gp49B-Mediated Negative Regulation of Antibody Production by Memory and Marginal Zone B Cells

This information is current as Saori Fukao, Kei Haniuda, Takuya Nojima, Toshiyuki Takai of October 2, 2021. and Daisuke Kitamura J Immunol 2014; 193:635-644; Prepublished online 16 June 2014; doi: 10.4049/jimmunol.1302772 http://www.jimmunol.org/content/193/2/635 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2014/06/16/jimmunol.130277 Material 2.DCSupplemental http://www.jimmunol.org/ References This article cites 67 articles, 37 of which you can access for free at: http://www.jimmunol.org/content/193/2/635.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 October 2, 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology gp49B-Mediated Negative Regulation of Antibody Production by Memory and Marginal Zone B Cells

Saori Fukao,*,1 Kei Haniuda,*,1 Takuya Nojima,*,2 Toshiyuki Takai,† and Daisuke Kitamura*

The rapid Ab responses observed after primary and secondary immunizations are mainly derived from marginal zone (MZ) and memory B cells, respectively, but it is largely unknown how these responses are negatively regulated. Several inhibitory receptors have been identified and their roles have been studied, but mainly on follicular B cells and much less so on MZ B, and never on memory B cells. gp49B is an Ig superfamily member that contains two ITIMs in its cytoplasmic tail, and it has been shown to negatively regulate mast cell, macrophage, and NK cell responses. In this study, we demonstrate that gp49B is preferentially expressed on memory and MZ B cells. We show that gp49B2/2 mice produce more IgM after a primary immunization and more IgM and IgG1 after a secondary immunization than gp49B+/+ mice in –dependent immune responses. Memory and MZ Downloaded from B cells from gp49B2/2 mice also produce more Abs upon in vitro stimulation with CD40 than those from gp49B+/+ mice. The in vitro IgM production by MZ B cells from gp49B+/+, but not gp49B2/2, mice is suppressed by interaction with a putative gp49B 2/2 ligand, the integrin avb3 heterodimer. In addition, gp49B mice exhibited exaggerated IgE production in the memory recall response. These results suggest that plasma cell development from memory and MZ B cells, as well as subsequent Ab production, are suppressed via gp49B. In memory B cells, this suppression also prevents excessive IgE production, thus curtailing allergic diseases. The Journal of Immunology, 2014, 193: 635–644. http://www.jimmunol.org/

ecretion of Abs by the plasma cell progeny of B cells is is attributed partly to expanded clone size and elevated reactivity the effector component of the humoral immune response. and affinity of memory B cells resulting from somatic hyper- S Follicular (Fo) B cells, which constitute most mature mutation. Memory B cells have been defined as resting B cells B cells in the secondary lymphoid organs, play a central role in the that express Ag-specific Igs of switched isotypes and have a immune response to T cell–dependent (TD) Ags. In TD responses, long life span in immunized mice, although memory B cells Ag-stimulated Fo B cells proliferate and differentiate into short- expressing unswitched receptors (IgM and IgD) have also been lived plasma cells or become (GC) B cells upon recognized (3–5). Memory B cells are also identified by coex- by guest on October 2, 2021 costimulation by cognate Th cells (1). In the GC, B cells undergo pression of several cell-surface markers, for example, CD38+, Ig isotype switching and somatic hypermutation and selection CD80+, and so on (6, 7). based on the affinity of their BCR for Ag. The selected high- Marginal zone (MZ) B cells are another type of mature B cells affinity B cells differentiate into long-lived plasma cells or mem- that are distinguishable from Fo B cells by their distinct cell-surface ory B cells, both of which contribute to long-lasting protection phenotype (CD21high CD23low) and their restricted localization in against infection by pathogens such as viruses (2). the splenic MZ, a region surrounding follicles in the . The memory recall response is characterized by rapid and aug- MZ B cells are known to respond to both TD and T cell–inde- mented production of Abs with increased affinity for Ag, which pendent (TI) Ags (8–11). MZ B cells have increased basal activity of Akt, Erk, Jnk, and c-Myc, and therefore respond to stimulation more rapidly than Fo B cells and are prone to differentiate into *Division of Molecular Biology, Research Institute for Biomedical Sciences, Tokyo plasma cells (9, 12, 13). MZ B cells share common features with University of Science, Noda, Chiba 278-0022, Japan; and †Department of Experi- mental Immunology, Institute of Development, Aging and Cancer, Tohoku Univer- memory B cells in that they live longer, have a lower threshold of sity, Sendai, Miyagi 980-8575, Japan activation, and produce Abs more quickly upon TD immunization 1S.F. and K.H. contributed equally to this work. than Fo B cells (9, 10, 14, 15). 2Current address: Astellas Pharma Inc., Tsukuba, Ibaraki, Japan. Despite their essential role in the host protection against Received for publication October 17, 2013. Accepted for publication May 13, 2014. pathogens, Abs may do harm to the host when their production is deregulated: Abs from polyreactive and self-reactive clones, which This work was supported by Grants-in-Aid for Scientific Research (B) and on Inno- vative Areas (to D.K.), the Japan Society for the Promotion of Science, and the are known to be included in the normal B cell repertoire, may Ministry of Education, Culture, Sports, Science and Technology of Japan. cause autoimmune diseases (16), and deregulated production of Address correspondence and reprint requests to Dr. Daisuke Kitamura, Research IgE may induce allergic diseases. Therefore, Ab production must Institute for Biomedical Sciences, Tokyo University of Science, Yamazaki 2669, be properly regulated. B cell activation is negatively regulated Noda, Chiba 278-0022, Japan. E-mail address: [email protected] through inhibitory receptors such as CD22, PIR-B, and Siglec-G The online version of this article contains supplemental material. (17, 18). CD22 is a dominant regulator of BCR-mediated calcium Abbreviations used in this article: AFC, Ab-forming cell; alum, aluminum hydroxide; B1-8 ki, B1-8 IgH knock-in; CGG, chicken g-globulin; Fo, follicular; GC, germinal signaling (19, 20), and CD22-deficient mice exhibit augmented center; iGB, induced GC B; iMB, induced memory-like B; MZ, marginal zone; NP, TD immune response and autoantibody production (19), although (4-hydroxy-3-nitrophenyl) acetyl; TD, T cell dependent; TI, T cell independent; TI- this phenotype was not reproduced in a different strain of CD22- II, T cell–independent type II. deficient mice (20). PIR-B controls B cell activation by inhibiting Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 BCR or TLR9-mediated signals (18, 21). Siglec-G inhibits the www.jimmunol.org/cgi/doi/10.4049/jimmunol.1302772 636 gp49B REGULATES MEMORY/MZ B CELL Ab PRODUCTION

BCR-mediated calcium signaling and physiologically regulates eBioscience, or Southern Biotech), streptavidin and anti-CD138 (Bio- the expansion of B1 cells, which express higher levels of Siglec-G legend or BD Biosciences), anti-CD43 (S7) and CD21 (BD Biosciences), ε than other B cell subsets (22). Thus, the roles played by inhibitory anti-CD23, CD19, CD5, CD38, Fc RIa, CD45.1, integrin av and integrin b3 (Biolegend), and NP9-BSA labeled with biotin or Alexa 647. All receptors on B cells have become evident, but less is known about samples were analyzed using a FACSCalibur flow cytometer or FACS- their involvement in MZ B cells (23), and no inhibitory receptors CantoII (BD Biosciences). The data were analyzed using FlowJo (Tree on memory B cells have so far been studied in rodent models. Be- Star). cause MZ B and memory B cell responses are characterized by a Cell purification and culture rapid burst of Ab production, strict regulation would seem to be re- quired. For B cell isolation, non-B cells were removed from spleen cells by in- In this study, we demonstrate that memory and MZ B cells cubation with the mixture of biotinylated Abs against CD4, CD8, DX5, Ter- 119 (Biolegend), and CD43 (S7; BD Biosciences), followed by sequential express gp49, a 49-kDa membrane protein belonging to the Ig negative sorting using the MACS system (Miltenyi Biotec) and iMag system superfamily (24). There are two subtypes of gp49, gp49A and (BD Biosciences). Fo and MZ B cells were sorted from pooled spleen cells gp49B, which have significant homology in their extracellular of mice (at least two) using FACSAria II (BD Biosciences). For memory +/+ 2/2 domains (25). We demonstrate in this article that gp49B (also B cell purification, splenic B cells sorted from gp49B or gp49B B1-8 ki CD45.1 mice were transferred into C57BL/6 mice and the recipients termed Lilrb4) is dominantly expressed on memory and MZ were immunized on the next day. Four weeks later, pooled spleen cells B cells. Although gp49A has a short cytoplasmic tail that lacks from 15 recipient mice were stained with biotin anti-CD45.2, and CD45.2+ any specific signaling motifs, gp49B contains two ITIMs in its cells were depleted using the MACS system. From the remaining cells, + + + + + cytoplasmic tail that mediate inhibition of cellular activation by memory B cells (CD45.1 B220 NP CD38 IgG1 ) were sorted using recruiting SHP-1 and SHP-2 (26–29). gp49B has been shown to FACSAria II as shown in Supplemental Fig. 3C. These procedure yielded

