(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Vaccination against microbiota motility protects mice from the detrimental impact of dietary emulsifier consumption [1]
['Melissa C. Kordahi', 'Inserm', 'Team', 'Mucosal Microbiota In Chronic Inflammatory Diseases', 'Cnrs Umr', 'Université Paris Cité', 'Paris', 'Clara Delaroque', 'Marie-Florence Bredèche', 'Robustness']
Date: 2023-09
Dietary emulsifiers, including carboxymethylcellulose (CMC) and polysorbate 80 (P80), perturb gut microbiota composition and gene expression, resulting in a microbiota with enhanced capacity to activate host pro-inflammatory gene expression and invade the intestine’s inner mucus layer. Such microbiota alterations promote intestinal inflammation, which can have a variety of phenotypic consequences including increased adiposity. Bacterial flagellin is a key mediator of emulsifiers’ impact in that this molecule enables motility and is itself a pro-inflammatory agonist. Hence, we reasoned that training the adaptive mucosal immune system to exclude microbes that express flagellin might protect against emulsifiers. Investigating this notion found that immunizing mice with flagellin elicited an increase in mucosal anti-flagellin IgA and IgA-coated microbiota that would have otherwise developed in response to CMC and P80 consumption. Yet, eliciting these responses in advance via flagellin immunization prevented CMC/P80-induced increases in microbiota expression of pro-inflammatory agonists including LPS and flagellin. Furthermore, such immunization prevented CMC/P80-induced microbiota encroachment and deleterious pro-inflammatory consequences associated therewith, including colon shortening and increased adiposity. Hence, eliciting mucosal immune responses to pathobiont surface components, including flagellin, may be a means of combatting the array of inflammatory diseases that are promoted by emulsifiers and perhaps other modern microbiota stressors.
Funding: This work was supported by a Starting Grant from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No. ERC-2018-StG- 804135 to BC), a Chaire d’Excellence from IdEx Université de Paris - ANR-18-IDEX-0001 to BC, an Innovator Award from the Kenneth Rainin Foundation to BC, an award from the Fondation de l'avenir (AP-RM-21-032 to BC), ANR grants EMULBIONT (ANR-21-CE15-0042-01 to BC) and DREAM (ANR-20-PAMR-0002 to BC) and the national program “Microbiote” from INSERM to BC. MK is supported by a Postdoc Fellowship from the Fondation pour la Recherche Médicale (FRM). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Based on these previous observations, we hypothesized here that inducing a mucosal flagellin-specific IgA response through purified flagellin immunization can prevent against dietary emulsifiers–induced detrimental consequences. We observed that flagellin immunization is sufficient to fully prevent emulsifiers-induced alterations in microbiota composition and localization. Furthermore, such immunization efficiently protected against dietary emulsifiers–induced low-grade intestinal inflammation and metabolic dysregulation. Hence, the protective potential of flagellin immunization supports the central role played by this bacterial appendix in promoting microbiota encroachment and downstream detrimental consequences in a way that can be used to combat modern dietary stressors known to perturb host–microbiota interactions.
It has also been observed that levels of bacterial flagellin—the main component of bacterial flagella—are usually low in a healthy intestine and increased in an inflamed microenvironment, such as in IBD [ 10 – 12 ]. In mice lacking the flagellin receptor Toll-like receptor 5 (TLR5), a loss of flagellin-specific immunoglobulins (Igs) response is associated with an increased proportion of flagellated bacteria in the intestinal tract in a way that associates with microbiota encroachment [ 13 ]. Importantly, previous work demonstrating that anti-flagellin Igs can directly down-regulate/shutdown flagellar gene expression and motility by select bacteria suggest a direct relationship between intestinal anti-flagellin and microbiota-derived flagellin expression [ 14 ]. Well aligned with this concept, inducing an intestinal flagellin-specific IgA response decreased levels of flagellated bacteria, reducing microbiota encroachment, which altogether protects against experimentally induced severe and low-grade inflammation [ 6 ]. Interestingly, the beneficial impact of such anti-flagella adaptative immune response appears to be important only before disease initiation, with the observation that established chronic intestinal inflammation associates with nonprotective immune reactivity against flagella likely due to microbiota breaching the epithelial lining [ 15 , 16 ].
