T Cell Reactivity + Autoantibodies and CD4 Therapy Is Mediated By

Home , Aggrecan, Proteoglycan

Suppression of Proteoglycan-Induced Arthritis by Anti-CD20 B Cell Depletion Therapy Is Mediated by Reduction in Autoantibodies and CD4 + T Cell Reactivity This information is current as of September 27, 2021. Keith Hamel, Paul Doodes, Yanxia Cao, Yumei Wang, Jeffrey Martinson, Robert Dunn, Marilyn R. Kehry, Balint Farkas and Alison Finnegan J Immunol 2008; 180:4994-5003; ; doi: 10.4049/jimmunol.180.7.4994 Downloaded from http://www.jimmunol.org/content/180/7/4994

References This article cites 48 articles, 20 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/180/7/4994.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Suppression of Proteoglycan-Induced Arthritis by Anti-CD20 B Cell Depletion Therapy Is Mediated by Reduction in Autoantibodies and CD4؉ T Cell Reactivity1

Keith Hamel,* Paul Doodes,* Yanxia Cao,† Yumei Wang,† Jeffrey Martinson,* Robert Dunn,§ Marilyn R. Kehry,§ Balint Farkas,‡ and Alison Finnegan2*†

B cells have been implicated in the pathogenesis of rheumatoid arthritis (RA) since the discovery of RA as an autoimmune disease. There is renewed interest in B cells in RA based on the clinical efficacy of B cell depletion therapy in RA patients. Although, reduced titers of rheumatoid factor and anti-cyclic citrullinated peptide Abs are recorded, the mechanisms that convey clinical improvement are incompletely understood. In the proteoglycan-induced arthritis (PGIA) mouse model of RA, we reported that Ag-specific B cells have two important functions in the development of arthritis. PG-specific B cells are Downloaded from required as autoantibody-producing cells as well as Ag-specific APCs. Herein we report on the effects of anti-CD20 mAb B cell depletion therapy in PGIA. Mice were sensitized to PG and treated with anti-CD20 Ab at a time when PG-specific autoantibodies and T cell activation were evident but before acute arthritis. In mice treated with anti-CD20 mAb, development of arthritis was significantly reduced in comparison to control mAb-treated mice. B cell depletion reduced the PG-specific autoantibody response. Furthermore, there was a significant reduction in the PG-specific CD4؉ T cell recall ؉ response as well as significantly fewer PG-specific CD4 T cells producing IFN-␥ and IL-17, but not IL-4. The reduction in http://www.jimmunol.org/ PG-specific T cells was confirmed by the inability of CD4؉ T cells from B cell-depleted mice to adoptively transfer disease into SCID mice. Overall, B cell depletion during PGIA significantly reduced disease and inhibited both autoreactive B cell and T cell function. The Journal of Immunology, 2008, 180: 4994–5003.

heumatoid arthritis (RA)3 is a chronic and degenerative CD20 is only expressed after the pro-B cell stage and is not autoimmune disease that affects the synovial joints (1). expressed on plasma cells (9). Although autoantibodies are re- R Inflammation leads to cartilage degradation and bone duced in most RA patients treated with anti-CD20, autoanti- erosion, resulting in debilitating pain, joint deformity, and immo- body reduction does not always correlate with efficacy, suggest- by guest on September 27, 2021 bility (2, 3). Although the precise factors that initiate disease have ing that loss of other B cell functions may contribute to not been identified, several cell types have been implicated as me- suppression of RA (8, 10–12). diators of disease, including neutrophils, macrophages, B cells, Several autoimmune diseases, including systemic lupus ery- and T cells (4). The identification of rheumatoid factor (RF) in thematosus (SLE) and RA, are characterized by the presence of RA as an autoantibody provided the first key indicator that RA autoantibodies (13). In RA, the presence of RF and anti-cyclic is an autoimmune disease (5). However, because RF is not al- citrullinated peptide autoantibodies in the serum are an early ways associated with RA and because RF could be produced in predictor of disease in 50% of RA patients by an average of 7.5 diseases other than RA, interest in B cells waned (6). Recently, years (14). Autoantibodies bind to target epitopes, forming im- the importance of B cells in RA has been resurrected based on mune complexes that activate complement components and Fc the clinical efficacy of B cell depletion therapy (rituximab) in receptor-bearing cells that contribute to RA pathogenesis (15, RA patients (7, 8). Rituximab is a chimeric anti-human CD20 16). In several murine models of RA, autoantibodies are patho- mAb that selectively depletes immature and mature B cells, as genic, particularly the K/BxN and collagen-induced arthritis (CIA) models where disease is initiated by transfer of autoan- *Department of Immunology and Microbiology, †Department of Internal Medicine, tibodies to GPI and collagen respectively (17–20). Unlike these Section of Rheumatology, and ‡Department of Orthopedic Surgery, Rush University models, in proteoglycan (PG)-induced arthritis (PGIA), autoan- Medical Center, Chicago, IL 60612; and §Biogen Idec, San Diego, CA 92122 tibodies are required for initiation of disease but are insufficient Received for publication December 11, 2007. Accepted for publication January 21, 2008. to transfer arthritis (19, 21, 22). In addition to producing Abs, there are several functions of B The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance cells that could contribute to autoimmunity (11). B cells are with 18 U.S.C. Section 1734 solely to indicate this fact. highly competent APCs and present Ag 103-104-fold more ef- 1 This research was supported by National Institutes of Health Grant AR47657 ficiently than do macrophages or dendritic cells (23). Costimu- (to A.F.). latory molecules CD80 and CD86 are up-regulated on activated 2 Address correspondence and reprint requests to Dr. Alison Finnegan, Department of Medicine, Section of Rheumatology, Rush University Medical Center, 1735 West B cells, permitting T–B cell interaction. B cells also express Harrison Avenue, Chicago, IL 60612. E-mail address: [email protected] proinflammatory cytokines that may contribute to inflammation 3 Abbreviations used in this paper: RA, rheumatoid arthritis; CIA, collagen-induced (24–26). PGIA is dependent on Ag-specific B cells, and auto- arthritis; PG, proteoglycan; PGIA, proteoglycan-induced arthritis; RF, rheumatoid reactive T cells are not sufficiently activated to induce arthritis factor; SLE, systemic lupus erythematosus. unless PG is presented by B cells (19). Selective deletion of Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 costimulatory molecules CD80 and CD86 on PG-specific B www.jimmunol.org The Journal of Immunology 4995

