Abstract
The regeneration of orodental and craniofacial tissues remains a primary clinical challenge owing to the requirement of harmonized integration of hard–soft tissues, immune modulation, neurogenesis, angiogenesis, and osteogenesis. Extracellular vesicles (EVs) are nanoparticles containing RNAs, metabolites, lipids, and proteins that have surfaced as vital paracrine regulators able to coordinate these intricate mechanisms without the intrinsic risks associated with cell transplantation. This review synthesizes extensive in vivo evidence and evolving clinical findings highlighting that EVs from immune cells, mesenchymal stem cells, dental stem cells, and engineered origins regulate periodontal and dental pulp regeneration, peri-implant osseointegration, temporomandibular joint disorders, root resorption, orthodontic tooth movement, and craniomaxillofacial bone repair. EVs function as context-based biological regulators: regenerative EVs enhance M2 (anti-inflammatory) macrophage polarization, neurogenesis, angiogenesis, and odonto-osteogenesis, while EVs from pathogenic, mechanically stressed, or inflamed cells can initiate inflammation, osteoclastogenesis, and tissue damage. EV long noncoding RNAs and microRNAs play a critical part in dictating reparative versus destructive outcomes by regulating the TGF-β/SMAD, PI3K/protein kinase B, NF-κB, and immune-metabolic cascades. Biomaterial-based delivery platforms and engineered EVs, such as surface-modified dental implants, scaffolds, and hydrogels, consistently improved local retention, efficacy, and multi-tissue repair. Crucially, human studies indicate clinically significant enhancements in periodontal and alveolar bone regeneration, emphasizing their translational promise. Nonetheless, clinical implementation is restricted by heterogeneity in EV origins, cargo variability, manufacturing methods, dosage, and delivery approaches, together with an incomplete comprehension of how to predict their bioactivity. Future advances will mandate Minimal Information for Studies of Extracellular Vesicles (MISEV)–guided optimization, functional release criteria, localized delivery systems, potency-driven classification, and scalable Good Manufacturing Practices production. Overall, current evidence positions EVs as cell-free, adaptable mediators with transformative potential for precision-guided craniofacial therapy and regenerative dentistry.
Keywords
Introduction
The regeneration of oral, dental, and craniofacial tissues continues to pose substantial clinical challenges owing to the intricate correlation between soft–hard tissue coupling, neurogenesis, angiogenesis, immune modulation, and inflammation (Yelick and Sharpe 2019). Extracellular vesicles (EVs) are lipid-bilayer–encapsulated Nanoparticles (NPs) released by cells and have only recently been recognized as central regulators of intercellular communication within pathological and physiological processes (Valadi et al 2007). These vesicles are usually categorized into exosomes, microvesicles, and apoptotic bodies, which are differentiated based on their function, cargo, size, secretion pathways, and biogenesis (Doyle and Wang 2019). Nonetheless, according to the recent Minimal Information for Studies of Extracellular Vesicles (MISEV) 2023 guidelines (Welsh et al 2024), the utilization of these biogenesis-based terms should be reserved for instances in which direct evidence shows the vesicle source and operational definitions based on measurable features, including cell or tissue origin, surface markers, density, and size.
EVs originating from progenitors residing in dental tissues, immune cells, and mesenchymal stem cells (MSCs) modulate extracellular matrix (ECM) remodeling, neurogenesis, osteogenesis, angiogenesis, macrophage polarization, and inflammation resolution across a variety of oral and craniofacial contexts (Miron et al 2024). Crucially, EVs signaling is bidirectional and context based (i.e., EVs secreted from engineered or healthy cells enhance repair; Y. Zhang et al 2024), while EVs from pathogen-exposed, mechanically stressed, or inflamed cells can actively initiate tissue deterioration (Zhang et al 2025). This duality establishes EVs as not merely therapeutic agents but also as mechanistic mediators and biomarkers of disease advancement.
This review mainly synthesizes preclinical (animal) studies, which presently represent most of the published literature. Although initial case reports and series indicate translational potential, well-designed clinical trials are limited, emphasizing the requirement for future investigation. By incorporating disease-targeted applications and mechanistic insights, this review aims to elucidate the reparative/regenerative capabilities of EVs, pinpointing key knowledge gaps and offering a logical framework to steer future optimization and clinical applications in regenerative dentistry.
Periodontal Regeneration
In vivo research has established EVs, especially from MSCs and immune cells, as master regulators of immune homeostasis, periodontal inflammation, and tissue regeneration. As paracrine mediators, EVs modulate several regenerative pathways, including neural repair, angiogenesis, immune equilibrium, osteoblast–osteoclast interaction, and macrophage polarization, hence enhancing the restoration of periodontal structure and function. Notably, the regenerative potential of EVs is mainly based on their source and microenvironmental context. EVs from engineered, reparative, or healthy cells consistently enhance periodontal regeneration by facilitating a prohealing condition, tissue integration, and coordinated soft- and hard-tissue reconstruction (Ahmad et al 2025b).
