View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS 29237 FEBS Letters 579 (2005) 1039–1046

Zizimin2: a novel, -related Cdc42 guanine exchange factor expressed predominantly in lymphocytes

Akihiko Nishikimia,1, Nahum Mellerb,1, Natsuko Uekawaa, Ken-ichi Isobec, Martin A. Schwartzb, Mitsuo Maruyamaa,* a Laboratory of Experimental Animal Model Research, National Institute for Longevity Sciences, National Center for Geriatrics and Gerontology, 36-3, Gengo, Morioka-Cho, Obu-city, Aichi 474-8522, Japan b Departments of Microbiology and Biomedical Engineering, Cardiovascular Research Center and Mellon Prostate Research Institute, University of Virginia, Charlottesville, VA 22908, USA c Department of Mechanism of Aging, National Institute for Longevity Sciences, National Center for Geriatrics and Gerontology, 36-3, Gengo, Morioka-Cho, Obu-city, Aichi 474-8522, Japan

Received 18 November 2004; revised 28 December 2004; accepted 4 January 2005

Available online 13 January 2005

Edited by Beat Imhof

filing techniques to alter their expression during transi- Abstract A novel superfamily of guanine nucleotide exchange factors for Rho GTPases includes DOCK180 and zizimin1. tion through GCs [3,4]. The zizimin subfamily includes three of which only zizi- The Rho family of low molecular weight GTPases are piv- min1 has been cloned. We report here the cloning of zizimin2, otal regulators of the actin cytoskeleton and many other cellu- identified in a screen for genes enriched in germinal center B lar processes [5–7]. Rho include twenty-two human cells. Zizimin2 and zizimin1 have similar primary structures genes that can be divided to six or more subfamilies [8]. The and both proteins bound and activated Cdc42 but not the three major subfamilies are the RhoA, Rac1 and Cdc42 pro- Cdc42-related proteins TC10 or TCL. Their tissue distributions teins. Cdc42 mediates cell polarity, expression, are distinct, however, with zizimin2 expressed predominantly in progression and cell–cell contacts [9,10]. In lymphocytes, lymphocytes and an opposite pattern for zizimin1. Zizimin3 Cdc42 was shown to be essential for chemokine-induced T cell was also analyzed and showed distinct GTPase specificity and migration and to regulate T cell polarization towards antigen- tissue distribution. 2005 Federation of European Biochemical Societies. Published presenting cells [11]. by Elsevier B.V. All rights reserved. Like other GTPases, Rho proteins are active when bound to GTP and inactive when bound to GDP. Conversion of the Keywords: GEF; Rho; ; DOCK10; DOCK11; CZH2 GDP-bound proteins to the active state is catalyzed by guanine nucleotide exchange factors (GEFs). The classical GEFs for Rho GTPases comprising a family of over sixty proteins [12], share a common motif, designated the Dbl-homology (DH) domain that mediates nucleotide exchange [13]. Additionally, 1. Introduction a second family of GEFs that utilize an alternative Rho- GEF domain was recently discovered. Why two different do- During versatile adaptive immune responses, germinal cen- mains evolved is an open question but the inflation in Rho ters (GCs) are formed in peripheral lymphoid tissues. GCs play GEFs may reflect the need for selective activation of Rho pro- central roles in differentiation of late B cells to antibody-secret- teins by different signaling pathways and under diverse ing plasma cells or to antigen-specific memory B cells. The dif- conditions. ferentiation processes include immunoglobulin class-switching, This new family of Rho-GEFs was named CZH because it in- somatic hypermutation and affinity maturation through B cell cludes CDM (Ced-5, DOCK180, Myoblast city) proteins that receptor-dependent signals by encountering the cognate anti- activate Rac, and zizimin1 that activates Cdc42 [14–17]. Using gens [1,2]. During these processes, the activated B lymphocytes information from genomic, and transcript databases, migrate from the central to the marginal zone of the GC. A the family was predicted to include 11 mammalian genes, many number of genes were recently shown by pro- of which have not been cloned [15,16]. Based on , CZH proteins can be divided into one subfamily re- lated to CDM proteins and a subfamily related to zizimin1. The two groups share two conserved domains that we named CZH1 *Corresponding author. Fax: +81 562 46 8459. and CZH2 for CDM-zizimin homology 1 and 2, respectively E-mail address: [email protected] (M. Maruyama). [16]. The CZH2 domain, also called DOCKER or DHR2, is 1 These authors contributed equally to this work. a GEF domain that shows no sequence homology to DH do- mains [14–16]. The function of the CZH1 domain, also named Abbreviations: GC, germinal center; BCR, B cell receptor; GEF, gu- DHR1, remains unknown [15,16]. Based on sequence similar- anine nucleotide exchange factor; CDM, CED-5, DOCK180, Myo- ity, domain structure and phylogenetic analysis the zizimin-re- blast city; CZH, CDM-zizimin homology; GFP, green fluorescence lated proteins can be divided to two groups. The first, named protein; PNA, peanuts agglutinin; PH, pleckstrin homology; GST, glutathione S-transferase; PAK, p21-activated kinase; PBD, p21 bin- zizimin proteins or DOCK-D, includes three proteins ding domain; PBL, peripheral blood lymphocytes zizimin1/DOCK9, zizimin2/DOCK11 and zizimin3/DOCK10,

