Plant Physiol. (1994) 106: 1073-1084
Cytoplasmic Ca2+, K+, CI', and NO,- Activities in the
Liverwort Conocephalum conicum 1. at Rest and during
Action Potentials'
Kazimierz Trebacz', Wilhelm Simonis, and Cerald Schonknecht*
Julius-von-Sachs-lnstitut für Biowissenschaften der Universitat Würzburg, Lehrstuhl Botanik I,
Mittlerer Dallenbergweg 64, 97082 Würzburg, Germany
initiated by an elevation of the free Ca2+concentration in the
cytoplasm due to opening of voltage-dependent Ca2+channels in the plasma membrane and in the tonoplast (Williamson and Ashley, 1982; Kikuyama and Tazawa, 1983). This
causes activation of Ca2+-dependent anion channels, and
massive efflux of chloride depolarizes the plasma membrane
(Gaffey and Mullins, 1958; Lunevsky et al., 1983; Okihara et
al., 1991). The depolarization leads to opening of K+ channels
(outward, delayed rectifiers), and as a consequence the K+
efflux repolarizes the plasma membrane again to the resting
potential (Oda, 1976; Kikuyama et al., 1984), which is established near the equilibrium potential for K+.
There is mostly indirect evidence confirming functioning
of individual elements of that model in excitable higher plants
such as Aldrovanda vesiculosa (Iijima and Sibaoka, 1985),
Dionaea muscipula (Hodick and Sievers, 1988), Mimosa pudica
(Abe, 1981; Samejima and Sibaoka, 1982), and Salix viminalis
(Fromm and Spanswick, 1993). A complete investigation of
the ion fluxes during APs in higher plants is still lacking. It
is not clear whether the scheme proposed for characean cells
can be adopted without changes to terrestrial higher plants.
There seems to be a general agreement as to the participation
of Caz+in the excitation (Iijima and Sibaoka, 1985; Hodick
and Sievers, 1988). A C1- efflux as a consequence of an
increased cytoplasmic free Ca2+ concentration was not recorded in a11 investigations (Minorsky and Spanswick, 1989)
even in characean algae (Kikuyama et al., 1984). In some
plants or in some circumstances Ca2+influx alone may depolarize the membrane during APs, whereas in other cases
Ca2+ mainly plays the role of an anion channel activator
(Beilby, 1984). It remains an open question to what extent
there are also ion fluxes across intemal membranes during
APs.
The exact knowledge of intracellular ion activities is important for the understanding of membrane transport (Sanders, 1990), metabolism, nutrition, and cellular signaling.
Although the intracellular activities for cations (H+, K+, Na',
and Ca2+) are well established (Wyn Jones et al., 1979;
Tyerman, 1992) and time-dependent changes for H+ and
lntracellular Caz+, K+, CI-, and NOs- activities were measured
with ion-selective microeledrodes in the liverwort Conocephalum
conicum 1. at rest, during dark/light changes, and in the couese of
adion potentials triggered by light or electrical stimuli. l h e average
free cytosolic Caz+concentration was 231 f 65 nM. We did not
observe any light-dependent changes of the free cytosolic Caz+
concentration as long as no adion potential was triggered. During
adion potentials, on average a 2-fold increase of the free cytoplasmic Caz+concentration was recorded. lntracellular K+ adivity
was 76 f 10 mM. It did not depend on K+ concentration changes
in the bath solution between 0.1 and 10 mM. The average equilibrium potential for K+ in the standard medium containing 1 mM K+
was -110 mV, which differed significantly from the resting potentia1 of -151 f 2 mV. During adion potentials, either a slight
decrease or no changes in intracellular K+ activity were recorded.
The average CI- adivity was 7.4 f 0.2 mM in the cytoplasm and
43.5 f 7 mM in the vacuole. l h e adivities of NO,- were 0.63 I:
0.05 mM in the cytoplasm and 3.0 f 0.3 mM in the vacuole. For
both anions the vacuolar activity was 5 to 6 times higher than the
cytoplasmic activity. After the light was switched off both the CIand the NOs- adivity showed either no change or a slight increase.
lllumination caused a gradual return to previous values or no
change. During action potentials a slight decrease of intracellular
CI- activity was recorded. It was concluded that in Conocephalum,
as in characean cells, chloride channels are involved in the depolarization phase of the action potentials. We discuss a model for
the ion fluxes during an action potential in Conocephalum.
APs in plants play an important role in intracellular signaling and in the regulation of different physiological processes (Davies, 1987; Dziubifiska et al., 1989; Wildon et al.,
1992; Fromm and Spanswick, 1993). To understand the
function of APs in plant physiology, their ion mechanism
has to be known in detail. Data conceming the sequence of
ion fluxes during APs in plants were obtained primarily on
the highly specialized giant characean cells (Beilby, 1984).
According to the present model, APs in these plants are
~
This work was supported by the Deutsche Forschungsgemeinschaft (SFB 176, TP B6) and an Alexander-von-Humboldt Fellowship
(K.T.).
'Permanent address: Department of Plant Physiology, Mana
Curie Sklodowska University, Akademicka 19, PL-20-033 Lublin,
Poland.
* Corresponding author; fax 49-931-71446.
Abbreviations: AP, action potential; E, membrane potential; E,=,,
equilibrium potential for C1-; Een, potential reflecting the activity
for the respective ion; EK, equilibrium potential for K+; pCa,
negative logarithm of the free Ca'+ concentration; TEA, tetraethylammoniumchloride.
1073
Trebacz et al.
1074
Ca2+can be measured with different techniques (Guem et
al., 1991; Felle, 1993; Graziana et al., 1993), little is known
about the intracellular activities for anions (Cl-, Nos-) (Wyn
Jones et al., 1979; Tyerman, 1992) and their time-dependent
Ichanges (Thaler et al., 1992).
The liverwort Conocephalum conicum L., the organism studied here, can be regarded as an intermediary form between
algae and higher plants. It is a multicellular plant but without
conducting bundles. Conocephalum generates APs singly or
in series consisting of several dozens of excitations, spontaneous or evoked by light, and electrical stimuli, or they can
be generated by wounding (Dziubinska et al., 1983; Trpbacz
and Zawadzki, 1985). These APs are quite long lasting, from
about 30 s to over 1 min, and therefore it is possible to apply
ion-selective microelectrodes to measure changes in ion activities accompanying them. It was demonstrated (Dziubiliska
et al., 1989) that in Conocephalum the rate of respiration
begins to increase several seconds after the stimulation and
transiently reaches twice the resting level. This effect is
observed regardless of whether damaging stimuli (incision of
i i thallus edge) or nondamaging electrical stimuli are applied
to trigger APs. No change in the rate of respiration is observed
when no AP is evoked by the respective stimuli either in
unexcitable plants or when excitability is blocked by TEA
(Dziubiliska et al., 1989). Since it was reported that Ca2+
activates respiration in animal cells (Fein and Tsacopoulos,
1988; Brustovetsky et al., 1993), a change in cytosolic Ca2+
might also be the reason for the increased rate of the respiration caused by APs in Conocephalum (Dziubinska et al.,
1989). The transmembrane potential of Conocephalum cells
at rest contains a large electrogenic component (Trgbacz et
al., 1989), which raises the question about the nature of the
repolarizing ion.
Here we report measurements of intracellular Ca2+, K+,
C1-, and NO3- activities in Conocephalum with ion-selective
microelectrodes at rest, during dark/light changes, and in the
course of APs. On this basis we calculate the gradients and
driving forces for these ions and discuss the ion fluxes during
APs in plants.
MATERIALS AND METHODS
Plant Material
The liverwort Conocephalum conicum L. was grown in a
greenhouse at moderate illumination, in the temperature
range 20 to 26OC at high humidity. Every 2 to 3 weeks young
parts of the thalli were put into pots with freshly prepared
soil. Ten- to 14-d-old thalli were used in the experiments.
Two to 4 h before the experiments were started, the thalli
were rinsed with water to remove soil, blotted on tissue
paper, and mounted horizontally in the experimental chamber in the standard experimental medium containing 1 mM
KCI, 0.1 mM CaC12, 2 mM Tris/Mes, pH 7.0. Two sharpened
silver wires 0. 1 mm in diameter were inserted into the thallus
5 to 10 mm from each other to serve as stimulation electrodes
later in the experiment (2 to 4 h).
