Synaptotagmin-like protein 5: a novel Rab27A effector with C-terminal tandem C2 domains

Biochemical and Biophysical Research Communications 293 (2002) 899–906 www.academicpress.com

Synaptotagmin-like protein 5: a novel Rab27A effector with C-terminal tandem C2 domainsq Taruho S. Kuroda,a,b Mitsunori Fukuda,a,* Hiroyoshi Ariga,b and Katsuhiko Mikoshibac,d,e a

Fukuda Initiative Research Unit, RIKEN (The Institute of Physical and Chemical Research), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan b Graduate School of Pharmaceutical Sciences, Hokkaido University, Kita-ku, Sapporo 060-0812, Japan c Laboratory for Developmental Neurobiology, Brain Science Institute, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan d ‘‘CALCIUM OSCILLATION,’’ International Cooperative Research Project, JST, Brain Science Institute, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan e Division of Molecular Neurobiology, Department of Basic Medical Science, The Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan Received 3 April 2002

Abstract Synaptotagmin-like proteins 1–4 (Slp1–4) are new members of the carboxyl-terminal-type (C-type) tandem C2 proteins and are classified as a subfamily distinct from the synaptotagmin and the Doc2 families, because the Slp family contains a unique homology domain at the amino terminus, referred to as the Slp homology domain (SHD). We previously showed that the SHD functions as a binding site for Rab27A, which is associated with human hemophagocytic syndrome (Griscelli syndrome) [J. Biol. Chem. 277 (2002) 9212; J. Biol. Chem. 277 (2002) 12432]. In the present study, we identified a novel member of the Slp family, Slp5. The same as other Slp family members, the SHD of Slp5 preferentially interacted with the GTP-bound form of Rab27A and marginally with Rab3A and Rab6A, both in vitro and in intact cells, but not with other Rabs tested (Rab1, Rab2, Rab4A, Rab5A, Rab7, Rab8, Rab9, Rab10, Rab11A, Rab17, Rab18, Rab20, Rab22, Rab23, Rab25, Rab28, and Rab37). However, unlike other members of the Slp family, expression of Slp5 mRNA was highly restricted to human placenta and liver. Expression of Slp5 protein and in vivo association of Slp5 with Rab27A in the mouse liver were further confirmed by immunoprecipitation. The results suggest that Slp5 might be involved in Rab27A-dependent membrane trafficking in specific tissues. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Synaptotagmin-like protein; Slp homology domain; Rab27A; C2 domain; Granuphilin-a; Griscelli syndrome

Synaptotagmin-like proteins 1–4 (Slp1–4) are recently identified members of the carboxyl-terminal-type (C-

q The nucleotide sequence reported in this paper is deposited in the DDBJ, EMBL, and GenBank nucleotide sequence databases with accession number of AB080222. Abbreviations: C, carboxyl; C-type, carboxyl-terminal-type; GST, glutathione S-transferase; GTPcS, guanosine 50 -O-(3-thiotriphosphate); HRP, horseradish peroxidase; N, amino; PAGE, polyacrylamide gel electrophoresis; PC, phosphatidylcholine; PCR, polymerase chain reaction; PS, phosphatidylserine; PVDF, polyvinylidene difluoride; RBD, Rab3 binding domain; RT, reverse transcriptase; SHD(s), Slp homology domain(s); Slac2, Slp homolog lacking C2 domains; Slp, synaptotagmin-like protein; Syt(s), synaptotagmin(s). * Corresponding author. Fax: +81-48-467-9744. E-mail address: [email protected] (M. Fukuda).

type) tandem C2 proteins [1,2]. The characteristic feature of the Slp family is that it contains a unique amino (N)terminal homology domain (referred to as Slp homology domain, SHD), which is not found in other C-type tandem C2 proteins including the synaptotagmin (Syt) family [3–6], the Doc2 family [7], B/K [8,9], and Tac2-N [10]. The SHD is also found in other proteins, Slac2-a (Slp homolog lacking C2 domains-a) and Slac2-b, which do not contain any C2 domains. The SHD is usually composed of SHD1, two zinc-finger motifs, and SHD2 (see Fig. 2), but the SHDs of Slp1, Slp2-a, and Slac2-b lack zinc-finger motifs between SHD1 and SHD2, and the two SHDs are linked together [2]. The SHD shows weak homology with the Rab3 binding domain (RBD) of RIM [11], rabphilin [12], and Noc2 [13], and it

0006-291X/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 0 0 0 6 - 2 9 1 X ( 0 2 ) 0 0 3 2 0 - 0

