JAB1 Is Essential for Development and Germinal Center Formation and Inversely Regulates and Bcl6 Expression This information is current as of September 30, 2021. Selina Sitte, Joachim Gläsner, Julia Jellusova, Florian Weisel, Martina Panattoni, Ruggero Pardi and André Gessner J Immunol 2012; 188:2677-2686; Prepublished online 10 February 2012; Downloaded from doi: 10.4049/jimmunol.1101455 http://www.jimmunol.org/content/188/6/2677 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2012/02/13/jimmunol.110145 Material 5.DC1 References This article cites 44 articles, 20 of which you can access for free at: http://www.jimmunol.org/content/188/6/2677.full#ref-list-1

Why The JI? Submit online. by guest on September 30, 2021

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

• 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 © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

JAB1 Is Essential for B Cell Development and Germinal Center Formation and Inversely Regulates Fas Ligand and Bcl6 Expression

Selina Sitte,* Joachim Gla¨sner,† Julia Jellusova,‡ Florian Weisel,x Martina Panattoni,{ Ruggero Pardi,{ and Andre´ Gessner†

Jun activation domain-binding 1 (JAB1) regulates ubiquitin-dependent protein degradation by deneddylation of cullin-based ubiquitin ligases and, therefore, plays a central role in regulating proliferation and apoptosis. Because these processes are decisive for B cell development, we investigated JAB1 functions in B cells by establishing a mouse strain with a B cell-specific JAB1 deletion. We show that JAB1 is essential for early B cell development, because the ablation of JAB1 expression blocks B cell development between

the pro-B and pre-B cell stages. Furthermore, JAB1 deletion leads to aberrant expression of the apoptosis-triggering protein Fas Downloaded from ligand in pro-B cells. Concomitant B cell-specific overexpression of the antiapoptotic protein Bcl2 partially reverses the block in B cell development; rescued JAB1-deficient B cells reach the periphery and produce protective class-switched Abs after Borrelia burgdorferi infection. Interestingly, B cell-rescued mice exhibit no germinal centers but a striking extrafollicular plasma cell accumulation. In addition, JAB1 is essential for Bcl6 expression, a transcriptional repressor required for germinal center formation. These findings identify JAB1 as an important factor in checkpoint control during early B cell development, as well as in fate decisions in mature Ag- primed B cells. The Journal of Immunology, 2012, 188: 2677–2686. http://www.jimmunol.org/

