A Syntaxin Homolog Encoded by VAM3 Mediates Down-regulation of a Yeast G Protein-coupled Receptor
1999; Elsevier BV; Volume: 274; Issue: 3 Linguagem: Inglês
10.1074/jbc.274.3.1835
ISSN1083-351X
AutoresChristopher J. Stefan, Kendall Blumer,
Tópico(s)Neurobiology and Insect Physiology Research
ResumoG protein-coupled receptors that transduce signals for many hormones, neurotransmitters, and inflammatory mediators are internalized and subsequently recycled to the plasma membrane, or down-regulated by targeting to lysosomes for degradation. Here we have characterized yeast α-factor receptors tagged with green fluorescent protein (Ste2-GFP) and used them to obtain mutants defective in receptor down-regulation. In wild type cells, Ste2-GFP was functional and localized to the plasma membrane and endocytic compartments. Although GFP was fused to the cytoplasmic tail of the receptor, GFP also accumulated in the lumen of the vacuole, suggesting that the receptor's extracellular and cytoplasmic domains are degraded within the vacuole lumen. Transposon mutagenesis and a visual screen were used to identify mutants displaying aberrant localization of Ste2-GFP. Mutants that accumulated Ste2-GFP in numerous intracellular vesicles carried disruptions of the VAM3/PTH1 gene, which encodes a syntaxin homolog (t-SNARE) required for homotypic vacuole membrane fusion, autophagy and fusion of biosynthetic transport vesicles with the vacuole. We provide evidence that Vam3 is required for the delivery of α-factor receptor-ligand complexes to the vacuole. Vam3 homologs in mammalian cells may mediate late steps in the down-regulation and lysosomal degradation pathways of various G protein-coupled receptors. G protein-coupled receptors that transduce signals for many hormones, neurotransmitters, and inflammatory mediators are internalized and subsequently recycled to the plasma membrane, or down-regulated by targeting to lysosomes for degradation. Here we have characterized yeast α-factor receptors tagged with green fluorescent protein (Ste2-GFP) and used them to obtain mutants defective in receptor down-regulation. In wild type cells, Ste2-GFP was functional and localized to the plasma membrane and endocytic compartments. Although GFP was fused to the cytoplasmic tail of the receptor, GFP also accumulated in the lumen of the vacuole, suggesting that the receptor's extracellular and cytoplasmic domains are degraded within the vacuole lumen. Transposon mutagenesis and a visual screen were used to identify mutants displaying aberrant localization of Ste2-GFP. Mutants that accumulated Ste2-GFP in numerous intracellular vesicles carried disruptions of the VAM3/PTH1 gene, which encodes a syntaxin homolog (t-SNARE) required for homotypic vacuole membrane fusion, autophagy and fusion of biosynthetic transport vesicles with the vacuole. We provide evidence that Vam3 is required for the delivery of α-factor receptor-ligand complexes to the vacuole. Vam3 homologs in mammalian cells may mediate late steps in the down-regulation and lysosomal degradation pathways of various G protein-coupled receptors. G protein-coupled receptors (GPCRs) 1The abbreviations used are: GPCR, G protein-coupled receptor; CMAC, 7-aminochloromethylcoumarin; CPY, carboxypeptidase Y; G protein, guanine nucleotide-binding regulatory protein; GFP, green fluorescent protein; PIPES, 1,4-piperazinediethanesulfonic acid; SNARE, solubleNSF attachment proteinreceptor. 