Bacterial Extracellular Vesicles in Aging: Mechanisms and Therapeutic Prospects

Introduction

In recent years, the rising prevalence of an aging population has led to a corresponding increase in aging-related diseases—such as Alzheimer’s disease (AD),¹ Parkinson’s disease (PD),² metabolic-associated steatotic liver disease (MASLD),³ and osteoporosis⁴—and has resulted in various socio-economic challenges.⁵ Consequently, it is crucial to investigate the impact of aging on these diseases and to develop specific strategies for their prevention and treatment.

The process of aging is intricate. It also involves a variety of biological changes, both pathological and physiological. Recent studies have revealed a significant link between the decline of beneficial microbes in the human gut and the aging process.^6,7 The gut microbiome is a diverse ecosystem that includes bacteria, fungi, viruses, archaea, and parasites. Its total cell count surpassing that of the human host.⁸ This microbiome is vital for maintaining the body’s internal balance. A well-functioning gut microbiome is crucial for providing energy, aiding in metabolic processes, supporting intestinal cell health, and defending against infections. But, an imbalance in gut microbiota may cause changes in its composition and functioning. It may result in a range of diseases associated with aging, such as obesity, type 2 diabetes (T2D), neurodegenerative conditions, cardiovascular diseases, and cancer.^9,10

While interventions such as fecal microbiota transplantation (FMT) and probiotics have shown demonstrated potential in ameliorating aging-related metabolic disorders by restoring gut microbial balance.^11,12 However, their precise molecular mechanisms remain unclear. Emerging evidence indicates that bacterial extracellular vesicles (bEVs)—nanometer-sized, membrane-bound particles released by gut microbes^13,14—may serve as the critical effectors mediating these benefits. Unlike host-derived vesicles, bEVs exhibit unique stability and can traverse biological barriers to deliver bioactive cargo (eg., proteins, lipids, nucleic acids) directly to distant host tissues. Nevertheless, the role of bEVs in aging is far from straightforward; they function as a double-edged sword, capable of either maintaining homeostasis or driving pathology, depending on the host’s physiological context.

Despite this potential, current understanding of bEVs in aging remains obscured by fragmented data and disease-specific silos. Most studies treat bEVs as static biomarkers or isolated mediators of single conditions, often overlooking the dynamic interplay between age-related physiological decline (eg., barrier leakage, immune senescence) and bEV pathogenicity. Specifically, three critical gaps persist: (1) a lack of integrated insight into how aging synergistically drives systemic bEV accumulation; (2) insufficient evaluation of the context-dependent nature of bEVs, where even probiotic-derived vesicles may exert deleterious effects in compromised hosts;¹⁵ and (3) the absence of a cohesive translational roadmap addressing methodological heterogeneity and safety concerns.

Bridging the gap between mechanistic understanding and clinical translation will likely require converging insights from adjacent fields such as biomaterials, nanotechnology, and tissue engineering. Recent advancements offer promising blueprints for overcoming the specific challenges of targeted delivery and microenvironment modulation inherent to bEV-based therapies. For instance, bioinspired gradient scaffolds that mimic native tissue heterogeneity could be adapted to create niche environments that enhance the clearance of pathogenic bEVs or support tissue recovery in aged organs.¹⁶ Similarly, the integration of immunomodulatory nanomaterials, such as hydrogenated-silicon nanosheets capable of scavenging reactive oxygen species (ROS), provides a strategy to neutralize pro-inflammatory effects while promoting regeneration.¹⁷ In the context of acute kidney injury and other systemic conditions, self-propelled nanomotors and light-driven nanozymes demonstrate the potential for robust, multifunctional carriers that can intercept bEV-mediated damage or deliver cargo to restore function in aged clearance cells.^18,19 Furthermore, the success of combination therapy platforms, such as cyanine-nanoparticle systems with high biocompatibility for synergistic treatment,²⁰ and novel nanoreactors like tungsten carbide-derived polyoxometalates that amplify therapeutic efficacy via thermo-oxidative coupling,²¹ reinforces this direction. Finally, the translation of stem cell potentials for neurodegenerative diseases²² and the use of intestinal stem cell-derived organoids for disease modeling²³ provide critical complementary tools. These technological innovations underscore the necessity of integrating engineered solutions with biological mechanisms to develop precise interventions for age-related pathologies.

To bridge these gaps, this review synthesizes disparate findings into a coherent, mechanism-driven narrative. Our literature search was conducted primarily using the PubMed database, covering publications from inception to January 2026. We utilized a comprehensive combination of keywords including “bacterial extracellular vesicles”, “gut microbiota”, “aging”, and specific age-related conditions (eg., AD, osteoporosis), along with relevant MeSH terms to ensure broad coverage. We explicitly detail how the convergence of increased intestinal permeability and diminished immune clearance creates a permissive environment for bEV retention, thereby accelerating multi-organ decline. By systematically dissecting the gut–brain, gut–metabolic, gut–cardiovascular, and gut–bone axes, we aim to resolve apparent contradictions in current literature and identify specific cargo molecules responsible for inflammaging. Critically, we move beyond cataloging associations to evaluate the evidence for causality, distinguishing drivers of pathology from bystander effects. Building on this mechanistic foundation, we propose a strategic framework for advancing bEV-based medicine. This includes outlining essential steps for standardizing isolation and quantification protocols, evaluating immunogenicity risks, and integrating multi-omics approaches with artificial intelligence to identify novel targets. Finally, we define key priorities for future clinical trials, offering a clear pathway to transform bEVs from experimental observations into viable, safe, and effective therapeutics for delaying aging and treating age-related diseases.

Gut Microbiota Dynamics in Aging: Metabolic and Immune Interfaces

Aging is characterized by a significant decline in gut microbiota diversity and stability, driven by physiological changes and modifiable lifestyle factors. While specific dietary patterns like the green Mediterranean diet can sustain diversity,²⁴ the prevalence of Western-style diets and age-related physiological shifts often lead to a marked reduction in microbial richness, particularly after age 65.^25,26 This loss of diversity is not merely compositional; it is functionally linked to weakened immune responses and metabolic dysregulation,²⁷ creating a susceptibility to conditions such as obesity, T2D, and inflammatory disorders.²⁸ Critically, these age-associated alterations foster a gut microenvironment prone to chronic inflammation and compromised barrier function, as summarized in Figure 1 and Table 1.

Table 1 A Summary of Age-Associated Intestinal Changes, Consequent Shifts in Gut Microbiota, and Underlying Mechanisms

Figure 1 Multiple factors contribute to gut dysbiosis in older adults. Aging disrupts gut microbiota homeostasis through four key drivers: (a) Sedentary lifestyle and dietary changes: Reduced physical activity and low fiber intake alter microbial composition and decrease short-chain fatty acid (SCFA) production. (b) Institutionalization: Long-term care residence correlates with standardized diets, limited environmental exposure, and increased antibiotic use, leading to reduced diversity. (c) Accumulated intestinal inflammation: Chronic “inflammaging” creates an oxidative environment that favors pathobiont expansion over beneficial taxa. (d) Intestinal mucosal thinning: Age-related barrier atrophy facilitates bacterial translocation and exacerbates local inflammation. Created with Adobe Illustrator.

The functional collapse of the aging microbiome directly undermines host metabolic and immune homeostasis. The depletion of beneficial taxa reduces the production of short-chain fatty acids (SCFAs), which are essential for maintaining intestinal epithelial cell integrity and regulating anti-inflammatory pathways.^37–39 Concurrently, dysregulated tryptophan metabolism shifts the balance from protective aryl hydrocarbon receptor ligands toward pro-inflammatory kynurenine pathways, impairing the development of regulatory T cells and promoting systemic immune activation.^35,40–44 Furthermore, the diminished synthesis of immunomodulatory metabolites like butyrate and propionate weakens the induction of transforming growth factor-β (TGF-β) and IL-10, exacerbating the risk of autoimmune and inflammatory conditions.^45–48 Collectively, this convergence of metabolic insufficiency and immune senescence results in increased intestinal permeability and a reduced capacity to clear microbial components.

This compromised gut interface fundamentally alters the behavior and impact of bEVs. In a healthy host, bEVs function locally to support homeostasis. However, in the aged gut—defined by the diversity loss, metabolic decline, and barrier dysfunction described above—the restrictive control over microbial translocation is lost. Consequently, bEVs gain enhanced access to systemic circulation while the aging immune system exhibits a diminished capacity for their clearance. Thus, the age-related deterioration of the gut ecosystem transforms bEVs from local symbiotic mediators into potential drivers of distal organ dysfunction and systemic inflammaging. This mechanistic shift, linking specific microbial declines to vesicle-mediated pathology, is the focus of the following section.

