Research progress on exosomes from different sources in osteoarthritis and cartilage injury

Abstract

Exosomes are nanosized membrane vesicles secreted by cells that have garnered considerable attention in the biomedical field in recent years. Osteoarthritis (OA) is a common disabling joint disease with limited therapeutic options in clinical practice. Recent studies have shown that exosomes, as crucial mediators of intercellular communication, have proven effective in treating osteoarthritis and cartilage injury in vitro and in animal models. With ongoing research, exosomes have demonstrated tremendous potential in treating and even reversing osteoarthritis and cartilage injury. They may become an emerging therapeutic strategy in the future. This review primarily discusses the composition of exosomes and explores the potential mechanisms by which exosomes from different sources influence osteoarthritis and cartilage injury in the context of ongoing advances in exosome research. Furthermore, the review highlights the progress in exosome engineering for osteoarthritis Treatment to enhance the therapeutic efficacy of exosomes.

Background

The pathogenesis of osteoarthritis (OA) involves a series of interconnected biological processes, including cartilage injury, subchondral bone remodeling (subchondral sclerosis and osteophyte formation), synovial inflammation, meniscal degeneration, and thickening of the joint capsule and ligaments, with cartilage injury being the key factor in the progression of OA [1]. (Fig. 1) Risk factors for OA include age, sex, genetic susceptibility, obesity, inflammation, and excessive mechanical load, all of which increase the likelihood of OA onset and progression [2]. Given the combined effects of global population aging, the obesity epidemic, and increasing joint injuries, OA is expected to become more prevalent [3]. Cartilage, as a special connective tissue, lacks blood vessels and lymphatics; nutrients diffuse from the perichondrium into the extracellular matrix (ECM), nourishing chondrocytes. Additionally, chondrocytes have limited regenerative capacity and a long repair cycle, making cartilage repair a significant challenge. Currently, OA treatment faces various challenges. Traditional non-surgical treatments, such as nonsteroidal anti-inflammatory drugs (NSAIDs), opioids, topical analgesics, corticosteroid injections, and hyaluronic acid injections, may alleviate symptoms to some extent but fail to regenerate cartilage or reverse disease progression [4]. Although surgical joint replacement can improve function and quality of life in the long term, instability and infection are common limiting factors, often requiring multiple revision surgeries. Moreover, the use of exogenous stem cells for tissue repair also has limitations, such as immune rejection, ethical concerns, microvascular occlusion, and tumorigenic risks [5–7]. However, exosomes offer several advantages, including non-toxicity, low immunogenicity, and easy storage potential, making them a promising new therapeutic alternative. Exosomes may expand their clinical application as a potential drug for treating OA. This review focuses on recent research on the mechanisms of exosome action and strategies to enhance their therapeutic effects, aiming to provide new directions for future research.

Fig. 1.

Pathogenesis diagram of osteoarthritis (a) Normal joint structure (b) Changes in the structures of various components of the joint in osteoarthritis

Introduction to exosomes

Recent studies have demonstrated the promising potential of exosomes in treating and potentially reversing the pathological changes associated with OA and cartilage injury, as well as alleviating clinical symptoms in both in vivo and in vitro models [8]. Therefore, exosomes may represent a novel therapeutic option. Exosomes are a type of extracellular vesicle, typically ranging in size from 30 to 150 nm, with a density of 1.13 to 1.19 g/mL [9]. These small vesicles, derived from the cell membrane, carry a variety of complex signaling molecules, including lipids, proteins, nucleic acids, and intracellular components, and exhibit cell-specific transport functions. Exosomes are released into the extracellular space through fusion with multivesicular bodies (MVBs) and are secreted by nearly all cell types, playing a key role in intercellular communication [10]. Subsequently, target cells internalize exosomes, leading to specific effects. Notably, compared to parent cells, exosomes have significantly higher concentrations of sphingolipids, sphingomyelin, cholesterol, and phosphatidylserine. These membrane lipids not only ensure the stability of exosomes in the ECM but also promote their uptake by target cells [11, 12]. Additionally, proteins derived from the exosome membrane play a crucial role in tissue repair processes, facilitating communication between exosomes and recipient cells [13]. Exosome cargo, such as DNA, miRNA, proteins, and other RNA species (e.g., mRNA, lncRNA, circRNA), has become a key focus of current research. Specifically, in the field of cartilage repair and regeneration, RNA within exosomes regulates genes involved in inflammation, cell proliferation, apoptosis, and ECM synthesis, which are critical for treating OA [14]. Therefore, the internal components of exosomes play a pivotal role in influencing the progression of OA.

Functions of exosomes derived from different sources and their mechanisms

The function of exosomes is primarily determined by their internal contents, which are secreted by the parent cells. Exosomes can be broadly classified based on the physiological state of their parent cells: Healthy exosomes: Derived from cells in normal or regenerative states, these exosomes have therapeutic effects.Conversely, pathological exosomes: Released by cells under disease conditions, these exosomes exacerbated the progress of OA. Therefore, exosomes from different sources exhibit distinct roles. (Table 1)Parent cells such as mesenchymal stem cells (MSCs) can secrete classic healthy exosomes that help delay the progression of OA and cartilage injury, while alleviating related symptoms. In addition to these classic sources, there are also non-classic sources of healthy exosomes, including exosomes derived from synovial fibroblasts (SFB-exos), platelets (Plt-exos), osteocytes (OCY-exos), cartilage stem/progenitor cells (CSPC-exos), antler stem cells (ASC-exos), dental pulp stem cells (DPSC-exos), garlic (GDE), and M2 macrophage (M2-exos), all of which have shown therapeutic effects in OA. However, exosomes released by certain parent cells may act as pathological exosomes, inhibiting cartilage repair, enhancing inflammatory responses, and further exacerbating the progression of OA. Therefore, a deeper investigation into the functions and mechanisms of exosomes from different sources will provide crucial insights for identifying potential therapeutic strategies for OA.

Table 1.

