Advertisement
Research Article
SNX9 regulates tubular invagination of the plasma membrane through interaction with actin cytoskeleton and dynamin 2
Narae Shin, Namhui Ahn, Belle Chang-Ileto, Joohyun Park, Kohji Takei, Sang-Gun Ahn, Soo-A Kim, Gilbert Di Paolo, Sunghoe Chang
Journal of Cell Science 2008 121: 1252-1263; doi: 10.1242/jcs.016709

Summary

Dynamic membrane remodeling during intracellular trafficking is controlled by the intricate interplay between lipids and proteins. BAR domains are modules that participate in endocytic processes by binding and deforming the lipid bilayer. Sorting nexin 9 (SNX9), which functions in clathrin-mediated endocytosis, contains a BAR domain, however, the properties of this domain are not well understood. Here we show that SNX9 shares many properties with other BAR domain-containing proteins, such as amphiphysin and endophilin. SNX9 is able to deform the plasma membrane, as well as liposomes, into narrow tubules and recruit N-WASP and dynamin 2 to these tubules via its SH3 domain. SNX9-induced tubulation is antagonized by N-WASP and dynamin 2 while it is enhanced by perturbation of actin dynamics. However, SNX9 also has several unique properties. The tubulating activity requires the BAR and PX domains, as well as the low-complexity (LC) domain, which binds the Arp2/3 complex. SNX9 also binds to PtdIns(4)P-5-kinases via its PX domain and its tubulating activity is regulated by phosphoinositides. In addition, the kinase activity of PtdIns(4)P-5-kinases is stimulated by interaction with SNX9, suggesting a positive feedback interaction between SNX9 and PtdIns(4)P-5-kinases. These results suggest that SNX9 functions in the coordination of membrane remodeling and fission via interactions with actin-regulating proteins, endocytic proteins and PtdIns(4,5)P2-metabolizing enzymes.

Introduction

Endocytic processes function to internalize both extracellular components and cell surface receptors, and participate in diverse signal-transduction pathways (Maxfield and McGraw, 2004; Slepnev and De Camilli, 2000). Clathrin-mediated endocytosis (CME) is the most thoroughly studied of the endocytic pathways. In CME, a complex network of proteins, including key components of the clathrin coat, has been shown to interact with each other and with membrane lipids (Cho and Stahelin, 2005; Di Paolo and De Camilli, 2006; Gruenberg, 2003; Martin, 2001; van Meer and Sprong, 2004). Recently, several accessory proteins have been reported to bind and transform spherical liposomes into long tubules in vitro, the diameter of which are similar in size to the stalks of plasma membrane invaginations seen during CME in intact cells (Dawson et al., 2006; Farsad and De Camilli, 2003; Gallop and McMahon, 2005; Itoh and De Camilli, 2006; Takei et al., 1999; Tsujita et al., 2006; Zimmerberg and McLaughlin, 2004). Thus, it has been proposed that membrane deformations mediated by these proteins may contribute to the initial phase of the endocytic process. Amphiphysin and endophilin are among the first identified endocytic proteins with membrane tubulating properties (Farsad et al., 2001; Masuda et al., 2006; Peter et al., 2004; Takei et al., 1999). These proteins share a similar organization, with an NH2-terminal Bin-Amphiphysin-Rvs (BAR) domain and a C-terminal Src homology (SH3) domain. Recent crystal structures of the BAR domains of amphiphysin and endophilin revealed that these BAR domains form a banana-shaped dimer and have been shown to preferentially bind to small, highly curved artificial liposomes, thereby demonstrating their ability to sense membrane curvature (Habermann, 2004; Peter et al., 2004; Weissenhorn, 2005). These BAR domains have been shown to be responsible for the ability to induce membranes to form tubules in vitro and in vivo (Farsad et al., 2001; Gallop et al., 2006; Masuda et al., 2006; Peter et al., 2004; Takei et al., 1999). Whereas the BAR domain in these proteins is responsible for membrane binding, the SH3 domain mediates interactions with other proteins, primarily endocytic proteins and actin regulatory proteins containing proline- and arginine-rich sequences, such as dynamin and synaptojanin (Cestra et al., 1999; Grabs et al., 1997; Ringstad et al., 1997; Simpson et al., 1999; Slepnev and De Camilli, 2000). Thus, these proteins may coordinate endocytic process with actin dynamics (Itoh et al., 2005; Kessels and Qualmann, 2004; Kovacs et al., 2006; Tsujita et al., 2006). More specifically, BAR proteins may coordinate different forces during CME, including deformation of membranes, the pushing of the vesicle away from the membrane, actin assembly, and fission of vesicles from the membrane (Kaksonen et al., 2006; Qualmann et al., 2000).

Recently, the F-BAR domain or the extended FC (EFC) domain, which contains a FER-CIP4 homology (FCH) domain and an additional coiled-coil (CC) region was shown to have functional similarity with BAR domains, including a membrane-tubulating activity (Itoh et al., 2005; Tsujita et al., 2006). Additional studies showed that most PCH proteins – a protein family to which FCH proteins belong – bind to dynamin and activated neural Wiskott-Aldrich syndrome protein (N-WASP) – a protein involved in actin polymerization. Noticeably, membrane tubulation by F-BAR/EFC domains as well as by BAR domains is enhanced by disruption of the actin cytoskeleton, coupling membrane deformation in endocytosis to actin dynamics (Itoh et al., 2005; Tsujita et al., 2006). The structure of the F-BAR/EFC domain has been solved and reveals a gently curved helical-bundle dimer of ∼200 Å in length, fitting a tubular membrane with a ∼600 Å diameter (Shimada et al., 2007).

Sorting nexin 9 (SNX9) is a ubiquitously expressed soluble protein that associates with membranes in cells (Howard et al., 1999; Worby and Dixon, 2002; Worby et al., 2001). It has been classified as a member of the sorting nexin family of proteins based on the presence of a variant of the Phox homology (PX) region – a defining characteristic of this family of proteins (Worby and Dixon, 2002). At present, 18 mammalian and 10 yeast SNXs have been identified (Carlton and Cullen, 2005; Worby and Dixon, 2002). Previous studies have shown that SNX9 is involved in CME through its interaction with endocytic proteins, such as dynamin, adaptor complex protein 2 (AP2) and clathrin (Lundmark and Carlsson, 2003; Soulet et al., 2005), and in actin dynamics through its interaction with N-WASP or WASP (Badour et al., 2007; Yarar et al., 2008; Yarar et al., 2007). SNX9 stimulates N-WASP–Arp2/3-mediated actin assembly, which is greatly enhanced by PtdIns(4,5)P2-induced SNX9 oligomerization (Yarar et al., 2007). Additionally, we report here that SNX9 interacts with the Arp2/3 complex and PtdIns(4)P-5-kinases. SNX9 shares several features with amphiphysin and endophilin, including the presence of an SH3 domain and a BAR domain (Peter et al., 2004). Both amphiphysin and endophilin also contain an N-terminal amphiphatic helix located immediately upstream of the BAR domain, thereby also classifying them as N-BAR proteins. The BAR domain of SNX9, however, is found at the C-terminus of the protein and a recently published structure of the BAR domain of SNX9 has shown that SNX9 does not contain an immediate upstream amphiphatic helix in front of the BAR domain (Pylypenko et al., 2007), thus how the BAR domain of SNX9 mediates endocytosis is not fully understood (Pylypenko et al., 2007; Shin et al., 2007; Yarar et al., 2008).

Here we show that SNX9 shares functional similarities with BAR/F-BAR domain proteins, including a powerful membrane-deforming activity in vitro and in vivo via its BAR domain, and interaction with dynamin 2 and N-WASP via its SH3 domain. Furthermore, tubulation by SNX9 is antagonized by N-WASP and dynamin 2 overexpression whereas it is enhanced by perturbation of actin dynamics with latrunculin or jasplakinolide. However, SNX9 has distinct properties from other BAR/F-BAR domain proteins. The low complex (LC) domain binds Arp2/3 complex via its A-like region and is required for its tubulating activity together with BAR and PX domains. Furthermore, SNX9 binds to PtdIns(4)P-5-kinases via its PX domain and coexpression of SNX9 with PtdIns(4)P-5-kinases markedly increases tubule formation whereas coexpression with the 5-phosphatase domain of synaptojanin inhibits it. Furthermore, SNX9 stimulates the kinase activity of PtdIns(4)P-5-kinases, thus suggesting a positive feedback interaction between SNX9 and PtdIns(4)P-5-kinases. Our results suggest that similarly to the BAR domains of other proteins, the BAR domain of SNX9 functions in tandem with its PX and LC domains in the recruitment of SNX9, and subsequently N-WASP, Arp2/3 complex and dynamin 2, to sites of clathrin-coated pit formation where it coordinates membrane invagination and fission events through interaction with the actin cytoskeleton and dynamin.

