Exercise-derived exosomes: molecular mediators of systemic health and disease therapy

Abstract

Exosomes serve as pivotal nanoscale messengers in intercellular communication by transporting bioactive molecules such as miRNAs, proteins, and lipids that regulate physiological and pathological processes. Emerging evidence highlights exercise as a potent modulator of exosome biogenesis, dynamically altering their release kinetics, molecular cargo, and bioactivity across tissues. Exercise-derived exosomes disseminate systemic adaptations by delivering regulatory signals to noncontractile organs, thereby coordinating multitissue responses that underlie the protective and reparative benefits of physical activity. This review synthesizes current knowledge on the dynamic effects of acute and chronic exercise on exosome profiles and their therapeutic potential in treating neurological, cardiovascular, metabolic, and musculoskeletal disorders. This review further discusses how exosome engineering and precision medicine could harness exosomes as “exercise mimetics,” offering cell-free therapeutics for mobility-limited populations. By integrating exercise physiology with translational medicine, this work pioneers a new therapeutic paradigm where exosome-based molecular therapies replicate exercise’s multisystem benefits.

Introduction

The human body is a highly coordinated and complex system. Multiple factors, including genetics, environment, and lifestyle, work together to maintain physiological homeostasis and regulate the onset and progression of diseases. Among numerous health intervention measures, appropriate exercise is a widely proven, effective strategy for promoting physical health and preventing and treating various diseases [1]. Exercise enhances muscle strength and physical fitness while improving the function of multiple systems, including the nervous, metabolic, vascular, and musculoskeletal systems [2–7]. How exercise transforms local mechanical or metabolic signals into adaptive changes in distant organs and throughout the entire body remains unclear. The precise molecular communication mechanisms require further study [8].

The role of exosomes in physiological and pathological processes has attracted widespread attention [9]. Exosomes are extracellular vesicles (EVs) that measure approximately 30–150 nm in diameter. After multivesicular bodies fuse with the cell membrane, they are released and carry specific proteins, nucleic acids, lipids, and other biologically active cargo from the source cell [10]. Exosomes are key messengers in intercellular communication, precisely transmitting information to distant target cells via the circulatory system. They play a crucial role in physiological and pathological processes, such as tissue repair, immune regulation, metabolic homeostasis, and neural plasticity, by regulating the expression of genes and functional state of target cells [11, 12].

Exercise releases exosomes that mediate intercellular communication and regulate metabolic homeostasis throughout the body [13]. Acute and long-term exercise profoundly effect the quantity, origin, and molecular composition of exosomes released into the circulatory system from different tissues [14]. Exercise-derived exosomes carry bioactive molecules that travel through the bloodstream and transmit exercise-induced adaptive signals to organs and tissues that do not directly participate in contraction. These signals activate downstream signaling pathways, mediating the protective and reparative effects of exercise on multiple systems [12].

The mechanisms of exercise-derived exosomes in the nervous, vascular, metabolic, and musculoskeletal systems are complex and diverse but rarely systematically summarized. This review thoroughly explores the regulatory effects of exercise on exosome secretion and cargo. It focuses on the therapeutic potential of exercise-derived exosomes in neurological, vascular, metabolic, and musculoskeletal disorders. Finally, it summarizes the current research status and challenges of exercise-derived exosome and discusses their application prospects in precision rehabilitation and regenerative medicine. The findings provide a theoretical basis and research directions for the therapeutic application of exercise-derived exosomes in multiple diseases.

Overview of exosomes

Exosomes are EVs that are approximately 30–150 nm in diameter. They are secreted by most eukaryotic cells and have a typical double-layered lipid membrane structure [15]. Exosomes are key messengers that facilitate long-distance communication between cells. They are present in various body fluids, including blood, saliva, and cerebrospinal fluid (CSF) [16]. The cargo carried by exosomes, such as microRNAs (miRNAs), proteins, lipids, and metabolites, selectively enriches specific biomolecules from source cells and thereby accurately reflect the cells’ physiological or pathological state [17–19]. This characteristic renders exosomes a potential biomarker for liquid biopsies and disease diagnosis. They also provide a vehicle for developing cell-free, targeted treatment strategies [20].

Exosome biogenesis is a precisely regulated, multistep process that depends on the endocytic system. First, early endosomes evolve into late endosomes, whose membranes bud inward to form intraluminal vesicles (ILVs). The endosomal sorting complex required for transport (ESCRT) protein family, which includes tumor susceptibility gene 101 (TSG101) and Apoptosis-Linked gene 2-Interacting protein X (ALIX), primarily drives this process by recognizing ubiquitinated proteins and remodeling the membrane [21, 22]. ESCRT-independent pathways, such as the ceramide-mediated reorganization of lipid rafts or generation of ILVs involving membrane proteins (cluster of differentiation 63 and 9 [CD63 and CD9]), are also implicated. Multivesicular bodies (MVBs), which are rich in ILVs, fuse with either lysosomes, leading to the degradation of their contents, or the cell membrane and release ILVs into the extracellular space, thereby forming exosomes [23]. Exosome secretion is dynamically regulated by cellular stress signals, such as hypoxia and inflammation, and Rab GTPases, such as Rab27a. This finding indicates that exosome release is closely related to the changes in the microenvironment [24].

In terms of molecular composition, exosomes exhibit significant heterogeneity and functional specificity. Their membranes are rich in membrane proteins, such as CD9, CD63, and CD81, and heat shock proteins (HSPs), such as HSP70 and HSP90. These proteins are commonly used as universal markers for identifying exosomes [25]. Its tissue-specific proteins, such as synaptophysin in neuronal exosomes and programmed death-1 (PD-L1) in tumor exosomes, provide targeted recognition capabilities [26]. Exosomes contain various bioactive molecules such as miRNA, proteins, lipids, and metabolites. These molecules reflect the physiological or pathological state of source cells and can be taken up by target cells through endocytosis, membrane fusion, or receptor-mediated mechanisms, thereby regulating gene expression and signal transduction [27, 28].

From a functional perspective, exosomes exhibit physiological and pathological effects. Under physiological conditions, they coordinate immune responses, promote neural synapse plasticity, and facilitate tissue repair [29]. They can also exacerbate the progression of pathological processes. For instance, tumor exosomes carry PD-L1, which inhibits T cell function [30], and the presence of amyloid-beta and phosphorylated tau proteins in circulating exosomes may exacerbate Alzheimer’s disease (AD) [31]. In metabolic diseases, adipocytes secrete exosomes that abnormally activate transforming growth factor beta (TGF-β) signaling in obese individuals through molecules such as miR-23b, thus promoting insulin resistance and nonalcoholic fatty liver disease (NAFLD) disease [32].

Owing to their biological characteristics, exosomes show great potential in the field of medicine. They can be used as diagnostic markers because they are easily acquired, highly stable, and specific to certain diseases. For instance, the presence of Wilms tumor 1 protein in urine exosomes can predict diabetic nephropathy [33], and high miR-21 levels in blood exosomes are associated with poor breast cancer prognosis [34]. In therapeutic applications, exosomal miR-1249-3p derived from natural killer cells can alleviate insulin resistance and inflammation in mice with type 2 diabetes mellitus (T2DM) [35]. Engineering modifications can overcome the natural limitations of exosomes. For instance, corynoxine-B encapsulated in Fe65-engineered neuronal exosomes can mitigate the pathological changes associated with AD [36].

In recent years, research on exercise-derived exosome therapies for multisystem diseases has increased greatly. Exercise-derived exosomes are a distinct subpopulation of EVs released into the circulatory system as a direct or indirect result of acute or chronic physical activity. These vesicles are not just resting-state basal exosomes. They are distinguished by a set of minimum criteria: (1) Temporal correlation with exercise: For acute exercise, samples are collected from 0 to 24 h post-exercise, when circulating exosome abundance peaks and gradually returns to baseline levels during the 90-minute to 72-hour recovery period. For chronic exercise, sampling occurs 24–48 h after the final training session to avoid acute stress interference [37]. (2) Change in abundance or cargo compared to baseline: Compared to resting conditions, changes in the abundance or cargo of exosomes are key evidence for determining whether they are “exercise-derived.” These alterations can include increases or decreases in total exosome abundance; changes in specific exosome subpopulations, such as CD63+/CD81+; and reprogramming of cargo molecules, including miRNAs, proteins, lipids, and myokines [38, 39]. (3) Tissue-of-origin evidence: Traceable to exercise-responsive tissues such as skeletal muscle, vascular endothelium, or cardiomyocytes through tissue-specific markers or lineage tracing [40]. The convergence of these temporal, compositional, and functional attributes defines exosomes as “exercise-derived.” Exercise-derived exosomes promote physical health and play a significant role in mitigating neurological disorders, vascular diseases, metabolic disorders, musculoskeletal conditions and cancers [41–43].

Effects of exercise on exosomes

Exercise can treat various diseases by regulating metabolic reprogramming and energy homeostasis, alleviating the inflammatory microenvironment, promoting tissue repair and regeneration, and maintaining the balance of oxidative stress [44–49]. Exosomes play an indispensable role in these processes. However, exercise is a complex physiological stimulus that can elevate catecholamine levels, induce transient hypoxia, and trigger inflammation simultaneously. Each of these stimuli can independently trigger exosome release. Therefore, a critical consideration is how to attribute exosomal alterations specifically to exercise-derived effects rather than to these overlapping pathways. Several experimental and conceptual approaches can achieve this distinction. (1) Differentiation based on cargo characteristics. Exercise-derived exosomes contain metabolic regulatory molecules, such as miR-342-5p, lactate dehydrogenase A (LDHA), and peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α)-related proteins. These molecules mediate adaptive responses, including muscle repair, anti-apoptosis, and enhanced glycolysis [50, 51]. Exosomes induced by hypoxia typically carry HIF-1α downstream targets, such as vascular endothelial growth factor (VEGF) and miR-210 [52]. Exosomes associated with inflammation are enriched with pro-inflammatory mediators, such as TNF-α, IL-6, and miR-155 [53]. In contrast, exosomes induced by catecholamines do not contain exercise-specific myokines or cytokines and fail to enrich muscle-derived miRNAs, such as miR-1 and miR-133a [54]. (2) Differentiation based on functional phenotype. Exercise-derived exosomes have tissue-protective effects, such as reducing the size of myocardial infarctions and promoting the osteogenic differentiation of bone marrow-derived mesenchymal stem cells (MSCs) [51]. Exosomes originating from co-stimulatory conditions exhibit context-dependent functions. For example, hypoxia-induced exosomes promote angiogenesis, yet they do not enhance systemic metabolic adaptation. Inflammation-induced exosomes exacerbate tissue damage [55]. (3) Experimentally designed validation. First, controlled exercise protocol. Employing low-to-moderate intensity exercise or progressive exercise models distinguishes between aerobic-phase, exercise-specific exosome release and anaerobic-phase, stress-induced exosome release [56]. Second, synchronous stimulation inhibition. Block specific pathways such as catecholamine β-adrenergic receptors or NF-κB pathway to validate that the function and cargo of exercise-induced exosomes remain intact [57]. Third, temporal separation. Compare exosomal profiles during exercise versus during the post-exercise recovery phase [58]. Exercise-derived exosomes persist during recovery with stable adaptive cargo. These criteria allow us to attribute specific exosome populations and their functions to “exercise-derived”.

As a physiological stimulus, exercise can significantly impact the biological characteristics of exosomes, including their release kinetics, spectrum of cell sources, and composition of molecular cargo. These changes ultimately affect their functional output [59]. The effect of exercise varies depending on the pattern, intensity, duration, and training status [60]. Acute and long-term exercise have different effects on exosome-mediated, interorgan communication as shown in Table 1.

Table 1.

