Defective early B cell tolerance checkpoints in patients with systemic sclerosis allow the production of self-antigen-specific clones

Abstract

Objective:

Early selection steps preventing autoreactive naïve B cell production are often impaired in patients with autoimmune diseases, but central and peripheral B cell tolerance checkpoints have not been assessed in patients with systemic sclerosis (SSc).

Methods:

Using an in-vitro PCR-based approach that allows the expression of recombinant antibodies cloned from single B cells, we tested the reactivity of antibodies expressed by 212 CD19⁺CD21^−/loCD10⁺IgM^hiCD27⁻ new emigrant/transitional and 190 CD19⁺CD21⁺CD10⁻IgM⁺CD27⁻ mature naïve B cells from ten patients with SSc.

Results:

Patients with SSc displayed elevated proportions of polyreactive and anti-nuclear new emigrant/transitional B cells that recognize topoisomerase I when compared to healthy donors, suggesting defective central B cell tolerance contributes to the production of serum autoantibodies characteristic of the disease. Frequencies of autoreactive mature naïve B cells were also significantly increased in SSc patients and revealed an impaired peripheral B cell tolerance checkpoint.

Conclusion:

Defective counterselection of developing autoreactive naïve B cells in SSc leads to the production of self-antigen specific B cells that may secrete autoantibodies and allow the formation of immune complexes, which promote fibrosis in SSc.

Scleroderma or Systemic sclerosis (SSc) is an autoimmune disease characterized by vascular abnormalities, fibroblast activation leading to extracellular matrix synthesis and fibrosis of the skin and internal organs, and dysregulated immunity (1, 2). Patients with SSc are classified into two main groups—diffuse and limited SSc. Diffuse SSc (dSSc) is characterized by skin fibrosis proximal to the elbows and knees and internal organ damage, especially to the lungs, resulting in pulmonary fibrosis, a major cause of disease-associated morbidity and mortality (1, 2). In contrast, patients with limited SSc (lSSc) usually suffer from skin alterations restricted to the hands and face and are less commonly affected by visceral fibrosis.

These two subgroups of patients with SSc are also characterized by the production of specific autoantibodies. Anti-topoisomerase 1/anti-SCL-70, anti-RNA polymerase III and anti-U3 RNP are usually found in patients with dSSc, whereas anti-centromere, anti-Pm/Scl, anti-Th/To, and anti-U1 RNP are often associated with lSSc (1, 2). All these autoantibodies that target nucleic acid containing self-antigens demonstrate a break in B cell tolerance in SSc; however, the origin of the autoreactive B cells secreting these serum autoantibodies remains unknown.

Analysis of patients with various autoimmune diseases, including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), Sjögren’s syndrome (SjS), type 1 diabetes (T1D), myasthenia gravis (MG), neuromyelitis optica spectrum disease (NMOSD), and multiple sclerosis (MS) revealed an impaired selection of developing autoreactive B cells in either (or both) the bone marrow and the periphery, resulting in the accumulation of large numbers of circulating autoreactive naïve B cells (3).

We report herein that both dSSc and lSSc patients display defective central and peripheral B cell tolerance checkpoints, leading to the production of autoreactive naïve clones expressing unmutated antibodies with self-antigen specificity characteristic of the disease. These autoreactive antibodies through the formation of immune complexes have recently been shown to promote fibrosis in SSc (4), thereby revealing that failed B cell tolerance mechansims play an essential role in disease pathophysiology.

PATIENTS AND METHODS

Patients with SSc according to current criteria were enrolled from the Yale ILD Center of Excellence and the Yale Scleroderma Center (Table S1). Most patients were naïve of any medication and all met diagnostic criteria for SSc (1). Characteristics of patients with SSc, their autoantibody profiles, and the presence of the 1858T PTPN22 polymorphism associated with the disease and leading to impaired early B cell tolerance checkpoints are summarized in Supplementary Table 1 (5-9). The study protocol was approved by the Institutional Review Board at Yale (HIC#1307012431, HIC#0906005336), and informed consent was obtained from all patients before participation.

