Download Membrane trafficking and osmotically induced

Journal of Experimental Botany, Vol. 57, No. 15, pp. 4123–4131, 2006
doi:10.1093/jxb/erl187 Advance Access publication 6 November, 2006
RESEARCH PAPER
Membrane trafficking and osmotically induced volume
changes in guard cells
Joseph C. Shope and Keith A. Mott*
Biology Department, Utah State University, Logan, UT 84322-5305, USA
Received 13 June 2006; Accepted 5 September 2006
Abstract
Guard cells rapidly adjust their plasma membrane
surface area while responding to osmotically induced
volume changes. Previous studies have shown that this
process is associated with membrane internalization
and remobilization. To investigate how guard cells
maintain membrane integrity during rapid volume
changes, the effects of two membrane trafficking
inhibitors on the response of intact guard cells of Vicia
faba to osmotic treatments were studied. Using confocal microscopy and epidermal peels, the relationship
between the area of a medial paradermal guard-cell
section and guard-cell volume was determined. This
allowed estimates of guard-cell volume to be made from
single paradermal confocal images, and therefore
allowed rapid determination of volume as cells
responded to osmotic treatments. Volume changes in
control cells showed exponential kinetics, and it was
possible to calculate an apparent value for guard-cell
hydraulic conductivity from these kinetics. Wortmannin
and cytochalasin D inhibited the rate of volume loss
following a 0–1.5 MPa osmotic treatment. Cytochalasin
D also inhibited volume increases following a change
from 1.5 MPa to 0 MPa, but wortmannin had no effect.
Previous studies showing that treatment with arabinanase inhibits changes in guard-cell volume in response
to osmotic treatments were confirmed. However, pressure volume curves show that the effects of arabinanase and the cytochalasin D were not due to changes
in cell wall elasticity. It is suggested that arabinanase,
cytochalasin D, and wortmannin cause reductions in
the hydraulic conductivity of the plasma membrane,
possibly via gating of aquaporins. A possible role for
aquaporins in co-ordinating volume changes with
membrane trafficking is discussed.
Key words: Aquaporins, guard cells, hydraulic conductivity,
membrane trafficking, stomata.
Introduction
Guard cells change volume in response to a number of
environmental and hormonal factors. These changes in
volume are caused primarily by ion transport at the plasma
membrane and the guard cell therefore controls its volume
via changes in its osmolyte content. The signal transduction
pathways responsible for these responses have been studied intensively (for reviews, see Assmann, 1993, 1999;
MacRobbie, 1998; Blatt, 2000a, b). Guard cells also change
volume in response to external osmotic solutions (Shope
et al., 2003). These responses are rapid and reversible,
which suggests that they do not involve changes in the
osmolyte content of the guard cell, and the change in
volume is controlled by the water potential gradient and the
hydraulic conductivity of the plasma membrane.
As guard cells change volume, in response to either ion
transport or changes in the external osmotic solution, they
must adjust the surface area of their plasma membrane to
maintain cell integrity. Guard-cell volume can increase up
to 50% as stomata open (Raschke and Dickerson, 1973;
Franks et al., 2001; Shope et al., 2003), and plasma membrane area changes by approximately the same percentage
(Shope et al., 2003). Since membrane elasticity is on the
order of 3% (Morris and Homann, 2001), it is unlikely
that membrane stretching could accommodate such large
changes in surface area. Guard-cell protoplasts have been
shown to adjust plasma-membrane surface area via vesicle
fusion and fission as they respond to osmotic treatments
(Homann, 1998; Homann and Thiel, 1999, 2002). More
recently, studies with intact guard cells have shown that
plasma membrane is internalized and remobilized as the
cells shrink and swell (Shope et al., 2003; Meckel et al.,
* To whom correspondence should be addressed. E-mail: [email protected]
ª The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
4124 Shope and Mott
2004), suggesting that similar processes also operate in
intact, turgid guard cells.
It is possible that membrane trafficking is tied to ion
transport processes during changes in guard-cell volume
that are mediated by ion transport. However, during volume
changes in response to external osmotic solutions this
cannot be the case, yet guard cells are remarkably robust to
large volume changes caused by hyper- and hypo-osmotic
treatments (Shope et al., 2003). These treatments produce
changes in volume and surface area of up to 20% in <3 min,
yet membrane integrity is rarely lost. Thus, membrane
trafficking must be able to operate independently from the
ion transport processes.
The processes involved in membrane trafficking to
maintain plasma membrane surface area have not been
well studied in plants (Blatt et al., 1999), and have received
only limited attention in animals. There is evidence from
animal cells (Morris and Homann, 2001) and plant cells
(Battey et al., 1999) that these processes are distinct from
those involved in secretion and endocytosis, and it has been
suggested that membrane trafficking to adjust surface area
is controlled by changes in membrane tension (Morris and
Homann, 2001). Such a mechanism would be consistent
with observations of membrane trafficking during volume
changes caused by external osmotic solutions.
