Plant Physiol. (1991) 97, 1551-1557
Received for publication May 20, 1991
Accepted July 21, 1991
0032-0889/91/97/1551 /07/$01 .00/0
Lidocaine and ATPase Inhibitor Interaction with the
Chloroplast Envelope1
Weihua Wu and Gerald A. Berkowitz*
Horticulture Department, Cook College, Rutgers-The State University of New Jersey,
New Brunswick, New Jersey 08903
ABSTRACT
the chloroplast envelope is modulated by several systems. K+
fluxes, either directly (14) or indirectly (3), are somehow
linked to H+ countermovement. The action of an envelope
ATPase (again, either directly or indirectly) is required to
maintain an alkaline stroma under certain conditions (14,
17). Chloroplast envelope- bound Mg2` also contributes to
the regulation of H+ flux, possibly at the site that links H+
and K+ counterfluxes (11, 14, 17, 20).
In a recent study from this laboratory, several inhibitors
were identified as affecting the ability of the chloroplast to
maintain an alkaline stromal pH in the light (17). The cardiac
glycoside digitoxin was found to cause stromal acidification
and photosynthetic inhibition. It was speculated that this
compound acted by binding to, and inhibiting, an envelope
ATPase in a fashion similar to the previously reported action
of oligomycin (14). Under conditions that facilitated net H+
influx, the amine anesthetic lidocaine was found to increase
stromal pH. Because this anesthetics's action in animal systems is mediated by the blockage of monovalent cation channels (2), it was speculated that lidocaine's effects on chloroplast metabolism were due to a restriction of K+ efflux through
an ion channel, indirectly resulting in a reduction of H+
import.
Identification of these pharmacological agents as affecting
chloroplast metabolism offers some new probes with which
the envelope proteins regulating H+ fluxes can be further
characterized. For example, digitoxin has been reported to
only affect the Na+/K+ pump in animal systems (21); Na+/
K+ ATPases have not been previously in plant membranes.
Identification and characterization of the digitoxin-binding
site on the chloroplast envelope could aid in characterizing
the nature of this ATPase. Ion channels have not been previously reported to be present in the chloroplast envelope.
Further work with lidocaine could aid in characterizing that
component of the envelope transport system. It was the
objective of the work reported in this study, then, to further
characterize how these pharmacological agents interact with
chloroplast metabolism.
Photosynthetic capacity of isolated intact chloroplasts is known
to be sensitive to K+ fluxes across the chloroplast envelope.
However, little is known about the system of chloroplast envelope
proteins that regulate this K+ movement. The research described
in this report focused on characterizing some of the components
of this transport system by examining inhibitor effects on chloroplast metabolism. Digitoxin, an inhibitor of membrane-bound
Na+/K+ ATPases, was found to reduce stromal K+ at a range of
extemal K+ and inhibit photosynthesis. Scatchard plot analysis
revealed a specific protein receptor site with a Km for digitoxin
binding of 13 nanomolar. Studies suggested that the receptor site
was on the interior of the envelope. The effect of a class of amine
anesthetics that are known to be K+ channel blockers on chloroplast metabolism was also studied. Under conditions that facilitate low stromal pH and concomitant photosynthefic inhibition,
the anesthetic, lidocaine, was found to stimulate photosynthesis.
This stimulation was associated with the maintenance of higher
stromal KV. Comparison of the effects on photosynthesis of lidocaine analogs which varied in lipophilicity suggested a lipophilic
pathway for anesthetic action. The results of experiments with
lidocaine and digitoxin were consistent with the hypothesis that
a K+ channel and a K+-pumping envelope ATPase contribute to
overall K+ flux across the chloroplast envelope. Under appropriate
assay conditions, photosynthetic capacity of isolated chloroplasts was shown to be much affected by the activity of these
putative envelope proteins.
The chloroplast inner envelope is an osmotic barrier, and
its free permeability is limited to small uncharged molecules
such as H20, NH3, CO2, and 02 (4). During light-induced
stromal alkalinization (8), a pH gradient is maintained between the stromal compartment and the cytoplasm. However,
H+ movement can occur across the envelope. Stroma to
cytoplasm H+ flux is facilitated, and regulated, by a not wellcharacterized system of membrane proteins. The integrated
action of these proteins results in the maintenance of a
transenvelope gradient (in the light) of nearly 1 pH unit (25).
Net efflux of H+ from the illuminated chloroplast may also
occur apparently against this chemical gradient (6, 8).
Previous work has established that overall H+ flux across
MATERIALS AND METHODS
Plant Material
Spinach (Spinacia oleracea var 'Melody') seeds were
planted in pots containing 2000 cm3 potting mix (1:1, v/v,
peat:vermiculite) and thinned to three to five plants/pot after
2 weeks. Pots were watered twice/week with tap water and
once/week with complete (Peter's Geranium Special plus
'
New Jersey Agricultural Experiment Station, publication No.
12149-5-91. This material is based upon work supported by the U.S.