1The abbreviations used are: GPCR, G protein-coupled receptor; CMAC, 7-aminochloromethylcoumarin; CPY, carboxypeptidase Y; G protein, guanine nucleotide-binding regulatory protein; GFP, green fluorescent protein; PIPES, 1,4-piperazinediethanesulfonic acid; SNARE, solubleNSF attachment proteinreceptor. function in signal transduction pathways that allow eukaryotic cells to respond to sensory stimuli and extracellular signals, including various hormones and neurotransmitters. In response to agonist stimulation, GPCRs are regulated by phosphorylation, sequestration, and down-regulation, which limit the strength or duration of physiological responses (1Kobilka B. Annu. Rev. Neurosci. 1992; 15: 87-114Crossref PubMed Scopus (312) Google Scholar, 2Carman C.V. Benovic J.L. Curr. Opin. Neurobiol. 1998; 8: 335-344Crossref PubMed Scopus (230) Google Scholar). Whereas down-regulation persistently attenuates signaling by targeting GPCRs for lysosomal degradation, sequestration apparently serves to reactivate desensitized receptors by allowing them to be dephosphorylated and recycled in active form to the cell surface (3Pippig S. Andexinger S. Lohse M.J. Mol. Pharmacol. 1995; 47: 666-676PubMed Google Scholar,4Krueger K.M. Daaka Y. Pitcher J.A. Lefkowitz R.J. J. Biol. Chem. 1997; 272: 5-8Abstract Full Text Full Text PDF PubMed Scopus (311) Google Scholar). Recent studies have begun to elucidate the mechanisms governing the sequestration and down-regulation of GPCRs (2Carman C.V. Benovic J.L. Curr. Opin. Neurobiol. 1998; 8: 335-344Crossref PubMed Scopus (230) Google Scholar). For example, β-arrestin acts as an adaptor that binds GPCRs and clathrin (5Goodman Jr., O.B. Krupnick J.G. Santini F. Gurevich V.V. Penn R.B. Gagnon A.W. Keen J.H. Benovic J.L. Nature. 1996; 383: 447-450Crossref PubMed Scopus (1140) Google Scholar), recruiting receptors into coated pits that pinch off by a dynamin-dependent mechanism (6Gagnon A.W. Kallal L. Benovic J.L. J. Biol. Chem. 1998; 273: 6976-6981Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar). In contrast, the molecular mechanisms governing the transport of GPCRs along subsequent steps of the endocytic pathways responsible for sequestration or down-regulation are less well understood. GPCRs activated by peptide mating pheromones in the yeastSaccharomyces cerevisiae have provided insights into the molecular mechanisms of receptor down-regulation (7Riezman H. Semin. Cell Dev. Biol. 1998; 9: 129-134Crossref PubMed Scopus (11) Google Scholar). Agonist binding triggers phosphorylation and subsequent ubiquitination of yeast α-factor receptors on their cytoplasmic C-terminal domains (8Reneke J.E. Blumer K.J. Courchesne W.E. Thorner J. Cell. 1988; 55: 221-234Abstract Full Text PDF PubMed Scopus (239) Google Scholar, 9Hicke L. Riezman H. Cell. 1996; 84: 277-287Abstract Full Text Full Text PDF PubMed Scopus (665) Google Scholar). Ubiquitinated receptors are internalized by processes that involve clathrin, actin, myosin, and fimbrin (10Riezman H. Munn A. Geli M.I. Hicke L. Experientia (Basel). 1996; 52: 1033-1041Crossref PubMed Scopus (82) Google Scholar), and subsequently transported through vesicular intermediates similar to early and late endosomes in mammalian cells (11Hicke L. Zanolari B. Pypaert M. Rohrer J. Riezman H. Mol. Biol. Cell. 1997; 8: 13-31Crossref PubMed Scopus (92) Google Scholar, 12Prescianotto-Baschong C. Riezman H. Mol. Biol. Cell. 1998; 9: 173-189Crossref PubMed Scopus (89) Google Scholar). In late endosomal/prevacuolar compartments, pheromone receptor trafficking is thought to converge with biosynthetic pathways that deliver precursors of resident vacuolar proteins (13Davis N.G. Horecka J.L. Sprague Jr., G.F. J. Cell Biol. 1993; 122: 53-65Crossref PubMed Scopus (189) Google Scholar, 14Piper R.C. Cooper A.A. Yang H. Stevens T.H. J. Cell Biol. 1995; 131: 603-617Crossref PubMed Scopus (340) Google Scholar, 15Rieder S.E. Banta L.M. Kohrer K. McCaffery J.M. Emr S.D. Mol. Biol. Cell. 1996; 7: 985-999Crossref PubMed Scopus (239) Google Scholar). Subsequent events transport pheromone-receptor complexes to the