The roles and mechanisms of exosomes from different sources

Types of exosomes	Mechanism of action	Function	Ref
Exosomes derived from bone marrow mesenchymal stem cells (BMSC-exos)	Sp1 expression is negatively modulated by miR-135b	Promoted chondrocyte proliferation	[15]
	Upregulation of the TGFB1 gene targets the Nrf2 signaling pathway to inhibit NET formation and the release of ROS	Promoted the generation of cartilage ECM, chondrocyte proliferation and migration, and inhibited cell apoptosis	[16]
	Negative modulation of miRNA-21 by circYAP1	Decreased IL-1β, TNF-α, ROS, GSH, and MDA; facilitates OA cartilage regeneration	[17]
	Through the upregulation of lncRNA TUC339 expression	Promoted M2 macrophage polarization, chondrocyte proliferation and migration; inhibited chondrocyte apoptosis and osteogenic gene expression	[18]
Exosomes derived from human Wharton's jelly mesenchymal stem cells (hWJMSC-exos)	By significantly upregulating the expression of miR-148a and miR-29b	Promoted chondrocyte proliferation and cartilage ECM remodeling	[19]
	TFRC/BMP2/RUNX2 axis	Alleviated IL-1β-induced cartilage damage, inhibited cell apoptosis, and enhanced chondrocyte survival and cell cycle progression	[20]
Exosomes derived from adipose-derived stem cells (ADSC-exos)	Targeting and inhibiting FEZ2 through miR-429	Enhanced chondrocyte autophagy, thereby promoting chondrocyte proliferation and type II collagen production, which contributes to the treatment of cartilage injury	[21]
	Upregulation of miR-199a-3p inhibits mTOR and its downstream p-p70S6K	Enhanced autophagy in rat OA cartilage, significantly alleviating cartilage injury and proteoglycan loss, while promoting COL2A1 expression and inhibiting MMP13 expression	[22]
	Inhibited ADAMTS9-induced activation of the PI3K/AKT/mTOR signaling pathway by miR-93-5p	Enhancing autophagy in chondrocytes effectively alleviates IL-1β-induced chondrocyte apoptosis and significantly suppresses the expression of pro-inflammatory cytokines, including IL-6, IL-1β, and TNF-α	[23]
Exosomes derived from human adipose-derived mesenchymal stem cells (hADSC-exos)	Inhibited WNT3 and WNT9a using miR-376c-3p to decrease β-catenin expression	Downregulating inflammatory and fibrotic markers and upregulating cartilage repair molecules markedly improved cartilage injury and suppressed synovial thickening	[24]
Exosomes derived from human amniotic fluid stem cells (AFSC-exos)	/	Restored pain threshold in OA mice, with new cartilage covering joint defects, widespread and uniform cartilage matrix distribution, inhibited fibrous tissue regeneration, and promoted M2 macrophage polarization	[25]
Exosomes derived from umbilical cord mesenchymal stem cells(UCMSC-exos)	Inhibition of p53 gene expression, along with downregulation of p21, p16, and TP53BP, and upregulation of Ki67 expression	Significantly enhanced the viability and proliferative capacity of OA chondrocytes	[26]
Exosomes derived from human umbilical cord mesenchymal stem cells (hUCMSC-exos)	Inhibiting MAPK4 expression and blocking NF - κ B signaling pathway activation by delivering miR-199a-3p	Reduction of chondrocyte inflammation and apoptosis	[27]
Exosomes derived from synovial mesenchymal stem cells (SMSC-exos)	NRP1 expression is inhibited and its downstream PI3K/Akt signaling pathway is inactivated by miR-485-3p	Reversing ECM degradation in OA chondrocytes and promoting cartilage injury repair	[28]
	The ADAM19-dependent Wnt signaling pathway is inhibited by miR-320c	Inhibited ECM degradation and chondrocyte apoptosis, thereby promoting cartilage repair	[29]
Exosomes derived from Platelet (Plt-exos)	/	Promoting chondrocyte proliferation and migration in vitro; repairing cartilage injury and preventing subchondral bone loss in vivo	[30]
Exosomes derived from Osteocyte (OCY-exos)	Inhibited of the Wnt signaling pathway by DLX2	Facilitating ECM deposition, enhancing chondrocyte viability and migration, and inhibiting cell apoptosis	[31]
Exosomes derived from cartilage stem/progenitor cells (CSPC-exos)	Upregulated the expression of protein CDK9	DLX2 treatment enhances chondrocyte viability, proliferation, and migration in vitro; alleviates subacute OA in rats, promotes cartilage repair, and facilitates remodeling of the cartilage matrix in vivo	[32]
Exosomes derived from antler stem cells (ASC-exos)	/	Promoting the proliferation and migration of chondrocytes and rat BMSCs, while significantly enhancing the expression of cartilage-related genes and proteins	[33]
Exosomes derived from antler stem cells dental pulp stem cells(DPSC-exos)	Activating TRPV4 to promote osteoclast activation	Alleviating abnormal subchondral bone remodeling, cortical bone sclerosis, and osteophyte formation in mouse knee OA; repairing the fractured cartilage layer; and inhibiting synovial tissue hyperplasia and infiltration of inflammatory cells	[34]
Exosomes derived from garlic (GDE)	Inactivating the MAPK pathway	Reversed cartilage ECM degradation and alleviated cartilage injury in OA	[35]
Exosomes derived from M2 macrophage (M2-exos)	Inhibited the activity of the TLR pathway by miR-26b-5p	Alleviating synovial inflammation, promoting hyaline cartilage regeneration, and improving gait in OA mice	[36]
Exosomes derived from human progenitor endothelial cells(hPESCs-exo)	Targeted recruitment of bone marrow mesenchymal stem cells to the site of injury by activating the TGF - β/Smad pathway and releasing chemokines such as SDF-1 to bind to CXCR4 receptor	Enhancing cartilage repair ability, alleviate the microenvironment of osteoarthritis, simultaneously inhibit catabolic metabolism and promote synthetic metabolism	[37]
Exosomes derived from IL-1β-treated fibroblast-like synoviocytes (OA-FLS-exos)	Inducing enhanced glycolysis in macrophages	Promoting M1 macrophage polarization and chondrocyte apoptosis	[38]
	Downregulated SLC7A11 by miR-19b-3p	Promoting ferroptosis in OA chondrocytes	[39]
Exosomes derived from osteoclasts (OC-exos)	Inhibited the Smad2-mediated TGF-β signaling pathway by let-7a-5p	Promoting hypertrophic differentiation of chondrocytes	[40]
	Inhibited Smad2 expression by miR-212-3p and activated the NF-κB pathway	Suppressing chondrocyte anabolic metabolism and enhancing catabolic metabolism, thereby exacerbating the inflammatory response	[41]
Exosomes derived from dysfunctional chondrocytes treated with IL-1β (DC-exos)	Upregulated the expression of the PI3K/AKT/mTOR signaling pathway by lncRNA-AC002091.2	Promoting M1 macrophage polarization and inducing OA inflammatory response	[42]
Exosomes derived from OA chondrocyte (OA-CDE)	Inhibited miR-1277 via circ-BRWD1, thereby upregulating TRAF6 expression	Decreasing chondrocyte viability, inducing cell apoptosis, inflammation, and ECM degradation	[43]
	Transferred miR-125 to osteoblasts	Increasing the levels of inflammatory cytokines, exacerbating the disruption of bone homeostasis, and worsening cartilage injury	[44]
Exosomes derived from synovial fluid (SF-exos)	Activation of IL-17 and TNF inflammatory pathways by upregulating complement components C3 and C4B, and exacerbation of complement cascade reaction	Deteriorating verification microenvironment	[45]

