Reactive oxygen species (ROS) in photosynthetic
plant cells under light are mainly formed in chloroplasts
where in an oxygen?containing atmosphere O2molecules
are reduced to superoxide anion radicals during the trans?
fer of electrons through the photosynthetic electron
transport chain (PETC) . Under physiological condi?
tions this process (the Mehler reaction) can play a signif?
icant role and amount up to 30% of the overall electron
flow in the PETC . Superoxide is transformed to H2O2
as a result of dismutation or reduction, for example, by
ferredoxin in chloroplast stroma, and by reduction by
plastohydroquinone in thylakoid membrane .
Hydrogen peroxide inhibits the Calvin cycle even in low
concentrations. As chloroplasts have no catalase, the level
of H2O2is supposed to be regulated by membrane?bound
and soluble ascorbate peroxidases . ROS can not only
inhibit enzymes and destroy biomolecules, but they also
may have signaling functions. H2O2is viewed as the most
universal signaling molecule .
One of the unsolved and crucial questions in analyz?
ing the mechanisms of intracellular signaling involving
H2O2remains the ascertainment of transfer of these mol?
ISSN 0006?2979, Biochemistry (Moscow), 2012, Vol. 77, No. 2, pp. 143?151. © Pleiades Publishing, Ltd., 2012.
Original Russian Text © I. A. Naydov, M. M. Mubarakshina, B. N. Ivanov, 2012, published in Biokhimiya, 2012, Vol. 77, No. 2, pp. 179?189.
Originally published in Biochemistry (Moscow) On?Line Papers in Press, as Manuscript BM11?236, January 8, 2012.
Abbreviations: BSA, bovine serum albumin; DCF, dichloroflu?
orescein; DCMU, 3?(3,4?dichlorophenyl)?1,1?dimethylurea
(diuron); FDA, fluorescein diacetate; H2DCF?DA, dihydro?
dichlorofluorescein diacetate; PETC, photosynthetic electron
transport chain; PS I(II), photosystem I(II); ROS, reactive
* To whom correspondence should be addressed.
Formation Kinetics and H2O2Distribution
in Chloroplasts and Protoplasts of Photosynthetic Leaf Cells
of Higher Plants under Illumination
I. A. Naydov, M. M. Mubarakshina, and B. N. Ivanov*
Institute of Basic Biological Problems, Russian Academy of Sciences,
142290 Pushchino, Moscow Region, Russia; E?mail: email@example.com
Received August 16, 2011
Revision received September 6, 2011
Abstract—The dye H2DCF?DA, which forms the fluorescent molecule DCF in the reaction with hydrogen peroxide, H2O2,
was used to study light?induced H2O2production in isolated intact chloroplasts and in protoplasts of mesophyll cells of
Arabidopsis, pea, and maize. A technique to follow the kinetics of light?induced H2O2production in the photosynthesizing
cells using this dye has been developed. Distribution of DCF fluorescence in these cells in the light has been investigated. It
was found that for the first minutes of illumination the intensity of DCF fluorescence increases linearly after a small lag both
in isolated chloroplasts and in chloroplasts inside protoplast. In protoplasts of Arabidopsis mutant vtc2?2 with disturbed
biosynthesis of ascorbate, the rate of increase in DCF fluorescence intensity in chloroplasts was considerably higher than in
protoplasts of the wild type plant. Illumination of protoplasts also led to an increase in DCF fluorescence intensity in mito?
chondria. Intensity of DCF fluorescence in chloroplasts increased much more rapidly than in cytoplasm. The cessation of
cytoplasmic movement under illumination lowered the rate of DCF fluorescence intensity increase in chloroplasts and
sharply accelerated it in the cytoplasm. It was revealed that in response to switching off the light, the intensity of fluores?
cence of both DCF and fluorescent dye FDA increases in the cytoplasm in the vicinity of chloroplasts, while it decreases in
the chloroplasts; the opposite changes occur in response to switching on the light again. It was established that these phe?
nomena are connected with proton transport from chloroplasts in the light. In the presence of nigericin, which prevents the
establishment of transmembrane proton gradients, the level of DCF fluorescence in cytoplasm was higher and increased
more rapidly than in the chloroplasts from the very beginning of illumination. These results imply the presence of H2O2
export from chloroplasts to cytoplasm in photosynthesizing cells in the light; the increase in this export falls in the same time
interval as does the cessation of cytoplasmic movement.
Key words: plants, protoplasts, chloroplasts, reactive oxygen species, hydrogen peroxide, confocal microscopy
144NAYDOV et al.
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
ecules from chloroplasts to cytoplasm . Our previous
study  indicated the exit of H2O2molecules from intact
isolated chloroplasts into the incubation medium as
detected using electron paramagnetic resonance (EPR)
spectroscopy. We also obtained the first evidence of their
exit from chloroplasts inside cells. In the present study the
H2DCF?DA (dihydro?dichlorofluorescein diacetate) dye
was used to detect H2O2in cells. In a recent review 
results obtained with this method were concluded to be
adequate, but the authors stressed the need to consider
the conditions of measurement. Upon entry into a cell,
this dye is de?esterified, turning into H2DCF (dihydro?
dichlorofluorescein), which passes membranes with
much more difficulty and thus it remains in cellular com?
partments where it reacts with H2O2, forming the fluores?
cent product DCF (dichlorofluorescein) . Taking into
account the significant level of superoxide dismutase in
chloroplasts and its elevated local concentration (1 mM)
on the stromal surface of thylakoid membrane , we can
exclude possible reaction of H2DCF with superoxide rad?
icals in stroma.
