Anti-CD79 Antibody Induces Anergy That Protects against Autoimmunity Ian R. Hardy, Nadia Anceriz, François Rousseau, Matt B. Seefeldt, Eric Hatterer, Magali Irla, Vanessa Buatois, This information is current as Laurence E. Chatel, Andrew Getahun, Ashley Fletcher, of September 29, 2021. Laura Cons, Guillemette Pontini, Nicole A. Hertzberg, Giovanni Magistrelli, Pauline Malinge, Mia J. Smith, Walter Reith, Marie H. Kosco-Vilbois, Walter G. Ferlin and John C. Cambier Downloaded from J Immunol published online 17 January 2014 http://www.jimmunol.org/content/early/2014/01/17/jimmun ol.1302672 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2014/01/20/jimmunol.130267 Material 2.DCSupplemental

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 29, 2021 • No Triage! Every submission reviewed by practicing scientists

• 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published January 17, 2014, doi:10.4049/jimmunol.1302672 The Journal of Immunology

Anti-CD79 Antibody Induces B Cell Anergy That Protects against Autoimmunity

Ian R. Hardy,*,1 Nadia Anceriz,†,1 Franc¸ois Rousseau,† Matt B. Seefeldt,* Eric Hatterer,† Magali Irla,‡ Vanessa Buatois,† Laurence E. Chatel,‡ Andrew Getahun,* Ashley Fletcher,* Laura Cons,† Guillemette Pontini,† Nicole A. Hertzberg,* Giovanni Magistrelli,† Pauline Malinge,† Mia J. Smith,* Walter Reith,‡ Marie H. Kosco-Vilbois,† Walter G. Ferlin,†,2 and John C. Cambier*,2

B cells play a major role in the pathogenesis of many autoimmune disorders, including rheumatoid arthritis, systemic eryth- ematosus, multiple sclerosis, and type I diabetes mellitus, as indicated by the efficacy of B cell–targeted therapies in these diseases.

Therapeutic effects of the most commonly used B cell–targeted therapy, anti-CD20 mAb, are contingent upon long-term depletion Downloaded from of peripheral B cells. In this article, we describe an alternative approach involving the targeting of CD79, the transducer subunit of the B cell AgR. Unlike anti-CD20 mAbs, the protective effects of CD79-targeted mAbs do not require cell depletion; rather, they act by inducing an anergic-like state. Thus, we describe a novel B cell–targeted approach predicated on the induction of B cell anergy. The Journal of Immunology, 2014, 192: 000–000.

cells have proven to be effective targets for the treatment direct modes of B cell depletion by rituximab: Ab-dependent cellular http://www.jimmunol.org/ of multiple autoimmune disorders and B-lineage can- cytotoxicity (ADCC), complement-dependent cellular cytotoxicity B cers (1, 2). The most widely used B cell–targeted drug is (CDC), and the direct induction of via CD20 cross-linking rituximab, which has been approved in the United States since (9–11). The primacy of these mechanisms in rituximab-induced 1997 for treatment of B cell and since 2006 for treat- B cell loss in humans is unclear. ment of rheumatoid arthritis. The therapeutic usefulness of ritux- Rituximab is not consistently efficacious even among auto- imab was shown recently in multiple other autoimmune diseases, immunities known to be Ab mediated. For example, in mouse such as multiple sclerosis and type I diabetes mellitus (3, 4). De- models of lupus in which B cells express human CD20, rituximab spite inconclusive data from phase III clinical trials in systemic was unable to efficiently deplete B cells from secondary lymphoid lupus erythematosus (SLE), rituximab continues to see significant tissues or affect the course of disease, despite depletion of pe- by guest on September 29, 2021 off-label use for treatment of this disease (5). ripheral blood B cells (12). Indeed, the very applicability of rit- Rituximab is a chimeric human/mouse IgG1 mAb that targets uximab in SLE remains controversial. Two large, double-blinded, CD20 and mediates long-lasting depletion of peripheral B cells (6). placebo-controlled studies in SLE patients found that rituximab CD20 is a surface that is abundantly expressed on B- does not have any benefit over placebo (5, 13). However, the re- lineage cells from the pre-B cell stage to the plasmablast stage (7). sults of a number of nonblinded clinical trials and off-label use of Because CD20 is not expressed on plasma cells, rituximab does not rituximab suggest that it has clinical efficacy in SLE, although impair established Ab-mediated immunity gained from past infec- perhaps less than that seen in rheumatoid arthritis (14–16). tions and vaccinations (8). Empirical evidence supports at least three CD79 (Ig-a/b) may emerge as an alternative target for the treat- ment of B cell–dependent autoimmunity (17). CD79 is a disulphide- linked heterodimer of CD79a (Ig-a) and CD79b (Ig-b)andis *Department of Immunology, University of Colorado and National Jewish Health, associated with membrane Ig on the surface of B-lineage cells. Denver, CO 80206; †NovImmune S.A., 1228 Plan-Les-Ouates, Geneva, Switzerland; and ‡Department of Pathology and Immunology, University of Geneva Medical Together, these components constitute the B cell AgR. Upon an School, 1211 Geneva 4, Switzerland Ag-induced BCR aggregation, CD79 is phosphorylated and ini- 1I.R.H. and N.A. contributed equally to this work. tiates a cascade of downstream signaling events. Thus, B cells are 2W.G.F. and J.C.C. contributed equally to this work. activated and ready to receive further coactivating signals that Received for publication October 2, 2013. Accepted for publication December 16, drive proliferation and differentiation, ultimately delivering a 2013. memory cell pool and an appropriate humoral response. During This work was supported by the Colorado Bioscience Discovery Evaluation grant this process, B cells become robust APCs and release cytokines program and National Institutes of Health Grants P01 AI022295 and R01 AI077597. that can influence the quality of the immune response. Address correspondence and reprint requests to Dr. John C. Cambier, University of Work in our laboratory and those of others defined and char- Colorado and National Jewish Health, 1400 Jackson Street, K803, Denver, CO acterized an alternate mode of BCR signaling that is induced by 80206. E-mail address: [email protected] chronic AgR stimulation and maintains a state of B cell unrespon- The online version of this article contains supplemental material. siveness termed “anergy” (18–23). Anergic B cells are charac- Abbreviations used in this article: ADCC, Ab-dependent cellular cytotoxicity; ASC, Ab-secreting cell; BM, bone marrow; CDC, complement-dependent cellular cytotox- terized by the partial downregulation of surface BCR and impaired icity; CIA, collagen-induced arthritis; CII, collagen type II; HEL, hen egg lysozyme; propagation of activating signals that normally emanate from CD79, hIgG, hamster IgG; LN, lymph node; m, mouse; NP, nitrophenyl; NP4OVA, nitrophenyl- including activation of the SYK tyrosine kinase and extracellular conjugated OVA; RT, room temperature; SLE, systemic lupus erythematosus. Ca2+ influx,andtheyhavealifespanthatisreducedfrom∼40 d for Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 a typical naive B cell to ∼5 d (19, 21, 24–26). We hypothesized that

