Advances on photosystem II investigation by measurement of delayed chlorophyll fluorescence by a phosphoroscopic methods.

Abstract

A method for data acquisition based on recording of light signal from a conventional phophoroscope fluorometer with high-speed digitalization is proposed to extract more information from a delayed chlorophyll a fluorescence (DF) signal. During the signal processing from all points registered by the fluorometer, we obtain simultaneously a large number of induction curves of DF decaying at different time ranges. In addition, it is possible to register a series of dark relaxation kinetics of DF, recorded at different moments during the induction period or at different temperatures. This allows the evaluation of the contribution of DF kinetic components during the induction period or at different temperatures and the comparison between DF signals registered with different phophoroscopes. With the phosphoroscope system used in this study, we have shown that the contribution of the millisecond components (with lifetimes 0.6 and 2-4 ms) predominates during the first second of the induction period. After 1 s of illumination, the amplitudes of the 0.6 ms and 2-4 ms components and of the slower one (with lifetime more than 10 ms) become approximately equal. The change in lifetime of the different components during the induction and during gradual heating is also observed. It is shown that all registered DF kinetic components have different temperature dependences.

Full text

Page 1

Photochemistry and Photobiology, 2003, 77(3):292–298
Advances on Photosystem II Investigation by Measurement of Delayed
Chlorophyll Fluorescence by a Phosphoroscopic Method{
Ivelina Zaharieva1and Vassilij Goltsev*2
1Institute of Biophysics, Bulgarian Academy of Sciences, Sofia, Bulgaria and
2Department of Biophysics and Radiobiology, Biological Faculty, University of Sofia, Sofia, Bulgaria
Received 20 August 2002; accepted 5 December 2002
ABSTRACT
A method for data acquisition based on recording of light
signal from a conventional phophoroscope fluorometer with
high-speed digitalization is proposed to extract more in-
formation from a delayed chlorophyll a fluorescence (DF)
signal. During the signal processing from all points registered
by the fluorometer, we obtain simultaneously a large number
of induction curves of DF decaying at different time ranges. In
addition, it is possible to register a series of dark relaxation
kinetics of DF, recorded at different moments during the
induction period or at different temperatures. This allows the
evaluation of the contribution of DF kinetic components
during the induction period or at different temperatures and
the comparison between DF signals registered with different
phophoroscopes. With the phosphoroscope system used in this
study, we have shown that the contribution of the millisecond
components (with lifetimes 0.6 and 2–4 ms) predominates
during the first second of the induction period. After 1 s of
illumination, the amplitudes of the 0.6 ms and 2–4 ms
components and of the slower one (with lifetime more than
10 ms) become approximately equal. The change in lifetime of
the different components during the induction and during
gradual heating is also observed. It is shown that all registered
DF kinetic components have different temperature depend-
ences.
INTRODUCTION
Delayed fluorescence (DF), or the so-called delayed light emission,
was discovered by Strehler and Arnold (1) in 1951. DF of green
plants, algae and photosynthetic bacteria is light emission from
chlorophyll a molecules, predominantly of the photosystem II
(PSII) antenna system. Its spectrum is similar to that of prompt
fluorescence (PF), but it occurs later than PF and can be detected
after the excitation light is turned off (2,3). DF involves an
insignificant loss of energy (0.03%) that could potentially be used
in photosynthesis (4), but it is a sensitive indicator of many
reactions in photosynthesis (5). DF has many components that are
characterized by different decay rates starting from 2–4 ns (6) and
100–200 ns (7) to microseconds (8,9), to milliseconds (10–12) and
to seconds and longer (2,13–16).
The difference between PF and DF is in the origin of the excited
singlet state of the emitting antenna chlorophyll a molecule. The
variable chlorophyll fluorescence is created either as a result of
equilibration between the antenna and the reaction center of PSII or
of exciton radical pair equilibration. Variable (prompt) fluores-
cence is used by hundreds of investigators as a probe for various
aspects of photosynthesis—from excitation energy transfer in
picosecond time scale to CO2fixation in minutes—see Govindjee
(17) for more details.
Earlier, DF emission has been attributed to chlorophyll triplet
recombination (18). According to the radical pair hypothesis, DF
originates from recombination of P6801I2and occurs with a 2–4 ns
lifetime (5). It has been shown that radical pair hypothesis may be
valid for explaining DF emitted in the microsecond time range
(19–21). The origin of DF decaying in the seconds-to-minutes time
scale (at 30 s, 1 min and 2 min after the flash) as well as of
thermoluminescence in leaves is ascribed to S2QB2or S3QB2
recombination (16). The decay kinetics of DF in a time range from
several microseconds to milliseconds after light excitation (decay
range investigated in this study) has been thought to reflect the
recombination between the reduced electron acceptor, QA2, and
the oxidized secondary electron donor, Z1, of PSII (22,23).
DF can be used for investigation of the plant physiological state,
in parallel with other techniques, as done by Srivastava et al. (24).
Strasser (25) used a special technical system with multibranched
fibre optics for simultaneous measurements of several parameters
on the sample (DF, PF, absorption and oxygen evolution). With the
fibre optic system, each measured point originated from the same
geometric constellation, which increases the quality (signal to noise
ratio) of the signal. Although DF could, in principle, bring valuable
information about structure and function of the photosynthetic
apparatus it has been far less investigated than PF. One possible
reason is that DF measurements are technically more difficult to
implement, even in their simplest modes, than PF measurements. It
requires at least some minimal mechanical means to strictly
separate the actinic light from the observation periods in the
measuring process. Furthermore, it is a rather weak signal; thus,
high detection sensitivity and total absence of stray light are
required (26).
{Posted on the website on 21 January 2003.
*To whom correspondence should be addressed at: Department of
Biophysics and Radiobiology, Biological Faculty, University of Sofia,
8, Dragan Tzankov Boulevard, 1164 Sofia, Bulgaria. Fax: 3592-656641;
e-mail: goltsev@biofac.uni-sofia.bg
Abbreviations: DF, delayed fluorescence; PF, prompt fluorescence; PQ,
plastoquinone; PSI, photosystem I; PSII, photosystem II.
? 2003 American Society for Photobiology 0031-8655/01$5.0010.00
292

