Plant Cell Physiol. 46(7): 1083–1092 (2005)
doi:10.1093/pcp/pci121, available online at www.pcp.oupjournals.org
JSPP © 2005
Caffeine Inhibits Callose Deposition in the Cell Plate and the Depolymerization
of Microtubules in the Central Region of the Phragmoplast
Hiroki Yasuhara *
Department of Biotechnology, Faculty of Engineering and High Technology Research Center (HRC), Kansai University, Yamate-cho, Suita, Osaka,
564-8680 Japan
;
Treatment of tobacco BY-2 cells with 10 mM caffeine
that was started after the cells had entered the mitotic
phase did not completely inhibit the deposition of callose in
the cell plate and allowed the centrifugal redistribution of
phragmoplast microtubules. On the other hand, when
treatment with caffeine was started before the cells entered
the mitotic phase, the deposition of callose was completely
inhibited and the redistribution of phragmoplast microtubules was also inhibited. As the inhibition of redistribution of phragmoplast microtubules seems to be caused by
the inhibition of depolymerization of microtubules at the
central region of the phragmoplast, these results strongly
suggest that the deposition of callose in the cell plate is
tightly linked with the depolymerization of phragmoplast
microtubules. Callose deposition was observed in phragmoplasts isolated from caffeine-treated cells as well as in those
isolated from non-caffeine-treated cells, and caffeine did
not inhibit callose synthesis in isolated phragmoplast, indicating that caffeine neither inhibits the accumulation of
callose synthase at the equatorial regions of the phragmoplast nor arrests callose synthase itself.
Keywords: Caffeine — Callose — Cell plate —– Cytokinesis
— Phragmoplast — Tobacco BY-2 cells.
Abbreviations: BFA, brefeldin A; DAPI, 4′,6′-diamino-2-phenylindole; MAPK, mitogen-activated protein kinase.
Introduction
In cytokinesis in higher plants, the parental cytoplasm is
divided into two by the centrifugally developed cell plate
which is formed under the control of the phragmoplast. The
cell plate is formed through the coalescence of Golgi-derived
vesicles that have been accumulated at the equatorial region of
the phragmoplast (Hepler 1982, Samuels et al. 1995), where
plus ends of oppositely oriented sets of microtubules are interdigitating (Euteneuer and McIntosh 1980). The phragmoplast
first appears between two daughter chromosomes in late anaphase as a compact cylindrically arranged microtubule array
(Zhang et al. 1990, Kumagai et al. 2001). Then the array
gradually develops into a barrel-shaped array by increasing the
*
number of microtubules. When a small cell plate at the center
of the array appears, microtubules begin to redistribute to
assume a characteristic donut-like array. The phragmoplast
microtubules continue to redistribute centrifugally as the cell
plate develops, until its outer edge fuses with the parental cell
wall.
The development of the phragmoplast is thought to be
caused by the depolymerization of microtubules at the central
region or the inner margin of the phragmoplast and the polymerization of microtubules at the outer margin (Gunning 1982).
In fact, the occurrence of rapid microtubule turnover in the
phragmoplast has been demonstrated using fluorescence recovery after a photobleaching technique in cells microinjected with
fluorescently labeled tubulin (Hush et al. 1994). The polymerization of microtubules at the outer margin of the phragmoplast
requires the depolymerization of pre-existing microtubules.
This was clearly shown by the finding that the inhibition of
microtubule depolymerization by taxol inhibited the centrifugal development of the phragmoplast (Yasuhara et al. 1993).
We previously demonstrated that treatment of BY-2 cells with
brefeldin A (BFA), a drug that disrupts the Golgi apparatus,
caused the formation of phragmoplasts with a few cell plate
vesicles (Yasuhara and Shibaoka 2000). In the vesicle-depleted
phragmoplast, the development of the phragmoplast was inhibited due to the inhibition of depolymerization of microtubules
at the central region of the phragmoplast. This result suggests
the presence of mechanisms that harmonize the progression of
development of the cell plate and of the phragmoplast. To harmonize the progression of two processes, one must trigger the
other or both must be triggered by the same signal. The question must be raised of what event(s) or signal(s) triggers the
redistribution of the phragmoplast microtubules. Recently,
Nishihama et al. (Nishihama et al. 2001, Nishihama et al. 2002)
reported that the inhibition of function of the NACK1–NPK1
complex by a dominant-negative strategy inhibited the redistribution of phragmoplast microtubules. Their results showed
the involvement of the mitogen-activated protein kinase
(MAPK) cascade in the control of the redistribution of
phragmoplast microtubule. However, events which link the
MAPK cascade with phragmoplast microtubule redistribution
remain to be identified.
The arrested redistribution of phragmoplast microtubules
similar to that observed in BFA-treated BY-2 cells has been
Corresponding author: E-mail, [email protected]; Fax, +81-06-6388-8609.