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1302672 2 THERAPEUTIC B CELL ANERGY the mechanism of B cell anergy might be harnessed for therapeutic RBCs and were resuspended in FACS buffer. To collect BM, the femurs inactivation of B cells. were flushed with the same medium as above using a syringe. Collected BM Recently, the therapeutic effectiveness of anti-CD79b mAb in cells were washed once in ACK and resuspended in FACS buffer. For CIA- related experiments, spleens or draining LNs were digested individually with the MRL/lpr mouse model of lupus was demonstrated (17). In the collagenase (2.4 mg/ml; Life Technologies/Invitrogen) and DNase (1 mg/ml; current study, we addressed the mechanism of anti-CD79b mAb- Sigma-Aldrich) for 30 min at RT. When appropriate, B cells were negatively mediated immune suppression. We report in this article that anti- selected using a CD43 B Cell Isolation (Miltenyi Biotec). CD79b mAb induces a polyclonal B cell anergy that is capable of Flow cytometric analysis preventing collagen-induced arthritis (CIA). These findings in- troduce a new strategy for therapeutic targeting of B cells that The following mAbs (from BD Biosciences unless otherwise stated) were does not require B cell depletion, but instead acts by disabling used for cell surface staining: anti-CD3–FITC (145-2C11), anti-CD4– allophycocyanin (RM4-5), anti-CD19–allophycocyanin (1D3), anti-CD19– AgR function. Pacific Blue (1D3; eBioscience), anti-CD43–Biotin (S7) detected with SA–Alexa Fluor 647 (Invitrogen), anti-CD43–FITC (S7), anti-CD44–FITC Materials and Methods (IM7), anti-CD45–PerCP (30F11), anti-CD69–PE (H1.2F3), anti-CD79 Mice (polyclonal rabbit) detected with goat anti-rabbit IgG–Alexa Fluor 649 (Molecular Probes), anti-CD79b–DyLight 488 (HM79), anti-CD79b– Unless otherwise noted, female mice were used at 2–6 mo of age. C57BL/6 DyLight 649 (HM79), anti-CD80–PE (1610A1), anti-CD86–PE (GL-1), mice, purchased from The Jackson Laboratory, were used as wild-type con- anti-B220–PerCP (RA3-6B2), anti-FcgRIIB–DyLight 488 (2.4G2), anti- trols. FcRg2/2 mice were a kind gift from the laboratory of Dr. E. Gelfand hIgG–FITC (G70-204/G94.56), anti-IA/IE FITC (2G9), anti-IgD–PE (11- (National Jewish Health). FcgRIIB2/2 mice were purchased from Taconic 26; Southern Biotech), anti-IgM–DyLight 649 (b-7-6), anti-MHCII–DyLight

Laboratories. These mice were bred and housed at the animal facility at 488 (D3.137), and anti-PNA FITC (Sigma-Aldrich). Cells were incubated with Downloaded from National Jewish Health, and the experiments were performed under Insti- Fc block (BD Biosciences) and stained in FACS buffer (PBS + 1% BSA + tutional Animal Care and Use Committee–approved protocols. CIA experi- 0.02% NaN3) to reduce nonspecific staining. Flow cytometry was performed ments were undertaken using adult 8-wk-old male DBA/1J mice. on cells using a CYAN flow cytometer (Beckman Coulter) or FACSCalibur flow cytometer (BD Biosciences) and analyzed using FlowJo (TreeStar), Induction of CIA CellQuest Pro (BD Biosciences), and PRISM (GraphPad) software. CIA was induced in male DBA/1J mice, as described (27). Briefly, mice Intracellular Ca2+ mobilization

were immunized with bovine collagen type II (CII) emulsified in CFA at http://www.jimmunol.org/ day 0. After 21 d, mice received a secondary immunization with CII Primary B cells were prepared for real-time intracellular calcium analysis, emulsified in IFA. as previously described (19). For stimulation in vitro, B cells were stim- Anti-CD79b, mouse (m)–anti-CD79b, m–anti-CD79b-D/A, anti-CD20, ulated with 3.0 mg b-7-6 (anti-m) or 1.5 mg 1-3-5 (anti-d) anti-Ig H chain anti–CD20-D/A, control hamster IgG (hIgG), and control mouse IgG2a Abs, 15.0 mg polyclonal rabbit anti-mouse–CD79, or ionomycin (1.0 mM) were administered s.c. on day 0. Two hours after mAb injection, the mice and analyzed by cytometry using an LSR II (BD Biosciences) and FlowJo were immunized with bovine CII. Clinical scores were assessed after the software (TreeStar). Experiments using B cells from CIA mice were secondary immunization on individual paws, applying a scale ranging from conducted as previously described; cells were resuspended in RPMI 1640 0 to 4, as previously described (28). supplemented with 0.1% BSA and 25 mM HEPES and loaded with fura 2-AM (1 mM; Invitrogen) for 30 min at 37˚C. After washing, the cells were BrdU labeling transferred to a 96-well plate. Stimulation was performed with 7.5 mg

polyclonal goat anti-mouse IgM using a FlexStation II device (Molecular by guest on September 29, 2021 For analysis of newly generated B cells, mice were given drinking water Devices) and analyzed by SoftMax Pro software. Following stimulation, the containing 0.8 mg/ml BrdU for 3 wk. Fresh BrdU-containing drinking water response was measured based on area under the curve of the relative intra- was protected from and refreshed every day. For the first week, the cellular calcium concentration minus that of the unstimulated background. BrdU-containing drinking water was supplemented with 1.0% glucose to avoid drinking aversion. Intracellular staining was conducted with BD Immunization with nitrophenyl-conjugated OVA Cytofix/Cytoperm and the FITC-conjugated anti-BrdU Ab, B44 (both from

BD Biosciences) To prepare nitrophenyl (NP)-conjugated OVA (NP4OVA) immunogen, a 1:1 volume mixture of 1 mg/ml NP4OVA and 10 mg/ml alum (Brenntag) was Histological assessment of arthritis incubated at RT on a rotator for 3 h. A total of 200 ml the mixture was injected Animals were sacrificed at experimental end point on day 41. The knee i.p. per mouse to achieve a final immunization with 100 mgNP4OVA and joints of four representative animals/group were collected, processed as 1 mg alum. previously described, and embedded in paraffin blocks. Serial midsagittal sections (8 mm thickness) of the whole knee joint were either stained with NP-specific ELISPOT assay H&E or Safranin O/fast green counterstaining (28). Histological sections For ELISPOT analysis, serial 2-fold dilutions of 1 3 106/ml splenocytes/ were graded as previously described (28). sample were resuspended and plated in triplicate in 96-well plates. Cells Measurement of free serum anti-CD79 mAb were incubated at 37˚C for 6 h. The plates were washed three times with PBS + 0.05% Tween, letting each wash sit for 10 min at RT. The plates One million C57BL/6 splenocytes/well were distributed in a V-bottom 96- were then washed twice with PBS. Next, 50 ml secondary goat anti-mouse well plate and blocked with anti-CD16/CD32 (BD Biosciences) for 20 min IgG (1:4000) with 2% BSA in PBS was added and incubated at 4˚C at 4˚C. After washing the cells, 5 ml whole mouse serum (or a dilution overnight. The plates were washed three times with PBS + 0.05% Tween. thereof) was diluted to a final volume of 50 ml and added to each well, Next, the plates were incubated with Elispot development buffer (25 mM followed by incubation for 30 min at 4˚C. After subsequent washing, cells 5-bromo-chloro-3-indolyl phosphate p-toluidine, 100 mM NaCl, 100 mM were stained with a secondary fluorescent Ab to mouse IgG2a and ana- Tris, 10 mM MgCl2 [at pH 9.5]) for 1 h. The reaction was stopped by lyzed by flow cytometry to quantitate B cell staining by the anti-CD79 washing the plate three times with double-distilled H2O. The number of mAb remaining in the original serum. spots at a cell dilution in the linear range was determined, and the number of Ab-secreting cells (ASCs) was calculated. Tissue harvesting Hen egg lysozyme–specific ELISPOT assay Spleens, lymph nodes (LNs), bone marrow (BM), and PBLs were prepared for analysis by a variety of methods. To harvest spleens and LNs, a single- For ELISPOT analysis, serial 2-fold dilutions of 1 3 106/ml splenocytes/ cell suspension was prepared by mechanical disruption in IMDM sup- sample were resuspended and plated in triplicate in 96-well plates. Plates plemented with 5% FCS, HEPES, Pen/Strep, and gentamicin. RBCs were were coated with 10 mg/ml hen egg lysozyme (HEL)/well, incubated at RT lysed using ACK (150 mM NH4Cl, 10 mM KHCO3, 100 mM Na2 EDTA) for 2 h, and blocked with 2% BSA in PBS for 2 h. Next, splenocytes from for 2 min at room temperature (RT). The cells were then resuspended in harvested mice were incubated at 37˚C for 6 h. The plates were then 103 volume of sterile PBS and washed in FACS buffer (1% BSA, 0.02% washed three times with PBS + 0.05% Tween, letting each wash sit for 10 NaN3 in PBS). To harvest PBLs, peripheral blood was collected in tubes min at RT. Finally, the plates were washed twice with PBS. Next, 50 ml containing 5 ml heparin. PBLs were washed twice with ACK to lyse the RS3.1-bt (anti-IgMa; 1:2000) with 2% BSA in PBS was added and incu- The Journal of Immunology 3 bated at 4˚C overnight. The plates were washed three times with PBS + Full-length 18B12 and chimeric HM79 H and L chain constructs were 0.05% Tween and then incubated with Streptavidin-AP (as directed by cotransfected in CHO cells, and stable cell lines were obtained by selection provider; Southern Biotech Cat. No. 7100-04) with 2% BSA in PBS and using blasticidin and zeocin (InvivoGen). Conditioned supernatants of Ig- incubated at 4˚C overnight. The plates were again washed three times with expressing CHO cells were collected, and rAbs were purified by protein G PBS + 0.05% Tween. Next, the plates were incubated with Elispot de- affinity column chromatography (GE Healthcare). velopment buffer (25 mM 5-bromo-chloro-3-indolyl phosphate p-toluidine, 100 mM NaCl, 100 mM Tris, 10 mM MgCl2 [pH 9.5]) for 1 h. The re- CD79a and CD79b cloning and expression in PEAK cells action was stopped by washing the plate three times with double-distilled The cDNAs encoding mouse CD79a (BC027633) in pCMV6-AC-GFP vector H2O. The number of spots at a cell dilution in the linear range was de- and mouse CD79b (NM_008339) in pCMV6 vector were purchased from termined, and the number of ASCs was calculated. OriGene Technologies. PEAK cells were cultured and transfected with these HEL immunization and B cell–adoptive transfer constructs, as previously described (30). The cells were used for FACS staining with anti-CD79b mAbs 48 h posttransfection. B cells from MD4 or MD4/ML5 mice treated with anti-CD79b mAb or control IgG were enriched by depletion of CD43+ cells with magnetic beads Statistics 6 (MACS anti-mouse CD43; Miltenyi Biotec). A total of 10 B cells in 200 ml Because of the nonnormal distributions of both the arthritis scores and the PBS was adoptively transferred by i.v. injection. One hour later, the recipient B cell counts, we used a nonparametric permutation testing approach to mice were immunized i.p. with HEL-conjugated SRBCs, as described pre- analyze the data in Figs. 1A, 7A, and 7B (31). Additionally, we adjusted viously (29). the p values from these tests to compensate for the multiple-significance testing. This adjustment used the permutation approach described by Measurement of IgG and NP-specific IgG Westfall and Young (32), a procedure that has the further advantage of Enzyme immunoassay/RIA 96-well plates (Costar) were coated with 50 ml taking into account the correlation structure of the data (multiple assess-

