Uveitis and the Gut Microbiota

Abstract

Uveitis is a heterogeneous collection of inflammatory diseases of the intraocular uveal tissues and adjacent structures and they collectively are a significant cause of visual morbidity. In recent years, investigating the contribution of the gut microbiota to autoimmunity, including in the development of uveitis has gained interest. Decreased disease severity has been observed in both the induced experimental autoimmune model (EAU) of uveitis and the spontaneous RI61H model of uveitis in mice treated with oral broad-spectrum antibiotics and raised in germ-free conditions, implicating a role for the gut microbiota in the development of disease in these models. Also, in support of these findings are the differences in the composition of the microbiota that have been reported in uveitis patients. Proposed mechanisms accounting for the microbiota triggering uveitis include antigenic mimicry and dysbiosis leading to dysregulation of the immune system. An improved understanding of these mechanisms will facilitate potential therapeutic approaches including alteration of the microbiota with probiotic treatment and fecal microbiota transplants.

Introduction

The term “uveitis” refers to a heterogeneous collection of inflammatory diseases of the intraocular uveal tissues and adjacent structures ¹. Uveitis is not only a significant cause of visual morbidity but can lead to sight-threating inflammation and has been estimated to account for up to 15% of blindness in the United States ^2–4. An imbalance between auto-reactive pathogenic effector T-cells, including T-helper (Th1) and Th17 lymphocytes with T-regulatory cells has been implicated in the pathogenesis of non-infectious uveitis ⁵.

The adult intestines are home to a complex microbial ecosystem consisting of an estimated 100 trillion organisms (Host-Bacterial Mutualism in the Human Intestine). There are numerous reported beneficial effects of the gut commensals which include facilitating metabolism of nutrients, maintaining the intestinal barrier and preventing pathologic bacterial colonies from colonizing ^{6, 7}. In recent years, the contribution of the gut microbiota on immune function has been recognized including their important role in regulating T-regulatory cell and Th17 cell mucosal responses ⁸. For example, the commensal Bacteroides fragilis has been shown to promote inducible T-regulatory cells within the colon ⁹. The United States National Institutes of Health (NIH) Human Microbiome Project was established in 2007 to study human commensal microbiota and has reported associations between some clinical phenotypes and microbiota composition ¹⁰. Ribosomal 16S RNA gene sequencing is permitting identification of the different species that constitute the gut microbiome ¹⁰. Interest has emerged on the role of the commensal microbiota and dysbiosis on the pathogenesis of a spectrum of immune-mediated diseases including in multiple sclerosis ¹¹, type I diabetes ¹², rheumatoid arthritis ¹³ and ankylosing spondylitis amongst others ¹⁴. Mirroring the interest that has arisen in the role of commensal bacteria in the development of non-ocular immune-mediated diseases, recent studies have explored the role the gut microbiota in the development of autoimmune uveitis ^15–21. Given that uveitis is a chronic disease and the lifetime burden of the disease to patients, many of whom typically require long-term treatment with immunomodulatory therapy ²², an improved understanding of the microbiota in the pathogenesis of the disease will ultimately facilitate therapeutic approaches. In this review, we will summarize the literature to date on this field, including reports from animal models of uveitis and observational clinical studies, as well as discuss proposed mechanisms of the gut microbiota triggering uveitis and future therapeutic approaches of altering the gut microbiota to treat uveitic diseases.

