Photosynthesis Research

, Volume 117, Issue 1, pp 529–546

Photosynthetic electron transport and specific photoprotective responses in wheat leaves under drought stress


  • Marek Zivcak
    • Department of Plant PhysiologySlovak Agricultural University
    • Department of Plant PhysiologySlovak Agricultural University
  • Zuzana Balatova
    • Department of Plant PhysiologySlovak Agricultural University
  • Petra Drevenakova
    • Department of Plant PhysiologySlovak Agricultural University
  • Katarina Olsovska
    • Department of Plant PhysiologySlovak Agricultural University
  • Hazem M. Kalaji
    • Department of Plant Physiology, Faculty of Agriculture and BiologyWarsaw Agricultural University SGGW
  • Xinghong Yang
    • State Key Laboratory of Crop Biology, Shandong Key Laboratory of Crop BiologyCollege of Life Sciences, Shandong Agricultural University
  • Suleyman I. Allakhverdiev
    • Institute of Plant Physiology, Russian Academy of Sciences
    • Institute of Basic Biological ProblemsRussian Academy of Sciences
Regular Paper

DOI: 10.1007/s11120-013-9885-3

Cite this article as:
Zivcak, M., Brestic, M., Balatova, Z. et al. Photosynth Res (2013) 117: 529. doi:10.1007/s11120-013-9885-3


The photosynthetic responses of wheat (Triticum aestivum L.) leaves to different levels of drought stress were analyzed in potted plants cultivated in growth chamber under moderate light. Low-to-medium drought stress was induced by limiting irrigation, maintaining 20 % of soil water holding capacity for 14 days followed by 3 days without water supply to induce severe stress. Measurements of CO2 exchange and photosystem II (PSII) yield (by chlorophyll fluorescence) were followed by simultaneous measurements of yield of PSI (by P700 absorbance changes) and that of PSII. Drought stress gradually decreased PSII electron transport, but the capacity for nonphotochemical quenching increased more slowly until there was a large decrease in leaf relative water content (where the photosynthetic rate had decreased by half or more). We identified a substantial part of PSII electron transport, which was not used by carbon assimilation or by photorespiration, which clearly indicates activities of alternative electron sinks. Decreasing the fraction of light absorbed by PSII and increasing the fraction absorbed by PSI with increasing drought stress (rather than assuming equal absorption by the two photosystems) support a proposed function of PSI cyclic electron flow to generate a proton-motive force to activate nonphotochemical dissipation of energy, and it is consistent with the observed accumulation of oxidized P700 which causes a decrease in PSI electron acceptors. Our results support the roles of alternative electron sinks (either from PSII or PSI) and cyclic electron flow in photoprotection of PSII and PSI in drought stress conditions. In future studies on plant stress, analyses of the partitioning of absorbed energy between photosystems are needed for interpreting flux through linear electron flow, PSI cyclic electron flow, along with alternative electron sinks.


Drought stressWheatPhotosynthetic electron transportCyclic electron transport around PSIPhotosystem stoichiometryChlorophyll fluorescenceAlternative electron sinks


\( A_{{{\text{CO}}_{ 2} }} \)

CO2 assimilation rate

Cyt b6/f

Cytochrome b6/f


Mesophyll conductance


Stomatal conductance


Light emitting diode


Light harvesting complex


Nonphotochemical quenching


Primary electron donor of PSI (reduced form)


Primary electron donor of PSI (oxidized form)


Photosynthetic active radiation




Photosystem I


Photosystem II


Primary PSII acceptor


pH-dependent energy dissipation


Ribulose 1,5-bisphosphate


Relative water content


Transthylakoid pH gradient


Water potential


Drought is one of the main factors negatively affecting the productivity of agricultural or natural ecosystems (Passioura 2007; Ciais et al. 2005) and the diversity of plant species (Engelbrecht et al. 2007). It has also a global impact on carbon gain (Buermann et al. 2007). Drought mostly affects plants through its effects on photosynthesis. However, photosynthetic responses to drought stress are complex, involving the interplay of different structural levels at different time scales in relation to plant development (Chaves et al. 2009).

The effects of drought stress on assimilation can be direct or indirect. Direct effects are through decreased CO2 availability to chloroplasts due to a decrease in stomatal and mesophyll conductances (Chaves 1991; Lal et al. 1996; Chaves et al. 2002; Flexas et al. 2004, 2012), or are a consequence of changes in photosynthetic metabolism (Tezara et al. 1999; Lawlor and Cornic 2002; Maroco et al. 2002; Parry et al. 2002; etc.). In conditions of limited CO2 diffusion, photorespiration lowers the energetic efficiency of photosynthesis in C3 plants (Ogren 1984). Indirect or secondary effects are through oxidative stress, which can contribute to the nonstomatal limitation of photosynthesis. This can seriously affect the leaf’s photosynthetic machinery, mostly as a result of the interaction between drought and excessive light or under multiple stress conditions (Ort 2001; Chaves and Oliveira 2004; Foyer and Noctor 2009).

Photosynthetic responses to water deficit are influenced by stress intensity, duration, and rate of progression—these factors determining whether mitigation processes associated with acclimation will be initiated (Chaves et al. 2009). The photosynthetic machinery is necessarily flexible, utilizing several mechanisms to prevent the harmful effects of highly reactive intermediates during the conversion of light into usable forms of energy. These mechanisms also need to insure that the output ratio of ATP/NADPH matches the demands of plant metabolism (Kramer et al. 2004; Cruz et al. 2005; Kramer and Evans 2011). The central process in higher plants is nonphotochemical quenching, by which excess light energy is harmlessly dissipated as heat (Muller et al. 2001). Nonphotochemical quenching typically dominates a rapidly reversible component called qE, resulting in the thermal dissipation of excess absorbed light energy in the light-harvesting antenna of PSII. qE is induced by a low thylakoid lumen pH and a high ΔpH that are generated by photosynthetic electron transport under excess light conditions.

These processes represent a form of feedback regulation of the light-dependent reactions of photosynthesis (Niyogi et al. 2005). The low pH of thylakoid lumen activates qE, by protonating the protein PsbS (Li et al. 2000) and by activating violaxanthin deepoxidase, which converts violaxanthin to antheraxanthin and zeaxanthin in xanthophyll cycle (Demmig-Adams 1990). There is also evidence for dissipation of excess energy in the reaction center of PSII where electron transfer is initiated (Ivanov et al. 2008). In addition, under low CO2 supply, energy from linear electron flow can in part be redirected to photorespiration (Kozaki and Takeba 1996) or to some alternative pathways—mainly the water–water cycle (Asada 1999). In photosystem I, the cyclic electron flow is activated (Golding and Johnson 2003; Golding et al. 2004; reviewed in Johnson 2005, 2011; Miyake 2010). There is controversy over the role and contribution of this process to photoprotection and balancing ATP/NADPH production ratio. Munekage and coworkers demonstrated the essential role of cyclic electron flow both for photoprotection and producing additional ATP for Calvin cycle (Munekage et al. 2002, 2004). However, the contribution of fast cyclic electron transport to the buildup of ΔpH on thylakoid membrane and ATP synthesis has been strongly questioned by researchers that still support the idea of photoprotective role in stress conditions (Laisk et al. 2007, 2010). Cruz et al. (2005) have argued for the important role of mechanisms (e.g., by changing the proton conductivity of the ATP-synthase) that modulate qE sensitivity to ΔpH; such mechanisms can efficiently modulate ATP/NADPH output ratio by interacting with linear or alternative electron flows.