Results

SNX9 has a functional BAR domain and induces massive tubulation of the plasma membrane

We first compared the sequence of the SNX9-BAR domain with that of well-established BAR-domain-containing proteins, such as endophilin and amphiphysin. Sequence alignment revealed that the BAR domain of SNX9 showed weak homology to other BAR domains, but residues inside of helix regions were well conserved. The basic residues in the SNX9-BAR domain were also well conserved with those of other BAR domains that have been shown to contribute to membrane binding via interaction with acidic phospholipids (Gallop et al., 2006) (Fig. 1A).

To test whether SNX9 could induce membrane tubules in cells, we performed time-lapse imaging analysis. We found that expression of GFP-SNX9 results in massive membrane tubulation in COS-7 cells (Fig. 1B). We next set out to identify the domain(s) of SNX9 that is (are) critical for this membrane tubulation activity by expressing various deletion mutants. We found that not only BAR domain (amino acids 385-595) but also that the PX (239-385) and LC domains (70-239) are required for this activity (Fig. 1C), whereas the SH3 domain (1-70) was not (Fig. 1D). Consistently, a recent structural study also showed that the PX domain is required for phospholipid binding and the LC domain (amino acids 201-215) is required for membrane penetration by forming an amphiphatic helix (Pylypenko et al., 2007). SNX9 membrane tubules colocalized with extracellularly applied fluorescent membrane tracers chloromethylbenzamido (CM)-DiI, cholera toxin B-Alexa Fluor 594 and FM 5-95, indicating that the observed tubules originated from the plasma membrane (Fig. 2A-C). Furthermore, time-lapse total internal reflection fluorescence (TIRF) microscopic analysis was used to observe tubules that originated from the vicinity of the plasma membrane (Fig. 2D).

BAR-domain proteins bind to phosphoinositides including PtdIns(4,5)P2, and this binding seems to be important for their membrane-tubulating activity (Farsad et al., 2001; Itoh et al., 2005; Kojima et al., 2004; Shinozaki-Narikawa et al., 2006; Tsujita et al., 2006). We thus investigated the lipid-binding profiles of SNX9. Both the protein-lipid overlay assay and the liposome co-sedimentation assay showed that SNX9 binds most phosphoinositides with a slight preference for PtdIns(3)P, PtdIns(3,4)P2, PtdIns(3,5)P2 and PtdIns(4,5)P2 (Fig. 3A,B).

SNX9 can tubulate the lipid bilayer in vitro

We next examined whether SNX9 can deform artificial liposomes similarly to other BAR domain-containing proteins, such as endophilin and amphiphysin (Farsad et al., 2001; Takei et al., 1998). Purified GST-SNX9 was incubated with artificial liposomes made from a crude brain lipid extract. Negative staining electron microscopy revealed that SNX9 efficiently tubulates liposome in vitro (Fig. 3C). The powerful liposome tubulation activity of SNX9 was confirmed through a light-microscopy-based assay using a lipid bilayer formed by drying and rehydrating lipid droplets in a chamber between two glass slides as previously described (Itoh et al., 2005). Fig. 3B shows that addition of GST-SNX9 induced massive tubulation of the membrane surface (Fig. 3D).

Fig. 1.
Fig. 1.

SNX9 has a functional BAR domain and induces massive membrane tubulation. (A) Multiple sequence alignment of the BAR domains in SNX9 and other proteins with Clustal X. Conserved residues are highlighted with the following color code: yellow, hydrophobic; green, polar; blue, basic; red, acidic. Secondary structure of rAmphiphysin2 was determined from the crystal structure and that of SNX9 was predicted using online program NPS@ (http://npsa-pbil.ibcp.fr). (B) COS-7 cells were transfected with GFP-SNX9 and time-lapse imaging was performed 20 hours after transfection. Images were captured every 5 seconds for 5 minutes. High-magnification views are of the regions enclosed in rectangles. Arrowheads indicate the ends of tubules that underwent elongation. Scale bars: 30 μm (low magnification) and 3 μm (high magnification). (C) Domain structures of SNX9 and its deletion mutants. Tubulation ability is indicated. (D) COS-7 cells were transfected with GFP-tagged SNX9 or SNX9-ΔSH3 and tubule formation and cellular localization were observed. Scale bars: 30 μm.

SNX9 binds to N-WASP and dynamin 2 as well as Arp2/3 complex

Pull-down assays followed by western blot analysis from HEK cell lysates confirmed that the major binding partner of SNX9 is dynamin 2 and that this interaction is mediated by the SH3 domain of SNX9 (Fig. 4A,B). Although we could not detect N-WASP in the affinity-purified material from a SH3-domain column based on Coomassie Blue staining of SDS-PAGE, immunoblotting with an anti-N-WASP antibody revealed the presence of this protein in the affinity purified material (Fig. 4A). Immunoprecipitation experiments confirmed the association between exogenously expressed Flag-tagged SNX9 and GFP-tagged N-WASP in COS-7 cells (Fig. 4B). Dual-color simultaneous time-lapse imaging provided further support for a physiological interaction of SNX9 with N-WASP and dynamin 2 during membrane tubule formation (Fig. 4C). Although a deletion of the SH3 domain of SNX9 allowed for membrane tubulation and the localization of truncated SNX9 to these tubules, N-WASP and dynamin 2 were dispersed in the cytosol and did not localize to these structures (Fig. 4D), indicating that N-WASP and dynamin 2 are recruited to plasma-membrane-derived tubules via an interaction with the SH3 domain of SNX9.

Fig. 2.
Fig. 2.

SNX9-induced tubules are elongated from the plasma membrane. (A-C) COS-7 cells expressing GFP-tagged SNX9 was stained with 2 μM CM-DiI (A), 2 μM cholera toxin B Alexa Fluor 594 (B) or 15 μM FM5-95 (C) for 10 minutes on ice and fixed for observation. SNX9-induced tubules (green) and membrane marker stained plasma membranes (red) were mostly colocalized at cell peripheries. High-magnification views are of the regions enclosed by rectangles. Scale bars: 30 μm (low magnification) and 3 μm (insets). (D) Tubules grown from the bottom of the cell in contact with the coverslip were imaged by TIRF microscopy. Arrowheads indicate tubules elongated from SNX9-positive spots at the plasma membrane. Insets are high-magnification views of the regions enclosed in rectangles. Scale bars: 30 μm (low magnification) and 8 μm (insets).

Since N-WASP is a major regulatory protein of Arp2/3-mediated actin polymerization (Machesky et al., 1999; Rohatgi et al., 1999), we investigated whether SNX9 also associates with Arp2/3 complex and showed that SNX9 interacts with Arp2/3 via its LC domain (Fig. 5A,C). Sequence alignment showed that SNX9 contains an Arp2/3 binding region (A-like domain) in its LC domain (residues 161-167, Fig. 5E) and that this A-like domain is rich in acidic residues and contains a conserved tryptophan residue, which has been shown to be sufficient for binding to the Arp2/3 complex (Marchand et al., 2001; Panchal et al., 2003). Accordingly, the LC1 (residues 74-151) subdomain, which does not contain an A-like domain fails to bind Arp3 whereas the LC2 and LC3 subdomains, which both contain an A-like domain are able to bind Arp3 (Fig. 5D). Immunostaining of endogenous SNX9 and Arp2/3 complex revealed that they were scattered in the cytoplasm in punctate structures and appeared to be mostly juxtaposed or partially colocalized (Fig. 5F). Thus, our results suggest that SNX9 binds and recruits the Arp2/3 complex to sites of clathrin-coated pit formation where N-WASP can interact with the Arp2/3 complex to induce actin polymerization.

