Abstract

Background

Water deficit (WD) decreases photosynthetic rate (A) via decreased stomatal conductance to CO2 (gs) and photosynthetic metabolic potential (Apot). The relative importance of gs and Apot, and how they are affected by WD, are reviewed with respect to light intensity and to experimental approaches.

Scope and Conclusions

With progressive WD, A decreases as gs falls. Under low light during growth and WD, A is stimulated by elevated CO2, showing that metabolism (Apot) is not impaired, but at high light A is not stimulated, showing inhibition. At a given intercellular CO2 concentration (Ci) A decreases, showing impaired metabolism (Apot). The Ci and probably chloroplast CO2 concentration (Cc), decreases and then increases, together with the equilibrium CO2 concentration, with greater WD. Estimation of Cc and internal (mesophyll) conductance (gi) is considered uncertain. Photosystem activity is unaffected until very severe WD, maintaining electron (e) transport (ET) and reductant content. Low A, together with photorespiration (PR), which is maintained or decreased, provides a smaller sink for e, causing over-energization of energy transduction. Despite increased non-photochemical quenching (NPQ), excess energy and e result in generation of reactive oxygen species (ROS). Evidence is considered that ROS damages ATP synthase so that ATP content decreases progressively with WD. Decreased ATP limits RuBP production by the Calvin cycle and thus Apot. Rubisco activity is unlikely to determine Apot. Sucrose synthesis is limited by lack of substrate and impaired enzyme regulation. With WD, PR decreases relative to light respiration (RL), and mitochondria consume reductant and synthesise ATP. With progressing WD at low A, RL increases Ci and Cc. This review emphasises the effects of light intensity, considers techniques, and develops a qualitative model of photosynthetic metabolism under WD that explains many observations: testable hypotheses are suggested.

INTRODUCTION

This review aims to advance understanding of the effects of relatively rapidly developing water deficit (WD) in plants on photosynthetic rate (A). It complements reviews by Chaves and Oliviera (2004) and Chaves et al. (2009), which provide a wider perspective. Considerable effort (see Boyer, 1990; Kramer and Boyer, 1995; Lawlor and Cornic, 2002) has already been devoted to analysing the effects of WD on A via stomatal conductance (gs) and photosynthetic potential (Apot) in different experimental systems, but there is lack of consensus over their relative importance. For example, Tezara et al. (1999) and Medrano et al. (2002) have concluded that gs and metabolism (RuBP and ATP supply) limit A at low WD, whereas Flexas et al. (2004a) and Chaves and Oliviera (2004) have concluded that gs decreases A and that metabolic limitations are unimportant, or only at severe WD.

Reductionist science might seem to require a single factor determining the effects of WD on photosynthetic rate and processes; however, finding such an elusive deus ex machina in very complex, partial and imperfect data requires some faith (or dogmatism). Identifying ‘a cause’ of decreased A, and Apot, under WD under all environmental conditions may not be possible because the concept is incorrect. Photosynthetic systems are varied, structurally and functionally dynamic, and dependent on the environment, which is also extremely dynamic (Lawlor, 2001). Control of A and Apot is distributed between many metabolic components and processes that vary in importance as conditions – environmental and within the plant – change (von Caemmerer, 2000). Probably control varies, depending on the plant and on environmental conditions during growth and under water deficit. Hence different experiments produce different answers. Understanding effects of WD on photosynthesis will come from better quantification of the interactions between WD (and other environmental conditions) and photosynthetic mechanisms.

Photosynthesis and water deficit: why the emphasis?

Photosynthetic CO2 assimilation per unit area and time, i.e. rate (A), is inhibited by rapidly developing WD in physiological studies, so it is assumed to be responsible for decreased dry matter production. However, leaf area is also very important for total production. This applies in the field, where slowly developing WD results in a small leaf area index, which often dominates production with only small effects on A (Legg et al., 1979; Sinclair and Purcell, 2005). However, the perceived need to apply understanding of photosynthesis to alleviation of practical problems such as loss of crop yield due to WD has increased interest in ‘water stress physiology’. Many basic questions remain about how cellular processes are regulated by WD. As A dominates cell metabolism, with very large fluxes of carbon, nitrogen and energy (Lawlor, 2001), it is potentially vulnerable to WD (Kramer and Boyer, 1995). It is integrated with respiration and aspects of electron transport (ET) and ATP synthesis in the mitochondria (Atkin and Macherel, 2009), and changes in A and energy are related, in ways still unclear, to accumulation of ‘stress metabolites’ (e.g. proline), gene expression and protein synthesis. It is now appreciated (Herbert, 2002; Scheibe et al., 2005; Rumeau et al., 2007) how tightly integrated photosynthetic metabolism is, and how difficult it will be, without understanding of the system and a clear model, to engineer plants for large biomass and yield production under WD (Sinclair and Purcell, 2005; Bohnert et al., 2006). It is also extremely doubtful if one model of photosynthesis and metabolism will suit all plant × environment combinations (Reynolds et al., 2005). Differences between plants grown in controlled environments and those in the field must be considered, especially when considering the potential for genetic modification (Sinclair and Purcell, 2005). Quantitative assessment of conditions in photosynthetic cells under WD is essential if the current flood of information from genomics, proteomics and metabolomics is to be used to improve plant production under WD (Flexas et al., 2004b; Chaves et al., 2009). First, general agreement on the qualitative processes involved is required, from which quantitative species × environment models may emerge.

Is there a standard experimental system?

Before discussing WD and its effects on A and Apot, it is essential to assess the conditions of experiments. Standardization is minimal: studies are very disparate in terms of species, environment and mode of applying stress and hence comparison of data is difficult (Table 1). Few studies have been done in variable field conditions (e.g. Legg et al., 1979; Wise et al., 1991, 1992; Escalona et al., 1999; Tezara et al., 2003; Ripley et al., 2007), some with plants grown in pots (Quick et al., 1992). Controlled environments are not necessarily qualitatively and quantitatively more uniform for plants grown in small pots. Disparities between experiments and selective emphasis on particular techniques (Table 1) generate very different data from which conflicting conclusions have been drawn, fuelling much competition between concepts (Lawlor, 2002; Flexas et al., 2004a).

Table 1.

Analysis of experiments from which data have been used to assess the effects of water deficit on photosynthetic metabolism. Particular attention should be paid to the conditions during growth (col. 3) and application of water deficit (col. 4). Note also that the types of measurements (col. 5) are related to the data derived and thus the interpretation, for example of sub-stomatal CO2 concentration (Ci). Chloroplastic CO2 concentration (Cc) and the ratio of Ci to atmospheric CO2 concentration (Ca) are indicated in columns 6–8. Col. 9 assesses the effects of removal of the epidermis and/or elevated CO2 on photosynthesis. Col. 10 shows the A/Ci response with ↓ indicating decreased slope and plateau. Col. 11 comments on the role of stomata in controlling photosynthesis under water deficit. Col. 12 indicates incorporation of alternative sinks into calculation of Cc by estimating electron flux. Col. 13 contains a general evaluation of experimental data

    Measurements taken?
 
      
Reference Species Growth conditions / PAR (μmol m−2 s−1) / Duration (h d−1Water deficit conditions / PAR / Duration CO2 / O2 / Fluorescence Ci Cc Ci/Ca CO2 / Epidermis removal stimulates AA/Ci gs controls AMetabolism inhibited? PR / Alt. sinks? Comments 
Lawlor & Fock, 1975 Helianthus CE / 400 / 16 PEG / hours Y / N / N      Y, mild WD Y, moderate/severe WD  PR decreasing, RD increasing 
Kaiser 1981, 1984 Spinacia Unkown Osmotic / 200 / mins Y / Y / N Leaf slices – Y / Y  Y, to severe WD Y, severe WD Photophosphorylation?  Conditions artificial Primary reactions insensitive. Calvin cycle decreased severe WD 
Dietz & Heber, 1983 Primula palinari Greenhouse / low ??? or 200? / ? Leaf pieces / 0 or 200 / fast Y / N / N Leaf slices – Y / Y  Y, to severe WD Y, at severe WD  Low light, rapid WD, effect on CO2 assimilation not light reactions or RD 
Sharp & Boyer, 1986 Helianthus CE / 900 /14 Soil / 4 d 40 or 2000 / 6 hr Y / N / N Attached leaf  Decrease/increase  ↓ Y, mild but not severe WD Y, moderate WD  Metabolic inhibition. No photoinhibition 
Cornic et al., 1987 Phaseolus CE / 310 / 16 Soil / 4–8 d Y / N / Y Attached leaf Decrease?/increase   Y, mild WD Y, moderate WD  Metabolic inhibition, Ci/Ca rises large WD 
Kirschbaum, 1987 Eucalyptus Greenhouse, natural Canberra / high? / 12 Soil Y / N / N Attached leaf Decrease/increase ↓ Y, mild WD Y, progressive  Decreased A/Ci, no major PI 
Cornic et al., 1989 Phaseolus CE / 200 / 16 Soil / 200? 15 d Y / Y / Y Attached leaf Y / Y / Y   Y, to severe WD Y, very severe WD  Ci misleading Random leaves measured over period 
 Elatostema CE / 40 / 16 Soil / 25 d Attached leaf      
Renou et al., 1990 Triticum CE / 600 / 14 PEG / hours Y / Y / N Y? ↓ to comp. pt Y / –  N, moderate WD Y, very severe WD Y/N Ci errors? Cc low assumes only PR 
Cornic & Briantais, 1991 Phaseolus Greenhouse / 350? / ? Soil / 8 d Y / N / Y Attached leaf Y ↓ to comp. pt Y ↓ to comp. pt Constant/Increase?   Y, to severe WD N, moderate WD Y, severe WD Y/Y? Fluorescence calculation gives low Cc but not measured 
Giménez et al., 1992 Helianthus Greenhouse + CE / 450 / 16 Soil / 450 / 4 d Y / N / N Attached leaf Decrease/increase N/- ↓ Y, mild WD Y, moderate WD  RuBP limiting A/Ci 
Martin & Ruiz-Torres, 1992 Triticum CE / 500 / 16 Soil / 500 / ?days Y / Y / N Attached leaf –  ↓ Y, mild WD Y, moderate WD  Metabolic limit Calvin cycle 
Lauer & Boyer, 1992 Helianthus Glycine, Phaseolus CE / 600–800 / 14 Soil / 12 d Y / N / N Attached leaf Small decrease/ then increase N/-  Y, mild WD Y, mild WD  Direct measure Ci, decrease/increase 
Quick et al., 1992 Lupinus, Helianthus, Vitus, Eucalyptus Field Portugal, natural / high? Soil pots / up to 6 d Y / N / N Large decrease Y/-  Y, mild to moderate WD N, mild to moderate WD  Stomatal control only Lupin, Helianthus, Eucalyptus. Metabolic Vitus, large sucrose 
Tourneux & Peltier, 1995 Solanum CE / 600 / 12 Much water Discs /dark 0 / hours ? N / Y / Y Discs Y ↓ to comp. pt   N, or very low WD Y/N Extreme conditions assumes PR only 
Escalona et al., 1999 Vitus Field / large/ 14 Drought vs.irrigated / days Y / N / N Attached leaf Decrease  ↓ Y, mild WD Y, moderate/severe WD  RuBP limiting A/Ci 
Tezara et al., 1999, 2008 Helianthus Greenhouse–CE / 500–700/ 14 Soil / 400 / 8 d Y / Y / Y Attached leaf Decrease/increase N / – ↓ Y, mild WD Y, moderate/severe WD Y/N ATP limiting RuBP and A/Ci 
Wingler et al., 1999 Hordeum CE / 460 / 12 Soil / 12 d Y / N / Y –   ?? Y, moderate WD  Ci measured not given. PR indirect evidence not increased 
Tang et al., 2002 Helianthus Glasshouse, large / ? / 14 Soil / several days N / Y / N – N / N  Y, mild WD Y, mild WD  Metabolic limit 
  CE / 700–900/ 12  Leaf pieces          
Haupt-Herting & Fock, 2002 Lycopersicum CE / 200 / 16 Soil/ 200 / 8 d Y / Y / Y Attached leaf – N / –  Y, mild WD Y, moderate WD Y/Y PR increased relative but not absolute sink for e with WD 
Botha et al., 2004 Rhamnus, Nicotiana, Vitus Greenhouse / 600 / 14? Soil / 600 / variable Y / N / Y Attached leaf Variable  ↓ Y, mild WD Y? severe WD, Rubisco limit  Combined data, large errors 
Ennahli & Earl, 2005 Gossypium Glasshouse / winter / 10 Soil / ? / several days Y / N / Y Attached leaf Y ↓ to comp. pt Decrease/increase ↓ Y, mild WD Y, progressive Y/N Calculation of Cc based on fluorescence assuming only PR. Conflicts in data 
Flexas et al., 2006 Glycine CE / 800–1000 / 14 Soil /? Y / N / Y Attached leaf Y ↓ to comp. pt –   Y, to severe WD Y, severe WD, Rubisco Y/N Calc Cc based on fluorescence assuming only PR 
 Nicotiana   Attached leaf          
Zhou et al., 2007 Oryza CE / 400 / 500 / 12 PEG / 500 / 2d Y / N / N Attached leaf Decrease/increase Y mild ↓ Y, mild WD Y, mild WD  ROS increase mild stress 
Ripley et al., 2007 Alloteropsis Field–polytunnel / large / 400 + / 12 Soil / 400 / several days Y / N / Y Attached leaf Decreased (mild WD) N / – ↓ Y, mild WD Y, mild/moderate WD,  Limited WD, high light, RuBP/Rubisco limiting 
    Measurements taken?
 