1083
1084
Callose deposition and phragmoplast development
shown in Tradescantia stamen hair cells treated with caffeine, a
well known inhibitor of plant cytokinesis (Valster and Hepler
1997). Samuels and Staehelin (1996) reported that treatment of
BY-2 cells with caffeine applied at metaphase inhibited the
deposition of callose in the cell plate and the conversion of the
fragile, thin, fusion tube-generated membrane network (FTN)
into the more stable tubulo-vesicular network (TVN) which
normally occurs in forming cell plates. However, they did not
mention the effects of caffeine on the redistribution of phragmoplast microtubules. In the present study, I re-examined the
effects of caffeine on the deposition of callose in the cell plate
and also examined the effects of caffeine on the redistribution
of phragmoplast microtubules in BY-2 cells.
The results obtained were very interesting. When caffeine
was applied to metaphase cells, the deposition of callose in the
cell plate was not completely inhibited and the redistribution of
phragmoplast microtubules took place. On the other hand,
when caffeine was applied to cells before the cells enter the
mitotic phase, the deposition of callose was completely inhibited and the redistribution of phragmoplast microtubules was
also inhibited. This suggests that the deposition of callose in
the cell plate is tightly related to the redistribution of phragmoplast microtubules. Furthermore, to determine the mechanisms
of inhibition of callose deposition by caffeine, the effects of
caffeine on the synthesis of callose in isolated phragmoplasts
were examined.
Results
Effects of caffeine on callose deposition and on the redistribution of phragmoplast microtubules
After the termination of propyzamide treatment, cells
were treated with or without 10 mM caffeine for 120 min and
stained for microtubules, nuclei and cell plates (Fig. 1a, protocol 1). Cell plates were stained with aniline blue, which is specific for callose (Wood and Fulcher 1984). In non-caffeinetreated control cells, aniline blue-positive cell plates were
observed between telophase nuclei (Fig. 1b, 2f, g) and the
phragmoplast developed to assume a ring, ‘C’ or band shape
(Fig. 1e, 2a, b). In a large portion of the cells treated with caffeine (Fig. 1a, protocol 1), signals of callose between telophase nuclei were still observed, but they were weaker than
those observed in untreated control cells (Fig. 1c, 2h), indicating that the caffeine treatment partially inhibited the deposition
of callose in the cell plate. This observation is consistent with
the results of Samuels and Staehelin (1996) who examined the
effects of caffeine on the deposition of callose by immunoelectron microscopy. They showed that the signal of callose in
the cell plate was significantly reduced in caffeine-treated cells
but it was still present. Although the deposition of callose in
the cell plate was inhibited, phragmoplasts in the caffeinetreated cells developed as those in the control cells (Fig. 1f,
2c). The redistribution of phragmoplast microtubules was not
affected by caffeine treatment in the case in which caffeine
treatment was started after the termination of treatment with
propyzamide.
In cells treated with caffeine before and during the mitotic
phase, namely in cells treated with caffeine during 5 h treatment with propyzamide and again 2 h after the termination of
treatment with propyzamide (Fig. 1a, protocol 2), the signal of
callose between telophase nuclei was rarely observed (Fig. 1d,
2i, j). The phragmoplast did not assume a ring, ‘C’ or band
shape, and microtubules in the central region of the phragmoplast were still present (Fig. 1g, 2d, e). At the time of the observation, namely 2 h after the termination of treatment with
propyzamide, a large portion of the phragmoplasts observed in
caffeine-treated cells (Fig. 2d, e) were similar in size; about
30 µm in diameter, which was almost the same as that of the
ring-shaped phragmoplast in the control cells (Fig. 2a). The
phragmoplast with microtubules in the central region never
exceeded this size. None of the cells was observed to have a
phragmoplast with microtubules in the entire region of the division plane even after prolonged caffeine treatment. Phragmoplasts in caffeine-treated cells resemble in shape those in cells
in which the supply of cell plate vesicles was completely
arrested by BFA treatment (Yasuhara and Shibaoka 2000).
Cells treated with caffeine according to the protocol
shown in Fig. 3a were sampled every 30 min after the termination of treatment with propyzamide, and double stained for
microtubules and nuclei. The percentage of cells at metaphase,
that of cells with phragmoplasts at early developmental stages
which had microtubules at central regions and that of cells with
phragmoplasts at late developmental stages which no longer
had microtubules in central regions was examined in each sample and plotted against time (Fig. 3b). Phragmoplasts that
assume a ring, ‘C’ or band shape were classified as those in the
late developmental stages. Irrespective of caffeine treatment,
the percentage of cells with phragmoplasts increased as the
percentage of metaphase cells decreased during the first
30 min. The increment of cells with phragmoplasts occurred
faster in caffeine-treated cells than in control cells, suggesting
that caffeine accelerated the transition of the microtubule array
from spindles to phragmoplasts. The percentage of cells with
phragmoplasts at early developmental stages increased with
time until 90 min and then decreased both in the control cells
and in caffeine-treated cells. However, the maximum value was
far higher in caffeine-treated cells than in control cells. These
results indicate that the phragmoplast at early developmental
stages rapidly developed into later stages in control cells but
not in caffeine-treated cells. In caffeine-treated cells, the transition of phragmoplast from early developmental stages to later
stages is delayed for nearly 1 h.