Page 2

The investigation of DF has been done in two different ways: (1)
time-resolved delayed light emission, when the kinetics of its dark
relaxation are first registered and then an attempt is made for the
correct identification of various decay components with particular
reactions (5), and (2) registration of induction curves (transients) of
DF, in parallel with the induction curves of PF; these reflect the
changes in the photosynthetic machinery during transition from
dark- to light-adapted state (27). Unfortunately, in most publica-
tions, these two measurements are not available simultaneously.
Because the registration of DF is based on the time separation of
DF from PF after illumination, a phosphoroscope method of DF
registration is frequently used for this purpose. The technique can
be described as follows: during the first period of a phosphoroscope
cycle, the actinic light is turned on and PF is measured while the
DF measuring device is off; in the next period, the positions are
reversed, with a short blank period separating the successive light
and dark phases (26). If phosphoroscopes are used for registration
of DF dark relaxation kinetics, it is usually after long illumination,
when DF has reached its stationary level (23), or after a single
flash. This method is often used for registration of DF induction
curves also, and in this case the signal is proportional to the
averaged light signal in each period of detection. Although the
mechanical phosphoroscope allows registration of relatively slow
DF components only (not faster than few tens of microseconds), it
allows simultaneous registration of DF and PF induction curves.
The main disadvantage of the method is that when the induction
curve of DF is registered, an overlapping of the fast and slow decay
components (2,28) occurs and this makes the induction complex.
The overlapping is due to averaging of the signal in the entire
period of DF detection when the DF intensity exponentially
decreases. Besides, a wide variety of phosphoroscopes are
described in literature, each with different durations of the working
cycle and periods of illumination of the sample and registration of
DF. To separate individual components that take part in the total
registered DF, different approaches have been used. The inter-
ference in the measurement of fast component due to overlapping
slow components in the technique with the phosphoroscope in
untreated chloroplasts can be avoided if the measurement is done at
low temperatures (29). At temperatures below 08C, the slow
components decrease and the induction of fast components with
about a 200 ls half decay time resembles that of PSII particles
at room temperature (30). Another approach was suggested by
Schreiber and Schliwa (31), who programmatically removed DF
intensity recorded at the end of each registration period to
minimize the contribution of slower decaying components. Un-
fortunately by this way the problem for overlapping of different
kinetic components is not solved.
The progress in technologies for biological signal processing
and data acquisition has been significant in recent times for the
improvement of photoluminescence methods to extend the current
knowledge of photosynthetic energy transformation. Further
experimental development in the field of DF and processing
methods are expected to allow simultaneous measurement of PF
and different components of DF (with different lifetimes) and the
study of influence of temperature on different DF components (32).
These measurements are expected to provide deeper insight into
the mechanism of DF emission as well as into the photosynthetic
process itself.
The aim of this article is to present a way for signal processing in
the detection of DF, which will extend the informational
capabilities of the method. It allows the evaluation of contribution
of different kinetic DF components during the induction period or
at different temperatures, and it can be applied for other
phosphoroscopes used for DF registration.
MATERIALS AND METHODS
Materials. Arabidopsis thaliana plants were grown for 1 month on soil in
a climatic chamber at controlled conditions of temperature (258C and 208C
during day and night, respectively) and light (luminescent lamp, 2000 lx, 16
h?d21). Before measurements were carried out, plants were kept in darkness
for 1 h and then the detached leaves were transferred to a measuring
chamber.
Measuring equipment. The DF and PF were measured using a Becquerel-
type phosphoroscope fluorometer FL-2006 (manufactured by ‘‘TEST,’’
Krasnojarsk, Russia) connected to a computer. One cycle of light and dark
period was 11 ms in duration: 0–5.15 ms for excitation and 5.15–11 ms for
darkness, as shown in Fig. 1. PF and DF signals were read by a two-channel
analogue–digital converter at approximately every 42 ls during the
corresponding registration period. PF was measured at 0.16 ms of the
working cycle, and DF was measured at 5.5–10.65 ms (0.35 ms after
cessation of excitation). Integral values of PF induction curve points were
obtained as a mean value of the 10th to 20th registered points during the
period of illumination of the sample. Integral values of DF induction curves
were obtained as a mean value from all (about 110) points or from chosen
sequence recorded points during the period of registration (dark period)
(Fig. 2). In each phosphoroscope working cycle, one DF dark relaxation
kinetic (of 100–120 points situated via the time interval of ;42 ls) was
available. It represents a sum of two fast kinetic components (s1; 200–900
ls and s2; 1.2–3.9 ms) and a ‘‘tail’’ of slow components (s3. 20 ms) (33).
Exciting light was obtained from a 75 W halogen tungsten lamp. It
provided maximum intensity of modulated visual light at the sample surface
of 1200 lmol photons?m22?s21. The front sharpness at the beginning of
illumination was 200 ls. Exciting light was perpendicular to the surface of
the sample, and PF and DF quanta were detected at an angle of 458 relative
to the exciting beam. To remove the scattered light from the exciting light,
cross filters were used. A blue–green SZS22 (manufactured by ZOS,
Krasnogorsk, Russia) filter with k < 660 nm was placed between the lamp
and the sample, whereas a far-red KS19 (ZOS, Krasnogorsk, Russia) filter
with k ? 680 nm was placed between the sample and photocathodes. The
PF and DF quanta were registered by two photomultipliers having
multialkaline photocathodes with S-20 spectral sensitivity.
With a new software, the phosphoroscope used in this study can register
simultaneously (1) prompt chlorophyll a fluorescence (PF) induction curves
with time resolution of 11 ms; (2) up to 100 delayed chlorophyll
a fluorescence induction curves (in the time range of 0.35–5.5 ms), also
Figure 1. Schematic representation of phosphoroscope-working diagram.
(A) Time course of actinic light intensities. One cycle of light and dark
period was 11 ms in duration: 0–5.15 ms for excitation (when PF is
detected) and 5.5–10.65 ms for darkness (when DF is detected). There are
two periods (about 0.35 ms each) between them, when the sample is not
illuminated and DF is not measured. (B) PF and DF light intensities; with
straight line over DF decay is presented the size of aperture (Ap) in front of
the DF detector.
Photochemistry and Photobiology, 2003, 77(3)293

