Extracellular vesicles from mesenchymal stem/stromal cells as emerging tools in wound healing: mechanisms and therapeutic potential

Abstract

Wound healing in adult mammals is a regulated but imperfect process in which fibrotic repair dominates over proper regeneration. Mesenchymal stem/stromal cells (MSCs) are considered promising in regenerative medicine; however, evidence suggests that their benefits are primarily mediated by paracrine signaling, including the secretion of extracellular vesicles (EVs). We conducted a structured review to evaluate the potential of MSC-derived EVs (MSC-EVs) in wound management. PubMed was searched for original studies published between January 2015 and July 2025. Of 171 records, 19 met the inclusion criteria. The data included the MSC source, EV isolation, dosing, and effects on wound repair. The review is organized into six themes. First, MSC-EVs consistently promoted the proliferation and migration of keratinocytes, fibroblasts, and endothelial cells, enhancing wound closure in vitro and in vivo. Second, pro-angiogenic effects were evident, often mediated through the PI3K/AKT/mTOR/HIF-1α signaling pathway, although tissue-specific responses were observed. Third, MSC-EVs exhibit immunomodulatory activity by reducing pro-inflammatory cytokines, inducing M2 macrophage polarization, and improving outcomes in chronic and diabetic wound models. Fourth, in extracellular matrix (ECM) remodeling, MSC-EVs increased collagen synthesis, regulated metalloproteinases, and reduced scar formation. Fifth, strategies such as MSC preconditioning and hydrogel-based delivery enhanced EV stability, prolonged activity, and improved therapeutic efficacy. Finally, we assessed methodological challenges, including heterogeneity in EV isolation and characterization, inconsistent dosing, limited adherence to ISCT and ISEV standards, and a lack of GMP-compliant protocols. In summary, MSC-EVs emerge as multifaceted acellular therapeutics that influence key phases of wound repair, including cell activation, angiogenesis, immune modulation, and ECM remodeling. They hold promise for accelerating healing and reducing fibrosis. However, substantial barriers remain before clinical translation, particularly the standardization of EV preparation and regulatory compliance. Addressing these gaps is essential for advancing MSC-EVs into safe, effective, and scalable therapies for wound healing.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-025-04813-5.

Background

Wound healing is a dynamic, multistep biological process that aims to restore the structure and function of damaged tissue through tightly regulated interactions among various cell types and signaling molecules [1–3]. This process can follow two primary paths: regeneration, which restores tissue with the same cell types, and repair, which replaces damaged tissue with fibrotic scar tissue [4]. The path taken depends mainly on the type of tissue involved and the extent of the injury [5].

Interestingly, despite their evolutionary advancement, mammals exhibit limited regenerative capacity compared with other organisms [6, 7]. In adult mammals, proper regeneration is typically restricted to specific tissues and minor injuries. This limitation is linked to the high specialization of mammalian cells and the absence of totipotent cells [8, 9]. However, in early to midgestational fetal skin, wounds heal rapidly and without scarring. This scarless healing is associated with enhanced fibroblast migration, quicker re-epithelialization, and matrix (ECM) deposition than in adults [10]. These observations have inspired efforts to mimic fetal-like regeneration in adult tissues, particularly through emerging strategies in tissue engineering, stem cell biology, and extracellular vesicle (EV) research [11].

In adult mammalian wound healing, the process is generally divided into three overlapping phases: inflammation and hemostasis, proliferation, and remodeling [12]. Hemostasis begins with clot formation to prevent bleeding and initiates the growth factor signaling pathway. The proliferative phase involves the formation of granulation tissue, angiogenesis, and epithelialization. Finally, during remodeling, the ECM is reorganized, and granulation tissue matures [13]. While fibrosis is a regular part of wound healing, excessive or dysregulated fibrosis can lead to poor clinical outcomes [14]. It is estimated that nearly half of the deaths in developed countries involve fibrotic complications. Excessive ECM deposition, typically resulting from prolonged or dysregulated inflammation, can lead to scarring, organ dysfunction, and chronic pain [15].

Mesenchymal stem/stromal cells (MSCs) have garnered considerable interest due to their potential to enhance wound healing [16]. These multipotent cells possess the ability to self-renew and can differentiate into various cell types [17]. More importantly, they secrete a broad array of bioactive molecules, including cytokines, chemokines, and ECM proteins, that modulate inflammation, promote regeneration, and inhibit fibrotic pathways [18, 19]. Due to their ability to modulate immune responses and inhibit profibrotic signaling pathways, MSCs are promising candidates for enhancing wound repair and improving tissue regeneration outcomes.

A growing body of research now focuses on extracellular vesicles (EVs), which are nano- to microscale particles naturally released by cells, as key mediators of MSC function. EVs, as defined by the International Society for Extracellular Vesicles (MISEV 2018), are membrane-bound structures incapable of self-replication and carry diverse cargos, such as proteins, lipids, and nucleic acids [20, 21]. They can transfer their contents to recipient cells, thereby influencing their behavior and facilitating intercellular communication [22, 23]. Many studies do not fully comply with the Minimal Information for Studies of Extracellular Vesicles (MISEV 2018/2023) guidelines, which recommend the use of specific markers and functional assays to confirm EV identity and activity. The lack of consistent reporting on EV size, concentration, protein content, and bioactivity hampers comparability between studies and limits reproducibility.

EVs are generally categorized based on their size and biogenesis. Exosomes (30–150 nm) originate from the inward budding of endosomal membranes, resulting in multivesicular bodies that fuse with the plasma membrane to release exosomes [24, 25]. Microvesicles (100–1000 nm), on the other hand, bud directly from the plasma membrane and have shown promise as targeted drug delivery vehicles because of their natural targeting properties [26]. Apoptotic bodies (50 nm to 5 μm) are formed during programmed cell death and contain cellular organelles, DNA fragments, and histones (Fig. 1) [27].

Fig. 1.

Extracellular vesicles derived from mesenchymal stem/stromal cells. MSCs Mesenchymal Stem/Stromal Cells, MVB Multi-Vesicular Body

Given their ability to influence critical phases of the wound healing cascade, including proliferation, angiogenesis, immune modulation, and ECM remodeling, MSC-EVs are gaining momentum as novel, acellular therapeutic platforms in regenerative medicine (Fig. 2) [28].

Fig. 2.

Influence of extracellular vesicles derived from mesenchymal stem/stromal cells on factors related to wound healing. TNF-α Tumor Necrosis Factor alpha, IL-6 Interleukin-6, IL-8 Interleukin-8, IL-1b Interleukin-1 beta, IL-10 Interleukin-10, NF-κB Nuclear Factor kappa-light-chain-enhancer of activated B cells, Cas-8 Caspase-8, M1 Macrophages with a pro-inflammatory phenotype, M2 Macrophages with anti-inflammatory phenotype, VEGF Vascular Endothelial Growth Factor, Ang-1 Angiopoietin 1, Ang-2 Angiopoietin-2, MMPs matrix metalloproteinases, TGFβ-1 Transforming Growth Factor, COL1A1 Collagen Type I Alpha 1 Chain; COL1A2 Collagen Type I Alpha 2 Chain, COL3A1 Collagen Type III Alpha 1 Chain, α-SMA Alpha-Smooth Muscle Actin, FGF2 Fibroblast Growth Factor 2, ELN Elastin, PI3K-AKT-mTOR-HIF-1α Signalling pathway involved in regulation of angiogenesis and wound healing, ITCH/JUNB/IRE1α Signalling pathway

This review offers a critical and integrative synthesis of current preclinical evidence on extracellular vesicles derived from human mesenchymal stem/stromal cells (MSC-EVs) in the context of skin wound healing. We focus on their biological effects across key phases of tissue healing, including cellular proliferation and migration, angiogenesis, immunomodulation, and extracellular matrix remodeling, while also evaluating their methodological quality, compliance with research standards, and emerging strategies to increase EV efficacy. By highlighting both the therapeutic potential and current limitations of MSC EVs, this review aims to inform future research and support the development of safe, effective, and standardized cell-free therapies for wound healing.

Materials and methods

A literature search was performed in PubMed to identify preclinical studies evaluating the effects of human mesenchymal stem/stromal cell-derived extracellular vesicles (MSC-EVs) on wound healing. Publications from January 2015 to July 2025 were screened via the keywords “extracellular vesicles” and “mesenchymal stem/stromal cells” and “wound healing”. Only original research articles published in English were included. Studies that used nonhuman MSCs, had unclear descriptions of purification/concentration methods, or lacked adequate statistical information were excluded. Publications that did not focus on wound healing or lacked functional outcome data were also removed. Among the 171 records initially identified, 19 studies [29–47] met the inclusion criteria. The screening process is summarized in the PRISMA flow diagram (Fig. 3). In addition, a search of ClinicalTrials.gov using the same keyword did not yield any registered studies on MSC-EVs in wound healing.

Fig. 3.

PRISMA flow diagram of the studies’ screening and selection

From each study, data were extracted on the MSC source and culture, EV purification and characterization methods, and dosing strategies. The following biological effects on key aspects of wound healing were reported: proliferation, angiogenesis, immune modulation, and ECM remodeling. Particular attention was also given to whether studies adhered to internationally recognized EV and MSC research standards, including the ISEV (MISEV 2018/2023) and ISCT guidelines [20, 21, 48].

Results

Extracellular vesicle sources

Human MSC-derived extracellular vesicles (hMSC-EVs) were most commonly obtained from adipose tissue (AD-MSCs, n = 5) and the umbilical cord (UC-MSCs, n = 5), placenta (P-MSCs, n = 3), and Wharton’s jelly (WJ-MSCs, n = 2). Single studies reported EVs from umbilical cord blood (UCB-MSCs), bone marrow (BM-MSCs), hair follicles (HF-MSCs), and foreskins (FS-MSCs) (Fig. 4A).

Fig. 4.

