The Gcs1 Arf-GAP Mediates Snc1,2 v-SNARE Retrieval to the Golgi in Yeast

INTRODUCTION

Protein and lipid transport between intracellular compartments is required for the functional and structural integrity of organelles in eukaryotic cells. This transport is mediated by carrier vesicles generated by protein-based coat complexes. The COPI coat consists of the Arf1 small GTPase and coatomer and confers intra-Golgi and Golgi-to-ER retrograde transport (Kreis et al., 1995; Kirchhausen, 2000; Spang, 2002; Nie et al., 2003). Coatomer consists of seven subunits: αCOP (Sec33, 160 kDa); βCOP (Sec26, 110 kDa); β′COP (Sec27, 102 kDa); γ COP (Sec21, 98 kDa); δCOP (60 kDa); ϵ COP (Sec28, 35 kDa); and ζCOP (20 kDa), which are conserved from yeast to mammals. Coatomer subunits can be divided into two subcomplexes: the B subcomplex (COPI B) composed of the α, β′, and ϵ subunits; and the F subcomplex (COPI F), consisting of the β, δ, γ, and ζ subunits (Eugster et al., 2000; McMahon and Mills, 2004). Interestingly, the γ subunit of COPI F shows structural similarity to components of the clathrin adaptor, AP2 (Hoffman et al., 2003), whereas COPI B has been suggested to be clathrinlike (McMahon and Mills, 2004).

Arf GTPases undergo a cycle of GTP binding and hydrolysis to regulate vesicle formation from a variety of intracellular compartments, such as the Golgi and endosomes (Kreis et al., 1995; Spang, 2002; Nie et al., 2003). In the GTP-bound and myristoylated form, Arf1 becomes membrane-bound and drives COPI vesicle formation in vitro from Golgi membranes and defined liposomes (Spang et al., 1998). Membrane-localized GTP-bound Arf1 recruits coatomer to create a molecular platform capable of deforming the lipid bilayer and leading to vesiculation. Thus, the regulation of Arf function is critical for the initial steps that lead to vesicle biogenesis.

GTP binding and hydrolysis on Arf is mediated by guanine nucleotide exchange factors and GTPase-activating proteins, known as Arf-GEFs and Arf-GAPs, respectively. Because vesicle uncoating is a prerequisite for fusion, Arf-GAP activity is thought to fulfill a role in the removal of coatomer, although roles in cargo packaging and vesicle formation have also been shown (Nickel et al., 1998; Pepperkok et al., 2000; Lanoix et al., 2001; Rein et al., 2002). Although in vitro studies suggest that myristoylated Arf1 and coatomer are sufficient to drive vesicle formation (Spang et al., 1998; Springer et al., 1999), these studies utilized amounts of protein that are unlikely to be physiological. Thus, other factors may be needed for Arf recruitment to membranes. These factors include p23 family members and SNAREs (Gommel et al., 2001; Rein et al., 2002). Recent work demonstrated that yeast Arf1 interacts directly with the SNAREs involved in ER-Golgi transport in a manner requiring Arf-GAP catalytic activity (Rein et al., 2002; Randazzo and Hirsch, 2004). Moreover, Arf-GAP function was shown to be sufficient for COPI coat recruitment even in the absence of activated Arf (Reinhard et al., 1999; Rein et al., 2002; Yang et al., 2002; Lee et al., 2004). Thus, it has been suggested that Arf-GAP is a component of coatomer (Yang et al., 2002; Lewis et al., 2004) and that its catalytic activity is necessary for coat formation and vesicle production (Lee et al., 2004).

Rein et al. (2002) found that yeast COPI (e.g., Arf1 and coatomer) bound to the Bet1, Bos1, and Sec22 ER-Golgi SNAREs in vitro (Rein et al., 2002). This binding occurred only after preincubation of the SNAREs with either of the two Arf-GAPs known to facilitate ER-Golgi transport (e.g., Gcs1 and Glo3; Poon et al., 1999). As SNAREs are central components of the vesicle docking and fusion machinery (reviewed in Chen and Scheller, 2001), Arf1-SNARE-coat interactions may be required to generate SNARE-equipped fusion-competent vesicles in vivo.

Because Arf1-GTP and coatomer binding to the SNAREs is Arf-GAP-dependent, it implies that Gcs1 and Glo3 may have two distinct functions. The first is to initiate Arf1-GTP binding, presumably by inducing a conformational change in SNARE structure and allowing for coat association. The second function is to uncoat the vesicle after budding has occurred by catalyzing GTP hydrolysis on Arf1. This additional role presumably allows the uncoated vesicles to undergo docking and fusion at the appropriate acceptor compartment.

We have been studying the Snc1 and Snc2 v-SNAREs that participate in both exocytosis and endocytosis in yeast (Protopopov et al., 1993; Gurunathan et al., 2000). These v-SNAREs partner with the Sso1,2 and Sec9 t-SNAREs to mediate exocytic functions (Brennwald et al., 1994; Couve and Gerst, 1994) and with the Tlg1,2 and Vti1 t-SNAREs to mediate endocytic functions (Bryant and James, 2003). Thus, the Snc v-SNAREs, which are members of the synaptobrevin/VAMP family, engage in multiple transport steps and recycle continually between the plasma membrane and trans-Golgi via early endosomes (Lewis et al., 2000; Hettema et al., 2003). Efficient Snc1 recycling to the early endosome requires the sorting nexin, Snx4, which is involved in protein retrieval from endosomes to the Golgi (Hettema et al., 2003). Here we show that the Snc1,2 v-SNAREs and Gcs1 Arf-GAP interact physically and genetically, leading to v-SNARE recycling to the trans-Golgi. This recycling process appears to involve coatomer, as well as Snx4, and thus, may represent a novel trafficking pathway from sorting endosomes back to the Golgi.

MATERIALS AND METHODS

Media, DNA, and Genetic Manipulations

Yeast were grown in standard growth media containing either 2% glucose or 3.5% galactose. Synthetic complete (SC) and drop-out media were prepared similar to that described (Rose et al., 1990). Standard methods were used for the introduction of DNA into yeast and the preparation of genomic DNA (Rose et al., 1990).

Growth Tests

Yeast were grown on synthetic and rich growth media (Rose et al., 1990). For cold-sensitive growth tests on plates, yeast were grown to stationary phase, normalized for optical density, diluted serially, and plated by drops onto solid medium preincubated at different temperatures. For growth tests involving sncΔ or sncΔ gcs1Δ cells, which carry a galactose-inducible form of SNC1, cells were first grown to stationary phase on galactose-containing synthetic medium. Next, a portion of the cells was shifted to glucose-containing medium for 24 h to induce the sncΔ phenotype. Cultures were then normalized for optical density, diluted serially, and plated by drops onto solid medium preincubated at different temperatures. For temperature-sensitive growth of COPI mutants, cells were grown to midlog phase on synthetic medium before normalization, serial dilution, and plating onto solid medium preincubated at different temperatures. Calcofluor resistance was measured by adding between 50 and 150 μg/ml fluorescent brightener 28 (Sigma, St. Louis, MO) to plates and plating serial dilutions of the different strains by drops.

Yeast Strains and Plasmids

Yeast strains are listed in Table 1. Vectors included: pRS313 (CEN, TRP1); pRS315 (CEN, LEU2); pRS316 (CEN, URA3); YCp50 (CEN, URA3); pAD11 (CEN, HIS3); pRS426 (2μ, URA3); pAD4Δ (2μ, LEU2, ADH1 promoter); and pAD54 and pAD6 (same as pAD4Δ, but containing sequences encoding the HA or Myc epitopes, respectively). Previously described SNC expression plasmids included: pADH-SNC1 (Gerst et al., 1992); pADH-SNC2, pADH-HASNC1, and pTGAL-SNC1 (Protopopov et al., 1993); and pADH-mycSNC2 and pADH-HASNC2 (Lustgarten and Gerst, 1999). Previously described plasmids for the bacterial expression of Gcs1 and Arf1 included pPPL21 and pET-Arf1H, respectively (Poon et al., 1996). A plasmid expressing recombinant N-myristoyltransferase in bacteria, pACY177/ET3d/yNMT (Haun et al., 1993), was used to create myristoylated Arf1, as described (Poon et al., 1996). Plasmid YIplac204-T/C-SEC7-dsRED.T4 was generously provided by B. Glick (University of Chicago, IL).

