B lymphocytes in neuromyelitis opticaB lymphocytes in neuromyelitis optica ABSTRACT Neuromyelitis optica (NMO) is an inflammatory autoimmune disorder of the CNS that predomi-nantly
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
VIEWS amp REVIEWS
Jeffrey L Bennett MDPhD
Kevin C OrsquoConnor PhDAmit Bar-Or MD FRCPScott S Zamvil MD
PhDBernhard Hemmer MDThomas F Tedder PhDH-Christian von
Buumldingen MDOlaf Stuve MD PhDMichael R Yeaman PhDTerry J Smith MDChristine Stadelmann
Neuromyelitis optica (NMO) is an inflammatory autoimmune disorder of the CNS that predomi-nantly affects the spinal cord and optic nerves A majority (approximately 75) of patients withNMO are seropositive for autoantibodies against the astrocyte water channel aquaporin-4(AQP4) These autoantibodies are predominantly IgG1 and considerable evidence supports theirpathogenicity presumably by binding to AQP4 on CNS astrocytes resulting in astrocyte injuryand inflammation Convergent clinical and laboratory-based investigations have indicated thatB cells play a fundamental role in NMO immunopathology Multiple mechanisms have beenhypothesized AQP4 autoantibody production enhanced proinflammatory B cell and plasmablastactivity aberrant B cell tolerance checkpoints diminished B cell regulatory function and loss ofB cell anergy Accordingly many current off-label therapies for NMO deplete B cells or modulatetheir activity Understanding the role and mechanisms whereby B cells contribute to initiationmaintenance and propagation of disease activity is important to advancing our understandingof NMO pathogenesis and developing effective disease-specific therapies Neurol Neuroimmunol
Neuroinflamm 20152e104 doi 101212NXI0000000000000104
GLOSSARYAIAQP4 5 AQP4 antibody index APRIL 5 a proliferation-inducing ligand AQP4 5 aquaporin-4 ASC 5 antibody-secretingcell BAFF 5 B cell-activating factor BBB 5 blood-brain barrier BCR 5 B cell receptor BM 5 bone marrow CR2 5 com-plement receptor 2 GM-CSF 5 granulocyte-macrophage colony-stimulating factor IL 5 interleukin MALT 5 mucosa-associated lymphoid tissue MHC 5 major histocompatibility complex MOG 5 myelin oligodendrocyte glycoprotein MS 5multiple sclerosis NMO 5 neuromyelitis optica OCB 5 oligoclonal band PB 5 peripheral blood QIgG 5 CSFserum IgGquotient RA 5 rheumatoid arthritis SLE 5 systemic lupus erythematosus TNF-a 5 tumor necrosis factor a
Neuromyelitis optica (NMO) is a rare demyelinating disorder of the CNS that is diagnosed by acombination of clinical imaging and laboratory criteria1 The most common manifestations arerecurrent optic neuritis and transverse myelitis however a broader range of cerebral dience-phalic and brainstem syndromes are now recognized2 Clinical and laboratory-based studiessupport a prominent role for B cells in disease pathogenesis Autoantibodies against theaquaporin-4 (AQP4) water channel (AQP4-IgG) are detected in approximately 75 of affectedindividuals (reviewed in reference 3) and additional neural and non-neural autoantibodies arefrequently observed in both seropositive (AQP4-IgG1) and seronegative (AQP4-IgG2) indi-viduals4 Both in vivo and in vitro AQP4-IgG has been shown to reproduce cardinal features ofdisease pathology56 supporting a direct role of this autoantibody in producing CNS injuryPlasmablasts are increased in the peripheral blood (PB) of patients with NMO and levels ofinterleukin (IL)-6 a cytokine that supports plasma cell differentiation and survival are elevated
From the Departments of Neurology and Ophthalmology and Neuroscience Program (JLB) University of Colorado Denver Department ofNeurology (KCO) Yale University School of Medicine New Haven CT Neuroimmunology Unit (AB-O) Montreal Neurological Instituteand Hospital McGill University Montreal Quebec Canada Department of Neurology (SSZ H-CvB) UCSF School of Medicine SanFrancisco CA Department of Neurology (BH) Technische Universitaumlt Muumlnchen Munich Cluster for Systems Neurology (SyNergy) MunichGermany Department of Immunology (TFT) Duke University Medical Center Durham NC Departments of Neurology and Neurothera-peutics (OS) University of Texas Southwestern Medical Center Dallas TX Department of Medicine (MRY) Divisions of Molecular Medicineand Infectious Diseases University of California Los Angeles Harbor-UCLA Medical Center (MRY) Torrance CA Departments of Oph-thalmology and Visual Sciences and Internal Medicine (TJS) University of Michigan Medical School Ann Arbor and Institute of Neuropa-thology (CS) University Medical Center Goumlttingen Germany
Funding information and disclosures are provided at the end of the article Go to Neurologyorgnn for full disclosure forms The Article ProcessingCharge was paid by The Guthy-Jackson Charitable Foundation
This is an open access article distributed under the terms of the Creative Commons Attribution-Noncommercial No Derivative 40 License whichpermits downloading and sharing the work provided it is properly cited The work cannot be changed in any way or used commercially
Neurologyorgnn copy 2015 American Academy of Neurology 1
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
in serum and CSF of both AQP4-IgG1 andAQP4-IgG2 patients7 In addition IL-138
and IL-59 also appear to be upregulated inNMO as compared with multiple sclerosis(MS) Together these observations areconsistent with a proinflammatory humoralresponse in NMO Moreover current empirictreatment regimens that reduce the frequencyof disease relapses directly deplete B cells (rit-uximab) or have relatively selective effects onlymphocytes (azathioprine mycophenolatemofetil and mitoxantrone) In patients withNMO disease activity can be reduced withoutsignificant reduction in AQP4-IgG titers10
suggesting that additional mechanismsbesides those associated with AQP4-IgGmay promote disease activity In this reviewwe examine potential mechanisms wherebyB cell dysfunction may contribute to NMOpathophysiology increased proinflammatoryB cell activity diminished B regulatory con-trol plasmablast expansion and autoantibodyproduction loss of B cell anergy and abnor-mal B cell tolerance Although many of thesemechanisms have yet to be directly implicatedin NMO pathology a critical assessment ofeach potential mechanism will help informdefinitive investigations Also while it isunderstood that many of these mechanisms
likely involve complex interactions with othercomponents of the adaptive immune responsethe focus of this review on B cells precludesdetailed discussion of each of thesecontributions
B CELLS PLASMA CELLS PLASMABLASTS ANDANTIBODIES B cells can perform a wide array ofnormal functions that when dysregulated may affectNMO disease activity antigen presentationproinflammatory and anti-inflammatory cytokineproduction and immunoglobulin productionWhile the role of B cells in autoimmune disordersmay change during different phases of the disease11
the apparent ability of B cell depletion to limitnew NMO disease activity implies an overallproinflammatory role for B cells in NMO possiblydue to altered numbers or abnormal activity ofproinflammatory or regulatory B cell subsets (table 1)Potential mechanisms include expansion of AQP4-specific plasmablast clones failure to eliminateautoreactive B cell subsets insufficient antigen-specificregulatory B cells andor the loss of anergicmaintenance (figure 1)
Plasma cells in bone marrow (BM) and mucosa-associated lymphoid tissue (MALT) are responsiblefor the IgG and IgA antibodies that provide long-term humoral immunity12 Circulating AQP4-IgGproduced by BM andor MALT plasma cells is pre-sumed to initiate CNS injury after gaining access tothe CNS compartment through the blood-brain bar-rier (BBB)1314 The contribution of intrathecally
Table 1 Circulating human B cell populations of potential relevance in NMO
Newly emigrant CD191CD272CD101IgM1 BM source of autoreactive B cells e-1 e-2
Mature CD191CD272CD102IgM1 PB source of autoreactive B cells e-1 e-2
Mature anergic CD191IgD1CD212CD272IgMlow Pool of silenced autoreactive B cells e-3
Transitional
T1 CD191CD24highCD38highIgMhiIgDlo CD272 B10 regulatory population e-4
T2 CD191CD24intCD38intIgMintIgD1 CD272 B10 regulatory population e-4
T3 CD191CD24intCD38intgMloIgD1 Anergic subset e-5
Memory
Unswitched CD191CD271IgD1IgM1 IgD1 autoreactive B cells Source of CD241
regulatory B10 subsete-6 e-7
Switched CD191CD271IgD2IgM2 Source of APCs or proinflammatory B cells e-8 e-9
Double-negative CD191CD272IgD2 Source of APCs or proinflammatory B cells e-10
Plasmablasts CD191CD202CD381 Production of immunoglobulin Induction ofregulatory cells Inductionof proinflammatory B cells
e-11 e-12
Abbreviations APC 5 antigen-presenting cell BM 5 bone marrow NMO 5 neuromyelitis optica PB 5 peripheral bloodThis list represents widely accepted minimal marker combinations but is not meant to be exhaustiveReferences e-1ndashe-12 are available at Neurologyorgnn
2 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
produced AQP4-IgG to the pool of pathogenic CNSantibody remains less clear Plasmablasts are elevatedin the CSF of patients with NMO during relapse515
and limited NMO cases have shown AQP4-IgGrestricted to the CSF16 In contrast measures of intra-thecal IgG synthesis such as CSF-restricted oligoclo-nal IgG bands (OCBs) elevated CSFserum IgGquotient (QIgG) or positive AQP4 antibody index(AIAQP4) are observed in only 164 8 and 43of patients with NMO respectively1314 Thesediscrepancies may arise in part from technical differ-ences clonally expanded AQP4-specific CSF plasma-blasts are detected using flow cytometry and
single-cell PCR whereas QIgG AIAQP4 and OCBsare measured using nephelometric immunofluores-cence and electrophoretic techniques These lattermethods may be less sensitive in detecting intrathecalimmunoglobulin production during active diseasewhen the BBB is dysfunctional Furthermore thelevel of intrathecal AQP4-IgG synthesis could be sig-nificantly underestimated due to tissue binding ofAQP4-IgG
Newly generated plasmablasts circulating in thePB of patients with NMO are likely precursors forantibody-producing plasma cells that reside in theBM and CNS17 Chihara et al18 have identified a
Figure 1 Potential roles of B cells in neuromyelitis optica pathogenesis
B cells may play proinflammatory and anti-inflammatory roles in neuromyelitis optica pathogenesis through various mech-anisms Autoreactive B cells may be generated by defective central tolerance (CT primary checkpoint in bone marrow) orperipheral tolerance (PT secondary checkpoint in secondary lymphoid tissue) Stimulated B cells leaving germinal centersmay differentiate into memory B cells or antibody-producing plasmablasts and plasma cells In addition to the production ofaquaporin-4 (AQP4)-IgG in the bone marrow and CNS plasma cells and plasmablasts may have additional proinflammatoryand anti-inflammatory functions Plasmablasts may secrete factors such as interleukin (IL)-17 tumor necrosis factor a
(TNF-a)nitrous oxide (NO) and granulocyte-macrophage colony-stimulating factor (GM-CSF) facilitating neutrophil andmacrophage CNS infiltration and heightening proinflammatory immune cell activity through modulation of gut microbiotaAlternatively anti-inflammatory plasma cells (Pregs) may suppress disease activity in part through the production of IL-10or IL-35 Memory B cells may further promote disease activity by antigen (Ag) presentation secretion of the proinflamma-tory cytokines lymphotoxin (LT) and TNF-a or facilitation of Th17 differentiation (IL-6 production) IL-10-producing B reg-ulatory cells may limit the immune response through antigen-specific or bystander suppression of proinflammatory T cellfunction Circulating AQP4-specific anergic B cells may provide a pool of autoreactive disease-relevant B cells that con-tribute to disease activity The pool of anergic B cells may be enhanced by deficient B cell tolerance release of anergic Bcells may be enhanced by antigen-complement adducts or decreased levels of IL-6 The location of germinal centers pro-ducing AQP4-reactive memory cells and plasmablasts remains unknown (asterisk) Blue arrows developmental pathwaysdashed green arrows and boxes stimulatory cytokines dashed red arrows and boxes inhibitory cytokines APRIL 5 aproliferation-inducing ligand BAFF 5 B cell-activating factor
Neurology Neuroimmunology amp Neuroinflammation 3
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
CD19intCD27highCD38highCD1802 B cell popula-tion that is selectively increased in NMO PB Thesecells have phenotypic features of plasmablasts secreteincreased AQP4-IgG following IL-6 stimulationin vitro and overlap with a similarly defined CSFB cell population1518 These PB plasmablasts maynot only repopulate or replenish BM plasma cellniches but also deliver antibody-secreting cells (ASCs)to the CNS compartment In addition subsets ofplasma cells may exacerbate or mitigate disease activ-ity through the secretion of proinflammatory or reg-ulatory cytokines (discussed below)19
The initial molecular characterization of thehumoral immune response in an early AQP4-IgG1NMO patient revealed a clonally expanded plasma-blast population primarily directed against AQP45 Incontrast to similar analyses in MS20 the repertoireshowed significant intraclonal variability and heavychain variable regions dominated by VH2 rather thanVH4 gene segments2021 These findings provide amolecular signature suggesting that the CNS B cellpopulation in NMO may emanate from recent ger-minal center activity that is driven by antigenic targetsthat are distinct from those in MS Further analyses ofthe B cell repertoires in AQP4-IgG1 and AQP4-IgG2 NMO could help to distinguish ASCs in thesetwo disease subsets and provide an avenue for differ-entiating NMO from other inflammatory demyelin-ating disorders A limited analysis of PB plasmablasts
and CSF B cells from the same patient has suggestedthat circulating plasmablasts may migrate betweenthese compartments during disease activity15 Giventhe results of similar studies in MS a comprehensiveanalysis of peripheral and CSF B cell repertoires usinghigh-throughput deep sequencing might clarify thepatterns of B cell migration in NMO21
