Gut Inflammation Precipitating Autoimmunity
by Adam Burrack, PhD
In previous posts in our “immunology of diabetes” series I have hinted at the critical role that inflammation and T cell responses within the gut tube play in the onset of autoimmunity leading to type 1 diabetes (T1D). I have covered the consequences in terms of the CD4 T cell response to insulin, auto-antibody production by B cells against the insulin molecule, and the CD8 T cell response to IGRP. I’ve described the association between HLA molecules and the tendency to develop T1D, and the potential role of the bacteria resident within the intestine in shaping the T cell responses to the gut and to proteins derived from gut contents. I have not previously addressed the physical barriers between the gut and the immune system, and the ways in which these barriers may be weakened, or the role this weakening may play in the onset of T1D. Today we will address this physical barrier and its role in preventing unintended T cell responses to gut contents.
The small intestine is where gut contents (non-self items such as food and intestine-resident bacteria) meets self (intestinal cells, digestive hormones, mucus lining the interior of the small intestine), requiring a very fine distinction between self and non-self to preserve immune tolerance and prevent autoimmune T cell responses leading to disorders including type 1 diabetes. Tolerance, or lack of destructive T cell responses, to self-peptides can break down at any level in the gut, from the small intestine (celiac disease, an innappropriate response to wheat-derived peptides) to the colon (colitis, autoimmune T cell responses against poorly-defined components of the colon).
Interestingly, recent estimates are 5-6% of people with T1D also develop celiac disease, an autoimmune T cell response against wheat-derived peptides derived from gluten, and modified within the gut by the enzyme tissue transglutaminase. Incidence of this disease is significantly higher in those with T1D than the general population, (approximately 1 in 20 people with T1D also have celiac disease), suggesting a mechanistic and/or genetic link between these aberrant T cell responses. The symptoms of celiac disease can be subtle in some patients: there has been a recent of diagnoses in the absence of clinical symptoms, based on T cell responses against wheat-derived peptides (as well as barley and rye).
Some clinicians and researchers have speculated that these two conditions are mechanistically linked through gut inflammation and genetically-determined tendencies to develop autoimmunity mediated through HLA types that are associated with the development of T1D. If true, this would suggest that genetically-determined tendencies to develop diabetes may represent a more general trend toward increased levels of gut inflammation, that diabetes and celiac disease are both potential outputs of this genetic tendency. This suggests a model in which genetics determine the tendency to develop autoimmunity (genetically-determined HLA molecules characteristic to an individual) and environment determines both whether autoimmunity develops (through the presence of inappropriately high levels of inflammation) as well as what type of autoimmunity (by the presence of “new” peptides in places they would otherwise not be found).
With the above scenario in mind, a potential disease mechanism for the development of T1D emerges in which genetic tendencies toward aberrant T cell responses combines with HLA-determined tendencies to show peptides to T cells in ways which the T cells were not made tolerant to during development in the thymus. The third required component to promote an autoimmune response to these so-called “neo-antigens” is inflammation within the gut itself, potentially triggered by bacteria within the gut as described in our post about Mark Atkinson’s research group at the University of Florida. For T1D to occur, keeping in mind our post about the research of Diane Mathis and Christophe, the inflammation appears to have to be specific to the gut tube and the antigen-presentation event to T cells specific to the pancreatic lymph node. We established in our post about the career of George Eisenbarth that the most likely “first target” of the autoimmune response leading to T1D is the insulin molecule itself.
Key remaining questions are how is the inflammation triggered and how might gut contents escape the gut tube to for an autoimmune T cell attack to be triggered?
Dr. Alessio Fasano and his research team first characterized the protein zonulin in the year 2000 at the University of Maryland Health Science Center. Zonulin regulates the permeability of tight junctions between cells, in particular in the digestive tract. Tight junctions between epithelial cells of the small intestine are particularly relevant to type 1 diabetes onset. Maintaining the integrity of tight junctions is critical to keep gut contents in the gut tube and cells of our immune system – in particular antigen-presenting cells and T cells – outside of the gut tube. A second key component of the intestinal barrier, which is not the focus of today’s post, is the layer of mucus that coats the interior wall of the gut tube.
In a 2005 manuscript in the Proceedings of the National Academy of the USA, Dr Fasano proposes that loss of zonulin-mediated integrity in the gut epithelial barrier predisposes the BB rat model of T1D to develop anti-insulin responses leading to the development of T1D. The biobreeding (BB) rat is an alternate rodent model to the NOD mouse which develops an autoimmune condition very similar to human T1D. This model has been used nearly as long as the NOD mouse, but is not the current favored model because there are more and better analytic tools to investigate diabetes and T cell responses in NOD mice.
In a follow-up 2006 manuscript in the journal Diabetes, the same group found that zonulin up-regulation in at-risk individuals correlated with later onset of T1D. This report described the correlation between levels of the zonulin protein in the serum of people at risk of T1D with intestinal permeability and later development of T1D in 70% of subjects studied over the following 3.5 years. This report provided clinical evidence that elevated serum levels of zonulin correlated with later T1D onset about as well as autoantibody levels.
In an exciting 2009 clinical follow-up, also published in the journal Diabetes, this same group narrowed down the correlation further to a link with the T1D-associated HLA-associated DR allele and demonstrated that HLA-DR4-expressing individuals at-risk of developing T1D developed T cell responses against wheat-derived peptides. This result provides a potential mechanistic link between T1D-associated HLA types and the additional propensity to inappropriately develop T cell responses against wheat-derived proteins leading to further medical complications such as celiac disease.
Taken together, these results provide an exciting example of the power of the combination of genetic sequencing tools and mechanistic animal studies to elucidate potential mechanisms of human disease. As such, this case provides a tremendous example of the utility of animal studies in basic research to determine disease mechanism. An important next step for the field will be to take these types of results and reverse engineer solutions – which would then need to be verified in animal models of disease before application to human patients in clinical trials.