      
Reference Species Growth conditions / PAR (μmol m−2 s−1) / Duration (h d−1Water deficit conditions / PAR / Duration CO2 / O2 / Fluorescence Ci Cc Ci/Ca CO2 / Epidermis removal stimulates AA/Ci gs controls AMetabolism inhibited? PR / Alt. sinks? Comments 
Lawlor & Fock, 1975 Helianthus CE / 400 / 16 PEG / hours Y / N / N      Y, mild WD Y, moderate/severe WD  PR decreasing, RD increasing 
Kaiser 1981, 1984 Spinacia Unkown Osmotic / 200 / mins Y / Y / N Leaf slices – Y / Y  Y, to severe WD Y, severe WD Photophosphorylation?  Conditions artificial Primary reactions insensitive. Calvin cycle decreased severe WD 
Dietz & Heber, 1983 Primula palinari Greenhouse / low ??? or 200? / ? Leaf pieces / 0 or 200 / fast Y / N / N Leaf slices – Y / Y  Y, to severe WD Y, at severe WD  Low light, rapid WD, effect on CO2 assimilation not light reactions or RD 
Sharp & Boyer, 1986 Helianthus CE / 900 /14 Soil / 4 d 40 or 2000 / 6 hr Y / N / N Attached leaf  Decrease/increase  ↓ Y, mild but not severe WD Y, moderate WD  Metabolic inhibition. No photoinhibition 
Cornic et al., 1987 Phaseolus CE / 310 / 16 Soil / 4–8 d Y / N / Y Attached leaf Decrease?/increase   Y, mild WD Y, moderate WD  Metabolic inhibition, Ci/Ca rises large WD 
Kirschbaum, 1987 Eucalyptus Greenhouse, natural Canberra / high? / 12 Soil Y / N / N Attached leaf Decrease/increase ↓ Y, mild WD Y, progressive  Decreased A/Ci, no major PI 
Cornic et al., 1989 Phaseolus CE / 200 / 16 Soil / 200? 15 d Y / Y / Y Attached leaf Y / Y / Y   Y, to severe WD Y, very severe WD  Ci misleading Random leaves measured over period 
 Elatostema CE / 40 / 16 Soil / 25 d Attached leaf      
Renou et al., 1990 Triticum CE / 600 / 14 PEG / hours Y / Y / N Y? ↓ to comp. pt Y / –  N, moderate WD Y, very severe WD Y/N Ci errors? Cc low assumes only PR 
Cornic & Briantais, 1991 Phaseolus Greenhouse / 350? / ? Soil / 8 d Y / N / Y Attached leaf Y ↓ to comp. pt Y ↓ to comp. pt Constant/Increase?   Y, to severe WD N, moderate WD Y, severe WD Y/Y? Fluorescence calculation gives low Cc but not measured 
Giménez et al., 1992 Helianthus Greenhouse + CE / 450 / 16 Soil / 450 / 4 d Y / N / N Attached leaf Decrease/increase N/- ↓ Y, mild WD Y, moderate WD  RuBP limiting A/Ci 
Martin & Ruiz-Torres, 1992 Triticum CE / 500 / 16 Soil / 500 / ?days Y / Y / N Attached leaf –  ↓ Y, mild WD Y, moderate WD  Metabolic limit Calvin cycle 
Lauer & Boyer, 1992 Helianthus Glycine, Phaseolus CE / 600–800 / 14 Soil / 12 d Y / N / N Attached leaf Small decrease/ then increase N/-  Y, mild WD Y, mild WD  Direct measure Ci, decrease/increase 
Quick et al., 1992 Lupinus, Helianthus, Vitus, Eucalyptus Field Portugal, natural / high? Soil pots / up to 6 d Y / N / N Large decrease Y/-  Y, mild to moderate WD N, mild to moderate WD  Stomatal control only Lupin, Helianthus, Eucalyptus. Metabolic Vitus, large sucrose 
Tourneux & Peltier, 1995 Solanum CE / 600 / 12 Much water Discs /dark 0 / hours ? N / Y / Y Discs Y ↓ to comp. pt   N, or very low WD Y/N Extreme conditions assumes PR only 
Escalona et al., 1999 Vitus Field / large/ 14 Drought vs.irrigated / days Y / N / N Attached leaf Decrease  ↓ Y, mild WD Y, moderate/severe WD  RuBP limiting A/Ci 
Tezara et al., 1999, 2008 Helianthus Greenhouse–CE / 500–700/ 14 Soil / 400 / 8 d Y / Y / Y Attached leaf Decrease/increase N / – ↓ Y, mild WD Y, moderate/severe WD Y/N ATP limiting RuBP and A/Ci 
Wingler et al., 1999 Hordeum CE / 460 / 12 Soil / 12 d Y / N / Y –   ?? Y, moderate WD  Ci measured not given. PR indirect evidence not increased 
Tang et al., 2002 Helianthus Glasshouse, large / ? / 14 Soil / several days N / Y / N – N / N  Y, mild WD Y, mild WD  Metabolic limit 
  CE / 700–900/ 12  Leaf pieces          
Haupt-Herting & Fock, 2002 Lycopersicum CE / 200 / 16 Soil/ 200 / 8 d Y / Y / Y Attached leaf – N / –  Y, mild WD Y, moderate WD Y/Y PR increased relative but not absolute sink for e with WD 
Botha et al., 2004 Rhamnus, Nicotiana, Vitus Greenhouse / 600 / 14? Soil / 600 / variable Y / N / Y Attached leaf Variable  ↓ Y, mild WD Y? severe WD, Rubisco limit  Combined data, large errors 
Ennahli & Earl, 2005 Gossypium Glasshouse / winter / 10 Soil / ? / several days Y / N / Y Attached leaf Y ↓ to comp. pt Decrease/increase ↓ Y, mild WD Y, progressive Y/N Calculation of Cc based on fluorescence assuming only PR. Conflicts in data 
Flexas et al., 2006 Glycine CE / 800–1000 / 14 Soil /? Y / N / Y Attached leaf Y ↓ to comp. pt –   Y, to severe WD Y, severe WD, Rubisco Y/N Calc Cc based on fluorescence assuming only PR 
 Nicotiana   Attached leaf          
Zhou et al., 2007 Oryza CE / 400 / 500 / 12 PEG / 500 / 2d Y / N / N Attached leaf Decrease/increase Y mild ↓ Y, mild WD Y, mild WD  ROS increase mild stress 
Ripley et al., 2007 Alloteropsis Field–polytunnel / large / 400 + / 12 Soil / 400 / several days Y / N / Y Attached leaf Decreased (mild WD) N / – ↓ Y, mild WD Y, mild/moderate WD,  Limited WD, high light, RuBP/Rubisco limiting 

Experimental conditions

From Table 1 (which focuses on C3 plants such as sunflower, wheat and bean) we identify four general types of experimental approaches. There is no preferred species, although sunflower is frequently used. Arabidopsis has been little used despite its importance in molecular biology, because of technical difficulties in measuring gas exchange. Thus no standard or model system has been adopted. The great importance of environmental factors, and their interactions with the plant, is often ignored (or not appreciated). Differences in duration and severity of WD interacting with the intensity and duration of light are particularly important. A standardized approach is urgently required (Blum, 1999) for physiological and molecular studies of WD if the current confusion in the literature is to be remedied.

  1. Plants are grown under glasshouse or controlled-environment conditions, often at low light, and samples of leaf are taken and subjected to WD under no or low light, resulting in rapid stress (Kaiser and Heber, 1981; Dietz and Heber, 1983; Kaiser 1984, 1987; Renou et al., 1990; Tourneux and Peltier, 1995).

  2. Plants are grown as above, but subjected to WD under particular conditions before sampling for experimentation: generally with stronger light (Tang et al., 2002).

  3. Plants are grown hydroponically as in (1) with relatively defined water status (Ψ, RWC) applied rapidly (over hours to days) using osmotica (e.g. polyethylene glycol); measurements are made on intact plants, light may differ between studies (Lawlor and Fock, 1975; Lawlor, 1976; Renou et al., 1990; Zhou et al., 2007).

  4. Plants dry a small volume of soil, resulting in progressive decrease in Ψ, RWC, osmotic potential and turgor over several days, during which measurements are made. The rates of decrease in Ψ and RWC depend, amongst other things, on environment, leaf area and gs. However, they do not change linearly with duration or with soil water content because of the soil water characteristic curve (Kramer and Boyer, 1995). This complicates experimentation (Sinclair and Purcell, 2005), requiring well-replicated measurements, sampling, etc, under comparable conditions, generally within a single experiment.

WATER DEFICIT AND PHOTOSYNTHESIS

The schematic in Fig. 1 emphasizes some of the most important cellular structures, metabolic processes and fluxes that determine photosynthesis and are affected by WD. Boyer and co-workers (see Kramer and Boyer, 1995; Tang et al., 2002) in particular have contributed greatly to analysis of the effects of WD on photosynthesis, and Cornic and Briantais (1991), Lawlor and Cornic (2002), Lawlor (2002), Flexas et al. (2004a) and Chaves et al. (2009), have also considered ‘the problem’. Here, we focus on:

  1. metabolic potential for photosynthesis (Apot), which is determined by the capacity of the system related to the amounts and activities of components of light-harvesting, electron transport and energy-transduction processes, and of carbon metabolism, including enzymes (e.g. Rubisco) and processes (RuBP synthesis), of the Calvin cycle.

  2. rate of photosynthesis (A), which is determined by stomatal and internal limitations (gs and gi, respectively) to CO2 diffusion and Apot (Tezara et al., 1999; Lawlor, 2002; Lawlor and Cornic, 2002).

Fig. 1.

Photosynthesis in a C3 leaf under water deficit involves all of the main structures in the compartments of the mesophyll cell. This greatly simplified schematic attempts to provide an overview (see Lawlor, 2001) of the structures and the associated energy, electron, carbon and oxygen fluxes. Light (yellow) is captured by chlorophyll in the antennae and photosystems (PS), and the energy excites the reaction centres. Excited PSII passes electrons (e, red arrows) to PSI via a chain of redox components, reducing them. Simultaneously, e is removed from water and passes to PSII and O2 is released (purple arrows). H+ accumulates in the thylakoid lumen (together with H+ transferred from the cytosol into the lumen by ET; light blue arrows). Passage of H+ through ATP synthase generates ATP from ADP and Pi. ET reduces ferredoxin (Fd) and then NADP+, forming NADPH: ATP and NADPH are used by the Calvin cycle to generate RuBP (dark blue), which reacts with CO2 from the atmosphere, catalysed by the enzyme Rubisco (carboxylase reaction shown as RuBIS–C). From the products of this reaction, the carbon flux (dark blue lines) is to sucrose, some of which may be used in darkness for glycolysis and cell respiration, but most is transported (transporter cross-hatched) to the rest of the plant. Rubisco also catalyses the reaction of RuBP with O2 (oxygenation, shown as RuBIS–O), which ultimately gives rise in the peroxisomes to glycine. This is decarboxylated in the mitochondria, the CO2 produced is photorespiration and e is transferred to the mitochondrial ET chain, where ATP is generated, before reducing O2. In addition, the tricarboxylic acid (TCA) cycle produces CO2 from substrates ultimately derived from sucrose. It may operate in darkness (dark respiration) or in the light (day respiration) depending on the activity of photosynthesis. Electrons from the TCA cycle enter the mitochondrial ET chain, passing to O2 and forming water and transporting H+, which is coupled to ATP synthesis. ATP is used for reactions in the cytosol or chloroplast. In addition reductant (red arrows) from the cytosol (and from the chloroplasts transferred by metabolite – particularly malate – shuttles) may also be oxidized in the mitochondria by NADH and NADPH dehydrogenases, with ET coupled to ATP synthesis. Under water deficit, the stomata close (and internal conductance gm,, termed gi in the text, may change) thus limiting the flux of CO2 to the Calvin cycle, leading to shortage of e acceptors and slowing ET, even if PR increases as a proportion of CO2 assimilation and consumes relatively more e. Excess excitation in the antennae and PSII leads, via the thylakoid H+ concentration, to activation of violaxanthin de-epoxidase, which converts excitation energy to heat by non-photochemical quenching (NPQ). Excess energy leads to reduction of O2 and formation of reactive oxygen species (ROS, large red arrows), which are partly detoxified (ROS W–W) but may accumulate sufficiently to damage components, e.g proteins of PSII and ATP synthase. This impairs ATP synthesis, decreasing RuBP production and hence Apot. Also, low ATP slows protein synthesis and decreases the cell's ability to repair damage caused by ROS, and affects regulation of ion transport. This schematic should be considered together with Fig. 2, which shows the changes in some components and fluxes of CO2, energy, etc.

Fig. 1.

Photosynthesis in a C3 leaf under water deficit involves all of the main structures in the compartments of the mesophyll cell. This greatly simplified schematic attempts to provide an overview (see Lawlor, 2001) of the structures and the associated energy, electron, carbon and oxygen fluxes. Light (yellow) is captured by chlorophyll in the antennae and photosystems (PS), and the energy excites the reaction centres. Excited PSII passes electrons (e, red arrows) to PSI via a chain of redox components, reducing them. Simultaneously, e is removed from water and passes to PSII and O2 is released (purple arrows). H+ accumulates in the thylakoid lumen (together with H+ transferred from the cytosol into the lumen by ET; light blue arrows). Passage of H+ through ATP synthase generates ATP from ADP and Pi. ET reduces ferredoxin (Fd) and then NADP+, forming NADPH: ATP and NADPH are used by the Calvin cycle to generate RuBP (dark blue), which reacts with CO2 from the atmosphere, catalysed by the enzyme Rubisco (carboxylase reaction shown as RuBIS–C). From the products of this reaction, the carbon flux (dark blue lines) is to sucrose, some of which may be used in darkness for glycolysis and cell respiration, but most is transported (transporter cross-hatched) to the rest of the plant. Rubisco also catalyses the reaction of RuBP with O2 (oxygenation, shown as RuBIS–O), which ultimately gives rise in the peroxisomes to glycine. This is decarboxylated in the mitochondria, the CO2 produced is photorespiration and e is transferred to the mitochondrial ET chain, where ATP is generated, before reducing O2. In addition, the tricarboxylic acid (TCA) cycle produces CO2 from substrates ultimately derived from sucrose. It may operate in darkness (dark respiration) or in the light (day respiration) depending on the activity of photosynthesis. Electrons from the TCA cycle enter the mitochondrial ET chain, passing to O2 and forming water and transporting H+, which is coupled to ATP synthesis. ATP is used for reactions in the cytosol or chloroplast. In addition reductant (red arrows) from the cytosol (and from the chloroplasts transferred by metabolite – particularly malate – shuttles) may also be oxidized in the mitochondria by NADH and NADPH dehydrogenases, with ET coupled to ATP synthesis. Under water deficit, the stomata close (and internal conductance gm,, termed gi in the text, may change) thus limiting the flux of CO2 to the Calvin cycle, leading to shortage of e acceptors and slowing ET, even if PR increases as a proportion of CO2 assimilation and consumes relatively more e. Excess excitation in the antennae and PSII leads, via the thylakoid H+ concentration, to activation of violaxanthin de-epoxidase, which converts excitation energy to heat by non-photochemical quenching (NPQ). Excess energy leads to reduction of O2 and formation of reactive oxygen species (ROS, large red arrows), which are partly detoxified (ROS W–W) but may accumulate sufficiently to damage components, e.g proteins of PSII and ATP synthase. This impairs ATP synthesis, decreasing RuBP production and hence Apot. Also, low ATP slows protein synthesis and decreases the cell's ability to repair damage caused by ROS, and affects regulation of ion transport. This schematic should be considered together with Fig. 2, which shows the changes in some components and fluxes of CO2, energy, etc.

