Extracellular vesicles from adipose-derived stem cells in bone regeneration: Mechanisms and therapeutic advances

Abstract

Extracellular vesicles (EVs) secreted by adipose-derived stem cells (ADSCs) have emerged as a promising cell-free therapeutic tool for bone regeneration. These EVs deliver a diverse array of bioactive molecules, including proteins, lipids, and nucleic acids, thereby modulating the bone microenvironment, activating key signaling pathways, and promoting bone regeneration. Innovative strategies involving preconditioning, genetic modification, and biomaterial-assisted delivery have been explored, with preclinical studies demonstrating synergistic effects that enhance targeting specificity and therapeutic efficacy. Functionally, EVs derived from ADSCs promote osteogenesis by enhancing osteoblast and mesenchymal stem cell activity, support angiogenesis through vascular endothelial growth factor signaling, and modulate inflammation by shifting macrophages from pro-inflammatory to anti-inflammatory phenotypes. In disease-specific contexts, they reduce cartilage degradation and support subchondral bone restoration in osteoarthritis, while in osteoporosis, they help restore the balance between bone formation and resorption and mitigate bone loss. Despite these promising developments, challenges remain in standardizing production protocols, optimizing delivery systems, and confirming long-term safety and efficacy in clinical settings. This review summarizes current insights into the mechanisms of EVs derived from ADSCs in bone-related diseases and highlights recent innovations and future directions that may accelerate their clinical application as a regenerative therapy.

Key Words: Adipose-derived stem cell; Extracellular vesicle; Bone regeneration; Fracture healing; Osteoarthritis; Osteoporosis; Therapeutic mechanism

Core Tip: Extracellular vesicles derived from adipose-derived stem cells exhibit significant potential in bone regeneration. They mediate osteogenesis, angiogenesis, and immunomodulation via bioactive molecules, with translational strategies including preconditioning, cargo engineering, and biomaterial integration. This review summarizes their efficacy in fractures, osteoarthritis, and osteoporosis, and discusses clinical translation challenges.

INTRODUCTION

The skeletal system plays a crucial role in providing structural support, protecting vital organs, and facilitating movement. It undergoes continuous remodeling through the meticulously balanced processes of bone formation and resorption[1]. Disruptions in this equilibrium, whether caused by aging, illness, or nutritional deficiencies, can impede bone repair mechanisms and contribute to a range of conditions, including fractures, osteoporosis (OP), and osteoarthritis (OA)[2]. According to the International Osteoporosis Foundation, OP alone affects approximately 200 million people worldwide, with one in three women and one in five men over the age of 50 years experiencing osteoporotic fractures in their remaining lifetime. OA is a leading cause of joint pain and disability, significantly impacting global health[3]. Current treatments, including pharmacological interventions and surgical procedures, often fail to reverse disease progression or promote tissue regeneration[4]. Therefore, there is an urgent need for novel therapeutic strategies, particularly those based on stem-cell-derived regenerative medicine approaches, to address these limitations and improve patient outcomes.

In recent years, cell-based therapies, particularly those using stem cells, have gained prominence in regenerative medicine for treating musculoskeletal disorders[5]. The regenerative potential of stem cells is due to their pluripotency, self-renewal ability, and the secretion of bioactive factors[6]. Of the stem cells available, adult stem cells such as mesenchymal stem cells (MSCs) are the most widely studied due to their accessibility and potential for differentiation into tissue-specific cells. Bone marrow-derived MSCs (BMSCs) are widely studied for musculoskeletal regeneration applications, owing to their potent osteogenic differentiation potential and capacity to generate functional osteoblasts[7]. However, procurement requires invasive techniques such as bone marrow aspiration, which are associated with patient discomfort and yield low cell numbers. Additionally, BMSC functionality demonstrates an age-dependent decline in therapeutic potency. Moreover, donor comorbidities further undermine BMSC quality and therapeutic efficacy[8]. In contrast, adipose-derived stem cells (ADSCs) present unique advantages for clinical use, enabled by their scalable isolation via minimally invasive techniques. This approach circumvents the supply constraints and ethical issues linked to BMSC utilization, offering a more practical alternative for regenerative medicine applications[9].

ADSCs have demonstrated the ability to differentiate into various cell lineages, including osteocytes, chondrocytes, and myocytes, making them suitable for applications in bone and cartilage regeneration[10]. Their immunomodulatory properties and secretion of anti-inflammatory cytokines enhance their therapeutic potential in treating bone diseases[11]. Additionally, ADSCs secrete extracellular vesicles (EVs) that encapsulate a diverse cargo of microRNAs (miRNAs), proteins, and lipids, which mediate their immunomodulatory and regenerative effects[12]. To maximize their clinical utility, the development of efficient isolation and characterization protocols remains essential. While alternative techniques, including size-exclusion chromatography, ultrafiltration, and immunoaffinity-based approaches[13] exist, ultracentrifugation persists as the predominant methodology for EV isolation[14]. For characterization, transmission electron microscopy allows for the visualization of EV morphology, nanoparticle tracking analysis quantifies size distribution, and flow cytometry evaluates surface marker expression[15]. These techniques ensure that EVs derived from ADSCs (ADSC-EVs) used in regenerative applications are well-defined and reproducible, which are essential criteria for clinical translation.

ADSC-EVs offer several advantages over traditional cell-based therapies. Notably, they mitigate risks associated with cell transplantation, such as tumorigenesis and immune rejection, due to the absence of replicating cells[16]. Additionally, their acellular nature allows for easier storage, preservation, and standardization, enhancing their clinical applicability[17]. In the context of bone regeneration, ADSC-EVs have demonstrated efficacy by promoting osteogenic differentiation, enhancing angiogenesis, and modulating immune responses. These mechanisms collectively contribute to improved bone repair outcomes[18]. For therapeutic optimization, investigators are actively exploring advanced modification strategies. These include parent cells preconditioning, cargo modification engineering, and integration with biomaterial scaffolds, all aimed at enhancing EV functionality and delivery precision[19].

In summary, ADSC-EVs represent a promising cell-free therapeutic approach for bone regeneration, offering advantages in safety, storage, and efficacy. Ongoing research focusing on understanding their mechanisms of action and developing strategies to enhance their clinical utility holds the potential to revolutionize treatments for bone-related diseases. This review first synthesizes advances in translational strategies for ADSC-EVs, including preconditioning, cargo engineering, and biomaterial integration, and then explores their therapeutic efficacy in bone regeneration across fractures, OA, and OP, while concurrently tackling translational barriers to facilitate clinical translation.

TRANSLATIONAL STRATEGIES OF ADSC-EVS

ADSC-EVs have garnered significant attention as a novel acellular therapeutic platform for bone regeneration applications. Nevertheless, native EV formulations frequently exhibit limited therapeutic potency, manifesting as inadequate in vivo persistence, accelerated clearance kinetics, and consequently reduced sustained bioactivity. These formulations also demonstrate suboptimal targeting precision to osseous defect sites and insufficient regenerative capacity to address complex bone healing demands[20]. Building on insights into their natural functions, multiple translational strategies have been developed to optimize ADSC-EVs’ biological activity, targeting specificity, and in vivo retention, thereby enhancing therapeutic efficacy and clinical feasibility. These strategies are categorized by optimization objectives: (1) Biological or physical preconditioning of ADSCs to enhance the intrinsic quality of secreted EVs; (2) Genetic and pharmacological engineering to modulate EV cargo composition; and (3) Biomaterial-assisted delivery systems to improve retention, targeting precision, and controlled release kinetics at defect sites (Figure 1). This section systematically reviews these strategies, highlighting how they leverage and amplify the intrinsic properties of ADSC-EVs to address clinical challenges in bone tissue engineering and regeneration.

