RGS13 Regulates Germinal Center B Lymphocytes Responsiveness to CXC Ligand (CXCL)12 and CXCL13

This information is current as Geng-Xian Shi, Kathleen Harrison, Gaye Lynn Wilson, of September 28, 2021. Chantal Moratz and John H. Kehrl J Immunol 2002; 169:2507-2515; ; doi: 10.4049/jimmunol.169.5.2507 http://www.jimmunol.org/content/169/5/2507 Downloaded from

References This article cites 39 articles, 17 of which you can access for free at: http://www.jimmunol.org/content/169/5/2507.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

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 28, 2021 *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 © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

RGS13 Regulates Germinal Center B Lymphocytes Responsiveness to CXC Chemokine Ligand (CXCL)12 and CXCL13

Geng-Xian Shi, Kathleen Harrison, Gaye Lynn Wilson, Chantal Moratz, and John H. Kehrl1

Normal lymphoid tissue development and function depend upon directed cell migration. Providing guideposts for cell movement and positioning within lymphoid tissues, signal through cell surface receptors that couple to heterotrimeric G , which are in turn subject to regulation by regulator of G signaling (RGS) proteins. In this study, we report that germinal center B lymphocytes and thymic epithelial cells strongly express one of the RGS family members, RGS13. Located between Rgs1 and Rgs2, Rgs13 spans 42 kb on mouse 1. Rgs13 encodes a 157-aa protein that shares 82% amino acid identity with

its 159-aa counterpart. In situ hybridization with sense and antisense probes localized Rgs13 expression to the germinal Downloaded from center regions of mouse and Peyer’s patches and to the thymus medulla. Affinity-purified RGS13 Abs detected RGS13- expressing cells in the light zone of the germinal center. RGS13 interacted with both Gi␣ and Gq␣ and strongly impaired signaling through Gi-linked signaling pathways, including signaling through the chemokine receptors CXCR4 and CXCR5. Prolonged CD40 signaling up-regulated RGS13 expression in human tonsil B lymphocytes. These results plus previous studies of RGS1 indicate the germinal center B cells use two RGS proteins, RGS1 and RGS13, to regulate their responsiveness to chemokines. The

Journal of Immunology, 2002, 169: 2507–2515. http://www.jimmunol.org/

any cellular stimuli elicit physiological responses in RGS proteins possessed GAP activity for Gi and Gq subfamily target cells by activating receptors that couple to het- members (8Ð10). Coding regions for ϳ25 human RGS proteins M erotrimeric G proteins. Activated receptors trigger G␣ have now been identified. Two Rho guanine nucleotide exchange subunits to exchange GTP for GDP, which leads to the dissociation factors, which have a divergent RGS domain, selectively act as ␤␥ of G␣ subunits from heterodimers. Both GTP-bound G␣ and GAPs for G12␣ and G13␣ and recently another subfamily of RGS ␤␥ free subunits activate downstream effectors. However, G␣ sub- proteins have been identified that act as GAPs for Gs␣ (11, 12). units possess an intrinsic GTPase activity that limits the duration Renowned for their roles in the positioning and migration of that they remain GTP bound and able to trigger signaling (re- lymphocytes within lymphoid tissues, chemokines signal thorough by guest on September 28, 2021 viewed in Refs. 1 and 2). Also limiting the duration of GTP-G␣, -coupled receptors (GPCRs) that use Gi and Gq, thereby members of the regulator of G protein signaling (RGS)2 protein suggesting that chemokine responses depend in part upon the num- family dramatically increase the intrinsic GTPase activity of G␣ ber and levels of RGS proteins that lymphocytes express (reviewed subunits, a property that defines them as GTPase-activating pro- in Refs. 13 and 14). RGS proteins may establish thresholds for teins (GAPs). Genetic studies in yeast, , responsiveness, provide a stop signal for migration, and/or con- and Aspergillus nidulans first identified such proteins (3Ð5). Sug- tribute to receptor desensitization. Experimentally, the introduc- gesting that they function by binding of G␣ subunits, a yeast two- tion of expression vectors for RGS1, RGS3, and RGS4 into B hybrid screen with a G␣ subunit identified a mammalian RGS pro- lymphocyte cell lines dramatically impaired chemokine-induced tein termed G␣-interacting protein (6). Cementing the functional cell migration (15Ð17). Among the known chemokines and their relationship between the yeast and mammalian proteins, human receptors, CXC chemokine ligand (CXCL)12 via CXCR4, RGS1, RGS2, RGS3, and RGS4 substituted to varying degrees for CXCL13 via CXCR4, and CXCL19 and CXCL21 via CCR7 pro- Sst2p, a yeast protein involved in the desensitization of pheromone vide critical positioning cues for B lymphocytes during de- signaling (7). Rapidly thereafter several studies demonstrated that velopment and/or B cell immune responses (18Ð24). Located within lymphoid tissues, germinal centers are sites crit- ical for the generation of B cells with high-affinity Ag receptors B Cell Molecular Immunology Section, Laboratory of Immunoregulation, National (reviewed in Ref. 25). During the establishment of germinal cen- Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892 ters, activated B cells and select T cells must migrate into the Received for publication May 3, 2002. Accepted for publication June 26, 2002. germinal center region from the B cell follicle. The acquisition of The costs of publication of this article were defrayed in part by the payment of page high-affinity Ag receptors, i.e., the affinity maturation of the B cell charges. This article must therefore be hereby marked advertisement in accordance Ab response, likely depends upon the recirculation of B lympho- with 18 U.S.C. Section 1734 solely to indicate this fact. cytes between the dark and light zones of the germinal center (26). 1 Address correspondence and reprint requests to Dr. John H. Kehrl, Building 10, Finally, select B cells leave the germinal center region destined to Room 11B10, Laboratory of Immunoregulation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, 9000 Rockville Pike, Bethesda, MD become plasma or memory B cells. The retention and migratory 20892-1876. E-mail address: [email protected] signals that orchestrate the movements of germinal center B cells 2 Abbreviations used in this paper: RGS, regulator of G protein signaling; GAP, remain only partially understood, although CXCL12/CXCR4 may GTPase-activating protein; GPCR, G protein-coupled receptor; GFP, green fluores- direct plasma cell precursors from the germinal center region (27). cent protein; HS, human serum; MAPK, mitogen-activated protein kinase; CHO, Among the RGS proteins expressed in B lymphocytes, RGS1 Chinese hamster ovary; EST, expressed sequence tag; PNA, peanut agglutinin; ϩ CXCL, CXC chemokine ligand. shows a prominent expression in germinal center B cells, CD38 /

Copyright © 2002 by The American Association of Immunologists, Inc. 0022-1767/02/$02.00 2508 RSG13 REGULATES GERMINAL CENTER B LYMPHOCYTE RESPONSIVENESS

