Once activated, T cells gain the ability to access both healthy and inflamed nonlymphoid tissues. They are then reactivated to remain in the tissue and exert their effector function only if they encounter their specific Ag. In this study, we set out to determine if the same is true for B cells using a mouse model of CNS autoimmunity that incorporates both T and B cell recognition of a myelin autoantigen. Both T and B cells were common infiltrates of spinal cords in diseased mice. However, unlike T cells, anti-myelin B cells were excluded from the inflamed tissue. Further, CNS B cells did not have a phenotype consistent with Ag-specific activation as it occurs in lymphatic tissue. Instead, they expressed elevated levels of CD80, indicating that B cells may contribute to local inflammation through nonantigen-specific mechanisms.

Multiple sclerosis (MS) is a chronic disease characterized by inflammation of the CNS and demyelination of axons, thought to be driven by an autoimmune response targeting myelin Ags (1, 2). The importance of B cells to ongoing disease was demonstrated by clinical studies using anti-CD20 Abs to deplete B cells in MS patients (3). This finding is complicated by subsequent studies using a different approach to deplete B cells through the blockade of BAFF and APRIL, key cytokines that drive B cell proliferation, maturation, and survival. In contrast to CD20-targeting therapies, there was no evidence of benefit and, instead, disease may have been accelerated (4). Therefore, although B cells as a whole are major players in disease, there is an urgent need to identify the pathogenic and protective subsets and the mechanisms by which they influence disease.

The B cell lineage is best known for production of Abs, via plasmablasts and plasma cells, which target immune effector mechanisms to specific Ags. However, plasma cells are not depleted by anti-CD20, nor are Ab levels effected in the therapeutic time frame (3, 5). Therefore, although Abs may contribute to MS, they are not the primary mechanism by which B cells drive disease. B cells may also contribute to MS by influencing inflammation and the autoimmune T cell response through production of cytokines or by acting as APC (6, 7). Particular attention has recently been paid to B cells that infiltrate the CNS in MS (8). Indeed, several studies of human tissue (912) and in animal models (1319) describe B cell accumulation specifically within the meninges, often forming clusters immediately adjacent to demyelinating lesions. Consensus is forming around the hypothesis that these are sites where activated autoimmune B cells promote pathology from within the CNS in MS (7, 8).

Some studies of human MS brain tissue report that meningeal B cells assemble into highly organized follicles separated from T cell zones (912). Like the structures in lymph nodes that they resemble, these so-called tertiary lymphatic tissues may be capable of sustaining germinal center responses in which Ag-specific B cell clones proliferate and differentiate into memory and effector subsets (20). Similarly, Ab-producing plasma cells have been described in MS brain tissue (9, 11, 12). In more recent studies, evidence of class switch and accumulation of somatic mutations was found in BCR genes isolated from MS brain tissue (7, 21), both of which are strong indicators that B cells in the CNS derive from a germinal center response. The apparent accumulation of Ag-experienced B cell clones indicates that they may be responding to an autoantigen present in the tissue.

Nevertheless, much less is known about B cell pathological mechanisms and B cell invasion of the inflamed CNS in autoimmune disease compared with what is known about T cells. It is well established that, as with other tissues, Ag-activated T cells are able to gain access to the CNS and that those T cells that encounter their specific Ag within the tissue are retained and reactivated to exert their effector function (22). The evidence presented above suggests that the same may also be true of B cells, but this has not yet been demonstrated. We set out to define the B cell relationship with the inflamed CNS in anti-myelin autoimmunity using a mouse model of experimental autoimmune encephalomyelitis (EAE) induced by immunization with a fusion-protein based on the extracellular domain of mouse myelin oligodendrocyte glycoprotein (mMOG) (23). We have previously demonstrated that, unlike most models used currently, this model incorporates both T and B cell recognition of the autoantigen (15). By tracking transferred T and B cells with known specificity for MOG, we were surprised to find that, although activated autoimmune T cells are enriched in the diseased CNS as expected, the opposite was true of myelin-specific B cells. This finding is a significant challenge to our assumptions about how B cells contribute to autoimmune disease pathology from within the CNS.

Wild-type C57BL/6, 2D2 TCR transgenic (24), and OTII TCR-transgenic (OTII) mice (4194;Tg(TcraTcrb)425Cbn/J) were purchased from the Jackson Laboratory. IgHMOG MOG-specific BCR knockin mice (25) were received as a gift from Dr. H. Wekerle. B1-8 mice (26) with a homozygous deletion of the Jκ locus (B1-8 Jκ−/−) (27) were a gift from Dr. A. Haberman. Mice expressing fluorescent proteins within all nucleated cells, either dsRed (6051; Tg(CAG-DsRedMST)1Nagy/J) under control of the β-Actin promoter or eGFP via the ubiquitin promoter [4353; Tg(UBC-GFP)30Scha/J] were obtained from the Jackson Laboratory. All mice were housed under specific pathogen-free conditions at the West Valley Barrier Facility at Western University Canada. Animal protocols (number 2011-047) were approved by the Western University Animal Use Subcommittee.

The following Abs were purchased from BD Biosciences: anti-CD4-V450 (RM4-5), anti-CD45R-V450 (RA3-6B2), anti-CD138-BV421 (281–2), anti-CD19-BV711 (1D3), anti-CD95-PE-Cy7 (Jo2), anti-Bcl6-A647 (K112-91), anti-CD4-A647 (RM4-5), anti-CD62L-A700 (MEL-14), and anti-IgG1-APC (A85-1). The following Abs were purchased from BioLegend: anti-IgKappa-Biotin (RMK-12), anti-IgLambda-Biotin (RML-42), anti-CD80-PECy7 (16-10A1), and anti-His Tag-purified (J099B12). The following Abs were purchased from eBioscience: anti-IgD-eF450 (11–26), anti-CD3-FITC (145-SC11), anti-CD38-PE (90), anti-CD4-PE-Cy5 (RM4-5), anti-IgM-PE-Cy5 (II/41), anti-CD279-Biotin (RMP1-30), and Streptavidin-APC. FluoroMyelin Red for myelin staining was purchased from Invitrogen.

Naive Ag-specific T and B cells were isolated from either RFP+ 2D2 and GFP+ IgHMOG or RFP+ IgHMOG mice, respectively, as previously described (28). Hapten-specific GFP+ B cells were similarly isolated from GFP+ B1-8 Jκ–deficient mice, and nonspecific T and B cells were isolated from RFP+ and GFP+ C57BL/6 mice respectively, when noted. Briefly, lymph nodes and spleens of donor mice (as indicated) were dissociated and T or B cells were isolated using EasySep Negative selection Mouse T and B cell Enrichment Kits (StemCell Technologies). Cells were transferred i.v. into wild-type C57BL/6 recipients, or in some experiments, OTII or B1-8 Jκ−/− mice. Unless otherwise stated, 5 × 105 T cells and 5 × 106 B cells per mouse were transferred 2 d prior to immunization.

To induce a T and B cell autoimmune response targeting MOG-self–antigen we used a novel fusion protein Ag based on the extracellular domain of mMOG [mMOGtag (15)].

mMOGtag protein was isolated and purified as previously described (23). Mice 6–8 wk old were immunized s.c. at two sites on each flank with a total of 0.5 mg of mMOGtag (and 0.5 mg nitrophenol [NP]-mMOGtag, when indicated) in CFA (Sigma-Aldrich). At the same time, mice were also administered 250 ng of pertussis toxin (List Biological Laboratories) i.p. and again 2 d later. Clinical disease was monitored daily and scored as follows: 0, no clinical signs; 1, tail paralysis; 2, tail paralysis and hind limb weakness; 3, hind limb paralysis; and 4, complete hind limb paralysis and front limb weakness. Half points were given for intermediate scores.

