Research ArticleMultiple Sclerosis

Activated leukocyte cell adhesion molecule regulates B lymphocyte migration across central nervous system barriers

See allHide authors and affiliations

Science Translational Medicine  13 Nov 2019:
Vol. 11, Issue 518, eaaw0475
DOI: 10.1126/scitranslmed.aaw0475

Crossing barriers

Increasing evidence supports a critical role for B lymphocytes in multiple sclerosis (MS). Extravasation of B lymphocytes into the central nervous system promotes disease initiation and progression, and B lymphocyte–targeting therapies have shown potential benefits in experimental models and clinical trials. However, the mechanisms mediating B lymphocyte extravasation are unclear. Now, Michel et al. show that the adhesion molecule ALCAM promoted B lymphocyte extravasation. ALCAM-expressing B cells were increased in serum and brain of patients with MS, and the frequency of ALCAM-expressing B lymphocytes correlated with disease severity in a mouse model of autoimmune encephalomyelitis. Blocking ALCAM reduced disease severity in mice.

Abstract

The presence of B lymphocyte–associated oligoclonal immunoglobulins in the cerebrospinal fluid is a classic hallmark of multiple sclerosis (MS). The clinical efficacy of anti-CD20 therapies supports a major role for B lymphocytes in MS development. Although activated oligoclonal populations of pathogenic B lymphocytes are able to traffic between the peripheral circulation and the central nervous system (CNS) in patients with MS, molecular players involved in this migration have not yet been elucidated. In this study, we demonstrated that activated leukocyte cell adhesion molecule (ALCAM/CD166) identifies subsets of proinflammatory B lymphocytes and drives their transmigration across different CNS barriers in mouse and human. We also showcased that blocking ALCAM alleviated disease severity in animals affected by a B cell–dependent form of experimental autoimmune encephalomyelitis. Last, we determined that the proportion of ALCAM+ B lymphocytes was increased in the peripheral blood and within brain lesions of patients with MS. Our findings indicate that restricting access to the CNS by targeting ALCAM on pathogenic B lymphocytes might represent a promising strategy for the development of next-generation B lymphocyte–targeting therapies for the treatment of MS.

INTRODUCTION

Multiple sclerosis (MS) is a chronic autoimmune disorder of the central nervous system (CNS) and the leading cause of nontraumatic neurological handicap in young adults (1). Although proinflammatory and encephalitogenic T lymphocytes are classically thought to drive the development of MS, multiple lines of evidence have recently pinpointed the key contribution of also B lymphocytes in its pathogenesis. The administration of B lymphocyte–depleting therapies, in the form of anti-CD20 monoclonal antibodies (mAbs), was demonstrated to decrease the clinical and radiological activity of the disease in patients with MS affected by both relapsing-remitting (RR) and progressive MS (26). The therapeutic benefit of the B lymphocyte–depleting therapies is thought to result from the elimination of B lymphocyte subsets with pathogenic properties from circulation (79). In the peripheral blood of patients with MS, pathogenic subsets of B lymphocytes were recently identified and characterized as proinflammatory granulocyte-macrophage colony-stimulating factor (GM-CSF)–secreting (10) and tumor necrosis factor–α (TNFα)–secreting B lymphocytes (11, 12).

Beyond the clinical evidence implicating B lymphocytes in MS pathogenesis, B lymphocytes were demonstrated to account for up to 10% of CNS-infiltrating leukocytes (13). Several studies have reported the presence of ectopic B lymphocyte–enriched follicles and antibody deposits in the meninges and cortex of MS-affected brains, respectively, and correlated their presence with disease severity (1416). In parallel, several groups have indicated that B lymphocytes derived from human peripheral blood and draining cervical lymph nodes share antigen specificity with intrathecal B lymphocytes (17, 18). Collectively, these data point to the presence of a blood-CNS axis that permits B lymphocyte infiltration into the CNS in MS. The physiological routes and the molecular players involved in this process are, however, unknown.

During neuroinflammation, pathogenic leukocytes in the peripheral blood use different trafficking molecules to cross the blood-brain barrier (BBB) (1928) or the blood-meningeal barrier (BMB) (2932). We have previously demonstrated that activated leukocyte cell adhesion molecule (ALCAM)/CD166 is a cell adhesion molecule that promotes the diapedesis of monocytes and CD4+ T lymphocytes across the BBB (21, 27). Traditionally, ALCAM was known as a costimulatory molecule involved in the formation and stabilization of immune synapse between leukocytes and antigen-presenting cells (3336). It can either bind to its ligand CD6 or engage in homotypic ALCAM-ALCAM interactions (37, 38). Following up on our previous observation that ex vivo B lymphocytes express ALCAM (21), we sought to investigate its role in mediating their trafficking into the CNS in MS.

