Download Early steps in cold sensing by plant cells: the role of actin

The Plant Journal (2000) 23(6), 785±794
Early steps in cold sensing by plant cells: the role of actin
cytoskeleton and membrane ¯uidity
BjoÈrn LaÂrus OÈrvar*,², Veena Sangwan, Franz Omann and Rajinder S. Dhindsa
Department of Biology, McGill University, 1205 Avenue Docteur Pen®eld, MontreÂal, QueÂbec H3A 1B1, Canada
Received 5 May 2000; revised 23 June 2000; accepted 30 June 2000.
*For correspondence (fax +35 457 07111; e-mail [email protected]).
²
Present address: Technological Institute of Iceland, Keldnaholt, IS-112 Reykjavik, Iceland.
Summary
Many plants acquire freezing tolerance through cold acclimatization (CA), a prolonged exposure to low
but non-freezing temperatures at the onset of winter. CA is associated with gene expression that requires
transient calcium in¯ux into the cytosol. Alfalfa (Medicago sativa) cells treated with agents blocking this
in¯ux are unable to cold-acclimatize. Conversely, chemical agents causing increased calcium in¯ux induce
cold acclimatization-speci®c (cas) gene expression in alfalfa at 25LC. How low temperature triggers
calcium in¯ux is, however, unknown. We report here that induction of a CA-speci®c gene (cas30), calcium
in¯ux and freezing tolerance at 4LC are all prevented by cell membrane ¯uidization, but, conversely, are
induced at 25LC by membrane rigidi®cation. cas30 expression and calcium in¯ux at 4LC are also prevented
by jasplakinolide (JK), an actin micro®lament stabilizer, but induced at 25LC by the actin micro®lament
destabilizer cytochalasin D (CD). JK blocked the membrane rigidi®er-induced, but not the calcium channel
agonist-induced cas30 expression at 25LC. These ®ndings indicate that cytoskeleton re-organization is an
integral component in low-temperature signal transduction in alfalfa cell suspension cultures, serving as a
link between membrane rigidi®cation and calcium in¯ux in CA.
Keywords: cold acclimatization, signalling, actin, membrane ¯uidity.
Introduction
Cold acclimatization (CA) in plants, resulting in acquisition
of increased freezing tolerance, requires exposure to low
but non-freezing temperatures over a periods of days or
weeks (Levitt, 1980). Transient in¯ux of Ca2+ into the
cytosol is an early event during CA (Knight et al., 1991;
Monroy and Dhindsa, 1995). Blocking this Ca2+ in¯ux
prevents cold-induced gene expression and development
of freezing tolerance in alfalfa (Medicago sativa) cells
(Monroy et al., 1993) and in Arabidopsis (Tahtiharju et al.,
1997). Conversely, chemical agents that increase Ca2+
in¯ux, such as the Ca2+ channel agonist Bay K 8644 or
the Ca2+ ionophore A23187, induce CA-speci®c (cas) gene
expression in alfalfa cells at 25°C (Monroy and Dhindsa,
1995). These results demonstrate that cold-induced in¯ux
of Ca2+ is an integral component of the low-temperature
signalling cascade during CA. However, the nature and
temporal sequence of events leading to this Ca2+ in¯ux are
largely unknown.
A direct and perhaps the earliest effect of an alteration of
temperature on cells is the change in ¯uidity of their
ã 2000 Blackwell Science Ltd
membranes (Levitt, 1980). A decrease in temperature
lowers membrane ¯uidity, whereas its increase enhances
membrane ¯uidity (Alonso et al., 1997; Mejia et al., 1995).
Previous studies (Murata and Los, 1997) have suggested
that the plasma membrane acts as the primary sensor of
temperature change through dynamic changes in its
physical state. Rigidi®cation of plasma membranes by
Pd-catalysed hydrogenation leads to the induction of the
cold-inducible gene, desA, encoding a fatty acid desaturase in the cyanobacterium, Synecocystis (Vigh et al., 1993).
It is not known if, and how, membrane ¯uidity modulates
ion channel activity. Mazars et al. (1997) have shown that
cold-induced Ca2+ in¯ux in tobacco protoplasts was
stimulated by disruption of microtubules and actin micro®laments. Recent studies have also demonstrated roles for
actin micro®laments in K+ channel activity in stomatal
opening (Hwang et al., 1997), and in mechanical stress
signalling in plants (Morelli et al., 1998). Although Ca2+
channel proteins have yet to be characterized in plants, a
plant gene (AKT1) encoding a putative K+ channel has
785
786 BjoÈrn LaÂrus OÈrvar et al.
been cloned (Sentenac et al., 1992). The deduced protein
contains ankyrin-like repeats suggesting that ion channel
proteins in plants may interact with the cytoskeleton
(Sentenac et al., 1992).
The above observations led us to explore whether
changes in membrane ¯uidity and re-organization of the
actin cytoskeleton have a role in the early stages of lowtemperature signalling in plant cells. We also investigated
whether these events are linked to Ca2+ in¯ux and CA. We
have used the transcript level of a CA-speci®c gene (cas30)
and the development of freezing tolerance in alfalfa cell
suspension cultures as end-point markers in low-temperature signalling. Our results suggest that cold signalling in
plants may proceed from membrane rigidi®cation through
actin micro®lament re-organization to Ca2+ in¯ux and CAspeci®c gene expression, eventually leading to CA and
increased frost tolerance.
Results
Chemical modulation of membrane ¯uidity in alfalfa
protoplasts
Both DMSO and benzyl alcohol (BA) have been
routinely used as a membrane rigidi®er and ¯uidizer,
respectively. Fluorescence polarization of 1,6-diphenyl1,3,5-hexatriene (DPH) was used to con®rm the effects
of these agents on alfalfa protoplasts, but the polarization index p is inversely related to membrane ¯uidity
(Choi and Yu, 1995). The data presented in Table 1 show
that DMSO treatment reduces membrane ¯uidity at 25°C
as demonstrated by an increase in the polarization
index p. Conversely, pre-treatment of protoplasts with
BA inhibits the cold-induced increase in p, or a decrease
in membrane ¯uidity.
