Invited Article: Inhibition of B cell functions

Emerging data from animal and human studies have revisited the importance of B cells in the pathophysiology of autoimmune neurologic disorders. The interest has peaked in the last 5 years because of rapid advances in developing new therapies against B cells functions. In neurology, perhaps more than rheumatology or other disciplines, B cells and autoantibodies are involved in the pathogenesis of a large number of neurologic diseases that affect all levels of the neuraxis, including brain and spinal cord (e.g., multiple sclerosis [MS], neuromyelitis optica, stiff-person syndrome, autoimmune encephalitis, and paraneoplastic disorders), dorsal root ganglia and peripheral nerves (e.g., acute and chronic autoimmune neuropathies), neuromuscular junction disorders (e.g., myasthenia gravis and myasthenic syndrome), and muscle (e.g., autoimmune myopathies). Understanding the B cell functions, therefore, will be essential in our efforts to identify targets of therapy applicable to a wide spectrum of autoimmune neurologic disorders.

This review discusses the biology and maturation of B cells, the trophic factors implicated in B cell survival within the CNS, the molecules facilitating their transmigration into the brain, and the therapeutic potential of present and future anti-B cell therapies in the management of autoimmune neurologic disorders.

B CELL FUNCTIONS

All humans generate autoreactive B cells with antiself reactivity which are kept in balance by complex mechanisms of negative selection and peripheral tolerance.^1,2 Loss of self-tolerance results in the production of autoantibodies to various self-antigens and to development of an autoimmune disease.^1–3 The pathogenic role of B cells has been historically connected to the presence of autoantibodies produced from the differentiation of antigen-specific mature B cells. It is arguably clear that circulating antibodies, if recognizing surface antigens on targeted tissues, can be pathogenic and initiate an acute inflammatory cascade by complement activation (figure 1A). Antibodies can also induce tissue injury by binding to Fc receptors on macrophages, neutrophils, and NK cells and attack their target via an antibody-dependent cell-mediated cytotoxic process (figure 1A). However, a number of autoantibodies commonly seen in autoimmune neurologic disorders, such as GM1, MOG, paraneoplastic or anti-GAD, are not pathogenic; they are simply markers of autoreactive B cells and signs of loss in self tolerance.^3,4 In those conditions, B cells contribute to systemic autoimmunity by at least two other mechanisms, namely antigen presentation and cytokine production. B cells are strong antigen-presenting cells, even more potent in antigen presentation than macrophages or dendritic cells, leading to T cell activation (figure 1B). Such B cell-T-cell interactions result in simultaneous expansion of antigen-specific B and T cells and perpetuate or enhance the immune response.^4,5 Further, activated B cells secrete a number of cytokines and chemokines including interleukins (IL-4, IL-6, IL-10, IL-12, IL-16, IL-23), tumor necrosis factor-α (TNFα), and interferon-γ, which alter the function of immunoregulatory T cells and affect the activation of macrophages.^5–7 Lymphotoxin β, a TNFα family member, is also expressed on the surface of activated B cells⁴ and contributes to the formation of ectopic germinal centers, as seen in the meninges of patients with MS and other neuroinflammatory disorders.^8–10 Collectively, B cells are attractive targets for immunotherapy because, in addition to their traditional antibody production, they participate in multiple levels of the immune response.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Mechanism of B cell functions

B cells contribute to the pathology of immune-mediated conditions by antibody production, complement activation, or antibody binding to macrophages in an antibody-dependent-cell mediated cytotoxicity (A); by acting as potent antigen-presenting cells resulting in clonal expansion of cytotoxic T cells (B); and by producing cytokines (IL6, IL-10, Il-12, Il-16, Il-23, interferon-γ, and TNFα), which affect activation of macrophages and various stages of immunoregulatory T cells.

