CXCR4 Expression on Activated B Cells Is Downregulated by CD63 and IL-21 Nobuya Yoshida, Daisuke Kitayama, Masafumi Arima, Akemi Sakamoto, Ayako Inamine, Haruko This information is current as Watanabe-Takano, Masahiko Hatano, Takao Koike and of September 24, 2021. Takeshi Tokuhisa J Immunol 2011; 186:2800-2808; Prepublished online 26 January 2011;

doi: 10.4049/jimmunol.1003401 Downloaded from http://www.jimmunol.org/content/186/5/2800

References This article cites 38 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/186/5/2800.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

by guest on September 24, 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 © 2011 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CXCR4 Expression on Activated B Cells Is Downregulated by CD63 and IL-21

Nobuya Yoshida,*,† Daisuke Kitayama,* Masafumi Arima,* Akemi Sakamoto,* Ayako Inamine,‡ Haruko Watanabe-Takano,x Masahiko Hatano,x Takao Koike,† and Takeshi Tokuhisa*

CXCR4 expression is critical for localization of centroblasts in the dark zone of germinal centers (GCs), and centrocytes down- regulate CXCR4 and thus leave the dark zone to reside in the zone. However, mechanisms governing CXCR4 downregula- tion on centrocytes are not known. In this study, we show that the amount of intracellular CXCR4 in centroblasts was similar to that in centrocytes, suggesting differential control of CXCR4 expression in these GC B cells. Restimulation of activated B cells with IL-21, which is a major cytokine produced by T follicular helper cells, accelerated CXCR4 internalization by inducing en- docytosis-related GRK6 expression. Although CXCR4 expression was downregulated on GC B cells by IL-21 stimulation, CXCR4low centrocytes developed in the spleens of IL-21R–deficient mice, suggesting other mechanisms for downregulation. The level of Downloaded from CD63 (which recruits CXCR4 to late endosome in CD4 T cells) in centrocytes was more than that in centroblasts and was strikingly elevated in activated Bcl6-deficient B cells. Bcl6, a transcriptional repressor, was detected on the chromatin of the CD63 in resting B cells, therefore CD63 is a molecular target of Bcl6. Downregulation of CD63 mRNA in activated Bcl6- deficient B cells by small interfering RNA upregulated CXCR4 expression on the B cells. Furthermore, addition of Bcl6 inhibitor to activated B cell cultures increased CD63 mRNA expression in (and downregulated CXCR4 expression on) those activated

B cells. Thus, CXCR4 can be downregulated on activated B cells by IL-21–induced endocytosis and CD63-mediated endosomal http://www.jimmunol.org/ recruitment, and these mechanisms may contribute to downregulation of CXCR4 on centrocytes. The Journal of Immunology, 2011, 186: 2800–2808. erminal centers (GCs) are the site for development of ceptor, CXCR4, on centroblasts is significantly more than that on high-affinity memory B cells and long-lived plasma cells centrocytes (4, 5), and CXCR4 plays an important role in segre- G (1, 2). After Ag-activated B cells collaborate with acti- gation of dark and light zones in GCs (4). Thus, CXCR4 ex- vated T follicular helper (Tfh) cells on the follicular border, some pression on centrocytes has to be downregulated to leave the dark of the activated B cells rapidly proliferate in the follicle to gen- zone to migrate to the light zone. However, the mechanism of its erate GCs. These proliferating B cells, centroblasts, undergo so- downregulation on centrocytes is not known. by guest on September 24, 2021 matic hypermutation and form the dark zone of GCs. Then, the Because CXCR4 is known as a coreceptor for HIV infection and centroblasts turn to differentiate to centrocytes with Ig class- as a key molecule for cancer metastasis, mechanisms of CXCR4 switching to IgG. The centrocytes migrate to an area adjacent to expression are mainly studied on CD4 T cells and cancer cell lines the dark zone of GCs. In the area called the light zone of GCs, (6–8). In addition to transcriptional control of the CXCR4 gene centrocytes express high-affinity IgG Abs that competitively bind (9–12), CXCR4 expression on the cell surface is regulated by to Ags on follicular dendritic cells and also collaborate with its endocytosis and exocytosis. The various endocytosis-related GC–Tfh cells, which produce IL-21 and IL-4 (3). These acti- molecules including GRK6, b-arrestin2 (Arrb2), and AIP4 are vated centrocytes scarcely proliferate and differentiate to memory inducible in CD4 T cells for the clathrin-dependent internalization B cells or long-lived plasma cells. The level of a chemokine re- of CXCR4 (13–16). Of the exocytosis-related molecules, CD63, a ubiquitously expressed , has been identified as a *Department of Developmental Genetics (H2), Graduate School of Medicine, Chiba negative regulator of CXCR4 exocytosis in CD4 T cells (17, 18). University, Chiba 260-8670, Japan; †Department of Medicine II, Hokkaido Univer- CD63 interacts with CXCR4 directly through the N-linked gly- ‡ sity Graduate School of Medicine, Sapporo 060-8638, Japan; Department of Oto- cans portion of CD63. The CD63–CXCR4 complex induces di- laryngology (J2), Graduate School of Medicine, Chiba University, Chiba 260-8670, Japan; and xDepartment of Biomedical Science (M14), Graduate School of Medicine, rection of CXCR4 trafficking to the endosomes/lysosomes, rather Chiba University, Chiba 260-8670, Japan than to the plasma membrane. Thus, activation of the endocytosis- Received for publication October 13, 2010. Accepted for publication December 22, related such as GRK6, Arrb2, and AIP4 and/or that of 2010. CD63 may be related to the downregulation of CXCR4 on cen- This work was supported in part by Grants-in-Aid from the Ministry of Education, trocytes in the light zone of GCs. Culture, Sports, Science and Technology of Japan and by the Global COE Program (Global Center for Education and Research in Immune System Regulation and Treat- We generated GC-like B cells in vitro by sequentially stimulat- ment), Ministry of Education, Culture, Sports, Science and Technology of Japan. ing splenic B cells with anti-IgM Abs and anti-CD40 Abs plus Address correspondence and reprint requests to Dr. Takeshi Tokuhisa, Department IL-4 and then with IL-21 or IL-4 after a 2-d interval (19). Using the of Developmental Genetics (H2), Graduate School of Medicine, Chiba University, in vitro stimulation system, we have reported that restimulation 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan. E-mail address: tokuhisa@faculty. chiba-u.jp of activated B cells with IL-21 or IL-4 induced proliferation low high Abbreviations used in this article: Ac-H3, acetylation of histone H3 lysine (K) 9 and and differentiation of CXCR4 or CXCR4 B cells, re- K14; Arrb2, b-arrestin2; ChIP, chromatin immunoprecipitation; Ct, threshold cycle; spectively. Thus, IL-21 from Tfh cells may play a role in down- GC, germinal center; siRNA, small interfering RNA; Tfh, T follicular helper; WT, regulation of CXCR4 on centrocytes in the light zone of GCs. In wild-type. this study, we show that restimulation of activated B cells with Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 IL-21 induces GRK6 expression to activate endocytosis of CXCR4. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1003401 The Journal of Immunology 2801