Using H2DCF?DA for detecting H2O2formation in
plants usually provides a static picture of DCF fluores?
cence in cells, fibrils, and whole leaves . In [10, 11]
H2DCF?DA was used for analyzing processes that initiate
and accompany apoptosis in guard cells of stomas in epi?
dermis of pea leaves. In  the CM?H2DCF?DA dye
with similar properties was used. H2O2was added to
preparations of onion epidermis, and it showed a direct
dependence of DCF fluorescence level on H2O2concen?
tration as well as a dependence of rate of DCF fluores?
cence growth in epidermis cells and whole leaves on tem?
perature and light intensity.
Earlier we found a significant difference in changes
in DCF fluorescence under illumination in chloroplasts
inside guard cells of stomas and inside mesophyll cells in
leaves . The former are known to possess specific car?
bon metabolism , and the latter are where the major
part of photosynthesis occurs. Thus, studying the forma?
tion and transport of H2O2in mesophyll leaf cells is nec?
essary for analyzing the mechanisms of regulation of this
The main problem in investigations of H2O2forma?
tion in vivo in experiments with whole leaves or plant tis?
sues with the use of H2DCF?DA is to guarantee dye entry
into cells. Even under vacuum infiltration of leaves the
DCF fluorescence is observed only in cells adjacent to
fibrous bundles and in epidermal lesions. Protoplasts –
the cells without a cellular wall – are a good object for
studying processes in vivo as long as they remain viable.
This study describes the characteristics of using
H2DCF?DA and confocal microscopy for investigating
the dynamics of H2O2production in chloroplasts and
mesophyll cell protoplasts in the light and provides rec?
ommendations for conducting suitable experiments with
these objects. Chloroplasts and cytoplasm of Arabidopsis
mutants with defective ascorbate biosynthesis were found
to accumulate hydrogen peroxide faster than chloroplasts
of wild type plants. The investigation of DCF fluores?
cence distribution in protoplasts of photosynthetic cells
in the light showed the presence of flow of H2O2mole?
cules from chloroplasts to cytoplasm under illumination.
The increase in the rate of H2O2accumulation near
chloroplasts was found to occur simultaneously with the
arrest of cytoplasmic movements.
MATERIALS AND METHODS
Protoplast isolation. The middle part of leaves was cut
into strips 1 mm wide in extraction medium, incubated for
5 min, and then put in 5 ml of medium with enzymes and
incubated for 2 h at a temperature of 28°C under light with
intensity of 80 µE·m–2·sec–1. After transferring to the basal
medium, protoplasts were extracted from them, and the
suspension was filtered through a nylon cloth with a pore
size of 80 µm. The filtrate was centrifuged for 5 min at 30g,
and the pellet was resuspended in a small amount of basal
medium containing 0.5 M sorbitol, 5 mM CaCl2, and
5 mM MES?KOH, pH 5.5. The extraction medium addi?
tionally included 5 mg/ml polyvinylpyrrolidone (to pre?
vent the destructive effect of phenolic compounds), 0.2%
BSA (to prevent proteases from functioning), and
0.25 mM EDTA (to bind heavy metal ions), and the fer?
mentation medium additionally contained 1% cellulose
(Sigma, USA), 0.2% Macerozyme R10 (Serva,
Germany), and 10 mM of sodium ascorbate. The process
of protoplast extraction is stressful for the cell and can lead
to its death. Dead protoplasts preserved an intact outer
membrane, but no cytoplasmic movement was observed,
and the mitochondria were round and gathered in groups
(see “Results and Discussion”). The mentioned additions,
especially ascorbate, allowed a significant quantity of
viable protoplasts to be obtained. The suspension of pro?
toplasts was kept at room temperature to prevent temper?
ature stress. DCMU (3?(3,4?dichlorophenyl)?1,1?
dimethylurea or diuron), nigericin, and valinomycin were
added to the protoplast suspension before taking an
aliquot of suspension for microscopy.
Isolation of intact chloroplasts. Chloroplasts were
isolated from spinach leaves and separated by centrifuging
in a step gradient of 40 and 80% Percoll [15, 16] with
some modifications. The medium with 40% Percoll con?
tained: 3.33 mM EDTA, 1.66 mM MgCl2, 83.3 mM
Hepes (pH 7.6), and 0.55 M sorbitol. The medium
with 80% Percoll contained: 10 mM EDTA, 5 mM
MgCl2, 250 mM Hepes (pH 7.6), and 1.65 M sorbitol.
The chloroplasts concentrated in the lower level were
washed free from Percoll in incubation medium, and the
percentage of intact organelles was estimated .
Suspensions with intact percentage near 100% were used
in the experiments. The medium for suspension and incu?
KINETICS OF H2O2 PRODUCTION IN ILLUMINATED PHOTOSYNTHESIZING CELLS 145
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
bation under the microscope contained 0.4 M sorbitol,
5 mM MgCl2, 20 mM NaCl, and 25 mM HEPES?KOH
Dye administration. Active H2DCF?DA solution was
prepared by dilution of the 20 mM solution in dimethyl?
sulfoxide with incubation medium to the concentration
of 100 µM. A 4?µl sample of protoplast or chloroplast sus?
pension was put on a glass slide, an equal volume of
H2DCF?DA was added, and a coverglass was placed on
top. The thickness of liquid layer under the coverglass
with a side of 18 mm is less than 40 µm, so to prevent
compression and crushing of protoplasts they were put on
the slide between spacers of appropriate thickness. The
protoplasts were viewed in passing green light (light filter
ZS 11) to prevent photoinduced formation of H2O2before
the beginning of observations. In confocal mode the pro?
toplasts were located by their chlorophyll fluorescence. A
minimal number of scan cycles was used so that the first
registered frames showed the “dark” state of the proto?
plasts. MitoTracker Red dye was added to the protoplast
incubation medium to final concentration 0.4 µM 20 min
before the microscopic observations. Hoechst33258 was
added to the protoplast extraction medium at concentra?
tion 10 µg/ml 1 h before the end of incubation and was
washed off during precipitation of the protoplasts and
their resuspension in basic medium.