The variability of AQP4-IgG titers in treated pa-tients22 and of CSF AQP4-IgG titers and plasma cellsduring disease relapse13 suggest that AQP4-specificASCs are a labile population that may be modulatedat the levels of production trafficking and mainte-nance IL-5 IL-6 tumor necrosis factor a (TNF-a)B cell-activating factor (BAFF) and a proliferation-inducing ligand (APRIL) are critical for isolatedplasma cell survival in vitro (table 2)23 ElevatedCSF levels of IL-6 BAFF APRIL and IL-5 likelyplay a role in facilitating AQP4-specific ASC survivalin the CNS of patients with NMO79 In BM eosi-nophils are the main source of APRIL and IL-6which are essential for the support of plasma cellniches24 Likewise CNS infiltration of eosinophilsmay facilitate plasma cell survival and IgG productionin NMO lesions Modulation of eosinophil numberand location or direct targeting of IL-6 signaling mayhave a direct effect on plasma cell survival autoanti-body production and NMO pathology For examplethe S1P1 receptor agonist fingolimod promotes reten-tion of eosinophils in BM and ASCs in secondary
Table 2 B cell cytokines implicated in NMO
Cytokine Levels in NMO NMO relevant source(s) Potential action References
IL-6 Increased serumIncreased CSF
Lymphocytes monocytes and macrophages endothelial cellsgranulocytes glial cells
Support plasmablast and plasma celldifferentiation and survival promote Th17differentiation
e-13ndashe-19
IL-5 Unchanged CSF T cells eosinophils mast cells astrocytes Maintenance of plasma cell niches andCNS antibody-secreting cells
Abbreviations APRIL 5 a proliferation-inducing ligand BAFF 5 B cell-activating factor GM-CSF 5 granulocyte-macrophage colony-stimulating factor IL 5
interleukin ND 5 not determined NK 5 natural killer NMO 5 neuromyelitis optica TNF-a 5 tumor necrosis factor a12 5 contradictory results References e-13ndashe-34 are available at Neurologyorgnn
4 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
lymphoid tissue2526 This effect could result inenhanced serum AQP4-IgG titers and CNS ASCmigration resulting in the exacerbation of NMO dis-ease activity reported with fingolimod treatment27 Itis interesting that natalizumab another MS therapynoted to exacerbate NMO disease activity and tissueinjury2829 increases circulating eosinophils3031 thatmay contribute directly to lesion formation32 or sup-port local ASCs In contrast inhibition of IL-6 sig-naling using the anti-IL-6 receptor monoclonalantibody tocilizumab reduces PB plasmablasts andAQP4-IgG titers in some patients and appears toreduce relapse activity in patients with NMO33
These observations are consistent with a role forIL-6 in the induction of plasma cells that produceautoantibodies against AQP4 and other relevantantigens23
While some studies have reported a correlationbetween AQP4-IgG titers and clinical relapse a con-sistent relationship with disease activity has yet toemerge2234 A threshold level of AQP4-IgG that trig-gers clinical relapse even in a given individual hasnot been identified and serum levels during relapsediffer widely within and between individuals More-over failure of rituximab has been documented insome patients with NMO experiencing relapsesdespite undetectable CD191 cells in PB and highlevels of AQP4-IgG serum titers35 In one studytreatment with anti-CD20 antibody (rituximab) ap-peared to result in a rapid and transient increase inserum AQP4-IgG titers and clinical relapse in tandemwith increased circulating BAFF levels36 however inanother study successful anti-CD20 therapy did notresult in reduced levels of AQP4-IgG BAFF orAPRIL10 Discordance between AQP4-IgG serumtiters and disease activity may also be explained bycompartmental localization at the site of tissue dam-age in the CNS In addition antibodies may differwith respect to affinity during the course of diseaseThese changes may be undetectable in current assaysIndeed in patients with NMO CSF AQP4-IgGtiters have been observed to correlate more closelywith clinical activity and neuroinflammation thanserum titers37 Alternatively antibodies targetedagainst other self-antigens may contribute to diseaseactivity For instance autoantibodies against myelinoligodendrocyte glycoprotein (MOG) have been de-tected in a small fraction of patients diagnosed withNMO who are AQP4-IgG2 There are also clinicaldifferences between the MOG-IgG1 and AQP4-IgG1 NMO patients although the findings arebased on small numbers of patients and require fur-ther study3839 Unlike antindashAQP4-IgG antindashMOG-IgG injected intracerebrally into murine brain withhuman complement does not reproduce NMOpathology40
PROINFLAMMATORY BCELLS IN NMO B cells mayfacilitate disease activity in NMO by stimulatingpathogenic T cell responses through antigen presenta-tion or cytokine secretion (table 2) B cells expressmajor histocompatibility complex (MHC) class IImolecules constitutively and serve as potentantigen-presenting cells41 B cell MHC II expressionalso contributes to the development of T follicularhelper cells that participate in B cell differentiationand immunoglobulin isotype switching4142 Thuscooperation between AQP4-specific B cells andAQP4-specific T cells may be particularly importantin ASC differentiation and the production of NMO-IgG Moreover the presence of AQP4-specific B cellsmay promote T cell responses against AQP4 thatcontribute to tissue damage B cells may produceproinflammatory cytokines through antigen-specificor polyclonal stimulation IL-6 secretion byproinflammatory memory B cells in NMO mayaggravate disease activity by promoting pathogenicTh17 differentiation Indeed during NMOexacerbations IL-17A and IL-23 are elevatedindicating enhanced Th17 pathway activity43 Areduction in IL-6 secretion following anti-CD20B cell depletion in patients with NMO may explainsome of the discordance between relapse activityreduction and persistently elevated AQP4-IgGtiters Reduced Th17 pathway activity resultingfrom reduction in the number of proinflammatoryCD201 memory B cells may significantly reduceclinical activity despite a minimal effect on AQP4-IgG production by CD202 tissue-resident plasmacells In addition alterations in the cytokine profileof B cells that reconstitute after rituximab treatmentmay explain some benefits of treatment In MS Bcells emerging after anti-CD20 depletion secretelower levels of IL-644 Also reconstituting B cellsmay have a reduced proinflammatory cytokineprofile with lower secretion of lymphotoxin orTNF-a upon activation45
Bystander activation may also result in the produc-tion of B cell cytokines that promote NMO diseaseactivity Cytokines such as TNF-a IL-17 andgranulocyte-macrophage colony-stimulating factor(GM-CSF) may be produced by various plasma cellsubsets and regulate humoral immunity alter inter-actions with commensal microbiota and modifyadaptive and innate immune responses Nitric oxideand TNF-a secreted by plasma cells in the laminapropria of the small intestine modulate IgA secretionand can alter the composition of the gut flora46 Giventhat AQP4-specific T cell responses in NMO displaycross-reactivity to Clostridium perfringens adenosinetriphosphate-binding cassette transporter permeasein vitro47 such changes in commensal microbiotamay be able to influence downstream T cell reactivity
Neurology Neuroimmunology amp Neuroinflammation 5
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
During bacterial and parasitic infections plasma cellsare a source of GM-CSF and IL-174849 Thus in thecontext of NMO increased levels of GM-CSF andIL-17 may result in increased granulocyte recruitmentto the CNS promoting disease relapse Future studiesare needed to directly determine the quantity pheno-type and activity of proinflammatory B cells andASCs in NMO and NMO spectrum diseases
REGULATORY B CELLS IN NMO Ongoing advancesin the understanding of B cell contributions to auto-immunity include the elucidation of a population ofB cells that can negatively regulate cellular immuneresponses and inflammation the most studied ofwhich are those secreting IL-10 termed B10 cells(tables 1 and 2)50 Recently IL-35 which isessential for the immunosuppressive function of Tregulatory cells51 was also noted to be necessary forthe suppressive function of B cells52 Mice lackingIL-35 in their B cells only were unable to recoverfrom experimental autoimmune encephalomyelitisand had higher macrophage and inflammatoryT cell activity IL-35 was also found to be criticalfor the induction of B regulatory cells53 It isinteresting that in both infectious and autoimmunemodels the major B cell sources of IL-10 and IL-35transcripts were plasmablasts and plasma cells52
suggesting that mature plasma cells expressing IL-35may be important for the maintenance of Bregulatory cell numbers Additional studies will beneeded to confirm plasmablast IL-10 expression theextent to which such expression may be antigen-restricted and whether this expression profile isevident during human plasma cell development
Human B cells with similar features to murineB10 cells have been identified and are expanded inseveral human autoimmune disorders including sys-temic lupus erythematosus (SLE) rheumatoid arthri-tis (RA) Sjoumlgren syndrome and bullous skindisease54 Human B10 regulatory B cells may playan important role in suppressing AQP4-specific andinnate immune responses in NMO (figure 1) Quanet al55 noted both fewer B10 regulatory cells(CD191CD24highCD38high) in relapsing AQP41NMO patients and reduced IL-10 induction follow-ing in vitro stimulation of CD191 B cells The studyhowever may have underestimated the circulatingB10 population as regulatory B10 cells do not haveclearly defined cell differentiation markers and theexpression of CD38high may be unreliable for identi-fying circulating B regulatory cells54 It is interestingthat a reduction in IL-10 expression by CD191 Bcells has been observed in patients with MS56 andthe beneficial effect of parasitic infection in patientswith MS has been associated with an increased pro-duction of B cellndashderived IL-1057 Therefore the
possibility that B cells play an important role in sup-pressing immune activation in NMO warrants fur-ther investigation
B CELL TOLERANCE AND NMO PATHOGENESISCENTRAL AND PERIPHERAL TOLERANCE ANDB CELL ANERGY Central and peripheral tolerance
During early B cell development immunoglobulinvariable region gene segments are stochastically re-combined to generate functional antibodies (B cellreceptors [BCRs]) that are expressed on the cell sur-face This process is fundamental for the generationof the wide diversity of the immunoglobulin reper-toire but also generates autoreactive B cells alongsidethose that comprise the nonself-reactive naive repertoireElimination of autoreactive B cells is controlled bytolerance mechanisms The majority of autoreactive Bcells are eliminated at 2 separate steps58 during B celldevelopment (figure 2) A central checkpoint in theBM between early immature and immature B cellstages removes a large population of developing B cellsthat express autoreactive and polyreactive antibodiesThese antibodies have been shown to display low-levelreactivity to multiple self-antigens by in vitro approachesAfter passing through the first checkpoint only a smallfraction of B cell clones with low levels of self-reactivitymigrate to the periphery59 The second B cell tolerancecheckpoint selects against these autoreactive newlyemigrant B cells before they enter the mature naiumlve Bcell compartment
The mechanisms underlying tolerance defects inautoimmune diseases have not been fully elucidatedCentral tolerance checkpoint integrity is thought tobe associated with B cell-inherent characteristics Thesemechanisms are specifically associated with signalingthrough the BCR60 Conversely the peripheral toler-ance checkpoint is thought to be dependent on T cellB cell interactions particularly those involving regula-tory T cells rather than B cell-inherent characteris-tics61 Dysregulation of central or peripheral B celltolerance may bemeasured by evaluating the frequencyof autoreactive B cells in the naive repertoire59 B celltolerance defects have been clearly demonstrated in anumber of autoimmune diseases For example patientswith RA SLE and type 1 diabetes exhibit defects inboth central and peripheral B cell tolerance check-points that result in the accumulation of self-reactivemature naiumlve B cells in their blood58 This autoreactiveB cell pool is thought to be the reservoir from whichdisease-associated autoantibodies are derived NMOpresents an opportunity to establish whether this con-cept is accurate as B cells that directly produce theautoantibodies have been isolated
Not all autoimmune diseases follow this pattern ofcombined central and peripheral tolerance defects Asubset of patients with MS harbor a naiumlve B cell
6 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
in serum and CSF of both AQP4-IgG1 andAQP4-IgG2 patients7 In addition IL-138
and IL-59 also appear to be upregulated inNMO as compared with multiple sclerosis(MS) Together these observations areconsistent with a proinflammatory humoralresponse in NMO Moreover current empirictreatment regimens that reduce the frequencyof disease relapses directly deplete B cells (rit-uximab) or have relatively selective effects onlymphocytes (azathioprine mycophenolatemofetil and mitoxantrone) In patients withNMO disease activity can be reduced withoutsignificant reduction in AQP4-IgG titers10
suggesting that additional mechanismsbesides those associated with AQP4-IgGmay promote disease activity In this reviewwe examine potential mechanisms wherebyB cell dysfunction may contribute to NMOpathophysiology increased proinflammatoryB cell activity diminished B regulatory con-trol plasmablast expansion and autoantibodyproduction loss of B cell anergy and abnor-mal B cell tolerance Although many of thesemechanisms have yet to be directly implicatedin NMO pathology a critical assessment ofeach potential mechanism will help informdefinitive investigations Also while it isunderstood that many of these mechanisms
likely involve complex interactions with othercomponents of the adaptive immune responsethe focus of this review on B cells precludesdetailed discussion of each of thesecontributions