Here, we demonstrated that ALCAM is highly expressed on human B lymphocytes with memory, effector, and proinflammatory phenotypes. Using a B lymphocyte–dependent mouse model of experimental autoimmune encephalomyelitis (EAE) induced via immunization with recombinant human myelin oligodendrocyte glycoprotein (rhMOG), we demonstrated that the presence of ALCAM+ B lymphocytes in the CNS correlates with disease development. In vitro, we demonstrated that ALCAM promotes the interaction of B lymphocytes with mouse BBB endothelial cells (ECs). In vivo, we found that blocking ALCAM impairs the trafficking of B lymphocytes into the CNS, particularly through the meningeal blood vessels, and reduces EAE disease severity. Corroborating our data in mice, we showed that ALCAM mediates the migration of B lymphocytes across primary cultures of human BBB-ECs and BMB-ECs. In patients with MS, not only is the expression of ALCAM on B lymphocytes increased in the peripheral blood, but it is also up-regulated on lesion-infiltrating B lymphocytes harvested from freshly autopsied MS brain. Together, our data identified ALCAM as a valuable therapeutic target for treating B lymphocyte–dependent CNS inflammatory disorders.

RESULTS

ALCAM is primarily expressed on effector memory B lymphocytes

Having previously detected ALCAM on ex vivo B lymphocytes (21), we sought to define its expression pattern on different B lymphocyte subsets. Flow cytometry analysis of circulating B lymphocytes isolated from healthy donors revealed that ALCAM is preferentially expressed on CD19+CD24hiCD38neg and CD19+CD27+ memory B lymphocytes, while being almost absent on CD19+CD24dimCD38dim naïve and CD19+CD24highCD38high transitional B lymphocytes (Fig. 1A). Moreover, the expression of ALCAM was increased on activated CD19+ B lymphocytes defined using the activation markers CD86, CD80, and CD95, and as compared with nonactivated B lymphocytes (Fig. 1A). In corroboration with these results, flow cytometry analysis of ALCAM+ versus ALCAMneg sorted ex vivo B lymphocytes revealed that the expression of ALCAM is indeed tightly associated with memory B lymphocyte phenotypes shown as CD24hiCD38neg and CD27+, and with activated B lymphocyte phenotypes shown as CD80+ and CD95+ cells (Fig. 1B). Compared with ALCAMneg B lymphocytes, ALCAM+ B lymphocytes also expressed higher amounts of the adhesion molecules VLA-4 and LFA-1 (Fig. 1, C and D) and the proinflammatory cytokines [TNFα, GM-CSF, and interleukin-6 (IL-6)] (Fig. 1, E to H), which are all molecules implicated in MS pathogenesis (912, 39). To explore whether B lymphocyte activation affects the expression of ALCAM, we stimulated ex vivo B lymphocytes with CD40L, recombinant human IL-4, and anti–immunoglobulin M for B cell receptor (BCR)–dependent activation or CpG for Toll-like receptor (TLR)–dependent activation for 48 hours and subsequently quantified the expression of ALCAM via flow cytometry and immunofluorescence microscopy. Both techniques revealed that ALCAM expression is up-regulated upon both BCR- and TLR-dependent B lymphocyte activation (Fig. 1, I to M, and fig. S1). Together, our observations demonstrate that ALCAM defines a subset of activated memory B lymphocytes with a proinflammatory profile that could be highly relevant for neuroinflammation.

Fig. 1 ALCAM identifies populations of pathogenic human B lymphocytes.

(A to H) Characterization of ex vivo B lymphocytes from the peripheral blood of healthy donors by flow cytometry. (A) Frequencies (mean ± SEM) of ALCAM expression on total CD19+, CD24dimCD38dim (naïve), CD24highCD38high (transitional), CD24highCD38neg and CD27+ (memory), and CD86+, CD80+, and CD95+ (activated) CD19+ B lymphocytes; n ≥ 12. ***P > 0.001, ****P > 0.0001 by one-way ANOVA with Bonferroni’s post hoc test. (B) Frequencies (mean ± SEM) of the expression of the same markers as in (A) on ALCAM+ versus ALCAMneg CD19+ B lymphocytes; n ≥ 12. *P > 0.05, ****P > 0.0001 by two-way ANOVA with Bonferroni’s post hoc test. (C and D) Fluorescence intensity (mean ± SEM) of VLA-4 and LFA-1 expression in ALCAM+ versus ALCAMneg B lymphocytes; n = 7. **P > 0.01, ****P > 0.0001 by paired t test. MFI, mean fluorescence intensity. (E) Representative dot plots and (F to H) quantification graphs of the expression of intracellular TNFα, GM-CSF, and IL-6 in ALCAM+ versus ALCAMneg B lymphocytes; n = 6. *P > 0.05, **P > 0.01 by paired t test. (I to K) Analysis of ALCAM and CD80 expression on B lymphocytes from the peripheral blood of healthy donors before (day 0, ex vivo) versus after (day 2) in vitro activation by flow cytometry; n ≥ 4. *P < 0.05 by paired t test. (L and M) Representative images and frequencies (mean ± SEM) of activated B lymphocytes immunostained for CD20 (yellow), ALCAM (cyan), and nuclei (TOPRO-3; magenta) and acquired by confocal microscopy. Scale bar, 10 μm; data shown are representative of n = 3.

