Plant Cell Physiol. 44(8): 795–802 (2003)
JSPP © 2003
Alpha Tonoplast Intrinsic Protein is Specifically Associated with Vacuole
Membrane Involved in an Autophagic Process
Yuji Moriyasu 1, Masaki Hattori 1, Guang-Yuh Jauh 2, 3 and John C. Rogers 2, 4
1
2
School of Food and Nutritional Sciences, University of Shizuoka, 52-1 Yada, Shizuoka, Japan
Institute of Biological Chemistry, Washington State University, Pullman, WA 99164-6340 U.S.A.
;
Protein markers that are specific for the internal contents
or the limiting membrane of a certain type of vacuole have
been used to define different vacuole types. For example, different tonoplast intrinsic protein (TIP) isoforms appear to correlate with different vacuole functions. Antibodies were prepared to peptides representing the C-terminal cytoplasmic tails
of a-, g-, and d-TIP, and affinity-purified antibodies were found
to be highly specific for the appropriate protein (Jauh et al.
1998, Jauh et al. 1999). These antibodies were used predominantly in immunofluorescence studies to compare tonoplast
labeling with vacuole contents. The barley cysteine protease,
aleurain, served as a lumenal marker for lytic vacuoles (Jauh et
al. 1999, Paris et al. 1996). For lumenal markers for vacuoles
storing vegetative storage proteins, antibodies to soybean vegetative storage proteins and antibodies to protease inhibitors in
tomato and potato were used (Jauh et al. 1998). For lumenal
markers for protein storage vacuoles (PSVs), antibodies to barley lectin were used to study barley root tip cells (Jauh et al.
1999, Paris et al. 1996), and antibodies to glucanase and chitinase were used to characterize tobacco and tomato seed PSVs
(Jiang et al. 2000). These studies in aggregate (Jauh et al. 1998,
Jauh et al. 1999, Jiang et al. 2001, Jiang et al. 2000) demonstrated that storage vacuole tonoplast contains d-TIP: PSVs
containing seed-type storage proteins are marked by a- and dor d- and d- plus g-TIP, whereas vacuoles storing vegetative
storage proteins and pigments are marked by d-TIP alone or dplus g-TIP. In contrast, those marked by g-TIP alone have characteristics of lytic vacuoles. In root tips, relatively undifferentiated cells that contained small vacuoles labeling separately for
each of the three TIPs were identified (Jauh et al. 1999). These
results argued that plant cells may have the ability to generate
and maintain three separate vacuole organelles, with each being
marked by a different TIP, and that the functional diversity of
the vacuolar system may be generated from different combinations of the three basic types (Jauh et al. 1999).
This hypothesis raises the following question. If storage
vacuoles are characterized by d-TIP alone or together with aor g-TIP, and if lytic vacuoles are characterized by g-TIP alone,
what kind of vacuole is marked by only a-TIP in its tonoplast?
One possible answer came from studies of vacuoles in barley
aleurone protoplasts (Swanson et al. 1998). Two distinct vacuoles were identified in those cells. In protoplasts isolated and
maintained in the presence of abscisic acid for 1 d, large PSVs
Autophagy in plant cells is induced by nutrient starvation. Initially, double membrane-bound organelles, termed
autophagosomes, enclose a portion of cytoplasm, and then
fuse with a vacuole or lysosome to give an autolysosome.
Autolysosomes can be visualized by incubating cells in the
presence of a membrane-permeable cysteine protease
inhibitor. The inhibitor presumably decreases proteolytic
degradation of the autolysosome contents that are composed of portions of cytoplasm enclosed by the membrane
originating from the inner membrane of autophagosomes,
and allows them to accumulate. The origin of membranes
that give rise to autophagosomes and autolysosomes is
unknown. Here we use an acidotropic fluorescent dye, LysoTracker Red, to label autolysosomes specifically. We demonstrate that autolysosome membranes are marked by the
presence of a-tonoplast intrinsic protein (a-TIP) but not by
g-TIP or d-TIP. The identification of a TIP specifically associated with membranes derived from an autophagic process may help our understanding of how plant cells generate
and maintain functionally distinct types of vacuoles.
Keywords: Autophagy — Barley — Tobacco — Tonoplast
intrinsic protein — Vacuole.
Abbreviations: PBS, phosphate-buffered saline; PSV, protein
storage vacuole; TIP, tonoplast intrinsic protein.
