B-lymphocyte depletion for the treatment of multiple sclerosis: now things really get interesting

Conventional wisdom is that B lymphocytes are a major driving force behind autoantibody-dependent autoimmune diseases, such as systemic lupus erythematosus, and the autoimmune blistering diseases, such as pemphigus vulgaris. Other autoimmune diseases have been classically viewed as predominantly T-cell-dependent, such as multiple sclerosis (MS), rheumatoid arthritis, systemic sclerosis and Type 1 diabetes. This conclusion derives from the finding that adoptive transfer of T cells from diseased animals can initiate disease symptoms in healthy recipients. Recent assessments of the role of B cells in the immune system have demonstrated that B cells serve essential functions in regulating immune responses that were not previously appreciated [1,2]. B cells contribute to disease pathogenesis not only by the production of autoantibodies, but through their function as cellular adjuvants for CD4⁺ T-cell activation [3] and their ability to regulate T-cell function and inflammation through cytokine production [4]. Clinically, pan-mature B-cell depletion in humans using a CD20 monoclonal antibody (mAb; rituximab) is effective for treating various autoimmune disorders in some patients, such as rheumatoid arthritis [2]. Furthermore, recent Phase I and II trials using rituximab suggest clinical efficacy in MS patients [5,6]. The human clinical trial findings have been reviewed extensively [7,8], so they will not be discussed here. Nonetheless, the mechanisms underlying the effect of pan-mature B-cell depletion on disease activity in humans remains predominantly unknown.

CD20 is a B-cell-specific cell surface molecule first expressed during the pre-B to immature B-cell transition and fully expressed by mature B-cell subsets, but is lost upon plasma cell differentiation in humans and mice [9]. In mice, CD20 mAb depletes normal and malignant B cells in vivo through antibody-dependent cellular cytotoxicity, without detectable contributions from the complement system [10–12]. Unexpectedly, B-cell depletion by CD20 mAb also significantly inhibits T-cell function in mouse autoimmune disease models, including collagen-induced arthritis and Type 1 diabetes [13–15]. CD20 mAb does not deplete T-cell or other non-B-cell leukocyte subsets in mice. In most cases, serum antibody and established autoantibody levels are also not affected by CD20⁺ B-cell depletion [16]. Moreover, the rapid therapeutic benefit of B-cell depletion in MS patients is unlikely to be explained by a reduction in pathogenic autoantibodies [5,6]. Rather, in these models and with other antigens, B cells are necessary for optimal CD4⁺ T-cell but not CD8⁺ T-cell activation during immune responses to low-dose antigens, with dendritic cells sharing this duty [3]. Thereby, the therapeutic effect of B-cell depletion in mice is attributable in part to reduced autoreactive CD4⁺ T-cell activation. Studies in humans have suggested similar effects [17,18].

B cells not only activate T cells, but appear to also have opposing negative regulatory properties that can significantly inhibit autoimmune disease and inflammation [19–22]. A phenotypically unique spleen CD1d^highCD5⁺ regulatory B-cell subset has recently been identified that is responsible for most B-cell IL-10 production [23,24]. We call these cells ‘B10 cells’, since other regulatory B-cell subsets may exist. Although B10 cells only represent 1–2% of spleen B cells, they dramatically inhibit the induction of antigen-specific inflammatory reactions and autoimmunity [22]. Thereby, B cells may regulate multiple components of the immune system and causes of autoimmune disease through combinations of their multipurpose cellular and humoral functions [2].

