- Email: [email protected]
Parameters of photosynthetic energy partitioning
Journal of Plant Physiology 175 (2015) 131–147
Contents lists available at ScienceDirect
Journal of Plant Physiology journal homepage: www.elsevier.com/locate/jplph
Review article
Parameters of photosynthetic energy partitioning Duˇsan Lazár ∗ Department of Biophysics, Center of the Region Haná for Biotechnological and Agricultural Research, Faculty of Science, Palack´ y University, Sˇ lechtitel˚ u 11, 783 71 Olomouc, Czech Republic
a r t i c l e
i n f o
Article history: Received 27 August 2014 Received in revised form 22 October 2014 Accepted 25 October 2014 Available online 29 November 2014 Keywords: Quantum yield Photochemical Non-photochemical Chlorophyll fluorescence Photosynthesis
a b s t r a c t Almost every laboratory dealing with plant physiology, photosynthesis research, remote sensing, and plant phenotyping possesses a fluorometer to measure a kind of chlorophyll (Chl) fluorescence induction (FLI). When the slow Chl FLI is measured with addition of saturating pulses and far-red illumination, the so-called quenching analysis followed by the so-called relaxation analysis in darkness can be realized. These measurements then serve for evaluation of the so-called energy partitioning, that is, calculation of quantum yields of photochemical and of different types of non-photochemical processes. Several theories have been suggested for photosynthetic energy partitioning. The current work aims to summarize all the existing theories, namely their equations for the quantum yields, their meaning and their assumptions. In the framework of these theories it is also found here that the well-known NPQ parameter (= (FM − FM )/FM ; Bilger and Björkman, 1990) equals the ratio of the quantum yield of regulatory light-induced non-photochemical quenching to the quantum yield of constitutive non-regulatory non-photochemical quenching (˚NPQ /˚f,D ). A similar relationship is also found here for the PQ parameter (˚P /˚f,D ). © 2014 Elsevier GmbH. All rights reserved.
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chl fluorescence and quantum yields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pulse amplitude modulation technique for measurement of Chl fluorescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The slow Chl FLI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Saturation pulse method for determination of Chl fluorescence quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The relaxation analysis following the quenching analysis to discriminate particular types of non-photochemical quenching . . . . . . . . . . . . . . . . . . . . Energy partitioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – employing the quenching analysis and consideration of active centers only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Demmig-Adams et al. (1996) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kato et al. (2003) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ishida et al. (2011) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kramer et al. (2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – employing the quenching and relaxation analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ahn et al. (2009) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kasajima et al. (2009) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Guadagno et al. (2010) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – discrimination of the sustained regulatory non-photochemical quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Porcar-Castell et al. (2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Porcar-Castell (2011) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – consideration of photoinactivated centers, in addition to active centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Tel.: +420 58 563 4164. E-mail address: [email protected] http://dx.doi.org/10.1016/j.jplph.2014.10.021 0176-1617/© 2014 Elsevier GmbH. All rights reserved.
132 132 133 133 134 136 138 138 138 138 138 139 139 140 140 140 141 141 141 142 142
132
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
Kornyeyev et al. (2001) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hikosaka et al. (2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hendrickson et al. (2005) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kornyeyev and Hendrickson (2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kornyeyev and Holaday (2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – connection of the theories employing the quenching and relaxation analyses with the theories of active and photoinactivated centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kornyeyev et al. (2013) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
142 143 143 143 144 144 144 145 146 146
Introduction Chlorophyll (Chl) fluorescence induction (FLI) is one of the most frequently used methods of all Chl fluorescence measurements. Therefore, it is widely and extensively used not only in plant physiology and photosynthesis research, but also in remote sensing and plant phenotyping. The Chl FLI was first observed by Kautsky and Hirsch (1931) by their own eyes; the curve constructed based on their observation started with a fast fluorescence rise (FLR) followed by a slow fluorescence decrease (FLD) to a steady state. The FLR and FLD are also known as the fast and the slow phases of FLI, respectively (e.g., Govindjee, 1995). Later, with improvements in fluorometers, use of the Chl FLI was in fact split into two types of measurements: (i) measurement of the Chl FLR with a high time resolution and usually under high intensity of excitation light (the O-J-I-P curve) and (ii) measurements of whole Chl FLI for a long time and applying different sources of excitation light (the quenching analysis). These two types of measurements are also related to different excitation and detection of fluorescence signal (see below). The fast Chl FLR was reviewed, e.g., by Lazár (2006), Stirbet and Govindjee (2011, 2012), and Schansker et al. (2014), while the slow Chl FLD and Chl FLI were reviewed, e.g., by Lazár (1999), Maxwell and Johnson (2000), Papageorgiou et al. (2007), Baker (2008), and Porcar-Castell et al. (2014). The Chl FLI was also theoretically modeled and this effort was reviewed by Lazár and Schansker (2009), Vredenberg and Práˇsil (2009), Rubin and Riznichenko (2009), and Stirbet et al. (2014). In the past two decades, a lot of effort has been made to define the so-called photosynthetic energy partitioning, i.e., to define parameters (quantum yields) which will reflect the fraction of absorbed light energy utilized by particular processes (photochemistry vs. non-photochemical dissipations). Irrespective of many publications on this topic, there exists no review article that summarizes all of the derived parameters. Therefore, a researcher can be confused by the existence of so many parameters, their interpretation and the assumptions under which the derivations are made. The reader thus should be introduced and correctly informed about differences among the parameters and their proper usage. Even experts are not familiar with all of the existing theories. It is, therefore, the main aim of this review to contribute a summary of all known approaches and parameters of photosynthetic energy partitioning. In the current review, the basics of the Chl fluorescence and quantum yields are described first, followed by a description of the measuring techniques together with explanation of basic fluorescence levels and quenching coefficients and parameters, followed by detailed description and discussion of the parameters (quantum yields) of the photosynthetic energy partitioning. A side product of reviewing the theories is that it is found here that the well-known NPQ and PQ parameters are equal to a ratio of two quantum yields (Eqs. (38) and (39); see below). The topic of this review is very closely related to the mechanisms of the light-induced regulatory non-photochemical Chl
Fig. 1. A scheme of energetic levels (the so-called Jablonski diagram) of a single molecule showing routes of formation and deactivation of the energetic levels. The thin horizontal lines show vibrational energy levels which are superimposed on electronic energy levels of the ground state (S0 ), excited singlet states (S1 and S2 ) and excited triplet state (T1 ) shown by thick horizontal lines. Vertical thick arrows symbolically show orientation of spin; when the arrows in the ground and exited states are in the opposite/same directions, the singlet/triplet excited state is realized. Vertical wavy arrows indicate radiative transitions by means of absorption of light (A) of different wavelengths (shorter wavelength (blue) means higher energy whereas longer wavelength (orange) means lower energy) and by means of emission of light by fluorescence (F; S1 –S0 transition) and by phosphorescence (P; T1 –S0 transition). Thin vertical or skew arrows indicate non-radiative transitions by means of vibrational relaxation (VR), internal conversion (IC), and inter-system crossing (ISC) which all represent a loss of energy by heat dissipation. For the case of a photosynthetic pigment molecule, additional deactivations of the excited states not shown in the figure can occur: via energy transfer to another molecule and via photochemistry (the primary charge separation). Each of all the above mentioned processes is characterized by its rate constant (not shown in the figure). In the dark-adapted state of healthy non-stressed sample the non-radiative transitions are the origins of the basal constitutive non-regulatory heat dissipations characterized by its overall apparent rate constant kD (see the text and Fig. 2). In a light-adapted state of the sample, light-induced changes (increase) in the values of the rate constants of the non-radiative transitions occur (with respect to their values in the dark-adapted state) which cause the light-induced regulatory heat dissipation characterized by its overall apparent rate constant kNPQ (see the text and Fig. 2). When the sustained regulatory non-photochemical quenching is present, it also affects the non-radiative transitions of the dark-adapted state (increase of kD ). The same is true when a photoinhibition occurs.
fluorescence quenching as such. Much discussion is still ongoing on these mechanisms. As this is beyond of the scope of this review, it does not deal with these mechanisms and the readers are referred to a recent review on the biodiversity of the non-photochemical Chl fluorescence quenching (Goss and Lepetit, 2014) and to a recent book on all aspects of the non-photochemical Chl fluorescence quenching edited by Demmig-Adams et al. (2014). Chl fluorescence and quantum yields To access the photosynthetic energy partitioning, the Chl fluorescence signal is used. By definition, fluorescence is a radiative deactivation of an excited singlet state (usually the first singlet
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
133
particular processes and therefore their values can be directly compared, opposite to other parameters (see below) whose relative changes can only be compared. Pulse amplitude modulation technique for measurement of Chl fluorescence
Fig. 2. Schematic representation of basic partitioning of incoming light energy occurring in photosystem II into photochemistry (characterized by its effective rate constant kP ; kP = kcs qP , where kcs is the intrinsic rate constant of charge separation in PSII and qP estimates fraction of the open PSII (Eqs. (4) and (6))), fluorescence emission (kf ), basal constitutive non-regulatory heat dissipation (kD ), and light-induced regulatory heat dissipation (kNPQ ). In a context of theories which do not consider the sustained non-photochemical quenching and the non-photochemical quenching of the photoinactivated centers, the light-induced regulatory non-photochemical dissipations is reversible and causes the qE and qT , and qI quenching. However, in the theories which consider the sustained non-photochemical quenching and the non-photochemical quenching of the photoinactivated centers, only the qE and qT quenching parts of the light-induced regulatory non-photochemical dissipation are considered as reversible and the qI quenching is considered as irreversible and, in fact, being the sustained non-photochemical quenching and/or reflecting formation of the photoinactivated centers.
state S1 ) of a molecule into the ground state as it is described by ´ what is now called the Jablonski diagram (Jabłonski, 1933) – see Fig. 1. In photosynthesis, Chl fluorescence changes upon illumination of a sample between minimal and maximal fluorescence levels. Thus changing Chl fluorescence is called the variable Chl fluorescence (see also the text below). It is generally assumed that the variable Chl fluorescence can be emitted only by photosystem II (PSII; reviewed in Lazár, 1999, 2006; Maxwell and Johnson, 2000; Papageorgiou et al., 2007; Baker, 2008; Stirbet and Govindjee, 2011, 2012; Schansker et al., 2014; Porcar-Castell et al., 2014). However, it was shown that photosystem I (PSI) can also, in principle, emit variable Chl fluorescence (Lazár, 2013 and references therein), but this fluorescence seems to be small when compared with variable PSII fluorescence. Emission of Chl fluorescence is only one route of usage of incoming (absorbed) light energy by PSII. This is schematically shown in Fig. 2 (see also Fig. 1). From a view of Chl fluorescence emission, the other routes represent competing processes that cause quenching of Chl fluorescence; Chl fluorescence is quenched by photochemistry and by non-photochemical processes. There are basically two types of non-photochemical quenching: basal constitutive non-regulatory heat dissipation and light-induced regulatory heat dissipation, see Fig. 2 and the legend of Fig. 1. A quantum yield of a given process, ˚i , is calculated as: ˚i =
ki , ˙k
(1)
where ki is the rate constant of the given process and ˙k is the sum of all the rate constants of all the processes of deactivation of the incoming light energy. As it is usually assumed that the rate constants kf and kD (see Fig. 2) are invariant in time and that they both represent basal always present non-regulatory types of light energy deactivation, they are grouped into one rate constant kf,D and a term basal constitutive non-regulatory dissipation is used. A quantum yield is a number in the interval 0, 1 and it holds that: ˙˚i = 1.
(2)
The advantage of an evaluation of the quantum yields of the photosynthetic energy partitioning is that the yields have the meaning of fractions of the usage of the incoming light energy by
The Chl FLI is mostly measured using the pulse amplitude modulation (PAM) technique. The PAM technique was commercially introduced by Schreiber et al. (1986) based on the work of Horton (1983) and Quick and Horton (1984). In the PAM technique, the measurement of Chl fluorescence is achieved by utilizing two light sources during the measurements: a light that drives photosynthesis and a light which excites (and “measure”) fluorescence. The light which drives photosynthesis can be continuous actinic light and/or a short (usually up to 1 s) saturating pulse or far-red continuous illumination (to oxidize the electron transport chain). The measuring light is in fact very short flashes (about 1–5 s long) whose intensity is always claimed in the literature to be very weak but this is not completely true. When necessary (measurement of minimal fluorescence – see below), the measuring flashes are applied with a low frequency, which leads to a low integrated intensity of excitation light of these measuring flashes. Therefore, the measuring flashes only “measure” fluorescence and do not drive photosynthesis (ideally). On the other hand, in case of background actinic or saturating illumination, the measuring flashes are applied with a high frequency to detect the fast fluorescence changes caused by the actinic/saturating illumination and, in this case, the resulting higher integrated energy of the measuring flashes is not a problem because the background actinic/saturating illumination to drive photosynthesis is already applied. During every measuring flash, the fluorescence signal is detected by a detector (photodiode) featuring a fast response and large linearity range (up to 109 ). The necessity of the first feature is because of the ability to detect fast changes in fluorescence signal and the second feature is necessary because the fluorescence signal is also detected a few microseconds after the measuring flash, i.e., either in the case of a darkness (measurement of minimal fluorescence – see below) when no fluorescence caused by actinic/saturating illumination driving photosynthesis is emitted, or in the case of the actinic/saturating illumination, when a high fluorescence is emitted. Then, the difference between the fluorescence signal detected during the measuring flash and fluorescence signal detected after the measuring flash is the output of the fluorometer. Therefore, the output fluorescence signal reflects fluorescence intensity but it is not directly proportional to the intensity of light used to drive photosynthesis, in contrast to the non-PAM fluorometers. This fluorescence intensity, however, is directly proportional to the quantum yield of Chl fluorescence at a given time. This fluorescence signal is sometimes called modulated fluorescence to convey that it is measured by the PAM technique. Because the modulated fluorescence represents the difference of the two signals, an effect of a possible background light (detected immediately after the measuring flash), e.g., when the measurement is performed in the field, is eliminated, which is the advantage of the PAM technique. At present, fluorometers made by several companies can measure Chl fluorescence utilizing the PAM technique. The slow Chl FLI Before evaluation of fluorescence quenching and energy partitioning, the FLI was measured. Typical slow FLIs are shown in Fig. 3. After peak P either one (M) or two (M1 , M2 ) maxima can appear in the FLD and sometimes, under special conditions, even oscillations can be measured until a steady-state terminal level
134
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
Fig. 3. Typical slow Chl FLI measured with dark-adapted pea leaves under 40 mol m−2 s−1 of red (maximum at about 650 nm) light. The O, P, S, M and T steps (curve a, shifted by 50 s to the right) and the O, P, S1 , M1 , S2 , M2 and T steps (curve b) are denoted. Data were taken from Lazár (1999).
T is reached (e.g., Papageorgiou and Govindjee, 1968; Yamagishi et al., 1978; Walker et al., 1983). With respect to the photosynthetic function, due to a long time scale of the measurements (about 10–30 min), the slow FLI reflects not only the light-induced electron transport reactions occurring in the thylakoid membrane and the non-photochemical Chl fluorescence quenching, but also the dark photosynthetic carbon reactions (Calvin–Benson cycle). The decrease of fluorescence intensity from P to T was used to evaluate the so-called vitality index, Rfd , as follows (Lichtenthaler et al., 1986): Rfd =
FP − FT , FT
(3)
where FP and FT denote fluorescence intensities at the P and T steps, respectively. When a plant is stressed, FP decreases and FT increases, hence Rfd decreases (e.g. Lichtenthaler et al., 2005). Alternatively, a stressed-induced increase in the non-photochemical Chl fluorescence quenching causes more pronounced decrease of the FP -level than of the FT -level leading again to the decrease of Rfd . Saturation pulse method for determination of Chl fluorescence quenching The FLD from P to T can be generally understood to be caused by different types of fluorescence quenching, either of photochemical or of non-photochemical nature. In addition to continuous actinic light, saturating pulses (usually 1 s duration with intensity of excitation light of about 10,000 mol photons of PAR m−2 s−1 ) are applied during the FLI to reveal and quantify the quenching. A typical measured curve is shown in Fig. 4, where particular fluorescence levels are also denoted. The technique was first introduced by Bradbury and Baker (1981) and spread in the course of time and became popular due to the availability of a commercial fluorometer, PAM 101 by Walz (Effeltrich, Germany) (Schreiber et al., 1986) followed by even advanced and more compact fluorometers by Walz, Hansatech (Norfolk, UK), Photon Systems Instruments (Brno, Czech Republic), and by other factories. Together with advancement in the instrumentation, the measuring approaches were also improved by application of far-red light (usually about several seconds) together with switching off of the continuous actinic light (van Kooten and Snel, 1990). The far-red light is predominantly absorbed by PSI, causing oxidation of the electron transport chain in front of PSI,
Fig. 4. Typical quenching analysis followed by the relaxation analysis measured with dark-adapted mature tobacco leaf under 200 mol m−2 s−1 of red (maximum at about 650 nm) actinic light. The time interval starting with blue arrow up and ending with blue arrow down indicates the interval when the measuring flashes were on. The same is true for the actinic light denoted by the green color. The red arrows indicate position of saturating pulses. Short horizontal lines indicate fluorescence levels: blue for minimal fluorescence for dark-adapted state (F0 ), green for actual fluorescence for light-adapted states (F(t)), and red for maximal fluorescence for dark-adapted states (FM ), light-adapted states (FM ), and during dark recovery of the ). The minimal sample, i.e. during relaxation of non-photochemical quenching (FM fluorescence of light-adapted states (F0 ) was not measured. (For interpretation of the references to color in this legend, the reader is referred to the web version of the article.)
consequently leading to formation of open PSII state which enables correct and a fast (compared to a case when the actinic light is switched off only) determination of the minimal Chl fluorescence level for a light-adapted state (see below). All of the so-called quenching coefficients and other parameters listed below are calculated from different basic fluorescence levels (see Fig. 4), which can be divided into three basic groups: • fluorescence levels determined for a dark-adapted state of the sample: ◦ F0 – minimal fluorescence for the dark-adapted state; ◦ FM – maximal fluorescence for the dark-adapted state; • fluorescence levels determined for a light-adapted state of the sample: ◦ F0 – minimal fluorescence for the light-adapted state; – maximal fluorescence for the light-adapted state; ◦ FM ◦ F(t) – actual fluorescence at time t during the slow FLI; in the literature this fluorescence level is often denoted as FS , where S stands for steady state, however, usage of F(t) has a more general meaning; • fluorescence levels determined in darkness, after previous illumination of the sample by actinic light: ◦ F0 – minimal fluorescence during dark recovery after previous illumination; – maximal fluorescence during dark recovery after previous ◦ FM illumination; where following lights are used to obtain particular fluorescence levels: • • • •
F0 – measuring flashes only; FM – measuring flashes and saturating pulse; F0 – far-red light and measuring flashes; – measuring flashes, actinic light and saturating pulse; FM
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
• F(t) – measuring flashes and actinic light; • F – far-red light and measuring flashes; 0 • F – measuring flashes and saturating pulse. M The reader should keep in mind that evaluation of almost all of the fluorescence levels mentioned above might have conceptual and practical problems in their determinations/measurements. For example, to saturate photochemistry, the saturating light pulse is used which causes a fast Chl FLR during the pulse (the so-called O-J-I-P curve) to a maximal fluorescence level. Conceptually it is assumed in all the theories presented in this review that photochemistry is reflected in a reduction of the first quinone electron acceptor of PSII, QA . According to the original hypothesis by Duysens and Sweers (1963), when QA is oxidized in all PSIIs, the minimal fluorescence signal is detected and when QA is reduced in all PSIIs, the maximal fluorescence signal is detected. When QA is oxidized, PSII is in the so-called open state whose fraction is estimated by qP (Eqs. (4) and (6)) and its derivatives (Eqs. (40)–(42) and (69)–(71)). However, it was found that the fast Chl FLR might reflect also other process than QA reduction (for reviews see Lazár, 2006; Stirbet and Govindjee, 2012; Schansker et al., 2014). Therefore, the maximal fluorescence levels might not reflect the QA reduction only leading to overestimation of the maximal fluorescence levels (for reviews see Samson and Bruce, 1996; Samson et al., 1999). On the other hand, practical determination of maximal fluorescence levels, mainly for light-adapted states, might suffer from underestimation because the intensity of saturating light pulse used for determination of the maximal fluorescence levels might not be high enough (due to existence of the light-induced non-photochemical Chl fluorescence quenching, see below). To avoid this problem, different routines for determination of the correct maximal fluorescence levels have been suggested (Earl and Ennahli, 2004; Loriaux et al., 2013). Another example of a possible problem is measurement of minimal fluorescence levels either for the dark-adapted or light-adapted states. Conceptually, a contribution of PSI Chl fluorescence (see Lazár, 2013 and references therein) should be taken into account so that the measured minimal fluorescence levels are not overestimated by the contribution of the PSI Chl fluorescence (reviewed, e.g., by Henriques, 2009). On the other hand, practically, the measuring flashes used for determination of the F0 -level must be of very low intensity so that the flashes do not cause any PSII closure (QA reduction), which would overestimate the real F0 -level. As for the F0 -level, the far-red light used to oxidize the electron transport chain and thus to determine quickly and correctly the F0 -level, must be of enough high intensity otherwise the F0 -level is overestimated by an insufficient oxidation of the electron transport chain by the used far-red light. Overall, because of the problems with correct determination of the Chl fluorescence levels, the related quantum yields of processes of photosynthetic energy partitioning might not be correct. Therefore, further research, which is beyond the scope of this review, is still needed on these topics. As it is not the goal of this review to discuss these issues, the reader is referred to other reviews which dealt with these topics (Samson and Bruce, 1996; Samson et al., 1999; Lazár, 2006; Stirbet and Govindjee, 2012; Harbinson, 2013; Schansker et al., 2014). From the above mentioned basic fluorescence levels some other fluorescence values are calculated:
• FV = FM − F0 – variable fluorescence for a dark-adapted state; • F = F − F – variable fluorescence for a light-adapted state; V M 0 • F = F = F − F(t) – difference in fluorescence between F and M M Q F(t);
135
• F = F − F – variable fluorescence during dark recovery after V M 0 previous illumination. The fluorescence levels and values mentioned above are then used for calculations of the quenching coefficients and other parameters. Even if these quenching coefficients and other parameters were successfully used to describe photosynthetic function, they do not represent quantum yields and are therefore hardly compared absolutely. The basic two quenching coefficients are coefficient of photochemical quenching qP and coefficient of nonphotochemical quenching qN calculated as (Schreiber et al., 1986): qP = qN =
F , FV
(4)
FV − FV FV
.
(5)
The quenching coefficients have meaning of the fractions of variable Chl fluorescence that is quenched by photochemical (qP ) or non-photochemical (qN ) processes. Note that in the original paper by Schreiber et al. (1986) the qP was denoted as qQ and instead of qN , denotation of qE was used, assuming that the energy dependent non-photochemical fluorescence quenching, characterized by its quenching coefficient qE (see below), is the main contributor to overall non-photochemical fluorescence quenching, characterized by its quenching coefficient qN . Also note that in the original paper by Schreiber et al. (1986) the far-red light was not applied to determine F0 and therefore the F0 level was approximated by the F0 level (i.e., the minimal fluorescence is not changed by light-adaptation), which in fact leads to the equations for the quenching coefficients listed below which are also used in the literature: qP = qN =
F −F , FM 0 FM − FM
FV
(6)
.
(7)
In addition to its meaning as mentioned above, qP (Eqs. (4) and (6)) estimates a fraction of “open” PSII centers (i.e., PSII with oxidized QA ) and was in fact deduced for the so-called puddle model of photosynthetic units (PSU), i.e., PSUs which are energetically separated and each PSU contains one reaction center and related antenna system (see Lazár, 1999; Stirbet, 2013 for reviews). Alternatives for the estimation of fraction of “open” PSII centers were also derived (see Eqs. (40)–(42) and (69)–(71) and the text below). It might happen that far-red light to measure F0 is not available in the fluorometer or the F0 level is overestimated (see above). In these cases, the F0 level can be calculated according to Oxborough and Baker (1997) as: F0 =
F0 ). (FV /FM ) + (F0 /FM
(8)
It was later shown that the same equation for F0 can be also derived in the framework of other theories (Klughammer and Schreiber, 2008; Kasajima et al., 2009). In addition, other parameters calculated from fluorescence levels measured during the slow FLI are frequently used: • relative change of minimal fluorescence characterized by the quenching coefficient q0 calculated as (Bilger and Schreiber, 1986): q0 =
F0 − F0 F0
,
(9)
136
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
• non-photochemical Chl fluorescence quenching (determined on only), NPQ, calculated as (Bilger and the basis of FM and FM Björkman, 1990): NPQ =
FM − FM FM
.
