Author's personal copy
FEBS Letters 582 (2008) 1558–1563
The vacuolar V1/V0-ATPase is involved in the release of the
HOPS subunit Vps41 from vacuoles, vacuole fragmentation and fusion
Kozue Takedaa, Margarita Cabreraa, Jan Rohdec, Dirk Bauschb, Ole N. Jensend,
Christian Ungermanna,*
a
Department of Biology, University of Osnabrück, Barbarastrasse 13, 49076 Osnabrück, Germany
b
Universitätsklinikum Freiburg, Freiburg, Germany
c
Rhein-Minapharm Biogenetics, 10th of Ramadan City, Egypt
d
Department of Biochemistry and Molecular Biology, University of Odense, Odense M, Denmark
Received 7 December 2007; revised 21 March 2008; accepted 28 March 2008
Available online 9 April 2008
Edited by Felix Wieland
Abstract At yeast vacuoles, phosphorylation of the HOPS subunit Vps41 depends on the Yck3 kinase. In a screen for mutants
that mimic the yck3D phenotype, in which Vps41 accumulates in
vacuolar dots, we observed that mutants in the V0-part of the V0/
V1-ATPase, in particular in vma16D, also accumulate Vps41.
This accumulation is not due to a phosphorylation defect, but
to reduced release of Vps41 from vma16D vacuoles. One reason
could be a connection to vacuole fission, which is blocked in
V-ATPase mutants. Vacuole fusion is not impaired between vacuoles lacking the V0-subunits Vma16 or Vma6 and wild-type
vacuoles, whereas fusion between mutant vacuoles is reduced.
Our data suggest a connection between vacuole biogenesis and
membrane fusion.
2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
Keywords: Yck3; Vps41; V-ATPase; Fusion; Vacuole
1. Introduction
Organelle biogenesis in eukaryotic cells is controlled by regulated fusion and fission events. Fusion depends on Rab-GTPases, tethers and SNAREs, while fission relies, among other
factors, on coat proteins [1,2]. The molecular mechanisms of
these reactions have been addressed with model systems like
the yeast vacuole. Fusion of yeast vacuoles occurs in a regulated cascade and can be monitored in vivo and in vitro [3].
It requires the consecutive function of the Rab GTPase Ypt7
and the hexameric HOPS complex to mediate tethering, followed by SNARE-dependent fusion. A number of additional
factors have been implicated in the fusion reaction, including
the membrane-embedded V0 part of the vacuolar V0/V1-ATPase and the fusion factor Vac8 [4,5].
Vacuole fusion has to be controlled during the cell cycle,
where vacuolar fragments are transported to the daughter cell,
and in response to osmotic stress, which leads to vacuole fragmentation [6,7]. A common mechanism of regulation is phosphorylation, and one kinase, the casein kinase Yck3, has
been linked to vacuole fusion [8]. Yck3 is a palmitoylated pro*
Corresponding author. Fax: +49 541 969 2884.
E-mail address: [email protected]
(C. Ungermann).
tein that is sorted to vacuoles via the AP-3 pathway [9]. This
pathway transports vesicular cargo directly from the Golgi
to the vacuole, thus bypassing the endosome [10].
We identified Yck3 as a kinase involved in the modification
of Vps41 [8]. Vps41 is a vacuole-specific subunit of the sixmembered HOPS complex, which is involved in vacuole fusion
and biogenesis [11,12]. Vps41 is not only required as a HOPS
subunit, but has also been discussed as part of the coat of AP-3
vesicles [13,14]. Several lines of evidence connect Yck3 and
Vps41 in vacuole biogenesis. In mutants lacking Yck3,
Vps41 is not phosphorylated and strongly accumulates in
dot-like structures on and between vacuoles in vivo [8]. If challenged by osmotic stress, yck3D vacuoles fragment like wildtype vacuoles, but do not maintain this phenotype. This suggests that Yck3 suppresses fusion in wild-type cells under these
conditions. Overexpression of Yck3 causes a reduction in fusion, while deletion of YCK3 results in slightly increased fusion. Interestingly, yck3D vacuoles do not efficiently release
Vps41 from their surface, and fusion is strongly resistant to
the Ypt7 inhibitor Gdi1. As Ypt7 is binding to HOPS [12],
and thus also to Vps41, it is likely that phosphorylation of
Vps41 is responsible for the mobility of the HOPS complex
in vivo and in vitro. Yck3 therefore seems to be as a negative
regulator of fusion [8].
