|Home | About | Journals | Submit | Contact Us | Français|
Understanding the pathogenesis of complex immunologic disorders such as multiple sclerosis (MS) is challenging. Abnormalities in many different cell types are observed in the immune system and CNS of patients with MS and the identification of the primary and secondary events is difficult. Recent studies suggest that the model of MS as a disorder mediated only by T cells is overly simplistic and propose an important role for B cells in the propagation of the disease. B-cell activation in the form of oligoclonal bands (OCB) production is the most consistent immunologic finding in patients with MS. Notably, markers of B-cell activation within the CSF of patients with MS predict conversion from clinically isolated syndrome to clinically definite MS and correlate with MRI activity, onset of relapses, and disability progression. In addition, the main genetic risk factor in MS is associated with OCB production, and environmental agents associated with MS susceptibility (vitamin D and the Epstein-Barr virus) influence B-cell proliferation and function. Finally, the only cell-specific treatments that are effective in patients with MS are monoclonal antibodies targeting the B-cell antigen CD20, suggesting a potentially causative role for B cells. Based on current evidence there is no longer doubt that B cells are relevant to the etiology and pathogenesis of MS. Elucidating the role of B cells in MS will be a fruitful strategy for disease prevention and treatment.Neurology® 2012;78:823–832
Multiple sclerosis (MS) is a debilitating disease of the CNS characterized by myelin loss, axonal pathology, and progressive neurologic dysfunction.1 Genetic, epidemiologic, and pathologic studies support the hypothesis that the neurologic manifestations of the disease arise from immune-mediated demyelination, which impairs neuronal transmission and results in axonal degeneration.2,3 However, understanding the factors driving this abnormal immune response is limited.
T cells and in particular CD4+ T helper cells (Th) have until recently been considered the primary immune drivers in MS. Evidence supporting this hypothesis includes that the main MS genetic risk factor resides in the major histocompatibility complex (MHC) class II region, which plays a central role in the development of T-cell central tolerance4,5 and that the animal model of MS, experimental autoimmune encephalomyelitis (EAE), can be adoptively transferred to mice by the injection of encephalitogenic myelin-specific Th cells.6
However, the notion of MS being primarily a Th-cell disease has been reassessed, and the role of CD8+ T cells, B cells, and innate immunity emphasized. We review here recent evidence that B cells are important in driving the pathologic immune response in MS.
Several features of MS plaque pathology cast doubt on the view that MS is a purely Th-cell–directed process. Lymphocytes may be absent in early MS lesions with extensive oligodendrocyte loss and demyelination.7,8 Conversely, inflammatory cuffs are often present in otherwise normal-appearing white matter contiguous with or distant from active MS lesions and unlike the lesions of EAE, CD4+ T cells are outnumbered by MHC class I–restricted, clonally expanded CD8+ cells in active MS lesions.9,10
The necessity to re-evaluate the role of different cell types in MS was also recently suggested when a phase II clinical trial testing the effect of Ustekinumab, an antibody directed against the common subunit of interleukin (IL)-12 and IL-23, relevant for Th1 and Th17 cell differentiation respectively, reported no benefit in relapsing-remitting MS (RRMS).11 Previously, an antibody directed against CD4, which successfully depleted CD4+ T cells, was also found to have no effect on disease activity or course in patients with MS.12 In clear contrast, analogous antibodies prevented the development of EAE.13,14 These trials demonstrate why, although animal models represent useful and essential tools in modern research, one needs to be aware of their distance from the human disease.15 This is especially true in complex immune disorders such as MS, in which primary causal and secondary events are difficult to separate.
The most consistent immunologic finding in patients with MS is the presence of oligoclonal bands (OCB) in the CSF. OCB arise from the intrathecal synthesis of clonal IgG and are present in more than 95% of patients with MS.16 Clonally expanded B cells are also located in brain parenchyma and are responsible for the production of OCB.17–19 Plasma cells are observed in large numbers in the perivascular spaces within subacute and chronic MS plaques, and it is likely that antigen processing and antibody synthesis takes place at these sites.8,20 A lymph node sinus-like organization of the perivascular spaces is seen within plaques of chronic patients with MS, with lymphocytes and resident macrophages located within, and plasma cells located outside, smooth-walled channels.20 Plasma cells are also present in smaller numbers in parenchymal plaque and periplaque white matter tissue, and the leptomeninges. This indicates a targeted B-cell response consequent to antigenic stimulation within the CNS.