To examine the dependence of the degree of the inhibition of callose deposition in the cell plate and of the development of the phragmoplast on the timing of caffeine treatment,
caffeine treatment was started at various points of time after the
termination of treatment with aphidicolin according to the
protocols shown in Fig. 4a. Caffeine-treated cells and untreated
Callose deposition and phragmoplast development
1085
Fig. 1 Effects of caffeine on callose deposition in the cell plate and on the development of the phragmoplast. (a) Protocols for caffeine treatment. aph., aphidicolin; prop., propyzamide. (b, c, d, h and i) Micrograph of cells stained for DNA with DAPI (blue) and for callose with aniline
blue (yellowish-white). (e, f, g, j and k) Micrograph of cells immunostained for microtubules. Scale bar: 50 µm. (l) Recovery of development of
the phragmoplast after the termination of treatment with caffeine. Meta., the percentage of metaphase cells; Ana., the percentage of anaphase
cells; Early phrag., the percentage of cells having phragmoplasts at early stages; Late phrag., the percentage of cells having phragmoplasts at late
stages; Others, the percentage of cells other than the above four cell types. Most of the cells belonging to ‘Others’ are interphase cells that have
not entered the mitotic phase during the period of the experiment. At least 500 cells were examined for each determination.
1086
Callose deposition and phragmoplast development
Fig. 2 Inhibition of callose deposition
in the cell plate and the depolymerization
of microtubules in the central region of
the phragmoplast by caffeine. Cells were
treated with or without (a, b, f, g, k and l)
caffeine according to protocol 1 (c, h and
m) or protocol 2 (d, e, i, j, n and o) in Fig.
1a. Cells harvested 120 min after the termination of treatment with propyzamide
were triple stained for microtubules,
nuclei and cell plates. (a–e) Images of
immunostained microtubules (red). (f–j)
Images of nuclei (blue) and cell plates
(yellowish-white). (k–o) Merged images.
Scale bar: 10 µm.
Fig. 3 Effects of caffeine on mitosis
and on the development of the phragmoplast. (a) Protocol for the caffeine treatment. Cells were sampled every 30 min
after the termination of treatment with
propyzamide. (b) The percentages of
metaphase cells (Meta.), those of cells
with phragmoplasts at early developmental stages (Early phrag.), those of
cells with phragmoplasts at late developmental stages (Late phrag.), those of
cells that have completed cytokinesis or
those of cells that have not completed
cytokinesis but have no phragmoplast
(No phrag.) were plotted against time.
control cells were harvested 120 min after the termination of
treatment with propyzamide, double stained for cell plates and
nuclei and the percentage of telophase cells with an aniline
blue-positive cell plate and that of telophase cells without an
aniline blue-positive cell plate (Fig. 4b) was examined, or they
were double stained for microtubules and nuclei and the per-
centage of cells with phragmoplast at early developmental
stages and that of cells with phragmoplast at late developmental stages was examined (Fig. 4c). In cells in which caffeine
treatment had been started at 7 h before the termination of
treatment with propyzamide, when the mitotic index was still
0%, the percentage of cells with aniline blue-positive cell
Callose deposition and phragmoplast development
plates was virtually 0% and that of cells with phragmoplasts at
late developmental stages was about 15%. As the start of treatment with caffeine was delayed, the percentage of cells with an
aniline blue-positive cell plate and that of cells with phragmoplasts at late developmental stages increased up to about 70 and
65%, respectively, in parallel with the increase in the mitotic
index examined at the time of the start of caffeine treatment.
These results indicate that in cells that have been treated with
caffeine before and during the mitotic phase, the deposition of
callose in the cell plate does not occur and the phragmoplast
1087
develops very slowly, but in cells treated with caffeine after the
cells entered the mitotic phase the deposition of callose in the
cell plate does occur and the phragmoplast develops rapidly as
in caffeine-untreated cells.