Membrane ¯uidization inhibits cold signalling in alfalfa
cells
To test the hypothesis that membrane ¯uidity plays a
fundamental role in low-temperature signal transduction
in alfalfa, transcript accumulation of a CA-speci®c gene,
cas30, and the development of freezing tolerance in alfalfa
cell suspension cultures were used as end-point markers.
The expression of cas30 is speci®cally induced by cold
treatment and is not responsive to other stresses investigated (Mohapatra et al., 1989), and therefore serves as a
reliable end-point marker for CA. Cold treatment causes a
rapid and reversible accumulation of cas30 transcripts
(Figure 1; data not shown). BA, a well-established membrane ¯uidizer (Carratu et al., 1996; Kitagawa and Hirata,
1992), was used to counter cold-induced membrane
rigidi®cation. When cells were pre-treated with BA for 1 h
at 25°C before being incubated at 4°C for 6 days in the
presence of the alcohol, the cold-induced cas30 expression
was inhibited at all concentrations tested (Figure 2a). BA
treatment also strongly inhibited the development of
freezing tolerance (Figure 2b) without affecting cell viability (data not shown). Similar results were obtained by
treating cells with ethanol, another membrane ¯uidizer
(Mrak, 1992) (data not shown). To investigate whether
lowering of the cas30 transcript levels by BA treatment is
due to a decrease of transcript levels in general, transcript
levels for the replacement histone H3.2 protein and for the
putative L5 ribosomal protein from alfalfa were determined. The H3.2 (Robertson et al., 1996) and L5 (Asemota
et al., 1994) transcripts are constitutively expressed in
alfalfa. Both the L5 and H3.2 transcript levels, which are
not detectably affected after 48 h CA or during deacclimatization (Figure 1), are also not affected by BA
treatment (Figure 2a). Thus the effects of BA on cas30
transcript levels do not re¯ect a general down-regulation
of transcription.
CA in alfalfa suspension cells involves a rapid in¯ux of
Ca2+ into the cytosol as determined by 45Ca2+ cation
exchange chromatography as previously described
(Monroy and Dhindsa, 1995). Isolated protoplasts were
used to measure Ca2+ in¯ux in order to avoid complications due to Ca2+-rich cell walls, but cold-induced accumulation of cas30 transcripts also occurs in isolated
protoplasts (data not shown). The source for the coldinduced Ca2+ in¯ux has been shown to be primarily
apoplastic in alfalfa cells and is strongly inhibited by Ca2+
chelators and Ca2+ channel blockers (Monroy and Dhindsa,
1995). Using 45Ca2+ cation exchange chromatography
(Monroy and Dhindsa, 1995), we examined the effects of
5 mM BA treatment on cold-induced Ca2+ in¯ux. This
concentration of BA was used since it had no effect on
protoplast viability, as determined by Evan's blue staining
(data not shown), and it inhibits cold-induced membrane
rigidi®cation (Table 1). Treatment of protoplasts with 5 mM
BA inhibited Ca2+ in¯ux (Figure 2c). Thus, at 4°C, Ca2+
in¯ux, cas30 gene expression and freezing tolerance are all
prevented by BA-induced cell membrane ¯uidization.
Membrane rigidi®cation activates cold signalling in
alfalfa cells at 25°C
We reasoned that if reduced membrane ¯uidity mediates
the cold-induced cas30 transcript accumulation, Ca2+ in¯ux
and increased freezing tolerance, then reduction of
membrane ¯uidity at 25°C by chemical agents should
mimic these responses. Treatment of cells at 25°C with
DMSO, a well-documented membrane rigidi®er (Lyman
and Preisler, 1976), at concentrations as low as 1%,
induced cas30 transcript accumulation (Figure 3a), but
did not affect the H3.2 and L5 mRNA levels (Figure 3a).
DMSO treatment at 25°C increased membrane rigidi®caã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
Early steps in cold sensing by plant cells 787
Table 1. Polarization index (p) of isolated protoplasts treated
with cold, BA, DMSO, JK or CD
Treatment
Temperature (°C)
Polarization index (p)
Experiment 1
Control
Control
25
4
0.0954 6 0.0073a
0.1315 6 0.0138b
Experiment 2
Control
Control
BA (5 mM)
25
4
4
0.0825 6 0.0043a
0.1173 6 0.0122b
0.0853 6 0.0064a
Experiment 3
Control
Control
DMSO (3%)
25
4
25
0.0870 6 0.0015a
0.1322 6 0.0048b
0.1003 6 0.0029c
Experiment 4
Control
Control
JK (2 mM)
25
4
4
0.0921 6 0.0109a
0.1555 6 0.0136b
0.1525 6 0.0085b
Experiment 5
Control
CD (50 mM)
25
25
0.1084 6 0.0115a
0.1053 6 0.0102a
Data are expressed as the mean 6 SD of three replicates. Values
with different letters are signi®cantly different (P < 0.05) in each
experiment as determined by Student's t-test.
tion in isolated protoplasts (Table 1), enhanced the freezing
tolerance of cells (Figure 3b) and promoted Ca2+ in¯ux
(Figure 3c). Cell and protoplast viability was not affected by
this chemical (data not shown). Therefore, the examined
events characteristic of CA, namely cold-induced cas30
transcript accumulation, Ca2+ in¯ux and increased freezing
tolerance, can be induced by membrane rigidi®cation at
25°C, or prevented at 4°C by membrane ¯uidization.
Figure 1. Cas30 transcript accumulation during 48 h cold acclimatization
(CA), and subsequent disappearance following de-acclimatization (DA) at
25°C.
Total RNA (10 mg) was hybridized to labelled cas30 cDNA. Filters were
stripped and re-probed with labelled L5 and H3.2 cDNAs. Equal loading
of the gel was con®rmed by negative imaging of the RNA blot.
plant actin micro®laments. Treatment with JK inhibits
cold-induced cas30 transcript accumulation (Figure 4a) and
Ca2+ in¯ux (Figure 4b), and lowers freezing tolerance
(Figure 4c). Conversely, treatment of cells with CD caused
both cas30 transcript accumulation (Figure 5a) and increased Ca2+ in¯ux (Figure 5b) at 25°C. However, CD
treatment at 25°C reduced the constitutive level of freezing
tolerance (Figure 5c). Cell or protoplast viability was
unaffected by treatment with JK or CD (data not shown).