Yck3 itself is phosphorylated [8] (K.T., M.C., unpublished
observations), which may be due to its activation in the cell.
Such an activation would be expected during osmotic stress,
when Vps41 is phosphorylated [8]. If activation of Yck3 is a prerequisite of its function, then mutants that accumulate Vps41
proximal to the vacuole might be involved in the Yck3 regulation network. Here we report that this phenotype is observed
for mutants in the V0-ATPase. In analyzing these mutants for
yck3D-like phenotypes we observe that Vps41 release, but not
its phosphorylation is impaired. Additional analysis of mutants
in the V0-ATPase revealed a defect in vacuole fission. We finally
analyzed in vitro fusion and observed that there is not symmetric
requirement of the V0-ATPase for fusion.
2. Materials and methods
2.1. Yeast strains and molecular biology
BY4741 deletion yeast strains were purchased from EUROSCARF.
For Vps41 localization screening, pRS415.NOP1pr-GFP-Vps41 [8]
was introduced into BY4741 wt, yck3D, and each BY4741 deletion
mutant and cells were maintained in SDC-Leu synthetic medium.
0014-5793/$34.00 2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.febslet.2008.03.055
Author's personal copy
K. Takeda et al. / FEBS Letters 582 (2008) 1558–1563
BY4741 pep4D yck3D and BY4741 pho8D yck3D have been described
previously [8]. VMA16 was deleted by homologous recombination
using PCR fragments. For constructing GFP-YCK3 strains, the
YCK3 ORF was cut with Salt–Not1 from pRS416.NOP1-GFPYCK3 [8], and was inserted into the pRS406.NOP1pr-GFP plasmid.
All strains carrying tagged proteins were confirmed by PCR and/or
Western blotting with antibodies against the altered products.
2.2. Antibodies
Polyclonal antibodies against Sec17, Ypt7, Nyv1 and Vam3 have
been previously described [12,15,16,23]. Antibodies against Vma6 were
raised in rabbits against the GST-tagged protein. Antibodies to Pep4
and Vph1 were kindly provided by W. Wickner (Dartmouth Medical
School, NH, USA) and A. Mayer (University of Lausanne, CH).
Mouse monoclonal antibodies against Vma1 and Pho8 were from
Invitrogen Inc. IgGs were prepared as previously described [15].
2.3. Microscopy
Staining of cells with FM4-64 was performed as described [8,27].
Yeast cells were grown to logarithmic phase, incubated with 30 lM
FM4-64 for 30 min, washed with medium, and chased for 1 h. For
GFP microscopy, cells were grown logarithmically in YPD or selective
medium, washed, resuspended in fresh medium, and analyzed by fluorescence microscopy. Images were acquired with a Leica DM5500
microscope and a SPOT Pursuit camera using GFP, FM4-64 or DIC
filters. Pictures were processed using Adobe Photoshop 7.0. Figures
show representative fields from multiple experiments.
2.4. Yeast cell lysis
Total protein extraction and subcellular fractionation from yeast were
done as described previously [8]. To analyze protein content of yeast
cells, cells were lysed in 0.24 N NaOH, 130 mM b-mercaptoethanol,
1.3 mM PMSF, followed by TCA (13% v/v) precipitation, and acetone
wash; equal amounts were analyzed by Western blot. For subcellular
fractionation, cultures (20 ml) were grown to OD600 = 1–1.5 in YPD.