Drought stress has also been observed to induce long-term changes in the structure, content, and activity of individual photosynthetic components (Kohzuma et al. 2009). Specifically, drought stress has been associated with a decrease in leaf absorbance i.e., a decrease of chlorophyll content (Balaguer et al. 2002; Maroco et al. 2002), an increase of the xanthophyll pool (Maroco et al. 2002), a decrease of expression of ATP synthase (Tezara et al. 1999; Kohzuma et al. 2009), a decrease in expression of cyt b6/f (Kohzuma et al. 2009) or expression of functional proteins (Zadraznik et al. 2013), etc.

In this article, we present data on the effects of drought stress on photosynthesis in wheat (T. aestivum L.). Wheat is one of the most important crops worldwide (Bonjean and Angus 2001). It is also interesting because, as grasses are evolutionary younger than dicots, they may differ in physiological responses to drought (e.g., Flexas et al. 2012). There are many studies showing the high acclimation capacity of wheat photosynthetic apparatus to environmental stresses (e.g., Gray et al. 1996; Mehta et al. 2010; Mathur et al. 2011; Vassileva et al. 2012; Brestic et al. 2012). Moreover, wheat and its relatives (e.g., barley) are frequently used in photosynthesis research, providing a broad baseline for comparisons as well as numerical inputs to photosynthetic models—although these are mostly valid only for non-stressed (NS) conditions (e.g., Yin et al. 2009).

Our study focuses on functional changes at the level of photosynthetic electron transport and related photosynthetic processes affected by different levels of prolonged drought stress. We use a relatively long duration of drought stress during moderate light intensity, allowing us to observe physiological changes that can be attributed to acclimation rather than to injury (e.g., by photoinhibition or by oxidative stress). Based on our results, we discuss the roles of alternative electron sinks and cyclic electron flow in the protection of PSII and PSI in conditions of limited water supply, and the effects of drought stress on the distribution of absorbed light energy between photosystems I and II.

Materials and methods

Plant material

After vernalisation at 6 °C for 2 months, plants of winter wheat (T. aestivum L., cv. Ilona) were transplanted and cultivated individually in pots (0.5 l) containing standard peat substrate Klassman TS1. The pots were regularly irrigated and occasionally fertilized using standard liquid fertilizer with micronutrients. Before and during the drought stress protocol, plants were grown in a growth chamber with artificial light provided by fluorescent tubes (Osram Fluora) with maxima in red and blue spectral regions; the incident PAR at leaf level was app. 200 μmol photons m−2 s−1. The photoperiod was 14-h light/10-h dark cycle with light intensity reduced by half during the first and the last hours of light period. Temperature ranged between 18 °C at night and 23 °C during the light period.

Drought stress treatment and measuring protocol

Drought stress in wheat plants was induced after the seventh leaf from the base appeared. Plants within the experiment were divided into two groups: “non-stressed” and “drought-stressed.” Before the start of the experiment, all pots were fully irrigated to achieve the maximum water capacity. All pots were then weighed. The soil of some pots was dried to get the percentage of soil dry mass in a water saturated soil; this was used to calculate dry mass of soil in each experimental pot. The weight of the empty pot with the dry mass was then subtracted from each actual pot weight to get the maximum water capacity (from fully irrigated pots) or actual water content. Soil water content (expressed as % of soil water capacity) in each pot was calculated as the ratio of actual to maximum water content. All pots were weighed daily, and water was supplied to 70 % of soil water capacity in control plants and 20 % of soil water capacity in stressed plants—the soil water content below ~15 % of soil water capacity caused wilting. After 14 days, the water supply for drought-stressed plants was stopped for 3 days. Measurements were taken from the leaves of drought-stressed plants during the 7th–14th day period (before the full withholding of irrigation). Leaves from this period are referred to as “initial drought stress” (ID) or “moderate drought stress” (MD), depending on leaf water status—water availability decreased during the day, see Fig. 1 for respective leaf water content. Measurements after full withholding of irrigation were taken from the 14th to 16th day of the experiment, referred to as the “severe drought” (SD) period. In addition to measurements on stressed plants, leaves of control plants were also tested and are labeled as “non-stressed.”
Fig. 1

Definition of drought stress levels based on the values of leaf relative water content (RWC), leaf water potential, and stomatal conductance. a Relationship between the leaf water potential (ΨW) and the leaf relative water content (RWC) measured in wheat leaves during drought stress. b Relationship between the leaf RWC and stomatal conductance (gs) measured in wheat leaves during drought stress. The four levels of drought stress were defined as follows (points within squares): NS no stress, ID initial drought, MD moderate drought, and SD severe drought

All measurements were taken from the last appearing leaf on the plant. The simultaneous measurements of CO2 assimilation with chlorophyll fluorescence, and simultaneous measurements of P700 and chlorophyll fluorescence were performed on the same leaf. Between the different measurements, the plants were exposed to ambient light for at least half an hour to eliminate the effect of previous measurements. The same leaf was then used for destructive determination of water potential and relative water content (RWC).

Measurements of the leaf water status

Leaf water status was measured both as the RWC and as the leaf water potential (ΨW). The samples were taken from the middle part of the leaf blade; this part was previously used also in all other measurements.

The RWC was determined using the fresh weight (FW) measured immediately after taking the sample. After that, the leaf sample was saturated in distilled water for ~12 h at a low temperature (4 °C) in the dark to obtain the saturated weight (SW). Finally, the sample was dried at 80 °C for 12 h to obtain dry weight (DW). RWC was calculated according to Turner (1981) as RWC (%) = (FW  DW)/(SW − DW)*100.

Leaf water potential (ΨW) of fresh leaf segments was measured by the psychrometric method using microvoltmeter Wescor HR-33 with measuring chamber C-52 (Wescor, USA). Three segments of each sample were measured, and the average value was used in the analyses.