The intestinal tract is colonized by a large and diverse collection of microbes referred to as the gut microbiota [ 1 ]. Under physiological conditions, the intestine is protected from its microbiota by a multilayered mucus structure that covers the intestinal surface, thus allowing the vast majority of gut bacteria to be kept at a safe distance from the intestinal epithelial lining [ 2 ]. We and others have reported that select dietary emulsifiers can promote low-grade inflammation (LGI) in the gut. Such LGI associates with, and may result from, emulsifiers changing gut microbiota. LGI can contribute to a variety of chronic diseases, including metabolic syndrome and can predispose to severe forms of inflammation including inflammatory bowel disease (IBD) [ 3 – 5 ]. We, for example, previously reported that 2 commonly used emulsifiers, namely, carboxymethylcellulose (CMC) and polysorbate 80 (P80), are sufficient to induce colitis in mice genetically prone to this disorder as well as to promote metabolic dysregulations in wild-type (WT) mice. Mechanistically, such alterations were associated with host–microbiota perturbations characterized by altered microbiota composition and function, especially the promotion of microbiota encroachment, hypothesized to be central for the observed host damages. Microbiota encroachment was previously reported to involve, at least in part, the flagella appendix expressed by select microbiota members [ 6 – 8 ] and responsible for bacterial motility [ 6 , 7 ]. We indeed previously observed a direct correlation between the severity of microbiota encroachment and the severity of emulsifiers-induced chronic intestinal inflammation in mice [ 3 ], as well as with the severity of type 2 diabetes in a human cohort [ 9 ]. Using gnotobiotic mouse models hosting a minimally complex intestinal microbiota, we observed that in the absence of microbiota encroachment, dietary emulsifiers are well tolerated and not associated with detrimental consequences on intestinal health. Altogether, these findings suggest that microbiota encroachment, likely mediated, at least in part, by flagella expression, is central for the subsequent development of chronic intestinal inflammation and metabolic dysregulations, in both preclinical models as well as in humans.
( A, B ) Body weight gain over time of mice immunized with either vehicle (PBS, A ) or purified flagellin ( B ). ( C ) Epididymal fat pad weight and ( D ) 15-hour fasting blood glucose level, measured at week 19. The underlying data for this figure can be found in S5 Data . Data are represented as means ± SEM. N = 4–5. For bar graphs, statistical analyses were performed using a one-way ANOVA and significant differences were recorded as follows: ns: non-significant, **p < 0.01, ****p < 0.0001. ANOVA, analysis of variance; CMC, carboxymethylcellulose; FliC, flagellin; P80, polysorbate 80.
We previously reported, in numerous models, that chronic intestinal inflammation can lead to metabolic dysregulations [ 3 , 17 ]. Hence, we next investigated the effect of dietary emulsifiers consumption on host metabolism, and the potential preventive effect of flagellin immunization. As reported in Fig 5 ( S5 Data ), we observed that either CMC or P80 consumption induced a greater body weight gain compared with the water-treated control group ( Fig 5A and S5 Data ). Such increased body weight gain was accompanied by significantly increased fat deposition and overnight fasting blood glucose levels in emulsifiers-treated mice compared to control mice ( Fig 5C and 5D and S5 Data ), further demonstrating that chronic consumption of dietary emulsifiers is sufficient to impair host metabolism. Immunization with purified flagellin fully abrogated various emulsifiers-induced alterations in metabolism, with fat deposition and overnight fasting blood glucose levels being fully normalized to water-treated control group ( Fig 5C and 5D and S5 Data ). Regarding overall weight gain, flagellin immunization was sufficient to prevent CMC-induced increased weight gain. Altogether, these data indicate that flagellin immunization prevents the chronic intestinal inflammation and improves some of its associated metabolic consequences that were otherwise observed in mice chronically exposed to dietary emulsifiers.