Page 3

with a time resolution of 11 ms; and (3) a series of dark relaxation kinetics
of 5.5 ms DF, recorded during specific phosphoroscope working cycles (11
ms, 5.5 ms dark period, about 110 points of DF at ;42 ls). In a second
mode, the fluorometer can also be used to measure thermograms—
simultaneous registration of stationary levels of PF and DF during gradual
heating from 58C to 808C. In the mode of thermoinduced change
measurements, in addition to PF and DF integral values, kinetics of DF
dark relaxation at each point of the temperature curve can also be registered.
RESULTS AND DISCUSSION
The phosphoroscope FL-2006 allows registration during the
induction period of DF, decaying in the 0.35–5.5 ms time range.
Simultaneously the PF induction curve is registered during the
periods of illumination of the sample (Fig. 3). Because the changes
that occur in the photosynthetic apparatus are much faster at the
beginning of the induction period, the DF and PF induction curves
are presented on a logarithmic time scale. In the DF induction
curve different maxima are observed, denoted earlier as I1, I2, ...,
I6(35,36).
The maxima observed in the DF intensity at the first 1 s during
the induction period (denoted as I1 and I2) coincide with the
moment of maximal rate of Fo–Fiand of Fi–Fpincrease in PF, and
their amplitudes reflect the redox transitions in the acceptor side of
PSII (35–37). This fast phase of DF induction also has been
attributed to the changes in the electrical potential (Dw), depending
on the state of the PSII reaction center (36,38,39). The appearance
of the late maxima in DF (I4 and I5) can be related to the
photoinduced proton gradient as well as to the initiation of the
photosynthetic dark reactions (38,40). This phase of DF induction
coincides with P-to-S decay in PF, which has been related to
internal proton concentration changes (41,42). A well-pronounced
deep, D2, is usually observed between the fast (I1–I2) and slow (I4–
I5) phases of the DF induction curves. It may be a result of closure
of the reaction centers (RC) of PSII during the plastoquinone (PQ)
pool reduction (35). The late maxima (I6and I7) possibly result
from activation of the Calvin–Benson cycle, which is initiated by
adenosine triphosphate and reduced nicotinamide adenine di-
nucleotide phosphate accumulation (43). The clarification of the
nature of the maxima observed, as well as of the differences
between the simultaneously registered DF induction curves,
requires further study.
As in other experiments using the phosphroscopic method, in
this DF induction curve, different kinetic components, which have
lifetimes that differ by an order of magnitude, are included. The
contribution of each component changes during the induction
period (23), but the separation of individual components in such
a curve is impossible. One possible way to do this is by the
simultaneous registration of induction curves of DF, decaying in
different subintervals of the dark relaxation period.
Induction kinetics of DF recorded in 50 sequencing intervals
during the period of registration, as well as the averaged DF
induction curve, are shown in Fig. 4. Each interval spans about 85
ls. With the slowdown of the DF components registered, the shape
of the DF induction curve changes. This is evidence for the
different contributions of fast- and slow-decaying DF components
at different times of the induction period. With increasing DF dark
relaxation time, the amplitude of the first two peaks (I1and I2) in
DF induction curves gradually decreases; I1disappears faster than
I2. It can be seen that in slow-decaying components concomitant
with the disappearance of the first two maxima in the induction
curve, an appearance of a new peak, I3, may be observed (Fig. 4,
inset). I3is absent in fast-decaying components and rarely could be
observed as a separate maximum in the total DF induction curve.
From a three-dimensional plot (not shown) it can be seen that this
maximum is not time-shifted I2, but is rather a new peak, which is
entirely the result of the slow (seconds)-decaying DF components.
The ratio between the peaks in the slow phase of DF induction in
some samples (maize, barley) also changes (data not shown). At
the fastest decaying registered DF components, the maximum I4
has higher amplitude than I5, whereas at slower decaying
Figure 2. Experimental data during integral PF and DF registration regime.
Integral values of PF induction curve points were obtained as a mean value
of 10th to 20th registered points during the period of illumination of the
sample. Integral values of DF induction curves were obtained as a mean
value from all (about 110) points or from a chosen sequence recorded
points during the period of registration (dark period). The values of DF at
two dark intervals are presented with squares and triangles.
Figure 3. Simultaneously registered induction curves of PF and of DF
decaying in the time range of 0.35 to 5.5 ms. Each PF point is obtained as
a mean value of ten (10th to 20th) consecutive points, registered from an
analogue–digital converter during the periods of illumination of the sample.
DF values are means of all the (about 110) points measured during each
period of registration (from 0.35 to 5.5 ms after the end of illumination).
Because the time needed to open the shutter is 0.2 ms and registration starts
afterward, the exact value of Focannot be measured. Instead of this, the
average of the first 10 points is stored as Foand this point corresponds to the
OJ rise observed with high-resolution fluorometers (34).
294Ivelina Zaharieva and Vassilij Goltsev

