Extracellular Vesicles in Periodontitis: Bidirectional Crosstalk Between Periodontal Inflammation and Systemic Diseases

Abstract

Periodontitis is a chronic inflammatory disease that leads to the destruction of tooth supporting tissues, including gingival inflammation, periodontal attachment loss, and alveolar bone resorption. Increasing evidence suggests that periodontitis also increases susceptibility to a range of systemic conditions and there is a reciprocal communication between the periodontum and systemic organs. Extracellular vesicles (EVs), as key mediators of intercellular communication, can transport proteins, lipids, and nucleic acids across biological barriers and may play an important role in this crosstalk. In this review, we synthesize current evidence for the involvement of EVs in communication between periodontitis and distant organs. Compared with previous reviews that mainly focus on Porphyromonas gingivalis EVs or one-way effects, we synthesize evidence across multiple EV sources (bacterial, host periodontal, and circulating biofluid EVs) and explicitly review both directions of signaling. At present, the evidence is considerably stronger for periodontal-to-systemic EV effects, particularly in endothelial dysfunction, metabolic reprogramming, osteoclastogenic signaling, and barrier disruption in distant organs, whereas the evidence of systemic EVs on periodontitis remains limited, fragmented, and largely preclinical. This marked imbalance identifies the systemic-to-oral EV pathway as a major knowledge gap in the field. Further studies are needed to define the sources, cargoes, and functional significance of EVs in oral-systemic crosstalk and to clarify their translational potential as biomarkers and therapeutic targets.

Graphical Abstract

Introduction

Periodontitis is a highly prevalent oral infectious and inflammatory disease and remains a major global public health challenge with important consequences for both oral and general health. Epidemiological studies indicate that its burden has increased over the past decade, with moderate-to-severe periodontitis affecting approximately 53.2% of the population and severe periodontitis affecting 23.6%.^1–3 In addition to causing progressive destruction of tooth-supporting tissues, including gingival inflammation, periodontal attachment loss, and alveolar bone resorption, periodontitis has also been associated with a broad range of systemic disorders, including cardiovascular disease, diabetes mellitus, adverse pregnancy outcomes, osteoporosis, and neurodegenerative disease. Among these, the most consistent epidemiological support is seen for cardiovascular and metabolic diseases. Meta-analyses have reported modest but reproducible increases in risk, such as a 31% higher risk of incident cardiovascular disease (HR=1.31, 95% CI 1.13–1.48) and a 26% higher risk of incident type 2 diabetes mellitus (T2DM) (summary relative risk=1.26, 95% CI 1.12–1.41) among individuals with periodontitis.^4,5 In many studies, these associations are stronger in severe than in moderate disease, suggesting a possible dose-response relationship, whereas associations with pregnancy complications, bone disease, and neurodegeneration remain more heterogeneous and, in some cases, less well supported by prospective evidence.^6–11 Despite these epidemiological associations, the biological mechanisms linking periodontitis to systemic disorders remain incompletely understood.

Extracellular vesicles (EVs) have emerged as important mediators of intercellular communication in both physiological and pathological settings.^12,13 These nanoscale vesicles transport proteins, lipids, nucleic acids, and other bioactive molecules that can reshape recipient cell behavior and transmit signals across local and distant tissue barriers. In periodontitis, EVs are released by a wide range of sources, including periodontal resident cells, infiltrating immune cells, body fluids, and periodontal microorganisms. Together, these vesicles form a heterogeneous communication network with the potential to connect local periodontal inflammation to systemic pathology. Their biological effects are determined not only by the cellular or microbial origin of the vesicles, but also by their cargo composition, which is shaped during biogenesis and further modified by the unique physicochemical and biological features of the periodontal pocket such as abundant proteolytic activity, altered pH and oxidative stress, and continuous exposure to host-derived cytokines and tissue-breakdown products.^14,15 A broad range of virulence-associated molecules, inflammatory mediators, lipids, metabolites, and regulatory nucleic acids have been identified in periodontitis-associated EVs, supporting their potential to influence distant recipient tissues after systemic dissemination. Importantly, EV-mediated oral-systemic communication may converge on several shared downstream mechanisms rather than acting as a series of isolated organ-specific events. These mechanisms include inflammatory amplification, immune reprogramming, endothelial dysfunction, metabolic dysregulation, osteoclastogenic signaling, and oxidative stress-related tissue injury, often involving pathways such as TLR2/TLR4-MyD88 signaling and NF-κB activation.^16–19 Through circulation in blood, and possibly through neural routes, periodontitis-associated EVs may therefore provide a unifying mechanistic framework linking local periodontal inflammation to systemic diseases.

Although recent studies have substantially expanded the EV field in periodontology, existing reviews have focused mainly on P. gingivalis-derived EVs or one-way effects.^20–22 By comparison, we summarize the roles of periodontitis-associated EVs derived from periodontal pathogens, periodontal tissues, and body fluids in communication with the distant organs (Table 1, the effects of periodontal pathogen-derived EVs on systemic diseases are separately detailed in Table 2). Conversely, we review current evidence that EVs originating from extraoral tissues may influence periodontal inflammation under pathological conditions. Within this bidirectional framework, the strength of evidence is not uniform across organ systems. For some targets, including the cardiovascular system, hepatic metabolic regulation, osteoclast-related bone remodeling, and blood-brain barrier injury, direct mechanistic support from in vitro and animal studies suggests that periodontitis-associated EVs can reach distant tissues and alter defined cellular pathways. In contrast, for those involving the effects of extraoral EVs on periodontitis, the available evidence remains largely indirect, preclinical, or inferential. When evaluating the strength, reproducibility, and physiological relevance of EV-mediated oral-systemic communication, an important consideration is the substantial methodological heterogeneity that complicates interpretation of the EV literature. Isolation methods such as ultracentrifugation, size-exclusion chromatography, and precipitation differ in yield and purity and may co-isolate non-vesicular components. EV dosage is also inconsistently reported across studies, limiting direct comparison of biological effects, and incomplete adherence to MISEV recommendations further raises concerns about contamination and incomplete vesicle characterization. These issues are particularly relevant in periodontitis, where the unique periodontal microenvironment and the highly bioactive nature of bacterial EV-associated components may amplify inflammatory responses and complicate attribution of observed effects to EVs themselves.

Table 1.

Effects of Periodontal EVs on Systemic Disease

Target Organ/System	Associated Diseases	EV Sources	Key Cargoes	Major Mechanisms	Main Pathological Effects
Cardiovascular system	Endothelial dysfunction, thrombosis, atherosclerosis, vascular calcification	Pgingivalis EVs; EVs released from LPS-stimulated endothelial cells; Circulating EVs from periodontitis patients	PPAD, gingipains, LPS, miR-155-5p	Activation of ERK1/2 and p38 MAPK signaling pathway;23 PPAD/CitH3/B3GAT1 pathway24 VE-cadherin and PECAM-1 degradation;16,25 Increase stress fibre formation;25 Suppression of DET1;26 Upregulation of Runx227	Promote vascular endothelial glycocalyx injury, impair endothelial repair and angiogenesis, increase vascular permeability, and promote platelet aggregation, foam cell formation, and osteogenic transition of VSMCs.
Skeletal system	Bone loss, osteoporosis	EVs from F. alocis, P. gingivalis, T. forsythia, and S. oralis; EVs from P. gingivalis-stimulated dendritic cells	Lipoproteins, LPS	Upregulation of NFATc1 and c-Fos;28,29 Induce ferroptosis via the Hippo-YAP pathway;30 Activation of SAA3/TLR4/MyD88/NF-κB signaling;31 Inhibiting Cyp4f40-mediated acetylcholine production32	Promote osteoclast differentiation, suppress osteogenic differentiation and enhance bone resorption.
Liver	NAFLD	EVs derived from macrophages or PDLFs stimulated by P. gingivalis-LPS; F.alocis EVs, P.gingivalis EVs,	PPAD	Upregulation of SREBP1, SCD-1, FASN and CD36;33 Inhibition of AMPK signaling pathway;33 Induce PAI-1 expression via TLR2-PI3K-AKT-NF-κB signaling pathway;34 Inhibit the expression of CYP4A10;35 Activate hepatic stellate cells through the THBS1/TGF-β pathway36	Promote triglyceride accumulation and lipid metabolic dysregulation, aggravating hepatic steatosis and liver injury.
Metabolic system	T2DM, hepatic insulin resistance	P. gingivalis EVs; Circulating EVs from periodontitis patients/rats	Gingipains	Inhibition of AKT-GSK-3β signaling pathway37,38	Exacerbate hepatic insulin resistance and worsen glycemic control.
Retina	Diabetic retinopathy	P. gingivalis EVs	/	Increased ICAM-1, MMP-9, ROS and mitochondrial dysfunction39	Disrupt the blood-retinal barrier, increase retinal permeability, and promote retinal inflammation and injury.
Renal tissue	Nephropathy	P. gingivalis EVs	/	Upregulate EMT- and fibrosis-related genes40	Exacerbate tubulointerstitial injury and interstitial fibrosis.
Adipose tissue	Adipose tissue dysfunction, insulin resistance	Gingival EVs from oral infection mice	miRNAs	Reduced AKT phosphorylation, dysregulation of adipokine and leptin41	Decrease adiponectin and leptin, suppress adipogenesis and lipid metabolism.
Brain	Neurodegenerative disorders	Pgingivalis EVs; A. actinomycetemcomitans EVs; Gingival EVs from periodontitis patients	Gingipains, LPS, msRNAs	Activation of NF-κB signaling pathway;42 Cleavage of caspase-3;43 Promote MAM formation44	Activate microglia, promote Aβ deposition and tau phosphorylation, disrupt the blood–brain barrier, and induce TG neuronal activation and neuroinflammation.
Placenta/pregnancy system	Preeclampsia, fetal growth restriction, miscarriage, preterm birth	P. gingivalis EVs	/	Increase sFlt-1/PlGF ratio, MCP-1 and decrease VEGF;45 Reduce GLUT1, glycolytic inhibition46	Inhibit trophoblast migration and invasion, impair angiogenesis and immune homeostasis, and cause placental structural alteration and APOs.