Plasmids created for this study are listed in Table 2. Sequences of the oligonucleotides used will be provided on request. Disruption constructs for GCS1 that do not interfere with adjacent open reading frames were created by amplifying a region corresponding to 1999 bp upstream of the start codon and 1455 bp downstream of the stop codon of GCS1 from genomic DNA. This PCR product was cloned into pGEM-T-Easy to give pGEM-GCS1. Next, a fragment containing either URA3 or LEU2 was cloned into the PmeI-XbaI sites of GCS1 in pGEM-GCS1. Digestion with PmeI and XbaI resulted in the removal of nucleotides 213-592 from the coding region of GCS1. Subsequent insertion of either the URA3 or LEU2 selectable marker gave plasmids pGCS1::URA3 and pGCS1::LEU2, respectively. A ∼6-kb disruption fragment was excised from pGCS1::URA3 or pGCS1::LEU2 by digestion with NotI and was used to transform both wild-type and sncΔ null cells.

Synthetic Genetic Analysis

For synthetic genetic analysis (SGA), a query strain (PPY169-4) was constructed by replacing the GCS1 gene with a nourseothricin-resistance cassette (Goldstein and McCusker, 1999) via homologous recombination. The cassette was created by PCR amplification using oligonucleotides bearing sequences flanking the GCS1 coding region and plasmid p4339 as a template (Tong et al., 2001). The replacement of GCS1 in PPY169-4 was verified by PCR analysis. An automated approach was then used to cross strain PPY169-4 with the yeast gene-deletion collection, and the resulting haploid double segregants were screened for synthetic-lethal combinations as described previously (Tong et al., 2001).

Microscopy

GFP and RFP fluorescence in strains expressing the appropriate GFP- and DsRedT4/mRFP-tagged fusion proteins was visualized by confocal microscopy (Bio-Rad, Hercules, CA).

Immunoprecipitation

Protein-protein interactions were monitored by the immunoprecipitation (IP) from cell extracts, as described in (Couve and Gerst, 1994) except that a 10 mM Tris (pH 7.5), 1 mM EDTA buffer was substituted for phosphate-buffered saline. Monoclonal antisera included anti-myc antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) and anti-HA antibodies (gift of Michael Wigler, Cold Spring Harbor Laboratory). Anti-myc antibodies were used for IP (4 μl per reaction) and detection (1:1000). Anti-HA antibodies were also used for IP (0.8 μl per reaction) and detection (1:7000). Polyclonal antibodies included anti-Gcs1 antibodies (1:2500). Samples of TCLs and immunoprecipitates were resolved by electrophoresis and detected by Western blotting. Detection was performed using enhanced chemiluminescence (Amersham).

Two-Hybrid Assay

Assessment of the Snc v-SNARE and Gcs1 Arf-GAP interaction in β-galactosidase assays (Figure 1, A and D) was performed using Y153 cells in the yeast two-hybrid assay (Durfee et al., 1993). Transformants were patched onto selective synthetic medium, before being subjected to lifts onto nitrocellulose filters and lysis in liquid nitrogen. β-Galactosidase assays were performed using standard procedures. Assessment of the Snc v-SNARE and Gcs1 Arf-GAP interaction in drop tests (Figure 1B) was performed using AH109 cells in the yeast two-hybrid assay, as described by Durfee et al. (1993). Transformants were grown in liquid selective medium, diluted serially, and plated by drops onto solid medium lacking histidine and containing 0-2 mM 3-aminotriazole.

In Vitro Arf-GAP Activity Assays

To assess the effects of Snc2 on Gcs1 Arf-GAP activity in vitro, recombinant His6-tagged Gcs1 (plasmid pPPL21) and His6-tagged Snc2 (plasmid pPPL92), which lacks the transmembrane domain, were expressed in Escherichia coli BL21 and purified in native form using standard protocols (Poon et al., 2001). First, 50 ng of purified Gcs1 was mixed with varying amounts of His6-Snc2, ranging from 50 ng to 25 μg, and incubated on ice in 80 μl of 12.5% glycerol, 0.125% bovine serum albumin (fraction V), 1.25 mM DTT, 1.25 mM ATP, 1.25 mM MgCl, 187 mM KOAc, and 31.3 mM MOPS buffer at pH 7.5. After 3 h of incubation, GAP activity was assayed by the addition of 20 μl γ-³²P-GTP-bound myristoylated Arf1, incubation at 30°C for 15 min, and subsequent assessment of GTP-hydrolysis, as previously described (Poon et al., 2001).

In Vitro Arf- and Coatomer-binding Assays

Truncated genes encoding SNAREs lacking their transmembrane domains were cloned into vector pETGEXCT (Sharrocks, 1994). N- and C-terminal GST-tagged SNARE fusion proteins were expressed in E. coli and purified using standard procedures. Gcs1 and Arf1ΔN17-Q71L expression in E. coli and purification were performed as described (Rein et al., 2002). Coatomer was purified from yeast as described (Hosobuchi et al., 1992). Pulldown assays employing immobilized SNARE-GST fusion proteins were performed essentially as described (Rein et al., 2002). In brief, 5 μg SNARE-GST fusion proteins were immobilized onto GSH-agarose (Sigma) and subsequently incubated for 1 h at 4°C with 20 nM recombinant Gcs1 in a total reaction volume of 100 μl in BBP (25 mM HEPES, pH 6.8, 300 mM KOAc, 1 mM DTT, 0.5 mM MgCl₂, and 0.2% Triton X-100). Gcs1 was removed from the reaction by three washes with BBP. The beads were incubated with 40 nM coatomer and 7.3 μM recombinant Arf1ΔN17-Q71L for 1 h at 4°C, washed three times with BBP and once with 20 mM HEPES, pH 6.8. The proteins bound to the beads were separated by SDS-PAGE, visualized by Fairbanks staining, and visualized using the Odyssey system (Li-Cor).

RESULTS

The Snc2 v-SNARE Interacts with the Gcs1 Arf-GAP in the Two-Hybrid Assay

The yeast two-hybrid screen was used to identify proteins that interact with the Gcs1 Arf-GAP. Full-length Gcs1 fused to the DNA binding domain of Gal4 was used as bait and screened with a yeast genomic library fused to the transactivating domain of Gal4. One of the candidate prey genes identified in this assay encoded the yeast synaptobrevin/VAMP ortholog, Snc2, as shown in Figure 1A. Interestingly, a systematic genome-wide two-hybrid screen also identified Snc2 as interacting with Gcs1 (Ito et al., 2001). Thus, independent two-hybrid screens suggest that this Arf-GAP and a post-Golgi v-SNARE interact physically. This interaction could be of consequence for post-Golgi vesicular transport as the Snc v-SNAREs mediate both endo- and exocytosis (Protopopov et al., 1993; Gurunathan et al., 2000) and cycle between the plasma membrane and the Golgi (Lewis et al., 2000). Indeed, a role for Gcs1 in post-Golgi vesicle biogenesis has already been proposed (Poon et al., 2001).

To further extend our observations using two-hybrid analysis, we assessed protein-protein interactions with either full-length Snc1 or Snc2 (e.g., Snc1^2-116 and Snc2^2-115) fused to the transactivating domain of Gal4 and Gcs1 fused to the DNA-binding domain (Figure 1B). When tested for the ability to confer growth in the absence of histidine and in the presence of a metabolic inhibitor of His3 (3-aminotriazole [3-AT]), we found that either v-SNARE could do so in the presence of Gcs1. Thus, both Snc1 and Snc2 interact with this Arf-GAP. However, the Snc1-Gcs1 interaction was more sensitive to higher concentrations of 3-AT (Figure 1B), indicating that it may be weaker than that of Snc2-Gcs1. Thus, both lines of experimentation verify an interaction between the Gcs1 Arf-GAP and the Snc1,2 v-SNAREs.

