You are here

Article Text

Article menu
Basic and translational research
Extended report
Rheumatoid arthritis synovial tissue harbours dominant B-cell and plasma-cell clones associated with autoreactivity
  1. M E Doorenspleet1,2,
  2. P L Klarenbeek1,2,
  3. M J H de Hair1,
  4. B D C van Schaik3,
  5. R E E Esveldt1,2,
  6. A H C van Kampen3,4,
  7. D M Gerlag1,
  8. A Musters1,
  9. F Baas5,
  10. P P Tak1,
  11. N de Vries1,2
  1. 1Department of Clinical Immunology & Rheumatology, Academic Medical Center of the University of Amsterdam, Amsterdam, The Netherlands
  2. 2Department of Experimental Immunology, Academic Medical Center of the University of Amsterdam, Amsterdam, The Netherlands
  3. 3Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Academic Medical Center of the University of Amsterdam, Amsterdam, The Netherlands
  4. 4Department of Biosystems Data Analysis, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands
  5. 5Department of Genome Analysis, Academic Medical Center of the University of Amsterdam, Amsterdam, The Netherlands
  1. Correspondence to Dr N de Vries, Department of Clinical Immunology and Rheumatology, Academic Medical Center/University of Amsterdam, Meibergdreef 9, Room F4-105, Amsterdam 1105 AZ, The Netherlands; n.devries{at}amc.nl

Abstract

Objective To identify potential autoreactive B-cell and plasma-cell clones by quantitatively analysing the complete human B-cell receptor (BCR) repertoire in synovium and peripheral blood in early and established rheumatoid arthritis (RA).

Methods The BCR repertoire was screened in synovium and blood of six patients with early RA (ERA) (<6 months) and six with established RA (ESRA) (>20 months). In two patients, the repertoires in different joints were compared. Repertoires were analysed by next-generation sequencing from mRNA, generating >10 000 BCR heavy-chain sequence reads per sample. For each clone, the degree of expansion was calculated as the percentage of the total number of reads encoding the specific clonal sequence. Clones with a frequency ≥0.5% were considered dominant.

Results Multiple dominant clones were found in inflamed synovium but hardly any in blood. Within an individual patient, the same dominant clones were detected in different joints. The majority of the synovial clones were class-switched; however, the fraction of clones that expressed IgM was higher in ESRA than ERA patients. Dominant synovial clones showed autoreactive features: in ERA in particular the clones were enriched for immunoglobulin heavy chain gene segment V4–34 (IGHV4–34) and showed longer CDR3 lengths. Dominant synovial clones that did not encode IGHV4–34 also had longer CDR3s than peripheral blood.

Conclusions In RA, the synovium forms a niche where expanded—potentially autoreactive—B cells and plasma cells reside. The inflamed target tissue, especially in the earliest phase of disease, seems to be the most promising compartment for studying autoreactive cells.

  • B cells
  • Rheumatoid Arthritis
  • Synovitis
  • Early Rheumatoid Arthritis

https://doi.org/10.1136/annrheumdis-2012-202861

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Autoreactive B lymphocytes are thought to play a cardinal role in autoimmune-mediated inflammatory diseases ranging from B-cell-mediated diseases (eg, systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA)) to diseases that until recently were thought to be mediated primarily by autoreactive T cells (eg, multiple sclerosis and diabetes mellitus type 1).1–4 B-cell-depletion therapy has shown that targeting B cells results in clinical improvement in several autoimmune diseases.5–10 Unfortunately, the effect is transient and disease relapse generally occurs.11 Hence, there is a need to identify individual autoreactive B cells and plasma cells to investigate whether targeting these specific autoreactive cells might lead to more permanent clinical improvement.

    Although many autoantibodies are known (eg, rheumatoid factor and anti-citrullinated protein antibodies (ACPAs)), isolation and characterisation of the plasma cells that produce the antibodies and their precursor B cells has proven difficult. This is mainly due to the low expression of surface immunoglobulin on plasma cells, the fact that multiple structurally different antibodies may recognise the same antigen, and the poor correlation of the primary structure of antibodies with specificity, the latter being defined in different experimental settings. Alternatively, sorting hundreds to thousands of cells and subsequently screening each for its reactivity is laborious especially when the autoantigens merely consist of candidate antigens. Therefore little is known about autoreactive B cells with regard to B-cell receptor (BCR) structural variation, phenotype and tissue distribution in disease.