The pathogenicity of these antibodies remains controversial. IgG deposition and complement-mediated demyelination are predominant features of brain lesions in patients with established disease.21 However, in newly forming lesions, oligodendrocyte apoptosis is a striking feature that does not appear to be triggered by immunoglobulin deposition, T cells, or macrophages.7,8 Additionally, the ability of these intrathecal antibodies to react against myelin antigens is unclear.22 For example, a recent study has shown that recombinant antibodies generated from clonally expanded plasma cell and B-cell clones can react against MS brain tissue.23 However, in a larger screen of more than 50 recombinant antibodies derived from the CSF of patients with MS, nearly all did not bind myelin basic protein (MBP), myelin oligodendrocyte glycoprotein (MOG), or proteolipid protein (PLP) (the main candidate autoantigens of MS), and hardly reacted with either control or MS brain tissues.24 This inconsistency indicates that autoantibody production within the CNS is probably not the initiating force in MS. However, there is no doubt the presence of OCB indicates abnormal B-cell activation within the CNS of patients with MS and suggests a role for B cells in the propagation of disease.
B-cell homeostasis and function within the CNS are relevant for the development and course of clinically definite MS (CDMS). In 85% of patients with MS the initial presentation is defined as a clinically isolated syndrome (CIS). Only about 60% of CIS cases will develop a second demyelinating event and therefore be diagnosed with CDMS over 20 years.25 Identifying factors driving the fate of patients with CIS is relevant not only for a better prognosis and early intervention strategies but also for the understanding of the biological mechanisms behind the development of CDMS.
B-cell counts are increased in the CSF of patients with CIS and peripheral B-cell subsets show a markedly higher expression of the α4 subunit of the VLA-4 receptor, needed for their migration across the blood–brain barrier.26 Abnormal B-cell activity is not only present at the first manifestation of the disease, but also predicts conversion to CDMS (table 1). CIS cases with OCB and polyspecific production of antibodies against measles, rubella, and varicella zoster viruses (the so-called “MRZ reaction”) and high levels of the B-cell attractant chemokine CXCL13 in the CSF are at an increased risk of a second demyelinating event.27–33 Other studies have also shown an overrepresentation of certain variable heavy chain sequences (VH4 and VH2 families) in B and plasma cells collected from CSF and plaques of patients with MS and that this feature may influence conversion to CDMS.34–36
The hypothesis that the presence of antibodies directed against myelin antigens in the serum of CIS cases could predict conversion to MS has also been investigated. While Berger et al.37 reported that risk of conversion was greater in CIS cases with serum antibodies directed against MBP and MOG, these results have not been confirmed by a larger multicenter trial,38 suggesting that anti-myelin antibody production is unlikely to mediate the role of B cells in CIS conversion.
Markers of B-cell presence and activation also correlate with disease activity and progression in CDMS (table 2). CSF CXCL13 levels are augmented in CIS, RRMS, secondary progressive MS (SPMS), and primary progressive MS (PPMS) and correlate with CSF total leukocyte and B-cell counts, intrathecal IgG synthesis, markers of demyelination and blood–brain barrier leakage, MRI activity, and relapse rate.33,39,40 Serum levels of CXCL13 are also higher during periods of MRI activity.41 Mature B-cell and plasma blast CSF counts correlate with MRI lesions.42 Furthermore, disease progression is slower in OCB-negative (OCB−) than in OCB-positive (OCB+) patients.43,44 Further evidence for a role of B cells in disease progression comes from pathologic studies showing the presence of lymphoid neogenesis in a substantial percentage of SPMS cases. Inflammatory aggregates organized as B-cell germinal center (GC)–like structures were observed in the subarachnoid space of SPMS cases and associated with features indicative of a more severe disease course.45–47 Notably, a colocalization between these B-cell aggregates and subpial cortical lesions46,47 and a gradient of neuronal loss in the cortical layers48 were found, suggesting the release of soluble factors across the inflamed meninges may cause cortical damage.
Complex disorders such as MS are conditions that have no single cause but result from a combination of genetic and environmental factors and their interactions.49 It would be logical to hypothesize that these factors would have a direct influence on the cell type triggering the abnormal immune response in MS. This does not prove causation but it is an unconditional requirement for plausibility.