MSC Culture and Characterization. A Sources of MSC-EVs; B GMP compliance; C Number of MSC passages; D Analysed MSC markers. AD-MSCs Adipose Tissue Mesenchymal Stem/Stromal Cells, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, P-MSCs Placental Mesenchymal Stem/Stromal Cells, WJ-MSCs Wharton’s Jelly Mesenchymal Stem/Stromal Cells; UCB-MSCs Umbilical Cord Blood Mesenchyma Stem/Stromal Cells, BM-MSCs Bone Marrow Stem/Stromal Cells, HF-MSCs Hair Follicle Mesenchymal Stem/Stromal Cells, FS-MSCs Foreskin Mesenchymal Stem/Stromal Cells, GMP Good Manufacturing Practise, HLA-DR Human Leukocyte Antigen- Antigen D Related

MSC culture conditions

In nearly all cases, MSCs were cultured under traditional conditions (5% CO₂, 37 °C) in standard media such as DMEM supplemented with fetal bovine serum (FBS) and antibiotics (n = 17) [29–45]. To reduce contamination by serum-derived EVs, some protocols used serum deprivation 6–72 h before EV purification (n = 6) [29, 35, 36, 38, 40, 43], EV-depleted serum by ultracentrifugation and/or 0.22 μm filtration (n = 7) [31–34, 37, 38, 41, 44], or commercial exosome-free serum (n = 3) [30, 39, 41]. These approaches limit the use of serum vesicles but do not remove xenogeneic components, making FBS use incompatible with GMP standards for clinical use.

The studies used conditioned medium from various passages, but most often from passages 3–5 (Fig. 4C), as cells at higher passages may lose their multipotential properties, affecting the quality and functionality of EVs.

Three of the studies focused on the use of hydrogels as carriers for EVs, highlighting their key role in tissue engineering and regenerative medicine [30, 36, 44]. In contrast, one study used a 3D spheroid cell culture model to optimize EV secretion, presenting an alternative approach to enhancing their therapeutic efficacy [46]. This methodological diversity demonstrates progress in the field, which is seeking new solutions for both the delivery and production of EVs.

Preconditioning of MSCs was explored in two studies: one used 3,2-dihydroxyflavone [31] to modify cellular secretome profiles, while the other applied biologically relevant stressors such as thrombin, lipopolysaccharide, H₂O₂, and hypoxia [47]. These strategies aimed to increase EV bioactivity by modulating the parent cell phenotype and signaling environment.

Among the 19 publications reviewed, 10 focused on confirming that the cells used in the study were mesenchymal stem/stromal cells [30, 31, 33–35, 37, 40, 42, 43, 45], according to ISCT guidelines [48] (Fig. 4D).

Only two studies reported compliance with GMP principles (Fig. 4B) [46, 47], and manufacturing protocols were not established in the remaining 17 cases, which raises doubts about the safety and reproducibility of the results and the establishment of validated release criteria for clinical application [49, 50].

EVs purification and concentration

According to MISEV guidelines (Supplementary Table 1), the choice of isolation method critically affects EV yield, purity, and biological relevance. In the analyzed studies (Fig. 5A), most studies employed size- and density-based separation, predominantly differential ultracentrifugation (n = 14) [31–39, 41, 43–45, 47], which is valued for its scalability but is prone to co-isolating nonvesicular material [51]. Precipitation (n = 3) and filter concentration (n = 5) were also used, alone or in combination, offering faster processing but lower specificity [29, 30, 40–42, 46]. Regardless of the method used, rigorous characterization is essential to ensure that the observed biological effects can be explicitly attributed to EVs [51].

Fig. 5.

Purification and characterization of EVs. A Analysed vesicular markers; B EV purification methods. LAMP1 Lysosomal Associated Membrane Protein 1, ALIX ALG-2 interacting protein X, TSG101 Tumor Susceptibility Gene 101 protein, GM130 Golgi Matrix Protein 130, APO1A Apolipoprotein A-I

EVs characterization

Characterization of EVs was typically performed via nanoparticle tracking analysis (NTA) to determine their concentration and size distribution and transmission electron microscopy (TEM) to assess their morphology. In addition, EV-specific marker expression was evaluated in accordance with the International Society for Extracellular Vesicles (ISEV) guidelines [20, 21].

According to the ISEV recommendations (Supplementary Table 1), at least one protein marker from the first three categories (membrane, cytoplasmic, contaminating) should be detected [20]. Despite these clear recommendations, only one study fully adhered to this standard, reporting CD63 as a marker from category 1, TSG101 from category 2, and albumin from category 3 [32]. (Fig. 5B). Accurate characterization of isolated MSC-EVs is a crucial step in understanding their molecular effects on various cells. This affects the comparability of results, as differences in the composition of MSC-EVs can lead to varying biological effects, as well as the reproducibility of studies, since it is challenging to replicate experiments without complete characterization. The lack of precise characterization is also a significant barrier to the development of medicinal products based on MSC-EVs.

Dosing of EVs

Accurate and standardized EV dosing is critical for reproducibility and meaningful comparisons across studies. The most precise method involves quantifying the number of EV particles per millilitre or per experimental animal (EVs/ml/animal), which directly reflects the actual dose administered. This method of determining the exact dose of EVs was employed in seven studies [29, 31, 35, 41–43, 46]. Considering only these publications, the number of MSC-EV doses used in in vitro studies ranged from 1 × 10⁷ to 1.5 × 10¹¹ particles/ml, with a mean dose of 2.9 × 10¹⁰ (SD = 4.8 × 10¹⁰). In contrast, in in vivo studies, the doses ranged from 1 × 10⁷ to 1 × 10¹¹ particles/ml, with a mean dose of 2.35 × 10¹⁰ (SD = 3.64 × 10¹⁰). In one case, the dose of MSC-EVs was given in terms of the number of particles per animal [46]. Alternatively, nine studies reported doses as protein quantities (µg EVs/ml or/animal) [30, 32–35, 37, 40, 42, 44], and one reported the number of MSC-EVs corresponding to the same number of cells. However, both methods are less precise and do not accurately reflect the actual number of EV particles administered, making comparisons and the reproducibility of results between experiments complex. Unfortunately, in two cases, the dose of MSC-EVs was not given at all, which also greatly limits the ability to standardize the protocol and analyze the results [36, 45].

The influence of MSC-EVs on the wound healing mechanism

MSC-EVs promote cell proliferation and migration

Cell proliferation and migration are fundamental biological processes that determine the efficiency of wound healing [52]. These mechanisms are fundamental during the proliferative phase; when granulation tissue forms, re-epithelialization is initiated, and the structural integrity of the tissue begins to be restored [53, 54]. In this context, mesenchymal stem/stromal cell-derived extracellular vesicles (MSC-EVs) have been widely studied for their ability to modulate these cellular responses. Among the 19 publications reviewed, 16 evaluated these effects in vitro, demonstrating that MSC-EVs were consistently found to stimulate these key regenerative processes (Table 1). These effects were observed in multiple cell types involved in wound healing, including human dermal fibroblasts (HDFs), endothelial cells, and epithelial cells.

Table 1.

The influence of mesenchymal stem/stromal cells extracellular vesicles on the proliferation and migration of different cell types

EV source	Target cells	EV dose	Test*	EV effect	Signalling pathway	References
AD-MSCs	HDFs	1 × 10¹⁰ particles/ml	CCK-8 Scratch test	120% ↑ proliferation and migration	-	Lee et al. 2023 [29]
AD-MSCs	HCEs	100 µg protein/ml	MTT Scratch test	↑ proliferation and migration	-	Tao et al. 2019 [33]
AD-MSCs	HUVECs	100 µg protein/ml	CCK-8 Scratch test	↑ proliferation ↑ migration	PI3K-AKT-mTOR-HIF-1α	Liu et al. 2021 [35]
UCB-MSCs	HUVECs, HDFs	ND	MTT Scratch test	↑ proliferation ↑ migration	-	Sung et al. [47]
UC-MSCs	HDFs	20 µg protein/ml	CCK-8 Transwell test	↑ proliferation ↑ migration	-	Tang et al. 2023 [36]
UC-MSCs	HUVECs	20 µg protein/ml	CCK-8 Scratch test	↑ proliferation 2,5-fold ↑ migration	-	Yang et al. 2020 [30]
WJ-MSCs	HDFs HaCaT	1 × 109 particles/ml	Scratch test	↑ proliferation ↑ migration	-	Kim et al. 2023 [31]
AD-MSCs	HaCaT	5 µg protein/ml	Scratch test CCK-8	↑ migration ↑ proliferation	-	Liao et al. 2022 [34]
UC-MSCs	HaCaT HSFs	ND	Scratch test Transwell CCK-8	↑ migration ↑ migration ↑ proliferation	ITCH/JUNB/IRE1α	Cheng et al. 2020 [37]
UC-MSCs	HUVECs	ND	CCK-8	↑ proliferation	-	Lu et al. 2024 [45]
HF-MSCs	HDFs	1.5 × 1011 particles/ml	CCK-8 Scratch test	↑ proliferation ↑ migration	-	Heras et al. 2022 [38]
P-MSCs	HDFs	4.56 × 10¹⁰ particles/ml	CCK-8 Scratch test Transwell	↑ proliferation ↑ migration ↑ migration	-	Su et al. 2023 [39]
AD-MSCs	HDFs	100 µg protein/ml	WST-1	↑ proliferation	-	Lin et al. 2023 [40]
P-MSCs	HECFCs	1 × 109 particles/ml	Cell proliferation assay	no significant differences in proliferation	-	Hao et al. 2021 [41]
AD-MSCs	HaCaT	250 µg protein/ml	PicoGreen Scratch test	↑ proliferation ↑ migration	-	Hodge JG et al. 2023 [42]
UC-MSCs	HUVECs	1 × 10¹⁰ particles/ml	CKK-8 Scratch test	↑ proliferation ↑ migration	-	Wu et al. 2024 [43]

AD-MSCs Adipose tissue Mesenchymal Stem/Stromal cells, UCB-MSCs Umbilical Cord Blood Mesenchymal Stem/Stromal Cells, WJ-MSCs Wharton’s Jelly Mesenchymal Stem/Stromal Cells, HaCaT Human immortal keratinocyte line, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, HDFs Human Dermal Fibroblasts, HCEs Human Corneal Epithelial Cells, HUVECs - Human Umbilical Vein Endothelial Cells, HSFs Human Skin Fibroblasts, WST-1 cell proliferation reagent, HECFCs Human Endothelial Colony Forming Cells, PI3K-AKT mTOR-HIF-1α Signalling pathway involved in regulation of angiogenesis and wound healing, ND no data,↑ increased,

*The principles underlying the applied tests are summarized in Supplementary Table 2

Lee et al. reported that EVs derived from adipose tissue mesenchymal stem/stromal cells (AD-MSC-EVs) promoted the proliferation and migration of human dermal fibroblasts (HDFs) in a dose-dependent manner, with a 120% increase in cell growth and faster closure of the wound gap compared to untreated controls at 1 × 10¹⁰ particles/ml [29]. Liao et al. reported a similar effect in human corneal epithelial cells (HCEs), where AD-MSC-EVs significantly accelerated cell migration, resulting in a reduced wound area compared to the control [33]. Moreover, MSC-EVs appear to restore the proliferation and migration of cells even under pathophysiological conditions.