Another central detrimental impact of dietary emulsifiers consumption is their ability to induce microbiota encroachment, with the observation of microbiota colonization of the normally sterile inner mucus layer, which can be quantified by measuring epithelium–bacteria distance [ 9 ]. Such microbiota encroachment is hypothesized to play a central role in emulsifiers-induced chronic low-grade intestinal inflammation and metabolic dysregulations [ 3 ]. Based on the role played by flagella appendix in microbiota encroachment phenomenon [ 7 , 8 ], we examined microbiota encroachment via confocal imaging of Carnoy-fixed colon specimens. We recapitulated observations that both CMC and P80 consumption induce stark microbiota encroachment, with the average bacteria/epithelium distance being reduced from 16.50 μm in water-treated mice to 8.70 μm and 6.20 in CMC- and P80-treated mice, respectively ( Fig 3C and 3D and S3 Data ). Flagellin immunization fully protected against emulsifiers-induced microbiota encroachment, with bacteria/epithelium distances of 17.90 μm, 16.00 μm, and 19.90 μm being observed for water-, CMC-, and P80-treated groups, respectively ( Fig 3C and 3D and S3 Data ), further supporting the hypothesis that immune responses to flagellin protect against chronic intestinal inflammation.
( A-C ) Fecal bioactive levels of pro-inflammatory microbiota-derived molecules flagellin ( A ) and LPS ( B ) measured at week 17 via use of TLR5 and TLR4 reporter cells. ( C ) Colons, collected at week 19, were subjected to immunostaining paired with FISH followed by confocal microscopy analysis of microbiota localization. Distances of closest bacteria to IECs per condition over 5 HPFs per mouse. ( D ) Representative pictures obtained from 5 biological replicates. MUC2, green; actin, purple; bacteria, red; and DNA, blue. Scale bar, 50 μm. The underlying data for this figure can be found in S3 Data . N = 4–5. Statistical analyses were performed using a one-way ANOVA and significant differences were recorded as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ANOVA, analysis of variance; CMC, carboxymethylcellulose; FISH, fluorescent in situ hybridization; FliC, flagellin; HPF, high-powered field; IEC, intestinal epithelial cell; LPS, lipopolysaccharide; P80, polysorbate 80.
Beside microbiota composition, functional assessment appears warranted to deeply investigate microbiota alterations with potential downstream detrimental consequences on intestinal health. For example, alterations in microbiota—including those induced by CMC and P80—can increase the levels of pro-inflammatory microbiota-derived molecules such as flagellin and lipopolysaccharide (LPS) [ 3 , 4 , 17 ]. Thus, we next quantified fecal bioactive levels of these pro-inflammatory molecules via the use of TLR5 and TLR4 reporter cells. This approach revealed significantly elevated flagellin and LPS levels in mice consuming emulsifiers ( Fig 3A and 3B and S3 Data ). Animals that had been immunized against bacterial flagellin were fully protected against such emulsifiers-induced increase in microbiota pro-inflammatory potential, as presented in Fig 3A and 3B ( S3 Data ).
Bacterial DNA was extracted from feces collected at weeks 1 and 17 and subjected to 16S rRNA gene sequencing. ( A-C ) PcoA of the Bray Curtis distance matrix of microbiota assessed by 16S rRNA gene sequencing at week 1 ( A ) and week 17 ( B and C ). Each dot represents an individual animal and is colored by experimental group (in A and B : blue, water; orange, CMC; purple, P80; light blue, water-FliC; light orange, CMC-FliC; light purple, P80-FliC. In C : blue, PBS control groups; red, flagellin-immunized groups). ( D ) Bray Curtis distance separating mice between experimental groups at week 17. The underlying data for this figure can be found in S2 Data . Data are represented as means ± SEM. Statistical analyses were performed using a one-way ANOVA, and significant differences were recorded as follows: **p < 0.01, ***p < 0.001, ****p < 0.0001. ANOVA, analysis of variance; CMC, carboxymethylcellulose; FliC, flagellin; PCoA, principal coordinate analysis; P80, polysorbate 80.
