Exosome-Mediated Regulation of Re-Epithelialization and Epithelial–Mesenchymal Plasticity in Diabetic Cutaneous Wounds: Current Advances and Translational Perspectives

Abstract

Diabetic cutaneous wounds, particularly diabetic foot ulcers, are common and refractory chronic complications of diabetes. Their impaired healing is closely associated with persistent inflammation, oxidative stress, insufficient angiogenesis, dysregulated extracellular matrix (ECM) remodeling, and defective re-epithelialization. Re-epithelialization requires keratinocytes to undergo coordinated, stage-dependent transitions among activation, migration, proliferation, redifferentiation, and barrier reconstruction. Compared with the classical concept of epithelial–mesenchymal transition, epithelial–mesenchymal plasticity (EMP) more accurately reflects the transient, reversible, and intermediate phenotypic states adopted by keratinocytes during cutaneous wound repair. Exosomes/small extracellular vesicles (sEVs) may support this process by exerting functional effects on keratinocyte repair programs and indirectly remodeling inflammatory, vascular, oxidative, and ECM microenvironments. However, most available studies assess wound closure, histological re-epithelialization, angiogenesis, inflammation, or collagen deposition, and direct evidence linking exosome/sEV treatment to coordinated EMP-associated state transitions, epithelial redifferentiation, and barrier recovery in diabetic wound keratinocytes remains limited. This review therefore reinterprets current exosome/sEV evidence through a keratinocyte-centered framework of “EMP imbalance–exosome-mediated intercellular communication–wound microenvironment remodeling.” We distinguish direct EMP evidence from functional keratinocyte effects, indirect microenvironmental support, and mechanistic inference, and discuss engineered vesicles, local delivery systems, evidence maturity, and translational risks. This framework clarifies both the therapeutic promise and the current evidentiary boundaries of exosome/sEV-based strategies for diabetic wound re-epithelialization.

Graphical abstract

Keywords

diabetic cutaneous wound / diabetic foot ulcer / exosomes / small extracellular vesicles / re-epithelialization / epithelial–mesenchymal plasticity / keratinocytes

Introduction

Diabetes-associated cutaneous wounds, particularly diabetic foot ulcers, are among the most common and refractory chronic complications of diabetes. Their development, progression, and recurrence are closely related to peripheral neuropathy, peripheral arterial disease, infection, abnormal foot pressure, and metabolic dysregulation, all of which substantially increase the risks of infection, amputation, and mortality and impose a considerable medical and socioeconomic burden[1]. Epidemiological studies have shown that diabetic foot ulcers are highly prevalent among patients with diabetes and are associated with poor long-term outcomes. Recent systematic evidence further indicates a markedly increased risk of death in patients with diabetic foot ulcers, with a 5-year mortality rate approaching 50%[2,3]. Therefore, diabetic cutaneous wounds represent not only a local disorder of tissue repair but also a major clinical problem affecting long-term survival and quality of life in patients with diabetes.

Compared with acute skin injuries, diabetic wounds persist in a pathological microenvironment characterized by hyperglycemia, accumulation of advanced glycation end products, sustained inflammation, oxidative stress, insufficient angiogenesis, and dysregulated extracellular matrix (ECM) remodeling. These abnormalities impair multiple repair-related cell types, including keratinocytes, fibroblasts, endothelial cells, and immune cells, ultimately leading to prolonged inflammation, poor granulation tissue formation, defective neovascularization, abnormal collagen deposition and remodeling, and delayed re-epithelialization[4–6]. Re-epithelialization is a critical step in wound closure and epidermal barrier restoration. It depends on the activation, migration, proliferation, redifferentiation, and reconstruction of cell–cell junctions and barrier structures by keratinocytes at the wound edge[7,8]. Accordingly, the ability of keratinocytes to acquire migratory competence after injury and to re-establish a stable epithelial architecture after wound coverage is a key determinant of effective diabetic wound healing.

Epithelial–mesenchymal transition (EMT) has traditionally been used to describe the acquisition of migratory properties by epithelial cells during wound repair. However, keratinocytes in cutaneous wound healing usually do not undergo a classical complete EMT. Instead, wound-edge keratinocytes more commonly adopt intermediate states that retain partial epithelial features while acquiring migratory capacity, and subsequently return to a stable epithelial phenotype after wound coverage to restore epidermal structure and barrier function[9–11]. Therefore, epithelial–mesenchymal plasticity (EMP), rather than a binary EMT-based model, more accurately describes the transient, reversible, and stage-dependent phenotypic transitions of keratinocytes during re-epithelialization[12]. In diabetic wounds, hyperglycemia, enhanced oxidative stress, sustained inflammation, and abnormal growth factor responses disrupt this dynamic process, resulting in insufficient keratinocyte migration, delayed epidermal reconstruction, and incomplete barrier recovery. Existing evidence suggests that hyperglycemia impairs functional wound closure and alters the role of EMT/EMP-related regulators, such as zinc finger E-box binding homeobox 1 (ZEB1), in wound angiogenesis and closure[13].

Exosomes are generally considered small extracellular vesicles (sEVs) associated with the endosomal pathway. They carry a variety of bioactive molecules, including proteins, lipids, mRNAs, miRNAs, lncRNAs, and circRNAs, and play important roles in intercellular communication and tissue repair. According to the Minimal Information for Studies of Extracellular Vesicles (MISEV2023) Guidelines, when the endosomal origin of vesicles has not been rigorously demonstrated, it is more appropriate to use terms such as extracellular vesicles (EVs), sEVs, or names based on experimental isolation characteristics[14]. Because the term “exosome” remains widely used in studies of diabetic wounds and skin repair, this review retains the terminology used in the original literature when discussing previous studies, while interpreting it mainly as exosome-like sEVs in mechanistic discussions. For clarity, the term “exosomes/sEVs” is used hereafter to refer to exosome-like sEVs unless otherwise specified.

In recent years, exosomes/sEVs from various cellular sources have shown considerable potential in diabetic wound repair, with mesenchymal stromal/stem cell (MSC)-derived exosomes/sEVs receiving particular attention. Current evidence suggests that their effects involve the attenuation of inflammation, promotion of angiogenesis, reduction of oxidative stress, regulation of collagen deposition and ECM remodeling, improvement of keratinocyte and fibroblast function, and acceleration of re-epithelialization[15,16]. These effects are not isolated events but converge on the remodeling of the pathological diabetic wound microenvironment. Because keratinocyte plasticity is influenced by inflammatory mediators, oxidative stress, vascular support, ECM cues, and paracrine signaling, exosomes/sEVs may contribute to diabetic wound re-epithelialization through functional effects on keratinocytes or through indirect remodeling of the surrounding wound microenvironment.

Although the beneficial effects of exosomes in diabetic wound healing have been widely investigated, most existing studies and reviews have focused on their general anti-inflammatory, pro-angiogenic, antioxidant, collagen-regulatory, and tissue-repair effects. In contrast, relatively few studies have systematically examined exosome/sEV involvement in EMP-related repair from the perspective of keratinocyte-driven re-epithelialization. Notably, most available evidence relies on wound closure rate, histological re-epithelialization, angiogenesis, and inflammatory markers as major endpoints, whereas direct links among exosome activity, EMP-associated and redifferentiation markers, keratinocyte state transitions, and barrier function recovery remain insufficiently established. Reconsidering exosomes/sEVs within the framework of “intercellular communication–microenvironment remodeling–keratinocyte plasticity” may therefore help clarify their mechanistic boundaries and translational potential in diabetic cutaneous wound repair.

Several recent reviews have comprehensively summarized the sources, cargos, engineering strategies, biomaterial delivery systems, and broad pro-healing effects of exosomes/sEVs in diabetic wounds[15–21]. The specific contribution of the present review is therefore not another source-by-source catalogue of vesicle therapies, but a keratinocyte-centered reinterpretation of the evidence through the linked processes of re-epithelialization and EMP. This distinction is important because accelerated wound closure, increased angiogenesis, reduced inflammatory cytokines, or altered collagen deposition do not, by themselves, demonstrate normalization of keratinocyte EMP trajectories and epithelial redifferentiation. Direct links among exosome/sEV exposure, keratinocyte state transitions, coordinated EMP-associated marker changes, redifferentiation, and functional barrier recovery remain insufficiently established.