Cell staining and sorting

Mononuclear cells from healthy donors and patients with SSc were enriched for B cells by magnetic separation with anti-CD20 microbeads (Miltenyi Biotech) and stained with anti-human CD19-Pacific Blue, anti-human CD27-PerCP Cy5.5, anti-human CD10-PE-Cy7, anti-human CD21-APC, anti-human IgM-FITC (Biolegend). Single CD19⁺CD21^lowCD10⁺IgM^hiCD27⁻ new emigrant/transitional and CD19⁺CD21⁺CD10⁻IgM⁺CD27⁻ mature naïve B cells were sorted on a FACSAria (BD Biosciences) into 96-well PCR plates and immediately frozen on dry ice.

The following antibodies were used for T cell phenotyping: anti-CD4 APC-Cy7, anti-CD25 PE-Cy7, anti-CD127 PerCP-Cy5.5 (all from Biolegend), and anti-CD3 eFluor 605NC (eBioscience). Intracellular staining with anti-FOXP3 Alexa Fluor 488 (clone PCH101; eBioscience) was performed using the FOXP3/Transcription Factor Staining Buffer Set in accordance with the manufacturer’s instructions (eBioscience).

cDNA synthesis, Ig genes amplification, antibody production, and purification

cDNA synthesis, RT-PCR reactions, primer sequences, cloning strategy, expression vectors, in vitro recombinant antibody production and purification were performed as previously described (10). In brief, cDNA was synthesized in the original 96-well PCR plate in which single B cells were sorted. RNA from single cells was reverse-transcribed in a 14μL volume at 37°C for 55 minutes. 3.5μL of cDNA or 1st-PCR product were used to amplify IgH, Igκ or Igλ transcripts by two rounds of PCR in 40μL reactions containing 20pM primers and 1.2U Hotstar Taq DNA polymerase (Qiagen). PCR products were then purified (Qiaquick, Qiagen), sequenced and analyzed by Ig BLAST comparison with GenBank. Since gene restriction sites were introduced by 2nd PCRs, digested IgH, Igκ and Igλ PCR products were purified (Qiaquick, Qiagen) and cloned into expression vectors containing human IgG1, Igκ or Igλ constant regions (10). 12.5μg of IgH and IgL chain encoding plasmid DNA was co-transfected with polyethylenimine in 293A human embryonic kidney fibroblasts washed with serum-free DMEM and thereafter cultured in DMEM supplemented with 1% Nutridoma SP (Roche). Supernatants were collected and titrated after 8-10 days of culture. HEp-2-reactivity ELISAs and IFAs were performed using antibodies purified on protein G Sepharose™ (Amersham Pharmacia Biosciences).

Repertoire analysis

Immunoglobulin sequences and mutation status were determined using Ig BLAST comparison with GenBank using the National Center for Biotechnology Information IgBlast server (http://www.ncbi.nlm.nih.gov/igblast/). Heavy chain complementarity determining region 3 was defined as the interval between amino acid at position 94 in the V_H framework 3 and the conserved tryptophan at position 103 in JH segments. Antibody sequences and reactivity are shown in Supplementary Tables 2 and 3.

ELISAs and IFAs

Recombinant antibody reactivity was assessed as previously reported using the highly polyreactive ED38 recombinant antibody as a positive control for HEp-2 reactivity and polyreactivity assays (10). Antibodies were considered polyreactive when they recognized all 3 distinct antigens: double-stranded (ds) DNA, insulin, and lipopolysaccharide (LPS). For indirect immunofluorescence assays, HEp-2 cell-coated slides (Bion Enterprises Ltd.) were incubated in a moist chamber at room temperature with purified recombinant antibodies at 50-100 μg/mL according to the manufacturer’s instructions.