EVs: Agents of Alveolar Bone Preservation
Alveolar bone preservation marks a vital early stage of periodontal regeneration, since maintaining bone volume and inhibiting osteoclast functionality are mandatory for subsequent tissue repair. EVs from healthy MSCs, macrophages, dental stem cells, and engineered sources contribute to periodontal regeneration by protecting alveolar bone architecture and restoring a proregenerative niche. Periodontal ligament stem cell (PDLSC)–EVs significantly enhanced collagen architecture, bone volume (BV)/total volume (TV), and bone mineral density (BMD) while alleviating osteoclast infiltration in rodent periodontitis models (Lei et al 2022). These effects underscore their involvement in coupling inflammation resolution with hard-tissue regeneration.
Similarly, differentiated dental pulp stem cell (DPSC)–EVs significantly improved osteocalcin (OCN) expression, inhibited inducible nitric oxide synthase (iNOS+) M1 macrophages, attenuated tumor necrosis factor–alpha (TNF-α), and averted alveolar bone loss (Qiao et al 2023), hence promoting a proregenerative microenvironment. When administered through injectable chitosan hydrogel, DPSC-EVs demonstrated superior antiosteoclastic and anti-inflammatory activity and extended local retention, modulated by miR-1246, showing that biomaterial-mediated delivery can further improve regenerative outcomes (Zhang et al 2021).
EVs from Schwann cells (SCs) (Cui et al 2025), dental follicle stem cells (DFSCs) (Liang et al 2024), and stem cells from human exfoliated deciduous teeth (SHEDs) (Yu et al 2024) enhanced osteodifferentiation and ECM deposition by upregulating OCN, collagen type I alpha-1 chain (COL1A1), and runt-related transcription factor-2 (RUNX2) while also promoting angiogenesis and, in the case of SC-EVs, neurogenesis to facilitate tissue regeneration. EVs from gingival mesenchymal stem cells (GMSCs) (J. Wang et al 2025) and bone marrow mesenchymal stem cells (BMSCs) (Dong et al 2025) exerted robust immunoregulatory effects by alleviating interleukin (IL)–6, IL-1β, and TNF-α; mitigating inflammatory infiltration; and enhancing the macrophage shift toward the M2 phenotype, partly via metabolic rewiring. miR-146a-5p further contributed by suppressing osteoclastogenesis through TNF receptor–associated factor 6 (TRAF6) signaling, hence promoting bone repair (Dong et al 2025).
Macrophage Polarization as Regenerative Axis
In the literature, the capability of therapeutic EVs to regulate macrophage phenotype is one of the most consistent themes. TNF-α–pretreated GMSC-EVs preserved alveolar bone via decreasing inflammation and stimulating anti-inflammatory (M2) macrophage infiltration (Nakao et al 2021). Genetic and molecular modification of EVs by overexpressing CXCR4 to enhance homing to inflamed periodontal tissues through the SDF-1/CXCL12 axis, and loading them with miR-126 to regulate target cell behavior, inhibited osteoclastogenesis and expedited periodontal repair (H. Luo et al 2023). Importantly, the regenerative ability of M2-EVs was diminished by IL-10 inhibition, confirming that EV-mediated macrophage polarization is functionally necessary for bone preservation (X. Chen et al 2022). IL-10 carried by M2-EVs regulated their regenerative potential by stimulating the IL-10/IL-10R pathway to promote osteogenesis and suppress osteoclastogenesis, and its suppression disrupted this equilibrium, hence alleviating bone preservation (X. Chen et al 2022).
Forkhead box O1 (FoxO1)–overexpressing PDLSC-EVs promoted M2 reprogramming while alleviating osteoclast counts and oxidative stress, correlating immune and metabolic reprogramming to periodontal regeneration (Niu et al 2024). FoxO1, a transcription factor, mediates oxidative stress, energy metabolism, apoptosis, and osteogenesis, and its overexpression in PDLSCs enriches the derived EVs with FoxO1 payload, promoting their capability to enhance osteogenesis, alleviate mitochondrial dysfunction and oxidative stress, and modulate M2 shift, hence promoting their immunoregulatory and regenerative potential (Niu et al 2024).