Introduction
The plant cell vacuole has been described as a multifunctional organelle (Wink 1993), but some proposed functions
seemed contradictory. However, over the past few years it has
become clear that apparently contradictory functions, such as
providing a lytic or digestive environment versus an environment where storage of proteins is optimized, are frequently partitioned in separate vacuoles (Hoh et al. 1995, Jauh et al. 1999,
Neuhaus and Rogers 1998, Paris et al. 1996). How the plant
cell organizes and maintains functionally distinct vacuoles, and
how membrane and proteins are directed to one but not another
vacuole, remain problems that are largely unexplained on the
molecular level.
3
4
Present address: Institute of Botany, Academia Sinica, Nankang, Taipei 11529, Taiwan
Corresponding author: E-mail: [email protected]; Fax, +1-509-335-7643.
795
796
Autophagic vacuoles
contained crystalloid and globoid inclusions and had a nearneutral lumenal pH (Swanson et al. 1998, Swanson and Jones
1996). In contrast, smaller (<10 mm) secondary vacuoles contained no inclusions, had a strongly acidic lumenal pH, and the
lumenal contents included active proteases as assessed from
studies using fluorogenic protease substrates introduced into
the vacuoles in living cells (Swanson et al. 1998). Interestingly, the limiting membranes of both PSVs and secondary
vacuoles were labeled with anti-a-TIP antibodies (Swanson et
al. 1998). The authors suggested that secondary vacuoles might
represent autophagic organelles having functions similar to animal lysosomes.
Formation of organelles termed autophagosomes or
autolysosomes has been well described under conditions of
nutrient starvation in plant cells (Aubert et al. 1996, Moriyasu
and Ohsumi 1996). In suspension cultured sycamore cells
starved of sucrose, these autophagosomes predominantly were
identified as double membrane-bound structures of 3–5 mm
diameter containing cytoplasm, that appeared to fuse with the
central vacuole (Aubert et al. 1996). In tobacco BY-2 suspension culture cells, treatment with the membrane-permeable
cysteine protease inhibitor E-64d was required to allow accumulation of starvation-induced autolysosomes (Moriyasu and
Ohsumi 1996). These 1–6 mm membrane-bound organelles
accumulated in the perinuclear cytoplasm and contained osmiophilic structures, including partially degraded mitochondria,
they also appeared to fuse with the central vacuole to release
their contents into the vacuole lumen. They were termed
“autolysosomes” because they had an acidic lumenal pH, as
judged from their accumulation of acidotropic dyes, and they
were labeled strongly by enzyme histochemistry for acid phosphatase, a marker for animal lysosomes (Moriyasu and Hillmer
2000, Moriyasu and Ohsumi 1996).
To clarify the terminology used in this paper, an
autophagosome is a double membrane-bound structure that
encloses a portion of cytoplasm but does not contain hydrolytic enzymes. Such an autophagosome fuses with a pre-existing lysosome or vacuole and transforms to an autolysosome. In
autolysosomes, enclosed cytoplasm and the membrane that
originates from the inner membrane of autophagosomes are
usually partially degraded and may be called an inclusion body.
Presumably, starvation-induced autolysosomes formed normally
in the absence of E-64d but were not visualized (Moriyasu and
Ohsumi 1996). We hypothesized that the protease inhibitor
stabilized the autolysosome membrane and allowed accumulation of an acidified internal environment that concentrated acidotropic dyes. This hypothesis assumes that the autolysosome
membrane would be rapidly degraded in the absence of the
protease inhibitor. Thus, as previously used (Moriyasu and
Ohsumi 1996), and as used in this paper, the term “autolysosome” refers specifically to a structure that is stabilized by use
of E-64d and thereby can be visualized and studied.
The ability to induce autolysosomes by nutrient starvation
provided an opportunity for us to determine if TIPs were abun-
dant in their membranes. We applied the methodology to barley root tip cells because the correlation between TIP isoforms
and vacuole functions was best established there. Here we
demonstrate that cells in barley root tips cultured in vitro form
cytoplasmic organelles that are morphologically similar to
autolysosomes in tobacco suspension culture cells, and contain
an acidic lumenal pH as judged by their accumulation of the
acidic organelle-specific fluorescent probe, LysoTracker Red.
When studied by immunofluorescence with antibodies for different TIP isoforms, these barley autolysosomes labeled specifically for a-TIP, but not for d- or g-TIP, and they did not contain aleurain, the lytic vacuole marker. Thus our results are
consistent with those from the barley aleurone protoplast system (Swanson et al. 1998), and support the hypothesis that aTIP is associated with a certain type of vacuole membrane that
originates from a process similar to autophagy.
