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.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 63, No. 1, January 2011, pp 63–72 DOI 10.1002/art.27767 © 2011, American College of Rheumatology 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 Tobias Scheel,1 Angelika Gursche,2 Josef Zacher,2 Thomas Häupl,3 and Claudia Berek1 Objective. To elucidate the development of synovial tissue–specific B cell immune responses, the clonality of individual naive B cells, memory B cells, and plasma cells and their organization and histologic localization in the inflamed tissue were investigated in patients with rheumatoid arthritis (RA). Methods. B and plasma cells were isolated by laser capture microdissection (LCM) from the synovial tissue of patients with RA. In addition, single naive B cells, memory B cells, and plasma cells were sorted from synovial tissue cell suspensions. RNA was extracted from the cells, and Ig VH genes were amplified, cloned, and sequenced. Results. Both LCM and single cell sorting analyses showed that naive and memory B cells infiltrated the RA synovial tissue. Comparison of the V-gene repertoire of B and plasma cells suggested that synovial plasma cells were generated, by and large, from locally activated B cells, indicating that a selected population of memory B cells differentiates into large plasma cell clones that then accumulate in the inflamed tissue. Clonally related plasma cells were isolated from separate and distinct localized areas of the tissue, suggesting that the newly generated plasma cells have a high migratory capacity. Conclusion. These results support the idea of a continuous activation of selected B cell clones, and hence a massive accumulation of plasma cells, in RA synovial tissue. As B cells and their secreted antibodies are an important factor in controlling inflammatory processes, patients with RA displaying intensive synovial tissue lymphocytic infiltrations might benefit from B cell depletion therapy. Early treatment will prevent accumulation of pathogenic plasma cells. Rheumatoid arthritis (RA) is a chronic inflammatory process that leads to swollen joints and, ultimately, to destruction of the joints. Normally, the synovial lining layer consists of a thin membrane of synoviocytes attached to an extracellular matrix, known as the sublining layer, in which blood vessels, fat cells, and fibroblasts are embedded. In patients with RA, a massive infiltration of macrophages, T cells, and B cells into the sublining layer is observed, and lymphocyte accumulation is supported by intensive vascularization (1,2). A striking histologic feature of RA is the formation of lymphocytic aggregates in the synovial tissue, which may resemble tertiary lymphoid structures (3–6). Whereas small aggregates consist mainly of T lymphocytes, in large aggregates, B cells are the dominant cell type. The importance of B cells in the immunopathologic processes of RA is still controversial (7,8), although there is an increasing body of evidence showing significant clinical improvement when patients with RA are treated with rituximab, a B cell–depleting anti-CD20 monoclonal antibody (9–11). B cells may contribute to the pathogenesis of RA in multiple ways. B cells are the precursors of plasma cells, which, through the production of autoantibodies, enhance the inflammatory processes. Furthermore, B cells, acting as antigen-presenting cells, support the polyclonal activation of T cells and thus enhance the production of cytokines and other proinflammatory molecules. B cells also play a key role in the development of Supported by the Bundesministerium für Bildung und Forschung (grant NGFN2). The Deutsches Rheuma-Forschungszentrum, a Leibniz Institute, is supported by the Berlin Senate of Research and Education. 1 Tobias Scheel, PhD, Claudia Berek, PhD: Deutsches Rheuma-Forschungszentrum, Berlin, Germany; 2Angelika Gursche, MD, Josef Zacher, MD: Helios-Kliniken Berlin-Buch, Berlin, Germany; 3Thomas Häupl, MD: Charité–University Medicine Berlin, Berlin, Germany. Address correspondence to Claudia Berek, PhD, Deutsches Rheuma-Forschungszentrum, Charitéplatz 1, 10117 Berlin, Germany. E-mail: firstname.lastname@example.org. Submitted for publication March 9, 2010; accepted in revised form September 21, 2010. 63 64 SCHEEL ET AL Table 1. Demographic and clinical characteristics of the 7 patients with rheumatoid arthritis (RA)* Patient Age, years Sex Age at onset of RA, years Disease duration, years Steroids/NSAIDs/analgesics Glucocorticoids, mg/day DMARD RF Anti-CCP DAS28 ESR, mm/hour CRP, mg/dl 1 2 3 4 5 6 7 69 F 61 8 Yes 5 MTX ⫹ ⫹ 5.19 70 3.62 73 M 67 6 Yes 5 LEF ⫹ ⫹ 4.48 10 1.54 62 F 58 4 No No MTX ⫹ ⫹ 5.39 24 2.5 54 M 37 17 Yes 5 LEF ⫹ ND 5.58 24 0.26 70 F 47 23 Yes 5 LEF ⫹ ⫹ 6.49 44 10.42 67 F 61 7 No 7.5 MTX ⫹ etan. ⫹ ⫹ 4.05 17 0.13 51 F 38 13 Yes 10 HCQ ⫹ ND 4.92 20 0.28 * NSAIDs ⫽ nonsteroidal antiinflammatory drugs; DMARD ⫽ disease-modifying antirheumatic drug; MTX ⫽ methotrexate; LEF ⫽ leflunomide; etan. ⫽ etanercept; HCQ ⫽ hydroxychloroquine; RF ⫽ rheumatoid factor; anti-CCP ⫽ anti–cyclic citrullinated peptide; ND ⫽ not determined; DAS28 ⫽ Disease Activity Score in 28 joints (range 0–10); ESR ⫽ erythrocyte sedimentation rate; CRP ⫽ C-reactive protein. organized lymphoid tissue, and in RA, they are a prerequisite for the development of ectopic lymphoid tissue (12,13). By inducing lymphoid neogenesis, which is associated with