![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 18, Issue 2, 646-657, February 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biochemistry and Molecular Biology, Mayo Clinic College of Medicine, Rochester, MN 55905
Submitted July 17, 2006;
Revised October 23, 2006;
Accepted November 22, 2006
Monitoring Editor: Sandra Lemmon
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Babst, 2005). MVBs form when the limiting membrane of the late endosome invaginates and buds into the lumen of the organelle, actively selecting a subset of the proteins from the limiting membrane in the process (Gorden et al., 1978; Haigler et al., 1979). The intralumenal vesicles within the MVB and their contents are exposed to hydrolases within the lysosome as a consequence of heterotypic fusion of these organelles. By contrast, proteins within the limiting membrane of the MVB are delivered to the limiting membrane of the lysosome/vacuole after heterotypic fusion (Odorizzi et al., 1998).
Sorting of cargo into the MVB pathway and heterotypic fusion with the lysosme/vacuole leads to delivery into the hydrolytic lumen of the organelle, and therefore entry into this pathway must be highly regulated. Ubiquitination is the best-characterized cis-acting signal mediating entry into the MVB pathway (for reviews, see Katzmann et al., 2002; Hicke and Dunn, 2003; Raiborg et al., 2003). Studies in organisms ranging from the yeast Saccharomyces cerevisiae, to mammalian cells have demonstrated that ubiquitin modification of a variety of MVB cargoes is a requisite for entry into this pathway, and ubiquitin modification of endosomal membrane proteins that are not normally MVB cargoes is sufficient to target them into this pathway (Katzmann et al., 2001; Reggiori and Pelham, 2001; Urbanowski and Piper, 2001; Raiborg et al., 2002). Ubiquitin-independent cargoes of the MVB pathway have also been described, but relevant signals for inclusion have not been precisely defined. In yeast, ubiquitin modification of the Sna3 protein has been reported to be dispensable for Sna3 entry into the MVB pathway (Reggiori and Pelham, 2001). In the case of the mammalian melanosomal protein Pmel17, the relevant sorting information for targeting to intralumenal vesicles resides within the lumenal domain of the protein (Theos et al., 2006), suggesting a novel mechanism by which this cargo is selected.
The sorting of all yeast MVB cargoes requires the function of the class E vacuolar protein sorting (Vps) proteins as the trans-acting machinery (Odorizzi et al., 1998). This machinery has been conserved through evolution from yeast to humans (for reviews, see Katzmann et al., 2002; Babst, 2005) and has been demonstrated to mediate the budding of retroviruses such as human immunodeficiency virus-1 (Morita and Sundquist, 2004). The class E Vps proteins recognize ubiquitinated MVB cargo proteins and actively sort them into the MVB pathway (for reviews, see Katzmann et al., 2002; Raiborg et al., 2003; Babst, 2005). In class E vps mutants, MVBs fail to form and therefore MVB cargoes fail to reach the vacuolar lumen. (Raymond et al., 1992; Odorizzi et al., 1998). Many of the class E Vps proteins form the endosmal sorting complexes required for transport (ESCRTs). ESCRT-I, -II, and -III, together with additional class E Vps proteins such as Vps27/Hrs and Vps4/SKD1, transiently associate with the endosomal membrane to execute the MVB sorting reaction (Katzmann et al., 2001; Babst et al., 2002a,b; Bache et al., 2003; Bilodeau et al., 2003; Katzmann et al., 2003; von Schwedler et al., 2003; Bowers et al., 2004). The class E proteins Vps23/Tsg101, Vps28, and Vps37 were described previously as subunits of the 350-kDa ESCRT-I complex, which plays a role in the recognition of ubiquitin-modified MVB cargoes through the ubiquitin E2 variant (UEV) domain of Vps23/Tsg101 (Babst et al., 2000; Bishop and Woodman, 2001; Katzmann et al., 2001; Bache et al., 2004; Teo et al., 2004). Recent structural determination of the ESCRT-I core suggests that Vps23, Vps28, and Vps37 may be present in equal molar ratios within this complex (Kostelansky et al., 2006; Teo et al., 2006); however, it is unclear how these subunits of 43, 28, and 25 kDa alone would comprise a complex with an apparent molecular mass of 350 kDa. ESCRT-I recruitment to the endosomal membrane is dependent upon its association with Vps27/Hrs, which seems to dictate site selection for the MVB sorting reaction by virtue of its ability to bind both the endosomally enriched lipid phosphatidylinositol-3-phosphate [PtdIns(3)P] as well as ubiquitin modified MVB cargo through ubiquitin interacting motifs (Bache et al., 2003; Bilodeau et al., 2003; Katzmann et al., 2003).
Cellular components involved in mediating ubiquitin-dependent entry into the MVB pathway have begun to be elucidated. In yeast, deletion of class E vps genes results in a dramatic missorting of all MVB cargoes; however, mutants specifically defective for the delivery of ubiquitin-dependent MVB cargoes retain the ability to sort ubiquitin-independent MVB cargoes, including Sna3 (Bilodeau et al., 2002; Katzmann et al., 2004). This suggests additional mechanisms of cargo selection by this pathway exist. To gain insight into this process, we performed a genetic screen for loci affecting the MVB sorting of Sna3, a putative ubiquitin-independent MVB cargo (Reggiori and Pelham, 2001; Katzmann et al., 2004). This screen identified Mvb12 as a critical factor for sorting a subset of MVB cargoes, including Sna3 and Ste3. However, trafficking of other MVB cargoes such as carboxypepsidase S (CPS) and Ste2 is largely unaffected in mvb12
cells, as is general endosomal function. Several lines of evidence indicate that Mvb12 impacts MVB cargo sorting as a novel subunit of the previously described ESCRT-I. The distinct phenotypes of mvb12 compared with deletion of other ESCRT-I subunits suggest that Mvb12 is a modulator of the Vps23/28/37 ESCRT-I core machinery's ability to sort specific MVB cargoes. These results identify Mvb12 as a novel subunit of ESCRT-I that modulates the cargo sorting activity of this complex.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) were used to introduce Sna3-GFP into the collection of viable deletion mutants (Tong et al., 2001) Double deletion mutants in which both deletions are marked with HIS3 were generated by performing crosses between opposite mating types containing the relevant deletions, followed by sporulation, tetrad dissection, and verification of genotype.