Simultaneous measurements of gas exchange and chlorophyll fluorescence

The induction curve was recorded using a Licor 6400 gasometer (Licor, USA) with simultaneous measurement of chlorophyll fluorescence. Before the measurements, plants were exposed to ambient light in the growth chamber for at least 30 min. Immediately before the measurements, plants were dark adapted for 20 min in dark box and for app. 3 min in the measuring head. The F0 and FM values were then determined using saturation flash (6,500 μmol photons m−2 s−1), and the actinic light provided by LED light unit (1,000 μmol photons m−2 s−1) was switched on. Within the measuring head, the following conditions were maintained: leaf temperature 20 °C, reference CO2 content 380 ppm, and ambient air humidity. Every 120 s, the gas exchange rates were measured followed by saturation pulse and far-red pulse for F0′ determination. The duration of the induction curve was at least 30 min—in some cases longer if it took more time to reach the steady state. The curves, as well as the steady-state values of photosynthetic parameters, were used in the analyses described below.

A range of measured and calculated fluorescence parameters were used in the analysis (Table 1). We also calculated electron transport fluxes based on gas exchange measurements. The rate of linear electron transport based on requirements to support photosynthesis was calculated according to Harley et al. (1992); gm was considered to be infinite: \( J_{\text{g}} = 4\left( {A_{{{\text{CO}}_{ 2} }} + R_{\text{L}} } \right)\left( {c_{\text{i}} + 2\Upgamma^{*} } \right)/\left( {c_{\text{i}} - \Upgamma^{*} } \right) \), or where gm equal to gs was used: \( J_{\text{g}} = 4\left[ {A_{{{\text{CO}}_{ 2} }} + R_{\text{L}} } \right]\left[ {\left( {c_{\text{i}} - A_{{{\text{CO}}_{ 2} }} /g_{\text{m}} } \right) + 2\Upgamma^{*} \left] / \right[\left( {c_{\text{i}} - A_{{{\text{CO}}_{ 2} }} /g_{\text{m}} } \right) - \Upgamma^{*} } \right] \), where ci represents intercellular content of CO2, \( A_{{{\text{CO}}_{2} }} \) represents measured CO2 assimilation (both values obtained from gas exchange measurements), RL represents mitochondrial respiration in light (the values reported by Yin et al. 2009 were used), and Γ* represents CO2 compensation point measured in the absence of respiration. We used for calculation the values of Γ* 37 μmol mol−1 in NS and ID leaves, 42 μmol mol−1 in MD leaves, and 64 μmol mol−1 in SD leaves as published by Galmes et al. (2006). This is in accordance with our unpublished measurements in wheat. The \( A_{{{\text{CO}}_{2} }} \) and RL values are used also for calculation of the rate of electron transport used in photorespiration: \( J_{\text{PR}} = 2/3\left[ {J_{\text{PSII}} - 4\left( {A_{{{\text{CO}}_{2} }} + R_{\text{L}} } \right)} \right] \) (Epron et al. 1995; Valentini et al. 1995); the JPSII is the electron transport calculated using the data obtained by chlorophyll fluorescence measurements: JPSII = αII · PAR · ΦPSII (PAR—photosynthetic active radiation incident on the surface of the leaf; ΦPSII—estimated effective quantum yield (efficiency) of PS II photochemistry at given PAR; αII, portion of incident PAR absorbed by PSII; the equal absorbance (0.84) of PAR by leaf was assumed in all samples; see also Table 1).
Table 1

Measured and calculated chlorophyll fluorescence parameters


Name and basic physiological interpretation

Measured or computed inputs for calculation of fluorescence and P700 parameters

 F, F

Fluorescence emission from dark- or light-adapted leaf, respectively


Minimum fluorescence from dark-adapted leaf, (PS II centers open)

 Fm, Fm

Maximum fluorescence from dark- or light-adapted leaf, respectively (PS II centers closed)

 FV = Fm − F0

Maximum variable fluorescence from dark-adapted leaf


Minimum fluorescence from light-adapted leaf


Steady state fluorescence at any light level


P700 absorbance at given light intensity

 Pm, Pm

Maximum P700 signal measured using saturation light pulse following after short far-red pre-illumination in dark (Pm) or light-adapted state

Chlorophyll fluorescence parameters derived from the saturation pulse analysis

 FV/Fm = 1 − (F0/Fm)

Estimated maximum quantum efficiency (yield) of PSII photochemistry (Kitajima and Butler 1975; Krause and Weis 1991; Schreiber et al. 1995)

 ΦPSII = (Fm − F′)/Fm

Estimated effective quantum yield (efficiency) of PSII photochemistry at given PAR (Genty et al. 1989)


Rate of linear electron transport in PSII at given PAR, and portion of PAR absorbed by PSII (αII) (Björkman and Demmig 1987; Genty et al. 1989)

 NPQ = (Fm − Fm′)/Fm

Nonphotochemical quenching of Fm (Schreiber et al. 1988; Walters and Horton 1991)

 SVo = F0/F0 − 1

Nonphotochemical quenching of F0 (Härtel and Lokstein 1995)

 qP = (Fm′ − Fs′)/(Fm′ − F0)

Coefficient of photochemical quenching based on the “puddle” model (i.e., unconnected PS II units) (Schreiber 1986; Björkman and Demmig 1987; Bilger and Björkman 1990)

 qL = qP·(F0/Fs′)

Coefficient of photochemical quenching based on the “lake” model (i.e., fully connected PS II units) (Kramer et al. 2004)

 ΦNO = 1/[NPQ + 1 + qL(Fm/F− 1)]

Quantum yield of nonregulated energy dissipation in PSII (Kramer et al. 2004)


Quantum yield of pH-dependent energy dissipation in PSII (Kramer et al. 2004)

PSI parameters derived from the saturation pulse analysis of P700 absorbance

 ΦPSI = (Pm′ − P)/Pm

Estimated effective quantum yield (efficiency) of PSI photochemistry at given PAR (Klughammer and Schreiber 1994)

 ΦND = P/Pm

Fraction of overall P700 that is oxidized in given state i.e., a lack of electrons coming from electron donors (Klughammer and Schreiber 1994)

 ΦNA = (Pm − Pm′)/Pm

Fraction of overall P700 that is oxidized in given state by saturation pulse i.e., a lack of electron acceptors (Klughammer and Schreiber 1994)

Simultaneous measurements of P700 redox state and chlorophyll fluorescence

The P700 redox state was measured using a Dual PAM-100 with a dual wavelength (830/875 nm) unit (following Klughammer and Schreiber 1994). Saturation pulses (10,000 μmol photons m−2 s−1), intended primarily for determination of chlorophyll fluorescence parameters, were also used for the assessment of P700 parameters. As in case of gas exchange measurements, analyzed plants were first exposed to ambient light in growth chamber for at least 30 min; immediately before the measurements, plants were dark adapted for 20 min in a dark box and for app. 2 min in the measuring head. After determination of F0, FM, and PM, the induction curve at light intensity similar to ambient (174 μmol photons m−2 s−1) was used for induction of photosynthesis. After a steady state was reached, a rapid light curve was initiated (light intensities 14, 30, 61, 103, 134, 174, 224, 347, 539, 833, 1036, 1295, 1602, and 1960 μmol photons m−2 s−1; duration of illumination at each light intensity was 30 s). After this, a light intensity of 174 μmol photons m−2 s−1 was applied again for a few minutes. The measuring protocol was finished by initiating another induction curve at moderate light intensity (539 μmol photons m−2 s−1) for 5 min, followed by measurements of dark recovery kinetics for 5 min, with saturation flashes every 20 s. In all chlorophyll fluorescence records, the correction for PSI fluorescence was done according to the method of Pfundel (1998).