Ϫ IgD , sorted from human tonsil, while similarly sorted mantle strate kit for peroxidase (Vector Laboratories). The slides were counter- zone B cells, CD38Ϫ/IgDϩ, lacked detectable RGS1 by immuno- stained with hematoxylin, dehydrated, cleared in ethanol and xylene, and blotting (16). Surprisingly, we find that germinal center B cells permanently mounted using Permount (Fisher Scientific, Pittsburgh, PA). prominently express another RGS protein, RGS13. Although Isolation of cells and their analysis by RT-PCR and Western RGS1 can be found in many other tissues beside B cells, RGS13 blotting possesses a very restricted range of tissue expression. Beside char- The tonsil B and T cells were isolated from tonsil mononuclear cells by two acterizing RGS13 in B cells and B cell lines, we have also exam- rounds of SRBC rosetting as previously described (16). The purity of tonsil ined what signals modulate its expression, determined its intracel- B cells was routinely Ͼ95%. A total of 1 ϫ 107 of the purified B cells were lular localization, and examined its ability to modulate signaling stimulated with various reagents for 4, 24, 48 h in RPMI 1640 supple- through a variety of GPCRs including CXCR4 and CXCR5. mented with 10% FCS, respectively, at final concentration of 100 ng/ml CXCL12 (R&D Systems, Minneapolis, MN), 1 ␮g/ml CXCL13 (R&D Systems), 1 ␮g/ml anti-CD40 or anti-CD95 mAb, 20 ␮mol/L 1-palmitoyl- Materials and Methods 2-hydroxy-sn-glycero-3-phosphocholine (16/0; Avanti, Alabaster, AL), 4 Plasmids and reagents ␮l/ml anti-human IgM (␮-chain-specific) antiserum, 100 ng/ml PMA, or 30 ␮mol/L L-␣-lysophosphatidic acid (Sigma-Aldrich). The stimulated ton- The coding region of mouse RGS13 was isolated by PCR from a cDNA sil B cells were harvested and the total RNA was isolated with TRIzol clone with GenBank number AW495950 and then subcloned into EcoRI/ reagents (Life Technologies) and then 2 ␮g total RNA was subsequently BamHI of p3XFLAG-CMV-14 (Sigma-Aldrich, St. Louis, MO) or used for reverse transcription (Omniscript; Qiagen, Cologne, Germany). pEGFP-N1 (Clontech Laboratories, Palo Alto, CA). The coding region of ThefollowingPCRprimerswereusedforthePCR:RGS13,ATGAGCAGGCG human RGS13 was inserted into EcoRI/BamHI sites of p3XFLAG-CMV- GAATTGTTGGA and GAAACTGTTGTTGGACTGCATA; RGS1, CCAG 14. Human RGS13 C-terminal tagged with green fluorescent protein (GFP) GAATGTTCTTCTCTGCTAACCCA and TCACTTTAGGCTATTAGCCT Downloaded from was prepared by inserting it into pcDNA3.1/CT-GFP-topoisomerase (In- GCAGG; and ␤-actin-GTTTGAGACCTTCAACACCC and ATACTCCT vitrogen, Carlsbad, CA) and subsequently into the EcoRI site of pAAV- GCTTGCTGATCC. MCS (Stratagene, La Jolla, CA). The coding region of CXCR5 was iso- Mouse bone marrow cells were isolated from femur bones. A sample lated by RT-PCR using RNA prepared from human serum (HS)-Sultan ϩ was taken for RNA isolation and the rest was used to isolate B220 bright cells and subcloned into the EcoRI site of pAAV-MCS. Dr. S. Gutkind ϩ or B220 dim from the lymphocyte-gated population by cell sorting. (National Institute of Dental Research, National Institutes of Health, Be- Briefly, the cells were incubated with mAbs to B220 and CD43, washed thesda, MD) kindly provided the expression vectors for the activated G ϩ ϩ ␣ with staining buffer (PBS without Ca2 /Mg2 plus 1% BSA fraction V), subunits. Human RGS1 and RGS3 C-terminal tagged with Flag were de- http://www.jimmunol.org/ and sorted on the FACStarPlus (BD Biosciences, Mountain View, CA). B scribed previously (7, 16). The Abs against the following were purchased: cells from mesenteric , peripheral lymph nodes, and Peyer’s FLAG (Sigma-Aldrich), phospho-p42/44 mitogen-activated protein kinase patches were isolated by negative depletion. A single-cell suspension was (MAPK; , Beverly, MA), p42/44 MAPK, G ,G and ␣i2 ␣q /11 made and treated with biotinylated Abs to CD4, CD8, CD11c, and GR-1 G (Santa Cruz Biotechnology, Santa Cruz, CA), CD11c, CD40, CD95, ␣s for 15 min at 4¡C rotating. After washing, the cells were resuspended with CD4, CD8, CD21, CD23, CD43, GR-1, and B220 (BD PharMingen, San streptavidin-conjugated Dynabeads (Dynal, Lake Success, NY) for 15 min Diego, CA). at 4¡C. Cell bead complexes were removed with a Dynal magnetic particle Cell lines and cell cultures concentrator. The B cell purity obtained with this depletion procedure was 90Ð95%. To obtain activated splenic B cells, mice were immunized i.p. The CA46 (a human Burkitt’s B cell lymphoma), Ramos (a human Bur- with a 10% SRBC solution and the spleens were harvested 7 days later. The

kitt’s B cell chronic lymphoma), Raji (a human Burkitt’s B cell chronic B cells were isolated as described above. The splenic B cell subpopulations by guest on September 28, 2021 lymphoma), HS-Sultan (a B lineage human plasmacytoma), HR1 (a mouse were isolated by staining single-cell suspensions of splenic B cells with normal line), KG-1 (a human acute myelogenous leukemia), mAbs CD21, CD23, and B220 and cell sorting on a FACStarPlus (BD HEL92.1.7 (a human erythroleukemia), K562 (a human chronic myelog- Biosciences). The populations collected were CD21ϩ/B220ϩ, CD21ϩ/ enous leukemia), HL-60 (a human promyelocytic leukemia), Jurkat (a hu- CD23ϩ/B220ϩ, and B220ϩ/CD21dim/CD23Ϫ. RNA from each cell popu- man acute T cell leukemia line), MOLT-4 (a human T lineage leukemia), lation was used to generate cDNAs using the Clontech Advantage RT-PCR U937 (a human histiocytic lymphoma), HeLa (a human epitheloid carci- kit (Clontech Laboratories), following kit methodology using the Rgs13- noma), 293T (a T Ag-transformed primary embryonal kidney line with specific primers GAAAATTGCTTCACGAAGGGG and GCATGTTT transformation of adenovirus type 5 DNA), Chinese hamster ovary (CHO)- GAGTGGGTTCACGAATG. In addition, tonsil T and B cells and mouse K1, COS (a SV40-transformed fibroblast like kidney line) were obtained spleen cell lysates were prepared using a kinase lysis buffer (20 mM from the American Type Culture Collection (Manassas, VA). The human HEPES (pH 7.4), 2 mM EGTA, 50 mM ␤-glycerophosphate, 1% Triton Burkitt’s lymphoma cell lines COMER and JDRB were kind gifts from Dr. X-100, 1 mM sodium orthovanadate, and 10% glycerol plus protease in- I. Magrath (National Cancer Institute, National Institutes of Health), and hibitors). Thirty-microgram proteins of human tonsil T and B cells and the human B cell line BJAB.6 and BHM23 were kind gifts from Dr. T. mouse spleen lysates was fractionated by SDS-PAGE and transferred to Folkes (Center for Disease Control and Prevention, Atlanta, GA). All of the pure nitrocellulose. The membrane was immunoblotted as below using a lymphoid cell lines and the CHO cells were maintained in RPMI 1640 1/500 diluted affinity-purified rabbit anti-mRGS13 polyclonal Ab. (Life Technologies, Gaithersburg, MD) supplemented with 10% FCS (Life Technologies), whereas the other nonlymphoid cells were maintained in Northern blots and in situ hybridization DMEM (Life Technologies) plus 10% FCS. For detecting human RGS13 expression, 20 ␮g total RNA was isolated Generation of polyclonal antisera to RGS13 and from 19 cell lines using TRIzol reagents. The RNAs were size fractionated immunohistochemistry staining and transferred to nitrocellulose. The membranes were hybridized with a 480-bp RGS13 cDNA fragment labeled with [␣-32P]dCTP using Prime-It A peptide based on a region of RGS13 amino acid sequence likely to be RmT Random Primer Labeling kit (Stratagene, La Jolla, CA) as a probe. highly antigenic, QPQSPREINIDSTT, was synthesized and conjugated to ␤-actin expression was used as a control. Hybridization was performed at keyhole limpet hemocyanin before injection into rabbits for the generation 68¡C for 2 h using QuikHyb (Stratagene), washed three times in 2ϫ SSC/ of polyclonal Abs. The rabbit serum was affinity purified using the RGS13 0.1% SDS for 15 min each at room temperature, and then in 0.1ϫ SSC/ peptide (Covance, Princeton, NJ). The Ab recognizes both RGS13-GFP 0.1% SDS at 60¡C for 30 min. For detecting mouse Rgs13 expression, and RGS13-Flag when they are expressed in COS or HEK293 cells. For mouse tissue and embryo blots were purchased (Clontech Laboratories). immunohistochemistry, frozen sections of spleen and thymus tissue were The membranes were hybridized and processed similar to the human blots prepared. The frozen slides were fixed in cold acetone for 2 min. Endog- with the exception that a 480-bp Rgs13 cDNA fragment was used. For in enous peroxidase was quenched with DAKO peroxidase blocking reagent situ hybridization, a 1076-bp fragment of the Rgs13 cDNA was subcloned (DAKO, Carpinteria, CA). Endogenous biotin in tissue was blocked with into PCR 4-topoisomerase (Invitrogen). Antisense and sense RGS13 ribo- an avidin-biotin blocking kit (Vector Laboratories, Burlingame, CA). The probes were transcribed with T3 and T7 RNA polymerases in the presence sections were preincubated in 5% goat serum in PBS and then overnight of 35S-labeled UTP (Amersham Pharmacia Biotech, Piscataway, NJ). with anti-RGS13 rabbit polyclonal Abs diluted 1/100 in PBS with 0.1% Mouse embryo sections of various developmental ages and sections of Triton X-100 and 1% goat serum. The slides were washed with PBS, in- adult mouse organs were prepared at Molecular Histology Laboratories cubated with goat anti-rabbit IgG (Vectastain Elite avidin-biotin complex (Gaithersburg, MD). In situ hybridization was performed as previously kit; Vector Laboratories), and then developed with diaminobenzidine sub- described (28). The Journal of Immunology 2509

