![[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. 19, Issue 12, 5327-5337, December 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Department of Neurobiology, Max Planck Institute for Biophysical Chemistry, Göttingen 37077, Germany; and European Neuroscience Institute, Göttingen 37077, Germany
Submitted May 6, 2008;
Revised August 26, 2008;
Accepted September 26, 2008
Monitoring Editor: Jean E. Gruenberg
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
; Klenchin and Martin, 2000).
Docking, defined as the close contact of two membranes in preparation of fusion (Schikorski and Stevens, 2001), is thought to be controlled by Rab/Ypt GTPases and initialized by specialized tethering molecules bridging the organelles (Sztul and Lupashin, 2006). Several such tethers (generally large coiled-coil proteins or multisubunit complexes) are known (Waters and Hughson, 2000). Tethers may be directly recruited by activated GTPases, as for example the GARP (Golgi-associated retrograde protein) complex in retrograde trafficking. Alternatively, they may contain a guanine nucleotide exchange factor (GEF) activity such as the transport protein particle (TRAPP 1) complex, which activates the GTPase Ypt1p and functions in ER-to-Golgi traffic. Both activities may also be combined, such as in the vacuolar HOPS (homotypic fusion and vacuole protein sorting) complex, which does not only act as a GEF but also as an effector for Ypt7p. Some of these tethers like the HOPS and Dsl1 complex have also been implicated in SNARE (N-ethylmaleimide-sensitive factor [NSF] attachment receptor) binding, which would provide a link to the core fusion machinery (for review see Cai et al., 2007).
SNARE proteins deliver the energy necessary for membrane fusion through the interaction of SNAREs from two opposing membranes and formation of a four-helical bundle: the trans-complex. Complex formation is initiated at the N-terminus of the SNARE motif, which is found distal from the membrane, and proceeds toward the membrane-proximal C-terminus (in a process termed "zippering"), thereby pulling the membranes together and overcoming the energy barrier. After fusion, all SNAREs of the complex are located in the same membrane, in what is termed cis-complexes. These are nonproductive in terms of fusion and need to be disassembled to provide free SNARE molecules for future rounds of fusion, a process achieved by the AAA-ATPase NSF and its cofactor 
SNAP (soluble NSF attachment protein), in an energy-dependent process (Jahn and Scheller, 2006).
One of the trafficking steps in which both docking and fusion have been intensively studied, and where probably most of the main players are known, is the fusion of early endosomes. Early endosomes form the first sorting station on which many routes of endocytosis converge (Maxfield and McGraw, 2004). They fuse not only with incoming endocytosed vesicles, but also with each other (referred to as homotypic fusion), which constitutes an important step in the recycling of material through the endosomal endomembrane system. Docking of early endosomes seems to be mainly regulated by the small GTPase Rab 5. Active (GTP-bound) Rab 5 is localized on the endosome membrane via its isoprenyl anchor. It recruits a complex of Rabaptin-5 and the Rab 5 guanine nucleotide exchange factor Rabex, which together drive the exchange of GDP to GTP in Rab 5, resulting in a positive feedback loop that induces the further accumulation of the active GTPase on the membrane (Stenmark et al., 1995; Lippe et al., 2001). Active Rab 5 also recruits the phospatidyl inositol (PI)-3 kinase hVps34 (Christoforidis et al., 1999b), which increases the local PI-3-phosphate concentration. Active Rab5 and PI(3)P then function as a "coincidence" signal for the recruitment of the long coiled-coil protein early endosomal autoantigen 1 (EEA1; Mu et al., 1995; Patki et al., 1997; Simonsen et al., 1998). EEA1 is a tethering protein essential for docking (Christoforidis et al., 1999a) that is thought to cross-link endosomes by the formation of homodimers (Callaghan et al., 1999).
Although there appears to be consensus that SNAREs are not involved in the initial tethering contact between organelles destined to fuse, it is debated whether they contribute by transinteractions to the establishment of a stably docked state in preparation for fusion. For instance, several recent studies suggest a function for the SNARE syntaxin 1 in the docking of synaptic vesicles (Hammarlund et al., 2007) and of dense core vesicles (Hammarlund et al., 2008) in neurons of Caenorhabditis elegans. Similarly, docking of chromaffin granules significantly reduced in chromaffin cells in which syntaxin 1 was destroyed by botulinum neurotoxins (de Wit et al., 2006). These findings disagree with earlier studies where syntaxin was perturbed without any measurable effects on docking (Broadie et al., 1995; O'Connor et al., 1997; Fujiwara et al., 2006), but it is possible that such changes are only detectable if high-pressure freezing procedures are used for the electron microscopic analysis (Hammarlund et al., 2007, 2008). In addition to exocytotic vesicles, several studies suggest that docking of yeast vacuoles (a compartment comparable to mammalian lysosomes; Wickner and Haas, 2000) depends at least partly on SNARE proteins (Ungermann et al., 1998; Laage and Ungermann, 2001).
Here we analyzed the involvement of SNARE proteins in the docking of mammalian early endosomes. We generated a new fluorescence-based in vitro assay for measuring docking of endosomes. The assay differentiates between docked and fused organelles, thereby allowing for studying both processes in parallel. As expected, both fusion and docking are active, energy-consuming processes. Also, perturbing Rab and EEA1 function reduced both fusion and docking. However, inhibiting SNARE function selectively perturbed fusion, but not docking.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
) and the polyclonal G96 anti-Synaptophysin antibody (Jahn et al., 1985) were described previously. The rabbit anti EEA1 serum used for the functional studies was raised against CLRRILQRTPGRV (Takamori et al., 2006). Antibodies used for Western blotting were monoclonals directed against Rab 5 (Cl 621.1; Fischer von Mollard et al., 1994) and EEA1, which was purchased from BD Biosciences (San Jose, CA). Fab fragments against Syntaxin 6, Syntaxin 13, Vti 1a, and Vamp 4 were a kind gift from Dr. D. Zwilling (University of California).