Effect of drought stress: water relations and stomatal responses

To impose water stress, the soil water content in pots was initially maintained at 20 % of water capacity. The water supply was sufficient for plants to survive. Thus, it was possible to observe the long-term effects of drought stress including acclimation of photosynthetic structures and processes.

This treatment led to decreases of leaf RWC and water potential (ΨW) compared with NS plants (Fig. 1a) in which leaf water status was kept in a narrow range (RWC from 86 to 92 %; ΨW from −0.6 to −1.2 MPa). As irrigation was performed in the late afternoon (after all measurements had been taken), leaf water status varied considerably during the day. During the period with limited water supply, leaf RWC ranged between 85 and 63 % and ΨW fluctuated from −1.2 to −2.2 MPa. In the last period, water was not supplied to plants, and the value of these parameters fell below −2.2 MPa, leading to almost full stomatal closure. Thus, we can consider this as a critical leaf water potential value (Van den Berg et al. 2002). Both the soil and leaf water statuses affected the activity of stomata; there was an exponential relationship between RWC and stomatal conductance (gs) values measured after 30 min of illumination by strong actinic light (Fig. 1b).

In NS plants, the gs values ranged between 0.5 and 1.2 mol m−2 s−1, indicating that the stomata were fully open. During the period of limited irrigation, two different conditions were distinguished depending on water availability. In early measurements, the leaf water content was sufficient to maintain a relatively high maximum of gs (0.3–0.5 mol m−2 s−1); such measurements were denoted as “initial drought” (ID). After plants had used the available water, leaf water content and stomatal conductance decreased indicating “moderate drought” (MD) conditions; the maximum of gs during 30 min of illumination in these plants was in a relatively narrow range (0.1 and 0.3 mol m−2 s−1). After the water supply was stopped, in addition to a decrease of water content below the critical level, leaf stomata were almost fully closed. Measurements taken during this period were denoted as “severe drought” (SD).

Simultaneous measurements of gas exchange and chlorophyll fluorescence

The photosynthetic induction in dark-adapted leaves was monitored by saturation pulse method of chlorophyll fluorescence measured together with CO2 and water vapor exchange (Fig. 2).
Fig. 2

The values of selected photosynthetic parameters derived from gas exchange and chlorophyll fluorescence measurement during photosynthetic induction following dark-adapted state (20 min in dark before the measurements) toward the steady state. The measurements were recorded in air CO2 concentration of 380 μmol mol−1, leaf temperature 20 °C, and PAR 1,000 μmol photons m−2 s−1. a CO2 assimilation rate (\( A_{{{\text{CO}}_{ 2} }} \)). b Leaf stomatal conductance (gs). c PSII electron transport rate (JPSII) calculated based on measurements of PSII quantum yields, assuming the equal distribution of absorbed light between PSI and PSII. d Rate of electron transport consumed by carboxylation plus oxygenation of RuBP (Jg), calculated by the data from gas exchange measurements. e Ratio of electron transport consumed by photorespiration (JPR) to the total PSII electron transport. f Nonphotochemical quenching of maximum fluorescence (NPQ). The average values ± standard errors from 8 to 10 plants are presented. NS non-stressed samples, ID initial drought, MD moderate drought, SD severe drought

Recordings were continued as long as necessary to obtain the steady state as indicated by stable values of \( A_{{{\text{CO}}_{2} }} \) and gs—the minimum time used was 30 min. Recording began after 20 min of dark adaptation and F0, Fm measurements. These measurements were necessary for calculation of quenching parameters by the saturation pulse method (Schreiber 1986).

In addition to different steady-state values of stomatal conductance, leaves subjected to different drought stress levels also differed in the times required for stomata to start opening (Fig. 2a, b). Moreover, there was clear shift in the amount of time needed to achieve the steady state (<10 min in NS vs. more than 30 min in SD). While the electron transport rate calculated from gas exchange data (JG) closely followed the trend of ACO2, JPSII values (calculated using chlorophyll fluorescence data) were substantially higher under drought stress than the JG, especially under the SD (Fig. 2c vs d). The higher values of JPSII may indicate the presence of alternative electron flow and/or imprecise estimates of electron transport rates; both cases are discussed below.

The distribution of electrons within linear electron transport between RuBP carboxylation and photorespiration (estimated using the formula of Epron et al. 1995 and Valentini et al. 1995) reached the steady state very quickly in NS and ID (Fig. 2e), but stabilized much later in MD and SD. After 30 min of photosynthetic induction at high light, significantly higher photorespiration remained in MD and SD. In severely stressed plants, the proportion of electrons consumed in oxygenation was two times higher than in NS plants. Surprisingly, high stress levels did not lead to a progressive increase in nonphotochemical quenching (Fig. 2f). No significant differences in regulated PSII dissipation in high light were observed between the leaves of NS, ID, and MD plants. However, NPQ was significantly enhanced in SD leaves: NPQ increased after ~12 min, and this increase was apparent until the end of recording.

The steady-state values of key photosynthetic parameters derived from induction curves together with leaf water status parameters (Supplementary Table 1) reflect the effects of different level of drought stress, with the exception of Fv/Fm parameter that was almost unaffected even in SD, as was demonstrated also in our previous studies (e.g., Zivcak et al. 2008).

Leaf water status (RWC) and selected photosynthetic parameters measured in the steady state (Fig. 3) showed linear correlations with \( A_{{{\text{CO}}_{ 2} }} \), JPSII, and proportion of total electron flow associated with photorespiration (JPR/JPSII).
Fig. 3

Relationships between leaf water status (relative water content, RWC and water potential, ΨW) or stomatal conductance (gs), and the steady-state values of photosynthetic parameters derived from simultaneous measurements of gas exchange and chlorophyll fluorescence: CO2 assimilation (\( A_{{{\text{CO}}_{ 2} }} \)), intercellular CO2 content (ci), PSII electron transport (JPSII), ratio of electron transport consumed by photorespiration (JPR) to the total PSII electron transport, and nonphotochemical quenching (NPQ). The best-fit curves are presented

Nonphotochemical quenching grew exponentially with water loss (in the range of our observations). The intercellular CO2 content linearly decreased with decreasing RWC (ΨW). However, below 65 % of RWC or (−2 MPa of ΨW) the ci values significantly increased.