Until lysis, cells were treated as for the vacuole purification, and adjusted
relative to the smaller culture size. Spheroplasts were resuspended in 1 ml
of lysis buffer (200 mM sorbitol, 50 mM KOAc, 2 mM EDTA, 20 mM
HEPES-KOH pH 6.8, 1· PIC [17], 1 mM PMSF, 1 mM DTT), and cells
were osmotically lysed in DEAE-dextran/HCl. The total extract was centrifuged at 300 · g, 3 min, 4 C, and the supernatant was then centrifuged for 15 min at 13 000 · g. The pellets (P13) were collected and
diluted to 0.3 mg/ml in PS buffer (20 mM PIPES/KOH, 200 mM sorbitol) containing 0.1· PIC.
2.5. Vps41 phosphorylation and release assay
P13 membrane fractions or isolated vacuoles were incubated in
150 ll reactions with or without ATP under standard fusion conditions
for 45 min supplemented with salts ([18]; 0.5 mM MgCl2, 0.5 mM
MnCl2, 150 mM KCl). Membranes were reisolated for 10 min at
20 000 · g and 4 C, and pellets were resuspended in PS buffer. Resuspended pellets and supernatants were TCA-precipitated (13% v/v),
washed with acetone and resuspended in SDS sample buffer. Vps41
was analyzed by Western blotting.
2.6. Vacuole fusion
Vacuoles were incubated at 26 C in PS buffer supplemented with
salts (5 mM MgCl2, 125 mM KCl), CoA (10 lM), with or without
an ATP-regenerating system (0.5 mM ATP, 40 mM creatine phosphate, 0.1 mg/ml creatine kinase). Vacuole fusion was measured by a
biochemical complementation assay as described previously [17]. Fusion was determined by measuring the absorbance of at least duplicate
samples at 400 nm. The results are expressed by subtracting the signal
from an incubation lacking ATP. (1 U of fusion activity is defined as
1 lmole p-nitrophenol developed per min and lg pep4 vacuoles.)
3. Results
3.1. Identification of mutants with accumulated Vps41
A deletion in the kinase yck3 results in a strong in vivo accumulation of GFP-tagged Vps41 proximal to the vacuole, which
1559
is most likely due to a lack of phosphorylation. As Yck3 is itself phosphorylated, and might require phosphorylation for its
activity, we expected that the deletion of proteins involved in
Yck3 activation would result in a similar Vps41 localization
proximal to the vacuole. Deletion of yck3 results in an enlarged vacuole phenotype, which is reminiscent to mutants in
class D proteins involved in endosome-vacuole transport
[8,19]. We therefore screened mutants with a similar phenotype
that were implicated in vacuole biogenesis in previous screens
(Suppl. Table 1) [8,19,20]. To do so, we transformed the strains
with GFP-tagged Vps41 and monitored its in vivo distribution
by fluorescence microscopy. When compared to the yck3D mutant, only mutants in the V0-subunits Vma16 and Vma2 of the
vacuolar V1/V0-ATPase showed a phenotype of dot-like accumulation of GFP-Vps41. The accumulation was not as striking
as observed in the yck3D mutants, but was clearly detectable
(Fig. 1B, Suppl. Fig. 1). Moreover, addition of the V-ATPase
pump activity inhibitor concanamycin A lead to a clear accumulation of GFP-Vps41, which was not as pronounced as in
vma16D mutant cells (Fig. 1C). Mutants in vph1, the a-subunit
of the V0-part, also showed an increased dot-like accumulation
of Vps41. However, it cannot be excluded that this positive
score for vph1D mutants is in part due to their fragmented vacuoles [21]. We confirmed that our V-ATPase mutants showed
the previously described pH-dependent growth phenotype
(Fig. 2A), and focused in further experiments primarily on
the vma16D mutant.