Effects of exercise on exosomes

Type of exercise	Exercise protocol	Species	Exosome source	Separation method	Characterization method	Mechanism	Biological effects	References
Acute heavy resistance exercise	Perform back squats using 75% of your one-rep maximum (1RM). There are six groups of ten repetitions each, with a two-minute rest between sets.	Human	Plasma	Size exclusion chromatography	Vesicle Flow Cytometry, Imaging Flow Cytometry	CD63↑, VAMP3↑, SGCA↑	Exosomes play a role in the body’s adaptive response to resistance training, which may be related to muscle metabolism or repair. Female skeletal muscles release more exosomes after resistance training.	[59]
Acute maximal treadmill running	After warming up, increase the exercise speed until it reaches 80% of the participants’ maximum heart rate within two minutes. Then, keep the speed constant and increase the slope by 2% every two minutes until the participants are exhausted.	Human	Plasma	Invitrogen™ exosomes separation kit	Western blot, Enzyme-Linked Immunosorbent Assay	LPO↑, TLR4/NF-κB↑, GSH↓	Obese individuals have a limited ability to repair oxidative damage after exercise; however, the lipid peroxidation reaction is enhanced. Additionally, untrained individuals are more sensitive to the antioxidant response of acute exercise.	[64]
Acute aerobic exercise	The duration is 40 min, with an intensity of 70% of heart rate reserve.	Human	Plasma	ExoQuick precipitation method	Nanoparticle tracking analysis, Small RNA-seq, Western blot	PTEN↓, FOXO↓, IGF-1↑, miR-486-5p↓, miR-206↓	It promotes muscle repair and growth and alleviates age-related muscle decline and metabolic disorders.	[38]
Aerobic combined with resistance exercise	First, complete a 45-minute, double-leg power cycling session at 55% of your maximum oxygen consumption. Immediately after, perform single-leg knee extension resistance training consisting of three sets of eight to twelve repetitions with a load of 55% of 1RM and a two-minute rest period between sets.	Human	Vastus lateralis	Tissue RNA extraction, Protein extraction, Exocrine enrichment	Quantitative real-time PCR, Western blot, Nanoparticle tracking analysis	Dicer↑, exportin-5↑, clathrin↑, Alix↑, Rab27A↑, PGC-1α↑, VEGF↑	Acute exercise increases the production and loading of cargo in skeletal muscle exosomes. This enhances communication between muscles and other tissues and participates in angiogenesis and metabolic regulation.	[50]
Moderate-intensity aerobic exercise	Exercise at 60% of your VO2 max intensity continuously for 30 min or until you reach voluntary exhaustion.	Human	Plasma	Exocrine enrichment	Nanoparticle tracking analysis, Flow cytometry	/	There was a significant reduction in platelet-derived exosomes 24 h after exercise and a significant increase in skeletal muscle-derived exosomes immediately after exercise.	[14]
Moderate-to-vigorous intensity aerobic exercise	Ride at 70% of your VO2max intensity for 20 min.	Human	Plasma	Differential ultracentrifugation	Nanoparticle tracking analysis, Western blot, Label-free quantitative protein omics	TSG101↑, syntenin-1↑, ANXA1↑, MPO↑, S100A8↑, LCP1↑, BASP1↑, HIST1H4A↑, LYZ↑, S100A9↑	Exercise triggers the release of immune proteins via exosomes, thereby enhancing antimicrobial defenses and reducing the risk of chronic inflammation. Healthier individuals and younger groups tend to respond more favorably.	[63]
Uphill running or downhill running	The speed is 20 m/min for 90 min.	Rat	Plasma and skeletal muscle	ExoQuick™ precipitation method	Real-time fluorescence quantitative PCR	miR-1↓, miR-133a↓, miR-499↑	Exosomes play a role in regulating the immune system after acute injury.	[60]
High-intensity aerobic exercise	Treadmill Test: The starting speed is 4–6 km/h and increases by 2 km/h every three minutes until exhaustion. Bicycle test: Starting at 40 W, increase by 40 W every three minutes until exhaustion.	Human	Plasma	Size exclusion chromatography, Immunoaffinity capture method	Western blot, Ultra-sensitive qPCR	CD9↑, CD41b↑	Extracellular DNA may play a role in post-exercise inflammatory signaling and tissue repair.	[135]
Exhaustive treadmill exercise	Begin at 10 m/min and increase by 2 m/min every 10 min until the mouse is exhausted, up to a maximum of 90 min.	Mouse	Plasma and skeletal muscle	Ultracentrifugation, PS affinity bead separation, Separation of CD81 affinity beads	Scanning electron microscope, Nanoparticle tracking analysis (NTA), Western blot, Proteomic profiling, qPCR	ATP2A1↑, β-enolase↑, miR-1↑, miR-206↑, Pax7↓, MyHC↑	Promote myogenesis and muscle repair.	[136]
Acute aerobic exercise	Ride for 60 min at 55% VO2max.	Human	Plasma	Size exclusion chromatography, ultracentrifugation	Nanoparticle tracking analysis, Transmission electron microscope, Western blot, Real-time quantitative PCR	TSG101↑, CD81↑, HSP60↑, Alix ↑	It regulates metabolic adaptation and participates in the pathophysiological process of prediabetes.	[61]
Muscle-damaging exercises, including plyometric jumping and downhill running	Plyometric jumping: Perform 10 sets of 10 repetitions, with a jumping height at 90% of maximum capacity. Rest for one minute between sets. Downhill running: Perform 10 sets of 10 repetitions with a 90% maximum jumping height and a 1-minute rest between sets. Then, perform 5 sets lasting 4 min each at a speed of 10 km/h with the treadmill incline set to -10%. Take a two-minute standing rest between sets.	Human	Plasma	Size exclusion chromatography	Nanoparticle tracking analysis, Transmission electron microscope, Real-time quantitative PCR	miR-31↓, Myf5↑, CK↑	It promotes satellite cell activation and participates in muscle repair.	[137]
Aerobic exercise	Initially, the speed was approximately 2 m/min, increasing to 3.25 m/min on the seventh day and 6.59 m/min on the fourteenth day. Each 20-minute session was performed twice daily for 28 days, with one day of rest per week.	Rat	Cerebrospinal fluid	Centrifugation and filtration	Nanoparticle tracking analysis, Transmission electron microscope, Western blot, Real-time quantitative PCR	/	Exercise promotes neurological recovery and prevents cerebral infarction lesions from growing larger.	[90]
High-intensity interval cycling exercise	Ride at VO₂ max intensity for 60 s, for a total of 10 sets. Rest for 75 s between each set.	Human	Plasma	Size exclusion chromatography	Nanoparticle tracking analysis, Transmission electron microscope, Western blot, Real-time quantitative PCR	miR-1-3p↑, miR-16-5p↑, miR-222-3p↑	MiRNAs in exercise-derived exosomes play a role in the systemic regulation of various organs.	[39]
Ergometer cycling or treadmill running	Ergometer cycling: Begin at 50 W and increase by 50 W every three minutes until exhaustion. Teadmill running: Begin at 6 km/h and increase the speed by 2 km/h every three minutes with a 1.5% slope until exhaustion.	Human	Plasma	Differential centrifugation and filtration	Nanoparticle tracking analysis, Transmission electron microscope, Western blot	/	Exercise-derived exosomes enter the bloodstream and facilitate intercellular signaling, including immune regulation and angiogenesis.	[65]
Low intensity resistance exercise training	Perform low-intensity resistance training three days a week for 12 weeks.	Human	Plasma and skeletal muscle	Size exclusion chromatography and ultracentrifugation	Nanoparticle tracking analysis, Western blot	Alix↑, TSG101↑, miR-23a↑, miR-27a↑, miR-199a↑, MuRF1↓, atrogin-1↓	Low-intensity resistance training promotes muscle synthesis and metabolism by regulating exosome cargo. This ultimately improves muscle quality and function.	[111]
Low-intensity progressive resistance training	Use resistance bands for resistance training. Perform 10–15 repetitions per set, three sets per exercise, three days per week for 12 weeks.	Human	Plasma and skeletal muscle	Size exclusion chromatography, ultracentrifugation, Immunoaffinity capture	Nanoparticle tracking analysis, Western blot, Transmission electron microscope	TSG101↑, Alix↑, miR-23a↑, miR-27a↑, MuRF1↓, atrogin-1↓	Resistance training improves muscle synthesis and metabolism, and can partially reverse age-related changes.	[110]
Acute aerobic exercise or chronic aerobic exercise	Acute aerobic exercise: Cycling at 70% VO2max intensity for 60 min. Chronic aerobic exercise: Cycling at 70% VO2max intensity for 30 min, three times a week for four weeks.	Human	Serum	/	Real-time quantitative PCR	miR-486↓, PTEN↑, PI3K/Akt↑	Exercise enhances muscle insulin sensitivity, promotes glucose uptake, and accelerates energy metabolism during exercise.	[13]
Concurrent aerobic and resistance training	The aerobic exercise intensity was 70% VO₂ max, and the resistance training heart rate reached 85% of the maximum heart rate. Each session included 20 min of aerobic exercise and 20 min of resistance training three times per week for six months.	Human	Serum	Total exosome isolation reagent precipitation method	Nanoparticle tracking analysis, Transmission electron microscope	let-7a-5p↓, miR-23a-3p↓, miR-451a-3p↓	Short-term exercise promotes the release of exosomes through muscle contraction. Long-term exercise, on the other hand, maintains miRNA profiles through adaptive regulation, thereby reducing the risk of age-related diseases.	[138]
Moderate-intensity aerobic exercise	Perform 60% VO2max exercise intensity for 20 min per day for two weeks.	Rat	serum	miRCURY™ Exosome Isolation Kit	Nanosight Nanoparticle Characterization System, Western blot	CD63↑, AChE↓, ROS↓	Exercise improves aging-related exosome dysfunction by regulating oxidative stress and the cholinergic system.	[139]
Sprint Interval Training	Sprint at full speed for 30 s. Then, recover for four minutes and 30 s. Repeat this process four times in week one, gradually increasing to nine times in week six. Continue this routine for six weeks.	Human	Plasma	Size exclusion chromatography	Nanoparticle tracking analysis, Western blot, Transmission electron microscope, Real-time quantitative PCR	TSG101↑, HSP60↑, miR-23a-3p↓, miR-21-5p↑	It promotes myocardial protection, angiogenesis, and metabolic adaptation.	[68]
Football training	Morning: Plyometric training, speed training, and technical training (85% HRmax, 80 min). Afternoon: Aerobic exercise, coordination, and technical training (70% HRmax, 90–100 min); small race (90% HRmax, 90 min); and interval running (80% HRmax, 90 min). For two months.	Human	Serum	Total Exosome Isolation Reagent	Real-time quantitative PCR	c-miR-27b↑, c-miR-29a↑, c-miR-27b↑	Promote skeletal muscle hypertrophy and regeneration.	[67]

Effects of acute exercise on exosomes

Single episodes of acute exercise, especially moderate-to-high intensity exercise, can rapidly alter the number and composition of exosomes in the circulatory system. In the short period after exercise, the concentration of circulating exosomes often transiently increases in healthy individuals. This phenomenon is believed to be an immediate cellular response to exercise-induced stressors, such as mechanical stretching, metabolic byproduct accumulation, oxidative stress, and hormonal fluctuations. These cells release signaling molecules to regulate homeostasis within the internal environment. However, this transient increase typically returns to baseline levels within 24 h post-exercise. For example, healthy individuals who cycled for 60 min in a normal oxygen environment experienced a significant increase in plasma exosomes after exercising for 30 min, but these levels returned to near baseline 60 min after the exercise. However, this effect was weakened or disappeared in individuals with prediabetes or those exercising in a hypoxic environment [61]. Therefore, training environments and diseases can impact exosomes. A combination of acute aerobic and resistance training can accelerate exosome release, thereby promoting vascular regeneration and adaptation [50].

Acute, moderate-intensity exercise can significantly alter the size distribution and cellular origin of circulating exosomes. Immediately following exercise, the total exosome levels decrease while the skeletal muscle-derived exosomes increase. Skeletal muscle-derived exosomes are actively released by skeletal muscles after exercise and participate in muscle repair or metabolic regulation. However, these changes are influenced by gender and body mass index (BMI) [14]. Exosome concentrations are lower in women than in men after exercise. In addition, platelet-derived exosomes decrease more significantly in women, suggesting that exercise may pose a greater cardiovascular risk to them. The exosomes released by obese individuals may be associated with metabolic abnormalities. Exercise reduces these exosomes and improves insulin sensitivity [14]. Following resistance training, the characteristics of EVs differed between men and women. Conkright et al. [59] found that acute resistance training increased the concentration of CD63-positive exosomes in males and females. However, they also found that males produced more small-sized exosomes with a sparser distribution of surface proteins, whereas females released more exosomes from skeletal muscles with lower protein expression per exosome.

The impact of acute exercise on exosomes is not merely a change in quantity. Rather, it lies in the reprogramming of exosomal cargo. D’Souza et al. [39] found that after 4 h of high-intensity interval training (HIIT), 29 miRNAs were upregulated in healthy males. Among these 29 target miRNAs, 11 were upregulated in muscle tissues, eight in plasma, and nine in exosome components. These exercise-upregulated exosomes returned to resting levels after recovery [62]. Chong et al. [63] discovered that 20 min of moderate-intensity cycling can stimulate the release of exosomes and alter their protein composition, thereby enhancing immune function. The protein composition of exosomes is significantly affected by age and aerobic fitness. Individuals with high fitness levels exhibit a robust immune-repair response, whereas aging and low fitness levels may impede this process. Oxidative stress markers in acute exercise-derived exosomes are associated with BMI. In one study, the obese group had a significantly higher concentration of lipid peroxidation products in their exosomes after acute exercise [64]. The normal weight group experienced a greater decrease in glutathione concentration in exosomes after acute exercise, which may be related to their efficient use of exosomes to transport antioxidants [64]. Training status also influences the cargo of exercise-induced exosomes. Nair et al. [38] found that for people who exercise regularly, seven circulating exosomal miRNAs were differentially expressed immediately after acute exercise and eight were present 3 h after exercise. These miRNAs target the insulin-like growth factor 1 (IGF-1) signaling pathway, suggesting that exercise activates muscle growth and repair mechanisms. For people who sit for long periods of time, nine circulating exosomal miRNAs were differentially expressed immediately after acute exercise. These miRNAs mainly target the insulin-like growth factor 1 receptor (IGF-1R) and protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathways, inhibiting anabolic signaling. Four more circulating exosomal miRNAs were differentially expressed 3 h after exercise, and they are related to inflammation and metabolic regulation [38]. Therefore, the circulating exosomal miRNAs released after acute exercise differs between the training and sedentary groups.

