Research ArticleDISEASES AND DISORDERS

BAFF inhibition attenuates fibrosis in scleroderma by modulating the regulatory and effector B cell balance

See allHide authors and affiliations

Science Advances  11 Jul 2018:
Vol. 4, no. 7, eaas9944
DOI: 10.1126/sciadv.aas9944

Abstract

Systemic sclerosis (SSc) is an autoimmune disease characterized by skin and lung fibrosis. More than 90% of patients with SSc are positive for autoantibodies. In addition, serum B cell activating factor (BAFF) level is correlated with SSc severity and activity. Thus, B cells are considered to play a pathogenic role in SSc. However, there are two opposing subsets: regulatory B cells (Bregs) and effector B cells (Beffs). Interleukin-10 (IL-10)–producing Bregs negatively regulate the immune response, while IL-6–producing Beffs positively regulate it. Therefore, a protocol that selectively depletes Beffs would represent a potent therapy for SSc. The aims of this study were to investigate the roles of Bregs and Beffs in SSc and to provide a scientific basis for developing a new treatment strategy targeting B cells. A bleomycin-induced scleroderma model was induced in mice with a B cell–specific deficiency in IL-6 or IL-10. We also examined whether BAFF regulates cytokine-producing B cells and its effects on the scleroderma model. IL-6–producing Beffs increased in number and infiltrated the inflamed skin in the scleroderma model. The skin and lung fibrosis was attenuated in B cell–specific IL-6–deficient mice, whereas B cell–specific IL-10–deficient mice showed more severe fibrosis. In addition, BAFF increased Beffs but suppressed Bregs. Furthermore, BAFF antagonist attenuated skin and lung fibrosis in the scleroderma model with reduction of Beffs but not of Bregs. The current study indicates that Beffs play a pathogenic role in the scleroderma model, while Bregs play a protective role. BAFF inhibition is a potential therapeutic strategy for SSc via alteration of B cell balance.

INTRODUCTION

B cells are important for antibody (Ab) production and for antigen presentation and cytokine production (1). In particular, cytokine-producing B cells play critical roles in multiple aspects of immunity. There are two opposing B cell subsets: regulatory B cells (Bregs) and effector B cells (Beffs) (2). Interleukin-10 (IL-10)–producing Bregs are now recognized as negative regulators of the immune system, inflammation, and autoimmunity based on studies with human subjects and mouse models of autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus (SLE), and multiple sclerosis (MS) (36). The phenotype of mouse splenic Bregs is derived from two different B cell subsets: marginal zone (MZ) and B1 B cells (7). Furthermore, it was reported that the CD9+ B cell subset is enriched in IL-10–producing Bregs (7, 8), since this subset includes both the MZ and B1 B cell subsets (9). By contrast, cytokine-producing Beffs can positively modulate the immune response through the production of various cytokines (2). For example, lymphotoxin-producing Beffs are essential for the ontogenesis, homeostasis, and activation of secondary lymphoid organs, as well as for the development of tertiary lymphoid tissues at ectopic sites (10). Other Beffs have been shown to modulate the development of effector and memory CD4+ T cell responses via the production of cytokines such as IL-6, interferon-γ, and tumor necrosis factor (11). Therefore, a protocol that selectively depletes Beffs while sparing Bregs would represent a potent therapy for autoimmune diseases.

Systemic sclerosis (SSc; also known as scleroderma) is a connective tissue disorder characterized by excessive fibrosis in the skin and various internal organs with an autoimmune etiology. More than 90% of SSc patients are positive for autoantibodies such as anti–DNA topoisomerase I, anticentromere, and anti–RNA polymerase Abs. One study demonstrated that SSc patients displayed distinct abnormalities of blood B cell compartments characterized by expanded naïve B cells and activated memory B cells (12). B cell activating factor (BAFF) has been shown to be present at elevated levels in patients with SSc and is correlated with disease severity (13). B cells of patients with SSc that were stimulated by BAFF exhibited an enhanced ability to produce IL-6 (13). In addition, B cells and BAFF were shown to promote collagen production by dermal fibroblasts in SSc (14). IL-6 plays an important role in tissue fibrosis and autoimmunity in the SSc pathogenesis (15, 16) and is thus considered a candidate therapeutic target. The skin fibrosis of two patients with diffuse SSc was markedly improved after treatment with the anti–IL-6 receptor Ab tocilizumab (17). A phase 2 trial of tocilizumab demonstrated clinically significant improvement of skin fibrosis in patients with SSc (18). Thus, IL-6–producing Beffs may play a critical role in the development of scleroderma; however, there is no reliable detection method for IL-6–producing Beffs, and the resulting phenotype remains unclear. By contrast, IL-10–producing Bregs were shown to suppress the skin fibrosis of the Scl-cGVHD model, an animal model for human SSc (19). IL-10–producing Bregs have been reported to be decreased in patients with SSc and were associated with disease activity (20).

Here, we investigated the intracellular staining and phenotype of IL-6–producing Beffs. Furthermore, we evaluated the role of IL-6–producing Beffs and IL-10–producing Bregs in the pathogenesis of scleroderma using B cell–specific cytokine-deficient mice. On the basis of these results, we propose a new potential therapeutic strategy for SSc via alteration of the Beff and Breg balance.

RESULTS

CD40 and lipopolysaccharide synergistically induce IL-6 production from B cells

To identify the stimulation condition of IL-6 production from B cells, we cultured B cells with various Toll-like receptor (TLR) agonists with or without agonistic CD40 monoclonal Ab (mAb). The TLR4 agonist [lipopolysaccharide (LPS)] and TLR9 agonist induced IL-6 production from B cells. Addition of the agonistic CD40 mAb in combination with LPS or TLR9 agonist signals significantly enhanced IL-6 production from B cells (Fig. 1A). Similar to the results for IL-6, IL-10 production was induced by LPS and the TLR9 agonist. By contrast, addition of agonistic CD40 mAb in combination with LPS signals significantly reduced IL-10 production from B cells (Fig. 1A). Thus, agonistic CD40 mAb accelerates IL-6 production from B cells stimulated with LPS, while agonistic CD40 mAb attenuates IL-10 production from B cells stimulated with LPS.

Fig. 1 CD40 and LPS synergistically induce IL-6 production from B cells.

