QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 16, Issue 9, 3951-3962, September 2005

This Article Abstract Full Text (PDF) All Versions of this Article: E05-02-0114v1 16/9/3951    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Li, Y. Articles by Peng, R. Search for Related Content PubMed PubMed Citation Articles by Li, Y. Articles by Peng, R.

Structure-based Functional Analysis Reveals a Role for the SM Protein Sly1p in Retrograde Transport to the Endoplasmic Reticulum

Max Planck Institute for Biophysical Chemistry, Department of Molecular Genetics, 37077 Göttingen, Germany

Submitted February 10, 2005; Revised May 26, 2005; Accepted June 8, 2005
Monitoring Editor: Akihiko Nakano

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Sec1p/Munc18 (SM) proteins are essential for membrane fusion events in eukaryotic cells. Here we describe a systematic, structure-based mutational analysis of the yeast SM protein Sly1p, which was previously shown to function in anterograde endoplasmic reticulum (ER)-to-Golgi and intra-Golgi protein transport. Five new temperature-sensitive (ts) mutants, each carrying a single amino acid substitution in Sly1p, were identified. Unexpectedly, not all of the ts mutants exhibited striking anterograde ER-to-Golgi transport defects. For example, in cells of the novel sly1-5 mutant, transport of newly synthesized lysosomal and secreted proteins was still efficient, but the ER-resident Kar2p/BiP was missorted to the outside of the cell, and two proteins, Sed5p and Rer1p, which normally shuttle between the Golgi and the ER, failed to relocate to the ER. We also discovered that in vivo, Sly1p was associated with a SNARE complex formed on the ER, and that in vitro, the SM protein directly interacted with the ER-localized nonsyntaxin SNAREs Use1p/Slt1p and Sec20p. Furthermore, several conditional mutants defective in Golgi-to-ER transport were synthetically lethal with sly1-5. Together, these results indicate a previously unrecognized function of Sly1p in retrograde transport to the endoplasmic reticulum.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Membrane fusion in eukaryotic cells is of fundamental importance for exo- and endocytosis and for maintaining the integrity of intracellular organelles. Essential components of the machinery that mediates lipid bilayer mixing or fusion are members of three protein families that are conserved from unicellular yeast to man: the soluble N-ethylmaleimide-sensitive fusion (NSF)-attachment protein receptors (SNAREs), the Sec1/Munc18 (SM) proteins and the Ypt/Rab GTPases. So far, evidence favors the view that SNAREs catalyze membrane fusion (Sutton et al., 1998; Weber et al., 1998; Schuette et al., 2004), whereas Ypt/Rab GTPases and SM proteins regulate upstream processes, termed "tethering," to program the membranes to be fused (Jahn et al., 2003; Pfeffer, 2001). Being first identified in screens for membrane-trafficking mutants in Caenorhabditis elegans (Brenner, 1974), all the SM family members are soluble, peripherally membrane-associated proteins with molecular masses of 60–90 kDa. There are four SM proteins in the budding yeast and seven in mammalian cells. Although the primary sequence homology within SM proteins is rather low, their three-dimensional structures are highly conserved, suggesting that SM proteins fulfill a general role in membrane fusion (Halachmi and Lev, 1996; Rizo and Sudhof, 2002; Gallwitz and Jahn, 2003).

The first evidence regarding SM protein function came from the finding that the mammalian N-Sec1/Munc18 protein binds with high affinity to the SNARE syntaxin1 exclusively in its closed conformation, therefore precluding syntaxin1 from SNARE complex assembly (Hata et al., 1993; Pevsner et al., 1994; Garcia et al., 1994; Dulubova et al., 1999; Yang et al., 2000). In agreement with this, overexpression of N-Sec1/Munc18 in Drosophila neurons inhibits exocytosis, but this effect is rescued by concomitant overexpression of syntaxin 1 (Wu et al., 1998). These findings have led to a model according to which SM proteins inhibit SNARE complex formation. However, not all the function of SM proteins can be explained by this model. First, deletions of SM genes in yeast, invertebrates, and vertebrates invariably lead to a block of intracellular fusion steps (Novick and Schekman, 1979; Ossig et al., 1991; Verhage et al., 2000), whereas overexpression of SM proteins does not inhibit membrane trafficking as the inhibitory model would predict. For instance, in chromaffin cells, overexpression of Munc18 increases rather than decreases exocytosis (Voets et al., 2001); and not all syntaxins are capable of adopting closed conformations (Dulubova et al., 2001, 2002). Importantly, the binding of SM proteins to their respective syntaxins does not necessarily preclude formation of fusogenic SNARE complexes. For example, yeast Sly1p, which in solution binds to the Golgi syntaxin Sed5p with high affinity (Grabowski and Gallwitz, 1997), also associates with a fully assembled SNARE complex and promotes the assembly of physiologically relevant SNARE complexes (Peng and Gallwitz, 2002), and yeast Sec1p binds to the exocytotic SNARE complex and promotes membrane fusion in vivo and in vitro (Carr et al., 1999; Scott et al., 2004). Similarly, Vps45p binds to a very short N-terminal peptide of the syntaxin Tlg2p, and from cell lysates it coprecipitates with syntaxin Tlg2p and its SNARE partners. This indicates that Vps45p, like Sly1p and Sec1p, not only binds to syntaxin but also to syntaxin-containing SNARE complexes (Nichols et al., 1998; Coe et al., 1999; Dulubova et al., 2002). Although the mechanism of SM/syntaxin binding is diverse, the structures of members of both protein families are conserved. This suggests that all the SM proteins might share a common role but the bimolecular binding of SM/syntaxin per se is not relevant or important for their function. In support of this, we have recently shown that in yeast, mutant Sly1p and Sed5p incapable to bind to each other were fully functional in vivo (Peng and Gallwitz, 2004). A similar observation has also been reported for N-Sec1/Munc18 in mammalian cells (Ciufo et al., 2005).

To look more closely into the structure-function relationship of SM proteins, we took advantage of the recently solved crystal structure of the yeast Sly1p and performed a thorough mutational analysis. In this study, we report the identification of five novel sly1 temperature-sensitive mutants. Extensive phenotypic analysis of these ts mutants leads to the conclusion that Sly1p plays a vital and previously unrecognized role in retrograde transport to the endoplasmic reticulum (ER).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Yeast Strains, Genetic Methods, and Plasmids
Yeast strains used in this study are listed in Table 1. Yeast transformation, mating, sporulation, and tetrad analysis were performed by using standard techniques (Sherman and Hicks, 1991; Gietz et al., 1992).