Electrodes and General Electrophysiology
Micropipettes were prepared from borosilicate glass capilIaries 1.5 mm in outer diameter (Hilgenberg, Malsfeld, Germany) with filament, or without filament for N03--selective
Plant Physiol. Vol. 106, 1994
microelectrodes. Micropipettes were pulled eitlier with a
L/M-3P-A puller (List Medical, Darmstadt, Germmy) or with
a Getra micropipette puller (Munich, Germany). Tip diameters were in the range of 0.2 pm (Thaler et al., 1992). The
potential electrodes were filled with 3 M KCl for Ca2+-selective-, 1 M NaCl for K+-selective-, and 0.5 M K2S#O4
plus 10
m~ KCl for C1-- and N03--selective microelectrodes. The
capillaries destined to become ion-selective microelectrodes
were silanized according to Chao et al. (1988). The silanized
micropipettes were filled with reference solution from their
blunt ends. The reference solution for Ca2+-selec:tivemicroelectrodes (Ammann et al., 1987a) contained 10 m~ EGTA,
6.28 mM CaCl,, 44.9 mM KOH, 55.1 mM KCl, and Mops, pH
7.4 (pCa 7.0; 100 mM K+ as interfering ion). K+-selective
microelectrodes contained 1000 mM KCI, 10 mM NaCI, 2 m~
Hepes/NaOH, pH 7.2. Cl--sensitive microelectrodes were
filled with 100 mM KC1, 10 mM Tris/H2S04,pH 7.4 (Thaler
et al., 1992), and N03--selective microelectrodes contained
100 mM NaN03, 100 mM KCl (Miller and Zlien, 1991).
Remaining air in the shank was pushed away by means of a
10-cm3syringe.
Two methods of filling the ion-selective cocktail into the
tip were applied (Felle, 1988). A small portion of the cocktail
was placed near the filament from the blunt end by means
of another glass pipette with a broken tip. The drop of cocktail
moved spontaneously along the filament toward the tip
within about 1 to 2 min. In the other method, the cocktail
was sucked up into the tip of the microelectrode by applying
negative pressure from its other end. The cockt#dlfor Ca2+selective microelectrodes (Ammann et al., l987a) was
composed of 5% (w/w) Caz+ ionophore I1 (ETH 129,
N,N,N’,N’-tetra-cyclohexyl-3-oxapentanediamid
e), 94% (w/
w) 2-nitrophenyl-octyl ether (o-NPOE), and 1%(w/w) sodium tetraphenylborate; 86% (w/w) of this solution was
added to 14% (w/w) polyvinylchloride and dissolved in two
volumes of tetrahydrofuran (a11 components frorn Fluka AG,
Buchs, Switzerland). For K+-selective microelectrodes (Ammann et al., 198%) the cocktail contained 5% (w/w) K+
ionophore I (valinomycin), 93% (w/w) 1,2-dime:hyl-3-nitrobenzene, and 2% (w/w) potassium tetralcis(4-chloropheny1)borate (a11 from Fluka). The cocktail for (-I--sensitive
microelectrodes was composed as described by Kondo et al.
(1989). For N03--selective microelectrodes the cocktail
(Miller and Zhen, 1991) contained (w/w): 28% polyvinylchloride (high mo1 wt polymer); 1%methyltriphenyl phosphonium bromide; 65% 2-nitrophenyl octyl ethc.r; 6% methyltridodecylammonium nitrate, dissolved in 3 volumes of
tetrahydrofuran (a11 from Fluka). CI--sensitivc? microelectrodes were preincubated in 100 mM KCl solution for at least
3 h before use.
The Ca2+-selectivemicroelectrodes were calibrated according to Ammann et al. (1987a) and the K+-selectivemicroelectrodes were calibrated with solutions containing 1, 10, 50,
100, or 1000 mM KCI. Ten millimolar NaCl and 2 mM Hepes
was added to each solution and the pH was adjusted to 7.2
with NaOH. Calibration solutions for Cl--sensitive microelectrodes were prepared according to Thaler cat al. (1992)
and for N03--selective microelectrodes according to Miller
and Zhen (1991). The selectivity coefficients wclre obtained
by the fixed interference method (Vaughan-Jones and Aickin,
lon Activities in Conocephalum
1987). The ion-selective microelectrodes were calibrated before and after each experiment. Data obtained by the recalibration were the basis of the calculations.
The electrodes were connected to the inputs of a highimpedance ( 1015 Q ) differential amplifier (FD 223, World
Precision Instruments, New Haven, CT). The amplifier subtracts the voltage of the potential electrode ( E ) from the
voltage of the ion-selective electrode ( E
Eio,). The three
Eion, and E,,, were fed to a pen
traces representing E, E
recorder (model314, Kontron, Munich, Germany) and E and
Eion parallel to a digital storage oscilloscope (HM 205-3,
Hameg, Frankfurt, Germany). The traces presented in the
figures were copied by a scanner from original records. For
the sake of clarity in the figures, only the voltage trace ( E )
and the trace for the ion activity (Ei,,,) are shown. An Ag/
AgCl reference electrode had contact with the plant via a salt
bridge and the flowing medium. The stimulation electrodes
were connected through a switch to a regulated DC current
source.
Illumination was provided by a cold light source equipped
with light-conducting fiber (KL 1500 Schott, Mainz, Germany). The photon fluence rate at the plant surface was kept
at 29 pE m-’ s-’ or 50 PE m-‘ s-’ as measured with a quantum
meter (LI-189, Li-Cor, Lincoln, NE).
+
+
lon-Selective Measurement
After the ion-selective microelectrode was calibrated, the
two microelectrodes, an ion-selective and a potential one,
were inserted perpendicularly into the tissue near to the
midrib. The two microelectrodes were located as close to each
other as possible (less than 0.1 mm) under dissection microscope (Leitz, Wetzlar, Germany) observation. The electrodes
were inserted 50 to 100 pm deep in the thallus, in the second
layer of cells, which are relatively large (100-300 pm). Therefore, the two microelectrodes were located in the same cell
or in neighboring cells. The magnitude of the electric potential
difference as well as the slope and direction of AP records
were good indicators of proper microelectrode insertion into
the cell. The distance between the stimulating electrodes and
the site of microelectrodes (ion-selective and potential) insertion was 7 to 10 mm.
Even after addition of polyvinylchlorideand with the small
tip diameters, the ion-selective microeleckodesused had only
a limited pressure resistance. Therefore, the cell turgor had
to be reduced by the addition of sorbitol (Thaler et al., 1992).
Thirty to 60 min before the microelectrodes were inserted,
250 to 320 mM sorbitol, depending on the actual water
potential of cells, was added to the medium perfused through
the measuring chamber at a rate of approximately50 mL h-’.
Sorbitol in this concentration range influenced neither the
resting potential nor the APs. To exclude the possible artifact
(Felle and Bertl, 1986a) that the turgor pushed back the ionselective cocktail inside the cell, the resistance of ion-selective
microelectrodes was measured before the impalement and
every 10 min during the experiments. Because the resistance
of the ion-selective electrodes (>10 G Q )is severa1 orders of
magnitude larger than the membrane resistance and the
resistance of the potential electrode (both in the MQ range),
the electrical resistance must not change during impalement;
1075
when the cocktail was pushed back the resistance decreased
strongly. Only measurements with stable resistances (ionselective microelectrode) and stable potentials, .which were
followed by a successful recalibration (see “Results”), were
evaluated. Anion chromatography was performed as described by Schroppel-Meierand Kaiser (1988).
RESULTS
Measurements of the Membrane Potential and the
Cytoplasmic lon Activity in the Thallus of Conocephalum
The transmembrane potential in cells of Conocephalum
differs significantly among different thalli, but it is constant
within one thallus. This is an important prerequisite for
measuring ion activities with an ion-selective microelectrode
and a separate potential electrode in a tissue where the
electrodes might be located in different cells. Because the ion
activity is calculated from the difference between the two
electrodes (the potential electrode measuring E and the ionselective one measuring E
Eion), the E sensed by the two
electrodes has to be identical. With a rate of AP transmission
of 10 to 30 cm min-’ (in the standard medium) and a distance
between the two microelectrodes of 0.1 mm, the lag time
between the electrodes is smaller than 100 ms, which is
below the time resolution of the measuring system. So, even
during the rapid potential changes of APs no significant
differences (in E ) for the two measuring electrodes are to be
expected.
