Download Evidence for an Extracellular Reception Site for Abscisic

Plant Physiol. (1994) 104: 1177-1 183
Evidence for an Extracellular Reception Site for
Abscisic Acid in Commelina Cuard Cells'
Britt E. Anderson, John M. Ward*, and Julian 1. Schroeder
Department of Biology and Center for Molecular Genetics, University of California, San Diego,
La Jolla, California 92093-01 16
The phytohormone abscisic acid (ABA) triggers stomatal closing
as a physiological response to drought stress. Severa1 basic questions limit an understandingof the mechanism of ABA reception in
guard c e k Whether primary ABA receptors are located on the
extracellular side of the plasma membrane, within the intracellular
space of guard cells, or both remains unknown. Furthermore, it is
not clear whether ABA must be transported into guard cells to
exert control over stomatal movements. In the present study, a
combination of microinjedion into guard cells and physi0logic-l
assays of stomatal movements have been performed to determine
primary sites of ABA reception in guard cells. Microinjedion of
ABA into guard cells of Commelina communis 1. resulted in injected cytosolic concentrations of 50 to 200 PM ABA and in additional experiments in lower concentrations Of approximately 1 PM
ABA. Stomata with ABA-loaded guard cells (n > 180) showed
opening similar to stomata with uninjected guard cek. The viability
of guard cells following ABA injedion was demonstrated by neutra1
red staining as well as monitoring of stomatal opening. Extracellular
application of 10 p~ ABA inhibited stomatal opening by 98% at
pH 6.15 and by 57% at pH 8.0. The pH dependenceof extracellular
ABA action may suggest a contribution of an intracellular ABA
receptor to stomatal regulation. The findings presented here show
that intracellular ABA alone does not suffice to inhibit stomatal
opening under the imposed conditions. Furthermore, these data
provide evidence that a reception site for ABA-mediated inhibition
of stomatal opening i s on the extracehlar Side of the Plasma
membrane of guard cells.
bounded by pairs of guard cells. Stomatal aperture is controlled by guard cell turgor, which is mediated by ionic fluxes
across p a r d cell plasma membranes.
Although the physiology of ABA action has been studied
in detail, the molecular mechanisms of ABA reception and
initial events in ABA-induced signal transduction remain
unknown. In guard cells, severa1key ion channels responsible
for ABA-induced 'tomata' closure have been characterized
(Schroeder et a1.j 1987; Schroeder and Ha@wara, 1989; for
reviewsj see Schroeder and He&chr 1989; Keams and
mann, 1993). Whether ABA or other plant hormones exert
their primary actions via activation of extracellular receptors
or by binding to intracellular receptors or both are fundamental unanswered questions in plant biology.
Evidence for the plant hormone auxin has shown that
extracellular binding sites exist that may contribute to auxin
action (Venis et al., 1990; Barbier-Brygooet al., 1991; Marten
et al., 1991; Jones and Herman, 1993). Results conceming
the primary site Of ABA reception~
have so far been
equivocal. As Yet unconfirmed data from exPeriments using
immunohistochemical staining or photoaffinity labeling have
suggested that ABA binds to proteins in the plasma membrane of guard cells (Homberg and Weiler, 1984; Curvetto
and Delmastro, 1986). However, evidente exists that ABA
uptake may be required for stomatal regulation. Cellular ABA
accumulation is known to depend on extracellular pH (Kaiser
and Hartung, 1981). Studies in which the extracellular pH
was varied showed that ABA was less effective at alkaline
pH (Ogunkami et al., 1973), indicating that ABA uptake may
be necessary for the regulation of stomatal aperture. A different study showed a lower threshold concentration for ABA
at an acidic pH of 5.5 but a more rapid ABA response at a
neutra1 pH of 6.8 (Paterson et al., 1988). In addition, other
studies show that alkaline pH does not reduce ABA effects
(Hartung, 1983).
Extracellular application of ABA to guard cells produces
early and rapid increases in cytosolic Ca2+(McAinsh et al.,
1990; Schroeder and Ha@wara, 1990), together with rapid
activation of nonselective ion channels, which provide a
rnechanism for ca2+ influx and, even in the absence of
extracellular Ca2+,for depolarization (Schroeder and Hagiwara, 1990). Concomitantly, ABA triggers a rapid transient
of K+ efflux (MacRobbie, 19901, which m p i t F S depolarization. In Vite Of the rapid nature Of these early responses~
which occur within seconds, these data have not excluded
the possibility that ABA uptake into guard cells may be
required for their activation.
howeverg
ABA is a vital plant hormone, which, among other functions, is known to induce dormancy in seeds and buds, inhibit
root elongation, and reduce the effects of drought stress on
plants (Walton, 1980). Studies of Arabidopsis show that two
genetic loci each affect the control of processes such as seed
gennination, seedling growth, gas exchange regulation, and
protein synthesis by ABA, indicating common initial signaling
mechanisms (Koomneef et al., 1984; Finkelstein and Somerd l e , 1990). During drought conditions, ABA acts on stomatal
pores in the leaf epidermis to cause closure or to inhibit the ,
opening of stomata. Stomatal pores, which regulate transpirational H20 loss and COz uptake from the atmosphere, are
' This research was supported by grant MCB-9004977 from the
~ t science
i ~ Foundation,
~ ~ l a ~ ~ t science
i
~ Foundation~
~
l
Research Expenence for Undergraduates supplement, U.S. Department of Agriculture grant 92-37304-7757 u.1.s.1,and a Nationa1
Science Foundation Plant Biology postdoctoral fellowship U.M.W.).