.95% purity of corresponding B cell subsets. Feeder cells were removed Downloaded from bind to the integrin av and b3 (avb3), which is expressed on a from cells of the induced GC B (iGB) cell culture by staining with anti- variety of cells including activated T cells, and this interaction H2Kd (Biolegend) Ab, followed by depletion using the iMag system (BD inhibits IgE-dependent degranulation of mast cells (30–32). Other Biosciences). studies showed that gp49B is also expressed on activated T cells Purified B cells were cultured with 10 mg/ml (or otherwise as indicated) anti-CD40 (Southern Biotech), 10 mg/ml LPS (Sigma), or 10 mg/ml anti- and NK cells, and regulates IFN-g production (32–34). However, IgM (Jackson Immunoresearch) in RPMI 1640 medium (Sigma) supple- the function of gp49B on any B cell subset has not been deter- mented with 10% FBS, 5.5 3 1025 M 2-ME, 10 mM HEPES, 1 mM mined. sodium pyruvate, 100 U/ml penicillin, and 100 mg/ml streptomycin (Life http://www.jimmunol.org/ 2/2 In this article, we show that gp49B mice produce higher Technologies) in a humidified atmosphere at 37˚C with 5% CO2. amounts of IgM after a primary immunization and of IgM, IgG1, Generation of BM chimeric mice and IgE after a secondary immunization than gp49B+/+ mice. In vitro analyses suggest that gp49B suppresses CD40-mediated Ab BM cells from mice (at least three) of each indicated genotype, mixed at the indicated ratios, were injected i.v. (5 3 106 cells/mouse) into lethally ir- production through interaction with the avb3 integrin. Thus, we radiated mice (5.5 Gy, twice at 3-h intervals) of the indicated genotype. At have found a novel regulatory mechanism that attenuates Ab least 8 wk after the transfer, the chimeric mice were used for immuniza- production by memory and MZ B cells, and this mechanism may tion. be required to control the rapid and intensive Ab responses Adoptive transfer by guest on October 2, 2021 commontothesecells. The indicated number of B cells purified from the indicated strains of mice was transferred i.v. into C57BL/6 or mMT mice, which were then im- Materials and Methods munized with NP-CGG/alum or soluble NP-CGG on the next day, as in- Mice and immunizations dicated. C57BL/6 mice were purchased from Sankyo Labo Service. B1-8 IgH ELISA and ELISPOT assays knock-in (B1-8 ki) mice (35) backcrossed to congenic C57BL/6-CD45.1 strains were used where indicated. We used heterozygous B1-8 knock-in NP-specific Ab-forming cells (AFCs) were detected by ELISPOT assay mice, in which the endogenous IgH allele is effectively excluded (36). using the MultiScreen 96-well filtration plate (Millipore) coated with 2/2 gp49B mice (37) and mMT mice (38) were also used. All mice were 10 mg/ml NP13.6-BSA in triplicate, as described previously (39). Anti-NP maintained in our mouse facility under specific pathogen-free conditions, IgM spots were revealed by HRP-conjugated goat anti-mouse IgM Ab and treated under the protocols approved by the Animal Care and Use (Southern Biotech) in conjunction with 3-amino-9-ethylcarbazole substrate Committee of the Tokyo University of Science. Mice were immunized i.p. (Dako). NP-specific IgM and IgG1 were detected by ELISA as described with 100 mg (4-hydroxy-3-nitrophenyl) acetyl (NP)35-chicken g-globulin previously (39). Total IgM was detected using 96-well plates coated with (CGG) precipitated in aluminum hydroxide (alum) or 50 mgNP50-Ficoll goat anti-mouse IgM Ab and with BSA as a blocking reagent. To measure (Biosearch Technologies) and boosted i.v. with 25 mg soluble NP32-CGG NP-specific IgE titers, we coated plates with purified rat anti-mouse IgE unless otherwise noted. To induce IgE responses, mice were primed s.c. (BD Biosciences), and bound Abs were revealed with biotinylated with 150 mgNP32-CGG precipitated in alum, 25 mg per site into bilateral NP9-BSA and HRP-conjugated streptavidin (Southern Biotech). lower flanks and the base of tail, and 50 mg into scruff of the neck. In the Western blotting secondary immunization, 25 mg each of soluble NP32-CGG was injected i.p. and into the scruff of the neck s.c., and 12.5 mg per site into bilateral Cell samples were lysed in 13 SDS-PAGE sample buffer, sonicated, lower flanks s.c. (75 mg in total per immunization). Mice were i.v. ad- boiled, and used for SDS-PAGE, followed by Western blotting. The fol- ministrated 25 mg each of soluble NP32-CGG as the tertiary immunization, lowing primary polyclonal Abs were used: rabbit anti–phospho-p44/42 to induce active anaphylaxis, and rectal temperature of the mice was MAPK (ERK1/2; Cell Signaling), rabbit anti-p44/42 MAPK (Cell Sig- measured thereafter. naling), rabbit anti–phospho-Akt (Ser473; Cell Signaling), and rabbit anti- Akt (Cell Signaling). Goat anti-rabbit IgG-HRP (Zymed) was used as Flow cytometry a secondary Ab. ECL Western Blotting Detection Reagents (Amersham) were used for detection. Single-cell suspensions from were depleted of RBCs by ammonium chloride lysis, incubated with anti-FcgRII/III Ab (2.4G2) to block FcgRs, Proliferation assay and stained for 30 min on ice in PBS containing 0.5% BSA, 2 mM EDTA, 0.05% sodium azide with different combinations of the following Abs and MZ B cells (1 3 105/well) were cultured in 96-well flat plates (BD reagents conjugated with FITC, PE, biotin, allophycocyanin, PE-Cy7, Falcon) as described earlier for 48 h, with the last 8 h in the presence of allophycocyanin-Cy7, or Pacific Blue: anti-IgG1 (Southern Biotech), anti- [3H]thymidine (1 mCi/well; PerkinElmer). Incorporated [3H]thymidine IgD and CD45.2 (Beckman Coulter or Biolegend), anti-CD45R/B220 was counted in a BetaPlate scintillation counter (Wallac, Gaithersburg, (Biolegend or eBioscience), anti-gp49 (eBioscience), anti-IgM (Biolegend, MD). These experiments were performed in triplicate. The Journal of Immunology 637

Preparation of 3T3 cells expressing integrin avb3 in any of these mice (37, 44, 45). Therefore, we analyzed B cell subsets in lymphoid organs by flow cytometry. The numbers of Fo The integrin b3 gene was retrovirally transduced into BALB/c 3T3 cells, which express only av chain intracellularly (data not shown). Mouse in- B and MZ B cells in the spleen (Fig. 2A) were equivalent between +/+ 2/2 tegrin b3 cDNA was cloned into pMX-IRES-ratCD2 vector (40). The gp49B and gp49B mice. The numbers of pro-B, pre-B, vector was transfected into PLAT-E packaging cell line (41) by using Lipo- immature B, and mature B cells in the were also fectamine 2000 (Invitrogen), and the supernatant was added to BALB/c normal in gp49B2/2 mice (Supplemental Fig. 1A). These data 3T3 fibroblasts (clone A31, provided by RIKEN BRC) in the presence of DOTAP Liposomal Transfection Reagent (Roche). indicate that early B cell development is largely normal in the absence of gp49B. iGB cell culture and iMB cell generation We next analyzed late B cell development during an immune iGB cell culture and iMB cell formation were performed as previously de- response. Two weeks after immunization with NP-CGG in alum, + scribed (42). In brief, purified B cells from CD45.1 mice were cultured for the number of NP-specific GC B cells determined by flow 4 d on irradiated 40LB cell feeder layer with rIL-4 (1 ng/ml; Peprotech). cytometry, and the size and frequency of GCs determined by Then the expanded B (iGB) cells were injected i.v. into g-irradiated (6.5 Gy) C57BL/6 mice. The iGB cell–derived (CD45.1+) B cells (iMB cells) were histological analysis were equivalent between the spleens of +/+ 2/2 sorted from pooled spleens of 10 recipient mice by FACSAria II. gp49B and gp49B mice (data not shown). Ninety days after the primary immunization, we did a secondary immunization. The Quantitative real-time PCR numbers of NP-specific IgG1+ memory B cells just before the RNA preparation, cDNA synthesis, and quantitative real-time PCR were secondary immunization, or after 5 d, did not differ significantly performed, and data were presented as described previously (42). Gene between gp49B+/+ and gp49B2/2 mice (Fig. 2B). To ask whether expression was normalized to that of gapdh. The following primer sets the lack of gp49B intrinsically affects memory B cell formation,

