Slac2-a/Melanophilin, the Missing Link between Rab27 and Myosin Va

Myosin Va is a member of the unconventional class V myosin family, and a mutation in themyosin Va gene causes pigment granule transport defects in human Griscelli syndrome and dilute mice. How myosin Va recognizes its cargo (i.e. melanosomes), however, has re- mained undetermined over the past decade. In this study, we discovered Slac2-a/melanophilin to be the “missing link” between myosin Va and GTP-Rab27A present in the melanosome. Deletion analysis and site-directed mutagenesis showed that the N-terminal Slp (synaptotagmin-like protein) homology domain of Slac2-a specifically binds Rab27A/B isoforms and that the C-terminal half directly binds the globular tail of myosin Va. The tripartite protein complex (Rab27A· Slac2-a·myosin Va) in melanoma cells was further confirmed by immunoprecipitation. The discovery that myosin Va indirectly recognizes its cargo through Slac2-a, a novel Rab27A/B effector, should shed light on molecular recognition of its specific cargo by class V myosin.

The Slp family contains two conserved domains at the N terminus, referred to as SHD1 and SHD2 (5). The SHD1 and SHD2 of Slp3-a and Slp4 are separated by a sequence containing two zinc-finger motifs, whereas Slp1 and Slp2-a lack such zinc-finger motifs, and their SHD1 and SHD2 are linked together (5). The SHD has also been found in other proteins, including Slac2-a (Slp homologue lacking C2 domains-a) and Slac2-b/KIAA0624 (11), suggesting a general role of the SHD in cellular signaling.
Two very recent important discoveries have been made concerning the functional relationship between the SHD and Rab27A, one of the small GTP-binding proteins believed to be essential for membrane trafficking in eukaryotic cells (12). The first was our discovery that the SHD of the Slp family and Slac2-a directly interact with the GTP-bound form of Rab27A both in vitro and in intact cells (13). Since mutations in the rab27A gene cause hemophagocytic syndrome (Griscelli syndrome), an uncontrolled T lymphocyte and macrophage activation syndrome in humans (14,15), and defects in granule exocytosis in cytotoxic T lymphocytes and melanosome transport in ashen mice (16 -19), we hypothesized that the Slp family and Slac2 are involved in such membrane trafficking (13). The second very recent important discovery is the identification of Slac2-a as melanophilin and that a mutation in the mlph gene causes defects in melanosome transport in leaden mice (20). Interestingly ashen mice, which carry a mutation in the rab27A gene, and dilute mice, which carry a mutation in the myosin Va gene, which encodes one of the actin-based motor proteins (21,22), have shown similar defects in pigment granule transport (i.e. clumping of melanosomes in the perinuclear region), and as a result ashen, leaden, and dilute mice all exhibit a similarly lighter coat color (14,16,19,(23)(24)(25). In addition, genetic analysis has shown that these three proteins function in the same or overlapping transport pathways (20), but the functional relationships between these three molecules, Rab27A, Slac2-a/ melanophilin, and myosin Va, in melanosome transport remained to be clarified (25).
In this study, we report on two domain structures of Slac2-a, the N-terminal SHD, which specifically interacts with the Rab27A and Rab27B isoforms, and the C-terminal domain, which directly interacts with the globular tail of myosin Va, and we discuss the role of the tripartite protein complex in melanosome transport based on our findings.