|
). Sna3KallR-GFP was constructed through mutagenesis of pRS416-Sna3-GFP (GeneTailor; Invitrogen). GFP-CPS and Ste2-GFP have been described previously (Odorizzi et al., 1998). Ste3-GFP and Ste3-GFP-Ub were graciously provided by Robert Piper (University of Iowa, Iowa City, IA). DsRed FYVE was described previously as an endosomal marker (Katzmann et al., 2003). pMB103 (pRS416-Vps4E233Q) expresses a dominant-negative form of Vps4 that has been described previously (Babst et al., 1997). The MVB12 open reading frame (ORF) was amplified from yeast genomic DNA and cloned using BamHI and XhoI sites into the pET28b (Novagen, Madison, WI) and pnTAP416 vectors digested with SalI and BamHI, for bacterial expression of His6-Mvb12 and yeast expression of TAP-Mvb12, respectively. To create pGPD414 HA-ubiquitin, pGDP416 HA-ubiquitin (Davies et al., 2003) was digested with SacI and KpnI and subcloned into pRS414. The TAP tag was cloned into the EcoRI site of pGPD416 (Mumberg et al., 1995) to create the pnTAP416 vector. All PCR-based plasmids were sequenced to ensure that no aberrant mutations were present.
Microscopy
Fluorescence microscopy was performed on live cells in minimal media, by using a Nikon fluorescence microscope with fluorescein isothiocyanate, rhodamine, green fluorescent protein (GFP), and DsRed filters and a digital camera (Coolsnap HQ; Photometrix, Melbourne, Australia). Images were deconvolved with Delta Vision software (Applied Precision, Seattle, WA). FM4-64 labeling was performed as described in Vida and Emr (1995). Kinetic analysis of FM4-64 uptake was performed by labeling cells on ice for 15 min, followed by chasing at room temperature.
Biochemical Analyses
Gel filtration analyses were performed as described previously (Katzmann et al., 2001). Bacterial expression of His6-Mvb12 was induced in the HMS174 DE3 strain at 37°C for 3.5 h with 0.5 mM isopropyl 
-D-thiogalactoside. Crude lysate subjected to a 100,000 x g clearing spin was used for gel filtration. Tagged proteins were visualized using monoclonal antibodies anti-HA.11 (Covance, Princeton, NJ), monoclonal anti-GFP AV JL-8 (BD Biosciences, San Jose, CA), or penta-His (QIAGEN, Valencia, CA). Pulse-chase analysis of CPS, carboxypeptidase Y (CPY), Sna3-GFP, and Sna3KallR-GFP was performed as described previously (Babst et al., 2002a), and quantitation was performed using phosphorimaging screens and a Storm 840 system (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom). Protein A purification using TAP-Mvb12 was performed as described in Azmi et al. (2006), with the exception that 15 OD600 of cells were lysed in phosphate-buffered saline. Proteins were visualized with polyclonal anti-Vps23 and anti-Vps28 (Katzmann et al., 2001). Ste3 immunoprecipitation was performed essentially as described in Katzmann et al. (2001), with all buffers containing 5 mM N-ethylmaleimide and incubations performed at 50°C. Yeast expressing Ste3-GFP and hemagglutinin (HA)-ubiquitin were trichloroacetic acid precipitated, processed, and lysed in urea cracking buffer with glass beads. Immunoprecipitation was performed with monoclonal anti-GFP AV JL-8 (BD Biosciences), and samples were subjected to SDS-PAGE and Western blotting. Ste3 was detected with anti-GFP, and the ubiquitination status was determined with monoclonal anti-HA.11 (Covance) to recognize HA-ubiquitin.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
); thus, trafficking of Sna3 fused to GFP (Sna3-GFP) was analyzed to identify novel cargo recognition factors and components of the MVB pathway. Methodologies described previously to perform genome-wide synthetic genetic interaction studies (Tong et al., 2001) were used to introduce Sna3-GFP into the collection of viable yeast deletion strains and fluorescence microscopy was used to analyze subcellular localization of Sna3-GFP in 4570 mutants. Whereas the total number of mutants within the collection is slightly higher than the number analyzed, mutants that failed to come through the selection procedure and were previously known to function in protein trafficking or seemed irrelevant were not pursued. For this reason, as well as essential genes not being included in this collection and the collection itself containing errors, this analysis is not intended to represent an exhaustive survey of all gene products required for the function of the MVB pathway. The majority of strains within the collection displayed wild-type Sna3-GFP distribution, indistinguishable from wild type wherein fluorescence is localized to the lumen of the vacuole (Figure 1; data not shown). However, a large number of previously known vps mutants were identified, validating the experimental procedure (Supplemental Figure 1). In addition to previously known trafficking factors, this approach identified two mutants that displayed MVB sorting defects distinct from previously characterized class E vps mutants: vta1 (Azmi et al., 2006) and deletion of YGR206w (Figure 1 and Supplemental Figure 1). YGR206w encodes a 12-kDa protein that plays a role in MVB sorting and is referred to as MVB12.
