The role of B cells in the immunopathogenesis of multiple sclerosis

Summary

There is ongoing debate on how B cells contribute to the pathogenesis of multiple sclerosis (MS). The success of B‐cell targeting therapies in MS highlighted the role of B cells, particularly the antibody‐independent functions of these cells such as antigen presentation to T cells and modulation of the function of T cells and myeloid cells by secreting pathogenic and/or protective cytokines in the central nervous system. Here, we discuss the role of different antibody‐dependent and antibody‐independent functions of B cells in MS disease activity and progression proposing new therapeutic strategies for the optimization of B‐cell targeting treatments.

B cells can contribute to MS pathogenesis via production of antibodies against CNS antigens. The success of B‐cell targeting therapies in MS highlighted the antibody‐independent functions of B cells in MS. B cells can contribute to MS through antigen presentation to T cells and modulation of the function of T cells and myeloid cells by secreting pathogenic and/or protective cytokines.

Abbreviations

APC

antigen‐presenting cell

CAAR

chimeric autoantibody receptor

CNS

central nervous system

CSF

cerebrospinal fluid

DMF

dimethyl fumarate

EAE

experimental autoimmune encephalomyelitis

GA

glatiramer acetate

GITRL

glucocorticoid‐induced tumor necrosis factor receptor‐ligand

GM‐CSF

granulocyte–macrophage colony‐stimulating factor

IL‐21

interleukin‐21

LT‐α

lymphotoxin α

MHC‐II

major histocompatibility complex II

MOG

myelin oligodendrocyte glycoprotein

MS

multiple sclerosis

OCBs

oligoclonal bands

PMN‐MDSCs

polymorphonuclear myeloid‐derived suppressor cells

SPAG16

sperm‐associated antigen 16

STAT

signal transducer and activator of transcription

TACI

transmembrane activator and CAML interactor

Tfh cell

follicular helper T cell

Th1

T helper type 1

Th17

T helper type 17

TNF‐α

tumor necrosis factor α

VLA‐4

very late antigen‐4

Introduction

Multiple sclerosis (MS) is an autoimmune demyelinating disease of the central nervous system (CNS) that causes myelin loss and axonal damage leading to severe neurological disability commonly in young adults. ¹ , ² , ³ , ⁴ The disease course is initially relapsing–remitting for most individuals with MS (about 85% of patients), which is characterized by episodes of acute neurological dysfunction followed by full or partial recovery. Most individuals with relapsing–remitting MS develop secondary progressive MS after 10–20 years of disease evolution, a progressive course that is characterized by continuous neurological deterioration. Approximately, 10% of patients exhibit a progressive disease course from clinical onset, which is called primary progressive MS. The least common form of disease (<5%) is similar to primary progressive MS but with overlapping relapses and is called progressive relapsing MS. ⁵ Although the precise etiology of MS is unclear, CNS antigen‐specific‏ T cells have been believed to have the key role in MS development. It is believed that following activation of these autoreactive T cells in the periphery through some infectious agents such as Epstein–Barr virus, which are molecular mimics of CNS antigens, they can enter the CNS and drive the inflammatory response leading to destruction of myelin and axon of nerve cells and formation of demyelinating plaques. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ Studies have demonstrated that T‐cell receptors from MS patients can recognize both DRB1*1501‐restricted myelin basic protein peptide and DRB5*0101‐restricted Epstein–Barr virus peptide, and CD4⁺ T cells isolated from the cerebrospinal fluid (CSF) of MS patients specific for Epstein–Barr virus peptide, can cross‐recognize an immunodominant myelin basic protein peptide highlighting the key role of T cells in MS pathogenesis. ⁶ Correspondingly, therapeutic approaches for decades have aimed to limit effector T‐cell responses and most approved MS therapies have focused on targeting T cells or correcting the balance between effector and regulatory T‐cell responses. ¹¹ , ¹²

