Emerging Role of Exosomes in Diagnosis and Treatment of Infectious and Inflammatory Bowel Diseases

To communicate with each other, cells release exosomes that transfer their composition, including lipids, proteins and nucleic acids, to neighboring cells, thus playing a role in various pathophysiological processes. During an infection with pathogenic bacteria, such as adherent-invasive E. coli (AIEC) associated with Crohn disease, exosomes secreted by infected cells can have an impact on the innate immune responses of surrounding cells to infection. Furthermore, inflammation can be amplified via the exosomal shuttle during infection with pathogenic bacteria, which could contribute to the development of the associated disease. Since these vesicles can be released in various biological fluids, changes in exosomal content may provide a means for the identification of non-invasive biomarkers for infectious and inflammatory bowel diseases. Moreover, evidence suggests that exosomes could be used as vaccines to prime the immune system to recognize and kill invading pathogens, and as therapeutic components relieving intestinal inflammation. Here, we summarize the current knowledge on the role of exosomes in bacterial infections and highlight their potential use as biomarkers, vaccines and conveyers of therapeutic molecules in inflammatory bowel diseases.


Introduction
Cell-to-cell communication is essential to maintain homeostasis in a multicellular organism. Exosomes, small extracellular vesicles of 30 to 100 nm produced by most cell types and delivered into bodily fluids, are a part of a larger network by which cells communicate with each other [1]. These vesicles arise from a unique biogenesis pathway starting with the inward budding of the plasma membrane to form membrane-enclosed compartment called early endosomes [2]. These early endosomes then mature into late endosomes that accumulate intraluminal vesicles (ILVs) in their lumen. These ILVs are formed by inward budding of the membrane of early endosomes, a process mediated by the endosomal sorting complex required for transport (ESCRT) or by ESCRT-independent mechanisms involving tetraspanins or lipids such as lysobisphosphatidic acid and ceramides [2]. During their formation, ILVs are loaded with biologically active molecules, including lipids [3], proteins [4] and nucleic acids [5]. Late endosomes then fuse either with lysosomes, leading to the degradation of their content, or with the plasma membrane to release ILVs into the extracellular space, generating vesicles referred to as "exosomes" [2].
Although initially considered as cellular waste [6], new conceptions supported by recent data propose that exosomes are an alternative way to eliminate waste products and maintain cellular homeostasis [7], and that these vesicles play a role in transmitting information and in activating biological responses in nearby or distant target cells [8]. Indeed, secreted exosomes may elicit cellular uninfected exosome-receiving macrophages is mostly due to the release of M. tuberculosis vesicles rather than host cell-derived exosomes into extracellular milieu [40]. Thus, further studies are needed to decipher the specific role of bacterial and host vesicles in the context of bacterial infection.
Exosomes secreted by macrophages infected with mycobacteria also carry functional bacterial RNAs as well as host cell mRNAs [24] and miRNAs that may target various metabolic pathways of the host following infection [25,41]. Indeed, upon M. tuberculosis infection, macrophages as well as exosomes secreted by these cells exhibit an increase in the level of miR-18a [42]. This miRNA favors M. tuberculosis survival in infected macrophages by counteracting autophagy, however, the impact of miR-18a released into exosomes from infected macrophages on M. tuberculosis survival in exosome-receiving cells has yet to be analyzed [42]. Similarly, the infection of human macrophages with Yersinia pestis and Bacillus anthracis leads to a variation in exosomal miRNA cargo in a pathogen-dependent manner [43], suggesting that the secretion of miRNAs in exosomes may play a role in infection. After LPS exposure, murine bone marrow-derived dendritic cells (BMDC) secrete exosomes containing miR-146a and miR-155. These miRNAs are efficiently transferred to recipient cells in which they modulate inflammatory gene expression and cell responses to endotoxins [44]. Similarly, exosomes derived from the serum of mice that exhibit acute sepsis-related lung injury induced by intraperitoneal LPS injection are selectively enriched in pro-inflammatory miR-155 [45]. These exosomes induce NF-κB activation and the production of TNF-α and IL-6 and promote proliferation of macrophages in vitro [45]. In vivo, the intravenous injection of these exosomes to naïve mice increases the recruitment of pro-inflammatory M1 macrophages to the lungs and induces lung inflammation [45]. Exosomes from macrophages infected with Helicobacter pylori also exhibit an increase in miR-155 level, which exacerbates inflammatory responses in recipient macrophages, thus limiting H. pylori replication and preventing H. pylori-induced gastritis [46].
Exosomes also play a role in the dissemination of bacterial virulence factors. Indeed, the major virulence factor cytotoxin-associated gene A (CagA) secreted by H. pylori was found in exosomes from gastric epithelial cells inducibly expressing the cagA gene as well as in exosomes from patients infected with a CagA-positive H. pylori strain. These exosomes may pass into systemic circulation and deliver CagA to tissues and organs nearby or at a distance from the site of infection [47]. At the systemic level in individuals infected with CagA-positive H. pylori strains, these exosomes may potentially contribute to an increased risk of colorectal cancer [48] and decrease the risk of pancreatic cancer [49]. Exosomes derived from gastric cells infected with H. pylori also transfer the functional activated mesenchymal-epithelial transition factor that is internalized by tumor-associated macrophages [50]. After internalization, this factor seems to modulate tumor-associated macrophages to promote gastric cancer progression [50]. In the stomach, exosomes derived from the serum of H. pylori-infected patients suffering from chronic gastritis can induce the expression of soluble IL-6 receptor, which increases expression of the pro-inflammatory cytokine IL-1α in gastric epithelial cells [51]. Similarly, Shiga toxin (Stx) 2a secreted by enterohemorrhagic E. coli is captured by kidney cells, transported to recycling endosomes and released into extracellular milieu, where it could be either free or associated with exosomes [52]. In mice, exosome-associated Stx2a was found to induce more critical lethality and kidney damages than the free form of Stx2a [52]. Finally, cells intoxicated with the lethal toxin virulence factor secreted by B. anthracis, a toxin that cleaves most mitogen-activated protein kinase (MAPK) kinases (MAPKKs) and nucleotide binding oligomerization domain (NOD)-like receptor family pyrin domain containing 1 (NLRP1), leading to the inhibition of most MAPK signaling pathways and activation of NLRP1 inflammasomes, secrete exosomes delivering this toxin to naïve cells [53]. These exosomes deliver the lethal factor to naïve cells at a distance from the infection site, thus contributing to the diffusion and persistence of these toxins within the host [53].