ALCAM on murine B lymphocytes correlates with EAE severity

In the meninges of patients with MS, the presence of B lymphocytes was demonstrated to correlate with the magnitude of neuroinflammation and with the clinical prognosis of the patient (15, 40, 41). To determine whether the expression of ALCAM on B lymphocytes was also associated with these parameters, we resorted to a B lymphocyte–dependent EAE model of MS, induced via immunization with rhMOG (42). Specifically, we used flow cytometry to analyze B lymphocytes isolated from secondary lymphoid organs and CNS tissue, before and after immunization at time points representing the different stages of B lymphocyte–dependent rhMOG-EAE (Fig. 2A) and MOG35–55–EAE (fig. S2). In both the spleen and inguinal lymph nodes, the frequency of ALCAM+ CD19+ B lymphocytes was substantially up-regulated just before onset of disease, reaching over 80% at peak of disease in rhMOG-EAE and in MOG35–55–EAE and remaining elevated in the chronic phase of the disease, compared with naïve animals (Fig. 2, B and C, and fig. S2, A and B). In the CNS, ALCAM+ B lymphocytes were detected in all compartments analyzed (spinal cord, brain, and meninges) but were preferentially localized in meningeal infiltrates (fig. S3). Their frequency also closely followed the disease course, reaching over 80% at peak of disease in both EAE models (Fig. 2, D and E, and fig. S2C). In comparison, the CNS of naïve mice contained just over 3000 B lymphocytes, of which only ~20% expressed ALCAM (fig. S4). No differences were observed in their coexpression of other adhesion molecules (fig. S5). Our data demonstrate that ALCAM on B lymphocytes is tightly associated with rhMOG-EAE disease and may thus be actively contributing to the development of neuroinflammation.

Fig. 2 ALCAM on B lymphocytes correlates with rhMOG-EAE severity.

(A) EAE clinical scores (daily, mean ± SEM) of rhMOG-immunized C57BL/6 mice; arrows indicate animal euthanasia and organ harvest for subsequent B lymphocyte analysis; n = 3 pooled experiments with a total of 31 animals. (B, C, and E) Frequencies (mean ± SEM) and (D) representative dot plots of ALCAM expression on CD19+ B lymphocytes isolated from (B) spleen, (C) inguinal lymph nodes, and (D and E) CNS of nonimmunized (NI) versus immunized animals at 10 (presymptomatic), 15 (peak), and 37 (chronic) days postimmunization (dpi), as analyzed by flow cytometry. For flow cytometry, n = 5 to 7 animals per group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way ANOVA with Bonferroni’s post hoc test.

ALCAM promotes the extravasation of murine B lymphocytes across BBB endothelium

Because ALCAM was previously shown to promote the migration of monocytes and CD4+ T lymphocytes across the BBB (21, 27), we hypothesized that ALCAM could also drive the trafficking of B lymphocytes into the CNS. To address this hypothesis, we investigated whether ALCAM plays a role in the early stages of B lymphocyte extravasation by conducting in vitro dynamic flow assays using ALCAM knockout (KO) versus wild-type (WT) B lymphocytes isolated from a previously characterized C57BL/6 mouse model (43). In the absence of ALCAM, the velocity of B lymphocytes flowing over monolayers of primary cultures of BBB-ECs was greater, indicating that ALCAM is involved in the rolling of B lymphocytes on BBB endothelium (Fig. 3, A and B). Accordingly, the absence of ALCAM also resulted in a reduced number of B lymphocytes arrested on the BBB endothelium (Fig. 3C). An in-depth analysis of the interaction between ALCAM on B lymphocytes and BBB endothelium over time revealed that ALCAM participates in rolling and adhesion of B lymphocytes on BBB-ECs beginning from as early as 10 min after beginning of flow and persists for over 30 min without any noteworthy rescue by compensatory mechanisms (Fig. 3, D and E). To evaluate whether ALCAM is also involved in the diapedesis of murine B lymphocytes across BBB endothelium, we conducted in vitro migration assays where we observed that its genetic ablation on B lymphocytes reduced the number of B lymphocytes that successfully transmigrated across monolayers of primary cultures of murine WT BBB-ECs (Fig. 3F). These observations demonstrate that ALCAM on B lymphocytes partakes in their early interaction with BBB endothelium as well as in their actual transmigration.

Fig. 3 ALCAM mediates the rolling and adhesion of murine B lymphocytes on BBB endothelium.