900

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906

contains a sequence similar to that of the Rab3-binding pocket of the RBD [14]. Very recently we discovered that the SHD of Slp1, Slp2-a, Slp3-a, Slac2-a, and Slac2-b functions as a Rab27A-specific binding domain and that the SHD of Slp4 and the RBD of rabphilin function as a Rab3A, Rab8, and Rab27A binding domain both in vitro and in intact cells [15]. The SHD of Slp and Slac2 directly and preferentially interacted with the GTP-bound, active form of Rab27A rather than its GDP-bound form [15], suggesting a role of the Slp family and Slac2 as Rab27A effector molecules in the Rab27A-dependent membrane transport mechanism. Mutation of the rab27A gene [16,17], which encodes a small GTP-binding protein [18,19], was recently identified as the cause of human hemophagocytic syndrome (Griscelli syndrome) and ashen mice, both of which exhibit defects in melanosome transport and in regulated granule exocytosis in cytotoxic T lymphocytes [20–28]. Interestingly, Slac2-a was recently identified as melanophilin, a mutation which causes a defect in melanosome transport in leaden mice [29]. Since mouse coat color mutation analyses have shown that products of three genes, rab27A (mutated in the ashen mouse), slac2a/mlph (mutated in the leaden mouse), and MYOVA (mutated in the dilute mouse), function in the same overlapping melanosome transport pathways [20,24,25, 29,30] and form a tripartite protein complex both in vitro and in melanoma cells [31], we hypothesized that the Slp family is also involved in Rab27A-dependent melanosome transport via the SHD. Indeed, we recently demonstrated that in the wild-type melanoma cells Slp1 and Slp2 are localized in the periphery of the cells with Rab27A and melanosomes, whereas they accumulate in the perinuclear region in the dilute mouse-derived S91/ Cloudman cells [15]. To thoroughly understand Rab27A-dependent membrane transport, it is important to determine how many SHD-containing proteins (i.e., Rab27A effector) are present in the body and whether they are functionally diversified or not (i.e., have the same or different biochemical properties). In this study, we searched for the human genome sequences and identified a novel member of the Slp family, Slp5. Like other members of the Slp family, Slp5 specifically interacted with the GTP-bound form of Rab27A both in vitro and in vivo, but in contrast to other members of the Slp family, expression of Slp5 mRNA was specific to the placenta and liver. Based on our findings, we discuss the function of Slp5 in Rab27Adependent membrane trafficking in specific tissues.

Materials and methods Materials. Ex Taq and recombinant Taq DNA polymerases were obtained from Takara Shuzo (Shiga, Japan) and Toyobo Biochemicals

(Tokyo, Japan), respectively. Horseradish peroxidase (HRP)-conjugated anti-T7 tag antibody and anti-T7 tag antibody-conjugated agarose were from Novagen (Madison, WI, USA). HRP-conjugated anti-FLAG tag antibody, anti-FLAG M2-agarose affinity gel, guanosine 50 -O(3-thiotriphosphate) (GTPcS), L -a-phosphatidylcholine (PC), dipalmitoyl, and L -a-phosphatidylserine (PS), dioleoyl were from Sigma Chemical (St. Louis, MO, USA). HRP-conjugated anti-glutathione S-transferase (GST) antibody was from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-Rab27 mouse monoclonal antibody was from Transduction Laboratories (Lexinton, KY, USA). All other chemicals were commercial products of reagent grade. Solutions were made up in deionized water prepared with an Elix10 Water Purification System and Milli-Q Biocel A10 System (Millipore; Bedford, MA, USA). Molecular cloning of human Slp5 cDNA. cDNA encoding the open reading frame of human Slp5 was amplified from human first-strand cDNAs prepared from placenta by polymerase chain reaction (PCR), as described previously [15,32], using the following primers with restriction enzyme sites (underlined) or stop codons (bold letter) designed on the basis of the human Slp5 genome sequence (GenBank accession number; AL121578): 50 -GCGGATCCATGTCTAAGAACT CAGAGTT-30 (Slp5-50 primer; sense), 50 -TTAGAGCCTACATTTTC CCA-30 (Slp5-30 primer; antisense). PCR was carried out in the presence of Perfect Match PCR Enhancer (Stratagene; La Jolla, CA, USA) for 40 cycles, each consisting of denaturation at 94 °C for 1 min, annealing at 55 °C for 2 min, and extension at 72 °C for 2 min. The PCR products were purified from an agarose gel on a Micro-Spin column (Amersham Biosciences; Buckinghamshire, UK) as described previously [15,32], and were directly inserted into the pGEM-T Easy vector (pGEM-T-Slp5) (Promega; Madison, WI, USA). Both strands of the cDNA inserts were completely sequenced. Addition of the T7-tag to the N-terminus of Slp5 (pEF-T7-Slp5) and construction of the expression vectors were performed as described previously [32–35]. Sequence analyses. Multiple sequence alignment and depiction of the phylogenetic tree of C-type tandem C2 proteins were performed by using the CLUSTALW program (http://hypernig.nig.ac.jp/homology/ clustalw.shtml) set at the default parameters (gapdist ¼ 8 and maxdiv ¼ 40). Reverse transcriptase (RT)-PCR analysis. Human first-strand cDNAs prepared from various tissues were obtained from Clontech Laboratories (Human MTC Panel I, Palo Alto, CA, USA). PCRs were carried out in the presence of Perfect Match PCR Enhancer for 40 cycles, each consisting of denaturation at 94 °C for 1 min, annealing at 55 °C for 2 min, and extension at 72 °C for 2 min. Slp5-50 primer and Slp5-30 primer were used for amplification. The PCR products were analyzed by 1% agarose gel electrophoresis and ethidium bromide staining. The authenticity of the products was verified by subcloning into a pGEM-T Easy vector and DNA sequencing. Preparation of GST fusion proteins. Construction of pGEX-4T-3 vector (Amersham Biosciences) carrying fragments of the Slp5-SHD, -C2A domain, or -C2B domain was essentially performed by PCR [34] using pGEM-T-Slp5 as a template. The following pairs of oligonucleotides with appropriate restriction enzyme sites (underlined) and/or termination codons (bold letters) were used for amplification: the Slp550 primer described above and Slp5-SHD-30 primer (antisense) 50 -GC GAATTCATGGACTTCTTCTTAAAATGG-30 ; Slp5-C2A-50 primer (sense) 50 -GCGGATCCAACGTGAAAGTCAGTGGTGA-30 and Slp5-C2A-30 primer (antisense) 50 -GCACTAGTCAGCCAATATCAG GAGCAAACT-30 ; Slp5-C2B-50 primer (sense) 50 -GCGGATCCAAA GGAGAGCTGACAGTTGT-30 and the Slp5-30 primer described above. The resulting pGEX-Slp5-SHD, -C2A, and -C2B were confirmed by DNA sequencing and transformed into Escherichia coli JM109. GST fusion proteins were expressed and purified on glutathione Sepharose (Amersham Biosciences) by the standard method [36]. GST–Slp5-SHD, -C2A, and -C2B encoded amino acids 1–148, 403–540, and 544–730 of human Slp5, respectively. Antibody production and immunoprecipitation. Anti-Slp5 antibody was prepared as described elsewhere (T.S. Kuroda, and M. Fukuda,