0014-5793/$30.00 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2005.01.006 1040 A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046 of which only zizimin1 had been cloned. The second group, nin (PNA; Vector Laboratories), followed by streptavidin magnetic high low named zir for zizimin-related, or DOCK-C, includes zir1/ beads. PNA or PNA B cells were obtained by auto MACS sort- ing. Total RNA preparation, cDNA synthesis and PCR-select cDNA DOCK6, zir1/DOCK7 and zir3/DOCK8 of which only high subtraction were carried out as described previously [22], the PNA DOCK8 was cloned [15,16,18]. B cells cDNA was used as tester and the PNAlow B cells cDNA was Using subtractive hybridization, in this study we cloned zizi- used as driver. min2 as a gene expressed at higher levels in GC B cells com- pared to non-GC B cells. Zizimin2 primary structure is 2.4. Cloning of zizimin2 similar to zizimin1, exhibiting 60% identity in amino acid se- Sequences obtained from the subtracted library were used for Gen- quence. We show that zizimin2 and zizimin1 bind selectively bank searches. A 99 bp clone completely matched sequences of the putative zizimin2/DOCK11 protein. The hypothetical FL cDNA se- to Cdc42 but not Rac Rho or the Cdc42-related proteins quence of zizimin2 was generated by combining sequences of overlap- TC10 and TCL. Overexpression of zizimin2 activates Cdc42 ping Genbank expressed sequence tag (EST) clones. To obtain the in cells. Zizimin2 is expressed predominantly in lymphocytes authentic sequence of zizinin2 as expressed in B cells, primers from in contrast to zizimin1 which is enriched in non-hematopoietic the putative beginning and end of zizimin2 open reading frame (50- 0 0 tissues. The GTPase specificity and tissue distribution of zizi- GAA AGT GCG CAA ATT CAC C-3 and 5 -TCA TAC TTC CGC GTA TCT TG-30) were designed and PCR was performed using min3 were also analyzed. murine splenic B cells cDNA as template. Consequently, a single prod- uct 6.2 kb long was amplified. Three independently amplified clones were sequenced to be verified. Finally the untranslated 50 region of Zizimin2 was determined using the GeneRacer rapid amplification of 2. Materials and methods cDNA end (RACE) kit (Invitrogen) with the following gene specific primers, 50-TGG CAA CAT CC GAA AGT CCC CAG A-30 and 2.1. Vectors and bacterial expression 50-TGG GCC CTC TTT TCA CAT CCT CTG G-30. The sequence pHis full-length (FL) zizimin2 was generated by inserting the PCR- of zizimin2 was deposited at the GenBank database with the accession amplified full length zizimin2 cDNA into pcDNA-4-His-max (Invitro- number AB116935. gen) using the NotI site. To construct a vector for expression of green fluorescence protein (GFP) fusion proteins with additional tags (pGT), 2.5. GTPases binding and activation assays a fragment coding for myc-flag-HA tags was cloned into XhoI/EcoRI Transfected cells were washed twice in cold PBS and lysed 10 min site of the pEGFP-C1 vector (Clontech). Zizimin2 (669-E), Zizimin2 on ice in 50 mM Tris, pH 7.5, 500 mM NaCl, 1% Triton X-100, 5 (1516-E) and Zizimin1 (1512-E) were amplified by PCR and cloned mM EDTA and 1% protein inhibitors mixture (Sigma # P8340). into pGT. pEF-oncoVav1 was described elsewhere [19]. HA-tagged The EDTA serves to deplete GTPases from by chelating human zizimin1 CZH2 (Zizimin1 1512-E) was described before [16]. Mg2+ ions that are essential cofactor for nucleotide binding. Lysates For HA-tagged human zizimin2 CZH2 construction, EST Genbank were cleared by 10 min centrifugation at 16 000 · g. A portion of this accession number H15608 was amplified by PCR using the primers total lysate was reserved and the rest incubated for 1 h at 4 C with 50-ATA GGT ACC ACC AAA AGG AAA ACC TTT TTG A-30 25 lg GST fused GTPases bound to glutathione agarose beads (Sig- (sense) and 50-AAA GAA TTC TCA CAC TTC AGC GTA TCT ma). The beads were washed 3 times in lysis buffer (LB) and either TGG G-30 (antisense), the product was digested with KpnI and EcoRI eluted directly in SDS–PAGE loading buffer (Fig. 2B) or eluted first and cloned into KpnI/EcoRI site of pEF4-HA-Kpn [16]. For HA- in LB lacking EDTA and containing 5 mM MgCl and 100 lM tagged human zizimin3 CZH2, EST Genbank accession number 2 GTPcS followed by a elution of the beads in SDS–PAGE loading BM905401 was amplified by PCR using the primers 50-ATA GGT buffer (Fig. 3). For assaying Cdc42 activation, we used the Cdc42 ACC AAC AAG CAG AAG TCA ATT GTC CG-30 (sense) and 50- activation assay kit (Cytoskeleton) according to the manufacturerÕs AAG GAT CCT CAG ACT TCA GCA CTA GAT GAG-30 (anti- instructions. sense). The product was digested with KpnI and BamHI and cloned into appropriate sites of the modified pEF4-HA-Kpn vector which possessed an additional BamHI site just prior to SfuI site. Glutathione 2.6. RNA analysis S-transferase (GST)-TC10 in the bacterial expression vector pGEX2T For Northern blotting, total RNA from various mouse tissues was was expressed and purified as described [20]. For GST-TCL protein, isolated with TriZol (Invitrogen) and then poly-A+ RNA was purified human TCL (Genbank Accession No. AJ276567) in pcDNA3.1, ob- with lMACS mRNA isolation kit (Miltenyi Biotec). Poly-A+ RNA tained from UMR cDNA resource center at the university of Mis- (0.5 lg) was separated by electrophoresis on 1.5% agarose, 6% formal- souri-Rolla (Rolla, MO), was subcloned into pGEX4T1 using dehyde gels and transferred onto Hybond-N+ membrane (Amersham), EcoRI/XhoI sites, expressed and purified as described for TC10 [20]. which was hybridized with [32P]-labeled PCR amplified murine zizi- min2, zizimin1, zizimin3 or b-actin cDNA probes. The following prim- ers were used for probe amplification: zizimin2, 50-GAC AAA ATA 2.2. Cell culture and gene introduction TTG TTA TCC-30 (sense) and 50-GCT GGC AAT GAT GGA BAF0/3 cells were maintained as described previously [21]. A20, EL- AGA AT-30 (antisense); zizimin1, 50-AAG GTC AAC CCG AAG 4, WEHI-231 cell lines and mouse primary splenic cells were cultured GAT CT-30 (sense) and 50-GGC TCC TTC TCT CTC CCA TT-30 in RPMI 1640 with 50 lM 2-mercaptoethanol. Other cell lines includ- (antisense); zizimin3, 50-TGG CCA AAG TCG AAA AAG TC-30 ing COS-1 and COS-7 cells were cultured in DMEM. All culture media (sense) and 50-GGC TCC ATC GTC ATC TGA AT-30 (antisense). was supplemented with 10% FCS, 50 units/ml penicillin and 50 lg/ml b-actin probe was purchased from Clontech. Probe hybridization streptomycin. Transient transfection of COS-1 cells was performed was performed with Perfecthyb (Sigma) as recommended by the with Fugene6 reagent (Roche Diagnostics). COS-7 cells were transfec- manufacturer. ted using the Effectene Reagent (Qiagen). Cells were harvested at 24– Quantitative real-time RT-PCR was carried out as described [22]. 48 h post transfection. Each RNA sample was analyzed in parallel for b-actin to control for differences in starting RNA amounts. The following gene specific prim- 2.3. Generation of subtracted cDNA library ers were used: human b-actin, 50-CCA AGG CCA ACC GCG AGA For GC and non-GC B cells preparation, C57bl/6J mice were immu- AG-30 (sense) and 50-GCC AGA GGC GTA CAG GGA TA-30 (anti- nized with 4-hydroxy-3-nitrophenyl acetyl coupled to chicken globulin sense); mouse and human zizimin2, 50-TTG CCT TTT ATG GCC (NP-CG) and single splenocyte suspensions were prepared from AGT CT-30 (sense) and 50-GAG CGA ATT TTG GAT CAA GC-30 7 days after priming. CD19+ B cells were isolated by auto (antisense); mouse zizimin1, 50-CCA AAT TCC CTG CAC ATC MACS (Miltenyi Biotec) by incubation with FITC-conjugated anti- TT-30 (sense) and 50-GGC TCC TTC TCT CTC CCA TT-30 (anti- CD19 antibody and anti-FITC magnetic beads, followed by removal sense); mouse b-actin, 50-CTA AGG CCA ACC GTG AAA AG-30 of the beads by anti-FITC isolation kit. The isolated CD19+ cells were (sense) and 50-ACC AGA GGC ATA CAG GGA CA-30 (antisense). found to be positive for B220 as well by FACS analysis (Becton Dick- Especially, for GC and non-GC B cell cDNA preparation in Fig. inson). The cells were then incubated with biotinylated peanut aggluti- 4A, four-color sorting (B220, CD19, PNA, and GL-7) was performed A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046 1041 on a BD FACS Vantage in Nippon Becton Dickinson Company, Ltd. and zizimin3 are located on 13 and 2, respec- Bioscience Kobe Laboratory and we prepared cDNAs using SMART tively. The genomic organization of murine zizimin2 was deter- PCR cDNA synthesis method (BD Clontech). mined by database search. The predicted zizimin2 gene spans 187,774 bp comprising of 53 exons (Fig. 1A).