Accordingly, we organize the literature around the conceptual axis of “EMP imbalance–exosome-mediated intercellular communication–wound microenvironment remodeling” and apply an explicit evidence hierarchy. We distinguish (1) direct EMP evidence showing stage- or trajectory-resolved changes in wound keratinocytes together with epithelial reconstruction or barrier recovery; (2) functional keratinocyte evidence demonstrating migration, proliferation, survival, or re-epithelialization without sufficient characterization of EMP-associated state transitions; (3) indirect microenvironmental support mediated through immune, vascular, redox, fibroblast, or ECM remodeling; and (4) mechanistic inference based on pathway associations without direct state-transition data. Using this framework, we examine natural and engineered exosomes/sEVs, local delivery systems, quality control, and translational maturity, while identifying the experiments needed to establish whether these vesicles truly regulate EMP rather than merely improve general wound-healing endpoints.

Impaired re-epithelialization and epithelial–mesenchymal plasticity imbalance in diabetic cutaneous wounds

Re-epithelialization during normal cutaneous wound healing

Re-epithelialization is a central event in cutaneous wound closure and epidermal barrier restoration. It primarily depends on the activation of keratinocytes located at the wound edge and in residual skin appendages, followed by their coordinated migration, proliferation, redifferentiation, and reconstruction of barrier structures[22]. During the early phase of acute skin injury, wound-edge keratinocytes sense inflammatory mediators, growth factors, ECM components, and mechanical cues. These signals induce partial loosening of cell–cell junctions, cytoskeletal remodeling, changes in adhesion patterns, and formation of a migratory front, thereby enabling keratinocytes to move across the wound bed[23].

Importantly, re-epithelialization is not a process in which keratinocytes fully convert into mesenchymal-like cells and migrate as isolated single cells. Evidence from three-dimensional skin models, intravital imaging, and studies of epidermal stem cell dynamics indicates that cutaneous wound repair involves tissue-scale coordination of cell migration, proliferation, and differentiation[24,25]. Keratinocytes close to the wound edge mainly contribute to migration and wound coverage, whereas cells located behind the leading edge provide proliferative support, replenish the epithelial sheet, and maintain epidermal continuity. These spatially distinct cell populations display dynamic functional specialization during repair[26]. Thus, keratinocytes generally achieve wound coverage through collective migration and epithelial sheet advancement rather than by fully losing epithelial organization and becoming free mesenchymal-like cells.

After the wound surface is covered by a new epithelial layer, keratinocytes must withdraw from the migratory state, re-establish cell–cell junctions, restore epithelial polarity and stratification, and complete terminal differentiation to regenerate a functional epidermal barrier. Normal re-epithelialization is therefore not merely a process of enhanced migration, but a continuous sequence involving keratinocyte activation, migration, proliferation, cessation of migration, redifferentiation, and barrier reconstruction. These stage-dependent, reversible, and state-specific phenotypic transitions provide the biological basis for interpreting diabetic wound re-epithelialization through the lens of epithelial–mesenchymal plasticity.

EMP rather than complete EMT predominates during cutaneous wound repair

EMT was originally used to describe the conversion of epithelial cells into mesenchymal cells during embryonic development, organ fibrosis, and cancer invasion and metastasis. Its classical features include loss of epithelial polarity, disruption of cell–cell junctions, downregulation of epithelial markers, and acquisition of mesenchymal migratory and invasive properties. However, keratinocytes involved in cutaneous wound repair do not usually undergo a classical complete EMT. During re-epithelialization, keratinocytes may acquire partial EMT-like features, such as junctional remodeling, altered adhesion, enhanced motility, and activation of selected EMT-related transcriptional programs. Nevertheless, these changes are typically transient, localized, and reversible, and they serve to promote wound coverage and subsequent barrier restoration rather than to establish a stable mesenchymal cell fate[9,10].

Compared with the traditional EMT concept, EMP emphasizes the ability of epithelial cells to occupy multiple intermediate states in response to changing microenvironmental cues, dynamically balancing retained epithelial traits, enhanced migratory capacity, and partial activation of mesenchymal-like programs[11]. This concept is particularly suitable for describing keratinocyte behavior during wound repair. Wound-edge keratinocytes must transiently reduce some static epithelial features to acquire motility, while preserving sufficient intercellular coordination and tissue continuity to support collective epithelial sheet movement. After wound closure, they must then restore a stable epithelial architecture and barrier function[26].

Accordingly, the key issue in cutaneous wound repair is not whether EMT occurs, but whether keratinocytes can enter, maintain, and exit a reparative plastic state at the appropriate time. Excessive maintenance of a static epithelial phenotype may limit migration and delay wound coverage, whereas prolonged persistence in a migratory but poorly differentiated intermediate state may impair epithelial stratification, junctional reconstruction, and barrier recovery. Recent work showing that grainyhead like transcription factor 2 (GRHL2) regulates keratinocyte EMT and the reverse mesenchymal–epithelial transition (MET), thereby influencing scar formation, further suggests that restoration of epithelial identity during the later phase of repair is essential for normal tissue reconstruction[27]. Thus, successful re-epithelialization depends not on simply enhancing or inhibiting EMT-like changes, but on enabling timely, reversible, and ordered switching between migratory and epithelial-rebuilding states.

Major manifestations and mechanisms of EMP imbalance in diabetic wounds

A major reason for the chronicity of diabetic cutaneous wounds is the disruption of keratinocyte plasticity required for normal re-epithelialization. Hyperglycemia, accumulation of advanced glycation end products, increased oxidative stress, persistent inflammation, insufficient angiogenesis, and microcirculatory impairment jointly alter the wound microenvironment. These pathological factors directly or indirectly compromise keratinocyte migration, proliferation, differentiation, and stress adaptation, ultimately delaying re-epithelialization. Previous studies have shown that hyperglycemia inhibits keratinocyte migration and is associated with abnormal p38 mitogen-activated protein kinase (MAPK) signaling, autophagy, and related pathways[28]. Diabetes may also restrict keratinocyte migration and wound re-epithelialization through epigenetic regulation of the forkhead box O1–matrix metalloproteinase 9 (FOXO1–MMP9) axis[29]. These findings indicate that the diabetic microenvironment not only reduces keratinocyte activity but also interferes with their ability to enter a migratory state, sustain migration, and complete wound coverage.

From the EMP perspective, abnormalities in diabetic wounds should not be simply defined as either insufficient or excessive EMT. Instead, they should be understood as disruption of the timing, reversibility, and microenvironmental dependence of keratinocyte state transitions. Sen and Roy proposed that functional wound closure requires not only epithelial coverage of the wound surface but also restoration of the cutaneous barrier, a process governed by glucose metabolism-dependent EMP dynamics[12]. Tan et al. further reported that hyperglycemia restricts acetylcholine-induced EMT-like changes in keratinocytes, suggesting that high-glucose conditions may impair the acquisition of migration-associated phenotypes[30]. In parallel, EMT/EMP-related regulators such as ZEB1 appear to exert glucose-state-dependent effects on wound angiogenesis and closure, indicating that diabetes disrupts the dynamic plasticity network required for repair rather than simply driving phenotypic conversion in one direction[13].

It should be noted that direct evidence for impaired MET or insufficient epithelial phenotype restoration in diabetic wounds remains limited. A more cautious interpretation is that diabetic keratinocytes may exhibit both inadequate initiation of migration-associated EMP programs and insufficient return to a stable epithelial phenotype after wound coverage. Studies on GRHL2-mediated regulation of keratinocyte EMT–MET dynamics and scar formation suggest that epithelial phenotype restoration during the later phase of repair is important for normal tissue reconstruction[27]. Although this mechanism requires further validation in diabetic wounds, it highlights the need to assess not only whether keratinocyte migration is enhanced, but also whether keratinocytes can exit the migratory state at the appropriate time and rebuild epithelial architecture. In addition, studies of chronic wounds indicate that aberrant DNA methylation can impair wound healing by affecting EMT/EMP-related programs, suggesting that epigenetic regulation may represent another mechanism contributing to EMP imbalance in diabetic and chronic wounds[31].

Beyond intrinsic keratinocyte abnormalities, persistent inflammation, insufficient angiogenesis, tissue hypoxia, and ECM disorganization further aggravate defective re-epithelialization in diabetic wounds. Keratinocyte dysfunction is not an isolated event; rather, it is intertwined with immune cell abnormalities, endothelial injury, fibroblast dysfunction, and aberrant ECM remodeling. Together, these changes impair the initiation of migration-associated phenotypes, delay wound coverage, and compromise subsequent barrier reconstruction[5–7]. Therefore, defective re-epithelialization in diabetic cutaneous wounds reflects a dynamic imbalance between keratinocyte plasticity and the surrounding wound microenvironment. Strategies that modulate intercellular communication and reshape the local microenvironment may help create conditions that support stage-appropriate keratinocyte plasticity. In this context, exosomes/sEVs have emerged as promising regulatory candidates because they can deliver diverse bioactive molecules and simultaneously affect inflammation, angiogenesis, oxidative stress, ECM remodeling, and epithelial repair[15].

