QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online September 2, 2003
doi: 10.1242/10.1242/jcs.00751

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend 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 Yeo, S. C. L. Articles by Munn, A. L. Search for Related Content PubMed PubMed Citation Articles by Yeo, S. C. L. Articles by Munn, A. L.

Vps20p and Vta1p interact with Vps4p and function in multivesicular body sorting and endosomal transport in Saccharomyces cerevisiae

Sebastian C. L. Yeo^1,, Linghui Xu^1,, Jihui Ren^2,*,, Victoria J. Boulton¹, Mahendra D. Wagle^1,, Cong Liu^1,, Gang Ren^1,, Peisze Wong¹, Regina Zahn^1,¶, Piriya Sasajala^1,**, Hongyuan Yang², Robert C. Piper³ and Alan L. Munn^1,2,4,,

¹ Institute of Molecular and Cell Biology, The National University of Singapore, Singapore, 117609, Singapore
² Department of Biochemistry, Faculty of Medicine, The National University of Singapore, Singapore, 119260, Singapore
³ Department of Physiology and Biophysics, University of Iowa, Iowa City, 52242, IA, USA
⁴ Institute for Molecular Bioscience, University of Queensland, Brisbane, QLD, 4072, Australia

View larger version (33K): [in a new window]   Fig. 1. Vps4p interacts with Vps20p and Vta1p. (A) A schematic showing the predicted domain structure of Vps20p, Vta1p and Vps4p, the fragments of Vps20p and Vta1p encoded by the two-hybrid library plasmids pB42AD-VPS20 (pAM349) and pB42AD-VTA1 (pAM398), and the various constructs used to map the domains that interact. Cross-hashed boxes represent domains predicted to have a high propensity (0.5) to form coiled-coil structure as assessed using the COILS algorithm (Lupas et al., 1991). Numbers refer to amino acid residues. Library plasmids are depicted using solid lines and constructs used for mapping using broken lines. (B) Yeast two-hybrid interactions of Vps4p with Vps20p and Vta1p. Yeast strain EGY48 carrying the reporter plasmid p8op-LacZ together with either pLexA vector alone or pLexA-VPS4 (pAM333), and either pB42AD vector alone, pB42AD-VPS20 or pB42AD-VTA1 were plated on SD complete medium lacking uracil, histidine and tryptophan to select for the p8op-LacZ, pLexA-based and pB42AD-based plasmids, respectively. After the strains had grown sufficiently they were replica plated to synthetic galactose/raffinose (SG) complete medium containing X-gal to test expression of the lacZ two-hybrid interaction reporter gene. The plates were photographed after 4 days at 30°C. Shown are patches representing four independent transformants for each plasmid combination (a-f). (C) Quantification of yeast two-hybrid interactions. Each of the strains in B was assayed for ß-galactosidase activity as a measure of the strength of two-hybrid interaction. Activities (in Miller units) represent the means obtained from assaying three independent transformants. Error bars, +/-s.e.m.

View larger version (21K): [in a new window]   Fig. 2. Vps4p in yeast lysates associates with Vps20p and Vta1p and association is ATP-independent. Vps20p and Vta1p were expressed as GST fusions in E. coli and purified on glutathione-agarose beads. Vps4p, either tagged with GFP or without tag, was expressed from a centromeric plasmid in vps4 (RH2906). Lysates were prepared from both strains and a 100,000 g supernatant (S3) was supplemented with 20 mM MgCl2 and incubated with beads bearing GST only or GST-Vps20p (A), or with beads bearing GST only or GST-Vta1p (B) with or without pretreatment of the lysates with apyrase to deplete endogenous ATP. Unbound proteins (unbound) were recovered in the supernatants. After washing the beads, the specifically bound proteins (bound) were eluted by heating in Laemmli sample buffer. Proteins in both bound and unbound samples were resolved by SDS-PAGE, and transferred to PVDF membranes. Vps4p-GFP was detected by immunoblotting with a GFP-specific polyclonal antiserum. In each set of experiments the exposure times for the gels containing bound and unbound samples were identical.

View larger version (31K): [in a new window]   Fig. 3. Vps4p directly binds Vps20p and Vta1p in vitro and binding is ATP-independent. Wild-type Vps4p, and both ATP hydrolysis mutant (E233Q) and ATP binding mutant (K179A) forms of Vps4p were tagged at the C-terminus with 6HIS and expressed in E. coli. Each protein was affinity purified using the 6HIS tag and incubated with beads bearing GST-Vps20p, GST-Vta1p or GST only in the presence or absence of added ATP. Unbound proteins (unbound) were recovered in the supernatants. After washing the beads, the specifically bound proteins (bound) were eluted by heating in Laemmli sample buffer. Proteins in both bound and unbound samples were resolved by SDS-PAGE, and transferred to PVDF membranes. Wild-type and mutant forms of Vps4p-6HIS were detected by immunoblotting with a pentaHIS-specific monoclonal antiserum. In each set of experiments the exposure times for the gels containing bound and unbound samples were identical.

View larger version (21K): [in a new window]   Fig. 4. Loss of Vps20p or Vta1p causes only slightly altered kinetics of -factor internalisation. Wild-type (RH1800) and vps20 (AMY174), and wild-type (AMY165) and vta1 (AMY162) cells were grown to early exponential phase and assayed for [35S]-factor internalisation at 30°C. After addition of the [35S]-factor, samples were taken in duplicate at various time points and washed in either phosphate buffer pH 6 (removes unbound [35S]-factor) or citrate buffer pH 1 (removes bound but non-internalised [35S]-factor). Shown is the internalised -factor as a percentage of the bound -factor at each time point.