Intracellular localization of RGS13 ting with anti-phospho-p42/44 MAPK mAb using detergent-soluble frac- tions of lysates after fractionation by SDS-PAGE. Cells (HeLa, 293T, COS, CHO) grown in chamber slides (80% confluent) were transfected with constructs expressing RGS13-GFP or GFP using Migration assay SuperFect (Qiagen) according to the manufacturer’s instruction, and the isolated and purified tonsil B cells were transfected with RGS13-GPF or CHO cells in six-well plates were transfected with or without constructs GFP using a human B cell Nuleofector kit (Amaxa Biosystems, Koln, expressing CXCR4 plus or not RGS13-GFP or RGS1-GFP for 36 h and Germany) according to the user manual. The translocation experiments then harvested. The cells were loaded into the insert (8-␮m pore size) of a were performed in HeLa cells transfected with constructs expressing Boyden chamber (Costar, Cambridge, MA) and fibronectin-coated bottom RGS13-GFP or GFP plus constructs that express different GTPase-defi- chamber (BD PharMingen), and the CXCL12 was added into the bottom cient G␣ subunits. The slides were observed with an Olympus BX-60 mi- chamber at a final concentration of 100 ng/ml. After a 6-h incubation at croscope equipped with a BX-FLA fluorescent attachment or a laser con- 37¡Cina5%CO2 incubator, the migrated cells were harvested and resus- focal microscope (National Institute of Allergy and Infectious Diseases pended in 300 ␮l FACS buffer. The GFP-positive cells (included both Imaging Unit, Dr. O. Schwartz). weak and strongly positive cells) were counted using a FACS (BD Phar- Mingen), and distribution of GFP expression in the loaded cells, migrated, Immunoblotting and immunoprecipitations and nonmigrated cells was determined. The percentage of migrating cells was calculated by dividing the number of GFP-positive cells in the mi- Cell lysates were prepared using an appropriate lysis buffer plus protease grated sample by the total number of GFP-positive cells loaded in the upper inhibitors for 20 min on ice. The detergent-insoluble materials were re- chamber. moved by microcentrifugation for 10 min at 4¡C. Equal amounts of pro- teins from each sample were fractionated by 10% SDS-PAGE and trans- ferred to pure nitrocellulose. Membranes were blocked with 5% BSA in Results Tween 20 plus Tris-buffered saline (TTBS) for 1 h and then incubated with Identification of an Rgs13 cDNA and mapping the Rgs13 locus an appropriate dilution of the primary Ab in 5% BSA in TTBS for2hor Downloaded from overnight. The blots were washed three times with TTBS before the ad- Based on the known DNA sequence of a RGS13 cDNA (GenBank dition of a biotinylated Ab (DAKO) diluted 1/5,000 in TTBS containing accession number AF030107), we screened the murine expressed 5% BSA for 1 h and then incubated with streptavidin conjugated to HRP sequence tag (EST) database at the National Center for Biotech- (DAKO) diluted 1/10,000 in TTBS containing 5% BSA for 1 h. The signal nology Information and identified several likely ESTs derived was detected by ECL according to the manufacturer’s instruction (Amer- from Rgs13. Complete nucleotide sequencing of a mouse B cell sham Pharmacia Biotech). The coimmunoprecipitation of RGS13 and G␣ subunits was conducted in 293T cells transfected with C-terminal FLAG- cDNA clone (GenBank accession number AW495950) revealed a http://www.jimmunol.org/ tagged constructs of RGS13 or RGS3. Briefly, 293T cells were cultured in 1262-bp cDNA predicted to encode a 157-aa protein 83% identical 100-mm plates and then transfected with 10 ␮g RGS constructs using to human RGS13 (GenBank accession number AF498319, Fig. 1). SuperFect (Qiagen). After a 36-h incubation in DMEM, 293T cells were We used the Rgs13 cDNA to construct a murine RGS13 expres- incubated in the presence or absence of aluminum fluoride (30 ␮M AlCl , 3 sion vector and amplified the coding region for human RGS13 10 mM NaF, and 10 mM MgCl2 in PBS) for 30 min on ice. The cells were lysed in kinase lysis buffer and anti-FLAG mAb was added, and the im- from reverse-transcribed human B cell mRNA to construct a hu- munoprecipitates were collected with antimouse Ig-conjugated magnetic man RGS13 expression vector. Human and mouse RGS13 are beads (Dynal). They were washed four times in lysis buffer, twice in lysis among the smallest of the RGS proteins composed largely of a buffer with 0.5 M NaCl, twice again in lysis buffer, and then fractionated by SDS-PAGE and analyzed by immunoblotting with appropriate anti-G RGS domain. Both human and mouse RGS13 have three cysteines protein Ab. The efficiency of immunoprecipitation was verified by immu- in a configuration of CXXCXXC beginning at amino acid by guest on September 28, 2021 noblotting with anti-FLAG mAb. position 6. Using the mouse genome sequencing consortium database, we MAPK assay determined the chromosomal location and genomic structure of COS cells were transfected with appropriate receptor expression constructs Rgs13. Composed of six exons, Rgs13 spans 42 kb on mouse chro- (0.5 ␮g, respectively) in the presence or absence of RGS13, RGS1, or mosome 1. A surprisingly large intron of 29 kb separates exon 4 RGS3 (2.0 ␮g, respectively) using SuperFect. After starving the cells using fresh medium without FCS for 6 h, cells were stimulated with CXCL12 from exon 5. Four of the six Rgs13 exons are coding, the last three (100 ng/ml) or CXCL13 (250 ng/ml) for varying durations and then lysed of which encode the RGS domain plus some flanking amino acid with 300 ␮l lysis buffer. MAPK activation was detected by immunoblot- sequence (Fig. 1). Similar to RGS13, which sits between RGS1 and