Department of Agriculture Competitive Research Grants Office under award No. 90-37262-5586.
1551
1 552
WU AND BERKOWITZ
micronutrients) fertilizer. Plants were grown in a growth
chamber with a 10-h light (480 umol/m2/s) period at
21°C (day)/16°C (night). Fully expanded, nonsenescing leaves
taken from 6- to 8-week-old plants were used for chloroplast
isolation.
Chloroplast Isolation
Intact (routinely >90%) chloroplasts were isolated from
spinach using procedures described previously (17). Briefly,
leaves were homogenized in 50 mL grind medium containing
0.33 M sorbitol, 50 mM Hepes-NaOH (pH 6.8), 2 mM
Na2EDTA, 1 mM MgC12, and 1 mM MnCl2. After centrifugation (750g for 50 s in a Sorvall RC5C centrifuge and SS-34
rotor), pelleted chloroplasts were resuspended in 5 mL grind
medium and carefully layered on top of an 8-mL cushion
(40% Percoll and 0.5% [w/v] BSA in grind medium) in 15mL tubes. These step gradients were then centrifuged (2000g
in an HS-4 rotor for 1 min). Intact chloroplasts were recovered
in the pellet and resuspended in a small volume (0.5-1.0 mL)
of grind medium. All steps in the isolation procedure were
carried out at 0 to 2°C.
Thylakoid Membrane Isolation
After isolation, intact chloroplasts were suspended in 20 to
30 mL buffer (20 mm Hepes-KOH, pH 7.6, with 4 mm
MgCl2), vortexed, and kept on ice for 2 min. Thylakoids
released from osmotically shocked chloroplasts were recovered by centrifugation at 15OOg (HS-4 rotor) for 4 min.
The green pellet was resuspended in 1 to 2 mL ATPase assay
medium as described by Hargarter and Ort (7) which contained 0.1 M sorbitol, 20 mm Hepes-KOH (pH 7.8), 5 mM
DTT, 20 mm KCl, and 4 mM MgCl2 and stored on ice until
used.
Photosynthesis
C02-supported 02 evolution of intact chloroplasts was
measured with Hansatech (Decagon Instr., Seattle, WA) 02
electrodes at 25°C and 1500 ,umol/m2/s PAR supplied by the
500 W bulb of a slide projector focused through a water heat
filter. Typically, 50 ML grind medium containing 20 to 30 ,g
Chl was added to 1 mL final volume photosynthesis reaction
medium (0.33 M sorbitol, 50 mM Hepes-NaOH [pH 7.6], 2
mM Na2EDTA, 1 mM MgCl2, 1 mM MnCl2, 0.25 mm NaH2PO4, 5 mm NaHCO3, and 1000 units/mL catalase) with
additions as noted in the text. In some experiments, the
chloroplasts were exposed to a "high Mg2+" treatment. In this
case, 5 mm extra Mg2+ was added (as the chloride salt) to the
standard reaction medium, which already contained 2 mM
divalent cations and 2 mM EDTA. In this case, the chloroplasts were exposed to high levels of free external Mg2+, which
increases the amount of chloroplast envelope-bound Mg2e
(20).
Stromal K
Stromal K+ at varying external K+ was ascertained using
silicone oil microcentrifugation. Four 400-AL microcentrifugation tubes were used as replicates for each measurement.
Plant Physiol. Vol. 97, 1991
Each microfuge tube contained a bottom layer of 20 ,L 14%
(v/v) HC104, 100 ,L silicone oil mixture (as described in ref.
17), and a top layer of 200 ,uL photosynthetic reaction medium with chloroplasts equivalent to approximately 20 Mig
Chl. The tubes were incubated at 25°C in a water bath with
1200 MUmol/m2/s PAR supplied by banks of 500 W incandescent floodlights placed on either side of the waterbath. After
10 min, tubes were microcentrifuged for 15 s in a Beckman
microfuge B. Frozen tubes were cut in the oil layer, and the
bottom portion of the tube with the HC104 fraction was
placed in a 1.5-mL microfuge tube with 480 uL water, vortexed, and centrifuged for 1 min. The supernatant was assayed
for K+ using a Perkin-Elmer 2280 atomic absorption spectrophotometer (Norwalk, CT). All samples and standards were
made up to contain 0.2% LaC13; absorbance was measured at
a wavelength of 766.5 nm. K+ in the HC104 fraction of the
microcentrifuge tubes that centrifuged with the chloroplasts,
but was external to the stroma, was estimated and subtracted
from the total in each tube to ascertain stromal K+ levels.
This was done by measuring the [14C]sorbitol-permeable
space, as described below, under the various treatments for
the chloroplast preparation and assuming that the external
K+ equilibrated into this volume. Stromal K+ concentration
was derived from a measure of the stromal volume (ascertained as described below) under the various treatments.