Figure 1 Translational strategies of adipose-derived stem cell-derived extracellular vesicles in therapeutic use. Adipose-derived stem cell-derived extracellular vesicles (EVs) are optimized through three core strategies to enhance their therapeutic efficacy. Biological or physical preconditioning of adipose-derived stem cells, including osteogenic induction, inflammatory priming with cytokines, pulsed electromagnetic field stimulation, and others, refines the regenerative capacity of secreted EVs. Genetic and pharmacological engineering strategies, such as microRNA overexpression, curcumin loading, surface modification with Cys-Arg-Glu-Lys-Ala peptide or biotin-avidin systems, and additional approaches, modulate EV cargo and targeting ability. Biomaterial-assisted delivery, via platforms like poly-lactic-co-glycolic acid/polydopamine scaffolds, decellularized extracellular matrix hydrogels, injectable gelatin-nanoparticle hydrogels, and more, improves EV retention and release kinetics at defect sites. These optimized EVs collectively contribute to bone regeneration, supporting their application in preclinical models and potential translation to clinical settings. Created in BioRender (Supplementary material). ADSC: Adipose-derived stem cell; ADSC-EVs: Adipose-derived stem cell-derived extracellular vesicles; PLGA: Poly-lactic-co-glycolic acid; PDA: Polydopamine; 3D: Three-dimensional.

Biological and physical preconditioning of ADSCs to optimize EV secretion

Preconditioning ADSCs with external stimuli can significantly modulate paracrine signaling pathways and enhance the regenerative capacity of secreted EVs, without requiring direct genetic manipulation. In bone regeneration contexts, osteogenic induction of ADSCs using osteoinductive media has been widely adopted to prime EVs with pro-osteogenic cargo. For instance, EVs derived from osteogenically induced ADSCs, when incorporated into a poly-lactic-co-glycolic acid/polydopamine scaffold, promoted enhanced bone formation, collagen deposition, and elevated expression of runt-related transcription factor 2 (RUNX2) and osteocalcin (OCN) in a mouse calvarial defect model[21]. Inflammatory priming with cytokines such as tumor necrosis factor-alpha (TNF-α) and interferon-γ has been shown to upregulate immunomodulatory and pro-regenerative miRNAs in ADSC-EVs. A study demonstrated that TNF-α/interferon-γ-stimulated ADSCs released EVs enriched in miR-27b-3p, which enhanced osteogenic and chondrogenic differentiation of BMSCs, promoted M2 macrophage polarization, and facilitated osteochondral repair when delivered via a silk fibroin scaffold in a rabbit temporomandibular joint OA model[22]. Additionally, biophysical stimulation of ADSCs using pulsed electromagnetic fields (PEMF; 15-75 Hz) augmented EV secretion and therapeutic efficacy. In an anterior cruciate ligament transaction-induced OA model, PEMF-stimulated ADSC-EVs reduced catabolic markers, including matrix metalloproteinases (MMP)-13 and interleukin (IL)-1β, preserved cartilage matrix, and enhanced histological scores[23]. These preconditioning strategies provide a straightforward yet effective approach to enhance the regenerative potential of EVs for bone and cartilage tissue repair.

Genetic and pharmacological engineering of ADSC-EVs for functional enhancement

To achieve precise molecular control over EV functionality, recent studies have employed genetic and pharmacological engineering approaches to selectively modulate the cargo composition of ADSC-EVs. A key strategy involves the transfection of ADSCs with miRNA mimics. For example, EVs derived from miR-486-5p-overexpressing ADSCs attenuated chondrocyte apoptosis, mitigated inflammatory responses, and maintained cartilage matrix integrity in a destabilization of the medial meniscus-induced OA model[24]. Similarly, miR-146a-enriched EVs from transfected ADSCs exhibited robust anti-inflammatory activity and enhanced bone mineral density (BMD) in a diabetic OP rat model via downregulation of the NOD-like receptor pyrin domain-containing 3 inflammasome pathway[25].

In bone defect scenarios, miR-375 overexpression in ADSCs augmented the osteogenic capacity of EVs through insulin-like growth factor binding protein-3 targeting, resulting in enhanced new bone formation and bone volume/total volume (BV/TV) in rat calvarial defects[26]. Beyond genetic modulation, pharmacological cargo loading represents another effective strategy. Curcumin-loaded EVs demonstrated augmented antioxidant and anti-apoptotic effects, leading to reduced cartilage degradation and pain in experimental OA models[27]. Surface engineering strategies, including Cys-Arg-Glu-Lys-Ala (CREKA) peptide conjugation[28] and biotin-avidin immobilization[29], have also been employed to enhance EV targeting precision and functional compatibility with implanted biomaterial scaffolds. These molecular engineering methods provide both flexibility and specificity, facilitating the design of EVs with customized therapeutic cargo to address complex bone regeneration challenges.

Biomaterial integration strategies for targeted and sustained EV delivery

While EVs exhibit intrinsic bioactivity, their rapid clearance and poor localization in vivo significantly limit therapeutic outcomes in bone repair. Consequently, integrating ADSC-EVs into biocompatible scaffolds or hydrogels has emerged as a crucial strategy to enhance EV retention kinetics, achieve sustained release profiles, and deliver mechanical and structural reinforcement to defect sites. For example, polydopamine-coated gelatin sponge scaffolds loaded with ADSC-EVs demonstrated significant enhancement in BV/TV, BMD, and osteogenic marker expression in a rat femoral defect model[30]. A bilayer decellularized extracellular matrix (ECM) hydrogel fabricated using three-dimensional bioprinting and loaded with ADSC-EVs supported concurrent cartilage and subchondral bone (SB) regeneration in a rat osteochondral defect model[31]. In another study, injectable gelatin-nanoparticle hydrogels enriched with miR-451a-containing EVs facilitated bone regeneration and M2 macrophage polarization in a rat calvarial defect model[32]. Moreover, scaffolds formulated with poly-lactic-co-glycolic acid and magnesium-gallic acid magnesium-organic frameworks, when combined with ADSC-EVs, augmented angiogenesis, osteogenesis, and anti-inflammatory responses in large bone defects[33]. These scaffold-based delivery systems not only optimize EV localization and activity durability but also function as osteoconductive platforms to support endogenous cell recruitment and tissue remodeling.

BONE FRACTURE

Epidemiology and clinical limitations

Bone tissue undergoes continuous remodeling, maintained by a balance between osteoblastic formation and osteoclastic resorption[34]. Factors such as aging, illness, or inadequate nutrition can disrupt this balance, impeding fracture healing. While younger individuals often experience efficient bone repair, these challenges are more pronounced in older populations or those with underlying health conditions[2]. Bone fractures represent a significant global health challenge, particularly as populations age. Clinically, fracture management is limited by delayed diagnosis, variable treatment protocols, and challenges in post-fracture rehabilitation. Addressing these limitations requires improved screening techniques, timely interventions, and personalized treatment strategies to reduce fracture risk and enhance recovery. Continued research is essential to develop more effective preventive measures and therapeutic approaches that can mitigate the growing impact of bone fractures worldwide.