Table 2.

Effect of periodontal pathogen-derived EVs on systemic diseases

Periodontal pathogen	EV cargoes	Major mechanisms	Main pathological effects	Systemic disease
P. gingivalis	Gingipains, PPAD, LPS	ERK1/2-p38 MAPK activation;23 PPAD/CitH3/B3GAT1;24 VE-cadherin/PECAM-1 degradation;16,25 RUNX2 upregulation27	Endothelial dysfunction; vascular permeability; foam cell formation; Calcification	Cardiovascular disease
	Lipoproteins, LPS	TLR2 activation;29 Hippo-YAP ferroptosis;30 SAA3/TLR4/MyD88/NF-κB;31 Cyp4f40 inhibition32	Osteoclast differentiation; impaired osteogenesis	Systemic bone loss
	PPAD	SREBP1/SCD-1/FASN upregulation;33 AMPK inhibition;33 PAI-1 induction;34 CYP4A10 inhibition;35 THBS1/TGF-β activation36	Hepatic steatosis; lipid dysregulation	NAFLD
	Gingipains	AKT-GSK-3β inhibition37	Hepatic insulin resistance	T2DM
	/	ICAM-1/MMP-9 upregulation; mitochondrial dysfunction39	Blood-retinal barrier disruption; retinal injury	Diabetic retinopathy
	/	EMT/fibrosis-related gene upregulation40	Renal fibrosis, tubulointerstitial injury	Diabetic nephropathy
	Gingipains, LPS	NF-κB activation;42 BDNF/NMDAR reduction;44 TGM2-MAM dysregulation47	BBB disruption, microglial activation, neuroinflammation, cognitive impairment	Neurodegenerative disorders
	/	Increased sFlt-1/PlGF and MCP-1, decreased VEGF;45 glycolysis inhibition46	Placental dysfunction, impaired trophoblast invasion, fetal growth restriction	Adverse pregnancy outcomes
A. actinomycetemcomitans	msRNAs	NF-κB activation;42 increased TNF-α and IL-648	BBB disruption, neuroinflammation	Alzheimer’s disease-related mechanisms
F. alocis	Lipoproteins	TLR2/NFATc1/c-Fos activation28	Osteoclast differentiation;	Systemic bone loss
	/	PAI-1 induction via TLR2-PI3K-AKT-NF-κB34	Hepatic steatosis	NAFLD
T. forsythia	Lipoproteins, LPS	TLR2 activation29	Osteoclast differentiation;	Osteoporosis/ systemic bone loss

EV Biogenesis

EVs are broadly divided into exosomes, microvesicles, and apoptotic bodies depending on their biogenetic origin. Exosomes, typically 40–160 nm in diameter, arise from the endosomal pathway and are released upon fusion of multivesicular bodies (MVBs) with the plasma membrane.⁴⁹ Biogenesis initiates with inward budding of the plasma membrane, giving rise to early sorting endosomes that mature into late endosomes, and ultimately MVBs containing intraluminal vesicles (ILVs). The formation of MVBs is largely governed by the endosomal sorting complex required for transport (ESCRT), which comprises four complexes, designated ESCRT-0 through ESCRT-III. ESCRT-0, -I, and -II cooperate to assemble a high-affinity platform for the recognition and recruitment of ubiquitinated cargo proteins, whereas ESCRT-III drives membrane budding and scission to facilitate ILV formation.^50,51 Upon the fusion of MVBs with the plasma membrane, ILVs are released into the extracellular milieu as exosomes. In addition to the ESCRT-dependent mechanism, exosome biogenesis can also proceed via ESCRT-independent pathways, including the phosphatidic acid pathway catalyzed by phospholipase D2 and lipid raft-dependent mechanisms mediated by Rab31 and flotillin.^52–54

Microvesicles originate from the remodeling of proteins and lipids in the plasma membrane, which leads to outward budding and subsequent membrane fission. These vesicles typically range from 50 to 1,000 nm in diameter and are released directly from the cell surface.⁵⁵ By contrast, apoptotic bodies are membrane-bound vesicles generated during apoptotic cell fragmentation and generally range from 1 to 5 μm in size.⁵⁶ Definitive biomarkers for the accurate identification and classification of EV subtypes are still lacking, and current approaches for isolating specific vesicles according to their biogenetic origin remain limited. Accordingly, the term “EVs” is used as a collective term.⁵⁷ EVs participate in a broad array of biological processes. EVs can function as antigen-presenting vehicles in immune regulation, enhance cell proliferation and migration, and promote neural stem cell differentiation and neurogenesis.^58–61 Owing to their stability, biocompatibility, and ability to traverse biological barriers, EVs are ideally suited to mediate intercellular communication in both physiological and pathological contexts.

Types and Functions of Periodontitis-Related EVs

Periodontitis profoundly reshapes the biogenesis, cargo, and functional output of EVs released by multiple cell types within the periodontal niche, which in turn reprograms recipient cell behavior and propagates local pathological signaling. For instance, EVs derived from lipopolysaccharide (LPS)-stimulated periodontal ligament fibroblasts (PDLFs) impair osteogenic differentiation of MG-63 cells by shifting the molecular balance from osteogenic markers, including alkaline phosphatase (ALP), runt-related transcription factor 2 (RUNX2), and osteoprotegerin (OPG), toward proinflammatory mediators such as interleukin 6 (IL-6) and tumor necrosis factor alpha (TNF-α).⁶² Similarly, EVs released by inflamed periodontal ligament stem cells (PDLSCs) promote macrophage polarization and amplify inflammatory responses.⁶³ Together, these observations identify EV remodeling as a central mechanism through which periodontitis disrupts periodontal homeostasis.

Disease progression is also accompanied by quantitative and qualitative alterations in the humoral EV profiles. The abundance of EVs in gingival crevicular fluid increases with the severity of periodontitis, while the molecular landscape of salivary EVs undergoes substantial remodeling.^64–66 Salivary EVs from patients with periodontitis exhibit a distinct transcriptomic signature characterized by increased TNF-α and reduced osterix expression, with high area under the curve (AUC) values supporting their potential as diagnostic biomarkers.⁶⁷ In parallel, programmed death-ligand 1 (PD-L1) mRNA is markedly enriched in salivary EVs from periodontitis patients.⁶⁸ Other salivary EV cargo molecules, including hsa-miR-140-5p, hsa-miR-146a-5p, hsa-miR-628-5p, and hsa-miR-125a-3p, are likewise significantly elevated in the context of periodontitis. In contrast, several miRNAs, including hsa-let-7d-5p, hsa-miR-103a-3p, hsa-miR-126-3p, hsa-miR-150-5p, hsa-miR-199a-3p, hsa-miR-4485-5p, hsa-miR-6088, and hsa-miR-6821-5p, were significantly downregulated in plasma-derived EVs from the affected individuals.⁶⁹

Among periodontitis-related vesicle populations, EVs from periodontal pathogens have received particular attention. Bacterial EVs are generated by outward blebbing of the membrane and typically contain concentrated virulence-associated cargoes, including LPS, lipoproteins, proteases, adhesins, and small RNAs, the latter of which have recently emerged as previously underappreciated regulators of host-pathogen interactions.^70–72 These EVs are especially relevant to major periodontal pathogens such as Porphyromonas gingivalis (P. gingivalis), Tannerella forsythia (T. forsythia), and Aggregatibacter actinomycetemcomitans (A. actinomycetemcomitans). Rather than being passive by-products of bacterial growth, EVs are increasingly recognized as specialized delivery systems that can modulate host immunity, barrier function, and intercellular signaling.^42,73,74 This has led to the view that bacterial EVs may help explain how periodontal pathogens exert systemic effects even when intact bacteria are not detectable at distal sites. Indeed, bacterial DNA and virulence-associated molecules derived from P. gingivalis have been detected in the liver, synovium, blood, and brain, frequently in the absence of cultivable bacteria.^75–77 Given their enrichment in pathogenic cargo and high efficiency in interacting with host cells, bacterial EVs provide a plausible vehicle for disseminating these inflammatory signals to distant tissues.