SNX9-induced membrane tubulation is inhibited by coexpression of N-WASP or dynamin 2

It was previously reported that the tubulation of F-BAR/EFC domain was antagonized by N-WASP and dynamin (Itoh et al., 2005; Tsujita et al., 2006). Since SNX9 also binds to N-WASP and dynamin 2, we examined the effects of N-WASP and dynamin 2 on SNX9-induced tubulation. Coexpression of high levels of N-WASP with SNX9 decreased the membrane tubulation induced by SNX9 whereas coexpression with N-WASP-ΔVCA, which cannot bind to the Arp2/3 complex, did not affect tubulation (Fig. 6A,C and supplementary material Fig. S2A,C). Likewise, when high levels of dynamin 2 were coexpressed with SNX9, tubulation was virtually absent (Fig. 6B and supplementary material Fig. S2B). As expected (Itoh et al., 2005; Tsujita et al., 2006), the antagonizing effect of dynamin 2 on tubulation was abolished with coexpression of dynamin 2 K44A, a GTPase-deficient mutant (Fig. 6D and supplementary material Fig. S2D).

We aimed to provide a quantitative analysis of the tubulation phenomenon in intact cells by measuring the length of induced tubules. Compared with cotransfections of GFP-SNX9 with N-WASP or dynamin 2, expression of SNX9-ΔSH3 in combination with these same proteins showed a significant increase in tubule length, which presumably results from the loss of interaction between SNX9 with N-WASP or dynamin 2 (Fig. 6E-I). Evidently, coexpression with N-WASP-ΔVCA or dynamin 2 K44A did not affect tubule length. In addition, the tubule length in cells with SNX9-ΔSH3 with empty vector was significantly longer than that in cells with full-length SNX9 and empty vector, probably because of the additional loss of interaction with endogenous N-WASP and dynamin 2 (Fig. 6I). Coexpression of N-WASP or dynamin 2 with SNX9, but not SNX9-ΔSH3, dramatically decreased the length of tubules (or inhibited tubulation completely in some cases) (Fig. 6A,B,I).

Fig. 3.
Fig. 3.

SNX9 exhibits phospholipid binding and powerful lipid tubulating activity in vitro. (A) Nitrocellulose filters spotted with various phospholipids (100 pmol of each phospholipid as indicated in this figure) were incubated with purified 0.5 μg/ml of GST-SNX9 and the binding of each protein to the specific phospholipid was detected with an anti-SNX9 antibody. (B) Synthetic PE and PC liposomes supplemented with 10% of the indicated lipid were incubated with SNX9, sedimented and stained with Coomassie Blue. Three independent experiments were performed and the protein intensity was measured. S, supernatant; P, pellet. (C) Negative stain electron microscopy of liposome composed of a brain lipid extract after addition of recombinant SNX9. Scale bar: 300 nm. (D) Lipid droplets are rehydrated to form a lipid bilayer sheet with solution containing either GST-alone or GST-tagged SNX9 during microscopic observation. Time-lapse images of GST-control (top) and GST-tagged SNX9 (bottom) using differential interference contrast (DIC) microscopy. Dotted lines indicate the edges of membrane sheet. Scale bars: 40 μm.

Perturbation of actin dynamics enhances membrane tubulation

Recent studies showed that tubulation by BAR and F-BAR/EFC proteins is enhanced by disruption of the actin cytoskeleton, indicating a close interplay between actin dynamics and endocytic events (Itoh et al., 2005; Tsujita et al., 2006). Since we also found that SNX9 binds N-WASP and the Arp2/3 complex – two of the most prominent proteins involved in actin polymerization – we examined the relationship between actin dynamics and SNX9-induced tubular invagination. Cells expressing low levels of SNX9 (i.e. which do not induce detectable membrane tubulation) were treated with the actin-depolymerizing agent, latrunculin B (LatB). Numerous membrane tubules started to elongate from preexisting fluorescent spots or from newly visible spots within seconds of LatB treatment (Fig. 7A,C).

We also treated cells with jasplakinolide, a cell-permeable actin-polymerizing agent. Within seconds of the treatment, membrane tubules started to grow from the spots but the morphology of tubules appeared to be different from that induced by LatB treatment (Fig. 7B). Although LatB-induced tubules were long and contained extended arm shapes, those promoted by jasplakinolide were short and circled around the SNX9-positive spots (Fig. 7A,B). These results indicate that any perturbation of regulated actin dynamics in cells could result in membrane tubulation. In addition, consistently with a previous study of other BAR and F-BAR domain proteins (Itoh et al., 2005), LatB treatment enhanced membrane tubule formation, even in cells expressing high levels of dynamin 2, suggesting that the antagonistic effect of dynamin 2 on SNX9-induced tubulation is blocked by actin depolymerization (supplementary material Fig. S3).

A recent study showed that SNX9 stimulates N-WASP–Arp2/3-mediated actin assembly, which is greatly enhanced by PtdIns(4,5)P2-induced SNX9 oligomerization (Yarar et al., 2007). We found here that SNX9 interacts via its PX domain with PtdIns(4)P-5-kinases Iα, Iβ and Iγ: major enzymes that synthesize PtdIns(4,5)P2 (Fig. 8). The coimmunoprecipitation of endogenous SNX9 with endogenous PtdIns(4)P-5-kinase Iα and GST-pull down assays with SNX9 with His-tagged PtdIns(4)P-5-kinases Iα, Iβ and Iγ (Fig. 8A-C) demonstrate their direct interaction. Accordingly, when cells expressing low levels of SNX9 (i.e. at levels insufficient to induce significant tubulation) were cotransfected with PtdIns(4)P-5-kinase Iα or PtdIns(4)P-5-kinase Iβ, numerous membrane tubules were formed around the cell periphery whereas the kinase dead mutant of PtdIns(4)P-5-kinase Iα failed to do so (Fig. 8D,E and supplementary material Fig. S4A,B). Depletion of PtdIns(4,5)P2 by the 5-phosphatase domain of synaptojanin inhibits tubule formation even with high levels of SNX9 expression (Fig. 8F and supplementary material Fig. S4C). We further showed that SNX9 increases the kinase activity of PtdIns(4)P-5-kinase Iγ in vitro and in vivo (Fig. 8G,H) suggesting a positive feedback interaction between SNX9 and PtdIns(4)P-5-kinases.

Fig. 4.
Fig. 4.

SNX9 binds to dynamin 2 and N-WASP through its SH3 domain and colocalizes on tubules. (A) Pull-down assay from HEK 293T cells with purified GST fusion of SNX9, SNX9-SH3, SNX9-ΔSH3 or GST-alone. The pull-downs were analyzed by Coomassie Blue staining. Arrow indicates major interacting protein predicted to be dynamin 2. Immunoblotting using anti-dynamin 2 or anti-N-WASP antibody was then performed. (B) COS-7 cells were cotransfected with Flag-tagged SNX9 or SNX9-ΔSH3 and GFP-tagged N-WASP or GFP-tagged dynamin 2, and immunoprecipitation (IP) was carried out with anti-Flag antibody, followed by immunoblotting (IB) with anti-Flag or anti-GFP antibody. TCL, total cell lysates. (C,D) Dual-color simultaneous time-lapse imaging of SNX9 or SNX9-ΔSH3 with dynamin 2 or N-WASP. GFP- or mRFP-tagged SNX9 or SNX9-ΔSH3 was cotransfected with mRFP-tagged dynamin 2 or GFP-tagged N-WASP, respectively. Live dual-color imaging was performed 24 hours after transfection. High-magnification views are of the regions enclosed in rectangles. Scale bars: 10 μm (low magnification), 2 μm (insets).

