Extracellular vesicles: a promising cell-free therapy for cartilage repair

Few effective therapies for cartilage repair have been found as cartilage has a low regenerative capacity. Extracellular vesicles (EVs), including exosomes, are produced by cells and contain bioactive components such as nucleic acids, proteins, lipids and other metabolites that have potential for treating cartilage injuries. Challenges like the difficulty in standardizing targeted therapy have prevented EVs from being used frequently as a treatment option. In this review we present current studies, mechanisms and delivery strategies of EVs. Additionally, we describe the challenges and future directions of EVs as therapeutic agents for cartilage repair.

EVs may interact with recipient cells via contact, membrane fusion or endocytosis as a mediator of intercellular communication. In contact, the membrane ligand on the EVs' surface interacts with the receptor in the cell membrane of the recipient cell and generates cell signals [42,43]. In this case, EVs will not be internalized by the target cells. Another mechanism of interaction between EVs and cell targets is membrane fusion, in which the EV membrane consists of a lipid bilayer that fuses with the cell membrane and releases the cargo into the cytosol. A study by Parolini et al. showed that exosome uptake by melanoma cells happened via fusion and increased at low pH [44]. The most commonly studied mechanism of EV internalization is the endocytosis pathway, in which the EVs enter the recipient cell by phagocytosis [45], macropinocytosis [46,47], clathrin-mediated endocytosis [46], caveolin-mediated endocytosis [48] or lipid raft-mediated endocytosis [49]. It is possible that a particular type of EV has more than one mechanism when interacting with a recipient cell. Because EVs have therapeutic potential, their interactions with cells should be studied to develop targeted therapies.
Stem cell-derived EVs are at least as good as, if not better than, stem cells when applied for therapeutic purposes. Overall, they demonstrate less negative potential. A study by Mohammed et al. showed that exosomes from ADSCs are more effective as an adjuvant treatment in dentistry for scaling and root planing [50]. Another study performed by Zavatti et al. compared AFSCs and their exosomes in animal models of osteoarthritis and found that AFSC-derived exosomes were more effective in treating cartilage damage than the cells [51]. When compared with cell-based therapy using stem cells, EVs have some distinct advantages. For example, EVs have simpler storage needs, allow allogeneic transplantation due to lack of MHC I and MHC II antigens and are less vulnerable to damage at the injury site; it is also possible they can reach a higher circulating dose than bigger cells [23]. Because EVs are non-self-replicating, the possibility of iatrogenic tumor growth is reduced.
MSC exosomes are effective in supporting cartilage repair and regeneration [52]. The application of EVs in cartilage repair has been investigated in vivo in many different animal models with a variety of concentrations (Table 1). Small animal models such as mice, rats and rabbits are used in current research on cartilage regeneration. However, more research with larger test animals is required to be clinically appropriate. To improve treatment efficacy, it will also be necessary to standardize the EVs dose calculation.