cells is essential for autoreactive T cell activation and the de- anti-CD20 mAb were dissected, fixed in formalin, decalcified, and embed- velopment of arthritis (27). The requirement for primed auto- ded in paraffin. The tissue sections were stained with H&E. reactive B and T cells to sustain PGIA suggests a need for Abs, anti-CD20 mAb treatment, and B cell depletion reciprocal interactions that maintain chronic autoreactivity (28). Several studies in animal models of autoimmune disease have In experiments to assess the kinetics of B cell depletion, naive cells from spleen, popliteal lymph nodes, bone marrow, and peritoneal cavity (PE) shed some light on the role of B cell depletion in autoimmunity lavages were collected at days 1, 3, 7, 14, and 21 following i.v. injection (29). In CIA, depletion of B cells with an Ab specific for mouse of 250 ␮g of anti-mouse CD20 mAb (18B12, IgG2a) (31) or control Ab CD20 administered before initiation of CIA delays the onset of anti-human CD20 mAb (2B8), an engineered IgG2a derivative of the 2B8 disease; however, B cell depletion was not therapeutic after IgG1 mouse parent of rituximab that has no cross-reactivity with mouse CD20. 18B12 anti-mouse CD20 was generated by immunization of collagen immunization (30). Recently, anti-CD20 mAb con- Ϫ Ϫ CD20 / mice with CD20-transfected cell lines and CD20-peptide-key- ferred resistance in a murine model of lupus when treatment hole limpet hemocyanin conjugate (36). The 18B12 IgG2a variant was began at an early age and was continued for several weeks (31). generated by cloning and expression of 18B12 variable domain with the The ability of anti-CD20 therapy to ameliorate or delay disease IgG2a constant region in CHO cells. Mouse B cell subsets were charac- in these models may be due to the timing of treatment. Deple- terized by flow cytometry after staining with Abs: anti-IgD FITC, anti- CD21 PE, anti-B220 PerCp, anti-IgM APC, anti-IgM PerCp-Cy5.5, anti- tion of B cells in both models was before any clinical signs of CD5 PE, anti-CD38 FITC, anti-B220 PE, anti-B220 FITC, anti-CD43 PE, Ϫ/Ϫ disease and may mimic the disease course of B cell mice, (BD Biosciences), and anti-CD23 APC (Abcam). Profiling of B cell de- thus suggesting that B cells are important in initiation of auto- pletion in PG-immunized mice was performed at days 7 and 21 after anti- immunity (32, 33). We were therefore interested in determining CD20 mAb treatment, and cells were isolated from spleen, popliteal lymph nodes, bone marrow, and PE lavages. Mice were administered anti-CD20 whether B cells are important in maintaining chronic autoim- Downloaded from mAb and control mAb 4 days following the second PG-DDA immuniza- mune disease. tion. B cell depletion was monitored weekly by bleeding mice from the PGIA develops in BALB/c mice after two to three injections orbital plexus and staining cells with anti-CD19 PE, anti-CD3 APC, and of PG in adjuvant, and systemic autoantibody and autoreactive anti-CD11b FITC (BD Biosciences). Percentage of B cell depletion was ϭ T cells are activated before any signs of arthritis. In the present calculated as (average of specific subset in anti-CD20 treated mice (n 3)/average of control Ab-treated mice (n ϭ 3)) ϫ 100. The PBMC B study, we depleted B cells with an Ab to CD20 at a time point cell percentage was determined after gating out CD11bϩ from total

after PG-specific Ab and T cell reactivity were present but be- viable cells. B cell subsets were determined by flow cytometry using a http://www.jimmunol.org/ fore any clinical evidence of arthritis. In mice administered anti- FACSCalibur II and FACSDiva software (BD Biosciences). Data rep- mCD20, joint inflammation stabilized and subsequently waned resent the means Ϯ SEM. ϩ along with autoantibody titers. CD4 T cells from B cell-de- Detection of anti-PG Abs by ELISA pleted mice were significantly suppressed in their ability to re- spond to PG and were not efficient in inducing arthritis when Immunized mice were bled from the orbital plexus before anti-CD20 or ϩ control Ab treatment and every week following Ab injection. Serum was adoptively transferred. Cytokine production by CD4 T cells obtained and examined for Abs against mouse and human PG by ELISA. indicated that although the Th2 cells were not affected, Th1 and EIA tissue culture 96 “half-area” plates (Costar, Corning) were coated Th17 cytokine production was significantly inhibited. This overnight with 0.5 ␮g human PG, or 0.75 ␮g native mouse PG in carbonate study demonstrates that B cells are important as autoantibody- buffer. Serially diluted (PBS/0.5% Tween 20) serum samples and internal by guest on September 27, 2021 standards (pooled sera from arthritic mice) were incubated with immobi- secreting cells and as activators of autoreactive T cells. Impor- lized PGs. Abs were detected colorimetrically with HRP-labeled anti- tantly, these experiments indicate that B cells are essential after mouse IgG1 or IgG2a (Invitrogen Life Technologies) and the substrate initial sensitization for maintaining T cell reactivity and suggest o-phenylenediamine. The relative concentrations of human and mouse PG that B cells are critical APCc for chronic maintenance of were calculated based on titration of known concentration of unlabeled autoimmunity. murine IgG1 and IgG2a bound to the ELISA plate (SouthernBiotech). Murine Abs were detected with HRP-labeled anti-IgG1 or anti-IgG2a to establish a standard curve. Based on the standard curve, the relative Materials and Methods concentration of anti-human and anti-mouse PG was determined. Data Ϯ Mice, Ag, and immunizations represent the means SEM. Female BALB/c and SCID mice were purchased from the National Cancer T cell proliferation assessment Institute and maintained under germ-free conditions in our animal care Spleens were isolated from PG-immunized mice treated with either anti- Ͼ ϩ facility (Rush University Medical Center, Chicago, IL). Mice 3moof CD20 mAb or control mAb 21 days after treatment. CD4 T cells were ␮ age were immunized i.p. with 150 g of human PG (aggrecan) measured purified by negative selection using microbeads (Miltenyi Biotec). Al- as protein emulsified in 2 mg of the synthetic adjuvant dimethyldioctade- ternatively, as indicated in the figure legends, T cells obtained from the ␮ cylammonium bromide (DDA) (Sigma-Aldrich) and boosted with 100 g PE exudates of PG-immunized mice were purified by nylon wool pu- PG/DDA on day 21 as described previously (34). PG was obtained from rification (pooled from 4–6 mice). CD4ϩ T cells (2.5 ϫ 105 cells/well) human cartilage of joint replacement surgeries, as approved by the Insti- were stimulated with PG (10 ␮g/ml) and an equal number of irradiated tutional Review Board of Rush University Medical Center, and purified as spleen cells (2500 rad) from naive wild-type BALB/c mice. Cells were previously described by extraction with 4 M guanidinium chloride fol- cultured in quadruplicate in 96-well falcon plates (Fisher Scientific) in lowed by cesium chloride gradient centrifugation and treatment with chon- RPMI 1640 media containing 10% FCS, 100 ␮g/ml penicillin, 100 ␤ droitinase ABC and endo- -galactosidase (35). Native mouse PG was from ␮g/ml streptomycin, 50 ␮M 2-ME, and 2 mM L-glutamine (complete epiphyseal cartilage from skeletal tissues of 4–7-day-old mice and pre- media). Cultures were pulsed overnight on day 4 with [3H]thymidine pared as described for human PG but not deglycosylated. All animal ex- and examined for proliferation. Data represent the means Ϯ SEM. periments were approved by the Animal Care and Use Committee at Rush University Medical Center. Assessment of Th1, Th2, and Th17 cytokine profile Assessment of arthritis Spleens from PG-immunized mice were harvested 21 days after treatment with anti-CD20 mAb or control mAb. CD4ϩ T cells were purified by Mice were scored for arthritis development by a blinded observer. The negative selection. CD4ϩ T cells from each group were subsequently as- scoring of PGIA was on a scale from 0 to 4 for each paw, with a possible sessed for the number of IL-4, IL-17, and IFN-␥ producers in response to total score of 0–16. Scoring of each paw was as follows: 0, normal; 1, mild PG restimulation in culture by ELISPOT. Sterile multiscreen plates (Mil- erythema and swelling of several digits; 2, moderate erythema and swell- lipore) were coated with capture Ab overnight at 4°C, and blocked for 1 h ing; 3, more diffuse erythema and swelling; and 4, severe erythema and before addition of CD4ϩ T cells (3 ϫ 105 cells/well) and irradiated naive swelling of complete paw with ankylosis. Scoring of joints was conducted spleen cells (2500 rad) (2 ϫ 105 cells/well) and PG (10 ␮g/ml). Cells were three times a week. The hindlimbs of mice treated with control mAb or incubated for 48 h to screen for IFN-␥, and 72 h to screen for IL-17 and 4996 SUPPRESSION OF AUTOANTIBODIES AND CD4ϩ T CELL REACTIVITY