FIGURE 1. Genomic structure of Rgs13 and amino acid comparison of mouse RGS13, RGS1, and RGS2 and hu- man RGS13. Exons are denoted by solid boxes and introns by solid lines. Introns drawn according to their relative sizes. The coding exons are indicated by their contribution to the RGS13 protein. 2510 RSG13 REGULATES GERMINAL CENTER B LYMPHOCYTE RESPONSIVENESS

RGS2 on human , Rgs13 resides between Rgs1 and RGS13 mRNA (Fig. 2). To determine Rgs13 expression during B Rgs2 on mouse chromosome 1 (29). cell development and in different B cell populations, we used RT- PCR. We purified total bone marrow cells, bone marrow B220ϩ Expression of Rgs13 mRNA and RGS13 protein cells, newly formed splenic B cells, splenic marginal zone B cells, As a first check of Rgs13 expression, we searched the murine EST splenic follicular B cells, and B cells from immunized spleens, database using the Rgs13 cDNA sequence. We found Rgs13 ex- Peyer’s patches, mesenteric lymph nodes, or peripheral lymph pressed sequence tags from murine B cells (accession number nodes. The analysis revealed very weak Rgs13 expression in total ϩ AW495950), thymus (accession number BG089493), gallbladder bone marrow cells, but not in B220 bone marrow cells. Because ϩ (accession number BB871265), and neonatal cerebellum (acces- of the lack of Rgs13 in the B220 cells in the bone marrow, we did sion number BB257313). Next, we examined Rgs13 expression in not examine Rgs13 expression in finer detail during B cell devel- a variety of mouse tissues and RGS13 expression in human lym- opment. Among the other cell fractions, we found the highest lev- phoid and nonlymphoid cell lines by Northern blot analysis and els of Rgs13 expression in B cells purified from immunized mouse RT-PCR and RGS13 expression in cell lysates from mouse spleen spleens and from Peyer’s patches (Fig. 2). cells, human tonsil B cells, and from HS-Sultan cells by immu- We raised and affinity-purified Abs against an RGS13 internal noblotting. Among the murine tissues tested, we found Rgs13 ex- peptide conserved between human and mouse. The affinity-puri- pressed in spleen, stomach, small intestine, and thymus as well in fied Abs recognized a 19-kDa band present in lysates prepared later stage mouse embryos. The Rgs13 mRNA transcripts ran as a from human tonsil B cells, an HS-Sultan cell line, immunized single band of ϳ1.6 kb, suggesting that Rgs13 does not generate mouse spleen, but not in a lysate from human tonsil T cells. The alternative mRNA transcripts (Fig. 2). Northern blot analysis of RGS13 Abs readily recognized a RGS13-GFP expressed in Downloaded from human cell lines revealed a restriction of RGS13 expression to B HEK293 cells as well as FLAG-tagged RGS13 (Fig. 2 and data not cell lines. Eight of the 19 cell lines tested were of B cell origin and shown). all of them expressed RGS13. BJA-B and several of the Burkitt Although Northern blot analysis detected Rgs13 expression in lymphoma cells lines expressed the highest levels of RGS13. In RNA prepared from developing mouse embryos, we did not visu- contrast, other hemopoietic cell lines, including T cell lines and alize any Rgs13 expression by in situ hybridization upon examin- nonhemopoietic cell lines, failed to express detectable levels of ing sagittal sections prepared from e12.5, e14.5, e17.5, and newborn http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 2. Analysis of Rgs13, and RGS13, and RGS13 expression. AÐC, Northern blots of Rgs13 expression in indicated tissues and mouse embryos. Tissues indicated as well as age of embryos. Blot rehybridized with a ␤-actin probe to verify equal loading. D, Northern blot of RGS13 expression in human cell lines. Cell lines analyzed are shown above the blot. ␤-actin levels shown below. E, RT-PCR analysis of Rgs13 expression in purified cell populations from spleen and bone marrow. Spleen B cell populations were sorted into newly formed (NF), follicular (Fol), and mantle zone (MZ) on the basis of B220, CD21, and CD23 expression. Follicular cells are from two separate experiments. Total B cells were prepared from the spleen of a 7-day immunized mouse by negative selection (IB). Peripheral lymph node B cells (PLN), mesenteric lymph node B cells (MLN), and Peyer’s patch B cells were prepared by negative selection. Bone marrow cells were either unfractionated or sorted on the basis of B220 expression. The RNAs were analyzed for both Rgs13 and ␤-actin expression. F, Immunoblot of protein lysates using affinity-purified anti-RGS13 peptide-specific Abs. Cell lysates were prepared from mouse spleen (postimmunization), human tonsil B cells, human tonsil T cells, a HS-Sultan cell subline that expresses high levels of endogenous RGS13, and COS cells transfected or not with a construct that expresses RGS13-GFP. The Journal of Immunology 2511 mice (data not shown). Apparently, there is insufficient Rgs13 ex- pression levels. We prepared RNA from purified B cells stimulated pression in embryonic tissues to allow detection by a standard in with CXCL12, CXCL13, anti-CD40, anti-␮, PMA, L-␣-lysophos- situ hybridization protocol. Nevertheless, we readily detected phatidic acid, anti-Fas, or 1-palmitoyl-2-hydroxy-sn-glycero-3- Rgs13 expression by in situ hybridization using sections prepared phosphocholine for 4, 24, and 48 h and subjected the RNA to from mouse spleen, ileum in the regions of Peyer’s patches, and RT-PCR using either RGS13- or RGS1-specific primers. We ob- thymus (Fig. 3). The spleen and thymus sections derived from served that either CD40 or PMA augmented RGS13 expression at nonimmunized, day 9, or day 14 postimmunized mice. We verified each of the time points while only PMA enhanced RGS1 expres- the development of germinal center in the B cell zones of the sion at all three time points, although anti-␮ enhanced RGS1 ex- mouse spleens by detecting peanut agglutinin (PNA) staining pression at 4 h (Fig. 4). CD40 stimulation had little effect on RGS1 within lymphoid follicles. Although we did not detect an Rgs13 expression as previously reported for Rgs1 (17). None of the other hybridization signal in spleens from unimmunized mice, we found inductive stimuli significantly altered either RGS13 or RGS1 ex- a strong signal in the germinal center regions of spleens from im- pression. Thus, RGS1 and RGS13 expression may respond to dif- munized mice. We also detected Rgs13 expression in mouse Pey- ferent stimuli, CD40 stimulation augments RGS13 while engage- er’s patches in the germinal center regions. Beside its expression in ment of the B cell Ag receptor acts to enhance RGS1 expression. germinal centers, a very localized Rgs13 in situ signal overlaid the epithelial cells in the medullary region of the thymus, which oc- Intracellular localization of RGS13-GFP depends upon the cell curred independent of the immunization status of the mice. Immu- type examined nohistochemistry with affinity-purified RGS13 anti-peptide Abs The intracellular location of RGS proteins varies among the par- also detected RGS13 expression in the germinal center region and ticular family members that have been examined. Constitutive Downloaded from the thymic medullary region (Fig. 3). The germinal center cells plasma membrane, intracellular vesicle, cytoplasm, and intranu- that reacted with the RGS13 Ab predominantly localized within clear expression have all been documented as well as the recruit- the light zone region. ment from the cytosol to plasma membranes upon cell activation (30Ð33). To examine RGS13 expression, we prepared expression Analysis of inductive stimuli for RGS13 expression vectors for human and murine RGS13 fused to GFP. We found in