3. Results 3.2. Zizimin2 activates Cdc42 via the CZH2 domain CZH proteins were shown previously to function as direct 3.1. Cloning of mouse zizimin2 and sequence analysis activators of Rho proteins [14–16]. The interaction is mediated To identify genes potentially involved in late B cell differen- via a novel GEF domain named CZH2, DOCKER or DHR2. tiation, we looked for genes that are upregulated in GC B cells For nucleotide exchange, GEFs first bind with low affinity to when compared to non-GC B cells. C57bl/6J mice were immu- the GDP-bound GTPase and induce dissociation of GDP from nized with NP-CG and single-splenocyte suspensions were pre- this complex, which leads to formation of a higher affinity pared. GC B cells (CD19+ B220+ PNAhigh) and non-GC B cells intermediate. This intermediate is then dissociated by binding (CD19+ B220+ PNAlow) were purified and total RNA was pre- of GTP [24]. GEFs can therefore be distinguished from other pared. cDNA was synthesized and a library prepared of GC-B GTPase interacting proteins by their ability to bind preferen- cells cDNA subtracted by non-GC B cells cDNA. Genbank tially to the nucleotide-depleted (ND) state compared to searches using a 99 bp sequence derived from the library GTP- or GDP-bound states [16,25]. We therefore studied po- matched sequences of the putative zizimin2/DOCK11 protein tential interaction of zizimin2 with Cdc42, Rac1 or RhoA by (Fig. 1A). Quantitative RT-PCR analysis was then employed assaying binding to the nucleotide depleted GTPases using to confirm higher expression of zizimin2 RNA in GC B cells truncated forms of zizimin2 described in Fig. 2A. As shown (2.6-fold; see Fig. 4A). The hypothetical zizimin2 open reading in Fig. 2B, a zizimin2 fragment containing the C-terminal frame (ORF) was identified and primers from the potential two thirds of the protein (Zizimin2669-E) bound Cdc42 but beginning and end of the ORF were designed. These were used not Rac1 or RhoA. The same GTPases specificity was previ- to amplify murine splenic B cells cDNA. A 6.2 kb fragment ously shown for zizimin1 [16]. A shorter fragment containing was obtained and cloned into vector. Finally, the sequence only zizimin2 CZH2 domain (Zizimin21512-E) was sufficient coding for the 50 untranslated region was identified by 50- to mediate Cdc42 binding (Fig. 3). The results therefore impli- RACE. cate zizimin2 as a potential Cdc42-GEF. The complete sequence obtained bears 6346 nucleotides cod- To examine whether zizimin2 can activate Cdc42 in cells, FL ing for a protein of 2073 amino acids with a calculated molec- zizimin2 was overexpressed in COS-1 cells and GTP-loaded ac- ular weight of 238 kDa. Murine zizimin2 shows high sequence tive Cdc42 was precipitated using the p21 binding domain similarity to human zizimin1, which was cloned previously (PBD) of p21-activated kinase (PAK). As shown in Fig. 2C [16]. Zizimin2 possesses an N-terminus pleckstrin homology (left panel), zizimin2 expression induced substantial activation (PH) domain (predicted by the Pfam program [23]), CZH1 do- of endogenous Cdc42. Expression of the isolated CZH2 do- main and CZH2 domain (Fig. 1A). The predicted human zizi- mains of zizimin2 or zizimin1 (as GFP fusions), was sufficient min2 (Genbank GI: 40068509) also contains 2073 amino acids to induce Cdc42 activation, further supporting the function of and displays 96% identity and 98% similarity to the murine this module as a novel GEF domain (Fig. 2C right panel). Sim- protein. ilar expression of GFP-zizimin2 CZH2, GFP-zizimin1 CZH2 A third zizimin isoform, named zizimin3 or DOCK10, was or the control GFP was confirmed by FACS analysis (data identified in human by searching protein and genomic dat- not shown). We note that previous studies in NIH-3T3 or abases with zizimin1 and DOCK180 [15,16]. To compare zizi- 293T cells failed to detect Cdc42 activation by the short min2 to the other two isoforms, we searched protein and EST CZH2 domain [15,16]. The higher expression level in COS-1 databases for the complete sequences of murine zizimin1 and cells may account for the discrepancy. zizimin3. Interestingly, several areas of alternative splicing were identified. Both zizimin1 and zizimin3 have alternative 3.3. Analysis of GTPase selectivity amino and carboxy termini while three additional sites of alter- Rho proteins can be grouped into subfamilies based on native splicing were identified in zizimin1 (Fig. 1C). All five structural and functional similarities. Based on the ability to alternatively spliced forms of murine zizimin1 are conserved promote formation of filopodia, five mammalian genes are in- in human. Database searching identified no splice isoforms cluded in the Cdc42-related subfamily, namely: Cdc42, TC10, of zizimin2 (murine or human). We assembled the murine zizi- TCL, Wrch-1 and Wrch-2 [8]. Wrch-1 and Wrch-2, however, min1 and zizimin3 isoforms that exhibit the highest similarity are structurally and phylogenetically quite distinct from the to zizimin2 and aligned the sequences (Fig. 1B). Murine zizi- other three members [8]. To further define the selectivity of min2 and zizimin1 are 60% identical and 75% similar in amino zizimin isoforms for different GTPases, we compared binding acid sequence while zizimin3 displays 50% identity and 67% zizimin1, 2 and 3 to Cdc42, TC10 and TCL. To this end, we similarity to zizimin2. Conservation is evident along most of cloned the CZH2 domain of zizimin3 as the FL protein has the sequence but is highest within the CZH2 domain with not been cloned yet. Lysates of COS-7 cells expressing the 71% identity/85% similarity between zizimin2 and zizimin1 CZH2 domains of zizimin1, 2 or 3 were incubated with nucle- and 66% identity/83% similarity between zizimin2 and zizi- otide-depleted GTPases bound to agarose beads, which were min3. Overall, the zizimin2 and zizimin1 sequences are more washed and eluted with GTP. The GTP elution served to related to each other than to zizimin3. Interestingly, the zizi- ascertain the specificity of the GEFs interaction with the nucle- min2 genes from both mouse and human are located on chro- otide-depleted GTPases, which should be reversed by GTP mosome X. Mouse zizimin1 and zizimin3 are located on [24]. As shown in Fig. 3 (left panel), both zizimin2 and zizi- chromosomes 14 and 1, respectively, while human zizimin1 min1 bound Cdc42 but not TC10 or TCL. Zizimin2 CZH2 1042 A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046