Molecular basis of bone healing

Bone healing is a highly orchestrated process that involves a cascade of molecular events. During the initial inflammatory phase, immune cells release proinflammatory cytokines such as IL-1β, TNFα, and IL-6 to clear damaged tissue and recruit MSCs to the injury site[2]. This is followed by the angiogenic phase, during which vascular endothelial growth factor (VEGF) stimulates the proliferation of endothelial cells and the formation of new blood vessels to ensure adequate nutrient and oxygen supply[35]. Concurrently, during the osteogenic phase, key signaling pathways, including the Wnt/β-catenin, bone morphogenetic protein (BMP), and transforming growth factor-β pathways, drive the differentiation of MSCs into osteoblasts and promote the deposition of new bone matrix, with essential transcription factors such as RUNX2 and Osterix (OSX) upregulating osteogenesisrelated genes like collagen type I (COL I)[36]. Finally, the remodeling phase refines the newly formed bone, restoring its original architecture and mechanical strength[2,36]. Recent studies indicate that ADSC-EVs facilitate bone fracture healing through multiple mechanisms, including promotion of angiogenesis, osteogenic differentiation, and modulation of inflammatory responses. These preclinical strategies and their underlying pathways are summarized in Table 1, providing an overview of how ADSC-EVs contribute to bone fracture healing.

Table 1 Preclinical strategies of adipose-derived stem cell-derived extracellular vesicles in bone defect.

Pretreated cells/modified EVs	Strategy	In vivo model	Delivery route	In vivo effects	Ref.
Osteogenically induced ADSC-EVs	PLGA/PDA scaffold for sustained release	Mouse (4 mm defect)	Scaffold implantation	↑Bone volume, collagen, RUNX2/OCN expression; ↑BMSC osteogenesis	[21]
ADSC-EVs enriched with miR-375 (genetically modified ADSCs)	MiR-375 overexpression targeting IGFBP3	Rat (5 mm defect)	Hydrogel implantation	↑BV/TV, BMD, new bone	[26]
CREKA-modified rat ADSC-EVs	CREKA peptide for fibrin targeting	Rat femoral condyle (2.8 mm × 3 mm)	Local injection (40 μL)	↑BV/TV, Tb.Th, OCN, CD31; ↑M2 macrophages; ↓inflammation	[28]
Biotinylated ADSC-EVs	Avidin-biotin complex system on Bio-Ppy-Ti surface	Subcutaneous ectopic model	Immobilization on Ti (not injected)	↑ALP, COL I, OCN; ↑osseointegration	[29]
ADSC-EVs (unmodified)	PDA-coated gelatin sponge scaffold	Rat femoral condyle (3 mm defect)	Scaffold implantation	↑New bone, BMD, BV/TV, osteogenic markers; ↓Tb.Sp	[30]
ADSC-EVs (unmodified)	3D-bioprinted bilayer dECM hydrogel	Rat knee joint (4 mm defect)	Bilayer hydrogel implantation	↑Cartilage/subchondral bone repair, biomechanics; ↓IL-1β	[31]
ADSC-EVs (unmodified)	Injectable GNP hydrogel via miR-451a/MIF	Rat (5 mm defect)	Hydrogel injection	↑Bone healing, BV/TV, Tb.N/Th; ↑M2 polarization	[32]
ADSC-EVs (unmodified)	PLGA/Mg-GA MOF scaffold	Rat (6.5 mm defect)	Scaffold implantation	↑BV/TV, BMD, angiogenesis, osteogenesis; ↓inflammation	[33]
ADSC-EVs (unmodified)	None	STZ-induced T2DM rat femoral fracture	Local injection (600 μL, 200 μg/mL, every 3 days)	↑Mineralization, callus formation, BV/TV, and fracture healing	[40]

EVs: Extracellular vesicles; ADSC-EVs: Adipose-derived stem cell-derived extracellular vesicles; PLGA: Poly(lactic-co-glycolic acid); PDA: Polydopamine; RUNX2: Runt-related transcription factor 2; OCN: Osteocalcin; BMSC: Bone marrow mesenchymal stem cells; miR-375: MicroRNA-375; ADSC: Adipose-derived stem cell; IGFBP3: Insulin-like growth factor binding protein 3; BV/TV: Bone volume/total volume; BMD: Bone mineral density; CREKA: Cys-Arg-Glu-Lys-Ala; Tb.Th: Trabecular thickness; CD31: Cluster of differentiation 31; ALP: Alkaline phosphatase; COL I: Collagen type I; Ppy: Polypyrrole; Ti: Titanium; Tb.Sp: Trabecular separation; dECM: Decellularized extracellular matrix; IL-1β: Interleukin-1β; GNP: Gelatin-nanoparticle; miR-451a: MicroRNA-451a; MIF: Migration inhibitory factor; Tb.N: Trabecular number; Mg-GA MOF: Magnesium-gallic acid metal-organic framework; STZ: Streptozotocin; T2DM: Type 2 diabetes mellitus.

Mechanisms

Promotion of angiogenesis: Angiogenesis, the formation of new blood vessels, is essential for delivering nutrients and oxygen to fracture sites, thereby creating a regenerative microenvironment that supports osteogenesis[37]. ADSC-EVs enhance angiogenesis via VEGF and von Willebrand factor upregulation, as demonstrated in human umbilical vein endothelial cell tube formation assays. CREKA-modified EVs improve neovascularization, with cluster differentiation 31 (CD31)+ vessel density increasing by 2.5-fold in rat defect models[28]. These CREKA-EVs maintain the intrinsic proangiogenic properties of native ADSC-EVs while exhibiting superior retention in fibrin-rich bone defect areas due to their enhanced fibrin-binding affinity. In vitro studies have confirmed that CREKA-EVs significantly upregulate angiogenic genes (VEGF and von Willebrand factor) and promote capillary-like loop formation in human umbilical vein endothelial cells, thereby enhancing endothelial differentiation and vascular morphogenesis. Importantly, these angiogenic effects are partially attributed to miR-21-5p, which is one of the most abundant miRNAs in ADSC-EVs. This miRNA enhances endothelial progenitor cell viability, migration, invasion, and tube formation, ultimately promoting angiogenesis both in vitro and in vivo. Mechanistically, miR-21-5p targets NOTCH1, a key inhibitor of angiogenic signaling, thereby suppressing the NOTCH1/DLL4 axis while upregulating VEGFA expression. This signaling cascade activation boosts endothelial progenitor cell-mediated neovascularization and significantly contributes to bone regeneration in rat cranial defect models[38]. To maximize therapeutic potential, ADSC-EVs are increasingly being incorporated into bioactive materials. For instance, gelatin methacrylate hydrogels loaded with ADSC-EVs offer both mechanical support and sustained release of angiogenic cargo, thereby synergistically promoting endothelial proliferation and functional vascularization in vitro and in vivo[39].