(10)
• in analogy with the NPQ parameter, photochemical Chl fluorescence quenching expressed by the PQ parameter (Porcar-Castell et al., 2014; see also Porcar-Castell, 2011 and Laisk et al., 1997), also denoted as PC parameter (Guadagno et al., 2010; see also Ahn et al., 2009) calculated as follows: PQ =
FM FM − . FM F(t)
(11)
Further analysis of Eqs. (10) and (11) reveals (not shown) that the NPQ and PQ parameters are defined on the same basis; they express, for the light-adapted state, values of relative rate constants (relative to sum of the rate constants of fluorescence and basal constitutive non-regulatory heat dissipation, i.e., rate constant of basal constitutive non-regulatory dissipation, kf,D ) of the total non-photochemical and photochemical processes, respectively. Therefore, NPQ and PQ can be directly compared. The other parameters which are calculated from fluorescence values measured during the slow FLI are the quantum yields but they were originally used separately without existence of a theory (photosynthetic energy partitioning) considering that the sum of all the quantum yields for a given state of the sample (dark- vs. light-adapted state) must be unity (Eq. (2)). These quantum yields are: • maximal (or potential) quantum yield of PSII photochemistry for light-adapted state, ˚PSII , calculated as (Oxborough and Baker, 1997): ˚PSII =
FV
FM
,
(12)
• actual quantum yield of PSII photochemistry for light-adapted state, ˚P , calculated as (Genty et al., 1989; the yield was denoted there as ˚C and the authors did not distinguish denotations of the minimal and maximal fluorescence levels for dark- and lightadapted states): ˚P =
F . FM
(13)
In interpretation of data it is also sometimes helpful to have in mind that (Genty et al., 1989; Havaux et al., 1991; Oxborough and Baker, 1997): ˚P = qP ˚PSII .
(14)
As stressed by Kornyeyev and Holaday (2008), it is important to note that ˚PSII describes the situation in a light-adapted sample when the actinic light is not present at the time of its determination, it means when all reaction centers of PSII are oxidized (far-red light is used instead of actinic light leading to oxidation of all electron transport chain, including PSII reaction centers), whereas ˚P describes the situation in a light-adapted sample when the actinic light is present at the time of its determination, it means when reaction centers of PSII are partly oxidized and partly reduced. A correct understanding of the difference between ˚PSII and ˚P is critical for correct interpretation of the parameters (quantum yields) of photosynthetic energy partitioning described in the further text.
Last but not least, the most often used basic Chl fluorescence parameter is the maximal quantum yield of PSII photochemistry for a dark-adapted state, ˚Po , defined as (Kitajima and Butler, 1975): ˚Po =
FM − F0 FV = . FM FM
(15)
Many parameters related to the quenching analysis have been derived. An (older) analysis of the parameters known at that time revealed that only ˚Po , qP , qN , and q0 are independent parameters and hence the other parameters do not bring any new information (Roháˇcek, 2002). At the end of this section it is important to note that, even if an effort for unified terminology and denotation was done (e.g., van Kooten and Snel, 1990), some of the parameters might be denoted in different literature in a little bit varied way than that used here. Alternatively, the same denotation can be used for different parameter(s) which is caused by the fact that too many parameters already exist and it is difficult to create so far unused denotations for them. Therefore the reader must be always aware what parameter is dealing with based on its definition and original references. The relaxation analysis following the quenching analysis to discriminate particular types of non-photochemical quenching To formally distinguish between different modes of measurements, two different terms are used: the quenching analysis and the relaxation analysis, however, both terms are related to the measurement of Chl fluorescence quenching. The term quenching analysis refers to the measurements when actinic light is on (maximally shortly switched off during measurements of F0 ), i.e., the measurements during slow Chl FLI as mostly described so far above, and when the saturation pulses are applied to measure -levels and consequently to reveal the quenching. On the other FM hand, the term relaxation analysis refers to the measurements of sample recovery during darkness (after previous illumination of the sample) for up to 1 h by application of the saturation light pulses -levels (see above and Fig. 4). The analysis of the to measure FM -levels either directly (Quick and Stitt, 1989) or time course of FM by means of their semilogarithmic plot (Hodges et al., 1989) or by fitting of the FV -values (Roháˇcek, 2010) provides information about relaxation of the non-photochemical quenching in darkness which enables discrimination of three types of the non-photochemical Chl fluorescence quenching based on their different relaxation times in darkness. It is important to note that this discrimination is empirical and the relaxation times might differ from species to species and in dependence on the duration and intensity of used previous actinic illumination (see the works described in the section “Energy partitioning – consideration of active centers only and employing the relaxation analysis”). The three types of the Chl fluorescence non-photochemical quenching and their features, as originally discriminated (Horton and Hague, 1988; Hodges et al., 1989; Quick and Stitt, 1989; Walters and Horton, 1991, 1993), are as follows: • energy dependent (also called as high energy state) nonphotochemical Chl fluorescence quenching characterized by its quenching coefficient qE – associated with light-induced acidification of lumen, half-time of the relaxation is about 30 s to 1 min. • state transition non-photochemical Chl fluorescence quenching characterized by its quenching coefficient qT – results from phosphorylation of light-harvesting complexes associated with PSII followed by binding of the light-harvesting complexes to PSI (the so-called state 1 to state 2 transition), half-time of the relaxation is 5–8 min.
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
• photoinhibitory non-photochemical Chl fluorescence quenching characterized by its quenching coefficient qI – results from photoinhibition of PSII photochemistry, half-time of the relaxation is about 20–40 min and more. Occurrence of a particular type of the quenching also depends on the intensity of excitation light; hence, for example, the photoinhibitory non-photochemical quenching occurs only during illumination by a high intensity of excitation light whereas the state transition non-photochemical quenching is largely suppressed during excitation by high light intensity. In the literature, the quenching coefficients are also used for reference to the given type of quenching as such, i.e., the terms qE quenching, qT quenching, and qI quenching are used. The relaxation analysis is usually not used so often because it takes more time to measure it. However, recent papers on energy partitioning also considered the relaxation analysis in their theories (see below). Based on measurements of the relaxation analysis performed after previous high light illumination, when qT quenching is supposed to be suppressed (see above), the following equations were used by Krause and Jahns (2003) to calculate the coefficients of the non-photochemical quenching related to the qE and qI quenching: qE = 1 − qI = 1 −
FV
FV FV
FV
,
(16)
,
(17)
where FV and FV were determined after 10 min of illumination and after 10 min of dark recovery, respectively. It then holds that (Krause and Jahns, 2003): 1 − qN = (1 − qE )(1 − qI ).
(18)
On the other hand, Johnson and Ruban (2011) used the following equation to calculate the coefficients of the non-photochemical quenching related to the qE quenching: qE =
− F FM M FM
,
(19)
and F were determined after 5 min of illumination and where FM M after 5 min of dark recovery, respectively. For the case of high-light illumination, when qT quenching is supposed to be suppressed, a similar set of equations was defined by Krause and Jahns (2003) also for the NPQ parameter (Eq. (10)) as follows:
NPQE =
FM FM − F , FM M
(20)
NPQI =
FM − 1, FM
(21)
were determined in the same way as menwhere the F0 and FM tioned above in the work by Krause and Jahns (2003) and where subscripts E and I indicate fractions of the overall NPQ parameter related to qE quenching and qI quenching, respectively, with the following relationship (Krause and Jahns, 2003):
NPQ = NPQE + NPQI .
(22)
Although the qE quenching is usually connected with formation of zeaxanthin from violaxanthin (e.g., Holt et al., 2005), it was recently found that zeaxanthin-dependent quenching, denoted as qZ , is a new type of the non-photochemical quenching (Holzwarth et al., 2009; Nilkens et al., 2010) different from qE , qT and qI quenching. The qZ quenching probably occurs in PSII attached minor antennae, is formed during 10–30 min, and relaxes within 10–60 min.
137
Two notes are also necessary for the qT quenching. The first note is that in the original concept of discrimination of the nonphotochemical quenching as mentioned above, the qT quenching is not true non-photochemical quenching, i.e., the increase of the rate constant(s) of non-radiative heat dissipation(s) (leading to the quenching of Chl excited states with consequent decrease of Chl fluorescence); during actinic illumination the Chl fluorescence decreases because less excitations are formed in PSII due to the movement and binding of the light harvesting complexes from PSII to PSI (the state 1 to the state 2 transition). And vice versa, during the qT quenching relaxation phase in darkness, more excitations are formed in PSII due to the movement and binding of the light harvesting complexes from PSI back to PSII (the state 2 to the state 1 transition). The second note is that Schansker et al. -levels (2006) found that the qT component of a relaxation of the FM reflects inactivation of ferredoxin–NADP+ -oxidoreductase (FNR). Consequently, as the activation state of FNR affects redox state of the electron transport chain and thus also of the fluorescence signal (if FNR starts to inactivate during the dark relaxation of the sample, the electron transport chain become more reduced, including QA , leading to increased fluorescence signal), the qT relaxation component therefore would be rather of photochemical than of non-photochemical nature. It is clear that more research is needed on the qT component of dark relaxation of the non-photochemical Chl fluorescence quenching. Another type of non-photochemical Chl fluorescence quenching occurs in overwintering evergreens (reviewed, e.g., by Verhoeven, 2014). It is related to Chl fluorescence relaxation by the fact that this quenching does not relax and lasts for the whole winter and it is therefore called the sustained non-photochemical quenching. It is important to note that it represents a regulatory type of non-photochemical quenching; therefore, for the purpose of this work it is called as sustained regulatory non-photochemical quenching. To distinguish this quenching from light-induced regulatory non-photochemical quenching which is switched on when the light is present, the sustained regulatory non-photochemical quenching is usually revealed by means of fluorescence measurements during night (dark-adapted state) when the light-induced regulatory non-photochemical quenching is not present. Therefore, action of the sustained regulatory non-photochemical quenching is mostly determined by measurements of the maximal quantum yield of PSII photochemistry for a dark-adapted state (˚Po , Eq. (15)), which is lowered by the presence of the sustained regulatory non-photochemical quenching. Consequently, recovery of this quenching is monitored by an increase of ˚Po upon removing of winter-stressed plants into room temperature and it has two phases: a fast one occurring during minutes to hours and a slow one occurring within days. Based on various findings it was suggested that the two recovery phases of the quenching reflect the sustained forms of qE and qZ quenching, respectively (Verhoeven, 2014). In some recent studies, also a state of overwintering evergreens (see below) in light was described by the NPQ parameter as (PorcarCastell, 2011; Porcar-Castell et al., 2012): NPQw =
s FM
w FM
− 1,
(23)
where superscripts w and s stand for winter and summer, respectively. The NPQw parameter was further distinguished into the sustained regulatory part, NPQw s , and light-induced (reversible) regulatory part, NPQw , as follows (Porcar-Castell, 2011): r NPQw s =
s FM
w FM
− 1,
(24)
138
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
NPQw r =
s FM
w
FM
s FM
−
w FM
,
(25)
D = 1 − ˚PSII = 1 −
where the following relationship is valid: w
NPQ =
NPQw s
+ NPQw r .
(26)
In analogy with the NPQw parameter, photochemical Chl fluorescence quenching in the light-adapted state of overwintering evergreens was expressed by the PQw parameter (Porcar-Castell, 2011; see also Laisk et al., 1997), as follows: PQw =
s FM w
F(t)
−
s FM w FM
.
the state is when the actinic light is present or not, was defined as:
(27)
As already mentioned above for the NPQ and PQ parameters, w w parameters are defined also the NPQw , NPQw s , NPQr , and PQ on the same basis; they express, for the light-adapted state, values of relative rate constants (relative to kf,D ) of the total non-photochemical, sustained regulatory non-photochemical, light-induced (reversible) regulatory non-photochemical, and photochemical processes, respectively. Therefore, they can be directly compared. Energy partitioning Starting from mid-90s of the last century an effort was done to explore the so-called energy partitioning, i.e., to define quantum yields of processes responsible for utilization of absorbed light energy basically without further discrimination of the non-photochemical quenching into its three (or more) parts as mentioned above. However, the discrimination is included in the energy partitioning concept when the relaxation analysis is also considered (see below). The determination of energy partitioning is very close to the Chl fluorescence quenching analysis because it utilizes the same techniques for the measurement and determination of the basic fluorescence levels. However, in the Chl fluorescence quenching analysis, the goal is mainly to determine the quenching coefficients whereas in the energy partitioning, the goal is to determine the quantum yields. The advantage of usage of the quantum yields instead of the quenching coefficients is that the quantum yields sum up for a given state of the sample (dark- vs. light-adapted state) to unity (Eq. (2)), and therefore the roles of particular processes can be quantified and compared absolutely. As a description of the energy partitioning is the key part of this review, the parameters defined in particular works are presented accordingly. Overall, the theories of energy partitioning can be divided into five groups in dependence if only active PSII centers are considered and the quenching analysis is imployed, if the quenching and relaxation analyses are employed, if the sustained regulatory non-photochemical quenching is discriminated, and if both active and photoinactivated centers are considered. The fifth group is a group which includes a work connecting the theories employing the quenching and relaxation analyses with the theories of active and photoinactivated centers. Energy partitioning – employing the quenching analysis and consideration of active centers only Demmig-Adams et al. (1996) One of the first works on photosynthetic energy partitioning was the paper by Demmig-Adams et al. (1996). In this work, in analogy with the meaning of ˚P (Eq. (13)), denoted as P in this paper, a parameter D, having the meaning of quantum yield of overall thermal dissipation for a light-adapted state, without a specification if
FV
FM
.
(28)
In this approach, both P and D should sum to 1 but this was not found and another parameter E (for excess) had to be considered, which was defined as: E = 1 − D − P = (1 − qP )
FV
FM
.
(29)
The E parameter has according to Demmig-Adams et al. (1996) a meaning of a quantum yield of process(es) of energy utilization different from photochemistry (the P parameter) and thermal dissipation (the D parameter). As the processes hidden behind the E parameter are difficult to “imagine”, this approach seems to be not correct from its beginning. The origin of the problem was found later on by Kornyeyev and Hendrickson (2007); the use of D and P in one balance equation is not correct because D and P describe the energy partitioning under conditions when all reaction centers of PSII are open (light-adapted state when the actinic light is not present) and partially open (light-adapted state when the actinic light is present), respectively, which consequently leads to the existence of the E parameter. Hence, the E parameter does not represent a real component of the energy partitioning but rather a difference between the potential (˚PSII , Eq. (12)) and actual (˚P , Eq. (13)) quantum yields of PSII photochemistry for the light-adapted state (Kornyeyev and Hendrickson, 2007; Kornyeyev and Holaday, 2008): E = 1 − D − P = 1 − (1 − ˚PSII ) − ˚P = ˚PSII − ˚P .
(30)
Kato et al. (2003) The thermal dissipation in the framework of the DemmigAdams theory was further elaborated by Kato et al. (2003); they first considered that the value by which the maximal quantum yield of PSII photochemistry for a dark-adapted state, ˚Po (Eq. (15)), is lower than unity, determines the quantum yield of thermal dissipation in the dark-adapted state, L, expressed as: L = 1 − ˚Po = 1 −
FV , FM
(31)
where L stands for “loss”. Similarly, the quantum yield of thermal dissipation in a light-adapted state, denoted as D again, is expressed as: D=
F FV − V . FM FM
(32)
It further holds that the sum of L and D described by Eqs. (31) and (32), respectively, equals to the original D parameter (Eq. (28)) by Demmig-Adams et al. (1996) describing overall thermal dissipation for a light-adapted state. The other quantum yields (P and E, Eqs. (13) and (29), respectively) were defined by Kato et al. (2003) in the same was as by Demmig-Adams et al. (1996). Ishida et al. (2011) Even if the theory of Demmig-Adams et al. (1996) seems to be not correct, it was further advocated by Ishida et al. (2011). On the basis of experiments with transgenic rice in which the psbS gene was silenced (PsbS protein is necessary for the qE quenching; Li et al., 2000), Ishida et al. (2011) showed that D of the DemmigAdams theory consists of light-independent dissipation present in both the transgenic and in wild-type rice and from light-dependent dissipation present only in wild-type rice. According to Ishida et al.
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
(2011), the light-independent part of D of the Demmig-Adams theory, denoted as A and also referred to as the constitutive component of the quenching (see the other theories below), probably occurs in antenna and its quantum yield is for dark-adapted state calculated as: A=
F0 . FM
(33)
The experimentally found division of original Demmig-Adams’ parameter D into light-independent and light-dependent parts (Ishida et al., 2011) is in agreement with the above mentioned results of Kato et al. (2003) who divided the original parameter D of the Demmig-Adams theory (Eq. (28)) into L and D (Eqs. (31) and (32), respectively) parameters. Moreover, the equation for the parameter A (Eq. (33)) of Ishida et al. (2011), expressing the lightindependent part of the original Demmig-Adams’ parameter D, equals to (is another form of) the equation for the parameter L (Eq. (31)) of Kato et al. (2003), expressing the quantum yield of thermal dissipation in dark-adapted state. Further, even if, on one hand, Ishida et al. (2011) experimentally found a linear dependence of D of the Demmig-Adams theory (Eq. (28)) on the NPQ parameter (Eq. (13)), they, on the other hand, also derived a nonlinear dependence between D and NPQ (and also another nonlinear dependence between D and qN ) in a form: D=
1 − FV . FV + F0 (NPQ + 1)
(34)
Ishida et al. (2011) further found that the E parameter increases in PsbS-silenced plants and they discussed that the E parameter must represent energy dissipation too and that this dissipation occurs in or around closed PSII centers. This agrees with the explanation of the E parameter as the difference between the potential (˚PSII , Eq. (12)) and actual (˚P , Eq. (13)) quantum yields of PSII photochemistry for light-adapted states (Eq. (30)). The difference can be caused only by energy dissipation and the dissipation can occur only in the closed center which causes a decrease of ˚P (PSII reaction centers are partially open and partially closed) with respect to ˚PSII (all PSII reaction centers are open). Further, according to Ishida et al. (2011), the E is basal constitutive heat dissipation (see the other theories below). Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) Another approach for the energy partitioning was started at about the same time, as the first one (Demmig-Adams et al., 1996), by Genty et al. (1996) and Cailly et al. (1996) and this theory was advocated later on by Hendrickson et al. (2004). The works by Genty et al. (1996) and Cailly et al. (1996) are only conference abstracts and therefore, these theories were almost lost in the past until rediscovered by Hendrickson et al. (2004). In this theory of energy partitioning, an effort is done to discriminate different types of non-photochemical losses. These losses can be divided into two processes: constitutive non-regulatory (basal or dark) dissipation processes consisting of fluorescence emission and heat dissipation (see Fig. 2 and the legend to Fig. 1), characterized by their quantum yield ˚f,D (also denoted as ˚cNP ) which is calculated as: ˚f,D
F(t) = FM
(35)
and regulatory light-induced non-photochemical quenching characterized by its quantum yield ˚NPQ (also denoted as ˚r NP ) which is calculated as: ˚NPQ =
F(t) F(t) . − F FM M
(36)
139
The actual quantum yield of PSII photochemistry for the lightadapted state, ˚P (Eq. (13)), was the third part of energy usage by PSII and the sum of all the quantum yields equals unity: ˚P + ˚f,D + ˚NPQ = 1.
(37)
Therefore, there is no parameter (the E parameter) with a nonphysiological meaning as in the Demmig-Adams theory. Another advantage of the theory of Genty et al. is that it does not require knowledge of the F0 fluorescence level (Hendrickson et al., 2004) which is sometimes difficult to determine correctly (see above). So far, it has not been revealed in the literature (some hints can be found in Porcar-Castell (2011); see also Laisk et al. (1997)) that a ratio of ˚NPQ to ˚f,D equals the NPQ parameter of Bilger and Björkman (1990) (Eq. (10)): − F(t)/F F(t)/FM FM − FM ˚NPQ M = = = NPQ. FM ˚f,D F(t)/FM
(38)
Hence, as revealed here, the NPQ parameter does not only reflect the regulatory light-induced non-photochemical quenching as it is usually used for in the literature but also a ratio of the quantum yield of this quenching to the quantum yield of the constitutive non-regulatory non-photochemical quenching. Therefore, the same value of the NPQ parameter can be obtained for different values of ˚NPQ and ˚f,D , which should be taken into account in the analysis of results. In connection to Eq. (38), a similar relationship is revealed here also for the PQ parameter (Eq. (11)): (F − F(t))/FM F FM FM ˚P FM F(t) FM − = PQ, = M = M − = FM FM F(t) FM F(t) F(t) ˚f,D F(t)/FM
(39)
it means that the PQ parameter is a ratio of the quantum yield of actual PSII photochemistry for a light-adapted state to the quantum yield of the constitutive non-regulatory non-photochemical quenching. It is worth to repeat that in this and the following theories, the non-photochemical quenching is divided into two parts: the quenching which is always present, even in healthy leaves because it is the consequence of basic physics (see Fig. 2 and the legend to Fig. 1), the constitutive non-regulatory (basal or dark) non-photochemical quenching/dissipation, and, on the other hand, the non-photochemical quenching which is formed only upon illumination to regulate the amount of absorbed light to avoid occurrence of damage processes, the regulatory light-induced nonphotochemical quenching/dissipation. Kramer et al. (2004) Assuming the so-called lake model for PSUs (one PSU contains several reaction centers which all share the same antenna system without any restrictions; see Lazár, 1999 and Stirbet, 2013 for reviews), which is claimed to describe the energetic connectivity in terrestrial plants better than the puddle model (Kramer et al., 2004), a fraction of open PSII centers at the time t can be estimated from the qL parameter defined as (Kramer et al., 2004): qL =
1/F(t) − 1/FM
1/F0
− 1/FM
=
− F(t) FM − F FM 0
(F0 /F(t)).
(40)
An equation for qL was also derived by Miyake et al. (2009) which was later shown by Kasajima et al. (2009) to be transformed and to be identical to the equation derived by Kasajima et al. (2009): qL =
1/F(t) − 1/FM 1/F0 − 1/FM
.
(41)
140
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
Eqs. (40) and (41) differ in replacement of F0 by F0 in the later equation. For the case of the connected PSUs (each reaction center possesses its own antenna system but with defined probability of energy transfer from one antenna to antenna of another PSU; see Lazár, 1999 and Stirbet, 2013 for reviews), Kramer et al. (2004) derived a fraction of open PSII centers in these PSUs, qcu , to be expressed as: qcu =
− F(t) FM
− F J(F(t) − F0 ) + FM 0
,
(42)
where J is the sigmoidicity parameter (Lavergne and Leci, 1993) related to the probability of energy transfer between PSUs, p (Joliot and Joliot, 1964), as follows: J=
p . 1−p
(43)
Kramer et al. (2004), however, have shown that for appropriate values of the connectivity parameter p (about 0.5–0.7), there is almost no difference between usage of qcu and qL . According to Kramer et al. (2004), the non-photochemical losses can be divided again into two processes with related quantum yields: ˚(NO) =
1 NPQ + 1 + qL ((FM /F0 ) − 1)
˚(NPQ) = 1 − ˚P − ˚(NO),
(44) (45)
where ˚(NO) reflects quantum yield of constitutive non-regulatory (basal or dark) non-photochemical dissipation processes and ˚(NPQ) reflects quantum yield of regulatory light-induced nonphotochemical quenching. It was further shown by Klughammer and Schreiber (2008) and by Kasajima et al. (2009) that after inserting the equation for F0 (Eq. (8)) into ˚(NO) (Eq. (44)) and ˚(NPQ) (Eq. (45)), these lead to the same equations as for ˚f,D (Eq. (35)) and ˚NPQ (Eq. (36)), respectively of the previous theory. Consequently, the relationship among NPQ, ˚NPQ , and ˚f,D (Eq. (38)), is valid also for NPQ, ˚(NPQ), and ˚(NO). The same is true also for the relationship among PQ, ˚P , and ˚f,D (Eq. (39)), and PQ, ˚P , and ˚(NO). Further, it was also shown (Klughammer and Schreiber, 2008) that the equations for ˚(NO) and ˚(NPQ) parameters are valid not only for the lake model of PSUs but also for the puddle model of PSUs. In experiments, ˚NPQ (Eq. (36)) of the previous theory gives almost identical values as ˚(NPQ) (Eq. (45)), however, ˚f,D (Eq. (35)) of the previous theory is significantly higher than ˚(NO) (Eq. (44)). The difference between ˚f,D and ˚(NO) was explained (Hendrickson et al., 2004) by an incorrectness in F0 determination. Kramer et al. (2004) also have shown that the ˚P parameter is valid for both the puddle and lake models of PSUs. It is important also to note (it already implies from Eq. (45)) that: ˚P + ˚(NO) + ˚(NPQ) = 1.
(46)
Hence, the same as in the case of the previous approach (compare with Eq. (37)), there is no need for the E parameter of the Demmig-Adams theory. Overall, even if the theory by Kramer et al. (2004) is based on different assumptions, it in fact leads to almost identical values of the quantum yields as in the theory by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004).
analyses. Ahn et al. (2009) used the same equations for ˚P (Eq. (13)), ˚f,D (Eq. (35)), and ˚NPQ (Eq. (36)) as derived previously by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004). However, Ahn et al. (2009) denoted ˚f,D as ˚C and expressed ˚NPQ as follows: ˚NPQ =
FM − FM F(t) , FM FM
(47)
which is another form of (equals to) Eq. (36). Ahn et al. (2009) further presented equations that express the quantum yields of the various processes that comprise ˚NPQ , namely quantum yield of the qE quenching, ˚qE , and quantum yield of the sum of qT and qI quenching, ˚qT+qI , as follows: ˚qE =
− F FM M F(t) , FM FM
˚qT +qI =
(48)
FM − FM F(t) , FM FM
(49)
is the threshold maximal fluorescence level determined where FM from the relaxation analysis. The threshold time of 10 min for deter was used for Arabidopsis plants. Ahn et al. (2009) mination of FM showed experimentally but it also implies from the above equations that:
˚NPQ = ˚qE + ˚qT +qI .