Recombinant Proteins
All constructs were described before: Syntaxin 6 (residues 169-234; Zwilling et al., 2007), Syntaxin 13 (residues 1-250), Vti1a (residues 1-192), and mutant SNAP L294A (Brandhorst et al., 2006). Recombinant proteins were expressed as His6- or glutathione S-transferase (GST)-tagged fusion proteins and purified by Ni2-agarose or glutathione-Sepharose, respectively. The tags of all proteins were removed by thrombin cleavage. All proteins were further purified by ion-exchange chromatography. GDP-dissociation inhibitor (GDI) was a kind gift from Roger Goody (MPI for Molecular Physiology, Dortmund, Germany).
Cell Culture and Internalization of Marker
PC12 cells (clone 251; Heumann et al., 1983) were grown to confluence on 15-cm-diameter culture dishes in DMEM (with 5% FCS, 10% horse serum, 4 mM glutamine, and 100 U/ml each of penicillin and streptomycin) at 37°C in 10% CO2. Cells were harvested by washing once with room-temperature PBS (150 mM NaCl/200 mM Na2HPO4, pH 7.4), adding trypsin/EDTA (2 ml per plate; Lonza, Cologne, Germany). Cells were washed with internalization medium (OptiMEM; Invitrogen, Karlsruhe, Germany; supplemented with 10 mM glucose), prewarmed, and incubated for 5 min with marker (10-kDa dextran labeled with Alexa 488 or Alexa 594, respectively; Invitrogen), dissolved in internalization medium. The internalization was stopped by diluting the cells in 10 ml of ice-cold PBS with 5 mg/ml BSA, and the cells were washed three times with the solution. For the Atto 647N-labeled dextran, we coupled amino-dextran (Sigma, Munich, Germany) with the NHS-ester of the dye (Atto-tec, Siegen, Germany), by using a conventional protocol (modified from Invitrogen).
Preparation of Subcellular Fractions and Rat Brain Cytosol
PC12 cells were homogenized as described (Holroyd et al., 1999) with slight modifications. Briefly, the cell pellet was resuspended 1:4 in homogenization buffer (250 mM sucrose/3 mM imidazole/HCl, pH 7.4) with protease inhibitors (0.2 mM PMSF, 1 µg/ml leupeptin, 1 µg/ml aprotinin, and 0.7 µg/ml pepstatin) and homogenized by 10 passages through a stainless-steel ball homogenizer with a clearance of 20 µm. The homogenates were centrifuged for 15 min at 1200 xg, and the resulting postnuclear supernatants (PNSs) were divided into aliquots and snap-frozen in liquid N2. PNS fractions labeled with Alexa 488 and Alexa 594 were used for the cell-free interaction assay. Rat brain cytosol was prepared from fraction S2 by centrifugation at 300,000 x g for 30 min (Huttner et al., 1983).
Cell-Free Interaction Assay
Reaction mixtures (50 µl volume) contained, as final concentrations, 4 mg/ml PNS, 2 mg/ml cytosol, 11.25 mM HEPES, pH 7.0, 1.35 mM magnesium acetate, 0.18 mM DTT, 45 mM potassium acetate, as an ATP-regenerating system, 3.2 mM ATP, 26 mM creatine phosphate, and 0.132 mg creatine kinase (800 U/mg; Roche, Basel, Switzerland) or, as an ATP-depleting system, 5 µl hexokinase (1500 U/ml dissolved in 250 mM glucose; Roche). If not indicated otherwise, the reaction time was 45–60 min. The reaction solutions were directly added onto coverslips (18-mm diameter; Marienfeld GmbH, Lauda-Königshofen, Germany) in 12-well plates (into 1 ml of PBS) and centrifuged at 5900 x g in a Multifuge4 centrifuge (Heraeus Instruments, Hanau, Germany) for 45 min. Coverslips were analyzed by using a Zeiss Axiovert 200M fluorescence microscope (Jena, Germany) with a 1.4 NA 100x objective and appropriate filter sets (see Brandhorst et al., 2006).
Clustering Assay
Endosomes from baby hamster kidney (BHK) 21 cells were used in the assay, because their larger size allowed for better imaging than for PC12 endosomes. Early endosomes were purified as described previously (Bethani et al., 2007) and incubated in vitro as indicated above for the cell-free interaction assay. Endosome concentration was adjusted to 0.1–0.2 mg/ml. After 45 min, aliquots of the reactions were centrifuged onto glass coverslips as above. The coverslips were then imaged in presence of the styryl dye FM 2–10 (Invitrogen), at a concentration of 20 µM. The images (captured via a CCD camera with a 1317 x 1035 Kodak chip) were analyzed as follows: the total fluorescence of each image was normalized to an arbitrary value (same for all images, in all experiments), the images were divided in 13 bins of 100-pixel width, and the SD value was calculated for each of the 13 bins. The mean SD was then calculated and was used as an indicator of clusters in the images. For conditions containing mainly single endosomes, the fluorescence signal was evenly distributed within the images, and the SD was low. The clusters generated bright spots in small areas in the images, thus causing a strong increase in the SD measurement.