|
by using the deletion collection, the MVB12 gene was deleted in our standard genetic background (SEY6210) and MVB sorting phenotypes were examined in greater detail. Cells were also stained with the fluorescent vacuolar dye FM4-64 to enable visualization of the limiting membrane of the vacuole. As in the BY4742 genetic background, analysis of Sna3-GFP distribution revealed an intermediate MVB sorting phenotype, with a portion mislocalized to perivacuolar structures (indicated by arrows in Figure 2). This Sna3-GFP localization pattern is distinct from both wild-type cells, wherein Sna3-GFP is largely within the vacuole lumen, or a class E vps mutant (vps23), in which Sna3-GFP localized predominantly to the aberrant class E compartment (Figure 2). Moreover, the colocalization of Sna3-GFP and FM4-64 apparent in vps23 cells was absent in the mvb12 and wild-type cells. This suggests that the compartments to which Sna3-GFP localized in an mvb12 mutant are distinct from those seen in a standard class E vps mutant. This intermediate phenotype is consistent with the partial Sna3-GFP sorting defects observed in the mvb12 strain in the BY4742 genetic background, although less severe, and defects in both strains could be corrected by transforming a MVB12 plasmid (Figure 3; data not shown).
|
|
cells, kinetic analysis of Sna3-GFP sorting was performed by pulse-chase analysis (Figure 3A and Supplemental Figure 2A). Sna3-GFP is degraded within the vacuole lumen subsequent to sorting into the MVB pathway; hence, its turnover serves as a measure of its MVB sorting (Oestreich et al., 2007). Consistent with microscopic analysis, loss of Mvb12 resulted in a decreased rate of turnover of Sna3-GFP (Figure 3A and Supplemental Figure 2A). Importantly, the presence of Sna3-GFP in the vacuolar lumen in the mvb12 cells, as well as the eventual maturation of Sna3-GFP, indicated that Mvb12 is not absolutely required for function of the MVB pathway but that Mvb12 does play a role in Sna3 MVB sorting. This is distinct from the class E vps mutant vps23 wherein Sna3-GFP is not observed within the vacuolar lumen and its turnover is stabilized (Figure 2 and Supplemental Figure 2A).
The role of Mvb12 in MVB sorting was explored by analyzing the trafficking of additional MVB cargoes in the mvb12 strain. CPS is a well-characterized MVB cargo in yeast (Odorizzi et al., 1998). Ubiquitin modification of CPS by the ubiquitin ligase Rsp5 confers ubiquitin-dependent MVB sorting of CPS in a manner dependent upon ESCRTs (Katzmann et al., 2001, 2004); hence, CPS sorting serves as an indicator of ubiquitin-dependent MVB sorting. Localization of a GFP-CPS reporter was examined in wild type, mvb12, and the class E mutant vps23 strains. Whereas the mvb12 strain exhibited pronounced Sna3-GFP localization to punctate structures, GFP-CPS localization was only subtly perturbed, displaying minor missorting to the limiting membrane of the vacuole and no obvious class E compartment in both genetic backgrounds (Figures 1 and 2). This can be readily observed in the merged images, wherein the vacuolar limiting membrane looks yellow in both mvb12 and vps23 cells (Figure 2). Kinetic analysis of CPS maturation in the mvb12 strain (JPY22) yielded a consistent result with only a modest delay evident (Figure 3B). Maturation of the soluble vacuolar protease CPY is also only partially impeded in the mvb12 strain (Figure 3C). Furthermore, the mvb12 strain displayed very low levels of secreted p2CPY similar to wild-type cells, in contrast to the large portion secreted in the class E vps mutant vps23 (Figure 3D). The kinetic delay of CPY processing observed for mvb12 cells in Figure 3C can also be observed in Figure 3D wherein p2CPY is present in the intracellular fraction at 20 min, but it has been converted to mature (and not secreted) by 40 min. In contrast, the vps23 strain exhibited extensive limiting membrane and class E compartment localization for both GFP-CPS and Sna3-GFP (Figure 2) and defects both in CPS and CPY maturation have been observed previously (Figure 3 and Supplemental Figure 2; Babst et al., 2000). As has been observed previously, the class E vps mutants processed CPS to an aberrant form that migrates distinctly from wild-type mature (m)CPS. However, mvb12 cells did not display this misprocessed form; instead, mvb12 cells displayed mCPS indistinguishable from wild-type cells (Babst et al., 2002a; Supplemental Figure 2D). These results indicate that mvb12 displays distinct phenotypes from previously characterized class E vps mutants.
To further explore these distinctions, endocytic trafficking of the G protein-coupled receptors Ste2 and Ste3 was examined in the mvb12 strains of the appropriate mating type. Ubiquitin modification of Ste2 has been demonstrated to play a critical role in its internalization from the cell surface and subsequent delivery into the MVB pathway, resulting in delivery to the vacuolar lumen (Terrell et al., 1998). Ste3, by contrast, seems to undergo recycling between the plasma membrane and an endosomal compartment until ubiquitination removes it from the recycling loop by targeting it into the MVB pathway (Chen and Davis, 2002). The intracellular trafficking itineraries of Ste2 and Ste3 seem to be distinct; thus, both cargoes were analyzed in the mvb12 background. Analysis of Ste2-GFP localization in the mvb12 strain indicated that Mvb12 does not play a crucial role in this process, because no difference was observed compared with wild-type cells (Figure 4A). By contrast, trafficking of Ste3-GFP to the vacuolar lumen is severely impaired in the mvb12 strain, displaying vacuolar limiting membrane and perivacuolar punctate structure localization (Figure 4, A and C). Again, whereas mvb12 cells display Ste3-GFP localization to perivacuolar compartments, these cells do not colocalize with FM4-64 to the same degree observed in the class E vps23 cells. However, fusion of ubiquitin to Ste3-GFP (Ste3-GFP-Ub) can bypass the requirement for Mvb12 in Ste3 trafficking (Figure 4A). These results suggest that loss of Mvb12 negatively impacts the MVB sorting of Sna3 and Ste3 to a greater degree than sorting of CPS, Ste2, or Ste3-Ub. One explanation for the differential defects on Ste3-GFP and Ste3-GFP-Ub in mvb12 cells could be a defect in ubiquitin modification of Ste3. This was directly analyzed by immunoprecipitation of Ste3-GFP from wild-type, mvb12, and vps23 cells expressing HA-ubiquitin, followed by Western blotting with either anti-GFP or anti-HA (Figure 4B). Although equivalent numbers of cells were used in for each immunoprecipitation, Western blotting with anti-GFP consistently revealed higher levels of Ste3-GFP in both mvb12 cells and vps23 cells compared with wild-type cells, consistent with a defect in its degradation in these backgrounds (Figure 4B). Regardless, mvb12 cells display an increased level of ubiquitinated Ste3-GFP compared with both wild-type and vps23 cells (Figure 4B), indicating that mvb12 cells are not defective for the ubiquitination of Ste3. Whereas the mechanism by which Mvb12 functions is not clear from these studies, these results indicate that that some MVB cargoes are more sensitive to Mvb12 function than others. This characteristic is distinct from prototypical class E vps mutants and suggests that Mvb12 may be functioning as an accessory factor involved in entry of a subset of cargoes into the MVB pathway.