were used: gapdh sense: 59-GGAGAAACCTGCCAAGT ATGA-39; gapdh Downloaded from antisense: 59-CCCTGTTGCTGTAGCCGTATT-39; gp49A sense: 59-CAAG- we generated mixed bone marrow chimeric mice by transferring TTCAACATTCCAAGC-39; gp49A antisense: 59-CTATGGATGAGCTGCA- equal numbers of bone marrow cells of CD45.2+ gp49B2/2 and AC-39; gp49B sense: 59-ACCAAGTTCAAAATTCGATTT-39; gp49B congenic CD45.1+ gp49B+/+ mice into lethally irradiated mMT antisense: 59-GTGTGGATGAGCTGCACT-39; Blimp1 sense: 59-GA- ACCTGCTTTTCAAGTATGCTG-39; Blimp1 antisense: 59-AGTGTA- mice (Supplemental Fig. 1B). These mice were immunized with GACTTCACCGATGAGG-39. NP-CGG, and their spleen cells were analyzed 4 and 14 wk later. At both time points, the proportion of gp49B2/2 (CD45.2+) cells

Statistical analysis http://www.jimmunol.org/ in the NP-specific IgG1+ memory B cell fraction was comparable Statistical analysis was performed using the two-tailed unpaired Student t with that in the naive B cell fraction (Supplemental Fig. 1C). test. When the normality test failed, a Wilcoxon–Mann–Whitney U test These data indicate that memory B cells that lack gp49B develop was performed by using R software. and are maintained normally. Results Enhanced Ab production in gp49B2/2 mice during immune gp49B is selectively expressed on memory and MZ B cells responses To identify molecules selectively expressed on memory B cells, we Memory and MZ B cells share the common feature of respond- performed microarray analysis using in vivo–induced memory-like ingtoTDAgs morequicklythanFo B cells toproduceAbs.Given 2 2 by guest on October 2, 2021 B cells, termed iMB cells (42), and naive Fo B cells for comparison. that the development of memory and MZ B cells in gp49B / We have shown previously that the iMB cells are functionally mice is normal, we next examined Ab production in these mice equivalent to, but far more abundant than, bona-fide memory B cells during a TD immune response. After primary immunization with 2 2 (42). Among genes encoding cell-surface receptors that are highly NP-CGG/alum, gp49B / mice produced NP-specific IgM and preferentially expressed on iMB cells, we focused on gp49B, be- IgG1 at similar levels as gp49B+/+ mice on day 7, but had sig- cause we have also found it to be one of the most differentially nificantly increased levels of IgM on day 14 (Fig. 3A). Upon 2 2 expressed genes in memory B cells as compared with naive B, GC secondary challenge with NP-CGG on day 50, gp49B / mice B, and plasma cells in a published data set (43). We confirmed by produced significantly higher levels of NP-specific IgM and quantitative RT-PCR that the expression of gp49B transcripts was IgG1, with the latter extending over 2 wk after the challenge markedly higher in iMB cells than in Fo B cells, whereas the ex- (Fig. 3A). Because MZ B cells are the major source of plasma pression of gp49A did not significantly differ (Fig. 1A). Using an cells in response to T cell–independent type II (TI-II) Ag (8), we 2 2 available gp49 mAb (H1.1), which recognizes a common epitope on also tested TI-II immune response of gp49B / mice. After gp49A and gp49B (33), we found that gp49 protein was expressed primary immunization with model TI-II Ag, NP-Ficoll, NP- 2 2 on iMB cells at a higher level than naive B cells (Fig. 1B). specific IgM was significantly higher in gp49B / mice than We next examined the expression of gp49 on bona-fide memory in wild-type mice (Supplemental Fig. 2A). B cells, as well as other B cell subsets taken from mice immunized Because gp49B is also expressed on dendritic cells and activated with a typical T-dependent Ag, NP-CGG in alum. Expression of T cells, it was unclear at first whether the enhanced Ab production 2 2 gp49 was higher on NP-binding IgG1+ memory B cells than on GC by gp49B / mice is due to a B cell defect. Therefore, we gen- or naive B cells (Fig. 1C), whereas that on plasma cells was as low erated mixed bone marrow chimeric mice in which B cells were 2 2 as on naive B cells (Fig. 1D). We also found that, among spleen derived from gp49B+/+ or gp49B / mice, and the vast majority of B cells from unimmunized mice, MZ B cells expressed more other leukocytes were from gp49B+/+ mice. The mouse group in abundant gp49 than Fo B cells (Fig. 1E). Finally, we demonstrated which B cells lacked gp49B produced a significantly larger that gp49B is the major subtype expressed on MZ B cells and number of NP-specific IgM-secreting AFCs than the control memory B cells, because the gp49 staining was mostly lost on these mouse group in response to NP-CGG and NP-Ficoll (Fig. 3B, cells from gp49B2/2 mice (Fig. 1F). Supplemental Fig. 2B). This result indicates that B cell defect is responsible for the augmented primary IgM responses to TD and B cell development and memory B cell formation are normal in 2 2 TI-II Ags in gp49B / mice. gp49B2/2 mice To determine whether the enhanced memory-recall response Although three strains of gp49B-null knockout mice have been in gp49B2/2 mice is also due to a B cell–intrinsic defect, we published, so far profiles of B cell subsets have not been reported transferred splenic B cells of NP-primed gp49B+/+ or gp49B2/2 638 gp49B REGULATES MEMORY/MZ B CELL Ab PRODUCTION

FIGURE 1. gp49B is expressed on memory and MZ B cells. (A and B) iMB cells (CD19+ CD45.1+) were purified from spleens of recipient mice (C57BL/6, CD45.2+) transferred with iGB cells (42) 2 wk previously. (A) Real-time PCR analysis of gp49A and gp49B mRNA in Fo B cells sorted from C57BL/6 mice [as shown in (E)] and iMB cells sorted from recipient mice. Gene amplification was normalized to that of the control gapdh gene, and the gene expression in iMB cells is presented as to that of Fo B cells. (B) The expression of gp49 protein on the iMB and naive B cells from the recipient mice (gating shown in left panel) was analyzed by flow cytometry. (C) The expression of gp49 on naive, GC, and memory B cells from spleens of mice im- munized with NP-CGG 4 wk earlier. The following populations were gated as shown in the left three panels: naive B cells, B220+ CD19+ IgG12 NP2;GC B cells, B220+ CD19+ IgG1+ NP+ CD38low; and memory B cells, B220+ CD19+ IgG1+ NP+ CD38high.(D) The expression of gp49 on naive B and plasma Downloaded from cells from spleens of C57BL/6 mice immunized with NP-CGG/alum 5 d before the analysis (right panel). Gating strategy for the naive B cells (B220+ CD1382) and plasma cells (B220dull CD138+) is shown in the left panel.(E) The expression of gp49 on mature Fo and MZ B cells from spleens of naive C57BL/6 mice (right panel). Gating strategy for the Fo (B220+ CD21+ CD23high) and MZ B cells (B220+ CD21high CD23low) are shown in the left two panels.(F) The expression of gp49 on Fo, MZ, and memory B cells from gp49B+/+ (black line) and gp49B2/2 (shaded histogram) mice. The memory B cells were purified from mice immunized with NP-CGG/alum 4 wk before the analysis. All data are representative of two to five independent experiments. http://www.jimmunol.org/ mice into naive recipients together with CGG-primed wild-type B cells produced ∼10 times larger amounts of IgM than those spleen cells, as a source of memory T cells. The recipient mice harboring gp49B+/+ Fo B cells after 1 wk, on average, indicating were challenged with soluble NP-CGG on the next day. As shown that MZ B cells play a dominant role in the TD IgM response, as in Fig. 3C, the recipients of gp49B2/2 B cells produced larger previously demonstrated in a similar system (10). Strikingly, 2/2 +/+ amounts of Ag-specific IgG1 (anti-NP13.6 IgG1), the majority of gp49B MZ B cells were far more competent than gp49B MZ which was high affinity (anti-NP1.1 IgG1), than the recipients of B cells in the IgM production, whereas the difference between gp49B+/+ B cells. As a negative control, recipients of naive B cells gp49B2/2 and gp49B+/+ Fo B cells was not significant (Fig. 4A). To did not respond to the same soluble Ag. These data indicate that assess whether the augmented IgM production is due to increased gp49B expressed on memory B cells directly suppresses Ab pro- number of plasma cells, we transferred Fo or MZ B cells from by guest on October 2, 2021 duction by these cells upon Ag restimulation. gp49B+/+ or gp49B2/2, CD45.1 mice, both of which carried the ∼ 2/2 B1-8 VH gene knock-in allele and thus possessed 5% NP binding Enhanced Ab production by gp49B MZ and memory B cells (l+) B cells, into naive C57BL/6 mice (CD45.2). The recipient mice We then asked whether the augmented primary IgM response in were immunized with NP-CGG/alum on the next day, and spleen gp49B2/2 mice is attributable to the defect in MZ B cells. To cells were analyzed by flow cytometry 8 d later. gp49B2/2 MZ address this, we transferred MZ or Fo B cells from gp49B+/+ or B cells produced ∼2-fold more NP-binding cells and plasmablasts gp49B2/2 into mMT mice, and immunized the recipient mice with than gp49B+/+ MZ B cells. By contrast, the number of these cells NP-CGG/alum on the next day. The mice harboring gp49B+/+ MZ was equivalent when gp49B+/+ and gp49B2/2 Fo B cells were