* Corresponding author; fax 1-619-534-7108.
~
1177
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.
Anderson et ai.
1178
Stomatal pores provide an opportune system to study ABA
reception and its initial signal transduction mechanisms for
several reasons: (a) the response of single guard cells can be
readily monitored by measuring stomatal aperture, (b) the
delivery of various physiological stimuli can be made to guard
cells within epidermal strips, and (c) several molecular mechanisms that control ABA-induced stomatal movements have
been identified. Furthermore, microinjections of dyes and
second messengers into guard cells have recently shown that
the cytosol of guard cells can be accessed using microelectrodes (Blatt et al., 1990; Gilroy et al., 1990, 1991; McAinsh
et al., 1990, 1992). In the present study we have used
iontophoretic microinjection to load ABA into the cytosol of
guard cells. In conjunction with monitoring the aperture of
stomatal pores in leaf epidermal strips, a potent cell biological
system is provided to gain initial insight into the question of
whether extracellular ABA reception is required for control
of stomatal movements.
MATERIALS A N D METHODS
Plant Material
Commelina communis L. plants were grown in a controlled
environment growth chamber (Conviron E15) at 25OC with
a 16-h light/8-h dark daylnight cycle, under fluorescent and
incandescent illumination at a photon fluence rate of 100 pE
m-2 s-'. Plants were grown in potting soil (Uni-Gro; L&L
Nursery Supply Inc., Chino, CA) and watered with deionized
water. Epidermal strips were isolated from the third fully
expanded leaf of 4- to 6-week-old plants.
Epidermal Strip Preparation and lncubation
Variability in the extent of stomatal responses to physiological stimuli has been reported (Gorton et al., 1989). To
minimize experimental variation, pairs of epidermal strips
from the same leaf were used for control experiments and
treatments. Furthermore, injected and uninjected guard cells
(approximately 10 injected and >20 uninjected guard cells
per epidermal strip) were mapped to allow repeated aperture
measurements of individual stomatal pores over time (Gorton
et al., 1989). Each leaf was halved vertically, and two identical
strips of abaxial epidennis were gently isolated under water
to minimize epidermal cell damage and desiccation. The
epidermal strips were anchored on two sides with highvacuum grease (Dow Coming, Midland, MI) in perfusion
chambers prior to microinjection. The epidermal strips were
incubated in 50 or 100 m KCl with either 10 mM Mes-Tris
(pH 6.15) or 10 m Hepes-KOH (pH 8.0) containing either O
or 10 p~ ABA ([a]cis-trans isomer, Sigma).In a11 experiments
stomata were initially closed. To promote stomatal opening,
the epidermal strips were incubated under white light with a
photon fluence rate of 500 pE m-2 s-'. In a11 experiments
epidermal strips from the same leaves were used for control
treatments. Data from experiments in which stomata of control treatments did not open in response to light were omitted
from further analysis. During the incubation period (up to
135 min), the chambers were perfused with COz-free air
(21% 0 2 , 79% N2; Parsons Medical, San Diego, CA), and the
temperature was controlled at 27OC. Measurements of sto-
Plant Physiol. Vol. 104, 1994
mata1 aperture were made using an ocular microrneter and
high-resolution, differential interference microscopy at the
times indicated in the figure legends.