In Figure 1, traces from two potential electrodes (both
without ion-selective cocktail) inserted into neighboring cells
and the difference of both signals are shown. Neither during
the relatively large light-dependent potential changes nor
during the electrically triggered AP were there significant
deviations between the potentials measured by the two electrodes. Rapid potential changes (light-dependent or APs) did
not cause deviations between the two electrodes larger than
2 mV in any of our control experiments ( n = 11).This allows
us to cany out measurements of ion activities with an ionselective microelectrode inserted into one cell and a potential
microelectrode not necessarily inserted in the same cell but
into one of the neighboring cells.
+
Measurements of the Free Cytoplasmic Ca2+
Concentration in Cells of Conocephalum
After fabrication, the Ca’+-selective microelectrodes had a
resistance between 40 and 100 GQ. Their response time to a
10-fold change in the free Ca2+concentration was usually 5
to 10 s. The slope of microelectrodes was nearly Nemstian
between pCa 5 and 7 and decreased to less than 20 mV in
most microelectrodes between pCa 7 and 8 (Fig. 2, a and b).
After impalement it took up to 15 min before a stable potentia1 was registered by the Ca’+-selective microelectrode. The
response rate of Ca2+-selectivemicroelectrodes allowed measurement of not only the resting leve1 of the free cytosolic
Ca2+concentration, but also its changes during the relatively
long-lasting APs. However, this took place in only approximately 50% of the experiments. In the remaining cases, after
impalement the resistance of the microelectrodes grew to
over 1000 GR, resulting in a drift of the potential and large
Trcbacz et al.
'I 076
-200
5 min
51
O
I
-
!I
I
-50
-1O0
-150
-200
Figure 1. Parallel potential measurements in two neighboring cells
both impaled with a potential microelectrode filled with 3 M KCI
(top and bottom), and the difference between the two potentials
(middle,note t h e 5-times expanded voltage axis for t h e difference
trace).The white and black bars indicate illumination and darkening
of the plant, respectively. The downward pointing arrow indicates
the moment of electrical stimulation to trigger an AP.
offsets during APs. The reason for this was probably clogging
of the microelectrode tips. These experiments were rejected.
Only measurements with stable potentials and resistances
that were followed by a calibration with slopes as given in
Figure 2b were evaluated. These measurements showed an
average cytoplasmic free Ca2+concentration in Conocephalum
of 231 k 65 nM in six different thalli at an externa1 Ca2+
concentration of 0.1 mM.
Although there were some differences in the free Ca2+
concentration for cbfferent thalli (from 100 up to 310 nM),
the free Ca2+concentration measured in a single thallus was
constant within 0.1 pCa (see Figs. 3 and 4). In none of the
experiments did the free Ca2+concentration exceed the submicromolar range, which indicates that the tip of Ca2+selective microelectrodes was always located in the cytoplasm
rather than in the vacuole. We think that Ca2+-selective
microelectrodes were sometimes inserted into the vacuole (as
were other ion-selective microelectrodes)but that they were
probably excluded from the vacuole (see below) before a
stable potential, and thus a reliable free Ca2+concentration,
could be measured. No significant change in the cytoplasmic
free Ca2+concentration was observed ( n = 12) after light was
tumed off (Figs. 3 and 4). Turning the light on evoked a
transient depolarization (approximately 10 mV) called the
generator potential, which could trigger an AP. In cases
where the generator potential was too small to evoke an AP,
Plant Physiol. Vol. 106, 1994
there was no detectable change in the free Ca2+coixentration
after illumination (n = 7) (Fig. 3).
Figure 4 shows an example of a parallel measiirement of
the free cytoplasmic Ca2+concentration and trans membrane
potential changes. The first two APs were triggered electrically (downward pointing arrows), AP number 1 in light and
number 2 in darkness. The third AP was evokr?d by light
(note the lack of an artifact from electrical stimulation). One
can observe a significant increase in the free cytoplasmic Ca2+
concentration during the APs that was prolonged for several
seconds after the repolarization. The increase of the free
cytoplasmic Ca2+concentration during an AP did not depend
on the stimulus (light or electric pulse) used to trigger the
AP. The average leve1 of the free cytoplasmic Ca2+concentration during 11 APs recorded in six different thalli was 477
f 114 nhi. The greatest value detected was 1458 nivr. It should
be stressed that the increase in the free cytophsmic Ca2+
activity recorded during the AP cannot be an artifact resulting
from the lower time resolution of 'the Ca2+-selectivemicroelectrode in comparison with the potential microel ectrode. At
depolarization cation-selective microelectrodes show an apparent decrease in ion activity (and an increase at hyperpolarization) when potential changes are too fast for them
to follow. The fast transients during rapid depolarization
and repolarization phases of the APs were caused by
such a difference in time resolution of the two types of
microelectrodes.
Measurements of the lntracellular K+ Activity wi th
K'-Selective Microelectrodes
The K+-selective microelectrodes had a lower resistance,
between 15 and 30 GQ, and an almost Nemstian slope (Fig.
2, c and d). They responded to a 10-fold change of the K+
concentration within 3 to 5 s. The response time W(IS probably
limited by the exchange of the solutions.
The average intracellular K+ activity in C. conicum immersed in the standard medium containing 1 rnM K+ was
76 f 10 mM (n = 8). The results could not be divided into
two separate groups. This means that either the tips of K+selective microelectrodes were located in a11 expximents in
the cytoplasm or there was no significant difference in K+
activity between the *cytoplasmand the vacuole. An increase
of K+ concentration in the extemal medium frorn 0.1 to 10
m~ had no significant influence on the intracellular K+ activity (Table I). The plasma membrane was depolaiized by 11
and 14 mV after the transitions from 0.1 to 1 antl from 1 to
10 mM K+ in the medium, respectively. The iequilibrium
potential for K+ came close to the resting potential at 0.1 mM
K+ in the extemal medium, but it differed significantly by 41
and 82 mV when the medium contained 1 and 10 m~ K+
even after several hours of perfusion. No chariges of the
intracellular K+ activity were recorded after tuming light off
or on in cases where no AP was evoked by illumination. The
amplitudes and the shapes of APs showed alrriost no dependente on the extemal K+ concentration. Duriiig APs, the
intracellular K+ activity underwent either a slight decrease or
no changes were recorded. In Figure 5, an example is shown
of APs accompanied by a slight (<8 mM) decrease in the
cytoplasmic K+ activity. Again, the observed activity changes
lon Activities in Conocephalum
did not depend on whether the APs were triggered electrically
(AP number 1, downward pointing arrow) or by light (AP
number 2).
>
i
2
E
1077
-1~~1
'1
-150
Measurements of intracellular CI- and NOsActivities with lon-Selective Microelectrodes
The C1--sensitive microelectrodes (ETH 500 based) had a
Nemstian slope in solutions containing 1 mM or more C1-,
but the slope was significantly smaller (20 mV) between 0.1
and 1 mM C1- (Fig. 2, e and f). The microelectrodes had a
high resistance, between 150 and 200 GQ, and therefore
responded relatively slowly to C1- activity changes. They
needed 5 to 20 s to reach 90% of the voltage after a 10-fold
change in C1- activity. The C1--sensitive cocktail used by us,
like most known anion sensors, had a higher selectivity
toward nitrate over C1-. Thus, it was important to know
exactly the selectivity coefficient of these microelectrodes. We
used the fixed interference method (Vaughan-Jones and
Aickin, 1987) to obtain it. Figure 2e presents original traces
from one of the C1--sensitive microelectrodes calibrated without and with 1.3 mM NO3- as interfering ion. The average
characteristics for five such microelectrodes are given in
Figure 2f. The selectivity coefficient Kcl calculated on the
basis of these curves was 3.83 f 0.9, which means that C1-sensitive microelectrodes were approximately 4 times more
selective for NO3- than for C1-. The interference of NO3with C1--sensitive microelectrodes might lead to a gross over-
Figure 3. Transmembrane potential (toptrace)and cytoplasmicfree
Caz+ concentration (bottom trace) in the thallus of C. conicum in
the absence of APs. A subthreshold electric stimulus (downward
pointing arrow) was followed by a light/dark and a dark/light change
(black and white bars) without induction of an AP. Even at the
enlarged pCa axis (cf. Fig. 4) there were no detectable changes in
the free cytoplasmic Caz+concentration (note expanded time axis
compared to the other figures).
estimation of the intracellular C1- content if the intracellular
Nos- activity is not known.