FIGURE 2. Normal B cell development and memory B cell formation in gp49B2/2 mice. (A) The frequency (representative data, center panel) and the number (right panel) of Fo (B220+ CD21+ CD23high) and MZ (B220+ CD21high CD23low) B cells (gating shown in the left and center panels) in the spleens of unimmunized gp49+/+ and gp492/2 mice. n =4.(B) gp49B+/+ and gp49B2/2 mice were immunized with NP-CGG/alum and boosted with soluble NP-CGG 3 mo later. Gating strategy for memory B cells (B220+ Dump2 IgG1+ NP+ CD38+; Dump: CD138, FcεRI, Ter119) is shown by the left three panels.(Right panel) Numbers of memory B cells in the spleen of each mouse 3 mo after the primary immunization (preboost) and 5 d after the booster im- munization. n =3. The Journal of Immunology 639

FIGURE 3. Augmented Ab responses to TD Ag in gp49B2/2 mice. (A)gp49B+/+ and gp49B2/2 mice received a primary immunization with NP-CGG/alum and 50 d later, they were boosted with soluble NP-CGG (n = 6). Anti-NP IgM (left panels) and IgG1 (right panels) concentrations in the sera 7 or 14 d after the Downloaded from primary (top panels) and the secondary (bottom panels) immunizations were determined by ELISA. (B) Pooled BM cells from gp49B+/+ or gp49B2/2 mice and mMT mice, mixed at a ratio of 1:4, or BM cells of mMT mice alone, were transferred (5 3 106 cells/mouse) i.v. into lethally irradiated C57BL/6 mice. Ten weeks later, chimeric mice were immunized with NP-CGG/alum (left panel). Seven days after the immunization, the numbers of anti-NP IgM AFCs in the spleens of the mice were determined by ELISPOT (right panel). n =4(mMT alone: n = 3). (C) Purified B cells (1 3 107) from pooled spleens of NP-CGG– primed or unimmunized gp49B+/+ or gp49B2/2 mice (five for each group), and pooled spleen cells (1 3 107) from three C57BL/6 mice primed with 100 mg CGG/alum 30 d earlier were cotransferred i.v. into C57BL/6 mice (primed B cells: n =5;naiveBcells:n = 3). One day after the transfer, the recipient mice http://www.jimmunol.org/ were immunized with soluble NP-CGG (left panel). Concentrations of anti-NP13.6 IgG1 (center panel) and anti-NP1.1 IgG1 (right panel) Abs in the sera 7 d after immunization were determined by ELISA. *p , 0.05, **p , 0.01. Data are representative of two to three independent experiments. transferred (Supplemental Fig. 3A). These data suggest that MZ estimated by ELISA. gp49B2/2 MZ B cells produced nearly B cells, but not Fo B cells, are responsible for the augmented IgM 3-fold more IgM than gp49B+/+ cells after stimulation with anti- production upon primary immunization in gp49B2/2 mice. CD40 Ab, but not significantly more upon LPS stimulation Next, we assessed whether the lack of gp49B directly affected (Fig. 4B). After 3 d of culture with anti-CD40 Ab, gp49B2/2 MZ MZ B cells in vitro. Sorted MZ and Fo B cells were stimulated 4 d B cells generated more plasma cells than gp49B+/+ cells, as

with anti-CD40 Ab or LPS, and IgM in the culture supernatants was assessed by flow cytometry and ELISPOT assays (Fig. 4C, 4D). by guest on October 2, 2021

FIGURE 4. gp49B suppresses Ab production by MZ and memory B cells. (A) Fo or MZ B cells sorted from pooled spleens of gp49B+/+ or gp49B2/2 mice were transferred i.v. into mMT mice (2 3 106 cells/mouse, left panel), and the recipients (n = 6) were immunized with NP-CGG/alum on the next day. IgM titers in the sera 7 or 14 d after immunizations were determined by ELISA (center and right panels). (B–E) MZ B cells sorted from pooled spleens of gp49B+/+ or gp49B2/2 mice were unstimulated (2) or stimulated with anti-CD40, anti-IgM Abs, or LPS, as indicated. IgM in culture supernatants was detected by ELISA on day 4 (B). The frequency of plasma cells (CD138+) in the cultures was analyzed by flow cytometry on day 2 and 3 (C). The numbers of IgM-secreting AFCs in the cultures were estimated by ELISPOT on day 3 (D). [3H]thymidine incorporation was determined on day 2 (E). (F) gp49B+/+ or gp49B2/2 NP-binding cells (IgG1+ memory, Fo, and MZ B cells) were purified as shown in Supplemental Fig. 3C, 3D. Sorted cells (10,000 cells/well) were stimulated with anti-CD40 Ab for 7 d in vitro, and the concentration of NP-specific IgG1 in the culture supernatant was determined by ELISA. *p , 0.05, **p , 0.01. All cultures were performed in triplicate. Data are representative of two to three independent experiments. n.s., not significant. 640 gp49B REGULATES MEMORY/MZ B CELL Ab PRODUCTION