Flow cytometry analysis of T cells and B cells harvested from mouse lymph nodes (inguinal, axillary, and cervical), spleen, blood, liver, intestines, and spinal cord was performed as previously described (16). Briefly, all tissues were harvested from mice after perfusion with ice-cold PBS. Individual spinal cords, livers, and intestines were additionally dissociated through a wire mesh after which leukocytes were isolated using a Percoll (GE Healthcare Life Sciences) gradient. Leukocytes were collected at the 37/90% Percoll interface. The spleen, blood, and liver were then lysed for 2 min at 37°C to remove RBCs. Both lymph node and isolated spinal cord cell suspensions were blocked with an anti-Fc-γ receptor (CD16/32 2.4G2) in PBS containing 1% FBS before further incubation with the listed combination of staining Abs. Dead cells were excluded by staining with the Fixable Viability Dye eFluor506 (eBioscience). Flow cytometry was performed on a LSRII cytometer (BD Immunocytometry Systems) and analyzed with FlowJo software (TreeStar).

At the end of the experiment, or earlier if a mouse reached a predetermined disease endpoint, spinal cords were extracted from mice and prepared as previously described (16). Briefly, five to nine evenly spaced spinal cord tissues spanning the lumbar to cervical regions were cut and frozen in OCT (Tissue-Tek) media. Serial cryostat sections (7 μm) were blocked in PBS containing 1% BSA, 0.1% Tween-20, and 10% rat serum before proceeding with staining. Sections were mounted with ProLong Gold Antifade Reagent (Invitrogen) and stored at −20°C. Tiled images of whole spinal cord sections (20×) were imaged using a DM5500B fluorescence microscope (Leica).

PRISM software (GraphPad) was used for all statistical analysis. A Student t test was used for single comparisons, and ANOVA followed by a Student t test with Bonferroni correction was used for multiple comparisons.

We recently demonstrated that immunization of C57BL/6 mice with a novel protein Ag based on the extracellular domain of mMOGtag results in chronic CNS autoimmunity driven by both anti-myelin T and B cells (15). To identify and track the location and differentiation of pathogenic autoimmune cells over the course of the disease, we transferred fluorescent anti-MOG T and B cells to nonfluorescent recipient mice prior to disease induction. Red fluorescent T cells were isolated from 2D2 TCR transgenic mice [specific for MOG35–55 peptide (24)] that also ubiquitously express RFP, and green fluorescent B cells isolated from IgHMOG BCR H chain knockin mice [with enriched specificity for MOG protein (25)] that also ubiquitously express GFP. These cells were transferred to wild-type C57BL/6 mice 2 d prior to immunization with mMOGtag in CFA and i.p. administration of pentoxifylline (PTX) to induce disease. We previously showed that immunization induces a large expansion of the transferred autoimmune T and B cells via a germinal center response (15).

Consistent with our previous observations (15), immunized mice developed physical signs of disease ∼5 d post immunization (Fig. 1A). Diseased mice were sacrificed 11 d post immunization in the acute phase of disease, and spinal cords were harvested for immunofluorescence analysis of pathology and inflammation. Extensive pathology was evident throughout the spinal cord featuring large regions of white matter demyelination associated with CD4+ T cell (Fig. 1B) and CD8+ T cell (data not shown) infiltration. CD45R+ B cells were largely confined to the meninges, often directly adjacent to underlying demyelinating lesions (Fig. 1B).

FIGURE 1.

CNS infiltrating T cells are enriched for autoimmune T cells. GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred into wild-type nonfluorescent C57BL/6 recipients 2 d prior to induction of EAE via immunization with mMOGtag in CFA and i.p. administration of PTX. Mice were monitored daily for signs of disease (A). Note that disease scores are shown only for mice that developed disease. (B) Representative immunofluorescence image of a spinal cord section collected 11 d post immunization from an EAE mouse demonstrating T and B cell infiltration and demyelination. Regions of myelin loss are starred. (C) Representative serial sections demonstrating extensive infiltration of CD4+ cells, many of which were also RFP+ (bottom), into a region of extensive myelin loss in the white matter (top; myelin loss is starred). (D) CD4+ cells in spinal cord sections of C57BL/6 or OTII-recipient EAE mice (F) were quantified and the percent of RFP+ cells is shown. Each symbol represents an individual mouse from three pooled independent experiments (C57BL/6) or a single OTII experiment. (E) Fluorescent MOG-specific T and B cells were transferred to wild-type C57BL/6 recipients prior to EAE induction as described above. Then 20 d post immunization, circulating blood, pooled lymph nodes (LN), draining inguinal lymph nodes (dLN), and spinal cord (SC) were analyzed by FACS. The percent of CD4+ cells that are also RFP+ is shown for each tissue. Each symbol represents an individual mouse. (F) Disease scores for OTII recipient EAE mice used for analysis of RFP+ T cell infiltration of the spinal cord shown in (D). Two mice reached disease severity endpoint and were sacrificed early (dotted arrow) and remaining mice were sacrificed at the solid arrow. Scale bars, 100 μm.

FIGURE 1.

CNS infiltrating T cells are enriched for autoimmune T cells. GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred into wild-type nonfluorescent C57BL/6 recipients 2 d prior to induction of EAE via immunization with mMOGtag in CFA and i.p. administration of PTX. Mice were monitored daily for signs of disease (A). Note that disease scores are shown only for mice that developed disease. (B) Representative immunofluorescence image of a spinal cord section collected 11 d post immunization from an EAE mouse demonstrating T and B cell infiltration and demyelination. Regions of myelin loss are starred. (C) Representative serial sections demonstrating extensive infiltration of CD4+ cells, many of which were also RFP+ (bottom), into a region of extensive myelin loss in the white matter (top; myelin loss is starred). (D) CD4+ cells in spinal cord sections of C57BL/6 or OTII-recipient EAE mice (F) were quantified and the percent of RFP+ cells is shown. Each symbol represents an individual mouse from three pooled independent experiments (C57BL/6) or a single OTII experiment. (E) Fluorescent MOG-specific T and B cells were transferred to wild-type C57BL/6 recipients prior to EAE induction as described above. Then 20 d post immunization, circulating blood, pooled lymph nodes (LN), draining inguinal lymph nodes (dLN), and spinal cord (SC) were analyzed by FACS. The percent of CD4+ cells that are also RFP+ is shown for each tissue. Each symbol represents an individual mouse. (F) Disease scores for OTII recipient EAE mice used for analysis of RFP+ T cell infiltration of the spinal cord shown in (D). Two mice reached disease severity endpoint and were sacrificed early (dotted arrow) and remaining mice were sacrificed at the solid arrow. Scale bars, 100 μm.