To investigate membrane trafficking during changes in
guard-cell volume caused by external osmotic solutions, the
effects of several compounds known to inhibit membrane
trafficking and/or stomatal opening in the light were
examined. In initial experiments, stomatal opening and
closing were used as an indicator of guard-cell volume
changes. Two membrane-trafficking inhibitors, wortmannin and cytochalasin D, were found to inhibit some
stomatal movements in response to external osmotic solutions, and these were studied in more detail using confocal
microscopy to track guard-cell volume changes and
membrane internalization. Cell volume was determined
by labelling the plasma membrane with a fluorescent dye
(FM4-64) and taking paradermal confocal images at 2 lm
intervals through the guard cells. Each stack of images was
assembled into a three-dimensional figure and volume was
determined by discrete integration as described previously
(Franks et al., 2001). Although this technique produces
accurate estimates of guard-cell volume (Shope et al.,
2003), it is too slow to capture the kinetics of water movement in response to the osmotic treatments because each
stack of images requires c. 1 min to acquire with the confocal microscope. To overcome this problem, the relationship between the medial paradermal area of a guard cell
and its volume was determined. This relationship was
found to be approximately linear for a single guard cell and
remarkably similar for all the guard cells measured over
a wide range of turgor pressures and volumes. This result
allowed medial paradermal sections of guard cells to be
acquired rapidly as they responded to osmotic treatments
and allowed the area of these sections to be used to estimate
guard-cell volume.
Materials and methods
Vicia faba plants were grown in a temperature-controlled greenhouse
in a soil-less medium containing perlite, peat, and vermiculite (1:1:1
by vol.). Pots (1 by wt. 1) were watered daily to excess with
a nutrient solution (Peters 20:10:20.0), and plants were used before
they flowered and before they became pot-bound. Epidermal strips
from the abaxial surface were isolated and floated on buffer (10 mM
MESKOH; 50 mM KCl; pH 6.15) for c. 2 h under a halogen lamp
that was filtered through water to remove the infrared. The air above
the buffer was flushed with CO2-free air at the beginning of the 2 h
incubation to promote stomatal opening. For stomatal aperture
studies, peels were imaged using a digital camera, and apertures were
measured using imaging software.
After a minimum of 2 h floating on buffer under the light, FM4-64
(5 mM in DMSO) was added to the buffer solution to bring the final
concentration to 5 lM. Peels were allowed to incubate for 30–60 min
with FM4-64 under the light to open stomata. Peels were then mounted
(cuticle facing the coverslip) at the bottom of a 100 ll well slide and
visualized with a Bio-Rad MRC 1024 confocal microscope. Osmotic
pressure in the well was changed via a gravity-fed perfusion system.
Medial paradermal images were acquired at time intervals after the
osmotic treatment, and the area of each medial section was used to
calculate the cell volume using the relationship shown in Fig. 2.
To estimate hydraulic conductivity in control experiments, volume
versus time data were fitted to the first-order rate equation shown
below.
Vt ¼ Vf þ ðVi Vf Þekt
ð1Þ
where Vt is the volume at any time t, Vf is the final volume, Vi is the
initial volume, and k is the first-order rate constant. Regression
software was used to fit Vf, Vi, and k. The lines for control treatments
in Figs 3, 4, and 6 show regressions of the data using equation 1.
The first-order rate constant (k) was converted to a hydraulic
conductivity by assuming that the cells were in water potential
equilibrium with the external solution at the beginning of the
experiment (when volume equalled Vi) and at the end of the
experiment (when volume equalled Vf). From the fitted values of Vi
and Vf it was therefore possible to calculate the total volume lost for
a change in 1.5 MPa. For example, in the control experiment shown
in Fig. 3, the fitted values of Vi and Vf are 6673 and 5767 lm3,
respectively, making the total volume change 906 lm3 for a 1.5 MPa
change in water potential (W). The elasticity coefficient for the cell
was determined as:
’ ¼ ðDP=DVÞV
ð2Þ
where V is the average volume during the response. Guard-cell
osmotic pressure (p) was estimated to be 2.5 MPa from the
relationship given for Vicia faba (Franks et al., 2001) and from
observations of plasmolysis at external osmotic pressures around this
value. Finally, hydraulic conductivity of the membrane was estimated
using the equation:
Lp ¼ Vk=Að’ þ pÞ
ð3Þ
where A is the area of the cell membrane as estimated from the
relationship given by Shope et al. (2003).
Guard-cell volume and medial section area
To determine guard-cell volumes, a series of paradermal confocal
images of the guard cells at 2 lm intervals was collected. These
Membrane trafficking in guard cells 4125
image stacks, each containing 15 images, were then assembled into
a 3-D image using image visualization software (Autovisualize; AQI,
Troy, NY, USA). The resulting 3-D images were rotated 90, and the
cross-sectional area of the cell was traced at 3 lm steps as described
previously (Shope et al., 2003). Volume was determined by discrete
integration of these measured sections. Previous studies have shown
that this technique provides reliable estimates of volume (Shope
et al., 2003). Different volumes were produced by using mannitol
solutions in 0.5 MPa increments and cell volume was allowed to
stabilize for 10 min prior to imaging. The area of a medial paradermal
section was determined by rotating the 3-D image appropriately and
finding the section with the largest area.