Promotion of osteogenic differentiation: ADSC-EVs orchestrate osteogenic differentiation through dual modulation of osteoblasts and BMSCs, leveraging their bioactive cargo (e.g., miRNAs or proteins) to activate key signaling pathways and synergize with biomaterials. These EVs act as versatile mediators, directly enhancing osteoblast function and driving BMSC commitment to osteogenic lineages. Their therapeutic potential is amplified in engineered biomaterial systems, where controlled EV delivery sustains pro-osteogenic signaling while overcoming pathological microenvironments (e.g., oxidative stress or diabetic conditions). This section dissects the molecular and functional mechanisms underlying the osteogenic efficacy of ADSC-EVs across cellular targets and fracture microenvironment dynamics.

ADSC-EVs directly enhance osteoblast survival, differentiation, and function through multiple mechanisms. Experimental studies demonstrate that ADSC-EVs significantly promote the survival of MG63 osteoblast-like cells under stress conditions while simultaneously upregulating osteogenic markers such as OCN and alkaline phosphatase (ALP), thereby validating their critical role in bone tissue repair[29,40]. Beyond cellular-level effects, the therapeutic potency of ADSC-EVs is driven by their molecular cargo. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses have revealed that miRNAs within ADSC-EVs are enriched in pathways essential for skeletal system development, including autophagy, mitogen-activated protein kinase (MAPK) signaling, and Rap1 signaling[41]. A key example is miRNA-375, which enhances osteogenesis by suppressing insulin-like growth factor binding protein-3, thereby activating the pro-osteogenic insulin-like growth factor pathway[26]. ADSC-EVs further exhibit antioxidative properties, mitigating oxidative stress in osteoblasts by delivering antioxidant enzymes and miRNAs. This reduces intracellular reactive oxygen species (ROS) levels, preserving osteogenic capacity in pathological microenvironments[42]. These findings collectively underscore that ADSC-EVs directly enhance osteoblast survival, differentiation, and function, demonstrated by increased expression of OCN and ALP, as well as ROS reduction in stressed osteoblasts.

Extensive studies have demonstrated the intrinsic osteogenic potential of ADSC-EVs by providing both molecular and functional evidence. Specifically, ADSC-EVs directly enhance osteogenic differentiation in BMSCs by upregulating key osteogenic genes, including RUNX2, OSX, OCN, osteopontin (OPN), and COL1A1, thereby promoting calcium nodule formation and bone matrix mineralization[30]. In vitro studies with BMSCs further corroborate this effect, showing elevated ALP activity and upregulated expression of osteogenic genes, including RUNX2 and OCN[28]. This osteogenic capacity of ADSC-EVs is further amplified when delivered via engineered biomaterials. For example, ADSC-EVs, when applied with gelatin sponge/polydopamine constructs, significantly upregulate osteogenic gene expression in BMSCs and promote new bone formation in a rat femoral defect model, as confirmed by micro-computed tomography and histology[30]. This enhanced efficacy extends to various delivery systems. When combined with mineral-doped poly-L-lactic acid scaffolds, ADSC-EVs enhanced BMSC osteogenic differentiation and markedly elevated expression of osteogenic markers, including COL I, OPN, OCN and bone sialoprotein, underscoring their adaptability in enhancing bone regeneration[43]. ADSC-EVs also enhance bone repair when paired with surface-modified implants. For instance, when loaded onto graphene-oxide-modified titanium, they stimulate BMSC adhesion, proliferation, and osteogenesis (COL I, OPN, and OCN upregulation), thereby enhancing osseointegration[44]. These findings collectively demonstrate that ADSC-EVs promote osteogenic differentiation of BMSCs through upregulation of key markers, including RUNX2, OSX, and OCN, with their intrinsic bioactivity enabling them to act as a universal enhancer across diverse delivery platforms, thereby supporting tailored strategies for critical-sized bone defects, diabetic complications, and implant osseointegration.

Regulation of the inflammatory response: The inflammatory response following bone fracture plays a pivotal role in determining healing outcomes, requiring precise regulation to balance tissue repair and pathological damage[36]. ADSC-EVs have emerged as potent immunomodulators capable of reshaping the inflammatory microenvironment, particularly in compromised conditions such as diabetic fractures. Under hyperglycemic stress, persistent activation of the nuclear factor kappa B (NF-κB) pathway drives excessive secretion of proinflammatory cytokines, including TNF-α and IL-1β, which disrupt normal bone regeneration. To counteract this dysregulation, ADSC-EVs deliver bioactive cargo, including miRNAs and proteins, that suppress NF-κB signaling and attenuate chronic inflammation, thereby restoring the regenerative capacity of the fracture site[40]. A specific example of this regulation is the delivery of miR-451a by ADSC-EVs, which targets and downregulates macrophage migration inhibitory factor (MIF); as a result, the expression of proinflammatory cytokines such as TNF-α and IL-6 is reduced, while the secretion of anti-inflammatory cytokine IL-10 is enhanced[32]. This shift in the cytokine milieu creates a more favorable environment for bone healing. Central to this immunomodulatory effect is the reprogramming of macrophage polarization: ADSC-EVs promote a phenotypic switch from pro-inflammatory M1 macrophages to anti-inflammatory M2 macrophages[32], which is critical for resolving inflammation and initiating subsequent tissue remodeling processes[45]. Mechanistically, suppression of MIF by miR451a delivered via ADSC-EVs may alleviate its inhibitory influence on several downstream signaling pathways involved in macrophage polarization. In particular, MIF downregulation has been shown in other cellular contexts to permit IL-4/IL-13-induced signal transducer and activator of transcription 6 phosphorylation and nuclear translocation, thereby promoting the transcription of M2-associated genes such as arginase 1 and IL-10, which reinforce the M2 phenotype and contribute to inflammation resolution and bone regeneration[46]. In addition, reduced MIF expression may facilitate activation of the phosphatidylinositol 3-kinase/protein kinase B signaling pathway, which not only promotes M2 macrophage polarization and dampens M1-driven inflammation, but also directly supports bone regeneration[47,48]. Notably, M2 macrophage-derived exosomes stimulated phosphatidylinositol 3-kinase/protein kinase B phosphorylation and accelerated fracture healing in a diabetic mouse model[49]. Although direct experimental evidence linking ADSC-EV-derived miR-451a to these downstream pathways remains limited, these interactions are biologically plausible and supported by mechanistic studies in related inflammatory and regenerative models. Moreover, ADSC-EVs contribute to the alleviation of oxidative stress, which often accompanies inflammation, by enhancing the activity of the nuclear factor-E2-related factor 2 (Nrf2)/heme oxygenase-1 (HO-1) pathway[50]. This antioxidant signaling cascade facilitates the scavenging of ROS and suppresses lipopolysaccharide-induced inflammatory cascades, further reinforcing the anti-inflammatory and pro-regenerative effects of ADSC-EVs[50].

In summary, ADSC-EVs orchestrate a multifaceted immunomodulatory response by suppressing NF-κB-driven inflammation, promoting M2 macrophage polarization via the miR-451a/MIF axis, and enhancing antioxidant defenses through Nrf2/HO-1 activation. These synergistic mechanisms collectively resolve pathological inflammation and establish a pro-regenerative microenvironment, positioning ADSC-EVs as a promising therapeutic strategy for bone repair in diabetic complications and implant integration challenges.