Next, we examined RGS13 expression in tonsil B cells stimulated HEK293T cells and in COS cells that mouse RGS13-GFP pre- http://www.jimmunol.org/ with a variety of stimuli and compared the results to RGS1 ex- dominantly resided within the nucleus although we also observed by guest on September 28, 2021

FIGURE 3. Analysis of Rgs13 expression by in situ hybridization and mouse RGS13 expression by immunohistochemistry. A, PNA staining of a spleen section from an unimmunized mouse (original magnification, ϫ400). B, PNA staining of a spleen section from a day 9-immunized mouse showing germinal center (GC) development (original magnification, ϫ400). CÐE, Rgs13 expression in unimmunized, day 9, and day 14 postimmunization spleen sections (original magnification, ϫ100, dark field). Germinal centers indicated by arrows. F, Sense probe applied to day 9-immunized spleen section (original magnification, ϫ100). G, Rgs13 in germinal center (original magnification, ϫ1000, bright field). H, Sense probe applied to day 9-immunized spleen section (original magnification, ϫ1000, bright field). I, RGS13 expression mouse spleen germinal center. Brown staining is observed in a portion of the germinal center closest to the marginal zone (original magnification, ϫ1000). J, Rgs13 expression in Peyer’s patches (PP) from mouse ileum (original magnification, ϫ200, dark field). Arrow indicated hybridization signal. K, Rgs13 expression in Peyer’s patches from mouse ileum (original magnification, ϫ1000, bright field). L, Rgs13 expression mouse thymus medulla (M) (original magnification, ϫ200, dark field). M, Rgs13 expression in mouse thymus (original magnification, ϫ1000, bright field). Thymus medulla (M) and cortex (C) indicated. N, Sense probe in mouse thymus (original magnification, ϫ1000, bright field). O, RGS13 expression in mouse thymus by immunocytochemistry (original magnification, ϫ400). P, Control Ab staining of mouse thymus (original magnification, ϫ400). In the dark field photographs shown, the hybridization signal is observed as reflected silver grains (white) while in the bright field photographs the hybridization signal is visualized as dark grains. 2512 RSG13 REGULATES GERMINAL CENTER B LYMPHOCYTE RESPONSIVENESS

(Fig. 5), although they failed to significantly alter the intracellular localization of GFP (data not shown). Coexpression with an active

form of Gi␣ caused RGS13-GFP to accumulate at intracellular junctions; coexpression with an active form of Gq␣ altered the morphology of the HeLa cells, but did not substantially change the localization of RGS13-GFP; surprisingly, the coexpression with an

active form of Gs␣ shifted RGS13-GFP from the cytoplasm to the nucleus; and coexpression with an active form of Go␣ resulted in a perinuclear accumulation of RGS13-GFP. To examine RGS13- GFP in human B cells, we introduced either human or mouse RGS13-GFP into purified human B cells and examined its intra- cellular location by confocal imaging. Like the results with HeLa cells RGS13-GFP localized predominantly in the cytoplasm of pri- mary human B cells (Fig. 5).

Interaction with G proteins and regulation of GPCR signaling by RGS13 Comparison of the RGS13 amino acid sequence with that of RGS proteins reveals that RGS13 has a number of substitutions in Downloaded from amino acids conserved among other family members. Among the 120-aa RGS domain 40 aa are strongly conserved between the RGS proteins that possess G and/or G GAP activity. Substitu- FIGURE 4. Analysis of RGS13 and RGS1 expression in human tonsil B i q cells. Purified tonsil B cells were stimulated or not with the indicated re- tions occur in 7 of those 40 aa in RGS13 while 0 substitutions agents for 4, 24, and 48 h. Purified RNA was subjected to RT-PCR using occur in RGS1 and only 3 in RGS2. However, G␣ pull-down and http://www.jimmunol.org/ RGS1, RGS13, and actin specific primers. Non-reverse transcriptase control functional signaling experiments indicate that RGS13 affects Gi␣ shown (lane 11). Positive control was the appropriate plasmid DNA (lane and Gq␣ signaling in a manner similar to that of other RGS pro- 12). One-kilobase ladder run in lanes adjacent to lanes 1 and 12. PCR teins. Pull-down experiments using the FLAG Ab to immunopre- products are visualized by ethidium staining. cipitate either FLAG-RGS13 or a control RGS protein, FLAG- RGS3, expressed in HEK293T cells demonstrated the interaction of RGS13 with both G and G (Fig. 6). As previously shown some plasma membrane expression, whereas in HeLa cells i␣ q␣ with other RGS proteins, demonstration of that interaction depends RGS13-GFP predominated in the cytoplasm (Fig. 5). Expression upon the prior treatment of the cell lysates with aluminum fluoride of GFP alone resulted in cytosolic expression in all of the cell (8, 9). When expressed in cells we found that the mouse RGS13- types (data not shown). Coexpression of different GTPase-defi- by guest on September 28, 2021 FLAG routinely ran as a doublet while human RGS13 ran as a cient G subunits altered RGS13-GFP localization in HeLa cells ␣ single band on SDS-PAGE. Next, we examined the effect of RGS13 on signaling through a number of different GPCRs using either COS or HEK293T trans- fected with an expression vector for an individual GPCR in the presence or absence of expression vectors for human RGS13,

FIGURE 5. Analysis of RGS13-GFP expression in various cell types.