(A to F) Interaction of B lymphocytes isolated from ALCAM knockout (KO) mice versus wild-type (WT) mice, with activated primary cultures of BBB-ECs isolated from WT mouse brain tissue, as assessed by dynamic flow assays. (A) Representative images of arrested B lymphocytes (white) on BBB-ECs after 30 min of constant flow. (B and D) Velocities (mean in μm/s ± SEM) of B lymphocytes (B) pooled over 30 min and (D) divided into 10-min intervals during constant flow. (C and E) Quantification of arrested (mean ± SEM) B lymphocytes per field of view (FOV) (C) after 30 min and (E) divided into 10-min intervals over 30 min of constant flow. *P < 0.05, ****P < 0.0001 by Student’s t test. (F) Absolute number of migrated B lymphocytes isolated from ALCAM KO versus WT mice across BBB-ECs isolated from WT mice. Scale bars, 120 μm. Data shown are representative of n = 3 independent experiments. *P < 0.05 by paired t test.

Blocking ALCAM hinders EAE progression and reduces the trafficking of B lymphocyte into the CNS

With our aforementioned data demonstrating that ALCAM is a potential contributor to the migration of B lymphocytes and to the development of neuroinflammation, we sought to define its role in vivo in B lymphocyte–dependent EAE. Analysis of CNS-infiltrating B lymphocytes in ALCAM KO versus WT mice immunized with rhMOG revealed a diminished presence of B lymphocytes in the CNS of ALCAM KO animals at the chronic phase of EAE (Fig. 4A), despite observing a similar number of total CNS-infiltrating leukocytes (fig. S6) and no abnormality in the peripheral B lymphocyte compartment of ALCAM KO versus WT animals (fig. S7). To determine whether ALCAM is a potential therapeutic target for treating neuroinflammation, we administered anti-ALCAM mAb or isotype control antibodies into mice immunized with rhMOG. Although blocking ALCAM did not affect disease onset, it impaired EAE progression from day 19 onward (Fig. 4B). Examination of CNS-infiltrating leukocytes via flow cytometry indicated that the reduced EAE severity mainly coincided with diminished B lymphocyte infiltration at the onset of the disease (days 12 to 13) (Fig. 4C). In contrast, the numbers of microglia, CD45+ cells, T cells, and CD3negB220neg cells all remained unchanged within the CNS (fig. S8) when blocking ALCAM in this B lymphocyte–dependent EAE model. Because B lymphocytes were demonstrated to preferentially accumulate within the leptomeninges in MS (15, 30, 40, 41, 44, 45), we sought to investigate whether ALCAM promotes this localized B lymphocyte infiltration in rhMOG-EAE. Intravital two-photon live imaging of the leptomeningeal vessels in the spinal cords of immunized animals revealed a >3-fold decrease in the number of infiltrating B lymphocytes at both the onset and chronic phase of the disease, respectively, upon ALCAM blockade (Fig. 4, D and E). Together, our data demonstrate that blocking ALCAM limits the extent of disability in the chronic phase and impairs the trafficking of B lymphocytes into the CNS, which occurs, at least partially, through the meningeal vascular structures.

Fig. 4 ALCAM contributes to pathogenic B lymphocyte infiltration in EAE.

(A) Frequencies (mean ± SEM) of CD19+B220+ B lymphocytes in CNS-infiltrating leukocytes in rhMOG-immunized ALCAM KO versus WT mice at 36 to 37 dpi (chronic phase); n = 10 per group. *P < 0.05 by Student’s t test. (B to E) rhMOG-EAE after the administration of anti-ALCAM antibody versus isotype control (Ctl) every other day from days 8 to 18 postimmunization. (B) EAE clinical scores (daily, mean ± SEM); gray area indicates treatment window; arrows indicate animal euthanasia and organ harvest for subsequent B lymphocyte analysis; n = 3 pooled experiments with a total of 17 to 20 animals per group. (C) Frequencies (mean ± SEM) of CD19+ B lymphocytes in CNS-infiltrating CD45+ cells at the onset and early chronic phase of the disease after treatment, as analyzed by flow cytometry; n ≥ 3 per group. (D) Quantifications (mean per mm2 ± SEM), and (E) representative images of B lymphocytes in the meninges of immunized animals at the onset and early chronic phase of the disease after treatment, via intravital two-photon live imaging; data shown are representative of n = 3 per group. (E) B lymphocytes are in magenta, and leptomeningeal vessels are in cyan. Scale bars, 100 μm. (B to D) *P < 0.05, **P < 0.01 by Wilcoxon-Mann-Whitney test.