Yeast strains expressing GFP-Sed5p was constructed as follows: The coding region of SED5 and the 5' and 3' adjacent sequences were amplified by the PCR from genomic DNA of Saccharomyces cerevisiae, subcloned, and sequenced to prevent possible errors introduced by PCR. A BamHI site was created after the start codon of translation (ATG) by Quick Change Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA), and an enhanced version of GFP (Stratagene) amplified as a BamHI fragment was then inserted into the BamHI of SED5, which was on an integrating vector pRS306 (CEN, URA3). The resulting pRS306-GFP-SED5 was used for yeast transformation after linearizing by digestion with Bsu36I. Transformants on SC-Ura dropouts were further grown on media containing 5'-FOA, which selected against URA3. Yeast strains losing wild-type SED5 but containing GFP-SED5 as the sole source for Sed5 protein was finally chosen for further study. The plasmid expressing GFP-Rer1p was kindly provided by A. Nakano (Tokyo, Japan).

pRS316-SLY1-LEU2 was constructed by subcloning the 2.2-kb SalI-XhoI fragment from YEp13 (New England Biolabs, Beverly, MA) containing the LEU2 gene into the SalI site created right behind the stop codon of SLY1. The DNA fragments encoding the entire cytoplasmic regions of Use1p (1–218), Sec20p (1–275) and Ufe1p (1–329) were amplified from genomic DNA prepared from MSUC3D by standard PCR. The PCR products were then subcloned into pGEX-KG or pGEX-KT vectors (Hakes and Dixon, 1992) for standard bacterial expression of GST-fusion proteins. Constructs expressing His₁₀-tagged Sly1p (full length), GST-Vti1p, and GST-Sec22p without their transmembrane domain were made as previously described (Peng and Gallwitz, 2002).

The point mutations in the SLY1 gene were generated with pRS316-SLY1-LEU2 by site-directed mutagenesis using the QuickChange Site-Directed Mutagenesis Kit (Stratagene). The plasmids carrying sly1 mutant genes, verified by DNA sequence analysis, were transformed into a diploid strain (RPY135) in which one copy of SLY1 was replaced by KanMX4. After selected on kanamycin-containing plates, transformants were sporulated and the tetrad progeny was dissected. The Leu⁺/Kan⁺ haploids with the genomic SLY1 disruption by KanMX4 but containing different mutant sly1 genes on plasmids were selected, and growth was followed at temperatures of 14, 25, and 37°C. The sly1 alleles that conferred temperature sensitivity were integrated into the genome for further study. Briefly, an XbaI I-XhoI fragment bearing a mutant sly1 allele, and the adjacent LEU2 gene was introduced into RPY135, which was resistant to G418. The resulting Leu⁺ transformants, in which the sly1::KanMX4 was replaced by a certain sly1::LEU2 mutant allele and became sensitive to G418, were identified. After sporulation and tetrad analysis, the Leu⁺ haploids bearing the mutant sly1 allele were finally selected, and their phenotypes were further characterized.

Cell Labeling and Invertase Assay
Cell labeling and immunoprecipitations were performed as described previously (Tsukada and Gallwitz; 1996). Newly synthesized CPY or alkaline phosphatase (ALP), precipitated with affinity-purified antibodies, was resolved by SDS-PAGE, followed by autoradiography. Invertase activity staining was carried out as described (Peng et al., 2000).

Protein Expression, Purification, and GST Pulldown Assay
Expression in Escherichia coli and purification of GST- and His-tagged proteins as well as GST pulldown experiments were performed as described (Peng and Gallwitz, 2002). GST pulldown experiments involved the following modifications: 10 µl of 50% glutathione Sepharose slurry (Pharmacia, Piscataway, NJ) were mixed with excessive amounts of GST fusion proteins at 4°C for 60 min to saturate the available binding sites on the beads. The beads were then washed three times with buffer F (50 mM Tris-HCl, pH 7.5, 150 mM KCl, 10% glycerol, 0.5% Triton X-100, 2 mM -mercaptoethanol) and mixed with purified Sly1p-His₁₀ in a total volume of 100 µl in buffer F adjusted to a final KCl concentration of 375 mM. Incubation proceeded for 60 min at 4°C before washing six times with buffer F (500 µl each per washing). Proteins bound on beads were eluted with SDS sample buffer, analyzed by SDS-PAGE, and viewed by Coomassie blue staining or Western blotting.

Antibodies, Immunoprecipitation, and Western Blot Analysis
Antibodies against Sly1p, Sec22p, and Bos1p have been described (Peng and Gallwitz, 2002). Antisera against Kar2p/BiP were raised in rabbits. The c-Myc antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA) Polyclonal antiserum against Use1p and Ufe1p was kindly provided by G. Fischer von Mollard (University Goettingen) and S. Jentsch (MPI for Biochemistry, Munich), respectively. Immunoprecipitation experiments and Western blotting analysis were performed as previously described (Peng and Gallwitz, 2002).

The quantification of the relative amounts of Sly1p on immunoblots was performed using a LumiImager (Boehringer, Mannheim, Germany).

Kar2p/BiP Secretion Assay
Cells were grown in YEPD at 25°C to midlog phase before harvested. Around 0.1 OD of cells were spotted onto YEPD plates and incubated at 25°C for 8–10 h. Nitrocellulose filters were directly placed onto the plates, and the plates were incubated for additional 2 h at 25°C or 37°C. The filters were then carefully peeled off, washed three times in Western blot buffer A (10 mM Tris-HCl, pH 7.4, 0.9% NaCl), and treated subsequently with antiserum against Kar2p.

Fluorescence Microscopy
Cells expressing GFP-fused proteins were grown in YPG or SC medium under the appropriate conditions. The fluorescence was viewed in living or fixed cells as indicated in the figure legends. For viewing living cells, the midlog phase cells were collected by centrifugation and washed once with sterile, distilled H₂O, and three times with ice-cold phosphate-buffered saline (PBS) buffer. The washed cells were then resuspended in appropriate volume of ice-cold PBS buffer. Cell suspension (1.5 µl) was dropped onto a 76 x 26-mm microscope slide, covered with a coverslip, and subjected to immediate viewing. For analyzing cells at specific temperatures, midlog phase cells grown at either 25°C or nonpermissive temperature (35 or 37°C) were collected by centrifugation, washed once in ice-cold PBS buffer, and fixed at –20°C in absolute methanol for 15 min. Cells were then washed once with acetone (precooled to –20°C) and three times with ice-cold PBS buffer. Further treatment was essentially as described (Carr et al., 1999). Fluorescence microscopy was performed with an Axioplan microscope equipped with a 100 x oil-immersion objective (Carl Zeiss, Thornwood, NY) and with a FITC filter (excitation 485 nm, emission 535 nm). Images were recorded with a photometric CCD camera and processed with the help of Photoshop software (Adobe, San Jose, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Directed Mutagenesis of SLY1 Yields Five New Temperature-sensitive Mutants
We undertook a systematic, genetic approach to explore the structure-function relationship of the yeast SM protein Sly1p. The available structures of the yeast Sly1p (Bracher and Weissenhorn, 2002) and the neuronal homologue N-Sec1/Munc18 (Misura et al., 2000) allowed for targeted mutagenesis by altering charged and highly conserved amino acids in order to identify residues or structural elements of physiological importance. Of the 666 amino acids of the whole protein, around seventy were selected for substitutions. As outlined in detail in Materials and Methods, sly1 mutant genes on a single-copy vector that were able to support the growth of haploid cells lacking the genomic SLY1 gene were identified. Growth properties of the different viable mutant strains (n = 62) were investigated at temperatures between 14 and 37°C. None of the sly1 mutants generated caused cold sensitivity, but five new temperature-sensitive alleles with a single amino acid substitution each (sly1-2, D59A; sly1-3, I153K; sly1-4, H276A; sly1-5, R452A; sly1-6, E618A) were found (Figure 1). These were used to replace the chromosomal SLY1 gene of a haploid wild-type strain and subjected to further study.