Sources of EVs
EVs can be obtained from almost all bodily fluids. Parental cell selection should account for the desired therapeutic function of the resultant EVs. EVs can be a therapeutic drug or they can act as a delivery vehicle for a specific drug. For instance, EVs isolated from bovine milk can be utilized to deliver exogenous hsa-miR148a-3p in RNAbased treatment [76]. Although cells from an injury site can produce EVs, they usually do not produce therapeutic EVs; rather, EVs from the cells in a cartilage injury site tend to aggravate the damage [77,78]. However, EVs from therapeutic cells at the same injury site can maintain chondrocyte homeostasis [78]. These therapeutic cells can be differentiated cells or stem cells. Ma et al. found that EVs released by chondrocytes induced proliferation and differentiation of umbilical cord MSCs into chondrocytes, indicating that EVs promote cartilage regeneration [79].
EVs from blood components are advantageous because blood collection is less invasive and safer than adipose tissue or bone marrow collection. Otahal et al. studied the use of EVs derived from blood for treatment of osteoarthritis [80]. These investigators demonstrated that EVs isolated from citrate-anticoagulated platelet-rich plasma-enhanced desirable chondrogenic gene expression changes in osteoarthritis and prevented proinflammatory cytokine release [80]. Another study by Liu et al. showed that EVs derived from platelet-rich plasma promoted proliferation and inhibited chondrocyte apoptosis via the Wnt/β-catenin signaling pathway [81].
Stem cells such as induced pluripotent stem cells and MSC have potential in tissue repair. As cell-based therapy, stem cells can be applied directly or serve as a cell source for tissue engineering. Various types of stem cells produce functional EVs with advantages for cartilage repair. EVs derived from AFSCs can repair cartilage damage in correlation with their TGF-β content [51]. MSCs, which are non-hematopoietic stem cells, are present in various body tissues and are multipotent. The therapeutic effect of MSCs depends on a paracrine mechanism mediated by their EVs [12]. EVs isolated from ADSCs prevent cartilage degeneration and attenuate the progression of osteoarthritis by modulating immune reactivity [20]. Another study, using EVs from BMSCs, showed that BMSCderived EVs promote ECM synthesis and protect against cartilage damage [65]. MSC-derived exosomes promote proliferation, migration and ECM synthesis, which helps to attenuate apoptosis and modulates immune reactivity in osteochondral defects [82].

Mechanism of EVs in cartilage regeneration
Common causes of cartilage damage are trauma and degenerative disease. In articular cartilage, damage often results from violent injury, chronic inflammatory disease or degenerative joint diseases [85]. According to Schulze-Tanzil [86], traumatic cartilage injury causes chondrocyte and synoviocyte stress that leads to inflammation, degradation of the cartilage's ECM and apoptosis. Inflammation in cartilage is often caused by inflammatory cytokines including IL-1β, TNF-α, IL-6, IL-15, IL-17 and IL-18 [87]. Cartilage damage has an effect on the quantity of chondrocytes by triggering cell death [88] and inducing chondrocyte apoptosis [89]. Additionally, injured ECM degrades faster than it can be synthesized. Understanding the pathogenesis of cartilage injury can help scientists develop specific therapies, including therapy for cartilage damage directed to overcome the results of homeostatic changes. Inflammation in cartilage tends to increase pain and disease progression. Inflammation is a phenomenon in traumatic cartilage injury [86]. If the damage is caused