Immunoprecipitation and Western blots

DLD1 cells grown to confluence in a T-25 flask (2-3 x 10⁶ cells) were lysed in M-PER (78501, ThermoFisher) for 5min on ice with 1X HALT Protease Inhibitor (87786, ThermoFisher). The lysate was clarified by centrifugation for 10min at 10,000 rpm at 4°C. 200μl of lysate was added to Protein A/G beads with 2μg K24 or the equivalent volume of PBS for 1hr rotation at 4°C. Beads were washed three times in Gentle Ag/Ab Binding Buffer pH 8.0 and eluted by the addition of 2X Laemmli buffer. Western blot of K24 input, DLD1 total lysate, and elutions from control and K24 beads was probed with primary monoclonal anti-topoisomerase I (sc-271285, Santa Cruz Biotechnology, Dallas, TX) and secondary HRP-linked anti-mouse IgG antibody (7076S, Cell Signaling Technology, Danvers, MA), with signal detection by ECL (ThermoFisher).

In vitro Treg suppression assay

CD4⁺ T cells were enriched using the EasySep Human CD4⁺T cell enrichment kit (STEMCELL Technologies). CD4⁺CD25^hiCD127^lo/− Tregs and CD3⁺CD4⁺CD25⁻ T responder (Tresp) cells were sorted by flow cytometry. Tresp cells were then labeled with CellTrace CFSE (InvivoGen) at 5 μM. Treg and Tresp cells were cocultured at a 1:1 ratio in the presence of beads loaded with anti-CD2, anti-CD3, and anti-CD28 (Treg suppression inspector human; Miltenyi Biotec) at a 1bead/cell ratio (11). On days 3.5 to 4.5, cocultures were stained for viability with the LIVE/DEAD kit (Invitrogen), and proliferation of the viable Tresp was assessed following CFSE dilution.

Statistical analysis

Statistical analysis was performed using GraphPad Prism (version 5.0; GraphPad, San Diego, CA). Differences between subject groups were tested for statistical significance with nonparametric Mann Whitney U tests or unpaired Z-tests. p-values ≤ 0.05 were considered significant.

RESULTS

Impaired central B cell tolerance in patients with SSc

Central B cell tolerance mediates the removal of developing polyreactive and anti-nuclear reactive immature B cells in the bone marrow (3, 10). To determine whether this early B-cell selection step is altered in SSc, we enrolled patients with SSc, many of which were treatment naïve (Table 1 and Table S1). We cloned and tested the reactivity by ELISA of 212 recombinant antibodies expressed by single CD19⁺CD21^loCD10⁺IgM^hiCD27⁻ transitional B cells that recently emigrated from the bone marrow and were isolated from ten of these patients (five with diffuse SSc and five with the limited form of the disease). Of note, flow cytometry analysis revealed that transitional B cell subsets including early T1 and more mature T2 were found at similar frequencies between healthy donors and patients with SSc (Supplementary Figure 1). In contrast, SSc patients displayed significantly increased proportions of CD19⁺CD21⁺CD10⁻IgM⁺CD27⁻ mature naïve B cells that differentiate from transitional B cells but decreased frequencies of CD19⁺CD21⁺CD10⁻CD27⁺ conventional memory B cells compared to controls (Supplementary Figure 1).

Table 1.

Characteristics of patients with SSc

Patients	dSSc (n=5)	lSSc (n=5)
Age	45.4 + 15.6	51.6 + 9.8
Gender
Female	5 (100%)	5 (100%)
Race
White	5 (100%)	5 (100%)
Anti-Scl70
Positive	4 (80%)	1 (20%)
Negative	1 (20%)	4 (80%)
Anti-centromere
Positive	1 (20%)	2 (40%)
Negative	4 (80%)	3 (60%)
Treatment naive	3 (60%)	4 (80%)
Clinically significant ILD	5 (100%)	3 (60%)
Group I PAH	0 (0%)	0 (0%)
GERD	5 (100%)	5 (100%)
MRSS	15	5

Data presented as mean +/− SD. ILD: interstitial lung disease. PAH: Pulmonary arterial hypertension. GERD: Gastroesophageal reflux disease. MRSS: Modified Rodnan Skin Score.