more severe synovial and systemic inflammation (14,15), B cells may play an important role in establishing and maintaining the processes of chronic inflammation (16). A number of approaches to the analysis of B cell responses in the synovial tissue of patients with RA have been described. In general, studies of enzymatically digested synovial tissue have analyzed the V-gene repertoire of synovial B cells (17–19). Microdissection of B cells from ectopic germinal centers has shown that synovial B cells take part in an antigen-specific immune response involving somatic hypermutation and, possibly, affinity-selected differentiation into plasma cells (5,19– 21). Previous studies utilizing a modified VH-gene fingerprinting method, in which the VH-gene repertoire of B cells was compared between the peripheral blood, synovial fluid, and synovial tissue of patients with RA, yielded findings suggesting that clonal B cell expansion is a characteristic feature of RA (19,22,23). In the present study, we analyzed the B cell immune responses in synovial tissue that lacked germinal center formation. B cells were microdissected directly from large B cell infiltrates and/or multiple plasma cell aggregates in the inflamed tissue of patients with RA. B and plasma cells were microdissected from distinct and separate locations, and their VH-gene repertoire was compared. The analysis showed that local B cell activation and direct differentiation into plasma cells may occur in RA synovial tissue. Intensive expansion of selected B cell clones was confirmed by an independent approach in which single naive B cells, memory B cells, and plasma cells were sorted from synovial cell suspensions. Our findings demonstrate that there is a continuous activation of B cells as well as differentiation of these cells into plasma cells, leading to an accumulation of large plasma cell clones, in RA synovial tissue. In patients with RA, B cell depletion, which interferes with plasma cell development and Ig secretion, may thus help to control the chronic inflammatory processes of this disease. PATIENTS AND METHODS Patients. Synovial tissue was obtained from 16 patients with a diagnosis of RA according to the American College of Rheumatology criteria (24). Synovial tissue samples from 3 patients with B cell and/or plasma cell infiltrates (samples from patients 1–3 [P1–P3]) were analyzed by laser capture microdissection (LCM). In addition, single B cells and plasma cells were sorted from synovial cell suspensions (samples P4–P7). The demographic and clinical features of all 7 of these patients with RA are summarized in Table 1. Immunohistochemistry. Frozen tissue sections were fixed in acetone at ⫺20°C for 10 minutes. Endogenous biotin was blocked using an avidin–biotin blocking kit (Vector), followed by a 20-minute blocking step with phosphate buffered saline/3% bovine serum albumin. To identify cell types and their location in the synovial tissue, tissue sections were stained with the following mouse anti-human antibodies: anti-CD20 (L26; Dako), anti-CD4 (MT310; Dako), anti-CD8 (DK25; Dako), anti-CD68 (EBM11; Dako), anti–HLA–DR (L243; DRFZ), anti-CD55 (MCA1614; Serotec), anti-CD31 (JC/70A; Dako), and anti-human FDC (8D6; BD PharMingen). To detect mouse primary antibodies, an alkaline phosphatase– CLONAL ACCUMULATION OF SYNOVIAL B AND PLASMA CELLS IN RA Table 2. 65 Clonal diversity of microdissected synovial B cells* No. of areas No. of caps Agg 7 22 157 33 Ass Agg 2 2 6 14 18 38 139 142 310 55 48 85 9 5 6 37 25 18 12 147 139 173 93 1,153 67 61 44 393 Patient, cell population Patient 1 Plasma cell Patient 2 B cell Inf Plasma cell Plasma cell Patient 3 B cell Inf Plasma cell Plasma cell Total No. of sequences No. of different V–D–J rearrangements No. of clones 10 35 11 Ass Agg 56 * Populations of B cell infiltrates (Inf), plasma cells associated with B cell infiltrates (Ass), and plasma cell aggregates (Agg) were isolated from rheumatoid arthritis synovial tissue by laser capture microdissection. labeled polymer (Dako) was used and visualized with the Alkaline Phosphatase Substrate I kit (Vector). Plasma cells were detected by staining with the biotinylated monoclonal antibody Wue-1 (DRFZ) (25), followed by detection with peroxidase-coupled streptavidin (Sigma) and visualization using Fast diaminobenzidine (Sigma). Hematoxylin (Merck) was used to stain the nuclei. LCM analysis. Frozen tissue sections (kept frozen at ⫺70°C) were immediately transferred to acetone at ⫺20°C for 3 minutes. To protect against RNA degradation, sections were stained using the HistoGene LCM Immunofluorescence Staining kit (Arcturus). Plasma cells were labeled with Alexa 488–Wue-1 antibodies (DRFZ) for 4 minutes at room temperature and dehydrated using 70% ethanol for 30 seconds, 95% ethanol for 30 seconds, 100% ethanol for 30 seconds, and a final step of xylene for 5 minutes. Clusters of B cells were identified on consecutive sections by staining with the mouse anti-human CD20 antibody L26 (Dako) and Alexa 488–labeled goat anti-mouse (Invitrogen) as a secondary antibody. For microdissection, the laser spot diameter was adjusted to 7 m, and cell samples were collected onto CapSure HS caps (Arcturus). B cell infiltrates were isolated as small areas of ⬃10–20 B cells. Two-to-eight adjacent areas (depending on the size of the B cell infiltrate) were dissected from each of the follicles analyzed. Plasma cells were