An array of chronic inflammatory disorders, including inflammatory bowel disease (IBD) and metabolic syndrome, are associated with disturbances in the intestinal microbiota and a breakdown in the normally beneficial host-microbiota relationship. Perturbations in the composition of microbiota and/or microbial gene expression not only mark the disease but have been shown to play a key role in driving it. One common feature of inflammation-associated microbiota is increased level of flagellin, a protein component of the flagellum structure that allows single celled organisms to be mobile. This flagellin protein has been shown to activate pro-inflammatory gene expression via TLR5 and the NLRC4 inflammasome. Its elevated levels of expression were found to reflect increased levels of motile bacteria in the gut that hold a high ability to penetrate the mucus layer that serves to protect the host against microbial onslaught. Host organism actively protects itself against microbial invasion, through coating of gut bacteria by flagellin-specific IgA, which in turn results in suppression of levels of flagellated bacteria and reduction of microbiota encroachment.

In this study the authors aimed to investigate whether boosting levels of mucosal flagellin-specific IgA by direct administration of purified flagellin, may help to keep flagellated bacteria in check and serve as a general vaccine against subsequent development of the chronic IBD and obesity.

The researchers have tested several routes and regimens of delivering of the flagellin to the gut. They showed that the best immuno-response was observed when mice were repeatedly injected with highly conserved monomeric Salmonella-derived flagellin (FilC) via intraperitoneal injections over the course of ten weeks. This regimen resulted in robust levels of systemic (serum) and mucosal (fecal) anti-flagellin IgA and IgG antibodies. The antibodies were observed in the lumen of various regions of the gut including the duodenum, jejunum, ileum and colon. The resultant immune responses depended on flagellin's ability to activate TLR5 and NLRC4 receptors, as the KO mice missing the receptors showed a near complete loss of anti-flagellin antibodies in their serum samples. In addition, anti-flagellin antibodies remained elevated three months after completion of the immunisation regimen indicating long-lasting response.

To assess the impact of the flagellin immunisation regimen on the bacteria gut composition, fecal samples were analysed using Illumina sequencing of the V4 region of 16S rRNA genes. Analysis showed that flagellin immunisation led to a strong alteration of microbiota composition in treated animals compared to PBS injected controls. Interestingly, the immunisation was sufficient to significantly increase the proportion of fecal IgA-coated bacteria. This strongly suggested that anti-flagellin IgAs, induced by immunisation with highly conserved flagellin (FliC) protein, were binding to variety of microbial members, confirming the general immune system-inducing properties of the FliC immunogen.

The level of flagellin was also reduced in feces from flagellin-immunised mice compared to PBS controls. This result suggested less bacterial load and/or less-motile microbiota being present in the gut of treated animals. Indeed, bacterial loads were substantially reduced in the fecal samples of immunised mice and the bacteria present in the gut were spatially localised further away from the epithelium of immunised animals than in the guts of control animals. This suggested that the microbiota present displayed less mobility and therefore possessed less potential to encroach upon the host epithelial cells and initiate inflammation.

The observed changes in localisation within the mucosal-epithelial layer of the gut resulted from anti-flagellin antibodies, rather than the flagellin protein itself, as immunised mice, which were not able to produce antibodies due to lack of mature B cells (µMT strain), did not display an increase in bacterial-epithelial distance.

Microbiota encroachment is not only a general feature of IBD, but also a metabolic syndrome, including obesity in humans. Therefore, the authors aimed to test whether the FliC immunisation regimen tested here which led to the increase of levels of anti-flagellin antibodies in serum and in mucosa and prevention of microbial encroachment in the gut, was able to protect against IL-10 deficiency-induced colitis and/or ameliorate diet-induced obesity in mice.

Indeed, the flagellin-immunised mice induced to produce a colitis phenotype by weekly injections of a IL-10 receptor-neutralising antibody, fared better on colitis disease scores including less colon shortening, increased fat tissue weight and smaller spleen weights, when compared to PBS-treated controls. Flagellin-immunisation had a significant impact on WT animals in protecting them against intestinal inflammation, whereas the protective effects were very modest in B cell deficient µMT mice.

Similarly, FliC immunised mice kept on a high fat diet gained less weight and exhibited less adiposity compared to non-immunised animals. In addition, flagellin immunisation was associated with decreased intestinal inflammation, as revealed by a protection against colon shortening and spleen enlargement that normally results from a high fat diet. Correlation analysis showed that the load of anti-flagellin IgA in the fecal samples correlated with the degree of protection against adiposity (measured by weight of adipose tissue) and inflammation (measured by weight of spleen and length of colon). Interestingly, obese human subjects exhibit increased levels of fecal flagellin and reduced levels of fecal flagellin-specific IgA, relative to normal weight subjects.

In summary, the authors proposed that the administration of flagellin has the potential to stimulate the adaptive immuno-system to produce beneficial anti-flagellin antibodies that prevent an array of diseases associated with and driven by gut inflammation related to gut microbiota. Although the vaccination regimen tested here of 10 injections of flagellin might not be practical in humans, the observation that almost all humans exhibit considerable antibody responses to flagellin leads the authors to speculate that humans might exhibit a memory-type response to an initial flagellin treatment, which will make the approach more effective in humans than in mice.