    Recent developments in next-generation sequencing (NGS) allow screening of the BCR repertoire to identify and quantify expanded B- and plasma-cell clones. This technology is rapid, allowing quantitative longitudinal and cross-sectional comparisons and enabling linkage of clonal expansion to inflammation. A major advantage of this technology is that the reactivity of the receptor does not need to be known. Previously, we and others have used such approaches to identify T-cell and B-cell clones.12–17

    Several features encoded in the BCR sequence may point to autoreactivity. Since murine studies showed that massive clonal expansions are formed against self-antigens,18 ,19 autoreactive clones have been thought to be often expanded. Furthermore, since chronic antigen stimulation leads to class-switching, memory B cells and plasma cells can often be class-switched —for example, in RA synovium.20–26 A more specific feature of autoreactive B cells is enrichment of the immunoglobulin heavy chain gene segment V4–34 (IGHV4–34), which has been associated with reactivity against self-epitopes and seems to be tightly regulated to prevent autoimmune responses. In SLE, such autoantibody-producing IGHV4–34-positive plasma cells are enriched in peripheral blood (PB) compared with healthy individuals and those with RA.27 Another more specific feature of B cells (reactive to nuclear antigens) is an increased number of amino acids in the most dominant antigen-binding site of the BCR (third complementary determining region (CDR3) of the heavy chain). These are normally counter-selected during B-cell development, but might be enriched in autoimmune disease.28 In this study, we use NGS to measure B-cell and plasma-cell clonal expansions and characterise their autoreactive properties in RA. We compare PB and synovial tissue (ST) to see whether there is specificity for one of these compartments. To evaluate whether B-cell clonality is variable during different phases of the disease, we compare patients with early (ERA) and established (ESRA) RA.

    Methods

    Patients

    Six patients with RA who were naïve to disease-modifying antirheumatic drugs, with disease duration <6 months, were included (ERA) as well as six patients with active disease despite methotrexate treatment (7.5–30 mg/week), with a disease duration >20 months (ESRA). All patients were naïve to treatment with biological agents, fulfilled 1987 American College of Rheumatology criteria for RA,29 were seropositive (ACPAs >25 kAU/l and/or rheumatoid factor >12.5 kU/l) and had active disease defined by a Disease Activity Score examined in 28 joints >3.2 (see online supplementary table S1). Paired synovial fluid (SF) was obtained from an additional ESRA patient (ESRA7), who had active disease despite treatment with rituximab (1000 mg rituximab, 100 mg methylprednisolone, infusion 9 months before sampling). The study was approved by the local medical ethics committee, and all patients gave written informed consent.

    Synovial biopsy/fluid and PB sampling

    In all patients, synovium was sampled from a clinically inflamed joint as described previously.30 In one ESRA patient, biopsy samples were obtained from both inflamed knees within 10 days of each other (ESRA6). Paired PB was sampled in four ERA and four ESRA patients. In ESRA7, SF was obtained by needle arthrocentesis simultaneously from an inflamed elbow and knee, together with paired PB. Storage of synovial biopsy samples and PB samples, isolation and quantification of RNA, and cDNA synthesis were performed as described previously.12

    Linear amplification and NGS

    Linear amplification was based on the protocol used previously.12 ,13 ,17 ,31 In short, a linear amplification of the BCR repertoire was performed using primers covering all functional Vheavy genes. The Vheavy primers contained a primer B sequence required for amplicon sequencing (Roche Diagnostics, Mannheim, Germany). Amplified products were purified and used in a generic PCR using primer B as forward primer and a reverse generic primer specific for all functional Jheavy genes, containing primer sequence A. Alternatively, a Cα,32 Cδ,33 Cγ or Cμ reverse primer was used, again containing primer sequence A.34 Samples were again purified, quantified, prepared for sequencing according to the manual for amplicon sequencing, and sequenced on a Roche Genome Sequencer FLX (titanium platform). For each sample, >10 000 BCRheavy sequences were analysed. NGS visualises expanded B cells as a deviation in the repertoire because they carry the same BCR sequence. Moreover, plasma cells can be identified because they produce increased amounts of BCR mRNA, producing a comparable deviation in the repertoire. For clarity, the term ‘dominant clone’ was used to denote unique BCR signals that are highly abundant within the repertoire.

    Bioinformatics and data and statistical analysis

    The bioinformatics pipeline used to obtain the BCR sequences was described previously in detail,31 performing Multplex IDentifier (MID) sorting, identification of gene segments, CDR3 detection and removal of artefacts. CDR3 lengths were defined according to the international ImMunoGeneTics information system.35 Values are expressed as mean and SD, or median and IQR, following criteria for (non-)parametric analysis. Differences between groups were analysed using Student t tests, Mann–Whitney U tests or one/two-way analysis of variance. Two-sided p values of <0.05 were considered significant. GraphPad Prism software version 5.1 was used to perform the analyses.

    Results

    B-cell clonal expansions in the synovium

    First, we investigated whether activated and therefore expanded B and plasma cells were present in the inflamed ST. To this end, the BCRheavy chain repertoire in the synovium was compared with that in PB.