Membrane Binding of Proteins
Reaction mixtures identical to the once in the interaction assay were incubated with the additions described in the respective experiments. The membrane was then pelleted by ultracentrifugation for 20 min at 135,000 x g in a TLA 55 rotor (Beckman, Krefeld, Germany). For the GDI samples we wanted to analyze the conditions directly at the start of the reaction and therefore minimized the centrifugation time to 10 min at 500,000 x g in a TLA 120.2 rotor (Beckman). Pellets were dissolved and analyzed by Western blotting. The signal was determined by densiometry as described (Bethani et al., 2007), normalized to the synaptobrevin signal, and then expressed as percent of the signal from the control reaction. Controls were incubated either 60 min for GTP
S and wortmannin or 10 min for GDI.
STED Microscopy
Samples were imaged using a TCS STED (Stimulated Emission Depletion) superresolution fluorescence microscope from Leica Microsystems GmbH (Mannheim, Germany), with a 1.4 NA 100x objective (Leica). Excitation was performed with a 635 nm diode laser, and depletion was achieved via a Spectra-Physics MaiTai tunable laser at 750 nm. Signal was detected by use of an Avalanche Photodiode. The system resolution limit is approximately 70–90 nm, measured by analysis of crimson-fluorescent beads (20 nm diameter, Invitrogen).
Data Analysis
Imaging was performed as previously described (Brandhorst et al., 2006; Rizzoli et al., 2006; Bethani et al., 2007). Images performed in the red and green channel were then analyzed automatically by use of a self-written routine in Matlab (The Mathworks, Natick, MA). The images were first high-pass filtered, to remove noise, then thresholds of 4–6 pixels above background were applied, and all objects above the thresholds (excluding single pixels) were used further in the analysis. The intensity centers of the objects were determined, corrected for the shift between the images by use of multicolor fluorescent beads (identified from images taken in the blue channel, where only the beads were seen), and all distances from green centers of intensity to red centers of intensity were determined. The distance to the closest red neighbor was then obtained for each green object and used further in the analysis. Endosomes whose red and green intensity centers were within 112.5 nm from each other were considered to be fused (see Figure 1D). The 25-nm distance from 112.5 to 137.5 nm was not analyzed because these organelles could not be unambiguously assigned to either fusion or docking. Endosomes whose red and green intensity centers were within 137.5–512.5 nm from each other were counted as docked. The value obtained for docking was then corrected for endosome density on the coverslip by subtracting a baseline due to random positioning of the endosomes on the coverglass (the average percentage for the distances from 512.5 to 1012.50 nm).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
; Gruenberg et al., 1989; Brandhorst et al., 2006). PNSs were then isolated from the cells and used in in vitro reactions, which contained rat brain cytosol and an ATP-regenerating system, i.e., conditions under which endosome fusion is known to occur (Brandhorst et al., 2006). Aliquots from the reactions were either spun down immediately onto coverslips (Figure 1A) or after 60 min of incubation (Figure 1B). The coverslips were then imaged by conventional epifluorescence microscopy. We have shown previously that organelles labeled by this method colocalize to a high extent with bona fide early endosomal markers, such as transferrin and the early endosomal, but not the ER-to-Golgi, SNAREs (Brandhorst et al., 2006). Many yellow (colocalized) spots were observed after incubation, compared with virtually none in the control.
|
). As a consequence emitted light is only collected from the center of the excited spot, which leads to a substantial increase in resolution (
70 nm in our hands). To be able to use STED microscopy, we coupled dextran to a STED-suitable fluorescent Atto dye, and used it for fluid-phase uptake as above. The size distribution is unimodal and centers around 200 nm, with an average size of 227 ± 17 nm (>3000 endosomes were measured, from six independent experiments; see Supplemental Figure S1).
To differentiate docking from fusion, we determined the centers of intensity for all spots both in the red and green channel and then measured the distances to the closest spot of the opposite color. To calibrate the assay, we performed two imaging experiments with fluorescent beads of a size similar to endosomes (200 nm in diameter, Figure 1, D and E; see also Supplemental Figure S1). First, we used multilabeled TetraSpeck beads (Invitrogen), to simulate fused endosomes and analyzed them as above. The histogram shows a single peak (r = 80 nm). Theoretical considerations reveal that this is to be expected if 1) the localization along the two orthogonal axes are following a Gaussian distribution, with the maximum at r = 0, and 2) the normal distributions along both axes are independent of each other (Rayleigh distribution, see Appendix). Although assumption 1 is probably only an approximation, we checked assumption 2 by testing whether the direction of the vector r is random in a given field of beads. This was indeed the case (data not shown). Second, we used a mixture of beads that were either green or red (also 200 nm in diameter), and we repeated the same imaging and analysis procedure. These beads show a broad distance distribution, with no overlap with the colocalization peak (Figure 1E).
Figure 1C shows a typical histogram (average of nine experiments) for an endosome (PNS) preparation incubated for 60 min at 37°C. The distance distribution shows two peaks. The first exhibits a maximum 75 nm and thus precisely corresponds to the peak of the multicolored bead control, documenting that this peak represents vesicle populations in which the two dyes are colocalized. The second peak spans between 150–500 nm, thus revealing a population of endosomes that are not fused but closely associated with each other. We conclude that the first peak of the endosome distance distribution (Figure 1C) represents fused endosomes (in agreement with previous investigations on the subject, (Brandhorst et al., 2006; Rizzoli et al., 2006; Bethani et al., 2007), whereas the second peak represents docked endosomes, thus allowing for differentiating docking from fusion.
