Plant Cell Physiol. 48(5): 753–761 (2007)
doi:10.1093/pcp/pcm045, available online at www.pcp.oxfordjournals.org
ß The Author 2007. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.
All rights reserved. For permissions, please email: [email protected]
Excessive Expression of the Plant Kinesin TBK5 Converts Cortical
and Perinuclear Microtubules into a Radial Array Emanating
From a Single Focus
Yuhei Goto
1
1, 2
and Tetsuhiro Asada
1,
*
Department of Biological Science, Graduate School of Science, Osaka University, Toyonaka, Osaka, 560-0043 Japan
Introduction
TBK5 is a plant-specific kinesin constantly expressed in
tobacco BY-2 cells. An analysis of the distribution of green
fluorescent protein-tagged TBK5 (GFP–TBK5) transiently
expressed in BY-2 protoplasts revealed that TBK5 could
associate with microtubules in vivo. GFP–TBK5 often
assembled to form a single particle when accumulated in
cells. The particle was located in close proximity to the
nucleus, and its formation was accompanied by the development of a radial array of microtubules emanating from it and
the loss of cortical microtubules. Microtubule depolymerization
by treatment with propyzamide inhibited particle formation
and stimulated the formation of dispersed aggregates of
GFP–TBK5. Through expression of different TBK5 mutants
as GFP fusions, the motor domain, two separated coiled-coil
domains and the C-terminal domain of TBK5 were identified
as the domains playing essential roles in particle formation.
Mutants with putatively non-motile motor domains or lacking
the C-terminal domain were localized to cortical and
perinuclear microtubules, whereas those lacking either of
the coiled-coil domains were preferentially distributed around
the nucleus and along perinuclar microtubules. Further, the
deletion of one of the coiled-coil domains or the C-terminal
domain was sufficient to inhibit the propyzamide-induced
formation of dispersed aggregates, whereas the mutation in
the motor domain was not. These results led us to propose a
model in which the particle is formed through the microtubulebased movement of GFP–TBK5 toward the nucleus and
subsequent microtubule-independent aggregation based on
coiled-coil interactions. The dramatic microtubule rearrangement would be explained if GFP–TBK5 relocated and
gathered newly formed microtubules and/or microtubulenucleating units.
The microtubule cytoskeleton in eukaryotic cells forms
the mitotic spindle and other cell type-specific arrays
required for motility, morphogenesis and division of cells.
The formation of microtubules in the cytoplasm relies on
microtubule-nucleating units to prepare microtubule seeds
for promoting tubulin polymerization (Hyman and
Karsenti 1998, Job et al. 2003). The formation of focused
microtubule arrays, seen in many animal cell types that
contain centrosomes, is based on the localization of
microtubule-nucleating units to the compact organelles
that function as microtubule-organizing centers (Lüders
and Stearns 2007). Land plant cells lack centrosomes or
their equivalents, and contain dispersed microtubulenucleating units to generate diverse microtubule arrays
with characteristically non-focused or parallel nature
(Vaughn and Harper 1998, Wasteneys 2002, Schmit 2002,
Chan et al. 2003, Murata et al. 2005, Lüders and Stearns
2007). One of the major problems in understanding the
mechanism of microtubule organization in land plant cells
lies in regulating the distribution of microtubule-nucleating
sites (Mazia 1984, Lambert 1995). Theoretically, modification of this regulation could result in concentration of
microtubule-nucleating sites to generate an unusually
focused array. Even when induced artificially, such a
phenomenon could provide a useful system for dissecting
the unknown mechanism of microtubule organization
in land plant cells.
The kinesin superfamily of eukaryotic microtubule
motors includes some subfamilies whose members were
initially identified in fungal and animal cells as key players
in the formation of the bipolar spindle (Vale 2003,
Lawrence et al. 2004) and then identified from plants or
cultured plant cells (Mitsui et al. 1993, Asada et al. 1997,
Oppenheimer et al. 1997, Reddy et al. 1997, Matsui et al.
2001, Preuss et al. 2003, Preuss et al. 2004, Ambrose et al.
2005). Interestingly, one of these subfamilies, kinesin-14,
Keywords: Cortical microtubules — Kinesin — Microtubule
— Microtubule-organizing center — Tobacco BY-2 cell.
Abbreviations: BSA, bovine serum albumin;
dimethylsulfoxide;
GFP,
green
fluorescent
PBS, phosphate-buffered saline.
DMSO,
protein;
2
Present address: Research and Development Division, Janssen Pharmaceutical Co. Ltd, Chiyoda, Tokyo, 101-0065 Japan.
*Corresponding author: E-mail, [email protected]; Fax, þ81-72-723-2061.
753
754
Artificially induced microtubule-aster formation in tobacco cells
is far more diverse in plants than in other eukaryotes
(Richardson et al. 2006) and includes plant-specific variants
such as TBK5 (Matsui et al. 2001). TBK5 was identified
from tobacco BY-2 suspension culture cells, where its
mRNA was constantly present at a high level (Matsui et al.