Importantly, bacterial EVs should not be viewed merely as secreted bacterial fragments or vehicles but as highly adapted virulence platforms that couple immune evasion with inflammatory amplification. P. gingivalis exemplifies this functional dichotomy. At the bacterial surface, nonphosphorylated tetra-acylated lipid A (NPLA) attenuates toll-like receptor 4 (TLR4) recognition, limits inflammasome activation, and increases resistance to cationic antimicrobial peptides, thereby favoring immune escape and persistence.^78–80 By contrast, EVs from P. gingivalis are enriched in C4’ monophosphoryl lipid A (C4’-MPLA), a potent agonist of the TLR4-TRIF-IFNβ axis, and are therefore equipped to induce robust TRIF-dependent IFNβ production and inflammasome-mediated interleukin-1β (IL-1β) release.⁸⁰ This partitioning of immune-silent and immune-stimulatory lipid A species suggests a strategic division of labor: reduced immune recognition at the primary site of infection coupled with efficient propagation of proinflammatory signals to distal tissues. In this framework, EVs are not ancillary to P. gingivalis pathogenicity but are central effectors that extend bacterial influence beyond the periodontal niche while amplifying inflammatory injury.

This vesicle-centered paradigm may be further expanded by evidence that P. gingivalis can hijack host EV pathways following macrophage invasion, enabling bacterial virulence factors to be repackaged into macrophage-derived EVs and redistributed systemically.^81,82 Such host-derived EVs may potentially extend the range and immune invisibility of pathogen-associated cargo. Although this model is attractive and supported by experimental studies, its prevalence and in vivo significance remain unclear. Overall, the current literature supports the view that periodontitis-related EVs are heterogeneous carriers of inflammatory and regulatory information, but the relative contributions of periodontitis-related EVs to systemic diseases remain unclear.

EV-Mediated Bidirectional Communication Between Periodontitis and Systemic Disease

Periodontal EVs and Cardiovascular Diseases

Cardiovascular disease is among the most intensively studied systemic conditions linked to periodontitis, and EVs have emerged as candidate mediators of this relationship.^83–85 Available data suggesting that periodontal EVs, particularly P. gingivalis EVs, may influence several processes relevant to vascular pathology, including endothelial dysfunction, barrier disruption, thromboinflammation, atherosclerosis, and vascular calcification.

In endothelial cells, P. gingivalis EVs impair reparative responses such as cell proliferation and capillary-like structure formation (Figure 1a).⁸⁶ They also suppressed endothelial nitric oxide synthase (eNOS) expression through Rho-associated kinase (ROCK)-dependent signaling, with downstream effects on ERK1/2 and p38 MAPK pathways (Figure 1a).²³ Peptidylarginine deiminase (PPAD) present in P. gingivalis EVs contributes to vascular endothelial glycocalyx injury by inducing histone H3 citrullination and suppressing β-1,3-glucuronosyltransferase (B3GAT1) expression (Figure 1a).²⁴ These findings are consistent with a model in which bacterial vesicles reduce endothelial resilience and promote a dysfunctional vascular phenotype.⁸⁷ In parallel, P. gingivalis EVs can weaken endothelial barrier integrity through multiple mechanisms, including degradation of platelet endothelial cell adhesion molecule-1 (PECAM-1), increasing stress fiber formation at intercellular junctions, and promoting the internalization and lysosomal degradation of VE-cadherin.^16,25 These convergent effects on cytoskeletal remodeling and junctional protein turnover establish a mechanistic basis for EV-induced vascular leakage. Such changes are accompanied by inflammatory activation and may create a prothrombotic microenvironment by exposing the subendothelial collagen and enhancing platelet reactivity. Consistently, P. gingivalis EVs have been shown to directly induce platelet aggregation, supporting their contribution to thromboembolic pathology (Figure 1b).^88,89

Figure 1.

Pathogenic role of periodontal EVs and associated mechanisms in cardiovascular disease. (a) P. gingivalis EVs impair vascular endothelial cell function by inhibiting cell proliferation, inducing oxidative stress and endothelial glycocalyx injury. (b) P. gingivalis EVs disrupt endothelial barrier through degrading adhesion molecule, increasing stress fiber formation and promote platelet aggregation, thereby contributing to thrombosis. (c) Plasma EVs of periodontitis patients as well as LPS-stimulated HUVEC-derived EVs, cause foam cell formation and atherosclerosis by translocating miR-155-5p.

Abbreviations: EVs, extracellular vesicles; P. gingivalis, Porphyromonas gingivalis; PPAD, peptidylarginine deiminase; CitH3, citrullinated histone H3; B3GAT1, β-1,3-glucuronosyltransferase; ROCK, rho-associated kinase; eNOS, endothelial nitric oxide synthase; PECAM-1, platelet endothelial cell adhesion molecule-1; HUVECs, human umbilical vein endothelial cells; HAECs, human aortic endothelial cells; VSMCs, vascular smooth muscle cells.

Periodontal EVs also appear to participate in atherosclerosis progression through both host-derived and bacteria-associated mechanisms. MiR-155-5p is elevated in plasma EVs of patients with severe periodontitis, particularly in those with concomitant carotid atherosclerosis, and the experimental delivery of miR-155-5p-enriched EVs accelerates atherosclerotic plaque formation and carotid hyperplasia in mice. This process may involve suppression of the target gene DET1, leading to enhanced foam cell formation and lipid accumulation.²⁶ In addition, LPS from periodontal pathogens induces human umbilical vein endothelial cells (HUVECs) to release EVs enriched in miR-155-5p that promote angiogenic activity and increase vascular permeability after transfer to human aortic endothelial cells (HAECs), consequently contributing to plaque progression.⁹⁰ Plasma EVs may further contribute to this process by delivering P. gingivalis LPS to distant tissue-resident macrophages, where they synergize with local inflammatory cues to promote foam cell formation and pericarotid sclerosis.²⁶ Beyond atherogenesis, P. gingivalis EVs also favor vascular calcification by activating ERK signaling in vascular smooth muscle cells (VSMCs), which in turn upregulates the osteogenic regulator RUNX2 and drives VSMC mineralization (Figure 1c).²⁷

Collectively, these findings suggest that periodontal EVs are active mediators of cardiovascular pathology, rather than passive correlates of periodontal inflammation. By impairing endothelial repair, disrupting barrier function, enhancing thromboinflammatory responses, promoting atherosclerosis, and driving vascular calcification, P. gingivalis and host-derived pathogenic EVs provide a mechanistic framework linking periodontitis to cardiovascular disease. However, it should be noted many findings derive from in vitro exposure systems or exogenous vesicle administration models that may not accurately reflect physiological EV concentrations, distribution, or vascular access in vivo.

Periodontal EVs and Systemic Bone Loss

Periodontitis is associated not only with local alveolar bone destruction but also with reduced bone mass at extraoral skeletal sites. It has been reported that periodontitis is associated with an approximately two-fold increased risk of osteoporosis.^91–94 Shared risk factors such as aging, obesity, diabetes, menopause, chronic inflammation, and inadequate vitamin D or calcium intake may partly explain this relationship, yet these factors do not fully address how an oral inflammatory lesion might influence systemic bone remodeling.^95–97 Bacterial EVs have therefore been proposed as one possible mechanism. Osteoclast precursors express TLRs, enabling them to sense pathogen-associated and microbe-associated molecular patterns.¹⁷ Bacterial EVs are well positioned to disseminate inflammatory and osteoclastogenic signals beyond the oral cavity and directly reprogram osteoclast precursors through TLR-dependent pathways.

Among oral pathogens, Filifactor alocis (F. alocis) has emerged as a keystone species linked to chronic and aggressive periodontitis, peri-implant inflammation, and endodontic infection.⁹⁸ EVs from F. alocis accumulated in the long bones of mice after intraperitoneal injection and promoted osteoclastogenesis and bone resorption. Mechanistically, this response depended on TLR2 and was accompanied by increased expression of nuclear factor activated T cells c1 (NFATc1) and c-Fos. This property extends beyond F. alocis. EVs from P. gingivalis, T. forsythia, and Streptococcus oralis (S. oralis) similarly activate osteoclast precursors via TLR2 and induce a broad pro-osteoclastogenic cytokine programme.²⁹ These effects depend on the various amphiphilic molecules derived from different bacterial origins. In EVs from the Gram-negative species P. gingivalis and T. forsythia, both lipoprotein lipase and polymyxin B attenuate osteoclastogenesis, consistent with the contributions from both lipoproteins and LPS. In contrast, the effects of F. alocis and S. oralis EVs were sensitive only to lipoprotein lipase, indicating a dominant role for lipoproteins in EVs from Gram-positive bacteria (Figure 2a).^28,29

Figure 2.

