B Cells as a Therapeutic Target for IFN-β in Relapsing–Remitting Multiple Sclerosis

Interferon-β-1b was the first approved therapy for relapsing–remitting multiple sclerosis (RR MS). Pivotal randomized, placebo-controlled clinical trials reported that IFN-β-1b reduced the frequency and severity of clinical relapses and decreased CNS inflammatory lesion formation in RR MS patients (1). Despite its established efficacy, IFN-β’s therapeutic mechanisms of action and the physiological role of endogenous type I IFNs in the control of the autoimmune response have not been completely elucidated. Previously proposed mechanisms of action include: inhibition of Ag presentation, T cell proliferation, modulation of cytokine production, and downregulation of adhesion molecules and metalloproteinases involved in lymphocyte migration across the blood–brain barrier (2). Studies of experimental autoimmune encephalomyelitis (EAE), an animal model of MS, have suggested that endogenous IFN-β suppresses Th17 cells, which have been linked to the pathogenesis of EAE and MS (3). A recent study has reported that active RR MS patients have a 7-fold increase in the percentage of Th17 cells compared with patients with inactive disease or healthy controls (HCs) (4). In a subset of active patients treated with IFN-β-1b who experienced suppression of their clinical disease activity, the percentage of IL-17–producing cells was decreased (4), suggesting that the number of Th17 cells correlates with the disease activity and that IFN-β-1b may selectively inhibit this T cell subset. Earlier studies have reported an increased expression of the Th17 cell-specific cytokine IL-17A in blood mononuclear cells during disease activity and in active MS lesions (5–7). We have recently reported a novel mechanism of IFN-β’s suppression of Th17 cell differentiation through the inhibition of IL-1β and IL-23 and the induction of IL-12 and IL-27 expression in dendritic cells (DC) (8). Durelli et al. (4) have proposed that IFN-β mediates apoptosis of Th17 cells because this cell subset exhibited an increased expression of IFN-type receptor chain 1 when compared with the Th1 cells.

B cells are proposed to play a dual role in the pathogenesis of MS. They contribute to the induction of the autoimmune response but also mediate the resolution of the CNS inflammatory infiltrate (9, 10). Their pathogenic function has been traditionally associated with Ab production, as supported by the presence of B cells, plasma cells, and myelin-specific IgG in active and chronic MS lesions and intrathecal production of oligoclonal IgG Abs by plasma cells (11). However, the recently introduced therapy with anti-CD20 mAb (rituximab), which rapidly depletes B cells and eliminates their pathogenicity in MS but does not affect Ab-producing CD20⁻ plasma cells, has demonstrated a strong therapeutic efficacy in RR MS (12). Although treated patients had stable plasma cell numbers and unchanged oligoclonal Ab production in the cerebrospinal fluid, they had a lower clinical relapse rate, and their brain scans exhibited significantly decreased numbers of active inflammatory lesions in comparison with placebo controls. The results imply that the most critical role of B cells in MS disease development is not linked to their autoantibody-producing capacity, but to their role in Ag presentation (13) and the regulation of T cell differentiation and effector functions (14) in the development of the autoimmune response.

B cells have long been recognized as APCs that are present in high numbers in the peripheral circulation in comparison with the more effective, but much less numerous, DCs (15). Moreover, Yan et al. (16) have demonstrated that B cells are the first to capture and present self-proteins to the autoreactive T cells, which is followed by DCs’ Ag presentation and stimulation of the same autoreactive T cells. Furthermore, B cells produce multiple cytokines, which regulate T cell differentiation (15, 17).

In addition to the B cell-mediated activation, priming, and differentiation of autoreactive T cells (13), B cells’ regulatory role is essential for the regulation of an autoimmune response (18). The regulatory function of B cells in CNS inflammatory demyelinating disease has been reported by several studies of the EAE mouse model. Wolf et al. (19) have postulated a regulatory role for B cells in EAE, because mice deficient in B cells suffered from an unusually severe and chronic form of EAE, whereas a more recent study has demonstrated that mice deficient for B cell surface marker CD19 were more susceptible to EAE (20). Duddy et al. (21) have demonstrated that B cells derived from MS patients produce lower amounts of IL-10 when compared with HCs, implicating that the deficiency in B cells’ IL-10 production may contribute to the pathogenesis of MS.