vacuole where they are degraded by vacuolar hydrolases (11Hicke L. Zanolari B. Pypaert M. Rohrer J. Riezman H. Mol. Biol. Cell. 1997; 8: 13-31Crossref PubMed Scopus (92) Google Scholar,16Singer B. Riezman H. J. Cell Biol. 1990; 110: 1911-1922Crossref PubMed Scopus (69) Google Scholar). Yeast pheromone receptors transit the endocytic pathway through the action of several gene products. For example, a small GTP-binding protein (Ypt5/Vps21) and an NSF homolog (Sec18), regulate fusion or maturation of early endosomes (11Hicke L. Zanolari B. Pypaert M. Rohrer J. Riezman H. Mol. Biol. Cell. 1997; 8: 13-31Crossref PubMed Scopus (92) Google Scholar, 17Singer-Kruger B. Stenmark H. Dusterhoft A. Philippsen P. Yoo J.S. Gallwitz D. Zerial M. J. Cell Biol. 1994; 125: 283-298Crossref PubMed Scopus (182) Google Scholar), and two syntaxin homologs, Tlg1 and Tlg2, are required for receptor down-regulation (18Holthuis J.C. Nichols B.J. Dhruvakumar S. Pelham H.R. EMBO J. 1998; 17: 113-126Crossref PubMed Scopus (215) Google Scholar). Here we have used yeast cells expressing green fluorescent protein fused to the α-factor receptor (Ste2-GFP) as a new tool to localize receptors in compartments of the endocytic pathway and to identify gene products involved in receptor down-regulation. We describe evidence that a syntaxin homolog encoded by the VAM3 gene is required for delivery of Ste2 to the vacuole, a late step in the down-regulation pathway. Enzymes used for recombinant DNA methods were purchased from commercial sources and used according to the suppliers' recommendations. Sources of growth media for yeast and bacterial cells have been described (19Stefan C.J. Overton M.C. Blumer K.J. Mol. Biol. Cell. 1998; 9: 885-899Crossref PubMed Scopus (51) Google Scholar). [35S]Na2SO4 (carrier-free) was obtained from ICN. 7-Aminochloromethylcoumarin (CMAC) and monoclonal antibodies specific for carboxypeptidase Y were obtained from Molecular Probes. The S. cerevisiaestrains used in these studies were JE115 (MAT a ura3–52 leu2–3, 112 his3Δ-1 trp1 ste2Δ::LEU2 sst1-Δ5; Ref. 8Reneke J.E. Blumer K.J. Courchesne W.E. Thorner J. Cell. 1988; 55: 221-234Abstract Full Text PDF PubMed Scopus (239) Google Scholar), KBY49 (vam3–100 derivative of JE115), KBY62 (vam3–101 derivative of JE115), KBY58 (ste2Δ derivative of JE115), and KBY65 (vam3Δ derivative of KBY58). The unmarked ste2Δ allele of KBY58 was introduced by two-step gene deletion using JE115 and ClaI-cut YIp5ste2Δ (19Stefan C.J. Overton M.C. Blumer K.J. Mol. Biol. Cell. 1998; 9: 885-899Crossref PubMed Scopus (51) Google Scholar), and confirmed by loss of LEU2. The VAM3 coding region was deleted usingApaI/SacI-cut pvam3::LEU2 (20Darsow T. Rieder S.E. Emr S.D. J. Cell Biol. 1997; 138: 517-529Crossref PubMed Scopus (294) Google Scholar), to create KBY65. To localize α-factor receptors in living cells, we constructed plasmid pRS314STE2-GFP, which expresses green fluorescent protein (GFP) fused to wild type receptors at their extreme C termini. Polymerase chain reaction was used to generate GFP(S65T) coding sequences flanked by BamHI restriction sites. This polymerase chain reaction product was digested with BamHI and cloned into pGEX2T (Amersham Pharmacia Biotech) that had been cut with BamHI to create pGEX2T-GFP. Plasmid pRS314STE2-GFP was created by inserting a BamHI fragment containing the GFP (S65T) coding region in the appropriate reading frame from pGEX2T-GFP into the BglII site of pRS314STE2Δter in which the wild type termination codon was inactivated (19Stefan C.J. Overton M.C. Blumer K.J. Mol. Biol. Cell. 1998; 9: 885-899Crossref PubMed Scopus (51) Google Scholar). To fuse GFP to α-factor receptors lacking their C-terminal cytoplasmic tails, we inserted the BamHI fragment containing the GFP(S65T) coding