Using leaf stomatal conductance (gs) as a reference parameter, a clear logarithmic relationship with \( A_{{{\text{CO}}_{ 2} }} \) and JPSII as well as an exponential decay of the JPR/JPSII ratio can be observed. The relationship between gs and ci is more complicated, with some bifurcation of the trend at low gs related to leaf water status. This can be attributed to nonstomatal effects in SD plants such as a decrease in mesophyll conductance (cf. Flexas et al. 2012 for review) or changes in Rubisco enzymatic activities (Lawlor and Cornic 2002), and ATP synthesis (Tezara et al. 1999).

Simultaneous measurements of chlorophyll fluorescence and P700 absorbance

Measurements based on the saturation pulse method allow for the calculation of PSII quantum yields (Schreiber 1986) and PSI quantum yields (Klughammer and Schreiber 1994). On each leaf, the induction kinetics at moderate light, the rapid light curve, recovery at moderate light, the induction curve at high light, and dark recovery kinetics were recorded subsequently. The same leaves were also used for gasometric measurements and for measurements of leaf water status (see “Materials and methods” for details).

Complementary quantum yields of PSII (graphs left) and PSI (graphs right) were recorded within rapid light curves (Fig. 4).
Fig. 4

Light responses of complementary quantum yields of PSII and PSI recorded in plants subjected to different levels of drought stress. a The effective quantum yield of PSII (ΦPSII); b the effective quantum yield of PSI (ΦPSI); c the quantum yield of regulated nonphotochemical quenching in PSII (ΦNPQ); d the quantum yield of the PSI nonphotochemical quenching caused by the donor-side limitation, i.e., the fraction of overall P700 that is oxidized in a given state (ΦND); e the fraction of energy captured by PSII passively dissipated in form of heat and fluorescence (ΦNO); f the quantum yield of the PSI nonphotochemical quenching caused by the acceptor-side limitation, i.e., the fraction of overall P700 that cannot be oxidized in a given state (ΦNA). The small plots within each graph show the initial phase of each light curve. The rapid light curves were obtained after previous induction at moderate light; the duration of each interval with a given light intensity was 30 s (see "Materials and methods" for details). The average values ± standard errors from 8 to 10 plants are presented. NS non-stressed samples, ID initial drought, MD moderate drought, SD severe drought

The sum of three complementary quantum yields of PSII (ΦPSII, ΦNO, and ΦNPQ) or PSI (ΦPSI, ΦNA, and ΦND) is defined as unity. As anticipated, the PSII and PSI effective quantum yields (Fig. 4a, b) show a similar trend, both in accordance with previously presented results. The curves of quantum yield of regulated dissipation processes in PSII (ΦNPQ) are also in accordance with previously presented results. While NS, ID, and MD leaves show very similar trends, a steeper increase in ΦNPQ as well as higher maximum values were observed in SD leaves (Fig. 4c). The curves of nonregulated dissipation (ΦNO) are not significantly different among drought stress levels. The ΦND (Fig. 4d) represents the fraction of total P700 which is oxidized in given state (P700+/P700T). Hence, it is a measure of the PSI donor-side limitation causing nonphotochemical energy dissipation in PSI (Klughammer and Schreiber 1994). The trend of this parameter is very similar to (ΦNPQ). However, this parameter was sensitive to moderate (and insignificantly also to initial) drought, which was not seen in ΦNPQ. Furthermore, there was a fraction of P700 that was not reduced by saturation flash in a given state i.e., PSI acceptors (ΦNA) that are not fully oxidized (Fig. 4f). The acceptor-side limitation was generally low, becoming negligible at moderate-to-high light intensities, while donor-side limitation shows an exponential increase.

In general, the major differences between observed drought stress levels were apparent in the initial part of the light response for the assessed parameters. At low-to-moderate light intensities, the PSII quantum yield (ΦPSII) continuously decreased with increasing light. In contrast, the PSI quantum yield (ΦPSI) temporarily stopped decreasing, leading to a significant increase in the ΦPSI to ΦPSII ratio at low light intensities (Fig. 5).
Fig. 5

Light responses of a nonphotochemical quenching in PSII (NPQ) and b the ratio of the effective quantum yield of PSI and PSII (ΦPSI/ΦPSII). The small plots within each graph show the initial phase of each light curve. The rapid light curves were performed after previous induction at moderate light; duration of each interval with a given light intensity was 30 s (see "Materials and methods" for details). The average values ± standard errors from 8 to 10 plants are presented. NS non-stressed samples, ID initial drought, MD moderate drought, SD severe drought

This phenomenon can be seen in all plants. However, in drought-stressed plants, it appears at lower light intensities (e.g., at 100 μmol photons m−2 s−1 in SD leaves, while only at 220 μmol photons m−2 s−1 in NS leaves). The initial trends of light dependencies of ΦPSI to ΦPSII ratio are very similar to those observed for NPQ (Fig. 5). As the increase of ΦPSI to ΦPSII ratio can be considered to be a symptom of enhancement of cyclic electron transport, this suggests that cyclic flow has a role in nonphotochemical energy dissipation (Heber and Walker 1992; Munekage et al. 2002; Livingston et al. 2010). To estimate correctly the rate of cyclic electron flow using PSII and PSI quantum yields, the values need to be corrected by the coefficient of absorbed energy distribution between PSI and PSII (Huang et al. 2010). Therefore, the decrease of ΦPSI to ΦPSII ratio that we observed does not necessarily mean that the rate of cyclic electron flow decreases, as the uncorrected ratio itself does not always represent the true estimate of cyclic electron flow, especially in high light intensities (see discussion below).

Additional information about photoprotective responses can be obtained from dark recovery kinetics after reaching the steady state (Supplementary Fig. 1). Here, the trends of parameters were similar to previous observations; however, some surprising contradictory trends were also observed. Specifically, an increase in NPQ associated with a decrease in the ΦPSI to ΦPSII ratio (small graph in Supplementary Fig. 1a) as well as with a slowly decreasing PSI donor-side limitation (small graph in Supplementary Fig. 1b). Using the dark relaxation kinetics, we estimated the energy-dependent (pH-dependent) fraction of nonphotochemical quenching (qE). We found that qE values increased with severity of drought stress. However, the differences were not fully proportional to the differences in the NPQ values; this indicates that there can be also differences in slow-relaxing components of nonphotochemical quenching, which also contribute to values of parameter NPQ.

