Small extracellular vesicles: the origins, current status, future prospects, and applications

Abstract

Small extracellular vesicles (sEVs) are membrane-bound vesicles with a size of less than 200 nm, released by cells. Due to their relatively small molecular weight and ability to participate in intercellular communication, sEVs can serve not only as carriers of biomarkers for disease diagnosis but also as effective drug delivery agents. Furthermore, these vesicles are involved in regulating the onset and progression of various diseases, reflecting the physiological and functional states of cells. This paper introduces the classification of extracellular vesicles, with a focus on the extraction and identification of sEVs and their significant role in repair, diagnosis, and intercellular communication. Additionally, the paper addresses the engineering modification of sEVs to provide a reference for enhanced understanding and application.

Introduction

Extracellular vesicles (EVs) are nano-scale membrane structures released by cells, which are widely found in body fluids such as blood, saliva, and urine. EVs are the key medium of cell-to-cell communication, which regulates the function of receptor cells by transmitting biomolecules. It can be used as a liquid biopsy marker for early detection of diseases. In addition, EVs can also be used for tissue repair, nerve regeneration, and as a natural drug carrier [1]. This paper focuses on small extracellular vesicles (sEVs) smaller than 200 nm.

Classification of EVs

EVs can be further subdivided into multiple subgroups based on biological characteristics, namely exosomes, microvesicles, and apoptotic bodies [2]. But with the continuous advancement of research methods, EVs can also be divided into other types based on different classification methods, such as migrasomes, cancer exosomes and so on [3] (Fig. 1).

Fig. 1.

Different types of EVs A. Biogenesis of exosomes; B. Biogenesis of microvesicle; C. Biogenesis of apoptotic bodies; D. Biogenesis of other vesicles

Exosomes

Exosomes are currently the most studied type of EVs, and their functions are relatively well understood. Exosomes are tiny vesicles that can be secreted by most cells in the body. They have a diameter of about 30–150 nm [1]. When observed under the transmission electron microscope, it shows a cup-shaped vesicle with a double-layer membrane structure.

Exosomes originate from the endocytic pathway of cells. The plasma membrane contains various components and invaginates, forming early sorting endosomes (ESE). During the subsequent maturation, these ESEs can exchange substances with other organelles or fuse with each other to form late sorting endosomes (LSEs). Then LSEs invaginated for the second time, forming multivesicular bodies (MVBs). MVBs can be degraded by fusion with lysosomes or fused with plasma membrane of cells to release intraluminal vesicles (ILVs) containing foreign bodies into extracellular space [4]. During the formation of intraluminal vesicles, the sorting of contents into the vesicles relies on the endosomal sorting complexes required for transport (ESCRT) [5]. There are also non-classical pathways that do not rely on the ESCRT mechanism, such as Rab31 GTPase-mediated exosome biogenesis [6]. There are many proteins involved in exosome biogenesis, including ESCRT proteins, and other proteins (CD9, CD81, CD63, TSG101, and so on), which are also used as exosome markers [7].

Exosomes are distributed in almost all body fluids, such as human blood and urine. Exosomes transfer substances carried internally or on their surface from donor cells to recipient cells through mechanisms such as direct membrane fusion, ligand-mediated interactions, and various intracellular endocytic pathways, participating in immune responses, antigen presentation, tumor invasion, and many other processes [8]. Exosomes play an important role in disease diagnosis and treatment research because of their unique phospholipid bilayer structure, wide sources, diverse contents, and powerful intercellular signal transduction function.

Microvesicles (MVs)

MVs are larger vesicles with a particle size of about 100–1000 nm. MVs are formed by sprouting from the cell surface and splitting [9]. The formation of MVs involves many factors. Some MVs germinate directly from the cell membrane mediated by ARRDC1 and TSG101 proteins, and the biogenesis of MV may be ESCRT-dependent or independent [10]. Additionally, the budding of MVs involves Ca²⁺-dependent cytoskeletal remodeling, during which Ca²⁺levels significantly increase, leading to the activation of calpain, as well as changes in disordered enzymes, thereby driving membrane asymmetry and cytoskeletal remodeling, resulting in the budding of the plasma membrane to form MVs [11]. Rho Small GTPase and Rho-associated coiled-coil containing kinases (ROCK) are essential for MVs’ release. Wang et al. [12] found that the Rho small GTPases CDC42 is a convergence point for various regulatory signals in the process of microvesicle generation.

Like exosomes, MVs contain genetic material, proteins, and lipids among other cellular contents [13]. MVs can transfer functional genes and protein molecules to target cells, mediating intercellular communication and maintaining tissue homeostasis, while playing an important role in organ development and regeneration, circulatory system renewal, immune regulation, and cancer treatment.

Apoptotic bodies (ABs)

ABs are EVs released by apoptotic cells. The particle size distribution is 0.5–5 μm. When cells are subjected to adverse external stimuli, they trigger programmed death pathways leading to apoptosis. In order to maintain the integrity of the cell membrane and avoid the possible tissue damage caused by the leakage of its contents, apoptotic cells will undergo a specific division process, thus forming apoptotic cell fragments and Abs. This is a defense mechanism evolved by the body to maintain homeostasis [14].

The process of ABs formation mainly consists of three steps: membrane vesicle formation, apoptotic membrane protrusion, and division to form apoptotic bodies. After a cell undergoes apoptosis, it begins to deform, mediating membrane bubbling through actin contraction. The repeated process of bubbling and contraction in apoptotic cells leads to the production of Abs [15]. At the molecular level, kinases are thought to control apoptotic membrane breaching. Among them, ROCK1 is a key regulatory factor in this process [16]. Pannexin 1(PANX1) channel, Plexin B2 receptor, cytoskeleton network structure, and vesicle transport process are involved in regulating the formation of apoptotic membrane processes. PANX1 is an important negative regulator of apoptotic cell catabolism. In addition to ABs, smaller EVs known as apoptotic vesicles (ApoEVs, with a size distribution of 0.1–1 μm) are also produced during apoptosis [15]. ABs also contain DNA, RNA, proteins, and lipid molecules, similar to other types of EVs, suggesting that they also have functions in mediating intercellular communication, signal transduction exchanges, tissue development and regeneration, and disease genesis and therapy [17].

Migrasome

Migrasome is a membrane structure produced at the tips or intersections of contractile filaments generated at the tail during cell migration [18]. It is an organelle with a diameter of 1–3 μm, featuring a single-layer vesicle, which also contains many vesicles. Under scanning electron microscopy, it presents a pomegranate-like structure [19]. Such vesicles are dependent on cell migration for their existence. They appear on contractile filaments, the tips and crossings of tubular structures left behind by migrating cells. As the cell continues to migrate, the contractile filaments break, and the vesicles are released into the extracellular space or are phagocytosed by the next cell to arrive at that location. The process by which cells release their contents is called migracytosis [20]. Through this mechanism, it recruits cellular contents and various cytokines, and by releasing these signaling molecules, it influences the physiological and pathological states of surrounding cells or remains on the migration path of the migrating cell to act as a localization signal, integrating temporal and spatial information to mediate intercellular communication [21].

Currently, it has been proved that migrasomes exist in various cells, body fluids, and tissues [22], and their physiological functions have also been preliminarily studied during embryonic development. The molecular and cellular biological functions of migrasomes have been continuously revealed, confirming the role of migrasomes in biological processes with active cell migration, such as regulating vascular homeostasis, immune response, tissue regeneration, and tumor metastasis.