Microinjection
Microelectrodes were pulled from filamented glass (Sutter
Instrumerits Co., San Rafael, CA, catalog No. QF 100-70-10)
to a tip diameter of approximately 250 nm. The electrode tip
was filled with 3 p L of a filtered solution of 10 mM ABA (free
acid), 20 I ~ KOH
M
(from a stock of 100 mM ABA dissolved
in 200 mM KOH), and 10 mM Mes. The final pH of the
microinjected solution was adjusted to 7.2 by alidition of
approximately 6 mM HCl. The osmolality was adjusted to
525 mmol kg-' by addition of D-mannitol. For experiments
in which the final injected cytosolic ABA concenhation was
approximately 1 p ~ the
,
electrode tip was filled with a
solution containing 100 PM ABA, 100 mM KCl, 10 mM Mes
(pH 7.5) (suggested by A. Schwartz and S.M. Assrriann). The
electrode was backfilled with 1 M KCl. Electrode maneuvering
and guard cell impalement were accomplished with a micromanipularor (Narishige, Tokyo, Japan) and an inverted
microscope (Diaphot; Nikon, Tokyo, Japan)under >:400 magnification using high-resolution, differential interferencecontrast o'ptics. Impalement of guard cells became apparent
both microscopically and by the simultaneous appearance of
hyperpolairized membrane potentials. One or two guard cells
per closed stoma were iontophoretically injected (approximately 10 stomata injected per epidermal strip). Iontophoretic
injection 'was accomplished using alternating 5-s pulses of
-200 and O pA of current applied through the electrode for
2 min (Gilroy et al., 1991) using an Axopatch-ID patchclamp amplifier in current clamp mode (Axon Instruments,
Inc., Foster City, CA). To avoid extemal effects of any ABA
possibly leaking from the injection microelectrode tluring the
microinjection procedure, the epidermal strip wa:; continuously perfused with incubation medium using a peristaltic
pump (Rainin; Woburn, MA). Epidermal strips were incubated in neutra1 red to verify the viability of injected guard
cells at the end of a11 experiments (Gilroy et al., 1991).
Carboxyfluorescein Experiments
To estirnate the approximate cytosolic concentration range
of ABA achieved by iontophoretic injection, microelectrodes
were filled with 10 p~ to 20 mM carboxyfluorescein adjusted
to pH 7.2 with KOH and with the osmolality adjusted to 525
mmol k g ' by the addition of mannitol. Guard cells were
iontophoretically injected (as above). Microelectrode and cell
fluorescence were recorded by passing light from a 100-W
mercury lamp through a fluorescein isothiocyanate filter set
(Omega Optical, excitation 480 nm, emission 535 nm). Fluorescence measurements were made using a pinhole diaphragm and a photo multiplier (Thom EM1 Electron Tubes,
Inc., Rockaway, NJ) and recorded and analyzed using an
Axolab interface (Axon Instruments, Inc.) and a 25-MHz 386AT microcomputer (Schroederand Hagiwara, 1990). Fluorescence intensities were measured from 9.6 pm2 of guard cell
cytoplasmic regions exposed through a pinhole ciiaphragm
(the less-Ihorescent areas directly over the vacuole and the
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.
Extracellular Guard Cell ABA Receptor
brightly fluorescent nucleus were avoided). The fluorescence
intensities of known concentrations, between 10 ^M and 20
mM, of carboxyfluorescein were measured in microelectrode
tips as above and were used to estimate cytoplasmic carboxyfluorescein concentration following iontophoretic microinjection. This calibration provided a coarse correction for fluorescence quenching in a volume of similar optical thickness
as guard cells. Based on the similarity of ABA and carboxyfluorescein in charge (both have a pKa of 4.7) and mol wt
(264.3 and 376.3, respectively), the mobility of these two
compounds during iontophoretic microinjection should be
similar (Pusch and Neher, 1988). Furthermore, experiments
were performed at a constant temperature of 25 °C in all
experiments to reduce variation in diffusion of dyes and
ABA. Therefore, estimates of the efficiency of carboxyfluorescein microinjection were used to determine the concentration range of injected cytosolic ABA concentrations following microinjection. The efficiency of ABA microinjection was
not determined by co-injection of carboxyfluorescein and
ABA in the same guard cells, since high concentrations of
injected carboxyfluorescein resulted in guard cells that did
not remain viable for the time required for our experiments
and showed stomatal closing of the half of the stomatal
apertures flanked by the injected guard cells in all experiments (n > 40).
RESULTS
Iontophoretic Microinjection of Commelina Guard Cells
To study the primary reception site of ABA in guard cells,
it was necessary to apply known concentrations of ABA either
extracellularly or to the guard cell cytosol. Single-cell microinjection was used to access the guard cell cytosol. The efficiency of the iontophoretic microinjection technique was
analyzed using the fluorescent dye carboxyfluorescein. Upon
impalement of guard cells with the microelectrode and injection of negative current, rapid appearance of the dye in the
cytoplasm was observed (Fig. 1). Localization of dye in the
cell periphery and in chloroplasts and the nucleus indicated
cytoplasmic rather than vacuolar injection. The observed
results of more than 40 injections of carboxyfluorescein into
stomatal guard cells indicated that the iontophoretic injection
technique used in this study allowed reproducible delivery
into the cytosol.