Biological basis of exosome/sEV-mediated intercellular communication in diabetic wounds

Basic characteristics and bioactive cargos of exosomes/sEVs

Exosomes are generally considered sEVs closely associated with the endosomal pathway. Their biogenesis involves plasma membrane invagination to form early endosomes, inward budding of the endosomal membrane to generate intraluminal vesicles, maturation into multivesicular bodies, and release into the extracellular space after fusion of multivesicular bodies with the plasma membrane[32]. Although exosomes have traditionally been described as vesicles approximately 30–150 nm in diameter, size, density, or selected marker proteins alone are insufficient to rigorously establish their endosomal origin. Therefore, findings on exosomes in diabetic wound studies should be interpreted in relation to the isolation methods, characterization criteria, and terminology used in the original reports.

Exosomes/sEVs are enclosed by a lipid bilayer and can carry proteins, lipids, mRNAs, miRNAs, lncRNAs, circRNAs, and various metabolism-related molecules. The composition of these cargos is shaped by the donor cell type, culture conditions, pathological state, and external stimuli, and largely determines the biological effects on recipient cells. Previous reviews have shown that exosomal miRNAs, lncRNAs, and circRNAs participate in diabetic wound repair by regulating inflammation, angiogenesis, re-epithelialization, and ECM remodeling[17]. In addition, functional proteins and lipid components in exosomes/sEVs may influence multiple repair-related cell types, including keratinocytes, fibroblasts, endothelial cells, and macrophages.

In the context of this review, the significance of exosomes/sEVs lies not simply in their ability to promote the proliferation or migration of a single cell type, but in their role as intercellular information carriers that integrate inflammatory, hypoxic, oxidative, angiogenic, and ECM-related cues. Because keratinocyte plasticity is shaped by macrophage polarization, endothelial cell function, fibroblast activation, ECM mechanics, and local paracrine signaling, exosomes/sEVs provide an important biological basis for connecting EMP imbalance with microenvironmental abnormalities in diabetic wounds[18].

Major sources of exosomes/sEVs relevant to diabetic cutaneous wound repair

The exosome/sEV sources most extensively investigated in diabetic wound repair are MSCs, particularly adipose-derived stromal/stem cells (ADSCs), bone marrow-derived MSCs, and human umbilical cord-derived MSCs (hUCMSCs). Their vesicles contain heterogeneous nucleic acid, protein, and lipid cargos and have been associated with immunomodulatory, pro-angiogenic, antioxidant, matrix-regulatory, and epithelial-repair effects[15,18,33–37]. However, vesicle composition and biological activity vary according to donor-cell origin, metabolic state, culture conditions, and preconditioning strategy. Source designation alone therefore does not establish a specific mechanism or level of evidence for EMP-related repair.

Skin-resident and wound-associated cells provide a second group of relevant vesicle sources. Epidermal stem cell-, keratinocyte-, and dermal fibroblast-derived EVs participate in local communication among epidermal, immune, and mesenchymal compartments[38–41]. Importantly, these vesicles are not invariably reparative. Their effects depend on the physiological or pathological state of the donor cell, and EVs released under diabetic or high-glucose conditions may propagate dysfunctional signals between keratinocytes and fibroblasts[42,43]. This context dependence is particularly relevant to EMP because local vesicle communication may either support coordinated epithelial reconstruction or reinforce a non-healing microenvironment.

Other investigated sources include platelet-rich plasma and endothelial progenitor cells, whose vesicles have been linked to re-epithelialization, fibroblast activation, and angiogenesis in diabetic wound models[44–46]. The representative sources, cargos, target cells, models, endpoints, and corresponding evidence levels are summarized in Table 1. To avoid a source-by-source repetition of biological effects, the following section instead focuses on how these vesicles influence re-epithelialization and EMP-related repair at different levels of mechanistic evidence.

Linking exosomes/sEVs to EMP-related repair through intercellular communication

Exosomes/sEVs may influence diabetic wound re-epithelialization through two broad routes: direct interaction with keratinocyte repair programs and indirect remodeling of the inflammatory, vascular, oxidative, fibroblast, and ECM microenvironments[16,18]. However, neither the cellular source of a vesicle nor improvement in a general wound-healing endpoint is sufficient to establish direct EMP regulation. The mechanistic interpretation must instead consider the recipient cell, model context, measured endpoints, temporal resolution, and whether coordinated keratinocyte state transitions were demonstrated. Section 4 therefore evaluates the available evidence using the hierarchy of direct EMP evidence, functional keratinocyte evidence, indirect microenvironmental support, and mechanistic inference.

Mechanisms by which exosomes/sEVs participate in EMP-related repair processes in diabetic wounds

EMP-related repair in diabetic wounds should be interpreted within a multicellular communication framework, but the evidentiary strength of different findings must be separated. In this review, “direct EMP evidence” requires exosome/sEV exposure to be linked to keratinocyte-specific, stage- or trajectory-resolved phenotypic transitions in a diabetic wound-relevant setting. Ideally, such evidence should demonstrate coordinated changes in epithelial identity and junctional/barrier features, migration-associated or mesenchymal-like programs, and subsequent redifferentiation, together with functional evidence of epithelial sheet advancement or barrier reconstruction. A single endpoint—such as wound closure, histological epithelial thickness, keratinocyte migration, or expression of one EMT marker—is insufficient to establish coordinated EMP dynamics.

Accordingly, the mechanistic evidence discussed below is classified into four levels: direct EMP trajectory evidence; functional keratinocyte evidence; indirect microenvironmental support; and mechanistic inference. This hierarchy is intended to prevent microenvironmental improvements in inflammation, angiogenesis, oxidative stress, or ECM remodeling from being interpreted as direct regulation of keratinocyte EMP. The integrated direct, indirect, and pathological EV communication routes are summarized in Figure 1, and the evidence level assigned to representative studies is presented in Table 1.

Functional effects on keratinocyte repair programs

The core of re-epithelialization is not simply the acquisition of motility. Wound-edge keratinocytes must enter a transient reparative state that permits collective migration and proliferation, and then appropriately exit that state after coverage to re-establish epithelial junctions, stratification, and barrier function. Exosome/sEV-mediated enhancement of keratinocyte migration or proliferation should therefore be described as functional keratinocyte evidence unless the study also characterizes the coordinated transition into and out of EMP-associated states.

For future mechanistic studies, a practical core EMP panel should combine epithelial identity and junction/barrier proteins (E-cadherin, claudin-1, and occludin), migration- or mesenchymal-associated markers (including vimentin), and transcriptional regulators that govern state transitions, particularly GRHL2, ZEB1, snail family transcriptional repressor 1 (SNAI1/Snail), and snail family transcriptional repressor 2 (SNAI2/Slug). These markers should be evaluated as coordinated temporal and spatial patterns rather than as isolated binary readouts, because regulators such as ZEB1 and GRHL2 exert stage- and context-dependent effects during wound repair[13,27]. Single-cell RNA sequencing, spatial transcriptomics, pseudotime or trajectory inference, lineage tracing, and intravital/live imaging can further resolve keratinocyte subpopulations, wound-edge spatial states, and the transition from migration to redifferentiation. Recent single-cell data from human diabetic foot ulcers (DFUs) demonstrating impaired epidermal differentiation and altered keratinocyte trajectories provide a relevant methodological precedent[47], while lineage-tracing and live-imaging studies in acute wounds illustrate how dynamic state transitions can be validated experimentally[24,25].

More targeted evidence comes from engineered vesicles designed to modulate re-epithelialization and tissue regeneration programs. Huang et al. reported that circular RNA derived from cyclin-dependent kinase 13 (circCDK13)-loaded sEVs accelerated wound healing and promoted skin appendage regeneration in db/db mice and streptozotocin-induced diabetic rats. Mechanistically, circCDK13 interacted with insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3) and regulated molecules related to cell migration and proliferation, including CD44 and c-MYC, thereby promoting reparative behaviors in human epidermal keratinocytes and dermal fibroblasts[48]. Platelet-rich plasma-derived exosomes were also shown to promote re-epithelialization in chronic diabetic wounds in rats through Yes-associated protein (YAP) activation, suggesting that exosomes can act on pathways related to migration, proliferation, and mechanosensing[44]. These studies indicate that exosomes/sEVs may enhance the ability of keratinocytes to enter a reparative plastic state by modulating signaling modules such as CD44/c-MYC and YAP. However, these findings should not yet be interpreted as definitive evidence that exosomes/sEVs coordinate the full sequence of EMP-associated state transitions and subsequent epithelial redifferentiation.