Liu et al. mimicked diabetic microenvironments by exposing human umbilical vein endothelial cells (HUVECs) to advanced glycation end products (AGEs), resulting in endothelial dysfunction. Treatment with AD-MSC-EVs reversed these deficits by activating the PI3K/AKT/mTOR/HIF-1α signaling pathway, leading to a significant increase in HUVEC proliferation and migration [35]. It is noteworthy that pre-conditioning of MSCs, as presented in the studies here, enhanced the ability of MSC-EVs to stimulate these processes, resulting in greater proliferation and faster scratch closure compared to controls. Sung et al. compared the effects of different methods of pre-conditioning MSCs (e.g., thrombin, hypoxia, LPS, and H₂O₂) on EV efficacy, demonstrating enhanced stimulation of HUVEC and HDF proliferation and faster wound closure in the pre-conditioned groups than in control groups. The best therapeutic effect was observed in the thrombin group [47].

Further augmentation of EV function has been achieved through the use of biomaterial carriers. Tang et al. [36] and Yang et al. [30] independently demonstrated that hydrogel-based delivery systems, such as GelMA (a hydrogel consisting of gelatin and methacrylic anhydride) and Pluronic F127, prolong the availability of EVs and enhance biological outcomes. In both studies, EVs-loaded hydrogels outperformed free EVs in promoting cell proliferation and migration, indicating a synergistic effect associated with the controlled release of EVs. In addition, several other studies provided further evidence supporting these observations. Kim et al. demonstrated that EVs from Wharton’s jelly MSCs (WJ-MSC-EVs) stimulated both HDFs and keratinocytes (HaCaT), significantly increasing proliferation and migration [31]. Liao et al. also confirmed the pro-regenerative effects of AD-MSC-EVs on HaCaT cells, with enhanced proliferation and migration at 5 µg/ml [34]. Cheng et al. reported that UC-MSC-EVs promoted proliferation and migration of HaCaT and human skin fibroblasts (HSFs), with involvement of the ITCH/JUNB/IRE1α signaling pathway [37]. Lu et al. showed that UC-MSC-EVs enhanced HUVEC proliferation [45], while Heras et al. found that hair follicle MSC-derived EVs (HF-MSC-EVs) promoted HDF proliferation and migration at high particle doses [38]. Su et al. demonstrated similar effects using placenta-derived MSC-EVs (P-MSC-EVs), which stimulated proliferation and migration in HDFs [39].

Interestingly, not all studies confirmed strong proliferative activity. Hao et al. reported that P-MSC-EVs had no significant effect on human endothelial colony-forming cells (HECFCs), suggesting that the impact of EVs may depend on both the MSC source and the recipient cell type [41]. In contrast, Hodge et al. reported that AD-MSC-EVs enhanced the proliferation and migration of HaCaT keratinocytes in a dose-dependent manner [42]. Finally, Wu et al. reported that UC-MSC-EVs at 1 × 10¹⁰ particles/ml promoted both the proliferation and migration of HUVECs, further supporting their pro-angiogenic and pro-regenerative potential [43].

Taken together, these studies provide evidence that MSC-derived EVs stimulate the proliferation and migration of various cells and that additional preconditioning of MSCs and the use of hydrogels enhance this effect.

These in vitro results have been substantiated by in vivo data. In 15 studies, MSC-EVs demonstrated accelerated wound closure, confirming the in vitro results that showed stimulation of cell proliferation and migration (Table 2). For instance, Lee et al. demonstrated that the combination of AD-MSC-EVs with hyaluronic acid (HA) stimulated wound healing in mice and 20 kg SPF minipigs. This model is classified as a large animal model and was chosen because of significant differences in anatomical, physiological, and skin tissue regeneration between humans and rodent models. The aim was to provide a more clinically relevant assessment of the effect on wound healing. In porcine models, EV-treated wounds demonstrated a ~ 20% greater closure rate after two weeks compared to HA alone; however, in mice, granulation and epithelial regeneration were significantly improved [29]. Tang et al. observed that UC-MSC-EVs embedded in GelMA hydrogels restored the wound-healing capacity of diabetic rats to levels comparable to those of non-diabetic controls, demonstrating 95–100% closure within 14 days. These findings suggest that the use of a hydrogel as an EV carrier accelerates the healing of diabetic wounds, which is promising for treating patients with this disease [36].

Table 2.

The influence of mesenchymal stem/stromal cell extracellular vesicles on the rate of wound closure in different animal models

EV source	Animal/wound model	EV dose	EV effect	Signalling pathway	References
AD-MSCs	Minipig (full-thickness excisional skin wound, 30 mm x 30 mm) Mice (dermal filler model)	Minipigs: 4.0 × 10¹⁰ particles/ml, 3 times a week Mice: 1.4 × 10⁹ particles/ml, given once	acceleration of wound healing	-	Lee et al. 2023 [29]
UC-MSCs	Rats (8 mm full-thickness excisional skin wounds)	ND	acceleration of wound healing by 95–100%	-	Tang et al. 2023 [36]
UC-MSCs	Mice (12 mm full-thickness excisional skin excisional wound)	200 µg protein/µl, every other day for eight days	acceleration of wound healing by 90%	ITCH/JUNB/IRE1α	Cheng et al. 2020 [37]
WJ-MSCs	Rats (8 mm full-thickness excisional skin wounds) Mice (full-thickness incisional skin wounds, ulceration created)	Rats: 2 × 10⁸ EVs/rat, daily for three days after the procedure. Mice: 1 × 10⁸ EVs/ mouse, daily for three days after the procedure	↑ re-epithelialization ↓ wound area size	-	Kim et al. 2023 [46]
UC-MSCs	Rats (10 mm full-thickness excisional skin wounds)	100 µg protein, single topical application	acceleration of wound healing	-	Yang et al. [30]
WJ-MSCs	Mice (8 mm full-thickness excisional skin wounds)	1 × 10⁹ particles/ml,	↓ of wound size	-	Kim et al., 2023 [31]
BM-MSCs	Diabetic mice (8 mm full-thickness excisional skin wounds) Rats (12 mm full-thickness excisional skin wounds)	50 µg protein, single dose on the third day after procedure	Mice: greater wound closure Rats: No statistically significant difference in wound closure	-	Born et al. 2022 [32]
AD-MSCs	Mice (2 × 1,5 cm full-thickness excisional skin wounds)	100 µg protein/ml, daily for 2 weeks	acceleration of wound healing by 90%	-	Liao et al. 2022 [34]
AD-MSCs	Diabetic rats (10 mm full-thickness excisional skin wounds)	100 µg protein/ml, single application	↑ in wound closure rate	PI3K-AKT-mTOR-HIF-1a	Liu et al. 2021 [35]
UC-MSCs	Mice (6 mm full-thickness excisional skin wounds)	7 × 1010 particles/ml, single application	acceleration of wound healing by 75–100%	-	Lu et al. 2024 [45]
UCB-MSCs	Rats (8 mm full-thickness excisional skin wounds)	20 µg protein/10 µl, frequency of administration not specified	acceleration of wound healing by 93%	-	Sung et al. 2019 [47]
P-MSCs	Mice (10 mm full-thickness excisional skin wounds)	4,56 × 1010 particles/ml, every other day for two weeks	acceleration of wound healing by 80–100%	-	Su et al. 2023 [39]
UC-MSCs	Rats (20 mm full-thickness excisional skin wounds)	50 µg protein/10 µl, single application	acceleration of wound healing by 90–100%	-	Wu et al. 2024 [43]

AD-MSCs Adipose tissue Mesenchymal Stem/Stromal Cells, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, WJ-MSCs Wharton’s Jelly Mesenchymal Stem/Stromal Cells, ITCH/JUNB/IRE1α Signalling pathway, ITCH Itchy E3 ubiquitin protein ligase, IRE1α inositol-requiring enzyme 1α, JUNB transcription factor; full-thickness excisional skin wound - damage affecting the entire thickness of the skin – from the epidermis, through the entire layer of the dermis, to the border with the subcutaneous tissue with causing a loss of tissue volume; full-thickness incisional skin wounds - an incisional wound is a type of full thickness wound that is created by a surgical blade or another sharp instrument during a surgical procedure, such as a scalpel, which cuts or parts the tissue without causing a loss of tissue volume, ND no data, ↑ increased, ↓ reduced

It has been found that molecules carried by MSC-EVs influence cell proliferation and migration through various mechanisms, including modulation of signaling pathways and gene expression. For example, miR-27b transported by MSC-EVs, as demonstrated by Cheng et al. significantly accelerates wound healing by affecting the ITCH/JUNB/IRE1α signalling pathway, which promotes cell proliferation and migration. Studies in mice in which miR-27b in EVs was blocked showed delayed healing, confirming the key role of miR-27b in accelerating this process [37].