region in the appropriate reading frame from pGEX2T-GFP into the BglII site of pRS314ste2-300ter (19Stefan C.J. Overton M.C. Blumer K.J. Mol. Biol. Cell. 1998; 9: 885-899Crossref PubMed Scopus (51) Google Scholar), to create plasmid pRS314ste2Δtail-GFP. To express VAM3, plasmid pRS313VAM3 was constructed by isolating a 1.8-kilobase pairXmnI-ApaI fragment encompassing theVAM3 locus from pVAM3.416 (20Darsow T. Rieder S.E. Emr S.D. J. Cell Biol. 1997; 138: 517-529Crossref PubMed Scopus (294) Google Scholar), and inserting it into pRS313 that had been cleaved with SmaI andApaI. The vam3–6 temperature-sensitive allele was expressed from pvam3–6.416 (20Darsow T. Rieder S.E. Emr S.D. J. Cell Biol. 1997; 138: 517-529Crossref PubMed Scopus (294) Google Scholar). Transposon insertion mutagenesis and a visual screen were performed to identify cells that displayed aberrant localization of Ste2-GFP. Two pools of a transposon-mutagenized yeast genomic DNA library (21Chun K.T. Goebl M.G. Genetics. 1996; 142: 39-50Crossref PubMed Google Scholar) cleaved withNotI were transformed into JE115 cells carrying theSTE2-GFP fusion on a centromeric plasmid. Approximately 6,000 transformants were treated individually with pheromone (1 μm α-factor, 3 h) and screened using fluorescence microscopy for abnormal distribution of Ste2-GFP. To identify genes affected in the mutants isolated from the visual screen, chromosomal DNA was prepared, digested with EcoRI,HindIII, or BamHI, which cleaves once in the transposable element and cleaves in the adjacent chromosomal DNA (21Chun K.T. Goebl M.G. Genetics. 1996; 142: 39-50Crossref PubMed Google Scholar). Fragments containing the bla gene from the transposable element and flanking yeast genomic DNA were cloned into pCRScriptCAM SK(+) (Stratagene) cleaved with EcoRI, HindIII, or BamHI. These plasmids were introduced intoEscherichia coli by selecting for resistance to ampicillin and chloramphenicol. Plasmid DNA was prepared and sequenced using a primer that anneals to the Tn3 38 base pair repeat at the left end of the transposon; this identified the chromosomal locus disrupted by the transposable element. Quantitative bioassays (halo assays) were used to measure the apparent sensitivity of cells to pheromone-induced growth arrest (22Sprague G.J. Methods Enzymol. 1991; 194: 77-93Crossref PubMed Scopus (224) Google Scholar). The relative responsiveness of cells was determined by comparing the amount of α-factor required to form a halo 20 mm in diameter, as indicated by interpolating dose-response curves. Quantitative mating assays were performed as described (22Sprague G.J. Methods Enzymol. 1991; 194: 77-93Crossref PubMed Scopus (224) Google Scholar). Methods used to purify [35S]α-factor and perform ligand binding assays with inviable, intact cells have been described (19Stefan C.J. Overton M.C. Blumer K.J. Mol. Biol. Cell. 1998; 9: 885-899Crossref PubMed Scopus (51) Google Scholar). Assays of cells expressing wild type or GFP-tagged receptors employed [35S]α-factor (30 Ci/mmol) at concentrations ranging from 1 to 10 nm. Ligand binding data were plotted according to the method of Scatchard and fitted by nonlinear least mean square regression. Nonspecific binding was determined in the presence of a 500-fold excess of unlabeled α-factor. Endocytosis of the α-factor receptor was determined by measuring the rates that cells internalize bound 35S-labeled α-factor, as described previously (23Dulic V. Egerton M. Elguindi I. Raths S. Singer B. Riezman H. Methods Enzymol. 1991; 194: 697-710Crossref PubMed Scopus (162) Google Scholar). Rates of degradation of internalized35S-labeled α-factor by wild type cells orvam3 mutants were determined as described previously (11Hicke L. Zanolari B. Pypaert M. Rohrer J. Riezman H. Mol. Biol. Cell. 1997; 8: 13-31Crossref PubMed Scopus (92) Google Scholar,23Dulic V. Egerton M. Elguindi I. Raths S. Singer B. Riezman H. Methods Enzymol. 