B CELLS PLASMA CELLS PLASMABLASTS ANDANTIBODIES B cells can perform a wide array ofnormal functions that when dysregulated may affectNMO disease activity antigen presentationproinflammatory and anti-inflammatory cytokineproduction and immunoglobulin productionWhile the role of B cells in autoimmune disordersmay change during different phases of the disease11
the apparent ability of B cell depletion to limitnew NMO disease activity implies an overallproinflammatory role for B cells in NMO possiblydue to altered numbers or abnormal activity ofproinflammatory or regulatory B cell subsets (table 1)Potential mechanisms include expansion of AQP4-specific plasmablast clones failure to eliminateautoreactive B cell subsets insufficient antigen-specificregulatory B cells andor the loss of anergicmaintenance (figure 1)
Plasma cells in bone marrow (BM) and mucosa-associated lymphoid tissue (MALT) are responsiblefor the IgG and IgA antibodies that provide long-term humoral immunity12 Circulating AQP4-IgGproduced by BM andor MALT plasma cells is pre-sumed to initiate CNS injury after gaining access tothe CNS compartment through the blood-brain bar-rier (BBB)1314 The contribution of intrathecally
Table 1 Circulating human B cell populations of potential relevance in NMO
Newly emigrant CD191CD272CD101IgM1 BM source of autoreactive B cells e-1 e-2
Mature CD191CD272CD102IgM1 PB source of autoreactive B cells e-1 e-2
Mature anergic CD191IgD1CD212CD272IgMlow Pool of silenced autoreactive B cells e-3
Transitional
T1 CD191CD24highCD38highIgMhiIgDlo CD272 B10 regulatory population e-4
T2 CD191CD24intCD38intIgMintIgD1 CD272 B10 regulatory population e-4
T3 CD191CD24intCD38intgMloIgD1 Anergic subset e-5
Memory
Unswitched CD191CD271IgD1IgM1 IgD1 autoreactive B cells Source of CD241
regulatory B10 subsete-6 e-7
Switched CD191CD271IgD2IgM2 Source of APCs or proinflammatory B cells e-8 e-9
Double-negative CD191CD272IgD2 Source of APCs or proinflammatory B cells e-10
Plasmablasts CD191CD202CD381 Production of immunoglobulin Induction ofregulatory cells Inductionof proinflammatory B cells
e-11 e-12
Abbreviations APC 5 antigen-presenting cell BM 5 bone marrow NMO 5 neuromyelitis optica PB 5 peripheral bloodThis list represents widely accepted minimal marker combinations but is not meant to be exhaustiveReferences e-1ndashe-12 are available at Neurologyorgnn
2 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
produced AQP4-IgG to the pool of pathogenic CNSantibody remains less clear Plasmablasts are elevatedin the CSF of patients with NMO during relapse515
and limited NMO cases have shown AQP4-IgGrestricted to the CSF16 In contrast measures of intra-thecal IgG synthesis such as CSF-restricted oligoclo-nal IgG bands (OCBs) elevated CSFserum IgGquotient (QIgG) or positive AQP4 antibody index(AIAQP4) are observed in only 164 8 and 43of patients with NMO respectively1314 Thesediscrepancies may arise in part from technical differ-ences clonally expanded AQP4-specific CSF plasma-blasts are detected using flow cytometry and
single-cell PCR whereas QIgG AIAQP4 and OCBsare measured using nephelometric immunofluores-cence and electrophoretic techniques These lattermethods may be less sensitive in detecting intrathecalimmunoglobulin production during active diseasewhen the BBB is dysfunctional Furthermore thelevel of intrathecal AQP4-IgG synthesis could be sig-nificantly underestimated due to tissue binding ofAQP4-IgG
Newly generated plasmablasts circulating in thePB of patients with NMO are likely precursors forantibody-producing plasma cells that reside in theBM and CNS17 Chihara et al18 have identified a
Figure 1 Potential roles of B cells in neuromyelitis optica pathogenesis
B cells may play proinflammatory and anti-inflammatory roles in neuromyelitis optica pathogenesis through various mech-anisms Autoreactive B cells may be generated by defective central tolerance (CT primary checkpoint in bone marrow) orperipheral tolerance (PT secondary checkpoint in secondary lymphoid tissue) Stimulated B cells leaving germinal centersmay differentiate into memory B cells or antibody-producing plasmablasts and plasma cells In addition to the production ofaquaporin-4 (AQP4)-IgG in the bone marrow and CNS plasma cells and plasmablasts may have additional proinflammatoryand anti-inflammatory functions Plasmablasts may secrete factors such as interleukin (IL)-17 tumor necrosis factor a
(TNF-a)nitrous oxide (NO) and granulocyte-macrophage colony-stimulating factor (GM-CSF) facilitating neutrophil andmacrophage CNS infiltration and heightening proinflammatory immune cell activity through modulation of gut microbiotaAlternatively anti-inflammatory plasma cells (Pregs) may suppress disease activity in part through the production of IL-10or IL-35 Memory B cells may further promote disease activity by antigen (Ag) presentation secretion of the proinflamma-tory cytokines lymphotoxin (LT) and TNF-a or facilitation of Th17 differentiation (IL-6 production) IL-10-producing B reg-ulatory cells may limit the immune response through antigen-specific or bystander suppression of proinflammatory T cellfunction Circulating AQP4-specific anergic B cells may provide a pool of autoreactive disease-relevant B cells that con-tribute to disease activity The pool of anergic B cells may be enhanced by deficient B cell tolerance release of anergic Bcells may be enhanced by antigen-complement adducts or decreased levels of IL-6 The location of germinal centers pro-ducing AQP4-reactive memory cells and plasmablasts remains unknown (asterisk) Blue arrows developmental pathwaysdashed green arrows and boxes stimulatory cytokines dashed red arrows and boxes inhibitory cytokines APRIL 5 aproliferation-inducing ligand BAFF 5 B cell-activating factor
Neurology Neuroimmunology amp Neuroinflammation 3
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
CD19intCD27highCD38highCD1802 B cell popula-tion that is selectively increased in NMO PB Thesecells have phenotypic features of plasmablasts secreteincreased AQP4-IgG following IL-6 stimulationin vitro and overlap with a similarly defined CSFB cell population1518 These PB plasmablasts maynot only repopulate or replenish BM plasma cellniches but also deliver antibody-secreting cells (ASCs)to the CNS compartment In addition subsets ofplasma cells may exacerbate or mitigate disease activ-ity through the secretion of proinflammatory or reg-ulatory cytokines (discussed below)19
The initial molecular characterization of thehumoral immune response in an early AQP4-IgG1NMO patient revealed a clonally expanded plasma-blast population primarily directed against AQP45 Incontrast to similar analyses in MS20 the repertoireshowed significant intraclonal variability and heavychain variable regions dominated by VH2 rather thanVH4 gene segments2021 These findings provide amolecular signature suggesting that the CNS B cellpopulation in NMO may emanate from recent ger-minal center activity that is driven by antigenic targetsthat are distinct from those in MS Further analyses ofthe B cell repertoires in AQP4-IgG1 and AQP4-IgG2 NMO could help to distinguish ASCs in thesetwo disease subsets and provide an avenue for differ-entiating NMO from other inflammatory demyelin-ating disorders A limited analysis of PB plasmablasts
and CSF B cells from the same patient has suggestedthat circulating plasmablasts may migrate betweenthese compartments during disease activity15 Giventhe results of similar studies in MS a comprehensiveanalysis of peripheral and CSF B cell repertoires usinghigh-throughput deep sequencing might clarify thepatterns of B cell migration in NMO21
The variability of AQP4-IgG titers in treated pa-tients22 and of CSF AQP4-IgG titers and plasma cellsduring disease relapse13 suggest that AQP4-specificASCs are a labile population that may be modulatedat the levels of production trafficking and mainte-nance IL-5 IL-6 tumor necrosis factor a (TNF-a)B cell-activating factor (BAFF) and a proliferation-inducing ligand (APRIL) are critical for isolatedplasma cell survival in vitro (table 2)23 ElevatedCSF levels of IL-6 BAFF APRIL and IL-5 likelyplay a role in facilitating AQP4-specific ASC survivalin the CNS of patients with NMO79 In BM eosi-nophils are the main source of APRIL and IL-6which are essential for the support of plasma cellniches24 Likewise CNS infiltration of eosinophilsmay facilitate plasma cell survival and IgG productionin NMO lesions Modulation of eosinophil numberand location or direct targeting of IL-6 signaling mayhave a direct effect on plasma cell survival autoanti-body production and NMO pathology For examplethe S1P1 receptor agonist fingolimod promotes reten-tion of eosinophils in BM and ASCs in secondary
Table 2 B cell cytokines implicated in NMO
Cytokine Levels in NMO NMO relevant source(s) Potential action References
IL-6 Increased serumIncreased CSF
Lymphocytes monocytes and macrophages endothelial cellsgranulocytes glial cells
Support plasmablast and plasma celldifferentiation and survival promote Th17differentiation
e-13ndashe-19
IL-5 Unchanged CSF T cells eosinophils mast cells astrocytes Maintenance of plasma cell niches andCNS antibody-secreting cells
Abbreviations APRIL 5 a proliferation-inducing ligand BAFF 5 B cell-activating factor GM-CSF 5 granulocyte-macrophage colony-stimulating factor IL 5
interleukin ND 5 not determined NK 5 natural killer NMO 5 neuromyelitis optica TNF-a 5 tumor necrosis factor a12 5 contradictory results References e-13ndashe-34 are available at Neurologyorgnn
4 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
lymphoid tissue2526 This effect could result inenhanced serum AQP4-IgG titers and CNS ASCmigration resulting in the exacerbation of NMO dis-ease activity reported with fingolimod treatment27 Itis interesting that natalizumab another MS therapynoted to exacerbate NMO disease activity and tissueinjury2829 increases circulating eosinophils3031 thatmay contribute directly to lesion formation32 or sup-port local ASCs In contrast inhibition of IL-6 sig-naling using the anti-IL-6 receptor monoclonalantibody tocilizumab reduces PB plasmablasts andAQP4-IgG titers in some patients and appears toreduce relapse activity in patients with NMO33
These observations are consistent with a role forIL-6 in the induction of plasma cells that produceautoantibodies against AQP4 and other relevantantigens23
While some studies have reported a correlationbetween AQP4-IgG titers and clinical relapse a con-sistent relationship with disease activity has yet toemerge2234 A threshold level of AQP4-IgG that trig-gers clinical relapse even in a given individual hasnot been identified and serum levels during relapsediffer widely within and between individuals More-over failure of rituximab has been documented insome patients with NMO experiencing relapsesdespite undetectable CD191 cells in PB and highlevels of AQP4-IgG serum titers35 In one studytreatment with anti-CD20 antibody (rituximab) ap-peared to result in a rapid and transient increase inserum AQP4-IgG titers and clinical relapse in tandemwith increased circulating BAFF levels36 however inanother study successful anti-CD20 therapy did notresult in reduced levels of AQP4-IgG BAFF orAPRIL10 Discordance between AQP4-IgG serumtiters and disease activity may also be explained bycompartmental localization at the site of tissue dam-age in the CNS In addition antibodies may differwith respect to affinity during the course of diseaseThese changes may be undetectable in current assaysIndeed in patients with NMO CSF AQP4-IgGtiters have been observed to correlate more closelywith clinical activity and neuroinflammation thanserum titers37 Alternatively antibodies targetedagainst other self-antigens may contribute to diseaseactivity For instance autoantibodies against myelinoligodendrocyte glycoprotein (MOG) have been de-tected in a small fraction of patients diagnosed withNMO who are AQP4-IgG2 There are also clinicaldifferences between the MOG-IgG1 and AQP4-IgG1 NMO patients although the findings arebased on small numbers of patients and require fur-ther study3839 Unlike antindashAQP4-IgG antindashMOG-IgG injected intracerebrally into murine brain withhuman complement does not reproduce NMOpathology40
PROINFLAMMATORY BCELLS IN NMO B cells mayfacilitate disease activity in NMO by stimulatingpathogenic T cell responses through antigen presenta-tion or cytokine secretion (table 2) B cells expressmajor histocompatibility complex (MHC) class IImolecules constitutively and serve as potentantigen-presenting cells41 B cell MHC II expressionalso contributes to the development of T follicularhelper cells that participate in B cell differentiationand immunoglobulin isotype switching4142 Thuscooperation between AQP4-specific B cells andAQP4-specific T cells may be particularly importantin ASC differentiation and the production of NMO-IgG Moreover the presence of AQP4-specific B cellsmay promote T cell responses against AQP4 thatcontribute to tissue damage B cells may produceproinflammatory cytokines through antigen-specificor polyclonal stimulation IL-6 secretion byproinflammatory memory B cells in NMO mayaggravate disease activity by promoting pathogenicTh17 differentiation Indeed during NMOexacerbations IL-17A and IL-23 are elevatedindicating enhanced Th17 pathway activity43 Areduction in IL-6 secretion following anti-CD20B cell depletion in patients with NMO may explainsome of the discordance between relapse activityreduction and persistently elevated AQP4-IgGtiters Reduced Th17 pathway activity resultingfrom reduction in the number of proinflammatoryCD201 memory B cells may significantly reduceclinical activity despite a minimal effect on AQP4-IgG production by CD202 tissue-resident plasmacells In addition alterations in the cytokine profileof B cells that reconstitute after rituximab treatmentmay explain some benefits of treatment In MS Bcells emerging after anti-CD20 depletion secretelower levels of IL-644 Also reconstituting B cellsmay have a reduced proinflammatory cytokineprofile with lower secretion of lymphotoxin orTNF-a upon activation45