3.2. vma16 mutants permit Vps41 phosphorylation, but show
impaired release from vacuoles
We decided to test the vma16D mutant for phenotypes described for yck3D. First, we analyzed the response of the
vma16D mutant to osmotic stress. In wild-type cells, vacuoles
fragment during osmotic stress and maintain this phenotype
for at least 60 min. We previously showed that vacuoles in
yck3 mutants fragment if stressed, but refuse to a large vacuole
by 60 min (Fig. 2B, Suppl. Fig. 2) [8]. In contrast, vacuoles in
vam16D cells remain large and do not fragment under osmotic
stress (Fig. 2B). Similar phenotypes were observed for other VATPase mutants, consistent with recent observations by
Mayer and colleagues [22]. Second, we analyzed if Vps41 is still
phosphorylated in vma16 mutants. Membrane fractions were
isolated from wild-type, vma16D and yck3D cells and incubated with ATP (Fig. 3A). Under these conditions, Vps41
shifts to a form of lower mobility in SDS–PAGE gels, which
represents the phosphorylated protein. Whereas yck3 mutants
were unable to phosphorylate Vps41, vma16 mutants did not
differ from wild-type. Consistent with this, Yck3 was found
on vacuoles in vma16D cells (Fig. 3B). As phosphorylation of
Vps41 is linked to its release from the vacuolar membrane,
we analyzed in a third assay whether Vps41 is still released
from vma16D vacuoles. Whereas Vps41 release was clearly observed from wild-type vacuoles, it was diminished in yck3D [8]
and vma16D cells, despite the fact that Vps41 was phosphorylated on vma16D membranes (Fig. 3C and D). The release does
not seem to be linked exclusively to the pump activity of the VATPase as the addition of concanamycin A did not affect this
reaction (not shown). In contrast, the ATP-dependent release
of Sec17 from vacuoles, which occurs as a consequence of
SNARE complex disassembly on vacuoles, was observed from
all membranes with equal efficiency (Fig. 3C). The accumulation of Vps41 dots proximal to the vacuole in vma16D cells
Author's personal copy
1560
K. Takeda et al. / FEBS Letters 582 (2008) 1558–1563
Fig. 1. Mutant analysis on GFP-Vps41 localization in vivo. (A) Screen for membrane associated GFP-Vps41. BY wild-type yeast and indicated
mutant cells expressing GFP-tagged Vps41 were grown logarithmically in SDC-Leu medium, and analyzed by fluorescence microscopy. For each
mutant the number of cells, which had a dot-like accumulation of GFP-Vps41, was counted and the relative cell number is represented as the
percentage to the total cells examined. Error bars are standard deviation (S.D.). *Dunnett test, P < 0.05 compared with wt. (B) GFP-Vps41
localization in wt, yck3D and indicated mutants. Size bar = 5 lm. (C) Inhibition of V-ATPase pump activity affects GFP-Vps41 localization in vivo.
BY wild-type cells containing GFP-Vps41 were treated with 1 lM concanamycin A or DMSO, and visualized by fluorescence microscopy after 1, 2
and 3 h. Only the 3-h time point is shown. As a comparison, vma16D cells treated in parallel are displayed. Size bar = 5 lm.
can therefore be explained by its slow release from vacuole
membranes.