EV miRNAs Dictate Destructive versus Regenerative Outcomes
EV microRNAs (miRNAs) function as molecular modulators that dictate whether EVs regulate repair or inflammation. Pathogenic EV miRNAs (miR-184, miR-155-5p, miR-143-3p) enhance nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) stimulation, macrophage inflammation, lipid loading, and osteoclastogenesis (W.W. Yang et al 2024). Periodontal destruction is reversed by suppressing these miRNAs. Contrarily, regenerative miRNAs (long noncoding RNA [lncRNA] HCP5, miR-203-3p, miR-146a-5p, miR-205-5p, miR-200b/c, miR-126, miR-1246) inhibit TRAF6 and NF-κB signaling, enhance osteoblast differentiation and angiogenesis, M2 polarization, and rebalance regulatory T cells (Tregs)/T helper 17 cells (Dong et al 2025). Knocking these miRNAs down diminished EVs-modulated bone repair, confirming their causal significance.
Engineered EVs Offer Superior Periodontal Regeneration
Surface-modified EVs, scaffolds, and hydrogels substantially enhanced EV residency and biological efficacy. Polyphenol- or nanocurcumin-containing EVs, injectable hydrogel-loaded umbilical cord mesenchymal stem cell (UCMSC)–EVs, hyaluronic acid–copper ion hydrogel-incorporating SHED-EVs, DPSC-EVs in chitosan, and alginate–gelatin hydrogel-containing PDLSC-EVs all yielded near-complete repair of collagen organization, cementum, PDL, and bone, significantly surpassing free EVs (Fan et al 2025).
Plant EVs are structurally similar to human EVs in that they are nanovesicles with a phospholipid bilayer capable of delivering biomolecules; nonetheless, their biogenesis and extraction differ, as they are not usually isolated from controlled cell culture secretion but are extracted from plant tissues using physical approaches. Compared with mammalian EVs, plant EVs can be generated at a larger scale, with higher stability and lower cost, and they naturally carry miRNAs, lipids, and flavonoids that can be uptaken by mammalian cells or even bacteria to exert antimicrobial, antioxidant, and anti-inflammatory effects, hence facilitating tissue regeneration. Plant EVs (garlic and ginger) conferred antimicrobial, antioxidant, mitochondrial-protective, and antimicrobial features, enabling the concurrent inhibition of Pg, osteoclastic function, and immune inflammation (Sundaram et al 2019; Xie et al 2024; Yu et al 2025).
EV-Mediated Multi-tissue Periodontal Regeneration
EVs coordinate multi-tissue regeneration. SC-EVs and polyphenol-modified EVs stimulate osteogenesis, angiogenesis, and neurogenesis simultaneously, causing functionally and structurally cohesive periodontal regeneration (Fan et al 2025). Likewise, PDLSC- and DFSC-EVs attract endogenous progenitor cells and facilitate innervated and vascularized bone and PDL production (Fig 1) (Liang et al 2024), a key factor for prolonged periodontal stability.

Comparison of extracellular vesicles (EVs) (exosomes; EXOs) and extruded nanovesicles (eNVs) in rat periodontal repair. (
Pulp Regeneration
EVs function as critical modulators of dentin regeneration, angiogenesis, immune regulation, pulp viability, and neurogenesis, rendering them as functional alternatives for SC treatment in endodontics (Ahmad et al 2025a) (Fig 2).

(
Oxidative Stress and Inflammation Resolution
DFSC-EVs significantly inhibited oxidative damage and acute pulp inflammation in a lipopolysaccharide-induced pulpitis rodent model, decreasing 8-hydroxy-2′-deoxyguanosine positive (a biomarker of oxidative DNA damage) in DPSCs (Li et al 2024). This protective effect permitted vigorous reparative dentinogenesis, characterized by enhanced dentin sialophosphoprotein (DSPP) expression and mineralized barrier generation (Li et al 2024). The administration of a reactive oxygen species–responsive injectable hydrogel enhanced these effects, facilitating inflammation resolution and enhancing BV/TV at the affected area (Li et al 2024).
Stem cells from apical papilla (SCAP)–EVs activated the aggregation of forkhead box p3-positive regulatory T cells (Foxp3+) in inflamed pulp, reinstating odontoblast functionality and immune tolerance (Yu et al 2022). Foxp3+ refers to cells expressing Foxp3, a transcription factor that defines Tregs, which play a vital part in inhibiting inflammation and preserving immune tolerance; hence, heightened Foxp3+ cell aggregation suggests improved anti-inflammatory modulation in the inflamed pulp (Yu et al 2022). Moreover, let-7c-5p (a member of the let-7 family of miRNAs) preserved pulp by inhibiting dentin matrix protein-1 (DMP1)–NF-κB (Yuan et al 2018) and high mobility group AT-hook 2 (HMGA2)– phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) inflammatory axes, hence protecting odontogenic activity under inflammatory conditions.