Furthermore, membrane ¯uidity (Table 1) and transcript
levels of H3.2 and L5 were apparently unaffected by either
JK or CD treatment (Figure 4a and Figure 5a). Thus, Ca2+
in¯ux and cas30 expression are induced by an actin
micro®lament destabilizing agent, and, conversely, inhibited by an actin micro®lament stabilizing agent.
Disruption of actin micro®laments is required for cold
signalling in alfalfa cells
Temporal order of membrane rigidi®cation and actin
micro®lament re-arrangement in cold signalling in alfalfa
cells
Yamamoto (1989) demonstrated that membrane rigidi®erinduced Ca2+ in¯ux in hepatocytes is accompanied by actin
micro®lament re-organization. It has also been shown that
cold-induced Ca2+ in¯ux in tobacco protoplasts can be
enhanced (almost doubled) by chemical such as
cytochalasin D (CD) that modulate actin micro®lament
organization (Mazars et al., 1997). We therefore examined
whether modulators of actin micro®lament re-organization
mimic the effect of cold-induced rigidi®cation of the
membrane on Ca2+ channels and cas30 expression. To
modulate actin micro®lament organization we used CD
(Cooper, 1987) and jasplakinolide (JK)(Bubb et al., 1994),
both membrane-permeating chemicals that have been
successfully used as a destabilizer (Hwang et al., 1997;
Mazars et al., 1997; Morelli et al., 1998) and stabilizer
(Mathur et al., 1999; Sawitzky et al., 1999), respectively, of
The results presented above suggest that cold-induced
rigidi®cation of membranes and actin micro®lament reorganization are required for Ca2+ in¯ux and cas30
transcript accumulation. However, the temporal order of
these events remained to be determined. By using a
combination of chemical treatments, we presumed that
promoting an earlier event while blocking a subsequent
one should prevent cas30 transcript accumulation.
Conversely, blocking an earlier event but promoting a
downstream one should result in cas30 transcript accumulation. It has been previously demonstrated that the
cold induction of cas30 requires Ca2+ in¯ux (Monroy et al.,
1993). The data in Figure 6 show that cas30 expression
induced at 25°C by either DMSO treatment (membrane
rigidi®er) or by CD treatment (micro®lament destabilizer) is
inhibited by the Ca2+ chelator 1,2-bis(2-aminophenoxy)-
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
788 BjoÈrn LaÂrus OÈrvar et al.
Figure 2. Membrane ¯uidization inhibits cold-triggered cas30 transcript
accumulation, development of freezing tolerance and Ca2+ in¯ux in
alfalfa cells.
(a) Inhibition of cold-induced cas30 transcript accumulation by the
membrane ¯uidizer BA. Cells were treated for 1 h at 25°C with the
indicated concentrations of BA and subsequently placed at 4°C for 24 h
before RNA extraction. Filters were stripped and re-probed with labelled
L5 and H3.2 cDNAs. (b) Inhibition of cold-induced development of
freezing tolerance by BA. Cells were treated for 1 h at 25°C with BA, then
incubated for 6 days at 4°C. Freezing tolerance is expressed as the ratio
(%) of BA-treated versus untreated cells. (c) BA inhibition of coldtriggered Ca2+ in¯ux. Protoplasts were treated with 5 mM BA for 30 min,
followed by a 20 min incubation with 45Ca2+ at 4°C. 45Ca2+ in¯ux is
expressed as the ratio (%) of BA-treated versus untreated (C) protoplasts,
both at 4°C. Although 20 mM BA treatment did not affect the viability of
intact cells, it reduced the viability of isolated protoplasts by 50%,
whereas 5 mM had no effects (data not shown). Error bars in (b) and (c)
represent the standard deviation of the mean of three and four replicates,
respectively.
ethane N,N,N¢,N¢-tetraacetic acid (BAPTA, Figure 6a,b).
These results indicate that Ca2+ in¯ux is downstream of
both membrane rigidi®cation and actin micro®lament reorganization, and further extend the results presented in
Figure 3. Membrane rigidi®cation induces cas30 transcript accumulation,
development of freezing tolerance and Ca2+ in¯ux in alfalfa cells at 25°C.
(a) Induction of cas30 transcript accumulation by the membrane rigidi®er
DMSO at 25°C. Cells were incubated for 3 h with DMSO before RNA
extraction. Filters were stripped and re-probed with labelled L5 and H3.2
cDNAs. (b) Induction of freezing tolerance by DMSO. Cells were treated
for 3 days with DMSO at 25°C. Freezing tolerance is expressed as the
ratio (%) of DMSO-treated versus untreated cells. (c) Induction of Ca2+
in¯ux by DMSO at 25°C, compared to cold-induced Ca2+ in¯ux.
Protoplasts were treated for 30 min with 3% DMSO before incubation
with 45Ca2+ at 25°C. Error bars in (b) and (c) represent the standard
deviation of the mean of three and four replicates, respectively.
Figures 3(c) and 5(b). The DMSO-induced cas30 expression
at 25°C is also inhibited by the actin micro®lament
stabilizer, JK (Figure 6c). However, the induction of cas30
at 25°C by the Ca2+ channel agonist Bay K 8644, although
sensitive to BAPTA, is not inhibited by JK (Figure 6d).
These data demonstrate that cold-induced re-organization
of actin micro®laments is located downstream of membrane rigidi®cation but upstream of Ca2+ in¯ux.
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
Early steps in cold sensing by plant cells 789
2+
Figure 4. Cold-induction of cas30 transcript accumulation, Ca in¯ux and
freezing tolerance are inhibited by the actin micro®lament stabilizer
jasplakinolide (JK).
(a) Inhibition of the cold-induced cas30 transcript accumulation by JK.
RNA was isolated from cells treated for 1 h at 25°C with JK (2 mM) or
solvent vehicle alone (C; 0.15% DMSO), followed by 24 h incubation at
4°C. Filters were stripped and re-probed with labelled L5 and H3.2
cDNAs. (b) Inhibition of cold-induced Ca2+ in¯ux by JK. Protoplasts were
treated for 30 min with JK (2 mM) or solvent vehicle alone (C) at 25°C
before incubation with 45Ca2+ at 4°C for 20 min. (c) Inhibition of coldinduced development of freezing tolerance by JK. Cells were treated for
1 h at 25°C with JK (2 mM) or solvent vehicle alone and then incubated for
6 days at 4°C. Freezing tolerance of JK-treated cells is expressed as the
ratio (%) of that of cells treated with solvent vehicle alone. Error bars in
(b) and (c) represent the standard deviation of the mean of four
replicates.
