Age-Associated B Cells with Proinflammatory Characteristics Are Expanded in a Proportion of Multiple Sclerosis Patients

Nele Claes, Judith Fraussen, Marjan Vanheusden, Niels Hellings, Piet Stinissen, Bart Van Wijmeersch, Raymond Hupperts and Veerle Somers
J Immunol December 15, 2016, 197 (12) 4576-4583; DOI: https://doi.org/10.4049/jimmunol.1502448

Abstract

Immune aging occurs in the elderly and in autoimmune diseases. Recently, IgD−CD27− (double negative, DN) and CD21−CD11c+ (CD21low) B cells were described as age-associated B cells with proinflammatory characteristics. This study investigated the prevalence and functional characteristics of DN and CD21low B cells in multiple sclerosis (MS) patients. Using flow cytometry, we demonstrated a higher proportion of MS patients younger than 60 y with peripheral expansions of DN (8/41) and CD21low (9/41) B cells compared with age-matched healthy donors (1/33 and 2/33, respectively), which indicates an increase in age-associated B cells in MS patients. The majority of DN B cells had an IgG+ memory phenotype, whereas CD21low B cells consisted of a mixed population of CD27− naive, CD27+ memory, IgG+, and IgM+ cells. DN B cells showed similar (MS patients) or increased (healthy donors) MHC-II expression as class-switched memory B cells and intermediate costimulatory molecule expression between naive and class-switched memory B cells, indicating their potential to induce (proinflammatory) T cell responses. Further, DN B cells produced proinflammatory and cytotoxic cytokines following ex vivo stimulation. Increased frequencies of DN and CD21low B cells were found in the cerebrospinal fluid of MS patients compared with paired peripheral blood. In conclusion, a proportion of MS patients showed increased peripheral expansions of age-associated B cells. DN and CD21low B cell frequencies were further increased in MS cerebrospinal fluid. These cells could contribute to inflammation by induction of T cell responses and the production of proinflammatory cytokines.

Introduction

Multiple sclerosis (MS) is an inflammatory disease of the CNS that is characterized by myelin loss and neurodegeneration, leading to increased disability. B cells are implicated in MS pathogenesis by autoantibody production, Ag presentation, costimulation and cytokine production such as lymphotoxin-α (LT-α) and TNF-α (1–3). Premature immune aging is seen in a proportion of MS patients and is thought to contribute to MS pathogenesis and progression (4). During aging, the immune system undergoes changes that lead to immune dysregulation or remodeling. Bacterial and viral infections increase whereas vaccine efficacy declines. Further, modifications of immune cell repertoires and functions are observed.

Humoral immunity is affected by aging, as a decline in the production of B cells leads to reduced protective Ab responses and increased autoantibody generation (5). BCR mutations accumulate during aging and lead to an alteration in the selection of memory B cells. Moreover, BCR diversity is decreased whereas clonal B cell expansion occurs, contributing to a restricted B cell repertoire (6, 7). Different populations of B cells have already been described in relation to immune aging, namely double negative (DN) IgD−CD27− B cells and CD21−CD11c+ (CD21low) B cells (8).

Increased frequencies of DN B cells were previously indicated in aged healthy donors and systemic lupus erythematosus (SLE) patients, where they were associated with increased disease-specific autoantibodies (9–11). Although DN B cells do not express the memory marker CD27, the majority of the cells were IgG+ and displayed significant BCR hypermutation, which suggests their selection in an Ag-driven process (11). DN B cells are thought to be senescent memory B cells that downregulate CD27 following chronic Ag stimulation. Alternatively, it is hypothesized that DN B cells develop independently of T cell help outside the germinal center (12). Further, DN B cells might initiate a germinal center reaction but fail to progress to CD27+ memory B cells.

CD21low B cells have been described in aged mice and in patients with SLE, common variable immunodeficiency disease (CVID), HIV and rheumatoid arthritis (RA) (13–16). The majority of CD21low B cells in healthy donors exhibited a memory phenotype whereas CD21low B cells were mainly naive in RA and CVID (13). CD21low B cells displayed polyreactivity and autoreactivity with up to 22% of CD21low B cells producing antinuclear Abs (13). CD21low B cells potentially originate from downregulation of the EBV receptor CD21 as a consequence of chronic EBV stimulation. Alternatively, CD21low B cells could develop after activation via TLR-7 (16).

The involvement of DN and CD21low B cells in MS remains unclear. Recently, Ig class-switched DN B cells in the peripheral blood of MS patients were found to be clonally related to intrathecal Ig repertoires, pointing to the involvement of DN B cells in MS pathology (17). In this study, we phenotypically analyzed DN and CD21low B cells in the peripheral blood and cerebrospinal fluid (CSF) of MS patients and healthy donors and functionally studied the cells ex vivo. Further characterization of these cells provides better insights into MS pathogenesis and can contribute to finding new therapeutic possibilities or screening tools related to age-associated B cells.