3.3. Vacuole fusion of V-ATPase mutants
The deletion of yck3 does not inhibit but rather enhances
vacuole fusion [8]. We therefore asked if vma16 mutants are affected in this assay. Vacuole fusion can be monitored in a content mixing assay using two types of vacuoles. One set of
vacuoles lacks Pho8, whereas the other is devoid of the Pho8
processing protease Pep4 and thus accumulates the inactive
pro-form. Fusion of vacuoles leads to processing of proPho8 to its active form, which can then by assayed. We generated pep4 and pho8 mutants in our vma16D background
strains. Vacuoles lacking Vma16 contain wild-type like
amounts of the SNARE Vam3 or the Rab GTPase Ypt7, as
well as Pho8 (in pep4D) and Pep4 (in pho8D) (Fig. 4A). How-
Author's personal copy
K. Takeda et al. / FEBS Letters 582 (2008) 1558–1563
Fig. 2. V-ATPase mutants are defective in vacuole fragmentation. (A)
Growth of wild-type and mutant strains at pH 7.5. Logarithmically
growing BY wild-type yeast strain and the indicated deletion strains
were transferred to YPD plates buffered to pH 5.5 or buffered to pH
7.5. The undiluted culture is shown at the left along with 10-fold serial
dilutions. (B) Vacuole salt stress response of BY wild-type, vma16D
and yck3D cells. Yeast cells were grown to logarithmically and stained
with FM4-64 as described in experimental procedures. After 10 or
60 min of incubation in YPD + 0.4 M NaCl, cells were immediately
analyzed by fluorescence microscopy.
ever, we could not detect the V0-subunits Vma6 (d) and Vph1
(a), as well as the V1-subunit Vma1 (A) on vma16D vacuoles,
whereas they were present on yck3D and wild-type vacuoles.
1561
Fig. 3. Inefficient Vps41 release from vma16D vacuoles. (A) Vps41
mobility shift in vitro. Isolated vacuoles from the indicated strains
were incubated in 150 ll reactions with or without ATP under
standard fusion conditions for 30 min. Membranes were reisolated
and Vps41 was analyzed by Western blotting. (B) Localization of
GFP-Yck3 in BY WT and vma16D cells. Yeast strains expressing GFPYck3 were analyzed by fluorescence microscopy. (C) Release of Vps41
from the vacuole. BY pep4D (wt), BY pep4D vma16D and BY pep4D
yck3D vacuoles were incubated in a Vps41 phosphorylation assay for
30 min. Proteins in pellet and supernatant were detected by Western
blot with antibodies against the indicated proteins. (D) Quantification
of experiment in (C). The signal was quantified by densitometric
scanning. All values were derived by using the Odyssey LiCor system.
The graph shows a percent release of Vps41 protein (errors are given as
S.D., n = 3).
Vacuoles isolated from tester strains lacking Yck3 fused efficiently with wild-type vacuoles, with pho8D vacuoles being
more fusion competent than pep4D vacuoles (Fig. 4B). When
both vacuoles were from yck3D cells, fusion was slightly increased when compared to wild-type (Fig. 4B), as previously
reported [8]. Vacuoles isolated from vma16D cells showed efficient fusion if combined with wild-type vacuoles. To our surprise, we could not detect any fusion if Vma16 was absent
Author's personal copy
1562
K. Takeda et al. / FEBS Letters 582 (2008) 1558–1563
Fig. 4. Vacuole fusion analysis of vma6 and vma16 mutants. (A) Vacuolar protein content on vma16D vacuoles. Vacuoles used in B were analyzed by
Western blot. (B) vma16D vacuole fusion. The averaged results from BY4741 wild-type (pep4D or pho8D), vma16D (vma16D pep4D or vma16D pho8D)
and yck3D (yck3D pep4D or yck3D pho8D) vacuole fusion assays were compared. wt (pep4D · pho8D) 100%: corresponds to an average of 1.0 ALP
units (OD400 = 0.25–0.3 at a background activity of 0.05; error are S.D., n = 3).
from both vacuoles. This asymmetric fusion phenotype has not
been observed to this degree in any previous mutant. We therefore asked if this is a general phenotype for V-ATPase mutants. Mutants in the d-subunit of the V0-part, Vma6, show
a similar accumulation of GFP-Vps41 in vivo (Suppl.
Fig. 1A). Similar to vma16D vacuoles, vacuoles that lack
Vma6 showed significant fusion when combined with wild-type
vacuoles, but poor fusion if Vam6 was lacking from both partners (Suppl. Fig. 3A). Of note, subunits of the V0-part and
Vtc4 can be isolated with vacuolar SNAREs, and antibodies
to Vma6 immunoprecipitates vacuolar SNAREs (Suppl.