2001). A recent comprehensive analysis of genomic
sequences by Richardson et al. (2006) allows identification
of its counterparts not only in Arabidopsis but also in Oryza
and Populus. In addition to the closely related motor
domain, TBK5 and other kinesin-14s share a similar neck
domain. Because of this similarity, TBK5 has been
predicted to be a minus end-directed motor (Matsui et al.
2001). The unique structural features that distinguish TBK5
polypeptide from other characterized kinesin-14s are the
presence of two coiled-coiled regions separated by a
centrally located motor domain and a positively charged
C-terminal region. The actual activities, subcellular localization, in vivo function and role of this kinesin remain to
be elucidated.
In this study, we analyzed the distribution of green
fluorescent protein (GFP)-tagged TBK5 transiently
expressed in tobacco BY-2 protoplasts. This analysis,
which originally aimed at obtaining information about
the subcellular localization of TBK5, demonstrated that the
expressed GFP-tagged TBK5 proteins often assembled
to form a single particle, with this generating an unusual
radial array of microtubules emanating from a focus.
Here we describe the detection and partial characterization
of this artificial phenomenon.
Results and Discussion
Protoplasts of tobacco BY-2 cells were transformed
with plasmids, which were designed to express the fulllength TBK5 transiently as GFP fusions under the control
of the cauliflower mosaic virus 35S promoter. Upon
plasmid introduction, approximately half of the GFPpositive cells contained a fibrous pattern of green fluorescence after being cultured for 15–25 h (Fig. 1a). Tubulin
immunostaining confirmed that this pattern corresponded
to that of cortical and perinuclear microtubules in cells
(Fig. 2a, left). The other half of the GFP-positive cells
contained a single green fluorescent particle that had either
an aster-like (Fig. 1b) or a globular shape (Fig. 1c).
This particle was always located in close proximity to the
nucleus. Distributions of the expressed N- and C-terminal
GFP fusions were indistinguishable, suggesting that the
GFP tag did not interfere with the original action of TBK5.
Labeling of the cells by green fluorescence appeared
at 8 h, peaked at 16–18 h and disappeared at 22 h after
plasmid introduction (Fig. 1d). All GFP-positive cells were
initially marked by a cytosolic green fluorescence in
which fibrous patterns were visible. The cells containing
fluorescent particles were detected 12 h after plasmid
introduction, and the ratio of those cells increased in a
consistent manner with time (Fig. 1d). Therefore, expressed
fusion proteins were initially distributed diffusely in the
cytoplasm and along microtubules before pooling to
generate the fluorescent particles. Cells marked by the
aster-shaped particle were rarely observed, especially in
the initial and terminal phases of transient expression.
Absence of the fluorescent particle in cells during the early
stage of transient expression suggests that the induction of
particle formation requires the presence of a high level
of GFP–TBK5. Since the changes were not reversed at the
late stage of transient expression, where the expression
level was probably reduced, the particles would have
probably remained static. Rather than at the beginning of
the expression, the fluorescent particles were always larger
toward the end of transient expression (data not shown),
where dead cells with protruding cytoplasm and fluorescent
particles were occasionally observed.
Indirect immunofluorescence using anti-a-tubulin
antibodies revealed that non-transformed control cells
typically contained plasma membrane-associated cortical
microtubules, and those distributed around the nucleus and
in cytoplasm between the nucleus and plasma membrane
(Fig. 2b). In cells with a relatively small GFP–TBK5
particle, the number of cortical microtubules was reduced,
and microtubules were arranged in an aster-shaped
array that sharply focused on the particle (Fig. 2a,
center). In cells with a larger GFP–TBK5 particle, while
cortical microtubules were depleted, microtubules arranged
in a radial array appeared to emanate from the particle
(Fig. 2a, right). Co-localization of GFP–TBK5 and microtubules was observed only on the particle.
Cytochalasin E, an inhibitor of actin filament assembly
(Kobayashi and Hakuno 2003), did not affect particle
formation (Fig. 1e). However, propyzamide, an inhibitor of
microtubule assembly (Akashi et al., 1988), not only
inhibited particle formation but also stimulated the formation of dispersed aggregates of GFP–TBK5 (Fig. 1e, f).
Indirect immunofluorescence using anti-a-tubulin antibodies confirmed that propyzamide treatment abolished microtubules in cells, and revealed that the dispersed aggregates
of GFP–TBK5 contained tubulin (Fig. 2d). In cells cultured
for 20 min after the termination of propyzamide treatment,
microtubules formed around the dispersed GFP–TBK5
aggregates (Fig. 2c). In the same cells, recovery of cortical
microtubule was less detectable (Fig. 2c). Propyzamide
treatments caused fragmentation of the previously formed
particles, while dispersed aggregates tended to coalesce
after the termination of propyzamide treatment (data
not shown).