1991; 194: 697-710Crossref PubMed Scopus (162) Google Scholar), with the following modifications. Cells were grown at 26 °C or 30 °C to a density of 107 cells/ml, washed, and suspended in YPB (10 mm PIPES, pH 6.0, 1 mmMgCl2, 0.1 mm EDTA, 1% yeast extract, 2% bactopeptone). 35S-Labeled α-factor (1 μCi, 100 nm final concentration) was bound to cells for 1 h under conditions that inhibit ligand-induced internalization of receptors (YPB at 0 °C). Unbound ligand was removed by pelleting cells, and cells were suspended in YPD medium, pH 6.0, at 26 °C, or 38 °C if vam3 temperature-sensitive mutants were analyzed. Receptor internalization was allowed to proceed. Aliquots (100 μl) were removed at various times and diluted into 10 ml 50 mm sodium citrate buffer, pH 1.1, containing 10 mm KF on ice to remove surface-bound α-factor. Cells were filtered, washed with 50 mm potassium phosphate buffer, pH 6.0, containing 10 mm NaN3 and 10 mm KF, and suspended in extraction buffer (23Dulic V. Egerton M. Elguindi I. Raths S. Singer B. Riezman H. Methods Enzymol. 1991; 194: 697-710Crossref PubMed Scopus (162) Google Scholar). Internalized 35S-labeled α-factor extracted from cells was subjected to thin layer chromatography to resolve intact and degraded α-factor species, as described previously (23Dulic V. Egerton M. Elguindi I. Raths S. Singer B. Riezman H. Methods Enzymol. 1991; 194: 697-710Crossref PubMed Scopus (162) Google Scholar). Cultures were grown to a density of 2 × 107 cells/ml in synthetic medium (SD-uracil) to select for plasmids pVAM3.416 and pvam3–6.416. Immunoblotting methods used to detect carboxypeptidase Y (CPY) in yeast whole-cell extracts were similar to those previously described (24Srivastava A. Jones E.W. Genetics. 1998; 148: 85-98Crossref PubMed Google Scholar). Cells were lysed by mechanical disruption in Laemmli sample buffer, and the protein concentration of the cleared lysates was determined by the Bradford method and adjusted to 1 mg/ml with Laemmli sample buffer prior to SDS-polyacrylamide gel electrophoresis. Yeast cell cultures were labeled with CMAC at a concentration of 100 μm for 15 min in synthetic medium at room temperature to examine vacuole morphology. To visualize α-factor receptors, cultures were grown to a density of 2 × 107 cells/ml in synthetic medium (SD-tryptophan) to select for plasmids that express GFP-tagged receptors. Cells were harvested by centrifugation, suspended in low fluorescence medium (25Waddle J.A. Karpova T.S. Waterston R.H. Cooper J.A. J. Cell Biol. 1996; 132: 861-870Crossref PubMed Scopus (182) Google Scholar), and observed under an Olympus epifluorescence microscope equipped with UG-1, BP490, and BP545 dichroic filters and a cooled CCD camera (Dage). To detect changes in the distribution of receptors upon treating cells with pheromone, we mixed MAT a cells (JE115) expressing GFP-tagged receptors (from pRS314STE2-GFP) with a 4-fold excess of MATα cells (RK537–3B) to provide a source of α-factor. The cell suspension (2 μl) was placed on an agarose pad containing medium as described previously (25Waddle J.A. Karpova T.S. Waterston R.H. Cooper J.A. J. Cell Biol. 1996; 132: 861-870Crossref PubMed Scopus (182) Google Scholar). Cells were imaged at room temperature on an Olympus epifluorescence microscope equipped with stage and shutter controllers. Time lapse fluorescence images were collected by acquisition of 10 focal planes (0.5 μm apart), and collapsed into a single two-dimensional image as described previously (25Waddle J.A. Karpova T.S. Waterston R.H. Cooper J.A. J. Cell Biol. 1996; 132: 861-870Crossref PubMed