Identification and characterization of SmD183119-reactive T cells that provide T cell help for pathogenic antidouble-stranded DNA antibodies.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 48, No. 2, February 2003, pp 475–485 DOI 10.1002/art.10762 © 2003, American College of Rheumatology Identification and Characterization of SmD1 -Reactive T Cells That Provide T Cell Help for Pathogenic Anti–Double-Stranded DNA Antibodies 83–119 Gabriela Riemekasten,1 Dirk Langnickel,2 Fanny M. Ebling,3 George Karpouzas,3 Jatinderpal Kalsi,3 Gunda Herberth,2 Betty P. Tsao,3 Peter Henklein,2 Sven Langer,2 Gerd-R. Burmester,2 Andreas Radbruch,4 Falk Hiepe,5 and Bevra H. Hahn3 Objective. The C-terminal peptide of amino acids 83–119 of the SmD1 protein is a target of the autoimmune response in human and murine lupus. This study was undertaken to test the hypothesis that SmD183–119-reactive T cells play a crucial role in the generation of pathogenic anti–double-stranded DNA (anti-dsDNA) antibodies. Methods. Splenic or lymph node T cells derived from unmanipulated as well as SmD183–119-immunized NZB/NZW mice were analyzed in vitro by enzyme-linked immunospot (ELISpot) assay to determine T cell help for anti-dsDNA generation induced by the SmD183–119 peptide. Cytokines expressed by these T cells were measured by ELISpot assay, enzyme-linked immunosorbent assay, and flow cytometry. SmD183–119- and ovalbumin-specific T cell lines were generated and characterized. Results. The SmD183–119 peptide, but not the control peptides, significantly increased the in vitro generation of anti-dsDNA antibodies in cultures from unmanipulated NZB/NZW mice. Interferon-␥ (IFN␥), interleukin-2 (IL-2), IL-4, transforming growth factor ␤, and IL-10 production increased in response to the peptide in young mice; only IFN␥ and IL-2 were increased in older, diseased mice. Activation of SmD183–119-reactive T cells by immunization of NZB/ NZW mice resulted in elevated anti-dsDNA synthesis and, later, increased antibodies to SmD183–119. Most cells in SmD183–119-specific CD4ⴙ T cell lines helping both antibodies had increased intracellular expression of IFN␥, and most expressed both IFN␥ and IL-4. Conclusion. The SmD183–119 peptide plays an important role in generating T cell help for autoantibodies, including anti-dsDNA, and activates different subsets of T cells as defined by distinct cytokine expression. This peptide is an interesting target structure for the modulation of autoreactive T cells, and its characterization may contribute to our understanding of the role of autoantigen-reactive T cells in the pathogenesis of SLE. Supported by grants from University Research Foundation of Humboldt University, the DFG (Ri 1056/2-1, SFB 421, C4 and C6), Merck Sharp & Dohme (Arthritis/Arthrose 2000 grant), the NIH (R37-AI-46776 and P60-AR-36834), and the Arthritis Foundation Southern California Chapter, and by gifts from the Paxson Family, the Mitchell Family, and the Dorough Foundation. 1 Gabriela Riemekasten, MD: University of California, Los Angeles and Charité University Hospital, Berlin, Germany; 2Dirk Langnickel, Dipl Biotech, Gunda Herberth, Dipl Biol, Peter Henklein, PhD, Sven Langer, Dipl Biol, Gerd-R. Burmester, MD: Charité University Hospital, Humboldt University, Berlin, Germany; 3Fanny M. Ebling, PhD, George Karpouzas, MD, Jatinderpal Kalsi, PhD, Betty P. Tsao, PhD, Bevra H. Hahn, MD: University of California, Los Angeles; 4Andreas Radbruch, PhD: Deutsches Rheumaforschungszentrum, Berlin, Germany; 5Falk Hiepe, MD: Charité University Hospital and Deutsches Rheumaforschungszentrum, Berlin, Germany. Address correspondence and reprint requests to Gabriela Riemekasten, MD, Department of Rheumatology and Clinical Immunology, Charité University Hospital, Schumannstrasse 20/21, D-10117 Berlin, Germany. E-mail: email@example.com. Submitted for publication May 29, 2002; accepted in revised form October 9, 2002. High-affinity antibodies against the Sm proteins as well as against double-stranded DNA (dsDNA) are a hallmark of the immune response in systemic lupus erythematosus (SLE) (1,2). Despite the importance of these antibodies as diagnostic and prognostic markers, the target structures and the mechanisms of autoantibody production are not well understood. The Sm antigens are part of the spliceosomal complex that plays an essential role in the generation of messenger RNA and DNA (for review, see ref. 3). It comprises at least 9 different polypeptides, designated B, B⬘, N, D1, D2, D3, E, F, and G (4). Anti-Sm antibodies are predominantly directed against the SmD1 protein (5). Anti-dsDNA as 475 476 RIEMEKASTEN ET AL well as anti-Sm antibody responses are polyclonal, antigen-driven, somatically mutated, and high-titered (6), suggesting that both responses are driven by autoreactive T helper cells. Since uncomplexed mammalian DNA is poorly immunogenic (7,8), and T cells are unlikely to recognize dsDNA in the context of the major histocompatibility complex (MHC) (9), models have emerged in which peptides of anti-dsDNA antibodies themselves (10), anti-dsDNA antibody binding peptides (11), or nucleosomal peptides (12) provide the T cell epitopes that drive anti-dsDNA production. As shown for other autoimmune diseases, the frequency of autoreactive T cells is low (13). Until now, data characterizing autoantigen-specific T cells were obtained by analyses of T cell clones or of T cells after immunization with the autoantigens (9,14). Recently, we identified the C-terminal peptide of amino acids (aa) 83–119 of the SmD1 protein as a major target for the autoimmune response in SLE (15,16). Approximately 70% of patients with SLE have antibodies to SmD183–119, compared with ⬍7% of disease controls and healthy individuals (17). These antibodies are closely associated with disease activity and with concentrations of anti-dsDNA in individual patients, suggesting a functional link between dsDNA and the SmD183–119 peptide. In the NZB/NZW mouse, a murine model of SLE, acceleration of lupus and increased concentration of pathogenic anti-dsDNA