Constructs expressing RGS13-GFP were transfected into 293T cells (A), FIGURE 6. Interaction of RGS13 with endogenous G␣ subunits. COS cells (B), HeLa (CÐG), or human tonsil B lymphocytes (H and I). HEK293T cells were transfected with expression vectors for RGS3-Flag Cells were cotransfected with expression vectors for constitutively active (lanes 2 and 6), mouse RGS13-Flag (lanes 3 and 7), or human RGS13-Flag forms of G␣i (D), G␣q (E), G␣s (F), or Go␣ (G). A construct expressing (lanes 4 and 8). The cell lysates were either not treated (lanes 1Ð4)or mouse RGS13-GFP was used with the exception of the cell shown in I treated with AlF4- (lanes 5Ð8). Flag immunoprecipitates (IP) were iso- where a construct expressing human RGS13-GFP was used instead. H and lated. Abs specific for G␣i2,G␣q /11,G␣s, or Flag were used to immunoblot I, Confocal images. the immunoprecipitates and the cell lysates. The Journal of Immunology 2513 mouse RGS13, or human RGS3. Following exposure to the ap- propriate ligand, we measured either the production of inositol triphosphates, the production of cAMP, the activation of a NF-␬B reporter , the activation of a serum-response element-depen- dent reporter gene, or the phosphorylation of endogenous p42 MAPK. RGS13 had no effect on ␤-adrenergic receptor-mediated cAMP production (Gs dependent); it inhibited angiotensin recep- ϳ tor-induced p42 MAPK phosphorylation (Gq dependent) by 60% at 5, 10, and 15 min following stimulation; it inhibited M1 mus- carinic receptor-induced inositol triphosphate production (Gq de- pendent) by ϳ40%, and it partially inhibited M1 muscarinic re- ceptor-mediated serum-response element and NF-␬B activation

(Gq and G12/13 dependent) by 35 and 60%, respectively. In each instance, human and murine RGS13 performed similarly, although in each case RGS3 expression provided a greater degree of inhi- bition than did RGS13 (data not shown). Expression of RGS13 also very modestly inhibited the activation of the serum-response element reporter following stimulation with activated forms of Downloaded from G12␣ or G13␣; however, whether this represents a significant effect on G12 and G13 signaling need further investigation (data not shown). Because of the likely importance of RGS13 in modulating the response of germinal center B cells to chemokines, we also com- pared the effect of RGS3 and RGS13 on signaling through the

CXCR4 and CXCR5 receptors. We transfected COS cells with http://www.jimmunol.org/ either the CXCR4 or CXCR5 receptors in the presence or absence of expression vectors for human RGS13, mouse RGS13, or human RGS3. We stimulated the cells with either CXCL12 or CXCL13 and measured the phosphorylation of endogenous p42 MAPK (Fig. 7). We found that both human and mouse RGS13 potently inhib- ited the production of phosphorylated p42 MAPK following ex- posure to CXCL12 (equivalent or even superior to RGS3) or CXCL13 (nearly equivalent to RGS3). Because germinal center B cells can express either or potentially both RGS13 and RGS1, we by guest on September 28, 2021 directly compared RGS1 and RGS13 on CXCR4 and CXCR5 sig- naling. At equivalent levels of expression RGS13 proved superior to RGS1 in its ability to inhibit CXCR4 and CXCR5 signaling to p42 MAPK activation 5 min poststimulation (Fig. 7). We observed similar results at 2 and 10 min following exposure to ligand (data not shown). Pertussis toxin, an inhibitor of Gi␣ signaling, also potently inhibited CXCR4 and CXCR5 induced p42 MAPK phos- FIGURE 7. RGS13 inhibits CXCL12 and CXCL13 signaling. A, Inhi- phorylation consistent with the know importance of G ␣ in che- i bition of CXCL12-induced phosphorylated MAPK. COS cells were trans- mokine signaling (34, 35). fected with an expression vector for CXCR4 (lanes 1Ð16) in the presence Finally, we examined the ability of RGS13 to inhibit cell mi- or absence of expression vectors for human RGS3 (hRGS3, lanes 4, 8, 12, gration by using a transient transfection system and CHO cells. We and 16), human RGS13 (hRGS13, lanes 2, 6, 10, and 14), or mouse RGS13 transfected CHO cells with an expression vector for CXCR4 in the (mRGS13, lanes 3, 7, 11, and 15). The cells (lanes 5Ð16) were stimulated presence of an expression vector for either GFP, mouse RGS13- with CXCL12 for 2, 5, or 10 min. The amount of phosphorylated p42 GFP, human RGS13-GFP, or human RGS1-GFP. We determined MAPK (phospho-p42 MAPK) induced was detected with a specificAbby the number of GFP-positive cells loaded in the upper well of a immunoblotting. The levels of human RGS3, human RGS13, mouse chamber and the number of GFP-positive cells in the RGS13, and p42/44 MAPK are shown. B, Inhibition of CXCL13-induced bottom chamber following a 6-h chemotaxis assay. We found that phospho-p42 MAPK. Similar experiment as shown in the first panel except CHO cells transfected with the CXCR4R migrated in response to the cells were transfected with CXCR5 rather than CXCR4. The cells were stimulated with CXCL13 for 5, 10, or 15 min. C, Comparison of RGS1 and stimulation with CXCL12 and that the transfection of either mouse RGS13 on chemokine signaling. Similar experimental design as first two or human RGS13-GFP or human RGS1-GFP substantially inhib- panels. The cells were transfected with the indicated expression vectors and ited CHO cell migration as compared with GFP alone (Fig. 8). stimulated with CXCL12 or CXCL13 for 5 min. The levels of phospho-p42 Thus, the expression of RGS13 can inhibit cell migration in re- MAPK, hRGS1, hRGS13, hRGS3, and p42/44 MAPK are shown. sponse to chemokine stimulation.

Discussion spleen, in Peyer’s patches, and in the thymus; however, not in In this report, we describe the isolation of Rgs13 cDNAs and the thymic lymphocytes. Despite a number of nonconservative substi- mapping of the Rgs13 genomic locus on chromosome 1. We show tutions in its RGS domain, RGS13 potently inhibits signaling that Rgs13 has a restricted range of tissue expression with B lym- through GPCRs that couple to Gi␣ and to a lesser extent to those phocytes, one of the major sites of its expression. We document that couple to Gq␣. A direct comparison with RGS1 indicates that strong Rgs13 expression in the germinal center regions of mouse at similar expression levels RGS13 more potently inhibits 2514 RSG13 REGULATES GERMINAL CENTER B LYMPHOCYTE RESPONSIVENESS

The major questions unanswered by this study are precisely what role Rgs13 plays in germinal B cell function and why should germinal center B cells use multiple RGS proteins. To begin to answer these questions, we need to unambiguously ascertain which germinal center B cells express Rgs13 and which Rgs1. The RGS13 Ab used in this study reacted with a limited number of cells in the light zone of the germinal center, yet the in situ hy- bridization study detected Rgs13 expression in a higher percentage of the germinal center cells. Varying sensitivities of the two meth- ods or protein instability may account for the failure of the mRNA and protein to perfectly colocalize; however, additional studies are needed to resolve this inconsistency. Providing a possible mech- anism to differentially express Rgs13 and Rgs1 during the recruit- ment and trafficking of B cells through the germinal center, CD40 FIGURE 8. RGS13 inhibits CXCL12-induced migration of CXCR4-ex- and Ig stimulation distinctively regulated RGS1 and RGS13 ex- pressing CHO cells. CHO cells were transfected with constructs expressing pression. Current models of germinal center formation and main- CXCR4 and either GFP, human RGS13-GFP (hRGS13-GFP), mouse tenance suggest a temporal dissociation between these signals RGS13-GFP (mRGS13-GFP), or human RGS1-GFP (hRGS1-GFP). The (36Ð38). Based on the known functions of RGS proteins and our numbers of GFP-positive cells loaded into the upper portion of chemotaxis Downloaded from chambers were enumerated as well as the number of GPF-positive cells current studies of RGS13, we can speculate on what functional that migrated into the CXCL12-containing bottom well. The percentage of roles RGS13 may have in germinal B cell function. Our current GFP-positive cells that migrated was calculated. The data are from one hypothesis based on both in vitro studies and ongoing studies of experiment of two performed. The controls are the mean Ϯ SEM of six RGS1Ϫ/Ϫ mice (C. Moratz, manuscript in preparation) is that a wells, while the others are the mean Ϯ SEM of three wells. major function of RGS proteins in lymphocytes is to help them negotiate a complex chemokine milieu. Since RGS proteins may exhibit some degree of receptor selectivity the simple expression http://www.jimmunol.org/ of RGS13 could bias a cell to preferentially respond to one source CXCL12 and CXCL13 signaling although RGS1 and RGS13 sim- of chemoattractant vs another. An intriguing idea is that RGS pro- ilarly inhibited the migration of CXCR4 transfected CHO cells. teins may be asymmetrically expressed within the cell. Such in- Based on the localization of a RGS13-GFP fusion protein, RGS13 tracellular localization of RGS13 could allow for directional mi- largely resides in the cytoplasm of B lymphocytes. However, anal- gratory responses in response to conflicting chemotactic signals. ysis of spleen sections from a recently immunized mouse sug- Another possibility is that RGS proteins such as RGS13 participate gested prominent membrane localization in germinal center B in desensitization, thereby allowing the cell to cells. We also show that CD40 signaling preferentially regulates