MATERIALS AND METHODS
Molecular Cloning of Mouse Rab27B, Rab34, and Myosin Va cDNAs-cDNA encoding a full open reading frame of mouse Rab27B, Rab34, and myosin Va was amplified from Marathon-Ready adult brain cDNA (CLONTECH) by reverse transcriptase PCR as described previously (26). The purified PCR products were directly inserted into the pGEM-T Easy vector (Promega, Madison, WI). Both strands were completely sequenced using a Hitachi SQ-5500 DNA sequencer. We identified one deletion as a result of alternative splicing (deletion of amino acids 1387-1411) and several amino acid differences (R695A, D904E, * This work was supported in part by grants from the Science and Technology Agency to Japan (to K. M.) and Grant 13780624 from the Ministry of Education, Science, and Culture of Japan (to M. F.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Direct Interaction between Slac2-a and Myosin Va-pEF-T7-GST-Slac2-a-⌬146, -T7-GST-Slac2-a, -T7-Rab27A, -T7-GST-FLAG-myosin Va-GT, and pEF-T7-GST vectors were constructed by PCR essentially as described elsewhere. 2 COS-7 cells (5 ϫ 10 5 cells (the day before transfection)/10-cm dish) were transfected with 4 g of plasmids as described previously (30). Three days after transfection, cells were harvested and homogenized in 1 ml of a buffer containing 50 mM HEPES-KOH, pH 7.2, 250 mM NaCl, 0.1 mM phenylmethylsulfonyl fluoride, 10 M leupeptin, and 10 M pepstatin A in a glass-Teflon Potter homogenizer with 10 strokes at 900 -1000 rpm. After solubilization with 1% Triton X-100, insoluble material was removed by centrifugation at 15,000 rpm for 10 min. Expressed GST fusion proteins were affinity-purified on glutathione-Sepharose beads (wet volume, 20 l; Amersham Biosciences) according to the manufacturer's recommendations. After extensively washing the beads with 10 mM HEPES-KOH, pH 7.2, 100 mM NaCl, and 0.2% Triton X-100, thrombin (1 unit, Sigma) digestion was performed on the same column at 25°C for 1 h. The eluate containing FLAG-myosin Va-GT protein (or Rab27A) was then incubated with benzamidine-Sepharose 6B (wet volume, 20 l; Amersham Biosciences) to remove the thrombin. Protein concentrations were estimated by 10% SDS-PAGE or determined with a Bio-Rad protein assay kit using bovine serum albumin as a reference.
The purified FLAG-myosin Va-GT protein and Rab27A were incubated with glutathione-Sepharose beads (wet volume, 20 l) either coupled with T7-GST-Slac2-a or T7-GST alone in 50 mM HEPES-KOH, pH 7.2, 100 mM NaCl, 1 mM MgCl 2 , and 0.2% Triton X-100 for 1 h at 4°C. After washing three times with 1 ml of the binding buffer without recombinant proteins, proteins trapped with the beads were analyzed by 10% SDS-PAGE followed by immunoblotting with horseradish (HRP)-conjugated anti-FLAG tag antibody (Sigma) and anti-Rab27 mouse monoclonal antibody (Transduction Laboratories, Lexington, KY) as described previously (26,30).

RESULTS AND DISCUSSION
The SHD of Slac2-a Specifically Interacts with Rab27A/B Isoforms-We recently discovered that the SHD of Slac2-a/b directly binds the GTP-bound form of Rab27A in vitro and in intact cells (13). However, it remained undetermined whether the full-length Slac2-a protein specifically recognizes the Rab27A molecule and whether Rab27A interacts with the SHD alone or also with the large Slac2-a C-terminal domain with unknown function. To address these issues, we first investigated the specific interaction of full-length Slac2-a with various Rab proteins (Rab1, Rab2, Rab3A, Rab4A, Rab5A, Rab6A, Rab7, Rab8, Rab9, Rab10, Rab11A, Rab17, Rab18, Rab20, Rab22, Rab23, Rab25, Rab27A, Rab27B, Rab28, Rab34, or Rab37) in intact cells by co-expression assay (30,35). In brief, T7-tagged Slac2-a and each of the FLAG-tagged Rabs were co-expressed in COS-7 cells, and T7-Slac2-a protein was immunoprecipitated with the anti-T7 antibody-conjugated agarose (26,30). As expected, T7-Slac2-a protein specifically co-immunoprecipitated with the FLAG-Rab 27A and -Rab27B isoforms but not with other Rabs (Fig. 1, B, lane 18, and D, lane 1). The faint signal observed in Fig. 1B, lane 20, was probably attributable to nonspecific interaction of Rab34 with the agarose beads.
