Gut microbiota-derived extracellular vesicles: bridging microbial-host crosstalk in metabolic disorders

Abstract

Gut microbiota-derived extracellular vesicles have emerged as crucial mediators in microbe-host communication, not only facilitating intracellular communication, quorum sensing, and horizontal gene transfer among bacteria but also playing a central role in cross-kingdom dialogue. In recent years, bacterial extracellular vesicles (BEVs) have attracted widespread attention due to their ability to carry a diverse array of bioactive molecules-such as proteins, lipids, and nucleic acids-and deliver them to host cells, thereby precisely regulating host metabolic and immune homeostasis. This review systematically elaborates the entire biological process of BEVs, from their biogenesis to functional interactions with host cells, with a specific emphasis on revealing their roles in the pathogenesis of various metabolic diseases-including obesity, type 2 diabetes (T2DM), metabolic dysfunction-associated steatotic liver disease (MASLD), atherosclerosis, and hypertension-at both molecular and cellular levels. Furthermore, leveraging their inherent stability, biocompatibility, and targeting capabilities, we discuss the translational potential and challenges of BEVs in the diagnosis and treatment of metabolic disorders. Beyond summarizing the latest research advances on BEVs in metabolic disorders, this review provides a critical analysis of current mechanistic insights and clinical translation pathways, aiming to establish a theoretical framework for developing novel microbiome-based metabolic interventions. Deciphering the BEV-mediated microbiota-host interaction network holds promise for pioneering new strategies for the precision prevention and treatment of metabolic disease.

Introduction

Metabolic disorders, including obesity, type 2 diabetes mellitus (T2DM), metabolic dysfunction-associated steatotic liver disease (MASLD), hypertension, and atherosclerosis, have become major global public health crises, yet current treatment strategies remain limited [1–3]. The gut microbiota, also referred to as the intestinal microbiome, is increasingly recognized as a central link in the pathogenesis of these metabolic diseases [4–6]. This complex ecosystem residing in the distal gut plays a crucial role in host metabolism, immune responses, and inflammatory processes [7, 8]. Dysbiosis of the gut microbiota has been closely associated with obesity [9], T2DM [10], MASLD [11], hypertension, and atherosclerosis [12], often accompanied by compromised gut barrier integrity and altered microbial composition [13]. Among the various mediators of microbe-host crosstalk, bacterial extracellular vesicles (BEVs) derived from gut microbes have emerged as pivotal effector molecules. In this review, the term “gut microbiota-derived extracellular vesicles” refers to the total pool of extracellular vesicles released by bacteria colonizing the human intestinal tract, representing a highly heterogeneous and dynamic mixture that reflects the real-time status of the gut microbiota and encompasses vesicles produced by both Gram-negative and Gram-positive bacteria through different biogenesis pathways [14]. This review systematically elaborates the entire biological process of BEVs, from their biogenesis to functional interactions with host cells, with a specific emphasis on revealing their roles in the pathogenesis of various metabolic diseases at both molecular and cellular levels, and further discusses their translational potential.

BEVs are predominantly secreted from the outer membrane or cytoplasmic membrane of gut microbiota and other microorganisms [15]. Recent studies have revealed that BEVs play essential roles in facilitating communication not only among microorganisms but also between microbes and their host [15]. Acting as versatile carriers of bioactive molecules, BEVs precisely modulate host metabolic and immune homeostasis [16]. They regulate host signaling pathways and biological activities of target cells primarily through the transfer and delivery of effector molecules to target cells or organs, thereby promoting microbial activity [17]. Although significant progress has been made in understanding the roles of BEVs in intestinal diseases and oncology, their mechanistic contributions to metabolic disorders-such as obesity, T2DM, and MASLD-remain incompletely characterized and require systematic elucidation. Furthermore, key questions—including how BEVs traverse the gut barrier, and the relative contributions of intact vesicles versus their cargo to systemic effects—remain subjects of active debate and substantial knowledge gaps. In this review, we not only focus on the biological properties of BEVs and their mechanistic roles in obesity, T2DM, MASLD, hypertension, and atherosclerosis but also provide an in-depth exploration of BEV-mediated inter-organ communication, particularly through the “gut-target organ axis”. We critically evaluate the limitations of current research models and theoretical frameworks, and further discuss the therapeutic potential of BEVs in clinical practice. Our aim is to clarify the specific mechanisms underlying BEV-facilitated microbiota-host dialogue in metabolic diseases and to propose novel targeted intervention strategies directed at the gut microbiome.

BEVs biological process

BEVs serve as key messengers mediating microbiota–host interactions. As previously mentioned, “gut microbiota-derived EVs” is an umbrella term encompassing membrane-bound vesicles of diverse sizes, bioactive cargoes, and functional roles, produced by all bacteria within the intestinal ecosystem. Their composition is not static but is profoundly influenced by the host’s diet, health status, and the ecological structure of the microbiota itself [18]. The functional journey of BEVs begins with biogenesis, followed by interaction with host cells, and initially manifests within the local gut microenvironment. This section systematically elaborates the complete biological process of BEVs, from their formation to functional execution, including their biogenesis mechanisms, routes of internalization, and core roles within the intestine, thereby laying the groundwork for understanding their systemic effects in metabolic diseases.

Gut dysbiosis and BEVs: a bidirectional relationship

A critical reciprocal interaction exists between gut microbiota dysbiosis and BEVs. On one hand, dysbiosis fundamentally reshapes the BEV landscape. Reduced microbial diversity, overgrowth of pathobionts (e.g., Enterobacteriaceae, Fusobacterium nucleatum), and a decreased Bacteroidetes/Firmicutes ratio leads to quantitative and qualitative changes in the BEV pool: increased production of immunostimulatory BEVs, altered cargo composition (enriched in LPS, CpG DNA, and virulence factors), and modified surface properties that may enhance translocation across the gut barrier [19]. On the other hand, BEVs themselves actively modulate gut microbiota composition. BEVs from beneficial bacteria can suppress pathogen growth via antimicrobial activity or quorum sensing interference, whereas pathogenic BEVs may promote dysbiosis by delivering virulence factors to neighboring commensals [17, 20]. Thus, BEVs act as both indicators and mediators of gut ecosystem stability—their molecular signatures may serve as real-time functional readouts of the intestinal environment, while their altered profiles in turn contribute to disease progression. This bidirectional relationship underscores the potential of BEV-based strategies for diagnosing and therapeutically rebalancing dysbiosis in metabolic disorders.

Biogenesis of BEVs

BEVs are nanoscale, lipid bilayer-enclosed particles secreted by both Gram-negative and Gram-positive bacteria, which play a pivotal role in mediating microbe-host cellular communication [21]. Gram-negative bacteria predominantly produce three distinct BEV subtypes [22]: (1) Outer membrane vesicles (OMVs; 20–250 nm) formed through outer membrane budding, enriched with outer membrane proteins, DNA, RNA, and lipid components [23]; (2) Outer-inner membrane vesicles (OIMVs) and explosive outer membrane vesicles (EOMVs), generated via phage endolysin-triggered cell lysis, which encapsulate randomly sorted cytoplasmic contents including outer/inner membrane proteins, virulence factors, and nucleic acids [24]. In contrast, Gram-positive bacteria generate cytoplasmic membrane vesicles (CMVs; 20–400 nm) derived from their plasma membrane and containing both membrane-associated and cytosolic components [25] (see Fig. 1). The diverse biogenesis pathways and specific cargo molecules carried by these vesicles determine their functional versatility.

Fig. 1.

Different biogenic mechanisms of BEVs. Gram‑negative bacteria produce BEVs through outer membrane budding (forming OMVs) or explosive cell lysis triggered by phage‑derived endolysins (forming EOMVs and OIMVs). Gram‑positive bacteria generate CMVs via endolysin‑induced blebbing and lysis of the plasma membrane

Internalization pathways of BEVs

BEVs interact with host cells through three primary mechanisms to deliver their bioactive cargo into the cell, thereby regulating host physiological and pathological processes. These mechanisms include: (1) direct activation of host receptors; (2) barrier translocation and cellular uptake of intact BEVs; and (3) delivery of BEVs cargo and activation of downstream signaling pathways [26]. Unlike free bacterial products, e.g., soluble lipopolysaccharide (LPS), DNA, or peptidoglycan. BEVs possess unique advantages as a delivery system: their lipid bilayer protects labile cargo from enzymatic degradation; they enable the co-delivery of multiple synergistic signals (e.g., LPS + DNA + protein toxins) to the same target cell; and critically, they facilitate access to specific intracellular compartments—such as the cytosol via endosomal escape or direct membrane fusion—that are otherwise inaccessible to free molecules [27, 28]. This vesicular delivery mechanism distinguishes BEVs from mere leakage of bacterial components and underlies their potent biological effects in metabolic diseases.