The release, clearance, and cargo changes of exosomes differ for each type of acute exercise. In the study of Frühbeis et al. [65], healthy individuals performed incremental exercise protocols involving cycling or running until exhaustion. Exosomes were then isolated from the plasma samples collected before and 90 min after exercise. The results showed that during the aerobic phase of exercise, exosomes were rapidly released into the circulation and were cleared from circulation during the early recovery period after cycling but remained elevated after running [65]. This difference may be due to the continued exosome production after running or the limited clearance rates. Acute uphill and downhill exercise regulate specific miRNAs in muscles and circulation via distinct mechanisms. Yin et al. [60] found that uphill running primarily involves local muscle adaptation, whereas downhill running triggers a strong systemic response. These dynamic changes may play a key role in muscle injury repair, fiber type conversion, and systemic adaptation. In addition, uphill exercise has no significant effect on exosome miRNA. After downhill exercise, most miRNAs increased significantly immediately and 1 h after exercise, with recovery occurring at 48 h [60]. Therefore, different types of acute exercise can cause the release of different cargoes by exosomes, thereby regulating physiological and pathological processes through enhanced immunity, skeletal muscle proliferation and differentiation regulation, and antioxidant effects.

Effects of long-term exercise on exosomes

Different from the immediate effects of acute exercise, long-term and regular exercise induces more persistent molecular reprogramming and functional adaptation of exosomes. Darragh et al. [66] found no significant difference in the total concentration of circulating exosomes in the basal state between individuals who regularly exercise and sedentary control groups. In addition, no significant differences were observed in resting circulating exosome abundance or metabolomic characteristics among long-term endurance training, strength training, and normally active males [66]. Therefore, basal exosome levels may not be affected by exercise training history. However, the composition of the cargo in exosomes has undergone a fundamental optimization. This chronic adaptation is primarily manifested by the selective enrichment of specific functional molecules, especially miRNAs, in exosomes. For instance, elderly males who engaged in long-term endurance training exhibited significant upregulation of three exosomal miRNAs (miR-486-5p, miR-215-5p, and miR-941) and significant downregulation of one miRNA (miR-151b) compared with the sedentary group [38]. The high expression of circulating exosomal miRNAs induced by long-term training may be associated with anti-aging and anabolic processes [38]. In addition, 2 months of soccer training significantly increased the expression of miR-27b and miR-29a in circulating exosomes (Domańska-Senderowska et al., 2017). This increased expression regulates collagen metabolism and mitochondrial function, ultimately contributing to improved cardiorespiratory endurance [67].

Different training environments have varying effects on circulating exosomes. For instance, 6 weeks of sprint interval training in a hypoxic environment significantly regulated miRNAs in exosomes, such as downregulating miR-23a-3p and upregulating miR-21-5p, whereas training at sea level primarily increased the number of exosomes and markers [68]. In addition, long-term exercise can protect various organs by regulating the exosomal cargo. For instance, long-term exercise exerted a cardioprotective effect by upregulating miR-342-5p in plasma exosomes and improved neurological outcomes by prompting skeletal muscle cells to secrete exosomes rich in miR-484 [40, 51]. Therefore, long-term exercise has a positive, systemic remodeling effect on exosomes. The molecules responsible for this effect and their anti-inflammatory, antioxidant, and anti-apoptotic properties are crucial for preventing and improving various chronic diseases and promoting physical health.

Exercise-derived Exosomal cargo

Exosomes serve as key carriers for intercellular communication and mediate the effects of exercise by carrying bioactive regulatory factors. Exercise-derived exosomes are the molecular basis for exercise’s remote effects on the entire organ system. Their cargo includes miRNAs, proteins, lipids, and myokines as shown in Fig. 1 [69]. Based on the roles of exosome cargo in the nervous, vascular, metabolic, and musculoskeletal systems, we have categorized these cargoes into five core signaling axes: the antioxidant axis, the anti-inflammatory axis, the pro-angiogenic axis, the anti-apoptotic axis, and the metabolic reprogramming axis. While some axes are present in multiple organ systems, others are more specific to certain organs and contexts [70].

Fig. 1.

Exercise-derived exosome cargo and its biological effects. Exercises such as running, cycling, swimming, and resistance training stimulate the brain, heart, liver, muscles, blood vessels, adipose tissue, and bones to secrete exosomes that are rich in miRNAs, proteins, and cytokines. These exosomes then target articular chondrocytes, vascular endothelial cell, hepatocytes, cardiac fibroblasts, macrophages, endothelial cells, pericytes, cardiomyocytes, intestinal epithelial cells, neurons, myoblasts, fat cells, satellite cells, muscle fiber cells, skeletal muscle cells, liver cells, nucleus pulposus cells and osteoblasts. They regulate multiple signaling pathways, including NF-κB, MAPK, VEGF, TGF-β, Smad3, PI3K, AKT, mTOR and Nrf2. Ultimately, these effects are beneficial for the brain, heart, skeletal muscles, adipose tissue, and liver

The antioxidant axis is one of the most widely applied signaling pathways. Exosomal cargo, including miR-484, miR-342-5p, superoxide dismutase 3 (SOD3), and unsaturated lipids, can mitigate ROS accumulation and lipid peroxidation. For example, skeletal muscle exosomes rich in miR-484 reduce mitochondrial ROS levels during cerebral ischemia [40]. Plasma exosomes enriched with miR-342-5p enhance myocardial antioxidant capacity [51]. SOD3-rich exosomes derived from swimming exercise improve redox imbalance in osteoarthritic cartilage [71]. The antioxidant axis is present in the nervous, musculoskeletal, and metabolic systems, reflecting the tissue-protective effects of exercise through the alleviation of oxidative stress.

The anti-inflammatory axis is mediated by exosomal cargo, including miR-146a-5p, miR-532-5p, and carnosine [72]. These molecules suppress NF-κB-related inflammatory cascades persistently in vascular, neural, and musculoskeletal tissues [73]. However, this regulation exhibits tissue specificity. For example, calmodulin-containing exosomes primarily modulate macrophage polarization in degenerative intervertebral discs [74], and miR-532-5p, which is derived from brain tissue, primarily stabilizes the blood-brain barrier (BBB) by reducing ephrin type-A receptor 4 (EPHA4) signaling in AD [75].

An angiogenic axis exists in ischemic brain tissue, the heart, and skeletal muscle. Exosomes that are rich in miR-206, miR-126, and miR-125a-5p enhance AKT/eNOS activity, stimulate VEGF signaling, and restore perfusion in hypoxic tissues [76, 77]. However, angiogenic outcomes vary greatly across organ environments. For example, the myocardium and skeletal muscle undergo vascular remodeling easily, while the central nervous system maintains the BBB and permits only highly regulated angiogenic responses [78].

The anti-apoptotic axis includes exosomal cargo, such as miR-342-5p, miR-124, HSP70, and carnosine. This cargo stabilizes mitochondrial function and suppresses caspase-mediated cell death [65]. This axis exerts protective effects in neurons, cardiomyocytes, chondrocytes, and myocytes. For example, extracellular miR-342-5p and miR-124 protect cells from programmed death in cases of myocardial infarction and cerebral ischemia by suppressing pro-apoptotic proteins [51, 79].

Finally, the metabolic reprogramming axis integrates multiple miRNAs and proteins that regulate glucose metabolism, mitochondrial biogenesis, and lipid homeostasis [80]. Exosomes enriched with miR-133b/206 suppress hepatic gluconeogenesis [81], and miR-324 reduces lipid accumulation in NAFLD by inhibiting Rho-associated coiled-coil-containing protein kinase 1 (ROCK1) [82]. Exercise-induced, miR-215-5p-rich exosomes enhance neuronal metabolism by inhibiting IDH1 and BCL2L11 [83], while exosomes that are rich in eNAMPT can enhance systemic energy homeostasis by increasing NAD⁺ levels [84]. Although metabolic improvements extend throughout the body, tissues respond through unique pathways shaped by local metabolic priorities. For example, the liver regulates FOXO1-dependent gluconeogenesis, and the brain modulates mitochondrial redox enzymes [85].

Thus, exercise-derived exosomes coordinate systemic adaptation through multiple molecular signaling pathways while maintaining target organ specificity via differentiated cargo composition, receptor accessibility, tissue metabolic demands, and microenvironmental barriers. Although research on exercise-derived exosomes is in its early stages, integrating multi-omics data with targeted tissue tracing technologies shows promise in elucidating the mechanisms and clinical value of these exosomal components in exercise medicine and disease treatment.

Potential applications of exercise-derived exosomes in disease treatment

Exercise-derived exosomes have therapeutic potential because they can activate the aforementioned core signaling pathways in a coordinated, multi-targeted way. As illustrated in Tables 2, 3, 4 and 5; Figs. 2, 3, 4 and 5, the relative importance of these signaling pathways and their combination patterns varies greatly depending on the pathophysiological environment of the target organ. In certain scenarios, a single signaling axis may dominate; in others, multiple axes contribute synergistically to therapeutic effects [62]. Subsequent subsections will detail these applications, emphasizing their common mechanisms and organ-specific adaptations in neurological, vascular, metabolic, and musculoskeletal disorders.

Table 2.

The role of exercise-derived exosomes in neurological diseases

Target organ /site	Species	Exercise protocol	Exosome source	Exosome cargo	Recipient cell	Disease	Mechanism	Functions	References
Blood–Brain Barrier	Mouse	Perform treadmill exercise five days a week for 16 weeks.	Brain tissue	miR-532-5p	Pericytes, endothelial cells	AD	miR-532-5p↑, EPHA4↓, ZO-1↑, claudin-5↑, PDGFRβ↑, NG2↑, LRP1↑	It improves blood-brain barrier function and cognitive ability.	[75]
Brain	Rat	Perform 20 min of treadmill exercise daily, seven days a week, for four weeks.	Skeletal muscle	miR-484	Neurons	Ischemic stroke	miR-484↑, ACSL4↓, LPO↓, GPX4↑, SLC7A11↑, ROS↓	It inhibits neuronal ferroptosis, alleviates cerebral ischemia damage, and improves neurological function.	[40]
Brain	Rat	Perform electric treadmill training at a speed of 12 rpm for 30 min, six days a week, for four weeks.	Blood	miR-124	Neurons	Cerebral ischemia-reperfusion injury	miR-124↑, STAT3↓, BCL-2↑, BAX↓	It inhibits neuronal apoptosis, improves neurological function, and alleviates cerebral ischemia-reperfusion injury.	[79]
Brain	Human	Endurance training	Circulating blood	miR-215-5p	Neurons	AD	miR-215-5p↑, IDH1↓, BCL2L11↓, SIRT1↓, RIP1/RIP3↓	It inhibits neuronal necroptosis and prevents AD.	[83]
Brain	Rat	Perform treadmill exercise at a speed of 12 m/min for 30 min, five times a week for four weeks.	Brain tissue	/	Microglia, neurons	Ischemic stroke	/	It inhibits microglial activation, regulates synaptic plasticity, and exerts neuroprotective effects.	[91]
Brain	Rat	Perform treadmill exercise at a speed of 12 m/min for 30 min, five times a week for four weeks.	Brain tissue	/	Neurons	Ischemic stroke	Syn↑, PSD-95↑	It reduces infarct volume, promotes synapse growth, and improves neural function.	[140]
Brain	Human	Perform Tai Chi training for 60 min each time, five times a week, for 12 weeks.	Serum	LRP1	Neurons	Amnestic mild cognitive impairment	LRP1↑, Aβ↓, tau↓	It enhances hippocampal plasticity, thereby improving memory and cognitive function.	[92]
Brain	Rat	Perform 30 min of treadmill exercise at 12 m/min daily for 14 days.	Bone marrow mesenchymal stem cells	/	Neurons	Ischemic stroke	JNK1/c-Jun↑, p-JNK1↑, p-c-Jun↑, Bcl-2↑, Bax↓, caspase-3↓, Synaptophysin↑, PSD-95↑, GAP-43↑, MAP-2↑, NF-200↑	It inhibits neuronal apoptosis, promotes axon regeneration, and improves neurological function following cerebral ischemia-reperfusion injury.	[89]
Brain	Rat	Perform treadmill exercise for 28 days, five days a week.	Cerebrospinal Fluid	rno-miR-138-5p, rno-miR-370-3p, rno-miR-665, etc.	Neurons, endothelial cells	Ischemic stroke	rno-miR-138-5p↑, rno-miR-370-3p↑, rno-miR-665↓, rno-miR-338-3p↓, cleaved caspase-3↓, Bcl-2↑, NF-κB↓, MAPK↑	It inhibits neuronal apoptosis, promotes angiogenesis and nerve regeneration, and improves nerve function.	[90]
Brain	Mouse	Perform treadmill exercise at a speed of 10 m/min for 30 min, five times a week for four weeks.	Circulating endothelial progenitor cells	miR-126	Neurons, endothelial cells	Ischemic stroke	miR-126↑, cleaved caspase-3↓, Bcl-2↑, PI3K/Akt↑, VEGF↑, BDNF↑, TrkB↑	It promotes angiogenesis and neurogenesis, improves sensory-motor function, and enhances neural plasticity.	[76]
Skeletal muscle	Human	Perform a 45-minute cycle at 70% of VO2max once a week.	Plasma	miR-877, miR-4433b, miR-486	Satellite cells, myoblasts	Cerebral palsy	Pax7↓, COL1A1↓, COL3A1↓, MYH1↓	The effects of acute exercise on EV size and miRNA profiles are limited.	[93]
Neurons	Mouse	Perform 60 min of moderate-intensity treadmill exercise, five days per week, for eight weeks.	Bone marrow-derived endothelial progenitor cells	miR-27a	Neuronal N2a cells	Hypertension-associated neuronal ischemic injury	miR-27a↑, Nox4↓, ROS↓, ECE1↓, p-AKT↑, eNOS↑	It promotes cell survival, reduces oxidative stress, and protects mitochondrial function.	[88]

Table 3.