Exosomes and Adaptive Immune Responses
Increasing evidence suggests that exosomes can modulate T and B cell differentiation, activation and proliferation by either directly interacting with B and T cells or indirectly via their capture by antigen-presenting cells (APCs) that present exosomal antigens to T cells, thus influencing adaptive immune responses. Indeed, it has been shown that exosomes secreted by dendritic cells and B lymphocytes express the major histocompatibility complex (MHC) class I and II loaded with antigens as well as CD86 and adhesion molecules on their surface [54]. In vitro, this exosomal complex can activate naïve or differentiated T lymphocytes [54]. Moreover, electronic microscopy analysis has revealed the presence of MHC-II molecules on the surface of follicular dendritic cells [55]. As these cells do not normally express MHC-II molecules, the latter probably come from the capture of exosomes secreted by adjacent B cells [55]. Similarly, dendritic cells infected with mycoplasma bacteria secrete exosomes that activate B cell proliferation without the help of T lymphocytes and in an antigen non-specific manner [33]. Tumor cells infected with Mycoplasma also secrete exosomes that activate B cells [33], which, in turn, secrete exosomes inhibiting T cell activation [56]. Because exosomes convey antigens and are enriched in MHC molecules, they may act as a direct or indirect antigen-presenting component.

Direct Antigen Presentation
Exosomes can directly present their MHC-loaded antigens to T cells without the help of dendritic cells. However, the ability of exosomes derived from APCs to stimulate T cells is 10 to 20 times lower than that of APCs themselves [57,58]. Exosomes from dendritic cells carrying ovalbumin on MHC-I molecules activate CD8 + T hybridoma cells [59]. Dendritic cells infected with M. tuberculosis secrete exosomes carrying M. tuberculosis peptide-MHC-II-complexes that may activate T lymphocytes [60]. In addition, activated T cells are more susceptible to transcriptional modulation in response to exosomes derived from dendritic cells than resting T cells [61]. This exosomal stimulation results in an increased expression of inhibitory cytokines by T cells [61]. Exosomes from dendritic cells also induce the activation and proliferation of natural killer (NK) cells via the IL-15Rα receptor on the exosomal surface that binds to IL-15, and via the UL16 binding protein 1 (ULBP1), a ligand for the natural killer group 2 member D (NKG2D) receptor expressed by NK cells [62].

Indirect Antigen Presentation
MHC-antigenic peptide complexes may be transported via exosomes to recipient APCs and, without the need of being processed by recipient cells, be used for the presentation of exogenous antigens to T lymphocytes and activation of these cells [63]. This indirect mode of antigen presentation is also called as MHC cross-dressing. Exosomes carrying antigens, in the presence of naïve dendritic cells, are able to activate T lymphocytes [64]. Although exosomes secreted by M. tuberculosisor M. bovisinfected macrophages contain bacterial antigens as well as MHC and co-stimulatory molecules and can directly stimulate CD4 + and CD8 + T cells isolated from mycobacteria-sensitized mice, maximal stimulation of these cells requires the prior incubation of exosomes with APCs such as dendritic cells [22]. Exosomes derived from murine dendritic cells contain antigens and MHC-II molecules and induce the activation of T lymphocytes in recipient mice [58]. However, the activation of T lymphocytes in vitro by these exosomes requires the presence of dendritic cells, which constitute essential intermediaries for exosomal antigenic presentation [58]. Finally, it was reported that mast cell-derived exosomes are able to induce the maturation and functional activation of dendritic cells through the cross-presentation of antigens to T cells in vivo [65].
Activation of T cells leads to an increased number of secreted exosomes and to the release of a distinct exosomal population, depending on their activation status [66]. Exosomes derived from activated CD3 + T lymphocytes, together with IL-2, stimulate and induce the proliferation of resting CD3 + T lymphocytes [67]. Moreover, T cell cultures pulsed with exosomes purified from stimulated CD3 + T lymphocytes and IL-2 showed an increase in CD8 + T cells [67]. During cognate immune interaction, miRNAs are transferred unidirectional from T cells to APCs via exosomes in an antigen-dependent manner [68]. Exosomes derived from CD8 + T lymphocytes carry miR-150 and modulate macrophages, which in turn induce regulatory T (Treg) cells and inhibit effector T cell proliferation, thus suppressing specific immune responses [69]. Finally, Treg cells release a larger number of exosomes compared to other T lymphocytes. These exosomes may carry the CD39 and CD73 surface proteins, which induce the production of adenosine, a nucleotide that plays a major role in anti-inflammatory response and T cell suppression [70], and transport the miRNA let-7d that suppresses T helper (Th) 1 proliferation and cytokine secretion, thus mediating immunosuppression [71].

