Regulation of endosomal clathrin and retromer-mediated endosome to Golgi retrograde transport by the J-domain protein RME-8
2009; Springer Nature; Volume: 28; Issue: 21 Linguagem: Inglês
10.1038/emboj.2009.272
ISSN1460-2075
AutoresAnbing Shi, Lin Sun, Riju Banerjee, Michael P. Tobin, Yinhua Zhang, Barth D. Grant,
Tópico(s)Calcium signaling and nucleotide metabolism
ResumoArticle17 September 2009free access Regulation of endosomal clathrin and retromer-mediated endosome to Golgi retrograde transport by the J-domain protein RME-8 Anbing Shi Anbing Shi Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Lin Sun Lin Sun Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Riju Banerjee Riju Banerjee Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Michael Tobin Michael Tobin Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Yinhua Zhang Yinhua Zhang Division of Parasitology, New England Biolabs, Inc., Ipswich, MA, USA Search for more papers by this author Barth D Grant Corresponding Author Barth D Grant Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Anbing Shi Anbing Shi Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Lin Sun Lin Sun Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Riju Banerjee Riju Banerjee Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Michael Tobin Michael Tobin Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Yinhua Zhang Yinhua Zhang Division of Parasitology, New England Biolabs, Inc., Ipswich, MA, USA Search for more papers by this author Barth D Grant Corresponding Author Barth D Grant Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA Search for more papers by this author Author Information Anbing Shi1, Lin Sun1, Riju Banerjee1, Michael Tobin1, Yinhua Zhang2 and Barth D Grant 1 1Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ, USA 2Division of Parasitology, New England Biolabs, Inc., Ipswich, MA, USA *Corresponding author. Department of Molecular Biology and Biochemistry, Rutgers University, Nelson Biological Labs, Room A307, 604 Allison Road, Piscataway, NJ 08854, USA. Tel.: +1 732 445 7340; Fax: +1 732 445 4213; E-mail: [email protected] The EMBO Journal (2009)28:3290-3302https://doi.org/10.1038/emboj.2009.272 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info After endocytosis, most cargo enters the pleiomorphic early endosomes in which sorting occurs. As endosomes mature, transmembrane cargo can be sequestered into inwardly budding vesicles for degradation, or can exit the endosome in membrane tubules for recycling to the plasma membrane, the recycling endosome, or the Golgi apparatus. Endosome to Golgi transport requires the retromer complex. Without retromer, recycling cargo such as the MIG-14/Wntless protein aberrantly enters the degradative pathway and is depleted from the Golgi. Endosome-associated clathrin also affects the recycling of retrograde cargo and has been shown to function in the formation of endosomal subdomains. Here, we find that the Caemorhabditis elegans endosomal J-domain protein RME-8 associates with the retromer component SNX-1. Loss of SNX-1, RME-8, or the clathrin chaperone Hsc70/HSP-1 leads to over-accumulation of endosomal clathrin, reduced clathrin dynamics, and missorting of MIG-14 to the lysosome. Our results indicate a mechanism, whereby retromer can regulate endosomal clathrin dynamics through RME-8 and Hsc70, promoting the sorting of recycling cargo into the retrograde pathway. Introduction RME-8 was originally identified in Caenorhabditis elegans in our earlier genetic screens, isolated as a temperature sensitive lethal mutant defective in endocytosis (Grant and Hirsh, 1999; Zhang et al, 2001). Further analysis showed that RME-8 is normally abundant on endosomes, but not the plasma membrane, suggesting a function for RME-8 in endosome function rather than plasma membrane uptake processes (Zhang et al, 2001). Subsequently, rme-8 mutants were also identified in genetic screens in Drosophila, showing endocytic defects in several tissues (Chang et al, 2004). Further analysis also showed that as in C. elegans, Drosophila RME-8 protein was associated with endosomes and not the plasma membrane (Chang et al, 2004). Likewise, the