The role of exercise-derived exosomes in vascular diseases

Target organ	Species	Exercise protocol	Exosome source	Exosome cargo	Recipient cell	Disease	Mechanism	Functions	References
Heart	Mouse	Perform 60 min of moderate-intensity aerobic exercise three times per week for two weeks.	Endothelial progenitor cells	miR-126	Cardiac fibroblasts	Cardiac fibrosis	miR-126↑, TGF-β↓, Smad3↓, α-SMA↓, Col1↓, Col3↓	It inhibits myocardial cell apoptosis, alleviates inflammatory responses, and improves cardiac function.	[94]
Heart	Rat	Perform swimming training for 90 min each time, twice a day, for four weeks.	Vascular endothelial cells	miR-342-5p	Cardiomyocytes, endothelial cells	Myocardial Ischemia/Reperfusion	miR-342-5p↑, Caspase 9↓, Jnk2↓, Ppm1f↓, p-Akt↑	It prevents myocardial cell apoptosis and improves cardiac function.	[51]
Hindlimb	Rat	Perform moderate-intensity, intermittent treadmill training at a speed of 30 m/min for 60 min, five days a week, for four weeks.	Skeletal muscle	miR-125a-5p	Endothelial cells	Hindlimb Ischemia	miR-125a-5p↑, ECE1↓, ET-1↓, eNOS↑, NO↑, p-AKT↑	It promotes vascular regeneration and improves blood flow and perfusion.	[78]
Heart	Mouse, human	Mouse: Perform swimming training for 90 min each time, twice a day, five days a week, for four weeks. Human: At least two years of jogging training.	Plasma	CRNDE	Cardiomyocytes	Myocardial infarction	CRNDE↑, Nrf2↑, HO-1↑, Keap1↓, ROS↓, cleaved-caspase-3↓, Bax↓, Bcl-2↑	It inhibits cell apoptosis, reduces oxidative stress, and improves cardiac function.	[73]
Skeletal Muscle	Mouse	Mice were fed drinking water containing doxycycline (0.5 mg/mL) for two weeks to simulate the effects of resistance training.	Serum	miR206-3p	Endothelial cells	Lower extremity arterial disease	miR206-3p↑, PI3K/Akt↑, IGF-1↑, VEGF↑, SDF1↑	It promotes endothelial cell angiogenesis and blood flow recovery in ischemic limbs.	[95]
Blood vessel endothelium	Mouse	Perform 60 min of low-intensity aerobic exercise at a speed of 5 m/min or 60 min of moderate-intensity aerobic exercise at a speed of 10 m/min, five days per week for four weeks.	Plasma	miR-126	Endothelial Cells	Vascular injury	miR-126↑, SPRED1↓, VEGF↑	It inhibits apoptosis, promotes angiogenesis, and alleviates endothelial damage.	[141]
Blood vessel endothelium	Mouse	Electrical pulse stimulation applied to C2C12 cells simulates muscle contraction.	Skeletal muscle	miR-130a	Endothelial Cells	Capillary rarefaction	miR-130a↑, Gax↓, p-NF-κB p65↑, ROS↑	It improves endothelial cell function and promotes angiogenesis.	[77]

Table 4.

The role of exercise-derived exosomes in metabolic diseases

Target organ	Species	Exercise protocol	Exosome source	Exosome cargo	Recipient cell	Disease	Mechanism	Functions	References
Hippocampus	Mouse	Perform 60 min of treadmill exercise three times a week for 12 weeks.	Serum	miR-382-3p, MALAT1, BDNF	Hippocampal neurons	T2DM with cognitive dysfunction	miR-382-3p↑, MALAT1↑, BDNF↑, PI3K/AKT↑, Ras/MAPK↑	It inhibits hippocampal neuronal apoptosis, promotes neuronal proliferation, and improves cognitive dysfunction in mice with T2DM.	[56]
Blood vessel endothelium	Mouse, human	Mouse: Two-week voluntary shift running program. Human: Forty-five minutes of moderate-intensity treadmill exercise (50% VO₂ max).	Plasma	SOD3, ATP7A	Vascular endothelial cell	T2DM	SOD3↑, ATP7A↑, VEGFR2↑	It promotes endothelial cell migration and capillary formation, thereby improving vascular regeneration capacity.	[71]
Hippocampus	Mouse	Perform forced treadmill training at a speed of 7 m/min for 30 min, five days a week, for four weeks.	Gastrocnemius muscle	miR-200a-3p	Hippocampal neurons	T2DM	miR-200a-3p↑, Keap1↓, Nrf2↑, Hsp90aa1↑, Mct2↑	It improves the spatial memory function of mice with T2DM.	[102]
White adipose tissue, skeletal muscle	Mouse	Perform one or two hours of non-weight-bearing swimming training daily, five days a week, for ten weeks.	Serum	miR-27a	Fat cells, skeletal muscle cells	Obesity	miR-27a↓, PPAR-γ↑, IRS-1↑, Akt↑, GLUT-4↑, UCP1↑, PRDM16↑, PGC-1α↑	It inhibits weight gain and fat accumulation caused by a HFD and reuces fat cell volume.	[99]
Heart	Human	Perform 60–70% of your heart rate reserve at a 60-minute session, three times per week for 12 weeks.	Plasma	/	Cardiomyocytes, vascular endothelial cells, adipocytes, hepatocytes, etc.	Obesity	/	It inhibits inflammation, regulates metabolism, and protects the heart.	[98]
Liver	Mouse	Perform 10 min of high-intensity interval training five days a week for five weeks.	Skeletal muscle, plasma	miR-133a, miR-133b	Hepatocytes	Diet-induced insulin resistance	miR-133a↑, miR-133b↑, FoxO1↓, G6PC↓, PCK1↓	It regulates liver lipid metabolism and promotes fatty acid oxidation.	[85]
Skeletal muscle	Human	Perform an acute cycling exercise at 55% of your VO2max for 60 min.	Plasma, Skeletal muscle	TSG101, ALIX, CD9, CD81, HSP60, TSG101, ALIX, CD9, CD81	Vascular endothelial cells, hepatocytes, etc.	Prediabetes	TSG101↑, CD81↑, HSP60↑, ALIX↑, CD9↑	It improves glucose tolerance and insulin sensitivity.	[61]
Liver, skeletal muscle, adipose tissue	Human	Perform 35 min of high-intensity interval training three times a week for 12 weeks.	Plasma	PRDX2, GPX3, PTGS1, MFN2, OPA1	Liver cells, skeletal muscle cells, fat cells	T2DM, insulin resistance	PRDX2↑, GPX3↑, PTGS1↑, MFN2↑, OPA1↑	It improves insulin sensitivity, reduces oxidative stress, and maintains mitochondrial respiratory function.	[101]
Visceral white adipose tissue, intestines, liver	Mouse	Perform 30 min of treadmill exercise, five days a week, for six weeks.	Plasma	CD11c, CD31, CD146, TIE2, PRDX2, GPX3, PTGS1	Adipocytes, macrophages, intestinal epithelial cells	Insulin resistance, obstructive sleep apnea-related metabolic disorders	pAKT/AKT↓, Firmicutes/Bacteroidetes↓, CD11c↓	It restores the balance of intestinal flora, reduces the pro-inflammatory components of exosomes, and reverses insulin resistance.	[142]
Adipose tissue	Rat	Perform treadmill exercises five days a week for nine weeks.	Stem cells derived from rat inguinal fat tissue	miR-323–5p, miR-433–3p, miR-874–3p	3T3-L1 preadipocytes	Obesity	miR-323–5p↑, DUSP3↓, p-ERK↑, p-PPARγ↑, C/EBPα↓	It reduces the number of fat cells in adipose tissue and slows the progression of obesity.	[100]
Liver	Human, mouse	Human: Perform 45–60 min of aerobic Ba Duan Jin training at 60–75% VO2max daily for 3–5 days per week for 12 weeks. Mouse: Perform 50 min of treadmill exercise daily, five days a week, for six weeks.	Plasma	miR-324-5p	Hepatocyte	Metabolic-related fatty liver disease	miR-324-5p↑, ROCK1↓, AKT/GSK3↑, PEPCK↓, ACC↓	It regulates glucose and lipid metabolism in the liver and improves insulin sensitivity.	[82]
Adipose tissue, Skeletal muscle	Human	Recreational physical activity lasting 30 years.	Serum	let-7a-5p, let-7 g-5p, and 12 other miRNAs	Adipocytes, skeletal muscle cells	/	let-7a-5p↓, let-7 g-5p↓, NDRG4↑, FAM13A↑, ST3GAL6↑, AFF1↑	It inhibits inflammation, maintains metabolic homeostasis, and protects the cardiovascular system.	[103]
Skeletal muscle, adipose tissue, hypothalamus, and other tissues related to systemic metabolism.	Human	Ride at 70% VO2max for 20 min.	Plasma	eNAMPT	C2C12 myoblasts	Age-related metabolic disorders	eNAMPT↑, NAD+↑, SIRT1↑	It regulates metabolism and slows down the aging process.	[84]

Table 5.

