The healthy gut microbiota is referred to as the microbial barrier, and comprises various species of commensal intestinal bacteria (43). These species either compete or cooperate to establish a balanced microbial community (43), which is critical for resisting the colonization, growth, and invasion of pathogenic microorganisms (44). BEVs can modulate the equilibrium of the gut microbiota in several ways ( Figure 2B ). First, BEVs promote the survival of their parent bacterium or other bacteria. For instance, P. aeruginosa OMVs carry many Pseudomonas quinolone signals, which can bind iron, an essential element for bacterial viability, and bring it to the outer membrane of the parent bacterium via fusion (18). Similarly, OMVs from Akkermansia muciniphila (A. muciniphila) can restore the disturbed balance of gut microbiota via selectively promoting the proliferation of beneficial bacteria through membrane fusion (45). Second, bacteria release BEVs as a defense mechanism against phage infections. For example, Manning et al. reported that the co-incubation of OMVs collected from Escherichia coli (E. coli) and T4 bacteriophages resulted in a significant reduction in the active phage number (46). Similarly, Reyes-Robles et al. found that Vibrio cholerae (V. cholerae) secreted OMVs carrying phage receptors as a defense mechanism that conferred protection against phage predation (47). Finally, BEVs function as tools to eliminate other bacteria. Li et al. reported that OMVs from 15 strains of gram-negative bacteria, including many commensal or pathogenic gut bacteria, such as Enterobacter, Escherichia, Morganella, Salmonella, and Shigella strains, could lyse many gram-positive and gram-negative cultures. Peptidoglycan hydrolases associated with BEVs are thought to account for bacterial lysis (48). Growing evidence has supported the antimicrobial functions of BEVs. For instance, OMVs from P. aeruginosa can kill competitor species such as S. aureus via peptidoglycan hydrolases, antimicrobial 4-hydroxy-3-methyl-2-(2-non-enyl)-quinoline, and rhamnolipid (49). OMVs from Lysobacter and Myxococcus contain a toxic mixture of bioactive compounds and lytic enzymes capable of killing the surrounding microbes (50).

The intestinal chemical barrier is composed of AMPs and other antibacterial substances, such as bile acids (41), and inhibits growth of certain bacteria and segregates intestinal bacteria from intestinal epithelial cells. Several studies have suggested that BEVs can disrupt the function of the chemical barrier ( Figure 2C ). Nakayama-Imaohji et al. reported that a Bacteroides fragilis (B. fragilis) strain with hypervesiculating mutants (which release more OMVs) showed higher resistance to treatment with AMPs, such as LL-37 and defensin-2 (51). Similarly, Urashima et al. found that the outer membrane protein T, which was specifically enriched in the OMVs of enterohemorrhagic E. coli (EHEC), broke down LL-37 and inhibited its antimicrobial activity, thereby enhancing EHEC survival and adaptation to the host gut environment (52). Moreover, the exposure of P. aeruginosa to lysozyme significantly enhanced OMVs release (by approximately 100-fold) (53). Analogously, in vitro studies have shown that E. coli upregulates OMVs secretion upon encountering AMPs, and the addition of E. coli OMVs has been demonstrated to increase bacterial survival in vitro when challenged with antibiotics, such as Polymyxin B and colistin (54). All in all, BEVs can digest or neutralize AMPs, potentially weakening the chemical barrier function.

Contrarily, some evidences indicated that BEVs act as stimulators, inducing the intestine to increase AMP expression, thereby enhancing the chemical barrier ( Figure 2C ). For instance, Kaparakis et al. discovered that OMVs from P. aeruginosa and Helicobacter pylori (H. pylori), which contained peptidoglycans, could induce epithelial cells to express human β-defensins (HBD), such as HBD2 and HBD3 (55). Lactobacillus derived CMVs have also been reported to stimulate the expression of the AMP REG3G, a c-type lectin, thus promoting the chemical barrier of the gastrointestinal tract and providing protection against pathogens (56). OMVs released from A. muciniphila were recently reported to stimulate goblet cells to produce mucus (45), which resisted the adhesion and stimulation of pathogenic bacteria to gut epithelial cells. These findings suggest that some BEVs can stimulate intestinal cells, leading to increased AMP and mucus production, thereby enhancing the chemical barrier.