Materials and Methods

Study subjects

This study was approved by the institute’s Medical Ethical Committee and informed consent was obtained from all participants. Peripheral blood was collected from 85 healthy donors and 64 MS patients for flow cytometry. MS patients were diagnosed according to the McDonald criteria (18). Cytokine production was studied using 11 out of 85 healthy donors and 13 out of 64 MS patients. Clinical data are provided in Table I. Paired blood and CSF samples were obtained from six patients undergoing lumbar puncture for diagnostic purposes. Costimulatory and HLA molecule expression was measured on B cells of 31 healthy donors and 47 untreated MS patients.

View this table:
Table I. Characteristics of MS patients and healthy donors

Cell isolation

PBMC were isolated by Ficoll density gradient centrifugation (Lympholyte; Cedarlane Laboratories, SanBio B.V., Uden, the Netherlands) and used for flow cytometric analysis or B cell isolation. CD19+ B cells were isolated from PBMC using negative magnetic selection (STEMCELL Technologies SARL, Grenoble, France) according to the manufacturer’s instructions. Purity of the isolated B cells was confirmed by flow cytometry.

CSF cells were collected by centrifugation of the CSF for 12 min at 250 × g at 4°C and immediately used for flow cytometry. CSF samples contaminated with RBCs were excluded from analyses.

Flow cytometry

Age-associated B cells were analyzed using anti-human CD19 PE-Cy7, CD27 allophycocyanin, IgD PE-CF594, IgM PerCP-Cy5.5, IgG FITC, CD21 PE (all from BD Biosciences, Erembodegem, Belgium), and CD11c allophycocyanin-Cy7 (BioLegend; ImTec Diagnostics, Antwerp, Belgium). Isotype controls were used for gating. All flow cytometry was performed on a FACSAria II flow cytometer and FACSDiva (BD Biosciences) or FlowJo software. The cut-off to identify donors with expanded DN or CD21low B cells was the mean percentage DN or CD21low B cells from healthy donors younger than 60 y plus two times the SD (7 and 3%, respectively). Expression of costimulatory and HLA molecules on DN B cells was assessed using anti-human CD19 PerCP-Cy5.5, IgD allophycocyanin-Cy7, CD27 PE-Cy7, HLA-DR/DP/DQ FITC, CD80 PE, and CD86 PE-CF594 (BD Biosciences).

Analysis of cytokine production

B cells were seeded in 48-well, flat-bottom plates (Nunc, Roskilde, Denmark) at 375 × 103 cells in RPMI 1640 (Lonza, Basel, Switzerland) with 10% FCS (Life Technologies), 1% nonessential amino acids, 1% sodium pyruvate (Sigma-Aldrich), 50 U/ml penicillin and 50 μg/ml streptomycin (Invitrogen, Carlsbad, CA).

CD40L stimulation (referred to as CD40L) was done by culturing B cells 15:1 on preadhered irradiated CD40L cells (a gift from Prof. Dr. B. ‘t Hart; University Medical Center Groningen, the Netherlands) together with IL-4 (10 ng/ml; R&D Systems, Minneapolis, MN). For triple B cell stimulation, B cells were stimulated for 24 h with anti-human IgM/IgG (1 μg/ml; Jackson ImmunoResearch Europe, Suffolk, U.K.) followed by CD40L and CpG oligonucleotide (1 μg/ml, ODN 2006; Invivogen, Toulouse, France) stimulation for 48 h. Granzyme-B production was measured after IL-21 stimulation (50 ng/ml; Life Technologies) and anti-IgM/IgG (6.5 μg/ml) for 72 h in AIM-V medium (Life Technologies).

For intracellular flow cytometry, cells were restimulated with PMA (25 ng/ml) and calcium ionomycin (1 μg/ml; both Sigma-Aldrich) with GolgiPlug (BD Biosciences) for 4 h. Fixation and permeabilization was done using Cytofix/Cytoperm (BD Biosciences). The Abs used were anti-human CD19 PE-CF594, CD27 allophycocyanin, IgD allophycocyanin-Cy7, IL-10 PE-Cy7, TNF-α PerCP-Cy5.5, LT-α PE, and granzyme-B FITC (BD Biosciences). Appropriate isotype controls were used for gating. To control for B cell activation, anti-human CD19 PE-Cy7, IgD PE-CF594, CD27 allophycocyanin, CD86 FITC, and CD25 PerCP-Cy5.5 (BD Biosciences) were used.

Statistical analyses

Statistical analyses were performed using GraphPad Prism version 6.01. For analysis of non-paired data, multiple groups were compared by Kruskal–Wallis with Tukey post hoc testing whereas a Mann–Whitney U test was used for comparing the two groups. Nonparametric Spearman correlation was used to assess correlations. An χ2 test was used for differences in proportions. When analyzing paired data, a Friedman test with Dunn post hoc testing was performed for multiple comparisons and Wilcoxon matched-pairs signed rank test was performed for comparison of the two groups. A p value <0.05 was considered significant.