RME-8 orthologue in mammals is tightly associated with the membranes of endosomes, but is not enriched on the plasma membrane (Girard et al, 2005; Fujibayashi et al, 2008). In plants, RME-8 is also endosome associated and is required for gravitropism, probably because of abnormalities in the vacuole of rme-8 mutant cells that alters amyloplast sedimentation (Silady et al, 2008). Despite all of these observations, an understanding of the precise membrane trafficking defect in rme-8 mutants has remained elusive. RME-8 in all these species is a large protein (>200 kDa) with highly conserved features including a central DNA J-domain and four repeated motifs (IWN repeats) of unknown function (Zhang et al, 2001; Chang et al, 2004; Silady et al, 2004; Girard et al, 2005). DNA J-domains are known to bind to the ubiquitous Hsc70 chaperone, and J-domains in general are a characteristic feature of Hsc70 co-chaperones, proteins that recruit Hsc70 to specific cellular compartments and stimulate Hsc70 ATPase activity (Walsh et al, 2004). For example, the auxillin DNA J-domain protein recruits and activates Hsc70 on clathrin-coated vesicles to facilitate uncoating, a prerequisite for downstream fusion steps (Ungewickell et al, 1995). Although auxillin and its paralogue GAK mediate this function for CCVs derived from the plasma membrane and Golgi (Greener et al, 2001; Eisenberg and Greene, 2007), RME-8 is the only DNA J-domain protein reported to localize to endosomes. Drosophila and human RME-8 J-domains have been shown to physically interact with Hsc70 (Chang et al, 2004; Girard et al, 2005). Drosophila Hsc70 mutants interact with rme-8 mutants genetically, and dominant negative Hsc70 expressed in mammalian cells impairs endosome function (Newmyer and Schmid, 2001; Chang et al, 2002). These results suggested that RME-8 could mediate its effects on endosomes through a role as an endosomal Hsc70 co-chaperone. The early endosome is known to be a hub for the sorting of membrane proteins after endocytosis. In the endosome, cargo proteins can either be delivered to lysosomes through multivesicular bodies (MVBs) for degradation, recycled to the plasma membrane directly or indirectly through the recycling endosome, or can be recycled through retrograde transport from endosomes to the trans-Golgi network (TGN). The retrograde pathway and retromer, the major regulatory complex associated with this pathway, are vital for the retrieval of the Golgi sorting receptors from endosomes to the TGN in yeast and mammalian cells (Bonifacino and Hurley, 2008). This includes receptors that transport degradative enzymes to the vacuole/lysosome such as Vps10 and CI-MPR (cation-independent mannose 6-phosphate receptor) (Marcusson et al, 1994; Arighi et al, 2004; Carlton et al, 2004; Seaman, 2004). Recently, retromer-dependent retrieval of the Wnt-ligand chaperone MIG-14/Wntless has also been shown (Belenkaya et al, 2008; Franch-Marro et al, 2008; Pan et al, 2008; Port et al, 2008; Yang et al, 2008). Retromer consists of a core complex Vps5-Vps17 (SNX1/2-SNX5/6 in mammals) and a cargo recognition complex Vps26-Vps29-Vps35 (Bonifacino and Hurley, 2008; Collins, 2008). In addition to retromer, clathrin and clathrin-related proteins (such as epsinR and AP-1) have also been reported to function as components of the retrograde transport machinery. Clathrin and clathrin adaptor epsinR have been shown to be required for retrograde transport of the Shiga toxin B subunit (Lauvrak et al, 2004; Saint-Pol et al, 2004). Clathrin adaptor AP-1 is also required for retrograde transport of mannose 6-phosphate receptor MPR46 (Meyer et al, 2000). Vps27/Hrs together with the ESCRT-I,II,III (endosomal sorting complex required for transport) complexes also functions on early endosomes and MVBs, but rather than promoting retrograde recycling, these proteins are best known for promoting degradation of integral membrane proteins (Katoh et al, 2003; Odorizzi et al, 2003). Importantly, Hrs has been linked to the accumulation of endosomal clathrin, promoting the formation of degradative ESCRT-enriched endosomal subdomains (Raiborg et al, 2001a, 2002). In this study, we analyse