Kolmogorov–Smirnov method. Statistical differences between groups were determined using Student’s t-test (2 groups) and one-way ANOVA (3 or more groups). In ANOVA, Bartlett’s Test verified the homogeneity of variances, and post-hoc tests were conducted using the Tukey method. Statistical significance was exhibited as: *, P < 0.05; **, P < 0.01; ***, P < 0.001; #, P < 0.0001; NS, no statistical significance. 3 Results and discussion 3.1 Derivation and characterization of iMyotubes In this study, iPSCs served as a renewable source of SkMCs, enabling the generation of a large number of source cells with consistent characteristics essential for the mass production of EVs. The differentiation of human iPSCs into myotubes was achieved through a stepwise induction process, whereby iPSCs were first directed into myogenic precursors, then into myoblasts, and finally into myotubes (Fig. S1A). To distinguish these iPSC-derived cells from primary myoblasts and myotubes, they were designated as iMyoblasts and iMyotubes throughout this study. Over the course of myogenic induction, the expression of the pluripotency marker Sox2 decreased substantially upon differentiation (Fig. S1B). Concurrently, the expression of Pax7, a marker of myogenic precursors, was markedly upregulated by Day 10 of differentiation. Following transition to myoblast induction medium, the cells exhibited a pronounced increase in the expression of the key myogenic regulatory factor MyoD1 by Day 18. The acquisition of iMyoblasts represented a critical step, as these cells provide a proliferative source for subsequent experiments before undergoing terminal differentiation into post-mitotic iMyotubes. Flow cytometry analysis confirmed the purity of iMyoblasts, demonstrating positive expression of CD82, Pax7, and MyoD1, and absence of CD31, CD45, and CD11b (Fig. S1C). The expression of Pax7 and MyoD1 in iMyoblasts was further validated by immunofluorescence staining (Fig. S1D). Subsequently, iMyoblasts were cultured in skeletal muscle differentiation medium until Day 26 to obtain iMyotubes. At this stage, the expression of myogenin (MyoG), which is critical for myotube formation and maintenance, was significantly elevated. This was accompanied by the appearance of a characteristic tube-like morphology under bright-field microscopy (Fig. S1E). Collectively, these results demonstrate the successful generation of iPSC-derived iMyoblasts and iMyotubes through a sequential differentiation protocol. 3.2 Preparation and characterization of EMEVs Various forms of mechanical stimulation have been described for cellular applications, including fluid shear stress, compression, tensile stretch, bending, vibration, and osmotic pressure[41]. Of these, stretching was identified as the most suitable modality for simulating muscle contraction. Accordingly, we employed a commercial Flexcell system to apply cyclic mechanical stretch to iMyotubes. As illustrated in Fig. S2A, iMyotubes were subjected to elongations of 10%, 15%, and 20% under sinusoidal waveforms at 0.25, 0.5, and 1 Hz, under varying static/strain cycle regimens. Among these conditions, iMyotubes stimulated with 0.5 Hz and 15% elongation exhibited the highest viability under both 6h static/6h strain and 12h static/12h strain cycles, whereas viability declined when elongation was increased to 20%. Based on maximal cellular tolerance, the 0.5 Hz, 15% elongation, and 12h static/12h strain condition was selected for subsequent experiments. It should be noted that other mechanical stimulation parameters may also prove beneficial and warrant further investigation. Previous studies have indicated that mechanical stimulation enhances EV secretion from engineered tissues[42]. Here, we assessed the influence of mechanical stretch on EV production and cargo composition in iMyotubes. Quantification of the exosomal marker CD63 and protein content revealed a 10-fold and 7.7-fold increase in EV yield under 0.5 Hz and 15% elongation stimulation compared to static conditions, respectively (Figs. S2b & S2c). Moreover, mechanical stimulation elevated the irisin content within EVs (Fig. S2d). Together, these findings confirm that stretching at 0.5 Hz and 15% elongation effectively mimics exercise-like stimulation in iMyotubes. For clarity, EVs derived from iMyotubes under this mechanical stimulation are designated as EMEVs, while those from static cultures are referred to as EVs (Fig. 1a). We characterized and compared the properties of EMEVs and EVs. Transmission electron microscopy (TEM) revealed that both EV types exhibited typical saucer-like morphologies (Fig. 1b). Dynamic light scattering (DLS) indicated similar particle sizes, zeta potentials, and size distributions between EMEVs and EVs (Fig. 1c & d). Western blot analysis confirmed the expression of exosomal markers CD63 and TSG101 in both EV types (Fig. 1e). Protein cargo was analyzed using ELISA and high-performance liquid chromatography-mass spectrometry (HPLC-MS). Notably, among numerous reported myokines, only irisin was detected in EMEVs (Table S2). In addition to conventional myokines, lactate dehydrogenase A (LDHA)—recently identified in SkMC-derived vesicles [34]—was also assessed via ELISA. Consistent with prior reports, both EMEVs and EVs carried LDHA, with EMEVs containing significantly higher levels than EVs (Fig. S3). These results indicate that irisin and LDHA represent major endogenous cargoes in EMEVs. LDHA has been implicated in glycolysis enhancement, promotion of myogenic differentiation, and bone remodeling [43, 44], while irisin is a well-established exercise-induced myokine with multi-tissue bioactivity [8, 45]. To elucidate the mechanisms underlying differences in EV production and cargo between EVs and EMEVs, we focused on the AMPK-PGC-1α pathway, which is known to regulate energy metabolism and exercise adaptation. Mechanical stretch activates AMPK, leading to enhanced energy production, cellular adaptation, and irisin expression[46]. Using an AMPK inhibitor, we suppressed this pathway during stretch stimulation. Inhibition resulted reduced phosphorylation of AMPK and expression of PGC-1α (Fig. 1f). Interestingly, AMPK inhibition did not significantly alter EMEV production but markedly reduced irisin content within EVs (Figs. 1g & h). In contrast, EV secretion was inhibited by calcium chelation. Previous work has shown that shear stress—a distinct form of mechanical stimulation—promotes EV biogenesis in mesenchymal stem cells via extracellular Ca²⁺ influx[47]. However, the effect of stretch stimulation on EV production in SkMCs remains poorly understood. Our results suggest that exercise-mimetic stretching enhances EV secretion via Ca²⁺ influx and promotes irisin enrichment via AMPK-PGC-1α activation in myotubes (Fig. 1i). 