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906

901

unpublished data). Since the C2 domains of Slp4 and Slp5 were highly conserved, the anti-Slp5 antibody also recognized Slp4. However, the expression of Slp4 was highly restricted to pancreatic b-cells and was not found in liver [37], we could use this antibody in the liver where the Slp5 mRNA was abundantly expressed (see Fig. 5). Whole heart and liver from adult female ICR mouse were homogenized in 1 ml of 50 mM Tris–HCl (pH 8.0), 150 mM NaCl, 0.1% SDS, 1% NP40, 0.5% deoxycholate, and protease inhibitors, and the homogenates were centrifuged at 14,000 rpm for 10 min at 4 °C. The supernatant was used for immunoprecipitation assay as described previously [31]. Miscellaneous procedures. GST pull-down assay, co-immunoprecipitation assay, and guanine nucleotide-selective binding assay were performed as described previously [15,32,35]. Preparation of liposomes consisting of PC and PS (1:1 w/w) and phospholipid binding assay were also performed as described previously [38,39]. The protein concentrations were determined with a BioRad protein assay kit (BioRad Laboratories; Hercules, CA, USA) using bovine serum albumin as a standard.

Results and discussion Molecular cloning of Slp5, a novel member of the synaptotagmin-like protein family To identify additional members of the Slp family, we performed a database search with the mouse Slp1–4 amino acid sequences as the query. The search yielded a human genomic sequence on chromosome region Xp21.1 (GenBank accession number; AL121578) that contained a putative amino acid sequence similar to that of Slp4/granuphilin-a [37]. Human first-strand cDNAs were used to amplify the full-length of the expected open reading frame by PCR, and it was cloned. The 2193-base sequence comprised a single open reading frame encoding 730 amino acids. The corresponding protein contained the SHD at the N-terminus and tandem C2 domains at the C-terminus (Fig. 1A), indicating that it should be classified as a novel member of the Slp family, and we designated it as Slp5. Consistent with this classification, the phylogenetic tree of all the C-type tandem C2 proteins reported to date revealed that Slp5 belongs to the same branch as other Slp family members (Fig. 1B, shaded box). Slp5 showed the highest homology with human Slp4/granuphilin-a (57% similarity and 44% identity in full-length), especially in the SHD region (61% similarity and 44% identity), the C2A domain (62% similarity and 50% identity), and the C2B domain (67% similarity and 56% identity) (Fig. 1A). The exon/ intron boundaries of the human slp5 gene and human slp4 gene (GenBank accession number; NT_011687) are shown in Fig. 1A (arrowheads), where both patterns of exon division were completely matched in the SHD and the C2A and C2B domains (Fig. 1A, closed arrowheads), but not in the spacer domains between the SHD and the C2A domain (Fig. 1A, open arrowheads), indicating that the SHDs and tandem C2 domains of Slp4 and Slp5 were derived from the same source. It

Fig. 1. Identification of Slp5 as a novel member of the Slp family. (A) Comparison of human Slp5 with human Slp4/granuphilin-a. Slp5 showed the highest homology with Slp4 (57% similarity in fulllength). The SHD is composed of SHD1 (black boxes), two zincfinger motifs (indicated as Zn2þ ), and SHD2 (black boxes). The similarities between each of the domains are indicated as percentages. Note that the SHD and two C2 domains (shaded boxes) are well conserved (higher than 60% similarity), while the spacer domain between the SHD and C2A domains is not conserved very well (45% similarity). The exon/intron boundaries are indicated by arrowheads, and the numbers above and below the arrowheads describe the position in the codon at which the cording sequence is separated by the intron (0 ¼ at the codon junction; 1 and 2 ¼ after the first codon position and the second codon position, respectively). Note that the pattern of exon/intron divisions is identical in the SHD, C2A, and C2B domains (closed arrowheads), but different in the spacer domain (open arrowheads). Amino acid numbers are given on both sides. (B) Phylogenetic tree of the C-type tandem C2 protein family reported to date. The phylogenetic tree is depicted as described under ‘‘Materials and methods’’. The C-type tandem C2 proteins are categorized into the synaptotagmin family [3–6], the Doc2 family [7], and the Slp family [1,2]. Note that Slp5 (shaded box) is classified into a branch of the Slp family.