Results

DN and CD21low B cells are increased in aged individuals

DN and CD21low B cells were described as age-associated B cell subtypes, as an increased frequency of these B cells was found in elderly healthy donors or aged mice (9, 12, 13, 16). To confirm this finding in our healthy donor cohort (Table I), we determined the frequency of DN and CD21low B cells in the peripheral blood of healthy donors younger (n = 33) and older (n = 52) than 60 y (Fig. 1A). In healthy donors older than 60 y, the percentage of DN (4.9 ± 0.34%) and CD21low B cells (2.24 ± 0.22%) in peripheral blood was significantly increased compared with healthy donors younger than 60 y (3.6 ± 0.28%, p = 0.014 and 1.45 ± 0.15%, p = 0.027, respectively; Fig. 1A, 1B). Further, in healthy donors, the percentage of DN and CD21low B cells was positively correlated with age (p = 0.0216 and p = 0.0301, respectively; Fig. 1C). A positive correlation between the percentage of DN B cells and CD21low B cells was also observed (p < 0.0001; Fig. 1D), suggesting that both populations expanded following a similar trigger during aging, possibly the proinflammatory milieu. These results indicate that DN and CD21low B cells are associated with aging in our healthy donor cohort and that these B cells could already be found in increased frequencies from the age of 60 y.

FIGURE 1.
FIGURE 1.

DN and CD21low B cells are increased in healthy donors older than 60 y. (A) Representative gating strategy to identify DN and CD21low B cells in the peripheral blood of healthy donors. DN and CD21low B cells were identified as depicted within CD19+ B cells. (B) Percentage of DN B cells and CD21low B cells in healthy donors older than 60 y (>60 y; n = 52) compared with healthy donors younger than 60 y (<60 y; n = 33). (C) Correlation between the percentage of DN/CD21low B cells and age in healthy donors (n = 85). (D) Correlation between the percentage of DN B cells and CD21low B cells. Mean levels + SEM are shown. Mann–Whitney U test and the Spearman rank correlation coefficient were used for statistics. *p < 0.05.

Increased expansion of DN and CD21low B cells in a proportion of MS patients

As DN and CD21low B cells were described to be expanded in (auto)immune diseases, we aimed to investigate whether these cells were expanded in MS patients. The proportion of MS patients younger than 60 y who presented with increased peripheral blood frequencies (>7% of CD19+ B cells) of DN B cells (8/41; 20%) was significantly elevated compared with age-matched healthy donors (1/33; 3%, p = 0.031; Fig. 2A, 2B). The same trend was observed for CD21low B cells that were expanded (>3% of CD19+ B cells) in the peripheral blood of 9/41 (22%) MS patients younger than 60 y compared with 2/33 (6%) healthy donors of the same age (p = 0.056; Fig. 2A, 2C). Interestingly, in contrast to healthy donors, no correlation was present between the percentage of DN or CD21low B cells and age in the MS cohort (p = 0.63 and p = 0.14; Fig. 2D). These observations indicate increased expansions of age-associated B cells in a proportion of MS patients.

FIGURE 2.
FIGURE 2.

An increased proportion of MS patients younger than 60 y present with expansions of DN and CD21low B cells. (A) Representative plots of MS patients younger than 60 y (<60 y) indicating the gating strategy. (B and C) Percentage of DN (B) and CD21low (C) B cells in healthy donors younger (n = 33) and older (n = 52) than 60 y and MS patients younger (n = 41) and older (n = 23) than 60 y. Black dotted line represents the cut off. (D) Correlation between the percentage of DN or CD21low B cells and age in MS patients (n = 64). (E) Correlation between the percentage of DN B cells and CD21low B cells in MS patients. Statistics were done by χ2 test and Spearman rank correlation coefficient. *p < 0.05, **p < 0.01, §p = 0.05, £p = 0.056.

Similar frequencies of age-associated B cells were measured in treated and untreated MS patients or in MS patients treated with different immunomodulatory treatments (Supplemental Fig. 1A, 1B). Longitudinal treatment effects can, however, not be excluded from these data. Although there was a gender bias in the MS patients, similar DN and CD21low B cell frequencies were observed in male and female patients as well as in different clinical MS subtypes (Supplemental Fig. 1C). The proportion of healthy donors older than 60 y with expansions of DN B cells (9/52; 17%) or CD21low B cells (15/52; 29%) was significantly increased compared with younger healthy donors (p = 0.046 and p = 0.01, respectively; Fig. 2B, 2C), which again confirmed the association of DN and CD21low B cells with aging. Although not significant, in the age group older than 60 y a decreased proportion of MS patients with expanded CD21low B cells was found compared with healthy donors (p = 0.05; Fig. 2C). This was probably due to the lower number of included MS patients older than 60 y.

The positive correlation between the percentage of DN and CD21low B cells in MS patients (p < 0.0001; Fig. 2E) suggests that both cell types can expand in a proportion of MS patients because of a similar trigger, possibly the proinflammatory milieu. In addition, percentages of DN and CD21low B cells were positively correlated with percentages of CD4+CD28− T cells in 72 included healthy donors and MS patients (Supplemental Fig. 2). CD4+CD28− T cells were previously linked to T cell aging and MS pathology (19), suggesting that both DN and CD21low B cells and CD4+CD28− T cells could expand in MS patients because of the autoreactive proinflammatory milieu.

Together, these data indicate increased expansion of age-associated B cells in a proportion of MS patients, which was most pronounced for DN B cells.