ALCAM mediates B lymphocyte migration across human CNS barriers

To establish whether ALCAM expression on human B lymphocytes serves the same function as in mouse, we investigated the migration of B lymphocytes across monolayers of primary cultures of human BBB-ECs and BMB-ECs using a modified Boyden chamber, as previously described (23, 2527, 46). The migration of B lymphocytes was twofold greater across BMB-ECs than across the BBB-ECs, suggesting that the BMB is a preferred access point to the CNS for B lymphocyte in human (Fig. 5A). Flow cytometric analysis of B lymphocytes in the lower versus upper chambers (migrated versus nonmigrated cells, respectively) demonstrated that ALCAM is in fact preferentially up-regulated on B lymphocytes that migrated across both barriers (Fig. 5, B and C). To investigate whether ALCAM plays an active role in human B lymphocyte transmigration, we repeated the migration assays, introducing an anti-ALCAM blocking mAb versus isotype control into the Boyden chamber system and comparing it with other blocking antibodies that have a known effect on the trafficking of B lymphocytes (47). Treatment with an anti-ALCAM blocking mAb reduced the migration of B lymphocytes across BMB-ECs and BBB-ECs comparable to anti–intercellular adhesion molecule–1 (anti–ICAM-1) mAb treatment, but slightly less than anti–VLA-4 mAb (Fig. 5, D and E, and fig. S9, A and B). To explore the role of ALCAM in the early stages of human B lymphocyte extravasation across CNS barriers, we also conducted in vitro dynamic flow assays using pretreated B lymphocytes with either anti-ALCAM blocking mAb or isotype control. Blocking ALCAM reduced B lymphocyte adhesion to the BBB-EC monolayer (fig. S10). These results demonstrate that human B lymphocytes migrate across both the BBB and BMB, with a preference for the BMB, and that ALCAM participates in the trafficking across both barriers.

Fig. 5 ALCAM mediates the migration of human B lymphocytes across human CNS barriers.

(A to E) Migration of ex vivo B lymphocytes from healthy donors across primary cultures of human BBB-EC or BMB-EC, using a modified Boyden chamber model. (A) Absolute numbers (mean ± SEM) of B lymphocytes that have migrated across human BBB-EC or BMB-EC; n = 14 per group, in triplicate. *P < 0.05 by Student’s t test. (B and C) Expression of ALCAM on nonmigrated (recovered from the upper chamber) versus migrated (recovered from the lower chamber) CD19+ B lymphocytes, as analyzed by flow cytometry; n ≥ 4 per group. *P < 0.05 by paired t test. (D and E) Frequencies (mean ± SEM) of migrated B lymphocytes across BBB-ECs and BMB-ECs. B lymphocytes and ECs were preincubated for 1 hour with isotype control versus blocking antibodies against ALCAM, ICAM-1, or VLA-4; n ≥ 4 per group. **P < 0.01, ****P < 0.0001 by one-way ANOVA with Dunnett’s post hoc test.

Peripheral and brain-infiltrating B lymphocytes up-regulate ALCAM in patients with MS

B lymphocytes have previously been detected within MS lesions (13, 40, 48, 49) and are believed to actively contribute to disease development. To examine whether ALCAM+ B lymphocytes are preferentially elevated in the periphery and/or brain lesions of patients with MS, we investigated their presence in the peripheral blood and within the CNS of patients with MS. First, we compared the expression of ALCAM on ex vivo B lymphocytes isolated from the blood of patients with MS versus healthy controls. The percentage of ALCAM expression on B lymphocytes was found to be increased in untreated patients, in stable patients with relapsing-remitting MS (RRMS), and in patients with secondary progressive MS (SPMS), compared with healthy controls (Fig. 6, A and B). In patients with active relapse (AR), the amount of ALCAM on the individual B lymphocytes was also observed as up-regulated (fig. S9C). ALCAM+ B lymphocytes from the blood of untreated patients with RRMS were also coexpressing adhesion molecules VLA-4, ICAM-1, and CD11a (fig. S11). Second, we quantified and validated the presence of ALCAM+ B lymphocytes in brain lesions of patients with MS by both flow cytometry and immunohistochemical staining, respectively. For flow cytometry, we generated a protocol for isolating and analyzing immune cells infiltrating active and chronic inactive lesions (as defined by luxol fast blue hematoxylin eosin staining and oil red O staining) extracted from postmortem MS brains within hours of the patient’s death. Overall, we analyzed two cortical lesions and three white matter lesions (active and chronic inactive) from one patient with MS. Among all examined lesions, cortical lesions contained the highest number of CD19+ B lymphocytes (Fig. 6C). The expression of ALCAM on B lymphocytes was higher in all types of lesions than on B lymphocytes derived from the peripheral blood of patients with MS (dashed red line) and was particularly high in cortical lesions (Fig. 6D). Last, using confocal microscopy imaging of frozen human MS CNS samples, we confirmed the presence of ALCAM+ B lymphocytes in meningeal (Fig. 6, E and F, and fig. S12, B and C) and parenchymal infiltrates (Fig. 6, G to K, and fig. S12A), most of them being CD3 negative (fig. S12). Additional histological costaining showed coexpression of ALCAM with other adhesion molecules (VLA-4 and CD11a) on infiltrating B lymphocytes (fig. S13). To conclude, these findings support the important role of ALCAM in MS pathogenesis and indicate a role for ALCAM in meningeal infiltration of B lymphocytes and in cortical lesion formation.