Direct activation of host receptors

BEVs carry a set of molecules known as microbe-associated molecular patterns (MAMPs), which can directly bind to various pattern recognition receptors (PRRs) expressed on host immune cells and other somatic cells. These include Toll-like receptors (TLRs) and cytosolic PRRs, such as NOD-like receptors and RIG-I-like receptors [28]. BEVs activate the NF-κB/MAPK pathway through LPS–TLR4 or peptidoglycan–NOD1/2 interactions, promoting the release of inflammatory cytokines, interfering with insulin signaling, and ultimately contributing to metabolic dysregulation [29]. These early signaling events, triggered by receptor activation, further facilitate the internalization of BEVs into host cells through specific endocytic pathways [30].

Gut barrier translocation and cellular uptake of BEVs

Translocation across the intestinal barrier is a critical step for BEVs to exert systemic effects, primarily occurring via transcellular and paracellular pathways [31]. The transcellular route relies on various endocytic mechanisms: after being internalized by intestinal epithelial cells, BEVs are transported and processed intracellularly through the endolysosomal system. The paracellular pathway becomes particularly significant under pathological conditions such as T2DM and obesity. Persistent metabolic stress compromises intestinal barrier integrity and downregulates tight junction protein expression, facilitating the passive diffusion of BEVs and their cargo (e.g., LPS and DNA) through intercellular spaces into the systemic circulation, thereby exacerbating systemic inflammation and metabolic disorders [31].

It is noteworthy that the precise mechanisms and extent of BEV translocation from the intestinal lumen to the systemic circulation remain a focus of current research. Two primary pathways are currently proposed: (1) translocation of intact BEVs, which may occur under conditions of impaired intestinal barrier function (e.g., “leaky gut” commonly seen in metabolic diseases) via paracellular leakage or M-cell transcytosis [32]; and (2) a “cargo delivery” model, wherein BEVs are internalized by intestinal epithelial cells and release bioactive molecules (e.g., LPS, DNA, miRNAs), which are then relayed by the host cell’s signaling systems to the circulation, indirectly mediating distal effects [27]. These findings indicate that the translocation and transport of bacterial components or metabolites contribute importantly to the pathogenic process.

Regarding cellular uptake, BEVs are recognized and internalized by host cells through seven principal mechanisms [33]: macropinocytosis, clathrin-mediated endocytosis, lipid raft-mediated uptake, caveolin-mediated endocytosis, and direct membrane fusion, scavenger receptor-mediated endocytosis, and phagocytosis [34] (see Fig. 2). Macropinocytosis, clathrin-mediated endocytosis, lipid raft-mediated uptake, caveolin-mediated endocytosis, and direct membrane fusion have been extensively reviewed elsewhere [34]. Briefly, macropinocytosis involves actin-driven membrane ruffling that non-selectively engulfs BEVs; clathrin-mediated endocytosis relies on AP2 adaptor proteins and dynamin for vesicle scission; lipid raft- and caveolin-mediated pathways are cholesterol-dependent and enriched in specific membrane microdomains; direct membrane fusion allows BEV cargo to be delivered directly into the host cytosol [35–38].

Fig. 2.

Seven pathways by which host cells internalize BEVs. Macropinocytosis: Effector molecules released upon cell contact induce reorganization of the host cell cytoskeleton. Actin filaments polymerize, forming membrane protrusions that engulf BEVs into large macropinosomes. Clathrin-Mediated Endocytosis: Ligand binding to receptors within clathrin-coated pits initiates invagination. With the aid of GTPase dynamin, pits pinch off to form clathrin-coated vesicles containing BEVs. Lipid Raft-Mediated Process: Cholesterol and other lipids aggregate within lipid raft domains, causing membrane curvature and invagination, allowing BEV entry without clathrin or caveolin. Caveolae-Mediated Endocytosis: Caveolae are sphingolipid- and cholesterol-rich subdomains of glycolipid rafts. Their invagination internalizes BEVs into caveolar vesicles. Direct Membrane Fusion: BEVs fuse directly with the host plasma membrane, delivering their cargo (e.g., proteins, nucleic acids) into the cytosol without passing through endosomes. Scavenger Receptor-Mediated Endocytosis: Scavenger receptors (e.g., SR-A, MARCO, CD36) on macrophages and dendritic cells recognize negatively charged BEV surface components, facilitating internalization often linked to immune tolerance rather than overt inflammation. Phagocytosis: Professional phagocytes (macrophages, neutrophils, dendritic cells) engulf large BEVs or BEV aggregates (> 0.5 μm) via actin-dependent membrane remodeling. The resulting phagosome matures into a phagolysosome, where BEVs are degraded and their contents presented as antigens

Scavenger receptors (e.g., scavenger receptor class A (SR-A), macrophage receptor with collagenous structure (MARCO), cluster of differentiation 36 (CD36)) expressed on macrophages and dendritic cells recognize these BEVs, facilitating their internalization without necessarily triggering robust inflammatory activation, thereby contributing to immune tolerance or antigen clearance [39, 40]. Phagocytosis is an actin-dependent process employed primarily by professional phagocytes such as macrophages, neutrophils, and dendritic cells to internalize larger BEVs (or BEV aggregates) > 0.5 μm in diameter. This process is initiated by the engagement of phagocytic receptors (e.g., Fc receptors, complement receptors, or scavenger receptors) and leads to the formation of a phagosome that subsequently matures into a phagolysosome, where BEVs are degraded and their contents processed for antigen presentation [41, 42].

For instance, BEVs can exploit cholesterol-rich microdomains of lipid rafts for internalization; Staphylococcus aureus BEVs use this route to deliver virulence factors to adipocytes, promoting TLR2-dependent inflammation [43]. Clathrin-mediated endocytosis serves as a key pathway for LPS delivery by BEVs; for example, BEVs from enterohemorrhagic E. coli (EHEC) activate TLR4 via AP2 adaptor proteins, driving metabolic inflammation [32]. Some BEVs (e.g., from Pseudomonas aeruginosa) can directly deliver proteins into hepatocytes via membrane fusion, interfering with mitochondrial β-oxidation and contributing to hepatic lipid metabolic disorders [44]. Collectively, these diverse internalization pathways underscore a key concept: BEVs function as nature’s nanoparticles that deliver bacterial signaling molecules into specific host cell compartments—a capability distinct from the passive diffusion of free bacterial products across damaged barriers or the random release of bacterial DNA into the circulation [45, 46]. This compartment-specific delivery, particularly the access to the cytosol and endosomal networks, allows BEVs to trigger signaling cascades (e.g., cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING), non-canonical inflammasome) that free LPS or DNA cannot efficiently activate [47, 48]. Importantly, the relative contribution of scavenger receptor-and phagocytosis-mediated uptake versus other pathways under different pathological conditions (e.g., obesity vs. steady state) remains an active area of investigation [17, 27]. Future studies should employ more precise tracing technologies—such as dual labeling of BEV membranes and cargo—to determine which translocation and uptake mechanisms dominate under specific pathological conditions.

Delivery of BEVs contents and activation of signaling pathways

Following endocytic internalization, BEVs release their bioactive cargo (nucleic acids, proteins, and metabolites) to directly modulate host cell metabolism and immune homeostasis. LPS carried by BEVs engages the TLR4/MD2 (myeloid differentiation factor 2) complex, initiating both MyD88 (myeloid differentiation primary response protein 88) and TRIF (TIR-domain-containing adapter-inducing interferon-β)-dependent signaling pathways. This leads to the activation of NF-κB and interferon regulatory factor 3 (IRF3), thereby inducing the production of pro-inflammatory cytokines and type I interferons [29, 49]. Bacterial DNA encapsulated within BEVs can be recognized by intracellular receptors such as TLR9 or cGAS. Upon sensing bacterial DNA, cGAS catalyzes the synthesis of the second messenger cyclic GMP-AMP (cGAMP), which activates STING and triggers a robust type I interferon response—a mechanism critically involved in pancreatic β-cell injury and adipose tissue inflammation [50]. Notably, BEVs can also deliver microbial miRNAs that mimic host miRNA functions. For example, BEV-delivered miR-122 directly suppresses the farnesoid X receptor signaling pathway in hepatocytes, disrupting bile acid and lipid metabolic homeostasis and exacerbating high-fat diet-induced hepatic steatosis [51]. Conversely, Akkermansia muciniphila-derived BEVs deliver functional proteins that enhance intestinal barrier integrity via TLR2 activation, consequently reducing circulating LPS levels and ameliorating obesity-associated metabolic endotoxemia [52].