|
strain was examined further. Sna3 sorting has been suggested to occur by a ubiquitin-independent process (Reggiori and Pelham, 2001). However, more recent observations indicate that Sna3 is ubiquitylated (Peng et al., 2003) and that both ubiquitin-dependent and ubiquitin-independent pathways can mediate Sna3 trafficking (Oestreich et al., 2007). Analysis of CPS, Ste2-GFP, and Ste3-GFP-Ub trafficking indicated that Mvb12 does not play an essential role in the general entry of ubiquitin-dependent MVB cargoes into the MVB pathway. Thus, we hypothesized that Mvb12 may contribute more to the sorting process that does not require Sna3 ubiquitination. To examine this possibility, trafficking of Sna3-GFP with all lysines mutated to arginine (Sna3KallR-GFP) was examined in the mvb12 strain. Whereas Sna3KallR-GFP sorted in a manner indistinguishable from Sna3-GFP in the wild-type strain (SEY6210 and BY4742), mutation of the lysine residues resulted in enhanced Sna3KallR-GFP localization to peri-vacuolar punctate structures upon deletion of MVB12 compared with Sna3-GFP (Figures 1 and 2) and a delay in Sna3KallR-GFP cleavage (Figure 3, A and B, and Supplemental Figure 2, A and B). These results are consistent with Sna3 trafficking into the MVB via alternative pathways involving either ubiquitination of Sna3 or a process that does not require ubiquitination of Sna3 (Oestreich et al., 2007), with loss of Mvb12 having a greater impact on the latter. The explanation for this observation is not clear at present, but it is interesting to note that Ste3 may also use this pathway. Alternatively, both Mvb12 and Sna3-ubiquitination may affect the kinetics of Sna3 entry into the MVB pathway and thus yield a synthetic phenotype when both are compromised. Although further dissection of Sna3 trafficking and Mvb12 function will be required to resolve these models, these observations are consistent with Mvb12 functioning to facilitate the sorting of a subset of MVB cargoes.
Endosomal Localization of Mvb12 Is Dependent upon the ESCRT-I Subunit Vp s23
Loss of Mvb12 function conferred several phenotypes associated with endosome-to-vacuole protein sorting defects, including partial defects in MVB sorting. Subcellular localization was addressed by integrating the GFP coding sequence in-frame with the chromosomal copy of MVB12, resulting in a functional Mvb12-GFP chimera (Figure 3, A–C). Visualization of cells expressing Mvb12-GFP revealed both cytoplasmic and intracellular punctate structures that colocalize with the endosomal marker DsRed-FYVE but not the Golgi marker Sec7-RFP (Figure 5). To further characterize these compartments, kinetic analysis of FM4-64 uptake was analyzed in cells expressing Mvb12-GFP (Figure 6). Colocalization of Mvb12-GFP with FM4-64 structures at early time points (1–3 min) was minimal, but increased from 5 to 7 min (Figure 6). At later time points (15–25 min), the discreet patterns of FM4-64 and Mvb12-GFP indicate that the Mvb12-GFP endocytic structures are largely perivacuolar. Colocalization with endosomes places Mvb12-GFP at a location where the MVB sorting reaction is occurring, consistent with the observed phenotypes upon its deletion. This localization is also consistent with results obtained in a previous proteome-wide localization analysis (Huh et al., 2003).