Quantitation of carboxyfluorescein microinjection by sin-
1179
Figure 2. Iontophoretic microinjection of ABA does not inhibit
stomatal opening. One guard cell (indicated by arrow) was microinjected using a microelectrode containing ABA. A, Stomata following
microinjection at time 0 min. B, The same stomata after 135 min of
incubation under light. The abaxial epidermal strip was incubated
in 50 HIM KCI, 10 mM Hepes-HCI (pH 8.0). This experiment is
representative of more than 100 guard cells microinjected with ABA
under these conditions. Bars indicate 20 ^m.
gle-cell fluorescence photometry was used to estimate the
efficiency of iontophoretic microinjection of ABA. From these
experiments it was estimated that iontophoretic injection of
ABA resulted in a concentration of 50 to 200 JIM ABA in the
guard cell cytosol. The efficiency of ABA microinjection
determined here correlates closely with microinjection efficiencies reported for guard cells in recent studies (Gilroy et
al., 1990, 1991; McAinsh et al., 1990; Lee et al., 1993). These
estimates give the cytosolic concentration of microinjected
ABA and do not include endogenous ABA in guard cells
(Harris et al., 1988). The injected concentration range of ABA
is severalfold higher than the concentration of ABA found in
Vicia faba guard cells following water stress, which was
within the range of 7 to 13 ^M (Harris et al., 1988).
To test whether guard cells microinjected with ABA would
remain viable for the duration of these experiments, epidermal strips were stained with the viability stain neutral red at
the completion of all experiments. Up to 2.5 h after microinjection of ABA, 80 to 90% of the cells remained viable as
determined by the ability to accumulate neutral red in the
vacuole and sustain visible cytoplasmic streaming. In addition, stomata with microinjected guard cells were competent
to open, which requires guard cell viability, as described
below. Approximately 10% of the injected guard cells showed
immediate cessation of cytoplasmic streaming during microinjection and a visible deflation in turgor, indicating a
vacuolar location of the electrode tip. Data from these cells
were not included in this study.
Microinjection of ABA
The question of whether ABA, applied directly to the guard
cell cytoplasm, could inhibit stomatal opening was addressed
using the iontophoretic microinjection technique. Aperture
measurements of noninjected stomata and of stomata within
epidermal strips from the same leaf externally treated with
ABA were made in parallel and served as controls for all
Figure 1. Carboxyfluorescein was iontophoretically microinjected
experimental trials. Guard cells were microinjected with ABA
into the cytosol of one stomatal guard cell in an epidermal strip
from the abaxial side of a Commelina leaf.
(50-200 JIM, one guard cell per stoma) and exposed to light
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.
Anderson et al.
1180
in the absence of external ABA (n > 180 ABA-injected guard
cells in this study). Figure 2A shows two initially closed
stomata following ABA microinjection and prior to exposure
to light. In Figure 2B, the same two stomata are shown
following incubation under white light for 135 min. In all
experimental trials, the stomata with injected guard cells (n
> 100) opened similarly to the uninjected control stomata.
Stomata in which both surrounding guard cells were microinjected with ABA (n = 10) also opened similarly to uninjected
control stomata (data not shown). These results indicated that
intracellular ABA alone could not inhibit stomatal opening.
In additional control experiments, stomata microinjected with
50 to 200 MM ABA and incubated in the presence of 10 MM
extracellular ABA at pH 6.15 remained closed (n = 20).
To quantify the effect of internal ABA on stomatal opening,
apertures were measured for stomata with guard cells injected
with 50 to 200 MM internal ABA throughout a 135-min
incubation. As shown in Figure 3, stomata with one guard
cell microinjected with ABA consistently had rates and maximum stomatal opening that were similar to uninjected controls. When guard cells were injected with lower cytosolic
ABA concentrations of approximately 1 MM (n = 24 guard
cells), stomatal pores also opened similarly to controls.
It is possible that microinjected ABA requires the presence
of additional CaCl2 in the bath solution for inhibition of
stomatal opening (Schwartz et al., 1988). Stomata of guard
cells injected with 50 to 200 MM ABA, with 0.5 HIM CaCl2
added to the bath solution, also opened similarly to controls
(n = 10). These data further support the observation (Fig. 2)
that internal ABA alone was not sufficient to inhibit stomatal
opening.
Extracellular Application of ABA
The ability of low concentrations of externally applied ABA
to inhibit stomatal opening was tested. To exclude the possibility that externally applied ABA could be taken up by
guard cells, the experiments were conducted at an extracellular pH of 8.0. ABA predominates in the deprotonated form
at pH 8.0 and uptake is eliminated (Kaiser and Hartung,
Unln|«ct*d
ABA InJccMd
50
100
Time (mln)
150
Plant Physiol. Vol. 104, 1994
Figure 4. Extracellular application of ABA inhibits stomatal opening.
Abaxial epidermal strips containing uninjected stomata were incubated as described in the legend to Figure 2 with the addition of
10 MM ABA to the perfusion chamber. A and B show the same two
stomatal complexes before ABA exposure (A) and 135 min after
ABA addition to the bath and exposure to white light (B). This figure
depicts an epidermal strip that showed a maximum response to
external ABA. Although 10 MM external ABA allowed partial stomatal
opening in most experiments at pH 8.0, stomata not exposed to
external ABA opened to a greater extent in all experiments (see text
and Fig. 5). Bars indicate 20 Mm.