Recovery from the caffeine-induced inhibition of callose deposition and of development of the phragmoplast after the termination of treatment with caffeine
Cells were cultured in the presence of caffeine and propyzamide for 5 h and then in the presence of only caffeine for
90 min. Cells were then cultured in the presence (Fig. 1a, protocol 3) or absence of caffeine (Fig. 1a, protocol 4). Cells were
sampled immediately before the termination of the first caffeine treatment, i.e. 90 min after the termination of propyzamide treatment, and 30 min after that time they were stained
for microtubules, nuclei and cell plates. During the 90 min culture after the termination of treatment with propyzamide in the
presence of caffeine, about 85% of cells formed phragmoplasts, but a majority of the phragmoplasts remained at early
developmental stages (Fig. 1j, l) and the signal of callose
between telophase nuclei was not observed (Fig. 1h). The signal of callose between telophase nuclei was still absent when
the cells were cultured again in the presence of caffeine, but the
signal appeared when the cells were cultured in the absence of
caffeine (Fig. 1i). The percentage of cells with phragmoplasts
at early developmental stages did not decrease greatly when the
cells were cultured again in the presence of caffeine (Fig. 1l),
but the percentage of phragmoplasts at early developmental
stages greatly decreased and, corresponding to this, the percentage of cells with phragmoplasts at late developmental
stages increased when the cells were cultured in the absence of
caffeine (Fig. 1l). The percentage of binucleated cells in cells
sampled 14.5 h after the termination of the first caffeine treatFig. 4 Effects of caffeine treatments that started at various time after
the termination of treatment with aphidicolin on the development of
the phragmoplast. (a) Protocols for the caffeine treatments. Cells were
sampled at the time indicated by ‘S’, i.e. 120 min after the termination
of treatment with propyzamide. (b and c) Results of the experiments
that were carried out according to each of the protocols –7 to 0 and of
the non-caffeine-treated control experiment (untreated). The experiments for (b) and (c) were carried out separately. (b) Cells were double stained with aniline blue and DAPI for callose in the cell plate and
nuclei and examined for the percentage of telophase cells with an aniline blue-positive cell plate (Telo +CP) and that of telophase cells
without an aniline blue-positive cell plate (Telo –CP) and the percentage of cells other than the above two types of cells (Others). (c) Cells
were double stained for microtubules and nuclei and examined for the
percentage of cells with phragmoplasts at late developmental stages
(Late phrag.), the percentage of cells with phragmoplasts at early
developmental stages (Early phrag.) and the percentage of cells other
than the above two types of cells (Others). Most of the cells belonging
to ‘Others’ are interphase cells that have not entered the mitotic phase
during the period of the experiment. The mitotic indices that were
determined immediately after the start of caffeine treatment (open circles) are also included. At least 500 cells were examined for each
determination.
1088
Callose deposition and phragmoplast development
Fig. 5 Effects of caffeine on the accumulation of vesicles at the equatorial region of the phragmoplast. Cells were harvested 120 min after the
termination of treatment with propyzamide and examined by electron microscopy. (a and b) Cells not treated with caffeine. (c and d) Cells treated
with caffeine according to protocol 1 in Fig. 1a. (e and f) Cells treated with caffeine according to protocol 2 in Fig. 1a. Arrows indicate the
absence of cell plate vesicles. (b, d and f) Higher magnification view of the growing margin of the phragmoplast boxed in (a), (c) and (e). The
accumulation of vesicles at the growing margins of the phragmoplast was not inhibited by caffeine. Scale bars: 2 µm (a, c and e) and 1 µm (b, d
and f).
ment was examined. The percentages of binucleated cells were
91.0% in cells cultured again in the presence of caffeine (protocol 3 in Fig. 1a) and 18.5% in cells cultured in its absence (protocol 4 in Fig. 1a). The percentage of binucleated cells in
control cells which had not been subjected to any caffeine
treatment was 10.9%. Therefore, only 7.6% of the cells did not
resume cytokinesis after removal of caffeine. These results
indicate that the inhibitory effects of caffeine on the deposition
of callose in the cell plate and on the development of the
phragmoplast are highly reversible.
Ultrasutucture of caffeine-treated cells
Since the inhibition of redistribution of phragmoplast
microtubules observed in caffeine-treated cells resembles that
observed in cells in which accumulation of cell plate vesicles is
completely arrested by BFA treatment (Yasuhara and Shibaoka
2000), examinations were made to ascertain whether or not the
accumulation of the cell plate vesicle is inhibited in caffeinetreated cells. Cells treated with caffeine according to protocol 1
and protocol 2 in Fig. 1a and non-treated cells were harvested
120 min after the termination of treatment with propyzamide
and then examined by electron microscopy. As shown in Fig. 5,
vesicles were present throughout the equatorial region of the
phragmoplast in non-treated cells (Fig. 5a, b). In caffeinetreated cells, many vesicles were present at the growing margin of the phragmoplast (Fig. 5c, d, e, f) but few vesicles were
present at the central region of the phragmoplast (Fig. 5c, e,
arrows). This is consistent with the previous report that caffeine inhibits the consolidation of the cell plate, and partly
formed cell plates degrade to vesicles and become resorbed
(Hepler and Bonsignore 1990). Although the maturation and
the consolidation of the cell plate were inhibited, the presence
of vesicles at the growing margin of the phragmoplast indicates that the accumulation of the cell plate vesicles itself was
Callose deposition and phragmoplast development
1089
plasts were stained for callose and DNA. The signal of callose
in the cell plate was rarely observed in phragmoplasts incubated without UDP-glucose (Fig. 6a). Callose in the cell plate
that had been synthesized in vivo before the isolation of the
phragmoplast was faintly or not stained in the isolated phragmoplasts, as has been reported by Kakimoto and Shibaoka (1992).