Findings by Kim et al. have demonstrated that EVs derived from three-dimensional (3D) cultures of mesenchymal stem/stromal cells (MSCs) Wharton’s jelly (WJ-MSCs), significantly accelerate wound healing. This is due to the increased secretion of EVs and their richer therapeutic content compared to EVs from traditional two-dimensional (2D) cultures. In studies conducted on rats and mice, the administration of MSC-EV purified from cells cultured in 3D significantly increased re-epithelialization and reduced wound area, regardless of the wound model or method of administration. These results clearly indicate that the cell culture method affects the quality and function of MSC-EVs [46].

Collectively, these studies provide evidence that EVs effectively accelerate the wound closure process regardless of the animal model used.

MSC-EVs influence angiogenesis

Angiogenesis - the formation of new blood vessels is essential for supplying cells with oxygen and nutrients, which is crucial for proper wound healing [55, 56]. Multiple in vitro studies have demonstrated that MSC-derived extracellular vesicles (MSC-EVs) actively support angiogenesis by enhancing endothelial cell organization, migration, and expression of proangiogenic mediators. Among the 19 analyzed studies, 10 directly investigated the proangiogenic effects of MSC-EVs in vitro, primarily through tube formation assays and gene expression analysis of angiogenesis-related factors (Table 3).

Table 3.

The influence of mesenchymal stem/stromal cell extracellular vesicles on the angiogenesis process

EV source	Target cells	EV dose	Test*	EV effect	Signalling pathway	References
WJ-MSCs	HUVECs	5 × 10⁸ particles/ml	Tube formation assay ELISA	↑ number of vascular structures ↑ expression of VEGF, Ang-1, Ang-2	-	Kim et al. 2023 [46]
AD-MSCs	HUVECs	100 µg protein/ml	Tube formation assay ELISA	restoration of the length of the tubular structures activating the PI3K-AKT-mTOR-HIF-1α signalling pathway	PI3K-AKT-mTOR-HIF-1α	Liu et al. 2021 [35]
UC-MSCs	HUVECs	ND	Test Matrigel	↑ the length of tubular structures ↑ GSH and MDA level	-	Lu et al. 2024 [45]
UC-MSCs	HUVECs	1 × 10¹⁰ particles/ml	Tube formation assay qRT-PCR Western blot	↑ in the number of coil nodes ↑ in VEGF factor expression level ↑ in VEGF protein level	-	Wu et al. 2024 [43]
P-MSCs	HCEs	100 µg protein/ml	qRT-PCR	↓ in VEGF, MMP2, MMP9 expression levels	-	Tao et al. 2019 [33]
HF-MSCs	HUVECs	2 × 10¹¹ particles/ml	Tube formation assay	↑ number of vascular structures higher number of branching points	-	Heras et al. 2022 [38]
UCB-MSCs	HUVECs	20 µg protein/10 µl	Tube formation assay	↑ number of vascular structures	-	Sung et al. 2019 [47]
P-MSCs	HECFCs	2 × 109 particles/ml	Tube formation assay qRT-PCR	↑ number of vascular structures ↑ expression of Ang-2	-	Hao et al. 2021 [41]
FS-MSCs	HUVECs	50 µg protein/ml	Tube formation assay	↑ number of vascular structures	-	Xu et al. 2024 [44]

WJ-MSCs Wharton’s Jelly Mesenchymal Stem/Stromal Cells, AD-MSCs Adipose Tissue Mesenchymal Stem/Stromal Cells, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, P-MSCs Placental Mesenchymal Stem/Stromal Cells, HF-MSCs Hair Follicle Mesenchymal Stem/Stromal Cells, FS-MSC Foreskin Mesenchymal Stem/Stromal Cells, HUVECs Human Umbilical Vein Endothelial Cells, HCEs Human Corneal Epithelial Cells, HECFCs human endothelial colony forming cells, VEGF Vascular Endothelial Growth Factor, Ang 1 Angiopoietin 1, Ang-2 Angiopoietin-2, PI3K-AKT-mTOR-HIF-1α Signalling pathway involved in regulation of angiogenesis and wound healing, GSH reduced glutathione, MDA malondialdehyde, MMP2 Matrix Metalloproteinases 2, MMP9 Matrix Metalloproteinases 9, ND no data, ↑ increased, ↓ decreased, *The principles underlying the applied tests are summarized in Supplementary Table 2

Kim et al. demonstrated that WJ-MSC-EVs significantly increased the number and complexity of capillary-like structures in HUVEC cultures. Additionally, EVs enhanced the expression of key proangiogenic factors, including VEGF, angiopoietin-1, and angiopoietin-2 [46]. Similar results were reported by Liu et al., who investigated the effect of AD-MSC-EV on angiogenesis in diabetic wound healing via the PI3K/AKT/mTOR/HIF-1α pathway. The use of AGEs inhibited angiogenesis in HUVEC cultures, shortening tubular structures and reducing the expression of HIF-1α and VEGF, while EVs reversed these effects. Further studies confirmed that the proangiogenic effect of AD-MSC-EV was mediated by the activation of this pathway [35]. Lu et al. confirmed that extracellular vesicles (EVs) promote angiogenesis in diabetic wound healing by counteracting the detrimental effects of neutrophil extracellular traps (NETs). NETs reduced the formation of tubular structures and induced endothelial ferroptosis, marked by decreased GSH and increased MDA levels. EV treatment reversed these effects, restoring angiogenic capacity. These findings suggest that EVs enhance angiogenesis by inhibiting NET-induced ferroptosis, providing a potential avenue for novel therapeutic strategies [45].

It was found that MSC-EVs purified from stem/stromal cells cultured without glucose and serum stimulate angiogenesis more effectively than those from cells cultured under standard conditions. This shows that the cell culture method is crucial for the function and effectiveness of vesicles, influencing the formation of new blood vessels [43].

Interestingly, the tissue-specific context appears to shape the angiogenic role of MSC-EVs. While most studies report enhancement of vessel formation, Tao et al. found that EVs derived from the placenta (P-MSC-EVs) inhibited angiogenesis in human corneal endothelial cells (HCEs) [33]. The differences between this study and others may be because the cornea is an avascular tissue, and its healing process aims to prevent neovascularization, which could lead to visual impairment. This highlights the targeted action of MSC-EVs, which are tailored to the regenerative needs of a specific tissue. However, it should be emphasized that this conclusion is based on only one study, and a more comprehensive analysis of this issue is needed [33].

According to reports by Heras et al., HF-MSC-EVs in the tube formation assay significantly increased the number of vascular structures and branching points in HUVECs [38]. Similar results were obtained by Sung et al. [47] and Xu et al. [44], who also demonstrated that UCB-MSC-EVs and FS-MSC-EVs significantly increase the formation of vascular structures in in vitro tests on these cells.

Hao et al. provide evidence showing that P-MSC-EVs not only increase the number of vascular structures but also enhance the expression of the Angiopoietin-2 (Ang-2) gene, a key factor in the process of vascular maturation and stabilization [41].

Collectively, these findings support the proangiogenic role of MSC-EVs in wound healing while also highlighting the importance of the cellular source, culture conditions, and tissue context. MSC-EVs act through both canonical angiogenic signaling (e.g., PI3K/AKT/mTOR/HIF-1α) and non-canonical mechanisms, such as the suppression of ferroptosis or NET-mediated injury.

In vivo data confirm the proangiogenic effect of MSC-EVs observed in vitro. Ten studies focused on the analysis of the effect of EVs on angiogenesis in animal tissue models (Table 4).

Table 4.

Effect of mesenchymal stem/stromal cell extracellular vesicles on angiogenesis in animal models

EV source	Animal/wound model	EV dose	EV effect	Signalling pathway	References
P-MSCs	Mice (hind limb ischemia)	1 × 10¹¹ particles/ml, single injection	↑ volume of blood vessels improvement of blood flow	-	Hao et al. 2022 [41]
UC-MSCs	Rats (10 mm full-thickness excisional skin wounds)	100 µg protein, single topical application	↑ microvessel density increase in VEGF and TGFβ-1 expression	-	Yang et al. [30]
BM-MSCs	Diabetic mice (8 mm full-thickness excisional skin wounds) Rats (12 mm full-thickness excisional skin wounds)	50 µg protein, single dose on the third day after procedure	↑ in the number of axial vessels	-	Born et al. 2022 [32]
P-MSCs	Mice (corneal alkaline burn)	33,33 µg protein/µl, three times a day/3 weeks	inhibition of angiogenesis ↓ in VEGF-α, MMP2 and MMP9 levels reduction in vessel density	-	Tao et al. 2019 [33]
AD-MSCs	Minipig (full-thickness excisional skin wound, 30 mm x 30 mm) Mice (dermal filler model)	Minipigs: 4.0 × 10¹⁰ particles/ml, 3 times a week Mice:1.4 × 10⁹ particles/ml, given once	↑ vascularization	-	Lee et al. 2023 [29]
WJ-MSCs	Mice (full-thickness incisional skin wounds, ulceration created)	1 × 10⁸ EVs/ mouse, daily for three days after the procedure	↑ expression of VEGF, Ang-1, Ang-2 ↑ in the number of axial vessels	-	Kim et al. 2023 [31]
AD-MSCs	Diabetic rats (10 mm full-thickness excisional skin wounds)	100 µg protein/ml, single application	↑ number of CD31-positive cells	PI3K-AKT-mTOR-HIF-1a	Liu et al. 2021 [35]
UC-MSCs	Rats (8 mm full-thickness excisional skin wounds)	ND	↑ in the number of blood vessels	-	Tang et al. 2023 [36]
UC-MSCs	Mice (6 mm full-thickness excisional skin wounds)	7 × 1010 particles/ml, single application	improved tissue perfusion and vessel density	-	Lu et al. 2024 [45]
FS-MSCs	Diabetic mice (6 mm full-thickness excisional skin wounds)	200 µg protein/mice, daily in the form of a solution	↑ number of CD31-positive cells restoration of the structural and functional network of blood vessels	-	Xu et al. 2024 [44]

P-MSCs Placental Mesenchymal Stem/Stromal Cells, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, BM-MSCs Bone Marrow Mesenchymal Stem/Stromal Cells, VEGF Vascular Endothelial Growth Factor; TGFβ-1 Transforming Growth Factor Beta 1, MMP2 Matrix Metalloproteinases 2, MMP9 Matrix Metalloproteinases 9, CD31 vascular endothelial cells marker; full-thickness excisional skin wound - damage affecting the entire thickness of the skin – from the epidermis, through the entire layer of the dermis, to the border with the subcutaneous tissue with causing a loss of tissue volume; full-thickness incisional skin wounds - an incisional wound is a type of full thickness wound that is created by a surgical blade or another sharp instrument during a surgical procedure, such as a scalpel, which cuts or parts the tissue without causing a loss of tissue volume, ND no data, ↑ increased, ↓ decreased

Hao et al. examined the effects of P-MSC-EVs conjugated with collagen-binding peptide SILY (SILY-EVs) on vascular remodelling, revascularization, and restoration of blood flow in the ischemic limb of mice. It was found that SILY-EVs treatment significantly increased the total blood vessel volume compared to treatment with unmodified EVs. SILY-EVs also improved the capillary density and the artery. These results suggest that conjugation of EVs with SILY peptide not only prolongs their retention but also translates into significantly better blood vessel regeneration and improved blood flow in the ischemic limb [41]. Yang et al. reported an increase in microvessel density and VEGF mRNA expression in a diabetic wound model in rats under the influence of UC-MSC-EVs. The study compared the effects of UC-MSC-EVs and UC-MSC-EVs combined with Pluronic F-127 hydrogel.