An alternative method for monitoring in vitro docking of organelles was previously described for yeast vacuoles. According to this procedure, clusters of organelles are differentiated from single organelles simply by addition of a fluorescent membrane probe (e.g., a styryl dye) at the end of the docking reaction, which allows for visualizing all membranes in the reaction mixture (Mayer and Wickner, 1997; Wang et al., 2002). However, when we monitored clustering of gradient-purified endosomes with this procedure, no good correlation between fusion and docking was observable (Supplemental Figure S2). Importantly, clustering was completely independent of ATP, raising doubts whether clustering as observed by this assay reports specific docking of endosomes (see Materials and Methods for details).
Basic Requirements of Endosome Docking
We first tested whether energy is required for docking. Removing ATP from the reaction mixture or incubating on ice inhibits fusion, as previously described (Gruenberg et al., 1989; Brandhorst et al., 2006) and also brings the docking value close to background. Very similar results were obtained when the endosome reaction mixtures were prepared separately and only mixed immediately before centrifugation onto coverslips (Figure 2A).
|
100 nm from each other) or docked (their intensity centers within 150–500 nm from each other). The docking value was also corrected for accidental colocalization of endosomes as described in Materials and Methods. About 6% of the endosomes were fused in our reactions, which, as expected, was reduced to background values (1%) upon ATP removal or incubation on ice (Figure 2B). Under conditions permissive for fusion, 15–20% of the endosomes were docked (but not fused). In contrast; no docking was observed in the absence of ATP and at low temperature (Figure 2C).
It has previously been shown that similar to yeast vacuoles (Peters and Mayer, 1998), fusion of early endosomes is dependent on the local release of Ca2+ ions because it is inhibited by the fast Ca2+-chelator BAPTA but not by the slower chelator EGTA (Holroyd et al., 1999), suggesting that in the presence of BAPTA the system is arrested at a docked state. As shown in Figure 4A, this is indeed the case: In the presence of BAPTA the pool of fused endosomes is virtually abolished, whereas the pool of docked vesicles is largely unaffected. As expected, the slower calcium chelator EGTA, which is known not to affect fusion in this system (Holroyd et al., 1999; Rizzoli et al., 2006) had no significant effect on docking (three independent experiments, data not shown).
Being able to prevent docked endosomes to progress to fusion allowed for investigating whether not only initiation but also maintenance of docking is dependent on ATP. We tested this by incubating the endosomes for 15 min in the presence of BAPTA. The reaction was then either 1) spun down immediately, 2) stopped by placing the tubes on ice, or 3) continued after addition of buffer, or 4) an ATP-depletion system (hexokinase solution) was added, and the reaction continued for 45 min (Figure 2D). As shown in Figure 2, E and F, the amount of docking obtained in the first 15 min in presence of ATP persisted even after 45 min of ATP depletion, indicating that ATP is only needed for initiating docking but not for its maintenance.
Next we analyzed how dilution of the endosomes affects the docking reaction. Interestingly, docking turned out to be rather insensitive to dilution up to 100x, whereas fusion was substantially reduced (Figure 3, A–C). This suggests that, although reducing the probability for endosomes to meet in the reaction tube inhibits their fusion ability, their docking ability still persists. This is in line with the hypothesis that endosomes need to meet relatively early after the start of incubation in order to proceed to fusion (Barbieri et al., 1998). We tested this by preincubating our endosomes separately for 10 min at 37°C in the reaction mixture before we combined them what yielded in a highly significant reduction of fusion but not docking (Figure 3, D–F). Apparently, the docking process is much more robust than fusion, with endosomes that meet late after the start of incubation being able to dock, but not to fuse.
|
First, we inhibited NSF by the addition of a mutant form of SNAP (-SNAP, which is unable to support ATP hydrolysis of NSF and thus functions as a dominant inhibitor of SNARE disassembly (Barnard et al., 1997). Addition of SNAP L294A (50 µM) strongly decreased fusion, but reduced docking only very mildly (Figure 4B), similar to the BAPTA effect.
|
; Zwilling et al., 2007), as competitive inhibitors. These proteins compete with the SNAREs on the early endosomal membrane for assembly and therefore reduce the amount of SNARE complex formation between the membranes destined to fuse (trans-complexes). We have shown previously that homotypic fusion of early endosomes is profoundly inhibited if a combination of the three Q-SNAREs syntaxin 13, syntaxin 6, and vti1a, is used. However, as shown in Figure 4C, docking remained basically unaltered although fusion was strongly reduced.
Finally, we added to the reaction a mixture of Fab fragments directed against all of the early endosomal SNAREs (2 µg of each). We have shown previously that SNARE-specific Fab fragments selectively block the function of SNAREs, thus providing convenient tools for interfering with specific SNAREs in in vitro reactions (Antonin et al., 2000). Again, only a very small effect (compare with Supplemental Figure S3) on docking was observed although fusion was drastically inhibited (Figure 4D).
In conclusion, transinteraction between SNAREs is mandatory for fusion, but is not required for the docking of early endosomes.
Molecular Characteristics of Endosome Docking
To gain further insight into the molecular mechanisms underlying docking of early endosomes, we addressed several proteins that are known to function in endosome recycling. First, the actin cytoskeleton does not seem to be involved in the docking process, because perturbing actin via latrunculin A (Coue et al., 1987) or phalloidin (Cooper, 1987) had no significant effects (Figure 5, A and B).
|
). EEA1 is a Rab5 effector that also needs to bind to PI3-phosphate via its FYVE domain for effective recruitment to the endosomal membrane. To interfere with EEA1-binding, we used the PI-3 kinase inhibitor wortmannin (50 nM), which resulted in a decrease of EEA1 membrane association by 70% (Figure 5C). Under these conditions, docking was strongly inhibited (60%). Fusion was also reduced but to a somewhat lower degree (Figure 5, A and B). Furthermore, incubation of endosomes with an EEA1 antiserum or with a FYVE peptide that competes with EEA1-binding to PI(3)P (McBride et al., 1999) reduced docking but had no major effects on fusion (see Discussion). The effect of the EEA1 antiserum was fully neutralized by applying the peptide against the antibody was raised (see Figure 5, A and B). Together, these data confirm that EEA1 function is instrumental for effective and/or stable docking.