Fig. 1. Primary structure of zizimin2 and its isoforms. (A) Schematic representation of murine zizimin2 protein and gene structures. Top diagram: protein structure, the location and boundaries of the PH domain, CDM-zizimin homology 1 (CZH1) domain and the CDM-zizimin homology 2 (CZH2) domain are displayed. The amino acid sequence derived from the cDNA fragment cloned by subtractive hybridization is shown above the CZH2 domain. Bottom diagram: gene structure, each exon is represented by a vertical bar, exons encoding the PH, CZH1 and CZH2 domains are indicated. Intron 1 is approximately 55 kb long. (B) Comparison of murine zizimin2, zizimin1 and zizimin3 amino acid sequences. The sequences were aligned using the Multalin program [31]. Residues in black shading are identical in all proteins and in bold are similar in 2 of the 3 proteins. The PH, CZH1 and CZH2 domains are boxed. The zizimin1 sequence was assembled from Genbank GI numbers 32469704, 51768647 and 314197757. The zizimin3 sequence was assembled from GI numbers 51705170 and 28972351 according to human sequence GI: 32469767. (C) Alternative splicing of zizimin proteins. The zizimin1 and zizimin3 sequences used in (B) are presented schematically and areas of alternative splicing are indicated below. The length of the alternative splices in amino acids (aa) is indicated. The boundaries of the splicing areas are indicated above the diagrams. (a) Replacement of the first 42 amino acids by MGCTTSVILFKGIRTVFERNCAYMCKQPGESNALEYTAYNWSKEDSELSIAFC according to Genebank GIs 28972594 and 32469704. (b) Deletion of amino acids 1353–1375 according to GI:28972594. (c) Insertion of EADLALQREPPAF- PYSHSTCQRKSWG after amino acid 1673 according to GIs 31419757, 34784430 and 28374168. (d) Insertion of AAQYQFTDSETDVE after amino acid 1788 according to GIs 28972594 and 26337589. (e) Replacement of the carboxy terminus (amino acids 2051-91) by MG according to GIs 28972594 and 26337589. (f) Replacement of the amino terminus by MESRLLWASGNPSASASDQGIPN according to GI 26334085. (g) Replacement of the carboxy terminus by LCRGPCLYSFCASVSSISLSTVSKSDYGQGRPSKVRSGATLHHTCN according to GI 289723. A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046 1043

