Impacts of microbiome metabolites on immune regulation and autoimmunity

Summary

A vast number of studies have demonstrated a remarkable role for the gut microbiota and their metabolites in the pathogenesis of inflammatory diseases, including multiple sclerosis (MS). Recent studies in experimental autoimmune encephalomyelitis, an animal model of MS, have revealed that modifying certain intestinal bacterial populations may influence immune cell priming in the periphery, resulting in dysregulation of immune responses and neuroinflammatory processes in the central nervous system (CNS). Conversely, some commensal bacteria and their antigenic products can protect against inflammation within the CNS. Specific components of the gut microbiome have been implicated in the production of pro‐inflammatory cytokines and subsequent generation of Th17 cells. Similarly, commensal bacteria and their metabolites can also promote the generation of regulatory T‐cells (Treg), contributing to immune suppression. Short‐chain fatty acids may induce Treg either by G‐protein‐coupled receptors or inhibition of histone deacetylases. Tryptophan metabolites may suppress inflammatory responses by acting on the aryl hydrocarbon receptor in T‐cells or astrocytes. Interestingly, secretion of these metabolites can be impaired by excess consumption of dietary components, such as long‐chain fatty acids or salt, indicating that the diet represents an environmental factor affecting the complex crosstalk between the gut microbiota and the immune system. This review discusses new aspects of host–microbiota interaction and the immune system with a special focus on MS as a prototype T‐cell‐mediated autoimmune disease of the CNS.

Abbreviations

AHR

aryl hydrocarbon receptor

CNS

central nervous system

EAE

experimental autoimmune encephalomyelitis

FFAR

free fatty acid receptor

GPR

G‐protein‐coupled receptors

HDACs

histone deacetylases

IFN

interferon

IL

interleukin

MS

multiple sclerosis

SCFAs

short‐chain fatty acids

Th1/17

T helper 1/17 cells

Treg

regulatory T‐cells

Introduction

The human diet is composed of a variety of nutrients, including a variety of different carbohydrates and fats. Changes in the proportion of those dietary components, such as today's widespread intake of excess fat, excess salt and reduced fibre content, have been implicated in the priming of autoimmune diseases, including multiple sclerosis (MS). In MS, several clinical trials are currently ongoing to test the effects of dietary interventions on the disease. So far, protective effects have been proposed for a Mediterranean diet,1 polyunsaturated fatty acids2 or vitamin D.3 In turn, high salt intake has been demonstrated to exacerbate disease severity in several animal models by expanding pro‐inflammatory T helper cell subsets.4, 5, 6 An observational human study showed similar trends in MS, suggesting that high salt intake worsens the disease.7 Moreover, the intake of high amounts of milk, meat or animal fat correlated with an increasing MS prevalence.8, 9 Coinciding with this, obesity as a result of high fat intake was shown to increase the risk for MS,10 especially during adolescence. More recent studies showed that the type of fatty acid rather than excess fat intake alone may affect the disease. Investigations on saturated fats in the animal model of MS revealed harmful as well as beneficial effects depending on the aliphatic chain length.11 Whereas medium‐ and long‐chain fatty acids with more than 12 carbon atoms induce pro‐inflammatory T‐cells and aggravate central nervous system (CNS) autoimmunity, short‐chain fatty acids (SCFAs) with three–five carbon atoms induce regulatory T‐cells (Treg) that ameliorate the disease.11 Interestingly, SCFAs are bacterial metabolites produced by commensal gut microbiota from otherwise indigestible fibre.12 Indeed, several studies investigating environmental factors in MS pathology demonstrate a role for the gut microbiome, one component of a complex communication system that exists between the gut and the CNS.13 Most of the studies demonstrate alterations in the interaction of the gut microbiome, the diet and CNS immunopathology, mostly by influencing immune cell priming or by attenuating pro‐inflammatory mediator production in the periphery. Moreover, there are direct interactions between metabolites generated from the gut microbiota and the CNS. Because the exact mechanisms of many CNS diseases are still unclear, the discovery of this complex relationship, the microbiota–gut–brain axis, has opened an unexpected avenue for the study of CNS diseases and represents a potential therapeutic target in diseases such as MS.