3.3 EMEVs promoted the proliferation of MuSCs and improved their mitochondrial function Primary MuSCs were utilized to evaluate the in vitro effects of EMEVs. Although C2C12 cells are commonly used as a model in studies of muscle development, disease, and metabolism, their responses may considerably diverge from those of primary cells, potentially introducing experimental bias[48]. The use of MuSCs from aged mice also posed significant challenges owing to reduced isolatable cell numbers and diminished proliferative capacity. Therefore, primary MuSCs from young mice were employed to assess the biological effects of EMEVs. Treatment with either EVs or EMEVs enhanced the expression of Pax7—a key factor maintaining regenerative capacity in MuSCs—with EMEVs exerting markedly stronger effects than EVs (Figs. 2a & 2b). To further investigate the mechanistic basis of this effect, loss-of-function experiments were conducted using EMEVs with irisin or LDHA knocked down via siRNA in iMyotubes [designated as EMEVs (irisin KD) and EMEVs (LDHA KD), respectively]. Treatment with EMEVs (irisin KD) abolished the promotive effect on Pax7 expression, whereas EMEVs (LDHA KD) retained this activity, indicating that irisin plays a primary role in mediating the upregulation of Pax7 by EMEVs (Fig. S4). Skeletal muscle atrophy during aging is commonly associated with systemically elevated TNF-α in serum, which contributes to reduced MuSC numbers[49]. To simulate this age-related proliferative inhibition, MuSCs were treated with TNF-α, resulting in substantially slowed gains in cellular viability. The addition of EVs partially alleviated TNF-α-induced proliferative suppression, whereas EMEVs nearly completely abolished the inhibitory effects of TNF-α (Fig. 2c). Similarly, EMEVs (irisin KD) failed to rescue the TNF-α-induced proliferative suppression, whereas EMEVs (LDHA KD) retained the ability to counteract this inhibition (Fig. S5). Consistent with these findings, flow cytometric cell cycle analysis revealed a TNF-α-induced reduction in S-phase MuSCs, which was reversed by EMEV treatment (Figs. 2d & 2e). These results indicate that EMEVs promote MuSC proliferation and counteract TNF-α-mediated inhibition. Normal mitochondrial function is critical for maintaining MuSC regenerative capacity. Mitochondrial impairment leads to dysfunctional oxidative phosphorylation and elevated oxidative stress, which can suppress MuSC proliferation and compromise muscle regeneration[50]. Irisin has been reported to enhance mitochondrial biogenesis, mitigate oxidative stress, and promote mitophagy[51-53]. Therefore, we examined mitochondrial morphology and mitophagy in MuSCs following TNF-α challenge and EMEV intervention. TNF-α treatment induced mitochondrial fragmentation and reduced the number of active mitochondria in MuSCs (Fig. 2f), alongside impaired mitophagy as visualized by TEM (Fig. 2g). EMEV supplementation restored mitophagy and increased the count of active mitochondria in TNF-α-treated MuSCs, whereas EVs failed to elicit similar effects. Moreover, EMEVs promoted the expression of PINK1 and Parkin (Fig. 2h), central regulators of mitophagy. PINK1 acts as a sensor of mitochondrial damage, localizing to the outer mitochondrial membrane to recruit Parkin, which subsequently initiates autophagosome and autolysosome formation[54]. Using TOMM20 staining, we demonstrated that EMEVs enhanced the co-localization of PINK1 and Parkin and their mitochondrial translocation more effectively than EVs (Fig. 2i). This process is essential for PINK1-Parkin-mediated maintenance of mitochondrial function and initiation of mitophagy in response to damage [54]. Failure to remove promptly damaged mitochondria can trigger various pro-inflammatory signaling pathways[55-57]. Among these, the NLRP3/Casp-1 and cGAS/STING pathways are well-established to inhibit cellular proliferation[58-62]. In TNF-α-treated MuSCs, we observed that mitochondrial damage prompted the release of reactive oxygen species (ROS) and mitochondrial DNA (mtDNA) into the cytosol, which are pivotal for NLRP3 inflammasome and STING activation. EMEV intervention effectively reduced the levels of GSDMD-N, NLRP3, cleaved Casp1, and IL-1β in TNF-α-treated MuSCs (Fig. S6), and also attenuated activation of the cGAS/STING pathway (Fig. S7). Together, these data demonstrate that EMEVs improve mitochondrial function in inflammatory-challenged MuSCs by promoting mitophagy and suppressing pro-inflammatory pathways activation triggered by mitochondrial damage (Fig. S8). 3.4 EMEVs prevented myotubes from atrophy and converted their cytokine profile Myocytes and myotubes are terminally differentiated cells with a limited lifespan and are incapable of self-renewal. Therefore, myotubes derived from myoblasts or MuSC fusion are commonly used to model myocytes in vitro . In this study, we employed MuSC-derived myotubes to examine alterations in their structural maintenance and cytokine profiles in response to EMEVs (Fig. 3a), which serve as key indicators of their potential to enhance muscle mass. Myosin heavy chain (MyHC), a component of the thick filaments in skeletal muscle, is critical for the formation and maintenance of functional myotubes. We found that myotubes treated with PBS or EVs exhibited a significant reduction in MyHC-positive area after three days, whereas those treated with EMEVs showed no notable change (Figs. 3b & 3c), indicating that EMEVs may prevent structural degeneration of myotubes. Consistently, EMEVs (irisin KD) failed to rescue the loss of myotube structure, whereas EMEVs (LDHA KD) retained this protective capacity (Fig. S9), suggesting that irisin played a primary role in mediating the anti-atrophic effects of EMEVs. Subsequently, MuSC-derived myotubes were exposed to low-intensity mechanical stimulation (10% and 15% elongation, 0.5 Hz) to simulate daily low-intensity exercise, along with incubation with EVs or EMEVs. As stimulus intensity increased, the activities of mitochondrial respiratory chain complexes I and II were elevated across all groups (Figs. 3d & 3e). Notably, the EMEV-treated group displayed the greatest exercise-potentiating effects, with activity levels consistently higher than those of EV-treated cells at the