However, proliferation upon stimulation with anti-CD40 Ab, CD40. To obtain the B cells expressing both gp49B and integrin avb3, anti-IgM, or LPS did not significantly differ between gp49B+/+ we cultured splenic B cells on a feeder layer of 40LB cells that and gp49B2/2 MZ B cells (Fig. 4E). In contrast, gp49B2/2 and express exogenous CD40L and BAFF (42) (Fig. 6A). Then these cells gp49B+/+ Fo B cells produced equivalent levels of IgM after were stimulated with anti-CD40 after starvation. Although the phos- stimulation with anti-CD40 Ab or LPS (Supplemental Fig. 3B). phorylation of Akt was comparable in gp49B+/+ and gp49B2/2 cells, We then examined the effect of gp49B on Ab production by mem- the levels of the induced Erk phosphorylation were higher in ory B cells in vitro. The mice that were transferred with gp49B+/+ or gp49B2/2 cells than in gp49B+/+ cells (Fig. 6B). In addition, the gp49B2/2, B1-8 ki B cells were immunized with NP-CGG, and expression of blimp1, the master regulator of plasma cell develop- donor-derived NP-specific IgG1+ memory B cells were sorted at 4 wk ment (51) known to be induced by ERK, was more strongly induced after the immunization, to obtain a sufficient number of memory by anti-CD40 stimulation in gp49B2/2 cells than in gp49B+/+ cells B cells (Supplemental Fig. 3C). The memory B cells, as well as naive (Fig. 6D). In contrast, the levels of Erk phosphorylation and blimp1 NP-specific Fo and MZ B cells (Supplemental Fig. 3D), were stim- expression were equivalent between gp49B+/+ and gp49B2/2 Bcells ulated in vitro with anti-CD40 Ab for 7 d. As shown in Fig. 4F, after stimulation with anti-IgM (Fig. 6C, 6D). Collectively, our results gp49B2/2 memory B cells produced ∼3-fold more NP-specific IgG1 suggest that gp49B attenuates a CD40 signaling pathway leading to than gp49B+/+ cells, whereas Fo and MZ B cells did not produce Erk activation and the following plasma cell differentiation. detectable amounts of NP-specific IgG1 under the same conditions. 2 2 Enhanced IgE production by gp49B / mice upon secondary Taken together, these data strongly suggest that gp49B expressed immunization on memory and MZ B cells negatively regulates the differentiation of + these cells into plasma cells that produce Abs in the immune response. It has been shown that IgG1 memory B cells are able to undergo secondary class switching to IgE and to generate a high-affinity IgE Downloaded from gp49B suppresses MZ B cell Ab production through interaction response upon repeated immunization (52). Alternatively, a recent a b with integrin v 3 study suggested that IgE+ memory B cells are generated through It was previously shown that gp49B interacts with the avb3 integ- a germinal center pathway and produce IgE in the recall response rin heterodimer, and thereby suppresses the degranulation of mast (53). Whatever the case, we considered the possibility that the ex- 2/2 cells (30). We found that the expression of integrin avb3 was in- aggerated memory B cell response in gp49B mice might also duced on both Fo and MZ B cells after stimulation with anti-CD40 result in a deregulated IgE recall response. To test this possibility, http://www.jimmunol.org/ Ab (Fig. 5A), suggesting that the gp49B-mediated suppression of we immunized gp49B+/+ or gp49B2/2 mice with NP-CGG s.c. and Ab production by MZ B cells in vitro might be caused by the i.p., routes known to induce an IgE response (54). As compared +/+ gp49B and integrin avb3 interaction. Thus, we cultured MZ with gp49B mice, NP-specific IgE production was slightly ele- B cells with anti-CD40 Ab on a feeder layer of BALB/c 3T3 cells vated in gp49B2/2 mice after the primary immunization, but expressing integrin avb3 or not, and measured IgM produced in markedly so after the secondary immunization (Fig. 7A). Pre- the culture supernatant (Fig. 5B). As shown in Fig. 5C, Ab pro- viously it was reported that the Th2-mediated IgE response to duction by gp49B+/+ MZ B cells was significantly suppressed on OVA induced by intranasal priming with LPS and OVA is aug- 2/2 the feeder cells expressing integrin avb3 as compared with control mented in gp49B mice, and this was suggested to be likely due feeder cells. By contrast, gp49B2/2 MZ B cells were not sup- to a dendritic cell defect (55). To clarify whether the augmented IgE by guest on October 2, 2021 2/2 pressed by the integrin avb3-expressing feeder cells. This result production in gp49B mice in our system was attributable to indicates that gp49B suppresses Ab production by MZ B cells memory B cells, we generated mixed bone marrow chimeric mice +/+ 2/2 through its interaction with integrin avb3. harboring gp49B or gp49B B cells, with the vast majority of other hematopoietic cells intact. Upon immunization by the same gp49B attenuates CD40-mediated ERK activation and blimp1 s.c./i.p. protocol, the mice with gp49B2/2 B cells produced sig- expression nificantly elevated levels of NP-specific IgE compared with the CD40 signaling is known to promote proliferation, Ab production, control mouse group in the recall response upon a challenge 6 wk and survival of B cells (46–48). Earlier mentioned results suggest after the primary immunization (Fig. 7B). Accordingly, when these that gp49B may specifically suppress a CD40 signaling pathway mice were induced an active anaphylaxis reaction by i.v. injection that promotes B cell differentiation to plasma cells. CD40 sig- of NP-CGG as the tertiary immunization, the mice with gp49B2/2 naling induces the activation of ERK and PI3K signaling, which B cells showed severer hypothermia than those with gp49B+/+ are also important for the plasma cell differentiation (49, 50). B cells (Fig. 7C). Notably, two of the mice with gp49B2/2 Bcells Therefore, we compared the phosphorylation of ERK and AKT on died during this response, whereas those with gp49B+/+ B cells were gp49B+/+ and gp49B2/2 B cells after the stimulation with anti- all alive. These results indicate that gp49B suppresses excessive

FIGURE 5. gp49B suppresses Ab production upon interaction with integrin avb3.(A) Fo and MZ B cells were stimulated with anti-CD40 Ab, or unstimulated (2) for 48 h, and the expression of integrin av and b3 was analyzed by flow cytometry. Data are representative of two independent experiments. (B) The integrin b3 cDNA was retrovirally transduced into BALB/c 3T3 cells. The expression of integrin av and b3 on the surface of these +/+ 2/2 cells (integrin b3) and the parental cells (3T3) was analyzed by flow cytometry. (C) MZ B cells sorted from pooled spleens of gp49B and gp49B mice were cultured with anti-CD40 Ab or without (2), on the BALB/c 3T3 transduced with integrin b3 or mock vector for 4 d. Relative titers of IgM in the culture supernatant were determined by ELISA. n =3.*p , 0.05. Data are representative of two independent experiments. n.s., not significant. The Journal of Immunology 641

FIGURE 6. gp49B suppresses CD40-mediated ERK activation and blimp1 induction. (A) Splenic B cells of gp49B+/+ and gp49B2/2 mice were cultured on 40LB cells for 3 d. Then feeder cells were depleted, and the remaining B cells were incubated in medium overnight. The expressions of gp49 and 2/2 integrin avb3 after incubation were analyzed by flow cytometry. Control: gp49B -derived B cells stained with isotype-matched control Abs. (B–D) The cells prepared as in (A) were stimulated with anti-CD40 (5 mg/ml) or anti-IgM (5 mg/ml) Abs for the indicated time periods. The cell lysates from the anti- CD40– (B) or anti-IgM–stimulated (C) cells were analyzed by Western blotting to detect phosphorylated and total ERK1/2 and Akt proteins, as indicated.

RNA samples from the cells stimulated for 8 h, or unstimulated (–), were subjected to real-time PCR analysis. Blimp1 mRNA amplification was normalized Downloaded from to that of the control gapdh mRNA, and the normalized values of gp49B2/2 cells are presented as relatives to those of gp49B+/+ cells (D). Data are representative of two independent experiments.

IgE production by memory B cells in the recall response, and compared with the IgM BCR on Fo B cells, the IgG BCR on IgG+ contribute to prevention of acquired allergic disease. memory B cells transduces stronger signals and induces robust

proliferation, cellular activation, and Ab production (57–59). MZ http://www.jimmunol.org/ Discussion B cells have increased basal signaling activity, and thus lower Memory B cells express higher levels of B7-1 and B7-2, and re- threshold for activation, compared with Fo B cells (9, 12, 13). In spond to Ags more quickly, than naive B cells (7, 56). In addition, contrast, GC B cells have inactive BCR signaling caused by in- by guest on October 2, 2021