1981; Baier and Hartung, 1991). Epidermal strips with initially closed stomata (Fig. 4A) were incubated under light in
the presence of 10 MM external ABA. As shown in Figure 4B,
after 135 min the stomata exposed to 10 MM external ABA
remained closed or did not open as wide as nontreated
controls (n > 400 stomata in 20 experiments). Parallel control
experiments in the absence of external ABA showed wider
stomatal opening than ABA-treated stomata. This confirmed
that ABA in the external solution contributes to inhibition of
stomatal opening.
Taken together, the above experiments (Figs. 2-4) demonstrate a requirement for external ABA in the inhibition of
stomatal opening under the imposed conditions. To further
analyze whether the uptake of externally applied ABA may
also contribute to the inhibition of stomatal opening, the
effect of ABA on stomatal aperture was quantitatively and
directly compared at pH 6.15 and 8.0 using epidermal strips
from the same leaves. Guard cells were exposed to 0 or 10
MM external ABA at pH 6.15 or 8.0, placed under light, and
perfused with CO2-free air for 2 h. External ABA at both pH
6.15 and 8.0 inhibited stomatal opening (Fig. 5), and inhibition was more effective at pH 6.15 in all experiments. Extracellular ABA inhibited stomatal opening without addition of
exogenous Ca2+ to the bath solution in all experiments. These
data suggest that basal free Ca2+ levels in the cell wall space
of guard cells and in the unbuffered bath solutions were
sufficient for ABA-induced inhibition of stomatal opening
under the imposed conditions and/or that Ca2+-independent
events contribute to inhibition of stomatal opening (MacRobbie, 1990). Stomatal opening was inhibited an average of
98% by 10 MM external ABA at pH 6.15 and 57% at pH 8.0.
These results may suggest that, in addition to an extracellular
ABA requirement, the pH-dependent uptake of ABA contributes to the inhibition of stomatal opening and/or that other
processes involved in stomatal regulation are pH dependent.
Further experiments were conducted to determine the ef-
Figure 3. Stomata with guard cells microinjected with 50 to 200 MM
ABA open at a similar rate and to a maximum aperture as uninjected
control stomata. Guard cells were microinjected and abaxial epidermal strips were incubated in 50 ITIM KCI, 10 HIM Mes-Tris (pH
6.15). D, Uninjected guard cells (n = 20); •, stomata with one guard
cell injected with ABA (50-200 MM; n = 10). Data are from one
representative experiment of nine. Error bars denote SE.
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.
Extracellular Guard Cell ABA Receptor
12
7
T
O
2
Time (hr)
Figure 5. Externa1 ABA at pH 6.15 and 8.0 inhibits stomatal opening.
Epidermal strips were incubated at pH 6.1 5 or 8.0 in the presence
and absence of 10 PM external ABA. Apertures of mapped stomata
( n = 40 for each bar) were measured before the addition of ABA
and after 2 h of exposure to extracellular ABA and incubation under
white light. Data are from one representative experiment of five.
Error bars denote SE.
fects of microinjected ABA on the time course of stomatal
opening at an extracellular pH of 8.0. Guard cells were
injected with ABA and incubated for 135 min under white
light at an externa1 pH of 8.0. Stomata with guard cells that
were microinjected with ABA opened at a rate and to an
extent that was similar to uninjected control stomata (Fig. 6).
These data may indicate that the decreased effectiveness of
external ABA in the inhibition of stomatal opening at pH 8.0
(Fig. 5) was not solely due to a decrease in ABA uptake and
a lack of intracellular ABA receptor activation. Although a
possibility of functional intracellular receptors for ABA was
not excluded, the presented data demonstrate a requirement
for extracellular ABA in the inhibition of stomatal opening.
DISCUSSION
Phytohormones such as auxin, GAs, and ABA have important functions in the control of plant growth, development, and responses to the environment. Functional reception sites at the extracellular side of the plasma membrane
have been implicated for the hormones auxin (Venis et al.,
1990; Barbier-Brygoo et al., 1991; Marten et al., 1991; Jones
and Herman, 1993) and GAs (Hooley et al., 1991). However,
the cellular location of ABA receptors has remained equivocal. Stomatal guard cells provide a well-suited system to
study ABA receptor location.