Close modal

It is well established that activated but not naive T cells are able to gain access to the noninflamed CNS, and that encountering a specific Ag is required for retention of a T cell in the tissue (29). Although the literature typically generalizes this to include the inflamed CNS in EAE, the evidence for this is limited. Using intravital microscopy of the pial microvasculature in a rat model of EAE, Bartholomäus et al. (30) observed endothelial transmigration of activated T cells regardless of specificity, but noted that only myelin-specific cells were able to migrate away from the immediate vicinity of the vessel. However, our analysis of T cell infiltration of the spinal cords of mice described above revealed that both RFP+ 2D2 and RFP endogenous CD4+ T cells were evident throughout the white matter and meninges (Fig. 1C). The proportion of infiltrating CD4+ T cells that were RFP+ was variable between individual mice (Fig. 1D), yet in most cases exceeded the proportion of RFP+ cells observed in the inguinal lymph nodes in which the anti-MOG response itself was initiated (never exceeding 2–44% as determined by FACS; data not shown). The relative enrichment for MOG-specific RFP+ T cells in the inflamed CNS was confirmed in a separate experiment that used FACS to quantify the accumulation of RFP+ T cells in the spinal cord, circulation, and lymphoid tissues of EAE mice. Indeed, RFP+ cells made up a greater percentage of the total CD4+ T cell population in the spinal cord compared with circulating blood, spleen, lymph nodes, or draining inguinal lymph nodes (Fig. 1E).

In the above experiments it is likely that recipient-derived nonfluorescent MOG-specific T cells also responded to mMOGtag immunization. Therefore, at least some of the infiltrating RFP T cells observed in the spinal cord may have also been MOG specific, potentially resulting in an underestimation of enrichment for MOG specificity. To largely limit the anti-MOG T cell response to transferred RFP+ 2D2 T cells, OTII mice in which the great majority of T cells express an irrelevant TCR against the foreign Ag OVA were used as recipients. Following transfer of fluorescent MOG-specific T and B cells as described above, OTII recipients developed disease in response to immunization with mMOGtag (Fig. 1F). Again, RFP+ cells made up a large proportion, but not all, of the infiltrating CD4+ T cells (Fig. 1D). Therefore, although MOG-specific T cells are highly enriched in the inflamed spinal cord in EAE, Ag specificity is not likely to be an absolute requirement.

FACS analysis revealed that spinal cord T cells express higher levels of the activation marker CD44 and lower levels of CD62L compared with lymph node T cells (Fig. 2A, 2B). By histology, a slight majority of both RFP+ 2D2 T cells and endogenous RFP T cells costained with CD44, regardless of whether C57BL/6 or OTII recipient mice were used (Fig. 2C). Similarly, both RFP+ and RFP T cells stained for intracellular Ki67 in approximately equal numbers (Fig. 2D), indicating that many were in cell cycle. However, we observed no evidence that T cells in either the meninges or infiltrating deeper into the CNS parenchyma expressed PD-1 (data not shown), excluding the possibility that they were either T follicular helper cells or exhausted T cells. Therefore, consistent with established rules for T cell surveillance of the healthy CNS (29), activated anti-myelin T cells are highly enriched in the inflamed CNS in autoimmunity.

FIGURE 2.

Infiltrating T cells have an activated phenotype regardless of anti-myelin specificity. GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred into C57BL/6 recipients, 2 d prior to induction of EAE. Lymph nodes and spinal cords were harvested from diseased mice 20 d post MOGtag immunization and CD4+ T cells were analyzed by FACS. (A) Representative histograms of the mean fluorescence intensity of CD44 (top) and CD62L (bottom) staining of CD4+ RFP+ cells isolated from the inguinal lymph node or the CNS of the same mouse. (B) Mean fluorescent intensity (MFI) of CD44 (top) and CD62L (bottom) staining of endogenous RFP and transfer-derived RFP+ CD4+ T cells from lymph nodes or spinal cords. Each symbol represents an individual mouse. *p < 0.05, **p < 0.01. (C) In two separate similar experiments in which MOG-specific T and B cells were transferred into either C57BL/6- (left) or OTII- (right) recipient mice, spinal cords were harvested and prepared for immunofluorescence staining to characterize infiltrating immune cells (disease score shown in Fig. 1F). The percentage of RFP and RFP+ CD4+ T cells costaining with CD44 was quantified. Each symbol represents an individual mouse. (D) Representative image from the C57BL/6-recpient experiment as described in Fig. 1A showing Ki67 staining of both RFP+ (squares) and RFP (circles) CD4+ T cells in the spinal cord of an EAE mouse. Scale bars, 100 μm.

FIGURE 2.

Infiltrating T cells have an activated phenotype regardless of anti-myelin specificity. GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred into C57BL/6 recipients, 2 d prior to induction of EAE. Lymph nodes and spinal cords were harvested from diseased mice 20 d post MOGtag immunization and CD4+ T cells were analyzed by FACS. (A) Representative histograms of the mean fluorescence intensity of CD44 (top) and CD62L (bottom) staining of CD4+ RFP+ cells isolated from the inguinal lymph node or the CNS of the same mouse. (B) Mean fluorescent intensity (MFI) of CD44 (top) and CD62L (bottom) staining of endogenous RFP and transfer-derived RFP+ CD4+ T cells from lymph nodes or spinal cords. Each symbol represents an individual mouse. *p < 0.05, **p < 0.01. (C) In two separate similar experiments in which MOG-specific T and B cells were transferred into either C57BL/6- (left) or OTII- (right) recipient mice, spinal cords were harvested and prepared for immunofluorescence staining to characterize infiltrating immune cells (disease score shown in Fig. 1F). The percentage of RFP and RFP+ CD4+ T cells costaining with CD44 was quantified. Each symbol represents an individual mouse. (D) Representative image from the C57BL/6-recpient experiment as described in Fig. 1A showing Ki67 staining of both RFP+ (squares) and RFP (circles) CD4+ T cells in the spinal cord of an EAE mouse. Scale bars, 100 μm.

Close modal

Our analysis above confirms that activated, myelin-specific T cells are highly enriched in the inflamed CNS in anti-myelin autoimmunity. To determine if the same is true for infiltrating B cells, spinal cord sections from the experiment described above (Fig. 1A) were separately analyzed for B cell infiltration and accumulation of GFP+ IgHMOG B cells. As noted above, CD45R+ B cells were largely confined to the meninges (Fig. 1A) and were only very rarely observed in the parenchyma (data not shown; also see below). Surprisingly, no GFP+ B cells were found in any of the sections we investigated (example shown in Fig. 3A). By FACS, GFP+ B cells were evident in the draining inguinal lymph nodes of the same mice (Fig. 3B), confirming that the transfer and activation of IgHMOG B cells was successful. Further, both RFP+ T cells and GFP+ B cells were clearly evident by histology in lymph nodes harvested from separate mice at a similar timepoint (Fig. 3C), confirming both that immunization with mMOGtag induced a germinal center response incorporating the progeny of the transferred MOG-specific T and B cells, and that GFP+ B cells can be detected by histology if they are present in the tissue. The exclusion of GFP+ B cells from the CNS was further confirmed in a separate experiment using FACS to analyze diseased spinal cord, circulating blood, and lymphatic tissue harvested from mMOGtag-induced EAE mice. Although GFP+ B cells were evident in peripheral tissues, they were completely absent from the CNS (Fig. 3D).

FIGURE 3.