ments recorded for each animal). All statistical analyses were done using Downloaded from 20 mg/ml NP20BSA or NP2BSA or 1 mg/ml goat anti-mouse IgG or goat anti-mouse IgM in ELISA capture buffer (PBS + 150 mM NaCl). Plates SAS 9.2. were allowed to incubate overnight at 4˚C. The next day, wells were as- All other data are depicted as means with SEM. Data were graphed using pirated, blocked with 2% BSA in PBS, and allowed to incubate overnight PRISM software. Excluding the statistics described above, p values were at 4˚C. Serum was added and serially diluted in 3-fold increments across calculated using PRISM using one-tailed or two-tailed t tests, where ap- the plate. The first dilutions were 1:500 for total IgM, 1:1000 for total IgG, propriate. and 1:200 for NP2BSA- or NP20BSA-specific detection. Plates were in- Study approval

cubated overnight at 4˚C and then washed three times with PBS + 0.05% http://www.jimmunol.org/ Tween. Depending upon the desired mode of detection, 100 ml secondary- All animal studies were conducted under the procedures approved by the detection Ab was added. These reagents included goat anti-mouse IgG1– Institutional Animal Care and Use Committee at National Jewish Health or alkaline phosphatase (1:2000), goat anti-mouse IgG–alkaline phosphatase according to license from the Swiss veterinary office for animal experi- (1:2000), or goat anti-mouse IgM–alkaline phosphatase (1:1000) in 1% mentation. BSA in PBS + 0.05% Tween. Cloning and expression of rIgG2a chimeric Ab HM79 and Results 18B12 Anti-CD79b mAb inhibits development of CIA 18B12 VH and VL nucleotide sequences were synthesized by DNA2.0, CIA was shown to require B cells and has been used to define the according to the sequences described in patent application US 2007/0136826 by guest on September 29, 2021 A1. For expression in mammalian cells, 18B12 VH and VL sequences were therapeutic actions of B cell–targeted mAbs (33). To study the subcloned in-frame of an IL-2 signal sequence into pFUSE vectors (InvivoGen) efficacy of anti-CD79b mAb in this model, 8-wk-old DBA/1 mice containing wild-type mouse IgG2a backbone (pFUSEss-CHIg-mG2a) and were treated with control polyclonal hIgG or anti-CD79b mAb mouse constant IgΚ (pFUSE2ss-CLIg-mK), respectively. pFUSEss-CHIg- prior to induction of CIA. Mice treated with a single dose of anti- mG2a was subjected to site-directed mutagenesis using a QuikChange Site- CD79b mAb exhibited a delay in arthritis onset, as well as reduced Directed Mutagenesis kit (Agilent) to introduce D265A mutation with the following primers: 59-CATGTGTGGTGGTGGCTGTGAGCGAGGAT- disease incidence and severity (Fig. 1A, left and middle panels). GAC-39 and 59-GTCATCCTCGCTCACAGCCACCACCACACATG-39. Reduced B cell numbers were seen in peripheral blood for 2 wk, A total of 200 mg purified hamster HM79 (IgG2/l) Ab was subjected to followed by a gradual, but not entirely complete, recovery for the SDS-PAGE and transferred to a polyvinylidene difluoride membrane. The duration of the experiment (Fig. 1A, right panel). There was re- membrane was stained using Ponceau Red. Higher and lower molecular bands, corresponding to H chains and L chains, respectively, were sub- duced synovial lymphocytic infiltration in anti-CD79b mAb– jected to N-terminal Edman sequencing. The EVRLLES sequence was treated mice, although this did not reach statistical significance obtained after seven cycles for the hamster mAb higher molecular band. (p = 0.054) (Fig. 1B). Reduced infiltration correlated with in- No data were obtained with hamster mAb lower molecular band, sug- creased preservation of cartilage in anti-CD79b mAb–treated gesting that the sample was N-terminally blocked. After total RNA ex- mice, as indicated by histological analysis (Fig. 1C). The reduced traction from the HM79 hybridoma cell line and cDNA synthesis, the VH was PCR amplified using a degenerate forward primer, based on the disease score anti-CD79b mAb–treated mice also correlated with EVRLLES sequence described in human and rodent VHs, 59-GAGGT- decreased serum CII-specific IgG, in particular the IgG2a and GCGGCTKTTGGARTCTGG-39, and reverse primer 59-CTACGTTGCAG- IgG2b subclasses (Fig. 1D). Interestingly, not all subclasses were GTGATGGGCTGCTTG-39. Purified PCR product was further cloned into pCR4- affected equally by anti-CD79b mAb treatment; IgG1 Ab levels TOPO vector (Invitrogen) before sequencing. Then, the HM79 VH was subcloned in-frame of an IL-2 signal sequence into pFUSEss-CHIg-mG2a were not reduced significantly. vectors containing mouse IgG2a wild-type and mouse IgG2a D265A Anti-CD79b mAb treatment leads to a reduction in peripheral backbones (mouse IgG2a D265A was generated by site-directed mu- tagenesis, as described above). For VL, part of the hamster constant l-chain blood, LN, and splenic B cells was cloned using degenerate forward primer 59-CAARGCYAC-MYTGG- Anti-CD79b mAb treatment reduced B cell numbers in peripheral TGTGTMYG-39 and reverse primer 59-GRVRCABTCWGCASGRGMC- blood, as was reported in studies using other B cell–targeted ARRCTC-39, on the basis of hamster H57 Fab protein sequence (1NFD_E, GI: 2914251). Then, a specific reverse primer annealing in the hamster treatments in both mice and humans (1–6, 17). To determine constant L-chain was designed: 59-GTTTCCCCTTCATGGGTAACTT- whether anti-CD79b mAb treatment affected B cell numbers in GGCAGG-39. Finally, HM79 VL nucleotide sequence was amplified by other organs, we treated 8–12-wk-old C57BL/6 mice with 0.5 mg RACE PCR using the SMART RACE cDNA Amplification Kit (Clontech), of control hIgG, anti-CD79b mAb, or anti-CD20 mAb and ana- according to the manufacturer’s instructions, and cloned into pCR4-TOPO vector for sequencing. The VL sequence was further subcloned in-frame of lyzed spleens, LNs, BM, and peripheral blood. One week post- an IL-2 signal sequence into pFUSE2ss-CLIg-mL1 vector containing administration, anti-CD79b mAb treatment reduced B cell numbers mouse constant Igl1 for expression in mammalian cells. in the spleen by ∼50% (Fig. 2A, 2B). No concomitant increase in 4 THERAPEUTIC B CELL ANERGY Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. Anti-CD79b mAb inhibits development of CIA. DBA/1 mice (n = 10/group) were treated with 1.0 mg of anti-CD79b mAb or hIgG as control 2 h before primary immunization with CII. Twenty-one days later, the mice received a second immunization with CII. (A) Arthritis incidence (left panel), arthritis score (middle panel), and B cell (CD19+) count in peripheral blood (right panel) were assessed. This experiment is representative of three independent experiments. (B) Representative H&E-stained joint sections at day 41 (original magnification 35; insets 320). (C) Representative Safranin O/fast green–stained joint sections at day 41 (original magnification 320). (D) Serum was collected from five mice in each group once a week from days 14 to 41, and the serum titers of anti-CII IgG, IgG1, IgG2a, and IgG2b were determined. Vertical brackets denote SEM. *p , 0.05, **p , 0.01, ***p , 0.001. the number of peripheral T cells was observed (Supplemental Fig. 1). CD79b mAb treatment had no effect on the number of marginal However, loss of B cells was much less complete than that induced zone B cells recovered (CD21+/CD1dhi) (Fig. 2E). No change by an anti-CD20 mAb (18B12), a mouse IgG2a specific for mouse was observed in developing B cell populations (CD232/CD24hi/ CD20 (Fig. 2A). Consistent with previously published reports (34, CD93+/B220+) in the BM, despite loss of the mature recircu- 35), anti-CD20 mAb treatment led to loss of . 95% of B cells lating B cell populations (CD23+/CD24int/CD932/B220+/CD19+) from spleen. B cell loss in the LNs and peripheral blood was more (Fig. 2F). efficient although still not approaching the depletion previously reported with anti-CD20 mAb treatment (Fig. 2B). Genetic ab- Targeting CD79 with mAb does not induce B cell activation lation of whose products are required for Ab-dependent B cell activation is also a potential consequence of BCR liga- cellular cytotoxicity (FcRg and FcgRIIb)ordisablingcomple- tion in vivo. To explore the status of B cells following anti-CD79b ment-dependent cytotoxicity did not impair the ability of anti- mAb treatment, we assessed expression of B cell activation markers CD79b mAb to induce B cell loss from these organs, suggesting ex vivo 12–48 h following administration of anti-CD79b mAb. that the decrease in B cell numbers may be due to effects of the B cells in mice treated with anti-CD79b mAb exhibited a slight Ab on B cells other than Fc engagement of cytotoxic effectors upregulation of CD69 and CD86, but not CD44, at 12 h (Fig. 3A). (Supplemental Fig. 2). However, by 24 h, expression of these markers returned to baseline. We observed no differential effect of anti-CD79 treatment on Analysis of cells 1 wk following treatment revealed no significant the numbers of splenic immature (CD93+), mature (CD932), or upregulation of MHC class II, CD80, CD44, or CD86 (Fig. 3B). follicular B cells (Fig. 2C, 2D). Surprisingly, however, anti- As an additional assay of B cell activation and differentiation to The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/