The enteral effect of BEVs

As the primary interface for microbiota-host interactions, the gut serves not only as a major site of BEV production but also as a key target for their initial biological effects. BEVs play an important role in maintaining metabolic homeostasis by coordinately regulating the intestinal physical barrier and local immune microenvironment. Conversely, dysregulation of BEV function represents a critical factor in the pathogenesis of metabolic diseases [53].

BEVs regulate intestinal barrier function

The modulation of barrier function represents an important property of BEVs, although their effects demonstrate significant strain dependency. BEVs derived from probiotic strains generally enhance barrier integrity. For instance, Akkermansia muciniphila-derived BEVs activate the adenosine monophosphate-activated protein kinase (AMPK) pathway, significantly upregulating the expression of tight junction proteins—such as Zonula Occludens Protein 1 (ZO-1) and occludin—in Caco-2 cells. This enhancement of epithelial barrier integrity effectively prevents the translocation of microbial products [52]. Similarly, BEVs from the probiotic Escherichia coli Nissle 1917 (EcN) reinforce barrier function by upregulating junction proteins including ZO-1 and ZO-2 [54]. In contrast, certain BEVs derived from pathogenic bacteria disrupt barrier integrity by targeting tight junction complexes. For example, Helicobacter pylori-derived BEVs carry virulence factors such as VacA toxin, which directly induces degradation and redistribution of key tight junction proteins—including occludin and claudin-4—in epithelial cells, thereby increasing intestinal permeability [55]. These findings collectively reveal the potential mechanisms through which BEVs modulate intestinal barrier homeostasis via multitargeted regulation, either preserving or compromising barrier function depending on their bacterial origin.

BEVs regulate the intestinal immune microenvironment

The regulation of the immune microenvironment further reflects the strain-dependent functionality of BEVs. BEVs carry pathogen-associated molecular patterns (PAMPs) on their surface, which are recognized by TLRs on intestinal epithelial and immune cells, enabling precise modulation of immune responses [56]. For instance, BEVs derived from the pathogenic bacterium Fusobacterium nucleatum activate the NF-κB pathway via the TLR4-MyD88 axis, promoting the secretion of proinflammatory cytokines [57, 58]. In contrast, BEVs produced by probiotics such as Propionibacterium freudenreichii suppress the NF-κB pathway, exerting anti-inflammatory effects [59]. It is noteworthy that BEVs from Akkermansia muciniphila not only enhance the physical barrier but also mediate immunomodulatory functions through specific proteins such as Amuc_1100 via TLR2 signaling, collectively ameliorating metabolic endotoxemia [52]. However, current understanding of BEV-mediated immune regulation in the gut remains limited. Most existing studies rely on in vitro cell models, and these findings require further validation within the complex context of the in vivo microenvironment. Moreover, BEVs from different bacterial strains exhibit substantial differences in their immunomodulatory components and functions, and the characterization of these specific components remains inadequate, constraining our insight into their precise regulatory mechanisms. Future studies should integrate multi-omics approaches such as proteomics and lipidomics to deeply resolve the functional composition of BEVs and validate these discoveries in human cohort studies. Such efforts will establish a more solid theoretical foundation for developing BEV-based immunomodulatory strategies.

Mechanistically, the HFD not only alters gut microbial composition by enriching Gram-negative opportunistic pathogens (e.g., Proteobacteria) known for high BEV production, but also directly modulates the biogenesis and cargo sorting of BEVs. Emerging evidence suggests that HFD triggers the expansion of BEV-producing pathobionts such as Pseudomonas panacis, whose BEVs activate TLR4/CD14 signaling to induce IR and adipose inflammation [60]. Conversely, the abundance of BEVs derived from beneficial bacteria like Akkermansia muciniphila, which normally enhance barrier integrity via AMPK-ZO-1 pathways, is reduced under HFD [61]. This compositional shift, combined with HFD-induced downregulation of tight junction proteins and remodeling of gut microbial metabolites (e.g., reduced SCFAs and increased LPS), tilts the BEV pool toward a pro-inflammatory profile. These “pathogenic” BEVs—enriched with LPS, bacterial DNA, and peptidoglycan—enter the systemic circulation via paracellular routes and directly activate adipose tissue TLR4/NF-κB and cGAS-STING signaling, establishing a feed-forward loop that perpetuates gut barrier dysfunction, systemic endotoxemia, and ultimately obesity [50, 52].

In summary, BEVs maintain intestinal homeostasis through a “dual regulatory” mechanism: they reinforce the physical barrier to prevent the translocation of harmful substances, while simultaneously modulating immune activity to maintain a controlled inflammatory tone. When this balance is disrupted—such as under high-fat diet conditions—the pro-inflammatory properties of BEVs become predominant. This leads to increased intestinal permeability, entry of inflammatory mediators such as LPS into the circulation, and ultimately contributes to systemic metabolic disturbances via the “gut-organ axis” (e.g., gut–liver axis, gut–adipose axis). These processes establish a pathological foundation for the development of diseases such as obesity and T2DM.

The mechanisms of BEVs in metabolic diseases

Building upon the biological properties of BEVs and their fundamental roles in the gut, their systemic effects extend to various metabolic disorders. This section will systematically elaborate the specific mechanisms of BEVs in obesity, T2DM, MASLD, atherosclerosis, and hypertension, based on disease pathophysiology. These discussions aim to unravel the “cross-organ communication” mediated by BEVs as key players linking the gut to distant organs.

The mechanism of BEVs in obesity

Gut microbiota-derived BEVs contribute to the initiation and progression of obesity through signaling along the “gut–adipose axis”. The central mechanism involves the translocation of BEVs and their bioactive cargo into the systemic circulation, entry into host cells via specific internalization pathways, disruption of metabolic homeostasis, direct targeting of adipose tissue, and the induction of chronic low-grade inflammation along with lipid metabolic disorders.

BEVs promote obesity-related inflammation through intestinal barrier damage

BEVs can directly impair intestinal barrier function and translocate into the bloodstream, triggering systemic inflammation [62]. In high-fat diet (HFD)-fed mice, increased production of BEVs derived from LPS-producing bacteria (e.g., Firmicutes) accompanies downregulation of tight junction proteins such as ZO-1 and occludin. These changes facilitate the uptake of BEVs and their carried inflammatory mediators (e.g., LPS) by intestinal epithelial cells via clathrin- or caveolin-mediated endocytosis, or enable their paracellular translocation into the circulation, directly contributing to metabolic endotoxemia [29, 61]. Beyond animal studies, clinical evidence also indicates altered BEV profiles in the blood of individuals with obesity, characterized by enriched Firmicutes-derived DNA, reduced Bacteroidetes-derived DNA, and a positive correlation between the LPS load in BEVs and systemic inflammation levels [63, 64].

BEVs mediate the inflammatory response of adipose tissue

BEVs that translocate into the systemic circulation can be internalized by adipose tissue macrophages and adipocytes through mechanisms such as lipid raft-mediated endocytosis [65]. The LPS carried by these BEVs activates the TLR4 receptor on adipocytes, recruiting the intracellular adaptor proteins MyD88 and TRIF, which in turn activate the NF-κB and IRF3 signaling pathways, respectively. This leads to the release of pro-inflammatory cytokines such as TNF-α and IL-6. These cytokines induce serine phosphorylation of insulin receptor substrate 1 (IRS-1), thereby inhibiting its normal tyrosine phosphorylation, disrupting insulin signaling, and exacerbating insulin resistance [66]. Furthermore, microbial DNA encapsulated within BEVs can be recognized by the cytosolic pattern recognition receptor cGAS, catalyzing the synthesis of the second messenger cGAMP and activating the STING protein. This triggers the TANK-binding kinase 1 (TBK1)–IRF3 signaling axis and induces a type I interferon response, further aggravating adipose tissue inflammation [50]. The effects of BEVs exhibit strain-specific characteristics: pathogenic BEVs deliver peptidoglycan (PGN) that can activate pro-inflammatory responses via TLR2, whereas BEVs derived from the probiotic Akkermansia muciniphila demonstrate anti-inflammatory properties [65].