Oncosomes

Oncosomes are vesicles secreted by tumors, and they are important vehicles for transmitting information between cells [23]. These large vesicles are called large oncosomes. Similar to exosomes, these two-layer membrane structures of oncosomes can transport proteins (including cancer proteins), cytokines, enzymes, and lipids, thus realizing the regulation of target cell phenotype by cancer cells and promoting the occurrence of tumor metastasis [24, 25].

Exopher

The neuron cells of Caenorhabditis elegans can excrete a huge membrane vesicle called exopher, which is about 4 microns in diameter. These membrane vesicles are filled with aggregation proteins and damaged organelles (including mitochondria). Nematode nerve cells can eliminate harmful substances in cells through it [26]. A similar mechanism exists in the human body. Cardiac muscle cells excrete dysfunctional mitochondria and other components through a large extracellular vesicle exophers. These vesicles are ingested and removed by macrophages resident in the heart, thus maintaining the health of myocardial cells and heart function [27].

Current research on exophers is still in its early stages. Researchers are exploring whether similar processes exist in other tissues, including the brain, to maintain the health of specific cells.

As an essential cellular secretion, EVs play a significant role in cell-to-cell communication, disease diagnosis, and treatment. With ongoing research, our understanding of EVs is evolving, offering new insights and methodologies for their clinical application. sEVs, which are vesicles with a diameter of less than 200 nm, are the focus of this paper. It primarily discusses sEVs, encompassing exosomes, and highlights the research progress in sEVs extraction and identification technologies, as well as their role in disease diagnosis and treatment.

Separation and extraction of sEVs

At present, the main methods for separating and extracting sEVs include centrifugation, ultrafiltration, polymer sedimentation, and so on [28]. Each separation and extraction method is based on the physical characteristics of sEVs, so there are differences in the quantity and purity of sEVs, having advantages and disadvantages, respectively (Table 1). At present, differential ultra-centrifugation is the most widely used method for separating and extracting sEVs, and it is considered the gold standard for sEVs extraction [1]. It is impossible to obtain pure sEVs by only one method. In order to effectively improve the purity and yield, different separation and extraction methods can be combined.

Table 1.

Extractions of small extracellular vesicles

Methods	Principle	Advantage	Disadvantage	Ref.
Ultracentrifugation	Separation is carried out alternately by low-speed and high-speed centrifugation.	1.Gold standard. 2.The extractable sample size is large.	1.Time consuming. 2.High-speed will damage sEVs.	[29]
Density gradient centrifugation	Ultracentrifugation makes sEV gather in a specific density class.	High purity.	1.Time consuming. 2.Repeated centrifugation will damage sEVs.	[30]
Polymer sedimentation	The solubility and dispersibility of sEVs are changed by hydrophobic polymer, so that sEVs can be precipitated under relatively low centrifugal force.	1.Easy to use. 2.Low cost.	1.Miscellaneous protein pollution is much. 2.Mechanical force or chemical additives can damage sEVs.	[31]
Ultrafiltration	Solvents and small molecular substances are filtered, while relatively large molecular substances are trapped on the ultrafiltration membrane.	1.Simple and fast. 2.Low cost. 3.Good portability.	1.Impurities of the same size are difficult to distinguish. 2.Low separation efficiency. 3.The purity is low.	[32]
Immunoaffinity	Separation of sEVs with magnetic beads coated with anti-marker antibodies.	1.High specificity. 2.High purity.	1.The efficiency and the recovery rate are low. 2.The biological activity of sEVs is susceptible to pH and salt concentration.	[33]
Microfluidic Platform	Particles with different characteristics are separated by different degrees of stress in different force fields.	1.Fast. 2.Low cost. 3.Easy automation.	1.Lack of standardization of technology. 2.Lack of large-scale clinical sample testing.	[34, 35]

Identification of sEVs

Although sEVs are difficult to detect because of the complex interfering components and small volume of body fluids. However, with the continuous progress of signal amplification technology, new functional materials, and sensor performance, it is now possible to achieve ultra-sensitive detection of sEVs, which helps us to better understand sEVs (Table 2).

Table 2.

Advantages and disadvantages of sEVs identification methods

Technology	Specific method	Superiority	Disadvantaged	Ref.
Electron Microscope (EM)	Scanning electron microscopy or transmission electron microscopy	1. 1.It can directly observe the structure and morphology.	1. Sample pretreatment is complex and demanding. 2. Unable to accurately measure sample concentration after pretreatment.	[36]
Nanoparticle Flow Cytometry (NanoFCM)	Vesicles expressing specific protein were sorted using fluorescent antigen-antibody reactions.	1. Fast, Qualcomm volume. 2. The size and volume of the particles can be analyzed. 3. The influence of the contained impurities can be avoided.	1. Expensive equipment 2. Time-consuming	[37]
Dynamic light scattering (DLS)	Measuring fluctuation data of light intensity	1. High measurement sensitivity with lower limit of 10 nm. 2. Simple sample preparation and fast measurement speed.	1. The light intensity fluctuation signal of large particles will mask the light intensity fluctuation signal of small particles. It is not suitable for the measurement of complex samples with different sizes. 2. Unable to measure sample concentration.	[38]
Nanoparticle Tracking Analysis (NTA)	Measuring fluctuation data of light intensity	1. The nanoparticle can be directly and real-timely observed. 2. The mono-dispersity and poly-dispersity particles are more accurate.	The source of sEVs cannot be determined and may include fragments of membranes and other cells, or lipoprotein complexes.	[39]
Atomic Force Microscope (AFM)	Detecting an atomic interaction force between the sample surface and the element	1. Abundance, morphology, biomechanics and bio-molecular composition of sEVs can be characterized.	1. Expensive instrument 2. Small imaging range	[40]
Surface-enhanced Raman spectroscopy (SERS)	Detecting the scattered photons with the frequency changed due to the interaction between the incident light and the medium molecules	1. High sensitivity 2. High selectivity 3. Fast analysis	1. Reliable calibration model is needed and result processing is complex.	[41]
Protein determination	Antigen-antibody reaction	1. Simple operation.	1. Overall mean method, free protein cannot be distinguished. 2. Time-consuming	[42]

Functional applications of sEVs

The research on sEVs is increasing day by day, and sEVs show important therapeutic potential in the field of cancer diagnosis and treatment. This is mainly due to the stability and non-immunogenicity of sEVs during delivery and treatment. sEVs can be produced by many types of cells. Different types of sEVs play a role in pathological processes such as cancer progression and inflammatory reaction. By analyzing sEVs in body fluids, we can understand the situation of diseases, thus providing suggestions for early diagnosis and intervention of diseases. At the same time, the heterogeneity of sEVs also brings some new opportunities for its application in drug delivery. According to the differences in composition and function of different types of sEVs, we can choose the appropriate sEVs type for specific diseases and treatment goals.

Tissue damage repair

Tissue repair is an extremely important part of clinical treatment, accompanied by regeneration, proliferation, differentiation, and migration of tissue cells; regulation of inflammation level; return of reactive oxygen species (ROS) content to homeostasis, and swelling and healing of the injured surface. sEVs can promote tissue repair and regeneration by promoting cell proliferation and migration, inhibiting apoptosis, and regulating inflammatory and immune responses, while mesenchymal stem cell(MSC)-derived sEVs have shown superior effects than sEVs from other cell sources [43].