same elongation; moreover, EMEVs at 10% elongation even surpassed EVs at 15% elongation. These results suggested that EMEV treatment prevented myotubes from atrophy and enhanced the adaptive response to exercise-mimetic mechanical stimulation. Exercise and mechanical stimuli are known to activate AMPK and inhibit NF-κB—key pathways regulating energy metabolism and inflammation, respectively[63, 64]. We therefore assessed AMPK and NF-κB signaling in myotubes subjected to concurrent mechanical stimulation and vesicle treatment by measuring levels of p-AMPK, nuclear NF-κB p65, and p50/p105. As stimulus intensity rose, p-AMPK levels increased accordingly (Fig. 3f). Remarkably, EMEV-treated cells again exhibited exercise-potentiating effects, with p-AMPK concentrations substantially exceeding those in EV-treated cells at matched elongation levels; EMEVs at 10% elongation even induced higher p-AMPK than EVs at 15%. Conversely, levels of p65 and p50/p105 decreased, reflecting NF-κB inhibition (Figs. 3g & 3h). Together, these data indicate that mechanical strain elicits beneficial exercise-like responses in myotubes, including AMPK activation and NF-κB suppression, and that EMEVs enhance cellular adaptation to such stimuli. AMPK and NF-κB signaling can drive distinct downstream events, including expression of irisin and proinflammatory cytokines. We thus evaluated the cytokine profiles of myotubes under mechanical stimulation with EV or EMEV treatment. Both 10% and 15% elongation stimulated increased irisin secretion and decreased release of MSTN, TNF-α, IFN-γ, IL-1α, and IL-1β, regardless of EMEV addition (Fig. 3i). However, EMEVs markedly amplified these mechanical stimulation-induced changes. Collectively, these findings indicate that EMEVs not only mimic exercise-induced effects but also potentiate the adaptive responses to mechanical stimulation. 3.5 EMEVs promoted osteogenesis in vitro Recent evidence indicates that EVs derived from skeletal muscle may promote osteogenesis[34], while atrophic skeletal muscle can impair osteogenic differentiation[35]. Given the abundance of irisin and LDHA in EMEVs observed in this study, we further investigated their potential role in facilitating osteogenesis in osteoblasts. MC3T3 pre-osteoblasts were used as an in vitro osteogenesis model and cultured with EMEVs or EVs under standard osteogenic induction conditions. Immunofluorescence staining after 3 and 7 days of induction revealed that EMEVs markedly enhanced the expression of key osteogenic transcription factors, Runx2 and Sp7 (Osterix), compared to EVs (Figs. 4a-4c). In contrast, EV-treated cells exhibited no significant increase in Runx2 or Sp7 expression relative to the control until day 7. These results suggest that EVs have a considerably weaker effect than EMEVs on initiating osteogenic transcription. Furthermore, we measured several osteokines in the conditioned medium using ELISA. Among them, osteocalcin (OCN) secreted by osteoblasts exerts multiple endocrine functions primarily through its uncarboxylated form (Glu-OCN)[65]. For example, Glu-OCN stimulates insulin secretion in pancreatic β-cells, promotes testosterone synthesis in Leydig cells, and enhances glucose uptake in white adipose and muscle tissues[66]. We therefore compared Glu-OCN levels across groups. As shown in Fig. S10, osteoblasts secreted higher amounts of total OCN and Glu-OCN following EMEV treatment, suggesting a potential enhancement of bone-derived endocrine signaling. However, no statistically significant differences were observed in other osteokines, including RANKL, OPG, and SLIT3. Consistent trends were observed in the transcriptional profiles of osteogenesis-related genes after 14 days of induction with EMEVs or EVs. qPCR analysis was performed for markers representing early (Runx2, Sp7, Col1a1), intermediate/late (Ocn, Opn) stages of osteogenesis. At day 3, only EMEV-treated cells showed pronounced upregulation of all early osteogenic genes, while EV treatment only elevated Sp7 expression (Fig. 4d). At days 7 and 14, EV-treated cells exhibited no significant differences in any osteogenic gene compared to the control, whereas EMEV-treated cells displayed time-dependent upregulation consistent with osteogenic progression, including stronger induction of late markers (Ocn and Opn) at day 14. Interestingly, this pro-osteogenic effect was significantly attenuated when using EMEVs with either irisin or LDHA knockdown (Fig. S11), indicating that both irisin and LDHA were essential bioactive components mediating the osteogenic promotion by EMEVs. In addition to gene expression, mineralization was evaluated by Alizarin Red S (ARS) staining (Figs. 4e & 4f). EMEV treatment significantly enhanced mineralized nodule formation, while EV treatment did not differ statistically from the control. The superior pro-osteogenic effects of EMEVs are likely attributable to their higher content of bioactive molecules such as irisin and LDHA. 3.6 Synthesis and characterization of DT-EMEVs Although EMEVs exerted beneficial effects on muscle cells and osteoblasts in monoculture systems, their therapeutic efficacy upon systemic administration in vivo could be limited by natural biodistribution patterns. Natural EVs tend to accumulate in highly vascularized and filtration-prone organs such as the liver, spleen, lungs, and kidneys, with only minimal quantities reaching musculoskeletal tissues[67, 68]. Specific surface modifications can enhance the targeted delivery of EVs to particular cell types or organs[67, 68]. For example, functionalization with SDSSD or ASSLNIA peptides has been shown to promote EV uptake by osteoblasts and muscle cells, respectively[69, 70]. Therefore, we conjugated SDSSD and ASSLNIA peptides to EMEVs via a diacyllipid insertion approach, as previously described[69]. After incubation, red fluorescently labeled SDSSD and green fluorescently labeled ASSLNIA demonstrated clear colocalization with EMEVs (Figs. 5a & 5b), confirming successful surface modification. These engineered vesicles, designated dual-targeted EMEVs (DT-EMEVs), were designed for simultaneous delivery to bone and muscle. DT-EMEVs displayed a morphology similar to unmodified EMEVs under TEM (Fig. 5c), with marginally larger sizes and stable physicochemical properties in terms of size and zeta potential (Figs. 5d & 5e). Prior to in vivo studies, we evaluated the in vitro uptake of DT-EMEVs by osteoblasts and myotubes. Both DT-EMEVs and bone-targeted EMEVs (BT-EMEVs, modified with SDSSD) showed