Autoimmune B cells are excluded from the inflamed spinal cords of mMOGtag-induced EAE. GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred into wild-type nonfluorescent C57BL/6 recipients 2 d prior to induction of EAE (disease score shown in Fig. 1A) (A) Representative immunofluorescence image of a spinal cord section showing a meningeal B cell cluster devoid of transfer-derived GFP+ B cells. (B) FACS analysis of the inguinal lymph nodes of the same mouse demonstrating the presence of GFP+ B cells. (C) In a separate experiment, RFP+ and GFP+ MOG-specific T and B cells were transferred into C57BL/6 recipient mice and 2 d later mMOGtag was administered via footpad injection. Immunofluorescence analysis of a draining inguinal lymph node harvested 10 d post immunization reveals the presence of MOG-specific RFP+ T cells and GFP+ B cells. (D) As described in Fig. 1E, fluorescent MOG-specific T and B cells were transferred to wild-type C57BL/6 recipients prior to EAE induction. 20 d post immunization, circulating blood (Blood), pooled lymph nodes (LN), draining inguinal lymph nodes (dLN), and spinal cord (SC) were analyzed by FACS. The percent of CD45R+ CD19+ cells that are also GFP+ is shown for each tissue. (E) MOG-specific RFP+ T cells and GFP+ B cells were transferred into nonfluorescent C57BL/6 recipient mice 2 d prior to induction of EAE. Then 8 d post immunization, cells isolated from the dLN, other pooled lymph nodes (LN), spleen (SP), intestine (In), liver (Li), and spinal cord (SC) were analyzed by FACS. Each symbol represents an individual mouse. (F) In two separate but repeat experiments, fluorescent MOG-specific T and B cells were transferred to B1-8 Jκ−/− recipient mice bearing a mutant BCR specific for an irrelevant Ag. One experiment with n = 4 ended 18 d post immunization (dotted arrow), whereas the other went out 34 d (n = 2). (G) Immunofluorescence was performed on spinal cords harvested 34 d post mMOGtag immunization. One representative image is shown of a typical meningeal B cell cluster devoid of GFP+ cells. One infiltrating parenchymal B cell (closed triangle) was evident in the white matter of a diseased mouse (see inset box shown at higher magnification, right). Scale bars, 100 μm.

FIGURE 3.

Autoimmune B cells are excluded from the inflamed spinal cords of mMOGtag-induced EAE. GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred into wild-type nonfluorescent C57BL/6 recipients 2 d prior to induction of EAE (disease score shown in Fig. 1A) (A) Representative immunofluorescence image of a spinal cord section showing a meningeal B cell cluster devoid of transfer-derived GFP+ B cells. (B) FACS analysis of the inguinal lymph nodes of the same mouse demonstrating the presence of GFP+ B cells. (C) In a separate experiment, RFP+ and GFP+ MOG-specific T and B cells were transferred into C57BL/6 recipient mice and 2 d later mMOGtag was administered via footpad injection. Immunofluorescence analysis of a draining inguinal lymph node harvested 10 d post immunization reveals the presence of MOG-specific RFP+ T cells and GFP+ B cells. (D) As described in Fig. 1E, fluorescent MOG-specific T and B cells were transferred to wild-type C57BL/6 recipients prior to EAE induction. 20 d post immunization, circulating blood (Blood), pooled lymph nodes (LN), draining inguinal lymph nodes (dLN), and spinal cord (SC) were analyzed by FACS. The percent of CD45R+ CD19+ cells that are also GFP+ is shown for each tissue. (E) MOG-specific RFP+ T cells and GFP+ B cells were transferred into nonfluorescent C57BL/6 recipient mice 2 d prior to induction of EAE. Then 8 d post immunization, cells isolated from the dLN, other pooled lymph nodes (LN), spleen (SP), intestine (In), liver (Li), and spinal cord (SC) were analyzed by FACS. Each symbol represents an individual mouse. (F) In two separate but repeat experiments, fluorescent MOG-specific T and B cells were transferred to B1-8 Jκ−/− recipient mice bearing a mutant BCR specific for an irrelevant Ag. One experiment with n = 4 ended 18 d post immunization (dotted arrow), whereas the other went out 34 d (n = 2). (G) Immunofluorescence was performed on spinal cords harvested 34 d post mMOGtag immunization. One representative image is shown of a typical meningeal B cell cluster devoid of GFP+ cells. One infiltrating parenchymal B cell (closed triangle) was evident in the white matter of a diseased mouse (see inset box shown at higher magnification, right). Scale bars, 100 μm.

Close modal

Finally, a similar experiment was performed to track GFP+ MOG-specific B cells in preclinical mice to determine if autoimmune cells are present in the CNS at an earlier timepoint. Again, although RFP+ MOG-specific T cells were readily apparent in the spinal cords of mMOGtag-immunized mice (data not shown), GFP+ cells were absent from the spinal cords (Fig. 3E), but detectable in the draining inguinal lymph nodes, other lymphatic tissues (other peripheral lymph nodes and spleen) and liver. Except for a single mouse, very few were observed in the intestines.

As for T cells, nonfluorescent, C57BL/6 recipient-derived B cells would also be able to respond to immunization with mMOGtag. Therefore, it is possible that GFP B cells infiltrating the spinal cord were myelin specific and somehow outcompeted the transferred GFP+ IgHMOG cells for access to the tissue. To exclude this possibility, fluorescent IgHMOG B cells were transferred to B1-8 Jκ−/− recipient mice in which >95% of B cells are specific for the irrelevant NP hapten Ag (27), effectively preventing an endogenous anti-MOG response. Disease was induced via immunization with mMOGtag, as described above (Fig. 3F). Again, meningeal B cell clusters were clearly evident in the spinal cords of diseased mice (Fig. 3G, left). CD45R+ cells were only very rarely observed in the white matter parenchyma (Fig. 3G, right). No GFP+ MOG-specific B cells were observed in any section imaged. Therefore, anti-MOG B cells are excluded from the inflamed tissue in a model of CNS autoimmunity that incorporates B cell activation and recognition of myelin Ag (15).

In the above experiments transferred GFP+ B cells may have been excluded due to their specificity for a CNS Ag or alternatively due to their activation status. To differentiate between these possibilities, we transferred both RFP+ IgHMOG MOG-specific B cells and GFP+ NP-specific B1-8 Jκ−/− B cells into nonfluorescent wild-type C57BL/6 mice. Recipients were immunized with both mMOGtag and NP-haptenated mMOGtag in CFA. Disease developed normally in these mice (Fig. 4A), and draining inguinal lymph nodes and spinal cords were harvested for analysis 21 d post immunization. FACS analysis of the lymph nodes confirmed that both transferred MOG-specific RFP+ and NP-specific GFP+ B cells were activated and proliferated in response to immunization with the combined Ags (Fig. 4B, left). Despite this, neither RFP+ nor GFP+ B cells were present in the spinal cords of diseased mice (Fig. 4B, right; 4C). This suggests that activated B cells, regardless of specificity, are excluded from the inflamed CNS.

FIGURE 4.

Activated B cells are excluded from the inflamed CNS in MOGtag-induced EAE. RFP+ IgHMOG and GFP+ B1-8 Jκ−/− B cells were transferred into C57BL/6 mice 2 d prior to immunization with both mMOGtag and NP-MOGtag in CFA, accompanied by PTX i.p. Mice were monitored daily for signs of disease (A). Then 21 d post immunization, lymph nodes and spinal cords were harvested from sick mice and analyzed by FACS for the presence of RFP+ and GFP+ B cells (B and C). One representative set of plots (n = 4) for a lymph node (left) and spinal cord (right) from the same mouse is shown in (B). Each data point represents an individual mouse in (C). *p < 0.05, **p < 0.01. (D) In a separate experiment, GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred to C57BL/6 recipients 2 d prior to EAE induction. Lymph nodes and spinal cords were harvested from sick mice 20 d post immunization and analyzed by FACS. Representative plots gated on B cells (left) show the presence of GFP+ B cells in lymph nodes (top) but not spinal cord (bottom). The CD38/CD95 expression profile of GFP B cells was further analyzed to determine activation status (right). (E) CD80 expression by naive/memory phenotype B cells (CD38hi CD95lo) from the lymph node and spinal cord was quantified. Each symbol represents an individual mouse. (F) Immunofluorescence staining of spinal cord sections taken from the experiment described in Fig. 1A showing no Ki67 staining of meningeal B cells. (G) IgD (bottom left) but no IgG (bottom right) staining, was observed in the spinal cord of diseased mice. Lymph nodes were stained as a positive controls (top). Scale bars, 100 μm.