FIGURE 1. Anti-mCD20 mAb B cell depletion in naive BALB/c mice. Aged-matched BALB/c mice received 250 ␮g of either anti-mCD20 or control mAb i.v. Presence of B cell subsets were assessed on 1, 3, 7, 14, and 21 days after mAb injection. Subsets assessed in- cluded: (A) CD19ϩ mature B cells of the blood and the mature B cells (B220highIgMϩ), immature B cells (B220low IgMϩCD43Ϫ), pre-B cells (B220lowIgMϪCD43Ϫ), and pro-B cells (B220lowIgMϪCD43ϩ) of the bone marrow; (B) germinal center (CD38ϩB220ϩ) and follicular (CD21ϩ CD23ϩ) B cells of the lymph nodes; (C) B1a (CD11blow CD5ϩIgMhighB220ϩ/Ϫ), B1b (CD11blowCD5ϪIgMhigh B220ϩ/Ϫ), and B2 (CD11bϪCD5ϪIgMϩ/ϪB220high)B cells of the peritoneal cavity; (D) T1 (B220ϩCD21Ϫ IgMhigh), T2 (B220ϩCD21highIgMhighIgDhigh), mature (B220ϩCD21ϩIgMϩIgDhigh),andmarginalzone(B220ϩ CD21ϩIgMhighIgDlow) B cells of the spleen. Histograms Downloaded from indicate the percentage of B cells from anti-mCD20 mAb compared with anti-hCD20 mAb-treated mice and represent the average of three independent experiments Ϯ SEM. http://www.jimmunol.org/

IL-4. Cytokines were detected using HRP-labeled Abs specific for IFN-␥, anti-mCD20 mAb or control mAb. This dose was based on pre- IL-4, or IL-17, and read using ELISPOT reader and software (ImmunoSpot liminary experiments in other strains of mice (data not shown). Analyzer and ImmunoSpot software, CTL Analyzers). Data represent the means Ϯ SEM. Blood, spleen, popliteal lymph nodes, bone marrow, and PE cells were harvested from mice 1, 3, 7, 14, and 21 days after ϩ Adoptive transfer of arthritis with CD4 T cells mAb treatment. In the blood, mature B cell (CD19ϩ) numbers by guest on September 27, 2021 For adoptive transfer experiments, spleens from naive mice and PGIA- were considerably reduced (90.7%) after 1 day, and by day 3 immunized mice treated with anti-mCD20 mAb or control mAb mice were the cells were entirely depleted (99.5%) (Fig. 1A). Pro-B cells ϩ harvested, and CD4 T cells were purified by negative selection. Simul- (B220lowIgMϪCD43ϩ) of the bone marrow were the only sub- taneously, spleens from another set of arthritic, PG-immunized mice were depleted of T cells using Thy1.2 (CD90) microbeads and used as a source set not depleted due to their lack of CD20 expression (Fig. 1A). of B cells and APCs. CD4ϩ-purified T cells (1 ϫ 107) from either anti- The increase in pro-B cell numbers during depletion may be due mCD20 mAb, control Ab-treated PGIA mice, or naive mice were mixed to an attempt to repopulate B cell compartments to homeostatic ϫ 7 ␮ Ϫ Ϫ with T-depleted spleen cells of PGIA mice (1.5 10 ) and 100 gofPG levels. Pre-B cells (B220lowIgM CD43 ) in the bone marrow in saline, and then injected i.p. into female SCID mice. Development of were maximally depleted (31.7%) on day 7, followed by a re- arthritis was monitored in SCID mice every 3 days after cell transfer. Data ϩ represent the means Ϯ SEM. covery of this subset (Fig. 1A). Immature B cells (B220lowIgM CD43Ϫ) in the bone marrow reached a maximum depletion of Statistical analysis 85.8% on day 7, followed by a recovery to 19.0% on day 21 The Mann-Whitney U test was used to compare nonparametric data for (Fig. 1A). Mature B cells (B220highIgMϩ) demonstrated a sim- statistical significance. Pearson’s correlation coefficient was used to calcu- ilar pattern of depletion and were rapidly depleted in the bone late the correlation between reduction in arthritis and PG-specific Ab titers. Significance is based on a value of p Ͻ 0.05. marrow at day 1 and reached a maximum depletion of 87.0% on day 3 (Fig. 1A). Results Possibly the most important sites of depletion are the two main B cell depletion of naive and PG-immunized BALB/c mice with sites of B cell activation and expansion, the spleen and lymph anti-mouse CD20 mAb nodes. In the spleen, T1 (B220ϩCD21ϪIgMϩ) cells were the most We have previously shown that PG-specific B cells are required difficult to deplete, having only reached 76.3% depletion on day for the development of autoreactive T cells in the absence of cir- 14, which may be due to the time needed to fully deplete existing culating Ab (22). These results indicate that B cells directly func- T1 cells and the influx of newly created B cells from the bone ϩ ϩ high high tion as APCs for T cell activation. What is not clear from these marrow (Fig. 1D). The T2 (B220 CD21 IgM IgD ) B cells ϩ studies is whether B cells are essential for maintaining T cell ac- in the spleen were depleted by day 7 (Fig. 1D). Total mature (B220 ϩ ϩ ϩ ϩ ϩ high tivation after initiation of disease (22). To address this question we CD21 IgM IgD ) and marginal zone (B220 CD21 IgM assessed the requirement for B cells after an immune response to IgDlow) B cells of the spleen were depleted 98.7 and 99.0%, re- PG was initiated by depleting B cells. spectively, on day 14 (Fig. 1D). The lymph nodes were the most ϩ We initially examined B cell subset depletion in naive BALB/c susceptibletoanti-CD20treatment,withthegerminalcenter(CD38 mice using an IgG2a anti-mouse CD20 mAb (anti-mCD20). Mice B220ϩ) and follicular (CD21ϩCD23ϩ) B cells being depleted by were treated with a single i.v. injection of 250 ␮g/mouse of either 93.5 and 92.6%, respectively, on day 7 (Fig. 1B). The Journal of Immunology 4997

FIGURE 2. Anti-mCD20 mAb depletion in PG-immunized BALB/c mice. Aged-matched BALB/c mice were boosted with PG in adjuvant at 3 wk and 4 days after they were injected i.v. with 250 ␮g of either anti- mCD20 mAb or control mAb. Blood and bone marrow (A), lymph nodes (B), peritoneal lavages (C), and spleens (D) were harvested at days 7 and 21 after Ab injection and were assessed for depletion of B cell subsets. Histograms indicate the percentage of B cells of anti-mCD20 mAb compared with control mAb treated mice and represent the average of three independent ex- Downloaded from periments Ϯ SEM. http://www.jimmunol.org/