Furthermore, CXCR4 expression was continuously downregulated allophycocyanin–anti-CXCR4 (BD Pharmingen), FITC–anti-CXCR4 (BD in activated B cells from Bcl6-deficient (Bcl62/2) mice with over- Pharmingen), FITC–rat IgG2b for an isotype control (BD Pharmingen), expression of CD63, and the CD63 gene is a molecular target of FITC–anti-GL7 (BD Pharmingen), and PE–anti-Fas (BD Pharmingen). Biotinylated Abs were detected by allophycocyanin–streptavidin (BD Bcl6. We discuss roles of IL-21 and CD63 in maintaining the Pharmingen). Flow cytometric analysis was performed with a FACSCali- downregulation of CXCR4 on centrocytes in the light zone of GCs. bur (Becton Dickinson) or a FACSCanto II (Becton Dickinson) using CellQuest software (Becton Dickinson) or FlowJo software (TOMY Dig- ital Biology), respectively. For intracellular staining, cells were stained Materials and Methods with FACS Permeabilizing Solution 2 (Becton Dickinson) according to the Mice manufacturer’s instructions. For CFSE staining, purified B cells were la- beled with CFSE (Molecular Probes) as described previously (24). C57BL/6 mice were purchased from Japan SLC (Hamamatsu, Japan). 2 2 2 2 Bcl6 / (20) and IL-21R–deficient (IL-21R / ) (21) mice were as de- Real-time quantitative RT-PCR scribed. All mice were maintained under specific pathogen-free conditions in the animal center of the Graduate School of Medicine, Chiba University. RT-PCR was performed as described elsewhere (25). Total RNA was The care of all animals used in the current study was in accordance with extracted from B cells with the TRIzol reagent (Life Technologies). Total Chiba University Animal Care Guidelines. RNA was reverse-transcribed using SuperScript III Reverse Transcriptase (Invitrogen) and Oligo (dT) (Pharmacia), and the cDNAs were used for Immunization and purification of GC B cells PCR. After an initial 5-min incubation at 94˚C, 50 cycles of PCR were carried out using the following conditions: denaturation at 94˚C for 60 s, Mice were immunized intraperitoneally with 50 mg alum-precipitated (4- annealing at 60˚C for 60 s, and polymerization at 72˚C for 60 s. RT-PCR hydroxy-3-nitrophenyl) acetyl25-chicken gammaglobulin (Biosearch Tech- primers for the cDNA amplification were the following: Bcl6,59-CCG- 7 nologies) and 1.0 3 10 PFU Bordetella pertussis (Nacalai Tesque, Kyoto, GCTCAATAATCTCGTGAA-39 and 59-GGTGCATGTAGAGTGGTGAG- Japan). GC B cells were isolated from the spleens of these mice 10 d TGA-39; GRK6,59-GGAATCGCAAAGGCAAGAGCAAGA-39 and 59-T- Downloaded from after immunization. Briefly, spleen cells were first blocked with un- CACGAAATAACAGGCGCCCAATG-39; AIP4,59-AGCGAAGTAAGA- conjugated anti-CD32/16 mAbs (2.4G2; BD Pharmingen), followed by GGAGTAGCAAGG-39 and 59-TTGAACTGGATCGCGATGTCCTCT-39; incubation with allophycocyanin Cy7–anti-B220 mAbs (BD Pharmingen), b-arrestin2 (Arrb2),59-AATTTGCCTTGCTCCGTCACACTG-39 and 59- FITC–anti-GL7 mAbs (BD Pharmingen), PE–anti-Fas mAbs (BD Phar- TGATGATAAGCCGCACAGAGTTCC-39; CXCR4,59-ACGGCTGTAG- mingen), and allophycocyanin–anti-CXCR4 mAbs (BD Pharmingen). GC AGCGAGTGTT-39 and 59-AGGGTTCCTTGTTGGAGTCA-39; CD63,59- + + + + + + high B cells (B220 GL7 Fas ), centroblasts (B220 GL7 Fas CXCR4 ), cen- TCAACATAACTGTGGGCTGTGGGA-39 and 59-AGCCACCAGCAGT- + + + low + - trocytes (B220 GL7 Fas CXCR4 ), and non-GC B cells (B220 GL7 ATGTTCTTCCT-39; CD23,59-CTAGAAAGCGTTGCTGCTGTGCAA-39 http://www.jimmunol.org/ - Fas ) were sorted by a FACSAria II (Becton Dickinson). Purity of each and 59-ATTCTTCTCCGTTTCCCAGTGCCA-39; IRF-4,59-GTGGAAA- FACS-sorted population was .99%. CACGCGGGCAAGC-39 and 59-GGCTCCTCTCGACCAATTCCTCA-39; b-actin,59-CCAGCCTTCCTTCTTGGGTAT-39, and 59-TGGCATAGAG- Splenic B cell culture GTCTTTACGGATGT-39. Spleen cell suspensions were treated with lysis buffer (0.155 M ammonium Immunoblot chloride, 0.1 M disodium EDTA, 0.01 M potassium bicarbonate) to lyse erythrocytes. Single-cell suspensions were prepared in staining solution Cultured B cells were washed with PBS and lysed with a lysis buffer (1% (PBS with 3% FCS), and treated with biotinylated anti-CD43 mAbs and Nonidet P-40, 5% glycerol, 50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 10 mg/ anti-IgG1 mAbs (BD Pharmingen) for 15 min on ice. After washing, the ml leupeptin, 0.1 mM PMSF, 1 mM DTT, 1 mg/ml pepstatin A, 10 mM

cells were resuspended with streptavidin-Micro Beads (Miltenyi Biotec, Na3VO4, and 10 mM NaF). Cell lysate was subjected to SDS-PAGE, and by guest on September 24, 2021 Germany) for 15 min on ice. Labeled cells were removed by a MACS immunoblot was carried out as described previously (26). Primary Abs system (Miltenyi Biotec). The resulting B cell fraction contained .95% used were goat polyclonal IgG anti-CD63 Abs (R-13, sc-31213; Santa B220+ B cells. Purified B cells were cultured in RPMI 1640 medium (Sigma Chemical) supplemented with 10% FCS (Intergen), 50 mM2- mercaptoethanol, 10 mM HEPES, 100 mg/ml streptomycin (Wako Chemical), and 100 U/ml penicillin G potassium (Banyu Pharmaceutical). For B cell activation, purified B cells were stimulated with F(ab9)2 frag- ments of rabbit anti-mouse IgM Abs (1.0 3 1025 mg/ml; Alpha Di- agnostic), anti-CD40 mAbs (1.0 mg/ml; BD Pharmingen), and rIL-4 (2 U/ ml) (22) at the beginning of culture and sequentially stimulated with rIL-4 (20 U/ml) or rIL-21 (30 ng/ml; R&D Systems) at day 2 of culture in a humidified atmosphere at 37˚C with 5% CO2. Live cell numbers were counted under a microscope after staining with trypan blue (Invitrogen). CXCR4high and CXCR4low cells at day 2 of culture were sorted by a FACSAria II. Bcl6-inhibitor (50–500 nM, pTAT-Bcl6; YGRKKRR- QRRRGGGGGLVATVKEAGRSIHELPREEL) (23), Dynamin-inhibitor (80 mM, Dynasore; Ascent Scientific), or human rCXCL12 (1 mg/ml, SDF-1; Peprotech) was added for the indicated experiments. FACS- isolated GC B cells were stimulated with rIL-21 (30 ng/ml; R&D Sys- tems) for the indicated experiments.