The significant signal of callose in the cell plate was observed
both in phragmoplasts isolated from non-caffeine-treated cells
(Fig. 6b) and in those isolated from caffeine-treated cells (Fig.
6d) after the incubation with callose synthesis buffer which
contained UDP-glucose, indicating that callose was synthesized even in the phragmoplasts isolated from caffeine-treated
cells. Furthermore, these results also indicate that caffeine does
not inhibit the transport of callose synthase to the equatorial
region of the phragmoplast. Since callose synthase has been
suggested to be transported to the equatorial region via cell
plate vesicles (Hong et al. 2001), the fact that caffeine does not
inhibit the transport of callose synthase to the equatorial region
is consistent with the view that caffeine did not inhibit the
accumulation of cell plate vesicles (Fig. 5). Caffeine added to
callose synthesis buffer did not weaken the signal of callose in
the cell plate observed after the incubation of isolated phragmoplast with the buffer (Fig. 6c, e), indicating that caffeine did not
inhibit callose synthesis in isolated phragmoplasts.
Fig. 6 Effects of caffeine on callose synthesis in isolated phragmoplasts. Phragmoplasts were isolated from cells treated with caffeine
according to protocol 2 in Fig. 1a (c and e) or non-caffeine-treated
cells (a, b and d). Isolated phragmoplasts were incubated with (b, c, d
and e) or without (a) UDP-glucose (UDPG) and stained for DNA with
DAPI (blue) and for callose with aniline blue (yellowish-white). Caffeine added to the callose synthesis buffer did not affect callose synthesis by isolated phragmoplasts (d and e). Scale bar: 10 µm.
not inhibited in caffeine-treated cells. The finding that the treatment with caffeine (protocol 2 in Fig. 1a) that did not cause the
inhibition of accumulation of vesicles caused an inhibition of
redistribution of phragmoplast microtubules indicates that the
accumulation of cell plate vesicles alone is insufficient for the
induction of redistribution of phragmoplast microtubules. The
inhibition of redistribution of phragmoplast microtubules does
not seem to be caused by the degradation of the cell plate by
caffeine, because the degradation of the cell plate occurred
both in cells treated with caffeine according to protocol 1 (Fig.
5c) and in cells treated with caffeine according to protocol 2
(Fig. 5e), while the redistribution of phragmoplast microtubules was inhibited in cells treated with caffeine according to
protocol 2 but not in cells treated with caffeine according to
protocol 1.
Effects of caffeine on callose synthesis in isolated phragmoplasts
Phragmoplasts were isolated from cells treated with or
without caffeine for 5 h during the treatment with propyzamide. After incubation with or without UDP-glucose, phragmo-
Discussion
The depolymerization of microtubules at the central region of
the phragmoplast is tightly related to callose deposition in the
cell plate
Cells whose cell cycle was synchronized by using aphidicolin and propyzamide were treated with caffeine during and
after the treatment with propyzamide. The deposition of callose in the cell plate was completely inhibited when treatment
with caffeine started before the mitotic index began to increase,
but it was partially inhibited when treatment with caffeine
started after the termination of treatment with propyzamide.
Bonsignore and Hepler (1985) showed that caffeine easily
reaches the target site and disrupts normal plate formation in
Tradescantia stamen hair cells. In their experiments, treatment
with caffeine that started 10 min before the onset of cell plate
formation inhibited cytokinesis in all the cells examined. In
BY-2 cells, treatment with caffeine that started after the termination of treatment with propyzamide completely inhibited the
completion of cell plate formation (Yasuhara unpublished
observation), indicating that caffeine also easily penetrates into
BY-2 cells. Therefore, the partial inhibition of callose deposition by caffeine treatment that started after the termination of
treatment with propyzamide would not be due to insufficient
penetration of caffeine into the cells. Possibly, caffeine-sensitive processes, which lead to callose deposition, start before or
the moment cells enter the mitotic phase. Thus, caffeine does
not completely inhibit callose deposition when administered
after the cells enter the mitotic phase.