Observation setup. For this study, we used a Leica
TCS SPE confocal microscope (Germany). To prevent
destructive processes, less than 10% of laser capacity was
used. Both DCF and chlorophyll fluorescence are quite
visible at this level, but the latter does not bleach and the
light does not cause cell death. Further decrease in laser
intensity requires further amplifying of the signal, which
reduces the image quality. Fluorescence was registered by
successively scanning the field of vision by appropriate
lasers. Chlorophyll fluorescence was observed with the fol?
lowing parameters: excitation at 635 nm, registration in
the 640?750?nm range; for DCF fluorescence: excitation
at 488 nm, registration in the 504?543?nm range; for
Hoechst33258 fluorescence: excitation at 405 nm, regis?
tration in the 410?480?nm range; for MitoTracker Red
fluorescence: excitation at 488 nm, registration in the 550?
605?nm range. The object was scanned with a 3?sec inter?
val. Observation time in experiments with protoplasts
without compression reached 10 min. Lasers of the micro?
scope were the source of photosynthesis?initiating light.
Oil immersion lenses were used in this study. They
provided a well?defined picture of fluorescence distribu?
tion in areas close to the lens. During observation of
deeper planes the shape of object becomes distorted along
the vertical axis and fluorescence intensity is decreased. A
water immersion lens does not distort the picture, but sig?
nificantly decreases resolution of the microscope.
Characteristics of water medium observation and cor?
rections for movements of the objects. During the observa?
tion water evaporates from under the cover glass, leading
to a decrease in thickness of the liquid layer between the
glasses. As a result, the chloroplast?containing layer of
protoplast adjacent to the cover glass moves 5?8 µm down
in 5 min, thus going out of the focal plate. To measure
dynamics of DCF fluorescence in a selected chloroplast
plane – so?called XYT series, where X and Y are coordi?
nates on the plane, and T is time – this movement should
be compensated by continuous focus correction either
manually, guided by the picture of chlorophyll fluores?
cence and adjusting the focus with the fine?tuning screw,
or automatically, if the microscope allows it. With manu?
al correction of the focus there is a danger to select differ?
ent focal planes at different stages of observation. We used
the automatic correction of focus, in which the micro?
scope performs several scans of the object before the reg?
istration of each new frame along the vertical axis within
a predetermined range relative to the focal position in the
previous frame, and finds the best focal position for the
next frame or a series of frames in accordance with the
chosen algorithm. With scan interval of 3 sec, 5?10 scans
in the range of 1 µm (0.5 µm above and below the previ?
ous position of the focal plane) are enough to find the best
focus. It should be taken into account that searching for
the best focus by laser scanning means additional illumi?
nation, which leads to H2O2formation.
Movements of chloroplasts inside protoplasts, as well
as movements of protoplasts in the space between glasses,
can influence the detected values when using XYT series.
Dynamics of changes in fluorescence can be registered in
the chloroplast volume – in XYZT series where Z is the
vertical axis. In XYZT experiment series fluorescence is
recorded in the whole layer of chloroplasts, and their
movements do not affect the dynamics of fluorescence
registered. However, in this case the interval between
frames is several times longer and the object moves a
greater distance, which requires increases in the range
and number of scans in search for the best focus, there?
fore decreasing the time resolution of fluorescence
Estimation of mean DCF fluorescence intensity inside
and around chloroplasts. We used a macro for the ImageJ
program to reveal specific dynamics of DCF fluorescence
in chloroplasts. Based on chlorophyll fluorescence pat?
tern for each frame, we created a mask for the area of the
image where the chlorophyll fluorescence is at least 25%
of the maximum value (chloroplasts region). The fluores?
cence values only from pixels present in the mask were
used to calculate the average DCF fluorescence in
chloroplasts in each frame. Values were averaged over the
area of the mask of a single optical slice in the case of XYT
series, or over all Z layers in the case of XYZT series. This
approach eliminates the influence of chloroplast move?
ment and changes in their relative area in the frame dur?
ing the monitoring on the measured fluorescence values.
An additional mask, based on the chloroplasts mask, was
built to estimate the fluorescence around the chloroplas?
146NAYDOV et al.
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
ts; it included the pixels within a short distance from the
border of the chloroplasts mask, and not overlapping with
RESULTS AND DISCUSSION
Objects observed in protoplasts of mesophyll leaf
cells. Areas corresponding to chloroplasts were identified
in the field of view based on chlorophyll fluorescence
(Fig. 1; see color insert). It is seen that the DCF fluores?
cence after 5?min illumination of protoplasts of meso?
phyll cells of Arabidopsis (Fig. 1a′) and maize (Fig. 1b′)
leaves is concentrated in the chloroplasts. A dark field in
the protoplast – area “V” in Fig. 1b′ – is the vacuole,
where living protoplasts never demonstrated DCF fluo?
rescence increase under illumination.
There is no indication in the literature on the nature
of the bright green DCF fluorescence areas within
chloroplasts that are clearly visible in Fig. 1a′.