Bone, cartilage repair and regeneration

Bone, cartilage repair and regeneration depend on the synergistic regulation of multiple cells and signal molecules. Endogenous repair and exogenous transplantation based on MSCs are new strategies for bone regeneration. However, under the influence of the diseased microenvironment, the regenerative effect it can exert is limited. sEVs, as a cell-free alternative to cell transplantation, have gradually come into people’s view with their ability to regulate the microenvironment and their regenerative potential [44]. The potential of sEVs to regulate the microenvironment and promote regeneration is gradually coming into view (Fig. 2).

Fig. 2.

Bone and cartilage repair of sEVs from different stem cells

Bone marrow mesenchymal stem cell-derived sEVs (BMSC-sEVs)

By preventing chondrocyte apoptosis, lowering extracellular matrix breakdown, and encouraging angiogenesis, BMSC-sEVs aid in bone healing and cartilage repair [45]. Liang et al. [46] pointed out that sEVs derived from low-dose dimethyl oxaloacetylglycine-pretreated BMSC-sEVs activate the AKT/mTOR pathway to stimulate angiogenesis in HUVECs, thereby promoting bone regeneration. In an osteoarthritis (OA) model, it was found that miR-326 released by BMSC-sEVs targeting to histone deacetylase-3 to activate STAT1/NF-κB p65 pathway and inhibit the focal death of chondrocytes and cartilage [47]. A similar mechanism was seen with miR-361-5p in BMSCs-sEVs, which attenuated chondrocyte damage by targeting Asp-Glu-Ala-Asp-box polypeptide 20 and inhibiting the NF-κB signal pathway [48].

Adipose stem cell-derived sEVs(ADSC - sEVs)

It has been demonstrated that ADSC-sEVs promote cartilage formation in periosteal cells by upregulating the expression of miR-145 and miR-221 in the cells [49]. By triggering the Wnt3a/β-catenin signaling pathway, ADSC-sEVs encourage osteogenic differentiation of BMSCs, which is beneficial for bone regeneration and repair [50]. MiR-130a-3p in sEVs from osteogenic differentiated ADSCs can regulate the proliferation and osteogenic differentiation of ADSCs by mediating the SIRT7/Wnt/β-catenin axis, which is an effective inducer of bone repair [51]. Similarly, kartogenin (KGN) has been studied by Xie et al. [52] and have also shown that KGN-pretreated ADSC-sEVs can enhance the chondrogenic differentiation of ADSCs by promoting cell proliferation and migration while inhibiting apoptosis. In an OA model, miR-376c-3p in ADSC-sEVs attenuated OA-induced chondrocyte degradation and synovial fibrosis by inhibiting the WNT-β-catenin pathway through targeting WNT3 or WNT9a [53]. OA-induced chondrocyte degradation and synovial fibrosis. In addition, miR-429 in ADSC sEVs not only promoted cartilage value-addition but also ameliorated cartilage damage by targeting FEZ2 and promoting autophagy [54].

Umbilical cord mesenchymal stem cell-derived sEVs (UCMSC- sEVs)

In terms of bone and cartilage repair, there have been previous studies confirming the effective role of UCMSC-sEVs in promoting cartilage damage repair [55]. The results of these studies are summarized as follows. Firstly, in terms of inflammation, UCMSC-sEVs exerted anti-inflammatory properties in cartilage injury and promoted the expression of chondrogenic gene (Col2a1) [56]. Some RNAs also play a reparative role. Long non-coding RNA H19, which is highly expressed in sEVs, promotes chondrocyte regeneration and repair through the miR-29b-3p/FoxO3 axis [57]. Zhang et al. [58] found that UCMSC-sEVs overexpressing miR-21 accelerated the process of bone regeneration by promoting angiogenesis through the notched gene 1/cellular transmembrane ligand 4/vascular endothelial growth factor 4 signal pathway. In femoral head necrosis, UCMSC-sEVs delivered miR-21-5p to target SRY-box transcription factor 5 and negatively regulate its expression to enhance angiogenesis and osteogenesis in vivo [59]. In biomedical engineering, sEVs have been combined with composite hydrogels to promote bone regeneration and repair [58, 60]. This new strategy shows great potential for future clinical applications.

Dental pulp stem cell-derived sEVs (DPSC- sEVs)

DPSC-sEVs have been shown to inhibit transient receptor potential vanilloid 4-mediated osteoclast activation, effectively ameliorating aberrant subchondral bone remodeling and inhibiting bone resorption and osteosclerosis [61]. DPSC-sEVs was also able to influence the osteogenic capacity of other stem cells, as DPSC-sEVs were efficiently taken up by jaw bone marrow-derived MSCs (JB-MSCs), which significantly promoted the expression of osteogenic genes and the osteogenic differentiation capacity of JB-MSCs [62]. DPSC-sEVs promoted the proliferation, migration, and osteogenesis of periodontal ligament stem cells in vitro [63]. In terms of resistance to inflammation, DPSC-sEVs inhibited the IL-6/JAK2/STAT3 signal pathway to regulate inflammation during acute inflammatory stress [63]. DPSC-sEVs also reduce the M1 polarization of macrophages through ROS-MAPK-NFκB P65 signaling pathway, showing anti-inflammatory characteristics [64].

Endothelial progenitor cell-derived sEVs(EPC-sEVs)

Distraction osteogenesis (DO) stimulates the potential of tissue regeneration by cutting the bone, thus prolonging the bone. EPC is closely related to the DO process. Jia et al. [65] evaluated the effect of EPC-sEVs on osteogenesis in a rat tibia DO model and found that transplantation of EPC-sEVs significantly accelerated the formation of healing tissues and ossification compared with transplantation of EPCs. These results indicated that EPC-sEVs could significantly accelerate bone formation during DO in rats. The in vitro results also showed that EPC-sEVs significantly increased the proliferation, migration, and angiogenesis of human umbilical vein endothelial cells (HUVECs) in a miR-126-dependent manner. It suggests that EPC-sEVs promote bone regeneration by stimulating angiogenesis.

Bone and cartilage injury is a common joint disease, which is relatively difficult to treat because of its special anatomical position. Conservative treatments, such as rest, physical therapy, and drug therapy, can alleviate symptoms, but they cannot be fundamentally repaired. Although surgical treatment can partially repair cartilage, it is risky, expensive, and takes a long time to recover after operations. Stem cell therapy, especially sEVs released by stem cells, as an innovative cell-free therapy strategy, shows great potential in promoting cartilage repair and regeneration. sEVs have the advantages of promoting cartilage repair, reducing inflammatory reactions, protecting joint structure, enhancing regeneration ability, and high safety. Among the existing MSCs sources, the bone marrow needed by BMSCs may come from individuals of different ages, and their age has a great influence on their biological activities [66]. The source of umbilical cord is safe and prolific, and there is no ethical problem. Compared with other MSCs, UCMSCs have a stronger ability to promote angiogenesis, which is helpful for rapid tissue reconstruction during bone regeneration. Compared with placenta, bone marrow, and other tissues, fat is more abundant, easy to extract, less traumatic to donors, self-transplantation, and high safety, and is also considered an ideal source for clinical application of MSCs. However, it should be noted that the application of sEVs in cartilage repair is still in the research stage, and its long-term effect and safety need further clinical trials to verify and evaluate.

Wound healing

Wound healing is a significant and complex process. Challenges of chronic wound healing include oxidative damage, inflammatory reactions, local infections, and decreased angiogenesis. MSC-sEVs accelerate wound healing and reduce scarring by promoting the proliferation of skin cells, inhibiting the release of inflammatory factors and promoting angiogenesis and maturation [67] (Fig. 3).