    In both ERA and ESRA, dominant clones were detected in ST but not PB (figure 1A,B). The distribution of clonal size did not differ significantly between ERA and ESRA patients (see online supplementary figure S1A,B). However, the pooled frequency distribution of the synovial repertoire contained significantly more high-frequency clones (right-skewed distribution) than that of PB. Based on this frequency distribution—approaching the x-axis in the stratum 0.4–0.5%—we arbitrarily defined dominant clones as clones with a frequency ≥0.5%. Using this threshold, we found that 1.9% of the ST clones were dominant, and 0.3% in PB (figure 1C). These synovial frequencies did not correlate with disease phase (see online supplementary figure S1A). Dominant synovial clones had a large impact on the total repertoire; combined, the 22 dominant clones accounted for 61% (median, IQR 42–64%) of the repertoire, whereas the 22 most dominant clones in blood accounted for 14% of the repertoire (median, IQR 7–22%) (figure 1D). Again, there was no correlation with disease phase (see online supplementary figure S1B). Collectively, these data show selective enrichment of multiple dominant clones in the inflamed synovium compared with PB.

    Figure 1
    Figure 1

    Clonal size and impact of B-cell clones in the synovium and peripheral blood. (A,B) Scatter plot of early (ERA1–6) and established (ESRA1–6) rheumatoid arthritis (RA) showing clones recovered from (A) synovium and (B) peripheral blood (ERA1–4 and ESRA1–4). Each dot represents one clone. The size of the clones is depicted as percentage of the total BCRheavy sequences. (C) Frequency distribution of synovial tissue (ST) clones (black bars) and peripheral blood (PB) clones (white bars) showing a right-skewed distribution in both compartments and significantly more dominant clones present in ST (bars show mean and SD; **p<0.01, ***p<0.001 using two-way analysis of variance). (D) Cumulative size of the 250 most dominant clones in ST of 12 RA patients (black line) and PB of eight RA patients (grey line, mean and SD are shown). The x-axis depicts the number of clones included starting from the most abundant clone; the y-axis shows the cumulative size of the clones as a percentage (***p<0.001 in all 22 clones included). BCR, B-cell receptor.

    Overlapping B-cell clones between multiple joints

    If dominant clones in ST were truly RA related, one would expect them to be present in multiple joints of the same patient and to be enriched in synovium compared with blood. To this end, we compared the repertoires in ST of both (knee) joints of one ESRA patient (ESRA6), and in SF of an elbow and knee joint (together with PB) of an additional ESRA patient (ESRA7). Furthermore, we compared the clones in ST and PB in ERA and ESRA patients.

    In the left knee of patient ESRA6, we detected 24 dominant clones. Ten of these clones were also detected in the right knee, of which three were dominant in both joints. The findings in the right knee were comparable (six overlapping out of 20 dominant clones). Together, all overlapping clones accounted for 43% of the repertoire in both joints (figure 2A). In SF, we detected four out of 17 dominant clones in the knee and elbow and five out of 16 clones from the elbow and the knee, of which three were dominant in both (figure 2B).

    Figure 2
    Figure 2

    Clonal overlap between different joints of the same patient, and between synovium and peripheral blood. (A) Overlapping clones between synovial tissue (ST) biopsy samples from two different joints (left and right knee) of the same patient. Each dot represents one overlapping clone. The clonal frequency (in percentage) in each compartment is depicted on both axes. The dotted lines all delineate 0.5%, and clones depicted above or to the right of these lines represent the clones that are among the most dominant in each joint. Clones in the upper right quadrant represent those that are the most dominant in both knees. Clones with a frequency of 0 are present in the other joint but not traced back in the joint studied. (B) Overlapping clones between synovial fluid (SF) samples from two different joints (elbow and knee) within one patient. Clones in the upper right quadrant represent those that are the most dominant in both joints. (C) Overlapping clones between peripheral blood and ST in a patient with established RA (four analyses in total, representative picture). (D,E) Overlapping clones between peripheral blood and SF from a knee joint (D), and SF from an elbow joint (E).

    Comparing the ST or SF clones with those in PB, only 0.15% of the synovial clones could be recovered in PB (median, IQR 0.03–0.42%) (figure 2C–E). Collectively, several overlapping dominant clones were observed between different joints, suggesting that specific B-cell and plasma-cell clones reside in multiple inflamed joints of the same patient. We found no evidence for extensive overlap of dominant clones between different patients (online supplementary table S3) or between synovium and blood within a patient.

    Class-switch recombination

    In line with the hypothesis that autoreactive B cells in synovium are specifically activated and expanded, it is likely that class-switching—from IgD+IgM+ to IgG+ or IgA+—occurs as well. We therefore analysed the immunoglobulin isotypes of synovial clones in ERA and ESRA. We found that the majority of the clones were class-switched (ERA 87.8% (SD 6%), ESRA 59.2% (SD 15%); p<0.0001 compared with PB of healthy individuals (19.3%), data not shown, online supplementary figure S2). Analysis in dominant clones showed identical results (ERA mean 81.9% and ESRA mean 56.0%; p<0.0001; figure 3A). Interestingly, significantly more clones were class-switched in ERA than ESRA (p=0.01). Accordingly, the ratio IgG+/IgM+ clones was altered in ERA (8.4, SD 1.6) and ESRA (1.7, SD 1.3) (p<0.001; figure 3B,C; online supplementary results section). Collectively, these data show that the majority of synovial clones have a switched isotype, although the ratio IgG+/IgM+ is different in ERA and ESRA.