The distinction between A and Apot is not just semantic and is sometimes confused as ‘photosynthesis’ may refer to both. WD affects both A via changes in gs (and possibly gi) and through Apot: the former is related largely to tissue water and the latter to metabolism under the conditions prevailing in the chloroplast and cell. Discrepancies in the literature concerning regulation of A revolve around the relative effects of WD on gs and gi and on Apot, and what causes them. Considering C3 mesophytes, in Fig. 2 we have summarized and simplified information from the literature (see Table 1) in order to obtain an overview of changes in amounts, fluxes, etc, of some key components in relation to RWC: this emphasizes general cell water relations and does not assume a mechanism. Recently, emphasis has been placed on gs (Medrano et al., 2002; Flexas et al., 2004a) because it is regarded as the controlling factor. This over-emphasizes gs and under-estimates cellular water status and metabolic factors, which are driving forces determining A and Apot. By its very nature, gs is highly variable, being affected by physiological state (e.g. leaf water status) and environment (e.g. water vapour pressure, CO2 concentration) and so is not a satisfactory basis for comparison. It is as if electrical circuits were analysed in terms of variable resistances, ignoring electrical potential (voltage). In addition, gs reaches a minimum below which much important cell activity occurs, so presenting data on this basis distorts the metabolic responses. Because of the (incompletely) known importance of cellular conditions, evaluation of all limiting factors (of both A and Apot) and cell metabolism is required, preferably by appropriate experimental and statistical techniques.

Fig. 2.

Representation of measured (or calculated) changes in processes (shown in Fig. 1) of leaves of C3 plants resulting from water deficit. As the basis of comparison, the relative water content (RWC) is used, as it indicates the sensitivity of processes to changing water status. Information is generalized from the literature (cited in Table 1), for intact (attached) leaves of plants relatively rapidly stressed, under moderate-to-strong light. Relative changes in the components were related to RWC (where necessary derived from ψ, etc., converted assuming published relationships), and averages used to produce the generalized responses shown. The aim is to indicate how processes change with WD: they should be treated as at best semi-quantitative. Accurate relationships of such types are required if analysis of the effects of WD on photosynthesis is to advance. (A) Leaf water potential, ψ, osmotic potential, π, and turgor pressure, Р. Zero Р is indicated by arrow. (B) Stomatal conductance, gs. (C) gross and net photosynthetic rate, A, photorespiration PR, and dark respiration, RD. (D). Concentration of CO2: substomatal (Ci) and chloroplastic (Cc), and the equilibrium CO2 compensation concentration (Γ). (E) Total Rubisco protein amount per unit leaf area, and initial Rubisco activity, ATP synthase amount and ATP content of whole leaf. (F) Content of sucrose, 3PGA and starch. (G) Content of RuBP, 3PGA (for comparison) and Apot derived from the plateau of A/Ci curves. (H) Total electron transport and its partitioning to Rubisco carboxylation and oxygenation, and to other sinks. (I) Change in NPQ (measured) and ROS (speculative). (J) Variable to maximal fluorescence (Fv/Fm; efficiency of energy capture of open PSII reaction centres) of dark-adapted leaves, photochemical quenching (qP), quantum efficiency of PSII (ϕPSII) and apparent quantum efficiency of CO2 assimilation (ϕCO2).

Fig. 2.

Representation of measured (or calculated) changes in processes (shown in Fig. 1) of leaves of C3 plants resulting from water deficit. As the basis of comparison, the relative water content (RWC) is used, as it indicates the sensitivity of processes to changing water status. Information is generalized from the literature (cited in Table 1), for intact (attached) leaves of plants relatively rapidly stressed, under moderate-to-strong light. Relative changes in the components were related to RWC (where necessary derived from ψ, etc., converted assuming published relationships), and averages used to produce the generalized responses shown. The aim is to indicate how processes change with WD: they should be treated as at best semi-quantitative. Accurate relationships of such types are required if analysis of the effects of WD on photosynthesis is to advance. (A) Leaf water potential, ψ, osmotic potential, π, and turgor pressure, Р. Zero Р is indicated by arrow. (B) Stomatal conductance, gs. (C) gross and net photosynthetic rate, A, photorespiration PR, and dark respiration, RD. (D). Concentration of CO2: substomatal (Ci) and chloroplastic (Cc), and the equilibrium CO2 compensation concentration (Γ). (E) Total Rubisco protein amount per unit leaf area, and initial Rubisco activity, ATP synthase amount and ATP content of whole leaf. (F) Content of sucrose, 3PGA and starch. (G) Content of RuBP, 3PGA (for comparison) and Apot derived from the plateau of A/Ci curves. (H) Total electron transport and its partitioning to Rubisco carboxylation and oxygenation, and to other sinks. (I) Change in NPQ (measured) and ROS (speculative). (J) Variable to maximal fluorescence (Fv/Fm; efficiency of energy capture of open PSII reaction centres) of dark-adapted leaves, photochemical quenching (qP), quantum efficiency of PSII (ϕPSII) and apparent quantum efficiency of CO2 assimilation (ϕCO2).

When water loss from leaves exceeds uptake, WD develops. With a small decrease in RWC of approx. 10–20 %, turgor (P) decreases from 0·7–0·9 to 0 MPa, and Ψ from 0 to –1 MPa (Fig. 2A). Concomitantly, gs (Fig. 2B) and, as a consequence of limitation to CO2 diffusion, A (Fig. 2C) decrease substantially (approx. 30–50 %). This much is generally agreed (Cornic et al., 1987; Cornic and Briantais, 1991; Cornic et al., 1992; Lawlor 1995, 2002; Tezara et al., 1999; Flexas et al., 2004a). However, the decrease in metabolism shown by Apot (Fig. 2G; see section on A/Ci curves, below) that occurs (Tezara et al, 1999; Lawlor 2002) is not observed or accepted by, for example, Cornic and Briantais (1991), Quick et al. (1992), Cornic (2000), Cornic and Fresneau (2002) and Flexas et al. (2004a), who consider that gs determines A, even at large WD. The view is that Apot is sustained, so that with small gs and gi, Ci and Cc decrease substantially, approaching or reaching the compensation point and are responsible for decreasing A (Cornic and Fresneau, 2002). Before discussing this, the role of stomata is considered.

Stomatal conductance under water deficit

Changes in gs depend on hydraulic factors (RWC, Ψ and turgor) in the stomatal apparatus, including transport of water across membranes (which involves aquaporins; Kaldenhoff et al., 2008), and metabolic (e.g. ABA-related) processes (Comstock, 2002; Buckley, 2005; Roelfsema and Hedrich, 2005). Changes in gs may be rapid, occurring within minutes of alterations in the atmospheric humidity or alterations in the root-medium water- or osmotic potential. This serves to regulate water loss in relation to uptake, so RWC decreases very little. In experimental system (4) listed above, an initial approx. 50 % decrease in gs is related to a decrease of approx. 10 % in RWC and 0·5 MPa in Ψ (Fig. 2A). Effects may be rapidly reversible, or longer-term and persistent. ABA, possibly at small concentrations either transported from roots or released from ‘storage’ in the leaf, may interact with hydraulic regulation initially. Most likely, ABA is synthesised de novo and accumulates substantially as turgor is lost (approx. 80 % RWC and –1 MPa decrease in Ψ; Pierce and Raschke, 1980; Cornish and Zeevaart, 1984) and may then dominate regulation, ensuring long-term closure. Regulation of gs is integrated with photosynthetic metabolism and the environment in ways not well understood (Buckley, 2005). A gs substantially smaller than the unstressed value is probably not a long-term solution to WD under field conditions (Legg et al., 1979), unless there is rapid and major adjustments in all aspects of the mechanisms for dealing with excess energy (see Noctor et al., 2002) and imbalance between supply and demand of assimilates and organ growth. Expansion of cells and tissues is rapidly slowed by a small WD and stops at zero P (80–90 % RWC). This must alter the balance between photosynthetic assimilate supply and demand. Under field conditions, with slowly developing WD, plants often do not exhibit large decrease in gs. Rather, long-term adjustment involves smaller leaf area index (LAI). For example, a barley crop growing in the field as WD developed slowly (over many weeks) avoided low RWC by decreasing LAI and thereby water loss, and by changing cell water balance rather than closing stomata. However, a much larger crop under rapid drought responded with decreased RWC and metabolic inhibition (Legg et al., 1979). The composition and function of the photosynthetic system changes during development, and depends ultimately on control of gene expression (Pfannschmidt et al., 2009), allowing some adaptation to conditions, including WD.

Effects of elevated Ca on A and A/Ci curves

Evaluation of the effects of WD on Apot has come from assessing the response of A to increasing external CO2 concentration (Ca) and calculated Ci. The view that gs determines A, even at substantial WD (70–60 % RWC or less), rests on studies where removing the epidermis (and thus gs) or increasing Ca overcomes the limiting gs (Kaiser and Heber, 1981; Dietz and Heber, 1983; Kaiser 1984, 1987; Quick et al., 1992; Tourneux and Peltier, 1995; Cornic, 2000). However, a pattern is apparent (see Table 1, and references therein): these studies used plants grown at low irradiance with WD that developed quickly in no or weak light before measurement. In contrast, similar experiments but using plants grown at higher irradiance (Table 1) show that removal of the epidermis and increasing Ca (Tang et al., 2002) do not restore A to unstressed values, showing that Apot is impaired – a result considered by Flexas et al. (2004a) to be caused by low RWC. Similarly, application of elevated Ca (up to such large concentrations that metabolism was inhibited) failed to reverse the decrease in A (Tezara et al., 1999). Haupt-Herting and Fock (2002) were unable to reverse decreased A of intact and attached leaves with a ten-fold increase in Ca. Zhou et al. (2007) reversed inhibition with increased Ca under mild WD but not severe. These results show that WD under low light had most effect via gs with little or no effect on Apot. In contrast, with stronger light A is not restored by elevated Ca, showing inhibition of Apot. An exception (Quick et al., 1992) is the maintenance of O2 production with very large (15 %) Ca in Eucalyptus, lupin and sunflower, but not in Vitis, possibly because of very active stomatal control. It is likely that the effects of gs and Apot depend on conditions, e.g. the rate and severity of the WD (which also depends on species) and the radiation, with Apot increasing in importance as WD increases under physiologically more relevant conditions (Lawlor, 2002).

Extension of such analyses by measurement of A/Ci responses (curves) on intact plants has provided valuable information. Many studies with relatively long periods of strong irradiance during growth and during slow WD in C3 plants (see Table 1, and Wise et al., 1991; Martin and Ruiz-Torres, 1992; Tezara et al., 1999; Ripley et al., 2007) have observed decreased slopes and plateaux of A/Ci curves, showing inhibition of Apot (Fig. 2G) even at very mild WD (10–15 % loss of RWC; Table 1). Cornic et al. (1987) also observed this, but did not consider that Apot was inhibited. Similarly, Ennahli and Earl (2005) demonstrated a progressive decrease in A/Ci curves with increasing WD; however, the evidence was rejected in favour of substantially decreasing Cc based on fluorescence data, although this may be erroneous [see section on intercellular (mesophyll) conductance, below]. There is further evidence of decreasing Apot from A/Ci curves in the C3 and C4 sub-species of Allopteris semialata, a grass of southern Africa, even with mild WD (Ripley et al., 2007). Rapidly stressed rice responded similarly (Zhou et al., 2007), showing metabolic limitation.

However, many reasons have been given for not accepting such a large body of consistent, reproducible evidence based on measurements using well-stirred air in leaf chambers on well-illuminated, intact leaves droughted when attached to plants. Decreased Apot has been rejected as an artefact of the calculation of Ci (see Cornic and Briantais, 1991; Ennahli and Earl, 2005). This view is supported by reversal of decreased A by elevated Ca on detached (and inevitably damaged) leaf pieces in often unstirred chambers, where water films, large boundary layers and weak light may play a significant role in the responses. Rejection of evidence from attached leaves cites potential errors in measurements of small CO2 and H2O fluxes at large WD, although the same methods are accepted for measuring small fluxes, e.g. at low CO2 and light. Conductance of the cuticle to CO2 becomes more important as stomata close (Boyer et al., 1997), but has only a small effect even at severe WD (Tezara et al., 1999; Flexas et al., 2004a). Other potential technical errors in gas-exchange measurements have been raised (Flexas et al., 2004a, 2007; Ennahli and Earl, 2005). Whilst techniques must be questioned, the errors suggested are largely overcome by application of ‘best practice’ (e.g. tests for leaks, good replication under the same conditions rather than repeats) and are small (Morison et al., 2005, 2007) given the magnitude of the effects of WD.

The most important criticism has come from the perceived role of heterogeneous gs (‘patchy stomata’), which may give erroneous Ci values and thus decrease the slopes and plateaux of the A/Ci curves. Yet theoretical analysis of the distribution of gs across leaves, and its effect on Ci, is inconclusive (see Buckley et al., 1997). Most importantly, there is no substantial experimental evidence of patchy stomata. Indeed, the evidence suggests (even if proof of a negative is not possible) absence of patchiness (Wise et al., 1991, 1992; Giménez, 1992; Gunasekera and Berkowitz, 1992; Martin and Ruiz-Torres, 1992; Osmond et al., 1999; Haupt-Herting and Fock, 2002). There is continued reluctance to accept this direct evidence (Ennahli and Earl, 2005), although Flexas et al. (2004a) concede that ‘patchiness and cuticular conductance may not totally prevent the usefulness of A/Ci curves’. We conclude that data from A/Ci curves are trustworthy and must be fully considered. They show unequivocally that Apot is impaired by increasing WD in many but not all studies, depending on conditions.

Intercellular and chloroplastic CO2 concentrations and intercellular (mesophyll) conductance

Consideration of how Ci and Cc change with WD is required as they are central to, and indicate the state of, photosynthetic metabolism.

Intercellular CO2 concentration

Assimilation of CO2 by well-watered leaves with large gs at a Ca of approx. 350 µmol mol−1 results in a calculated Ci of aprox. 0·7–0·8 of Ca. As gs restricts the supply of CO2, Ci falls to approx. 0·6–0·7 (Fig. 2D), as often observed (e.g. Cornic et al., 1987; Martin and Ruiz-Torres, 1991; Tezara et al., 1999), indicating that Apot is maintained relative to gs. Hence increasing Ca increases Ci and the plateaux of A/Ci curves, showing stomatal control. However, Ci may not always decrease, e.g. sunflower stressed in the field kept a rather constant Ci as A decreased (Wise et al., 1991). Frequently, following an initial fall, Ci/Ca remains rather constant, and increasing Ci with large Ca does not increase A (Martin and Ruiz-Torres, 1992; Tezara et al., 1999; Haupt-Herting and Fock, 2002), indicating inhibition of Apot. The magnitude of the decrease in Ci differs, presumably because of the relative effects of WD on A, Apot, gs and respiration, etc. With further loss of RWC (below approx. 70 %) Ci may increase to approach Ca (Cornic et al., 1987) or reach (Ennahli and Earl, 2005) and eventually exceed it (Lawlor, 1976; Tezara et al., 2008) as CO2 from respiration is emitted in the light (Lawlor and Fock, 1975). Such changes in Ci/Ca have been directly measured (Lauer and Boyer, 1992), so confirming the general validity. However, as discussed, there has been concern about errors in measurements, calculations, etc, so the changes in Ci have not been accorded due weight. This has lead to contradictions in studies, e.g. Ennahli and Earl (2005), where Ci remained high and at large WD CO2 was evolved from leaves in the light (a qualitative effect not easily explained by errors in gas exchange) so Ci must have been greater than Ca. The simplest explanation is that the initially large Apot and small gs decreased Ci, but then inhibition of Apot, with maintenance of RL, increased Ci. This is a consequence of the greater sensitivity of A than RL to WD (see section on respiration, below). In addition, the data of Cornic et al. (1987) were later rejected on the basis of alternative calculations (see Cornic and Fresneau, 2002).