FIGURE 2. Anti-CD79b mAb treatment leads to a reduction in peripheral B cells. Wild-type C57BL/6 mice were injected with 0.5 mg of control by guest on September 29, 2021 polyclonal hIgG or anti-CD79b mAb (HM79), or 0.25 mg of anti-CD20 mAb (18B12). Tissues were harvested for analysis 1 wk later. (A) Percentage of B cells (CD19+) among splenocytes (left panel), absolute number of splenocytes (middle panel), and absolute number of splenic B cells (CD19+)(right panel). Data are the total of five to seven independent experiments. (B) Percentage of B cells (CD19+) in peripheral blood (left panel) and LNs (middle panel). Absolute number of B cells (CD19+) in the LN (right panel). (C) Frequency (left panel) and absolute numbers (right panels) of immature (CD93+) and mature (CD932) B cells among the total splenic B cells (CD19+). (D) Frequency (left panel) and absolute number (right panel) of follicular B cells (CD23+, CD21+) among splenic B cells (CD19+). (E) Frequency (left panel) and absolute number (right panel) of marginal zone (CD21+, CD1dhi) B cells among splenic B cells (CD19+). (F) Frequency of immature B cells (CD232/CD24hi/CD93+/CD19+)(left panel) and recirculating mature B cells (CD23+/ CD24int/CD932/CD19+/B220+)(right panel) among total BM cells. Experiments in (B–F) are representative of three independent experiments (n = 3–5 mice/ group). Vertical brackets denote SEM. *p , 0.05, **p , 0.01, ***p , 0.001. ns, Not significant.

Ig-secreting cells, we analyzed serum levels of total IgM and m–anti-CD79b mAbs retained the ability to bind CD79b-expressing IgG in anti-CD79b mAb–treated mice. We detected no change in cells ex vivo and reduced B cell populations similarly to the pa- serum IgM or IgG levels relative to controls (Fig. 3C). Thus, BCR rental hamster mAbs in vivo (Supplemental Fig. 3). signaling induced in vivo by anti-CD79b does not cause signifi- To test the efficacy of these Abs, we examined their ability to cant B cell activation. modulate arthritis induced in DBA/1 mice, as described in Fig. 1. Mice were injected once at the beginning of the experiment with Protection from autoimmunity is not dependent on control mouse IgG2a, m–anti-CD20 mAb, m–anti-CD20-D/A anti-CD79b–induced reduction of peripheral B cells mAb, m–anti-CD79b mAb, or m–anti-CD79b-D/A mAb. B cell To determine whether the therapeutic activity of anti-CD79b mAb is numbers in peripheral blood and disease activity were assessed at dependent on Fc domain function, we engineered anti-CD79b and regular intervals thereafter. As previously reported, mice treated anti-CD20 mAbs to eliminate ADCC and CDC activity and tested with m–anti-CD20 mAb did not develop arthritis and had remark- their effectiveness in preventing disease. Variable regions of the ably reduced peripheral B cells for the duration of the experiment hamster anti-CD79b Ab, HM79, were cloned into mouse l-chain (Fig. 4A). Importantly, introduction of the D265A mutation and IgG2a H chain C region backbones to create a chimeric ham- eliminated the protective effects of m–anti-CD20 mAb, whereas ster/m–anti-CD79b. We then engineered an alanine substitution at its ability to deplete peripheral B cells was partially retained aspartic acid 265 in the chimeric anti-CD79b mAb to generate (Fig. 4A). These data suggest that the protective effects of anti- m–anti-CD79b–D/A and, as a control, into the anti-CD20 mAb to CD20 mAb are Fc dependent and predicated upon efficient and generate anti-CD20–D/A, 18B12. The 265 aspartic acid residue was complete peripheral B cell clearance. shown to be required for the functional engagement of IgG Fc with Similar to the results shown in Fig. 1, the incidence of and clinical FcgR and complement (36). Importantly, both mutant and nonmutant scores for CIA were reduced in mice treated with m–anti-CD79b-D/A 6 THERAPEUTIC B CELL ANERGY