(A) B cells were isolated from spleens of naïve mice by magnetic sorting based on CD19 expression. Sorted B cells were cultured for 72 hours with media alone or media containing anti-CD40 mAb, along with the indicated TLR agonists. After in vitro stimulation for 72 hours, IL-6 (left) and IL-10 (right) levels in supernatants were quantified by enzyme-linked immunosorbent assay (ELISA). Bars represent the means ± SD from three independent experiments (n = 3 mice). Significant differences between means of media alone and individual stimuli are indicated: *P < 0.001, **P < 0.0001, analysis of variance (ANOVA) followed by Tukey’s multiple comparison test. Significant differences between cultures with or without anti-CD40 mAb are indicated: #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001, Student’s t test. (B) IL-6–producing B cells were determined after in vitro stimulation by LPS, anti-CD40 mAb, and LPS + anti-CD40 mAb, with PIB [phorbol 12-myristate 13-acetate (PMA), ionomycin, and brefeldin A] added during the final 5 hours of cultures (5 to 48 hours). Isotype control Ab was used as negative controls for IL-6 staining. Percentages indicate the frequencies of cytoplasmic IL-6+ B cells within the indicated gates among total CD19+ B cells. Bars represent the means ± SD from three independent experiments (n = 3 mice). *P < 0.0001, ANOVA followed by Tukey’s multiple comparison test. (C) Representative cell surface phenotype of spleen IL-6–producing B cells after stimulation with LPS + anti-CD40 mAb for 24 hours with PIB added during the final 5 hours of culture. Cultured cells were stained for viability and cell surface molecule expression (using LEGENDScreen Mouse PE Kit from BioLegend), permeabilized, stained with anti–IL-6 mAb, and analyzed by flow cytometry. Representative cell surface molecule expression by IL-6+ (red line) and IL-6 (black line) CD19+ B cells from three individuals is shown. Shaded histograms represent isotype-matched control mAb staining.

To visualize IL-6–producing B cells, we established a detection method of intracellular IL-6 staining by fluorescence-activated cell sorting (FACS). We cultured splenocytes with LPS, agonistic CD40 mAb, or LPS + agonistic CD40 mAb for various time courses (5, 12, 24, or 48 hours). We added PIB during the final 5 hours of cultures. In line with the results described above, LPS and agonistic CD40 mAb signals cooperatively induced the IL-6 production of B cells (Fig. 1B). In addition, the 24-hour culture was found to be the best condition for the detection of IL-6–producing B cells, and approximately 40% of the B cells produced IL-6 (Fig. 1B). Therefore, the culture with LPS and agonistic CD40 mAb for 24 hours appears to be the best condition for visualizing IL-6–producing B cells.

MZ B cell-related cell surface markers are highly expressed in IL-6–producing B cells

To identify whether IL-6–producing B cells represent a unique or known B cell subset, we analyzed the cell surface phenotype. We assessed the phenotype of IL-6–producing B cells following 24 hours of culture with LPS and agonistic CD40 mAb, along with 5 hours of PIB stimulation. On average, IL-6+ B cells expressed higher densities of CD1d, CD9, CD21, CD23, CD25, CD80, CD86, CD150 [SLAM (Signaling lymphocyte activation molecule)], CD155 [PVR (Poliovirus receptor)], CD200 (OX2), and CD267 [TACI (transmembrane activator and calcium-modulator and cyclophilin ligand interactor)], which is one of the BAFF receptors, when compared with IL-6 B cells (Fig. 1C and fig. S1). Although naïve B cells do not express CD25, we induced most of the B cells to express CD25 after 24 hours of culture with LPS and agonistic CD40 mAb, followed by 5 hours of PIB stimulation (fig. S2). By contrast, on average, IL-6+ B cells expressed lower densities of CD43, immunoglobulin M (IgM), and Ly-6D when compared with IL-6 B cells (fig. S1). The higher expression of CD1d, CD9, and CD21 on IL-6+ B cells suggests that IL-6+ B cells might be predominantly found within the MZ B cell subset. Furthermore, IL-6+ B cells show higher expression of TACI, a receptor of BAFF.

The MZ B cell subset is a major source of IL-6–producing B cells

We previously reported that IL-10–producing Bregs were predominantly found within the splenic CD1dhi MZ and CD5+ B1 B cell subsets (7). To determine which B cell subsets secreted IL-6, we stained and sorted splenic B cells into three fractions [follicular B cells (CD1dintCD5), MZ B cells (CD1dhiCD5), and B1 B cells (CD1dintCD5+)] before in vitro 24-hour stimulation with LPS and agonistic CD40 mAb, followed by 5 hours of PIB stimulation (Fig. 2A). The frequency of IL-6–producing B cells among sorted follicular B cells was comparable with that detected in pan-B cells. The frequency of IL-6–producing B cells among sorted B1 B cells was decreased compared with that of pan-B cells. By contrast, the frequency of IL-6–producing B cells in sorted MZ B cells was significantly increased compared with that in other B cell subsets (Fig. 2, B and C). As previously reported, MZ and B1 B cell subsets produced IL-10, while the follicular B cell subset did not produce IL-10 (Fig. 2, B and C). In addition, the frequency of IL-10–producing B cells in B1 B cells was significantly increased compared with that in MZ B cell subsets (Fig. 2, B and C). Next, to examine whether B cells simultaneously produce IL-6 and IL-10, we stained B cells with both IL-6 and IL-10 and analyzed them by FACS. We found that IL-6– and IL-10–producing B cells exist mutually exclusively (fig. S3). Thus, MZ B cells represent a major subset of IL-6–producing B cells in the spleen.

Fig. 2 The MZ B cell subset is a major source of IL-6–producing B cells.

(A) Splenic B cells from wild-type mice were isolated by Miltenyi MACS enrichment; stained for CD1d, CD5, and CD19 expression; and sorted into follicular B cell (CD1dintCD5), MZ B cell (CD1dhiCD5), and B1 B cell (CD1dintCD5+) populations before stimulation. (B) Sorted B cells were cultured with LPS + anti-CD40 mAb for 24 hours with PIB added during the final 5 hours of culture (for IL-6) or LPS + PIB for 5 hours (for IL-10). IL-6+ or IL-10+ B cells derived from each purified population were then analyzed by flow cytometry. All data are representative of two independent experiments. (C) Bars represent the means ± SD from four mice in each group. Significant differences between pan-B cell versus other B cell subsets are indicated: *P < 0.05, **P < 0.01, ***P < 0.001, #P < 0.05, ##P < 0.01, ###P < 0.001, ANOVA followed by Tukey’s multiple comparison test.