OA

Epidemiology and clinical limitations

OA is a widespread chronic degenerative joint disorder affecting approximately 528 million people globally, as reported by the World Health Organization. Its increasing prevalence is driven by aging populations, rising obesity rates, and a higher incidence of joint injuries[3]. Current treatment approaches primarily involve physical therapy, pharmacological management, and surgical interventions, all of which aim to alleviate symptoms and slow disease progression. However, none offer a definitive cure or restore lost articular cartilage[51-54]. In OA-affected joints, elevated oxidative stress disrupts chondrocyte homeostasis, leading to accelerated apoptosis of chondrocytes and degradation of the ECM[55,56]. This cascade of events further deteriorates joint integrity and function, exacerbating disability and reducing overall quality of life. Therefore, there is a significant unmet need for innovative therapeutic strategies that address the underlying pathophysiological mechanisms.

Pathophysiological mechanisms of OA

OA manifests pathologically as progressive SB loss (accelerated by aging), osteophyte proliferation at joint margins, subchondral sclerosis, and chronic synovitis[57]. These structural and inflammatory alterations collectively drive the disease’s pathophysiology through intertwined inflammatory, metabolic, and biomechanical mechanisms, wherein persistent synovial inflammation propagates cartilage degradation while dysregulating bone remodeling[3,58,59]. Mechanical overloading from joint injuries or obesity initiates the release of catabolic enzymes, such as MMPs, which degrade the cartilage ECM[60,61]. Concurrently, chronic low-grade inflammation driven by cytokines such as IL-1β and TNF-α activates catabolic pathways, increasing MMPs and a disintegrin and metalloproteinase with thrombospondin motifs 5 expression, which degrade cartilage ECM[56]. In OA-affected joints, increased oxidative stress disrupts chondrocyte homeostasis, leading to enhanced apoptosis and impaired reparative function[62]. Additionally, age-related chondrocyte senescence contributes to a pro-inflammatory secretory phenotype, intensifying local tissue degradation[63-65]. Notably, these inflammatory and degenerative processes extend into the osteochondral junction, where vascular invasion from SB into articular cartilage enhances crosstalk between bone and cartilage, thereby facilitating increased transport of cytokines and amplifying tissue damage. This disrupts SB architecture, perturbs the balance of bone remodeling, and impairs the functional coupling of bone resorption and formation across the osteochondral unit, as evidenced by coordinated inflammatory responses and oxidative stress in both cartilage and SB compartments under pro-inflammatory stimuli like lipopolysaccharide[66].

However, current pharmacological therapies, including bisphosphonates and anti-receptor activator of NF-κB ligand (anti-RANKL) antibodies, primarily target bone resorption but fail to address the underlying regenerative deficits. This gap underscores the need for approaches that simultaneously target inflammation, cellular senescence, and impaired bone remodeling. ADSC-EVs, with their multifactorial regulatory capacity, have emerged as promising candidates. By targeting core pathological features such as MMP overexpression, chondrocyte apoptosis, senescence-driven inflammation, and SB imbalance, ADSC-EVs offer the potential for disease-modifying OA therapy. The subsequent sections elaborate on the mechanistic basis of ADSC-EV-mediated bone repair, including their roles in SB regeneration, osteoblast rejuvenation, and immune microenvironment reprogramming. To synthesize these insights, these preclinical strategies and their underlying pathways are summarized in Table 2, providing an overview of how ADSC-EVs contribute to OA.

Table 2 Preclinical Strategies of adipose-derived stem cell-derived extracellular vesicles in osteoarthritis.

Pretreated cells/modified EVs	Strategy	In vivo model	Delivery route	Effects in vivo	Ref.
ADSCs stimulated with ↓TNF-α & IFN-γ to ADSC-EVs	MiR-27b-3p enriched EVs + SF scaffold	Rabbit TMJ condylar defect	Scaffold implantation	↑BMSC proliferation, osteo-/chondrogenesis; ↑M2 macrophages; ↓inflammation; ↑bone/cartilage repair	[22]
PEMF-stimulated rat ADSC-EVs	PEMF stimulation (15-75 Hz) to enhance EVs	ACLT rat OA model	IA injection (40 μL, weekly × 4)	↑Anabolic cartilage markers; ↓MMP-13, IL-1β; ↑histology scores; ↓inflammation	[23]
ADSC-EVs enriched with miR-486-5p (miR-486-5p mimic-transfected ADSCs)	Engineered EVs delivering miR-486-5p	DMM rat OA model	IA injection (20 μL, weekly × 10)	↓Chondrocyte apoptosis, MMP-13, inflammation; ↑COL II, cartilage matrix; ↓OARSI score	[24]
Curcumin-loaded ADSC-EVs	Drug-loading to enhance antioxidant & anti-apoptosis	ACLT mouse model	IA injection (10 μL, biweekly × 4)	↓Cartilage degeneration, oxidative stress, apoptosis, pain; ↑ECM components (COL II, aggrecan)	[27]
ADSC-EVs enriched with miR-376c-3p	MiR-376c-3p delivery targeting WNT/β-catenin	MIA rat OA model	IA injection (100 μg EVs, single dose)	↓Cartilage degradation, fibrosis, inflammatory cytokines; ↑COL II; WNT signaling inhibition	[90]
ADSC-EVs (unmodified)	Native EVs isolated by TFF	MIA-induced rat OA model	IA injection (30 μL, weekly × 3 or 6)	↓Mankin score, cartilage erosion, inflammatory markers; ↑cartilage matrix protection	[91]
ADSC-EVs (unmodified)	Native EVs isolated by TFF	DMM mouse OA model	IA injection (6 μL, weekly × 6)	↓OARSI score, synovial inflammation, M1 macrophages; ↑COL II, chondrocyte proliferation/migration	[91]

EVs: Extracellular vesicles; ADSC: Adipose-derived stem cell; TNF-α: Tumor necrosis factor-alpha; IFN-γ: Interferon-gamma; ADSC-EVs: Adipose-derived stem cell-derived extracellular vesicles; miR-27b-3p: MicroRNA-27b-3p; SF: Silk fibroin; TMJ: Temporomandibular joint; BMSC: Bone marrow mesenchymal stem cells; PEMF: Pulsed electromagnetic field; ACLT: Anterior cruciate ligament transection; OA: Osteoarthritis; IA: Intra-articular; MMP-13: Matrix metalloproteinase-13; IL-1β: Interleukin-1β; miR-486-5p: MicroRNA-486-5p; DMM: Destabilization of the medial meniscus; COL II: Collagen type II; OARSI: Osteoarthritis Research Society International; ECM: Extracellular matrix; miR-376c-3p: MicroRNA-376c-3p; MIA: Monosodium iodoacetate; TFF: Tangential flow filtration.