Fig. 1 (continued)

Fig. 2. Zizimin2 binds and activates Cdc42. (A) Schematic representation of the zizimin fragments. Boxes indicate the functional domains as described. (B) Zizimin2 binds Cdc42. COS-1 cells were transfected with pGT-Zizimin2(669-E) and equal amount of cell lysates were incubated with nucleotide-depleted Cdc42, Rac1 or RhoA attached to beads. The beads were washed and bound proteins eluted in SDS sample buffer, resolved by SDS–PAGE and detected by Western blotting with anti-HA (upper panel). A portion of the elutes was resolved by SDS–PAGE and stained by coomassie to detect the GST-Rho proteins (lower panel). Molecular weight standards in kDa are indicated on the left. (C) Cdc42 activation by zizimin2. pcDNA4-His-max (vector) or full-length (FL) zizimin2 in pcDNA4-His-max (left panel); pGT (GFP), pGT-Zizimin2(1516-E) or pGT- Zizimin1(1512-E) (right panel), were transfected into COS-1 cells. Activation of endogenous Cdc42 was assayed by precipitating GTP-loaded Cdc42 with bead-bound GST-PAK1-PBD, and detection of Cdc42 in pull downs (upper panels) or total cell lysates (lower panels) by western blotting with anti-Cdc42. Molecular weight in kDa is indicated on the left. was detected as two protein bands in total lysates, only the nel). Zizimin1 was shown previously to promote filopodia lower band, however, bound Cdc42. For zizimin3, which has formation, indicating that its target GTPase is from the not previously been analyzed, we also tested binding to Rac1 Cdc42 subfamily [16]. These results strongly support the idea and RhoA. Under the same binding conditions, none of the that Cdc42 is the target for zizimin1 and zizimin2. GTPases showed consistent binding to zizimin3 CZH2 domain (data not shown), however, when zizimin3 levels were in- 3.4. Tissue distribution of zizimin2 and its isoforms creased by fourfold, nucleotide-sensitive binding to Cdc42, Having cloned zizimin2 by subtractive hybridization as a TCL and to lesser extent TC10 and Rho was evident. As a po- gene potentially upregulated in GC B cells, we wished to com- sitive control, we used oncoVav1 which was shown to interact pare expression of zizimin2 in GC versus non-GC B cells. The in vitro with Rac, Rho and Cdc42 [26]. OncoVav1 indeed two cell populations were isolated from mouse and zizi- bound Rac, Rho and Cdc42 as well as TC10 and TCL, demon- min2 expression was measured by quantitative RT-PCR. The strating activity of the GTPases in this assay (Fig. 3 right pa- results demonstrated that zizimin2 is indeed enriched in GC 1044 A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046