Results
Autolysosomes accumulate in barley root tip cells treated with
the membrane-permeable cysteine protease inhibitor E-64d
When tobacco suspension culture cells are put under starvation conditions, autophagy is induced. A cysteine protease
inhibitor blocks proteolysis with the autophagic organelles, and
as a result autolysosomes having incompletely degraded cytoplasmic materials accumulate in the cells. We reasoned that
placing excised barley root tips in culture might result in nutrient starvation, similar to that observed with the tobacco suspension culture cells in the absence of sucrose. Alternatively, if
autophagy is an ongoing process in barley root tip cells under
normal developmental conditions, it might be detected by the
accumulation of autolysosomes in the presence of cysteine protease inhibitors.
To test these possibilities, barley root tips were cultured in
vitro for 24 h in the presence of the membrane permeable
cysteine protease inhibitor E-64d, chemically fixed, and then
plastic-embedded sections were surveyed for the presence of
autolysosomes in the cells by light microscopy (Fig. 1). In pilot
experiments, no differences were seen when root tips were cultured in the presence of sucrose versus in the absence of
sucrose (data not presented). This result indicated that nutrient
starvation alone did not result in accumulation of autolysosomes, or that root tip cells in culture are starved even if
sucrose is added to the culture medium. In subsequent experiments, root tips were routinely cultured in the presence of
sucrose. Various concentrations of E-64d were tested in order
to examine the possibility that E-64d has some general adverse
effects on barley root tips. In sections from controls, epidermal
and neighboring cortical cells near the meristem were immature and contained many small vacuoles, some of which contained one or a few particles which were strongly stained with
toluidine blue (Fig. 1A, examples indicated by arrows). Similar
particles have previously been described in vacuoles in vegetative tissues and probably represent storage proteins (Shumway
Autophagic vacuoles
Fig. 1 Effect of the membrane-permeable cysteine protease inhibitor
E-64d on the morphology of barley root tips. Barley root tips were
kept in MS medium containing 3% (w/v) sucrose in the presence (B)
or absence (A) of 100 mM E-64d for 24 h. Methanol was used as a solvent control in A. The root tips were then fixed with glutaraldehyde
and embedded in Spurr’s resin. Sections (1 mm thick) prepared from
the blocks were stained with toluidine blue and observed under a light
microscope. (A) Arrow indicates a small vacuole containing a darkly
stained inclusion. (B) Open arrow indicates a large vacuole with multiple darkly stained inclusions. For both: RC, root cap; E, epidermal
layer; C, subepidermal or cortical layer; bars = 20 mm.
et al. 1970). E-64d did not have any significant effect on the
morphology of root tips at 10 mM (data not shown). However,
when the root tips were treated with 50, 100, 200 or 400 mM E64d, the immature cells developed relatively large vacuoles that
contained numerous toluidine blue-stained particles (Fig. 1B,
open arrow). No reproducible differences in morphology were
observed in response to these four treatments except that root
797
tips became swollen in the presence of 400 mM E-64d (data not
shown). We judged that 400 mM E-64d may have toxic effects
on the cells, and therefore used the inhibitor at 100 mM in subsequent experiments.
We further characterized the ultrastructure of the cells in
thin sections by transmission electron microscopy (Fig. 2). At
lower magnification the images were similar to those obtained
by light microscopy as shown in Fig. 1. Immature epidermal
and cortical cells of control root tips contained several small
vacuoles, some of which included electron-dense particles (Fig.
2A, arrow). In contrast, immature epidermal and their neighboring cortical cells in root tips treated with E-64d had large
vacuoles containing numerous electron dense particles (Fig.
2B, arrow). Comparing these light and electron microscopic
images, we judged that the electron-dense particles in electron
microscopic images corresponded to toluidine-blue stained particles in light microscopic images. The electron-dense particles
in the control cells differed from those in E-64d-treated cells in
that the surface of the particles appeared smoother (compare
Fig. 2A versus B).
To clarify further the morphology of vacuoles induced by
treatment with E-64d, we observed their development over a
time course (Fig. 3). Root tip epidermal cells treated with E64d for 2 h (Fig. 3A) were similar to untreated controls. After
4 h treatment, the number of vacuoles appeared to increase and
most vacuoles contained large osmiophilic inclusions (Fig. 3B,
arrow). The morphology of cells treated for 8 h was similar to
the 4 h sample except that the vacuoles appeared larger and
contained numerous inclusions (Fig. 3C, arrow). Between 8 h
and 24 h (compare Fig. 3C with Fig. 2B), both the size of the
vacuoles and the number of inclusions per vacuole increased
greatly.
The morphology of the large vacuoles that resulted from
treatment with E-64d was similar to that of autolysosomes in
tobacco cells which also contained electron dense particles
with rough-appearing surfaces (Moriyasu and Ohsumi 1996).