Page 4

components the I4and I5amplitudes become equal and I5becomes
the more expressed maximum in the slow DF induction phase.
The software developed allows simultaneous recording (using
a phosphoroscope) not only of PF and DF (decaying in different
time ranges) induction curves but also of overbearing numbers of
DF relaxation kinetics, registered during the dark periods of the
phosphoroscope working cycle. During the induction period,
changes are observed in the DF components: in the first few
seconds after the beginning of the illumination, the fast-decaying
components predominate over the slow-decaying ones. After
prolonged illumination, the participation of the slow-decaying
DF component rises (23,37,44–47). Thus, this leads to difference
in the shape of the relaxation kinetics during the induction period,
which may be used to evaluate the contributions of different kinetic
components to the total DF emitted during the induction period.
The operator sets in advance in the experimental setup the number
of DF registration intervals, the number of saved DF dark
relaxation kinetics as well as their distribution during the induction
period—the desired number of DF relaxation kinetics is distributed
regularly in logarithmic or linear time scale or the fixed position of
the DF relaxation kinetics are recorded. There is a possibility to
save all relaxation kinetics; however, after the first 1 s of the
induction period, the rate of the changes in the photosynthetic
apparatus decreases and recording of such a large number of
kinetics (about 16000 for 3 min measuring time) becomes un-
necessary. It is more appropriate to save only a few tens of
DF relaxation kinetics, regularly distributed in logarithmic time
scale. Each of the saved DF kinetics is averaged with previously
missed relaxation kinetics, thus decreasing the noise in the saved
kinetics as well. The operator has the possibility to determine the
maximal number of DF kinetics, which may be averaged, e.g. up to
100 (they are measured for about 1 s). In this manner, we can avoid
averaging relaxation kinetics of different shapes. Figure 5 presents
13 relaxation kinetics curves registered at different times in the first
50 s after the beginning of intermittent illumination. All missed
relaxation kinetic curves are averaged programmatically in the next
saved curve.
Even in this restricted time window (0.35–5.5 ms), the DF decay
displays polyphasic behavior. Such types of decays are commonly
analyzed by a variety of programs in terms of several exponential
components, usually no more than three (26). When every re-
gistered polyexponential decay is presented as a sum of in-
dividual exponents, we can obtain some information about the
changes in concentrations and lifetimes of light-emitting states of
the PSII reaction center because it can be assumed that the
amplitudes of these exponents correspond to the concentration of
different light-emitting states of the reaction center and that the
lifetimes characterize their stability. The induction curves of
lifetimes and amplitudes of the different kinetic components, which
can be registered with the FL-2006 apparatus, are presented in
Fig. 6. From comparison of rate constants of electron transport
reactions on the PSII acceptor side with lifetimes of measured DF
kinetic components, we have supposed relationships between the
particular states of the PSII reaction center, which are able to
recombine with the emission of DF, and the parameters of DF
decays. The emission, decaying in the submillisecond domain (L1),
is connected with the recombination of Z1P680QA2QB2, and this
decayed in millisecond time domain (L2)—with the recombination
of Z1P680QA2QB5. The lifetimes (s1and s2) are determined by the
probabilities of direct and back electron transport in these states
(33). These assumptions were fitted well for interpretation of
experiments done with modificators of PSII electron transport (33)
or of thylakoid membrane fluidity (48), as well as for interpretation
of temperature dependences of DF decay parameters (43). The
millisecond characteristic time (s2) increases approximately two-
fold during the first second of illumination (Fig. 6). This increase
observed at the beginning of the induction period is possibly
connected with the photoinduced PQ pool reduction (33) because
with the closure of PQ pool, one of the ways for disappearance of
the Z1P680QA2QB5state (namely, dissociation of the PQ molecule
from the QBbinding site) becomes blocked. After the first few
seconds of illumination, s2only slightly decreases. So it can be
concluded that after the photosystem I (PSI) activation, the PQ
pool is only partially reoxidized. The main part of DF increase in
this moment is probably due to the photoinduced proton gradient
but not to the reopening of PSII reaction centers (36). The PQ pool
maintains some level of reduction after the PSI activation. At the
Figure 4. Simultaneously registered induction curves of DF, decaying in
50 consecutive time intervals, each with a duration of about 100 ls. The
bold black line represents the induction curve of DF, decaying in the total
available time range (0.35 to 5.5 ms), obtained from the same measurement.
The inset shows the induction curves of DF recorded in intervals from 0.5
to 0.6 ms (solid line) and 5.1 to 5.2 ms (dotted line). Both curves in the inset