enhanced uptake in osteoblasts compared to unmodified EMEVs within 12 hours of co-incubation (Fig. 5f). However, after 24 hours, uptake levels were comparable across all EMEV groups. Similar trends were observed in myotubes treated with DT-EMEVs and muscle-targeted EMEVs (MT-EMEVs, modified with ASSLNIA). This time-dependent effect may be attributed to the inherent biocompatibility of EMEVs, whose membrane composition resembles that of natural cell membranes, enabling eventual uptake even in the absence of targeting ligands. Nonetheless, DT-EMEVs exhibited significantly improved targeting efficiency within the first 12 hours. We next examined the in vivo biodistribution of DiR-labeled EMEVs following systemic administration via tail vein injection (Fig. 5g). Mice were euthanized at 4- and 8-hours post-injection, and femur, tibia, gastrocnemius muscle, and major organs were harvested for ex vivo imaging. Both modified and unmodified EMEVs showed negligible signal in the brain and heart (Figs. 5h & 5i), consistent with the barriers posed by the blood-brain barrier and high perfusion rates[71]. BT-EMEVs and MT-EMEVs exhibited reduced accumulation in off-target organs (liver, spleen, lungs, kidneys) and targeted enrichment in bone and muscle tissues, respectively. DT-EMEVs showed combined biodistribution, with significant accumulation in both bone and muscle. These results were corroborated by fluorescence imaging of thigh sections (Figs. 5j & 5k), wherein DT-EMEV-injected mice displayed strong red fluorescence in both musculoskeletal tissues, confirming their dual-targeting capability and potential for tailored biodistribution in the treatment of OSP. 3.7 Anti-sarcopenia effects of DT-EMEVs in aged mice An in vivo OSP model was established using 18-month-old male C57BL/6J mice. Aged mice were randomly assigned to five groups: an old control (PBS-treated), and groups treated with EMEVs, MT-EMEVs, BT-EMEVs, and DT-EMEVs. Young (3-month-old) mice treated with PBS served as healthy controls (Fig. 6a). After eight weeks of weekly tail vein injections, aged mice (approximately 20 months old) underwent functional assessments followed by tissue collection for histological evaluation of muscle morphology and function. Muscle strength was evaluated using a grip test (Fig. 6b), and muscle endurance was assessed via hanging and exhaustive running tests (Figs. 6c & S12). Mice treated with DT-EMEVs were the only group that exhibited significantly improved muscle strength and endurance compared to the Old group, though values did not reach those of the Young group. All other treatment groups performed similarly to the Old group. DT-EMEV treatment did not significantly alter body weight or gastrocnemius muscle weight in aged mice (Figs. 6d & 6e); however, the ratio of gastrocnemius weight to body weight was significantly increased only in the DT-EMEV group (Fig. 6f). Transverse sections of gastrocnemius muscle were subjected to H&E staining to evaluate myofiber morphology. ImageJ was used to quantify myofiber cross-sectional area and maximum diameter (Figs. 6g-6i). Based on these measurements, the DT-EMEV group displayed the largest myofiber size and was the only group not significantly different from Young mice in both parameters. H&E staining also revealed substantial lipid infiltration among myofibers in the Old group, which was markedly reduced in the DT-EMEV group to a extent comparable to the Young group (Fig. 6i). Immunofluorescence staining for PINK1, ROS, and Pax7 showed that the DT-EMEV group had the highest PINK1 signal among treatment groups, similar to Young mice, suggesting enhanced PINK1-mediated mitophagy (Fig. 6j). ROS levels were most reduced in the DT-EMEV group, while Pax7 expression was highest, both approaching levels observed in Young mice, indicating attenuated oxidative stress and increased satellite cell activation, respectively (Figs. 6k & 6l). Additionally, serum irisin concentrations were significantly elevated in DT-EMEV-treated mice compared to the Old group, whereas MSTN levels were reduced (Fig. S13). This shift in myokine profile may further contribute to the preservation of muscle and bone mass and overall metabolic health. 3.8 Anti-osteoporosis effects of DT-EMEVs in aged mice The same aged mouse OSP model was employed to evaluate the anti-osteoporotic effects of the various EME treatment groups. Following 8 weeks of weekly administration, bone mass and microarchitecture were assessed using μCT. Severe age-related bone loss was observed in the Old group, which was modestly improved by EMEVs or MT-EMEVs/BT-EMEVs treatment, but most effectively reversed by DT-EMEVs (Fig. 7a). Correspondingly, key bone morphometric parameters reflecting bone mass, including bone mineral density (BMD) and trabecular bone volume per tissue volume (Tb.BV/TV), were significantly enhanced in the DT-EMEVs group compared to all other treatments (Fig. 7b). Similarly, microstructural parameters such as trabecular number (Tb. N) and trabecular separation (Tb. Sp) showed the most pronounced improvement in DT-EMEV-treated animals (Fig. 7c). The biomechanical properties of mouse tibiae, evaluated by three-point bending test, demonstrated that DT-EMEVs treatment resulted in the highest values for maximum load and stiffness (Fig. 7d). The mineral apposition rate (MAR), determined by calcein double labelling, was highest in mice receiving BT-EMEVs and DT-EMEVs (Figs. 7e & 7f). Immunofluorescence staining for osteogenic markers OCN and Sp7 also revealed the strongest expression in the DT-EMEVs group (Fig. 7g), indicating markedly enhanced osteogenesis in vivo. Serum levels of bone turnover markers, including OCN, TRAP5b, β-CTX, P1NP, and BALP, were measured. Both total OCN and Glu-OCN were elevated in the BT-EMEVs and DT-EMEVs groups (Fig. 7h), suggesting a systemic metabolic benefit in aged mice. Furthermore, DT-EMEVs treatment significantly increased the bone formation marker P1NP and decreased the bone resorption marker β-CTX (Fig. S14), collectively supporting an anti-osteoporotic effect. Other markers such as BALP and TRAP5b also trended lower, though not significantly compared to other groups. qPCR analysis of osteogenic genes (Runx2, Sp7, Alp, Col1a, Ocn, Opn) in bone tissue confirmed the superior bone-preserving effects of DT-EMEVs, showing the highest expression levels among all groups (Figs. 7i-7k). Finally, blood biochemical analysis indicated excellent biosafety profiles for DT-EMEVs and all other EME treatment groups. Parameters including ALT, AST, BUN, Scr, and TBil showed no significant differences compared to Old and Young controls, confirming preserved hepatic and renal function (Fig. S15). 