Fig. 3B and C). This suggests that the V-ATPase is linked to
the SNARE machinery.
In sum, mutants in the V0-ATPase affect the mobility of
Vps41, but not its phosphorylation, and – in contrast to yck3D
mutants – show a strong vacuole fusion defect if the V-ATPase
is defective on both membranes.
4. Discussion
Our data reveal novel aspects of Vps41 dynamics at the yeast
vacuole. We report the identification of mutants in the V0-part of
the V1/V0-ATPase that promote a dot-like localization of Vps41
close to the vacuole. This phenotype is reminiscent of Vps41Õs
localization in the yck3D mutant. We assume that these dots represent fusion competent regions as Vps41 is also enriched between apposed vacuoles in vivo [8]. The localization of Vps41
is explained by its reduced release from vma16D vacuoles
(Fig. 3). However, vma16 mutants show additional deficiencies
not observed in yck3D mutants, including a defect in osmotic
stress-induced vacuole fragmentation and vacuole fusion.
Our initial screen was based on the fact that Vps41 becomes
phosphorylated in a Yck3-dependent manner under osmotic
stress conditions. We considered it likely that changes in osmolarity are sensed by plasma membrane associated proteins,
which then transmit the signal to Yck3 via a kinase cascade.
Activated Yck3 would then phosphorylate Vps41 and block
fusion. With GFP-Vps41 as a reporter, we have not yet succeeded in identifying such a signal transduction pathway, nor
do we have any idea on the mechanism of Yck3 activation.
It is well possible that essential kinases or phosphatases, which
we did not analyze in our screen, are linked to Yck3 activation.
How can we explain our observations? Our study shows that
mutations in the V-ATPase affect the release of Vps41. Even
though we focused primarily on the V0-part in our studies,
and the strongest phenotype has been observed for vma16D,
V1-mutants also affect Vps41 localization. It is therefore possible that the in vivo accumulation of Vps41 and the reduced release is linked to the proton pump activity of the V-ATPase.
Indeed, treatment of cells with concanamycin A leads to some
accumulation of GFP-Vps41 proximal to the vacuole, whereas
the release of Vps41 from isolated vacuoles was not blocked
(Fig. 1C). This suggests that the dynamics of Vps41 in vivo
are linked to V-ATPase pump activity, whereas it can by bypassed in vitro. Because yck3D vacuoles fuse in vitro, but do
not release Vps41, release does not seem to be a prerequisite
of fusion [8]. It is, however, possible that the failure to release
phosphorylated Vps41, as observed here for V-ATPase mutants, affects fusion activity.
As the V-ATPase and its proton pump activity are also required for vacuole fragmentation [22] (our study), it is not unlikely that the localization of GFP-Vps41 to dot-like structures
proximal to the vacuole reflects not only docking zones, but
also fission zones. A link between vacuole fusion and fission
has been postulated for the dynamin-like protein Vps1 [24].
Interestingly, Vps41 has also been implicated in fusion (as part
of the HOPS complex) and fission (as a potential coat of AP-3
vesicles) [11–13]. The V-ATPase function may be required indirectly to allow for lipid asymmetry as a prerequisite of fusion
and fission. It is also possible that the V-ATPase is directly involved in these processes as implicated by the association with
SNAREs (our findings; [25]).