To solve this problem we applied N03--selective microelectrodes recently developed by Miller and Zhen (1991),
which are much more selective toward Nos- over C1- than
the C1--sensitive microelectrodes described above. These mi-
100-
'";
-1 50 -200-100
tkl
7GZ
e
9
'Oo0.1
-50-
-50-
+78mMCI-
20
-
100
100
::h
-j\
50K
-100-
-100-
f
5 min
h
1001
lO0l
W
O
-1 50
-100
-100
-50
-50
O
-25
5
6 7 0
-log ([Ca2']/M)
-200
o
1
2
-log ([K']/M)
3
-100
1
2
3
4
1
2
3
4
-lOg ([CI-]/M)
Figure 2. Calibration data of ion-selective microelectrodes. a, Recalibration trace of the Caz+-selectivemicroelectrode
used in the experiment presented in Figure 4.The numbers 5, 6, 7, and 8 denote pCa of calibration solutions prepared
according to Ammann et al. (1987a). b, Averaged calibration curves of Caz+-selectivemicroelectrodes: A, first calibration
(dashed line); O, recalibration (solid line) resulting from measurements as shown in a. For recalibrations SE values are
given. Average slope of nine measurements was 33 mV between pCa 5 and 6,28 mV between pCa 6 and 7, and 16 mV
between pCa 7 and 8. c, Calibration traces of K+-selectivemicroelectrode used in the experiment presented in Figure 5.
Reference and potential electrodes were filled with 1 M NaCI. 1, 10, 50, 100, and 1000 denote concentrations of K+ in
calibration solutions in mM. Left, Calibration; right, recalibration. d, Averaged calibration curves of nine K+-selective
microelectrodes (as in b). The abscissa was corrected for K+ activity according to Ammann (1986).e, Original calibration
traces of a CI--sensitive microelectrode in solutions containing 0.1, 1, 10, and 100 mM (concentration)KCI (left trace)
and in the same solutions with addition of 1.3 m M NO,- (right trace). f, Average calibration curves of five CI--sensitive
microelectrodes: O, in solutions without N o s - (solid line); O, with addition of 1.3 m M NO3- as interfering ion (dashed
line) resulting from measurements as shown in e. g, Original traces obtained during calibration of a NO,--selective
microelectrode in solutions containing 0.1, 1, 10, 20, and 100 mM (activity) NO3- (left trace) and in the same solutions
with addition of 78 mM CI- (activity)(right trace). h, Average calibration curves of five N03--selective microelectrodes:
O, in solutions without CI- (solid line); O, with addition of 78 m M CI- as interfering ion (dashed line).
Trcbacz et al.
1078
Plant Physiol. Vol. 106, 1994
-50
-50
-100
-1O0
>
G
E
-150
iz
I
-150
-200
h
5
(u
2
1
5 min
6.5
7.0
2
1
v
=
11'
-200
3
o)
5 min
7.5
Figure 4. Transmembrane potential (top trace) and cytoplasmic free
Ca2+concentration (bottom trace) in Conocephalum during APs. All
traces shown here were obtained during a single measurement on
one thallus lasting 28 min. APs 1 and 2 were evoked by electrical
stimuli: 1 in light, 2 in darkness; AP 3 was triggered by light on. The
fast transients in t h e bottom traces during fast voltage changes are
artifacts caused by different time resolutions of the ion-selective
and the potential microelectrode. Other denotations are as in
Figure 1 .
croelectrodes had a low resistance, from 5 to 20 GQ, and
responded within 2 to 3 s to 10-fold changes in NO3activities. They showed a slight shift from a Nernstian slope
between 0.1 and 1 mM NO3- (Fig. 2, g and h). The average
slope in this range of activities was 36 mV. The selectivity
coefficient K N 0 3 obtained by the fixed interference method
was 0.023 f 0.004, which means that N03--selective microelectrodes were approximately 43 times more selective for
NO3- than for C1-. Figure 2g shows an example of calibration
traces of NO,--selective microelectrodes obtained without
and with 78 mM C1- as interfering anion in the calibration
solutions. The respective calibration curves are presented in
Figure 2h.
The apparent intracellular C1- activity measured with C1-sensitive microelectrodes, (Cl), is the sum of the real C1activity, [Cl], and the real NO3- activity, [NO3],multiplied by
the selectivity coefficient Kcl:
(Cl) = [Cl]
+ 3.83 X [NO,]
Figure 5. Transmembrane potential (top traces) and c!/toplasmic K+
activity (bottom traces) in Conocephalum during APs obtained
during a single measurement. AP 1 was triggered by electrical
stimulation and AP 2 by light on. Other denotations are as in
Figure 1 .
Similarly, the intracellular nitrate activity measured with
N03--selective microelectrodes, (Nos), was grea ter than the
real one by the factor K N 0 3 X [Cl]:
(NO,) = [NO,]
+ 0.023 x [Cl]
Solving this equation system, we obtained the following
formulas for the calculation of intracellular Cl-. and NO3activities:
[NO3] = 1.1 X (NO3) - 0.025
Cytoplasmic K+
Activitv
Calculated
mM
mM
73 f 9 ( n = 8)
76+lO(n=8)
78 f 10 ( n = 9)
o. 1
1
10
(C1;i
[Cl] = 1.1 X (CI) - 4.2 x (NO,)
The lntracellular
CI- and NO3- Activities in Cells of
Conocephalum
Applying the above calculations, we obtained NO different
groups of values for both the C1- and the NO,'- activity. It
seems reasonable to postulate that the lower values were
measured in the cytoplasm and the higher values were measured in the vacuole (Table 11). During single measurements,
transitions from high to low values (but never reverse
Table 1. 7he cytoplasmic K+ activity of C. conicum, the EK calculated from cytoplasmic and external
K+ activities, the resting potential, and the AP amplitudes are compared at external media containing
O.1, I , and 10 mM KCl
A constant ion strength of the medium was adjusted with NaCI.
Externa1 K+
Concentration
X
Resting Potential
Amplitude of AP
mV
mV
mV
-166f 3
-110f3
-55 f 3
-162 f 2 ( n = 22)
-151 f 2 ( n = 2 1 )
-137 t 2 ( n = 22)
108 f 2 ( n = 22)
104f2(n=21)
103 2 ( n = 22)
EK
*
1079
lon Activities in Conocephalum
Table II. The intracellular Cl- and NO3- activities in C. conicum
The cytoplasmic and vacuolar activities resulted from measurements with ion-selective microelectrodes. T h e whole cell activities were calculated from t h e data obtained with ion-selective microelectrodes (ISM) on the basis of 20% cytoplasm and 80% vacuole (v/v).Additionally, t h e whole cell
concentrations were determined bv anion chromatoaraphv (AC) of the cell sap of whole thalli.
Whole
[CI-l/mM
[No,-]/m~
Cytoplasm (activity)
Vacuole (activity)
7.4 f 0.2 ( n = 15)
0.63 f 0.05 ( n = 16)
43.5 f 7 ( n = 4)
3.0 f 0.3 ( n = 5)
changes) were frequently observed (see, for instance, Fig. 7b)
that were probably caused by the exclusion of the microelectrode tip by the tonoplast from the vacuole (Thaler et al.,
1992). On the other hand, the membrane potential (obtained
by potential microelectrodes) was quite constant. We never
recorded a sudden potential decrease, which would be expected if the potential electrode were excluded from a vacuole
with a positive potential inside. This indicates that the potentia1 difference across the tonoplast in Conocephalum is negligible. In approximately 50% of the experiments we succeeded
in measuring C1- and NO3- activities on the same thallus. In
the remaining cases C1- and NO3- activities were measured
on different thalli and the average activities of the interfering
ion from a11 measurements was taken for calculation. The
average cytoplasmic and vacuolar activities for a11 thallus
investigated (Table 11) were 7.4 f 0.2 mM ( n = 15) and 43.5
f 7 mM ( n = 4) for C1-, and 0.63 f 0.05 mM ( n = 16) and
3.0
0.3 mM ( n = 5) for NO3-, respectively. The average
cytoplasmic C1- and Nos- activities calculated on the basis
of the results obtained on the same thalli equalled 7.2 0.3
mM ( n = 7) and 0.53 k 0.09 mM ( n = 8), respectively, which
is not significantly different from the C1- and NO3- activities
that were obtained on different thalli. The ion-selective microelectrodes were only occasionally located in the vacuole,
so that in none of experiments were vacuolar C1- and N03activities obtained on the same thallus.
The cell sap from complete thalli of Conocephalum was
analyzed by anion chromatography. The concentrations ob-
*
+
>
5
-100
-150
1
5 min
36.3
2.5
38.3 f 4 ( n = 4)
2.2 f 0.9 ( n = 3)
conditions of plant culture must be well controlled. Cultivating plants in a soil containing nitrate fertilizers causes accumulation of NO3- and a decrease of C1- contents (data not
shown).