The most resistant B cell subsets to depletion were of the peri- on 7 and 21 days, although not significant, did show more re- toneal cavity with B1a (CD11blowCD5ϩIgMhighB220ϩ/Ϫ), B1b sistance to depletion than did those of naive mice (Fig. 2, B and (CD11blowCD5ϪIgMhighB220ϩ/Ϫ),andB2(CD11bϪCD5ϪIgMhigh D). The amount of depletion of all subsets continued to increase by guest on September 27, 2021 B220ϩ/Ϫ) B cells being only 25.4, 28.7, and 67.1% depleted, re- from 7 to 21 days in both naive and immunized mice. The spectively (Fig. 1C). Similar results have been seen by others and resistance observed of immunized mouse B cells to depletion by were due to inaccessibility of Fc receptor-bearing cells to the peri- anti-mCD20 mAb treatment may be due to a higher rate of B toneal cavity (37) and not due to insufficient coating of cells with cell output from the bone marrow, which accounted for a faster anti-CD20 mAb (38). repopulation of these B cell compartments. Also, an inherent Based on the B cell-depletion profile in nonimmune mice, we property of activated B cells may confer a higher resistance to assessed the profile of B cell depletion in BALB/c mice immu- depletion, as described in a spontaneous model of SLE (31). nized with PG. Naive mice or mice immunized twice with PG In a surprise reaffirmation of the mechanism of resistance by in adjuvant were treated with either anti-mCD20 mAb or con- peritoneal cavity B cells to anti-CD20 mAb depletion in naive trol mAb 4 days after the second immunization. Tissues and mice, immunized mice were completely depleted for all peri- cells were harvested on days 7 and 21 after mAb treatment and toneal cavity B cell subsets by day 7. PG-DDA i.p. immunized stained to assess depletion of the different B cell populations. mice showed a 99.5, 99.0, and 98.7% depletion of B1a, B1b, Mature B cells of the bone marrow were similarly depleted on and B2 subsets, respectively (Fig. 2C). The increased suscep- day 7 in both naive and immunized mice, but by day 21 mature tibility of immunized mouse peritoneal cavity B cell subsets to B cells of immunized mice recovered significantly faster ( p ϭ depletion may be due to an influx of Fc receptor-bearing cells 0.046) than did those of naive mice (Fig. 2A). Similarly, the such as macrophages in response to immunization. DDA has levels of CD19ϩ B cells in the blood of immune mice also been shown to increase the attraction of macrophages to sites of reflected an accelerated repopulation in comparison to naive injection, which may increase Fc receptor-mediated depletion mice ( p ϭ 0.037), although both were still almost completely (39). depleted at day 21 (Fig. 2A). Interestingly, B cell subsets at sites of activation in secondary Cellular autoreativity in PGIA before B cell depletion lymphoid tissues were more resistant to anti-mCD20 depletion To study the mechanism by which B cell depletion affects an in immune mice than in naive mice. The splenic subsets includ- ongoing autoimmune response, a treatment period was chosen ing follicular, total mature, and particularly marginal zone B based on priming B and T cell responses to PG, but before cells proved the most resistant to depletion in immunized mice development of maximum arthritis. Preliminary experiments at both 7 and 21 days ( p ϭ 0.05) (Fig. 2D). Follicular B cells showed that 4 days following the second immunization of PG- of popliteal lymph nodes in immunized mice were also more DDA, mice developed significantly higher levels of anti-mouse challenging to deplete than in naive mice on day 7 ( p ϭ 0.046), PG IgG1 ( p ϭ 0.023) and IgG2a ( p ϭ 0.023), as well as anti- and on day 21 they were also less depleted than in naive mice human PG IgG1 ( p ϭ 0.025) and IgG2a ( p ϭ 0.023) Ab titers (Fig. 2B). The remaining splenic and lymph node B cell subsets than did naive BALB/c mice (Fig. 3, B and C). The presence of 4998 SUPPRESSION OF AUTOANTIBODIES AND CD4ϩ T CELL REACTIVITY

FIGURE 3. Presence of PG-specific T and B cell reactivity at the time of B cell depletion. PE cells and serum were collected from BALB/c mice 1 wk following the second PG-DDA immunization. A, T cells from peritoneal lavages of PG-immunized mice (n ϭ 3) were purified with nylon wool separation and subsequently restimulated with or without 10 ␮g/ml PG in vitro culture for 5 days. Proliferation was determined by 24 h [3H]thymidine incorporation. B, Serum PG-specific mouse IgG1 (mPG IgG1) and IgG2a (mPG IgG2a) and (C) human PG-specific IgG1 (hPG IgG1) and IgG2a (hPG IgG2a) titers of Downloaded from naive (n ϭ 3) and immunized (n ϭ 7–10) mice were measured by ELISA. Values are means Ϯ SEM and are representative of three independent .p Ͻ 0.05 ,ء .experiments anti-mouse and anti-human PG Abs indicated that autoreactive mice treated with control mAb experienced a progressive in- B cells had been activated. Similarly, T cells obtained at the crease in PGIA and more aggressive disease, whereas in mice http://www.jimmunol.org/ same time from the peritoneal cavity demonstrated a significant treated with anti-mCD20 mAb arthritis did not progress or PG-specific recall response ( p ϭ 0.021) when compared with waned in severity (Fig. 4A). The incidence of disease was 100% naive mice (Fig. 3A). Although both autoreactive B and T cells in both B cell-depleted and control mAb-treated groups by day were activated at this point, severe inflammation of the periph- 19 after mAb treatment (Fig. 4B). The ability of B cell depletion eral joints had not been initiated (data not shown). to suppress inflammation remained relatively constant, even as circulating B cells (CD19ϩ) recovered in the blood. Nearly Anti-mCD20 mAb treatment during development of PGIA complete depletion was maintained until week 4 after mAb suppresses arthritis severity treatment, when B cells steadily increased in the circulation,

The ability of PGIA to develop in mice treated with anti- ultimately achieving a level of 73.8% of control mAb-treated by guest on September 27, 2021 mCD20 mAb was assessed by following the procedures out- mice by 11 wk, when the mice were sacrificed (Fig. 4C). lined above. Four days following the second PG immunization, To determine the extent of inflammation, we assessed ankle female BALB/c mice were injected i.v. with either 250 ␮g/ joint histology in mAb-treated mice (Fig. 5). In control mAb- mouse anti-mCD20 mAb or control mAb and observed for pro- treated mice, mononuclear and polymorphonuclear infiltration gression of PGIA by scoring fore- and hindlimbs for inflam- were profuse in tissues and joint spaces. Edema in the synovial mation. The development of PGIA in mice depleted of B cells and periarticular tissues was accompanied by synovial hyper- or treated with control Ab was indistinguishable until day 19 plasia. Cartilage damage and bone erosion were also evident after mAb treatment, when both groups had developed an av- (Fig. 5, B and D). In contrast, cellar infiltration and histopatho- erage PGIA score of ϳ2.5 (Fig. 4A). Subsequently, after day 19, logical signs of arthritis were decreased in B cell depleted mice