Exosomes and IBD
IBD, which comprises CD and UC, are a family of chronic immune-related disease of the gastrointestinal tract characterized by defects in intestinal homeostasis leading to chronic relapsing intestinal inflammation. The exact etiology of IBD is unclear but seems to result from an abnormal and prolonged inflammatory response to intestinal microbiota in genetically susceptible individuals [18]. Diagnosis of IBD currently relies on a combination of symptoms, biological, endoscopic, radiologic and histologic criteria and on their evolution over time [72], and available drugs only aim at controlling relapses, limiting recurrence and improving the quality of patients' life but fail to definitively cure IBD [73]. The dysregulation of mucosal immunity plays a major role in IBD development and progression. The larger part of pro-inflammatory cytokines is secreted by CD4 + T cells, including Th1, Th17 and Th2. While elevated levels of Th1 cytokines, such as TNF-α, IL-12 and IFN-γ, and Th17 cytokines, such as IL-17, IL-21 and IL-22, were reported in the intestinal mucosa of CD patients, UC patients rather exhibit an increase in Th2 cytokines, such as IL-5 and IL-13, and Th17 cytokines [74].
It was reported that exosomes isolated from the colonic lumen of IBD patients contain higher amounts of IL-6, IL-8, IL-10 and TNF-α compared to exosomes from healthy subjects, and the levels of these pro-inflammatory molecules were positively correlated with CD severity score [75]. Moreover, exosomes from IBD patients induce the activation of colonic epithelial cells in vitro, which then produce a greater amount of IL-8 [75]. Exosomes from IBD patients and intestinal epithelial cells treated with these exosomes induce the migration of a higher number of macrophages in vitro than untreated intestinal epithelial cells [75]. Similarly, exosomes purified from the serum of mice treated with dextran sulfate sodium (DSS) to induce colitis contain 56 differentially expressed proteins compared to exosomes isolated from untreated mice and are able to activate the MAPK pathway and to induce the production of TNF-α in naïve macrophages [76]. Intestinal microbiota play a major role in the development of IBD [18]. Enteropathogenic bacteria, such as Bacteroides fragilis, secrete outer membrane vesicles that stimulate the release of intestinal mucosa-derived exosomes carrying an elevated level of C-C motif chemokine 20 (CCL20) and prostaglandin E2 (PGE2). CCL20 and PGE2 induce the recruitment and proliferation, respectively, of Th17 cells through the MyD88-mediated pathway, and cause inflammation [77]. The bacterial stimulation of human neutrophils induces the release of microvesicles that exert antibacterial effects in a manner that depends on β2 integrin function, continuous actin remodeling and glucose supply [78]. An increase in human proteins associated with antimicrobial defense, and more specifically to oxidative stress, was reported in extracellular vesicles (including exosomes, microvesicles and bacterial outer membrane vesicles) found on the interface between intestinal mucosa and intestinal lumen in IBD patients compared to control subjects [79]. In addition, an increase in proteins related to oxidative antimicrobial activity was shown in IBD patients and was correlated with an alteration in microbial functions [79]. Thus, the modification of the protein composition of extracellular vesicles may be associated with an alteration in microbial functions and may play a role in the abnormal interactions observed between the host and its microbiota in IBD patients [79].
Exosomes purified from intestinal epithelial cells or macrophages infected with CD-associated AIEC can activate NF-κB and MAPK pathways in naïve macrophages, leading to an increased secretion of the pro-inflammatory cytokines IL-6 and TNF-α [20] (Figure 1). Exosomes derived from AIEC-infected intestinal epithelial cells can also activate the NF-κB pathway in uninfected intestinal epithelial cells, thereby increasing IL-8 secretion. Moreover, exosomes derived from AIEC-infected intestinal epithelial cells and macrophages promote AIEC replication inside exosome-receiving cells [20] ( Figure 1). Our unpublished data showed that AIEC-infected intestinal epithelial cells secrete exosomes that have high levels of miR-30c and miR-130a. The exosomal miR-30c and miR-130a are then transferred via exosomes to recipient cells, where they target the mRNAs encoding the autophagy proteins ATG5 and ATG16L1, inhibiting ATG5 and ATG16L1 expression, respectively. This consequently leads to impaired autophagy-mediated clearance of intracellular AIEC, enhancing AIEC intracellular replication ( Figure 1). The inhibition of these miRNAs in AIEC-infected cells abolishes the increase in miR-30c and miR-130a levels in exosomes secreted by these cells and in exosome-receiving cells, thus suppressing the inhibitory effect of exosomes on ATG5 and ATG16L1 expression and on autophagy-mediated AIEC clearance in exosome-receiving cells (unpublished data). In the ileal mucosa of CD patients, increased miR-30c and miR-130a levels and decreased ATG5 and ATG16L1 mRNA expression levels were observed compared to control subjects or UC patients [80], however, the levels of these miRNAs in CD patient-derived exosomes remain to be investigated. The infection of CEABAC10 transgenic mice expressing human carcinoembryonic antigen-related cell adhesion molecule (CEACAM) 6, a receptor expressed by intestinal epithelial cells for AIEC binding [81], with the AIEC reference strain LF82 induces the secretion of exosomes in intestinal lumen. Exosomes isolated from the luminal content of AIEC LF82-infected CEABAC10 transgenic mice trigger a pro-inflammatory response in naïve macrophages in vitro [20]. However, the exosomal components responsible for the induction of pro-inflammatory response in recipient cells remain to be identified. Together, these data highlight that exosomes are important mediators of host-AIEC interaction with their capacity to modulate innate immune responses.
Cells 2020, 9, x FOR PEER REVIEW 7 of 20 exosomes that have high levels of miR-30c and miR-130a. The exosomal miR-30c and miR-130a are then transferred via exosomes to recipient cells, where they target the mRNAs encoding the autophagy proteins ATG5 and ATG16L1, inhibiting ATG5 and ATG16L1 expression, respectively. This consequently leads to impaired autophagy-mediated clearance of intracellular AIEC, enhancing AIEC intracellular replication ( Figure 1). The inhibition of these miRNAs in AIEC-infected cells abolishes the increase in miR-30c and miR-130a levels in exosomes secreted by these cells and in exosome-receiving cells, thus suppressing the inhibitory effect of exosomes on ATG5 and ATG16L1 expression and on autophagy-mediated AIEC clearance in exosome-receiving cells (unpublished data). In the ileal mucosa of CD patients, increased miR-30c and miR-130a levels and decreased ATG5 and ATG16L1 mRNA expression levels were observed compared to control subjects or UC patients [80], however, the levels of these miRNAs in CD patient-derived exosomes remain to be investigated. The infection of CEABAC10 transgenic mice expressing human carcinoembryonic antigen-related cell adhesion molecule (CEACAM) 6, a receptor expressed by intestinal epithelial cells for AIEC binding [81], with the AIEC reference strain LF82 induces the secretion of exosomes in intestinal lumen. Exosomes isolated from the luminal content of AIEC LF82-infected CEABAC10 transgenic mice trigger a pro-inflammatory response in naïve macrophages in vitro [20]. However, the exosomal components responsible for the induction of pro-inflammatory response in recipient cells remain to be identified. Together, these data highlight that exosomes are important mediators of host-AIEC interaction with their capacity to modulate innate immune responses. Similarly, increases in AIEC intracellular replication and the activation of NF-κB and MAPK pathways, leading to enhanced secretion of the pro-inflammatory cytokines IL-6 and TNF-α, were observed in macrophages receiving exosomes derived from IECs and macrophages infected with AIEC.