Confocal microscopy and kinetics of guard-cell volume
changes
To determine hydraulic conductivity of guard cells, epidermal peels
were equilibrated at the bottom of a well slide containing c. 100 ll
buffer for c. 10 min. The buffer solution was then replaced with 1.5
MPa mannitol (in buffer) using a gravity-fed perfusion system and
a flow rate of c. 10 ml min1. Standard washout kinetics show that the
solution in the well reached 99% of 1.5 MPa in c. 3 s. This treatment
was carried out while imaging the medial plane of a pair of guard cells
using the confocal microscope.
For experiments involving pharmacological treatments, the
following incubations were used: arabinanase (endo-arabinanase;
Megazyme International, Wicklow, Ireland), 10 U ml1 for 60 min;
wortmannin (Sigma Chemical, St Louis, MO, USA), 30 lM for 30–
45 min; cytochalasin D (Sigma Chemical), 20 lM for 30–45 min;
HgCl2 (Merck & Co., Rahway, NJ, USA), 50 lM for 5–30 min;
trifluralin (Supelco, Bellefonte, PA, USA), 10 lM for 30 min.
Pressure probe
Epidermal strips from the abaxial surface of a leaf were mounted to
a well slide using VALAP (Vaseline:lanolin:paraffin; 1:1:1 by vol.)
with the cuticle side away from the coverslip. A standard cell pressure
probe filled with silicon oil was inserted into a guard cell, and the
volume manipulated via a piston coupled to a micrometer. Cells were
allowed to equilibrate for 5 min between changes before the pressure
at each volume was recorded (for details, see Franks et al., 2001). At
each volume/pressure value, a digital image was captured and saved
for measurement of the medial paradermal area.
Results
Three of the compounds tested—arabinanase, cytochalasin
D, and trifluralin—were found to inhibit stomatal opening
in light substantially (Fig. 1, a). The remaining compound—wortmannin—had no effect (Fig. 1, a). By
contrast, arabinanase, cytochalasin D, and wortmannin
were found to inhibit stomatal closing in response to a 1.5
MPa mannitol solution, but trifluralin had no effect (Fig. 1,
b). In these experiments, stomata were first opened under
the light for several hours, then treated with the compound
of interest for different periods of time (as noted in the
Materials and methods) before treatment with the osmotic
solution. Apertures were measured 10 min later. To
examine volume increases in response to osmotic treatments, stomata were first opened in the light, then
transferred to a 1.5 MPa mannitol solution to close them,
and then reopened by transferring them back to buffer.
Fig. 1. Changes in stomatal aperture in response to (a) light, (b) a change
from buffer to 1.5 M mannitol, and (c) a change from 1.5 M mannitol to
buffer. Each point represents an average of 8–12 stomata, and the error
bars represent the standard deviation. Each experiment was repeated three
times with similar results. For (a), peels were incubated in the compound
of interest as described in the Materials and methods before they were
placed in light. For (b), stomata were first opened in light for several
hours, then incubated in the compound of interest as described in the
Materials and methods, and then placed in 1.5 mannitol. Measurements
were taken 10 min later. For (c), stomata were first opened in light for
several hours, then incubated in 1.5 M mannitol for 15 min to shrink the
guard cells. The compound of interest was then added as described in the
Materials and methods, and then the peels were returned to buffer.
Measurements were taken 10 min later.
Fifteen minutes were allowed for stomata to close in the 1.5
MPa solution before the compound of interest was added.
Peels were incubated with the compound as described in the
Materials and methods section and then the osmotic
solution was replaced with buffer (c. 0 MPa) and apertures
were measured 10 min later (Fig. 1, c). In these experiments, only cytochalasin D was found to inhibit increases
in guard-cell volume; wortmannin, trifluralin, and arabinanase had no effect. None of the treatments discussed above
appeared to damage the guard cells when viewed under the
light microscope. The effects of wortmannin and cytochalasin D on the kinetics of guard-cell volume changes
were investigated in more detail using confocal microscopy.
Measurement of guard-cell volume using confocal
microscopy
It was impossible to capture images rapidly enough to
produce 3-D images of guard cells as they shrank or swelled
in response to osmotic solutions. To overcome this problem,
guard-cell medial area was used as a proxy for volume. To
determine the relationship between medial area and volume,
guard cells were imaged while in equilibrium mannitol solutions of various concentrations (see Materials and methods).
In total, 10 guard cells were measured. Only two values of
volume and medial area were obtained for five of these cells,
but three or more values were obtained for the other five
cells. The points on Fig. 2 show volume and medial area for
all 10 guard cells. Maximum guard-cell volumes (in buffer
with no mannitol, 0 MPa) were between 6000 and 9000 lm3,
4126 Shope and Mott
and maximum medial areas were between 600 and 900 lm2.
Both volume and medial area declined as external osmotic
pressure was increased, and the relationship between these
two parameters was found to be approximately linear over
the range tested (Fig. 2). The heavy line is a linear regression of all points, and the lighter lines show linear
regressions for the five guard cells for which three or more
data points were obtained; the R2 for each of these lines was
higher than for the overall regression.
returned to near zero, volume increased rapidly and
returned to a value very close to the original. In a few
experiments (e.g. Fig. 3), volume decreased by a small
amount after the initial stable reading was achieved and, in
these cases, volume did not return completely to the
original value when the osmotic concentration was returned
Kinetics of guard-cell responses to osmotic solutions
When guard cells were treated with a 1.5 MPa osmotic
solution, their volume decreased by c. 10% within 10 min
and stabilized at a lower value within a few minutes (Fig.