Quantification of PSII electron transport to alternative electron sinks

To estimate the fraction of linear electron transport that is not used for carboxylation or oxygenation of RuBP, the difference between electron transport calculated using the quantum yields of PSII photochemistry from chlorophyll fluorescence measurements were compared, and the electron transport calculated from gas exchange measurements, both measured simultaneously (data presented in Fig. 2). Hence, the area between the curves represents the measure of electrons released by PSII that must be utilized by alternative electron sinks (Fig. 6). In this figure, we present the data assuming equal distribution of absorbed light between PSII and PSI.
Fig. 6

The values of linear electron transport rate calculated from PSII effective quantum yields (JF = JPSII) calculated assuming equal distribution of electrons between PSI and PSII and the electron transport rate used either for carboxylation or oxygenation of RuBP calculated using data obtained by gas exchange measurements (JG). Both JF and JG were calculated using records of simultaneous measurements of gas exchange and chlorophyll fluorescence within photosynthetic induction curve (see "Materials and methods" for details). The difference J− JG represents a common way to estimate additional electron flux in other pathways besides photosynthesis and photorespiration; hence, the color fill between JF and JG illustrates the total sum of electrons used in alternative electron pathways. The average values from 8 to 10 plants are presented. NS non-stressed samples, ID initial drought, MD moderate drought, SD severe drought

The data indicate that in the NS plants outside of photochemistry associated with photosynthesis, there is little additional electron sinks, while with increasing drought stress, there is significant increase in additional electron sinks.


Stomatal and nonstomatal effects at different level of drought stress

A decrease in leaf water content due to drought stress mainly affects photosynthesis through stomatal closure, which causes a shortage in CO2 supply leading to a decrease in linear electron transport (Lawlor and Cornic 2002). Our results indicate that stomatal responses vary in degree, becoming more pronounced with increasing severity of stress. The effect of CO2 shortage on linear electron flow was partly ameliorated by an increase in photorespiration rate. Photorespiration may itself be part of the dissipatory mechanisms adopted by plants to mitigate oxidative damage resulting from insufficient utilization of electrons (Kozaki and Takeba 1996). However, there are also experimental results showing a negligible protective effect of photorespiration in conditions of drought (Brestic et al. 1995).

In addition to RuBP, there are several alternative acceptors of electrons from linear electron transport, such as oxygen in the Mehler reaction (Asada 1999) or nitrite (Eichelmann et al. 2011). Nevertheless, the most efficient way to prevent oxidative damage is through the precise control of electron transport rate associated with well-regulated dissipation of excessive light energy in the process of nonphotochemical quenching (reviewed in Müller et al. 2001).

We observed direct correlations between leaf water status and CO2 assimilation, PSII electron transport rate, photorespiration rate, and nonphotochemical quenching (Fig. 3). However, all these parameters were correlated with stomatal conductance, which decreased with the increasing stress, resulting in a decrease in ci until severe water stress. This finding supports the view that the dominant limitation of photosynthesis is through stomatal effects (Cornic and Massacci 1996).

Particularly interesting is the relationship between the steady-state values of NPQ and stomatal conductance. NPQ appears to remain constant until some threshold value of gs is reached, after which there is a large increase in NPQ. It should be noted that the measurements were done at high light intensity which would induce maximum capacity for NPQ. However, NPQ gradually increased with decreasing RWC, suggesting it is more associated with leaf water status.

The threshold level of low gs which induced a sharp rise in NPQ was ~0.12 mmol m−2 s−1 when CO2 assimilation rate was reduced to about 50 % of its NS value (~12 μmol CO2 m−2 s−1); leaf RWC was below 70 % and the leaf water potential was below −1.5 MPa. Such values of RWC and ΨW are considered to be threshold levels at which the major nonstomatal limitation of photosynthesis occurs (Lawlor and Cornic 2002). The most frequently reported processes contributing to a decrease of photosynthesis not caused by stomata are impaired ATP synthesis (Tezara et al. 1999; Cruz et al. 2005), decrease in Rubisco activity (Holaday et al. 1992; Wingler et al. 1999), RuBP availability (Flexas et al. 2004), and impairment of photochemistry (Björkman and Powles 1984). The typical symptom of nonstomatal limitation is the occurrence of a ci inflection point. This is clearly visible in our measurements (Fig. 3). Lal et al. (1996) attributed the occurrence of ci inflection points either to a decrease of mesophyll conductance or to stomatal patchiness. The interaction between stomatal conductance and leaf water status (which we documented by divergence of ci values at low gs) has not been commonly described in research articles (Flexas and Medrano 2002). Nevertheless, it supports the idea of the direct effect of low water potential on metabolic processes, independent of stomatal closure (Ortiz-Lopez et al. 1991).

Alternative ways of PSII protection in drought stress

Chloroplasts have evolved significant flexibility in mechanisms to cope with changes in demands for energy (Kramer and Evans 2011), which are activated under specific environmental and metabolic challenges. Their role is to prevent the production of deleterious reactive oxygen species and to optimize the ATP to NADPH ratio to cover the demand of metabolic processes, especially CO2 carboxylation (Kramer and Evans 2011). The chloroplast must balance the production and consumption of ATP and NADPH by augmenting production of intermediates or preventing the accumulation of excess intermediates. Mismatches in regulation will limit photosynthesis (Cruz et al. 2005; Kramer and Evans 2011). The excitation of photosystems must be well regulated, and this is achieved by several mechanisms.

During drought stress and relatively high light conditions, substantial structural rearrangements for light protection are anticipated to be a major mechanism affecting photosystem stoichiometry (Bailey et al. 2001). Degradation of D1/D2 proteins of PSII has been shown to increase under water stress (Yuan et al. 2005). Moreover, Eichelmann and Laisk (2000) observed that stress conditions tend to decrease light harvesting antenna of PSII more than PSI. Such changes must be considered when converting photochemical efficiency of both photosystems estimated from chlorophyll fluorescence or P700 absorbance records into a rate of electron transport (Kramer and Evans 2011). Measurements of the quantum yield of PSII are frequently utilized to predict rates of linear electron transport. However, in most cases an equal distribution of light between photosystems is assumed (Genty et al. 1989; Krall and Edwards 1992). The PSI and PSII electron transport rates calculated assuming equal distribution of the absorbed light between PSI/PSII are shown in Fig. 7a, c. The difference between electron transport rates of PSI and PSII (Fig. 7e) represents an estimate of the electron transport rate of cyclic electron flow (Yamori et al. 2011). Hence, calculation assuming an equal distribution between photosystems would suggest that the highest cyclic electron transport will be in NS plants and that there will be a decrease or complete cessation of cyclic electron flow in drought-stressed plants at high light levels.
Fig. 7

Calculated values of electron transport rate through PSII and PSI as well as the difference between electron transport rates in wheat subjected to different level of drought stress. The values in graphs left (a, c, e) were calculated assuming equal distribution of absorbed energy between PSII and PSI. Graphs right (b, d, f) represent hypothetical values of electron fluxes calculated assuming unequal distribution, where the ratios between PSII and PSI absorbance was recalculated to obtain stable difference (JPSI − JPSII) at high light intensities. The average values of PSII and PSI quantum yields were used for calculation, and the ratios of total PAR absorbed by PSII (αII) and PSI (αI) are presented in legend

Low or moderate drought stress effects on PSII effective quantum yields seemed to be insignificant, supporting the existence of an intact PSII electron transport in initial phases of drought stress (Brestic et al. 1995; Lawlor and Cornic 2002). The obtained results also support the claim of an essential role of cyclic electron transport for ATP formation (Heber and Walker 1992), which is needed more in plants running the carboxylation (NS) than in plants with closed stomata (discussed in Laisk et al. 2010). However, these results contradict reports showing high cyclic electron flow during drought stress (Golding and Johnson 2003; Huang et al. 2012) and an increase of cyclic electron flow at light intensities exceeding saturation level (Clarke and Johnson 2001; Brestic et al. 2008; Huang et al. 2012).