Discussion

In this study, we reveal that SNX9 shares some properties with BAR/F-BAR domain proteins. Like BAR/F-BAR domain proteins, SNX9 has membrane-deforming activity in vitro and in vivo via its BAR domain. It also binds dynamin 2 and N-WASP, and this interaction regulates the tubulating activity of SNX9 in conjunction with the actin cytoskeleton. We showed that the overexpression of dynamin 2 prevents tubulation whereas the GTPase-deficient mutant of dynamin 2 does not (Fig. 6 and supplementary material Fig. S2). This antagonistic effect of dynamin 2 on tubulation can be explained by the role of dynamin 2 in the fission of endocytic vesicles. We also demonstrated that N-WASP coexpression inhibits SNX9-induced tubule formation whereas coexpression with N-WASP-ΔVCA, which cannot bind to the Arp2/3 complex, did not affect tubulation (Fig. 6 and supplementary material Fig. S2). This effect of N-WASP may be caused by its role in actin polymerization thereby acting to link the endocytic vesicle with the actin cytoskeleton and causing an application of a pulling tension on the tubule/vesicle. This notion is further substantiated by experiments with LatB (Fig. 7). Relieving the tension by depolymerization of cortical actin cytoskeleton probably contributes to the dramatic increase in tubule formation observed after LatB treatment. In addition, LatB treatment blocked the antagonistic effect of dynamin 2 on tubule formation (supplementary material Fig. S3), thus suggesting an interconnection between the function of dynamin and the actin cytoskeleton.

SNX9 also has unique properties that have not been shown with other BAR/F-BAR proteins. The LC and PX domains bind the Arp2/3 complex and PtdIns(4)P-5-kinases, respectively, and these interactions regulate tubule formation (Figs 5, 8 and supplementary material Fig. S4). SNX9 also stimulates the kinase activity of PtdIns(4)P-5-kinases, thus suggesting a positive feedback interaction between SNX9 and PtdIns(4)P-5-kinases (Fig. 8).

Interestingly, although the perturbation of actin dynamics by both LatB and jasplakinolide enhances membrane tubulation, the morphologies of the resulting tubules differ. Extended long tubules are major forms in LatB-treated cells whereas short tubules circling the GFP-SNX9-positive spots are most frequent in jasplakinolide-treated cells. Although the exact mechanism underlying these differences is a subject for future study, it should be noted that jasplakinolide is known to paradoxically disrupt actin filaments in vivo and induce disordered polymerization of monomeric actin into amorphous masses (Bubb et al., 2000). Thus, tubules may start to grow abruptly at first, but later, as jasplakinolide stabilizes the actin cytoskeleton, further growth of tubules is blocked, and thus, it seems to `hold' membrane tubules at the spots where tubulation starts. LatB, however, depolymerizes the actin cytoskeleton, thus relieving the tension between tubules and plasma membrane around the spots and resulting in long and extended tubules.

Fig. 5.
Fig. 5.

SNX9 binds to the Arp2/3 complex via its LC domain. (A) COS-7 cell lysates were immunoprecipitated with anti-Arp3 antibody and immunoblotted (IB) with anti-SNX9 antibody. (B) Schematic diagrams of the GST-tagged truncated LC domains of SNX9 constructs. (C) In vitro binding assays were carried out with purified GST-SNX9 protein fragments (full-length, SH3, LC or PXBAR) and Arp2/3 complex. The bound protein was verified using anti-Arp3 antibody. (D) In vitro binding assays were carried out with various truncated mutant of LC domain and the bound protein was immunoblotted with anti-Arp3 antibody. The VCA domain of N-WASP (GST-VCA) was used as a positive control for Arp2/3 binding. (E) Comparison of the SNX9 A-like domain with the A-like domains of N-WASP, WASP, Scar, Cortactin, Myo1b and ActA. Black box with an asterisk indicates conserved tryptophan residue and gray boxes indicate conserved aspartate or glutamate residues. (F) Endogenous SNX9 and Arp2/3 complex in Cos-7 cells were stained with anti-SNX9 antibody and anti-Arp2/3 antibody, respectively. High-magnification views of the regions enclosed by rectangles are shown below each image. Arrows indicate where two proteins were juxtaposed or overlap. Scale bars: 30 μm (low magnification), 5 μm (high magnification).

SNX9, also known as SH3PX1, is a member of the SNX protein family, which is characterized by the presence of a variant of the phospholipid-binding module, the PX domain (Howard et al., 1999; Worby and Dixon, 2002; Worby et al., 2001). SNX9 was initially identified as an interacting partner of the metalloproteases MDC9 and MDC15 (Howard et al., 1999). It was also found to interact with Dock (the fly orthologue of mammalian NcK) and Dscam (Down syndrome cell adhesion molecule), thereby forming a complex involved in axonal guidance in Drosophila (Worby et al., 2001). Recent studies showed that the PRD domain of activated Cdc42-associated kinase 2 (ACK2) binds to the SH3 domain of SNX9 and that, upon EGF stimulation, ACK2, clathrin and phosphorylated SNX9 form a complex involved in the degradation of the EGF receptor (Lin et al., 2002). SNX9 also plays a role in CME through its interaction with components of the clathrin-mediated endocytic machinery (Lundmark and Carlsson, 2003). SNX9 forms a complex with dynamin 2 in the cytosol and regulates the recruitment of dynamin 2 to the membrane (Lundmark and Carlsson, 2004). A recent finding showed that SNX9 interacts with WASP and is involved in CD28 endocytosis and T-cell signaling (Badour et al., 2007). Furthermore, SNX9 enhances dynamin assembly and increases its GTPase activity (Soulet et al., 2005). Importantly, key elements of the clathrin coat, such as AP2 and clathrin itself, also bind to the LC region of SNX9 in a cooperative manner (Lundmark and Carlsson, 2003). The critical role of dynamin in endocytosis is further substantiated by experiments that show that abnormal levels of SNX9 by either overexpression or knockdown of SNX9 inhibit transferrin uptake in cells (supplementary material Fig. S5).

Fig. 6.
Fig. 6.

Antagonistic effects of N-WASP and dynamin 2 on SNX9-induced tubulation. (A,B) When high levels of GFP-tagged N-WASP or mRFP-tagged dynamin 2 were coexpressed with GFP- or mRFP-tagged SNX9 in COS-7 cells, tubular invaginations were significantly reduced or even absent in many cases. (C,D) Coexpression with N-WASP-ΔVCA, a mutant that cannot interact with the Arp2/3 complex, failed to affect tubulation by SNX9. Likewise, coexpression of SNX9 with dynamin 2 K44A, a GTPase-deficient mutant, abolished the antagonistic effect of dynamin 2 on tubulation. (E-H) Representative pictures from various combinations of SNX9 or SNX9-ΔSH3 with dynamin2 or N-WASP. Scale bars: 30 μm. (I) Quantitative analysis of tubule length in various experimental schemes. The length of tubules was measured manually using a MetaMorph software. Three rectangular areas (20×20 μm) near the periphery of each cell were drawn, and within these areas, the tubules that were well separated from other tubules were selected and their lengths were measured in the x-y dimension. Three independent experiments were performed and the data were pooled and normalized against those of full-length SNX9 with mRFP empty vector. Error bar indicates s.d. n=50 cells. *P<0.05, significantly different from SNX9-GFP or -mRFP control or each other, ANOVA and Tukey's HSD post hoc test for several different groups and Student's t-test for two different groups.

The SH3 domain of SNX9 binds to N-WASP and dynamin 2, and this interaction is required for the recruitment of N-WASP and dynamin 2 to SNX9-induced tubules (Fig. 4). SNX9 also interacts with the Arp2/3 complex via its LC domain (Fig. 5). Interestingly, the LC domain is the binding region of clathrin and AP2 to SNX9 (Lundmark and Carlsson, 2003). Thus, SNX9 appears to use the same domains to bind either endocytic proteins or actin regulating proteins: for endocytic proteins, SH3 to dynamin 2 and LC to clathrin and AP2; for actin-regulating proteins, SH3 to N-WASP and LC to the Arp2/3 complex. Taken together, it seems that SNX9 can provide links between different forces during CME: membrane deformation, fission and actin assembly, which are mediated by SNX9, dynamin 2, N-WASP and the Arp2/3 complex. Our results suggest that any imbalance between these forces can result in either the increase or decrease in the tubulation of the membrane associated with SNX9.

In summary, we find that, similarly to the F-BAR/EFC domain and other BAR-domain-containing proteins (Farsad et al., 2001; Gallop and McMahon, 2005; Itoh et al., 2005; Takei et al., 1999; Tsujita et al., 2006), the BAR domain of SNX9 coordinates membrane invagination and fission events through interaction with the actin cytoskeleton via N-WASP, and membrane dynamics via dynamin 2 (supplementary material Fig. S7). Thus, our study provides further evidence for an important role of BAR-domain-containing proteins as coordinators of membrane invagination, fission and actin dynamics during CME.