4. Conclusions This study developed an exercise-mimetic therapy using engineered EVs produced from mechanically stimulated iPSC-derived myotubes. Enriched with irisin and LDHA, these vesicles activated exercise-related pathways and promoted myogenesis and osteogenesis in vitro . Dual-targeted EMEVs specifically accumulated in muscle and bone tissue in vivo , and restored muscle function, bone density, and strength in an aged murine model of osteosarcopenia, demonstrating a novel targeted strategy for musculoskeletal disorders. Supplementary Information The online version contains supplementary material available. Figures Scheme 1. Schematic illustration depicting the preparation of EMEVs, DT-EMEVs and their therapeutic mechanism for OSP. a Human iPSCs are differentiated into myotubes and subjected to mechanical strain, releasing exercise mimetic extracellular vesicles (EMEVs) enriched with exercise-inducible cargos (e.g., irisin and LDHA). b In vitro, EMEVs exert multi-targeted effects: they promote proliferation and enhance mitochondrial function in muscle satellite cells (MuSCs); attenuate atrophy and remodel cytokine secretion profiles in myotubes; and stimulate osteogenic differentiation while increasing secretion of uncarboxylated osteocalcin (Glu-OCN) in pre-osteoblasts. c To improve bone and muscle targeting in vivo , EMEVs are engineered into dual-targeted EMEVs (DT-EMEVs) via surface conjugation of bone-targeted peptide (SDSSD) and muscle-targeted peptide (ASSLNIA). Systemically administered DT-EMEVs accumulate in target tissues, restoring muscle and bone mass in aged OSP rat model. Fig. 1 Preparation and characterization of EMEVs. a Schematic illustration of the origins of EVs and EMEVs. b TEM photographs of EVs and EMEVs. Scale bar = 100 nm. c Comparison of mean sizes (P value = 0.69) and zeta potentials (P value = 0.22) of EVs and EMEVs. d Size distribution percentages of EVs and EMEVs by DLS analysis. e Exosomal protein markers of EVs and EMEVs and their parent cells determined by western blotting. f iMyotubes cultured in the static, exercise mimic, and exercise mimic plus AMPK inhibitor conditions for 24 h, followed by western blotting for detecting the p‐AMPK, total AMPK and PGC-1α protein expression. β-Actin served as the loading control. g Quantification of EV production by determining the total protein content. h Quantification of EV irisin content by ELISA analysis. i Schematic illustration of the exercise mimic stimulation promoting EV production by Ca 2+ influx and AMPK/PGC-1α activation. (NS, no significant difference; #, p < 0.0001). Fig. 2 EMEVs promoted the proliferation of MuSCs and improved their mitochondrial function. a Immunofluorescence staining for Pax7 in MuSCs after PBS, EVs and EMEVs treatment for 48 h. Scale bar = 100 μm. b Quantification of Pax7 fluorescence intensity. c Cell viability of MuSCs for 4 days after TNF-α plus PBS/EVs/EMEVs treatment. d Flow cytometry analysis of MuSCs cell cycle after TNF-α plus PBS/EVs/EMEVs treatment for 48 h. PBS treatment served as control. e Quantification of cell cycle by flow cytometry analysis. f Confocal images of active mitochondria (red) stained by Mitotracker and F-actin cytoskeleton stained by phalloidine (green). Scale bar = 10 μm. The morphology of mitochondria was analysed using ImageJ software. g Autophagosomes observed under TEM. Scale bar = 1 μm. h The expression of PINK and Parkin determined by western blotting. i Confocal images of immunofluorescence staining for TOMM20 (red) and PINK (green, top line)/Parkin (green, bottom line). Co-localization shown in yellow. Scale bar = 10 μm. j Schematic illustration of mechanism by which EMEVs may improve the mitochondrial function of MuSCs and alleviate the subsequent inflammation, through promoting PINK/Parkin dependent mitophagy. NS, no significant difference; *, p < 0.05; **, p < 0.01; #, p < 0.0001. Fig. 3 EMEVs prevented myotubes from atrophy and converted their cytokine profile a Schematic illustrating the experiment design. b Immunofluorescence staining for MyHC (red) and DAPI (blue) in myotubes formed from MuSCs after incubating with PBS/EVs/EMEVs for 3 days. Scale bar = 10 μm. c Quantification of MyHC positive areas. d Relative enzymatic activities of respiratory chain complex I after PBS/EVs/EMEVs incubation for 24 h under static, 10% elongation and 15% elongation strain stimulation. e Relative enzymatic activities of respiratory chain complex II after the above treatment. f-h Quantification of ( f ) p-AMPK, ( g ) nuclear p65 and ( h ) p50/p105 concentrations in myotubes formed from MuSCs after the above treatment, n=5. i Heatmap of cytokine profile in myotubes formed from MuSCs after different treatments, n=5. NS, no significant difference; *, p < 0.05; **, p < 0.01; #, p < 0.0001. Fig. 4 EMEVs promoted osteogenesis in vitro . a Confocal image of immunofluorescence staining for F-actin cytoskeleton stained by phalloidine (green), and Runx2/Sp7 stained by antibodies (red) in MC3T3 cells after EVs and EMEVs treatment for 3 and 7 days. PBS treated cells served as control. Scale bar = 10 μm. b, c Quantification of Runx2/Sp7 at Day 3 and 7. d Relative expression of osteogenic differentiation related genes following osteogenic induction (n = 3). e Representative microscopy images of ARS staining at Day 7 and 14 of osteogenic induction. Scale bar = 100 μm. f Quantitative analysis of ARS staining (n = 3). NS, no significant difference; *, p < 0.05; **, p < 0.01; ***, p < 0.001; #, p < 0.0001. Fig. 5 Synthesis and characterization of DT-EMEVs. a Confocal images of DT-EMEVs. Red fluorescence: Cy5 labelled SDSSD; Green fluorescence: FITC labelled ASSLNIA. Scale bar = 10 μm. b Quantification of red and green fluorescence at each coordinate. c TEM images of DT-EMEVs. Scale bar = 100 nm. d Size distribution percentages of DT-EMEVs by DLS analysis. e The diameter and zeta potential variation of DT-EMEVs detected by DLS over one week. f The uptake percentage of EMEVs/BT-EMEVs/MT-EMEVs/DT-EMEVs detected by flow cytometry in osteoblasts and myotubes during incubation for 24 h. g Schematic illustration of experiment design for in vivo biodistribution. h Biodistribution of DiR-labelled EMEVs/BT-EMEVs/MT-EMEVs/DT-EMEVs at 4 and 8 h after injection. i Quantification of DiR fluorescence intensity (n = 3). j Microscopy images of murine thighs