by degenerative disease, such as in the intervertebral disc (IVD), inflammation is caused by an imbalance of the ECM catabolic and anabolic pathways [90]. Treatment using EVs can inhibit the inflammatory cascade. A study by Zhang et al. indicated that MSC-derived exosomes reduced IL-1β [74]. IL-1β, as the most important proinflammatory mediator, is also involved in inflammatory responses 10 during disc degeneration [91]. Another study showed that MSC-derived exosomes slowed the progression of IVD degeneration by suppressing inflammatory mediators and NLRP3 inflammasome activation [70]. Suppressing the NLRP3 pathway can prevent pyroptosis, a programmed cell death triggered by proinflammatory signals. Exosomal miR-410 from MSC inhibits the NLRP3 pathway and regulates pyroptosis [92]. Recently, treatments for osteoarthritis have focused on macrophage polarization. Macrophages are immune cells found in the synovial lining that complete a variety of tasks depending on their subtype; they may be proinflammatory (M1) or antiinflammatory (M2) [93]. A study by Zhang et al. demonstrated that exosomes isolated from BMSCs reduced inflammation by regulating macrophage polarization, inhibiting M1 macrophage production and promoting M2 macrophage generation [60]. The purpose of therapy in cartilage repair is to restore the chondrocyte ECM to its original state. ECM components, like collagen type II and proteoglycan, play a role in regulating chondrocyte functions. Therapy can be aimed at synthesizing those specific ECMs. He et al. reported that BMSC-derived exosomes upregulated collagen type II production and downregulated MMP13 protein expression in an animal model of osteoarthritis [65]. Another study found that BMSC-derived exosomes promoted ECM production in degenerated nucleus pulposus cells in vitro [94]. Thus it appears that EVs play defined roles in recovering cartilage ECM.
Chondrocytes play a role in cartilage regeneration by synthesizing ECM, despite their low number in normal cartilage. Because cartilage injury further diminishes the number of chondrocytes, a therapeutic method is required to maintain their population. The number of chondrocytes can be maintained by several mechanisms, one of which is to differentiate stem cells into chondrocytes. EVs from nucleus pulposus cells induce differentiation of MSC into nucleus pulposus-like cells by inhibiting the Notch1 pathway [95]. One component of EVs, miRNA, can also target the pathway in chondrogenic differentiation. Li et al. showed that miR-8485 from exosomal chondrocytes activated the Wnt/β-catenin pathways to stimulate differentiation of BMSCs into chondrocytes [96]. In addition to chondrogenic differentiation of stem cells, increasing chondrocyte proliferation in the injury site improves cartilage regeneration. Some studies have shown that EV cargo -for instance, miRNA [61,62,97 -can promote chondrocyte proliferation.
Chondrocyte loss that is caused by apoptosis and autophagy can be overcome using EV therapy. Cheng et al. reported that miR-21 in MSC-derived exosomes prevented nucleus pulposus cell apoptosis [98]. Similarly, studies have shown the utility of EVs in the inhibition of apoptosis induced by endoplasmic reticulum in IVD degeneration [99,100]. Inhibiting apoptosis and increasing cell proliferation in cartilage repair will maintain the number of chondrocytes.
The role of EVs in cartilage treatment is to restore cartilage homeostasis by maintaining the number of chondrocytes and balancing the metabolism of specific ECMs (Figure 1). EVs deliver functional cargo, such as miRNA, for cartilage regeneration (Table 2). Additionally, EV cargo modulates inflammation at the injury site.