B CELL-TROPHIC FACTORS AND RECEPTORS, RELEVANT IN AUTOIMMUNE NEUROLOGIC DISORDERS

B lymphocytes arise from hematopoietic stem cells in the bone marrow where they mature independently of an antigen, into pro-B cells, pre-B cells, and immature B cells (table 1). The positively selected B cells, when re-stimulated with the relevant antigen, enter the antigen-dependent phase in the peripheral lymphoid tissues where they clonally expand and give rise, sequentially, to mature B cells expressing surface IgM and IgD, activated B cells in the germinal centers, memory B cells, early and late plasmablasts, and, finally, antigen-specific, antibody-producing plasma cells.^2,5,11 Specific cluster of differentiation (CD) markers define distinct or transitional phases of differentiation (table 1). In particular, CD19 and CD20 distinguish the mature B cells from stem cells and plasma cells; CD27 is a marker for memory B cells; CD27/CD38/BAFF-R identify early plasmablasts; CD27/CD38/CD138/BAFF-R are markers for late plasmablasts; and CD138 identify plasma cells (table 1).

Two members of the TNF-a family, BAFF and APRIL, have emerged as critical factors for B cell survival and differentiation. B cell activating factor (BAFF) (synonym = Blys) promotes the survival and differentiation of B cells from a transitional to a mature B cell stage^4,11,12 and sustains germinal center formation, thereby enhancing immunoglobulin production in vivo.¹³ BAFF exerts these actions by binding to three different receptors on B cells, BAFF-R, BCMA, and TACI (figure 2). A proliferating inducing ligand (APRIL) has effects similar to BAFF, but it exerts a more prominent role on stimulating plasma cells and enhancing IgM production.^13,14 Both BAFF and APRIL are produced by monocytes, macrophages, and dendritic cells, and circulate in a trimeric form.^12–14 The mRNA of BAFF-R and APRIL is increased in the monocytes and B cells from patients with MS.^15,16 BAFF is also increased in the muscles of patients with myositis (Raju and Dalakas MC, unpublished observations), where a large number of B cells and plasma cells has now been observed.¹⁷ Most importantly, BAFF and APRIL are produced by astrocytes and they are upregulated in MS lesions promoting the survival and clonal expansion of B cells within the CNS.^15,18 Therapies targeting BAFF or APRIL, therefore, are attractive because they have the potential to halt the clonal expansion of B cells and suppress the autoimmune process in several immune-mediated neurologic disorders.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 B cell activating factor (BAFF), a proliferating inducing ligand (APRIL), and their receptors

BAFF and APRIL are fundamental in B cell differentiation and immunoglobulin production. They exert their action by binding to three receptors, BAFF-R, BCMA, and TACI (for BAFF), and BCMA and TACI (for APRIL). BCMA = B cell maturation antigen; TACI = transmembrane activator and CAML interactor.

MOLECULES INVOLVED IN THE TRANSMIGRATION OF B CELLS TO THE NERVOUS SYSTEM, AND IN SITU CLONAL EXPANSION

Human B cells constitutively express the adhesion molecules VLA-4 and LFA-1. These molecules are upregulated on activated B cells and bind to their counter-receptors VCAM-1 and ICAM-1 on the blood–brain barrier endothelial cell wall allowing for B cell transmigration (figure 3*). During a neuroinflammatory process, the chemokines MCP-1 and IL-8 are also upregulated on the endothelial cells and interact with their respective receptors CCR2 and CXCL13 on activated B cells, enhancing the transmigration of B cells within the CNS (figure 3**).^2,11 The CXCL13 chemokine, which is fundamental for B-cell recruitment into lymphoid follicles, is strongly upregulated in the meningeal B cell aggregates of EAE mice¹⁹ and in the brains of patients with MS⁹ and has emerged as a key molecule for B cell migration into the CNS.^10,11,19

• Download figure
• Open in new tab
• Download powerpoint

Figure 3 Transmigration and persistence of B cells within the CNS

After activation by an antigen, B cells proliferate, release cytokines, and transform into antibody secreting cells (ASC). Activated B cells express the adhesion molecules VLA-4 and LFA-1 which bind to counterreceptors VCAM-1 and ICAM on the endothelial cell wall and transmigrate (*). Activated B cells also express the chemokines CCR2 and CXCL13 which bind to chemokine receptors MCP-1 and Il-8 on the endothelial cell wall and transmigrate within the CNS (**). Once within the CNS, activated B cells transform to ASC, when they encounter their antigen, and produce antibodies in situ (***). The B cell activating factor and a proliferating inducing ligand, secreted by astrocytes, enhance the clonal expansion of B cells and their maturation to memory B cells, late plasmablasts, and long-lived plasma cells with further production of autoantibodies in situ (****).