Materials and Methods

DNA constructs

SNX9 was amplified by PCR and the PCR product was subcloned into pEGFP (Clontech, Mountain View, CA), mRFP vector (a kind gift from Roger Tsien, UCSD, CA) and pGEX-4T1 vectors (Amersham Biosciences, Piscataway, NJ). The following deletion constructs were PCR amplified and subcloned into expression vectors: SNX9-SH3 domain (1-67), SNX9-ΔSH3 domain (68-595), SNX9-PXBAR domain (246-595), SNX9-BAR domain (357-595), and SNX9-ΔBAR domain (1-356). PtdIns(4)P-5-kinases, GFP-Dynamin-2 and GFP-Dynamin 2 K44A, synaptojanin were kindly provided by Pietro De Camilli (Yale University, New Haven, CT) and were subcloned into expression vectors. The kinase dead mutant of PtdIns(4)P-5-kinase Iα (D309N/R427Q) was provided by Richard Anderson (University of Wisconsin, Madison, WI). GFP-N-WASP was provided by Woo Keun Song (GIST, Korea) and full-length N-WASP and N-WASP-ΔVCA domain (1-391) were subcloned into expression vectors. All DNA constructs were confirmed by DNA sequencing.

Antibodies and reagents

The following antibodies were used. Anti-SNX9 (provided by Sven R. Carlsson, Umeå University, Umeå, Sweden or Santa Cruz Biotechnology, Santa Cruz, CA), anti-Hudy2 and anti-Arp3 (Upstate Biotechnology, Lake Placid, NY), anti-N-WASP and anti-PtdIns(4)P-5-kinase Iα, Iβ and Iγ (Santa Cruz Biotechnology), anti-GFP (Abcam, Cambridge, UK), anti-Flag (Sigma, St. Louis, MO), anti-His (Qiagen, Valencia, CA), anti-HA (Covance, Princeton, NJ) and anti-EEA1 (BD biosciences, San Diego, CA). Secondary antibodies were obtained from Jackson ImmunoResearch. Lysotracker, CM-DiI, Cholera Toxin B Alexa Fluor 594, and Texas Red-transferrin were from Molecular Probes (Eugene, OR). Latrunculin B was from Calbiochem (San Diego, CA), Jasplakinolide was from Invitrogen (San Diego, CA) and all other reagents were from Sigma.

Fig. 7.
Fig. 7.

Perturbation of actin dynamics enhances tubule formation by SNX9. (A) COS-7 cells were transfected with low levels of GFP-SNX9, which does not induce membrane tubulation (left). After addition of latrunculin B (1 μM), tubular invaginations can be observed within a few seconds (right). Insets are high-magnification views and below are time-lapse images of the regions enclosed in rectangles. (B) Cells expressing low levels of GFP-SNX9 were imaged before and after the addition of jasplakinolide (2 μM). Scale bar: 30 μm; 2 μm for insets and time-lapse images. (C) COS-7 cells expressing GFP-tagged SNX9 was stained with 15 μM FM5-95 (C) for 10 minutes on ice and treated with latrunculin B. SNX9-induced tubules (green) and FM5-95-stained plasma membranes (red) were mostly colocalized. High-magnification views are of the regions enclosed in rectangles. Scale bars: 30 μm (low magnification), 5 μm (insets).

GST pull-down assays

The GST-SNX9 and its truncated mutant or His-PtdIns(4)P-5-kinases vector were transformed into E. coli BL-21 and cultured in 2-YT medium supplemented with ampicillin. After overnight induction with 0.5 mM IPTG at 25°C, the cells were sonicated in the lysis buffer (1% Triton X-100, 0.5% sodium deoxycholate, 20 mM Tris, pH 8.0, 150 mM NaCl, 1 mM MgCl2, 1 mM EGTA, 0.1 mM PMSF for GST-fusion protein or 50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 10 mM imidazole for His-tagged protein), centrifuged for 15 minutes at 12,000 g, the supernatant cell lysates were incubated with glutathione-agarose-4B beads (Amersham Biosciences) for GST-fusion protein or Ni-NTA agarose (Qiagen) for His-tagged protein at 4°C for 30 minutes. After washing with lysis buffer three times, the beads fused with GST fusion proteins were incubated for 2 hours at 4°C with brain lysates, which were extracted from rat brain with lysis buffer. The beads were then washed extensively with lysis buffer and analyzed by SDS-PAGE and used for immunoblotting. To purify GST-fusion or His-tagged protein, after washed with lysis buffer, the protein was eluted with an elution buffer (50 mM Tris-HCl pH 8.0, 0.15 M NaCl, 20 mM reduced glutathione for GST-fusion protein, 50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 3 M imidazole for His-tagged protein). The purified proteins were analyzed by SDS-PAGE and proteins were quantified with BSA as the standard protein.

Immunoprecipitation

HEK 293T cells were transfected with Flag-SNX9 or Flag-SNX9-ΔSH3 with GFP-N-WASP or with GFP-dynamin 2 using Lipofectamine 2000 (Invitrogen). Cells were washed twice with cold PBS and extracted for 1 hour at 4°C in a modified RIPA buffer (50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 150 mM NaCl, 1% NP-40, 1 mM sodium orthovanadate, 1 mM PMSF, 10 mM leupeptin, 1.5 mM pepstatin and 1 mM aprotinin). The extracts were then clarified by centrifugation at 12,000 g for 10 minutes, and concentration of the protein in the supernatants was determined using Bradford Protein Assay Reagent kit (Bio-Rad, Hercules, CA). Samples containing 1 mg total protein were then taken for subsequent immunoprecipitation for 9 hours with anti-Flag antibody, followed by an additional 4 hours of incubation at 4°C with protein-G-Sepharose beads (Amersham Biosciences). The immunoprecipitates were extensively washed with the lysis buffer and then subjected to SDS-PAGE and were used for immunoblot analysis.

In vitro tubulation of liposome

Liposome preparation, in vitro tubulation reaction, negative staining and electron microscopy were performed as previously described (Farsad et al., 2001; Takei et al., 1998) with modified buffer A (20 mM HEPES-NaOH pH 7.4, 100 mM NaCl, 1 mM MgCl2, 1 mM DTT).

In vitro tubulation of flat membrane sheets

In vitro tubulation of flat membrane sheets were performed as previously described (Itoh et al., 2005). Briefly, 1 μl droplets of lipid (10 mg/ml Brain Polar Lipid Fraction in chloroform; Avanti Polar Lipids, Alabaster, AL) were spotted on coverslips and allowed to dry under vacuum (0.2 millitorr) for at least 1 hour. After prehydrating the lipids for 20-30 minutes in an incubator (37°C, 10% CO2, 100% humidity), lipids were fully rehydrated by injecting 15-20 μl buffer A containing 0.1 mg/ml casein (Sigma). Just before imaging, 5 μl protein solution (final concentration of 1 mg/ml) was injected into the chamber and membrane deformation was imaged with Olympus IX-71 microscope (Olympus, Tokyo, Japan) in differential interference contrast (DIC) mode.

Protein-lipid overlay assay

Protein-lipid overlay assay was performed using PIP strips (Echelon Research Laboratories, Salt Lake City, UT). Membranes were blocked with 3% (w/v) BSA in Tris-buffered saline-Tween (TBS-T) for 1 hour at room temperature. Membranes were then incubated overnight at 4°C in TBS-T containing 3% BSA, containing 0.5 μg/ml purified GST-SNX9 protein. After incubation, membranes were washed with TBS-T three times for 10 minutes and incubated with an anti-SNX9 antibody for 1 hour at room temperature. Membranes were then probed with HRP-conjugated anti-rabbit antibody (Jackson ImmunoResearch, West Grove, PA), and the immunoreactivity was visualized by enhanced chemiluminescence (Amersham).