FIGURE 4. Suppression of PGIA by anti-mCD20 treatment. BALB/c mice were immunized i.p. with human PG in DDA twice at 3-wk intervals. Four days following the second PG-DDA immunization mice were either injected i.v. with anti-mCD20 mAb (n ϭ 7–10) or control mAb (n ϭ 7–10) and were monitored for arthritis. Arthritis severity (A) is the sum of the inflamed paws in individual mice divided by the number of mice with inflamed paws. Incidence (B) denotes the percentage of mice that develop PGIA. B cell recovery (C) was assessed weekly after Ab injection by examining the percentage of CD19ϩ B cells in the blood in anti-mCD20 mAb (n ϭ 5–10) and control mAb (n ϭ 5–10) mice by flow cytometry. Histograms indicate the percentage of B cells in anti-mCD20 mAb-treated mice compared with control mAb-treated mice. Values are means Ϯ SEM and are representative of three independent .p Ͻ 0.05 ,ء .experiments The Journal of Immunology 4999

correlate with a reduction in disease. In combining two separate experiments a positive correlation (r ϭ 0.82) ( pϽϭ0.01) was found between weeks 2 and 7 after anti-mCD20 mAb treatment. CD4ϩ T cell autoreactivity is inhibited in B cell-depleted mice Ag presentation by B cells is critical for the activation of an autoreactive CD4ϩ T cell population and the subsequent development of PGIA (19, 22). The need to focus Ag to B cells for presentation to T cells indicates that B cell activation of T cells is critical for the development of disease (19, 22). The loss of B cells in this autoreactive cycle could result in a decrease in autoreactive T cell activation, thereby quelling the inflammatory response. To examine whether there was a reduction in PG-specific T cell activation, B cell depletion therapy was be- gun 4 days after the second immunization. After 21 days, splenic CD4ϩ T cells were purified, restimulated with PG in the presence of irradiated naive spleen cells and PG (10 ␮g/ml), ϩ and assessed for PG-specific T cell proliferation. CD4 T cells Downloaded from from anti-mCD20 mAb-treated mice proliferated less efficiently than did T cells from control mAb-treated mice, indicating that either fewer autoreactive CD4ϩ T cells were primed or they were less responsive to Ag (Fig. 7A). To characterize the effect of B cell depletion on Th cell cy- ϩ

tokine profiles, purified CD4 T cells from mAb-treated mice http://www.jimmunol.org/ were assessed for the number of IL-4 (Th2), IFN-␥ (Th1), and IL-17 (Th17) by ELISPOT. Although all cultures responded to PG restimulation, only IFN-␥-(p ϭ 0.042) and IL-17- ( p ϭ 0.05) producing cells were diminished in B cell-depleted mice (Fig. 7, B–D), indicating that the Th1 and Th17, but not the Th2, compartments in arthritic mice were inhibited by B cell FIGURE 5. Suppression of histopathology in the joints of anti-mCD20 depletion. mAb-treated mice. Hindlimbs of mice were dissected, fixed in formalin, decalcified, and embedded in paraffin. Tissue sections were stained with ϩ B cell depletion prevents CD4 T cell adoptive transfer by guest on September 27, 2021 H&E. Sections are from representative ankle joints from control mAb- of PGIA treated (B and D) and anti-mCD20 mAb-treated mice (A and C). Magni- ϩ fications are ϫ4 objective (A and B) and ϫ10 objective (C and D). Cellular The reduction in PG-specific CD4 T cell proliferation and infiltration (E) was measured on a scale of 0–4 by a blinded observer, and cytokine production in vitro suggests that priming of autoreac- values represent means Ϯ SEM of 10–15 sections from three independent tive helper T cells in B cell-depleted mice was suboptimal. To p Ͻ 0.05. confirm that the reduction in priming of CD4ϩ T cells correlates ,ء .experiments with a reduction in arthritogenic potential, the ability of CD4ϩ T cells to adoptively transfer disease into SCID mice was (Fig. 5, A, C, and E). These data demonstrate that B cell de- assessed. Spleen cells from B cell-depleted or control mAb- pletion after the development of a PG-specific immune response treated mice were obtained 20 days after mAb treatment. Puri- suppresses arthritis, indicating that B cells are necessary to fied CD4ϩ T cells combined with B cells and APCs (T cell- maintain the development of arthritis. depleted spleen cells) from arthritic mice were transferred into SCID recipients. Development of arthritis was monitored over B cell depletion suppresses anti-mouse and anti-human time. Purified CD4ϩ splenic T cells from arthritic mice treated PG IgG1 Abs with PBS or those from naive nonimmunized mice acted as In PGIA, autoantibodies are critical for the initiation of disease, positive and negative control groups, respectively. CD4ϩ T solidifying their involvement in PGIA pathogenesis (22). To cells obtained from PG-immunized PBS-treated mice induced determine whether B cell depletion therapy reduced autoanti- an aggressive inflammation beginning 7 days after transfer, body titers, mice were bled 1 day before mAb treatment and leading to severe inflammation and ankylosis in 80% of recip- every week thereafter for 7 wk. Serum samples were assayed ients, whereas CD4ϩ T cells from naive mice were unable to for IgG1 and IgG2a anti-mouse and anti-human PG titers. Be- transfer arthritis (Fig. 8, A and B). PG-immunized mice treated fore B cell depletion, mice developed similar levels of IgG1 and with control mAb generated CD4ϩ T cells equivalent to PBS IgG2a autoantibodies. After treatment with anti-CD20 or con- treated mice in their ability to transfer disease with similar se- trol mAb, anti-mouse and anti-human PG IgG2a were not sig- verity and incidence. CD4ϩ T cells originating from B cell- nificantly different (Fig. 6, B and D). In contrast to the IgG2a depleted PG-immunized mice were significantly reduced in Ab titers, both anti-mouse and anti-human PG IgG1 titers were their ability to transfer PGIA, and those mice that did develop significantly reduced in the B cell-depleted mice in comparison PGIA (40%) experienced a very mild disease in comparison to to control mAb group (Fig. 6, A and C). Autoantibody titers SCID mice that received CD4ϩ T cells from control mAb- or waned with time and those of B cell-depleted mice reached a PBS-treated mice (Fig. 8, A and B). lower maximum level than did those of control Ab-treated mice The inability to transfer PGIA with CD4ϩ T cells from B cell- (Fig. 6). Autoantibody titers of PGIA mice depleted of B cells depleted mice was not due to a defect or a suppression of primed 5000 SUPPRESSION OF AUTOANTIBODIES AND CD4ϩ T CELL REACTIVITY

FIGURE 6. Suppression of PG-specific Ab titers in anti-mCD20 mAb-treated mice. Serum from PG-DDA- immunized mice treated with anti-mCD20 mAb (n ϭ 5–10) or control mAb (n ϭ 5–10) were collected before mAb injection and then weekly for 7 wk. PG-specific human IgG1 (A) and IgG2a (B) and mouse PG-specific IgG1 (C) and IgG2a (D) Abs were measured by ELISA. Ϯ Values are means SEM and are representative of Downloaded from .p Ͻ 0.05 ,ء .three independent experiments http://www.jimmunol.org/

B cell activity, as autoantibody titer levels of SCID recipient mice ferred from B cell-depleted PGIA mice as compared with anti-PG by guest on September 27, 2021 were comparable between groups (Fig. 8, C and D). Surprisingly, levels from control mAb-treated PGIA mice (Fig. 8C). These re- an increase in anti-human PG IgG1 ( p ϭ 0.016) and IgG2a ( p ϭ sults demonstrate that B cell depletion therapy during development 0.032) was observed in SCID recipients of CD4ϩ T cells trans- of arthritis suppresses the autoreactive T cell response.