Despite the evidence for B cells’ role in the development and regulation of the autoimmune inflammatory response in MS (22), little is known about the IFN-β’s effects on B cell functions. We have characterized in this study the in vitro and in vivo IFN-β-1b’s effect on the human B cells’ stimulatory capacity toward T cells and cytokine secretion in the context of the autoimmune response in RR MS.

Peripheral blood samples were collected upon obtaining Institutional Review Board-approved informed consent from 49 RR MS patients and 23 HCs. The patients were not treated with immunomodulatory therapies at the time of blood sample collection, and their treatment-free period was as reported in our previous studies (23). PBMC were separated using Ficoll-Paque (GE Healthcare), and B cells were purified (>95% CD19⁺) using magnetic beads negative separation (Miltenyi Biotec). For the in vitro B cell expansion, B cells were isolated from seven RR MS patients and expanded according to the protocol of von Bergwelt-Baildon et al. (24). Briefly, CD40L-expressing L cells (a kind gift from Dr. Bar-Or, McGill University, and Dr. Liu, DYNAX) were irradiated (10,000 rad) and cultured with B cells at a 1:10 ratio in RPMI 1640 complete medium supplemented with 10% FBS, 2 mM glutamine, and 100 U/ml penicillin G and streptomycin (Invitrogen) in the presence of IL-4 (10 ng/ml). Every 10 d, the B cells were washed and restimulated with irradiated CD40L-expressing L cells, until a sufficient number of B cell were obtained. The in vitro-expanded B cells were used, after extensive washing and IL-4 cytokine removal, as APCs 7–10 d after restimulation. They were >99% CD19⁺, MHC class I⁺ and class II⁺, with high expression of CD80 and CD40 and no detected CD14- or CD11c-positive monocytes or DCs (Supplemental Fig. 1).

The in vitro-expanded B cells were washed and deprived overnight of IL-4 and CD40L stimulation. The next day, the B cells were stimulated with 0.5 μg/ml anti-IgG/IgM mAb (Jackson ImmunoResearch Laboratories) and 2.0 μg/ml rCD40L (R&D Systems, Minneapolis, MN) (21) in the absence or presence of IFN-β-1b (1–1000 IU/ml; Bayer). Following 24 h culture, the cells were washed and irradiated with 3000 rad to inhibit B cell proliferation and the production of proinflammatory cytokines (25, 26). The B cells were then pulsed with 10 μg/ml myelin basic protein (MBP)_83–99 or influenza virus hemagglutinin (FluHA)_306–318, and cocultured at a 1:1 ratio with Ag-specific T cell lines (27) for 48 h. For the STAT1 inhibition, the B cells were pretreated with 20 μM fludarabine (FLUD) (Sigma-Aldrich) for 2 h and then stimulated in the absence or presence of IFN-β-1b for 24 h, after which the cells were washed and cocultured with MBP_83–99-specific T cell lines. [³H]Thymidine (1 μCi/well) was added after 48 h, and thymidine incorporation was determined after 16 h in triplicate wells.

CD40, CD80, CD86, MHC class I, and class II surface expression was assessed by flow cytometry on B cells derived from 10 RR MS patients and 10 HCs 24 h after anti-IgG/IgM mAb plus CD40L stimulation in the absence or presence of IFN-β-1b. B cells were stained with fluorescently conjugated anti-CD19, -CD40, -CD80, -CD86, HLA-ABC, and HLA-DR mAbs (BD Pharmingen) according to the manufacturer’s recommendations and analyzed using FACSCalibur (BD Biosciences). Background staining was assessed using isotype-matched control Abs.

B cells derived from 10 RR MS patients and 10 HCs were stimulated with anti-IgG/IgM mAb and rCD40L in the absence or presence of IFN-β-1b for 6 h prior to RNA isolation with an RNeasy kit (Qiagen). For the STAT1 inhibition, the B cells were pretreated with 20 μM FLUD for 1 h and cultured in the absence or presence of IFN-β-1b for 6 h. Each sample was analyzed in triplicate, and relative gene expression was normalized against 18S RNA.