DN B cells exhibit a memory profile, whereas CD21low B cells include naive and memory cells

Because there is no agreement on the phenotype of CD21low B cells (13) and CD21low and DN B cells have not been phenotypically characterized in MS patients, we determined the memory or naive status of DN and CD21low B cells (Fig. 3A). Although DN B cells do not express the memory marker CD27, ∼45% of DN B cells were IgG+, whereas only 10% were IgM+ both in healthy donors and MS patients (mean values, Fig. 3B). These results pointed to a memory phenotype for DN B cells. Similar mean frequencies of IgG+ (32%) and IgM+ (31%) CD21low B cells were present in the peripheral blood of healthy donors and MS patients, indicating a mixed CD21low B cell phenotype between donors (Fig. 3C). DN and CD21low B cells that were negative for both IgG and IgM probably belonged to the IgA or IgE isotype. Further, CD21low B cells consisted of ∼60% CD27− naive B cells and ∼40% CD27+ memory B cells (Fig. 3D).

FIGURE 3.
FIGURE 3.

DN B cells exhibit a memory B cell profile whereas CD21low B cells include both naive and memory B cells. (A) Gating strategy to identify IgM+ and IgG+ cells based on isotype controls (dashed lines). (B and C) Percentage of IgG+ and IgM+ DN B cells (B) and CD21low B cells (C) in the peripheral blood of healthy donors (n = 85) and MS patients (n = 64). (D) Percentage of CD27+ and CD27− CD21low B cells of healthy donors and MS patients. (E) CD21 expression (MFI) on DN B cells is depicted for healthy donors and MS patients. Mean + SEM is depicted. One-way ANOVA (Kruskal–Wallis) with Tukey’s post-test and Mann–Whitney U test were used for statistics. ****p < 0.0001.

As 20% of DN B cells could be retraced to the CD21low B cell gate, we analyzed CD21 expression on DN B cells. A significantly lower CD21 expression (mean fluorescence intensity, MFI) was observed on DN B cells of MS patients compared with healthy donors (p = 0.0002; Fig. 3E). Notably, decreased CD21 expression was also observed on total B cells, naive B cells, non–class-switched memory B cells and class-switched memory (CSM) B cells (p < 0.0001 for total B cells and all B cell subtypes) of MS patients (data not shown). As low CD21 expression on B cells has previously been associated with autoreactivity, these data could suggest the increased autoreactive nature of DN B cells and B cells in MS (13).

The above results indicate that DN B cells are memory B cells with low CD21 expression, whereas CD21low B cells comprise both naive and memory cells.

DN B cells have a role in Ag presentation, costimulation and proinflammatory functions

Next, functional properties of DN B cells that could contribute to MS pathology were analyzed and compared with CSM B cells. CSM B cells are highly implicated in MS pathology via cytokine production (20) and show a close relationship with DN B cells in clonal analysis (21), whereas naive B cells are involved in primary immune responses and are thus less important in MS pathogenesis. In addition, frequencies of DN B cells in the CSF of MS patients were determined.

Ag presentation and costimulation.

Expression of Ag presentation (HLA-DR/DP/DQ) and costimulatory (CD80, CD86) molecules was measured on DN, naive and CSM B cells in the peripheral blood of untreated MS patients (n = 47, 48 ± 13.3 y old, 70% female) and healthy donors (n = 31, 31 ± 12.5 y old, 61% female). DN B cells showed similar (MS patients) or increased (healthy donors) expression of HLA-DR/DP/DQ as CSM B cells and intermediate CD80 and CD86 expression between naive and CSM B cells in healthy donors and MS patients (Fig. 4A). These results provide evidence for the potential of DN B cells to stimulate (proinflammatory) T cell responses.

FIGURE 4.
FIGURE 4.

DN B cells have functional properties that could contribute to MS pathology. (A) Expression levels (MFI) of HLA-DR/DP/DQ, CD80 and CD86 are shown for naive, DN and CSM B cells from healthy donors (n = 31) and MS patients (n = 47). (B) The percentage of LT-α and TNF-α positive total, DN and CSM B cells after B cell CD40L or triple stimulation is depicted for healthy donors (n = 10) and MS patients (n = 13). Granzyme B was measured after B cell stimulation with IL-21 + anti-human IgM/IgG. (C) Percentages of CD86+ cells within total B cells and DN B cells are depicted following CD40L and triple stimulation for healthy donors (n = 10) and MS patients (n = 13). (D) The percentage of DN and CD21low B cells is shown for paired PBMC and CSF cells from six MS patients. Friedman test with Dunn’s post hoc testing, Kruskal–Wallis test of one-way ANOVA or Wilcoxon matched-pairs signed rank test were used. Mean + SEM is depicted (A–C). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Proinflammatory cytokine production.