Reshaping inflammation and the immune microenvironment

Persistent inflammation is a major driver of defective re-epithelialization in diabetic wounds. In diabetic wounds, macrophages, neutrophils, and multiple inflammatory mediators remain abnormally activated, leading to sustained increases in tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), interleukin-6 (IL-6), reactive oxygen species (ROS), and proteolytic enzymes. These factors impair keratinocyte migration, proliferation, and differentiation, thereby further delaying re-epithelialization[6]. Therefore, the significance of exosome/sEV-mediated inflammatory regulation should not be limited to “anti-inflammation.” Rather, it should be understood as immune microenvironment remodeling that relieves the inhibitory effects of chronic inflammation on keratinocyte plasticity and barrier reconstruction.

Multiple studies have shown that MSC- and epidermal stem cell-derived exosomes can shift the diabetic wound immune environment from a sustained pro-inflammatory state toward a more reparative state. ADSC-derived exosomes promote macrophage polarization toward an M2-like reparative phenotype through the circ-Rps5/miR-124-3p axis, reducing inflammation and improving collagen generation and diabetic wound healing[49]. Epidermal stem cell-derived exosomes induce M2-like macrophage polarization through the miR-203a-3p/suppressor of cytokine signaling 3 (SOCS3) axis and thereby promote diabetic wound repair[39]. Bone marrow MSC-derived exosomal miR-146a-5p regulates the M1/M2 macrophage polarization balance and promotes wound healing in diabetic mice[50]. hUCMSC-derived exosomes also promote M2 macrophage polarization, angiogenesis, and collagen deposition, thereby accelerating diabetic wound repair[37]. In addition, keratinocyte-derived exosomes enhance the pro-reparative function of macrophages through the metastasis associated lung adenocarcinoma transcript 1 (MALAT1)/miR-1914-3p/milk fat globule-EGF factor 8 (MFGE8) axis, further supporting the existence of exosome-mediated bidirectional communication between epidermal and immune cells[40].

From the EMP perspective, these studies are best interpreted as indirect evidence that immune microenvironment remodeling may support reparative keratinocyte plasticity. If a wound remains in a pro-inflammatory and high-ROS state, keratinocytes may fail to complete the ordered transition from a migratory state to an epithelial-rebuilding state even if they retain intrinsic migratory potential. Thus, exosome/sEV-mediated immunomodulation should be regarded as indirect microenvironmental support for keratinocyte repair rather than as evidence that these vesicles directly regulate the keratinocyte EMP program.

Promoting angiogenesis and reducing hypoxia/oxidative stress

Insufficient angiogenesis, tissue hypoxia, and excessive oxidative stress are key microenvironmental factors contributing to defective re-epithelialization in diabetic wounds. Keratinocyte migration, proliferation, redifferentiation, and barrier reconstruction all require adequate oxygen supply, nutrient delivery, and redox balance. When wounds remain ischemic, hypoxic, and exposed to excessive ROS, keratinocytes are more likely to show impaired migration, increased stress injury, and failure of barrier reconstruction. By promoting angiogenesis and reducing oxidative stress, exosomes/sEVs may provide the metabolic and structural support required for EMP-related repair processes.

ADSC-derived exosomes have been extensively studied in relation to angiogenesis and oxidative stress in diabetic wounds. Hypoxia-preconditioned ADSC-derived exosomes promote high-quality diabetic wound healing through pathways including phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt) and reduce local inflammation[33]. ADSC-derived exosomes improve high-glucose-associated oxidative stress and angiogenic impairment through the sirtuin 3/superoxide dismutase 2 (SIRT3/SOD2) axis[34]. Nuclear factor erythroid 2-related factor 2 (Nrf2)-overexpressing ADSC-derived exosomes accelerate wound repair in a diabetic foot ulcer rat model by promoting vascularization[51]. In addition, ADSC-derived exosomes enhance angiogenesis in diabetic wounds through the miR-146a-5p/juxtaposed with another zinc finger gene 1 (JAZF1) axis, suggesting that noncoding RNA cargos are important mediators of exosome-driven vascular microenvironment improvement[52].

Exosomes from other cellular sources can also improve wound repair through angiogenic and antioxidant mechanisms. hUCMSC-derived exosomes reduce oxidative stress injury and promote angiogenesis, thereby accelerating diabetic cutaneous wound repair[36]. Epidermal stem cell-derived exosomes deliver miR-200b-3p to diabetic wounds and alleviate excessive autophagy-induced endothelial cell apoptosis, improving angiogenesis and wound healing[53]. Endothelial progenitor cell-derived exosomal miR-221-3p promotes angiogenesis-related repair responses and accelerates skin wound healing in diabetic mice[46]. These findings indicate that exosomes/sEVs do not act only on keratinocytes themselves. Instead, they also support the “migration–coverage–reconstruction” sequence by improving vascular supply, redox balance, and endothelial cell function.

Regulating fibroblast activation and ECM remodeling

Fibroblasts and the ECM are not passive backgrounds in wound repair. They actively regulate keratinocyte migration direction, migration efficiency, migration termination, and subsequent epithelial reconstruction. During normal wound healing, appropriate ECM deposition and remodeling provide a scaffold for keratinocyte migration and regulate epithelial behavior through matrix composition, mechanical properties, and adhesion signals. In diabetic wounds, insufficient ECM deposition, disorganized collagen alignment, abnormal protease activity, and imbalanced matrix remodeling may impair keratinocyte migration, delay wound coverage, and compromise barrier reconstruction. Therefore, exosomes/sEVs may provide indirect support for EMP-related repair in diabetic wounds by regulating fibroblast function and the ECM microenvironment.

ADSC-derived exosomes improve diabetic wound healing and inhibit scar fibrosis through the miR-128-1-5p/transforming growth factor-β1 (TGF-β1)/Smad axis, indicating that exosomes may not only promote wound closure but also limit aberrant transforming growth factor-β (TGF-β)/Smad activation and prevent excessive fibrosis or abnormal matrix deposition[54]. Platelet-rich plasma-derived exosomes have also been reported to promote diabetic wound repair by acting on fibroblasts[45]. In addition, autologous dermal fibroblast-derived exosomes promote diabetic cutaneous wound healing through the Akt/β-catenin pathway, suggesting that fibroblasts themselves may participate in communication between matrix repair and epithelial repair through exosomal signaling[41].

The vicious cycle of pathological EV communication and EMP impairment

The biological effects of EVs in diabetic wounds are bidirectional. In addition to the administration of exogenous pro-reparative vesicles, endogenous EVs released by cells exposed to hyperglycemia, oxidative stress, persistent inflammation, hypoxia, or matrix disorganization may carry altered cargos and transmit pathological information to neighboring cells. These vesicles may therefore function not merely as passive biomarkers of cellular stress, but as active components of the diabetic wound microenvironment that sustain intercellular dysfunction.

Evidence of such pathological communication has been reported in both directions across the epidermal–dermal interface. sEVs released by high-glucose-exposed keratinocytes induced autophagy in recipient fibroblasts and inhibited fibroblast migration and collagen synthesis, thereby impairing tissue repair[42]. Conversely, EVs derived from diabetic skin fibroblasts transferred long noncoding RNA upregulated in diabetic skin (lnc-URIDS) to keratinocytes, increased keratinocyte MMP-9 expression, and delayed diabetic wound healing[43]. These studies provide direct evidence of pathological EV-mediated communication between keratinocytes and fibroblasts. However, because they did not resolve coordinated entry into and exit from EMP-associated states, they should not be interpreted as direct evidence of complete EMP trajectory disruption.

Together, these findings support a potential vicious cycle: diabetic metabolic and inflammatory stress alters EV secretion and cargo composition; pathological EVs then impair keratinocyte and fibroblast functions, increase proteolytic activity, disrupt matrix production, and delay epithelial coverage; the resulting persistent inflammation, matrix instability, and incomplete barrier recovery further reinforce cellular stress and abnormal EV communication. In this model, pathological EVs may impair EMP-related repair both by restricting the acquisition of an effective migratory phenotype and by preventing the subsequent restoration of epithelial organization. This interpretation remains partly inferential and requires validation using temporally resolved keratinocyte markers, spatial analysis, and barrier-function endpoints.

Recognizing this negative communication network provides an important conceptual contrast with therapeutic exosomes/sEVs. Exogenous or engineered vesicles are intended to interrupt the pathological cycle by improving inflammatory resolution, vascular supply, redox balance, fibroblast function, and keratinocyte repair. Their therapeutic efficacy should therefore be evaluated not only by accelerated wound closure, but also by whether they suppress endogenous pathogenic EV signaling and support the temporal coordination between keratinocyte migration, wound coverage, redifferentiation, and barrier reconstruction.