FIGURE 4.

Activated B cells are excluded from the inflamed CNS in MOGtag-induced EAE. RFP+ IgHMOG and GFP+ B1-8 Jκ−/− B cells were transferred into C57BL/6 mice 2 d prior to immunization with both mMOGtag and NP-MOGtag in CFA, accompanied by PTX i.p. Mice were monitored daily for signs of disease (A). Then 21 d post immunization, lymph nodes and spinal cords were harvested from sick mice and analyzed by FACS for the presence of RFP+ and GFP+ B cells (B and C). One representative set of plots (n = 4) for a lymph node (left) and spinal cord (right) from the same mouse is shown in (B). Each data point represents an individual mouse in (C). *p < 0.05, **p < 0.01. (D) In a separate experiment, GFP+ IgHMOG B cells and RFP+ 2D2 T cells were transferred to C57BL/6 recipients 2 d prior to EAE induction. Lymph nodes and spinal cords were harvested from sick mice 20 d post immunization and analyzed by FACS. Representative plots gated on B cells (left) show the presence of GFP+ B cells in lymph nodes (top) but not spinal cord (bottom). The CD38/CD95 expression profile of GFP B cells was further analyzed to determine activation status (right). (E) CD80 expression by naive/memory phenotype B cells (CD38hi CD95lo) from the lymph node and spinal cord was quantified. Each symbol represents an individual mouse. (F) Immunofluorescence staining of spinal cord sections taken from the experiment described in Fig. 1A showing no Ki67 staining of meningeal B cells. (G) IgD (bottom left) but no IgG (bottom right) staining, was observed in the spinal cord of diseased mice. Lymph nodes were stained as a positive controls (top). Scale bars, 100 μm.

Close modal

Phenotypic analysis of CNS-infiltrating B cells supports the contention that they are not activated in a conventional way as occurs in response to specific Ag in lymphatic tissue. By FACS, spinal cord B cells were exclusively CD38hi CD95lo, a phenotype shared by naive and memory B cells (Fig. 4D). However, compared with lymph node B cells of the same CD38hi CD95lo phenotype, spinal cord B cells expressed elevated levels of surface CD80 (Fig. 4E), indicating that they are activated to some degree. Importantly, this phenotype is identical to that of spinal cord B cells, which we recently described in a different B cell–dependent EAE model that develops spontaneously in mice expressing both the 2D2 TCR and IgHMOG BCR (16), therefore it is not an artifact unique to the mMOGtag-induced model.

Follow-up analysis by histology did not find any evidence of class switch in meningeal B cells, as they were exclusively IgD+ and no IgG1+ cells were observed (Fig. 4G). Parenchymal B cells were too rare to analyze reliably. Further, we did not find any evidence of plasma cells in spinal cord tissue as indicated by the absence of cells with the intracellular L chain (data not shown), an indicator of large-scale Ab production, which is a fundamental property of plasma cells. In contrast to T cells (see above), very few B cells in the meninges or rare parenchymal cells stained with Ki67 (Fig. 4F), indicating that the large majority were not proliferating.

To determine if unactivated, naive B cells are able to access the CNS in EAE and to confirm that B cell exclusion from the CNS was not due to their expression of GFP or due to the method of cell transfer, B cells were isolated from GFP-expressing but otherwise wild-type mice, and transferred to already immunized mice at the preclinical stage of disease (day 6 post immunization). Two days post transfer, draining inguinal lymph nodes, pooled lymph nodes, and spinal cords were harvested for analysis by FACS. As expected, GFP+ B cells were recovered from lymph nodes (Fig. 5A), consistent with normal homing of naive cells to lymphatic tissue. Small numbers of GFP+ cells were also recovered from spinal cords and, although very rare in absolute terms (data not shown), the proportion of total B cells isolated from spinal cords that were GFP+ was not significantly different from lymphatic tissue. Similarly, in a separate experiment RFP+ IgHMOG and GFP+ B1-8 Jκ−/− B cells were isolated from unimmunized mice and transferred into recipients with already established mMOGtag-induced EAE (Fig. 5B). Two days post transfer, draining inguinal lymph nodes, pooled lymph nodes, and spinal cords were harvested for analysis by FACS. Both GFP+ and RFP+ B cells were recovered from lymph nodes, (Fig. 5C) and, consistent with the previous experiment, even though in absolute terms the number of transferred B cells recovered from the spinal cord was very small (Fig. 5C, left), they made up a similar proportion of total B cells recovered from both tissues (Fig. 5C, right). No significant bias between NP- or MOG-specific B cells was observed, suggesting that the activation state rather than specificity is the primary factor limiting initial access to the tissue. Together, these experiments suggest that, unlike activated B cells, small numbers of unactivated cells are recruited to the inflamed spinal cord in EAE. However, because transferred B cells made up a similar proportion of the total B cell pool, the relative rate of recruitment is similar in the spinal cord and lymphatic tissue. The complete absence of MOG-specific B cells observed in previous experiments could be due to the subsequent removal or loss of cells that encounter Ag once in the CNS, or because all Ag-specific cells were activated by immunization.

FIGURE 5.

Naive B cells can access the inflamed CNS. EAE was induced in C57BL/6 mice and 6 d post immunization wild-type RFP+ T cell and GFP+ B cells were transferred i.v. into recipient mice (A). Then 2 d later mice were sacrificed and inguinal lymph nodes that drain the site of immunization (dLN), axial and brachial lymph nodes (LN), and spinal cords (SC) were analyzed by FACS for the presence of nonspecific GFP+ B cells. (B) In a separate experiment EAE was induced in C57BL/6 mice. Once disease was established, RFP+ IgHMOG B cells and GFP+ B1-8 Jκ−/− B cells were isolated from healthy donor mice and transferred i.v. into the already sick EAE recipient mice (timepoint of transfer indicated by the arrow). (C) After 2 d mice were sacrificed and LN, dLN, and the SC were analyzed by FACS for the presence of both the absolute number (left) and percent of (right) RFP+ and GFP+ B cells. Each symbol represents an individual mouse.

FIGURE 5.

Naive B cells can access the inflamed CNS. EAE was induced in C57BL/6 mice and 6 d post immunization wild-type RFP+ T cell and GFP+ B cells were transferred i.v. into recipient mice (A). Then 2 d later mice were sacrificed and inguinal lymph nodes that drain the site of immunization (dLN), axial and brachial lymph nodes (LN), and spinal cords (SC) were analyzed by FACS for the presence of nonspecific GFP+ B cells. (B) In a separate experiment EAE was induced in C57BL/6 mice. Once disease was established, RFP+ IgHMOG B cells and GFP+ B1-8 Jκ−/− B cells were isolated from healthy donor mice and transferred i.v. into the already sick EAE recipient mice (timepoint of transfer indicated by the arrow). (C) After 2 d mice were sacrificed and LN, dLN, and the SC were analyzed by FACS for the presence of both the absolute number (left) and percent of (right) RFP+ and GFP+ B cells. Each symbol represents an individual mouse.