The role of exercise-derived exosomes in musculoskeletal diseases

Target organ	Species	Exercise protocol	Exosome source	Exosome cargo	Recipient cell	Disease	Mechanism	Functions	References
Bone tissue	Mouse	Perform a one-hour treadmill exercise at a speed of 10 m/min daily for 10 weeks.	Myotubes derived from skeletal muscle	FNDC5, irisin	Osteoblasts	Postmenopausal osteoporosis	PGC1α↑, FNDC5↑, irisin↑, AMPKα↑, Nrf2↑, HMOX1↑, Fpn↑, Cyclin A2↑, CDK2↑	It promotes osteoblast proliferation, inhibits osteoblast ferroptosis, and improves bone mass and structure.	[104]
Intervertebral disc	Human	Simulate movement through mechanical stretching. Parameters: Elongation of 15%, frequency of 0.5 Hz, stimulation for several hours per day for two days.	iMyotubes	Irisin	Nucleus pulposus cells, macrophages	Intervertebral disc degeneration	Irisin↑, AMPK↑, SOX9 ↑, ACAN↑, MMP13↓, ADAMTS5↓, NF-κB↓, TNF-α↓, IL-1β↓, iNOS↓, CD86↓, Arg1↑, CD206↑	It reduces inflammatory infiltration and improves intervertebral disc function.	[109]
Skeletal muscle	Mouse	Perform treadmill exercise at a speed of 10 m/min and an incline of 10° for one hour daily over a period of ten weeks.	White adipose tissue	miR-146a-5p	Skeletal muscle cells	Skeletal muscle atrophy	miR-146a-5p↑, IGF-1R↓, PI3K/AKT/mTOR↓, FoxO3↓, Fbx32↓, MuRF↓, Cyclin A2↓, MyoD↑, MyoG↑, IL-6↓, TNF-α↓	It inhibits muscle atrophy, promotes muscle differentiation, and enhances muscle function.	[97]
Bone tissue	Rat	Neural muscle electrical stimulation was administered using continuous square waves with a pulse width of 10 ms, a stimulation frequency of 2 Hz, and a current intensity of 5–10 mA. Treatment was administered for 30 min per day, six days per week, for eight weeks.	Skeletal muscle	Multiple miRNAs, including miR-17-5p, miR-19b-3p, and miR-27a-3p	Osteoblast precursor cells	Denervation-induced osteoporosis	miR-30a-5p↑, miR-27a-3p↑, Smad7↓, Smurf2↓, ACVR1↓, HDAC4↓, Runx2↑, Osterix↑, ALP↑, COL-1↑, OCN↑, Cathepsin K↓	It promotes osteogenic differentiation, improves bone microstructure, and regulates the balance of bone metabolism.	[105]
Skeletal muscle	Mouse	Mechanical strain was used to simulate exercise. Low strain, long duration: 0%-15% strain at 0.5 Hz for 24 h. High strain, short duration: 12%-22% strain at 1 Hz with 50 min of rest after each 10-minute loading cycle for a total of 24 h.	Muscle progenitor cells	Multiple miRNAs such as miR-146b-5p and miR-183-5p	C2C12 myoblasts	Sarcopenia	miR-146b-5p↑, miR-183-5p↑, BMP↑, FoxO3↓, Smad7↓	It promotes cell proliferation and enhances myogenic differentiation and muscle regeneration capacity.	[106]
Skeletal muscle	Mouse	Use neuromuscular electrical stimulation. Use a symmetrical waveform with a 9 mA amplitude, a 150 µs pulse duration, and a 50 Hz frequency. Stimulate for 5 s and rest for 10 s. Do two sets of 10 repetitions per hind limb with 5 min of rest between sets for a total of 5 sets.	Serum	/	Skeletal muscle cells, muscle satellite cells	Aging-related impaired skeletal muscle regeneration	/	It promotes muscle function recovery.	[107]
Knee articular cartilage	Mouse	Perform 30 min of non-weight-bearing swimming training five days a week for eight weeks.	Serum	miR-146a-5p, SOD3	Articular chondrocytes	Osteoarthritis	SOD3↑, ROS↓, CCAAT/EBPβ↓, APOE↓, MMP13↓, ADAMTS5↓, COL2↑, ACAN↑	It protects the structure of the cartilage and promotes the recovery of knee joint function.	[108]
Skeletal muscle	Human	Plyometric Jumping: Perform 10 repetitions at 90% of your maximum jump height. Rest for one minute between sets. Complete 10 sets total. Downhill running: 10 km/h speed, four minutes per set, two minutes rest between sets, five sets total.	Plasma	Multiple miRNAs such as miR-1 and miR-31	Satellite cells, muscle fiber cells	Exercise-induced muscle damage	miR-31↓, Myf5↑	It regulates satellite cell activity and maintains muscle homeostasis.	[137]
Skeletal muscle	Human	Perform resistance training for 30 weeks.	Serum	miR-362-3p, miR-424-5p, and other miRNAs	Skeletal muscle cells	Aging-related sarcopenia	miR-362-3p↑, PTPN1↓, CD82↓, CLR↓, Th17↓, Hippo↑, Wnt↑	It relieves sarcopenia related to aging.	[143]
Skeletal muscle	Human	Perform 12 weeks of home-based resistance training, three times a week.	Plasma	miR-23a, miR-27a, miR-146a, miR-92a	Skeletal muscle cells	Aging-related sarcopenia	miR-23a↑, miR-27a↑, miR-146a↑, miR-92a↑, TSG101↑, NF-κB↓, MAPK↓, Atrogin-1↓	It relieves sarcopenia related to aging.	[110]

Fig. 2.

Mechanisms of exercise-derived exosomes in neurological diseases. Exercise triggers the release of exosomes from the brain, blood, muscles, and bone marrow mesenchymal stem cells. These exosomes target neurons, endothelial cells, pericytes, satellite cells, and myoblasts, alleviating neurological diseases such as AD, ischemic stroke, cerebral ischemia-reperfusion injury, amnestic mild cognitive impairment, and neuronal ischemic injury. This figure illustrates the core mechanisms of neurological diseases. Table 2 provides detailed experimental evidence and supporting references

Fig. 3.

Mechanism of exercise-derived exosomes in vascular diseases. Exercise triggers the release of exosomes from endothelial progenitor cells, vascular endothelial cells, skeletal muscle, and blood. These exosomes target cardiac fibroblasts, endothelial cells, and cardiomyocytes, alleviating vascular diseases such as cardiac fibrosis, myocardial ischemia/reperfusion, hindlimb ischemia, myocardial infarction, lower extremity arterial disease, vascular injury, and capillary rarefaction. This figure illustrates the core mechanisms of vascular diseases. Table 3 provides detailed experimental evidence and supporting references

Fig. 4.

Mechanism of exercise-derived exosomes in metabolic diseases. Exercise triggers the release of exosomes from blood, muscle, and stem cells. These exosomes target hippocampal neurons, vascular endothelial cell, hepatocytes, fat cells, skeletal muscle cells, 3T3-L1 preadipocytes, macrophages, and intestinal epithelial cells, alleviating metabolic diseases such as T2DM, obesity, insulin resistance, and metabolic-related fatty liver disease. This figure illustrates the core mechanisms of metabolic diseases. Table 4 provides detailed experimental evidence and supporting references

Fig. 5.

Mechanism of exercise-derived exosomes in musculoskeletal diseases. Exercise triggers the release of exosomes from skeletal muscle, blood, white adipose tissue, and iMyotubes. These exosomes target osteoblasts, osteoblast precursor cells, C2C12 myoblasts, skeletal muscle cells, nucleus pulposus cells, macrophages, satellite cells, muscle fiber cells, and articular chondrocytes, alleviating musculoskeletal diseases such as osteoporosis, sarcopenia, intervertebral disc degeneration, skeletal muscle atrophy, muscle damage, and osteoarthritis. This figure illustrates the core mechanisms of musculoskeletal diseases. Table 5 provides detailed experimental evidence and supporting references

Neurological diseases

Neurological diseases are difficult to treat because of their complex pathological microenvironment, which includes blood–brain barrier obstruction, inflammatory cascade reactions, and irreversible neuronal damage [86]. Owing to their natural ability to penetrate the blood–brain barrier, low immunogenicity, and multitarget regulatory properties, exercise-derived exosomes play an active role in nerve repair [87]. However, their efficacy is significantly influenced by delivery timing, disease type, and exercise pattern. In addition, the mechanisms are complex and diverse.

Exercise-derived exosomes offer new therapeutic targets for neurological disorders such as AD, ischemic stroke, mild cognitive impairment, and cerebral palsy. They regulate neurovascular homeostasis, inhibit neuronal death cascades, and promote synaptic network remodeling (Tables 2 and Fig. 2). Liang et al. [75] demonstrated that exercise-induced, miR-532-5p-rich brain-derived exosomes repair the BBB and alleviate AD through gain- and loss-of-function approaches. In vitro experiments revealed that miR-532-5p mimics increased the expression of BBB-related markers and enhanced cell viability in brain endothelial cells and pericytes. Conversely, miR-532-5p inhibitors reduced the protective effects of exercise-derived exosomes. In vivo experiments showed that adeno-associated virus (AAV)-mediated overexpression of miR-532-5p recapitulated the BBB stabilization phenotype observed after prolonged exercise. Thus, miR-532-5p serves as a key mediator of the beneficial effects of exercise [75]. Subsequently, dual luciferase reporter assays confirmed the direct binding of miR-532-5p to the 3’ untranslated region (3’-UTR) of EPHA4. EPHA4 protein levels were significantly downregulated in cells treated with exercise-derived exosomes and in mice overexpressing miR-532-5p. Furthermore, long-term exercise reduced EPHA4 expression in AD mice, and miR-532-5p inhibition reversed this downregulation. These results confirm EPHA4 as a direct downstream target of miR-532-5p [75]. Additionally, in vitro experiments using PKH26-labeled exosomes demonstrated distinct intracellular localization within endothelial cells and pericytes. Combined with CISH technology and vascular cell markers, these studies confirmed enhanced miR-532-5p signaling in BBB-associated cells following exercise. Therefore, exercise-derived, miR-532-5p-rich exosomes exert protective effects on the BBB and lower Aβ levels by directly inhibiting EPHA4 [75].

In cerebral ischemic injury, Huang et al. [40] measured the expression levels of pre-mir-484 and mature mir-484 in multiple tissues, including the heart, liver, spleen, lungs, kidneys, skeletal muscles, and blood, following pre-adaptation exercise. The results showed no significant changes in other tissues, but pre-miR-484 and miR-484 expression were significantly increased in skeletal muscle. This tissue-specific upregulation indicates an exercise-induced enrichment of miR-484 in skeletal muscle exosomes [40]. Further analysis revealed that locally inhibiting the expression of miR-484 via the bilateral injection of an adeno-associated virus carrying a short hairpin RNA targeting miR-484 (AAV-sh-mir-484) into the gastrocnemius muscles significantly reduced the levels of exosomal miR-484 in the CSF post-exercise. These results demonstrate that the enriched skeletal muscle exosomal miR-484 mediates neuroprotective effects [40]. Additionally, the study found that fluorescently labeled Mu-Exos were internalized by neurons in vitro and in vivo, confirming that exosomes carry miR-484 to target cells. Prior to exercise, tissue-specific AAV vectors were used to silence miR-484 in skeletal muscle. This intervention eliminated exercise-induced upregulation of miR-484 in CSF exosomes and suppressed its protective effects simultaneously. Therefore, exercise induces the secretion of skeletal muscle-derived, miR-484-enriched exosomes that target ACSL4-mediated ferroptosis, ultimately mitigating cerebral ischemic injury [40]. Chen et al. [88] isolated EPCs from bone marrow and identified them using immunofluorescence techniques. Their results showed that endothelial progenitor cell (EPC)-derived exosomes (EPC-EXs) expressed both universal exosome markers (CD63 and Tsg101) and EPC-specific markers (CD34 and VEGFR2). Further studies showed exercise reduces ROS production by upregulating miRNA-27a in EPC-EXs. This enhances mitochondrial function and decreases the expression of the pro-apoptotic proteins cytochrome c and NADPH oxidase 4 (Nox4). Ultimately, this alleviates hypertension-associated neuronal ischemic injury [88].

In ischemia–reperfusion injury, the increased miRNA-124 expression in circulating exosomes derived from exercise inhibits the signal transducer and activator of transcription 3 signaling pathway, thus reducing neuronal apoptosis and increasing the expression of B-cell lymphoma 2 [79]. Jiang et al. [89] discovered that a combined intervention of exercise and MSC-derived exosomes increases the expression levels of synaptic proteins (PSD-95 and synaptophysin) and axonal markers (NF-200, MAP-2, and GAP-43) in the perilesional cortex by regulating the JNK1/c-Jun pathway. This suggests that neuronal and synaptic structures are essential functional targets. Thus, Exercise combined with MSC-derived exosomes can treat ischemic cerebrovascular disease [89]. During the ischemic stroke recovery phase, Huang et al. [90] identified miRNA-132-3p as a key cargo in exercise-induced exosomes in CSF. Bioinformatics analysis and in vitro experiments revealed that miR-132-3p inhibits RhoA. Overexpression of miR-132-3p in primary cortical neurons increased synaptic spine density and promoted PSD-95 aggregation. However, this effect was reversed with RhoA overexpression. These results confirm that exercise-derived exosomes enhance synaptic plasticity and alleviate ischemic stroke via the miR-132-3p/RhoA/actin cytoskeletal pathway [90].

Exercise-derived exosomes play a key role in preventing and treating mild cognitive impairment. Low-intensity resistance exercise increases miRNA-215-5p expression in skeletal muscle exosomes, thereby inhibiting isocitrate dehydrogenase 1 (IDH1) and BCL2-like 11. This phenomenon reduces the risk of neuronal cell death and forms a negative feedback regulatory network with the transcription factors CCAAT/enhancer-binding protein beta and GATA-binding protein 6, ultimately maintaining brain homeostasis [91, 92]. In patients with cerebral palsy, the exosomes derived from aerobic exercise improved the neurodevelopmental microenvironment by regulating the expression of multiple miRNAs [93]. The myokines they carry, such as IL-6, can also promote neural regeneration by activating the phosphatidylinositol 3-kinase (PI3K)/AKT signaling pathway.

Exercise-derived exosomes exhibit synergistic, multitarget therapeutic effects on neurological disorders. Wang et al. [76] isolated circulating EPCs from mouse peripheral blood and specifically captured EPC-EXs using anti-CD34- and anti-vascular endothelial growth factor receptor 2 (VEGFR2)-conjugated beads to separate them from non-EPC-EXs. In vitro studies revealed that exercise-derived EPC-EXs enriched with miRNA-126 greatly enhanced endothelial cell migration and tubule formation, mitigated hypoxia-induced apoptosis, and promoted axonal growth. In vivo, Wang et al. [76] used a mouse middle cerebral artery occlusion model to simulate cerebral ischemia. Exercise-derived EPC-EXs greatly increased microvascular density around the infarct zone and promoted angiogenesis. This improved cerebral perfusion and neurological function. Bioinformatics analysis revealed that the miRNA profile of exosomes in CSF changes after exercise, such as miR-132-3p and miR-124 being upregulated [90]. These changes regulate the pathways related to neuroinflammation and synaptic remodeling [90]. Further research is needed to clarify the molecular mechanisms of exosome-targeted delivery to promote its clinical application in the treatment of neurological diseases.