B cell supernatants (SNs) were collected 24 h following anti-IgG/IgM mAb and CD40L B cell activation in the absence or presence of IFN-β-1b, and samples were analyzed in duplicate. IL-1β, IL-10, and IL-12p70 secretion was measured using OptEIA (BD Pharmingen) and IL-23p19/p40 using an eBioscience ELISA kit according to the manufacturer’s protocol (eBioscience). IL-27 was assayed using ELISA plates coated with 0.4 μg/ml anti-human IL-27 mAb (R&D Systems) overnight. Plates were washed and blocked with OptEIA blocking buffer (BD Pharmingen), and samples were incubated for 2 h at room temperature. Detection was conducted with biotinylated anti-human IL-27 mAb (R&D Systems) for 1 h at room temperature.

CD19⁺ B cells were separated from the peripheral blood of six untreated and six IFN-β-1b–treated RR MS patients using negative magnetic bead separation. The B cells were cultured for 24 h, and the SNs were collected for cytokine secretion measurements using ELISA, as above. IL-27 was measured in the B cells derived from untreated and IFN-β-1b–treated patients using an IL-27 ELISA kit (R&D Systems), which has recently become commercially available.

The in vitro-expanded B cells were extensively washed and stimulated with anti-IgG/IgM mAb and rCD40L in the absence or presence of IFN-β-1b for 48 h before SN collection. IFN-β in the SNs was neutralized with anti–IFN-β Ab (4000 IU/ml, goat anti-human polyclonal Ab; R&D Systems) 2 h prior to the SN transfer. IL-1β and IL-23 (R&D Systems) were added to the SN at 10 ng/ml and 50 ng/ml, respectively. IL-27 was neutralized with blocking mAb (goat anti-human IgG Ab, R&D Systems) at 10 μg/ml. CD45RA⁺ naive T cells were separated from PBMCs using magnetic beads negative selection (>95% purity, Miltenyi Biotec) and stimulated at 0.5 × 10⁶ cells/ml with immobilized anti-CD3 (1 μg/ml, clone UCHT; R&D Systems) and anti-CD28 mAb (5 μg/ml, clone 28.2; BD Pharmingen). Activated CD45RA⁺ T cells were cultured with 250 μl SN for 3 d, prior to RNA harvesting for the gene expression assays and SN collection for the cytokine measurements.

A total of 2 × 10⁶ in vitro-expanded B cells derived from three RR MS patients were stimulated with anti-IgG/IgM mAb and CD40L in the absence or presence of IFN-β-1b for 1 or 6 h. For the inhibition of STAT1, B cells were pretreated with FLUD (20 μM) for 1 h and cultured in the absence or presence of IFN-β-1b for 1 h. The cell lysates were used for Western blotting analysis of total and phosphorylated STAT1 and STAT3, suppressor of cytokine signaling (SOCS)1, and SOCS3. The lysates were resolved on a 5–15% gradient SDS-PAGE (Bio-Rad) and transferred to a polyvinylidene difluoride membrane. The membranes were blocked with 5% milk in TBS (20 mM Tris and 500 mM NaCl) and 0.1% Tween 20 at room temperature for 2 h, followed by overnight incubation at 4°C with primary Abs against p-STAT1 (rabbit polyclonal Ab), total STAT1 (clone 9H2), p-STAT3 (rabbit polyclonal Ab), total STAT3 (rabbit polyclonal Ab), SOCS1 (rabbit polyclonal Ab), and SOCS3 (rabbit polyclonal Ab); all Abs were purchased from Cell Signaling Technology. Secondary HRP-conjugated Ab was added at 1/2000 dilution for 1 h, and the protein bands were detected with an ECL Detection System (Santa Cruz Biotechnology).

The B cells were negatively isolated using magnetic bead separation (Miltenyi Biotec) from seven untreated RR MS patients and seven patients receiving IFN-β-1b therapy for >6 mo. The B cells were stimulated with anti-IgM/IgG mAb plus CD40L for 24 h, washed, irradiated (3000 rad), and cocultured with PBMCs derived from HCs at a ratio of 1:10 for 72 h. The proliferating effector PBMCs for all experiments were derived from two HCs to provide the same readout for testing B cells derived from treated and untreated RR MS patients. MLR cultures were pulsed with [³H]thymidine for 16 h before harvesting. The proliferation is expressed as δ (Δ) cpm (B cell + PBMC coculture cpm − unstimulated PBMCs cpm).