We next investigated the role of Rab GTPases on docking. In the fusion of early endosomes, Rab5 has been shown to orchestrate the assembly of docking complexes and to function upstream of EEA1. We added GTPS to the reactions to increase the fraction of the active, GTP-bound conformation of Rab proteins. Surprisingly, we observed that docking was reduced rather than potentiated. This finding was explained by the fact that GTPS caused a loss of EEA1 from the membranes (Figure 5C), although the fraction of membrane-bound Rab 5 did increase. Finally, the combination of wortmannin with GTPS reduced docking even further. Thus, although GTPS increased the amount of active Rab 5, it nevertheless reduced the membrane-bound fraction of EEA1. This suggests that binding of EEA1 to Rab 5 may not be sufficient for its recruitment to the membrane in absence of other interactions (Jones et al., 1998; Simonsen et al., 1998; Lawe et al., 2002). Why membrane levels of EEA1 are reduced in the GTPS-treated reactions remains obscure, with one possible explanation being that the PI-3-kinase (hVps34) needs to be activated by a process involving a GTPase cycle that would be blocked by GTPS.
The presence of Rab proteins on the membrane is, however, necessary for docking. This became evident when we treated our PNS with 10 µM GDI, which binds to Rab proteins in the GDP-bound (inactive) form and dissociates them from the membrane. Samples were preincubated for 30 min on ice in the presence of GDI and were then either centrifuged directly or incubated for 5 min at 37°C. At the start of the reaction, i.e., after preincubation with GDI, the membrane-bound fraction of Rab 5 is already reduced by over 60% compared with the control. Under these conditions, docking was significantly inhibited. The loss of Rab 5 from the membrane was accompanied by a comparable loss of EEA1, in line with the view that Rab5 mediates docking via its effector EEA1 (Raiborg et al., 2001).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
As our assay depends on endocytotic labeling of endosomes, it only reports docking of functionally defined, bona fide early endosomes. With this approach, we avoided a significant problem associated with organelle clustering assays, where the identity of the organelles is difficult to determine (i.e., in such assays the organelles visualized may all be endosomes, but not necessarily recently endocytosed, or active, endosomes). Indeed, organelle clustering was clearly not energy-dependent, unlike fusion or bona fide docking. This finding correlates well with results from other in vitro systems, where clustering only shows an ATP dependence when the assay conditions are substantially changed from those of a normal fusion reaction (e.g., through a strong reduction of the salt concentration, Mayer and Wickner, 1997).
Interestingly, our data also show that fusion and docking do not correlate perfectly. Although only 6% of the endosomes have fused after 60 min, about three times more remain in the docked state (see Figures 2
–4 and Supplemental Figure S3) and are apparently unable to proceed further to fusion, as fusion has already largely ceased at this point (data not shown and Brandhorst et al., 2006). In addition, some factors like wortmannin and the FYVE peptide inhibited docking much strongly than fusion (see Figure 5, A and B), indicating that not all endosomes that are able to dock are equally able to fuse. Conversely, fusion was reduced by dilution, but docking was not (see Figure 3, A–C).
The dilution results indicate that in these conditions the endosomes were able to meet and dock, but were no longer able to proceed to fusion. This correlates well with the observation that early endosomes are less fusogenic after separate preincubation at 37°C (Barbieri et al., 1998 and unpublished observations), which suggests that the endosomes are primed for fusion at the start of incubation, and then must meet and fuse within a limited amount of time. Dilution (or separate incubation; see Figure 3, D–F) prolongs this period, and results in nonfusogenic endosomes, a hypothesis first put forward in the yeast vacuole system (Mayer and Wickner, 1997; Xu et al., 1997). This also provides a relatively simple explanation for the substantial amount of docked, but not fused endosomes: they would be organelles that met too late in the reaction and were beyond the point of fusion.
The specificity of the docking process is underlined by the fact that although the labeled endosomes constitute only a very small fraction of all organelles (data not shown), a substantial percentage (20–25%) still manage to find each other, even when they are strongly diluted (Figure 3). This is all the more remarkable when one considers that our assay only measures the amount of green endosomes interacting with red ones, while ignoring the interactions between endosomes labeled with the same dye. Including these interactions would bring the fraction of endosomes fusing or docking to each other to 60–75%. Furthermore, we only measure one round of docking/fusion in our assay (an endosome fusing or docking onto a green-red pair would not be accounted for), and thus even this estimate is likely lower than the total fraction of active organelles.
These observations are in line with the current understanding of docking/tethering (Sztul and Lupashin, 2006; Cai et al., 2007) as a process mediated by an interplay of small GTPases like Rabs and long tethering molecules or mulitsubunit complexes. Indeed, the results from our assay suggest that fusion, but not docking, required SNARE function. Interestingly, our findings differ somewhat from those in one in vitro system where fusion has been thoroughly investigated, the yeast vacuole (Wickner, 2002). Although the differences between the systems may explain the different results (with the yeast vacuole corresponding to the late endosome/lysosome of the mammalian cell, and not to the early endosome), it is also likely that a number of experimental details contribute to the different interpretations.