Comparison of DCF fluorescence intensity in chloro?
plasts of Arabidopsis, kept one or two days under continu?
ous illumination, which promotes the formation of starch
(Fig. 2a; see color insert), or the same time in the dark,
which encourages the use of starch in plant metabolism
(Fig. 2b; see color insert), as well as the absence of such
zones in the chloroplasts of mesophyll cells of maize,
which are not able to form starch (Fig. 1b′), identified
these areas as starch grains. Higher level of DCF fluores?
cence was detected in starch grain?containing chloroplas?
ts due to nonspecific adsorption and less shielding of
DCF from the excitation light by chlorophyll. This
should be considered when comparing DCF fluorescence
in cells with different number and size of starch grains in
chloroplasts, as different levels of fluorescence may not
reflect actual differences in H2O2formation. The concen?
tration of DCF in starch grains can distort significantly
the results of static experiments, when DCF fluorescence
is measured only once after an experimental influence.
Previously, using MitoTracker Red dye, we identified
small organelles that are visible in Fig. 2b between the
chloroplasts as mitochondria . The fact that mito?
chondria can be observed when recording DCF fluores?
cence is because DCF fluorescence is growing in these
organelles when the protoplasts are illuminated, albeit
more weakly than in the chloroplasts. The most likely
explanation for this growth is increased H2O2formation
in the mitochondria as a result of photorespiration .
In protoplasts, which were chosen for observations, the
mitochondria were elongated and moved quickly with
cytoplasmic flow. The movement of the cytoplasm, cyclo?
sis, is an important characteristic of intact plant cells. In
protoplasts that were isolated from leaves with the pre?
cautions described in Methods we observed, using dye
Hoechst33258 that binds to DNA, a compact nucleus
with clear boundaries characteristic of living cells (not
Dynamic of DCF fluorescence in chloroplasts under
illumination. Isolated intact chloroplasts as well as chloro?
plasts inside protoplasts exhibit nearly linear increase in
DCF fluorescence after illumination, following a small
lag period (Fig. 3). Due to fast destruction of chloroplas?
ts in the microscopic field of view, only a few chloroplas?
ts exhibited DCF fluorescence for two minutes, as shown
Fig. 3. a) Change in DCF fluorescence under light in isolated intact spinach chloroplasts (curves 1?4 correspond to chloroplasts that are marked
in the inset). b) Change in DCF fluorescence under light in chloroplasts inside the protoplast (curves 1?5 correspond to the chloroplasts that
are marked in the inset) and DCF fluorescence changes, averaged over all chloroplasts of the protoplast (dashed curve). Protoplasts were in a
medium with pH 5.5. Insets show images after 2 min (a) and 10 min (b) of illumination, the brighter regions corresponding to higher DCF flu?
orescence intensity; the scale is 10 µm. The moment of switching on the illumination corresponds to the beginning of the horizontal axis.
Fluorescence, arbitrary units
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
Fig. 1. (I. A. Naydov et al.) Fluorescence of chlorophyll (left, in
red) and DCF (right, in green) in protoplasts after 5 min of illu?
mination. a) Arabidopsis; b) corn. V, vacuole. The scale is 10 µm.
Fig. 4. (I. A. Naydov et al.) DCF fluorescence (green) in
Arabidopsis protoplasts, wild?type (a) and mutant vtc2?2 (b), after
5 min of illumination.
Fig. 5. (I. A. Naydov et al.) DCF fluorescence in Arabidopsis pro?
toplasts 30 sec after turning off a 5?min illumination (a) and after
15 sec of re?illumination (b).
Fig. 2. (I. A. Naydov et al.) Identification of bright green areas
inside chloroplasts as starch grains. DCF fluorescence (green)
after 5 min of illumination of protoplasts from leaves of
Arabidopsis plants that were under continuous illumination for 2
days (a) or 2 days in the dark (b). The scale is 10 µm.
Fig. 6. (I. A. Naydov et al.) FDA fluorescence (green) at the
beginning of illumination of Arabidopsis protoplasts (a, b) and
after 15 sec of illumination (a′, b′) in the absence (a, a′) and in the
presence of 1 µM nigericin (b, b′).
KINETICS OF H2O2 PRODUCTION IN ILLUMINATED PHOTOSYNTHESIZING CELLS147
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
in Fig. 3a. Chloroplast destruction is visible as complete
vanishing of DCF fluorescence in separate chloroplasts
between two scans (3 sec). In spots where DCF fluores?
cence disappeared, a compact zone of chlorophyll fluo?
rescence remained for a long time (not shown). Discrete
disappearance of DCF fluorescence indicates breaching
of the chloroplast membrane and dye exit into the medi?
um where in becomes diluted. When glutathione cycle
components get washed away, the amount of accumulat?
ed hydrogen peroxide should increase, but it cannot be
observed because of dilution. Fast destruction of isolated
intact chloroplasts limits their use for analyzing H2O2
production in these organelles.
Kinetics of DCF fluorescence increase in chloro?
plasts inside protoplasts (Fig. 3b) in the beginning of illu?
mination is similar to that in isolated chloroplasts. It was
practically the same when protoplasts were put into medi?
um with pH 5.5 (Fig. 3b), which imitates the conditions
of apoplast that surrounds cells in natural conditions, as
well as in medium with pH 7.6, which was used in exper?
iments with intact chloroplasts (not shown). When
chloroplasts were exposed to medium with pH 7.6 due to
breaching of protoplasts, DCF fluorescence continued in
chloroplasts for almost 5 min. Breaching was discovered
only when distinct borders of protoplasts disappeared. In
the medium with pH 5.5, complete disappearance of
DCF fluorescence inside chloroplasts after breaching of
the plasma membrane occurred in less than 30 sec. This
allowed keeping track of protoplast intactness during
We showed earlier that under illumination in the
presence of DCMU, an inhibitor of photosynthetic elec?
tron transport, no DCF fluorescence increase occurred
under the conditions used, neither in isolated chloroplas?
ts, nor in chloroplasts and mitochondria inside protoplas?
ts . This indicated not only a dependence of observed
H2O2formation on photosynthetic electron transport, but
also an absence of ROS generation as a result of photody?
namic processes, which occurs in the light in the presence
of dyes. Deceleration of increase in DCF fluorescence in
chloroplasts after 5?6 min of illumination (Fig. 3b) was
not a result of dye depletion, as indicated by a linear
increase in fluorescence after a two?fold increase in light
intensity (not shown). This decrease could be due to
acceleration of exit of H2O2molecules from chloroplasts
as they accumulate in the chloroplast stroma.