Liposome co-sedimentation assay

PE/PC liposomes consisted of PE (70%), PC (20%) and 10% various phosphoinositides. PE, PC and various phosphoinositides were purchased from Sigma-Aldrich and Cell Signals (Columbus, OH). These lipid mixtures were dried under nitrogen gas and resuspended in 50 μl buffer (1 mg/ml in 0.1 M sucrose, 20 mM HEPES, pH 7.4, 100 mM KCl and 1 mM EDTA) for 1-2 hours. To remove aggregated proteins, purified proteins were subjected to centrifugation at 150,000 g for 30 minutes at 4°C. 5 μg proteins were incubated with 100 μg liposomes in buffer for 15 minutes at RT and centrifuged at 100,000 g for 15 minutes at 25°C. The intensity of protein bands was measured using Image J software (National Institutes of Health).

Cell culture, transfection and immunocytochemistry

COS-7 cells were cultured at 37°C and 5% CO2 in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT). Transfection was carried out using Lipofectamine 2000 (Invitrogen), and cells were observed after 16-24 hours. Fluorescence images were acquired on a Olympus IX-71 inverted microscope with a 60× or 100× 1.4 NA oil-immersion lens using a CoolSNAP-HQ CCD camera (Roper Scientific) driven by MetaMorph Imaging software (Universal Imaging Cooperation, West Chester, PA). For immnocytochemistry, cells were fixed in 4% formaldehyde, 4% sucrose, PBS for 15 minutes, permeabilized for 5 minutes in 0.25% Triton X-100, PBS and blocked for 30 minutes in 10% BSA, PBS at 37°C. The cells were incubated with primary antibodies, 3% BSA, PBS for 2 hours at 37°C or overnight at 4°C, washed in PBS, and incubated with secondary antibodies, 3% BSA, PBS for 45 minutes at 37°C.

Fig. 8.
Fig. 8.

SNX9 interacts via its PX domain with PtdIns(4)P-5-kinases Iα, Iβ and Iγ, which regulate the tubulation activity of SNX9 and whose kinase activity is increased by SNX9. (A) COS-7 cell lysates were immunoprecipitated with anti-SNX9 antibody and immunoblotted (IB) with anti-SNX9 antibody or anti-PtdIns(4)P-5-kinase Iα antibody. TCL, total cell lysates. (B) In vitro GST-pull down assay was performed with purified GST-fused SNX9 and its truncated mutants (2 μg each) with His-tagged PtdIns(4)P-5-kinase Iα (200 ng). (C) COS-7 cells were cotransfected with GFP-tagged SNX9, SNX9-ΔPX or SNX9-PX with HA-tagged PtdIns(4)P-5-kinase Iα, Iβ or Iγ and immunoprecipitation (IP) was carried out with anti-GFP antibody, followed by immunoblotting (IB) with anti-HA or anti-GFP antibody. (D) Representative images of cells that were cotransfected with GFP-tagged SNX9 and HA-tagged PtdIns(4)P-5-kinase Iα, Iβ or empty vector. When HA-tagged PtdIns(4)P-5-kinase Iα or Iβ were coexpressed with low levels of GFP-tagged SNX9 in COS-7 cell, tubular invaginations were significantly increased around the cell periphery. (E) The kinase dead mutant of PtdIns(4)P-5-kinase Iα (D309N/R427Q) failed to increase tubulation by low levels of GFP-tagged SNX9 in COS-7 cells. (F) Coexpression of mRFP-tagged SNX9 with GFP-tagged 5-phosphatase domain of synaptojanin inhibits tubule formation even with high levels of SNX9 expression. Scale bars: 30 μm (low magnification), 2 μm (insets). (G) Recombinant PtdIns(4)P-5-kinase Iγ-90 (PIPKγ) was pre-incubated with two concentrations of SNX9 and its activity measured using a PtdIns(4,5)P2-radiolabeling assay. At 1.0 μM SNX9, the kinase activity of PIPKγ was not significantly increased, but at 5.0 μM, there was a significant 2.5-fold increase in activity (**P<0.0001). (H) PtdIns(4)P-5-kinase Iγ (both the 87 kDa and 90 kDa isoforms) was immunoprecipitated from rat brain cytosol and pre-incubated with two concentrations of SNX9 and its activity assayed as in G. At both 1.0 and 5.0 μM SNX9, the immunoprecipitated kinase showed a twofold (**P<0.001) and fivefold (**P<0.05) significant increase in kinase activity, respectively.

Live-cell imaging and image analysis

The transfected COS-7 cells were observed 16 hours after transfection. Immediately before image acquisition, cells were mounted in a temperature-controlled perfusion chamber (Chamlide CF-S-10, LCI, Seoul, Korea) on the stage of an Olympus IX-71 microscope.

Images were acquired with a 100× oil-immersion objective lens, N.A. 1.4 using a CoolSNAP-ES CCD camera driven by MetaMorph Imaging software with a GFP- or mRFP-optimized filter set (Omega Optical, Brattleboro, VT). Simultaneous dual color imaging was performed using a Dual-View (Optical Insights, Tucson, AZ) with a GFP- or mRFP-optimized dual filter set (Chroma, Rochingham, VT). TIRF microscope imaging was performed using objective-type TIRFM set-up from Olympus or from Zeiss with a Cascade EMCCD camera (Roper Scientific). The 488 nm line from a 20 mW argon laser (Melles Griot, Carlsbad, CA) is focused into the back focal plane of a 60× objective lens (NA 1.40). The beam exited the lens at an incident angle, which gives a calculated penetration depth around 200 nm. The length of tubules was measured manually using a MetaMorph software. Three rectangular areas (20×20 μm) near the periphery per cell were selected, and the length of the tubule which was well separated from other tubules was measured in x-y dimension. Since we disregarded the tubule length in the x-z dimension, the measurement would be a conservative estimate. Since tubules grow over time, the figures that we have represented in Fig. 6 were at stages when tubulation occurred almost at a saturated level. If we took the picture earlier when the tubules had just started to grow, we could measure the length of tubules with considerably less variation. Three independent experiments were performed and the data were pooled and normalized against those of full-length SNX9.

siRNA protein knockdown and transferrin uptake assay

siRNA for SNX9 was designed from nucleotides 577-597 of the human SNX9 cDNA sequence (GenBank accession number NM_016224). Complementary oligonucleotides were synthesized separately, with the addition of an Acc65I site at the 5′ end and a HindIII site at the 3′ end. The target sequence was 5′-AACAGTCGTGCTAGTTCCTCA-3′ whereas 5′-AATTCTCCGAACGTGTCACGT-3′ was used for control. The annealed cDNA fragment was cloned into the Acc65I-HindIII sites of the vector psiRNA-hH1GFPzeo G2 vector (InvivoGen, San Diego, CA). After transfection, the cells were incubated for 48 hours for immunoblotting and transferrin uptake assay. Transferrin-uptake assay was carried out as previously described (Kim et al., 2006).

Lipid kinase PI(4,5)P2 radiolabeling assay

Immunoprecipitation of PtdIns(4)P-5-kinase Iγ (both the 87 kDa and 90 kDa isoforms) were carried out using a rabbit polyclonal antibody raised against the C-terminus of PtdIns(4)P-5-kinase Iγ-90 (a kind gift from Pietro De Camilli, Yale University) and rat brain cytosol as the protein source. For recombinant proteins, His-tagged PtdIns(4)P-5-kinase Iγ-90 was purified using immobilized metal affinity chromatography (TALON resin, BD Biosciences) and GST-tagged human SNX9 was purified using glutathione beads (GE Healthcare). In the lipid kinase assay, after pre-incubation with SNX9 on ice, PtdIns(4)P-5-kinase Iγ immunoprecipitates or recombinant protein were incubated for 15 minutes at 37°C in a total reaction volume of 50 μl kinase buffer (25 mM HEPES pH 7.4, 100 mM KCl, 1 mM EGTA, 2 mM MgCl2) containing 60 μg brain lipids and phosphoinositides (Sigma B1502 and P6023) in micelles, 5 μCi [γ-32P]ATP and 50 μM cold ATP. Samples were then processed as previously described (Di Paolo et al., 2002).

Acknowledgments

This research was supported by a grant from Korea Research Foundation (KRF-2006-C00515) to S.C. funded by Korean Government (MOEHRD, Basic Research Promotion Fund), and by a grant from the Molecular and Cellular BioDiscovery Research Program (M10601000102-06N0100-10210) to S.C. funded by the Ministry of Science and Technology, the Republic of Korea. G.D.P. and B.C-I. are funded by NIH grants R01 NS056049 and F31 NS058096, respectively.