Healthy exosomes from classic sources

MSCs possess multi-lineage differentiation potential and have been utilized in intra-articular injection therapies for OA treatment [46]. However, MSCs also have several limitations, including cell heterogeneity, inconsistent stemness, variability in differentiation capacity, limited homing ability, and potential adverse effects such as immune incompatibility, tumorigenesis risk, and chromosomal abnormalities [47]. Nonetheless, the efficacy of many MSC-based treatments is primarily attributed to the secretion of paracrine factors, particularly exosomes [48].

BMSC-exos

Previous studies have shown that exosomes derived from bone marrow mesenchymal stem cells (BMSC-exos) enhance chondrocyte proliferation, promote ECM synthesis, suppress inflammation, and alleviate pain in knee OA [49]. Wang et al. found that BMSC-exos significantly express miR-135b, which negatively regulates Sp1 (a transcription factor), thereby reversing the inhibitory effects on chondrocyte proliferation [15]. More recently, exosomes from BMSCs overexpressing miR-135b were found to promote M2 macrophage polarization via targeting MAPK6, downregulating inflammatory factors such as IL-1β and PGE2, thus improving cartilage injury [50]. Neutrophil extracellular traps (NETs) have been implicated in cartilage degradation driven by neutrophil elastase [51], while Nrf2, a transcriptional regulator involved in oxidative stress responses and antioxidant defenses, plays a significant role. BMSC-exos upregulates TGFB1 to target the Nrf2 pathway, inhibiting NET formation and reactive oxygen species (ROS) release, thereby promoting cartilage ECM generation, chondrocyte proliferation, migration, and reducing apoptosis, which is beneficial for cartilage regeneration [16]. Additionally, circular RNA Yes-associated protein 1 (circYAP1) negatively regulates miRNA-21 to silence its pro-inflammatory effects [52], and BMSC-exos utilizes this mechanism to significantly reduce OA progression in rats, with notable increases in cartilage thickness and reductions in pro-inflammatory factors (IL-1β, TNF-α) and oxidative stress markers, including ROS, glutathione (GSH), and malondialdehyde (MDA), highlighting its potent anti-inflammatory and antioxidant potential [17]. Another study reported that BMSC-exos significantly expressed long non-coding RNA TUC339, which promoted M2 macrophage polarization, chondrocyte proliferation and migration, and inhibited chondrocyte apoptosis and osteogenesis-related gene expression [18].

hWJMSC-exos

Beyond BMSC-derived exosomes, similar immunomodulatory and regenerative capacities have been observed in exosomes from other mesenchymal stem cell sources.Jiang et al. found that exosomes derived from human Wharton’s jelly mesenchymal stem cells (hWJMSC-exos) promoted chondrocyte regeneration and BMSC migration in a dose-dependent manner in vitro and induced M2 macrophage polarization [19]. Moreover, hWJMSC-exos are enriched in miR-148a and miR-29b, which have been shown to promote chondrocyte proliferation and ECM remodeling [53, 54], suggesting that these miRNAs may serve as new therapeutic targets for OA. Bone morphogenetic protein (BMP) promotes cartilage regeneration [55], and the runt-related transcription factor 2 (RUNX2) plays a key role in endochondral and membranous ossification [56]. Transferrin receptor (TFRC) binds iron-loaded transferrin and mediates the delivery of extracellular iron to the cell via clathrin-dependent endocytosis [57]. Chen et al. found that hWJMSC-exos promoted the TFRC/BMP2/RUNX2 axis, significantly improving IL-1β-induced cartilage injury, inhibiting cell apoptosis, and enhancing chondrocyte survival and cell cycle progression [20].

ADSC-exos

Clinically, adipose tissue is relatively easy to obtain, which has led to increasing research on exosomes derived from adipose-derived stem cells (ADSC-exos). Meng et al. attributed the protective effects of ADSC-exos on cartilage to high levels of miR-429 [21]. In their study, ADSC-exos inhibited FEZ2 through miR-429, increasing the number of autophagosomes, upregulating autophagy-related proteins Beclin 1 and LC3-II/I, and enhancing chondrocyte proliferation and type II collagen production, thus treating cartilage injury caused by OA. mTOR (mammalian target of rapamycin) is a key negative regulator in the autophagy process [58]. Zhao et al. showed that ADSC-exos upregulated miR-199a-3p, targeting mTOR and inhibiting its expression, which decreased p-p70S6 (a downstream signaling molecule of mTOR) [22]. During the continuous autophagy mediation by ADSC-exos, LC3B protein levels (a key component of autophagosomes) were significantly upregulated, improving autophagy levels in OA cartilage in rats, and significantly alleviating cartilage injury and proteoglycan loss. The expression of COL2A1 (an anabolic gene) increased, while MMP13 (a catabolic gene) decreased, mitigating OA pathology. Additionally, the ADAMTS molecular mechanism mediates changes in cell apoptosis and homeostatic processes, including autophagy [59], and inhibiting the PI3K/AKT/mTOR signaling pathway can induce chondrocyte autophagy, thus alleviating OA [60]. Li et al. discovered that miR-93-5p, a critical miRNA in ADSC-exos, mediated the inhibition of ADAMTS9-activated PI3K/AKT/mTOR signaling, upregulated autophagy levels, alleviated IL-1β-induced chondrocyte apoptosis, and downregulated pro-inflammatory factor expression (IL-6, IL-1β, TNF-α) [23]. The WNT-β-catenin signaling pathway plays a significant role in the pathogenesis of OA, and inhibition of the WNT pathway can promote cartilage growth to reduce OA symptoms [61]. miR-376c-3p, a conserved miRNA, has been shown to play an important role in cell apoptosis and injury processes [62]. Recently, Li et al. found that exosomes derived from human adipose-derived mesenchymal stem cells (hADSC-exos) significantly express miR-376c-3p, which targets WNT3 and WNT9a, negatively regulating these genes, leading to reduced β-catenin expression and downregulation of OA-related inflammatory factors (IL-1β, iNOS, TNF-α, IL-6, and IFN-γ) and fibrosis-related molecules (α-SMA, Col1a3, MMP3, MMP13, TIMP1, TIMP2, ADAMTS4, and ADAMTS5), while upregulating cartilage injury repair-related molecules (Col2a1, Sox9, Comp, and ACAN), significantly improving cartilage injury and inhibiting synovial thickening [24].