Pathogenic roles of periodontal EVs and associated mechanisms in systemic bone loss, NAFLD, T2DM, and adipose tissue dysfunction. (a) Periodontal pathogen-derived EVs activate TLR2 on osteoclast precursors and promote osteoclast differentiation; P. gingivalis EVs also induce ferroptosis and inflammation in BMSCs via the Hippo-YAP and SAA3/TLR4/MyD88/NF-κB pathways, inhibiting osteogenic differentiation. In addition, P. gingivalis EVs induce osteoclast differentiation by suppressing Cyp4f40-mediated acetylcholine production. (b) EVs from macrophages and PDLFs stimulated with P. gingivalis LPS upregulate lipogenesis-related genes; P. gingivalis and F. alocis EVs increase PAI-1 expression, whereas P. gingivalis EVs downregulate CYP4A10 and activate hepatic stellate cells via the THBS1/TGF-β pathway, promoting hepatic inflammation and lipid accumulation and contributing to NAFLD. (c) P. gingivalis EVs and plasma EVs from periodontitis patients aggravate hepatic insulin resistance and contribute to T2DM. (d) P. gingivalis EVs disrupt the blood-retinal barrier and induce retinal inflammation, oxidative stress, and apoptosis, contributing to diabetic retinopathy; they also upregulate fibrosis- and EMT-related genes, exacerbating renal fibrosis and tubulointerstitial injury. (e) Gingival EVs from orally infected periodontitis mice induce adipose tissue dysfunction by downregulating adipogenic genes.

Abbreviations: EVs, extracellular vesicles; P. gingivalis, Porphyromonas gingivalis; T. forsythia, Tannerella forsythia; F. alocis, Filifactor alocis; S. oralis, Streptococcus oralis; TLR2, toll-like receptor 2; NFATc1, nuclear factor‑activated T cells c1; BMSCs, bone marrow mesenchymal stem cells; ACSL4, acyl‑CoA synthetase long-chain family member 4; TFRC, transferrin receptor; PTGS2, prostaglandin-endoperoxide synthase 2; SAA3, serum amyloid A3; Cyp4f40, cytochrome P450, family 4, subfamily f, polypeptide 40; AKT, protein kinase B; GSK-3β, glycogen synthase kinase 3β; NAFLD, nonalcoholic fatty liver disease; SREBP1, sterol regulatory element-binding protein 1; SCD-1, stearoyl‑CoA desaturase 1; FASN, fatty acid synthase; PPAD, peptidylarginine deiminase; PAI-1, plasminogen activator inhibitor-1; CYP4A10, cytochrome P450, family 4, subfamily A, polypeptide 10; NPAS2, neuronal PAS family member 2; THBS1, thrombospondin 1; TGF-β, transforming growth factor β; ICAM-1, intercellular adhesion molecule 1; MMP-9, matrix metalloproteinase 9; ROS, reactive oxygen species; EMT, epithelial-to-mesenchymal transition; TGF-β, transforming growth factor β. The red upward/downward arrow indicates increased/decreased levels, respectively.

In addition, P. gingivalis EVs promote bone resorption and impair bone regeneration via multiple mechanisms. They induce ferroptosis in bone marrow mesenchymal stem cells (BMSCs) via the Hippo-YAP pathway, in which nuclear translocation of YAP drives the expression of ferroptosis-associated genes such as ACSL4, TFRC, PTGS2, and inhibiting bone regeneration.³⁰ Moreover, P. gingivalis EVs suppresses the osteogenic differentiation of BMSCs through the SAA3/TLR4/MyD88/NF-κB axis.³¹ In vivo, P. gingivalis-derived EVs accumulate in long bones, where they inhibit Cyp4f40-mediated acetylcholine production, thereby relieving acetylcholine-dependent suppression of osteoclastogenesis.³² These findings support the concept that bacterial EVs function as mobile osteoimmunological cues that impair bone regeneration (Figure 2a).

Host-derived EVs can amplify these effects. EVs released from P. gingivalis-stimulated dendritic cells have been reported to induce senescence-associated secretory phenotype (SASP) and worsen bone loss in recipient systems, suggesting that secondary host EV responses may extend the skeletal consequences of periodontal infection.⁹⁹ Altogether, these observations posit oral bacterial EVs as active disseminators of bone-destructive signals by transporting microbial TLR ligands to distant bone. They offer a mechanistic framework linking periodontal infection to systemic bone loss and identify EV-associated lipoproteins and/or LPS as candidate therapeutic targets in inflammation-driven skeletal disease. Even so, several caveats remain, as many studies rely on nonphysiological exposure routes, such as intraperitoneal injection of isolated bacterial EVs, and the extent to which naturally circulating oral EVs accumulate in the bone at physiologically relevant levels remains unclear.

Periodontal EVs and Nonalcoholic Fatty Liver Disease (NAFLD)

Mounting epidemiological and experimental evidence points to a meaningful association between periodontitis and metabolic liver disease, particularly NAFLD.^100,101 In cohort studies, the incidence of NAFLD increased in a clear severity-dependent manner among individuals with periodontitis, with incidence rate ratios of 1.11 (95% CI 0.56–2.21; p=0.7) for moderate periodontitis and 1.72 (95% CI 0.92–3.21; p=0.09) for severe periodontitis relative to periodontally healthy controls.¹⁰² Notably, this relationship appears to be at least partly reversible, as nonsurgical periodontal therapy has been shown to improve hepatic biochemical parameters in patients with concomitant periodontitis and NAFLD, resulting in reduced serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels.¹⁰³ Experimental data further reinforce this link, with ligature-induced periodontitis in rats causing clear elevations in circulating AST and ALT.¹⁰⁴

In this context, EVs are emerging as central effectors of this oral-liver crosstalk. EVs derived from macrophages and PDLFs stimulated with P. gingivalis LPS are taken up by hepatocytes and induce the expression of lipogenesis-associated genes including sterol regulatory element-binding protein 1 (SREBP1), stearoyl-CoA desaturase 1 (SCD-1), and fatty acid synthase (FASN). The resulting metabolic reprogramming promotes triglyceride accumulation and accelerates the progression of NAFLD. Intravenous delivery of these EVs increases hepatic SCD-1 expression, suppresses AMPK signaling, and aggravates steatosis in vivo.³³ Plasminogen activator inhibitor-1 (PAI-1) has been linked to NAFLD through its role in hepatic lipid accumulation. EVs from periodontal pathogens, including F. alocis and P. gingivalis, induce PAI-1 expression via TLR2-PI3K-AKT-NF-κB signaling pathway in hepatocytes, promoting steatosis.³⁴ In addition, P. gingivalis EVs deliver PPAD, which can translocate into the nucleus and interact with neuronal PAS family member 2 (NPAS2), downregulating the downstream fatty acid metabolism gene CYP4A10 and exacerbating metabolic dysfunction-associated fatty liver disease.³⁵ P. gingivalis EVs have also been reported to activate hepatic stellate cells through the THBS1/TGF-β pathway, further implicating periodontal EVs in liver metabolic pathology (Figure 2b).³⁶ These findings identified periodontitis-associated EVs as plausible systemic drivers of hepatic lipid dysregulation and liver disease progression. Nevertheless, current evidence is based largely on preclinical models and selected inflammatory stimuli rather than the full complexity of human periodontitis. More work is needed to determine whether circulating periodontal EVs materially contribute to liver disease progression in physiological administration models.

Periodontal EVs and Type 2 Diabetes Mellitus (T2DM)

The relationship between periodontitis and T2DM is widely considered bidirectional, with diabetes increasing periodontal susceptibility and periodontitis being associated with poorer glycemic control.^105–107 Compared with diabetic patients without periodontitis, those with concomitant periodontal disease exhibit higher glycated hemoglobin A1c (HbA1c) and homeostasis model assessment of insulin resistance (HOMA-IR) values and often require more intensive glucose-lowering therapy.^108,109 These observations position periodontitis not merely as a comorbidity, but as a clinically relevant modifier of metabolic control in diabetes.

Periodontal EVs have emerged as plausible mediators of this effect. Systemic dissemination of periodontal bacteria and their vesicles is likely facilitated by diabetes-associated endothelial dysfunction and microvascular impairment, which together increase vascular permeability and compromise barrier integrity. Among the best-characterized examples are P. gingivalis EVs, which deliver gingipains to the liver and disrupt insulin signaling pathways, thereby suppressing hepatic glycogen synthesis.³⁷ Consistent with this, circulating EVs derived from periodontitis in both experimental models and patients impair glucose handling, attenuate insulin signaling, and reduce glycogen synthesis in hepatocytes, with associated decreases in the phosphorylation of protein kinase B (AKT) and glycogen synthase kinase 3β (GSK-3β).³⁸ These findings support the view that inflammatory EVs generated in periodontitis contribute directly to systemic insulin resistance and metabolic dysregulation (Figure 2c).