Fig. 3. Selectivity of zizimin proteins for Cdc42-related GTPases HA-tagged CZH2 domains of human zizimin proteins (Zizimin11512-E or the corresponding regions in zizimin2 and ziziminin3) were expressed in COS-7 cells. Cell lysates were incubated with nucleotide-depleted Rho proteins attached to beads. Bound proteins were eluted in buffer containing 5 mM MgCl2 and 100 lM GTPcS. The beads underwent a second elution in SDS sample buffer to release the GST–GTPase fusion proteins. The GTPcS elutes were resolved by SDS–PAGE and zizimin proteins were detected using Western blotting with anti-HA (left top panel). Total lysates were analyzed in parallel (lysate). The secondary elutes were resolved by SDS–PAGE and the GST-Rho proteins detected by Coomassie staining (bottom panels). As a positive control, binding of oncoVav1 to the GTPases was assayed in parallel and detected by Western blotting with anti-myc (top right panel). For zizimin1, zizimin2 and Vav the portions of the total lysate loaded were 5.3% of the amount used in the pull downs. For zizimin3 fourfold more lysate was used, hence the lysate portion loaded represent 1.3% of the amount used in the pull down.

B cells (2.6-fold increase over non-GC B cells; Fig. 4A left pa- malignancies including acute lymphoblastic leukemia, adult T nel). Comparison of zizimin2 expression in splenic mature B cell leukemia, non-HodgkinÕs lymphoma or multiple and T lymphocytes revealed a similar expression levels (Fig. myeloma. 4A right panel). To gain further insight into the cell-type specificity of zizi- min2 expression, we examined murine cell lines from various 4. Discussion origins by quantitative RT-PCR. Zizimin2 mRNA level in the three B cell lines tested (BAF0/3, WEHI-231 and A20) The zizimin subfamily of non-conventional Rho-GEFs in- was 5–7-fold higher than in NIH-3T3 fibroblasts or Neuro- cludes three genes named zizimin1/DOCK9, zizimin2/ 2A neuroblastoma cells, while C2C12 myoblasts exhibited an DOCK11 and zizimin3/DOCK10 [15,16]. We report here the intermediate level (Fig. 4B). Expression in EL-4 T lymphoma cloning of zizimin2 in a screen for genes enriched in GC B cells. cells was about twofold lower than in the B cell lines (Fig. 4B). Zizimin2 and zizimin1 are similar along their entire length with To examine zizimin2 expression in vivo, we used quantita- 60% identity and 75% similarity in amino acid sequence (Fig. tive RT-PCR to analyze normal mouse tissues and cells. 1B). While the target GTPase for both zizimin1 and zizimin2 Expression of zizimin2 was evident in spleen, , bone appear to be Cdc42 (Figs. 2B and 3), the tissue distribution marrow and to higher degree in purified peripheral blood lym- of the two isoforms is remarkably different. Zizimin2 is ex- phocytes (PBL) but was not detected in heart, brain or kidney pressed predominantly in lymphocytes and zizimin1 is pre- (Fig. 4C). Zizimin1 expression showed an opposite pattern, dominantly expressed in non-hematopoietic tissues and cells with high expression in brain and kidney compared to hemato- (Fig. 4). This may suggest that zizimin2 was adapted to engage poietic tissues and cells (Fig. 4C). To further substantiate these in lymphocyte-specific signaling pathways. A similar situation findings and to compare expression of the three zizimin iso- exists in the DOCK180 subfamily, where DOCK180 is ex- forms in a more rigorous way, we studied mRNA expression pressed predominantly in adherent cells and DOCK2 in hema- in tissues of 4-week-old mice by Northern blotting (Fig. 4D). topoietic cells [27,28]. Nevertheless, both DOCK180 and The results confirmed predominant expression of zizimin2 in DOCK2 activate Rac proteins and were shown to regulate cell hematopoietic organs although low expression level was de- migration [29,30]. We should note that while in previous study tected in some of the non-hematopoietic tissues. Zizimin1 high expression level of zizimin1 was observed in human heart was detected most strongly in brain, lung, and kidney as com- and skeletal muscle tissues [16], in this study, done with mouse pared to heart, liver, skeletal muscle or hematopoietic organs, tissues, very little or no expression was detected in these tissues and zizimin3 was enriched in brain, lung, spleen and thymus. (Fig. 4C and D). Taken together, these results document predominant expres- To gain insight into the GTPase specificity and expression sion of zizimin2 in B and T lymphocytes while zizimin1 is ex- pattern for the complete zizimin1 subfamily, we also cloned pressed mainly in non-hematopoietic tissues and cells. the CZH2 domain of zizimin3 and studied zizimin3 tissue dis- Finally, zizimin2 expression level in normal lymphocytes tribution. The zizimin3 CZH2 domain interacted with Rho versus lymphoma cell lines of different differentiation/diag- GTPases as expected, that is, it bound to the nucleotide-free nostic stage was measured by RT-PCR (Fig. 4E). As a gen- GTPases and was eluted by GTP. Binding, however, appeared eral rule zizimin2 expression was similar in normal and to be with lower affinity and was more promiscuous (Fig. 3). transformed cells. Zizimin2 expression was reduced in some The tissue distribution of zizimin3 is distinct from those of zizi- of the lymphomas compared to normal cells, however, this ef- min1 and zizimin2 and partially overlaps with both (Fig. 4D). fect was not consistent and did not correlate with any of the We hypothesize that the target for zizimin3 may be another A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046 1045