(50)
Ahn et al. (2009) emphasized that the above equations for ˚NPQ , ˚qE , and ˚qT+qI (Eqs. (47)–(49)), but also for ˚P and ˚f,D , are valid for the light-adapted state when the actinic light is present and thus when a certain fraction of the PSII centers is in the open and the rest in the closed state. Therefore the quantum yields are in fact actual quantum yields. To obtain the quantum yields during the saturation pulse, i.e., for the same conditions of measurements as for the NPQ parameter (Eq. (10)), it is in the case when all reaction centers are closed, following formulae were presented: ˚NPQ,RCC = ˚qE,RCC =
FM − FM
FM
− F FM M
˚qT +qI,RCC =
FM
,
(51)
,
(52)
F FM − FM M
FM
FM
,
(53)
where subscript RCC means “reaction centers closed” and it is again valid that: ˚NPQ,RCC = ˚qE,RCC + ˚qT +qI,RCC .
(54)
It is interesting to note that the set of quantum yields represented by Eqs. (51)–(53) is the only published set of the quantum yields for a light-adapted state when light is switched on and all reaction centers are closed. However, to have a complete set of the quantum yields when all the reaction centers are closed, it is noticed here that the following quantum yields must be added: ˚P,RCC = 0, ˚f,D,RCC = 1 − ˚P,RCC − ˚NPQ,RCC = 1 − 0 −
(55) FM − FM
FM
=
FM
FM
.
(56)
Kasajima et al. (2009) Energy partitioning – employing the quenching and relaxation analyses Ahn et al. (2009) The paper by Ahn et al. (2009) represents the first theory of energy partitioning which employs the quenching and relaxation
Kasajima et al. (2009) derived the same equations for ˚P (Eq. (13)) and for ˚f,D (Eq. (35)) as derived previously by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004). Further, Kasajima et al. (2009) were not so strict in discrimination of the regulatory light-induced non-photochemical quenching with respect to its origin as Ahn et al. (2009) were and divided it originally
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
only into fast and slow components, again based on the relaxation analysis. However, Kasajima et al. (2009) concluded that the fast component corresponded to the qE quenching whereas the slow component approximately corresponded to the sum of qT and qI quenching. The threshold time to distinguish between the fast and slow components of non-photochemical quenching depended on plant species and was determined based on the relaxation analysis to be 1 min for Arabidopsis (compare with 10 min for Arabidopsis used by Ahn et al., 2009) and 5 min for rice. Then, the quantum yields of the fast and slow components of the regulatory lightinduced non-photochemical quenching, ˚NPQ fast and ˚NPQ slow , respectively, were derived to be as follows: ˚NPQ fast =
− 1/F 1/FM M
˚NPQ slow =
1/F(t)
,
− 1/F 1/FM M
1/F(t)
,
(57) (58)
is the threshold maximal fluorescence level determined where FM from the relaxation analysis as mentioned above. It further holds that sum of ˚NPQ fast (Eq. (57)) and ˚NPQ slow (Eq. (58)) equals to ˚NPQ (Eqs. (36) and (47)) of Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004). Also notice that Eqs. (57) and (58) are another forms of (equal to) Eq. (48) and (49), respectively, of Ahn et al. (2009) and are also another forms of (equal to) Eq. (104) and (105), respectively, of Kornyeyev et al. (2013) (see below). Interestingly, these equations were used by Chen et al. (2012) who, however, referred to wrong papers and not to Ahn et al. (2009) or Kasajima et al. (2009).
Guadagno et al. (2010) The work by Guadagno et al. (2010) presents the third and the last theory which employs the quenching and relaxation analyses. Guadagno et al. (2010) paid more attention to finding the threshold -levels during dark relaxation. By plottimes of variations of the FM -levels were used instead ting logarithm of a NPQ parameter (the FM -levels) vs. time during relaxation analysis measured with of the FM beet plants, they have found that there are two variations of the slope of the NPQ curve, revealing thus two threshold times of 2 and 20 min. There was no variation of the slope at 10 min as used for the threshold time by Ahn et al. (2009) and at 1 or 5 min as used for the threshold time by Kasajima et al. (2009). It is important to note here that it was later found that analysis of time dependence of the NPQ data, as was done by Guadagno et al. (2010), distorts real kinetics of the non-photochemical Chl fluorescence quenching (Holzwarth et al., 2013), therefore, the real threshold times might be different than found by Guadagno et al. (2010). Guadagno et al. (2010) then used the ˚P (Eq. (13)), ˚f,D (Eq. (35)), and ˚NPQ (Eq. (36)) as derived previously by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) with the same re-denotation of ˚f,D as ˚C and expression of ˚NPQ (Eq. (47)) as used by Ahn et al. (2009). Thanks to the existence of the two threshold times in the relaxation analysis, Guadagno et al. (2010) were able to divide ˚NPQ into quantum yield of the qE quenching, ˚qE , quantum yield of the qT quenching, ˚qT , and quantum yield of the qI quenching, ˚qI , according to the following formulae:
˚qE =
x − F FM M F(t) , FM Fx
(59)
M
y
˚qT =
x FM − FM F(t)
y
x FM
FM
,
(60)
y
˚qI =
FM − FM F(t) , y FM F M
(61)
141
y
x and F where FM are the maximal fluorescence levels of the first M and second, respectively, threshold times determined from the x represents maximal fluorescence relaxation analysis. In fact, FM signal during the dark recovery of the sample when qE quenching was completely relaxed and qT and qI quenching was still y present whereas FM represents a state of the sample when qE and qT quenching were relaxed and only qI quenching was still present. Guadagno et al. (2010) showed experimentally but it also implies from the above equations that:
˚NPQ = ˚qE + ˚qT + ˚qI .
(62)
The theory used by Guadagno et al. (2010) is so far the most complete theory of energy partitioning employing the quenching and relaxation analyses. Energy partitioning – discrimination of the sustained regulatory non-photochemical quenching Porcar-Castell et al. (2008) Because in overwintering evergreens the sustained regulatory non-photochemical quenching does not relax in winter, even during nights (dark-adapted state), estimation of the quantum yields with respect to energy partitioning and the sustained regulatory non-photochemical quenching was first derived for dark-adapted state (Porcar-Castell et al., 2008). In fact, it is interesting to note that the set of quantum yields in this theory represents the only published set for a dark-adapted state. First, the authors derived the maximal quantum yield of PSII photochemistry for a dark-adapted state, which is identical to Eq. (15). But because the fluorescence values are measured in winter when the sustained regulatory nonphotochemical quenching is present, the quantum yield represents the maximal quantum yield of PSII photochemistry for a darkadapted state when the quenching is present (as mentioned above, the quantum yield is smaller in winter than in summer). To distinguish it from Eq. (15), superscript w (for winter) is used in denotation of the quantum yield and fluorescence levels: ˚w Po =
w − Fw FM 0 w FM
=
FVw w FM
.
(63)
The other quantum yields, for the constitutive non-regulatory (basal or dark) dissipation, ˚w , and for the sustained regulatory f,D
, were derived as follows: non-photochemical quenching, ˚w NPQ ˚w = f,D
F0w As as s Aw aw , FM
˚w NPQs =
F0w w FM
−
(64)
F0w As as s Aw aw , FM
(65)
where A is sample absorptance and a is a fraction of the absorbed light that is captured by PSII. The superscript s indicates reference levels determined in summer. As, usually, the measurements are done with mature leaves where Chl concentration is high enough causing saturated absorptance levels (Porcar-Castell et al., 2008), the changes in absorptance can be omitted. Further, it is assumed that PSII captures the same fraction of the absorbed light as PSI and that this fraction is not changed during seasons (Porcar-Castell et al., 2008), therefore also the changes in fraction of the absorbed light that is captured by PSII can be omitted. Then, Eqs. (64) and (65) can be rewritten to: ˚w = f,D
F0w s FM
˚w NPQs =
,
F0w w FM
(66) −
F0w s FM
.
(67)
142
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
It can be easily checked that the sum of all the quantum yields defined by Eqs. (63), (66) and (67) is unity: w w ˚w Po + ˚f,D + ˚NPQs = 1.
(68)
It is worth to stress that because all the quantum yields (Eqs. (63), (66) and (67)) were determined for a dark-adapted state during which the sustained regulatory non-photochemical quenching persists, the quantum yield of this quenching (Eq. (67)) is directly accessible and no discrimination of the quantum yields of regulatory non-photochemical quenching into its parts is necessary as it is done in the next work (Porcar-Castell, 2011). It is also worth to mention that in the original work by PorcarCastell et al. (2008) was a typographical error (confirmed by a communication with Dr. Porcar-Castell) in the equation for ˚w NPQs (Eq. (9) in Porcar-Castell et al., 2008). Porcar-Castell (2011) The next work by Porcar-Castell (2011) is based on the work by Kramer et al. (2004); the qL parameter was used for estimation of a fraction of total open PSII centers at the time t in overwintering evergreens, qw , as follows (Porcar-Castell, 2011): L w 1/F(t)w − 1/FM
qw L =
,
s 1/F0s − 1/FM
(69)
where superscripts w and s indicate values at winter and summer, respectively. To discriminate the total regulatory nonparameter was photochemical quenching into its parts, the qw L further divided into fractions of open PSII centers where the sustained regulatory non-photochemical quenching is present, qw , and where the light-induced (reversible) regulatory nonL,s , as follows (Porcar-Castell, photochemical quenching is present, qw L,r 2011):
qw L,s = qw L,r =
w 1/F0w − 1/FM
1/F0s
,
w 1/F(t)w − 1/FM
w 1/F0w − 1/FM
.
=
F w w FM
.
(73)
photochemical quenching, ˚w , are almost the same as the yields NPQt from the theory of Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) (Eqs. (35) and (36)) but the reference dark-adapted maximal fluorescence level determined in summer is considered (superscript s), i.e.:
˚w NPQt =
F(t) w FM
(78)
w w ˚w NPQs + ˚NPQr = ˚NPQt .
(79)
w FM
−
where it holds:
The quantum yields of the sustained regulatory non, according to theory of photochemical quenching, ˚w NPQs Porcar-Castell et al. (2008) and Porcar-Castell (2011) differ in usage of the dark-adapted minimal fluorescence level, F0w , in the first case (Eq. (67)) instead of actual fluorescence level, F(t)w , in the second case (Eq. (77)), reflecting the fact that the theories were derived for the dark- and light-adapted states, respectively. On the other hand, the equation for the quantum yield of the lightinduced regulatory non-photochemical quenching, ˚w , (Eq. NPQr (78)) is identical to the equation for the yield according to theory of Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) (Eq. (36)) (when in situ fluorescence levels are considered omitting thus the w superscripts in Eq. (78)). Again, it is worth noting that in the original work by PorcarCastell (2011) there was a typographical error (confirmed by a communication with Dr. Porcar-Castell) in the equation for ˚w f,D (Eq. (24) in Porcar-Castell, 2011).
As already shown above, the quantum yield of the photoinhibitory non-photochemical Chl fluorescence quenching (˚qI ) can be defined on the basis of the relaxation analysis of Chl fluorescence in darkness (Eq. (61)). However, the rationale of ˚qI of the photoinactivated centers formed by illumination was explicitly first defined in the framework of the Demmig-Adams theory (DemmigAdams et al., 1996) by Kornyeyev et al. (2001). The authors defined the P, D, and E parameters in the same way as in the original Demmig-Adams theory. Moreover, the D parameter (quantum yield of overall thermal dissipation) was suggested to be expressed as a sum of three quantum yields: DCON for constitutive always present dissipation, DREG for regulatory light-induced dissipation, and DPI for photoinhibitory dissipation, i.e., thermal dissipation in photoinactivated centers. Therefore: D = DCON + DREG + DPI .
(74)
−
w
F(t) . s FM
(75)
(80)
If DCON is the quantum yield for the only existing dissipation then it would be expressed as (ideal DCON ): ideal DCON = 1 −
F(t)w = , s FM w
F(t)w
(77)
F(t)w w , FM
˚w NPQr =
Kornyeyev et al. (2001)
The quantum yields of the constitutive non-regulatory (basal or dark) dissipation, ˚w , and of the total regulatory nonf,D
˚w f,D
F(t)w F(t)w − . s w FM FM
(71)
The first from the quantum yields for the light-adapted state when the actinic light is present is the actual quantum yield of PSII photochemistry for the light-adapted state which is identical to ˚P (Eq. (13)) but to distinguish that it was measured in winter, it is again denoted by superscript w:
FM
˚w NPQs =
Energy partitioning – consideration of photoinactivated centers, in addition to active centers
(72)
w
(76)
The quantum yield of the total regulatory non-photochemical quenching was then discriminated into the quantum yield of sustained regulatory non-photochemical quenching, ˚w , and of NPQs light-induced regulatory (called reversible in Porcar-Castell, 2011) : non-photochemical quenching, ˚w NPQr
(70)
w w qw L = qL,s qL,r .
w − F(t)w FM
w w ˚w P + ˚f,D + ˚NPQt = 1.
s − 1/FM
It further holds that:
˚w P =
It again holds that:
FV . FM
(81)
But because of the existence of the other dissipation processes, which affect DCON , the ideal DCON is corrected for existence of these other dissipations. To account for these processes, the ideal DCON is multiplied by the maximal quantum yield of PSII photochemistry
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
for the light-adapted state (˚PSII , Eq. (12)) relative to that one for the dark-adapted state (˚Po , Eq. (15)) which leads to: DCON =
) (1 − FV /FM )(FV /FM
FV /FM
.
(82)
Action of photoinhibition was estimated from a change in the ˚Po parameter measured after ((FV /FM )PI ) photoinhibition relative to that one measured before (FV /FM ) photoinhibition. Therefore, quantum yield for ideal photoinhibitory dissipation equals to: ideal DPI = 1 −
) (1 − (FV /FM )PI /(FV /FM ))(FV /FM
(FV /FM )PI
(83)
.
(84)
The DREG parameter is then found from an equation defining D as a sum of the free quantum yields (Eq. (80)) leading to: DREG = 1 −
) (FV /FM
(FV /FM )PI
.
(85)
As discussed later by Kornyeyev and Hendrickson (2007), the above defined DPI and DREG parameters reflect the quantum yields in light-adapted state in the absence of actinic light but the P parameter (˚P ) reflects the quantum yield in light-adapted state in the presence of actinic light. Consequently, also this theory as the original Demmig-Adams theory, suffers from mismatching the quantum yields for light-adapted states measured with and without presence of actinic light which results in necessity of consideration of the non-physiological E parameter. Hikosaka et al. (2004)
(FV /FM )PI , FV /FM
(86)
where (FV /FM )PI is the ˚Po parameter determined for the photoinhibited sample (means after the photoinhibition). Note that in the original paper by Hikosaka et al. (2004), the maximum quantum yield of PSII photochemistry of photoinactivated and of active centers are denoted as Fv /Fm , and Fv /FmM , respectively. Also note that B (Eq. (86)) of Hikosaka et al. (2004) is identical to ideal DPI (Eq. (83)) but different to DPI (Eq. (84)) of the theory by Kornyeyev et al. (2001). Consequently B of Hikosaka et al. (2004) expresses the quantum yield of heat dissipation by photoinactivated centers in dark-adapted state whereas DPI of Kornyeyev et al. (2001) reflects the quantum yield of heat dissipation by photoinactivated centers in light-adapted state but in the absence of actinic light (Kornyeyev and Hendrickson, 2007). Then the quantum yield of energy loss (thermal dissipation) in dark-adapted state, DD , was defined as:
DD = (1 − B) 1 −
FV FM
DL =
F V
FM
PI
−
FV
FM
.
(88)
PI
The other parameter, ˚P , denoted by Hikosaka et al. (2004) as C, was defined as usually (Eq. (13)). The E parameter (excess) then equals to:
(87)
(89)
that leads to the same formula for E (Eq. (29)) as defined by Demmig-Adams et al. (1996). Even if the approaches by Kornyeyev et al. (2001) and by Hikosaka et al. (2004) are improved versions of the original Demmig-Adams theory (Demmig-Adams et al., 1996) by consideration of the photoinactivated centers, they still operate with the E parameter and therefore these approaches do not seem to be the good ones. Hendrickson et al. (2005) The third approach, by Hendrickson et al. (2005), which considers the photoinactivated non-functional centers, is based on the theory previously presented by the same authors (Hendrickson et al., 2004). In the later approach, quantum yield of heat dissipation by the photoinactivated non-functional centers, ˚NF , is defined in the same way as B (Eq. (86)) in the theory by Hikosaka et al. (2004) which considered the photoinactivated non-functional centers. The other quantum yields in the theory by Hendrickson et al. (2005) are the same as in their original approach but modified with respect to existence of the photoinactivated centers, it means, the quantum yields are multiplied by the fraction of active centers described by the (FV /FM )PI /(FV /FM ), ratio. Therefore, ˚P =
The Demmig-Adams theory (Demmig-Adams et al., 1996) was also extended for explicit existence of the photoinactivated nonfunctional centers later on in another way by Hikosaka et al. (2004). Similarly as in the theory of Kornyeyev et al. (2001), Hikosaka et al. (2004) assumed that relative change in the ˚Po parameter (Eq. (15)) reflects a measure of photoinhibition, therefore, fraction of photoinactivated centers and quantum yield of utilization of excitation energy by these centers (as heat dissipation) as well, B, is expressed as: B=1−
and the quantum yield of thermal dissipation in light-adapted sample, DL , was defined as:
E = 1 − B − DD − DL − C,
(FV /FM )PI . FV /FM
Then, similarly as in the case of DCON , ideal DPI is corrected for existence of the other dissipative processes accounted for again by change of ˚PSII relative to ˚Po but measured after photoinhibition (i.e., expressed by (FV /FM )PI ) which leads to: DPI =
143
)(F /F ) (F/FM V M PI
˚f,D =
FV /FM
,
(90)
(F(t)/FM )(FV /FM )PI , FV /FM
˚NPQ =
− F(t)/F )(F /F ) (F(t)/FM M V M PI
FV /FM
(91) .
(92)
All the quantum yields sum to unity, that is: ˚NF + ˚P + ˚f,D + ˚NPQ = 1.
(93)
Kornyeyev and Hendrickson (2007) The fourth approach which considered the photoinactivated non-functional centers is by Kornyeyev and Hendrickson (2007). It is also based on the theory of Hendrickson et al. (2004) but the role of photoinactivation is implemented into the theory in a different way than in the work of Hendrickson et al. (2005). Kornyeyev and Hendrickson (2007) assumed a pool of PSII which contains both active and photoinactivated centers. Considering that the presence of photoinactivated non-functional centers can be reflected only in fluorescence levels obtained for light-adapted states, i.e., in FS , and by expressing the quantum yields of the Hendrickson and FM et al. (2004) theory by means of the involved rate constants, it was found that all the quantum yields (˚P , ˚f,D and ˚NPQ ; Eqs. (13), (35) and (36), respectively) of Hendrickson et al. (2004) are functions of active and photoinactivated centers as well and hence, can be used for description of energy partitioning in a pool of both active and photoinactivated centers. However, ˚NPQ of Hendrickson et al. (2004) was found to be in fact a sum of the quantum yield of regulatory non-photochemical quenching (in active centers), ˚NPQ , and of
144
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
the quantum yield of heat dissipation by the photoinactivated nonfunctional centers, ˚NF . Further analysis revealed that if FM in the original equation for ˚NPQ by Hendrickson et al. (2004) (Eq. (36)) is exchanged with (FM )PI (FM determined in darkness after previous illumination causing photoinactivation), then a correct equation for quantum yield of regulatory non-photochemical quenching is obtained and it is as follows: ˚NPQ
F(t) F(t) = − . FM (FM )PI
(94)
Based on the fact that all the quantum yields sum to unity, the quantum yield of heat dissipation by the photoinactivated nonfunctional centers, ˚NF , then equals to: ˚NF = 1 − ˚P − ˚NPQ − ˚f,D =
F(t) F(t) − . FM (FM )PI
(95)
It also implies from the summation of all the quantum yields to unity that sum of all the non-photochemical quantum yields can ratio. be expressed by the F(t)/FM Kornyeyev and Holaday (2008) The theory of Kornyeyev and Holaday (2008) is the most advanced theory existing so far for the energy partitioning occurring in active and photoinactivated centers. The authors analyzed all the previous theories which consider the photoinactivated nonfunctional centers and when necessary they performed related corrections of the parameters. Consequently, Kornyeyev and Holaday (2008) defined two sets of parameters of energy partitioning, one set for the case when the sample is in the light-adapted state and in the presence of actinic light (reaction centers of PSII are partially open and partially closed) and the other set for the case when the sample is in the light-adapted state but in the absence of actinic light (all reaction centers of PSII are open). As for the first case (the sample is in the light-adapted state and in the presence of actinic light), it was already noted by Kornyeyev and Hendrickson (2007) that DPI (Eq. (84)) and DREG (Eq. (85)) reflect the quantum yields in light-adapted state but in the absence of actinic light. To obtain equations for the quantum yields in the presence of actinic illumination, the formulae for the DPI (Eq. (84)) and DREG (Eq. (85)) must be multiplied by F(t)/F0 ratio which has already been mentioned before by Kornyeyev and Hendrickson (2007) for the case of the DPI parameter. Then, the corrected parameters are as follows:
˚NF =
(FV /FM )PI 1− FV /FM
˚NPQ =
1−
FV /FM
FV /FM
F(t) , (FV /FM )PI F0
(FV /FM )PI
F(t) . F0
(96)
(97)
As the above parameters apply to the case of the light-adapted state and in the presence of actinic light, the quantum yield of PSII photochemistry for this case is ˚P (Eq. (13)). The last quantum yield, ˚f,D , is obtained from the fact that all the quantum yields sum to unity and therefore: ˚f,D =
F /F F(t) F(t) F(t) + V M − . FM F0 FV /FM F0
(98)
For the second case when the sample is in the light-adapted state but in the absence of actinic light, the quantum yield of PSII photochemistry is expressed by ˚PSII (Eq. (12)) and the other quantum yields are as follows:
˚NF =
(FV /FM )PI 1− FV /FM
FV /FM
(FV /FM )PI
,
(99)
˚NPQ = 1 −
˚f,D =
FV /FM
(FV /FM )PI
,
) (1 − FV /FM )(FV /FM
FV /FM
(100)
.
(101)
It is important to stress that the above two sets of equations (Eqs. (96)–(98) vs. (99)–(101)) differ not only in the formulation of the quantum yields other than of photochemistry of PSII but also in the quantum yield of photochemistry of PSII itself; in the first case, Eq. (13) must be applied because actinic light is switched on and in the second case Eq. (12) must be used because actinic light is switched off (actinic light must be switched off for measurement of F0 ). Overall, the equations presented in Kornyeyev and Holaday (2008) represent the most comprehensive theory for energy partitioning considering the active and photoinactivated centers published so far. Kornyeyev and Holaday (2008) further noted that in the case of field measurements it is difficult or in principle impossible to measure correct ˚Po (=FV /FM , Eq. (15)) because the sample should not have any damage in the case of ˚Po measurement which is not met in the field conditions where the sample is exposed to various environmental stresses (e.g., photoinhibition, high temperature stress, drought stress, etc.). To avoid this problem, ˚Po should be replaced by its standard value of 0.8 in Eqs. (96) and (99). Knowledge of ˚Po is not necessary for calculation of Eqs. (97) and (100), the same as for calculation of Eqs. (13) and (12). Even if knowledge of ˚Po is further necessary for calculation of Eqs. (98) and (101), their numerical values can be obtained on the basis of the fact that sum of all the quantum yields in given set of equations equals to unity.