    Figure 3
    Figure 3

    Class-switch recombination of synovial clones in early RA (ERA) and established RA (ESRA). (A) The percentage of the repertoire occupied by all class-switched clones in synovium (total) compared with the percentage of class-switched clones present among the dominant clones (dom.) in ERA and ESRA (bars show mean and SD). (B) Percentage of the clones occupied by the different immunoglobulin isotypes in synovium in ERA (black bars) and ESRA (white bars) (bars show mean and SD, ***p<0.001 for both IgG and IgM using two-way analysis of variance). (C) The ratio between the number of clones with an IgG isotype and an IgM isotype in synovium in ERA versus ESRA (lines show mean and SD; ***p<0.0001 using the Student t test).

    IGHV4–34 use

    It is thought that the IGHV4–34 gene segment conveys reactivity against self-epitopes and is therefore counter-selected through the differentiation from B cell to plasma cell.27 As we found expanded B- and plasma-cell clones in ST of ERA and ESRA patients but not in PB, we hypothesised that the inflamed ST might be enriched for IGHV4–34, especially the dominant clones. To this end, we analysed synovial clones for the presence of the IGHV4–34 in comparison with PB.

    In ERA, 4.1% of all synovial clones expressed IGHV4–34 (mean, SD 2.5%), compared with 2.0% in ESRA (mean, SD 0.9%). Similar numbers were found in PB (4.6% in ERA and 5.2% in ESRA, p=0.61 and p=0.12, respectively) compared with ST (figure 4A). Among the dominant synovial clones in ERA, 12.3% expressed IGHV4–34 (median, IQR 8.3–17.1%), while in ESRA only one patient had one IGHV4–34-positive dominant clone (median 0, IQR 0–1.4%; p=0.002) (figure 4B). The results remained significant if the antinuclear antibody-positive (ANA+) RA patients were excluded (p=0.009; data not shown). Almost all dominant IGHV4–34-expressing clones were IgG+ (median 100%; data not shown). In conclusion, among dominant clones, we observed an enrichment of clones expressing the IGHV4–34 gene in ST of ERA patients, but not in ST of ESRA patients, or in PB.

    Figure 4
    Figure 4

    IGHV4–34 clones in synovial tissue (ST) and peripheral blood (PB) in early RA (ERA) and established RA (ESRA). (A) Percentage of total clones that are IGHV4–34 bearing in ST compared with those in PB in ERA and ESRA (median and IQR are depicted). (B) Percentage of all dominant clones that are IGHV4–34 bearing in ST, comparing ERA (ERA-ST) and ESRA (ESRA-ST) (median and IQR are depicted; **p<0.01 using Mann–Whitney U test).

    Increased CDR3 length

    Longer BCRheavy CDR3 sequences have also been associated with self-reactive antibodies.28 ,36–39 We analysed BCRheavy CDR3 lengths, and observed that synovial clones have longer CDR3 sequences than PB, in both ERA and ESRA. Dominant synovial clones showed longer CDR3s in ERA compared with ESRA (p=0.05); this difference disappeared if the IGHV4–34-bearing clones were excluded from the analysis (figure 5 and see online supplementary results). As a result, in ERA, dominant IGHV4–34-positive clones showed significantly longer CDR3s than IGHV4–34-negative clones (p=0.005). Collectively, these data show that synovial clones have longer CDR3 sequences than blood.

    Figure 5
    Figure 5

    CDR3 length of clones present in synovial tissue (ST) compared with peripheral blood (PB) in early RA (ERA) and established RA (ESRA). The CDR3 length (number of amino acids) of all synovial clones (total ST), all dominant synovial clones (dominant ST), all dominant clones using the IGHV4–34 gene if available (IGHV4–34(+) ST), all dominant clones that do not use the IGHV4–34 gene (IGHV4–34(−) ST) and all clones from PB in ERA as well as ESRA. All bars show mean and SD (*p=0.05, **p<0.01, ***p<0.001 using one-way analysis of variance).

    Discussion

    Here we describe the findings of a quantitative repertoire analysis to find potential autoreactive B cells and plasma cells in synovium and blood in RA. Our findings suggest that the inflamed synovium forms a niche where dominant, class-switched B cells and plasma cells are retained. This notion is supported by the finding that the dominant clones can be found in multiple joints of the same patient, but not in PB. Especially in ERA, these dominant clones are enriched for the IGHV4–34 gene, which has longer CDR3 sequences. Taken together, these findings suggest local accumulation of B cells and plasma cells with autoreactive features in the inflamed synovium.