Macrophages: Architects of Pulp Repair
Bone morphogenetic protein-2 (BMP2) and DSP are crucial modulators of reparative dentinogenesis via their complementary involvement in cell differentiation and ECM deposition. BMP2 is a strong odonto/osteogenic growth factor that stimulates mothers against decapentaplegic homolog (SMAD) signaling to activate DPSCs to differentiate into odontoblast-like cells, hence propagating dentin repair. DSP, a breakdown product of DSPP, is a primary noncollagenous protein of dentin involved in ECM organization and mineralization, facilitating the production and maturation of newly deposited dentin. DPSC-EVs polarized pulp macrophages toward a CD163+ M2 phenotype via EV miR-125a-3p–modulated targeting inhibitor of nuclear factor kappa-B kinase subunit beta (IKBKB), enhancing BMP2 release and inhibiting NF-κB/toll-like receptor (TLR) signaling (Zheng et al 2020). This formed a pro-odontogenic immune niche with increased DMP1/DSPP and reparative dentinogenesis.
EVs Regulate Pulp Revascularization
Pulp regeneration necessitates revascularization. M2-EVs produced vascular endothelial growth factor (VEGF)– and CD31+-enriched dense microvasculature in matrigel plugs, while M1-EVs suppressed vascularization (Wang et al 2024). Hypoxia pretreatment refers to culturing cells under a low-oxygen environment, inducing the HIF-1α pathway, and changing their EV cargo. In this state, lysyl oxidase-like protein-2 (LOXL2) is upregulated and abundant in EVs, where it contributes to angiogenesis by enhancing ECM remodeling, endothelial cell proliferation and migration, and vascularization; accordingly, LOXL2 knockdown undermines tube formation, facilitating its mechanistic involvement in the proangiogenic effects of hypoxia EVs. Hypoxia-pretreated DPSC-EVs demonstrated improved angiogenic signatures associated with LOXL2 overexpression in inflamed pulp (B. Li et al 2022). SHED-EVs transported miR-26a to stimulate the mothers against decapentaplegic homolog 2/3 (SMAD2/3)/transforming growth factor–beta (TGF-β); pathway (M. Wu et al 2021), mediating robust pulp angiogenesis and endothelial differentiation.
Integrated Vascular-Odontogenic Repair
Naïve EVs were outperformed by odontogenically differentiated (OD)-EVs (EVs from the cells pretreated under odontogenic differentiation conditions) (Huang et al 2016). OD-DPSC-EVs produced highly vascularized, structured pulplike tissue with erythrocyte-containing vessels and robust dentin phosphoprotein/DMP1 expression (Huang et al 2016). Extended-release EV microspheres formed consistent tertiary dentin characterized by tubule-like organization and collagen-abundant neodentin in rodent pulp-capping models (Swanson et al 2020a). Coapplication of OD-EVs and DPSCs produced the most comprehensive dentin-pulp complex, with NF200+ sensory nerve ingrowth, CD31+ vessels, and tubular neodentin (Y. Wang et al 2025a).
Cell Homing–Based Pulp Regeneration
Dentin–pulp tissue (DPT)-EVs allowed cell-homing regeneration, efficaciously filling entire root canals with innervated, vascularized pulplike tissue and propagating predentin production (Y. Chen et al 2022). Likewise, SCAP-EVs regulated pulp–dentin complex generation, featuring improved vascularization and polarized odontoblasts (Zhuang et al 2020). DPSC-EVs released during apoptosis surprisingly showed regenerative potential, inducing Tu translation elongation factor–based endothelial autophagy via TFEB stimulation and hence regulating revascularized pulp–dentin repair in dogs and rodents (Z. Li et al 2022).
Biomaterial-Based EV Treatments
Biomaterial-based EV delivery significantly enhanced therapeutic results. Embedding DPSC-EVs within an injectable hydroxypropyl-modified chitosan/chitin whiskers hydrogel extended their secretion and formed almost complete root canal filling, characterized by vascularized odontoblast-like lining and pulplike tissue (Wang et al 2023). SHED-EVs, administered via gelatin methacrylol (GelMA), improved mineralized dentin–pulp production (Lu et al 2024). Furthermore, EV and photobiomodulation treatment amplified odontogenic signaling and mineralization (Abdelgawad et al 2022).
Quantitative Modulation of Regenerative EVs
The hyperactivation of mechanistic target of rapamycin complex-1 (mTORC1) signaling diminished EV release by odontoblasts; reduced EV availability inhibited DPSC odontogenesis, demonstrating the quantity of EVs as a central modulator of dentin homeostasis (X. Luo et al 2023).