Scopus (182) Google Scholar). To characterize the endocytic trafficking of yeast α-factor receptors and to identify mutants defective in this process, we needed a rapid and reliable means of examining receptor localization. Accordingly, we constructed a centromeric plasmid that expresses green fluorescent protein (GFP, S65T) fused to the extreme C-terminal cytoplasmic tail of the α-factor receptor, Ste2-GFP (Fig.1). We also constructed a similar plasmid that expresses GFP fused to a truncated α-factor receptor lacking its cytoplasmic C-terminal tail, Ste2Δtail-GFP, which should not be internalized from the cell surface (8Reneke J.E. Blumer K.J. Courchesne W.E. Thorner J. Cell. 1988; 55: 221-234Abstract Full Text PDF PubMed Scopus (239) Google Scholar, 26Konopka J.B. Jenness D.D. Hartwell L.H. Cell. 1988; 54: 609-620Abstract Full Text PDF PubMed Scopus (166) Google Scholar). These two forms of Ste2-GFP were expressed from the STE2 promoter in a ste2null mutant and characterized with regard to signaling activity, agonist binding affinity, cell surface expression, and internalization kinetics. Tagging full-length α-factor receptors with GFP did not appear to affect receptor signaling, agonist binding affinity, cell surface expression, or internalization. Tagged and untagged receptors displayed similar agonist binding affinities (K d = 4.9 and 5.3 nm for Ste2-GFP and Ste2, respectively; Fig. 1 A) and cell surface expression levels (B max = 15,000 and 19,000 sites/cell respectively, Fig. 1 A). Cells expressing Ste2-GFP responded to pheromone with nearly normal efficiency (Fig. 1 B) and mated as efficiently as cells expressing untagged receptors (data not shown). Furthermore, cells expressing Ste2-GFP internalized receptors at a rate indistinguishable from that of cells expressing full-length untagged receptors, as indicated by measuring rates of 35S-labeled α-factor uptake (Fig. 1 C). Cells expressing receptors lacking their cytoplasmic C-terminal domains (Ste2Δtail-GFP) were defective in this assay (Fig. 1 C) and conferred a pheromone supersensitive phenotype identical to untagged truncated receptors (data not shown), as expected (8Reneke J.E. Blumer K.J. Courchesne W.E. Thorner J. Cell. 1988; 55: 221-234Abstract Full Text PDF PubMed Scopus (239) Google Scholar, 26Konopka J.B. Jenness D.D. Hartwell L.H. Cell. 1988; 54: 609-620Abstract Full Text PDF PubMed Scopus (166) Google Scholar). We subsequently examined the localization of full-length Ste2-GFP in the absence or presence of mating pheromone. Consistent with our previous immunofluorescence studies using fixed cells expressing myc-tagged receptors (19Stefan C.J. Overton M.C. Blumer K.J. Mol. Biol. Cell. 1998; 9: 885-899Crossref PubMed Scopus (51) Google Scholar), Ste2-GFP was present at the cell surface (Fig.2 A), particularly in daughter cells, probably due to polarization of the secretory pathway. Ste2-GFP was also detected at the tips of cell surface projections induced 2 h after exposing cell cultures to pheromone (data not shown), similar to what has been reported using untagged receptors (27Jackson C.L. Konopka J.B. Hartwell L.H. Cell. 1991; 67: 389-402Abstract Full Text PDF PubMed Scopus (94) Google Scholar). Ste2-GFP also localized to two morphologically distinct types of intracellular vesicular structures. Smaller vesicles (usually >5/cell; Fig. 2 A, single arrowhead) were present throughout the cytoplasm. The second type of vesicular structures were larger (Fig.2 A, double arrow), less numerous (typically 2–5/cell), and nearly always located next to the vacuole, which was stained with the vital dye CMAC (28Stewart A. Deacon J.W. Biotech. Histochem. 