Bystander activation may also result in the produc-tion of B cell cytokines that promote NMO diseaseactivity Cytokines such as TNF-a IL-17 andgranulocyte-macrophage colony-stimulating factor(GM-CSF) may be produced by various plasma cellsubsets and regulate humoral immunity alter inter-actions with commensal microbiota and modifyadaptive and innate immune responses Nitric oxideand TNF-a secreted by plasma cells in the laminapropria of the small intestine modulate IgA secretionand can alter the composition of the gut flora46 Giventhat AQP4-specific T cell responses in NMO displaycross-reactivity to Clostridium perfringens adenosinetriphosphate-binding cassette transporter permeasein vitro47 such changes in commensal microbiotamay be able to influence downstream T cell reactivity
Neurology Neuroimmunology amp Neuroinflammation 5
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
During bacterial and parasitic infections plasma cellsare a source of GM-CSF and IL-174849 Thus in thecontext of NMO increased levels of GM-CSF andIL-17 may result in increased granulocyte recruitmentto the CNS promoting disease relapse Future studiesare needed to directly determine the quantity pheno-type and activity of proinflammatory B cells andASCs in NMO and NMO spectrum diseases
REGULATORY B CELLS IN NMO Ongoing advancesin the understanding of B cell contributions to auto-immunity include the elucidation of a population ofB cells that can negatively regulate cellular immuneresponses and inflammation the most studied ofwhich are those secreting IL-10 termed B10 cells(tables 1 and 2)50 Recently IL-35 which isessential for the immunosuppressive function of Tregulatory cells51 was also noted to be necessary forthe suppressive function of B cells52 Mice lackingIL-35 in their B cells only were unable to recoverfrom experimental autoimmune encephalomyelitisand had higher macrophage and inflammatoryT cell activity IL-35 was also found to be criticalfor the induction of B regulatory cells53 It isinteresting that in both infectious and autoimmunemodels the major B cell sources of IL-10 and IL-35transcripts were plasmablasts and plasma cells52
suggesting that mature plasma cells expressing IL-35may be important for the maintenance of Bregulatory cell numbers Additional studies will beneeded to confirm plasmablast IL-10 expression theextent to which such expression may be antigen-restricted and whether this expression profile isevident during human plasma cell development
Human B cells with similar features to murineB10 cells have been identified and are expanded inseveral human autoimmune disorders including sys-temic lupus erythematosus (SLE) rheumatoid arthri-tis (RA) Sjoumlgren syndrome and bullous skindisease54 Human B10 regulatory B cells may playan important role in suppressing AQP4-specific andinnate immune responses in NMO (figure 1) Quanet al55 noted both fewer B10 regulatory cells(CD191CD24highCD38high) in relapsing AQP41NMO patients and reduced IL-10 induction follow-ing in vitro stimulation of CD191 B cells The studyhowever may have underestimated the circulatingB10 population as regulatory B10 cells do not haveclearly defined cell differentiation markers and theexpression of CD38high may be unreliable for identi-fying circulating B regulatory cells54 It is interestingthat a reduction in IL-10 expression by CD191 Bcells has been observed in patients with MS56 andthe beneficial effect of parasitic infection in patientswith MS has been associated with an increased pro-duction of B cellndashderived IL-1057 Therefore the
possibility that B cells play an important role in sup-pressing immune activation in NMO warrants fur-ther investigation
B CELL TOLERANCE AND NMO PATHOGENESISCENTRAL AND PERIPHERAL TOLERANCE ANDB CELL ANERGY Central and peripheral tolerance
During early B cell development immunoglobulinvariable region gene segments are stochastically re-combined to generate functional antibodies (B cellreceptors [BCRs]) that are expressed on the cell sur-face This process is fundamental for the generationof the wide diversity of the immunoglobulin reper-toire but also generates autoreactive B cells alongsidethose that comprise the nonself-reactive naive repertoireElimination of autoreactive B cells is controlled bytolerance mechanisms The majority of autoreactive Bcells are eliminated at 2 separate steps58 during B celldevelopment (figure 2) A central checkpoint in theBM between early immature and immature B cellstages removes a large population of developing B cellsthat express autoreactive and polyreactive antibodiesThese antibodies have been shown to display low-levelreactivity to multiple self-antigens by in vitro approachesAfter passing through the first checkpoint only a smallfraction of B cell clones with low levels of self-reactivitymigrate to the periphery59 The second B cell tolerancecheckpoint selects against these autoreactive newlyemigrant B cells before they enter the mature naiumlve Bcell compartment
The mechanisms underlying tolerance defects inautoimmune diseases have not been fully elucidatedCentral tolerance checkpoint integrity is thought tobe associated with B cell-inherent characteristics Thesemechanisms are specifically associated with signalingthrough the BCR60 Conversely the peripheral toler-ance checkpoint is thought to be dependent on T cellB cell interactions particularly those involving regula-tory T cells rather than B cell-inherent characteris-tics61 Dysregulation of central or peripheral B celltolerance may bemeasured by evaluating the frequencyof autoreactive B cells in the naive repertoire59 B celltolerance defects have been clearly demonstrated in anumber of autoimmune diseases For example patientswith RA SLE and type 1 diabetes exhibit defects inboth central and peripheral B cell tolerance check-points that result in the accumulation of self-reactivemature naiumlve B cells in their blood58 This autoreactiveB cell pool is thought to be the reservoir from whichdisease-associated autoantibodies are derived NMOpresents an opportunity to establish whether this con-cept is accurate as B cells that directly produce theautoantibodies have been isolated
Not all autoimmune diseases follow this pattern ofcombined central and peripheral tolerance defects Asubset of patients with MS harbor a naiumlve B cell
6 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
produced AQP4-IgG to the pool of pathogenic CNSantibody remains less clear Plasmablasts are elevatedin the CSF of patients with NMO during relapse515
and limited NMO cases have shown AQP4-IgGrestricted to the CSF16 In contrast measures of intra-thecal IgG synthesis such as CSF-restricted oligoclo-nal IgG bands (OCBs) elevated CSFserum IgGquotient (QIgG) or positive AQP4 antibody index(AIAQP4) are observed in only 164 8 and 43of patients with NMO respectively1314 Thesediscrepancies may arise in part from technical differ-ences clonally expanded AQP4-specific CSF plasma-blasts are detected using flow cytometry and
single-cell PCR whereas QIgG AIAQP4 and OCBsare measured using nephelometric immunofluores-cence and electrophoretic techniques These lattermethods may be less sensitive in detecting intrathecalimmunoglobulin production during active diseasewhen the BBB is dysfunctional Furthermore thelevel of intrathecal AQP4-IgG synthesis could be sig-nificantly underestimated due to tissue binding ofAQP4-IgG
Newly generated plasmablasts circulating in thePB of patients with NMO are likely precursors forantibody-producing plasma cells that reside in theBM and CNS17 Chihara et al18 have identified a
Figure 1 Potential roles of B cells in neuromyelitis optica pathogenesis
B cells may play proinflammatory and anti-inflammatory roles in neuromyelitis optica pathogenesis through various mech-anisms Autoreactive B cells may be generated by defective central tolerance (CT primary checkpoint in bone marrow) orperipheral tolerance (PT secondary checkpoint in secondary lymphoid tissue) Stimulated B cells leaving germinal centersmay differentiate into memory B cells or antibody-producing plasmablasts and plasma cells In addition to the production ofaquaporin-4 (AQP4)-IgG in the bone marrow and CNS plasma cells and plasmablasts may have additional proinflammatoryand anti-inflammatory functions Plasmablasts may secrete factors such as interleukin (IL)-17 tumor necrosis factor a
(TNF-a)nitrous oxide (NO) and granulocyte-macrophage colony-stimulating factor (GM-CSF) facilitating neutrophil andmacrophage CNS infiltration and heightening proinflammatory immune cell activity through modulation of gut microbiotaAlternatively anti-inflammatory plasma cells (Pregs) may suppress disease activity in part through the production of IL-10or IL-35 Memory B cells may further promote disease activity by antigen (Ag) presentation secretion of the proinflamma-tory cytokines lymphotoxin (LT) and TNF-a or facilitation of Th17 differentiation (IL-6 production) IL-10-producing B reg-ulatory cells may limit the immune response through antigen-specific or bystander suppression of proinflammatory T cellfunction Circulating AQP4-specific anergic B cells may provide a pool of autoreactive disease-relevant B cells that con-tribute to disease activity The pool of anergic B cells may be enhanced by deficient B cell tolerance release of anergic Bcells may be enhanced by antigen-complement adducts or decreased levels of IL-6 The location of germinal centers pro-ducing AQP4-reactive memory cells and plasmablasts remains unknown (asterisk) Blue arrows developmental pathwaysdashed green arrows and boxes stimulatory cytokines dashed red arrows and boxes inhibitory cytokines APRIL 5 aproliferation-inducing ligand BAFF 5 B cell-activating factor
Neurology Neuroimmunology amp Neuroinflammation 3
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
CD19intCD27highCD38highCD1802 B cell popula-tion that is selectively increased in NMO PB Thesecells have phenotypic features of plasmablasts secreteincreased AQP4-IgG following IL-6 stimulationin vitro and overlap with a similarly defined CSFB cell population1518 These PB plasmablasts maynot only repopulate or replenish BM plasma cellniches but also deliver antibody-secreting cells (ASCs)to the CNS compartment In addition subsets ofplasma cells may exacerbate or mitigate disease activ-ity through the secretion of proinflammatory or reg-ulatory cytokines (discussed below)19
The initial molecular characterization of thehumoral immune response in an early AQP4-IgG1NMO patient revealed a clonally expanded plasma-blast population primarily directed against AQP45 Incontrast to similar analyses in MS20 the repertoireshowed significant intraclonal variability and heavychain variable regions dominated by VH2 rather thanVH4 gene segments2021 These findings provide amolecular signature suggesting that the CNS B cellpopulation in NMO may emanate from recent ger-minal center activity that is driven by antigenic targetsthat are distinct from those in MS Further analyses ofthe B cell repertoires in AQP4-IgG1 and AQP4-IgG2 NMO could help to distinguish ASCs in thesetwo disease subsets and provide an avenue for differ-entiating NMO from other inflammatory demyelin-ating disorders A limited analysis of PB plasmablasts
and CSF B cells from the same patient has suggestedthat circulating plasmablasts may migrate betweenthese compartments during disease activity15 Giventhe results of similar studies in MS a comprehensiveanalysis of peripheral and CSF B cell repertoires usinghigh-throughput deep sequencing might clarify thepatterns of B cell migration in NMO21
The variability of AQP4-IgG titers in treated pa-tients22 and of CSF AQP4-IgG titers and plasma cellsduring disease relapse13 suggest that AQP4-specificASCs are a labile population that may be modulatedat the levels of production trafficking and mainte-nance IL-5 IL-6 tumor necrosis factor a (TNF-a)B cell-activating factor (BAFF) and a proliferation-inducing ligand (APRIL) are critical for isolatedplasma cell survival in vitro (table 2)23 ElevatedCSF levels of IL-6 BAFF APRIL and IL-5 likelyplay a role in facilitating AQP4-specific ASC survivalin the CNS of patients with NMO79 In BM eosi-nophils are the main source of APRIL and IL-6which are essential for the support of plasma cellniches24 Likewise CNS infiltration of eosinophilsmay facilitate plasma cell survival and IgG productionin NMO lesions Modulation of eosinophil numberand location or direct targeting of IL-6 signaling mayhave a direct effect on plasma cell survival autoanti-body production and NMO pathology For examplethe S1P1 receptor agonist fingolimod promotes reten-tion of eosinophils in BM and ASCs in secondary
Table 2 B cell cytokines implicated in NMO
Cytokine Levels in NMO NMO relevant source(s) Potential action References
IL-6 Increased serumIncreased CSF
Lymphocytes monocytes and macrophages endothelial cellsgranulocytes glial cells
Support plasmablast and plasma celldifferentiation and survival promote Th17differentiation
e-13ndashe-19
IL-5 Unchanged CSF T cells eosinophils mast cells astrocytes Maintenance of plasma cell niches andCNS antibody-secreting cells
Abbreviations APRIL 5 a proliferation-inducing ligand BAFF 5 B cell-activating factor GM-CSF 5 granulocyte-macrophage colony-stimulating factor IL 5
interleukin ND 5 not determined NK 5 natural killer NMO 5 neuromyelitis optica TNF-a 5 tumor necrosis factor a12 5 contradictory results References e-13ndashe-34 are available at Neurologyorgnn
4 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
lymphoid tissue2526 This effect could result inenhanced serum AQP4-IgG titers and CNS ASCmigration resulting in the exacerbation of NMO dis-ease activity reported with fingolimod treatment27 Itis interesting that natalizumab another MS therapynoted to exacerbate NMO disease activity and tissueinjury2829 increases circulating eosinophils3031 thatmay contribute directly to lesion formation32 or sup-port local ASCs In contrast inhibition of IL-6 sig-naling using the anti-IL-6 receptor monoclonalantibody tocilizumab reduces PB plasmablasts andAQP4-IgG titers in some patients and appears toreduce relapse activity in patients with NMO33
These observations are consistent with a role forIL-6 in the induction of plasma cells that produceautoantibodies against AQP4 and other relevantantigens23