The intestinal epithelial barrier, a physical partition separating the body’s internal environment from the lumen, is composed of a single layer of epithelial cells interconnected via tight junction proteins, such as occludin, claudins, and zonula occludens (39, 57). Although this physical barrier effectively limits the intrusion of most harmful substances, BEVs have been shown to internalize or permeate it (58). BEVs penetrate non-phagocytic host cells via five primary mechanisms: clathrin-mediated endocytosis, caveolin-mediated endocytosis, lipid raft-mediated endocytosis, macropinocytosis, and membrane fusion (17, 59). BEVs can concurrently utilize one or more pathways to infiltrate host cells, depending on their size and components. For example, OMVs derived from H. pylori were found to enter epithelial cells via four different mechanisms (55, 60, 61).

Once internalized, BEVs traverse the endolysosomal pathway and are subsequently degraded in lysosomes or autophagosomes (62); however, recent studies suggest that some BEVs can escape degradation and deliver their cargos into cells. Bielaszewska et al. demonstrated that after the OMVs of EHEC O157 were internalized in early endosomes through a process reliant on dynamin-dependent endocytosis, virulence factors, including Shiga toxin 2a (Stx2a), cytolethal distending toxin V (CdtV), and EHEC hemolysin, were separately transported from the vesicles via intracellular trafficking (63). Although the precise mechanisms of BEVs internalization and cargo transport remain unclear, BEVs are surmised act as a significant cargo delivery system to intestinal epithelial cells, influencing the function and integrity of the physical barrier of the gut.

Evidence suggests that some gut pathogenic bacteria can damage the intestinal barrier via BEVs ( Figure 2D ). Upon internalization in human intestinal epithelial cells, gram-negative bacterial OMVs release lipopolysaccharides (LPS) into the cytosol (64), facilitated by sorting nexin 10 (SNX10), which activates caspase-5. This leads to Lyn phosphorylation, subsequently down-regulating E-cadherin expression and impairing the intestinal barrier (64). EHEC O157 OMVs can disrupt the barrier through two pathways: the release of hemolysin, which increases mitochondrial permeability and triggers apoptosis (65), and the discharge of CdtV-B, which causes DNA damage and induces apoptosis (66). Moreover, OMVs of the pathogen Fusobacterium nucleatum (F. nucleatum) can activate the FADD-RIPK1-cCASP-3 signaling pathway, decreasing ZO-1 protein and increasing apoptosis, thereby damaging the epithelial barrier (67). Similarly, Campylobacter jejuni releases OMVs containing toxins that harm cellular DNA and impair the intestinal barrier (68–70). V. cholerae OMVs carry active proteases that induce apoptosis or necrosis, causing epithelial barrier loss during infection (71). Finally, Enterotoxigenic B. fragilis releases OMVs along with B. fragilis toxin, which disrupt the intestinal barrier via cleaving E-cadherin and affecting the zonula adherens and tight junctions in the intestinal epithelium (72).

Despite their disruptive potential, BEVs do not always impair the intestinal epithelial barrier ( Figure 2D ). The probiotic E. coli Nissle 1917 and commensal ECOR63 enhance barrier function via increasing tight junction protein expression (73, 74). Furthermore, OMVs produced by E. coli C25, a commensal bacterium, trigger a moderate release of the proinflammatory interleukin 8 (IL-8) and stimulate the transcriptional upregulation of Toll-like receptors (TLRs) in intestinal epithelial cell lines, subsequently enhancing the barrier function of epithelial cells and inhibiting bacterial internalization (75). Similarly, OMVs released from A. muciniphila help maintain the integrity of the intestinal barrier via penetrating the intestinal epithelial cells and boosting the expression of tight junction proteins and mucus (45, 76).

The gut immune barrier, primarily comprising immune cells including macrophages, dendritic cells (DCs), lymphocytes, mast cells, and natural killer cells, resides in the lamina propria or Peyer’s patch, situated beneath the physical barrier (77). They can gather information from the intestinal epithelial cells which produce a range of immunoregulatory signals (78). Furthermore, they directly recognize and accept certain bacterial components that permeate this barrier (79).