desensitize to a primary chemokine gradient and response to a by guest on September 28, 2021 RGS13 expression whereas signaling through the B cell Ag recep- second one. A likely site of conflicting chemoattractants gradients tor preferentially regulates RGS1 expression. is the light zone of the germinal center, where B cells have several Beside Rgs13, lymphocytes also express Rgs1 and Rgs2. Inter- alternative fates. They may undergo apoptosis, re-enter the dark estingly, these three are clustered together on chromosome 1, both in mice and (30). Rgs1 and Rgs13 are separated by zone, or leave the germinal center. RGS13 may assist the B cells 48 kb while Rgs13 and Rgs2 are located 139 kb apart. There are no in making these decisions. known intervening genes between these loci. A search of the Our results differ slightly from those recently published that mouse EST database using the intervening DNA sequence be- characterized human RGS13 (39). The expression data reported in tween Rgs1 and Rgs13 failed to identify any novel coding regions that study agreed with our results although we have provided more (J. Kehrl, unpublished observation). Determining the DNA regu- detailed information and focused on Rgs13 expression. In contrast, latory regions that control expression of these three genes should to Johnson and Druey (39), we failed to detect any effect of RGS13 provide some insights into the mechanisms that account for their on either signaling through Gs-coupled receptors or G␣qQL sig- overlapping and restricted patterns of expression. The close jux- naling, although we agree on the effect of RGS13 on Gi- and Gq- taposition of Rgs1, Rgs13, and Rgs2 complicates the creation of coupled receptors. In addition, we examined the effects of RGS13 double or triple null animals, likely requiring independent target- on signaling through GPCRs used by B lymphocytes. We ex- ing of each locus. Also, it may pose difficulties for the analysis of pressed both human and mouse RGS13 as Flag-tagged proteins, mice with targeted disruption of one of these loci because of po- whereas Johnson and Druey (39) had difficulties in detecting a tential effects on the adjacent loci or possible compensatory tagged version of human RGS13 and resorted to using a GPF fu- mechanisms. sion protein. Perhaps contributing to our success, we used a C- RGS13 inhibited both the activation of MAPK in response to terminal triple Flag tag. The difference between RGS13-GFP and CXCR4 and CXCR5 signaling in COS cells and the migration of RGS13-Flag may account for some of the differences in signaling CXCR4 expressing CHO cells. In preliminary experiments, we data. We have also repeated some of the signaling experiments have found that the introduction of RGS13 into a human B cell line expressing untagged versions of human and mouse RGS13 with as a GFP fusion protein reduces the migration of the derived cell similar results (G. Shi, unpublished data). Also in contradistinction lines to either CXCL12 or CXCL13 when compared with control to the Johnson and Druey study (39), we could detect endogenous cell lines. In addition, the introduction of an RGS13 cDNA in an RGS13 by immunoblotting. We made affinity-purified anti-peptide antisense orientation into a RGS13-expressing B cell line enhances Abs against three RGS13 peptides: a C-terminal, N-terminal, and the migration of the derived cell lines to the same chemokines an internal peptide. Only those made against the internal RGS13 when compared with control cells (G.-X. Shi, unpublished obser- peptide proved useful for immunoblotting and immunohistochem- vations). Thus, a major role for RGS13 may be to regulate the istry. Finally, while similar to Johnson and Druey (39), we noted responsiveness of germinal center B cells to chemokines. that some cell types express RGS13-GFP within their nuclei, and The Journal of Immunology 2515 we found significant differences in its intracellular localization de- chemoattractant receptors by RGS (regulator of G-protein signaling) family pending upon the cell line chosen and no evidence that B lympho- members. J. Biol. Chem. 273:28040. 16. Moratz, C., V. H. Kang, K. M. Druey, C. S. Shi, A. Scheschonka, P. M. Murphy, cytes express RGS13 within their nuclei. Thus, the intracellular T. Kozasa, and J. H. Kehrl. 2000. Regulator of G protein signaling 1 (RGS1) localization of an RGS-GFP fusion protein in heterologous cells markedly impairs Gi␣ signaling responses of B lymphocytes. J. Immunol. 164: may not help predict the intracellular location of the endogenous 1829. 17. Reif, K., and J. G. Cyster. 2000. RGS molecule expression in murine B lympho- RGS proteins. cytes and ability to down-regulate chemotaxis to lymphoid chemokines. J. Im- In conclusion, Rgs13 possesses one of the most limited patterns munol. 164:4720. 18. Nagasawa, T., S. Hirota, K. Tachibana, N. Takakura, S. Nishikawa, Y. Kitamura, of expression of the known RGSs. Notable among the cell types N. Yoshida, H. Kikutani, and T. Kishimoto. 1996. Defects of B-cell lymphopoi- that express Rgs13 are germinal center B lymphocytes. The com- esis and bone-marrow myelopoiesis in mice lacking the CXC chemokine PBSF/ plicated trafficking pattern that B cells undergo during their transit SDF-1. Nature 382:635. 19. Zou, Y. R., A. H. Kottmann, M. Kuroda, I. Taniuchi, and D. R. Littman. 1998. through the germinal center along with the known ability of RGS Function of the chemokine receptor CXCR4 in and in cerebellar proteins to regulate chemokine receptor signaling implicates development. Nature 393:595. Rgs13 in the control of B cell migration within this microenviron- 20. Forster, R., A. E. Mattis, E. Kremmer, E. Wolf, G. Brem, and M. Lipp. 1996. A putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid ment. Yet germinal center B cells do not rely on Rgs13 alone, but organs and specific anatomic compartments of the spleen. Cell 87:1037. also express Rgs1 and likely other RGSs. Germinal center B cells 21. Gunn, M. D., V. N. Ngo, K. M. Ansel, E. H. Ekland, J. G. Cyster, and likely employ these RGS proteins to provide stop signals, set L. T. Williams. 1998. A B-cell-homing chemokine made in lymphoid follicles activates Burkitt’s lymphoma receptor-1. Nature 391:799. thresholds for responses, and assist in chemokine desensitization, 22. Legler, D. F., M. Loetscher, R. S. Roos, I. Clark-Lewis, M. Baggiolini, and a likely requirement for cells to negotiate complex chemokine B. Moser. 1998. B cell-attracting chemokine 1, a human CXC chemokine ex- networks. pressed in lymphoid tissues, selectively attracts B lymphocytes via BLR1/ Downloaded from CXCR5. J. Exp. Med. 187:655. 23. Forster, R., A. Schubel, D. Breitfeld, E. Kremmer, I. Renner-Muller, E. Wolf, and Acknowledgments M. Lipp. 1999. CCR7 coordinates the primary immune response by establishing We thank Mary Rust for her editorial assistance, Dr. Louis Staudt for functional microenvironments in secondary lymphoid organs. Cell 99:23. 24. Reif, K., E. H. Ekland, L. Ohl, H. Nakano, M. Lipp, R. Forster, and J. G. Cyster. disclosing array data that supported an up-regulation of RGS13 expression 2002. Balanced responsiveness to chemoattractants from adjacent zones deter- in germinal center B cells, and Dr. Anthony Fauci for his continued mines B-cell position. Nature 416:94.