un activation domain-binding protein 1 (JAB1) is a highly results in a partial block in development as a result of in- conserved protein that was initially identified as a coactivator creased apoptosis and accumulation of E3-ligase substrates (9). J of AP-1 transcription factors (1). Furthermore, JAB1 (syn- Apoptosis is a programmed cell death that is initiated either by onym CSN5) is a subunit of the COP9 signalosome (2, 3) that intrinsic or extrinsic signals. A well-characterized extrinsic apo- plays a central role in numerous cellular processes (4). Within the ptosis signal is the Fas–Fas ligand (FasL) interaction, which plays COP9 signalosome, JAB1 is the only catalytic subunit and reg- an important role in the immune system (10, 11). The expression ulates the activity of cullin-based ubiquitin ligases by deneddy- of FasL is tightly regulated and mainly found on hematopoietic lation (5). Therefore, JAB1 contributes to the degradation of cells, such as activated T cells and NK cells, but its expression on by guest on September 30, 2021 various and influences the regulation of the cell cycle. B cells and the contribution of JAB1 have not been addressed. Constitutive deletion of JAB1 in Drosophila and Dictyostelium Proliferation and apoptosis are tightly regulated during B cell leads to severe developmental defects (6, 7). Although many development (12), as well as during subsequent maturation and former studies addressed the functions of JAB1 in mammalian differentiation. When pro-B cells, the first discrete B cells, have cells in vitro, relatively little is known about the in vivo functions rearranged the encoding IgH, they express the pre-BCR on of JAB1 in mammals. Constitutive JAB1 knockout (KO) mice are their surfaces and, therefore, are designated pre-B cells. These embryonically lethal because of impaired proliferation and mas- early large pre-B cells divide several times to become small qui- sive apoptosis (8). The T cell-specific deletion of JAB1 in mice escent pre-B cells. If pro-B cells express no functional pre-BCR, they are eliminated by apoptosis. Immature B cells expressing a completely assembled BCR are checked for autoreactivity and are eliminated by apoptosis if it is detected (13). If the B cells *Microbiological Institute–Clinical Microbiology, Immunology and Hygiene, Univer- sity Hospital Erlangen, 91054 Erlangen, Germany; †Institute for Medical Microbiology survive, they proliferate for clonal expansion. B cell maturation and Hygiene, University of Regensburg, 93053 Regensburg, Germany; ‡Department of after Ag activation can follow two distinct pathways. x Genetics, University of Erlangen, 91058 Erlangen, Germany; School of Medicine, Yale First, the activated B cells can migrate into the B cell follicle University, New Haven, CT 06519; and {School of Medicine and Scientific Institute, Vita-Salute San Raffaele University, San Raffaele, 20132 Milan, Italy within secondary lymphatic organs and contribute germinal center Received for publication May 9, 2011. Accepted for publication January 12, 2012. (GC) formation, with subsequent Ab affinity maturation and generation of long-lived plasma cells or memory B cells (14). This work was supported by Deutsche Forschungsgemeinschaft Grants SFB643/B1 and Ge-671/12-2 (to A.G.). B cell leukemia/ (Bcl) 6 is a transcriptional repressor Address correspondence and reprint requests to Prof. Andre´ Gessner, Institute for essential for GC formation and somatic hypermutations (15–17). Medical Microbiology and Hygiene, University of Regensburg, Franz-Josef- It suppresses DNA damage responses in GC B cells (18). The Strauss-Allee 11, 93053 Regensburg, Germany. E-mail address: andre.gessner@ regulation of Bcl6 occurs at the posttranscriptional level (19), but klinik.uni-regensburg.de the exact mechanisms and the involvement of JAB1 are unknown. The online version of this article contains supplemental material. Second, the B cells can migrate to extrafollicular sites and Abbreviations used in this article: AID, activation-induced cytidine deaminase; Bcl, B cell leukemia/lymphoma; FasL, Fas ligand; fr., Hardy fraction; GC, germinal differentiate into plasma cells, leading to extrafollicular Ab re- center; ICD, intracellular domain; JAB1, Jun activation domain-binding protein 1; sponse (20). KO, knockout; MZ, marginal zone; p.i., postinfection; PNA, peanut agglutinin; WT, The murine model of Lyme disease caused by infection with wild-type. Borrelia burgdorferi is well suited for analyzing humoral immune Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 responses, because the clearance of B. burgdorferi is strictly de- www.jimmunol.org/cgi/doi/10.4049/jimmunol.1101455 2678 JAB1 FUNCTIONS IN B CELL BIOLOGY pendent on Abs (21), and the bacteria express lipoproteins with rabbit anti-human Bcl6 (Santa Cruz Biotechnology), rabbit anti-mouse very strong B cell mitogenic activity (22). During the course of HRP, and donkey anti-rabbit HRP (Dianova). infection, B. burgdorferi disseminate in the organism and are Quantitative RT-PCR preferentially found in the collagenous connective tissue of heart, Pro-B cells of Hardy fraction (fr.) A (B220+, CD43+, BP-12, CD24int) joints, and other organs (23). The main clinical outcome of ex- + + +/2 + and fr. B/C (B220 , CD43 , BP-1 , CD24 ) were sorted with a MoFlo perimental Lyme disease is arthritis, and it depends on the genetic (Dako), using the following Abs: B220-PerCP, CD43-PE (BD), BP-1– background of the mice: susceptible mice, such as C3H/He mice, FITC, and CD24-allophycocyanin (BioLegend). For quantification of Bcl6 2 develop arthritis, whereas C57BL/6 mice are resistant to it (24). and Aicda transcript levels, splenic non-GC B cells (B220+ PNA )orWT + + Previously, B cell-deficient mice (Igh6 KO) were analyzed in GC B cells (B220 PNA ) were sorted after immunization with SRBCs. Total RNA was isolated using the RNeasy Mini (QIAGEN), according murine Lyme disease; although they were on an arthritis-resistant to the manufacturer’s protocol. cDNA was prepared using SuperScript C57BL/6 background, they developed persistent arthritis (21). Reverse Transcriptase (Invitrogen). Quantitative RT-PCR was performed in Given that JAB1 regulates proliferation and apoptosis, crucial a LightCycler (Roche) with the LightCycler FastStart DNA MasterPLUS processes in B cell development, we addressed the roles of JAB1 SYBR Green I kit (Roche) in a 10-ml reaction volume using the following Aicda 9 in early B cell development by generating B cell-specific JAB1- primer pairs: (activation-induced cytidine deaminase [AID]) 5 - GCCACCTTCGCAACAAGTCT-39 and 59-CCGGGCACAGTCATAGC- deficient mice and analyzing B cell development, as well as the AC-39; Bcl6 59-TTAGAGCCCATAAGACAGTGCT-39 and 59-TCA- humoral immune response, after B. burgdorferi infection. We show GGTTGAGCCTAGATGTGT-39; Cops5 (JAB1) 59-GGCGGCGAAACCC- that JAB1 is essential for early B cell development and that TGGACTAA-39 and 59-CCGGAGAGCCAGCAGCCATAAC-39; Fasl 59- 9 9 overexpression of Bcl2 partially rescues this defect. Furthermore, TCCGTGAGTTCACCAACCAAA-3 and 5 -GGGGGTTCCCTGTTAA- ATGGG-39; and Hprt1 59-GTTGAATACAGGCCACACTTTGTTG-39 and JAB1 is essential for GC formation; thus, we identify JAB1 as an 59-GATTCAACTTGCGCTCATCTTAGGC-39. Hprt1 was amplified for Downloaded from important factor in fate decisions of Ag-primed B cells. normalization. For measurement of Cops5 (JAB1) absolute quantification was applied using the cloned cDNA, and for Aicda and Bcl6, relative Materials and Methods quantification with the comparative cycle threshold method was applied. Mouse strains IL-7 culture and proliferation assay Mice with floxed alleles of JAB1 on a C57BL6 background were described B cells from bone marrow were isolated by magnetic cell sorting with before (9). Mb1-cre mice on a BALB/c background were kindly provided CD19-coated magnetic beads (Miltenyi Biotec) on an Auto MACS device http://www.jimmunol.org/ by Michael Reth, Max-Planck Institute of Immunobiology and Epi- (Miltenyi Biotec). Cells were cultured in Opti-MEM (Invitrogen) supple- genetics, Freiburg, Germany (25). Both strains were crossed to selectively mented with the indicated concentrations of IL-7–containing supernatant del/del delete JAB1 in B cells (JAB1 ). The Em bcl2-22 strain (26) was used from J558L cells (a gift from Dr. F. Melchers, University of Basel). Pro- to introduce bcl2 as a B cell-specific transgene in rescue experiments. liferation was determined by incorporation of 3H-labeled thymidine after Experiments were performed according to the guidelines for animal wel- 3 d. For investigation of STAT5 phosphorylation, CD19-enriched B cells fare and legal requirements. were starved for 1.5 h in Opti-MEM with 0.5% FCS and then pulsed with Opti-MEM supplemented with 25 ng/ml rIL-7 and 10% FCS for 15 min. Flow cytometry and Abs Cells were immediately fixed after the IL-7 pulse. FACS analyses were performed using the following Abs and reagents: B220- Cell cycle analysis PerCP,CD5-biotin,CD19-biotin, CD23-FITC,CD43-PE, CD127(IL-7R)–PE, by guest on September 30, 2021 CD178 (FasL)-biotin, IgM-biotin (BD), CD19-FITC, CD25-allophycocyanin, Bone marrow cells were stained with Abs, as described in the flow streptavidin-allophycocyanin (Invitrogen), CD21-PE (eBioscience), IgM- cytometry section. Cells were then fixed with 2% paraformaldehyde for 10 FITC (Southern Biotech), and peanut agglutinin (PNA)-FITC (Vector Labo- min, washed, and permeabilized with 70% methanol in PBS overnight at ratories). For intracellular staining of m H chain, the Fix-and-Perm kit (ADG) 4˚C. To stain the DNA, cells were incubated with 1 mg/ml DAPI (Sigma) was used, according to the manufacturer’s protocol. For investigation of in PBS with 0.01% Tween 20 for 30 min at room temperature. Cells were STAT5 phosphorylation, the BD Phosflow system (with fixation buffer Lyse/ analyzed on a FACSCantoII flow cytometer (BD). Duplets were excluded Fix Buffer and permeabilization buffer Perm Buffer III) was used, according prior to analysis of DNA content. to the manufacturer’s protocol. Cells were stained with Phospho–STAT5– FITC (BD). Cells were analyzed on a FACSCalibur flow cytometer (BD) ELISA using CellQuest software (BD) or on a FACSCantoII (BD) using FACSDiva (BD) or FlowJo software (Tree Star) for analysis. Ab titers of naive mice were measured by standard ELISA method. Briefly, plates were coated with anti-Ig, and sera were serially diluted 1:3 in PBS/1% Histology BSA. Alkaline phosphatase-conjugated anti-Ig (Southern Biotec) was used for detection. Spleens were embedded in Tissue-Tek (Sakura Finetek). Cryostat sections were fixed in acetone, rehydrated in PBS, and blocked with PBS/10% FCS/ Serum transfer 1% BSA/2% rat serum/1% Fc block (BD). Sections were stained for 60 min del/del del/del with anti–IgM-Cy3 or -Cy5 (Dianova), anti–IgD-biotin (eBioscience), anti- Sera from B. burgdorferi-infected WT, JAB1 , and Bcl2 JAB1 mice were diluted 3-fold in sterile PBS and injected i.p. in a volume of 200 (MOMA-1)-biotin (Biomedicals), PNA-FITC (Vector Labo- m ratories), or rabbit anti-Bcl6 (Santa Cruz); washed; and stained for 30 min, l into Rag1 KO mice. After 24 h, mice were infected with B. burgdorferi. as indicated, with streptavidin-Cy3 or -Cy5 (Amersham Biosciences) or At day 28 postinfection (p.i.), pathogen burden was determined, as de- anti-rabbit IgG-Cy3 (Jackson ImmunoResearch). Immunofluorescence scribed previously (27). microscopy was performed using an Axioplan2 microscope (Zeiss) and OpenLab software (Improvision). Results del/del Infection with B. burgdorferi JAB1 mice exhibit a block in early B cell development B. burgdorferi infection, measurement of joint swelling, and detection of To examine the role of JAB1 in B cell development, we generated B. burgdorferi-specific Abs were performed, as previously described (27). B cell-specific JAB1 KO mice. A mouse strain expressing the cre The LightCycler FastStart DNA MasterPLUS SYBR Green I kit (Roche) recombinase under the control of the mb1 promoter on a BALB/c m was used for quantitative PCR in a volume of 10 l. background (25) was crossbred with a mouse strain with floxed Immunoblot analysis alleles of JAB1 on a C57BL/6 background (9). In the experiments,