Fig. 6 ALCAM is expressed by B lymphocytes in patients with MS.

(A) Representative dot plots and (B) quantification graphs of frequencies (mean ± SEM) of ALCAM expression on CD19+ B lymphocytes in the peripheral blood of healthy controls (HC), patients with MS with active relapse (AR), relapsing-remitting disease (RRMS) and secondary progressive disease (SPMS), as measured by flow cytometry; n = 7 to 25 donors per group. (C) Absolute numbers (mean per gram of tissue ± SEM) of CD19+ B lymphocytes and (D) frequencies (mean ± SEM) of ALCAM expression on CD19+ B lymphocytes recovered from cortical lesions (CL), as well as active (A) and chronic inactive (Ci) white matter lesions (WML), analyzed by flow cytometry; n = 1 to 4. *P < 0.05 by one-way ANOVA with Dunnett’s post hoc test. (E to J) Representative images of meningeal (E and F) and perivascular infiltrates (G to J), (E, G, and I) stained with luxol fast blue (blue), hematoxilyn (violet), and eosin (pink) in sections obtained from patients with MS. Black squares in (E), (G), and (I) are magnified areas shown respectively in (F), (H), and (J) regions, with immunostainings for CD20 (yellow), ALCAM (cyan), and nuclei (TOPRO-3; magenta), captured by confocal microscopy. Scale bar, 100 μm; n = 3. (K) Quantification (mean ± SEM) of CD20+ cells that are positive for ALCAM by confocal microscopy; n = 12 images from four different patients with MS.

DISCUSSION

B lymphocyte–depleting therapies provide strong evidence that B lymphocytes are involved in MS disease activity. B lymphocytes are found in the inflamed meninges of patients with MS, and this inflammation is associated with adjacent cortical demyelination and neurodegenerative changes in progressive forms of the disease and also early after diagnosis, suggesting a pathogenic role for meningeal B cells (14, 15, 30, 41, 44, 45, 50). Specific subsets of regulatory B lymphocytes were shown to be able to infiltrate the CNS to counteract neuroinflammation (5153), but their presence was reported to be reduced in patients with MS (54). The proposed mechanisms by which B lymphocytes influence MS pathogenesis and progression include antigen presentation (10, 55, 56), autoantibody production (57, 58), and/or cytokine secretion (10, 5153). However, the mechanisms, molecular players, and cellular signals required for B lymphocytes to infiltrate the CNS remain unclear (31).

In the present manuscript, we demonstrate that ALCAM expression is greater on proinflammatory memory and mature naïve B lymphocytes, is rapidly up-regulated upon both BCR- and TLR-dependent B lymphocyte activation, and is increased on B lymphocytes derived from patients with MS. With GM-CSF believed to be highly implicated in MS pathogenesis (5964) and GM-CSF+ B lymphocyte amounts reported to be increased in patients with MS (10), our current data collectively demonstrate that B cell–intrinsic ALCAM expression is likely to be involved in one or more stages of MS pathogenesis. In support of this notion, allele polymorphisms in both ALCAM and its ligand CD6 have been found associated with an increased risk of developing MS (6568). However, because our cohorts do not feature patients with other neurological and/or inflammatory diseases, we cannot determine whether the pathological regulation of ALCAM extends to other diseases as well. This will need to be addressed on a per-disease basis.

We and others have previously demonstrated that ALCAM is involved in the CNS recruitment of both monocytes and CD4+ T lymphocytes (21, 28). At the same time, we recently showed that ALCAM also plays a major role in maintaining BBB integrity (27). The dual functions of ALCAM may be attributed to its cellular localization as a cell surface adhesion molecule or due to its association with junctional proteins. Here, we focus on the function of ALCAM as an extracellular adhesion molecule and delineate its role in the migration of B lymphocytes across different CNS barriers. First, we provide evidence that human B lymphocytes migrate in greater numbers across meningeal vascular structures than across parenchymal microvessels, which might explain the relative paucity of B lymphocytes in white matter lesions and the presence of B lymphocyte aggregates in meningeal structures and in cortical lesions (14, 15). Second, we show that pharmacologically blocking ALCAM or neutralizing its expression by genetic knockout impairs the capacity of B lymphocytes to roll over, adhere to, and transmigrate across CNS ECs. Unlike this multifunctional role, ALCAM plays a rather specific role in mediating CD4+ T lymphocyte extravasation; it only participates in their transmigration across BBB endothelium but plays no role in also mediating their initial interaction with the BBB (21, 28). Last, we demonstrate that the frequency of ALCAM+ B lymphocytes within the CNS increases presymptomatically and correlates with EAE disease severity and that targeting ALCAM in vivo reduces the number of infiltrating B lymphocytes within the meningeal structures of EAE animals while also alleviating EAE symptoms in the chronic phase of the disease. The absence of beneficial effect at earlier disease time points can simply reflect our anti-ALCAM treatment window, starting presymptomatically when ALCAM+ B lymphocytes might have already begun invading the CNS. Together, our findings establish ALCAM as a major cell adhesion molecule involved in B lymphocyte extravasation into the CNS. They do not completely rule out its implication in promoting B lymphocyte trafficking under steady state for immunosurveillance but strongly demonstrate its pathological contribution to CNS inflammation. Our different in vitro and in vivo experimental approaches were carefully selected to validate the precise role of ALCAM on B lymphocyte and minimize, if not eliminate, any caveat that may stem from compensatory mechanisms in genetically KO animals or from targeting ALCAM+ cellular populations beside B lymphocytes.