obtained by microdissecting neighboring single cells to avoid contamination of juxtaposed B cells. For further analysis, ⬃10 cells were collected onto one cap. To remove co-isolated cells, the cap surface was pressed onto a sterile tissue preparation strip (Arcturus). RNA was immediately extracted using the PicoPure RNA isolation kit (Arcturus), and integrity of the RNA was controlled with the use of the Agilent 2100 Bioanalyzer. When RNA was prepared from the remaining tissue sections after 2.5 hours of microdissection, RNA integrity numbers between 5.7 and 6.9 were obtained. The complementary DNA (cDNA) of rearranged Ig genes was synthesized using the SensiScript reverse transcription (RT) kit (Qiagen) with primers specific for the different Ig isotypes: for C␥, 5⬘-GAGGGCGCCAGGGGGAAGAC; for C␣, 5⬘-GAGGCTCAGCGGGAAGACCTTG; and for C, 5⬘ACGGGGAATTCTCACAGGAGAC. V-region genes were similarly amplified using a mix of VH primers (26) together with the isotype-specific primers. The amplifications were done with 1.25 units of AmpliTaq Gold enzyme (Applied Biosystems). The cycle program consisted of an initial denaturation step at 95°C for 10 minutes, followed by 50 cycles of 95°C for 30 seconds, 68°C for 30 seconds, and 72°C for 40 seconds, and a final extension step of 72°C for 10 minutes. Amplified Ig cDNA was separated on a 2% low-melting agarose gel, and bands in the range of 350 bp were excised and purified using the NucleoSpin Extract II kit (MachereyNagel). Enzymatic digestion of synovial tissue and single cell sorting. Samples of synovial tissue were cut into small pieces (maximum size 1 mm3) and digested using Collagenase D (Roche) for 90 minutes. After Ficoll centrifugation, cells were stained using the monoclonal antibodies anti-CD19– allophycocyanin-Cy7 (BD PharMingen), anti-CD27–Cy5 (DRFZ), and anti-CD38–fluorescein isothiocyanate (BD PharMingen). Before staining, Fc receptors were blocked using anti-CD16/anti-CD32 (Miltenyi). Single CD27⫺CD38⫺ naive B cells, CD27⫹CD38⫺ memory B cells, and CD27highCD38⫹ plasma cells were sorted into 96-well plates, and cDNA was synthesized using a Qiagen one-step RT– polymerase chain reaction (PCR) kit. For amplification, primers specific for the constant-region genes of H- and L-chain sequences (for CH1, C 5⬘-GGGAATTCTCACAGGAGACGA, C ␥ 5⬘-GGAAGGTGTGCACGCCGCTGGTC, and C␣ 5⬘-TGGGAAGTTTCTGGCGGTCACG; for C, 5⬘-GTTTCTCGTAGTCTGCTTTGCTCA; and for C , 5⬘-CACCAGTGTGGCCTTGTTGGCTTG) together with primers specific for the different V-region genes (26) were used. Using the CH1 C␥ primer, the heavy-chain isotypes IgG1 and IgG2 could not be distinguished. In addition, using the CH1 C␣ primer, the IgA1 and IgA2 isotypes were indistinguishable. Cells showing positive reactions on PCR were purified using ExoSAP-IT (USB). Sequencing of the immunoglobulin V-region genes. PCR products were cloned into the TOPO TA cloning vector (Invitrogen). Randomly picked clones were sequenced with an automated DNA sequencer (ABI Prism 310 Genetic Analyzer; Applied Biosystems) using the BigDye Terminator Cycle Sequencing kit (version 1.1; Applied Biosystems). Putative germline gene sequences were identified using the V BASE Se- 66 SCHEEL ET AL Table 3. Clonal diversity of sorted single synovial B cells Patient, cell population Patient 4 Plasma cells Patient 5 Naive B cells Memory B cells Plasma cells Patient 6 Plasma cells Patient 7 Naive B cells Memory B cells Plasma cells Total RESULTS No. of No. of No. of No. of sorted cells VH sequences VL sequences clones 46 19 9 96 96 96 46 40 72 35 32 51 1 96 55 34 6* 96 96 96 718 31 30 60 353 19 21 36 237 7 * VH and VL gene sequences derived from clones P6-1 and P6-2 differed by somatic diversity. quence Directory (provided by I. Tomlinson, Cambridge, UK). The third complementarity-determining region (CDR3) length of the VH region was determined according to the sequences described by Kabat et al (27). The sequences can be found in GenBank, under the accession numbers HM994882– HM996558. Statistical analysis. Statistical significance was calculated using the chi-square test (for differences in VH-gene repertoire and in Ig-class distribution), the Mann-Whitney test (for differences in numbers of somatic mutations), or Student’s t-test (for differences in CDR3 length). P values less than 0.05 were considered significant. Heterogeneous phenotype of the chronically inflamed synovial tissue of patients with RA. Synovial tissue samples were obtained during surgery from 16 patients with established RA (Table 1). All samples were assessed by immunohistologic staining, and only 7 of the 16 tissue samples were found to have substantial numbers of B and/or plasma cells. In samples P1 and P6, only plasma cell aggregations were found, whereas in samples P2–P5 and P7, large follicle-like structures surrounded by plasma cells were also seen. However, none of the characterized tissue samples showed evidence of ectopic germinal centers. The synovial B cell immune response was investigated in the synovial tissue of those 7 patients, using 2 different approaches. In samples P1–P3, B cells and plasma cells were isolated by LCM, while samples P4–P7 were used to sort single naive B cells, memory B cells, and/or plasma cells from synovial tissue cell suspensions (Tables 2 and 3). A typical example of B and plasma cell distribution in the inflamed synovial tissue is shown in Figure 1 (for sample P2). B and plasma cells were