Effect of ascorbate content in plant on DCF fluores?
cence under illumination of photosynthetic cell protoplasts.
The kinetics of DCF fluorescence increase in chloroplas?
ts under illumination reflects the rate of accumulation of
H2O2molecules, reacting with H2DCF, which competes
with other reactions of these molecules. The H2O2con?
tent in chloroplasts in vivo, according to established con?
ceptions, is kept at a very low level by the functioning of
soluble and membrane forms of ascorbate?peroxidase .
The activity of these peroxidases is high, and the efficien?
cy of H2O2removal, if current theories are correct, should
depend primarily on the amount of available ascorbate
. Therefore, the growth rate of DCF fluorescence
should depend not only on the rate of H2O2production,
but also on the degree of competition of ascorbate and
H2DCF for H2O2molecules. Indeed, the accumulation of
DCF was much higher in the chloroplasts of protoplasts
extracted from Arabidopsis mutant vtc2?2 leaves with a
low content of ascorbic acid in the cells (10% of its con?
tent in wild?type cells)  than in wild type after the
same period of illumination (Fig. 4; see color insert). In
previous work it was found that the addition of ascorbate
to protoplasts suspension of this mutant decreases the rate
of DCF fluorescence in their chloroplasts, becoming the
same as that of wild?type . These results demonstrate
for the first time by in vivo observations the role of ascor?
bate as primary antioxidant in chloroplasts and suggest
that rate and level of H2O2accumulation in these
organelles in its absence may increase significantly. In
protoplasts from vtc2?3 mutant leaves, which contain up
to 30% of the amount of ascorbate in wild type, the
growth rate of DCF fluorescence in chloroplasts was not
significantly different from that observed in wild type for
5 min (data not shown). This fact shows that even this
content of ascorbic acid is sufficient during the observa?
tion period to maintain steady?state concentration of
H2O2at low levels, and supports the view that chloroplas?
ts contain an “excess” amount of ascorbate, judging from
the values of Kmreactions in which it participates . It
is obvious that an excess of ascorbate is a necessary safety
margin in case of malfunction of its regeneration system,
taking into account the fact that it can be used in chloro?
plasts for detoxification of not only H2O2, but also other
Changes in DCF fluorescence intensity in chloroplas?
ts and cytoplasm during light–dark and dark–light transi?
tions. Outflow of H2O2from intact spinach chloroplasts
(capable of CO2fixation) into the incubation medium in
the light was shown in . Increase in DCF fluorescence
intensity, which should reflect the accumulation of H2O2,
in the cytoplasm around the chloroplasts was low in illu?
minated Arabidopsis protoplasts (Figs. 1a and 2).
However, DCF fluorescence increased near the chloro?
plasts after extinguishing the light, but decreased in
response to the reintroduction of light, while increasing
in the chloroplasts simultaneously (Fig. 5; see color
The observed phenomenon can influence the inter?
pretation of observed changes in DCF fluorescence in
photosynthetic cells under illumination. Increase in DCF
fluorescence in chloroplasts after re?introduction of light
might be connected with the beginning of PETC func?
tioning and additional production of H2O2, but the
decrease of DCF fluorescence near the chloroplasts
allows several explanations. To elucidate the mechanism
of this phenomenon, we used FDA (fluorescein diacetate)
148NAYDOV et al.
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
dye, whose properties are close to those of H2DCF?DA,
but which is fluorescent in its native form, i.e. without
reaction with H2O2. Distribution of its fluorescence in the
cytoplasm without illumination (Fig. 6a; see color insert)
was similar to the distribution of DCF fluorescence in
Fig. 5a. Intensity of FDA fluorescence under illumination
decreased outside chloroplasts and increased in chloro?
plasts (Fig. 6a′; see color insert), as occurred after the re?
introduction of light in the experiments with DCF.
DCMU completely prevented the changes in fluores?
cence of both FDA and DCF in the chloroplasts and
around them, indicating the dependence of this phenom?
enon on electron transport and associated processes.
Valinomycin, discharging the transmembrane electrical
potential difference, did not affect the changes in FDA
fluorescence in the chloroplasts and around them under
illumination (data not shown), but these changes were
not observed in the presence of nigericin, which prevents
the development of transmembrane proton gradient (Fig.
6, b and b′; see color insert).
The intensity of FDA fluorescence depends on pH,
and FDA has been used, for example, to estimate pH
changes in roots and near?root environment .