Chloroplast Volume
The dual-label silicone oil microcentrifugation technique
developed by Heldt et al. (8) as modified by Peters and
Berkowitz (17) was used to ascertain stromal volume and
sorbitol-permeable space associated with chloroplasts. Chloroplasts were incubated in 200 uL photosynthetic reaction
medium which contained 12.5 MCi/mL ['4C]sorbitol and 5
MCi/mL [3H]H20 above layers of 100 ,L silicone oil and 20
,ML HC104 in 400-ML tubes. After 5 min, chloroplasts were
separated from the medium above the silicone oil layer by
microcentrifugation for 15 s. 3H and 14C in the supernatant
and the HC104 fraction of the tubes were ascertained using
dual-label dpm programs with external standards ratio quench
correction on a Beckman 3801 liquid scintillation spectrophotometer. The 3H20 volume associated with the chloroplasts represented the sum of the stromal and sorbitol-permeable space. The ['4C]sorbitol space represented the sorbitolpermeable (i.e. extrastromal) volume of reaction medium
associated with the chloroplasts in the HC104 fraction. The
difference between the two was taken to be a measure of
stromal volume.
Digitoxin-Binding Studies
Digitoxin binding to the chloroplast envelope was estimated
using silicone oil microcentrifugation of chloroplasts incubated in solutions containing [3H]digitoxin and [14C]sorbitol.
Intact chloroplasts were incubated under the same conditions
used for the stromal K+ determinations except labeled sorbitol
(at 12.5 ,Ci/mL) and digitoxin (at 26.1 Ci/mmol) were added
to the photosynthetic reaction medium. In addition to the
labeled sorbitol and digitoxin, a second set of tubes was made
up with unlabeled cardiac glycoside added above that supplied
INHIBITOR INTERACTION WITH THE CHLOROPLAST ENVELOPE
1553
Table I. Mg2+, ATPase Inhibitor, and Lidocaine Effects on Photosynthesis of Isolated Chloroplasts
In this table, the D and 0 treatments refer to 6.5 Mm digitoxin and 20 Ag/mL oligomycin, respectively.
Lidocaine was at 100 AM and Mg2+ was supplied to the reaction medium at 5 mm (above the level
chelated by EDTA). Results of three separate experiments are presented; data are shown as means +
SE (n = 4).
Treatment
Photosynthesis
% Inhibitiona
Experiment 1 Experiment 2 Experiment 3 Experiment 1 Experiment 2 Experiment 3
AMoI 02/mg Chl/h
Control
143 ± 2.7a 101 ± 3.2 170 ± 4.5
0
0
0
Mg2+
103 ± 3.2
69 ± 2.5 122 ± 3.8
28
32
28
+D,
68 ± 2.4
48 ± 3.0
87 ± 3.2
52
52
49
+D, lidocaine
83 ± 2.4
58 ± 2.5 116 ± 3.6
42
42
32
+0
46±1.9
32±2.0
76±2.6
68
68
55
67 ± 7.0
+0, lidocaine
54 ± 3.5
94 ± 4.2
53
46
45
a
For each experiment, the rate under the various treatments is compared with the control rate.
by the addition of labeled digitoxin. Chloroplasts in the 400,uL microfuge tubes were incubated in the light at 25°C for 5
min prior to centrifugation for 15 s. The ['4C]sorbitol pelleting
with the chloroplasts in each tube was used to estimate and
subtract out the [3H]digitoxin in the HC104 fraction that was
not associated with the chloroplast. The amount of labeled
digitoxin bound to the chloroplast envelope in the presence
of both labeled and nonlabeled (10 uM) digitoxin was subtracted from the amount of labeled digitoxin bound to the
chloroplast envelope in the presence of only labeled digitoxin.
The differences represent specific binding of a ligand to a
receptor molecule and were used in Scatchard Plot analyses
(1) to distinguish between specific binding of digitoxin to a
receptor molecule and nonspecific association ofthis inhibitor
with the chloroplast envelope.
Reagents
['4C]sorbitol and [3H]H20 were obtained from ICN (Irvine,
CA) and [3H]digitoxin was from New England Nuclear (Boston, MA). Silicone oils were from William F. Nye Co. (New
Bedford, MA), and Hepes was from Research Organics (Cleveland, OH). Bupivacaine was obtained as a gift from the
manufacturer (Sterling-Winthrop Research Institute, Rensselaer, NY). Lidocaine analogs QX-222 and QX-314 were obtained as a gift from the manufacturer (Astra Pharmaceutical,
Westborough, MA). Lidocaine prepared as the HCI salt was
obtained from Research Biochemicals (Natick, MA). All other
reagents were from Sigma (St. Louis, MO).