Vascular diseases

Exercise-derived exosomes can prevent and treat vascular diseases, such as coronary heart disease, myocardial infarction, and hypertension, by regulating angiogenesis, inhibiting myocardial fibrosis, and improving mitochondrial function (Table 3; Fig. 3). Long-term aerobic exercise triggers endothelial progenitor cells (EPCs) to release exosomes containing miRNA-126. Smad3 is the only direct target of miR-126 validated by luciferase assays, with this effect confirmed in both cardiomyocytes and endothelial cells. miR-126 exerts anti-fibrotic and cardioprotective effects by inhibiting Smad3. Eliminating Smad3 produces the same protective effects as miR-126, while overexpressing Smad3 eliminates the ability of exercise-derived exosomes to reduce cardiac fibrosis and collagen deposition [94]. TGF-β, an upstream activator of Smad3, when pharmacologically inhibited, similarly eliminates miR-126-mediated cardiac protection. Thus, these exercise-derived exosomes suppress the transdifferentiation of cardiomyocytes into myofibroblasts by downregulating the TGF-β/Smad3 signaling pathway. This ultimately reduces collagen fiber deposition and restores cardiac diastolic function, resulting in a 30% reduction in infarct size [94]. Furthermore, moderate-intensity exercise promotes the release of miR-342-5p from exosomes. Luciferase assays revealed that caspase 9, JNK2, and Ppm1f are all direct targets of miR-342-5p [51]. On the one hand, miR-342-5p activates the AKT pathway by inhibiting Ppm1f, thereby exerting anti-apoptotic and cardioprotective effects. However, when AKT is inhibited, the reduction in the area of myocardial infarction and the suppression of apoptosis induced by miR-342-5p are eliminated. On the other hand, miR-342-5p exerts anti-apoptotic effects by targeting caspase 9 and JNK2. Ultimately, this increases myocardial cell survival rates by 40% in ischemia-reperfusion injury [51].

Endothelin convertase 1 (ECE1) was validated as a direct target of miR-125a-5p via luciferase assays. During ischemic vascular regeneration, exercise-derived exosomes containing miR-125a-5p activate the AKT/endothelial nitric oxide synthase (eNOS) pathway by inhibiting ECE1. This process promotes angiogenesis and perfusion recovery, ultimately increasing vascular density by 50% [78]. Inhibiting AKT or eNOS eliminated the ability of miR-125a-5p to induce endothelial cell proliferation, tubule formation, and hindlimb vascular regeneration. Similarly, ECE1 overexpression reversed these angiogenic effects, confirming the indispensable role of the AKT/eNOS pathway in angiogenesis [78]. Bioinformatics predictions indicate that ROCK1 is the primary target of miR-206, though this has not been validated experimentally. Hayashi et al. [95] discovered that miR-206 in skeletal muscle exosomes derived from exercise can act on vascular endothelial cells, promoting angiogenesis by activating the VEGF pathway. In addition, C2C12 muscle progenitor cells secrete exosomes after mechanical stretching; these exosomes can activate the PI3K/AKT pathway in endothelial cells by releasing miR-206, thus accelerating vascular reconstruction in ischemic myocardium [77]. Inhibiting either the VEGF or PI3K/AKT pathway eliminates the sprouting of endothelial spheroids and recovery of post-ischemic limb blood flow induced by miR-206 [77].

The exosomes derived from different types of exercise have varying vascular protective effects. Denham et al. [96] found that long-term endurance athletes had significantly higher expression levels of miR-1, miR-486, and miR-494 in their blood than non-athletes. However, a single bout of maximal acute exercise led to an immediate decrease in the expression levels of miR-1, miR-133a, and miR-486. Furthermore, correlation analysis revealed that the expression of miR-1 and miR-486 positively correlated with maximal oxygen uptake, while miR-486 also negatively correlated with resting heart rate. These associations remained significant even after controlling for confounding factors such as age, gender, and body mass index [96]. In addition, long-term exercise continuously regulates the lncRNA CRNDE-miR-489-3p-nuclear factor E2-related factor 2 (Nrf2) axis, thus increasing Nrf2-mediated heme oxygenase-1 (HO-1) expression by 2.8 times and ultimately enhancing myocardial antioxidant capacity [73]. In the future, combining exosome-targeted delivery technology with other therapies will further promote its substantial application in the clinical translation of vascular diseases.

Metabolic diseases

Exercise-derived exosomes demonstrate therapeutic potential in metabolic diseases such as diabetes, obesity, and NAFLD by regulating insulin signaling pathways, adipocyte differentiation, and hepatic metabolic homeostasis (Table 4; Fig. 4). Wang et al. [56] discovered that long-term aerobic exercise triggers the release of serum exosomes that are rich in MALAT1. The StarBase and TargetScan platforms identified complementary binding sites between MALAT1 and miR-382-3p, and between miR-382-3p and the 3’-UTR of the brain-derived neurotrophic factor (BDNF). Dual luciferase reporter assays revealed that co-transfection of miR-382-3p mimics with either MALAT1 or BDNF 3’-UTR plasmids significantly reduced luciferase activity, confirming their direct binding. RNA immunoprecipitation revealed the co-localization of MALAT1, miR-382-3p, and BDNF within RNA-induced silencing complexes, which confirms their endogenous interaction. Knockdown of MALAT1 reduced BDNF expression and IRS-1/AKT pathway activation, promoting hippocampal neuronal apoptosis. Overexpression of miR-382-3p inhibited BDNF and the IRS-1/AKT signaling pathway. Suppression of miR-382-3p reversed these effects [56]. On the one hand, the miR-382-3p/MALAT1 axis regulates adiponectin secretion, insulin sensitivity, and prevents muscle atrophy via the IRS-1/AKT signaling pathway. On the other hand, exercise-derived, MALAT1-rich exosomes can cross the BBB to deliver MALAT1 to hippocampal neurons. MALAT1 competitively inhibits miR-382-3p, thereby upregulating BDNF and activating the PI3K/AKT/Ras-MAPK signaling pathway. This pathway reduces neuronal apoptosis and improves cognitive function [56]. Castaño et al. [85] found that HIIT-derived myogenic exosomes are enriched with miR-133a/b and miR-206. MiRmap identified conserved binding sites for these miRNAs within the 3’-UTR of the forkhead box protein O1 (FoxO1). Transfection of miR-133b mimics into 3T3-L1 cells and primary hepatocytes significantly decreased FoxO1 expression. Injection of miR-133b-loaded exosomes reduced hepatic FoxO1 expression and suppressed the expression of gluconeogenesis-related genes, such as G6pc and PCK1 [85]. This improved glucose tolerance in sedentary mice. Silencing FoxO1 recapitulated the metabolic benefits of exercise-derived exosomes, confirming FoxO1 as a functional target. Therefore, miR-133b and miR-206 inhibit FoxO1-mediated gluconeogenesis, reduce hepatic glucose output, suppress adipogenic differentiation in white adipose tissue, and indirectly promote muscle glucose uptake by regulating systemic metabolism [85]. Zhang et al. [82] discovered that long-term aerobic exercise produces exosomes that are rich in miR-324. Predictions from miRanda and TargetScan suggest that there is complementary binding between miR-324 and the 3’-UTR of the ROCK1. Dual luciferase reporter assays revealed that co-transfection of miR-324 mimics with ROCK1 3’-UTR plasmids reduced luciferase activity, confirming direct targeting [82]. Overexpression of miR-324 downregulates ROCK1 protein levels, activates the AKT/GSK3 signaling pathway, and reduces hepatic triglyceride accumulation in mice fed a high-fat diet (HFD). Silencing ROCK1 produced effects similar to those of miR-324 on insulin sensitivity and lipid metabolism, thus validating ROCK1 as a functional mediator. Therefore, miR-324 improves hepatic insulin sensitivity and reduces lipid accumulation in NAFLD by inhibiting ROCK1 [82].

In obese individuals, the fat-derived exosomes released during low-intensity resistance exercise activate IGF-1R by releasing miRNA-146a-5p, thereby inhibiting skeletal muscle atrophy and improving systemic insulin sensitivity through the regulation of adiponectin secretion [97]. In this study, the fat-derived exosomes was validated using the following approaches. First, the researchers isolated exosomes directly from white adipose tissue and detected exosome-specific markers, such as Alix, TSG101, CD63, and CD9, via Western blot. They also excluded the expression of endoplasmic reticulum markers, such as calnexin, thereby confirming the exosomal characteristics of the extracted vesicles. Second, the study used a mouse model with adipose tissue-specific miR-146a-5p knockout to further confirm the fat-derived exosomes by comparing the expression and functional differences of miR-146a-5p between fat-derived exosomes and controls. Additionally, in vitro experiments used co-cultures of 3T3-L1 adipocytes and C2C12 myoblasts that were transfected with miR-146a-5p mimics. These experiments demonstrated that miR-146a-5p can be transferred from adipocytes to myoblasts and can modulate their function [97]. Regarding exosome dose standardization, the study relied primarily on total protein concentration (e.g., 10 µg/mL for cell treatment) as the benchmark. However, the study did not standardize exosomes based on particle counting or levels of specific exosome markers. This normalization strategy, which relies on protein content rather than particle count, may lead to an incorrect estimation of the number of exosomes delivered. Consequently, it could affect the accurate interpretation of dose-response relationships, particularly when exosome protein content and particle counts vary across samples [97]. Estébanez et al. [98] conducted a 12-week training program that combined aerobic exercise and resistance training for obese individuals. Their results revealed that certain proteins carried by exosomes, such as CD81 and flotillin-1, were associated with cardiac metabolic parameters, including VO2max and left ventricular diastolic diameter. Exercise-derived exosomes may improve cardiac metabolic health in obese individuals by facilitating intercellular communication. Different types of exercise-derived exosomes exert antidiabetic effects through distinct mechanisms. Abdelsaid et al. [71] found that 2 weeks of voluntary rotational exercise and a single 45-minute 50% VO₂ max treadmill exercise session significantly increased SOD3 and ATPase copper transporting alpha (ATP7A) expression in plasma exosomes from T2DM model mice and healthy individuals. SOD3 activity depends on ATP7A-mediated copper ion transport, and exercise activates SOD3 function by restoring ATP7A levels in exosomes [71]. Furthermore, activated SOD3 catalyzes the dismutation of superoxide anion (O₂⁻) into hydrogen peroxide (H₂O₂). Elevated local H₂O₂ levels activate the VEGFR2 signaling pathway, which ultimately repairs impaired angiogenesis and skin wound healing in T2DM mice. However, when exercise intervention was applied to SOD3-knockout T2DM mice, the anti-diabetic, vascular-protective effect of exosomes disappeared. This indicates that the protective effect of exercise-derived exosomes is highly dependent on SOD3 [71]. In mice with obesity-associated insulin resistance induced by a HFD, 10 weeks of moderate-intensity swimming exercise (1–2 h per day consisting of continuous exercise in five-minute segments) exerted an anti-diabetic effect by regulating miR-27a in exosomes [99]. Specifically, the levels of miR-27a in the serum exosomes of obese mice were significantly higher than those of normal mice. However, long-term, moderate-intensity swimming exercise significantly down-regulated the expression of miR-27a. Dual luciferase reporter assays confirmed that miR-27a directly inhibits peroxisome proliferator-activated receptor gamma (PPARγ) [99]. Consequently, downregulation of miR-27a relieves its transcriptional repression of PPAR-γ. PPAR-γ activates brown adipose tissue-related pathways, such as PRDM16/PGC-1α/UCP1, which increase energy expenditure and improve obesity-related metabolic disorders. On the other hand, PPARγ also activates insulin signaling pathways in skeletal muscle, including IRS-1/Akt/GLUT4, which promotes the translocation of glucose transporter 4 (GLUT4) to the cell membrane, thereby enhancing insulin sensitivity. The 2-hour/day moderate-intensity swimming exercise group demonstrated superior miR-27a downregulation, PPAR-γ activation efficiency, and metabolic improvement compared to the 1-hour/day group. This suggests that the duration of the exercise intervention may positively correlate with the efficiency of exosomal miRNA regulation [99].

Exercise-derived exosomes can reverse the disruption of metabolic homeostasis during aging. Following aerobic exercise, the plasma exosomes of aged mice exhibited increased levels of miR-323-5p, which balances adipocyte differentiation and lipolysis by promoting PPARγ phosphorylation [100]. Chong et al. [84] also found that 20 min of moderate-intensity cycling can promote the release of extracellular nicotinamide phosphoribosyltransferase in exosomes, maintaining tissue NAD⁺ levels through systemic transmission and ultimately alleviating age-related metabolic disorders.

These studies reveal the cross-organ regulatory mechanisms of exercise-derived exosomes and provide new targets for the precise treatment of metabolic diseases. For instance, the sustained high expression of SOD3 in circulating exosomes after exercise helps maintain vascular function in patients with diabetes [101]. In patients with diabetes, engineered modified exosomes delivering miR-200a-3p can enhance lactate transport in the hippocampus by activating the Kelch-like ECH-associated protein 1/heat shock protein 90 alpha family class A member 1 pathway, thus improving cognitive function and glucose metabolism [102]. In the future, spatiotemporal targeted delivery technology can be combined with exosomes to advance the clinical application of exercise-derived exosomes for treating metabolic system diseases [103].