Characteristics and Sources of EVs

Definition and Classification of EVs

Extracellular vesicles (EVs) encompass all membrane-bound structures released from cells.^49,50 Recent studies indicate that both eukaryotic and prokaryotic cells have the ability to produce EVs, a trait that has been preserved throughout evolutionary history.^51,52 Host cell-derived EVs can be divided into two main categories: exosomes and microvesicles.⁵³ Exosomes, ranging from 30 to 150 nanometers in size, are typically present in several bodily fluids such as blood, urine, and saliva. These vesicles arise from the endosomal pathway in cells and are secreted when multivesicular bodies merge with the membrane of the cell. Exosomes can be further classified into several types, such as those derived from tumors, immune cells, and stem cells, each playing unique roles in cell communication and the progression of diseases.^54,55 On the other hand, microvesicles, which range from 50 to 1000 nanometers in size, are produced through the outward budding and fission of the cell membrane, making their release process more straightforward.⁵⁶ Exosomes and microvesicles serve as crucial signaling molecules that facilitate communication between cells. They carry various bioactive compounds, encompassing proteins, lipids, and nucleic acids, such as miRNA.⁵⁷ These vesicles play a significant role in the regulation of many biological processes, including cell growth, differentiation, and immune responses.^58–61 Beyond exosomes and microvesicles, the human physiological fluid landscape harbors additional EV subpopulations with distinct characteristics, further underscoring the profound heterogeneity of the EV spectrum. Apoptotic bodies represent a major subclass generated during the late stages of programmed cell death.⁶² Typically exceeding 1 μm in diameter, these large vesicles form concomitant with the fragmentation of the nucleus and organelles, which are subsequently encapsulated by the budding plasma membrane. Characterized by the extensive externalization of phosphatidylserine and the inclusion of cellular debris, apoptotic bodies primarily function to facilitate the efficient recognition and clearance of dying cells by phagocytes, thereby preserving tissue homeostasis.⁶² Another emerging subpopulation of significant interest is the large oncosome. These are exceptionally large EVs (ranging from 1 to 10 μm in diameter) shed specifically by aggressive tumor cells.⁶³ While their biogenesis is hypothesized to involve direct plasma membrane blebbing, the precise molecular mechanisms governing their formation remain incompletely elucidated. Accumulating evidence indicates that large oncosomes are enriched in various malignancies, including prostate cancer, glioblastoma, and breast cancer. Notably, their proteomic and RNA cargo profiles differ substantially from those of smaller EVs, such as exosomes, suggesting unique roles in tumor progression and intercellular communication within the cancer microenvironment.^64,65

Characteristics of bEVs

bEVs, produced by gut microorganisms like bacteria and fungi, exhibit distinct biological characteristics. Originally, bEVs were recognized as outer membrane vesicles (OMVs) because they are created by the budding and fission processes of the outer membrane in Gram-negative bacteria.^66–69 As research advanced, two additional vesicle types were identified resulting from cell lysis caused by phage infections: Outer–Inner Membrane Vesicles^70,71 and Explosive Outer Membrane Vesicles.⁷² Furthermore, investigations over the last twenty years have shown that Gram-positive bacteria can also generate EVs. The mechanism by which the EVs penetrate the thick cell wall is still debated.^73,74 Given that the membrane composition of these vesicles resembles that of the cell membrane, they are thought to originate from it and are commonly known as Cell Membrane Vesicles⁷⁵ (Figure 2).

Figure 2 Schematic illustration of the diverse biogenesis and release pathways of bacterial extracellular vesicles. The diagram depicts the known mechanisms of EV release in both Gram-negative and Gram-positive bacteria. (Left) A Gram-negative bacterium releasing outer membrane vesicles (OMVs) via direct blebbing from the cell surface. (Center) A bacterium undergoing lysis following bacteriophage infection, resulting in the release of phage particles alongside outer-inner membrane vesicles (OIMVs) and explosive outer membrane vesicles (EOMVs). (Right) A Gram-positive bacterium releasing cytoplasmic membrane vesicles (CMVs) through the thick peptidoglycan layer. Created with Adobe Illustrator.

bEVs serve two main purposes: Firstly, they contribute to the self-defense of bacteria by encapsulating harmful elements such as bacteriophages, antibiotics, reactive oxygen species, and antimicrobial peptides within their structure, allowing for their expulsion to reduce potential harm to the bacteria.⁷⁶ One more point, bEVs can act as decoys, absorbing the impact of antibiotics that target bacterial membranes.⁷⁶ Some bEVs also carry enzymes that can break down antibiotics and harbor antibiotic resistance genes, which can be shared between species through these vesicles.^77–79 Finally, let us look at this point, bEVs play a role in shaping the intestinal microenvironment. They mediate interbacterial communication and can be internalized by host cells, thereby indirectly modulating the gut microenvironment to favor the growth of their parent bacteria through alterations in host cell states.^80–83

Human vs. Bacterial EVs: A Dichotomous Comparison of Molecular Signatures and Biological Functions

The functional heterogeneity of EVs is fundamentally rooted in the origin of their parent cells. Human host-derived EVs and gut bacterial EVs—particularly OMVs from Gram-negative bacteria—exhibit intrinsic distinctions in their molecular cargo composition, biogenesis mechanisms, and subsequent modes of immune interaction. Elucidating these disparities is pivotal for deciphering the mechanisms underlying the “host-microbe” dialogue within the gut microecosystem.

Fundamental Divergences in Molecular Cargo Profiles and Membrane Architecture

The marked differences in protein, nucleic acid, and lipid compositions between human and bacterial EVs directly reflect the profound divergence in biogenesis pathways between eukaryotic and prokaryotic cells.

Regarding protein and lipid components, human EVs originate primarily from the endosomal system or via direct budding from the plasma membrane. Their membrane structure is enriched with cholesterol, sphingomyelin, and members of the tetraspanin superfamily (eg., CD9, CD63, CD81), forming stable “lipid raft” microdomains that facilitate signal transduction and membrane fusion.^84,85 Their protein cargo highly mirrors the state of the parent cell, encompassing antigen-presenting molecules, cell adhesion molecules, and specific signaling pathway proteins, all aimed at mediating precise intercellular communication.⁸⁶ In stark contrast, bacterial EVs (specifically OMVs) are formed by direct budding from the bacterial outer membrane. Their membrane skeleton is rich in lipopolysaccharides (LPS), peptidoglycan, and porins; while devoid of cholesterol, they possess potent immunogenicity.⁸⁷ The protein cargo of bacterial EVs consists mainly of virulence factors (eg., hemolysins, proteases), adhesins, and metabolic enzymes. For instance, heat shock proteins (such as GroEL) carried by Helicobacter pylori OMVs can drive host inflammation via molecular mimicry, whereas probiotic-derived EVs may carry specific anti-inflammatory enzymes to maintain microbial homeostasis.^88,89

In terms of nucleic acid cargo, the two exhibit distinctly different functional orientations. Human EVs carry a complex eukaryotic transcriptome, including mRNA, miRNA, lncRNA, and DNA fragments. Upon uptake by recipient cells, these molecules can directly reprogram gene expression profiles; for example, miR-146a-5p enriched in skeletal muscle stem cell-derived EVs inhibits adipogenic differentiation,⁹⁰ while specific mRNAs in tumor-derived EVs serve as liquid biopsy biomarkers^91,92 Conversely, bacterial EVs primarily encapsulate bacterial genomic DNA fragments, plasmid DNA, tRNA, and small non-coding RNAs. Notably, the packaging of plasmid DNA renders bacterial EVs highly efficient vectors for horizontal gene transfer, enabling the dissemination of antibiotic resistance genes or virulence factors among bacterial populations, thereby accelerating adaptive evolution.⁸⁹ Furthermore, bacterial sRNAs can be directly injected into host cells to modulate host immune responses, constituting a unique mechanism of cross-species gene regulation.⁹³

Duality in Immune Interaction Modes and Pathological Roles

Driven by the aforementioned molecular disparities, human and bacterial EVs play distinct, often opposing, biological roles within the host: the former primarily acting as endogenous homeostatic regulators, and the latter frequently serving as exogenous immune triggers or pathogenic mediators.

The core function of human EVs lies in maintaining physiological homeostasis or mediating endogenous disease progression. Under healthy conditions, host EVs promote immune tolerance and tissue repair by delivering anti-inflammatory mediators (eg., mesenchymal stem cell-derived EVs).⁸⁵ However, under pathological conditions, they can transform into drivers of disease: tumor-derived EVs fuel cancer metastasis by remodeling the microenvironment, promoting angiogenesis, and facilitating immune escape;⁹⁴ in neurodegenerative disorders, host EVs act as vehicles for the inter-neuronal propagation of toxic proteins such as α -synuclein (α -syn) and amyloid- β (Aβ), exacerbating the progression of PD and AD.^95,96 Overall, the pathological impact of human EVs largely stems from dysfunctions within the host’s own cellular machinery.

Bacterial EVs, conversely, primarily elicit host immune responses or mediate infection-related diseases through the delivery of “danger signals.” Due to the abundance of pathogen-associated molecular patterns (PAMPs) like LPS on their surface, bacterial EVs potently activate the host innate immune system, particularly through the Toll-like receptor 4 (TLR4) pathway. This “double-edged sword” can either train the immune system to maintain homeostasis or, in cases of dysbiosis, induce chronic inflammation.^14,87 Mechanistically, bacterial EVs function as “remote toxin delivery systems,” capable of transporting virulence factors to deep tissues without direct bacterial contact. They participate in biofilm formation, disrupt epithelial barriers, and can even traverse the blood-brain barrier to induce systemic diseases such as sepsis and systemic bone loss.^14,97 For example, EVs from oral pathogens can trigger bursts of pro-inflammatory cytokines leading to endocarditis, while the abnormal translocation of gut microbiota-derived EVs is closely linked to metabolic syndrome and neuroinflammation. Thus, the pathological role of bacterial EVs predominantly reflects the challenge and manipulation of the host defense system by exogenous microbes.