Therapeutic strategies and translational challenges of exosomes/sEVs in diabetic wound re-epithelialization

From natural to engineered exosomes: improving controllability and mechanistic specificity

Natural exosomes/sEVs have several advantages as cell-free therapeutic candidates for diabetic wounds, including low immunogenicity, the capacity to deliver multiple bioactive cargos, and the ability to act on several repair-related cell types simultaneously. In particular, MSC-derived exosomes/sEVs have become one of the most extensively investigated vesicle types in diabetic wound research because they can regulate immune responses, angiogenesis, fibroblast function, and tissue repair processes[15,18,55]. However, natural exosomes are not standardized drugs with fixed composition and clearly defined mechanisms. Their bioactive cargos are influenced by donor cell source, donor status, culture conditions, preconditioning strategy, isolation and purification methods, and storage conditions, leading to considerable batch-to-batch variability. In addition, natural exosomes have limitations such as restricted yield, variable cargo loading, insufficient tissue targeting, unclear in vivo fate, and co-isolation of impurities, all of which hinder direct clinical translation.

For these reasons, engineered exosomes have become an important direction in diabetic wound therapy. Current engineering strategies include preconditioning donor cells with hypoxia, inflammatory factors, drugs, or other stimuli to optimize vesicle cargos; genetic modification to enrich selected miRNAs, circRNAs, or proteins; membrane modification to enhance targeting and tissue retention; and integration with biomaterials such as hydrogels, dressings, microneedles, or silk fibroin patches to improve local release and therapeutic stability[19–21]. The common goal of these approaches is not simply to increase vesicle quantity, but to transform exosomes from natural paracrine products into therapeutic tools that are designable, controllable, and evaluable.

Preclinical studies have provided initial evidence supporting engineered exosomes for diabetic wound repair. miR-146a-loaded engineered exosomes released from a silk fibroin patch promote diabetic wound healing by targeting interleukin-1 receptor-associated kinase 1 (IRAK1) and improving the inflammatory microenvironment[56]. Engineered miR-31-enriched exosomes have also been reported to accelerate diabetic wound healing by enhancing angiogenesis, fibrogenesis, and re-epithelialization[57]. These studies suggest that engineered exosomes can be designed to target key pathological barriers in diabetic wounds, including persistent inflammation, insufficient angiogenesis, excessive oxidative stress, abnormal matrix remodeling, and delayed re-epithelialization.

However, enhanced potency should not be equated with biological precision. Engineered miR-31-enriched exosomes have been shown to promote angiogenesis, fibrogenesis, and re-epithelialization in diabetic wounds[57], and miR-31 itself enhances keratinocyte proliferation and migration during cutaneous repair[58]. Nevertheless, the biological effects of a single miRNA are pleiotropic and context dependent. For example, nuclear factor κB (NF-κB)-induced miR-31 promotes keratinocyte hyperproliferation and epidermal hyperplasia in inflammatory skin disease[59]. These observations do not demonstrate that miR-31-engineered exosomes cause adverse epidermal effects in diabetic wounds; rather, they illustrate why sustained or supraphysiological enrichment of a single regulatory cargo should not automatically be assumed to produce stage-appropriate repair.

From an EMP perspective, an intervention that strongly enhances migration and proliferation during the early phase may become counterproductive if the same signal persists after wound coverage and interferes with keratinocyte growth arrest, redifferentiation, junctional reconstruction, or barrier maturation. Similar concerns apply to cargos that broadly stimulate angiogenesis, fibroblast activation, or TGF-β-related signaling, because excessive or prolonged activity could alter vascular maturation, matrix organization, or scar formation. Potential over-engineering risks therefore include off-target regulation across multiple recipient-cell types, loss of the physiological balance among vesicle cargos, excessive cargo exposure, inappropriate duration of signaling, and mismatch between therapeutic activity and wound-healing stage.

Accordingly, the value of engineered exosomes should be judged not only by increased therapeutic efficacy, but also by whether engineering improves mechanistic specificity, temporal controllability, and testability. Rational design should define the intended recipient cell, therapeutic window, cargo dose, duration of exposure, and expected stage-specific effect. For EMP-relevant applications, early-phase assays should assess keratinocyte activation, collective migration, and wound coverage, whereas later-phase assays should determine whether the intervention can be withdrawn or attenuated sufficiently to permit redifferentiation, restoration of E-cadherin, claudin-1, and occludin, epithelial stratification, and functional barrier recovery.

Safety assessment should similarly extend beyond acute cytotoxicity and short-term wound closure. Relevant evaluations include cargo copy number and loading consistency, dose–response relationships, vesicle biodistribution and persistence, unintended uptake by non-target cells, late inflammatory or fibrotic effects, epidermal architecture, scar quality, recurrence, and the reversibility of engineered signaling. At present, most engineered exosome studies remain at the cell or rodent level and do not establish whether enhanced early repair is accompanied by normal late epithelial maturation. The goal should therefore be controlled correction of a stage-specific pathological barrier rather than maximal and persistent activation of a single pro-repair pathway.

Local delivery systems: a realistic route for translating exosome-based diabetic wound therapy

A major advantage of exosome/sEV-based therapy for diabetic cutaneous wounds is the feasibility of local administration. Diabetic wounds are usually superficial, visible, and accessible for repeated treatment, allowing exosomes/sEVs to act locally without relying entirely on systemic delivery. Compared with direct application or injection of free exosomes, local delivery systems can increase retention at the wound site, reduce rapid clearance, protect vesicle bioactivity, and enable sustained, controlled, or responsive release. Thus, combining exosomes with hydrogels, microneedles, dressings, patches, or composite scaffolds may represent one of the most practical routes for translating exosome-based approaches into local diabetic wound therapy.

Hydrogels are among the most widely used material platforms for exosome delivery. Their high water content, favorable biocompatibility, and tunable mechanical properties allow them to create a relatively moist and stable wound microenvironment while protecting exosomes and enabling controlled release. hUCMSC-derived exosomes combined with Pluronic F127 thermosensitive hydrogel have been shown to promote chronic diabetic wound healing and skin regeneration in rats[60]. A matrix metalloproteinase (MMP)-responsive poly(ethylene glycol) (PEG) hydrogel loaded with ADSC-derived exosomes enables enzyme-responsive release in the MMP-rich diabetic wound microenvironment and promotes wound repair by optimizing cellular functions and alleviating oxidative stress[61]. In addition, hypoxia-preconditioned ADSC-derived exosome-embedded hydrogels promote angiogenesis and accelerate diabetic wound healing[62]. These studies indicate that hydrogels are not merely passive vesicle carriers, but can enhance exosome efficacy by modulating release kinetics, maintaining a moist wound environment, and prolonging local bioactivity.

When evaluating “exosome plus hydrogel” strategies, it is important to distinguish evidence for material feasibility from evidence for diabetic wound therapy. For example, exosome-loaded alginate hydrogel promotes wound closure, collagen synthesis, and angiogenesis in a full-thickness skin wound model, but this model was not specific to diabetic wounds[63]. Such findings are therefore better interpreted as material-based feasibility evidence rather than direct evidence of efficacy in diabetic wounds. The diabetic wound microenvironment is characterized by hyperglycemia, chronic inflammation, oxidative stress, increased protease activity, and insufficient angiogenesis, all of which may affect material degradation, exosome stability, and local biological effects. Therefore, diabetic wound-specific validation remains necessary.

Microneedle and patch systems are also promising local delivery tools. Microneedles can penetrate the stratum corneum or superficial wound barriers in a minimally invasive manner, enabling exosomes and co-delivered agents to enter local tissues more effectively and allowing controlled release through material design. A gelatin methacryloyl/poly(ethylene glycol) diacrylate (GelMA/PEGDA) microneedle patch loaded with human umbilical vein endothelial cell (HUVEC)-derived exosomes and tazarotene promotes cell migration, angiogenesis, and wound repair in diabetic wound models through controlled release of both vesicles and drug[64]. M2 macrophage-derived exosome-encapsulated microneedles combined with mild photothermal therapy also suppress excessive inflammation and promote vascular regeneration in diabetic rat wounds[65]. These strategies improve exosome delivery efficiency and provide a basis for constructing composite therapeutic platforms such as “exosome plus drug,” “exosome plus immunomodulation,” and “exosome plus physical stimulation.”

From the perspective of EMP-related re-epithelialization, stage-matched release is an important design objective but has not yet been fully achieved for exosome-based diabetic wound therapy. Existing MMP-responsive exosome-loaded hydrogels demonstrate that vesicle release can be coupled to a protease-rich diabetic wound microenvironment[61]. In parallel, non-exosomal biomaterial studies have established the broader feasibility of programmed release across different wound-healing stages. A microenvironment-responsive hydrogel achieved spatiotemporal sequential delivery in infected chronic diabetic wounds[66], whereas a near-infrared-responsive bilayer hydrogel enabled an early antibacterial phase followed by a later pro-angiogenic phase[67]. These platforms provide proof of material feasibility, but they should not be regarded as direct evidence that exosome delivery can yet control keratinocyte EMP trajectories.