    In the literature, a disputed finding is whether individual B-cell clones can be retrieved from and are expanded in multiple joints of the same patient.40 ,41 Here we can benefit from the high resolution of the NGS approach, which gives a full-repertoire perspective. Our data indicate that only a few synovial clones are dominant, and that these can be found in multiple joints of the same patient. These results are not due to a more generalised sharing of clones between blood and synovium, since dominant synovial clones are hardly traceable in blood, and when they are, they are present in very low frequencies, showing up to a 1000-fold difference in clonal frequency between synovium and blood. In fact, the true differences in clonal frequencies between synovium and blood might be even larger if some of the overlap in our samples is the result of traces of blood in the synovial biopsy samples. Interestingly, not all dominant clones were detectable in multiple joints. This suggests that both ‘public’ clones (shared with other joints) and ‘private’ clones (specific to the individual joint) exist. One can speculate whether these differences arise randomly, or point to the existence of private antigenic targets in different joints. Additional studies are required to provide further insight into the exact roles and reactivities of the shared and joint-specific clones.

    We chose to study the B-cell repertoire by measuring BCR heavy-chain mRNA levels to analyse the abundance of each clone in the context of the full repertoire. This approach has the advantage that it can start from very low amounts of cells, requires limited amplification of BCR mRNA in the middle of the human exome, and only identifies expressed BCRs. However, it is important to notice that these mRNA molecules code for antigen receptors on different cell types in the B-cell lineage, including B cells and plasma cells. These cells might express different levels of BCR mRNA, resulting in erroneous estimates of cell frequencies.

    For this study, one may argue that the BCR signal in PB originates from B cells, since plasmacytoid cells are relatively rare, whereas in the synovium it originates from plasma cells. Previous publications indicated that plasma cells and B cells show a 5–50-fold difference in the amount of BCR mRNA (depending on the activation status of the B cell).42 ,43 Our own preliminary estimates suggest a median difference of three (range 2–13-fold). In this context, 1000-fold differences between clonal signals in synovium and blood indicate that some of the differences can be attributed to the selective clonal enrichment of individual clones in the joint. Taken together, these data imply that the synovium is selectively enriched in specific B-lineage clones, indicating that it forms a niche for disease-associated B cells and plasma cells.

    In this study, we were unable to obtain phenotypic or genomic data on the synovial clones, because isolation of these cells is technically challenging. It is important to overcome this hurdle, as it would allow comparison of different B-cell subsets in synovium and blood. This would answer the question whether the difference in BCR signals arises from particular subsets of cells. Moreover, it would allow further characterisation of expanded clones to see whether they are truly pathogenic, or of a more regulatory phenotype.

    Different lines of evidence suggest that autoreactive B cells reside in RA synovium. Several studies have previously reported that memory B cells and plasma cells are present in RA synovium and sometimes in blood.20 ,23 ,40 ,41 ,44 ,45 More recently, it has been shown that synovial immunoglobulins recognise multiple self-antigens.46 Interestingly, the IGHV4–34 gene segment has been shown to be used more often in PB plasma cells in ANA+ SLE patients, which has been implied in autoreactivity.27 This could not be confirmed in RA. We chose to study this finding in RA synovium, and observed increased usage of IGHV4–34 in dominant clones in ST. Since most of our patients did not have ANAs, this fuels the hypothesis that the expanded clones are autoreactive in RA also, albeit with reactivity to different autoantigens. The finding that expanded synovial clones also have increased CDR3 lengths—another feature previously associated with autoreactivity—supports this claim. Collectively, these findings suggest that some of the dominant clones in the inflamed synovium may be autoreactive. However, the mechanism underlying this association remains elusive.

    A striking finding was the difference in IgG and IgM expression between ERA and ESRA. This might be explained by the loss of dominance of T cells necessary for class-switching in germinal centres. Moreover, the increased IGHV4–34 usage and the longer CDR3 sequences were preferentially found in ERA. These findings are all in line with our recent study on T cells, which also pinpoint the most striking changes in clonality to ERA.12 A potential mechanism underlying these differences may be epitope spreading in which more and more autoantigens become involved during chronic inflammation. Such a scenario is further supported by the broadening of ACPA specificities before clinical onset of the disease as well as during the transition of undifferentiated arthritis into RA.47–49 This strongly suggests that identifying autoreactive clones in the earliest phase of disease will probably offer most insight into disease pathogenesis.

    In this study, we show how novel NGS-based repertoire analysis may help to identify, track and characterise disease-associated B- and plasma-cell clones. One of the next challenges will be to define the antigen specificity of these clones. As novel approaches become available, quantitative high-throughput screening of immunoglobulin reactivity will be possible.50 Coupling these techniques to BCR-repertoire screens will allow the coupling of an autoreactive phenotype to BCR reactivity. Our study clearly suggests that such studies should focus on the earliest phase of tissue inflammation. The identification of autoreactive B cells and plasma cells and their autoantigens will help to further elucidate the pathogenesis of RA and develop new therapeutic strategies for selectively targeting these cells.