|
|
, hse1 and the ESCRT-III-like mutants (fti1/vps46/did2 and vps60/mos10), Mvb12-GFP distribution seemed indistinguishable from wild-type cells (Supplemental Figure 3A). This suggests that these factors play no role in the endosomal association of Mvb12. However, loss of ESCRT-II (vps25), ESCRT-III (vps20, vps24, or snf7) or the AAA-ATPase Vps4 (vps4) resulted in enhanced endosomal localization of Mvb12-GFP (Figure 7A and Supplemental Figure 3B). These results indicate that these factors are not required for the recruitment of Mvb12 to the endosomal membrane and are consistent with the idea that Mvb12 removal from the endosomal membrane requires the function of ESCRT-II, -III, and Vps4. In striking contrast, loss of Vps27 or the ESCRT-I subunits Vps23 and Vps37 resulted in a dramatic reduction in the amount of endosome-associated Mvb12-GFP (Figure 7A and Supplemental Figure 3A). To further address the redistribution of Mvb12 upon loss of the ESCRT-I subunit Vps23, a previously characterized dominant-negative form of Vps4 (Vps4E233Q) (Babst et al., 1997) was expressed in the vps23 strain. Expression of Vps4E233Q in SEY6210 resulted in the enhanced accumulation of Mvb12-GFP on endosomes as well as reduced cytoplasmic distribution, consistent with Mvb12 localization in the vps4 strain; however, this accumulation was blocked in the vps23 strain, suggesting that even transient Mvb12 membrane association is absent without Vps23 (Figure 7B). However, loss of the ESCRT-I subunit Vps28 conferred increased membrane association of Mvb12-GFP, as observed upon loss of ESCRT-II, -III, and Vps4 (Figure 7A and Supplemental Figure 3B). This effect phenocopies the redistribution of Vps23-GFP that has been observed in the vps28 strain (Li et al., 1999). These observations are consistent with Mvb12 membrane association requiring the function of Vps27 and the ESCRT-I subunits Vps23 and Vps37. ESCRT-I membrane association is dependent upon Vps27, but Vps27 membrane association is independent of ESCRT-I (Katzmann et al., 2003). Similarly, association of GFP-Vps27 with endosomes marked by DsRed-FYVE is not disrupted in mvb12 cells (Supplemental Figure 3C). Thus, Mvb12 behaves in a manner similar to the ESCRT-I subunit Vps23. These observations raised the possibility that Mvb12 is a novel subunit of ESCRT-I that facilitates its recognition of a subset of MVB cargoes.
|
; Katzmann et al., 2001). Monomeric Mvb12 would be predicted to behave as a species smaller than 45 kDa with either epitope and was not detected in yeast extracts (Figure 8A; data not shown), suggesting that it is stably associated with this higher molecular mass complex in vivo. Furthermore, bacterially expressed His6-Mvb12 behaves as a species of approximately 18 kDa when subjected to gel filtration (Figure 8A). Given the separation range of the Sephacryl S-300 gel filtration column, this observation is consistent with bacterially expressed Mvb12 behaving as a monomer. These results suggest that Mvb12 is stably associated with a higher molecular mass complex in yeast extracts.
|
, vps25 (ESCRT-II), snf7 or vps24 (ESCRT-III), or vps4 strains (Figure 8A). However, elimination of the ESCRT-I subunits (vps23, vps28, and vps37) reduced the apparent molecular mass to 
200 kDa in Vps23 or Vps28 and dramatically destabilized the Mvb12-GFP protein to the point that it was undetectable with loss of Vps37 (Figure 8A; data not shown). This result indicated that the formation of the Mvb12-containing complex depends on the function of the ESCRT-I complex. Because ESCRT-I is a 350-kDa complex, the simplest interpretation is that Mvb12 associates with ESCRT-I itself. Moreover, gel filtration analysis of Vps23 and Vps28 in extracts generated from the mvb12 strain revealed a reduction in the apparent molecular mass of ESCRT-I to less than 200 kDa (Figure 8B). For comparison, loss of Vps28 shifts the apparent molecular mass of Vps23 to approximately 200 kDa, whereas loss of Vps37 has an even greater effect on its apparent mass (Figure 8B). These observations support the conclusion that Mvb12 is a stable subunit of ESCRT-I.
To directly address the association of Mvb12 with ESCRT-I, isolation of a tandem affinity purification (TAP)-tagged form of Mvb12 was performed. TAP-Mvb12 was expressed in wild-type, vps23, and vps37 strains, purified from cleared lysates under native conditions using IgG-Sepharose, and the isolated material was subjected to Western analysis with anti-Vps23 and anti-Vps28 antisera (Figure 8C). The ESCRT-I components Vps23 and Vps28 copurified with TAP-Mvb12 from the wild-type lysate (lane 6), but they did not copurify with the TAP-tag alone (lane 5). This result indicated that Mvb12 is physically associated with ESCRT-I. Elimination of Vps23 did not reduce TAP-Mvb12 isolation of Vps28 (lane 7) nor did loss of Vps28 eliminate TAP-Mvb12 isolation of Vps23 (Supplemental Figure 4). However, isolation of the remaining ESCRT-I components with TAP-Mvb12 was compromised in the vps37 strain (Figure 8C and Supplemental Figure 4). Loss of another component of the MVB sorting machinery (vps4) did not affect the association between Mvb12 and ESCRT-I (Supplemental Figure 4), consistent with the gel filtration analysis of the Mvb12-containing complex in vps4 lysates. These results confirmed that Mvb12 is a novel component of ESCRT-I and are consistent with the isolation of Vps23 and Vps37 with Mvb12-TAP in the yeast proteome-wide analysis (Krogan et al., 2006). However, the phenotypes of mvb12 differ from the phenotypes of vps23, vps28 and vps37. These results lead us to propose that Mvb12 is modulating the ability of ESCRT-I to mediate MVB sorting of specific cargoes rather than being a core component of ESCRT-I.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Hicke and Dunn, 2003; Raiborg et al., 2003; Babst, 2005). ESCRT-I was first identified as a complex that is capable of interacting with ubiquitin-modified MVB cargo via the UEV domain in Vps23 (Katzmann et al., 2001). Subsequent work has revealed the presence of additional members of the class E Vps proteins that contain ubiquitin-interacting domains, including Vps27/Hrs, which participates in recruitment of ESCRT-I (Bilodeau et al., 2002; Shih et al., 2002; Bache et al., 2003; Bilodeau et al., 2003; Katzmann et al., 2003). Together, Vps27 and ESCRT-I seem to play a critical role in the selection of ubiquitin-modified MVB cargoes. The reason behind this apparently complicated mechanism of cargo recognition is not entirely obvious; however, a variety of explanations including site of action, degree/linkage of cargo ubiquitination, and even affinity have merits. Furthermore, cargoes that are capable of entering the MVB pathway independent of ubiquitin modification exist and may require additional mechanisms of recognition by the MVB sorting machinery.