BEVs regulate adipose tissue metabolic homeostasis

BEVs directly regulate core pathways of lipid metabolism through their bioactive cargo. Components such as LPS modulate fat accumulation and lipolysis by influencing short-chain fatty acid (SCFA)-mediated activation of GPR43 and GPR41 receptors [67]. Concurrently, BEV-delivered PGN directly activates adipocyte TLR2 receptors, thereby inhibiting white adipocyte browning and promoting the “whitening” of brown adipocytes, thereby reducing energy expenditure [68]. In contrast, BEVs derived from probiotics such as Akkermansia muciniphila significantly upregulate mRNA expression of PPAR-α and PPAR-γ in epididymal adipose tissue, enhancing fatty acid oxidation and energy consumption, which contributes to the amelioration of obesity [69]. BEVs from EcN have also been shown to improve glucose metabolism, reducing body weight and blood glucose levels in obese mice [70].

However, significant knowledge gaps remain in understanding how BEVs regulate adipose metabolism. Although BEVs are known to influence overall adipose tissue function, the molecular mechanisms by which they specifically target different adipocyte subtypes (e.g., white, brown, beige) and precisely regulate their differentiation and function remain unclear. Moreover, most current studies administer BEVs via direct injection into the circulation, bypassing the critical physiological process of intestinal barrier translocation. This approach may not fully replicate the natural pathophysiology of BEV translocation from the gut in human metabolic diseases. Therefore, future research should distinguish the respective contributions of intact BEVs versus their cargo to metabolic regulation and further investigate the efficiency and regulatory mechanisms of BEV translocation into adipose tissue under physiological and pathological conditions. Such studies will provide important insights into the comprehensive role of BEVs in energy homeostasis. Alterations in BEV composition resulting from gut microbiota dysbiosis can thereby affect systemic energy balance through these mechanisms. Together, these findings highlight BEVs as essential mediators in microbe–host crosstalk, playing a key role in the regulation of adipose tissue metabolic homeostasis through multiple targets and pathways.

The mechanism of BEVs in T2DM

Elevated levels of BEVs have been detected in the feces, serum, and urine of patients with T2DM compared to healthy individuals, strongly suggesting that gut microbiota-derived BEVs actively contribute to the disease process [71, 72]. BEVs promote the progression of T2DM through multiple mechanisms, including disruption of the intestinal barrier, direct impairment of β-cell function, interference with insulin signaling in peripheral tissues, and modulation of metabolic–immune networks [50, 73, 74].

BEVs aggravate β-cell dysfunction through the gut-pancreatic islet axis

In T2DM, the hyperglycemic environment disrupts tight junctions in intestinal epithelial cells via a GLUT2-dependent pathway, significantly increasing intestinal permeability and creating conditions for the translocation of BEVs and their bioactive cargo into the systemic circulation, where they can target pancreatic islets [73]. Clinical investigations have consistently demonstrated elevated BEV levels in the blood, feces, and urine of T2DM patients compared to healthy individuals, supporting the association between impaired intestinal barrier function and the pathological role of BEVs [72]. Translocated BEVs directly impair islet function through their cargo. Animal studies demonstrate that BEVs enter the islets through the compromised intestinal barrier, where microbial DNA carried by BEVs is recognized by cytosolic cGAS in pancreatic β-cells, triggering the cGAS–STING innate immune pathway [50]. Activation of this pathway leads to IRF3 phosphorylation and robust production of type I interferon (IFN-β), inducing islet inflammation and impairing insulin secretion from β-cells [74]. It is important to note that the effects of BEVs exhibit significant strain specificity. For instance, BEVs derived from certain pathogenic bacteria can directly induce β-cell apoptosis or dysfunction through bacterial nucleic acids or inflammatory mediators such as LPS [50]. In contrast, BEVs from probiotics such as Akkermansia muciniphila may confer protection indirectly by modulating tight junction proteins and thereby improving intestinal barrier integrity [52]. Moreover, the pathogenic impact of BEVs is not limited to the pancreas. Chen PP et al. demonstrated that BEVs may translocate across the compromised intestinal vascular barrier and accumulate in the renal tubulointerstitium of diabetic rats [75], where they activate the local NLRP3 inflammasome. This finding highlights that the “gut-organ axis” translocation mechanism also contributes to the development of diabetic complications.

BEVs mediate insulin resistance and disorders of glucose metabolism

As critical mediators in T2DM pathogenesis, BEVs deliver their bioactive cargo to peripheral tissues such as the liver and muscle, directly interfering with insulin signaling and glucose metabolism. For example, BEVs derived from Pseudomonas panacis activate the TLR4/CD14 pathway via their carried LPS, inducing insulin resistance and resulting in impaired insulin signal transduction and disrupted GLUT4 translocation in myotubes [60]. TLR activation leads to the activation of IKKβ and JNK. These kinases promote serine phosphorylation of IRS-1, thereby interrupting the transmission of insulin signals to downstream Akt [76]. It is noteworthy that BEVs from different bacterial strains exhibit markedly distinct effects. For instance, BEVs from pathogenic bacteria such as Porphyromonas gingivalis deliver virulence factors (e.g., gingipains) to the liver, where they degrade IRS-1 in a proteasome-dependent manner and inhibit the activation of the Akt/GSK-3β signaling pathway in hepatocytes, thereby reducing hepatic glycogen synthesis and exacerbating fasting hyperglycemia [77]. In contrast, BEVs derived from EcN modulate gut microbiota composition by enriching beneficial bacteria and SCFA-producing species, thereby improving metabolic function [70].

BEVs regulate the progression of T2DM through metabolism-immune interaction

BEVs play a significant role in the systemic progression of T2DM by modulating gut microbiota composition and host immunometabolic signaling. As demonstrated by Shi J et al., Escherichia coli-derived BEVs significantly improve gut microbiota balance by reducing the Firmicutes/Bacteroidetes ratio while enriching beneficial taxa such as Lachnospiraceae and Oscillospira -microbial restructuring strongly correlated with metabolic improvement [70]. Moreover, BEVs indirectly regulate metabolic inflammation by influencing host gene expression programs. Studies suggest that BEV-mediated abnormal recruitment of inflammatory cell and metabolic dysfunction may be associated with the modulation of circadian clock gene expression, providing novel insights into the systemic dysregulation characteristic of T2DM [78]. Particularly noteworthy is the unique therapeutic potential of Akkermansia muciniphila-derived BEVs, which not only enhance intestinal barrier integrity via the AMPK pathway but also modulate systemic inflammatory status, thereby breaking the vicious cycle of insulin resistance [52]. However, despite these mechanistic advances, the causal role of BEVs in human T2DM pathogenesis requires further validation. Most current evidence originates from cellular or animal models, while direct evidence demonstrating BEV functionality in human tissues remains relatively limited. A key unresolved question is whether elevated circulating BEV levels are a cause of T2DM pathogenesis, a consequence of diabetic gut microenvironment dysbiosis, or part of a vicious cycle between the two. Addressing this will require future studies to integrate advanced technologies such as single-cell sequencing and spatial transcriptomics with organoid co-culture models to further elucidate the cell-specific mechanisms of BEVs in human T2DM. Additionally, exploring whether BEVs indirectly contribute to T2DM pathogenesis by regulating gut hormone (e.g., GLP-1) secretion or influencing gut-brain axis signaling represents a promising research direction.

The mechanism of BEVs in atherosclerosis

Atherosclerosis is a chronic inflammatory disease of the arterial wall characterized by the accumulation of lipid-rich plaques [79]. Its pathophysiology involves endothelial dysfunction, infiltration of macrophages and oxidized low-density lipoprotein (oxLDL), foam cell formation, smooth muscle cell proliferation, and ultimately plaque rupture leading to thrombotic events such as myocardial infarction and stroke [80]. Patients with atherosclerosis exhibit significantly elevated levels of bacterial DNA within circulating BEVs, suggesting their active involvement in disease progression. BEVs contribute to the multifaceted acceleration of atherosclerosis by damaging vascular endothelium, driving plaque formation, and promoting thrombotic events.