RESULTS AND DISCUSSION
Envelope Protein Inhibitor Effects on Photosynthesis
Incubation of chloroplasts in solutions containing millimolar levels of unchelated Mg2+ leads to stromal acidification
in vitro, resulting in photosynthetic inhibition (1 1, 17). It can
be speculated that, under this condition, the activity of chloroplast envelope proteins that facilitate H+ efflux would be
necessary to maintain maximal rates of photosynthesis. Conversely, inhibition of an envelope protein whose activity results in H+ influx should increase photosynthesis when an H+
extrusion system has been "turned off." These relationships
are demonstrated in the series of experiments shown in Table
I. In three separate experiments, addition of Mg2" to the
reaction medium inhibited photosynthesis by about 30%. In
the presence of added Mg2", exposure of chloroplasts to the
ATPase inhibitor digitoxin or oligomycin resulted in further
photosynthetic inhibition. These ATPase inhibitors decrease
the stromal pH of the illuminated chloroplast (11, 17). The
data in Table I are consistent with previous assertions in the
literature (14, 15) which postulate that ATPase-facilitated H+
efflux from the stroma is necessary for optimal photosynthesis. In further studies, it was determined that, after isolation, chloroplasts become increasingly sensitive to the Mg2"
and the ATPase inhibitor treatments. During a 2.5-h period
when chloroplasts were left in grind medium at 0C, photosynthetic capacity (measured at 25°C) was hardly diminished;
however, the photosynthetic inhibition caused by Mg2+ and
digitoxin greatly increased (data not shown). Possibly, the
increased sensitivity to these treatments with time may reflect
a general collapse in the ability of the illuminated chloroplast
to maintain high stromal pH, which may be related to a loss
in stromal K+ (Fig. 1).
It should be noted that not all previous studies support the
contention that the activity of an envelope ATPase contributes to the maintenance of an alkaline stromal pH in the
light. Reduction of the ATP level (using coupling factor
inhibitors) in the illuminated chloroplast was not found to
affect stromal pH in two different studies (5, 19). However,
stromal [ATP] in the presence of the inhibitors was appreciable, 30% in one study (19) and in the other (5), 70% of the
content found under control conditions. Possibly, envelope
ATPase activity was not much affected by this reduction in
ATP level. The finding (as shown in Table I) that several
chloroplast envelope ATPase inhibitors inhibit photosynthesis
and that this inhibition is reversed by treatments (lidocaine)
that increase stromal pH (17) suggests an important physiological role for the oligomycin- and digitoxin-sensitive ATPase. In the presence of Mg2+ and digitoxin, photosynthetic
stimulation due to lidocaine addition averaged 26% in the
three experiments shown in Table I. Lidocaine stimulated
photosynthesis by an average 46% in the presence of Mg2e
2
E
+
4
400
10
AA
O~~~~~~~
0
LI)
Plant Physiol. Vol. 97, 1991
WU AND BERKOWITZ
1 554
Control
/ ~ I--,'00-0
A-A Lidocaine
5020en
0
i
5
10
15
20
25
30
35
40
EXTERNAL K+(mM)
E
+
-J
0
co
H
0
1i0
20
30
40
50
60
70
80
90
EXTERNAL K+(mM)
Figure 1. Stromal K+ at varying external K+. Measurements were
made after 15 min illumination in photosynthetic reaction medium. All
measurements are presented as means ± SE (n = 4); in many cases,
SE bars are covered by the data point. A, Effect of the presence or
absence of 100 Mm lidocaine on stromal K+. Mg2+ (5 mM) and 10 Mm
digitoxin were added to the reaction medium in both cases. B, Effect
of the presence or absence of digitoxin (10 AM) on stromal K+
concentration.
and oligomycin in these experiments. Because lidocaine is
known to be a K+ channel blocker (2), the photosynthetic
stimulation caused by this anesthetic (Table I) suggests that
preventing K+ loss from the stroma by restricting flux through
an ion channel may greatly affect photosynthesis. Further
studies addressed this hypothesis.
Measurement oflidocaine effects on stromal K+ did suggest
that at least one K+ transport site across the chloroplast
envelope may be a K+ channel. Previous work has shown that
100 mM K+ or above is required in incubation media to
prevent K+ loss from isolated chloroplasts and to maintain
high rates of photosynthesis (3, 18). In the presence of Mg2+
and the ATPase inhibitor digitoxin, exposure of chloroplasts
to lidocaine did result in higher stromal K+ concentrations at
a wide range of external K+ (Fig. IA). The effect of lidocaine
on photosynthesis (Table I), then, is likely associated with a
reduction in K+ loss from the stroma (Fig. 1A).