antibodies after immunization with the SmD183–119 peptide indicate a role of this peptide in the generation of anti-dsDNA antibodies (17). We have hypothesized that the highly positively charged SmD1 peptide could be a further candidate for the activation of T cell help for dsDNAspecific B cells and could therefore be responsible for the association between anti-dsDNA and anti-Sm antibodies (17). The present study examined this hypothesis. MATERIALS AND METHODS Mice. (NZB ⫻ NZW)F1 mice were obtained from Jackson Laboratories (Bar Harbor, ME) or from Harlan Winkelmann (Borchen, Germany). Strains were bred either in the University of California, Los Angeles Rheumatology Vivarium or in the Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin in Berlin-Marienfelde, Germany. Only female mice were used in these experiments. Antigens. The SmD1 83–119 peptide (VEPKVKSKKREAVAGRGRGRGRGRGRGRGRGRGGPRR), as well as a randomized peptide with amino acids identical to the SmD1 83–119 peptide (CREKGRVGRGRPAVGRRGVGRPGRRGSRARGEGKGRK), used as control antigen, were synthesized according to a protocol described previously (17). For T cell epitope mapping, 7 overlapping 15-mer peptides, peptides aa 83–97, aa 85–99, aa 87–101, aa 90–104, aa 93–107, aa 97–111, and aa 105–119, were synthesized in the same way. Other control peptides were histone peptide 3 (91–105; QSSAVMALQEASEAY), Ro 480 peptide (CAIALREYRKKMDI), a peptide based on complementaritydetermining region 1 of a murine anti-dsDNA monoclonal antibody that bears the common idiotype (16/6Id ), and pCONS (FIEWNKLRFRQGLEW), an artificial peptide based on an algorithm describing T cell stimulatory sequences from the VH regions of murine IgG antibodies to DNA (19). In vivo activation of SmD183–119-reactive T cells by immunization of NZB/NZW mice. For stimulation of SmD183–119-reactive T cells and for T cell epitope mapping, NZB/NZW mice at different ages were immunized subcutaneously (SC) in the hind limbs with either the SmD183–119 peptide or the randomized peptide, each emulsified with Freund’s complete adjuvant (CFA) (1:1). After 9–10 days, cells from draining popliteal and inguinal lymph nodes and spleens were harvested. Six additional NZB/NZW mice (6 weeks old) were repeatedly immunized SC (in the neck and finally in the hind limbs) with 100 g keyhole limpet hemocyanin– SmD183–119 peptide and subsequently boosted with 50 g SmD183–119 emulsified in incomplete Freund’s adjuvant (IFA) once per week (5 times during a period of 9 weeks). Nine days after the last immunization, at age 14 weeks, the mice in this group were killed and serum, splenic, and lymph node T cells were harvested. For intracellular cytokine determination, 12-week-old NZB/NZW mice were immunized SC in the neck with 100 g of the SmD183–119 peptide in CFA, followed 1 week later by 2 weekly injections of 50 g of the SmD183–119 peptide emulsified with IFA. As control, mice were immunized with the randomized peptide of SmD183–119. Spleen cells were harvested 14 days after the last immunization (when the age of the mice was ⬃16 weeks). Enzyme-linked immunosorbent assay (ELISA). An anti-dsDNA ELISA was used with calf thymus as the source of dsDNA; it was performed as described previously (20). AntiSmD183–119 peptide antibodies were detected by ELISA using polystyrene-coated SmD183–119 peptide as described (17). Supernatants or sera were analyzed in triplicate in 1:4 and 1:100 dilutions, respectively. Purification of T and B cells. Single cell suspensions derived from spleens and lymph nodes were enriched for Thy-1,2⫹ and B220⫹ cells with the Vario MACS magnetic purification system (Miltenyi, Auburn, CA), using appropriate microbead-coated antibodies. Purity of the isolated cell populations as determined by fluorescence-activated cell sorter (FACS) varied from 92% to 99%. For enzyme-linked immunospot (ELISpot) assay, B and T cells were cultured at a ratio of 1:10 in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum (FCS) plus 20% concanavalin A supernatant with or without peptides at optimal concentrations, in a 12-well culture plate (Gibco BRL Life Technologies, Paisley, UK) at 37°C. After 5 days, cells were separated by centrifugation and cultured in fresh DMEM with 10% FCS and used for ELISpot assay. Supernatants were harvested for detection of antibodies. Determination of T cell help for anti-dsDNA and anti-SmD183–119 secretion by ELISpot assay. B cells (105) and T cells (106) cultured in complete medium (DMEM with 10% FCS) with various concentrations of peptides were added to ROLE OF SmD183–119-REACTIVE T CELLS IN ANTI-dsDNA GENERATION each well of a 96-well plate coated with dsDNA or the SmD183–119 peptide, respectively, and blocked for 2 hours with sterile 3% bovine serum albumin (Sigma, St. Louis, MO) in phosphate buffered saline at room temperature. After 8–10 hours of incubation at 37°C, cells were poured off and plates were washed 12–16 times. Spots were developed as described previously (21). Antibody-forming cells (AFCs) were enumerated as blue spots using an inverted microscope (Leitz, Wetzlar, Germany) and recorded as the number of IgG-producing AFCs/105 B cells. Results of ELISpot assays were further confirmed in simultaneous ELISAs of the supernatants derived from the cultures. Cytokine ELISAs. T cells (2 ⫻ 105) and B cells (2 ⫻ 4 10 ) were cultured in serum-free AIM V medium (Gibco BRL Life Technologies) with or without peptides. Interleukin-2 (IL-2), interferon-␥ (IFN␥), and IL-4 (all from PharMingen, San Diego, CA) were measured in individual supernatants using commercial ELISA kits, according to the manufacturer’s instructions. Each sample of supernatants was assayed in triplicate in 2 or 3 separate experiments, at a dilution of 1:4. A standard curve was constructed for each experiment, using recombinant cytokines (PharMingen). Cytokine ELISpot assays. Ninety-six–well plates (Costar, Cambridge, MA) were coated with purified anticytokine antibodies (anti-IFN␥, anti–IL-4, anti–IL-10, anti– transforming growth factor ␤ [anti-TGF␤]) (100 l/well; all from