Snc v-SNAREs Coimmunoprecipitate with Gcs1

Because the Snc v-SNAREs interact physically with Gcs1 in the two-hybrid assay (Figure 1, A and B), we next examined whether these proteins coimmunoprecipitate. In the absence of a functional copy of GCS1, yeast cells are unable to reenter the cell cycle at 14°C and are rendered cold-sensitive (Ireland et al., 1994). For the coimmunoprecipitation experiments, we used an HA-tagged version of Gcs1 that was deemed functional by virtue of its ability to confer cold-resistant growth to gcs1-1 and gcs1Δ mutant cells (our unpublished observations). We found that HA-tagged Gcs1 coprecipitated with myc-tagged Snc2 from lysates prepared from wild-type cells expressing these proteins (Figure 1C). A single band corresponding to a molecular mass of about 40 kDa was observed in precipitates from cells expressing both proteins. This signal was specific, but weak, being eliminated by relatively low concentrations of salt (i.e., 130 mM NaCl; our unpublished observations). Thus, these proteins interact physically, but perhaps not tightly, in vivo. Similar results were obtained with myc-Snc1 and HA-Gcs1 (our unpublished observations).

The Arf-GAP Domain of Gcs1 Is Not Required for the Interaction with Snc2

To determine which regions of Gcs1 and Snc2 are required for their physical association, truncated forms of the proteins were tested by two-hybrid analysis (Figure 1D). Deletions in the amino terminus of Gcs1 (Gcs1^49-352, Gcs1^117-352, and Gcs1^137-352), which effectively remove the Arf-GAP domain (amino acid residues 8-129) or portions thereof, did not abolish and even enhanced the interaction with Snc2 isolated in the initial screen (e.g., Snc2^8-115). In contrast, the amino terminus of Gcs1 alone (Gcs1^1-139 and Gcs1^1-226) conferred no β-galactosidase activity when coexpressed with Snc2. This lack of interaction was not due to insufficient protein expression because Western blot analysis indicated that the amino terminal Gcs1 fusion protein was as abundant as the full-length Gcs1 fusion (our unpublished observations). These results suggest that the Arf-GAP domain of Gcs1 is dispensable for the interaction with Snc2.

Deletion of the first 51 amino acids of Snc2 did not abolish the interaction with either full-length Gcs1 or its amino terminus-truncated forms. Thus, the Gcs1-interacting domain resides in the carboxy terminus of the v-SNARE, which encompasses the SNARE domain (Fasshauer et al., 1998). As access to the transmembrane region of Snc2 is unlikely in vivo, this result implies that Gcs1 binds to the t-SNARE-interacting domain of the protein.

A Genetic Interaction between the SNC and GCS1 Genes

We next assessed the significance of the Gcs1-Snc interaction in vivo using a genetic approach. We examined whether the combined disruption of the SNC and GCS1 genes leads to synthetic defects, which would indicate a related function in post-Golgi transport. We first disrupted GCS1 in the same genetic background (SP1 wild-type cells) as that used to generate the sncΔ strain (Protopopov et al., 1993). The disruption of GCS1 led to cold sensitivity on synthetic and rich medium (YPD) at 15°C in the SP1 background (Figure 2A and our unpublished results), as shown previously in the W303 background (Ireland et al., 1994). The disruption of GCS1 in SP1 cells also resulted in an inhibition of growth at 37°C on rich medium (our unpublished results).

We examined whether overexpression of the SNC genes or reexpression of GCS1 could confer cold-resistant growth to gcs1Δ cells. We found that HA-GCS1 expression from a multicopy plasmid or GCS1-RFP expression from a single copy plasmid could confer growth at 15°C to gcs1Δ cells. In contrast, overexpression of the SNC genes had no effect and could not confer growth at 15°C (Figure 2A). This verified that the GCS1 expression constructs are functional and could restore cold-resistant growth to gcs1Δ cells. We note that overexpression of either HA-GCS1 (Figure 2A) or native GCS1 (our unpublished observations) from multicopy plasmids has a mild inhibitory effect on wild-type cells at lower temperatures (15 and 18°C; Figure 2A). This is likely to result from a general inhibitory effect the Arf-GAP has on Arf function.

Next, we disrupted GCS1 in sncΔ null cells, which are temperature-sensitive on synthetic medium and unable to grow on amino acid-rich medium (Protopopov et al., 1993). To maintain viability, SNC1 was expressed in the sncΔ gcs1Δ strain under the control of an inducible GAL promoter and the cells grown on galactose-containing medium. On shifting the cells to glucose-containing medium the sncΔ phenotype becomes apparent after 12 h (Protopopov et al., 1993). We grew sncΔ and sncΔ gcs1Δ cells to stationary phase and examined their growth upon plating onto solid medium at different temperatures. Cells lacking the SNC genes alone were both cold- and temperature-sensitive for growth on glucose-containing medium, as previously shown (Protopopov et al., 1993 and our unpublished results), but could grow slowly at 18-35°C. In contrast, cells disrupted in both the SNC v-SNARE genes and the GCS1 Arf-GAP gene were unable to grow on glucose-containing medium at any temperature (Figure 2B). However, both sncΔ and sncΔ gcs1Δ cells were fully viable when maintained on galactose-containing medium (Figure 2B), whereon Snc1 is expressed. Thus, synthetic lethality is observed between the gcs1Δ and sncΔ mutations and suggests that these gene products provide related functions that allow for an essential transport activity.

Gcs1 Colocalizes with the Snc v-SNAREs

As the Snc v-SNAREs and Gcs1 interact both physically and genetically, we examined whether these proteins colocalize in yeast (Figure 3). We used functional (Figure 2A and our unpublished observations) green and red fluorescent protein (GFP and RFP, respectively) derivatives expressed from low copy plasmids. Both GFP-Snc1 and GFP-Snc2 strongly labeled the yeast plasma membrane and some cytoplasmic structures and weakly labeled vacuolar membranes (Figure 3A). This corresponds well with the pattern of labeling described for GFP-Snc1 (Gurunathan et al., 2000; Lewis et al., 2000). The cytoplasmic structures observed earlier with GFP-Snc1 included the trans-Golgi and endosomes (Lewis et al., 2000; Galan et al., 2001). Importantly, these v-SNAREs have been demonstrated to recycle back to the trans-Golgi via early endosomes in a manner dependent on the involvement of Rcy1, Ric1, and Ypt6, which mediate endosome-Golgi transport (Lewis et al., 2000; Galan et al., 2001; Lafourcade et al., 2004). Rcy1 is an F-box protein involved in recycling at endosomes (Wiederkehr et al., 2000), whereas Ric1 is part of the GEF complex for the rab-GTPase Ypt6 at the TGN (Siniossoglou et al., 2000). In our colabeling experiments, we found that both GFP-Snc1 and GFP-Snc2 localized with Gcs1-RFP at a subset of large punctate structures present in the cytoplasm, but not at the plasma membrane (Figure 3A). The intracellular localization of Gcs1 was previously unknown, though its role in post-Golgi vesicle biogenesis (Poon et al., 2001) suggested a possible Golgi or endosomal localization. Thus, the internal compartments colabeled by Gcs1-RFP and GFP-Snc1 (or GFP-Snc2) are likely to be late Golgi or endosomal in nature.