In vitro docking of yeast vacuoles was suggested to depend on transinteractions between SNARE proteins mainly because 1) transinteraction between SNAREs was observed when fusion was inhibited by late stage (i.e., post docking) inhibitors such as microcystin-LR (Ungermann et al., 1998), 2) vacuole clustering depends on ATP/NSF activity (Mayer and Wickner, 1997), and 3) one may be able to disengage putative trans-SNARE complexes by excess NSF without influencing fusion (Ungermann et al., 1998, 1999; Peters et al., 2001). However, as suggested above, some of these results may depend strongly on the experimental conditions. When a new fusion assay was used, microcystin-LR did not seem to inhibit vacuole fusion, but rather the activation/activity of alkaline phosphatase (ALP), which was used as a marker for vacuolar fusion in the original assays (Mayer and Wickner, 1997; Jun and Wickner, 2007). Also, in one report vacuole clustering was shown to be SNARE independent (Wang et al., 2003), and vacuole clustering also seems to be independent of ATP under buffer conditions used for the fusion reaction (Jun and Wickner, 2007). Finally, it is still debated whether the NSF-sensitive SNARE complexes monitored in the work mentioned were bona fide trans-complexes or rather cis-complexes formed in the (previously) fused organelles.
Although our data show that endosome docking is not affected by interference with SNARE function, we do not believe that there is no functional cross-talk between the molecular machinery involved in docking and SNARE proteins. Several lines of evidence indicate that docking proteins directly bind to SNAREs. For instance, EEA1 has been shown to interact both with syntaxin 6 (Mills et al., 2001) and syntaxin 13 (McBride et al., 1999), i.e., two of the SNAREs involved in the fusion of early endosomes (Brandhorst et al., 2006). It is conceivable that such binding plays a role in regulating the transition from tethering to fusion (Sztul and Lupashin, 2006). Furthermore, we cannot exclude that SNAREs are involved in stabilizing and maintaining a docked state. This may be particularly relevant in cases where progress from docking toward fusion does not occur constitutively (as is the case for endosome fusion) but is regulated, resulting in prolonged and reversible docked states (as in synapses: Hammarlund et al., 2007, 2008) or in certain neuroendocrine cells).
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
):
 |
We assume (as experimentally confirmed) that Px and Py are uncorrelated. In this case, the probability distribution of the distance r = 
is described by a Rayleigh distribution
 |
(the value where the first derivative = 0):
|
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Address correspondence to: Reinhard Jahn (rjahn{at}gwdg.de)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
Barbieri, M. A., Hoffenberg, S., Roberts, R., Mukhopadhyay, A., Pomrehn, A., Dickey, B. F., and Stahl, P. D. (1998). Evidence for a symmetrical requirement for Rab5-GTP in in vitro endosome-endosome fusion. J. Biol. Chem 273, 25850–25855.
Barnard, R. J., Morgan, A., and Burgoyne, R. D. (1997). Stimulation of NSF ATPase activity by alpha-SNAP is required for SNARE complex disassembly and exocytosis. J. Cell Biol 139, 875–883.
Bethani, I., Lang, T., Geumann, U., Sieber, J. J., Jahn, R., and Rizzoli, S. O. (2007). The specificity of SNARE pairing in biological membranes is mediated by both proof-reading and spatial segregation. EMBO J 26, 3981–3992.[CrossRef][Medline]
Brandhorst, D., Zwilling, D., Rizzoli, S. O., Lippert, U., Lang, T., and Jahn, R. (2006). Homotypic fusion of early endosomes: SNAREs do not determine fusion specificity. Proc. Natl. Acad. Sci. USA 103, 2701–2706.
Broadie, K., Prokop, A., Bellen, H. J., O'Kane, C. J., Schulze, K. L., and Sweeney, S. T. (1995). Syntaxin and synaptobrevin function downstream of vesicle docking in Drosophila. Neuron 15, 663–673.[CrossRef][Medline]
Cai, H., Reinisch, K., and Ferro-Novick, S. (2007). Coats, tethers, Rabs, and SNAREs work together to mediate the intracellular destination of a transport vesicle. Dev. Cell 12, 671–682.[CrossRef][Medline]
Callaghan, J., Simonsen, A., Gaullier, J. M., Toh, B. H., and Stenmark, H. (1999). The endosome fusion regulator early-endosomal autoantigen 1 (EEA1) is a dimer. Biochem. J 338, (Pt 2), 539–543.[CrossRef][Medline]
Christoforidis, S., McBride, H. M., Burgoyne, R. D., and Zerial, M. (1999a). The Rab5 effector EEA1 is a core component of endosome docking. Nature 397, 621–625.[CrossRef][Medline]
Christoforidis, S., Miaczynska, M., Ashman, K., Wilm, M., Zhao, L., Yip, S. C., Waterfield, M. D., Backer, J. M., and Zerial, M. (1999b). Phosphatidylinositol-3-OH kinases are Rab5 effectors. Nat. Cell Biol 1, 249–252.[CrossRef][Medline]
Cooper, J. A. (1987). Effects of cytochalasin and phalloidin on actin. J. Cell Biol 105, 1473–1478.