BEVs induce vascular endothelial injury and lipid metabolism disorders

BEVs directly impair vascular endothelial function. For example, BEVs derived from Porphyromonas gingivalis significantly enhance the responsiveness of aortic endothelial cells to inflammatory stimuli. Pre-exposure of endothelial cells to these BEVs upregulates their own TLR expression through TLR2/4 signaling, leading to exaggerated secretion of monocyte chemoattractant protein-1(MCP-1) in response to subsequent stimulation by biomolecules such as LPS [81, 82]. As a key chemoattractant for monocytes/macrophages, the overexpression of MCP-1 directly promotes the infiltration of macrophages into the vascular wall, which is a hallmark feature of early atherogenesis development [83]. It is noteworthy that BEVs from Escherichia coli exhibit stronger pro-inflammatory effects than free LPS and cytolysin A protein. They more effectively inducing intercellular adhesion molecule 1 (ICAM-1) expression and leukocyte adhesion while promoting interleukin 8 (IL-8)/C-X-C motif chemokine ligand 1 (CXCL1) release via the TLR4/NF-κB pathway, thereby recruiting neutrophils and exacerbating vascular inflammation [84]. Collectively, these changes contribute to vascular endothelial dysfunction and disrupted lipid metabolism, establishing a foundation for atherosclerotic development.

BEVs drive the imbalance between plaque development and stability

BEVs directly participate in the formation and progression of atherosclerotic plaques by modulating key cellular behaviors within the plaque microenvironment. Porphyromonas gingivalis-derived BEVs can precisely deliver virulence factors such as LPS to the arterial wall, where they stimulate macrophages to phagocytose oxidized LDL and suppress cholesterol efflux through downregulation of ABCA1/G1 expression, thereby promoting foam cells formation—a characteristic pathological feature of atherosclerotic plaque development [85]. In a seminal study utilizing ApoE-/-murine models, Wang N’s team demonstrated that treatment with BEVs from cytotoxin-associated gene A (CagA)-positive Helicobacter pylori significantly elevated cholesterol levels, increased arterial apoptosis, and accelerated atherosclerotic plaque formation. This process may be associated with activation of the ROS/NF-κB signaling pathway and subsequent release of IL-6 and TNF-α [86].

BEVs promote thrombosis and systemic inflammation

During the terminal stage of the disease, BEVs directly contribute to acute clinical events. Porphyromonas gingivalis-derived BEVs exhibit potent platelet-activating capacity and can directly induce platelet aggregation via TLR2/4 signaling, contributing to acute thrombotic events [87]. Simultaneously, proinflammatory cytokines (such as IL-6, TNF-α) stimulated by BEVs not only exacerbate local vascular inflammation but also trigger systemic inflammatory responses through the circulatory system. This “local-systemic” inflammatory network ultimately accelerates the onset and progression of atherosclerotic cardiovascular events [88]. It is particularly important to note that current research on BEVs in atherosclerosis faces a critical methodological challenge: how to distinguish the respective contributions of orally derived versus intestinally derived BEVs. Although the role of BEVs from oral pathogens like P. gingivalis is relatively well-established, this may partially obscure the independent effects of gut-derived BEVs. Future studies should employ more precise source-tracking techniques (e.g., strain-specific labeling) and epidemiological analyses (incorporating both oral and intestinal health) to elucidate the relative contributions of BEVs from different origins. Furthermore, the exact mechanisms of interaction between BEVs and components of the vascular microenvironment—such as host lipoproteins and extracellular matrix—require deeper investigation. These studies will provide critical insights into the comprehensive understanding of the role of BEVs in atherosclerotic thrombosis.

The mechanism of BEVs in MASLD

Gut dysbiosis-derived BEVs translocate in substantial quantities to the liver via the portal circulation, where they play an important role in the initiation and progression of MASLD, as well as its transition to hepatitis and fibrosis [89]. The underlying mechanism involves a continuous pathological sequence, starting with hepatocyte metabolic dysregulation, progressing to systemic inflammation, and ultimately leading to structural tissue alterations [90].

BEVs mediate the imbalance of lipid metabolism in hepatocytes

BEVs directly target hepatocytes and disrupt lipid metabolic homeostasis through multiple signaling pathways, serving as the initiating factor in driving hepatic steatosis. Jin M et al. discovered through subcutaneous infusion of purified PGN that PGN can be recognized by NOD2, a pattern recognition receptor, within hepatocytes. This interaction activates the NOD2-RIPK2-NF-κB signaling axis, upregulating the expression of key lipogenic genes such as PPARγ and SREBP1. This process acts as a “switch” for hepatic lipid synthesis, leading to substantial accumulation of triglycerides in hepatocytes [91]. Concurrently, the LPS enriched in Gram-negative bacterial BEVs activates intracellular Ca²⁺ signaling and the JNK/p38 MAPK inflammatory cascade via the TLR4/CD14 complex in hepatocytes. Sustained activation of this pathway indirectly suppresses mitochondrial β-oxidation and reduces fatty acid breakdown, further exacerbating lipid accumulation in hepatocytes [89, 92]. The significant attenuation of hepatic steatosis observed in LBP (LPS-binding protein) knockout mouse models provides reverse genetic evidence supporting the critical role of the BEV-mediated LPS–LBP–CD14 axis in regulating hepatocellular lipid metabolism.

BEVs exacerbates metabolic disorders and insulin resistance

BEVs exert systemic effects beyond the liver, amplifying metabolic damage through complex network interactions. The absence of CD14-a critical receptor for the LPS/LBP complex—not only alleviates hepatic steatosis but also markedly improves systemic glucose homeostasis in diet-induced obese mice, underscoring the central role of BEVs in systemic metabolic dysregulation [93]. This protective effect extends to the nervous system; CD14-deficient mice exhibit reduced sympathetic nerve activity and improved blood pressure, suggesting that BEVs may exert remote effects via neuro-immune-metabolic networks such as the “gut-liver-brain axis” [94]. The dual role of BEVs is clearly demonstrated in these mechanisms: whereas BEVs derived from pathogenic bacteria exacerbate metabolic disorders, those from the probiotic Akkermansia muciniphila significantly attenuate hepatic fibrosis and inflammatory markers by enhancing gut barrier function, modulating microbiota composition, and exerting direct anti-inflammatory effects. These observations provide direct evidence supporting the potential of BEV-based therapeutic strategies [95].

BEVs drive the processes of liver inflammation and fibrosis

BEVs drive the progression from simple hepatic steatosis to inflammation and fibrosis by directly activating resident immune cells in the liver. TLR4 activation by BEVs not only induces the production of pro-inflammatory cytokines in liver sinusoidal endothelial cells but, more critically, promotes the activation of hepatic stellate cells (HSCs), leading to their transdifferentiation into pro-fibrogenic myofibroblasts. These cells initiate collagen deposition and thereby trigger the fibrotic process [96]. Recent studies have revealed a deeper mechanism: BEVs can serve as vectors that deliver bacterial DNA directly into hepatocytes and HSCs. This foreign DNA acts as a danger signal recognized by cytosolic cGAS, activating the cGAS/STING innate immune pathway. This triggers a robust type I interferon response mediated by TBK1–IRF3 signaling, along with NF-κB-driven release of inflammatory cytokines, ultimately resulting in hepatocyte injury and sustained HSC activation [90].

It should be noted that while the “gut-liver axis” provides a relatively clear research framework for understanding the role of food-derived extracellular vesicles in MASLD, the core mechanism of action lies in the remodeling of the intestinal microenvironment by BEVs, rather than their direct translocation. Recent studies support this view. A comprehensive review by Zhang T et al. highlighted that Akkermansia muciniphila and its derived components, including extracellular vesicles, can reshape gut microbiota composition and enhance intestinal barrier integrity, thereby indirectly ameliorating metabolic dysfunction-associated liver diseases [95]. This finding directs our focus to several key scientific questions: How do BEV-induced changes in the gut microbiota specifically alter the composition of microbial products (such as LPS and other PAMPs) and host metabolites delivered to the liver via the portal vein? Is it primarily these gut-derived signals, or secondary signals generated by the interaction of BEVs with immune cells within the gut, that play the dominant role in initiating and modulating hepatic inflammation? Currently, many experimental designs struggle to distinguish the relative contributions of BEVs’ direct intestinal effects from subsequent systemic consequences. Future research requires carefully designed experiments, such as comparing the effects on hepatocytes of gut microbial metabolites from BEV-pretreated sources versus those of purified LPS or intact BEVs themselves. Furthermore, research on the specific interactions between these gut-derived signaling molecules and various hepatic non-parenchymal cells (e.g., liver sinusoidal endothelial cells and distinct macrophage subsets) remains relatively scarce. The crosstalk between these cells likely determines the intensity, duration, and final outcome of the inflammatory response. Notably, the reduction of the protective Vsig4⁺ macrophage population in the liver leads to a loss of suppression against these gut-derived inflammatory signals, thereby exacerbating inflammation and fibrosis progression [90]. Clinical intervention studies provide a compelling translational perspective for this indirect mechanism of action. Polymyxin B treatment, which targets Gram-negative bacteria (the source of LPS and BEVs), effectively reduces circulating TNF-α and LPS levels and reverses hepatic steatosis [97]. These results not only demonstrate the feasibility of targeting harmful gut-derived signals but also indirectly support the hypothesis that BEVs may exert their effects by modulating such gut microbial products [97].