To better address the intracellular localization of Gcs1, we examined the location of Gcs1 tagged either at the carboxy terminus with RFP or at the amino terminus with GFP along with markers of other endomembrane compartments or structures (Figure 3B). We found that Gcs1 colocalized in part with both Sed5 and Sec7, which are early and late Golgi markers, respectively (Franzusoff et al., 1991; Hardwick and Pelham, 1992). This colocalization was limited to a subset of punctate structures and correlated well with the known functional overlap between Gcs1 and the Glo3 and Age2 Arf-GAPs, which facilitate Golgi-ER and post-Golgi transport, respectively (Poon et al.,1999, 2001). In addition, Gcs1 also colocalized to a subset of compartments labeled by Tlg2 (Figure 3B), a t-SNARE involved in endocytosis and the delivery of proteins to endosomes and the vacuole (Abeliovich et al., 1998; Holthuis et al., 1998; Seron et al., 1998). In contrast, Gcs1 did not colocalize with a preautophagosomal marker, Aut7/Apg8 (Kim et al., 2001; our unpublished observations). Together, these results imply that Gcs1 resides in both Golgi and endosomal compartments.

We also found that Snc2 colocalized well with Sec7 (Figure 3B), as predicted (Lewis et al., 2000; Galan et al., 2001). This colocalization was observed at numerous cytoplasmic structures that are thought to correspond to the trans-Golgi (Lewis et al., 2000; Galan et al., 2001), but which may include early endosomes. In contrast, there was little to no colocalization between GFP-Snc1 (or GFP-Snc2) and the Age2 Arf-GAP (Figure 3C). This implies that the site of interaction between the Snc v-SNAREs and Gcs1 may be distinct from the intracellular locale governed by Age2. Although we cannot exclude the possibility that fluorescent protein-tagged Gcs1, as well as the tagged organellar markers, do not induce changes in the morphology and distribution of intracellular trafficking compartments, our results are consistent with Gcs1 and Snc v-SNAREs colocalizing at late or post-Golgi (endosomal) structures.

Snc-Gcs1 Interactions Do Not Alter Arf-GAP Activity In Vitro

Because the Snc v-SNAREs and Gcs1 interact and colocalize at an endosomal compartment, we examined the functional consequences of this interaction. First we examined whether Snc2 v-SNARE binding to Gcs1 alters its ability to activate GTP hydrolysis by Arf1. Recombinant His₆-Snc2, lacking the transmembrane domain, was mixed with His₆-tagged Gcs1 in the presence of recombinant myristoylated Arf1 prebound to GTP and subsequent GTP hydrolysis was measured in vitro. It was found that the presence of Snc2 had no effect on GTP hydrolysis by Arf1 (Figure 4A). This suggests that the Snc v-SNAREs do not alter Gcs1-mediated GTP-hydrolysis.

Snc-Gcs1 Interactions Promote Arf1 Binding and Coatomer Recruitment In Vitro

Because the Snc v-SNAREs physically interact with Gcs1 (Figure 1), but do not regulate Gcs1 GAP activity in vitro (Figure 4A), we examined whether Gcs1 modulates the binding of Arf1 to the v-SNARE. Previous work suggested that Arf-GAP-SNARE interactions prime the v-SNARE to bind Arf1 and allow for subsequent coat recruitment (Rein et al., 2002). Specifically, the catalytic interaction of either Glo3 or Gcs1 with the ER-Golgi v-SNAREs (e.g., Bet1, Bos1, and Sec22) resulted in the binding of Arf1ΔN17-Q71L and the acquisition of coat in a nucleotide- and GAP activity-independent manner in vitro. Thus, Arf-GAP-SNARE interactions have been proposed to recruit both SNAREs and coat proteins to the sites of vesicle formation (Rein et al., 2002).

To determine whether Gcs1 fulfills a similar role with the Snc v-SNAREs, we substituted Snc1 or Snc2 for the ER-Golgi v-SNAREs in this in vitro binding assay. First, we examined whether GST-tagged Snc1,2 v-SNAREs could precipitate Gcs1 in vitro (Figure 4B). We found that Snc1-GST and Snc2-GST could precipitate recombinant Gcs1 as well as Sec22-GST (Figure 4B). Next, when used in the in vitro binding assay either GST-Snc1 or Snc1-GST was readily able to recruit purified coatomer in a Gcs1-dependent manner (Figure 4C; lanes 4, 6, 10, and 12). This reaction does not appear to require Arf1, as COPI was recruited in its absence (see lanes 4 and 10, Figure 4C). This property was observed previously with the ER-Golgi SNAREs (Rein et al., 2002). Arf1ΔN17-Q71L binding to Snc1 or to Snc2 (our unpublished observations), however, was dependent on the addition of Gcs1 to the in vitro assay (Figure 4C; lanes 2, 6, 8, and 12). In contrast, GST alone was unable to recruit either Arf1ΔN17-Q71L or coatomer (Figure 4D). Thus, the Snc v-SNAREs interact with both Arf and coatomer, which is suggestive of a post-Golgi role for this complex in vivo.

Synthetic Lethal Interactions with gcs1Δ

To identify factors that facilitate Gcs1-mediated post-Golgi transport, we used a nonbiased screen to select for gene deletions that are synthetically lethal in combination with gcs1Δ. By exhaustive screening against a yeast deletion library of nonessential genes, we found that a number of deletions are synthetically lethal in combination with the gcs1Δ mutation. These included deletions in genes encoding the other known Arf-GAPs, GLO3 and AGE2 (Poon et al.,1999, 2001). In addition, the screen identified numerous genes encoding factors involved in Golgi-endosome transport, including: TLG2, VPS1, VPS51, and YPT31 (see Table 3). These findings are consistent with a role for Gcs1 in post-Golgi transport.

Snc v-SNARE Overexpression Inhibits the Growth of Certain Coatomer Mutants

Because the Snc v-SNAREs recruit COPI coat components in a Gcs1-dependent manner in vitro (Figure 4C), we examined the significance of this in vivo. Overexpression of the SNC genes is known to rescue mutations affecting partner t-SNAREs from the plasma membrane (i.e., sec9-4 and sso2-1) as well as a mutation in a SNARE regulator (sec1-1; Couve and Gerst, 1994; Gerst, 1997). We decided to examine the effect of SNC1 and SNC2 overexpression in mutants of coatomer (i.e., sec21-2, sec27-1, and sec33-1). Interestingly, the growth of sec27-1 and sec33-1 cells, which express mutated components of the clathrinlike B subcomplex of COPI (McMahon and Mills, 2004), was significantly inhibited by the overexpression of either SNC1 or SNC2 (Figure 4E). In contrast, the overexpression of either SNC1 or SNC2 in sec21-2 cells, which expresses a mutated component of the adaptor-like F subcomplex of COPI (McMahon and Mills, 2004), had no deleterious effect (Figure 4F). Thus, we could identify genetic interactions between the Snc v-SNAREs and components of the COPI B subcomplex. Because Gcs1 interacts physically with the Snc proteins, we addressed the possibility that overproduction of the v-SNAREs could cause a decrease in the Arf-GAP available for membrane transport. However, an increase in GCS1 expression did not alleviate the effects of SNC overexpression in the COPI B mutants (Figure 4, D and E).

The Deletion of GCS1 Alters GFP-Snc1 Recycling

Because of a functional overlap with Age2, Gcs1 was proposed to play a role in post-Golgi protein sorting (Poon et al., 2001). In addition, the present study shows the colocalization of Gcs1 with Golgi and endosomal markers, as well as physical and genetic interactions with v-SNAREs that facilitate post-Golgi transport. Because GFP-Snc1 recycles through early endosomes back to the Golgi (Lewis et al., 2000), we examined the trafficking of this v-SNARE in cells lacking the GCS1 gene (Figure 5A). In addition, we also examined GFP-Snc1 localization in rcy1Δ cells, which are defective in early endosome-Golgi sorting (Galan et al., 2001), and end4-1 cells, which are defective in the endocytosis of endocytic markers such as GFP-Snc1 (Lewis et al., 2000). Unlike in wild-type cells, we found that GFP-Snc1 accumulated in intracellular compartments in both rcy1Δ and gcs1Δ cells, while being restricted to the plasma membrane in end4-1 cells (Figure 5A). Thus, proper GFP-Snc1 recycling is largely inhibited in the absence of Gcs1. This result is identical to that shown earlier for rcy1Δ and other mutations in proteins that facilitate early endosome-Golgi transport (Galan et al., 2001).