To better understand the mechanisms of cargo selection by the MVB pathway we performed a genetic screen for factors impacting Sna3 sorting, identifying mvb12 in the process. Deletion of MVB12 confers cargo-specific defects in MVB sorting, a feature that makes it distinct from previously described class E vps deletion mutants that display unilateral defects in MVB cargo sorting. This feature is particularly interesting considering that our data indicate that Mvb12 is a member of the ESCRT-I complex. But given that loss of the core ESCRT-I subunits (Vps23, Vps28, and Vps37) confer significantly more severe phenotypes, Mvb12 is not critical for all functions executed by ESCRT-I. Based on these observations, we suggest that Mvb12 plays a role in assisting ESCRT-I during sorting of a subset of MVB cargoes.
Although further investigation is required to distinguish the mechanism by which this occurs, several possibilities exist that could explain this role in cargo-specific function. One potential model is that Mvb12 facilitates the binding of ESCRT-I to a subset of MVB cargoes, including Ste3 and Sna3, but it is dispensable for binding of ESCRT-I to CPS and Ste2. Although it is attractive to speculate that Mvb12 could directly bind to Ste3 and Sna3 and mediate their interaction with ESCRT-I, it is equally likely that Mvb12 is not involved in directly binding these cargoes but helps to generate the Ste3 and Sna3 binding site(s) within ESCRT-I indirectly. How would this model explain the suppression of Ste3 trafficking defects in the mvb12 strain by direct fusion to ubiquitin (Ste3-GFP-Ub)? Although Ste3 has been demonstrated to be a ubiquitin-dependent MVB cargo, there may be ways in which the ubiquitinated form of Ste3 is recognized by the Mvb12-containing ESCRT-I that are distinct from the ESCRT-I recognition of the Ste3-Ub fusion, possibly through differential recognition of ubiquitin itself. Further experimentation will be required to address the model that Mvb12 allows ESCRT-I binding, either directly or indirectly, to a subset of MVB cargoes.
A second model for the role of Mvb12 in the MVB sorting of Ste3 and Sna3 but not CPS or Ste2 involves the trafficking and MVB sorting of distinct cargoes via discrete endosomal compartments. Recent evidence in mammalian cells is consistent with the idea that there are distinct classes of MVBs being formed at distinct sites (White et al., 2006). Similar phenomenon may be occurring in yeast. In this second model, Ste3 and Sna3 would traffic to the vacuole via an endosomal compartment that is distinct from that transited by CPS and Ste2. Although ESCRT-I mediates MVB sorting at both compartments, Mvb12 may be required for ESCRT-I localization or function at the Ste3/Sna3 compartment but not the Ste2/CPS compartment. Direct fusion of ubiquitin to the carboxy terminus of Ste3 (Ste3-Ub) may alter Ste3-Ub trafficking to the Ste2/CPS-sorting compartment, allowing MVB sorting in the absence of Mvb12. This idea seems plausible in that Ste2 is thought to internalize in a ubiquitinated form, whereas Ste3 is thought to recycle between an endosomal compartment and the cell surface until ubiquitination removes it from this recycling loop (Terrell et al., 1998; Chen and Davis, 2002). It is possible that endocytosed cargoes that are ubiquitinated bypass a recycling endosome. At present, characterization of the yeast endosomal system is not sufficient to address the possibility of discrete endosomes for sorting distinct MVB cargoes. Further experimentation will be required to map out the potential subtleties of the yeast endosomal system and cargo-specific MVB sorting.
A third model for the role of Mvb12 in cargo-specific sorting involves some more general form of modulation of ESCRT-I function, perhaps through associations with additional ESCRT machinery. For example, ESCRT-I efficiency could be impacted through Mvb12-modulated interactions with components that act upstream or downstream of ESCRT-I. How might this lead to cargo-specific defects upon loss of Mvb12? Different MVB cargoes, such as Sna3 and Ste3 may be more sensitive to the level of ESCRT-I function, thereby revealing differential sorting phenotypes when ESCRT-I function is compromised by loss of Mvb12. However, in our experience GFP-CPS is the most sensitive cargo that can be used to address function of the MVB pathway (Azmi et al., 2006; our unpublished data). Thus, the observation that mvb12 has only a subtle impact on the sorting of GFP-CPS seems to cast doubt upon this third model. However, further experimentation will be required to definitively address this role for Mvb12 as a general modulator of ESCRT-I function.
Loss of Mvb12 function confers partial defects in the sorting of cargoes into the MVB pathway. Although the mechanism is unclear, Mvb12 seems to play a role in assisting ESCRT-I to sort a subset of MVB cargoes as a stable component of ESCRT-I. This begs the question as to how this was missed during the initial characterization of ESCRT-I. One possibility is that the small size of the Mvb12 protein (12 kDa) precluded its identification in the isolation procedure by using protein A-tagged Vps23 (Katzmann et al., 2001). Although the reason that Mvb12 was not previously identified remains unclear, the more intriguing question that arises is whether mammalian ESCRT-I may also contain accessory subunits that have eluded detection to this point. Although Mvb12 has been conserved among fungi, sequence homologues in higher eukaryotes have escaped our detection. However, it seems likely that functional homologues exist that modulate the function of mammalian ESCRT-I with respect to specific MVB cargoes as well. Additional studies will provide a deeper understanding of the role of Mvb12 in cargo-specific ESCRT-I function during the MVB sorting reaction and may provide insight regarding mammalian counterparts as well.