Discussion
The underlying mechanisms of low temperature sensing
and signal transduction in plants are poorly understood.
Here we demonstrate that, in alfalfa cells, (a) cold-induced
Ca2+ in¯ux, cas30 transcript accumulation and development of freezing tolerance are inhibited by chemicals that
either ¯uidize membranes or stabilize actin micro®laments; (b) chemicals that either rigidify membranes or
destabilize actin micro®laments cause Ca2+ in¯ux and
cas30 expression at 25°C; and (c) the role of actin
cytoskeleton in low-temperature signal transduction is
located downstream of membrane rigidi®cation but upstream of Ca2+ in¯ux.
The importance of membrane ¯uidity in low-temperature perception has long been recognized (Levitt, 1980;
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
Figure 5. cas30 transcripts accumulation and Ca2+ in¯ux are induced by
the actin micro®lament destabilizer cytochalasin D (CD).
(a) CD-induced cas30 mRNA accumulation at 25°C. RNA was isolated
from cells treated for 3 h at 25°C with CD (50 mM) or solvent vehicle alone
(C; 0.64% DMSO). (b) Induction of Ca2+ in¯ux by CD. Protoplasts were
treated for 30 min with CD (50 mM) or solvent vehicle alone (C) at 25°C
before incubation with 45Ca2+ for 20 min. (c) Inhibition of cold-induced
development of freezing tolerance by CD. Cells were treated for 3 days
with CD (50 mM) or solvent vehicle alone at 25°C. Freezing tolerance was
calculated as the percentage of that of cells subjected to solvent vehicle
treatment alone. Error bars in (b) and (c) represent the standard deviation
of the mean of four replicates.
Murata and Los, 1997). However, its relationship to
subsequent responses to cold has been unclear. Our
results show that DMSO treatment induces membrane
rigidi®cation and mimics the effects of cold on gene
expression, Ca2+ in¯ux and cold acclimatization.
Furthermore, counteracting the cold-induced membrane
rigidi®cation with chemical ¯uidizers halts all examined
responses to cold, suggesting that membrane rigidi®cation is not only a suf®cient but a necessary condition for
cold signalling in plant cells. It should be noted, however,
that although 3% DMSO has almost saturating effects on
freezing tolerance, its effect on cas30 expression is not
saturated. While the reasons for this are unclear, it is
possible that DMSO induces a battery of cas genes,
besides cas30, leading to enhanced freezing tolerance.
Alternatively, DMSO may also be exerting its well-known
cryoprotective effect which may or may not be related to
its effect on membrane rigidi®cation.
It is widely accepted that Ca2+ in¯ux into the cytosol is
triggered by cold (Knight et al., 1991) and is required for
790 BjoÈrn LaÂrus OÈrvar et al.
Figure 6. Cold-induced re-organization of actin micro®laments provides a
link between membrane rigidi®cation and Ca2+ in¯ux.
(a) Induction of cas30 transcript accumulation by DMSO at 25°C is
inhibited by the Ca2+ chelator BAPTA. Cells were treated for 1 h with
BAPTA (BP; 2 mM) and then for 3 h with DMSO (DM; 3%). Control cells
were treated with DMSO (3%) or culture medium alone (C) for 3 h. (b)
Induction of cas30 transcript accumulation by CD is inhibited by BAPTA.
Cells were treated for 1 h with BAPTA (2 mM) and then for 3 h with CD
(50 mM). Control cells were treated with CD (50 mM) or solvent vehicle
alone (C; 0.64% DMSO) for 3 h. (c) Induction of cas30 transcript
accumulation by DMSO at 25°C is inhibited by JK. Cells were treated for
1 h with JK (1 mM) and then for 3 h with DMSO (3%) at 25°C. Control cells
were treated as described in (a). (d) Induction of cas30 transcript
accumulation at 25°C by the Ca2+ channel agonist Bay K 8644 is inhibited
by BAPTA but not by JK. Cells were treated for 1 h with either JK (1 mM)
or BAPTA (2 mM) and then for 3 h with Bay K 8644 (BK; 100 mM). Control
cells were treated with Bay K 8644 (100 mM) or solvent vehicle alone (C)
alone for 3 h.
cold-induced gene expression and development of freezing tolerance (Knight et al., 1996; Monroy and Dhindsa,
1995; Monroy et al., 1993; Tahtiharju et al., 1997). The data
presented here indicate that cold-triggered Ca2+ in¯ux is
downstream of membrane rigidi®cation, because the
in¯ux is inhibited by membrane ¯uidization (Figure 2c)
but triggered at 25°C by membrane rigidi®cation
(Figure 3c). Although membrane rigidi®cation-induced
Ca2+ in¯ux has not been reported in plants, DMSO-induced
membrane rigidi®cation induces Ca2+ in¯ux in hepatocytes
(Yamamoto, 1989) and parathyroid cells (Nygren et al.,
1987). Cholesterol, another frequently used modulator of
membrane ¯uidity, increases membrane rigidity and
causes Ca2+ in¯ux in rabbit smooth muscle cells
(Gleason et al., 1991). Thus, membrane rigidi®cation and
Ca2+ in¯ux appear to be coupled. It should be noted that
our data indicate little about the changes in actual
concentration of Ca2+ in the cytosol and refer strictly to
in¯ux of extracellular Ca2+ into the cell. The transient rise
in cytosolic Ca2+ concentration due to cold shock is well
established through extensive research by many researchers, notably by Trewavas and colleagues (Knight et al.,
1991).
In our study, treatment with CD causes both Ca2+ in¯ux
(Figure 5b) and cas30 expression at 25°C (Figure 5a).
Conversely, treatment with JK inhibits both these events
at 4°C (Figure 3a,b). JK can simultaneously stabilize one
population of actin micro®laments and destabilize another, or rapidly induce nucleation of actin polymerization,
depending on the cellular domain (Spector et al., 1999).