The inhibition of particle formation by propyzamide
indicated that intact microtubules were required for
Artificially induced microtubule-aster formation in tobacco cells
755
a
b
c
N
globular particle
aster-shaped particle
other
100
5
80
4
60
3
40
2
20
1
0
8
10
12
14
16
18
20
Time after transformation (hour)
22
24
0
GFP-positive cells
(percent of total cells)
d
Fluorescent patterns
(percent of GFP-positive cells)
N
Fluorescent patterns
(percent of GFP-positive cells)
e
globular particle
aster-shaped particle
other
100
f
80
60
Dispersed aggregates
40
20
0
+DMSO
+DMSO
+DMSO
+cytochalasin +propyzamide
Fig. 1 Distribution of transiently expressed GFP–TBK5 fusion proteins in tobacco BY-2 cells. (a) Fibrous patterns of fluorescence observed
in the periphery (left) and center (right) of cells 16 h after transformation. Cells with an aster-shaped fluorescent particle (b) or a globular
fluoresent particle (c) were observed. The photographs were prepared by superimposing the images of differential interference contrast
microscopy and those of fluorescence microscopy (inserts show only the fluorescence image). N, nucleus (d) A time course of changes in
the percentage of each green fluorescent pattern (bar) and GFP–TBK5-expressing cells (line). (e) Effects of propyzamide on the formation of
GFP–TBK5 particles. Transformed cells were cultured for 8 h in normal conditions, and further cultured for 4 h in the presence of 1%
DMSO, 1% DMSO plus 100 mM cytochalasin E or 1% DMSO plus 100 mM propyzamide. (f) Dispersed aggregates of GFP–TBK5 generated
by the propyzamide treatment. Scale bar ¼ 5 mm.
756
Artificially induced microtubule-aster formation in tobacco cells
a
Cell center
Tubulin
Tubulin
c
TBK5
-GFP
Cell center
b
TBK5
-GFP
Cell periphery
Cell periphery
TBK5
-GFP
d
Tubulin
TBK5
-GFP
Tubulin
TBK5
-GFP
Fig. 2 Microtubule reorganization associated with the formation
and growth of the GFP–TBK5 particle. (a) Distributions of
microtubules (white) and GFP–TBK5 (green) are indicated accordingly. Cells with fibrous microtubule patterns (left), a relatively
small particle (center) and a larger particle (right) of GFP–TBK5
were stained with anti-a-tubulin antibodies. (b) Microtubule
arrangement in a non-transformed cell that had been cultured
overnight after cell wall digestion. (c) Microtubule regeneration
from the GFP–TBK5 aggregates. Transformed cells were cultured
for 4 h in the presence of 100 mM propyzamide, and further
incubated for 20 min without propyzamide. Immunostaining with
anti-a-tubulin antibodies visualized microtubules (red), which
extended from the GFP–TBK5 aggregates. (d) Detection of tubulins
in GFP–TBK5 aggregates. Transformed cells were cultured for 4 h
in the presence of 100 mM propyzamide and analyzed by
indirect immunofluorescence using anti-a-tubulin antibodies.
Scale bar ¼ 10 mm.
concentrating expressed GFP–TBK5 into single particles,
while the propyzamide-stimulated formation of dispersed
aggregates suggested that some mechanisms existed through
which GFP–TBK5 can locally and efficiently be gathered
without the involvement of intact microtubules. To understand the mechanism by which GFP–TBK5 proteins
assembled to form a particle or aggregates, we prepared
six different mutant TBK5 proteins (Fig. 3b). TBK5N
lacks a small N-terminal domain, whereas TBK5C lacks
the C-terminal domain with a calculated pI of 10.74.
TBK5T182N is a putative non-motile, rigor-type mutant
that has a point mutation within an ATP-binding consensus
motif located in the motor domain (Nakata and Hirokawa
1995). TBK5NC lacks the coiled-coil domain located
on the N-terminal side of the motor domain, whereas
TBK5CC is deficient of the other coiled-coil domain.
TBK5C consists of the basic C-terminal domain.
The distribution of GFP–TBK5N was indistinguishable from that of GFP–TBK5 (Fig. 3c). Expression of
GFP–TBK5T182N yielded fibrous patterns of green fluorescence (Fig. 3d), without the formation of fluorescent
particles (Fig. 3c). When GFP–TBK5T182N-expressing
cells were treated with propyzamide, however, these cells
generated dispersed aggregates of the fluorescent protein
(Fig. 3c, e). Thus, the TBK5 motor domain might actually
exhibit ATP hydrolysis-dependent activity, with this
being essential for particle formation, but not for local
aggregation.
The expression of GFP–TBK5NC and GFP–
TBK5CC resulted in fluorescent labeling preferentially
distributed around the nucleus and along the perinuclear
microtubules (Fig. 3d) without yielding green fluorescent
particles (Fig. 3c). Propyzamide treatment of these cells did
not induce formation of dispersed aggregates (Fig. 3c).
Thus, coiled-coil interactions on both sides of the motor
domain seem to be the essential events for particle
formation and local microtubule-independent aggregation.
Propyzamide-treated GFP–TBK5NC- and GFP–
TBK5CC-expressing cells devoid of any bright fluorescent
dots were marked with cytosolic fluorescence, when
observed under conventional fluorescent microscopy (data
not shown). Observations of the propyzamide-treated cells
with a microscope equipped with an image deconvolution
system, however, clearly visualized that fluorescent fibers
were located around the nucleus (Fig. 3e). This observation
suggests that association of the fusion protein increases
the stability of microtubules.