are normalized to the maximal DF value.
Figure 5. Typical record of DF relaxation kinetics during the induction
period. One cycle of light and dark period of the phosphoroscope was
11 ms in duration: 0–5.15 ms for excitation and 5.15–11 ms for darkness.
Photochemistry and Photobiology, 2003, 77(3)295

Page 5

end of the period of registration, a sharp decrease in s2can be
observed, which coincides with the activation of the Calvin–
Benson cycle enzymes and reopening of PSII reaction centers. This
is evidence that the PF decrease after the maximal level Fp
(observed at this moment of induction period) is due to the
nonphotochemical quenching rather than PQ pool reoxidation.
This is in agreement with the works of Briantais et al. (41,42),
shown that P-to-S decay in PF is a result of energy-dependent
quenching by photoinduced lumen acidification.
Elevated temperatures increase the rate of electron transport
reactions in the forward as well as in the backward direction. This
causes changes in amplitudes and lifetimes of different kinetic
components (23). For this purpose, valuable information about
single-kinetic components can be obtained from registration of DF
and PF stationary levels at different temperatures. A typical record
of temperature curves of PF and DF decaying in the entire interval
of registration with FL-2006 is presented in Fig. 7. The thermo-
gram of PF (in this study the Fsvalue is recorded) is similar but not
identical to the frequently registered thermograms of Fo(49). From
temperature dependence of DF, we can obtain the temperature of
half-inactivation of DF, which is usually connected with the
inhibition of the PSII oxygen-evolving system (23,50).
During gradual heating of the sample, besides the integral PF
and DF values, the entire DF relaxation kinetics is recorded at each
point. Decomposition of registered DF relaxation kinetics at
different temperatures into simple components provides us the
opportunity to obtain the temperature dependences of stationary
concentration and lifetimes of different light-emitted states (Fig. 8).
The amplitudes, as well the lifetimes of the three kinetic
components, have various temperature dependences, so they are
determined from different electron transfer reactions in PSII, which
are characterized by different rate constants and different back-
reaction activation energies. The temperature dependence of
integral DF is determined by at least two processes: (1) the
temperature dependence of rate constants of back electron transport
reactions (because of the high activation energy required); and (2)
the changes in the relative part of different DF components in the
total DF emission. These are dependent on the influence of
temperature on many processes in the photosynthetic apparatus
(reaction in Calvin–Benson cycle, thylakoid membrane energiza-
tion, etc.). By applying some inhibitors (e.g. DCMU) or electron
donors and acceptors to the photosynthetic apparatus to eliminate
particular light-emitted states of PSII reaction center, it is possible
to obtain energies of activation for separate reactions in the PSII
acceptor side (19).
CONCLUSIONS
In summary, the presented method for data processing is very
useful for providing us the DF and PF integral induction curves
from a single measurement, a large number of DF induction curves
in different intervals of DF decay as well as kinetics of these
decays. It increases the amount of data collected from one
measurement, decreases errors due to manual dark adaptation
period determination, decreases errors in data deviation from dif-
ferent samples and automates the measuring protocol.
Furthermore, the new program has increased the informative
value of the data obtained and will increase the basic knowledge
about DF itself. Simultaneous registration of DF in many intervals
will allow more profound understanding of the nature of DF
induction maxima. Using this method a new DF maximum (I3) has
been observed, which is caused by increasing the contribution of
slow (seconds time) DF kinetic components. Increased number of
DF dark relaxation kinetics is expected to help us in the analysis
of photoinduced dynamics of the three major DF components. In
addition, changes in the kinetic characteristics of the individual
components can be easily studied. Finally, the thermograms of DF
dark relaxation kinetic parameters are now available.
The software developed gives the possibility to follow the
kinetic characteristics of different DF components during the
induction as well as during gradual heating. The suggested way for
Figure 6. Time courses of dark relaxation parameters of delayed
chlorophyll fluorescence in leaf disks from Arabidopsis thaliana. The
parameters were calculated by fitting of experimental DF decay kinetics
using equation DF(t) 5 L1? e2t/s11 L2? e2t/s21 L3. (A) Amplitudes L1,
L2and L3. (B) Lifetimes, s1and s2, of the kinetic components.
Figure 7. Thermograms of PF and DF stationary levels. The PF and DF
values are obtained as described in Fig. 3. Leaf disks from Arabidopsis
thaliana were preilluminated for 2 min with actinic light (1200 lmol
photons?m22?s21) at 208C; then the samples were cooled to 58C and heated
with the rate of about 38C?s21. The PF and DF were registered at every 18C
during the heating.
296Ivelina Zaharieva and Vassilij Goltsev