Characterization of Amine Anesthetic Action
Physiological effects of local amine anesthetics are related
to their ability to reach the receptor site of the cation channel,
which resides deep within the channel pore toward the interior
surface of the membrane (9, 22). It is currently theorized that
the receptor site is most readily reached by a compound such
as lidocaine via a hydrophobic route involving diffusion across
the lipid portion of the membrane (10, 22). Characterization
of the interaction of anesthetics with ion channels, therefore,
has been facilitated by comparison of the relative efficacy of
different analogs that vary in lipophilicity (9, 10, 26). Analogs
of lidocaine that are less lipophilic have been found not to
penetrate as well through the membrane to the receptor site;
their pharmacological action is less pronounced (10). The
experimental compounds QX-222 and QX-314 have structures similar to the tertiary amine lidocaine, except they are
quaternary amines and, therefore, have a fixed charge. The
fixed positive charge at the amine nitrogen is surrounded by
ethyl groups in the structure of QX-3 14, making this quaternary amine substantially more lipophilic than QX-222 despite
its cationic property (10). In the structure of the local amine
anesthetic bupivacaine, the 3° amine nitrogen resides in a
piperidine ring and is substituted with a butyl group, making
it substantially more lipophilic than lidocaine (23, 26). Lidocaine hydrochloride is a preparation of the anesthetic that,
when delivered to an aqueous solution at approximately pH
8, delivers a slightly lower level of the pharmacologically active
free base form. The relative lipophilicity of these anesthetic
analogs, from lowest to greatest, is as follows: QX-222 << QX314 < lidocaine HCI < lidocaine << bupivacaine. In a series
of experiments, the relative ability of these anesthetic analogs
to stimulate photosynthesis in the presence of the ATPase
inhibitors was found to be correlated with their relative lipophilicities. A compilation of results form four such experiments is shown in Figure 2. Photosynthesis in the presence of
Mg2' and digitoxin or oligomycin averaged 57% of the control
rate in these experiments. As compared with this inhibited
rate, the anesthetics QX-222, QX-314, lidocaine HCI, lido-
Li)
_)
1
11
63
z
I)
0
0
42
~~~~~~~~~~~~~41
90
4
40- 0 0 !dil u1
8070-
H
W
z
0
bLJ
H
60
50- Z
bJ
CONTROL
7;
ui
w
uw
0~~~~~~~~~N
x
a-4 x
~~~m
IN THE PRESENCE OF ATPase INHIBITORS
Figure 2. Reversal of ATPase inhibitor effects on photosynthesis by
amine anesthetic analogs. These data are a compilation of four
experiments, two with digitoxin and two with oligomycin. Bars, photosynthesis under various treatments compared with control rate,
which was measured in standard photosynthetic reaction medium
with no addition. Numbers above bars, percentage of stimulation due
to anesthetic compared with rate in the presence of ATP inhibitors
alone. The control photosynthetic rate for the four expeiernets was
98, 133, 84, and 63 Mmol 02/mg Chl/h.
INHIBITOR INTERACTION WITH THE CHLOROPLAST ENVELOPE
caine, and bupivacaine stimulated photosynthesis by 0, 41,
42, 48, and 63%, respectively. In the absence of the acidifying
treatments (Mg2+, digitoxin, oligomycin), the anesthetics had
no significant effect on photosynthesis in these experiments
(data not shown).
Direct effects of these anesthetic analogs on the putative
chloroplast envelope ion channel were not monitored in these
experiments. The effects these compounds have on photosynthesis as shown in Figure 2 likely result from their effects on
stromal K+ currents as noted in Figure IA. The differential
effects these local amine anesthetics have on photosynthesis
allow for some preliminary speculations concerning the molecular nature ofthe ion channel. In all likelihood, the channel
structure is similar to amine anesthetic- sensitive channels in
animal systems. The drug receptor site is readily accessible
either from within the membrane through the channel walls
or from the aqueous phase on the interior side of the membrane; in either case the pathway involves movement through
a hydrophobic phase. The analog QX-222 had no effect on
the chloroplast, supporting this notion. Bupivacaine was
found to work as well or better than lidocaine. The 10-fold
increase in the lipid solubility of bupivacaine over lidocaine
(23) further supports the notion of a lipophilic pathway for
access to the receptor site in the channel. All the anesthetics
that did affect the chloroplast in these studies (QX-314, lidocaine, bupivacaine) have a common dimethyl-benzene ring
in the lipophilic portion of the molecule away from the
substituted amine. This ring structure is thought to afford
tight binding to a hydrophobic component within the channel
pore (9).
Preliminary Characterization of the Digitoxin Receptor
We speculate that digitoxin inhibits photosynthesis (Table
I) because of inhibition of an envelope ATPase. However, the
possibility remains that the site of digitoxin inhibition is the
thylakoid ATPase. This issue was addressed by comparing the
effects of digitoxin and oligomycin, along with a known
inhibitor of coupling factor activity (N,N'-dicyclohexylcarbodiimide; 13) on ATPase activity of purified thylakoid membranes using the ATPase assay decribed by LeBel et al. (12)
and light-activating the thylakoid preparation as described by
Hargarter and Ort (7). Whereas N,N'-dicyclohexylcarbodiimide substantially inhibited ATPase activity of these preparations, digitoxin and oligomycin had no effect (data not
shown), supporting the contention that the effects these inhibitors have on photosynthesis were mediated by an inhibition
of the chloroplast envelope ATPase.
Digitoxin (along with other compounds such as ouabain
and digoxin which make up a class of drugs known as cardiac
glycosides) is known to specifically inhibit only the Na+/K+
ATPase, which is a ubiquitous component of all known
animal cell plasma membranes (21). The effect this inhibitor
has on photosynthesis (Table I) raises the intriguing possibility
that its receptor site on the chloroplast envelope is an Na+/
K+ ATPase. Therefore, a series of experiments were undertaken to further characterize the nature of the interaction of
this inhibitor with the chloroplast.