IL-6 is increased in bleomycin-induced scleroderma

To determine whether and when the levels of IL-6 and IL-10 are increased in the bleomycin-induced scleroderma model, we measured the serum cytokine levels and the frequency of splenic cytokine-producing B cells. Serum IL-6 levels were gradually increased along with bleomycin treatment, while there was no change in serum IL-10 levels with bleomycin treatment (Fig. 3A). In addition, the frequency of splenic IL-6–producing B cells at 3 weeks after bleomycin treatment was significantly increased compared with that detected at 3 weeks after phosphate-buffered saline (PBS) treatment (Fig. 3B).

Fig. 3 IL-6 is increased in bleomycin-induced scleroderma.

(A) Serum samples were collected from bleomycin-induced scleroderma mice. Serum IL-6 or IL-10 levels were determined by ELISA. Bars represent the means ± SD from four mice in each group. Significant differences between means of naïve mice and bleomycin (Bleo)–treated mice are indicated: *P < 0.05, **P < 0.001, ANOVA followed by Tukey’s multiple comparison test. (B) Splenocytes were isolated from bleomycin-induced scleroderma mice on day 21 after treatment. Splenocytes were cultured with LPS + anti-CD40 mAb for 24 hours with PIB added during the final 5 hours of culture (for IL-6) or LPS + PIB for 5 hours (for IL-10). Left: Percentages indicate the frequencies of cytoplasmic IL-6+ or IL-10+ B cells within the indicated gates among total CD19+ B cells. Right: Bars represent the means ± SD from four mice in each group. *P < 0.001, Student’s t test. (C) Skin-infiltrating cells were isolated from bleomycin-induced scleroderma mice on day 21 after treatment. Lymphocytes were cultured with LPS + anti-CD40 mAb for 24 hours with PIB added during the final 5 hours of culture (for IL-6) or LPS + PIB for 5 hours (for IL-10). Left: Percentages indicate the frequencies of cytoplasmic IL-6+ or IL-10+ B cells within the indicated gates among total CD19+ B cells. Right: Bars represent the means ± SD from four mice in each group. *P < 0.001, Student’s t test. (D and E) Inflamed skin sample was collected from mice treated with bleomycin. Immunofluorescence histology of frozen skin sections detecting IL-6 production by B220+ skin B cells. DAPI (4′,6-diamidino-2-phenylindole) was used to visualize nuclei. Scale bars, 25 μm (D) and 5 μm (E). All data are representative of two independent experiments.

Next, to determine whether IL-6–producing Beffs infiltrate the inflamed skin, we analyzed the cytokine-producing B cells by FACS and immunohistochemistry. For FACS analysis, we collected B cells infiltrating the inflamed skin and stimulated them in vitro. The number of IL-6–producing Beffs in inflamed skin with bleomycin treatment was significantly increased compared with that in the skin of mice that received PBS treatment (Fig. 3C). The numbers of IL-10–producing Bregs in inflamed skin with bleomycin treatment were also significantly increased compared with those in the skin of PBS-treated mice (Fig. 3C). In addition, immunohistochemical analysis revealed the presence of IL-6–producing Beffs (IL-6+B220+) in the inflamed skin in the bleomycin-induced scleroderma model (Fig. 3, D and E). Unlike in the spleen, most IL-6–producing Beffs were found in the CD1dintCD5, not in the CD1dhi B cell subset (fig. S4). Thus, IL-6–producing Beffs are increased and infiltrated the inflamed skin in the bleomycin-induced scleroderma model.

Skin and lung fibrosis are attenuated in mice with B cell–specific IL-6 deficiency

We next evaluated whether cytokine-producing B cells regulate the skin fibrosis of the bleomycin-induced scleroderma model. To this end, we generated mixed bone marrow chimeric mice with a B cell–specific deficiency in IL-6 production (B-IL-6−/−) or IL-10 production (B-IL-10−/−), together with control chimeras (control). The schema and validation of these mixed bone marrow chimeric mice are outlined in fig. S5. Although B cells showed complete lack of IL-6 or IL-10 production in B-IL-6−/− or B-IL-10−/− mice, B-IL-6−/− or B-IL-10−/− mice demonstrated 20% IL-6 or IL-10 deficiency in all hematopoietic cells. To compensate for the effect of this deficiency outside the B cell compartment, we generated control mice against B-IL-6−/− or B-IL-10−/− mice as IL-620% or IL-1020% chimeras, respectively, which showed 20% IL-6 or IL-10 deficiency in all hematopoietic compartments. We left the hematopoietic compartment for 8 to 10 weeks to repopulate, and then, we injected the chimeric mice with bleomycin to induce scleroderma. IL-6 deficiency of B cells caused significant reduction in dermal thickness, type 1 collagen mRNA expression, and lung fibrosis (Fig. 4, A and B, and fig. S6, A to C). In contrast, B cell IL-10 deficiency significantly augmented the dermal thickness, type 1 collagen mRNA expression, and lung fibrosis (Fig. 4, C and D, and fig. S6, D to F).

Fig. 4 Skin fibrosis is attenuated in mice with B cell–specific IL-6 deficiency.