BEVs can stimulate intestinal epithelial cells to secrete various cytokines and chemokines that play pivotal roles in modulating intestinal immune functions ( Figure 2E ). For instance, F. nucleatum releases OMVs that stimulate epithelial cells, thereby increasing the activation of p-ERK, p-CREB, and NF-κB signaling pathways. This activation subsequently upregulates proinflammatory cytokines, including tumor necrosis factor, keratinocyte chemoattractant, IL-6, interferon (IFN)-γ, and monocyte chemoattractant protein (MCP)-1 (55). Thapa et al. analyzed the effect of BEVs derived from 32 different gut bacteria (26 gram-negative and six gram-positive bacteria) on intestinal epithelial cells. Their findings revealed that BEVs could induce species-specific immune responses in these cells. OMVs from gram-negative bacteria were found to trigger a stronger proinflammatory response than CMVs from gram-positive bacteria. A large proportion of the BEVs induced a significant increase in CCL20, IL-8, and CXCL1 levels in epithelial cell lines. Their research also identified LPS as the dominant proinflammatory bacterial effector that activated the caspase- and RIPK2-dependent pathways (80). OMVs can stimulate immune responses via cell surface and intracellular receptors in epithelial cells. For example, EHEC O157 OMVs induce IL-8 production in human intestinal epithelial cells via stimulating TLR4 and TLR5 (cell surface receptors), thus activating the nuclear factor NF-κB (81). In addition, these OMVs can deliver peptidoglycan into the host cell cytosol, thereby inducing innate immune responses through a NOD1 (intracellular receptor)/NF-κB dependent, but TLR-independent, mechanism (55, 59).

BEVs can also directly engage with intestinal immune cells (59), particularly macrophages, which play vital roles in the immune barrier ( Figure 2E ). Research shows that macrophages can uptake gram-negative OMVs via clathrin-mediated endocytosis. LPS from these OMVs can escape from early endosomes into the cytosol, triggering the caspase-11-dependent release of IL-1β and cell death in a dose-dependent manner (82). Previous studies found that guanylate-binding proteins recognized LPS, bound to the OMV surface, and mediated activation of the caspase-11 non-canonical inflammasome (83). Similarly, Bitto et al. reported that OMVs from P. aeruginosa directly activated the inflammasome in macrophages (84), which were dependent on caspase-5, a human homolog of murine caspase-11, highlighting another pathway of activation of immune responses in mice and humans via OMVs (83). Moreover, gram-positive bacteria can also initiate an immune response in macrophages via signaling pathways that differ significantly from those used by OMVs. Wang et al. discovered that S. aureus released CMVs that interacted with TLR2, thereby activating the NLRP3 inflammasome via potassium efflux. This led to the recruitment of apoptosis-associated speck-like protein containing a caspase recruitment domain and caspase-1 activation, resulting in the cellular release of mature cytokines IL-1β and IL-18 and the induction of pyroptosis (85). Conversely, CMVs from Pediococcus pentosaceus demonstrated potent anti-inflammatory properties. These CMVs facilitate the differentiation of bone marrow precursors into myeloid-derived suppressor-like cells and promote M2 macrophage polarization in vitro and in vivo (85, 86).

DCs are phagocytes and antigen-presenting cells that regulate the activation of adaptive immune responses, particularly T-helper and regulatory T (Tregs) cells ( Figure 2E ) (87). Shen et al. and Chu et al. demonstrated that DCs could detect polysaccharide found in B. fragilis OMVs via TLR2, which then activated growth arrest and DNA damage-inducible protein (Gadd45a) signaling, resulting in increased proliferation of Tregs and secretion of the anti-inflammatory cytokine IL-10 (88). This immune response process protected mice from severe experimental colitis (89). However, deficiencies in ATG16L1 or NOD2, two genes associated with IBD, disrupt DC-Treg cell interactions, thereby obstructing the protective function of B. fragilis OMVs (89). Additionally, E. coli OMVs induce DCs to generate T-helper cell responses in a strain-specific manner. Non-pathogenic E. coli strains, E. coli Nissle 1917 (probiotic) and ECOR63 (commensal), trigger increased secretion of Th1 polarizing cytokines (IFN-γ and IL-12) from DCs. Conversely, OMVs from ECOR12 (commensal) or ECOR53 (pathogenic) stimulate the production of higher levels of Treg-related cytokines (IL-10 and TGF-β). Despite the differences between strains, all OMVs enhance the secretion of Th17/Th22 priming cytokines (IL-6, IL-23, tumor necrosis factor-α, and IL-1β) (58).