Next, as DN B cells showed the ability to produce proinflammatory TNF-α and cytotoxic granzyme-B in healthy donors (22, 23), we measured the frequency of LT-α, TNF-α, IL-10, and granzyme B positive DN B cells of MS patients and healthy donors in comparison with CSM B cells. Hereby, we determined whether DN B cells have a proinflammatory cytokine profile similar to CSM B cells, which could contribute to MS pathology. Cytokine producing DN B cells were analyzed from healthy donors (n = 11) and MS patients (n = 13) following CD40L and triple (CD40L + BCR crosslinking + CpG) stimulation (LT-α, TNF-α, IL-10) or IL-21 and BCR crosslinking (granzyme B). Compared with CSM B cells, the percentage of TNF-α+ DN B cells was increased after CD40L stimulation in MS patients (p = 0.0006) and equal after triple stimulation in both healthy donors and MS patients (Fig. 4B). Further, although the percentage of LT-α+ cells was lower in DN B cells compared with total B cells in MS patients after triple activation, similar frequencies of LT-α+ cells were demonstrated for DN B cells and CSM B cells after CD40L and triple stimulation. The percentage of granzyme-B positive cells was also increased in DN B cells compared with CSM B cells, which was significant in healthy donors (p = 0.036, Fig. 4B). IL-10 production was absent in our culture system, both after CD40L and triple activation (data not shown). The limited sample size could account for the lack of significant differences in TNF-α production for healthy donors and granzyme-B production for MS patients.

DN B cells appeared to be less responsive to stimulation than total B cells as CD86+ cell percentages were significantly decreased for DN B cells compared with total B cells, both following triple and CD40L stimulation (Fig. 4C). Interestingly, after CD40L stimulation, the percentage of CD86+ DN B cells was significantly higher in MS patients (53.8%) than healthy donors (33.4%, p = 0.0095), indicating an increased ability of DN B cells to become activated in MS.

Thus, DN B cells presented with a proinflammatory cytokine profile that was comparable to or even more pronounced than that of CSM B cells. DN B cells further showed distinct activation mechanisms or thresholds in MS patients and healthy donors.

DN B cell expansion in the CSF of MS patients.

We finally determined the frequency of DN and CD21low B cells in paired PBMC and CSF from a small cohort of six MS patients (45 ± 6.26 y old, 33% female) to analyze the involvement of age-associated B cells in MS pathology. Increased percentages of both DN B cells (p = 0.0313) and CD21low B cells (p = 0.0156) were present in the CSF compared with the peripheral blood (Fig. 4D).

Together, these results suggest that DN B cells play a role in Ag presentation and costimulation and produce proinflammatory cytokines and granzyme B. In MS patients, DN B cells showed a higher potential to become activated and were present in high frequencies in the CSF where they could contribute to MS pathology.

Discussion

Immune aging was observed in healthy aged individuals, and characteristics of an aged immune system were observed in a proportion of patients with different (auto)immune diseases. Two main B cell subtypes related to an aged immune system are IgD−CD27− (DN) B cells and CD21low B cells. In this study, we confirmed that peripheral blood DN B cells and CD21low B cells were increased in aged healthy donors and demonstrated that an increased proportion of MS patients younger than 60 y showed peripheral expansion of these cells, indicating increased accumulation of age-associated B cells in these MS patients. DN B cells demonstrated similar expression of Ag presentation molecules as CSM B cells and costimulatory molecule expression intermediate between naive and CSM B cells. Further, DN B cells showed a proinflammatory cytokine profile that could contribute to MS pathology.

The decline of total B cell numbers, a restricted B cell repertoire, and autoantibody generation in the elderly has indicated B cell aging (7). An increased percentage of DN and CD21low B cells was previously described in the elderly (9, 12, 16). We confirmed this finding in our healthy donor population and showed a positive correlation between the frequency of these age-associated B cells and age. Of note, DN B cell percentages were also significantly increased in healthy donors older than 75 y, which is an established age category for investigating immune aging (data not shown) (9, 24). The expansion of age-associated B cells in the elderly is potentially related to the typical inflammatory microenvironment, characterized by a general increase in proinflammatory cytokines and other inflammatory mediators, called inflamm-aging (25, 26).

The presence of age-associated B cells, including DN and CD21low B cells, has further been described in (auto)immune diseases like SLE, CVID and RA (10, 11, 14–16). To our knowledge, we have demonstrated for the first time that an increased proportion of MS patients younger than 60 y presented with an expanded population of DN or CD21low B cells when compared with age-matched healthy donors. A possible cause could be the repeated Ag challenges that result in exhaustion of the humoral immune response. Additionally, as observed with aged healthy individuals, the proinflammatory milieu caused by chronic inflammation in MS patients could contribute to the development of DN and CD21low B cells. The exact mechanisms leading to expansion of age-associated B cells remain to be investigated.

Phenotypical characterization indicated that DN B cells in our cohort were mainly IgG+, indicating a memory phenotype. A similar DN B cell phenotype was previously described in aged healthy donors and in SLE, where they were associated with higher disease activity and autoantibody titers (9, 11, 22). In the CD21low B cell population, similar percentages of IgG+ and IgM+ cells and a mixed population of CD27− and CD27+ cells were present. This is in agreement with observations in SLE (14) but contrasts the IgD−IgM−IgG+CD27high phenotype of peripheral CD21−CD11c+ B cells from aged humans (16). A recent publication indicated the memory phenotype of CD21low CD23− B cells in the peripheral blood of healthy donors (27). Conclusive evidence of the maturation status of DN and CD21low B cells can be obtained by deep sequencing of Ig BCR genes or proliferation analysis using κ-deleting recombination excision circles, if sufficient cell numbers are available. Further, we observed a decreased CD21 expression on DN B cells of MS patients compared with healthy donors, which could suggest the autoreactive nature of DN B cells as CD21low B cells have been shown to contain mostly autoreactive unresponsive clones (13). Additional research is necessary to directly link decreased DN B cell CD21 expression with autoreactivity.