A future phase-specific exosome system might provide anti-inflammatory, antioxidant, or migration-supporting activity during the early wound phase and subsequently reduce, terminate, or switch its activity to favor keratinocyte redifferentiation, junctional reconstruction, and barrier maturation. Achieving this goal will require more than sustained release. A credible on-demand platform should demonstrate that a defined wound-associated signal—such as protease activity, pH, ROS, glucose concentration, or an externally applied stimulus—predictably alters release kinetics while preserving vesicle integrity and biological potency. Studies should compare responsive delivery with free vesicles and conventional sustained release, and should link release profiles to temporally resolved keratinocyte, EMP/MET, and barrier-function endpoints. Until these requirements are met, phase-specific exosome delivery should be presented as a promising engineering direction rather than an established therapeutic capability.

Translational challenges: standardization, quality control, efficacy endpoints, and evidence levels

Although preclinical studies remain the dominant component of the evidence base, clinical evaluation of exosome/sEV-based therapy for diabetic foot ulcers has now begun to emerge. Systematic reviews and meta-analyses support an overall pro-healing effect across experimental diabetic wound studies, including improvements in wound closure, angiogenesis, re-epithelialization, and collagen-related outcomes[16,68]. Nevertheless, substantial heterogeneity remains in vesicle source, isolation and characterization, dose definition, administration route, delivery material, animal model, and endpoint selection. These pooled preclinical results therefore support biological and therapeutic potential but do not, by themselves, establish a standardized or clinically mature treatment. The current translational evidence landscape and the principal requirements for progression are summarized in Table 2.

First, inadequate standardization of nomenclature, isolation, and characterization remains a major barrier to reproducibility. According to MISEV2023-related principles, extracellular vesicle studies should improve consistency in nomenclature, sample source, isolation method, particle characterization, purity assessment, marker detection, and functional validation[14]. However, in diabetic wound research, studies differ markedly in donor cell type, culture conditions, preconditioning strategy, isolation and purification methods, particle size range, marker combinations, and quantification approaches. Some studies still use the term “exosomes” broadly without adequately proving endosomal origin or excluding contamination by other extracellular vesicles or protein complexes. Future studies should therefore use more standardized EV or sEV terminology whenever appropriate and clearly report isolation methods, particle characteristics, marker profiles, and purity controls to improve comparability across studies.

Second, key quality attributes and potency assays remain insufficiently defined. As complex bioactive products, exosomes/sEVs should be evaluated not only by particle size, concentration, morphology, and marker expression, but also by RNA, protein, and lipid cargo profiles, sterility and endotoxin control, impurity residues, storage stability, batch-to-batch consistency, and biological potency.

For diabetic wound therapy, potency assays should not be limited to cell proliferation, migration, or wound closure rate. They should also incorporate inflammation regulation, angiogenesis, oxidative stress, ECM remodeling, quality of re-epithelialization, temporally resolved EMP-associated and redifferentiation markers, and barrier function recovery. Only when potency assays are aligned with therapeutic mechanisms can exosomes/sEVs move from experimental pro-healing products to evaluable, releasable, and reproducibly manufactured therapeutic products.

Third, dose, administration mode, and efficacy endpoints have not been standardized. Current diabetic wound studies vary widely in exosome dose, whether dose is expressed by particle number or protein amount, dosing frequency, administration route, delivery material, observation period, and endpoint selection. Some studies use local injection, whereas others use topical application or material-based loading; some quantify exosomes by protein content, whereas others use particle counts. These differences make cross-study comparison difficult. For clinical translation, future studies need to define the minimum effective dose, optimal dosing frequency, local delivery route, treatment schedule, and safety range. Efficacy endpoints should also extend beyond reduction of wound area to include re-epithelialization quality, vascular maturation, infection control, pain, recurrence, amputation risk, quality of life, and long-term safety.

Fourth, human clinical evidence is now available but remains early and should be interpreted separately from direct evidence of EMP regulation. A randomized controlled clinical trial enrolled 110 patients with persistent diabetic foot ulcers and compared topical Wharton’s jelly-derived MSC (WJ-MSC) exosome gel plus standard of care with standard care alone and a vehicle control[69]. The study reported a higher complete-healing rate and a shorter time to complete epithelialization in the exosome-treated group. However, only 85 participants were included in the final per-protocol analysis, baseline ulcer areas differed among groups, and the study did not assess keratinocyte-state trajectories, EMP/MET markers, epidermal-barrier quality, or product-specific potency biomarkers. It therefore provides encouraging clinical wound-healing evidence, but not clinical evidence of dynamic EMP regulation.

A separate single-center phase I/II open-label study treated ten patients with refractory diabetic foot ulcers using perilesional injections of allogeneic hUCMSC derivatives containing conditioned medium, extracellular vesicles, and exosomes[70]. Complete ulcer closure and an absence of treatment-related serious adverse events were reported, with no recurrence documented during follow-up. However, the small uncontrolled design and the use of a composite MSC-derivative preparation prevent the therapeutic contribution of exosomes/sEVs from being isolated. Collectively, these studies indicate that clinical translation has moved beyond a purely hypothetical stage, while emphasizing the need for independently replicated trials using compositionally defined vesicle products, standardized dosing, rigorous controls, and mechanistically aligned endpoints.

Fifth, regulatory pathways and product classification remain to be clarified. Exosomes/sEVs contain bioactive cargos and are derived from cells; when combined with hydrogels, patches, microneedles, or other materials, they may become combination products involving both cell-derived biologics and medical devices. Different countries or regions may classify exosome products differently, such as biologics, drugs, cell-derived products, or combination therapies. Therefore, clinical translation of exosomes/sEVs requires not only mechanistic evidence of wound healing efficacy but also compliance with manufacturing and regulatory requirements, including sterility control, endotoxin testing, impurity assessment, viral safety, batch consistency, storage stability, release criteria, and traceability.

For engineered or material-combined exosomes, additional risks related to genetic modification, cargo loading, material degradation products, and local immune responses must also be evaluated.

Finally, successful translation will require models and clinical trial designs that more closely reflect real diabetic wounds. Diabetic foot ulcers are often accompanied by infection, ischemia, neuropathy, abnormal pressure, and multiple comorbidities, whereas most animal models cannot fully recapitulate these complex conditions. Future studies should validate exosome/sEV efficacy and safety in chronic wound models that better approximate clinical scenarios and should independently replicate the emerging clinical findings through rigorously designed multicenter trials with adequate sample sizes, standardized vesicle products, clearly defined endpoints, and sufficient follow-up. Only after standardization of preparation, mechanistic validation, quality control, delivery strategy, and clinical evidence have been established, can exosomes/sEVs become a meaningful component of cell-free therapy for supporting re-epithelialization in diabetic cutaneous wounds.

Conclusion and perspectives

Diabetic cutaneous wound healing cannot be adequately evaluated by wound-area reduction alone. Effective repair requires keratinocytes to acquire a transient migratory and proliferative state, collectively cover the wound surface, subsequently withdraw from that state, and restore epithelial junctions, stratification, differentiation, and barrier function. EMP provides a more biologically appropriate framework than a binary EMT model for describing these reversible and stage-dependent keratinocyte transitions. The principal contribution of this review is to place keratinocyte-centered re-epithelialization and EMP at the core of exosome/sEV research, rather than treating epithelial repair as one endpoint within a general catalogue of anti-inflammatory, pro-angiogenic, antioxidant, and matrix-regulatory effects.

When interpreted using the proposed evidence hierarchy, the current literature supports the therapeutic potential of exosomes/sEVs but also reveals a substantial mechanistic gap. Several studies provide functional evidence that vesicles enhance keratinocyte migration, proliferation, or histological re-epithelialization, whereas a larger body of evidence demonstrates indirect support through macrophage regulation, angiogenesis, oxidative-stress reduction, fibroblast function, and ECM remodeling. However, no representative study currently provides sufficiently resolved evidence that exosome/sEV treatment coordinates the complete temporal sequence of EMP-associated keratinocyte state transitions and subsequent epithelial redifferentiation in diabetic wounds. Improved wound closure or expression of an isolated EMT-related marker should therefore not be interpreted as proof of dynamic EMP regulation.