1090
Callose deposition and phragmoplast development
Fig. 7 Schematic representations of two hypothetical mechanisms
which link callose deposition with microtubule depolymerization. (a)
The caffeine-sensitive signaling process might induce callose synthesis, and the signal of callose deposition might transfer to the system for
microtubule depolymerization. (b) The caffeine-sensitive signaling
process might be linked with both a pathway for induction of callose
deposition and a pathway for induction of microtubule depolymerization. The NPK1-mediated signaling process, which might induce
microtubule depolymerization, seems to be downstream of the caffeine-sensitive process. See text for further explanations.
The redistribution of the phragmoplast microtubules
occurred in cells in which deposition of callose in the cell plate
occurred, i.e. in cells without caffeine treatment and in cells in
which caffeine treatment started after the mitotic index began
to increase. However, the redistribution of the phragmoplast
microtubules is greatly retarded in cells in which the deposition of callose in the cell plate did not occur, i.e. in cells in
which caffeine treatment started before the mitotic index began
to increase. In caffeine-treated cells, microtubules were present
in the central region of the phragmoplast whose diameter was
almost the same as that of the ring-shaped phragmoplast in
untreated cells (Fig. 2), indicating that the depolymerization of
microtubules in the central region of the phragmoplast was
inhibited, while the polymerization of microtubules at the outer
margin occurred until free tubulin in the cells was exhausted.
The cessation of the further growth of the phragmoplast in caffeine-treated cells seems to be due to the inhibition of the depolymerization of microtubules in the central region, which shut
off the supply of tubulin for the polymerization of microtubules at the outer margin. The deposition of callose in the cell
plate seems to be tightly related to the depolymerization of
microtubules at the central region of the phragmoplast.
Mechanisms which link callose deposition with microtubule
depolymerization
The following three possible mechanisms can be postulated to account for the concomitant inhibition of callose deposition and the phragmoplast microtubule depolymerization.
First, caffeine inhibits callose deposition that might trigger the
depolymerization of phragmoplast microtubules. Secondly, caffeine inhibits the depolymerization of phragmoplast micro-
tubules that might trigger the callose deposition. Thirdly,
caffeine might inhibit some signaling steps that trigger both the
callose deposition and the microtubule depolymerization. The
second possibility seems to be untenable because callose has
already been deposited at the equatorial region of the phragmoplast at early developmental stages where the depolymerization of microtubules has not yet occurred (Samuels and
Staehelin 1996, Yasuhara and Shibaoka 2000). In knolle and
keule mutant cells, which accumulate unfused vesicles at the
equatorial region of the phragmoplast, the depolymerization
of phragmoplast microtubules at the central region occurs
(Waizenegger et al. 2000, Strompen et al. 2002). Since the vesicle fusion step occurs before the deposition of callose in the
forming cell plate (Samuels et al. 1995, Samuels and Staehelin
1996), the deposition of callose could be inhibited in these
mutant cells. If the deposition of callose in the cell plate is
completely inhibited in the mutant cells, the occurrence of
microtubule depolymerization at the central region of the
phragmoplast indicates that the callose deposition is not necessary for the depolymerization of phragmoplast microtubules
and thus the first possibility would be untenable. It would be
important to examine whether callose is present in the equatorial region of the phragmoplast of knolle and keule cells. The
induced expression of NPK1KW, a kinase-negative version
of the tobacco MAPK kinase kinase, NPK1, in BY-2 cells
has been shown to cause the inhibition of redistribution of
phragmoplast microtubules similar to that observed in caffeinetreated cells (Nishihama et al. 2001). However, accumulation
of callose in the incomplete cross wall was observed in cells
expressing NPK1KW (Nishihama et al. 2001). This suggests
that the NPK1-mediated signaling process is downstream of the
caffeine-sensitive signaling process. Therefore, if the first possibility is the case, the signal of callose deposition might transfer from the system for microtubule depolymerization via the
NPK1-mediated signaling process (Fig. 7a). Alternatively, if
the third possibility is the case, the caffeine-sensitive signaling
process might be linked with the two pathways, namely a pathway for induction of callose deposition and that for induction
of microtubule depolymerization, and only the latter might be
controlled by the NPK1-mediated signaling process (Fig. 7b).
What are the caffeine-sensitive signaling processes that
might trigger callose synthesis and/or the depolymerization of
microtubules? Although there is no direct evidence, it has been
proposed that caffeine inhibits cytokinesis through the disruption of the mechanisms which regulate the level of calcium ion.