We reasoned that the increase in size of vacuoles in the barley
epidermal cells induced by treatment with E-64d might result
Fig. 2 Electron microscopic observation of barley root tips, untreated or treated with E-64d.
Sections were made for electron microscopy from
the same blocks as those used in Fig. 1, stained
with KMnO4 and uranyl acetate, and observed in
an electron microscope. The panels show an epidermal cell (E) positioned adjacent to the beginning of the root cap (RC). (A) Methanol-treated
root tip. Arrow indicates a small vacuole with an
osmiophilic inclusion. (B) E-64d-treated root tip.
Arrow indicates a large vacuole containing many
osmiophilic inclusions. For both: N = nucleus;
bar = 5 mm.
798
Autophagic vacuoles
1999), we hypothesized that these putative barley autolysosomes would have a unique TIP profile. We therefore established a system where putative autolysosomes could be identified in fixed cells, so that their associated TIP isoforms could
be defined by immunofluorescence labeling.
LysoTracker Red dye is concentrated in autolysosomes
In order to localize autolysosomes, we tried to stain them
with a fluorescent dye. We chose LysoTracker Red, a dye that
is concentrated in mammalian lysosomes (Lepperdinger et al.
1998, Tarasova et al. 1997), presumably because of their acidic
pH. It has the additional advantage that it can be crosslinked
with aldehyde to fix it in place, and remains fluorescent after
fixation. The ability of the dye to label autolysosomes was
tested in tobacco cells that were accumulating starvationinduced autolysosomes. The dye was concentrated in
organelles having the typical appearance of autolysosomes
localized in a perinuclear distribution (Fig. 4A, arrow), while
control cells grown in the presence of sucrose did not concentrate the dye (Fig. 4B). This ability of autolysosomes to concentrate LysoTracker Red is consistent with our previous observation that they concentrated the acidotropic dyes quinacrine
and neutral red (Moriyasu and Ohsumi 1996).
Similar results were obtained when barley root tip cells
were studied. Barley root tips were incubated in MS medium
containing sucrose in the presence (Fig. 4C) or absence (Fig.
4D) of E-64d for 24 h, labeled with LysoTracker Red, fixed
with paraformaldehyde, and then processed to release individual cells. E-64d-treated cells contained numerous cytoplasmic
organelles that concentrated LysoTracker Red (Fig. 4C, arrow)
that were similar in appearance to the starvation-induced
autolysosomes in BY-2 cells. In contrast, control cells showed
little concentration of the dye and no autolysosome-like structures were visible (Fig. 4D).
Fig. 3 The development of autophagic vacuoles in barley root tip
cells treated with E-64d. Barley root tips were kept in MS medium
containing 3% (w/v) sucrose in the presence of 100 mM E-64d for 2 h
(A), 4 h (B), and 8 h (C). They were thereafter fixed with glutaraldehyde and embedded in blocks of Spurr’s resin. Sections made from
these blocks were stained with KMnO4 and uranyl acetate, and
observed in an electron microscope. Arrows indicate autolysosomes;
bar = 5 mm.
from fusion of autolysosomes with each other or with preexisting vacuoles in the cell, similar to the process observed in
sucrose-deprived tobacco cells. Hence we concluded that these
barley root tip cell vacuoles were likely to be autolysosomes.
Based on our previous findings that different TIP isoforms correlated with different functional types of vacuoles (Jauh et al.
Alpha-TIP is specifically associated with the membrane surrounding autolysosomes in barley root tip cells
Barley root tips were incubated with E-64d for 24 h,
labeled with LysoTracker Red, fixed with paraformaldehyde,
and then processed to release individual cells that were permeabilized and labeled with anti-peptide antibodies specific for
different TIP isoforms, or with anti-aleurain antibodies (Jauh et
al. 1999). We limited our observations to small cells having a
square or rectangular shape that were similar in size and
appearance to epidermal cells and/or subepidermal cells (cortical cells just beneath the epidermal cells) near the meristem, as
shown in Fig. 2 and 3. We concentrated on these cells because
they lacked large, pre-existing vacuoles that might otherwise
interfere with characterization of membranes associated with
newly formed autolysosomes.
We reasoned initially that autolysosomes would contain an
acidic environment with proteases that were inhibited by E64d, and therefore would most likely resemble lytic vacuoles.