AFSC-exos

Beyond ADSC-exos, exosomes from other classic sources, such as amniotic fluid stem cells (AFSC-exos), have also demonstrated therapeutic potential in OA.Zavatti et al. previously reported that AFSC-exos not only exhibit low immunogenicity but also secrete immunoregulatory factors targeting autoimmune diseases or allogeneic implants, indicating their potential therapeutic value [63]. Recently, they found that AFSC-exos restored the pain threshold in OA mice after 6 weeks, with newly-formed cartilage almost covering the joint defects, and a widespread and uniform cartilage matrix distribution. Additionally, AFSC-exos inhibited fibrous tissue regeneration and promoted M2 macrophage polarization [25]. However, research on AFSC-exos is still in its early stages, and further investigation is needed to understand their role in OA pathogenesis and treatment. The p53 pathway governs cellular aging, responding to various internal and external signals to regulate cellular stability mechanisms involved in DNA replication, chromosome segregation, and cell cycle control.

UCMSC-exos

Recent studies on umbilical cord mesenchymal stem cell-derived exosomes (UCMSC-exos) have directly linked this pathway to OA therapy.Cao et al. reported that UCMSC-exos may intervene in chondrocyte senescence in OA by inhibiting the p53 pathway [26]. They found that UCMSC-exos suppressed the expression of p53 and three related genes (p21, p16, and TP53BP) and upregulated the expression of cell proliferation-related genes (Ki67), significantly enhancing the survival and proliferative capacity of OA chondrocytes. Furthermore, UCMSC-exos downregulated SASP genes that promote inflammation and aging. Previous studies have shown that miR-485-3p can promote chondrocyte proliferation and inhibit apoptosis in OA [64]. Inhibition of NRP1, which is highly expressed in OA cartilage, suppresses MMP13 transcription and inactivates the PI3K/AKT pathway, thus alleviating OA [65]. Exosomes derived from human umbilical cord mesenchymal stem cells (hUCMSC-exos)targeted the inhibition of MAPK4 expression by delivering miR-199a-3p, thereby blocking the activation of the NF - κ B signaling pathway, reducing chondrocyte inflammation and apoptosis, and alleviating the pathological process of osteoarthritis [27].

SMSC-exos

Synovial mesenchymal stem cell-derived exosomes (SMSC-exos) have emerged as a pivotal therapeutic agent in OA by orchestrating multi-pathway regulation through specific miRNAs. Recent studies demonstrate their ability to counteract IL-1β-induced chondrocyte dysfunction and ECM degradation via targeted molecular interventions [28]. Mechanistically, miR-485-3p in SMSC-exos mediates the inhibition of the target gene NRP1, thereby inactivating the downstream PI3K/Akt signaling pathway. In earlier studies, miR-320c in exosomes was identified as a key gene promoting the treatment of OA by SMSC [66]. Recently, Kong et al. found that SMSC-exos targets ADAM19-dependent Wnt signaling through miR-320c, inhibiting ECM degradation and chondrocyte apoptosis, thus promoting cartilage injury repair [29].

Exosomes derived from classic MSC sources exhibit multifaceted therapeutic effects in OA, including promoting chondrocyte proliferation, enhancing ECM synthesis, and modulating inflammatory responses. Their mechanisms involve miRNA-mediated regulation of key pathways and interaction with immune cells. These findings highlight their potential as a cell-free therapeutic strategy, though further studies are needed to optimize their specificity and efficacy.

Non-classical sources of healthy exosomes

SFB-exos

In conventional studies, exosomes used for treating OA have primarily been derived from MSCs. However, exosomes secreted by other cell types, serving as non-classical sources of healthy exosomes, also display comparable biological activities. For example, SFB-exos serve as important regulators of joint-local inflammation mediated by SFB and chondrocytes. These exosomes have the ability to transfer miRNAs and inflammatory proteins to tissues [67]. Lai et al. found that, compared to healthy individuals, OA patients undergoing total knee replacement had significantly downregulated expression of miR-214-3p in SFB-exos [68]. Overexpression of miR-214-3p in SFB-exos significantly promoted chondrocyte proliferation, downregulated inflammatory cytokines, and inhibited apoptosis. In vivo, this treatment protected the subchondral bone of OA rats, increased cartilage thickness, and restored a smooth, regular cartilage surface, significantly improving cartilage injury. Additionally, it effectively suppressed synovial thickening and alleviated synovitis.

Plt-exos

Platelet-rich plasma (PRP) has been shown to repair OA cartilage [69], and increasing evidence suggests that the regenerative effects of PRP are primarily attributed to exosomes derived from Platelet (Plt-exos) [70]. Plt-exos are not only readily accessible but also offer a viable and highly safe alternative to exosomes derived from mesenchymal stem cells. Xu et al. demonstrated that Plt-exos promoted chondrocyte proliferation and migration in vitro, highlighting their regenerative potential. In vivo, Plt-exos improved cartilage thickness and bone microstructure in OA mice, preventing cartilage degradation and subchondral bone loss [30].