The microvessel density and the expression of VEGF and TGFβ-1 mRNA were significantly higher in the UC-MSC-EVs/PF-127 and UC-MSC-EVs groups compared to the controls. However, the best effects were observed in the case of EVs combined with hydrogel, suggesting that the use of such a combination leads to significantly increased angiogenesis in diabetic wounds [30].

Studies conducted by Born et al. showed that BM-MSC-EVs containing increased amounts of HOX transcript antisense RNA (HOTAIR-MSC-EVs) stimulate the angiogenesis process. The effect of HOTAIR-MSC-EVs in healthy rats and diabetic mice was compared. The number of axial vessels in the HOTAIR-MSC-EVs group was significantly increased compared with the control groups. HOTAIR-MSC-EVs effectively promoted the formation of new blood vessels in animal models by the increased density of CD31 + vessels in the tissue [32].

Tao et al. showed that EVs derived from P-MSCs inhibited the process of corneal neovascularization, a desirable effect for preserving corneal transparency, underscoring the context-dependent role of MSC-EVs in angiogenesis [33].

Lee et al. observed that AD-MSC-EVs increase vascularization in miniature pig skin wounds as well as in a mouse skin filler model [29]. Similar results were obtained in studies on rats, where UC-MSC-EVs led to an increase in the number of blood vessels [36], and in mice, where EVs from the same cells improved tissue perfusion and blood vessel density [45].

Kim et al. who demonstrated that WJ-MSC-EVs increase the expression of key angiogenic factors such as VEGF, Ang-1, and Ang-2, resulting in an increase in the number of axial blood vessels [46]. This is also confirmed by studies on diabetic wound models: Liu et al. observed that AD-MSC-EVs increase the number of CD31-positive cells, which is an indicator of new vessel formation, involving the PI3K-AKT-mTOR-HIF-1a signaling pathway in this process [35]. In turn, Xu et al. report that FS-MSC-EVs also increase the number of CD31-positive cells, leading to the structural and functional reconstruction of the blood vessel network [44].

This underscores the context-dependent effects of MSC-EVs and their potential for targeted modulation of biological processes depending on the source and specific tissue needs.

MSC-EVs modulate immune response

Persistent inflammation is a hallmark of chronic, non-healing wounds and represents a major barrier to effective tissue regeneration [3, 57]. Mesenchymal stromal cell-derived extracellular vesicles (MSC-EVs) have emerged as potent immunomodulatory agents capable of reprogramming inflammatory responses in injured tissues. Nine reviewed studies demonstrated that extracellular vesicles derived from mesenchymal stem/stromal cells modulate inflammatory processes in vitro, showing an inhibitory effect on inflammation induced by various factors (Table 5).

Table 5.

Influence of mesenchymal stem/stromal extracellular vesicles on the immune response in different cell types

EV source	Target cells	EV dose	Test*	EV effect	References
AD-MSCs	HCEs	100 µg protein/ml	PCR	↓ in expression levels of IL-1β, IL-8, TNF-α, NF-κB, IL-10 and Cas-8 increased in expression level of IL-10	Tao et al. 2019 [33]
WJ-MSCs	RAW 264.7	1 × 10⁹ particles/ml	Griess test ELISA	reduced the level of NO reduced the level of IL-1β, IL-6, TNF-α	Kim et al. 2023 [31]
WJ-MSCs	RAW264.7	5 × 10⁸ particles/ml	ELISA Griess test flow cytometry	↓ in expression levels of IL-1β, IL-6, TNF-α, ↑ in expression level of IL-10 reduced the level of nitric oxide ↓ in expression level of CD80 marker ↑ expression level of CD206 marker	Kim et al. 2023 [46]
AD-MSCs	HDFs	ND	qPCR	↓ expression level of IL-6 reduction of apoptotic cells	Lin et al. 2023 [40]
UC-MSCs	neutrophils	ND	flow cytometry	inhibition of NETs formation	Lu et al. 2024 [45]
P-MSCs	HECFCs	1 × 109 particles/ml	qPCR	↓ IFN-γ gene expression ↑ level of IL10	Hao et al. 2021 [41]
FS-MSCs	RAW264.7	100 µg protein/ml	flow cytometry	induction of macrophage phenotypic change from M1 to M2	Xu et al. 2024 [44]

AD-MSCs Adipose tissue Mesenchymal Stem/Stromal Cells, WJ-MSCs Wharton’s Jelly Mesenchymal Stem/Stromal Cells, HCEs Human Corneal Epithelial Cells, RAW264.7 Macrophages , HDFs Human Dermal Fibroblasts, IL-1β Interleukin-1 beta, IL-8 Interleukin-8, IL-6 Interleukin-6, TNF-α Tumor Necrosis Factor alpha, NF-κB Nuclear Factor kappa-light-chain-enhancer of activated B cells, IL-10 Interleukin-10, IFN-γ interferon gamma, Cas-8 Caspase-8, NO Nitric Oxide, macrophage phenotype M1 - proinflammatory; macrophage phenotype M2 - anti-inflammatory/repairing, NETs Neutrophil Extracellular Traps, ND no data, ↑ increased, ↓ decreased, *The principles underlying the applied tests are summarized in Supplementary Table 2

The study by Tao et al. showed that P-MSC-EVs significantly reduced the expression of proinflammatory cytokines and cell apoptosis-related genes (IL-1β, IL-8, TNF-α, NF-κB, IL-10, and Cas-8) in human corneal epithelial cells (HCEs) compared to the control.

The expression of IL-10 was increased, indicating the anti-inflammatory effect of extracellular vesicles [33].

Kim et al. investigated the anti-inflammatory effects of extracellular vesicles (EVs) from Wharton’s jelly-derived mesenchymal stem/stromal cells (WJ-MSCs), including those preconditioned with the flavonoid 3,2’-dihydroxyflavone (3,2’-DHF). EVs from both preconditioned and unpreconditioned cells significantly reduced nitric oxide (NO) production and proinflammatory cytokine expression, with more potent effects observed for preconditioned EVs [31].

Further studies confirmed that WJ-MSC-EVs suppressed inflammation in LPS-stimulated RAW264.7 macrophages [58]. Additionally, EV treatment promoted a shift in macrophage phenotype from pro-inflammatory (M1) to anti-inflammatory (M2), as evidenced by decreased CD80 expression and increased CD206 expression. ELISA results also showed reduced levels of IL-6, TNF-α, and IL-1β, alongside an increase in IL-10, further confirming the immunomodulatory potential of WJ-MSC-EVs [46].

According to the findings of Lin et al. AD-MSC-EVs reduced inflammation in fibroblasts exposed to radiation. This process resulted in a threefold increase in IL-6 expression in fibroblasts, indicating the induction of inflammation. It also increased the number of apoptotic cells ~ 4-fold. The effects of EVs derived from adipose tissue stem/stromal cells (AD-MSC-EVs) and conditioned medium (AD-MSC-CM) were compared. AD-MSC-EVs reduced IL-6 expression by ~ 60%, and ADSC-CM by ~ 50%. Subsequent results showed a reduction in apoptotic cells in both groups. These findings suggest that AD-MSC conditioned medium has an anti-inflammatory effect, but EVs have a stronger effect on inhibiting this process. This may be because EVs are also contained in the conditioned medium, but in lower concentrations [40].

Lu et al. report that UC-MSC-EVs inhibit the formation of extracellular neutrophil networks (NETs), which is crucial for reducing inflammation [45]. Hao et al. observed that P-MSC-EVs reduce IFN-γ gene expression and increase IL-10 levels, thereby promoting an anti-inflammatory response [41]. Furthermore, Xu et al. demonstrated that EVs from FS-MSCs induce a shift in macrophage phenotype from pro-inflammatory (M1) to anti-inflammatory (M2), which supports the suppression of inflammatory responses.

The anti-inflammatory activity of MSC-EVs has also been confirmed in vivo in eight studies (Table 6). H&E staining, assessment of macrophage polarization and analysis of the mRNA expression level of factors such as TGFβ-1, TNF-α, NF-κB and interleukins, including IL-6, IL-1β, IL-8 and IL-10 were employed to characterize these effects.

Table 6.