Craniomaxillofacial Regeneration
Across in vivo craniomaxillofacial models, EV treatment showed enhanced vascularized bone regeneration via incorporating (1) proangiogenic coupling, (2) immune modulation, and (3) osteogenic signaling, with cargo engineering and delivery systems emerging as vital efficacy enhancers.
Microsphere- and Hydrogel-Mediated Sustained Delivery
Microsphere- and hydrogel-based delivery consistently yielded effective defect bridging (Swanson et al 2020b). A CP05 (CD63-binding EV-anchoring peptide)–modified hydrogel model transporting EVs engineered to carry BMP2 and nonsense-mediated mRNA decay-associated protein significantly enhanced COL-1A1/OCN expression, trabecular number (Tb.N), BV/TV, and calvarial osteogenesis over 2 mo (Yang et al 2023).
EV mimetics are artificially produced nanovesicles designed to replicate the morphology and functionality of natural EVs, usually generated via chemical or physical approaches (cell extrusion) to attain controllable cargo and higher yield. Injectable hydrogels supplementing EV mimetics demonstrated dose-dependent near-complete critical defect repair (Xu et al 2025). EV mimetics from osteogenic MSCs with noggin suppression (ie, suppression of the BMP antagonist to promote BMP/SMAD signaling) mediated strong OCN positivity and lamellar bone formation (Fan et al 2020), whereas EV mimetics from standard osteogenic MSC hydrogels yielded superior BMD and coverage at high dosages (Xu et al 2025). Numerous “release-regulating” delivery systems, including EV-Yoda1–loaded hydrogels (He et al 2024), small intestinal submucosa–calcium alginate hydrogels functionalized with CP05 tethered EVs (Ma et al 2022), β-glycerophosphate/CS hydrogels for BMSC-EVs (Wu et al 2022), and polydopamine-modified porous microspheres containing hypoxia-preconditioned SHED-EVs (Gao et al 2022), improved vascularization, BV/TV, calcium formation, and collagen architecture, reflecting that extended release and residency enhance biocompatibility and osteogenesis.
EV Integration and Scaffold Functionalization
EV coating and scaffold functionalization augmented regeneration. The exosome-loaded poly(lactic-co-glycolic acid)/magnesium-gallium metal organic framework scaffolds enhanced BMD and BV/TV, together with increasing OCN, RUNX2, and alkaline phosphatase (ALP) and CD31/VEGF expression, suggesting synchronized angiogenesis–osteogenesis (Kang et al 2022). EV-integrated PCL/Ca-Si scaffolds enhanced osteogenesis, along with mTOR, AKT, and PI3K pathway stimulation, neovascularization, and connective tissue production (Yun et al 2024). Nanohydroxyapatite scaffolds incorporating EVs improved angiogenesis, collagen, and osteon production; osteoblast counts; and osteogenesis with increased osteopontin (OPN) (Youseflee et al 2023). Decellularized ECM-functionalized silk fibroin scaffolds incorporating osteoblast MSC-EVs resulted in ~80% defect repair versus ~6% controls and ~53% scaffold alone (Wang et al 2022). Nanofibrous PLA scaffolds embedded in osteogenic-stimulated EV-loaded microspheres substantially improved early ectopic mineralization and provided superior calvarial defect regeneration, featuring marrow-incorporating bone integrated within the host (Swanson et al 2020b).
The origin of the EVs greatly affects osteoinductive capability since its payload reflects the functional state and phenotype of the parent cells. Osteoblast-EVs are generally abundant in late-phase osteogenic factors that directly enhance ECM formation and mineralization, hence facilitating osteogenesis. Contrarily, BMSC-EVs carry an array of modulatory molecules that can both regulate and propagate osteogenesis, including cues that mediate SC recruitment, proliferation, and early differentiation, together with immunoregulatory factors that establish a proregenerative niche. The osteoinductive capability of the ceramic scaffold relied on EV origin. Mineralizing OD-EVs/osteoblast-derived exosomes–grafted tricalcium phosphate hydroxyapatite regulated mineralized tissue generation in ectopic and orthotopic aspects, while BMSC-EVs failed to stimulate ectopic bone production (Guerrero et al 2025). In mandibular defects, DPSC-EVs–loaded xenografts promoted trabecular structure, BMD, and osteogenesis (Gönen et al 2025).