OCY-exos

Previous studies have shown that exosomes derived from Osteocyte (OCY-exos) are involved in bone metabolism and diseases, such as periodontitis [71]. However, the potential mechanisms of OCY-exos in OA have not yet been clarified. Recently, Xu et al. reported that OCY-exos, both in vitro and in vivo, regulate ECM deposition, enhance chondrocyte viability, migration, and inhibit apoptosis through modulation of the DLX2/Wnt signaling pathway [31]. DLX2 expression is notably elevated in OCY-exos, accompanied by a suppression of the Wnt signaling pathway.

CSPC-exos

CDK9, a key regulator of cell proliferation, plays a critical role in the transcriptional process mediated by RNA polymerase II, particularly in cellular growth [72]. Chen et al. identified CDK9 as a crucial protein in CSPC-exos, showing significant promise for OA cartilage injury repair [32]. CSPC-exos have been shown to enhance chondrocyte viability, proliferation, and migration in vitro, while alleviating subacute OA in rats and promoting cartilage regeneration at the injury site. This leads to the remodeling of cartilage matrix. Furthermore, CDK9 stimulates the expression of genes involved in chondrocyte growth and migration, such as Tgfa, Myocd, Gpnmb, and Rufy3.

ASC-exos

Interestingly, antler cartilage, under natural conditions, not only exhibits self-repair capabilities but also demonstrates remarkable growth rates of up to 2 cm per day [73]. A study revealed that ASC-exos promote the proliferation and migration of chondrocytes and rat bone marrow stem cells (BMSCs), significantly upregulating cartilage-associated genes and proteins [33]. Additionally, ASC-exos are enriched in proteins such as transferrin (TF), S100 calcium-binding protein A4 (S100A4), and insulin-like growth factor 1 (IGF1), which play key roles in promoting cell proliferation, migration, and cartilage biosynthesis.

DPSC-exos

multi-lineage differentiation capacity [74]and immunomodulatory effects in the microenvironment [75], highlighting the clinical promise of DPSC-exos in OA treatment [76]. The TRPV4 mechanosensitive calcium channel has been implicated in the regulation of bone mass through enhanced osteoclast differentiation [77]. Recently, Fu et al. demonstrated that DPSC-exos activate TRPV4 to promote osteoclast activation, thereby ameliorating abnormal subchondral bone remodeling, cortical bone sclerosis, and osteophyte formation in OA mice. This results in the repair of damaged cartilage, suppression of synovial hyperplasia, and reduction of inflammatory cell infiltration [34].

GDE

Matrix-degrading enzyme transcription is finely regulated through multiple signaling pathways, with the MAPK signaling pathway being a prime example [78]. Liu et al. found that GDE improved chondrocyte viability and significantly inhibited ECM-degrading enzymes, such as matrix metalloproteinases (MMPs) and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS), through MAPK pathway inhibition. This reversal of ECM degradation alleviates cartilage injury in OA [35].

M2-exos

Emerging evidence underscores the dual therapeutic potential of M2-exos in OA, which synergistically enhance lymphatic regeneration and resolve inflammatory cascades through targeted immunomodulation.For instance, Song et al. demonstrated that M2-exos enhance the viability of lymphatic endothelial cells (LECs) and promotes their proliferation and migration, thereby facilitating lymphangiogenesis. Additionally, M2-exos aid in the clearance of inflammatory mediators, such as cytokines and chemokines, in OA, resulting in reduced local inflammation levels [79]. Another study revealed that M2-exos induced macrophage polarization towards the M2 phenotype, alleviating synovial inflammation, facilitating hyaline cartilage regeneration, and improving gait in mice [36]. This effect may be attributed to the high levels of miR-26b-5p, which suppresses TLR signaling activity, thereby preventing M1 macrophage polarization and targeting COL10A1.

hPESCs-exos

hPESCs-exo promoted the synthesis of key extracellular matrix components such as collagen II and proteoglycans in chondrocytes by activating the TGF - β/Smad pathway, enhancing cartilage repair ability [37]. In addition, extracellular vesicles reduce the pathological degradation of extracellular matrix by inhibiting the activity of matrix metalloproteinases (MMPs), and selectively recruit stem cells to migrate to the site of injury to participate in tissue regeneration by releasing chemokines (such as SDF-1) and binding to CXCR4 receptors on the surface of bone marrow mesenchymal stem cells. This multi-channel synergistic effect significantly alleviates the inflammatory microenvironment of osteoarthritis, synchronously inhibits catabolism and promotes synthetic metabolism, ultimately achieving the repair of cartilage structure and function [37].

Non-classical exosome sources demonstrate unique therapeutic potential by targeting synovial inflammation, subchondral bone remodeling, and cartilage regeneration. However, their long-term safety and mechanisms require further validation in clinical settings.

Pathological exosomes in the development and progression of OA

Exosomes play a significant role not only in the treatment of OA but also in exacerbating the progression of OA through what are termed pathological exosomes. Recently, Liu et al. found that exosomes derived from IL-1β-treated fibroblast-like synoviocytes (OA-FLS-exos) promote M1 macrophage polarization and chondrocyte apoptosis, thereby aggravating the inflammatory microenvironment and exacerbating OA pathology [38]. OA-FLS-exos induce an OA-like phenotype by enhancing macrophage glycolysis, a process closely associated with the activation of HIF1A.

Ferroptosis, a novel form of regulated cell death, is closely related to redox balance, iron metabolism, and mitochondrial function. Studies have shown that OA-FLS-exos further promote ferroptosis in OA chondrocytes, a process potentially mediated by miR-19b-3, which targets and downregulates the ferroptosis regulator SLC7A11 [39]. Additionally, exosomes derived from osteoclasts (OC-exos) inhibit the TGFβ signaling pathway by targeting Smad2 through let-7a-5p, promoting hypertrophic differentiation of chondrocytes [40]. Another study indicated that the high expression of miR-212-3p in OC-exos targets Smad2 to suppress chondrocyte anabolic metabolism and promote catabolic metabolism, while also activating the NF-kB pathway, which further amplifies the inflammatory response and accelerates OA progression [41]. Certain long non-coding RNAs (lncRNAs) in OA cartilage have also been shown to participate in various pathological processes, including ECM degradation, inflammation, and apoptosis [80]. Lv et al. reported that exosomes derived from dysfunctional chondrocytes treated with IL-1β (DC-exos) promote M1 macrophage polarization and induce OA-related inflammatory responses through lncRNA-AC002091.2 (OANCT) [42]. The downstream signaling of OANCT involves the PI3K/AKT/mTOR pathway, where OANCT interacts with FTO (a demethylase) to mediate the demethylation of PIK3R5 mRNA, leading to its upregulation.