Nucleofection of small interfering RNA Predesigned CD63 small interfering RNA (siRNA) (s63677, s63678, and s63679; Silencer Select Predesigned siRNA; Ambion) and scramble siRNA (Silencer Select Negative Control #1 siRNA; Ambion) were pur- chased, and 500 pM of each siRNA was transfected to purified B cells FIGURE 1. Restimulation of activated B cells with IL-21 and IL-4 in- previously stimulated with anti-IgM Abs and anti-CD40 mAbs plus rIL-4 (2 U/ml), using the nucleofector I device (Amaxa). Nucleofection of each versely regulates CXCR4 expression on those B cells in vitro. Naive B siRNA was performed according to the manufacturer’s recommendations. cells were cultured with anti-IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 7 d. IL-21 or IL-4 was added at day 2 of culture. A, IgG1 Flow cytometry analysis and CXCR4 expression on activated B220+ B cells at day 4 of culture were Cells were blocked with unconjugated anti-CD32/16 mAbs followed analyzed by flow cytometry. The numbers in the plot indicate the per- by incubation with mAbs as indicated: allophycocyanin–anti-IgG1 (BD centage of each gate. Data are presented as a representative of five in- high low Pharmingen), FITC–anti-IgG1 (BD Pharmingen), PE–anti-B220 (BD dependent experiments. B, Numbers of CXCR4 and CXCR4 B cells Pharmingen), FITC–anti-B220 (BD Pharmingen), allophycocyanin Cy7– in the culture were calculated by the flow cytometry profiles. Open and anti-B220 (BD Pharmingen), biotinylated anti-CXCR4 (BD Pharmingen), filled circles indicate IL-4 and IL-21 restimulated B cells, respectively. 2802 DOWNREGULATION OF CXCR4 ON ACTIVATED B CELLS