Insights from mouse models of uveitis

The observations that depleting the gut microbiota attenuates disease severity in mouse models of autoimmune uveitis have implicated a role for the gut microbiota in the pathogenesis of uveitis ^{17, 23, 24}. In the experimental autoimmune (EAU) model of uveitis, uveitis is induced through immunizing mice with interphotoreceptor binding protein peptide (IRBP) and Freund’s adjuvant ²⁵. Nakamura and colleagues reported a significant decrease in the clinical severity of EAU in B10.RIII mice who received combination treatment with oral broad-spectrum antibiotics (vancomycin, metronidazole, neomycin and ampicillin) starting one-week prior to immunization with IRBP. Notably, no effect was seen in mice that received intraperitoneal antibiotics, dissuading from a generalized anti-inflammatory effect of antibiotic treatment. Further, an increase in the frequencies of T-regulatory cells was observed within the gastrointestinal lamina propria as well as the cervical and mesenteric lymph nodes of the antibiotic treated mice ¹⁷. In accord with these findings, a reduction in EAU severity was also observed in C57BL/6J mice raised in germ-free conditions ²³. Despite the value of these observations, these studies are limited by the demonstration that the immunization process itself can affect the microbiome ¹⁷. In addition to the classic EAU model, observations from studies involving the spontaneous R161H model of uveitis have also suggested a role for the microbiome in the pathogenesis of uveitis. The R161H mice have a B10.RIII background and spontaneously develop uveitis due to their expression of a transgenic T-cell receptor that is specific to IRBP. Of their CD4+ T-cell compartment, 20–30% of peripheral CD4+ cells are IRBP specific and consequently, 100% of mice develop uveitis by 2-months of age ²⁶. Both rearing R161H mice under germ-free conditions and treating mice prior to birth with a combination of broad-spectrum antibiotics resulted in significantly decreased severity of the disease and was associated with decreased frequencies of IRBP-specific T cells in the intestinal lamina propria ²⁴. The observation of decreased disease activity in both spontaneous and inducible models of uveitis (in which disease induction is dependent on an immunization trigger) is suggestive of a multifactorial role of the microbiota in the pathogenesis of this disease.

Observational studies in patients with uveitis

To date, a few observational studies have compared the microbiota of uveitis patients to normal healthy controls. Huang and co-authors studied the microbiota composition and fecal metabolites in patients with anterior uveitis compared to normal volunteers. A significant difference in the diversity of microbiota or microbial composition was not observed after correction methods were employed although their study was limited by their sample size of 38 patients. However, a significant difference between the two groups was observed when examining fecal metabolites ^27. Chakravarthy and colleagues studied the gut microbiota of a heterogenous group of 13 uveitis patients in India, and compared this to 13 healthy controls with a similar ethnic and dietary attributes. They reported decreased microbiota diversity in their uveitis cohort. Notably, they also demonstrated a reduction in the diversity of several purported anti-inflammatory microbes (including Faecalibacterium, Bacteroides) and an increase in the pro-inflammatory Prevotella in uveitis patients ¹⁶. This group also investigated the gut fungal microbiome within this cohort of uveitis patients and similarly observed a decrease in the number of fungal species and diversity in uveitis patients compared to healthy controls ¹⁵. However, whether and how much corticosteroids or immunosupressives these patients were on was not specified. The microbiota of Chinese patients with active Vogt-Koyanagi-Harada (VKH) disease was studied by Ye et al., who similarly to Huang and colleagues, observed no differences in diversity measures between VKH patients and healthy controls. They did however report relative enrichment of Paraprevotella spp. and depletion of Clostridium spp., Bifidobacterium spp., Candidatus Methanomethylophilus alvus and Methanoculleus sp. CAG:1088 in the active VKH patients compared to the control group while the enterotypes showed no significant difference between patients and controls. They also showed that these differences were reduced following treatment with immunomodulatory therapy and that Alistipes and Bacteroides spp. were associated with good response to imunosupressive therapy. Using 37 species, they developed a classifier that distinguished patients with good prognosis from those that relapsed with high sensitivity and specificity. They also performed fecal microbiota transplants (FMT) pooled from active, treatment-naive VKH patients into antibiotic treated B10RIII mice, prior to EAU induction. These mice exhibited a significant increase in both clinical and retinal histological scores compared to mice that received FMTs from normal healthy volunteers or PBS treated mice ²¹. Interestingly, this group also reported a similar finding of increased disease severity in EAU mice following FMTs from patients with Bechet’s disease compared to control mice ²⁰.

Although these studies undoubtedly offer insights into the role of the microbiota in uveitis, they have their inherent limitations. Faecal samples from patients may not be fully representative of their gut microbiota, although clearly the non-invasive nature of this sampling technique is beneficial. Secondly, although these observational studies demonstrate some associations between the microbiota and uveitis, they do not demonstrate causality. Similar to other autoimmune diseases, it remains unclear whether the observed differences are etiological factors in the development of the disease or rather the result of disease itself ²⁸. Lastly, more studies are required to better characterize the patterns of dysbiosis in uveitis patients.

Proposed Mechanisms of the Microbiota triggering uveitis

Several non-mutually exclusive mechanisms have been proposed for the involvement of the microbiota in the pathogenesis of uveitis as well as other autoimmune diseases, including: 1) antigenic mimicry; 2) dysbiosis (including HLA-associated dysbiosis) leading to impaired microbiome-dependent immune homeostasis; 3) dysbiosis resulting in disruption of intestinal barrier function and 4) migration of gut mucosal-associated lymphocytes to peripheral sites.