What’s more, BEVs can access Peyer’s patches and then directly interact and activate the immune cells. Wang et al. found A. muciniphila OMVs are able to enter Peyer’s patches after direct delivery into the intestinal lumen, and induce higher production of immune active DCs with CD80 expression (45). Consequently, with the help of activated DCs, the productions of CD69⁺ B cells and IgA⁺ plasma cells along with total B cells are significantly augmented, thereby increasing intestinal IgA concentration (45, 90). This process is believed to reduce the relative abundance of harmful pathogens in the gut microbiota.

In conclusion, the influence of BEVs on intestinal barrier regulation is complex, with some enhancing barrier function, and others contributing to its impairment. This highlights the intricate interplay between the gut microbiota and their multifaceted effects on human health and diseases.

Numerous studies have highlighted the role of gut bacteria in exploiting BEVs to infect and harm hosts. BEVs can neutralize AMP activation, potentially undermining the effectiveness of the host chemical barriers and enhancing their susceptibility to pathogenic infections (53, 54). Furthermore, BEVs can foster the formation of biofilms and complex microbial communities, which are implicated in gastrointestinal infections and other diseases (22, 104, 105). A significant proportion of biofilm matrix-associated proteins originate from BEVs such as OMVs produced by P. aeruginosa (91). Several intestinal pathogens employ BEVs as vehicles for delivering toxins to gut cells during infection. V. cholerae, a noninvasive gram-negative pathogen, causes cholera via colonizing the small intestine and releasing a potent enterotoxin called cholera toxin (CT). Chatterjee et al. discovered that V. cholerae OMVs carried copious amounts of CT and could be internalized by intestinal epithelial cells, subsequently increasing cyclic adenosine monophosphate levels in a ganglioside GM1 (CT receptor)-dependent manner (92). Similarly, L. monocytogenes, a gram-positive intracellular pathogen, utilizes CMVs to release toxins, including listeriolysin O and phosphatidylinositol-specific phospholipase C, which cause mammalian cytotoxicity (93).

However, not all BEV effects on infections are harmful. Evidence suggests that BEVs can bolster gut defense against bacterial infections. During these infections, OMVs deliver LPS to macrophages, inducing a caspase-11-mediated inflammatory reaction that aids the host in pathogen clearance (82). Patten et al. observed that pre-incubation of intestinal epithelial cells with E. coli C25-derived OMVs impeded the internalization of the parent bacterium. They suggested that this was due to the mild proinflammatory response induced by OMVs in epithelial cells, which enhanced their ability to combat infection (75).

Moreover, BEVs are critical mediators in immune training and fortifying antiviral defenses. Bhar et al. found that co-inoculation of BEVs with murine norovirus led to the enhanced production and release of proinflammatory cytokines in macrophages, suggesting the potential role of BEVs in promoting an antiviral response (94). Erttmann et al. reported that gut microbiota depletion lowered systemic tonic IFN-I levels and antiviral priming, rendering the mice more susceptible to systemic viral infections. They found that the gut microbiota released DNA-containing BEVs that could permeate the intestinal barrier and circulate in the blood, delivering foreign DNA to distal host cells, thereby activating the cGAS-STING-IFN-I-dependent pathway to protect against RNA viruses (95). Additionally, Frantz et al. identified a specific small RNA (sRNA), rli32, partially derived from L. monocytogenes CMVs, that could infiltrate mammalian cell lines and increase IFN-I expression in a RIG-I-dependent manner (96).

Numerous studies have suggested that gut barrier dysfunction can exacerbate IBD progression. Current research indicates that BEVs contribute to IBD via damaging both physical and immune barriers, particularly the epithelial and immune cells ( Figure 3A ). The increased proportion of gram-negative bacteria observed in patients with IBD typically release excessive OMVs laden with LPS (64). These OMVs infiltrate epithelial cells and their LPS translocate into the cytosol, instigating immune reactions, downregulating E-cadherin expression, and causing intestinal barrier dysfunction (64). Tulkens et al. clinically investigated and revealed a significant correlation between the levels of BEV-associated LPS in the plasma and impaired barrier integrity in patients with intestinal mucositis, including IBD (20, 106). As a result, bacterial BEVs significantly stimulate peripheral blood mononuclear cells to secrete proinflammatory cytokines such as IL-6, IL-8, MCP-1, and macrophage inflammatory protein-1α. Specific OMVs released by gut pathogens are also associated with IBD (20). Liu et al. reported that F. nucleatum OMVs significantly exacerbated dextran sulfate sodium (DSS)-induced colitis symptoms in mice via activating receptor-interacting protein kinases 1 and 3 and inducing epithelial necroptosis. This process resulted in significant epithelial barrier loss and oxidative stress-related damage (67). Engevik et al. supported this result and, in addition, they found that F. nucleatum OMVs also activated TLR4 and downstream targets signal-regulated kinase, cAMP response element binding Protein, and NF-κB, thereby promoting proinflammatory cytokine production and leading to increased immune cell infiltration (107).