Exosomes as Promising Diagnostic Tools
Exosomes found in the biological fluids of patients with various diseases such as cancer, inflammatory diseases and infections exhibit modifications in their composition compared to those from healthy subjects. Thus, these vesicles represent an interesting tool for the identification of biomarkers using minimally-or non-invasive techniques [28].
The use of exosomes as biomarkers has been considered in the context of infectious diseases and especially in mycobacterial infection. Twenty mycobacterial proteins known to be involved in M. tuberculosis intracellular survival were identified in exosomes derived from the serum of M. tuberculosis-infected patients [82]. Moreover, the mRNA [83] and miRNA [25,84] profiles in exosomes derived from the serum differ between healthy subjects, patients with active tuberculosis and patients with latent tuberculosis and could therefore serve as biomarkers. Similarly, variation in the percentage of small RNA derived from repeated sequences of the genome in exosomes purified from the serum of tuberculosis patients may constitute a biomarker for the progression of this pathology [84]. Different levels of exosomal miR-148a-3p, miR-451a and miR-150-5p were also reported in pleural effusions of tuberculous lesions, making these miRNAs promising biomarkers for this disease [85]. Finally, a recent study suggested that a combination of the profiling of six differentially expressed miRNAs in plasma-derived exosomes and electronic health records may improve pulmonary tuberculosis and tuberculous meningitis diagnosis compared with miRNA or electronic health records alone [86].
The use of exosomes as biomarkers for non-cancerous inflammatory diseases such as rheumatoid arthritis, acute kidney failure, or IBD has also been investigated [87]. Concerning IBD (Figure 2), an increase in the level of exosomes containing annexin A1, a protein that stimulates intestinal mucosa wound repair in a murine model of colitis [88] and promotes the resolution of inflammation by binding to formyl peptide receptors expressed on immune [89] and epithelial [90] cells, was observed in the serum of IBD patients compared to healthy subjects and may be used as a potential biomarker of intestinal mucosa inflammation [15]. More than 2000 proteins were detected in exosomes purified from the saliva of IBD patients, a much larger amount than that found in healthy subjects [91]. Among these proteins, the most promising biomarker seems to be proteasome subunit alpha type 7 (PSMA7), which exhibits increased expression in IBD patients compared to healthy individuals [91]. This protein is involved in the ubiquitin-proteasome pathway and was previously reported to promote NOD1 degradation, thus limiting NOD1-mediated apoptosis, NF-κB activation and inflammation in the colon carcinoma cell line [92]. Exosomes from the colonic lumen of IBD patients contain significantly higher mRNA and protein levels of IL-6, IL-8, IL-10 and TNF-α compared with those from healthy individuals, and thus may contribute to IBD diagnosis [75].