Energy partitioning – connection of the theories employing the quenching and relaxation analyses with the theories of active and photoinactivated centers Kornyeyev et al. (2013) Kornyeyev et al. (2013) analyzed the previous approaches to define a universal theory for energy partitioning. Generally, they compared two approaches which differ in the fact that determination of physiological state of the sample is based on changes in values of maximal fluorescence or on changes in values of maximal quantum yield of PSII photochemistry (i.e., when all reaction centers of PSII are open, irrespective if the sample is in darkadapted or light-adapted state or is in the recovery state). In the first case, Kornyeyev et al. (2013) used the fact that when the photoinactivated centers are considered and maximal fluorescence is measured in darkness after photoinhibitory illumination ((FM )PI ), it is in fact equivalent to measurement of maximal ). Then, after a fluorescence during the relaxation analysis (FM period of dark recovery, the quantum yield of the regulatory nonphotochemical quenching processes which relax, ˚qE+qT , and which do not relax (the quenching of photoinactivated centers), ˚qI , can be expressed from the theory of active and photoinactivated centers by Kornyeyev and Hendrickson (2007) as: ˚qE+qT =
˚qI =
F(t) F(t) − F , FM M
F(t) F(t) , − F FM M
(102)
(103)
which are rewritten Eqs. (94) and (95), respectively, with (FM )PI re . On the other hand, according to theory of active denoted as FM centers employing relaxation analysis of Ahn et al. (2009), the
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
quantum yields of qE quenching (Eq. (48)) and sum of qT and qI quenching (Eq. (49)) can be rewritten into (equal to): ˚qE =
F(t) F(t) − F , FM M
(104)
F(t) F(t) . − F FM M
˚qT +qI =
(105)
Looking at Eqs. (102)–(105), the right sides of Eqs. (102) and (104) are identical, the same as the right sides of Eqs. (103) and (105) but the time of dark relaxation (the threshold time) , which separates the groups of nonprior to measurements of FM photochemical quenching components, was different in the two approaches; 3 h for separation of sum of qE and qT quenching from qI quenching (Eqs. (102) and (103)) and 10 min for separation of qE quenching from sum of the qT and qI quenching (Eqs. (104) and (105)). By combination of the theories, the quantum yield of the qT quenching can be obtained by subtracting Eq. (104) from Eq. (102) which leads to: ˚qT =
F(t) 10 min FM
−
F(t)
,
3h FM
(106)
where superscripts “10 min” and “3 h” indicate that the maximal fluorescence was measured after 10 min and 3 h, respectively, of dark recovery of the sample. The previous equations (Eqs. (102)–(105)) were constructed based on assumption/existence of only one threshold time in the sample recovery considered in particular theories, however, Guadagno et al. (2010) revealed two threshold times in the dark recovery of the sample, based on which they defined Eqs. (59)–(61)) which can be rewritten into (equal to): ˚qE = ˚qT = ˚qI =
F(t) F(t) , − x FM FM F(t)
x FM
F(t) y
FM
− −
F(t)
(107)
,
(108)
F(t) , FM
(109)
y
FM
y
x and F where FM are the maximal fluorescence values of the first M and second, respectively, threshold times determined from the relaxation analysis (see above). Note that determination of the two threshold times might be distorted by analysis of the NPQ time course as found by Holzwarth et al. (2013); see above. Further, x is an analog of F 10 min and F y is an analog of assuming that FM M M 3h then Eq. (108) is identical with Eq. (106), the same as Eq. FM (107) with Eqs. (104) and (109) with Eq. (103). Then, from mathematical point of view, there is no difference if only one threshold time or two threshold times are considered in particular theories because the related equations are essentially the same in both the cases. However, the physiological meaning of the parameters is absolutely related to the length of dark recovery chosen for calculation. Therefore, the threshold times should be carefully determined for a particular study by preliminary measurements and by a cor -levels directly and not from NPQ values rect procedure (from FM -levels). calculated from FM - and FM In the case when maximal quantum yields of PSII photochemistry are used for determination of physiological state of the sample, during dark Kornyeyev et al. (2013) evaluated the course of FV /FM recovery and analyzed it in the same way as Guadagno et al. (2010) analyzed the NPQ course (see above). Kornyeyev et al. (2013) determined two threshold times reflecting variations in the slope of the course and denoted the related F /F values as (F /F )x FV /FM V M V M )y , (F /F )y being higher than (F /F )x because q and and (FV /FM T V M V M y qE quenching are relaxed when (FV /FM ) is measured whereas only
145
)x is measured. By compariqE quenching is relaxed when (FV /FM son of the FV /FM values and considering also the F(t)/F0 ratio to account for a closure of a fraction of PSII centers, following equations were introduced by Kornyeyev et al. (2013) for the sum of quantum yields of qE and qT quenching, and the quantum yield for the qE quenching itself:
˚qE+qT =
1−
˚qE =
1−
FV /FM
)y (FV /FM
FV /FM
) (FV /FM
x
F(t) , F0
(110)
F(t) . F0
(111)
The quantum yield of the qT quenching can be obtained by subtracting Eq. (111) from Eq. (110) which leads to:
˚qT =
FV /FM
) (FV /FM
x
−
FV /FM
) (FV /FM
y
F(t) . F0
(112)
The quantum yield of the qI quenching can be estimated from:
˚qI =
1−
) (FV /FM
FV /FM
y
FV /FM
)y (FV /FM
F(t) . F0
(113)
Kornyeyev et al. (2013) showed experimentally with beet leaves that the quantum yields of the non-photochemical quenching processes calculated according to the methods based on values of maximal fluorescence signals (Eqs. (107)–(109)) and based on values of maximal quantum yield of PSII photochemistry (Eqs. (111)–(113)) give very similar results and can be used as alternatives. The small deviations between the results were assigned to inaccuracy of the F0 measurements. Because of prac method is less sensitive to changes in the tical reasons (the FV /FM place of measurements on the leaf surface and experimental errors method), in measurements of the fluorescence levels than the FM method in Kornyeyev et al. (2013) recommended using the FV /FM the field. On the other hand, the equations used for definition of method lack physiologithe quantum yields based on the FV /FM cal background and the above mentioned good correlation of these quantum yields (Eqs. (111)–(113)) with the quantum yields based method (Eqs. (107)–(109)) might be accidental. on the FM Summary All the theories of the photosynthetic energy partitioning are in fact based on two original approaches: by Demmig-Adams et al. (1996) on one hand and by Genty et al. (1996) and Cailly et al. (1996) on the other hand. From these two approaches the first one seems to be not completely correct because it includes the E parameter which has no physiological meaning. Then, all the theories which are based on the Demmig-Adams approach (Kornyeyev et al., 2001; Kato et al., 2003; Hikosaka et al., 2004; Ishida et al., 2011) should be avoided because they carry the same problematic issue (presence of the E parameter) as the original one. On the other hand, the original approach by Genty et al. (1996) and Cailly et al. (1996), later re-discovered by Hendrickson et al. (2004), and the similar approach by Kramer et al. (2004), do not include the E parameter and are, together with their improved derivatives (Hendrickson et al., 2005; Kornyeyev and Hendrickson, 2007; Kornyeyev and Holaday, 2008; Porcar-Castell et al., 2008; Ahn et al., 2009; Kasajima et al., 2009; Guadagno et al., 2010; Porcar-Castell, 2011; Kornyeyev et al., 2013), thus more suitable for description of photosynthetic energy partitioning. From the above mentioned more suitable theories it is not possible to say which the only valid one is because they describe different states of the photosynthetic apparatus caused by different conditions (active centers only vs. active and inactive centers vs. centers with sustained non-photochemical quenching) and they
146
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
consider different levels of resolution of particular processes (total regulatory non-photochemical quenching vs. resolution of the qE and qT quenching, in addition to the qI quenching of the photoinactivated centers). The highest amount of information, i.e., resolution of the quantum yield of non-photochemical quenching into quantum yields of qE , qT , and qI quenching, is obtained when also the relaxation analysis is followed after the quenching analysis (Ahn et al., 2009; Kasajima et al., 2009; Guadagno et al., 2010). These measurements are, however, more time consuming when compared with the measurements of the quenching analysis alone, and therefore, they may not be suitable for some kind of experiments. Similarly, resolution of the photoinactivated centers (Kornyeyev et al., 2001; Hikosaka et al., 2004; Hendrickson et al., 2005; Kornyeyev and Hendrickson, 2007; Kornyeyev and Holaday, 2008) requires more time than the quenching analysis alone because the FV /FM ratio must be measured in darkness after the light treatment, which might require up to several hours of darkness. However, if a high intensity of excitation light is used for illumination, promoting formation of photoinactived centers, these theories are the only way for correct determination of photosynthetic energy partitioning. When overwintering evergreens are explored, the two theories by Porcar-Castell and coworkers (PorcarCastell et al., 2008; Porcar-Castell, 2011) can be used from which the later theory of Porcar-Castell (2011) is the better one because it provides in addition to the quantum yield of the sustained regulatory non-photochemical quenching also information about the quantum yield of the light-induced regulatory non-photochemical quenching. On the other hand, if the researcher needs only the basic information about the photosynthetic energy partitioning and/or there is no time to measure dark-relaxed maximal fluorescence, the two theories (Genty et al., 1996; Cailly et al., 1996; Hendrickson et al., 2004 vs. Kramer et al., 2004) are enough to be used. In summary, many theories of the energy partitioning in photosynthesis have been developed so far and a researcher can find the approach which is the most suitable for his/her requirements and possibilities. Acknowledgements This work was supported by the Czech National Program of Sustainability I, grant number LO1204. D.L. thanks the anonymous reviewers for their useful comments. References Ahn TK, Avenson TJ, Peers G, Li Z, Dall’Osto L, Bassi R, et al. Investigating energy partitioning during photosynthesis using an expanded quantum yield convention. Chem Phys 2009;357:151–8. Baker NR. Chlorophyll fluorescence: a probe of photosynthesis in vivo. Annu Rev Plant Biol 2008;59(89):113. Bilger W, Björkman O. Role of the xanthophyll cycle in photoprotection elucidated by measurements of light-induced absorbance changes, fluorescence and photosynthesis in leaves of Hedera canariensis. Photosynth Res 1990;25:173–85. Bilger W, Schreiber U. Energy-dependent quenching of dark-level chlorophyll fluorescence in intact leaves. Photosynth Res 1986;10:303–8. Bradbury M, Baker NR. Analysis of the slow phases of the in vivo chlorophyll fluorescence induction curve. Changes in the redox state of photosystem II electron acceptors and fluorescence emission from photosystems I and II. Biochim Biophys Acta 1981;13:542–51. Cailly AL, Rizzal F, Genty B, Harbinson J. Fate of excitation at PS II in leaves, the nonphotochemical side. Plant Physiol Biochem 1996;86 [special issue]. Chen C, Zhang D, Li P, Ma F. Partitioning of absorbed light energy differed between the sun-exposed side and the shaded side of apple fruits under high light conditions. Plant Physiol Biochem 2012;60:12–7. Demmig-Adams B, Adams WW, Barker DH, Logan BA, Bowling BA, Verhoeven AS. Using chlorophyll fluorescence to assess the fraction of absorbed light allocated to thermal dissipation of excess excitation. Physiol Plant 1996;98:253–64. Demmig-Adams B, Garab G, Govindjee Adams W III, editors. Non-photochemical quenching and energy dissipation in plants, algae and cyanobacteria. Dordrecht: Kluwer; 2014. Duysens LNM, Sweers HE. Mechanism of the two photochemical reactions in algae as studied by means of fluorescence. In: Japanese society of plant physiologists,
editor. Studies on microalgae and photosynthetic bacteria. Tokyo: University of Tokyo Press; 1963. p. 353–72. Earl HJ, Ennahli S. Estimating photosynthetic electron transport via chlorophyll fluorometry without photosystem II light saturation. Photosynth Res 2004;82:177–86. Genty B, Briantais J-M, Baker NR. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochim Biophys Acta 1989;990:87–92. Genty B, Harbinson J, Cailly AL, Rizza F. Fate of excitation at PS II in leaves: the non-photochemical side. In: Book of abstract from the third BBSRC Robert Hill symposium on photosynthesis; 1996, Abstract no. P28. Goss R, Lepetit B. Biodiversity of NPQ. J Plant Physiol 2014;172:13–32. Govindjee. Sixty-three years since Kautsky: chlorophyll a florescence. Aust J Plant Physiol 1995;22:131–60. Guadagno CR, Virzo De Santo A, D’Ambrosio N. A revised energy partitioning approach to assess the yields of non-photochemical quenching components. Biochim Biophys Acta 2010;1797:525–30. Harbinson J. Improving the accuracy of chlorophyll fluorescence measurements. Plant Cell Environ 2013;36:1751–4. Havaux M, Strasser RJ, Greppin H. A theoretical and experimental analysis of the qP and qN coefficients of chlorophyll fluorescence quenching and their relation to photochemical and nonphotochemical events. Photosynth Res 1991;27:41–55. Hendrickson L, Furbank RT, Chow WS. A simple alternative approach to assessing the fate of absorbed light energy using chlorophyll fluorescence. Photosynth Res 2004;82:73–81. Hendrickson L, Föerster B, Pogson BJ, Chow WS. A simple chlorophyll fluorescence parameter that correlates with the rate coefficient of photoinactivation of photosystem II. Photosynth Res 2005;84:43–9. Henriques FS. Leaf chlorophyll fluorescence: background and fundamentals for plant biologists. Bot Rev 2009;75:249–70. Hikosaka K, Kato MC, Hirose T. Photosynthetic rates and partitioning of absorbed light energy in photoinhibited leaves. Physiol Plant 2004;121:699–708. Hodges M, Cornic G, Briantais J-M. Chlorophyll fluorescence from spinach leaves: resolution of non-photochemical quenching. Biochim Biophys Acta 1989;974:289–93. Holt NE, Zigmantas D, Valkunas L, Li X-P, Niyogi KK, Fleming GR. Carotenoid cation formation and the regulation of photosynthetic light harvesting. Science 2005;307:433–6. Holzwarth AR, Miloslavina Y, Nilkens M, Jahns P. Identification of two quenching sites active in the regulation of photosynthetic light-harvesting studied by timeresolved fluorescence. Chem Phys Lett 2009;483:262–7. Holzwarth AR, Lenk D, Jahns P. On the analysis of non-photochemical chlorophyll fluorescence quenching curves I. Theoretical considerations. Biochim Biophys Acta 2013;1827:786–92. Horton P. Relations between electron transport and carbon assimilation; simultaneous measurement of chlorophyll fluorescence, transthylakoid pH gradient and O2 evolution in isolated chloroplasts. Proc R Soc Lond B 1983;217:405–16. Horton P, Hague A. Studies on the induction of chlorophyll fluorescence in isolated barley protoplasts IV. Resolution of non-photochemical quenching. Biochim Biophys Acta 1988;932:107–15. Ishida S, Merita K, Kishine M, Takabayashi A, Murakami R, Takeda S, et al. Allocation of absorbed light energy in PSII to thermal dissipations in the presence or absence of PsbS subunits of rice. Plant Cell Physiol 2011;52:1822–31. ´ Jabłonski A. Efficiency of anti-stokes fluorescence in dyes. Nature 1933;131:839–40. Johnson MP, Ruban A. Restoration of rapidly reversible photoprotective energy dissipation in the absence of PsbS protein by enhanced pH. J Biol Chem 2011;286:19973–81. Joliot A, Joliot P. Étude cinétique de la réaction photochimique libérant l‘oxygéne au cours de la photosynthése. C R Acad Sci Paris 1964;258:4622–5. Kasajima I, Takahara K, Kawai-Yamada M, Uchimiya H. Estimation of the relative sizes of rate constants for chlorophyll de-excitation processes through comparison of inverse fluorescence intensities. Plant Cell Physiol 2009;50:1600–16. Kato MC, Hikosaka K, Hirotsu N, Makino A, Hirose T. The excess light energy that is neither utilized in photosynthesis nor dissipated by photoprotective mechanisms determines the rate of photoinactivation in photosystem II. Plant Cell Physiol 2003;44:318–25. Kautsky H, Hirsch A. Neue Versuche zur Kohlensureassimilation. Naturwissenschafte 1931;19:964. Kitajima M, Butler WL. Quenching of chlorophyll fluorescence and primary photochemistry in chloroplasts by dibromothymoquinone. Biochim Biophys Acta 1975;376:105–15. Klughammer C, Schreiber U. Complementary PS II quantum yields calculated from simple fluorescence parameters measured by PAM fluorometry and the saturation pulse method. PAM Appl Notes 2008;1:27–35. Kornyeyev D, Hendrickson L. Energy partitioning in photosystem II complexes subjected to photoinhibitory treatment. Funct Plant Biol 2007;34:214–20. Kornyeyev D, Holaday AS. Corrections to current approaches used to calculate energy partitioning in photosystem 2. Photosynthetica 2008;46:170–8. Kornyeyev D, Logan BA, Patron P, Allen RD, Holaday AS. Enhanced photochemical light utilization and decreased chilling-induced photoinhibition of photosystem II in cotton overexpressing genes encoding chloroplast-targeted antioxidant enzymes. Physiol Plant 2001;113:323–31. Kornyeyev D, Guadagno CR, D’Ambrosio N. Assessment of energy partitioning in PSII complexes using chlorophyll fluorescence: reviewing the different approaches toward the definition of a unified method. Photosynthetica 2013;51: 477–80.
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147 Kramer DM, Johnson G, Kiirats O, Edwards GE. New fluorescence parameters for the determination of QA redox state and excitation energy fluxes. Photosynth Res 2004;79:209–18. Krause HG, Jahns P. Pulse amplitude modulated chlorophyll fluorometry and its application in plant science. In: Green BR, Parson WW, editors. Light-harvesting antennas in photosynthesis. Dordrecht: Kluwer; 2003. p. 373–99. Laisk A, Oja V, Rasulov B, Eichelmann H, Sumberg A. Quantum yields and rate constants of photochemical and nonphotochemical excitation quenching. Experiment and model. Plant Physiol 1997;11:803–15. Lavergne J, Leci E. Properties of inactive photosystem II centers. Photosynth Res 1993;35:323–43. Lazár D. Chlorophyll a fluorescence induction. Biochim Biophys Acta 1999;1412:1–28. Lazár D. The polyphasic chlorophyll a fluorescence rise measured under high intensity of exciting light. Funct Plant Biol 2006;33:9–30. Lazár D. Simulations show that a small part of variable chlorophyll a fluorescence originates in photosystem I and contributes to overall fluorescence rise. J Theor Biol 2013;335:249–64. Lazár D, Schansker G. Models of chlorophyll a fluorescence transients. In: Laisk A, Nedbal L, Govindjee, editors. Photosynthesis in silico: understanding complexity from molecules to ecosystems. Dordrecht: Springer; 2009. p. 85–123. Li X-P, Björkman O, Shih C, Grossmann AR, Rosenquist M, Jansson S, et al. A pigmentbinding protein essential for regulation of photosynthetic light harvesting. Nature 2000;403:391–5. Lichtenthaler HK, Buschmann C, Rinderle U, Schmuck G. Application of chlorophyll fluorescence in ecophysiology. Radiat Environ Biophys 1986;25:297–308. Lichtenthaler HK, Langsdorf G, Lenk S, Buschmann C. Chlorophyll fluorescence imaging of photosynthetic activity with the flash-lamp fluorescence imaging system. Photosynthetica 2005;43:355–69. Loriaux SD, Avenson TJ, Welles JM, Mcdermitt DK, Eckles RD, Riensche B, et al. Closing in on maximum yield of chlorophyll fluorescence using a single multiphase flash of sub-saturating intensity. Plant Cell Environ 2013;36:1755–70. Maxwell K, Johnson GN. Chlorophyll fluorescence – a practical guide. J Exp Bot 2000;51:659–68. Miyake C, Amako K, Shiraishi N, Sugimoto T. Acclimation of tobacco leaves to high light intensity drives the plastoquinone oxidation system – relationship among the fraction of open PSII centers, non-photochemical quenching of chl fluorescence and the maximum quantum yield of PSII in the dark. Plant Cell Physiol 2009;50:730–43. Nilkens M, Kress E, Lambrev P, Miloslavina Y, Müller M, Holzwarth AR, et al. Identification of a slowly inducible zeaxanthin-dependent component of nonphotochemical quenching of chlorophyll fluorescence generated under steadystate conditions in Arabidopsis. Biochim Biophys Acta 2010;1792:466–75. Oxborough K, Baker NR. Resolving chlorophyll a fluorescence images of photosynthetic efficiency into photochemical and non-photochemical components without measuring F0 . Photosynth Res – calculation of qP and FV /FM 1997;54:135–42. Papageorgiou G, Govindjee. Light-induced changes in the fluorescence yield of chlorophyll a in vivo II. Chlorella pyrenoidosa. Biophys J 1968;8:1316–28. Papageorgiou GC, Tsimilli-Michael M, Stamatakis K. The fast and slow kinetics of chlorophyll a fluorescence induction in plants, algae and cyanobacteria: a viewpoint. Photosynth Res 2007;94:275–90. Porcar-Castell A. A high-resolution portrait of the annual dynamics of photochemical and non-photochemical quenching in needles of Pinus sylvestris. Physiol Plant 2011;143:139–53. Porcar-Castell A, Juurola E, Nikinmaa E, Berninger F, Ensminger I, Hari P. Seasonal acclimation of photosystem II in Pinus sylvestris I. Estimating the rate constants of sustained thermal energy dissipation and photochemistry. Tree Physiol 2008;28:1474–82. Porcar-Castell A, Garcia-Plazaola JI, Nichol CJ, Kolari P, Olascoaga B, Kuusinen N, et al. Physiology of the seasonal relationship between the photochemical reflectance index and photosynthetic light use efficiency. Oecologia 2012;170:313–23.
147
Porcar-Castell A, Tyystjärvi E, Atherton J, van der Tol C, Flexas J, Pfündel EE, et al. Linking chlorophyll a fluorescence to photosynthesis for remote sensing applications: mechanisms and challenges. J Exp Bot 2014;65:4065–95. Quick WP, Horton P. Studies on the induction of chlorophyll fluorescence in barley protoplasts II: resolution of fluorescence quenching by redox state and the transthylakoid pH gradient. Proc R Soc Lond B 1984;220:371–82. Quick WP, Stitt M. An examination of factors contributing to non-photochemical quenching of chlorophyll fluorescence in barley leaves. Biochim Biophys Acta 1989;977:287–96. Roháˇcek K. Chlorophyll fluorescence parameters: the definitions, photosynthetic meaning, and mutual relationships. Photosynthetica 2002;40:13–29. Roháˇcek K. Method for resolution and quantification of components of the nonphotochemical quenching (qN ). Photosynth Res 2010;105:101–13. Rubin A, Riznichenko G. Modeling of the primary processes in a photosynthetic membrane. In: Laisk A, Nedbal L, Govindjee, editors. Photosynthesis in silico: understanding complexity from molecules to ecosystems. Dordrecht: Springer; 2009. p. 151–76. Samson G, Bruce D. Origins of the low yield of chlorophyll a fluorescence induced by single turnover flash in spinach thylakoids. Biochim Biophys Acta 1996;1276:147–53. Samson O, Práˇsil O, Yaakoubd B. Photochemical and thermal phases of chlorophyll a fluorescence. Photosynthetica 1999;37:163–82. Schansker G, Tóth SZ, Strasser RJ. Dark recovery of the Chl a fluorescence transient (OJIP) after light adaptation: the qT-component of non-photochemical quenching is related to an activated photosystem I acceptor side. Biochim Biophys Acta 2006;1757:787–97. Schansker G, Tóth SZ, Holzwarth AR, Garab G. Chlorophyll a fluorescence: beyond the limits of the QA model. Photosynth Res 2014;120:43–58. Schreiber U, Schliwa U, Bilger W. Continuous recording of photochemical and non-photochemical chlorophyll fluorescence quenching with a new type of modulation fluorometer. Photosynth Res 1986;10:51–62. Govindjee Stirbet A. Chlorophyll a fluorescence induction: a personal perspective of the thermal phase, the J-I-P rise. Photosynth Res 2012;113:15–61. Stirbet A. Excitonic connectivity between photosystem II units: what is it, and how to measure it? Photosynth Res 2013;116:189–214. Stirbet A, Govindjee. On the relation between the Kautsky effect (chlorophyll a fluorescence induction) and photosystem II: basics and applications of the OJIP fluorescence transient. J Photochem Photobiol B 2011;104:236–57. Stirbet A, Riznichenko GY, Rubin AB, Govindjee. Modeling chlorophyll a fluorescence transient: relation to photosynthesis. Biochemistry (Moscow) 2014;79:291–323. van Kooten O, Snel J. The use of chlorophyll nomenclature in plant stress physiology. Photosynth Res 1990;25:147–50. Verhoeven A. Sustained energy dissipation in winter evergreens. New Phytol 2014;201:57–65. Vredenberg W, Práˇsil O. Modeling of chlorophyll a fluorescence kinetics in plant cells: derivation of a descriptive algorithm. In: Laisk A, Nedbal L, Govindjee, editors. Photosynthesis in silico: understanding complexity from molecules to ecosystems. Dordrecht: Springer; 2009. p. 125–49. Walker DA, Sivak MN, Prinsley RT, Cheesbrough JK. Simultaneous measurement of oscillations in oxygen evolution in chlorophyll a fluorescence in leaf pieces. Plant Physiol 1983;73:542–9. Walters RG, Horton P. Resolution of components of non-photochemical chlorophyll fluorescence quenching in barley leaves. Photosynth Res 1991;27:121–33. Walters RG, Horton P. Theoretical assessment of alternative mechanisms for nonphotochemical quenching of PS II fluorescence in barley leaves. Photosynth Res 1993;36:119–39. Yamagishi A, Satoh K, Katoh S. Fluorescence induction in chloroplasts isolated from the green alga Bryopsis maxima III. A fluorescence transient indicating proton gradient across the thylakoid membrane. Plant Cell Physiol 1978;19: 17–25.