Effect of mesenchymal stem/stromal cell extracellular vesicles on the immune response in different animal models

EV source	Animal/wound model	EV dose	EV effect	Signalling pathway	References
UCB-MSCs	Rats (8 mm full-thickness excisional skin wounds)	20 µg protein/10 µl, frequency of administration not specified	↓ expression of TNF-α, IL-6	-	Sung et al. 2019 [47]
WJ-MSCs	Mice (full-thickness incisional skin wounds, ulceration created)	1 × 10⁸ EVs/ mouse, daily for three days after the procedure	↓ expression of IL-1b, IL-6, TNF-α ↑ expression of IL-10	-	Kim et al. 2023 [46]
AD-MSCs	Mice (corneal alkaline burn)	100 µg protein/ml, three times a day/3 weeks	↓ expression of IL-1β, TNF-α	-	Tao et al. 2019 [33]
UC-MSCs	Rats (10 mm full-thickness excisional skin wounds)	100 µg protein, single topical application	reduce signs of inflammation	-	Yang et al. 2020 [30]
AD-MSCs	Minipig (full-thickness excisional skin wound, 30 mm x 30 mm) Mice (dermal filler model)	Minipigs: 4 × 10¹⁰ particles/ml, 3 times a week Mice:1.4 × 10⁹ particles/ml, given once	reduce signs of inflammation	-	Lee et al. 2023 [29]
UC-MSCs	Rats (8 mm full-thickness excisional skin wounds)	ND	reduction of inflammatory cell infiltration	-	Tang et al. 2023 [36]
UC-MSCs	Mice (6 mm full-thickness excisional skin wounds)	7 × 1010 particles/ml, single application	inhibition of NETs formation	-	Lu et al. 2024 [45]
P-MSCs	Mice (hind limb ischemia)	1 × 10¹¹ particles/ml, single injection	induction of macrophage phenotypic change from M1 to M2 reduce signs of inflammation	-	Hao et al. 2022 [41]

UCB-MSCs Umbilical Cord Blood Mesenchymal Stem/Stromal Cells, WJ-MSCs Wharton’s Jelly Mesenchymal Stem/Stromal cells, AD-MSCs Adipose tissue Mesenchymal Stem/Stromal Cells, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, TNF-α Tumor Necrosis Factor alpha; IL-6 - Interleukin-6, IL-1b Interleukin-1 beta, IL-10 Interleukin-10; macrophage phenotype M1 - proinflammatory; macrophage phenotype M2 - anti-inflammatory/repairing, NETs Neutrophil Extracellular Traps; full-thickness excisional skin wound - damage affecting the entire thickness of the skin – from the epidermis, through the entire layer of the dermis, to the border with the subcutaneous tissue with causing a loss of tissue volume; full-thickness incisional skin wounds - an Incisional wound is a type of full thickness wound that is created by a surgical blade or another sharp instrument during a surgical procedure, such as a scalpel, which cuts or parts the tissue without causing a loss of tissue volume, ND no data, ↑ increased, ↓ decreased

Sung et al. evaluated EVs derived from umbilical cord blood MSCs (UCB-MSCs) preconditioned under various conditions (thrombin, hypoxia, LPS, H₂O₂). In rats with.

full-thickness skin wounds, EVs from thrombin-preconditioned MSCs most effectively suppressed TNF-α and IL-6 levels, indicating an enhanced therapeutic profile. The results suggest that preconditioning of MSCs increases the anti-inflammatory potential of UCB-MSC-EVs during wound healing [47].

Kim et al. further demonstrated that EVs from 3D cultures of WJ-MSCs reduced inflammation in a mouse skin wound model. Treatment resulted in decreased expression of IL-1β, IL-6, and TNF-α, and a concurrent increase in IL-10, compared to untreated wounds [46].

Tao et al. treated mice with alkaline corneal burns using P-MSC-derived EVs and analyzed pro-inflammatory gene expression (IL-1β, TNF-α) and inflammatory cell infiltration in the cornea. They observed a significant reduction in cytokine expression at both early and late stages of therapy. H&E staining showed that P-MSC-EVs alleviated histopathological changes, while the control group exhibited stromal edema and inflammatory infiltration on days 7 and 14 post-injury [33].

Yang et al. demonstrated that UC-MSC-derived EVs reduce inflammation in diabetic rats, with enhanced effects when combined with Pluronic F127 hydrogel. Histological analysis revealed reduced inflammatory cell infiltration, particularly in the EV + hydrogel group, compared to the controls. These findings confirm the anti-inflammatory potential of EVs and suggest that the hydrogel may prolong their local effect [30].

According to reports by Lee et al. AD-MSC-EVs applied to full-thickness skin wound models in miniature pigs, as well as in a skin filler model in mice, reduce inflammatory symptoms [29]. Similar effects were observed in a study by Tang et al. UC-MSC-EVs led to a reduction in inflammatory cell infiltration in full-thickness skin wounds in rats [36]. Lu et al. extended these observations by demonstrating that UC-MSC-EVs inhibit the formation of extracellular neutrophil networks (NETs) in skin wounds in mice, which also contributes to the reduction of inflammation [45].

Hao et al. provided a mechanistic explanation, demonstrating that P-MSC-EVs in a mouse model of hind limb ischemia induce a phenotypic shift of macrophages from pro-inflammatory (M1) to anti-inflammatory (M2) [41].

Taken together, these results highlight the potential of MSC-EVs as therapeutic agents in regenerative medicine. They have been shown to have a stimulating effect on reducing inflammation, which is a promising aspect of their use in wound treatment.

MSC-EVs remodel extracellular matrix

Another key factor in effective wound healing is the remodeling of the extracellular matrix (ECM) – a dynamic process in which the ECM is constantly broken down, rebuilt, and modified [59]. In this context, out of 19 publications, 6 addressed the role of mesenchymal stromal cell-derived extracellular vesicles (MSC-EVs) in ECM remodeling in vitro, while 10 additional studies reported in vivo effects (Tables 7 and 8).

Table 7.

Effect of mesenchymal stem/stromal cell extracellular vesicles on extracellular matrix remodelling, in vitro studies

EV source	Target cells	EV dose	Test*	EV effect	Signalling pathway	Reference
AD-MSCs	HDFs	1.5 × 10¹⁰ particles/ml	Procollagen Type 1 C-peptide (PIP) test qRT-PCR	↑ collagen synthesis ↑ in expression levels of COL1A1, COL3A1, α-SMA, FGF2 and ELN	-	Lee et al. 2023 [29]
AD-MSCs	HDFs	100 µg protein/ml	Western blot Picrosirius red staining RT-qPCR	↑ in expression levels of COL1A1, COL1A2, COL3A1 ↑ collagen synthesis	-	Lin et al. 2023 [40]
UC-MSCs	HaCaT HSFs	10 µg protein	Western blot RT-qPCR	↓ expression of ubiquitin ligase (ITCH) ↑ JUNB and IRE1α proteins levels	ITCH/JUNB/IRE1α	Cheng et al. 2020 [37]
P-MSCs	HECFCs	5 × 10⁷ particles/ml	NTA	↑ EVs adhesion to the type I collagen surface	-	Hao et al. 2022 [41]
AD-MSCs	HCEs	100 µg protein/ml	RT-qPCR H&E	reduced levels of MMP2 and MMP9 mRNA organization of the matrix structure in the proper layer of the cornea	-	Tao et al. 2019 [33]

AD-MSCs Adipose tissue Mesenchymal Stem/Stromal Cells, UC-MSCs Umbilical Cord Mesenchymal Stem/Stromal Cells, P-MSCs Placental Mesenchymal Stem/Stromal Cells, HDFs Human Dermal Fibroblasts, HaCaT Human immortal keratinocyte line, HSFs Human Skin Fibroblasts, HECFCs Human Endothelial Colony Forming Cells, NTA Nanoparticle Tracking Analysis, RT-qPCR Quantitative reverse transcription polymerase chain reaction, COL1A1 Collagen Type I Alpha 1 Chain, COL1A2 Collagen Type I Alpha 2 Chain, COL3A1 Collagen Type III Alpha 1 Chain, α-SMA Alpha-Smooth Muscle Actin, FGF2 Fibroblast Growth Factor 2, ELN Elastin, ITCH Itchy E3 ubiquitin protein ligase, IRE1α inositol-requiring enzyme 1α, JUNB transcription factor, MMP2 Matrix Metalloproteinases 2, MMP9 Matrix Metalloproteinases 9, ND no data, ↑ increased, ↓ decreased; *The principles underlying the applied tests are summarized in Supplementary Table 2

Table 8.

Effect of mesenchymal stem/stromal cell extracellular vesicles on ECM remodelling in different types of animal models

EV source	Animal/wound model	EV dose	EV effect	Signalling pathway	Reference
P-MSCs	Mice (10 mm full-thickness excisional skin wounds)	4.56 × 10¹⁰ particles/ml, every two days/two weeks	reducing the number of aging fibroblasts	-	Su et al. 2023 [39]
AD-MSCs	Diabetic rats (10 mm full-thickness excisional skin wounds)	100 µg protein/ml, single application	↑ collagen volume fraction	PI3K-AKT-mTOR-HIF-1a	Liu et al. 2021 [35]
AD-MSCs	Mice (corneal alkaline burn)	100 µg protein/ml, three times a day/3 weeks	reduced levels of MMP2 and MMP9 mRNA more regular arrangement of the corneal matrix	-	Tao et al. 2019 [33]
AD-MSCs	Minipig (full-thickness excisional skin wound, 30 mm x 30 mm) Mice (dermal filler model)	Minipigs: 4.0 × 10¹⁰ particles/ml, 3 times a week Mice:1.4 × 10⁹ particles/ml, single application	↑ levels of type III collagen thicker layers of tissue	-	Lee et al. 2023 [29]
WJ-MSCs	Mice (8 mm full-thickness excisional skin wounds)	1 × 10⁹ particles/ml, single application	↑ collagen volume fraction	-	Kim et al. 2023 [31]
UC-MSCs	Rats (8 mm full-thickness excisional skin wounds)	ND	↑ collagen volume fraction	-	Tang et al. 2023 [36]
UC-MSCs	Mice (12 mm full-thickness excisional skin excisional wound)	200 µg protein/µl, every other day for eight days	induction of bigger collagen fiber formation	ITCH/JUNB/IRE1α	Cheng et al. 2020 [37]
P-MSCs	Mice (hind limb ischemia)	1 × 10¹¹ particles/ml, single injection	↑ collagen volume fraction	-	Hao et al. 2022 [41]
UC-MSCs	Rats (20 mm full-thickness excisional skin wounds)	50 µg protein/10 µl, single application	greater and more extensive collagen deposition reduction of scar formation	-	Wu et al. 2024 [43]
FS-MSCs	Diabetic mice (6 mm full-thickness excisional skin wounds)	200 µg protein/mice, daily in the form of a solution	↑ collagen volume fraction	-	Xu et al. 2024 [44]