retrograde transport in C. elegans and identify an earlier unsuspected mechanism for the regulation of endosomal clathrin that is required for retrograde transport from endosomes to the Golgi. We show that the J-domain protein RME-8 binds to SNX-1, and that loss of function in rme-8 or snx-1, or depletion of C. elegans Hsc70 (HSP-1) by RNAi, disrupts endosome to Golgi transport of the retromer-dependent cargo protein MIG-14/Wntless. In the absence of any of RME-8, SNX-1, or HSP-1/Hsc70, MIG-14 is missorted to the late endosome and lysosome and is depleted from the Golgi. Furthermore, we show that loss of any of these three proteins leads to accumulation of clathrin on endosomes and loss of endosomal clathrin dynamics. Our work shows that retromer, through RME-8 and Hsc70, acts to limit clathrin accumulation, a prerequisite for the recycling of retrograde cargo. In the absence of this regulation, the retrograde transport route is lost and cargo that should be retrieved from the endosome is instead degraded. Results RME-8 physically interacts with SNX-1 To better understand the function of RME-8 as an endosomal regulator, we screened for interacting proteins using the yeast two-hybrid system. The bait construct included part of the DNA J-domain and the entire C-terminal half of the RME-8 protein (amino-acids 1337–2279; Figure 1C) including IWN repeats 3 and 4 and the intervening Armadillo (ARM)-like region. One clone encoding a portion of the retromer component SNX-1 was recovered from this interaction screen (see Materials and methods). Further analysis showed that full-length SNX-1 also interacts with RME-8 in this assay (data not shown). Successively, smaller regions of RME-8 were used as bait, narrowing the interacting region to amino-acids 1388–1950 (Figure 1B and C). Deletion of amino-acids 1388–1619 or 1732–1950 of RME-8 abrogated the interaction, indicating a requirement for RME-8 sequences corresponding to C-terminal region beyond the DNA J-domain, including the adjacent linker between the J-domain and IWN3, IWN3 itself, and the ARM-like domain (Figure 1B and C). Sequences C-terminal to the ARM-like domain, including IWN4, were dispensable for the interaction. Figure 1.RME-8 physically interacts with SNX-1. (A) The interaction between RME-8 and SNX-1 requires C-terminal SNX-1 residues span 423–466. RME-8 (residues 1337–2279) was expressed in a yeast reporter strain as a fusion with the DNA-binding domain of LexA (bait). SNX-1 and its truncated forms were expressed in the same yeast cells as fusions with the B42 transcriptional activation domain (prey). Interaction between bait and prey was assayed by complementation of leucine auxotrophy (LEU2 growth assay). Colonies were diluted in liquid and spotted on solid growth medium directly or after further 10X dilution. (B) The interaction of RME-8 with SNX-1 requires RME-8 sequences C-terminal to the DNA J-domain. SNX-1 (residues 221–472) was expressed in a yeast reporter strain, a fusion with the B42 transcriptional activation domain (prey). Mutant forms of RME-8 were expressed in the same yeast cells as a fusion with the DNA-binding domain of LexA (bait). Interaction between bait and prey was assayed by complementation of leucine auxotrophy (LEU2 growth assay) as above. (C) Schematic representations of SNX-1 and RME-8 and the regions of each used in the Y2H analysis. Protein domains are displayed as boxes (white for the ARM-like domain, dark for others) above protein sequences used in the study (shown as dark lines). Amino-acid numbers are indicated. (D) Glutathione beads loaded with recombinant GST or GST-SNX-1(271–472) were incubated with in vitro expressed HA-RME-8(1337–2279), and then washed to remove unbound proteins. Bound proteins were eluted and analysed by western blot using anti-HA (top) or anti-GST (bottom) antibodies. Input lane contains in vitro expressed HA-RME-8(1337–2279) used in the binding assays (10%). Download figure Download PowerPoint Residues 423–466 of SNX-1, corresponding to a portion of helix 3 of the SNX-1 VPS5/BAR domain, were necessary and sufficient for the interaction in the yeast two-hybrid system (Figure 1A and C). This region of