Delivery strategies of EVs in cartilage repair
It is necessary to design suitable EVs that are functional therapeutic agents and deliver them to enhance their effectiveness and efficiency in treating damage. For cartilage repair, EVs can be obtained from bodily fluid, tissue or cell culture and delivered by local or intravenous administration. Woo et al. isolated EVs from ADSCs and used them to treat osteoarthritis in rats [20]. They found that ADSC-derived EVs enhanced proliferation and migration of chondrocytes, regulated the expression of catabolic and anticatabolic factors and inhibited macrophage infiltration into synovium, thereby modulating immune reactivity [20]. He et al. also demonstrated that EVs derived from BMSCs and injected intra-articularly stimulated cartilage regeneration and ECM synthesis, as well as reducing knee discomfort, in an osteoarthritis model [65].

Engineering cells & their EVs
Engineering parental cells or their EVs can enhance the effectiveness of EVs in therapy. Changes in the cell microenvironment -such as the pretreatment medium, oxygen level and mechanical stimulation -influence cell behavior and affect EV characteristics and functions. For instance, kartogenin has been used to improve stem cell proliferation and chondrogenic differentiation in cartilage regeneration [102]. An examination of EVs from cells pretreated with kartogenin revealed a paracrine change of the cells in chondrogenesis. Liu et al. reported that EVs derived from kartogenin-preconditioned BMSCs enhanced chondral matrix synthesis and reduced degradation; thus this approach appears more effective for cartilage repair than the use of EVs from BMSCs without pretreatment with kartogenin [103]. A study using infrapatellar fat pad MSCs showed a similar result: EVs pretreated with kartogenin more effectively promoted articular cartilage defect repair [56]. Thus by altering the cellular environment through the addition of chemical compounds to the cell culture medium, the efficiency of the resultant EVs is improved. Hypoxic preconditioning of stem cells also affects the efficiency of EVs. Hypoxic pretreatment of BMSCs enhances their release of EVs that increase proliferation, migration and apoptosis inhibition of chondrocytes through the miR-216a-5p/JAK2/STAT3 signaling pathway [62]. The cell microenvironment can also be modified through mechanical stimulation. A study by Yan et al. showed that mechanical stimulation using a rotary cell culture system enhanced the yield of EVs from umbilical cord MSC-derived EVs and found that EV function on 10.2144/fsoa-2021-0096 Future Sci. OA (2021) FSO774 future science group cartilage repair was enhanced through upregulation of lncRNA H19 [61]. Modification of parental cells can also be accomplished by genetic engineering; Thomas et al. successfully engineered L-cells with WNT3a and isolated EVs that were able to heal osteochondral defects [55]. Another target of modification for targeted therapy, besides parental cells, is the EV itself, via a method called post-secretion modification. The aim of engineering EVs is to make them a functional drug delivery system. The drug loaded in EVs can be a natural component of therapeutic cell-derived EVs or another chemical agent. Loading a drug into EVs increases its in vivo stability, circulation in blood, and cell targeting efficiency [104]. Combining EVs with drugs promoting cartilage regeneration, such as kartogenin, enhances their function. Post-secretion modification of EVs is more efficient than engineering parental cells to deliver drugs. For example, even though kartogenin is beneficial in cartilage regeneration, it has low water solubility. Xu et al. isolated EVs from dendritic cells and engineered them to be a delivery agent for kartogenin [63]. They showed that this treatment increased the effectiveness of synovial fluid-derived MSCs to differentiate into chondrocytes [63]. Post-secretion modification can also be performed on the EV surface by adding specific ligands. Engineering the natural surface increased targeting efficiency in vivo [105].
For the same dose, delivering EVs through intravenous administration is less effective than local administration in cartilage repair. The half-life of exosomes in blood circulation is about 2 min [106]; healing a cartilage injury requires more time due to the characteristics of cartilage. However, local administration methods such as intra-articular injection require frequent injections that make the patient uncomfortable. Combining EVs with biomaterials or scaffolds could reduce treatment frequency, as the biomaterial will ensure that the EVs remain at the defect site.

EVs embedded in biomaterials
The scaffold acts as a time-controlled delivery system for EVs in cartilage injury, trapping them at the injury site and periodically releasing them. The release of drugs or EVs from a scaffold can be caused by diffusion, polymer dissolution and degradation, or swelling [107]. Scaffolds are defined by their ability to retain EVs at the injury site, gradually release them into the matrix and integrate with the damaged tissue to promote surrounding cell migration [108].
Scaffolds for cartilage regeneration can be made from synthetic or natural materials. Some common synthetic polymer materials are poly(lactic-co-glycolic acid) and polymer of lactic acid [109]. Synthetic materials have the advantages of reproducibility, structure and customizable characteristics. However, synthetic materials are more expensive than natural ones and they have weak cell attachment [109]. Additionally, natural scaffolds such as collagen, fibroin and chitosan tend to be safer because of their biocompatibility and reduced toxicity. The drawbacks of natural scaffolds are their source-dependent mechanical and physical properties [110].
The scaffold form needed to trap EVs and maintain their release can be solid or hydrogel. Hydrogel, a hydrophilic polymer, is widely used in cartilage tissue engineering. Its mechanical behavior permits its use as an articular cartilage substitute [111]. Hydrogel can be fabricated from natural materials or synthetic polymers to mimic the natural ECM and will control the release of EVs embedded in it. Stem cell-derived EVs can be incorporated into a photoinduced crosslinking hydrogel to retain the exosomes inside and enhance cartilage repair [112]. Chen et al. showed that EV-impregnated scaffolds from the cartilage ECM and gelatin methacrylate hydrogel promoted cartilage regeneration [71]. Thus studies indicate that biomaterial has a significant role in the delivery of EVs for repairing cartilage damage. Further research should be conducted to explore the various biomaterials that may be used for EV delivery in cartilage repair and regeneration. Figure 2 summarizes therapeutic strategies utilizing EVs for cartilage repair. There are numerous delivery methods for EVs used to repair cartilage damage, with the primary goal being the restoration of cartilage homeostasis. The simplest method is to use naive EVs isolated from bodily tissue or cell cultures. However, EV parental cell types must be considered, as they affect EV bioactivity. By retaining EVs within the biomaterial and controlling their release, implanting EV-loaded biomaterials may be a way to enhance therapeutic effects. Furthermore, it does not require frequent administration when dosed appropriately.