Figure 2. Diagnostic potential of exosomes in IBD.
Exosomes derived from saliva, intestinal lumen as well as serum, which contain PSMA7, pro-inflammatory cytokine mRNAs and proteins, and annexin A1, respectively, could be used as promising biomarkers for IBD diagnosis.

Vaccination against Bacterial Infection with Exosomes
Exosomes can modulate immune responses and play a role as antigen presenters. Thus, their use as "vaccines" in infectious diseases has been suggested. A first study conducted in the context of bacterial infection reported that exosomes secreted by M. bovis-or M. tuberculosis-infected cells can activate antigen-specific CD4 + and CD8 + T cells in vivo and promote the activation and maturation of dendritic cells [22]. Macrophages treated with M. tuberculosis proteins release exosomes that recruit dendritic cells to the lungs [27] and activate dendritic cells, CD4 + and CD8 + T cells [23] as well as Th1 immune response protecting against M. tuberculosis infection [93] after intranasal administration into mice. In vivo, infection with M. tuberculosis leads to the secretion of exosomes that contribute to T cell responses, suggesting that the cross presentation of antigens may play a significant role in eliciting acquired immune responses during an infection with these bacteria [94].
Dendritic cells treated with diphtheria toxin secrete exosomes which, after intravenous injection into mice, stimulate a specific IgG response against this toxin [95]. Similarly, exosomes released by dendritic cells carry an antigen that exhibits a cross reactivity with the capsular polysaccharide 14 (Cps14) antigen of Streptococcus pneumoniae type 14. These exosomes stimulate the establishment of IgM and IgG responses against this antigen, thus protecting mice from an infection with S. pneumoniae type 14 [96].

Therapeutic Potential of Exosomes in IBD
Exosomes secreted by various cell types can modulate T lymphocyte activation and increase intestinal epithelial barrier function, thus constituting a new therapeutic strategy to reduce inflammation associated with IBD ( Figure 3). Indeed, exosomes released by myeloid-derived suppressor cells (MDSC), a heterogeneous population of immature myeloid cells that regulates immune response and has been described in IBD [97], reduce the severity of DSS-induced colitis by inhibiting Th1 proliferation and promoting Treg cell expansion [98]. Similarly, exosomes secreted by dendritic cells treated with IL-10 inhibit colitis development in 2,4,6 trinitrobenzene sulfonic acid (TNBS)-treated rats by stimulating CD4 + CD25 + Treg cells [99]. In the same way, the administration of exosomes secreted by dendritic cells expressing TGF-β1 alleviates DSS-induced colitis, which is mediated by Th17 responses, by inducing CD4 + Foxp3 + Treg cell activation [100]. Intravenous injection of exosomes released by intestinal epithelial cells from untreated mice to DSS-treated mice reduces the severity of experimental colitis by decreasing the number of CD4 + T lymphocytes and by activating Treg cells and immunosuppressive dendritic cells, while the inhibition of exosome Exosomes derived from saliva, intestinal lumen as well as serum, which contain PSMA7, pro-inflammatory cytokine mRNAs and proteins, and annexin A1, respectively, could be used as promising biomarkers for IBD diagnosis.

Vaccination against Bacterial Infection with Exosomes
Exosomes can modulate immune responses and play a role as antigen presenters. Thus, their use as "vaccines" in infectious diseases has been suggested. A first study conducted in the context of bacterial infection reported that exosomes secreted by M. bovisor M. tuberculosisinfected cells can activate antigen-specific CD4 + and CD8 + T cells in vivo and promote the activation and maturation of dendritic cells [22]. Macrophages treated with M. tuberculosis proteins release exosomes that recruit dendritic cells to the lungs [27] and activate dendritic cells, CD4 + and CD8 + T cells [23] as well as Th1 immune response protecting against M. tuberculosis infection [93] after intranasal administration into mice. In vivo, infection with M. tuberculosis leads to the secretion of exosomes that contribute to T cell responses, suggesting that the cross presentation of antigens may play a significant role in eliciting acquired immune responses during an infection with these bacteria [94].
Dendritic cells treated with diphtheria toxin secrete exosomes which, after intravenous injection into mice, stimulate a specific IgG response against this toxin [95]. Similarly, exosomes released by dendritic cells carry an antigen that exhibits a cross reactivity with the capsular polysaccharide 14 (Cps14) antigen of Streptococcus pneumoniae type 14. These exosomes stimulate the establishment of IgM and IgG responses against this antigen, thus protecting mice from an infection with S. pneumoniae type 14 [96].