Coue, M., Brenner, S. L., Spector, I., and Korn, E. D. (1987). Inhibition of actin polymerization by latrunculin A. FEBS Lett 213, 316–318.[CrossRef][Medline]
de Wit, H., Cornelisse, L. N., Toonen, R. F., and Verhage, M. (2006). Docking of secretory vesicles is syntaxin dependent. PLoS ONE 1, e126.[CrossRef]
Edelmann, L., Hanson, P. I., Chapman, E. R., and Jahn, R. (1995). Synaptobrevin binding to synaptophysin: a potential mechanism for controlling the exocytotic fusion machine. EMBO J 14, 224–231.[Medline]
Fischer von Mollard, G., Stahl, B., Walch-Solimena, C., Takei, K., Daniels, L., Khoklatchev, A., De Camilli, P., Sudhof, T. C., and Jahn, R. (1994). Localization of Rab5 to synaptic vesicles identifies endosomal intermediate in synaptic vesicle recycling pathway. Eur. J. Cell Biol 65, 319–326.[Medline]
Fujiwara, T., Mishima, T., Kofuji, T., Chiba, T., Tanaka, K., Yamamoto, A., and Akagawa, K. (2006). Analysis of knock-out mice to determine the role of HPC-1/syntaxin 1A in expressing synaptic plasticity. J. Neurosci 26, 5767–5776.
Gruenberg, J., Griffiths, G., and Howell, K. E. (1989). Characterization of the early endosome and putative endocytic carrier vesicles in vivo and with an assay of vesicle fusion in vitro. J. Cell Biol 108, 1301–1316.
Gruenberg, J., and Howell, K. E. (1987). An internalized transmembrane protein resides in a fusion-competent endosome for less than 5 minutes. Proc. Natl. Acad. Sci. USA 84, 5758–5762.
Hammarlund, M., Palfreyman, M. T., Watanabe, S., Olsen, S., and Jorgensen, E. M. (2007). Open syntaxin docks synaptic vesicles. PLoS Biol 5, e198.[CrossRef][Medline]
Hammarlund, M., Watanabe, S., Schuske, K., and Jorgensen, E. M. (2009). CAPS and syntaxin dock dense core vesicles to the plasma membrane in neurons. J. Cell Biol 180, 483–491.[CrossRef]
Heumann, R., Kachel, V., and Thoenen, H. (1983). Relationship between NGF-mediated volume increase and "priming effect" in fast and slow reacting clones of PC12 pheochromocytoma cells. Role of cAMP. Exp. Cell Res 145, 179–190.
Holroyd, C., Kistner, U., Annaert, W., and Jahn, R. (1999). Fusion of endosomes involved in synaptic vesicle recycling. Mol. Biol. Cell 10, 3035–3044.
Huttner, W. B., Schiebler, W., Greengard, P., and De Camilli, P. (1983). Synapsin I (protein I), a nerve terminal-specific phosphoprotein. III. Its association with synaptic vesicles studied in a highly purified synaptic vesicle preparation. J. Cell Biol 96, 1374–1388.
Jahn, R., and Scheller, R. H. (2006). SNAREs—engines for membrane fusion. Nat. Rev. Mol. Cell. Biol 7, 631–643.[CrossRef][Medline]
Jahn, R., Schiebler, W., Ouimet, C., and Greengard, P. (1985). A 38,000-dalton membrane protein (p38) present in synaptic vesicles. Proc. Natl. Acad. Sci. USA 82, 4137–4141.
Jones, A. T., Mills, I. G., Scheidig, A. J., Alexandrov, K., and Clague, M. J. (1998). Inhibition of endosome fusion by wortmannin persists in the presence of activated Rab5. Mol. Biol. Cell 9, 323–332.
Jun, Y., and Wickner, W. (2007). Assays of vacuole fusion resolve the stages of docking, lipid mixing, and content mixing. Proc. Natl. Acad. Sci. USA 104, 13010–13015.
Kawasaki, F., Mattiuz, A. M., and Ordway, R. W. (1998). Synaptic physiology and ultrastructure in comatose mutants define an in vivo role for NSF in neurotransmitter release. J. Neurosci 18, 10241–10249.
Klenchin, V. A., and Martin, T. F. (2000). Priming in exocytosis: attaining fusion-competence after vesicle docking. Biochimie 82, 399–407.
Laage, R., and Ungermann, C. (2001). The N-terminal domain of the t-SNARE Vam3p coordinates priming and docking in yeast vacuole fusion. Mol. Biol. Cell 12, 3375–3385.
Lawe, D. C., Chawla, A., Merithew, E., Dumas, J., Carrington, W., Fogarty, K., Lifshitz, L., Tuft, R., Lambright, D., and Corvera, S. (2002). Sequential roles for phosphatidylinositol 3-phosphate and Rab5 in tethering and fusion of early endosomes via their interaction with EEA1. J. Biol. Chem 277, 8611–8617.
Lippe, R., Miaczynska, M., Rybin, V., Runge, A., and Zerial, M. (2001). Functional synergy between Rab5 effector Rabaptin-5 and exchange factor Rabex-5 when physically associated in a complex. Mol. Biol. Cell 12, 2219–2228.
Maxfield, F. R., and McGraw, T. E. (2004). Endocytic recycling. Nat. Rev. Mol. Cell Biol 5, 121–132.[CrossRef][Medline]
Mayer, A., and Wickner, W. (1997). Docking of yeast vacuoles is catalyzed by the Ras-like GTPase Ypt7p after symmetric priming by Sec18p (NSF). J. Cell Biol 136, 307–317.