On the molecular level, ALCAM was initially described as a ligand for CD6, a costimulatory molecule involved in the formation of immune synapses between T lymphocytes and antigen-presenting cells (3336). ALCAM can also form homotypic ALCAM-ALCAM interactions, although ALCAM-CD6 binding was later found to be the stronger of the two (37, 38). More recently, the list of ALCAM’s binding partners was expanded to include at least eight other molecules (27, 69, 70), suggesting that ALCAM may be more promiscuous than initially thought. The cell membrane microdomains known as lipid rafts, where we initially identified ALCAM as a cell adhesion molecule (21) and which are implicated in many physiological processes including leukocyte extravasation, are known to also harbor many important cell adhesion molecules such as E-selectin, ICAM-1, and vascular cell adhesion molecule–1 (VCAM-1) (71, 72). It is therefore very plausible that ALCAM promotes leukocyte extravasation while cooperating through direct or indirect interactions with other molecules and structures involved in cell adhesion and migration. Although we did not directly address this hypothesis in the present manuscript, our current observations showing that ALCAM+ B lymphocytes feature high coexpression of both ICAM-1 and VLA-4 indeed support it and point to the need to potentially decipher immune migration into the CNS through a wider lens that includes the collective and perhaps synergetic function of major adhesion molecules.

The central disadvantages of currently used therapeutic strategies in MS reside in their safety profile. Highly effective therapies, whether immune cell–depleting agents (rituximab, ocrelizumab, and alemtuzumab) or immune cell migration blockers (natalizumab), carry important risk of developing opportunistic infections, cancer, or autoimmune disorders (73, 74). This is probably in direct relation with their potential to cause profound immunosuppression due to their unspecific targeting of a vast array of immune cell populations. Our goal is to define a cell adhesion molecule signature profile to specifically define pathogenic immune cell subtypes and thus to specifically deplete them from the circulation or to prevent them from migrating into the target organs. Even if we did not directly investigate in this study the health consequences of targeting ALCAM+ B lymphocytes in neuroinflammation, our data provide conclusive evidence that ALCAM is a key mediator of B lymphocyte adhesion and migration across CNS vascular barriers, and more so across the BMB. We postulate that a pharmacological neutralization of ALCAM could prevent, or at least delay, cortical pathology and disease progression in humans affected with MS.

MATERIALS AND METHODS

Study design

This study investigated the role of the adhesion molecule ALCAM in B lymphocyte migration across CNS barriers. This was evaluated by flow cytometry and confocal microscopy using peripheral blood mononuclear cells (PBMCs) and brain-infiltrating immune cells from patients with MS and healthy controls. Primary cultures of human brain and meningeal ECs were used to explore in vitro the role of ALCAM in B lymphocyte migration, using >3 biological repeats and conducted in >3 independent experiments. EAE mouse models were used for in vivo experiments, using >5 animals per condition treated and scored by an investigator blinded to the treatment group and conducted in >3 independent experiments. Clinical scores were evaluated by a blinded investigator. Primary data are reported in data file S1.

Statistical analysis

Statistical analyses were performed using Prism (GraphPad Software, San Diego, CA), and results are presented as means ± SEM. Student’s t test, one-way analysis of variance (ANOVA) with Bonferroni’s or Dunnett’s post hoc test, and two-way ANOVA with Bonferroni’s post hoc test were performed when appropriate. Only P values <0.05 were considered statistically significant.

SUPPLEMENTARY MATERIALS

stm.sciencemag.org/cgi/content/full/11/518/eaaw0475/DC1

Materials and Methods

Fig. S1. Activated B lymphocytes up-regulate ALCAM upon both BCR- and TLR-dependent activation.

Fig. S2. ALCAM expression on peripheral and CNS-infiltrating B lymphocytes is increased during the course of MOG35–55–EAE.

Fig. S3. ALCAM+ B lymphocytes are preferentially localized in meningeal infiltrates in EAE.

Fig. S4. Few immune cells and B lymphocytes are isolated from the CNS of naïve WT mice.

Fig. S5. CNS-infiltrating B lymphocytes do not coexpress cell adhesion molecules in EAE.