In addition to its detrimental impact on glycemic control, periodontitis may also contribute to the pathogenesis of diabetic microvascular complications, particularly diabetic retinopathy (DR).¹¹⁰ An expanding body of epidemiological evidence has increasingly implicated periodontitis in retinal injury and other microvascular abnormalities associated with diabetes, independent of classical risk factors. This association is particularly compelling given that elevated HbA1c accounts for only 6.6% of DR risk, highlighting the existence of additional pathogenic determinants.¹¹¹ Consistent with this view, periodontal indices have been correlated with early retinal vascular alterations, while experimental studies have further demonstrated that P. gingivalis EVs exacerbate blood-retinal barrier breakdown in diabetic mice, increasing retinal permeability and inducing pathological retinal alterations characterized by retinal pigment epithelial exudation, diminished microvascular calibre, and an increased number of fundus microaneurysms. At the cellular level, these vesicles drive inflammatory activation of human retinal microvascular endothelial cells, upregulating adhesion molecules and matrix-remodeling factors, such as intercellular adhesion molecule 1 (ICAM-1) and matrix metalloproteinase 9 (MMP-9), while also inducing oxidative stress, mitochondrial dysfunction, and apoptosis (Figure 2d).³⁹ Furthermore, P. gingivalis EVs can enter the renal tissue via circulation and exacerbate renal dysfunction, tubulointerstitial injury, and fibrosis in mice. These EVs upregulate fibrosis- and EMT-related genes, including TGF-β1, Acta2, vimentin, collagen I, β-catenin, and Snail1, inducing renal fibrosis and tubular epithelial-to-mesenchymal transition (EMT), potentially contributing to diabetic complications, such as diabetic nephropathy (Figure 2d).⁴⁰ Together, these observations suggest that periodontal EVs not only worsen hepatic insulin resistance, but also directly target vulnerable vascular beds and renal tissue to accelerate diabetic end-organ damage. However, this interpretation remains provisional, as much of the available mechanistic evidence is derived from non-physiological models. Although highly informative, these systems may not fully reproduce the endogenous abundance, cargo composition, biodistribution, and tissue exposure of periodontal EVs in patients with naturally occurring periodontitis and diabetes.

Periodontal EVs and Adipose Tissue Dysfunction

Adipose tissue is increasingly recognized as an active participant in systemic inflammation and insulin resistance, making it a plausible target of periodontal EV signaling.^112,113 In experimental models, oral inoculation with P. gingivalis and Fusobacterium nucleatum induces adipose tissue abnormalities characterized by impaired AKT phosphorylation, altered adipokine expression, and dysregulation of adipogenic genes.^114,115 Gingival EVs isolated from orally infected periodontitis mice can reproduce phenotypes of adipose dysfunction when transferred to healthy animals, suggesting that vesicle-mediated communication may contribute to oral-adipose crosstalk.⁴¹ The miRNA cargo of these EVs is enriched for regulators of insulin signaling and adipose homeostasis, further supporting a role for gingival vesicle miRNAs in inter-organ crosstalk and metabolic impairment (Figure 2e). However, these findings are based primarily on non-physiological inoculation and EV transfer models, which may not accurately reflect endogenous EV production, distribution, and target engagement in naturally occurring periodontitis.

Periodontal EVs and Neurodegenerative Disorders

Alzheimer’s disease and related dementias (ADRD) affect nearly 50 million people worldwide, yet their initiating drivers remain incompletely defined, and effective disease-modifying therapies are still lacking.^116–119 This has attracted much attention on modifiable exposures that may shape disease risk long before clinical onset. Among these, periodontitis has emerged as a credible contributor, particularly in aging populations.^120,121

ADRD is characterized by extracellular β-amyloid (Aβ) deposition and intracellular aggregates of hyperphosphorylated tau, two pathological processes that are tightly intertwined with chronic microglial activation.¹²² Once activated, microglia release cytokines, complement factors, chemokines, and reactive oxygen species (ROS) that amplify neuronal stress and further promote amyloid accumulation, establishing a feed-forward inflammatory loop. EVs from P. gingivalis contain cargoes with clear neuroinflammatory potential, including lipopolysaccharide (LPS) and gingipains. Experimental studies indicate that these vesicles can activate microglia, enhance inflammatory signaling, and promote processes linked to amyloid accumulation and tau pathology.^19,43 P. gingivalis EVs appear to compromise the blood-brain barrier (BBB) integrity by disrupting endothelial tight junctions and promoting vascular dysfunction (Figure 3a).^123,124 Current evidence further suggests that gingival EVs derived from periodontitis patients can access the brain and directly influence neuroimmune and cerebrovascular compartments. These vesicles exhibit pronounced accumulation in hippocampal microglia, implicating them in critical processes such as immune surveillance, synaptic remodeling, and amyloid clearance.⁴⁴ The emerging picture thus extends beyond microbial spillover, highlighting targeted vesicle-mediated reprogramming that destabilizes both microglial homeostasis and barrier function, two processes central to the progression of neurodegenerative diseases. Furthermore, P. gingivalis EVs may impair cognition by activating microglial nuclear factor kappa-B (NF-κB) pathways while reducing brain-derived neurotrophic factor (BDNF) and N-methyl-D-aspartate receptor (NMDAR) expression. They reach the brain through trigeminal and hematogenous routes and contribute to trigeminal nerve-mediated cognitive impairment (Figure 3b).⁴⁴ In addition, P. gingivalis EVs promote neuroinflammation and neuronal ferroptosis through TGM2-mediated dysregulation of mitochondria-associated endoplasmic reticulum membrane (MAM). MAMs, the dynamic contact sites between the mitochondria and endoplasmic reticulum, are critical regulators of mitochondrial Ca²⁺ overload and metabolic failure, both of which are implicated in neuroinflammation and neuronal loss (Figure 3c).⁴⁷

Figure 3.

Pathogenic role of periodontal EVs and associated mechanisms in neurodegenerative disorders. (a and b) EVs originating from diverse periodontal pathogens such as P. gingivalis and A. actinomycetemcomitans possess the remarkable ability to enter the systemic circulation, penetrate the BBB and reach the brain through trigeminal routes. (c and d) These vesicles transport inflammatory components, such as LPS, gingipain and msRNA into the brain, where they can influence key mechanisms involved in neurodegenerative disorders. In particular, periodontal pathogen-derived EVs may promote β-amyloid aggregation and deposition, exacerbate neuroinflammation, and contribute to cognitive impairment.

Abbreviations: EVs, extracellular vesicles; P. gingivalis, Porphyromonas gingivalis; A. actinomycetemcomitans, Aggregatibacter actinomycetemcomitans; BBB, blood–brain barrier; Aβ, β-amyloid; LPS, lipopolysaccharide; BDNF, brain-derived neurotrophic factor; NMDAR, N-methyl-D-aspartate receptor; msRNAs, miRNA-sized sRNAs; RISC, RNA-induced silencing complex; NF-κB, nuclear factor kappa-B; TNF-α, tumor necrosis factor alpha. The red upward/downward arrow indicates increased/decreased levels, respectively.

Similarly, EVs from A. actinomycetemcomitans displayed clear neuroinvasive and neuroinflammatory properties. Following gingival exposure, these vesicles accumulate in the brain and elicit the expression of cerebral TNF-α and IL-6. Their proinflammatory activity is largely driven by TLR4- and TLR8-MyD88 signaling.¹²⁵ Furthermore, gingival injection of A. actinomycetemcomitans EVs facilitated their retrograde transport from the trigeminal ganglion (TG) neuron axon terminals to cell bodies, leading to TG neuronal activation, altered action potentials, and neural inflammation.⁴⁸ This supports a model in which A. actinomycetemcomitans EVs access the central nervous system through both circulatory and neural routes and engage innate immune pathways to amplify neuroinflammatory responses. An additional layer of pathogenicity may be conferred by small bacterial RNAs. EVs derived from A. actinomycetemcomitans can deliver miRNA-sized sRNAs (msRNAs) into host cells, where they are incorporated into the endogenous RNA-induced silencing complex (RISC), thereby engaging the host RNA silencing machinery. Through coordinated vesicular delivery and downstream activation of NF-κB signaling, these cargoes provide a plausible mechanism by which periodontal bacterial EVs reprogram the host inflammatory pathways at the post-transcriptional level (Figure 3d). Their capacity to penetrate the brain further raises the possibility that bacterial RNA cargo contributes directly to chronic neuroimmune activation that underlies ADRD.⁴²

Taken together, these findings support a broader view that periodontal EVs act as mobile neuroinflammatory effectors. By transporting proteases, endotoxins, and regulatory RNAs from the oral cavity to the brain, they provide a mechanistic framework that links chronic periodontitis to microglial dysfunction, BBB breakdown, and progressive neurodegeneration. However, although some studies use gingival injection or gingival gel application to mimic the route by which locally derived periodontal EVs may affect brain, such approaches still do not fully recapitulate the endogenous trafficking patterns or chronic systemic exposure observed in human periodontitis. Given the limited number of studies and the possibility of confounding factors, these findings should therefore be interpreted with caution.