Page 6

signal processing could be applied for most of the phosphoro-
scopes used in our days and also for an optoelectronic-based
apparatus, which does not use a phophoroscope. If this way of data
acquisition is applied to another apparatus, it will allow registered
data to be comparable with the data obtained from other authors
because of the simultaneous registration of DF, decaying in dif-
ferent time ranges.
Acknowledgements—We thank P. Chernev for help in writing the Win-
dows-based software and Prof. Govindjee for critical reading of the man-
uscript. This work is supported by the Swiss National Scientific Fund
(SCOPES 2000–2003 grant 7BUPJ062408.00/1).
REFERENCES
1. Strehler, B. L. and W. Arnold (1951) Light production by green plants.
J. Gen. Phys. 34, 809–820.
2. Lavorel, J. (1975) Luminescence. In Bioenergetics of Photosynthesis
(Edited by Govindjee), pp. 223–317. Academic Press, New York.
3. Markovic, D. Z., C. N. Radenovic, L. Rafailovic, S. A. Zerajic, M.
Jeremic and M. B. Markovic (1999) Temperature dependence of
delayed fluorescence induction curve transients. Gen. Physiol. Biophys.
18, 257–267.
4. Govindjee and P. A. Jursinic (1979) Photosynthesis and fast changes in
light emission by green plants. In Photochemical and Photobiological
Reviews, Vol. 4 (Edited by K. C. Smith), pp. 125–205. Plenum Press,
New York.
5. Jursinic, P. (1986) Delayed fluorescence: current concepts and status. In
Light Emission by Plants and Bacteria (Edited by Govindjee, J. Amesz
and D. J. Fork), pp. 291–328. Academic Press, Orlando, Florida.
6. Godik, V. I. and A. Yu. Borisov (1980) Short-lived delayed
luminescence of photosynthesizing organisms. II. The ratio between
delayed and prompt fluorescence as studied by the modulation method.
Biochim. Biophys. Acta 590, 182–193.
7. Sonneveld, A., L. N. M. Duysens and A. Moerdijk (1981) Sub-
microsecond chlorophyll delayed fluorescence from photosystem I.
Magnetic-field induced increase of the emission yield. Biochim.
Biophys. Acta 636, 39–49.
8. Haveman, J. and J. Lavorel (1975) Identification of the 120 ls phase in
the decay of delayed fluorescence in spinach chloroplasts and
subchloroplast particles as the intrinsic back reaction. The dependence
of the level of this phase on the thylakoid internal pH. Biochim.
Biophys. Acta 408, 269–283.
9. Holzapfel, C. and A. Haug (1974) Time course of microsecond-delayed
light emission from Scenedesmus obliquus. Biochim. Biophys. Acta
333, 52–58.
10. Barber, J. and J. Neumann (1974) An energy-conservation site between
H2O and DBMIB: evidence from msec delayed light and chlorophyll
fluorescence studies in chloroplasts. FEBS Lett. 40, 186–189.
11. Bertsch, W. (1969) Millisecond delayed light emission from photore-
action II: two models for the reaction center. In Progress in
Photosynthesis Research, Vol. 2 (Edited by H. Metzner), pp. 996–
1005. H. Laupp Jr., Tu ¨bingen, Germany.
12. Hipkins, M. F. and J. Barber (1974) Estimation of the activation energy
for millisecond delayed fluorescence from uncoupled chloroplasts.
FEBS Lett. 42, 289–292.
13. Barbieri, G., R. Delosme and P. Joliot (1970) Comparaison entre
l’emission d’oxygene et l’emission de luminescence a la suite d’une
serie d’eclairs saturants. Photochem. Photobiol. 12, 197–206.
14. Joliot, P., A. Joliot, B. Bouges and G. Barbieri (1971) Studies of system