Despite considerable evidence, such as infiltration of B cells in brain lesions and detectable oligoclonal bands (OCBs) in the CSF of approximately 90% of MS patients, which strongly suggests an involvement of B cells in MS, their role in MS has been underestimated and incompletely understood for decades. ¹³ , ¹⁴ Recently, an updated insight into the role of B cells in the immunopathogenesis of MS has emerged following the success of studies testing B‐cell‐depleting therapies with anti‐CD20 monoclonal antibodies as a therapeutic approach, which suppressed inflammatory disease activity and limited new MS relapses. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ Interestingly, these clinical studies have revealed that while T cells decrease after treatment with anti‐CD20 monoclonal antibodies, serum and CSF antibodies and long‐lived plasma cells remain unchanged, suggesting involvement of B cells in MS pathogenesis through antibody‐independent mechanisms such as antigen presentation to T cells and/or secretion of pro‐inflammatory or anti‐inflammatory cytokines. ²⁰ , ²¹ , ²² , ²³ , ²⁴ These findings triggered numerous investigations and debates among scientists regarding the different roles of B cells in MS pathogenesis and progression. In the following sections, we review the antibody‐dependent and antibody‐independent roles of B cells in MS and will discuss the mechanisms by which the clinical trial findings on different B‐cell‐depleting therapies can support basic investigations to delineate a concept for the role of B cells in MS.

The role of antibodies in MS

The involvement of B cells and antibodies in the pathogenesis of MS was demonstrated first by the discovery of OCB as a result of elevated IgG and IgM production in the CSF of individuals with MS. ²⁵ , ²⁶ The IgG OCBs can be found in approximately 90% of MS patients but IgM OCB can be found in only 30%–40% of patients, and are commonly associated with disease activity and therapeutic responses to B‐cell‐depleting therapy. ²⁷ , ²⁸ Elevated numbers of clonally expanded B cells and oligoclonal immunoglobulin bands in the CSF are common diagnostic hallmarks of multiple sclerosis. ²⁹ By comparing the immunoglobulin transcriptomes of B cells with the corresponding immunoglobulin proteomes in the CSF of MS patients, Obermeier et al. demonstrated that the source of these intrathecally produced antibodies was the clonally expanded B cells in the CSF that can also be found in other CNS subcompartments such as meninges and parenchyma in the same patient but that differ among patients. ²⁹ Studies have reported evidence for somatic hypermutation and affinity maturation of these B cells within the CSF. ³⁰ , ³¹ Moreover intrathecal somatic hypermutation studies have shown the existence of identical B‐cell clones in the CNS and the periphery, indicating bidirectional trafficking of these B‐cell clones both into and out of the CNS. ³² , ³³ , ³⁴ This trafficking is regulated by chemoattractants such as CXCL10, CXCL12 and CXCL13, which have been shown to be elevated in the CSF of MS patients. ³⁵

Further evidence for the involvement of antibodies in MS pathogenesis is provided by the identification of antibodies that bind to myelin fragments within phagocytes and also complement depositions in perivascular MS lesions. ³⁶ , ³⁷ Moreover, studies have shown that antibodies produced by clonally expanded plasma cells in the CSF of MS patients can bind to CNS tissue, causing complement‐mediated demyelination ³⁸ , ³⁹ (Fig. 1a). Some CNS auto‐antigens have been identified as targets for serum or CSF antibodies, including myelin oligodendrocyte glycoprotein (MOG), myelin basic protein myelin‐derived lipids, neurofascin, contactin‐2, KIR4·1 and also some intracellular auto‐antigens such as DNA and RNA. ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ , ⁴⁷ Moreover, studies have identified antibodies of CSF OCBs that primarily recognize ubiquitous intracellular self‐proteins, suggesting that the generation of CSF OCBs might be as a response against dead‐cell debris. ⁴⁸ , ⁴⁹ Another target that has been identified for CSF autoantibodies in MS is sperm‐associated antigen 16 (SPAG16). ⁵⁰ Increased plasma levels of anti‐SPAG16 antibodies with 95% specificity have been reported in 21% of MS patients and their seropositivity has been shown to be associated with an increased Expanded Disability Status Scale in MS patients and with an increased Progression Index in primary progressive MS. ⁵¹ , ⁵² It is believed that during CNS inflammation and astrocytic damage, the release of intracellular SPAG16 from astrocytes causes the formation of anti‐SPAG16 antibodies in the CNS, which can damage neurons through complement activation. ⁵² These findings confirm the role of B cells and antibody‐dependent CNS damage in MS disease severity and progression.

Figure 1.