Fig. 4. Expression profiles of zizimin2 and its isoforms. (A) Expression of zizimin2 mRNA in C57Bl/6J mice splenic lymphocytes. Left panel: Quantitative real time RT-PCR was carried out on GC B cells (B220+, CD19+, PNAhigh, GL-7+) and non-GC B cells (B220+, CD19+, PNAlow, GL- 7). In each experiment, the increase in zizimin2 mRNA level over the level in non-GC B cells was calculated. Means ± S.D. values of three experiments are presented. Right panel: Analysis of splenic mature B (B220+) and T cells (CD4+ or CD8+) was done as above, the increase in zizimin2 level over the level in B cells is presented. (B) Expression of zizimin2 mRNA in murine cell lines of different origin. Real time RT-PCR analysis was done as in (A). The bars represent fold increase of zizimin2 mRNA level over the level in NIH-3T3 cells. (C) Expression of zizimin2 and zizimin1 mRNAs in murine tissues. RT-PCR analysis of 4-week-old C57Bl/6J mice tissues was done at the linear range of cycle numbers. The products were resolved on 2.0% agarose gels and stained by ethidium bromide. (D) Tissue distribution of murine zizimin isoforms. mRNAs from mice tissues as in (C) was analyzed by Northern blotting. A common membrane was hybridized in succession with 32P-labeled probes from a unique region of each zizimin isoform or b-actin. RNA molecular size in kb is indicated on the left. (E) Expression of zizimin2 RNAs in human lymphoma cell lines. Quantitative real time RT-PCR was done as in (A). The bars represent the increase of zizimin2 mRNA level over the level in normal peripheral blood lymphocytes. The abbreviations below the cell line indicate the diagnosis and lineage of lymphoma. ALL; acute lymphoblastic leukemia, ATL; adult T cell leukemia, NHL; non-HodgkinÕs lymphoma, MM; multiple myeloma. (T); T lymphoma and (B); B lymphoma.

Rho protein that was not investigated. The cellular signaling [2] Kelsoe, G. (1996) Life and death in germinal center (redux). pathways controlled by zizimin proteins are currently un- Immunity 4, 107–111. known. Future studies will address this intriguing question. [3] Feldhahn, N., Schwering, I., Lee, S., Wartenberg, M., Klein, F., Wang, H., Zhou, G., Wang, S.M., Rowley, J.D., Hescheler, J., Kronke, M., Rajewsky, K., Kuppers, R. and Muschen, M. (2002) Acknowledgements: We are grateful to Mr. H. Kuriki for his excellent Silencing of B cell receptor signals in human naive B cells. J. Exp. technical assistance in cell sorting. We thank Drs. S. Iida and R. Ueda Med. 196, 291–305. for providing RNA of various human lymphocyte cell lines. We also [4] Klein, U., Tu, Y., Stolovitzky, G.A., Keller, J.L., Haddad, J.J., thank Dr. I. Macara (Center for University of Virginia, Miljkovic, V., Cattoretti, G., Califano, A. and Dalla-Favera, R. Charlottesville, VA) for GST-TC10 expression vector. This work was (2003) Transcriptional analysis of the B cell germinal center supported in part by grants-in-aid for scientific research from the Min- reaction. Proc. Natl. Acad. Sci. USA 100, 2639–2644. istry of Education, Culture, Sports, Science and Technology of Japan, [5] Etienne-Manneville, S. and Hall, A. (2002) Rho GTPases in cell the fund for Comprehensive Research on Aging and Health, a research biology. Nature 420, 629–635. grant for Longevity Sciences (16A-2) from the Ministry of Health, La- [6] Erickson, J.W. and Cerione, R.A. (2004) Structural elements, bor and Welfare of Japan, and USPHS Grant RO1 47214 (to M.A.S.) mechanism, and evolutionary convergence of Rho protein-guan- A. N. is a research fellow of the Japan Society for the Promotion of ine nucleotide exchange factor complexes. Biochemistry 43, 837– Science. 842. [7] Aznar, S., Fernandez-Valeron, P., Espina, C. and Lacal, J.C. (2004) Rho GTPases: potential candidates for anticancer therapy. Cancer Lett. 206, 181–191. References [8] Wennerberg, K. and Der, C.J. (2004) Rho-family GTPases: itÕs not only Rac and Rho (and I like it). J. Cell Sci. 117, 1301–1312. [1] MacLennan, I.C. (1994) Germinal centers. Annu. Rev. Immunol. [9] Bishop, A.L. and Hall, A. (2000) Rho GTPases and their effector 12, 117–139. proteins. Biochem. J. 348, 241–255. 1046 A. Nishikimi et al. / FEBS Letters 579 (2005) 1039–1046