P-MSCs Placental Mesenchymal Stem/Stromal Cells, AD-MSCs Adipose Tissue Mesenchymal Stem/Stromal Cells, MMP2 Matrix Metalloproteinases 2, MMP9 Matrix Metalloproteinases 9, PI3K-AKT-mTOR-HIF-1a Signalling pathway involved in regulation of angiogenesis and wound healing, full-thickness excisional skin wound - damage affecting the entire thickness of the skin – from the epidermis, through the entire layer of the dermis, to the border with the subcutaneous tissue with causing a loss of tissue volume, ND no data, ↑ increased

Multiple studies have demonstrated that MSC-EVs stimulate the synthesis of ECM components, particularly collagen, elastin, and fibronectin. For instance, Lee et al. reported that adipose tissue-derived MSC-EVs (AD-MSC-EVs) increased collagen synthesis and the expression of key ECM-related genes including COL1A1, COL3A1, α-SMA, FGF2 and ELN. This suggests that EVs have a beneficial effect on the remodelling of the ECM, which is of great importance in the wound healing process [29]. Lin et al. showed similar effects in irradiated fibroblasts, where EVs stimulated radiation-suppressed collagen synthesis and reduced IL-6 expression. Increased expression of COL1A1, COL1A2, and COL3A1 proteins was also observed, as confirmed by Western blot [40].

Cheng et al. further elucidated the underlying molecular mechanisms, demonstrating that miR-27b-enriched MSC-EVs regulate ECM remodeling via the ITCH/JUNB/IRE1α pathway. EVs containing miR-27b reduced the expression of ITCH ubiquitin ligase, which inhibited JUNB protein degradation and promoted extracellular matrix (ECM) remodeling. The use of EVs increased the levels of JUNB and IRE1α proteins, indicating improved skin structural integrity, ECM component synthesis, and endoplasmic reticulum homeostasis, thereby favorably influencing the ECM remodeling process [37].

According to Tao et al., the application of EVs from AD-MSCs to human corneal epithelial cells (HCEs) at a concentration of 100 µg/ml results in a reduction in the mRNA levels of matrix metalloproteinases (MMP2 and MMP9). This, in turn, promotes the proper organization of the matrix structure in the corneal stroma, which is essential for maintaining its integrity and transparency [33].

In studies conducted by Lin et al., the same EVs at a similar concentration of 100 µg/ml, after administration to human dermal fibroblasts (HDFs), produced a similar effect. Increased collagen synthesis and elevated gene expression levels for type I (COL1A1, COL1A2) and type III (COL3A1) collagen were observed, as assessed by Picrosirius red staining and RT-qPCR [40].

Beyond composition, ECM targeting strategies may also enhance MSC-EV retention and function. Research by Hao et al. showed that extracellular vesicles conjugated with the SILY peptide (SILY-EVs), which have a high affinity for collagen, are more effective in promoting ECM remodeling than standard EVs. Conjugation with SILY significantly increased the adhesion of EVs to type I collagen. SILY-EVs improved the viability and proliferation of HECFCs. The results suggest that targeting EVs to the ECM through increased affinity for collagen can effectively support tissue healing processes by modulating ECM remodelling [41].

These in vitro findings were corroborated by in vivo studies (Table 8), which collectively confirmed that MSC-EVs promote ECM remodeling, increase collagen deposition, and reduce matrix degradation in cutaneous wound models.

In full-thickness skin wounds in mice, Su et al. reported that EVs derived from placental mesenchymal stem/stromal cells (P-MSC-EVs) enriched in miR-145-5p enhanced ECM remodeling by reducing fibroblast senescence and promoting collagen fiber organization, effects likely mediated by Erk/Akt signaling [39].

The stimulating effect of AD-MSC-EVs on collagen synthesis, which contributes to improved wound healing, is partly dependent on the activation of the PI3K-AKT-mTOR-HIF-1 signaling pathway [35].

MSC-EVs may also suppress excessive ECM degradation. In a mouse model of alkaline corneal burns, Tao et al. demonstrated that P-MSC-EVs led to a significant reduction in mRNA levels for MMP2 and MMP9. qRT-PCR analysis confirmed the modulatory effect of EVs on the expression of these enzymes, which may limit ECM degradation. Additionally, histological analysis revealed a more regular structure of the corneal matrix in the EV-treated animals, suggesting their positive effect on the remodeling or preservation of ECM structure through the regulation of metalloproteinase activity during pathological healing [33].

In both minipig and mouse models, the therapeutic effects of a combination of AD-MSC-EVs and HA were investigated. Full-thickness skin wounds in minipigs were treated topically, while mice received subcutaneous injections of the same combination. Histological analysis of the results revealed that the therapy led to an increase in type III collagen levels. The treatment also promoted tissue healing and regeneration, supporting ECM remodelling and scar-free healing. These effects were observed in both animal models, with mice exhibiting a thickening of tissue layers and an increased expression of type III collagen. No significant changes in type I or total collagen were noted [29].

Many experiments have reported increased collagen volume within skin wounds, regardless of the MSC donor species. For example, both EVs derived from Wharton’s jelly (WJ-MSCs) and umbilical cord (UC-MSCs) induced an increase in the collagen fraction in full-thickness skin wound models in mice and rats [31, 36]. Similar results were obtained for EVs derived from the placenta (P-MSCs), where a single injection led to an increase in the collagen fraction in a limb ischemia model in mice [41].

Importantly, some studies have described additional mechanisms and effects associated with ECM remodeling. Cheng et al. demonstrated that EVs from UC-MSCs not only increased collagen fiber volume in a mouse wound model but also induced the formation of larger fibrous structures through activation of the ITCH/JUNB/IRE1α axis [37].

Wu et al. demonstrated that in full-thickness wounds in rats, EVs from UC-MSCs not only promoted greater and more extensive collagen deposition, but also reduced scar formation [43]. In a diabetic model, the use of EVs derived from foreskin MSCs (FS-MSCs) increased the collagen fraction, suggesting their potential in treating difficult-to-heal wounds [44].

Overall, the results support the hypothesis that EVs stimulate ECM remodelling, and thus influence the process of tissue regeneration and proper wound healing. However, despite promising initial results, the role of MSC-EVs in ECM remodelling remains underexplored, likely due to the complexity of the involved signalling pathways in ECM remodelling. They represent a major research challenge that requires sophisticated experimental models. In addition, the ECM is a complex structure, and the interactions of MSC-EVs with ECM compartments are multifaceted, which also creates difficulties in related experiments.

Cumulative evidence from in vitro and in vivo studies confirms that MSC-derived extracellular vesicles (MSC-EVs) exert regenerative effects at multiple levels of the wound healing process. They enhance cell proliferation and migration, stimulate angiogenesis, modulate inflammatory responses, and support extracellular matrix remodeling, thereby contributing to accelerated and more organized tissue healing. These effects are mediated through diverse molecular pathways and are influenced by factors such as MSC source, culture conditions, and EV delivery methods. Preconditioning strategies and biomaterial-based carriers further optimize EV function, expanding their therapeutic potential. Taken together, MSC-EVs represent a multifaceted and adaptable tool for promoting wound healing.

Challenges in MSC-EVs clinical translation

Although MSC-EVs consistently demonstrate promising therapeutic effects in preclinical studies, their successful clinical translation remains limited. The main barrier is the absence of clearly defined specifications and release criteria for EV-based products, which are fundamental for ensuring identity, purity, potency, safety, and batch-to-batch consistency. Current ISEV and ISCT recommendations provide valuable technical guidance but have not yet been adopted as formal regulatory standards. As emphasized by Takakura et al. [60], no harmonized analytical framework currently exists that would satisfy regulatory expectations for the qualification and release of EV-based therapeutics. The present status of proposed release criteria is summarized in Table 9.

Table 9.

Challenges of manufacturing of MSC-EV- based products for clinical application

Manufacturing Process Development of MSC-EV based Product
A. Justification of specifications and release criteria
Criterion	Description	Methods used for EVs (Pharpacopoeial texst if available)	Research status on EVs	Acceptance criteria
Identity	Confirmation that the preparation contains the correct cell-derived EVs with characteristic markers and parameters.	Western blot / ELISA (markers CD9, CD63, CD81, TSG101, Alix); NTA (size, particle number); TEM / Cryo-EM (morphology); Bead-based flow cytometry (ExoView, MACSPlex).	Well-developed studies; adherence to MISEV2023 standards.	Acceptance criteria limits/ranges necessary to determine.
Purity	Assessment of the proportion of actual EVs relative to contaminants (proteins, lipids, microparticles, culture medium residues).	NTA (particle/protein ratio – purity index); Western blot for negative markers (calnexin, GRP94); SEC chromatography and protein profile analysis.	Lack of full standardization.	Acceptance criteria limits/ranges necessary to determine.
Potency	Confirmation of biological activity consistent with the drug’s mechanism of action (e.g., regeneration, immunomodulation).	In vitro functional assays (HUVEC migration, fibroblast proliferation, cytokine inhibition); Immunological tests (T cell suppression, NK cell activation); Measurement of effector gene or protein expression (RT-qPCR, ELISA).	Many studies are being conducted; however, there is no universal test, and each application requires its own.	Acceptance criteria limits/ranges necessary to determine.
Safety	Absence of bacteria, fungi, mycoplasma, and endotoxins in the product.	Sterility test (Ph. Eur. 2.6.1); LAL test for endotoxins (Ph. Eur. 2.6.14); PCR/mycoplasma culture (Ph. Eur. 2.6.7).	Test methods internationally harmonized outlined in pharmacopoeias like the USP and European Pharmacopoeia (Ph. Eur.), requiring validation for EVs based products.	Well defined acceptance criteria limits/ranges: the EV- based products has to be sterile, negative for mycoplasma and viruses; endotoxin limits depend on the administration routes (ex. 5EU per kg body weight per hour for intravenous administration).
	Absence of adventitious viruses in source cells, raw materials, and the product.	Viral safety (Ph.Eur 5.1.7)