Different roles of B cells in the pathogenesis of multiple sclerosis. (a) B cells can differentiate into plasma cells and secrete autoantibodies that contribute to central nervous system (CNS) inflammation via opsonization of CNS antigens and complement fixation. (b) B cells can recognize and internalize the CNS antigens and act as antigen‐presenting cells (APCs) to activate CNS‐specific pathogenic T cells. (c) Different subsets of B cells in the CNS can modulate T‐cell and myeloid cell functions by secretion of pro‐inflammatory and anti‐inflammatory cytokines. B cells also produce interleukin‐6 (IL‐6) and present it in trans through their own IL‐6Rα, inducing Ly6G⁺ cell differentiation into polymorphonuclear myeloid‐derived suppressor cells (PMN‐MDSCs) and these MDSCs suppress B‐cell activation and cytokine production in a negative feedback loop.

However, despite remarkable investigations, the definition of different specific antigens in the CNS that can be recognized by CSF antibodies and the pathogenic roles of peripheral anti‐myelin antibodies in MS have been elusive. CNS‐specific antibodies in MS were conventionally believed to enhance demyelination in CNS tissue. ⁵³ However, recent studies have indicated that peripheral anti‐myelin antibodies can activate peripheral myelin‐reactive T cells by opsonization of CNS antigens in deep cervical lymph nodes. ⁵⁴ , ⁵⁵ Overall, investigations are ongoing in this field to find the target antigens for different local and peripheral CNS‐specific antibodies to exactly describe the role of antibodies in MS.

The role of B cells as antigen‐presenting cells in MS

In addition to producing antibodies and secreting cytokines, B cells can also act as antigen‐presenting cells (APCs) (Fig. 1b). The APC capability of B cells has been demonstrated in many human diseases, such as infectious diseases and autoimmune diseases. ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ Disease‐relevant memory B cells in the spleen and lymph nodes of MS patients can act as very effective APCs by binding to specific antigens through their B‐cell receptor, internalizing and presenting them to CNS‐specific pathogenic T cells. Pathogenic T cells in MS and experimental autoimmune encephalomyelitis (EAE) including T helper‐17 (Th17), T helper‐1 (Th1), follicular helper T (Tfh) and Tfh‐like cells are activated in the CNS‐draining lymph nodes, and B cells as APCs in cooperation with these T cells drive disease activity and progression in MS and EAE. Recently, we investigated the role of interleukin‐21 (IL‐21)‐producing T cells in MS and found that secretion of IL‐21 by Th17, Tfh and Tfh‐like cells can promote disease severity and progression, likely by helping B cells to produce antibodies and driving the inflammatory B‐cell response in the CNS, indicating the importance of T cell–B cell cooperation in MS pathogenesis. ⁶⁰ , ⁶¹ The expression of co‐stimulatory CD80, CD86 and CD40 molecules by B cells can effectively induce activation, expansion and differentiation of CNS‐specific pathogenic T cells. Compared with professional APCs, B cells are more efficient at presenting protein antigens and appear to be the best APCs in low antigen concentrations as they can recognize neuroantigens by their specific receptor. ⁶² , ⁶³

Compared with healthy individuals, in MS patients, most memory B cells have been demonstrated to be CNS antigen‐specific B cells, which can potently present myelin antigens, such as MOG and myelin basic protein to T cells and activate them. ⁶⁴ , ⁶⁵ Memory B cells have also been reported to be involved in antigen transportation to follicular dendritic cells in germinal centers. ²⁰ In the interactions between APCs and T cells, co‐stimulatory molecules such as CD80 and CD86 play a pivotal role in T‐cell activation and proliferation. In comparison with the B cells of healthy individuals, higher expression of such co‐stimulatory molecules has been reported in B cells of MS patients, which can induce activation and proliferation of T cells in response to CNS antigens such as myelin basic protein. ⁶⁶ The frequency of CD80⁺ B cells also has been shown to be abnormally increased in MS patients with active disease. ⁶⁷ B cells in the CNS have been demonstrated to express co‐inhibitory molecules such as programmed death‐ligand 1 and glucocorticoid‐induced tumor necrosis factor receptor‐ligand (GITRL), which have a role in down‐regulation of T‐cell responses. ⁶⁸ , ⁶⁹ GITRL expression by B cells has been shown to induce differentiation of regulatory T cells. ⁶⁹ In vivo studies in EAE mice also supported APC function of B cells. Mice with selective knockout of major histocompatibility complex II (MHC‐II) in B cells have been shown to be resistant to MOG‐induced EAE and to have diminished Th1 and Th17 responses. ⁵⁸ , ⁷⁰ Although B‐cell‐specific knocking out of MHC‐II causes a decrease of anti‐MOG production by EAE mice, anti‐MOG administration only partially restored EAE susceptibility, highlighting the MHC‐II‐dependent APC function of B cells in EAE. ⁵⁸ Moreover, selective knockout of co‐stimulatory CD80 and CD86 genes in B cells has been shown to decrease T‐cell responses, highlighting the significant role of B‐cell–T‐cell interactions and APC functions of B cells in MS. ⁷¹ Overall, these findings confirm the concept that antigen‐specific B cells in the CNS can function as potent APCs in MS pathogenesis.