Statistical analysis for the proliferation assays, quantitative RT-PCR, flow cytometry, and ELISA results were performed using a Wilcoxon rank-sum test. Statistical analyses of the gene expression in the activated CD45RA⁺ cells cultured with SNs from IFN-β-1b–treated B cells in the absence or presence of the indicated cytokines and blocking Abs were performed using a repeated-measures ANOVA Friedman test (InStat; GraphPad). A p value <0.05 was considered significant.

In this study, we have investigated the effect of IFN-β-1b on B cells’ stimulatory capacity using the in vitro-expanded B cells stimulated via the anti-IgG/IgM mAb BCR ligation and CD40L costimulatory signal, typically provided by activated T cells. Our results demonstrated that B cells derived from RR MS patients expressing the disease-associated MHC class II DR2 molecule, which binds autoantigen MPB_83–99 and the control FluHA_306–318 peptide with high affinity, effectively present these peptides to the Ag-specific T cell lines derived from the DR2⁺ donors (Fig. 1A). IFN-β-1b treatment of B cells inhibits their stimulatory capacity for T cell proliferation in a dose-dependent fashion, as assessed by [³H]thymidine incorporation (Fig. 1B). To confirm that the IFN-β-1b’s inhibition of T cell Ag-specific proliferation is mediated via its inhibition of B cell to T cell contact, and not via its effect on the B cells’ cytokine secretion, we compared the T cell proliferative responses in the presence (Fig. 1B, left panel) or absence of B cells’ SNs (Fig. 1B, right panel). Our results identified a comparable dose-dependent inhibition of the T cell proliferative response in both conditions, confirming that IFN-β-1b inhibits B cell to T cell contact, whereas the changes in secreted cytokines did not affect T cell proliferation. As expected, SN transfer from the IFN-β-1b–pretreated B cells to the Ag-specific memory T cells did not affect their proliferative response (data not shown). Because an IFN-β-1b dose of 1000 IU/ml induced a statistically significant (p < 0.05) inhibition of T cell proliferation, this dose was used in all subsequent experiments in this study.

To elucidate the mechanisms of IFN-β-1b–mediated inhibition of B cells’ stimulatory capacity, we investigated its effects on the CD40, CD80, CD86, MHC class I, and class II surface expression in anti-IgG/IgM mAb and CD40L-activated B cells derived from 10 RR MS patients and 10 HCs. The treatment significantly decreased CD40 expression by 1.3-fold in the RR MS patients and by 1.6-fold in the HCs; CD80 expression was significantly decreased only in the RR MS patients, whereas the expression of CD86, MHC class I, and II were not significantly inhibited upon IFN-β treatment (Fig. 2A and data not shown).

We then investigated the effect of IFN-β-1b on the B cell expression of SOCS1, which has been reported to inhibit CD40 expression in macrophages (28). Our results in the same cohort of RR MS patients and HCs revealed that IFN-β-1b significantly induces the expression of SOCS1 (by 7.4-fold and 7.9-fold, respectively) (Fig. 2B). Next, we demonstrated that IFN-β treatment of activated B cells induces early STAT1 phosphorylation, which is paralleled by an increased SOCS1 expression, as confirmed at the protein level by Western blotting (Fig. 2B). To confirm that IFN-β–induced inhibition of B cells’ CD40 expression is mediated via STAT1 phosphorylation, we employed the selective STAT1 inhibitor FLUD. Our results in the in vitro-expanded B cells derived from seven RR MS patients revealed that the IFN-β-1b–inhibited CD40 gene expression, as well as the T cell proliferative response, were reversed with the FLUD (Fig. 2C).

We have investigated the effect of IFN-β-1b on the production of IL-1β and IL-23, cytokines that promote Th17 differentiation (29), in the B cells derived from 10 RR MS patients and 10 HCs. We found that IFN-β-1b significantly downregulated IL-1β gene expression in the activated B cells derived from both RR MS patients and HCs (2.7- and 3.8-fold, respectively). The IL-23 gene expression was also inhibited in both patients and HCs (−3.2- and −5.2-fold, respectively) (Supplemental Fig. 2A). In the same B cell cultures, IFN-β-1b inhibited IL-1β secretion in both MS patients and HCs (−2.1-fold in both) and IL-23 secretion (−3.8- and −5.1-fold, respectively) (Fig. 3A).

