B cells from rheumatoid arthritis patients show important alterations in the expression of CD86 and FcγRIIb, which are modulated by anti-tumor necrosis factor therapy

Rheumatoid arthritis (RA) is a chronic, inflammatory, and autoimmune disease that affects mainly synovial joints, leading to progressive destruction, pain, and disability. It is well known from mouse models that B cells play a pivotal role in the development of the autoimmune process as a precursor of antibody-secreting cells but also as antigen-presenting cells (APCs) [1, 2].

Immune cells express an array of receptors that bind the Fc portion of IgG-containing immune complexes (FcγRs). Particularly, it has been stated that B cells and plasma cells express only the low-affinity receptor FcγRIIb, which, in contrast to FcγRIIa, has an immunoreceptor tyrosine-based inhibitory motif on the cytoplasmic domain. This characteristic confers an inhibitory function to the receptor which is essential in several checkpoint stages in which abnormal humoral responses are quenched by mechanisms that include the deletion of autoreactive clones and feedback inhibition of IgG secretion [3].

Given this property, it is not surprising that these molecules have been involved in autoimmune processes. Autoimmune-susceptible mice present several polymorphisms in the regulatory regions of the FcγRIIb gene, which result in a reduced expression of the receptor on germinal center B cells [4]. Moreover, depending on the strain, mice deficient in FcγRIIb can spontaneously develop a lupus-like syndrome, become susceptible to collagen-induced arthritis (CIA), or develop a severe phenotype of CIA or experimental autoimmune encephalomyelitis [5‐8]. In contrast, overexpression of FcγRIIb on B cells, but not on macrophages, leads to an early resolution of CIA and reduced spontaneous lupus [9].

On the other hand, human autoimmune diseases characterized by a deregulated secretion of autoantibodies, such as systemic lupus erythematosus (SLE) and RA, have been associated with abnormalities in FcγRIIb regulation. Polymorphisms in the promoter region as well as in the transmembrane domain of the FcγRIIb gene have been described to affect the expression and function of this receptor, respectively [10‐12]. While both polymorphisms in FcγRIIb are associated with SLE occurrence [10, 13], the one on the transmembrane domain is also associated with joint damage in RA [14]. Although alterations in the expression of FcγRIIb on B cells have been described for other autoimmune diseases [15‐18], no data about RA are available.

The aim of our study was to evaluate the phenotype of B cells from RA patients, focusing on their activation status and their expression of FcγRIIb. These parameters were compared with those obtained from B cells of healthy individuals. In addition, we followed up on these patients during anti-tumor necrosis factor (anti-TNF) therapy and assessed the phenotype of their B cells after 6 months of treatment. Our findings show that B cells from RA patients are activated, as reflected by the expression of CD86. We have also observed an altered expression of FcγRIIb, which is associated with the presence of autoantibodies. These abnormalities were shown to be partially reverted by anti-TNF therapy.

In the present work, we provide evidence that phenotypic alterations on B cells from RA patients affect key molecules involved in the regulation of antigen presentation and antibody secretion functions. A major role of B cells in the development and perpetuation of RA has been consistently demonstrated with the appearance of B cell-depleting therapy and its impressive results in reducing symptoms and preventing disease progression [27]. Studies on murine models have suggested that antigen-specific B cells are required as APCs for the induction of autoimmune arthritis, owing to their expression of MHC (major histocompatibility complex) class II and co-stimulatory molecules CD80 and CD86 [28, 29]. CD86 is upregulated on activated B cells upon B-cell receptor (BCR) and CD19/CD21 complex engagement [24, 25]. Our results show that RA patients have a higher proportion of naïve and memory B cells expressing CD86 than healthy controls, reflecting an expanded activated status within these subpopulations, which is likely to favor a more productive interaction with pathogenic T cells. Analogous results have been reported for other inflammatory diseases, such as SLE [30‐33], systemic sclerosis [34], asthma [35], and irritable bowel syndrome [36]. Presumably, constant stimulation of autoantigens through BCR and the influence of other proinflammatory signals, such as stimulation through Toll-like receptors (TLRs) [37], give rise to this activated phenotype. Also, it is probable that the effect of these activation stimuli could be attenuated by the cross-linking of IgG-containing immune complexes to their inhibitor receptor, FcγRIIb, since studies on dendritic cells (DCs) have demonstrated that exposure to immune complexes together with a blockage of FcγRIIb is sufficient to induce an increase in the expression of CD86 [38, 39], whereas the overexpression of this receptor on murine B cells reverses the induction of CD86 triggered via BCR [9]. However, in a multifactorial complex disease such as RA, it is expected that the expression of this or other activation markers is influenced by a variety of factors as it is suggested by the absence of correlation between CD86-expressing B cells and FcγRIIb expression levels found in our patients.