Emerging evidence suggests that endogenous alcohol produced by gut microbiota may also disrupt BEV homeostasis, potentially contributing to metabolic disease pathogenesis. High-alcohol-producing bacteria, particularly Klebsiella pneumoniae and certain yeast species, generate substantial amounts of endogenous ethanol from carbohydrate fermentation—a process implicated in MAFLD progression [98, 99]. Although direct evidence linking endogenous alcohol to BEV alterations remains limited, several plausible mechanisms can be proposed: (i) Direct membrane perturbation: Ethanol intercalates into bacterial lipid bilayers, altering membrane fluidity and curvature, which could affect the budding and scission processes during BEV biogenesis, potentially leading to increased release of BEVs with altered cargo composition [38, 100, 101]. At low concentrations (< 30 mol%), ethanol increases the size of bacterial vesicles with discernible shape deformations; at higher concentrations, it compromises vesicle integrity and induces lipid-lipid aggregation [100]. (ii) Induction of oxidative stress: Ethanol metabolism generates reactive oxygen species (ROS) and acetaldehyde, which can damage bacterial DNA and proteins [102]. These damaged cellular components might be preferentially packaged into BEVs, creating “neoantigen” -enriched vesicles with heightened immunogenicity [103]. (iii) Increased gut permeability: Endogenous ethanol disrupts intestinal tight junctions via activation of the TLR4/NF-κB pathway and myosin light chain kinase (MLCK), enhancing paracellular translocation of BEVs into the portal circulation [104–106]. Chronic ethanol exposure reduces expression of tight junction proteins (e.g., ZO-1, occludin) and increases bacterial translocation in a TLR4-dependent manner [104], while parallel MLCK up-regulation and MAPK activation directly mediate ethanol-induced barrier disruption in humans [105]. (iv) Altered host immune recognition: Acetaldehyde-protein adducts could modify BEV surface proteins, potentially altering their recognition by host pattern recognition receptors (e.g., scavenger receptor A) and exacerbating inflammatory responses [107, 108]. Specifically, malondialdehyde-acetaldehyde (MAA)-adducted proteins bind directly to scavenger receptor A and stimulate PKC-dependent proinflammatory cytokine release in epithelial cells [107], and similar modifications have been shown to promote macrophage inflammatory responses [108]. Future studies specifically comparing BEV profiles in HiAlc bacterial carriers versus non-carriers, coupled with in vitro ethanol exposure experiments, are urgently needed to clarify whether endogenous alcohol contributes to metabolic disease through BEV-mediated mechanisms (see Fig. 3).

Fig. 3.

Schematic diagram illustrating the mechanistic roles of BEVs in metabolic diseases. The figure is organized into four distinct panels: (A) Obesity, (B) T2DM, (C) MASLD, and (D) Atherosclerosis. This figure summarizes the core mechanisms of BEVs in these four conditions. A Obesity: BEVsregulate metabolism via the “gut–adipose axis”. (1) Direct effects: BEVs translocating to adipose tissue (e.g., those derived from Akkermansia muciniphila) can directly activate inflammatory pathways in adipocytes through membrane-bound TLR receptors, suppressing beige adipogenesis andpromoting the release of inflammatory cytokines. (2) Indirect effects: Components carried by BEVs, such as LPS, indirectly influence the production ofgut microbiota metabolites (e.g., SCFAs), which participate in the regulation of fat accumulation and lipolysis via receptors including GPR41/43. B T2DM: BEVs affect glucose homeostasis through the “gut–islet axis” and peripheral tissues. (1) Direct effects: Circulating BEVs can be directlyinternalized by pancreatic β-cells or skeletal muscle cells. Their cargo, such as LPS, interferes with insulin signaling (e.g., by inducing IRS-1phosphorylation) via the TLR4/MyD88 pathway. (2) Indirect effects: Certain BEVs (e.g., those derived from Escherichia coli Nissle 1917) can indirectlyreshape the gut microbiota composition (e.g., by reducing the Firmicutes/Bacteroidetes ratio), thereby improving systemic metabolic conditions. C MASLD: BEVs translocate to the liver via the portal vein. (1) Direct effects: BEVs are taken up directly by liver sinusoidal endothelial cells, Kupffer cells,and HSCs. Their carried PAMPs (e.g., LPS) activate downstream inflammatory responses through TLR4/CD14 signaling, promoting hepatocyte steatosisand fibrosis. (2) Indirect effects: BEVs can exacerbate liver injury indirectly by inducing systemic inflammation. D Atherosclerosis: BEVs participate invascular inflammation and plaque formation. (1) Direct effects: Certain BEVs (e.g., those derived from Porphyromonas gingivalis) are directly internalized by endothelial cells or macrophages. Their carried virulence factors (e.g., gingipains) promote foam cell formation. (2) Indirect effects: BEVs canaccelerate the progression of atherosclerosis indirectly by activating platelets or stimulating the liver to produce acute-phase proteins

The mechanism of BEVs in hypertension

Gut microbiota dysbiosis, particularly through Gram-negative bacteria-derived BEVs (e.g., Prevotella, Klebsiella), represents an important contributing factor in hypertension development by triggering chronic low-grade inflammation and immune imbalance via their carried LPS [109]. These BEVs contribute to blood pressure dysregulation through three sequential mechanisms: impairing vascular endothelial function, promoting structural remodeling of blood vessels, and disrupting neuro-immune-metabolic homeostasis. These processes form a comprehensive pathological cascade underlying BEV-mediated hypertension.

BEVs induce vascular endothelial dysfunction

Under hypertensive conditions, gut microbiota-derived BEVs impair vascular endothelial function through multiple molecular pathways. leading to disrupted vasomotor balance. LPS carried by Gram-negative bacterial BEVs activates TLR4 receptors on vascular endothelial cells, which on one hand suppresses the activity and expression of endothelial nitric oxide synthase (eNOS), thereby reducing the production of the potent vasodilator nitric oxide (NO) [110]. On the other hand, it stimulates the oversecretion of endothelin-1 (ET-1) a strong vasoconstrictor. This dual disruption—reduced vasodilation and enhanced vasoconstriction—increases vascular tone and elevates blood pressure. Notably, these effects are strain-specific. For instance, administration of BEVs derived from Akkermansia muciniphila significantly reduces systolic blood pressure in spontaneously hypertensive rats (SHR) and decreases levels of the proinflammatory cytokines IL-1β and IL-6 levels in both thoracic aorta and serum, thereby improving endothelial function through anti-inflammatory mechanisms [111]. Within this cytokine network, IL-1β acts as an initiator of inflammatory cascades, stimulating the release of IL-6 and IL-17A. IL-6, in turn, promotes the differentiation of T cells into proinflammatory Th17 cells, which are closely associated with vascular dysfunction and hypertension [112].

BEVs drive vascular remodeling and increased stiffness

BEVs contribute to the sustained progression of hypertension and target organ damage by inducing structural alterations in the vascular wall. An abnormal elevated Firmicutes/Bacteroidetes ratio observed in hypertensive rat models leads to modifications in the properties of BEVs [113]. These altered BEVs are associated with vascular remodeling phenomena, including increased medial thickness, collagen deposition, and fragmentation of elastic fibers [114]. Notably, under obesity-related hypertensive conditions, BEVs infiltrate the adrenal glands and facilitate the accumulation of microbial DNA in the adrenal medulla. This foreign DNA activates the cGAS/STING innate immune pathway, triggering local inflammation and dysregulated catecholamine secretion. This newly identified mechanism provides a novel perspective on how BEVs promote vascular remodeling and elevated blood pressure via endocrine glands.