Fig. 3.

Wound repair effects of sEVs from different stem cells

BMSC-sEVs

In terms of inflammation regulation, BMSCs-sEVs can promote M2 macrophage polarization and anti-inflammatory factor expression, alleviating the inflammatory milieu and reducing muscle contusion [68]. Angiogenesis is an important stage. New sEVs with special treatment can activate PI3K/Akt and ERK1/2 signal pathways. MiR-21-5p contained in it can promote angiogenesis and fibroblast proliferation by targeted regulation of SPRY2 [69]. Cellular proliferation also provides essential conditions for wound healing. Li et al. [70] reported that the binding of long non-coding RNA H19 in BMSC-sEVs to miRNA-152-3p resulted in the up-regulation of phosphatase tension proteins, which promoted the proliferation and migration of fibroblasts. It was discovered that the transfer of miR-4645-5p by BMSC-sEVs deactivated AKT-mTORC1 signaling in keratinocytes, hence promoting wound healing and activating keratinocyte autophagy, proliferation [71]. Jiang et al. [72] showed that BMSCs-sEVs reduced the expression of transforming growth factor 1 and promoted wound healing by inhibiting the transforming growth factor β/Smad signal pathway while promoting the secretion of transforming growth factor 3 and reducing scar formation. In addition, Shen et al. [73] found that BMSC-sEVs could accelerate wound healing by up-regulating miR-93-3p to regulate the APAF1 axis.

ADSC-sEVs

In skin and wound repair, it has been found that ADSC-sEVs can show better ability to promote wound healing compared to BMSC-sEVs [74]. The presence of inflammation is a risk factor for wound healing. CCL1 is a typical inflammatory factor, and its expression is detrimental to wound healing. MiR-19b in ADSC-sEVs promotes skin wound healing by targeting CCL1 to regulate the TGF-β pathway [75]. In addition to miRNA, GAS5, a long non-coding RNA contained in ADSC-sEVs, also inhibits inflammation [76]. Different pathways also play important roles in inflammation suppression. ADSCs-sEVs inhibit the PI3K/AKT/mTOR signal pathway, thereby promoting mitochondrial autophagy, reducing inflammation, and inhibiting fibrosis [77]. They can also exert anti-inflammatory effects by targeting the ROCK1/PTEN pathway [78]. Unlike BMSC-sEVs, which are more proliferative, ADSC-sEVs exhibit a greater effect on angiogenesis [79]. The ability of MSC-sEVs to promote enhanced blood vessel formation and angiogenesis is critical to the wound healing process. MicroRNAs (miR-132 and miR-146a) in sEVs upregulate the expression of pro-angiogenic genes and exert pro-angiogenic effects [78]. ADSC-sEVs significantly promoted cell proliferation, migration, and angiogenesis, which correlated with the EGR-1/lncRNA-SENCR /DKC1/VEGF-A axis [80]. The ADSC-sEVs fraction also regulates angiogenesis through the miR-146a-5p/JAZF1 axis [81]. In another signal pathway, ADSC-sEVs accelerate angiogenesis through the HSP90/LRP1/AKT signal pathway [82]. In addition, hypoxic preconditioning has been used as a tool for the regulation of angiogenesis, and hypoxic adipose stem cell-derived sEVs carrying a high abundance of USP22 promote angiogenesis and cutaneous wound healing through upregulation of lncRNA H19 [83]. The combination of hypoxia-treated adipose stem cell sEVs and hydrogel can significantly promote wound healing [84] that provides a new strategy for research. Cell proliferation, migration, proteoglycan synthesis, and the production of type 1 and type 3 procollagen are essential features of wound healing. ADSC-sEVs promotes wound healing by regulating the activation of the Akt/HIF-1α signal pathway to facilitate the proliferation and migration of human keratinocytes cells [85].

UCMSC-sEVs

UCMSC-sEVs improve the wound healing process by attenuating the inflammatory response. Interleukins are increased in UCMSC-sEVs, in which IL-4 can inhibit the expression of pro-inflammatory cytokines [86]. In addition, UCMSC-sEVs can also reduce the pro-inflammatory ability of M1 macrophages, while the anti-inflammatory ability of M2a and M2c macrophages increases [86]. UCMSC-sEVs significantly enhance human corneal epithelial cell function and attenuate apoptosis and inflammation through activation of the autophagy signal pathway AMPK-mTOR-ULK1 [87]. UCMSC-sEVs also promote angiogenesis during wound healing through various mechanisms. UCMSCs regulate oxidative stress injury in endothelial cells through sEVs, promoting angiogenesis and accelerating skin wound healing [88]. UCMSCs-sEVs contain angiopoietin-2 (Ang-2), which promotes angiogenesis through the action of Ang-2, thereby healing wounds [89]. UCMSC sEVs also promote cell proliferation and inhibit scar formation during skin injury repair. hUCMSC-derived sEVs have been shown to increase fibroblast migration and proliferation in vitro, thereby promoting wound healing [90]. MiR-21 in hUCMSC-sEVs inhibits PTEN and activates the PI3K/Akt pathway, which promotes human corneal epithelial cell proliferation and wound repair [91]. In addition, hypoxia plays a positive role in wound repair. MiR-125b in sEVs derived from hypoxic UCMSCs can promote cell proliferation and migration through targeted inhibition of the TP53INP1 pathway in endothelial cells, thus promoting wound healing [92].

DPSC-sEVs

In wound repair, DPSC-sEVs has also shown potential for application. DPSC-sEVs accelerate skin wound healing by enhancing the angiogenic properties of HUVECs through the Cdc42/p38 MAPK signal pathway [93]. Hypoxia treatment often positively affects its effects, and hypoxia pretreated DPSCs-sEVs enhanced angiogenic potential by upregulating lysyl oxidase-like 2 [94].

EPC-sEVs

In wound repair, EPC-sEVs can regulate the function of vascular endothelial cells and play an important role in vascular injury repair. Zhao et al. [95] ‘s study showed that EPC-sEVs could promote angiogenesis through the hsa_circ_0093884/miR-145/SIRT1 axis. sEVs from endothelial progenitor cells can also repair injured vascular endothelial cells through the Bcl2/Bax/Caspase-3 pathway and enhance endothelial cell function [96]. It was also found that endothelial progenitor cell-derived sEVs containing miR-221-3p alleviated diabetic ulcers in a diabetic mouse model [97]. HIPK2 as a direct target of miR-221-3p may become a new therapeutic direction. EPC-sEVs can also down-regulate PPARG by increasing the expression of miR-182-5p in vitro, which promotes cell proliferation, migration, and wound healing. wound healing [98]. In addition, sEVs of EPCs with high miR-126-3p expression were able to target and inhibit the expression of the PIK3R2 gene by regulating PI3KR2/SPRED1 signal and inhibiting diabetic endothelial cell death [99].