To investigate the molecular mechanisms of IFN-β-1b–mediated inhibition of IL-1β and IL-23 secretion in B cells, we measured the gene expression of a SOCS3. In a study of the same 10 RR MS patients and 10 HCs, we found that IFN-β-1b treatment increased SOCS3 gene expression (Fig. 3B). At the protein level, we detected an early phosphorylation of STAT3, a transcription factor for SOCS3, as well as an increase in SOCS3 expression in the IFN-β-1b–treated B cells derived from three RR MS patients (Fig. 3B).

We next examined the effect of IFN-β-1b on the secretion of IL-12 and IL-27, cytokines that inhibit Th17 cell differentiation (29), in B cells derived from 10 RR MS patients and 10 HCs. IFN-β-1b induced IL-12p35 gene expression by 2.6-fold in the RR MS patients and by 2.2-fold in the HCs. IFN-β-1b also induced IL-27p28 gene expression in both RR MS patients and HCs (15.6-fold and 10.5-fold, respectively) (Supplemental Fig. 2B). The secretion of the IL-12p70 functional heterodimer was increased by 2.3-fold in the RR MS patients and by 2.1-fold in the HCs in comparison with the untreated samples, whereas the IFN-β-1b B cell treatment induced IL-27 secretion in both RR MS patients and HCs (4.0-fold and 3.1-fold, respectively) (Fig. 4).

We next sought to characterize the molecular mechanisms involved in the IFN-β-1b–induced IL-12 and IL-27 secretion in B cells. As presented in Fig. 2B, we found that STAT1 phosphorylation, which has been reported to induce the secretion of IL-12 and IL-27 (30), was induced by IFN-β-1b treatment in activated B cells. Fig. 2B presents one Western blot, which is representative of three similar experiments performed on the B cells derived from the RR MS patients.

We examined to what extent IFN-β-1b treatment-mediated changes in B cell cytokine secretion suppress the differentiation of Th17 cells. To address this question, the SNs from the in vitro-expanded B cells derived from seven RR MS patients were collected after washing, activation, and culturing of B cells for 48 h in the absence or presence of IFN-β-1b. IFN-β-1b treatment of the in vitro-expanded B cells induced a pattern of cytokine secretion changes similar to that in the ex vivo-treated B cells. It decreased the secretion of IL-1β by 9.5-fold and IL-23 by 10.2-fold, whereas the secretion of IL-27 was increased by 3.4-fold (all p values <0.05; data not shown). To test to what extent the IFN-β-1b–induced changes in B cells’ cytokine production affect the Th17 cells’ differentiation, naive CD45RA⁺ T cells derived from seven RR MS patients were stimulated with anti-CD3 and anti-CD28 mAbs and cultured with B cells’ SNs. Prior to the SN transfer, the IFN-β-1b added to the B cells was blocked with anti–IFN-β–neutralizing Abs to inhibit IFN-β-1b’s direct effect on the Th17 cell differentiation (8). The SNs derived from IFN-β-1b–treated B cells inhibited Th17 cell differentiation, as evidenced by the significantly decreased gene expression of the Th17-specific cytokine IL-17A (10.1-fold) in comparison with the SNs from untreated B cells (Fig. 5A). The inhibitory effects of the SNs from B cells treated with IFN-β-1b on the Th17 cell differentiation were reversed by the addition of IL-1β, IL-1β plus IL-23, anti–IL-27p28 mAbs, or a combination thereof, as evidenced by an increased/reversed IL-17A gene expression (Fig. 5A). A similar pattern of changes in the expression of retinoic acid-related orphan nuclear hormone receptor C and CD161 Th17 cell markers was detected in the experiment (data not shown).

In the same experiments, the measurements of secreted IL-17A demonstrated a significant (−28.1-fold) decrease in the presence of IFN-β-1b–treated B cells’ SNs, whereas the effects were reversed by the addition of IL-1β, IL-23, anti–IL27 mAb, or a combination thereof (Fig. 5B). In summary, our results demonstrate that IFN-β-1b modifies the secretion of multiple cytokines by B cells, which collectively inhibit the Th17 cell differentiation.