Periodontal EVs and Adverse Pregnancy Outcomes

Accumulating clinical evidence and meta-analyses link periodontal disease to adverse pregnancy outcomes (APOs), including preeclampsia and preterm birth, both of which contribute substantially to maternal and neonatal morbidity and mortality.^126–128 A plausible mechanistic basis for this association has emerged from studies implicating P. gingivalis in placental dysfunction. P. gingivalis impairs the invasive capacity of human trophoblasts, a process essential for normal placental development, and systemic exposure to P. gingivalis LPS in pregnant animals induces fetal malformation, intrauterine growth restriction, and embryo resorption in a dose-dependent manner.¹²⁵ However, the relatively low placental burden of P. gingivalis argues against extensive direct bacterial colonization as the principal driver of pathology.¹²⁹ This discrepancy has shifted attention toward microbial EVs as more plausible mediators of the oral-placental axis. P. gingivalis EVs can be efficiently translocated to the placenta during pregnancy, where they perturb trophoblast homeostasis (Figure 4a). In early pregnancy, P. gingivalis EVs increase oxidative stress in human trophoblasts and skew placental signaling toward an anti-angiogenic and proinflammatory state. Specifically, they decreased the levels of IL-8, placental growth factor (PlGF), and vascular endothelial growth factor (VEGF), while increasing soluble fms-like tyrosine kinase 1 (sFlt-1) and monocyte chemoattractant protein-1 (MCP-1), resulting in an elevated sFlt-1/PlGF ratio, which is a hallmark biomarker of preeclampsia (Figure 4b).^45,130 Under physiological conditions, trophoblast-derived signals coordinate immune and endothelial cell behavior to support maternal-fetal homeostasis. However, P. gingivalis EVs disrupt trophoblast-centered crosstalk, impairing communication with both neutrophils and endothelial cells. Trophoblasts primed with P. gingivalis EVs enhanced neutrophil chemotaxis while losing their normal immunomodulatory capacity and simultaneously suppressing endothelial migration and activation, which restricts maternal angiogenesis (Figure 4c).¹³⁰ These findings support the view that P. gingivalis EVs promote placental pathology by subverting trophoblast functions that normally sustain the immune balance and vascular remodeling.

Figure 4.

Pathogenic role of periodontal EVs and associated mechanisms in adverse pregnancy outcomes. (a) P. gingivalis EVs can penetrate the placental barrier and reach the placenta. (b) P. gingivalis EVs increase oxidative stress in trophoblasts and shift placental signaling toward an anti-angiogenic and proinflammatory state. (c) P. gingivalis EVs disrupt trophoblast crosstalk with both neutrophils and endothelial cells. (d) P. gingivalis EVs inhibit glycolysis and suppress trophoblast cell invasion.

Abbreviations: EVs, extracellular vesicles; P. gingivalis, Porphyromonas gingivalis; PlGF, placental growth factor; VEGF, vascular endothelial growth factor; sFlt-1, soluble fms-like tyrosine kinase 1; MCP-1, monocyte chemoattractant protein-1. The red upward/downward arrow indicates increased/decreased levels, respectively.

In addition, P. gingivalis EVs also appear to be capable of reprogramming trophoblast metabolism by suppressing glycolysis and inducing a more quiescent functional state that preserves viability while impairing invasive capacity, with downstream negative effects on placental and fetal growth (Figure 4d). These functional defects are accompanied by structural alterations in the placental labyrinth, including increased cytoplasmic density and trophoblast hypertrophy, together with subtle expansion of the junctional zone.⁴⁶ Notably, the negative consequences appear dose-dependent; lower EV exposure impairs fetal growth and placental integrity without overt fetal vascular defects or miscarriage, whereas higher exposure results in complete pregnancy loss. Overall, a coherent model is emerging in which periodontal EVs act as systemic placental stressors. By reprogramming trophoblast metabolism, angiogenic signaling, immune crosstalk, and invasive capacity, they offer a biologically plausible explanation for how oral dysbiosis contributes to preeclampsia, fetal growth restriction, and other APOs. However, the evidence remains limited and is complicated by model heterogeneity, uncertain physiological dosing, and mixed interventional findings.^8–11 Although observational studies suggest an association between periodontitis and APOs, the inconsistency of intervention results indicates that the underlying mechanisms are likely multifactorial and not yet fully explained by EV-mediated pathways.

Effect of Extraoral EVs on Periodontitis

Compared with the extensive focus on periodontal EVs acting on distant organs, the reverse direction of communication has been studied less thoroughly. Most work on extraoral EVs in periodontitis has emphasized therapeutic or regenerative applications, particularly those derived from adipose tissue- and bone marrow-derived stem cells. By contrast, the pathogenic roles of extraoral EVs on periodontitis remain far less well defined. Emerging evidence suggests that EVs originating from metabolically and immunologically active tissues, including the liver, adipose tissue, gut microbiota, and bone marrow (as detailed in Table 3 and Figure 5), can modulate periodontal inflammation, pointing to a broader network of interorgan communication in periodontitis.

Table 3.

Effects of Extraoral EVs on Periodontitis

EV-Source	Pathological Background	Functional Cargoes	Mechanisms	Effects on Periodontitis
Liver	T2DM	FASN	FASN-mediated ectopic fatty acid synthesis; NLRP3 inflammasome activation; GSDMD cleavage in PDLSCs	Aggravate periodontal inflammation and alveolar bone loss131
Liver	Hepatocellular Carcinoma	TNF-α	Upregulated osteoclast markers TRAF6 and CTSK expressions	Promote osteoclast differentiation132
Liver	NAFLD/obesity	miR-1, miR-122	miR-1 induces inflammation via KLF4 suppression and NF-κB activation; miR-122 induces mitochondrial dysfunction via Arl2 suppression	Potentially pro-inflammatory133,134
Adipose -derived mesenchymal stem cells	Reparative state	/	Inhibit TRAF6/COX-2/RANK signaling axis	Inhibit bone resorption135
Adipose tissue macrophages	Obesity	Inflammatory miRNAs such as miR-30e-5p	Jazf1 promoter hypermethylation and Jazf1 downregulation in macrophages; Activate NF-κB pathway	Promote periodontal macrophage activation and M1 polarization, exacerbate periodontal inflammation136
Adipose tissue	Obesity/diabetes	miR-155, miR-27a	Activate NF-κB pathway; JAK-STAT pathway; RANK/RANKL/OPG pathway	Exacerbate inflammation and bone loss137–141
Gut microbiota	Gut microbiota dysbiosis	/	Treg/Th17 imbalance; inducing trained immunity	May aggravate periodontitis142–146
Bone Marrow Mesenchymal Stem Cell	/	miR-1246	OPG-RANKL-RANK axis; ROS/JNK signaling; restore Th17/Treg balance	Promote periodontal regeneration147–152
Bone Marrow	Traumatic Brain Injury	miR-1224	Activate NF-κB signaling	Promote osteoclast differentiation and bone loss153

Figure 5.

The integrative role of extraoral-derived EVs in periodontitis pathogenesis. Extraoral EVs originating from diverse peripheral sources such as liver, adipose tissue, and gut microbiota, bone marrow are depicted as pivotal entities in the reverse crosstalk between the systemic organ and the periodontitis. These vesicles deliver FASN, miR-3e-5p, and other molecular cargo that can influence periodontitis.

Abbreviations: EVs, extracellular vesicles; T2DM, type 2 diabetes mellitus; FASN, fatty acid synthase; PDLCs, periodontal ligament cells; GSDMD, gasdermin D; E. coli, Escherichia coli; L. animalis, Lactobacillus animalis; ADSCs; adipose-derived stem cells. The red upward/downward arrow indicates increased/decreased levels, respectively.

Among these extraoral sources, liver-derived EVs appear to be particularly relevant across a range of systemic diseases, including diabetes, NAFLD, obesity, and cancer. Under these conditions, hepatocytes are subjected to chronic low-grade inflammation, steatosis, lipotoxicity, oxidative stress, and mitochondrial dysfunction, all of which reshape EV biogenesis and cargo sorting. As a result, the liver releases vesicle populations enriched in inflammatory mediators, metabolic enzymes, and regulatory miRNAs that can transmit hepatic stress signals to peripheral tissues, including the periodontium. A central mechanism by which these vesicles may exacerbate periodontitis is through the induction of inflammatory and metabolic reprogramming in recipient periodontal cells. Particularly compelling evidence comes from diabetic liver-derived EVs, which preferentially accumulate in periodontal tissues and deliver FASN to periodontal ligament cells (PDLCs), thereby inducing ectopic fatty acid synthesis, activating the NLRP3 inflammasome, promoting gasdermin D (GSDMD) cleavage, and triggering pyroptosis. Hepatic deletion of FASN attenuates these effects and mitigates periodontal bone loss, supporting a direct pathogenic role for liver-derived EVs in diabetes-associated periodontal destruction.¹³¹ Furthermore, EVs derived from hepatocellular carcinoma cells have been shown to promote osteoclast differentiation, as reflected by an increased number of TRAP-positive multinucleated osteoclasts, at least in part through the transfer of TNF-α.¹³² Additional studies suggest that miRNAs from hepatocyte-derived EVs under NAFLD or obesity may further amplify periodontal inflammation. For instance, miR-1-containing EVs released from steatotic hepatocytes can activate NF-κB signaling through repression of kruppel-like factor 4, whereas miR-122-rich EVs from fatty hepatocytes impair mitochondrial function in distant tissues by targeting metabolic regulators such as ARL2.^133,134 Through these mechanisms, hepatocyte-derived EVs may enhance inflammatory tone and thus represent a potential mediator linking hepatic metabolic dysfunction to periodontitis.