+ del/del we compared littermates heterozygous for mb1-cre (designated as CD19 splenic B cells from naive WT and Bcl2 JAB1 mice were del/del sorted with a MoFlo (Dako) and lysed in standard sample buffer. Immu- JAB1 ) or lacking mb1-cre (designated as wild-type [WT]). noblot analysis was performed using the following Abs: mouse anti-human First, we performed FACS analyses of peripheral organs of WT and JAB1 (GeneTex), mouse anti-human Bcl2, rabbit anti-mouse actin (Sigma), JAB1del/del mice to investigate whether the B cell compartment is The Journal of Immunology 2679 affected by JAB1 deletion. Peripheral B cells were absent in Aberrant FasL expression in JAB1-deficient pro-B cells del/del JAB1 mice: no immature B cells were found in the spleen We then analyzed the expression of JAB1-deficient pro- + 2 (Fig. 1A), both B1a (CD5 ) and B1b (CD5 ) B cells were missing B cells. Although mb1-cre mediates deletion very early in B cell from the peritoneum, and no B cells were detected in blood (data development (25), we proposed that efficient JAB1 deletion not shown). T cell subsets were not altered (data not shown). might not occur until the late pro-B cell stage. Therefore, pro- del/del Next, we analyzed bone marrow of WT and JAB1 mice to B cells of WT and JAB1del/del micewereseparatedbycell identify the stage at which the B cell development is blocked. We sorting into fr. A and fr. B/C (28). Cells of fr. A represent early high + observed that the pro-B cell compartment (B220 /CD43 ) and pro-B cells, and cells of fr. B/C represent late pro-B cells. In + + the portion of very early pro-B cells (B220 /ckit ) were relatively JAB1del/del pro-B cells of fr. B/C, the JAB1 mRNA level was del/del enlarged in JAB1 mice (Fig. 1B, upper panels). However, the reduced to 30% compared with the WT level; however, in pro- + + del/del pre-B cell portion (CD19 /CD25 ) was reduced in JAB1 mice. B cells of fr. A, the JAB1 mRNA level was comparable between del/del Moreover, a small portion of the remaining B cells in JAB1 WT and JAB1del/del cells (Fig. 2A). Previously, p53 was shown mice were intracellularly positive for IgM H chain, indicating that to accumulate in JAB1-deficient blastocytes (8). Therefore, ex- few B cells reached the pre-B cell stage (Fig. 1B, lower panels). We pression of p53 target genes in pro-B cells of fr. B/C was ana- del/del conclude that, in JAB1 mice, B cell development is blocked lyzed. We found that Fasl mRNA, a p53 target gene (29), was between the late pro-B cell and pre-B cell stages. not detectable in WT cells but was highly expressed in JAB1del/del Histology of spleen sections confirmed the total absence of cells (data not shown). This finding was confirmed on the pro-

del/del Downloaded from peripheral B cells in JAB1 mice (Fig. 1C). WT spleens dis- tein level by flow cytometry: WT pro-B cells of fr. B/C expressed played normal architecture, with regularly formed follicles sur- no surface FasL, whereas the majority of JAB1del/del cells showed rounded by the marginal zone (MZ) and IgM-high+ plasma cells surface expression of FasL (Fig. 2B). However, expression dispersed in the red pulp. In contrast, spleens from JAB1del/del of additional known target genes of p53 were not or only mar- mice showed a strongly disturbed architecture without either ginally changed, as determined by real-time RT-PCR (Supple- B cell-containing follicles or dispersed plasma cells. mental Fig. 1). http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 1. JAB1del/del mice have a block in early B cell development. (A and B) Flow cytometry was used to compare B cell numbers in WT and JAB1del/del mice. Lymphocytes were pregated based on their forward/ side scatter profile. Numbers refer to percentage values of the respective B cell subset out of total lymphocytes. Blots are representative of at least four independent experiments each, with at least three mice per group. (A) Spleen cells (left panels) and peritoneal exudate cells (PEC; right panels) were stained with B cell- specific Abs. (B) Bone marrow cells were stained with Abs that identify different developmental B cell stages. m H chain was stained intracellularly. Pro-B cells: B220hi/CD43+; early pro-B cells: B220+/ckit+; pre-B cells: CD19+/m H chain+ and CD19+/CD25+.(C) Splenic cryosections were stained with anti-sialoadhe- sin (MOMA-1; red) to identify metallophilic macro- phages and anti-IgM (green) to identify B cells. Scale bar, 500 mm. 2680 JAB1 FUNCTIONS IN B CELL BIOLOGY

FIGURE 2. JAB1-deficient B cells show aber- rant FasL expression. (A) JAB1 mRNA levels in sorted pro-B cells of WT and JAB1del/del mice measured by quantitative RT-PCR. Cells of fr. A represent early pro-B cells, and cells of fr. B/C represent late pro-B cell. (B) FasL surface ex- pression of fr. B/C pro-B cells of WT and JAB1del/del mice determined by flow cytometry. Pro-B cells fr. B/C (B220+, CD43+, CD24+) were pregated. Graphs are representative of at least five mice analyzed per group.