IFN-β-1b’s therapeutic effect in RR MS patients is characterized by an increased expression of IL-10 (2), for which production is associated with the regulatory B cell function (18). We therefore investigated the effect of IFN-β-1b on the expression of the anti-inflammatory cytokine IL-10 in B cells derived from 10 RR MS patients and 10 HCs upon stimulation via BCR and CD40 ligation. IFN-β-1b significantly induced IL-10 gene expression in the RR MS patients by 1.6-fold and in HCs by 2.4-fold (Supplemental Fig. 2C). These findings were consistent with the IFN-β-1b–induced upregulation of IL-10 secretion by 1.3-fold in the RR MS patients and by 1.4-fold in HCs (Fig. 6).

The effect of the in vivo IFN-β-1b treatment on the B cells’ stimulatory capacity toward T cells was tested in allogenic MLR using B cells derived from seven IFN-β-1b–treated and seven untreated RR MS patients. The results revealed a significant inhibition of the stimulatory capacity of B cells derived from the IFN-β-1b–treated patients in comparison with the untreated patients, measured by the proliferation response of allogenic PBMCs (Fig. 7A).

Further studies of B cells separated from RR MS patients receiving IFN-β-1b therapy revealed a significantly decreased secretion of IL-1β and IL-23 and higher secretion of IL-12 and IL-27 in comparison with the B cells derived from the untreated MS patients (Fig. 7B, 7C), confirming that IFN-β-1b’s effects on B cell cytokine secretion contribute to its in vivo operative therapeutic effects.

A recent clinical trial has demonstrated that depletion of B cells with anti-CD20 mAb provides rapid and effective disease activity suppression in patients with RR MS (12). The results suggest, in support of previously reported animal studies, that B cells may play a prominent role in the pathogenesis of MS through their role in Ag presentation (22), induction of T cell differentiation (31), and the regulation of autoimmune inflammatory responses (10). Duddy et al. (21) have recently reported that in vitro-activated B cells derived from MS patients treated with mitoxantrone, an approved therapy for aggressive MS, secrete significantly higher amounts of IL-10 in comparison with untreated patients, implicating this as a therapeutic mechanism. IFN-β-1b has been used for many years as a treatment for RR MS; however, very few studies have addressed its effect on the B cells. One of the proposed IFN-β therapeutic mechanisms is linked to its ability to downregulate Ag presentation by B cells. Jiang et al. (13) have demonstrated that IFN-β-1b reduced the IFN-γ–induced Ag-presenting capacity for tetanus toxoid recall Ag in human B cells. In the current study, we investigated the effect of IFN-β-1b on the B cells’ capacity to present the autoantigen MBP_83–99 and found that activated B cells effectively present MBP_83–99 and FluHA_306–318 control peptide to the Ag-specific T cells. IFN-β-1b treatment of B cells inhibited T cell proliferation in a dose-dependent manner even after removal of the SNs from cocultures, implying that the IFN-β-1b inhibition of B cell to T cell contact suppresses T cell proliferation even in the absence of IFN-β-1b’s effect on the B cell cytokine secretion.

Liu et al. (32) have shown that IFN-β-1b treatment of RR MS patients decreases the expression of costimulatory molecule CD40 on B cells. Although there are no data on the mechanisms of CD40 inhibition in B cells, Qin et al. (28) have reported that in macrophage IFN-β induced SOCS1, which downregulated the CD40 expression. Our current study has demonstrated that IFN-β-1b decreases the CD40 expression on B cells derived from both RR MS patients and HCs (Fig. 2A). Furthermore, we found in the same cohort of RR MS patients and HCs that IFN-β-1b significantly increased SOCS1 expression in B cells (Fig. 2B), which is typically induced upon STAT1 phosphorylation (28). Because our results demonstrated that IFN-β-1b induces B cell STAT1 phosphorylation, we propose that the STAT1-induced SOCS1 expression downregulates the CD40 expression, which in turn inhibits Ag-specific T cell proliferative response (Fig. 2C). Finally, using the STAT1 inhibitor FLUD, we confirmed that IFN-β-1b’s suppression of the CD40 expression and subsequent Ag-specific T cell proliferation is mediated via IFN-β-1b–induced STAT1 phosphorylation (Fig. 2C).