CNS autoimmunity and neuroinflammation

It was long believed that the CNS is an immune privileged system that is protected by the blood–brain barrier, a physical barrier that prevents the invasion of circulating immune cells and pathogens. However, it has been shown that immune cells are able to cross this barrier14 and, upon injury, immune cells are actively recruited to the CNS.15 Moreover, immune cells can attack the CNS resulting in neuroinflammation, a process observed in autoimmune diseases such as MS.

MS is the most prevalent non‐traumatic neurological disease affecting young people in Western countries. The disease affects approximately 2·5 million patients worldwide, and women twice as frequently as men.16, 17 Although the aetiology of the disease is not entirely clear, a growing body of evidence suggests a combination of genetic and environmental risk factors.17 Essentially, disruption of the blood–brain barrier leads to a massive CNS massive infiltration of peripheral immune cells, including T lymphocytes and myeloid cells.18 MS pathology involves demyelination, axonal loss, reduced oligodendrocyte numbers, gliosis and astrocyte activation. Until now, the exact causal relationship between these disease components remains unknown. Current studies, demonstrating the presence of inflammatory cells and their products in CNS lesions, led to the general accepted hypothesis that MS is triggered by pathogenic T‐cells reactive against CNS constituents.19 Within the CNS, T‐cells get re‐activated by local antigen‐presenting cells (APCs), such as macrophages or microglia, that present the appropriate antigen by MHC class II molecules and thereby recruit other inflammatory cells, leading to the destruction of myelin sheaths and consequently axon damage.20, 21 The mechanisms that underlie the peripheral activation of myelin‐reactive T‐cells are less well understood. It is suggested that the initial activation of T‐cells may occur through T‐cell receptor cross‐reactivity with viral or bacterial proteins having a structural similarity to myelin antigens (molecular mimicry).22 However, CD4⁺ T‐cells reactive to myelin antigens were found in similar proportion in the blood of patients with MS as compared with healthy controls. Nonetheless, a notable number of publications demonstrated that myelin‐specific T‐cells obtained from patients with MS display an activated phenotype23, 24 and an increased production of pro‐inflammatory cytokines, such as interferon gamma (IFN‐γ) and tumour necrosis factor (TNF).25 Factors that cause this activation of CNS reactive T‐cells in patients with MS are still unknown. However, recent studies revealed that genetic as well as environmental factors play a role. Besides stress, smoking and viral infections, mainly dietary habits and the gut microbiota are implicated in immune cell activation during MS, thus representing potential risk factors affecting the disease.

Much of our understanding of the immunopathology of MS is derived from animal models mimicking several aspects of the disease. The experimental autoimmune encephalomyelitis (EAE) is the most commonly used animal model of MS. It is mainly induced in rodents by active immunization with myelin or myelin peptides (e.g. MOG, myelin oligodendrocyte glycoprotein), or by adoptive transfer of activated myelin‐specific CD4⁺ T lymphocytes in naïve recipients. The resulting T‐cell‐mediated autoimmune reaction against myelin in the CNS induces some similar symptoms to those seen in MS and causes some pathological hallmarks of MS, including neuroinflammation, active demyelination, and oligodendrocyte and axonal loss (reviewed in Ref. 26) Based on EAE models, it was assumed that CD4⁺ T helper (Th) cells play a key role in MS. Th1 cells producing IFN‐γ were assumed to have a pathogenic role, while Th2 cells primarily producing interleukin‐4 (IL‐4) or IL‐10 exert modulatory functions and a protective role.27 Subsequent work showed that also Th17 cells, secreting IL‐17 amongst others, are involved in the pathogenesis of the disease.28, 29 Mice lacking IL‐23 production, a cytokine that induces Th17 cell differentiation, were shown to be protected from EAE.28 In accordance, the transfer of myelin‐specific Th17 cells was very potent in inducing EAE in naive recipient mice.30, 31 Moreover, IL‐17‐deficient mice showed delayed and reduced symptoms, but no complete protection.32, 33 It is thus well accepted that Th17 cells together with Th1 cells are the main constituents of CD4⁺ effector T‐cells that drive disease pathology.34 Th17 cells were already implicated in the development of inflammatory bowel disease,35 rheumatoid arthritis36, 37 and psoriasis.37 The importance of Th17 cells was linked to MS, as lesions of patients with MS contained an increased number of IL‐17‐producing CD4⁺ T‐cells.38 In contrast to Th cells, Treg play a central role for immune‐regulation and suppression of autoreactive immune cells. Once activated, Treg‐expressing forkhead family transcription factor Foxp3 exert their suppressive functions via the release of anti‐inflammatory cytokines like IL‐10 and transforming growth factor (TGF)‐β in addition to cell–cell contact‐dependent mechanisms.39 In EAE, adoptive transfer of Treg cells improved disease symptoms, while ablation led to worsening of the disease.39 Importantly, an impairment of Treg cell function is believed to be a major cause for the disruption of immune homeostasis, further contributing to autoimmune reactivity.40 The loss of Treg suppressive capacity might be related to the potential of Treg cells to convert into Th1‐like Treg cells, secreting IFN‐γ,41, 42, 43 as well as Th17‐like Treg cells, secreting IL‐17 with pro‐inflammatory potential.44, 45, 46, 47 In patients with MS, IFN‐γ‐secreting Treg cells were found to be increased, and Treg cells displayed lower expression levels of FoxP3 and an impaired suppressive capacity.48, 49, 50 It is thus believed that the balance between pro‐inflammatory Th cells and anti‐inflammatory Treg cells plays a major role in MS pathology.