should be noted that both the human slp5 gene (mapped to Xp21.1, LocusID; 94122) and the human slp4 gene (mapped to Xq21.33, LocusID; 94121) are located on chromosome X, suggesting that the slp4 gene or the slp5 gene may be produced by gene duplication. Another open reading frame encoding a different Slp5 splicing isoform which lacked SHD1 but contained the histone H2A motif at the N-terminus (GenBank accession number; XM_066572) was predicted from the human slp5 gene. However, since none of the corresponding product was amplified by RT-PCR under our

902

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906

experimental conditions (data not shown), we concluded that the mRNA of this Slp5 isoform is very low-copy, or not transcribed. The SHD of Slp5 functions as a Rab27A binding domain In our previous study, the SHDs of Slp1, Slp2-a, Slp3-a, Slac2-a, and Slac2-b specifically interacted with Rab27A, but not with other Rabs, whereas the SHD of Slp4 and the RBD of rabphilin interacted with Rab3A, Rab8, and Rab27A [15]. The SHDs are composed of SHD1, two zinc-finger motif sequences (absent in the SHDs of Slp1, Slp2-a, and Slac2-b), and SHD2 [2] (Fig. 2). The SHD of Slp5 contains two zinc-finger motifs and shows high sequence similarity with other SHDs, suggesting the possible function of Slp5-SHD as a specific Rab effector. To identify the Rab proteins that interact with Slp5, we tested the binding of GST–Slp5-SHD and 20 different FLAG-tagged Rab proteins expressed in COS-7 cells (Rab1, Rab2, Rab3A, Rab4A, Rab5A, Rab6A, Rab7, Rab8, Rab9, Rab10, Rab11A, Rab17, Rab18, Rab20, Rab22, Rab23, Rab25, Rab27A, Rab28, and Rab37) by GST pull-down assay as described previously [15]. As expected, the SHD of Slp5 preferentially associated with Rab27A, and very weakly with Rab3A and Rab6A (Fig. 3A).

To confirm the association between Slp5 and Rab3A, Rab6A, or Rab27A in intact cells, we next performed a co-immunoprecipitation assay. T7-tagged Slp5 and FLAG-tagged Rabs were transiently expressed in COS-7 cells, and solubilized T7-Slp5 was immunoprecipitated with anti-T7 tag antibody-conjugated agarose, and co-immunoprecipitated Rab proteins were detected by anti-FLAG antibody. Consistent with the GST pull-down assay, Slp5 strongly interacted with Rab27A and marginally with Rab3A and Rab6A (Fig. 3B), suggesting that Rab27A is the physiological binding partner of Slp5. Although Slp4, the closest homolog of Slp5 in the Slp family, has been shown to associate with Rab8 [15], no interaction between Slp5 and Rab8 was detected (Fig. 3B, lane 3), indicating that Slp5 is one of the Rab27A-specific members of the Slp/ Slac2 family (i.e., Slp1, Slp2-a, Slp3-a, Slac2-a, and Slac2-b) and different from the Rab3A, Rab8, and Rab27A-specific members (including Slp4 and rabphilin). Slp5 Interacts with an active form of Rab27A The function of Rab proteins is generally regulated by the guanine nucleotide exchange system: the GTPbound form is active, and the GDP-bound form is

Fig. 2. Sequence alignment of the Slp homology domains (SHD) and the Rab3 binding domains (RBD). Sequence alignment of the SHD of Slp1, Slp2-a, Slp3-a, Slp4, Slp5, Slac2-a, and Slac2-b, and the RBD of RIM1D57–106, RIM2, rabphilin, and Noc2. Among the splicing isoforms of RIM1, RIM1D57–106 [11] shows the highest sequence homology with other SHDs or RBDs. Residues in the sequences that are conserved and similar are shown against a black background and against a shaded background, respectively. The solid lines indicate two SHDs. The pound signs (#) indicate the sequence corresponding to the Rab3 binding site (SGAWFF) of rabphilin [14]. The conserved Cys residues corresponding to two zinc-finger motifs are indicated by asterisks (*). Amino acid numbers are indicated at the right of each line.