View larger version (51K): [in this window] [in a new window]   Figure 1. The growth phenotype of temperature-sensitive (ts) sly1 mutants and the stability of mutant Sly1 proteins. (A) Location of single amino acid substitutions in six sly1 ts mutants. I, II, III, domain I (a.a.1–161), domain II (a.a.162–268 and a.a.582–666) and domain III (a.a.269–581) of Sly1p. (B) Three µl of 10-fold diluted log-phase cultures, starting with 106 cells/ml, were spotted onto YPD plates and incubated for 3 d at the indicated temperatures. (C) Cells of wild-type and mutant sly1 strains grown at the permissive temperature (25°C) to midlog phase were shifted to and incubated at 37°C for 1, 2, 3, and 4 h. Total proteins were prepared from these cells by lysis in NaOH/-mercaptoethanol and precipitated by trichloroacetic acid. After separation of proteins by SDS-PAGE, Sly1p was detected by anti-Sly1p antibodies. The Sly1 protein amounts from immunoblots were calculated by LumiImager analysis, and the loading materials were evaluated by Bos1p signals on the same blot.

The five newly identified ts mutants (sly1-2, sly1-3, sly1-4, sly1-5, sly1-6), together with the previously identified sly1-1 allele (Mizuta and Warner, 1994; Cao et al., 1998), exhibited growth arrest at different temperatures when grown on rich media (Figure 1B). Cells of sly1-4 and sly1-6 exhibited the most severe growth defect and almost failed to proliferate already at 30°C. Cells of sly1-1 and sly1-2 grew well at 30°C, but extremely slow at 35°C. Although the growth of all ts mutants ceased at 37°C, the mutant proteins in sly1-3, sly1-4, and sly1-5 cells were stable for at least 2 h at the nonpermissive temperature (Figure 1C). However, Sly1 mutant proteins decreased with time in sly1-1, sly1-2, and sly1-6 cells, but 4 h after cell growth at 37°C 25% (sly1-1), 15% (sly1-2), and 32% (sly1-6) of the protein level seen at 25°C was still detectable (Figure 1C).

Protein Traffic Defects in Newly Identified sly1 Temperature-sensitive Mutants
Sly1p acts early in the secretory pathway, downstream of the GTPase Ypt1p and the vesicle tethering factor Uso1p (Dascher et al., 1991; Ossig et al., 1991; Sapperstein et al., 1996; Cao and Barlowe, 2000) and has been shown to promote the assembly of ER/Golgi SNARE complexes (Kosodo et al., 2002; Peng and Gallwitz, 2002; Peng and Gallwitz, 2004). The newly identified temperature-sensitive sly1 mutants were first tested for their defects in protein transport by following the fate of newly synthesized lysosomal enzymes carboxypeptidase Y (CPY) and ALP, and of the secreted invertase. The three enzymes are N-linked glycoproteins that undergo core glycosylation in the ER and outer chain elongation in the Golgi before being transported to their final destination. Different glycosylated forms, and mature lysosomal enzymes generated by proteolytic cleavage, can be easily identified by gel electrophoretic analysis. Surprisingly, at the nonpermissive temperature of 37°C, not all of the sly1 mutants exhibited striking anterograde ER-to-Golgi transport defects as expected. In sly1-1 and sly1-2, and especially in sly1-5 mutant cells, only a slight delay of CPY and ALP maturation was observed (Figure 2A). However, most of the newly synthesized enzymes were properly and efficiently processed in sly1-5 within 30 min at 37°C, indicating that they reached their final destination. In sharp contrast, ER-to-Golgi protein transport in sly1-3, sly1-4, and sly1-6 mutant cells was completely blocked as in sec23-1 mutant cells that are defective in vesicle formation at the ER (Figure 2A). Similar results were obtained with invertase, whose intracellular and periplasmic portions were assessed in non-denaturating acrylamide gels. Whereas sly1-3, sly1-4, and sly1-6 mutant cells were almost completely defective in invertase secretion and accumulated the ER core-glycosylated enzyme intracellularly at 37°C, sly1-1, and especially sly1-2 and sly1-5 mutant cells secreted a significant portion of the newly synthesized invertase, although in underglycosylated form (Figure 2B).

View larger version (50K): [in this window] [in a new window]   Figure 2. Analysis of CPY transport and invertase secretion in temperature-sensitive sly1 mutants. (A) Wild-type (MUSC3D) and various mutant cells were grown at 25°C, shifted to 37°C for 30 min, and then labeled with Tran-35S-label at 37°C for 7 min and chased for 0, 10, and 30 min. CPY was immunoprecipitated from cell lysates, analyzed by SDS-PAGE, and viewed by autoradiography. ER core-glycosylated (p1), Golgi-modified (p2), and the mature form (m) of CPY are indicated. (B) Cells of wild-type and mutant strains grown at 25°C were transferred to 0.1% glucose medium to induce invertase synthesis for 60 min at either 25 or 37°C. Spheroplasts were prepared and pelleted to obtain the periplasmic (E) and intracellular (I) fractions. Proteins of both fractions were separated on a nondenaturing polyacrylamide gel and subjected to activity staining of invertase. S, secreted invertase; ER, ER core-glycosylated invertase; C, cytoplasmic form of invertase.

According to the transport defects observed, the sly1 mutants could be teamed into two subgroups: one group (sly1-3, sly1-4, and sly1-6) exhibited striking inhibition of ER-to-Golgi transport, whereas the other, particularly sly1-5, showed only mild anterograde transport defects, resembling mutants whose gene products are directly involved in Golgi-to-ER transport. This suggested that Sly1p might also have a function in retrograde protein transport.

Golgi-to-ER Retrograde Traffic Is Defective in Some sly1 Mutants
Anterograde and retrograde traffic between the ER and the Golgi is tightly coupled. A block in the retrograde transport has an indirect but rapid effect on anterograde transport (Lewis and Pelham, 1996). Several assays have been developed to distinguish these transport steps. Kar2p/BiP is a soluble ER protein, which is being retrieved from the Golgi when escaped from the ER (Semenza et al., 1990). However, Kar2p is secreted into the medium when retrograde traffic is blocked. We therefore searched for Kar2p secretion of the newly identified sly1 mutants using a qualitative immunological detection method with cells grown on filters both at permissive and nonpermissive temperatures. As shown in Figure 3A, Kar2p was missorted in sly1-5 cells at 25°C, but much more pronounced at 37°C. Kar2p was also secreted from sly1-1 and sly1-2 cells, especially at 37°C, but less efficiently than from sly1-5 mutant cells. Low amounts of the ER-resident protein appeared to also reach the outside of the sly1-3 and sly1-6 mutant cells at both temperatures, whereas sly1-4 behaved like wild-type cells. Weak Kar2p signals on filters might have been caused by cell breakage, although the test for cytoplasmic hexokinase in the same filter assay did not indicate that this was a serious problem.

