Commensal microbiota as a potential trigger of autoimmune uveitis

KEYWORDS:

• Autoimmune uveitis
• autoreactive T cells
• commensal microbiota
• gut
• immune privilege
• retina

1. Introduction

Uveitis encompasses inflammatory diseases of the uveal tract and is responsible for 10–15% of severe visual handicap in the Western world [1,2]. Uveitis affects people at a young age and therefore the lifetime burden of disease is possibly greater than in diseases of aging such as age-related macular degeneration (AMD) or glaucoma. Uveitis can be associated with a systemic autoimmune syndrome, but in about 50% of the cases the eye is the only clinically affected organ. Our understanding of the etiology of disease, its driving mechanisms and treatment options remain limited, although anecdotal evidence has linked flares of autoimmune uveitis to microbial infections in some cases [3]. In recent years, it has been recognized that the commensal microbiome has a huge impact on all aspects of human biology, including immune function and disease [4]. This is easy to conceptualize for intestinal diseases, but less so for diseases at sites distant from the gut. Using a spontaneous autoimmune uveitis model developed in our laboratory, we demonstrated that retina-specific T cells receive an activation signal in the gut from commensal microbiota-derived antigen(s) and trigger autoimmune uveitis [5]. We propose that gut commensal microbiota play a role as a trigger of autoimmune uveitis and discuss potential therapeutic approaches for uveitis with our current knowledge.

2. Human microbiome and autoimmune diseases

Human body is inhabited by a diverse microbial community, and the dynamic crosstalk between the host and the microbiota is important in maintaining homeostasis in health and disease [6]. Although microbial exposures have long been associated anecdotally with various human disorders, the commensal microbiome has begun to be examined in this regard only recently. In 2007, the National Institutes of Health initiated the Human Microbiome Project (HMP) to identify and characterize the microorganisms in healthy adults and in patients with selected diseases [7]. The human gut microbiota is dominated by Firmicutes and Bacteroidetes with variations in relative proportions between individuals. Clinical studies increasingly support a contributory role of changes in commensal gut microbiota (dysbiosis) to human autoimmune diseases including rheumatoid arthritis (RA), spondyloarthritides, lupus, diabetes, inflammatory bowel disease (IBD), and multiple sclerosis (MS) and have started unraveling associations between the relative abundance of several bacterial taxa and autoimmune diseases. These studies have also indicated varying enterotypes based on disease stage, chronicity, or treatment. Of note, in some cases these were similar in patients and the corresponding animal model of the disease: e.g. Prevotella copri was more abundant among (treatment-naïve) RA patients than in healthy controls, and animals colonized with P. copri became more susceptible to arthritis, supporting a pro-inflammatory role for this organism. Both new onset RA patients and psoriatic arthritis patients showed reduced microbiome diversity [8]. More recently, there has been a surge in metagenomic and metabolomic studies. For example, a metagenomic sequencing study successfully differentiated IBD patients from healthy individuals based on bacterial species abundance. Importantly, new unpublished data (presented at conferences) are beginning to support the notion that human microbiota can promote development of pathology in animal models of the corresponding disease, validating the critical role of these models to study effects of the human microbiome.

As the first gut microbiome study in human uveitis, our preliminary data from a small cohort of chronic autoimmune posterior segment uveitis patients, whose disease was controlled by treatment, indicated a changed microbial composition between uveitis and healthy controls. There were significant differences in genus-level abundance of several bacteria (Sen et al., unpublished data). Although further validation is required, including potential effects of the treatment that these patients were receiving, this supports our hypothesis that uveitis patients have an altered gut microbial composition. A larger study to characterize the gut microbiome in different types of uveitis is currently underway.