BEVs mediate the disorder of the neuro-immune-metabolic network

BEVs systemically regulate blood pressure homeostasis by influencing complex immune-metabolic-endocrine cross-talk. Chronic low-grade inflammation resulting from increased Gram-negative bacteria and the translocation of their BEVs serves as a central link connecting gut dysbiosis to hypertension, forming the foundation for this network-based regulation [109]. In terms of immune modulation, BEVs affect cytokine levels such as IL-6 and IL-1β, promoting the differentiation of T cells into pro-inflammatory Th17 cells while potentially reducing the population of regulatory T cells (Tregs). This disruption of immune balance exacerbates vascular inflammation and endothelial dysfunction [112]. The antihypertensive effect of Akkermansia muciniphila-derived BEVs is closely associated with their ability to restore T-cell equilibrium by increasing the Treg/Th17 ratio [111]. From metabolic and endocrine perspectives, it is noteworthy that increased BEV translocation in obesity-related hypertension may enhance sodium reabsorption in renal tubules by altering the activity of renal Na⁺/K⁺-ATPase, thereby contributing to elevated blood pressure [115].

However, the specific molecular targets through which BEVs exert these effects—such as the particular sodium channels or transporters involved—remain unclear, presenting a highly promising area for future investigation. Furthermore, although the discovery that BEVs infiltrate the adrenal glands and disrupt catecholamine secretion via the cGAS/STING pathway provides novel mechanistic insight into the “gut-adrenal axis”, the reproducibility and generalizability of these findings across different experimental models require further validation. It is important to emphasize that research on BEVs in the context of hypertension is still relatively emerging, and many mechanistic hypotheses necessitate more robust experimental support. Future studies should focus on conducting associational analyses between BEV omics profiles and clinical phenotypes in hypertensive human populations. Such efforts will help establish a solid evidence base for these proposed mechanisms and further elucidate the pivotal role of BEVs as key mediators within the “gut-vascular axis” and the “gut-adrenal axis”.

Progress in clinical applications of BEVs

Based on the aforementioned core mechanistic roles of BEVs in metabolic diseases, their potential for clinical translation has become increasingly evident [116]. BEVs not only hold promise as disease-specific biomarkers, but their inherent biological properties also render them uniquely advantageous for applications in antimicrobial therapy, drug delivery, and vaccine development. However, the transition of BEVs from basic research to clinical application faces multiple challenges, necessitating a more critical evaluation of current advances and limitations [117].

BEVs as diagnostic biomarkers

BEVs have been extensively studied as biomarkers for the monitoring and diagnosis of various diseases [118]. Researchers have discovered that unique molecular cargoes within BEVs-such as proteins, mRNA, lipids, and metabolites-isolated from biological fluids, including serum, urine, and feces, can serve as diagnostic biomarkers for different diseases [119]. Shin C et al. reported an inverse correlation between abdominal obesity risk and Proteobacteria abundance, coupled with a positive association with Firmicutes, while demonstrating that urine-derived BEVs could predict 10-year abdominal obesity risk-establishing their value as early screening tools for obesity-related complications [120]. Metagenomic analyses further revealed significantly reduced levels of Akkermansia muciniphila-derived BEVs in fecal samples from obese and T2DM patients compared to healthy controls, providing robust evidence for BEVs as diagnostic biomarkers in metabolic disorders [121]. Furthermore, studies have revealed that serum-derived BEVs from MASLD patients contain significantly higher levels of 16S rDNA compared to healthy controls, and this elevation correlates positively with the severity of liver inflammation and fibrosis. These findings suggest that circulating BEV load holds promise as a noninvasive biomarker for monitoring MASLD progression [122]. Importantly, distinguishing the unique value of BEVs as biomarkers from other microbiome-based approaches is critical, as exemplified by Yie et al. (2025), who revealed that urine BEVs exhibit stronger associations with dietary and metabolic factors than conventional stool-based microbiome profiling [123]. Compared to fecal microbiome sequencing (which provides taxonomic composition but not functional output or systemic exposure), BEVs offer a functional “snapshot” of the active microbial products that actually translocate across the gut barrier and enter the host circulation [124, 125]. Unlike measuring free circulating bacterial products (e.g., soluble LPS or cell-free DNA), which are rapidly cleared or degraded and lack source specificity, BEVs protect their cargo from degradation, enabling detection of stable and source-traceable signals [126]. Furthermore, BEV surface markers (e.g., O-antigens, lipoproteins) can be leveraged for strain-specific capture and quantification, allowing identification of the bacterial origin of pathogenic signals—a feature not possible with free LPS or DNA measurements [127, 128]. Thus, BEVs serve as integrative biomarkers that reflect both microbial dysbiosis and host–microbe interface permeability, offering advantages in specificity, stability, and functional relevance. However, most current research remains at the discovery stage, and clinical translation faces considerable challenges. The most significant bottleneck is the lack of standardized methods for the isolation and characterization of BEVs. Different extraction techniques yield heterogeneous populations of vesicles, leading to poor comparability across studies. Future efforts are urgently needed to validate the specificity, sensitivity, and clinical relevance of these BEV-based biomarkers through large-scale clinical cohorts, particularly in relation to disease staging and prognostic outcomes.

BEVs as antimicrobial agents

Antibiotics are the most important class of antimicrobial drugs, second only to vaccines. The inherent antimicrobial properties of BEVs, demonstrated in interspecies competition, have sparked interest in exploring natural BEVs as potential antimicrobial agents [20]. OMVs isolated from Cystobacter velatus Cbv34, a representative strain of the Myxococcota phylum, have been confirmed to suppress the growth of Escherichia coli at levels comparable to those achieved by gentamicin, a clinically established antibiotic [129]. A study developed an anti-adhesion nanodrug platform using Helicobacter pylori, known as bacteria-mimicking nanoparticles (OM-NPs). OM-NPs demonstrated the ability to bind to gastric epithelial cells (AGS cells) and reduce Helicobacter pylori adhesion in AGS cell cultures [130]. This anti-adhesion effect was achieved by coating the nanoparticles with membrane molecules from BEVs, mimicking bacterial properties. Nevertheless, safety remains a primary concern in this field of application. BEVs derived from pathogenic bacteria may carry virulence factors, and their strong immunogenicity has the potential to trigger excessive inflammatory responses. Future strategies should therefore prioritize the use of probiotics or engineered non-pathogenic bacterial strains as platforms for BEV production, thereby mitigating these risks at the source.

BEVs as platforms for drug delivery

BEVs represent a highly promising natural drug delivery system due to their inherent biocompatibility, low immunogenicity, and intrinsic targeting capabilities [131]. Their lipid bilayer structure enables efficient encapsulation of both hydrophilic and hydrophobic drugs while protecting bioactive molecules from degradation during circulation. For instance, studies have successfully loaded small interfering RNA (siRNA) into engineered E. coli OMVs, achieving effective delivery to tumor cells and subsequent gene silencing. This provides proof-of-concept for utilizing BEVs to deliver nucleic acid therapeutics for metabolic diseases [132]. Furthermore, BEVs derived from probiotic strains are widely explored owing to their favorable safety profile. For example, Lactobacillus-derived BEVs have been shown to alleviate intestinal inflammation by modulating the gut microbiota and restoring the mucus barrier [133]. These findings directly demonstrate the applicability of natural BEVs as therapeutic carriers. In parallel, the unique properties of BEVs have inspired the development of novel bioinspired delivery systems. For instance, Zhang et al. created biomimetic hybrid nanovesicles (MLP-NVs) by fusing M2 macrophage membranes with lipopeptides. In an ApoE^−/− mouse model, this system exhibited excellent targeting ability, accumulating preferentially in atherosclerotic plaques and significantly attenuating inflammation and plaque progression [134]. However, the application of BEVs in treating metabolic diseases still faces significant challenges related to delivery efficiency and targeting specificity. Systemically administered BEVs are rapidly cleared by the mononuclear phagocyte system in the liver and spleen, limiting their accumulation in specific metabolic organs. Overcoming this bottleneck requires functional modification of BEV membranes—such as conjugating specific ligands—to enable precise programming of tissue-targeting capabilities [16].