We therefore tested antibodies to aleurain and g-TIP as lytic
Autophagic vacuoles
799
Fig. 4 Concentration of LysoTracker Red in autolysosomes. (A, B) Tobacco BY-2 suspension culture cells were kept under sucrose starvation
conditions for 24 h in the presence (A) or absence (B) of E-64c, an analog of E-64d, and then labeled with LysoTracker Red. After fixation the
cells were observed by confocal microscopy. In the cells treated with E-64c, autolysosomes accumulated and concentrated LysoTracker Red as
indicated by the arrow. (C, D) Cells from barley root tips incubated in MS medium containing sucrose in the presence (C) or absence (D) of E64d; the arrow indicates organelles with an appearance similar to that of autolysosomes that concentrate LysoTracker Red. n = nucleus; bar =
20 mm (A, B); 10 mm (C, D).
vacuole markers (Jauh et al. 1999). As shown in Fig. 5A, control cells had only occasional structures that concentrated LysoTracker Red (arrow), but these were not labeled with antialeurain antibodies (green). Thus, even in the absence of E64d, aleurain did not appear to be associated with compartments that concentrated LysoTracker Red. Similarly, large clusters of autolysosomes containing LysoTracker Red showed no
labeling by anti-aleurain antibodies in all five E-64d-treated
cells studied (Fig. 5B). Two different patterns were observed in
treated cells labeled with anti-g-TIP antibodies. In three cells,
only occasional vacuoles were labeled for g-TIP and no association with LysoTracker Red-containing vacuoles was present
(Fig. 5C). Alternatively, in three cells, more numerous vacuoles
labeled for g-TIP and, in some areas of the cell, these were
closely associated with LysoTracker Red-containing vacuoles
(Fig. 5D, open arrow). However, in other areas (open triangle)
the LysoTracker positive-vacuoles showed no association with
g-TIP. We then tested d-TIP as a marker for storage vacuoles
and found no association with LysoTracker-positive vacuoles
in all five cells studied (Fig. 5E).
Much different results were obtained when cells were
labeled for a-TIP. Fig. 6 presents three patterns that were representative of all eight cells studied. In Fig. 6A, prominent labeling for a-TIP appears to surround each individual autolysosomal inclusion as marked by LysoTracker Red labeling,
although occasionally a-TIP labeling seemed to form a larger
structure that included more than one inclusion (arrow). In a
second pattern, labeling for a-TIP again surrounded all LysoTracker-positive vacuoles, but the structures defined by a-TIP
were large and included many inclusions (Fig. 6B, arrow). A
third variation also showed all LysoTracker-positive vacuoles
surrounded by a-TIP, with an approximately equal frequency
800
Autophagic vacuoles
Fig. 5 Barley root tip cells labeled with LysoTracker (LT, red) and antibodies to vacuolar proteins (green). Each set presents red and green
images individually, and the two images combined
(Merge). (A) Control cell not treated with E-64d
labeled with anti-aleurain (Aleu) antibodies. Arrow
indicates a small LysoTracker-positive organelle.
(B–E) Cells treated with E-64d for 24 h. (B) Antialeurain antibodies. (C) Anti-g-TIP antibodies. (D)
Anti-g-TIP antibodies. Open arrow: autolysosomes
associated with g-TIP; open triangle: autolysosomes showing no association with g-TIP. (E) Antid-TIP. The green signal associated with the cell
periphery at the top and the bottom represents
cross-reactivity of the anti-d-TIP peptide antibody
with barley microtubules (Jauh et al. 1999). For all:
n, position of nucleus; bar = 5 mm.
Fig. 6 Barley root tip cells labeled with LysoTracker and anti-a-TIP antibodies. Red, green, and
merged images, and labels are as for Fig. 5; cells
were incubated with E-64d for 24 h. (A) Two cells
where predominantly individual autolysosomes
appear to be surrounded by a-TIP-containing
membrane; arrow: a larger structure containing
multiple red autolysosomal inclusions. (B) A cell
where multiple inclusions (arrow) are incorporated
into large vacuoles marked by a-TIP in their membranes. (C) A cell where some vacuoles appear to
contain only individual autolysosomal inclusions,
while others appear to contain multiple inclusions
(arrow). For all: n, position of nucleus; bar = 5 mm.
of a-TIP structures containing multiple inclusions (arrow) versus those with only one inclusion. These patterns were uniform in multiple optical sections collected from the same cell,
although only single sections are presented here.
Discussion
Although a limited number of cells were studied by
immunofluorescence, the patterns obtained were striking in
their reproducibility and absence of variability. In every
instance, a-TIP-containing membranes surrounded all visible
Autophagic vacuoles
LysoTracker-containing autolysosomes. In contrast, vacuoles
containing aleurain and vacuoles marked by d-TIP showed no
association with autolysosomes, and vacuoles marked by g-TIP
in some instances were associated with autolysosomes in one
part of the cell while in other parts of the same cell there was
no association with autolysosomes. Thus we conclude that aTIP in the absence of d- or g-TIP is specifically associated with
membranes that surround autolysosomes in this barley root tip
cell system.