The first evidence suggesting that central B cell tolerance was not properly established in SSc came from the Ig repertoire analysis of their new emigrant/transitional B cells (Supplementary Figure 2 and Supplementary Table 2). Pooled heavy chain gene (IgH) sequences from their new emigrant/transitional B cells revealed a significantly higher frequency of long IgH complementarity determining regions 3 (CDR3s), a feature that favors antibody self-reactivity (10, 12), whereas the density of positive charges in IgH CDR3 was similar between healthy donors and patients (Supplementary Figure 2). The reactivities of antibodies expressed by new emigrant/transitional B cells from 10 patients with SSc were then compared to 13 healthy donor controls (HD) previously studied (Figure 1A-C) (3). We found that the frequencies of new emigrant/transitional B cells expressing polyreactive antibodies in both patients with diffuse and limited SSc were significantly increased and averaged 22.1% and 26.9%, respectively, compared to only 7.1 % in HD controls, revealing that central B cell tolerance is impaired in SSc (P < 0.0001; Figure 1A and B, Supplementary Figure 3 and Supplementary Table 2). The proportion of anti-nuclear clones in new emigrant/transitional B cells from SSc patients was also significantly increased compared to healthy donors, further demonstrating the impaired removal of developing autoreactive B cells in the bone marrow of these patients (P = 0.036, Figure 1C and D). Anti-nuclear reactive new emigrant/transitional B cells from patients with SSc recognized diverse structures in the nucleus, as illustrated by the different anti-nuclear staining patterns shown for clones neSSc065 κ28, neSSc210 κ24, and neSSc114 κ45 (Figure 1D). Both polyreactive and anti-nuclear reactive new emigrant B cells from patients with SSc were enriched in clones that displayed positively charged amino acids in their IgH CDR3s, whereas long IgH CDR3 favored polyreactivity but not anti-nuclear reactivity (Supplementary Figure 2). Of note, similar autoreactive B cell frequencies were observed in patients with SSc whether or not they harbored PTPN22 risk allele which is associated with this disease and results in impaired early B cell tolerance checkpoints (Supplementary Figure 2) (5-9). Hence, other polymorphisms or factors may result in altered autoreactive B cell counterselection in these patients. Altogether, our data show that central B cell tolerance is defective in both dSSc and lSSc patients.

Figure 1. Defective central B cell tolerance in SSc patients.

(A) Antibodies cloned from single new emigrant/transitional B cells from a representative healthy donor (HD10), patients with diffuse (dSSc114) or limited (lSSc368) SSc, were tested by ELISA for reactivity against double-stranded DNA, insulin, and LPS. Dotted lines show ED38-positive control. Horizontal lines show cutoff OD405 for positive reactivity. The frequencies of polyreactive and non-polyreactive clones are summarized in pie charts, with the number of antibodies tested indicated in the centers. Frequencies of polyreactive (B) and antinuclear (C) new emigrant B cells were compared between 13 HD and 10 patients with SSc (5 dSS and 5 lSSc that were also represented separately), and statistically significant differences are indicated (Mann–Whitney U test; *P < 0.05; ***P < 0.001; ****P < 0.0001). Each diamond represents an individual, and the median is shown with a bar. (D) Anti-nuclear antibodies show various patterns of HEp-2 staining. Original magnification, x 40. (E) Topoisomerase I western blot for K24 input supernatant (first lane), DLD1 total lysate (second lane); no antibody or neSSc210 K24 included for the immunoprecipitation (third and fouth lanes). Results shown are representative of three experiments and demonstrate that neSSc210 K24 binds topoisomerase I.

Identification of an anti-topoisomerase I/Scl-70 reactive clone in a patient with dSSc

neSSc210 κ24 was of particular interest due to its strong, compact fine speckled pattern since it resembled staining characteristic of anti-topoisomerase I/Scl-70, suggesting that this clone may recognize this self-antigen targeted in dSSc (Figure 1D) (13). To determine whether neSSc210 κ24 is an anti-topoisomerase I reactive antibody, we performed immunoprecipitation experiments with this recombinant antibody using the DLD1 cell line. Indeed, the presence of topoisomerase I was revealed using a monoclonal anti-human topoisomerase I antibody (Figure 1E). We found that neSSc210 κ24 bound topoisomerase I, the major self-antigen in dSSc, in the absence of somatic hypermutation, which normally improves antibody affinity (Figure 1E). Thus, defects in central B cell tolerance in patients with SSc may result in the production of autoreactive clones that recognize self-antigen specifically targeted in this disease.