Conversely, several studies have reported the protective role of BEVs against IBD ( Figure 3B ). The probiotic E. coli Nissle 1917 enhances gut physical barrier integrity via upregulating the tight junction proteins ZO-1, ZO-2, and claudin-14, thereby attenuating DSS colitis in mice (73). The commensal bacterium B. fragilis secretes OMVs that interact with DCs, triggering immune tolerance and thereby protecting animals from 2,4,6-trinitrobenzenesulfonic acid solution-induced colitis and intestinal inflammation (88). A. muciniphila OMVs are reported to ameliorate DSS-induced colitis using several mechanisms, including restoring the disturbed balance of the gut microbiota, maintaining the integrity of the intestinal barrier, and activating B cells and DCs (45). The IBD-associated genes ATG16L1 and NOD2 are crucial for OMV-mediated activation of colitis protection. ATG16L1 T300A transgenic mice did not exhibit protection from 2,4-dinitrobenzene sulfonic acid-induced colitis. Individuals with Crohn’s disease, a subtype of IBD, typically carry the ATG16L1 major risk variant T300A. This finding suggests a potential target for the early genetic diagnosis of IBD (89).

The balance of gut microbiota significantly influences host metabolic homeostasis, and BEVs are crucial in this process. A significant increase in OMVs from Pseudomonas panacis was observed in a high-fat diet-induced type 2 diabetes mouse model. Subsequent studies confirmed that these OMVs could block the insulin signaling pathway in skeletal muscles and adipose tissues (97). Conversely, A. muciniphila-derived OMVs ameliorate high-fat diet-induced obesity via various mechanisms, which include improved intestinal barrier integrity, reduced inflammation, balanced energy, and improved blood parameters (98). This contrasting effect of different BEVs on metabolic homeostasis emphasizes the complex and multifaceted roles of these entities in maintaining host health. It is thought that the characteristics of the parent bacteria determine whether their BEVs are harmful or beneficial, and normal quantity of BEVs could maintain immunological activity while excess amounts would be harmful.

Numerous reports have highlighted the influence of BEVs on cancer development and metastasis in the gastrointestinal tract. OMVs from E. coli MG1655 have been shown to deliver a tRNA fragment termed as Ile-tRF-5X, into human colorectal carcinoma cells (HCT116). This interaction promotes the expression of mitogen-activated protein kinase 3, thereby enhancing cell proliferation (99). F. nucleatum, widely recognized as a pathogen that promotes colorectal cancer (CRC) development, utilizes various mechanisms for this process, including OMVs. Proteomic analysis using mass spectrometry revealed an abundance of virulence factors and biologically active proteases present or selectively enriched in these OMVs (100). The specific roles of OMVs in CRC include inducing IL-8 expression (100, 101), which fosters a pro-inflammatory microenvironment favoring tumor growth; and reducing E-cadherin and cadherin-1 gene expression to promote an epithelial-to-mesenchymal transition-like genotype in tumor cells (100, 102), which ultimately promotes the migration and invasion of cancer cells in vivo (108).

Conversely studies have explored the potential of BEVs as therapeutic agents for cancer treatment via immunotherapy. Kim et al. found that gram-negative bacterial OMVs specifically targeted and accumulated in tumor tissues of a syngeneic mouse colonic tumor model, subsequently triggering the production of antitumor cytokines CXCL10 and IFN-γ. This indicates that BEVs represent a promising new approach for cancer immunotherapy (103). Additionally, A. muciniphila OMVs have been found to enhance programmed cell death protein-1-based immunotherapy of CRC in mouse models. This suggests a potential clinical application of OMVs in improving the efficacy of immunotherapy by targeting programmed cell death protein-1 (45). These varied findings demonstrate the significant and multifaceted roles of BEVs in gastrointestinal cancer progression and potential therapeutic strategies.