Musculoskeletal diseases

Exercise-derived exosomes exert therapeutic effects on musculoskeletal system diseases, such as osteoporosis, skeletal muscle atrophy, osteoarthritis, intervertebral disc degeneration, and sarcopenia, by regulating bone metabolic balance, muscle regeneration microenvironment, and joint oxidative stress (Table 5; Fig. 5). FNDC5/irisin, carried by skeletal muscle-derived exosomes, binds to caveolin-1 to activate the AMP-activated protein kinase/Nrf2 signaling pathway for the prevention and treatment of osteoporosis. It promotes osteoblast proliferation and inhibits ferroptosis by upregulating heme oxygenase 1 and ferroportin. Tao et al. [104] found that it improved bone density and microstructure in ovariectomized mice. Neuromuscular electrical stimulation (NMES) can prompt the release of miR-17-5p and miR-19b-3p from skeletal muscle exosomes. These miRNAs inhibit adipocyte differentiation and reduce the damage caused by bone marrow fat deposition to the bone microenvironment, ultimately alleviating osteoporosis [105]. This muscle–bone interaction has also been studied in in vitro exercise simulation experiments. Following mechanical stretching, C2C12 muscle progenitor cell-derived exosomes activate the PI3K/AKT signaling pathway in vascular endothelial cells by releasing miR-146b-5p and miR-183-5p, thus accelerating angiogenesis in ischemic muscles [106]. In addition, NMES activates IGF-1R signaling by increasing the expression of miR-146a-5p in serum exosomes. As a result, skeletal muscle atrophy is inhibited, and the contractile function of aged muscles is improved by approximately 30% [97, 107].

In osteoarthritis, the exosomes derived from swimming exercise decrease interleukin-1 beta (IL-1β) expression and increase aggrecan expression in synovial fluid by reshaping the oxidative-lipid network [108]. This effect is partly attributed to SOD3, which is carried by exosomes and can clear ROS in the joint cavity. SOD3 can also inhibit abnormal cholesterol deposition mediated by the CCAAT/enhancer-binding protein beta-apolipoprotein E axis and ultimately delay cartilage matrix degradation [108]. The core pathological features of intervertebral disc degeneration include an imbalance in extracellular matrix metabolism and inflammatory macrophage infiltration, which are triggered by nucleus pulposus cells (NPCs). Exercise-derived exosomes can disrupt this cycle. NPCs and macrophages can internalize exercise-derived exosomes [109]. In NPCs, exercise-derived exosomes inhibit the NF-κB signaling pathway by releasing irisin. This process upregulates the expression of anabolic factors, such as sex-determining region Y-related high-mobility group box 9 and aggrecan, and downregulates the expression of degradative enzymes, such as matrix metalloproteinase 13 and a disintegrin and metalloproteinase with thrombospondin motifs 5. Exercise-derived exosomes also reduce the release of pro-inflammatory factors, such as tumor necrosis factor-alpha and IL-1β, and chemokine factors, such as CC chemokine ligand 2. In macrophages, exercise-derived exosomes release irisin to regulate polarization, thus reducing the infiltration of M1-type macrophages and promoting the formation of M2-type macrophages. Ultimately, local inflammatory responses and lumbar disc degeneration are alleviated [109].

Exercise-derived exosomes play an important role in age-related sarcopenia. Sarcopenia is characterized by the progressive loss of skeletal muscle mass and function. Aging significantly alters exosome secretion and cargo composition. For instance, the expression of exosome markers, such as CD81 and CD9, is reduced in the circulation of older adults, and the levels of miRNAs associated with muscle health, such as miR-23a, miR-27a, and miR-133a, are significantly decreased in exosomes. These miRNAs inhibit muscle atrophy and promote myocyte differentiation [110]. Low-intensity resistance training for 12 weeks partially reversed these changes by restoring the levels of miR-23a and miR-27a in circulating exosomes and increased the expression of proteins related to exosome biogenesis, such as TSG101 and Alix [110, 111]. Therefore, exercise may improve sarcopenia by enhancing the secretion and function of exosomes.

Exercise-derived exosomes carry various bioactive molecules, including miRNAs, proteins, and lipids. They simultaneously act on multiple signaling pathways, cell types, and tissues/organs, ultimately treating diseases of the musculoskeletal system. For instance, mechanical stress-derived exosomes promote myoblast proliferation by inhibiting the mTOR pathway [106] and prevent osteoclast differentiation by transporting miR-27a-3p [105]. This cross-organizational regulation is crucial for repairing aging muscles. NMES-induced exosomes can restore the dysregulated miRNA profile in aged skeletal muscle and indirectly improve neuromuscular junction function by activating IDH1 to inhibit neuronal necrotic apoptosis [107]. Therefore, exercise-derived exosomes are used to treat musculoskeletal diseases via various mechanisms.

Cross-diseases regulatory network

In various diseases, exercise-derived exosomes exert therapeutic effects through a cross-disease regulatory network characterised by a core signaling axis and tissue-specific targets, rather than through isolated molecules and signaling pathways. The five core regulatory axes are antioxidant, anti-inflammatory, anti-apoptotic, pro-angiogenic and metabolic reprogramming. These axes are widely present in the nervous, cardiovascular, metabolic and musculoskeletal systems, forming common modules of exercise-derived exosome protective effects. For example, molecules such as miR-342-5p, SOD3 and miR-146a-5p appear repeatedly across multiple disease models [51, 71, 97]. By inhibiting apoptosis, scavenging oxidative stress, or reducing inflammation, they confer identical protective effects across different organs. However, molecules associated with angiogenesis and metabolic reprogramming, such as miR-126, miR-133b/206, eNAMPT and miR-324, exhibit differences in tissue-specific targeting [76, 77, 81, 82]. The nervous system relies more on BBB stability and synaptic plasticity; the heart focuses on antifibrosis and microvascular regeneration; and the liver and adipose tissue depend primarily on balancing energy metabolism and maintaining lipid homeostasis.

The key miRNAs and proteins found in most exercise-derived exosomes tend to exert antioxidant, anti-apoptotic, anti-fibrotic, pro-angiogenic and metabolic regulatory functions by modulating signaling pathways such as PI3K/AKT, Nrf2/HO-1, VEGF/eNOS, TGF-β/Smad and FOXO1 [32, 73, 77]. Differences across diseases are primarily determined by tissue microenvironments, receptor distribution, barrier structures, and inflammation or hypoxia. Consequently, the same core molecule may exert tissue-specific effects through distinct downstream targets in different pathologies. For example, in addition to its anti-fibrotic and cardioprotective effects, which are achieved by inhibiting Smad3 [94], miR-126 also promotes microvascular angiogenesis in the cardiovascular system by activating the AKT/eNOS pathway [76, 77]. Taken together, these cross-disease regulatory networks reveal a unified biological logic for exercise-derived exosomes across multiple organs and pathological states, whereby systemic protection is achieved through a relatively constant set of core signalling axes, followed by differentiated therapeutic effects via tissue-specific molecules.

Applications of engineered exosomes

Although exercise-derived exosomes have advantages as “exercise mimetics”, including low immunogenicity, natural tissue permeability and systemic regulatory capabilities, several limitations hinder their clinical translation. These include insufficient targeting specificity towards pathological tissues, a low capacity to carry therapeutic molecules endogenously, and poor stability in vivo. Engineered exosomes have therefore emerged as a key strategy to enhance therapeutic efficacy, with surface modification and payload loading forming the two core technological pillars.

Surface modification strategies: enhancing targeting specificity and bioavailability

The surface modification of exosome-ribonucleoprotein complexes is intended to improve the precision with which they target diseased cells, prolong their circulation time and strengthen their interactions with recipient cells, by functionalising the lipid bilayer with exogenous molecules. Current strategies include ligand-mediated targeting modifications, membrane fusion modifications and polyethylene glycol (PEG) modifications [112].

Ligand-mediated targeted modification exploits the specific binding of ligands to receptors on target cells. This enables the precise targeting of exosome-RNA complexes to pathological tissues. Thanks to its high specificity and biocompatibility, this method is now the most widely adopted approach for surface modification. For example, Wang et al. [113] used amide bonding technology to attach the arginylglycylaspartic acid (RGD) peptide to the surface of exosomes. In a mouse model of myocardial infarction, RGD-modified exosomes accumulated significantly more in ischaemic myocardium than unmodified exosomes. Delivering miR-210 via these exosomes promoted angiogenesis and ultimately improved cardiac function recovery [114]. In neurological disorders, Cui et al. [115] utilised rabies virus glycoprotein (RVG) peptide modification on the surface of MSC-derived exosomes. This facilitated the crossing of the BBB by the exosomes and the delivery of BDNF, alleviating cognitive impairment in mice with AD by reducing β-amyloid deposition [116].

Membrane fusion modification involves integrating functional molecules from donor membranes into exosomes by fusing them with membranes from other cellular sources or synthetic lipid vesicles [117]. This approach preserves the native structure of the exosomes while imparting new targeting or functional properties. For example, one study involved fusing exosomes with macrophage membranes to improve their ability to recognise pro-inflammatory M1 macrophages in the synovium of patients with rheumatoid arthritis (RA). These fused exosomes accumulated specifically in inflamed joints, re-polarising M1 macrophages to an anti-inflammatory M2 phenotype and greatly reducing synovial inflammation and joint damage in RA mice [118]. Furthermore, fusion with platelet membranes improves the blood compatibility of exosomes, reducing their clearance by the immune system [119].

PEG is a well-established approach to enhancing the stability of nanocarriers in vivo. Covalently attaching polyethylene glycol molecules to the surface of vesicles forms a hydration layer that reduces phagocytosis by reticuloendothelial system cells, thereby prolonging circulation time [120]. In the treatment of T2DM with miR-34a inhibitors, polyethylene glycol-modified exosomes demonstrated greater accumulation in the liver and adipose tissue, thereby improving insulin resistance [121]. Notably, PEGylation can be combined with targeted ligands to achieve both prolonged circulation and specific targeting simultaneously, thus optimising the therapeutic performance of cell-derived exosomes further [122].

Cargo loading strategies: precise regulation of therapeutic function

Payload loading involves encapsulating exogenous therapeutic molecules, such as miRNAs, proteins and small-molecule drugs, into exosomes. This process can either enhance their inherent therapeutic efficacy or confer novel functions. Payload loading uses non-covalent interactions or physical techniques to encapsulate payloads within exosomes, providing a simple and versatile approach for a wide range of payload types. Common methods include electroporation, ultrasound-assisted loading and extrusion [123]. Electroporation involves creating transient pores in the exosome membrane using brief electrical pulses to allow payload molecules to enter. For example, Bheri et al. [124] used electroporation to load the pro-angiogenic miR-126 into exosomes, and the resulting miR-126-loaded exosomes greatly enhanced tube formation in cultured cardiac endothelial cells in vitro. In a rat myocardial ischaemia-reperfusion model, miR-126-loaded exosomes greatly improved cardiac function, reduced infarct size and increased vascular density [124]. Ultrasound-assisted loading uses low-intensity ultrasound to create cavitation effects that enhance membrane permeability without damaging the exosome structure. Ultrasonic treatment creates transient pores in the exosome membrane, thereby improving drug permeation efficiency [125]. The extrusion method involves forcing exosome vesicles and cargo through a porous membrane under pressure to achieve encapsulation through membrane rupture and reorganization. This method is suitable for large-scale production and generates exosome mimics of uniform size [126].

Therapeutic applications of engineered exosomes in diseases

Engineered exosomes offer numerous advantages for treating diseases. They combine the excellent biocompatibility and low immunogenicity of natural vesicles, circulate more stably in vivo and are more easily taken up by cells [127]. Furthermore, unlike natural exosomes, which lack clear targeting capabilities, engineered exosomes can significantly enhance their enrichment efficiency in diseased tissues, thereby improving therapeutic efficacy [128]. Combining surface modification with payload delivery allows engineered exosomes to demonstrate significant therapeutic effects in neurological, vascular, metabolic, and musculoskeletal disorders. In neurological disorders, RVG-modified exosomes loaded with BDNF can cross the BBB efficiently, thereby promoting neurogenesis and synaptic plasticity in mouse models of AD while enhancing memory function [115, 116]. In cardiovascular diseases, RGD-modified exosomes loaded with VEGF promote ischaemic cardiac revascularisation, reducing myocardial fibrosis by 42% and improving cardiac function in mice following myocardial infarction [113, 129, 130]. In metabolic diseases, exosomally delivered miR-146a suppresses inflammation and collagen deposition by targeting IRAK1, thereby promoting wound healing in diabetic conditions [131]. Despite the promising potential of engineered exosomes for treating diseases, there are still numerous challenges to overcome in the transition from laboratory to clinical application. Most culture systems struggle to achieve industrial-scale amplification while maintaining exosome yield and purity [123]. Furthermore, the complex microenvironment of the human body affects the efficiency with which engineered exosomes can be targeted, and their long-term safety is uncertain [132].