To examine whether GFP-Snc1 reaches the late Golgi in gcs1Δ cells we performed a colocalization study with GFP-Snc1 and Sec7-RFP in both wild-type and gcs1Δ cells (Figure 5B). We found that Snc1 could not colocalize effectively with Sec7 in the absence of GCS1. In contrast, these proteins readily colocalize in wild-type cells (Figures 3B and 5B). Similar results were obtained using another Golgi marker, Yif1 (Matern et al., 2000), that colocalizes in part with Gcs1 (our unpublished observations). We found that GFP-Yif1 could not colocalize with mRFP-Snc1 in the absence of GCS1 (Figure 5C). In contrast, partial colocalization is observed between mRFP-Snc1 and GFP-Yif1 in wild-type cells (Figure 5C). Thus, the ability of Snc1 to recycle to the trans-Golgi (e.g., Sec7 compartment) is dependent on the Gcs1 Arf-GAP. We note that both Golgi markers (Sec7, Yif1) appeared to be more widely distributed and less punctate in gcs1Δ cells, indicating a possible alteration in Golgi morphology in the absence of Gcs1 function.

GFP-Snc1 Localization Is Altered in COPI B and ESCRT Mutants

As GFP-Snc1 retrieval to the Golgi is blocked in gcs1Δ cells (Figure 5) and the SNC genes interact genetically with mutations in COPI B (Figure 4E), we tested whether mutants in the COPI B subcomplex play a role in Snc1 recycling. We followed the localization of GFP-Snc1 in COPI B mutants and a variety of other cell types (Figure 6). In wild-type cells, we found that GFP-Snc1 gave typical plasma membrane staining that was slightly bud-enriched (Lewis et al., 2000), a process requiring endocytosis (Valdez-Taubas and Pelham, 2003). Similar results were observed in COPI B mutants (e.g., sec27-1, sec28Δ) and an ESCRT-I mutant (vps23Δ at 26°C; however, the extent of plasma membrane labeling seen on the buds of these mutants was considerably stronger than that observed on the buds of wild-type cells. This bud-enriched pattern of labeling differed greatly from GFP-Snc1 labeling of the entire plasma membrane in endocytosis-deficient end4-1 cells. In addition, it differed from the extensive pattern of internal GFP-Snc1 labeling seen in a COPI F mutant (sec21-2), as well as sec33-1 cells. This is probably because both sec21-2 and sec33-1 mutants are impaired in transport through the early secretory pathway at 26°C (Wuestehube et al., 1996), unlike sec27-1 and sec28Δ cells. The internal pattern of GFP-Snc1 labeling was reminiscent of that seen in rcy1Δ mutants, which are defective in Snc1 recycling from early endosomes to the Golgi (Galan et al., 2001). Our results imply that Snc1 v-SNARE retrieval and recycling through early endosomes to the Golgi is also impaired in sec27-1 and sec28Δ cells, resulting in their retargeting to the bud plasma membrane.

gcs1Δ Cells Are Calcofluor-sensitive

Mutations in both the GCS1 and AGE2 Arf-GAP genes lead to defects in post-Golgi transport. In particular, combined gcs1 and age2 mutations led to the impaired delivery of internalized Ste3 mating-factor receptor and the vital dye, FM4-64, to the vacuole (Poon et al., 2001). This result suggests that Gcs1 might facilitate protein retrieval from endosomes to other organelles.

Here we show that the deletion of GCS1 strongly affects the recycling of GFP-Snc1 from early endosomes to the Golgi (Figure 5). Yet, cells lacking GCS1 tend to grow normally on synthetic medium at temperatures above 15°C (Figure 2A) and have only minor defects in protein trafficking to the cell surface (Poon et al., 1996). This suggests that protein export pathways are not markedly affected by loss of the Gcs1 Arf-GAP, as long as the Age2 Arf-GAP is present. To verify that export from the Snc1 recycling compartment is not abolished, we examined whether gcs1Δ cells are resistant to calcofluor, a molecule that binds to chitin and inhibits cell growth. Cells that are resistant to calcofluor have either a mutation in chitin synthase III (Chs3) or are defective in Chs3 export from the chitosome, which is analogous to the early endosome (Valdivia et al., 2002). However, we found that gcs1Δ cells are generally sensitive to calcofluor, unlike control chs6Δ cells (our unpublished observations). This implies that the loss of Gcs1 function does not alter export of Chs3 to the cell surface.

Gcs1 Coimmunoprecipitates with Snx4

Normal cycling of the Snc v-SNAREs requires additional proteins, including the sorting nexins that play a role in retrieval, often as part of multiprotein complexes (Carlton et al., 2005). In particular, the Snx4 sorting nexin is required for the retrieval of GFP-Snc1 from post-Golgi endosomes (Hettema et al., 2003). Because Gcs1 and Snx4 both mediate GFP-Snc1 recycling and physically interact with Snc1 (Hettema et al., 2003 and this study), we determined whether Gcs1 and Snx4 form a complex. We expressed both myc-tagged Snx4 and HA-tagged Gcs1 in wild-type cells and examined whether they can coimmunoprecipitate from cell lysates (Figure 7). We found that Gcs1 coprecipitates with Snx4 in a specific manner. No band corresponding to Gcs1 was detected in precipitates formed in the absence of either myc-Snx4 or HA-Gcs1. In contrast, a control reaction employing the anti-myc antibody to bring down myc-Snc2 demonstrated that it specifically precipitated an HA-tagged protein, in this case the Sso1 t-SNARE (Figure 7). Thus, Gcs1 and Snx4 interact in a specific manner.

DISCUSSION

Coatomer recruitment to Golgi membranes is necessary for formation of the COPI vesicles involved in intra-Golgi and Golgi-ER retrograde transport in yeast and mammals (Kirchhausen, 2000; Spang, 2002; McMahon and Mills, 2004). Despite the established role for coatomer, studies in mammalian cells also describe a post-Golgi role for coatomer and COPI in the transport of proteins to endosomes and multivesicular bodies (Whitney et al., 1995; Aniento et al., 1996; Gu and Gruenberg, 2000; Faure et al., 2004). Coatomer binding in all systems is Arf-dependent and, thus, a mechanism for the recruitment of Arf and Arf-like proteins to different membranes also must depend on specific recruiting factors.

Here we demonstrate that the yeast exo- and endocytic v-SNAREs, Snc1 and Snc2, interact genetically and physically with the Gcs1 Arf-GAP (Figures 1 and 2) and colocalize with Gcs1 to late Golgi and endosomal compartments (Figure 3, A and B). This Arf-GAP has been previously shown to facilitate ER-Golgi and post-Golgi transport (Poon et al.,1999, 2001). Thus, we hypothesize that the Snc v-SNAREs are actively involved in recruiting Gcs1 and, subsequently, Arf1 to these membranes. This idea is supported by in vitro binding data demonstrating the recruitment of coatomer to the Arf-GAP-v-SNARE complex (Figure 4C). One functional consequence of this interaction is retrieval of the Snc v-SNAREs and, perhaps, other cargo proteins to the trans-Golgi. Indeed, Snc1 does not reach the trans-Golgi, as visualized by Sec7-RFP or GFP-Yif1, in cells lacking GCS1 (Figure 5, B and C). This role for Gcs1 is similar to that described for Rcy1 and Snx4, which also interact with Snc1 and mediate its retrieval to the trans-Golgi (Galan et al., 2001; Hettema et al., 2003; Chen et al., 2005). Consistent with a role for Gcs1 in Snc1,2 recycling, we demonstrate that Gcs1 binds to the Snx4 sorting nexin (Figure 7). This suggests that the v-SNARE recruits a complex involving a sorting nexin, an Arf-GAP, Arf, and a coat for retrieval to the Golgi from endosomal compartments (see model, Figure 8). Thus, in the absence of any of these factors (i.e., Gcs1, Snx4, COPI B, etc.), Snc v-SNARE retrieval to the Golgi is altered.