Next, expression of Ag presentation and costimulatory molecules and cytokine production of DN B cells was analyzed to determine how DN B cells could contribute to MS pathology. The finding of similar HLA-DR/DP/DQ expression levels as CSM B cells and CD80/CD86 expression intermediate between naive and CSM B cells pointed to the potential of DN B cells to induce (proinflammatory) T cell responses. CD21low B cells also expressed high levels of CD80, CD86, and MHC-II, underlining their potential to stimulate T cells (14, 16). B cell induced autoreactive T cell responses have already been demonstrated in MS and experimental autoimmune encephalomyelitis (28, 29). Further, DN B cells produced LT-α and TNF-α after CD40L or triple stimulation. In comparison with CSM B cells, the percentage of TNF-α+ and LT-α+ DN B cells was equal or even increased, pointing to their proinflammatory nature by which DN B cells can contribute to MS pathology. Although we did not observe differences in the percentages of LT-α and TNF-α producing DN B cells between MS patients and healthy donors, the increased percentage of CD86+ DN B cells following bystander stimulation and the decreased CD21 expression on DN B cells of MS patients indicated distinct activation and/or functional mechanisms of DN B cells in MS. Finally, granzyme-B positive (cytotoxic) DN B cells were evidenced in both MS patients and healthy donors, the latter presenting with increased percentages of granzyme-B positive DN B cells compared with CSM B cells. B cell and DN B cell granzyme-B production was described by others in the context of viral Ag recognition or BCR crosslinking and in the presence of IL-21 (23, 30). B cells producing granzyme-B also upregulated costimulatory molecules, MHC-II and cell adhesion molecules, which suggests that DN B cells could further increase their potential to stimulate T cell responses in a proinflammatory milieu with IL-21 (31). CD21low B cells from mice showed expression of genes involved in cytotoxicity such as perforin and granzyme-A (13). Although we did not measure cytokine production by CD21low B cells, we could speculate that these cells could contribute to the pathology of MS by the production of cytotoxic molecules.

To contribute to MS pathology, age-associated B cells should be able to migrate into the CNS. We showed increased frequencies of DN and CD21low B cells in the CSF of MS patients compared with peripheral blood, indicating that age-associated B cells are present in the CNS of MS patients. A recent report also showed peripheral Ig class-switched DN B cells in four out of eight MS patients that were clonally related to intrathecal Ig repertoires (17). Furthermore, DN B cells showed CCR7, CCR6 and CXCR3 expression, whereas CD21low B cells exhibited increased levels of CXCR6, suggesting their capacity to migrate to inflammatory sites (14, 23).

Whether these DN and CD21low B cells represent true exhausted cells or B cell subtypes with alternative developmental pathways needs to be further investigated. Their contribution to pathology in ∼20% of MS patients with increased age-associated B cell frequencies could be a bystander effect of the chronic immune stimulation or directly driven by pathogenic processes. Therefore, the findings in this study provide interesting keys for future study.

In conclusion, this study provided proof of the increased expansion of age-associated B cell populations in a proportion of MS patients. These age-associated B cells were increased in the CSF of MS patients and showed T cell stimulatory and proinflammatory functions that could contribute to MS pathology. DN B cells in MS patients could be discriminated from their counterparts in healthy donors by a decreased CD21 expression, which is associated with autoreactive B cells, and an increased activation potential. Research focusing on the role of DN and CD21low B cells in the development and progression of MS could have added value in understanding the etiology of the disease and in finding new therapeutic strategies.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Igna Rutten, Kim Ulenaers, Anne Bogaers, Dr. Van Woensel (Hasselt University and UBiLim, Hasselt), Anita Knevels (Revalidation and MS Center, Overpelt), Bertine Timmermans, Ingrid Mevissen, and all MS nurses (Zuyderland Medical Center, Sittard) for patient recruitment and sample collection.

Footnotes

  • This work was supported by Hasselt University and the Belgian Charcot Foundation. J.F. is a postdoctoral fellow of the Fund for Scientific Research, Flanders.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    CD21low
    CD21−CD11c+
    CSF
    cerebrospinal fluid
    CSM
    class-switched memory
    CVID
    common variable immunodeficiency disease
    DN
    double negative
    LT-α
    lymphotoxin-α
    MFI
    mean fluorescence intensity
    MS
    multiple sclerosis
    RA
    rheumatoid arthritis
    SLE
    systemic lupus erythematosus.

  • Received November 19, 2015.
  • Accepted October 15, 2016.