Chloroplast CO2 concentration

Correct values of Cc are essential for understanding photosynthetic metabolism (von Caemmerer, 2000), and they are also required for the correct calculation of gi. Warren (2006) has critically evaluated the several methods used for Cc, and thus gi, calculation. All suffer from substantial and similar assumptions, so whilst providing apparently independent checks they tend to reinforce erroneous interpretations. Methods used to calculate Cc include the following. A major assumption is in apportioning ET to photorespiration (PR), which is determined by deducting the carboxylation flux from the total, assuming alternative fluxes to be zero or very small and constant. Thus, from O2 exchange, Renou et al. (1990), and Tourneux and Peltier (1995) calculated that Cc approached the compensation point, only increasing at very small WD depending on assumptions about respiration. However, judged from isotope O2 and CO2 exchange data (Haupt-Herting and Fock, 2002), not all ET is to PR even in turgid cells, and certainly not with WD. The small decrease in measured Ci with WD suggests that the ratio of oxygenation to caboxylation is smaller than Cornic and Briantais (1991) and Cornic and Fresneau (2002) determined. A constant flux of e to PR is unlikely from current knowledge of respiration and a decrease is also likely (see section on photorespiration, below). Therefore, under WD over-estimation of ET by fluorescence and under-estimation of alternative sinks probably over-estimates PR and so under-estimates Cc, the values of which must be treated with considerable caution for, as Warren (2006) says, the e flux is ‘at best, a semi-quantitative estimate of the rate of linear electron transport’. Hence, it is legitimate to conclude that the very substantial decrease in calculated Cc with WD (Cornic, 2000; Flexas et al., 2008) is an artefact, requiring rigorous examination. We suggest that with relatively small WD, Cc does decrease as Ci falls (Fig. 2D), but as WD decreases Apot further, maintenance of respiration increases Cc and Ci.

  1. Direct measurement of exchange of C and O isotopes. This allows fluxes to be determined under WD, and from them the sinks for e (Haupt-Herting and Fock, 2000, 2002) and Cc (Renou et al, 1990; Tourneux and Peltier, 1995). Assumptions about the nature of the sinks are very important and are discussed later.

  2. Chlorophyll fluorescence. Changes in fluorescence have provided great insight into photosynthetic metabolism, including under WD (e.g. Cornic and Briantais, 1991; Cornic and Fresneau, 2002). The ‘variable J’ and ‘constant J’ methods are widely used to calculate Cc. Linear e transport is calculated from the quantum yield of PSII energy conversion, ϕPSII = (Fm′ – F)/Fm′), derived from maximal and steady-state fluorescence (Fm′ and F, respectively), the total leaf absorptance (α) and the distribution of energy between PSII and PSI (f, usually taken as 0·5). The e transport is apportioned to Rubisco caboxylation and oxygenation, which, together with specificity of Rubisco, allows calculation of Cc. There are assumptions and errors in the methods, suggesting that estimation of Cc is not as quantitative as assumed. (Warren, 2006). Fluorescence is measured from chloroplasts near the tissue surface and so may not represent the population within the leaf, and there is strong evidence that it over-estimates e flux (Haupt-Herting and Fock, 2002; Tezara et al., 2008). Values of α and f require measurement. Fluorescence may over-estimate the total electron transport (ET) and thus the flux to Rubisco oxygenation.

  3. A ‘calibration curve’ (see Lal et al., 1996; Warren, 2006, 2008) of fluorescence against Ca under non-photorespiratory conditions (1–2 % O2) is used to estimate alternative sinks (which all involve e transport to O2) and to draw conclusions (Lal et al., 1996) about Cc under WD. This is illogical as it assumes that fluorescence measured under WD shows Cc. There is sound, independent evidence that the relation between A and metabolism changes substantially compared to the well-watered state. As an example of the problems with the technique, and conclusions drawn from it, we hypothesize that such ‘calibration’ explains the absence of an effect of vapour pressure deficit, but a large effect of soil WD, on calculated Cc and gi in three species (Warren, 2008). Clearly effects of WD differ from changes caused by [CO2] alone, e.g. ATP decreases under WD (often assumed to cause low Cc) but increases with CO2 deficiency (Wormuth et al., 2006). Reliance on techniques that have limited independence, many basic and untested assumptions, and ignore changes in metabolism under WD is not justified.

Intercellular (mesophyll) conductance (gi)

Estimating the conductance, gi, of the path (cell wall, plasmalemma and chloroplast membranes) of transport of CO2 between the intercellular spaces (at Ci) to Rubisco in the chloroplast at Cc (Evans and von Caemmerer, 1996) is currently of major interest (Flexas et al., 2008), and depends on calculation of Cc. Rapid and substantial changes in estimated gi occur according to conditions, and are used to explain many features of CO2 exchange under WD, for example the apparently large difference between Cc and Ci in the study by Ennahli and Earl (2005). Considering the possible role of gi, from the model in Fig. 1, all the CO2 released and contributing to Ci is extra-chloroplastic (no chlororespiration), so to maintain a very low Cc and large Ci would require that the large decrease in gi would be in the chloroplast envelope, not in the plasmalemma or wall (to allow CO2 movement to the intercellular space), requiring a specific mechanism. The uncertainty about estimating Cc must apply to gi. Accepting that gi reflects a real ‘state’ in cells, and ignoring the question as to why such a basic process should be so highly variable and sensitive to conditions, what determines gi is still much debated. It has physical characteristics including solubility of CO2, surface area of intercellular spaces, walls and cytosol, and dimensions of the intercellular spaces, which change as tissues and cells shrink with WD. In addition, gi has a metabolic component (e.g. carbonic anhydrase, which facilitates CO2 movement to Rubisco active sites: aquaporins may act as CO2 channels), which would be changed by WD. The evidence suggests that gi is not greatly affected by WD.

We conclude that responses of A and Apot to gs and elevated CO2 under WD are likely to be a continuum, depending on species, growth conditions, severity and duration of WD, and the environment. At low light, metabolism is not greatly affected, nor may it be in some studies at moderate WD (e.g. Quick et al., 1992) if adaptation of the stomatal response has occurred. However, with higher light, Apot is generally inhibited.

METABOLIC CAUSES OF DECREASED Apot

Here, we consider the effects of WD on light reactions and electron transport, including generation of reactive oxygen species and photoinhibition, followed by ATP synthesis, then Calvin cycle function, including Rubisco functions, RuBP synthesis and photorespiration. An overview of these processes is given in Lawlor (2001) and summarized in Fig. 1.

Light capture, energy use and dissipation

Light capture and energy use are central to any discussion of A and Apot under WD (see Sharp and Boyer, 1985, 1986; Kirschbaum, 1987; Cornic and Briantais, 1991). When leaves are exposed to radiation, photons excite chlorophyll to the singlet excited state (1chl*), which is quenched by several processes (Avenson et al., 2004; Baker et al., 2007), as follows. (1) Variable fluorescence from chlorophyll a associated with PSII. (2) Formation of triplet states of chlorophyll (3chl*) by intersystem cross-over. (3) Energy-dependent quenching (qE), shown by non-photochemical quenching (NPQ) with energy from the antenna chlorophyll of PSII being transferred to zeaxanthin (Z) and dissipated as heat (Müller et al., 2001; Kanazawa and Kramer, 2002; Niyogi et al., 2005), shown as ‘VAZ’ in Fig. 1. Zeaxanthin accumulates when there is excess energy by conversion of violaxanthin (V) via antheraxanthin (A) catalysed by violaxanthin de-epoxidase (VDE; Fig. 1). Conversion of V to Z requires low pH (large [H+]) in the thylakoid lumen, which activates VDE. Accumulation of H+, and thus NPQ, is stimulated by a large proton gradient (ΔpH) across the thylakoid membrane, indicating decreased H+ transport. This may occur as a consequence of inadequate ADP or Pi in normal metabolism, e.g. when CO2 is limited and ATP concentration is large, but it may also occur when ATP synthase is not activated, e.g. by redox regulation of the γ-subunit of ATP synthase (Kanazawa and Kramer, 2002 ). Such a mechanism explains the strong negative correlation between NPQ and ATP under WD (Tezara et al., 2008). (4) Photochemistry. Excitation of the reaction centres of the photosystems induces ET, with water-splitting evolving O2 and releasing H+ to the lumen and transport of e to ferredoxin and synthesis of NADPH. ET also generates the proton motive force, including ΔpH, across the thylakoid membrane required for flux of H+ through ATP synthase, resulting in ATP synthesis (see Avenson et al., 2005). NADPH and ATP are used predominantly in the Calvin cycle for CO2 assimilation. When energy capture is in balance with photochemistry (and so there is little excess energy) fluorescence and NPQ are very small, and 1chl* is rapidly quenched, minimizing the probability of generating reactive oxygen species (ROS: see Noctor et al., 2002).

In unstressed leaves with rapid A, even quite substantial radiation flux can be used in photochemistry without causing accumulation of excess energy, and fluorescence and NPQ are very small. Complex regulation is required to ensure that these fundamental processes function under a wide range of conditions (Scheibe et al., 2005; Rumeau et al., 2007). However, with small WD where A is decreased, the same radiation may exceed the capacity of photochemistry and NPQ rises (Fig. 2I), indicating that the ET chain and redox components are over-reduced compared to the normal state (Cornic and Briantais, 1991). Increased NPQ shows that the lumen pH is very acidic and that transport of H+ through ATP synthase is limiting. Under progressive WD this is not likely to be due to inadequate Pi (Kanazawa and Kramer, 2002; Avenson et al., 2004, 2005) because ATP concentration also decreases (Tezara et al., 1999) and A is small without accumulation of metabolites. In this respect the effects of WD differ from inadequate CO2 supply (Wormuth et al., 2006). We conclude that WD and small A induce over-energization of the thylakoids.

Electron transport

Energy transfer to the reaction centres of PSII and PSI results in ET to ferredoxin and then reduction of NADP+. Because PSII activity is substantially maintained under WD, the potential e flux to acceptors is large (Cornic and Fresneau, 2002). However, as A progressively decreases with WD, so must consumption of NADPH. Thus, with increasing WD total ET decreases (Fig. 2H) as sink capacity falls. The reduced pyridine nucleotide content is remarkably similar with and without WD (Lawlor and Khanna-Chopra, 1984; Tezara et al., 2008), not decreased as Flexas et al. (2004a) state. This is independent evidence that a crucial feature of WD is maintenance of light reactions, ET and reductant status, but with impaired ATP metabolism. We conclude that under WD, as A decreases substantially, ET to carboxylation falls, both absolutely and relatively to PR, decreasing these sinks for e. Then ET to O2 and consumption of reductant by mitochondrial dehydrogenases (see later) become more important sinks.

Oxygen metabolism and electron transport to O2

Electrons from the water-splitting complex enter the photosynthetic ET chain and H+ and O2 are released (Eo, gross O2 evolution; the O2 ‘photosynthesis’ of Tourneux and Peltier, 1995). This is the sole source of O2. However, e reduces O2 via several processes, as follows. (1) Photorespiration results in the transfer of e to O2 via the mitochondrial ET chain, and ATP is generated. Measurements of A, PR and ET and partitioning by mass spectroscopy of O and C isotopes in tomato leaves with progressive WD (Haupt-Herting and Fock, 2002) have demonstrated that Eo and gross O2 uptake (Ou) decreased but that Ou/Eo was greater with WD, showing that O2 reduction increased relative to O2 evolution. Net CO2 uptake (A), gross CO2 uptake (total photosynthesis, TPS) and gross CO2 evolution (all CO2 released in the light, Ec) decreased substantially with increasing WD. Although Ec fell by approx. 40 %, Ec/TPS increased, showing greater respiratory activity, and recycling of evolved CO2 doubled. However, A decreased more than Eo at severe WD so ET was maintained relative to A. Thus, although total ET decreases under WD it is dissociated, in part, from A as it reduces O2. Tourneux and Peltier (1995) also observed, under extreme conditions, decreased Eo with increasing stress, but Ou/Eo increased from approx. 50 to 100 % with WD equivalent to 80 % RWC, below which both decreased in parallel. Such a large increase in O2 uptake was not shown by Haupt-Herting and Fock (2002): Ou/Eo changed from approx. 50 to 60 % with WD under physiologically realistic conditions. Haupt-Herting and Fock (2002) showed that ET to O2 increased relative to gross PR, and was not constant as expected if PR were solely responsible. Sinks for e include the Mehler reaction and the Asada water–water cycles, which may increase as the system becomes more reduced, but probably not greatly. A potential sink not yet explored under WD is oxidation of NADH and NADPH [transferred from the chloroplast by metabolite shuttles (Stitt, 1997) via transporters] by mitochondrial dehydrogenases (Fig. 1; see ‘mitochondrial activity and water deficit’, below). Quantification of sinks for e over a range of WD, in relation to light, is required.

Generation of reactive oxygen species (ROS)

Earlier work has been thoroughly reviewed by Smirnoff (1993), Mittler (2002) and Demmig-Adams et al. (2006). With WD, despite increased qE, components of light-harvesting, photosystems and the ET chain are produced with very negative redox potentials (Mittler, 2002; Apel and Hirt, 2004; Baier and Dietz, 2005). They react with O2 from water-splitting (and intermediates of the process), generating ROS, including singlet oxygen (1O*2) that results from 3chl* donating energy to molecular O2 (Krieger-Liszkay, 2005); superoxide is also formed. This reacts with H+ in the presence of superoxide dismutase (SOD), generating hydrogen peroxide, H2O2, which is convert to water and O2 by peroxidases. Perhydroxy radical, hydrogen peroxide and hydroxyl radical are also synthesized. ROS react with proteins and lipids, causing damage to cellular structures and metabolism, especially associated with photosynthesis. The mitochondrial ET chain and other parts of cell metabolism also produce ROS; systems for dissipation exist (Møller, 2001; Mittler, 2002) but the magnitude compared to chloroplasts under WD is unknown. Probably, generation is much greater in chloroplasts because of their larger, fluctuating energy loads. PR and the Mehler reactions generate H2O2 (Noctor et al., 2002; Luna et al., 2005), the latter using e from ferredoxin, a potentially important reaction during induction of A in allowing ET and development of the H+ gradient for ATP synthesis (Haupt-Herting and Fock, 2002; Noctor et al., 2002). Detoxification of ROS involves reactions with reduced compounds such as ascorbate and glutathione: 1O*2 is removed by reaction with tocopherol (see Asada, 2000; Mittler, 2002; and Noctor et al., 2002, for detailed discussions). Detoxification mechanisms consume reducing power and form water (the ‘water–water cycle’, shown as ‘ROS W–W’ in Fig. 1; see Asada, 2000, for details). The normal capacity of the Mehler reaction to consume e is probably small (Biehler and Fock, 1996; Badger et al., 2000; Haupt-Herting and Fock, 2002), and the water–water cycle also (Noctor et al., 2002). The increase in ROS formation and concentration, and thus potential for damage, depends on the capacity of both synthesis and removal. With rapid development of WD in tissue not adjusted to the energy imbalance caused by large changes in A, ROS accumulation clearly depends on the balance between synthesis and dissipation, all dependent on growth conditions, rate and duration of WD, etc. However, potential damage related to ROS under WD is difficult to assess as this intricate system has not been quantified under clearly defined irradiance and WD. It is unclear if production of ROS increases substantially together with NPQ under WD, or is delayed until NPQ cannot maintain the energy status below a threshold. Blokhina et al. (2003) suggest that as NPQ increases so does ROS accumulation (Fig. 2I). Therefore, when WD develops over days under relatively bright light, ROS-induced damage is observed (Demmig-Adams et al., 2006). Importantly, increased ROS production and the high redox state of the ET chain, etc, induces expression of genes coding for components of energy-dissipating and regulation systems in chloroplasts, allowing acclimation to conditions (Pfannschmidt et al., 2009).