It is clearly documented that the digitoxin-binding site on
the Na+/K+ ATPase in animal membranes in completely
1 555
accessible from the external solution (21). It is for this reason
that, in studies of cardiac glycoside effects on this ATPase,
ouabain is typically used in place of digitoxin (24). Ouabain
is highly water soluble (lipid insoluble), and digitoxin is highly
lipid soluble (24). This extreme difference in lipid solubility
does not affect the relative efficacy of these cardiac glycosides;
this finding has been used to support the contention that the
inhibitor-binding site on this ATPase is on the outer surface
of animal cell membranes (24).
Cardiac glycoside interaction with the chloroplast envelope
ATPase may be qualitatively different. Whereas digitoxin
routinely inhibits photosynthesis of isolated chloroplasts
(Table I), ouabain at concentrations up to 10-fold the effective
concentration of digitoxin had no effect on chloroplast photosynthesis (Table II). These data suggest that the receptor site
on the chloroplast envelope ATPase is not accessible to ouabain but is reached by digitoxin, which can readily diffuse
within the interior of the chloroplast envelope membrane.
Further studies were undertaken to examine digitoxin binding to a receptor site on the chloroplast envelope. Increasing
concentrations of digitoxin resulted in an increasing extent of
[3H]digitoxin associated with the chloroplast (Fig. 3A). Significantly, the presence of an excess of unlabeled ouabain had
no effect on the extent of [3H]digitoxin bound to the chloroplast (Fig. 3A). These data are consistent with the hypothesis
that the cardiac glycoside receptor site on the chloroplast
envelope ATPase is not accessible to the nonlipophilic compound ouabain.
Despite the finding of substantial effects of digitoxin on
chloroplast photosynthesis (e.g. Table I), no data have been
presented to demonstrate that digitoxin association with the
chloroplast is a function of binding of this ligand to a specific
receptor site of a protein on the chloroplast envelope. Especially because of the lipophilic nature of this inhibitor, the
possibility exists that digitoxin association with the chloroplast
could be nonspecific; the compound could simply diffuse into
the envelope interior and be associated with the chloroplast
in a nonspecific manner.
Attempts were made
to demonstrate specific ligand to
receptor site binding of this ATPase inhibitor with the chlo-
roplast using Scatchard Plot analyses which allow for identification of binding to a specific receptor despite the occurrence
of a large excess of nonspecific binding (1, 16). These experiments involve the quantification of [3H]digitoxin binding to
the chloroplast in the presence and absence of additions of
Table II. Comparison of Cardiac Glycoside Effects on
Photosynthesis in the Presence of 5 mM Mg2+
Cardiac glycoside concentrations are in parentheses.
Treatment
Control
Mg2+
Photosynthesis
MmoI 02/mg ChIl/h
114
67
42
% Inhibitiona
0
41
63
Mg2+, digitoxin (6.5 Mm)
Mg2+, ouabain (6.5 MM)
68
40
67
41
Mg2+, ouabain (65 Mm)
a Rate under the various treatments is compared with the control.
_
la.)0
.51
4-,um diameter, these experiments yield a digitoxin receptor
T~~~~~~~~~~~
site (i.e. possible envelope ATPase) concentration of 5.2 x
L
A
IO' binding sites/cm2 of the chloroplast inner envelope.
9
E
.6,
A-A Digitoxin and 1mM ouabaln
0-0 Digttoxin only
X
z
A
.4
0
0
O~~~~/
0~~~00
.2
0
z
0
cx
0
m
0 1
3
5
7
9
EXTERNAL DIGITOXIN (/M)
0
I-
1)E
0
E
Plant Physiol. Vol. 97, 1991
WU AND BERKOWITZ
1 556
30t
B
25
The data presented in Figures 3 and 4 indicate that a
chloroplast ATPase other than the coupling factor likely has
a receptor site that binds specifically to digitoxin. Although
the issue is still not resolved, at least some previous studies
have shown that chloroplast envelope ATPase activity contributes to the maintenance of high stromal pH and optimum
photosynthetic capacity (14, 17). The experiments in this
study that suggest that this ATPase has a specific binding site
for digitoxin lead to the intriguing speculation that the physiological effects of this envelope ATPase on chloroplast metabolism may be mediated by its action as a K+ pump. Data
shown in Figure lB support this contention. At a range of
external K+ concentrations, exposure of chloroplasts to digitoxin resulted in a decrease in stromal K+. Maury et al. (14)
previously noted a reduction in K+ uptake due to treatment
of intact chloroplasts with oligomycin. This body of work,
then, suggests that the activity of a K'-pump ATPase in the
20
0
CL
0
0
z
15
A
9,
10.