In addition to Ag presentation, B cells secrete multiple cytokines that regulate T cell differentiation (33). We demonstrated that IFN-β-1b in vitro treatment inhibit B cells’ IL-1β and IL-23 and induces IL-12 and IL-27 secretion, similar to our previous report on the IFN-β-1b modification of cytokine secretion in DCs (8). It has been previously demonstrated by our laboratory and others (8, 33) that IFN-β induces SOCS3 expression in DCs, which inhibited IL-1β (34, 35) and IL-23 secretion (34, 36). In an attempt to elucidate the molecular mechanism of IFN-β–induced cytokine changes in B cells, we demonstrated that IFN-β-1b induces STAT3 phosphorylation, a transcription factor for SOCS3 (37). We therefore postulate that IFN-β-1b–mediated suppression of IL-1β and IL-23 secretion by B cells occurs through the induction of STAT3 phosphorylation and subsequent SOCS3 expression, whereas the secretion of IL-12 and IL-27 is induced by IFN-β-1b–mediated STAT1 phosphorylation.

Collectively, this pattern of changes in B cell cytokine production inhibits Th17 cell differentiation, as demonstrated in the SN transfer experiments. B cell-mediated Th17 cell differentiation has been reported by two earlier studies (17, 38). However, these studies did not address the involvement of B cell-secreted cytokines in the differentiation of this pathogenic T cell subset. We demonstrate that the IFN-β-1b–modulated B cell cytokine secretion profile inhibits Th17 cell differentiation, as evidenced by the suppression of IL-17A (Fig. 5A), retinoic acid-related orphan nuclear hormone receptor C and CD161 gene expression (data not shown), and IL-17A cytokine secretion (Fig. 5B). The addition of IL-1β and IL-23, neutralization of IL-27, and combinations thereof reversed the suppressive effect of IFN-β-1b–mediated changes in B cell cytokine secretion on the Th17 cell differentiation, confirming that IFN-β-1b suppresses B cell-induced Th17 cell differentiation through the downregulation of IL-1β or IL-23 and the induction of IL-27 secretion.

One of the hallmarks of IFN-β’s therapeutic effect in MS has been an increased expression of the immunosuppressive cytokine IL-10 (2). The regulatory role of B cells in the resolution of autoimmune inflammatory responses has been associated with their IL-10 secretion in EAE, inflammatory bowel syndrome, and collagen-induced arthritis (10, 39). Our study has demonstrated that IFN-β-1b induced IL-10 gene expression and cytokine secretion in B cells derived from both RR MS patients and HCs.

Studies of the IFN-β-1b–treated RR MS patients revealed that B cells derived from treated patients, upon ex vivo separation and short-term activation, induced a significantly lower T cell proliferative response in allogenic MLR in comparison with the B cells derived from untreated patients. The results confirmed that the in vitro-demonstrated IFN-β–induced inhibition of CD40 and CD80 expression may contribute to the in vivo operative IFN-β-1b’s therapeutic mechanisms. Furthermore, studies of the B cells derived from the IFN-β-1b–treated patients revealed that they secrete significantly lower levels of IL-1β and IL-23 and higher levels of IL-12 and IL-27 in comparison with the B cells derived from the untreated patients. Importantly, studies of the B cells derived from IFN-β–treated patients confirmed that the in vitro-described IFN-β–induced changes in B cell cytokine secretion may be operative in the IFN-β-1b’s therapeutic effect in treated patients.

In summary, we report that IFN-β-1b treatment of B cells derived from RR MS patients and HCs inhibits their multiple proinflammatory functions, including T cell stimulatory capacity and cytokine secretion, whereas it induces regulatory IL-10 cytokine production. Studies of the RR MS patients treated with IFN-β-1b confirmed that its in vivo operative therapeutic effects include inhibition of B cell stimulatory capacity and the B cells’ cytokine secretion changes, which may selectively inhibit Th17-mediated autoimmune response in RR MS. Identification of novel mechanisms of action of this immunomodulatory therapy targeting B cells, for which contribution to the development of the autoimmune response in MS has recently received significant attention, may uncover additional mechanisms of the B cell contribution to the autoimmune response and provide new targets for future selective treatment of MS.

We thank Dr. Xin Zhang for help with the statistical analysis.

The authors have no financial conflicts of interest.