PharMingen) overnight according to the instructions of the manufacturer and blocked as described above. T cells (2 ⫻ 105) derived from spleens or lymph nodes of treated or untreated mice, as well as B cells (2 ⫻ 104) from diseased NZB/NZW mice exhibiting proteinuria of ⬎100 mg/dl were suspended in AIM V medium and cocultured with or without peptides (5 g/well). After 72 hours in 5% CO2 at 37°C, cells were removed and the reaction was visualized by addition of the individual biotinylated anticytokine antibody and subsequent alkaline phosphatase–conjugated streptavidin. Spots representing cytokine-expressing cells were detected as was done for AFCs. The expression index was calculated as the quotient between numbers of spots in T and B cell cultures with peptide and spots in T and B cell cultures without peptide. Intracellular cytokine detection by FACS. Cells were fixed, permeabilized, and stained for intracellular cytokines and surface markers as described previously (22). Monoclonal antibodies (all from PharMingen) were used for intracellular and surface staining. Cells were divided into 2 groups for further staining. The first group of cells, which had been marked with biotinylated CD69, were stained using fluorescein isothiocyanate (FITC)–conjugated anti-IFN␥, phycoerythrin (PE)–conjugated anti-CD4, and Cy5-stained anti-CD3. The second subset were stained with FITC-conjugated anti–tumor necrosis factor ␣ (anti-TNF␣), PE-conjugated anti–IL-4, anti– IL-2 antigen-presenting cells (APCs), and biotinylated antiCD4 according to the instructions of the manufacturer (Becton Dickinson). Completely unstained controls and stained isotype controls were also used. Samples were analyzed by 4-color analysis on a FACSCalibur flow cytometer (Becton Dickinson), using FCS-Express software, version 1.0. Generation of T cell lines. T cells derived from splenic cells of unmanipulated 8–10-week-old female NZB/NZW mice were cultured with irradiated (10 minutes, 30 Gy) CD90depleted splenocytes (APCs) from 5-month-old untreated fe- 477 Figure 1. In vitro T cell help for anti–double-stranded DNA (antidsDNA) antibodies by the SmD183–119 peptide in untreated NZB/ NZW mice. Addition of SmD183–119 to a culture of splenic T cells from 10–15-week-old nonimmunized prenephritic NZB/NZW mice plus B cells from older nonimmunized mice resulted in an increase in anti-dsDNA antibody–forming cells (AFCs), compared with the spots obtained from T and B cells cultured without the peptide. Cultures of SmD183–119 with B cells alone, with the addition of the randomized peptide (rand. pept.) to B cells alone, and with T ⫹ B cells did not influence the generation of anti-dsDNA AFCs. Values are the mean ⫾ SEM from 3 independently performed experiments (n ⫽ 8 in each). P values are versus all cultures without SmD183–119. male NZB/NZW mice at a ratio of 1:4 in appropriate culture wells at 37°C/5% CO2 in the presence of 5 g/ml SmD183–119. As a control, ovalbumin (OVA) was used for repeated in vitro restimulation of T cells derived from mice previously immunized with OVA (50 g, emulsified in CFA). As shown in our previous work (17), OVA immunization has no influence on anti-dsDNA response and survival. CD90 depletion was performed by MACS technology according to the instructions of the manufacturer (Miltenyi). Every 14–17 days, cells were restimulated with irradiated CD90-depleted APCs as well as with 5 g/ml SmD183–119 peptide or OVA. Three days after restimulation, cells were stimulated with 50 units/ml of IL-2; this was repeated every 3–4 days, followed by a resting period of 4–5 days. Characterization of T cells was performed after 6 weeks of culture or later. T cells were analyzed by flow cytometry after staining with carboxyfluorescein diacetate succinimidyl ester (Molecular Probes, Eugene, OR) for 5 days with or without stimulation either with the corresponding antigen or with anti-CD3 or SmD183–119. Simultaneously, T cell lines were functionally analyzed in vitro for help for anti-dsDNA and anti-SmD183–119 antibodies, by ELISpot assay. Statistical analysis. Comparisons between groups were performed by Student’s t-test and by nonparametric 478 RIEMEKASTEN ET AL Figure 2. In vitro activation of SmD183–119-reactive T cells by 5 g/ml SmD183–119 peptide in immunized 7–12-week-old NZB/NZW mice, and the influence on in vitro help for autoantibody generation. A, Increased T cell help for anti-dsDNA AFCs after immunization with the SmD183–119 peptide, as detected by enzyme-linked immunospot (ELISpot) assay. T cells were obtained from draining lymph nodes 9 days after subcutaneous immunization. B cells were from unmanipulated 35-week-old mice with proteinuria. B, Detection of anti-SmD183–119 AFCs, by ELISpot assay, in 15-week-old NZB/NZW mice (n ⫽ 5) after repeated immunization with SmD183–119 (5 times during 9 weeks). T cells were obtained from draining lymph nodes after paw immunization. Values are the mean ⫾ SEM from 3 independently performed experiments (n ⫽ 8 in each). P values are versus cultures of T and B cells without any peptide. See Figure 1 for other definitions. Mann-Whitney test. Values are presented as the mean ⫾ SEM. P values less than 0.05 were considered significant. RESULTS The SmD183–119 peptide activates in vitro T cells derived from untreated NZB/NZW mice to help in the generation of anti-dsDNA–producing plasma cells. The addition of the SmD183–119 peptide to a culture of splenic T cells harvested from nonimmunized prenephritic NZB/NZW mice plus B cells from older nonimmunized NZB/NZW mice resulted in an increase in anti-dsDNA AFCs, compared with the spots obtained from T and B cells cultured without the peptide (P ⬍ 0.001 for cultures with 5 g or 20 g of the SmD183–119 peptide) (Figure 1). Cultures of SmD183–119 peptide with B cells alone, with the addition of the randomized peptide to B cells alone, and with T ⫹ B cells did not influence the generation of anti-dsDNA AFCs. AntiSmD183–119 antibodies were not found in the supernatants of T ⫹ B cells cocultured for 5 days, while anti-dsDNA antibodies were readily