FIGURE 7. Exaggerated IgE production in gp49B2/2 mice in the memory recall response. (A) gp49B+/+ and gp49B2/2 mice received a primary immunization with NP-CGG/alum s.c., and were boosted 3 wk later with soluble NP-CGG s.c. and i.p. as detailed in Materials and Methods. Serum anti-NP IgE concentrations on the indicated days after immunization were determined by ELISA. n = 9. Data are representative of two inde- pendent experiments. (B) BM chimeric mice containing gp49B+/+ or gp49B2/2 B cells were generated as shown in Fig. 3B. Ten weeks later, these chimeric mice were immunized and then boosted 6 wk after that with NP- CGG, as in (A). Serum anti-NP IgE concentrations were determined as in (A). n =9.(C) The mice shown in (B) were injected i.v. with soluble NP-CGG 5 mo after the booster immunization, and rectal temperature of these mice was measured at the indicated time after the tertiary immunization. Two gp49B2/2 mice died as indicated by dagger (†). * p , 0.05, **p , 0.01. 642 gp49B REGULATES MEMORY/MZ B CELL Ab PRODUCTION creased phosphatase activity and are not activated but rather un- B cells in the immune response (49). Therefore, gp49B is likely to dergo apoptosis upon BCR stimulation in vivo (60–62). Because regulate the plasma cell differentiation by suppressing ERK of their elevated responsiveness with lower activation thresholds, activation. Because ERK phosphorylation was not affected in memory and MZ B cells might need a strong regulatory system gp49B2/2 B cells when stimulated with anti-IgM Abs, it is sug- distinct from that of naive Fo or GC B cells. In this report, we gested that gp49B regulates ERK activation specifically signaled demonstrated that gp49B is expressed at significantly higher levels from CD40. on memory and MZ B cells compared with Fo or GC B cells, and Although CD40 is generally considered to be involved in TD regulates plasma cell differentiation and Ab production of mem- immune response, our data indicated that gp49B also attenuates ory and MZ B cells during TD immune responses. Thus, such IgM production from B cells during TI-II immune responses. Be- “activation-prone” cells among B-lineage cells appear to be spe- cause MZ B cells play a major role in the TI IgM response (8), it is cially equipped with the gp49B negative regulator. plausible that gp49B negatively regulates Ab production by MZ The number of MZ B and other B-lineage cells was normal in B cells. Recent study suggests that localizing at the gp49B2/2 mice, as was the number of memory B cells after pri- peri-MZ induce the Ab production and class switching of MZ mary and secondary immunization. In addition, gp49B2/2 memory B cells by providing BAFF, APRIL, and IL-21 (63). Because these B cells were normally developed in early and late phases of the neutrophils also express CD40L, they may activate MZ B cells via immune response in mixed BM chimeric mice. Thus, gp49B is not CD40L. Thus, gp49B may suppress the signaling from CD40 on involved in the development or the maintenance of memory and MZ B cells in TI immune response, as well as in TD immune MZ B cells. In contrast, the production of IgM was increased in response. gp49B2/2 mice after primary immunization. It is known that MZ We demonstrated that gp49B transduces an inhibitory signal in Downloaded from B cells dominantly contribute to primary IgM production during MZ B cells upon interaction with integrin avb3, as previously TD immune responses (10), which was confirmed by our B cell shown for mast-cell degranulation (30). Although gp49B has no transfer experiments (Fig. 4A). We demonstrated that gp49B2/2 RGD sequence, which commonly binds to integrins, it has been MZ B cells, but not Fo B cells, produce more IgM Ab than gp49B+/+ suggested to interact with integrin through Leu-Asp-Ser-Gln cells upon immunization, suggesting that gp49B negatively regulates (30), a sequence that conforms to the integrin-binding consensus Ab production by MZ B cells in the primary response. It is of note motif, aliphatic-Asp/Glu-Ser/Thr-Pro/hydrophilic, found in other http://www.jimmunol.org/ that the enhanced IgM response was evident on day 7 after primary Ig superfamily members (64). Integrin avb3 is expressed on var- immunization with either TD or TI-II Ags in the bone marrow ious cell types, and its interaction with ligands is tightly controlled chimeras where gp49B was largely intact in non-B cells, whereas it by its adhesion capacity (30), which is typically upregulated after was not on day 7, but only after 2 wk in gp49B2/2 mice (Fig. 3A cellular activation (65). Therefore, gp49B-mediated suppression is and 3B and Supplemental Fig. 2A, 2B). Further study is necessary presumably regulated redundantly by its expression, as well as the to solve this apparent inconsistency, but it would be possible to activation state of the avb3 integrin on the cells with which the explain that the early IgM response is negatively regulated by some B cells interact. Considering the additional involvement of various non-B cells, the function of which is attenuated by gp49B. cytokines, among others, the regulation of Ab production by MZ IgG1 titers were equivalent in gp49B+/+ and gp49B2/2 mice in and memory B cells may be more complex in a physiological by guest on October 2, 2021 the primary response, but higher in gp49B2/2 mice in the sec- context than has generally been appreciated. ondary memory response. In addition, memory B cells derived In this study, we have identified, for the first time as far as we are from gp49B2/2 mice produced higher levels of IgG1 Ab after aware, an inhibitory receptor that is expressed and functions on adoptive transfer into recipient mice upon immunization with memory B cells. What would be the physiological significance soluble Ag. Higher production of IgM in gp49B2/2 mice in the of gp49B-mediated inhibition of recall Ab response by memory secondary response against soluble Ag is likely due to IgM+ B cells? Peritoneal B1 B cells, which are typically weakly self- memory B cells. These data indicate that gp49B also regulates the reactive, are known to produce autoantibodies in vivo in the ab- recall response of memory B cells. sence of inhibitory receptors such as PIR-B or Siglec-G (21, 22). We also demonstrated that upon stimulation by CD40 in It was reported that, in humans, about one third of circulating IgG+ vitro, gp49B2/2 MZ and memory B, but not Fo B, cells gener- memory B cells, but much fewer mature naive B cells from ated more Ab-producing plasma cells than gp49B+/+ cells, despite healthy donors, express self-reactive BCR encoded by somatically their equivalent proliferation. These results suggest that gp49B hypermutated V genes (66). As in the case of B1 cells, negative regulates differentiation of MZ and memory B cells into plasma regulation of Ab production by such self-reactive memory B cells cells induced by CD40 stimulation during immune responses. would be important to prevent unintentional activation of such CD40 signaling is known to promote proliferation, Ab produc- cells and the resultant autoimmunity. tion, and survival of B cells (46–48). Thus, gp49B may regulate Another possibility is that gp49B suppresses IgE production a specific pathway downstream of CD40, namely, a pathway in the recall response, as indicated in this study. The enhanced promoting B cell differentiation to plasma cells. gp49B contains IgE produced from memory B cells after secondary immuniza- two ITIMs in its cytoplasmic tail and has been shown to inhibit tion resulted in exacerbated active anaphylaxis in gp49B2/2 mice, cellular activation by recruiting SHP-1 and SHP-2 in mast cells perhaps through its augmented deposition on mast cells. Recent (26–29). A similar mechanism may regulate the CD40 signaling reports indicated that both IgE and IgG1 memory B cells might pathway in MZ and memory B cells. contribute to the production of IgE in the recall response (52, 53, Although CD40 stimulation is known to activate the NF-kB 67). Thus, negative regulation of the production of Ag-specific IgE pathway, the amount of p52/p100 and the degradation of IkBa from the memory B cell pool could prevent immediate hyper- was equivalent between gp49B+/+ and gp49B2/2 B cells after the sensitivity diseases such as asthma and allergy. stimulation with anti-CD40 (data not shown). Instead, the phos- phorylation of ERK and the expression of blimp1 were enhanced Acknowledgments 2/2 in gp49B B cells. ERK signaling is essential for the induc- We thank Drs. I. Shiratori (NEC Soft, Ltd.) for discussion and technical tion of Blimp1 by stimulation with anti-CD40, and ERK signaling advice, T. Azuma (Research Institute for Biomedical Sciences) for reagents plays a critical role for the plasma cell differentiation of memory and advice on immunization, K. Rajewsky (Max-Delbruck-Center€ for The Journal of Immunology 643