While some studies have reported a correlationbetween AQP4-IgG titers and clinical relapse a con-sistent relationship with disease activity has yet toemerge2234 A threshold level of AQP4-IgG that trig-gers clinical relapse even in a given individual hasnot been identified and serum levels during relapsediffer widely within and between individuals More-over failure of rituximab has been documented insome patients with NMO experiencing relapsesdespite undetectable CD191 cells in PB and highlevels of AQP4-IgG serum titers35 In one studytreatment with anti-CD20 antibody (rituximab) ap-peared to result in a rapid and transient increase inserum AQP4-IgG titers and clinical relapse in tandemwith increased circulating BAFF levels36 however inanother study successful anti-CD20 therapy did notresult in reduced levels of AQP4-IgG BAFF orAPRIL10 Discordance between AQP4-IgG serumtiters and disease activity may also be explained bycompartmental localization at the site of tissue dam-age in the CNS In addition antibodies may differwith respect to affinity during the course of diseaseThese changes may be undetectable in current assaysIndeed in patients with NMO CSF AQP4-IgGtiters have been observed to correlate more closelywith clinical activity and neuroinflammation thanserum titers37 Alternatively antibodies targetedagainst other self-antigens may contribute to diseaseactivity For instance autoantibodies against myelinoligodendrocyte glycoprotein (MOG) have been de-tected in a small fraction of patients diagnosed withNMO who are AQP4-IgG2 There are also clinicaldifferences between the MOG-IgG1 and AQP4-IgG1 NMO patients although the findings arebased on small numbers of patients and require fur-ther study3839 Unlike antindashAQP4-IgG antindashMOG-IgG injected intracerebrally into murine brain withhuman complement does not reproduce NMOpathology40
PROINFLAMMATORY BCELLS IN NMO B cells mayfacilitate disease activity in NMO by stimulatingpathogenic T cell responses through antigen presenta-tion or cytokine secretion (table 2) B cells expressmajor histocompatibility complex (MHC) class IImolecules constitutively and serve as potentantigen-presenting cells41 B cell MHC II expressionalso contributes to the development of T follicularhelper cells that participate in B cell differentiationand immunoglobulin isotype switching4142 Thuscooperation between AQP4-specific B cells andAQP4-specific T cells may be particularly importantin ASC differentiation and the production of NMO-IgG Moreover the presence of AQP4-specific B cellsmay promote T cell responses against AQP4 thatcontribute to tissue damage B cells may produceproinflammatory cytokines through antigen-specificor polyclonal stimulation IL-6 secretion byproinflammatory memory B cells in NMO mayaggravate disease activity by promoting pathogenicTh17 differentiation Indeed during NMOexacerbations IL-17A and IL-23 are elevatedindicating enhanced Th17 pathway activity43 Areduction in IL-6 secretion following anti-CD20B cell depletion in patients with NMO may explainsome of the discordance between relapse activityreduction and persistently elevated AQP4-IgGtiters Reduced Th17 pathway activity resultingfrom reduction in the number of proinflammatoryCD201 memory B cells may significantly reduceclinical activity despite a minimal effect on AQP4-IgG production by CD202 tissue-resident plasmacells In addition alterations in the cytokine profileof B cells that reconstitute after rituximab treatmentmay explain some benefits of treatment In MS Bcells emerging after anti-CD20 depletion secretelower levels of IL-644 Also reconstituting B cellsmay have a reduced proinflammatory cytokineprofile with lower secretion of lymphotoxin orTNF-a upon activation45
Bystander activation may also result in the produc-tion of B cell cytokines that promote NMO diseaseactivity Cytokines such as TNF-a IL-17 andgranulocyte-macrophage colony-stimulating factor(GM-CSF) may be produced by various plasma cellsubsets and regulate humoral immunity alter inter-actions with commensal microbiota and modifyadaptive and innate immune responses Nitric oxideand TNF-a secreted by plasma cells in the laminapropria of the small intestine modulate IgA secretionand can alter the composition of the gut flora46 Giventhat AQP4-specific T cell responses in NMO displaycross-reactivity to Clostridium perfringens adenosinetriphosphate-binding cassette transporter permeasein vitro47 such changes in commensal microbiotamay be able to influence downstream T cell reactivity
Neurology Neuroimmunology amp Neuroinflammation 5
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
During bacterial and parasitic infections plasma cellsare a source of GM-CSF and IL-174849 Thus in thecontext of NMO increased levels of GM-CSF andIL-17 may result in increased granulocyte recruitmentto the CNS promoting disease relapse Future studiesare needed to directly determine the quantity pheno-type and activity of proinflammatory B cells andASCs in NMO and NMO spectrum diseases
REGULATORY B CELLS IN NMO Ongoing advancesin the understanding of B cell contributions to auto-immunity include the elucidation of a population ofB cells that can negatively regulate cellular immuneresponses and inflammation the most studied ofwhich are those secreting IL-10 termed B10 cells(tables 1 and 2)50 Recently IL-35 which isessential for the immunosuppressive function of Tregulatory cells51 was also noted to be necessary forthe suppressive function of B cells52 Mice lackingIL-35 in their B cells only were unable to recoverfrom experimental autoimmune encephalomyelitisand had higher macrophage and inflammatoryT cell activity IL-35 was also found to be criticalfor the induction of B regulatory cells53 It isinteresting that in both infectious and autoimmunemodels the major B cell sources of IL-10 and IL-35transcripts were plasmablasts and plasma cells52
suggesting that mature plasma cells expressing IL-35may be important for the maintenance of Bregulatory cell numbers Additional studies will beneeded to confirm plasmablast IL-10 expression theextent to which such expression may be antigen-restricted and whether this expression profile isevident during human plasma cell development
Human B cells with similar features to murineB10 cells have been identified and are expanded inseveral human autoimmune disorders including sys-temic lupus erythematosus (SLE) rheumatoid arthri-tis (RA) Sjoumlgren syndrome and bullous skindisease54 Human B10 regulatory B cells may playan important role in suppressing AQP4-specific andinnate immune responses in NMO (figure 1) Quanet al55 noted both fewer B10 regulatory cells(CD191CD24highCD38high) in relapsing AQP41NMO patients and reduced IL-10 induction follow-ing in vitro stimulation of CD191 B cells The studyhowever may have underestimated the circulatingB10 population as regulatory B10 cells do not haveclearly defined cell differentiation markers and theexpression of CD38high may be unreliable for identi-fying circulating B regulatory cells54 It is interestingthat a reduction in IL-10 expression by CD191 Bcells has been observed in patients with MS56 andthe beneficial effect of parasitic infection in patientswith MS has been associated with an increased pro-duction of B cellndashderived IL-1057 Therefore the
possibility that B cells play an important role in sup-pressing immune activation in NMO warrants fur-ther investigation
B CELL TOLERANCE AND NMO PATHOGENESISCENTRAL AND PERIPHERAL TOLERANCE ANDB CELL ANERGY Central and peripheral tolerance
During early B cell development immunoglobulinvariable region gene segments are stochastically re-combined to generate functional antibodies (B cellreceptors [BCRs]) that are expressed on the cell sur-face This process is fundamental for the generationof the wide diversity of the immunoglobulin reper-toire but also generates autoreactive B cells alongsidethose that comprise the nonself-reactive naive repertoireElimination of autoreactive B cells is controlled bytolerance mechanisms The majority of autoreactive Bcells are eliminated at 2 separate steps58 during B celldevelopment (figure 2) A central checkpoint in theBM between early immature and immature B cellstages removes a large population of developing B cellsthat express autoreactive and polyreactive antibodiesThese antibodies have been shown to display low-levelreactivity to multiple self-antigens by in vitro approachesAfter passing through the first checkpoint only a smallfraction of B cell clones with low levels of self-reactivitymigrate to the periphery59 The second B cell tolerancecheckpoint selects against these autoreactive newlyemigrant B cells before they enter the mature naiumlve Bcell compartment
The mechanisms underlying tolerance defects inautoimmune diseases have not been fully elucidatedCentral tolerance checkpoint integrity is thought tobe associated with B cell-inherent characteristics Thesemechanisms are specifically associated with signalingthrough the BCR60 Conversely the peripheral toler-ance checkpoint is thought to be dependent on T cellB cell interactions particularly those involving regula-tory T cells rather than B cell-inherent characteris-tics61 Dysregulation of central or peripheral B celltolerance may bemeasured by evaluating the frequencyof autoreactive B cells in the naive repertoire59 B celltolerance defects have been clearly demonstrated in anumber of autoimmune diseases For example patientswith RA SLE and type 1 diabetes exhibit defects inboth central and peripheral B cell tolerance check-points that result in the accumulation of self-reactivemature naiumlve B cells in their blood58 This autoreactiveB cell pool is thought to be the reservoir from whichdisease-associated autoantibodies are derived NMOpresents an opportunity to establish whether this con-cept is accurate as B cells that directly produce theautoantibodies have been isolated
Not all autoimmune diseases follow this pattern ofcombined central and peripheral tolerance defects Asubset of patients with MS harbor a naiumlve B cell
6 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
CD19intCD27highCD38highCD1802 B cell popula-tion that is selectively increased in NMO PB Thesecells have phenotypic features of plasmablasts secreteincreased AQP4-IgG following IL-6 stimulationin vitro and overlap with a similarly defined CSFB cell population1518 These PB plasmablasts maynot only repopulate or replenish BM plasma cellniches but also deliver antibody-secreting cells (ASCs)to the CNS compartment In addition subsets ofplasma cells may exacerbate or mitigate disease activ-ity through the secretion of proinflammatory or reg-ulatory cytokines (discussed below)19
The initial molecular characterization of thehumoral immune response in an early AQP4-IgG1NMO patient revealed a clonally expanded plasma-blast population primarily directed against AQP45 Incontrast to similar analyses in MS20 the repertoireshowed significant intraclonal variability and heavychain variable regions dominated by VH2 rather thanVH4 gene segments2021 These findings provide amolecular signature suggesting that the CNS B cellpopulation in NMO may emanate from recent ger-minal center activity that is driven by antigenic targetsthat are distinct from those in MS Further analyses ofthe B cell repertoires in AQP4-IgG1 and AQP4-IgG2 NMO could help to distinguish ASCs in thesetwo disease subsets and provide an avenue for differ-entiating NMO from other inflammatory demyelin-ating disorders A limited analysis of PB plasmablasts
and CSF B cells from the same patient has suggestedthat circulating plasmablasts may migrate betweenthese compartments during disease activity15 Giventhe results of similar studies in MS a comprehensiveanalysis of peripheral and CSF B cell repertoires usinghigh-throughput deep sequencing might clarify thepatterns of B cell migration in NMO21
The variability of AQP4-IgG titers in treated pa-tients22 and of CSF AQP4-IgG titers and plasma cellsduring disease relapse13 suggest that AQP4-specificASCs are a labile population that may be modulatedat the levels of production trafficking and mainte-nance IL-5 IL-6 tumor necrosis factor a (TNF-a)B cell-activating factor (BAFF) and a proliferation-inducing ligand (APRIL) are critical for isolatedplasma cell survival in vitro (table 2)23 ElevatedCSF levels of IL-6 BAFF APRIL and IL-5 likelyplay a role in facilitating AQP4-specific ASC survivalin the CNS of patients with NMO79 In BM eosi-nophils are the main source of APRIL and IL-6which are essential for the support of plasma cellniches24 Likewise CNS infiltration of eosinophilsmay facilitate plasma cell survival and IgG productionin NMO lesions Modulation of eosinophil numberand location or direct targeting of IL-6 signaling mayhave a direct effect on plasma cell survival autoanti-body production and NMO pathology For examplethe S1P1 receptor agonist fingolimod promotes reten-tion of eosinophils in BM and ASCs in secondary
Table 2 B cell cytokines implicated in NMO
Cytokine Levels in NMO NMO relevant source(s) Potential action References
IL-6 Increased serumIncreased CSF
Lymphocytes monocytes and macrophages endothelial cellsgranulocytes glial cells
Support plasmablast and plasma celldifferentiation and survival promote Th17differentiation
e-13ndashe-19
IL-5 Unchanged CSF T cells eosinophils mast cells astrocytes Maintenance of plasma cell niches andCNS antibody-secreting cells
Abbreviations APRIL 5 a proliferation-inducing ligand BAFF 5 B cell-activating factor GM-CSF 5 granulocyte-macrophage colony-stimulating factor IL 5
interleukin ND 5 not determined NK 5 natural killer NMO 5 neuromyelitis optica TNF-a 5 tumor necrosis factor a12 5 contradictory results References e-13ndashe-34 are available at Neurologyorgnn
4 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
lymphoid tissue2526 This effect could result inenhanced serum AQP4-IgG titers and CNS ASCmigration resulting in the exacerbation of NMO dis-ease activity reported with fingolimod treatment27 Itis interesting that natalizumab another MS therapynoted to exacerbate NMO disease activity and tissueinjury2829 increases circulating eosinophils3031 thatmay contribute directly to lesion formation32 or sup-port local ASCs In contrast inhibition of IL-6 sig-naling using the anti-IL-6 receptor monoclonalantibody tocilizumab reduces PB plasmablasts andAQP4-IgG titers in some patients and appears toreduce relapse activity in patients with NMO33