McBride, H. M., Rybin, V., Murphy, C., Giner, A., Teasdale, R., and Zerial, M. (1999). Oligomeric complexes link Rab5 effectors with NSF and drive membrane fusion via interactions between EEA1 and syntaxin 13. Cell 98, 377–386.[CrossRef][Medline]
Mills, I. G., Urbe, S., and Clague, M. J. (2001). Relationships between EEA1 binding partners and their role in endosome fusion. J. Cell Sci 114, 1959–1965.[Abstract]
Mu, F. T., Callaghan, J. M., Steele-Mortimer, O., Stenmark, H., Parton, R. G., Campbell, P. L., McCluskey, J., Yeo, J. P., Tock, E. P., and Toh, B. H. (1995). EEA1, an early endosome-associated protein. EEA1 is a conserved alpha-helical peripheral membrane protein flanked by cysteine "fingers" and contains a calmodulin-binding IQ motif. J. Biol. Chem 270, 13503–13511.
O'Connor, V. et al. (1997). Disruption of syntaxin-mediated protein interactions blocks neurotransmitter secretion. Proc. Natl. Acad. Sci. USA 94, 12186–12191.
Patki, V., Virbasius, J., Lane, W. S., Toh, B. H., Shpetner, H. S., and Corvera, S. (1997). Identification of an early endosomal protein regulated by phosphatidylinositol 3-kinase. Proc. Natl. Acad. Sci. USA 94, 7326–7330.
Peters, C., Bayer, M. J., Buhler, S., Andersen, J. S., Mann, M., and Mayer, A. (2001). Trans-complex formation by proteolipid channels in the terminal phase of membrane fusion. Nature 409, 581–588.[CrossRef][Medline]
Peters, C., and Mayer, A. (1998). Ca2+/calmodulin signals the completion of docking and triggers a late step of vacuole fusion. Nature 396, 575–580.
Raiborg, C., Bremnes, B., Mehlum, A., Gillooly, D. J., D'Arrigo, A., Stang, E., and Stenmark, H. (2001). FYVE and coiled-coil domains determine the specific localisation of Hrs to early endosomes. J. Cell Sci 114, 2255–2263.[Medline]
Rizzoli, S. O., Bethani, I., Zwilling, D., Wenzel, D., Siddiqui, T. J., Brandhorst, D., and Jahn, R. (2006). Evidence for early endosome-like fusion of recently endocytosed synaptic vesicles. Traffic 7, 1163–1176.[CrossRef][Medline]
Schikorski, T., and Stevens, C. F. (2001). Morphological correlates of functionally defined synaptic vesicle populations. Nat. Neurosci 4, 391–395.[CrossRef][Medline]
Simonsen, A., Lippe, R., Christoforidis, S., Gaullier, J. M., Brech, A., Callaghan, J., Toh, B. H., Murphy, C., Zerial, M., and Stenmark, H. (1998). EEA1 links PI(3)K function to Rab5 regulation of endosome fusion. Nature 394, 494–498.[CrossRef][Medline]
Stenmark, H., Vitale, G., Ullrich, O., and Zerial, M. (1995). Rabaptin-5 is a direct effector of the small GTPase Rab5 in endocytic membrane fusion. Cell 83, 423–432.[CrossRef][Medline]
Sztul, E., and Lupashin, V. (2006). Role of tethering factors in secretory membrane traffic. Am J. Physiol. Cell Physiol 290, C11–C26.
Takamori, S. et al. (2006). Molecular anatomy of a trafficking organelle. Cell 127, 831–846.[CrossRef][Medline]
Ungermann, C., Sato, K., and Wickner, W. (1998). Defining the functions of trans-SNARE pairs. Nature 396, 543–548.
Ungermann, C., Wickner, W., and Xu, Z. (1999). Vacuole acidification is required for trans-SNARE pairing, LMA1 release, and homotypic fusion. Proc. Natl. Acad. Sci. USA 96, 11194–11199.
Wang, L., Merz, A. J., Collins, K. M., and Wickner, W. (2003). Hierarchy of protein assembly at the vertex ring domain for yeast vacuole docking and fusion. J. Cell Biol 160, 365–374.
Wang, L., Seeley, E. S., Wickner, W., and Merz, A. J. (2002). Vacuole fusion at a ring of vertex docking sites leaves membrane fragments within the organelle. Cell 108, 357–369.[CrossRef][Medline]
Waters, M. G., and Hughson, F. M. (2000). Membrane tethering and fusion in the secretory and endocytic pathways. Traffic 1, 588–597.[CrossRef][Medline]
Wickner, W. (2002). Yeast vacuoles and membrane fusion pathways. EMBO J 21, 1241–1247.[CrossRef][Medline]
Wickner, W., and Haas, A. (2000). Yeast homotypic vacuole fusion: a window on organelle trafficking mechanisms. Annu. Rev. Biochem 69, 247–275.[CrossRef][Medline]
Willig, K. I., Rizzoli, S. O., Westphal, V., Jahn, R., and Hell, S. W. (2006). STED microscopy reveals that synaptotagmin remains clustered after synaptic vesicle exocytosis. Nature 440, 935–939.
Xu, Z., Mayer, A., Muller, E., and Wickner, W. (1997). A heterodimer of thioredoxin and I(B)2 cooperates with Sec18p (NSF) to promote yeast vacuole inheritance. J. Cell Biol 136, 299–306.
Zwilling, D., Cypionka, A., Pohl, W. H., Fasshauer, D., Walla, P. J., Wahl, M. C., and Jahn, R. (2007). Early endosomal SNAREs form a structurally conserved SNARE complex and fuse liposomes with multiple topologies. EMBO J 26, 9–18.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  S. V. Barysch, S. Aggarwal, R. Jahn, and S. O. Rizzoli Sorting in early endosomes reveals connections to docking- and fusion-associated factors PNAS, June 16, 2009; 106(24): 9697 - 9702. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