II photocenters by comparative measurements of luminescence,
fluorescence, and oxygen emission. Photochem. Photobiol. 14,
287–305.
15. Rutherford, A. W. and Y. Inoue (1984) Oscillation of delayed
luminescence from PS II: recombination of S2QB2and S3QB2. FEBS
Lett. 165, 163–170.
16. Rutherford, A. W., Govindjee
accumulation and photochemistry in leaves studied by thermolumines-
cence and delayed light emission. Proc. Natl. Acad. Sci. USA 81,
1107–1111.
17. Govindjee (1995) Sixty-three years since Kautsky: chlorophyll
a fluorescence. Aust. J. Plant Physiol. 22, 131–160.
18. Stacy, W. T., T. Mar, C. E. Swenberg and Govindjee
analysis of a triplet exciton model for the delayed light in Chlorella.
Photochem. Photobiol. 14, 197–219.
19. Jursinic, P. and Govindjee (1977) Temperature dependence of delayed
light emission in the 6 to 340 microsecond range after a single flash in
chloroplasts. Photochem. Photobiol. 26, 617–628.
20. Jursinic, P., Govindjee and C. A. Wraight (1978) Membrane potential
and microsecond to millisecond delayed light emission after a single
excitation flash in isolated chloroplasts. Photochem. Photobiol. 27,
61–71.
21. Wong, D., Govindjee and P. Jursinic (1978) Analysis of microsecond
fluorescence yield and delayed light emission changes after a single
flash in pea chloroplasts: effects of mono- and divalent cations.
Photochem. Photobiol. 28, 963–974.
22. Van Gorkom, H. J. and M. Donze (1973) Charge accumulation in the
reaction center of photosystem 2. Photochem. Photobiol. 17, 333–342.
23. Veselovskii, V. A. and T. V. Veselova (1990) Luminescent character-
istics of plants photosynthetic apparatus. In Luminestsentsiya Rastenii,
pp. 8–78. Nauka, Moskva. (In Russian)
24. Srivastava, A., R. J. Strasser and Govindjee (1999) Greening of peas:
parallel measurements of 77 K emission spectra, OJIP chlorophyll
a fluorescence transient, period four oscillation of the initial
fluorescence level, delayed light emission, and P700. Photosynthetica
37, 365–392.
25. Strasser, R. J. (1974) Studies on the oxygen evolving system in flashed
leaves. In Proceedings of the Third International Congress on
Photosynthesis (Edited by M. Avron), pp. 497–503. Elsevier Scientific
Publishing Company, Amsterdam.
26. Lavorel, J., J. Breton and M. Lutz (1986) Methodological principles of
measurment of light emitted by photosynthetic systems. In Light
Emission by Plants and Bacteria (Edited by Govindjee, J. Amesz and
D. J. Fork), pp. 57–98. Academic Press, Orlando, Florida.
27. Duysens, L. N. M. and H. E. Sweers (1963) Mechanism of two
photochemical reactions in algae as studied by means of fluorescence.
In Studies on Microalgae and Photosynthetic Bacteria (Edited by
Japanese Society of Plant Physiologists), pp. 353–372. University of
Tokyo Press, Tokyo.
28. Itoh, S. and N. Murata (1974) Studies on the induction and decay
kinetics of delayed light emission in spinach chloroplasts. In
Proceedings of the Third International Congress on Photosynthesis
(Edited by M. Avron), pp. 115–126. Elsevier, Amsterdam.
and Y. Inoue (1984) Charge
(1971) An
Figure 8. Temperature dependence of dark relaxation parameters of
delayed chlorophyll fluorescence in leaf disks from Arabidopsis thaliana.
The parameters were calculated as described in Fig. 6. The measuring
procedure is as that in Fig. 7.
Photochemistry and Photobiology, 2003, 77(3)297