View larger version (36K): [in a new window]   Fig. 5. Loss of Vps20p or Vta1p causes defects in -factor degradation. Wild-type (RH1800) and vps20 (AMY174) (A), and wild-type (AMY165) and vta1 (AMY162) (B) cells were grown to early exponential phase and assayed for [35S]-factor transport to the vacuole and degradation at 30°C. [35S]-factor was prebound to the cells on ice and then the cells were harvested at 4°C and resuspended in fresh YPUAD and incubated at 30°C. At the time points shown, samples were taken in duplicate and washed in either phosphate buffer pH6 (removes unbound [35S]-factor) or citrate buffer pH 1 (removes bound non-internalised [35S]-factor). Lysates were prepared from each sample of cells and subjected to thin layer chromatography to separate intact (i) and degraded (d) [35S]-factor, which were visualised by fluorography at -80°C. 6, washed in pH 6 buffer; 1, washed in pH 1 buffer; o, origin where samples were loaded.

View larger version (72K): [in a new window]   Fig. 6. Loss of Vps20p or Vta1p causes defects in Ste3p localisation to the vacuole. Wild-type (SF838-9D), vps20 (SF838-9Dvpl10) and vta1 (PLY3046) cells were transformed with a low-copy plasmid encoding Ste3-GFP in combination with an empty URA3 containing centromeric plasmid or centromeric plasmid containing the wild-type VTA1 or VPS20 gene as indicated. The localisation of Ste3-GFP was then assessed by fluorescence microscopy together with DIC imaging to identify yeast vacuoles. Cells were resuspended in 1% sodium azide, 1% sodium fluoride, 100 mM Tris pH 8.0 prior to fluorescence and DIC microscopy. Bar, 5 µm.

View larger version (49K): [in a new window]   Fig. 7. Loss of Vps20p or Vta1p causes secretion of Golgi-modified p2 CPY precursor and degradation of Vps10p. Wild-type (SF838-9D), vps20 (SF838-9Dvpl10) and vta1 (PLY3046) cells were converted to PEP4 as described in Materials and Methods. They were pulse-labelled with [35S]methionine/cysteine for 10 minutes at 30°C and then chased with excess unlabelled methionine/cysteine at the same temperature. Samples were taken at 0 or 60 minutes of chase, and further membrane transport was stopped by addition of sodium azide and sodium fluoride to 20 mM. The cells in each sample were converted to spheroplasts and fractionated into intracellular (I) and extracellular (E) fractions. CPY was immunoprecipitated from half the intracellular (I) and extracellular (E) fractions of the 60-minute chase samples (A). Vps10p was immunoprecipitated from the intracellular (I) fractions of the 0- and 60-minute chase samples (B). Indicated are the mature (mCPY) and Golgi-modified (p2CPY) forms of CPY, and both full-length Vps10p (Vps10p) and the lower protease-cleaved band of Vps10p (*).

View larger version (90K): [in a new window]   Fig. 8. Vps20p and Vta1p are required for multivesicular body sorting and LY accumulation in the vacuole. The two MVB reporter proteins, Fth1-GFP-Ub (left) and Sna3-GFP (centre) were localised in wild-type (SF838-9D), vps20 (SF838-9Dvpl10) and vta1 (PLY3046) cells. For visualisation of Fth1-GFP-Ub, cells were grown in 100 µM of the iron chelator BPS for 6 hours prior to microscopy. Cells were resuspended in 1% sodium azide, 1% sodium fluoride, 100 mM Tris pH 8.0 prior to microscopy. Endocytosis of LY (right) was measured by incubating cells in media containing LY for 60 minutes at 30°C, washing and viewing the cells by DIC and fluorescence microscopy. Bar, 5 µm.

View larger version (118K): [in a new window]   Fig. 9. Vps20p and Vta1p are required for MVB lipid sorting. Sorting of the lipid dye NBD-PC was followed in wild-type (SF838-9D), vps20 (SF838-9Dvpl10) and vta1 (PLY3046) cells. Cells labelled at 30°C with FM4-64 (2 µM) in potassium phosphate-buffered YPUAD (pH 7.0) for 20 minutes followed by incubation with NBD-PC (100 µM) for an additional 20 minutes. Cells were washed and resuspended in SD media and incubated for an additional 30 minutes at 30°C prior to fluorescence microscopy. Cells were resuspended in 1% sodium azide, 1% sodium fluoride, 100 mM Tris pH 8.0 prior to fluorescence and DIC microscopy. Bar, 5 µm.

View larger version (159K): [in a new window]   Fig. 10. Vps20p and Vta1p are required for transport of FM4-64 from late endosomes to the vacuole. Wild-type (SF838-9D), vps20 (SF838-9Dvpl10) and vta1 (PLY3046) cells were grown to early exponential phase and then incubated with 2 µM FM4-64 at 0°C for 30 minutes. Cells were washed in ice-cold YPUAD and resuspended in YPUAD at 30°C without FM4-64 (0'). Cell aliquots were removed at the indicated times after shift to 30°C (5', 10', etc.), washed in 1% sodium azide, 1% sodium fluoride, 100 mM Tris pH 8.0, and viewed by fluorescence and DIC microscopy. Bar, 5 µm.