The role of cytokines secreted by B cells in MS

Numerous studies have reported the distinct cytokine profile of B cells and their abnormal pro‐inflammatory and anti‐inflammatory cytokine balance in MS. ⁷² , ⁷³ , ⁷⁴ , ⁷⁵ , ⁷⁶ , ⁷⁷ B cells of MS patients have been shown to produce abnormally high levels of IL‐6, tumor necrosis factor α (TNF‐α), lymphotoxin α (LT‐α), and granulocyte–macrophage colony‐stimulating factor (GM‐CSF) compared with normal individuals. ⁷² , ⁷⁵ , ⁷⁷ , ⁷⁸ Enhanced production of these pro‐inflammatory cytokines by B cells can activate other immune cells leading to disease progression (Fig. 1c). IL‐6 secreted by B cells can induce Th17 cell differentiation and conversely can inhibit differentiation of regulatory T cells. ⁷⁹ , ⁸⁰ Correspondingly, ablation of IL‐6‐producing B cells has been shown to result in reduced Th17 cell differentiation and EAE amelioration. ⁵⁸ , ⁷³ Both LT‐α and TNF‐α are pro‐inflammatory cytokines produced by B cells of MS patients in high amounts. ⁷⁴ Overexpression of microRNA‐132 in B cells has been reported to play an important role in abnormally high production of LT‐α and TNF‐α by these cells in MS patients. ⁷⁸ Bar‐Or et al. have shown that while B‐cell depletion using rituximab significantly diminishes pro‐inflammatory Th1 and Th17 responses, soluble products from activated B cells of untreated MS patients can reconstitute the reduced pro‐inflammatory T‐cell response, which appeared to be mainly mediated by LT‐α and TNF‐α secreted by B cells. ⁷²

Studies also have demonstrated that B cells are capable of regulating immune responses by producing anti‐inflammatory cytokines such as IL‐10, IL‐35 and transforming growth factor‐β. ²³ Interestingly, evidence from clinical trials has shown that B cells that reconstitute after B‐cell depletion therapy are different from the B cells of untreated patients and tend to produce high levels of anti‐inflammatory IL‐10 but low levels of pro‐inflammatory cytokines such as IL‐6, TNF, LT and GM‐CSF and contribute to persistent tolerance. ⁷³ , ⁷⁴ , ⁷⁷ Understanding how these reconstituting B cells lose their pro‐inflammatory characteristic and exert regulatory functions requires further investigation. Moreover, adoptive transfer of IL‐10‐secreting B cells has been reported to suppress EAE in an IL‐10‐dependent manner. ⁸¹ , ⁸² Consistently, Fillatreau et al. have shown that selective knocking out of B‐cell IL‐10 in mice increases EAE severity and prevents recovery from disease, suggesting the regulatory properties of B cells in EAE. ⁸³

Alteration of the gut microbiome has also been reported to induce IL‐10‐producing B cells, resulting in decreased EAE severity. ⁸⁴ By using IL‐10 reporter mice, Matsumoto et al. demonstrated that plasmablasts in the draining lymph nodes, but not splenic B cells, express IL‐10 and limit autoimmune inflammation, indicating the importance of plasmablasts as IL‐10‐producing regulatory B cells. ⁸⁵ They also showed that interferon regulatory factor 4 could positively regulate IL‐10 production and inhibit the generation of pathogenic T cells by suppressing dendritic cell functions. Increased numbers of IL‐17‐ and interferon‐γ‐producing T cells have been found following suppression of these IL‐10‐producing regulatory B cells, demonstrating that these B cells can regulate the pathogenic T‐cell response in EAE. ⁸³ Similar naive and memory B cells capable of secreting IL‐10 have also been described in humans. ⁷⁵ , ⁸⁶ , ⁸⁷ IL‐10‐producing B cells generated from CD27⁺ memory B cells through Toll‐like receptors 4 and 9 stimulation (so called B10 cells) have been shown to inhibit production of TNF by monocytes through an IL‐10‐dependent route. ⁸⁸ Human naive CD27^– B cells have also been shown to produce IL‐10 following CD40 engagement. ⁷² , ⁷⁴ , ⁷⁸ , ⁸⁹