    Acknowledgments

    We thank M E Jakobs and E J Bradley for technical assistance.

    References

    1. ↵
      1. Franciotta D,
      2. Salvetti M,
      3. Lolli F,
      4. et al
      . B cells and multiple sclerosis. Lancet Neurol 2008;7:852–8.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Fraussen J,
      2. Vrolix K,
      3. Martinez-Martinez P,
      4. et al
      . B cell characterization and reactivity analysis in multiple sclerosis. Autoimmun Rev 2009;8:654–8.
      OpenUrlCrossRefPubMed
    3. ↵
      1. Hu CY,
      2. Rodriguez-Pinto D,
      3. Du W,
      4. et al
      . Treatment with CD20-specific antibody prevents and reverses autoimmune diabetes in mice. J Clin Invest 2007;117:3857–67.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Serreze DV,
      2. Chapman HD,
      3. Varnum DS,
      4. et al
      . B lymphocytes are essential for the initiation of T cell-mediated autoimmune diabetes: analysis of a new “speed congenic” stock of NOD.Ig mu null mice. J Exp Med 1996;184:2049–53.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Cohen SB,
      2. Emery P,
      3. Greenwald MW,
      4. et al
      . Rituximab for rheumatoid Arthritis Refractory to anti-tumor necrosis factor therapy: results of a multicenter, randomized, double-blind, placebo-controlled, phase III trial evaluating primary efficacy and safety at twenty-four weeks. Arthritis Rheum 2006;54:2793–806.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Edwards JC,
      2. Szczepanski L,
      3. Szechinski J,
      4. et al
      . Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis. N Engl J Med 2004;350:2572–81.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Emery P,
      2. Fleischmann R,
      3. Filipowicz-Sosnowska A,
      4. et al
      . The efficacy and safety of rituximab in patients with active rheumatoid arthritis despite methotrexate treatment: results of a phase IIB randomized, double-blind, placebo-controlled, dose-ranging trial. Arthritis Rheum 2006;54:1390–400.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Hauser SL,
      2. Waubant E,
      3. Arnold DL,
      4. et al
      . B-cell depletion with rituximab in relapsing-remitting multiple sclerosis. N Engl J Med 2008;358:676–88.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Tamimoto Y,
      2. Horiuchi T,
      3. Tsukamoto H,
      4. et al
      . A dose-escalation study of rituximab for treatment of systemic lupus erythematosus and Evans’ syndrome: immunological analysis of B cells, T cells and cytokines. Rheumatology (Oxford) 2008;47:821–7.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Vigna-Perez M,
      2. Hernandez-Castro B,
      3. Paredes-Saharopulos O,
      4. et al
      . Clinical and immunological effects of Rituximab in patients with lupus nephritis refractory to conventional therapy: a pilot study. Arthritis Res Ther 2006;8:R83.
      OpenUrlCrossRefPubMed
    11. ↵
      1. Leandro MJ,
      2. Cambridge G,
      3. Ehrenstein MR,
      4. et al
      . Reconstitution of peripheral blood B cells after depletion with rituximab in patients with rheumatoid arthritis. Arthritis Rheum 2006;54:613–20.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Klarenbeek PL,
      2. de Hair MJ,
      3. Doorenspleet ME,
      4. et al
      . Inflamed target tissue provides a specific niche for highly expanded T-cell clones in early human autoimmune disease. Ann Rheum Dis 2012;71:1088–93.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Cantaert T,
      2. Doorenspleet ME,
      3. Francosalinas G,
      4. et al
      . Increased numbers of CD5+ B lymphocytes with a regulatory phenotype in spondylarthritis. Arthritis Rheum 2012;64:1859–68.
      OpenUrlCrossRefPubMedWeb of Science
      1. Freeman JD,
      2. Warren RL,
      3. Webb JR,
      4. et al
      . Profiling the T-cell receptor beta-chain repertoire by massively parallel sequencing. Genome Res 2009;19:1817–24.
      OpenUrlAbstract/FREE Full Text
      1. Robins HS,
      2. Campregher PV,
      3. Srivastava SK,
      4. et al
      . Comprehensive assessment of T-cell receptor beta-chain diversity in alphabeta T cells. Blood 2009;114:4099–107.
      OpenUrlAbstract/FREE Full Text
      1. Weinstein JA,
      2. Jiang N,
      3. White RA III.,
      4. et al
      . High-throughput sequencing of the zebrafish antibody repertoire. Science 2009;324:807–10.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Maillette de Buy Wenniger LJ,
      2. Doorenspleet ME,
      3. Klarenbeek PL,
      4. et al
      . IgG4+ clones identified by next-generation sequencing dominate the b-cell receptor repertoire in IgG4-associated cholangitis. Hepatology Published Onine First: 8 Jan 2013. doi:10.1002/hep.26232
    18. ↵
      1. Shan H,
      2. Shlomchik MJ,
      3. Marshak-Rothstein A,