Molecular Medicine) and M. Maruyama (National Institute for Longevity inhibitory receptors of the immunoglobulin superfamily. Immunogenetics 50: Sciences) for B1-8 ki mice, Y. Kawai (Research Institute for Biomedical 286–294. 26. Katz, H. R., E. Vivier, M. C. Castells, M. J. McCormick, J. M. Chambers, and Sciences) for mouse care, and P. D. Burrows for critical comments on the K. F. Austen. 1996. Mouse mast cell gp49B1 contains two immunoreceptor manuscript. tyrosine-based inhibition motifs and suppresses mast cell activation when coli- gated with the high-affinity Fc receptor for IgE. Proc. Natl. Acad. Sci. USA 93: 10809–10814. Disclosures 27. Kuroiwa, A., Y. Yamashita, M. Inui, T. Yuasa, M. Ono, A. Nagabukuro, The authors have no financial conflicts of interest. Y. Matsuda, and T. Takai. 1998. Association of tyrosine phosphatases SHP-1 and SHP-2, inositol 5-phosphatase SHIP with gp49B1, and chromosomal assignment of the gene. J. Biol. Chem. 273: 1070–1074. 28. Wang, L. L., J. Blasioli, D. R. Plas, M. L. Thomas, and W. M. Yokoyama. 1999. References Specificity of the SH2 domains of SHP-1 in the interaction with the immunor- 1. Pereira, J. P., L. M. Kelly, and J. G. Cyster. 2010. Finding the right niche: B-cell eceptor tyrosine-based inhibitory motif-bearing receptor gp49B. J. Immunol. migration in the early phases of T-dependent antibody responses. Int. Immunol. 162: 1318–1323. 22: 413–419. 29. Lu-Kuo, J. M., D. M. Joyal, K. F. Austen, and H. R. Katz. 1999. gp49B1 inhibits IgE- 2. Tarlinton, D., and K. Good-Jacobson. 2013. Diversity among memory B cells: initiated mast cell activation through both immunoreceptor tyrosine-based inhibitory origin, consequences, and utility. Science 341: 1205–1211. motifs, recruitment of src homology 2 domain-containing phosphatase-1, and sup- 3. Toyama, H., S. Okada, M. Hatano, Y. Takahashi, N. Takeda, H. Ichii, pression of early and late calcium mobilization. J. Biol. Chem. 274: 5791–5796. T. Takemori, Y. Kuroda, and T. Tokuhisa. 2002. Memory B cells without somatic 30. Castells, M. C., L. B. Klickstein, K. Hassani, J. A. Cumplido, M. E. Lacouture, hypermutation are generated from Bcl6-deficient B cells. Immunity 17: 329–339. K. F. Austen, and H. R. Katz. 2001. gp49B1-alpha(v)beta3 interaction inhibits 4. Dogan, I., B. Bertocci, V. Vilmont, F. Delbos, J. Me´gret, S. Storck, -induced mast cell activation. Nat. Immunol. 2: 436–442. C. A. Reynaud, and J. C. Weill. 2009. Multiple layers of B cell memory with 31. Moulder, K., K. Roberts, E. M. Shevach, and J. E. Coligan. 1991. The mouse different effector functions. Nat. Immunol. 10: 1292–1299. vitronectin receptor is a T cell activation antigen. J. Exp. Med. 173: 343–347. 5. Pape, K. A., J. J. Taylor, R. W. Maul, P. J. Gearhart, and M. K. Jenkins. 2011. 32. Rojo, S., D. N. Burshtyn, E. O. Long, and N. Wagtmann. 1997. Type I trans- Different B cell populations mediate early and late memory during an endoge- membrane receptor with inhibitory function in mouse mast cells and NK cells. J. Downloaded from nous immune response. Science 331: 1203–1207. Immunol. 158: 9–12. 6. Ridderstad, A., and D. M. Tarlinton. 1998. Kinetics of establishing the memory 33. Wang, L. L., D. T. Chu, A. O. Dokun, and W. M. Yokoyama. 2000. Inducible B cell population as revealed by CD38 expression. J. Immunol. 160: 4688–4695. expression of the gp49B inhibitory receptor on NK cells. J. Immunol. 164: 5215– 7. Anderson, S. M., M. M. Tomayko, A. Ahuja, A. M. Haberman, and 5220. M. J. Shlomchik. 2007. New markers for murine memory B cells that define 34. Gu, X., A. Laouar, J. Wan, M. Daheshia, J. Lieberman, W. M. Yokoyama, mutated and unmutated subsets. J. Exp. Med. 204: 2103–2114. H. R. Katz, and N. Manjunath. 2003. The gp49B1 inhibitory receptor regulates 8. Martin, F., A. M. Oliver, and J. F. Kearney. 2001. Marginal zone and B1 B cells the IFN-gamma responses of T cells and NK cells. J. Immunol. 170: 4095–4101. unite in the early response against T-independent blood-borne particulate anti- 35. Lam, K. P., R. Kuhn,€ and K. Rajewsky. 1997. In vivo ablation of surface im- gens. Immunity 14: 617–629. munoglobulin on mature B cells by inducible gene targeting results in rapid cell http://www.jimmunol.org/ 9. Oliver, A. M., F. Martin, and J. F. Kearney. 1999. IgMhighCD21high death. Cell 90: 1073–1083. enriched in the splenic marginal zone generate effector cells more rapidly than the 36. Sonoda, E., Y. Pewzner-Jung, S. Schwers, S. Taki, S. Jung, D. Eilat, and bulk of follicular B cells. J. Immunol. 162: 7198–7207. K. Rajewsky. 1997. B cell development under the condition of allelic inclusion. 10. Song, H., and J. Cerny. 2003. Functional heterogeneity of marginal zone B cells Immunity 6: 225–233. revealed by their ability to generate both early antibody-forming cells and ger- 37. Kasai, S., M. Inui, K. Nakamura, Y. Kakizaki, S. Endo, A. Nakamura, S. Ito, and minal centers with hypermutation and memory in response to a T-dependent T. Takai. 2008. A novel regulatory role of gp49B on dendritic cells in T-cell antigen. J. Exp. Med. 198: 1923–1935. priming. Eur. J. Immunol. 38: 2426–2437. 11. Phan, T. G., S. Gardam, A. Basten, and R. Brink. 2005. Altered migration, re- 38. Kitamura, D., and K. Rajewsky. 1992. Targeted disruption of mu chain mem- cruitment, and somatic hypermutation in the early response of marginal zone brane exon causes loss of heavy-chain allelic exclusion. Nature 356: 154–156. B cells to T cell-dependent antigen. J. Immunol. 174: 4567–4578. 39. Haniuda, K., T. Nojima, K. Ohyama, and D. Kitamura. 2011. Tolerance induc- 12. Hampel, F., S. Ehrenberg, C. Hojer, A. Draeseke, G. Marschall-Schro¨ter, tion of IgG+ memory B cells by T cell-independent type II . J. Immunol. R. Kuhn,€ B. Mack, O. Gires, C. J. Vahl, M. Schmidt-Supprian, et al. 2011. 186: 5620–5628. by guest on October 2, 2021 CD19-independent instruction of murine marginal zone B-cell development by 40. Yamamoto, M., K. Hayashi, T. Nojima, Y. Matsuzaki, Y. Kawano, constitutive Notch2 signaling. Blood 118: 6321–6331. H. Karasuyama, R. Goitsuka, and D. Kitamura. 2006. BASH-novel PKC-Raf-1 13. Meyer-Bahlburg, A., A. D. Bandaranayake, S. F. Andrews, and D. J. Rawlings. pathway of pre-BCR signaling induces kappa gene rearrangement. Blood 108: 2009. Reduced c-myc expression levels limit follicular mature B cell cycling in 2703–2711. response to TLR signals. J. Immunol. 182: 4065–4075. 41. Nosaka, T., T. Kawashima, K. Misawa, K. Ikuta, A. L. Mui, and T. Kitamura. 14. Martin, F., and J. F. Kearney. 2000. B-cell subsets and the mature preimmune 1999. STAT5 as a molecular regulator of proliferation, differentiation and apo- repertoire. Marginal zone and B1 B cells as part of a “natural immune memory”. ptosis in hematopoietic cells. EMBO J. 18: 4754–4765. Immunol. Rev. 175: 70–79. 42. Nojima, T., K. Haniuda, T. Moutai, M. Matsudaira, S. Mizokawa, I. Shiratori, 15. Hao, Z., and K. Rajewsky. 2001. Homeostasis of peripheral B cells in the ab- T. Azuma, and D. Kitamura. 2011. In-vitro derived germinal centre B cells sence of B cell influx from the bone marrow. J. Exp. Med. 194: 1151–1164. differentially generate memory B or plasma cells in vivo. Nat. Commun. 2: 16. Wardemann, H., S. Yurasov, A. Schaefer, J. W. Young, E. Meffre, and 465. M. C. Nussenzweig. 2003. Predominant autoantibody production by early human 43. Bhattacharya, D., M. T. Cheah, C. B. Franco, N. Hosen, C. L. Pin, W. C. Sha, B cell precursors. Science 301: 1374–1377. and I. L. Weissman. 2007. Transcriptional profiling of antigen-dependent murine 17. Jellusova, J., and L. Nitschke. 2011. Regulation of B cell functions by the sialic B cell differentiation and memory formation. J. Immunol. 179: 6808–6819. acid-binding receptors siglec-G and CD22. Front. Immunol. 2: 96. 44. Rojo, S., C. C. Stebbins, M. E. Peterson, D. Dombrowicz, N. Wagtmann, and 18. Ujike, A., K. Takeda, A. Nakamura, S. Ebihara, K. Akiyama, and T. Takai. 2002. E. O. Long. 2000. Natural killer cells and mast cells from gp49B null mutant -/- Impaired dendritic cell maturation and increased TH2 responses in PIR-B mice. mice are functional. Mol. Cell. Biol. 20: 7178–7182. Nat. Immunol. 3: 542–548. 45. Daheshia, M., D. S. Friend, M. J. Grusby, K. F. Austen, and H. R. Katz. 2001. 19. O’Keefe, T. L., G. T. Williams, S. L. Davies, and M. S. Neuberger. 1996. Hy- Increased severity of local and systemic anaphylactic reactions in gp49B1- perresponsive B cells in CD22-deficient mice. Science 274: 798–801. deficient mice. J. Exp. Med. 194: 227–234. 20. Nitschke, L., R. Carsetti, B. Ocker, G. Ko¨hler, and M. C. Lamers. 1997. CD22 is 46. Armitage, R. J., W. C. Fanslow, L. Strockbine, T. A. Sato, K. N. Clifford, a negative regulator of B-cell receptor signalling. Curr. Biol. 7: 133–143. B. M. Macduff, D. M. Anderson, S. D. Gimpel, T. Davis-Smith, C. R. Maliszewski, 21. Kubo, T., Y. Uchida, Y. Watanabe, M. Abe, A. Nakamura, M. Ono, S. Akira, and et al. 1992. Molecular and biological characterization of a murine ligand for CD40. T. Takai. 2009. Augmented TLR9-induced Btk activation in PIR-B-deficient B-1 Nature 357: 80–82. cells provokes excessive autoantibody production and autoimmunity. J. Exp. 47. Kehry, M. R. 1996. CD40-mediated signaling in B cells. Balancing cell survival, Med. 206: 1971–1982. growth, and death. J. Immunol. 156: 2345–2348. 22. Hoffmann, A., S. Kerr, J. Jellusova, J. Zhang, F. Weisel, U. Wellmann, 48. Foy, T. M., D. M. Shepherd, F. H. Durie, A. Aruffo, J. A. Ledbetter, and T. H. Winkler, B. Kneitz, P. R. Crocker, and L. Nitschke. 2007. Siglec-G is a B1 R. J. Noelle. 1993. In vivo CD40-gp39 interactions are essential for - cell-inhibitory receptor that controls expansion and calcium signaling of the B1 dependent . II. Prolonged suppression of the humoral immune cell population. Nat. Immunol. 8: 695–704. response by an antibody to the ligand for CD40, gp39. J. Exp. Med. 178: 1567– 23. Samardzic, T., D. Marinkovic, C. P. Danzer, J. Gerlach, L. Nitschke, and 1575. T. Wirth. 2002. Reduction of marginal zone B cells in CD22-deficient mice. Eur. 49. Yasuda, T., K. Kometani, N. Takahashi, Y. Imai, Y. Aiba, and T. Kurosaki. 2011. J. Immunol. 32: 561–567. ERKs induce expression of the transcriptional repressor Blimp-1 and subsequent 24. Castells, M. C., X. Wu, J. P. Arm, K. F. Austen, and H. R. Katz. 1994. Cloning of plasma cell differentiation. Sci. Signal. 4: ra25. the gp49B gene of the immunoglobulin superfamily and demonstration that one 50. Omori, S. A., M. H. Cato, A. Anzelon-Mills, K. D. Puri, M. Shapiro-Shelef, of its two products is an early-expressed mast cell surface protein originally K. Calame, and R. C. Rickert. 2006. Regulation of class-switch recombination described as gp49. J. Biol. Chem. 269: 8393–8401. and plasma cell differentiation by phosphatidylinositol 3-kinase signaling. Im- 25. McCormick, M. J., M. C. Castells, K. F. Austen, and H. R. Katz. 1999. The munity 25: 545–557. gp49A gene has extensive sequence conservation with the gp49B gene and 51. Shaffer, A. L., K. I. Lin, T. C. Kuo, X. Yu, E. M. Hurt, A. Rosenwald, provides gp49A protein, a unique member of a large family of activating and J. M. Giltnane, L. Yang, H. Zhao, K. Calame, and L. M. Staudt. 2002. Blimp-1 644 gp49B REGULATES MEMORY/MZ B CELL Ab PRODUCTION