3). There was clear membrane internalization associated
with cell shrinkage (Fig. 4). This lower volume was stable
for up to 20 min, and when osmotic concentration was
Fig. 2. Relationship between surface area and volume for guard cells of
Vicia faba. Data are from 10 guard cells. The heavy line is a linear
regression of all points (R2¼0.8382). The lighter lines are regressions for
five individual cells for which three or more points were taken. The R2
values for the individual cells were all higher than the value for all cells
together.
Fig. 3. Kinetics of volume loss and gain by a control guard cell exposed
to 1.5 MPa osmotic treatment. Osmotic pressure of the bathing solution
was increased to 1.5 MPa at time zero and then returned to c. 0 MPa at
6.33 min. Points are measured volumes; lines are best fits to equation 1.
Fig. 4. Kinetics of volume loss and images of membrane internalization
for control guard cells and those treated with wortmannin or cytochalasin
D. Osmotic pressure of the bathing solution was increased to 1.5 MPa at
time zero. Points are measured volumes; lines for controls are best fit to
equation 1, and Lp values calculated from these data are given in the text.
Lines for treatments are smoothed lines through the data. Initial volumes
were between 6700 and 7800 lm3.
Membrane trafficking in guard cells 4127
to zero. Kinetic data show that volume decreased
approximately exponentially, and when these data were
regressed using equation 1, R2 values were >0.98. Values
for t1/2 and Lp varied from 0.2 to 1.5 min and from
1.53109 to 4.53109 m s1 MPa1, respectively.
Treatment with wortmannin or cytochalasin D for 30–45
min (with no osmotic treatment) had no visual effect on the
guard cells; the cells remained turgid, there was no loss of
membrane integrity or internalization of membrane, and the
pore aperture remained open (pictures before osmotic
treatment; Fig. 4). However, when these cells were then
treated with 1.5 MPa mannitol they lost volume much more
slowly than control cells (Fig. 4). Volume loss was so slow
that even after 10 min it was difficult to detect changes in
guard-cell shape or pore aperture visually (pictures in Fig.
4). These cells showed no detectable membrane internalization, and there was no sign that plasma membrane
integrity was lost in any of the experiments. Data for cells
treated with wortmannin or cytochalasin D did not fit
equation 1 as well as data for controls (R2 values were lower
than 0.95), so values for Lp were not calculated. The relative
change in volume in 1 min has been used as a means of
comparing these treatments with controls. These data are
shown in Fig. 5.
To examine the effects of wortmannin and cytochalasin
D on osmotically induced volume increases, cells were first
treated with 1.5 MPa mannitol to reduce their volume, and
then treated with wortmannin and cytochalasin D for 30–45
min. As with experiments above, there was no visible effect
of the treatment with wortmannin or cytochalasin D. Cells
were then returned to buffer (0 MPa). In control experiments,
cells increased in volume rapidly following the return to
buffer (Fig. 6), and previously internalized membrane
disappeared, presumably to be reincorporated into the
Fig. 5. Relative change in volume after 1 min for controls and for cells
treated with wortmannin or cytochalasin D. Error bars show 1 SD from
the mean. Closing treatments received 1.5 MPa mannitol at time zero.
Opening treatments received 1.5 MPa mannitol for 10 min before
addition of cytochalasin D or wortmannin for 30–45 min; osmotic
pressure was then returned to 0 MPa at time zero.
plasma membrane (Shope et al., 2003). Cells treated with
wortmannin showed variable results but, in general, their
increase in volume was comparable in size and kinetics to
Fig. 6. Kinetics of water uptake following an increase in the osmotic
pressure of the solution from 1.5 MPa to approximately zero. Pictures
show the effect of a 1.5 MPa osmotic treatment on control, wortmannintreated, and cytochalasin D-treated guard cells. Points are measured
volumes; lines for controls are best fit to equation 1 and Lp values
calculated from these data are given in the text. Lines for treatments are
smoothed lines through the data. Initial volumes were between 5300 and
6700 lm3.
4128 Shope and Mott
that of the controls. However, cells treated with cytochalasin D swelled slowly or not at all. In some experiments, such as the one shown in Fig. 6, the cells actually
lost a small amount of volume in response to the treatment,
and occasionally cells visibly lysed in response to the
osmotic treatment. The relative volume increase in 1 min
for controls and treatments is shown in Fig. 5.
To test whether the inability of guard cells to change
volume in response to these compounds was caused by
changes in wall properties, the cell-pressure probe was used
to determine pressure–volume curves for guard cells
(Franks et al., 2001). Figure 7 shows that there was no
difference in pressure–volume relationships between control cells and those treated with arabinanase or cytochalasin
D. When inflated with the pressure probe, these cells
underwent changes in shape and volume that were
indistinguishable from control cells (Fig. 7). It was not
Fig. 7. Pressure–volume relationships for guard cells treated with
arabinanase and cytochalasin D. The pictures show the effect of pressure
on the size and shape of control and arabinanase-treated guard cells. For
the control, the lower of the two guard cells is impaled by the probe; for
the arabinanase treatment, the upper of the two guard cells is impaled.