How might a-TIP-containing membranes be associated
with autolysosomes? Aubert et al. (1996) showed that most
starvation-induced autophagosomes in suspension cultured sycamore cells are enclosed by a double membrane. The outer
membrane of these organelles appeared to fuse with the central
vacuole membrane to release a single membrane-bound vesicle into its lumen. If this model is applied to our results, one
possibility would be that autophagosomes selectively fuse with
pre-existing vacuoles in the plant cell that are marked with aTIP alone. This explanation would be consistent with the
appearance of vacuoles indicated by arrows in Fig. 6, where the
vacuoles appear to contain multiple autolysosomes. However,
in previous studies (Jauh et al. 1999) we found that root tip
cells containing vacuoles marked only by a-TIP, in the absence
of d- or g-TIP, were infrequent. For example, where triple labeling for the three TIP isoforms was performed on pea root tip
cells, only 3% had vacuoles marked by a-TIP alone. This
explanation would appear to require the generation of large
amounts of vacuole membrane in association with starvation
conditions and the presence of E-64d. We have not been able to
follow the organization of vacuoles in single living barley root
tip cells as they are placed in culture and treated with E-64d.
Thus, the question of whether a-TIP-containing membranes are
generated by our experimental conditions remains unclear.
A second explanation would postulate that membranes
containing a-TIP participate directly in the process of formation of autolysosomes. Such a process could explain the results
shown for the cells in Fig. 6 if a-TIP were specifically associated with the outer membrane of an autophagosome. In the
process of transformation of autophagosomes to autolysosomes, a-TIP on the outer membrane would be transferred to
the tonoplast of the vacuole with which the autophagosomes
fused. A rigorous test of this hypothesis will require immunolocalization of a-TIP in autophagosome and autolysosome membranes at the electron microscope level.
Our finding, that autophagy is involved in formation of a
certain functional type of vacuole, would be consistent with
results from previous studies. Autophagy has long been postulated to participate in vacuole biogenesis (Moriyasu and
Hillmer 2000). A process by which vegetative vacuoles were
formed from membranes derived from the trans-Golgi network in a manner that involved autophagic envelopment of
cytoplasm was described (Marty 1978, Marty 1997). Other
workers subsequently presented evidence for the origination of
vegetative vacuoles from endoplasmic reticulum-derived mem-
801
branes that assembled around an area of cytoplasm in an
autophagic-like manner (Hilling and Amelunxen 1985). It
seems reasonable to hypothesize that one functional vacuole
type, such as a lytic vacuole, might originate from one membrane source, e.g. the trans-Golgi network, while a different
type of vacuole, such as a protein storage vacuole might originate from a different membrane source, e.g. the endoplasmic
reticulum. The earlier studies did not identify the functional
type of vacuole formed.
Subsequently, more evidence for participation of endoplasmic reticulum-derived membrane in protein storage vacuole biogenesis has been presented. In studies of biogenesis of
protein storage vacuoles, Robinson et al. (1995) demonstrated
that protein storage vacuole membrane originated from tubular
dilations of endoplasmic reticulum. This tubular membrane
system was marked by immunolabeling with anti-a-TIP antibodies, and gradually expanded to envelope and compress coexisting vegetative vacuoles in a manner that had some similarities to an autophagic process (Hoh et al. 1995). Thus a-TIP
and endoplasmic reticulum-derived membrane appear to be
associated with protein storage vacuoles.
Our previous work indicated that storage vacuoles were
marked by the presence of d-TIP in their membranes, and that
the combination of d-TIP and a-TIP was characteristic of a
specific type of storage vacuole, the protein storage vacuole
(Jauh et al. 1998, Jauh et al. 1999, Jiang et al. 2000). Thus, the
vacuoles involved in autophagy described here differ in that dTIP was not associated with their membranes. Perhaps formation of a certain vacuole type by autophagy involves a-TIPcontaining membranes, and the function of the vacuole may be
subsequently modified for storage by addition of membrane
elements that carry d-TIP as a marker (Jauh et al. 1999).