Our data also shed light on the role of the V-ATPase in vacuole fusion. Previous data were consistent with the V0-part as a
pore forming fusogen that acts after SNAREs [25]. As vma16
Author's personal copy
K. Takeda et al. / FEBS Letters 582 (2008) 1558–1563
or vma6 mutant vacuoles could fuse efficiently with wild-type
vacuoles (Fig. 4, Suppl. Fig. 3), we consider a symmetric pore
requirement of the V-ATPase for fusion unlikely. Similar fusion results were obtained by Mayer and colleagues on the
vma6 deletion mutant [22]. We could not detect selected subunits of the V-ATPase on vacuoles lacking Vma6 or Vma16,
indicating that an assembled V-ATPase is not present on vacuoles in these mutants. Fusion was, however, strongly reduced
if the V-ATPase was missing from both tester vacuoles. The
phenotypes differ for the vma16 and vma6 mutant. The complete lack of a fusion signal in the vma16 mutant suggested
an essential asymmetric requirement, whereas some fusion
was detected with vma6 mutants. This indicates that the VATPase is required at least on one vacuole to allow for efficient
vacuole fusion, and is linked to the SNARE machinery (Suppl.
Fig. 3) [25]. It is possible that Vph1 plays a critical role within
this complex, because mutants in vph1 show the most severe
fusion defect [21,22,26].
It is presently not known how phosphorylation affects Vps41
function. Future studies are needed to identify the phosphorylation site in Vps41 and clarify how Yck3 activity is controlled.
Point mutants in Yck3 might facilitate future studies.
Acknowledgements: We thank Kathrin Auffarth and Angela Perz for
expert technical assistance, Helmut Wiecorek and Markus Huss for
concanamycin A, and Andreas Mayer, William Wickner, and Michael
Knop for plasmids and antibodies. This work was supported by the
DFG (SFB431), the Fonds of the Chemische Industrie, the Alexander
von Humboldt foundation (to K.T.), the Fundación Ramón Areces (to
M.C.), and the Hans-Mühlenhoff foundation (to C.U.).
Appendix A. Supplementary material
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.febslet.2008.
03.055.
References
[1] Bonifacino, J.S. and Glick, B.S. (2004) The mechanisms of vesicle
budding and fusion. Cell 116, 153–166.
[2] Markgraf, D.F., Peplowska, K. and Ungermann, C. (2007) Rab
cascades and tethering factors in the endomembrane system.
FEBS Lett. 581, 2125–2130.
[3] Ostrowicz, C.W., Meiringer, C.T. and Ungermann, C. (2008)
Yeast vacuole fusion: a model system for eukaryotic endomembrane dynamics. Autophagy 4, 5–19.
[4] Subramanian, K., Dietrich, L.E., Hou, H., Lagrassa, T.J.,
Meiringer, C.T. and Ungermann, C. (2006) Palmitoylation
determines the function of Vac8 at the yeast vacuole. J. Cell Sci.
119, 2477–2485.
[5] Reese, C. and Mayer, A. (2005) Trans-SNARE pairing can
precede a hemifusion intermediate in intracellular membrane
fusion. J. Cell Biol. 171, 981–990.
[6] Wang, Y.X., Kauffman, E.J., Duex, J.E. and Weisman, L.S.
(2001) Fusion of docked membranes requires the armadillo repeat
protein Vac8p. J. Biol. Chem. 276, 35133–35140.
[7] Weisman, L.S. (2006) Organelles on the move: insights from yeast
vacuole inheritance. Nat. Rev. Mol. Cell Biol. 7, 243–252.
1563
[8] Lagrassa, T.J. and Ungermann, C. (2005) The vacuolar kinase
Yck3 maintains organelle fragmentation by regulating the HOPS
tethering complex. J. Cell Biol. 168, 401–414.
[9] Sun, B., Chen, L., Cao, W., Roth, A.F. and Davis, N.G. (2004)
The yeast casein kinase Yck3p is palmitoylated, then sorted to the
vacuolar membrane with AP-3-dependent recognition of a YXXPhi adaptin sorting signal. Mol. Biol. Cell 15, 1397–1406.
[10] Cowles, C.R., Odorizzi, G., Payne, G.S. and Emr, S.D. (1997)
The AP-3 adaptor complex is essential for cargo-selective transport to the yeast vacuole. Cell 91, 109–118.