View larger version (61K): [in this window] [in a new window]   Figure 3. Golgi-to-ER trafficking is defective in some sly1 mutant cells. (A) Cells of wild-type, sly1 and sec20-1 (RH236–3A) mutant strains were grown at 25°C to midlog phase before they were spotted onto YPD plates and incubated at 25°C for 8 h. The plates were then overlaid with nitrocellulose filters, and cells were allowed to grow at 25 or 37°C for additional 2 h. The Kar2p secreted and bound to the nitrocellulose filter was detected by immunoblotting with anti-Kar2p antibodies. (B) Cells of sec23-1 and double sly1-5/sec23-1 mutant cells stably expressing GFP-Sed5p (strain RPY174 and YLY25, respectively) were grown at 25°C and then shifted to and incubated at 37°C for 1 h. Samples of the cells were fixed before viewing the fluorescence under the microscope. To see whether the wild-type SLY1 gene can rescue the defect of Sed5p cycling between the Golgi and the ER, a single copy vector bearing the wild-type SLY1 was introduced into the sly1-5/sec23-1 double mutant, and the GFP-Sed5p signal was viewed as above. (C) The dynamics of Rer1p localization in sly1 mutants. The SLY1/pep4 (PS42-1), sly1-5/pep4 (YLY23), and sly1-6/pep4 (YLY24) strains expressing GFP-Rer1p from a single copy vector were grown, treated, and observed microscopically as in B. DIC images are shown underneath the fluorescently labeled cells.

The same conclusion could be reached from experiments in which we followed the dynamics of Rer1p localization in sly1-5 mutant cells. Rer1p is a Golgi-localized receptor capturing and relocalizing several ER membrane proteins, and itself cycles through to the ER (Boehm et al., 1997; Sato et al., 2001). A construct expressing GFP-Rer1p from a single-copy vector (Sato et al., 2001) in yeast strains lacking vacuolar Pep4 protease (pep4) was used to follow the localization of the receptor protein. As shown in Figure 3C, when expressed in cells in the presence of wild-type SLY1 (SLY1/pep4), GFP-Rer1p exhibited normal Golgi localization at both permissive (25°C) and nonpermissive temperature (37°C). In sharp contrast, when expressed in sly1-5/pep4 mutant cells, typical Golgi staining was only scarcely seen, but instead the fluorescence was mostly confined to vacuoles. This indicated that GFP-Rer1p was missorted to the vacuole as a result of the blockage of the retrograde Golgi-to-ER pathway. Vacuolar staining was already visible at 25°C, but it was much more pronounced at 37°C. Importantly, vacuolar staining by missorted GFP-Rer1p was observed only in sly1-5/pep4 but not in sly1-6/pep4 mutant cells, which are primarily defective in anterograde ER-to-Golgi transport. However, in the latter mutant, the number and brightness of punctate Golgi structures decreased at nonpermissive temperature, and a diffuse fluorescence of the entire cytoplasm increased. This might have been the result of an accumulation of anterograde COPII vesicles formed at the ER but unable to dock to and fuse with Golgi membranes.

Together, these findings are in accord with the observed protein transport defects (Figure 2) and support the conclusion that in sly1-5 mutant cells, defects in retrograde transport outweigh those in anterograde transport between the ER and the Golgi complex. This clearly indicates that Sly1p has a function in both trafficking steps.

Sly1p Directly Binds to Nonsyntaxin ER-SNAREs and Is Associated with SNARE Complexes on ER Membranes
It has been reported that in vitro, Sly1p binds to the ER-syntaxins Ufe1p from yeast and syntaxin18 from vertebrate cells and that Sly1p coimmunoprecipitates with ER-SNAREs from mammalian cell lysates (Yamaguchi et al., 2002; Hirose et al., 2004). These studies suggested a possible role for Sly1p in retrograde transport although direct evidence remained to be established. Our recent work demonstrated that Sly1p not only binds to the Golgi syntaxin Sed5p but also directly and specifically interacts with nonsyntaxin SNAREs on Golgi membranes (Peng and Gallwitz, 2004).

To address whether Sly1p also binds to nonsyntaxin SNAREs on ER membranes, we screened with purified proteins for possible interactions between Sly1p and the three nonsyntaxin ER-SNARES, Sec20p, Sec22p, and Use1p (Lewis et al., 1997; Burri et al., 2003; Dilcher et al., 2003). For this purpose, the cytosolic regions of the three nonsyntaxin SNAREs and of the syntaxin Ufe1p were fused to the C-terminus of GST. The fusion proteins were immobilized on glutathione Sepharose beads and incubated with N-terminally His₁₀-tagged Sly1p at salt concentration of 375 mM. After extensive washing of the beads, significant binding of Sly1p was observed with the nonsyntaxin SNAREs Sec20p and Use1p and, as expected from a previous study (Yamaguchi et al., 2002), with Ufe1p. As previously seen (Peng and Gallwitz, 2004), Sly1p did not bind to Sec22p. Vti1p, a nonsyntaxin SNARE that acts in the transport from Golgi to endosomes and to the vacuole, did not interact with Sly1p, indicating that the observed interactions were specific (Figure 4A).

View larger version (27K): [in this window] [in a new window]   Figure 4. Sly1p directly binds to nonsyntaxin ER-SNAREs and is associated with a SNARE complex on the ER. (A) Bacterially produced GST fusions of SNAREs involved in Golgi-to-ER retrograde transport were immobilized on glutathione Sepharose beads and incubated with purified Sly1p. After washing, proteins bound to beads were analyzed by SDS-PAGE and Coomassie blue staining (top panel) or by immunoblotting using anti-Sly1p antibodies (bottom panel). (B) Cells of a HA-Ufe1p-expressing strain (YLY17) were grown at 25°C to midlog phase. Detergent lysates prepared from the strain were subjected to immunoprecipitation by using anti-HA antibodies. HA-Ufe1p and its coimmunoprecipitated proteins were analyzed on immunoblots.