Therapeutic Potential of Exosomes in IBD
Exosomes secreted by various cell types can modulate T lymphocyte activation and increase intestinal epithelial barrier function, thus constituting a new therapeutic strategy to reduce inflammation associated with IBD ( Figure 3). Indeed, exosomes released by myeloid-derived suppressor cells (MDSC), a heterogeneous population of immature myeloid cells that regulates immune response and has been described in IBD [97], reduce the severity of DSS-induced colitis by inhibiting Th1 proliferation and promoting Treg cell expansion [98]. Similarly, exosomes secreted by dendritic cells treated with IL-10 inhibit colitis development in 2,4,6 trinitrobenzene sulfonic acid (TNBS)-treated rats by stimulating CD4 + CD25 + Treg cells [99]. In the same way, the administration of exosomes secreted by dendritic cells expressing TGF-β1 alleviates DSS-induced colitis, which is mediated by Th17 responses, by inducing CD4 + Foxp3 + Treg cell activation [100]. Intravenous injection of exosomes released by intestinal epithelial cells from untreated mice to DSS-treated mice reduces the severity of experimental colitis by decreasing the number of CD4 + T lymphocytes and by activating Treg cells and immunosuppressive dendritic cells, while the inhibition of exosome production in vivo increases the severity of DSS-induced colitis [101]. Moreover, exosomes secreted by intestinal epithelial cells after DSS treatment carry a higher amount of TGF-β1 and adhere to the epithelial cell adhesion molecule (EpCAM) expressed by intestinal epithelial cells to alleviate DSS-induced colitis [101].
Cells 2020, 9, x FOR PEER REVIEW 10 of 20 production in vivo increases the severity of DSS-induced colitis [101]. Moreover, exosomes secreted by intestinal epithelial cells after DSS treatment carry a higher amount of TGF-β1 and adhere to the epithelial cell adhesion molecule (EpCAM) expressed by intestinal epithelial cells to alleviate DSS-induced colitis [101].