A possible explanation of this contradiction is that an incorrect energy distribution factor was used in calculation, as we used a default value of 0.5, i.e., equal distribution of absorbed energy between PSII and PSI. Therefore, rather than using the default value of 0.5 assuming equal distribution, we adjusted the partitioning of absorbed energy between the photosystems with increasing drought to determine if the results would be more consistent with data and conclusions which could be drawn. We hypothesized that the rate of cyclic electron flow increases with increasing light and reaches a maximum, constant rate. We then recalculated the distribution factors for each treatment and the rates of electron transport (Fig. 7b, d, f). By decreasing the fraction of absorbance by PSII with increasing drought, the differences between PSII electron transport rates between the NS and drought stress treatments became greater. In contrast, the PSI electron transport appeared to be less affected by drought, resulting in the highest predicted cyclic electron flow in severely stressed plants (SD).

Besides the fact that such results on cyclic electron flow are in concordance with many reports (Golding and Johnson 2003; Rumeau et al. 2007; Lehtimäki et al. 2010; Huang et al. 2012; etc.), this change in partitioning between photosystems is also supported by the fact that the new estimated values of PSII electron transport rate (JPSII) are much closer to the values of JG calculated using the gas exchange measurements (Supplementary Fig. 2); this issue will be discussed later.

Therefore, we suggest that adjustment of light harvesting (what we see as an alteration of PSII/PSI energy distribution factor) can be considered as an important protective response, as illustrated by a decrease in the total pool of electrons, which are in excess of the needs of carboxylation or oxygenation (Supplementary Fig. 2). This can avoid the overreduction of PQ pool and, thus, it can prevent the PSII damage.

The logical question is whether the proposed altered distribution of energy between photosystems is due to changes in protein complexes (i.e., state-transitions or long-term responses, etc.) or it is the spillover of energy. It is broadly accepted that the state transitions are associated with low light conditions (Dietzel et al. 2008). However, there are several recent studies suggesting that the activity of protein kinases in leaves (mainly STN7) can function even in high light conditions which could result in disconnection of LHCII from PSII, providing thus the protection against high light (Fristedt and Vener 2011). Specifically, in monocots (such as wheat) exposed to drought or salt stress, it has been observed that the CP29 protein of LHCII is phosphorylated, with its subsequent lateral migration from grana stacks to stroma lamellae; in addition, the LHCII can be disconnected (Liu et al. 2009; Chen et al. 2013). Therefore, kinases may contribute to alteration of PSI/PSII stoichiometry in high light conditions.

Another possible explanation for the observed decrease of JPSIIJPSI with increasing water stress is that there are some alternative auxiliary electron sinks on PSII acceptor side. In this respect, Ivanov et al. (2012), in high light and low-temperature conditions, documented the important role of the plastid terminal oxidase (PTOX) as an alternative O2-dependent electron sink that can accept electrons directly from PSII (Cournac et al. 2000). The PTOX is involved also in chloroplast redox signaling pathways (Aluru and Rodermel 2004; McDonald et al. 2011) and in chlororespiration (Rumeau et al. 2007). The increase of PTOX was observed in conditions of stress, e.g., salinity (Stepien and Johnson 2009), high temperature (Diaz et al. 2007) or low temperature (Savitch et al. 2010; Ivanov et al. 2012). PTOX activity avoids the overreduction of plastoquinone pool by its oxidation (Joët et al. 2002; Josse et al. 2003). Moreover, Ivanov et al. (2012) also suggest the involvement of PTOX in modulation of cyclic electron flow and regulation of cyclic electron pathway in stress conditions. The activity of PTOX (or other auxiliary electron acceptor) as an electron sink as well as its regulatory function can well explain some of our observations; however, to explain fully the faster increase in JPSII compared to JPSI with increasing light (Fig. 7a, c, e), the rate of electron uptake in high light must be rather high (30 μmol e m−2 s−1 or more). Our measurements, indeed, indicated in drought-stressed leaves such a high electron transport that needs to be utilized by alternative electron sinks (Fig. 6). On other hand, some studies on the function of PTOX as PSII electron acceptor challenges such high rates (discussed well in Stepien and Johnson 2009). However, we can not exclude that this, or some similar mechanism may be contributing as an important alternative sink in wheat when water supply is limiting.

There are also other alternative sinks that can accept the electrons released by PSII. But it was shown that in C3 leaves the water–water cycle contributed <5 % to linear electron flow, even when CO2 fixation was inhibited (Ruuska et al. 2000; Clarke and Johnson 2001). Similarly, the published data on the contribution of nitrite reduction in high light (Eichelmann et al. 2011) does not explain the high alternative electron flow, especially in drought stress conditions with low nitrogen uptake.

Even with a decrease in the fraction of energy absorbed by PSII (Fig. 7b, d, f), there is still an excess pool of electron that can be potentially harmful and need to be utilized by alternative electron sinks. Therefore we suggest that both adjustment of light harvesting and alternative electron flows (including auxiliary sinks) contribute to protection of PSII against photoinhibition.

In addition to mechanistic issues, our analyses may have practical and technical implications. The commonly used approach of calculating electron transport rate by assuming equal distribution between PSI and PSII may overestimate rate of linear electron flow under drought stress conditions. This could result in overestimation of electron flow to alternative electron acceptors and underestimation of cyclic electron flow. Furthermore, the PSII electron transport is used in some photosynthetic models, e.g., for estimation of mesophyll conductance. The potential errors due to inaccurate estimates of PSII electron transport has been highlighted in several technical or methodical articles (e.g., Yin et al. 2009). Our analyses show how the use of a default value assuming equal distribution of energy absorption between the photosystems, versus a reduction in energy absorption by PSII with increasing drought can affect rates of photochemistry and the potential contribution of cyclic photophosphorylation. Likewise, activity of auxiliary electron acceptors directly from PSII, such as PTOX could affect the relative ratio of cyclic and linear electron flow.