Current Methodological Challenges in bEV Research

While the biological significance of bEVs is increasingly recognized, the interpretation of existing literature is constrained by significant methodological heterogeneity. A primary challenge lies in distinguishing bEVs from host-derived vesicles, as their overlapping physical properties often lead to cross-contamination in standard isolation protocols, potentially skewing functional attributions.^98,99 Furthermore, the lack of consensus on isolation techniques—ranging from ultracentrifugation to microfluidics—results in the selective enrichment of distinct subpopulations, complicating direct comparisons across studies.^98–100 Coupled with the absence of universal bacterial surface markers for precise quantification, these technical limitations suggest that reported variations in bEV abundance and cargo may partly reflect methodological biases rather than purely biological differences. Therefore, the findings summarized in the following sections should be interpreted within the context of these evolving analytical frameworks.

The Role of bEVs in Aging-Related Diseases

Recent research indicates that bEVs can penetrate the human bloodstream rather than being confined to the gastrointestinal tract, leading to heightened interest among researchers regarding their effects and the substances they carry on human health. Notably, there has been a surge in findings related to bEVs and their role in triggering aging-related diseases. This review will examine several established studies and those with promising implications for clinical applications. The mechanisms by which bEVs affects aging-related diseases are summarized in Table 2.

Table 2 Gut–Organ Axis Signaling via bEVs: Cargo, and Roles in Aging-Related Diseases

Gut–Brain Axis

Recent research has emphasized the link between the gut microbiome and the central nervous system, referred to as the gut–brain axis, which significantly contributes to neurodegenerative conditions such as AD and PD.^114–116 This discovery has enhanced our understanding of the development of these neurological disorders and has created new possibilities for treatment methods. Considering the expanding research on the gut–brain axis and its crucial effects on health, we will start by examining the role of bEVs in affecting neurodegenerative illnesses. The mechanisms by which bEVs mediate gut–brain communication, including their translocation across the intestinal and blood–brain barriers and their modulation of neuroinflammation, are summarized in Figure 3.

Figure 3 Bacterial extracellular vesicles (bEVs) cross the blood-brain barrier and contribute to aging-related neurodegenerative diseases. In aging, gut dysbiosis leads to reduced short-chain fatty acid (SCFA) production and increased intestinal inflammation, compromising epithelial barrier integrity. This allows bEVs to translocate into the bloodstream or travel via the vagus nerve to the brain. Once in the central nervous system, bEVs from Paenalcaligenes hominis activate microglia through lipopolysaccharides (LPS)–Toll-like receptor 4 (TLR4)–Nuclear factor kappa-light-chain-enhancer of activated B cells (NF–κB) signaling, triggering neuroinflammation. H. pylori-derived bEVs are taken up by astrocytes and microglia, promoting amyloid-β accumulation via the C3–C3aR axis and enhancing tau phosphorylation through glycogen synthase kinase-3β activation. Furthermore, E coli bacterial extracellular vesicles deliver curli to colonic epithelial cells, upregulating death-associated protein kinase 1 and facilitating α-syn phosphorylation, which may propagate to the brain and contribute to PD pathology. Created with Adobe Illustrator.

Alzheimer’s Disease

AD, the most prevalent neurodegenerative disorder in the elderly, is characterized by progressive cognitive decline driven primarily by aging.¹ Its incidence rises exponentially after age 65, posing a severe global health burden.¹¹⁷ Pathologically, AD is defined by the accumulation of extracellular Aβ plaques and intracellular tau neurofibrillary tangles, which synergistically induce neuronal injury.^118,119 While the precise etiology remains elusive, emerging evidence suggests a critical link between AD onset and extracellular bEVs.^6,120

LPS, a unique component located in the outer membrane of Gram-negative bacteria, is often called endotoxin. OMVs develop by the membrane’s outward budding process, and almost all OMVs include LPS.¹²¹ Additionally, LPS functions as a PAMP, which human cells use to identify Gram-negative bacteria and trigger immune responses. When LPS binds to the TLR4 on immune cell surfaces, it facilitates the movement of the transcription factor Nuclear factor kappa-light-chain-enhancer of activated B cells (NF–κB) from the cytoplasm into the nucleus,^122–124 ultimately resulting in the secretion of inflammatory substances such as tumor necrosis factor–α (TNF–α), interleukin–1β, interleukin–6 (IL–6), and nitric oxide, thereby causing inflammation.¹²⁵ In older adults, the abundance of Paenalcaligenes hominis and Escherichia coli in the gastrointestinal tract is significantly higher—8.7 and 7.7 times, respectively—compared to younger individuals. Research by Lee et al demonstrated that administering EVs and LPS derived from Paenalcaligenes hominis and Escherichia coli to SPF C57BL/6J mice can lead to colitis and AD, with LPS being a critical factor. Further investigations revealed that LPS activates the NF–κB pathway, contributing to neuroinflammation.⁸³ Similarly, Helicobacter pylori is found in greater abundance in the intestines of older adults.¹²⁶ Its secreted EVs can penetrate the brain, where they are absorbed by astrocytes and microglia, enhancing their communication through the C3–C3aR pathway, which activates microglia and promotes Aβ accumulation.¹⁰¹ Furthermore, EVs from Helicobacter pylori can stimulate microglia and affect the GSK–3β pathway, significantly worsening neuroinflammation and phosphorylation of tau,^102,127 which are vital in cognitive decline associated with AD.^128,129

At present, there are two prevailing viewpoints regarding the mechanisms by which bEVs penetrate the brain from the intestine. These viewpoints are not mutually exclusive; rather, they likely represent distinct routes for bEV entry into the brain. One viewpoint posits that bEVs traverse both the intestinal epithelial barrier and the blood–brain barrier in succession to reach the brain. The intestinal epithelial barrier serves to differentiate the internal and external environments of the intestine, typically permitting only certain solutes and water to pass.^130,131 Tight junctions are the specific areas where solutes and water navigate through this barrier, known as the paracellular pathway. Claudin is a key element of this pathway, influencing the flow of substances. Additionally, Occludin and ZO–1 contribute to the stability of the paracellular pathway and support transcellular components like Claudin.¹³² Akkermansiaceae can release bEVs that enhance the expression of various tight junction proteins, such as Claudin–1, 4, 5, ZO–1/2, and Occludin, in IECs, thereby bolstering the barrier’s integrity.^133,134 However, the abundance of these bacteria significantly declines in the gastrointestinal tract of older adults.¹³⁵ Moreover, SCFAs, generated from cellulose fermentation by specific gut bacteria, are the main energy source for colon cells, with butyrate directly fueling IECs.¹³⁶ Elevating SCFA levels in the gut can also strengthen the intestinal epithelial barrier, improve colitis, and support the blood–brain barrier’s integrity by increasing claudin–5 expression.¹³⁷ The human body lacks the ability to produce the glycosidases necessary for cellulose degradation; instead, these enzymes are generated by gut microbiota and subsequently released through transmembrane transport. The bacteria that produce these glycosidases typically depend on SCFAs for energy. However, as people age, the abundance of essential cellulose-degrading families in the gut—such as Bifidobacteriaceae, Lachnospiraceae, and Clostridiaceae^133,138—as well as Akkermansiaceae, which not only breaks down cellulose but also enhances tight junction protein levels, diminishes due to the thinning of the intestinal mucosa. This reduction leads to compromised integrity of both the intestinal epithelial barrier and the blood–brain barrier. Furthermore, as noted earlier, the aging process is associated with an increase in certain pro-inflammatory bacteria like Paenalcaligenes hominis and Escherichia coli. The OMVs they produce contain LPS, which can trigger colitis and reduce claudin–5 expression in hippocampal capillaries, thereby heightening the permeability of both the intestinal epithelial barrier and the blood–brain barrier.¹⁰² Numerous studies indicate that individuals with colitis, especially during active disease phases, experience cognitive deficits.^139,140 Additionally, cognitive and emotional issues in these patients improve with colitis treatment.^141,142 The beneficial bacteria Lactobacillaceae, known for lowering IL–8 levels and reducing oxidative stress, are diminished due to the chronic inflammation present in the gut.¹⁴³

An alternative viewpoint indicates that certain bEVs access the brain through a pathway reliant on the vagus nerve. In older adults, the abundance of Paenalcaligenes hominis in the gut is 8.7 times higher than in younger populations. Notably, mice that were given Paenalcaligenes hominis or its derived EVs displayed signs of intestinal inflammation, cognitive impairment, activation of the brain’s immune response, and neuroinflammation.⁸³ Conversely, vagotomized mice treated similarly only exhibited intestinal inflammation, with minimal brain pathology. When Paenalcaligenes hominis was substituted with Escherichia coli or LPS, both vagotomized and non-vagotomized mice experienced comparable levels of intestinal and brain damage. This suggests that EVs from Paenalcaligenes hominis penetrate the brain through a vagus nerve-dependent route, while those from Escherichia coli and LPS do, consequently, independently of the vagus nerve.⁸³ Nonetheless, research on PD has shown that vagotomy can significantly reduce the entry of bEVs from orally ingested Escherichia coli into the brain.¹⁰³ The debate over which bEVs can access the brain via the vagus nerve remains unresolved, and investigating the underlying factors contributing to these differences may be crucial for understanding gut–brain axis communication in the future.