Figure 3. Therapeutic potential of exosomes in IBD. Exosomes derived from human cells, plants or worms have been shown to modulate various cell types and pathways to inhibit intestinal inflammation. A. Exosomes secreted by intestinal epithelial cells (IECs) [101], dendritic cells (DCs)
[100] and myeloid-derived suppressor cells (MDSCs) [98] are able to limit the expansion of CD4 + T cells, and of Th1 and Th17 cells. These exosomes and exosomes derived from IL-10-treated DCs or TGF-β-expressing DCs and exosome-like nanoparticles from curcuma increase Treg cell development, while exosomes derived from IECs and exosome-like nanoparticles derived from curcuma favor regulatory DC expansion. Finally, broccoli-derived nanoparticles induce tolerant DCs [102]. Although the components responsible for the effects of exosome-like nanoparticles from curcuma or exosomes from MDSCs and DCs remain to be identified, the effects of exosomes derived from IECs seem to be mediated by exosomal TGF-β. B. Exosomes derived from DCs, IECs and ginger-derived exosome-like nanoparticles may improve intestinal barrier function. While specific ginger exosome-like nanoparticle component(s) involved in these effects have not yet been identified, it was shown that IEC-derived exosomes contain annexin A1 (ANXA1), a protein known to favor intestinal epithelial repair. Upon substance P stimulation, IECs also produce exosomes enriched in miR-21. These exosomes promote cell proliferation and migration, thus improving  [98] are able to limit the expansion of CD4 + T cells, and of Th1 and Th17 cells. These exosomes and exosomes derived from IL-10-treated DCs or TGF-β-expressing DCs and exosome-like nanoparticles from curcuma increase Treg cell development, while exosomes derived from IECs and exosome-like nanoparticles derived from curcuma favor regulatory DC expansion. Finally, broccoli-derived nanoparticles induce tolerant DCs [102]. Although the components responsible for the effects of exosome-like nanoparticles from curcuma or exosomes from MDSCs and DCs remain to be identified, the effects of exosomes derived from IECs seem to be mediated by exosomal TGF-β. B. Exosomes derived from DCs, IECs and ginger-derived exosome-like nanoparticles may improve intestinal barrier function. While specific ginger exosome-like nanoparticle component(s) involved in these effects have not yet been identified, it was shown that IEC-derived exosomes contain annexin A1 (ANXA1), a protein known to favor intestinal epithelial repair. Upon substance P stimulation, IECs also produce exosomes enriched in miR-21. These exosomes promote cell proliferation and migration, thus improving intestinal barrier function. C. Exosomes may have an impact on inflammation and macrophage modulation. Mesenchymal stem cell (MSC)-derived exosomes may limit colitis development by modulating protein ubiquitination and by increasing the number of anti-inflammatory M2 macrophages. MSC-derived exosomes that contain miR-146a and curcuma-derived exosome-like nanoparticles are able to inhibit NF-κB activation, thus limiting inflammation. In association with melatonin, MSC-derived exosomes also limit the activation of various pro-inflammatory pathways. Finally, exosome-like nanoparticles derived from ginger, curcuma and N. brasiliensis limit pro-inflammatory cytokine secretion and increase anti-inflammatory cytokine secretion. D. Exosome-like nanoparticles derived from grapes favor intestinal stem cell proliferation, thus limiting colitis development. E. Exosome-like nanoparticles derived from curcuma increase the abundance of butyrate-producing bacteria, while exosome-like nanoparticles from ginger are captured by Lactobacillaceae and modulate bacterial gene expression, thus increasing indole-3-carboxaldehyde expression that induces the production of IL-22 in the colon, leading to improved intestinal barrier function and decreased colitis development. NOX: NADPH oxidase; MMP-9: matrix metallopeptidase-9; iNOS: inducible nitric oxide synthase; ICAM-1: intercellular adhesion molecule-1; COX-2: cyclooxygenase-2.
Maintaining a functional intestinal barrier and epithelial restoration following injuries are also important factors to prevent intestinal inflammation ( Figure 3). Exosomes derived from dendritic cells improve intestinal barrier function in a murine model of DSS-induced colitis by activating NF-κB via the exosomal miR-146b [44,103]. Annexin A1, which is overexpressed during pro-inflammatory response and plays a role in healing mucosal damages, is found in exosomes derived from intestinal epithelial cells [15]. The annexin A1-containing exosomes are able to activate mucosal healing in vitro and ex vivo [15]. Finally, it was recently shown that substance P, a neuropeptide, regulates the production of exosomes selectively enriched in miR-21 by intestinal epithelial cells [104]. In human colonocytes and mouse colonic crypts, these exosomes promote cell proliferation and migration, however, their effects on colitis remain to be investigated [104].
Several studies have reported that exosomes derived from mesenchymal stem cells (MSCs) represent a promising therapeutic tool ( Figure 3). Indeed, the intraperitoneal administration of exosomes secreted by MSCs derived from bone marrow [105][106][107] or umbilical cord [108] to DSS-treated mice decreases disease activity score, weight loss, colon length shortening and histological score by increasing anti-inflammatory and reducing pro-inflammatory responses [105][106][107][108] more efficiently than MSCs alone [105][106][107]. It has been suggested that these exosomes may exert their effects on DSS-induced colitis by regulating protein ubiquitination [107] and by inducing the generation of immunosuppressive IL-10-producing M2 macrophages in the colon [108]. In vitro results confirmed that exosomes derived from MSCs inhibit inflammatory cytokine production by colonic macrophages stimulated with LPS and promote the polarization of these macrophages into M2 phenotype [108]. Moreover, MSC-derived exosomes alleviate colitis in vivo by inhibiting expression of IL-7 and inducible nitric oxide synthase (iNOS) in mouse colonic macrophages [105], thus limiting the production of nitric oxide and the expression of TNF-α, IL-6 and IL-1β and increasing IL-10 secretion [105,109]. MSC-derived extracellular vesicles, including exosomes and microparticles, also decrease the severity of TNBS-induced colitis in rats by suppressing activation of the NF-κB pathway [110], an effect probably mediated by the inhibition of TNF receptor associated factor 6 (TRAF6) and IL-1 receptor associated kinase 1 (IRAK1) by the anti-inflammatory miR-146a carried by these exosomes [111]. This suppression limits pro-inflammatory cytokine production and reduces nitric oxide production in colonic macrophages [110,111]. Finally, the injection of exosomes secreted by adipose tissue-derived MSCs combined with melatonin, an anti-inflammatory hormone, could limit inflammation caused by DSS-induced colitis in rat [112].
Some edible plants and food products, such as Curcuma longa, ginger, grapes, broccoli or bovine milk, contain exosomes and exosome-like nanoparticles that can relieve intestinal inflammation ( Figure 3). Exosome-like nanoparticles carrying curcumin and derived from Curcuma longa limit DSS-induced colitis development in mice by suppressing NF-κB activation and by increasing Treg cell and regulatory dendritic cell expansion in the colonic mucosa [113]. Broccoli-derived nanoparticles protect mice from DSS-induced colitis by triggering the activation of adenosine monophosphate-activated protein kinase (AMPK) in dendritic cells, thus preventing dendritic cell activation and inducing tolerant dendritic cells [102]. Exosome-like nanoparticles purified from grape juice, when orally administrated to DSS-treated mice, are captured by LGR5 + intestinal stem cells and promote proliferation of these cells, thus enhancing epithelial tissue renewal and protecting mice from DSS-induced colitis [114]. Moreover, nanoparticles derived from grapefruit are taken up by intestinal macrophages and improve experimental colitis in mice by upregulating the expression of heme oxygenase-1 (HO-1) and inhibiting the production of IL-1β and TNF-α [115]. Mass spectrometry analysis of these grapefruit-derived exosomes revealed that they carry numerous proteins involved in the regulation of carbohydrates and lipid metabolism, however, their impact on macrophage physiology remains to be investigated [115]. Exosome-like nanoparticles derived from ginger are mostly captured by intestinal epithelial cells and macrophages [116]. In vivo, these vesicles alleviate DSS-induced colitis, promote intestinal healing by increasing survival and proliferation of intestinal epithelial cells, limit pro-inflammatory cytokine secretion and increase the release of anti-inflammatory cytokines, suggesting their potential in the prevention and treatment of IBD [116]. Finally, exosomes derived from bovine milk increase the levels of Muc2, RegIIIγ, Myd88 and Gata4 gene expression, as well as IgA and sIgA levels in the intestine, thus playing a role in maintaining the integrity of the mucus layer and intestinal barrier function in mice [117].
Nanoparticles derived from food products have also an impact on intestinal microbiota. Indeed, treatment with nanoparticle curcumin increases the abundance of butyrate-producing bacteria and fecal butyrate level in both the inflamed colon and normal colon in mice [113]. Similarly, bovine milk-derived nanoparticles alter gut microbiota composition and modulate their production of some short-chain fatty acids in murine gastrointestinal tract, enhancing epithelial barrier [117]. Finally, ginger-derived exosome-like nanoparticles are preferentially captured by Lactobacillaceae in a nanoparticle-lipid dependent manner and contain miRNAs that modulate bacterial gene expression [118]. Among them, ginger nanoparticle-derived mdo-miR7267-3p targets a monooxygenase (ycnE) of Lactobacillus rhamnosus (LGG), leading to the increased expression of indole-3-carboxaldehyde (I3A) [118]. I3A acts as a ligand for aryl hydrocarbon receptor and induces the production of IL-22 in the colon of DSS-treated mice, thus improving the maintenance of intestinal barrier function and decreasing the severity of experimental colitis [118]. These findings suggested that food product-derived nanoparticles may be used to target specific microbiota components and alleviate intestinal inflammation. Extracellular vesicles derived from intestinal parasites could allow the development of new therapeutic strategies for IBD ( Figure 3). Indeed, some intestinal parasites, and more particularly hookworms may have immunosuppressive properties and have been already used in clinical trials for IBD treatment [119,120]. Exosome-like extracellular vesicles released by the rodent intestinal parasite Nippostrongylus brasiliensis contain proteins and miRNAs potentially carrying immunomodulatory properties. Intraperitoneal injection of these vesicles to mice treated with TNBS decreases the level of the pro-inflammatory cytokines IL-6, IL-1β, IFN-γ and IL-17a and increases the level of the anti-inflammatory cytokine IL-10 in colonic tissue, thus preventing colitis development [121]. Parasite-derived extracellular vesicles could thus be used as a therapy for the treatment of IBD [121].