JAB1del/del mice develop persistent arthritis after (m H chain2). In Bcl2 JAB1del/del mice, the frequencies of early B. burgdorferi infection and immature B cells were comparable to WT mice, whereas the We also analyzed JAB1del/del mice in the murine model of Lyme frequency of T1 B cells was reduced 6-fold compared with WT disease, because we hypothesized that putative residual B cells in mice, and recirculating B cells were almost completely missing JAB1del/del mice would be activated and massively expanded by B. (Fig. 4A, lower panels). These findings prompted us to analyze the peripheral B cell burgdorferi, an extremely strong B cell mitogen (22). Therefore, del/del to exclude the presence of functional relevant numbers of residual compartments of Bcl2 JAB1 mice to investigate the distri- Downloaded from B cells in JAB1del/del mice, WT and JAB1del/del mice were infected with B. burgdorferi to induce Lyme disease. The clinical outcome of disease was monitored by measurement of joint swelling that normally emerges first at the site of inoculation and eventually affects the contralateral site if inflammation persists. WT and del/del JAB1 mice were on an arthritis-resistant genetic background. http://www.jimmunol.org/ By day 7 p.i., JAB1del/del mice displayed arthritis of the tibiotarsal joints of the inoculated (right) site, whereas the contralateral joints were still unaffected (Fig. 3A). By day 22 p.i., the left joints of JAB1del/del mice also developed arthritis, which persisted for $114 d. In contrast, in WT mice, we observed only a moderate swelling of the right joints; the left joints were not affected. The pathogen burden was comparable in WT and JAB1del/del mice at day 22 p.i. (Fig. 3B). At later time points (days 129 and 162 p.i.), JAB1del/del mice showed a strongly increased pathogen burden, by guest on September 30, 2021 whereas in WT mice, Borrelia numbers were just above the de- tection limit. The pathogen burden was also very high in the hearts of JAB1del/del mice and was not detectable in WT hearts (data not shown). Because the clearance of B. burgdorferi is dependent on Abs (21), we measured B. burgdorferi-specific Ab titers in sera of infected mice. In WT sera, B. burgdorferi-specific Abs of the classes IgM, IgG1, and IgG2a were detected, whereas in sera of JAB1del/del mice, no B. burgdorferi-specific Abs were found (Fig. 3C). Nevertheless, JAB1del/del mice mounted a strong antibacterial Th1 cell response, similar to that of WT mice (data not shown). Taken together, these findings clearly demonstrate that JAB1del/del mice suffered from persistent Lyme disease, emphasizing the complete B cell defect in these mice. B cell development is partially rescued by B cell-specific Bcl2 overexpression We next asked whether the observed block in B cell development of del/del JAB1 mice might be rescued by antiapoptotic strategies. Bcl2 FIGURE 3. JAB1del/del mice develop chronic arthritis after B. burg- is an antiapoptotic protein that inhibits the intrinsic apoptosis dorferi infection. WT and JAB1del/del mice were infected s.c. into the right pathway (30, 31). Therefore, we crossbred the JAB1del/del mouse hind footpad with B. burgdorferi. Data are representative of three inde- strain with a mouse strain overexpressing human Bcl2 as a B cell- pendent experiments, with at least three mice/group. (A) The clinical specific transgene (26) and designated the new mouse strain Bcl2 manifestation of disease was monitored by repeatedly measuring the JAB1del/del. First, bone marrow of WT, JAB1del/del, and Bcl2 thickness of the right and left tibiotarsal joints. Each data point represents JAB1del/del mice were analyzed by flow cytometry to follow early the joint thickness of an individual mouse, and horizontal lines indicate the mean of the values for each group (n = 3–5). (B) The pathogen burden was B cell development. The overall percentage of CD19+ cells did not del/del determined on days 22, 129, and 162 p.i. by quantitative PCR using DNA differ between WT and Bcl2 JAB1 mice (Fig. 4A, upper del/del del/del extracted from the inoculated right tibiotarsal joints of WT and JAB1 panels). However, in Bcl2 JAB1 bone marrow, the frequency mice. Copy numbers of the B. burgdorferi flaB gene were measured, and + of pre-B cells (m H chain ) was reduced ∼50% compared with murine Nid1 was amplified for normalization. (C) B. burgdorferi-specific del/del WT bone marrow, but it was 10-fold higher than in JAB1 IgM, IgG1, and IgG2a Ab titers in sera of infected mice were measured by bone marrow; the remaining CD19+ cells represented pro-B cells ELISA. Triangles indicate samples with titers below the detection limit. The Journal of Immunology 2681 Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 4. Bcl2 overexpression partially rescues B cell development. (A and B) Flow cytometry was used to compare B cell numbers in WT, JAB1del/del, and Bcl2 JAB1del/del mice. Lymphocytes were pregated based on forward/side scatter profile. Numbers refer to percentage values of the respective B cell subset out of total lymphocytes. Blots are representative of at least four independent experiments each, with at least three mice/group. (A) Bone marrow cells were stained with Abs against CD19 and intracellular m H chain (upper panels) or against B220 and surface IgM (lower panels). The latter staining discriminates between early B cells (B220+/IgM2), immature (imm.) B cells (B220+/IgM+), T1 B cells (B220+/IgMhigh), and recirculating (recir.) B cells (B220high/IgM+). (B) Peripheral B cell compartments were analyzed. Blood cells, spleen cells, and peritoneal exudate cells (PEC) were stained with B cell- specific Abs. In addition, the frequencies of follicular (CD21+/CD23+) and MZ (CD21high/CD23low) B cells in spleens were determined. For this purpose, B220+ cells were pregated. bution of rescued B cells in the periphery. In blood and spleen of overexpressing mouse strain without JAB1 deletion (data not Bcl2 JAB1del/del mice, we indeed found immature IgM+ B cells; shown) and might represent immature progenitor B cells escaping however, their frequencies were reduced 3- and 16-fold, respec- from the bone marrow. In contrast, the numbers of peritoneal tively, compared with WT frequencies (Fig. 4B, upper panels). In B cells and MZ B cells in the spleen were not corrected by Bcl2 addition, IgM2 B cells were found in blood and spleen of Bcl2 overexpression (Fig. 4B, lower panels). The MZ B cell defect of JAB1del/del mice. Those cells were also detected in the Bcl2- our Bcl2 JAB1del/del strain is in line with the results of a previous 2682 JAB1 FUNCTIONS IN B CELL BIOLOGY study showing that the original Bcl2-overexpressing mouse strain bone marrow B cells proliferated upon IL-7 stimulus in a dose- harbors no MZ B cells (32). dependent manner, whereas JAB1del/del and Bcl2 JAB1del/del To investigate the spleen architecture of Bcl2 JAB1del/del mice, we cells showed no proliferation (Fig. 5C), and IL-7R was expressed performed immunofluorescence staining of spleen sections. WT at similar levels on B cells of each genotype (Supplemental Fig. spleens showed a normal architecture, whereas JAB1del/del spleens 3). STAT5 phosphorylation was analyzed to investigate the un- contained no B cells (Supplemental Fig. 2). Interestingly, in Bcl2 derlying mechanism of the observed proliferation defect. STAT5 JAB1del/del spleens, we detected neither follicular nor MZ B cells, is the key transcription factor mediating IL-7 signaling in develop- but IgM+ cells were dispersed in the red pulp. Those cells probably ing B cells (34). STAT5 phosphorylation in bone marrow B cells represented plasma cells. The absence of MZ B cells in Bcl2 was analyzed by flow cytometry with a phosphorylation-specific JAB1del/del spleens correlates with the results of our FACS analyses STAT5 Ab after a short IL-7 pulse. Stimulation of WT cells leads in which we detected no MZ B cells (Fig. 4B, lower panels). In to a small, but detectable, signal of STAT5 phosphorylation addition, fewer metallophilic macrophages (MOMA-1+)were compared with unstimulated cells (Fig. 5D). JAB1del/del and Bcl2 present in spleens of both JAB1del/del and Bcl2 JAB1del/del mice. JAB1del/del cells also exhibited STAT5 phosphorylation after IL-7 The reduced number of metallophilic macrophages in spleens of stimulation. This indicates that the deletion of JAB1 affects IL-7 JAB1del/del and Bcl2 JAB1del/del mice is in line with the results of signaling downstream of STAT5. previous studies showing that B cells are crucial for the development JAB1 deletion leads to G /M phase cell cycle arrest of MZ cells, including metallophilic macrophages (33). 2 Because JAB1 was shown to have a huge impact on the cell cycle, Rescued peripheral B cells lack JAB1 protein but show we performed cell cycle analyses of bone marrow B cells. In the Downloaded from aberrant FasL expression case of pre-B cells (intracellularly positive for IgM H chain = mH), Next, we asked whether the rescued B cells indeed lack the JAB1 the cell cycle phases were distributed similarly in WT and Bcl2 del/del del/del protein to exclude the possibility that Bcl2 JAB1 B cells, which JAB1 cells (85.1% and 82.1% for G0/G1 phase; 7.3% and might have escaped mb1-cre–mediated deletion, could have accu- 6.5% for S phase; 6.9% and 10.9% for G2/M phase) (Fig. 6A). In del/del mulated in the periphery because of prevented apoptosis. An im- contrast, fewer JAB1 cells were in G0/G1 phase (53.4%) and munoblot analysis of sorted splenic B cells from WT and Bcl2 more were in S phase (23.5%) and G2/M phase (22.7%) compared http://www.jimmunol.org/ del/del JAB1 animals demonstrated that JAB1 protein was completely with WT cells. This indicates that JAB1del/del cells are arrested in del/del + absent in Bcl2 JAB1 B cells, whereas human Bcl2 was highly S/G2/M phase. In the case of immature B cells (surface IgM ), expressed in these rescued B cells (Fig. 5A). Therefore, we con- 95.3% of WT cells were in G0/G1 phase, and 1.7% were in G2/M firmed successful JAB1 deletion in rescued B cells at the protein phase (Fig. 6B). Interestingly, only 84.9% of Bcl2 JAB1del/del cells level. Interestingly, murine Bcl2 was not detectable in the rescued were in G0/G1 phase, and 9.6% were in G2/M phase. The per- B cells (data not shown), which might be the result of an autor- centage of cells in S phase was not different between WT and Bcl2 egulatory feedback suppression by overexpressed human Bcl2. JAB1del/del cells (1.9% and 3.2%, respectively). Thus, rescued del/del We next asked whether the aberrant FasL expression of JAB1 B cells seem to be arrested in G2/M phase. pro-B cells (Fig. 2B) was compensated for in rescued peripheral by guest on September 30, 2021 B cells. Surprisingly, splenic B cells of Bcl2 JAB1del/del mice Ab production is independent of JAB1 showed enhanced FasL surface expression compared with WT To investigate whether the rescued B cells are functional in terms of B cells (Fig. 5B). Thus, Bcl2 overexpression does not abrogate Ab production, we measured serum Ig concentrations of naive WT, aberrant FasL expression caused by JAB1 deletion. JAB1del/del, and Bcl2 JAB1del/del mice by ELISA. On average, in To test the proliferative capacity of the rescued B cells in vitro, WT sera, the concentrations were 298 mg/ml for IgM, 800 mg/ml bone marrow B cells were cultured in the presence of IL-7. WT for IgG1, and 791 mg/ml for IgG2b (Fig. 7). In sera from JAB1del/del