Inefficient Coprecipitation of Sly1-5p with Ufe1p and ER-Nonsyntaxin SNAREs from sly1-5 Mutant Cell Lysates
As described above, the most severe defects of retrograde transport were observed in sly1-5 mutant cells. Because Sly1p also binds to ER-nonsyntaxin SNAREs (Figure 4, A and B), we reasoned that this binding might be affected in sly1-5 mutant cells. Therefore, immunoprecipitates with anti-Sly1p antibodies were prepared from detergent lysates of wild-type and sly1-5 mutant cells that had been grown for 60 min at 37°C. Coprecipitation of Sly1 proteins with Ufe1p and with several ER- and Golgi-nonsyntaxin SNAREs was analyzed by immunoblotting. As shown in Figure 5, the Golgi syntaxin Sed5p as well as nonsyntaxin SNAREs specific for anterograde ER-to-Golgi traffic (Bos1p and Bet1p) were coprecipitated with the SM protein with similar efficiency from extracts of wild-type and sly1-5 cells. However, compared with Sly1p from wild-type cell extracts, coprecipitation with Sly1-5p of the ER syntaxin Ufe1p and of nonsyntaxin SNAREs specific for retrograde transport (Use1p and Sec20p) was significantly reduced, indicating a specific defect in the formation of ER SNARE complexes. As can be seen from Figure 5, the significant reduction of ER-SNAREs coprecipitating with Sly1-5p from mutant cell lysates is not the result of lower intracellular levels of these SNAREs. As Ufe1p appears to be a permanent ER resident (Sato and Nakano, 2004), a hypothetical defect of sly1-5 mutant cells in sorting of the ER syntaxin into COPI vesicles would be highly unlikely to account for its highly reduced coprecipitation with Sly1-5p and other ER SNAREs.

View larger version (76K): [in this window] [in a new window]   Figure 5. Inefficient coprecipitation of Sly1-5p with SNARE proteins of the retrograde Golgi-to-ER trafficking. Cells of wild-type (MSUC3D) and of sly1-5 (YLY26) were grown at 25°C before transfer to and incubation at 37°C for 60 min. Detergent lysates were prepared from the cells, and immunoprecipitation was performed with affinity-purified Sly1p antibodies. Proteins of the precipitates were separated by SDS-PAGE, followed by immunoblotting with antibodies against Sly1p, Bet1p, Bos1p, Sed5p, Ufe1p, Use1p, and Myc (Sec20p). Compared with wild-type, 23, 34, and 24% of Ufe1p, Sec20p, and Use1p, respectively, coprecipitated with Sly1 protein from lysates of sly1-5 mutant cells (LumiImager analysis, corrected for different amounts of lysates; see below). Note that coprecipitation with Sly1 protein of Golgi SNAREs was not impaired in the mutant cells. Also note that the material above Sly1p that cross-reacted with anti-Sly1p antibodies (input and supernatant) is absent from immunoprecipitates and that, according to LumiImager analysis, 40% more lysates from sly1-5 as compared with wild-type cells was used for immunoprecipitation in the experiment shown (compare intensity of bands in WT and sly1-5 lanes of input and supernatant).

Genetic Interactions between sly1-5 and Mutants Defective in Golgi-to-ER Retrograde Transport
Genetic interactions provide a means to identify components of a common pathway. A functional relationship of gene products is likely if the combination of two conditional mutations aggravates the growth defect of a single mutant. We applied this method to analyze if components of the retrograde transport pathway have close functional connections with Sly1p. To observe such synthetic effects, we created heterozygous diploids by mating haploid sly1-5 cells with other single mutants defective in Golgi-to-ER retrograde transport: sec20-1 and sec22-5, both expressing mutant nonsyntaxin ER-SNAREs (Lewis et al., 1997; Ballensiefen et al., 1998), tip20-5, expressing defective Sec20p-interacting protein (Cosson et al., 1997), and dsl1-22, encoding a defective COPI-interacting protein (Andag et al., 2001; Reilly et al., 2001). The resulting diploids were sporulated, tetrads were dissected, and the growth properties of the resulting spores were analyzed. As shown in Table 3, spores carrying both sly1-5 and either sec20-1, sec22-5, tip20-5, or dsl1-22 were not viable, demonstrating that the double mutants were synthetically lethal. In contrast, combinations of sly1-5 either with sec23-1 or with sec24-11, genes encoding COPII components, or with bos1-1, encoding a Golgi-SNARE protein defective in anterograde ER-to-Golgi transport, were viable. These results also support the conclusion that Sly1p also plays a vital role in retrograde transport to the ER.

View larger version (38K): [in this window] [in a new window]   Figure 6. Sly1p cycles between the Golgi and the ER. Cells of sec12 stably expressing GFP-Sly1p (strain RPY152) were grown at 25°C and then shifted to and incubated at 35°C for the indicated time periods. Samples of cells were fixed before viewing under the microscope. To see if the temperature effect was reversible, samples treated at 35°C for 60 min were quickly cooled to 25°C and further incubated at 25°C for 60 min. Cells were then fixed and viewed by fluorescence microscopy.

In a further attempt to prove this, we followed the fate of GFP-tagged Sly1p in a sec12 mutant that is unable to form transport vesicles at the ER but still allows retrograde Golgi-to-ER transport at nonpermissive conditions (Schröder et al., 1995). As shown in Figure 6, cells at permissive conditions exhibited the Golgi-typical dot-like fluorescence, but after a shift to 35°C, Sly1p accumulated in the ER, which is easily seen by fluorescence of the nuclear rim and adjacent to the plasma membrane. Full recovery of Golgi localization was observed within minutes after shifting cells back to 25°C. This clearly demonstrates that Sly1p in yeast, like its homologue in mammalian cells (Rowe et al., 1998), cycles between the Golgi apparatus and the ER, and it suggests that the SM protein is likely to be available at any given time to act as a component of the machinery needed to fuse Golgi-derived vesicles with the ER membrane.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
The purpose of our study was to shed new light on the functioning of SM proteins in vesicular trafficking by directed mutagenesis of the Sly1 protein that according to present knowledge acts close to membrane fusion in anterograde transport of vesicles between the ER and Golgi. Selected amino acid substitutions of the Sed5p-binding region of yeast Sly1p have already shown the value of such an approach because it was discovered that 1) the high-affinity binding of the syntaxin Sed5p to Sly1p does not serve an essential function and 2) that Sly1p does also bind nonsyntaxin SNAREs in a specific manner and, apparently, at region(s) of the SM protein different from the Sed5p binding pocket (Peng and Gallwitz, 2004). Furthermore, a single amino acid change within domain III in Sly1-20p was previously shown to make yeast cells independent of the need of the GTPase Ypt1p (Dascher et al., 1991; Ossig et al., 1991) and the tethering protein Uso1p (Sapperstein et al., 1996), strongly suggesting that the essential Sly1 protein interacts functionally with other components of the membrane fusion machinery.

Because of the lack of known interaction partners of Sly1p and of critical regions of the protein for functional integrity, we focused on amino acid substitutions, single or in combination, of conserved residues in potentially important secondary structure elements, and of residues forming conserved salt bridges or hydrogen bonds (Misura et al., 2000; Bracher and Weissenhorn, 2002).