The data in the figure show the relationship between pressure and volume
for control cells and cells treated with arabinanase and cytochalasin D.
Data are for one cell per treatment, but partial curves for each treatment
were obtained for two or three other cells before the seal was lost with
the pressure probe. These partial curves were consistent with the data
shown.
possible to obtain pressure–volume data for cells treated
with wortmannin.
Mercury had no apparent effect on the kinetics of guardcell volume changes (Fig. 8). For the experiments shown in
Fig. 8, the cells were incubated in 50 lM HgCl2 for 20 min,
and experiments with 5 min and 30 min incubations
showed similar results.
Discussion
This study confirmed that guard cells shrink and swell
rapidly in response to osmotic solutions, and that plasma
membrane is internalized and remobilized as their surface
area changes. The data are consistent with the assumption
that these changes in volume are the result of osmotic water
loss and are therefore independent of ion transport processes in the guard cells. However, this study shows that
these volume changes can be inhibited by cytochalasin D,
wortmannin, and arabinanase. It is important to note that
these compounds had no apparent effect on guard cells until
the cells were challenged with an osmotic solution, and even
after treatment with osmotic solutions, they did not cause
membrane integrity to fail. Indeed, there were no apparent
cellular consequences of the treatment other than the slow
change in volume and lack of membrane internalization.
Cells did lyse in a few experiments involving volume
increases in the presence of cytochalasin D, but this event
was obvious as massive membrane internalization, and
clearly did not happen in most experiments. It is therefore
unlikely that the effects of these compounds were the result
of general loss of cell integrity. It is possible, but unlikely,
that the effects of these compounds were caused by solute
movement because only massive fluxes of solutes could
have prevented water flow following the osmotic treatments
used in this study. Furthermore, permeabilization of the
Fig. 8. Kinetics of volume loss by guard cells treated with 0.1 M HgCl2 for
20 min before time zero and for controls. All cells were exposed to 1.5 MPa
osmotic treatment at time zero. Points are measured volumes; lines are
best fits to equation 1. Initial volumes were between 6000 and 6200 lm3.
Membrane trafficking in guard cells 4129
membrane to solutes would have resulted in loss of guardcell turgor and stomatal closure.
It is suggested, therefore, that there are only two plausible explanations for the effects of wortmannin, cytochalasin D, and arabinanase on guard-cell volume changes
in response to external osmotic solutions. First, the hydraulic conductivity of the plasma membrane could have been
reduced by the treatment or, secondly, the cell-wall elasticity could have been reduced such that a small reduction in
volume produced a large change in turgor pressure. The
latter explanation has been proposed as the explanation for
the effect of arabinanase on guard cells (Jones et al., 2003).
It would allow the cells to reach hydraulic equilibrium
with the external solution with a much smaller decrease in
volume. However, the present experiments show that
guard cells treated with cytochalasin D or arabinanase can
be easily inflated with a pressure probe (Fig. 7), and no
differences were found in wall properties between control
and treated cells. It is therefore suggested that the simplest
explanation for the effects of wortmannin and cytochalasin
D is that the hydraulic conductivity of the guard-cell
plasma membrane decreased substantially in response to
the treatments.
Two of the compounds used in this study—trifluralin and
wortmannin—have been shown to affect stomatal movements in response to external stimuli in previous studies.
Trifluralin depolymerizes microtubules by preventing the
addition of tubulin dimers, and it has been shown to inhibit
stomatal opening in response to light (Marcus et al., 2001).
Wortmannin is a PI 3-kinase inhibitor that has been shown
to block endocytosis in animal cells (Clague et al., 1995; Li
et al., 1995) and plant cells (Emans et al., 2002). It has been
shown to inhibit closing in response to ABA (Jung et al.,
2002). In previous studies, the effects of trifluralin and
wortmannin on stomatal movements were attributed to
interruptions in signalling cascades between the external
signal (ABA or light) and the activation of ion pumps. The
present data for trifluralin are consistent with this
conclusion—it inhibited stomatal opening in response to
light, but it had no effect on responses to osmotic solutions.
However, the present data show that wortmannin also
inhibits decreases in guard-cell volume in response to
external osmotic solutions, indicating that its effects are not
exclusively on signalling components between the stimulus
and ion transport processes. It is interesting that while
wortmannin inhibited decreases in guard-cell volume, it did
not substantially inhibit light-induced or osmotically
induced increases in guard-cell volume. This result is
consistent with previous studies (Jung et al., 2002) showing
that wortmannin inhibited ABA-induced stomatal closure,
but did not inhibit light-induced opening. Since wortmannin is an endocytosis inhibitor, it seems possible that it may
affect stomatal movements by inhibiting the membrane
internalization necessary to maintain cell integrity as guard
cells lose volume.