Contents lists available at ScienceDirect
Journal of Plant Physiology journal homepage: www.elsevier.com/locate/jplph
Review article
Parameters of photosynthetic energy partitioning Duˇsan Lazár ∗ Department of Biophysics, Center of the Region Haná for Biotechnological and Agricultural Research, Faculty of Science, Palack´ y University, Sˇ lechtitel˚ u 11, 783 71 Olomouc, Czech Republic
a r t i c l e
i n f o
Article history: Received 27 August 2014 Received in revised form 22 October 2014 Accepted 25 October 2014 Available online 29 November 2014 Keywords: Quantum yield Photochemical Non-photochemical Chlorophyll fluorescence Photosynthesis
a b s t r a c t Almost every laboratory dealing with plant physiology, photosynthesis research, remote sensing, and plant phenotyping possesses a fluorometer to measure a kind of chlorophyll (Chl) fluorescence induction (FLI). When the slow Chl FLI is measured with addition of saturating pulses and far-red illumination, the so-called quenching analysis followed by the so-called relaxation analysis in darkness can be realized. These measurements then serve for evaluation of the so-called energy partitioning, that is, calculation of quantum yields of photochemical and of different types of non-photochemical processes. Several theories have been suggested for photosynthetic energy partitioning. The current work aims to summarize all the existing theories, namely their equations for the quantum yields, their meaning and their assumptions. In the framework of these theories it is also found here that the well-known NPQ parameter (= (FM − FM )/FM ; Bilger and Björkman, 1990) equals the ratio of the quantum yield of regulatory light-induced non-photochemical quenching to the quantum yield of constitutive non-regulatory non-photochemical quenching (˚NPQ /˚f,D ). A similar relationship is also found here for the PQ parameter (˚P /˚f,D ). © 2014 Elsevier GmbH. All rights reserved.
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chl fluorescence and quantum yields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pulse amplitude modulation technique for measurement of Chl fluorescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The slow Chl FLI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Saturation pulse method for determination of Chl fluorescence quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The relaxation analysis following the quenching analysis to discriminate particular types of non-photochemical quenching . . . . . . . . . . . . . . . . . . . . Energy partitioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – employing the quenching analysis and consideration of active centers only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Demmig-Adams et al. (1996) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kato et al. (2003) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ishida et al. (2011) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kramer et al. (2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – employing the quenching and relaxation analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ahn et al. (2009) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kasajima et al. (2009) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Guadagno et al. (2010) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – discrimination of the sustained regulatory non-photochemical quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Porcar-Castell et al. (2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Porcar-Castell (2011) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – consideration of photoinactivated centers, in addition to active centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Tel.: +420 58 563 4164. E-mail address: [email protected] http://dx.doi.org/10.1016/j.jplph.2014.10.021 0176-1617/© 2014 Elsevier GmbH. All rights reserved.
132 132 133 133 134 136 138 138 138 138 138 139 139 140 140 140 141 141 141 142 142
132
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
Kornyeyev et al. (2001) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hikosaka et al. (2004) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hendrickson et al. (2005) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kornyeyev and Hendrickson (2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kornyeyev and Holaday (2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Energy partitioning – connection of the theories employing the quenching and relaxation analyses with the theories of active and photoinactivated centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kornyeyev et al. (2013) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
142 143 143 143 144 144 144 145 146 146
Introduction Chlorophyll (Chl) fluorescence induction (FLI) is one of the most frequently used methods of all Chl fluorescence measurements. Therefore, it is widely and extensively used not only in plant physiology and photosynthesis research, but also in remote sensing and plant phenotyping. The Chl FLI was first observed by Kautsky and Hirsch (1931) by their own eyes; the curve constructed based on their observation started with a fast fluorescence rise (FLR) followed by a slow fluorescence decrease (FLD) to a steady state. The FLR and FLD are also known as the fast and the slow phases of FLI, respectively (e.g., Govindjee, 1995). Later, with improvements in fluorometers, use of the Chl FLI was in fact split into two types of measurements: (i) measurement of the Chl FLR with a high time resolution and usually under high intensity of excitation light (the O-J-I-P curve) and (ii) measurements of whole Chl FLI for a long time and applying different sources of excitation light (the quenching analysis). These two types of measurements are also related to different excitation and detection of fluorescence signal (see below). The fast Chl FLR was reviewed, e.g., by Lazár (2006), Stirbet and Govindjee (2011, 2012), and Schansker et al. (2014), while the slow Chl FLD and Chl FLI were reviewed, e.g., by Lazár (1999), Maxwell and Johnson (2000), Papageorgiou et al. (2007), Baker (2008), and Porcar-Castell et al. (2014). The Chl FLI was also theoretically modeled and this effort was reviewed by Lazár and Schansker (2009), Vredenberg and Práˇsil (2009), Rubin and Riznichenko (2009), and Stirbet et al. (2014). In the past two decades, a lot of effort has been made to define the so-called photosynthetic energy partitioning, i.e., to define parameters (quantum yields) which will reflect the fraction of absorbed light energy utilized by particular processes (photochemistry vs. non-photochemical dissipations). Irrespective of many publications on this topic, there exists no review article that summarizes all of the derived parameters. Therefore, a researcher can be confused by the existence of so many parameters, their interpretation and the assumptions under which the derivations are made. The reader thus should be introduced and correctly informed about differences among the parameters and their proper usage. Even experts are not familiar with all of the existing theories. It is, therefore, the main aim of this review to contribute a summary of all known approaches and parameters of photosynthetic energy partitioning. In the current review, the basics of the Chl fluorescence and quantum yields are described first, followed by a description of the measuring techniques together with explanation of basic fluorescence levels and quenching coefficients and parameters, followed by detailed description and discussion of the parameters (quantum yields) of the photosynthetic energy partitioning. A side product of reviewing the theories is that it is found here that the well-known NPQ and PQ parameters are equal to a ratio of two quantum yields (Eqs. (38) and (39); see below). The topic of this review is very closely related to the mechanisms of the light-induced regulatory non-photochemical Chl
Fig. 1. A scheme of energetic levels (the so-called Jablonski diagram) of a single molecule showing routes of formation and deactivation of the energetic levels. The thin horizontal lines show vibrational energy levels which are superimposed on electronic energy levels of the ground state (S0 ), excited singlet states (S1 and S2 ) and excited triplet state (T1 ) shown by thick horizontal lines. Vertical thick arrows symbolically show orientation of spin; when the arrows in the ground and exited states are in the opposite/same directions, the singlet/triplet excited state is realized. Vertical wavy arrows indicate radiative transitions by means of absorption of light (A) of different wavelengths (shorter wavelength (blue) means higher energy whereas longer wavelength (orange) means lower energy) and by means of emission of light by fluorescence (F; S1 –S0 transition) and by phosphorescence (P; T1 –S0 transition). Thin vertical or skew arrows indicate non-radiative transitions by means of vibrational relaxation (VR), internal conversion (IC), and inter-system crossing (ISC) which all represent a loss of energy by heat dissipation. For the case of a photosynthetic pigment molecule, additional deactivations of the excited states not shown in the figure can occur: via energy transfer to another molecule and via photochemistry (the primary charge separation). Each of all the above mentioned processes is characterized by its rate constant (not shown in the figure). In the dark-adapted state of healthy non-stressed sample the non-radiative transitions are the origins of the basal constitutive non-regulatory heat dissipations characterized by its overall apparent rate constant kD (see the text and Fig. 2). In a light-adapted state of the sample, light-induced changes (increase) in the values of the rate constants of the non-radiative transitions occur (with respect to their values in the dark-adapted state) which cause the light-induced regulatory heat dissipation characterized by its overall apparent rate constant kNPQ (see the text and Fig. 2). When the sustained regulatory non-photochemical quenching is present, it also affects the non-radiative transitions of the dark-adapted state (increase of kD ). The same is true when a photoinhibition occurs.
fluorescence quenching as such. Much discussion is still ongoing on these mechanisms. As this is beyond of the scope of this review, it does not deal with these mechanisms and the readers are referred to a recent review on the biodiversity of the non-photochemical Chl fluorescence quenching (Goss and Lepetit, 2014) and to a recent book on all aspects of the non-photochemical Chl fluorescence quenching edited by Demmig-Adams et al. (2014). Chl fluorescence and quantum yields To access the photosynthetic energy partitioning, the Chl fluorescence signal is used. By definition, fluorescence is a radiative deactivation of an excited singlet state (usually the first singlet
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
133
particular processes and therefore their values can be directly compared, opposite to other parameters (see below) whose relative changes can only be compared. Pulse amplitude modulation technique for measurement of Chl fluorescence
Fig. 2. Schematic representation of basic partitioning of incoming light energy occurring in photosystem II into photochemistry (characterized by its effective rate constant kP ; kP = kcs qP , where kcs is the intrinsic rate constant of charge separation in PSII and qP estimates fraction of the open PSII (Eqs. (4) and (6))), fluorescence emission (kf ), basal constitutive non-regulatory heat dissipation (kD ), and light-induced regulatory heat dissipation (kNPQ ). In a context of theories which do not consider the sustained non-photochemical quenching and the non-photochemical quenching of the photoinactivated centers, the light-induced regulatory non-photochemical dissipations is reversible and causes the qE and qT , and qI quenching. However, in the theories which consider the sustained non-photochemical quenching and the non-photochemical quenching of the photoinactivated centers, only the qE and qT quenching parts of the light-induced regulatory non-photochemical dissipation are considered as reversible and the qI quenching is considered as irreversible and, in fact, being the sustained non-photochemical quenching and/or reflecting formation of the photoinactivated centers.
state S1 ) of a molecule into the ground state as it is described by ´ what is now called the Jablonski diagram (Jabłonski, 1933) – see Fig. 1. In photosynthesis, Chl fluorescence changes upon illumination of a sample between minimal and maximal fluorescence levels. Thus changing Chl fluorescence is called the variable Chl fluorescence (see also the text below). It is generally assumed that the variable Chl fluorescence can be emitted only by photosystem II (PSII; reviewed in Lazár, 1999, 2006; Maxwell and Johnson, 2000; Papageorgiou et al., 2007; Baker, 2008; Stirbet and Govindjee, 2011, 2012; Schansker et al., 2014; Porcar-Castell et al., 2014). However, it was shown that photosystem I (PSI) can also, in principle, emit variable Chl fluorescence (Lazár, 2013 and references therein), but this fluorescence seems to be small when compared with variable PSII fluorescence. Emission of Chl fluorescence is only one route of usage of incoming (absorbed) light energy by PSII. This is schematically shown in Fig. 2 (see also Fig. 1). From a view of Chl fluorescence emission, the other routes represent competing processes that cause quenching of Chl fluorescence; Chl fluorescence is quenched by photochemistry and by non-photochemical processes. There are basically two types of non-photochemical quenching: basal constitutive non-regulatory heat dissipation and light-induced regulatory heat dissipation, see Fig. 2 and the legend of Fig. 1. A quantum yield of a given process, ˚i , is calculated as: ˚i =
ki , ˙k
(1)
where ki is the rate constant of the given process and ˙k is the sum of all the rate constants of all the processes of deactivation of the incoming light energy. As it is usually assumed that the rate constants kf and kD (see Fig. 2) are invariant in time and that they both represent basal always present non-regulatory types of light energy deactivation, they are grouped into one rate constant kf,D and a term basal constitutive non-regulatory dissipation is used. A quantum yield is a number in the interval 0, 1 and it holds that: ˙˚i = 1.
(2)
The advantage of an evaluation of the quantum yields of the photosynthetic energy partitioning is that the yields have the meaning of fractions of the usage of the incoming light energy by
The Chl FLI is mostly measured using the pulse amplitude modulation (PAM) technique. The PAM technique was commercially introduced by Schreiber et al. (1986) based on the work of Horton (1983) and Quick and Horton (1984). In the PAM technique, the measurement of Chl fluorescence is achieved by utilizing two light sources during the measurements: a light that drives photosynthesis and a light which excites (and “measure”) fluorescence. The light which drives photosynthesis can be continuous actinic light and/or a short (usually up to 1 s) saturating pulse or far-red continuous illumination (to oxidize the electron transport chain). The measuring light is in fact very short flashes (about 1–5 s long) whose intensity is always claimed in the literature to be very weak but this is not completely true. When necessary (measurement of minimal fluorescence – see below), the measuring flashes are applied with a low frequency, which leads to a low integrated intensity of excitation light of these measuring flashes. Therefore, the measuring flashes only “measure” fluorescence and do not drive photosynthesis (ideally). On the other hand, in case of background actinic or saturating illumination, the measuring flashes are applied with a high frequency to detect the fast fluorescence changes caused by the actinic/saturating illumination and, in this case, the resulting higher integrated energy of the measuring flashes is not a problem because the background actinic/saturating illumination to drive photosynthesis is already applied. During every measuring flash, the fluorescence signal is detected by a detector (photodiode) featuring a fast response and large linearity range (up to 109 ). The necessity of the first feature is because of the ability to detect fast changes in fluorescence signal and the second feature is necessary because the fluorescence signal is also detected a few microseconds after the measuring flash, i.e., either in the case of a darkness (measurement of minimal fluorescence – see below) when no fluorescence caused by actinic/saturating illumination driving photosynthesis is emitted, or in the case of the actinic/saturating illumination, when a high fluorescence is emitted. Then, the difference between the fluorescence signal detected during the measuring flash and fluorescence signal detected after the measuring flash is the output of the fluorometer. Therefore, the output fluorescence signal reflects fluorescence intensity but it is not directly proportional to the intensity of light used to drive photosynthesis, in contrast to the non-PAM fluorometers. This fluorescence intensity, however, is directly proportional to the quantum yield of Chl fluorescence at a given time. This fluorescence signal is sometimes called modulated fluorescence to convey that it is measured by the PAM technique. Because the modulated fluorescence represents the difference of the two signals, an effect of a possible background light (detected immediately after the measuring flash), e.g., when the measurement is performed in the field, is eliminated, which is the advantage of the PAM technique. At present, fluorometers made by several companies can measure Chl fluorescence utilizing the PAM technique. The slow Chl FLI Before evaluation of fluorescence quenching and energy partitioning, the FLI was measured. Typical slow FLIs are shown in Fig. 3. After peak P either one (M) or two (M1 , M2 ) maxima can appear in the FLD and sometimes, under special conditions, even oscillations can be measured until a steady-state terminal level
134
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
Fig. 3. Typical slow Chl FLI measured with dark-adapted pea leaves under 40 mol m−2 s−1 of red (maximum at about 650 nm) light. The O, P, S, M and T steps (curve a, shifted by 50 s to the right) and the O, P, S1 , M1 , S2 , M2 and T steps (curve b) are denoted. Data were taken from Lazár (1999).
T is reached (e.g., Papageorgiou and Govindjee, 1968; Yamagishi et al., 1978; Walker et al., 1983). With respect to the photosynthetic function, due to a long time scale of the measurements (about 10–30 min), the slow FLI reflects not only the light-induced electron transport reactions occurring in the thylakoid membrane and the non-photochemical Chl fluorescence quenching, but also the dark photosynthetic carbon reactions (Calvin–Benson cycle). The decrease of fluorescence intensity from P to T was used to evaluate the so-called vitality index, Rfd , as follows (Lichtenthaler et al., 1986): Rfd =
FP − FT , FT
(3)
where FP and FT denote fluorescence intensities at the P and T steps, respectively. When a plant is stressed, FP decreases and FT increases, hence Rfd decreases (e.g. Lichtenthaler et al., 2005). Alternatively, a stressed-induced increase in the non-photochemical Chl fluorescence quenching causes more pronounced decrease of the FP -level than of the FT -level leading again to the decrease of Rfd . Saturation pulse method for determination of Chl fluorescence quenching The FLD from P to T can be generally understood to be caused by different types of fluorescence quenching, either of photochemical or of non-photochemical nature. In addition to continuous actinic light, saturating pulses (usually 1 s duration with intensity of excitation light of about 10,000 mol photons of PAR m−2 s−1 ) are applied during the FLI to reveal and quantify the quenching. A typical measured curve is shown in Fig. 4, where particular fluorescence levels are also denoted. The technique was first introduced by Bradbury and Baker (1981) and spread in the course of time and became popular due to the availability of a commercial fluorometer, PAM 101 by Walz (Effeltrich, Germany) (Schreiber et al., 1986) followed by even advanced and more compact fluorometers by Walz, Hansatech (Norfolk, UK), Photon Systems Instruments (Brno, Czech Republic), and by other factories. Together with advancement in the instrumentation, the measuring approaches were also improved by application of far-red light (usually about several seconds) together with switching off of the continuous actinic light (van Kooten and Snel, 1990). The far-red light is predominantly absorbed by PSI, causing oxidation of the electron transport chain in front of PSI,
Fig. 4. Typical quenching analysis followed by the relaxation analysis measured with dark-adapted mature tobacco leaf under 200 mol m−2 s−1 of red (maximum at about 650 nm) actinic light. The time interval starting with blue arrow up and ending with blue arrow down indicates the interval when the measuring flashes were on. The same is true for the actinic light denoted by the green color. The red arrows indicate position of saturating pulses. Short horizontal lines indicate fluorescence levels: blue for minimal fluorescence for dark-adapted state (F0 ), green for actual fluorescence for light-adapted states (F(t)), and red for maximal fluorescence for dark-adapted states (FM ), light-adapted states (FM ), and during dark recovery of the ). The minimal sample, i.e. during relaxation of non-photochemical quenching (FM fluorescence of light-adapted states (F0 ) was not measured. (For interpretation of the references to color in this legend, the reader is referred to the web version of the article.)
consequently leading to formation of open PSII state which enables correct and a fast (compared to a case when the actinic light is switched off only) determination of the minimal Chl fluorescence level for a light-adapted state (see below). All of the so-called quenching coefficients and other parameters listed below are calculated from different basic fluorescence levels (see Fig. 4), which can be divided into three basic groups: • fluorescence levels determined for a dark-adapted state of the sample: ◦ F0 – minimal fluorescence for the dark-adapted state; ◦ FM – maximal fluorescence for the dark-adapted state; • fluorescence levels determined for a light-adapted state of the sample: ◦ F0 – minimal fluorescence for the light-adapted state; – maximal fluorescence for the light-adapted state; ◦ FM ◦ F(t) – actual fluorescence at time t during the slow FLI; in the literature this fluorescence level is often denoted as FS , where S stands for steady state, however, usage of F(t) has a more general meaning; • fluorescence levels determined in darkness, after previous illumination of the sample by actinic light: ◦ F0 – minimal fluorescence during dark recovery after previous illumination; – maximal fluorescence during dark recovery after previous ◦ FM illumination; where following lights are used to obtain particular fluorescence levels: • • • •
F0 – measuring flashes only; FM – measuring flashes and saturating pulse; F0 – far-red light and measuring flashes; – measuring flashes, actinic light and saturating pulse; FM
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
• F(t) – measuring flashes and actinic light; • F – far-red light and measuring flashes; 0 • F – measuring flashes and saturating pulse. M The reader should keep in mind that evaluation of almost all of the fluorescence levels mentioned above might have conceptual and practical problems in their determinations/measurements. For example, to saturate photochemistry, the saturating light pulse is used which causes a fast Chl FLR during the pulse (the so-called O-J-I-P curve) to a maximal fluorescence level. Conceptually it is assumed in all the theories presented in this review that photochemistry is reflected in a reduction of the first quinone electron acceptor of PSII, QA . According to the original hypothesis by Duysens and Sweers (1963), when QA is oxidized in all PSIIs, the minimal fluorescence signal is detected and when QA is reduced in all PSIIs, the maximal fluorescence signal is detected. When QA is oxidized, PSII is in the so-called open state whose fraction is estimated by qP (Eqs. (4) and (6)) and its derivatives (Eqs. (40)–(42) and (69)–(71)). However, it was found that the fast Chl FLR might reflect also other process than QA reduction (for reviews see Lazár, 2006; Stirbet and Govindjee, 2012; Schansker et al., 2014). Therefore, the maximal fluorescence levels might not reflect the QA reduction only leading to overestimation of the maximal fluorescence levels (for reviews see Samson and Bruce, 1996; Samson et al., 1999). On the other hand, practical determination of maximal fluorescence levels, mainly for light-adapted states, might suffer from underestimation because the intensity of saturating light pulse used for determination of the maximal fluorescence levels might not be high enough (due to existence of the light-induced non-photochemical Chl fluorescence quenching, see below). To avoid this problem, different routines for determination of the correct maximal fluorescence levels have been suggested (Earl and Ennahli, 2004; Loriaux et al., 2013). Another example of a possible problem is measurement of minimal fluorescence levels either for the dark-adapted or light-adapted states. Conceptually, a contribution of PSI Chl fluorescence (see Lazár, 2013 and references therein) should be taken into account so that the measured minimal fluorescence levels are not overestimated by the contribution of the PSI Chl fluorescence (reviewed, e.g., by Henriques, 2009). On the other hand, practically, the measuring flashes used for determination of the F0 -level must be of very low intensity so that the flashes do not cause any PSII closure (QA reduction), which would overestimate the real F0 -level. As for the F0 -level, the far-red light used to oxidize the electron transport chain and thus to determine quickly and correctly the F0 -level, must be of enough high intensity otherwise the F0 -level is overestimated by an insufficient oxidation of the electron transport chain by the used far-red light. Overall, because of the problems with correct determination of the Chl fluorescence levels, the related quantum yields of processes of photosynthetic energy partitioning might not be correct. Therefore, further research, which is beyond the scope of this review, is still needed on these topics. As it is not the goal of this review to discuss these issues, the reader is referred to other reviews which dealt with these topics (Samson and Bruce, 1996; Samson et al., 1999; Lazár, 2006; Stirbet and Govindjee, 2012; Harbinson, 2013; Schansker et al., 2014). From the above mentioned basic fluorescence levels some other fluorescence values are calculated:
• FV = FM − F0 – variable fluorescence for a dark-adapted state; • F = F − F – variable fluorescence for a light-adapted state; V M 0 • F = F = F − F(t) – difference in fluorescence between F and M M Q F(t);
135
• F = F − F – variable fluorescence during dark recovery after V M 0 previous illumination. The fluorescence levels and values mentioned above are then used for calculations of the quenching coefficients and other parameters. Even if these quenching coefficients and other parameters were successfully used to describe photosynthetic function, they do not represent quantum yields and are therefore hardly compared absolutely. The basic two quenching coefficients are coefficient of photochemical quenching qP and coefficient of nonphotochemical quenching qN calculated as (Schreiber et al., 1986): qP = qN =
F , FV
(4)
FV − FV FV
.
(5)
The quenching coefficients have meaning of the fractions of variable Chl fluorescence that is quenched by photochemical (qP ) or non-photochemical (qN ) processes. Note that in the original paper by Schreiber et al. (1986) the qP was denoted as qQ and instead of qN , denotation of qE was used, assuming that the energy dependent non-photochemical fluorescence quenching, characterized by its quenching coefficient qE (see below), is the main contributor to overall non-photochemical fluorescence quenching, characterized by its quenching coefficient qN . Also note that in the original paper by Schreiber et al. (1986) the far-red light was not applied to determine F0 and therefore the F0 level was approximated by the F0 level (i.e., the minimal fluorescence is not changed by light-adaptation), which in fact leads to the equations for the quenching coefficients listed below which are also used in the literature: qP = qN =
F −F , FM 0 FM − FM
FV
(6)
.
(7)
In addition to its meaning as mentioned above, qP (Eqs. (4) and (6)) estimates a fraction of “open” PSII centers (i.e., PSII with oxidized QA ) and was in fact deduced for the so-called puddle model of photosynthetic units (PSU), i.e., PSUs which are energetically separated and each PSU contains one reaction center and related antenna system (see Lazár, 1999; Stirbet, 2013 for reviews). Alternatives for the estimation of fraction of “open” PSII centers were also derived (see Eqs. (40)–(42) and (69)–(71) and the text below). It might happen that far-red light to measure F0 is not available in the fluorometer or the F0 level is overestimated (see above). In these cases, the F0 level can be calculated according to Oxborough and Baker (1997) as: F0 =
F0 ). (FV /FM ) + (F0 /FM
(8)
It was later shown that the same equation for F0 can be also derived in the framework of other theories (Klughammer and Schreiber, 2008; Kasajima et al., 2009). In addition, other parameters calculated from fluorescence levels measured during the slow FLI are frequently used: • relative change of minimal fluorescence characterized by the quenching coefficient q0 calculated as (Bilger and Schreiber, 1986): q0 =
F0 − F0 F0
,
(9)
136
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
• non-photochemical Chl fluorescence quenching (determined on only), NPQ, calculated as (Bilger and the basis of FM and FM Björkman, 1990): NPQ =
FM − FM FM
.