Why would formation of structures with close similarities
to previously defined autolysosomes (Moriyasu and Ohsumi
1996) be induced solely by the presence of the membranepermeable cysteine protease inhibitor, E-64d, in barley root tips
incubated in sucrose-containing medium? In barley aleurone
protoplasts, secondary vacuoles, which are thought to be similar to autolysosomes, were shown to contain high levels of protease activity that could be inhibited by the cysteine protease
inhibitors leupeptin and E-64 (Swanson et al. 1998). Thus it is
reasonable to hypothesize that E-64d could diffuse into living
root tip cells and inhibit protease activity in autolysosomes that
might normally be present in small numbers, as shown by
accumulation of LysoTracker Red in a few small structures in
control cells (e.g. Fig. 5A). Although sucrose was present in
the incubation medium, it is likely that the nutritional status of
cells in root tips incubated in vitro would be more tenuous than
that of cells in root tips attached to a normal plant vascular system. We speculate that inhibition of normal, low level,
autophagic activity by E-64d would limit recycling of carbon
and nitrogen from endogenous protein turnover, and would
thereby further stress the cells nutritionally. This nutritional
stress could then stimulate formation of more autolysosomes,
802
Autophagic vacuoles
as was shown for tobacco BY-2 cells cultured in the absence of
sucrose (Moriyasu and Ohsumi 1996).
Materials and Methods
Plant materials
Barley (cv. Himalaya or cv. Doriru) seeds were sterilized for
about 20 min with 0.16% (v/v) Bleach that was commercially available. The seeds were washed with sterile water, and then they were put
on 1% agar in Petri dishes with an embryo side facing up. The dishes
were put under continuous light from fluorescent lamps at 25–26°C.
After 1–2 d, when roots came out, the dishes were slanted so that the
roots grew straight in parallel with the surface of agar. Root tips (about
10 mm long) were excised from the seedlings grown for 3–4 d. Six
root tips were kept in 3 ml of culture medium consisting of 1´ MS salt
mixture (M7150, Sigma) and 3% (w/v) sucrose at 25°C with rotation
at 100 rpm. Since E-64d was prepared as 100´ stock solutions in
methanol, the final concentration of methanol in the incubation
medium became 1%. After treatment with E-64d, the root tips were
stained with 1 mM LysoTracker Red (Molecular Probes) in 10 mM
HEPES-Na (pH 7.5) for 1 h with rotation and thereafter fixed in a
solution containing 50 mM Na phosphate (pH 7.0), 5 mM EGTA, 3.7%
(w/v) formaldehyde, and 0.02% (w/v) NaN3.
To isolate the cells from root tips, root tips (1–2 mm from the
tips) were further excised from the fixed root tips (10 mm long) and
washed with 50 mM Na phosphate (pH 7.0) containing 5 mM EGTA
and 0.02% (w/v) NaN3. They were treated with 1% (w/v) cellulase
(Cellulysin, Calbiochem) in the same buffer for 20 min at room temperature. Epidermal and subepidermal cells were released from root
tips by tapping root tips with a plastic tip of a Pipetman in a small
amount of the buffer. Released cells were collected by centrifugation.
Cells suspended in a small amount of the buffer were observed by confocal laser microscopy (LSM510, Zeiss or MRC1024, Bio-Rad).
Immunostaining of root tip cells
Fixed root tip cells were treated with 0.5% (w/v) Triton X-100
for 5 min at room temperature and then kept in a blocking buffer consisting of 0.25% (w/v) BSA, 0.25% (w/v) gelatin, 0.05% (w/v) Nonidet P-40 and 0.02% (w/v) NaN3 in phosphate-buffered saline (PBS) for
30 min at room temperature. The cells were kept in the blocking buffer
containing a primary antibody (1/100 dilution) at 4°C overnight. They
were washed with the blocking buffer three times each for 5 min and
once for 30 min, and then kept in PBS containing a secondary antibody (1/200 dilution) at room temperature for 1 h. The cells were
washed with PBS overnight. Aliquots were taken onto a glass slide,
covered with Mowiol dissolved in glycerol, and sealed with a cover
slip. Samples were observed under a confocal laser microscope
(MRC1024, Bio-Rad).
Electron microscopy
Root tips treated in various ways were fixed with 2.5% (w/v)
glutaraldehyde in Millonig’s phosphate buffer consisting of 100 mM
Na phosphate (pH 7.4) and 86 mM NaCl for 1 h at room temperature,
and overnight at 4°C after changing the fixative. Samples were postfixed with 1% (w/v) osmium tetroxide for 3 h at room temperature.
They were dehydrated using various concentrations of ethanol and
propylene oxide and then they were embedded in Spurr’s resin. Sections were stained by keeping on a drop consisting of 2 : 1 mixture of
4% (w/v) uranyl acetate and 1% KMnO4 (w/v) for 5 min at room temperature and washed with water. They were viewed in an electron
microscope (H-7000, Hitachi, Japan).
Acknowledgments
This work was supported by grants from the National Science
Foundation (MCB-9974429), Department of Energy (DE-FG0397ER20277), and Human Frontier Science Program (RG0018/2000) to
J.C.R., and by a grant from the Japan Space Forum oriented to utilization of the space station to Y.M.