References

  1. ↵
    1. Bar-Or A.,
    2. L. Fawaz,
    3. B. Fan,
    4. P. J. Darlington,
    5. A. Rieger,
    6. C. Ghorayeb,
    7. P. A. Calabresi,
    8. E. Waubant,
    9. S. L. Hauser,
    10. J. Zhang,
    11. C. H. Smith
    . 2010. Abnormal B cell cytokine responses a trigger of T cell-mediated disease in MS? Ann. Neurol. 67: 452–461.
    OpenUrlCrossRefPubMed
    1. Cross A. H.,
    2. J. L. Trotter,
    3. J. Lyons
    . 2001. B cells and antibodies in CNS demyelinating disease. J. Neuroimmunol. 112: 1–14.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Fraussen J.,
    2. N. Claes,
    3. L. de Bock,
    4. V. Somers
    . 2014. Targets of the humoral autoimmune response in multiple sclerosis. Autoimmun. Rev. 13: 1126–1137.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Thewissen M.,
    2. L. Linsen,
    3. V. Somers,
    4. P. Geusens,
    5. J. Raus,
    6. P. Stinissen
    . 2005. Premature immunosenescence in rheumatoid arthritis and multiple sclerosis patients. Ann. N. Y. Acad. Sci. 1051: 255–262.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Weksler M. E.,
    2. P. Szabo
    . 2000. The effect of age on the B-cell repertoire. J. Clin. Immunol. 20: 240–249.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Chong Y.,
    2. H. Ikematsu,
    3. K. Yamaji,
    4. M. Nishimura,
    5. S. Kashiwagi,
    6. J. Hayashi
    . 2003. Age-related accumulation of Ig V(H) gene somatic mutations in peripheral B cells from aged humans. Clin. Exp. Immunol. 133: 59–66.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Gibson K. L.,
    2. Y. C. Wu,
    3. Y. Barnett,
    4. O. Duggan,
    5. R. Vaughan,
    6. E. Kondeatis,
    7. B. O. Nilsson,
    8. A. Wikby,
    9. D. Kipling,
    10. D. K. Dunn-Walters
    . 2009. B-cell diversity decreases in old age and is correlated with poor health status. Aging Cell 8: 18–25.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Montoya-Ortiz G.
    2013. Immunosenescence, aging, and systemic lupus erythematous. Autoimmune Dis. 2013: 267078.
    OpenUrl
  9. ↵
    1. Colonna-Romano G.,
    2. M. Bulati,
    3. A. Aquino,
    4. M. Pellicanò,
    5. S. Vitello,
    6. D. Lio,
    7. G. Candore,
    8. C. Caruso
    . 2009. A double-negative (IgD-CD27-) B cell population is increased in the peripheral blood of elderly people. Mech. Ageing Dev. 130: 681–690.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Sanz I.,
    2. C. Wei,
    3. F. E. Lee,
    4. J. Anolik
    . 2008. Phenotypic and functional heterogeneity of human memory B cells. Semin. Immunol. 20: 67–82.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Wei C.,
    2. J. Anolik,
    3. A. Cappione,
    4. B. Zheng,
    5. A. Pugh-Bernard,
    6. J. Brooks,
    7. E. H. Lee,
    8. E. C. Milner,
    9. I. Sanz
    . 2007. A new population of cells lacking expression of CD27 represents a notable component of the B cell memory compartment in systemic lupus erythematosus. J. Immunol. 178: 6624–6633.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Bulati M.,
    2. S. Buffa,
    3. G. Candore,
    4. C. Caruso,
    5. D. K. Dunn-Walters,
    6. M. Pellicanò,
    7. Y. C. Wu,
    8. G. Colonna Romano
    . 2011. B cells and immunosenescence: a focus on IgG+IgD-CD27- (DN) B cells in aged humans. Ageing Res. Rev. 10: 274–284.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Isnardi I.,
    2. Y. S. Ng,
    3. L. Menard,
    4. G. Meyers,
    5. D. Saadoun,
    6. I. Srdanovic,
    7. J. Samuels,
    8. J. Berman,
    9. J. H. Buckner,
    10. C. Cunningham-Rundles,
    11. E. Meffre
    . 2010. Complement receptor 2/CD21- human naive B cells contain mostly autoreactive unresponsive clones. Blood 115: 5026–5036.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Wehr C.,
    2. H. Eibel,
    3. M. Masilamani,
    4. H. Illges,
    5. M. Schlesier,
    6. H. H. Peter,
    7. K. Warnatz
    . 2004. A new CD21low B cell population in the peripheral blood of patients with SLE. Clin. Immunol. 113: 161–171.
    OpenUrlCrossRefPubMed
    1. Rakhmanov M.,
    2. B. Keller,
    3. S. Gutenberger,
    4. C. Foerster,
    5. M. Hoenig,
    6. G. Driessen,
    7. M. van der Burg,
    8. J. J. van Dongen,
    9. E. Wiech,
    10. M. Visentini,
    11. et al
    . 2009. Circulating CD21low B cells in common variable immunodeficiency resemble tissue homing, innate-like B cells. Proc. Natl. Acad. Sci. USA 106: 13451–13456.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Rubtsov A. V.,
    2. K. Rubtsova,
    3. A. Fischer,
    4. R. T. Meehan,
    5. J. Z. Gillis,
    6. J. W. Kappler,
    7. P. Marrack
    . 2011. Toll-like receptor 7 (TLR7)-driven accumulation of a novel CD11c⁺ B cell population is important for the development of autoimmunity. Blood 118: 1305–1315.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Palanichamy A.,