In summary, current preclinical evidence robustly supports a dual-mechanism model where LPS-rich bEVs drive neuroinflammation either by compromising barrier integrity (E. coli) or via vagus nerve signaling (P. hominis), with the latter gaining strong experimental validation from vagotomy studies. However, significant uncertainties remain: the apparent contradiction regarding E. coli’s reliance on the vagus nerve between AD and PD models suggests that entry routes may be strain-specific or context-dependent, a hypothesis yet to be resolved. Furthermore, while these mechanistic pathways are well-defined in murine systems, their direct translation to human AD pathophysiology remains largely inferential, relying primarily on correlative microbiota shifts and indirect biomarkers rather than definitive proof of bEV translocation in patient brains.

Parkinson’s Disease

PD is the second most common neurodegenerative disorder globally, characterized by motor symptoms such as resting tremors, bradykinesia, and rigidity.² The prevalence of PD increases markedly with age, particularly affecting individuals over 65.^144,145 Pathologically, the disease is defined by the intracellular aggregation of α -syn into Lewy bodies, leading to dopaminergic neuron loss in the substantia nigra.¹⁴⁶ While PD etiology involves a complex interplay of genetic susceptibility and environmental exposures, emerging research highlights a pivotal role for the gut-brain axis.^147–149 Specifically, bacterial bEVs derived from the gut microbiota have been implicated in modulating α -syn pathology and driving disease progression.^150–152

LRRK2 is a prevalent gene associated with an increased risk of PD in Asia.¹⁵³ Research conducted by Liang et al reveals that individuals with the LRRK2 mutation have a sevenfold higher abundance of Escherichia coli compared to those without the mutation who are unaffected.¹⁰³ This may be attributed to the ability of E. coli to generate distinctive curli-containing EVs. Curli is an amyloid protein found on the surface of bacteria like E. coli.¹⁵⁴ When bEVs deliver curli to the epithelial cells of the colonic mucosa, there is a notable rise in DAPK1 expression, which in turn enhances the phosphorylation of α-syn in the intestines. Furthermore, curli has been identified as a potential catalyst for the misfolding of α-syn, which may lead to increased α-syn aggregation.^104,105 The R1628P variant in the LRRK2 gene obstructs the degradation of α-syn in the gastrointestinal system, while the interaction of these factors leads to the build-up of phosphorylated α-syn within the intestines.¹⁰³ Additionally, research indicates that EVs from Escherichia coli can transport α-syn from the gut to the brain through the vagus nerve, potentially triggering symptoms akin to PD.^103,106,155

Collectively, these findings establish a compelling mechanistic link in LRRK2-mutant models, where E. coli-derived curli on bEVs acts as a specific seed for α-syn misfolding and facilitates its retrograde transport via the vagus nerve. However, it is critical to distinguish that this “gene-environment” synergy has been primarily validated in transgenic murine systems. Whether this specific curli-mediated seeding mechanism operates with similar potency in sporadic PD cases (without LRRK2 mutations) or in the complex human gut environment remains speculative. Current human evidence is largely limited to correlative increases in Enterobacteriaceae, lacking direct demonstration of curli-bearing bEVs traversing the human vagus nerve. Thus, while the pathway is experimentally robust in genetically susceptible hosts, its generalizability as a universal driver of human PD pathology awaits confirmation in diverse clinical cohorts.

Mechanistic Gaps in Tissue Tropism and Molecular Homing

While emerging evidence confirms that bEVs can translocate to distant organs such as the brain, adipose tissue, and bone, the precise molecular determinants governing their tissue tropism remain largely elusive. Unlike mammalian exosomes, where specific integrins and tetraspanins dictate organ-specific homing, the “zip codes” for bacterial vesicles are yet to be fully decoded. Current data suggest a dual-mode entry mechanism: passive translocation driven by barrier compromise (eg., LPS-mediated disruption of tight junctions) versus active, receptor-mediated transport.

A critical area for future investigation lies in the specific molecular interactions between bEV surface components and host endothelial receptors. For instance, Outer membrane protein A, a conserved porin found on many Gram-negative bEVs, is known to facilitate bacterial invasion of the blood-brain barrier (BBB) by binding to host glycoprotein 96.¹⁵⁶ However, whether bEV-associated outer membrane protein A retains this functional conformation to mediate specific transcytosis across brain microvascular endothelial cells, independent of the parent bacterium, remains an open question. Similarly, the accumulation of bEVs in adipose tissue suggests potential interactions with scavenger receptors or lipid raft domains on adipocytes, yet no specific ligand-receptor pair has been identified.¹⁵⁷

It is hypothesized that the heterogeneous surface proteome and lipid composition of bEVs—shaped by their bacterial origin and environmental stressors—create distinct “molecular signatures” that dictate their biodistribution.^158–160 Future studies employing cryo-electron microscopy to map bEV surface topography, combined with genetic knockout models of candidate bacterial surface proteins (eg., ompA-null strains) and host endothelial receptors, are urgently needed to elucidate these homing mechanisms. Until then, the organ-specific accumulation of bEVs should be interpreted as a complex interplay of passive leakage, immune cell-mediated transport, and potentially undiscovered active targeting pathways.^161–163

Gut–Metabolism Axis

The relationship between gut microbiota and metabolic regulation, referred to as the gut–metabolism axis, is vital in the context of conditions such as T2D, obesity, and MASLD. These health concerns are especially common in older adults and are linked to a heightened risk of various comorbidities, posing a significant challenge to public health.^164,165 Investigations into how gut microbiota influences metabolic health started early and have produced a wealth of findings, ranking just behind studies on the gut–brain axis. Emerging evidence points to bEVs as pivotal mediators in this axis, orchestrating a complex balance between metabolic deterioration and homeostasis through diverse molecular cargos.

Metabolic Dysfunction-Associated Fatty Liver Disease

MASLD, formerly known as non-alcoholic fatty liver disease (NAFLD), is the most prevalent chronic liver condition globally, characterized by hepatic fat accumulation accompanied by cardiometabolic risk factors.^3,166 The prevalence of MASLD rises steadily with age, and critically, the risk of progressing to severe liver fibrosis increases significantly in older populations.^167,168 While the disease burden is substantial, recent research has identified a novel mechanism driving this age-related progression: bEVs. Emerging evidence suggests that bEV-mediated signaling exacerbates hepatic inflammation and fibrosis, particularly in the context of aging.^169,170

Fizanne et al conducted a seminal study on bEVs obtained from the feces of healthy subjects, individuals with MASLD, and those suffering from metabolic dysfunction-associated steatohepatitis (MASH) Their research revealed that these bEVs, produced by gut microbiota,¹⁰³ could diminish the levels of critical tight junction proteins, specifically occludin and ZO–1, via a pathway reliant on non-muscular myosin light chain kinase (nmMLCK).^171,172 This molecular alteration directly enhances the permeability of the intestinal epithelial barrier, facilitating the translocation of microbial products. Furthermore, obesity in MASLD patients significantly increases the risk of intestinal barrier damage,^173–177 a condition often compounded by a compromised hepatic clearance system. Specifically, barrier dysfunction and systemic endotoxemia can lead to a decrease in Vsig4+ Kupffer cells, thus hindering the liver’s capacity to eliminate circulating EVs.^108,178 These Vsig4+ Kupffer cells play a vital role in safeguarding host cells from the detrimental effects of bEVs by clearing them from circulation through a complement C3-mediated opsonization mechanism.^178,179 However, studies in C57BL/6J mice have revealed that the abundance of Vsig4+ Kupffer cells is markedly diminished in aged animals¹⁸⁰ and in those subjected to a prolonged Western diet rich in sugars, fats, and refined carbohydrates.¹⁰⁸ This indicates that both aging and a high-calorie diet may synergistically contribute to a rise in circulating bEVs.

Once these harmful bEVs penetrate the compromised intestinal barrier and enter the bloodstream, their surface PAMPs, such as LPS, interact with TLR4 on macrophages, initiating a robust inflammatory response.^181–184 Additionally, some bEVs can provoke inflammation directly in hepatic sinusoidal endothelial cells, activate hepatic stellate cells (HSCs), and elevate fibrosis levels.¹⁰⁷ Beyond protein-mediated signaling, the Cyclic GMP–AMP Synthase–Stimulator of Interferon Genes (cGAS/STING) pathway is essential for the metabolic dysfunction caused by bacterial DNA cargo. Heightened STING expression in liver tissue leads to pronounced inflammation in hepatocytes and liver fibrosis.¹⁰⁷ Specific bEVs can transport bacterial DNA into hepatocytes and HSCs, activating the cGAS/STING pathway,¹⁵² which results in increased platelet-derived growth factor β and inflammatory mediators. This cascade causes hepatocyte inflammation and drives the activation and proliferation of HSCs. Such inflammation in hepatocytes can accelerate the progression of MASLD to MASH,^185–187 while the activation of HSCs into myofibroblasts and subsequent extracellular matrix deposition remains a central event in the development of liver fibrosis.^188,189 The mechanisms are summarized in Figure 4.