TBK5 was predicted to be a minus-end directed motor,
because its motor and neck domains resemble those of other
kinesin-14s (Case et al. 1997, Endow and Waligora 1998,
Endow and Higuchi 2000, Matsui et al. 2001). These data
can be combined with results of the expression of GFP–
TBK5T182N, GFP–TBK5NC and GFP–TBK5CC to
Artificially induced microtubule-aster formation in tobacco cells
a
757
Coiled coil
N
c
C
Ncd-like "neck"
Motor
Particle
Dispersed
aggregates
P.D.N.*
GFP-TBK5
+
+
−
C TBK5-GFP
+
+
−
C
GFP-TBK5T182N
−
+
−
C
GFP-TBK5∆N
+
+
−
pl=10.74
b
N
C
GFP
N
N
GFP
182
xT N
GFP
N
GFP
N
GFP
C
GFP-TBK5∆NC
−
−
+
N
GFP
C
GFP-TBK5∆CC
−
−
+
TBK5∆C-GFP
−
−
−
GFP-TBK5C
−
−
−
N
C
GFP
N
GFP
C
*preferential distribution around the nucleus
d
e
f
GFP-TBK5C
GFP-TBK5T182N
(cell center)
(cell periphery)
GFP-TBK5T182N
GFP-TBK5∆NC
GFP-TBK5∆CC
GFP-TBK5∆NC
TBK5∆C-GFP
(cell center)
(cell periphery)
GFP-TBK5∆CC
TBK5∆C-GFP
(cell center)
(cell periphery)
Fig. 3 Distribution of various GFP-tagged TBK5 mutants transiently expressed in BY-2 protoplasts. Schematic representations of the
primary structures of TBK5 (a) and its mutants fused with GFP (b). (c) The ability of each fusion protein to form a particle, aggregates or to
yield strong perinuclear labeling. (d) Typical distributions of GFP–TBK5T182N, GFP–TBK5NC, GFP–TBK5CC and TBK5C–GFP.
N, nucleus. (e) The fluorescent patterns in propyzamide-treated GFP–TBK5T182N-, GFP–TBK5NC- and GFP–TBK5CC-expressing cells.
(f) Preferential distribution of stably expressed GFP–TBK5C on the phragmoplast. Densely packed fluorescent fibers, probably
corresponding to phragmoplast microtubules, were observed in the region between two separated groups of chromosomes.
Scale bar ¼ 10 mm.
758
Artificially induced microtubule-aster formation in tobacco cells
N
Step 1:
Microtubule-dependent
relocation of GFP-TBK5
T182N
∆C
N
Step 2:
Microtubule-independent
aggregation of GFP-TBK5
∆NC
∆CC
(∆C)
N
Fig. 4 A model for the formation of GFP–TBK5 particles. Step 1:
a process where expressed GFP–TBK5 fusion protein (filled circle)
moves along the microtubules (thin lines) towards the nucleus (N).
Step 2: a process where GFP–TBK5 fusion protein aggregates with
coiled-coil interactions occurring on both sides of the motor
domain, independently of the presence of intact microtubules. Step
2 may start before the completion of Step 1. The model assumes
that propyzamide induces the formation of dispersed aggregates
of GFP–TBK5 by inhibiting Step 1 but not Step 2. In addition, the
model also explains that the distributions of GFP–TBK5T182N and
TBK5C–GFP (Fig. 3d) result from an inhibition of Step 1 and
those of GFP–TBK5NC and GFP–TBK5CC (Fig. 3d) result from
an inhibition of Step 2. It is possible that the GFP–TBK5NC and
GFP–TBK5CC proteins are partially aggregated in the perinuclear
region. For Step 1 to occur, the minus end of microtubules around
the nucleus is oriented towards the nucleus to facilitate GFP–TBK5
accumulation in the perinuclear region via microtubule-based
TBK5 mobility. GFP–TBK5 might co-assemble with microtubules
and/or microtubule-nucleating units to generate the single
dominant microtubule-organizing center.
furnish the following model of particle formation (Fig. 4).
Non-cortical microtubules were arranged in a polarized
array with their minus ends located on the nuclear
membrane. Because of this microtubule arrangement with
TBK5 movement directed to the minus end, GFP–TBK5
accumulated in the perinuclear region (Fig. 4, Step 1).
This process was followed by microtubule-independent
aggregation, where accumulated GFP–TBK5 proteins
assembled on the basis of coiled-coil interactions to
form a particle that served as a microtubule-organizing
center (Fig. 4, Step 2). The expressed GFP–TBK5 may
co-assemble with endogenous TBK5. If the aggregation
first occurred locally at different sites, multiple GFP–TBK5
particles and microtubule-radiating foci would appear.