Photosystem activity and photoinhibition of PSII and ATP synthase

Effects of WD were analysed by Sharp and Boyer (1986), Kirschbaum (1987) and Demmig-Adams et al. (2006). Over a wide range of WD, excitation of the antenna chlorophylls and PS reaction centres is maintained, although some decrease in energy transfer, shown by smaller Fo during illumination, may occur in the antenna (Haupt-Hertung and Fock, 2002). Efficiency of PSII measured in dark-adapted leaves by Fv/Fm is generally unimpaired by WD (Fig. 2J; see Tezara et al., 1999) unless severe. Photochemical quenching (qP), a measure of efficiency of PSII, decreases at 60 % RWC, when A is very small. Thus PSII is not impaired by relatively severe WD: this also applies to PSI. Cornic and Briantais (1991) concluded that PSII activity is much less affected by WD than other partial processes in photosynthesis, justifying their view that ‘photosynthesis’ is not sensitive to WD. Insensitivity of PSII to WD is surprising as it is susceptible to damage (photoinhibition, PI), by 1O*2 attack on its D1 (32 kDa) protein, which is very labile and rapidly turned-over (half-life approx. 2 h). The propensity to PI, seen at low CO2 and O2 in bright light, is regarded as a feature of WD, on the assumption (e.g. Kanazawa and Kramer, 2002) that Cc is close to the compensation point. However, evidence for PI during WD is equivocal. Sharp and Boyer (1986) demonstrated in sunflower that quantum yields of CO2 fixation and rates of light- and CO2-saturated A decreased substantially with WD, but were not affected by the light intensity during increasing WD, so that PI did not occur. However, PI developed when CO2 and O2 were almost absent. Kirschbaum (1987) observed that WD decreased the A/Ci relationship in Eucalyptus pauciflora but PI was not a major contributor to it. In contrast, Lu and Zhang (1998) concluded that when WD was imposed gradually on wheat at low light, A and gs decreased significantly without affecting PSII photochemistry or maximal efficiency, and without damaging reaction centres or antennae. However, photochemistry was affected after light adaptation, with decreased efficiency of excitation-energy capture by open PSII reaction centres and quantum yield of PSII ET, and without a significant increase in NPQ and increased PI. Giardi et al. (1996) observed damage to the D1 protein, indicated by decreased qP and Fv/Fm, under WD, with excessive energy load.

Damaged D1 protein is rapidly degraded and new replacement is synthesized and incorporated into PSII reaction centres by mechanisms involving chaperones. Re-synthesis is an important rate-limiting step (see Nishiyama et al., 2001; Yokthongwattana and Melis, 2006; Takahashi et al., 2007; Saibo et al., 2009) but has not been examined under WD. The gene PsbA is chloroplast encoded, suggesting that regeneration might be very susceptible to conditions in the chloroplast, particularly if ATP is limiting under WD. Perhaps damage to PSII is limited by NPQ, or locally by cyclic ET (Rumeau et al., 2007) and by the relatively protective lipid membrane. Regulation of chloroplast (and PSII) energetics is complex (Avenson et al., 2005; Rumeau et al., 2007), with cyclic ET around PSI and modulation of H+ efflux through ATP synthase by ‘sensing’ of stromal metabolites. As metabolites change drastically with WD, the potential for unbalanced regulation is large (Joët et al., 2002). Differences between experiments might be related to differences in the radiation load, susceptibility to damage and rate of repair significantly interacting with WD.

Photoinhibitory damage to ATP synthase is a recently described phenomenon, of great potential importance under WD. The γ-subunit of the ATP synthase complex was preferentially attacked by 1O*2 in a conditional mutant flu of Arabidopsis (Mahler et al., 2007), which accumulated protochlorophyllide in darkness and so generated 1O*2 upon illumination. Damage was close to two regulatory cysteine molecules C178 and C184: the γ-subunit is not surrounded by other proteins and is thus potentially exposed to attack. Damage correlated strongly with a decrease in ATP hydrolysis activity and with increased NPQ (Mahler et al., 2007). As ATP hydrolysis correlates strongly with ATP synthase activity, it suggests that loss of ATP synthase activity may occur under high light and WD when ROS is generated. Direct evidence that WD damages chloroplastic ATP synthase, with the ε-subunit lost from thylakoids to the stroma, is provided by Kohzuma et al. (2008). Normally the ε-subunit binds to the γ-subunit and suppresses ATPase activity, and may have a role in relaxation of the hyper-energized state and regulation of proton movement through the complex. With over-energized conditions, as with WD, loss of the ε-subunit may allow relaxation of hyper-energization and dissipation of the ΔpH and proton motive force, changing energy coupling (Akashi et al., 2004).

Different species of ROS affect other subunits of ATP synthase, e.g. H2O2 impairs the α- and β-subunits, but more slowly than 1O2 damages the γ-subunit.

However, PI damage to ATP synthase remains to be demonstrated under WD. We hypothesize that ATP synthase is damaged and then removed from thylakoids, resulting in the decreased content observed by Tezara et al. (1999). Possibly, the repair cycle for ATP synthase components is not as active as for D1 protein (Nishiyama et al., 2001) or inhibition decreases ATP synthesis, so slowing and disrupting the re-synthesis and repair cycles. PI-related inhibition, resulting in damage to ATP synthase with relatively mild WD observed in isolated chloroplasts (e.g. Keck and Boyer, 1974) and in intact leaves where ATP synthase protein is lost (Tezara et al., 1999), would explain decreased ATP content and Apot, accounting for the differences associated with dim and bright light during WD (Table 1). This testable hypothesis requires that ATP synthase is more sensitive to PI than is PSII, because loss of ATP synthase was detected at WD where Fv/Fm was unaffected, and qP was still large (Tezara et al., 1999). ATP synthase may be more sensitive than PSII to attack by ROS, either because of differences in molecular structure, or it is more accessible to ROS then D1 protein. In contrast with the lipid environment of PSII, the progressive increase in ionic concentrations, particularly of Mg2+ during WD (Younis et al., 1979, 1983), in the aqueous environment of the ATP synthase complex may enhance damage. Interestingly, this may relate to the question of why chloroplast genes have migrated to the nucleus and the role of stress conditions in the process (Cullis et al., 2009). There is a prima facie case that ATP synthase is inhibited by conditions that occur under WD. A detailed, objective examination of the problem is needed.

ATP metabolism under water deficit

We consider that ATP synthesis is of crucial importance to understanding effects of WD. Early evidence of impaired ATP synthase in isolated chloroplasts (see Keck and Boyer, 1974; Tang et al., 2002) was largely ignored or dismissed, but substantiated by evidence from intact leaves (Barlow et al., 1976; Lawlor and Khanna-Chopra, 1984; Tezara et al., 1999, 2008), including for marama bean under WD (M. Searson and M. J. Paul, pers. comm.; see Mitchell et al., 2005). That WD has different effects from changing Ci is shown by the data of Wormuth et al. (2006) in a study of gene expression and metabolic regulation in Arabidopsis subjected to different Ca over several hours, where concentrations of ATP and ATP/ADP ratio were large with zero CO2 but decreased in elevated CO2. ATP accumulation is to be expected where the main sink for e (CO2) is removed yet ET and H+ transport are unaffected. Decreased Apot (Fig 2G) correlated strongly with decreased ATP (Fig. 2E; Tezara et al., 1999). The evidence and explanation (Lawlor, 1995: Tezara et al., 1999; Lawlor 2002), that ATP content ultimately limits Apot, is not accepted by Flexas et al. (2004a, 2006), mainly on the grounds that ATP has not been sufficiently measured (suggesting the need for rectification using proper experimentation and sampling), or decreases only at very large WD and so has no importance for A. They also believe that PR increases, thus increasing ATP, although ATP from dark respiration decreases due to inhibition of mitochondrial ATP synthase, rather than that of chloroplasts.

An indirect measure of ATP synthesis failed to demonstrate any effect of WD in a study by Ortiz-Lopes et al. (1991) on sunflower growing in the field. They demonstrated activation of ATP synthase at severe WD from measurement of the decay in the flash-induced electrochromic absorption change at 518 nm from cytochrome, caused by H+ flux from the inner thylakoid space to the stroma through the activated ATP synthase (CFo–CF1). However, it is not clear that change in the signal is quantitatively related to ATP synthesis. Loss or malfunction of some ATP synthase complexes, with activation of those remaining, albeit with altered decay kinetics (which were observed), could decrease ATP synthesis. No confirmatory measurements of ATP content were made. These interpretations cannot be reconciled with the evidence of a decrease in ATP measured under WD. Clearly, as the main sinks (CO2 assimilation, protein synthesis) for ATP are strongly decreased under WD, yet ATP content decreases, inhibition of ATP synthesis occurs, rather than increased consumption.

Role of ATP synthase in metabolic regulation

Increasingly, ATP synthase activity is regarded as regulating A, ET, energy, NADPH and ATP balance under greatly and rapidly changing environmental conditions (Herbert, 2002), and we suggest particularly so under WD. Mechanisms are discussed by Dal Bosco et al. (2004), Avenson et al. (2005), Wu et al. (2007) and Takizawa et al. (2008). Using spectroscopic techniques, Kanazawa and Kramer (2002) demonstrated that the H+ flux through ATP synthase is slowed by factors (possibly stromal metabolites or Pi) other than the redox regulation of the γ-subunit when Ca is altered and A of leaves decreases. Lumen pH is responsible for the large NPQ so a strong inverse correlation between NPQ and ATP, as observed by Tezara et al. (2008), is expected if impaired ATP synthesis slowed H+ transport. Dal Bosco et al. (2004) showed that inactivation of the ATP synthase γ-subunit prevented ATP synthesis under reducing conditions in Arabidopsis and increased NPQ substantially due to a large H+ gradient. This is further evidence that damage to ATP synthase would produce responses similar to WD.

Calvin cycle under WD

Function of the Calvin cycle is central to CO2 assimilation (von Caemmerer, 2000; Lawlor, 2001). Carbon flux through this complex system depends on many processes and is highly regulated. Analysis has been limited, with focus on amounts of RuBP and 3PGA, and on Rubisco. Measurements have demonstrated that RuBP decreases with WD (Giménez et al., 1992; Gunasekera and Berkowitz, 1992; Tezara et al., 1999), and 3PGA also (Lawlor and Fock, 1977; Tezara et al, 1999). However, Flexas et al. (2004a, 2006) did not consider that WD affected the RuBP content. Wingler et al. (1999) measured approx. 25–30 % decrease in A and 50–60 % in RuBP and 3PGA, although they considered RuBP not to be limiting as it was above estimated concentrations of Rubisco binding sites. With WD, if Apot is not affected, decreased A as a consequence of low gi should increase RuBP and, particularly, ATP contents as they are not consumed, provided that amounts and activities of Calvin cycle enzymes are maintained (Fig. 1). If the enzymes of the cycle are impaired then RuBP should decrease and ATP rise; however, the decrease in RuBP and ATP suggests that the supply of ATP is inadequate. Regulation was discussed by Tezara et al. (1999) and Lawlor (2002), who concluded that ATP supply was the limiting factor.

Rubisco amount and activity

Rubisco amount and, particularly, activity under WD have been extensively studied (see Parry et al., 2002; Flexas et al., 2006). Changes in Rubisco amount and activity are variable between studies and not well correlated with changes in A or metabolites. Generally, Rubisco protein content per unit area of young, mature leaves does not decrease until WD is severe (Fig. 2E; e.g. Giménez et al., 1992; Wingler et al., 1999; Tezara et al., 1999); indeed, it may increase due to leaf shrinkage, although it does decrease in some studies, for reasons unknown (Tezara et al., 2002). Rubisco activity (initial) does decrease with severe WD in some studies (see Flexas et al., 2004a, 2006), suggesting that it may limit A. Wingler et al. (1999) found no decrease in Rubisco activity or activation state despite a large decrease in A with WD in barley. Bota et al. (2004) concluded that Rubisco was more affected by WD than RuBP synthesis, but conditions, sampling, etc., were probably inadequate. There were no data on ATP. Flexas et al. (2006) measured initial activity of Rubisco in Glycine max and Nicotiana tabacum and correlated it with calculated Cc, substantiating their view that low Cc inhibits Rubisco. Judged from the very similar changes in RuBP and 3PGA (Fig. 2G), and RuBP and ATP (Fig. 2E, G; e.g. Tezara et al., 1999), but not in Rubisco, the enzyme does not alter the flux in the Calvin cycle under WD, but other conditions are probably also important (e.g. nitrogen supply).

Analysing regulation of Rubisco is difficult because both amount (determined by breakdown and synthesis) and enzymatic activity change, the latter particularly rapidly. Inhibitors such as analogues of RuBP bind to its active sites, especially at sub-saturating RuBP concentration under WD (Giménez et al., 1992; von Caemmerer, 2000). They are not easily displaced, so that Rubisco activity diminishes. It is restored and regulated by Rubisco activase, a catalytic molecular chaperone that removes inhibitors from the active sites, and requires a large ATP/ADP ratio: as this drops Rubisco activase and Rubisco activity decrease, so slowing Rubisco (Parry et al., 2002; Portis, 2003). Activase is also regulated by redox changes mediated by thioredoxin-f, which alters the response to the ATP/ADP ratio (Portis, 2003). To summarize: loss of Rubisco activity in WD seems more likely to be related to Rubisco activase and lack of ATP than to changes in protein, although the latter is possible.