BEVs as nano-vaccines

As natural immunomodulators, BEVs carry abundant PAMPs and exhibit significant similarity to their parent bacteria in immunoregulatory components, including lipoproteins, lipids, peptidoglycan, and genetic material. The unique dual properties of BEVs-being both antigenic and adjuvant-position them as promising vaccine candidates [117]. Nakao R et al. demonstrated that a Porphyromonas gingivalis BEV-based vaccine significantly reduces malondialdehyde-modified low-density lipoprotein (MDA-LDL, the most atherogenic form of oxidized LDL) levels and arterial wall thickness in rats, highlighting the potential of BEV immunization for preventing pathogen-accelerated atherosclerotic progression [135] (see Fig. 4). In this vaccine platform, BEVs derived from P. gingivalis serve as both the immunogen and the adjuvant. The BEVs carry bacterial surface antigens—including outer membrane proteins, LPS, and virulence factors such as gingipains—which are naturally enriched in the vesicle structure. Intranasal immunization with these BEVs elicits production of P. gingivalis-specific antibodies in both serum (IgG) and saliva (sIgA), which can neutralize the pathogen’s proteolytic activity and reduce bacterial viability [135]. It is important to clarify that this vaccine does not target LDL or MDA-LDL directly; rather, by generating immunity against P. gingivalis, it prevents the pathogen-accelerated inflammatory processes that promote LDL oxidation and atherosclerosis [135]. Thus, the observed reduction in MDA-LDL is an indirect consequence of pathogen-specific immune protection.

Fig. 4.

The medical potential of BEVs in drug delivery, vaccine platforms, antimicrobial agents, and biomarkers. Created using BioRender.com

Beyond infectious diseases, BEVs have also shown considerable promise in cancer immunotherapy. As illustrated in Fig. 4, bioengineered OMVs derived from a low-endotoxicity mutant E. coli strain have been developed as cell-specific drug-delivery vehicles for cancer therapy [117]. These engineered OMVs were surface-decorated with an affibody targeting human epidermal growth factor receptor 2 (HER2), enabling specific recognition of HER2-overexpressing breast cancer cells, and encapsulated with small interfering RNA (siRNA) targeting kinesin spindle protein (KSP). Systemic administration of these siRNA-loaded OMVs achieved targeted gene silencing and induced significant tumor growth regression in a xenograft mouse model, without observable nonspecific side effects [117]. This proof-of-concept study highlights the versatility of BEVs as a platform not only for prophylactic vaccines against infectious pathogens but also for therapeutic cancer vaccines and targeted drug delivery systems.

Translational potential and challenges

Despite the broad clinical prospects of BEVs, their translation into clinical applications faces a series of urgent challenges. The primary obstacle lies in the lack of standardized and scalable production processes. BEVs produced under different bacterial strains and culture conditions exhibit significant heterogeneity in size, cargo, and functionality [136], which not only hinders the comparison of results across studies but also poses substantial difficulties for quality control and large-scale manufacturing. Future research must urgently establish standardized production systems based on Good Manufacturing Practice (GMP) guidelines, coupled with multi-omics technologies for more precise characterization of BEVs.

Safety and immunogenicity represent another critical challenge requiring careful evaluation. While the endogenous origin of BEVs confers good biocompatibility, their systemic application in patients with metabolic diseases demands thorough assessment of safety and immune responses. BEVs derived from Gram-negative bacteria carry LPS, which may activate the TLR4 pathway and trigger strong immune reactions or even sepsis-like responses, potentially exacerbating the pre-existing chronic low-grade inflammatory state in patients [137]. Although certain probiotic-derived BEVs (e.g., those from Akkermansia muciniphila) have demonstrated favorable safety profiles in preclinical studies, the potential long-term effects of their use remain unknown.

Furthermore, delivery efficiency and targeting specificity remain crucial bottlenecks for achieving therapeutic benefits. Although BEVs possess inherent targeting capabilities, they are rapidly cleared by the mononuclear phagocyte system—particularly in the liver and spleen—after systemic administration, resulting in limited accumulation in specific metabolic organs. Precisely controlling their in vivo distribution and enhancing delivery efficiency to specific diseased cells are key to translating their therapeutic potential into clinical outcomes [138].

Moving forward, advancing the clinical translation of BEVs will require close interdisciplinary collaboration among microbiology, nanomedicine, bioengineering, and clinical medicine. By addressing these critical challenges, BEVs are poised to evolve from a powerful tool in basic research into a revolutionary diagnostic and therapeutic modality for patients with metabolic diseases.

Conclusion

This review systematically elucidates the key mechanistic roles of BEVs in various metabolic diseases and their clinical translational potential. A central concept emerging from this analysis is that BEVs are not merely passive carriers of bacterial products but rather sophisticated delivery vehicles that protect their cargo, enable synergistic co-delivery, and—most importantly—grant access to intracellular compartments (e.g., the cytosol) that free bacterial molecules cannot reach. BEVs, carrying a diverse array of bioactive molecules, play a dual role in the pathogenesis of metabolic disorders—acting both as drivers of disease progression and as promising therapeutic agents.

The primary challenges in current BEV research lie in methodological limitations, particularly the lack of standardized methods for the strain-specific isolation and characterization of BEVs, which hinders direct comparison of results across different laboratories. Specifically, for isolation, there is no consensus on whether to use ultracentrifugation, size-exclusion chromatography, tangential flow filtration, or precipitation-based kits—each method yields distinct vesicle subpopulations with varying purity, yield, and integrity [139]. For characterization, standardized protocols are missing for quantifying particle size and concentration (e.g., nanoparticle tracking analysis vs. tunable resistive pulse sensing vs. cryo-electron microscopy), as well as for assessing cargo composition (proteomics, lipidomics, RNA-seq) [140]. For quality control, there are no established benchmarks for acceptable endotoxin levels, protein-to-lipid ratio, batch-to-batch consistency, or sterility testing specifically tailored for BEV-based products [141, 142]. Moreover, the mechanisms through which bacterial growth conditions influence the molecular composition and activity of BEVs remain unclear, posing significant obstacles to quality control. From a mechanistic perspective, although studies have demonstrated that BEVs can translocate across the gut barrier, their translocation efficiency and functional pathways under physiological and pathological conditions require further elucidation. Most importantly, the clinical translation of BEVs faces considerable bottlenecks: standardized manufacturing processes and quality control systems have yet to be established, and the majority of current research remains at the basic stage, lacking robust clinical data and ongoing clinical trials to substantiate their applicability [143].

Looking ahead, breakthroughs are needed across multiple dimensions in BEV research. Technologically, there is an urgent need to develop advanced analytical methods to decipher BEV heterogeneity and to establish standardized protocols for isolation, characterization, and quality control. Mechanistic investigations should focus on elucidating the interactions between BEVs and host cells, particularly the processes governing cargo delivery and downstream molecular pathways [144]. Genetic engineering technologies could be leveraged to construct engineered BEVs for precision drug delivery. On the clinical front, priority should be given to conducting human studies to validate the sensitivity and specificity of BEVs as disease biomarkers, and to advancing BEV-based therapeutic strategies from preclinical research into clinical trials.

As our understanding of BEV mechanisms deepens and technological innovations continue to emerge, BEVs are expected to transition from basic research to clinical applications, offering novel strategies for the prevention, diagnosis, and treatment of metabolic diseases. However, achieving this goal will require close collaboration across multiple disciplines—including microbiology, nanomedicine, bioengineering, and clinical medicine—to jointly advance the field and ultimately realize the promise of BEV-based precision medicine in metabolic disorders.

Author’s contributions

Conceptualization: Kaile Ma, Qiqi Zhang, Zishan Jin. Writing-original draft: Kaile Ma, Qiqi Zhang, Zishan Jin. Figure preparation: Kaile Ma, Qiqi Zhang. Writing-review and editing: Kaile Ma, Qiqi Zhang, Zishan Jin, Rui Hao, Xiao Sun, Lijuan Zhou, Min Li. Funding acquisition: Min Li. All authors have read and approved the article.

Funding

This study was supported by the Clinical Research Center for Metabolic Diseases (Integrated Traditional Chinese and Western Medicine) of Chongqing (No. SQMS2026DXB-002); the NATCM Initiative for Strengthening TCM Evidence‑Based Research, NATCM-STER; the Scientific and technological innovation project of China Academy of Chinese Medical Sciences, CI2021A01606; and the 'Guidelines for Integrated Traditional Chinese and Western Medicine Treatment of Type 2 Diabetes' issued by the National Administration of Traditional Chinese Medicine (Project No. ZYZB-2022–798).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.