Mechanisms

Structural restoration and bone microarchitecture repair: SB remodeling dysfunction serves as a key pathological driver of articular cartilage degeneration, characterized by distinct stages: Initial bone mass reduction due to enhanced osteoclastic activity, subsequent subchondral sclerosis in advanced phases, and osteophyte formation resulting from aberrant bone anabolism[67]. ADSC-EVs exhibit potent efficacy in restoring SB microarchitecture in OA. In a transgenic pig model of cartilage-bone injury, ADSC-EVs restored SB plate (SBP) thickness to physiological levels (0.72 ± 0.18 mm vs healthy 0.75 ± 0.21 mm) and normalized BV fraction (BV/TV: 18.3% ± 5.2%) by harmonizing osteoclast-osteoblast activity, as demonstrated by reduced tartrate-resistant acid phosphatase positive (TRAP+)/OCN+ cells. Micro-computed tomography confirmed stabilization of SBP and trabecular microarchitecture, with 50% defect filling by fibrocartilage and no neoplastic changes[68]. Similarly, in rat osteochondral defects, ADSC-EVs delivered through collagen hydrogel significantly enhanced trabecular bone formation, increasing BV/TV and BMD while reducing trabecular separation (Tb.Sp). By 8 weeks, EV-treated groups achieved superior osseous healing and optimized trabecular microarchitecture beneath cartilage defects[69]. Notably, ADSC-EVs demonstrate superior osteogenic potential compared to BMSC/synovial MSC-EVs. In ectopic osteogenesis models, ADSC-EVs induced the formation of Haversian-like structures, marked calcium mineralization (Alizarin red staining), and elevated COL I expression[70]. Hypoxia pretreatment further amplified these effects in spinal OA mice, restoring trabecular integrity (↑BV/TV, ↓synovial MSC) and normalizing H-type vasculature, which synergized with balanced bone remodeling (↓TRAP+/OCN+ cells) to alleviate pain[71].

Anti-senescence and osteogenic signaling: ADSC-EVs ameliorate osteoblast senescence and metabolic dysfunction in OA through multifaceted signaling pathways. They attenuate inflammatory responses and oxidative stress in osteoarthritic osteoblasts, reversing senescence hallmarks and correcting metabolic perturbations[72]. Bioinformatics analyses reveal that IL-1β-primed ADSC-EVs modulate key signaling pathways (Wnt, transforming growth factor-β, VEGF, Hippo) associated with bone regeneration[73]. Mechanistically, MAPK pathway inhibition enhances BMP-2 signaling, leading to RUNX2/OSX phosphorylation and upregulation of osteogenic genes (ALP, OCN), thereby accelerating bone matrix mineralization[74]. Preconditioning strategies further augment efficacy: PEMF optimizes ADSC-EV functionality, with 75 Hz frequency demonstrating superior efficacy in preserving subchondral microarchitecture (↑trabecular connectivity, ↓sclerosis) and mitigating synovial hyperplasia in OA models[23]. These mechanistic synergies are supported by in vitro evidence showing ADSC-EVs significantly enhance BMSC viability (> 24 hours/48 hours), migration (1.5-fold vs other EVs), and osteogenic marker expression (Alpl, COL alpha 1, RUNX2/OCN), collectively promoting bone matrix synthesis[70].

ADSC-EVs promote SB regeneration in OA via structural restoration of SBP thickness, BV/TV, and trabecular microarchitecture[68,69,71], coupled with cellular rejuvenation by counteracting osteoblast senescence[72]. These effects are augmented by molecular reprogramming through activation of osteogenic pathways (BMP-2/RUNX2/OSX)[70,74] and potentiated by preconditioning strategies (hypoxia, PEMF, IL-1β) that enhance bioactivity for bone-anabolic outcomes[23,71,73]. Notably, ADSC-EVs also exert profound effects on OA-related bone regeneration through modulation of the immune microenvironment, including regulation of macrophage polarization, balancing pro-/anti-inflammatory cytokines and inhibiting osteoclastogenesis via immune cell crosstalk. This critical mechanism will be elaborated in the subsequent “IMMUNE MICROENVIRONMENT REGULATION” section, integrating with related findings in OP to highlight shared regulatory principles.

OP

Epidemiology and clinical limitations

OP is a systemic skeletal disorder characterized by a reduction in and a deterioration of bone microarchitecture, which collectively lead to enhanced bone fragility and an elevated risk of fractures[75]. A study reported that in 2010, an estimated 5.5 million men in the European Union were affected by OP, with the prevalence among women being approximately four times that of men[76]. Current clinical therapies, such as bisphosphonates, denosumab, and teriparatide, primarily target either bone resorption or formation. Although effective in reducing fracture risk, these treatments have limitations, including adverse effects, poor long-term adherence, and rebound bone loss upon discontinuation[77]. EV-based therapies may overcome current treatment limitations by restoring the delicate balance of bone remodeling, offering a novel and holistic approach to OP management.

Molecular mechanisms in OP

OP often stems from an imbalance between bone formation and resorption, regulated by osteoblasts (derived from BMSCs) and osteoclasts (originating from bone marrow-derived macrophages)[34]. Factors such as estrogen deficiency, aging, disuse, medication, and malnutrition contribute to dysregulation in osteoblast and osteoclast activity, as well as osteoblast and adipocyte differentiation, thereby accelerating OP[78]. Traditionally, OP has been attributed primarily to endocrine factors, with estrogen deficiency leading to secondary hyperparathyroidism and, when combined with insufficient vitamin D and calcium intake, serving as the key etiological determinants of the disease[79]. At the molecular level, estrogen deficiency upregulates RANKL and pro-inflammatory cytokines, including IL-1, IL-6, and TNF-α, leading to increased osteoclastogenesis, while simultaneously impairing osteoblast differentiation via downregulation of the Wnt/β-catenin pathway[80]. Furthermore, excessive ROS contributes to oxidative stress, exacerbating osteoblast apoptosis and stimulating osteoclast activity[81]. To elucidate the relevance of how ADSC-EVs address these pathological mechanisms, these preclinical strategies and their underlying pathways are summarized in Table 3, providing an overview of how ADSC-EVs contribute to OP.

Table 3 Preclinical strategies of adipose-derived stem cell-derived extracellular vesicles in osteoporosis.

Pretreated cells/modified EVs	Mechanism	Animal model	Delivery	In vivo effects	Ref.
MiR-146a-overexpressing ADSC-EVs	MiR-146a delivery to suppress inflammasome	STZ-induced diabetic rat	Tail vein, every 2 days	↑Bone density; ↓inflammation; best effect with miR-146a	[25]
ADSC-EVs (unmodified)	Activates Nrf2/HO-1	Dexamethasone rat	Tail vein, 3 ×/week	↑BMD, BV/TV, Tb.Th, Tb.N; ↓apoptosis; ↑osteogenesis	[83]
ADSC-EVs enriched in OPG, miR-21-5p	Inhibit osteoclastogenesis; promote MSC migration	OVX mouse	IV, 3 ×/week	↑BMD, BV/TV, Tb.Th, Tb.N; ↓osteoclasts; ↑MSC migration	[85]
WSTLZT-treated ADSC-EVs (miR-122-5p)	Regulate bone-fat balance via SPRY2/MAPK	OVX mouse	Tail vein, weekly	↑BMD, BV/TV, Tb.Th; ↓Tb.Sp, marrow adipose; ↑osteogenesis	[94]
ADSC-EVs (unmodified)	Suppress NLRP3 inflammasome	STZ-induced diabetic rat	Tail vein, every 2 days	↑Bone mineral density; ↓IL-1β, IL-18; ↓osteoclasts	[95]

EVs: Extracellular vesicles; miR-146a: MicroRNA-146a; ADSC-EVs: Adipose-derived stem cell-derived extracellular vesicles; STZ: Streptozotocin; Nrf2: Nuclear factor-E2-related factor 2; HO-1: Heme oxygenase-1; BMD: Bone mineral density; BV/TV: Bone volume/total volume; Tb.Th: Trabecular thickness; Tb.N: Trabecular number; OPG: Osteoprotegerin; miR-21-5p: MicroRNA-21-5p; MSC: Mesenchymal stem cell; OVX: Ovariectomized; IV: Intravenous; WSTLZT: Wen-Shen-Tong-Luo-Zhi-Tong; miR-122-5p: MicroRNA-122-5p; SPRY2: Sprouty homolog 2; MAPK: Mitogen-activated protein kinase; Tb.Sp: Trabecular separation; NLRP3: NOD-like receptor pyrin domain-containing protein 3; IL-1β: Interleukin-1β; IL-18: Interleukin-18.