Our data show that RA patients present reduced levels of FcγRIIb on memory B cells and plasmablasts compared with healthy donors. This phenomenon can be explained by the abnormal downregulation of this receptor from naïve to memory B cells which we observed in our RA group. These results are concordant with those seen in SLE and chronic inflammatory demyelinating polyneuropathy, other autoimmune diseases characterized by uncontrolled secretion of autoantibodies [15‐18].

It has been postulated that FcγRIIb upregulation might constitute a critical checkpoint in peripheral tolerance by providing an inhibitory feedback that limits the ongoing humoral response to self-antigens. In fact, in lupus-prone mice, the restoration of FcγRIIb levels on B cells can revert the secretion of autoantibodies and renal disease [40]. Furthermore, the expression of this inhibitory receptor specifically on B cells, but not on macrophages, can be determinant for controlling autoimmunity in models of arthritis and lupus [9]. Through this work, we provide new evidence that may help to reinforce this concept as we have revealed, for the first time, an association between high levels of FcγRIIb on memory B cells and no or low titers of specific autoantibodies, anti-MCV antibodies. This interesting result appears to be exclusive for autoimmune responses since we did not find a similar association when analyzing total IgG levels. It is noteworthy to consider that in this study we have examined only the expression of FcγRIIb, but not its function, which if altered could also affect the regulatory ability over the humoral response against citrullinated proteins. Likewise, a recent publication has demonstrated an association between a functional polymorphism for FcγRIIb and anti-ccp (+) RA in an Asian population [41].

After patients underwent 6 months of therapy with an anti-TNF antibody, we observed a decrease in CD86-expressing memory B cells, reflecting an attenuation of the B-cell activated status. Paradoxically, FcγRIIb expression on the naïve B-cell subset also decreased significantly. It has been demonstrated that FcγRIIa and FcγRIIb on human monocytes are differentially regulated by Th1/Th2 cytokines, with interferon-gamma favoring the activator receptor and interleukin-4 favoring the inhibitory receptor [42]. On the other hand, TNF downmodulates FcγRIIb and FcγRIIa on monocytes, not affecting FcγRIIa but reducing FcγRIIb expression on DCs, while increasing FcγRIIa without changing FcγRIIb on neutrophils, indicating a tight cell-specific regulation [43‐47]. To our knowledge, no studies addressing the effect of TNF over FcγRIIb on human B cells have been published, but our results strongly suggest that TNF or other downstream cytokines may influence the expression of this receptor on B lymphocytes. In regard to RA, exposure of DCs to synovial fluid from RA patients has been shown to lead to an upregulation of FcγRIIb [14] and an elevated expression of FcγRIIb has been demonstrated on RA synovial tissue, probably counteracting the upregulation of other activating receptors [48]. Some studies have reported that FcγR expression levels on leukocytes can vary with anti-rheumatic drugs, which would promote a more inhibitory profile [48‐52]. It is possible that the reduction in FcγRIIb expression that we observed on naïve B cells as a consequence of anti-TNF therapy is accompanied by a decrease in the expression of activating receptors on these cells, like TLR9 and CD21, which would determine a restoration of the protective activator/inhibitor balance, but this issue needs to be investigated. The effect of TNF blockage, however, was not sufficient to prevent the downregulation of FcγRIIb from naïve to memory B cells. These results are in accordance with the fact that our group of patients did not achieve a reduction in anti-MCV titers after 6 months of therapy. Others have reported that changes in these antibodies become significant after 18 months of anti-TNF therapy [53], so it is conceivable that a full normalization of B-cell phenotype may become apparent only over longer follow-up periods.

Our data demonstrate the existence of important alterations in the phenotype of peripheral B cells from RA patients, involving the expression of the co-stimulatory molecule CD86 and the inhibitory receptor FcγRIIb, the latter being associated with high titers of autoantibodies. We consider that our study contributes relevant evidence to a better comprehension of the molecular mechanisms that are implied in the regulation of B cells and the role that they play in the autoimmune response elicited in RA.