Further studies have found that miR-1277, through interactions with MMP13 and ADAMTS5, alleviates ECM degradation in IL-1β-treated chondrocytes [81]. Circular RNAs (circRNAs) regulate gene expression by competitively binding miRNAs [82]. For example, Guo et al. demonstrated that exosomes derived from OA chondrocyte (OA-CDE) significantly express circ-BRWD1 [43]. In this study, circ-BRWD1 targeted and inhibited miR-1277, leading to the upregulation of its direct target gene TRAF6, which resulted in decreased chondrocyte viability, apoptosis, inflammatory responses, and ECM degradation.

The widespread distribution of sympathetic nerves in the synovium, cartilage, and subchondral bone suggests that sympathetic nerve activity may influence OA progression. Recent studies have shown that increased sympathetic nerve activity in OA mice accelerates disease progression, including cartilage injury and subchondral bone ossification [44]. One mechanism involves the activation of sympathetic nerves, which facilitates the transfer of miR-125 from OA-CDE to osteoblasts, upregulating inflammatory factor expression and exacerbating bone homeostasis disruption and cartilage injury. SF-exos activate inflammatory signaling pathways (such as IL-17 and TNF pathways) by upregulating C3 and C4B, enhancing complement cascade reactions, thereby worsening the inflammatory microenvironment and exacerbating the progression of OA [45].

Exosome engineering for osteoarthritis treatment

Exosomes exhibit considerable potential for the treatment of OA and cartilage injury; however, their native form presents certain limitations in targeting specificity, cargo capacity, and in vivo stability. These constraints reduce the therapeutic efficacy of exosomes in OA treatment, and overcoming these challenges is essential before their clinical application. To optimize the therapeutic outcomes, it is crucial to engineer and modify exosomes purposefully. Incorporating additional processing strategies can further enhance their effectiveness in treating OA. (Figure 2).

Fig. 2.

Steps for engineering exosome modification (A) The porcine cartilage was fragmented and subjected to decellularization to obtain an ACECM suspension. The suspension was then placed in a cylindrical mold for deep freezing and lyophilization, followed by crosslinking with water-soluble carbodiimide to finally produce a porous scaffold with a vertically oriented structure. (B) Cas9 protein and guide RNA (sgRNA) are mixed with liposomes to form the liposome-Cas9/sgRNA complex, which then fuses with exosomes. Through endocytosis, the complex enters target cells and ultimately exerts its effect on the target gene. (C) A plasmid containing the CAP sequence was first constructed, encoding lysosome-associated membrane glycoprotein 2b (Lamp2b), with the CAP sequence (DWRVIIPPRPSA) inserted at its N-terminus. The plasmid also includes a glycosylation sequence (GNSTM) and a glycine-serine (Gly-Ser) spacer sequence to protect the CAP sequence from degradation and ensure its effective presentation on the exosome surface. After successful construction, the plasmid was transfected into specific cells, which, upon expressing Lamp2b and the CAP sequence, naturally secreted exosomes displaying the CAP sequence

Decellularized cartilage extracellular matrix (ACECM) scaffold

Exosomes encapsulated in scaffolds can be confined to the defect site, minimizing loss during application. Jiang et al. demonstrated that in OA rats, the hWJMSC-exos + ACECM scaffold group successfully induced hyaline cartilage regeneration in the cartilage defect area, with no distinct boundary observed. The combination of exosomes and the scaffold exhibited synergistic effects, showing superior repair outcomes compared to the hWJMSC-exos group alone [19]. Another study used 3D printing to construct both 3D ACECM scaffolds and 2D ACECM membranes to culture human umbilical mesenchymal stem cells (hUMSCs) and extract exosomes. Compared to Exos and 2D-Exos, the 3D-Exos significantly enhanced BMSC chondrogenic differentiation, cartilage cell proliferation and migration, inhibited cartilage cell apoptosis, and promoted macrophage M2 polarization [83].

Gene editing

Gene therapy has emerged as a promising option for treating OA. Several genes closely associated with OA pathology have been identified in chondrocytes, making gene editing a potential long-term therapeutic strategy for OA patients [84]. MMP-13, a protease that degrades type II collagen in the ECM, was targeted by Liang et al. through gene editing, using exosomes combined with Cas9 sgMMP-13 plasmids to suppress MMP-13 expression. This approach restored type II collagen expression, protected the ECM, and led to successful cartilage repair in OA mice, with a smooth surface and structurally intact cartilage similar to normal tissue [85]. FGF18 has also been reported to promote cartilage repair and effectively inhibit OA progression [86]. Recently, Chen et al. used gene editing to activate the FGF18 gene, upregulating cartilage regeneration-associated genes such as SOX9, Col2a1, and ACAN while downregulating the inflammatory gene IL-1β. This strategy promoted chondrocyte proliferation, improved joint structure, and alleviated OA symptoms in OA rats [87].

Cartilage targeting peptide (CAP)

Natural exosomes exhibit limited targeting ability. CAP peptides can enhance exosome recognition and binding to specific cells or tissues, improving targeting efficiency and ensuring that exosomes effectively reach the desired location. In studies on cartilage-targeted exosomes, one study reported that the binding of CAP peptides to mouse dendritic cells enabled exosomes to specifically target chondrocytes, showing significantly better therapeutic effects in a rat OA model compared to unmodified exosomes [88]. Another study demonstrated that exosomes engineered with CAP peptides enhanced miRNA delivery efficiency in treating OA in mice, effectively slowing the progression of the disease [22], and it also integrates miRNA overexpression techniques (such as overexpression of miR-199a-3p). Additionally, Cao et al. found that CAP-modified exosomes significantly prolonged retention time within the joint, allowing for prolonged therapeutic effects, including improvements in chondrocyte function and promotion of cartilage matrix synthesis, which ultimately ameliorated cartilage injury [26], and combined with delivery system optimization (such as cartilage targeted polymer modification and thiolated hyaluronic acid micro gel coating). While applying CAP, other technologies such as miRNA overexpression or delivery system optimization can also be integrated to synergistically improve targeting efficiency and therapeutic efficacy.