50
Slop. (-1/MKm)
Km-12.30MM
L.LILL .002
0
-
-0.0027
C]
0
m
0
0
60
120
180
240
300
[3H]-DIGITOXIN IN MEDIUM (nM)
Figure 3. Measurement of bound [3H]digitoxin in the presence and
absence of an excess of unlabeled ouabain (A) or unlabeled digitoxin
(B). Binding in the presence of only labeled digitoxin (0) is compared
with binding in the presence of 1 mm ouabain (A) in A. In B, binding
of labeled digitoxin in the absence (0) and presence (E) of 10 AM
added unlabeled digitoxin was measured. The difference (V) is the
calculated specific binding (1). It should be noted that, although the
maximum ouabain concentration can be made relatively high (i.e. at
1 mM), digitoxin at 10 ,uM approaches the limit of solubility in aqueous
solutions. The experiment shown in B was repeated a total of three
times; results in each case were similar to those shown. All data are
shown as means ± SE (n = 4).
z
o .001.
m
--f.
Rt- 1.29
t
i-
2
3
4
CI
BOUND [3H]-DIGITOXIN (pmol/mg Chl)
.002
llJ
LL-
cc
0
D
.001
0
unlabeled digitoxin; the specific binding component of total
[3H]digitoxin association with the chloroplast can be deduced
from this analysis as shown in Figure 3B. In the Scatchard
Plot analyses shown in Figure 4, a high affinity specific
receptor site for digitoxin was revealed. The average Km for
the two experiments shown of 12.68 nm is similar to the Km
for cardiac glycoside binding to the Na+/K+ ATPase in animal
cell membranes, which ranges between 2 and 23 nM, depending on the particular mix of cations present during the assay
(21). Scatchard Plot analysis also allows for estimation of the
concentration of receptor sites for the ligand (Rt in Fig. 4).
The average value of the two experiments shown; 1.30 x 10-3
nmol/mg Chl, is equivalent to 7.83 x 10" receptor sites/mg
Chl. Assuming a total chloroplast stromal volume of 20 ,L/
mg Chl and individual chloroplast shape as a sphere with a
0
1
2
3
4
5
BOUND [3H]-DIGITOXIN (pmol/mg Chi)
Figure 4. Scatchard Plot analysis of specific binding of digitoxin to a
receptor site on a protein of the envelope. Data shown in A are
recalculated from the experiment shown in Figure 3B. Data in B are
from a second experiment. The slope of the linear regression (i.e. the
broken line) represents the negative reciprocal of the Km for ligand to
receptor interaction. Abscissa intercept of the regression (labeled Rt)
yields the total receptor number/mg Chi. Each data value was calculated from binding data with four replications per concentration and
are shown as means ± SE.
INHIBITOR INTERACTION WITH THE CHLOROPLAST ENVELOPE
envelope may be important for the normal metabolic function
of the chloroplast. No information is presented in this report,
however, to delineate how ATPase-driven K+ fluxes are linked
to H+ counterflow. Digitoxin-sensitive K+ uptake, as shown
in Figure 1B, could only be indirectly linked to H+ counterflow. This possibility could explain why some researchers
have concluded that envelope ATPase activity is not essential
to maintain high stromal pH. With the isolated chloroplast in
vitro, high stromal pH may be dependent on K+ uptake only
under certain incubation conditions as discussed previously
(14, 17, 20). It should also be pointed out that no data have
been presented here to delineate whether or not the oligomycin- and digitoxin-sensitive chloroplast envelope ATPases
are the same proteins. Because these inhibitors both cause
photosynthetic inhibition that is linked to stromal acidification (17) and reduction of stromal K+ (Fig. 7, also see ref.
14), this possibility appears plausible.
of the 4th International Congress on Photosynthesis. Balaban
International Science Services, Philadelphia, pp 591-598
6. Gimmler H, Schafer G, Heber U (1974) Low permeability of the
chloroplast envelope towards cations. In M Avron, ed, Proceedings of the 3rd International Congress on Photosynthesis.
7.
8.
9.
10.
11.
CONCLUSION
Data presented in this report has focused on substantiating
the assertion that local amine anesthetics, such as lidocaine,
and the cardiac glycoside, digitoxin, affect chloroplast metabolism because of interaction with chloroplast envelope proteins. Documentation of these interactions is significant for
several reasons. K+ movements across the chloroplast envelope greatly affect photosynthetic capacity due to concomitant
H+ counterfluxes (3, 11, 14, 17). The use of these inhibitors
as probes in this study has allowed the identification of a K+
channel and a K+ pump (which could be an Na+/K+ or a K+/
H+ pump) ATPase as possibly contributing to the regulation
of K+ exchange across the envelope. These proteins may play
an important function in maintaining high stromal pH in
the light and, hence, allowing for maximal photosynthetic
activity.