EV communication in diabetic wounds is also bidirectional and context dependent. Therapeutic vesicles may improve the local environment and support epithelial repair, whereas endogenous EVs released by cells exposed to metabolic or inflammatory stress may transfer pathological cargos, impair epidermal–dermal communication, increase proteolytic activity, and reinforce chronic non-healing. Engineered vesicles and local delivery systems offer opportunities to improve cargo selection, tissue retention, and release control, but greater potency does not necessarily ensure appropriate biological timing. Early stimulation of migration and proliferation must ultimately be compatible with signal withdrawal, keratinocyte redifferentiation, and barrier maturation. Emerging human studies indicate that clinical translation has begun, but the available evidence remains preliminary, product heterogeneous, and mechanistically disconnected from EMP-specific endpoints[69,70].

Future studies should integrate coordinated epithelial, migration-associated, transcriptional, and barrier markers, including E-cadherin, vimentin, GRHL2, ZEB1, SNAI1/Snail, SNAI2/Slug, claudin-1, and occludin, with temporally and spatially resolved methods such as single-cell and spatial transcriptomics, trajectory analysis, lineage tracing, and live imaging. These approaches should be combined with functional evaluation of collective migration, epithelial stratification, junctional reconstruction, transepidermal barrier recovery, scar quality, and wound recurrence. Translation will additionally require standardized vesicle nomenclature and characterization, defined critical quality attributes and potency assays, reproducible dosing, stage-matched delivery, long-term safety assessment, and rigorously controlled clinical trials. Exosomes/sEVs should therefore be viewed not as established EMP-directed therapeutics, but as promising and potentially designable regulators of intercellular communication whose effects on keratinocyte plasticity remain to be directly demonstrated.

References

[1] Armstrong DG, Tan TW, Boulton AJM, Bus SA. Diabetic foot ulcers: a review. JAMA. 2023;330(1):62-75.

[2] Zhang P, Lu J, Jing Y, Tang S, Zhu D, Bi Y. Global epidemiology of diabetic foot ulceration: a systematic review and meta-analysis. Ann Med. 2017;49(2):106-116.

[3] Chen L, Sun S, Gao Y, Ran X. Global mortality of diabetic foot ulcer: a systematic review and meta-analysis of observational studies. Diabetes Obes Metab. 2023;25(1):36-45.

[4] Holl J, Kowalewski C, Zimek Z, et al. Chronic diabetic wounds and their treatment with skin substitutes. Cells. 2021;10(3):655.

[5] Deng L, Du C, Song P, et al. The role of oxidative stress and antioxidants in diabetic wound healing. Oxid Med Cell Longev. 2021;2021:8852759.

[6] Wu X, He W, Mu X, et al. Macrophage polarization in diabetic wound healing. Burns Trauma. 2022;10:tkac051.

[7] Fang WC, Lan CE. The epidermal keratinocyte as a therapeutic target for management of diabetic wounds. Int J Mol Sci. 2023;24(5):4290.

[8] Rousselle P, Braye F, Dayan G. Re-epithelialization of adult skin wounds: cellular mechanisms and therapeutic strategies. Adv Drug Deliv Rev. 2019;146:344-365.

[9] Haensel D, Dai X. Epithelial-to-mesenchymal transition in cutaneous wound healing: where we are and where we are heading. Dev Dyn. 2018;247(3):473-480.

[10] Vu R, Dragan M, Sun P, Werner S, Dai X. Epithelial-mesenchymal plasticity and endothelial-mesenchymal transition in cutaneous wound healing. Cold Spring Harb Perspect Biol. 2023;15(8):a041237.

[11] Yao W, Wang Z, Ma H, et al. Epithelial-mesenchymal plasticity (EMP) in wound healing: exploring EMT mechanisms, regulatory network, and therapeutic opportunities. Heliyon. 2024;10(14):e34269.

[12] Sen CK, Roy S. The hyperglycemia stranglehold stifles cutaneous epithelial‒mesenchymal plasticity and functional wound closure. J Invest Dermatol. 2021;141(6):1382-1385.

[13] Singh K, Sinha M, Pal D, et al. Cutaneous epithelial to mesenchymal transition activator ZEB1 regulates wound angiogenesis and closure in a glycemic status-dependent manner. Diabetes. 2019;68(11):2175-2190.

[14] Welsh JA, Goberdhan DCI, O’Driscoll L, et al. Minimal information for studies of extracellular vesicles (MISEV2023): from basic to advanced approaches. J Extracell Vesicles. 2024;13(2):e12404.

[15] Jiang M, Jiang X, Li H, et al. The role of mesenchymal stem cell-derived EVs in diabetic wound healing. Front Immunol. 2023;14:1136098.

[16] Qiao Z, Wang X, Zhao H, et al. The effectiveness of cell-derived exosome therapy for diabetic wound: a systematic review and meta-analysis. Ageing Res Rev. 2023;85:101858.

[17] Shen J, Zhao X, Zhong Y, et al. Exosomal ncRNAs: the pivotal players in diabetic wound healing. Front Immunol. 2022;13:1005307.

[18] Li Y, Zhu Z, Li S, et al. Exosomes: compositions, biogenesis, and mechanisms in diabetic wound healing. J Nanobiotechnology. 2024;22(1):398.

[19] Liu L, Liu D. Bioengineered mesenchymal stem cell-derived exosomes: emerging strategies for diabetic wound healing. Burns Trauma. 2024;12:tkae030.

[20] Guo L, Xiao D, Xing H, Yang G, Yang X. Engineered exosomes as a prospective therapy for diabetic foot ulcers. Burns Trauma. 2024;12:tkae023.

[21] Ye H, Wang F, Xu G, Shu F, Fan K, Wang D. Advancements in engineered exosomes for wound repair: current research and future perspectives. Front Bioeng Biotechnol. 2023;11:1301362.

[22] Pastar I, Stojadinovic O, Yin NC, et al. Epithelialization in wound healing: a comprehensive review. Adv Wound Care (New Rochelle). 2014;3(7):445-464.

[23] Safferling K, Sütterlin T, Westphal K, et al. Wound healing revised: a novel reepithelialization mechanism revealed by in vitro and in silico models. J Cell Biol. 2013;203(4):691-709.

[24] Aragona M, Dekoninck S, Rulands S, et al. Defining stem cell dynamics and migration during wound healing in mouse skin epidermis. Nat Commun. 2017;8:14684.

[25] Park S, Gonzalez DG, Guirao B, et al. Tissue-scale coordination of cellular behaviour promotes epidermal wound repair in live mice. Nat Cell Biol. 2017;19:155-163.

[26] Sun X, Joost S, Kasper M. Plasticity of epithelial cells during skin wound healing. Cold Spring Harb Perspect Biol. 2023;15(5):a041232.

[27] Chen T, Zhang B, Xie H, Huang C, Wu Q. GRHL2 regulates keratinocyte EMT-MET dynamics and scar formation during cutaneous wound healing. Cell Death Dis. 2024;15(10):748.

[28] Li L, Zhang J, Zhang Q, et al. High glucose suppresses keratinocyte migration through the inhibition of p38 MAPK/autophagy pathway. Front Physiol. 2019;10:24.

[29] Yang B, Alimperti S, Gonzalez MV, et al. Reepithelialization of diabetic skin and mucosal wounds is rescued by treatment with epigenetic inhibitors. Diabetes. 2024;73(1):120-134.

[30] Tan MWY, Tan WR, Kong ZQ, et al. High glucose restraint of acetylcholine-induced keratinocyte epithelial-mesenchymal transition is mitigated by p38 inhibition. J Invest Dermatol. 2021;141(6):1438-1449.e9.

[31] Singh K, Rustagi Y, Abouhashem AS, et al. Genome-wide DNA hypermethylation opposes healing in patients with chronic wounds by impairing epithelial-mesenchymal transition. J Clin Invest. 2022;132(17):e157279.

[32] Kalluri R, LeBleu VS. The biology, function, and biomedical applications of exosomes. Science. 2020;367(6478):eaau6977.

[33] Wang J, Wu H, Peng Y, et al. Hypoxia adipose stem cell-derived exosomes promote high-quality healing of diabetic wound involves activation of PI3K/Akt pathways. J Nanobiotechnology. 2021;19(1):202.

[34] Zhang Y, Bai X, Shen K, et al. Exosomes derived from adipose mesenchymal stem cells promote diabetic chronic wound healing through SIRT3/SOD2. Cells. 2022;11(16):2568.

[35] Ren S, Chen J, Guo J, et al. Exosomes from adipose stem cells promote diabetic wound healing through the eHSP90/LRP1/AKT axis. Cells. 2022;11(20):3229.