These observations are consistent with a role forIL-6 in the induction of plasma cells that produceautoantibodies against AQP4 and other relevantantigens23
While some studies have reported a correlationbetween AQP4-IgG titers and clinical relapse a con-sistent relationship with disease activity has yet toemerge2234 A threshold level of AQP4-IgG that trig-gers clinical relapse even in a given individual hasnot been identified and serum levels during relapsediffer widely within and between individuals More-over failure of rituximab has been documented insome patients with NMO experiencing relapsesdespite undetectable CD191 cells in PB and highlevels of AQP4-IgG serum titers35 In one studytreatment with anti-CD20 antibody (rituximab) ap-peared to result in a rapid and transient increase inserum AQP4-IgG titers and clinical relapse in tandemwith increased circulating BAFF levels36 however inanother study successful anti-CD20 therapy did notresult in reduced levels of AQP4-IgG BAFF orAPRIL10 Discordance between AQP4-IgG serumtiters and disease activity may also be explained bycompartmental localization at the site of tissue dam-age in the CNS In addition antibodies may differwith respect to affinity during the course of diseaseThese changes may be undetectable in current assaysIndeed in patients with NMO CSF AQP4-IgGtiters have been observed to correlate more closelywith clinical activity and neuroinflammation thanserum titers37 Alternatively antibodies targetedagainst other self-antigens may contribute to diseaseactivity For instance autoantibodies against myelinoligodendrocyte glycoprotein (MOG) have been de-tected in a small fraction of patients diagnosed withNMO who are AQP4-IgG2 There are also clinicaldifferences between the MOG-IgG1 and AQP4-IgG1 NMO patients although the findings arebased on small numbers of patients and require fur-ther study3839 Unlike antindashAQP4-IgG antindashMOG-IgG injected intracerebrally into murine brain withhuman complement does not reproduce NMOpathology40
PROINFLAMMATORY BCELLS IN NMO B cells mayfacilitate disease activity in NMO by stimulatingpathogenic T cell responses through antigen presenta-tion or cytokine secretion (table 2) B cells expressmajor histocompatibility complex (MHC) class IImolecules constitutively and serve as potentantigen-presenting cells41 B cell MHC II expressionalso contributes to the development of T follicularhelper cells that participate in B cell differentiationand immunoglobulin isotype switching4142 Thuscooperation between AQP4-specific B cells andAQP4-specific T cells may be particularly importantin ASC differentiation and the production of NMO-IgG Moreover the presence of AQP4-specific B cellsmay promote T cell responses against AQP4 thatcontribute to tissue damage B cells may produceproinflammatory cytokines through antigen-specificor polyclonal stimulation IL-6 secretion byproinflammatory memory B cells in NMO mayaggravate disease activity by promoting pathogenicTh17 differentiation Indeed during NMOexacerbations IL-17A and IL-23 are elevatedindicating enhanced Th17 pathway activity43 Areduction in IL-6 secretion following anti-CD20B cell depletion in patients with NMO may explainsome of the discordance between relapse activityreduction and persistently elevated AQP4-IgGtiters Reduced Th17 pathway activity resultingfrom reduction in the number of proinflammatoryCD201 memory B cells may significantly reduceclinical activity despite a minimal effect on AQP4-IgG production by CD202 tissue-resident plasmacells In addition alterations in the cytokine profileof B cells that reconstitute after rituximab treatmentmay explain some benefits of treatment In MS Bcells emerging after anti-CD20 depletion secretelower levels of IL-644 Also reconstituting B cellsmay have a reduced proinflammatory cytokineprofile with lower secretion of lymphotoxin orTNF-a upon activation45
Bystander activation may also result in the produc-tion of B cell cytokines that promote NMO diseaseactivity Cytokines such as TNF-a IL-17 andgranulocyte-macrophage colony-stimulating factor(GM-CSF) may be produced by various plasma cellsubsets and regulate humoral immunity alter inter-actions with commensal microbiota and modifyadaptive and innate immune responses Nitric oxideand TNF-a secreted by plasma cells in the laminapropria of the small intestine modulate IgA secretionand can alter the composition of the gut flora46 Giventhat AQP4-specific T cell responses in NMO displaycross-reactivity to Clostridium perfringens adenosinetriphosphate-binding cassette transporter permeasein vitro47 such changes in commensal microbiotamay be able to influence downstream T cell reactivity
Neurology Neuroimmunology amp Neuroinflammation 5
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
During bacterial and parasitic infections plasma cellsare a source of GM-CSF and IL-174849 Thus in thecontext of NMO increased levels of GM-CSF andIL-17 may result in increased granulocyte recruitmentto the CNS promoting disease relapse Future studiesare needed to directly determine the quantity pheno-type and activity of proinflammatory B cells andASCs in NMO and NMO spectrum diseases
REGULATORY B CELLS IN NMO Ongoing advancesin the understanding of B cell contributions to auto-immunity include the elucidation of a population ofB cells that can negatively regulate cellular immuneresponses and inflammation the most studied ofwhich are those secreting IL-10 termed B10 cells(tables 1 and 2)50 Recently IL-35 which isessential for the immunosuppressive function of Tregulatory cells51 was also noted to be necessary forthe suppressive function of B cells52 Mice lackingIL-35 in their B cells only were unable to recoverfrom experimental autoimmune encephalomyelitisand had higher macrophage and inflammatoryT cell activity IL-35 was also found to be criticalfor the induction of B regulatory cells53 It isinteresting that in both infectious and autoimmunemodels the major B cell sources of IL-10 and IL-35transcripts were plasmablasts and plasma cells52
suggesting that mature plasma cells expressing IL-35may be important for the maintenance of Bregulatory cell numbers Additional studies will beneeded to confirm plasmablast IL-10 expression theextent to which such expression may be antigen-restricted and whether this expression profile isevident during human plasma cell development
Human B cells with similar features to murineB10 cells have been identified and are expanded inseveral human autoimmune disorders including sys-temic lupus erythematosus (SLE) rheumatoid arthri-tis (RA) Sjoumlgren syndrome and bullous skindisease54 Human B10 regulatory B cells may playan important role in suppressing AQP4-specific andinnate immune responses in NMO (figure 1) Quanet al55 noted both fewer B10 regulatory cells(CD191CD24highCD38high) in relapsing AQP41NMO patients and reduced IL-10 induction follow-ing in vitro stimulation of CD191 B cells The studyhowever may have underestimated the circulatingB10 population as regulatory B10 cells do not haveclearly defined cell differentiation markers and theexpression of CD38high may be unreliable for identi-fying circulating B regulatory cells54 It is interestingthat a reduction in IL-10 expression by CD191 Bcells has been observed in patients with MS56 andthe beneficial effect of parasitic infection in patientswith MS has been associated with an increased pro-duction of B cellndashderived IL-1057 Therefore the
possibility that B cells play an important role in sup-pressing immune activation in NMO warrants fur-ther investigation
B CELL TOLERANCE AND NMO PATHOGENESISCENTRAL AND PERIPHERAL TOLERANCE ANDB CELL ANERGY Central and peripheral tolerance
During early B cell development immunoglobulinvariable region gene segments are stochastically re-combined to generate functional antibodies (B cellreceptors [BCRs]) that are expressed on the cell sur-face This process is fundamental for the generationof the wide diversity of the immunoglobulin reper-toire but also generates autoreactive B cells alongsidethose that comprise the nonself-reactive naive repertoireElimination of autoreactive B cells is controlled bytolerance mechanisms The majority of autoreactive Bcells are eliminated at 2 separate steps58 during B celldevelopment (figure 2) A central checkpoint in theBM between early immature and immature B cellstages removes a large population of developing B cellsthat express autoreactive and polyreactive antibodiesThese antibodies have been shown to display low-levelreactivity to multiple self-antigens by in vitro approachesAfter passing through the first checkpoint only a smallfraction of B cell clones with low levels of self-reactivitymigrate to the periphery59 The second B cell tolerancecheckpoint selects against these autoreactive newlyemigrant B cells before they enter the mature naiumlve Bcell compartment
The mechanisms underlying tolerance defects inautoimmune diseases have not been fully elucidatedCentral tolerance checkpoint integrity is thought tobe associated with B cell-inherent characteristics Thesemechanisms are specifically associated with signalingthrough the BCR60 Conversely the peripheral toler-ance checkpoint is thought to be dependent on T cellB cell interactions particularly those involving regula-tory T cells rather than B cell-inherent characteris-tics61 Dysregulation of central or peripheral B celltolerance may bemeasured by evaluating the frequencyof autoreactive B cells in the naive repertoire59 B celltolerance defects have been clearly demonstrated in anumber of autoimmune diseases For example patientswith RA SLE and type 1 diabetes exhibit defects inboth central and peripheral B cell tolerance check-points that result in the accumulation of self-reactivemature naiumlve B cells in their blood58 This autoreactiveB cell pool is thought to be the reservoir from whichdisease-associated autoantibodies are derived NMOpresents an opportunity to establish whether this con-cept is accurate as B cells that directly produce theautoantibodies have been isolated
Not all autoimmune diseases follow this pattern ofcombined central and peripheral tolerance defects Asubset of patients with MS harbor a naiumlve B cell
6 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
lymphoid tissue2526 This effect could result inenhanced serum AQP4-IgG titers and CNS ASCmigration resulting in the exacerbation of NMO dis-ease activity reported with fingolimod treatment27 Itis interesting that natalizumab another MS therapynoted to exacerbate NMO disease activity and tissueinjury2829 increases circulating eosinophils3031 thatmay contribute directly to lesion formation32 or sup-port local ASCs In contrast inhibition of IL-6 sig-naling using the anti-IL-6 receptor monoclonalantibody tocilizumab reduces PB plasmablasts andAQP4-IgG titers in some patients and appears toreduce relapse activity in patients with NMO33
These observations are consistent with a role forIL-6 in the induction of plasma cells that produceautoantibodies against AQP4 and other relevantantigens23
While some studies have reported a correlationbetween AQP4-IgG titers and clinical relapse a con-sistent relationship with disease activity has yet toemerge2234 A threshold level of AQP4-IgG that trig-gers clinical relapse even in a given individual hasnot been identified and serum levels during relapsediffer widely within and between individuals More-over failure of rituximab has been documented insome patients with NMO experiencing relapsesdespite undetectable CD191 cells in PB and highlevels of AQP4-IgG serum titers35 In one studytreatment with anti-CD20 antibody (rituximab) ap-peared to result in a rapid and transient increase inserum AQP4-IgG titers and clinical relapse in tandemwith increased circulating BAFF levels36 however inanother study successful anti-CD20 therapy did notresult in reduced levels of AQP4-IgG BAFF orAPRIL10 Discordance between AQP4-IgG serumtiters and disease activity may also be explained bycompartmental localization at the site of tissue dam-age in the CNS In addition antibodies may differwith respect to affinity during the course of diseaseThese changes may be undetectable in current assaysIndeed in patients with NMO CSF AQP4-IgGtiters have been observed to correlate more closelywith clinical activity and neuroinflammation thanserum titers37 Alternatively antibodies targetedagainst other self-antigens may contribute to diseaseactivity For instance autoantibodies against myelinoligodendrocyte glycoprotein (MOG) have been de-tected in a small fraction of patients diagnosed withNMO who are AQP4-IgG2 There are also clinicaldifferences between the MOG-IgG1 and AQP4-IgG1 NMO patients although the findings arebased on small numbers of patients and require fur-ther study3839 Unlike antindashAQP4-IgG antindashMOG-IgG injected intracerebrally into murine brain withhuman complement does not reproduce NMOpathology40
PROINFLAMMATORY BCELLS IN NMO B cells mayfacilitate disease activity in NMO by stimulatingpathogenic T cell responses through antigen presenta-tion or cytokine secretion (table 2) B cells expressmajor histocompatibility complex (MHC) class IImolecules constitutively and serve as potentantigen-presenting cells41 B cell MHC II expressionalso contributes to the development of T follicularhelper cells that participate in B cell differentiationand immunoglobulin isotype switching4142 Thuscooperation between AQP4-specific B cells andAQP4-specific T cells may be particularly importantin ASC differentiation and the production of NMO-IgG Moreover the presence of AQP4-specific B cellsmay promote T cell responses against AQP4 thatcontribute to tissue damage B cells may produceproinflammatory cytokines through antigen-specificor polyclonal stimulation IL-6 secretion byproinflammatory memory B cells in NMO mayaggravate disease activity by promoting pathogenicTh17 differentiation Indeed during NMOexacerbations IL-17A and IL-23 are elevatedindicating enhanced Th17 pathway activity43 Areduction in IL-6 secretion following anti-CD20B cell depletion in patients with NMO may explainsome of the discordance between relapse activityreduction and persistently elevated AQP4-IgGtiters Reduced Th17 pathway activity resultingfrom reduction in the number of proinflammatoryCD201 memory B cells may significantly reduceclinical activity despite a minimal effect on AQP4-IgG production by CD202 tissue-resident plasmacells In addition alterations in the cytokine profileof B cells that reconstitute after rituximab treatmentmay explain some benefits of treatment In MS Bcells emerging after anti-CD20 depletion secretelower levels of IL-644 Also reconstituting B cellsmay have a reduced proinflammatory cytokineprofile with lower secretion of lymphotoxin orTNF-a upon activation45