The main genetic locus in MS is located within the MHC class II region and corresponds to the HLA-DRB1*1501 class II allele.4 This allele has been found to increase the risk of MS in most populations studied and its presence in homozygosity increases the risk of MS more than 6-fold.2,4
A further stratification of patients based on the presence or absence of OCB has demonstrated that the DRB1*1501 allele is strongly associated with the OCB+ subpopulation whereas the association tends to disappear in OCB− patients.50–52 It has therefore been hypothesized that OCB− patients may represent a phenotypically similar but immunologically distinct entity.50 However, OCB status can change during the course of MS,53 perhaps suggesting that OCB negativity represents particular phases of the disease, or individuals with a weaker tendency to B-cell activation rather than a distinct condition. The DRB1*1501 allele may be involved in this scenario, since MHC class II mediated presentation of antigens from B cells to CD4+ T cells is important for B-cell differentiation into GC B cells and plasma cells.54
Several non-MHC genes have been associated with MS susceptibility,55–57 some of which, such as CD40, CXCR4, and CXCR5, appear particularly relevant for B-cell homeostasis and function. The transmembrane protein CD40 is constitutively present on B cells. Its cognate receptor CD154 is expressed in a wide range of cell types (mainly activated CD4+ and CD8+ T cells) and its binding with CD40 induces B-cell proliferation, chemokine and cytokine production, GC formation, differentiation into memory B and plasma cells and B-cell–mediated T-cell activation by upregulating MHC class II and costimulatory molecules expression.58 In addition, the B-cell attractant CXCL13 acts through binding the receptor CXCR5, which is highly expressed on B cells and to a lesser extent T cells.59 Together with CXCR4, these 2 genes are directly involved in GC development and their finely regulated expression in centroblasts and centrocytes is critical for correct GC organization.60
The environment also exerts a significant influence on MS susceptibility.e1 A wealth of studies strongly support vitamin D deficiency as a key factor for MS. The prevalence of MS correlates with latitude and UV radiatione2,e3 and both vitamin D intake and low vitamin D levels are inversely associated with risk of MS.e4,e5 Possibly the strongest evidence for a role for vitamin D is the association of 2 genes involved in vitamin D biology with MS.57
As in other immune cell types, vitamin D influences development and functionality of B cells.e6 This pleiotropic hormone plays an important role in B-cell homeostasis and function by decreasing cell proliferation, inducing apoptosis, and inhibiting plasma cell differentiation.e7
Using chromatin immunoprecipitation followed by massively parallel DNA sequencing (ChIP-seq) in B-cell lines, our group has shown the presence of 2,776 different vitamin D responsive elements (VDREs) through the entire genome bound by the vitamin D receptor (VDR). Genetic loci associated with MS are strikingly enriched for VDR binding sites. In other words, vitamin D regulates the expression of genes which are associated with MS susceptibility in B cells including HLA-DRB1, CD40, CXCR4, and CXCR5.e8
The Epstein-Barr virus (EBV) may also be involved in the pathophysiology of MS. Although a great majority (more than 90%) of the general population appears to encounter EBV at some point during life, nearly all patients with MS (>99.5%) have been infected with EBV.2 The risk of MS is increased in individuals with either high anti-EBV antibody titers or a history of infectious mononucleosis (IM).2e9 These observations do not arise from a shared genetic susceptibility since the HLA-DRB1*1501 class II allele is not associated with IM.e10
This is relevant as EBV is a DNA human γ herpesvirus which primarily infects B cells and is able to immortalize them in vitro and induce lymphoproliferative disorders in vivo.e11 Different proteins encoded by EBV, in particular members of the Epstein-Barr nuclear antigen (EBNA) and latent membrane protein (LMP) families, influence the expression of a number of genes involved in cell adhesion or signaling, transcription, RNA processing, immune processes, and cell-cycle regulation.e11-e14
Interestingly, a common thread seems to link EBV and vitamin D B-cell gene regulation pathways. One way EBV influences gene expression is by an EBNA-3 mediated blockage of the VDR.e15 As an exploratory analysis, we used the expression profiles of lymphoblastoid cell lines (LCLs) obtained by infecting primary B cells with an EBV mutant strain lacking the EBNA-3 genee13 and our VDR ChIP-Seq mape8 to explore to what extent EBNA-3 may influence the expression of vitamin D–responsive genes. Almost 30% of the genes which are regulated by EBNA-3 are characterized by the presence of a VDRE, which is much greater than expected by chance (p = 0.003). In a gene ontology analysis we found that these genes with both an EBNA-3 and VDR influence are involved in cell proliferation, apoptosis, and immune response.