Close modal

Together, our findings suggest that B cells have a very different relationship with the inflamed CNS in autoimmunity than T cells do, in that Ag-specific cells actively participating in the anti-myelin response are excluded from the tissue rather than enriched. The absence of activated, autoimmune B cells in the inflamed CNS in two mouse models of anti-myelin autoimmune disease [in this study and (16)] may be interpreted as being at odds with findings from studies of human MS. However, close reading of this literature suggests that this may not be the case. Descriptions of B cell infiltration in MS brain tissues have largely focused on the most highly organized structures resembling lymphoid tissue that may be able to sustain a germinal center, however, it is clear that most clusters of B cells in the meninges are much less organized. Not all studies of human MS brain tissue found evidence of organized tertiary tissues (31), and even those that did report that most B cells were in less organized clusters not containing germinal centers (912), perhaps more reminiscent of the clusters we describe in our models [in this study and (15, 16)]. Similarly, although plasma cells have been reported to be in MS brain tissue (9, 11, 12, 32), this too is not a universal finding (11, 21).

Nevertheless, studies analyzing BCR genes cloned from MS brains report that they are often class switched and show evidence of accumulating somatic mutations (7, 21, 33), both of which are very strong indications that they come from B cells derived from a germinal center response. Indeed, a recent important study from Stern et al. (21) traced the clonal lineage of brain B cells and showed that some were related to B cells found in deep cervical lymph nodes, building on their previous study demonstrating that B cells in the meninges are also clonally related to those deeper parenchyma (33). Importantly, they found no evidence of germinal center responses within the CNS tissues they analyzed, nor did they observe plasma cells. Therefore, although they provide clear evidence that the B cells found in MS brains had previously been activated and likely derived from germinal center responses in cervical lymph nodes, the cells in the tissue were not themselves in an activated state, at least not in a conventional sense as we understand Ag-specific B cell activation in lymph nodes. Most critically, in a recent follow-up study from the same group, no evidence was found that BCRs isolated from MS brains recognized CNS or MS-specific Ags (34). Therefore, instead of contradicting the human literature, we believe that our experimental findings in mouse models are entirely consistent with observations of human MS brain tissue.

Our studies employ a unique experimental system, adapted from our studies of B cell responses to foreign model Ags (28), which allows us to identify and track myelin-specific cells throughout the developing autoimmune response and associated disease pathology. This system depends on the transfer of small numbers of fluorescent T and B cells with known anti-MOG specificity to nonfluorescent recipients, which are then activated and expanded to large numbers in lymph nodes via immunization with Ag. The success of this transfer is demonstrated by the presence of GFP+ cells in draining popliteal lymph nodes as, in our experience, without immunization transferred naïve B cells are lost and not detectable in any tissue after a small number of days. Further, as appreciable numbers of GFP+ B cells were also found in other tissues including noninvolved lymphoid tissue and the liver, the egress and subsequent recirculation of the progeny of transferred GFP+ B cells is also not defective. Exclusion from the inflamed CNS may be determined by recruitment mechanisms limiting that access of newly activated B cells. Alternatively, myelin-reactive cells may be induced to leave the tissue or die after encountering specific Ag, although this would likely also be tied to activation status as naive MOG-specific B cells were detectable in the inflamed CNS for at least 2 d post transfer.

In the work presented in this study we are not able to differentiate between naive and IgM memory B cells. By definition, memory B cells derive from previously activated cells but are themselves quiescent (35). Unless they have undergone class switch, memory B cells are very difficult to differentiate from naive B cells phenotypically. In humans, CD27 is often used as a surface marker to identify memory B cells and CD27+ B cells have been reported to be present or even enriched in meningeal clusters in MS (36). Unfortunately, no equivalent memory marker exists in the mouse model system. Nevertheless, because we observed virtually no class switch in the meningeal B cells in our model, we suspect that most were naive, rather than memory. This does contrast with human MS studies that, in addition to IgM+ B cells, also note IgG and IgA switched cells (21, 37). However, although this does represent a discrepancy between human MS and our model, it should be noted that humans accumulate large numbers of memory B cells over decades of exposure to numerous pathogens. Like central memory T cells, memory B cells are thought to have homing properties similar to naive B cells (38) and therefore would accumulate in the same tissues. Our short-lived mouse models housed under very clean conditions will not accumulate B cell memory to nearly the same degree and therefore do not recapitulate this aspect of human disease. This could explain why we only observe apparently naive B cells in our mouse models, whereas memory cells are reported to be common in human MS tissue.

The distribution of B cells in the inflamed CNS may represent another discrepancy between our mouse model and human MS. In our models, B cells in the spinal cord are largely restricted to the meninges with only very rare cells found deeper in the parenchyma. Some histological studies of human MS brain tissue do report B cells in the parenchyma and white matter lesions (33, 39, 40). However, as for descriptions of plasma cells (see above), parenchymal B cells are not always observed (37). This is further complicated by descriptions of parenchymal cells as perivascular (33, 36), as these can be considered to be associated with the pia mater. In the few cases where the distinction is made, parenchymal B cells not directly associated with a venule are very sparse, as we observe in our models. Regardless, analysis of BCR genes isolated from the peripheral meninges and from deeper parenchymal MS brain tissue demonstrated that they were clonally related (33). Similarly, we did not observe any evidence of multiple subsets of B cells in the spinal cord by FACS (which would also incorporate rare parenchymal cells), nor was there any evidence by histology that the rare parenchymal B cells were activated any more than meningeal cells (by Ki67 staining, for example; data not shown).

Regardless of whether they are naive or memory, it is not clear how Ag-nonspecific meningeal B cells could promote CNS pathology. B cells are capable of presenting Ag to T cells and the physical association between T and B cells in meningeal clusters and the elevated expression of CD80 by B cells in the CNS [Fig. 4E and (16)] strongly suggests that this is a role they play within the CNS. However, B cell presentation of Ag is best understood in the context of the germinal center response where B cells internalize Ag that binds their own specific BCR for presentation to T cells specific for the same Ag. Little is known about nonspecific Ag presentation by B cells. Further research is required to determine how B cells acquire their semiactivated phenotype and if their APC function is indeed enhanced. Although it is possible that they arrive already stimulated, we believe it is more likely that they are activated by locally produced cytokines within the inflamed CNS following recruitment. Also, future work will need to understand the mechanisms and outcomes of noncognate interactions with autoimmune T cells in the CNS, as these may differ considerably from those that occur in the GC response in lymphatic tissue.

Finally, our work presented in this study has important implications for our understanding of the mechanistic contribution(s) of autoimmune B cells to CNS autoimmunity. Both we and others have demonstrated that anti-myelin B cells do indeed contribute to the incidence and/or severity of disease in models that accommodate B cell recognition of Ag (15, 16, 4143). Nevertheless, contrary to our expectations, autoimmune B cells are excluded from the inflamed site. More studies will be required to determine if this is a general feature of B cell responses or if it is unique to autoimmunity or the CNS environment. Regardless, we propose a model where B cells contribute to disease through (at least) two different mechanisms exerted from different anatomical sites. First, anti-myelin B cells may drive disease from the periphery, perhaps by influencing T cell activation and in some cases Ab production. Second, Ag-nonspecific cells may contribute to local pathology from within the CNS, although we have not yet confirmed that this is the case. Potential pathogenic mechanisms include noncognate Ag presentation to T cells and production of inflammatory mediators. Investigations of the activation status of meningeal B cells will aid greatly in resolving this important issue.