However, since the formation of multiple particles and
microtubule-radiating foci was never detected in our assays,
aggregation might have first generated only a single perinuclear particle that subsequently served as a focus for
accumulation of fusion proteins through further aggregation. It remains to be determined whether or not the particle
formation can be induced only at the onset of the G1 phase
(Kumagai et al. 2004) and/or during the G2 phase (Katsuta
et al. 1990) when many microtubules are present in the
region between the nucleus and cell periphery.
Expression of TBK5C–GFP yielded fibrous patterns
of green fluorescence (Fig. 3d) without inducing formation
of fluorescent particles (Fig. 3c), and propyzamide
treatment did not result in the formation of dispersed
aggregates of fluorescent proteins (Fig. 3c, e). The results
from experiments with propyzamide indicated that the
C-terminal domain played an essential role in aggregating
GFP–TBK5. The same domain may also be required for
TBK5 mobility because, unlike GFP–TBK5NC and
GFP–TBK5CC, TBK5C–GFP was not preferentially
distributed around the nucleus (Fig. 3d). Therefore, in the
case of TBK5C–GFP, the failure in particle formation
may be due to both poor mobility and deficient local
aggregation (Fig. 4). The apparent microtubule association
of TBK5C–GFP possibly involves an interaction between
TBK5C and endogenous TBK5, although these data
would also be explained if TBK5C itself binds directly
to the microtubules.
To understand the function of the C-terminal domain,
we analyzed TBK5C (basic C-terminal region) fused with
GFP. Transient expression of GFP–TBK5C resulted in
clear accumulation of green fluorescence within the nucleus
(data not shown), an event probably due to the basic pI and
small size of the TBK5C protein. However, analysis
using stable GFP–TBK5C-expressing transformants,
aimed at determining the distribution of green fluorescence
in dividing cells in which the nuclear envelope has broken
down, revealed that the GFP–TBK5C was localized on the
phragmoplast (Fig. 3f). Thus, the C-terminal domain
may be able to associate with microtubules in vivo. If the
binding of tubulins or microtubule fragments to this
C-terminal domain induces the aggregation process,
the inability of TBK5C–GFP to form aggregates and
particles would then be obvious. In addition, the fact that
Artificially induced microtubule-aster formation in tobacco cells
propyzamide treatment stimulates the formation of
dispersed aggregates (Fig. 1e, f) would also be explained,
because propyzamide would have increased the cytosolic
concentration of tubulins and/or microtubule fragments.
Propyzamide-treated TBK5C–GFP-expressing cells
devoid of any bright fluorescent dots (Fig. 3c) contained
cytosolic fluorescence when observed under conventional
fluorescent microscopy (data not shown). However, observations using a microscope equipped with an image
deconvolution system occasionally visualized thin and
short fibers at the cell cortex (Fig. 3e). This debris probably
corresponded to the remnants of cortical microtubules.
Thus, association of TBK5C–GFP may increase the
stability of microtubules.
Although different microtubule-associated proteins
have been analyzed in plant cells with GFP fusions, particle
formation or remodeling of overall microtubule organization in patterns similar to those observed in this study have
not been reported to date. It seems worth exploring the
possible mechanism of the microtubule rearrangement
induced by GFP–TBK5 expression, since this endeavor
might offer an insight on the essential mechanism for land
plant cells to generate and maintain their non-focused or
parallel microtubule arrays. In animal cells, formation
of radial microtubule arrays involves the action of a minus
end-directed microtubule motor (cytoplasmic dynein)
in transporting and gathering centrosomal components
(Young et al. 2000) or membrane-bound organelles
responsible for microtubule nucleation (Vorobjev et al.
2001). This knowledge led us to hypothesize that the
microtubule rearrangements we observed was caused by
co-assembly of the GFP–TBK5 fusion proteins with
endogenous microtubule-nucleating units. The finding
that dispersed aggregates of GFP–TBK5 were able to
regenerate microtubule asters (Fig. 2c, d) is consistent
with the hypothesized co-assembly of GFP–TBK5 and
microtubule-nucleating units. However, the observed
microtubule regeneration would not necessarily indicate
that microtubule-nucleating units were collected as aggregates, since the same microtubule regeneration would be
expected if the aggregates contained microtubule seeds as
they were formed in the presence of propyzamide, and since
the aggregates did in fact contain tubulins after propyzamide treatment (Fig. 2d). In addition, our immunofluorescence assays occasionally failed to detect the expected
vigorous microtubule regeneration from the aggregates
(data not shown). Therefore, at this stage, we can merely
hypothesize that the microtubule rearrangement associated
with the particle formation is a result of accumulation of
active microtubule-nucleating units without strong support.