orchestrates plasma cell differentiation by extinguishing the mature B cell gene inhibited by CD22 and promotes the development of B cells whose survival is expression program. Immunity 17: 51–62. less dependent on Ig alpha/beta. J. Exp. Med. 204: 747–758. 52. Erazo, A., N. Kutchukhidze, M. Leung, A. P. Christ, J. F. Urban, Jr., 60. Khalil, A. M., J. C. Cambier, and M. J. Shlomchik. 2012. B cell receptor signal M. A. Curotto de Lafaille, and J. J. Lafaille. 2007. Unique maturation program of transduction in the GC is short-circuited by high phosphatase activity. Science the IgE response in vivo. Immunity 26: 191–203. 336: 1178–1181. 53. Talay, O., D. Yan, H. D. Brightbill, E. E. Straney, M. Zhou, E. Ladi, W. P. Lee, 61. Shokat, K. M., and C. C. Goodnow. 1995. Antigen-induced B-cell death and J. G. Egen, C. D. Austin, M. Xu, and L. C. Wu. 2012. IgE⁺ memory B cells and elimination during germinal-centre immune responses. Nature 375: 334–338. plasma cells generated through a germinal-center pathway. Nat. Immunol. 13: 62. Pulendran, B., G. Kannourakis, S. Nouri, K. G. Smith, and G. J. Nossal. 1995. 396–404. 54. Yang, Z., B. M. Sullivan, and C. D. Allen. 2012. Fluorescent in vivo detection Soluble antigen can cause enhanced apoptosis of germinal-centre B cells. Nature reveals that IgE(+) B cells are restrained by an intrinsic cell fate predisposition. 375: 331–334. Immunity 36: 857–872. 63. Puga, I., M. Cols, C. M. Barra, B. He, L. Cassis, M. Gentile, L. Comerma, 55. Breslow, R. G., J. J. Rao, W. Xing, D. I. Hong, N. A. Barrett, and H. R. Katz. A. Chorny, M. Shan, W. Xu, et al. 2011. B cell-helper neutrophils stimulate the 2010. Inhibition of Th2 adaptive immune responses and pulmonary inflamma- diversification and production of immunoglobulin in the marginal zone of the tion by leukocyte Ig-like receptor B4 on dendritic cells. J. Immunol. 184: 1003– spleen. Nat. Immunol. 13: 170–180. 1013. 64. Clements, J. M., P. Newham, M. Shepherd, R. Gilbert, T. J. Dudgeon, 56. Kurosaki, T., Y. Aiba, K. Kometani, S. Moriyama, and Y. Takahashi. 2010. L. A. Needham, R. M. Edwards, L. Berry, A. Brass, and M. J. Humphries. 1994. Unique properties of memory B cells of different isotypes. Immunol. Rev. 237: Identification of a key integrin-binding sequence in VCAM-1 homologous to the 104–116. LDV active site in fibronectin. J. Cell Sci. 107: 2127–2135. 57. Martin, S. W., and C. C. Goodnow. 2002. Burst-enhancing role of the IgG 65. van Kooyk, Y., and C. G. Figdor. 2000. Avidity regulation of integrins: the membrane tail as a molecular determinant of memory. Nat. Immunol. 3: 182– driving force in leukocyte adhesion. Curr. Opin. Cell Biol. 12: 542–547. 188. 66. Tiller, T., M. Tsuiji, S. Yurasov, K. Velinzon, M. C. Nussenzweig, and H. Wardemann. 58. Horikawa, K., S. W. Martin, S. L. Pogue, K. Silver, K. Peng, K. Takatsu, and + C. C. Goodnow. 2007. Enhancement and suppression of signaling by the con- 2007. Autoreactivity in human IgG memory B cells. Immunity 26: 205–213. served tail of IgG memory-type B cell antigen receptors. J. Exp. Med. 204: 759– 67. He, J. S., M. Meyer-Hermann, D. Xiangying, L. Y. Zuan, L. A. Jones, 769. L. Ramakrishna, V. C. de Vries, J. Dolpady, H. Aina, S. Joseph, et al. 2013. The 59. Waisman, A., M. Kraus, J. Seagal, S. Ghosh, D. Melamed, J. Song, Y. Sasaki, distinctive germinal center phase of IgE+ B lymphocytes limits their contribution Downloaded from S. Classen, C. Lutz, F. Brombacher, et al. 2007. IgG1 B cell receptor signaling is to the classical memory response. J. Exp. Med. 210: 2755–2771. http://www.jimmunol.org/ by guest on October 2, 2021