the SNX-1 BAR domain is phylogenetically highly conserved, and molecular modelling of the SNX-1 BAR domain suggests that it would be exposed to the cytoplasm, away from the membrane-binding surface, in which it could interact with other proteins (data not shown). We also confirmed the binding and the specificity of the RME-8/SNX-1 interaction using an independent glutathione S-transferase (GST)-pull down assay. Recombinant GST-SNX-1 (aa 271–472), GST-Y59A8B.22/Snx5, GST-LST-4/Snx9, or GST only was immobilized on glutathione sepharose beads and incubated with in vitro transcribed and translated HA-tagged RME-8 (aa 1337–2279). Only GST-SNX-1, but not other SNX proteins, pulled down RME-8 (Figure 1D; Supplementary Figure 1). RME-8 colocalizes with SNX-1 SNX-1 is the only C. elegans homologue of mammalian Sorting Nexin 1 and Sorting Nexin 2, PX and BAR domain proteins that function in endosome to Golgi retrograde transport with VPS-26, VPS-29, and VPS-35 as part of the retromer complex (Bonifacino and Hurley, 2008). The physical association of RME-8 with SNX-1 suggested that RME-8 might function with the retromer complex on endosomes to mediate retrograde transport. If the RME-8/SNX-1 physical interaction is functionally relevant in vivo, then the two proteins would be expected to colocalize on endosomes. Indeed, we found that mCherry-RME-8 colocalized extensively with GFP-SNX-1 in several different tissues, including the intestine, the hypodermis, and coelomocytes (Figure 2A–A″; Supplementary Figure 2A–B″). Figure 2.RME-8 colocalizes with SNX-1 and RAB-5 on early endosomes. Representative images from deconvolved 3D image stacks are shown. All images were acquired in intact living animals expressing GFP and mCherry-tagged proteins specifically in intestinal epithelial cells. (A–A″) mCherry-RME-8 colocalizes with GFP-SNX-1. Arrowheads indicate endosomes labelled by both GFP-SNX-1 and mCherry-RME-8. (B–B″) mCherry-SNX-1 colocalizes with early endosome marker GFP-RAB-5. Arrowheads indicate endosomes labelled by both GFP-RAB-5 and mCherry-SNX-1. (C–C″) mCherry-SNX-1 does not colocalize well with Golgi marker MANS-GFP. Arrowheads and the inset indicate mCherry-SNX-1 positive endosomes juxtaposed to MANS-GFP-labelled Golgi ministacks. (D–D″) mCherry-RME-8 colocalizes with early endosome marker GFP-RAB-5. Arrowheads indicate endosomes labelled by both GFP-RAB-5 and mCherry-RME-8. (E–E″) mCherry-RME-8 does not colocalize well with Golgi marker MANS-GFP. Arrowheads and the inset indicate mCherry-RME-8 positive endosomes juxtaposed to MANS-GFP-labelled Golgi ministacks. Enlarged images (× 4) of boxed regions are shown in the insets. In each image, autofluorescent lysosome-like organelles can be seen in all three channels with the strongest signal in blue, whereas GFP appears only in the green channel and mCherry only in the red channel. Signals observed in the green or red channels that do not overlap with signals in the blue channel are considered bone fide GFP or mCherry signals, respectively. Scale bar represents 10 μm. Download figure Download PowerPoint RME-8 and SNX-1 localize to early endosomes but not Golgi In other organisms, the retromer complex, including Snx1 and Snx2, is found primarily on early endosomes, and to a lesser extent on or near the Golgi apparatus (Bonifacino and Hurley, 2008). As expected, we found that in C. elegans, mCherry-SNX-1 colocalized well with early endosomal markers GFP-EEA-1 and GFP-RAB-5 in several tissues of living animals (Figure 2B–B″; Supplementary Figure 3). GFP-RME-8 also colocalized well with early endosome markers (Figure 2D–D″; Supplementary Figure 4). Similar to all invertebrate Golgi, C. elegans Golgi appears as dispersed ministacks throughout the cell rather than in one large juxtanuclear stack (Grant and Sato, 2006). In the intestine, the Golgi ministacks are similar in size to early endosomes (Chen et al, 2006). We found very little direct overlap of mCherry-SNX-1 or mCherry-RME-8 with Golgi marker Mannosidase (MANS)-GFP (Figure 2C–C″ and E–E″). Rather, we noted that RME-8 and SNX-1-labelled endosomes were very often directly juxtaposed to MANS-labelled Golgi, a localization