In further exploring how microtubule rearrangements
occurred in GFP–TBK5-expressing cells, we note that
microtubule nucleation in non-dividing tobacco BY-2 cells
759
depends on placement of microtubule-nucleating units on
existing cortical microtubules and that it usually yields
branches of new microtubules extending from the preexisting microtubules (Murata et al. 2005). If there is a
system that could continuously collect the microtubule
seeds or branches to form a small domain, this domain
would function as a dominant microtubule-organizing
center, since the collected microtubules could either
elongate to yield radiating microtubules directly or provide
new sites for microtubule nucleation. In fact, it is highly
possible that the GFP–TBK5 fusion proteins would
co-assemble with microtubules, since anti-a-tubulin antibodies recognized the particle (Fig. 2a, right) and dispersed
aggregates (Fig. 2d). Therefore, it seems rational to
hypothesize that the observed microtubule rearrangement
is based on the actions of accumulated GFP–TBK5 for
relocating and gathering microtubule seeds or branches
in a manner either together with or separately from
microtubule-nucleating units. For microtubule seeds or
branches to be relocated by the expressed GFP–TBK5
fusion proteins through the proposed microtubule-based
mechanism (Fig. 4), GFP–TBK5 would have to select a
component of the microtubule-nucleating units and/or new
microtubules as a cargo in order to transport them
along the pre-existing microtubules. The finding that
GFP–TBK5C expression in dividing cells yielded labeling
of the phragmoplast supports the previous hypothesis that
TBK5 binds acidic tubulins at the extremely basic
C-terminal domain possibly to transport microtubules
against neighboring microtubules (Matsui et al. 2001).
The finding that the GFP–TBK5 fusion proteins
co-localized with microtubules in the early stages of
transient expression leads us to speculate that endogenous
TBK5 is distributed along microtubules in non-transformed
tobacco BY-2 cells. Determining the subcellular localization
of endogenous TBK5 and discriminating the functional
similarities/differences between endogenous TBK5 in nontransformed cells and accumulated GFP–TBK5 fusion
proteins in transformed cells may provide a clue to
understanding the molecular mechanisms that regulate the
distribution and arrangement of microtubules in land plant
cells. As illustrated by a previous study on NACK1, a
tobacco kinesin (Nishihama et al. 2001), inducible expression of mutant proteins as a dominant-negative inhibitor
is useful for identifying the processes involving the
corresponding endogenous proteins. Mutant TBK5
proteins created in the present study (i.e. TBK5T182N,
TBK5NC, TBK5CC and TBK5C) are candidates
for expressing the dominant-negative inhibitor of TBK5,
since their actions and distributions were found to be
distinct from those of full-length TBK5. Beside analyses
using RNA interference technology or arabidopsis insertion
lines with disrupted TBK5 ortholog genes, analysis in which
760
Artificially induced microtubule-aster formation in tobacco cells
any of the four TBK5 mutants are expressed in a
regulated fashion is expected to be useful for elucidating
the function and role of endogenous TBK5 proteins in
non-transformed cells.
Materials and Methods
Cell culture and transformation
Tobacco BY-2 cells (Nicotiana tabacum L. cv. Bright
Yellow 2) were maintained using standard procedures (Nagata
et al. 1982). Protoplasts of 3-d-old subcultured BY-2 cells were
transformed using a PEG-mediated method (Maas et al. 1995).
Transformation efficiency estimated from the ratio of GFPpositive cells was about the same for each experiment, although
not always determined. Transformed protoplasts were suspended
in MS medium supplemented with 0.5 M mannitol, spread on 1%
agarose, and incubated at 288C in the dark. Stable BY-2
transformants were prepared by the Agrobacterium-mediated
method using the LBA4404 strain.
Preparation of plasmids
All plasmids for transient expression were prepared by
inserting a DNA fragment that encoded TBK5 or its variants
into the 50 or 30 end of the GFP-coding region of the CaMV35SsGFP(S65T)-nos3 plasmid (Niwa et al. 1999). DNA encoding the
mutant TBK5 protein with a point mutation or without one of the
coiled-coil domains was prepared by recombinant PCR. In
preparing a binary vector construct that expresses GFP–TBK5C,
the DNA cassette designed to overexpress GFP–TBK5C was
inserted into an Xba site of pPZP112 (Hajdukiewicz et al. 1994).
GFP–TBK5,
GFP–TBK5T182N,
GFP–TBK5N,
GFP–
TBK5NC, GFP–TBK5CC and GFP–TBK5C contained a
linker peptide (GGGKVEGG) between the N-terminally located
GFP and TBK5 sequences. TBK5–GFP and TBK5C–GFP
contained a linker peptide (GGGKDPSM) between the TBK5
and C-terminally located GFP sequences. Constructs were verified
by DNA sequencing exploiting the use of a dye deoxy-terminator
cycle sequencing kit [Applied Biosystems (ABI), Tokyo, Japan]
with an ABI model-310 sequencing system. Details of plasmid
construction and DNA constructs used in this study are available
upon request.
Treatments with cytochalasin E or propyzamide
For experiments examining the effects of cytochalasin E and
propyzamide on particle formation, transformed cells were first
cultured for 8 h in normal conditions, and then subjected to a
further 4 h culture with 1% dimethylsulfoxide (DMSO), 1%
DMSO plus 100 mM cytochalasin E or 1% DMSO plus 100 mM
propyzamide. For experiments examining the effects of propyzamide on the distribution of GFP-tagged mutant TBK5 proteins,
transformed cells were first cultured under normal conditions until
the appearance of GFP-positive cells, and then cultured in the
presence of 1% DMSO plus 100 mM propyzamide for 42 h before
observation.