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906

903

Fig. 3. Slp5 interacts with GTP-bound form of Rab27A. (A) Specific interaction of Slp5 with Rab27A in GST pull-down assay. FLAG-tagged Rab proteins were expressed in COS-7 cells, and the proteins expressed were solubilized with 1% Triton X-100. Expression of recombinant Rab proteins with FLAG tag in the reaction mixtures was confirmed by immunoblotting with HRP-conjugated anti-FLAG tag antibody (1/10,000 dilution, top panel). GST–Slp5-SHD was used to perform a GST pull-down assay as described previously [15]. The proteins trapped with the beads were subjected to 12.5% SDS–polyacrylamide gel electrophoresis (PAGE) and transferred to the PVDF (polyvinylidene difluoride) membrane. The blot was first probed with HRP-conjugated anti-FLAG tag antibody (middle panel). The same blot was then stripped [46] and reprobed with HRP-conjugated anti-GST antibody to ensure that the same amounts of GST–Slp5-SHD proteins were loaded (1/10,000 dilution, bottom panel). We confirmed that none of the Rab proteins were trapped with the beads coupled with GST alone (data not shown). Note that Slp5-SHD specifically interacted with Rab27A, and marginally with Rab3A and Rab6A. Molecular weight markers (103 ) are shown at the left. (B) Specific interaction of Slp5 with Rab27A in intact cells. pEF-T7-Slp5 and pEF-FLAG-Rab3A, -Rab6A, -Rab8, or -Rab27A were cotransfected into COS-7 cells. The proteins expressed were solubilized with 1% Triton X-100 and immunoprecipitated with anti-T7 tag antibody-conjugated agarose as described previously [15,32]. Co-immunoprecipitated FLAG-Rabs were first detected by HRP-conjugated anti-FLAG tag antibody (middle panel). The same blot was stripped and reprobed with HRP-conjugated anti-T7 tag antibody to ensure loading of the same amounts of T7-Slp5 proteins (1/10,000 dilution, bottom panel). Total expressed FLAG-Rab proteins (1/100 volumes of reaction mixtures; input) used for immunoprecipitation are shown in top panel. Note that the full-length Slp5 specifically interacted with Rab27A, and very weakly with Rab3A and Rab6A even in the intact cells. Molecular weight markers (103 ) are shown at the left. (C) Guanine nucleotide-selective interaction of Rab27A with the Slp5. The GTP- or GDP-bound form of FLAG-Rab27A coupled with the agarose beads was prepared as described previously [15], and T7-Slp5 expressed in COS-7 cells was incubated with the beads for 2 h at 4 °C. After washing the beads five times, co-precipitated T7-Slp5 was detected by immunoblotting with HRP-conjugated anti-T7 tag antibody (middle panel). The same blot was stripped and reprobed with HRP-conjugated anti-FLAG tag antibody to ensure the same amount of FLAG-Rab27A protein (bottom panel). Total T7-Slp5 proteins (1/100 volumes of reaction mixtures; input) used for assay are shown in top panel. Note that Slp5 preferentially interacted with GTP-bound form of Rab27A beads, but not the beads alone. The results shown are representative of at least two or three independent experiments.

inactive. Almost all Rab effectors show higher affinity for the Rab-GTP form than for the Rab-GDP form [18]. Because the Slp family and Slac2 have been shown to preferentially interact with Rab27A-GTP [15], we investigated whether Slp5 also shows higher affinity for the GTP-bound form of Rab27A. Recombinant FLAGRab27A coupled with anti-FLAG tag antibody-conju-

gated agarose was first incubated with 0.5 mM GTPcS and then with T7-Slp5 in the presence of GTPcS, or first with 1 mM GDP and then with T7-Slp5 in the presence of GDP. The proteins bound to the beads were analyzed by immunoblotting with HRP-conjugated anti-T7 tag antibody. As shown in Fig. 3C, Slp5 exhibited higher affinity for Rab27A-GTP than for Rab27A-GDP,

904

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906

suggesting that Slp5 functions as a Rab27A-GTP effector molecule. Phospholipid binding properties of the Slp5 C2 domains Next, we focused on the tandem C2 domains, because the C2 domain was originally found in Ca2þ -dependent protein kinase C and is thought to function as a phospholipid binding site (reviewed in [40]). In Syt I, a wellcharacterized C-type tandem C2 protein, the C2A domain binds negatively charged phospholipids in a Ca2þ dependent manner, which is thought to be crucial for Ca2þ -dependent neurotransmitter release [41–43]. Crystallographic and mutational analyses of the Syt I C2A domain indicated that five aspartate (or glutamate) residues at the top of the C2 b-sandwich structure are essential for Ca2þ binding [44,45]. Sequence alignment revealed that Slp5 lacks such conserved aspartate or glutamate residues (His at amino acid positions 501 [H501] and S509 in the Slp5-C2A domain, and R603, and I672 in the Slp5C2B domain), suggesting that Slp5 may be classified as an Ca2þ -independent type. To address this, phospholipid (PS/PC liposomes) binding assay was performed using GST–Slp5-C2A and -C2B (Fig. 4). Unexpectedly, however, the C2A domain of Slp5 showed Ca2þ -dependent phospholipid binding activity (Fig. 4, top panel), and the Slp5-C2B bound phospholipid irrespective of the presence of Ca2þ (Fig. 4, bottom panel). Tissue distribution of Slp5 The tissue distribution of human Slp5 mRNA was investigated by RT-PCR with specific primers (Fig. 5). Expression of the Slp5 mRNA was highest in placenta,