ABA Reception by Guard Cells
Under drought stress conditions, ABA accumulates rapidly
in leaves (for review, see Walton, 1980) and can decrease
transpirational water loss by triggering stomatal closure. The
physiological response of stomatal guard cells to exogenously
applied ABA is well characterized. Studies have shown that
an extracellular concentration of 1 ~ L MABA is sufficient to
1181
induce maximum stomatal closure in Commelina (McAinsh et
al., 1990). Although externally applied ABA triggers stomatal
closure in excised epidermal strips, it has remained unclear if
binding to an extracellular receptor contributes to this response. High-affinity ABA-binding sites have been indicated
in the plasma membrane of V. faba guard cells (Hornberg
and Weiler, 1984; Curvetto and Delmastro, 1986). However,
it is not clear if these binding sites represent ABA uptake
transporters or ABA receptors. Severa1 distinct mechanisms
for ABA reception are therefore possible, including: (a) intracellular ABA receptors requiring ABA uptake into guard cells
and cytosolic ABA accumulation, (b) ABA receptors at the
extracellular side of the plasma membrane that trigger intracellular signaling events, and (c) extracellular ABA receptors
acting in parallel with intracellular receptors, resulting in a
requirement for both extracellular and cytosolic ABA. To gain
insight into which of these ABA reception mechanisms can
be excluded and which mechanisms function in the inhibition
of stomatal opening, we utilized the technique of microinjection of ABA into the cytosol of guard cells and the physiological assay of stomatal aperture measurement.
Extracellular Location of Guard Cell ABA Receptors
As reported here, stomata with one or both guard cells
microinjected cytosolically with ABA consistently opened at
a similar rate and to a similar maximum aperture as uninjected stomata (Figs. 2, 3, and 6). These data indicate that
intracellular ABA alone was not sufficient to inhibit stomatal
opening. Externally applied ABA (10 PM) inhibited stomatal
opening (Figs. 4 and 5). Taken together, these data demonstrate that extracellular ABA was required to inhibit stomatal
opening under the conditions imposed here. However, the
possibility that, following external application, ABA uptake
had an additional role in the inhibition of opening could not
be ruled out by these observations (Fig. 5). The experimental
approach presented here focuses on whether extracellular
ABA is required. Data using co-injection of ABA and carboxyfluorescein have indicated a contribution of intracellular
ABA to stomatal regulation using a complementary approach
o+
O
I
.
50
1O0
150
Time (min)
Figure 6. Microinjected ABA does not inhibit stomatal opening at
an extracellular pH of 8.0. Guard cells were microinjected with
ABA and abaxial epidermal strips were incubated as described in
the legend to Figure 2 . O, Uninjected guard cells (n = 20); m,
stomata with one guard cell injected with ABA (50-200 PM; n =
10). Data are from one representative experiment of seven. Error
bars denote SE.
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.
1182
Anderson et al.
that focuses on a requirement for intracellular ABA (Schwartz
et al., 1993).
In the present study inhibition of stomatal opening by ABA
was analyzed rather than ABA-mediated stomatal closing to
ensure viability of guard cells after microinjection. In calibration experiments, high levels of microinjected carboxyfluorescein inhibited stomatal opening and reduced cell viability
as judged by neutral red staining (n > 40, see "Materials and
Methods").Further experiments would therefore be required
to determine if the presented results also apply to ABAinduced stomatal closing.
Dependence of ABA Effectiveness on Extracellular pH
At an extracellular pH of 8.0, a condition that eliminates
ABA uptake (Kaiser and Hartung, 1981; Baier and Hartung,
1991), externally applied ABA consistently inhibited stomatal
opening (Fig. 5), although to a lesser extent than at pH 6.15.
Furthermore, stomata with guard cells microinjected with
ABA and incubated at pH 8.0 opened at a rate similar to
uninjected control stomata (Fig. 6 ) .These data further support
a requirement for an ABA receptor at the extracellular side
of the plasma membrane that functions in the regulation of
stomatal aperture.
However, the data reported here do not completelyexclude
the presence of an additional intracellular ABA receptor.
Alternate explanations for the decreased inhibition of stomatal opening at pH 8.0 are pH effects on plasma membrane
ion channels and pumps that control stomatal movements or
a direct pH dependence of ABA binding to receptors at the
plasma membrane. Previous studies have suggested that a
decreased effectiveness of extracellular ABA to induce stomatal closure at alkaline pH is correlated with a decreased
rate of ABA binding to high-affinity sites at the plasma
membrane of guard cells (Hornberg and Weiler, 1984).
Studies of the Arabidopis thaliana ABA-insensitive mutants
abil and abi2 have shown that early ABA-induced signaling
events are shared in different physiological responses of
different cell types (Finkelstein and Somerville, 1990). Recent
studies using different techniques to those used here, specifically of a-amylase secretion in barley aleurone cells (Gilroy
and Jones, 1994) and ABA-induced expression of the wheat
Em gene in a transient assay (R. Quatrano, personal communication), also suggest an extracellular reception site for
ABA.
Stability of Microinjeded ABA
The possibilities that microinjected ABA was either rapidly
degraded in guard cells, leaked out of guard cells, or was
microinjected at concentrations too high to be effective in
this study need to be considered. Stomatal opening was
observed in response to ABA injection, which requires guard
cell viability. Additional experiments, in which ABA was
injected at approximately 100-fold lower concentrations (1
PM), also did not show an inhibition of stomatal opening.
