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

Expansion of Localized Autoimmunity to Joints and Distal Sites

As mentioned above, antibodies are released in the peripheral blood and circulated in the body for years before any clinical sign of joint inflammation.[5–7,112–115] This strongly suggests that antibodies are generated at extra-articular sites, but does not completely exclude the possibility that they might still be produced in the joints in the absence of any macroscopic and/or microscopic signs of inflammation. These healthy individuals lack not only clinical complaints, but also any clear-cut sign of joint inflammation,[9,116] raising the possibility that antibodies are passive bystanders of the disease. However, ACPAs are able to exacerbate existing minimal joint disease by passive transfer in mice[117] and possess several effector pathogenic functions. ACPAs activate the complement system[118] and promote macrophage activation when incorporated in immune complexes via either Fcγ receptors or TLR-4-dependent mechanisms.[119] More recently, antibodies against mutated citrullinated vimentin were shown to promote bone resorption in vitro and to induce osteoclastogenesis by adoptive transfer into mice.[120] ACPAs can also stimulate neutrophils to release neutrophil-derived extracellular traps and promote inflammation[121] (Fig. 2).

Figure 2.

A schematic representation of how mucosal disequilibrium might lead to generation of autoimmunity and later to joint disease development
Complex mechanisms dependent on environmental exposure and the host microbiome are responsible for maintaining homeostasis at mucosal surfaces (such as the respiratory and gastrointestinal tract). In genetically susceptible hosts, failure of these mechanisms may lead to mucosal disequilibrium and molecular changes such as post-translational modifications (citrullination), with subsequent antigen presentation by professional antigen presenting cells, activation of immune effector T cells (such as Th17) and relative deficiency of Tregs. These changes lead to activation of B cells and generation of antibodies (such as ACPAs) by plasma cells. These antibodies undergo somatic hypermutation and epitope spreading, leading to joint disease initiation and perpetuation through several mechanisms (complement activation, cell surface receptor activation, ligation of cell surface components and neutrophil activation).

Despite these advances in understanding how ACPAs might contribute to perpetuation of joint inflammation, it remains unclear what event or series of events is required for the delayed initiation of this joint inflammation, although there are several possibilities. First, it is plausible that a yet unidentified second hit (such as minor trauma or transient infection/microbiota community alteration) could lead to expression of citrullinated proteins in an otherwise citrullinated protein-poor healthy joint[67] and that these antigens would then be targeted by the pre-existing circulating autoantibodies, ultimately leading to clinical signs of joint inflammation. This would imply that circulating autoantibodies that may have been generated in response to mucosal antigens would target similar antigens in the joint. Supporting this, it has recently been shown that the same citrullinated peptides could be identified by mass spectrometry in the lungs and joints of RA patients.[122] Second, it is possible that sites other than the synovial membrane are the first joint component to be affected, with secondary synovial involvement being an epiphenomenon. In line with this, recent studies utilizing microcomputer tomography showed that signs of bone destruction are present before clinical onset of synovial inflammation.[121] Third, progressive epitope spreading[123] and the emergence of subclinical inflammation (as seen for certain cytokines and chemokines)[124,125] might be needed to alter the number and/or specificity profile required by ACPAs to gain these effector functions. In this context, it is worth mentioning that all available evidence showing direct pro-arthritogenic properties of ACPAs have been obtained using antibodies purified from peripheral blood and/or SF of patients with already established disease. This leaves open the possibility that antibodies might emerge from mucosal interactions first as non-pathogenic immunoglobulins during the preclinical phase only to gain arthritogenic properties through epitope spreading or pathogenic changes in avidity, affinity or Fc function at a later time point. These changes could potentially allow ACPAs to target the joints and cause inflammation. These issues remain to be addressed in high-quality natural history studies of RA that can include broad evaluations of innate and adaptive immunity at mucosal, systemic and joint sites.