Figure 4 Bacterial extracellular vesicles (bEVs) evade phagocytosis by Vsig4+ Kupffer cell and promote aging-related metabolic disorders. In healthy individuals, bEVs entering the liver via the portal circulation are primarily cleared by Vsig4-expressing Kupffer cells through opsonic phagocytosis. However, in aging, the number of these functional Kupffer cells is markedly reduced, leading to impaired clearance of bEVs. Accumulated bEVs can deliver bacterial DNA to hepatocytes and hematopoietic stem cells (HSCs), activating the GMP–AMP Synthase–Stimulator of Interferon Genes (cGAS/STING) pathway. This triggers the release of pro-inflammatory cytokines and upregulates platelet-derived growth factor-β (PDGF-β), promoting hepatocyte inflammation and activation of hematopoietic stem cells, thereby exacerbating the risk of liver fibrosis. Created with Adobe Illustrator.

In conclusion, the mechanistic framework established in murine models—where aging and Western diet synergistically impair bEV clearance (via Vsig4+ Kupffer cell loss) and enhance gut permeability, driving liver inflammation via the cGAS/STING pathway—is biologically compelling. However, translating these findings to human MASLD requires caution due to profound species-specific differences in metabolism, bile acid composition, and dietary complexity. Unlike the controlled, high-fat diets used in mice, human dietary patterns are highly heterogeneous, and the impact of specific bEV cargos on human HSCs may differ significantly from murine responses. Furthermore, while correlative studies link gut dysbiosis to human liver fibrosis, direct causal evidence demonstrating that circulating bEVs drive fibrosis progression in patients—distinct from other metabolic confounders like insulin resistance or visceral adiposity—remains elusive. Thus, while the “gut-liver bEV axis” is a robust driver in experimental steatohepatitis, its precise contribution to the multifactorial pathogenesis of human MASLD warrants validation in longitudinal cohorts with detailed dietary and metabolomic profiling.

Obesity and Type 2 Diabetes

T2D has emerged as a global health crisis,¹⁹⁰ with a prevalence that rises sharply with age; notably, nearly half of all affected individuals are now aged 65 or older.^191,192 Characterized by insulin resistance and impaired beta-cell function, T2D is frequently comorbid with metabolic disorders such as MASLD and cardiovascular disease.^193–195 While traditional views focus on systemic metabolic dysfunction, recent research highlights a critical role for the gut microbiome in disease progression. Specifically, bEVs have been identified as key mediators in host-microbe crosstalk, carrying bioactive cargo that significantly influences insulin sensitivity and metabolic homeostasis.^196–198

Research conducted by Choi et al revealed that a diet high in fats significantly increased the abundance of Proteobacteria, especially Paenalcaligenes hominis, in mice. Furthermore, the bEVs generated by P. hominis are capable of traversing the intestinal barrier to access key organs like the liver, adipose tissue, and skeletal muscles. This interaction disrupts insulin signaling pathways and hinders glucose absorption in skeletal muscles, leading directly to insulin resistance and glucose intolerance.¹⁰⁹ Alongside the rise in detrimental bacteria, unhealthy eating patterns can also diminish beneficial bacteria, resulting in a decline in advantageous bEVs, which may contribute to the onset of T2D. Studies indicate that obese and type 2 diabetic C57BL/6J mice exhibit a notably lower presence of Akkermansia muciniphila compared to their healthy counterparts, and the ingestion of prebiotics can restore its abundance.¹⁹⁹ Additionally, bEVs from A. muciniphila found in the feces of healthy individuals are significantly more prevalent than those in patients with T2D.¹³³

Subsequent research concerning this bacterium has revealed that the delivery of bEVs originating from A. muciniphila to diabetic mice fed a high-fat diet can lead to an increase in the expression of the tight junction protein occludin, enhance the integrity of tight junctions, decrease weight gain, and improve glucose tolerance. Similarly, EVs derived from Escherichia coli Nissle 1917 (E. coli Nissle 1917) can aid in alleviating obesity and type 2 diabetes by reducing the Firmicutes/Bacteroidetes ratio in the gut, promoting the proliferation of SCFA-producing bacteria, and elevating SCFA concentrations in the intestines. These short-chain fatty acids are subsequently transported to the liver via the hepatic portal vein, where their heightened levels may inhibit COX2, resulting in a downregulation of the metabolism of ω–6 unsaturated fatty acids, thereby improving conditions associated with obesity and type 2 diabetes.¹¹⁰

Collectively, the studies delineated above underscore a sophisticated bidirectional role for bEVs in metabolic disorders, governed by a distinct dichotomy between pathogenic and protective vesicles. Harmful bEVs, typified by those derived from P. hominis or dysbiotic communities prevalent in aging and Western diet contexts, operate through a “leak-inflame-damage” axis. They actively degrade intestinal barrier integrity by downregulating tight junction proteins (eg., occludin, ZO-1) via nmMLCK-dependent pathways, thereby permitting the systemic translocation of pro-inflammatory cargos such as LPS and bacterial DNA. Once in circulation, these vesicles trigger innate immune receptors (TLR4, cGAS/STING) in metabolic tissues, sustaining chronic low-grade inflammation that impairs insulin signaling in muscle and adipose tissue and drives fibrogenesis in the liver. Conversely, protective bEVs, exemplified by those from A. muciniphila and E. coli Nissle 1917, function via a “seal-modulate-resolve” mechanism. They reinforce the epithelial barrier by upregulating tight junction components, effectively preventing endotoxin leakage. Moreover, they modulate the gut ecosystem to favor SCFA production, which exerts systemic anti-inflammatory effects (eg., COX2 inhibition) and directly enhances metabolic flexibility. This delicate equilibrium between detrimental and beneficial bEV populations is dynamically regulated by host factors; aging and poor diet tip the scale towards pathogenic bEV dominance, whereas prebiotic interventions and healthy lifestyles can restore the protective bEV milieu. Understanding this balance is crucial for developing targeted therapies that not only suppress harmful vesicle production but also augment the release of therapeutic bEVs.

However, a critical gap exists between these proof-of-concept findings in controlled murine models and their clinical applicability in humans. Unlike the genetically homogeneous and diet-controlled mice used in these studies, human T2D patients exhibit vast heterogeneity in genetic background, long-term dietary patterns, medication history, and baseline microbiota composition. Consequently, the therapeutic efficacy of single-strain bEV supplementation observed in mice may be diluted or altered in the complex human gut ecosystem, where colonization resistance and inter-individual variability are significant. Furthermore, while correlative data link specific bEV profiles to human insulin resistance, longitudinal evidence demonstrating that modulating bEV cargo can sustainably reverse T2D progression in patients—independent of caloric restriction or pharmacological intervention—remains speculative. Thus, while the “bEV-metabolism axis” offers a promising mechanistic target, translating these insights into robust clinical therapies requires rigorous validation in diverse human cohorts that account for the multifactorial nature of the disease.

Gut–Cardiovascular Axis

The interplay between gut microbiota and cardiovascular conditions, including atherosclerosis, is highlighted by the gut–cardiovascular axis. In older adults, cardiovascular diseases are prevalent and have a significantly high mortality rate, often linked to disorders of the metabolic system. The mechanisms underlying gut cardiovascular axis and the mechanisms underlying gut bone axis mentioned later are summarized in Figure 5.

Figure 5 Distinct cargos in Bacterial extracellular vesicles differentially modulate aging-related gut and bone diseases, exerting either protective or detrimental effects. Helicobacter pylori-derived extracellular vesicles carry lipopolysaccharide and cytotoxin-associated gene A into systemic circulation, inducing ROS production and activating Nuclear factor kappa-light-chain-enhancer of activated B cells (NF–κB) signaling in vascular endothelial cells. This promotes secretion of pro-inflammatory cytokines (e.g., IL–6, TNF–α), leading to impaired endothelial proliferation and accelerated apoptosis—key events in atherogenesis. In contrast, Lactobacillus rhamnosus GG (LGG)-derived EVs accumulate in vascular smooth muscle cells, where cargo such as ribonucleases and heat shock proteins activate the protein kinase B pathway, enhancing intracellular calcium deposition and expression of osteogenic markers, thereby contributing to vascular calcification. Moreover, LGG fosters the expansion of Akkermansia muciniphila, whose EVs translocate to bone tissue, promoting osteoblast activity and suppressing osteoclastogenesis, thus supporting bone integrity. Created with Adobe Illustrator.