Exosomes and Administration of Therapeutic Molecules
Exosomes are resistant to lysis by complement [122] and by RNases [5], which increases their stability in vivo by protecting their content. Therefore, the use of exosomes for the administration of therapeutic molecules has been considered [123]. Several strategies allow the incorporation of therapeutic molecules into exosomes, such as the direct modification of exosomal content by direct incubation, electroporation, transfection, ultrasound treatment, exosomal membrane permeabilization with saponins, sonication or freeze-thaw cycles [124]. To date, in the context of bacterial infection, only the antibacterial compound linezolid has been successfully incorporated in naïve exosomes by direct incubation of these vesicles with the molecule and has been shown to exert its effect in vivo [125]. The addition of a therapeutic molecule in exosomes can also be obtained by modifying the genome of exosome-donor cells to induce the expression and capture of the therapeutic compound in the secreted exosomes. This strategy has been successfully employed in the context of IBD. Exosomes secreted by BMDC genetically modified to express the gene encoding TGF-β1 (TGF-β/BMDC) were able to increase Treg cells and suppress Th17 response in DSS-treated mice, thus protecting against experimental colitis [100]. TGF-β was detected in exosomes released by TGF-β/BMDC but not in exosomes from non-modified BMDC. The intravenous administration of 10 µg of exosomes containing TGF-β to mice before DSS treatment helps to prevent weight loss, to reduce intestinal bleeding and disease activity index, while the administration of TGF-β alone does not confer any protection against DSS-induced colitis [100]. This suggested that the protective effects of exosomes on DSS-induced colitis could probably come from the increased stability of TGF-β in exosomes [100]. However, although able to delay the development of colitis, exosomes containing TGF-β did not inhibit the progression of colitis after disease onset [100].

Conclusions
A growing amount of evidence suggests that exosomes play a role in bacterial infection as well as in IBD progression, and a better understanding of the underlying mechanisms may provide new insights for de development of future diagnosis and therapeutic strategies. The presence of PAMPs and other immunomodulatory molecules in exosomes and the ability of these vesicles to modulate immune responses raise the potential of development of exosome-based "vaccines" for infectious diseases. Similarly, exosomes derived from different mammalian cell types, from parasites but also from edible plants, have exerted the potential to limit inflammation associated with IBD. Moreover, the stability of exosomes in various body fluids and the presence of some disease-associated components in these vesicles make the analysis of exosomal composition a promising strategy to identify minimally-or non-invasive biomarkers for diseases. Finally, the stability and packaging properties of exosomes point to their use as vectors of therapeutic molecules. However, further investigations are required to better understand the role of these vesicles in infection and IBD, their potential use as a diagnostic tool, to manage the incorporation of therapeutic molecules into these vesicles, to identify the most efficient targeting strategy and to evaluate the safety of these drug-loaded vesicles in vivo.

Conflicts of Interest:
The authors declare no conflict of interest.