Cruz Biotechnology) and goat polyclonal IgG anti-actin Abs (C-11, sc- [Input]). The normalized ChIP fraction Ct value was adjusted for the 1615; Santa Cruz Biotechnology). normalized background [rabbit IgG] fraction Ct value (DDCt [ChIP/IgG] = DCt [normalized ChIP] 2 DCt [normalized IgG]). ChIP fold enrichment Chromatin immunoprecipitation assay above the sample specific background was calculated as 22DDCt [ChIP/IgG]. The following three sets of primers were used in the ChIP assays: CD63-1, Chromatin immunoprecipitation (ChIP) assay was performed using a ChIP 9 9 9 assay (Upstate Biotechnology) and conducted according to the manu- 5 -TGTGCCTTAAACAACTTCCCAGCC-3 and 5 -AAAGGCGAGAG- TTAGGTCAGATCC-39; CD63-2,59-GAGGGTTAGCCGGGATAGAAG- facturer’s recommendations as described (27). Briefly, protein and DNA in 9 9 9 9 3 6 ATT-3 and 5 -CCCACAGCCCACAGTTATGTTGAT-3 ; CD63-3,5-TA- B cells (3 10 /ml) were cross-linked by adding formaldehyde solution 9 9 (37%; Fisher Scientific) directly, and then these cells were lysed by SDS AAGTGGAGGTTTCCCTCCCATC-3 and 5 -CCTTGAAATCATTCCC- ACAGCCCA-39. lysis buffer containing protease inhibitors. The lysates were subjected to sonication to reduce DNA length to less than 1 kb. Chromatin immuno- Statistical analysis precipitation was performed using a specific Ab to the protein of interest overnight at 4˚C. The amount of the objective DNA region in the immune- Statistical analysis was done using unpaired Student t test, and p values precipitated chromatin was analyzed by quantitative PCR analysis. Real- ,0.05 were considered to be significant. time PCR was performed in 20-ml reaction volumes containing iQ SYBR- Green Supermix (Bio-Rad), 200 nM of each primer, and 1 ml each ChIP DNA fraction using the CFX96 real-time PCR detection system (Bio-Rad). Results PCR cycle parameters were 3 min at 95˚C, 45 cycles of 30 s at 95˚C, 30 s Downregulation of CXCR4 on activated B cells restimulated at 60˚C, and 30 s at 72˚C, followed by melting curve analysis. The with IL-21 is due to acceleration of CXCR4 endocytosis threshold cycle (Ct) (i.e., the cycle number at which the amount of am- plified DNA of interest reached a fixed threshold) was determined sub- When splenic B cells were sequentially stimulated with anti-IgM sequently. Relative quantification of ChIP DNA was calculated by the Abs and anti-CD40 mAbs plus a low dose of IL-4 (which did not Downloaded from comparative Ct method described elsewhere (28). Each ChIP DNA frac- induce Ig class-switching) and with IL-21 or IL-4 after a 2-d tions’ Ct value was normalized to the input DNA fraction Ct value for the same quantitative PCR assay (DCt [normalized ChIP] = Ct [ChIP] 2 Ct http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 3. Restimulation of activated B cells with IL-21 accelerates CXCR4 endocytosis. Naive B cells were cultured with anti-IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 4 d. IL-21 or IL-4 was se- quentially added at day 2 of culture. A, Cell surface and intracellular CXCR4 expression of these B cells was analyzed by flow cytometry. Filled and open histograms indicate IL-21 and IL-4 restimulated B cells, re- spectively. The isotype control of each CXCR4 staining is represented as FIGURE 2. Restimulation of activated B cells with IL-21 downregulates an open histogram with a thin broken line. The numbers in each histogram CXCR4 expression on those B cells. Naive B cells were cultured with anti- indicate the mean fluorescence intensity (isotype controls ,13). Data are IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 4 d. A, CFSE- presented as a representative of five independent experiments. B, Dynasore labeled naive B cells were cultured. IL-21 (filled histograms), IL-4 (open (Dyn) was added in the culture at day 2 of culture. CXCR4 expression on histograms), or nothing (open histogram with a broken line) was sequen- these B cells was analyzed by flow cytometry. Filled and open histograms tially added at day 2 of culture. B220+ cells in the culture were analyzed by indicate activated B cells cultured with and without Dyn, respectively. The flow cytometry. Data are presented as a representative of three independent numbers in each histogram indicate the mean fluorescence intensity. Data experiments. B, CXCR4low and CXCR4high B cells at day 2 of culture were are presented as a representative of four independent experiments. C, sorted by FACS and restimulated with IL-21 or IL-4 for another 2 d. Expression of GRK6 and AIP4 mRNA in these activated B cells was an- CXCR4 expression on these B cells was analyzed by flow cytometry. The alyzed by real-time quantitative RT-PCR. Results represent means 6 SD of number in each histogram indicates the mean fluorescence intensity. Data three independent experiments. Bars in figures represent mean values 6 are presented as a representative of four independent experiments. SD. *p , 0.05. N.S., not significant. The Journal of Immunology 2803 interval, these restimulations induced two distinct subsets of IgG1 surface CXCR4 (Fig. 3A). These results suggested that restimula- B cells, CXCR4low or CXCR4high IgG1 B cells, respectively (Fig. tion of activated B cells with IL-21 downregulates CXCR4 expres- 1A). Although the number of the activated B cells restimulated sion on IgG1 B cells by endocytosis. To examine CXCR4 endocy- with IL-21 decreased after day 5 of culture because of the dif- tosis in activated B cells restimulated with IL-21, Dynasore, a ferentiation to plasma cells, kinetic studies of activated B cells dynamin inhibitor, was added at day 2 of the culture. Dynasore in- clearly demonstrated that restimulation of activated B cells with creased the amount of CXCR4 on activated B cells restimulated IL-21 or IL-4 induced proliferation of CXCR4low or CXCR4high with IL-21 but not with IL-4, indicating an acceleration of CXCR4 activated B cells including IgG1 B cells, respectively (Fig. 1B). endocytosis in activated B cells restimulated with IL-21 (Fig. 3B). To examine the relationship between proliferation and CXCR4 Then, we measured expression of GRK6, AIP4,andArrb2 mRNA expression on activated B cells, naive B cells labeled with CFSE in sequentially activated B cells at day 4 of culture by real-time were sequentially stimulated for 4 d. As shown in Fig. 2A, being quantitative RT-PCR. The amount of GRK6 mRNA but not that associated with cell cycle progression, restimulation of activated of AIP4 or Arrb2 (data not shown) increased in activated B cells B cells with IL-21 or IL-4 decreased or increased the amount of restimulated with IL-21 but not with IL-4 (Fig. 3C). CXCR4 on those B cells, respectively. To confirm the regulation of CXCR4 expression on activated B cells by IL-21 or IL-4, we The amount of CD63 mRNA in centrocytes is more than that in isolated CXCR4high and CXCR4low activated B cells at day 2 of centroblasts culture before restimulation by FACS. These isolated B cells were In GCs, upregulation and downregulation of CXCR4 allows GC restimulated with IL-21 or IL-4 for 2 d. The IL-21 or IL-4 B cells to localize in dark and light zones, respectively (4). To restimulation clearly decreased or increased the amount of examine the effect of IL-21 on CXCR4 downregulation in GC Downloaded from CXCR4 on activated B cells, respectively (Fig. 2B). B cells, wild-type (WT) and IL-21R2/2 GC B cells were sorted When we examined the amount of intracellular CXCR4 protein from spleen cells 10 d after immunization and cultured with IL-21 in sequentially stimulated B cells and compared it with that of sur- for 48 h. CXCR4 expression on WT GC B cells but not on IL- face CXCR4 on these cells at day 4 of culture, the difference in 21R2/2 GC B cells was downregulated with IL-21 stimulation amounts of intracellular CXCR4 between IgG1 B cells restimulated (Fig. 4A). However, when we analyzed CXCR4 expression on GC 2/2 low with IL-21 or IL-4 was much less than that for amounts of cell B cells from immunized IL-21R mice, CXCR4 GC B cells http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 4. CD63 mRNA is upregulated in centrocytes. IL-21R2/2 and WT mice were immunized with alum-precipitated (4-hydroxy-3-nitrophenyl) 2/2 + + acetyl25-chicken gammaglobulin. Splenic B cells from IL-21R and WT mice 10 d after immunization were analyzed. A, GC B cells (Fas GL7 ) isolated from splenic B cells of WT and IL-21R2/2 (KO) mice were cultured with (filled histograms) or without (open histograms) IL-21 for 48 h. CXCR4 ex- pression on these cultured GC B cells was analyzed by flow cytometry. The numbers in each histogram indicate the mean fluorescence intensity. Data are presented as a representative of three independent experiments. B, Cell surface and intracellular CXCR4 expression of GC (Fas+GL7+) B cells from spleens of WT (open histograms) and IL-21R2/2 (KO, filled histograms) mice was analyzed by flow cytometry. The isotype control of each intracellular CXCR4 staining is represented as an open histogram with a thin broken line. The numbers in each histogram indicate the mean fluorescence intensity (isotype controls ,134). Data are presented as a representative of three independent experiments. C, Centroblasts (Fas+GL7+CXCR4high) and centrocytes (Fas+ GL7+CXCR4low) were sorted from splenic B cells of WT (open bars) and IL-21R2/2 (filled bars) mice by FACS. Expression of Bcl6, IRF-4, CXCR4, GRK6, and CD63 mRNA in these isolated B cells was measured by real-time quantitative RT-PCR. Results represent means 6 SD of four independent experiments. Bars in figures represent mean values 6 SD. *p , 0.05. N.S., not significant. 2804 DOWNREGULATION OF CXCR4 ON ACTIVATED B CELLS were detected in the spleens of IL-21R2/2 mice, and the amount of CXCR4 expression on CXCR4low IL-21R2/2 GC B cells was almost equivalent to that on CXCR4low WT GC B cells (Fig. 4B). These results suggested that the CXCR4 downregulation on cen- trocytes in the light zone of GCs could not be explained only by the acceleration of CXCR4 endocytosis by IL-21. To elucidate the mechanism of CXCR4 downregulation on centrocytes, we isolated CXCR4high GC B cells (centroblasts) and CXCR4low GC B cells (centrocytes) from immunized WT and IL- 21R2/2 mice and analyzed expression of Bcl6, IRF-4, and CXCR4 mRNA in those GC B cells by real-time quantitative RT-PCR. The amount of Bcl6 mRNA in centroblasts was more than that in centrocytes from both WT and IL-21R2/2 mice, and the amount of IRF-4 mRNA was inversely correlated with that of Bcl6 (Fig. 4C). Surprisingly, the amount of both CXCR4 mRNA and in- tracellular CXCR4 protein (Fig. 4B) in the centroblasts was sim- ilar to that in the centrocytes from both WT and IL-21R2/2 mice, suggesting acceleration of endocytosis and/or deceleration of exo- cytosis of CXCR4 in centrocytes. However, amounts of GRK6 Downloaded from mRNA in centrocytes from WT and IL-21R2/2 mice were not more than those in centroblasts. Because CD63 is known to traffic CXCR4 protein to late endosome in CD4 T cells (17), we ex- amined CD63 mRNA expression in these GC B cells. Indeed, CD63 mRNA expression was upregulated in the centrocytes of 2/2 both WT and IL-21R mice (Fig. 4C), suggesting that CD63 http://www.jimmunol.org/ downregulates CXCR4 expression on centrocytes.