Substitutions of the highly conserved aspartic acid located as single residue between -strand 1 and the following -helix in N-Sec1 (Misura et al., 2000) as well as in Sly1p (Bracher and Weissenhorn, 2002; Arac et al., 2005) have previously been found to interfere with the binding of cognate syntaxins and with the function of the mutant proteins in vesicular transport (Harrison et al., 1994; Dulubova et al., 2003). As we have shown here, mutation of the analogous aspartate (D59A) in yeast Sly1p generated a conditional mutant, sly1-2, with a relatively unstable protein at elevated temperatures. In contrast to the analogous mutation (D44N) in rat Sly1p, which completely abolished binding of syntaxin 5 (Dulubova et al., 2003), yeast Sly1p domain I carrying the D59A substitution still interacted with Sed5p in vitro, although less efficiently than the same fragment of the wild-type protein (our unpublished observation). As D59A is not in direct contact with the bound syntaxin, less efficient binding of Sly1(D59A) to Sed5p in vitro and destabilization of the mutant protein in vivo may have been caused by an alteration of the conserved five-stranded -sheet and its adjacent helices, of which helix 6 and -strand 5 form the hydrophobic pocket accommodating Sed5p (Bracher and Weissenhorn, 2002). Likewise, mutation I153K (sly1-3) in -strand 5 may significantly alter the structural arrangement of the 5-stranded -sheet by generating a patch of three adjacent lysine residues and resulting in protein instability. Because our recent work disclosed that high-affinity Sed5p binding to Sly1p is not essential for the functioning of the two proteins (Peng and Gallwitz, 2004), the loss of Sed5p binding resulting from the I153K substitution within the hydrophobic binding pocket cannot explain the conditional defect of the sly1-3 mutant.

Amino acid residue R266 when mutated to lysine (sly1-1 allele) was previously shown to result in temperature-sensitive cell growth (Cao et al., 1998). R266, which is located in a short -helix (²) near the N-terminal insertion point of Sly1p domain III into domain II, forms a conserved salt bridge with E618 that interconnects the two parts of domain II. Our finding that the substitution E618A also causes temperature sensitivity (sly1-6 allele) underlines the importance of this salt bridge for structural integrity of the protein. This is further illustrated by the relative lability of both mutant proteins whose intracellular level after 1 h at restrictive temperature dropped to <50% compared with the level at 25°C. However, the two proteins were stable at permissive temperature (Figure 1C), and even after 4 h at 37°C, 25 and 32% of Sly1-1p and Sly1-6p, respectively, were still detectable on protein blots, indicating that loss of function was not simply caused by protein degradation. Although the E618A substitution in sly1-6 is expected to interrupt the salt bridge with R266, the conservative mutation in Sly1-1p (R266K) might interfere primarily with other polar interactions that the guanidinium group of R266 is engaged in with several backbone carbonyl groups (Bracher and Weissenhorn, 2002). Different structural alterations of the SM protein induced by the amino acid substitutions in Sly1-1p and Sly1-6p may explain the different functional alterations exhibited by sly1-1 and sly1-6 mutant cells at restrictive conditions.

A second conserved salt bridge between R258 and D586 that interconnects the N- and C-terminal parts of domain II appears to be less important for the arrangement of the protein's domain structure because substitutions with alanine of neither of the two amino acid residues impaired cell growth at temperatures ranging from 14 to 37°C. Likewise, interference with a conserved salt bridge by mutating E167 of domain II and W328 of domain III proved functionally insignificant, and disruption of conserved hydrogen bonds discovered in the Sly1p crystal structure between D325 and H400, and between K394 and D398 of domain III, were functionally neutral (see Table 2).

On the other hand, substitutions of some amino acid residues involved in conserved hydrogen bonds within domain III of Sly1p (H276/D449 and D286/R452) resulted in conditional mutants, or even in lethality. The H276A mutation caused temperature-sensitive growth already at 30°C and a complete block in ER-to-Golgi forward transport. In vivo, the mutant protein, Sly1-4p, was perfectly stable for at least 4 h at 37°C. As the substitution D449A, which should affect hydrogen bond formation with H276, did not impair cell growth at various temperatures, it seems most likely that the side chain of H276, a residue within the loop between the structurally conserved -helices 10 and 11 (Bracher and Weissenhorn, 2002), serves an important role for functional interactions of the Sly1 protein.

Previous studies have shown that the yeast Sec1 protein with a R432P substitution (sec1-1 allele) causes temperature sensitivity and an accumulation of Golgi-derived vesicles at restrictive temperature (Novick et al., 1979; Brummer et al., 2001). The arginine analogous to R432 in yeast Sec1p is conserved in other yeast, in plant and in mammalian SM proteins. A comparison of crystal structures of Sly1p with those of neuronal Sec1 proteins from rat (Misura et al., 2000) and squid (Bracher and Weissenhorn, 2001) revealed that this highly conserved arginine is positioned in the middle of an -helix (17 in Sly1p; 15 in N-Sec1p) in a larger region of domain III that shares the same topology in both proteins. Our study showed that the substitution R452P in Sly1p, which is analogous to the mutation in Sec1-1p, is lethal. As in general, proline residues are not tolerated in the central part of -helices in globular proteins (Presta and Rose, 1988), this mutation may have destabilized Sly1p and caused its fast degradation. Significant destabilization of Sec1-1p (R432P substitution) and very high reversion frequency of sec1-1 underlines the severity of this mutation (Brummer et al., 2001). However, Sly1p can obviously accommodate glutamate instead of R452 without impairment of its function, suggesting that the -helical arrangement might not be disturbed in such a mutant. In contrast, removal of the large hydrophilic side chain through the substitution of alanine for R452 generated a viable, but temperature-sensitive mutant (sly1-5). The Sly1-5 protein was rather stable, and after 2 h following the shift of mutant cells to the restrictive temperature of 37°C, the level of the mutant protein was still 80% of that of wild type.

The sly1-5 mutant turned out to be of special interest. At nonpermissive conditions, ER-to-Golgi transport of newly synthesized proteins was only slightly affected, as shown by the efficiency with which lysosomal enzymes reached the vacuole or secreted invertase arrived in the periplasmic space, the latter, however, in underglycosylated form. Importantly, sly1-5 mutant cells heavily secreted the ER-resident Kar2p, like the sec20 mutant, which is known to be defective in retrograde transport from the Golgi to the ER. Indications for defects in Golgi-to-ER transport were also obtained with sly1-1 and sly1-2 mutants, but less pronounced than with sly1-5. We further inquired into a possible role of Sly1p in retrograde protein transport by following the intracellular localization of the Golgi protein Rer1p, which is known to cycle through the ER and which was previously shown to be missorted to the vacuole via the multivesicular body (MVB)-sorting pathway in COPI mutants (Sato et al., 2001). Mislocalization of a GFP-Rer1p fusion protein to the vacuole in the sly1-5 mutant at nonpermissive conditions, but not in the forward transport-defective mutant sly1-6, strongly supported the conclusion that Sly1p has an essential function in retrograde transport too. Likewise, disruption of cycling through the ER of the Golgi syntaxin Sed5p in sly1-5 mutant cells led to the same conclusion. As Sly1p was shown to bind to the ER syntaxin Ufe1p (Yamaguchi et al., 2002), although less efficiently than to the Golgi syntaxin Sed5p (Grabowski and Gallwitz, 1997; Peng and Gallwitz, 2004), an additional role of the SM protein in retrograde Golgi-to-ER protein transport was not unexpected but had not been demonstrated until now. In strong support of this newly discovered functional role of Sly1p are the synthetically lethal interactions observed when the sly1-5 allele was combined specifically with various mutant alleles of genes having a clearly defined function in retrograde transport (Table 3).