Hydrogels

Natural exosomes have a fast metabolism and a short turnover rate, leading to rapid clearance from the body, which limits their retention at the targeted lesion site and hinders long-term therapeutic effects.(Table 2)Zhou et al. introduced chondroitin sulfate (OCS) into methacrylate gelatin (GM) to construct GMOCS hydrogels and incorporated BMSC-derived exosomes (GMOCS-Exos) [16]. Compared to other groups, GMOCS-Exos significantly promoted chondrocyte proliferation, reversing cartilage degeneration in OA rats, including degradation of cartilage matrix, chondrocyte death, and osteophyte formation. Li et al. used a synthetic hydrogel scaffold composed of methacrylate gelatin, oxidized hyaluronic acid, and dopamine-conjugated hyaluronic acid, combining it with decellularized cartilage extracellular matrix (dECM) and exosomes, which promoted cartilage-specific gene expression, inhibited hypertrophic differentiation of chondrocytes, and filled cartilage defects in rats [89]. Hyaluronic acid methacrylate (HAMA) acted as a lubricant, while gelatin methacrylate (GelMA) provided a supportive growth environment. Yang et al. constructed a hydrogel carrier (Par) using HAMA and GelMA, combined with exosomes (Exo), to form hydrogel particles [90]. The porous structure of Par@Exo promoted migration and chondrogenic differentiation of stem cells such as BMSCs, significantly enhancing the expression of cartilage cell markers. In vivo, this treatment alleviated cartilage erosion, enhanced joint stability, and improved the therapeutic efficacy of exosomes. Lymphatic endothelial cells (LEC), the fundamental units of the synovial lymphatic system (SLS), are essential for lymphatic drainage in OA. OA-induced reductions in synovial lymphatic vessels impair drainage function, causing inflammatory factor accumulation and accelerating OA progression [91]. Song et al. used thermosensitive hydrogels (HP) to load M2-exos, promoting 3D lymphatic vessel formation, improving lymphatic drainage, and facilitating cartilage regeneration in OA rats. This resulted in cartilage matrix remodeling and increased mechanical strength [79]. Notably, HP undergoes a reversible sol-gel phase transition at specific temperatures, allowing rapid exosome release at lower temperatures, while gelation at higher temperatures enables sustained exosome release.

Table 2.

The release time of exosomes loaded in hydrogels

Hydrogel component	Types of exosomes	Sustained release time of exosomes	Ref
Methacryloyl Gelatin + Chondroitin Sulfate	BMSC-exo	14days	[16]
Methacrylated Gelatin + Oxidized Hyaluronic Acid + Dopamine-conjugated Hyaluronic Acid	BMSC-exo	24days	[89]
Hyaluronic Acid Methacryloyl + Gelatin Methacryloyl	UCMSC-exo	12days	[90]
Hyaluronic Acid + Pluronic F-127	M2-exo	10days	[79]

Pre-treatment

Preprocessing refers to the use of specific methods and techniques to regulate the microenvironment of cells or tissues during experimental processes, in order to optimize their function and behavior and enhance their effectiveness in specific applications. In the study of extracellular vesicles, the role of pretreatment is particularly important, such as by regulating the microenvironment of mother cells (such as hypoxia, drug stimulation, or biomolecule induction), thereby affecting the characteristics and functions of extracellular vesicles, enhancing their therapeutic effect on OA, and providing new strategies and ideas for the treatment of OA. This approach offers a novel strategy and perspective for the treatment of OA. However, the precise mechanisms through which pre-treated exosomes exert their protective effects in OA remain incompletely understood.One study demonstrated that hypoxic pre-treatment enhanced the therapeutic effects of ADSC-exos, significantly improving pain and cartilage degeneration in lumbar facet joint OA (LFJ OA). This treatment also helped preserve the cartilage ECM and alleviated abnormal H-type angiogenesis in the subchondral bone [92]. Another study demonstrated that, under hypoxic conditions, pre-treatment of BMSC-exos enhanced OA chondrocyte activity and promoted the expression of cartilage-specific phenotypes, while also inhibiting cell apoptosis, compared to normoxic conditions. Hypoxia upregulated HIF-1α levels, which induced a significant expression of miR-140-3p in BMSC-exos, thereby promoting chondrocyte migration and proliferation [93]. Kartogenin (KGN) is known for its potent ability to induce mesenchymal stem cells to differentiate into chondrocytes [94]. Shao et al. reported that exosomes derived from infrapatellar fat pad mesenchymal stem cells (IPFP-exos), when pre-treated with KGN, further enhanced their ability to promote cartilage repair, significantly upregulating cartilage-related genes [95]. In another study, TNF-α pre-treatment increased the release of IPFP-exos, which also improved gait and alleviated synovial inflammation [96]. Zhang et al. found that dECM pre-treatment of BMSC-exos significantly upregulated miR-3473b and negatively regulated the PTEN/AKT signaling pathway, a critical pathway in cartilage repair [97]. Fan et al. previously demonstrated that tropoelastin (TE) not only inhibits the progression of OA but also promotes the adhesion and migration of IPFP cells [98]. More recently, they found that ADSC pre-treated with TE significantly enhanced exosome release, effectively maintaining the cartilage phenotype and reversing cartilage injury in rat models of OA [99]. Furthermore, TE pre-treatment upregulated miR-451-5p and 232 potential target genes associated with chondrocyte proliferation, differentiation, and ECM synthesis.Previous studies have highlighted the role of TRAF6 (TNF receptor-associated factor 6) as a key inflammatory mediator that exacerbates OA [100]. Recently, Dong et al. reported that quercetin (QUE) pre-treatment of BMSC-derived exosomes significantly inhibited the NF-κB/p65 signaling pathway, which is positively regulated by TRAF6, enhancing chondrocyte viability, inhibiting apoptosis, and reducing OA-related inflammation [101]. Cinnamaldehyde (CA) has been shown to alleviate OA symptoms by modulating the PI3K/Akt/mTOR, NF-κB, and MMP pathways [102]. Sankaranarayanan et al. found that CA pre-treatment enhanced BMSC-exos’ ability to suppress pro-inflammatory signaling pathways, such as NF-κB and MAPK, thereby reducing inflammation in OA chondrocytes [103].