Additional Note: The laboratories of both Markus Babst (University of Utah, Salt Lake City, UT) and Scott Emr (University of California, San Diego, San Diego, CA) have independently identified the gene product of YGR206w as a subunit of the ESCRT-I complex. Because it is a 12-kDa protein involved in MVB sorting, we have agreed on the designation Mvb12 for the gene product of YGR206w.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org). 
* These authors contributed equally to this work.
Address correspondence to: David J. Katzmann (katzmann.david{at}mayo.edu)
Abbreviations used: CPS, carboxypeptidase S; CPY, carboxypeptidase Y; ESCRT, endosomal sorting complex required for transport; MVB, multivesicular body; ORF, open reading frame; PtdIns(3)P, phosphatidylinositol-3-phosphate; TAP, tandem affinity purification; Ub, ubiquitin; VPS, vacuolar protein sorting
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Babst, M. (2005). A protein's final ESCRT. Traffic 6, 2–9.[CrossRef][Medline]
Babst, M., Katzmann, D. J., Estepa-Sabal, E. J., Meerloo, T., Emr, S. D. (2002a). Escrt-III: an endosome-associated heterooligomeric protein complex required for mvb sorting. Dev. Cell 3, 271–282.[CrossRef][Medline]
Babst, M., Katzmann, D. J., Snyder, W. B., Wendland, B., Emr, S. D. (2002b). Endosome-associated complex, ESCRT-II, recruits transport machinery for protein sorting at the multivesicular body. Dev. Cell 3, 283–289.[CrossRef][Medline]
Babst, M., Odorizzi, G., Estepa, E. J., Emr, S. D. (2000). Mammalian tumor susceptibility gene 101 (TSG101) and the yeast homologue, Vps23p, both function in late endosomal trafficking. Traffic 1, 248–258.[CrossRef][Medline]
Babst, M., Sato, T. K., Banta, L. M., Emr, S. D. (1997). Endosomal transport function in yeast requires a novel AAA-type ATPase, Vps4p. EMBO. J 16, 1820–1831.[CrossRef][Medline]
Bache, K. G., Brech, A., Mehlum, A., Stenmark, H. (2003). Hrs regulates multivesicular body formation via ESCRT recruitment to endosomes. J. Cell Biol 162, 435–442.
Bache, K. G., Slagsvold, T., Cabezas, A., Rosendal, K. R., Raiborg, C., Stenmark, H. (2004). The growth-regulatory protein HCRP1/hVps37A is a subunit of mammalian ESCRT-I and mediates receptor down-regulation. Mol. Biol. Cell 15, 4337–4346.
Bilodeau, P. S., Urbanowski, J. L., Winistorfer, S. C., Piper, R. C. (2002). The Vps27p Hse1p complex binds ubiquitin and mediates endosomal protein sorting. Nat. Cell Biol 4, 534–539.[Medline]
Bilodeau, P. S., Winistorfer, S. C., Kearney, W. R., Robertson, A. D., Piper, R. C. (2003). Vps27-Hse1 and ESCRT-I complexes cooperate to increase efficiency of sorting ubiquitinated proteins at the endosome. J. Cell Biol 163, 237–243.
Bishop, N. and Woodman, P. (2001). Tsg101/mammalian vps23 and mammalian vps28 interact directly and are recruited to vps4-induced endosomes. J. Biol. Chem 276, 11735–11742.
Bonangelino, C. J., Chavez, E. M., Bonifacino, J. S. (2002). Genomic screen for vacuolar protein sorting genes in Saccharomyces cerevisiae. Mol. Biol. Cell 13, 2486–2501.
Bowers, K., Lottridge, J., Helliwell, S. B., Goldthwaite, L. M., Luzio, J. P., Stevens, T. H. (2004). Protein-protein interactions of ESCRT complexes in the yeast Saccharomyces cerevisiae. Traffic 5, 194–210.[CrossRef][Medline]
Chen, L. and Davis, N. G. (2002). Ubiquitin-independent entry into the yeast recycling pathway. Traffic 3, 110–123.[CrossRef][Medline]
Davies, B. A., Topp, J. D., Sfeir, A. J., Katzmann, D. J., Carney, D. S., Tall, G. G., Friedberg, A. S., Deng, L., Chen, Z., Horazdovsky, B. F. (2003). Vps9p CUE domain ubiquitin binding is required for efficient endocytic protein traffic. J. Biol. Chem 278, 19826–19833.
Gorden, P., Carpentier, J. L., Cohen, S., Orci, L. (1978). Epidermal growth factor: morphological demonstration of binding, internalization, and lysosomal association in human fibroblasts. Proc. Natl. Acad. Sci. USA 75, 5025–5029.
Gruenberg, J. and Stenmark, H. (2004). The biogenesis of multivesicular endosomes. Nat. Rev. Mol. Cell Biol 5, 317–323.[CrossRef][Medline]
Haigler, H. T., McKanna, J. A., Cohen, S. (1979). Direct visualization of the binding and internalization of a ferritin conjugate of epidermal growth factor in human carcinoma cells A-431. J. Cell Biol 81, 382–395.
Hicke, L. and Dunn, R. (2003). Regulation of membrane protein transport by ubiquitin and ubiquitin-binding proteins. Annu. Rev. Cell Dev. Biol 19, 141–172.[CrossRef][Medline]
Huh, W. K., Falvo, J. V., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S., O'Shea, E. K. (2003). Global analysis of protein localization in budding yeast. Nature 425, 686–691.[CrossRef][Medline]
Katzmann, D. J., Babst, M., Emr, S. D. (2001). Ubiquitin-dependent sorting into the multivesicular body pathway requires the function of a conserved endosomal protein sorting complex, ESCRT-I. Cell 106, 145–155.[CrossRef][Medline]
Katzmann, D. J., Odorizzi, G., Emr, S. D. (2002). Receptor downregulation and multivesicular-body sorting. Nat. Rev. Mol. Cell Biol 3, 893–905.[CrossRef][Medline]
Katzmann, D. J., Sarkar, S., Chu, T., Audhya, A., Emr, S. D. (2004). Multivesicular body sorting: ubiquitin ligase Rsp5 is required for the modification and sorting of carboxypeptidase S. Mol. Biol. Cell 15, 468–480.