dissected at 4 and 8 h after injection. Red fluorescence: DiR; Blue fluorescence: DAPI. k Quantification of DiR fluorescence intensity in the sections (n = 3). NS, no significant difference; *, p < 0.05; **, p < 0.01; ***, p < 0.001; #, p < 0.0001. Fig. 6 Anti-sarcopenia effects of DT-EMEVs in aged mice. a Schematic illustration depicting the in vivo study design using the OSP model. b, c Quantification of grip strength and hanging time after treatment (n = 6). d-f Quantification of body weight (d), gastrocnemius weight (e) and gastrocnemius weight percentage (f) after treatment (n = 6). g, h Quantification of myofiber area (g) and maximum myofiber diameter (h) (n = 6). i Representative microscopy images of H&E staining, immunofluorescence staining for PINK, ROS staining, and immunofluorescence staining for Pax7. j-l Quantification of PINK (j), ROS (k) and PAX7 (l) fluorescence intensity (n = 3). NS, no significant difference; *, p < 0.05; **, p < 0.01; ***, p < 0.001; #, p < 0.0001. Fig. 7 Anti-osteoporosis effects of DT-EMEVs in aged mice. a Representative μCT images of 3D reconstructed bone trabeculae. b Quantitation of BMD and Tb. BV/TV (n = 6). c Quantitation of Tb. N and Tb. Sp (n = 6). d Quantitation of tibia maximum load and stiffness measured by three-point bending test (n = 6). e Representative microscopy images of calcein double-labelling (scale bar = 100 μm) and immunofluorescence staining for OCN and Osterix (scale bar = 100 μm). f Quantitation of MAR (n = 3). g Quantitation of OCN and Sp7 fluorescence intensity (n = 3). h Quantitation of total OCN and Glu-OCN concentration (n = 6). i Relative expression of Runx2 and Sp7 in the femur tissues (n = 3). j Relative expression of Alp and Col1a in the femur tissues (n = 3). k Relative expression of Ocn and Opn in the femur tissues (n = 3). NS, no significant difference; *, p < 0.05; **, p < 0.01; ***, p < 0.001; #, p < 0.0001. References [1] B. Kirk, J. Zanker, G. Duque, Osteosarcopenia: epidemiology, diagnosis, and treatment-facts and numbers. J Cachexia Sarcopenia Muscle. 11, 609-618 (2020). https://doi.org/10.1002/jcsm.12567 [2] S.S.Y. Yeung, E.M. Reijnierse, V.K. Pham, et al., Sarcopenia and its association with falls and fractures in older adults: A systematic review and meta-analysis. J Cachexia Sarcopenia Muscle. 10, 485-500 (2019). https://doi.org/10.1002/jcsm.12411 [3] W.S. Lim, C.Y. Cheong, J.P. Lim, et al., Singapore Clinical Practice Guidelines For Sarcopenia: Screening, Diagnosis, Management and Prevention. J Frailty Aging. 11, 348-369 (2022). https://doi.org/10.14283/jfa.2022.59 [4] M. Gomarasca, G. Banfi, G. Lombardi, Myokines: The endocrine coupling of skeletal muscle and bone. Adv Clin Chem. 94, 155-218 (2020). https://doi.org/10.1016/bs.acc.2019.07.010 [5] A.M. Gonzalez-Gil, L. Elizondo-Montemayor, The Role of Exercise in the Interplay between Myokines, Hepatokines, Osteokines, Adipokines, and Modulation of Inflammation for Energy Substrate Redistribution and Fat Mass Loss: A Review. Nutrients. 12, 1899 (2020). https://doi.org/10.3390/nu12061899 [6] F. Wang, K.-F. So, J. Xiao, et al., Organ-organ communication: The liver’s perspective. Theranostics. 11, 3317-3330 (2021). https://doi.org/10.7150/thno.55795 [7] M.C.K. Severinsen, B.K. Pedersen, Muscle–Organ Crosstalk: The Emerging Roles of Myokines. Endocrine Reviews. 41, 594-609 (2020). https://doi.org/10.1210/endrev/bnaa016 [8] J. Zhang, P. Valverde, X. Zhu, et al., Exercise-induced irisin in bone and systemic irisin administration reveal new regulatory mechanisms of bone metabolism. Bone Research. 5, 16056 (2017). https://doi.org/10.1038/boneres.2016.56 [9] C. Vinel, L. Lukjanenko, A. Batut, et al., The exerkine apelin reverses age-associated sarcopenia. Nat Med. 24, 1360-1371 (2018). https://doi.org/10.1038/s41591-018-0131-6 [10] C. Chi, H. Fu, Y.-H. Li, et al., Exerkine fibronectin type-III domain-containing protein 5/irisin-enriched extracellular vesicles delay vascular ageing by increasing SIRT6 stability. European Heart Journal. 43, 4579-4595 (2022). https://doi.org/10.1093/eurheartj/ehac431 [11] L.S. Chow, R.E. Gerszten, J.M. Taylor, et al., Exerkines in health, resilience and disease. Nature Reviews Endocrinology. 18, 273-289 (2022). https://doi.org/10.1038/s41574-022-00641-2 [12] L. Tao, J. Wang, K. Wang, et al., Exerkine FNDC5/irisin‐enriched exosomes promote proliferation and inhibit ferroptosis of osteoblasts through interaction with Caveolin‐1. Aging Cell. 23, e14181 (2024). https://doi.org/10.1111/acel.14181 [13] R. Sheng, M. Cao, M. Song, et al., Muscle-bone crosstalk via endocrine signals and potential targets for osteosarcopenia-related fracture. Journal of Orthopaedic Translation. 43, 36-46 (2023). https://doi.org/10.1016/j.jot.2023.09.007 [14] J. Suh, N.K. Kim, S.H. Lee, et al., GDF11 promotes osteogenesis as opposed to MSTN, and follistatin, a MSTN/GDF11 inhibitor, increases muscle mass but weakens bone. Proc Natl Acad Sci U S A. 117, 4910-4920 (2020). https://doi.org/10.1073/pnas.1916034117 [15] X.H. Liu, J.P. Pan, W.A. Bauman, et al., Myostatin inhibits insulin-like growth factor 1-dependent citrate secretion and osteogenesis via nicotinamide adenine dinucleotide phosphate oxidase-4 in a mouse mesenchymal stem cell line. Ann N Y Acad Sci. 1517, 203-212 (2022). https://doi.org/10.1111/nyas.14894 [16] B. Dankbar, M. Fennen, D. Brunert, et al., Myostatin is a direct regulator of osteoclast differentiation and its inhibition reduces inflammatory joint destruction in mice. Nat Med. 21, 1085-1090 (2015). https://doi.org/10.1038/nm.3917 [17] L. Tang, T. Zhao, Y. Kang, et al., MSTN is an important myokine for weight-bearing training to attenuate bone loss in ovariectomized rats. J Physiol Biochem. 78, 61-72 (2022). https://doi.org/10.1007/s13105-021-00838-5 [18] J. Hatakeyama, S. Inoue, H. Jiang, et al., Exercise-induced interactions between skeletal muscle and bone via myokines and osteokine in mice: Role of FNDC5/irisin, IGF-1, and osteocalcin. Bone. 190, 117314 (2025). https://doi.org/10.1016/j.bone.2024.117314 [19] H. Chen, B. Hu, X. Lv, et al., Prostaglandin E2 mediates sensory nerve regulation of bone homeostasis. Nat Commun. 10, 181 (2019). https://doi.org/10.1038/s41467-018-08097-7 [20] D. Zhao, M.A. Riquelme, T. Guda, et al., Connexin hemichannels with prostaglandin release in anabolic function of bone to mechanical loading. Elife. 11, e74365 (2022). https://doi.org/10.7554/eLife.74365 [21] N. Bonnet, L. Bourgoin, E. Biver, et al., RANKL inhibition improves muscle strength and insulin sensitivity and restores bone mass. J Clin Invest. 