Immunochemical estimation of F-actin levels has shown
that JK pre-treatment causes a dramatic increase in F-actin
levels in cold-shocked cells (OÈrvar et al., unpublished data),
indicating that JK stabilizes the micro®laments. However,
the mechanism of CD action is quite complex. CD can cap
the barbed end of actin micro®laments, severe the
micro®laments and sequester actin monomers, and thus
affect the abundance and organization of micro®laments
(Cooper, 1987; Goddette and Frieden, 1986; Sampath and
Pollard, 1991). Although latrunculin B, a powerful actin
micro®lament-disrupting agent (Spector et al., 1983), has
similar effects on cas30 expression as does CD (data not
shown), it is also plausible that the cold-induced Ca2+ entry
mimicked by CD treatment (Figure 5b), does not require
depolymerizaton of micro®laments, but instead remodelling of the pre-existing network. To date, although lowtemperature disruption of the actin cytoskeleton has been
demonstrated in animal cells (Callaini et al., 1991; Watts
and Howard, 1992), there is no direct evidence that low
temperature disrupts the actin cytoskeleton in plants.
However, it should be noted that a wheat gene encoding
an actin depolymerization factor (ADF/co®lin) is rapidly upregulated by cold (Danyluk et al., 1996). It is noteworthy, in
this respect, that a recently characterized actin-defective
yeast mutant is cold-sensitive (McCollum et al., 1999).
Treatment of cells with 2 mM JK inhibits cold-induced
Ca2+ in¯ux, cas30 expression and freezing tolerance
substantially but not completely (Figure 3). JK concentrations from 0.1±10 mM had similar effects on cas30 expression (data not shown). This raises the possibility that these
cold-induced processes may be at least partially modulated through, as yet unidenti®ed, mechanisms that
operate independently of re-organization in the actin
cytoskeleton. The substantial effects of CD and JK on
Ca2+ in¯ux through opening of Ca2+ channels observed in
this study indicate that such Ca2+ in¯ux requires the reorganization of the cytoskeleton. However, it should be
emphasized that the cytoskeleton may also be a platform
for several other physiological functions involved in cold
acclimatization, such as protein traf®cking and modulation
of activities of protein kinases/phosphatase that may or
may not be dependent on Ca2+ in¯ux.
An intriguing observation of the present study is that CD
treatment at 25°C reduces the constitutive level of freezing
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
Early steps in cold sensing by plant cells 791
tolerance (Figure 5c) although it induces Ca2+ in¯ux and
cas30 expression (Figure 5a,b). While the reasons for this
are presently unclear, it is possible that triggering of CA
requires only a transient disruption of the actin micro®lament network, and completion of the CA processes
depends upon the re-organization of the actin cytoskeleton
in an altered pattern. The required re-assembly of the actin
cytoskeleton might be prevented by the continued presence of CD. The constitutive level of freezing tolerance is
therefore lost by disruption of the pre-existing actin
cytoskeleton coupled with a failure to re-organize. We are
currently investigating the dynamics of actin cytoskeletal
changes (F-actin levels) during cold shock as well as
following treatments with modulators of membrane
¯uidity and actin micro®lament stabilization.
What is the nature of mechanistic relationship between
membrane rigidi®cation, actin micro®lament re-organization, and Ca2+ channel opening? The effects of CD and JK
on Ca2+ in¯ux, observed in the present study, suggest that
micro®laments are directly or indirectly connected to the
channel, keeping it under tension in the closed state. It is
tempting to suggest that cold- or chemically induced reorganization of actin micro®laments can dissipate the
tension forces that keep the channel closed, thereby
allowing the channel to open. This may explain why
treatment with the actin micro®lament stabilizer JK would
maintain the tension forces that keep the channel closed
whereas CD could dissipate them. Our data suggest that
re-organization of actin micro®laments in cold signalling is
located downstream of membrane rigidi®cation, since the
cas30 transcript accumulation induced by cold or by
DMSO at 25°C is inhibited by JK (Figure 6c). Membrane
rigidi®cation-induced re-organization of the actin micro®laments and Ca2+ channel opening would appear to
require that the actin cytoskeleton be coupled to the
plasma membrane as well as to Ca2+ channels. It is known
that cytoskeletal components are attached to the plasma
membrane and keep it under tension (Ding and Pickard,
1993; Schafer and Schroer, 1999). This observation
assumes particular signi®cance because cold-activated
Ca2+ channels in plants have been shown to be stretchactivated (Ding and Pickard, 1993), indicating that tensile
forces are involved in the activation of these channels. In
®broblasts, controlled stretch forces applied to the plasma
membrane cause large Ca2+ transients (Glogauer et al.,
1995). It has also been shown that CD treatment induces
Ca2+ transients through stretch-activated Ca2+ channels
(Glogauer et al., 1995; Howarth et al., 1998; Wu et al., 1999;
Xu et al., 1997). These Ca2+ transients do not occur
following disruption of microtubules (Wu et al., 1999; Xu
et al., 1997). Interestingly, Mazars et al. (1997) demonstrated that destabilization of microtubules with oryzalin
increased Ca2+ in¯ux in cold-shocked tobacco protoplasts.
We did not investigate the role of microtubules in lowã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
temperature signalling, but alfalfa suspension cells treated
with oryzalin at 25°C do accumulate cas30 transcripts
(OÈrvar et al., unpublished data), indicating that re-organization of both micro®laments and microtubules is involved
in low-temperature sensing in these cells. The interaction
between micro®laments and microtubules is well established and therefore it is possible that both these
cytoskeletal components are involved in low-temperature
signal transduction.
To the best of our knowledge, this is the ®rst report
demonstrating the signalling role of the actin cytoskeleton
in regulating plant gene expression. It should be pointed
out that the results presented here were derived from cell
suspension cultures. Whether they can be extended to
intact plants awaits further investigation. It has been
proposed that the cytoskeleton acts as a scaffold where
physical forces are transduced into biochemical signals
(Schmidt and Hall, 1998). How temperature-induced
changes in the physical state of the membrane regulate
actin cytoskeleton organization remains to be determined.