The process of skin wound healing is complicated. Some refractory wounds often cause a slow or stagnant healing process due to a local blood circulation disorder and a lack of new blood vessels. These wounds often require longer treatment and are prone to recurrence. MSC-sEVs have low immunogenicity and high safety and can promote wound healing by promoting angiogenesis, accelerating the proliferation and migration of skin cells, and inhibiting excessive inflammatory reactions and scar formation. In addition, compared with traditional cell transplantation, sEVs therapy have the advantages of convenient administration and time saving. It can be administered by intravenous injection, local injection, and other ways without complicated cell culture and amplification. This makes the sEVs treatment more flexible and convenient, and is suitable for the treatment of different parts and types of wounds. ADSC-sEVs have been proved to regulate cell proliferation, migration, and apoptosis, as well as regulate immune and inflammatory reactions and promote angiogenesis. BMSC-sEVs is also involved. However, the abundance of adipose-derived stem cells is almost 500 times that of the same number of bones, and the sampling method is simple and safe [100]. After local transplantation, UCMSC-sEVs can aggregate to the ischemic area of the wound, secrete cytokines at the same time, promote the regeneration of vascular smooth muscle cells and vascular reconstruction, and create a good environment for wound repair. In addition to the sEVs from MSCs found in umbilical cord, bone marrow, and fat, other sEVs from MSCs, such as endothelium and gingiva, have also been proved to play a role in wound repair. A large number of studies have proved that sEVs have the advantages of multi-target treatment in wound surface. However, at present, many experimental designs are mostly to study the overall mechanism of sEVs, and there are few further studies on the inclusion, which leads to the unclear specific mechanism and target cell target.

Nerve repair

In terms of neurological repair, MSC-sEVs have the functions of protecting neuronal function by promoting axonal regeneration and protecting the integrity of the blood-spinal cord barrier, which provides ideas for the treatment of a large number of neurodegenerative diseases (Fig. 4).

Fig. 4.

Nerve repair effect of sEVs from different stem cells

BMSC-sEVs

In terms of resisting inflammation, BMSC-sEVs can reduce inflammation by regulating the M1/M2 phenotype of microglia [101]. Pretreated BMSC-sEVs also act on macrophages and participate in regulating inflammation through the TSG-6/NF-κB/NLRP3 signaling pathway [102]. In an ischemic stroke model, it was found that ZFAS1, a long-chain non-coding RNA carried by BMSC-sEVs, had an inhibitory effect on microRNA-15a-5p, which reduced the level of oxidative stress in ischemic stroke, decreased apoptosis, and inhibited inflammation [103]. sEVs also play a role in promoting nerve regeneration. MiR-17-92 in BMSC-sEVs significantly reduced the expression of PTEN protein in rat brain tissues, leading to an increase in the expression of Akt signal pathway and downstream proteins, which in turn promoted axonal regeneration [104]. Another study also found that the combination of BMSC-sEVs and hydrogels can promote axonal regeneration by increasing the expression of neurofilament and synaptophysin [105].

ADSC-sEVs

In nerve repair, ADSCs are recognized as a very promising stem cell therapeutic material due to their potent neurotrophic, immunomodulatory effects and safety [106]. The following are some of the most promising stem cell therapeutic materials. ADSC-sEVs have similar effects as ADSC, and ADSC-sEVs are rich in bioactive molecules that promote nerve repair through the delivery of neurotrophic factors, alleviate inflammation by delivering inflammation-modulating molecules, and promote the survival of injured cells in neural tissues [107]. Inhibition of Inflammation Inhibition of inflammation is a prerequisite for nerve repair to occur, and ADSC-sEVs can prevent inflammation and reduce nerve damage by inhibiting microglia activation through inhibition of NF-kB and MAPK pathways [108]. sEVs pretreated by hypoxia can improve nerve injury by delivering circRNAs and changing the polarization of microglia [109]. The stable delivery of neurotrophic factors provides the basis for nerve repair. Engineering sEVs derived from adipose stem cells can efficiently deliver and translate neurotrophic factors to recipient cells to enhance nerve regeneration [110].

UCMSC-sEVs

In terms of nerve repair, UCMSC-sEVs promote skin nerve regeneration by recruiting fibroblasts and stimulating their secretion of nerve growth factor (NGF) [111]. They can also control inflammation by inhibiting microglia proliferation [112]. Some pre-treatments also affect the repair effect, and infarct pre-treated umbilical cord MSC-sEVs can promote nerve function recovery by enhancing vascular endothelial remodeling [113].

DPSC-sEVs

In peripheral nerve injury, DPSC-sEVs can promote myelin regeneration by inhibiting autophagy in endogenous Schwann cells via the miR-122-5p/P53 pathway [114] in peripheral nerve injury. In addition, DPSC-sEVs attenuated brain injury in mice with cerebral ischemia-reperfusion injury, and it was speculated that it might exert anti-inflammatory effects by inhibiting the HMGB1/TLR4/MyD88/NF-κB pathway [115].

EPC-sEVs

In neural repair, EPC-sEVs ameliorated neuronal apoptosis by decreasing bcl2l11 expression in oxygen and glucose deprivation induced neuronal apoptosis through delivery of miR-221-3p [116]. A new study found that EPC-sEVs affected macrophage phenotype through the SOCS3/JAK2/STAT3 axis, inhibited neuroinflammation, and ameliorated spinal cord injury [117]. With the deepening of EPC-sEVs-related research, the discovery of the mechanism of EPC-sEVs promoting vascular regeneration has received more and more widespread attention. However, the specific downstream molecular mechanism of EPC-sEVs for vascular regeneration is still not fully understood, and an effective method for the purification and amplification of EPC-sEVs has not been established.

Nerve injury has always been a big problem that puzzles clinical treatment. The limited regeneration ability of neurons, microenvironment obstacles, and the complexity of nerve connection leading to the traditional conventional treatment methods cannot play an effective radical effect. Stem cells can proliferate indefinitely and have obvious pluripotency and can differentiate into different cell types in the body. sEVs derived from stem cells have the ability of tissue repair, regeneration, and protection similar to stem cells, and can cross the blood-brain barrier. These sEVs can reduce nerve injury by promoting the growth and repair of neurons and reducing inflammatory reactions. sEVs from different stem cells also have differences in nerve repair. For example, due to the diversity of contents, UCMSC-sEVs are considered to have greater therapeutic potential and can provide a large number of bioactive molecules to promote repair. In contrast, the yield of BMSC-sEVs is low, but it still has a certain nerve repair effect. sEVs promote nerve repair through various channels.

Disease prediction and diagnosis

sEVs, as an important medium of intercellular communication, carry a variety of bioactive molecules from mother cells. These molecules can reflect the physiological state, pathological changes, and interaction with other cells of mother cells. They provide rich biological information for disease diagnosis.

Compared with the detection of free molecules in body fluids, the molecules in sEVs are encapsulated and protected by vesicles, so they are more stable and have higher detection sensitivity than those in body fluids. Free molecules in body fluids are easily affected by degradation, dilution, and other factors, which leads to the decrease of detection sensitivity. Secondly, the molecules in sEVs are derived from specific cell types, so they have higher specificity. Free molecules may come from many cell types or tissues, and their specificity is relatively low. sEVs carry many types of biomolecules, which can provide more comprehensive disease information. Free molecules can usually only provide a single biomarker of information, and the amount of information is limited.

Compared with tissue biopsy, sEVs detection is noninvasive and convenient.

To sum up, molecular detection of sEVs has unique advantages and potential in disease diagnosis. However, sEVs detection still faces some technical challenges and limitations, and related technical methods need to be further improved and optimized.

Tumors

Early diagnosis and treatment of tumors can significantly improve the survival rate of patients, and the lack of appropriate diagnostic markers for most tumors causes patients to miss the best treatment time. Therefore, the exploration and application of tumor diagnostic markers are of great significance as a new target for tumor liquid biopsy. sEVs contain much tumor bioinformation, which has important application value for early diagnosis and prognostic evaluation of tumors (Table 3).

Table 3.