The direct impact of dietary emulsifiers on the intestinal microbiota plays a central role in promoting bacterial encroachment, intestinal inflammation, and its downstream consequences [ 3 , 4 ]. Hence, we next examined the extent to which flagellin immunization might prevent emulsifiers-induced alterations in intestinal microbiota composition. Use of 16S rRNA gene sequencing followed by PCoA of the Bray Curtis distance revealed that mice included in the study had homogeneous baseline microbiota composition prior to the start of treatment (week 1, Fig 2A ). In contrast, such approach found that 10 weeks of exposure to CMC or P80 resulted in a clear treatment-based microbiota clustering (week 17, Fig 2B ), indicating that both CMC and P80 markedly impacted intestinal microbiota composition. Such observation was confirmed through Bray Curtis distance computing between groups, revealing a significant increase between emulsifiers-treated and water-treated mice ( Fig 2D and S2 Data ). Next, we observed that flagellin immunization is by itself also sufficient to clearly impact microbiota composition, with a distinct clustering being observed between immunized and nonimmunized mice ( Fig 2C ), with a PERMANOVA p-value of 0.001. Such microbiota composition investigation finally revealed that flagellin immunization is not sufficient to prevent emulsifiers-induced alterations in microbiota composition, with the observation of distinct clusterings ( Fig 2B ) and a significantly increased Bray Curtis distance between water- and P80-treated mice ( Fig 2D and S2 Data ). Altogether, these findings indicate that while flagellin immunization is sufficient to impact intestinal microbiota composition, it fails to prevent emulsifiers-induced microbiota alteration, suggesting that protection conferred by immunization on intestinal inflammatory tone and metabolic dysregulations does not solely rely on microbiota composition normalization.
We next performed cecal IgA+ and IgA− bacterial population sorting and 16S sequencing, as previously reported [ 6 ], in order to characterize the impact of flagellin immunization on emulsifiers-induced modulation of the IgA-coated microbial population. As presented in S3 Fig ( S7 Data ), such an approach showed a great range of IgA indices in all the tested experimental groups and without evident global effect of emulsifiers consumption nor flagellin immunization. Accordingly, principal coordinate analysis (PCoA) of Euclidean distances using computed IgA index (log (IgA + abundance / IgA − abundance)) revealed that both CMC and P80 consumption significantly impacted IgA coating of the intestinal microbiota, with clear distinct clustering between groups ( Fig 1E ). As presented in S4 Fig ( S8 Data ), both CMC and P80 consumption induced alterations in the IgA coating of numerous Clostridiales microbiota members. Such emulsifiers-induced alteration in the IgA-coated microbial population was mostly prevented by flagellin immunization, as indicated by the absence of treatment-based clustering ( Fig 1F ). Flagellin immunization impacted the IgA index from various members of the intestinal microbiota, including Lachnospiraceae, Ruminococcaceae, and Bacteroidaceae ( S5 Fig and S9 Data ). Thus, the impact of emulsifiers consumption on the IgA–microbiota interaction was not observed in immunized mice, suggesting that training the immune system to target flagellin prevented these compounds from destabilizing microbiota–immune system homeostasis.
( A-C ) Fecal levels of anti-flagellin IgA at weeks 5 and 17, with data being expressed as relative values compared to week 5 nonimmunized group, defined as 100%. ( D-F ) Cecal contents, collected at week 19, were sorted for IgA-positive and IgA-negative bacterial populations. DNA was extracted from sorted cells and subjected to 16S rRNA sequencing. ( D ) Bar graph representing the percentage of IgA-coated bacteria in the caecum content at week 19. ( E, F ) PCoA of the Euclidean distance, at week 19, using IgA indices with dots being colored by treatment ( E , water = blue; CMC = orange; P80 = purple, F, water-FliC = light blue; CMC-FliC = light orange; P80-FliC = light purple). The underlying data for this figure can be found in S1 Data . N = 4–5. For bar graphs, statistical analyses were performed using a t test and one-way ANOVA. For line charts, a two-way ANOVA or a mixed model was performed. Significant differences were recorded as follows: CMC vs. water, *p < 0.05, **p < 0.01, ****p < 0.0001. ANOVA, analysis of variance; CMC, carboxymethylcellulose; FliC, flagellin; IgA, immunoglobulin A; PCoA, principal coordinate analysis; P80, polysorbate 80.