1995; 70: 57-65Crossref PubMed Scopus (25) Google Scholar). Two approaches were used to determine whether the intracellular vesicles observed in cells expressing Ste2-GFP were secretory or endocytic. First, we examined the localization of Ste2Δtail-GFP, which fails to undergo endocytosis. Ste2Δtail-GFP was present at the cell surface, but it did not localize to intracellular vesicular structures (Fig. 2 A). A small proportion of Ste2Δtail-GFP also localized to a perinuclear ring similar to the endoplasmic reticulum, suggesting that this fusion protein has a slight biosynthetic defect. This ring structure was not the vacuole membrane because it did not stain with CMAC. Second, because receptor internalization is stimulated by pheromone, we determined whether the number of intracellular vesicles increases when cells are exposed to the pheromone secreted by a neighboring cell of opposite mating type. For these experiments, MAT acells expressing Ste2-GFP and MATα cells that do not express a GFP-tagged protein were mixed and mounted on a medium-containing agarose pad. Time-lapse fluorescence video microscopy was used to follow the distribution of Ste2-GFP in single cells over time. Consistent with an endocytic origin, the number of vesicles containing Ste2-GFP increased 2–4-fold as cells responded over time to the pheromone produced by neighboring cells of opposite mating type (Fig. 2 B). Increases in vesicle number were detected 20 min after cells were first exposed to pheromone, before receptor expression is induced. Results of these two experiments, and previous immunofluorescence studies of pheromone receptors (11Hicke L. Zanolari B. Pypaert M. Rohrer J. Riezman H. Mol. Biol. Cell. 1997; 8: 13-31Crossref PubMed Scopus (92) Google Scholar, 14Piper R.C. Cooper A.A. Yang H. Stevens T.H. J. Cell Biol. 1995; 131: 603-617Crossref PubMed Scopus (340) Google Scholar, 15Rieder S.E. Banta L.M. Kohrer K. McCaffery J.M. Emr S.D. Mol. Biol. Cell. 1996; 7: 985-999Crossref PubMed Scopus (239) Google Scholar), indicate that small vesicles containing Ste2-GFP are probably early endosomes, and the larger vesicles are late endosome/prevacuolar compartments. Although Ste2-GFP was expected to be associated with the plasma membrane, cell surface projections, and endocytic compartments, it was surprising that Ste2-GFP-derived fluorescence was also present in the lumen of the vacuole (Fig. 2 A, top row). GFP accumulated in the vacuole of most cells, although the level differed significantly from cell to cell (also see Fig. 3, top row), possibly reflecting different extents of basal internalization of Ste2-GFP, rates of GFP degradation, or plasmid copy number. Localization to the vacuole lumen was confirmed by co-staining with CMAC. Accumulation of Ste2-GFP or its degradation products in the vacuole lumen was unexpected because the GFP domain was fused to the cytoplasmic tail of the receptor, which should remain cytoplasmically disposed if endocytic trafficking to the vacuole exclusively involves fusion events between unilamellar donor and acceptor membranes. The potential significance of this observation with respect to the mechanisms of endocytic transport is presented under "Discussion." To identify gene products that participate in endocytic trafficking of α-factor receptors, we used a visual screen to isolate mutants in which the intracellular distribution of Ste2-GFP was abnormal. Transposon insertion mutagenesis was used because insertion sites are easily determined with the aid of the completed yeast genome sequence. However, this method is likely to reveal only a subset of the genes involved in endocytic trafficking because transposon insertions occur at nonrandom sites, insertions into essential genes cannot be recovered, and subtle but significant differences in receptor localization could be overlooked in a visual screen. Mutants were identified