Cold-induced membrane rigidi®cation is believed to occur
in distinct microdomains of the plasma membrane (Murata
and Los, 1997). These microdomains may provide the
loci within which low-temperature-induced membrane
rigidi®cation modulates actin micro®lament organization.
In various mammalian cell types, e.g. epithelial cells,
Triton X-100-insoluble membrane microdomains called
lipid rafts have been identi®ed (Anderson, 1998). These
cholesterol-rich rafts move within the lipid bilayer and may
serve as platforms for both transmembrane signal transduction and cytoskeletal re-arrangement (Anderson, 1998;
Simons and Ikonen, 1997). Whether similar platforms or
microdomains for transmembrane signalling and cytoskeletal remodelling exist in plants, and are important in
cold sensing, awaits further investigation.
Experimental procedures
Plant material and cold acclimatization
Conditions for growth of alfalfa (Medicago sativa ssp. falcata cv
Anik) cell suspension cultures were as previously described
(Monroy and Dhindsa, 1995). In all experiments, 7- or 8-day-old
cultures were used. CA was administered by cooling the cells
from 25 to 4°C (cooling rate 1.2°C min±1) under low light
(20 mmol photons m±2 sec±1) and de-acclimatization by returning
the cells to 25°C at 100 mmol photons m±2 sec±1 illumination.
Administration of chemicals
For modulation of membrane ¯uidity, benzyl alcohol (BA; Carratu
et al., 1996; Kitagawa and Hirata, 1992) and DMSO (Lyman and
Preisler, 1976) were added directly to the cell suspension at the
concentrations indicated. The role of actin micro®lament reorganisation in the cold response was investigated using JK
(Molecular Probes, Eugene, Oregon, USA) or CD (Sigma, Oakville,
792 BjoÈrn LaÂrus OÈrvar et al.
Ontario, Canada). Stock solutions of both chemicals were made in
50% DMSO to ®nal concentrations of 0.7 and 3.94 mM, respectively. The ®nal treatment solutions of these chemicals contained
0.15% and 0.64% DMSO, respectively. To prevent Ca2+ in¯ux, the
Ca2+ chelator 1,2-bis-(2-aminophenoxy)ethane N,N,N¢,N¢-tetraacetic acid (BAPTA; Sigma, Oakville, Ontario, Canada), dissolved
in H2O, was used. For promoting Ca2+ in¯ux at 25°C, the Ca2+
agonist Bay K 8644 (Calbiochem, San Diego, California, USA) and
the Ca2+ ionophore A23187 (ICN, Costa Mesa, California, USA),
both dissolved in DMSO, were used as previously described
(Monroy and Dhindsa, 1995). In each experiment, the cell
suspension cultures were ®rst treated for 1 h with the indicated
chemical at 25°C before incubation at 25°C or 4°C. In each
experiment when relevant, control cultures were treated with the
DMSO solvent concentration alone, as indicated in the ®gure
legends.
Cell viability and freezing tolerance
Freezing tolerance of cell suspension cultures was determined by
the 2,3,5-triphenyl tetrazolium chloride (TTC) reduction method
(Monroy et al., 1993) from two independent experiments and
expressed as a percentage of viability of frozen, untreated control
cells.
RNA gel blot analysis
Total RNA was isolated using TRIzol (Gibco BRL, Burlington,
Ontario, Canada) according to the manufacturer's instructions,
and 10 mg subjected to blot analysis with high-stringency
hybridization and washing as described (Davis et al., 1994). For
cas30 hybridization, the 5¢ end (480 bp BamHI/KpnI fragment) of
cas30 corresponding to the previously reported cDNA clone
pSM2358 (Mohapatra et al., 1989) was used as probe. To
determine whether the chemical treatments investigated in this
study affect cas30 speci®cally, the expression of two control
genes was studied. A 250 bp Bsu36I/MunI cDNA fragment for the
replacement histone H3.2 protein from alfalfa (Robertson et al.,
1996), and a 721 bp cDNA fragment for a putative ribosomal
protein L5 from alfalfa (Asemota et al., 1994), were used as probes
to determine the transcript levels of their respective genes. All
probes were radiolabelled with 32P by random priming (T7
QuickPrime Kit; Amersham Pharmacia, Baie d'UrfeÂ, QueÂbec,
Canada) according to the manufacturer's instructions. Negative
digital images of EtBr-stained blots were used as loading controls.
Transcript levels following each treatment described were
determined in at least three independent experiments. The results
from one representative experiment are shown.
Ca2+ in¯ux measurements
Protoplasts, isolated from cell suspension cultures as described
by Monroy and Dhindsa (1995) were used for Ca2+ in¯ux studies.
The analysis of 45Ca2+ in¯ux by cation exchange chromatography
was as previously described (Monroy and Dhindsa, 1995), except
for the addition of 0.5 mM Ca2+ to stabilize membranes, and the
use of 250±400 protoplasts per ml in each experiment. Protoplasts
(200 ml) were ®rst treated for 30 min at 25°C with the chemical
agents using gentle shaking (60 rev min±1), before incubation
for 20 min (without shaking) with 45Ca2+ at 4°C (cooling rate
4°C min±1) or 25°C. Ca2+ in¯ux was compared to in¯ux in nontreated or DMSO-treated (solvent vehicle alone) protoplasts. Ca2+
in¯ux measurements were determined in at least three inde-
pendent experiments, each with three replicates, using three
different protoplast preparations. The results from one representative experiment are shown.
Membrane ¯uidity measurements
Membrane ¯uidity was determined using 1,6-diphenyl-1,3,5hexatriene (DPH, Sigma, Oakville, Ontario, Canada), prepared in
azetonitrile, as probe. Protoplasts were treated at 25°C for 30 min,
incubated with DPH (6 mM) for 30 min, followed by 20 min
incubation at 4°C or 25°C. Steady-state ¯uorescence polarization
measurements were carried out as described previously
(Raymond and Plaa, 1996), using a Shimadzu RF-540 spectro¯uorophotometer with slit bandwidth of 10 nm and excitation and
emission wavelengths of 365 and 432 nm, respectively. All
experiments were repeated at least three times. The results from
one representative experiment are shown and expressed as
values of the ¯uorescence polarization index (p) that is inversely
related to membrane ¯uidity (Choi and Yu, 1995).