(10)
• in analogy with the NPQ parameter, photochemical Chl fluorescence quenching expressed by the PQ parameter (Porcar-Castell et al., 2014; see also Porcar-Castell, 2011 and Laisk et al., 1997), also denoted as PC parameter (Guadagno et al., 2010; see also Ahn et al., 2009) calculated as follows: PQ =
FM FM − . FM F(t)
(11)
Further analysis of Eqs. (10) and (11) reveals (not shown) that the NPQ and PQ parameters are defined on the same basis; they express, for the light-adapted state, values of relative rate constants (relative to sum of the rate constants of fluorescence and basal constitutive non-regulatory heat dissipation, i.e., rate constant of basal constitutive non-regulatory dissipation, kf,D ) of the total non-photochemical and photochemical processes, respectively. Therefore, NPQ and PQ can be directly compared. The other parameters which are calculated from fluorescence values measured during the slow FLI are the quantum yields but they were originally used separately without existence of a theory (photosynthetic energy partitioning) considering that the sum of all the quantum yields for a given state of the sample (dark- vs. light-adapted state) must be unity (Eq. (2)). These quantum yields are: • maximal (or potential) quantum yield of PSII photochemistry for light-adapted state, ˚PSII , calculated as (Oxborough and Baker, 1997): ˚PSII =
FV
FM
,
(12)
• actual quantum yield of PSII photochemistry for light-adapted state, ˚P , calculated as (Genty et al., 1989; the yield was denoted there as ˚C and the authors did not distinguish denotations of the minimal and maximal fluorescence levels for dark- and lightadapted states): ˚P =
F . FM
(13)
In interpretation of data it is also sometimes helpful to have in mind that (Genty et al., 1989; Havaux et al., 1991; Oxborough and Baker, 1997): ˚P = qP ˚PSII .
(14)
As stressed by Kornyeyev and Holaday (2008), it is important to note that ˚PSII describes the situation in a light-adapted sample when the actinic light is not present at the time of its determination, it means when all reaction centers of PSII are oxidized (far-red light is used instead of actinic light leading to oxidation of all electron transport chain, including PSII reaction centers), whereas ˚P describes the situation in a light-adapted sample when the actinic light is present at the time of its determination, it means when reaction centers of PSII are partly oxidized and partly reduced. A correct understanding of the difference between ˚PSII and ˚P is critical for correct interpretation of the parameters (quantum yields) of photosynthetic energy partitioning described in the further text.
Last but not least, the most often used basic Chl fluorescence parameter is the maximal quantum yield of PSII photochemistry for a dark-adapted state, ˚Po , defined as (Kitajima and Butler, 1975): ˚Po =
FM − F0 FV = . FM FM
(15)
Many parameters related to the quenching analysis have been derived. An (older) analysis of the parameters known at that time revealed that only ˚Po , qP , qN , and q0 are independent parameters and hence the other parameters do not bring any new information (Roháˇcek, 2002). At the end of this section it is important to note that, even if an effort for unified terminology and denotation was done (e.g., van Kooten and Snel, 1990), some of the parameters might be denoted in different literature in a little bit varied way than that used here. Alternatively, the same denotation can be used for different parameter(s) which is caused by the fact that too many parameters already exist and it is difficult to create so far unused denotations for them. Therefore the reader must be always aware what parameter is dealing with based on its definition and original references. The relaxation analysis following the quenching analysis to discriminate particular types of non-photochemical quenching To formally distinguish between different modes of measurements, two different terms are used: the quenching analysis and the relaxation analysis, however, both terms are related to the measurement of Chl fluorescence quenching. The term quenching analysis refers to the measurements when actinic light is on (maximally shortly switched off during measurements of F0 ), i.e., the measurements during slow Chl FLI as mostly described so far above, and when the saturation pulses are applied to measure -levels and consequently to reveal the quenching. On the other FM hand, the term relaxation analysis refers to the measurements of sample recovery during darkness (after previous illumination of the sample) for up to 1 h by application of the saturation light pulses -levels (see above and Fig. 4). The analysis of the to measure FM -levels either directly (Quick and Stitt, 1989) or time course of FM by means of their semilogarithmic plot (Hodges et al., 1989) or by fitting of the FV -values (Roháˇcek, 2010) provides information about relaxation of the non-photochemical quenching in darkness which enables discrimination of three types of the non-photochemical Chl fluorescence quenching based on their different relaxation times in darkness. It is important to note that this discrimination is empirical and the relaxation times might differ from species to species and in dependence on the duration and intensity of used previous actinic illumination (see the works described in the section “Energy partitioning – consideration of active centers only and employing the relaxation analysis”). The three types of the Chl fluorescence non-photochemical quenching and their features, as originally discriminated (Horton and Hague, 1988; Hodges et al., 1989; Quick and Stitt, 1989; Walters and Horton, 1991, 1993), are as follows: • energy dependent (also called as high energy state) nonphotochemical Chl fluorescence quenching characterized by its quenching coefficient qE – associated with light-induced acidification of lumen, half-time of the relaxation is about 30 s to 1 min. • state transition non-photochemical Chl fluorescence quenching characterized by its quenching coefficient qT – results from phosphorylation of light-harvesting complexes associated with PSII followed by binding of the light-harvesting complexes to PSI (the so-called state 1 to state 2 transition), half-time of the relaxation is 5–8 min.
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
• photoinhibitory non-photochemical Chl fluorescence quenching characterized by its quenching coefficient qI – results from photoinhibition of PSII photochemistry, half-time of the relaxation is about 20–40 min and more. Occurrence of a particular type of the quenching also depends on the intensity of excitation light; hence, for example, the photoinhibitory non-photochemical quenching occurs only during illumination by a high intensity of excitation light whereas the state transition non-photochemical quenching is largely suppressed during excitation by high light intensity. In the literature, the quenching coefficients are also used for reference to the given type of quenching as such, i.e., the terms qE quenching, qT quenching, and qI quenching are used. The relaxation analysis is usually not used so often because it takes more time to measure it. However, recent papers on energy partitioning also considered the relaxation analysis in their theories (see below). Based on measurements of the relaxation analysis performed after previous high light illumination, when qT quenching is supposed to be suppressed (see above), the following equations were used by Krause and Jahns (2003) to calculate the coefficients of the non-photochemical quenching related to the qE and qI quenching: qE = 1 − qI = 1 −
FV
FV FV
FV
,
(16)
,
(17)
where FV and FV were determined after 10 min of illumination and after 10 min of dark recovery, respectively. It then holds that (Krause and Jahns, 2003): 1 − qN = (1 − qE )(1 − qI ).
(18)
On the other hand, Johnson and Ruban (2011) used the following equation to calculate the coefficients of the non-photochemical quenching related to the qE quenching: qE =
− F FM M FM
,
(19)
and F were determined after 5 min of illumination and where FM M after 5 min of dark recovery, respectively. For the case of high-light illumination, when qT quenching is supposed to be suppressed, a similar set of equations was defined by Krause and Jahns (2003) also for the NPQ parameter (Eq. (10)) as follows:
NPQE =
FM FM − F , FM M
(20)
NPQI =
FM − 1, FM
(21)
were determined in the same way as menwhere the F0 and FM tioned above in the work by Krause and Jahns (2003) and where subscripts E and I indicate fractions of the overall NPQ parameter related to qE quenching and qI quenching, respectively, with the following relationship (Krause and Jahns, 2003):
NPQ = NPQE + NPQI .
(22)
Although the qE quenching is usually connected with formation of zeaxanthin from violaxanthin (e.g., Holt et al., 2005), it was recently found that zeaxanthin-dependent quenching, denoted as qZ , is a new type of the non-photochemical quenching (Holzwarth et al., 2009; Nilkens et al., 2010) different from qE , qT and qI quenching. The qZ quenching probably occurs in PSII attached minor antennae, is formed during 10–30 min, and relaxes within 10–60 min.
137
Two notes are also necessary for the qT quenching. The first note is that in the original concept of discrimination of the nonphotochemical quenching as mentioned above, the qT quenching is not true non-photochemical quenching, i.e., the increase of the rate constant(s) of non-radiative heat dissipation(s) (leading to the quenching of Chl excited states with consequent decrease of Chl fluorescence); during actinic illumination the Chl fluorescence decreases because less excitations are formed in PSII due to the movement and binding of the light harvesting complexes from PSII to PSI (the state 1 to the state 2 transition). And vice versa, during the qT quenching relaxation phase in darkness, more excitations are formed in PSII due to the movement and binding of the light harvesting complexes from PSI back to PSII (the state 2 to the state 1 transition). The second note is that Schansker et al. -levels (2006) found that the qT component of a relaxation of the FM reflects inactivation of ferredoxin–NADP+ -oxidoreductase (FNR). Consequently, as the activation state of FNR affects redox state of the electron transport chain and thus also of the fluorescence signal (if FNR starts to inactivate during the dark relaxation of the sample, the electron transport chain become more reduced, including QA , leading to increased fluorescence signal), the qT relaxation component therefore would be rather of photochemical than of non-photochemical nature. It is clear that more research is needed on the qT component of dark relaxation of the non-photochemical Chl fluorescence quenching. Another type of non-photochemical Chl fluorescence quenching occurs in overwintering evergreens (reviewed, e.g., by Verhoeven, 2014). It is related to Chl fluorescence relaxation by the fact that this quenching does not relax and lasts for the whole winter and it is therefore called the sustained non-photochemical quenching. It is important to note that it represents a regulatory type of non-photochemical quenching; therefore, for the purpose of this work it is called as sustained regulatory non-photochemical quenching. To distinguish this quenching from light-induced regulatory non-photochemical quenching which is switched on when the light is present, the sustained regulatory non-photochemical quenching is usually revealed by means of fluorescence measurements during night (dark-adapted state) when the light-induced regulatory non-photochemical quenching is not present. Therefore, action of the sustained regulatory non-photochemical quenching is mostly determined by measurements of the maximal quantum yield of PSII photochemistry for a dark-adapted state (˚Po , Eq. (15)), which is lowered by the presence of the sustained regulatory non-photochemical quenching. Consequently, recovery of this quenching is monitored by an increase of ˚Po upon removing of winter-stressed plants into room temperature and it has two phases: a fast one occurring during minutes to hours and a slow one occurring within days. Based on various findings it was suggested that the two recovery phases of the quenching reflect the sustained forms of qE and qZ quenching, respectively (Verhoeven, 2014). In some recent studies, also a state of overwintering evergreens (see below) in light was described by the NPQ parameter as (PorcarCastell, 2011; Porcar-Castell et al., 2012): NPQw =
s FM
w FM
− 1,
(23)
where superscripts w and s stand for winter and summer, respectively. The NPQw parameter was further distinguished into the sustained regulatory part, NPQw s , and light-induced (reversible) regulatory part, NPQw , as follows (Porcar-Castell, 2011): r NPQw s =
s FM
w FM
− 1,
(24)
138
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
NPQw r =
s FM
w
FM
s FM
−
w FM
,
(25)
D = 1 − ˚PSII = 1 −
where the following relationship is valid: w
NPQ =
NPQw s
+ NPQw r .
(26)
In analogy with the NPQw parameter, photochemical Chl fluorescence quenching in the light-adapted state of overwintering evergreens was expressed by the PQw parameter (Porcar-Castell, 2011; see also Laisk et al., 1997), as follows: PQw =
s FM w
F(t)
−
s FM w FM
.
the state is when the actinic light is present or not, was defined as:
(27)
As already mentioned above for the NPQ and PQ parameters, w w parameters are defined also the NPQw , NPQw s , NPQr , and PQ on the same basis; they express, for the light-adapted state, values of relative rate constants (relative to kf,D ) of the total non-photochemical, sustained regulatory non-photochemical, light-induced (reversible) regulatory non-photochemical, and photochemical processes, respectively. Therefore, they can be directly compared. Energy partitioning Starting from mid-90s of the last century an effort was done to explore the so-called energy partitioning, i.e., to define quantum yields of processes responsible for utilization of absorbed light energy basically without further discrimination of the non-photochemical quenching into its three (or more) parts as mentioned above. However, the discrimination is included in the energy partitioning concept when the relaxation analysis is also considered (see below). The determination of energy partitioning is very close to the Chl fluorescence quenching analysis because it utilizes the same techniques for the measurement and determination of the basic fluorescence levels. However, in the Chl fluorescence quenching analysis, the goal is mainly to determine the quenching coefficients whereas in the energy partitioning, the goal is to determine the quantum yields. The advantage of usage of the quantum yields instead of the quenching coefficients is that the quantum yields sum up for a given state of the sample (dark- vs. light-adapted state) to unity (Eq. (2)), and therefore the roles of particular processes can be quantified and compared absolutely. As a description of the energy partitioning is the key part of this review, the parameters defined in particular works are presented accordingly. Overall, the theories of energy partitioning can be divided into five groups in dependence if only active PSII centers are considered and the quenching analysis is imployed, if the quenching and relaxation analyses are employed, if the sustained regulatory non-photochemical quenching is discriminated, and if both active and photoinactivated centers are considered. The fifth group is a group which includes a work connecting the theories employing the quenching and relaxation analyses with the theories of active and photoinactivated centers. Energy partitioning – employing the quenching analysis and consideration of active centers only Demmig-Adams et al. (1996) One of the first works on photosynthetic energy partitioning was the paper by Demmig-Adams et al. (1996). In this work, in analogy with the meaning of ˚P (Eq. (13)), denoted as P in this paper, a parameter D, having the meaning of quantum yield of overall thermal dissipation for a light-adapted state, without a specification if
FV
FM
.
(28)
In this approach, both P and D should sum to 1 but this was not found and another parameter E (for excess) had to be considered, which was defined as: E = 1 − D − P = (1 − qP )
FV
FM
.
(29)
The E parameter has according to Demmig-Adams et al. (1996) a meaning of a quantum yield of process(es) of energy utilization different from photochemistry (the P parameter) and thermal dissipation (the D parameter). As the processes hidden behind the E parameter are difficult to “imagine”, this approach seems to be not correct from its beginning. The origin of the problem was found later on by Kornyeyev and Hendrickson (2007); the use of D and P in one balance equation is not correct because D and P describe the energy partitioning under conditions when all reaction centers of PSII are open (light-adapted state when the actinic light is not present) and partially open (light-adapted state when the actinic light is present), respectively, which consequently leads to the existence of the E parameter. Hence, the E parameter does not represent a real component of the energy partitioning but rather a difference between the potential (˚PSII , Eq. (12)) and actual (˚P , Eq. (13)) quantum yields of PSII photochemistry for the light-adapted state (Kornyeyev and Hendrickson, 2007; Kornyeyev and Holaday, 2008): E = 1 − D − P = 1 − (1 − ˚PSII ) − ˚P = ˚PSII − ˚P .
(30)
Kato et al. (2003) The thermal dissipation in the framework of the DemmigAdams theory was further elaborated by Kato et al. (2003); they first considered that the value by which the maximal quantum yield of PSII photochemistry for a dark-adapted state, ˚Po (Eq. (15)), is lower than unity, determines the quantum yield of thermal dissipation in the dark-adapted state, L, expressed as: L = 1 − ˚Po = 1 −
FV , FM
(31)
where L stands for “loss”. Similarly, the quantum yield of thermal dissipation in a light-adapted state, denoted as D again, is expressed as: D=
F FV − V . FM FM
(32)
It further holds that the sum of L and D described by Eqs. (31) and (32), respectively, equals to the original D parameter (Eq. (28)) by Demmig-Adams et al. (1996) describing overall thermal dissipation for a light-adapted state. The other quantum yields (P and E, Eqs. (13) and (29), respectively) were defined by Kato et al. (2003) in the same was as by Demmig-Adams et al. (1996). Ishida et al. (2011) Even if the theory of Demmig-Adams et al. (1996) seems to be not correct, it was further advocated by Ishida et al. (2011). On the basis of experiments with transgenic rice in which the psbS gene was silenced (PsbS protein is necessary for the qE quenching; Li et al., 2000), Ishida et al. (2011) showed that D of the DemmigAdams theory consists of light-independent dissipation present in both the transgenic and in wild-type rice and from light-dependent dissipation present only in wild-type rice. According to Ishida et al.
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
(2011), the light-independent part of D of the Demmig-Adams theory, denoted as A and also referred to as the constitutive component of the quenching (see the other theories below), probably occurs in antenna and its quantum yield is for dark-adapted state calculated as: A=
F0 . FM
(33)
The experimentally found division of original Demmig-Adams’ parameter D into light-independent and light-dependent parts (Ishida et al., 2011) is in agreement with the above mentioned results of Kato et al. (2003) who divided the original parameter D of the Demmig-Adams theory (Eq. (28)) into L and D (Eqs. (31) and (32), respectively) parameters. Moreover, the equation for the parameter A (Eq. (33)) of Ishida et al. (2011), expressing the lightindependent part of the original Demmig-Adams’ parameter D, equals to (is another form of) the equation for the parameter L (Eq. (31)) of Kato et al. (2003), expressing the quantum yield of thermal dissipation in dark-adapted state. Further, even if, on one hand, Ishida et al. (2011) experimentally found a linear dependence of D of the Demmig-Adams theory (Eq. (28)) on the NPQ parameter (Eq. (13)), they, on the other hand, also derived a nonlinear dependence between D and NPQ (and also another nonlinear dependence between D and qN ) in a form: D=
1 − FV . FV + F0 (NPQ + 1)
(34)
Ishida et al. (2011) further found that the E parameter increases in PsbS-silenced plants and they discussed that the E parameter must represent energy dissipation too and that this dissipation occurs in or around closed PSII centers. This agrees with the explanation of the E parameter as the difference between the potential (˚PSII , Eq. (12)) and actual (˚P , Eq. (13)) quantum yields of PSII photochemistry for light-adapted states (Eq. (30)). The difference can be caused only by energy dissipation and the dissipation can occur only in the closed center which causes a decrease of ˚P (PSII reaction centers are partially open and partially closed) with respect to ˚PSII (all PSII reaction centers are open). Further, according to Ishida et al. (2011), the E is basal constitutive heat dissipation (see the other theories below). Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) Another approach for the energy partitioning was started at about the same time, as the first one (Demmig-Adams et al., 1996), by Genty et al. (1996) and Cailly et al. (1996) and this theory was advocated later on by Hendrickson et al. (2004). The works by Genty et al. (1996) and Cailly et al. (1996) are only conference abstracts and therefore, these theories were almost lost in the past until rediscovered by Hendrickson et al. (2004). In this theory of energy partitioning, an effort is done to discriminate different types of non-photochemical losses. These losses can be divided into two processes: constitutive non-regulatory (basal or dark) dissipation processes consisting of fluorescence emission and heat dissipation (see Fig. 2 and the legend to Fig. 1), characterized by their quantum yield ˚f,D (also denoted as ˚cNP ) which is calculated as: ˚f,D
F(t) = FM
(35)
and regulatory light-induced non-photochemical quenching characterized by its quantum yield ˚NPQ (also denoted as ˚r NP ) which is calculated as: ˚NPQ =
F(t) F(t) . − F FM M
(36)
139
The actual quantum yield of PSII photochemistry for the lightadapted state, ˚P (Eq. (13)), was the third part of energy usage by PSII and the sum of all the quantum yields equals unity: ˚P + ˚f,D + ˚NPQ = 1.
(37)
Therefore, there is no parameter (the E parameter) with a nonphysiological meaning as in the Demmig-Adams theory. Another advantage of the theory of Genty et al. is that it does not require knowledge of the F0 fluorescence level (Hendrickson et al., 2004) which is sometimes difficult to determine correctly (see above). So far, it has not been revealed in the literature (some hints can be found in Porcar-Castell (2011); see also Laisk et al. (1997)) that a ratio of ˚NPQ to ˚f,D equals the NPQ parameter of Bilger and Björkman (1990) (Eq. (10)): − F(t)/F F(t)/FM FM − FM ˚NPQ M = = = NPQ. FM ˚f,D F(t)/FM
(38)
Hence, as revealed here, the NPQ parameter does not only reflect the regulatory light-induced non-photochemical quenching as it is usually used for in the literature but also a ratio of the quantum yield of this quenching to the quantum yield of the constitutive non-regulatory non-photochemical quenching. Therefore, the same value of the NPQ parameter can be obtained for different values of ˚NPQ and ˚f,D , which should be taken into account in the analysis of results. In connection to Eq. (38), a similar relationship is revealed here also for the PQ parameter (Eq. (11)): (F − F(t))/FM F FM FM ˚P FM F(t) FM − = PQ, = M = M − = FM FM F(t) FM F(t) F(t) ˚f,D F(t)/FM
(39)
it means that the PQ parameter is a ratio of the quantum yield of actual PSII photochemistry for a light-adapted state to the quantum yield of the constitutive non-regulatory non-photochemical quenching. It is worth to repeat that in this and the following theories, the non-photochemical quenching is divided into two parts: the quenching which is always present, even in healthy leaves because it is the consequence of basic physics (see Fig. 2 and the legend to Fig. 1), the constitutive non-regulatory (basal or dark) non-photochemical quenching/dissipation, and, on the other hand, the non-photochemical quenching which is formed only upon illumination to regulate the amount of absorbed light to avoid occurrence of damage processes, the regulatory light-induced nonphotochemical quenching/dissipation. Kramer et al. (2004) Assuming the so-called lake model for PSUs (one PSU contains several reaction centers which all share the same antenna system without any restrictions; see Lazár, 1999 and Stirbet, 2013 for reviews), which is claimed to describe the energetic connectivity in terrestrial plants better than the puddle model (Kramer et al., 2004), a fraction of open PSII centers at the time t can be estimated from the qL parameter defined as (Kramer et al., 2004): qL =
1/F(t) − 1/FM
1/F0
− 1/FM
=
− F(t) FM − F FM 0
(F0 /F(t)).
(40)
An equation for qL was also derived by Miyake et al. (2009) which was later shown by Kasajima et al. (2009) to be transformed and to be identical to the equation derived by Kasajima et al. (2009): qL =
1/F(t) − 1/FM 1/F0 − 1/FM
.
(41)
140
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
Eqs. (40) and (41) differ in replacement of F0 by F0 in the later equation. For the case of the connected PSUs (each reaction center possesses its own antenna system but with defined probability of energy transfer from one antenna to antenna of another PSU; see Lazár, 1999 and Stirbet, 2013 for reviews), Kramer et al. (2004) derived a fraction of open PSII centers in these PSUs, qcu , to be expressed as: qcu =
− F(t) FM
− F J(F(t) − F0 ) + FM 0
,
(42)
where J is the sigmoidicity parameter (Lavergne and Leci, 1993) related to the probability of energy transfer between PSUs, p (Joliot and Joliot, 1964), as follows: J=
p . 1−p
(43)
Kramer et al. (2004), however, have shown that for appropriate values of the connectivity parameter p (about 0.5–0.7), there is almost no difference between usage of qcu and qL . According to Kramer et al. (2004), the non-photochemical losses can be divided again into two processes with related quantum yields: ˚(NO) =
1 NPQ + 1 + qL ((FM /F0 ) − 1)
˚(NPQ) = 1 − ˚P − ˚(NO),
(44) (45)
where ˚(NO) reflects quantum yield of constitutive non-regulatory (basal or dark) non-photochemical dissipation processes and ˚(NPQ) reflects quantum yield of regulatory light-induced nonphotochemical quenching. It was further shown by Klughammer and Schreiber (2008) and by Kasajima et al. (2009) that after inserting the equation for F0 (Eq. (8)) into ˚(NO) (Eq. (44)) and ˚(NPQ) (Eq. (45)), these lead to the same equations as for ˚f,D (Eq. (35)) and ˚NPQ (Eq. (36)), respectively of the previous theory. Consequently, the relationship among NPQ, ˚NPQ , and ˚f,D (Eq. (38)), is valid also for NPQ, ˚(NPQ), and ˚(NO). The same is true also for the relationship among PQ, ˚P , and ˚f,D (Eq. (39)), and PQ, ˚P , and ˚(NO). Further, it was also shown (Klughammer and Schreiber, 2008) that the equations for ˚(NO) and ˚(NPQ) parameters are valid not only for the lake model of PSUs but also for the puddle model of PSUs. In experiments, ˚NPQ (Eq. (36)) of the previous theory gives almost identical values as ˚(NPQ) (Eq. (45)), however, ˚f,D (Eq. (35)) of the previous theory is significantly higher than ˚(NO) (Eq. (44)). The difference between ˚f,D and ˚(NO) was explained (Hendrickson et al., 2004) by an incorrectness in F0 determination. Kramer et al. (2004) also have shown that the ˚P parameter is valid for both the puddle and lake models of PSUs. It is important also to note (it already implies from Eq. (45)) that: ˚P + ˚(NO) + ˚(NPQ) = 1.
(46)
Hence, the same as in the case of the previous approach (compare with Eq. (37)), there is no need for the E parameter of the Demmig-Adams theory. Overall, even if the theory by Kramer et al. (2004) is based on different assumptions, it in fact leads to almost identical values of the quantum yields as in the theory by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004).
analyses. Ahn et al. (2009) used the same equations for ˚P (Eq. (13)), ˚f,D (Eq. (35)), and ˚NPQ (Eq. (36)) as derived previously by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004). However, Ahn et al. (2009) denoted ˚f,D as ˚C and expressed ˚NPQ as follows: ˚NPQ =
FM − FM F(t) , FM FM
(47)
which is another form of (equals to) Eq. (36). Ahn et al. (2009) further presented equations that express the quantum yields of the various processes that comprise ˚NPQ , namely quantum yield of the qE quenching, ˚qE , and quantum yield of the sum of qT and qI quenching, ˚qT+qI , as follows: ˚qE =
− F FM M F(t) , FM FM
˚qT +qI =
(48)
FM − FM F(t) , FM FM
(49)
is the threshold maximal fluorescence level determined where FM from the relaxation analysis. The threshold time of 10 min for deter was used for Arabidopsis plants. Ahn et al. (2009) mination of FM showed experimentally but it also implies from the above equations that:
˚NPQ = ˚qE + ˚qT +qI .