When B cells encounter their antigen within the CNS, they differentiate into antibody secreting cells (ASC) producing immunoglobulins in situ²⁰ (figure 3***). Because BAFF and APRIL are highly expressed and secreted by the astrocytes of MS brains, they likely play a role in the survival of BAFF-receptor expressing B cells and their transformation into plasma cells promoting the continuous production of immunoglobulins (i.e., persisting oligoclonal bands) within the CNS (figure 3****).¹⁵

Ectopic lymphoid follicles are also formed in the meninges of patients with secondary progressive MS^8–10 through a recapitulation of all stages of B cell differentiation, similar to the one observed in secondary lymphoid organs.⁹ High number of B cells, including memory B cells and plasma cells, is also seen in the CSF of patients with MS and may be involved in maintaining the intrathecal production of immunoglobulins.²¹ Lymphotoxin β is probably a key factor driving the formation of ectopic follicular centers, as mentioned earlier.

TARGETING B CELLS USING ANTI-B CELL MOLECULES AND TROPHIC FACTORS

A series of monoclonal antibodies or fusion proteins can target B cells, either at their surface molecules or at their trophic factors and receptors. The transmigration process of B cells across the endothelial cell wall can be also targeted by natalizumab, a monoclonal antibody against integrins approved for MS. Natalizumab affects the transmigration not only of T cells but also B cells.²² The following B cell molecules are targets of new humanized monoclonal antibodies or fusion proteins, currently in clinical trials (table 2):

1. BAFF, also called Blys (B lymphocyte stimulator), and BAFF-R (BAFF-receptor). Anti-BAFF strategies affect the survival and proliferation of B cells and result in B cell depletion not only in the periphery but also in the lymph nodes and spleen. Mice deficient in BAFF or BAFF-R, as well as wild type mice with a blockade of the BAFF/BAFF-R, demonstrate marked reduction of B cells. Considering that BAFF is increased in the serum and tissues of patients with autoimmune neurologic diseases, anti-BAFF agents may provide a promising therapy. The following anti-BAFF agents are currently undergoing phase I or II clinical trials in rheumatoid arthritis and lupus:

   1. Anti-Blys (LymphoStat-B) (belimumab), a fully humanized monoclonal antibody against soluble BAFF that neutralizes Blys.

   2. Anti BAFF-R-Ig fusion protein, directed against the soluble BAFF-R IgG.

   3. AMG g23, a Blys antagonist.

2. BCMA. Recombinant BCMA IgG fusion protein binds with high affinity to APRIL and neutralizes APRIL and, to a lesser degree, BAFF. In contrast to rituximab, BCMA IgG reduces plasma cells and immunoglobulin levels because BCMA is essential for the survival of long-lived bone marrow plasma cells.²³

3. TACI. TACI IgG fusion protein (atacicept) neutralizes Blys, APRIL, and Blys/APRIL heterodimers²⁴ and is currently undergoing clinical trials in systemic lupus erythematosus and rheumatoid arthritis.