IMMUNE MICROENVIRONMENT REGULATION

ADSC-EVs demonstrate multifaceted therapeutic efficacy across OA and OP by modulating key pathological processes, including inflammation, oxidative stress, apoptosis, and bone remodeling pathways. These vesicles are enriched with functional non-coding RNAs, such as miRNAs and long non-coding RNAs (lncRNAs), which mediate critical regulatory functions in immune modulation and cellular protection. For instance, lncRNA KCNQ1OT1 within ADSC-EVs attenuates apoptotic processes by downregulating pro-apoptotic factors (e.g., Bax, cleaved caspase-3, and caspase-9) and upregulating anti-apoptotic proteins such as Bcl-2[82], thereby enhancing viability and survival in chondrocytes and osteoblasts. Concurrently, ADSC-EVs activate the Nrf2/HO-1 antioxidant pathway, conferring cytoprotection against oxidative damage and facilitating osteogenic differentiation[83]. Immunoregulatory effects are exerted through macrophage polarization shifts from a pro-inflammatory M1 phenotype to an anti-inflammatory M2 phenotype, characterized by reduced inducible nitric oxide synthase and elevated CD163 expression[24]. This reprogramming is accompanied by decreased secretion of pro-inflammatory cytokines (e.g., IL-6, TNF-α), fostering an immune milieu conducive to bone repair[24]. ADSC-EVs further modulate bone metabolic homeostasis by rebalancing the RANKL/OPG signaling axis. Under hypoxic and serum-deprived conditions, osteocytes treated with ADSC-EVs exhibit suppressed RANKL expression at both transcriptional and translational levels, accompanied by elevated OPG mRNA levels. This biochemical shift results in a decreased RANKL/OPG ratio[84]. The functional consequence manifests as attenuated osteoclast differentiation, as evidenced by diminished TRAP+ multinucleated cell formation when bone marrow monocytes are exposed to conditioned medium from ADSC-EV-treated osteocytes[84]. Additionally, ADSC-EVs inhibit osteoclast differentiation via delivery of miRNAs, including miR-21-5p and let-7b-5p. MiR-21-5p downregulates Acvr2a to block osteoclastogenesis, while let-7b-5p markedly suppresses osteoclastogenic genes (e.g., Acp5, Ctsk, Mmp9, Nfatc1), shifting bone metabolic balance toward anabolic outcomes[85]. Collectively, these mechanisms enable ADSC-EVs to protect bone cells from inflammation-induced damage while directly augmenting regenerative capacity. These actions collectively highlight the critical role of ADSC-EVs in orchestrating immune microenvironment remodeling to drive bone regeneration.

In OA, synovial inflammation is initiated when synovial fibroblasts and macrophages recognize cartilage-derived damage-associated molecular patterns via pattern recognition receptors, activating innate immune responses, stimulating chemokine and cytokine release, and upregulating MMPs in chondrocytes, thereby establishing a self-perpetuating cycle of ECM degradation and chronic inflammation[86,87]. Resident chondrocytes, synovial fibroblasts, and infiltrating immune cells (e.g., monocytes, macrophages), sustain synovitis through these pathological interactions[88]. ADSC-EVs mitigate this inflammatory microenvironment by modulating both molecular and cellular components within the osteoarthritic joint. Their miR-486-5p cargo downregulates endoplasmic reticulum (ER) stress markers CCAAT/enhancer-binding protein homologous protein and glucose-regulated protein 78 in chondrocytes, thereby attenuating ER stress-induced apoptosis and suppressing inflammatory signaling cascades[24]. Simultaneously, ADSC-EVs reprogram synovial macrophages toward an M2 phenotype, reducing M1 markers like inducible nitric oxide synthase and increasing M2 markers like CD163. These changes are associated with decreased IL-6 and TNF-α, and increased IL-10 Levels, as validated in OA synovial fluid-stimulated models[24]. Beyond these effects, ADSC-EVs further contribute to bone regeneration in OA by modulating crosstalk between inflammatory resolution and osteogenic processes. Their cargo, including miRNAs such as miR-127-5p and miR-376c-3p, promotes chondrogenesis and osteoblast activity[89,90]. Specifically, miR-376c-3p directly inhibits the Wnt/β-catenin pathway, a critical regulator of bone remodeling. In vitro validation through gain- and loss-of-function experiments demonstrated that miR-376c-3p mimic transfection reduced β-catenin protein levels (western blot) and suppressed transcriptional activity of Wnt target genes (e.g., c-Myc, cyclin D1) in luciferase reporter assays; conversely, miR-376c-3p inhibition restored β-catenin expression and pathway activity. This regulatory mechanism mediates miR-376c-3p-induced upregulation of chondrogenic markers (e.g., COL II, SOX9) and osteogenic transcription factors (e.g., RUNX2), while suppressing inhibitors of bone formation[90]. Additionally, by attenuating chronic inflammation and ECM degradation, ADSC-EVs establish a conducive microenvironment for SB remodeling - promoting equilibrium between osteoclast-mediated resorption and osteoblast-driven formation[70,91]. This coordinated modulation of inflammatory pathways and osteogenic programs highlights the therapeutic potential of ADSC-EVs to not only mitigate synovitis but also promote functional bone regeneration in osteoarthritic joints.

Effective OP management requires precise regulation of osteoblast-osteoclast activity balance and osteogenesis-adipogenesis coupling[92]. ADSC-EVs demonstrate therapeutic potential in glucocorticoid-induced OP by mitigating osteoblast apoptosis and restoring osteogenic function. In such models, they alleviate dexamethasone-induced oxidative stress in osteoblasts, reducing intracellular ROS levels, restoring superoxide dismutase activity, and lowering malondialdehyde content. Anti-apoptotic effects are mediated through Bcl-2 upregulation and Bax/cleaved caspase-3 downregulation. ADSC-EVs also restore the osteogenic capacity of MC3T3-E1 cells, enhancing ALP activity, mineralization (ARS staining), and expression of osteogenic markers (Runx2, Bmp2, and Opn). In vivo, 100 μg ADSC-EVs significantly improved BMD, increased BV/TV, trabecular thickness, trabecular number, and decreased Tb.Sp in the distal femur of treated rats. These effects are attributed to Nrf2/HO-1 axis activation, as Nrf2 knockdown partially abolished these protective outcomes[83]. ADSC-EVs enriched with lncRNA KCNQ1OT1 play a pivotal role in immune microenvironment modulation to support bone health, particularly in counteracting inflammation-induced damage. In TNF-α-stimulated primary osteoblasts, KCNQ1OT1-containing EVs exhibit superior inhibition of cytotoxicity and apoptosis compared to unmodified ADSC-EVs, achieved through competitive binding to miR-141-5p - a pro-inflammatory/pro-apoptotic miRNA upregulated by TNF-α in a dose-dependent manner. This interaction reduces Bax and cleaved caspase-3 expression while enhancing cell viability, thereby protecting osteoblasts from inflammatory impairment[82]. This interaction reduces the expression of pro-apoptotic proteins Bax and cleaved caspase-3, while enhancing cell viability, thereby protecting osteoblasts from inflammation-mediated impairment[82].