Protoplast viability
The viability of treated protoplasts was determined by vital
staining with Evan's blue (Kanai and Edwards, 1973). Brie¯y,
protoplasts where treated with BA (5 or 20 mM), DMSO (3%), JK
(2 mM), CD (50 mM) or solvent alone, for 30 min at 25°C, with gentle
shaking (60 rev min±1), before incubation at 4°C or 25°C. The
number of protoplasts were counted in 16 squares in a haemocytometer and the average viability 6 SD of eight replicates
calculated.
Acknowledgements
We thank Drs B. Tuchweber and M. Audet for their valuable help
with the membrane ¯uidity measurements, Dr R. Esnault for the
L5 cDNA, Drs T. Kapros and J.H. Waterborg for the H3.2 cDNA,
and Drs T. Bureau, B. Suter, Ronald Poole and U. Thorsteinsdottir
for the critical reading of the manuscript. This work was
supported by Natural Sciences and Engineering Research Council
of Canada.
References
Alonso, A., Queiroz, C.S. and Magalhaes, A.C. (1997) Chilling
stress leads to increased membrane rigidity in roots of coffee
(Coffea arabica L.) seedlings. Biochim. Biophys. Acta, 1323,
75±84.
Anderson, R.G.W. (1998) The caveolae membrane system. Annu.
Rev. Biochem. 67, 199±225.
Asemota, O., Breda, C., Sallaud, C., el Turk, J., de Kozak, I.,
Buffard, D., Esnault, R. and Kondorosi, A. (1994) Cloning and
expression of a cDNA encoding a cytoplasmic L5 ribosomal
protein from alfalfa (Medicago sativa L.). Plant Mol. Biol. 26,
1201±1205.
Bubb, M.R., Senderowicz, A.M., Sausville, E.A., Duncan, K.L. and
Korn, E.D. (1994) Jasplakinolide, a cytotoxic natural product,
induces actin polymerization and competitively inhibits the
binding of phalloidin to F-actin. J. Biol. Chem. 269, 14869±
14871.
Callaini, G., Dallai, R. and Riparbelli, M.G. (1991) Micro®lament
distribution in cold-treated Drosophila embryos. Exp. Cell Res.
194, 316±321.
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
Early steps in cold sensing by plant cells 793
Carratu, L., Franceschelli, S., Pardini, C.J., Kobayashi, G.S.,
Horvath, I., Vigh, L. and Maresca, B. (1996) Membrane lipid
perturbation modi®es the set point of the temperature of heat
shock response in yeast. Proc. Natl Acad. Sci. USA, 93, 3870±
3875.
Choi, J.H. and Yu, B.P. (1995) Brain synaptosomal aging: free
radicals and membrane ¯uidity. Free Rad. Biol. Med. 18, 133±
139.
Cooper, J.A. (1987) Effects of cytochalasin and phalloidin on actin.
J. Cell Biol. 105, 1473±1478.
Danyluk, J., Carpentier, E. and Sarhan, F. (1996) Identi®cation and
characterization of a low temperature regulated gene encoding
an actin-binding protein from wheat. FEBS Lett. 389, 324±327.
Davis, L., Kuehl, M. and Battey, J. (1994) Preparation and analysis
of RNA from eucaryotic cells. In Basic Methods in Molecular
Biology. Norwalk, Connecticut: Appleton and Lange, pp. 350±
355.
Ding, J.P. and Pickard, B.G. (1993) Modulation of
mechanosensitive calcium-selective cation channels by
temperature. Plant J. 3, 713±720.
Gleason, M.M., Medow, M.S. and Tulenko, T.N. (1991) Excess
membrane cholesterol alters calcium movements, cytosolic
calcium levels, and membrane ¯uidity in arterial smooth
muscle cells. Circ. Res. 69, 216±227.
Glogauer, M., Ferrier, J. and McCulloch, C.A. (1995) Magnetic
®elds applied to collagen-coated ferric oxide beads induce
stretch-activated Ca2+ ¯ux in ®broblasts. Am. J. Physiol. 269,
1093±1104.
Goddette, D.W. and Frieden, C. (1986) Actin polymerization. The
mechanism of action of cytochalasin D. J. Biol. Chem. 261,
15974±15980.
Howarth, F.C., Boyett, M.R. and White, E. (1998) Rapid effects of
cytochalasin-D on contraction and intracellular calcium in
single rat ventricular myocytes. P¯ugers Arch. 436, 804±806.
Hwang, J.-U., Suh, S., Hanju, Y., Kim, J. and Lee, Y. (1997) Actin
®laments modulate both stomatal opening and inward K+channel activities in guard cells of Vicia faba L. Plant Physiol.
115, 335±342.
Kanai, R. and Edwards, G.E. (1973) Puri®cation of enzymatically
isolated mesophyll protoplasts from C3, C4, and Crassulacean
acid metabolism plants using an aqueous dextran±
polyethylene glycol two-phase system. Plant Physiol. 52, 484.
Kitagawa, S. and Hirata, H. (1992) Effects of alcohols on
¯uorescence anisotropies of diphenylhexatriene and its
derivatives in bovine blood platelets: relationships of the
depth-dependent change in membrane ¯uidity by alcohols
with their effects on platelet aggregation and adenylate cyclase
activity. Biochim. Biophys. Acta, 1112, 14.
Knight, H., Trewavas, A.J. and Knight, M.R. (1996) Cold calcium
signaling in Arabidopsis involves two cellular pools and a
change in calcium signature after acclimation. Plant Cell, 8,
489±503.
Knight, M.R., Campbell, A.K., Smith, S.M. and Trewavas, A.J.
(1991) Transgenic plant aequorin reports the effects of touch
and cold-shock and elicitors on cytoplasmic calcium. Nature,
352, 524±526.
Levitt, J. (1980) Freezing resistance ± types, measurement and
changes. In Responses of Plants to Environmental Stress, Vol.
1: Chilling, Freezing, and High Temperature Stress. New York:
Academic Press, pp. 137±141.
Lyman, G.H. and Preisler, H.D. (1976) Membrane action of DMSO
and other chemical inducers of Friend leukaemic cell
differentiation. Nature, 262, 360±363.