4. Lymphotoxin-β receptor (LTβR). An antibody against LTβR, the LTβR-IgG, blocks the formation of ectopic architecture in the lymphoid follicles and disrupts the formation of ectopic germinal centers. It is currently undergoing phase II trials in rheumatoid arthritis.⁴

5. CD22. This molecule transmits survival signals on B cells and is expressed on immature and mature pro-B and pre-B cells, but is weakly expressed on germinal center B cells. A monoclonal antibody directed against the CD22 molecule, called epratuzumab, causes B cell depletion and is currently in phase III trials in patients with lupus.⁴

6. CD20. The monoclonal antibody against CD20, rituximab, is the only anti-B cell agent on the market and the only one currently undergoing clinical trials in autoimmune neurologic disorders, as described below. Its humanized form, ocrelizumab, may also become available. The anti-CD20 antibodies cause severe and prolonged B cell depletion without affecting the long-lived, autoantibody-producing, plasma cells.^25,26

INHIBITING B CELLS AND AUTOANTIBODIES IN THE TREATMENT OF NEUROLOGIC DISORDERS

The neurologic diseases in which autoreactive B cells and autoantibodies play a central role include the following: 1) MS, where B cells and antibodies are involved in different stages or disease subgroups²⁷; 2) neuromyelitis optica, with autoantibodies against the aquaporin-4 water channel²⁸; 3) paraneoplastic neurologic syndromes, where autoantibodies are directed against antigens co-expressed on cancer and the nervous system; 4) stiff-person syndrome, with circulating and intrathecally produced antibodies against glutamic acid decarboxylase (GAD)²⁹; 5) myasthenia gravis and Lambert-Eaton myasthenic syndrome, where pathogenic antibodies are directed against the muscle acetylcholine receptors or the voltage-gated calcium channel³⁰; 6) neuromyotonia and limbic encephalitis, characterized by antibodies against voltage gated potassium channels; 7) autoimmune neuropathies, with antibodies against gangliosides or glycolipids; and 8) inflammatory myopathies, where high number of B cells and plasma cells are present within the endomysial inflammatory infiltrates.¹⁷

From all the anti B-cell agents described earlier, only two have been used in autoimmune neurologic disorders: hBCMA-Fc and rituximab. The hBCMA-Fc has been shown to exert an impressive effect in animals with MOG-induced EAE, not only in preventing disease development but also in ameliorating symptoms in already weak animals.³¹ In these animals, the hBCMA-Fc antibody was capable of depleting CD19+B cells from the blood, spleen, and lymph nodes, and effectively reduced the anti-MOG-specific IgG antibody titers. There was also suppression of the inflammation and demyelination within the CNS. Rituximab, originally approved for B cell lymphomas, has now been approved by the Food and Drug Administration for rheumatoid arthritis and is currently explored in neurology, as described below.

THERAPEUTIC CONSIDERATIONS AND SAFETY CONCERNS OF RITUXIMAB

Rituximab is a chimeric antibody consisting of human IgG and kappa constant regions attached to a mouse variable region from a hybridoma directed against the human CD20 molecule.⁴ CD20 is a 297 amino acid transmembrane associated phosphoprotein of 33–37 KD present on all B cells, except stem cells, pro-B cells, and plasma cells.⁴ In contrast to BAFF and APRIL, CD20 is not secreted and it is not shed or endocytosed when exposed to the antibody.

Kinetics of circulating B cells and serum immunoglobulins after rituximab.

Following an injection with anti-CD20, the rituximab-coated B cells in the periphery are rapidly depleted to very low levels. As shown in figure 4, B cells become undetectable 1 month after rituximab infusion and remain low for several months; they start re-appearing after the sixth month, but even by the 10th month they are below baseline. Rituximab depletes B cells by an antibody-dependent cellular cytotoxicity process, combined with complement-dependent cytotoxicity and induction of apoptosis.^4,25,26 Stem cells in the bone marrow that do not express CD20 are spared, thereby allowing for the generation of new B cells. The plasma cells and the germinal center B cells in the lymph nodes, spleen, CSF, and meningeal spaces are also unaffected and the intrathecally synthesized oligoclonal IgG bands, derived from the short or long-lived plasma cells in the CSF, remain unchanged.³²

• Download figure
• Open in new tab
• Download powerpoint

Figure 4 Kinetics of B cells after one series of rituximab infusions

Note the depletion of B cells from the circulation 1 month after administration in patients receiving rituximab compared to those receiving placebo. Repopulation begins after 6–8 months

