Characterisation of epitope-specific B cells and the role of IgG Fc sialylation in murine arthritis models

Abstract

Autoimmune diseases are a major burden for society and economy. One of the most common autoimmune diseases is rheumatoid arthritis (RA) characterized by joint inflammation, swelling and pain. If left untreated, it ultimately leads to disability.

B cells play a major role in rheumatoid arthritis as shown by numerous studies and the success of anti-CD20 B cell depletion therapy in clinical use. In healthy individuals, B cells originate and develop in the bone marrow before they exit into the periphery where they mature and differentiate. B cells act in the body in a dual role, on the one hand as professional antigen-presenting cells (APCs), taking up antigen and presenting it to cognate T cells via MHC. On the other hand, B cells can generate and mature specific antibodies against almost any given antigenic target. This diversity is achieved by a process called V(D)J recombination. V, (D) and J segments are recombined during early B cell development, creating a vast diversity of unique sequences. Subsequently, the BCR diversity can be increased even further upon B cell activation by somatic hypermutation (SHM) and class switch recombination (CSR) in the germinal center (GC) reaction.

However, the stochastic process of gene rearrangement and mutations also leads to autoreactive BCRs that upon encounter of self-antigen could initiate an autoreactive immune response. This fundamental dogma of autoimmunity has been postulated in many autoimmune diseases including RA. Therefore, it is important to understand the role and regulation of self-reactive B cells in arthritis.

In study I, we investigated the role of Ncf4 dependent intracellular reactive oxygen species (ROS) in controlling the differentiation from B cells into antibody-producing plasma cells. By employing Ncf4 mutated (R58A) mice we could show that lower levels of intracellular ROS induce plasma cell formation, increased CXCR3 and decreased CXCR4 expression. Increased plasma cell formation led to higher self-reactive anti-Col2 antibody production and enhanced disease severity. We concluded that intrinsic ROS can substantially modulate self-reactive B cell differentiation and thereby the disease phenotype.

In study II, we mapped the pathogenic B cell epitopes on glucose-6-phosphate isomerase (GPI), an autoantigen in human RA and mouse models of arthritis. We identified a single peptide (GPI293-307) as the most dominant epitope targeted both in human and mouse models. Following the observation of the early emergence of antibodies against these epitopes, we found that GPI293-307-specific B cells are not deleted during B cell development and activated in the pathogenesis of arthritis due to encounter of a structurally modified form of GPI protein on the articular cartilage surface exposing the GPI293-307 neoepitope. We concluded that this finding has important implications about the understanding of the role of naturally existing self-reactive B cells in GPI-mediated arthritis.

In study III, we studied the B cell response to Collagen type II, another important autoantigen in human RA and mouse models of arthritis. The focus was on the C1 epitope, a triple-helical peptide sequence originally identified in collagen-dependent mouse models. Using both B cell receptor knock-in mouse models, healthy human donors, and RA patients, we could identify and characterise for the first time a population of physiologically occurring self-reactive B cells that exhibit an immune-suppressive phenotype. We conclude that B suppressor cells (Bsups) represent a new subpopulation of B cells tolerizing T cells against self-antigens.

Finally, in study IV, we changed the focus from the specificity of self-reactive BCRs as conferred by the Fab fragment of immunoglobulins (Igs) to the importance of the Fc fragment. More specifically, we addressed the question of the interactions between the Immunoglobulin G (IgG) isotype, its Fc glycosylation, and the consequent effector functions in different murine arthritis models. To this end, we made use of available well- characterized pathogenic monoclonal antibodies (mAbs) causing arthritis in susceptible mouse strains. A glycoengineering method was applied on these antibodies to yield an array of different glycovariants to be tested in murine models. We showed that while the Fc glycan does not play a role in the collagen antibody-induced arthritis (CAIA) model, there seems to be an isotype -/and sialylation-dependent function in the regulation of arthritis severity in the collagen-induced arthritis (CIA) model.

In conclusion, we contributed towards the understanding of naturally existing autoreactive B cells in arthritis autoimmunity.