Defective peripheral B cell tolerance checkpoint in patients with SSc.

Autoreactive B cells that recognize peripheral self-antigens are normally eliminated at a second B cell tolerance checkpoint before entering the long-lived mature naïve B cell compartment (3, 10). As a consequence, mature naïve B cells from both healthy donors and patients with SSc displayed shorter and less positively charged IgH CDR3s than new emigrant/transitional B cell counterparts (Supplementary Figures 2 and 5). However, mature naïve B cells from patients with SSc were significantly enriched in clones with longer IgH CDR3s compared to counterparts in healthy donors, suggesting that the peripheral B cell tolerance checkpoint may not be properly regulated in SSc (Supplementary Figure 5A). We therefore investigated this peripheral B cell selection step by testing the self-reactivity of antibodies expressed by 190 mature naïve B cells from the same 10 patients with SSc using various ELISAs and indirect immunofluorescence (10). We found that the frequencies of mature naïve B cells expressing antibodies reactive to Human Epithelial type 2 (HEp-2) cell lysate, a commonly used ELISA for the detection of autoreactive immunoglobulins (10), were significantly increased in both dSSc and lSSc patients (42.9%–56.5%) compared with healthy donors (16.0%–26.3%, P < 0.0001) (Figure 2A, Supplementary Figure 6 and Supplementary Table 3). Peripheral B cell tolerance checkpoint defects in patients with SSc were further evidenced by the increased frequency of mature naïve B cells expressing polyreactive antibodies compared with HD (P = 0.0004), whether or not they carried the PTPN22 risk allele (Figure 2B and Supplementary Figures 4 and 7). The proportion of anti-nuclear clones in mature naïve B cells from patients with SSc were also increased compared to HD, but differences failed to reach significance (Figure 2C and D). Anti-nuclear clones showed nucleolar (mnSSc232 κ64 and mnSSc362 κ49) or speckled (mnSSc210 κ65) staining patterns (Figure 2D). Remarkably, the serum of patient SSc232 displayed anti-nuclear autoantibodies (Table S1), suggesting that the impaired peripheral B cell tolerance checkpoint in SSc may also contribute to the break in B cell tolerance and autoantibody secretion in this disease. Analysis of heavy chain sequences from mature naïve B cells revealed that long IgH CDR3s favored polyreactivity, whereas the presence of positively charged amino acids was associated with HEp-2 reactivity, polyreactivity, and anti-nuclear reactivity (Supplementary Figure 5). Since the regulation of the peripheral B cell tolerance checkpoint involves T cells and potentially regulatory T cells (Tregs) (14-17), we assessed Treg frequency and suppressive function in patients with SSc (Figure 3). We found that the proportion of Tregs in the blood of 16 patients with SSc was globally similar and potentially increased in patients with dSSc compared to HDs as previously reported (18, 19) (Figure 3A and B). HD and SSc patient Tregs similarly suppressed in vitro the induced proliferation of CD4⁺CD127⁺ T responder cells (Tresp) isolated from either HDs or patients with SSc (Figure 3C and D). In contrast, we observed that Tresp from patients with SSc, which included dSSc210, dSSc366, lSSc368, lSSc369, who displayed defective autoreactive B cell selection, were refractory to suppression by either autologous Tregs isolated from the same patients or heterologous Tregs isolated from HDs (Figure 3D and E). Hence, altered T cell responses in patients with SSc correlate with the impaired selection of autoreactive B cells in the periphery. B cell–activating factor (BAFF) is a molecule that controls the number of peripheral B cells and may interfere with peripheral B cell tolerance (20). We found that patients with SSc displayed serum BAFF concentrations similar to those in healthy donors, suggesting that BAFF is not responsible for peripheral B cell selection defects in SSc (Supplementary Figure 8). We conclude that patients with SSc suffer from a dysfunctional peripheral B cell tolerance checkpoint.

Figure 2. Impaired peripheral B cell tolerance checkpoint in SSc patients.