Another significant outcome of our study is the finding that, similar to the interaction of Sly1p seen with nonsyntaxin SNAREs involved in ER-to-Golgi and intra-Golgi transport (Peng and Gallwitz, 2004), Sly1p could also be demonstrated to bind specifically to nonsyntaxin SNAREs essential for retrograde transport (Sec20p and Use1p) and to be part of assembled SNARE complexes at the ER. Importantly, we observed that compared with wild-type cells, the fraction of the ER-nonsyntaxin SNAREs (Use1p and Sec20p) that coprecipitated with Sly1p from lysates of sly1-5 mutant cells was significantly reduced, whereas complexes of ER-to-Golgi SNAREs (Bet1p and Bos1p) with Sly1p in wild-type and sly1-5 cell lysates were almost indistinguishable. Because coprecipitation with Sly1-5p of the ER syntaxin Ufe1p was similarly diminished, it seems likely that in the sly1-5 mutant, a failure in the efficient generation of SNARE complexes at ER membranes might have caused the rather specific inhibition of retrograde vesicular transport between Golgi compartments and the ER. We have not yet investigated with purified proteins whether the direct binding of Sly1-5 mutant protein to ER-SNAREs might be impaired. However, because it has been demonstrated previously that Ufe1p, like Sed5p, binds to a short N-terminal peptide region of Sly1p domain I (Yamaguchi et al., 2002), it is highly unlikely that Ufe1p binding to Sly1-5p with its single amino acid substitution in domain III could be affected. The significant reduction of ER-SNARE complexes in sly1-5 cells again is in support of a role of Sly1p in retrograde transport from the Golgi to the ER, which appears to be specifically affected by the R452A substitution of the SM protein. Given the observation that R432P in yeast Sec1-1p apparently affects the interaction of the SM protein with other binding partners (Brummer et al., 2001), one might assume that, especially at elevated temperature, the R452A substitution in Sly1p alters the -helical arrangement of this structurally conserved region in such a way that the interaction with protein(s) of the fusion machinery at the ER is no longer possible.

It is widely accepted that SM proteins are essential components of the membrane fusion machineries at all steps of vesicular transport. In yeast, there are only four SM proteins, Sly1p, Sec1p, Vps45p, and Vps33p. Vps33p has been suggested to act between Golgi and endosomes as well as in vacuole fusion (Subramanian et al., 2004). Our finding that the function of Sly1p is not only connected to fusion of ER-derived vesicles with Golgi membranes but also of Golgi-derived vesicular carriers with the ER is an instructive example for a single SM protein acting in more than one fusion step through specific interaction with subsets of single SNAREs, and, possibly, their assembled complexes.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
We thank Winfried Weissenhorn for suggestions and discussions at the early stage of this study, Akihiko Nakano, Hans-Dieter Schmitt, Gabriele Fischer von Mollard, Sigurd Braun, and Stefan Jentsch for plasmids, yeast strains, and antibodies. We are indebted to Peter Mienkus and Heike Behr for competent technical assistance. This work was supported by the Max Planck Society and by grants to D.G. from the Human Frontier Science Program and Fonds der Chemischen Industrie.

	Footnotes

 
This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E05–02–0114) on June 15, 2005.

Abbreviations used: ER, endoplasmic reticulum; SM, Sec1/Munc18 proteins; SNARE, soluble N-ethylmaleimide-sensitive fusion (NSF)-attachment protein receptors; CPY, carboxypeptidase; ALP, alkaline phosphatase.

Address correspondence to: Renwang Peng (rpeng{at}gwdg.de) or Dieter Gallwitz (Dieter.Gallwitz{at}mpi-bpc.mpg.de).

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Andag, U., Neumann, T., and Schmitt, H. D. ((2001). ). The Coatomer-interacting protein Dsl1p is required for Golgi-to-endoplasmic reticulum retrieval in yeast. J. Biol. Chem. 276, , 39150–39160.[Abstract/Free Full Text]

Arac, D., Dulubova, I., Pei, J., Huryeva, I., Grishin, N. V., and Rizo, J. ((2005). ). Three-dimensional structure of the rSly1 N-terminal domain reveals a conformational change induced by binding to Syntaxin 5. J. Mol. Biol. 346, , 589–601.[CrossRef][Medline]

Ballensiefen, W., Ossipov, D., and Schmitt, H. D. ((1998). ). Recycling of the yeast v-SNARE Sec22p involves COPI-proteins and the ER transmembrane proteins Ufe1p and Sec20p. J. Cell Sci. 111, , 1507–1520.[Abstract]

Boehm, J., Letourneur, F., Ballensiefen, W., Ossipov, D., Demolliere, C., and Schmitt, H. D. ((1997). ). Sec12p requires Rer1p for sorting to coatomer (COPI)-coated vesicles and retrieval to the ER. J. Cell Sci. 110, , 991–1003.[Abstract]

Bracher, A., and Weissenhorn, W. ((2002). ). Structural basis for the Golgi membrane recruitment of Sly1p by Sed5p. EMBO J. 21, , 6114–6124.[CrossRef][Medline]

Brenner, S. ((1974). ). The genetics of Caenorhabditis elegans. Genetics 77, , 71–94.[Abstract/Free Full Text]

Brummer, M. H., Kivinen, K. J., Jantti, J., Toikkanen, J., Soderlund, H., and Keranen, S. ((2001). ). Characterization of the sec1-1 and sec1-11 mutations. Yeast 18, , 1525–1536.[CrossRef][Medline]

Burri, L., Varlamov, O., Doege, C. A., Hofmann, K., Beilharz, T., Rothman, J. E., Sollner, T. H., and Lithgow, T. ((2003). ). A SNARE required for retrograde transport to the endoplasmic reticulum. Proc. Natl. Acad. Sci. USA 100, , 9873–9877.[Abstract/Free Full Text]

Cao, X., and Barlowe, C. ((2000). ). Asymmetric requirements for a Rab GTPase and SNARE proteins in fusion of COPII vesicles with acceptor membranes. J. Cell Biol. 149, , 55–66.[Abstract/Free Full Text]

Cao, X., Ballew, N., and Barlowe, C. ((1998). ). Initial docking of ER-derived vesicles requires Uso1p and Ypt1p but is independent of SNARE proteins. EMBO J. 17, , 2156–2165.[CrossRef][Medline]

Carr, C. M., Grote, E., Munson, M., Hughson, F. M., and Novick, P. ((1999). ). Sec1p binds to SNARE complexes and concentrates at sites of secretion. J. Cell Biol. 146, , 333–344.[Abstract/Free Full Text]

Ciufo, L. F., Barclay, J. W., Burgoyne, R. D., and Morgan, A. ((2005). ). Munc18-1 regulates early and late stages of exocytosis via syntaxin-independent protein interactions. Mol. Biol. Cell 16, , 470–482.[Abstract/Free Full Text]

Coe, J. G., Lim, A. C., Xu, J., and Hong, W. ((1999). ). A role for Tlg1p in the transport of proteins within the Golgi apparatus of Saccharomyces cerevisiae. Mol. Biol. Cell 10, , 2407–2423.[Abstract/Free Full Text]

Cosson, P., Schroder-Kohne, S., Sweet, D. S., Demolliere, C., Hennecke, S., Frigerio, G., and Letourneur, F. ((1997). ). The Sec20/Tip20p complex is involved in ER retrieval of dilysine-tagged proteins. Eur. J. Cell Biol. 73, , 93–97.