Although the level of serum immunoglobulins is not expected to change after rituximab, reductions of antibody levels, especially IgM, have been inconsistently noted in several series.^{4,25,26,33,34} For example, in rheumatoid arthritis, the rheumatoid factor titers may decrease by two- to threefold.^4,25,26,34 In our patients with IgM anti-MAG demyelinating neuropathy, the IgM level and the anti-MAG titers were decreased by 30–50%.³³ Such a reduction is probably due to depletion of memory B cells, the precursors of short-lived plasma cells.^4,34 The memory CD20+ CD27+ B cells are depleted 1 month after rituximab infusions and remain undetectable for 6 months; these cells start re-appearing by the eighth month, coinciding with the slow rise of IgM level.³³ It seems that the depletion of CD27+ memory B cells may affect the replenishment of short-lived, but not the long-lived, plasma cells.^25,26,34 Because the reconstituting B cells are naïAdive B cells possessing a new and diverse immunoglobulin rearrangement pattern,^26,34 it may take some time for the newly appearing B cells to be restimulated by the original antigen. Antibody titers, therefore, may fall after rituximab at a rate controlled by the half life of Ig and plasma cells.²⁶

EFFICACY OF RITUXIMAB IN NEUROLOGY

Preliminary results with rituximab in autoimmune neurologic disorders are encouraging and suggest that further investigations in controlled trials are warranted. Case reports or prospective open-label studies have shown that rituximab can improve neurologic symptoms in the treatment of a range of diseases, such as dermatomyositis,³⁶ myasthenia gravis,³⁷ demyelinating IgM neuropathies, CIDP and multifocal motor neuropathies,^38,39 certain paraneoplastic autoimmune neurologic conditions,⁴⁰ Devic disease,⁴¹ and MS.⁴² Encouraging results from these studies have prompted ongoing controlled trials.

In a recently completed randomized trial in 104 patients with relapsing-remitting MS, a 58% relative reduction in the proportion of patients with relapse was noted after 24 weeks of therapy.⁴³ The mean number of gadolinium-enhancing lesions on MRI scans (the study’s primary endpoint) was reduced at 24 weeks compared to placebo (p < 0.0001). Rituximab was also effective in a placebo-randomized trial involving 26 patients with anti-MAG demyelinating neuropathy demonstrating reduction of the mean INCAT disability leg scores, 8 months after therapy (p < 0.05).³³ This is the first drug to demonstrate efficacy in this neuropathy. A 50% reduction in the MAG IgM titers was also noted at month 8. In Devic NMO, rituximab lowered the relapse rate significantly, compared to pretreatment data, in a total of 34 patients and stabilized or improved the EDSS score in 91% of them.^44,45 In a controlled study of patients with stiff person syndrome, encouraging results were observed in some patients but the results are still being analyzed (Dalakas MC et al., unpublished observations).

CONCLUSIONS AND PROSPECTS OF THERAPY WITH RITUXIMAB

Anti B cell therapy, as currently represented by rituximab, appears a promising tool in the immunotherapy of neurologic diseases. Apart from its very good safety profile, rituximab induces long-lasting benefits. Remissions lasting up to 12 months after therapy, as noted in patients with autoimmune neuropathy, represent a major step toward achieving temporary tolerance without targeting the specific autoreactive B cells. The notion that inhibition of B cells does not affect antibody production or increase the susceptibility to infections appears at first counterintuitive. The biology of B cells and the modes of action of rituximab, however, are teaching us that the immune system has redundant signals and plasticity to compensate for the peripherally depleted B cells. The reported excellent tolerance of combination therapy with methotrexate, as applied in rheumatoid arthritis, are early indications that rituximab may have a synergistic effect with the other immunosuppressants, without the need to stop them before initiating therapy. Vigilance and close monitoring are required to identify promptly any long-term sequelae.

ACKNOWLEDGMENT

The author thanks Dr. Raju from the Neuromuscular Diseases Section National Institute of Neurological Disorders and Stroke for performing the B cell counts during the clinical trials with rituximab.