Furthermore, there is no evidence for reduced ABA effectiveness on stomatal regulation or other responses at high ABA
concentrations (Walton, 1980). For example, induction of Pglucuronidase activity from ABA-inducible promoter-P-glu-
Plant Physiol. Vol. 104, 1994
curonidase constructs in transformed rice protoplasts was
proportional to ABA concentrations from 1 to 100 PM (Marcotte et al., 1988). The stability of ABA in guard cells is
demonstrated by its accumulation for up to 8 h in V. faba
guard cells to concentrations of 7 to 13 PM following dehydration stress of leaves (Harris et al., 1988; Harris and Outlaw, 1991). Therefore, it is unlikely that ABA was microinjected at concentrations that were too high to inhibit stomatal
opening.
ABA is metabolized slowly to phaseic acid and 4 '-dihydrophaseic acid with a half-time of 6 to 8 h in leaf tfiscs of V.
faba (Mertens et al., 1982). Upon the remova1 of dehydration
stress, the ABA concentration of V. faba guard cells returned
to prestress levels with a half-time of 2 h (Harris and Outlaw,
1991). However, responses of guard cells to ABA have been
reported to occur more rapidly. Recent data show that a brief
(2 min) exposure to extracellular ABA is sufficiení to trigger
the signaling cascade leading to long-term K+ efflux from
guard cells and a resulting decrease in stomatal aperture
(MacRobbie, 1990). Furthermore, the initial rates of stomatal
opening were not measurably affected by ABA microinjection
(Figs. 3 and 6). Therefore, the possibility that rapid degradation of A13A resulted in the complete lack of effecíiveness of
microinjected ABA in this study is unlikely. Since guard cells
remained viable throughout the duration of the experiments
reported here, significant leakage of ABA from guard cells,
within severa1 minutes, was unlikely. In addition, any leakage
of injected ABA may contribute to an inhibition of stomatal
opening that was never observed in this study.
In conclusion, this report provides strong eviderice for the
requirement of an ABA receptor at the extracellular side of
the plasma membrane of guard cells which functions in the
regulatiori of stomatal aperture. Furthermore, data show that
intracellular ABA alone does not suffice to inhibit stomatal
opening under the imposed conditions. The identification of
ABA receptors and components of ABA-induced signaling
pathways using cell biological, molecular biological, and genetic approaches should provide further insight into the
mechanism of ABA reception.
ACKNOWLEDCMENTS
Jennifer Sheaffer is gratefully acknowledged for performing severa1 control experiments. We thank Drs. Karen Schumaker and
Douglas Bush for critica1comments on the manuscript and Dr. Simon
Gilroy for ,adviceconceming neutral red staining.
Received September 7 , 1993; accepted December 14, 19513.
Copyright Clearance Center: 0032-0889/94/104/1177/07.
LITERATURE ClTED
Baier M, Hartung W (1991) Cytoplasmic pH and abscisic acid
transport across guard cell membranes. Z Naturforsch 46 106-110
Barbier-Brygoo H, Ephritikhine G, Kliimbt D, Maurel C, Palme
K, Schell J, Guern J (1991) Perception of the auxin signal at the
plasma membrane of tobacco mesophyll protoplasts. Plant J 1:
83-93
Blatt MR, Thiel G, Trentham DR (1990) Reversible inactivation of
K+ channels of Vicia stomatal guard cells following the photolysis
of caged inositol 1,4,5-triphosphate. Nature 346 766-'768
Curvetto NR, Delmastro SE (1986) Ultrastructural evidence of abscisic acid binding sites in guard cells of Vicia faba. Micrcmc Electron
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.