FIGURE 3. Targeting CD79 with mAb does not induce B cell activation. Wild-type, C57BL/6 mice (n = 2 or 3/group) were injected with 0.5 mg of control polyclonal hIgG (shaded graph) or anti-CD79b mAb (HM79) (black trace). (A) Tissues were harvested from mice at 12, 24, and 48 h postinjection with Ab, and splenic B cells (CD19+) were stained for surface expression of CD69, CD86, and CD44, as described in Materials and Methods.(B) Tissues were harvested 1 wk after injection with mAb, and splenic B cells (CD19+) were stained for surface expression of MHC class II, CD44, CD80, and CD86. (C) Serum was collected 1 wk after injection with 0.5 mg of control polyclonal hIgG (solid line) or anti-CD79b mAb (HM79) (dashed line) and analyzed for the levels of serum IgM and IgG. Downloaded from mAb, despite the fact that this treatment caused less B cell loss Consistent with a nondepleting mode of action of m–anti-CD79b- than the functionally ineffective m–anti-CD20-D/A Ab (Fig. 4B). D/A Ab, peripheral blood B cell numbers returned to near normal http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 4. Protection from autoimmunity is not dependent on anti-CD79b–induced reduction of peripheral B cells. Arthritis was initiated in DBA/1 mice, as described in Materials and Methods.(A and B) On the same day as the first CII immunization, mice (n = 10) were treated with 1.0 mg of mouse IgG2a, anti- CD20 mAb, anti-CD20-D/A mAb, m–anti-CD79b mAb, or m–anti-CD79b–D/A mAb. Arthritis incidence (left panels), arthritis score of arthritic animals (middle panels), and B cell count in peripheral blood (right panels) were assessed. Statistical significance between the isotype control and the D/A mAb (see vertical bars) is as indicated by asterisks. This experiment is representative of two independent experiments. (C) Serum was collected from five mice in each group from days 4 to 41 and used to stain freshly isolated target B cells. Vertical brackets denote SEM. *p , 0.05, **p , 0.01, ***p , 0.001. ns, Not significant. The Journal of Immunology 7 levels within 2 wk of treatment (Fig. 4B). Despite complete recovery anergy in mediating the effects of anti-CD79b mAb, we deter- oftheperipheralBcellpoolpriortothe secondary collagen challenge mined whether in vivo treatment affects the ability of B cells to on day 21, m–anti-CD79b–D/A mAb treatment impaired the onset signal through their AgRs. B cells recovered from mice 7 d following of CIA and reduced disease severity. These observations are in a single administration of anti-CD79b mAb expressed reduced contrast to those seen with anti-CD20–D/A mAb, which, despite surface levels of IgM, IgD, and CD79b, consistent with ligand en- more efficient peripheral B cell depletion, did not suppress ar- counter (Fig. 5B). These recovered B cells were coated with anti- thritis development. Thus, depletion of peripheral B cells cannot CD79b mAb, as indicated by staining with anti-hamster Ig Ab account for the therapeutic efficacy of anti-CD79b mAb. (Fig. 5B). Control experiments indicated that binding of staining To assess the effect of D/Z mutations on mAb half-life, we reagents, including the affinity-purified rabbit anti-CD79 used to determined the serum levels of free mAb in mice from Fig. 4A and measure CD79 levels, was not significantly hindered by mono- 4B. As expected, the levels of free serum m–anti-CD20 mAb were clonal anti-CD79 adsorbed to the B cells. not altered by the D265A mutation, which lies outside of the FcRn Previous work from our laboratory demonstrated that chronic binding site (Fig. 4C, right panel) (37). Unexpectedly, the D265A engagement of the BCR in anergic B cells results in elevated basal mutation significantly reduced the half-life of free serum m–anti- Ca2+ and dampened BCR-mediated Ca2+ influx (19). Analysis of CD79b (Fig. 4D, left panel). Like m–anti-CD20 and m–anti- B cells recovered from m–anti-CD79b mAb and m–anti-CD79–D/ CD20–D/A, m–anti-CD79b mAb was detectable in the serum A–treated mice 1 d following treatment revealed elevated basal of treated mice for .2 wk following injection, whereas m–anti- Ca2+ and suppressed BCR-mediated Ca2+ influx (Fig. 5C). One CD79b–D/A was cleared about a week earlier. Thus, m–anti-CD79– week later, B cells from anti-CD79b–treated animals remained

D/A retains protective activity despite having a reduced half-life hyporesponsive, whereas those from m–anti-CD79–D/A–treated Downloaded from and truncated biological activity relative to m–anti-CD20–D/A. animals had partially recovered responsiveness (Fig. 5C). This is likely a consequence of the reduced half-life of the m–anti-CD79– Anti-CD79b mAb induces B cell AgR desensitization D/A mAb relative to the intact m–anti-CD79b mAb. Finally, The effectiveness of m–anti-CD79b–D/A mAb in CIA suggests B cells recovered following anti-CD20 mAb treatment did not that the therapeutic benefit may derive solely from the interaction display impaired BCR-mediated Ca2+ mobilization or increased 2+ of the mAb with its Ag, CD79b. Thus, we hypothesized that ex- basal intracellular Ca concentration (Fig. 5D). Thus, treatment http://www.jimmunol.org/ posure of B cells to a chronic AgR stimulus derived from the anti- of mice with anti-CD79b mAb, but not anti-CD20 mAb, leads CD79b mAb may mimic the tolerizing signals that occur naturally to B cell unresponsiveness to subsequent BCR stimulation, as is and drive autoreactive cells into a silenced state, termed “B cell characteristic of anergic B cells. anergy” (20, 23, 24). Consistent with this hypothesis, anti-CD79b mAb was able to productively engage BCR signaling, as dem- Anti-CD79b mAb induces B cell anergy onstrated by Ca2+ mobilization in B cells, whereas anti-CD20 B cells from mice treated with anti-CD79b mAb share many mAb was not (Fig. 5A). To further explore the possible role of features with anergic B cells, including phenotypic markers and by guest on September 29, 2021

2+ 2+ FIGURE 5. Anti-CD79b mAb induces B cell AgR desensitization. (A) Analysis of intracellular concentration of Ca ([Ca ]i) of wild-type, splenic B cells (B220+) in response to 10 mg/ml of anti-CD79b mAb or anti-CD20 mAb. Splenocytes were prepared as described in Materials and Methods. This experiment is representative of two independent experiments. (B) Wild-type C57BL/6 mice (n = 3/group) were injected with 0.5 mg of hIgG (shaded graph) or anti-CD79b Ab (open graph). One week later, tissues were harvested for analysis. Splenic B cells (CD19+) were assessed for surface expression of bound hIgG, IgM, IgD, and CD79, as described in Materials and Methods. This experiment is representative of three independent experiments. (C) Wild-type C57BL/6 mice were injected with 1.0 mg of control mouse IgG (heavy line), m–anti-CD79b mAb (solid line), or m–anti-CD79b–D/A mAb (light line). One week later, tissues were harvested, and intracellular Ca2+ influx of B cells (B220+) in response to BCR cross-linking was determined by flow cytometry. (D) Wild-type C57BL/6 mice were injected with 0.5 mg of control polyclonal hIgG (solid line) or 0.25 mg of anti-CD20 Ab (dashed line). One week later, tissues were harvested, and intracellular Ca2+ influx of B cells (B220+) in response to BCR cross-linking was determined by flow cytometry. 8 THERAPEUTIC B CELL ANERGY blunted B cell AgR–mediated Ca2+ signaling. To determine suggesting that the anti-CD79b mAb treatment led to a B cell– whether these characteristics extend to impairment of their ability intrinsic hyporesponsiveness that was functionally equivalent to to mount Ab responses in vivo, we treated mice with anti-CD79b anergy (Fig. 6C). mAb or control polyclonal hIgG and immunized them i.p. with Reduced recovery of B cells following anti-CD79b mAb NP4OVA precipitated in alum. On day 15 following immunization, serum levels of NP-specific IgG were measured. Ab responses treatment reflects transient relocalization of cells were reduced by 75% in anti-CD79b mAb–treated mice (Fig. 6A, Repopulation of the B cell compartment following treatment with upper panel). Interestingly, high-affinity NP-specific IgG was re- anti-CD20 requires weeks in mice and months in humans (38, 39). duced by .90% by anti-CD79b mAb treatment, suggesting se- However, in mice treated with m–anti-CD79b–D/A mAb, splenic lective impairment of affinity-matured responses (Fig. 6A, lower B cells returned to near-normal levels within 14 d of maximal panel). ELISPOT enumeration of total and high-affinity NP- B cell reduction, suggesting that treatment does not cause B cell specific IgG ASCs revealed a similar anti-CD79–induced sup- death (Fig. 7A). To determine whether this repopulation was due pression. Control IgG–treated mice had a mean of 117 ASC/106 to de novo generation of new B cells, mice were treated with Abs splenocytes, and ∼43% of these were secreting high-affinity Ab, and then continuously fed BrdU-containing drinking water for as indicated by detection with low-valency Ag. In contrast, anti- 3 wk to label B cells produced during the recovery period. Ap- CD79b–treated mice produced half as many total IgG+ anti-NP proximately 31.5% of the mature (CD932) splenic B cells from ASCs; of these, only 12% were high affinity (Fig. 6B). control mice had incorporated BrdU, consistent with generation B cell anergy is a cell-intrinsic phenomenon that is due, in part, to due to normal B cell turnover (Fig. 7B). In mice that had been altered BCR signaling. To determine whether anti-CD79b mAb– sublethally irradiated to kill most mature B cells, .80% of B cells Downloaded from induced B cell suppression was also cell intrinsic, we used the MD4/ incorporated BrdU during de novo generation in the BM. Sur- ML5 mouse model of B cell tolerance. MD4-transgenic mice ex- prisingly, only 10.9% of mature B cells from m–anti-CD79b– press a HEL-specific BCR and contain a mostly naive peripheral treated mice incorporated BrdU, despite near-complete restoration B cell population. ML5-transgenic mice express soluble HEL and, of the B cell compartment (Fig. 7B). No differences were ob- when crossed to the MD4 transgene, it results in the expression of served in BrdU incorporation between immature B cells and non-

