The term IBD mainly refers to UC and Crohn’s disease. These are chronic inflammatory disorders of the intestinal tract that may cause life-threatening complications and have an increasing incidence worldwide. They are multifactorial diseases, and although their etiology remains poorly understood, they involve a dysregulated immune response against commensal gut microbes in genetically susceptible individuals with impaired intestinal epithelial integrity (Zhang and Li, 2014). Moreover, several studies have reported that IBD patients have lower microbiota diversity than healthy subjects, and an altered intestinal microbiota balance, which is known as dysbiosis (Ott et al., 2004; Sokol et al., 2008; Manichanh et al., 2012).

Conventional IBD therapies are complex, and not all patients show complete beneficial effects when treated with any of the drugs currently available to manage these intestinal conditions, including salicylates, corticoids, immunosuppressants and biological agents (Ko and Auyeung, 2014). Since dysbiosis is common in IBD, various therapeutic approaches targeting the modulation of the gut microbiota have been explored (Qiao et al., 2016). In this context, many studies have evaluated the therapeutic potential of certain bacteria, including commensal and probiotic strains, to ameliorate IBD in clinical trials (Fedorak, 2010; Wasilewski et al., 2015) or in animal models of colitis (Ewaschuk et al., 2008; Garrido-Mesa et al., 2011; Shen et al., 2012; Kang et al., 2013; Martín et al., 2015; Souza et al., 2016). Overall, these studies have shown the ability of these bacteria to exert beneficial effects on parameters related with gut function, including improvement of gut permeability, reduction of inflammatory cytokine production and/or release, and amelioration of the histological alterations observed in these inflammatory conditions. In some cases, these effects may be mediated, at least in part, by bacterial secreted factors (Ewaschuk et al., 2008; Martín et al., 2015) or by released membrane vesicles (Shen et al., 2012; Kang et al., 2013).

The probiotic EcN positively affects gastrointestinal homeostasis and microbiota balance. In fact, various clinical trials have reported its therapeutic benefits in inducing and maintaining remission of UC (Scaldaferri et al., 2016). EcN shows similar efficacy to the aminosalicylate mesalazine (Losurdo et al., 2015), which supports its use in the treatment of human UC (Floch et al., 2011). Similarly, the intestinal anti-inflammatory effects of this probiotic have been reported in experimental models of colitis in mice and rats (Grabig et al., 2006; Ukena et al., 2007; Arribas et al., 2009; Garrido-Mesa et al., 2011; Souza et al., 2016). Several mechanisms have been reported to be involved in the beneficial effects of this probiotic in IBD, such as its ability to modulate the host immune response toward an anti-inflammatory balance (Trebichavsky et al., 2010). In addition, EcN reinforces the intestinal epithelial barrier by strengthening TJs between intestinal epithelial cells (Ukena et al., 2007; Zyrek et al., 2007; Hering et al., 2014) and increasing the expression of antimicrobial factors such as microcins (Sassone-Corsi et al., 2016) and β-defensin-2 (Schlee et al., 2007). It is important to note that these effects were mainly established when live probiotic suspensions were administered. There is little reported information describing the bacterial factors that could mediate these effects. In this context, membrane vesicles released by commensal bacteria are currently receiving great attention as key players in signaling processes in the intestinal mucosa (Kaparakis-Liaskos and Ferrero, 2015; Olsen and Amano, 2015). The release of bacterial vesicles provides a mechanism for the delivery of microbial proteins and active compounds directly to the host, thus avoiding direct intercellular contact. Recently, it has been reported that OMVs released by the probiotic EcN enter intestinal epithelial cells via clathrin-mediated endocytosis, and thereby can trigger host immune and defense responses (Cañas et al., 2016). In fact, in vitro and ex vivo studies have shown that EcN OMVs induce the expression of antimicrobial peptides and modulate the cytokine/chemokine response of gut epithelial and immune cells (Fábrega et al., 2016). In addition, in intestinal epithelial cell lines, these vesicles promote upregulation of the TJ-proteins ZO-1 and claudin-14, but down-regulation of claudin-2, thus reinforcing intestinal barrier functions and reducing gut permeability (Alvarez et al., 2016).