Both Th17 and Treg cells were shown to frequently occur in the intestine. Commensal microbiota can induce the differentiation of Th17 cells, and the majority of Th17 cells primarily accumulate in the gut‐associated lymphatic tissue under steady‐state conditions.51, 52 Moreover, Treg cells are found twice as frequently in the lamina propria of the small intestine and colon as compared with most other organs, constituting approximately 20%–30% of the CD4⁺ T‐cells.53 The gut thus represents an anatomic site possibly linking immune regulation, microbiota and autoimmunity.

The microbiome–gut–brain axis in autoimmune neuroinflammation

Within the past 10 years, it has become increasingly clear that the gut and the residing microbiota play a key role in various diseases, including MS. The gut harbours trillions of different microbes, mainly bacteria, that approximately equal the number of human cells.54 In humans and other mammals, colonization of the gut is thought to already begin at birth.55 The adult microbiome is largely defined by two dominant phylotypes, namely Bacteroidetes and Firmicutes.56 Numerous studies identified fundamental roles for the gut microbiota, for example the protection from invasive pathogens, the processing of nutrients to bioactive molecules, such as neurotransmitters, vitamins and fatty acids.57 Moreover, the microbiota is essential in priming the host's immune system, one part of the complex communication network that exists between the gut and the CNS, the so‐called gut–brain axis or microbiome–gut–brain axis. This bi‐directional communication system further comprises the autonomic nervous system, the enteric nervous system, the vagus nerve and the hypothalamic pituitary adrenal axis (reviewed in Ref. 58). There are numerous mechanisms through which the gut microbiota interacts with the CNS. These include direct activation of the vagus nerve, leading to the secretion of acetylcholine or catecholamines,59 or the interaction with enteroendocrine cells, resulting in the production of various neuropeptides and the production of gut hormones, neurotransmitters or microbial‐associated molecular patterns (MAMPs). Moreover, the intestinal microbiome provides immune‐stimulatory signals, which may activate innate and adaptive immune responses. Vice versa, the adaptive immune system was shown to control the intestinal microbiome composition and diversity.60 Hence, the immune system plays an important role in the dynamic equilibrium that exists between the gut and all major organs including the brain.61 It is thus not surprising that alterations in the commensal gut microbiota composition have been related to numerous autoimmune diseases, including MS.62, 63 Recently published studies reported the presence of a dysbiotic gut microbiota in patients with MS, characterized by a reduction of species belonging to Clostridia XIVa and IV clusters.63 Experimental studies investigating the influence of the gut microbiome in CNS autoimmunity have involved on the one hand germ‐free animals, the use of microbiota eradication using antibiotics, and the use of probiotics and fecal transplants on the other. It was shown in mice that the gut microbiota is essential in driving inflammatory processes in the CNS. Germ‐free mice, which are bred and raised in a sterile environment, display significantly attenuated EAE severity after active immunization with myelin peptide.64 Furthermore, spontaneous EAE incidence was severely reduced in T‐cell receptor transgenic mice that were kept under germ‐free conditions.64, 65 Consistently, oral antibiotic treatment significantly reduced bacterial populations in wild‐type mice and thereby impaired the development of EAE.66 This protection was associated with a reduction of pro‐inflammatory cytokines, and increases in IL‐10 and IL‐13 production.66 However, several studies also delineated beneficial effects of the gut microbiome and revealed its necessity for proper immunity. While the microbiota does not appear to affect the numbers of most innate immune cells, it does affect their function. For instance, neutrophils from germ‐free mice displayed reduced killing capacity of pathogens.67 Moreover, accumulating evidence indicates that the colonization with Clostridia species induces the differentiation of peripheral Treg cells that have a critical role in the suppression of inflammatory responses.68 This process involves SCFAs, metabolites produced by bacterial fermentation of dietary fibre. Importantly, the composition of the intestinal microbiota is highly individual and influenced by numerous exogenous factors throughout our lifetime,69 including antibiotic treatment, various pathogens but also dietary habits that may cause a state of dysbiosis. Gut bacteria respond to dietary components on the compositional level, but may also alter the production of metabolites directed towards immune cells, including SCFAs, vitamins and aryl hydrocarbon receptor (AHR) ligands (Fig. 1).