FIGURE 5. Rescued peripheral B cells lack JAB1 protein but show aberrant FasL expres- sion. (A) Immunoblot analysis of sorted splenic B cells from naive WT and Bcl2 JAB1del/del mice. (B) FasL expression on splenic B cells of WT and Bcl2 JAB1del/del mice was determined by flow cytometry. CD19+ cells were pregated. Graphs are representative of at least five mice analyzed per group. (C) IL-7–dependent pro- liferation of CD19+ MACS-purified bone mar- row B cells was determined by incorporation of 3H-labeled thymidine. Data are representative of three independent experiments. Error bars indicate SEM. (D) STAT5 phosphorylation of CD19+ MACS-purified bone marrow B cells was analyzed after a 15-min IL-7 pulse by flow cytometry. Graphs show STAT5 phosphoryla- tion of total cells stimulated with 25 ng/ml IL-7 or unstimulated (us) as control. The Journal of Immunology 2683

FIGURE 6. JAB1 deletion leads to G2/M phase cell cycle arrest. Cell cycle phase distribution of bone marrow B cells was determined by measuring DNA

content in a flow cytometer. Left panels, Summarized results (error bar, SEM; n = 4–6; two independent experiments). Right panels, Representative FACS Downloaded from graphs. (A) Cell cycle analyses of pre-B cells. Pregated B220/mH+ cells are shown in the DAPI graphs. (B) Cell cycle analyses of immature B cells. Pregated B220/IgM+ cells are shown in the DAPI graphs. ***p , 0.001, two-way ANOVA test with Bonferroni posttest. mice, very low concentrations of Igs were detected (3 mg/ml IgM, JAB1del/del mice showed a high bacterial burden in the inoculated 19 mg/ml IgG1, and 2 mg/ml IgG2b). In contrast, sera from Bcl2 right tibiotarsal joints (Fig. 8C). In contrast, no B. burgdorferi JAB1del/del mice showed elevated Ig concentrations (16 mg/ml DNA was detected in the joints of Rag1 KO mice, which had http://www.jimmunol.org/ IgM, 148 mg/ml IgG1, and 35 mg/ml IgG2b). The very low Ig received serum from WT or Bcl2 JAB1del/del mice. As demon- concentrations in sera of some JAB1del/del mice might be produced strated by these results, Borrelia-specific Abs produced by JAB1- either by B cells that escaped the cre-mediated JAB1 deletion and/ deficient B cells are protective against B. burgdorferi after transfer or, in the case of IgG1 and IgG2b, represent residual maternal into immunodeficient mice. Therefore, the lack of endogenous Abs. These data indicate that Ab production of naive B cells is protection in Bcl2 JAB1del/del mice is most likely a consequence of independent of JAB1. too few B cells and subthreshold levels of antibacterial Abs during the early dissemination phase of the pathogen. Bcl2 JAB1del/del mice produce B. burgdorferi-specific protective Abs JAB1 is essential for GC formation, as well as the induction of by guest on September 30, 2021 To further investigate the immunological capacities of the rescued Bcl6 and AID expression JAB1-deficient B cells, we infected Bcl2 JAB1del/del mice with B. To further characterize the Bcl2-overexpressing, JAB1-deficient burgdorferi. The clinical outcome of Lyme arthritis did not differ B cells, we studied GC formation after B. burgdorferi infection. between JAB1del/del and Bcl2 JAB1del/del mice; both strains de- First, immunofluorescence analyses of spleen sections of infected veloped persistent arthritis affecting the right and left tibiotarsal mice were performed. In WT spleens, GCs (PNA+, IgD2) were joints (data not shown). However, Bcl2 JAB1del/del mice produced detected at day 84 p.i. (Fig. 9A). In addition, extrafollicular foci B. burgdorferi-specific Abs at day 29 p.i., albeit at reduced levels of plasma cells (IgM+) were found. In contrast, the spleens of compared with WT mice, whereas JAB1del/del mice did not pro- JAB1del/del mice lacked all B cells, consistent with the above- duce B. burgdorferi-specific Abs (Fig. 8A). The B. burgdorferi- mentioned B cell defect of these mice. Strikingly, spleens of specific Abs of Bcl2 JAB1del/del mice were not able to efficiently Bcl2 JAB1del/del mice contained no GCs but showed massive control the pathogen burden in the joint or heart tissue of Bcl2 extrafollicular accumulation of IgM+ cells. We performed flow JAB1del/del mice (Fig. 8B), which might result from the relatively cytometry analysis to confirm this finding. In WT spleens, 0.5% low titers during the early course of infection (e.g., day 4) in these GC B cells were identified by staining with PNA, whereas, in animals (Fig. 8A). To examine whether the B. burgdorferi-specific contrast, no GC B cells were detected in Bcl2 JAB1del/del spleens Abs of Bcl2 JAB1del/del mice have antibacterial protective capacity (Supplemental Fig. 4). Therefore, we conclude that in the absence if already present at the onset of infection, sera of B. burgdorferi- of JAB1, specific antibacterial humoral immunity is generated infected Bcl2 JAB1del/del mice were transferred into Rag1 KO solely extrafollicularly. The defect in GC formation is not caused mice prior to infection. The pathogen burden was determined 4 wk by Bcl2 overexpression, because the original Bcl2-transgenic mouse p.i. Rag1 KO mice that had received no serum or serum from strain shows normal GC formation after immunization (35).