References
Aubert, S., Gout, E., Bligny, R., Marty-Mazars, D., Barrieu, F., Alabouvette, J.,
Marty, F. and Douce, R. (1996) Ultrastructural and biochemical characterization of autophagy in higher plant cells subjected to carbon deprivation: control by the supply of mitochondria with respiratory substrates. J. Cell Biol.
133: 1251–1263.
Hilling, B. and Amelunxen, F. (1985) On the development of the vacuole. II.
Further evidence for endoplasmic reticulum origin. Eur. J. Cell Biol. 38: 195–
200.
Hoh, B., Hinz, G., Jeong, B.-K. and Robinson, D.G. (1995) Protein storage
vacuoles form de novo during pea cotyledon development. J. Cell Sci. 108:
299–310.
Jauh, G.-Y., Fischer, A.M., Grimes, H.D., Ryan, C.A. and Rogers, J.C. (1998) dTonoplast intrinsic protein defines unique plant vacuole functions. Proc. Natl.
Acad. Sci. USA 95: 12995–12999.
Jauh, G.-Y., Phillips, T. and Rogers, J.C. (1999) Tonoplast intrinsic protein isoforms as markers for vacuole functions. Plant Cell 11: 1867–1882.
Jiang, L., Phillips, T.E., Hamm, C.A., Drozdowicz, Y.M., Rea, P.A., Maeshima,
M., Rogers, S.W. and Rogers, J.C. (2001) The protein storage vacuole: a
unique compound organelle. J. Cell Biol. 155: 991–1002.
Jiang, L., Phillips, T.E., Rogers, S.W. and Rogers, J.C. (2000) Biogenesis of the
protein storage vacuole crystalloid. J. Cell Biol. 150: 755–769.
Lepperdinger, G., Strobl, B. and Kreil, G. (1998) HYAL2, a human gene
expressed in many cells, encodes a lysosomal hyaluronidase with a novel
type of specificity. J. Biol. Chem. 273: 22466–22470.
Marty, F. (1978) Cytochemical studies on GERL, provacuoles, and vacuoles in
root meristematic cells of Euphorbia. Proc. Natl. Acad. Sci. USA 75: 852–856.
Marty, F. (editor) (1997) The Biogenesis of Vacuoles: Insights from Microscopy.
Academic Press, New York.
Moriyasu, Y. and Hillmer, S. (2000) Autophagy and vacuole formation. In Vacuolar Compartments. Edited by Robinson, D.G. and Rogers, J.C. pp. 71–89.
Sheffield Academic Press, Sheffield.
Moriyasu, Y. and Ohsumi, Y. (1996) Autophagy in tobacco suspension-cultured
cells in response to sucrose starvation. Plant Physiol. 111: 1233–1241.
Neuhaus, J.M. and Rogers, J.C. (1998) Sorting of proteins to vacuoles in plant
cells. Plant Mol. Biol. 38: 127–144.
Paris, N., Stanley, C.M., Jones, R.L. and Rogers, J.C. (1996) Plant cells contain
two functionally distinct vacuolar compartments. Cell 85: 563–572.
Robinson, D.G., Hoh, B., Hinz, G. and Jeong, B.-K. (1995) One vacuole or two
vacuoles: do protein storage vacuoles arise de novo during pea cotyledon
development? J. Plant Physiol. 145: 654–664.
Shumway, L.K., Rancour, J.M. and Ryan, C.A. (1970) Vacuolar protein bodies
in tomato leaf cells and their relationship to storage of chymotrypsin inhibitor I protein. Planta 93: 1–14.
Swanson, S.J., Bethke, P.C. and Jones, R.L. (1998) Barley aleurone cells contain two types of vacuoles: Characterization of lytic organelles using fluorescent probes. Plant Cell 10: 685–698.
Swanson, S.J. and Jones, R.L. (1996) Gibberellic acid induces vacuolar acidification in barley aleurone. Plant Cell 8: 2211–2221.
Tarasova, N.I., Stauber, R.H., Choi, J.K., Hudson, E.A., Czerwinski, G., Miller,
J.L., Pavlakis, G.N., Michejda, C.J. and Wank, S.A. (1997) Visualization of G
protein-coupled receptor trafficking with the aid of the green fluorescent protein. J. Biol. Chem. 272: 14817–14824.
Wink, M. (1993) The plant vacuole: A multifunctional compartment. J. Exp.
Bot. 44 (Supp.): 231–246.
(Received March 3, 2003; Accepted May 13, 2003)