Other findings supporting a link between B cells, EBV, and MS come from pathologic studies reporting the presence of markers of latent EBV infection in a very high percentage of brain-infiltrating B and plasma cells in nearly all MS samples examined. The persistence of EBV was particularly enriched in meningeal B-cell follicles where viral reactivation (as defined by the presence of the early lytic protein BFRF1) was also observed.e16 By using laser microdissection and preamplifying EBV transcripts, subsequent experiments significantly increased the sensitivity of EBV detection.e17 EBV-positive B cells were also found to express the B-cell activating factor, which previous studies had shown to be upregulated by EBV proteins in B-cell lines and overexpressed in MS brain.e17-e19
However, regardless of how attractive the underlying biological rationale of the EBV presence in the CNS of patients with MS is, caution is needed since other groups have not been able to replicate these findings.e20-e25 For example, Sargsyan et al.e25 did not detect any EBV transcript in MS CSF B and plasma cells or in most of the actively demyelinating MS plaques analyzed. A single EBV-specific transcript (EBER-1) was found in a subset of MS plaques previously shown to be EBV DNA positive indicating the absence of EBV reactivation or abnormal latency programs.e25 Differences in the ascertainment of cases, sample preparation, and methods may be possible confounding factors,e20,e26 but replication is still needed to claim the presence and activation of EBV in MS brain as consistent features of the disease.
Another observation that needs replication is decreased CD8+ T-cell reactivity to autologous EBV-infected lymphoblastoid cell lines in patients with MS compared with healthy subjects.e27 This may predispose to the development of MS by allowing the accumulation of EBV-infected B cells in the MS brain. However, these data are contrary to other findings of higher EBV-specific CD8+ T-cell response in the blood of patients with CIS and normal response in patients with MS.e28
The development of monoclonal antibodies has represented one of the most significant changes in MS therapy. To date, several different molecules have been tested and the therapeutic efficacy of many of them promises to be much greater than currently available treatments.e29 Although the antigens targeted by such molecules are known, the identification of the mechanism underlying their therapeutic action is unclear since generally these antibodies target a wide range of immune cells.
For example, the monoclonal antibody natalizumab acts by blocking the α4 subunit of the VLA-4 receptor, which is needed for most leukocyte migration across the blood–brain barrier. Similarly, the impressively effective alemtuzumab causes complete depletion of all cells expressing the CD52 molecule which comprises T, B, natural killer and dendritic cells, most monocytes, and macrophages.e29
Interestingly, the only cell-specific antibody which has proved to be highly efficient in RRMS, rituximab, acts on the B-cell compartment, particularly by depleting the B cells positive for the CD20 antigen.e30 This transmembrane protein is expressed in different stages of B-cell differentiation, from pre-B cells to naive and memory B cells, but is absent in earlier stages (pro-B) and plasma cells.e30,e31 Clinical trials in RRMS have shown that rituximab causes a nearly complete depletion of CD20+ B cells and reduces the number of new Gd-enhancing lesions and the proportion of patients experiencing relapses.e32-e34 Similar results have been obtained in trials testing the second generation of anti-CD20 molecules (ocrelizumab and ofatumumab).e30
The fact that treatments specifically targeting B cells are effective in RRMS provides not only further support for the pathogenic role of B cells in MS but also the opportunity to elucidate how B cells may be acting. Notably, anti CD20 treatment does not target plasma cells and the influence on CSF IgG levels and OCB is minimal or absent.e35,e36 Therefore, it is unlikely that autoantibody production mediates the pathogenic role of B cells in MS.
Rather than producing autoantibodies, B cells may play a central role in influencing the T-cell response in patients with MS. As described above, B cells can function as professional antigen-presenting cells able to initiate T-cell–specific responses.e37,e38 Studies have suggested that a proportion of peripheral B cells in patients with RRMS can elicit CD4+ T-cell activation and proliferation in response to myelin antigens via direct antigen presentation.e39 However, B cells could also influence the T-cell response through cytokine and chemokine production (bystander mechanisms). These hypotheses find support in both animal and human studies of B-cell depletion.