Bcl6 negatively regulates CD63 mRNA expression in activated B cells To elucidate the mechanism of CD63 upregulation in centrocytes, we first focused on CXCL12 (SDF-1) stimulation. Because CXCL12 can transiently downregulate CXCR4 expression on CD4 T cells (29) and was suggested to be a possible regulator of CXCR4 expression on GC B cells (5), CXCL12 stimulation might by guest on September 24, 2021 induce CD63 mRNA to downregulate CXCR4 on centrocytes. Thus, naive B cells stimulated with anti-IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 2 d were restimulated with FIGURE 5. Bcl6 negatively regulates CD63 expression in activated B cells. A, Naive WT B cells were cultured with anti-IgM Abs and anti- CXCL12, and the amount of CXCR4 on their cell surfaces was CD40 mAbs plus a low dose of IL-4 for 4 d. CXCL12 was added at day 2 analyzed by flow cytometry. Though the amount of CXCR4 on of culture. CXCR4 expression on and CD63 mRNA expression in these activated B cells was downregulated 6 h after CXCL12 restim- activated B cells 6, 24, and 48 h after addition were analyzed by flow ulation, expression completely recovered 24 h after restimulation cytometry and by real-time quantitative RT-PCR, respectively. The number (Fig. 5A). Moreover, CD63 mRNA expression in these B cells was in each histogram indicates the mean fluorescence intensity. Data are not induced after CXCL12 restimulation. presented as a representative of three independent experiments. Results Then, we focused on Bcl6, a transcriptional repressor, to elu- represent means 6 SD of triplicate culture. B and C, Naive B cells from 2 2 cidate the mechanism of CXCR4 downregulation. We noted that WT (open bars) and Bcl6 / (KO, filled bars) mice were cultured with expression of Bcl6 and CD63 mRNA was inversely correlated in anti-IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 4 d. IL-21 centroblasts and centrocytes, suggesting downregulation of CD63 or IL-4 was sequentially added at day 2 of culture. B, CD63 mRNA ex- pression in naive and these activated B cells was measured by real-time mRNA in centroblasts by Bcl6. Thus, we examined expression of quantitative RT-PCR. Results represent means 6 SD of triplicate culture. CD63 2/2 mRNA and protein in activated Bcl6 B cells restimu- Bars in figures represent mean values 6 SD. *p , 0.05, **p , 0.01. C, lated with IL-21 or IL-4. Amounts of CD63 mRNA and protein CD63 protein in these B cells was detected by Western blot analysis. Data 2/2 were strikingly elevated in activated Bcl6 B cells restimulated are presented as a representative of three independent experiments. D, with IL-21 or IL-4 (Fig. 5B,5C). The amounts of CD63 mRNA Naive B cells from WT (open bars) and Bcl62/2 (KO, filled bars) mice 2 2 and protein in naive Bcl6 / B cells were also more than those in were cultured with anti-IgM Abs and anti-CD40 mAbs plus a low dose of naive WT B cells, suggesting CD63 as a molecular target of Bcl6. IL-4 for 3 d. The top figure indicates the CD63 gene map (boxes; exons) We looked for Bcl6-binding sequences in the CD63 gene. Within and putative Bcl6/STAT (BS)-binding sites (triangles). Relative amounts of the 4.8-kb sequence spanning 21.0 kb upstream to +1.0 kb Ac-H3 at and Bcl6 binding to the BS sites were measured by ChIP assay. 6 downstream from the CD63 locus (MGI: 99529), four putative Data are indicated by mean values SD of triplicate real-time quantitative RT-PCR. Results are presented as representative of three independent Bcl6-binding sequences (BS1: 2846 to 2838, 59-TTCTGTGAA- experiments. *p , 0.05. N.S., not significant. 39; BS2: 1492 to 1500, 59-TTCCTGGAG-39; BS3: +2288 to +2296, 59-TTCCTAGAA-39 and +2409 to +2417, 59-TTCAAGGAA-39 2 2 from the CD63 transcription start site) were identified by using Bcl6 / and WT B cells by ChIP assay. Bcl6 binding was detected match module of gene regulation (http://www.gene-regulation.com) at BS2 and BS3 of WT B cells. Levels of Ac-H3 at BS2 and BS3 of 2 2 (Fig. 5D). Then, we investigated Bcl6 binding and acetylation of Bcl6 / B cells were more than those of WT B cells although those histone H3 lysine (K) 9 and K14 (Ac-H3) at each BS site of naive at BS1 were similar between them. When we examined Bcl6 The Journal of Immunology 2805 binding and Ac-H3 at the CD63 locus of activated Bcl62/2 and WT termine whether or not Bcl6 deficiency affects CXCR4 in- B cells, Bcl6 binding was diminished in activated WT B cells, and ternalization. Although CXCR4 expression on activated Bcl62/2 the level of Ac-H3 increased at the locus in activated WT B cells B cells restimulated with IL-21 was slightly increased by the but not in activated Bcl62/2 B cells. These results strongly sug- addition of Dynasore, CXCR4 on activated Bcl62/2 B cells gested that the CD63 gene is a molecular target of Bcl6. restimulated with IL-4 was not increased at all (Fig. 6E). To confirm the CD63-mediated CXCR4 downregulation on Bcl6 upregulates CXCR4 expression on activated B cells by activated Bcl62/2 B cells, we examined the effect of CD63 siRNA silencing the CD63 gene on CXCR4 expression. Naive Bcl62/2 B cells were cultured with We examined CXCR4 expression on activated B cells from Bcl62/2 anti-IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 4 mice. Naive Bcl62/2 B cells were sequentially stimulated with d. Three types of predesigned CD63 siRNA or scramble siRNA anti-IgM Abs and anti-CD40 mAbs plus IL-4 and with IL-21 or were transfected to activated Bcl62/2 B cells at day 2 of culture. IL-4 after a 2-d interval. CXCR4 expression was downregulated CXCR4 expression on Bcl62/2 B cells transfected with each on activated Bcl62/2 B cells after restimulation not only with IL- CD63 siRNA increased compared with that on Bcl62/2 B cells 21 but also with IL-4 (Fig. 6A). Although the number of the IL- transfected with scramble siRNA at day 4 of culture (Fig. 7A). We 21–restimulated Bcl62/2 B cells decreased after day 4 of culture confirmed the downregulation of CD63 mRNA in CD63 siRNA because of the differentiation to plasma cells and the apoptosis, (s63679)-transfected Bcl62/2 B cells (Fig. 7B). kinetic studies of activated Bcl62/2 B cells clearly demonstrated Furthermore, we examined the effect of Bcl6 inhibitor on cell that restimulation with IL-21 or IL-4 induced proliferation of surface CXCR4 expression and CD63 mRNA expression in acti-