B‐cell‐derived transforming growth factor‐β₁ has also been shown to limit the early induction phase of EAE through inhibition of Th1/Th17 responses and dendritic cell APC activity. ⁹⁰ IL‐35‐producing B cells have been reported that can suppress pro‐inflammatory immune responses in EAE, as it was observed that mice lacking IL‐35 production by only B cells was unable to recover from the T‐cell‐mediated EAE. ⁹¹ These IL‐35‐producing B cells can regulate immune responses through induction of IL‐10‐producing B cells, suggesting that IL‐35 can be used to induce regulatory B cells to treat autoimmune diseases. ⁹² Overall, understanding how different subsets of B cells serve a role in prevention or development of MS and how B‐cell targeting therapies affect different B‐cell functions can lead to the development of more specific therapeutics with better outcome and safety in the future.

The interactions between B cells and myeloid cells in MS

Myeloid cells, including monocytes, dendritic cells and microglia, play important roles in MS pathogenesis. Myeloid cells not only function as APCs to activate T cells, but also can produce cytokines and chemokines, which can affect other cells and inflammatory processes in the CNS. Recently several studies have demonstrated an intricate interaction between B cells and myeloid cells in MS.

Li et al. have reported increased frequencies of pro‐inflammatory GM‐CSF‐producing B cells in MS. ⁷⁷ They showed that deletion of this subset of B cells resulted in a reduced pro‐inflammatory myeloid immune response in a GM‐CSF‐dependent manner, indicating that these B cells can enhance the myeloid cell inflammatory response by producing GM‐CSF implicating a pro‐inflammatory B‐cell/myeloid‐cell axis in MS. Moreover, they showed that signal transducer and activator of transcription 5 (STAT5) and STAT6 signaling were essential for B cells to produce GM‐CSF, and reciprocally regulate the generation of regulatory IL‐10‐secreting B cells. Conversely, Lehmann‐Horn et al. have shown that depletion of unactivated B cells causes enhanced production of pro‐inflammatory TNF by monocytes, resulting in EAE exacerbation. ⁹³ Moreover, they showed that anti‐CD20 treatment increased the relative frequency of monocytes and production of TNF by these cells in individuals with neuroimmunological disorders, indicating that the TNF‐mediated pro‐inflammatory activity of monocytes might be controlled by a subset of B cells. These conflicting results indicate that more investigation in this field is needed to define the function of different subsets of B cells and suggest selective targeting of pathogenic B cells, which can be more safe and effective.

On the other hand, recently Knier et al. have described an interaction between a subset of IL‐6‐ and GM‐CSF‐producing B cells and polymorphonuclear myeloid‐derived suppressor cells (PMN‐MDSCs) in the CNS, which are linked in a negative feedback loop. They reported that Ly6G⁺ myeloid cells in the CNS of EAE mice can differentiate into PMN‐MDSCs and inhibit the recruitment, local proliferation and cytokine secretion of CD138⁺ pro‐inflammatory B cells, leading to recovery from EAE. They showed that Ly6G⁺ cells that were recruited to the CNS interacted physically with GM‐CSF‐ and IL‐6‐producing B cells and acquired the properties of PMN‐MDSCs in the CNS in a gp130‐STAT3‐dependent manner. They also reported that B cells can produce IL‐6 and present it in trans through their own IL‐6Rα, inducing Ly6G⁺ cell differentiation into MDSCs, which in turn would suppress B‐cell activation and cytokine production ⁹⁴ (Fig. 1c). These results show that the interactions between myeloid cells, especially PMN‐MDSCs, and B cells may regulate B‐cell accumulation and cytokine production in the CNS of MS patients and therapeutic interventions targeting B‐cell/myeloid‐cell interactions might suppress CNS inflammation.