Fluorescence emerged from monoanion (pK 5.0) and
dianion (pK 6.4) forms of this dye (the dianion form has
a higher quantum yield), but not from the fully protonat?
ed form. Consequently, a significant change in the fluo?
rescence level due to changes in its quantum yield may
occur at pH below 7.7, where the fraction of dianion form
(95% at this pH) decreases rapidly. The pK value of DCF
is lower than that for FDA, and its fluorescence depends
strongly on pH at values below 7.0. The effect of nigericin
on changes in FDA fluorescence inside and around
chloroplasts under illumination suggests that these
changes depend on photoinduced pH fluctuations inside
and outside chloroplasts. It is known that the pH in stro?
ma of chloroplast located in cytoplasm shifted to the
alkaline side under illumination: from 7.3?7.4 to 7.7?7.8
at normal concentration of CO2in the air ([23, 24],
Heber, personal communication). Growth of DCF and
FDA fluorescence in chloroplast stroma under illumina?
tion can be partly explained by pH increase there. This
applies mostly to FDA, because the quantum yield of
DCF fluorescence in this pH range increases very slight?
ly, and DCF fluorescence increase in chloroplasts is
apparently caused by H2O2production in the light. It is
unlikely that this increase is due to return of some DCF
molecules from the cytoplasm (which has pH 7.2?7.4) to
chloroplast stroma. Theoretically, anions of weak acids
accumulate in the compartment with higher pH, but in
this case DCF molecules have to move through the
A significant fluorescence decrease in the layer adja?
cent to chloroplasts could not be the result of dye mole?
cule movement into the chloroplast since, as calculations
show, their concentration in the cytoplasm is almost with?
out change in response to a slight pH change in the
chloroplast stroma, the volume of which is much less than
the volume of cytoplasm. It is more likely that the
decrease in DCF and FDA fluorescence in the vicinity of
the chloroplast under illumination is the result of a signif?
icantly lower pH in this zone. The phenomenon of proton
outflow from intact chloroplasts to the outside medium
has long been known (see review in ), although there
is no conventional view on its mechanism yet. The proton
outflow from the chloroplast may be either the result of
functioning of ATPase, located in the membrane of
chloroplast and using thylakoid?produced ATP to pump
protons from the chloroplast to the cytoplasm , or –
as suggested in  and clearly demonstrated for chloro?
plasts in cytoplasm  – caused by proton diffusion
directly from the lumen, where they come from stroma by
electron transfer through the PETC in the light. Thus, pH
lowering under light near the chloroplast masks proton
accumulation in this area, reducing the DCF fluores?
cence quantum yield. When the light is turned off, the pH
increases and DCF fluorescence increases.
When nigericin was introduced in a protoplast sus?
pension under illumination, DCF fluorescence increase
around chloroplasts in response to switching on the light
was faster than inside the chloroplasts, which is evident
from the comparison of the values of fluorescence inten?
sity ratio inside the chloroplasts and in the surrounding
layer of cytoplasm in the absence and in the presence of
nigericin (Fig. 7). In the absence of nigericin the rate of
fluorescence increase in the cytoplasm becomes faster
than inside the chloroplasts after 3?4 min of illumination,
Fig. 7. Effect of 1 µM nigericin on the ratio of DCF fluorescence
intensity in chloroplasts (Fchl) to DCF fluorescence intensity in
the adjacent layer of cytoplasm (Fcyt) during illumination of
Arabidopsis protoplasts. For determination of mean fluorescence
intensity of the dye inside and around the chloroplasts, see
“Materials and Methods”. The moment of switching on the light
corresponds to the beginning of the horizontal axis.
KINETICS OF H2O2 PRODUCTION IN ILLUMINATED PHOTOSYNTHESIZING CELLS149
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
as indicated by the beginning of decrease in this ratio,
which, however, is still greater than 1. In the presence of
nigericin, this trend can be observed from the very first
minutes of illumination. Experiments in the presence of
nigericin showed that increase in DCF fluorescence in
the cytoplasm cannot be the result of the outflow of these
molecules to cytoplasm, but reflects their appearance in
the reaction of H2DCF with H2O2, which leaves the
chloroplasts. In the case when DCF could freely diffuse
through the membrane, the inhibition of pH changes
near the chloroplast in the presence of nigericin (see
above) would have lead to synchronous increase in the
rates of increase in DCF fluorescence in the cytoplasm
The fact that the ratio of fluorescence intensities in
the chloroplast and cytoplasm is less than 1 in the pres?
ence of nigericin and that it decreases during illumination
shows that H2O2molecules come into the cytoplasm
around the chloroplast faster than in the chloroplast stro?
ma. We emphasize that the increase in DCF fluorescence
does reflect not the accumulation of H2O2, but the accu?
mulation of DCF molecules, emerging in reaction of
H2DCF with the appearing molecules of H2O2. It has also
been shown in experiments with intact isolated chloro?
plasts  that H2O2exit from the chloroplast to the
incubation medium is accelerated in the presence of
nigericin. The reason for this is unclear. Perhaps this is
caused by functioning of two ways of hydrogen peroxide
exit from the chloroplast: 1) diffusion of H2O2molecules
from stroma, and 2) the diffusion of H2O2molecules
formed in the thylakoid membrane  through chloro?
plast lumen, as was discussed above for protons.
Stimulation of the second path in the presence of
nigericin would at the same time lead to a decrease in
intake of membrane?formed H2O2molecules into the
stroma, whose detoxification system operates to further
reduce the possibility of their reaction with H2DCF.
Transition from increase to decrease in the ratios of fluo?
rescence intensities in the chloroplast and the cytoplasm
observed in the absence of nigericin may also be caused by
a gradual increase in outflow to the cytoplasm of mem?
brane H2O2molecules specifically, while the speed of
their intake to stroma becomes saturated, which eventu?
ally manifests itself in slower growth of DCF fluorescence
in chloroplasts (Fig. 3b).