Adipose tissue represents another major extraoral source of EVs with potential relevance to periodontitis. Beyond its role in energy storage, adipose tissue functions as a metabolically and immunologically active endocrine organ that contributes to the regulation of systemic inflammatory tone. Under physiological conditions, EVs released from adipose-derived stem cells (ADSCs) exert anti-inflammatory and reparative effects.¹⁵⁴ Indeed, EVs from adipose-derived stem cells have been shown to attenuate periodontal inflammatory osteoclastogenesis, reduce alveolar bone resorption, and suppress the expression of osteoclast-related mediators such as TRAF6, COX-2, and RANK, supporting a protective role in the repair of periodontal bone defects.¹³⁵ By contrast, in obesity and diabetes, adipose tissue undergoes profound inflammatory remodeling characterized by dysregulated adipokine secretion, altered macrophage polarization, and oxidative imbalance, all of which reshape EV cargo composition. In accumulated visceral adipose tissue, macrophages acquire an activated phenotype through hypermethylation of the Jazf1 promoter. EVs released from these macrophages are enriched in inflammatory miRNAs, including miR-30e-5p, which can promote NF-κB signaling activation and M1 polarization in the periodontium, reinforcing chronic periodontal inflammation.¹³⁶

In addition, EVs derived from obese adipose tissue are enriched in mediators such as miR-155 and miR-27a, which regulate NF-κB and JAK/STAT signaling pathways.^137–140 Upon reaching periodontal tissues, these vesicles may stimulate gingival epithelial cells, PDLCs, and resident macrophages to produce proinflammatory cytokines, including IL-1β, IL-6, and TNF-α, thereby amplifying local inflammatory responses and disrupting extracellular matrix homeostasis. At the same time, they may reduce the expression of osteogenic markers such as RUNX2, ALP, and OPG, while shifting the RANKL/OPG balance towards osteoclastogenesis, ultimately promoting alveolar bone resorption.¹⁴¹

Although the role of gut microbiota-derived EVs in periodontitis remains poorly defined, these vesicles have emerged as key regulators of immunity, metabolism, and bone homeostasis. Importantly, evidence suggests that they can modulate osteogenic and osteoclastic processes as well as inflammatory set points, raising the possibility of an indirect connection between the gut microbiota and periodontitis pathogenesis.^155–157 Under homeostatic conditions, gut microbiota-derived vesicles may contribute to skeletal integrity, whereas dysbiosis is likely to shift the gut bacterial EV pool toward a more pro-resorptive and proinflammatory profile. For example, Lactobacillus animalis-derived EVs, which contain a variety of functional proteins, can access the femoral head and promote angiogenesis, osteogenesis, and anti-apoptotic signaling.¹⁴² In contrast, long-term glucocorticoid treatment induces gut dysbiosis, reduces the abundance of L. animalis and its secreted vesicles, and thereby impairs their bone-protective effects.¹⁴² Similarly, EVs derived from Escherichia coli have been shown to disrupt the Treg/Th17 balance and promote IL-17-driven osteoclastic bone resorption.¹⁴³ Although these findings are biologically suggestive, they do not directly establish a role for gut microbiota-derived EVs in periodontitis. Gut microbiota-derived EVs may also influence periodontal pathology through trained immunity.¹⁴⁴ By reprogramming bone marrow immune cells to mount exaggerated inflammatory responses upon secondary stimulation, these vesicles could potentially exacerbate periodontal tissue destruction. Given the emerging role of trained immunity in mediating the bidirectional interplay between periodontitis and systemic diseases,^145,146 gut microbial EVs may therefore represent a candidate mechanism linking intestinal dysbiosis to periodontal pathology. Nonetheless, direct experimental evidence in periodontitis remains limited, and this proposed pathway should be regarded as hypothesis.

Bone marrow-derived EVs, in contrast, have been studied mainly in the context of periodontal repair and regeneration. Bone marrow mesenchymal stem cell (BMSC)-derived EVs have been shown to attenuate inflammation and promote osteogenic differentiation in periodontal tissues. Mechanistically, these vesicles modulate pathways involved in inflammation, Th17/Treg balance, and bone formation, including the OPG-RANKL-RANK axis and JNK signaling, while also influencing macrophage polarization, TGF-β1 expression, and mitochondrial function.^147–150 Through these actions, BMSCs derived EVs enhance the proliferation and migration of PDLSCs, alleviate oxidative stress, suppress osteoclast formation, and create a tissue environment that is more conducive to periodontal healing. Notably, the biological activity of these EVs appears to depend on the status of the donor cells, as inflammatory priming or hypoxic preconditioning can further enhance their reparative potential.^151,152 Thus, unlike the predominantly pathogenic EVs released from metabolically diseased liver or adipose tissue, bone marrow-derived EVs have generally been associated with the maintenance of periodontal homeostasis and regenerative responses. However, under certain pathological conditions, bone marrow-derived EVs may also exert detrimental effects on bone remodeling. For example, traumatic brain injury has been associated with a persistent proinflammatory bone marrow milieu and progressive bone loss, with miR-1224 implicated in NF-κB activation and altered osteoblast differentiation.¹⁵³ Although these findings were not obtained in periodontitis models, they support the broader concept that pathological alterations in the bone marrow environment can modify EV composition and function, thereby influencing bone remodeling pathways in ways that could plausibly affect alveolar bone susceptibility.

Importantly, it should be noted that despite the emerging framework, direct evidence for a specific role of these extraoral EV populations in periodontitis remains limited. Much of the current mechanistic understanding is derived from studies performed in non-periodontal target tissues, systemic metabolic disease models, bone remodeling systems outside the periodontal niche, and the extrapolation of these findings to periodontitis therefore remains provisional, more precise evidence is needed to define their actual contribution in periodontal disease.

Periodontal EVs for Systemic Therapy

Beyond their pathogenic relevance, EVs are being actively explored as therapeutic platforms owing to their relative stability, low immunogenicity, and intrinsic biocompatibility. EVs derived from gingival mesenchymal stem cells (GMSCs) and PDLSCs have attracted particular interest in periodontal research. These vesicles may recapitulate some of the immunomodulatory and reparative functions of their parent cells while offering advantages in delivery, safety, and engineering flexibility. Such engineering flexibility is likely to be increasingly important for periodontal therapy, given that recent non-EV studies have shown that rational nanoengineering strategies, such as interfacial engineering of Zn(OH)F/CaF2 nanoparticles and atomically engineered asymmetric nanozymes, can substantially improve antibacterial efficacy, inflammatory control, and protection against periodontal tissue destruction.^158,159 These advances indirectly support the promise of engineered EVs as next-generation multifunctional therapeutic platforms.

GMSC-derived EVs are emerging as versatile therapeutic effectors, with both local and systemic applications. In experimental periodontitis, local administration of GMSC-EVs attenuates alveolar bone loss and reduces osteoclast activity, supporting their role in controlling inflammatory bone destruction.^160–162 Beyond the periodontium, GMSC-EVs have demonstrated regenerative efficacy in multiple tissue, including skin and peripheral nerve. Their therapeutic activity is thought to depend in part on the suppression of NF-κB signaling and the consequent dampening of the inflammatory microenvironment. In models of cutaneous injury, GMSC-EVs accelerate wound closure and improve repair, even under diabetic conditions.¹⁶³ In peripheral nerve injury models, they also promote a repair-supportive Schwann cell phenotype, enhance axonal growth, and improve functional recovery.¹⁶⁴

PDLSC-derived EVs display considerable therapeutic potential, particularly owing to their combined immunoregulatory, osteogenic, and pro-angiogenic properties. PDLSC-EVs restore immune homeostasis by modulating the Th17/Treg balance through the miR-155-5p-SIRT1 pathway and suppressing inflammasome-associated signaling, as shown in experimental autoimmune encephalomyelitis.^165,166 In parallel, they enhance osteogenic activity by promoting the migration, differentiation, and mineralization of osteogenic precursor cells, with upregulation of canonical osteogenic markers. In vivo, these effects translate into increased new bone formation in calvarial defect models, supported by the activation of AMPK, MAPK, and insulin signaling pathways.¹⁶⁷ During osteogenic differentiation, their pro-regenerative profile may be shaped by dynamic changes in EV-non-coding RNA cargo, including circRNAs and lncRNAs.¹⁶⁸ Under inflammatory conditions, PDLSC-EVs also enhance endothelial angiogenic responses by upregulating the expression of the vascular markers CD31 and VEGF-ɑ, further supporting tissue repair.¹⁶⁹ In addition, engineered PDLSC-EVs have shown value as drug delivery vehicles; for example, emodin-loaded PDLSC-EVs enhance antimicrobial photodynamic therapy and suppress biofilm formation and metabolic activity of Streptococcus mutans and Lactobacillus acidophilus.¹⁷⁰ Collectively, these findings identify PDLSC-EVs as a promising platform for periodontal regeneration, bone repair, and antimicrobial intervention. Although substantial challenges remain in large-scale production, cargo standardization, tissue specificity, and in vivo delivery, current evidence supports the view that engineered periodontal EVs could extend the therapeutic reach of oral stem cell biology far beyond the periodontium.