Fig. 5. Tissue distribution of human Slp5. RT-PCR analysis of Slp5 expression in various tissues (heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas) (top panel). Note that the Slp5 mRNA is highly expressed in placenta and liver. RT-PCR analysis of G3PDH expression was also performed (bottom panel) to ensure that equivalent amounts of first strand cDNA were used for RT-PCR analysis. ‘‘)’’ means without templates as a negative control. The size of the molecular weight markers (k=StyI) is shown at the left of the panel. The results shown are representative of two or three independent experiments.

moderate in liver, lowest in brain, lung, and kidney, and absent in heart, skeletal muscle, and pancreas (Fig. 5, top panel). Since human Rab27A mRNA was also expressed in placenta and liver [16], it is possible that Slp5 and Rab27A function in the same membrane transport pathway in placenta and liver. The highest expression of Slp5 mRNA in placenta implies possible involvement of Slp5 in hormone release in the placenta. It should be noted that no mRNA of the Slp family other than the Slp5 mRNA was detected well in liver. Although Slp4 and Slp5 were most likely to be derived from the same ancestor, their mRNA expression patterns were completely different, with the mRNA of Slp4 being selectively expressed in pancreatic b cells [37]. Thus, Slp4 and Slp5 may function differently as Rab27A effectors in their specific tissues. Formation of Slp5–Rab27A complex in the mouse liver

Fig. 4. Phospholipid binding properties of the Slp5 C2 domains. Liposomes and GST fusion proteins were incubated in 50 mM HEPES-KOH, pH 7.2, in the presence of 2 mM EGTA or 1 mM Ca2þ for 15 min at room temperature. After centrifugation at 12,000g for 10 min, the supernatants (S, non-binding fraction) and pellets (P, phospholipid binding fraction) were separated as described previously [38,39]. Equal proportions of the supernatants and pellets were subjected to 10% SDS–PAGE and then stained with Coomassie brilliant blue R-250. Note that Slp5-C2A and -C2B showed Ca2þ -dependent and Ca2þ -independent phospholipid binding, respectively. The results shown are representative of three independent experiments.

Finally, immunoprecipitation analysis was performed to investigate whether the Slp5–Rab27A complex is formed in vivo. As shown in Fig. 6A, anti-Slp5 antibody detected a single immunoreactive band with the same size of T7-tagged Slp5 in liver, but not in heart. This immunoreactive band was specifically concentrated by immunoprecipitation with the anti-Slp5 antibody, but not control IgG, from the mouse liver lysates (Fig. 6B, top panel), indicating that the immunoreactive band is endogenous Slp5 protein. It should be noted that Rab27A was also co-immunoprecipitated with the antiSlp5 antibody, but not control IgG (Fig. 6B, bottom panel). Thus, we concluded that the Slp5–Rab27A complex is indeed formed at physiological conditions. Conclusion In summary, we cloned and characterized a novel member of the Slp family, Slp5, and found that it

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906

Fig. 6. Formation of Slp5–Rab27A complex in the mouse liver. (A) Expression of Slp5 in the mouse liver. The recombinant T7-Slp5 expressed in COS-7 cells (lane 1) and total homogenates of adult mouse heart (lane 2) and liver (lane 3) were loaded on 10% SDS– polyacrylamide gel and immunoblotted with anti-Slp5 antibody. Note that the anti-Slp5 antibody recognized an immunoreactive band with the same size of T7-Slp5. (B) Co-immunoprecipitation of Rab27A with the anti-Slp5 antibody. Immunoprecipitates of the anti-Slp5 antibody or control IgG were subjected to 10% SDS–PAGE and immunoblotted with the anti-Slp5 (top panel) or anti-Rab27 antibody (bottom panel). The open arrowhead indicates IgG light chain (IgG LC). Molecular weight markers (103 ) are shown at the left.

showed the greatest similarity to Slp4. The same as other Slp family members, the SHD of Slp5 bound specifically to Rab27A both in vitro and in vivo. Slp5 preferentially interacted with the GTP-bound form of Rab27A, suggesting that Slp5 functions as a effector molecule for activated Rab27A. Further work is needed to determine whether Slp5 is actually involved in Rab27A-dependent membrane trafficking, especially in placenta and liver.

Acknowledgments This work was supported in part by grants from the Science and Technology Agency to Japan (to M.F. and K.M.) and Grants 13780624 from the Ministry of Education, Science, and Culture of Japan (to M.F.). We thank Eiko Kanno and Yukie Ogata for technical assistance.

References [1] M. Fukuda, K. Mikoshiba, Synaptotagmin-like protein 1–3: a novel family of C-terminal-type tandem C2 proteins, Biochem. Biophys. Res. Commun. 281 (2001) 1226–1233. [2] M. Fukuda, C. Saegusa, K. Mikoshiba, Novel splicing isoforms of synaptotagmin-like proteins 2 and 3: identification of the Slp homology domain, Biochem. Biophys. Res. Commun. 283 (2001) 513–519.