Concurrently, in OA contexts, ADSC-EVs exert immunomodulatory effects via miRNA cargo (miR-24-3p, miR-222-3p, and miR-193b-3p), promoting synovial macrophage polarization from M1 to M2 phenotype. This shift correlates with reduced IL-6/TNF-α secretion and elevated anti-inflammatory mediators, collectively fostering an immune milieu that supports bone regeneration and limits excessive bone resorption. These mechanisms collectively highlight how ADSC-EVs, through diverse RNA cargo, coordinate immune regulation and cytoprotection to maintain bone homeostasis across OA and OP[93]. Notably, although ADSC-EVs demonstrate standalone efficacy, complementary strategies like traditional Chinese medicine formulations (e.g., Wen-Shen-Tong-Luo-Zhi-Tong decoction) may further enhance their therapeutic impact. Such formulations modulate adipocyte-derived exosomes to deliver miR-122-5p, which targets sprouty homolog 2 to activate the MAPK pathway, thereby promoting BMSC osteogenic differentiation (↑RUNX2, OSX) and inhibiting adipogenic differentiation (↓CCAAT/enhancer binding protein alpha, peroxisome proliferator-activated receptor gamma 2). In ovariectomized mice, this approach improved bone microarchitecture (↑BMD, BV/TV, trabecular thickness, trabecular number; ↓Tb.Sp) and reduced marrow adipose tissue, with MAPK/extracellular signal-regulated kinase signaling mediating these effects[94]. Collectively, these findings underscore the potential of ADSC-EVs, particularly when enhanced through genetic or physical preconditioning, to prevent bone loss and restore homeostasis via antioxidant, anti-apoptotic, and pro-osteogenic mechanisms.

DISCUSSION

Building on the clinical limitations of current therapies and the promise of ADSC-EVs outlined in the introduction, we now synthesize mechanistic insights and preclinical evidence to map their translational trajectory in bone regeneration. We explored the therapeutic potential of ADSC-EVs in treating bone-related diseases, including fractures, OA, and OP. ADSC-EVs present a compelling, cell-free alternative to conventional therapies due to their regenerative and immunomodulatory properties. However, advancing therapeutic options for bone-related diseases requires a comprehensive understanding of EV components, particularly those unique to individual sources. At present, a standardized protocol for the characterization and preparation of ADSC-EVs is lacking. Establishing such a protocol, potentially using advanced detection techniques such as gene sequencing, would help ensure consistency and reproducibility across studies.

In summary, while ADSC-EVs demonstrate significant therapeutic potential for bone fractures, OA, and OP, each condition presents unique challenges. For fracture repair, the synergy between EVs and biomaterial scaffolds is promising, yet optimal delivery methods require further optimization. In OA, the modulation of inflammatory responses by EVs holds potential, although long-term cartilage regeneration remains to be fully validated. Finally, in OP, the restoration of bone homeostasis via EV-mediated pathways is encouraging, but standardization in EV production and dosage remains a critical barrier to clinical translation. For fracture repair, ADSC-EVs enhance both angiogenesis and osteogenic differentiation, especially when combined with bioactive scaffolds, supporting tissue regeneration at the fracture site. Their involvement in key pathways, such as Wnt/β-catenin, facilitates effective bone matrix formation and repair.

In OA, ADSC-EVs exhibit strong anti-inflammatory effects by promoting M2 macrophage polarization and reducing chondrocyte apoptosis. They help maintain cartilage integrity by regulating oxidative stress and upregulating chondrogenic markers (e.g., Sox9 and Col II), thus supporting cartilage stability. Integrating ADSC-EVs with bioactive materials—either through encapsulation techniques or scaffold-based delivery systems—may enhance their stability and bioavailability, offering a synergistic approach that could amplify therapeutic outcomes.

For OP, ADSC-EVs effectively promote osteogenesis and limit bone resorption. They enhance osteoblast differentiation through pathways such as Nrf2/HO-1 and inhibit osteoclastogenesis by downregulating RANKL. Additionally, ADSC-EVs modulate BMSC differentiation and reduce adipogenesis via miRNAs, creating a favorable environment for bone formation. To enhance the osteogenic and therapeutic effects of ADSC-EVs, strategies such as preconditioning ADSCs with specific chemical or physical stimuli, genetic modification, and the targeted selection of optimal ADSC clusters have shown promise. Amplifying specific osteogenic miRNAs or using direct genetic manipulation may increase therapeutic efficacy.

Despite the promising preclinical results, several challenges remain in the clinical translation of ADSC-EVs. First, the safety and long-term effects of ADSC-EVs need to be thoroughly evaluated, particularly in terms of potential immune responses and tumorigenic risks. Second, the optimal dosage and delivery methods for ADSC-EVs in different bone-related diseases require further investigation. Third, the mechanisms of action of ADSC-EVs, particularly the specific roles of their cargo (e.g., miRNAs and proteins), need to be elucidated to enhance their therapeutic efficacy. Additionally, the generalizability of current preclinical findings requires careful consideration. Many studies, including those involving rat skull defect models, utilize relatively simple two-dimensional bone defects that do not fully replicate the complex three-dimensional architecture and biomechanical environment of human bone injuries. This discrepancy may limit the direct translation of therapeutic outcomes observed in animal models to clinical practice. Future investigations using large animal models or more clinically relevant defect geometries are necessary to validate the efficacy of ADSC-EVs in real-world settings. Finally, establishing standardized protocols (e.g., MISEV2023 guidelines[13]) for EV characterization is critical. The storage of ADSC-EVs is also crucial for ensuring consistency and reproducibility in clinical applications.

These findings collectively establish a translational framework that bridges the molecular mechanisms outlined earlier with their potential clinical applications. Given the growing demand for effective cell-free therapies in bone and joint diseases, the immunomodulatory and regenerative functions of ADSC-EVs position them as a promising therapeutic strategy. However, to realize this potential, ongoing basic research must be closely aligned with translational efforts, thereby enabling a progressive transition from laboratory discoveries to real-world clinical interventions.

CONCLUSION

In conclusion, ADSC-EVs show substantial promise across various bone-related conditions by modulating inflammation, enhancing bone regeneration, and reducing bone resorption. While ADSC-EVs universally target inflammation and oxidative stress, their disease-specific mechanisms vary: In fractures, they primarily enhance angiogenesis via VEGF; in OA, they suppress ER stress via miR-486-5p; and in OP, they restore RANKL/OPG balance. Future studies should prioritize comparisons of EV cargo across these contexts. The use of a multilayered strategy that combines biological enhancement with materials science, such as optimizing miRNA profiles and integrating ADSC-EVs with bioactive materials, may further establish ADSC-EVs as viable, effective interventions in bone regeneration and repair.