    2. L. Apeltsin,
    3. T. C. Kuo,
    4. M. Sirota,
    5. S. Wang,
    6. S. J. Pitts,
    7. P. D. Sundar,
    8. D. Telman,
    9. L. Z. Zhao,
    10. M. Derstine,
    11. et al
    . 2014. Immunoglobulin class-switched B cells form an active immune axis between CNS and periphery in multiple sclerosis. Sci. Transl. Med. 6: 248ra106.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Polman C. H.,
    2. S. C. Reingold,
    3. B. Banwell,
    4. M. Clanet,
    5. J. A. Cohen,
    6. M. Filippi,
    7. K. Fujihara,
    8. E. Havrdova,
    9. M. Hutchinson,
    10. L. Kappos,
    11. et al
    . 2011. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann. Neurol. 69: 292–302.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Broux B.,
    2. S. Markovic-Plese,
    3. P. Stinissen,
    4. N. Hellings
    . 2012. Pathogenic features of CD4+CD28- T cells in immune disorders. Trends Mol. Med. 18: 446–453.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Duddy M.,
    2. M. Niino,
    3. F. Adatia,
    4. S. Hebert,
    5. M. Freedman,
    6. H. Atkins,
    7. H. J. Kim,
    8. A. Bar-Or
    . 2007. Distinct effector cytokine profiles of memory and naive human B cell subsets and implication in multiple sclerosis. J. Immunol. 178: 6092–6099.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Wu Y. C.,
    2. D. Kipling,
    3. D. K. Dunn-Walters
    . 2011. The relationship between CD27 negative and positive B cell populations in human peripheral blood. Front. Immunol. 2: 81.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Buffa S.,
    2. M. Bulati,
    3. M. Pellicanò,
    4. D. K. Dunn-Walters,
    5. Y. C. Wu,
    6. G. Candore,
    7. S. Vitello,
    8. C. Caruso,
    9. G. Colonna-Romano
    . 2011. B cell immunosenescence: different features of naive and memory B cells in elderly. Biogerontology 12: 473–483.
    OpenUrlPubMed
  23. ↵
    1. Bulati M.,
    2. S. Buffa,
    3. A. Martorana,
    4. G. Candore,
    5. D. Lio,
    6. C. Caruso,
    7. G. Colonna-Romano
    . 2014. Trafficking phenotype and production of granzyme B by double negative B cells (IgG(+)IgD(-)CD27(-)) in the elderly. Exp. Gerontol. 54: 123–129.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Ligthart G. J.,
    2. J. X. Corberand,
    3. C. Fournier,
    4. P. Galanaud,
    5. W. Hijmans,
    6. B. Kennes,
    7. H. K. Müller-Hermelink,
    8. G. G. Steinmann
    . 1984. Admission criteria for immunogerontological studies in man: the SENIEUR protocol. Mech. Ageing Dev. 28: 47–55.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Franceschi C.,
    2. M. Capri,
    3. D. Monti,
    4. S. Giunta,
    5. F. Olivieri,
    6. F. Sevini,
    7. M. P. Panourgia,
    8. L. Invidia,
    9. L. Celani,
    10. M. Scurti,
    11. et al
    . 2007. Inflammaging and anti-inflammaging: a systemic perspective on aging and longevity emerged from studies in humans. Mech. Ageing Dev. 128: 92–105.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Singh T.,
    2. A. B. Newman
    . 2011. Inflammatory markers in population studies of aging. Ageing Res. Rev. 10: 319–329.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Thorarinsdottir K.,
    2. A. Camponeschi,
    3. N. Cavallini,
    4. O. Grimsholm,
    5. L. Jacobsson,
    6. I. Gjertsson,
    7. I. L. Mårtensson
    . 2016. CD21(-/low) B cells in human blood are memory cells. Clin. Exp. Immunol. 185: 252–262.
    OpenUrl
  28. ↵
    1. Molnarfi N.,
    2. U. Schulze-Topphoff,
    3. M. S. Weber,
    4. J. C. Patarroyo,
    5. T. Prod’homme,
    6. M. Varrin-Doyer,
    7. A. Shetty,
    8. C. Linington,
    9. A. J. Slavin,
    10. J. Hidalgo,
    11. et al
    . 2013. MHC class II-dependent B cell APC function is required for induction of CNS autoimmunity independent of myelin-specific antibodies. J. Exp. Med. 210: 2921–2937.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Harp C. T.,
    2. A. E. Lovett-Racke,
    3. M. K. Racke,
    4. E. M. Frohman,
    5. N. L. Monson
    . 2008. Impact of myelin-specific antigen presenting B cells on T cell activation in multiple sclerosis. Clin. Immunol. 128: 382–391.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Hagn M.,
    2. B. Jahrsdörfer
    . 2012. Why do human B cells secrete granzyme B? Insights into a novel B-cell differentiation pathway. OncoImmunology 1: 1368–1375.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Hagn M.,
    2. E. Schwesinger,
    3. V. Ebel,
    4. K. Sontheimer,
    5. J. Maier,
    6. T. Beyer,
    7. T. Syrovets,
    8. Y. Laumonnier,
    9. D. Fabricius,
    10. T. Simmet,
    11. B. Jahrsdörfer
    . 2009. Human B cells secrete granzyme B when recognizing viral antigens in the context of the acute phase cytokine IL-21. J. Immunol. 183: 1838–1845.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section