Sucrose synthesis

Sucrose, starch and 3PGA contents usually fall (Fig. 2F) with progressive WD and decreasing A in sunflower (e.g. Lawlor and Fock, 1977). In Phaseolus (Vessey and Sharkey, 1989) a moderate WD (−1 MPa) decreased A and sucrose synthesis by 70 %, starch by 12-fold, and substantially decreased the A/Ci response, removing O2 sensitivity of A. Sucrose phosphate synthase (SPS) activity fell by 60%. Low Ca also decreased SPS in well-watered tissue; this was reversible by large Ca but not under WD. However, Quick et al. (1989) observed stimulation of SPS activity with WD at large Ca. The contradiction was explained by SPS responding to A, with gs the main control. Vessey et al. (1991) concluded that CO2 supply‘… explains away the last support … for direct effects of water stress on photosynthesis … ’but this is not justified. Flexas et al. (2004a) found the reasons for the greater inhibition of starch than sucrose synthesis with WD ‘not clear’. We suggest that the most important factor is the substantial fall in A, limiting synthesis of substrates, followed by changed regulation, consistent with the known properties of enzymes. Starch synthesis falls precipitately as [3PGA] (Lawlor and Fock, 1977) and ATP decrease and Pi rises (likely but not proven) due to inhibition of ADP glucose pyrophosphorylase activity. Sucrose synthesis decreases because the small concentration of glucose-6-phosphate (G6P) activates SPS kinase, and increased Pi inactivates SPS phosphatase: both effects inactivate SPS (Huber and Huber, 1996). Large Ca at such small WD would probably increase G6P, and thus SPS activity. As WD increases, so inadequate ATP and G6P and increased Pi become more dominant, in addition to low A. The system will be very dependent on conditions. Experimental evidence about metabolites and enzyme activity is consistent with known regulatory mechanisms, and shows that enzyme activities follow metabolism under WD, not regulate it. The conclusion of Flexas et al. (2004a) that lack of response to elevated CO2 is only explicable by ‘a functional limitation … [of] starch-sucrose synthesis’ is not justified and is supported by flawed modelling, with many erroneous assumptions.

Photorespiration (PR)

As A is decreased by falling gs, PR (which is approx. 25 % of A under normal conditions) becomes relatively more important under WD (Fig. 2C; Lawlor and Fock, 1975; Lawlor, 1976; Wingler et al., 1999; Haupt-Herting and Fock, 2002; Noctor et al., 2002) or may entirely replace the decreased A (Cornic and Fresneau, 2002). PR (and PR/A) rise as a consequence of the Rubisco oxygenase reaction, which is determined by the ratio of [O2]/[CO2] at the catalytic sites of Rubisco (see Fig. 1). Phosphoglycollate from the oxygenase reaction is metabolized to glycine, which is decarboxylated by glycine decarboxylase in the mitochondria, producing serine, with release of CO2 (PR) and e: the latter are transferred to O2 via the mitochondrial ET chain thus generating ATP (Fig. 1). As mentioned earlier, the view of Flexas et al. (2004a) and Ribo-Carbo et al. (2005) is that ATP production by mitochondria decreases with WD. Experimental evidence suggests that PR is not as large an absolute sink for e as once thought, but the assumption (or dogma) that it increases substantially is largely unquestioned (e.g. Kanazawa and Kramer, 2002). PR either remained rather constant over a wide range of WD or decreased when measured on rapidly stressed sunflower leaves by means of CO2 exchange with 21 and 1 % O2 – allowing and preventing PR, respectively, although PR/A increased (Lawlor, 1976; Lawlor and Fock, 1975). Using 12CO2 and 13CO2 exchange, Haupt-Herting and Fock (2000, 2002) showed that PR changed little, although PR/A increased. Given the very large decrease in A, the total sink (A + PR) for e is much decreased. Earlier measurements of respiration in the light by CO2 exchange did not adequately account for light respiration, which increased as a proportion of total respiratory CO2 emission determined by 14CO2 measurements and the decrease in specific radioactivity of CO2 emitted (Lawlor and Fock, 1975) and this overestimated PR under WD. This shows that stored C reserves were consumed with WD, in agreement with a substantial fall in sucrose content (Lawlor and Fock, 1977). In addition, the increase in equilibrium CO2 compensation concentration with WD (Γ, Fig. 2D) was unaffected by 1 % O2, i.e. the CO2 was not from PR (Lawlor, 1976). The evidence of Wingler et al. (1999), based on enzyme and metabolite assays, that PR increased in barley under WD is not consistent with smaller glycine and serine contents and a constant gly/ser ratio: PR CO2 exchange was not measured. As RuBP synthesis decreases with WD so both oxygenase and carboxylase reactions will decrease, although the O2/CO2 ratio affects PR/A. We conclude that the total sink for e provided by PR does not increase with progressive WD, so its role (whilst becoming relatively more important than A) in energy regulation is over-estimated. Respiration from the carboxylic acid cycle in the mitochondria becomes relatively more important.

Mitochondrial activity and water deficit

The vital role of mitochondria in photosynthetic carbon metabolism of leaves experiencing WD is considered by Atkin and Macherel (2009), so it is only briefly emphasized here. Mitochondria are responsible for respiration [tricarboxylic acid (TCA) cycle] in darkness – dark respiration, RD – and in the light (RL) and PR, and e from these processes reduces O2, forming water (Fig. 1) and transporting H+, which ultimately results in ATP synthesis (Vedel et al., 1999). Another essential feature is that mitochondria use NADH and NADPH (including from the chloroplast), transferring e, and are thus particularly important in redox regulation (Rasmusson and Møller, 1990; Lin et al., 2008). The multiplicity of dehydrogenases in plant mitochondria and how they function in redox regulation is discussed by Rasmusson et al. (2008). The redox state (NADH + NADPH/NAD+ + NADP+) under WD is comparable to normal conditions (Lawlor and Khanna-Chopra, 1984; Tezara et al., 2008), yet A is inhibited, suggesting over-reduction. We suggest that NAD(P)H is dissipated by mitochondria even at rather mild stress and is probably a major sink for electrons originating in the light reactions, with shuttle systems transferring reductant across the chloroplast envelope to the mitochondria (Stitt, 1997), although there is lack of evidence under WD. Synthesis of ATP in mitochondria is probably very important for ion transport and protein synthesis in the cytosol and it may be available to chloroplasts via efficient membrane transporters (Stitt, 1997).

If the flux of reductant from the chloroplast exceeds the capacity of the mitochondria to generate ATP, reductant may still be dissipated by the mitochondrial alternative oxidase (AOX), which uses e to reduce O2 but without coupling to ATP synthesis. AOX activity depends on a highly reduced state (Vedel et al., 1999) so it becomes more important under WD. There is clear evidence of this in wheat leaves, as the amount of reduced and active AOX protein increased substantially (Bartoli et al., 2005). Inhibition of AOX did not affect fluorescence in well-illuminated and well-watered leaves, but with WD, ϕPSII and qP decreased and NPQ increased greatly, especially when AOX was inhibited, although Fv/Fm was unaffected (Bartoli et al., 2005), indicating that AOX maintains photosynthesis under WD. More active AOX, by dissipation of e, decreases ROS production, whilst decreasing oxidative phosphorylation (Ribo-Carbo et al., 2005). This is taken as evidence by Flexas et al. (2006) that WD inhibits ATP synthesis by mitochondria. It is more likely that ATP synthesis is maximal from mitochondria and that the excess e is used by the AOX when ubiquinone is over-reduced. Further control of the redox and ATP status of the cell and organelles is provided by mitochondrial uncoupler proteins (UCP), which allow H+ to flow without passing through ATP synthase, so preventing ATP synthesis if ATP consumption is inadequate (Sweetlove et al., 2006), but this is not the case under WD. However, UCP is also activated by superoxide (ROS), and is required for the oxidation of glycine to serine in the photorespiratory pathway. Thus, conditions in the cell under WD might activate AOX and UCP as part of regulation (Clifton et al., 2006). We conclude that mitochondria provide several ‘safety valves’ allowing balance to be achieved between e and H+ fluxes, PR and TCA respiratory pathways, and between NAD(P)H and ATP synthesis. It is a matter of urgency to properly evaluate the magnitude of the different sinks for e.

Chloroplasts are much more sensitive to WD than mitochondria because when A stops RL is maintained (Lawlor and Fock, 1975), although ROS is formed in both organelles, and Bartoli et al. (2004) consider mitochondria to be sensitive to ROS. Regulatory systems for dealing with excess e are different in the two organelles: multiple regulatory pathways in mitochondria may provide greater protection. Perhaps the range of energy states is smaller in mitochondria than in chloroplasts, which experience large and rapidly changing radiant energy fluxes. Chloroplasts (particularly those not adjusted to strong light, WD, etc) may have inadequate mechanisms to prevent the accumulation of H+ and to dissipate the H+ gradient if it is large, except through ATP synthase. ATP synthases from both organelles, which have a different evolutionary origins and structure in plants (Hamasur and Glaser, 1992), may also differ in susceptibility to their environment, e.g. ROS and ion concentrations. How conditions in the organelles affect use of reductant and generation of ATP is not known: these topics deserve more attention in the context of WD.

SUMMARY OF REGULATION OF PHOTOSYNTHETIC METABOLISM UNDER DROUGHT STRESS

This analysis of the literature shows that the relative effects of stomatal and metabolic limitations (gs and Apot) depend on species and conditions of growth and experimentation. It has led to development of a qualitative ‘conceptual model’ that accommodates many experimental data and generates specific hypotheses. Under WD the balance between energy capture and metabolism is disturbed, as photochemistry decreases and energy dissipation increases. With mild, relatively rapid WD and decreased gs, A falls substantially and Ci and Cc a little. The magnitude of the change in Cc may be exaggerated because of assumptions and errors in measurements, so gi is not reliable. Light reactions, ET and NADP+ reduction are maintained, causing energy imbalance under mild stress. We hypothesize that this results in synthesis of ROS, which damages ATP synthase due to an interaction with increased ion concentrations in the chloroplast. Damaged complexes are removed from thylakoids while undamaged complexes continue to transport H+ and synthesize ATP. The ensuing decrease in ATP is crucial to chloroplast functions, slowing RuBP synthesis, which then results in loss of metabolic potential (Apot) with an accompanying decrease in Rubisco activity. Changes in Rubisco are not sufficiently well correlated with A to suggest that it has a primary role under WD. The model accounts for decreased A, Apot and increased NPQ, as well as loss of ATP synthase and low ATP content. It also explains changes in metabolites. Stimulation of A by elevated Ca occurs in weak light, where damage to ATP synthase would be minimal, but does not occur in strong light, where ATP synthase is damaged. As A drops, RL is maintained (or may increase), eventually exceeding A: consequently Ci/Ca increases greatly. As a consequence of slowed A combined with decreased ATP, metabolites of the Calvin cycle decrease, and sucrose also. There is a large increase in energy dissipation (NPQ), and maintenance of reductant content and a substantial decrease in ATP/reductant. A crucial test of the model would be to increase ROS production (e.g. by using the conditional flu mutant; Mahler et al., 2007) or to decrease energy dissipation using a mutant, and subject plants to controlled WD under a range of defined radiation. Increased ROS or decreased dissipation would greatly increase sensitivity to WD at low light, and particularly so at high light. The effect would be seen in increased ROS, which would precede the decrease in ATP synthase amount and activity, and decreased ATP content. As a result, A/Ci responses and metabolites would be altered as described. Concomitant measurement of Rubisco amount and activity would test if it changes in relation to A and Apot. We have focused on C3 plants, but the metabolic responses and sensitivity of C4 plants to WD, reviewed by Ghannoum (2009), suggest that the C3 cycle in C4 metabolism is impaired. We hypothesize that damage to ATP synthesis is the determining process in C4 photosynthesis: this is similarly testable. Explanation of the effects of WD and tests thereof requires the correct measurement of, amongst other things, A, Apot and NPQ, and analysis of metabolites, particularly RuBP, ATP, ADP and Pi, using rapid freeze-clamping, extraction, etc, under strictly comparable conditions in order to allow objective comparison of data. This model is very dynamic, considers environmental factors, and explains many observations in the literature without precluding any: it introduces flexibility into current interpretation.

LITERATURE CITED

Akashi
K
Hisabori
T
Ueoka-Nakanishi
H
Ingaki
NI
Yokota
A
A novel behaviour of ε-subunit of chloroplast ATP synthase
13th International Congress on Photosynthesis
 , 
2004
 
Apel
K
Hirt
H
Reactive oxygen species: metabolism, oxidative stress and signal transduction
Annual Review of Plant Biology
 , 
2004
, vol. 
55
 (pg. 
373
-
399
)
Asada
K
The water–water cycle as alternative photon and electron sink
Philosophical Transactions of the Royal Society, Series B, Biological Science
 , 
2000
, vol. 
355
 (pg. 
1419
-
1431
)
Atkin
OK
Macherel
D
The crucial role of plant mitochondria in orchestrating drought tolerance
Annals of Botany
 , 
2009
, vol. 
103
 (pg. 
581
-
597
)
Avenson
TJ
Cruz
JA
Kramer
DM
Modulation of energy-dependent quenching of excitons in antennae of higher plants
2004
, vol. 
101
 