129, 3214-3223 (2019). https://doi.org/10.1172/jci125915 [22] F. Pin, A.J. Jones, J.R. Huot, et al., RANKL Blockade Reduces Cachexia and Bone Loss Induced by Non-Metastatic Ovarian Cancer in Mice. J Bone Miner Res. 37, 381-396 (2022). https://doi.org/10.1002/jbmr.4480 [23] S.N. Jayash, D. Hamoudi, L.A. Stephen, et al., Anti-RANKL Therapy Prevents Glucocorticoid-Induced Bone Loss and Promotes Muscle Function in a Mouse Model of Duchenne Muscular Dystrophy. Calcif Tissue Int. 113, 449-468 (2023). https://doi.org/10.1007/s00223-023-01116-w [24] T. Ono, M. Hayashi, F. Sasaki, et al., RANKL biology: bone metabolism, the immune system, and beyond. Inflamm Regen. 40, 2 (2020). https://doi.org/10.1186/s41232-019-0111-3 [25] J. Chen, R. Zhou, Y. Feng, et al., Molecular mechanisms of exercise contributing to tissue regeneration. Signal Transduct Target Ther. 7, 383 (2022). https://doi.org/10.1038/s41392-022-01233-2 [26] M. Gleeson, N.C. Bishop, D.J. Stensel, et al., The anti-inflammatory effects of exercise: mechanisms and implications for the prevention and treatment of disease. Nature Reviews Immunology. 11, 607 (2011). https://doi.org/10.1038/nri3041 [27] D. Walzik, T.Y. Wences Chirino, P. Zimmer, et al., Molecular insights of exercise therapy in disease prevention and treatment. Signal Transduct Target Ther. 9, 138 (2024). https://doi.org/10.1038/s41392-024-01841-0 [28] V.A. Narkar, M. Downes, R.T. Yu, et al., AMPK and PPARdelta agonists are exercise mimetics. Cell. 134, 405-415 (2008). https://doi.org/10.1016/j.cell.2008.06.051 [29] C. Billon, E. Schoepke, A. Avdagic, et al., A Synthetic ERR Agonist Alleviates Metabolic Syndrome. J Pharmacol Exp Ther. 388, 232-240 (2024). https://doi.org/10.1124/jpet.123.001733 [30] T. Ono, R. Denda, Y. Tsukahara, et al., Simultaneous augmentation of muscle and bone by locomomimetism through calcium-PGC-1α signaling. Bone Research. 10, 52 (2022). https://doi.org/10.1038/s41413-022-00225-w [31] S. Deng, H. Cao, X. Cui, et al., Optimization of exosome-based cell-free strategies to enhance endogenous cell functions in tissue regeneration. Acta Biomater. 171, 68 (2023). https://doi.org/10.1016/j.actbio.2023.09.023 [32] M.A. Kumar, S.K. Baba, H.Q. Sadida, et al., Extracellular vesicles as tools and targets in therapy for diseases. Signal Transduct Target Ther. 9, 27 (2024). https://doi.org/10.1038/s41392-024-01735-1 [33] J.S. Choi, H.I. Yoon, K.S. Lee, et al., Exosomes from differentiating human skeletal muscle cells trigger myogenesis of stem cells and provide biochemical cues for skeletal muscle regeneration. J Control Release. 222, 107-115 (2016). https://doi.org/10.1016/j.jconrel.2015.12.018 [34] S. Ma, X. Xing, H. Huang, et al., Skeletal muscle-derived extracellular vesicles transport glycolytic enzymes to mediate muscle-to-bone crosstalk. Cell Metabolism. 35, 2028 (2023). https://doi.org/10.1016/j.cmet.2023.10.013 [35] H. Huang, S. Ma, X. Xing, et al., Muscle-derived extracellular vesicles improve disuse-induced osteoporosis by rebalancing bone formation and bone resorption. Acta Biomaterialia. 157, 609 (2023). https://doi.org/10.1016/j.actbio.2022.12.019 [36] J. He, W. Ren, W. Wang, et al., Exosomal targeting and its potential clinical application. Drug Deliv Transl Res. 12, 2385-2402 (2022). https://doi.org/10.1007/s13346-021-01087-1 [37] L. Caron, D. Kher, K.L. Lee, et al., A Human Pluripotent Stem Cell Model of Facioscapulohumeral Muscular Dystrophy-Affected Skeletal Muscles. Stem Cells Transl Med. 5, 1145 (2016). https://doi.org/10.5966/sctm.2015-0224 [38] Y. Feng, X.Y. Tian, P. Sun, et al., Simultaneous Study of Mechanical Stretch-Induced Cell Proliferation and Apoptosis on C2C12 Myoblasts. Cells Tissues Organs. 205, 189 (2018). https://doi.org/10.1159/000490239 [39] M. Borja-Gonzalez, J.C. Casas-Martinez, B. McDonagh, et al., Inflamma-miR-21 Negatively Regulates Myogenesis during Ageing. Antioxidants (Basel). 9, 345 (2020). https://doi.org/10.3390/antiox9040345 [40] G.K. Shang, L. Han, Z.H. Wang, et al., Sarcopenia is attenuated by TRB3 knockout in aging mice via the alleviation of atrophy and fibrosis of skeletal muscles. J Cachexia Sarcopenia Muscle. 11, 1104-1120 (2020). https://doi.org/10.1002/jcsm.12560 [41] M.M. Maneshi, P.A. Gottlieb, S.Z. Hua, A Microfluidic Approach for Studying Piezo Channels. Curr Top Membr. 79, 309 (2017). https://doi.org/10.1016/bs.ctm.2016.11.005 [42] S. Guo, L. Debbi, B. Zohar, et al., Stimulating Extracellular Vesicles Production from Engineered Tissues by Mechanical Forces. Nano Letters. 21, 2497 (2021). https://doi.org/10.1021/acs.nanolett.0c04834 [43] J. Liu, J. Wang, Z. Wang, et al., PGC-1α/LDHA signaling facilitates glycolysis initiation to regulate mechanically induced bone remodeling under inflammatory microenvironment. Bone. 185, 117132 (2024). https://doi.org/10.1016/j.bone.2024.117132 [44] Z. Zhang, W. Luo, G. Chen, et al., Chicken muscle antibody array reveals the regulations of LDHA on myoblast differentiation through energy metabolism. Int J Biol Macromol. 254, 127629 (2024). https://doi.org/10.1016/j.ijbiomac.2023.127629 [45] H. Zhang, X. Wu, J. Liang, et al., Irisin, an exercise-induced bioactive peptide beneficial for health promotion during aging process. Ageing Res Rev. 80, 101680 (2022). https://doi.org/10.1016/j.arr.2022.101680 [46] X. Chen, J. Yan, F. He, et al., Mechanical stretch induces antioxidant responses and osteogenic differentiation in human mesenchymal stem cells through activation of the AMPK-SIRT1 signaling pathway. Free Radic Biol Med. 126, 187 (2018). https://doi.org/10.1016/j.freeradbiomed.2018.08.001 [47] Y. Seo, H. Kang, J. Park, Shear stress-induced influx of extracellular calcium ions: a pivotal trigger amplifying the production of mesenchymal stem cell-derived extracellular vesicles. Nanoscale. 17, 3861 (2024). https://doi.org/10.1039/d4nr01215a [48] C.Y. Wong, H. Al-Salami, C.R. Dass, C2C12 cell model: its role in understanding of insulin resistance at the molecular level and pharmaceutical development at the preclinical stage. J Pharm Pharmacol. 72, 1667-1693 (2020). https://doi.org/10.1111/jphp.13359 [49] J. Li, X. Yi, Z. Yao, et al., TNF Receptor-Associated Factor 6 Mediates TNFalpha-Induced Skeletal Muscle Atrophy in Mice During Aging. J Bone Miner Res. 35, 1535-1548 (2020). https://doi.org/10.1002/jbmr.4021 [50] X. Hong, J. Isern, S. Campanario, et al., Mitochondrial dynamics maintain muscle stem cell regenerative competence throughout adult life by regulating metabolism and mitophagy. Cell Stem Cell. 29, 1298-1314.e1210 (2022). https://doi.org/10.1016/j.stem.2022.07.009 [51] G. Li, Z. Jian, H. Wang, et al., Irisin Promotes Osteogenesis by Modulating Oxidative Stress and Mitophagy through SIRT3 Signaling under Diabetic Conditions. Oxid Med Cell Longev. 