(A to D) Bleomycin-induced scleroderma was induced in mice with B cell–specific IL-6 deficiency (B-IL-6−/−; wild-type mice lethally irradiated and reconstituted with 80% μMT plus 20% Il6−/− bone marrow) or B cell–specific IL-10 deficiency (B-IL-10−/−; wild-type mice lethally irradiated and reconstituted with 80% μMT plus 20% Il10−/− bone marrow) and control chimera groups [wild-type mice lethally irradiated and reconstituted with 80% wild-type plus 20% Il6−/− or Il10−/−bone marrow (IL-620% or IL-1020%, respectively)]. (A and C) Skin samples were harvested 4 weeks after PBS or bleomycin treatment. Masson’s trichrome stain. Representative images. Arrows indicate dermis. Scale bars, 100 μm. Right: Dermal thickness (distance from the dermal-epidermal junction to the adipose layer), shown as the means ± SD of triplicate determinations per hpf from 10 mice per group. (B and D) Expression of col1a2 mRNA in the skin was measured by real-time polymerase chain reaction (PCR), shown as the means ± SD of triplicate determinations per hpf (high power field) from 10 mice per group. Open circles, PBS; closed circles, bleomycin. *P < 0.05, ***P < 0.001, ****P < 0.0001, one-way ANOVA followed by Tukey’s multiple comparison test. (E and F) B cells and fibroblasts were cocultured. Collagen release by fibroblasts was determined by the Sirius red assay in 72-hour culture supernatants of fibroblast cultured alone, cocultured with B cells, or recombinant TGF-β1 (5 ng/ml). IL-6 was determined by ELISA in 72-hour culture supernatants. B cells from wild-type, Il6−/−, or Il10−/− mice were isolated by Miltenyi MACS enrichment. B cells were either in cell-cell contact with fibroblast (contact) or seeded in the upper chamber of a Transwell culture insert (Transwell). Bars represent the means ± SD from two independent experiments (n = 4 mice). Significant differences between fibroblast only versus other groups are indicated: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001, ANOVA followed by Tukey’s multiple comparison test.

IL-6–producing Beffs promote collagen secretion by fibroblasts

To further determine the role of cytokine-producing B cells in the regulation and development of skin fibrosis, we cocultured and analyzed B cells and fibroblasts (Fig. 4E). B cells from wild-type mice strongly induced collagen secretion by fibroblasts (white bars in “Fibro-contact B cell” versus “Fibro only”), with a magnitude comparable to that observed under stimulation with recombinant transforming growth factor–β (TGF-β; black bar in “Fibro + TGF-β”). In addition, B cells from IL-6−/− mice showed significantly decreased levels of collagen secretion by fibroblasts compared with B cells from wild-type mice (P < 0.001; white versus blue bars in “Fibro-contact B cell”; Fig. 4E). By contrast, B cells from IL-10−/− mice showed significantly increased collagen secretion by fibroblasts compared with those from wild-type mice (P < 0.05; white versus red bars in “Fibro-contact B cell”). Similarly, IL-6 secretion into the supernatant was significantly induced in coculture of fibroblasts and B cells from wild-type mice (P < 0.0001; white bars in “Fibro-contact B cell” versus “Fibro only”; Fig. 4F). IL-6 production was significantly lower in the coculture with B cells from IL-6−/− mice than with those of wild-type mice (P < 0.05; white versus blue bars in “Fibro-contact B cell”; Fig. 4F), while IL-6 production was stronger with B cells from IL-10−/− mice than with those from wild-type mice (P < 0.01; white versus red bars in “Fibro-contact B cell”; Fig. 4F). However, we could not detect IL-10 production in the supernatant of these coculture systems. We then evaluated whether the increased collagen secretion induced by B cells is dependent on cell-cell contact through interactions between B cells and fibroblasts. B cell–induced collagen production by fibroblasts was significantly inhibited when we used Transwells (P < 0.0001, B cells from wild-type mice; P < 0.0001, B cells from IL-10−/− mice; Fig. 4E). Similarly, IL-6 production was significantly inhibited when we used Transwells (P < 0.0001, B cells from wild-type mice; P < 0.0001, B cells from IL-10−/− mice; Fig. 4F). Thus, IL-6–producing Beffs promote collagen secretion by fibroblasts through the interaction with B cells and fibroblasts.

BAFF increases IL-6–producing Beffs but attenuates IL-10–producing Bregs

BAFF exhibits a strong costimulatory function for B cell activation in vitro (21). We found that the IL-6+ B cells showed high expression levels of the BAFF receptor (Fig. 1C). To determine whether BAFF modulates the cytokine production of B cells, we cultured splenic B cells with BAFF along with LPS and/or CD40. BAFF significantly enhanced IL-6 production from B cells stimulated with LPS alone and LPS + CD40 (P < 0.01; Fig. 5A), although only BAFF stimulation failed to induce IL-6 from B cells. By contrast, BAFF significantly inhibited IL-10 production from B cells stimulated with LPS (P < 0.0001; Fig. 5A).

Fig. 5 BAFF enhances IL-6 production from B cells, while BAFF attenuates IL-10 production from B cells.

(A) B cells were isolated from spleens of naïve mice by magnetic sorting based on CD19 expression. Sorted B cells were cultured for 72 hours with or without BAFF, along with LPS, anti-CD40 mAb, or LPS + anti-CD40 mAb. After in vitro stimulation for 72 hours, IL-6 (left) and IL-10 (right) levels in supernatants were quantified by ELISA. Bars represent the means ± SD from three independent experiments (n = 3 mice). Significant differences between means of media alone and individual stimuli are indicated: *P < 0.001, **P < 0.0001, ANOVA followed by Tukey’s multiple comparison test. Significant differences between cultures with or without BAFF are indicated: #P < 0.01, ##P < 0.0001, Student’s t test. (B) Wild-type mice were treated with BAFFR-Fc or Fc control protein (control). Spleens were collected 1 week after treatment. IL-6– or IL-10–producing B cells were determined after in vitro stimulation. CD1dhiMZ B cells and CD5+ B1 B cells from spleens of mice treated with BAFFR-Fc or Fc control protein were examined by flow cytometry. Percentages indicate the frequencies of various B cell subsets within the indicated gates among total CD19+ B cells. Bars represent the means ± SD from four mice. NS, not significant. *P < 0.001, **P < 0.0001, Student’s t test. All data are representative of two independent experiments.

We next evaluated whether BAFF inhibition would modulate cytokine production from B cells in vivo. To this end, we administered BAFFR (BAFF-receptor)–Fc, which neutralizes BAFF, or Fc control protein (control) to naïve mice. The frequency and number of IL-6–producing Beffs in the BAFFR-Fc–treated mice were significantly decreased compared with those of control mice (Fig. 5B). By contrast, the frequency of IL-10–producing Bregs in the BAFFR-Fc–treated mice was significantly increased compared with that in control mice, although the number of IL-10–producing Bregs did not differ between the BAFFR-Fc–treated mice and control mice (Fig. 5B). In addition, BAFFR-Fc significantly decreased the number of MZ B cells, the major subset of IL-6–producing Beffs (Fig. 5B), but did not influence the number of B1 B cells, the major subset of IL-10–producing Bregs (Fig. 5B). These results suggest that BAFF increases IL-6–producing Beffs but attenuates IL-10–producing Bregs.