detectable (data not shown). Extended cultivation of T and B cells with the SmD183–119 peptide over 9 days resulted in detectable anti-dsDNA as well as anti-SmD183–119 in the supernatant (data not shown). Comparative potencies of various self peptides to induce T cell help for anti-dsDNA. The capability of different peptides to induce T cell help for anti-dsDNA antibodies in T cells derived from unmanipulated mice was studied. In 3 independently performed experiments, the effect of the SmD183–119 peptide on anti-dsDNA generation was slightly higher than that induced by the pCONS peptide, but in all of these experiments, the T cell help was clearly increased by both peptides (data not shown). The histone peptide 3 (91–105; QSSAVMALQEASEAY) was also compared in 3 independently performed experiments with the SmD183–119 peptide, showing in all experiments a lower capability to induce anti-dsDNA antibodies in vitro. As a negative peptide, the Ro 480 peptide (CAIALREYRKKMDI), showing no T cell help for the generation of anti-dsDNA antibodies, was used (data not shown). In all of these experiments, 8-week-old mice were used as T cell donors; the results might vary with donors at older ages. ROLE OF SmD183–119-REACTIVE T CELLS IN ANTI-dsDNA GENERATION In vivo stimulation of SmD183–119-reactive T cells and their influence on autoantibodies. A single in vivo immunization of 7-week-old mice with the SmD183–119 peptide and subsequent in vitro restimulation with SmD183–119 enhanced the generation of anti-dsDNA– secreting plasma cells. T cells derived from draining lymph nodes that were stimulated in vitro with the SmD183–119 peptide enhanced the number of antidsDNA AFCs compared with spontaneous T cell help (⬃5-fold; P ⬍ 0.001) (Figure 2A). In contrast, antidsDNA–specific AFCs were not significantly increased by immunization and subsequent in vitro restimulation of T cells with the randomized peptide as compared with cultures of B and T cells without any peptide. T cell help for anti-dsDNA AFCs was also increased by immunization of 27-week-old NZB/NZW mice with SmD183–119. However, anti-dsDNA AFC numbers increased only 1.7-fold (data not shown), which was not as great an increase as that observed with young NZB/NZW mice. After single immunization, no anti-SmD183–119 antibodies were detectable in the supernatants of lymph nodes or in spleen cell cultures. Repeated immunization (up to 5 times) of NZB/NZW mice with the SmD183–119 peptide emulsified in CFA or IFA resulted in significant T cell help for anti-SmD183–119 AFCs (P ⫽ 0.03 with 5 g) by T cells from draining lymph nodes, as detected by ELISpot (Figure 2B) and confirmed by ELISA (data not shown), as well as help for anti-dsDNA antibody generation. The sera from these repeatedly immunized mice contained high levels of anti-SmD183–119 and antidsDNA antibodies compared with sera from untreated NZB/NZW mice (data not shown). In an analysis of the fine specificity of the SmD183–119 peptide in lymph node T cells from SC SmD183–119–immunized NZB/NZW mice, using overlapping 15-mer peptides spanning the SmD183–119 peptide, the largest response was found for the N-terminus of the peptide. This was observed for both splenic T cells (data not shown) and lymph node T cells (Figure 3). Peptide aa 87–101 induced the highest response. Cytokine profile of SmD183–119-reactive T cells derived from untreated NZB/NZW mice. To analyze the cytokine expression pattern of SmD183–119-reactive T cells, splenic Thy-1,2⫹ T cells derived from untreated NZB/NZW mice of different ages were reactivated for 72 hours with SmD183–119 ex vivo in the presence of B cells from older nonimmunized NZB/NZW mice and analyzed by cytokine ELISpot. T cells from 7-week-old mice without proteinuria and from 27-week-old mice with proteinuria (⬎100 mg/dl) were compared (Table 1). 479 Figure 3. T cell fine epitope mapping by 7 overlapping 15-mer peptides spanning the whole SmD183–119 peptide, after subcutaneous immunization of NZB/NZW mice with SmD183–119. T cells were obtained from lymph nodes 9 days after immunization. The dominant T cell epitope of the SmD183–119 peptide was found in the N-terminus. P values are versus T and B cells without any peptide. aa ⫽ amino acid (see Figure 1 for other definitions). In young prenephritic mice, SmD183–119 resulted in expression (above background levels) of various cytokines, including IFN␥, IL-2, IL-4, and IL-10. In response to the SmD183–119 peptide, TGF␤ expression was detectable only in young prenephritic NZB/NZW mice; however, the number of spots was few (1/105 B cells). In 27-week-old mice with clinical nephritis, baseline production of all cytokines was higher than in young mice, and addition of SmD183–119 resulted in increases of IFN␥ and IL-2 only. The numbers of cells expressing IL-4 decreased, and IL-10 secretion did not change (Table 1). The randomized peptide control did not induce significantly increased levels of IFN␥ or IL-4 above background. B cells alone or stimulated with SmD183–119 did not contain any ELISpot-forming cells. Activation of SmD183–119 peptide–reactive T cells by in vivo immunization with the SmD183–119 peptide and its influence on cytokine expression. Splenic T cells from 7-week-old and 27-week-old NZB/NZW mice that had been immunized with the SmD183–119 peptide were analyzed for cytokine expression 9 days after immuniza- 480 RIEMEKASTEN ET AL Table 1. Number of cytokine-specific spots, as detected by enzyme-linked immunospot assay, in cells from young and nephretic unmanipulated and from young and nephritic SmD183–119-immunized female NZB/NZW mice, with and without in vitro stimulation with the SmD183–119 peptide* Cytokine-specific spots/2 ⫻ 105 cells/ml 7-week-old mice T⫹B Unmanipulated IFN␥ IL-4 IL-10 IL-2 Immunized IFN␥ IL-4 IL-10 IL-2 6.3 68 8.8 52 1 14 21 17 T ⫹ B ⫹ SmD1 27-week-old nephritic mice 83–119 T⫹B T ⫹ B ⫹ SmD183–119 26 (4.1 [P ⬍ 0.001]) 104 (1.5 [P ⬍ 0.004]) 22.3 (2.5 [P ⫽ 0.0016]) 75 (1.4 [P ⫽ 0.02]) 19 85 22 75 34 (1.7 [P ⫽ 0.01]) 68 (0.8 [P ⫽ 0.02]) 24 (1.1) 110 (1.5 [P ⫽ 0.04]) 15 (15 [P ⬍ 