Thus, a low DCF fluorescence in cytoplasm of pro?
toplasts in the absence of nigericin is related, first of all,
to effective scavenging of H2O2by chloroplast ascorbate
peroxidase (especially in the first minutes of illumina?
tion), which reduces the total outflow of H2O2from
chloroplasts. Second, lowering pH around chloroplasts
decreases (due to decrease in quantum yield) the fluores?
cence intensity of DCF molecules emerging in this area
in the reaction of H2DCF with H2O2molecules emerging
from the chloroplast. In addition, we cannot exclude
degradation of H2O2molecules in chloroplasts by cyto?
Change in DCF fluorescence distribution in chloro?
plasts and cytoplasm after the arrest of cytoplasmic move?
ment under illumination. It was found in the study of DCF
fluorescence distribution in protoplasts from wild?type
plants that arrest of the movement of cytoplasm coincides
with a sharp acceleration of DCF fluorescence increase in
the cytoplasm near the chloroplasts, and a slowdown in
DCF fluorescence increase in chloroplasts (Fig. 8).
Additionally, besides the fact that mitochondria stop
moving with the flow of cytoplasm, they undergo round?
ing and adhesion, which are the signs of apoptosis .
Then the protoplasts, which showed arrest of cyclosis and
mitochondrial shape changed, can be regarded as cells
with a developing apoptosis process. At the same time,
electron transfer reactions in the chloroplasts of these
protoplasts continue normally.
Considering the ratio of pH values in the cytoplasm
and in chloroplasts in the light (see above), DCF fluores?
cence increase in the cytoplasm, observed after cyclosis
arrest, with simultaneous reduction of its rate of increase
in the chloroplast, is most likely a result of increased out?
flow of H2O2molecules from the chloroplast. This would
also explain the decrease in the rate of DCF fluorescence
increase in the chloroplast. DCF fluorescence increase
near chloroplasts may also be partly caused by increased
production of H2O2in mitochondria, which are located
in the vicinity of chloroplasts in living cells .
Previously, we found that the destruction of the protoplast
leads to rapid growth of DCF fluorescence in mitochon?
dria coupled with a decrease in DCF fluorescence in the
chloroplasts . Studies of the connection between
arrest of cytoplasmic movements and increased H2O2
Fig. 8. Effect of cytoplasm flow arrest (arrow) on the kinetics of
DCF fluorescence in chloroplasts (1) and in the surrounding layer
of cytoplasm (2). The moment of switching on the light corre?
sponds to the beginning of the horizontal axis.
Fluorescence, arbitrary units
150 NAYDOV et al.
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
content in cytoplasm may be important for understanding
mechanisms of signal transmission from chloroplasts to
other systems of cells and to other cells. It is possible that
outflow of H2O2molecules from chloroplasts enhances a
cell death signal. This assumption corresponds to data
obtained in  showing that not only mitochondria, but
also chloroplasts are involved in the development of
Cytoplasmic flow is related to a phenomenon impor?
tant for understanding of intracellular regulation
described in [32, 33], namely, the appearance of a high?
conductivity plasmalemma zone near the illuminated
area of a Chara algal cell, and on the same side where
cytoplasm moves but not on the opposite side . Zones
with altered plasmalemma conductance disappear when
cytoplasmic flow is stopped . It is possible that mole?
cules of H2O2, emerging from the illuminated chloroplas?
ts, are a signal to processes that regulate the opening of
membrane channels to plasmalemma. In this case, stop?
ping the cytoplasm reduces the flow of these molecules,
which affects the properties of plasmalemma.
The results obtained in this study show the possibility
of using H2DCF?DA to investigate the kinetics of H2O2
formation in protoplasts of photosynthetic leaf cells
exposed to light. Light is the main source of energy for
plants, and photosynthetic cells adjust their metabolism to
use this energy optimally. The primary step in such adjust?
ment is H2O2formation in the PETC, which functions not
only as a converter of electromagnetic light energy into
chemical energy, but as a sensor of both the behavior pat?
tern and the very possibility of this transformation.
“Dumping” of electrons to oxygen and H2O2production
may reflect a balance between the arrival of the light ener?
gy and capacities of the PETC, Calvin cycle, and assimi?
late transport system to use this energy for CO2fixation.
The quantitative measurement of H2O2content in
cells in vivo is an almost impossible task , so the study
on the kinetics of its formation can give valuable informa?
tion about the implementation of the signaling function
of these molecules. The use of colored and fluorescent
indicators for various ROS is a technique that seemed to
hold great promise for study of the signaling role of ROS.
However, not only are new opportunities for such studies
found, but also their limitations. For example, nitrotetra?
zolium blue, which could have been applied – and this
was suggested in some papers – as a detector of superox?
ide anion radicals in living cells, is a good indicator of the
ascorbate content ; we have indeed found that a
change in nitrotetrazolium blue color reflects changes in
ascorbate content in cell compartments (unpublished).
Use of H2DCF?DA for registration of H2O2formation
turned out to be more promising. However, even in this
case it was found that there are processes that should be
taken into consideration for correct interpretation of the
observed H2O2formation kinetics. Some of these process?
es, which can be referred to as instrumental, are described
in the “Materials and Methods” section. The other, bio?
logical processes, which obviously not all have been iden?
tified, include, for example, the concentration of DCF in
starch grains, which is characteristic for plant cells, as
well as changes in the fluorescence distribution of dye in
response to pH shift in cell compartments.
With the use of methodological improvements devel?
oped in the course of this study, as well as computer data
processing, we were able to record the phenomenon of
H2O2content increase in the cytoplasm after arrest of its
movements in the light. There is evidence in the literature
that ROS, and in particular H2O2can affect, either direct?
ly or through redox?sensitive enzymes, actin cytoskeleton
system , and that depolymerization of cytoskeleton
microtubules under the effect of H2O2initiates the
expression of some genes of protection against pathogens
. We found a phenomenon consistent with the idea
that H2O2specifically is the chemical basis of the primary
signal – which is able to transform into cell signals of
other nature – that is sent by photosynthetic cells to other
cells using the PETC, its most powerful energy converter.