FIGURE 7. Ab production is independent of JAB1. Serum Ig concentrations of naive mice were measured by ELISA. Horizontal lines indicate the mean of each group (n = 6–9). 2684 JAB1 FUNCTIONS IN B CELL BIOLOGY

FIGURE 8. Rescued B cells produce B. burgdorferi- specific, protective Abs. (A) B. burgdorferi-specific Ab titers were measured by ELISA. Horizontal lines indi- cate the mean of each group (n = 3–8). (B) The path- ogen burden in right tibiotarsal joints and hearts of WT, JAB1del/del, and Bcl2 JAB1del/del mice was determined at day 84 p.i. (n = 3–4; error bar, SEM). Data are representative of two independent experiments. (C) Serum transfer in Rag1 KO mice and subsequent B. burgdorferi infection. Pathogen burden was determined at day 28 p.i. in inoculated right tibiotarsal joints (n =3; error bar, SEM). Triangles indicate samples with copy numbers of flaB below the detection limit of the quantitative PCR. Downloaded from

Because the transcriptional repressor Bcl6 plays an essential role pression is restricted to B cells and CD4+ follicular Th cells within in GC formation (15–17), we investigated Bcl6 expression in the GCs (37). In WT spleens, we found a dotted Bcl6 staining sorted splenic B cells of WT and Bcl2 JAB1del/del mice after im- restricted to the GCs (Fig. 9D). In contrast, in spleens of Bcl2 munization with SRBCs. The Bcl6 transcript level at day 10 after JAB1del/del mice, very few cells within the follicle were positive http://www.jimmunol.org/ immunization was identical in WT and Bcl2 JAB1del/del mice (Fig. for Bcl6 staining, probably representing follicular Th cells (Fig. 9B). In contrast, Bcl6 protein was absent in Bcl2 JAB1del/del cells, 9D, arrows). We conclude that the absence of Bcl6 protein and but it was clearly detectable in WT cells (Fig. 9C). Another GC- Aidca mRNA in JAB1-deficient Bcl2-overexpressing B cells lead specific transcript, Aidca (encoding AID) (36), was downregulated to the observed defect in GC formation. 10-fold in B cells of Bcl2 JAB1 mice after SRBC immunization and upregulated 37-fold in WT GC B cells compared with WT Discussion non-GC B cells (Fig. 9B). We investigated JAB1 functions in B cells by cell type-specific

To further examine Bcl6 protein expression, we performed deletion and found that JAB1 is essential for early B cell devel- by guest on September 30, 2021 immunofluorescence staining of spleen sections after B. burg- opment. The B cell development is blocked between late pro-B and dorferi infection. Previous studies showed that Bcl6 protein ex- pre-B cell stages, and no peripheral B cells were found (Fig. 1).

FIGURE 9. JAB1 is essential for GC formation and Bcl6 expression. (A) Splenic cryosections of B. burg- dorferi-infected WT, JAB1del/del, and Bcl2 JAB1del/del mice at day 84 p.i. were stained with PNA (green) and anti-IgD (red) to identify GCs, as well as with anti-IgM (blue). Scale bar, 500 mm. Relative Bcl6 and Aidca transcript levels determined by quantitative RT-PCR (B) and Bcl6 protein level detected by immunoblot (C) in sorted splenic B cells of WT and Bcl2 JAB1del/del mice at day 10 after immunization with SRBCs. (D) Splenic cryosections of B. burgdorferi-infected WT and Bcl2 JAB1del/del mice at day 84 p.i. were stained with PNA (green), anti-IgD (blue), and anti-Bcl6 (red). Arrows indicate Bcl6-positive cells. Scale bar, 100 mm. Data are representative of two experiments with three to six mice/group. The Journal of Immunology 2685

This is in line with previous results showing that JAB1 is im- JAB1 is not essential for the proper immunological function of portant for thymocyte development (9). T cell-specific JAB1 de- B cells, as the Ab titers in sera of rescued mice implicate (Fig. 7). In letion resulted in partial impairment of thymocyte development addition, these rescued B cells can produce Ag-specific protective and a notable fraction of T cells emerges at the periphery. In Abs, indicating a dispensable role for JAB1 in specific humoral contrast, in our B cell-specific JAB1 KO mice, peripheral B cells immune responses. These findings extend those previously re- were absolutely absent, presumably due to the high efficiency of ported for T cells (9), because in that study, the functional prop- mb1-cre–mediated deletion (25). erties of the rescued T cells were not analyzed. Gene expression analysis of JAB1-deficient pro-B cells revealed Strikingly, the Abs produced by rescued B cells originate solely that JAB1 suppresses FasL expression (Fig. 2). It was shown that from extrafollicular plasma cells; no GCs were found in these mice JAB1 deficiency causes accumulation of p53, although proliferative (Fig. 9). The mechanisms determining extrafollicular versus GC defects observed in JAB1-deficient cells were not dependent on immune responses are largely unclear (20). Our results suggest functional p53 (38, 39). In JAB-deficient B cell precursors we that JAB1 might be a crucial factor in this decision. No Bcl6 detected no or only marginally changed expression levels of several protein was detected after immunization in rescued B cells (Fig. p53-dependent target genes, whereas FasL was markedly upregu- 9C, 9D), whereas the transcript level was similar to WT. To our lated. So we propose that FasL be defined as a novel p53-independent knowledge, our findings demonstrate for the first time that JAB1 JAB1 target protein. FasL is involved in the major apoptosis pathway mediates the previously described posttranscriptional regulation of in immune cells triggering cell death of Fas-bearing cells (11). Fur- Bcl6 (19). It remains to be investigated whether JAB1 enhances thermore, reverse signaling by FasL was shown to activate multiple Bcl6 translation and/or protein stability. A recent study showed signaling molecules in T cells, which leads to apoptosis or survival, a critical role for Bcl6 in early B cell development (44). We Downloaded from depending on cell type and conditions (40). It is tempting to speculate speculate that the observed B cell development block might be that reverse FasL signaling contributes to the termination of B cell partially caused by a putative absence of Bcl6 in JAB1-deficient development in the absence of JAB1. Interestingly, Kirkin et al. (41) early B cells. observed an 11-kDa FasL intracellular domain (ICD), which was In summary, we demonstrate that JAB1 is essential for early generated via sequential proteolysis and subsequently translocates to B cell development and that Bcl2 overexpression can partially