that could potentially facilitate retrograde transport (Figure 2C″ and E″, note insets). GFP-RAB-5 also shows a similar high incidence of juxtaposition to MANS-labelled Golgi, further suggesting a close association of a population of endosomes with Golgi ministacks (data not shown). MIG-14/Wntless is missorted to the late endosome and lysosome in rme-8 and snx-1 mutants Earlier analysis of Golgi-resident retromer-dependent cargo proteins, including the mammalian CI-MPR (cation-independent mannose-6-phosphate receptor) and the C. elegans, Drosophila, and human MIG-14/Wntless proteins, showed that in the absence of retromer function, such cargos were depleted from the Golgi and missorted to the late endosome and lysosome (Belenkaya et al, 2008; Franch-Marro et al, 2008; Pan et al, 2008; Port et al, 2008; Yang et al, 2008). The resulting increase in CI-MPR and MIG-14/Wntless lysosomal degradation leads to reduced steady-state levels of these cargo proteins in the cell (Rojas et al, 2007). Thus, we reasoned that if RME-8 functions with SNX-1 in retrograde transport, rather than in endocytic uptake at the plasma membrane, then rme-8 mutants should missort cargo in the same manner as snx-1 mutants. Conversely, defective endocytosis of such cargo would be expected to lead to cargo accumulation at the plasma membrane, as was earlier observed for MIG-14-GFP in dpy-23/mu2-adaptin mutants (Pan et al, 2008). To date, the only retromer-dependent cargo protein known in C. elegans is MIG-14 (Pan et al, 2008; Yang et al, 2008). Thus, to determine whether RME-8 is required for retrograde transport, we assayed for changes in MIG-14-GFP endocytic sorting, comparing wild-type animals with rme-8 mutants. We also assayed for changes in MIG-14-GFP localization in snx-1 deletion mutants. As expected in wild-type animals, MIG-14-GFP colocalized well with the early endosome marker mCherry-RAB-5 and the Golgi marker MANS-mCherry, but not with mCherry-RAB-7, a marker for late endosomes and lysosomes (Supplementary Figure 5A–A″, E–E″; Figure 3O–O″). Consistent with the proposition that RME-8 functions with SNX-1 in retrograde transport, both rme-8 and snx-1 mutants displayed a greater than 10-fold reduction in average intestinally expressed MIG-14-GFP fluorescence intensity compared with wild-type controls (Figure 3A–C″, quantified in Figure 3H). Furthermore, overexpression of RFP fused to SNX-1(423–466), the 43 amino-acid fragment of SNX-1 identified above that interacts with RME-8, resulted in a three-fold reduction in MIG-14-GFP fluorescence, presumably by interfering with the association of the endogenous SNX-1 and RME-8 proteins (Supplementary Figure 11A–C). The reduction in MIG-14-GFP signal in all these backgrounds is equivalent to the phenotype earlier reported for MIG-14-GFP in vps-35 mutants (Pan et al, 2008; Yang et al, 2008). In the case of the rme-8 mutant, which is temperature sensitive, the dramatic redistribution of MIG-14-GFP only occurred at the restrictive temperature (data not shown). We observed similar reductions in MIG-14-GFP levels in rme-8 mutant early embryos, indicating that the requirement for RME-8 in sorting of retrograde cargo is not cell-type specific (Supplementary Figure 5I and J). Figure 3.MIG-14 recycling requires RME-8 and SNX-1. Panels A-G′ show confocal micrographs of intestinally expressed MIG-14-GFP in top and middle focal planes in intact living animals. Note the reduced MIG-14-GFP intensity, and altered subcellular localization, in rme-8 and snx-1 mutants (A–H). RNAi-mediated depletion of the mu2 subunit of AP-2 (DPY-23) restored MIG-14-GFP fluorescence, and enhanced cell surface localization, in both rme-8 and snx-1 mutants (D–E′, H), arrowheads indicate lateral membrane, arrows indicate basal membrane. RNAi-mediated depletion of lysosome biogenesis protein CUP-5/mucolipin1 inhibited loss of MIG-14-GFP signal in rme-8 and snx-1 mutants (F–G′). Bar graph indicating average MIG-14-GFP fluorescence intensity calculated as described in Materials and methods (H). Asterisks indicate a significant difference in the one-tailed Student's t-test (P<0.01). P-values for MIG-14-GFP were 6.58 × 10–7 for wild type (WT) versus rme-8(b1023) and 5.43 × 10–7 for WT versus snx-1(tm847). Cargo proteins that recycle through the recycling endosome, the human transferrin receptor (hTfR-GFP), and the IL-2 receptor alpha chain (hTAC-GFP) were not affected in rme-8 and snx-1 mutants (I–N). For panels O–R″ images from deconvolved 3D image stacks are shown. All images were acquired in intact living animals expressing GFP and mCherry-tagged proteins specifically in intestinal epithelial cells. Autofluorescent lysosome-like organelles are shown in blue. In rme-8(b1023), snx-1(tm847), and hsp-1/hsc70(RNAi) animals, but not WT animals (O–O″), MIG-14-GFP colocalizes with late endosome and lysosome marker mCherry-RAB-7 (P–R″). MIG-14-GFP localizes to the apparent limiting membrane of the late endosomes/lysosomes as well as apparent intralumenal structures. Note that the signal intensity for MIG-14-GFP in mutant and RNAi animals was boosted to allow visualization and colocalization. Arrowheads indicate late endosomes/lysosomes labelled by both mCherry-RAB-7 and MIG-14-GFP. Insets show enlargements (× 3) of the boxed area. Scale bar represents 10 μm. Download figure Download PowerPoint As controls, we assayed the distribution of earlier characterized model transmembrane cargos that recycle through the recycling endosome and not the Golgi: the human transferrin receptor (hTfR-GFP) and the IL-2 receptor alpha chain (hTac-GFP) (Chen et al, 2006; Shi et al, 2007). The localization and steady-state levels of these receptors in the intestine were unaffected by mutation of rme-8 or snx-1 (Figure 3I–N). Consistent with our results, transferrin endocytosis in mammalian cells is normal after depletion of RME-8 by RNAi (Girard et al, 2005). In subsequent experiments, we boosted the remaining signal for intestinal MIG-14-GFP in mutant animals to determine the fate of MIG-14 under these conditions (Figure 3P–R″; Supplementary Figure 5B–D″ and F–H″). Most of the remaining MIG-14-GFP protein in rme-8 and snx-1 mutants was found in ring-like late endosomes and lysosomes positive for mCherry-RAB-7, a localization not found in wild-type animals (Figure 3P–Q″). In rme-8 and snx-1 mutant cells, MIG-14-GFP colocalized with mCherry-RAB-7 on the apparent limiting membrane of these large endosomes and on mCherry-RAB-7 negative puncta apparently contained within the endosome, possibly intralumenal vesicles (Figure 3P–Q″). MIG-14-GFP could also still be observed in enlarged mCherry-RAB-5 labelled early endosomes in rme-8 and snx-1 mutants (Supplementary Figure 5B–C″). However, even with the boosted MIG-14-GFP signal, most MANS-labelled Golgi displayed only weak or adjacent MIG-14-GFP labelling in rme-8 and snx-1 mutants (Supplementary Figure 5F–G″). MIG-14/Wntless trafficking defects occur after endocytosis in rme-8 and snx-1 mutants To further establish the altered trafficking itinerary of MIG-14-GFP in rme-8 and snx-1 mutants, we blocked endocytic uptake of MIG-14-GFP in the rme-8 and snx-1 mutant backgrounds by RNAi-mediated depletion of the mu2 subunit of the clathrin adaptor complex AP-2 (DPY-23) (Pan et al, 2008). This blockade at the cell surface restored MIG-14-GFP fluorescence to levels equal to or above wild type and enhanced cell surface localization of MIG-14-GFP in both rme-8 and snx-1 mutants (Figure 3D–E′ and H), indicating that the trafficking defects associated with MIG-14 in rme-8 and snx-1 mutants occur after endocytosis, consistent with a function for both proteins in endosomal sorting events. We also inhibited lysosome-mediated degradation by depletion of CUP-5/mucolipin1, a transmembrane protein required for normal lysosome biogenesis and normal levels of hydrolytic activity (Treusch et al, 2004). Such depletion of CUP-5 by RNAi blocked much of the abnormal degradation of MIG-14-GFP in rme-8 and snx-1 mutants (Figure 3F–G′ and H), indicating that loss of rme-8 or snx-1 leads to degradation of MIG-14 through missorting into the lysosomal pathway. Similarly, the loss of MIG-14-GFP in rme-8 and snx-1 mutants was also ameliorated by RNAi-mediated depletion of VPS-37, a component of the ESCRT-I complex (data not shown). The increased degradation of MIG-14-GFP