The importance of the evidence of localized calcium gradients
in cell plate formation has been shown by experiments which
employed BAPTA-type calcium ion buffers. The buffers which
remove such gradients caused the disintegration of the cell
plate (Jürgens et al. 1994). The requirement for calcium ions in
callose synthesis at the cell plate has been pointed out by
Samuels and Staehelin (1996) in BY-2 cells and demonstrated
in phragmoplast isolated from BY-2 cells by Kakimoto and
Shibaoka (1992). Thus, the speculation that the inhibition of
Callose deposition and phragmoplast development
callose synthesis by caffeine is due to the changes in membrane-associated calcium at the cell plate seems appropriate. It
has been argued that the endoplasmic reticulum in the
phragmoplast may serve as a reservoir of calcium and play an
important role in regulating the level of free calcium in the
vicinity of the cell plate and, thereby, in controlling vesicle
fusion which leads to cell plate formation (Hepler 1982). However, no experimental evidence which supports this argument
has been provided. To clarify the role of the endoplasmic
reticulum in the phragmoplast in cytokinesis, we should study
the function of the endoplasmic reticulum isolated from isolated phragmoplasts.
The results of experiments in which isolated phragmoplasts were employed are very suggestive. In the presence of
UDP-glucose, phragmoplasts isolated from caffeine-treated
cells synthesized callose as much as those isolated from noncaffeine-treated cells and caffeine did not inhibit callose synthesis in isolated phragmoplasts. These results indicate that
caffeine inhibits neither the transport of callose synthase to the
equatorial regions of the phragmoplasts nor the activity of
callose synthase transported to the equatorial regions. Therefore, the author was tempted to hypothesize that caffeine inhibited callose deposition in living cells by depleting the supply of
substrates for callose synthesis, namely UDP-glucose. To
examine this hypothesis, experiments to determine whether or
not caffeine lowers the intracellular level of UDP-glucose are
now in progress.
Materials and Methods
Plant materials
Suspension-cultured cells, BY-2, of tobacco (Nicotiana tabacum
‘Bright Yellow 2’) were cultured in modified Linsmaier and Skoog’s
medium (modified LS medium) at 27°C as described by Nagata et al.
(1981)
Synchronization of the cell cycle and treatment with caffeine
The progression of the cell cycle was synchronized by sequential treatment with aphidicolin and propyzamide as described previously (Yasuhara and Shibaoka 2000). The mitotic index was
determined as described previously (Yasuhara and Shibaoka 2000).
Caffeine (caffeine, anhydrous; Wako Pure Chemical Industries, Ltd,
Osaka, Japan) was dissolved in modified LS medium at 100 mM and
used at 10 mM. In the experiments in which cells were treated with
caffeine during and after the treatment with propyzamide, cells were
washed with 3% sucrose that contained 10 mM caffeine to terminate
the treatment with propyzamide.
Staining for cell plate, nuclei and microtubules
Double staining for cell plates and nuclei and triple staining for
microtubules, nuclei and cell plates were performed as described previously (Yasuhara and Shibaoka 2000).
Electron microscopy
Cells were fixed by the osmium tetroxide–potassium ferricyanide fixation method and examined under an electron microscope
(JEM 1210; Jeol, Tokyo, Japan) as described previously (Yasuhara and
Shibaoka 2000).
1091
Isolation of phragmoplasts and synthesis of callose in isolated
phragmoplasts
Protoplasts at anaphase–telophase were isolated from cells
treated with or without caffeine by the procedure described previously
(Yasuhara et al. 2002) except for the modifications described below. In
isolating protoplasts from caffeine-treated cells, 10 mM caffeine was
added to the cell wall-digesting enzyme solution. Protoplasts were harvested 90 min after the termination of treatment with propyzamide.
Phragmoplasts were isolated from the protoplasts by passing the protoplasts through nylon mesh and were incubated with callose synthesis
buffer that contained UDP-glucose and calcium ions by the procedure
described by Kakimoto and Shibaoka (1992).
Acknowledgments
The author gratefully acknowledges Professor H. Shibaoka of
Osaka University for critical reading of the manuscript. This work was
supported in part by MEXT. HAITEKU (2002–2006).
References
Bonsignore, C.L. and Hepler, P.K. (1985) Caffeine inhibition of cytokinesis:
dynamics of cell plate formation-deformation in vivo. Protoplasma 129: 28–
35.
Euteneuer, U. and McIntosh, J.R. (1980) Polarity of midbody and phragmoplast
microtubules. J. Cell Biol. 87: 509–515.
Gunning, B.E.S. (1982) The cytokinetic apparatus: its development and spatial
regulation. In The Cytoskeleton in Plant Growth and Development. Edited by
Lloyd, C.W. pp. 229–292. Academic Press, London,
Hepler, P.K. (1982) Endoplasmic reticulum in formation of the cell plate and
plasmodesmata. Protoplasma 111: 121–133.