separately microdissected from the different locations. Plasma cells associated with B cell infiltrates (Figure 1, arrowheads) and plasma cells found in small aggregates along the vasculature (Figure 1, arrows) were isolated, and the VH-gene repertoire of the different cell populations was determined. Figure 1. Immunohistologic assessment of a representative synovial tissue sample from a patient with rheumatoid arthritis (sample P2). Tissue sections were stained with hematoxylin, and plasma cells (black staining) were labeled with the Wue-1 antibody. B cell infiltrates (Inf), associated plasma cells (Ass) (arrowheads), and plasma cell aggregates (Agg) (arrows) were isolated by laser capture microdissection. Insets, Higher-magnification views of the boxed areas. Bar ⫽ 100 m; original magnification of insets ⫻ 200. CLONAL ACCUMULATION OF SYNOVIAL B AND PLASMA CELLS IN RA Migration of naive and memory B cells into the synovial tissue. The analysis of B cells microdissected from the lymphocytic infiltrates showed that 55% of the synovial B cells expressed IgM, while 29% expressed IgG, and 16% expressed IgA (Figure 2A). Only one-half of the IgM sequences were unmutated. Naive B cells are defined as cells expressing unmutated immunoglobulins of the IgM isotype. On this basis, only a minority (24%) of the B cells in the lymphocytic infiltrates were of the naive phenotype. Similarly, when naive and memory B cells were sorted from synovial cell suspensions, again only a minor fraction (⬍10%) of B cells were of the naive phenotype (in samples P4, P5, and P7). A different result was found in plasma cells, since the analysis showed that practically all of the plasma cells expressed IgG and IgA (Figure 2A). Overall, the distribution of the Ig classes reflected the antibody levels in the serum, as previously described (28,29). Only 9% of the plasma cells expressed IgM, and with the exception of 2 sequences, all of them carried nucleotide exchanges, suggesting that synovial plasma cells are derived from memory B cells rather than from naive B cells. The finding that, independent of their localization, practically all synovial plasma cells expressed a memory phenotype is evidence against the possibility of polyclonal activation of naive B cells and direct differentiation into antibody-secreting cells. Somatic diversity of B and plasma cells. The mean mutation frequency in B cells was 9.1 nucleotide exchanges per mutated VH-gene segment. In plasma cells, on average, a slightly higher mutation frequency was found (Figure 2B). Sample P2 showed an average of 10.1 mutations per VH segment, while samples P1 and P3 had average mutation frequencies of 17.1 and 17.8, respectively. When the somatic diversity was compared between sequences derived from plasma cells associated with B cell infiltrates and sequences derived from plasma cell aggregates, no significant difference was seen (Figure 2B). Expression of a selected VH-gene repertoire by synovial plasma cells. A sequence analysis showed that the VH-gene repertoire of synovial B cells was not significantly different from that of peripheral blood B cells in healthy donors (30) (Figure 2C). Both in naive B cells and in memory B cells, the usage of VH-, D-, and JH-gene segments and the length of the CDR3 (Figure 2D) were comparable. In contrast, plasma cells differed from synovial B cells in their VH-gene repertoire. An overrepresentation of VH1 rearrangements and an underrepresentation of VH4 rearrangements were observed in plasma cells 67 Figure 2. A–C, Sequence analysis of microdissected B and plasma cells in synovial tissue from patients with rheumatoid arthritis, showing A, the Ig class distribution, B, the number of somatic mutations, and C, the VH-gene repertoire of B cell infiltrates (Inf) and plasma cell subsets (associated plasma cells [Ass] and aggregates [Agg]). ⴱ ⫽ P ⬍ 0.05. In C, the solid black bars indicate the VH-gene repertoire of peripheral blood naive B cells (as reported in ref. 30). D, Length of the third complementarity-determining region (CDR3) of the VH genes, determined for all V–D–J rearrangements that were isolated (samples P1–P3; shaded bars). In contrast, sequences isolated from plasma cell aggregates in sample P3 (solid line) showed a different distribution of the VH-CDR3 length. aa ⫽ amino acids. (Figure 2C). The alterations in the VH-gene usage were more pronounced in the plasma cell aggregates, with differences reaching statistical significance (P ⬍ 0.001). Evidence of selection was also seen in the analysis of the CDR3 of plasma cell VH sequences. Although there was 68 SCHEEL ET AL Figure 3. Sequence alignments (left) and schematic presentation of the sequence analysis (right) of microdissected B and plasma cells in rheumatoid arthritis synovial tissue sample P2, showing clonal B cell expansion and V-gene diversification. Right, Only those V–D–J rearrangements that were isolated repeatedly from independent polymerase chain reactions are depicted (●). Thin lines connect the different locations in which sequences with an identical rearrangement were found (B cell infiltrates [Inf; white ovals], associated plasma cells [Ass; grey ovals], and plasma cell aggregates [Agg; grey rectangles]). The indicated areas correspond to those described in Figure 1. Left, Sequences from clones P2-1 and P2-2 (indicated by thick lines on the schematic), both of which showed intraclonal diversity, are compared. All