(A) Antibodies cloned from single mature naïve B cells from a representative healthy donor (HD10) and patients dSSc114 and lSSc368 were tested by ELISA for reactivity against HEp-2 cell lysate. Dotted lines show ED38-positive control. Horizontal lines show cutoff OD405 for positive reactivity. The frequencies of HEp-2-reactive (in black) and non-HEp-2-reactive (in white) clones are summarized in pie charts with the number of antibodies tested indicated in the centers. The frequencies of HEp-2-reactive (A, right), polyreactive tested against dsDNA, insulin, and LPS (B), and antinuclear (C) clones in mature naïve B cells of 10 patients with SSc (5 dSS and 5 lSSc that were also represented separately), were compared with those in 13 HDs. Each diamond represents an individual, and the median is shown with a bar. Statistically significant differences between patients and HD are indicated (Mann–Whitney U test; *P < 0.05; ***P < 0.001; ****P < 0.0001). (D) Examples of mature naïve B cells from patients with SSc that expressed antibodies recognizing either cytoplasmic (mnSSc210 λ81, mnSSc369 κ15, and mnSSc114 κ62) or nuclear (mnSSc232 κ64, mnSSc210 κ65, and mnSSc362 κ49) structures. Original magnification, x 40.

Figure 3. Patients with SSc display T cells refractory to in vitro Treg suppression.

(A) CD25 vs CD127 and CD25 vs intracellular FOXP3 staining on CD3⁺CD4⁺ T cells from a representative healthy donor (HD), a patient with diffuse SSc (dSSc) or limited SSc (lSSc). (B) Summary of CD3⁺CD4⁺CD25^hiCD127^loFOXP3⁺ Treg frequencies in 44 HD and 16 patients with SSc (7dSSc and 9 lSSc). Bars represent the median. P values are shown when significant (Mann–Whitney U test; **P < 0.01). (C) Representative histograms of Treg-mediated suppression of autologous and heterologous CFSE-labeled Tresp cells on day 3.5 from one patient with dSSc and one with lSSc compared to a healthy donor. Dashed line shows non-stimulated Tresp. (D) shows the summary for autologous and heterologous suppressive ability of Tregs of HD (n=4) and patients with SSc (dSSc210, dSSc366, lSSc368, lSSc369, and dSSc370, n=5) and (E) combined Treg suppression data for Tresp from either HD or patients with SSc. Lines show the median. P values are shown when significant (Mann–Whitney U test, **P < 0.01).

Failure in early B cell tolerance checkpoints in patients with SSc is a feature common to most patients with autoimmune diseases.

We compared the frequencies of polyreactive and HEp-2-reactive clones in the new emigrant/transitional and mature naïve B cell compartments of patients with SSc to those in patients with other autoimmune diseases that we previously tested (3). We found that the elevated frequencies of polyreactive new emigrant/transitional B cells in SSc differed those in HD and MS but were similar to those in T1D, RA, pediatric SLE, and SjS, revealing that central B cell tolerance defects are a common feature of these autoimmune diseases—excluding MS (Figure 4A) (3). In contrast, all patients with autoimmune diseases including MS displayed an impaired peripheral B cell tolerance checkpoint, resulting in the accumulation of both polyreactive and HEp-2-reactive B cells in their blood (Figure 4B and C). Hence, the dysregulated removal of developing autoreactive naïve B cells continuously produced throughout life correlates with autoimmunity.

Figure 4. Defective central and peripheral B cell tolerance checkpoints are features common to many autoimmune diseases.

The frequencies of (A) polyreactive new emigrant/transitional B cells, (B) polyreactive mature naïve B cells, and (C) HEp-2-reactive mature naïve B cells in 10 patients with SSc were compared with those from 13 healthy donors (HD), 7 patients with multiple sclerosis (MS), 8 patients with type 1 diabetes (T1D), 10 patients with rheumatoid arthritis (RA), 3 patients with pediatric systemic lupus erythematosus (pediatric SLE) and 5 patients with primary Sjögren’s syndrome (SjS). Each diamond represents an individual, and the median is shown with a bar. Mann–Whitney U test; **P < 0.01; ***P < 0.001; ****P < 0.0001.