905

[3] M. Fukuda, K. Mikoshiba, The function of inositol high polyphosphate binding proteins, Bioessays 19 (1997) 593–603. [4] G. Schiavo, S.L. Osborne, J.G. Sgouros, Synaptotagmins: more isoforms than functions? Biochem. Biophys. Res. Commun. 248 (1998) 1–8. [5] B. Marqu
eze, F. Berton, M. Seagar, Synaptotagmins in membrane traffic: which vesicles do the tagmins tag? Biochimie 82 (2000) 409–420. [6] T.C. S€ udhof, Synaptotagmins: why so many? J. Biol. Chem. 277 (2002) 7629–7632. [7] R.R. Duncan, M.J. Shipston, R.H. Chow, Double C2 protein. A review, Biochimie 82 (2000) 421–426. [8] O.J. Kwon, H. Gainer, S. Wray, H. Chin, Identification of a novel protein containing two C2 domains selectively expressed in the rat brain and kidney, FEBS Lett. 378 (1996) 135–139. [9] M. Fukuda, K. Mikoshiba, The N-terminal cysteine cluster is essential for membrane targeting of B/K protein, Biochem. J. 360 (2001) 441–448. [10] M. Fukuda, K. Mikoshiba, Tac2-N, an atypical C-type tandem C2 protein localized in the nucleus, FEBS Lett. 503 (2001) 217–218. [11] Y. Wang, S. Sugita, T.C. S€ udhof, The RIM/NIM family of neuronal C2 domain proteins. Interactions with Rab3 and a new class of Src homology 3 domain proteins, J. Biol. Chem. 275 (2000) 20033–20044. [12] T. Yamaguchi, H. Shirataki, S. Kishida, M. Miyazaki, J. Nishikawa, K. Wada, S. Numata, K. Kaibuchi, Y. Takai, Two functionally different domains of rabphilin-3A, Rab3A p25/smg p25A-binding and phospholipid- and Ca2þ -binding domains, J. Biol. Chem. 268 (1993) 27164–27170. [13] K. Kotake, N. Ozaki, M. Mizuta, S. Sekiya, N. Inagaki, S. Seino, Noc2, a putative zinc finger protein involved in exocytosis in endocrine cells, J. Biol. Chem. 272 (1997) 29407–29410. [14] C. Ostermeier, A.T. Brunger, Structural basis of Rab effector specificity: crystal structure of the small G protein Rab3A complexed with the effector domain of rabphilin-3A, Cell 96 (1999) 363–374. [15] T.S. Kuroda, M. Fukuda, H. Ariga, K. Mikoshiba, The Slp homology domain of synaptotagmin-like proteins 1–4 and Slac2 functions as a novel Rab27A binding domain, J. Biol. Chem. 277 (2001) 9212–9218. [16] D. Chen, J. Guo, T. Miki, M. Tachibana, W.A. Gahl, Molecular cloning and characterization of rab27a and rab27b, novel human rab proteins shared by melanocytes and platelets, Biochem. Mol. Med. 60 (1997) 27–37. [17] T. Tolmachova, J.S. Ramalho, J.S. Anant, R.A. Schultz, C.M. Huxley, M.C. Seabra, Cloning, mapping and characterization of the human RAB27A gene, Gene 239 (1999) 109–116. [18] M. Zerial, H. McBride, Rab proteins as membrane organizers, Nat. Rev. Mol. Cell Biol. 2 (2001) 107–117. [19] J.B. Pereira-Leal, M.C. Seabra, Evolution of the Rab family of small GTP-binding proteins, J. Mol. Biol. 313 (2001) 889–901. [20] S.M. Wilson, R. Yip, D.A. Swing, T.N. O’Sullivan, Y. Zhang, E.K. Novak, R.T. Swank, L.B. Russell, N.G. Copeland, N.A. Jenkins, A mutation in Rab27a causes the vesicle transport defects observed in ashen mice, Proc. Natl. Acad. Sci. USA 97 (2000) 7933–7938. [21] G. Menasche, E. Pastural, J. Feldmann, S. Certain, F. Ersoy, S. Dupuis, N. Wulffraat, D. Bianchi, A. Fischer, F. Le Deist, G. de Saint Basile, Mutations in RAB27A cause Griscelli syndrome associated with haemophagocytic syndrome, Nat. Genet. 25 (2000) 173–176. [22] P. Bahadoran, E. Aberdam, F. Mantoux, R. Busca, K. Bille, N. Yalman, G. de Saint-Basile, R. Casaroli-Marano, J.P. Ortonne, R. Ballotti, Rab27a: a key to melanosome transport in human melanocytes, J. Cell Biol. 152 (2001) 843–850. [23] A.N. Hume, L.M. Collinson, A. Rapak, A.Q. Gomes, C.R. Hopkins, M.C. Seabra, Rab27a regulates the peripheral distribu-

906

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33] [34]

[35]