Limitations
EVs, particularly exosomes, have potential in cell-free therapy for cartilage repair and regeneration. Numerous in vitro and in vivo studies have delineated the composition of EVs and their role in tissue repair. However, a search of clinicaltrials.gov gave only one result, which involved the use of platelet-rich plasma enriched with exosomes in the treatment of chronic low back pain [113]. It is critical to have appropriate identity and potency parameters when  studying EVs to ensure their quality control and reproducibility. Those studying EVs and their effects should refer to the International Society of Extracellular Vesicles' guidelines [25] to promote reproducibility. While evidence for the use of EVs in cartilage repair is convincing, several factors must be considered prior to initiating clinical trials. Larger animal models should be studied before EVs are used in the clinic. The examination of partial-thickness and full-thickness chondral repair, as well as osteochondral repair, is possible in large animal models with thicker articular cartilage [114]. Further investigation into the choice of EV parental cells and their maintenance is required due to the heterogeneity of EV content. As no single drug fits all diseases, targeted therapy is important. Another consideration is the optimization of large-scale production of EVs, because it is challenging to isolate EVs with high purity in high yields. The most common method for isolating EVs is ultracentrifugation, but this has limitations of low purity and EV aggregation [115]. Tangential flow filtration can be an alternative to achieve reproducible large-scale production [116]. Scale-up methods to produce EVs as a therapeutic agent for cartilage repair need standardization.
The choice of whether or not to engineer EVs for targeted cartilage therapy will depend in part on further research to guarantee their efficacy and safety. Proper EV dose and delivery strategies are also important. EVs wash out easily in the circulatory system, necessitating a higher dose or entrapment in biomaterial. Scaffold in the form of hydrogel is a good candidate as a delivery agent. While live cell transplantation is already widely used, EV-based therapy has a greater potential for repair due to the absence of cells. EV-loaded scaffolds can be adapted to the current surgical techniques applied to repair cartilage defects by implanting the EV-loaded scaffolds in the defect site. It is hoped that this procedure will eliminate the need for repeated operations by optimizing the EV dosage in the scaffold and will increase patient comfort. Additionally, when EVs are used therapeutically, such as in an articular cartilage injury, they can regenerate hyaline cartilage.
EV pharmacokinetics also needs to be considered for therapeutic development. Furthermore, it is necessary to standardize the quality of EVs as a product, such as storage conditions (e.g., temperature and expiration date). Although there are many challenges in the clinical application of EVs for cartilage repair, the evidence on the function of EVs in healing cartilage injury is promising. A better understanding of the potential of EVs in therapy and their greater accessibility may significantly reduce related healthcare costs.

Conclusion & future perspective
EVs, including exosomes, can be obtained from any cell source. Determination of parental cells and therapeutic strategies are important in making EV therapy effective and efficient. EV-based therapy has the potential to repair cartilage damage by maintaining cartilage homeostasis. To optimize the therapeutic effects of EVs, they can be engineered and loaded with biomaterials to control their release. Proper strategies will lead to an increased accessibility and effectiveness of EV therapy for cartilage repair. To be clinically applicable, the standardization of EV products must be considered to ensure their safety.

Executive summary
• Extracellular vesicles (EVs), including exosomes, have the potential to treat cartilage damage by restoring cartilage homeostasis. • Due to the heterogeneity of EV content, selection of parental cells and appropriate therapeutic strategies are important in targeted therapy. • Loading biomaterials into EVs optimizes their effectiveness in cartilage repair.
• Some challenges in large scale production of EVs need to be addressed to facilitate their clinical application.