Atherosclerosis

Atherosclerosis is a chronic, lipid-driven inflammatory disease characterized by endothelial dysfunction and the accumulation of arterial plaques, which can progress to calcification, rupture, and thrombosis.²⁰⁰ While traditionally recognized as an age-associated condition, its prevalence rises markedly in middle-aged and older adults globally, with distinct trajectories in men and women post-menopause.^201,202 Beyond classical risk factors, emerging research highlights the gut microbiome as a critical modulator of vascular inflammation. Specifically, bEVs have gained attention for their potential to transport pro-inflammatory cargo, thereby influencing plaque stability and disease progression.^203,204

Healthy vascular endothelial cells play a vital role in ensuring the proper functioning of vascular smooth muscle cells,²⁰⁵ thereby helping to avert atherosclerosis. In contrast, when vascular endothelial cells are compromised, they release inflammatory substances that draw monocytes to the injury site. These monocytes then absorb low-density lipoproteins, transforming into foam cells and contributing to plaque development.²⁰⁶ Recent findings have identified the presence of Helicobacter pylori DNA²⁰³ and its virulence factor Cytotoxin-associated gene A (CagA)²⁰⁴ in atherosclerotic plaques. Research by Wang et al revealed that EVs from Helicobacter pylori carry LPS and CagA into the bloodstream. This process heightens reactive oxygen species (ROS) levels in vascular endothelial cells and activates the NF–κB signaling pathway, resulting in increased inflammatory markers such as IL–6 and TNF–α. Consequently, these changes hinder the proliferation of vascular endothelial cells and hasten their apoptosis, thereby facilitating the progression of atherosclerotic plaque formation.¹¹¹

Recent studies have indicated that the probiotic Lactobacillus rhamnosus GG (LGG) may contribute to the development of atherosclerosis.¹⁵ This probiotic has gained considerable interest for its effectiveness in treating various conditions, including cancer,²⁰⁷ acute gastroenteritis,²⁰⁸ and bone density loss,²⁰⁹ as well as for its role in preventing pneumonia²¹⁰ in critically ill individuals. Nonetheless, there is a scarcity of information regarding its potential toxic effects. Individuals suffering from CKD (chronic kidney disease) are at an increased risk for atherosclerosis due to inflammatory reactions linked to immune system imbalances and disruptions in lipid and cholesterol metabolism.²¹¹ Research conducted by Wei et al revealed that CKD rats that received LGG exhibited a markedly greater extent of atherosclerosis than those that did not receive the probiotic. This phenomenon occurs because the bEVs released by LGG accumulate in vascular smooth muscle cells, where they trigger the AKT signaling pathway through components like ribonucleases and heat shock proteins, leading to elevated intracellular calcium levels, enhanced expression of bone-related proteins, and increased vascular calcification.¹⁵

Ultimately, specific YRNAs and their derivatives, known as YsRNAs, found within hEVs are crucial in the development of atherosclerosis. These small RNAs derived from YRNA can trigger the activation of caspase 3 and the NF–κB signaling pathways, leading to increased cell death and inflammatory reactions, which worsen atherosclerosis.²¹² Additionally, YRNA and YsRNA have been detected in certain bEVs, suggesting they may also contribute to atherosclerosis through comparable mechanisms.^213,214

Furthermore, this study extends its scope to the regulatory role of bioactive substances secreted by specific gut microbiota in the progression of atherosclerosis. Notably, certain commensal bacteria, such as Clostridium clusters and Roseburia species, have been shown to modulate host lipid metabolism and inflammatory responses through both their metabolic byproducts (eg., butyrate) and secreted extracellular vesicles.^215,216 While butyrate is known to act as a histone deacetylase inhibitor to suppress the NF-κB signaling pathway and mitigate vascular inflammation, recent evidence suggests that extracellular vesicles derived from beneficial bacteria (eg., Lactobacillus species) can also deliver functional cargo to host cells to regulate cholesterol efflux and immune polarization. These secreted factors are capable of traversing the intestinal mucus layer and entering the systemic circulation, thereby influencing endothelial function within the arterial wall. This finding augments our understanding of the “gut-vessel axis,” suggesting that the combined action of metabolites and vesicles from specific bacterial strains serves as a critical mediator linking gut dysbiosis to the pathogenesis of atherosclerosis.^215,217

In summary, the emerging paradigm of bEVs in atherosclerosis reveals a context-dependent duality: while pathogens like H. pylori deliver direct virulence factors (eg., CagA) to destabilize plaques, even traditionally beneficial strains like L. rhamnosus GG can exert paradoxical pro-calcific effects under specific pathological conditions such as CKD. However, translating these mechanistic insights to human cardiovascular health requires navigating significant complexity and heterogeneity. Unlike the controlled, single-strain interventions in murine models, the human vascular niche is exposed to a dynamic mixture of bEVs from hundreds of microbial species, making it difficult to isolate the net effect of any single vesicle population. Furthermore, the tipping point at which a “beneficial” bEV cargo becomes detrimental—dependent on host factors like renal function, lipid profiles, and baseline inflammation—remains undefined in clinical populations. Current evidence is largely restricted to animal models of comorbidities or in vitro systems; thus, longitudinal human studies are critically needed to determine whether circulating bEVs serve as causal drivers of plaque progression or merely as biomarkers of underlying gut dysbiosis and systemic inflammation.

Gut–Bone Axis

While osteoporosis is frequently observed in older adults, studies focusing on the link between the gut and bone health began later than investigations in other fields, and general awareness remains limited. Critically, direct evidence demonstrating a causal role for bEVs in bone homeostasis is still scarce, with most current insights derived from indirect associations or murine models. However, as understanding of the gut microbiome’s systemic impact expands, the “gut-bone axis” mediated by bEVs is increasingly recognized as a potential regulator of skeletal health.

Osteoporosis

Osteoporosis stands as the most prevalent bone disorder globally4, marked by a decrease in bone density resulting from an imbalance in bone maintenance. This maintenance, known as bone homeostasis, involves a delicate equilibrium between the activity of osteoclasts, which break down bone, and osteoblasts, which create bone matrix, all governed by intricate regulatory mechanisms.²¹⁸ When the breakdown of bone surpasses its formation, osteoporosis develops.²¹⁹ Recent research has highlighted the significant influence of gut microbiota on the development of osteoporosis,^220–223, yet the specific contribution of bEVs remains an emerging and understudied frontier.

The gut plays a significant role in bone density regulation via the immune system. Research indicates that gut microbiota can trigger chronic inflammatory bowel diseases, which in turn lead to inflammation that contributes to bone degradation.^224,225 The primary forms of inflammatory bowel disease (IBD), Crohn’s disease and ulcerative colitis, are both linked to decreased bone mineral density in the femoral neck and trabecular bone.²²⁶ A study by Peek et al suggests a potential mechanism: intestinal inflammation alters the expression of osteoclast precursors (OCPs), crucial for osteoclast differentiation. OCPs from mice with intestinal inflammation exhibited marked increases in osteoclast differentiation ex vivo, correlating with elevated pro-osteoclast factors.¹¹² Given that older adults are at increased risk for IBD and that EVs from elderly-enriched bacteria like Escherichia coli and Paenalcaligenes hominis can induce colitis in mice,⁸³ it is plausible but not yet definitively proven that bacterial EVs may indirectly contribute to osteoporosis by exacerbating gut inflammation and subsequent systemic immune activation. Direct isolation and characterization of bEVs from IBD patients and their specific effects on human bone cells remain areas requiring urgent investigation.

In contrast to indirect inflammatory pathways, emerging data suggest a potential direct role for specific beneficial bEVs. EVs from Akkermansia muciniphila have been shown to infiltrate and accumulate within bone tissue in mice, mimicking estrogenic effects by promoting bone formation and suppressing bone resorption. Administering A. muciniphila directly to ovariectomized mice reversed osteoporosis.¹¹³ This indicates a potential therapeutic avenue via FMT or targeted probiotics. Notably, A. muciniphila is prevalent in younger individuals (average relative abundance ~9.2%) but drops significantly in older adults (~0.4%).¹²⁶ This age-related decline correlates with increased osteoporosis susceptibility, suggesting that the loss of protective bEVs might be a contributing factor alongside hormonal changes. However, it must be explicitly acknowledged that these findings are primarily based on ovariectomized mouse models. Whether human osteoblasts and osteoclasts respond to bacterial vesicles with similar sensitivity, and whether bEVs can truly replicate complex steroid hormone signaling in humans, remains speculative without direct clinical evidence.

Collectively, these findings propose a tentative dual-mechanism model for bEVs in osteoporosis: pathogenic vesicles may indirectly drive bone loss via gut-mediated systemic inflammation, while beneficial vesicles (eg., from A. muciniphila) could potentially target bone tissue to exert anabolic effects. Nevertheless, the field is in its infancy. Despite the established role of the gut microbiota in bone health, research specifically focusing on bEVs as mediators of the gut-bone axis remains remarkably scarce. Current understanding is largely confined to preliminary animal models, with a critical paucity of human clinical data characterizing the biodistribution, bioavailability, and direct cellular impacts of bEVs on human skeletal tissue. Bridging these murine insights to human pathophysiology requires addressing the profound impact of age-related microbial succession, species-specific hormonal contexts, and confounding lifestyle factors. The dramatic decline of A. muciniphila observed in aging mirrors human trends, yet quantifying the causal contribution of this specific bEV loss to human bone density—distinct from the effects of menopause, lifelong nutrition, and polypharmacy—remains highly challenging. Furthermore, the “bEV-as-hormone” hypothesis, while compelling in mice, lacks validation in elderly human cohorts. Thus, while the “gut-bone axis” mediated by bEVs offers a novel theoretical paradigm for age-related bone loss, its translational potential hinges on rigorous future studies to validate these mechanisms in humans and decipher the precise interplay between declining beneficial bEVs and rising inflammatory burdens. Until such direct evidence emerges, the role of bEVs in human bone homeostasis should be interpreted with caution.

A Unified Mechanistic Model: Gut Dysbiosis Drives Harmful bEV Accumulation in Aging

Synthesizing the evidence from the gut–brain, gut–metabolic, gut–cardiovascular, and gut–bone axes, we propose a unified conceptual framework to explain how bEVs mediate multi-organ decline in aging. The core of this model is that aging-associated gut dysbiosis drives a systemic shift in the bEV landscape, tipping the balance from homeostatic regulation to pathological propagation.