Interestingly, the effects of B cell depletion on EAE models range from EAE prevention to exacerbation depending on the timing of anti-CD20 treatment and the peptide used to induce CNS inflammation.e40,e41 These findings result from B cells having either regulatory or proinflammatory phenotypes and influencing the T-cell response through cytokine production and antigen presentation.e40,e41
Similarly to EAE, in human MS rituximab treatment influences the interplay between B and T cells. Studies have shown that rituximab administration reduces CSF counts of both B cells and T cells (95% and 50% mean reductions, respectively).e35,e42 This decrease in T-cell counts correlated with a reduction of CXCL13 levels between pre and post treatment CSF paired samples.e42 Since CXCR5 (the receptor for CXCL13) is found on B cells and ~20% of CD4+ T cells in blood and CSF of patients with MS59 and B cells can promote maturation of follicular dendritic cells and CXCL13 production in positive feedback fashion,e43 the authors speculated that lack of B cells may decrease the number of T cells within the CNS by reducing chemokine secretion.e42
Another hypothesis is that B-cell depletion may impair T-cell activation and trafficking across the blood–brain barrier. In line with this, Bar Or et al. have shown that CD4+ and CD8+ T cells from patients with MS exhibited reduced proliferation and IFN-γ and IL-17 production after B cell depletion as compared to pretreatment levels. Notably, these effects could be partially reversed by adding the supernatant of activated B-cell cultures obtained from untreated patients, implicating the presence of soluble factors able to activate T cells. Further experiments identified the B-cell–produced molecules lymphotoxin and tumor necrosis factor–α as relevant in mediating this mechanism.e44
Although these observations indicate that T cells may be partly mediating the pathogenic role of B cells in MS, the secretion of soluble factors other than antibodies from B cells, located in either meninges or brain parenchyma, may directly induce inflammation and consequent demyelination. Notably, these hypotheses are not mutually exclusive and multiple mechanisms are likely to take place simultaneously.
Establishing a causal relationship is not easy and requires a number of observations. In 1965, Sir Austin Bradford Hille45 outlined a number of criteria with the intent to define “what aspects of that association should we especially consider before deciding that the most likely interpretation of it is causation.”
As Hill wanted to clarify, “none of these 9 viewpoints can bring indisputable evidence for or against the cause and effect hypothesis and none can be required as a sine qua non.” However, it is interesting to see how many and how well these criteria are met by B cells in MS today (figure and table 3). Strikingly, B cells satisfy 8 out of the 9 features of causation and in particular fulfill the main requirements that B-cell abnormalities are strongly associated with MS, are influenced by MS risk factors, precede its clinical onset, influence disease course, and can be modified by a specific intervention strategy.
Current studies unequivocally indicate that the general view of MS as a condition mediated only by T cells is overly simplistic and that B cells influence the pathogenesis of MS. Although the understanding of MS causation will require the unification of all the pieces of a complex immunologic puzzle, particularly with regard to B–T cell interactions, these observations suggest that further elucidation of the role of B cells in MS will lead to the development of novel disease treatment and prevention strategies.
Study concept and design: Drs. Disanto and Ramagopalan. Drafting of the manuscript: Dr. Disanto. Critical revision of the manuscript for important intellectual content: Drs. Morahan, Barnett, Giovannoni, and Ramagopalan.
Dr. Disanto is funded by a research fellowship FISM-Fondazione Italiana Sclerosi Multipla-Cod.: 2010/B/5. Dr. Morahan is funded by the MS Society of Australia and the UK. Dr. Barnett has served on scientific advisory boards for and received speaker honoraria from Bayer Schering Pharma, Merck Serono, Novartis, and sanofi-aventis; and received research support from Multiple Sclerosis Research Australia and the Nerve Research Foundation, The University of Sydney. Dr. Giovannoni serves on scientific advisory boards for Merck Serono and Biogen Idec and Vertex Pharmaceuticals; served on the editorial board of Multiple Sclerosis; has received speaker honoraria from Bayer Schering Pharma, Merck Serono, Biogen Idec, Pfizer Inc, Teva Pharmaceutical Industries Ltd.–sanofi-aventis, Vertex Pharmaceuticals, Genzyme Corporation, Ironwood, and Novartis; has served as a consultant for Bayer Schering Pharma, Biogen Idec, GlaxoSmithKline, Merck Serono, Protein Discovery Laboratories, Teva Pharmaceutical Industries Ltd.–sanofi-aventis, UCB, Vertex Pharmaceuticals, GW Pharma, Novartis, and FivePrime; serves on the speakers bureau for Merck Serono; and has received research support from Bayer Schering Pharma, Biogen Idec, Merck Serono, Novartis, UCB, Merz Pharmaceuticals, LLC, Teva Pharmaceutical Industries Ltd.–sanofi-aventis, GW Pharma, and Ironwood. Dr. Ramagopalan receives research support from the Multiple Sclerosis Society of Canada Scientific Research Foundation and the Multiple Sclerosis Society of the United Kingdom.