0.001]) 67 (4.8 [P ⬍ 0.001]) 27 (1.3) 31 (1.8 [P ⬍ 0.01]) 2 42 66 53 25 (12.5 [P ⬍ 0.001]) 74 (1.8 [P ⫽ 0.02]) 66 (1) 150 (2.8 [P ⬍ 0.02]) * Values are the median number of spots, from triplicate specimens pooled from 2 mice in 1 representative experiment (3 experiments [total of 6 mice; 2 in each experiment] were performed, and all had similar results). Values in parentheses are the expression index (EI), estimated as the number of cytokine-expressing cells after peptide stimulation divided by the number of cytokine-expressing cells without peptide stimulation. When the EI was significant, the P value is indicated. IFN␥ ⫽ interferon-␥; IL-4 ⫽ interleukin-4. tion (Table 1). Immunization with the SmD183–119 peptide influenced the number of T cells expressing a given cytokine after restimulation with SmD183–119 peptide, but it also affected the secretion of T cells without any in vitro peptide restimulation (T ⫹ B cells alone). The absolute number of T cells expressing a distinct cytokine does not reflect the SmD183–119-specific stimulation after immunization. Therefore, cytokine expression after restimulation with SmD183–119 was compared with basal secretion of B and T cells without SmD183–119 and described by an index of cytokine expression. In 7-week-old female NZB/NZW mice, the expression index of IFN␥ was strongly increased in immunized mice compared with untreated mice (P ⬍ 0.001). Immunization with the SmD183–119 peptide markedly increased secretion of IL-4 (P ⬍ 0.001), with a 5-fold increased expression above background. IL-2 expression was not significantly increased by immunization with SmD183–119 compared with nonimmunized mice. TGF␤ expression was lower in immunized mice compared with nonimmunized controls (P ⫽ 0.03). IL-10 expression was slightly, but not significantly, lower in SmD183–119immunized mice than in untreated animals. Modulation of T cells by immunization in 27-week-old nephritic NZB/NZW mice confirmed the results obtained by immunization of young NZB/NZW mice. The results of the ELISpots are summarized in Table 1. Analysis of supernatants of B and T cell cultures from repeatedly immunized mice cocultured with SmD183–119 confirmed the enhancement of IFN␥, and to a lesser degree of IL-2 expression, detected by ELISpot assay ex vivo. Both CD4⫹ and CD8⫹ T cells produced IFN␥ (data not shown), whereas IL-4 was not detectable in culture supernatants. Intracellular cytokine expression of SmD183–119reactive T cells after immunization with SmD183–119, as detected by flow cytometry. Two weeks after repeated immunizations with the SmD1 83–119 peptide, CD4⫹,CD3⫹ T cells were analyzed cytometrically and the results compared with those found in age-matched (16-week-old) mice without treatment. In unmanipulated mice, there was no increase in IFN␥ expression in T cells after stimulation with the SmD183–119 peptide, compared with unstimulated T cells (Figures 4a and b). In contrast, in SmD183–119-immunized mice, the frequency of IFN␥-expressing T cells was increased to 0.08% after stimulation with the SmD183–119 peptide, while in unstimulated T cells and in T cells stimulated with the randomized peptide, no IFN␥-expressing T cells could be detected. (Figures 4c–f). CD69 expression did not serve as an appropriate marker to reflect autoantigen-mediated stimulation. As suggested previously (23), CD69 expression is disturbed in lupus, and the permanent in vivo activation by autoantigens probably prevents further up-regulation. In 16-week-old SmD183–119-treated mice, the frequency of IL-2–expressing cells among SmD183–119restimulated T cells increased to 0.08%, but no increase was observed in unstimulated T cells or in T cells that had been restimulated with the randomized peptide ROLE OF SmD183–119-REACTIVE T CELLS IN ANTI-dsDNA GENERATION 481 with the SmD183–119 peptide in vitro to generate T cell lines. T cell help for anti-dsDNA and anti-SmD183–119 synthesis was detected in vitro by ELISpot. After stimulation with the SmD183–119 peptide, SmD183–119specific lines cocultured with B cells increased help for both anti-dsDNA and anti-SmD183–119. In contrast, an OVA-specific T cell line, when stimulated with SmD183–119 (Figure 5), but not with OVA (data not shown), provided slightly increased help for anti-dsDNA antibody–secreting cells, but still, no anti-SmD183–119 response could be detected by ELISpot. In the SmD183–119-specific T cell line derived from unmanipulated mice, 92% of the SmD183–119reactive T cells were CD4⫹, whereas a minority (4%) were CD8⫹. The remaining 4% of the cells were double negative. A high proportion of CD4⫹ T cells in the SmD183–119-specific T cell line that helped anti-dsDNA synthesis expressed intracellular IFN␥, and the majority of these IFN␥⫹ T cells also expressed cytoplasmic IL-4 after stimulation with anti-CD3 (Figure 6). Only a minority of T cells (⬍4%) expressed IL-4 without IFN␥. IL-2 expression was detectable in ⬃50% of the T cells and IL-10 in 10–15% (Figure 6). Similar results were Figure 4. Intracellular cytokine expression, as measured by flow cytometry, in CD4⫹,CD3⫹ splenic T cells from 16-week-old unmanipulated NZB/NZW mice (a and b) and from age-matched NZB/NZW mice that were subcutaneously immunized 3 times with the SmD183–119 peptide (c–f). T cells were stained with no stimulation (a and c) or after stimulation with SmD183–119 (b and d), the randomized (rand.) peptide of SmD183–119 (e), or phorbol myristate acetate (PMA)/ionomycin (f). After immunization of mice with the SmD183–119 peptide, interferon-␥ (IFN␥) expression was stimulated by in vitro culture with the SmD183–119 peptide. In contrast, immunization with the randomized peptide and in vitro restimulation with this peptide did not result in IFN␥ expression. IFN␥ expression was not increased in unmanipulated mice after 3 hours of restimulation with SmD183–119. PerCP ⫽ peridin chlorophyll protein; FITC ⫽ fluorescein isothiocyanate. (data not shown). Restimulation of T cells from immunized mice with the SmD183–119 peptide also resulted in an increased frequency of cells coexpressing