Targeting B cells in MS

B‐cell‐depleting therapies

The first B‐cell‐depleting clinical study testing rituximab, an anti‐CD20 chimeric monoclonal antibody, as a therapeutic approach with the initial hypothesis of decreasing MS‐related pathogenic antibodies showed good efficacy, decreasing the CNS inflammation and limiting MS relapses. As plasma cells are CD20 negative and are not eliminated by anti‐CD20 therapy, the success of this approach updated the understanding about the role of B cells in MS and encouraged scientists to further investigate the antibody‐independent functions of B cells in MS such as antigen presentation and cytokine production. ¹⁷ After the success of a phase II trial of rituximab, scientists have focused on anti‐CD20 antibodies to optimize their efficacy and safety. In order to decrease the antibody response against anti‐CD20, a humanized anti‐CD20 monoclonal antibody, ocrelizumab, was developed and tested in a phase II trial with 600‐mg and 200‐mg doses, which showed 89% and 96% reduction in gadolinium‐enhancing lesions, respectively. ¹⁹ Moreover, the efficacy and safety of a fully humanized anti‐CD20 antibody, ofatumumab, has been evaluated in a phase II study, which showed suppression of new MRI lesion development while an antibody response against human anti‐CD20 antibodies was not seen in treated patients. ¹⁸

Conversely, a clinical study testing another B‐cell‐related therapy atacicept [a fusion protein of Transmembrane activator and CAML interactor (TACI) and Fc fragment Of IgG which targets B cells and plasma cells but not memory B cells] resulted in adverse outcomes, suggesting memory B cells as a relevant disease‐promoting B‐cell subset and the existence of a functional heterogeneity among B cells and their capacity to be pathogenic or protective in MS. ⁹⁵ In addition to involvement of B cells in the pathogenesis of relapsing–remitting MS, there is an ongoing debate about the involvement of these cells in primary progressive MS based on the results of a successful clinical study testing ocrelizumab in individuals with primary progressive MS (ORATORIO). ⁹⁶ Studies have demonstrated the role of both systemic and compartmentalized inflammatory processes in MS pathogenesis. ² , ⁹⁷ , ⁹⁸ Systemic inflammation, which mainly contributes to relapsing–remitting MS, involves activation of autoreactive T cells in the periphery and subsequent infiltration to the CNS where they can be reactivated and can damage neurons that correspond with focal inflammation, magnetic resonance imaging‐detectable lesions, and relapses while compartmentalized inflammation seems to contribute in the progressive form of MS and its exact process is not well understood. ⁹⁷ , ⁹⁸ , ⁹⁹ Studies have shown that aggregates of B cells in the subarachnoid space play a key role in CNS‐compartmentalized inflammation. ⁹⁷ , ⁹⁸ , ⁹⁹ As B‐cell depletion prevents the formation of the above‐mentioned focal CNS lesions, its therapeutic efficiency is assumed to be mainly based on the abrogation of the B‐cell properties in the periphery but it does not entirely stop chronic progression. ²² Nevertheless, ORATORIO findings showed that individuals with primary progressive MS experienced a modestly less severe worsening of disability compared with placebo‐treated patients, indicating the role of B cells in chronic non‐relapsing CNS‐compartmentalized inflammation that may underlie progressive tissue injury and worsening of disability in MS independent of focal inflammation. B cells may contribute to chronic progression by producing antibodies that cause complement‐mediated injury as well as through an abnormal pro‐inflammatory response mediated by cytokines produced by these cells. Moreover, B cells may contribute to chronic progressive tissue injury through secretion of products that may be directly toxic to oligodendrocytes and neurons in progressive MS. ¹⁰⁰ , ¹⁰¹ Hence, these results may indicate the capability of anti‐CD20 therapy in managing progression independent of focal inflammation. Overall studies have focused on underlying cellular and molecular mechanisms by which B cells contribute to compartmentalized inflammation in the CNS and on clarifying the different pro‐inflammatory and anti‐inflammatory subsets of B cells to target them selectively to effectively treat MS.

Although the above‐mentioned different B‐cell‐depleting approaches are effective, the main trouble with these therapies is the elimination of all B cells, including pathogenic and non‐pathogenic B cells, which cause general immunosuppression and have some adverse effects. Hence, an ideal B‐cell‐directed therapy for autoimmune diseases such as MS should eliminate pathogenic autoimmune cells while sparing other B cells.