Contradictory roles for B cells have been demonstrated in experimental autoimmune encephalomyelitis (EAE), a mouse model of MS [25–28]. EAE studies using double-transgenic mice with myelin oligodendrocyte glycoprotein (MOG)-specific T- and B-cell antigen receptors have shown that B cells can function as antigen-presenting cells during EAE initiation [25,26]. More than 50% of these mice develop inflammatory demyelinating lesions in the CNS, while disease incidence is only approximately 5% in MOG-specific T-cell receptor-transgenic mice. By contrast, congenitally B-cell-deficient mice develop a severe nonremitting form of EAE [27,28]. B-cell production of IL-10 can also inhibit EAE development [28]. These apparently contradictory results suggest either multiple roles for B cells during EAE pathogenesis or the involvement of different B-cell subsets. Using CD20 mAb, we have found that EAE disease initiation and progression are differentially influenced by the depletion of B cells from mice with otherwise intact immune systems [24]. B-cell depletion before EAE induction significantly exacerbates disease symptoms and increases encephalitogenic T-cell influx into the CNS. Worsening disease results from depletion of the B10-cell subset since the adoptive transfer of splenic CD20-deficient B10 cells before EAE induction normalizes immunopathology in otherwise B-cell-depleted mice. Regulatory B10 cells are maximally effective during early EAE initiation, but have no obvious role during disease progression, potentially owing to the emergence of regulatory T-cell function. Rather, CD20⁺ B-cell depletion during EAE progression dramatically suppresses disease symptoms. Specifically, B cells are required for CNS autoantigen specific-CD4⁺ T-cell generation and encephalitogenic T-cell entry into the CNS during disease progression. Thus, the balance of at least two opposing B-cell functions shapes the normal course of EAE immunopathogenesis. The therapeutic benefit of B-cell depletion may thereby depend on the relative contributions and timing of these opposing B-cell functions during the course of autoimmune disease.

The amelioration of disease progression in mice when CD20⁺ B cells are depleted after the onset of EAE symptoms may parallel the therapeutic benefit of rituximab for relapsing–remitting MS in humans. However, all immune-mediated inflammatory diseases may not behave the same after B-cell depletion, as rituximab treatment was recently suggested to exacerbate ulcerative colitis and trigger psoriasis, which predominately depend on T-cell-mediated mechanisms [29,30]. In addition, one case report suggests that B-cell depletion might have induced MS relapses in a patient with an 18-year history of MS who developed antimyelin-associated glycoprotein polyneuropathy that was treated with rituximab [31]. Although the B10-cell subset has yet to be definitively established in human disease, the available mouse studies raise the intriguing possibility that B10-cell depletion may induce or exacerbate autoimmunity in some susceptible individuals. In MS patients, B-cell IL-10 production is significantly lower than in healthy controls and is upregulated following therapy [32]. In addition, helminth infections induce regulatory B cells in MS patients and suppress disease activity [33], which may explain environment-related suppression of MS in areas with low disease prevalence. It therefore remains possible that pan-mature B-cell depletion may exacerbate MS occurrence in some undiagnosed cases or promote relapses. As a result, pan-B-cell-depletion therapies may actually represent a double-edged sword in the treatment of some autoimmune individuals [34].

Although the currently recognized therapeutic benefit of pan-B-cell depletion using currently available drugs (rituximab) appears to outweigh the potential negative consequences of this therapy, additional mouse studies suggest that prolonged and complete B-cell depletion may also be required to induce optimal long-term therapeutic benefit [13,14]. These collective observations open the door to future therapies designed to augment, eliminate or impair B-cell subset-specific functions [35]. Currently, B-cell subset-directed therapies targeting CD22, BLyS and CD40 are in development, as well as other potential therapies directed at specific B-cell functions. It may also be possible to identify pathways that regulate B10-cell activation, expansion and function, which will allow this potent B-cell subset to be manipulated for therapeutic benefit. For example, the selective expansion of autoantigen-specific B10 cells may be sufficient to blunt the induction of autoimmune disease. One might even envision the use of this approach in preclinical autoimmune diseases, such as Type 1 diabetes. Alternatively, an approach that selectively depletes mature B cells while sparing regulatory B10 cells may offer a particularly potent therapy for MS and other immune-mediated inflammatory diseases. Now that the door for B-cell-directed therapies has been opened, new treatments addressing the regulatory complexity of the immune system appear more promising, with the likelihood that combination therapies directed at both the T- and B-cell arms of the immune system will be needed to address the complexity of autoimmune disease.

Financial & competing interests disclosure

This work was supported by NIH grants (CA105001, CA96547 and AI56363). Takashi Matsushita is supported by fellowship from Japan Society for the Promotion of Science. Thomas Tedder is a paid consultant for MedImmune, Inc. and Angelica Therapeutics, Inc. and shareholder of Angelica Therapeutics. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.