Fig. S6. Immune cells infiltrate CNS of ALCAM KO and WT mice during the chronic phase of the disease.

Fig. S7. Absolute numbers of B lymphocytes are similar in lymphoid organs of ALCAM KO and WT mice in the presymptomatic phase.

Fig. S8. Non–B lymphocyte immune cell migration into the CNS is not affected by ALCAM blockade in B lymphocyte–dependent rhMOG EAE model.

Fig. S9. ALCAM promotes the migration of human B lymphocytes across CNS barriers and is up-regulated in relapsing patients with MS.

Fig. S10. ALCAM mediates the adhesion of human B lymphocytes on BBB endothelium.

Fig. S11. Cell adhesion molecules are coexpressed on ex vivo B lymphocytes from PBMCs of patients with MS.

Fig. S12. Most CD20+ cells in MS brain immune infiltrates are not CD3+ cells.

Fig. S13. Cell adhesion molecules are coexpressed on B lymphocytes in MS lesions.

Data file S1. Raw data (provided as separate Excel file).

References (7582)

REFERENCES AND NOTES

Acknowledgments: We thank J. A. Weiner for the gift of the ALCAM KO mice. We thank the cytometry facility and the cell imaging facility of the CRCHUM for their availability and their help in carrying out the experiments. We thank A. Michallet Roy for technical help at the animal facilities. We also thank R.-M. Rébillard for helping us with the patients’ characterization. Funding: This study was supported by an operating grant from the Research Foundation of the MS Society of Canada (MSSC) for the “Canadian B cells in MS” team (A.P., A.B.-O., and J.L.G.). A.P. holds the Senior Canada Research Chair in Multiple Sclerosis. C.G. holds a graduate scholarship from the MSSC. E.P. held a postdoctoral scholarship from the MSSC and the Canadian Institute of Health Research (CIHR). L.M. held a postdoctoral scholarship from the MSSC and the Fonds de Recherche Santé Québec (FRSQ). M.C. held a doctoral scholarship from the FRSQ and is now funded by the MSSC. T.D. held a fellowship from the FRSQ. M.-A.L. held a doctoral scholarship from the MSSC. Author contributions: L.M. and A.P. designed the study. R.M., A.B., B.L., and P.D. were involved in the collection of human samples and clinical characterization of patients. L.M., C.G., M.C., M.-A.L., S.Z., T.D., J.I.A., S.L., L.B., and E.P. performed the experiments. L.B. and S.L. performed blood collection from healthy controls and helped with PBMC isolations. L.M., C.G., and R.L. did the B lymphocyte in vitro characterization by flow cytometry and confocal microscopy. L.M., C.G., and E.P. did the in vitro activation of B lymphocytes. L.B. and E.P. took care of primary culture of human BBB-ECs and BMB-ECs from their isolation to the day of the experiments. L.M. and J.I.A. performed in vitro transmigration assays of human B lymphocytes across BBB-ECs and BMB-ECs. C.G. performed in vitro flow adhesion assays of human B lymphocytes on BBB-ECs. S.Z., T.D., E.P., and C.G. performed ALCAM expression characterization in lesions of patients with MS. S.L. was holding mouse colonies. L.M., C.G., E.P., S.L., and L.B. contributed to EAE experiments and in vivo two-photon acquisitions. L.M., C.G., S.L., and L.B. contributed to the euthanization of animals. L.M., C.G., and L.B. performed mouse cell isolations and flow cytometry acquisition and analysis. M.-A.L., M.C., and C.G. performed mouse brain EC isolation and culture for flow adhesion assays and migration assays. L.M. and M.-A.L. performed flow adhesion assays of mouse B lymphocytes on mouse ECs. C.G. performed in vitro migration assays of B lymphocytes across mouse brain ECs. J.L.G. provided isolated and purified rhMOG and provided support with experimental design and data analyses. L.M., C.G., S.Z., T.D., E.P., and J.I.A. analyzed the data. A.B.-O., J.L.G., and E.P. provided key scientific input. L.M., C.G., M.C., and A.P. interpreted the data and wrote the manuscript. A.B.-O., J.L.G., and A.P. secured funding. Competing interests: L.M. received honoraria for consulting from Biogen, Merck Serono, Roche, Novartis, Sanofi Genzyme, and Teva. J.L.G. has sponsored research agreements from Roche, Merck, and Novartis and consults for Roche and Visterra Inc. A.B.-O. and A.P. have participated as speakers in meetings sponsored by and received consulting fees and/or grant support from Actelion, Atara Biotherapeutics, Biogen Idec, Celgene/Receptos, Genentech/Roche, Teva, MAPI, Medimmune, Merck/EMD Serono, Novartis, and Sanofi Genzyme. Data and materials availability: All data associated with this study are present in the paper or the Supplementary Materials.

Stay Connected to Science Translational Medicine

Navigate This Article