Ellebrecht et al. administered engineered human T cells expressing a chimeric autoantibody receptor (CAAR), which can bind to the B‐cell receptors of autoreactive B cells and eliminate them. These CAARs consist of desmoglein 3, a pemphigus vulgaris autoantigen, fused to CD137‐CD3ζ signaling domains. They showed that desmoglein 3 CAAR‐T cells were specifically cytotoxic against B cells expressing anti‐desmoglein 3 B‐cell receptors in vitro and in vivo. These findings show that CAAR‐T cells can be an effective strategy for targeting autoreactive B cells in other autoimmune diseases such as MS, which can be investigated in future studies. ¹⁰²

The effects of other therapies on B cells in MS

Studies have demonstrated that other approved MS therapies that were primarily designed to suppress effector T‐cell responses can partially or indirectly affect B‐cell function (Table 1). For instance, interferon‐β, the first therapeutic that was approved for MS, has been shown to have anti‐inflammatory effects on different immune cells, such as T cells and myeloid cells, by regulating bystander responses and preventing their trafficking into the CNS. ¹⁰³ , ¹⁰⁴ Treatment with interferon‐β has been shown to reduce the percentages of C‐C chemokine receptor type 5‐positive and CD86‐positive naive B cells, resulting in reduction of co‐stimulatory signals and antigen presentation in MS patients. ¹⁰⁵ Moreover, in MS patients treated with interferon‐β, the frequency of CD27⁺ memory B cells has been shown to reduce through induction of apoptosis. ¹⁰⁶ Conversely, increased frequency of IL‐10‐producing regulatory B cells has been reported in the peripheral blood of patients treated with interferon‐β indicating a shift from pro‐inflammatory to anti‐inflammatory phenotype. ¹⁰⁷

Table 1.

The effects of different multiple sclerosis treatments on B cells

Drug	Effects on B cells	References
Anti‐CD20	↓ pre‐B cells, transitional B cells, naive B cells and memory B cells in blood ↓ B cells in CSF	21, 73, 75, 77, 136
Alemtuzumab (anti‐CD52)	↓ B cells in blood	138
Daclizumab (anti‐CD25)	↓ B cells in blood and CSF	138, 139
Natalizumab (anti‐α 4‐integrins)	↓ B cells, plasma cells in CSF, IgM and IgG synthesis ↑ pre‐B cells, memory B cells and CXCR3+ B cells in blood	125, 129, 140, 141
Fingolimod (FTY720)	↓ naive B cells, memory B cells, TNF, HLA‐DR and CD80 in blood ↑ transitional B cells in blood, TGF‐β, IL‐10	108, 109, 142, 143, 144
Dimethyl fumarate	↓ naive B cells and memory B cells in blood ↓ GM‐CSF, TNF, IL‐6 ↑ transitional B cells	121, 122, 123, 124
IFN‐β	↓memory B cells in blood, IL‐12, CD80, CD40 ↑ transitional B cells in blood, TGF‐β, IL‐10	106, 145, 146, 147
Teriflunomide	↓ B cells in blood, B cell proliferation	148, 149, 150, 151
Glatiramer acetate	↓ naive B cells, memory B cells and pre‐B cells in blood ↓LT‐α, ICAM‐1, IL‐6, immunoglobulin production ↑ IL‐10	119, 145, 152, 153
atacicept (Anti‐BAFF/APRIL fusion protein)	↓ transitional B cells, naive B cells, plasma cells and IgM in blood ↑ IL‐15	154, 155, 156, 157
Mitoxantrone	↑ B‐cell apoptosis; ↓ LT‐α; ↑ IL‐10	74, 158, 159, 160, 161

APRIL, A proliferation‐inducing ligand; BAFF, B‐cell‐activating factor; CSF, cerebrospinal fluid; GM‐CSF, granulocyte–macrophage colony‐stimulating factor; ICAM, intracellular adhesion molecule; IL‐10, interleukin‐10; LT‐α, lymphotoxin‐α; TGF‐β, transforming growth factor‐β; TNF, tumor necrosis factor.