(50)
Ahn et al. (2009) emphasized that the above equations for ˚NPQ , ˚qE , and ˚qT+qI (Eqs. (47)–(49)), but also for ˚P and ˚f,D , are valid for the light-adapted state when the actinic light is present and thus when a certain fraction of the PSII centers is in the open and the rest in the closed state. Therefore the quantum yields are in fact actual quantum yields. To obtain the quantum yields during the saturation pulse, i.e., for the same conditions of measurements as for the NPQ parameter (Eq. (10)), it is in the case when all reaction centers are closed, following formulae were presented: ˚NPQ,RCC = ˚qE,RCC =
FM − FM
FM
− F FM M
˚qT +qI,RCC =
FM
,
(51)
,
(52)
F FM − FM M
FM
FM
,
(53)
where subscript RCC means “reaction centers closed” and it is again valid that: ˚NPQ,RCC = ˚qE,RCC + ˚qT +qI,RCC .
(54)
It is interesting to note that the set of quantum yields represented by Eqs. (51)–(53) is the only published set of the quantum yields for a light-adapted state when light is switched on and all reaction centers are closed. However, to have a complete set of the quantum yields when all the reaction centers are closed, it is noticed here that the following quantum yields must be added: ˚P,RCC = 0, ˚f,D,RCC = 1 − ˚P,RCC − ˚NPQ,RCC = 1 − 0 −
(55) FM − FM
FM
=
FM
FM
.
(56)
Kasajima et al. (2009) Energy partitioning – employing the quenching and relaxation analyses Ahn et al. (2009) The paper by Ahn et al. (2009) represents the first theory of energy partitioning which employs the quenching and relaxation
Kasajima et al. (2009) derived the same equations for ˚P (Eq. (13)) and for ˚f,D (Eq. (35)) as derived previously by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004). Further, Kasajima et al. (2009) were not so strict in discrimination of the regulatory light-induced non-photochemical quenching with respect to its origin as Ahn et al. (2009) were and divided it originally
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
only into fast and slow components, again based on the relaxation analysis. However, Kasajima et al. (2009) concluded that the fast component corresponded to the qE quenching whereas the slow component approximately corresponded to the sum of qT and qI quenching. The threshold time to distinguish between the fast and slow components of non-photochemical quenching depended on plant species and was determined based on the relaxation analysis to be 1 min for Arabidopsis (compare with 10 min for Arabidopsis used by Ahn et al., 2009) and 5 min for rice. Then, the quantum yields of the fast and slow components of the regulatory lightinduced non-photochemical quenching, ˚NPQ fast and ˚NPQ slow , respectively, were derived to be as follows: ˚NPQ fast =
− 1/F 1/FM M
˚NPQ slow =
1/F(t)
,
− 1/F 1/FM M
1/F(t)
,
(57) (58)
is the threshold maximal fluorescence level determined where FM from the relaxation analysis as mentioned above. It further holds that sum of ˚NPQ fast (Eq. (57)) and ˚NPQ slow (Eq. (58)) equals to ˚NPQ (Eqs. (36) and (47)) of Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004). Also notice that Eqs. (57) and (58) are another forms of (equal to) Eq. (48) and (49), respectively, of Ahn et al. (2009) and are also another forms of (equal to) Eq. (104) and (105), respectively, of Kornyeyev et al. (2013) (see below). Interestingly, these equations were used by Chen et al. (2012) who, however, referred to wrong papers and not to Ahn et al. (2009) or Kasajima et al. (2009).
Guadagno et al. (2010) The work by Guadagno et al. (2010) presents the third and the last theory which employs the quenching and relaxation analyses. Guadagno et al. (2010) paid more attention to finding the threshold -levels during dark relaxation. By plottimes of variations of the FM -levels were used instead ting logarithm of a NPQ parameter (the FM -levels) vs. time during relaxation analysis measured with of the FM beet plants, they have found that there are two variations of the slope of the NPQ curve, revealing thus two threshold times of 2 and 20 min. There was no variation of the slope at 10 min as used for the threshold time by Ahn et al. (2009) and at 1 or 5 min as used for the threshold time by Kasajima et al. (2009). It is important to note here that it was later found that analysis of time dependence of the NPQ data, as was done by Guadagno et al. (2010), distorts real kinetics of the non-photochemical Chl fluorescence quenching (Holzwarth et al., 2013), therefore, the real threshold times might be different than found by Guadagno et al. (2010). Guadagno et al. (2010) then used the ˚P (Eq. (13)), ˚f,D (Eq. (35)), and ˚NPQ (Eq. (36)) as derived previously by Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) with the same re-denotation of ˚f,D as ˚C and expression of ˚NPQ (Eq. (47)) as used by Ahn et al. (2009). Thanks to the existence of the two threshold times in the relaxation analysis, Guadagno et al. (2010) were able to divide ˚NPQ into quantum yield of the qE quenching, ˚qE , quantum yield of the qT quenching, ˚qT , and quantum yield of the qI quenching, ˚qI , according to the following formulae:
˚qE =
x − F FM M F(t) , FM Fx
(59)
M
y
˚qT =
x FM − FM F(t)
y
x FM
FM
,
(60)
y
˚qI =
FM − FM F(t) , y FM F M
(61)
141
y
x and F where FM are the maximal fluorescence levels of the first M and second, respectively, threshold times determined from the x represents maximal fluorescence relaxation analysis. In fact, FM signal during the dark recovery of the sample when qE quenching was completely relaxed and qT and qI quenching was still y present whereas FM represents a state of the sample when qE and qT quenching were relaxed and only qI quenching was still present. Guadagno et al. (2010) showed experimentally but it also implies from the above equations that:
˚NPQ = ˚qE + ˚qT + ˚qI .
(62)
The theory used by Guadagno et al. (2010) is so far the most complete theory of energy partitioning employing the quenching and relaxation analyses. Energy partitioning – discrimination of the sustained regulatory non-photochemical quenching Porcar-Castell et al. (2008) Because in overwintering evergreens the sustained regulatory non-photochemical quenching does not relax in winter, even during nights (dark-adapted state), estimation of the quantum yields with respect to energy partitioning and the sustained regulatory non-photochemical quenching was first derived for dark-adapted state (Porcar-Castell et al., 2008). In fact, it is interesting to note that the set of quantum yields in this theory represents the only published set for a dark-adapted state. First, the authors derived the maximal quantum yield of PSII photochemistry for a dark-adapted state, which is identical to Eq. (15). But because the fluorescence values are measured in winter when the sustained regulatory nonphotochemical quenching is present, the quantum yield represents the maximal quantum yield of PSII photochemistry for a darkadapted state when the quenching is present (as mentioned above, the quantum yield is smaller in winter than in summer). To distinguish it from Eq. (15), superscript w (for winter) is used in denotation of the quantum yield and fluorescence levels: ˚w Po =
w − Fw FM 0 w FM
=
FVw w FM
.
(63)
The other quantum yields, for the constitutive non-regulatory (basal or dark) dissipation, ˚w , and for the sustained regulatory f,D
, were derived as follows: non-photochemical quenching, ˚w NPQ ˚w = f,D
F0w As as s Aw aw , FM
˚w NPQs =
F0w w FM
−
(64)
F0w As as s Aw aw , FM
(65)
where A is sample absorptance and a is a fraction of the absorbed light that is captured by PSII. The superscript s indicates reference levels determined in summer. As, usually, the measurements are done with mature leaves where Chl concentration is high enough causing saturated absorptance levels (Porcar-Castell et al., 2008), the changes in absorptance can be omitted. Further, it is assumed that PSII captures the same fraction of the absorbed light as PSI and that this fraction is not changed during seasons (Porcar-Castell et al., 2008), therefore also the changes in fraction of the absorbed light that is captured by PSII can be omitted. Then, Eqs. (64) and (65) can be rewritten to: ˚w = f,D
F0w s FM
˚w NPQs =
,
F0w w FM
(66) −
F0w s FM
.
(67)
142
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
It can be easily checked that the sum of all the quantum yields defined by Eqs. (63), (66) and (67) is unity: w w ˚w Po + ˚f,D + ˚NPQs = 1.
(68)
It is worth to stress that because all the quantum yields (Eqs. (63), (66) and (67)) were determined for a dark-adapted state during which the sustained regulatory non-photochemical quenching persists, the quantum yield of this quenching (Eq. (67)) is directly accessible and no discrimination of the quantum yields of regulatory non-photochemical quenching into its parts is necessary as it is done in the next work (Porcar-Castell, 2011). It is also worth to mention that in the original work by PorcarCastell et al. (2008) was a typographical error (confirmed by a communication with Dr. Porcar-Castell) in the equation for ˚w NPQs (Eq. (9) in Porcar-Castell et al., 2008). Porcar-Castell (2011) The next work by Porcar-Castell (2011) is based on the work by Kramer et al. (2004); the qL parameter was used for estimation of a fraction of total open PSII centers at the time t in overwintering evergreens, qw , as follows (Porcar-Castell, 2011): L w 1/F(t)w − 1/FM
qw L =
,
s 1/F0s − 1/FM
(69)
where superscripts w and s indicate values at winter and summer, respectively. To discriminate the total regulatory nonparameter was photochemical quenching into its parts, the qw L further divided into fractions of open PSII centers where the sustained regulatory non-photochemical quenching is present, qw , and where the light-induced (reversible) regulatory nonL,s , as follows (Porcar-Castell, photochemical quenching is present, qw L,r 2011):
qw L,s = qw L,r =
w 1/F0w − 1/FM
1/F0s
,
w 1/F(t)w − 1/FM
w 1/F0w − 1/FM
.
=
F w w FM
.
(73)
photochemical quenching, ˚w , are almost the same as the yields NPQt from the theory of Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) (Eqs. (35) and (36)) but the reference dark-adapted maximal fluorescence level determined in summer is considered (superscript s), i.e.:
˚w NPQt =
F(t) w FM
(78)
w w ˚w NPQs + ˚NPQr = ˚NPQt .
(79)
w FM
−
where it holds:
The quantum yields of the sustained regulatory non, according to theory of photochemical quenching, ˚w NPQs Porcar-Castell et al. (2008) and Porcar-Castell (2011) differ in usage of the dark-adapted minimal fluorescence level, F0w , in the first case (Eq. (67)) instead of actual fluorescence level, F(t)w , in the second case (Eq. (77)), reflecting the fact that the theories were derived for the dark- and light-adapted states, respectively. On the other hand, the equation for the quantum yield of the lightinduced regulatory non-photochemical quenching, ˚w , (Eq. NPQr (78)) is identical to the equation for the yield according to theory of Genty et al. (1996), Cailly et al. (1996), and Hendrickson et al. (2004) (Eq. (36)) (when in situ fluorescence levels are considered omitting thus the w superscripts in Eq. (78)). Again, it is worth noting that in the original work by PorcarCastell (2011) there was a typographical error (confirmed by a communication with Dr. Porcar-Castell) in the equation for ˚w f,D (Eq. (24) in Porcar-Castell, 2011).
As already shown above, the quantum yield of the photoinhibitory non-photochemical Chl fluorescence quenching (˚qI ) can be defined on the basis of the relaxation analysis of Chl fluorescence in darkness (Eq. (61)). However, the rationale of ˚qI of the photoinactivated centers formed by illumination was explicitly first defined in the framework of the Demmig-Adams theory (DemmigAdams et al., 1996) by Kornyeyev et al. (2001). The authors defined the P, D, and E parameters in the same way as in the original Demmig-Adams theory. Moreover, the D parameter (quantum yield of overall thermal dissipation) was suggested to be expressed as a sum of three quantum yields: DCON for constitutive always present dissipation, DREG for regulatory light-induced dissipation, and DPI for photoinhibitory dissipation, i.e., thermal dissipation in photoinactivated centers. Therefore: D = DCON + DREG + DPI .
(74)
−
w
F(t) . s FM
(75)
(80)
If DCON is the quantum yield for the only existing dissipation then it would be expressed as (ideal DCON ): ideal DCON = 1 −
F(t)w = , s FM w
F(t)w
(77)
F(t)w w , FM
˚w NPQr =
Kornyeyev et al. (2001)
The quantum yields of the constitutive non-regulatory (basal or dark) dissipation, ˚w , and of the total regulatory nonf,D
˚w f,D
F(t)w F(t)w − . s w FM FM
(71)
The first from the quantum yields for the light-adapted state when the actinic light is present is the actual quantum yield of PSII photochemistry for the light-adapted state which is identical to ˚P (Eq. (13)) but to distinguish that it was measured in winter, it is again denoted by superscript w:
FM
˚w NPQs =
Energy partitioning – consideration of photoinactivated centers, in addition to active centers
(72)
w
(76)
The quantum yield of the total regulatory non-photochemical quenching was then discriminated into the quantum yield of sustained regulatory non-photochemical quenching, ˚w , and of NPQs light-induced regulatory (called reversible in Porcar-Castell, 2011) : non-photochemical quenching, ˚w NPQr
(70)
w w qw L = qL,s qL,r .
w − F(t)w FM
w w ˚w P + ˚f,D + ˚NPQt = 1.
s − 1/FM
It further holds that:
˚w P =
It again holds that:
FV . FM
(81)
But because of the existence of the other dissipation processes, which affect DCON , the ideal DCON is corrected for existence of these other dissipations. To account for these processes, the ideal DCON is multiplied by the maximal quantum yield of PSII photochemistry
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
for the light-adapted state (˚PSII , Eq. (12)) relative to that one for the dark-adapted state (˚Po , Eq. (15)) which leads to: DCON =
) (1 − FV /FM )(FV /FM
FV /FM
.
(82)
Action of photoinhibition was estimated from a change in the ˚Po parameter measured after ((FV /FM )PI ) photoinhibition relative to that one measured before (FV /FM ) photoinhibition. Therefore, quantum yield for ideal photoinhibitory dissipation equals to: ideal DPI = 1 −
) (1 − (FV /FM )PI /(FV /FM ))(FV /FM
(FV /FM )PI
(83)
.
(84)
The DREG parameter is then found from an equation defining D as a sum of the free quantum yields (Eq. (80)) leading to: DREG = 1 −
) (FV /FM
(FV /FM )PI
.
(85)
As discussed later by Kornyeyev and Hendrickson (2007), the above defined DPI and DREG parameters reflect the quantum yields in light-adapted state in the absence of actinic light but the P parameter (˚P ) reflects the quantum yield in light-adapted state in the presence of actinic light. Consequently, also this theory as the original Demmig-Adams theory, suffers from mismatching the quantum yields for light-adapted states measured with and without presence of actinic light which results in necessity of consideration of the non-physiological E parameter. Hikosaka et al. (2004)
(FV /FM )PI , FV /FM
(86)
where (FV /FM )PI is the ˚Po parameter determined for the photoinhibited sample (means after the photoinhibition). Note that in the original paper by Hikosaka et al. (2004), the maximum quantum yield of PSII photochemistry of photoinactivated and of active centers are denoted as Fv /Fm , and Fv /FmM , respectively. Also note that B (Eq. (86)) of Hikosaka et al. (2004) is identical to ideal DPI (Eq. (83)) but different to DPI (Eq. (84)) of the theory by Kornyeyev et al. (2001). Consequently B of Hikosaka et al. (2004) expresses the quantum yield of heat dissipation by photoinactivated centers in dark-adapted state whereas DPI of Kornyeyev et al. (2001) reflects the quantum yield of heat dissipation by photoinactivated centers in light-adapted state but in the absence of actinic light (Kornyeyev and Hendrickson, 2007). Then the quantum yield of energy loss (thermal dissipation) in dark-adapted state, DD , was defined as:
DD = (1 − B) 1 −
FV FM
DL =
F V
FM
PI
−
FV
FM
.
(88)
PI
The other parameter, ˚P , denoted by Hikosaka et al. (2004) as C, was defined as usually (Eq. (13)). The E parameter (excess) then equals to:
(87)
(89)
that leads to the same formula for E (Eq. (29)) as defined by Demmig-Adams et al. (1996). Even if the approaches by Kornyeyev et al. (2001) and by Hikosaka et al. (2004) are improved versions of the original Demmig-Adams theory (Demmig-Adams et al., 1996) by consideration of the photoinactivated centers, they still operate with the E parameter and therefore these approaches do not seem to be the good ones. Hendrickson et al. (2005) The third approach, by Hendrickson et al. (2005), which considers the photoinactivated non-functional centers, is based on the theory previously presented by the same authors (Hendrickson et al., 2004). In the later approach, quantum yield of heat dissipation by the photoinactivated non-functional centers, ˚NF , is defined in the same way as B (Eq. (86)) in the theory by Hikosaka et al. (2004) which considered the photoinactivated non-functional centers. The other quantum yields in the theory by Hendrickson et al. (2005) are the same as in their original approach but modified with respect to existence of the photoinactivated centers, it means, the quantum yields are multiplied by the fraction of active centers described by the (FV /FM )PI /(FV /FM ), ratio. Therefore, ˚P =
The Demmig-Adams theory (Demmig-Adams et al., 1996) was also extended for explicit existence of the photoinactivated nonfunctional centers later on in another way by Hikosaka et al. (2004). Similarly as in the theory of Kornyeyev et al. (2001), Hikosaka et al. (2004) assumed that relative change in the ˚Po parameter (Eq. (15)) reflects a measure of photoinhibition, therefore, fraction of photoinactivated centers and quantum yield of utilization of excitation energy by these centers (as heat dissipation) as well, B, is expressed as: B=1−
and the quantum yield of thermal dissipation in light-adapted sample, DL , was defined as:
E = 1 − B − DD − DL − C,
(FV /FM )PI . FV /FM
Then, similarly as in the case of DCON , ideal DPI is corrected for existence of the other dissipative processes accounted for again by change of ˚PSII relative to ˚Po but measured after photoinhibition (i.e., expressed by (FV /FM )PI ) which leads to: DPI =
143
)(F /F ) (F/FM V M PI
˚f,D =
FV /FM
,
(90)
(F(t)/FM )(FV /FM )PI , FV /FM
˚NPQ =
− F(t)/F )(F /F ) (F(t)/FM M V M PI
FV /FM
(91) .
(92)
All the quantum yields sum to unity, that is: ˚NF + ˚P + ˚f,D + ˚NPQ = 1.
(93)
Kornyeyev and Hendrickson (2007) The fourth approach which considered the photoinactivated non-functional centers is by Kornyeyev and Hendrickson (2007). It is also based on the theory of Hendrickson et al. (2004) but the role of photoinactivation is implemented into the theory in a different way than in the work of Hendrickson et al. (2005). Kornyeyev and Hendrickson (2007) assumed a pool of PSII which contains both active and photoinactivated centers. Considering that the presence of photoinactivated non-functional centers can be reflected only in fluorescence levels obtained for light-adapted states, i.e., in FS , and by expressing the quantum yields of the Hendrickson and FM et al. (2004) theory by means of the involved rate constants, it was found that all the quantum yields (˚P , ˚f,D and ˚NPQ ; Eqs. (13), (35) and (36), respectively) of Hendrickson et al. (2004) are functions of active and photoinactivated centers as well and hence, can be used for description of energy partitioning in a pool of both active and photoinactivated centers. However, ˚NPQ of Hendrickson et al. (2004) was found to be in fact a sum of the quantum yield of regulatory non-photochemical quenching (in active centers), ˚NPQ , and of
144
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
the quantum yield of heat dissipation by the photoinactivated nonfunctional centers, ˚NF . Further analysis revealed that if FM in the original equation for ˚NPQ by Hendrickson et al. (2004) (Eq. (36)) is exchanged with (FM )PI (FM determined in darkness after previous illumination causing photoinactivation), then a correct equation for quantum yield of regulatory non-photochemical quenching is obtained and it is as follows: ˚NPQ
F(t) F(t) = − . FM (FM )PI
(94)
Based on the fact that all the quantum yields sum to unity, the quantum yield of heat dissipation by the photoinactivated nonfunctional centers, ˚NF , then equals to: ˚NF = 1 − ˚P − ˚NPQ − ˚f,D =
F(t) F(t) − . FM (FM )PI
(95)
It also implies from the summation of all the quantum yields to unity that sum of all the non-photochemical quantum yields can ratio. be expressed by the F(t)/FM Kornyeyev and Holaday (2008) The theory of Kornyeyev and Holaday (2008) is the most advanced theory existing so far for the energy partitioning occurring in active and photoinactivated centers. The authors analyzed all the previous theories which consider the photoinactivated nonfunctional centers and when necessary they performed related corrections of the parameters. Consequently, Kornyeyev and Holaday (2008) defined two sets of parameters of energy partitioning, one set for the case when the sample is in the light-adapted state and in the presence of actinic light (reaction centers of PSII are partially open and partially closed) and the other set for the case when the sample is in the light-adapted state but in the absence of actinic light (all reaction centers of PSII are open). As for the first case (the sample is in the light-adapted state and in the presence of actinic light), it was already noted by Kornyeyev and Hendrickson (2007) that DPI (Eq. (84)) and DREG (Eq. (85)) reflect the quantum yields in light-adapted state but in the absence of actinic light. To obtain equations for the quantum yields in the presence of actinic illumination, the formulae for the DPI (Eq. (84)) and DREG (Eq. (85)) must be multiplied by F(t)/F0 ratio which has already been mentioned before by Kornyeyev and Hendrickson (2007) for the case of the DPI parameter. Then, the corrected parameters are as follows:
˚NF =
(FV /FM )PI 1− FV /FM
˚NPQ =
1−
FV /FM
FV /FM
F(t) , (FV /FM )PI F0
(FV /FM )PI
F(t) . F0
(96)
(97)
As the above parameters apply to the case of the light-adapted state and in the presence of actinic light, the quantum yield of PSII photochemistry for this case is ˚P (Eq. (13)). The last quantum yield, ˚f,D , is obtained from the fact that all the quantum yields sum to unity and therefore: ˚f,D =
F /F F(t) F(t) F(t) + V M − . FM F0 FV /FM F0
(98)
For the second case when the sample is in the light-adapted state but in the absence of actinic light, the quantum yield of PSII photochemistry is expressed by ˚PSII (Eq. (12)) and the other quantum yields are as follows:
˚NF =
(FV /FM )PI 1− FV /FM
FV /FM
(FV /FM )PI
,
(99)
˚NPQ = 1 −
˚f,D =
FV /FM
(FV /FM )PI
,
) (1 − FV /FM )(FV /FM
FV /FM
(100)
.
(101)
It is important to stress that the above two sets of equations (Eqs. (96)–(98) vs. (99)–(101)) differ not only in the formulation of the quantum yields other than of photochemistry of PSII but also in the quantum yield of photochemistry of PSII itself; in the first case, Eq. (13) must be applied because actinic light is switched on and in the second case Eq. (12) must be used because actinic light is switched off (actinic light must be switched off for measurement of F0 ). Overall, the equations presented in Kornyeyev and Holaday (2008) represent the most comprehensive theory for energy partitioning considering the active and photoinactivated centers published so far. Kornyeyev and Holaday (2008) further noted that in the case of field measurements it is difficult or in principle impossible to measure correct ˚Po (=FV /FM , Eq. (15)) because the sample should not have any damage in the case of ˚Po measurement which is not met in the field conditions where the sample is exposed to various environmental stresses (e.g., photoinhibition, high temperature stress, drought stress, etc.). To avoid this problem, ˚Po should be replaced by its standard value of 0.8 in Eqs. (96) and (99). Knowledge of ˚Po is not necessary for calculation of Eqs. (97) and (100), the same as for calculation of Eqs. (13) and (12). Even if knowledge of ˚Po is further necessary for calculation of Eqs. (98) and (101), their numerical values can be obtained on the basis of the fact that sum of all the quantum yields in given set of equations equals to unity.