BAFF inhibition attenuates the skin and lung fibrosis of bleomycin-induced scleroderma

Dysregulation of serum BAFF levels in Tsk/+ mice, a genetic mouse model for scleroderma, has been demonstrated (15). Thus, we explored the timing of changes in serum BAFF in bleomycin-induced scleroderma using ELISA. Serum BAFF levels in the bleomycin-induced scleroderma model gradually increased after bleomycin treatment (Fig. 6A). To determine whether BAFF inhibition affects the skin fibrosis in the bleomycin-induced scleroderma model, we treated the mice with BAFFR-Fc, which neutralizes BAFF, or Fc control protein (control) three times a week for 4 weeks. The skin and lung fibrosis in the mice that received BAFFR-Fc treatment was significantly attenuated compared to that of the control group (Fig. 6, B to E). In addition, BAFFR-Fc significantly decreased the number of IL-6–producing Beffs but did not influence the number of IL-10–producing Bregs (Fig. 6F). These results suggest that BAFF inhibition is a potential therapeutic strategy for SSc via alteration of the Beff and Breg balance (Fig. 6G).

Fig. 6 BAFF inhibition attenuates skin fibrosis in bleomycin-induced scleroderma.

(A) Serum samples were collected from bleomycin-induced scleroderma mice. Serum BAFF levels were determined by ELISA. Bars represent the means ± SD from five mice in each group. Significant differences between means of naïve mice and bleomycin (Bleo)–treated mice are indicated: *P < 0.001, ANOVA followed by Tukey’s multiple comparison test. (B) Bleomycin-induced scleroderma was induced in mice treated with BAFFR-Fc or Fc control protein (control). Skin samples were harvested 4 weeks after bleomycin treatment. Left: Masson’s trichrome stain. Representative images. Arrows indicate dermis (distance from the dermal-epidermal junction to the adipose layer). Scale bar, 100 μm. (B and C) Analysis of dermal thickness (B, right) and expression of col1a2 mRNA in the skin (C). (D) Lung samples were harvested 4 weeks after bleomycin treatment. Left: Hematoxylin and eosin (H&E). Representative images. Scale bar, 100 μm. (D and E) Analysis of the lungs for determination of lung fibrosis scores (D, right) and collagen content (E). (B to D) Values are means ± SD of five mice per group. Open circles, PBS; closed circles, bleomycin. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, one-way ANOVA followed by Tukey’s multiple comparison test. (F) Spleens from bleomycin-induced scleroderma mice were collected 4 weeks after treatment. IL-6– or IL-10–producing B cells were determined after in vitro stimulation. Percentages indicate the frequencies of IL-6– or IL-10–producing B cells within the indicated gates among total CD19+ B cells. Bars represent the means ± SD from five mice. *P < 0.0001, Student’s t test. All data are representative of two independent experiments. (G) BAFF increased the IL-6–producing Beffs but suppressed the IL-10–producing Bregs. Furthermore, Beffs play a pathogenic role in scleroderma, while Bregs play a protective role.

DISCUSSION

Despite their recognized importance, the phenotype and function of IL-6–producing Beffs have thus far remained poorly understood, although B cells are known to produce large amounts of IL-6. Here, we demonstrated that CD40 and LPS synergistically induce IL-6 production from B cells and that the MZ B cell subset is a major source of IL-6–producing B cells. Furthermore, IL-6–producing Beffs play a pathogenic role in the bleomycin-induced scleroderma model, whereas IL-10–producing Bregs play a protective role. Moreover, we found that BAFF inhibition attenuates skin and lung fibrosis in the bleomycin-induced scleroderma model with reduction of IL-6–producing Beffs. Collectively, these findings suggest that B cells have reciprocal roles in the pathogenesis of SSc, exhibiting both pathogenic and protective functions (Fig. 6G).

CD40 is expressed on the surface of B cells, as well as dendritic cells and monocytes/macrophages, while the CD40 ligand (CD40L) is expressed on activated T cells. The CD40-CD40L interaction induces B cell survival, Ig class switching, and cytokine production and has been shown to play an important role in the pathogenesis of SSc (22). The CD40-CD40L interaction and antigen-specific signals were reported to be essential for IL-10 production from B cells (3), while TLR signals strongly augment the production of IL-10 in mice (23) and humans (24). The current study and previous ones have shown that CD40 and a TLR signal (LPS) synergistically promote IL-6 production from B cells, whereas the CD40 signal inhibited IL-10 production (11, 23). Although the fact that we found the MZ B cell subset to be a major source of IL-6–producing B cells is consistent with the literature (11, 25), this subset is also known to be a major component of IL-10–producing Bregs (5, 7, 26). Thus, the CD1dhigh MZ B cell subset has the ability to become either IL-10–producing Bregs or IL-6–producing Beffs, and its fate depends on the stimulation by CD40 and TLR signals. Thus, it is difficult to evaluate the adaptive transfer of the CD1dhigh MZ B cell effect in the bleomycin-induced scleroderma model. Together, these results suggest that TLR stimulation induces B cells into Bregs, while the T cell–B cell interaction upon TLR stimulation induces B cells into IL-6–producing Beffs.