1. Ivanov, B., and Khorobrykh, S. (2003) Antioxid. Redox
Signal., 5, 43?53.
2. Kuvykin, I. V., Vershubskii, A. V., Ptushenko, V. V., and
Tikhonov, A. N. (2008) Biochemistry (Moscow), 73, 1063?
3. Mubarakshina, M. M., and Ivanov, B. N. (2010) Physiol.
Plant., 140, 103?110.
4. Asada, K. (1999) Annu. Rev. Plant Physiol. Plant Mol. Biol.,
5. Foyer, C., and Noctor, G. (2009) Antioxid. Redox Signal.,
6. Mubarakshina, M. M., Ivanov, B. N., Naydov, I. A.,
Hillier, W., Badger, M. R., and Krieger?Liszkay, A. (2010)
J. Exp. Bot., 61, 3577?3587.
7. Swanson, S., Choi, W.?G., Chanoca, A., and Gilroy, S.
(2011) Ann. Rev. Plant Biol., 62, 273?297.
8. LeBel, C. P., Ischiropoulos, H., and Bondy, S. C. (1992)
Chem. Res. Toxicol., 5, 227?231.
9. Rodriguez, A. A., Grunberg, K. A., and Taleisnik, E. L.
(2002) Plant Physiol., 129, 1627?1632.
10. Samuilov, V. D., Kiselevsky, D. B., Sinitsyn, S. V., Shestak,
A. A., Lagunova, E. M., and Nesov, A. V. (2006)
Biochemistry (Moscow), 71, 384?394.
11. Samuilov, V. D., Kiselevsky, D. B., Shestak, A. A., Nesov,
A. V., and Vasil’ev, L. A. (2008) Biochemistry (Moscow), 73,
12. Kristiansen, K. A., Jensen, P. E., Moller, I. M., and
Schulz, A. (2009) Physiol. Plant., 136, 369?383.
13. Naidov, I. A., and Ivanon, B. N. (2008) Proc. Int. Conf.
Organization of Plants”, Yekaterinburg, p. 291.
14. Gotow, K., Taylor, S., and Zeiger, E. (1988) Plant Physiol.,
KINETICS OF H2O2 PRODUCTION IN ILLUMINATED PHOTOSYNTHESIZING CELLS151 Download full-text
BIOCHEMISTRY (Moscow) Vol. 77 No. 2 2012
15. Laasch, H. (1987) Planta, 171, 220?226.
16. Mullet, J. E., and Chua, N. H. (1983) Methods Enzymol.,
17. Heber, U., and Santarius, K. A. (1970) Z. Naturforsch., 25,
18. Naidov, I. A., Mudrik, V. A., and Ivanov, B. N. (2010) RAS
Reports, 432, 834?837.
19. Ivanov, B. N. (2000) Free Radical Res., 33, 217?227.
20. Conklin, P. L., Saracco, S. A., Norris, S. R., and Last, R.
L. (2000) Genetics, 154, 847?856.
21. Ivanov, B. N. (1998) Biochemistry (Moscow), 63, 133?138.
22. Monshausen, G. B., Bibikova, T. N., Weisenseel, M. H.,
and Gilroy, S. (2009) The Plant Cell, 21, 2341?2356.
23. Hauser, M., Eichelmann, H., Heber, U., and Laisk, A.
(1995) Planta, 196, 199?204.
24. Oja, V., Savchenko, G., Jakob, B., and Heber, U. (1999)
Planta, 209, 239?249.
25. Heber, U., and Heldt, H. W. (1981) Ann. Rev. Plant
Physiol., 32, 139?168.
26. Berkowitz, G. A., and Peters, J. S. (1993) Plant Physiol.,
27. Svintitskikh, V. A., Andrianov, V. K., and Bulychev, A. A.
(1985) J. Exp. Bot., 36, 1414?1429.
28. Mubarakshina, M., Khorobrykh, S., and Ivanov, B. (2006)
Biochim. Biophys. Acta, 1757, 1496?1503.
29. Matsuyama, S., Llopis, J., Deveraux, Q. L., Tsien, R. Y.,
and Reed, J. C. (2000) Nature Cell Biol., 2, 318?325.
30. Yoshinaga, K., Arimura, S.?I., Niwa, Y., Tsutsumi, N.,
Uchimiya, H., and Kawai?Yamada, M. (2005) Ann. Bot.,
31. Naidov, I. A., and Ivanov, B. N. (2009) Proc. All?Russ. Conf.
Irkutsk, pp. 310?311.
32. Lucas,W. J., and Dainty,J. (1977) J. Membr. Biol., 32, 75?92.
33. Bulychev, A. A., and Vredenberg, W. J. (2003) Planta, 218,
34. Bulychev, A. A., and Dodonova, S. O. (2011) Fiziol. Rast.,
35. Bulychev, A. A., and Krupenina, N. A. (2009) Plant. Signal.
Behav., 4, 24?31.
36. Queval, G., Hager, J., Gakiere, B., and Noctor, G. (2008)
J. Exp. Bot., 59, 135?146.
37. Dalle?Donne, I., Rossi, R., Milzani, A., Di Simplicio, P.,
and Colombo,R. (2001) Free Rad. Biol. Med., 31, 1624?1632.
38. Lin?Lin Yao, Qun Zhou, Bao?Lei Pli, and Ying?Zhang Li
(2011) Plant Cell Environ., 34, 1586?1598.
to Difficult Environment”,