Proceedings of the National Academy of Science
USA
(pg. 
5530
-
5535
)
Avenson
TJ
Kanazawa
A
Cruz
JA
Takizawa
K
Ettinger
WE
Kramer
DM
Integrating the proton circuit into photosynthesis: progress and challenges
Plant, Cell and Environment
 , 
2005
, vol. 
28
 (pg. 
97
-
109
)
Badger
MR
von Caemmerer
S
Ruuska
S
Nakano
H
Electron flow to oxygen in higher plants and algae: rates and control of direct photoreduction (Mehler reaction) and rubisco oxygenase
Philosophical Transactions of the Royal Society, Series B, Biological Science
 , 
2000
, vol. 
355
 (pg. 
1433
-
1446
)
Baier
M
Dietz
K-J
Chloroplasts a source and target of cellular redox regulation: a discussion on chloroplast redox signals in the context of plant physiology
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
1449
-
1462
)
Baker
NR
Harbinson
J
Kramer
DM
Determining the limitation and regulation of photosynthetic energy transduction in leaves
Plant, Cell and Environment
 , 
2007
, vol. 
30
 (pg. 
1107
-
1125
)
Barlow
EWR
Ching
TM
Boersma
L
Leaf growth in relation to ATP level in water-stressed corn plants
Crop Science
 , 
1976
, vol. 
16
 (pg. 
405
-
407
)
Bartoli
CG
Gómez
F
Martínez
DE
Guiamet
JJ
Mitochondria are the main target for oxidative damage in leaves of wheat (Triticum aestivum L.)
Journal of Experimental Botany
 , 
2004
, vol. 
55
 (pg. 
1663
-
1669
)
Bartoli
CG
Gómez
F
Gergoff
G
Guiamet
JJ
Puntarulo
S
Up-regulation of the mitochondrial alternative oxidase pathway enhances photosynthetic electron transport under drought conditions
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
1269
-
1276
)
Biehler
K
Fock
H
Evidence for the contribution of the Mehler peroxidase reaction in dissipating excess electrons in drought-stressed wheat
Plant Physiology
 , 
1996
, vol. 
112
 (pg. 
265
-
272
)
Blokhina
O
Virolainen
E
Fagerstedt
KV
Antioxidants, oxidative damage and oxygen deprivation stress: a review
Annals of Botany
 , 
2003
, vol. 
91
 (pg. 
179
-
194
)
Blum
A
Ribaut
J-M
Poland
D
Towards standard assays of drought resistance in crop plants
Molecular approaches for the genetic improvement of cereals for stable production in water-limited environments
 , 
1999
Mexico
CIMMYT
(pg. 
29
-
35
International workshop, June 1999
Bohnert
HJ
Gong
Q
Li
P
Ma
S
Unravelling abiotic stress tolerance mechanisms – getting genomics going
Current Opinion in Plant Biology
 , 
2006
, vol. 
9
 (pg. 
180
-
188
)
Bota
J
Medrano
H
Flexas
J
Is photosynthesis limited by Rubisco activity and RuBP content under progressive water stress?
New Phytologist
 , 
2004
, vol. 
162
 (pg. 
671
-
681
)
Boyer
JS
Wong
SC
Farquhar
GD
CO2 and water vapour exchange across leaf cuticle (epidermis) at various water potentials
Plant Physiology
 , 
1997
, vol. 
114
 (pg. 
185
-
191
)
Buckley
TN
The control of stomata by water balance
New Phytologist
 , 
2005
, vol. 
168
 (pg. 
275
-
292
)
Buckley
TN
Farquhar
GD
Mott
KA
Qualitative effects of patchy stomatal conductance distribution features on gas-exchange calculations
Plant, Cell and Environment
 , 
1997
, vol. 
20
 (pg. 
867
-
880
)
von Caemmerer
S
Biochemical models of leaf photosynthesis
 , 
2000
Collingwood, Australia
CSIRO Publishing
Chaves
MM
Oliveira
MM
Mechanisms underlying plant resilience to water deficits: prospects for water-saving agriculture
Journal of Experimental Botany
 , 
2004
, vol. 
55
 (pg. 
2365
-
2384
)
Chaves
MM
Flexas
J
Pinheiro
C
Photosynthesis under drought and salt stress: regulation mechanisms from whole plant to cell
Annals of Botany
 , 
2009
, vol. 
103
 (pg. 
551
-
560
)
Clifton
R
Millar
AH
Whelan
J
Alternative oxidases in Arabidopsis: a comparative analysis of differential expression in the gene family provides new insights into function of non-phosphorylating bypasses
Biochimica Biophysica Acta
 , 
2006
, vol. 
1757
 (pg. 
730
-
741
)
Comstock
JP
Hydraulic and chemical signalling in the control of stomatal conductance and transpiration
Journal of Experimental Botany
 , 
2002
, vol. 
53
 (pg. 
195
-
200
)
Cornic
G
Drought stress inhibits photosynthesis by decreasing stomatal aperture – not by affecting ATP synthesis
Trends in Plant Science
 , 
2000
, vol. 
5
 (pg. 
187
-
188
)
Cornic
G
Briantais
J.-M
Partitioning of photosynthetic electron flow between CO2 and O2 reduction in a C3 leaf (Phaseolus vulgaris L.) at different CO2 concentrations and during drought stress
Planta
 , 
1991
, vol. 
183
 (pg. 
178
-
184
)
Cornic
G
Fresneau
C
Photosynthetic carbon reduction and carbon oxidation cycles are the main electron sinks for Photosystem II activity during a mild drought
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
887
-
894
)
Cornic
G
Papgeorgiou
I
Louason
G
Effect of a rapid and slow drought cycle followed by rehydration on stomatal and non-stomatal components of leaf photosynthesis in Phaseolus vulgaris L
Journal of Plant Physiology
 , 
1987
, vol. 
126
 (pg. 
309
-
318
)
Cornic
G
Le Gouallec
J-L
Briantais
JM
Hodges
M
Effect of dehydration and high light on photosynthesis of two C3 plants (Phaseolus vulgaris L. and Elatostema repens (Lour.) Hall f.)
Planta
 , 
1989
, vol. 
177
 (pg. 
84
-
90
)
Cornic
G
Ghashghaie
J
Genty
B
Briantais
J-M
Leaf photosynthesis is resistant to a mild drought stress
Photosynthetica
 , 
1992
, vol. 
27
 (pg. 
295
-
309
)
Cornish
K
Zeevart
JAD
Abscisic acid metabolism in relation to water stress and leaf age in Xanthium strumarium
Plant Physiology
 , 
1984
, vol. 
76
 (pg. 
1029
-
1035
)
Cullis
CA
Vorster
BJ
Van Der Vyver
C
Kunert
KJ
Transfer of genetic material between the chloroplast and nucleus: how is it related to stress in plants?
Annals of Botany
 , 
2009
, vol. 
103
 (pg. 
625
-
633
)
Dal Bosco
C
Lezhneva
L
Biehl
A
, et al.  . 
Inactivation of the chloroplast ATP synthase γ-subunit results in high non-photochemical fluorescence quenching and altered nuclear gene expression in Arabidopsis thaliana
Journal of Biological Chemistry
 , 
2004
, vol. 
279
 (pg. 
1060
-
1069
)
Demmig-Adams
B
Adams
WW
III
Mattoo
A
Photoprotection, photoinhibition, gene regulation and environment
 , 
2006
, vol. 
vol. 21
 
Dordrecht
Springer
 
Advances in photosynthesis and respiration
Dietz
K-J
Heber
U
Carbon dioxide gas exchange and the energy status of leaves of Primula palinuri under water stress
Planta
 , 
1983
, vol. 
158
 (pg. 
349
-
356
)
Ennahli
S
Earl
HJ
Physiological limitations to photosynthetic carbon assimilation in cotton under water stress
Crop Science
 , 
2005
, vol. 
45
 (pg. 
2374
-
2382
)
Escalona
JM
Flexas
J
Medrano
H
Stomatal and non-stomatal limitations of photosynthesis under water stress in field-grown grape vines
Australian Journal of Plant Physiology
 , 
1999
, vol. 
26
 (pg. 
421
-
433
)
Evans
R
von Caemmerer
S
Carbon dioxide diffusion inside leaves
Plant Physiology
 , 
1996
, vol. 
110
 (pg. 
339
-
346
)
Flexas
J
Bota
J
Loreto
F
Cornic
G
Sharkey
TD
Diffusive and metabolic limitations to photosynthesis under drought and salinity in C3 plants
Plant Biology
 , 
2004
, vol. 
a 6
 (pg. 
1
-
11
)
Flexas
J
Bota
J
Cifre
J
, et al.  . 
Understanding down-regulation of photosynthesis under water stress: future prospects and searching for physiological tools for irrigation management
Annals of applied Biology
 , 
2004
, vol. 
b 144
 (pg. 
273
-
283
)
Flexas
J
Ribas-Carbó
M
Bota
J
, et al.  . 
Decreased Rubisco activity during water stress is not induced by decreased relative water content but related to conditions of low stomatal conductance and chloroplast CO2 concentration
New Phytologist
 , 
2006
, vol. 
172
 (pg. 
73
-
82
)
Flexas
J
Díaz-Espejo
A
Berry
JA
, et al.  . 
Analysis of leakage in IRGA's leaf chambers of open gas exchange systems: quantification and its effects in photosynthesis parameterization
Journal of Experimental Botany
 , 
2007
, vol. 
58
 (pg. 
1533
-
1543
)
Flexas
J
Ribas-Carbó
M
Diaz-Espejo
A
Galmés
J
Medrano
H
Mesophyll conductance to CO2: current knowledge and future prospects
Plant, Cell and Environment
 , 
2008
, vol. 
31
 (pg. 
602
-
621
)
Ghannoum
O
C4 photosynthesis and water stress
Annals of Botany
 , 
2009
, vol. 
103
 (pg. 
635
-
644
)
Giardi
MT
Cona
A
Geiken
B
Kucera
T
Masojidek
J
Mattoo
AK
Long-term drought stress induces structural and functional reorganization of photosystem II
Planta
 , 
1996
, vol. 
199
 (pg. 
118
-
125
)
Giménez
C
Mitchell
V
Lawlor
DW
Regulation of photosynthetic rate of two sunflower hybrids under water stress
Plant Physiology
 , 
1992
, vol. 
98
 (pg. 
516
-
524
)
Gunasekera
D
Berkowitz
GA
Heterogeneous stomatal closure in response to leaf water deficits is not a universal phenomenon
Plant Physiology
 , 
1992
, vol. 
98
 (pg. 
660
-
665
)
Hamasur
B
Glase
E
Plant mitochondrial F0-F1 ATP synthases: identification of the individual subunits and properties of the purified spinach leaf mitochondrial ATP synthase
European Journal of Biochemistry
 , 
1992
, vol. 
205
 (pg. 
409
-
416
)
Haupt-Herting
S
Fock
HP
Exchange of oxygen and its role in energy dissipation during drought stress in tomato plants
Physiologia Plantarum
 , 
2000
, vol. 
110
 (pg. 
489
-
495
)
Haupt-Herting
S
Fock
HP
Oxygen exchange in relation to carbon assimilation in water stressed leaves during photosynthesis
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
851
-
854
)
Herbert
SK
A new regulatory role for the chloroplast ATP synthase
Proceedings of the National Academy of Science of the USA
 , 
2002
, vol. 
99
 (pg. 
12518
-
12519
)
Huber
SC
Huber
JL
Role and regulation of sucrose-phosphate synthase in higher plants
Annual Review of Plant Physiology and Plant Molecular Biology
 , 
1996
, vol. 
47
 (pg. 
431
-
444
)
Joët
T
Cournac
L
Peltier
G
Havaux
M
Cyclic electron flow around photosystem I in C3 plants. In vivo control by the redox state of chloroplasts and involvement of the NADH-dehydrogenase complex
Plant Physiology
 , 
2002
, vol. 
128
 (pg. 
760
-
769
)
Kaiser
WM
Response of photosynthesis and dark-CO2-fixation to light, CO2 and temperature in leaf slices under osmotic stress
Journal of Experimental Botany
 , 
1984
, vol. 
35
 (pg. 
1145
-
1155
)
Kaiser
WM
Effects of water deficits on photosynthetic capacity
Physiologia Plantarum
 , 
1987
, vol. 
71
 (pg. 
142
-
149
)
Kaiser
WM
Heber
U
Photosynthesis under osmotic stress
Planta
 , 
1981
, vol. 
153
 (pg. 
423
-
429
)
Kaldenhoff
R
Ribas-Carbo
M
Flexas Sans
J
Lovisolo
C
Heckwolf
M
Uehlein
N
Aquaporins and plant water balance
Plant, Cell and Environment
 , 
2008
, vol. 
31
 (pg. 
658
-
666
)
Kanazawa
A
Kramer
DM
In vivo modulation of nonphotochemical quenching (NPQ) by regulation of chloroplast ATP synthase
Proceedings of the National Academy of Science of the USA
 , 
2002
, vol. 
99
 (pg. 
12789
-
12794
)
Keck
RW
Boyer
JS
Chloroplast response to low leaf water potentials. III. Differing inhibition of electron transport and photophosphorylation
Plant Physiology
 , 
1974
, vol. 
53
 (pg. 
474
-
479
)
Kirschbaum
MU
Water stress in Eucalyptus pauciflora: comparison of effects on stomatal conductance with effects on the mesophyll capacity for photosynthesis, and investigation of a possible involvement of photoinhibition
Planta
 , 
1987
, vol. 
171
 (pg. 
466
-
473
)
Kohzuma
K
Akashi
K
Munekage
Y
, et al.  . 
Allen
JF
Gantt
E
Golbeck
JH
Osmond
B
Preferential decay of the CF1-ε subunit induces thylakoid uncoupling in wild watermelon under drought stress
Photosynthesis. Energy from the sun: 14th International Congress on Photosynthesis.
 , 
2008
Springer
Dordrecht
(pg. 
617
-
621
)
Kramer
PJ
Boyer
JS
Water relations of plants and soils.
 , 
1995
San Diego
Academic Press, Inc
Krieger-Liszkay
A
Singlet oxygen production in photosynthesis
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
337
-
346
)
Lal
A
Ku
MSB
Edwards
GE
Analysis of inhibition of photosynthesis due to water stress in the C3 species Hordeum vulgare and Vicia faba; electron transport, CO2 fixation and carboxylation capacity
Photosynthesis Research
 , 
1996
, vol. 
49
 (pg. 
57
-
69
)
Lauer
MJ
Boyer
JS
Internal CO2 measured directly on leaves. Abscisic acid and leaf water potential cause opposing effects
Plant Physiology
 , 
1992
, vol. 
98
 (pg. 
1310
-
1316
)
Lawlor
DW
Water stress induced changes in photosynthesis, photorespiration, respiration and CO2 compensation concentration of wheat
Photosynthetica
 , 
1976
, vol. 
10
 (pg. 
378
-
387
)
Lawlor
DW
Smirnoff
N
Effects of water deficit on photosynthesis
Environment and plant metabolism: flexibility and acclimation
 , 
1995
Oxford
Bios Scientific Publishers
(pg. 
129
-
160
)
Lawlor
DW
Photosynthesis
 , 
2001
Oxford
Bios Scientific Publishers
Lawlor
DW
Limitations to photosynthesis in water-stressed leaves: stomatal vs
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
871
-
885
metabolism and the role of ATP
Lawlor
DW
Cornic
G
Photosynthetic carbon assimilation and associated metabolism in relation to water deficits in higher plants
Plant, Cell and Environment
 , 
2002
, vol. 
25
 (pg. 
275
-
294
)
Lawlor
DW
Fock
H
Photosynthetic and photorespiratory CO2 evolution of water-stressed sunflower leaves
Planta
 , 
1975
, vol. 
126
 (pg. 
247
-
258
)
Lawlor
DW
Fock
H
Water-stress induced changes in the amounts of some photosynthetic assimilation products and respiratory metabolites of sunflower leaves
Journal of Experimental Botany
 , 
1977
, vol. 
28
 (pg. 
329
-
337
)
Lawlor
DW
Khanna-Chopra
R
Sybesma
C
Regulation of photosynthesis during water stress
Advances in photosynthetic research
 , 
1984
, vol. 
vol. IV
 