Research converges on immune-angiogenic interactions as necessary for robust intramembranous ossification. Nuclear enriched abundant transcript-1 (NEAT1), a lncRNA abundant in human umbilical vein endothelial cell (HUVEC)–EVs, modulate macrophage shift, mainly enhancing the M2 phenotype, which promotes BMSC osteodifferentiation and migration. NEAT1 exerts these effects by regulating the DDX3X/ NOD-like receptor pyrin domain-containing-3 (NLRP3) cascade: NLRP3 is a central part of the inflammasome complex that mediates inflammatory responses and osteoclast-associated bone loss, and its stimulation expedites inflammation. GelMA/alginate hydrogel-supplementing HUVEC-EVs promoted RUNX2, OCN, and ALP expression; Tb.N; BV/TV; and angiogenesis. Knocking NEAT1 down diluted these results, whereas free EVs inhibited NLRP3 inflammasome functionality and enhanced M2 polarization (Chen et al 2023).
BMSC-EVs stimulated by Fe3O4 NPs and static magnetic fields demonstrated improved angiogenic and osteoinductive effects compared with standard BMSC-EVs owing to increased loading of miR-1260a, which enhanced angiogenesis (via COL4A2 inhibition) and osteogenesis (via HDAC7 suppression) (D. Wu et al 2021). Temporal secretion of DPSC-EVs and VEGF from gelatin+VEGF+calcium–magnesium ion hydrogels yielded almost complete defect repair with increased the CD31 region (Han et al 2023). Hydrogen-sulfide–preconditioned M2-EVs enhanced arginase-1 penetration and osteogenesis via enhanced β-catenin signaling and moesin internalization (Y.-k. Zhou et al 2024). Moreover, EVs from keratin-14+ and cathepsin K+ cells regulated vascular density (VEGFA and CD31), together with greater mineral density and BV (R. Zhou et al 2024). Furthermore, immune polarization emerged in adipose-derived mesenchymal stem cell (ADMSC)–EVs secreted from gelatin NP hydrogels, enhancing M1 to M2 polarization by macrophage migration inhibitory factor/miR-451a and permitting near-complete defect repair (measured as the percentage or area of the bone defect filled with newly formed tissue) at 2 mo (R. Li et al 2022).
Importantly, the efficacy of EV therapy is shaped by cargo quality and context. In an aged calvarial model, UCMSC-EVs reinstated osteogenesis by upregulating osterix (OSX), OPN, OCN, and ALP (H. Li et al 2025), whereas in diabetic conditions, normal-glucose BMSC-EVs enhanced regeneration, but high-glucose EVs compromised repair, both mediated by the SMAD7/miR-17 axis (Li et al 2023). Odonto/osteogenic pretreatment improved efficacy. 2,3,5,4′-Tetrahydroxystilbene-2-O-β-D-glucoside–activated DPSC-EVs enhanced OPN expression, proliferation, BV/TV, and angiogenesis in alveolar bone defects, while OD-DPSC-EVs–loaded alginate–oligomer scaffolds enhanced vessels, collagen, and bone, accompanied by shifts in pro-osteogenic miRNA signatures (S. Zhang et al 2024). Moreover, research suggests that systemic mechanobiology affecting regenerative pathways is associated with serum EVs sourced from distraction osteogenesis, promoting the migration and osteogenesis of jaw-derived BMSCs by calcium- and VEGF-associated signaling (Zhang et al 2022). Crucially, “cell-free dose effects” were revealed beyond cranial areas. Muscle cell type-4 (preosteoblastic mouse cell line)–EVs enhanced trabecular accrual predominantly via alleviating F4/80+ macrophages and tartrate-resistant acid phosphatase osteoclasts (Hakki et al 2024). SHED-EXOs delivered TFAM mRNA to target cells, facilitating mitochondrial oxidative phosphorylation and hence promoting osteodifferentiation and osteogenesis (Fig 3) (Guo et al 2022).

(
Temporomandibular Joint Disorders
Osteoarthritis of the temporomandibular joint (TMJ-OA) represents cartilage degradation and subchondral bone remodeling; nonetheless, chondrocyte-EVs show a dual role, contributing to both tissue regeneration and disease advancement (Q. Liu et al 2022). Strontium-preconditioned synovial membrane stem cell–derived EVs suppressed chondrocyte ferroptosis by ALG-2-interacting protein X–mediated miRNA loading, diminished pain, enhanced subchondral bone organization, and alleviated cartilage breakdown, partly via miR-143-3p targeting major facilitator superfamily domain-containing protein-8 (MFSD8) (Yuan et al 2025). DPSC-EVs, hydrogel administered or free, decreased subchondral bone destruction, protected joint organization, and alleviated synovial inflammation (Fig 4) (Diez-Guardia et al 2024). Likewise, MSC-EVs restored cartilage and subchondral bone integrity while attenuating fibrosis and inflammation through adenosine receptor–linked AMPK, ERK, and AKT pathways (Zhang et al 2019).