3. Animal studies on microbiota and uveitis

A causal role for the microbiome in autoimmune disease in animal models is now well established. Examples include experimental autoimmune encephalomyelitis (model for MS) [9], arthritis, colitis, diabetes [10,11], and autoimmune uveitis, the latter in two different models: (i) spontaneous uveitis in R161H mice that express a transgenic T cell receptor (TCR) for the interphotoreceptor retinoid binding protein (IRBP, a target autoantigen in autoimmune uveitis) [12] and (ii) the ‘classic’ uveitis model involving active immunization with IRBP in complete Freund’s adjuvant [13]. R161H mice spontaneously develop uveitis with 100% incidence by 2 months of age, making them a robust and reproducible model of the disease [12]. Long-term antibiotic treatment (starting before birth), or rearing under germ-free conditions, resulted in protection. Further studies indicated that retina-specific T cells receive a signal through their clonotypic TCR in the gut lamina propria and convert to Th17 and Th1 cells, which are considered pathogenic effectors in uveitis. This occurred even in the absence of the endogenous antigen IRBP, suggesting that these cells were triggered by a surrogate antigen present in the gut environment. Additional support for a microbiota-derived antigenic signal derives from the finding that protein extracts from microbiota-rich intestinal contents activated retina-specific R161H T cells, making them pathogenic enough to transfer disease in naïve wild-type (WT) recipients [5]. These findings strongly support a need for a TCR-driven (antigenic) signal, but they do not negate a requirement for innate adjuvant effects, which are ‘built into’ all microorganisms, including gut commensals. Although we were unable to separate the microbial mimic from putative microbial adjuvant components in the intestinal content extracts, a microbial mimic of a type 1 diabetes antigen was recently identified [14]. Thus, antigenic mimicry by commensals may be a more frequent trigger of autoimmune disease than was hitherto appreciated.

Importantly, commensal microbes may also affect progression of uveitis, once induced. This was demonstrated in IRBP-induced experimental autoimmune uveitis (EAU) model. Although in our hands antibiotic-treated WT mice (littermates of antibiotic-treated R161H mice) that were actively immunized for EAU developed full-blown disease [5], Nakamura et al. reported a different outcome [13]. Using a short-term course of the same antibiotic mix given orally, disease in the immunization-induced model was temporarily ameliorated, which the authors felt could be attributed at least in part by emergence of T regulatory cells in the intestine of antibiotic-treated mice, possibly as a result of the altered microflora. In contrast, broad-spectrum antibiotics did not affect progression of spontaneous uveitis in a recent report using the model of combined mutations in hypomorphic AIRE function and LYN deficiency [15]. The length of antibiotic treatment that was used in different studies (weeks vs. months) and/or the specific microbial environments in various facilities may underlie these differences. These results also highlight the notion that uveitis is a heterogeneous disease with potentially different environmental, immunologic, and genetic influencing factors.

Thus, alterations of commensal communities may contribute to disease by a combination of adaptive and innate pathways including microbial mimics of autoantigens, innate microbial stimuli, loss of microbiota that produce ‘anti-inflammatory’ metabolites such as short chain fatty acids [16], and/or by emergence of pathogenic bacteria which may disrupt intestinal barrier and stimulate production of inflammatory mediators.

4. Future directions and potential therapeutic strategies

The HMP and European Metagenomics of the Human Intestinal Tract project characterized the composition, diversity, and functionality of the healthy gut microbiome, and clinical studies have shown associations of taxonomic abundance with some clinical phenotypes. However, comprehensive studies of the microbiome, metagenome, and metabolome in uveitis patients to understand the role of commensal microbiota in induction or propagation of disease are still lacking. Although a role for gut commensals in animal models of autoimmune uveitis has been strongly supported, it is not known to what extent dysbiosis in the gut might affect human uveitic disease.

It is important to identify the putative microbial mimic(s) in the commensal flora, not only from mice but also in human microbiota. To this end, we are using bioinformatics approaches to identify candidate antigenic mimics in the microbial protein databases, but so far this approach has not been successful. We also attempted to narrow down the bacterial species by treating R161H mice with individual antibiotics from the cocktail of four, but no single antibiotic significantly reduced disease. For the human study, we are cataloging the flora of uveitis patients compared to healthy controls to first examine the associations, and subsequently will reconstitute germ-free R161H mice with human commensals from healthy donors and patients. If they promote the development of disease, we will attempt to identify the microorganisms involved and analyze them. Identification of the bacterium (or bacteria) involved in triggering or ameliorating autoimmune uveitis could open the door to manipulate these taxa for therapeutic purposes through antibiotic, probiotic, and prebiotic approaches. Information from these studies may also be relevant to other immune-driven ocular diseases such as AMD, glaucoma, and diabetic retinopathy.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

The work of Drs. Horai, Sen and Caspi is supported by the National Eye Institute, National Institutes of Health Intramural Research Program.