T.S. Kuroda et al. / Biochemical and Biophysical Research Communications 293 (2002) 899–906 tion of melanosomes in melanocytes, J. Cell Biol. 152 (2001) 795– 808. M.S. Marks, M.C. Seabra, The melanosome: membrane dynamics in black and white, Nat. Rev. Mol. Cell. Biol. 2 (2001) 738–748. J.A. Hammer III, X.S. Wu, Rabs grab motors: defining the connections between Rab GTPases and motor proteins, Curr. Opin. Cell Biol. 14 (2002) 69–75. J.C. Stinchcombe, D.C. Barral, E.H. Mules, S. Booth, A.N. Hume, L.M. Machesky, M.C. Seabra, G.M. Griffiths, Rab27a is required for regulated secretion in cytotoxic T lymphocytes, J. Cell Biol. 152 (2001) 825–834. E.K. Haddad, X. Wu, J.A. Hammer III, P.A. Henkart, Defective granule exocytosis in Rab27a-deficient lymphocytes from ashen mice, J. Cell Biol. 152 (2001) 835–842. X. Wu, K. Rao, M.B. Bowers, N.G. Copeland, N.A. Jenkins, J.A. Hammer III, Rab27a enables myosin Va-dependent melanosome capture by recruiting the myosin to the organelle, J. Cell Sci. 114 (2001) 1091–1100. L.E. Matesic, R. Yip, A.E. Reuss, D.A. Swing, T.N. O’Sullivan, C.F. Fletcher, N.G. Copeland, N.A. Jenkins, Mutations in Mlph, encoding a member of the Rab effector family, cause the melanosome transport defects observed in leaden mice, Proc. Natl. Acad. Sci. USA 98 (2001) 10238–10243. J.A. Mercer, P.K. Seperack, M.C. Strobel, N.G. Copeland, N.A. Jenkins, Novel myosin heavy chain encoded by murine dilute coat colour locus, Nature 9 (1991) 709–713. M. Fukuda, T.S. Kuroda, K. Mikoshiba, Slac2-a/Melanophilin, the missing link between Rab27 and Myosin Va: Implications of a tripartite protein complex for melanosome transport, J. Biol. Chem. 277 (2002) 12432–12436. M. Fukuda, E. Kanno, K. Mikoshiba, Conserved N-terminal cysteine motif is essential for homo- and heterodimer formation of synaptotagmins III, V, VI, and X, J. Biol. Chem. 274 (1999) 31421–31427. S. Mizushima, S. Nagata, pEF-BOS, a powerful mammalian expression vector, Nucleic Acids Res. 18 (1990) 5322. M. Fukuda, J. Aruga, M. Niinobe, S. Aimoto, K. Mikoshiba, Inositol-1,3,4,5-tetrakisphosphate binding to C2B domain of IP4BP/synaptotagmin II, J. Biol. Chem. 269 (1994) 29206– 29211. M. Fukuda, K. Mikoshiba, Distinct self-oligomerization activities of synaptotagmin family: unique calcium-dependent oligomeriza-

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

tion properties of synaptotagmin VII, J. Biol. Chem. 275 (2000) 28180–28185. D.B. Smith, K.S. Johnson, Single-step purification of polypeptides expressed in Escherichia coli as fusions with glutathione S-transferase, Gene 67 (1988) 31–40. J. Wang, T. Takeuchi, H. Yokota, T. Izumi, Novel rabphilin-3like protein associates with insulin-containing granules in pancreatic beta cells, J. Biol. Chem. 274 (1999) 28542–28548. M. Fukuda, T. Kojima, K. Mikoshiba, Phospholipid composition dependence of Ca2þ -dependent phospholipid binding to the C2A domain of synaptotagmin IV, J. Biol. Chem. 271 (1996) 8430– 8434. M. Fukuda, T. Kojima, K. Mikoshiba, Regulation by bivalent cations of phospholipid binding to the C2A domain of synaptotagmin III, Biochem. J. 323 (1997) 421–425. E.A. Nalefski, J.J. Falke, The C2 domain calcium-binding motif: structural and functional diversity, Protein Sci. 5 (1996) 2375– 2390. K. Mikoshiba, M. Fukuda, J.E. Moreira, F.M.T. Lewis, M. Sugimori, M. Niinobe, R. Llinas, Role of the C2A domain of synaptotagmin in transmitter release as determined by specific antibody injection into the squid giant synapse preterminal, Proc. Natl. Acad. Sci. USA 92 (1995) 10703–10707. S. Mochida, M. Fukuda, M. Niinobe, H. Kobayashi, K. Mikoshiba, Roles of synaptotagmin C2 domains in neurotransmitter secretion and inositol high-polyphosphate binding at mammalian cholinergic synapses, Neuroscience 77 (1997) 937–943. R. Fernandez-Chac on, A. K€ onigstorfer, S.H. Gerber, J. Garcıa, M.F. Matos, C.F. Stevens, N. Brose, J. Rizo, C. Rosenmund, T.C. S€ udhof, Synaptotagmin I functions as a calcium regulator of release probability, Nature 410 (2001) 41–49. X. Shao, B.A. Davletov, R.B. Sutton, T.C. S€ udhof, J. Rizo, Bipartite Ca2þ -binding motif in C2 domains of synaptotagmin and protein kinase C, Science 273 (1996) 248–251. R.B. Sutton, B.A. Davletov, A.M. Berghuis, T.C. S€ udhof, S.R. Sprang, Structure of the first C2 domain of synaptotagmin I: a novel Ca2þ /phospholipid-binding fold, Cell 80 (1995) 929– 938. M. Sugimori, C.-K. Tong, M. Fukuda, J.E. Moreira, T. Kojima, K. Mikoshiba, R. Llinas, Presynaptic injection of syntaxin-specific antibodies blocks transmission in the squid giant synapse, Neuroscience 86 (1998) 39–51.