Next we investigated the possible involvement of the C terminus of Slac2-a in Rab27A/B binding in the same co-transfection assay. The SHD domain alone efficiently co-immunoprecipitated with both the Rab27A and Rab27B isoforms (Fig. 1, C and D, lane 2), but the C-terminal domain lacking the SHD (⌬146) did not (Fig. 1, C and D, lane 3). In addition, the Slac2-a mutant carrying the SLEWY-to-ALEAA substitutions in SHD2 (referred to as Slac2-a(A4), see Fig. 1A) dramatically reduced the Rab27A/B binding activity (Fig. 1, C and D, lane 4). Consistent with this, crystallographic analysis has shown the corresponding sequence (SGAWFF) in rabphilin-3 directly interacts with Rab3A (36). These findings indicated that the SHD of Slac2-a specifically binds Ra27A/B isoforms but that the large C-terminal domain is not involved in the recognition of Rab27A/B molecules and might have different functions.

Two Domain Structures of Slac2-a: the N-terminal SHD Responsible for Rab27A/B Binding and the C-terminal Domain
Responsible for Myosin Va Binding-The results of a genetic analysis comparing dilute, ashen, and leaden mice have indicated that myosin Va, Rab27A, and Slac2-a function in the same or overlapping transport pathways in melanosome transport (20). Consistent with this, myosin Va in extracts from melanocytes has been shown to co-immunoprecipitate with Rab27A (19). However, when FLAG-Rab27A and T7-myosin Va were co-expressed in COS-7 cells, no Rab27A-myosin Va complex was detected in the cell lysates ( Fig. 2A, lane 8), indicating that an additional protein must link Rab27A and myosin Va.
Since the SHD of Slac2-a specifically binds Rab27A (Fig. 1), we hypothesized that Slac2-a is the missing link between Rab27A and myosin Va in melanosome transport. To test this hypothesis, three proteins (FLAG-Rab27A, T7-Slac2-a, and T7-myosin Va) were co-expressed in COS-7 cells, and their associations were analyzed by immunoprecipitation as described above (26,35). As expected, in the presence of full-length T7-Slac2-a, T7-myosin Va co-immunoprecipitated with FLAG-Rab27A ( Fig. 2A, lane 5), whereas in the absence of T7-Slac2-a, T7myosin Va was undetectable in the anti-FLAG antibody immunoprecipitants ( Fig. 2A, lane 8). Interestingly the SHD alone (Slac2-a-SHD) or the C-terminal half alone (Slac2-a-⌬146) failed to mediate co-immunoprecipitation of T7-myosin Va with FLAG-Rab27A ( Fig. 2A, lanes 6 and 7), suggesting that different domains of Slac2-a may be involved in Rab27A binding and myosin Va binding. Similar results were obtained when FLAG-Rab27B was used instead of FLAG-Rab27A (data not shown).
We next sought to identify the myosin Va-binding site in Slac2-a by dual tag (T7 and FLAG) co-expression assay. When T7-Slac2-a deletion mutants and FLAG-myosin Va were coexpressed in COS-7 cells, T7-Slac2-a-⌬146, but not T7-Slac2-a-SHD, co-immunoprecipitated with FLAG-myosin Va (Fig. 2B), indicating the two domain structures of Slac2-a: the N-terminal SHD involved in the GTP-bound form of Rab27A binding and the C-terminal domain involved in myosin Va binding.   2. The tripartite protein complex formed by Slac2-a, Rab27A, and myosin Va. A, Slac2-a bridges the gap between Rab27A and myosin Va. pEF-FLAG-Rab27A, pEF-T7-Slac2-a mutants, and/or pEF-T7-myosin Va were cotransfected into COS-7 cells. The proteins expressed were immunoprecipitated by anti-FLAG tag antibody-conjugated agarose. Co-immunoprecipitated (IP) T7-myosin Va and -Slac2-a were first detected by HRP-conjugated anti-T7 antibody (top and middle panels). The same blots were then stripped and reprobed with HRP-conjugated anti-FLAG tag antibody (bottom panel). Lanes 1-4 show the total expressed T7-myosin Va and -Slac2-a (1/80 volume of the reaction mixture) used for immunoprecipitation. Note that only the full-length Slac2-a (closed arrowhead) forms the link between Rab27A and myosin Va (lane 5), although T7-Slac2-a-SHD (arrow), but not T7-Slac2-a-⌬146 (open arrowhead), efficiently immunoprecipitated with FLAG-Rab27A. B, two domain structures of Slac2-a. FLAG-Slac2-a deletion mutants and T7-myosin Va were co-expressed in COS-7 cells, and their associations were analyzed by immunoprecipitation as described above. Note that the C-terminal half of Slac2-a, but not the SHD, is essential for myosin Va binding. The positions of the molecular weight markers (ϫ 10 Ϫ3 ) are shown on the left.