triplets differing from the nucleotide sequences of the putative VH-germline genes are shown (the third complementarity-determining region [CDR3] is given in full length). The name of the sequence indicates the area (Agg1, Agg2, or Agg4, and Inf1 or Inf2), the cap, and the sequence number. In addition, the Ig class is indicated. Numbering and position of the CDRs were determined according to the sequences described by Kabat et al (27). no significant difference in the usage of amino acids in the CDR3 of B cells compared with that in the CDR3 of plasma cells, the length of the CDR3 region was prolonged in plasma cell VH sequences of sample P3 (Figure 2D). These findings support the idea that antigen-specific activation of B cells and selective accumulation of plasma cells are present in RA synovial tissue. Clonal expansion of activated synovial B cells. Ig heavy-chain sequences are determined by a specific V–D–J rearrangement and, hence, a unique CDR3. This feature can be used as a clonal marker of B cell development. The analysis of the VH-gene repertoire showed that 10 V–D–J rearrangements (30.3%) in sample P1, 35 (18.6%) in sample P2, and 11 (6.4%) in sample P3 were isolated repeatedly (Table 2). Clonally related cells were found when B cells were dissected from different areas of the same infiltrate and their cDNA was amplified independently. They were also seen when B and plasma cells were prepared from separate locations, as shown by the schematic presentation of B cell expansion (Figure 3). Clonally related B and plasma cells were found when plasma cells were isolated from areas directly associated with a B cell infiltrate (Inf1 and Ass1 and Inf2 and Ass2) (Figure 3). In addition, clonally related cells were detected in separate and distinct plasma cell aggregates (Inf1 and Agg4). In some areas, plasma cell aggregates were in the direct neighborhood of the B cell infiltrate (Inf1 and Agg2). However, B and plasma cells with an identical V–D–J rearrangement were also detected in aggregates that were more than 1 mm apart from the B cell infiltrate (Inf1 and Agg1 and Agg3) (Figures 1 and 3). These findings show that newly generated plasma cells have the capacity to migrate deeply into the synovial tissue, survive, and thus accumulate at these sites. Evidence of intensive clonal expansion was also seen in sample P1, in which only plasma cells were detected. Microdissection showed that within one cluster of plasma cells, multiple cells with an identical V–D–J rearrangement were found, and similar to that observed in sample P2, descendents of the same B cell clone were detected in areas far apart from each other. One such V–D–J rearrangement was isolated from 3 different plasma cell aggregates. In addition, single naive B cells, memory B cells, and plasma cells were sorted from synovial cell suspensions, and the V-gene repertoire was determined. In CLONAL ACCUMULATION OF SYNOVIAL B AND PLASMA CELLS IN RA samples P5 and P7, no identical V–D–J rearrangements were seen when naive and memory B cells were compared, and none of the plasma cells showed an identical rearrangement (Table 3). However, within the plasma cell population, clonally related cells were detected. Two plasma cells in sample P7 did express the same V–D–J rearrangement. Even more surprising was the result obtained in sample P6. In this patient, ⬃100 pieces of synovial tissue were obtained during wrist surgery. After the tissue was digested, only 96 plasma cells were randomly sorted from the cell suspension, and therefore the chance to find clonally related cells was extremely low. Nevertheless, within this small group of plasma cells, 6 V–D–J rearrangements were isolated repeatedly. Sequences belonging to clone P6-1 and also sequences belonging to clone P6-2 were both isolated from 4 single sorted plasma cells (Table 3). In summary, using 2 different approaches, we obtained evidence of intensive B cell activation and accumulation of large plasma cell clones in RA synovial tissue. Switching of immunoglobulin class and intraclonal diversity. In general, sequences with an identical V–D–J rearrangement did not differ in their Ig class. Evidence of Ig class switching was found in only a few clones isolated from sample P2 (Figure 3) and sample P3. In the majority of cases, a switch to IgG1/2 and IgG4 was evident. However, class switching to IgG1/2 and IgA was also documented, as was that from IgM to IgG3 and IgG1/2. In addition, in most cases, sequences with an identical rearrangement showed an identical pattern of somatic mutations. In only 7 (11%) of 63 isolated clones was there evidence of somatic diversification, suggesting that B cell activation in synovial infiltrates does not, by and large, induce hypermutation. Somatic diversification seems to require the microenvironment of the germinal center. To some extent, intraclonal diversity was seen in sequences from sample P2 (Figure 3). Cells belonging to clone P2-1 were found in 3 of 6 plasma cell aggregates analyzed (Figure 3, indicated by bold line connecting Agg1, Agg2, and Agg4). A comparison with the putative germline gene VH1-2 showed that these V-region sequences had only 6 nucleotide exchanges in common but differed by up to 20 somatic mutations (as shown by the upper sequence alignment in Figure 3). Two sequences with an identical V-region were isolated as IgG1/2 and IgG4 (Figure 3). Similarly, cells belonging to clone P2-2 