The hypothesis that wortmannin affects stomatal movements by inhibiting membrane trafficking is supported by
the results with cytochalasin D and arabinanase, both of
which also inhibited stomatal responses to osmotic
solutions. Cytochalasin D is an inhibitor of actin filament
formation, and it is an inhibitor of endocytosis in animals
and in plants (Samaj et al., 2004). More specifically, cytochalasin D has been shown to affect vesicle fusion with
plasma membrane in guard-cell protoplasts (Bick et al.,
2001). Arabinanase cleaves (1–5)-a-L-arabinans, which
commonly occur as side chains on cell wall polymers. This
has led to speculation that arabinanase causes ‘walllocking’, which prevents guard cells from changing volume
(Jones et al., 2003). However, as noted above, the present
data show that wall properties are not changed by treatment
with arabinanase, making this explanation unlikely. On the
other hand, since the cell wall is connected to the
cytoskeleton via integral proteins in the plasma membrane
(Baluska et al., 2003), it is plausible that cytoskeleton
properties—and therefore membrane trafficking—could be
altered by arabinanase treatments.
Detailed analyses of the kinetics for volume changes in
guard cells are consistent with a reduction in plasma
membrane hydraulic conductivity after treatment with
arabinanase, wortmannin, or cytochalasin D. Volume
changes in control cells were approximately exponential,
suggesting that water transport was first-order and that the
hydraulic conductivity of plasma membrane was constant,
but data for cells treated with wortmannin or cytochalasin D
were less obviously exponential. There are several possible
explanations for this finding such as: (i) volume changes in
these cells were so much slower than control cells that it is
difficult to determine if the data are exponential or not; (ii) it
is possible that the hydraulic conductivity of the membrane
was not constant during the experiment (see below).
The values of Lp calculated for guard cells in the present
study are lower than values that have been reported for
isolated cells such as Chara corallina (Wendler and
Zimmerman, 1982), and they are slightly lower than values
for the epidermis of Elodea densa (Steudle et al., 1982).
This result is not surprising since most of the surface area of
the guard cell is cutinized and/or suberized, and is therefore
not exchanging water with the external medium. In
addition, it should be noted that the values for Lp reported
in this study are subject to error because of unstirred layers
that exist near the plasma membrane. Since the effect of
such unstirred layers is to cause underestimation of the true
hydraulic conductivity (Tyree et al., 2005), the values
presented in this study may be lower than the true Lp.
Finally, it has been shown that high osmotic concentrations
inhibit aquaporin function (Steudle and Tyerman, 1983; Ye
et al., 2004), and this might also have contributed to the low
Lp values found in this study. While these factors may have
caused an underestimation of the absolute values of Lp for
control cells, they cannot explain the effects of wortmannin,
4130 Shope and Mott
cytochalasin D, and arabinanase since control and treated
cells received identical osmotic treatments.
As noted above, the simplest explanation for the present
results with wortmannin and cytochalasin D is that the
hydraulic conductivity of the guard-cell plasma membrane
decreased substantially in response to the treatments. It is
now generally accepted that the hydraulic conductivity of
biological membranes is largely determined by the presence
and activity of aquaporins (Kjellbom et al., 1999; Tyerman
et al., 1999, 2002; Johansson et al., 2000; Chrispeels
et al., 2001; Baiges et al., 2002). Furthermore, large, rapid
changes in the osmotic water permeability of plasma membranes from maize protoplasts have been reported following treatment with osmotic solutions (Moshelion et al.,
2004). These changes in permeability were attributed to
changes in aquaporin activity, and extremely low values of
membrane permeability were reported immediately following an osmotic treatment. Since aquaporins have been
shown to be expressed in guard cells, it is possible that
aquaporin activity could be involved in the response of
guard cells to hyper- and hypo-osmotic treatments. Although aquaporins have been shown to be gated by several
factors, including phosphorylation (Maurel et al., 1995),
osmotic concentrations (Vera-Estrella et al., 2004), and possibly pH (Zeuthen and Klaerke, 1999), as far as is known,
cytochalasin D and wortmannin have not been shown to
have an effect on aquaporins. There are several possible
routes by which wortmannin and cytochalasin D might
affect aquaporin activity indirectly. Wortmannin is an
inhibitor of PI 3-kinase, and will therefore affect a number
of signal transduction cascades that could ultimately affect
aquaporin gating. Similarly, cytochalasin D is an inhibitor
of actin polymerization, and could affect numerous signalling pathways. However, one common effect of these two
inhibitors is their effect on membrane trafficking, and it is
suggested that the effects of wortmannin, cytochalasin D,
and arabinanase were caused by an inhibition in membrane
trafficking. It is further speculated that aquaporins in guard
cells may be gated directly or indirectly by membrane
tension, closing when membrane tension increases beyond
some threshold value, and reopening when membrane tension returns to some acceptable range. Such a system would
help prevent guard-cell volume from changing more rapidly
than membrane trafficking could operate to maintain membrane integrity.
There is evidence that aquaporins can be gated by
membrane tension or mechanical stimuli (Soveral et al.,
1997a, b; Wan et al., 2004). Furthermore, the idea that
aquaporin gating could be influenced by membrane tension
is supported by data showing that changes in turgor
pressure can cause sudden reductions in hydraulic conductivity (Cosgrove and Steudle, 1981), and by recent data
from another study that show rapid, reversible changes in
membrane Lp in response to pressure pulses (Wan et al.,
2004). Small pulses produced no changes in Lp, but larger
pulses produced decreases in Lp by a factor of 4–23. In the
latter study, very large turgor pulses produced nonreversible decreases in Lp. These authors concluded that
the decrease in Lp was not caused by stretch inactivation of
aquaporins, but instead was caused by changes in
aquaporin structure due to the high velocity of water flow
induced by higher pressure pulses. In the present study,
however, aquaporin closure could not have been caused by
high flows because it was found that control cells were able
to support high rates of water movement; it was only in the
presence of membrane-trafficking inhibitors that Lp declined. This suggests that changes in Lp in the present study
were not directly caused by high rates of water flow or
directly by the osmotic treatments used.