Activity of the cyclic electron flow in drought stress

PSI cyclic electron flow has been proposed to function under drought stress, which could function to induce qE by increased proton motive force and NPQ (Figs. 4, 5, Supplementary Fig. 1). Both the down-regulation of linear electron flow and up-regulation of cyclic electron flow avoid the excessive reduction of PSI electron acceptors. Such a reduction can result in the production of reactive species and therefore photo-oxidation of protein components (Tikkanen et al. 2006). Our results indicate that even at PAR up to levels equivalent to full sunlight (2,000 μmol photons m−2 s−1; ten times higher than that used during growth of the plants) the acceptor side was not overreduced—as indicated by parameter ΦNA (Fig. 4f). In such extreme conditions, it seems improbable that a low acceptor-side reduction can be maintained without high cyclic electron flow in drought-stressed leaves, as shown in Fig. 7b, d, f.

The high ΦPSI/ΦPSII ratio which occurs under lower light, particularly under SD, is consistent with function of cyclic electron flow (Fig. 4f). With increasing light, as soon as this ratio started to increase (indicating an increase of cyclic electron flow), the ΦNA decreased. This result is in agreement with the recent identification of the direct role of cyclic electron flow in photoprotection (Laisk et al. 2010), as it ameliorates the excess of electrons on PSI acceptor side. In addition to a decrease of ΦNA, the initiation of cyclic electron flow (suggested by an increase in ΦPSI/ΦPSII ratio) was also associated with (i) an increase of nonphotochemical quenching (NPQ); (ii) the accumulation of oxidized P700+ (indicated by the parameter ΦND); and (iii) the increase of the donor-side limitation as a result of down-regulation of electron transport between PSI and PSII. This phenomenon is also related to cyclic electron flow building the proton motive force, resulting in lumen acidification and subsequent slowing of electron transfer at the cyt b6f complex (Kramer and Evans 2011). The drought-induced decrease of electron transport rate between PSII and PSI was found also in dark-adapted samples (Goltsev et al. 2012). Although the increase in donor-side limitation was observed in both control and drought-stressed samples, the relationship of PSII electron transport rate with P700+/P700tot (Fig. 8a) indicates that the demand for electrons at PSI (redox poise of PSI) resulted in reduced electron transport from PSII in drought-stressed plants. The trend of QA/QAtot versus JPSII gave similar results (not shown here). Hence, the restriction on donation of electrons from PSII to PSI was increasing with increasing stress. The rate of cyclic electron flow calculated as a difference between JPSI and JPSII (from the adjusted partitioning of absorbance between photosystems, Fig. 8b) linearly correlated with the accumulation of oxidized P700+ until the maximum rate of cyclic electron flow was reached.
Fig. 8

a Relationship between the fraction of overall P700 that is oxidized in a given state (P700+/P700T, ΦND) and the electron transport rate through PSII (JPSII). b Relationships between the fraction of overall P700 that is oxidized in a given state (P700+/P700T, ΦND) and the rate of cyclic electron transport around PSI (estimated as JPSI − JPSII). The parameters were calculated from the data obtained by simultaneous measurements of chlorophyll fluorescence and P700 in light curve records (see "Materials and methods"); the data with unequal absorbance ratios (αI, αII) are used as shown in Fig. 7. The average values ± standard errors from 8 to 10 plants are presented. NS non-stressed samples, ID initial drought, MD moderate drought, SD severe drought

The initial slope of the correlation was similar in all groups; this close relationship also supports the role of cyclic electron transport in PSII protection through a buildup of ΔpH in thylakoids. This process enhances protonation of the lumen thereby limiting electron transport and triggering NPQ (Rumeau et al. 2007). By regulating electron transport into PSI via the cytb6f complex, cyclic electron flow also minimizes the probability of the formation of reactive oxygen species (ROS) such as superoxide on the acceptor side of PSI (Foyer and Noctor 2009) and induces the expression of stress response genes (Tikkanen et al. 2006).

The PSII electron transport rate at high light intensities surprisingly increases more quickly than donor-side limitation (Fig. 8a). We hypothesize that this is not a true increase, but this may be a result of further changes of PSI/PSII energy distribution due to disconnection of PSII antennae in high light or other processes leading to overestimation of PSII electron transport (and hence underestimation of the cyclic electron transport rate) at high light intensities.

In conclusion, the results of this study indicate that photosynthesis in wheat is limited mainly by stomatal effects in conditions of low-to-moderate drought stress. Enhancement of the nonstomatal effects in severely stressed plants seemed to be induced by low leaf water potential rather than the very low stomatal conductance. Drought stress gradually decreased the PSII electron transport, but the capacity for nonphotochemical quenching increased only in severely stressed plants where photosynthetic rate was decreased by half or more. Our results indicate that in drought-stressed leaves the electron balance can be partially equalized through alteration of energy distribution between PSII and PSI; in this case, the calculated PSII electron transport can be overestimated, and the real pool of nonutilized electrons would decrease compared with the model assuming the same energy distribution. Thus, in future studies, it will be important to utilize methods to measure the distribution of the absorbed energy between the photosystems using the methods of Laisk and Loreto (1996), Cardol et al. (2008), or other reliable method. It is evident that in drought-stressed plants, there is a relatively large portion of electrons that need to be utilized by alternative electron sinks. In addition to water–water cycle and nitrite reduction, we suggest also activity of auxiliary electron sinks on acceptor side of PSII (such as PTOX), which was previously shown to play a role as a safety valve protecting the photosynthetic apparatus. This would prevent overreduction of the PQ pool and protect the PSII against oxidative damage. In parallel, we propose that, with drought stress, the fraction of the absorbed light by PSII is decreased, with an increase in the ratio of the yields of PSI/PSII, which would coincide with an increase of nonphotochemical quenching, an increase in the donor-side limitation and a decrease in the acceptor-side limitation of PSI. In summary, our results demonstrate that in drought stress conditions, there is a complex of mutually interconnected and precisely regulated photoprotective responses.


The authors thank Dr Richard J. Ladle (School of Geography and the Environment, Oxford University, UK, and Institute of Biological and Health Sciences, Federal University of Alagoas, Praça Afrânio Jorge, s/n, Prado, Maceió, AL, Brazil) for reviewing and improving the English of the manuscript. The research described here has been supported by grant APVV-0197-10 and APVV-0661-10. This study was also supported by grants from the Russian Foundation for Basic Research and Molecular and Cell Biology Programs of the Russian Academy of Sciences to SIA.

Supplementary material

11120_2013_9885_MOESM1_ESM.pdf (208 kb)
Supplementary material 1 (PDF 208 kb)

Copyright information

© Springer Science+Business Media Dordrecht 2013