Page 7

29. Itoh, S. (1980) Correlation between the time course of millisecond
delayed fluorescence and that of prompt fluorescence at low
temperature in uncoupled spinach chloroplasts. Plant Cell Physiol.
21, 873–884.
30. Itoh, S. and N. Murata (1974) Temperature dependence of delayed light
emission in spinach chloroplasts. Biochim. Biophys. Acta 333, 525–534.
31. Schreiber, U. and U. Schliwa (1987) A solid-state, portable instrument
for measurment of chlorophyll luminescence induction in plants.
Photosynth. Res. 11, 173–182.
32. Radenovic, C., D. Markovic and M. Jeremic (1994) Delayed
chlorophyll fluorescence in plant models. Photosynthetica 30, 1–24.
33. Goltsev, V., L. Traikov and V. Hristov (1998) Effects of exogenous
electron acceptors on kinetic characteristics of prompt and delayed
fluorescence in atrazine inhibited thylakoid membranes. In Photosyn-
thesis: Mechanisms and Effects (Edited by G. Garab), pp. 3885–3888.
Kluwer Academic, Dordrecht, The Netherlands.
34. Strasser, R. J., A. Srivastava and Govindjee
chlorophyll a fluorescence transient in plants and cyanobacteria.
Photochem. Photobiol. 61, 32–42.
35. Goltsev, V. N. and I. T. Yordanov (1997) Mathematical model of
prompt and delayed chlorophyll fluorescence induction kinetics.
Photosynthetica 33, 571–586.
36. Goltsev, V., I. Zaharieva, P. Lambrev, I. Yordanov and R. Strasser
(2003) Simultaneous analysis of prompt and delayed chlorophyll
a fluorescence in leaves during the induction period of dark to light
adaptation. J. Theor. Biol. (In press)
37. Itoh, S. and N. Murata (1973) Correlation between delayed light
emission and fluorescence of chlorophyll a in system II particles
derived from spinach chloroplasts. Photochem. Photobiol. 18, 209–218.
38. Wraight, C. A. and A. T. Crofts (1971) Delayed fluorescence and the
high-energy state of chloroplasts. Eur. J. Biochem. 19, 386–397.
39. Satoh, K. and S. Katoh (1983) Induction kinetics of millisecond-
delayed luminescence in intact Bryopsis chloroplasts. Plant Cell
Physiol. 24, 953–962.
40. Gaevskii, N. A. and V. N. Morgun (1993) Use of variable fluorescence
and delayed light emission to studies of plant photosynthesis. Fiziol.
Rast. 40, 136–145. (In Russian)
(1995) Polyphasic
41. Briantais, J.-M., C. Vernotte, M. Picaud and G. H. Krause (1979) A
quantitative study of the slow decline of chlorophyll a fluorescence in
isolated chloroplasts. Biochim. Biophys. Acta 548, 128–138.
42. Briantais, J.-M., C. Vernotte, M. Picaud and G. H. Krause (1980)
Chlorophyll fluorescence as a probe for the determination of the photo-
induced proton gradient in isolated chloroplasts. Biochim. Biophys. Acta
591, 198–202.
43. Zaharieva, I., St. G. Taneva and V. Goltsev (2001) Effect of
temperature on the luminescent characteristics in leaves of Arabidopsis
mutants with decreased unsaturation of the membrane lipids. Bulg. J.
Plant Physiol. 27, 3–18.
44. Clayton, R. K. (1969) Characteristics of prompt and delayed
fluorescence from spinach chloroplasts. Biophys. J. 9, 60–76.
45. Etienne, A. L. (1974) Effects of carbonyl cyanide m-chlorophenylhy-
drazone and hydroxylamine on the photosystem II electron exchange
mechanism in 3-(3,4-dichlorophenyl)-1,1-dimethylurea treated algae
and chloroplasts. Biochim. Biophys. Acta 333, 497–508.
46. Malkin, S. and J. Barber (1978) Induction patterns of delayed
luminescence from isolated chloroplasts. I. Response of delayed
luminescence to changes in the prompt fluorescence yield. Biochim.
Biophys. Acta 502, 524–541.
47. Mar, T., J. Brebner and G. Roy (1975) Induction kinetics of delayed
light emission in spinach chloroplasts. Biochim. Biophys. Acta 376,
345–353.
48. Zaharieva, I., M. Velitchkova and V. Goltsev (1998) Effect of
cholesterol and benzyl alcohol on prompt and delayed chlorophyll
fluorescence in thylakoid membranes. In Photosynthesis: Mechamisms
and Effects (Edited by G. Garab), pp. 1827–1830. Kluwer Academic,
Dordrecht, The Netherlands.
49. Schreiber, U. and J. A. Berry (1977) Heat-induced changes of
chlorophyll fluorescence in intact leaves correlated with damage of
the photosynthetic apparatus. Planta 136, 233–238.
50. Goltsev, V., I. Yordanov, D. Stefanov, I. Zaharieva, P. Lambrev and R.
Strasser (2001) Simultaneous analysis of variable and delayed
chlorophyll fluorescence during induction period in photosynthetic
apparatus. In PS2001 Proceedings: 12th International Congress on
Photosynthesis. CSIRO Publishing, Melbourne, Australia.
298 Ivelina Zaharieva and Vassilij Goltsev