Antigenic mimicry is the process whereby autoreactive T-cells are generated through cross-reactivity of gut microbial peptides with self-antigens ²⁹. The observation that the adoptive transfer of R161H derived T-cells activated in vitro by intestinal contents can induce disease in 40–86% of naïve wild-type mice within 6–10 days (compared to no disease induction within this time frame following the adoptive transfer of T-cells cultured in the absence of intestinal contents) is supportive of a candidate microbial mimic ²⁴. Although specific microbial antigens triggering uveitis through mimicry have yet to be identified, a candidate antigen from Fusobacteria mimicking islet-specific glucose-6-phosphate catalytic subunit-related protein (IGRP) has been identified in non-obese diabetic mice ³⁰. Microbial cross-reactivity has also been demonstrated in patients with systemic lupus erythematosus. Commensals have been identified in the skin, oral mucosa and gut of lupus patients and healthy controls that are orthologs of the RNA binding autoantigen Ro60. One of these identified commensals, Propionibacterium propionicum, activated Ro60-specific CD4 memory T-cells from lupus patients ³¹. Similarly, three species of gut commensal bacteria (Prevotella, Parabactereroides and Butyrucumonas) were identified that share homology with the autoantigens N-acetylglucosamine-6-sulfatase and filament A in rheumatoid arthritis patients ³².

Given the importance of the commensal microbiota in immune homeostasis ⁸, it has been proposed that a relative dysbiosis (which refers to imbalances in the microbiome ecosystem) may result in maladaptive immune regulation including an imbalance in mucosal Th17/T-regulatory cells. Decreased levels of “anti-inflammatory” microbial metabolites have also been implicated in dysbiosis ³³. Thus, dysbiosis could potentially lower the threshold for activating pathogenic autoimmune Th17 effector cells through, for example, an innocuous antigen triggering a response through microbial-associated molecular patterns (MAMPs) or danger-associated molecular patterns (DAMPs) receptors ¹⁹.

The importance of microbial-derived metabolites in immune homeostasis is emerging. One such example are short chain fatty acids (SCFAs), which are derived from fermentation of intestinal microbial metabolites and have been shown to increase colonic T-regulatory cell frequencies in the gut in mice following administration of this metabolite ³⁴. A possible role of SCFAs in the development of uveitis is supported by the observation of a beneficial effect on clinical EAU scores following the administration of the SCFA propionate in drinking water prior to immunization of C57BL/6 EAU mice but not in B10.RIII mice. Increased frequencies of T-regulatory cell within the intestinal lamina propria were also observed. The SCFAs acetate and butyrate also resulted in T-regulatory cell induction but did not alter the clinical course of EAU ¹⁸.

Genetic factors such as HLA haplotypes can influence the gut microbiota, which is significant given the association of certain uveitic diseases with HLA haplotypes. Experiments in transgenic rats expressing human HLA-B27 revealed alteration of the gut microbiome compared to littermate controls ^{35, 36}. Notably, HLA-B27 is highly associated with spondyloarthropathies and anterior uveitis ³⁷. The demonstration of increased expression of bactericidal antimicrobial peptides (AMPs) and increased numbers of secretory IgA-coated bacteria in the faeces of these HLA-B27 transgenic mice have been implicated as a possible causative mechanism of dysbiosis ³⁵.

Disruption of the intestinal barrier through dysbiosis has also been proposed as a causative mechanism and may result in the translocation of either bacteria or microbial antigens into the blood or lymphatic system. This could lead to induction of an immune response either through an adjuvant mechanism or via the initiation of a local response to these microbes. This hypothesis has been supported by the demonstration of increased intestinal permeability ³⁸ and the detection of bacteria within the synovial fluid ³⁹ of ankylosing spondylitis patients. Despite this data from rheumatological diseases ⁴⁰ the altered integrity of the gut barrier in uveitis not reported to date.

Lastly, the migration of intestinal cells to peripheral sites has been supported by the report demonstrating the trafficking of leukocytes from the gut to the eye in an EAU mouse model using transgenic mice which express the photoconverting fluorescent reporter Kaede ¹⁸.

Despite this data, the relative contribution of the different proposed mechanisms to the development of uveitis remains uncertain.