Other components are also involved in GFP-Snc1 recycling to the Golgi (Lafourcade et al., 2004), including the Ypt31,32 GTPases that facilitate Golgi export (Jedd et al., 1997) and act upstream of Rcy1 (Chen et al., 2005). In earlier work, a correlation between GFP-Snc1 phosphorylation and its presence at the cell surface was demonstrated (Galan et al., 2001). Recently Chen et al. (2005) suggested that Ypt31,32 regulate the phosphorylation state of the Snc v-SNAREs, implying that phosphorylation targets these v-SNAREs for recycling (Chen et al., 2005). This method of targeting would seem to be an important mechanism for controlling SNARE recycling and a potential means for facilitating interactions with either Gcs1 or Snx4, for example. However, we have been unable to demonstrate the phosphorylation of either endogenous or epitope-tagged Snc proteins expressed in yeast, either by in vivo labeling or by mobility shift analysis employing alkaline phosphatase treatment using wild-type, rcy1Δ, or end4-1 cells (Couve et al., 1995 and our unpublished observations). In contrast, GFP-Snc1 can clearly be modified into a form whose mobility is altered by phosphatase treatment, as demonstrated by several studies (Galan et al., 2001; Hettema et al., 2003; Chen et al., 2005) as well as by us (our unpublished observations). This finding is consistent with the idea that GFP, and not the v-SNARE, is the phosphorylated substrate observed under these conditions. Thus, the role of phosphorylation as a recycling signal for v-SNAREs warrants further study.

In addition to its well-described role in Golgi-ER transport, the involvement of coatomer has also been demonstrated at the level of protein sorting to endosomes and multivesicular bodies in mammals (Whitney et al., 1995; Aniento et al., 1996; Gu and Gruenberg, 2000; Faure et al., 2004). We show here that certain COPI subunits in yeast may assume this additional role by mediating protein retrieval from endosomes to the Golgi. For example, recombinant Snc v-SNAREs recruit purified coatomer in an in vitro binding assay (Figure 4C). Also, the SNC genes interact genetically with those encoding COPI B subunits, but not a COPI F subunit (Figure 4, E and F). Finally, GFP-Snc1 labeling of the plasma membrane of the growing bud is noticeably heightened in certain COPI B mutants, but not in a COPI F mutant (Figure 6). As a similar result was obtained in the ESCRT I mutant, vps23Δ (Figure 6), which is likely to be defective in its ability to target GFP-Snc1 for vacuolar degradation, it suggests that Snc1 recycles to the plasma membrane under conditions where trafficking to endosomal compartments is affected (see model, Figure 8). Unlike the gcs1Δ mutant, wherein recycling GFP-Snc1 accumulates in early endosomes (Figure 5), the enhanced bud localization observed with sec27-1 and sec28Δ cells suggests an additional role for COPI (but not Gcs1) in sorting to multivesicular bodies (G. Gabriely and J. E. Gerst, our unpublished observations). These results suggest a broad role for COPI in endosome-Golgi transport and warrant further investigation.

Our findings support earlier studies that demonstrate a connection between Arf1 GTPases and v-SNAREs in conferring coat recruitment (Gommel et al., 2001; Rein et al., 2002; Lee et al., 2004). This connection may guarantee incorporation of a v-SNARE into nascent vesicles in order to make them fusion-competent at the appropriate acceptor compartment. The classical view for Arf-GAP function has been that GAP-mediated hydrolysis of GTP on Arf1 leads to vesicle uncoating to allow for subsequent fusion (Tanigawa et al., 1993; Bigay et al., 2003; Reinhard et al., 2003). Yet, other studies support the idea that Arf-GAP activity is required for the packaging of cargo into COPI vesicles (Nickel et al., 1998; Pepperkok et al., 2000; Lanoix et al., 2001; Rein et al., 2002; Lee et al., 2004) and for COPI vesicle biogenesis (Yang et al., 2002; Lee et al., 2004). Our findings neither contradict nor reconcile these differing views, but may suggest additional functions for Gcs1 that are independent of its Arf-GAP activity. This is based on the fact that Snc binding does not involve the Arf-GAP domain of Gcs1 (Figure 1D) nor alters Gcs1 Arf-GAP activity (Figure 4A).

Finally, it has been suggested that Arf-GAPs themselves are coat components (Yang et al., 2002; Lewis et al., 2004). For example, interactions between Arf-GAPs and coat proteins that function in post-Golgi transport, including the GGA and AP3 clathrin adaptor proteins, have been demonstrated (Randazzo and Hirsch, 2004). Arf-GAPs have been shown to interact physically with coat proteins by two-hybrid analysis and coimmunoprecipitation. In particular, Glo3, a yeast Arf-GAP that acts with Gcs1 upon Golgi-ER transport (Poon et al., 1999) binds to coatomer in both in vitro and two-hybrid assays (Eugster et al., 2000; Lewis et al., 2004). Although Gcs1 is not readily detected in COPI vesicles (Lewis et al., 2004), this Arf-GAP may play a role in post-Golgi transport along with members of the COPI coat, as suggested here for the COPI B subcomplex. The possibility exists that multiple subpopulations of COPI occur in yeast, as was recently shown for mammalian cells (Malsam et al., 2005). These different coats may define distinct sorting routes (i.e., Golgi-ER, intra-Golgi, early endosome-Golgi, and late endosome-MVB).