[10] Etienne-Manneville, S. (2004) Cdc42 – the centre of polarity. J. cellular senescence of mouse embryonic fibroblasts. FEBS Lett. Cell. Sci. 117, 1291–1300. 575, 30–34. [11] Cantrell, D.A. (2003) GTPases and T cell activation. Immunol. [23] Bateman, A., Birney, E., Durbin, R., Eddy, S.R., Finn, R.D. and Rev. 192, 122–130. Sonnhammer, E.L. (1999) Pfam 3.1: 1313 multiple alignments and [12] Schmidt, A. and Hall, A. (2002) Guanine nucleotide exchange profile HMMs match the majority of proteins. Nucleic Acids Res. factors for Rho GTPases: turning on the switch. Genes Dev. 16, 27, 260–262. 1587–1609. [24] Cherfils, J. and Chardin, P. (1999) GEFs: structural basis for their [13] Cerione, R.A. and Zheng, Y. (1996) The Dbl family of oncogenes. activation of small GTP-binding proteins. Trends Biochem. Sci. Curr. Opin. Cell Biol. 8, 216–222. 24, 306–311. [14] Brugnera, E., Haney, L., Grimsley, C., Lu, M., Walk, S.F., [25] Hart, M.J., Sharma, S., elMasry, N., Qiu, R.G., McCabe, P., Tosello-Trampont, A.C., Macara, I.G., Madhani, H., Fink, G.R. Polakis, P. and Bollag, G. (1996) Identification of a novel guanine and Ravichandran, K.S. (2002) Unconventional Rac-GEF activ- nucleotide exchange factor for the Rho GTPase. J. Biol. Chem. ity is mediated through the Dock180 ELMO complex. Nat. Cell 271, 25452–25458. Biol. 4, 574–582. [26] Han, J., Das, B., Wei, W., Van Aelst, L., Mosteller, R.D., [15] Cote, J.F. and Vuori, K. (2002) Identification of an evolutionarily Khosravi-Far, R., Westwick, J.K., Der, C.J. and Broek, D. (1997) conserved superfamily of DOCK180-related proteins with guan- Lck regulates Vav activation of members of the Rho family of ine nucleotide exchange activity. J. Cell. Sci. 115, 4901–4913. GTPases. Mol. Cell. Biol. 17, 1346–1353. [16] Meller, N., Irani-Tehrani, M., Kiosses, W.B., Del Pozo, M.A. and [27] Nishihara, H., Kobayashi, S., Hashimoto, Y., Ohba, F., Moc- Schwartz, M.A. (2002) Zizimin1, a novel Cdc42 activator, reveals hizuki, N., Kurata, T., Nagashima, K. and Matsuda, M. (1999) a new GEF domain for Rho proteins. Nat. Cell Biol. 4, 639–647. Non-adherent cell-specific expression of DOCK2, a member of [17] Wu, Y.C. and Horvitz, H.R. (1998) C. elegans phagocytosis and the human CDM-family proteins. Biochim. Biophys. Acta 1452, cell-migration protein CED-5 is similar to human DOCK180. 179–187. Nature 392, 501–504. [28] Hasegawa, H., Kiyokawa, E., Tanaka, S., Nagashima, K., Gotoh, [18] Ruusala, A. and Aspenstrom, P. (2004) Isolation and character- N., Shibuya, M., Kurata, T. and Matsuda, M. (1996) DOCK180, isation of DOCK8, a member of the DOCK180-related regulators a major CRK-binding protein, alters cell morphology upon of cell morphology. FEBS Lett. 572, 159–166. translocation to the . Mol. Cell. Biol. 16, 1770– [19] Deckert, M., Tartare-Deckert, S., Couture, C., Mustelin, T. and 1776. Altman, A. (1996) Functional and physical interactions of Syk [29] Grimsley, C.M., Kinchen, J.M., Tosello-Trampont, A.C., Brug- family kinases with the Vav proto-oncogene product. Immunity 5, nera, E., Haney, L.B., Lu, M., Chen, Q., Klingele, D., 591–604. Hengartner, M.O. and Ravichandran, K.S. (2004) Dock180 [20] Neudauer, C.L. and Macara, I.G. (2000) Purification and and ELMO1 proteins cooperate to promote evolutionarily biochemical characterization of TC10. Methods Enzymol. 325, conserved Rac-dependent cell migration. J. Biol. Chem. 279, 3–14. 6087–6097. [21] Miyazaki, T., Liu, Z.J., Kawahara, A., Minami, Y., Yamada, K., [30] Fukui, Y., Hashimoto, O., Sanui, T., Oono, T., Koga, H., Abe, Tsujimoto, Y., Barsoumian, E.L., Permutter, R.M. and Tanigu- M., Inayoshi, A., Noda, M., Oike, M., Shirai, T. and Sasazuki, chi, T. (1995) Three distinct IL-2 signaling pathways mediated by T. (2001) Haematopoietic cell-specific CDM family protein bcl-2 c-myc, and lck cooperate in hematopoietic cell proliferation. DOCK2 is essential for lymphocyte migration. Nature 412, Cell 81, 223–231. 826–831. [22] Uekawa, N., Nishikimi, A., Isobe, K., Iwakura, Y. and Maruy- [31] Corpet, F. (1988) Multiple sequence alignment with hierarchical ama, M. (2004) Involvement of IL-1 family proteins in p38 linked clustering. Nucleic Acids Res. 16, 10881–10890.