Figure 6 shows an example of parallel measurement of the
transmembrane potential and the cytoplasmic C1- activity.
After tuming off the light we observed either no change of
the cytoplasmic C1- activity or a slow increase by 1 to 2 mM.
Reillumination caused a gradual retum to the previous value
or no change in the C1- activity. Similarly, only slight or no
changes of the intracellular NO3- activity were recorded after
transitions between darkness and light (Fig. 7).
In 30% of our experiments the dynamics of C1--sensitive
microelectrodes allowed us to measure changes in the C1activity during APs. In these cases a small decrease, 1.5 k
0.7 mM ( n = 12), of C1- activity was recorded. An example
of C1- activity changes during an AP is presented in Figure
6. The transients accompanying fast depolarization and repolarization phases of the AP were caused by the longer
response time of the C1--sensitive microelectrode in comparison with the potential one. In most experiments (70%), the
inertia of C1--sensitive microelectrodes was too high. They
could not resolve the rapid potential changes during the AP.
This was manifested by large offsets in the differential signal,
which lasted almost as long as APs. Results of these experiments were only taken into account for calculations of the
Figure 6 . Original traces of transmembrane potential changes (upper trace) and the cytoplasmic CI- activity (lower trace). An action
potential was triggered electrically. Other denotations are as in Figure 1 .
,
-
(concentration)
tained were 38.3 f 4 mM ( n = 4) for C1- and 2.2 f 0.9 mM
( n = 3) for NO3- (Table 11). One has to point out that the
n/ I
W
Whole Cell (AC)
(ISM)
(activity)
Trebacz et al.
1080
Figure 7. Examples of parallel measurements
of the transmembrane potential (upper traces)
and the intracellular NO,- activity (lower
traces). a, The NO,--selective microelectrode
was located in the cytoplasm. b, The NO,-selective microelectrode was inserted into the
vacuole and was spontaneously excluded into
the cytoplasm. Traces in a and b were obtained
on two different thalli. APs were triggered electrically. Other denotations are as in Figure 1.
Plant Physiol. Vol. 106, 1994
a
-50
>
-100
Lu
-150
I
$
2.2 -
z
3.0-
om
2.6-
3.4average steady-state C1- activity. Almost no change in NOsactivity was recorded during APs, no matter whether the ionselective microelectrode was located in the cytoplasm or in
the vacuole (Fig. 7).
DISCUSSION
In this paper we describe the measurement of cytosolic
Ca2+, K+, C1-, and NO3- activities in intact thalli of the
liverwort C. conicum by ion-selective microelectrodes. This
technique allows the registration of changes of the ion activity
of a single cell within an intact Chl-containing tissue, while
simultaneously, the E is monitored. This is hardly possible
with other methods using photoproteins, fluorescent dyes
(Read et al., 1993), or classical isotope techniques. Instead of
double-barreled microelectrodes, which are much harder to
handle (Felle, 1988), we used two separate electrodes, one
ion-selective and one potential. This method is applicable
only in cases in which the system is under good control (no
differences in membrane potential of neighboring cells) (Felle,
1988), as was the case here (Fig. 1; see also Trgbacz, 1992).
It might not be applicable to other plants, especially to those
with complex tissues built up from different cell types. In our
case the impalement of two neighboring cells might even
have the advantage of causing less damage to the cells
compared with the impalement of one cell on two microelectrodes. Additionally, the method was insensitive to distortions that might be caused by electrolyte leakage from the
potential electrode.
The Cytoplasmic Free Caz+Concentration in C. conicum
The concentration of free Ca2+under steady-state conditions in the cytoplasm of the liverwort C. conicum cells
(231 & 65 nM, n = 6) was similar to those recorded in the
liverwort Riccia fluitans (127 & 33 nM) (Felle, 1988) and in
other plant species (Bush, 1993). We did not investigate to
what extent the cytoplasmic free Ca2+concentration varied
during changes of the extemal Ca2+concentration, but it was
shown by Felle (1988, 1991) that the cytoplasmic free Ca2+
concentration is tightly regulated in the liverwort R. fluitans
I
=
I
I
5 min
7
A
.
and in other plant cells (Felle et al., 1992). When the lightdependent potential changes in Conocephalum were not large
(and thus the artifacts caused by the different time resolution
of the electrodes were relatively small; see Figs. 3 and 4), the
resolution of Ca2+-selectivemicroelectrodes for related activity changes was well below 0.1 pCa. Within thc,se limits we
did not detect significant changes in the free cytoplasmic
Ca2+concentration after tuming light on and off, as long as
no AP was induced. This is in contrast to Miller and Sanders
(1987), who measured long-lasting and relatively large lightdependent changes in the free cytoplasmic Ca2'+concentration in Nitellopsis.
The Relationship between the Cytoplasmic Frec! Ca*+
Concentration and the AP
The results presented here show a significant increase of
the free Ca2+concentration during the AP. This increase was
on average 2-fold (from 231 f 65 nM to 477 k '114 nM), and
an up to 6-fold increase was registered. Williamson and
Ashley (1982) estimated that in Chara a 30-fold and in Nitella
a 42-fold increase of the free Caz+ concentration in the
cytoplasm occurred during an AP. The 10-fold lower increase
measured in our case may be caused by the different plant
species or by different methods; Williamson and Ashley
(1982) used aequotin. In principle, it is also possible that the
increase in the free cytoplasmic Ca2+concentration in Conocephalum during the AP was underestimated if there was a
very rapid and transient increase to which Ca2+-selective
microelectrodes responded too slowly to sense. The relatively
high variability of the free CaZ+concentration chmges during
APs measured by us may be caused by the different location
of microelectrode tips within an inhomogeneous cytoplasm
possessing local Ca2+ gradients (Gilroy et al., 1991). The
increase of the free Ca2+ concentration lasted only a few
seconds longer than the AP did. This is similar in Conocephalum and in intact characean cells (Williamson and Ashley,
1982) but at variance with the measurements by Kikuyama
and Tazawa (1983) in tonoplast-free preparations, in which
the free Ca2+concentration remained high severa1 minutes
after the repolarization.
lon Activities in Conocephalum
In Conocephalum, as in other species, the role of Ca2+during
the AP may be rather complex (Beilby, 1984). A depolarizing
Ca2+influx across the plasmalemma seems to be essential,
since different Ca2+channel inhibitors such as La3+, Mn2+,
and verapamil entirely cancel the excitability of Conocephalum
(Trebacz et al., 1989). But the observed increase in the free
cytoplasmic Ca2+concentration might also result from a Ca2+
efflux from interna1 stores into the cytoplasm (Forster, 1990;
Bush, 1993). The increased Ca2+activity might be the trigger
that opens ion channels in the plasmalemma to further
depolarize the membrane. These could be Caz+-dependent
C1- channels, as in characean cells (Lunevsky et al., 1983;
Shiina and Tazawa, 1987; Okihara et al., 1991) (see below),
or Ca2+-dependentK+ channels, as in Eremosphaeva (Thaler
et al., 1989).
Maintenance of Ca2+activity on a low level is an element
of the strategy developed by cells to respond to different
stimuli whose consequence is an increase in the cytoplasmic
Ca2+ activity (Bush, 1993). The increase of the free Ca2+
concentration in the cytoplasm of Conocephalum during the
AP seems sufficient to evoke physiological effects such as the
observed enhancement of the rate of respiration accompanying the AP (Dziubinska et al., 1989). It is worth mentioning
that the increase in the free cytoplasmic CaZ+concentration
was clearly related to the AP itself and not to the stimuli used
to trigger it; this is also the case for the described increase in
the rate of respiration (Dziubinska et al., 1989). It was previously demonstrated using antimony-filled H+-selectivemicroelectrodes that in Conocephalum the pH changes in the
course of APs did not exceed 0.05 pH unit and thus seem to
be too small to influence the activity of respiratory enzymes
(Trgbacz, 1992).
The Cytoplasmic K+ Activity and Its Relation to the
Transmembrane Potential
The intracellular K+ activity in the intact thallus of the
liverwort C. conicum (76 f 10 mM, n = 8) was in the same
range as in other plant species (Leigh and Wyn Jones, 1984;
Maathuis and Sanders, 1993). It was slightly lower than in
the thallus of the liverwort R. fluitans (100-130 mM) (Felle
and Bertl, 1986b; Felle, 1988), where, in contrast to our
measurements, light-dependent changes of the cytoplasmic
K+ activity were registered (Felle and Bertl, 1986b). In Conocephalum the cytoplasmic K+ activity did not significantly
change at a 10-fold increase or decrease of the extemal K+
concentration (Table I).