Diagnostic role of small extracellular vesicles in cancer diseases

Detecting disease	Biomarker	Source	Patient expression	Diagnostic yield	Ref.
Lung cancer	circRAPGEF5	Serum	Raise	AUC = 0.847, Sensitivity:64.90%, Specificity:95.60%	[118]
	circ__0048856	Serum	Raise	AUC = 0.943. Sensitivity:0.880 Specificity:0.800	[119]
	miR-200b-3p miR-3124-5p miR-92b-5p	Serum	miR-92b-5p ↓ miR-200b-3p↑ miR-3124-5p ↑	AUC = 0.93	[120]
	miR-1290 miR-29c-3p	Plasma	miR-1290↑ miR-29c-3p↓	AUC1 = 0.934 AUC2 = 0.868	[121]
Gastric cancer	miR-10401-3p miR-1255b-5p miR-6736-5p	Serum	Decline	AUC1 = 0.8333, AUC2 = 0.8316 AUC3 = 0.8142	[122]
	miR-31 miR-192 miR-375	Plasma	Raise	AUC1 = 0.734 AUC2 = 0.812 AUC3 = 0.745	[123]
Liver cancer	miR-10b-5p	Serum	Raise	AUC = 0.934, Sensitivity:90.7%; Specificity:75.0%.	[124]
Breast cancer	miR-18a-3p	Serum	Raise	AUC = 0.83	[125]
	miR-1246	Plasma	Raise	Sensitivity:97.30% Specificity:93.94%	[126]
	H19	Serum	Raise	Sensitivity:87.0% Specificity:70.6%	[127]
	annexin A2	Serum	Raise	AUC = 1.00	[128]
Colorectal cancer	miR-3937	Plasma	Raise	AUC = 0.827	[129]
	miR-874	Serum	Decline	AUC = 0.818 Sensitivity:80.8% Specificity:78.6%	[130]
	circGAPVD1	Plasma	Raise	AUC = 0.7662, Sensitivity:75.64% Specificity:71.79%	[131]

Other non-oncological diseases

As an important carrier of intercellular information transmission, small extracellular vesicles carry biochemical molecules such as proteins, nucleic acids and lipids, which not only show great potential in the diagnosis of tumor diseases but also have broad application prospects in the diagnosis of non-tumor diseases (Table 4).

Table 4.

Diagnostic role of small extracellular vesicles in non-cancer diseases

Affiliation system	Detecting disease	Biomarker	Source	Patient expression	Diagnostic yield	Ref.
Respiratory system	Allergic asthma	miR-223 miR-142a	Alveolar lavage fluid	miR-223↑ miR-142a↑	/	[132]
Nervous system	Parkinson’s disease	miR-23b-3p	Plasma	Raise	AUC = 0.8086	[133]
		miR-223-3p	Serum	Raise	Sensitivity:71.1% Specificity:79.9% AUC = 0.771	[134]
	Senile dementia	miR-135a	Serum	Raise	/	[135]
		Amyloid β-protein and miR-384	Plasma	Raise	AUC1 = 0.969 AUC2 = 0.906	[136]
Urinary system	IgA Kidney Disease	lncRNA G21551	Plasma	Decline	/	[137]
		miR-451a	Urine	Raise	AUC = 0.805	[138]
	Diabetic nephropathy	Circ_0008529	Urine	Raise	Sensitivity:70.21% Specificity:86.67%	[139]
Endocrine system	Type 1 diabetes	miR-146a-5p miR-223-3p miR-21-5p	Plasma	Raise	AUC1 = 0.978 AUC2 = 0.96 AUC3 = 0.809	[140]

sEVs is rich in disease-related proteins, lipids, and nucleic acids, which can be used as biomarkers for disease. In the process of detecting sEVs in vitro, researchers use various technical means to collect and process samples, separate and purify sEVs, and identify and analyze them later (Fig. 5). Through in-depth study and continuous optimization of the detection process, we will be able to deeply explore the specific mechanism of sEVs in intercellular communication and hopefully open up new ways and methods for the diagnosis and treatment of diseases.

Fig. 5.

Summary of methods that sEVs can be used for in vitro diagnosis

Intercellular signal transduction and disease development

sEVs regulate cancer cell growth and proliferation

Extracellular vesicles can activate immune cells, triggering the immune response and influencing tumor development. Tumor-derived extracellular vesicles are reported to be rich in PD-L1, and PD-L1-carrying extracellular vesicles inhibit T-cell activation, which is a major regulator of tumor development. Zhou et al. [141] found that small extracellular vesicles miR-424-5p could increase the secretion of pro-inflammatory cytokines IFN-γ, TNF-α, and IL-6, and down-regulate the expression of PD-L1 in cancer patients, which in turn promotes apoptosis of tumor cells. Serum sEVs target LIM and SH3 protein-1 to prevent cell invasion and proliferation and trigger cell death [142]. This will inhibit the growth in diffuse large B-cell lymphoma. In addition, miRNAs can also induce apoptosis, and miR-122-5p in gastric cancer cells can induce apoptosis and inhibit the proliferation of gastric cancer cells by decreasing the level of G protein-coupled receptor kinase 2 interacting protein 1 in gastric cancer cells [143].

Tumor cell growth and proliferation are important factors influencing the oncogenic process. For example, small extracellular vesicle miR-183-5p inhibits the expression of PPP2CA and increases the secretion of pro-inflammatory cytokines, which in turn promotes tumor progression in breast cancer models [144]. Tumor-derived sEVs can modulate the interaction between tumor cells and immune cells. Renal clear cell carcinoma-derived sEVs promote differentiation of M2 macrophages, which in turn promotes renal cell carcinoma progression [145]. Huang et al. [146] found that CircSAFB2 can mediate macrophage polarization through the mir-620/JAK1/STAT3 axis. This will lead to RCC cell immune escape and then promote the progress of renal cell carcinoma. Zhao et al. [147] found that the sEVs miR-424 inhibits T cell and dendritic cell activation by suppressing the CD28-CD80/86 co-stimulatory pathway in T cells and dendritic cells, leading to immune evasion and suppression of the body’s anti-tumor immunity, which in turn promotes cancer progression.

The above results suggest that regulating the expression of specific molecules in small extracellular vesicles has an effect on cancer cell growth.

sEVs regulate cancer cell metastasis

A significant proportion of cancer patients develop tumor metastasis. Therefore, inhibition of tumor metastasis is a crucial step in the treatment of malignant tumors, and sEVs are closely related to it. CBX7 gene is an oncogene, and CBX7 deficiency is closely associated with the invasiveness and poor prognosis of ovarian cancer. Zhang et al. [148] found through in vitro studies that adipocyte-derived miR-421 in sEVs could target and down-regulate CBX7 gene expression and promote the metastatic potential of ovarian cancer cells. MiR-552-5p in sEVs promotes tumorigenesis and migration in gastric cancer through downregulation of PTEN, tumor suppressor TOB1, and caspase-3 levels via the PTEN/TOB1 axis [149]. MiR-374a-5p was absorbed by gastric cancer cells, upregulating the levels of gastric cancer cell adhesion molecules by targeting HAPLN1 and promoting gastric cancer cell migration [150]. Inhibiting cancer cell metastasis through sEVs is an important direction for future exploration. Wang et al. [151] found that sEVs miR-363-5p were able to down-regulate the expression of PDGFB and thus inhibit breast cancer cell metastasis.

The role of sEVs in cancer migration has been widely recognized, and its research not only helps to reveal the molecular mechanism of cancer migration, but also provides important clues for developing new cancer treatment methods.