Probiotics are Generally Regarded As Safe (GRAS), yet some concerns about the potential risk associated with their use should be considered, especially in immunocompromised individuals and neonates. In conditions of altered intestinal barrier, the probiotic may cause invasive infections leading to sepsis (Ledoux et al., 2006; Bleich et al., 2008; Gronbach et al., 2010). In this context, the use of probiotic- derived bioactive factors, such as OMVs, could potentially have healing benefits, while avoiding the risks associated with the administration of live bacteria.

Based on previous in vitro studies (Alvarez et al., 2016; Fábrega et al., 2016) we hypothesized that EcN OMVs may contribute to the immunomodulatory and barrier strengthening effects of EcN in the gut. The aim of the present study was to evaluate the effects of OMVs isolated from EcN in the dextran sodium sulfate (DSS) model of mouse colitis, an experimental model that is widely used in preclinical assays to study new potential treatments for human IBD (Claes et al., 2011).

The etiology of IBD has not been completely elucidated to date. However, it has been proposed that intestinal dysbiosis is crucial, since it can promote an inappropriate immune response in genetically susceptible individuals, leading to the inflammatory status that characterizes these intestinal conditions (Becker et al., 2015). Accordingly, restoration of intestinal microbiota composition through the administration of probiotics could have beneficial effects against the inflammatory response, as has been shown both in experimental models and in human IBD (Claes et al., 2011; Bellaguarda and Chang, 2015). EcN is one of the probiotics reported to have a beneficial effect in IBD, and plays a prominent role in therapy for this group of intestinal conditions in humans. In fact, the 2011 update of the American Recommendations for Probiotic Use gave very strong “A” recommendations for its use to maintain remission in human UC (Floch et al., 2011). The therapeutic efficacy of EcN has been more recently supported in a systematic review and meta-analysis (Losurdo et al., 2015). Among the mechanisms that might be involved in the beneficial effects exerted by EcN, its immunomodulatory properties could play a key role. In the gut, the probiotic can counteract altered cytokine production by immune cells (Sturm et al., 2005; Guzy et al., 2008) and restore damaged epithelium through the modulation of TJ and zonula occludens proteins (Ukena et al., 2007; Zyrek et al., 2007). Although direct interaction between the bacteria EcN and the immune or epithelial cells could be responsible for the observed effects, other mechanisms involving the production and release of secreted bacterial factors cannot be ruled out. Among these secreted factors, membrane vesicles play a key role in bacteria–host communication, allowing the delivery of effector molecules upon interaction and internalization into the host cells. In fact, OMVs enclose many of the biological components that can be found in the bacterium and, specifically, ligands of pattern recognition receptors such as LPS (TLR-4 ligand), peptidoglycan (NOD-1/NOD-2 ligand) or DNA (TLR-9 ligand), among others. These receptors, which are expressed in epithelial and immune cells, are key components of innate immunity as they sense gut microbes and trigger appropriate immune responses. Therefore, bacteria-released OMVs can mediate most of the effects of the bacterium. Remarkably, OMVs as non-replicative bacterial forms further reduce the low risk of adverse events associated with the administration of the probiotic (Bleich et al., 2008; Gronbach et al., 2010; Olier et al., 2012).

Previous studies performed in human cell lines and human colonic explants have shown the immunomodulatory and barrier strengthening potential of EcN OMVs (Alvarez et al., 2016; Fábrega et al., 2016). Since orally administered live EcN suspensions ameliorate DSS-induced colitis in mice (Grabig et al., 2006; Garrido-Mesa et al., 2011; Olier et al., 2012), we sought to explore whether EcN OMVs could prevent colonic damage in the same experimental model of intestinal inflammation. This is a well-established animal model that resembles human UC, and it has been validated for the translation of mouse data to humans (Jiminez et al., 2015).

The present study demonstrates that oral administration of OMVs isolated from EcN has intestinal anti-inflammatory properties in the DSS experimental model of colitis, as previously reported when a suspension of the viable probiotic was administered (Garrido-Mesa et al., 2011). The beneficial effects of EcN OMVs were already observed during the course of the experiment. In fact, reduction in weight loss and DAI evolution in mice treated with OMVs, compared to the control colitic group, clearly evidenced amelioration of the impact of the colitic process and improvement in the health status of mice.