(
BMSC-EVs further enhanced regeneration by decreasing disc degeneration, mitigating apoptosis and inflammatory cytokines, and improving expression of COL2A1, aggrecan, and SRY-Box transcription factor 9 (SOX9), leading to increased cellularity and disc thickness (Chen et al 2025). Further reports showed protection of disc fibroblasts and inhibition of inflammatory responses after TMJ injury (Y. Yang et al 2024b). EVs secreted by SCs isolated from human urine preserved TMJ structure by suppressing osteoclastogenesis, downregulating matrix metalloproteinase–13, RUNX2, IL-1β, and IL-6, increasing cartilage proliferation and thickness, and reestablishing ECM homeostasis. This source was selected owing to its practicality (no surgical harvesting, low cost), high proliferative potential, and established paracrine effects (J.P. Zhou et al 2024).
Orthodontic Tooth Movement
EVs are central regulators of force-mediated periodontal remodeling, functioning as mechanotransduction signals that orchestrate osteogenesis on the tension side and, importantly, cementum/root preservation. Mechanical force–stimulated macrophage-EVs improved alveolar bone generation during orthodontic tooth movement, BV/TV, and elevated ALP-positive surfaces in tension/interradicular sites (Pu et al 2023).
Beyond remodeling, the integrity of cementum is dictated by the local immune niche, which plays a crucial role in cementum regeneration. M2-EVs were predominant in tension areas, promoting mineralization and facilitating cementum preservation and regeneration (Zhao et al 2022).
Furthermore, EV-modulated regulation contributes to regenerated periodontal tissue stability. PDLSC-EVs, especially simvastatin-containing EVs, and enhanced COL1A1, OSX, ALP, and RUNX2 expression, inhibited orthodontic relapse, forming denser alveolar bone and decreasing resorption, suggestive of improved structural integrity of regenerated tissues (X. Liu et al 2022). Moreover, MSC-EV–enriched medium promoted tissue remodeling while restricting radicular injury, indicating that precise EV signaling can coordinate bone remodeling with periodontal preservation during regeneration (Zwiri et al 2025).
Peri-implant Tissue Regeneration
Peri-implant regeneration, specifically under impaired conditions, including diabetes, involves not only osseointegration but also synchronized soft-tissue repair, immune balance, and neurovascular integration. SC-EVs under low-glucose conditions reinstated peri-implant osteogenesis in the type 2 diabetes mellitus (T2DM) rodent model, enhancing mineralization, trabecular indices, BV/TV, and BMD (Y. Wang et al 2025b), while also restoring peri-implant myelinated fibers and nerve density, underscoring an osteo-neurogenic aspect of regenerative osseointegration. These effects were recapitulated by miR-15b-5p agomir therapy, where miR-15b-5p—a modulatory miRNA—enhances osteogenesis by inhibiting TXNIP, a regulator of inflammation and oxidative stress, hence relieving TXNIP-regulated suppression of BMSC osteodifferentiation (Y. Wang et al 2025b).
TNF-α–pretreated ADMSC-EVs–coated titanium yielded superior peri-implant repair in T2DM mice, exhibiting elevated ALP expression, improved bone-to-bone contact, and favorable macrophage polarization toward CD206+ M2 phenotypes, along with diminished iNOS+ M1 macrophages (Y. Yang et al 2024a). Micro–computed tomography analysis confirmed improved trabecular structure (Y. Yang et al 2024a). Crucially, such EV-modulated immunoregulation also contributes to soft-tissue repair by restoring a balanced inflammatory niche that facilitates connective tissue integration and stable peri-implant sealing. Human gingival fibroblast–derived microvessel-functionalized titania nanopore implants showed this dual advantage by promoting bone-to-bone contact and soft-tissue attachment with mature collagen formation, facilitating concurrent soft- and hard-tissue integration (Fig 5) (Han et al 2026).

Histological analysis of bone–implant osseointegration and peri-implant soft-tissue generation. (
Root Resorption
Regarding resorption, root and periodontal structure regeneration encompasses tissue integrity preservation while enhancing repair via immunoregulation, osteoclast functionality, and angiogenesis. Mechanical force–induced DFSC-EVs substantially decreased osteoclast-modulated root resorption in rodents by suppressing inflammation and osteoclastogenesis (X. Li et al 2025). Mechanistically, EV-delivered miR-140-3p plays a vital role by targeting DNA (cytosine-5)-methyltransferase 1 (DNMT1), causing the suppression of the NF-κB pathway and upregulation of suppressor of cytokine signaling 1 (SOCS1). This pathway reprograms the M1 toward the M2 phenotype, inhibits NLRP3-modulated pyroptosis in PDLCs, and eventually suppresses osteoclast differentiation and functionality, hence preserving cementum and restricting root resorption (X. Li et al 2025). Hypoxia-pretreated SHED-EVs promoted vascularization, that is, elevated CD31+ structures, amplified VEGF, and increased microvasculature density (P. Liu et al 2022), further facilitating tissue repair in vascular-compromised periodontal niches.