CXCR4low B cells including IgG1 B cells (Fig. 6B). When we vated WT B cells. Naive WT B cells were cultured with anti-IgM Downloaded from analyzed the amount of intracellular CXCR4 protein in activated Abs and anti-CD40 mAbs plus a low dose of IL-4 for 4 d. Bcl6 Bcl62/2 B cells restimulated with IL-21 or IL-4, these activated inhibitor was added to the WT B cell culture twice a day for 4 d. Bcl62/2 B cells produced similar amounts of CXCR4 mRNA (Fig. CXCR4 expression was partially downregulated on activated 6C) and protein (Fig. 6D) compared with those of activated WT B cells cultured with the Bcl6 inhibitor (Fig. 7C). CD63 mRNA B cells. Furthermore, we added Dynasore to the culture to de- expression increased in the activated B cells after addition, and the http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 6. CXCR4 expression is downregulated on activated Bcl62/2 B cells in vitro. Naive Bcl62/2 (A–E) and WT (C, D) B cells were cultured with anti-IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 7 d. IL-21 or IL-4 was added at day 2 of culture. A, IgG1 and CXCR4 expression on activated B220+ Bcl62/2 B cells at day 4 of culture was analyzed by flow cytometry. The numbers in the flow cytometry profile indicate the percentage of each gate. Data are presented as a representative of five independent experiments. B, Numbers of CXCR4high and CXCR4low Bcl62/2 B cells in the culture were calculated by the flow cytometry profiles. Open and filled circles indicate IL-4 and IL-21 restimulated B cells, respectively. C, CXCR4 mRNA expression in activated B cells was measured by real-time quantitative RT-PCR. Filled and open bars indicate Bcl62/2 and WT B cells, respectively. Results represent means 6 SD of triplicate culture. Bars in figures represent mean values 6 SD. N.S., not significant. D, Cell surface and intracellular CXCR4 expression of activated B cells was analyzed by flow cytometry. Filled and open histograms indicate Bcl62/2 and WT B cells, respectively. The isotype control of each CXCR4 staining is represented as an open histogram with a thin broken line. The numbers in each histogram indicate the mean fluorescence intensity (isotype controls ,12). Data are presented as a representative of five independent experiments. E, Dynasore (Dyn) was added in the culture at day 2 of culture. CXCR4 expression on these activated Bcl62/2 B cells was analyzed by flow cytometry. Filled and open histograms indicate Bcl62/2 B cells cultured with and without Dyn, respectively. The numbers in each histogram indicate the mean fluorescence intensity. Data are presented as a representative of three independent experiments. 2806 DOWNREGULATION OF CXCR4 ON ACTIVATED B CELLS

possible to explain the continuous downregulation of CXCR4 expression on centrocytes by CXCL12 stimulation. A previous article reported that CXCR4 expression on GC B cells was reg- ulated by the level of its transcription (5). However, we have shown here that the amount of CXCR4 mRNA in centroblasts was similar to that in centrocytes of WT mice. Although it has been proved that CXCR4 mRNA expression is mainly controlled by nuclear respiratory factor-1 and yin-yang 1 in CD4 T cells (9–11), neither nuclear respiratory factor-1 mRNA nor yin-yang 1 mRNA was induced in activated B cells or GC B cells (data not shown). These results strongly suggested that CXCR4 expression on GC B cells is regulated at the post-transcriptional level. We showed two alternative mechanisms that downregulate CXCR4 expression on activated B cells. First, we showed that IL-21 stimulation accelerated CXCR4 endocytosis in activated B cells by increasing GRK6 expression. Because IL-21 produced by Tfh cells (3) is required for GC B cells to maintain the size of GCs (30) and for centrocytes to differentiate to long-lived plasma cells (31, 32),

the IL-21–induced CXCR4 downregulation may explain the main- Downloaded from tenance of CXCR4 downregulation on centrocytes in the light zone. Indeed, IL-21 stimulation downregulated CXCR4 expres- sion on FACS-isolated GC B cells. However, the IL-21–mediated CXCR4 downregulation was not clearly detected on GC B cells FIGURE 7. Bcl6 positively regulates CXCR4 expression on activated 2/2 2 2 from WT mice compared with that from IL-21R mice at day 7 B cells. Naive B cells from Bcl6 / and WT mice were cultured with anti-