Likewise, OMVs administration to colitic mice significantly reduced colonic damage. This was shown macroscopically by a significant reduction in the colonic weight/length ratio, which indicated amelioration of the tissue oedema that characterizes this colonic inflammatory process, and microscopically with recovery of the altered tissue histology associated with DSS colitis. The OMV-mediated anti-inflammatory effect was clearly associated with an improvement in the altered immune response accompanying DSS-induced colonic damage, thus mimicking the immunomodulatory properties described for this probiotic both in vivo and in vitro (Kruis et al., 2004; Grabig et al., 2006; Arribas et al., 2009; Garrido-Mesa et al., 2011; Scaldaferri et al., 2016). It is known that the acute mucosal inflammation induced by DSS results in activation of several cells involved in the intestinal innate immune response, including epithelial cells, macrophages and dendritic cells, leading to upregulated expression and/or production of pro-inflammatory cytokines, such as IL-1β, TNF-α and IL-6. This subsequently facilitates the sustained inflammatory response associated with an imbalance in Th1/Th17 and Treg cell responses, which results in increased expression and/or production of other cytokines such as IFN-γ, IL-12 (Th1) and IL-17 (Th17), together with a reduction in the anti-inflammatory cytokine IL-10 (Treg) (Neurath, 2014). Our results show that administration of EcN OMVs ameliorates the altered cytokine profile observed in colitic mice treated with DSS.

Different mechanisms could be involved in these immunomodulatory properties. Of particular significance is the ability to restore intestinal epithelial integrity. In this context, it has been reported that the intestinal anti-inflammatory properties of EcN are related to its ability to reinforce TJs of intestinal epithelial cells, thus counteracting the altered permeability status in the intestinal epithelium that characterizes intestinal inflammation (Ukena et al., 2007). Notably, impairment of the epithelial barrier function has been proposed as one of the first events to occur in intestinal inflammation. This facilitates the access of antigens from the intestinal lumen and triggers the exacerbated immune response. Consequently, rapid promotion of mucosal healing has been considered crucial in the management of intestinal inflammation (Dave and Loftus, 2012). Under conditions of intact epithelial barrier EcN OMVs (100 μg/ml) upregulate expression of claudin-14 and ZO-1 in T-84 and Caco-2 monolayers (Alvarez et al., 2016). However, oral administration of EcN OMVs does not prevent DSS-induced downregulation of ZO-1 mRNA. Thus, the effects previously observed in in vitro models of intact epithelial barrier are not appreciated in the in vivo model of experimental colitis. This suggests that different regulatory mechanisms could be activated by EcN OMVs in the presence of highly expressed inflammatory mediators. Notably, ZO-1 is a target of post-translational modifications that regulate the intracellular fate of this protein. Reinforcement of the epithelial barrier can also be triggered by stimuli that promote ZO-1 location at the cell boundaries, and hence its association with TJ structures. In this context, it has been described that TFF-3 promotes redistribution of ZO-1 from the cytoplasmic compartment to the intercellular junctions in Caco-2 cell monolayers, which results in an increased level of co-localisation with occludin. These regulatory effects were not accompanied by changes in ZO-1 or occludin protein levels (Buda et al., 2012). Expression of TFF-3 is downregulated in intestinal inflammation (Lozano-Pérez et al., 2014), as observed in the present study. In connection with this, here we show that EcN OMVs increase colonic expression of this bioactive peptide involved in epithelial protection and repair (Podolsky et al., 2009), and preserve the damaged colonic cytoarchitecture of the mucosa in colitic mice. In addition to beneficial effects on epithelial integrity restoration, increased expression of TFF-3 can help to ameliorate the altered immune response exerted by EcN OMVs in experimental colitis, since overexpression of this peptide has been reported to abolish the IL-1β-induced upregulation of the pro-inflammatory cytokines IL-6, IL-8 and TNFα (Lin et al., 2013). Another plausible mechanism exploited by EcN OMVs to protect intestinal barrier function in colitic mice is downregulation of MMP-9 expression. Upregulation of this gelatinase contributes to the pathogenesis of IBD by disrupting TJs between intestinal epithelial cells, leading to increased intestinal permeability and the subsequent pro-inflammatory effects (Nighot et al., 2015). Moreover, treatment of colitic mice with EcN OMVs tends to increase, or at least preserve, MMP-2 expression, thus protecting against DSS-associated intestinal barrier loss (Garg et al., 2009).