The resting potential in Conocephalum in the standard
medium containing 1 mM K+ ranged from -140 to -170 mV
in different thalli; occasionally values up to -210 mV were
measured. On the other hand, the mean value of EK was only
-110 mV under this condition (Table I). This indicates that
the transmembrane potential is to a great extent established
by an electrogenic pump (Trgbacz et al., 1989), comparable
to results in the liverwort R. fluitans (Felle and Bentrup, 1976).
One has to point out that at rest there exists a driving force
for K+ directed toward the cytoplasm. Thus, there must be a
1081
mechanism that keeps intracellular K+ concentration at a
physiological level.
A possible consequence of a less-negative E K in comparison
to the resting potential might be that in an early depolarization phase of the AP, K+ channels that carry K+-influxtoward
the cell (inward rectifying) may open, and this could contribute to an initial depolarization. The lack of a corresponding,
transient increase in the cytoplasmic K+ activity in the early
depolarization phase of the AP might be caused by the fact
that it was too small to be detected by the K+-selective
microelectrode. During the repolarization phase of the AP at
potentials less negative than E K , outward-rectifying K+ channels may open, which would cause a K+ efflux and a return
to E K . A slight decrease of intracellular K+ activity during APs
was frequently recorded. The complete repolarization to the
resting potential is probably established by an electrogenic
pump. The amplitudes and the shapes of APs showed no
dependence on the externa1K+ concentrations tested (O. 1-10
mM). Direct characterization of K+ channels in this plant by
voltage-clamp or patch-clamp measurements should help to
clarify this picture.
The lntracellular CI- and NO3- Activities and Their
Relationship to the AP
Most, if not all, anion sensors used for fabrication of ionselective microelectrodes have higher selectivity for NO3than for C1- (Ammann, 1986). Direct measurements of intracellular C1- activity with C1--sensitive microelectrodes give
reliable values only when the NO3- activity is negligibly low,
as in Eremosphaera uiridis (Thaler et al., 1992). Here we used
a method that allows the measurement of intracellular C1and NO3- activities when significant amounts of both anions
are present inside the cells. The purpose of this method is to
measure ion activities in the same plant with two different
anion-sensitive microelectrodes whose selectivity coefficients
toward the respective anions are different and well determined. We obtained cytoplasmicand vacuolar anion activities
of 7.4 f 0.2 mM ( n = 15) and 43.5 f 7 m~ ( n = 4) for C1-,
and 0.63 k 0.05 mM ( n = 16) and 3.0 f 0.3 m~ ( n = 5) for
NO,-, respectively (Table 11). Taking into account that the
cytoplasm in Conocephalum cells constitutes about 20% of the
cell volume, as estimated by centrifugation, one can calculate
an activity for C1- and NOs- of 36.3 f 7 mM and 2.5 f 0.3
m ~ respectively,
,
for the whole cells. This is comparable to
the ion activities calculated for the whole thallus from anion
chromatographic analysis: 28.3 f 3 mM ( n = 4) and 1.6 f 0.6
m~ ( n = 3), respectively. The concentrations obtained by
anion chromatography (given in Table 11) were converted
into activities with an activity coefficient of 0.74 (Ammann,
1986).
Our data (Table 11) show relatively low cytoplasmic C1and NOs- activities. Very divergent data conceming the
cytoplasmic C1- concentration of glycophytes ranging from 1
to 60 mM (and up to 500 mM for halophytes) have been
published (Wyn Jones et al., 1979; Sanders, 1984). More
recent data (Schroppel-Meier and Kaiser, 1988; Thaler et al.,
1992) indicate that the cytoplasmic C1- activity is probably
4 0 mM, at least in glycophytes. Compartmental analyses
with thalli of the liverwort R. fluitans give 11 and 30 m~ for
Trebacz et al.
1082
Plant Physiol. Vol 106, 1994
Table 111. The intracellular ion activities in C. ronicum (the pH data are unpublished)
PH
Cytoplasm
Vacuole
7.16 k 0.05
5.83 k 0.23
[K + l / m ~
[Ca”l/n~
[Cl-]/m~
76 k 10
231 k 65
7.4 f 0.2
43.5 f 7
(about 76)
the cytoplasmic and vacuolar C1- concentration, respectively
(Felle and Bentrup, 1976). The cytoplasmic NOs- activity in
Conocephalum was at the lower end of recently published
values (Miller and Zhen, 1991; King et al., 1992). Our data
point to a 6- and 5-fold concentration gradient for C1- and
NO3- across the tonoplast, respectively. We have good indications that the electrical potential across the tonoplast is
negligibly small. This would mean a net driving force for
both anions from the vacuole into the cytoplasm and thus an
active accumulation of these anions in the vacuole.
We measured no or very slight changes in the anion
activities after light/dark transitions. It was recently reported
(Thaler et al., 1992) that in Eremosphaera darkening resulted
in an increase of C1- activity. Reillumination caused a return
to the light value. It was postulated that changes in the
cytoplasmic C1- activity reflected ion fluxes compensating for
light-induced pH changes (Thaler et al., 1992). In Conocephalum such pH changes are much smaller than in Eremosphaera (Trpbacz, 1992), and C1- fluxes accompanying them
might be too small to be detected with ion-selective microelectrodes, especially on the higher background (7.4 compared to 2.2 mM) of C1- activity.
Our results indicate that in the standard experimental
medium the Ecl (across the plasmalemma) is +50 mV. At
about -150 mV E there is a large driving force for C1- out of
the cell, and opening of C1- channels would result in C1efflux. We frequently measured a small decrease of the
cytoplasmic C1- activity during APs. C1- channels probably
opened during the AP in Conocephalum, allowing C1- efflux
and giving rise to a depolarization.
A Model for the lon Fluxes during an AP
Table I11 summarizes the intracellular ion activities in Conocephalum known so far. Together with the extemal ion
Figure 8. A scheme illustrating the sequence
of ion fluxes across the plasma membrane of
Conocephalum in the steady state and during
APs. T h e transient depolarization of t h e E during an AP is shown at the bottom in an expanded time scale (compared to the other figures). In steady state the transmembrane
potential is established by an electrogenic
p u m p that extrudes H+ from t h e cytoplasm at
t h e expense of ATP. Most ion channels remain
closed in steady state. In t h e depolarization
phase of APs, Ca2+ influx and CI- efflux take
place. K+ efflux and t h e activity of t h e H+ p u m p
cause the repolarization of the transmembrane
potential; t h e complete return to resting potentia1 is governed by H+ p u m p activity.
IN
OUT
B
[N O a - l / m ~
0.63 ? 0.05
3.0 k 0.3
concentrations of the standard experimental medium (pH
7.0, 1 mM K+, 0.1 mM Ca2+,1.2 mM C1-, and no h 0 3 -added)
and the E (about -150 mV), this gives the framework of ion
gradients and driving forces one needs to unders.and APs in
Conocephalum. On the basis of previous studies (Trpbacz et
al., 1989) and the data presented here, we suggest a basic
model for the ion fluxes across the plasma ml.mbrane of
Conocephalum in the steady state and during AR; (Fig. 8). In
the steady state the transmembrane potential is established
by the electrogenic H+ pump extruding H+ out of the cytoplasm (Trpbacz et al., 1989), comparable to what happens in
the liverwort R. fluitans (Felle and Bentrup, 1976). At the
large negative resting potential most physiologically important ions are kept far away from their electrochemical equilibrium (Table 111). Strong driving forces promote Ca2+influx
and anion, especially C1-, efflux. We demonstrated that the
free cytoplasmic Ca2+ concentration increased significantly
and the cytoplasmic C1- activity decreased duIing an AP.
This points to Ca2+influx and C1- efflux during ihe depolarization phase of APs in Conocephalum. Probably Ca2+ and
anion channels are opened during the first ph,ise of APs,
causing the depolarization and the respective iori fluxes. C1seems to play a straightforward part as a depolarizing efflux,
the role of Ca2+is probably more complex (Beilby, 1984).