IL-6 is a multifunctional cytokine produced by various cell types such as B cells, T cells, monocytes, natural killer cells, and fibroblasts, although B cells are the major source of IL-6 (11, 25). Serum IL-6 levels are elevated in patients with diffuse SSc and are associated with the extent of skin thickness (27), and IL-6 plays a critical role in tissue fibrosis and autoimmunity in mouse models of SSc (15, 16). Inhibition of IL-6 suppresses skin fibrosis in a mouse model of SSc (28), and a phase 2 trial of tocilizumab showed clinically significant improvement of skin fibrosis in patients with SSc (18). Notably, the current study revealed that IL-6 deficiency in B cells alone attenuates the skin fibrosis in the bleomycin-induced scleroderma model, while IL-10 deficiency in B cells augments the skin fibrosis. B cells were shown to promote collagen production by dermal fibroblasts of SSc patients (14). IL-6–producing Beffs could infiltrate the inflamed skin tissue and promote collagen secretion by fibroblasts, although the phenotype of IL-6–producing Beffs in the skin was not consistent with that in the spleen. In contrast, the current study showed that IL-10–producing Bregs were increased in inflamed skin. Therefore, Bregs may infiltrate into the inflamed skin to attenuate the inflammation. The current study also revealed that IL-6–producing Beffs promote collagen secretion by fibroblasts through interaction with B cells and fibroblasts. B cells without stimulation did not produce IL-6; however, B cells cocultured with fibroblasts did produce IL-6. Thus, cell-cell contact with fibroblasts induces IL-6 production from B cells and augments collagen secretion by fibroblasts. Since these effects were inhibited when we used Transwells, additional factors, such as adhesion molecules, may play a role in these effects. In addition, IL-6 production was increased in coculture with B cells from IL-10−/− mice, resulting in increased collagen secretion by fibroblasts. Collectively, these results suggest that IL-6 and IL-10 secretion from B cells promotes and inhibits the fibrosis in SSc, respectively.

BAFF plays a critical role in the survival, maturation, and activation of B cells (21), and the serum BAFF levels are elevated in patients with various diseases, including SSc (13). B cells from SSc patients stimulated by BAFF exhibited enhanced ability to produce IL-6 (13). Consistent with these previous findings, in the present study, BAFF increased the numbers of IL-6–producing Beffs but attenuated IL-10–producing Bregs. However, Yang et al. (29) reported that BAFF enhanced IL-10 production of B cells from DBA/1J mice. This discrepancy may be due to the different mouse genetic backgrounds. By contrast, BAFF inhibition decreased IL-6–producing Beffs, but not IL-10–producing Bregs. Similarly, BAFF receptor–deficient mice showed depletion of B2 but not B1a B cells, a major subset of IL-10–producing Bregs (30). BAFF inhibition attenuated the skin fibrosis of the present SSc mouse model, which confirms the results of a previous study (15).

There has been some success with respect to targeting B cells for SSc therapy. B cell depletion therapy with rituximab, a CD20 mAb that depletes human pan-B cells, has shown beneficial effects on skin and lung fibrosis in patients with SSc (31, 32); however, a phase 3 randomized controlled study is required to confirm the efficacy and safety of rituximab for the treatment of SSc. Two large randomized controlled trials of rituximab have been conducted in patients with SLE, with the expectation that it would be effective; however, these trials failed to achieve the primary end points (33, 34). Our present findings suggest that this ineffectiveness may have been due to depletion of not only Beffs but also Bregs. Thus, the outcome of pan-B cell depletion depends on the balance between Beffs and Bregs in a given patient. Moreover, the effect of B cell depletion was shown to be dependent on the timing and balance of the two opposing B cell functions in mouse models of SLE and MS (6, 35). By contrast, phase 3 clinical trials demonstrated the efficacy of belimumab, an anti-BAFF Ab, in patients with SLE (36, 37); belimumab has been approved by the U.S. Food and Drug Administration. Our results further shed light on the most likely reason for the superior effects of partial B cell depletion with BAFF inhibition compared to pan-B cell depletion with a CD20 mAb, given that BAFF inhibition therapy selectively depletes Beffs while sparing Bregs. However, a phase 2 study of atacicept, an inhibitor of BAFF and a proliferation-inducing ligand, did not reveal efficacy in the patients with MS (38). The effect of selective B cell depletion is not always beneficial, and it may depend on the contribution of B cells to the particular disease. BAFF inhibition therapy, rather than pan-B cell depletion, could be a potent therapeutic strategy for SSc (Fig. 6G).

Here, we have investigated the intracellular staining and phenotype of IL-6–producing Beffs. IL-6–producing Beffs play a pathogenic role in scleroderma, whereas IL-10–producing Bregs play a protective role. Furthermore, BAFF contributes to SSc pathogenesis by modulating cytokine-producing B cells. Thus, our study reveals that BAFF inhibition is a potential therapeutic strategy for SSc via alteration of the Beff and Breg balance.

MATERIALS AND METHODS

Study design

We performed this study to determine whether cytokine-producing B cells control the development of scleroderma in a mouse model. To establish the model, we treated B cell–specific cytokine-deficient mice with bleomycin. The skin and lung fibrosis of the bleomycin-induced scleroderma model was attenuated in B cell–specific IL-6–deficient mice, while B cell–specific IL-10–deficient mice showed more severe skin and lung fibrosis. Subsequent histological analysis confirmed these findings. Sample sizes and end points were selected on the basis of our extensive experience with these systems. In selected experiments, the mice were randomly assigned to treatment groups, and the researchers were blinded to the treatment group during experimental procedures and raw data analysis. All animal experiments were performed according to institutionally approved protocols and in compliance with the guidelines of the Committee on Animal Experimentation of the Kanazawa University Graduate School of Medical Sciences. No animals or potential outliers were excluded from the data sets analyzed and presented herein. All in vitro studies were performed in replicates (n = 3, unless otherwise specified).

Mice

Wild-type C57BL/6 mice, Il10−/− mice, and μMT mice were obtained from the Jackson Laboratory. Il6−/− mice were generated as previously reported (39). All mice were on the C57BL/6 background. For experiments, all mice used were 8 to 10 weeks of age and housed in a specific pathogen–free barrier facility.

Generation of mixed bone marrow chimeras

Mice with B cell–specific IL-6 deficiency or B cell–specific IL-10 deficiency were generated using the mixed bone marrow chimera system, as described previously (3, 10). Briefly, recipient wild-type mice received 1000 cGy of x-ray irradiation. One day later, the recipients were reconstituted with a mixed inoculum of 80% μMT bone marrow cells supplemented with 20% bone marrow cells from Il6−/− or Il10−/− mice (a total of 2 × 106 cells). Control groups received 80% wild-type and 20% bone marrow cells from Il6−/− or Il10−/− mice (a total of 2 × 106 cells). Chimeric mice were left to fully reconstitute their lymphoid system for at least 8 to 10 weeks before bleomycin treatment. Chimerism was confirmed by B cell cytokine production using ELISA. Characterization of chimeras is outlined in fig. S5.