Figure 1.

Communication pathways of the microbiota–gut–brain axis. There are numerous mechanisms through which the gut can signal to the brain. These include interaction with enteroendocrine cells, resulting in the production of neurotransmitters and hormones, or activation of the vagus nerve. Moreover, the gut microbiota produces microbial metabolites that can activate immune cells to produce cytokines that induce differentiation of immune cells, also in the periphery. Short‐chain fatty acids (SCFAs) may induce regulatory T‐cells (Treg) either by G‐protein‐coupled receptors (GPRs), inhibition of histone deacetylases (HDACs) or stimulation of dendritic cells (DCs) to produce retinoic acid (RA), whereas tryptophan metabolites may suppress pro‐inflammatory activities by acting on aryl hydrocarbon receptors (AHR) and T helper (Th) 17 cells, or astrocytes. Secretion of these metabolites can be impaired by excess consumption of dietary components, such as long‐chain fatty acids or salt, indicating that the diet represents an environmental factor affecting the complex crosstalk between the gut microbiota and the immune system, potentially contributing to neuroinflammatory diseases such as multiple sclerosis (MS). IL‐10, interleukin‐10; MAMP, microbial‐associated molecular pattern; NFκB, nuclear factor‐κB; TGF, transforming growth factor; TNF, tumour necrosis factor.

Gut microbiota‐derived SCFAs in autoimmune neuroinflammation

Multiple bacterial species present in our large and small intestine produce SCFAs, mainly by fermentation of dietary fibre and otherwise indigestible starches. In the human gut, bacteria of the Bacteroidetes phylum secrete high levels of acetate and propionate, whereas those of the Firmicutes phylum generate large amounts of butyrate.70 SCFA concentrations in the gut can range from 20 to 140 mm,71 depending on the microbiota composition, the intestinal transit time, and the fibre content of the host diet. These microbiota‐derived metabolites are important energy sources not only for the gut microbiota itself but also for intestinal epithelial cells. In addition to their role as local energy carriers for the microbiota and the gut epithelial cells, SCFAs have diverse regulatory functions on host physiology and immunity. It was shown that SCFAs act as inhibitors of histone deacetylases (HDACs) and ligands for G‐protein‐coupled receptors (GPRs). The inhibition of HDACs was shown to induce an anti‐inflammatory phenotype that is crucial for maintaining immune homeostasis. Similar to the exposure to global HDAC inhibitors, SCFAs can inhibit nuclear factor‐κB (NF‐κB) signaling and downregulate the production of TNF in neutrophils and peripheral blood mononuclear cells.72, 73 Additionally, SCFAs induce anti‐inflammatory effects by HDAC inhibition in macrophages,74 dendritic cells75, 76 and T‐cells,77 mainly influencing the induction of Treg cells. We have recently demonstrated that administration of the SCFA propionate during EAE ameliorates disease symptoms by increasing Treg cell frequencies in the small intestine, thus leading to a more anti‐inflammatory environment in the gut.11 Indeed, most data discussing the mechanism for the effect of SCFAs demonstrate a pivotal involvement of Treg cells. In a murine model of inflammatory bowel disease, the administration of SCFAs increased the level of Treg cells in the gut.12 Additionally, the administration of butyrate in germ‐free mice mimicked the effect of Clostridium colonization and increased Treg levels in colon lamina propria.78 Moreover, butyrate was shown to induce IL‐10 secretion in dendritic cells and macrophages, a cytokine that is implicated in regulatory immune functions. The butyrate‐induced Treg differentiation was driven by augmented histone H3 acetylation at the FoxP3 promoter,12, 78 a transcription factor that induces the differentiation of naïve CD4⁺ T‐cells towards Treg cells. Moreover, HDAC inhibition in Treg cells enhanced their suppressive capacity, thus attenuating colitis.79 Because the suppressive capacity of Treg cells is disturbed in patients with MS,48, 49, 50 decreased intestinal colonization with SCFA‐producing bacteria and the resulting augmented HDAC inhibition may represent one possible mechanism driving inflammatory immune processes in MS.