support. 25. MacLennan, I. C. 1994. Germinal centers. Annu. Rev. Immunol. 12:117. http://www.jimmunol.org/ 26. Meyer-Hermann, M., A. Deutsch, and M. Or-Guil. 2001. Recycling probability and dynamical properties of germinal center reactions. J. Theor. Biol. 210:265. References 27. Hargreaves, D. C., P. L. Hyman, T. T. Lu, V. N. Ngo, A. Bidgol, G. Suzuki, 1. Hepler, J. R., and A. G. Gilman. 1992. G proteins. Trends Biochem. Sci. 17:383. Y. R. Zou, D. R. Littman, and J. G. Cyster. 2001. A coordinated change in 2. Neer, E. J. 1995. Heterotrimeric G proteins: organizers of transmembrane signals. chemokine responsiveness guides plasma cell movements. J. Exp. Med. 194:45. Cell 80:249. 28. Hong, J. X., G. L. Wilson, C. H. Fox, and J. H. Kehrl. 1993. Isolation and 3. Dietzel, C., and J. Kurjan. 1987. Pheromonal regulation and sequence of the characterization of a novel B cell activation gene. J. Immunol. 150:3895. SST2 gene: a model for desensitization to pheromone. 29. Sierra, D. A., D. J. Gilbert, D. Householder, N. V. Grishin, K. Yu, P. Ukidwe, Mol. Cell. Biol. 7:4169. S. A. Barker, W. He, T. G. Wensel, G. Otero, et al. 2002. Evolution of the 4. Koelle, M. R., and H. R. Horvitz. 1996. EGL-10 regulates G protein signaling in regulators of G-protein signaling multigene family in mouse and human. Genom- the C. elegans nervous system and shares a conserved domain with many mam- ics 79:177. malian proteins. Cell 84:115. 30. Druey, K. M., B. M. Sullivan, D. Brown, E. R. Fischer, N. Watson, K. J. Blumer, by guest on September 28, 2021 5. Lee, B. N., and T. H. Adams. 1994. Overexpression of flbA, an early regulator of C. R. Gerfen, A. Scheschonka, and J. H. Kehrl. 1998. Expression of GTPase- Aspergillus asexual sporulation, leads to activation of brlA and premature initi- deficient Gi␣2 results in translocation of cytoplasmic RGS4 to the plasma mem- ation of development. Mol. Microbiol. 14:323. brane. J. Biol. Chem. 273:18405. 6. De Vries, L., M. Mousli, A. Wurmser, and M. G. Farquhar. 1995. GAIP, a protein 31. De Vries, L., E. Elenko, J. M. McCaffery, T. Fischer, L. Hubler, T. McQuistan, that specifically interacts with the trimeric G protein G␣i3, is a member of a N. Watson, and M. G. Farquhar. 1998. RGS-GAIP, a GTPase-activating protein with a highly conserved core domain. Proc. Natl. Acad. Sci. USA for G␣i heterotrimeric G proteins, is located on clathrin-coated vesicles. Mol. 92:11916. Biol. Cell 9:1123. 7. Druey, K. M., K. J. Blumer, V. H. Kang, and J. H. Kehrl. 1996. Inhibition of 32. Dulin, N. O., A. Sorokin, E. Reed, S. Elliott, J. H. Kehrl, and M. J. Dunn. 1999. G-protein-mediated MAP kinase activation by a new mammalian gene family. RGS3 inhibits G protein-mediated signaling via translocation to the membrane Nature 379:742. and binding to G␣11. Mol. Cell. Biol. 19:714. 8. Berman, D. M., T. M. Wilkie, and A. G. Gilman. 1996. GAIP and RGS4 are 33. Chatterjee, T. K., and R. A. Fisher. 2000. Cytoplasmic, nuclear, and Golgi lo- GTPase-activating proteins for the G subfamily of G protein ␣ subunits. Cell i calization of RGS proteins: evidence for N-terminal and RGS domain sequences 86:445. as intracellular targeting motifs. J. Biol. Chem. 275:24013. 9. Watson, N., M. E. Linder, K. M. Druey, J. H. Kehrl, and K. J. Blumer. 1996. RGS family members: GTPase-activating proteins for heterotrimeric G-protein ␣-sub- 34. Arai, H., C. L. Tsou, and I. F. Charo. 1997. Chemotaxis in a lymphocyte cell line transfected with C-C chemokine receptor 2B: evidence that directed migration is units. Nature 383:172. ␤␥ 10. Hunt, T. W., T. A. Fields, P. J. Casey, and E. G. Peralta. 1996. RGS10 is a mediated by dimers released by activation of G␣i-coupled receptors. Proc. Natl. Acad. Sci. USA 94:14495. selective activator of G␣i GTPase activity. Nature 383:175. 11. Kozasa, T., X. Jiang, M. J. Hart, P. M. Sternweis, W. D. Singer, A. G. Gilman, 35. Neptune, E. R., and H. R. Bourne. 1997. Receptors induce chemotaxis by re- ␤␥ G. Bollag, and P. C. Sternweis. 1998. p115 RhoGEF, a GTPase activating protein leasing the subunit of Gi, not by activating Gq or Gs. Proc. Natl. Acad. Sci. USA 94:14489. for G␣12 and G␣13. Science 280:2109. 12. Zheng, B., Y. C. Ma, R. S. Ostrom, C. Lavoie, G. N. Gill, P. A. Insel, 36. Kosco-Vilbois, M. H., J. Y. Bonnefoy, and Y. Chvatchko. 1997. The physiology of murine germinal center reactions. Immunol. Rev. 156:127. X. Y. Huang, and M. G. Farquhar. 2001. RGS-PX1, a GAP for G␣S and sorting nexin in vesicular trafficking. Science 294:1939. 37. Choi, Y. S. 1997. Differentiation and apoptosis of human germinal center B- 13. Ansel, K. M., and J. G. Cyster. 2001. Chemokines in lymphopoiesis and lym- lymphocytes. Immunol. Res. 16:161. phoid organ development. Curr. Opin. Immunol. 13:172. 38. Jones, D. 2002. Dismantling the germinal center: comparing the processes of 14. Kehrl, J. H. 1998. Heterotrimeric G protein signaling: roles in immune function transformation, regression, and fragmentation of the lymphoid follicle. Adv. Anat. and fine-tuning by RGS proteins. Immunity 8:1. Pathol. 9:129. 15. Bowman, E. P., J. J. Campbell, K. M. Druey, A. Scheschonka, J. H. Kehrl, and 39. Johnson, E. N., and K. M. Druey. 2002. Functional characterization of the G E. C. Butcher. 1998. Regulation of chemotactic and proadhesive responses to protein regulator RGS13. J. Biol. Chem. 1:1.