Bleomycin-induced scleroderma model

Bleomycin (Nippon Kayaku) was dissolved in sterile saline at a concentration of 1 mg/ml. The mice were treated with intradermal injections of either bleomycin or saline (300 μl; administered using a 27-gauge needle) into their shaved backs (the para-midline, lower back region) every other day for 4 weeks.

BAFFR-Fc treatment

In vivo treatment murine BAFFR-Fc Chimera (BioLegend), which were made by fusing their extracellular domains to the Fc portion of human IgG1 and neutralize murine BAFF, and Fc control protein (BioLegend) were used in this study. To neutralize BAFF in vivo, mice received either murine BAFFR-Fc (2.5 mg/g, intraperitoneally, three times per week) or the same dose of Fc control protein.

Determination of collagen content

The collagen content of the culture supernatant was determined using QuickZyme Soluble Collagen Assay (QuickZyme Biosciences), according to the manufacturer’s instructions. The collagen content of the mouse and lung tissues was determined using QuickZyme Total Collagen Assay (QuickZyme Biosciences), according to the manufacturer’s instructions. Total right lungs were used.

Histological examination of skin and lung fibrosis

All skin sections were obtained from the bleomycin-injected region of the lower back, as full-thickness sections extending down to the body wall musculature. Lung sections were obtained from the bleomycin-induced scleroderma model. The skin and lung samples were fixed in formalin, dehydrated, embedded in paraffin, and used for immunostaining. Sections (6 μm thick) were stained with H&E and Masson’s trichrome to identify collagen deposition in the skin and lung. Dermal thickness, which was defined as the thickness of skin from the top of the granular layer to the junction between the dermis and intradermal fat, was evaluated. The severity of lung inflammation was determined by a semiquantitative scoring system, as previously described (40). Briefly, lung fibrosis in randomly chosen fields of sections from the left middle lobe examined at 100× magnification was graded on a scale of 0 (normal lung) to 8 (total fibrous obliteration of fields). All sections were evaluated independently by two investigators (Kie Mizumaki and Miyu Kano), in a blinded manner.

Flow cytometry and intracellular cytokine staining analysis

Single-cell leukocyte suspensions from spleens were generated by gentle dissection. The following mAbs were used: fluorescein isothiocyanate–, PE (phycoerythrin)–, PE-Cy5–, PE-Cy7–, PerCP-Cy5.5–, APC (allophycocyanin)–, APC-PECy7–, and BV421-conjugated mAbs to mouse CD4 (RM4-5), CD8 (53-6.7), CD11b (M1-70), CD19 (1D3), CD25 (MF-14), CD44 (IM7), IL-10 (MP5-20F3), and IL-10 (JES5-16E3), using LEGENDScreen Mouse PE Kit from BioLegend; CD1d (1B1), CD5 (53-7.3), CD21/CD35 (7G6), CD23 (B3B4), CD24 (M1/69), and CD45R/B220 (RA3-6B2) from BD Biosciences; and BAFF (121808) from R&D Systems. For two- to six-color immunofluorescence analysis, single-cell suspensions (106 cells) were stained at 4°C using predetermined optimal concentrations of mAb for 20 min. Blood erythrocytes were lysed after staining using FACS Lysing Solution (Becton Dickinson). For intracellular staining, cells were fixed and permeabilized with a Cytofix/Cytoperm kit (BD Biosciences). Dead cells were detected by using LIVE/DEAD Fixable Aqua Dead Cell Stain Kit (Invitrogen Molecular Probes) before cell surface staining. Cells with the forward and side light scatter properties of lymphocytes were analyzed using a BD FACSCanto II (BD Biosciences). Data were analyzed with FlowJo software (version 10.2; Tree Star).

B cell stimulation

Splenic B cells were purified with anti-CD19 mAb–coated microbeads or the Pan B cell isolation kit (Miltenyi Biotec) or by means of cell sorting with the FACSAria Fusion (BD Bioscience). B cells (2 × 106 cells/ml) were resuspended in complete medium [RPMI 1640 media containing 10% fetal bovine serum, penicillin (200 μg/ml), streptomycin (200 U/ml), 4 mM l-glutamine, and 5 × 10−5 M 2-mercaptoethanol (all from Gibco)]. B cells were stimulated with an agonistic anti-CD40 mAb (10 μg/ml; FGK45, Enzo Life Sciences), BAFF (100 ng/ml; R&D Systems), LPS (10 μg/ml; Escherichia coli serotype 0111: B4, Sigma-Aldrich), or other TLR agonists (TLR1, Pam3CSK4, 300 ng/ml; TLR2, heat-killed Listeria monocytogenes, 108 cells/ml; TLR3, polyriboinosinic acid/polyribocytidylic acid, 10 μg/ml; TLR5, Salmonella Typhimurium flagellin, 1 μg/ml; TLR6, Pam2CGDPKHPKSF, 100 ng/ml; TLR7, ssRNA40/LyoVec, 5 μg/ml; and TLR9, ODN1826, 5 μM; Invivogen).

Enzyme-linked immunosorbent assay

IL-6, IL-10, or BAFF levels were determined by specific ELISA kits (R&D Systems)

Intracellular cytokine staining

Intracellular cytokine expression was visualized by immunofluorescence staining and analyzed by flow cytometry as previously described. For IL-6 detection, cells (2 × 106 cells/ml) were cultured with LPS (10 μg/ml) and anti-CD40 mAb (10 μg/ml; FGK45) for 24 hours with PIB [PMA (50 ng/ml; Sigma-Aldrich), ionomycin (1 μg/ml; Sigma-Aldrich), and brefeldin A (supplied as 1000× solution; c)] added during the final 5 hours of culture. PIB [PMA (50 ng/ml; Sigma-Aldrich), ionomycin (1 μg/ml; Sigma-Aldrich), and brefeldin A (supplied as 1000× solution; BioLegend)] was added for 5 hours. For IL-10 detection, cells (2 × 106 cells/ml) were cultured with LPS (10 μg/ml) and PIB for 5 hours. Fc receptors were blocked with mouse Fc receptor mAb (2.4G2, BD Pharmingen) with dead cells detected using a LIVE/DEAD Fixable Aqua Dead Cell Stain Kit (Invitrogen) before cell surface staining. Stained cells were fixed and permeabilized using a Cytofix/Cytoperm kit (BD Pharmingen) according to the manufacturer’s instructions and stained with APC-conjugated mouse anti–IL-6 or anti–IL-10 mAb.