HEL as a neocognate self-Ag (20). B cells from MD4/ML5 mice are B cells. In treated and control animals, BrdU incorporation in the http://www.jimmunol.org/ rendered anergic, as evidenced by reduced peripheral B cell num- LN B cells was similarly affected by the respective treatments (Fig. bers, lower surface IgM expression, and failure to mobilize calcium 7B). These data formally demonstrate that the B cell-“depleting” efficiently upon BCR cross-linking (20). We treated MD4 mice with effects of m–anti-CD79–D/A mAb treatment do not reflect B cell anti-CD79b mAb; 1 d later, 1 3 106 splenic B cells from these donor death. Rather, the treatment must reflect transient relocation of mice were purified and transferred into wild-type C57BL/6 (IgMb+) some B cells to other niches from which they return following the recipients. As control, we transferred anergic B cells isolated from clearance of m–anti-CD79b–D/A mAb from the animal. anti-HEL/soluble HEL (MD4/ML5)-transgenic mice. A naive B cell control was provided by B cells from untreated anti-HEL (MD4)- Discussion transgenic mice. One hour after transfer, recipient mice were Previous work by us (17) demonstrated that anti-CD79b mAbs are by guest on September 29, 2021 immunized by i.p. injection of SRBC-HEL. Five days after cell therapeutic in the MRL/lpr mouse model of lupus. In this article, transfer, spleens from recipient mice were harvested, and HEL- we report analysis of the mechanisms that underlie the efficacy of specific IgMa+ ASCs were measured by ELISPOT assay. Anergic anti-CD79b mAb treatment. Unlike other B cell–targeted bio- B cells from MD4/ML5 mice and B cells from MD4 mice treated logics that induce B cell death by ADCC, complement fixation, or with anti-CD79b mAb prior to transfer exhibited a similarly im- survival factor starvation–mediated cell death, anti-CD79b mAb paired immune response relative to control-treated MD4 B cells, appears to exert its therapeutic effect by inducing a transient state

FIGURE 6. Anti-CD79b mAb treatment induces B cell anergy. Wild-type, C57BL/6 mice (n = 3) were injected with 0.5 mg of hIgG or anti-CD79b mAb and immunized i.p. with 100 mgofNP4OVA in alum. On day 15, serum was collected, and tissues were harvested for analysis. (A) The relative amounts of total (upper panel) and high-affinity (lower panel) NP-specific IgG from mice treated with control polyclonal hIgG (solid line) or anti-CD79b mAb (dashed line). (B) Number of total and high-affinity NP-specific IgG+ ASCs/106 splenocytes. The experiment is representative of two independent experiments. (C) MD4 mice were injected with hIgG or anti-CD79b, and MD4/ML5 mice were injected with hIgG, as previously described. One day later, B cells were purified, and 106 B cells were adoptively transferred into recipient wild-type C57BL/6 mice (n = 5). One hour posttransfer, 100 ml of 10% SRBCs conjugated to HEL was injected i.p. Five days later, tissues were harvested for analysis, and the number of splenic HEL-specific IgMa+ ASCs was enu- merated. Vertical brackets denote SEM. *p , 0.05, ***p , 0.001. The Journal of Immunology 9

FIGURE 7. Loss of B cells following anti-CD79b mAb treatment reflects transient relocalization of cells rather than B cell death. (A) C57BL/6 wild-type mice (n = 5) were injected with 1.0 mg of control mouse IgG or m–anti-CD79b–D/A mAb. Three weeks later, they were harvested for analysis, splenocytes were enumerated, and the proportion of B cells was determined by FACS. (B) C57BL/6 wild-type mice (n = 4) were injected with 1.0 mg of control mouse IgG or m–anti-CD79b–D/A mAb or were sublethally irradiated. The mice were then fed drinking water containing BrdU (0.8 mg/ml) for 3 wk, after which the mice were sacrificed and harvested, as described in Materials and Methods. The percentages of BrdU+ mature B cells (CD19+/CD932), immature B cells (CD19+/CD93+), and non-B cells (CD19) were determined by intracellular BrdU staining and FACS analysis. Vertical brackets denote SEM. *p , 0.05, **p , 0.01, ***p , 0.001. Downloaded from of B cell anergy. Analogous to naturally occurring anergic B cells, ripheral B cell redistribution and the contributions of this phe- B cell exposure to anti-CD79b mAb in vivo leads to reduced nomenon to the disease-preventive activity of anti-CD79b mAb. It surface Ig, impaired BCR-triggered calcium mobilization, and is interesting that FcR binding–incompetent anti-CD3 mAbs in- failure to mount a complete and robust response to Ag challenge. duce a similarly transient redistribution of T cells in treated ani-