Energy partitioning – connection of the theories employing the quenching and relaxation analyses with the theories of active and photoinactivated centers Kornyeyev et al. (2013) Kornyeyev et al. (2013) analyzed the previous approaches to define a universal theory for energy partitioning. Generally, they compared two approaches which differ in the fact that determination of physiological state of the sample is based on changes in values of maximal fluorescence or on changes in values of maximal quantum yield of PSII photochemistry (i.e., when all reaction centers of PSII are open, irrespective if the sample is in darkadapted or light-adapted state or is in the recovery state). In the first case, Kornyeyev et al. (2013) used the fact that when the photoinactivated centers are considered and maximal fluorescence is measured in darkness after photoinhibitory illumination ((FM )PI ), it is in fact equivalent to measurement of maximal ). Then, after a fluorescence during the relaxation analysis (FM period of dark recovery, the quantum yield of the regulatory nonphotochemical quenching processes which relax, ˚qE+qT , and which do not relax (the quenching of photoinactivated centers), ˚qI , can be expressed from the theory of active and photoinactivated centers by Kornyeyev and Hendrickson (2007) as: ˚qE+qT =
˚qI =
F(t) F(t) − F , FM M
F(t) F(t) , − F FM M
(102)
(103)
which are rewritten Eqs. (94) and (95), respectively, with (FM )PI re . On the other hand, according to theory of active denoted as FM centers employing relaxation analysis of Ahn et al. (2009), the
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
quantum yields of qE quenching (Eq. (48)) and sum of qT and qI quenching (Eq. (49)) can be rewritten into (equal to): ˚qE =
F(t) F(t) − F , FM M
(104)
F(t) F(t) . − F FM M
˚qT +qI =
(105)
Looking at Eqs. (102)–(105), the right sides of Eqs. (102) and (104) are identical, the same as the right sides of Eqs. (103) and (105) but the time of dark relaxation (the threshold time) , which separates the groups of nonprior to measurements of FM photochemical quenching components, was different in the two approaches; 3 h for separation of sum of qE and qT quenching from qI quenching (Eqs. (102) and (103)) and 10 min for separation of qE quenching from sum of the qT and qI quenching (Eqs. (104) and (105)). By combination of the theories, the quantum yield of the qT quenching can be obtained by subtracting Eq. (104) from Eq. (102) which leads to: ˚qT =
F(t) 10 min FM
−
F(t)
,
3h FM
(106)
where superscripts “10 min” and “3 h” indicate that the maximal fluorescence was measured after 10 min and 3 h, respectively, of dark recovery of the sample. The previous equations (Eqs. (102)–(105)) were constructed based on assumption/existence of only one threshold time in the sample recovery considered in particular theories, however, Guadagno et al. (2010) revealed two threshold times in the dark recovery of the sample, based on which they defined Eqs. (59)–(61)) which can be rewritten into (equal to): ˚qE = ˚qT = ˚qI =
F(t) F(t) , − x FM FM F(t)
x FM
F(t) y
FM
− −
F(t)
(107)
,
(108)
F(t) , FM
(109)
y
FM
y
x and F where FM are the maximal fluorescence values of the first M and second, respectively, threshold times determined from the relaxation analysis (see above). Note that determination of the two threshold times might be distorted by analysis of the NPQ time course as found by Holzwarth et al. (2013); see above. Further, x is an analog of F 10 min and F y is an analog of assuming that FM M M 3h then Eq. (108) is identical with Eq. (106), the same as Eq. FM (107) with Eqs. (104) and (109) with Eq. (103). Then, from mathematical point of view, there is no difference if only one threshold time or two threshold times are considered in particular theories because the related equations are essentially the same in both the cases. However, the physiological meaning of the parameters is absolutely related to the length of dark recovery chosen for calculation. Therefore, the threshold times should be carefully determined for a particular study by preliminary measurements and by a cor -levels directly and not from NPQ values rect procedure (from FM -levels). calculated from FM - and FM In the case when maximal quantum yields of PSII photochemistry are used for determination of physiological state of the sample, during dark Kornyeyev et al. (2013) evaluated the course of FV /FM recovery and analyzed it in the same way as Guadagno et al. (2010) analyzed the NPQ course (see above). Kornyeyev et al. (2013) determined two threshold times reflecting variations in the slope of the course and denoted the related F /F values as (F /F )x FV /FM V M V M )y , (F /F )y being higher than (F /F )x because q and and (FV /FM T V M V M y qE quenching are relaxed when (FV /FM ) is measured whereas only
145
)x is measured. By compariqE quenching is relaxed when (FV /FM son of the FV /FM values and considering also the F(t)/F0 ratio to account for a closure of a fraction of PSII centers, following equations were introduced by Kornyeyev et al. (2013) for the sum of quantum yields of qE and qT quenching, and the quantum yield for the qE quenching itself:
˚qE+qT =
1−
˚qE =
1−
FV /FM
)y (FV /FM
FV /FM
) (FV /FM
x
F(t) , F0
(110)
F(t) . F0
(111)
The quantum yield of the qT quenching can be obtained by subtracting Eq. (111) from Eq. (110) which leads to:
˚qT =
FV /FM
) (FV /FM
x
−
FV /FM
) (FV /FM
y
F(t) . F0
(112)
The quantum yield of the qI quenching can be estimated from:
˚qI =
1−
) (FV /FM
FV /FM
y
FV /FM
)y (FV /FM
F(t) . F0
(113)
Kornyeyev et al. (2013) showed experimentally with beet leaves that the quantum yields of the non-photochemical quenching processes calculated according to the methods based on values of maximal fluorescence signals (Eqs. (107)–(109)) and based on values of maximal quantum yield of PSII photochemistry (Eqs. (111)–(113)) give very similar results and can be used as alternatives. The small deviations between the results were assigned to inaccuracy of the F0 measurements. Because of prac method is less sensitive to changes in the tical reasons (the FV /FM place of measurements on the leaf surface and experimental errors method), in measurements of the fluorescence levels than the FM method in Kornyeyev et al. (2013) recommended using the FV /FM the field. On the other hand, the equations used for definition of method lack physiologithe quantum yields based on the FV /FM cal background and the above mentioned good correlation of these quantum yields (Eqs. (111)–(113)) with the quantum yields based method (Eqs. (107)–(109)) might be accidental. on the FM Summary All the theories of the photosynthetic energy partitioning are in fact based on two original approaches: by Demmig-Adams et al. (1996) on one hand and by Genty et al. (1996) and Cailly et al. (1996) on the other hand. From these two approaches the first one seems to be not completely correct because it includes the E parameter which has no physiological meaning. Then, all the theories which are based on the Demmig-Adams approach (Kornyeyev et al., 2001; Kato et al., 2003; Hikosaka et al., 2004; Ishida et al., 2011) should be avoided because they carry the same problematic issue (presence of the E parameter) as the original one. On the other hand, the original approach by Genty et al. (1996) and Cailly et al. (1996), later re-discovered by Hendrickson et al. (2004), and the similar approach by Kramer et al. (2004), do not include the E parameter and are, together with their improved derivatives (Hendrickson et al., 2005; Kornyeyev and Hendrickson, 2007; Kornyeyev and Holaday, 2008; Porcar-Castell et al., 2008; Ahn et al., 2009; Kasajima et al., 2009; Guadagno et al., 2010; Porcar-Castell, 2011; Kornyeyev et al., 2013), thus more suitable for description of photosynthetic energy partitioning. From the above mentioned more suitable theories it is not possible to say which the only valid one is because they describe different states of the photosynthetic apparatus caused by different conditions (active centers only vs. active and inactive centers vs. centers with sustained non-photochemical quenching) and they
146
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147
consider different levels of resolution of particular processes (total regulatory non-photochemical quenching vs. resolution of the qE and qT quenching, in addition to the qI quenching of the photoinactivated centers). The highest amount of information, i.e., resolution of the quantum yield of non-photochemical quenching into quantum yields of qE , qT , and qI quenching, is obtained when also the relaxation analysis is followed after the quenching analysis (Ahn et al., 2009; Kasajima et al., 2009; Guadagno et al., 2010). These measurements are, however, more time consuming when compared with the measurements of the quenching analysis alone, and therefore, they may not be suitable for some kind of experiments. Similarly, resolution of the photoinactivated centers (Kornyeyev et al., 2001; Hikosaka et al., 2004; Hendrickson et al., 2005; Kornyeyev and Hendrickson, 2007; Kornyeyev and Holaday, 2008) requires more time than the quenching analysis alone because the FV /FM ratio must be measured in darkness after the light treatment, which might require up to several hours of darkness. However, if a high intensity of excitation light is used for illumination, promoting formation of photoinactived centers, these theories are the only way for correct determination of photosynthetic energy partitioning. When overwintering evergreens are explored, the two theories by Porcar-Castell and coworkers (PorcarCastell et al., 2008; Porcar-Castell, 2011) can be used from which the later theory of Porcar-Castell (2011) is the better one because it provides in addition to the quantum yield of the sustained regulatory non-photochemical quenching also information about the quantum yield of the light-induced regulatory non-photochemical quenching. On the other hand, if the researcher needs only the basic information about the photosynthetic energy partitioning and/or there is no time to measure dark-relaxed maximal fluorescence, the two theories (Genty et al., 1996; Cailly et al., 1996; Hendrickson et al., 2004 vs. Kramer et al., 2004) are enough to be used. In summary, many theories of the energy partitioning in photosynthesis have been developed so far and a researcher can find the approach which is the most suitable for his/her requirements and possibilities. Acknowledgements This work was supported by the Czech National Program of Sustainability I, grant number LO1204. D.L. thanks the anonymous reviewers for their useful comments. References Ahn TK, Avenson TJ, Peers G, Li Z, Dall’Osto L, Bassi R, et al. Investigating energy partitioning during photosynthesis using an expanded quantum yield convention. Chem Phys 2009;357:151–8. Baker NR. Chlorophyll fluorescence: a probe of photosynthesis in vivo. Annu Rev Plant Biol 2008;59(89):113. Bilger W, Björkman O. Role of the xanthophyll cycle in photoprotection elucidated by measurements of light-induced absorbance changes, fluorescence and photosynthesis in leaves of Hedera canariensis. Photosynth Res 1990;25:173–85. Bilger W, Schreiber U. Energy-dependent quenching of dark-level chlorophyll fluorescence in intact leaves. Photosynth Res 1986;10:303–8. Bradbury M, Baker NR. Analysis of the slow phases of the in vivo chlorophyll fluorescence induction curve. Changes in the redox state of photosystem II electron acceptors and fluorescence emission from photosystems I and II. Biochim Biophys Acta 1981;13:542–51. Cailly AL, Rizzal F, Genty B, Harbinson J. Fate of excitation at PS II in leaves, the nonphotochemical side. Plant Physiol Biochem 1996;86 [special issue]. Chen C, Zhang D, Li P, Ma F. Partitioning of absorbed light energy differed between the sun-exposed side and the shaded side of apple fruits under high light conditions. Plant Physiol Biochem 2012;60:12–7. Demmig-Adams B, Adams WW, Barker DH, Logan BA, Bowling BA, Verhoeven AS. Using chlorophyll fluorescence to assess the fraction of absorbed light allocated to thermal dissipation of excess excitation. Physiol Plant 1996;98:253–64. Demmig-Adams B, Garab G, Govindjee Adams W III, editors. Non-photochemical quenching and energy dissipation in plants, algae and cyanobacteria. Dordrecht: Kluwer; 2014. Duysens LNM, Sweers HE. Mechanism of the two photochemical reactions in algae as studied by means of fluorescence. In: Japanese society of plant physiologists,
editor. Studies on microalgae and photosynthetic bacteria. Tokyo: University of Tokyo Press; 1963. p. 353–72. Earl HJ, Ennahli S. Estimating photosynthetic electron transport via chlorophyll fluorometry without photosystem II light saturation. Photosynth Res 2004;82:177–86. Genty B, Briantais J-M, Baker NR. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochim Biophys Acta 1989;990:87–92. Genty B, Harbinson J, Cailly AL, Rizza F. Fate of excitation at PS II in leaves: the non-photochemical side. In: Book of abstract from the third BBSRC Robert Hill symposium on photosynthesis; 1996, Abstract no. P28. Goss R, Lepetit B. Biodiversity of NPQ. J Plant Physiol 2014;172:13–32. Govindjee. Sixty-three years since Kautsky: chlorophyll a florescence. Aust J Plant Physiol 1995;22:131–60. Guadagno CR, Virzo De Santo A, D’Ambrosio N. A revised energy partitioning approach to assess the yields of non-photochemical quenching components. Biochim Biophys Acta 2010;1797:525–30. Harbinson J. Improving the accuracy of chlorophyll fluorescence measurements. Plant Cell Environ 2013;36:1751–4. Havaux M, Strasser RJ, Greppin H. A theoretical and experimental analysis of the qP and qN coefficients of chlorophyll fluorescence quenching and their relation to photochemical and nonphotochemical events. Photosynth Res 1991;27:41–55. Hendrickson L, Furbank RT, Chow WS. A simple alternative approach to assessing the fate of absorbed light energy using chlorophyll fluorescence. Photosynth Res 2004;82:73–81. Hendrickson L, Föerster B, Pogson BJ, Chow WS. A simple chlorophyll fluorescence parameter that correlates with the rate coefficient of photoinactivation of photosystem II. Photosynth Res 2005;84:43–9. Henriques FS. Leaf chlorophyll fluorescence: background and fundamentals for plant biologists. Bot Rev 2009;75:249–70. Hikosaka K, Kato MC, Hirose T. Photosynthetic rates and partitioning of absorbed light energy in photoinhibited leaves. Physiol Plant 2004;121:699–708. Hodges M, Cornic G, Briantais J-M. Chlorophyll fluorescence from spinach leaves: resolution of non-photochemical quenching. Biochim Biophys Acta 1989;974:289–93. Holt NE, Zigmantas D, Valkunas L, Li X-P, Niyogi KK, Fleming GR. Carotenoid cation formation and the regulation of photosynthetic light harvesting. Science 2005;307:433–6. Holzwarth AR, Miloslavina Y, Nilkens M, Jahns P. Identification of two quenching sites active in the regulation of photosynthetic light-harvesting studied by timeresolved fluorescence. Chem Phys Lett 2009;483:262–7. Holzwarth AR, Lenk D, Jahns P. On the analysis of non-photochemical chlorophyll fluorescence quenching curves I. Theoretical considerations. Biochim Biophys Acta 2013;1827:786–92. Horton P. Relations between electron transport and carbon assimilation; simultaneous measurement of chlorophyll fluorescence, transthylakoid pH gradient and O2 evolution in isolated chloroplasts. Proc R Soc Lond B 1983;217:405–16. Horton P, Hague A. Studies on the induction of chlorophyll fluorescence in isolated barley protoplasts IV. Resolution of non-photochemical quenching. Biochim Biophys Acta 1988;932:107–15. Ishida S, Merita K, Kishine M, Takabayashi A, Murakami R, Takeda S, et al. Allocation of absorbed light energy in PSII to thermal dissipations in the presence or absence of PsbS subunits of rice. Plant Cell Physiol 2011;52:1822–31. ´ Jabłonski A. Efficiency of anti-stokes fluorescence in dyes. Nature 1933;131:839–40. Johnson MP, Ruban A. Restoration of rapidly reversible photoprotective energy dissipation in the absence of PsbS protein by enhanced pH. J Biol Chem 2011;286:19973–81. Joliot A, Joliot P. Étude cinétique de la réaction photochimique libérant l‘oxygéne au cours de la photosynthése. C R Acad Sci Paris 1964;258:4622–5. Kasajima I, Takahara K, Kawai-Yamada M, Uchimiya H. Estimation of the relative sizes of rate constants for chlorophyll de-excitation processes through comparison of inverse fluorescence intensities. Plant Cell Physiol 2009;50:1600–16. Kato MC, Hikosaka K, Hirotsu N, Makino A, Hirose T. The excess light energy that is neither utilized in photosynthesis nor dissipated by photoprotective mechanisms determines the rate of photoinactivation in photosystem II. Plant Cell Physiol 2003;44:318–25. Kautsky H, Hirsch A. Neue Versuche zur Kohlensureassimilation. Naturwissenschafte 1931;19:964. Kitajima M, Butler WL. Quenching of chlorophyll fluorescence and primary photochemistry in chloroplasts by dibromothymoquinone. Biochim Biophys Acta 1975;376:105–15. Klughammer C, Schreiber U. Complementary PS II quantum yields calculated from simple fluorescence parameters measured by PAM fluorometry and the saturation pulse method. PAM Appl Notes 2008;1:27–35. Kornyeyev D, Hendrickson L. Energy partitioning in photosystem II complexes subjected to photoinhibitory treatment. Funct Plant Biol 2007;34:214–20. Kornyeyev D, Holaday AS. Corrections to current approaches used to calculate energy partitioning in photosystem 2. Photosynthetica 2008;46:170–8. Kornyeyev D, Logan BA, Patron P, Allen RD, Holaday AS. Enhanced photochemical light utilization and decreased chilling-induced photoinhibition of photosystem II in cotton overexpressing genes encoding chloroplast-targeted antioxidant enzymes. Physiol Plant 2001;113:323–31. Kornyeyev D, Guadagno CR, D’Ambrosio N. Assessment of energy partitioning in PSII complexes using chlorophyll fluorescence: reviewing the different approaches toward the definition of a unified method. Photosynthetica 2013;51: 477–80.
D. Lazár / Journal of Plant Physiology 175 (2015) 131–147 Kramer DM, Johnson G, Kiirats O, Edwards GE. New fluorescence parameters for the determination of QA redox state and excitation energy fluxes. Photosynth Res 2004;79:209–18. Krause HG, Jahns P. Pulse amplitude modulated chlorophyll fluorometry and its application in plant science. In: Green BR, Parson WW, editors. Light-harvesting antennas in photosynthesis. Dordrecht: Kluwer; 2003. p. 373–99. Laisk A, Oja V, Rasulov B, Eichelmann H, Sumberg A. Quantum yields and rate constants of photochemical and nonphotochemical excitation quenching. Experiment and model. Plant Physiol 1997;11:803–15. Lavergne J, Leci E. Properties of inactive photosystem II centers. Photosynth Res 1993;35:323–43. Lazár D. Chlorophyll a fluorescence induction. Biochim Biophys Acta 1999;1412:1–28. Lazár D. The polyphasic chlorophyll a fluorescence rise measured under high intensity of exciting light. Funct Plant Biol 2006;33:9–30. Lazár D. Simulations show that a small part of variable chlorophyll a fluorescence originates in photosystem I and contributes to overall fluorescence rise. J Theor Biol 2013;335:249–64. Lazár D, Schansker G. Models of chlorophyll a fluorescence transients. In: Laisk A, Nedbal L, Govindjee, editors. Photosynthesis in silico: understanding complexity from molecules to ecosystems. Dordrecht: Springer; 2009. p. 85–123. Li X-P, Björkman O, Shih C, Grossmann AR, Rosenquist M, Jansson S, et al. A pigmentbinding protein essential for regulation of photosynthetic light harvesting. Nature 2000;403:391–5. Lichtenthaler HK, Buschmann C, Rinderle U, Schmuck G. Application of chlorophyll fluorescence in ecophysiology. Radiat Environ Biophys 1986;25:297–308. Lichtenthaler HK, Langsdorf G, Lenk S, Buschmann C. Chlorophyll fluorescence imaging of photosynthetic activity with the flash-lamp fluorescence imaging system. Photosynthetica 2005;43:355–69. Loriaux SD, Avenson TJ, Welles JM, Mcdermitt DK, Eckles RD, Riensche B, et al. Closing in on maximum yield of chlorophyll fluorescence using a single multiphase flash of sub-saturating intensity. Plant Cell Environ 2013;36:1755–70. Maxwell K, Johnson GN. Chlorophyll fluorescence – a practical guide. J Exp Bot 2000;51:659–68. Miyake C, Amako K, Shiraishi N, Sugimoto T. Acclimation of tobacco leaves to high light intensity drives the plastoquinone oxidation system – relationship among the fraction of open PSII centers, non-photochemical quenching of chl fluorescence and the maximum quantum yield of PSII in the dark. Plant Cell Physiol 2009;50:730–43. Nilkens M, Kress E, Lambrev P, Miloslavina Y, Müller M, Holzwarth AR, et al. Identification of a slowly inducible zeaxanthin-dependent component of nonphotochemical quenching of chlorophyll fluorescence generated under steadystate conditions in Arabidopsis. Biochim Biophys Acta 2010;1792:466–75. Oxborough K, Baker NR. Resolving chlorophyll a fluorescence images of photosynthetic efficiency into photochemical and non-photochemical components without measuring F0 . Photosynth Res – calculation of qP and FV /FM 1997;54:135–42. Papageorgiou G, Govindjee. Light-induced changes in the fluorescence yield of chlorophyll a in vivo II. Chlorella pyrenoidosa. Biophys J 1968;8:1316–28. Papageorgiou GC, Tsimilli-Michael M, Stamatakis K. The fast and slow kinetics of chlorophyll a fluorescence induction in plants, algae and cyanobacteria: a viewpoint. Photosynth Res 2007;94:275–90. Porcar-Castell A. A high-resolution portrait of the annual dynamics of photochemical and non-photochemical quenching in needles of Pinus sylvestris. Physiol Plant 2011;143:139–53. Porcar-Castell A, Juurola E, Nikinmaa E, Berninger F, Ensminger I, Hari P. Seasonal acclimation of photosystem II in Pinus sylvestris I. Estimating the rate constants of sustained thermal energy dissipation and photochemistry. Tree Physiol 2008;28:1474–82. Porcar-Castell A, Garcia-Plazaola JI, Nichol CJ, Kolari P, Olascoaga B, Kuusinen N, et al. Physiology of the seasonal relationship between the photochemical reflectance index and photosynthetic light use efficiency. Oecologia 2012;170:313–23.
147
Porcar-Castell A, Tyystjärvi E, Atherton J, van der Tol C, Flexas J, Pfündel EE, et al. Linking chlorophyll a fluorescence to photosynthesis for remote sensing applications: mechanisms and challenges. J Exp Bot 2014;65:4065–95. Quick WP, Horton P. Studies on the induction of chlorophyll fluorescence in barley protoplasts II: resolution of fluorescence quenching by redox state and the transthylakoid pH gradient. Proc R Soc Lond B 1984;220:371–82. Quick WP, Stitt M. An examination of factors contributing to non-photochemical quenching of chlorophyll fluorescence in barley leaves. Biochim Biophys Acta 1989;977:287–96. Roháˇcek K. Chlorophyll fluorescence parameters: the definitions, photosynthetic meaning, and mutual relationships. Photosynthetica 2002;40:13–29. Roháˇcek K. Method for resolution and quantification of components of the nonphotochemical quenching (qN ). Photosynth Res 2010;105:101–13. Rubin A, Riznichenko G. Modeling of the primary processes in a photosynthetic membrane. In: Laisk A, Nedbal L, Govindjee, editors. Photosynthesis in silico: understanding complexity from molecules to ecosystems. Dordrecht: Springer; 2009. p. 151–76. Samson G, Bruce D. Origins of the low yield of chlorophyll a fluorescence induced by single turnover flash in spinach thylakoids. Biochim Biophys Acta 1996;1276:147–53. Samson O, Práˇsil O, Yaakoubd B. Photochemical and thermal phases of chlorophyll a fluorescence. Photosynthetica 1999;37:163–82. Schansker G, Tóth SZ, Strasser RJ. Dark recovery of the Chl a fluorescence transient (OJIP) after light adaptation: the qT-component of non-photochemical quenching is related to an activated photosystem I acceptor side. Biochim Biophys Acta 2006;1757:787–97. Schansker G, Tóth SZ, Holzwarth AR, Garab G. Chlorophyll a fluorescence: beyond the limits of the QA model. Photosynth Res 2014;120:43–58. Schreiber U, Schliwa U, Bilger W. Continuous recording of photochemical and non-photochemical chlorophyll fluorescence quenching with a new type of modulation fluorometer. Photosynth Res 1986;10:51–62. Govindjee Stirbet A. Chlorophyll a fluorescence induction: a personal perspective of the thermal phase, the J-I-P rise. Photosynth Res 2012;113:15–61. Stirbet A. Excitonic connectivity between photosystem II units: what is it, and how to measure it? Photosynth Res 2013;116:189–214. Stirbet A, Govindjee. On the relation between the Kautsky effect (chlorophyll a fluorescence induction) and photosystem II: basics and applications of the OJIP fluorescence transient. J Photochem Photobiol B 2011;104:236–57. Stirbet A, Riznichenko GY, Rubin AB, Govindjee. Modeling chlorophyll a fluorescence transient: relation to photosynthesis. Biochemistry (Moscow) 2014;79:291–323. van Kooten O, Snel J. The use of chlorophyll nomenclature in plant stress physiology. Photosynth Res 1990;25:147–50. Verhoeven A. Sustained energy dissipation in winter evergreens. New Phytol 2014;201:57–65. Vredenberg W, Práˇsil O. Modeling of chlorophyll a fluorescence kinetics in plant cells: derivation of a descriptive algorithm. In: Laisk A, Nedbal L, Govindjee, editors. Photosynthesis in silico: understanding complexity from molecules to ecosystems. Dordrecht: Springer; 2009. p. 125–49. Walker DA, Sivak MN, Prinsley RT, Cheesbrough JK. Simultaneous measurement of oscillations in oxygen evolution in chlorophyll a fluorescence in leaf pieces. Plant Physiol 1983;73:542–9. Walters RG, Horton P. Resolution of components of non-photochemical chlorophyll fluorescence quenching in barley leaves. Photosynth Res 1991;27:121–33. Walters RG, Horton P. Theoretical assessment of alternative mechanisms for nonphotochemical quenching of PS II fluorescence in barley leaves. Photosynth Res 1993;36:119–39. Yamagishi A, Satoh K, Katoh S. Fluorescence induction in chloroplasts isolated from the green alga Bryopsis maxima III. A fluorescence transient indicating proton gradient across the thylakoid membrane. Plant Cell Physiol 1978;19: 17–25.