Footnotes

  • Accepted January 24, 2008.

References

  1. ↵
    Badour, K., McGavin, M. K., Zhang, J., Freeman, S., Vieira, C., Filipp, D., Julius, M., Mills, G. B. and Siminovitch, K. A. (2007). Interaction of the Wiskott-Aldrich syndrome protein with sorting nexin 9 is required for CD28 endocytosis and cosignaling in T cells. Proc. Natl. Acad. Sci. USA 104, 1593-1598.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Bubb, M., Spector, I., Beyer, B. and Fosen, K. (2000). Effects of jasplakinolide on the kinetics of actin polymerization. An explanation for certain in vivo observations. J. Biol. Chem. 275, 5163-5170.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Carlton, J. G. and Cullen, P. J. (2005). Sorting nexins. Curr. Biol. 15, R819-R820.
    OpenUrlCrossRefPubMed
  4. ↵
    Cestra, G., Castagnoli, L., Dente, L., Minenkova, O., Petrelli, A., Migone, N., Hoffmuller, U., Schneider-Mergener, J. and Cesareni, G. (1999). The SH3 domains of endophilin and amphiphysin bind to the proline-rich region of synaptojanin 1 at distinct sites that display an unconventional binding specificity. J. Biol. Chem. 274, 32001-32007.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Cho, W. and Stahelin, R. V. (2005). Membrane-protein interactions in cell signaling and membrane trafficking. Annu. Rev. Biophys. Biomol. Struct. 34, 119-151.
    OpenUrlCrossRefPubMedWeb of Science
  6. ↵
    Dawson, J. C., Legg, J. A. and Machesky, L. M. (2006). Bar domain proteins: a role in tubulation, scission and actin assembly in clathrin-mediated endocytosis. Trends Cell Biol. 16, 493-498.
    OpenUrlCrossRefPubMedWeb of Science
  7. ↵
    Di Paolo, G. and De Camilli, P. (2006). Phosphoinositides in cell regulation and membrane dynamics. Nature 443, 651-657.
    OpenUrlCrossRefPubMedWeb of Science
  8. ↵
    Di Paolo, G., Pellegrini, L., Letinic, K., Cestra, G., Zoncu, R., Voronov, S., Chang, S., Guo, J., Wenk, M. R. and De Camilli, P. (2002). Recruitment and regulation of phosphatidylinositol phosphate kinase type 1 gamma by the FERM domain of talin. Nature 420, 85-89.
    OpenUrlCrossRefPubMedWeb of Science
  9. ↵
    Farsad, K. and De Camilli, P. (2003). Mechanisms of membrane deformation. Curr. Opin. Cell Biol. 15, 372-381.
    OpenUrlCrossRefPubMedWeb of Science
  10. ↵
    Farsad, K., Ringstad, N., Takei, K., Floyd, S. R., Rose, K. and De Camilli, P. (2001). Generation of high curvature membranes mediated by direct endophilin bilayer interactions. J. Cell Biol. 155, 193-200.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Gallop, J. L. and McMahon, H. T. (2005). BAR domains and membrane curvature: bringing your curves to the BAR. Biochem. Soc. Symp. 72, 223-231.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Gallop, J. L., Jao, C. C., Kent, H. M., Butler, P. J., Evans, P. R., Langen, R. and McMahon, H. T. (2006). Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J. 25, 2898-2910.
    OpenUrlCrossRefPubMedWeb of Science
  13. ↵
    Grabs, D., Slepnev, V. I., Songyang, Z., David, C., Lynch, M., Cantley, L. C. and De Camilli, P. (1997). The SH3 domain of amphiphysin binds the proline-rich domain of dynamin at a single site that defines a new SH3 binding consensus sequence. J. Biol. Chem. 272, 13419-13425.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Gruenberg, J. (2003). Lipids in endocytic membrane transport and sorting. Curr. Opin. Cell Biol. 15, 382-388.
    OpenUrlCrossRefPubMedWeb of Science
  15. ↵
    Habermann, B. (2004). The BAR-domain family of proteins: a case of bending and binding? EMBO Rep. 5, 250-255.
    OpenUrlAbstract
  16. ↵
    Howard, L., Nelson, K. K., Maciewicz, R. A. and Blobel, C. P. (1999). Interaction of the metalloprotease disintegrins MDC9 and MDC15 with two SH3 domain-containing proteins, endophilin I and SH3PX1. J. Biol. Chem. 274, 31693-31699.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Itoh, T. and De Camilli, P. (2006). BAR, F-BAR (EFC) and ENTH/ANTH domains in the regulation of membrane-cytosol interfaces and membrane curvature. Biochim. Biophys. Acta 1761, 897-912.
    OpenUrlCrossRefPubMedWeb of Science
  18. ↵
    Itoh, T., Erdmann, K. S., Roux, A., Habermann, B., Werner, H. and De Camilli, P. (2005). Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins. Dev. Cell 9, 791-804.
    OpenUrlCrossRefPubMedWeb of Science
  19. ↵
    Kaksonen, M., Toret, C. P. and Drubin, D. G. (2006). Harnessing actin dynamics for clathrin-mediated endocytosis. Nat. Rev. Mol. Cell Biol. 7, 404-414.
    OpenUrlCrossRefPubMedWeb of Science
  20. ↵
    Kessels, M. M. and Qualmann, B. (2004). The syndapin protein family: linking membrane trafficking with the cytoskeleton. J. Cell Sci. 117, 3077-3086.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Kim, S., Kim, H., Chang, B., Ahn, N., Hwang, S., Di Paolo, G. and Chang, S. (2006). Regulation of transferrin recycling kinetics by PtdIns[4,5]P2 availability. FASEB J. 20, 2399-2401.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Kojima, C., Hashimoto, A., Yabuta, I., Hirose, M., Hashimoto, S., Kanaho, Y., Sumimoto, H., Ikegami, T. and Sabe, H. (2004). Regulation of Bin1 SH3 domain binding by phosphoinositides. EMBO J. 23, 4413-4422.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Kovacs, E. M., Makar, R. S. and Gertler, F. B. (2006). Tuba stimulates intracellular N-WASP-dependent actin assembly. J. Cell Sci. 119, 2715-2726.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Lin, Q., Lo, C. G., Cerione, R. A. and Yang, W. (2002). The Cdc42 target ACK2 interacts with sorting nexin 9 (SH3PX1) to regulate epidermal growth factor receptor degradation. J. Biol. Chem. 277, 10134-10138.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Lundmark, R. and Carlsson, S. R. (2003). Sorting nexin 9 participates in clathrin-mediated endocytosis through interactions with the core components. J. Biol. Chem. 278, 46772-46781.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Lundmark, R. and Carlsson, S. R. (2004). Regulated membrane recruitment of dynamin-2 mediated by sorting nexin 9. J. Biol. Chem. 279, 42694-42702.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Machesky, L. M., Mullins, R. D., Higgs, H. N., Kaiser, D. A., Blanchoin, L., May, R. C., Hall, M. E. and Pollard, T. D. (1999). Scar, a WASp-related protein, activates nucleation of actin filaments by the Arp2/3 complex. Proc. Natl. Acad. Sci. USA 96, 3739-3744.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Marchand, J. B., Kaiser, D. A., Pollard, T. D. and Higgs, H. N. (2001). Interaction of WASP/Scar proteins with actin and vertebrate Arp2/3 complex. Nat. Cell Biol. 3, 76-82.
    OpenUrlCrossRefPubMedWeb of Science
  29. ↵
    Martin, T. F. (2001). PI(4,5)P(2) regulation of surface membrane traffic. Curr. Opin. Cell Biol. 13, 493-499.
    OpenUrlCrossRefPubMedWeb of Science
  30. ↵
    Masuda, M., Takeda, S., Sone, M., Ohki, T., Mori, H., Kamioka, Y. and Mochizuki, N. (2006). Endophilin BAR domain drives membrane curvature by two newly identified structure-based mechanisms. EMBO J. 25, 2889-2897.
    OpenUrlCrossRefPubMedWeb of Science
  31. ↵
    Maxfield, F. R. and McGraw, T. E. (2004). Endocytic recycling. Nat. Rev. Mol. Cell Biol. 5, 121-132.
    OpenUrlCrossRefPubMedWeb of Science
  32. ↵
    Panchal, S. C., Kaiser, D. A., Torres, E., Pollard, T. D. and Rosen, M. K. (2003). A conserved amphipathic helix in WASP/Scar proteins is essential for activation of Arp2/3 complex. Nat. Struct. Biol. 10, 591-598.
    OpenUrlCrossRefPubMedWeb of Science
  33. ↵
    Peter, B. J., Kent, H. M., Mills, I. G., Vallis, Y., Butler, P. J., Evans, P. R. and McMahon, H. T. (2004). BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 303, 495-499.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Pylypenko, O., Lundmark, R., Rasmuson, E., Carlsson, S. R. and Rak, A. (2007). The PX-BAR membrane-remodeling unit of sorting nexin 9. EMBO J. 26, 4788-4800.
    OpenUrlCrossRefPubMedWeb of Science
  35. ↵
    Qualmann, B., Kessels, M. M. and Kelly, R. B. (2000). Molecular links between endocytosis and the actin cytoskeleton. J. Cell Biol. 150, F111-F116.
    OpenUrlFREE Full Text
  36. ↵
    Ringstad, N., Nemoto, Y. and De Camilli, P. (1997). The SH3p4/Sh3p8/SH3p13 protein family: binding partners for synaptojanin and dynamin via a Grb2-like Src homology 3 domain. Proc. Natl. Acad. Sci. USA 94, 8569-8574.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Rohatgi, R., Ma, L., Miki, H., Lopez, M., Kirchhausen, T., Takenawa, T. and Kirschner, M. W. (1999). The interaction between N-WASP and the Arp2/3 complex links Cdc42-dependent signals to actin assembly. Cell 97, 221-231.
    OpenUrlCrossRefPubMedWeb of Science
  38. ↵
    Shimada, A., Niwa, H., Tsujita, K., Suetsugu, S., Nitta, K., Hanawa-Suetsugu, K., Akasaka, R., Nishino, Y., Toyama, M., Chen, L. et al. (2007). Curved EFC/F-BAR-domain dimers are joined end to end into a filament for membrane invagination in endocytosis. Cell 129, 761-772.
    OpenUrlCrossRefPubMedWeb of Science
  39. ↵
    Shin, N., Lee, S., Ahn, N., Kim, S. A., Ahn, S. G., YongPark, Z. and Chang, S. (2007). Sorting nexin 9 interacts with dynamin 1 and N-WASP and coordinates synaptic vesicle endocytosis. J. Biol. Chem. 282, 28939-28950.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Shinozaki-Narikawa, N., Kodama, T. and Shibasaki, Y. (2006). Cooperation of phosphoinositides and BAR domain proteins in endosomal tubulation. Traffic 7, 1539-1550.
    OpenUrlCrossRefPubMedWeb of Science
  41. ↵
    Simpson, F., Hussain, N. K., Qualmann, B., Kelly, R. B., Kay, B. K., McPherson, P. S. and Schmid, S. L. (1999). SH3-domain-containing proteins function at distinct steps in clathrin-coated vesicle formation. Nat. Cell Biol. 1, 119-124.
    OpenUrlCrossRefPubMedWeb of Science
  42. ↵
    Slepnev, V. I. and De Camilli, P. (2000). Accessory factors in clathrin-dependent synaptic vesicle endocytosis. Nat. Rev. Neurosci. 1, 161-172.
    OpenUrlPubMed
  43. ↵
    Soulet, F., Yarar, D., Leonard, M. and Schmid, S. L. (2005). SNX9 regulates dynamin assembly and is required for efficient clathrin-mediated endocytosis. Mol. Biol. Cell 16, 2058-2067.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Takei, K., Haucke, V., Slepnev, V., Farsad, K., Salazar, M., Chen, H. and De Camilli, P. (1998). Generation of coated intermediates of clathrin-mediated endocytosis on protein-free liposomes. Cell 94, 131-141.
    OpenUrlCrossRefPubMedWeb of Science
  45. ↵
    Takei, K., Slepnev, V. I., Haucke, V. and De Camilli, P. (1999). Functional partnership between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat. Cell Biol. 1, 33-39.
    OpenUrlCrossRefPubMedWeb of Science
  46. ↵
    Tsujita, K., Suetsugu, S., Sasaki, N., Furutani, M., Oikawa, T. and Takenawa, T. (2006). Coordination between the actin cytoskeleton and membrane deformation by a novel membrane tubulation domain of PCH proteins is involved in endocytosis. J. Cell Biol. 172, 269-279.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    van Meer, G. and Sprong, H. (2004). Membrane lipids and vesicular traffic. Curr. Opin. Cell Biol. 16, 373-378.
    OpenUrlCrossRefPubMedWeb of Science
  48. ↵
    Weissenhorn, W. (2005). Crystal structure of the endophilin-A1 BAR domain. J. Mol. Biol. 351, 653-661.
    OpenUrlCrossRefPubMedWeb of Science
  49. ↵
    Worby, C. A. and Dixon, J. E. (2002). Sorting out the cellular functions of sorting nexins. Nat. Rev. Mol. Cell Biol. 3, 919-931.
    OpenUrlCrossRefPubMedWeb of Science
  50. ↵
    Worby, C. A., Simonson-Leff, N., Clemens, J. C., Kruger, R. P., Muda, M. and Dixon, J. E. (2001). The sorting nexin, DSH3PX1, connects the axonal guidance receptor, Dscam, to the actin cytoskeleton. J. Biol. Chem. 276, 41782-41789.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    Yarar, D., Waterman-Storer, C. M. and Schmid, S. L. (2007). SNX9 couples actin assembly to phosphoinositide signals and is required for membrane remodeling during endocytosis. Dev. Cell 13, 43-56.
    OpenUrlCrossRefPubMedWeb of Science
  52. ↵
    Yarar, D., Surka, M. C., Leonard, M. C. and Schmid, S. L. (2008). SNX9 activities are regulated by multiple phosphoinositides through both PX and BAR domains. Traffic 9, 133-146.
    OpenUrlCrossRefPubMedWeb of Science
  53. ↵
    Zimmerberg, J. and McLaughlin, S. (2004). Membrane curvature: how BAR domains bend bilayers. Curr. Biol. 14, R250-R252.
    OpenUrlCrossRefPubMedWeb of Science
View Abstract
Back to top