Fluorescence microscopy
Cells were observed by using a conventional fluorescence
microscope (Olympus BH-2, Tokyo, Japan) or a microscopy
system equipped with an image deconvolution system (Applied
Precision DeltaVision, Washington, DC, USA). Conventional
microscopy was used to observe living GFP-positive cells
(Fig. 1b, c, f and results summarized in Fig. 3c). The DeltaVision
microscopy system was used to observe living GFP-positive cells
(Figs. 1a, 3d, e, f) and immunostained cells (Fig. 2). Figs. 1a, 2c,
and 3d, e were produced with the deconvolution program of the
DeltaVision microscopy system; other photographs are raw
images. For immunostaining, cells were suspended in a fixation
medium (50 mM PIPES, KOH, pH 7.0, 5 mM EGTA, 2 mM
MgSO4, 4% formaldehyde, 0.5M mannitol) for 45 min, and then in
a fixation medium containing 0.05% Triton X-100 for 15 min.
Fixed cells were washed three times with a washing buffer (50 mM
PIPES, KOH, pH 7.0, 5 mM EGTA, 2 mM MgSO4, 0.5 M
mannitol). After two washes with PBS-BSA (phosphate-buffered
saline containing 1% BSA), cells were further incubated for 1 h
with PBS-BSA containing mouse monoclonal antibodies against
chick brain a-tubulin (Sigma, St Louis, MO, USA). Cells were then
washed twice with PBS containing 0.02% Tween-20, and incubation with PBS-BSA containing Alexa 586-labeled goat anti-mouse
IgG (Molecular probes) for 1 h. Stained cells were washed with
PBS containing 0.02% Tween-20 before observation.
Achnowledgements
We thank Dr. Y. Niwa of Shizuoka University for kindly
providing the CaMV35S-SGFP-TYG-nos vector, Dr. P. Melga
of the State University of New Jersey for the pPZP vectors,
Dr. H. Yasuhara of Kansai University for advice on culturing
of protoplasts, and Dr. D. Collings of the Australian National
University for proofreading of this manuscript. We also wish to
thank two anonymous referees who gave important comments
on an earlier version of this paper. This study was supported by
research grants from the Asahi Glass Foundation and the Ministry
of Education, Science, Sports and Culture of Japan.
References
Akashi, T., Izumi, K., Nagano, E., Enomoto, M., Mizuno, K. and
Shibaoka, H. (1988) Effects of propyzamide on tobacco cell microtubules
in vivo and in vitro. Plant Cell Physiol. 29: 1053–1062.
Ambrose, J.C., Li, W., Marcus, A., Ma, H. and Cyr, R. (2005) A minusend-directed kinesin with plus-end tracking protein activity is involved in
spindle morphogenesis. Mol. Biol. Cell 16: 1584–1592.
Asada, T., Kuriyama, R. and Shibaoka, H. (1997) TKRP125, a kinesinrelated protein involved in the centrosome-independent organization of
the cytokinetic apparatus in tobacco BY-2 cells. J. Cell Sci. 110: 179–189.
Case, R.B., Pierce, D.W., Hom-Booher, N., Hart, C.L. and Vale, R.D.
(1997) The directional preference of kinesin motors is specified by an
element outside the motor catalytic domain. Cell 90: 959–966.
Chan, J., Calder, G.M., Doonan, J.H. and Lloyd, C.W. (2003) EB1 reveals
mobile microtubule nucleation sites in Arabidopsis. Nat. Cell Biol. 5:
967–971.
Endow, S.A. and Higuchi, H. (2000) A mutant of the motor protein kinesin
that moves in both directions on microtubules. Nature 406: 913–916.
Endow, S.A. and Waligora, K.W. (1998) Determinants of kinesin motor
polarity. Science 281: 1200–1202.
Hajdukiewicz, P., Svab, Z. and Maliga, P. (1994) The small, versatile pPZP
family of Agrobacterium binary vectors for plant transformation.
Plant Mol. Biol. 25: 989–994.
Hyman, A. and Karsenti, E. (1998) The role of nucleation in patterning
microtubule networks. J. Cell Sci. 111: 2077–2083.
Job, D., Valiron, O. and Oakley, B. (2003) Microtubule nucleation.
Curr. Opin. Cell Biol. 15: 111–117.
Katsuta, J., Hashiguchi, Y. and Shibaoka, H. (1990) The role of the
cytoskeleton in positioning of the nucleus in premitotic tobacco BY-2
cells. J. Cell Sci. 95: 413–422.
Artificially induced microtubule-aster formation in tobacco cells
Kobayashi, I. and Hakuno, H. (2003) Actin-related defense mechanism to
reject penetration attempt by a non-pathogen is maintained in tobacco
BY-2 cells. Planta 217: 340–345.
Kumagai, F., Yoneda, A., Kutsuna, N. and Hasezawa, S. (2004) Dynamic
behavior of microtubules and vacuoles at M/G1 interphase observed in
living tobacco BY-2 cells. Biotech. Agric. For. 53: 81–97.