The fact that HgCl2 had no apparent effect on the
hydraulic conductivity of guard cells is surprising. However,
not all aquaporins are sensitive to HgCl2 (Daniels et al.,
1994). Additionally, Wan et al. (2004) found that HgCl2 had
no effect on the hydraulic conductivity of cortical cells of
corn roots when the cells had been subjected to a large
pressure pulse. They attributed this to the fact that the
aquaporins were already inhibited by the pressure pulse.
Since the measurements of hydraulic conductivity reported
in the present study involve a large change in turgor and
water potential of the guard cells, it seems possible that some
of the aquaporins in the guard cells were already at least
partially inhibited during the measurements and therefore
did not show inhibition by HgCl2. However, it is difficult to
reconcile this idea with the effects of wortmannin and
cytochalasin on the guard cells.
In summary, it was found that guard cells undergo large
(c. 310) reductions in plasma membrane hydraulic
conductivity in response to several compounds known to
inhibit membrane trafficking. It is suggested that these
changes in apparent Lp may serve to prevent cells from
changing volume faster than they can traffic membrane
to preserve cell integrity. In addition, the present results
also show that the previously reported effect of arabinanase on guard-cell movements is not due to changes
in wall properties, and it is suggested that these effects
are also caused by changes in membrane Lp.
Acknowledgement
We thank Rand Hooper for expert technical assistance.
References
Assmann SM. 1993. Signal transduction in guard cells.
Annual Review of Cellular and Developmental Biology 9,
345–375.
Assmann SM. 1999. The cellular basis of guard cell sensing of
rising CO2. Plant, Cell and Environment 22, 629–637.
Baiges I, Schaffner AR, Affenzeller MJ, Mas A. 2002. Plant
aquaporins. Physiologia Plantarum 115, 175–182.
Membrane trafficking in guard cells 4131
Baluska F, Samaj J, Wojtaszek P, Volkmann D, Menzel D. 2003.
Cytoskeleton–plasma membrane–cell wall continuum in plants:
emerging links revisited. Plant Physiology 133, 482–491.
Battey NH, James NC, Greenland AJ, Brownlee C. 1999.
Exocytosis and endocytosis. The Plant Cell 11, 643–659.
Bick I, Thiel G, Homann U. 2001. Cytochalasin D attenuates the
desensitisation of pressure-stimulated vesicle fusion in guard cell
protoplasts. European Journal of Cell Biology 80, 521–526.
Blatt MR. 2000a. Ca2+ signalling and control of guard-cell volume in
stomatal movements. Current Opinion in Plant Biology 3, 196–204.
Blatt MR. 2000b. Cellular signaling and volume control in stomatal
movements in plants. Annual Review of Cell and Developmental
Biology 16, 221–241.
Blatt MR, Leyman B, Geelen D. 1999. Tansley Review No. 108.
Molecular events of vesicle trafficking and control by SNARE
proteins in plants. New Phytologist 144, 389–418.
Clague JJ, Thorpe C, Jones AT. 1995. Phosphatidylinositol 3kinase regulation of fluid phase endocytosis. FEBS Letters 367,
272–274.
Cosgrove DJ, Steudle E. 1981. Water relations of growing pea
epicotyl segments. Planta 153, 343–350.
Chrispeels MJ, Morillon R, Maurel C, Gerbeau P, Kjellbom P,
Johansson I. 2001. Aquaporins of plants: structure, function,
regulation, and role in plant water relations. In: Hohmann S, Agre P,
Nielsen S, eds. Aquaporins. New York, NY: Academic Press, 277–
344.
Daniels MJ, Mirkov TE, Chrispeels MJ. 1994. The plasmamembrane of Arabidopsis thaliana contains a mercury-insensitive
aquaporin that is a homolog of the tonoplast water channel protein
tip. Plant Physiology 106, 1325–1333.
Emans N, Zimmermann S, Fischer R. 2002. Uptake of a fluorescent
marker in plant cells is sensitive to brefeldin A and wortmannin.
The Plant Cell 14, 71–86.
Franks PJ, Buckley TN, Shope JC, Mott KA. 2001. Guard cell
volume and pressure measured concurrently by confocal microscopy and the cell pressure probe. Plant Physiology 125, 1588–1584.
Homann U. 1998. Fusion and fission of plasma-membrane material
accommodates for osmotically induced changes in the surface area
of guard-cell protoplasts. Planta 206, 329–333.
Homann U, Thiel G. 1999. Unitary exocytic and endocytic events in
guard-cell protoplasts during osmotically driven volume changes.
FEBS Letters 460, 495–499.
Homann U, Thiel G. 2002. The number of K+ channels in the
plasma membrane of guard cell protoplasts changes in parallel with
the surface area. Proceedings of the National Academy of Sciences,
USA 99, 10215–10220.
Johansson I, Karlsson M, Johanson U, Larsson C, Kjellbom P.