DISCUSSION

We reported herein that patients with SSc display impaired central and peripheral B cell tolerance checkpoints, which result in the increased production of autoreactive naïve B cells. In addition, we show that V(D)J recombination of germline Ig gene segments can produce immature B cells that recognize SSc-specific self-antigens (i.e., topoisomerase I) in the absence of somatic hypermutation or affinity maturation. In agreement with this observation, it was recently reported that the activation of transitional B cells isolated from patients with SSc could lead to the detection of anti-topoisomerase I antibody secretion (21). Reversion experiments that consisted of removing somatic hypermutation in self-antigen specific mutated antibodies previously suggested that circulating pathogenic autoantibodies targeting the aquaporin-4 water channel in patients with NMOSD can originate from autoreactive naïve B cells, which escape early B cell tolerance checkpoints (22). Our data further support a direct contribution of impaired central and peripheral B cell tolerance checkpoints in promoting autoimmune diseases through the production of self-antigen specific naïve B cells that may not only present self-antigens to T cells and initiate autoimmune responses, but also be the precursors of the plasma cells secreting autoantibodies involved in the formation of immune complexes. Defective autoreactive B cell selection likely plays an important role in SSc pathophysiology since it was recently reported that the secretion of autoantibodies and the formation of immune complexes in patients with SSc promote fibrosis through the activation of fibroblasts via IL-6 -and M-CSF-induced osteopontin secretion by macrophages (4). It remains to be determined if autoantibodies in SSc may originate from the activation of autoreactive new emigrant/transitional or mature naïve B cells. The identification of anti-topoisomerase I reactive clones in the new emigrant/transitional B cell compartment of patients with SSc further supports a direct involvement of transitional B cells in autoantibody secretion in autoimmune diseases (this study and (21)). Several reports have suggested that antinuclear autoantibodies in SLE may be secreted upon transitional B cell activation via TLR7 and IFNα (23-25). Since a type I IFN-induced gene signature typically associated with SLE is also present in about half of the patients with SSc, it is plausible that the latter scenario may also promote autoantibody production in SSc (26, 27). In addition, plasmacytoid dendritic cell IFNα secretion is enhanced in SSc due to dysregulated and enhanced TLR function (28, 29).

Transitional B cells can also be stimulated by TLR9 ligands, which induce cell proliferation, somatic hypermutation and antibody secretion (30). However, B-cell receptor (BCR)/Toll-like receptor 9 (TLR9) co-triggering normally prevents the production of autoreactive antibodies targeting DNA-containing self-antigens by inducing B cell death (31). Thus, anti-topoisomerase I transitional B cells from patients with SSc may develop in antibody-secreting cells if TLR9 tolerogenic function is impaired. While TLR9 function is defective in naïve B cells from patients with SLE who also secrete autoantibodies targeting DNA-containing antigens such as anti-dsDNA and anti-histones (32), Taher et al. suggested that TLR9 responses in total B cells from patients with SSc may also be decreased compared to their HD counterparts (21). Hence, impaired TLR9 function following the co-crosslinking of autoreactive BCRs with TLR9 by DNA-containing self-antigens may allow for the survival of anti-topoisomerase I transitional B cells in patients with dSSc and anti-centromere reactive clones in patients with lSSc, and may lead to the secretion of autoreactive antibodies in the serum of these patients. We cannot rule out that autoantibodies may also emerge from the activation of mature naïve B cells, which also contained anti-nuclear reactive clones in patients with SSc or other autoimmune diseases. Impaired early B cell tolerance checkpoints and the production of autoreactive B cells may also favor autoimmunity via the secretion of pro-inflammatory cytokines such as GM-CSF in SSc and MS (33, 34) or IL-6 in SSc and RA (21, 35). B cell production of lymphotoxin β may also favors tertiary lymphoid organ formation (36). Regardless, rituximab, which eliminates B cells, has shown therapeutic efficacy in SSc, further solidifying the importance of B cells in SSc pathophysiology (37, 38). Alternatively, restoring central B cell tolerance may represent a novel alternative therapeutical strategy to prevent autoantibody secretion and the formation of immune complexes, which may slow down or stop fibrotic processes in SSc (9).