FOOTNOTES

FOOTNOTES

ACKNOWLEDGMENTS

• Abeliovich, H., Grote, E., Novick, P., and Ferro-Novick, S. (1998). Tlg2p, a yeast syntaxin homolog that resides on the Golgi and endocytic structures. J. Biol. Chem. 273, 11719-11727. Crossref, Medline, Google Scholar
• Aniento, F., Gu, F., Parton, R. G., and Gruenberg, J. (1996). An endosomal beta COP is involved in the pH-dependent formation of transport vesicles destined for late endosomes. J. Cell Biol. 133, 29-41. Crossref, Medline, Google Scholar
• Bigay, J., Gounon, P., Robineau, S., and Antonny, B. (2003). Lipid packing sensed by ArfGAP1 couples COPI coat disassembly to membrane bilayer curvature. Nature 426, 563-566. Crossref, Medline, Google Scholar
• Brennwald, P., Kearns, B., Champion, K., Keranen, S., Bankaitis, V., and Novick, P. (1994). Sec9 is a SNAP-25-like component of a yeast SNARE complex that may be the effector of Sec4 function in exocytosis. Cell 79, 245-258. Crossref, Medline, Google Scholar
• Bryant, N. J., and James, D. E. (2003). The Sec1p/Munc18 (SM) protein, Vps45p, cycles on and off membranes during vesicle transport. J. Cell Biol. 161, 691-696. Crossref, Medline, Google Scholar
• Carlton, J., Bujny, M., Rutherford, A., and Cullen, P. (2005). Sorting nexins—unifying trends and new perspectives. Traffic 6, 75-82. Crossref, Medline, Google Scholar
• Chen, S. H., Chen, S., Tokarev, A. A., Liu, F., Jedd, G., and Segev, N. (2005). Ypt31/32 GTPases and their novel F-box effector protein Rcy1 regulate protein recycling. Mol. Biol. Cell 16, 178-192. Link, Google Scholar
• Chen, Y. A., and Scheller, R. H. (2001). SNARE-mediated membrane fusion. Nat. Rev. Mol. Cell Biol. 2, 98-106. Crossref, Medline, Google Scholar
• Couve, A., and Gerst, J. E. (1994). Yeast Snc proteins complex with Sec9. Functional interactions between putative SNARE proteins. J. Biol. Chem. 269, 23391-23394. Crossref, Medline, Google Scholar
• Couve, A., Protopopov, V., and Gerst, J. E. (1995). Yeast synaptobrevin homologs are modified posttranslationally by the addition of palmitate. Proc. Natl. Acad. Sci. USA 92, 5987-5991. Crossref, Medline, Google Scholar
• Durfee, T., Becherer, K., Chen, P. L., Yeh, S. H., Yang, Y., Kilburn, A. E., Lee, W. H., and Elledge, S. J. (1993). The retinoblastoma protein associates with the protein phosphatase type 1 catalytic subunit. Genes Dev. 7, 555-569. Crossref, Medline, Google Scholar
• Eugster, A., Frigerio, G., Dale, M., and Duden, R. (2000). COP I domains required for coatomer integrity, and novel interactions with ARF and ARF-GAP. EMBO J. 19, 3905-3917. Crossref, Medline, Google Scholar
• Fasshauer, D., Sutton, R. B., Brunger, A. T., and Jahn, R. (1998). Conserved structural features of the synaptic fusion complex: SNARE proteins reclassified as Q- and R-SNAREs. Proc. Natl. Acad. Sci. USA 95, 15781-15786. Crossref, Medline, Google Scholar
• Faure, J., Stalder, R., Borel, C., Sobo, K., Piguet, V., Demaurex, N., Gruenberg, J., and Trono, D. (2004). ARF1 regulates Nef-induced CD4 degradation. Curr. Biol. 14, 1056-1064. Crossref, Medline, Google Scholar
• Franzusoff, A., Redding, K., Crosby, J., Fuller, R. S., and Schekman, R. (1991). Localization of components involved in protein transport and processing through the yeast Golgi apparatus. J. Cell Biol. 112, 27-37. Crossref, Medline, Google Scholar
• Galan, J. M., Wiederkehr, A., Seol, J. H., Haguenauer-Tsapis, R., Deshaies, R. J., Riezman, H., and Peter, M. (2001). Skp1p and the F-box protein Rcy1p form a non-SCF complex involved in recycling of the SNARE Snc1p in yeast. Mol. Cell Biol. 21, 3105-3117. Crossref, Medline, Google Scholar
• Gerst, J. E. (1997). Conserved alpha-helical segments on yeast homologs of the synaptobrevin/VAMP family of v-SNAREs mediate exocytic function. J. Biol. Chem. 272, 16591-16598. Crossref, Medline, Google Scholar
• Gerst, J. E., Rodgers, L., Riggs, M., and Wigler, M. (1992). SNC1, a yeast homolog of the synaptic vesicle-associated membrane protein/synaptobrevin gene family: genetic interactions with the RAS and CAP genes. Proc. Natl. Acad. Sci. USA 89, 4338-4342. Crossref, Medline, Google Scholar
• Goldstein, A. L., and McCusker, J. H. (1999). Three new dominant drug resistance cassettes for gene disruption in Saccharomyces cerevisiae. Yeast 15, 1541-1553. Crossref, Medline, Google Scholar
• Gommel, D. U., Memon, A. R., Heiss, A., Lottspeich, F., Pfannstiel, J., Lechner, J., Reinhard, C., Helms, J. B., Nickel, W., and Wieland, F. T. (2001). Recruitment to Golgi membranes of ADP-ribosylation factor 1 is mediated by the cytoplasmic domain of p23. EMBO J. 20, 6751-6760. Crossref, Medline, Google Scholar
• Gu, F., and Gruenberg, J. (2000). ARF1 regulates pH-dependent COP functions in the early endocytic pathway. J. Biol. Chem. 275, 8154-8160. Crossref, Medline, Google Scholar
• Gurunathan, S., Chapman-Shimshoni, D., Trajkovic, S., and Gerst, J. E. (2000). Yeast exocytic v-SNAREs confer endocytosis. Mol. Biol. Cell 11, 3629-3643. Link, Google Scholar
• Hardwick, K. G., and Pelham, H. R. (1992). SED5 encodes a 39-kD integral membrane protein required for vesicular transport between the ER and the Golgi complex. J. Cell Biol. 119, 513-521. Crossref, Medline, Google Scholar
• Haun, R. S., Tsai, S. C., Adamik, R., Moss, J., and Vaughan, M. (1993). Effect of myristoylation on GTP-dependent binding of ADP-ribosylation factor to Golgi. J. Biol. Chem. 268, 7064-7068. Crossref, Medline, Google Scholar
• Hettema, E. H., Lewis, M. J., Black, M. W., and Pelham, H. R. (2003). Retromer and the sorting nexins Snx4/41/42 mediate distinct retrieval pathways from yeast endosomes. EMBO J. 22, 548-557. Crossref, Medline, Google Scholar
• Hoffman, G. R., Rahl, P. B., Collins, R. N., and Cerione, R. A. (2003). Conserved structural motifs in intracellular trafficking pathways: structure of the gammaCOP appendage domain. Mol. Cell 12, 615-625. Crossref, Medline, Google Scholar
• Holthuis, J. C., Nichols, B. J., Dhruvakumar, S., and Pelham, H. R. (1998). Two syntaxin homologues in the TGN/endosomal system of yeast. EMBO J. 17, 113-126. Crossref, Medline, Google Scholar
• Hosobuchi, M., Kreis, T., and Schekman, R. (1992). SEC21 is a gene required for ER to Golgi protein transport that encodes a subunit of a yeast coatomer. Nature 360, 603-605. Crossref, Medline, Google Scholar
• Ireland, L. S., Johnston, G. C., Drebot, M. A., Dhillon, N., DeMaggio, A. J., Hoekstra, M. F., and Singer, R. A. (1994). A member of a novel family of yeast `zn-finger' proteins mediates the transition from stationary phase to cell proliferation. EMBO J. 13, 3812-3821. Crossref, Medline, Google Scholar
• Ito, T., Chiba, T., Ozawa, R., Yoshida, M., Hattori, M., and Sakaki, Y. (2001). A comprehensive two-hybrid analysis to explore the yeast protein interactome. Proc. Natl. Acad. Sci. USA 98, 4569-4574. Crossref, Medline, Google Scholar
• Jedd, G., Mulholland, J., and Segev, N. (1997). Two new Ypt GTPases are required for exit from the yeast trans-Golgi compartment. J. Cell Biol. 137, 563-580. Crossref, Medline, Google Scholar
• Kim, J., Huang, W. P., and Klionsky, D. J. (2001). Membrane recruitment of Aut7p in the autophagy and cytoplasm to vacuole targeting pathways requires Aut1p, Aut2p, and the autophagy conjugation complex. J. Cell Biol. 152, 51-64. Crossref, Medline, Google Scholar
• Kirchhausen, T. (2000). Three ways to make a vesicle. Nat. Rev. Mol. Cell Biol. 1, 187-198. Crossref, Medline, Google Scholar
• Kreis, T. E., Lowe, M., and Pepperkok, R. (1995). COPs regulating membrane traffic. Annu. Rev. Cell Dev. Biol. 11, 677-706. Crossref, Medline, Google Scholar