FIGURE 7. Reduction in PG-specific T cell autoreactivity in B cell- depleted mice. Splenic CD4ϩ T cells of either anti-mCD20 mAb- or control mAb-treated mice were purified 21 days after Ab treatment and mixed with naive irradiated APCs with or without PG for 5 days and measured for proliferation (A)by[3H]thymidine incorporation for the final 24 h. Sim- ilar cultures were established for the number of cytokine secreting CD4ϩ T cells, with detection by ELISPOT for IL-17 (B) and IL-4 (C)at72hand IFN-␥ at 48 h (D). Values indicate the means Ϯ SEM of quadruplicate cul- .p Ͻ 0.05 ,ء .tures The Journal of Immunology 5001

FIGURE 8. CD4ϩ T cells from B cell-depleted mice are unable to transfer arthritis. Naive SCID mice received i.p. a mixture of PG, T cell- depleted spleen cells from arthritic mice and puriﬁed splenic CD4ϩ T cells from either naive (n ϭ 5), anti- mCD20 mAb- (n ϭ 5), control mAb- (n ϭ 5), or PBS- (n ϭ 5) treated mice 20 days after treatment. Following transfer, mice were monitored for PGIA severity of arthritis severity (A) and incidence (B). Serum samples collected from each group (n ϭ 5)

were analyzed for reconstituted Downloaded from SCIDs. Sera were analyzed by ELISA for IgG1 and IgG2a Ab-spe- ciﬁc human (C) or mouse (D) PG. Values are means Ϯ SEM and are representative of two independent exper- .p Ͻ 0.05 ,ء .iments http://www.jimmunol.org/

Discussion in a similar manner. Following day 19, progression of arthritis in Animal models of disease play an important role in understanding the anti-CD20 mAb-treated mice was prevented. Although B cell by guest on September 27, 2021 mechanisms of disease. In several autoimmune diseases B cell repopulation occurred over time, arthritis did not relapse in the depletion therapy has proven efficacious. The present study was time interval the mice were monitored. In CIA, B cell depletion designed to define the role of B cell depletion in a model that is before sensitization to collagen resulted in arthritis relapse very similar to B cell depletion therapy in RA. Our initial studies as- late (30). Thus, it is possible that if we had waited longer similar sessed the ability of an anti-mCD20 mAb of the IgG2a isotype to results would have been observed. deplete B cells. As expected from reports by others on anti-mCD20 Similar to B cell depletion in RA, PG-specific Ab levels de- depletion, by 14 days after treatment peripheral blood, germinal creased, and this decrease correlated with a reduction in disease. center, follicular, T2, mature, and marginal zone B cells were sub- However, the reduction in Abs may be only one mechanism by stantially depleted, whereas in the peritoneal cavity B1a, B1b, and which B cell depletion contributes to the reduction in arthritis. B2 B cells were relatively resistant. However, after immunization Clinical success of CTLA-4Ig, a CD80/86 blocking agent, has with PG, marginal zone and mature bone marrow B cells were highlighted the complicity of costimulation in the perpetuation of more resistant, and peritoneal B1a and B1b B cells were more autoimmunity (40). We have previously shown that B cells are sensitive to depletion. The sensitivity to depletion of peritoneal B critical as APCs in the development of PGIA (19). B cells must cells is likely a result of activated macrophages in the peritoneal recognize and present PG to T cells for arthritis to develop. Fur- cavity due to the i.p. immunization with PG in adjuvant. The sub- thermore, B cell expression of CD80/86 is essential for induction sequent inflammatory response in the peritoneal cavity exposes the of disease but dispensable for Ab production (27). In these previ- peritoneal B cells to activated macrophages mediating Fc␥R-de- ous experiments, the defects in B cell APC activity preceded sen- pendent B cell depletion (37). The increased resistance of marginal sitization with PG, and therefore we were unable to distinguish zone B cells may be due to their activation state, physical niche, or an increase in survival factors. Recent data from Gong et al. indi- between the effects of B cells on T cell priming or on perpetuation cate that B cell activating factor plays an important role in deter- of T cell activation. In the present study, PG-specific T cells were mining the susceptibility of mouse marginal zone B cells to de- present at the time of B cell depletion, as reflected in the secondary recall response of CD4ϩ T cells to PG (Fig. 3A). The reduction in pletion using an anti-hCD20 mAb in human CD20 transgenic mice ϩ ϩ (29). CD4 T cell proliferation (Fig. 7A) and the inability of CD4 T To determine whether B cells are important to maintain a state cells from B cell-depleted mice to transfer disease into SCID re- of chronic activation in PGIA, B cells were depleted after the cipients (Fig. 8) indicate that there was a reduction in the PG- initiation of disease at a time point when PG-specific autoantibod- specific effector or memory T cell pools generated in the absence ies and T cell reactivity were established but inflammation in the of B cells. This work extends the observations of B cell depletion joints was not yet apparent. After treatment, arthritis developed in in a model of SLE (31), and it supports the idea that B cells are both anti-CD20 mAb- and control mAb-treated mice for ϳ19 days involved in chronic T cell activation. 5002 SUPPRESSION OF AUTOANTIBODIES AND CD4ϩ T CELL REACTIVITY