McCollum, D., Balasubramanian, M. and Gould, K. (1999)
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794
Identi®cation
of
cold-sensitive
mutations
in
the
Schizosaccharomyces pombe actin locus. FEBS Lett. 451, 321±
326.
Mathur, J., Spielhofer, P., Kost, B. and Chua, N.-H. (1999) The
actin cytoskeleton is required to elaborate and maintain spatial
patterning during trichome cell morphogenesis in Arabidopsis
thaliana. Development, 126, 5559±5568.
Mazars, C., Thion, L., Thuleau, P., Graziana, A., Knight, M.R.,
Moreau, M. and Ranjeva, R. (1997) Organization of cytoskeleton
controls changes in cytosolic calcium of cold-shocked
Nicotiana plumbaginifolia protoplasts. Cell Calcium, 22, 413±
420.
Mejia, R., Gomez-Eichelmann, M.C. and Fernandez, M.S. (1995)
Membrane ¯uidity of Escherichia coli during heat-shock.
Biochim. Biophys. Acta, 1239, 195±200.
Mohapatra, S.S., Poole, R.J. and Dhindsa, R.J. (1989) Molecular
cloning and relationship to freezing tolerance of coldacclimation-speci®c genes of alfalfa. Plant Physiol. 84, 1172±
1176.
Monroy, A.F. and Dhindsa, R.S. (1995) Low-temperature signal
transduction: induction of cold acclimation-speci®c genes of
alfalfa by calcium at 25°C. Plant Cell, 7, 321±331.
Monroy, A.F., Farhan, S. and Dhindsa, R.S. (1993) Cold-induced
changes in freezing tolerance, protein phosphorylation, and
gene expression. Plant Physiol. 102, 1227±1235.
Morelli, J.K., Zhou, W., Yu, J., Lu, C. and Vayda, M.E. (1998) Actin
depolymerization affects stress-induced translational activity of
potato tuber tissue. Plant Physiol. 116, 1227±1237.
Mrak, R.E. (1992) Opposite effects of dimethyl sulfoxide and
ethanol on synaptic membrane ¯uidity. Alcohol, 9, 513±517.
Murata, N. and Los, D.A. (1997) Membrane ¯uidity and
temperature perception. Plant Physiol. 115, 875±879.
Nygren, P., Larsson, R., Rastad, J., Akerstrom, G. and Gylfe, E.
(1987) Dimethyl sulfoxide increases cytoplasmic Ca2+
concentration and inhibits parathyroid hormone release in
normal bovine and pathological human parathyroid cells.
Biochim. Biophys. Acta, 928, 194±198.
Raymond, P. and Plaa, G.L. (1996) Ketone potentiation of
haloalkane-induced hepatotoxicity: CCl4 and ketone treatment
on hepatic membrane integrity. J. Toxicol. Environ. Health, 49,
285±300.
Robertson, A.J., Kapros, T., Dudits, D. and Waterborg, J.H. (1996)
Identi®cation of three highly expressed replacement histone H3
genes of alfalfa. DNA Seq. 6, 137±146.
Sampath, P. and Pollard, T.P. (1991) Effects of cytochalasin,
phalloidin, and pH on the elongation of actin ®laments.
Biochemistry, 30, 1873±1980.
Sawitzky, H., Liebe, S., Willingale-Theune, J. and Menzel, D.
(1999) The anti-proliferative agent jasplakinolide rearranges
the actin cytoskeleton of plant cells. Eur. J. Cell Biol. 78,
424±433.
Schafer, D.A. and Schroer, T.A. (1999) Actin-related proteins.
Annu. Rev. Cell Dev. Biol. 15, 341±363.
Schmidt, A. and Hall, M.N. (1998) Signaling to the actin
cytoskeleton. Annu. Rev. Cell Dev. Biol. 14, 305±338.
Sentenac, H., Bonneaud, N., Minet, M., Lacroute, F., Salmon,
J.-M., Gaymard, F. and Grignon, C. (1992) Cloning and
expression in yeast of a plant potassium ion transport
system. Science, 256, 663±665.
Simons, K. and Ikonen, E. (1997) Functional rafts in cell
membranes. Nature, 387, 569±572.
Spector, I., Shochet, N.R., Kashman, Y. and Groweiss, A. (1983)
Latrunculins: novel marine toxins that disrupt micro®lament
organization in cultured cells. Science, 219, 493±495.
794 BjoÈrn LaÂrus OÈrvar et al.
Spector, I., Braet, F., Shochet, N.R. and Bubb, M.R. (1999) New
anti-actin drugs in the study of the organization and function of
the actin cytoskeleton. Microsc. Res. Tech. 47, 18±37.
Tahtiharju, S., Sangwan, V., Monroy, A.F., Dhindsa, R.S. and
Borg, M. (1997) The induction of kin genes in cold-acclimating
Arabidopsis thaliana. Evidence of a role for calcium. Planta,
203, 442±447.
Vigh, L., Los, D.L., Horvath, I. and Murata, N. (1993) The primary
signal in the biological perception of temperature: Pd-catalyzed
hydrogenation of membrane lipids stimulated the expression
of the desA gene in Synechocystis PCC6803. Proc. Natl Acad.
Sci. USA, 90, 9090±9094.
Watts, R.G. and Howard, T.H. (1992) Evidence for a gelsolin-rich,
labile F-actin pool in human polymorphonuclear leukocytes.
Cell Motil. Cytoskeleton, 21, 25±37.
Wu, Z., Wong, K., Glogauer, M., Ellen, R.P. and McCulloch, C.A.
(1999) Regulation of stretch-activated intracellular calcium
transients by actin ®laments. Biochem. Biophys. Res.
Commun. 261, 419±425.
Xu, W.X., Kim, S.J., So, I. and Kim, K.W. (1997) Role of actin
micro®lament in osmotic stretch-induced increase of voltageoperated calcium channel current in guinea-pig gastric
myocytes. P¯ugers Arch. 434, 502±504.
Yamamoto, N. (1989) Effect of dimethyl sulfoxide on cytosolic
ionized calcium concentration and cytoskeletal organization of
hepatocytes in a primary culture. Cell Struct. Funct. 14, 75±85.
ã Blackwell Science Ltd, The Plant Journal, (2000), 23, 785±794