The “Dysbiosis-Permeability-Clearance” Axis

In a homeostatic (young/healthy) state, the gut microbiota is dominated by beneficial taxa that release protective bEVs (eg., those producing SCFAs or reinforcing tight junctions), which are efficiently cleared by a robust reticuloendothelial system. However, as individuals age, physiological decline (eg., intestinal barrier thinning, immunosenescence) combined with lifestyle factors often prevalent in older populations (eg., reduced physical activity, low fiber intake, and cumulative antibiotic exposure) creates an intestinal environment conducive to pathobiont expansion. For instance, Lee et al observed that while oral administration of Paenalcaligenes hominis temporarily increased its abundance in mice, levels reverted to baseline upon cessation.⁸³ This suggests that the aged gut environment selectively favors specific bacteria capable of thriving under chronic inflammatory conditions, while failing to support beneficial taxa.

In this dysbiotic state, the production of harmful bEVs—enriched with virulence factors like LPS, CagA, and pro-inflammatory RNAs—increases, exacerbating intestinal inflammation and barrier permeability.^83,102 Simultaneously, the abundance of protective bEVs derived from beneficial taxa (eg., those producing SCFAs or reinforcing tight junctions) diminishes significantly.^{30,126,133,138,227} This dual imbalance results in a massive influx of detrimental bEVs into the circulation. Crucially, this influx is compounded by an age-related decline in systemic clearance mechanisms. Studies indicate that aging¹⁸⁰ and obesity¹⁰⁸ reduce the population of Vsig4+ Kupffer cells, the primary scavengers responsible for clearing circulating bEVs. Consequently, harmful bEVs accumulate in the bloodstream, are taken up by distant host cells, and trigger the diverse array of aging-related diseases detailed in Gut–Brain Axis–Gut–Bone Axis (Figure 6).

Figure 6 Bacterial extracellular vesicles escape from damaged intestinal epithelial barriers into the bloodstream and cause aging-related diseases. Aging disrupts gut–liver axis homeostasis: intestinal dysbiosis and reduced short-chain fatty acids/sIgA compromise epithelial barrier function, facilitating translocation of bacterial extracellular vesicles. Concurrently, diminished Vsig4+ Kupffer cells in the aging liver impair systemic clearance of bacterial extracellular vesicles, enabling their dissemination to peripheral organs and driving inflammaging. Created with Adobe Illustrator.

Implications for Intervention: Restoring Equilibrium

This unified model posits that the pathology of aging-related diseases is driven not merely by the presence of specific pathogens, but by a systemic failure of the “bEV homeostatic axis.” By identifying bEVs as the central effector molecules linking gut dysbiosis to distant organ damage, the therapeutic paradigm must shift from simply altering microbial composition to directly modulating vesicle cargo and enhancing host clearance capacity. Restoring the equilibrium between harmful and protective bEVs—thereby re-establishing the lost homeostatic state—thus emerges as a critical, yet largely unexplored, frontier for developing next-generation interventions against multi-organ decline.

bEVs Pave the Way for Next-Generation Microbiome Therapeutics and Diagnostics

The identification of bEVs as key mediators in the gut–system axis marks a paradigm shift in aging research, offering new paths for therapy and diagnosis. To provide a comprehensive view of this rapidly evolving field, we organize our discussion around four core pillars: therapeutic interventions, diagnostic potential, translational barriers, and future directions.

Therapeutic Interventions: From Dietary Modulation to Engineered Vesicles

Modifying the gut microbiota to alter the bEV landscape is gaining attention as a novel treatment approach with significant potential. Strategies range from broad lifestyle modifications to precise biological engineering, each offering distinct mechanisms for restoring host health.

Dietary Modulation and Longevity

Adjusting the composition of one’s diet—specifically the ratios of fiber, protein, carbohydrates, and fats—can significantly restore the structure of the gut microbiota and its vesicle output. Interestingly, the prevalence of various aging-related diseases does not merely rise with advancing years; instead, it often follows an inverted “U” pattern, diminishing among those who live exceptionally long lives. These individuals, regarded as having aged “successfully,” often possess dietary habits that foster a healthy gut microbiome.²²⁸ A prime example is Sardinia, one of the global “Blue Zones” noted for extended lifespans.²²⁹ Research there shows that the local Mediterranean diet, abundant in vegetables, fruits, legumes, olive oil, and fish, is particularly rich in fiber and polysaccharides. The fermentation of these components leads to the production of SCFAs, which serve as an energy source for the Verrucomicrobia phylum. Consequently, there is a notably higher presence of Verrucomicrobia, including Akkermansia muciniphila, in the intestines of Sardinia’s long-lived residents compared to the general population.²³⁰ Similarly, a study of centenarians in Bama County, China, found their diets are higher in fiber, enriching their gut microbiota with Ruminococcaceae capable of producing and utilizing SCFAs, thus supporting a healthy microbial environment.²³¹ These cases underscore diet as a primary driver of a protective bEV profile.

bEVs as Therapeutic Agents

Beyond diet, bEVs themselves are emerging as promising cell-free therapeutic agents. Akkermansia muciniphila-derived EVs (AEVs) have been the focus of extensive research. Upon uptake by IECs, AEVs significantly enhance both the quantity (increasing from 5.11% to 19.89%) and functionality of mucus-secreting goblet cells, while substantially thickening the intestinal mucus layer.²³² Furthermore, AEVs lower the expression of Toll-Like Receptor 2 and TLR4 in mouse colonic tissues, mitigating inflammation and reducing epithelial barrier permeability.¹³⁵ Research by Kang et al indicates that oral administration of AEVs alleviates colitis in C57BL/6J mice.²³³ Mechanistically, AEVs release encapsulated glycosidases that degrade cellulose, contributing to SCFA generation and addressing deficits from inadequate cellulose consumption.²³⁴ Crucially, AEVs promote the growth of beneficial phyla such as Firmicutes (including Lachnospiraceae, Lactobacillaceae, and Ruminococcaceae) and Bacteroidetes (like Bacteroides spp. and Alistipes spp.), while decreasing potentially harmful bacteria like Proteobacteria (Klebsiella pneumoniae).²³² This selective modulation occurs via membrane fusion, a process linked to the hydrodynamic diameter and zeta potential of the bEVs, where Bacteroidetes readily absorb AEVs while Salmonella and E. coli show minimal uptake.^232,235 Additionally, AEVs taken up by Peyer’s patches enhance mucosal IgA production.²³² Specific bacteria like B. acidifaciens and B. thetaiotaomicron can absorb mucin AEVs, stimulate their own growth, convert IgM to IgA, and facilitate its transport, further elevating IgA levels essential for clearing pathobionts.^236–241 Notably, Shin et al revealed that A. muciniphila abundance in intestinal mucus decreases significantly with age (from 9.2% in young mice to 0.4% in older mice); oral administration of A. muciniphila or its EVs markedly alleviates aging-related intestinal dysfunction and extends healthspan83. Similarly, bEVs from the next-generation probiotic Faecalibacterium prausnitzii possess the capability to restore the intestinal environment.²⁴²

Fecal Microbiota Transplantation

Moreover, FMT has shown promising results in managing stubborn cases of Crohn’s disease and ulcerative colitis by resetting the entire microbial and vesicle ecosystem.²⁴³ Tailored interventions targeting the gut microbiota for particular health conditions, including specific dietary and supplementation strategies based on individual microbiome profiles, are increasingly recognized as vital for enhancing health benefits. However, despite these advances, critical challenges remain in determining the optimal dosage, frequency, and delivery vehicles (eg., enteric coating) for oral bEV administration to ensure stability through the gastrointestinal tract. Furthermore, it remains unclear whether bEV therapies can effectively reverse established age-related tissue damage or if they are primarily prophylactic; defining the therapeutic window and potential synergistic effects with existing geriatric medications is essential for clinical translation.

Diagnostic Potential: BEVs as Biomarkers of Biological Age and Frailty

bEVs are crucial in understanding the aging process, especially for evaluating biological age. As people age, metabolic functions and immune responses undergo notable transformations, influencing health and indicating biological age through changes in gut microbiota composition and activity. Studies show a strong link between bEV levels and biological age, suggesting bEVs could be promising new biomarkers for determining an individual’s aging condition.²⁴⁴

The gut microbiome modulates cellular signaling and immune functions by releasing bEVs, a mechanism intricately linked to age-related physiological transformations.²⁴⁵ Bioactive elements in bEVs, including proteins, lipids, and nucleic acids, facilitate host-microbial interactions that influence metabolic and immune conditions, potentially fostering or hindering aging-associated diseases.^225,246 For instance, SCFAs present in bEVs are vital for managing inflammatory processes and metabolic well-being, impacting the evaluation of biological age.²¹⁸ Consequently, examining the quantity and makeup of bEVs yields valuable insights for determining biological age and assisting in the early identification of health threats.