Engineering sEVs

sEVs have significant advantages and potential in the process of repair. Small extracellular vesicles, as natural nanocarriers, can stably deliver bio-active molecules without toxicity or immunogenicity. In addition, it also has a high degree of targeting, which can accurately reach the damaged parts and play a role in repairing.

But sEVs are easily influenced by the parent cells and target cells in terms of targeted delivery efficiency. Additionally, due to the immune system’s defense mechanisms and the filtering effects of the liver and kidneys, these vesicles may be rapidly cleared before reaching the target tissue. At the same time, small extracellular vesicles have complex compositions and release mechanisms, and their biological distribution and transport are affected by various factors, making their tissue specificity not obvious. Compared to natural sEVs, engineered sEVs employ bioengineering techniques, resulting in a longer circulation half-life and enhanced targeting ability. They also reduce systemic drug distribution and adverse reactions, thereby improving therapeutic effects [152].

The engineering strategy of sEVs mainly includes two aspects: surface modification and drug loading of sEVs.

Drug loading of sEVs

There are two categories of drug loading strategies employed by sEVs. One is the exogenous drug delivery route, where sEVs are first extracted and purified before therapeutic drugs are encapsulated within them. This method offers the advantage of simple preparation [153]. Currently, commonly used exogenous drug delivery methods include electroporation, co-incubation, ultrasound, chemical transfection, and repeated freezing and thawing [154]. The other is the endogenous drug-carrying pathway, which involves introducing target molecules into donor cells through genetic engineering technology or co-incubation. These molecules are then packaged into vesicles, and sEVs are secreted from the donor cells. Finally, drug-loaded vesicles are recovered through separation and purification [155]. Each method has its own set of advantages and disadvantages. In the actual experimental process, it is necessary to choose the appropriate method based on the characteristics of the drugs being loaded. (Table 5).

Table 5.

sEVs drug loading methods and their advantages and disadvantages

Loading method	Method	Loading types	Advantage	Disadvantage	Ref.
Exogenous	electroporation	Small molecule material	High load rate	The sEVs aggregate easily and have low membrane stability	[156]
	co-incubation	Drugs, small molecular substances	1. The method is simple. 2. The sEVs structure is complete.	Low efficiency	[157]
	ultrasound	Drugs, and small molecular substances	High loading rate	The membrane structure is vulnerable	[153]
	Chemical transfection	Peptides, proteins, and nucleic acids	The method is simple.	There is a genetic risk	[154]
	freeze thawing	Drugs, and small molecular substances	High loading efficiency	The sEVs can cluster easily	[155]
Endogenous	Genetic engineering, co-incubation	Low-grade cytotoxic drugs	Retains the integrity and functionality of the sEVs.	Complex operation and high preparation cost.	[153]

Surface modification of sEVs

The loading process inside sEVs is the primary strategy for achieving high efficiency targeted delivery, while surface modification is the main method to improve vesicle targeted delivery efficiency. The main goal of surface engineering is to endow vesicles with targeting specificity and increase their circulation time in the blood.

Surface modification techniques include genetic modification and chemical modification. Genetic modification refers to the genetic recombination of the parent cells that produce sEVs, enabling them to produce sEVs with modified peptide segments or proteins, while chemical modification refers to the use of chemical methods to modify the surface of sEVs after their separation and purification (Fig. 6).

Fig. 6.

Surface modification technology of sEVs

Genetic engineering

Genetic engineering modification of sEVs is an effective method to functionalize their surface. This process often involves plasmid transfection, guiding the gene sequence of a protein or peptide to fuse with the gene sequence of the selected membrane protein, resulting in sEVs that contain the target protein. Studies have found that the sEVs membrane is composed of transmembrane proteins, such as LAMP-2B, GPI, CD63, CD9, and CD81, which can bind to targeting adaptors to enhance the specific delivery of sEVs [158]. Currently, LAMP-2B is the most widely used sEVs surface protein [159]. Research has involved fusing Her2 to the N-terminus of LAMP-2B, allowing the expression of the Her2-LAMP2 fusion protein on the sEVs surface, which accelerates targeted cell uptake through epidermal growth factor receptor-mediated endocytosis in tumor cells, effectively targeting colorectal cancer cells [160].

Additionally, CD63, CD9, and CD81 can also be linked to targeting sequences or peptide segments. Research has indicated that transfecting HEK293T cells with the plasmid pc-DNA encoding CD9-Hu R has resulted in the creation of novel CD9-Hu R surface-functionalized sEVs, which enhance the targeting specificity and drug loading efficiency of sEVs [161].

Covalent modification

Click chemistry

Click chemistry is a prevalent method for surface modification. The reaction’s benefits include high yield and exceptional functional group selectivity [162]. The most frequent modification involves the addition of an alkyne group. In a particular study, superparamagnetic iron oxide nanoparticles and curcumin were initially loaded into sEVs, and subsequently, the sEVs’ membrane was conjugated with a neural ciliary protein-1 targeting peptide via click chemistry. This resulted in sEVs that target glioblastoma and possess both imaging and therapeutic capabilities [163].

Metabolic engineering

The parental cells of sEVs can also be equipped with new components and functions via metabolic engineering. Metabolic engineering primarily entails altering the inherent synthesis and modification processes within the cells, thus eliminating the need for genetic manipulation. For instance, when ManNAz is introduced into the culture medium of parental cells, azide-containing sugars are incorporated into the glycans on the surface of the sEVs secreted by these cells [164], thereby functionalizing sEVs through bio-orthogonal chemistry.

Covalent modification is anticipated to alter the activity sites of proteins expressed on the surface of sEVs. Nevertheless, these modifications necessitate further experimental evaluation.

Noncovalent modification

Electrostatic effect

The electrostatic effect operates on the principle of attraction between opposite charges; as long as conditions permit, cations can attract as many anions as possible around them. Researchers have applied the same charge to the surface of sEVs, thereby enhancing their targeting efficiency towards biofilms with different charges [165]. Studies have shown that surface modification of sEVs can be achieved through a simple mixture of native sEVs and cationic amylopectin, as well as the electrostatic interaction between the two substances. The modified sEVs have the ability to target hepatic cells expressing sialic acid glycoprotein receptors, thereby targeting damaged liver tissue and enhancing the therapeutic effect of sEVs [158].

Hydrophobic effect/membrane fusion

The hydrophobic effect describes the tendency of globular proteins to fold in a water medium, burying hydrophobic residues within the molecule. The presence of a phospholipid bilayer structure on the surface allows sEVs to interact with liposomes via hydrophobic forces. Research indicates that surface modifications of sEVs can be accomplished through the hydrophobic insertion of lipids conjugated with folic acid. Such modifications enable sEVs to be tailored for targeted delivery of doxorubicin to tumors, significantly improving their surface functionality [159].

Adapter-receptor binding

An aptamer is a single-stranded DNA or RNA molecule that has been developed through exponential enrichment technology. It possesses the capability to bind specifically to various components, including peptides, proteins, cells, or tissues, with a high degree of targeting specificity and affinity [166]. Comparable to antibody molecules, aptamers frequently serve as targeting ligands characterized by their high affinity and specificity. Researchers have engineered an octapeptide, cNGQGEQc, as a targeting ligand for non-small cell lung cancer (NSCLC). Integrin α3β1 is overexpressed on NSCLC cells, and this ligand exhibits specific binding to integrin α3β1 [167].