Preparation of skin cell suspensions for flow cytometry

A 1 cm × 1 cm piece of the bleomycin-injected skin region was minced and then digested in 7 ml of RPMI 1640 and 10% fetal bovine serum containing collagenase (2 mg/ml; Sigma-Aldrich), hyaluronidase (1.5 mg/ml; Sigma-Aldrich), and deoxyribonuclease (DNase) I (0.03 mg/ml; Sigma-Aldrich) at 37°C for 90 min. Digested cells were then passed through a 70-μm cell Falcon Cell Strainer (BD Biosciences) to generate single-cell suspensions. The cell suspension was centrifuged at 300g for 10 min. The pellet was resuspended in 70% Percoll solution (GE Healthcare) and then overlaid by 37% Percoll solution followed by centrifugation at 500g for 20 min at room temperature. Cells were aspirated from the Percoll interface and passed through a 70-μm cell strainer. Subsequently, the cells were harvested by centrifugation and washed.

Immunohistochemical staining of mouse skin

Mice were treated with bleomycin for 2 weeks, and brefeldin A (250 μg per mouse; Sigma-Aldrich) was intravenouslly injected 6 hours before sample prepararion. The skin samples from bleomycin-injected mice were removed and frozen in liquid nitrogen using embedding medium for frozen tissue specimens [Tissue-Tek OCT (optimum cutting temperature) Compound, Sakura Finetek] and stored at −70°C until use. Frozen sections (5 μm thick) were immediately fixed in cold acetone and were incubated with rat anti-mouse B220 mAb (RA3-6B2 clone, BD Biosciences) and polyclonal goat anti-mouse IL-6 Ab (R&D Systems). Donkey anti-rat IgG with Alexa Fluor 488 or donkey anti-goat IgG with Alexa Fluor 594 (Thermo Fisher Scientific) were used as secondary Abs of rat anti-mouse B220 mAb or goat anti-mouse IL-6 mAb, respectively. Coverslips were mounted by using ProLong Diamond Antifade Mountant with DAPI (Thermo Fisher Scientific). Fluorescence microscopy was performed using a KEYENCE BZ-X710 fluorescence microscope (KEYENCE).

Fibroblast culture

Skin samples were taken from E14.5 embryos of naïve wild-type mice. To obtain fibroblasts, the skin tissue was cut into 1-mm3 pieces, placed in sterile plastic dishes, and cultured in Dulbecco’s modified Eagle’s medium (Invitrogen) containing 10% heat-inactivated fetal bovine serum, penicillin (100 U/ml), and streptomycin (100 μg/ml; Invitrogen) at 37°C in a humidified 5% CO2 atmosphere. After 2 to 3 weeks of incubation, outgrowing fibroblasts were detached by brief trypsin treatment and recultured in the medium. Fibroblasts (105 cells) were cultured without or with B cells (5 × 105 cells) in 24-well plates for 3 days. For Transwell experiments, B cells (5 × 105 cells) and fibroblasts (105 cells) were seeded in the upper and lower chambers, respectively, of a 0.4-μm polycarbonate membrane Transwell (Nunc Dominique Dutscher). For fibroblast stimulation, cells were cultured with recombinant TGF-β1 (5 ng/ml; R&D Systems). The supernatant was harvested after culture. All experiments used fibroblasts between passages 2 and 5, depending on the number of cells obtained initially from the tissue samples. Cultured fibroblasts were adherent to the dish and maintained the typical spindle-shaped aspect. The purity of fibroblasts, as confirmed by flow cytometry, was >99%, with no leukocytes found in the harvested cells. In each experiment, all the cell lines were examined at the same time and under the same conditions of culture (for example, cell density, passage, and days after plating).

Reverse transcription polymerase chain reaction

Total RNA was isolated from inflamed skin using RNeasy spin columns (Qiagen) and digested with DNase I (Qiagen) to remove chromosomal DNA. Total RNA was reverse-transcribed to a cDNA using a reverse transcription system with random hexamers (Promega). Cytokine mRNA was analyzed using real-time reverse transcription PCR (RT-PCR) quantification (Applied Biosystems). Real-time RT-PCR was performed on an ABI Prism 7000 sequence detector (Applied Biosystems). TaqMan probes and primers for collagen alpha-2(I) (col1a2) and glyceraldehyde-3-phosphate dehydrogenase (gapdh) were purchased from Applied Biosystems. GAPDH was used to normalize the mRNA. The relative expression of real-time RT-PCR products was determined according to the ΔΔCt method to compare target gene and GAPDH mRNA expression.

Statistical analysis

Data are presented as means ± SD. Two-tailed Student’s t test was used for comparisons between two groups, and P < 0.05 was considered significant. Comparisons among three or more groups were performed with ANOVA followed by Tukey’s multiple comparison test. Data were analyzed with GraphPad Prism (version 7; GraphPad Software).

Study approval

Animal studies were approved by the Committee on Animal Experimentation of the Kanazawa University Graduate School of Medical Sciences.

SUPPLEMENTARY MATERIALS

Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/7/eaas9944/DC1

Fig. S1. Phenotypes of IL-6–producing B cells after in vitro culture.

Fig. S2. CD25 expression on B cell is enhanced after stimulation.

Fig. S3. IL-6– and/or IL-10–producing B cells.

Fig. S4. Phenotype of IL-6–producing B cells in the skin.

Fig. S5. Scheme for the generation of B-IL-6−/− or B-IL-10−/− and corresponding control mice.

Fig. S6. Lung fibrosis is attenuated in mice with B cell–specific IL-6 deficiency.

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.

REFERENCES AND NOTES

Acknowledgments: We thank M. Matsubara, Y. Yamada, and Y. Iwauchi for technical assistance. Funding: This work was supported by Japan Society for the Promotion of Science KAKENHI (grant no. JP16K10147). Author contributions: T.M., T.K., K.M., M.K., T.S., M.T., and A.O. contributed to data collection, analysis, and interpretation. Y.I. generated IL-6–deficient mice. T.M., Y.H., M.H., M.F., and K.T. designed the study and wrote the manuscript. All authors discussed the results and commented on the manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.
View Abstract

Navigate This Article