Discussion
Microbiota dysbiosis is thought to play a central role in driving intestinal inflammation and, consequently, numerous chronic diseases with an inflammatory component [1]. Features of microbiota dysbiosis include alterations in species composition with an enrichment in flagellated bacteria, which can, for example, result from increases in Gamma-Proteobacteria, including motile pathobiont Escherichia coli strains [19,20] but can also result from other classes of bacteria, especially Firmicutes, up-regulating motility-related gene expression [14]. These disease-associated microbiotas expressing high levels of flagellin are also characterized by an increased capacity to penetrate the normally sterile inner mucus layer, a feature referred to as microbial encroachment. Such encroaching microbiotas are thought to play a significant role in driving gut inflammation, with, for example, the previous observation of a positive correlation between microbiota encroachment and the severity of intestinal inflammation in mice models as well as with the severity of metabolic dysregulation in humans [3,21]. While there is likely a broad array of underlying factors inducing microbiota dysbiosis and encroachment, various evidence supports a major role for environmental (i.e., nongenetic) determinants. For instance, we and others have previously shown that consumption of dietary emulsifiers can induce altered microbiota composition and encroachment, resulting in colitis in genetically susceptible mice and in LGI and metabolic syndrome in WT mice [3,4].
In this study, we importantly report that flagellin immunization is sufficient to prevent numerous detrimental consequences normally induced by dietary emulsifiers consumption, while such immunization did not appear sufficient to fully prevent emulsifiers-induced compositional changes in the microbiota. Hence, what we believe to be important regarding immunization-induced microbiota modulation is related to functional aspects, including microbiota localization and pro-inflammatory potential, rather than compositional aspects. Moreover, recent work from Clasen and colleagues report a high heterogeneity in the ability of a given flagellin to activate the TLR5 receptor, suggesting that a given microbiota could have a relatively high proportion of microbiota members expressing flagellin and yet have a weak TLR5 activation potential [22]. Hence, precise identification of the impact of the immunization protocol on microbiota composition, and more specifically on the flagellated bacterial population, will require follow-up studies. Future work will, for example, involve laser-capture microdissection-based collection of the inner intestinal mucus layer, in order to focus on mucus-associated microbiota, followed by shotgun metagenomics and identification of the flagellated bacterial population. In the current study, together with a decrease in flagellin bioactive levels in immunized mice, we also observed a decrease in fecal bioactive LPS levels, and the exact underlying mechanism will also deserve further investigations.
While the intestinal tract possesses various innate and adaptive immune mechanisms to keep the inner mucus layer sterile, for example, through the secretion of various antimicrobial peptides, our group has previously indicated that adaptive immunity, in particular mucosal production of flagellin-specific IgA, plays a key role in keeping motile bacteria in check [6]. We have indeed observed that eliciting anti-flagellin antibodies via immunization is an efficient strategy to protect against colitis as well as against diet-induced obesity [6]. Although antibodies directed against bacterial flagella could be highly species specific, many anti-flagellin antibodies can recognize highly conserved flagellin epitopes, such that inoculation of mice with Salmonella-derived flagellin generate antibodies that exhibit considerable cross-reactivity with other flagellins such as Clostridia flagellin [23]. Moreover, we have previously reported that flagellin derived from various microbiota members can bind TLR5 receptor as well as be recognized by antibodies raised against recombinant flagellin peptides from Firmicute Roseburia hominis [14]. However, the cross-reactivity of Salmonella-derived flagellin antibodies against various microbiota members, for example, Bacteroidetes species, deserves further investigations. Hence, even if the select microbiota members encroaching upon the epithelium upon dietary emulsifiers consumption have not yet been identified, and regardless of the fact that they will likely be different in various hosts, we hypothesize that eliciting a robust anti-flagellin response will nonetheless provide a degree of protection against the deleterious effects of chronic exposure to dietary emulsifiers.