We thank the veterinarians and animal care staff at the West Valley Barrier Facility for excellent husbandry of experimental animals. We would also like to thank Heather Craig for technical help and expertise.

This work was supported by a grant from the Canadian Institutes of Health Research. Y.T. is the recipient of an Ontario Graduate Scholarship. R.J. is the recipient of a studentship from the Multiple Sclerosis Society of Canada.

Abbreviations used in this article:

EAE

experimental autoimmune encephalomyelitis

MOG

mylelin oligodendrocyte glycoprotein

mMOG

mouse myelin oligodendrocyte glycoprotein

MS

multiple sclerosis

NP

nitrophenol

OTII

OTII TCR-transgenic

PTX

pentoxifylline.

1
Compston
,
A.
,
A.
Coles
.
2008
.
Multiple sclerosis.
Lancet
372
:
1502
1517
.
2
Ransohoff
,
R. M.
,
D. A.
Hafler
,
C. F.
Lucchinetti
.
2015
.
Multiple sclerosis-a quiet revolution.
Nat. Rev. Neurol.
11
:
134
142
.
3
Barun
,
B.
,
A.
Bar-Or
.
2012
.
Treatment of multiple sclerosis with anti-CD20 antibodies.
Clin. Immunol.
142
:
31
37
.
4
Kappos
,
L.
,
H.-P.
Hartung
,
M. S.
Freedman
,
A.
Boyko
,
E.-W.
Radü
,
D. D.
Mikol
,
M.
Lamarine
,
Y.
Hyvert
,
U.
Freudensprung
,
T.
Plitz
,
J.
van Beek
;
ATAMS Study Group
.
2014
.
Atacicept in multiple sclerosis (ATAMS): a randomised, placebo-controlled, double-blind, phase 2 trial.
Lancet Neurol.
13
:
353
363
.
5
Krumbholz
,
M.
,
T.
Derfuss
,
R.
Hohlfeld
,
E.
Meinl
.
2012
.
B cells and antibodies in multiple sclerosis pathogenesis and therapy.
Nat. Rev. Neurol.
8
:
613
623
.
6
Claes
,
N.
,
J.
Fraussen
,
P.
Stinissen
,
R.
Hupperts
,
V.
Somers
.
2015
.
B cells are multifunctional players in multiple sclerosis pathogenesis: insights from therapeutic interventions.
Front. Immunol.
6
:
642
.
7
Michel
,
L.
,
H.
Touil
,
N. B.
Pikor
,
J. L.
Gommerman
,
A.
Prat
,
A.
Bar-Or
.
2015
.
B cells in the multiple sclerosis central nervous system: trafficking and contribution to CNS-compartmentalized inflammation.
Front. Immunol.
6
:
636
.
8
Pikor
,
N. B.
,
A.
Prat
,
A.
Bar-Or
,
J. L.
Gommerman
.
2016
.
Meningeal tertiary lymphoid tissues and multiple sclerosis: a gathering place for diverse types of immune cells during CNS autoimmunity.
Front. Immunol.
6
:
657
.
9
Magliozzi
,
R.
,
O.
Howell
,
A.
Vora
,
B.
Serafini
,
R.
Nicholas
,
M.
Puopolo
,
R.
Reynolds
,
F.
Aloisi
.
2007
.
Meningeal B-cell follicles in secondary progressive multiple sclerosis associate with early onset of disease and severe cortical pathology.
Brain
130
:
1089
1104
.
10
Magliozzi
,
R.
,
O. W.
Howell
,
C.
Reeves
,
F.
Roncaroli
,
R.
Nicholas
,
B.
Serafini
,
F.
Aloisi
,
R.
Reynolds
.
2010
.
A gradient of neuronal loss and meningeal inflammation in multiple sclerosis.
Ann. Neurol.
68
:
477
493
.
11
Serafini
,
B.
,
B.
Rosicarelli
,
R.
Magliozzi
,
E.
Stigliano
,
F.
Aloisi
.
2004
.
Detection of ectopic B-cell follicles with germinal centers in the meninges of patients with secondary progressive multiple sclerosis.
Brain Pathol.
14
:
164
174
.
12
Howell
,
O. W.
,
C. A.
Reeves
,
R.
Nicholas
,
D.
Carassiti
,
B.
Radotra
,
S. M.
Gentleman
,
B.
Serafini
,
F.
Aloisi
,
F.
Roncaroli
,
R.
Magliozzi
,
R.
Reynolds
.
2011
.
Meningeal inflammation is widespread and linked to cortical pathology in multiple sclerosis.
Brain
134
:
2755
2771
.
13
Kuerten
,
S.
,
A.
Schickel
,
C.
Kerkloh
,
M. S.
Recks
,
K.
Addicks
,
N. H.
Ruddle
,
P. V.
Lehmann
.
2012
.
Tertiary lymphoid organ development coincides with determinant spreading of the myelin-specific T cell response.
Acta Neuropathol.
124
:
861
873
.
14
Magliozzi
,
R.
,
S.
Columba-Cabezas
,
B.
Serafini
,
F.
Aloisi
.
2004
.
Intracerebral expression of CXCL13 and BAFF is accompanied by formation of lymphoid follicle-like structures in the meninges of mice with relapsing experimental autoimmune encephalomyelitis.
J. Neuroimmunol.
148
:
11
23
.
15
Dang
,
A. K.
,
R. W.
Jain
,
H. C.
Craig
,
S. M.
Kerfoot
.
2015
.
B cell recognition of myelin oligodendrocyte glycoprotein autoantigen depends on immunization with protein rather than short peptide, while B cell invasion of the CNS in autoimmunity does not.
J. Neuroimmunol.
278
:
73
84
.
16
Dang
,
A. K.
,
Y.
Tesfagiorgis
,
R. W.
Jain
,
H. C.
Craig
,
S. M.
Kerfoot
.
2015
.
Meningeal infiltration of the spinal cord by non-classically activated B cells is associated with chronic disease course in a spontaneous B cell-dependent model of CNS autoimmune disease.
Front. Immunol.
6
:
470
.
17
Peters
,
A.
,
L. A.
Pitcher
,
J. M.
Sullivan
,
M.
Mitsdoerffer
,
S. E.
Acton
,
B.
Franz
,
K.
Wucherpfennig
,
S.
Turley
,
M. C.
Carroll
,
R. A.
Sobel
, et al
.
2011
.
Th17 cells induce ectopic lymphoid follicles in central nervous system tissue inflammation.
Immunity
35
:
986
996
.
18
Walker-Caulfield
,
M. E.
,
J. K.
Hatfield
,
M. A.
Brown
.
2015
.
Dynamic changes in meningeal inflammation correspond to clinical exacerbations in a murine model of relapsing-remitting multiple sclerosis.
J. Neuroimmunol.
278
:
112
122
.
19
Bielecki
,
B.
,
I.
Jatczak-Pawlik
,
P.
Wolinski
,
A.
Bednarek
,
A.
Glabinski
.
2015
.
Central nervous system and peripheral expression of CCL19, CCL21 and their receptor CCR7 in experimental model of multiple sclerosis.
Arch. Immunol. Ther. Exp. (Warsz.)
63
:
367
376
.
20
Aloisi
,
F.
,
R.
Pujol-Borrell
.
2006
.