B. Justification of the storage conditions and batch to batch consistency
Criterion	Description	Methods used for EVs	Research status on EVs
Stability	Determination of the stability of the EV- based product during storage and transport.	Stability studies at different storage temperatures (4 °C, -20 °C, − 80 °C) and time points. Analysis of freeze-thaw cycles. Confirmation of the acceptance criteria for identity, purity, potency and safety tests following EV-based product storage.	Limited data, single studies.
Batch- to bach consistency	Determination of the batch to bach consistency of EV-based product.	Confirmation of the acceptance criteria for identity, purity, potency, and safety tests for EV-based products obtained from different donors.	Limited data, single studies. The properties of MSC-EVs may differ depending on donor. There are no clear inclusion/exlusion criteria of MSC-EV donor.
C. Non-clinical studies
Criterion	Description	Methods used for EVs	Research status on EVs
Pharmacokinetics and biodistribution	Physical and chemical characterization of the particles.	Labelling of EVs with fluorochromes (DiR, PKH26); Isotope labelling (e.g., 99mTc, ¹¹¹In); Reporter protein labelling (luciferase, GFP).	Limited data, single studies. There are technical limitations in the evaluation of pharmacokinetics for EVs-based product. Artificially-labelled EVs may not reflect physiological dynamics by any visualization method.
Toxicology	Evaluation of acute and repeated- dose toxicity following single or multiplate EV based product administrations	In vivo single and repeated administration, single-dose studies with an appropriately extended post-dose observation; histopathological and biochemical analysis of major organs.	Limited data available. Only isolated short-term observations; comprehensive long-term safety studies are urgently required for regulatory approval.
Cancirogenity/tumourigenicity	Evaluation of the potential of EV- based product to promote uncontrolled cell proliferation or tumor formation.	Long-term animal studies assessing tumor incidence and histopathology; expression of proliferation markers (Ki-67, PCNA); oncogene and tumor suppressor gene expression profiling.	No systematic data available; carcinogenic risk remains unexplored. Long-term safety testing is required prior to clinical translation.
Genotoxicology	Assessment of EV capacity to induce DNA damage or chromosomal instability.	In vitro comet assay, micronucleus test; in vivo genotoxicity models assessing chromosomal aberrations.	No validated data for EVs; genotoxicity testing not yet performed in standardized systems.
Immunotoxicity and immunogenicity	Assessment of immune system activation, cytokine response, and potential immunotoxic effects following EVs administration.	Cytokine release assay; immune cell activation Histological analysis of immune system activation both locally and systemically.	Preliminary studies indicate low immunogenicity, an inhibitory effect on inflammation induced by various factors, but lack of systematic evaluation under GMP conditions.

This regulatory gap is compounded by substantial methodological variability across studies. Differences in MSC source, donor characteristics, culture conditions, and EV isolation methods lead to inconsistent purity and biological activity of the final products. Moreover, there is no standardized or validated assay to determine potency, and therefore, the biological activity of EV preparations cannot be reliably compared or predicted. Safety testing is similarly inconsistent, usually limited to sterility and endotoxin assessment, while data on viral safety, product stability, or long-term biological effects remain scarce (Table 9).

A further limitation concerns dosing strategies, which vary widely between studies and lack standardization. EV quantities are reported using different units—total protein, particle number, or volume—without a defined therapeutic threshold or established dose–response relationship [29–35, 37, 40–43]. The most common metric, protein concentration (µg protein/ml), is a poor surrogate for vesicle number or bioactive cargo, as it also captures co-isolated proteins and other contaminants. This heterogeneity makes it impossible to compare study outcomes directly or to design reproducible, clinically relevant dosing regimens.

Another critical barrier is the limited understanding of the pharmacokinetics and pharmacodynamics of MSC-EVs. Only a few experimental studies have examined their biodistribution, clearance, or tissue accumulation after administration. There are virtually no quantitative data on EV half-life, systemic exposure, or biological persistence. Without these data, rational dosing and therapeutic monitoring cannot be established. Equally concerning is the absence of comprehensive toxicological and carcinogenicity data. No long-term studies have evaluated the potential tumor-promoting effects, chronic toxicity, or immunogenicity of MSC-EVs after repeated administration. Given their complex bioactive composition and capacity to modulate cell signaling and proliferation, these gaps raise unresolved safety concerns that preclude the initiation of first-in-human trials. In conclusion, the clinical translation of MSC-EVs therapy is currently limited by the absence of standardized specifications and release criteria, non-uniform dosing parameters, and insufficient non-clinical pharmacokinetic and toxicological data. Progress in this field requires harmonized manufacturing protocols and validated analytical procedures that comply with GMP standards, as well as robust long-term non-clinical studies providing information for estimating the safe and biologically effective dose(s) to be used in clinical trials. Only the establishment of these frameworks will enable MSC-EVs to transition from experimental applications to safe, reproducible, and clinically approved therapeutics.

Conclusions

MSC-derived extracellular vesicles represent a promising acellular therapy for wound healing, offering a targeted and biologically active approach to tissue regeneration. By engaging multiple mechanisms - including immunomodulation, angiogenesis, and ECM remodelling-MSC-EVs support each phase of the healing cascade [29–47]. Future progress will depend on the standardization of EV characterization, scalable GMP-compliant production, and rigorous validation in clinically relevant models. With these advances, MSC-EV-based therapies may become a valuable addition to the regenerative medicine arsenal, particularly in wound treatment.

Abbreviations

AD-MSCs

Adipose tissue mesenchymal stem/stromal cells

AGEs

Advanced glycation end products

ALIX

ALG-2 interacting protein X

Ang-1

Angiopoietin 1

Ang-2

Angiopoietin-2

APO1A

Apolipoprotein A-I

BM-MSCs

Bone marrow mesenchymal stem/stromal cells

Cas-8

Caspase-8

CCK-8

Cell Counting Kit-8

CD

Cluster of Differentiation

COL1A1

Collagen Type I Alpha 1 Chain

COL1A2

Collagen Type I Alpha 2 Chain

COL3A1

Collagen Type III Alpha 1 Chain

DMEM

Dulbecco’s Modified Eagle Medium

ECM

Extracellular matrix

ELN

Elastin

EVs

Extracellular vesicles

FBS

Fetal bovine serum

FGF2

Fibroblast Growth Factor 2

FS-MSCs

Foreskin mesenchymal stem/stromal cells

GM130

Golgi Matrix Protein 130

GMP

Good Manufacturing Practice

GSH

Reduced glutatione

HA

Hyaluronic acid

HaCaT

Human immortal keratinocyte Line

HCEs

Human corneal epithelial cells

HDFs

Human dermal fibroblasts

HECFCs

Human Endothelial Colony Forming Cells

HF-MSCs

Hair follicle mesenchymal stem/stromal cells

HIF-1α

Hypoxia-inducible Factor 1-alpha

HLA-DR

Human Leukocyte Antigen – Antigen D Related

HOTAIR

HOX transcript antisense RNA

HSFs

Human skin fibroblasts

HUVECs

Human umbilical vein endothelial cells

IL-10

Interleukin-10

IL-1β

Interleukin-1 beta

IL-6

Interleukin-6

IL-8

Interleukin-8

IRE1α

Inositol-requiring enzyme 1α

ISCT

International Society for Cell & Gene Therapy

ISEV

International Society for Extracellular Vesicles

ITCH

Itchy E3 ubiquitin protein ligase

LAMP1

Lysosome-Associated Membrane Protein 1

MDA

Malondialdehyde

MISEV

Minimal Information for Studies of Extracellular Vesicles

MMP2

Matrix metalloproteinases 2

MMP9

Matrix metalloproteinases 9

MPC

2-methacryloyloxyethylphosphorylcholine polymer

MSC-EVs

Mesenchymal stem/stromal cells extracellular vesicles

MSCs

Mesenchymal stem/stromal cells

mTOR

Mechanistic Target of Rapamycin

MVB

Multi-Vesicular Body

ND

No data

NETs

Neutrophil Extracellular Traps

NF-κB

Nuclear Factor kappa-light-chain-enhancer of activated B cells

NO

Nitric oxide

NTA

Nanoparticle Tracking Analysis

PI3K

Phosphatidylinositol 3-kinase

P-MSCs

Placental mesenchymal stem/stromal cells

RT-qPCR

Quantitative reverse transcription polymerase chain reaction

SGD-EVs

EVs derived from serum and glucose deprivated

TEM

Transmission Electron Microscope

TGFβ-1

Transforming Growth Factor beta 1

TNF-α

Tumor Necrosis Factor Ralpha

TSG101

Tumor Susceptibility Gene 101 protein

UCB-MSCs

Umbilical cord blood mesenchymal stem/stromal cells

UC-MSCs

Umbilical cord mesenchymal stem/stromal cells

VEGF

Vascular Endothelial Growth Factor

WJ-MSCs

Wharton’s jelly mesenchymal stem/stromal cells

α-SMA

Alpha-Smooth Muscle Actin

↑

Increased

↓

Reduced

Author contributions

MP: Writing - review and editing, Supervision, Conceptualization, Funding acquisition.WU: Writing - original draft, Methodology, Data curation, Visualization.ZF: Writing - original draft, Methodology, Data curation, Visualization, Formal analysis.DK: Writing - review and editing, Methodology, Data curation.All authors read and approved the final manuscript.

Funding

This work was supported under Agreement No. FENG.02.01-IP.05-0324/23 as part of the “ANFIBIOM - an anti-inflammatory, anti-fibrotic biological medicinal product modulating wound healing” project carried out within the 02.02 First TEAM programme of the Foundation for Polish Science, co-financed by the European Union under the European Funds for Smart Economy 2021–2027 (FENG).

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.