The significance of bEVs in forecasting frailty is gaining acknowledgment. With an aging population, frailty has emerged as a crucial element influencing quality of life. Studies suggest that examining bEVs can offer vital biomarkers to pinpoint those at risk of developing frailty before clinical symptoms appear, allowing for timely intervention.²¹⁸ By tracking bEV concentrations and components, one can evaluate metabolic health, inflammation levels, and immune response, providing strong evidence for anticipating frailty and associated conditions like cardiovascular issues and neurodegenerative diseases. Additionally, properties of bEVs are linked to specific illnesses; for example, a notable correlation exists between gut microbiota imbalances and AD. Serving as information carriers, bEVs can indicate the physiological status of such imbalances, laying the groundwork for early detection.²⁴⁴

It is also important to note that host cell-derived EVs play a key role in aging by mediating intercellular communication and transferring bioactive molecules (proteins, lipids, miRNAs, lncRNAs) that propagate senescence, inflammation, and oxidative stress. EVs from aged cells can exacerbate dysfunction by promoting inflammatory responses and impairing autophagy, often through pathways like Nrf2. Simultaneously, age-associated molecular changes in host EVs, such as altered miRNA profiles, offer potential biomarkers for early detection of conditions like AD, highlighting their dual role as both drivers of aging and tools for diagnosis.²⁴⁷ Nevertheless, a major hurdle is the lack of standardized reference ranges for bEV biomarkers across diverse populations, accounting for variables like diet, geography, and medication use. Additionally, longitudinal studies are urgently needed to distinguish whether specific bEV signatures are causal drivers of frailty and disease progression or merely reactive byproducts, which is essential for validating their utility as predictive rather than just descriptive markers.

Translational Barriers: Heterogeneity, Immunogenicity, and Safety

The purification, quantification, and functional assessment of bEVs pose significant challenges. Variations in properties and constituents occur due to differing experimental setups and sample origins, hindering a comprehensive understanding of their roles. Studies indicate that bEVs are structurally and functionally diverse, transporting a range of bioactive substances. This diversity makes consistent comparisons and reproducibility among research facilities difficult.²⁴⁸ Additionally, the absence of standardized methods for effective isolation (eg., ultracentrifugation vs. size-exclusion chromatography) and quantification (particle count vs. protein mass) restricts their use in preclinical and clinical research. To progress, the scientific community must develop cohesive standards and protocols encompassing all facets of sample collection, processing, and analysis.²⁴⁵

Beyond technical standardization, evaluating immunogenic potential and long-term safety is crucial. bEVs originate from various intestinal bacteria, and these different origins can provoke distinct immune reactions. Recent studies suggest EVs from specific sources may possess immunogenic properties, activating the immune system and potentially leading to adverse effects.²⁴⁶ For example, when biological barriers are crossed, bEVs might trigger an antibody response, diminishing therapeutic effectiveness or increasing side effects. Thus, thorough immunogenicity evaluations are vital. Furthermore, the long-term impact of bEVs on the host is not fully understood, especially in chronic disease management where extended use could lead to unforeseen side effects, such as the horizontal transfer of antibiotic resistance genes or ecological disruption of the resident microbiome.^249,250 Consequently, systematic and prolonged clinical investigations are necessary to assess safety and efficacy. Establishing universal standards for bEV isolation, purity assessment, and potency testing is therefore an urgent priority for the field to enable cross-study comparisons. Equally critical is the need for rigorous long-term toxicology studies in large animal models to evaluate immunogenic risks, biodistribution, and the potential for unintended ecological disruption within the host microbiome following chronic therapy.

Future Directions: Multi-Omics Integration and Clinical Validation

To investigate the significance of bEVs in age-related diseases, upcoming studies must concentrate on multi-omics integrated analysis. This method offers an in-depth comprehension of bEV roles. These vesicles contain various bioactive substances facilitating intercellular communication and regulating biological processes like metabolism and immune responses.²⁵¹ Combining data from genomics (to identify source bacteria), transcriptomics (for RNA cargo), proteomics (for surface proteins), and metabolomics (for lipid/small molecule cargo) enhances our understanding of bEV mechanisms in diverse physiological contexts.²²⁵

Research indicates bEVs affect aging by modulating host endocrine and metabolic conditions.²⁵² In neurodegenerative diseases, bEVs might reduce disease advancement by influencing neuronal oxidative stress and inflammatory reactions.²⁵³ By combining omics fields, it becomes possible to pinpoint significant biomarkers within bEVs, offering fresh perspectives for early detection and management. Moreover, leveraging artificial intelligence and machine learning to analyze extensive multi-omics datasets can improve insights, propelling personalized healthcare. By mapping correlations between bEVs and host biomarkers, novel therapeutic targets can be discovered.

Ultimately, a key area for future investigation involves initiating clinical trials. Numerous efforts have shown bEVs may be beneficial in addressing metabolic disorders,²⁵⁴ neurodegenerative illnesses,²⁴⁷ and immune-related diseases.²⁵⁵ However, clinical use remains restricted due to insufficient systematic data. Upcoming studies need to focus on creating extensive randomized controlled trials to assess the effectiveness of bEVs from various origins. For instance, studies involving bEVs for AD could confirm efficacy by monitoring alterations in cognitive abilities, inflammation indicators, and relevant biomarkers.²⁹ Research must also prioritize assessing safety, investigating adverse effects, and establishing acceptable risk profiles for prolonged usage. Key questions remain regarding how multi-omics data can be effectively translated into actionable clinical algorithms for patient stratification and real-time monitoring. Furthermore, identifying the most viable and sensitive endpoints for early-phase clinical trials is crucial to demonstrate the efficacy of bEV therapies in slowing or reversing human aging processes, given the slow progression of age-related phenotypes.

Limitations and Future Perspectives

Despite the compelling evidence presented in this review, translating the “Dysbiosis-Permeability-Clearance” model into clinical practice faces significant hurdles. While we have synthesized a unified mechanism where aging compromises both intestinal barrier integrity and systemic clearance (specifically via the loss of Vsig4+ Kupffer cells), several critical limitations currently impede the validation and therapeutic exploitation of this axis. Addressing these challenges defines the immediate research agenda for the field.

Methodological Heterogeneity and the Lack of Bacterial-Specific Markers

A primary bottleneck, as highlighted in our discussion of conflicting reports (eg., the paradoxical pro-calcific effects of LGG-derived bEVs in CKD versus their benefits elsewhere²¹¹), is the absence of standardized protocols for distinguishing bEVs from host-derived EVs. Current isolation techniques (ultracentrifugation, size-exclusion chromatography) often yield preparations contaminated with host lipoproteins or exosomes, and the lack of universal bacterial surface markers complicates precise quantification.²⁵⁶ This methodological noise may partly explain the discrepancies in bEV abundance reported across different aging studies. Future efforts must prioritize the development of consensus guidelines specifically for bacterial vesicles, akin to MISEV²⁵⁷ but tailored to prokaryotic origins. Crucially, the discovery and validation of unique bacterial EV markers (distinct from common LPS or generic protein stains) are essential to accurately track bEV translocation from the gut to distant organs like the brain and liver without cross-contamination artifacts.

Inter-Individual Variability and the Context-Dependent Nature of bEVs

The dual nature of bEVs—acting as homeostatic mediators in youth but drivers of pathology in aging—is profoundly context-dependent, posing a challenge for “one-size-fits-all” therapies. As illustrated by the strain-specific entry routes of P. hominis (vagus nerve-dependent) versus E. coli (potentially vagus-independent) into the brain,^83,103 the impact of bEVs varies not only by bacterial species but also by the host’s physiological state (eg., renal function, immune senescence, genetic background like LRRK2 mutations).^103,180 The dramatic decline of protective taxa like Akkermansia muciniphila in the elderly further complicates the baseline bEV landscape, creating high inter-individual variability.²⁵⁸ Future research must move beyond murine models to embrace precision medicine frameworks. Large-scale, longitudinal human cohort studies are needed to map how bEV signatures evolve with age across diverse populations, accounting for variables such as diet, geography, and medication history. Only by stratifying patients based on their specific “bEV-clearance capacity” (eg., Kupffer cell function) and microbiome profile can we predict who will benefit from bEV-targeted interventions and who might be at risk of adverse effects.

The Gap Between Correlation and Causation in Human Aging

Perhaps, the most critical limitation is that while our proposed unified model is robust in preclinical systems, direct causal evidence in humans remains scarce. Most current human data are correlational, linking shifts in microbiota composition or circulating EV levels to diseases like AD, MASLD, and osteoporosis.^259,260 We still lack definitive proof that circulating bEVs are the drivers of human tissue decline rather than bystanders of general dysbiosis and inflammation. For instance, while the loss of Vsig4+ Kupffer cells is well-documented in aged mice,¹⁸⁰ its extent and functional consequence in aging humans remain to be quantified. Rigorous mechanistic studies are urgently required to bridge this gap. This includes the development of advanced non-invasive imaging to track bEV biodistribution in humans and the use of humanized mouse models or organ-on-a-chip systems that recapitulate the aged human immune environment. Establishing clear causal links is the prerequisite for validating bEVs as reliable therapeutic targets.

Concluding Remarks

In conclusion, while the path forward involves navigating complex methodological and biological challenges, the potential for studying bEVs in the context of aging and associated illnesses is extensive. Elucidating the precise role of bEVs in host pathophysiology not only offers fresh insights into the underlying mechanisms of aging but also paves the way for transformative clinical applications—from novel diagnostic biomarkers to next-generation therapeutics. As research progresses through interdisciplinary collaboration and rigorous standardization, we expect this field to yield more efficient approaches for preventing and managing age-related conditions. Ultimately, unlocking the secrets of bEVs holds the promise of enhancing the well-being and health span of our rapidly aging global population.