Lymphoid neogenesis in chronic inflammatory diseases.
Nat. Rev. Immunol.
6
:
205
217
.
21
Stern
,
J. N.
,
G.
Yaari
,
J. A.
Vander Heiden
,
G.
Church
,
W. F.
Donahue
,
R. Q.
Hintzen
,
A. J.
Huttner
,
J. D.
Laman
,
R. M.
Nagra
,
A.
Nylander
, et al
.
2014
.
B cells populating the multiple sclerosis brain mature in the draining cervical lymph nodes.
Sci. Transl. Med.
6
:
248ra107
.
22
Goverman
,
J.
2009
.
Autoimmune T cell responses in the central nervous system.
Nat. Rev. Immunol.
9
:
393
407
.
23
Jain
,
R. W.
,
A. K.
Dang
,
S. M.
Kerfoot
.
2016
.
Simple and efficient production and purification of mouse myelin oligodendrocyte glycoprotein for experimental autoimmune encephalomyelitis studies.
J. Vis. Exp.
116
:
e54727
.
24
Bettelli
,
E.
,
M.
Pagany
,
H. L.
Weiner
,
C.
Linington
,
R. A.
Sobel
,
V. K.
Kuchroo
.
2003
.
Myelin oligodendrocyte glycoprotein-specific T cell receptor transgenic mice develop spontaneous autoimmune optic neuritis.
J. Exp. Med.
197
:
1073
1081
.
25
Litzenburger
,
T.
,
R.
Fässler
,
J.
Bauer
,
H.
Lassmann
,
C.
Linington
,
H.
Wekerle
,
A.
Iglesias
.
1998
.
B lymphocytes producing demyelinating autoantibodies: development and function in gene-targeted transgenic mice.
J. Exp. Med.
188
:
169
180
.
26
Maruyama
,
M.
,
K. P.
Lam
,
K.
Rajewsky
.
2000
.
Memory B-cell persistence is independent of persisting immunizing antigen [Published erratum appears in 2001 Nature 409: 382.].
Nature
407
:
636
642
.
27
Chen
,
J.
,
M.
Trounstine
,
F. W.
Alt
,
F.
Young
,
C.
Kurahara
,
J. F.
Loring
,
D.
Huszar
.
1993
.
Immunoglobulin gene rearrangement in B cell deficient mice generated by targeted deletion of the JH locus.
Int. Immunol.
5
:
647
656
.
28
Kerfoot
,
S. M.
,
G.
Yaari
,
J. R.
Patel
,
K. L.
Johnson
,
D. G.
Gonzalez
,
S. H.
Kleinstein
,
A. M.
Haberman
.
2011
.
Germinal center B cell and T follicular helper cell development initiates in the interfollicular zone.
Immunity
34
:
947
960
.
29
Hickey
,
W. F.
2001
.
Basic principles of immunological surveillance of the normal central nervous system.
Glia
36
:
118
124
.
30
Bartholomäus
,
I.
,
N.
Kawakami
,
F.
Odoardi
,
C.
Schläger
,
D.
Miljkovic
,
J. W.
Ellwart
,
W. E. F.
Klinkert
,
C.
Flügel-Koch
,
T. B.
Issekutz
,
H.
Wekerle
,
A.
Flügel
.
2009
.
Effector T cell interactions with meningeal vascular structures in nascent autoimmune CNS lesions.
Nature
462
:
94
98
.
31
Haugen
,
M.
,
J. L.
Frederiksen
,
M.
Degn
.
2014
.
B cell follicle-like structures in multiple sclerosis-with focus on the role of B cell activating factor.
J. Neuroimmunol.
273
:
1
7
.
32
Prineas
,
J. W.
,
R. G.
Wright
.
1978
.
Macrophages, lymphocytes, and plasma cells in the perivascular compartment in chronic multiple sclerosis.
Lab. Invest.
38
:
409
421
.
33
Lovato
,
L.
,
S. N.
Willis
,
S. J.
Rodig
,
T.
Caron
,
S. E.
Almendinger
,
O. W.
Howell
,
R.
Reynolds
,
K. C.
O’Connor
,
D. A.
Hafler
.
2011
.
Related B cell clones populate the meninges and parenchyma of patients with multiple sclerosis.
Brain
134
:
534
541
.
34
Willis
,
S. N.
,
P.
Stathopoulos
,
A.
Chastre
,
S. D.
Compton
,
D. A.
Hafler
,
K. C.
O’Connor
.
2015
.
Investigating the antigen specificity of multiple sclerosis central nervous system-derived immunoglobulins.
Front. Immunol.
6
:
600
.
35
Tarlinton
,
D.
,
K.
Good-Jacobson
.
2013
.
Diversity among memory B cells: origin, consequences, and utility.
Science
341
:
1205
1211
.
36
Serafini
,
B.
,
M.
Severa
,
S.
Columba-Cabezas
,
B.
Rosicarelli
,
C.
Veroni
,
G.
Chiappetta
,
R.
Magliozzi
,
R.
Reynolds
,
E. M.
Coccia
,
F.
Aloisi
.
2010
.
Epstein-Barr virus latent infection and BAFF expression in B cells in the multiple sclerosis brain: implications for viral persistence and intrathecal B-cell activation.
J. Neuropathol. Exp. Neurol.
69
:
677
693
.
37
Henderson
,
A. P. D.
,
M. H.
Barnett
,
J. D. E.
Parratt
,
J. W.
Prineas
.
2009
.
Multiple sclerosis: distribution of inflammatory cells in newly forming lesions.
Ann. Neurol.
66
:
739
753
.
38
Weisel
,
F.
,
M.
Shlomchik
.
2017
.
Memory B cells of mice and humans.
Annu. Rev. Immunol.
35
:
255
284
.
39
Magliozzi
,
R.
,
B.
Serafini
,
B.
Rosicarelli
,
G.
Chiappetta
,
C.
Veroni
,
R.
Reynolds
,
F.
Aloisi
.
2013
.
B-cell enrichment and Epstein-Barr virus infection in inflammatory cortical lesions in secondary progressive multiple sclerosis.
J. Neuropathol. Exp. Neurol.
72
:
29
41
.
40
Meinl
,
E.
,
M.
Krumbholz
,
R.
Hohlfeld
.
2006
.
B lineage cells in the inflammatory central nervous system environment: migration, maintenance, local antibody production, and therapeutic modulation.
Ann. Neurol.
59
:
880
892
.
41
Bettelli
,
E.
,
D.
Baeten
,
A.
Jäger
,
R. A.
Sobel
,
V. K.
Kuchroo
.
2006
.
Myelin oligodendrocyte glycoprotein-specific T and B cells cooperate to induce a devic-like disease in mice.
J. Clin. Invest.
116
:
2393
2402
.
42
Krishnamoorthy
,
G.
,
H.
Lassmann
,
H.
Wekerle
,
A.
Holz
.
2006
.
Spontaneous opticospinal encephalomyelitis in a double-transgenic mouse model of autoimmune T cell/B cell cooperation.
J. Clin. Invest.
116
:
2385
2392
.
43
Oliver
,
A. R.
,
G. M.
Lyon
,
N. H.
Ruddle
.
2003
.
Rat and human myelin oligodendrocyte glycoproteins induce experimental autoimmune encephalomyelitis by different mechanisms in C57BL/6 mice.
J. Immunol.
171
:
462
468
.

The authors have no financial conflicts of interest.