Direct Interaction between the Globular Tail of Myosin Va and the C-terminal Domain of Slac2-a-Since the globular tail of myosin Va was thought to be essential for cargo recognition (21) (Fig. 3A), we investigated the interaction between Slac2-a and the myosin Va globular tail by using the same dual tag (T7 and FLAG) co-expression assay, and as shown in Fig. 3B, Slac2-a-⌬146 interacted with the globular tail of myosin Va in intact cells. We used purified recombinant proteins (GST-Slac2-a (or -⌬146), Rab27A, and FLAG-myosin Va-GT) for an in vitro binding assay to further confirm the direct interaction between the C terminus of Slac2-a and the globular tail of myosin Va as well as the in vitro formation of a tripartite protein complex from purified components. As expected, FLAGmyosin Va-GT bound GST-Slac2-a-⌬146 but not GST alone (Fig. 3C, lane 2, arrow), and full-length Slac2-a bound both FLAG-myosin Va-GT and Rab27A (Fig. 3D, arrows).
The Tripartite Protein Complex (Rab27A⅐Slac2-a⅐Myosin Va) in Melanoma Cells-Lastly, immunoprecipitation analysis was performed to investigate whether the tripartite protein complex (Rab27A⅐Slac2-a⅐myosin Va) is formed at physiological conditions. As shown in Fig. 3F, both myosin Va and Rab27A were co-immunoprecipitated with anti-Slac2-a-specific antibody (Fig. 3E), but not control IgG, from melanoma cell lysates. Thus, the tripartite protein complex (Rab27A⅐Slac2-a⅐myosin Va) demonstrated by in vitro binding experiments should be physiologically relevant.
Conclusions-The results of a recent biochemical analysis have suggested that the tail domain of myosin Va (or Vb) recognizes its cargo by directly binding to the proteins present in the cargo (e.g. Rab11, Rab25, and synaptobrevin⅐synaptophysin complex) (37)(38)(39). However, since myosin Va did not directly recognize Rab27A in the melanosomes, an additional linker protein was proposed to assist melanosome recognition in melanosome transport (25). In the present study, we discovered that Slac2-a is a missing link between Rab27A and myosin Va in melanosome transport and demonstrated how myosin Va recognizes its specific cargo (i.e. melanosomes) by its globular tail domain. The possible role of the tripartite protein complex (Slac2-a, Rab27A, and myosin Va) in melanosome capture in actin-rich cell periphery is summarized in Fig. 4. The SHD of Slac2-a specifically binds the GTP-Rab27A in the melanosomes, and the C terminus of Slac2-a binds the globular tail of myosin Va, which binds actin filament via the head domain. Since Slac2-a is expressed in various tissues, including the brain (20), the Slac2-a⅐Rab27⅐myosin Va complex may be involved in other types of membrane trafficking. For instance, Slac2-a⅐Rab27B⅐myosin Va may be involved in endoplasmic reticulum transport to dendrites in neurons because the inositol 1,4,5-trisphosphate receptor on the endoplasmic reticulum   (15,19,42). Slac2-a binds the GTP-bound form of Rab27A through its SHD (Fig. 1) (13) and also directly binds the globular tail of myosin Va through the C-terminal domain (Fig. 3). The head (motor) domain of myosin Va interacts with actin filaments. Thus, the tripartite protein complex of Rab27A, Slac2-a/melanophilin, and myosin Va is essential to the capture and local movement of melanosomes in the actin-rich cell periphery (25).
does not migrate to the postsynaptic spines in dilute mice (40,41). Further work is necessary to define the universality and/or specialty of the tripartite protein complex, Slac2-a, Rab27A/B, and myosin Va, in membrane trafficking.