Hepler, P.K. and Bonsignore, C.T. (1990) Caffeine inhibition of cytokinesis:
ultrastructure of cell plate formation/degradation. Protoplasma 157: 182–192.
Hong, Z., Delauney, A.J. and Verma, D.P.S. (2001) A cell plate specific callose
synthase and its interaction with phragmoplastin. Plant Cell 13: 755–768.
Hush, J.M., Wadsworth, P., Callaham, D.A. and Hepler, P.K. (1994) Quantification of microtubule dynamics in living plant cells using fluorescence redistribution after photobleaching. J. Cell Sci. 107: 775–784.
Jürgens, M., Hepler, L.H., Rivers, B.A. and Hepler, P.K. (1994) BAPTA-calcium
buffers modulate cell plate formation in stamen hairs of Tradescantia: evidence for calcium gradients. Protoplasma 183: 86–99.
Kakimoto, T. and Shibaoka, H. (1992) Synthesis of polysaccharides in phragmoplasts isolated from tobacco BY-2 cells. Plant Cell Physiol. 33: 353–361.
Kumagai, F., Yoneda, A., Tomida, T., Sano, T., Nagata, T. and Hasezawa, S.
(2001) Fate of nascent microtubules organized at the M/G1 interface, as visualized by synchronized tobacco BY-2 cells stably expressing GFP–tubulin:
time-sequence observations of the reorganization of cortical microtubules in
living plant cells. Plant Cell Physiol. 42: 723–732.
Nagata, T., Okada, K., Takebe, I. and Mitsui, C. (1981) Delivery of tobacco
mosaic virus RNA into plant protoplasts mediated by reverse-phase evaporation vesicles (liposomes). Mol. Gen. Genet. 184: 161–165.
Nishihama, R., Ishikawa, M., Araki, S., Soyano, T., Asada, T. and Machida, Y.
(2001) The NPK1 mitogen-activated protein kinase kinase kinase is a regulator of cell-plate formation in plant cytokinesis. Genes Dev. 15: 352–363.
Nishihama, R., Soyano, T., Ishikawa, M., Araki, S., Tanaka, H., et al. (2002)
Expansion of the cell plate in plant cytokinesis requires a kinesin-like protein/
MAPKKK complex. Cell 109: 87–99.
Samuels, A.L., Giddings, T.H. and Staehelin, L.A. (1995) Cytokinesis in
tobacco BY-2 and root tip cells: a new model of cell plate formation in higher
plants. J. Cell Biol. 130: 1345–1357.
Samuels, A.L. and Staehelin, L.A. (1996) Caffeine inhibits cell plate formation
by disrupting membrane reorganization just after the vesicle fusion step.
Protoplasma 195: 144–155.
Strompen, G., El Kasmi, F., Richter, S., Lukowitz, W., Assaad, F.F., Jürgens, G.
and Mayer, U. (2002) The Arabidopsis HINKEL gene encodes a kinesinrelated protein involved in cytokinesis and is expressed in a cell cycledependent manner. Curr. Biol. 12: 153–158.
1092
Callose deposition and phragmoplast development
Valster, A.H. and Hepler, P.K. (1997) Caffeine inhibition of cytokinesis: effect
on the phragmoplast cytoskeleton in living Tradescantia stamen hair cells.
Protoplasma 196: 155–166.
Waizenegger, I., Lukowitz, W., Assaad, F., Schwarz, H., Jürgens, G. and Mayer,
U. (2000) The Arabidopsis KNOLLE and KEULE genes interact to promote
vesicle fusion during cytokinesis. Curr. Biol. 10: 1371–1374.
Wood, P.J. and Fulcher, R.G. (1984) Specific interaction of aniline blue with
(1→3)-β-D-glucan. Carbohydr. Polym. 4: 49–72.
Yasuhara, H., Muraoka, M., Shogaki, H., Mori, H. and Sonobe, S. (2002)
TMBP200, a microtubule bundling polypeptide isolated from telophase BY-2
cells is a MOR1 homologue. Plant Cell Physiol. 43: 595–603.
Yasuhara, H. and Shibaoka, H. (2000) Inhibition of cell-plate formation by
brefeldin A inhibited the depolymerization of microtubules in the central
region of the phragmoplast. Plant Cell Physiol. 41: 300–310.
Yasuhara, H., Sonobe, S. and Shibaoka, H. (1993) Effects of taxol on the development of the cell plate and of the phragmoplast in tobacco BY-2 cells. Plant
Cell Physiol. 34: 21–29.
Zhang, D., Wadsworth, P. and Hepler, P.K. (1990) Microtubule dynamics in living dividing plant cells: confocal imaging of microinjected fluorescent brain
tubulin. Proc. Natl Acad. Sci. USA 87: 8820–8824.
(Received November 10, 2004; Accepted April 28, 2005)