Characterization of the interaction of these inhibitors with
chloroplast envelope proteins is also significant in that an
extensive body of medical literature exists regarding the nature
of both lidocaine and digitoxin interaction with membrane
proteins. This information should allow for further insights
to be made regarding both the chloroplast envelope ATPase
and the K+ channel.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
LITERATURE CITED
1. Beck JS, Goren HJ (1983) Simulation of association curves and
'Scatchard' plots ofbinding reactions where ligand and receptor
are degraded or internalized. J Receptor Res 3: 561-577
2. Dawson DC (1987) Properties of epithelial potassium channels.
In F Bronner, A Kleinzeller, G Giebisch, eds, Potassium Transport: Physiology and Pathophysiology. Academic Press, New
York, pp 41-71
3. Demmig B, Gimmier H (1983) Properties of the isolated intact
chloroplast at cytoplasmic K+ concentrations. I. Light-induced
cation uptake into intact chloroplasts is driven by an electrical
potential difference. Plant Physiol 73: 169-174
4. Douce R, Joyard J (1990) Biochemistry and function of the
plastid envelope. Annu Rev Cell Biol 6: 173-216
5. Enser U, Heber U (1981) Maintenance of a pH gradient across
the chloroplast envelope. In G Akoyunoglou, ed, Proceedings
1 557
22.
23.
24.
25.
26.
Elsevier Scientific Publishing Co., Amsterdam, The Netherlands, pp 138 1-1392
Hargarter R, Ort DR (1986) The relationship between lightinduced increases in the H+ conductivity of thylakoid membranes and activity of the coupling factor. Eur J Biochem 158:
7-12
Heldt HW, Werdan KW, Milovancev M, Geller G (1973) Alkalization of the chloroplast stroma caused by light-dependent
proton flux into the thylaloid space. Biochim Biophys Acta
314: 224-241
Hille B (1977) Local anesthetics: hydrophilic and hydrophobic
pathways for the drug-receptor reaction. J Gen Physiol 69:
497-515
Hille B (1984) Local anesthetics give use-dependent block. In B
Hille, ed, Ionic Channels of Excitable Membranes. Sinauer
Associates Inc., Sunderland, MA, pp 285-289
Huber SC, Maury W (1980) Effects of magnesium on intact
chloroplasts. I. Evidence for activation of (sodium) potassium/
proton exchange across the chloroplast envelope. Plant Physiol
65: 350-354
LeBel D, Poirier GG, Beaudoin AR (1978) A convenient method
for the ATPase assay. Anal Biochem 85: 86-89
Linnett PE, Beechey RB (1979) Inhibitors of the ATP synthetase
system. Methods Enzymol 55: 472-518
Maury WJ, Huber SC, Moreland DE (1981) Effects of magnesium on intact chloroplasts. II. Caton specificity and involvement of the envelope ATPase in (sodium) potassium/proton
exchange across the envelope. Plant Physiol 68: 1257-1263
McCarty DR, Keegstra K, Selman BR (1984) Characterization
and localization of the ATPase associated with pea chloroplast
envelope membranes. Plant Physiol 76: 584-588
Neuburger M, Joyard J, Douce R (1977) Strong binding of
cytochrome c on the envelope of spinach chloroplasts. Plant
Physiol 59: 1178-1181
Peters JS, Berkowitz GA (1991) Studies on the system regulating
proton movement across the chloroplast envelope. Plant Physiol 95: 1229-1236
Robinson SP (1986) Improved rates of C02-fixation by intact
chloroplasts isolated in media with KCl as the osmoticum.
Photosynth Res 10: 93-100
Robinson SP (1985) The involvement of stromal ATP in maintaining the pH gradient across the chloroplast envelope in the
light. Biochim Biophys Acta 806: 187-194
Sen Gupta A, Berkowitz GA (1989) Development and use of
chlorotetracycline fluorescence as a measurement assay of
chloroplast envelope-bound Mg2e. Plant Physiol 89: 753-761
Schwartz A, Whitmer K, Grupp G, Grupp I, Adams RJ, Lee SW
(1982) Mechanism of action of digitalis: is the Na, K-ATPase
the pharmacological receptor? In E Carafoli, A Scarpa, eds,
Transport ATPases. New York Academy of Sciences, New
York, pp 253-271
Strichartz GR, Ritchie JM (1987) The action of local anesthetics
on ion channels of exitable tissues. In GR Strichartz, ed, Local
Anesthetics, Handbook of Experimental Pharmacology, Vol
81. Springer Verlag, New York, pp 21-52
Tucker GT, Mather LE (1979) Clinical pharmacokinetics of local
anesthetics. Clin Pharmacokinet 4: 241-248
Wallick ET, Schwartz A (1988) Interaction of cardiac glycosides
with Na+,K+-ATPase. Methods Enzymol 156: 201-213
Werdan K, Heldt HW, Milovancev M (1975) The role of pH in
the regulation of carbon fixation in the chloroplast stroma.
Studies on CO2 fixation in the light and dark. Biochim Biophys
Acta 396: 276-192
Wheeler DM, Bradley EL, Woods WT (1988) The electrophysiologic actions of lidocaine and bupivacaine in the isolated,
perfused canine heart. Anesthesiology 68: 201-212