(data not shown), day 10, and day 14 (data not shown) after im- http://www.jimmunol.org/ IgM Abs and anti-CD40 mAbs plus a low dose of IL-4 for 4 d. A and B, CD63 siRNA (s63677, s63678, and s63679) or scramble siRNA was munization. We tried to detect CXCR4 endocytosis in centrocytes transfected to these Bcl62/2 B cells at day 2 of culture. A, CXCR4 ex- in vitro, but FACS-isolated GC B cells could not be cultured with pression was measured by flow cytometry at day 4 of culture. Filled and Dynasore for the period of time long enough to detect CXCR4 open histograms indicate scramble siRNA and CD63 siRNA transfected upregulation. Although we were not able to show clear results B cells, respectively. Data are presented as a representative of three in- proving IL-21–induced CXCR4 downregulation in centrocytes dependent experiments. B, CD63 mRNA expression in s63679-transfected in vivo, IL-21 has a potential to control CXCR4 downregulation not (filled bar) and scramble siRNA-transfected (Sc, open bar) B cells at day 3 only on activated B cells but also on GC B cells. of culture was measured by real-time quantitative RT-PCR. Results rep- Second, we showed that CD63, which traffics CXCR4 to late 6 resent means SD of triplicate culture. Bars in figures represent mean endosome in CD4 T cells (17), downregulates CXCR4 expression by guest on September 24, 2021 values 6 SD. *p , 0.05. C, Bcl6 inhibitor (100 nM) was administered in on activated B cells. The amount of CD63 mRNA was inversely the WT B cell culture twice a day for 4 d. CXCR4 expression was mea- sured by flow cytometry. Filled and open histograms indicate B cells with correlated with that of Bcl6 mRNA in GC B cells, and the amount 2/2 and without Bcl6 inhibitor administration, respectively. Data are presented of CD63 protein was strikingly augmented in activated Bcl6 as a representative of three independent experiments. D, Bcl6 inhibitor was B cells. Bcl6 binding was detected on the CD63 gene locus of added in the WT B cell culture at day 2 and day 2.5 of culture. Expression naive B cells. Thus, the CD63 gene is a molecular target of Bcl6. of CD63 and CD23 mRNA at day 3 of culture was measured by real-time It should be noted, however, that CD63 mRNA was detected in quantitative RT-PCR. Results represent means 6 SD of triplicate culture. centroblasts, which express a large amount of Bcl6. CD63, Bars in figures represent mean values 6 SD. *p , 0.05, **p , 0.01. N.S., a ubiquitously expressed tetraspanin, may be required in centro- not significant. blasts. Because the amount of CD63 mRNA in centrocytes was more than that in centroblasts, the larger amount of CD63 can amounts of mRNA were correlated with the dose of Bcl6 inhibitor contribute to the CXCR4 downregulation on centrocytes. There- (Fig. 7D). The amount of CD23 mRNA, whose gene is a molec- fore, IL-21–induced and CD63-mediated CXCR4 downregulation ular target of Bcl6 (23), also increased in the activated B cells may contribute to maintain CXCR4 downregulation on centro- cultured with the Bcl6 inhibitor. These results strongly suggested cytes in the light zone. that Bcl6 positively regulates CXCR4 expression on activated We showed that IL-21 restimulation induced GRK6 mRNA to B cells by silencing the CD63 gene. accelerate endocytosis of CXCR4 in activated B cells. Because IL-21 stimulation does induce Bcl6 in activated B cells (19, 33, Discussion 34), CXCR4 expression might be upregulated on activated B cells CXCR4 expression on GC B cells plays an important role in by Bcl6-mediated CD63 downregulation. However, CD63 was controlling dark and light zone segregation (4). CXCR4high cen- not significantly downregulated in activated B cells restimulated troblasts localized in the dark zone differentiate to centrocytes, with IL-21 compared with that in activated B cells without re- downregulate CXCR4, and move to the light zone in GCs. When stimulation. IL-21 stimulation also induces Blimp-1 in activated centrocytes were cultured in vitro without any stimulation, B cells, and Blimp-1 mutually represses Bcl6 expression in acti- CXCR4 was upregulated on centrocytes within 4 h after culture vated B cells (33). Indeed, the restimulation of activated B cells (5). Thus, CXCR4 is actively downregulated on centrocytes. with IL-21 induced Blimp-1 (19) and protected against the Bcl6- However, the mechanism of CXCR4 downregulation on cen- mediated downregulation of CD63. Therefore, the restimulation of trocytes is not fully understood. Although CXCL12 stimulation activated B cells with IL-21 downregulates CXCR4 expression on transiently downregulates CXCR4 expression on CD4 T cells activated B cells by acceleration of the GRK6-mediated endocy- (29) and activated B cells, expression was completely recovered tosis and CD63-induced endosome trafficking. IL-21 acts as the on activated B cells within 24 h after stimulation. Thus, it is im- inducer of Bcl6 in early phase of GC B cells such as centroblasts The Journal of Immunology 2807 and as the inducer of Blimp-1 in the late phase of GC B cells such 7. Zlotnik, A. 2006. Involvement of chemokine receptors in organ-specific me- tastasis. Contrib. Microbiol. 13: 191–199. as centrocytes (34). Thus, these two regulatory mechanisms can 8. Tsutsumi, H., T. Tanaka, N. Ohashi, H. Masuno, H. Tamamura, K. Hiramatsu, cooperate together to maintain CXCR4 downregulation on acti- T. Araki, S. Ueda, S. Oishi, and N. Fujii. 2007. Therapeutic potential of the vated B cells and probably on centrocytes. chemokine receptor CXCR4 antagonists as multifunctional agents. Biopolymers 88: 279–289. Signal transduction pathways of the IL-21–induced and the 9. Moriuchi, M., H. Moriuchi, W. Turner, and A. S. Fauci. 1997. Cloning and CD63-mediated CXCR4 downregulation on activated B cells may analysis of the promoter region of CXCR4, a coreceptor for HIV-1 entry. J. be new targets for therapy of HIV infection and WHIM syndrome. Immunol. 159: 4322–4329. 10. Wegner, S. A., P. K. Ehrenberg, G. Chang, D. E. Dayhoff, A. L. Sleeker, and CXCR4 is a part of the receptor for HIV, and the downregulation N. L. Michael. 1998. Genomic organization and functional characterization of of CXCR4 on CD4 T cells protected against HIV infection (35). the chemokine receptor CXCR4, a major entry co-receptor for human immu- Because these two regulatory mechanisms may work in CD4 nodeficiency virus type 1. J. Biol. Chem. 273: 4754–4760. 11. Moriuchi, M., H. Moriuchi, D. M. Margolis, and A. S. Fauci. 1999. USF/c-Myc T cells, activation of these two mechanisms in CD4 T cells could enhances, while Yin-Yang 1 suppresses, the promoter activity of CXCR4, protect against HIV infection. In WHIM syndrome, heterozygous a coreceptor for HIV-1 entry. J. Immunol. 162: 5986–5992. 12. Jourdan, P., J. P. Vendrell, M. F. Huguet, M. Segondy, J. Bousquet, J. Pe`ne, and truncating mutations in CXCR4 have been proposed to lead to H. Yssel. 2000. Cytokines and cell surface molecules independently induce altered lymphocyte trafficking (36). Leukocytes from WHIM CXCR4 expression on CD4+ CCR7+ human memory T cells. J. Immunol. 165: patients show impaired association of GRK6 with CXCR4 and 716–724. 13. Fong, A. M., R. T. Premont, R. M. Richardson, Y. R. Yu, R. J. Lefkowitz, and delayed recruitment of Arrb2 to CXCR4 leading to slower in- D. D. Patel. 2002. Defective lymphocyte chemotaxis in beta-arrestin2- and ternalization of the receptor (15). The impaired CXCR4 signal- GRK6-deficient mice. Proc. Natl. Acad. Sci. USA 99: 7478–7483. ing in WHIM syndrome is thought to result in defective B-cell 14. Woerner, B. M., N. M. Warrington, A. L. Kung, A. Perry, and J. B. Rubin. 2005. Widespread CXCR4 activation in astrocytomas revealed by phospho-CXCR4- functions (36). Thus, IL-21 stimulation induces GRK6 expression, specific antibodies. Cancer Res. 65: 11392–11399. Downloaded from which may enhance GRK6–CXCR4 association to weaken the 15. McCormick, P. J., M. Segarra, P. Gasperini, A. V. Gulino, and G. Tosato. 2009. abnormality of B-cell functions in WHIM patients. Impaired recruitment of Grk6 and b- 2 causes delayed internalization and desensitization of a WHIM syndrome-associated CXCR4 mutant receptor. It has been shown that coengagement of CD3 and CD63 induces PLoS ONE 4: e8102 . a potent costimulatory signal in T cells (37), and that the en- 16. Marchese, A., C. Raiborg, F. Santini, J. H. Keen, H. Stenmark, and J. L. Benovic. gagement of CD63 resulted in rapid translocation of MHC class II 2003. The E3 ubiquitin ligase AIP4 mediates ubiquitination and sorting of the G protein-coupled receptor CXCR4. Dev. Cell 5: 709–722. molecules to the endocytic pathway in dendritic cells (38), sug- 17. Yoshida, T., Y. Kawano, K. Sato, Y. Ando, J. Aoki, Y. Miura, J. Komano, http://www.jimmunol.org/ gesting that CD63 downregulation participates in Ag presentation Y. Tanaka, and Y. Koyanagi. 2008. A CD63 mutant inhibits T-cell tropic human immunodeficiency virus type 1 entry by disrupting CXCR4 trafficking to the on Ag-activated B cells. We found that expression of MHC class II plasma membrane. Traffic 9: 540–558. 2/2 on activated Bcl6 B cells was lower than that on activated WT 18. Yoshida, T., H. Ebina, and Y. Koyanagi. 2009. N-linked glycan-dependent in- B cells (data not shown). This observation suggests that the du- teraction of CD63 with CXCR4 at the Golgi apparatus induces downregulation 2/2 of CXCR4. Microbiol. Immunol. 53: 629–635. ration of T–B cell contact is impaired in Bcl6 mice. Thus, 19. Saito, T., D. Kitayama, A. Sakamoto, N. Tsuruoka, M. Arima, M. Hatano, CD63 downregulation by Bcl6 may be important not only for M. Miyazaki, and T. Tokuhisa. 2008. Effective collaboration between IL-4 and CXCR4 expression on centroblasts but also for Ag presentation on IL-21 on B cell activation. Immunobiology 213: 545–555. 20. Yoshida, T., T. Fukuda, M. Hatano, H. Koseki, S. Okabe, K. Ishibashi, S. Kojima, Ag-activated B cells. M. Arima, I. Komuro, G. Ishii, et al. 1999. The role of Bcl6 in mature cardiac In summary, we have found two novel mechanisms of CXCR4 myocytes. Cardiovasc. Res. 42: 670–679. by guest on September 24, 2021 downregulation on activated B cells. These two mechanisms 21. Ozaki, K., R. Spolski, C. G. Feng, C. F. Qi, J. Cheng, A. Sher, H. C. Morse, III, C. Liu, P. L. Schwartzberg, and W. J. Leonard. 2002. A critical role for IL-21 in may contribute to the maintenance of CXCR4 downregulation on regulating immunoglobulin production. Science 298: 1630–1634. centrocytes. Further study of these two mechanisms may provide 22. Karasuyama, H., and F. Melchers. 1988. Establishment of mouse cell lines which constitutively secrete large quantities of interleukin 2, 3, 4 or 5, using modified new insights to development of high-affinity memory B cells and cDNA expression vectors. Eur. J. Immunol. 18: 97–104. long-lived plasma cells in GCs. 23. Polo, J. M., T. Dell’Oso, S. M. Ranuncolo, L. Cerchietti, D. Beck, G. F. Da Silva, G. G. Prive, J. D. Licht, and A. Melnick. 2004. Specific peptide interference reveals BCL6 transcriptional and oncogenic mechanisms in B-cell lymphoma Acknowledgments cells. Nat. Med. 10: 1329–1335. We thank Dr. M. Osawa (Chiba University), Dr. A. Iwama (Chiba University), 24. Hasbold, J., A. B. Lyons, M. R. Kehry, and P. D. Hodgkin. 1998. Cell division and H. Satake for technical assistance, Dr. M.J. Grusby (Harvard School number regulates IgG1 and IgE switching of B cells following stimulation by of Public Health) for IL-21R2/2 mice, Dr. D. Tumes (Chiba University) for CD40 ligand and IL-4. Eur. J. Immunol. 28: 1040–1051. 25. Takamori, M., M. Hatano, M. Arima, A. Sakamoto, L. Fujimura, T. Hartatik, intensive proofreading, and S. Nakamura for secretarial assistance. T. Kuriyama, and T. Tokuhisa. 2004. BAZF is required for activation of naive CD4 T cells by TCR triggering. Int. Immunol. 16: 1439–1449. Disclosures 26. Watanabe-Takano, H., K. Takano, E. Keduka, and T. Endo. 2010. M-Ras is activated by bone morphogenetic protein-2 and participates in osteoblastic de- The authors have no financial conflicts of interest. termination, differentiation, and transdifferentiation. Exp. Cell Res. 316: 477– 490. 27. Arima, M., H. Toyama, H. Ichii, S. Kojima, S. Okada, M. Hatano, G. Cheng, References M. Kubo, T. Fukuda, and T. Tokuhisa. 2002. A putative silencer element in the 1. Manz, R. A., A. Thiel, and A. Radbruch. 1997. Lifetime of plasma cells in the IL-5 gene recognized by Bcl6. J. Immunol. 169: 829–836. 28. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression bone marrow. Nature 388: 133–134. data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. 2. Han, S., K. Hathcock, B. Zheng, T. B. Kepler, R. Hodes, and G. Kelsoe. 1995. Methods 25: 402–408. Cellular interaction in germinal centers. Roles of CD40 ligand and B7-2 in 29. Tarasova, N. I., R. H. Stauber, and C. J. Michejda. 1998. Spontaneous and J. Immunol. established germinal centers. 155: 556–567. ligand-induced trafficking of CXC-chemokine receptor 4. J. Biol. Chem. 273: 3. Chtanova, T., S. G. Tangye, R. Newton, N. Frank, M. R. Hodge, M. S. Rolph, 15883–15886. and C. R. Mackay. 2004. T follicular helper cells express a distinctive tran- 30. Linterman, M. A., L. Beaton, D. Yu, R. R. Ramiscal, M. Srivastava, J. J. Hogan, scriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide N. K. Verma, M. J. Smyth, R. J. Rigby, and C. G. Vinuesa. 2010. IL-21 acts help for B cells. J. Immunol. 173: 68–78. directly on B cells to regulate Bcl-6 expression and germinal center responses. J. 4. Allen, C. D., K. M. Ansel, C. Low, R. Lesley, H. Tamamura, N. Fujii, and Exp. Med. 207: 353–363. J. G. Cyster. 2004. Germinal center dark and light zone organization is mediated 31. Zotos, D., J. M. Coquet, Y. Zhang, A. Light, K. D’Costa, A. Kallies, by CXCR4 and CXCR5. Nat. Immunol. 5: 943–952. L. M. Corcoran, D. I. Godfrey, K. M. Toellner, M. J. Smyth, et al. 2010. IL-21 5. Caron, G., S. Le Gallou, T. Lamy, K. Tarte, and T. Fest. 2009. CXCR4 ex- regulates germinal center B cell differentiation and proliferation through pression functionally discriminates centroblasts versus centrocytes within human a B cell-intrinsic mechanism. J. Exp. Med. 207: 365–378. germinal center B cells. J. Immunol. 182: 7595–7602. 32. Ozaki, K., R. Spolski, R. Ettinger, H. P. Kim, G. Wang, C. F. Qi, P. Hwu, 6. Feng, Y., C. C. Broder, P. E. Kennedy, and E. A. Berger. 1996. HIV-1 entry D. J. Shaffer, S. Akilesh, D. C. Roopenian, et al. 2004. Regulation of B cell cofactor: functional cDNA cloning of a seven-transmembrane, G protein- differentiation and plasma cell generation by IL-21, a novel inducer of Blimp-1 coupled receptor. Science 272: 872–877. and Bcl-6. J. Immunol. 173: 5361–5371. 2808 DOWNREGULATION OF CXCR4 ON ACTIVATED B CELLS