2000. The role of aquaporins in cellular and whole plant water
balance. Biochimica et Biophysica Acta–Biomembranes 1465,
324–342.
Jones L, Milne J, Ashford D, McQueen-Mason SJ. 2003. Cell wall
arabinan is essential for guard cell function. Proceedings of the
National Academy of Sciences, USA 100, 11783–11788.
Jung JY, Kim YW, Kwak JM, Hwang JU, Young J, Schroeder JI,
Hwang I, Lee Y. 2002. Phosphatidylinositol 3- and 4-phosphate
are required for normal stomatal movements. The Plant Cell 14,
2399–2412.
Kjellbom P, Larsson C, Johansson I, Karlsson M, Johanson U.
1999. Aquaporins and water homeostasis in plants. Trends in Plant
Science 4, 308–314.
Li G, D’Souza-Schorey C, Barbieri MA, Roberts RL, Klippel A,
Williams LT, Stahl PD. 1995. Evidence for phsophatidylinositol
3-kinase as a regulator of endocytosis via activation of Rab5.
Proceedings of the National Academy of Sciences, USA 92,
10207–10211.
MacRobbie EAC. 1998. Signal transduction and ion channels in
guard cells. Philosphical Transactions of the Royal Society
London, Series B 353, 1475–1488.
Marcus AI, Moore RC, Cyr RJ. 2001. The role of microtubules in
guard cell function. Plant Physiology 125, 387–395.
Maurel C, Kado RT, Guern J, Chrispeels MJ. 1995. Phosphorylation regulates the water channel activity of the seed-specific
aquaporin alpha-TIP. EMBO Journal 14, 3028–3035.
Meckel T, Hurst AC, Thiel G, Homann U. 2004. Endocytosis
against high turgor: intact guard cells of Vicia faba constitutively
endocytose fluorescently labelled plasma membrane and GFPtagged K+-channel KAT1. The Plant Journal 39, 182–193.
Morris CE, Homann U. 2001. Cell surface area regulation and
membrane tension. Journal of Membrane Biology 179, 79–102.
Moshelion M, Moran N, Chaumont F. 2004. Dynamic changes in
the osmotic water permeability of protoplast plasma membrane.
Plant Physiology 135, 2301–2317.
Raschke K, Dickerson M. 1973. Changes in shape and volume of
guard cells during stomatal movement. Plant Research 1972, 149–153.
Samaj J, Baluska F, Voight B, Schlicht M, Volkmann D, Menzel D.
2004. Endocytosis, actin cytoskeleton, and signaling. Plant
Physiology 135, 1150–1161.
Shope JC, DeWald DB, Mott KA. 2003. Changes in surface area of
intact guard cells are correlated with membrane internalization.
Plant Physiology 133, 1314–1321.
Soveral G, Macey RI, Moura TF. 1997a. Membrane stress causes
inhibition of water channels in brush border membrane vesicles
from kidney proximal tubule. Biology of the Cell 89, 275–282.
Soveral G, Macey RI, Moura TF. 1997b. Water permeability of
brush border membrane vesicles from kidney proximal tubule.
Journal of Membrane Biology 158, 219–228.
Steudle E, Tyerman SD. 1983. Determination of permeability
coefficients, reflection coefficients, and hydraulic conductivity of
Chara corallina using the pressure probe: effects of solute
concentration. Journal of Membrane Biology 25, 85–96.
Steudle E, Zimmerman U, Zillikens J. 1982. Effect of cell turgor
on hydraulic conductivity and elastic modulus of Elodea leaf cells.
Planta 154, 371–380.
Tyerman SD, Bohnert HJ, Maurel C, Steudle E, Smith JAC.
1999. Plant aquaporins: their molecular biology, biophysics and
significance for plant water relations. Journal of Experimental
Botany 50, 1055–1071.
Tyerman SD, Niemietz CM, Bramley H. 2002. Plant aquaporins:
multifunctional water and solute channels with expanding roles.
Plant, Cell and Environment 25, 173–194.
Tyree MT, Koh S, Sands P. 2005. The determination of membrane
transport parameters with the cell pressure probe: theory suggests
that unstirred layers have significant impact. Plant, Cell and
Environment 28, 1475–1486.
Vera-Estrella R, Barkla BJ, Bohnert HJ, Pantoja O. 2004. Novel
regulation of aquaporins during osmotic stress. Plant Physiology
135, 2318–2329.
Wan XC, Steudle E, Hartung W. 2004. Gating of water channels
(aquaporins) in cortical cells of young corn roots by mechanical
stimuli (pressure pulses): effects of ABA and HgCl2. Journal of
Experimental Botany 55, 411–422.
Wendler S, Zimmerman U. 1982. A new method for the
determination of hydraulic conductivity and cell volume of plant
cells by pressure clamp. Plant Physiology 69, 998–1003.
Ye Q, Wiera B, Steudle E. 2004. A cohesion/tension mechanism
explains the gating of water channels (aquaporins) in Chara internodes
by high concentration. Journal of Experimental Botany 55, 449–461.
Zeuthen T, Klaerke DA. 1999. Transport of water and glycerol
in aquaporin 3 is gated by H+. Journal of Biological Chemistry
274, 21631–21636.