Limitations and future perspectives

Currently, there is no effective therapeutic strategy that can significantly slow the progression of OA and cartilage injury. However, exosomes have demonstrated considerable potential in treatment. Despite being an emerging therapeutic approach, exosome-based treatments still face several limitations. It is important to note that current isolation techniques (e.g., ultracentrifugation, size-exclusion chromatography) cannot completely separate exosomes from other extracellular vesicles (EVs), such as microvesicles (100–1000 nm), due to overlapping size ranges and shared biophysical properties. While exosomes are defined as 30–150 nm vesicles derived from multivesicular bodies, microvesicles are larger particles shed directly from the plasma membrane. This technical limitation implies that observed therapeutic effects in exosome studies may partially arise from co-isolated EVs. Future advancements in purification methods (e.g., immunoaffinity-based approaches targeting specific surface markers) are needed to definitively attribute functional outcomes to exosome-specific cargo. While exosomes exhibit tremendous therapeutic potential, their functional outcomes are highly dependent on experimental variables such as isolation methods (e.g., ultracentrifugation vs. size-exclusion chromatography), dosage, and administration routes. Standardization of these parameters across preclinical and clinical studies will be essential to advance exosome-based therapies for OA.

The mechanisms through which exosomes exert their effects on OA and cartilage injury involve multiple signaling pathways, which remain incompletely understood. Additionally, there is limited research on the potential side effects associated with exosome therapy. Healthy exosomes derived from various sources can target multiple signaling pathways, such as NF-κB and PI3K/Akt. However, due to the complexity of the human body and individual differences in clinical treatments (e.g., age, gender), these signaling pathways are not exclusively active in the diseased areas of OA or cartilage injury. Therefore, when targeting specific signaling pathways with exosomes, it is crucial to minimize off-target effects and ensure that exosomes remain localized to the diseased regions. This strategy will help mitigate unnecessary side effects. The same consideration applies to “detrimental exosomes.” Personalized treatments are essential, which include precise control over the source, mechanism of action, and dosage of exosomes. Exosomes primarily mediate intercellular communication by delivering miRNAs to target cells. These miRNAs can function as both tumor suppressors and promoters. Therefore, the therapeutic use of exosomes overexpressing specific miRNAs requires rigorous preclinical validation to exclude potential oncogenic risks [104]. In-depth studies on the specific mechanisms of each miRNA within particular miRNA families could mitigate the risks of tumor formation and other adverse effects. Furthermore, given the multifactorial nature of OA and the diverse functional roles of exosomes derived from different cell sources, combination therapy using multiple types of exosomes may offer a synergistic approach to address overlapping pathological pathways. For instance, pairing BMSC-exosomes (promoting chondrocyte proliferation and ECM synthesis) with ADSC-exosomes (enhancing autophagy and suppressing Wnt/β-catenin signaling) and M2 macrophage-derived exosomes (alleviating synovial inflammation) could simultaneously target cartilage regeneration, oxidative stress, and inflammatory responses. Similarly, combining CSPC-exosomes (enhancing chondrocyte viability) and DPSC-exosomes (ameliorating subchondral bone remodeling) might address both cartilage defects and structural joint damage in advanced OA. However, such strategies require rigorous preclinical validation to ensure compatibility, avoid conflicting signaling, and optimize dosage ratios. Hydrogel-based delivery systems (e.g., GMOCS or HAMA-GelMA) could further enhance the retention and sustained release of combined exosomes, maximizing therapeutic efficacy. Future studies should prioritize identifying complementary exosome combinations while excluding detrimental sources (e.g., OA-FLS-exosomes) and addressing potential safety concerns related to off-target effects.

Future research should also explore other components within exosomes, such as proteins, circRNAs, and lncRNAs. It is important to note that clinical trials involving exosomes are still limited, and the optimal concentration for their therapeutic use remains undetermined.While exosomes from various sources exhibit common therapeutic effects for OA and cartilage injury—such as promoting cartilage regeneration, inhibiting inflammation, and preventing ECM degradation—they also display distinct differences. Currently, there is a lack of comparative studies on the therapeutic effects of exosomes from different sources. Future comparative studies should standardize variables such as dosage, concentration, and sample characteristics to better define their similarities and differences, ultimately optimizing treatment strategies to minimize side effects and enhance efficacy. Recent research has shown that exosomes therapy is most effective in the early stages of OA, which may be related to the high potential for cartilage microenvironment repair; The weekly medication regimen, which reduces the risk of synovial injury, has significantly better efficacy than high-frequency treatment, and has a certain indicative effect on future clinical applications [105].

After injection into the body, exosomes are rapidly cleared from the bloodstream and tend to accumulate in organs such as the liver, lungs, spleen, and gastrointestinal tract [106]. Hydrogels, with their excellent biocompatibility, can effectively protect exosomes from rapid clearance within OA joints and provide a sustained release effect. Future research should focus on enhancing the exosome loading capacity and precisely controlling their release kinetics in OA joints with varying degrees of pathology. Given the complex structure of joints, particularly those affected by OA, 3D printing technology can be employed to construct hydrogels in various shapes to adapt to the joint environment, further improving the therapeutic efficacy of exosomes.Gene editing technology offers high precision, and selectively editing exosomes can significantly improve their ability to alleviate OA progression and repair cartilage injury. However, the current efficiency of gene editing remains relatively low. Future advancements in gene editing could involve modifying the parent cells that release exosomes to generate clonal cell lines with high exosome production potential, ensuring large-scale exosome production. Furthermore, pre-treating exosomes is an effective strategy to enhance their therapeutic effects on OA and cartilage injury. The development of various methods or substances for exosome pre-treatment will be a key focus for future research.Lastly, the current method of administering exosomes is primarily through injection, which may increase the risk of infection or exacerbate inflammation. Therefore, developing novel administration methods will be an important area of future research.

Author contributions

Zilong Yang and Zihao Deng wrote the main manuscript text and prepared figures and tables. Wentao Gao, Wei Zhang, Xuxuan Fan and Ke Su supervised manuscript.

Funding

This research was funded by the Natural Science Foundation of Hunan Province (Grant No.2025JJ70014).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.