The Hague
Martinus-Nijhoff/Dr. W. Junk Publishers
(pg. 
379
-
382
)
Legg
BJ
Day
D
Lawlor
DW
Parkinson
KJ
The effect of drought on barley growth: models and measurements showing the relative importance of leaf area and photosynthetic rate
Journal of Agricultural Science, Cambridge
 , 
1979
, vol. 
92
 (pg. 
703
-
716
)
Lu
C
Zhang
J
Effects of water stress on photosynthesis, chlorophyll fluorescence and photoinhibition in wheat plants
Australian Journal of Plant Physiology
 , 
1998
, vol. 
25
 (pg. 
883
-
892
)
Lui
Y-J
Norberg
FEB
Szilágyi
A
De Paepe
R
Åkerlund
HE
Rasmusson
AG
The mitochondrial external NADPH dehydrogenase modulates the leaf NADPH/NADP+ ratio in transgenic Nicotiana sylvestris
Plant Cell Physiology
 , 
2008
, vol. 
49
 (pg. 
251
-
263
)
Luna
CM
Pastori
GM
Driscoll
S
Groten
K
Bernard
S
Foyer
CH
Drought controls H2O2 accumulation, catalase (CAT) activity and CAT gene expression in wheat
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
417
-
423
)
Mahler
H
Wuennenberg
P
Linder
M
, et al.  . 
Singlet oxygen affects the activity of the thylakoid ATP synthase and has a strong impact on its γ subunit
Planta
 , 
2007
, vol. 
225
 (pg. 
1073
-
1083
)
Martin
B
Ruiz-Torres
NA
Effects of water-deficit stress on photosynthesis, its components and component limitations, and on water use efficiency of wheat (Triticum aestivum L.)
Plant Physiology
 , 
1992
, vol. 
100
 (pg. 
733
-
739
)
Medrano
H
Escalona
JM
Bota
J
Gulias
J
Flexas
J
Regulation of photosynthesis of C3 plants in response to progressive drought: stomatal conductance as a reference parameter
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
895
-
905
)
Mitchell
RAC
Keys
AJ
Madgwick
PJ
Parry
MAJ
Lawlor
DW
Adaptation of photosynthesis in marama bean Tylosema esculentum (Burchell A. Schreiber) to a high temperature, high radiation, drought-prone environment
Plant Physiology and Biochemistry
 , 
2005
, vol. 
43
 (pg. 
969
-
976
)
Mittler
R
Oxidative stress, antioxidants and stress tolerance
Trends in Plant Science
 , 
2002
, vol. 
7
 (pg. 
405
-
410
)
Morison
Jl
Gallouët
E
Lawson
T
Cornic
G
Herbin
R
Baker
NR
Lateral diffusion of CO2 in leaves is not sufficient to support photosynthesis
Plant Physiology
 , 
2005
, vol. 
139
 (pg. 
254
-
266
)
Morison
Jl
Lawson
T
Cornic
G
Lateral diffusion of CO2 inside dicotyledonous leaves can be substantial: quantification in different light intensities
Plant Physiology
 , 
2007
, vol. 
145
 (pg. 
680
-
690
)
Møller
IM
Plant mitochondria and oxidative stress: electron transport, NADPH turnover and metabolism of reactive oxygen species
Annual Review of Plant Physiology and Plant Molecular Biology
 , 
2001
, vol. 
52
 (pg. 
561
-
591
)
Müller
P
Li
X-P
Niyogi
KK
Non-photochemical quenching
Plant Physiology
 , 
2001
, vol. 
125
 (pg. 
1558
-
1566
A response to excessive light energy
Nishiyama
Y
Yamamoto
H
Allakhverdiev
SI
Inaba
M
Yokota
A
Murata
N
Oxidative stress inhibits the repair of photodamage to the photosynthetic machinery
EMBO Journal
 , 
2001
, vol. 
20
 (pg. 
5587
-
5594
)
Niyogi
KK
Li
X-P
Rosenberg
V
Jung
H-S
Is PsbS the site of non-photochemical quenching in photosynthesis?
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
375
-
382
)
Noctor
G
Veljovic-Javanovic
S
Driscoll
S
Novitskaya
L
Foyer
CH
Drought and oxidative load in the leaves of C3 plants: a predominant role for photorespiration
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
841
-
850
)
Ortiz-Lopez
A
Ort
DR
Boyer
JS
Photophosphorylation in attached leaves of Helianthus annuus at low water potential
Plant Physiology
 , 
1991
, vol. 
96
 (pg. 
1018
-
1025
)
Osmond
CB
Kramer
D
Lüttge
U
Reversible, water stress-induced non-uniform chlorophyll fluorescence quenching in wilted leaves of Potentilla reptans may not be due to patchy stomatal responses
Plant Biology
 , 
1999
, vol. 
1
 (pg. 
618
-
624
)
Parry
MAJ
Andralojc
PJ
Khan
S
Lea
P
Keys
AJ
Rubisco activity: effects of drought stress
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
833
-
839
)
Pfannschmidt
T
Bräutigam
K
Wagner
R
, et al.  . 
Potential regulation of gene expression in photosynthetic cells by redox and energy state: approaches towards better understanding
Annals of Botany
 , 
2009
, vol. 
103
 (pg. 
599
-
607
)
Pierce
ML
Raschke
K
Correlation between loss of turgor and accumulation of abscisic acid in detached leaves
Planta
 , 
1980
, vol. 
148
 (pg. 
174
-
182
)
Portis
AR
Rubisco activase – Rubisco's catalytic chaperone
Photosynthesis Research
 , 
2003
, vol. 
75
 (pg. 
11
-
27
)
Quick
WP
Chaves
MM
Wendler
R
, et al.  . 
The effect of water stress on photosynthetic carbon metabolism in four species grown under field conditions
Plant, Cell and Environment
 , 
1992
, vol. 
15
 (pg. 
25
-
35
)
Rasmusson
AG
Møller
IM
NADP-utilizing enzymes in the matrix of plant mitochondria
Plant Physiology
 , 
1990
, vol. 
94
 (pg. 
1012
-
1018
)
Rasmusson
AG
Geisler
DA
Møller
IM
The multiplicity of dehydrogenases in the electron transport chain of plant mitochondria
Mitochondrion
 , 
2008
, vol. 
8
 (pg. 
47
-
60
)
Renou
J-L
Gerbaud
A
Just
D
André
M
Differing substomatal and chloroplastic CO2 concentrations in water-stressed wheat
Planta
 , 
1990
, vol. 
182
 (pg. 
415
-
419
)
Reynolds
MP
Mujeeb-Kazi
A
Sawkins
M
Prospects for utilizing plant-adaptive mechanisms to improve wheat and other crops in drought- and salinity-prone environments
Annals of Applied Biology
 , 
2005
, vol. 
146
 (pg. 
239
-
259
)
Ribo-Carbo
M
Taylor
NL
Giles
L
, et al.  . 
Effects of water stress on respiration in soybean leaves
Plant Physiology
 , 
2005
, vol. 
139
 (pg. 
466
-
472
)
Ripley
BS
Gilbert
ME
Ibrahim
DG
Osborne
CP
Drought constraints on C4 photosynthesis: stomatal and metabolic limitations in C3 and C4 subspecies of Alloteropsis semialata
Journal of Experimental Botany
 , 
2007
, vol. 
58
 (pg. 
1351
-
1363
)
Roelfsema
MRG
Hedrich
R
In the light of stomatal opening: new insights into ‘The Watergate’
New Phytologist
 , 
2005
, vol. 
167
 (pg. 
665
-
691
)
Rumeau
D
Peltier
G
Cournac
L
Chlororespiration and cyclic electron flow around PSI during photosynthesis and plant stress responses
Plant, Cell and Environment
 , 
2007
, vol. 
30
 (pg. 
1041
-
1051
)
Saibo
NJM
Lourenço
T
Oliveira
MM
Transcription factors and regulation of photosynthetic and related metabolism under environmental stresses
Annals of Botany
 , 
2009
, vol. 
103
 (pg. 
609
-
623
)
Scheibe
R
Backhausen
JE
Emmerlich
V
Holtgrefe
S
Strategies to maintain redox homeostasis during photosynthesis under changing conditions
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
1481
-
1489
)
Sharp
RE
Boyer
JS
Key
JL
Kosuge
T
Loss in chloroplast activity at low water potentials in sunflower: the significance of photoinhibition
Cellular and molecular biology of plant stress
 , 
1985
, vol. 
vol. 22
 
New York
AR Liss Inc
(pg. 
41
-
49
UCLA Symposium on Cellular and Molecular Biology, New Series
Sharp
RE
Boyer
JS
Photosynthesis at low water potentials in sunflower: lack of photoinhibitory effects
Plant Physiology
 , 
1986
, vol. 
82
 (pg. 
90
-
95
)
Sinclair
TR
Purcell
LC
Is a physiological perspective relevant in a
Journal of Experimental Botany
 , 
2005
, vol. 
56
 (pg. 
2777
-
2782
‘genocentric’ age?
Smirnoff
N
The role of active oxygen in the response of plants to water deficit and desiccation
New Phytologist
 , 
1993
, vol. 
125
 (pg. 
27
-
58
)
Stitt
M
Dennis
DT
Turpin
DH
Lefebvre
DD
Layzel
DB
The flux of carbon between the chloroplast and cytoplasm
Plant metabolism
 , 
1997
2nd edn.
Harlow, UK
Addison, Wesley, Longman
(pg. 
382
-
400
)
Sweetlove
LJ
Lytovchenko
A
Morgan
M
, et al.  . 
Mitochondrial uncoupling protein is required for efficient photosynthesis
Proceedings of the National Academy of Science of the USA
 , 
2006
, vol. 
103
 (pg. 
19587
-
19592
)
Takahashi
S
Bauwe
H
Badger
M
Impairment of the photorespiratory pathway accelerates photoinhibition of photosystem II by suppression of repair but not acceleration of damage processes in Arabidopsis
Plant Physiology
 , 
2007
, vol. 
144
 (pg. 
487
-
494
)
Takizawa
K
Kanazawa
A
Kramer
DM
Depletion of stromal Pi induces high ‘energy-dependent’ antenna exciton quenching (qE) by decreasing proton conductivity at Cf0-CF1 ATP synthase
Plant, Cell and Environment
 , 
2008
, vol. 
31
 (pg. 
235
-
243
)
Tang
A-C
Kawamitsu
Y
Kanechi
M
Boyer
JS
Photosynthetic oxygen-evolution at low water potential in leaf discs lacking an epidermis
Annals of Botany
 , 
2002
, vol. 
89
 (pg. 
861
-
870
)
Tezara
W
Mitchell
VJ
Driscoll
SD
Lawlor
DW
Water stress inhibits plant photosynthesis by decreasing coupling factor and ATP
Nature
 , 
1999
, vol. 
401
 (pg. 
914
-
917
)
Tezara
W
Mitchell
V
Driscoll
SP
Lawlor
DW
Effects of water deficit and its interaction with CO2 supply on the biochemistry and physiology of photosynthesis in sunflower
Journal of Experimental Botany
 , 
2002
, vol. 
53
 (pg. 
1781
-
1791
)
Tezara
W
Martínez
D
Rengufo
E
Herrera
A
Photosynthetic responses of the tropical spiney shrub Lycium nodosum (Solanaceae) to drought, soil salinity and salt spray
Annals of Botany
 , 
2003
, vol. 
92
 (pg. 
757
-
65
)
Tezara
W
Driscoll
S
Lawlor
DW
Partitioning of photosynthetic electron flow between CO2 assimilation and O2 reduction in sunflower plants under water deficit
Photosynthetica
 , 
2008
, vol. 
46
 (pg. 
127
-
134
)
Tourneux
C
Peltier
G
Effect of water deficit on photosynthetic oxygen exchange measured using 18O2 and mass spectrometry in Solanum tuberosum L. leaf discs
Planta
 , 
1995
, vol. 
195
 (pg. 
570
-
577
)
Vedel
F
Lalanne
E
Sabar
M
Chétrit
P
De Paepe
R
The mitochondrial respiratory chain and ATP synthase complexes: composition, structure and mutational studies
Plant Physiology and Biochemistry
 , 
1999
, vol. 
37
 (pg. 
629
-
643
)
Vessey
TL
Sharkey
TD
Mild water stress of Phaseolus vulgaris plants leads to reduced starch synthesis and extractable sucrose phosphate synthase activity
Plant Physiology
 , 
1989
, vol. 
89
 (pg. 
1066
-
1077
)
Vessey
TL
Quick
WP
Sharkey
TD
Stitt
M
Water stress, carbon dioxide, and light effects on sucrose-phospate synthase activity in Phaseolus vulgaris
Physiologia Plantarum
 , 
1991
, vol. 
81
 (pg. 
37
-
44
)
Warren
C
Estimating the internal conductance to CO2 movement
Functional Plant Biology
 , 
2006
, vol. 
33
 (pg. 
431
-
442
)
Warren
C
Soil water deficit decreases the internal conductance to CO2 transfer but atmospheric water deficits do not
Journal of Experimental Botany
 , 
2008
, vol. 
59
 (pg. 
327
-
334
)
Wise
RR
Sparrow
DH
Ortiz-Lopez
A
Ort
DR
Biochemical regulation during the mid-day decline of photosynthesis in field-grown sunflower
Plant Science
 , 
1991
, vol. 
74
 (pg. 
45
-
52
)
Wise
RR
Ortiz-Lopez
A
Ort
DR
Spatial distribution of photosynthesis during drought in field-grown and acclimated and non-acclimited growth in chamber-grown cotton
Plant Physiology
 , 
1992
, vol. 
100
 (pg. 
26
-
32
)
Wingler
A
Quick
WP
Bungard
RA
Bailey
KJ
Lea
PJ
Leegood
RC
The role of photorespiration during drought stress: an analysis utilizing barley mutants with reduced activities of photorespiratory enzymes
Plant, Cell and Environment
 , 
1999
, vol. 
22
 (pg. 
361
-
373
)
Wormuth
D
Baier
M
Kandlbinder
A
Scheibe
R
Hartung
W
Dietz
K-J
Regulation of gene expression by photosynthetic signals triggered through modified CO2 availability
BMC Plant Biology
 , 
2006
, vol. 
6
 pg. 
15
  
doi:10.1186/1471-2229-6-15
Wu
G
Ortiz-Flores
G
Ortiz-Lopez
A
Ort
DR
A point mutation in the atpC1 raises the redox potential of the Arabidopsis chloroplast ATP synthase γ-subunit regulatory disulphide above the range of thioredoxin modulation
Journal of Biological Chemistry
 , 
2007
, vol. 
282
 (pg. 
36782
-
36789
)
Yokthongwattana
K
Melis
A
Demmig-Adams
B
Adams
WW
iii
Mottoo
AK
Photoinhibition and recovery in oxygenic photosynthesis: mechanisms of a photosystem II damage and repair cycle
Photoprotection, photoinhibition, gene regulation and environment
 , 
2006
Dordrecht, The Netherlands
Springer
(pg. 
175
-
191
)
Younis
HM
Boyer
JS
Govindjee
Conformation and activity of chloroplast coupling factor exposed to low chemical potential of water in cells
Biochimica Biophysica Acta
 , 
1979
, vol. 
548
 (pg. 
328
-
340
)
Younis
HM
Weber
G
Boyer
JS
Activity and conformational changes in chloroplast coupling factor induced by ion binding: formation of a magnesium-phosphate complex
Biochemistry
 , 
1983
, vol. 
22
 (pg. 
2505
-
2512
)
Zhou
Y
Lau
HM
Zhang
J
Inhibition of photosynthesis and energy dissipation induced by water and high light in rice
Journal of Experimental Botany
 , 
2007
, vol. 
58
 (pg. 
1207
-
1217
)

Comments

0 Comments