the nucleus where it modulates gene transcription. Furthermore, the rescue the JAB1-deletion phenotype. Secretion of antibacterial http://www.jimmunol.org/ analysis of knockin mice lacking the FasL ICD revealed that plasma class-switched Igs is not JAB1 dependent, whereas JAB1 is in- cell numbers, generation of GC B cells, and, consequently, produc- dispensable for GC formation. Furthermore, we identified FasL and tion of Ag-specific Abs in response to immunization with T cell- Bcl6 as novel target proteins of JAB1-dependent regulation. dependent or T cell-independent Ag are negatively regulated by FasL ICD-dependent signals (42). Because of the lack of suitable Acknowledgments reagents allowing the complete functional blockade of FasL, in- We thank Michael Reth for mb1-cre mice, Thomas Winkler for Bcl2-22 cluding retrograde signaling by FasL ICD in vivo, we have initiated mice, Wolfgang Schuh for assistance in cell sorting, Edith Roth for help mouse breedings to generate FasL-deficient mice with a B cell- with IL-7 cultures, Hans-Martin Ja¨ck and Reinhard Voll for providing specific JAB1 deletion to address the role of overexpressed FasL Abs, and Martha O¨ lke and Christina Becker for excellent technical assis- by guest on September 30, 2021 during B cell development. tance. Furthermore, we thank the members of the Deutsche Forschungsge- Overexpression of the antiapoptotic protein Bcl2 led to partial meinschaft Research Group 832 for helpful discussions. rescue of the block in B cell development. B cells, which lack JAB1, reach the periphery. To our knowledge, this shows for the Disclosures first time that cells can develop and survive without JAB1. The authors have no financial conflicts of interest. Interestingly, the aberrant FasL expression of JAB1-deficient pro-B cells is not abrogated by Bcl2 overexpression, because FasL is highly expressed on the surface of rescued peripheral References B cells (Fig. 5B). It remains to be investigated whether this en- 1. Claret, F.-X., M. Hibi, S. Dhut, T. Toda, and M. Karin. 1996. A new group of hanced FasL expression might lead to enhanced apoptosis of Fas+ conserved coactivators that increase the specificity of AP-1 transcription factors. Nature 383: 453–457. non-B cells in the steady state or during an immune response in 2. Seeger, M., R. Kraft, K. Ferrell, D. Bech-Otschir, R. Dumdey, R. Schade, the B cell-rescued mice. C. Gordon, M. Naumann, and W. Dubiel. 1998. A novel protein complex in- del/del volved in signal transduction possessing similarities to 26S proteasome subunits. Neither isolated rescued pro-B cells nor the JAB1 cells FASEB J. 12: 469–478. proliferate in vitro after IL-7 stimulus, demonstrating a nonre- 3. Wei, N., T. Tsuge, G. Serino, N. Dohmae, K. Takio, M. Matsui, and X.-W. Deng. dundant role for JAB1 in IL-7R-driven proliferation in vitro. As 1998. The COP9 complex is conserved between plants and mammals and is related to the 26S proteasome regulatory complex. Curr. Biol. 8: 919–922. analysis of STAT5 phosphorylation revealed, the signaling defect 4. Wei, N., G. Serino, and X.-W. Deng. 2008. The COP9 signalosome: more than in the absence of JAB1 is downstream of STAT5. However, a protease. Trends Biochem. Sci. 33: 592–600. in vivo, where factors in addition to IL-7 influence the developing 5. Cope, G. A., G. S. B. Suh, L. Aravind, S. E. Schwarz, S. L. Zipursky, E. V. Koonin, and R. J. Deshaies. 2002. Role of predicted metalloprotease motif B cells (43), some of the signals might operate in the absence of of Jab1/Csn5 in cleavage of Nedd8 from Cul1. Science 298: 608–611. JAB1 when Bcl2 is overexpressed. In fact, analysis of the cell 6. Oron, E., M. Mannervik, S. Rencus, O. Harari-Steinberg, S. Neuman-Silberberg, D. Segal, and D. A. Chamovitz. 2002. COP9 signalosome subunits 4 and 5 cycle revealed that JAB1-deficient pre-B cells are arrested in S/G2/ regulate multiple pleiotropic pathways in Drosophila melanogaster. Develop- M phase, whereas rescued pre-B cells proceed normally in the cell ment 129: 4399–4409. cycle (Fig. 6A). At the subsequent developmental stage, immature 7. Rosel, D., and A. R. Kimmel. 2006. The COP9 signalosome regulates cell proliferation of Dictyostelium discoideum. Eur. J. Cell Biol. 85: 1023–1034. rescued B cells also show abnormal cell cycle phase distribution; 8. Tomoda, K., N. Yoneda-Kato, A. Fukumoto, S. Yamanaka, and J. Y. Kato. 2004. rescued B cells are arrested in G2/M phase. We hypothesize that, Multiple functions of Jab1 are required for early embryonic development and in rescued B cells, the defects caused by JAB1 deletion become growth potential in mice. J. Biol. Chem. 279: 43013–43018. del/del 9. Panattoni,M.,F.Sanvito,V.Basso,C.Doglioni,G.Casorati,E.Montini,J.R.Bender, evident later than in JAB1 cells. The findings of the cell cycle A. Mondino, and R. Pardi. 2008. Targeted inactivation of the COP9 signalosome analyses correlate with those in JAB1-deficient T cells, which are impairs multiple stages of T cell development. J. Exp. Med. 205: 465–477. 10. Ju, S. T., D. J. Panka, H. Cui, R. Ettinger, M. el-Khatib, D. H. Sherr, arrested in S/G2/M phase (9). Thus, JAB1 is an important cell B. Z. Stanger, and A. Marshak-Rothstein. 1995. Fas(CD95)/FasL interactions cycle regulator in developing lymphocytes. required for programmed cell death after T-cell activation. Nature 373: 444–448. 2686 JAB1 FUNCTIONS IN B CELL BIOLOGY

