Gene, Environment, Microbiome and Mucosal Immune Tolerance in Rheumatoid Arthritis

Anca I. Catrina; Kevin D. Deane; Jose U. Scher


Rheumatology. 2016;55(3):391-402. 

In This Article

Environmental and Genetic Factors Contributing to the Generation of Autoimmunity at Mucosal Sites

Several indirect lines of evidence support mucosal surfaces as sites of generation of RA-related autoimmunity. In particular, Barra et al.[61] demonstrated that the proportion of IgA ACPAs was higher than IgG in subjects at risk for future RA; intriguingly, in patients with established RA, the proportion of IgG ACPAs was higher than IgA. This finding supports the notion that early RA-related autoimmunity may be triggered by mucosal processes generating IgA-related autoimmunity, later transitioning into an IgG response that eventually leads to clinical manifestations of disease (i.e. synovitis). There have also been associations (albeit controversial) between serological evidence for certain infections, including Proteus and Mycoplasma species, and risk for RA.[62–64] In addition, Mikuls et al.[65] demonstrated an association between RA-related autoantibodies in healthy, first-degree relatives of probands with RA and serum antibody titres against Porphyromonas gingivalis—an organism implicated in human periodontal disease (PD)—suggesting that immune responses to P. gingivalis may play a role in early RA-related autoimmunity.

There are several mechanisms by which the mucosa might contribute to the development of RA-associated autoimmunity (Table 1). Cross-reactivity may occur between mucosal antigens (which may be microbial in origin) and self-proteins, a process thought to occur in rheumatic fever, where pharyngeal infection with certain species of Streptococcus leads to the generation of antibodies that cross-react with self-antigens in the heart and other tissues.[66] In addition, mucosal generation of neo-antigens (e.g. citrullination) by pathogen-mediated inflammation and peptidylarginine deiminase activation, or carbamylation, could occur at mucosal sites through microbial-related respiratory burst.[67–69]

In RA, the finding of local enrichment of ACPA in both bronchoalveolar fluid of early untreated RA patients and in induced sputum of arthritis-free individuals at risk of developing RA[70] (in some cases, even in the absence of the same antibodies in the blood) supports the notion that autoimmunity might indeed originate at mucosal sites. In line with this, tissue inflammation and ectopic lymphoid structures were described in bronchial biopsies of patients with early untreated ACPA-positive RA in the absence of any other associated lung disease.[71] This was also shown in lung biopsies of ACPA-positive individuals with chronic lung disease but no signs of joint inflammation.[72] Interestingly, similar structures containing citrullinated protein-binding B cells have also been reported in patients with established RA and associated lung disease.[73] Signs of lung inflammation have further been described by high-resolution CT imaging of both ACPA-positive healthy individuals at high risk for RA development, even in the absence of smoking,[8] and patients with early untreated ACPA-positive RA.[74] An alternative explanation for ectopic lymphoid tissue formation is a secondary immune injury of the lungs induced by ACPA in the presence of citrullinated proteins in the lungs. However, these two different scenarios are complementary and not mutually exclusive.

In additional support of a mucosal-based trigger for RA, exposure to tobacco smoke is the strongest environmental risk factor associated with RA, with some estimates that smoking explains ~30% of the risk for ACPA-positive RA.[75] However, despite this association, the role of smoking in disease pathogenesis is still uncertain. Original studies showed that smoking increases the expression of citrullinated proteins in the lungs of healthy smokers.[76] This was later confirmed for early untreated RA patients as well.[74] Interestingly, increased expression of citrullinated proteins was present not only in smokers, but also in ACPA-positive non-smokers, suggesting that factors other than smoking might also contribute to the generation of citrullinated epitopes in the lungs.

The presence of susceptibility genes (and in particular HLA-DR SE) in smokers further increases the risk of developing seropositive RA, as demonstrated by several large epidemiological investigations.[77–83] This might be explained through specific interaction of the susceptibility genes with citrullinated but not native (i.e. arginine) epitopes.[84–86] In addition, peripheral blood B cells of both RA patients and healthy individuals carrying the SE alleles show increased ACPA production when exposed to smoking, suggesting that both environment and genetics influence the pool of autoreactive B cells in healthy subjects.[87] Somewhat controversial data have recently emerged from studies in animal models exposed to chronic cigarette smoking.[88] In these experiments, exposure of HLA-DR4 transgenic mice to cigarette smoking unexpectedly suppressed CIA, while enhancing innate immunity and mounting a robust response to citrullinated vimentin. In contrast, similar exposure in HLA-DQ8 transgenic mice (occurring in linkage with DR4 in humans) not only augmented the antigen-specific adaptive T cell responses to native and citrullinated proteins, but also worsened the course of CIA.

One interpretation of these findings is that DR4 contributes to autoimmunity by enhancing innate immune responses potentially following bacterial challenge, while DQ8 might contribute to antigen-specific autoreactive processes. However, the arthritis-suppressive effect of smoking in HLA-DR transgenic mice and the relevance of these findings in human disease remain to be elucidated.