Bystander activation may also result in the produc-tion of B cell cytokines that promote NMO diseaseactivity Cytokines such as TNF-a IL-17 andgranulocyte-macrophage colony-stimulating factor(GM-CSF) may be produced by various plasma cellsubsets and regulate humoral immunity alter inter-actions with commensal microbiota and modifyadaptive and innate immune responses Nitric oxideand TNF-a secreted by plasma cells in the laminapropria of the small intestine modulate IgA secretionand can alter the composition of the gut flora46 Giventhat AQP4-specific T cell responses in NMO displaycross-reactivity to Clostridium perfringens adenosinetriphosphate-binding cassette transporter permeasein vitro47 such changes in commensal microbiotamay be able to influence downstream T cell reactivity
Neurology Neuroimmunology amp Neuroinflammation 5
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
During bacterial and parasitic infections plasma cellsare a source of GM-CSF and IL-174849 Thus in thecontext of NMO increased levels of GM-CSF andIL-17 may result in increased granulocyte recruitmentto the CNS promoting disease relapse Future studiesare needed to directly determine the quantity pheno-type and activity of proinflammatory B cells andASCs in NMO and NMO spectrum diseases
REGULATORY B CELLS IN NMO Ongoing advancesin the understanding of B cell contributions to auto-immunity include the elucidation of a population ofB cells that can negatively regulate cellular immuneresponses and inflammation the most studied ofwhich are those secreting IL-10 termed B10 cells(tables 1 and 2)50 Recently IL-35 which isessential for the immunosuppressive function of Tregulatory cells51 was also noted to be necessary forthe suppressive function of B cells52 Mice lackingIL-35 in their B cells only were unable to recoverfrom experimental autoimmune encephalomyelitisand had higher macrophage and inflammatoryT cell activity IL-35 was also found to be criticalfor the induction of B regulatory cells53 It isinteresting that in both infectious and autoimmunemodels the major B cell sources of IL-10 and IL-35transcripts were plasmablasts and plasma cells52
suggesting that mature plasma cells expressing IL-35may be important for the maintenance of Bregulatory cell numbers Additional studies will beneeded to confirm plasmablast IL-10 expression theextent to which such expression may be antigen-restricted and whether this expression profile isevident during human plasma cell development
Human B cells with similar features to murineB10 cells have been identified and are expanded inseveral human autoimmune disorders including sys-temic lupus erythematosus (SLE) rheumatoid arthri-tis (RA) Sjoumlgren syndrome and bullous skindisease54 Human B10 regulatory B cells may playan important role in suppressing AQP4-specific andinnate immune responses in NMO (figure 1) Quanet al55 noted both fewer B10 regulatory cells(CD191CD24highCD38high) in relapsing AQP41NMO patients and reduced IL-10 induction follow-ing in vitro stimulation of CD191 B cells The studyhowever may have underestimated the circulatingB10 population as regulatory B10 cells do not haveclearly defined cell differentiation markers and theexpression of CD38high may be unreliable for identi-fying circulating B regulatory cells54 It is interestingthat a reduction in IL-10 expression by CD191 Bcells has been observed in patients with MS56 andthe beneficial effect of parasitic infection in patientswith MS has been associated with an increased pro-duction of B cellndashderived IL-1057 Therefore the
possibility that B cells play an important role in sup-pressing immune activation in NMO warrants fur-ther investigation
B CELL TOLERANCE AND NMO PATHOGENESISCENTRAL AND PERIPHERAL TOLERANCE ANDB CELL ANERGY Central and peripheral tolerance
During early B cell development immunoglobulinvariable region gene segments are stochastically re-combined to generate functional antibodies (B cellreceptors [BCRs]) that are expressed on the cell sur-face This process is fundamental for the generationof the wide diversity of the immunoglobulin reper-toire but also generates autoreactive B cells alongsidethose that comprise the nonself-reactive naive repertoireElimination of autoreactive B cells is controlled bytolerance mechanisms The majority of autoreactive Bcells are eliminated at 2 separate steps58 during B celldevelopment (figure 2) A central checkpoint in theBM between early immature and immature B cellstages removes a large population of developing B cellsthat express autoreactive and polyreactive antibodiesThese antibodies have been shown to display low-levelreactivity to multiple self-antigens by in vitro approachesAfter passing through the first checkpoint only a smallfraction of B cell clones with low levels of self-reactivitymigrate to the periphery59 The second B cell tolerancecheckpoint selects against these autoreactive newlyemigrant B cells before they enter the mature naiumlve Bcell compartment
The mechanisms underlying tolerance defects inautoimmune diseases have not been fully elucidatedCentral tolerance checkpoint integrity is thought tobe associated with B cell-inherent characteristics Thesemechanisms are specifically associated with signalingthrough the BCR60 Conversely the peripheral toler-ance checkpoint is thought to be dependent on T cellB cell interactions particularly those involving regula-tory T cells rather than B cell-inherent characteris-tics61 Dysregulation of central or peripheral B celltolerance may bemeasured by evaluating the frequencyof autoreactive B cells in the naive repertoire59 B celltolerance defects have been clearly demonstrated in anumber of autoimmune diseases For example patientswith RA SLE and type 1 diabetes exhibit defects inboth central and peripheral B cell tolerance check-points that result in the accumulation of self-reactivemature naiumlve B cells in their blood58 This autoreactiveB cell pool is thought to be the reservoir from whichdisease-associated autoantibodies are derived NMOpresents an opportunity to establish whether this con-cept is accurate as B cells that directly produce theautoantibodies have been isolated
Not all autoimmune diseases follow this pattern ofcombined central and peripheral tolerance defects Asubset of patients with MS harbor a naiumlve B cell
6 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
During bacterial and parasitic infections plasma cellsare a source of GM-CSF and IL-174849 Thus in thecontext of NMO increased levels of GM-CSF andIL-17 may result in increased granulocyte recruitmentto the CNS promoting disease relapse Future studiesare needed to directly determine the quantity pheno-type and activity of proinflammatory B cells andASCs in NMO and NMO spectrum diseases
REGULATORY B CELLS IN NMO Ongoing advancesin the understanding of B cell contributions to auto-immunity include the elucidation of a population ofB cells that can negatively regulate cellular immuneresponses and inflammation the most studied ofwhich are those secreting IL-10 termed B10 cells(tables 1 and 2)50 Recently IL-35 which isessential for the immunosuppressive function of Tregulatory cells51 was also noted to be necessary forthe suppressive function of B cells52 Mice lackingIL-35 in their B cells only were unable to recoverfrom experimental autoimmune encephalomyelitisand had higher macrophage and inflammatoryT cell activity IL-35 was also found to be criticalfor the induction of B regulatory cells53 It isinteresting that in both infectious and autoimmunemodels the major B cell sources of IL-10 and IL-35transcripts were plasmablasts and plasma cells52
suggesting that mature plasma cells expressing IL-35may be important for the maintenance of Bregulatory cell numbers Additional studies will beneeded to confirm plasmablast IL-10 expression theextent to which such expression may be antigen-restricted and whether this expression profile isevident during human plasma cell development
Human B cells with similar features to murineB10 cells have been identified and are expanded inseveral human autoimmune disorders including sys-temic lupus erythematosus (SLE) rheumatoid arthri-tis (RA) Sjoumlgren syndrome and bullous skindisease54 Human B10 regulatory B cells may playan important role in suppressing AQP4-specific andinnate immune responses in NMO (figure 1) Quanet al55 noted both fewer B10 regulatory cells(CD191CD24highCD38high) in relapsing AQP41NMO patients and reduced IL-10 induction follow-ing in vitro stimulation of CD191 B cells The studyhowever may have underestimated the circulatingB10 population as regulatory B10 cells do not haveclearly defined cell differentiation markers and theexpression of CD38high may be unreliable for identi-fying circulating B regulatory cells54 It is interestingthat a reduction in IL-10 expression by CD191 Bcells has been observed in patients with MS56 andthe beneficial effect of parasitic infection in patientswith MS has been associated with an increased pro-duction of B cellndashderived IL-1057 Therefore the
possibility that B cells play an important role in sup-pressing immune activation in NMO warrants fur-ther investigation
B CELL TOLERANCE AND NMO PATHOGENESISCENTRAL AND PERIPHERAL TOLERANCE ANDB CELL ANERGY Central and peripheral tolerance
During early B cell development immunoglobulinvariable region gene segments are stochastically re-combined to generate functional antibodies (B cellreceptors [BCRs]) that are expressed on the cell sur-face This process is fundamental for the generationof the wide diversity of the immunoglobulin reper-toire but also generates autoreactive B cells alongsidethose that comprise the nonself-reactive naive repertoireElimination of autoreactive B cells is controlled bytolerance mechanisms The majority of autoreactive Bcells are eliminated at 2 separate steps58 during B celldevelopment (figure 2) A central checkpoint in theBM between early immature and immature B cellstages removes a large population of developing B cellsthat express autoreactive and polyreactive antibodiesThese antibodies have been shown to display low-levelreactivity to multiple self-antigens by in vitro approachesAfter passing through the first checkpoint only a smallfraction of B cell clones with low levels of self-reactivitymigrate to the periphery59 The second B cell tolerancecheckpoint selects against these autoreactive newlyemigrant B cells before they enter the mature naiumlve Bcell compartment
The mechanisms underlying tolerance defects inautoimmune diseases have not been fully elucidatedCentral tolerance checkpoint integrity is thought tobe associated with B cell-inherent characteristics Thesemechanisms are specifically associated with signalingthrough the BCR60 Conversely the peripheral toler-ance checkpoint is thought to be dependent on T cellB cell interactions particularly those involving regula-tory T cells rather than B cell-inherent characteris-tics61 Dysregulation of central or peripheral B celltolerance may bemeasured by evaluating the frequencyof autoreactive B cells in the naive repertoire59 B celltolerance defects have been clearly demonstrated in anumber of autoimmune diseases For example patientswith RA SLE and type 1 diabetes exhibit defects inboth central and peripheral B cell tolerance check-points that result in the accumulation of self-reactivemature naiumlve B cells in their blood58 This autoreactiveB cell pool is thought to be the reservoir from whichdisease-associated autoantibodies are derived NMOpresents an opportunity to establish whether this con-cept is accurate as B cells that directly produce theautoantibodies have been isolated
Not all autoimmune diseases follow this pattern ofcombined central and peripheral tolerance defects Asubset of patients with MS harbor a naiumlve B cell
6 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
repertoire that indicates the presence of a defect exclu-sively in peripheral tolerance central tolerance appearsto be intact and functional62 The similarities betweenNMO and MS raise intriguing questions regarding therole of abnormal B cell tolerance in NMO pathogenesisIndeed 2 studies suggest that dysfunctional B cell tol-erance may contribute to the production of AQP4-IgGand other autoreactive peripheral B cells4 Since diseaseactivity is related to the production of AQP4 autoanti-body in the majority of patients with NMO modula-tion in the function of central and peripheral B cellcheckpoints may directly influence disease activityRecombinant antibodies produced from newlyemigrant (CD21lowCD101IgMhighCD27ndash) and mature(CD211CD102IgM1CD27ndash) naiumlve NMO B cellscould be tested for reactivity against AQP4 and otherself-antigens to determine whether the frequency ofautoreactive B cells is greater in seropositive patientsand whether diminished central or peripheral toler-ance is responsible for both AQP4-specific and otherautoreactivities
B cellndashdepleting therapies such as rituximabreduce the relapse rate in NMO (table 3)