TNF␣ and IL-2 (0.05%; data not shown). In general, the frequencies of cytokine-expressing cells were low. IL-4 expression was not detectable in immunized or untreated mice after 3 hours of restimulation. Characterization of SmD183–119-reactive T cell lines. T cells from unmanipulated 8-week-old NZB/ NZW mice were harvested and repeatedly restimulated Figure 5. In vitro T cell help for anti-dsDNA and anti-SmD183–119 antibodies by SmD183–119-specific T cell lines. T cell lines were obtained from 8–10-week-old unmanipulated female NZB/NZW mice and restimulated in vitro with the SmD183–119 peptide and irradiated antigen-presenting cells several times for up to 8 weeks. As a control, ovalbumin (OVA)–specific T cell lines were obtained from OVAimmunized NZB/NZW mice. SmD183–119-specific as well as OVAspecific T cell lines were stimulated with SmD183–119 peptide. In the presence of SmD183–119, SmD183–119-specific T cells gave help for anti-dsDNA as well as anti-SmD183–119 synthesis. In contrast, OVAspecific cell lines gave only slightly increased help for anti-dsDNA synthesis, but not for anti-SmD183–119 antibodies. Restimulation of OVA-specific T cell lines with OVA did not result in significant help for anti-dsDNA or anti-SmD183–119 antibodies. See Figure 1 for other definitions. 482 RIEMEKASTEN ET AL Figure 6. Cytokine expression in SmD183–119-specific T cell lines, as detected by flow cytometry. SmD183–119-specific T cells from lines that provide help for both anti–double-stranded DNA (anti-dsDNA) and anti-SmD183–119 antibodies were analyzed. As a control, ovalbumin (OVA)–specific T cell lines were measured for intracellular cytokine expression. After stimulation with anti-CD3, interferon-␥ (IFNg) was the cytokine dominantly expressed in SmD183–119-specific T cells. The majority of these T cells coexpressed IFNg and interleukin-4 (IL-4). In contrast, OVA-specific T cell lines revealed a completely different cytokine pattern, with only a few IFNg/IL-4 double-positive T cells. SmD183–119-specific T cells also expressed other cytokines, such as IL-2 and IL-10. FL ⫽ fluorescence; PE ⫽ phycoerythrin; APC ⫽ antigenpresenting cells. obtained after stimulation with SmD183–119, with a high frequency of IFN␥⫹,IL-4⫹ T cells (data not shown). In contrast, OVA-specific T cell lines exhibited a completely different cytokine pattern, with only a few IFN␥/ IL-4 double-positive T cells (Figure 6). Another SmD183–119-specific T cell line expressing only IFN␥ and no IL-4 did not provide T cell help for autoantibody generation in vitro (data not shown). DISCUSSION In previous work, we hypothesized that SmD183–119-reactive T cells play a crucial role in the generation of pathogenic anti-dsDNA antibodies via T cells (17). In the present study, we characterized SmD183–119-reactive T cells and demonstrated that they indeed promote the development of anti-dsDNA antibodies. Involvement of the protein SmD1 in the generation of anti-dsDNA antibodies presents a novel concept of interaction between the immune responses toward small nuclear RNP and dsDNA via T cell recognition and could explain the close relationship between antiSm and anti-dsDNA antibodies. To our knowledge, the present investigation is the first in which in vitro stimulation of freshly isolated T cells with a defined autoantigen has been used to characterize direct cytokine expression by autoantigen-specific T cells in unmanipulated mice with SLE. We previously found that SmD183–119 strongly influences the generation of anti-dsDNA antibodies in immunized NZB/NZW mice (17). We have now demonstrated that in cultures of B cells derived from diseased NZB/NZW mice, with SmD183–119 in the absence of T cells, plasma cells specific for either dsDNA or SmD183–119 were not generated. This rules out the polyclonal activation of dsDNA-specific B cells by SmD183–119. In contrast, SmD183–119-reactive T cells restimulated with SmD183–119 did increase the frequency of dsDNA-specific plasma cells ⬃3-fold, suggesting that SmD183–119-reactive T cells provide T cell help for anti-dsDNA–specific plasma cells. Our objective was to determine the mechanism by which SmD183–119-reactive T cells are activated. In vivo experiments and in vitro cultures demonstrated that the effect of SmD183–119 on the anti-dsDNA response is stronger than that on the anti-SmD183–119 response. Therefore, the dsDNA-specific B cells themselves could play a decisive role in the activation of SmD183–119-reactive T cells. Such B cells could initially be activated in a T cell–independent manner (e.g., by crosslinkage of surface antigen receptors) or independent of antigen receptors (e.g., by activation of Toll-like receptor 9, which has been suggested to play a role in lupus pathogenesis [24,25]). A precondition for these models is that the highly positively charged SmD183–119 or SmD1 molecules bind to DNA and do not critically impair the binding of DNA to anti-dsDNA antibodies on the B cell surface. Several groups (26,27) have hypothesized that such SmD1–dsDNA complexes are targets for both anti-dsDNA and anti-Sm responses. Our previous studies demonstrated that the addition of dsDNA to SmD1 increased the frequency of SmD1-reactive T cells in human SLE (17). In the experimental setting described here for the in vitro cultures, dsDNA for the ROLE OF SmD183–119-REACTIVE T CELLS IN ANTI-dsDNA GENERATION 483 Figure 7. Possible mechanism for the generation of anti-dsDNA and anti-SmD183–119 antibodies. After activation of dsDNA-specific B cells in a T cell–independent manner, these cells become potent antigen-presenting cells for dsDNA-associated proteins such as SmD1, accompanied by up-regulation of major histocompatibility complex and costimulatory molecules such as interferon-␥ (IFN␥) or IL-4. As a result, SmD183–119 could be presented to SmD183–119-reactive T cells, and T cells would be activated and in turn provide help for cells secreting high-affinity anti-dsDNA as well as anti-SmD183–119 antibodies. TLR9 ⫽ Toll-like receptor 9; TcR ⫽ T cell receptor; CD40L ⫽ CD40 ligand (see Figure 