 Download PDF

Email
Research Article
SNX9 regulates tubular invagination of the plasma membrane through interaction with actin cytoskeleton and dynamin 2
Narae Shin, Namhui Ahn, Belle Chang-Ileto, Joohyun Park, Kohji Takei, Sang-Gun Ahn, Soo-A Kim, Gilbert Di Paolo, Sunghoe Chang
Journal of Cell Science 2008 121: 1252-1263; doi: 10.1242/jcs.016709
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Citation Tools
Alerts

Article navigation

Other journals from The Company of Biologists

Advertisement

2020 at The Company of Biologists

Despite the challenges of 2020, we were able to bring a number of long-term projects and new ventures to fruition. While we look forward to a new year, join us as we reflect on the triumphs of the last 12 months.

Mole – The Corona Files

"This is not going to go away, 'like a miracle.' We have to do magic. And I know we can."

Mole continues to offer his wise words to researchers on how to manage during the COVID-19 pandemic.

Cell scientist to watch – Christine Faulkner

In an interview, Christine Faulkner talks about where her interest in plant science began, how she found the transition between Australia and the UK, and shares her thoughts on virtual conferences.

Read & Publish participation extends worldwide

“The clear advantages are rapid and efficient exposure and easy access to my article around the world. I believe it is great to have this publishing option in fast-growing fields in biomedical research.”

Dr Jaceques Behmoaras (Imperial College London) shares his experience of publishing Open Access as part of our growing Read & Publish initiative. We now have over 60 institutions in 12 countries taking part – find out more and view our full list of participating institutions.

JCS and COVID-19

For more information on measures Journal of Cell Science is taking to support the community during the COVID-19 pandemic, please see here.

If you have any questions or concerns, please do not hestiate to contact the Editorial Office.