What are the origins of defective central and peripheral B cell tolerance checkpoints in SSc? The analysis of early B cell tolerance checkpoints in primary immunodeficient patients with rare genetic mutations has revealed that decreased signaling from receptors that recognize self-antigens at the immature B cell stage, i.e., the BCR and potentially TLRs, resulted in an impaired central B cell tolerance, whereas decreased Treg numbers, defective Treg suppressive function or T cells refractory to Treg suppression were associated with an increase in autoreactive clones in the mature naïve B cell compartment (3, 14-17). In line with these observations, the 1858T polymorphism in the PTPN22 gene identified by genome-wide association studies (GWAS) is associated with T1D, RA, SLE, and SSc (5-7), decreases BCR and TCR signaling in human cells and interferes with developing autoreactive B cell counterselection (8, 9, 39, 40). Three out of the ten patients with SSc enrolled in this study harbored the 1858T PTPN22 variant, which likely accounts for their defective early B cell tolerance checkpoints (8, 9). Other gene variants associated with SSc, SLE, and other autoimmune diseases encoding B cell-specific scaffold protein with ankyrin repeats (BANK1) and B lymphocyte kinase (BLK), both of which regulate BCR and TLR signaling, may also contribute to the alteration of central B cell tolerance in SSc (41-44). In contrast to central B-cell tolerance, the peripheral B-cell tolerance checkpoint relies on B-cell extrinsic factors such as T cells. Indeed, the absence of T cells in CD3-deficient patients or defective Treg function in FOXP3-deficient patients or in other patients with primary immunodeficiencies, resulted in an impaired peripheral B-cell tolerance checkpoint (14-17). T cells may also provide survival signals to autoreactive B cells and favor the accumulation of autoreactive B cells. For instance, defective thymocyte selection in AIRE deficiency which favors the accumulation of autoreactive T cells in the conventional CD4+ rather than in the Treg cell compartment, results in the expansion of anti-insulin mature naïve B cells that are not detected in the absence of T cells in patients with CD3D or CD3E deficiency (14, 45). While Treg numbers and suppression function in SSc appeared unaffected, we found that Tresp from patients with SSc were refractory in vitro to Treg suppression. Similarly, patients with either T1D or X-linked lymphoproliferative disease (XLP) caused by SLAM-associated protein (SAP) deficiency also displayed Tresp cells that are refractory to Treg suppression in vitro, a feature that correlated with a defective peripheral B-cell tolerance checkpoint (11, 46). The increased production of various cytokines in SSc, especially IL-4 and IL-6, may alter Treg suppression in vitro and favor the fibrotic process (11, 47). In addition to rendering Tresp cells refractory to Tregs and perhaps directly affecting Treg suppressive function, Th2/Tfh cytokines in SSc including IL-4, IL-6, IL-10 and IL-21, favor B-cell activation, class switch recombination, plasma cell development, and the production of anti-topoisomerase I autoantibodies (48-50). Moreover and in line with the type I IFN signature common in SSc, polymorphisms identified by GWAS in several interferon regulatory factor (IRF) genes are associated with SSc and may also favor autoantibody secretion (51).

In conclusion, our studies demonstrate that patients with SSc display multiple defective B cell tolerance mechanisms, ultimately resulting in the production of autoreactive naïve B cells and secretion of autoantibodies that target specific self-antigen. Impaired early B cell tolerance checkpoints in SSc may be caused by polymorphisms or rare gene mutations as is frequently the case in other autoimmune diseases and primary immunodeficiencies. Alternatively, it is conceivable that cytokines or chemokines, both of which are routinely found at elevated concentrations in the serum of patients with autoimmunity, may also interfere with the counterselection of autoreactive B cells either in the bone marrow or periphery. Indeed, SSc is associated with a common type I IFN signature and aberrant cytokine and/or chemokine profiles that may influencing B cell selection and disease progression. Additional investigations are therefore warranted to explore such hypotheses.