were isolated from 2 separate B cell infiltrates and, in addition, from an area of plasma cells (Figure 3, indicated by bold line connecting Inf1, Inf2, and Agg4). The 69 B cell–derived sequences differed by up to 9 somatic mutations from the putative germline V-region gene VH1-69 (as shown by the lower sequence alignment in Figure 3). In contrast, one plasma cell sequence (Agg1.4.1 in clone P2-2) was identical to one of the B cell sequences, although it differed in Ig class. Similar results were demonstrated in sample P3. Long-term reactivation of memory B cells. Extreme intraclonal diversity in the V-gene repertoire of single plasma cells was found in sample P6 (supplementary results available from the corresponding author at http://www.drfz.de/index.php?id⫽848). Altogether, in this sample, 96 plasma cells were sorted, and 55 VH sequences, 33V sequences, and 1 V sequence were determined (Table 3). The sequence analysis showed that 55 of the plasma cells analyzed could be grouped into 44 independent clones. Although identical V–D–J sequences were found repeatedly, the possibility of contamination was excluded, because identical rearrangements were seen for both VH and V genes, and because the sequences differed substantially in their pattern of somatic mutations (supplementary results available from the corresponding author at http://www.drfz.de/ index.php?id⫽848). Thus, VH regions from clones P6-1 and P6-2 differed by more than 70 nucleotides from the respective germline gene sequences of VH4.39 and VH349. Similarly, the V sequences differed by 23–52 nucleotides from the respective germline sequences. Only a continuous activation of B cells and generation of plasma cells over a long period of time could explain the accumulation of these exceedingly high numbers of somatic mutations in the V-region genes. The intensive somatic diversification of the underlying B cell clones may have taken place in the peripheral secondary lymphoid organs. It is also possible that these large plasma cell clones were derived from previous ectopic germinal center reactions. DISCUSSION Using 2 different approaches, B and plasma cells were isolated from synovial tissue samples, and their V-gene repertoire was determined. Both the LCM protocol and the single cell sorting protocol revealed an intensive expansion of selected B cell clones within RA synovial tissue. With the LCM technique, which has the advantage that cells of defined localization can be isolated, we directly demonstrated activation of B cells, their local differentiation into plasma cells, and accumulation of plasma cells within the inflamed synovial tissue over time. 70 Both techniques showed the presence of naive and memory B cells in synovial lymphocytic infiltrates. Proportions of ⬃25% naive B cells and ⬃75% memory B cells were found, which supports the idea that there is a specific recruitment of memory B cells into the synovial tissue. Enhanced responsiveness to inflammatory chemokines may favor migration of memory B cells (31,32). In the synovial tissue, memory B cells are more likely to be activated by local self antigens, which thus allows them to receive signals that promote their maintenance within the synovial tissue. In addition to selective migration, there may be a preferential expansion of memory B cells, an interpretation that was supported by the finding of clonally related memory B cells within the synovial infiltrates (Figure 3). B and plasma cells with an identical rearrangement and an identical pattern of somatic mutations were found not only in plasma cells in close association with infiltrates but also in plasma cells isolated from independent aggregates. In some cases, clonally related plasma cells were microdissected from areas more than 1 mm apart, which suggests that within the inflamed synovial tissue, the locally generated plasmablasts and/or mature plasma cells have a high migratory capacity. Thus, accumulation of plasma cells within the synovial tissue is supported by migration of plasmablasts from the peripheral lymphoid organs and, in addition, by generation of new plasma cells within the inflamed synovial tissue. The comparison of the VH-gene repertoire of B and plasma cells suggested that there is an antigen-dependent activation of synovial B cells and an antigen-selected differentiation into plasma cells in RA synovial tissue. This interpretation is in line with recent evidence showing that memory B cells are not activated by bystander inflammatory signals, and furthermore, in vivo experiments have shown that self-renewal and differentiation of memory B cells requires antigen-dependent signals (33). In some synovial tissue, chronic inflammation supports the development of organized lymphoid structures and allows the formation of germinal centers. The analysis of germinal center B cells microdissected from synovial tissue samples showed B cell proliferation and somatic diversification of the V-gene repertoire in these structures (5,19,20). In the present study, however, tissue samples that did not show evidence of ectopic germinal center formation were analyzed, and in general, sequences with an identical V–D–J rearrangement showed the same pattern of somatic mutations, even when the V-region gene was joined to a different C-region gene. In a single infiltrate, only rarely were we able to isolate the sequences, such as Inf1.3.1 and Inf1.5.1 from clone SCHEEL ET AL