Therapeutic approaches

A number of therapeutic approaches to the treatment of uveitis through alteration of the microbiota and addressing dysbiosis have been proposed.

Probiotics are thought to alter the immune responses in gut mucosal tissues through interactions with mucosal dendritic cells and stimulation of anti-inflammatory cytokines secretion such as IL-10 and TGF-B ⁴¹. Using an EAU model, Kim et al. observed decreased retinal histological scores following three weeks of treatment with probiotics through oral gavage (L. casei, L. acidophilus, L. reuteri, B. bifidum, and S. thermophilus) following immunization with IRBP in C57BL/6 mice ⁴².

Since diet is known to influence the microbiota ⁴³, specific dietary modifications may theoretically, minimize dysbiosis, induce beneficial bacteria, and lead to the production of metabolites that promote immune-regulation. In addition to modification of diet, the direct ingestion of SCFAs or other beneficial bacterial metabolites may have a direct effect on modulating the gut mucosal immunity. Indeed, ingestion of the SCFA propionate has been reported to decreased the severity of EAU ¹⁸.

Fecal microbiota transplants (FMT) involve placing stool from a normal donor into the gastrointestinal tract of a patient and is used in the treatment of recurrent Clostridium difficile colitis ⁴⁴ successfully and its use has also been studied in inflammatory bowel disease ⁷ where its effects have been more modest. To date, there have been no trials investigating the effect of FMTs in uveitis patients ⁴⁵. Despite the observation of worsening severity of EAU following FMTs from patients with active VKH and Behcet’s disease into EAU mice ^{20, 21}, it cannot be assumed that the reverse would necessarily occur and that alteration in clinical disease would be seen following FMTs from normal healthy volunteers to patients with active uveitis.

Other possible approaches include targeting specific pathogenic causative bacteria (assuming they can be identified) with antibiotics or treating with medications to reduce intestinal permeability, which may be effective in preventing the dissemination of bacterial products.

Future Directions

Although the microbiome is emerging as an importance influence of autoimmunity, the relative importance of the microbiota to the development uveitis compared to other established genetic and environmental factors is unclear. Further, the majority of data to date on the involvement of gut microbiota in uveitis has arisen from mouse models of uveitis, which despite the value of these observations, are limited in the applicability due to the heterogenous spectrum of human uveitic diseases. The identification of microbial mimics, the genetic and environmental factors that promote dysbiosis, the relative contribution of dysbiosis to the development of clinical uveitic diseases and the extent to which alteration of commensal microbiota can lead to modification of disease activity in patients with established disease remains unknown are some of the critical questions that need to be addressed. Further studies are required to more completely characterize the microbiota in uveitis patients, including classifying constituent microbes into beneficial or maladaptive in order to achieve an improved understanding of these mechanisms, which will ultimately facilitate future therapeutic approaches.

Summary

The importance of the gut microbiota in the pathogenesis of uveitis is emerging. Results from mouse models of uveitis have reported decreased disease severity in mice raised in germ-free conditions or following treatment with oral broad-spectrum antibiotics, implicating a role for the gut microbiota in the development of uveitis. The role of the gut microbiota has also been implicated by the reported differences in the composition of the microbiota in uveitis patients compared to normal controls. The mechanisms involved are still to be fully elucidated but antigenic mimicry and dysbiosis leading to dysregulation of the immune system and disruption of the intestinal barrier have been proposed. Further understanding of the microbiota in the development of uveitis is necessary for the development of therapeutic strategies to alter the microbiota and address dysbiosis.

Research Agenda

• Characterization of the commensal microbiota in uveitis patients;

• Understanding the mechanisms through which dysbiosis can contribute or trigger uveitis including antigenic mimicry and dysregulation of the mucosal immune responses;

• The development of future therapeutic approaches, which include the role of probiotics and fecal microbiota transplants and whether alterations of the microbiota in patients with established uveitis can modify the course of disease activity

Practice Points

• The term uveitis refers to a heterogeneous collection of inflammatory diseases of the intraocular uveal tissues and adjacent structures;

• Uveitis is a significant cause of visual morbidity and has been estimated to account for up to 15% of blindness in the United States

• An imbalance between auto-reactive pathogenic effector T-cells, including T-helper (Th1) and Th17 lymphocytes with T-regulatory cells has been implicated in the pathogenesis of non-infectious uveitis;

• The gut microbiota has also been implicated in the pathogenesis of uveitis