2022, 3319056 (2022). https://doi.org/10.1155/2022/3319056 [52] X. Zhang, L. Liang, F. Wang, et al., Irisin‐Encapsulated Mitochondria‐Targeted Biomimetic Nanotherapeutics for Alleviating Acute Kidney Injury. Advanced Science. 11, e2402805 (2024). https://doi.org/10.1002/advs.202402805 [53] C. Zhao, Y. Wu, S. Zhu, et al., Irisin Protects Musculoskeletal Homeostasis via a Mitochondrial Quality Control Mechanism. International Journal of Molecular Sciences. 25, 10116 (2024). https://doi.org/10.3390/ijms251810116 [54] D.P. Narendra, R.J. Youle, The role of PINK1–Parkin in mitochondrial quality control. Nature Cell Biology. 26, 1639-1651 (2024). https://doi.org/10.1038/s41556-024-01513-9 [55] J.L. Turley, H.B.T. Moran, C.P. McEntee, et al., Chitin-derived polymer deacetylation regulates mitochondrial reactive oxygen species dependent cGAS-STING and NLRP3 inflammasome activation. Biomaterials. 275, 120961 (2021). https://doi.org/10.1016/j.biomaterials.2021.120961 [56] Z. Zhong, S. Liang, E. Sanchez-Lopez, et al., New mitochondrial DNA synthesis enables NLRP3 inflammasome activation. Nature. 560, 198-203 (2018). https://doi.org/10.1038/s41586-018-0372-z [57] H. Xian, K. Watari, E. Sanchez-Lopez, et al., Oxidized DNA fragments exit mitochondria via mPTP- and VDAC-dependent channels to activate NLRP3 inflammasome and interferon signaling. Immunity. 55, 1370-1385.e1378 (2022). https://doi.org/10.1016/j.immuni.2022.06.007 [58] F. Chen, G. Jiang, H. Liu, et al., Melatonin alleviates intervertebral disc degeneration by disrupting the IL-1β/NF-κB-NLRP3 inflammasome positive feedback loop. Bone Res. 8, 10 (2020). https://doi.org/10.1038/s41413-020-0087-2 [59] Y. Jin, Y. Liu, L. Xu, et al., Novel role for caspase 1 inhibitor VX765 in suppressing NLRP3 inflammasome assembly and atherosclerosis via promoting mitophagy and efferocytosis. Cell Death & Disease. 13, 512 (2022). https://doi.org/10.1038/s41419-022-04966-8 [60] W. Zhang, G. Li, R. Luo, et al., Cytosolic escape of mitochondrial DNA triggers cGAS-STING-NLRP3 axis-dependent nucleus pulposus cell pyroptosis. Experimental & Molecular Medicine. 54, 129-142 (2022). https://doi.org/10.1038/s12276-022-00729-9 [61] J.I. Jiménez-Loygorri, B. Villarejo-Zori, Á. Viedma-Poyatos, et al., Mitophagy curtails cytosolic mtDNA-dependent activation of cGAS/STING inflammation during aging. Nature Communications. 15, 830 (2024). https://doi.org/10.1038/s41467-024-45044-1 [62] Q. Guo, D. Zhu, Y. Wang, et al., Targeting STING attenuates ROS induced intervertebral disc degeneration. Osteoarthritis and Cartilage. 29, 1213-1224 (2021). https://doi.org/10.1016/j.joca.2021.04.017 [63] B. Egan, J.R. Zierath, Exercise metabolism and the molecular regulation of skeletal muscle adaptation. Cell Metab. 17, 162-184 (2013). https://doi.org/10.1016/j.cmet.2012.12.012 [64] Y. Sun, Y. Wu, Y. Jiang, et al., Aerobic exercise inhibits inflammatory response in atherosclerosis via Sestrin1 protein. Exp Gerontol. 155, 111581 (2021). https://doi.org/10.1016/j.exger.2021.111581 [65] G. Karsenty, Osteocalcin: A Multifaceted Bone-Derived Hormone. Annu Rev Nutr. 43, 55-71 (2023). https://doi.org/10.1146/annurev-nutr-061121-091348 [66] C. Shan, J. Yue, W. Liu, Broadening the role of osteocalcin in the hypothalamic-pituitary-gonadal axis. J Endocrinol. 249, R43-r51 (2021). https://doi.org/10.1530/joe-20-0203 [67] Y. Liang, L. Duan, J. Lu, et al., Engineering exosomes for targeted drug delivery. Theranostics. 11, 3183-3195 (2021). https://doi.org/10.7150/thno.52570 [68] J. Wang, X. Li, S. Wang, et al., Bone-Targeted Exosomes: Strategies and Applications. Adv Healthc Mater. 12, e2203361 (2023). https://doi.org/10.1002/adhm.202203361 [69] Y. Cui, Y. Guo, L. Kong, et al., A bone-targeted engineered exosome platform delivering siRNA to treat osteoporosis. Bioact Mater. 10, 207-221 (2022). https://doi.org/10.1016/j.bioactmat.2021.09.015 [70] S.D. Jativa, N. Thapar, D. Broyles, et al., Enhanced Delivery of Plasmid DNA to Skeletal Muscle Cells using a DLC8-Binding Peptide and ASSLNIA-Modified PAMAM Dendrimer. Mol Pharm. 16, 2376-2384 (2019). https://doi.org/10.1021/acs.molpharmaceut.8b01313 [71] Q. Xu, Z. Xiao, Q. Yang, et al., Hydrogel-based cardiac repair and regeneration function in the treatment of myocardial infarction. Mater Today Bio. 25, 100978 (2024). https://doi.org/10.1016/j.mtbio.2024.100978 Information & Authors Information Version history V1 Version 1 26 October 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords exercise-mimetic therapy extracellular vesicles muscle-bone crosstalk Authors Affiliations Yongzhi Cui 0000-0002-7532-0151 Shanghai 6th Peoples Hospital Affiliated to Shanghai Jiao Tong University View all articles by this author Qian Xiang Peking University Third Hospital Department of Orthopaedics View all articles by this author Ke Zhao China Academy of Chinese Medical Sciences Wangjing Hospital View all articles by this author Jiao Jiao Li 0000-0002-3584-6765 University of Technology Sydney Faculty of Engineering and Information Technology View all articles by this author Benchi Che Shanghai 6th Peoples Hospital Affiliated to Shanghai Jiao Tong University View all articles by this author Kaiwen Zheng Shanghai 6th Peoples Hospital Affiliated to Shanghai Jiao Tong University View all articles by this author Yanchun Gao Shanghai 6th Peoples Hospital Affiliated to Shanghai Jiao Tong University View all articles by this author Shengrong Guo Shanghai Jiao Tong University School of Pharmacy View all articles by this author Dehao Fu [email protected] Shanghai 6th Peoples Hospital Affiliated to Shanghai Jiao Tong University View all articles by this author Metrics & Citations Metrics Article Usage 435 views 181 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Yongzhi Cui, Qian Xiang, Ke Zhao, et al. Dual-targeted exercise mimetic extracellular vesicles regulate the muscle-bone crosstalk to treat osteosarcopenia. Authorea . 26 October 2025. DOI: https://doi.org/10.22541/au.176150016.60795129/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176150016.60795129/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fed738feed9c13d',t:'MTc3OTMwMjIzMw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();