Short‐chain fatty acids are rapidly metabolized, and their inhibition of HDACs was shown to be concentration dependent. Only high concentrations of SCFAs are sufficient to affect HDAC function,80 and their effects may require additional transporters.76 SCFAs can also indirectly suppress HDACs through GPR‐dependent mechanisms,81 with SCFAs differing in their affinity to known GPRs. Various studies identified that SCFAs can mediate their effects on host immunity by acting on the free fatty acid receptors (FFAR) GPR43 (also known as FFAR2), GPR41 (also known as FFAR3) and GPR109A (also known as HCAR2). It was shown that GPR43 expression is necessary for the SCFA‐induced expansion and suppressive function of Treg cells in colitis.82 Moreover, SCFA‐mediated activation of GPR109A, a receptor that responds to both niacin and butyrate, increased the expression of anti‐inflammatory effector molecules by monocytes and induced the differentiation of Treg cells and IL‐10 production.83 Under normal physiological conditions, niacin levels are not high enough to activate the receptor. Yet, butyrate levels, as present in the gut environment, are sufficient to stimulate a GPR‐mediated response.84 GPR‐dependent effects of SCFAs also extend to the CNS. Recent studies have demonstrated that the host microbiota controls maturation and function of microglia in the CNS.85 Microglia, the resident macrophages of the brain, are crucial for maintaining tissue homeostasis, the clearance of debris, dying cells and pathogens,86 and play a central role in brain development.87 Germ‐free mice display global defects in microglia, with altered cell percentages and impaired innate immune responses. Recolonization with complex microbiota partially restored microglia features, and it was shown that SCFAs regulated microglia homeostasis. Interestingly, mice lacking GPR43 mirrored the microglia defect of germ‐free mice, thus indicating that the maintenance of microglia homeostasis requires SCFAs and GPR43.85

Short‐chain fatty acids may also stimulate the production of retinoic acid (RA) by epithelial cells,88 a vitamin A‐derived metabolite that cooperates with TGF‐β to enhance Treg differentiation.89 Alongside its ability to enhance Treg cell differentiation, RA was observed to prevent Th17 cell differentiation,90 thus potentially promoting beneficial effects in neuroinflammation. Indeed, experimental studies investigating the effect of RA in neuroinflammation demonstrated a beneficial effect on the clinical course of EAE diseased mice.91, 92 Importantly, the actions of SCFAs are not limited to the intestine. SCFAs were shown to enter the systemic circulation by passing across the mucosa into the lamina propria.93 While butyrate enacts strong effects in the gut, its levels in the circulation are often undetectable throughout the body. However, acetate and propionate are readily detectable in the peripheral circulation, indicating that the microbiota may regulate the immune system beyond the gut and that distinct SCFAs exert different functions in the immune system.

In summary, SCFAs are naturally occurring fermentation products that may represent an attractive therapeutic option for autoimmune diseases like MS. It seems possible that the gut microbiota and its metabolism constitute interesting targets for the treatment of MS and potentially other autoimmune diseases. Such a goal may be achieved by direct intake of single SCFA or via the appropriate types of dietary fibre enabling the growth and proliferation of SCFA‐producing bacteria, thus allowing for the differentiation of naive CD4⁺ T‐cells into Treg cells.