B cell Ag presentation, more than other APCs, may be critically References ϩ important for CD4 T cell activation in chronic disease. This 1. Goronzy, J. J., and C. M. Weyand. 2005. Rheumatoid arthritis. Immunol. Rev. could be due to a continuous expansion of a high-affinity popula- 204: 55–73. 2. Lee, D. M., and M. E. Weinblatt. 2001. Rheumatoid arthritis. Lancet 358: tion of autoantigen-specific B cells in secondary lymphoid tissues. 903–911. These high-affinity B cells may be important in driving the in- 3. Dunlop, D. D., L. M. Manheim, E. H. Yelin, J. Song, and R. W. Chang. 2003. The creased expansion of pathogenic autoreactive T cells. The residual costs of arthritis. Arthritis Rheum. 49: 101–113. 4. Firestein, G. S. 2005. Immunologic mechanisms in the pathogenesis of rheuma- inflammation in B cell-depleted mice (Fig. 4A) indicates that T cell toid arthritis. J. Clin. Rheumatol. 11: S39–S44. activation by other APCs also occurs but may not be sufficient for 5. Franklin, E. C., H. R. Holman, H. J. Muller-Eberhard, and H. G. Kunkel. 1957. severe arthritis. Some RA patients that develop more aggressive An unusual protein component of high molecular weight in the serum of certain patients with rheumatoid arthritis. J. Exp. Med. 105: 425–438. symptoms of disease have tertiary lymphoid structures similar to 6. Janossy, G., G. Panayi, O. Duke, M. Bofill, L. W. Poulter, and G. Goldstein. germinal centers within the synovium that are comprised of B 1981. Rheumatoid arthritis: a disease of T-lymphocyte/macrophage immunoregu- lation. Lancet 2: 839–842. cells, T cells, and dendritic cells (41–43). Depletion of B cells 7. Edwards, J. C., L. Szczepanski, J. Szechinski, A. Filipowicz-Sosnowska, within the synovium could disrupt the formation of these structures P. Emery, D. R. Close, R. M. Stevens, and T. Shaw. 2004. Efficacy of B-cell- and subsequent T and B cell interaction (43). targeted therapy with rituximab in patients with rheumatoid arthritis. N. Engl. J. Med. 350: 2572–2581. The ability of B cells to dictate an immune response is described 8. Martin, F., and A. C. Chan. 2004. Pathogenic roles of B cells in human autoim- by their cytokine profiles. Be1 B cells have been shown to produce munity: insights from the clinic. Immunity 20: 517–527. IFN-␥ and IL-12 under certain circumstances that can orchestrate 9. Grillo-Lopez, A. J., E. Hedrick, M. Rashford, and M. Benyunes. 2002. Ritux- ϩ imab: ongoing and future clinical development. Semin. Oncol. 29: 105–112. a Th1 CD4 T cell response (44). Herein we demonstrate that the 10. Edwards, J. C., and G. Cambridge. 2006. B-cell targeting in rheumatoid arthritis depletion of B cells during PGIA results in a reduction in both and other autoimmune diseases. Nat. Rev. Immunol. 6: 394–403. Downloaded from ␥ ϩ 11. St. Clair, E. W., and T. F. Tedder. 2006. New prospects for autoimmune disease IFN- - and IL-17-producing CD4 T cells (Fig. 7, B and D). therapy: B cells on deathwatch. Arthritis Rheum. 54: 1–9. These data suggest that the loss of cytokine production by B cells 12. Martin, F., and A. C. Chan. 2006. B cell immunobiology in disease: evolving might be a factor responsible for clinical efficacy. concepts from the clinic. Annu. Rev. Immunol. 24: 467–496. 13. Lim, P. L., and M. Zouali. 2006. Pathogenic autoantibodies: emerging insights Although the decrease in autoantibody correlates with a reduc- into tissue injury. Immunol. Lett. 103: 17–26. tion in disease in RA and PGIA (Fig. 6), it is yet to be determined 14. Nielen, M. M., D. van Schaardenburg, H. W. Reesink, R. J. van de Stadt, I. E. van der Horst-Bruinsma, M. H. de Koning, M. R. Habibuw, whether autoantibody is a crucial component in the perpetuation of http://www.jimmunol.org/ J. P. Vandenbroucke, and B. A. Dijkmans. 2004. Specific autoantibodies precede RA, or merely a consequence of continuous B and T cell activa- the symptoms of rheumatoid arthritis: a study of serial measurements in blood tion. However, in PGIA, PG-specific Abs are essential for devel- donors. Arthritis Rheum. 50: 380–386. 15. Tsokos, G. C., and S. D. Fleming. 2004. Autoimmunity, complement activation, opment of arthritis, and thus the substantial reduction in Ab titers tissue injury and reciprocal effects. Curr. Dir. Autoimmun. 7: 149–164. in B cell-depleted mice likely contributes to the reduction in ar- 16. Kaplan, C. D., Y. Cao, J. S. Verbeek, M. Tunyogi-Csapo, and A. Finnegan. 2005. thritis. In our adoptive transfer studies, T cells from B cell-de- Development of proteoglycan-induced arthritis is critically dependent on Fc␥ receptor type III expression. Arthritis Rheum. 52: 1612–1619. pleted mice were unable to transfer disease, but autoantibodies 17. Stuart, J. M., and F. J. Dixon. 1983. Serum transfer of collagen-induced arthritis were equivalent to those in control mAb-treated mice (Fig. 8). This in mice. J. Exp. Med. 158: 378–392. prevention of arthritis without a disruption of autoantibody sug- 18. Maccioni, M., G. Zeder-Lutz, H. Huang, C. Ebel, P. Gerber, J. Hergueux, P. Marchal, V. Duchatelle, C. Degott, M. van Regenmortel, et al. 2002. Arthri- by guest on September 27, 2021 gests that the reduction of autoantibody during arthritis observed togenic monoclonal antibodies from K/BxN mice. J. Exp. Med. 195: 1071–1077. with B cell depletion in RA patients may be secondary to the effect 19. O’Neill, S. K., M. J. Shlomchik, T. T. Glant, Y. Cao, P. D. Doodes, and A. Finnegan. 2005. Antigen-specific B cells are required as APCs and autoanti- of lowered autoreactive T cell activation. The ability of antibody-producing cells for induction of severe autoimmune arthritis. J. Immunol. mCD20 mAb treatment to suppress T cell activation supports a 174: 3781–3788. role for B cells in maintaining chronic activation of T cells beyond 20. Petkova, S. B., K. N. Konstantinov, T. J. Sproule, B. L. Lyons, M. A. Awwami, and D. C. Roopenian. 2006. Human antibodies induce arthritis in mice deficient the point of autoimmune initiation (28). in the low-affinity inhibitory IgG receptor Fc␥RIIB. J. Exp. Med. 203: 275–280. ϩ The inability of the autoreactive CD4 T cell response to opti- 21. Wooley, P. H., S. W. Siegner, J. D. Whalen, R. L. Karvonen, and mally develop in B cell-depleted mice may be due to increased F. Fernandez-Madrid. 1992. Dependence of proteoglycan induced arthritis in BALB/c mice on the development of autoantibodies to high density proteogly- regulation. It has been reported that loss of costimulatory signals, cans. Ann. Rheum. Dis. 51: 983–991. which are common to B cells and other APCs, results in the con- 22. Mikecz, K., T. T. Glant, E. Buzas, and A. R. Poole. 1990. Proteoglycan-induced ϩ polyarthritis and spondylitis adoptively transferred to naive (nonimmunized) version of CD4 effector T cells toward a more regulatory popu- BALB/c mice. Arthritis Rheum. 33: 866–876. lation due to induction of anergy or up-regulation of FoxP3 (45, 23. Brennan, F. R., G. Negroiu, E. I. Buzas, C. Fulop, K. Hollo, K. Mikecz, and 46). Also, B cell expansion blocks T cell regulatory activity, sug- T. T. Glant. 1995. Presentation of cartilage proteoglycan to a T cell hybridoma derived from a mouse with proteoglycan-induced arthritis. Clin. Exp. Immunol. gesting that depletion of B cells could expand a regulatory T cell 100: 104–110. population (47). Expansion of regulatory T cells has also recently 24. Roosnek, E., and A. Lanzavecchia. 1991. Efficient and selective presentation of been observed in lupus nephritis patients treated with rituxmab antigen-antibody complexes by rheumatoid factor B cells. J. Exp. Med. 173: 487–489. (48). 25. Krzysiek, R., E. A. Lefevre, W. Zou, A. Foussat, J. Bernard, A. Portier, In summary, our results demonstrate that B cell depletion has a P. Galanaud, and Y. Richard. 1999. Antigen receptor engagement selectively induces macrophage inflammatory protein-1␣ (MIP-1␣) and MIP-1␤ chemokine dual effect on the development of arthritis. Depletion of B cells production in human B cells. J. Immunol. 162: 4455–4463. inhibits autoantibody production and prevents chronic activation 26. Yellin, M. J., J. Sinning, L. R. Covey, W. Sherman, J. J. Lee, E. Glickman-Nir, of autoreactive T cells. K. C. Sippel, J. Rogers, A. M. Cleary, M. Parker, et al. 1994. T lymphocyte T cell-B cell-activating molecule/CD40-L molecules induce normal B cells or chronic lymphocytic leukemia B cells to express CD80 (B7/BB-1) and enhance their costimulatory activity. J. Immunol. 153: 666–674. 27. O’Neill, S. K., Y. Cao, K. M. Hamel, P. D. Doodes, G. Hutas, and A. Finnegan. Acknowledgments 2007. Expression of CD80/86 on B cells is essential for autoreactive T cell ac- We thank Dr. Jeffrey Oswald and all the staff of the Comparative Research tivation and the development of arthritis. J. Immunol. 179: 5109–5116. Center of Rush University Medical Center for their expert technical 28. Shlomchik, M. J., J. E. Craft, and M. J. Mamula. 2001. From T to B and back assistance. again: positive feedback in systemic autoimmune disease. Nat. Rev. Immunol. 1: 147–153. 29. Gong, Q., Q. Ou, S. Ye, W. P. Lee, J. Cornelius, L. Diehl, W. Y. Lin, Z. Hu, Y. Lu, Y. Chen, et al. 2005. Importance of cellular microenvironment and cir- culatory dynamics in B cell immunotherapy. J. Immunol. 174: 817–826. Disclosures 30. Yanaba, K., Y. Hamaguchi, G. M. Venturi, D. A. Steeber, E. W. St. Clair, and Drs. Dunn and Kehry are employees of Biogen-Idec. T. F. Tedder. 2007. B cell depletion delays collagen-induced arthritis in mice: The Journal of Immunology 5003