6 for other definitions). initial activation of B cells could have been supplied by dead cells, or preactivation of B cells could be a result of culture. Once activated, the dsDNA-specific B cell blasts could become potent antigen-presenting cells, thereby up-regulating MHC and costimulatory molecules. SmD1 as a DNA-associated protein could be presented either directly or indirectly to SmD183–119-reactive T cells. The activated T helper cells could in turn help the activated B cells to perform hypermutation, class switching, and differentiation into plasma cells, resulting in the generation of secreted, high-affinity anti-dsDNA antibodies, as observed in the present study. The activated SmD183–119reactive T cells could then further activate SmD183–119specific B cells. This would depend on T cell help since SmD183–119 is not likely to activate B cells directly via repetitive epitopes. In the case of the mice that received repeated immunizations, the SmD183–119-reactive T cells had been activated in vivo and could directly mediate the rapid production of anti-SmD183–119 by B cells. In this model, the same SmD183–119-reactive T cells provide the driving force for both anti-dsDNA and anti-SmD183–119 antibody production, as was previously shown for histone-reactive T cells that drive antihistone, antinucleosome, anti-dsDNA, and anti–single-stranded DNA responses (12). Figure 7 illustrates a proposed model for the activation of dsDNA- and SmD183–119-specific B cells by SmD183–119-reactive T cells. However, further studies are needed to determine whether naive T cells can also be activated by such B cell blasts, or only by T cells that have been preactivated by dendritic cells in vivo. Moreover, the mechanisms by which autoimmunization against the peptide SmD183–119 specifically occurs in human and murine lupus is still unclear. Studies to characterize SmD183–119-reactive T cells, especially after immunization, have shown a dominance of cells expressing the Th1 cytokine IFN␥, as can readily be demonstrated by ELISpot, flow cytometry, and cytokine ELISA of supernatants from short-term cultures (72 hours) and T cell lines. T cell–derived IFN␥ is believed to play a key role in the pathogenesis of lupus 484 since IFN␥ hyperproduction is a consistent finding in murine (28,29) and human (14) lupus. Treatment of NZB/NZW mice with IFN␥ has been shown to accelerate the manifestations of lupus, whereas treatment with anti-IFN␥ antibodies or soluble IFN␥ receptors early in life delayed disease progression (30,31). Both antidsDNA and anti-SmD183–119 antibodies contain the IgG2a subtype (results not shown), further suggesting an important role of IFN␥, which activates synthesis of this subtype (32). Data on the role of IL-4 in murine lupus are somewhat contradictory. Ectopic expression of IL-4 prevented the development of lethal lupus-like glomerulonephritis in a (NZW ⫻ C57BL/6.Yaa)F1 murine lupus model (33). Conflicting evidence provided by Nakajima and colleagues suggests that systemic blocking of IL-4 by monoclonal antibodies can prevent the onset of lupus nephritis in NZB/NZW mice (34). In the present study, the ex vivo addition of IL-4 and anti-IFN␥ to cultures of T and B cells derived from unmanipulated prenephritic NZB/NZW mice resulted in an increase in the SmD183–119-related anti-dsDNA antibody response. These findings demonstrate the relevance of IL-4 (formerly described as B cell–stimulating factor 1 ), at least in a period of disease development, for the generation of plasma blasts that secrete dsDNA-specific antibodies. Furthermore, T helper cell cultures in SmD183–119specific lines derived from young prenephritic NZB/ NZW mice were rich in cells exhibiting cytoplasmic expression of both IFN␥ and IL-4. Th cells that produce both Th1 and Th2 cytokines have been described and classified as Th0 (36,37). It is still unclear whether they represent a developmental stage occurring during Th1/ Th2 differentiation, a stable differentiated population, or a mixture of various subpopulations (38). However, the majority of histone-specific T cell clones derived from SLE patients display such an unrestricted cytokine secretion pattern, suggesting a possible role of this Th0 subset in the pathogenesis of lupus (14). Such T cells would also explain the activation of both humoral and inflammatory processes during SLE. In conclusion, our findings suggest a new mechanism for functional linkage of anti-Sm and anti-dsDNA responses. T helper cells from lupus-prone NZB/NZW mice can be activated by the peptide SmD183–119, an autoantigen unrelated to nucleosomes. It promotes antiDNA production and probably serves as a link between the anti-dsDNA and antispliceosomal response. The role of SmD183–119-reactive T cells in the generation of pathogenic anti-dsDNA antibodies suggests that these RIEMEKASTEN ET AL cells may be useful targets for autoantigen-specific treatment strategies. ACKNOWLEDGMENTS We are grateful to Claudia Klein for technical support, and to Rudi Manz and Thomas Kamradt for advice and support. REFERENCES 1. Tan EM. Antinuclear antibodies: diagnostic markers for autoimmune diseases and probes for cell biology. Adv Immunol 1989;44:93–151. 2. Hahn BH. Antibodies to DNA. N Engl J Med 1998;338:1359–65. 3. Mattaj IW, Tollervey D, Seraphin B. Small nuclear RNAs in messenger RNA and ribosomal RNA processing. FASEB J 1993; 7:47–53. 4. Lehmeier T, Foulaki K, Lührmann R. Evidence for three distinct D proteins, which direct differentially with anti-Sm autoantibodies, in the core of the major snRNPs U1, U2, U4/6, and U5. Nucleic Acids Res 1990;18:6475–84. 5. Hoch SO, Eisenberg RA, Sharp GC. Diverse antibody recognition patterns of the multiple SmD antigen polypeptides. Clin Immunol 1999;92:203–8. 6. Maddison PJ, Reichlin M. Quantitation of precipitating antibodies to certain soluble nuclear antigens in SLE: their contribution to hypergammaglobulinemia. Arthritis Rheum 1977;20:819–24. 7. Madaio MP, Datta SK. Journal club: major peptide autoepitopes for nucleosome-specific T cells of human SLE. Am J Kidney Dis 2000;35:992–6. 