We specifically observed that while fecal flagellin levels were increased by emulsifiers exposure, immunization with purified flagellin fully prevented such effect. More importantly, the impact of both CMC and P80 on microbiota localization, fecal IgA response, intestinal inflammatory tone, and metabolism was all prevented in immunized mice. Thus, flagellin immunization appears as an efficient way to prevent the detrimental consequences of emulsifiers consumption. A possible explanation for such observations is that the primary mechanism of action for flagellin immunization is through the stabilization of the anti-flagellin IgA response normally produced while consuming emulsifiers, hence prohibiting the microbiota from penetrating the inner mucus layer coating the colon and activating pro-inflammatory genes. Moreover, while our IgA-Seq approach performed herein clearly suggests that flagellin immunization is sufficient to prevent emulsifiers-induced alteration in the IgA-coated microbiota composition, we do not know yet if this is occurring through modulation of the global microbiota composition or through targeted modulation of flagellin expression by various microbiota members.
Previous quantitative reverse transcription PCR (qRT-PCR)–based analysis in the colon of mice consuming dietary emulsifiers revealed only subtle modifications of select cytokines, suggesting the induction of only low-grade chronic intestinal inflammation. We suspect that only few intestinal immune cell populations are impacted, in the gastrointestinal tract, by dietary emulsifiers consumption. Hence, future work will involve performing single-cell RNA-seq analysis in order to finely characterize emulsifiers-induced alteration of the intestinal immune landscape. Moreover, intraperitoneal injection of flagellin also leads to pro-inflammatory response through activation of the innate immune system via TLR5 and/or NLRC4 in the peritoneum but also likely systemically, which could play a role in the observed prevention of detrimental phenotypes following administration of emulsifiers. Of note, we previously reported that in μMT mice unable to produce antibodies owing to their lack of mature B cells, flagellin immunization regimen no longer results in the beneficial modulation of the intestinal microbiota, hence arguing that a significant portion of flagellin immunization’s impact on the microbiota is mediated by anti-flagellin antibodies [6]. While we anticipate that the same observation will hold true concerning the protection of emulsifiers-treated mice, such aspect will deserve further investigations. Moreover, irrespective of the exact underling mechanism, more targeted immunization of the intestinal mucosa, for example, through targeted delivery of recombinant flagellin, should now be deployed.
To conclude, our data presented here suggest that antibacterial immunization could be an efficient way to prevent microbiota encroachment in a way that will subsequently prevent the development of chronic debilitating diseases. Of note, the regimen used herein, with repeated injection of purified flagellin, is not applicable in clinical settings. However, based on the presence of basal levels of anti-flagellin antibodies in humans, we speculate that humans might exhibit a memory-type response to exogenously administered flagellin, which will likely make targeted mucosal immunization with recombinant flagellin effective [24,25]. Additionally, as flagellin immunization was observed to be sufficient to protect against subsequently administered dietary emulsifiers, we have yet to investigate the therapeutic potential of such immunization regimen in established chronic low-grade intestinal inflammation. However, our results herein suggest that increased fecal flagellin in response to emulsifiers consumption in nonimmunized mice is also associated with an increase in anti-flagellin IgA (Fig S2 and S6 Data), which does not appear to be sufficient to prevent the deleterious effects induced by emulsifiers consumption. Similarly, Crohn’s disease patients harbor high levels of anti-flagellin Igs, which are not sufficient to be protective [26], suggesting that anti-flagellin Igs might be protective only when elicited prophylactically.
The various points raised above highlight the need for extensive preclinical development in order to harness the use of microbiota-derived antigens to vaccinate against modern chronic diseases involving alterations in the intestine–microbiota interaction. Nonetheless, our results suggest that this approach has high potential to prevent chronic inflammatory diseases. Should the elicitation of flagellin-specific mucosal antibodies keep motile bacteria in check, prevent microbiota encroachment, and result in a generally less pro-inflammatory microbiota in humans, we submit that this approach might be an innovative prophylactic/therapeutic venue for the protection against a broad array of inflammatory diseases including IBD and metabolic syndrome.
[END]
---
[1] Url:
https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002289
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