arthritis induction requires synergy between humoral and cell-mediated immu- 2006. Effects of abatacept in patients with methotrexate-resistant active rheuma- nity. J. Immunol. 179: 1369–1380. toid arthritis: a randomized trial. Ann. Intern. Med. 144: 865–876. 31. Ahuja, A., J. Shupe, R. Dunn, M. Kashgarian, M. R. Kehry, and M. J. Shlomchik. 41. Weyand, C. M., and J. J. Goronzy. 2003. Ectopic germinal center formation in 2007. Depletion of B cells in murine lupus: efficacy and resistance. J. Immunol. rheumatoid synovitis. Ann. NY Acad. Sci. 987: 140–149. 179: 3351–3361. 42. Takemura, S., A. Braun, C. Crowson, P. J. Kurtin, R. H. Cofield, W. M. O’Fallon, 32. Svensson, L., J. Jirholt, R. Holmdahl, and L. Jansson. 1998. B cell-deficient mice J. J. Goronzy, and C. M. Weyand. 2001. Lymphoid neogenesis in rheumatoid do not develop type II collagen-induced arthritis (CIA). Clin. Exp. Immunol. 111: synovitis. J. Immunol. 167: 1072–1080. 521–526. 33. Shlomchik, M. J., M. P. Madaio, D. Ni, M. Trounstein, and D. Huszar. 1994. The 43. Takemura, S., P. A. Klimiuk, A. Braun, J. J. Goronzy, and C. M. Weyand. 2001. role of B cells in lpr/lpr-induced autoimmunity. J. Exp. Med. 180: 1295–1306. T cell activation in rheumatoid synovium is B cell dependent. J. Immunol. 167: 34. Hanyecz, A., S. E. Berlo, S. Szanto, C. P. Broeren, K. Mikecz, and T. T. Glant. 4710–4718. 2004. Achievement of a synergistic adjuvant effect on arthritis induction by ac- 44. Harris, D. P., L. Haynes, P. C. Sayles, D. K. Duso, S. M. Eaton, N. M. Lepak, tivation of innate immunity and forcing the immune response toward the Th1 L. L. Johnson, S. Swain, and F. E. Lund. 2000. Reciprocal regulation of polarized phenotype. Arthritis Rheum. 50: 1665–1676. cytokine production by effector B and T cells. Nat. Immunol. 1: 475–481. 35. Glant, T. T., A. Finnegan, and K. Mikecz. 2003. Proteoglycan-induced arthritis: 45. Zhai, Y., L. Meng, F. Gao, Y. Wang, R. W. Busuttil, and J. W. Kupiec- immune regulation, cellular mechanisms, and genetics. Crit. Rev. Immunol. 23: Weglinski. 2006. CD4ϩ T regulatory cell induction and function in transplant 199–250. recipients after CD154 blockade is TLR4 independent. J. Immunol. 176: 36. O’Keefe, T. L., G. T. Williams, S. L. Davies, and M. S. Neuberger. 1998. Mice 5988–5994. carrying a CD20 gene disruption. Immunogenetics 48: 125–132. 46. Oderup, C., H. Malm, H. Ekberg, Z. Qi, B. Veress, F. Ivars, and M. Corbascio. 37. Hamaguchi, Y., Y. Xiu, K. Komura, F. Nimmerjahn, and T. F. Tedder. 2006. 2006. Costimulation blockade-induced cardiac allograft tolerance: inhibition of T Antibody isotype-specific engagement of Fc␥ receptors regulates B lymphocyte ϩ ϩ cell expansion and accumulation of intragraft cD4 Foxp3 T cells. Transplan- depletion during CD20 immunotherapy. J. Exp. Med. 203: 743–753. tation 82: 1493–1500. 38. Hamaguchi, Y., J. Uchida, D. W. Cain, G. M. Venturi, J. C. Poe, K. M. Haas, and T. F. Tedder. 2005. The peritoneal cavity provides a protective niche for B1 and 47. Olson, T. S., G. Bamias, M. Naganuma, J. Rivera-Nieves, T. L. Burcin, W. Ross, conventional B lymphocytes during anti-CD20 immunotherapy in mice. J. Im- M. A. Morris, T. T. Pizarro, P. B. Ernst, F. Cominelli, and K. Ley. 2004. Ex- munol. 174: 4389–4399. panded B cell population blocks regulatory T cells and exacerbates ileitis in a Downloaded from 39. Gonggrijp, R., W. J. Mullers, H. F. Dullens, and C. P. van Boven. 1985. Anti- murine model of Crohn disease. J. Clin. Invest. 114: 389–398. bacterial resistance, macrophage influx, and activation induced by bacterial rRNA 48. Sfikakis, P. P., V. L. Souliotis, K. G. Fragiadaki, H. M. Moutsopoulos, with dimethyldioctadecylammonium bromide. Infect. Immun. 50: 728–733. J. N. Boletis, and A. N. Theofilopoulos. 2007. Increased expression of the FoxP3 40. Kremer, J. M., H. K. Genant, L. W. Moreland, A. S. Russell, P. Emery, functional marker of regulatory T cells following B cell depletion with rituximab C. Abud-Mendoza, J. Szechinski, T. Li, Z. Ge, J. C. Becker, and R. Westhovens. in patients with lupus nephritis. Clin. Immunol. 123: 66–73. http://www.jimmunol.org/ by guest on September 27, 2021