      4. et al
      . The mechanism of autoantibody production in an autoimmune MRL/lpr mouse. J Immunol 1994;153:5104–20.
      OpenUrlAbstract
    19. ↵
      1. Shlomchik M,
      2. Mascelli M,
      3. Shan H,
      4. et al
      . Anti-DNA antibodies from autoimmune mice arise by clonal expansion and somatic mutation. J Exp Med 1990;171:265–92.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Gause A,
      2. Gundlach K,
      3. Zdichavsky M,
      4. et al
      . The B lymphocyte in rheumatoid arthritis: analysis of rearranged V kappa genes from B cells infiltrating the synovial membrane. Eur J Immunol 1995;25:2775–82.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Marion TN,
      2. Bothwell AL,
      3. Briles DE,
      4. et al
      . IgG anti-DNA autoantibodies within an individual autoimmune mouse are the products of clonal selection. J Immunol 1989;142:4269–74.
      OpenUrlAbstract
    22. ↵
      1. Randen I,
      2. Thompson KM,
      3. Thorpe SJ,
      4. et al
      . Human monoclonal IgG rheumatoid factors from the synovial tissue of patients with rheumatoid arthritis. Scand J Immunol 1993;37:668–72.
      OpenUrlCrossRefPubMed
    23. ↵
      1. Scheel T,
      2. Gursche A,
      3. Zacher J,
      4. et al
      . V-region gene analysis of locally defined synovial B and plasma cells reveals selected B cell expansion and accumulation of plasma cell clones in rheumatoid arthritis. Arthritis Rheum 2011;63:63–72.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Schroder AE,
      2. Greiner A,
      3. Seyfert C,
      4. et al
      . Differentiation of B cells in the nonlymphoid tissue of the synovial membrane of patients with rheumatoid arthritis. Proc Natl Acad Sci USA 1996;93:221–5.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Shlomchik MJ,
      2. Marshak-Rothstein A,
      3. Wolfowicz CB,
      4. et al
      . The role of clonal selection and somatic mutation in autoimmunity. Nature 1987;328:805–11.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Winkler TH,
      2. Fehr H,
      3. Kalden JR
      . Analysis of immunoglobulin variable region genes from human IgG anti-DNA hybridomas. Eur J Immunol 1992;22:1719–28.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Pugh-Bernard AE,
      2. Silverman GJ,
      3. Cappione AJ,
      4. et al
      . Regulation of inherently autoreactive VH4-34 B cells in the maintenance of human B cell tolerance. J Clin Invest 2001;108:1061–70.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Wardemann H,
      2. Yurasov S,
      3. Schaefer A,
      4. et al
      . Predominant autoantibody production by early human B cell precursors. Science 2003;301:1374–7.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Arnett FC,
      2. Edworthy SM,
      3. Bloch DA,
      4. et al
      . The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Gerlag DM,
      2. Tak PP
      . How to perform and analyse synovial biopsies. Best Pract Res Clin Rheumatol 2009;23:221–32.
      OpenUrlCrossRefPubMed
    31. ↵
      1. Klarenbeek PL,
      2. Tak PP,
      3. van Schaik BD,
      4. et al
      . Human T-cell memory consists mainly of unexpanded clones. Immunol Lett 2010;133:42–8.
      OpenUrlCrossRefPubMed
    32. ↵
      1. Keats JJ,
      2. Reiman T,
      3. Maxwell CA,
      4. et al
      . In multiple myeloma, t(4;14)(p16;q32) is an adverse prognostic factor irrespective of FGFR3 expression. Blood 2003;101:1520–9.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Aarts WM,
      2. Willemze R,
      3. Bende RJ,
      4. et al
      . VH gene analysis of primary cutaneous B-cell lymphomas: evidence for ongoing somatic hypermutation and isotype switching. Blood 1998;92:3857–64.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. van Dongen JJ,
      2. Langerak AW,
      3. Bruggemann M,
      4. et al
      . Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia 2003;17:2257–317.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Giudicelli V,
      2. Chaume D,
      3. Lefranc MP
      . IMGT/GENE-DB: a comprehensive database for human and mouse immunoglobulin and T cell receptor genes. Nucleic Acids Res 2005;33:D256–61.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Aguilera I,
      2. Melero J,
      3. Nunez-Roldan A,
      4. et al
      . Molecular structure of eight human autoreactive monoclonal antibodies. Immunology 2001;102:273–80.
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Crouzier R,
      2. Martin T,
      3. Pasquali JL
      . Heavy chain variable region, light chain variable region, and heavy chain CDR3 influences on the mono- and polyreactivity and on the affinity of human monoclonal rheumatoid factors. J Immunol 1995;154:4526–35.
      OpenUrlAbstract
    38. ↵
      1. Ichiyoshi Y,
      2. Casali P