Katzmann, D. J., Stefan, C. J., Babst, M., Emr, S. D. (2003). Vps27 recruits ESCRT machinery to endosomes during MVB sorting. J. Cell Biol 162, 413–423.
Kostelansky, M. S., Sun, J., Lee, S., Kim, J., Ghirlando, R., Hierro, A., Emr, S. D., Hurley, J. H. (2006). Structural and functional organization of the ESCRT-I trafficking complex. Cell 125, 113–126.[CrossRef][Medline]
Krogan, N. J., et al. (2006). Global landscape of protein complexes in the yeast Saccharomyces cerevisiae. Nature 440, 637–643.[CrossRef][Medline]
Li, Y., Kane, T., Tipper, C., Spatrick, P., Jenness, D. D. (1999). Yeast mutants affecting possible quality control of plasma membrane proteins. Mol. Cell. Biol 19, 3588–3599.
Morita, E. and Sundquist, W. I. (2004). Retrovirus budding. Annu. Rev. Cell Dev. Biol 20, 395–425.[CrossRef][Medline]
Mumberg, D., Muller, R., Funk, M. (1995). Yeast vectors for the controlled expression of heterologous proteins in different genetic backgrounds. Gene 156, 119–122.[CrossRef][Medline]
Odorizzi, G., Babst, M., Emr, S. D. (1998). Fab1p PtdIns(3)P 5-kinase function essential for protein sorting in the multivesicular body. Cell 95, 847–858.[CrossRef][Medline]
Oestreich, A. J., Aboian, M., Lee, J., Azmi, I., Payne, J., Issaka, R., Davies, B. A., Katzmann, D. J. (2007). Characterization of multiple multivesicular body sorting determinants within Sna3: a role for the ubiquitin ligase Rsp5. Mol. Biol. Cell 18, 707–720.
Peng, J., Schwartz, D., Elias, J. E., Thoreen, C. C., Cheng, D., Marsischky, G., Roelofs, J., Finley, D., Gygi, S. P. (2003). A proteomics approach to understanding protein ubiquitination. Nat. Biotechnol 21, 921–926.[CrossRef][Medline]
Raiborg, C., Bache, K. G., Gillooly, D. J., Madshus, I. H., Stang, E., Stenmark, H. (2002). Hrs sorts ubiquitinated proteins into clathrin-coated microdomains of early endosomes. Nat. Cell Biol 4, 394–398.[CrossRef][Medline]
Raiborg, C., Rusten, T. E., Stenmark, H. (2003). Protein sorting into multivesicular endosomes. Curr. Opin. Cell Biol 15, 446–455.[CrossRef][Medline]
Raymond, C. K., Howald-Stevenson, I., Vater, C. A., Stevens, T. H. (1992). Morphological classification of the yeast vacuolar protein sorting mutants: evidence for a prevacuolar compartment in class E vps mutants. Mol. Biol. Cell 3, 1389–1402.[Abstract]
Reggiori, F. and Pelham, H. R. (2001). Sorting of proteins into multivesicular bodies: ubiquitin-dependent and -independent targeting. EMBO J 20, 5176–5186.[CrossRef][Medline]
Shih, S. C., Katzmann, D. J., Schnell, J. D., Sutanto, M., Emr, S. D., Hicke, L. (2002). Epsins and Vps27p/Hrs contain ubiquitin-binding domains that function in receptor endocytosis. Nat. Cell Biol 4, 389–393.[CrossRef][Medline]
Teo, H., Gill, D. J., Sun, J., Perisic, O., Veprintsev, D. B., Vallis, Y., Emr, S. D., Williams, R. L. (2006). ESCRT-I core and ESCRT-II GLUE domain structures reveal role for GLUE in linking to ESCRT-I and membranes. Cell 125, 99–111.[CrossRef][Medline]
Teo, H., Veprintsev, D. B., Williams, R. L. (2004). Structural insights into endosomal sorting complex required for transport (ESCRT-I) recognition of ubiquitinated proteins. J. Biol. Chem 279, 28689–28696.
Terrell, J., Shih, S., Dunn, R., Hicke, L. (1998). A function for monoubiquitination in the internalization of a G protein-coupled receptor. Mol. Cell 1, 193–202.[CrossRef][Medline]
Theos, A. C., Truschel, S. T., Tenza, D., Hurbain, I., Harper, D. C., Berson, J. F., Thomas, P. C., Raposo, G., Marks, M. S. (2006). A lumenal domain-dependent pathway for sorting to intralumenal vesicles of multivesicular endosomes involved in organelle morphogenesis. Dev. Cell 10, 343–354.[CrossRef][Medline]
Tong, A. H., et al. (2001). Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science 294, 2364–2368.
Urbanowski, J. L. and Piper, R. C. (2001). Ubiquitin sorts proteins into the intralumenal degradative compartment of the late-endosome/vacuole. Traffic 2, 622–630.[CrossRef][Medline]
Vida, T. A. and Emr, S. D. (1995). A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast. J. Cell Biol 128, 779–792.
von Schwedler, U. K., et al. (2003). The protein network of HIV budding. Cell 114, 701–713.[CrossRef][Medline]
White, I. J., Bailey, L. M., Aghakhani, M. R., Moss, S. E., Futter, C. E. (2006). EGF stimulates annexin 1-dependent inward vesiculation in a multivesicular endosome subpopulation. EMBO J 25, 1–12.[CrossRef]
).