Lambert, A.M. (1995) Microtubule-organizing centers in higher plants:
evolving concepts. Bot. Acta 108: 535–537.
Lawrence, C.J., Dawe, R.K., Christie, K.R., Cleveland, D.W.,
Dawson, S.C., et al. (2004) A standardized kinesin nomenclature.
J. Cell Biol. 167: 19–22.
Lüders, J. and Stearns, T. (2007) Microtubule-organizing centers: a
re-evaluation. Nat. Rev. Mol. Cell Biol. 8: 161–167.
Maas, C., Reichel, C., Schell, J. and Steinbiss, H.H. (1995) Preparation and
transformation of monocot protoplasts. Methods Cell Biol. 59: 393–399.
Matsui, K., Collings, D.A. and Asada, T. (2001) Identification of a plantspecific kinesin-like protein that is highly expressed in interphase tobacco
BY-2 cells. Protoplasma 215: 105–115.
Mazia, D. (1984) Centrosome and mitotic pole. Exp. Cell Res. 153: 1–15.
Mitsui, H., Yamaguchi-Shinozaki, K., Shinozaki, K., Nishikawa, K. and
Takahashi, H. (1993) Identification of a gene family (kat) encoding
kinesin-like proteins in Arabidopsis thaliana and the characterization of
secondary structure of KatA. Mol. Gen. Genet. 238: 362–368.
Murata, T., Sonobe, S., Baskin, T.I, Hyodo, S., Hasezawa, S., Nagata, T.,
Horio, T. and Hasebe, M. (2005) Microtubule-dependent microtubule
nucleation based on recruitment of g-tubulin in higher plants. Nat. Cell
Biol. 7: 961–968.
Nakata, T. and Hirokawa, N. (1995) Point mutation of adenosine
triphosphate-binding motif generated rigor kinesin that selectively
blocks anterograde lysosome membrane transport. J. Cell Biol. 131:
1039–1053.
Nagata, T., Okada, K. and Takebe, I. (1982) Mitotic protoplasts and their
infection with tobacco mosaic virus RNA encapsulated in liposomes.
Plant Cell Rep. 1: 250–252.
Niwa, Y., Hirano, T., Yoshimoto, K., Shimizu, M. and Kobayashi, H.
(1999) Non-invasive quantitative detection and applications of
761
non-toxic, S65T-type green fluorescent protein in living plants. Plant J.
18: 455–463.
Nishihama, R., Soyano, T., Ishikawa, M., Araki, S., Tanaka, H., et al.
(2002) Expansion of the cell plate in plant cytokinesis requires a kinesinlike protein/MAPKKK complex. Cell 109: 87–99.
Oppenheimer, D.G, Pollock, M.A, Vacik, J., Szymanski, D.B., Ericson, B.,
Feldmann, K. and Marks, M.D. (1997) Essential role of a kinesin-like
protein in Arabidopsis trichome morphogenesis. Proc. Natl Acad. Sci.
USA 94: 6261–6266.
Preuss, M.L., Delmer, D.P. and Liu, B. (2003) The cotton kinesin-like
calmodulin-binding protein associates with cortical microtubules in
cotton fibers. Plant Physiol. 132: 154–160.
Preuss, M.L., Kovar, D.R., Lee, Y.R., Staiger, C.J., Delmer, D.P. and
Liu, B. (2004) A plant-specific kinesin binds to actin microfilaments and
interacts with cortical microtubules in cotton fibers. Plant Physiol. 136:
3945–3955.
Reddy, A.S., Narasimhulu, S.B. and Day, I.S (1997) Structural organization
of a gene encoding a novel calmodulin-binding kinesin-like protein from
Arabidopsis. Gene 204: 195–200.
Richardson, D.N., Simmons, M.P. and Reddy, A.S.N (2006)
Comprehensive comparative analysis of kinesins in photosynthetic
eukaryotes. BMC Genomics 7(18).
Schmit, A.C. (2002) Acentrosomal microtubule nucleation in higher plants.
Int. Rev. Cytol. 220: 257–289.
Vale, R. (2003) The molecular motor toolbox for intracellular transport.
Cell 112: 467–480.
Vaughn, K.C. and Harper, J.D (1998) Microtubule-organizing centers and
nucleating sites in land plants. Int. Rev. Cytol. 181: 75–149.
Vorobjev, I., Malikov, V. and Rodionov, V. (2001) Self-organization
of a radial microtubule array by dynein-dependent nucleation of
micortubules. Proc. Natl. Acad. Sci. USA 98: 10160–10165.
Wasteneys, G.O. (2002) Microtubule organization in the green kingdom:
chaos or self-order? J. Cell Sci. 115: 1345–1354.
Young, A., Dictenberg, J.B., Purohit, A., Tuft, R. and Doxsey, S.J. (2000)
Cytoplasmic dynein-mediated assembly of pericentrin and g-tubulin onto
centrosomes. Mol. Biol. Cell 11: 2047–2056.
(Received January 15, 2007; Accepted April 12, 2007)