8. Pisetsky DS. Immune responses to DNA in normal and aberrant immunity. Immunol Res 2000;22:119–26. 9. Khalil M, Inaba K, Steinman R, Ravetsch J, Diamond B. T cell studies in a peptide-induced model of systemic lupus erythematosus. J Immunol 2001;166:1667–74. 10. Ebling FM, Tsao BP, Singh RR, Sercarz E, Hahn BH. A peptide derived from an autoantibody can stimulate T cells in the (NZB ⫻ NZW)F1 mouse model of systemic lupus erythematosus. Arthritis Rheum 1993;36:355–64. 11. Puttermann C, Diamond, B. Immunization with a peptide surrogate for double stranded DNA (dsDNA) induces autoantibody production and renal immunoglobulin deposition. J Exp Med 1998;188:29–38. 12. Datta SK, Kaliyaperumal A. Nucleosome-driven autoimmune response in lupus: pathogenic T helper cell epitopes and costimulatory signals. Ann N Y Acad Sci 1997;815:155–70. 13. Kuwana M, Feghali CA, Medsger TA Jr, Wright TM. Autoreactive T cells to topoisomerase I in monozygotic twins discordant for systemic sclerosis. Arthritis Rheum 2001;44:1654–9. 14. Voll RE, Roth EA, Girkontaite I, Fehr H, Herrmann M, Lorenz H-M, et al. Histone-specific Th0 and Th1 clones derived from systemic lupus erythematosus patients induce double-stranded DNA antibody production. Arthritis Rheum 1997;40:2162–71. 15. Riemekasten G, Marell J, Trebeljahr G, Klein R, Hausdorf G, Häupl T, et al. A novel epitope on the C-terminus of SmD1 is recognized by the majority of sera from patients with systemic lupus erythematosus. J Clin Invest 1998;102:754–63. 16. Riemekasten G, Weiss C, Schneider S, Thiel A, Bruns A, Schumann F, et al. T cell reactivity against the SmD183–119 C-terminal peptide in patients with systemic lupus erythematosus. Ann Rheum Dis 2002;61:779–85. 17. Riemekasten G, Kawald A, Wei␤ C, Meine A, Marell J, Klein R, et al. Strong acceleration of murine lupus by injection of the SmD183–119 peptide. Arthritis Rheum 2001;44:2435–45. ROLE OF SmD183–119-REACTIVE T CELLS IN ANTI-dsDNA GENERATION 18. Waisman A, Ruiz PJ, Israeli E, Eilat E, Konen-Waisman S, Zinger H, et al. Modulation of murine systemic lupus erythematosus with peptides based on complementarity determining regions of a pathogenic anti-DNA monoclonal antibody. Proc Natl Acad Sci U S A 1997;94:4620–5. 19. Hahn BH, Singh RR, Wong WK, Tsao BP, Bulpitt K, Ebling FM. Treatment with a consensus peptide based on amino acid sequences in autoantibodies prevents T cell activation by autoantigens and delays disease onset in murine lupus. Arthritis Rheum 2001;44:432–41. 20. Panosian-Sahakian N, Klotz J, Ebling F, Kronenberg M, Hahn BH. Diversity of Ig V gene segments found in anti-DNA autoantibodies from a single (NZB ⫻ NZW)F1 mouse. J Immunol 1989;142:4500–6. 21. Singh RR, Hahn BH, Tsao BP, Ebling FM. Evidence for multiple mechanisms of polyclonal T cell activation in murine lupus. J Clin Invest 1998;102:1841–9. 22. Assenmacher M, Schmitz J, Radbruch A. Flow cytometric determination of cytokines in activated murine T helper lymphocytes: expression of interleukin-10 in interferon- and in interleukin-4expressing cells. Eur J Immunol 1994;24:1097–101. 23. Crispin JC, Martinez A, de Pablo P, Velasquillo C, Alcocer-Varela J. Participation of the CD69 antigen in the T-cell activation process of patients with systemic lupus erythematosus. Scand J Immunol 1998;48:196–200. 24. Takeshita F, Leifer CA, Gursel I, Ishii KJ, Takeshita S, Gursel M, et al. Cutting edge: role of toll-like receptor 9 in CpG DNAinduced activation of human cells. J Immunol 2001;167:3555–8. 25. Krieg AM. CpG DNA: a pathogenic factor in systemic lupus erythematosus? J Clin Immunol 1995;15:284–92. 26. Reyes PA, Tan EM. DNA-binding properties of Sm nuclear antigen. J Exp Med 1977;145:749–54. 27. Dempsey LA, Hanakahi LA, Meizels N. A specific isoform of hnRNP D interacts with DNA in the LR1 heterodimer: canonical RNA binding motifs in a sequence-specific duplex DNA binding protein. J Biol Chem 1998;273:29224–9. 28. Murray LJ, Lee R, Martens C. In vivo cytokine gene expression in T cell subsets of autoimmune MRL/Mp-lpr/lpr mouse. Eur J Immunol 1990;20:163–170. 485 29. Prud’homme GJ, Kono DH, Theofilopoulos AN. Quantitative polymerase chain reaction analysis reveals marked overexpression of interleukin-1, interleukin-10, and interferon-gamma mRNA in the lymph nodes of lupus-prone mice. Mol Immunol 1995;32: 495–503. 30. Jacob CO, van der Meide PH, McDevitt HO. In vivo treatment of (NZB ⫻ NZW)F1 lupus-like nephritis with monoclonal antibody to ␥ interferon. J Exp Med 1987;166:798–803. 31. Ozmen L, Roman D, Fountoulakis M, Schmidt G, Ryffel B, Garotta G. Experimental therapy of systemic lupus erythematosus: treatment of NZB/W mice with mouse soluble interferon-gamma receptor inhibits the onset of glomerulonephritis. Eur J Immunol 1995;25:6–12. 32. Snapper CM, Paul WE. Interferon-gamma and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 1987; 236:944–7. 33. Santiago ML, Fossati L, Jacquet C, Muller W, Izui S, Reininger L. Interleukin-4 protects against a genetically linked lupus-like autoimmune syndrome. J Exp Med 1997;185:65–70. 34. Nakajima A, Hirose S, Yagita H, Okumura K. Roles of IL-4 and IL-12 in the development of lupus in NZB/W F1 mice. J Immunol 1997;158:1466–72. 35. Rabin EM, Ohara J, Paul WE. B-cell stimulatory factor 1 activates resting B cells. Proc Natl Acad Sci U S A 1985;82:2935–9. 36. Assenmacher M, Löhning M, Scheffold A, Richter A, Miltenyi S, Schmitz J, et al. Commitment of individual Th-1 like lymphocytes to expression of IFN␥ versus IL-4 and IL-10: selective induction of IL-10 by sequential stimulation of naı̈ve Th cells with IL-12 and IL-4. J Immunol 1998;161:2825–32. 37. Firestein GS, Roeder WD, Laxer JA, Townsend KS, Weaver ST, Hom JT, et al. A new murine CD4⫹ T cell subset with an unrestricted cytokine profile. J Immunol 1989;143:518–25. 38. Löhning M, Grogan JL, Coyle AJ, Yazdanbakhsh M, Meisel C, Gutierrez-Ramos JC, et al. T1/ST2 expression is enhanced on CD4⫹ T cells from schistosome egg-induced granulomas: analysis of Th cell cytokine coexpression ex vivo. J Immunol 1999;162: 3882–9.