Other markers of inflammation and tissue damage that are over-expressed in DSS-induced colitis are COX-2 and iNOS (Kolios et al., 2004). These enzymes are in fact defense mechanisms against the injury, inflammation and infection that occur in pathological conditions, and are highly expressed at damaged sites. Increases in COX-2 levels are seen in chronic intestinal inflammation, which is one of the risk factors for colorectal cancer (Wang and DuBois, 2010). Expression of iNOS is induced by certain pro-inflammatory cytokines such as TNF-α or IFN-γ, as well as by bacterial products. Although increased NO production through upregulation of iNOS is part of the intestinal antibacterial response, excess NO has been associated with intestinal inflammation in IBD patients and attributed to the immune dysregulation that occurs in these intestinal disorders (Kolios et al., 2004). In fact, elevated iNOS expression has been demonstrated in clinical IBD, caused mainly by infiltrating macrophages in intestinal mucosa (Rachmilewitz et al., 1995; Palatka et al., 2005). In addition, excessive NO production mainly through iNOS has been correlated with tissue injury and mucosal lesions observed in both human IBD and DSS-induced colitic mice. Consequently, probiotics or natural compounds with proven efficacy in attenuating colitis in DSS-treated mice reduce iNOS expression (Camuesco et al., 2004; Garrido-Mesa et al., 2015; Ren et al., 2015; Senol et al., 2015; Utrilla et al., 2015). EcN is among these probiotics (Garrido-Mesa et al., 2011). Here we show for the first time the ability of OMVs from a probiotic strain, specifically EcN, to counteract DSS-induced expression of both iNOS and COX-2. This correlates with reduced expression of the pro-inflammatory cytokines TNF-α and IFN-γ, and lower inflammatory cell infiltration in OMVs-treated mice.

At present, the molecules in EcN OMVs responsible for the anti-inflammatory and/or barrier protective effects remain unknown. Olier et al. (2012) reported that the genotoxic polyketide colibactin is required for the in vivo anti-inflammatory effects of EcN. Moreover, deficiency in colibactin biosynthesis leads to exacerbation of colitis severity in DSS-treated mice. These facts lead authors to considered colibactin as an immunomodulin. Whether this bacterial effector is exported and delivered into the host cells remains elusive. As EcN OMVs exert anti-inflammatory effects in experimental colitis, we may speculate that colibactin could be secreted through OMVs, at least in part. However, the contribution of other specific probiotic OMV-associated factors cannot be ruled out. Regarding mediators of intestinal barrier protection, it has been shown that the secreted protein TcpC contributes to the in vitro strengthening ability of EcN when the epithelial barrier is intact (Hering et al., 2014). However, the effect of TcpC is not associated with OMVs (Alvarez et al., 2016). Thus, OMV-linked factors other than TcpC should contribute to the modulation of mucosal healing markers like TFF-3 and MMP-9, observed in this experimental murine colitis model.

Recently acquired knowledge shows that microbiota plays an important role in the pathogenesis of IBD; dysbiosis is a common feature of these inflammatory conditions. Nowadays, potential clinical applications of probiotics are being explored to restore microbiota imbalances and take advantage of their health-promoting capacities. However, the translation of probiotics or microbiota-based drugs to human health-care requires deep knowledge of the molecular mechanisms involved in microbiota–host interactions (Jia et al., 2008; Shanahan, 2011). In this regard, bacterial vesicles are key players in signaling processes in the intestinal mucosa, as they act as a secretion and delivery pathway for selected bacterial proteins and active compounds directly to the host cells. Among probiotics with promising results in IBD clinical assays and experimental colitis models, EcN has beneficial effects on the remission of UC. Our study proves that oral administration of OMVs isolated from this probiotic has intestinal anti-inflammatory effects in the DSS experimental model of colitis, similarly as previously reported for the administration of viable probiotic suspensions. Therefore, OMVs could represent a safe probiotic-derived strategy targeting intestinal inflammatory processes as they are free from bacteria and can elicit the same effect on gastrointestinal health as the probiotic itself.

LB, JG, JB and RG contributed to the conception and design of the study. M-JF and AR-N were the responsible to perform all the experiments and analysis of data. JG, JG-M, and FA contributed to the histological studies. JG, LB, JB, and RG supervised the work, interpreted the data and wrote the drafted manuscript. All the authors revised the manuscript and approved the final version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.