in the following way. Pools of a transposon-mutagenized yeast DNA library were transformed into aste2 null mutant (JE115) expressing Ste2-GFP from a centromeric plasmid; this eliminates the possibility that localization of Ste2-GFP could be affected by the presence of untagged receptors expressed from the chromosome. Transformants were screened individually in the absence and presence of pheromone for alterations in receptor distribution. Two transformants displayed a similar aberrant pattern of Ste2-GFP localization, either in the absence or presence of pheromone (data not shown). In contrast to wild type cells, these mutants displayed intense fluorescence in numerous punctate intracellular patches or vesicles (Fig. 3, second and fourth rows). Furthermore, both mutants possessed fragmented vacuoles (as revealed by CMAC staining). In contrast, when these mutants expressed endocytosis-defective receptors, Ste2Δtail-GFP, they lacked fluorescence in intracellular vesicles (data not shown), suggesting that the localization of full-length Ste2-GFP within these vesicles requires receptor endocytosis. Thus, the mutants displayed defects in vacuolar morphology and/or targeting of endocytic vesicles containing Ste2-GFP to vacuoles. Analysis of transposon insertion sites revealed that both mutants carried disruptions of the VAM3/PTH1 gene, which encodes a protein of 283 amino acids with similarity to yeast and mammalian syntaxins or t-SNAREs (20Darsow T. Rieder S.E. Emr S.D. J. Cell Biol. 1997; 138: 517-529Crossref PubMed Scopus (294) Google Scholar, 24Srivastava A. Jones E.W. Genetics. 1998; 148: 85-98Crossref PubMed Google Scholar, 29Nichols B.J. Ungermann C. Pelham H.R. Wickner W.T. Haas A. Nature. 1997; 387: 199-202Crossref PubMed Scopus (379) Google Scholar, 30Wada Y. Nakamura N. Ohsumi Y. Hirata A. J. Cell Sci. 1997; 110: 1299-1306Crossref PubMed Google Scholar). In one mutant (Fig. 3,second row) the insertion occurred at codon 112, near sequences encoding the second predicted coiled-coil domain of Vam3; this was termed the vam3–100 allele. In the second mutant (Fig. 3, fourth row) the insertion occurred at codon 268, within sequences encoding the transmembrane domain of Vam3; this was termed the vam3–101 allele. These mutations appeared to cause the mutant phenotypes because introduction of the wild type VAM3 gene on a centromeric plasmid corrected the defects in Ste2-GFP localization and vacuolar morphology (Fig. 3, third and fifth rows). Interestingly, however, the wild type VAM3 plasmid only partially corrected the phenotype of the vam3–101 mutant, because some cells in the population exhibited a wild type distribution of Ste2-GFP and intact vacuoles whereas others had a mutant phenotype (Fig. 3,fifth row). Thus, a mutation truncating Vam3 near its transmembrane domain may be partially dominant-negative. Similar truncations of other SNARE proteins are dominant-negative (31Burd C.G. Peterson M. Cowles C.R. Emr S.D. Mol. Biol. Cell. 1997; 8: 1089-1104Crossref PubMed Scopus (139) Google Scholar, 32Rossi G. Salminen A. Rice L.M. Brunger A.T. Brennwald P. J. Biol. Chem. 1997; 272: 16610-16617Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar). Several observations suggested that thevam3–100 and vam3–101 mutations preserve receptor expression, signaling and internalization. First, wild type cells and vam3–100 mutants expressed receptors having similar agonist binding affinities (4.9 and 6.2 nm, respectively) and cell surface expression levels (B max = 14,000 and 15,000 sites/cell, respectively) (Fig. 1 A). Second, vam3–100 andvam3–101 mutants expressing Ste2-GFP responded to pheromone with efficiencies similar to wild type cells expressing Ste2-GFP (Fig.1 B) and mated with normal efficiency (data not
Referência(s)