Further, we demonstrate that this polyclonal B cell anergy inhibits mals; this is associated with cell localization in the gut and is http://www.jimmunol.org/ pathogenic immune responses in the absence of complete or long- apparently the consequence of TCR signaling (45). Whether a sim- term B cell depletion. ilar response is activated in B cells by anti-CD79 Abs is currently The unique mechanism of action of anti-CD79b mAb relative to under study. anti-CD20 mAb derives from the distinct biologic function of its It is not clear how similar the molecular underpinnings of anti- target Ag, CD79. CD20 is a membrane tetraspanner that was CD79b mAb–induced unresponsiveness are to naturally occurring originally thought to have ion channel activity (40). Mice deficient anergy. Some minor differences do exist. For example, treatment in CD20 have relatively intact B cell development and humoral with anti-CD79b mAb causes downregulation of all components immunity (7). In contrast, CD79 is an integral component of the of the BCR, whereas, in most cases, anergic B cells display down- B cell AgR complex and is required for B cell development and regulation of IgM, with IgD remaining at near-normal levels (24). by guest on September 29, 2021 survival in the periphery (41–43). Signals emanating from CD79 Additionally, preliminary results suggest that one of the appar- are critical determinants of B cell fate decisions, among them ently redundant regulatory enzymes that is important in main- peripheral B cell anergy. taining anergy, the inositol 5-phosphate phosphatase SHIP-1, is In mice, peripheral B cell depletion by CD20-targeted mAb is dispensable for anti-CD79b mAb–induced desensitization, at primarily mediated by the interaction of the mAb’s Fc domain least in vitro (I.R. Hardy and J.C. Cambier, unpublished obser- with host FcgRs and clearance of mAb-coated B cells by ADCC vations) (18). It is important to note that anergy is a rapidly re- (34). When the concentration of anti-CD20 mAb falls below the versible phenomenon, the maintenance of which requires chronic threshold necessary to maintain clearance, B cells newly pro- AgR occupancy (19). The anergic-like state induced by anti-CD79b duced in BM repopulate the periphery, followed soon by a return therapy appears similarly reversible, and this would be advan- of pathology (27, 38, 39). Some reports (9, 11) suggest that tageous when patients develop disadvantageous side effects or certain anti-CD20 mAbs can induce tolerance or apoptosis di- opportunistic infections. rectly by cross-linking of CD20. However, considering the com- We suggest that with the development of additional strategies pelling evidence that ADCC is still critical for B cell depletion for suppression of pathogenic B cell function there may come a by these anti-CD20 mAbs, the direct induction of apoptosis is situation where multiple B cell–targeted therapies gain acceptance likely to be of only minor significance in anti-CD20 mAb ac- in the clinical market. As we better understand the mode of action tion (44). of B cell–targeted therapies in autoimmunity and, thus, determi- If the activity of anti-CD79b mAb does not require B cell de- nants of efficacy, it may be possible to predict an optimal course of pletion, then how is the Ab protective in CIA? Experiments re- B cell–targeted therapy based on the integrated calculus of disease ported in this article shed additional light on this question. The state and the genetics of the patient. One clear aim of next-gen- introduction of the D265A mutation, which ablates mAb ADCC eration B cell–targeted therapy is achievement of greater speci- and CDC activity, virtually eliminates the ability of anti-CD20 ficity, thus improving upon the current paradigm of disease control mAb to modify the course of CIA. In contrast to the anti-CD20 at the expense of global immune suppression. Anti-CD79b mAb mAb, anti-CD79b mAb possessing the same mutation retains therapy, with its potential to allow rapid immune recovery, may the ability to impair development of CIA, despite the return of represent a step toward this goal. B cells to peripheral lymphoid organs prior to disease onset. In conclusion, our studies define a novel, potentially therapeutic Experiments addressing the contribution of newly generated B cells approach that silences B cells en masse by inducing an Ag- to peripheral pools indicate that the disease-preventive effects of unresponsive state reminiscent of anergy. These findings have m–anti-CD79b–D/A mAb are associated with transient B cell far-reaching implications for the design of therapeutics targeting relocalization in the animal, rather than B cell death. Additional the many other receptors that are subject to ligand-induced de- studies are required to elucidate the basis of this induced pe- sensitization. 10 THERAPEUTIC B CELL ANERGY