      . Analysis of the structural correlates for antibody polyreactivity by multiple reassortments of chimeric human immunoglobulin heavy and light chain V segments. J Exp Med 1994;180:885–95.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Klonowski KD,
      2. Primiano LL,
      3. Monestier M
      . Atypical VH-D-JH rearrangements in newborn autoimmune MRL mice. J Immunol 1999;162:1566–72.
      OpenUrlAbstract/FREE Full Text
    40. ↵
      1. Itoh K,
      2. Patki V,
      3. Furie RA,
      4. et al
      . Clonal expansion is a characteristic feature of the B-cell repetoire of patients with rheumatoid arthritis. Arthritis Res 2000;2:50–8.
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Voswinkel J,
      2. Weisgerber K,
      3. Pfreundschuh M,
      4. et al
      . The B lymphocyte in rheumatoid arthritis: recirculation of B lymphocytes between different joints and blood. Autoimmunity 1999;31:25–34.
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Eckmann L,
      2. Huang GT,
      3. Smith JR,
      4. et al
      . Increased transcription and coordinate stabilization of mRNAs for secreted immunoglobulin alpha heavy chain and kappa light chain following stimulation of immunoglobulin A expressing B cells. J Biol Chem 1994;269:33102–8.
      OpenUrlAbstract/FREE Full Text
    43. ↵
      1. Xiang SD,
      2. Benson EM,
      3. Dunn IS
      . Tracking membrane and secretory immunoglobulin alpha heavy chain mRNA variation during B-cell differentiation by real-time quantitative polymerase chain reaction. Immunol Cell Biol 2001;79:472–81.
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      1. Mcgee B,
      2. Small RE,
      3. Singh R,
      4. et al
      . B lymphocytic clonal expansion in rheumatoid arthritis. J Rheumatol 1996;23:36–43.
      OpenUrlPubMed
    45. ↵
      1. Williams DG,
      2. Taylor PC
      . Clonal analysis of immunoglobulin mRNA in rheumatoid arthritis synovium: characterization of expanded IgG3 populations. Eur J Immunol 1997;27:476–85.
      OpenUrlCrossRefPubMed
    46. ↵
      1. Monach PA,
      2. Hueber W,
      3. Kessler B,
      4. et al
      . A broad screen for targets of immune complexes decorating arthritic joints highlights deposition of nucleosomes in rheumatoid arthritis. Proc Natl Acad Sci USA 2009;106:15867–72.
      OpenUrlAbstract/FREE Full Text
    47. ↵
      1. Sokolove J,
      2. Bromberg R,
      3. Deane KD,
      4. et al
      . Autoantibody epitope spreading in the pre-clinical phase predicts progression to rheumatoid arthritis. PLoS ONE 2012;7:e35296.
      OpenUrlCrossRefPubMed
      1. van de Stadt LA,
      2. de Koning MH,
      3. van de Stadt RJ,
      4. et al
      . Development of the anti-citrullinated protein antibody repertoire prior to the onset of rheumatoid arthritis. Arthritis Rheum 2011;63:3226–33.
      OpenUrlCrossRefPubMedWeb of Science
    49. ↵
      1. van der Woude D,
      2. Rantapaa-Dahlqvist S,
      3. Ioan-Facsinay A,
      4. et al
      . Epitope spreading of the anti-citrullinated protein antibody response occurs before disease onset and is associated with the disease course of early arthritis. Ann Rheum Dis 2010;69:1554–61.
      OpenUrlAbstract/FREE Full Text
    50. ↵
      1. Larman HB,
      2. Zhao Z,
      3. Laserson U,
      4. et al
      . Autoantigen discovery with a synthetic human peptidome. Nat Biotechnol 2011;29:535–41.
      OpenUrlCrossRefPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Handling editor Tore K Kvien

    • Contributors MED, PLK, FB, PPT and NdV were responsible for the conception and design of the study. All authors contributed to the analysis and interpretation of the data. MED and NdV drafted the article, and PLK, MJHdH, BDCvS, REEE, AHCvK, DMG, AM, FB, and PPT revised the manuscript criticially for important intellectual content. All authors finally aproved the version to be published. The research leading to these results has received support from the Innovative Medicines Initiative Joint Undertaking under grant agreement n° 115142-2, resources of which are composed of financial contribution from the European Union's Seventh Framework Programme (FP7/2007-2013) and EFPIA companies' in kind contribution.

    • Funding REEE and NdV were supported by BTCURE, a research project from the Innovative Medicines Initiative Joint Undertaking (grant No 115142-2). PLK was supported by an AMC Graduate school PhD Scholarship from the Academic Medical Center and University of Amsterdam. MJHdH was supported by the Dutch Arthritis Association (grant No 06-1-303).

    • Competing interests None.

    • Ethics approval Institutional medical ethics committee (Academic Medical Center).

    • Patient consent Obtained.

    • Provenance and peer review Not commissioned; externally peer reviewed.