[11] Wurmser, A.E., Sato, T.K. and Emr, S.D. (2000) New component
of the vacuolar class C-Vps complex couples nucleotide exchange
on the ypt7 GTPase to SNARE-dependent docking and fusion. J.
Cell Biol. 151, 551–562.
[12] Seals, D.F., Eitzen, G., Margolis, N., Wickner, W.T. and Price, A.
(2000) A Ypt/Rab effector complex containing the Sec1 homolog
Vps33p is required for homotypic vacuole fusion. Proc. Natl.
Acad. Sci. USA 97, 9402–9907.
[13] Rehling, P., Darsow, T., Katzmann, D.J. and Emr, S.D. (1999)
Formation of AP-3 transport intermediates requires Vps41
function. Nat. Cell Biol. 1, 346–353.
[14] Darsow, T., Katzmann, D.J., Cowles, C.R. and Emr, S.D. (2001)
Vps41p function in the alkaline phosphatase pathway requires
homo-oligomerization and interaction with AP-3 through two
distinct domains. Mol. Biol. Cell 12, 37–51.
[15] Ungermann, C., Nichols, B.J., Pelham, H.R. and Wickner, W.
(1998) A vacuolar v-t-SNARE complex, the predominant form
in vivo and on isolated vacuoles, is disassembled and activated for
docking and fusion. J. Cell Biol. 140, 61–69.
[16] Price, A., Seals, D., Wickner, W. and Ungermann, C. (2000) The
docking stage of yeast vacuole fusion requires the transfer of
proteins from a cis-SNARE complex to a Rab/Ypt protein. J. Cell
Biol. 148, 1231–1238.
[17] Haas, A. (1995) A quantitative assay to measure homotypic
vacuole fusion in vitro. Meth. Cell Sci. 17, 283–294.
[18] Mayer, A., Wickner, W. and Haas, A. (1996) Sec18p (NSF)driven release of Sec17p (alpha-SNAP) can precede docking and
fusion of yeast vacuoles. Cell 85, 83–94.
[19] Seeley, E.S., Kato, M., Margolis, N., Wickner, W. and Eitzen, G.
(2002) Genomic analysis of homotypic vacuole fusion. Mol. Biol.
Cell 13, 782–794.
[20] Bonangelino, C.J., Chavez, E.M. and Bonifacino, J.S. (2002)
Genomic screen for vacuolar protein sorting genes in Saccharomyces cerevisiae. Mol. Biol. Cell 13, 2486–2501.
[21] Bayer, M.J., Reese, C., Buhler, S., Peters, C. and Mayer, A. (2003)
Vacuole membrane fusion: V0 functions after trans-SNARE
pairing and is coupled to the Ca2+-releasing channel. J. Cell Biol.
162, 211–222.
[22] Baars, T.L., Petri, S., Peters, C. and Mayer, A. (2007) Role of the
V-ATPase in regulation of the vacuolar fission–fusion equilibrium. Mol. Biol. Cell 18, 3873–3882.
[23] Nichols, B.J., Ungermann, C., Pelham, H.R., Wickner, W.T. and
Haas, A. (1997) Homotypic vacuolar fusion mediated by t- and vSNAREs. Nature 387, 199–202.
[24] Peters, C., Baars, T.L., Buhler, S. and Mayer, A. (2004) Mutual
control of membrane fission and fusion proteins. Cell 119, 667–
678.
[25] 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.
[26] Thorngren, N., Collins, K.M., Fratti, R.A., Wickner, W. and
Merz, A.J. (2004) A soluble SNARE drives rapid docking,
bypassing ATP and Sec17/18p for vacuole fusion. EMBO J. 23,
2765–2776.
[27] Peplowska, K., Markgraf, D.F., Ostrowicz, C.W., Bange, G. and
Ungermann, C. (2007) The CORVET tethering complex interacts
with the yeast Rab5 homolog Vps21 and is involved in endolysosomal biogenesis. Dev. Cell 12, 739–750.