Dascher, C., Ossig, R., Gallwitz, D., and Schmitt, H. D. ((1991). ). Identification and structure of four yeast genes (SLY) that are able to suppress the functional loss of YPT1, a member of the RAS superfamily. Mol. Cell Biol. 11, , 872–885.[Abstract/Free Full Text]

Dilcher, M., Veith, B., Chidambaram, S., Hartmann, E., Schmitt, H. D., and Fischer von Mollard, G. ((2003). ). Use1p is a yeast SNARE protein required for retrograde traffic to the ER. EMBO J. 22, , 3664–3674.[CrossRef][Medline]

Dulubova, I., Yamaguchi, T., Arac, D., Li, H., Huryeva, I., Min, S. W., Rizo, J., and Sudhof, T. C. ((2003). ). Convergence and divergence in the mechanism of SNARE binding by Sec1/Munc18-like proteins. Proc. Natl. Acad. Sci. USA 100, , 32–37.[Abstract/Free Full Text]

Dulubova, I., Yamaguchi, T., Gao, Y., Min, S. W., Huryeva, I., Sudhof, T. C., and Rizo, J. ((2002). ). How Tlg2p/syntaxin 16 `snares' Vps45. EMBO J. 21, , 3620–3631.[CrossRef][Medline]

Dulubova, I., Yamaguchi, T., Wang, Y., Sudhof, T. C., and Rizo, J. ((2001). ). Vam3p structure reveals conserved and divergent properties of syntaxins. Nat. Struct. Biol. 8, , 258–264.[CrossRef][Medline]

Dulubova, I., Sugita, S., Hill, S., Hosaka, M., Fernandez, I., Sudhof, T. C., and Rizo, J. ((1999). ). A conformational switch in syntaxin during exocytosis: role of munc18. EMBO J. 18, , 4372–4382.[CrossRef][Medline]

Gallwitz, D., and Jahn, R. ((2003). ). The riddle of the Sec1/Munc-18 proteins— new twists added to their interactions with SNAREs. Trends Biochem. Sci. 28, , 113–116.[CrossRef][Medline]

Garcia, E. P., Gatti, E., Butler, M., Burton, J., and De Camilli, P. ((1994). ). A rat brain Sec1 homologue related to Rop and UNC18 interacts with syntaxin. Proc. Natl. Acad. Sci. USA 91, , 2003–2007.[Abstract/Free Full Text]

Gietz, D., St. Jean, A., Woods, R. A., and Schiestl, R. H. ((1992). ). Improved method for high efficiency transformation of intact yeast. Nucleic Acids Res. 20, , 1425.[Free Full Text]

Grabowski, R., and Gallwitz, D. ((1997). ). High-affinity binding of the yeast cis-Golgi t-SNARE, Sed5p, to wild-type and mutant Sly1p, a modulator of transport vesicle docking. FEBS Lett. 14, , 169–172.

Hakes, D. J., and Dixon, J. E. ((1992). ). New vectors for high level expression of recombinant proteins in bacteria. Anal. Biochem. 202, , 293–298.[CrossRef][Medline]

Halachmi, N., and Lev, Z. ((1996). ). The Sec1 family: a novel family of proteins involved in synaptic transmission and general secretion. J. Neurochem. 66, , 889–897.[Medline]

Harrison, S. D., Broadie, K., van de Goor, J., and Rubin, G. M. ((1994). ). Mutations in the Drosophila Rop gene suggest a function in general secretion and synaptic transmission. Neuron 13, , 555–566.[CrossRef][Medline]

Hata, Y., Slaughter, C. A., and Sudhof, T. C. ((1993). ). Synaptic vesicle fusion complex contains unc-18 homologue bound to syntaxin. Nature 366, , 347–351.[CrossRef][Medline]

Hirose, H., Arasaki, K., Dohmae, N., Takio, K., Hatsuzawa, K., Nagahama, M., Tani, K., Yamamoto, A., Tohyama, M., and Tagaya, M. ((2004). ). Implication of ZW10 in membrane trafficking between the endoplasmic reticulum and Golgi. EMBO J. 23, , 1267–1278.[CrossRef][Medline]

Jahn, R., Lang, T., and Sudhof, T. C. ((2003). ). Membrane fusion. Cell 112, , 519–533.[CrossRef][Medline]

Kosodo, Y., Noda, Y., Adachi, H., and Yoda, K. ((2002). ). Binding of Sly1 to Sed5 enhances formation of the yeast early Golgi SNARE complex. J. Cell Sci. 115, , 3683–3691.[Abstract/Free Full Text]

Lewis, M. J., and Pelham, H. R. ((1996). ). SNARE-mediated retrograde traffic from the Golgi complex to the endoplasmic reticulum. Cell 19, , 205–215.

Lewis, M. J., Rayner, J. C., and Pelham, H.R.B. ((1997). ). A novel SNARE complex implicated in vesicle fusion with the endoplasmic reticulum. EMBO J. 16, , 3017–3024.[CrossRef][Medline]

Misura, K. M., Scheller, R. H., and Weis, W. I. ((2000). ). Three-dimensional structure of the neuronal-Sec1-syntaxin 1a complex. Nature 404, , 355–362.[CrossRef][Medline]

Mizuta, K., and Warner, J. R. ((1994). ). Continued functioning of the secretory pathway is essential for ribosome synthesis. Mol. Cell Biol. 14, , 2493–2502.[Abstract/Free Full Text]

Nichols, B. J., Ungermann, C., Pelham, H. R., Wickner, W. T., and Haas, A. ((1997). ). Homotypic vacuolar fusion mediated by t- and v-SNAREs. Nature 387, , 199–202.[CrossRef][Medline]

Nichols, B. J., Holthuis, J. C., and Pelham, H. R. ((1998). ). The Sec1p homologue Vps45p binds to the syntaxin Tlg2p. Eur. J. Cell Biol. 77, , 263–268.[Medline]

Novick, P., and Schekman, R. ((1979). ). Secretion and cell-surface growth are blocked in a temperature-sensitive mutant of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 76, , 1858–1862.[Abstract/Free Full Text]

Ossig, R., Dascher, C., Trepte, H. H., Schmitt, H. D., and Gallwitz, D. ((1991). ). The yeast SLY gene products, suppressors of defects in the essential GTP-binding Ypt1 protein, may act in endoplasmic reticulum-to-Golgi transport. Mol. Cell Biol. 11, , 872–885.

Peng, R., and Gallwitz, D