suggested that MVB-mediated transport of membrane proteins to the lysosome does not require RME-8 or SNX-1. To examine this point further, we also assayed degradation of CAV-1-GFP in early embryos (Sato et al, 2006; Shi et al, 2007). CAV-1 is a known transmembrane cargo protein that is degraded in the one cell embryo after the metaphase to anaphase transition (Sato et al, 2006, 2008a; Bembenek et al, 2007). Degradation of CAV-1-GFP requires endocytosis and ESCRT-mediated endosomal sorting (Sato et al, 2006; Audhya et al, 2007). CAV-1-GFP degradation was unaffected in rme-8(b1023) and snx-1(tm847) mutants, further suggesting that RME-8 and SNX-1 are not required for ESCRT-mediated degradation of integral membrane proteins (Supplementary Figure 8A–C′). Neuronal cell polarity is impaired in rme-8 and snx-1 mutants The observed abnormal degradation of MIG-14 in rme-8 and snx-1 mutants would be expected to impair Wnt signalling, as shown earlier for vps-35 mutants. To test for an effect of loss of RME-8 and SNX-1 on Wnt signalling, we examined the control of mechanosensory neuron polarity, a process that requires MIG-14 and retromer regulated Wnt signalling (Prasad and Clark, 2006; Pan et al, 2008). As rme-8(b1023) is lethal at 20°C or higher, we examined the rme-8 effect on this phenotype at the permissive temperature (15°C), in which rme-8 mutants display partially penetrant phenotypes. We found that rme-8(b1023) and snx-1(tm847) display defective ALM posterior processes at a penetrance similar to that shown earlier for vps-35 mutants (Supplementary Figure 6A–C and G) (Pan et al, 2008). Defective PLM posterior processes were also observed in both mutants, although in this case, the snx-1 null mutant showed a higher penetrance than the partial loss of RME-8 (Supplementary Figure 6D–F and H). These results are consistent with the requirements we identified for RME-8 and SNX-1 in the control of MIG-14 trafficking. Depletion of C. elegans Hsc70 (HSP-1) causes MIG-14/Wntless missorting to the late endosome and lysosome We next sought to better understand the mechanism by which RME-8 influences endosome function and regulates retrograde transport. Recent work has shown that retrograde transport of the glycosphingolipid-binding bacterial Shiga toxin (STxB), and another retrograde cargo protein TGN46, from the early endosome to the Golgi requires endosomal clathrin and the clathrin adaptor epsinR (Saint-Pol et al, 2004; Popoff et al, 2007). Depletion of RME-8 from the cells of several organisms has been shown to produce abnormal clathrin distribution in vivo (Chang et al, 2004; Girard et al, 2005), and similar to human and Drosophila RME-8, we found that the C. elegans RME-8 J-domain binds to HSP-1/Hsc70 in vitro (Supplementary Figure 11D). Thus, to better understand how RME-8 functions, we examined the relationship between endosomal clathrin, RME-8, and SNX-1. One simple model that could explain these observations would be that SNX-1 cooperates with RME-8 to activate endosomal Hsc70, and in turn Hsc70 chaperone activity on the endosome controls endosomal clathrin dynamics to promote retrograde sorting. Consistent with this model, we found that depletion of C. elegans Hsc70 (HSP-1) by RNAi resulted in aberrant sorting of MIG-14-GFP, very similar to the phenotype observed for rme-8 and snx-1 mutants (Figure 3R–R″; Supplementary Figure 5D–D″, H–H″). After HSP-1 RNAi, MIG-14-GFP levels were reduced and most of the remaining MIG-14-GFP signal was found within large mCherry-RAB-7-labelled endosomes/lysosomes (Figure 3R–R″). Similar to the effects observed in rme-8 and snx-1 mutants, hsp-1(RNAi) led to MIG-14-GFP labelling of the mCherry-RAB-7-labelled limiting membrane and smaller mCherry-RAB-7 negative puncta that appeared to be within the RAB-7-labelled ring (Figure 3R–R″). GFP-SNX-1 and GFP-VPS-35 protein levels were not reduced in rme-8(b1023) or hsp-1(RNAi) animals, suggesting that the defects in retrograde transport were not through effects on retromer protein stability (Supplementary Figure 7A and B). The congruence of snx-1, rme-8, and hsp-1 phenotypes indicates that all three proteins function in a common
Referência(s)