Extracellular Guard Cell ABA Receptor
Biol Cel 10: 115-123
Finkelstein RR, Somerville CR (1990) Three classes of abscisic acid
(ABA)-insensitivemutants of Arabidopsis define genes that control
overlapping subsets of ABA responses. Plant Physiol 9 4
1172-1 179
Gilroy S, Fricker MD, Read ND, Trewavas AJ (1991) Role of
calcium in signal transduction of Commelina guard cells. Plant Cell
3 333-344
Gilroy S, Jones RL (1994) Perception of gibberellin and abscisic
acid at the externa1 face of the plasma membrane of barley
(Hordeum vulgare) aleurone protoplasts. Plant Physiol 1 0 4
1185-1 192
Gilroy S, Read ND, Trewavas AJ (1990) Elevation of cytoplasmic
calcium by caged calcium or caged inositol triphosphate initiates
stomatal closure. Nature 346: 769-771
Gorton HL, Williams WE, Binns ME (1989) Repeated measurements of aperture for individual stomates. Plant Physiol 8 9
387-390
Harris MJ, Outlaw WH (1991) Rapid adjustment of guard-cell
abscisic acid levels to current leaf-water status. Plant Physiol 9 5
171-173
Harris MJ, Outlaw WH, Mertens R, Weiler EW (1988) Waterstress-induced changes in the abscisic acid content of guard cells
and other cells of Vicia faba L. as determined by enzyme-amplified
immunoassay. Proc Natl Acad Sci USA 8 5 2584-2588
Hartung W (1983) The site of action of abscisic acid at the guard cell
plasmalemma of Valerianellu locustu. Plant Cell Environ 6
427-428
Hooley R, Beale MH, Smith SJ (1991) Gibberellin perception at the
plasma membrane of Avena fatua aleurone protoplasts. Planta 183
274-280
Hornberg C, Weiler EW (1984) High-affinity binding sites for
abscisic acid on the plasmalemma of Vicia faba guard cells. Nature
310 321-324
Jones AM, Herman EM (1993) KDEL-containingauxin-binding protein is secreted to the plasma membrane and cell wall. Plant Physiol
101: 595-606
Kaiser WM, Hartung W (1981) Uptake and release of abscisic acid
by isolated photoautotrophic mesophyll cells, depending on pH
gradients. Plant Physiol68 202-206
Kearns EV, Assmann SM (1993) The guard cell-environment connection. Plant Physiol102: 711-715
Koornneef M, Reuling G, Karssen CM (1984) The isolation of
abscisic acid-insensitive mutants of Arabidopsis thaliana. Physiol
Plant 1 5 473-497
Lee HJ, Tucker EB, Crain RC, Lee Y (1993) Stomatal opening is
induced in epidermal peels of Commelina communis L. by GTP
analogs or pertussis toxin. Plant Physiol102 95-100
1183
MacRobbie EAC (1990) Calcium-dependent and calcium-independent events in the initiation of stomatal closure by abscisic acid.
Proc R SOCLond B 241: 214-219
Marcotte WR, Bayley CC, Quatrano RS (1988) Regulation of a
wheat promoter by abscisic acid in rice protoplasts. Nature 335:
454-457
Marten I, Lohse G, Hedrich R (1991) Plant growth hormones control
voltage-dependent activity of anion channels in plasma membrane
of guard cells. Nature 353: 758-762
McAinsh MR, Brownlee C, Hetherington AM (1990) Abscisic acidinduced elevation of guard cell cytosolic CaZ+precedes stomatal
closure. Nature 343: 186-188
McAinsh MR, Brownlee C, Hetherington AM (1992) Visualizing
changes in cytosolic-free Ca2+during the response of stomatal
guard cells to abscisic acid. Plant Cell4 1113-1122
Mertens R, Stiining M, Weiler EW (1982) Metabolism of tritiated
enantiomers of abscisic acid prepared by immunoaffinity chromatography. Natunvissenschaften 6 9 595-597
Ogunkami AB, Tucker DJ, Mansfield TA (1973) An improved
bioassay for abscisic acid and other antitranspirants. New Phytol
7 2 277-278
Paterson NW, Weyers JDB, A'Brook R (1988) The effect of pH on
stomatal sensitivity to abscisic acid. Plant Cell Environ 11: 83-90
Pusch M, Neher E (1988) Rates of diffusional exchange between
small cells and a measuring patch pipette. Pfluegers Arch 411:
204-21 1
Schroeder JI, Hagiwara S (1989) Cytosolic caldum regulates ion
channels in the plasma membrane of Vicia faba guard cells. Nature
338: 427-430
Schroeder JI, Hagiwara S (1990) Repetitive increases in cytosolic
CaZ+of guard cells by abscisic acid activation of non-selective
Ca2+-permeablechannels. Proc Natl Acad Sci USA 87: 9305-9309
Schroeder JI, Hedrich R (1989) Involvement of ion channels and
active transport in osmoregulation and signaling of higher plant
cells. Trends Biochem Sci 1 4 187-192
Schroeder JI, Raschke K, Neher E (1987) Voltage dependence of K+
channels in guard cell protoplasts. Proc Natl Acad Sci USA 8 4
4108-41 12
Schwartz A, Ilan N, Grantz DA (1988) Ca2+ modified stomatal
responses to light, KCl and CO2. Plant Physiol87: 583-587
Schwartz A, Wu W-H, Tucker E, Assmann S (1993) Intracellular
Receptors for ABA in Stomatal Guard Cells (abstract No. 6.1.5-6).
XVth Intemational Botanical Congress, Yokohama, Japan
Venis MA, Thomas EW, Barbier-Brygoo H, Ephritikhine G (1990)
Impermeant auxin analogs have auxin activity. Planta 182:
313-318
Walton DC (1980) Biochemistry and physiology of abscisic acid.
Annu Rev Plant Physiol31: 453-489
Downloaded from www.plantphysiol.org on July 1, 2014 - Published by www.plant.org
Copyright © 1994 American Society of Plant Biologists. All rights reserved.