P2-2, that showed intraclonal diversity (Figure 3). These data suggest that in synovial tissue without ectopic germinal centers, B cells will not accumulate substantial somatic mutations, as has been found in murine models of RA (34). Indeed, the few sequences with different patterns of somatic mutations may result from germinal center reactions in the peripheral lymphoid organs or from previous, no longer detectable, synovial germinal centers. Nevertheless, although there was little evidence of B cell proliferation in the absence of synovial germinal center structures, it is possible that there are short phases in which B cells intensively proliferate and differentiate into plasma cells. If the synovial tissue supports long-term plasma cell survival, then one would expect accumulation of these cells over time. This interpretation is supported by the fact that the frequency of clonally related cells was much higher within the plasma cell population than in memory B cells. Furthermore, a comparison of Ig classes expressed in clonally related B and plasma cells indicated that the plasma cells that were present at the time point of the analysis may have been the result of previous phases of B cell activation. With knowledge of the organization of the CH-region genes, one would expect cells expressing IgG1 or IgG2 to switch to IgG4 and then to IgA. The finding of plasma cells expressing IgG1 and the corresponding B cells expressing IgG4 supports the notion that plasma cells have not been generated only recently but rather have accumulated over a long period of time. In synovial tissue without B cells, such as RA synovial tissue samples P1 and P6, plasma cells may be generated within the synovial tissue and may reflect previous phases of synovial B cell activation. The high degree of somatic diversity found in plasma cells supports the idea that there is a continuous activation of selected B cell clones over a long period of time. It would be important to identify the specificity of these clones and to determine whether autoantigens are responsible for the continuous activation of synovial B cells and their differentiation into plasma cells. The method of sorting single cells has the advantage in that VH and VL sequences from individual B and plasma cells can be cloned into expression vectors, and thus recombinant antibodies can be generated. Preliminary results show that the highly mutated plasma cell clones P6-1 and P6-2 are not rheumatoid factors and do not have specificity for citrullinated peptide antigens. However, these B cells may have lost their autospecificity by the accumulation of excessive numbers of somatic mutations. CLONAL ACCUMULATION OF SYNOVIAL B AND PLASMA CELLS IN RA When the sequences belonging to clones P6-1 and P6-2 were analyzed, it was found that the frequency of somatic mutations in the V-region genes was of the same order of magnitude as that seen in highly mutated IgD sequences isolated from tonsillar germinal center B cells. It has been suggested that these hypermutated IgD sequences are unable to mature into circulating memory B cells (35). In our study, in contrast, we found that hypermutated B cells had differentiated into plasma cells, whose V-region amino acid sequences showed less than 60% homology to the corresponding germline sequence. Thus, the chronic activation of the immune system in patients with RA might be seen as a consequence of B cell activation and differentiation of these cells into plasma cells. B cell depletion is known to interrupt the continuous generation of new plasma cells and hence prevents their accumulation in the synovial tissue. Therefore, the presence of large B cell infiltrates and/or high frequencies of synovial plasma cells could thus define a profile that would identify those patients who might benefit from therapy that targets the B cell compartment. 6. 7. 8. 9. 10. 11. 12. 13. 14. ACKNOWLEDGMENTS We are grateful to T. Kaiser, G. Steinhauser, and S. Schürer for providing technical support. We also thank R. S. Jack for critical reading of the manuscript. 15. 16. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Berek had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Scheel, Häupl, Berek. Acquisition of data. Scheel, Gursche, Zacher, Häupl. Analysis and interpretation of data. Scheel, Berek. 17. 18. 19. REFERENCES 1. Zvaifler NJ. The immunopathology of joint inflammation in rheumatoid arthritis. Adv Immunol 1973;16:265–336. 2. Yanni G, Whelan A, Feighery C, Bresnihan B. Analysis of cell populations in rheumatoid arthritis synovial tissues. Semin Arthritis Rheum 1992;21:393–9. 3. Weyand CM, Klimiuk PA, Goronzy JJ. Heterogeneity of rheumatoid arthritis: from phenotypes to genotypes. Springer Semin Immunopathol 1998;20:5–22. 4. Krenn V, Morawietz L, Burmester GR, Kinne RW, MuellerLadner U, Muller B, et al. Synovitis score: discrimination between chronic low-grade and high-grade synovitis. Histopathology 2006; 49:358–64. 5. Schroder AE, Greiner A, Seyfert C, Berek C. Differentiation of B cells in the nonlymphoid tissue of the synovial membrane of 20. 21. 22. 23. 24. 71 patients with rheumatoid arthritis. Proc Natl Acad Sci U S A 1996;93:221–5. Randen I, Mellbye OJ, Forre O, Natvig JB. The identification of germinal centres and follicular dendritic cell networks in rheumatoid synovial tissue. Scand J Immunol 1995;41:481–6. Chaiamnuay S, Bridges SL Jr. The role of B cells and autoantibodies in rheumatoid arthritis. Pathophysiology 2005;12:203–16. Goronzy JJ, Weyand CM. Developments in the scientific understanding of rheumatoid arthritis. Arthritis Res Ther 2009;11:249. Edwards JC, Szczepanski L, Szechinski J, Filipowicz-Sosnowska A, Emery P, Close DR, et al. Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis. N Engl J Med 2004;350:2572–81. Emery P, Fleischmann R, Filipowicz-Sosnowska A, Schechtman J, Szczepanski L, Kavanaugh A, et al, for the DANCER Study Group. 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, doseranging trial. Arthritis Rheum 2006;54:1390–400. Cohen SB, Emery P, Greenwald MW, Dougados M, Furie RA, Genovese MC, et al, for the REFLEX Trial Group. Rituximab for rheumatoid arthritis refractory to anti–tumor necrosis factor therapy: results of a multicenter, randomized, double-blind, placebocontrolled, phase III trial evaluating primary efficacy and safety at twenty-four weeks. Arthritis Rheum 2006;54:2793–806. Kratz A, Campos-Neto A, Hanson MS, Ruddle NH. Chronic inflammation caused by lymphotoxin is lymphoid neogenesis. J Exp Med 1996;183:1461–72. Liu YJ, Banchereau J. Mutant mice without B lymphocyte follicles. J Exp Med 1996;184:1207–11. Duke O, Panayi GS, Janossy G, Poulter LW. An immunohistological analysis of lymphocyte subpopulations and their microenvironment in the synovial membranes of patients with rheumatoid arthritis using monoclonal antibodies. Clin Exp Immunol 1982;49: 22–30. Young CL, Adamson TC III, Vaughan JH, Fox RI. Immunohistologic characterization of synovial membrane lymphocytes in rheumatoid arthritis. Arthritis Rheum 1984;27:32–9. Thurlings RM, Vos K, Wijbrandts CA, Zwinderman AH, Gerlag DM, Tak PP. Synovial tissue response to rituximab: mechanism of action and identification of biomarkers of response. Ann Rheum Dis 2008;67:917–25. Bridges SL Jr, Lee SK, Koopman WJ, Schroeder HW Jr. Analysis of immunoglobulin gamma heavy chain expression in synovial tissue of a patient with rheumatoid arthritis. Arthritis Rheum 1993;36:631–41. Williams DG, Taylor PC. Clonal analysis of immunoglobulin mRNA in rheumatoid arthritis synovium: characterization of expanded IgG3 populations. Eur J Immunol 1997;27:476–85. Gause A, Gundlach K, Zdichavsky M, Jacobs G, Koch B, Hopf T, et al. The B lymphocyte in rheumatoid arthritis: analysis of rearranged V genes from B cells infiltrating the synovial membrane. Eur J Immunol 1995;25:2775–82. Kim HJ, Krenn V, Steinhauser G, Berek C. Plasma cell development in synovial germinal centers in patients with rheumatoid and reactive arthritis. J Immunol 1999;162:3053–62. Humby F, Bombardieri M, Manzo A, Kelly S, Blades MC, Kirkham B, et al. Ectopic lymphoid structures support ongoing production of class-switched autoantibodies in rheumatoid synovium. PLoS Med 2009;6:e1. Itoh K, Patki V, Furie RA, Chartash EK, Jain RI, Lane L, et al. Clonal expansion is a characteristic feature of the B-cell repertoire of patients with rheumatoid arthritis. Arthritis Res 2000;2:50–8. Voswinkel J, Weisgerber K, Pfreundschuh M, Gause A. The B lymphocyte in rheumatoid arthritis: recirculation of B lymphocytes between different joints and blood. Autoimmunity 1999;31:25–34. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, 72 25. 26. 27. 28. 29. 30. SCHEEL ET AL Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24. Greiner A, Neumann M, Stingl S, Wassink S, Marx A, Riechert F, et al. Characterization of Wue-1, a novel monoclonal antibody that stimulates the growth of plasmacytoma cell lines. Virchows Arch 2000;437:372–9. Kim HJ, Berek C. Single cell analysis of synovial tissue B-cells. Methods Mol Med 2007;136:25–37. Kabat EA, Wu TT, Perry HM, Gottesman KS, Foeller C. Sequences of proteins of immunological interest. NIH Publication No. 91-3242. Vol. 2, 5th ed. Bethesda (MD): Department of Health and Human Services, Public Health Service, National Institutes of Health; 1991. p. 91. French M. Serum IgG subclasses in normal adults. Monogr Allergy 1986;19:100–7. Schroeder HW Jr, Cavacini L. Structure and function of immunoglobulins. J Allergy Clin Immunol 2010;125:S41–52. Brezinschek HP, Foster SJ, Brezinschek RI, Dorner T, DomiatiSaad R, Lipsky PE. Analysis of the human VH gene repertoire: differential effects of selection and somatic hypermutation on 31. 32. 33. 34. 35. human peripheral CD5⫹/IgM⫹ and CD5⫺/IgM⫹ B cells. J Clin Invest 1997;99:2488–501. Henneken M, Dorner T, Burmester GR, Berek C. Differential expression of chemokine receptors on peripheral blood B cells from patients with rheumatoid arthritis and systemic lupus erythematosus. Arthritis Res Ther 2005;7:R1001–13. Nanki T, Takada K, Komano Y, Morio T, Kanegane H, Nakajima A, et al. Chemokine receptor expression and functional effects of chemokines on B cells: implication in the pathogenesis of rheumatoid arthritis. Arthritis Res Ther 2009;11:R149. Benson MJ, Elgueta R, Schpero W, Molloy M, Zhang W, Usherwood E, et al. Distinction of the memory B cell response to cognate antigen versus bystander inflammatory signals. J Exp Med 2009;206:2013–25. William J, Euler C, Christensen S, Shlomchik MJ. Evolution of autoantibody responses via somatic hypermutation outside of germinal centers. Science 2002;297:2066–70. Liu YJ, de Bouteiller O, Arpin C, Briere F, Galibert L, Ho S, et al. Normal human IgD⫹IgM⫺ germinal center B cells can express up to 80 mutations in the variable region of their IgD transcripts. Immunity 1996;4:603–13.