Peptide anchoring of membrane CD63

Besides genetic engineering modification, CD63 can also participate in noncovalent modification. CD63 molecule is a transmembrane protein highly expressed on the surface of extracellular vesicles, and the CP05 peptide has a high affinity for the CD63 molecule. Since there are many CD63 molecules on the surface of extracellular vesicles, it can serve as an anchoring point between targeting peptides or therapeutic molecules and the surface of extracellular vesicles. Researchers have identified the extracellular vesicle anchoring peptide CP05 for extracellular vesicle surface markers, enabling efficient loading of various biopharmaceuticals into extracellular vesicles using CP05 [168]. This has been applied in targeted therapy for tumor immunotherapy and Duchenne muscular dystrophy, establishing a functionalized surface technology for extracellular vesicles.

Application of engineering sEVs

Engineering sEVs brings new insights to the development of the next generation drug delivery system. They shows great potential in cancer treatment, cardiovascular diseases and tissue regeneration and repair. For example, in NSCLC, its exploration may bring new directions for clinical treatment.

For the treatment of NSCLC, Engineering sEVs can be used as carriers of chemotherapy drugs. Researchers [162] combined the macrophage-derived sEVs loaded with paclitaxel with AA-PEG to target the sigma receptor overexpressed in lung cancer cells. The results show that this new drug preparation not only has higher bioavailability and drug loading capacity but also can be enriched in tumor cells, which provides a new strategy for the treatment of NSCLC.

Furthermore, in the treatment of NSCLC, inhibitors such as programmed cell death 1 (PD-1) and programmed cell death ligand-1 (PD-L1) can reverse the immunosuppressive state and enhance anti-tumor immunity by activating anti-tumor T cells. However, the efficacy of immune checkpoint inhibitors in some patients is poor, which may be associated with the emergence of an inhibitory tumor microenvironment (TME) [169]. M2 tumor-associated macrophages are the primary subgroup of inhibitory immune cells and play a role in the formation of TME alongside regulatory T cells. Additionally, the innate immunity mediated by cyclic guanosine monophosphate-adenosine monophosphate synthase/stimulator of interferon gene (STING) presents a promising application prospect in tumor treatment. The TME can be remodeled through the engineering of small extracellular vesicles (sEVs). Cheng et al. [170] fused sEVs from M1 macrophages with sEVs from genetically engineered tumor cells expressing CD47. They successfully designed and synthesized genetically engineered hybrid sEVs, encapsulating them with DNA-damaging chemotherapeutic drugs SN38 and MnO₂ to activate the STING pathway, thereby achieving cancer immunotherapy. The aforementioned research indicates that improving the TME through the engineering of sEVs is a viable strategy in anti-tumor immunotherapy.

A multi-functional lung cancer treatment strategy can be developed by loading targeted molecules on sEVs and modifying and optimizing them. This strategy can be used not only for the regulation of immune cell activity but also for targeting specific cells or tumors and drug loading and delivery. These innovative application prospects provide new ideas for the treatment. However, its clinical transformation still faces many problems and challenges, including the heterogeneity of engineering sEVs, whether it will cause the initial biological changes of sEVs after transformation, and how to standardize batch production and storage.

Conclusion

As an important medium of intercellular communication, sEVs have shown attractive application prospects in many fields, such as disease diagnosis, drug delivery, tissue repair.

In terms of tissue repair, previous studies mostly focused on the repair function of MSCs, but MSCs have some problems, such as self-apoptosis, uncontrollable preparation quality, possible immune response to allograft, and ethical debate [43]. sEVs therapy may avoid the above shortcomings. However, despite the remarkable potential of sEVs in tissue repair, the safety risk of their clinical application still needs to be systematically evaluated. Firstly, the heterogeneity of mother cell origin may lead to sEVs carrying potential pathological factors. For example, sEVs released in a pathological state may transmit some oncogenic proteins [171]. Secondly, the risk of immunogenicity cannot be ignored. Although sEVs are naturally low in immunogenicity, sEVs from different sources may activate immunity through surface major histocompatibility complex molecules [164]. Finally, the standardization of production technology needs to be solved urgently, and the existing separation technology cannot achieve industrial production. Different culture conditions will also affect the contents of sEVs [171].

In addition, engineering sEVs are also demonstrating its ability to deliver drugs. At present, many clinical trials are evaluating its safety and effectiveness (such as NCT03608031, NCT05043181, NCT04747574, etc.). Although engineering sEVs show broad prospects as a drug delivery system, its clinical transformation still faces challenges. First of all, sEVs have not solved the problem of large-scale production and lack an industrial solution that takes into account the yield, purity, and economy [1]. Secondly, the drug loading efficiency of sEVs restricts its therapeutic effect, especially the loading of macromolecular drugs [165]. The existing methods are difficult to achieve the balance between high drug loading and drug loading stability. In addition, the biological safety of engineered sEVs has not been fully clarified, including the immunogenicity changes caused by engineering modification and the risk of accumulation in the body after long-term use [7]. The blank of the supervision system further delays the industrialization process. At present, the standardized quality control systems of sEVs products has not been established in the world [165].

In conclusion, it is necessary to establish a safety evaluation framework for sEVs, including donor cell screening, pathogen detection, immunogenicity testing, and long-term in vivo tracking, and promote the formulation of standardized production and regulatory guidelines. In the future, through interdisciplinary cooperation and in-depth research, we are expected to solve various challenges faced by sEVs in the process of preparation, analysis, and application.

Abbreviations

sEVs

Small extracellular vesicles

EVs

Extracellular vesicles

ESE

Early sorting endosome

LSE

Late sorting endosome

MVB

Multivesicular body

ILV

Intraluminal vesicle

ESCRT

Endosomal sorting complexes required for transport

MV

Microvesicle

ROCK

Rho-associated coiled-coil containing kinases

AB

Apoptotic Body

PANX1

Pannexin 1

EM

Electron microscope

NanoFCM

Nanoparticle flow cytometry

DLS

Dynamic light scattering

NTA

Nanoparticle tracking analysis

AFM

Atomic force microscope

SERS

Surface-enhanced raman spectroscopy

MSC

Mesenchymal stem cell

BMSC

Bone marrow mesenchymal stem cell

ADSC

Adipose derived stem cell

UCMSC

Umbilical cord mesenchymal stem cell

DPSC

Dental pulp stem cell

EPC

Endothelial progenitor cell

OA

Osteoarthritis

KGN

Kartogenin

JB-MSC

Jaw bone marrow-derived MSC

DO

Distraction osteogenesis

Ang-2

Angiopoietin-2

NGF

Nerve growth factor

NSCLC

Non-small cell lung cancer

PD-1

Programmed cell death 1

PD-L1

Programmed cell death ligand-1

STING

Stimulator of interferon gene

TME

Tumor microenvironment

Author contributions

XY Zhou, J Huang and DQ Zhang have contributed equally to this work: conception and design, collection and/or assembly of data, data analysis and interpretation, visualization, manuscript writing, and final approval of the manuscript. ZY Qian: collection and/or assembly of data. X Zuo and YX Sun: conception and design, financial support, administrative support, provision of study material, supervision, collection and/or assembly of data, data analysis and interpretation, visualization, manuscript writing, and final approval of the manuscript. All authors reviewed the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by National Natural Science Youth Foundation of China (Grant 82202618, 82401900), Top Talent Support Program for young and middle-aged people of Wuxi Health Committee(BJ2023104), Medical Education Collaborative Innovation Fund of Jiangsu University (No.JDY2022017), Scientific Research Program of Wuxi Health Commission (No. Q202342). All figures are created with BioRender.com.

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.