• Lafourcade, C., Galan, J. M., Gloor, Y., Haguenauer-Tsapis, R., and Peter, M. (2004). The GTPase-activating enzyme Gyp1p is required for recycling of internalized membrane material by inactivation of the Rab/Ypt GTPase Ypt1p. Mol. Cell Biol. 24, 3815-3826. Crossref, Medline, Google Scholar
• Lanoix, J., Ouwendijk, J., Stark, A., Szafer, E., Cassel, D., Dejgaard, K., Weiss, M., and Nilsson, T. (2001). Sorting of Golgi resident proteins into different subpopulations of COPI vesicles: a role for ArfGAP1. J. Cell Biol. 155, 1199-1212. Crossref, Medline, Google Scholar
• Lee, M. C., Miller, E. A., Goldberg, J., Orci, L., and Schekman, R. (2004). Bi-directional protein transport between the ER and Golgi. Annu. Rev. Cell Dev. Biol. 20, 87-123. Crossref, Medline, Google Scholar
• Lewis, M. J., Nichols, B. J., Prescianotto-Baschong, C., Riezman, H., and Pelham, H. R. (2000). Specific retrieval of the exocytic SNARE Snc1p from early yeast endosomes. Mol. Biol. Cell 11, 23-38. Link, Google Scholar
• Lewis, S. M., Poon, P. P., Singer, R. A., Johnston, G. C., and Spang, A. (2004). The ArfGAP Glo3 is required for the generation of COPI vesicles. Mol. Biol. Cell 15, 4064-4072. Link, Google Scholar
• Lustgarten, V., and Gerst, J. E. (1999). Yeast VSM1 encodes a v-SNARE binding protein that may act as a negative regulator of constitutive exocytosis. Mol. Cell Biol. 19, 4480-4494. Crossref, Medline, Google Scholar
• Malsam, J., Satoh, A., Pelletier, L., and Warren, G. (2005). Golgin tethers define subpopulations of COPI vesicles. Science 307, 1095-1098. Crossref, Medline, Google Scholar
• Matern, H., Yang, X., Andrulis, E., Sternglanz, R., Trepte, H. H., and Gallwitz, D. (2000). A novel Golgi membrane protein is part of a GTPase-binding protein complex involved in vesicle targeting. EMBO J. 19, 4485-4492. Crossref, Medline, Google Scholar
• McMahon, H. T., and Mills, I. G. (2004). COP and clathrin-coated vesicle budding: different pathways, common approaches. Curr. Opin. Cell Biol. 16, 379-391. Crossref, Medline, Google Scholar
• Nickel, W., Malsam, J., Gorgas, K., Ravazzola, M., Jenne, N., Helms, J. B., and Wieland, F. T. (1998). Uptake by COPI-coated vesicles of both anterograde and retrograde cargo is inhibited by GTPgammaS in vitro. J. Cell Sci. 111, 3081-3090. Crossref, Medline, Google Scholar
• Nie, Z., Hirsch, D. S., and Randazzo, P. A. (2003). Arf and its many interactors. Curr. Opin. Cell Biol. 15, 396-404. Crossref, Medline, Google Scholar
• Pepperkok, R., Whitney, J. A., Gomez, M., and Kreis, T. E. (2000). COPI vesicles accumulating in the presence of a GTP restricted arf1 mutant are depleted of anterograde and retrograde cargo. J. Cell Sci. 113, 135-144. Crossref, Medline, Google Scholar
• Poon, P. P., Cassel, D., Spang, A., Rotman, M., Pick, E., Singer, R. A., and Johnston, G. C. (1999). Retrograde transport from the yeast Golgi is mediated by two ARF GAP proteins with overlapping function. EMBO J. 18, 555-564. Crossref, Medline, Google Scholar
• Poon, P. P., Nothwehr, S. F., Singer, R. A., and Johnston, G. C. (2001). The Gcs1 and Age2 ArfGAP proteins provide overlapping essential function for transport from the yeast trans-Golgi network. J. Cell Biol. 155, 1239-1250. Crossref, Medline, Google Scholar
• Poon, P. P., Wang, X., Rotman, M., Huber, I., Cukierman, E., Cassel, D., Singer, R. A., and Johnston, G. C. (1996). Saccharomyces cerevisiae Gcs1 is an ADP-ribosylation factor GTPase-activating protein. Proc. Natl. Acad. Sci. USA 93, 10074-10077. Crossref, Medline, Google Scholar
• Protopopov, V., Govindan, B., Novick, P., and Gerst, J. E. (1993). Homologs of the synaptobrevin/VAMP family of synaptic vesicle proteins function on the late secretory pathway in S. cerevisiae. Cell 74, 855-861. Crossref, Medline, Google Scholar
• Randazzo, P. A., and Hirsch, D. S. (2004). Arf GAPs: multifunctional proteins that regulate membrane traffic and actin remodelling. Cell Signal. 16, 401-413. Crossref, Medline, Google Scholar
• Rein, U., Andag, U., Duden, R., Schmitt, H. D., and Spang, A. (2002). ARF-GAP-mediated interaction between the ER-Golgi v-SNAREs and the COPI coat. J. Cell Biol. 157, 395-404. Crossref, Medline, Google Scholar
• Reinhard, C., Harter, C., Bremser, M., Brugger, B., Sohn, K., Helms, J. B., and Wieland, F. (1999). Receptor-induced polymerization of coatomer. Proc. Natl. Acad. Sci. USA 96, 1224-1228. Crossref, Medline, Google Scholar
• Reinhard, C., Schweikert, M., Wieland, F. T., and Nickel, W. (2003). Functional reconstitution of COPI coat assembly and disassembly using chemically defined components. Proc. Natl. Acad. Sci. USA 100, 8253-8257. Crossref, Medline, Google Scholar
• Rose, M. D., Winston, F., and Hieter, P. (1990). Methods in Yeast Genetics, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Google Scholar
• Seron, K., et al. (1998). A yeast t-SNARE involved in endocytosis. Mol. Biol. Cell 9, 2873-2889. Link, Google Scholar
• Sharrocks, A. D. (1994). A T7 expression vector for producing N- and C-terminal fusion proteins with glutathione S-transferase. Gene 138, 105-108. Crossref, Medline, Google Scholar
• Siniossoglou, S., Peak-Chew, S. Y., and Pelham, H. R. (2000). Ric1p and Rgp1p form a complex that catalyses nucleotide exchange on Ypt6p. EMBO J. 19, 4885-4894. Crossref, Medline, Google Scholar
• Spang, A. (2002). ARF1 regulatory factors and COPI vesicle formation. Curr. Opin. Cell Biol. 14, 423-427. Crossref, Medline, Google Scholar
• Spang, A., Matsuoka, K., Hamamoto, S., Schekman, R., and Orci, L. (1998). Coatomer, Arf1p, and nucleotide are required to bud coat protein complex I-coated vesicles from large synthetic liposomes. Proc. Natl. Acad. Sci. USA 95, 11199-11204. Crossref, Medline, Google Scholar
• Springer, S., Spang, A., and Schekman, R. (1999). A primer on vesicle budding. Cell 97, 145-148. Crossref, Medline, Google Scholar
• Tanigawa, G., Orci, L., Amherdt, M., Ravazzola, M., Helms, J. B., and Rothman, J. E. (1993). Hydrolysis of bound GTP by ARF protein triggers uncoating of Golgi-derived COP-coated vesicles. J. Cell Biol. 123, 1365-1371. Crossref, Medline, Google Scholar
• Tong, A. H., et al. (2001). Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science 294, 2364-2368. Crossref, Medline, Google Scholar
• Valdez-Taubas, J., and Pelham, H. R. (2003). Slow diffusion of proteins in the yeast plasma membrane allows polarity to be maintained by endocytic cycling. Curr. Biol. 13, 1636-1640. Crossref, Medline, Google Scholar
• Valdivia, R. H., Baggott, D., Chuang, J. S., and Schekman, R. W. (2002). The yeast clathrin adaptor protein complex 1 is required for the efficient retention of a subset of late Golgi membrane proteins. Dev. Cell 2, 283-294. Crossref, Medline, Google Scholar
• Whitney, J. A., Gomez, M., Sheff, D., Kreis, T. E., and Mellman, I. (1995). Cytoplasmic coat proteins involved in endosome function. Cell 83, 703-713. Crossref, Medline, Google Scholar
• Wiederkehr, A., Avaro, S., Prescianotto-Baschong, C., Haguenauer-Tsapis, R., and Riezman, H. (2000). The F-box protein Rcy1p is involved in endocytic membrane traffic and recycling out of an early endosome in Saccharomyces cerevisiae. J. Cell Biol. 149, 397-410. Crossref, Medline, Google Scholar
• Wuestehube, L. J., Duden, R., Eun, A., Hamamoto, S., Korn, P., Ram, R., and Schekman, R. (1996). New mutants of Saccharomyces cerevisiae affected in the transport of proteins from the endoplasmic reticulum to the Golgi complex. Genetics 142, 393-406. Crossref, Medline, Google Scholar
• Yang, J. S., Lee, S. Y., Gao, M., Bourgoin, S., Randazzo, P. A., Premont, R. T., and Hsu, V. W. (2002). ARFGAP1 promotes the formation of COPI vesicles, suggesting function as a component of the coat. J. Cell Biol. 159, 69-78. Crossref, Medline, Google Scholar