Probiotics and gut microbiota‐derived tryptophan metabolites in autoimmune neuroinflammation

The finding that gut microbiota are able to elicit pro‐ and anti‐inflammatory reactions has raised considerable interest in the administration of various bacteria for the treatment of CNS inflammation. Moreover, it was recently shown that increased Th17 cell frequencies in the gut correlate with alterations of the gut microbiota composition and high disease activity in patients with MS.94 These findings suggest that manipulation of the gut microbiome by the use of probiotics could potentially be beneficial to patients with MS. First studies in mice revealed that prophylactic treatment with Bifidobacterium animalis decreases the duration of EAE symptoms.95 Additionally, administration of the human gut‐derived commensal Prevotella suppressed EAE in a human leucocyte antigen (HLA) class II transgenic mouse model by decreasing pro‐inflammatory Th1 and Th17 cells, and increasing the frequencies of Treg cells, tolerogenic dendritic cells and suppressive macrophages.96 Moreover, three commercially available probiotic drinks containing strains of Lactobacillus casei also ameliorated EAE.97 Prophylactic use of Lactobacilli monostrains reduced autoreactive T‐cells and prevented EAE, whereas the administration of a combined mixture of three Lactobacilli strains was able to suppress disease progression therapeutically.97 This beneficial effect was IL‐10 dependent, and treatment with Lactobacilli induced Treg cells in the mesenteric lymph nodes, but also in the periphery. A pilot study in humans demonstrated the relevance of probiotics as immunomodulatory agents in patients with MS and healthy controls.98 Administration of VSL3, a probiotic cocktail of eight bacteria, induced changes in the gut microbiota composition that were associated with anti‐inflammatory immune responses in the periphery. VSL3 enriched Lactobacillus, Streptococcus and Bifidobacterium species, which was paralleled by decreased frequencies of intermediate monocytes.

Recent work linked the disease‐modifying effect of Lactobacillus to pro‐inflammatory Th17 cells.99 We have recently demonstrated that high‐salt concentrations, typically found in our ‘Western diets’, aggravate EAE by the induction of pathogenic Th17 cells.4 Moreover, high‐salt concentrations diminish the suppressive capacity of Treg cells, further favouring a pro‐inflammatory environment.100 Macrophages respond to increased salt concentrations by the induction of a pro‐inflammatory M1 phenotype and a decreased ability of regulatory M2 macrophages to suppress effector T‐cell proliferation.101 These effects were recently linked to the gut. High‐salt consumption significantly increased fecal sodium concentrations, which was paralleled by alterations in the fecal metabolome and gut microbiome composition.99 Above all, a Lactobacillus strain was suppressed, coinciding with increased numbers of intestinal and systemic Th17 cells. Administration of this Lactobacillus strain during EAE prevented the salt‐induced aggravation of EAE and decreased Th17 cell differentiation in the gut. Indicating a potential relevance in humans, we showed that a salt challenge in healthy humans also affects the abundance of intestinal Lactobacilli, alongside increased frequencies of pro‐inflammatory Th17 cells in the blood. Mechanistic studies in mice revealed that the fecal tryptophan metabolite indole‐3‐lactic acid (ILA) may sense the salt‐induced aggravation of experimental neuroinflammation.99 We also showed that ILA inhibited Th17 cell polarization in vitro, thus linking the high‐salt‐diet‐induced suppression of Lactobacillus to the induction of Th17 cells.