Ca2+ influx across the plasmalemma would depolarize the
membrane, yet Ca2+ might also be necessary to open C1chan els of the plasma membrane, as in characean cells
(Lunevsky et al., 1983; Shiina and Tazawa, 1987. Okihara et
al., 1991). Peak depolarization during APs in Ccnocephalum
never reaches, at least in the standard medium, ihe Ecl. This
is possibly because of a rapid inactivation of aniim channels
and/or the opening of other ion channels. We measured a
slight decrease of the cytoplasmic K+ activity during APs,
which suggests a K+ efflux. The opening of K+ channels at
potentials more positive than the EK would cause a K+ efflux
IN
IN
OUT
2
Ca2+4 x
OUT
CI-
1
H+
1 min
lon Activities in Conocephalum
and a membrane repolarization before Ecl is reached. This K+
efflux can repolarize the membrane to EK, but the resting
potential more negative than EK can be achieved only by an
electrogenic ion pump (cf. Table III), probably a H+ pump
(Trsbacz et al., 1989). Meanwhile, the membrane conductance for K+ has to decrease again to prevent a K+ influx. This
ion mechanism for the repolarization phase of APs is probably common for a11 plant species whose resting potential
has a large “metabolic”component.
With the above model and the data given in Table 111, some
quantitative considerations become meaningful. The microelectrodes were inserted into the cells of the second layer of
the thallus of Conocephalum, which have a length of about
150 to 200 pm and a diameter of about 80 pm. This gives a
surface of about 6 X 104pm2 and a volume of about 106pm3.
In these cells the cytoplasm constitutes about 20% of the cell
volume, i.e. a cytoplasmic volume of 2 x 105 pm3, or 0.2 nL.
Taking the well-establishedvalue of 1 pF/cm2 for the specific
capacitance of biological membranes (Cole, 1970), one calculates that for a depolarization of 100 mV only about 6 X
10-l6 mo1 ions have to cross a membrane with a surface of
6 X 104 pm2. This is equivalent to a concentration change of
about 3 p~ for a cytoplasmic volume of 0 . 2 nL. Therefore,
the flow of ions needed to cause the observed depolarization
had an insignificant effect on cytosolic free concentration and
thus would not have been detected by our measurements.
This also holds for Ca2+, since about 99.9% of the total
cytosolic Ca2+ are bound and not free (Carafoli, 1987). It
seems to be characteristic for plant APs (in contrast to animal
APs) that the ion fluxes are much larger than required for
loading the membrane capacitance; this is also the case for
different giant alga1 cells (Gaffey and Mullins, 1958; Kikuyama et al., 1984; Mummert and Gradmann, 1991) and
probably for higher plants as well (Abe, 1981; Fromm and
Spanswick, 1993). It has even been suggested that the origin
of electrical excitation in plants is more a matter of osmoregulation and that AP in plants could well be an epiphenomenon of the osmotic events (Mummert and Gradmann, 1991;
Gradmann et al., 1993). However, there are different reports
in which the role of APs for intercellular signaling in plants
was established (Dziubiríska et al., 1989; Wildon et al., 1992;
Fromm and Eschrich, 1993). The large ion currents occumng
during an AP have to flow in parallel, and the K+ efflux
probably electrically balances most (99.9%)of the Ca2+influx
and the C1- efflux. The arrows in Figure 8 may be interpreted
as the very small portion of ion fluxes not balanced by counter
ions and causing the transient depolarization.
It is worth mentioning that the activity changes we measured during APs were transient for a11 ions. In a11 cases the
activity before and after an AP did not differ significantly.
On the other hand, for Characeae it is well established that
during an AP there is a net uptake of Ca2+and a net release
of K+ and C1- from the cells (Gaffey and Mullins, 1958; Oda,
1976; Kikuyama et al., 1984). The data available suppose that
this is also the case for higher plants (Samejima and Sibaoka,
1980; Abe, 1981; Fromm and Spanswick, 1993). Net ion
fluxes from the cell to the extemal medium without corresponding permanent activity changes in the cytoplasm point
to intemal organelles, probably mainly the vacuole, as source
for the ions involved. In tonoplast-free Characeae cells no
1083
net C1- efflux could be detected (Kikuyama et al., 1984), and
the free Ca2+concentration remained increased severa1 minutes after the repolarization (Kikuyama and Tazawa, 1983).
So there are probably also ion fluxes across the tonoplast
during APs in plants, an efflux of anions from the vacuole
into the cytoplasm driven by the electrochemical gradient
(Table 111) and electrically compensated by a parallel K+
efflux.
The rapid retum of the free cytoplasmic Ca2+concentration
to the low resting leve1 during the repolarization phase (Fig.
4) could be achieved by the plasma membrane CaZ+-ATPase
(Felle et al., 1992) or by an active uptake of Ca2+into the
vacuole or another intemal Ca2+ store (Chanson, 1993;
Pfeiffer and Hager, 1993). Therefore, the transient changes
in the cytoplasmic ion activities reported here probably indicate the net efflux of K+ and C1- from the vacuole to the
extemal medium accompanied by an opposite Ca2+uptake.
As the measured activity changes are by orders of magnitude
larger than necessary for the depolarization of the plasma
membrane, the APs, besides their role in signaling (Dziubiiíska et al., 1989; Wildon et al., 1992; Fromm and Eschrich,
1993), likely contribute to osmoregulation (Mummert and
Gradmann, 1991; Gradmann et al., 1993).
ACKNOWLEDCMENTS
We thank Dr. W.M. Kaiser, Wiirzburg, Germany, for analyses by
anion chromatography and B. Bethmann and M. Thaler for helpful
discussion and advice.
Received March 14, 1994; accepted June 6, 1994.
Copyright Clearance Center: 0032-0889/94/l06/1073/12.
LITERATURE ClTED
Abe T (1981) Chloride ion efflux during an action potential in the
main pulvinus of Mimosa pudica. Bot Mag 9 4 379-383
Ammann D (1986) Ion-Selective Microelectrodes. Principles, Design
and Application. Springer-Verlag,Berlin
Ammann D, Biihrer T, Schefer U, Miiller M, Simon W (1987a)
Intracellular neutra1 camer-based Ca2+microelectrode with subnanomolar detection limit. Pfliigers Arch Eur J Physiol 409
223-228
Ammann D, Chao P, Simon W (1987b) Valinomycin-based K+
selective microelectrode with low electrical membrane resistance.
Neurosci Lett 7 4 221-226
Beilby MJ (1984) Calcium and plant action potential. Plant Cell
Environ 7: 415-421
Brustovetsky NN,Egorova MV, Gnutov DY (1993) The mechanism
of calcium-dependent activation of respiration in liver mitochondria from hibemating ground squirrels, Citellus undulatus. Comp
Biochem Physiol106B 423-426
Bush DS (1993) Regulation of cytosolic calcium in plants. Plant
PhysiollO3 7-13
Carafoli E (1987) Intracellular calcium homeostasis. Annu Rev
Biochem 5 6 395-433
Chanson A (1993) Active transport of proton and calcium in higher
plant cells. Plant Physiol Biochem 31: 943-955
Chao P, Ammann D, Oesch U, Simon W, Lang F (1988) Extra- and
intracellular hydrogen ion-selective microelectrode based on neutral camers with extended pH response range in acid media.
Pfliigers Arch Eur J Physiol411: 216-219
Cole KS (1970) Dielectric properties of living membranes. In F Snell,
J Wolken, G Iverson, J Lam, eds, Physical Principles of Biological
Membranes. Gordon & Breach Science Publishers, New York, pp
1-16
Trrbacz e t al.
1084
Davies E (1987)Action potentials as multifunctional signals in plants:
a unifying hypothesis to explain apparently disparate wound responses. Plant Cell Environ 1 0 623-631
Dziubihska H, Paszewski A, Trpbacz K, Zawadzki T (1983) Electrical activity of the livenvort Conocephalum conicum: the all-ornothing law, strength-duration relation, refractory periods and
intracellular potentials. Physiol Plant 57: 279-284
Dziubibka H, Trpbacz K, Zawadzki T (1989) The effect of excitation on the rate of respiration in the livenvort Conocephalum
conicum. Physiol Plant 7 5 417-423
Fein A, Tsacopoulos M (1988) Activation of mitochondrial oxidative
metabolism by calcium ions in Limulus ventral photoreceptors.