Challenges and Future Perspectives in EV-Mediated Intercellular Communication Between Periodontitis and Systemic Diseases

Although interest in EV-mediated oral-systemic communication has grown rapidly, the field remains methodologically and conceptually immature. Most available evidence comes from in vitro studies or animal models that are indispensable for mechanistic exploration but do not fully capture the chronicity, heterogeneity, microbial complexity, and comorbidity burden of human periodontitis. Consequently, a substantial gap remains between mechanistic plausibility and clinical relevance. In addition, publication bias toward positive findings, limited reporting of negative or conflicting results, and substantial heterogeneity in experimental design may exaggerate the apparent consistency of proposed mechanisms. Consequently, although EVs provide a compelling conceptual framework for linking periodontitis with extraoral pathology, many of these pathways remain best viewed as experimentally supported hypotheses rather than conclusively established causal mechanisms.

Source attribution is a central, unresolved issue. In complex biological fluids and periodontal tissues, EV populations are derived from multiple host and microbial sources; however, current approaches often lack robust markers and analytical resolution to assign vesicles to specific cellular or microbial origins with confidence. This challenge is particularly acute in periodontitis, where bacterial EVs, host-derived EVs from inflamed periodontal tissues, and vesicles originating from distant organs may coexist and interact, complicating the interpretation of biomarker studies and mechanistic experiments. Although direct analysis of EVs from periodontal tissues could provide a more precise view of local pathogenic processes, overcoming the current technical barriers will require new high-resolution approaches. Emerging technologies such as proximity barcoding assays now enable single-EV surface protein profiling,^171,172 offering new opportunities to resolve heterogeneous EV subpopulations across diverse cell types and tissues, clarifying their functional roles in disease, and broadening the current focus beyond bacterial vesicles, especially those from tractable pathogens such as P. gingivalis, toward host-derived EVs from periodontal tissues and biofluids.

Standardization is the second major challenge. Isolation methods, normalization strategies, marker selection, and functional assays vary widely across studies, making comparisons difficult and resulting in uneven reproducibility. The common practice of classifying vesicles as exosomes or microvesicles on size alone further complicates the literature. Consistent operational terminology, adherence to EV reporting guidelines, and improved characterization pipelines are critical for advancing the field.

Another limitation concerns in vivo tracking and administration methods. Current EV labeling and tracking strategies are limited by their low specificity, dye-associated artifacts, and poor resolution of endogenous EV dynamics in vivo.¹⁷³ As a result, many studies on inter-organ EV trafficking rely on indirect evidence or exogenous administration models rather than direct observation of physiological EV secretion and uptake. Developing robust approaches for real-time tracking of endogenous EV secretion, biodistribution, and uptake in vivo would markedly strengthen the causal inference. Furthermore, in addition to conventional administration methods, such as intraperitoneal or intravenous injection, simulating the EV transfer pathway in periodontitis through gingival administration strategies may better recapitulate the systemic effects of periodontitis on distant diseases.

From a translational perspective, EVs occupy a unique position in the interface between oral and systemic health. In such two-way relationships, a single move can affect the entire situation. Do EVs have the therapeutic potential to kill two birds with one stone? Periodontal stem cell-derived EVs are promising not only for periodontal regeneration, but also for nerve repair, bone healing, and broader immunomodulation, raising the possibility that EV-based interventions could simultaneously address oral disease and its systemic sequelae. Engineered EVs further expand the therapeutic landscape by enabling precise tissue targeting and cargo delivery. Looking ahead, progress will depend on integrating mechanistic studies with multi-omics profiling, bioengineering innovation, and well-designed clinical investigations. Defining EV cargo dynamics across tissues and disease stages, establishing standardized analytical pipelines, and validating findings in human cohorts are essential.

Conclusions

In summary, emerging evidence supports a role for EVs in mediating communication between periodontal tissues and distant organs, thereby contributing to the systemic implications of periodontitis. Among the currently proposed associations, mechanistic evidence is most developed in cardiovascular and systemic bone-related contexts, where EV-mediated effects on inflammation, endothelial dysfunction, osteoclastogenesis, and bone remodeling have been most consistently described, whereas evidence in other organ systems remains more limited. Importantly, this review highlights a bidirectional framework, encompassing not only the potential contribution of periodontal EVs to extraoral diseases but also the reciprocal influence of EVs derived from distant tissues on periodontal homeostasis and pathology. Nevertheless, support for this bidirectional EV-mediated crosstalk remains uneven, with stronger evidence currently available for the effects of periodontal EVs on systemic diseases than for the reverse direction. Moreover, the existing literature is still derived predominantly from preclinical studies. Further mechanistic, translational, and clinical investigations are needed to validate these pathways and determine their relevance in human disease.

Funding Statement

This work was supported by Shandong Provincial Natural Science Foundation grant ZR2024QH504.

Abbreviations

A. Actinomycetemcomitans, Aggregatibacter actinomycetemcomitans; Aβ, β-amyloid; ACSL4, Acyl-CoA Synthetase Long-Chain Family Member 4; ADRD, Alzheimer’s disease and related dementias; ADSCs, Adipose-derived stem cells; AKT, Protein kinase B; ALP, Alkaline phosphatase; APOs, Adverse pregnancy outcomes; AUC, Area under the curve; B3GAT1, β-1,3-glucuronosyltransferase; BBB, Blood–brain barrier; BDNF, Brain-derived neurotrophic factor; BMSCs, Bone marrow mesenchymal stem cells; C4’-MPLA, C4’monophosphoryl lipid A; CitH3, Citrullinated histone H3; CYP4A10, Cytochrome P450, family 4, subfamily A, polypeptide 10; Cyp4f40, Cytochrome P450, family 4, subfamily f, polypeptide 40; DR, Diabetic retinopathy; EMT, Epithelial-to-mesenchymal transition; eNOS, Endothelial nitric oxide synthase; ESCRT, Endosomal sorting complex required for transport; EVs, Extracellular vesicles; F. alocis, Filifactor alocis; FASN, Fatty acid synthase; GMSCs, Gingival mesenchymal stem cells; GSDMD, Gasdermin D; GSK-3β, Glycogen synthase kinase 3β; HAECs, Human aortic endothelial cells; HbA1c, Glycated hemoglobin A1c; HOMA-IR, Homeostasis model assessment of insulin resistance; HUVECs, Human umbilical vein endothelial cells; ICAM-1, Intercellular adhesion molecule 1; IL-1β, Interleukin-1β; IL-6, Interleukin 6; ILVs, Intraluminal vesicles; LPS, Lipopolysaccharide; MAM, Mitochondria-associated endoplasmic reticulum membrane; MCP-1, Monocyte chemoattractant protein-1; MMP-9, Matrix metalloproteinase 9; msRNAs, miRNA-sized sRNAs; MVBs, Multivesicular bodies; NAFLD, Nonalcoholic fatty liver disease; NFATc1, Nuclear factor-activated T cell c1; NF-κB, Nuclear factor kappa-B; NMDAR, N-methyl-D-aspartate receptor; NPAS2, Neuronal PAS family member 2; NPLA, Nonphosphorylated tetra-acylated lipid A; OPG, Osteoprotegerin; P. gingivalis, Porphyromonas gingivalis; PAI-1, Plasminogen activator inhibitor-1; PD-L1, Programmed death-ligand 1; PDLCs, Periodontal ligament cells; PDLFs, Periodontal ligament fibroblasts; PDLSCs, Periodontal ligament stem cells; PECAM-1, Platelet endothelial cell adhesion molecule-1; PlGF, Placental growth factor; PPAD, Peptidylarginine deiminase; PTGS2, Prostaglandin-endoperoxide synthase 2; RISC, RNA-induced silencing complex; ROCK, Rho-associated kinase; ROS, Reactive oxygen species; RUNX2, Runt-related transcription factor 2; S.oralis, Streptococcus oralis; SAA3, Serum Amyloid A3; SASP, Senescence associated secretory phenotype; SCD-1, Stearoyl-CoA desaturase 1; sFlt-1, Soluble fms-like tyrosine kinase 1; SREBP1, Sterol regulatory element-binding protein 1; T. forsythia, Tannerella forsythia; T2DM, Type 2 diabetes mellitus; TFRC, Transferrin receptor; TG, Trigeminal ganglion; TGF-β, Transforming growth factor β; TGM2, Transglutaminase 2; THBS1, Thrombospondin 1; TLR4, Toll-like receptor 4; TNF-α, Tumor necrosis factor α; VEGF, Vascular endothelial growth factor; VSMCs, Vascular smooth muscle cells.

Data Sharing Statement

Data sharing is not applicable to this article, as no new data were created or analyzed in this study.

Disclosure

The authors report no conflicts of interest in this work.