[36] Yan C, Xv Y, Lin Z, et al. Human umbilical cord mesenchymal stem cell-derived exosomes accelerate diabetic wound healing via ameliorating oxidative stress and promoting angiogenesis. Front Bioeng Biotechnol. 2022;10:829868.

[37] Teng L, Maqsood M, Zhu M, et al. Exosomes derived from human umbilical cord mesenchymal stem cells accelerate diabetic wound healing via promoting M2 macrophage polarization, angiogenesis, and collagen deposition. Int J Mol Sci. 2022;23(18):10421.

[38] Wang P, Theocharidis G, Vlachos IS, et al. Exosomes derived from epidermal stem cells improve diabetic wound healing. J Invest Dermatol. 2022;142(9):2508-2517.e13.

[39] Yang H, Xu H, Wang Z, et al. Analysis of miR-203a-3p/SOCS3-mediated induction of M2 macrophage polarization to promote diabetic wound healing based on epidermal stem cell-derived exosomes. Diabetes Res Clin Pract. 2023;197:110573.

[40] Kuang L, Zhang C, Li B, Deng H, Chen R, Li G. Human keratinocyte-derived exosomal MALAT1 promotes diabetic wound healing by upregulating MFGE8 via microRNA-1914-3p. Int J Nanomedicine. 2023;18:949-970.

[41] Han X, Wu P, Li L, et al. Exosomes derived from autologous dermal fibroblasts promote diabetic cutaneous wound healing through the Akt/β-catenin pathway. Cell Cycle. 2021;20(5-6):616-629.

[42] Hong X, Cai L, Li L, et al. Keratinocyte-derived small extracellular vesicles delay diabetic wound healing by triggering fibroblasts autophagy. Arch Physiol Biochem. 2025;131(1):11-23.

[43] Wu Y, Wu X, Wang J, et al. Fibroblast-derived extracellular vesicle-packaged long noncoding RNA upregulated in diabetic skin enhances keratinocyte MMP-9 expression and delays diabetic wound healing. Lab Invest. 2023;103(3):100019.

[44] Guo SC, Tao SC, Yin WJ, Qi X, Yuan T, Zhang CQ. Exosomes derived from platelet-rich plasma promote the re-epithelization of chronic cutaneous wounds via activation of YAP in a diabetic rat model. Theranostics. 2017;7(1):81-96.

[45] Cao W, Meng X, Cao F, Wang J, Yang M. Exosomes derived from platelet-rich plasma promote diabetic wound healing via the JAK2/STAT3 pathway. iScience. 2023;26(11):108236.

[46] Xu J, Bai S, Cao Y, et al. miRNA-221-3p in endothelial progenitor cell-derived exosomes accelerates skin wound healing in diabetic mice. Diabetes Metab Syndr Obes. 2020;13:1259-1270.

[47] Liu Y, Wang P, Li J, et al. Single-cell RNA sequencing reveals the impaired epidermal differentiation and pathological microenvironment in diabetic foot ulcer. Burns Trauma. 2025;13:tkae065.

[48] Huang Q, Chu Z, Wang Z, et al. circCDK13-loaded small extracellular vesicles accelerate healing in preclinical diabetic wound models. Nat Commun. 2024;15(1):3904.

[49] Yin D, Shen G. Exosomes from adipose-derived stem cells regulate macrophage polarization and accelerate diabetic wound healing via the circ-Rps5/miR-124-3p axis. Immun Inflamm Dis. 2024;12(6):e1274.

[50] Zhou X, Ye C, Jiang L, et al. The bone mesenchymal stem cell-derived exosomal miR-146a-5p promotes diabetic wound healing in mice via macrophage M1/M2 polarization. Mol Cell Endocrinol. 2024;579:112089.

[51] Li X, Xie X, Lian W, et al. Exosomes from adipose-derived stem cells overexpressing Nrf2 accelerate cutaneous wound healing by promoting vascularization in a diabetic foot ulcer rat model. Exp Mol Med. 2018;50(4):1-14.

[52] Che D, Xiang X, Xie J, Chen Z, Bao Q, Cao D. Exosomes derived from adipose stem cells enhance angiogenesis in diabetic wound via miR-146a-5p/JAZF1 axis. Stem Cell Rev Rep. 2024;20(4):1026-1039.

[53] Xu H, Yang H, Wang Z, et al. Epidermal stem cell derived exosomes alleviate excessive autophagy induced endothelial cell apoptosis by delivering miR200b-3p to diabetic wounds. J Invest Dermatol. 2024;144(5):1134-1147.e2.

[54] Liang Q, Zhou D, Ge X, et al. Exosomes from adipose-derived mesenchymal stem cell improve diabetic wound healing and inhibit fibrosis via miR-128-1-5p/TGF-β1/Smad axis. Mol Cell Endocrinol. 2024;588:112213.

[55] Soltani S, Zahedi A, Vergara AJS, et al. Preclinical therapeutic efficacy of extracellular vesicles derived from adipose-derived mesenchymal stromal/stem cells in diabetic wounds: a systematic review and meta-analysis. Stem Cell Rev Rep. 2024;20(8):2016-2031.

[56] Li Q, Hu W, Huang Q, et al. MiR146a-loaded engineered exosomes released from silk fibroin patch promote diabetic wound healing by targeting IRAK1. Signal Transduct Target Ther. 2023;8(1):62.

[57] Huang J, Yu M, Yin W, et al. Development of a novel RNAi therapy: engineered miR-31 exosomes promoted the healing of diabetic wounds. Bioact Mater. 2021;6(9):2841-2853.

[58] Li D, Li XI, Wang A, et al. MicroRNA-31 promotes skin wound healing by enhancing keratinocyte proliferation and migration. J Invest Dermatol. 2015;135(6):1676-1685.

[59] Yan S, Xu Z, Lou F, et al. NF-κB-induced microRNA-31 promotes epidermal hyperplasia by repressing protein phosphatase 6 in psoriasis. Nat Commun. 2015;6:7652.

[60] Yang J, Chen Z, Pan D, Li H, Shen J. Umbilical cord-derived mesenchymal stem cell-derived exosomes combined pluronic F127 hydrogel promote chronic diabetic wound healing and complete skin regeneration. Int J Nanomedicine. 2020;15:5911-5926.

[61] Jiang T, Liu S, Wu Z, et al. ADSC-exo@MMP-PEG smart hydrogel promotes diabetic wound healing by optimizing cellular functions and relieving oxidative stress. Mater Today Bio. 2022;16:100365.

[62] Hu N, Cai Z, Jiang X, et al. Hypoxia-pretreated ADSC-derived exosome-embedded hydrogels promote angiogenesis and accelerate diabetic wound healing. Acta Biomater. 2023;157:175-186.

[63] Shafei S, Khanmohammadi M, Heidari R, et al. Exosome loaded alginate hydrogel promotes tissue regeneration in full-thickness skin wounds: an in vivo study. J Biomed Mater Res A. 2020;108(3):545-556.

[64] Yuan M, Liu K, Jiang T, et al. GelMA/PEGDA microneedles patch loaded with HUVECs-derived exosomes and Tazarotene promote diabetic wound healing. J Nanobiotechnology. 2022;20(1):147.

[65] Zeng J, Sun Z, Zeng F, Gu C, Chen X. M2 macrophage-derived exosome-encapsulated microneedles with mild photothermal therapy for accelerated diabetic wound healing. Mater Today Bio. 2023;20:100649.

[66] Hu C, Liu W, Long L, et al. Microenvironment-responsive multifunctional hydrogels with spatiotemporal sequential release of tailored recombinant human collagen type III for the rapid repair of infected chronic diabetic wounds. J Mater Chem B. 2021;9(47):9684-9699.

[67] Huang X, Xu L, Yu X, et al. Near-infrared light-responsive multifunctional hydrogel releasing peptide-functionalized gold nanorods sequentially for diabetic wound healing. J Colloid Interface Sci. 2023;639:369-384.

[68] Chen L, Liu J, He Y, Zeng C, Liao W, Luo C. A systematic review and meta-analysis to investigate the effectiveness of exosome for diabetic wounds. J Tissue Viability. 2025;34(3):100917.

[69] Kishta MS, Hafez AM, Hydara T, et al. The transforming role of Wharton’s jelly mesenchymal stem cell-derived exosomes for diabetic foot ulcer healing: a randomized controlled clinical trial. Stem Cell Res Ther. 2025;16(1):559.

[70] Jafar H, Almousa R, Alhawari H, et al. Human umbilical cord mesenchymal stromal cells derivatives in treating diabetic foot ulcers: a phase I/II safety and efficacy trial. Stem Cell Res Ther. 2025;16(1):657.

Rights & permissions

The Author(s) 2026. This article is published by Higher Education Press on behalf of People’s Medical Publishing House.