Nature 331: 437-440
Felle H (1988) Cytoplasmic free calcium in Riccia fluitans L. and Zea
mays L.: interaction of CaZ+and pH? Planta 176 248-255
Felle H (1991) Aspects of Ca2+homeostasis in Riccia fluitans: reactions to perturbations in cytosolic-free Ca2+.Plant Sci 7 4 27-33
Felle H, Bentrup F-W (1976) Effect of light upon membrane potential, conductance, and ion fluxes in Riccia fluitans. J Membr Biol
27: 153-170
Felle H, Bertl A (1986a) The fabrication of H+-selectivemicroelectrodes for use in plant cells. J Exp Bot 37: 1416-1428
Felle H, Bertl A (1986b) Light-induced cytoplasmicpH changes and
their interrelation to the activity of the electrogenic proton pump
in Riccia fluitans. Biochim Biophys Acta 848: 176-182
Felle H, Tretyn A, Wagner G (1992) The role of the plasmamembrane CaZ+-ATPasein Ca2+homeostasis in Sinapis alba root
hairs. Planta 1 8 8 306-313
Felle HH (1993) Ion-selective microelectrodes: their use and importance in modem plant cell biology. Bot Acta 106: 5-12
Forster B (1990) Injected inositol 1,4,5-trisphosphate activates Ca2+sensitive K+ channels in the plasmalemma of Eremosphaera viridis.
FEBS Lett 269 197-201
Fromm J, Eschrich W (1993) Electric signals released from roots of
willow (Salix viminalis L) change transpiration and photosynthesis.
J Plant Physiol 141: 673-680
Fromm J, Spanswick R (1993) Characteristics of action potentials in
willow (Salix viminalis L.). J Exp Bot 4 4 1119-1125
Gaffey CT, Mullins LJ (1958) Ion fluxes during the action potential
of Chara. J Physiol144 505-524
Gilroy S, Fricker MD, Read ND, Trewavas AJ (1991) Role of
calcium in signal transduction of Commelina guard cells. Plant Cell
3 333-344
Gradmann D, Blatt MR, Thiel G (1993) Electrocoupling of ion
transporters in plants. J Membr Biol 136 327-332
Graziana A, Bono J-J, Ranjeva R (1993) Measurement of cytoplasmic calcium by optical fluorescence in plant systems. Plant
Physiol Biochem 31: 277-281
Guern J, Felle H, Mathieu Y, Kurkdjian A (1991) Regulation of
intracellular pH in plant cells. Int Rev Cytol 127: 111-173
Hodick D, Sievers A (1988) The action potential of Dionea muscipula
Ellis. Planta 174 8-18
Iijima T, Sibaoka T (1985) Membrane potentials in excitable cells of
Aldrovanda vesiculosa trap-lobes. Plant Cell Physiol26 1-13
Kikuyama M, Oda K, Shimmen T, Hayama T, Tazawa M (1984)
Potassium and chloride effluxes during excitation of Characeae
cells. Plant Cell Physiol 2 5 965-974
Kiku ama M, Tazawa M (1983) Transient increase of intracellular
Ca dunng excitation of tonoplast-free Chara cells. Protoplasma
117: 62-67
King BJ, Siddiqi MY, Glass ADM (1992) Studies of the uptake of
nitrate in barley. V. Estimation of root cytoplasmic nitrate concentration using nitrate reductase activity: implications for nitrate
influx. Plant Physiol99 1582-1589
Kondo Y, Bührer T, Seiler K, Fromter E, Simon W (1989) A new
double-barrelled, ionophore-based microelectrode for chloride
ions. Pfliigers Arch Eur J Physiol414 663-668
Leigh RA, Wyn Jones RG (1984) A hypothesis relating critical
potassium concentrations for growth to the distribution and function of this ion in the plant cell. New Phytol97: 1-13
Lunevsky VZ, Zherelova OM, Vostrikov IY, Berestovsky GN
(1983) Excitation of Characeae cell membranes as a result of
activation of calcium and chloride channels. J Membr Biol 72:
43-58
7
+
'
Plant Physiol. Vol. 106, 1994
Maathuis FJM, Sanders D (1993) Energization of potassium uptake
in Arabidopsis thaliana. Planta 191: 302-307
Miller AJ, Sanders D (1987) Depletion of cytosolic free calcium
induced by photosynthesis. Nature 326 397-400
Miller AJ, Zhen R-G (1991) Measurement of intracellular nitrate
concentrations in Chara using nitrate-selective mi1:roelectrodes.
Planta 184: 47-52
Minorsky PV, Spanswick RM (1989) Electrophysiolotjcal evidence
for a role for calcium in temperature sensing by rooh of cucumber
seedlings. Plant Cell Environ 1 2 137-143
Mummert H, Gradmann D (1991) Action potentials in Acetabularia:
measurement and simulation of voltage-gated fluxes. J Membr Biol
124 265-273
Oda K (1976) Simultaneous recording of potassium and chloride
effluxes during action potentials in Chara corallinz. Plant Cell
Physiol17: 1085-1088
Okihara K, Ohkawa T, Tsutsui I, Kasai M (1991) A Ca2+- and
voltage-dependent C1--sensitive anion channel in the Chara plasmalemma: a patch-clamp study. Plant Cell Physiol32 593-601
Pfeiffer W, Hager A (1993) A Ca2+-ATPaseand a hlg2+/H+-antiporter are present on tonoplast membranes from roois of Zea mays
L. Planta 191: 377-385
Read ND, Allan WTG, Knight H, Knight MR, Malh.6 R, Russell
A, Shacklock PS, Trewavas AJ (1992) Imaging and :neasurement
of cytosolic free calcium in plant and funga1 cells. J Microsc 166
57-86
Samejima M, Sibaoka T (1980) Changes in extracellularion concentration in the main pulvinus of Mimosa pudica during rapid movement and recovery. Plant Cell Physiol21: 467-479
Samejima M, Sibaoka T (1982) Membrane potentials and resistance
of excitable cells in the petiole and main pulvinus of Mimosa
pudica. Plant Cell Physiol23 459-465
Sanders D (1984) Gradient-coupled chloride transport in plant cells.
In GA Gerencser, ed, Chloride Transport Coupling in Biological
Membranes and Epithelia. Elsevier, Amsterdam, pp 63-120
Sanders D (1990) Kinetic modeling of plant and fungd membrane
transport. Annu Rev Plant Physiol Plant Mo1 Biol41: 77-107
Schroppel-Meier G, Kaiser WM (1988) Ion homeosta,jis in chloroplasts under salinity and mineral deficiency. I. So1u:e concentrations in leaves and chloroplasts from spinach plants under NaCl
or NaN03 salinity. Plant Physiol87: 822-827
Shiina T, Tazawa M (1987) Ca2+-activatedC1- channel in plasmalemma of Nitellopsis obtusa. J Membr Biol99 137-146
Thaler M, Simonis W, Schonknecht G (1992) Light-dependent
changes of the cytoplasmic H+ and C1- activity in the green alga
Eremosphaera viridis. Plant Physiol99 103-1 10
Thaler M, Steigner W, Forster B, Kohler K, Simonis W, Urbach W
(1989) Calcium activation of potassium channels in the plasmalemma of Eremosphaera viridis. J Exp Bot 4 0 1195-1203
Trpbacz K (1992) Measurements of intra- and extractillular pH in
the livenvort Conocephalum conicum during action pottmtials. Physiol Plant 84: 448-452
Trpbacz K, Tarnecki R, Zawadzki T (1989) The efficts of ionic
channel inhibitors and factors modifying metabolism on the excitability of the livenvort Conocephalum conicum. Physiol Plant 7 5
24-30
Trpbacz K, Zawadzki T (1985) Light-triggered action potentials in
the livenvort Conocephalum conicum. Physiol Plant 64: 482-486
Tyerman SD (1992) Anion channels in plants. Anni Rev Plant
Physiol Plant Mo1 Biol43 351-373
Vaughan-Jones RD, Aickin CC (1987) Ion-selectiv': microelectrodes. In NB Standen, PTA Gray, MJ Whitaker, eds, Microelectrode Techniques: The Plymouth Workshop Handbook. The Company of Biologists Limited, Cambridge, UK, pp 137-167
Wildon DC, Thain JF, Minchin PEH, Gubb IR, Reilly AJ, Skipper
YD, Doherty HM, O'Donnell PJ, Bowles DJ (1992) Electrical
signalling and systemic proteinase inhibitor indwtion in the
wounded plant. Nature 360 62-65
Williamson RE, Ashley CC (1982) Free CaZ+ and cytoplasmic
streaming in the alga Chara. Nature 296: 647-651
Wyn Jones RG, Brady CJ, Speirs J (1979) Ionic and osm(3ticrelations
in plant cells. In DL Laidman, RG Wyn Jones, eds, Recent Advances
in the Biochemistry of Cereals. Academic Press, London, pp
63-103