Distinct mucosal endotypes as initiators and drivers of rheumatoid arthritis

Abstract

Rheumatoid arthritis (RA) is a potentially devastating autoimmune disease. The great majority of patients with RA are seropositive for anti-citrullinated protein antibodies (ACPAs), rheumatoid factors, or other autoantibodies. The onset of clinically apparent inflammatory arthritis meeting classification criteria (clinical RA) is preceded by ACPA seropositivity for an average of 3-5 years, a period that is designated as 'at-risk' of RA for ACPA-positive individuals who do not display signs of arthritis, or 'pre-RA' for individuals who are known to have progressed to developing clinical RA. Prior studies of individuals at-risk of RA have associated pulmonary mucosal inflammation with local production of ACPAs and rheumatoid factors, leading to development of the 'mucosal origins hypothesis'. Recent work now suggests the presence of multiple distinct mucosal site-specific mechanisms that drive RA evolution. Indicatively, subsets of individuals at-risk of RA and patients with RA harbour a faecal bacterial strain that has exhibited arthritogenic activity in animal models and that favours T helper 17 (T_H17) cell responses in patients. Periodontal inflammation and oral microbiota have also been suggested to promote the development of arthritis through breaches in the mucosal barrier. Herein, we argue that mucosal sites and their associated microbial strains can contribute to RA evolution via distinct pathogenic mechanisms, which can be considered causal mucosal endotypes. Future therapies instituted for prevention in the at-risk period, or, perhaps, during clinical RA as therapeutics for active arthritis, will possibly have to address these individual mechanisms as part of precision medicine approaches.

Fig. 1 ∣. Transitions from pre-RA to clinical RA during the natural history of rheumatoid arthritis.

Model of the stages of disease in individuals with seropositive rheumatoid arthritis (RA) and in whom the initial break in tolerance to autoantigens begins in one or more of the three mucosal sites illustrated. For clarity, in this Review we define pre-RA as the period before the clinical diagnosis of RA is made. Patients with multiple risk factors but without any signs of clinical disease are considered ‘at-risk’ of RA. It is likely that the fine characteristics of the circulating autoimmunity are associated with the unique mucosal site that is involved in autoantibody formation. In addition, it is possible that the initial mucosal immune alterations affect not only the initiation of autoimmunity but also the eventual heterogenous clinical disease state, which for uncertain reasons is characterized by marked variability in the response to DMARD treatment and clinical outcomes. ACPA, anti-citrullinated protein antibody.

Fig. 2 ∣. Mechanisms potentially involved in the lung mucosal endotype of rheumatoid arthritis.

Illustration of potential mechanisms that underlie the initial mucosal inflammatory response in the lungs of a subset of individuals with preclinical rheumatoid arthritis (RA). Causal associations appear to be present between the increased local production of cytokines, chemokines, and complement activation fragments and the local production of anti-citrullinated protein antibodies (ACPAs) and rheumatoid factors (RFs). Drivers of these immune alterations and effector mechanisms are under active investigation. Potential mechanisms include the presence of dysbiosis, the composition and immune recognition of local citrullinated proteins and peptides, and the expansion of specific immune cell populations in the airways. Notably, the presence of local pulmonary mucosal inflammation seems to be associated with the future development of clinical RA, potentially through the generation of a systemic autoimmune response, as illustrated bottom right. FcR, Fc receptor; NET, neutrophil extracellular trap.

Fig. 3 ∣. Mechanisms linking dysbiosis and the gut mucosal endotype to the systemic spread of autoimmunity.

Subdoligranulum didolesgii strain 7, Prevotella copri and other bacteria that are specific to the gut traverse the mucosal barrier and cause local immune activation. In addition, bacterial antigens translocate in tertiary lymphoid structures, where they induce the generation of autoantibodies and T cells that are cross-reactive to self-antigens. In some instances, these autoantibodies and self-reactive T cells can transit to the synovium. Several cytokines such as IL-2, IL-12, IL-17 and IL-23, as well as anti-citrullinated protein antibodies (ACPAs) and rheumatoid factors (RFs), amplify the immune response, both in intestinal and in tertiary lymph tissues. Additional factors, especially metabolites generated by gut bacteria, are likely to influence these responses. T_FH, T follicular helper cell; T_H17, T helper 17 cell; T_PH, T peripheral helper cell.

Fig. 4 ∣. Mechanisms involved in the oral and periodontal endotype.

Citrullinated bacterial proteins that are recognized by anti-citrullinated protein antibodies (ACPA) are present in the oral and periodontal mucosa. Specific oral bacterial strains (listed in Table 1) are found both in the inflamed periodontal site and in the circulation of patients with rheumatoid arthritis (RA) shortly before they present a clinical flare. The proteins of such periodontal bacteria have been shown to undergo citrullination owing to the local presence of peptidyl arginine deiminases. The movement of oral commensal bacteria into the circulation is likely to be due to barrier breaks associated with periodontitis. Although the model was established by studies of patients with RA undergoing a flare, a subset of individuals at-risk of RA might exhibit the same mechanism prior to the development of synovitis. CitH, citrullinated histone; NETosis, neutrophil extracellular trap-associated cell death.