Acknowledgments 21. Healy, J. I., R. E. Dolmetsch, L. A. Timmerman, J. G. Cyster, M. L. Thomas, G. R. Crabtree, R. S. Lewis, and C. C. Goodnow. 1997. Different nuclear signals We thank Robert Dunn, Marilyn Kehry, and Biogen Idec for generous pro- are activated by the B cell during positive versus negative signaling. vision of the 18B12 mAb used in these experiments. We thank Les Huson Immunity 6: 419–428. for assistance with the statistical analyses of several experiments and Sandra 22. O’Neill, S. K., A. Getahun, S. B. Gauld, K. T. Merrell, I. Tamir, M. J. Smith, Duran for assistance in preparing the manuscript. J. M. Dal Porto, Q. Z. Li, and J. C. Cambier. 2011. Monophosphorylation of CD79a and CD79b ITAM motifs initiates a SHIP-1 phosphatase-mediated in- hibitory signaling cascade required for B cell anergy. Immunity 35: 746–756. Disclosures 23. Nossal, G. J., and B. L. Pike. 1980. Clonal anergy: persistence in tolerant mice of The authors have no financial conflicts of interest. -binding B lymphocytes incapable of responding to antigen or mitogen. Proc. Natl. Acad. Sci. USA 77: 1602–1606. 24. Benschop, R. J., K. Aviszus, X. Zhang, T. Manser, J. C. Cambier, and References L. J. Wysocki. 2001. Activation and anergy in bone marrow B cells of a novel immunoglobulin transgenic mouse that is both hapten specific and autoreactive. 1. Edwards, J. C., L. Szczepanski, J. Szechinski, A. Filipowicz-Sosnowska, Immunity 14: 33–43. P. Emery, D. R. Close, R. M. Stevens, and T. Shaw. 2004. Efficacy of B-cell- 25. Healy, J. I., R. E. Dolmetsch, R. S. Lewis, and C. C. Goodnow. 1998. Quantitative targeted therapy with rituximab in patients with rheumatoid arthritis. N. Engl. J. and qualitative control of antigen receptor signalling in tolerant B lymphocytes. Med. 350: 2572–2581. Novartis Found. Symp. 215: 137–144; discussion 144–145, 186–190. ´ 2. Maloney, D. G., A. J. Grillo-Lopez, C. A. White, D. Bodkin, R. J. Schilder, 26. Vilen, B. J., S. J. Famiglietti, A. M. Carbone, B. K. Kay, and J. C. Cambier. 1997. J. A. Neidhart, N. Janakiraman, K. A. Foon, T. M. Liles, B. K. Dallaire, et al. B cell antigen receptor desensitization: disruption of receptor coupling to tyro- 1997. IDEC-C2B8 (Rituximab) anti-CD20 monoclonal antibody therapy in sine kinase activation. J. Immunol. 159: 231–243. patients with relapsed low-grade non-Hodgkin’s lymphoma. Blood 90: 2188– 27. Anolik, J. H., D. Campbell, R. E. Felgar, F. Young, I. Sanz, J. Rosenblatt, and 2195. R. J. Looney. 2003. The relationship of FcgammaRIIIa genotype to degree of 3. Hauser, S. L., E. Waubant, D. L. Arnold, T. Vollmer, J. Antel, R. J. Fox, A. Bar- B cell depletion by rituximab in the treatment of systemic lupus erythematosus. Or, M. Panzara, N. Sarkar, S. Agarwal, et al; HERMES Trial Group. 2008. B-cell Arthritis Rheum. 48: 455–459. depletion with rituximab in relapsing-remitting multiple sclerosis. N. Engl. J. Downloaded from 28. Lissilaa, R., V. Buatois, G. Magistrelli, A. S. Williams, G. W. Jones, S. Herren, Med. 358: 676–688. L. Shang, P. Malinge, F. Guilhot, L. Chatel, et al. 2010. Although IL-6 trans- 4. Pescovitz, M. D., C. J. Greenbaum, H. Krause-Steinrauf, D. J. Becker, signaling is sufficient to drive local immune responses, classical IL-6 signaling is S. E. Gitelman, R. Goland, P. A. Gottlieb, J. B. Marks, P. F. McGee, obligate for the induction of -mediated autoimmunity. J. Immunol. 185: A. M. Moran, et al; Type 1 Diabetes TrialNet Anti-CD20 Study Group. 2009. 5512–5521. Rituximab, B-lymphocyte depletion, and preservation of beta-cell function. N. 29. Getahun, A., M. J. Smith, I. Kogut, L. F. van Dyk, and J. C. Cambier. 2012. Engl. J. Med. 361: 2143–2152. Retention of anergy and inhibition of antibody responses during acute g her- 5. Merrill, J. T., C. M. Neuwelt, D. J. Wallace, J. C. Shanahan, K. M. Latinis, pesvirus 68 infection. J. Immunol. 189: 2965–2974. J. C. Oates, T. O. Utset, C. Gordon, D. A. Isenberg, H. J. Hsieh, et al. 2010. 30. Muraki, M., and S. Honda. 2010. Efficient production of human http://www.jimmunol.org/ Efficacy and safety of rituximab in moderately-to-severely active systemic lupus erythematosus: the randomized, double-blind, phase II/III systemic lupus extracellular domain-human IgG1 heavy chain Fc domain fusion protein using erythematosus evaluation of rituximab trial. Arthritis Rheum. 62: 222–233. baculovirus/silkworm expression system. Protein Expr. Purif. 73: 209–216. 6. Reff, M. E., K. Carner, K. S. Chambers, P. C. Chinn, J. E. Leonard, R. Raab, 31. Good, P. 1994. Permutation Tests: A Practical Guide to Resampling Methods for R. A. Newman, N. Hanna, and D. R. Anderson. 1994. Depletion of B cells Testing Hypotheses. Springer Series in Statistics. Springer Publishing, New York, in vivo by a chimeric mouse human monoclonal antibody to CD20. Blood 83: NY. 435–445. 32. Westfall, P. H., and S. S. Young. 1993. Resampling-Based Multiple Testing: 7. Uchida, J., Y. Lee, M. Hasegawa, Y. Liang, A. Bradney, J. A. Oliver, K. Bowen, Examples and Methods for p-Value Adjustment. Wiley Series in Probability and D. A. Steeber, K. M. Haas, J. C. Poe, and T. F. Tedder. 2004. Mouse CD20 Statistics. John Wiley and Sons, Hoboken, NJ. expression and function. Int. Immunol. 16: 119–129. 33. Yanaba, K., Y. Hamaguchi, G. M. Venturi, D. A. Steeber, E. W. St Clair, and 8. Cambridge, G., M. J. Leandro, J. C. Edwards, M. R. Ehrenstein, M. Salden, T. F. Tedder. 2007. B cell depletion delays collagen-induced arthritis in mice: M. Bodman-Smith, and A. D. Webster. 2003. Serologic changes following arthritis induction requires synergy between humoral and cell-mediated immu- by guest on September 29, 2021 B lymphocyte depletion therapy for rheumatoid arthritis. Arthritis Rheum. 48: nity. J. Immunol. 179: 1369–1380. 2146–2154. 34. Uchida, J., Y. Hamaguchi, J. A. Oliver, J. V. Ravetch, J. C. Poe, K. M. Haas, and 9. Byrd, J. C., S. Kitada, I. W. Flinn, J. L. Aron, M. Pearson, D. Lucas, and T. F. Tedder. 2004. The innate mononuclear phagocyte network depletes J. C. Reed. 2002. The mechanism of tumor cell clearance by rituximab in vivo in B lymphocytes through -dependent mechanisms during anti-CD20 patients with B-cell chronic lymphocytic : evidence of caspase acti- antibody immunotherapy. J. Exp. Med. 199: 1659–1669. vation and apoptosis induction. Blood 99: 1038–1043. 35. Yu, S., R. Dunn, M. R. Kehry, and H. Braley-Mullen. 2008. B cell depletion 10. Clynes, R. A., T. L. Towers, L. G. Presta, and J. V. Ravetch. 2000. Inhibitory Fc inhibits spontaneous autoimmune thyroiditis in NOD.H-2h4 mice. J. Immunol. receptors modulate in vivo cytotoxicity against tumor targets. Nat. Med. 6: 443– 180: 7706–7713. 446. 36. Baudino, L., Y. Shinohara, F. Nimmerjahn, J. Furukawa, M. Nakata, 11. Shan, D., J. A. Ledbetter, and O. W. Press. 1998. Apoptosis of malignant human E. Martı´nez-Soria, F. Petry, J. V. Ravetch, S. Nishimura, and S. Izui. 2008. B cells by ligation of CD20 with monoclonal antibodies. Blood 91: 1644–1652. Crucial role of aspartic acid at position 265 in the CH2 domain for murine IgG2a 12. Ahuja, A., J. Shupe, R. Dunn, M. Kashgarian, M. R. Kehry, and M. J. Shlomchik. and IgG2b Fc-associated effector functions. J. Immunol. 181: 6664–6669. 2007. Depletion of B cells in murine lupus: efficacy and resistance. J. Immunol. 37. Carver, D. J., M. J. Aman, and K. S. Ravichandran. 2000. SHIP inhibits Akt 179: 3351–3361. activation in B cells through regulation of Akt membrane localization. Blood 96: 13. Rovin, B. H., R. Furie, K. Latinis, R. J. Looney, F. C. Fervenza, J. Sanchez- 1449–1456. Guerrero, R. Maciuca, D. Zhang, J. P. Garg, P. Brunetta, and G. Appel; LUNAR 38. Beers, S. A., R. R. French, H. T. Chan, S. H. Lim, T. C. Jarrett, R. M. Vidal, Investigator Group. 2012. Efficacy and safety of rituximab in patients with active S. S. Wijayaweera, S. V. Dixon, H. Kim, K. L. Cox, et al. 2010. Antigenic proliferative lupus nephritis: the Lupus Nephritis Assessment with Rituximab modulation limits the efficacy of anti-CD20 antibodies: implications for antibody study. Arthritis Rheum. 64: 1215–1226. selection. Blood 115: 5191–5201. 14. Leandro, M. J., J. C. Edwards, G. Cambridge, M. R. Ehrenstein, and 39. Roll, P., A. Palanichamy, C. Kneitz, T. Dorner, and H. P. Tony. 2006. Regen- D. A. Isenberg. 2002. An open study of B lymphocyte depletion in systemic eration of B cell subsets after transient B cell depletion using anti-CD20 anti- lupus erythematosus. Arthritis Rheum. 46: 2673–2677. bodies in rheumatoid arthritis. Arthritis Rheum. 54: 2377–2386. 15.Perrotta,S.,F.Locatelli,A.LaManna,L.Cennamo,P.DeStefano,andB.Nobili. 40. Bubien, J. K., L. J. Zhou, P. D. Bell, R. A. Frizzell, and T. F. Tedder. 1993. 2002. Anti-CD20 monoclonal antibody (Rituximab) for life-threatening auto- Transfection of the CD20 cell surface molecule into ectopic cell types generates immune haemolytic anaemia in a patient with systemic lupus erythematosus. Br. a Ca2+ conductance found constitutively in B lymphocytes. J. Cell Biol. 121: J. Haematol. 116: 465–467. 1121–1132. 16. Weide, R., J. Heymanns, A. Pandorf, and H. Ko¨ppler. 2003. Successful long-term 41. Gong, S., and M. C. Nussenzweig. 1996. Regulation of an early developmental treatment of systemic lupus erythematosus with rituximab maintenance therapy. checkpoint in the B cell pathway by Ig beta. Science 272: 411–414. Lupus 12: 779–782. 42. Pelanda, R., U. Braun, E. Hobeika, M. C. Nussenzweig, and M. Reth. 2002. 17. Li, Y., F. Chen, M. Putt, Y. K. Koo, M. Madaio, J. C. Cambier, P. L. Cohen, and B cell progenitors are arrested in maturation but have intact VDJ recombination R. A. Eisenberg. 2008. B cell depletion with anti-CD79 mAbs ameliorates au- in the absence of Ig-alpha and Ig-beta. J. Immunol. 169: 865–872. toimmune disease in MRL/lpr mice. J. Immunol. 181: 2961–2972. 43. Bannish, G., E. M. Fuentes-Panana´, J. C. Cambier, W. S. Pear, and J. G. Monroe. 18. Browne, C. D., C. J. Del Nagro, M. H. Cato, H. S. Dengler, and R. C. Rickert. 2001. Ligand-independent signaling functions for the B lymphocyte antigen 2009. Suppression of phosphatidylinositol 3,4,5-trisphosphate production is a receptor and their role in positive selection during B lymphopoiesis. J. Exp. Med. key determinant of B cell anergy. Immunity 31: 749–760. 194: 1583–1596. 19. Gauld, S. B., R. J. Benschop, K. T. Merrell, and J. C. Cambier. 2005. Mainte- 44. de Haij, S., J. H. Jansen, P. Boross, F. J. Beurskens, J. E. Bakema, D. L. Bos, nance of B cell anergy requires constant antigen receptor occupancy and sig- A. Martens, J. S. Verbeek, P. W. Parren, J. G. van de Winkel, and J. H. Leusen. naling. Nat. Immunol. 6: 1160–1167. 2010. In vivo cytotoxicity of type I CD20 antibodies critically depends on Fc 20. Goodnow, C. C., J. Crosbie, S. Adelstein, T. B. Lavoie, S. J. Smith-Gill, receptor ITAM signaling. Cancer Res. 70: 3209–3217. R. A. Brink, H. Pritchard-Briscoe, J. S. Wotherspoon, R. H. Loblay, K. Raphael, 45. Waldron-Lynch, F., O. Henegariu, S. Deng, P. Preston-Hurlburt, J. Tooley, R. et al. 1988. Altered immunoglobulin expression and functional silencing of self- Flavell, and K. C. Herold. 2012. Teplizumab induces human gut-tropic regula- reactive B lymphocytes in transgenic mice. Nature 334: 676–682. tory cells in humanized mice and patients. Sci. Transl. Med 4: 118ra12.