Conclusion and prospects

Exosomes have been extensively studied as important mediators of intercellular communication. However, the regulatory roles of exercise-derived exosomes in various physiological and pathological states have never been summarized comprehensively. This review summarizes research on exercise-derived exosomes and demonstrates their role as key mediators of systemic health and disease treatment. Exosomal cargo activates core signaling pathways, including antioxidant, anti-inflammatory, pro-angiogenic, anti-apoptotic, and metabolic reprogramming, thereby providing multiple benefits. The antioxidant, anti-inflammatory, and anti-apoptotic pathways are highly similar across the neural, vascular, metabolic, and musculoskeletal systems, generating broad protective effects. Conversely, the pro-angiogenic and metabolic reprogramming pathways often demonstrate stronger organ specificity, enabling precise regulation tailored to the unique functional demands of tissues such as ischemic myocardium, the diabetic liver, and aging bone [62, 133].

Numerous studies have demonstrated the therapeutic potential of exercise-derived exosomes for diseases such as AD, atherosclerosis, T2DM, and osteoarthritis. However, translating these findings into clinical applications remains challenging. First, the relationship among the dose, type, intensity, and duration of exercise and exosome characteristics is unclear. Second, individual differences among different populations may significantly impact the biological functions of exosomes. Finally, issues such as the efficiency of exosome delivery in vivo, immunogenicity, and standardized preparation processes hinder their clinical translation. For example, there is currently no unified standard for the isolation and quality control of exosomes. Different isolation approaches produce varying yields and levels of purity. Although differential ultracentrifugation typically achieves high particle recovery, it also co-enriches cellular debris, protein aggregates and lipoproteins, resulting in a low particle-to-protein ratio. Size-exclusion chromatography outperforms direct sedimentation in removing soluble proteins and small-molecule impurities, yet it exhibits limited resolution for non-exosomal nanoparticles of similar size. Due to methodological variations, both upstream pretreatment and downstream characterisation can have a significant impact on results.

Future research can be conducted in the following three areas. First, gene engineering must be utilized to modify the surface receptors of exosomes to develop exercise-derived exosomes targeting specific tissues. Second, a multimodal separation and purification strategy is employed during exosome isolation to enhance purity and ensure scalability. Adherence to the Minimum Information Standards for Extracellular Vesicles (MISEV) guidelines ensures rigorous reporting and reproducible results. Third, single-cell sequencing technologies must be combined to explore the heterogeneity of exosome-producing cells and clarify the cellular origin and functional roles of exercise-derived exosomes. Fourth, a predictive model based on the miRNA profile of exosomes must be established. For example, the miR-23a and miR-27a levels in circulating exosomes must be monitored to predict the metabolic response of patients with diabetes to exercise intervention [18]. With the integration of exosome engineering technology and precision medicine, exercise-derived exosomes are expected to become molecular substitutes for exercise. They will provide individualized, cell-free treatment options for patients who cannot tolerate traditional exercise and will drive the development of sports medicine from behavioral intervention therapy to molecular targeted therapy [134].

Abbreviations

CD63

Cluster of Differentiation 63

VAMP3

Vesicle-Associated Membrane Protein 3

SGCA

Sarcoglycan Alpha

LPO

Lipid Peroxidation

TLR4

Toll-Like Receptor 4

NF-κB

Nuclear Factor Kappa-B

GSH

Glutathione

PTEN

Phosphatase and Tensin Homolog

FOXO

Forkhead Box O

IGF-1

Insulin-Like Growth Factor 1

Alix

ALG-2 Interacting Protein X

Rab27A

Ras-Related Protein Rab27A

PGC-1α

Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1-Alpha

VEGF

Vascular Endothelial Growth Factor

TSG101

Tumor Susceptibility Gene 101

ANXA1

Annexin A1

MPO

Myeloperoxidase

S100A8

S100 Calcium Binding Protein A8

LCP1

Lipocortin 1

BASP1

Brain Acid Soluble Protein 1

HIST1H4A

Histone Cluster 1 H4A

LYZ

Lysozyme

S100A9

S100 Calcium Binding Protein A9

CD9

Cluster of Differentiation 9

CD41b

Cluster of Differentiation 41b

ATP2A1

ATPase Plasma Membrane Ca²⁺ Transporting 1

β-enolase

Beta-Enolase

Pax7

Paired Box 7

MyHC

Myosin Heavy Chain

CD81

Cluster of Differentiation 81

HSP60

Heat Shock Protein 60

Myf5

Myogenic Factor 5

CK

Creatine Kinase

PI3K

Phosphatidylinositol 3-Kinase

Akt

Protein Kinase B

AchE

Acetylcholinesterase

ROS

Reactive Oxygen Species

EPHA4

Ephrin type-A receptor 4

ZO-1

Zonula occludens-1

PDGFRβ

Platelet-derived growth factor receptor beta

NG2

Neuron-glial antigen 2

LRP1

Low-density lipoprotein receptor-related protein 1

ACSL4

Acyl-CoA synthetase long-chain family member 4

GPX4

Glutathione peroxidase 4

SLC7A11

Solute carrier family 7 member 11

STAT3

Signal transducer and activator of transcription 3

BCL-2

B-cell lymphoma 2

BAX

BCL2-associated X protein

IDH1

Isocitrate dehydrogenase 1

BCL2L11

BCL2-like 11

SIRT1

Sirtuin 1

RIP1

Receptor-interacting serine/threonine-protein kinase 1

RIP3

Receptor-interacting serine/threonine-protein kinase 3

Syn

Alpha-synuclein

PSD-95

Postsynaptic density protein 95

Aβ

Amyloid-beta

Tau

Microtubule-associated protein tau

JNK1

c-Jun N-terminal kinase 1

c-Jun

Transcription factor AP-1 subunit c-Jun

Caspase-3

Cysteine-aspartic acid protease 3

GAP-43

Growth-associated protein 43

MAP-2

Microtubule-associated protein 2

NF-200

Neurofilament heavy polypeptide

BDNF

Brain-derived neurotrophic factor

TrkB

Tropomyosin receptor kinase B

MAPK

Mitogen-activated protein kinase

COL1A1

Collagen type I alpha 1 chain

COL3A1

Collagen type III alpha 1 chain

MYH1

Myosin heavy chain 1

Nox4

NADPH oxidase 4

ECE1

Endothelin-converting enzyme 1

eNOS

Endothelial nitric oxide synthase

TGF-β

Transforming Growth Factor-Beta

Smad3

Mothers Against Decapentaplegic Homolog 3

α-SMA

Alpha-Smooth Muscle Actin

Caspase 9

Cysteinyl Aspartate-Specific Proteinase 9

JNK2

c-Jun N-Terminal Kinase 2

Ppm1f

Protein Phosphatase 1 F

ET-1

Endothelin-1

NO

Nitric Oxide

CRNDE

Colorectal Neoplasia Differentially Expressed

Nrf2

Nuclear Factor Erythroid 2-Related Factor 2

HO-1

Heme Oxygenase-1

Keap1

Kelch-Like ECH-Associated Protein 1

GF-1

Insulin-Like Growth Factor 1

SDF1

Stromal Cell-Derived Factor 1

SPRED1

Sprouty-Related EVH1 Domain-Containing Protein 1

Gax

GATA Binding Protein 6 (GATA6) Alternative Splice Variant

MALAT1

Metastasis-Associated Lung Adenocarcinoma Transcript 1

Ras

Rat Sarcoma Virus Oncogene Homolog

SOD3

Superoxide Dismutase 3

VEGFR2

Vascular Endothelial Growth Factor Receptor 2

Hsp90aa1

Heat Shock Protein 90 Alpha, Class A Member 1

Mct2

Monocarboxylate Transporter 2

PPAR-γ

Peroxisome Proliferator-Activated Receptor Gamma

IRS-1

Insulin Receptor Substrate 1

GLUT-4

Glucose Transporter Type 4

UCP1

Uncoupling Protein 1

PRDM16

PR Domain Containing 16

FoxO1

Forkhead Box Protein O1

G6PC

Glucose-6-Phosphatase, Catalytic Subunit

PCK1

Phosphoenolpyruvate Carboxykinase 1

PRDX2

Peroxiredoxin 2

GPX3

Glutathione Peroxidase 3

PTGS1

Prostaglandin-Endoperoxide Synthase 1

MFN2

Mitofusin 2

OPA1

Optic Atrophy 1

CD11c

Cluster of Differentiation 11c

DUSP3

Dual Specificity Phosphatase 3

p-ERK

Phosphorylated Extracellular Signal-Regulated Kinase

p-PPARγ

Phosphorylated Peroxisome Proliferator-Activated Receptor Gamma

C/EBPα

CCAAT/Enhancer-Binding Protein Alpha

ROCK1

Rho-Associated Coiled-Coil Containing Protein Kinase 1

GSK3

Glycogen Synthase Kinase 3

PEPCK

Phosphoenolpyruvate Carboxykinase

ACC

Acetyl-CoA Carboxylase

NDRG4

N-Myc Downstream Regulated Gene 4

FAM13A

Family With Sequence Similarity 13, Member A

ST3GAL6

ST3 Beta-Galactoside Alpha-2,3-Sialyltransferase 6

AFF1

A-Kinase Anchoring Protein Family Member 1

eNAMPT

Extracellular Nicotinamide Phosphoribosyltransferase

NAD+

Nicotinamide Adenine Dinucleotide

PGC1α

Peroxisome Proliferator-Activated Receptor-γ Coactivator 1α

FNDC5

Fibronectin Type III Domain Containing 5

AMPKα

AMP-Activated Protein Kinase Alpha Subunit

HMOX1

Heme Oxygenase 1

Fpn

Ferroportin

CDK2

Cyclin-Dependent Kinase 2

SOX9

SRY-Box Transcription Factor 9

ACAN

Aggrecan

MMP13

Matrix Metalloproteinase 13

ADAMTS5

A Disintegrin and Metalloproteinase with Thrombospondin Motifs 5

TNF-α

Tumor Necrosis Factor Alpha

IL-1β

Interleukin-1 Beta

iNOS

Inducible Nitric Oxide Synthase

CD86

Cluster of Differentiation 86

Arg1

Arginase 1

CD206

Cluster of Differentiation 206

IGF-1R

Insulin-Like Growth Factor 1 Receptor

mTOR

Mammalian Target of Rapamycin

FoxO3

Forkhead Box O3

Fbx32

F-Box Protein 32

MuRF

Muscle Ring Finger Protein

MyoD

Myoblast Determination Protein 1

MyoG

Myogenin

IL-6

Interleukin-6

Smad7

SMAD Family Member 7

Smurf2

Smad Ubiquitin Regulatory Factor 2

ACVR1

Activin A Receptor Type 1

HDAC4

Histone Deacetylase 4

Runx2

Runt-Related Transcription Factor 2

ALP

Alkaline Phosphatase

OCN

Osteocalcin

BMP

Bone Morphogenetic Protein

EBPβ

CCAAT/Enhancer-Binding Protein Beta

APOE

Apolipoprotein E

PTPN1

Protein Tyrosine Phosphatase Non-Receptor Type 1

CD82

Cluster of Differentiation 82

CLR

C-Type Lectin Receptor

Th17

T Helper 17 Cell

Flot-1

Flotillin-1

HSP70

Heat Shock Protein 70

CD14

Cluster of Differentiation 14

CD142

Cluster of Differentiation 142

MCH-I

Major Histocompatibility Complex Class I

ICAM-1

Intercellular Adhesion Molecule 1

LAMP-1

Lysosome-Associated Membrane Protein 1

tPA

Tissue Plasminogen Activator

Endothelial Progenitor Cells

EPCs

BBB

blood-brain barrier

ATP7A

ATPase copper transporting alpha

T2DM

type 2 diabetes mellitus

AD

Alzheimer’s disease

CSF

cerebrospinal fluid

Author contributions

H.K., J.L. and M.H. conceived and drafted the manuscript. H.K., Z.Z., Y.L. and X.T. finished the figures. J.H., X.M., X.W. and Y.F. revised the figures. H.K., J.L., Y.L. and J.H. were the major contributors to modifying and finalizing the manuscript. M.H. and X.Z. contributed greatly to revising the language and guiding revision ideas. All authors reviewed and approved the manuscript.

Funding

The work was supported by the National Natural Science Foundation of China under Grant [number 32371184, 32371244], the National Key Research and Development Program of China under Grant [number 2024YFC3607304], Shanghai Eastern Talent Plan Leading Project (2023). The work was also supported by innovative research team of high-level local universities in Shanghai (SHSMUZDCX20212000 and SHSMU-ZDCX20211202), Shanghai Frontiers Science Center of Cellular Homeostasis and Human Diseases, and the Fundamental Research Funds for the Central Universities to M.H. lab. The work was also supported by the Natural Science Foundation of Guangxi (2025GXNSFAA069105).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors have given their consent for publication.

Competing interests

The authors declare no competing interests.