Treatment of MS patients with another compound, fingolimod, a modulator of sphingosine 1 phosphate receptor, has been shown to decrease total numbers of B cells in the circulation, especially memory B cells, but has little effect on CSF B cells. ¹⁰⁸ , ¹⁰⁹ , ¹¹⁰ Moreover, fingolimod markedly causes the cytokine profile to shift from a pro‐inflammatory toward an anti‐inflammatory phenotype. Fingolimod has also been reported to inhibit the formation of B‐cell aggregates in the meninges in an EAE model, but the numbers of plasma cells and infiltrating B cells was not altered in the CNS parenchyma. ¹¹¹ Moreover, the migratory ability of regulatory B cells exposed to fingolimod has been demonstrated to be enhanced in vitro. ¹⁰⁸

Glatiramer acetate (GA) is an FDA‐approved immunomodulatory drug that modulates encephalitogenic T‐cell responses by suppressing different immune cells such as pro‐inflammatory APCs, resulting in the development of Th2 cells. ¹¹² , ¹¹³ , ¹¹⁴ , ¹¹⁵ , ¹¹⁶ Studies have shown that administration of GA in an EAE model increased the development of regulatory B cells. ¹¹⁷ , ¹¹⁸ Treatment of MS patients with GA causes an increase of IL‐10 and a decrease of IL‐6 and LT‐α production by B cells. ¹¹⁹ Moreover, GA has been shown to regulate the profile of adhesion molecules in B cells, inhibiting their migration into the CNS of individuals with relapsing–remitting MS. ¹²⁰

Dimethyl fumarate (DMF), a new drug for MS treatment, is the methyl ester of fumaric acid, and its mechanisms of effect in MS are not clearly understood. Studies have shown that treatment of MS patients with DMF decreases the number of all peripheral B cells, especially memory B cells, through induction of apoptosis in these cells. ¹²¹ , ¹²² , ¹²³ , ¹²⁴ DMF has been shown to reduce production of IL‐6, GM‐CSF and TNF‐α by B cells and shift their cytokine profile towards a less pro‐inflammatory and more regulatory phenotype in vitro and in vivo. ¹²¹ , ¹²³ , ¹²⁴

Another efficacious therapeutic for MS, natalizumab, is a monoclonal antibody against the α ₄ subunit of the integrin very late antigen‐4 (VLA‐4) that is expressed on most leukocytes, especially B and T cells. Natalizumab blocks the interaction of VLA‐4 with its ligand vascular cell adhesion molecule 1 on endothelial cells and prevents leukocyte infiltration into the CNS. Natalizumab has been shown to reduce the B‐cell frequency within the CNS tissue and CSF, and conversely increase their frequency in the peripheral blood of MS patients. ¹²⁵ , ¹²⁶ , ¹²⁷ , ¹²⁸ , ¹²⁹ Intrathecal IgG production is also reduced and OCB may disappear after treatment with natalizumab. ¹²⁹ , ¹³⁰ The recurrence of disease activity after cessation of natalizumab treatment was attributed to memory B‐cell subsets, which are accumulated in the periphery during treatment. ¹²⁶ , ¹²⁷ Consistently, conditional deletion of VLA‐4 on B cells in the EAE model has been shown to prevent migration of B cells to the CNS and reduce disease severity, highlighting the role of B cells in MS pathogenesis. ¹³¹ As the migration of regulatory B cells into the CNS is also concomitantly blocked by natalizumab, ¹³² the exact effect of this agent on B cells and MS disease demands further investigation to clarify the role of different subsets of B cells in MS.

Alemtuzumab, is another FDA‐approved monoclonal antibody directed against CD52 for MS treatment. ¹³³ , ¹³⁴ Alemtuzumab treatment causes depletion of both T and B cells as well as some other leukocytes as CD52 is expressed at a lesser degree on monocytes and granulocytes. ¹³⁵ Hyperpopulation of immature B cells in the relative absence of regulatory T cells and memory B cells has been reported following CD52 depletion, which causes secondary B‐cell autoimmunity and anti‐drug antibodies. ¹³⁵ The effect of different MS therapies on B cells is described in Table 1.

Conclusion

The results of different B‐cell‐depleting therapies in MS patients demonstrated the existence of pro‐inflammatory and anti‐inflammatory subpopulations of B cells. Hence, the next strategy of B‐cell‐depleting therapy can be the application of anti‐CD20 along with an approach that can prevent the elimination of anti‐inflammatory B cells. Moreover, investigation and development of CAAR‐T cells, which can specifically target pathogenic autoreactive B cells, may provide an effective approach to avoid the adverse effects of B‐cell depletion with antibodies. Further clarifying the exact mechanisms by which different pro‐inflammatory and anti‐inflammatory subsets of B cells contribute to MS disease activity and progression, and understanding how these cells modulate T‐cell and myeloid cell functions in the CNS, and also development of biomarkers related to the function of different B‐cell subsets, will help clinicians to better optimize the different therapeutic modalities.

Disclosure

The authors declare no conflict of interest.