11. Strasser, A., P. J. Jost, and S. Nagata. 2009. The many roles of 29. Fukazawa, T., T. Fujiwara, Y. Morimoto, J. Shao, M. Nishizaki, Y. Kadowaki, signaling in the immune system. Immunity 30: 180–192. A. Hizuta, L. B. Owen-Schaub, J. A. Roth, and N. Tanaka. 1999. Differential 12. Herzog, S., M. Reth, and H. Jumaa. 2009. Regulation of B-cell proliferation and involvement of the CD95 (Fas/APO-1) receptor/ligand system on apoptosis in- differentiation by pre-B-cell receptor signalling. Nat. Rev. Immunol. 9: 195–205. duced by the wild-type p53 gene transfer in human cancer cells. Oncogene 18: 13. Lu, L., and D. G. Osmond. 2000. Apoptosis and its modulation during B lym- 2189–2199. phopoiesis in mouse bone marrow. Immunol. Rev. 175: 158–174. 30. Vaux, D. L., S. Cory, and J. M. Adams. 1988. Bcl-2 gene promotes haemopoietic 14. MacLennan, I. C. M. 1994. Germinal centers. Annu. Rev. Immunol. 12: 117–139. cell survival and cooperates with c-myc to immortalize pre-B cells. Nature 335: 15. Dent, A. L., A. L. Shaffer, X. Yu, D. Allman, and L. M. Staudt. 1997. Control of 440–442. inflammation, cytokine expression, and germinal center formation by BCL-6. 31. Cory, S., and J. M. Adams. 2002. The Bcl2 family: regulators of the cellular life- Science 276: 589–592. or-death switch. Nat. Rev. Cancer 2: 647–656. 16. Fukuda, T., T. Yoshida, S. Okada, M. Hatano, T. Miki, K. Ishibashi, S. Okabe, 32. Brunner, C., D. Marinkovic, J. Klein, T. Samardzic, L. Nitschke, and T. Wirth. H. Koseki, S. Hirosawa, M. Taniguchi, et al. 1997. Disruption of the Bcl6 gene 2003. B cell-specific transgenic expression of Bcl2 rescues early B lympho- results in an impaired germinal center formation. J. Exp. Med. 186: 439–448. poiesis but not B cell responses in BOB.1/OBF.1-deficient mice. J. Exp. Med. 17. Ye, B. H., G. Cattoretti, Q. Shen, J. Zhang, N. Hawe, R. de Waard, C. Leung, 197: 1205–1211. M. Nouri-Shirazi, A. Orazi, R. S. Chaganti, et al. 1997. The BCL-6 proto- 33. Crowley, M. T., C. R. Reilly, and D. Lo. 1999. Influence of lymphocytes on the oncogene controls germinal-centre formation and Th2-type inflammation. Nat. presence and organization of dendritic cell subsets in the spleen. J. Immunol. Genet. 16: 161–170. 163: 4894–4900. 18. Ranuncolo, S. M., J. M. Polo, J. Dierov, M. Singer, T. Kuo, J. Greally, R. Green, 34. Goetz, C. A., I. R. Harmon, J. J. O’Neil, M. A. Burchill, and M. A. Farrar. 2004. M. Carroll, and A. Melnick. 2007. Bcl-6 mediates the germinal center B cell STAT5 activation underlies IL7 receptor-dependent B cell development. J. phenotype and lymphomagenesis through transcriptional repression of the DNA- Immunol. 172: 4770–4778. damage sensor ATR. Nat. Immunol. 8: 705–714. 35. Smith, K. G. C., U. Weiss, K. Rajewsky, G. J. V. Nossal, and D. M. Tarlinton. 19. Allman,D.,A.Jain,A.Dent,R.R.Maile,T.Selvaggi,M.R.Kehry,andL.M.Staudt. 1994. Bcl-2 increases memory B cell recruitment but does not perturb selection 1996. BCL-6 expression during B-cell activation. Blood 87: 5257–5268. in germinal centers. Immunity 1: 803–813. 20. MacLennan, I. C. M., K.-M. Toellner, A. F. Cunningham, K. Serre, D. M.-Y. Sze, 36. Muramatsu, M., V. S. Sankaranand, S. Anant, M. Sugai, K. Kinoshita, E. Zu´n˜iga, M. C. Cook, and C. G. Vinuesa. 2003. Extrafollicular antibody N. O. Davidson, and T. Honjo. 1999. Specific expression of activation-induced responses. Immunol. Rev. 194: 8–18. cytidine deaminase (AID), a novel member of the RNA-editing deaminase Downloaded from 21. McKisic, M. D., W. L. Redmond, and S. W. Barthold. 2000. Cutting edge: T cell- family in germinal center B cells. J. Biol. Chem. 274: 18470–18476. mediated pathology in murine Lyme borreliosis. J. Immunol. 164: 6096–6099. 37. Cattoretti, G., C.-C. Chang, K. Cechova, J. Zhang, B. H. Ye, B. Falini, 22. Yang, L., Y. Ma, R. Schoenfeld, M. Griffiths, E. Eichwald, B. Araneo, and D. C. Louie, K. Offit, R. S. K. Chaganti, and R. Dalla-Favera. 1995. BCL-6 J. J. Weis. 1992. Evidence for B-lymphocyte mitogen activity in Borrelia protein is expressed in germinal-center B cells. Blood 86: 45–53. burgdorferi-infected mice. Infect. Immun. 60: 3033–3041. 38. Fukumoto, A., K. Tomoda, N. Yoneda-Kato, Y. Nakajima, and J. Y. Kato. 2006. 23. Barthold, S. W., D. H. Persing, A. L. Armstrong, and R. A. Peeples. 1991. Ki- Depletion of Jab1 inhibits proliferation of pancreatic cancer cell lines. FEBS netics of Borrelia burgdorferi dissemination and evolution of disease after in- Lett. 580: 5836–5844.

tradermal inoculation of mice. Am. J. Pathol. 139: 263–273. 39. Yoshida, A., N. Yoneda-Kato, M. Panattoni, R. Pardi, and J. Y. Kato. 2010. http://www.jimmunol.org/ 24. Barthold, S. W., D. S. Beck, G. M. Hansen, G. A. Terwilliger, and K. D. Moody. CSN5/Jab1 controls multiple events in the mammalian cell cycle. FEBS Lett. 1990. Lyme borreliosis in selected strains and ages of laboratory mice. J. Infect. 584: 4545–4552. Dis. 162: 133–138. 40. Voss, M., M. Lettau, M. Paulsen, and O. Janssen. 2008. Posttranslational regu- 25. Hobeika, E., S. Thiemann, B. Storch, H. Jumaa, P. J. Nielsen, R. Pelanda, and lation of Fas ligand function. Cell Commun. Signal. 6: 11. M. Reth. 2006. Testing gene function early in the B cell lineage in mb1-cre mice. 41. Kirkin, V., N. Cahuzac, F. Guardiola-Serrano, S. Huault, K. Lu¨ckerath, Proc. Natl. Acad. Sci. USA 103: 13789–13794. E. Friedmann, N. Novac, W. S. Wels, B. Martoglio, A. O. Hueber, and 26. Strasser, A., S. Whittingham, D. L. Vaux, M. L. Bath, J. M. Adams, S. Cory, and M. Zo¨rnig. 2007. The Fas ligand intracellular domain is released by ADAM10 A. W. Harris. 1991. Enforced BCL2 expression in B-lymphoid cells prolongs and SPPL2a cleavage in T-cells. Cell Death Differ. 14: 1678–1687. antibody responses and elicits autoimmune disease. Proc. Natl. Acad. Sci. USA 42. Lu¨ckerath, K., V. Kirkin, I. M. Melzer, F. B. Thalheimer, D. Siele, W. Milani, 88: 8661–8665. T. Adler, A. Aguilar-Pimentel, M. Horsch, G. Michel, et al. 2011. Immune 27. Gla¨sner, J., H. Blum, V. Wehner, H. U. Stilz, J. D. Humphries, G. P. Curley, modulation by Fas ligand reverse signaling: lymphocyte proliferation is atten-

A. P.Mould, M. J. Humphries, R. Hallmann, M. Ro¨llinghoff, and A. Gessner. 2005. A uated by the intracellular Fas ligand domain. Blood 117: 519–529. by guest on September 30, 2021 small molecule alpha 4 beta 1 antagonist prevents development of murine Lyme 43. Nagasawa, T. 2006. Microenvironmental niches in the bone marrow required for arthritis without affecting protective immunity. J. Immunol. 175: 4724–4734. B-cell development. Nat. Rev. Immunol. 6: 107–116. 28. Hardy, R. R., C. E. Carmack, S. A. Shinton, J. D. Kemp, and K. Hayakawa. 44. Duy, C., J. J. Yu, R. Nahar, S. Swaminathan, S.-M. Kweon, J. M. Polo, E. Valls, 1991. Resolution and characterization of pro-B and pre-pro-B cell stages in L. Klemm, S. Shojaee, L. Cerchietti, et al. 2010. BCL6 is critical for the de- normal mouse bone marrow. J. Exp. Med. 173: 1213–1225. velopment of a diverse primary B cell repertoire. J. Exp. Med. 207: 1209–1221.