33. Tsuruoka, N., M. Arima, E. Arguni, T. Saito, D. Kitayama, A. Sakamoto, 36. Mc Guire, P. J., C. Cunningham-Rundles, H. Ochs, and G. A. Diaz. 2010. Oli- M. Hatano, and T. Tokuhisa. 2007. Bcl6 is required for the IL-4-mediated rescue goclonality, impaired class switch and B-cell memory responses in WHIM of the B cells from apoptosis induced by IL-21. Immunol. Lett. 110: 145–151. syndrome. Clin. Immunol. 135: 412–421. 34. Good-Jacobson, K. L., and M. J. Shlomchik. 2010. Plasticity and heterogeneity 37. Pfistershammer, K., O. Majdic, J. Sto¨ckl, G. Zlabinger, S. Kirchberger, in the generation of memory B cells and long-lived plasma cells: the influence of P. Steinberger, and W. Knapp. 2004. CD63 as an activation-linked T cell co- germinal center interactions and dynamics. J. Immunol. 185: 3117–3125. stimulatory element. J. Immunol. 173: 6000–6008. 35. BouHamdan, M., D. S. Strayer, D. Wei, M. Mukhtar, L. X. Duan, J. Hoxie, and 38. Mantegazza, A. R., M. M. Barrio, S. Moutel, L. Bover, M. Weck, P. Brossart, R. J. Pomerantz. 2001. Inhibition of HIV-1 infection by down-regulation of the J. L. Teillaud, and J. Mordoh. 2004. CD63 tetraspanin slows down cell migration CXCR4 co-receptor using an intracellular single chain variable fragment against and translocates to the endosomal-lysosomal-MIICs route after extracellular CXCR4. Gene Ther. 8: 408–418. stimuli in human immature dendritic cells. Blood 104: 1183–1190. Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021