Tryptophan is an essential amino acid that is metabolized by the gut microbiota into potent immune‐modulating products that bind the AHR. The AHR is a ligand‐inducible transcription factor that is expressed by immune cells and epithelial cells.102 Depending on the acting ligand, activation of AHR has been shown to either promote Th17 or Treg cell differentiation. The tryptophan metabolite FICZ [6‐formylindolo (3‐2b) carbazole] increased IL‐17 and IL‐22 producing CD4⁺ T‐cells in MOG‐immunized mice, thereby accelerating EAE onset and CNS pathology.103 In addition, FICZ treatment was shown to induce Th1 and Th17 cells in the spleen, coinciding with severe EAE symptoms.103 In contrast, tryptophan‐derived ITE [2‐(1'H‐indole‐3′carbonyl)‐thiazole‐4‐carboxylic acid methyl ester] was shown to ameliorate EAE symptoms by the induction of Treg cells.104 Coinciding with this, exposure of mice to TCDD, an environmental toxin, leads to profound immunosuppression resulting in reduced EAE severity.105 However, all these studies injected exogenous AHR ligands rather than investigating dietary or microbiota‐derived endogenous metabolites. Tryptophanase‐positive bacteria, such as Lactobacillus reuteri, generate various indole metabolites from dietary tryptophan that can stimulate AHR activity.106 Indoles were shown to cross the blood–brain barrier and suppress pro‐inflammatory activities by activating AHR in astrocytes.107 Lack of dietary tryptophan or deficiency of AHR in astrocytes caused severe EAE symptoms. This effect was reversed by tryptophan supplementation in control mice, but not in astrocyte‐specific AHR knockout mice.107 Interestingly, patients with MS show decreased concentrations of AHR ligands in the peripheral blood,108 indicating that patients with MS seem to harbour deficits in the generation, uptake or stability of these anti‐inflammatory metabolites, resulting in a decrease in their levels in AHR‐dependent immune regulation.

Other microbiota and bacterial metabolites affecting autoimmune neuroinflammation

Gut microbes may produce other metabolites than SCFAs or tryptophan metabolites, for example Bacillus‐derived poly‐gamma‐glutamic acid (gamma‐PGA). Gamma‐PGA is able to stimulate dendritic cells to favour the polarization of naïve CD4⁺ T‐cells towards Th1 rather than Th2 cells.109 There is evidence that gamma‐PGA may also promote the differentiation of Treg cells and suppresses the differentiation of Th17 cells.110 The initiation of FoxP3 expression by gamma‐PGA was partially attributed to TGF‐β induction via a TLR‐4 pathway. Intriguingly, in vivo administration of gamma‐PGA attenuated EAE symptoms and reduced Th17 cell infiltrates in the CNS.110 Another bacterial metabolite inducing Treg cells is polysaccharide A (PSA) isolated from human commensal Bacteroides fragilis. PSA was shown to mediate the conversion of CD4⁺ T‐cells into IL‐10‐producing Foxp3⁺ Treg cells via TLR2 and suppresses Th17 responses, thereby preventing EAE.111 In addition to bacterial metabolites, bacteria also produce MAMPs, which can be recognized by the host immune system via pattern recognition receptors, such as Toll‐ or Nod‐like receptors.112 Moreover, microbiota may influence neuroinflammation through cell wall components, such as peptidoglycan (PGN). PGN can be phagocytized by APCs, modulate dendritic cells and induce Th1 cells.113 Both in EAE and MS, PGN was found in APCs located in the CNS,114, 115 which inversely correlated with myelin density.116 These data further highlight the relevance of intestinal microbiota in shaping our immune system and CNS autoimmunity.

Conclusion

The gut is the most prominent anatomic site to investigate the interaction between commensal bacteria, diet and autoimmunity. Numerous bacteria and dietary interventions may affect pathogenic processes during neuroinflammation. Consequently, several studies implied alterations in the composition of the human gut microbiome in MS pathology. So far, available data suggest that SCFA‐producing bacteria may be reduced in patients with MS, whereas potentially pro‐inflammatory Methanobrevibacter and Enterobacteriaceae may be increased. If and how single bacterial species mechanistically contribute to MS, which may then be therapeutically targeted, remains to be investigated. To date, available data suggest multiple components in this axis may be disrupted, and therapeutic approaches will have to manipulate the gut microbiota on different levels. Multiple scenarios appear feasible, but remain to be mechanistically proven. Probiotics involve the treatment with potentially beneficial live microorganisms, being absent in the diseased host, as may be the case for butyrate‐producing bacteria. Alternatively, prebiotics such as starches and dietary fibre could be used to enrich strains that are deficiently present in the host. Antibiotics might be used to remove unwanted bacterial species. Yet, this treatment also targets many strains that may elicit beneficial effects, including SCFA‐producing bacteria. The direct supplementation of metabolites of bacterial origin, such as SCFAs, vitamins or tryptophan metabolites, may circumvent the underlying dysbiosis and have beneficial effects by increasing regulatory mechanisms. Finally, fecal microbiota transplantation in patients with MS from healthy donors may be conceivable, if other approaches fail to achieve the desired benefit.

Disclosures

The authors declare no competing financial interest.