Gaps and Outlook
Despite compelling evidence, EV treatment in regenerative dentistry continues to be hindered by immature manufacturing and context-based bioactivity. Notably, pioneering clinical reports by our research team in periodontitis patients have shown enhanced probing depth alleviation, clinical attachment gain, and radiographic bone fill after EV-based interventions (Appendix Table S1) (Estrin et al 2025; Froum et al 2026). A key gap is the field’s restricted capability to anticipate and govern whether EVs are regenerative or destructive. Future research should establish potency profiles (i.e., osteogenic/proangiogenic miRNA signatures, macrophage polarization ability), differentiating “repair-ready” from “disease-initiating” EV preparations, thereby informing origin/donor selection, pretreatment approaches (odonto/osteogenic induction, hypoxia), and engineering methods. Variable EV populations, inconsistent characterization/extraction protocols, and heterogeneous quantification limit translational reliability and comparability across investigations. A thorough implementation of Minimal Information for Studies of Extracellular Vesicles (MISEV)–guided protocols, together with orthogonal release criteria (such as functional assays, marker panels, and particle-to-protein ratios), is required to allow quality control and reproducible dosages. Biodistribution is origin and route based, and primary factors, including long-term safety, retention, and optimal dosages, particularly with repeated applications, are not fully standardized. Priority research comprises route-matched administration platforms, chronic toxicity investigations, and longitudinal tracking. Translation will mandate regulatory protocols specific to EV variability, stability-standardized preparations, and scalable Good Manufacturing Practices generation. The fastest clinical success probably lies in localized and biomaterial-incorporated EV delivery, where safety, dosage control, and retention margins are most effectively managed. While EV-based treatments are promising, they may pose immunological risks that depend on origin, dose, cargo, and route of administration. EVs can stimulate complement, activate antigen-presenting cells, and affect T-cell responses, potentially causing modified immunomodulation, alloimmunity, and inflammation. Moreover, poorly characterized or tumor-derived vesicles may deliver proinflammatory or immunosuppressive cues that could undermine safety. Hence, careful origin selection, rigorous purification, and optimized immunogenicity assessments are imperative before clinical application (Karasu et al 2018).
Conclusion
This review synthesized compelling in vivo and incipient clinical evidence establishing EVs as key mediators of orodental and craniofacial tissue regeneration. EVs serve as context-based biological regulators integrating odonto-osteogenesis, neurogenesis, angiogenesis, and immune regulation across an array of dental conditions. Their capability to mimic the paracrine activity of SCs while bypassing the intrinsic risks related to cell transplantation, renders them especially attractive for clinical applications. Crucially, recent first-in-human studies suggest that EV-mediated interventions can offer substantial clinical enhancements (Fig 6) (Estrin et al 2025; Froum et al 2026), highlighting their translational potential.

Clinical outcomes of gingival recession treatment using the minimally invasive vestibular incision technique with extracellular vesicle (EV)–solid platelet-rich fibrin (PRF). Panel 1: (
A primary unresolved issue lies in the challenge of reliably predicting, optimizing, and controlling EV bioactivity, as these nanovesicles can be either destructive or regenerative based on their cargo composition, surrounding microenvironment, and cellular source. Moreover, variability in the protocols of extraction, characterization, dosing, and delivery restricts reproducibility and regulatory preparedness. Future advancements will rely on the MISEV guidelines in manufacturing, the classification of potency, and the application of biomaterials for localized delivery. These advancements will be critical for scalable, safe, and precision-mediated clinical incorporation of EV interventions in regenerative dentistry.
Author Contributions
P. Ahmad, contributed to conception and design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript; N.E. Estrin, contributed to design, data acquisition and interpretation, drafted and critically revised the manuscript; R.J. Miron, contributed to design, data acquisition and analysis, drafted and critically revised the manuscript. All authors gave final approval and agreed to be accountable for all aspects of the work.
Supplemental Material
sj-docx-1-jdr-10.1177_00220345261448514 – Supplemental material for Extracellular Vesicles in Regenerative Dentistry
Supplemental material, sj-docx-1-jdr-10.1177_00220345261448514 for Extracellular Vesicles in Regenerative Dentistry by P. Ahmad, N.E. Estrin and R.J. Miron in Journal of Dental Research
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Data Availability Statement
No new data were created or analyzed in this study. Data sharing does not apply to this article.
A supplemental appendix to this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