Understanding the role of both central and peripheraltolerance in the counterselection of AQP4-reactive Bcells in NMO may guide administration of B cellndashdepleting therapies If central B cell tolerance isimpaired in NMO then the production of a largepopulation of autoreactive B cells in the BM could leadto a rapid relapse of disease activity during B cell recon-stitution Conversely if only peripheral B cell toleranceis abnormal as appears to be the case in MS normal Bcell repertoires should be established following deple-tion and reconstitution Hence understanding B celltolerance in NMO may be critical for determiningwhether induction or continuous B cell depletion isneeded to maintain treatment response62
Anergy B cell anergy is another mechanism that contrib-utes to silencing humoral autoimmunity The role ofB cell anergy in NMO and human autoimmune diseaseis relatively unexplored63 During disease relapse anergicescape may increase the frequency of circulating AQP4-IgG1 plasmablasts and thereby increase AQP4-IgGtiters182234 In combination with central or peripheraldeficiencies of B cell tolerance anergic escape may
Figure 2 B cell development tolerance checkpoints and neuromyelitis optica immunotherapy
Antibodies are generated during early B cell development by random joining of immunoglobulin gene segments The arbi-trary joining is the basis for the vast diversity of the B cell repertoire needed for complete immunity but this comes at aprice the developing B cell repertoire includes autoreactive antibodies B cell tolerance ensures that potentially detrimentalautoreactive B cells are cleared from the B cell repertoire During B cell development autoreactive B cells are removed at 2separate steps In the first a central checkpoint in the bone marrow prior to the transition to immature B cells removes thelarge portion of developing B cells that express polyreactive antibodies leaving a smaller fraction of clones with low levels ofpolyreactivity to migrate from the bone marrow to the periphery The second tolerance checkpoint residing in the peripheryfurther counterselects autoreactive new emigrant B cells before they differentiate into mature naiumlve B cells B cell lineagesthat may be directly affected by potential neuromyelitis optica treatment modalities are shown in the row below the sche-matic Surface markers indicated in the bottom part of the schematic are not intended to be comprehensive and do notinclude all reported subsets Adapted by permission from Macmillan Publishers Ltd Nature Immunology (Meffre ECasellas R Nussenzweig MC Antibody regulation of B cell development 20001379ndash385)
Neurology Neuroimmunology amp Neuroinflammation 7
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
increase the number of self-reactive B cells Analysis ofthe scope and composition of the anergic B cellpopulation in patients with NMO may help tounderstand whether increased humoral autoimmunityin NMO is a result of deficient screening of self-reactiveB cells or abnormal anergic silencing
Human naiumlve B cells with low surface IgM expres-sion (CD191IgD1CD212CD272IgMlow) exhibit lim-ited BCR signaling weak proliferation limiteddifferentiation and poor immunoglobulin productionfollowing stimulation but normal IgM expression con-sistent with an anergic state64 In SLE naive IgMlow Bcells display increased CD95 and decreased CD22expression suggesting increased activation and loss ofanergic silencing Activated complement componentsmay augment B cell responses through interaction withB cell surface complement receptors For exampleC3dg a cleavage product of complement componentC3 has been demonstrated to amplify BCR signalingthrough binding to complement receptor 2 (CR2)65
In NMO complement-mediated injury of astrocytesmay lead to generation of C3dg fragments covalentlyattached to AQP4 or other CNS proteins TheseldquoC3dg-antigen adductsrdquo may allow coligation of CR2and BCR reducing the threshold for B cell activationAs a result naiumlve IgMlow B cells may escape anergicsilencing Therefore theoretically therapies aimed atlimiting complement activation in NMO (table 3)may facilitate anergic silencing
THERAPEUTIC PERSPECTIVES Improved under-standing of the role of B cells in NMO pathogenesis
and a burgeoning array of immunotherapeutics (table3) provide a promising environment for the develop-ment and evaluation of targeted therapeutics Asnoted previously strategies involving B cell depletionor modulation may alter multiple B cell functions sotheir effects cannot easily be attributed to particularsubpopulations that are solely defined by surfacemarkers Consequently B cell subset analysis usingmultiparameter flow cytometry cytokine profilesand functional assays must be integrated to betterunderstand the potential multiple biologic effects ofthese investigational therapies Prospective studiesincorporating high-quality biological assays in well-characterized patient cohorts participating intherapeutic trials with these B cell targetingexperimental agents would be most informative
SUMMARY AND FUTURE DIRECTIONS Emergingevidence suggests that B lineage cells may bemultipurpose contributors toNMO spectrum disordersAQP4-IgG is detectable in the serum and CSF of themajority of patients with NMO and reproducesdisease-specific pathology The contribution of B cellsto NMO pathogenesis however may extend beyondthe production of AQP4-IgG to include an imbalanceof proinflammatory and anti-inflammatory B cellfunctions Accumulating evidence points toward theimportance of antigen presentation and cytokinesecretion however impaired B cell tolerance andaberrant anergic silencing warrant further study Basicand translational research closely associated withclinical studies should be leveraged toward definingthe mechanisms of communication between B cellsand other immune cells that drive NMO pathogenesisin order to identify novel targets for therapeuticintervention
AUTHOR CONTRIBUTIONSDr JL Bennett participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr KC OrsquoConnor participated in the analysis interpretation writing
and critical review of the manuscript for important intellectual content
Dr A Bar-Or participated in the analysis interpretation writing and
critical review of the manuscript for important intellectual content
Dr SS Zamvil participated in the critical review of the manuscript for
important intellectual content Dr B Hemmer participated in the critical
review of the manuscript for important intellectual content Dr TF Tedder
participated in the critical review of the manuscript for important intellectual
content Dr H-C von Budingen participated in the critical review of the
manuscript for important intellectual content Dr O Stuve participated in
the critical review of the manuscript for important intellectual content
Dr MR Yeaman participated in the critical review of the manuscript for
important intellectual content Dr TJ Smith participated in the critical review
of the manuscript for important intellectual content Dr C Stadelmann
participated in the critical review of the manuscript for important intellectual
content
ACKNOWLEDGMENTThe authors thank The Guthy-Jackson Charitable Foundation for its
support in organizing the NMO International Clinical Consortium amp
Biorepository Contributors from the Guthy-Jackson Charitable
Table 3 B cell therapeutics in NMO treatment pipeline
Agent(s) Target Potential mechanism(s)
Rituximabofatumumabocrelizumab
Anti-CD20-mediatedB cell depletion
Reduction in proinflammatory B cells may result intransient BAFF elevation
Medi-551 Anti-CD19-mediatedB cell depletion
Reduction in proinflammatory B cell and APCnumbers may reduce regulatory B cell numbersand activity
Atacicept APRILTACI scavengerreceptor
Reduction of APCs may reduce regulatory B cellnumbers and activity
Belimumab Anti-BAFF antibody Reduction of APCs may reduce regulatory B cellnumbers and activity
Anti-IL-17 Anti-IL-17 antibody Reduce T cell proinflammatory activity reduceCNS neutrophil recruitment reduce B cell survival
Anti-CD22 Anti-CD22-mediateddepletion
Reduction in anergic B cells
Tocilizumab SA237 Anti-IL-6 receptorantibody
Reduction in APC survival autoantibodyproduction reduction in proinflammatory B cellactivity
Anti-IL-5 Anti-IL-5 antibody Reduction of APCs reduction of CNS eosinophilinfiltration
Abbreviations APC 5 antibody-producing cell APRIL 5 a proliferation-inducing ligandBAFF 5 B cell-activating factor IL 5 interleukin NMO 5 neuromyelitis optica TACI 5
transmembrane activator and CAML interactorReferences e-35 and e-36 at Neurologyorgnn provide a recent review of NMO treatmentdata
8 Neurology Neuroimmunology amp Neuroinflammation
ordf 2015 American Academy of Neurology Unauthorized reproduction of this article is prohibited
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
Foundation NMO International Clinical Consortium and Biorepository
and Biorepository Oversight Committee are recognized in alphabetic
order by institution Tanuja Chitnis (Brigham and Womenrsquos Hospital
and Massachusetts General Hospital Boston USA) Raffaele Iorio
(Catholic University Rome Italy) Jens Wuumlrfel (Chariteacute Universitaumltsme-
dizin Berlin University Medicine Goumlttingen Germany) Friedemann
Paul (Chariteacute University Medicine Berlin Berlin Germany) Philippe
Cabre (CHRU Pierre Zobda-Quitman Fort de France Martinique
French West Indies) Danielle van Pelt and Rogier Hintzen (Department
of Neurology Erasmus MC Rotterdam) Romain Marignier (Hocircpital
Neurologique Pierre Wertheimer Hospices Civils de LyonCentre des
Neurosciences de Lyon Lyon France) Pablo Villoslada (Hospital Clinic
and Institute of Biomedical Research August Pi Sunyer (IDIBAPS)
Barcelona Spain) Michael Levy (Johns Hopkins University Baltimore
USA) Lekha Pandit (KS Hegde Medical Academy Nitte University
Mangalore India) Eric Klawiter (Massachusetts General Hospital
Harvard Medical School Boston MA) Brian Weinshenker (Mayo
Clinic Rochester MN USA) Dean Wingerchuk (Mayo Clinic Scotts-
dale AZ USA) Ho Jin Kim (National Cancer Center Goyang-si
Gyeonggi-do Republic of Korea) Silvia Tenembaum (National Pediatric
Hospital Dr Juan P Garrahan Buenos Aires Argentina) Jacqueline
Palace and Maria Isabel Leite (Oxford University Hospitals Trust
Oxford UK) Metha Apiwattanakul (Prasat Neurological Institute
Bangkok Thailand) Simon Broadley (School of Medicine Griffith Uni-
versity Gold Coast Campus QLD Australia amp Department of Neurol-
ogy Gold Coast University Hospital Southport QLD Australia)
Naraporn Prayoonwiwat (Siriraj Hospital Mahidol University Bangkok
Thailand) Kerstin Hellwig (St Josef Hospital Bochum Bochum
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
aquaporin-4 antibody assays Clin Exp Neuroimmunol
20145290ndash303
4 Pittock SJ Lennon VA de Seze J et al Neuromyelitis
optica and non organ-specific autoimmunity Arch Neurol
20086578ndash83
5 Bennett J Lam C Kalluri S et al Intrathecal pathogenic
anti-aquaporin-4 antibodies in early neuromyelitis optica
Ann Neurol 200966617ndash629
6 Saadoun S Waters P Bell BA Vincent A Verkman AS
Papadopoulos MC Intra-cerebral injection of neuromye-
litis optica immunoglobulin G and human complement
produces neuromyelitis optica lesions in mice Brain 2010
133349ndash361
7 Iccediloumlz S Tuumlzuumln E Kuumlrtuumlncuuml M et al Enhanced IL-6 pro-
duction in aquaporin-4 antibody positive neuromyelitis
optica patients Int J Neurosci 201012071ndash75
8 Uzawa A Mori M Arai K et al Cytokine and chemokine
profiles in neuromyelitis optica significance of interleukin-6
Mult Scler 2010161443ndash1452
9 Correale J Fiol M Activation of humoral immunity and
eosinophils in neuromyelitis optica Neurology 200463
2363ndash2370
10 Pellkofer HL Krumbholz M Berthele A et al Long-term
follow-up of patients with neuromyelitis optica after repeated
therapy with rituximab Neurology 2011761310ndash1315
11 Matsushita T Yanaba K Bouaziz JD Fujimoto M
Tedder TF Regulatory B cells inhibit EAE initiation in
mice while other B cells promote disease progression
J Clin Invest 20081183420ndash3430
12 McMillan R Longmire RL Yelenosky R Lang JE
Heath V Craddock CG Immunoglobulin synthesis by
human lymphoid tissues normal bone marrow as a major
site of IgG production J Immunol 19721091386ndash1394
13 Jarius S Paul F Franciotta D et al Cerebrospinal fluid
findings in aquaporin-4 antibody positive neuromyelitis
optica results from 211 lumbar punctures J Neurol Sci
201130682ndash90
14 Jarius S Franciotta D Paul F et al Cerebrospinal fluid
antibodies to aquaporin-4 in neuromyelitis optica and
related disorders frequency origin and diagnostic rele-
vance J Neuroinflammation 2010752
15 Chihara N Aranami T Oki S et al Plasmablasts as migra-
tory IgG-producing cells in the pathogenesis of neuromy-
elitis optica PLoS One 20138e83036
16 Klawiter EC Alvarez E Xu J et al NMO-IgG detected in
CSF in seronegative neuromyelitis optica Neurology
2009721101ndash1103
17 Radbruch A Muehlinghaus G Luger EO et al Compe-
tence and competition the challenge of becoming a long-
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
40 Saadoun S Waters P Owens GP Bennett JL Vincent A
Papadopoulos MC Neuromyelitis optica MOG-IgG
causes reversible lesions in mouse brain Acta Neuropathol
Commun 2014235
41 Molanarfi N Schulze-Topphoff U Weber MS et al MHC
class II-dependent B cell APC function is required for induc-
tion of CNS autoimmunity independent of myelin-specific
antibodies J Exp Med 201320132921ndash2937
42 Deenick EK Chan A Ma CS et al Follicular helper T cell
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm
DOI 101212NXI000000000000010420152 Neurol Neuroimmunol Neuroinflamm
Jeffrey L Bennett Kevin C OConnor Amit Bar-Or et al B lymphocytes in neuromyelitis optica
This information is current as of May 7 2015
ServicesUpdated Information amp
httpnnneurologyorgcontent23e104fullhtmlincluding high resolution figures can be found at
Supplementary Material httpnnneurologyorgcontentsuppl2015050723e104DC1
httpnnneurologyorgcgicollectionall_immunologyAll Immunologyfollowing collection(s) This article along with others on similar topics appears in the
Permissions amp Licensing
httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in
Reprints
httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online
2015 American Academy of Neurology All rights reserved Online ISSN 2332-7812Published since April 2014 it is an open-access online-only continuous publication journal Copyright copy
is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm