- © 2009 American Society of Plant Biologists
Abstract
The effects of short-term salt stress on gas exchange and the regulation of photosynthetic electron transport were examined in Arabidopsis (Arabidopsis thaliana) and its salt-tolerant close relative Thellungiella (Thellungiella halophila). Plants cultivated on soil were challenged for 2 weeks with NaCl. Arabidopsis showed a much higher sensitivity to salt than Thellungiella; while Arabidopsis plants were unable to survive exposure to greater than 150 mm salt, Thellugiella could tolerate concentrations as high as 500 mm with only minimal effects on gas exchange. Exposure of Arabidopsis to sublethal salt concentrations resulted in stomatal closure and inhibition of CO2 fixation. This lead to an inhibition of electron transport though photosystem II (PSII), an increase in cyclic electron flow involving only PSI, and increased nonphotochemical quenching of chlorophyll fluorescence. In contrast, in Thellungiella, although gas exchange was marginally inhibited by high salt and PSI was unaffected, there was a large increase in electron flow involving PSII. This additional electron transport activity is oxygen dependent and sensitive to the alternative oxidase inhibitor n-propyl gallate. PSII electron transport in Thellungiella showed a reduced sensitivity to 2′-iodo-6-isopropyl-3-methyl-2′,4,4′-trinitrodiphenylether, an inhibitor of the cytochrome b6f complex. At the same time, we observed a substantial up-regulation of a protein reacting with antibodies raised against the plastid terminal oxidase. No such up-regulation was seen in Arabidopsis. We conclude that in salt-stressed Thellungiella, plastid terminal oxidase acts as an alternative electron sink, accounting for up to 30% of total PSII electron flow.
Salinity in soils is a major global problem and is one that is of growing importance (Pitman and Läuchli, 2002). Research in this area has been limited by the lack of a suitable salt-tolerant genetic model (Bressan et al., 2001; Flowers and Colmer, 2008). A salt-tolerant Arabidopsis (Arabidopsis thaliana) relative, Thellungiella (Thellungiella halophila), is now promising to help in salt stress tolerance research (Volkov et al., 2003; Amtmann et al., 2005). Salinity tolerance is a complex phenomenon, brought about by adaptations in a range of physiological processes. Plants have developed a complex defense system, including ion homeostasis, osmolyte biosynthesis, compartmentation of toxic ions, and reactive oxygen species (ROS) scavenging systems (Hasegawa et al., 2000; Mittova et al., 2004; Stepien and Klobus, 2005; Flowers and Colmer, 2008). Of paramount importance is the process of photosynthesis that is well established as a primary target of many forms of environmental stress, including salinity (Garcia-Sanchez et al., 2002; Liska et al., 2004; Stepien and Klobus, 2006).
Soil salt prevents plants from taking up water, exposing them to drought stress. To conserve water, they close their stomata. This simultaneously restricts the entry of CO2 into the leaf, reducing photosynthesis. At higher concentrations, NaCl may also directly inhibit photosynthesis. When such inhibition occurs, the plant is liable to suffer from oxidative stress. Absorption of sunlight leads to ROS formation, mainly in the chloroplast, either via photoreduction of O2 to form superoxide (the Mehler reaction) or through the interaction of triplet-excited chlorophyll to form singlet excited oxygen (Asada, 2000; Foyer et al., 2002). ROS are highly reactive and can cause widespread damage to membranes, proteins, and DNA. To prevent such damage, there are a number of enzymatic processes in chloroplast to scavenge ROS (Asada, 2000). These are energetically demanding, requiring the synthesis of high concentrations of antioxidants and enzymes.
An alternative strategy, placing less of a metabolic burden on plants, would be to avoid the production of ROS. This can be achieved by regulation of photosynthetic electron transport (Johnson, 2005). Although the role of salt stress in inducing oxidative damage has been widely studied (Hernàndez et al., 2001; Bor et al., 2003; Stepien and Klobus, 2005), the extent to which regulatory processes are induced under such conditions, and the extent to which variation in their capacity determines the degree of damage incurred by plants exposed to salt have not been widely investigated.
Studies so far reported using Thellungiella as a model for salt tolerance have focused on short-term responses to salinity, in particular examining changes in gene expression (Inan et al., 2004; Kant et al., 2006; Wong et al., 2006). Fewer studies have examined the physiology of salt tolerance in this plant and none the effects of salt on leaf physiology. Here, we describe an investigation into the effects of salt stress on the regulation of photosynthesis in Arabidopsis and Thellungiella. We show that these plants respond to salt stress in highly contrasting ways. We discuss the implications of these results for our understanding of salt tolerance.
RESULTS
Ion Concentrations and Chlorophyll Content
Plants of Arabidopsis and Thellungiella were grown for 4 or 6 weeks, before being exposed to a range of salt concentrations. Exposure of Arabidopsis to NaCl concentrations higher than 150 mm resulted in plants dying before the end of the experiment and so higher concentrations were not used. Exposure of Thellungiella to NaCl concentrations up to 500 mm did not result in significant mortality, in line with previous reports (Inan et al., 2004; Taji et al., 2004). The Na+ level determined in control leaf tissue was considerably higher in Thellungiella than in Arabidopsis (Fig. 1A ). This difference disappeared after exposure to salinity, due to a rapid increase in sodium concentration in leaves of Arabidopsis. The accumulation of Na+ in leaves of Thellungiella was much lower at external concentrations between 0 and 150 mm NaCl. Sodium accumulation increased sharply in Arabidopsis leaves over the experiment, whereas leaf Na+ content in Thellungiella increased less, even at higher external concentrations of NaCl. The level measured after 2 weeks salt treatment in Thellungiella subjected to 250 and 500 mm NaCl was similar to that of Arabidopsis exposed to 150 mm NaCl. The two species also differed in their potassium accumulation. The leaf tissue K+ concentration in plants watered with salt-free medium was found to be 30% to 40% higher in Thellungiella (Fig. 1B). Salt treatment and sodium accumulation resulted in a large reduction of K+ content in leaves of Arabidopsis. In contrast, Thellungiella revealed only a limited decline in K+ level. Even in Thellungiella challenged with severe salinity the leaf potassium content was no lower than that in control Arabidopsis.
Changes in leaf Na+ (A) and K+ (B) content over time in Arabidopsis (hatched bars) and Thellungiella (dotted bars). Four-week-old Arabidopsis and 6-week-old Thellungiella were exposed to salt for up to 2 weeks. Plants were subjected (in following sequence on graph) to: 0, 100, and 150 mm NaCl for Arabidopsis, and 0, 100, 150, 250, and 500 mm NaCl for Thellungiella. Data represent the means ± se of at least five replicates. [See online article for color version of this figure.]
Control leaves of Thellungiella had a chlorophyll content that was around 30% higher than in Arabidopsis (Fig. 2 ). Exposure of Arabidopsis to salt resulted in a progressive decline in chlorophyll content. The total chlorophyll measured after 10 d of salt treatment with 100 and 150 mm NaCl dropped by 32% and 48%, respectively. Exposure of Thellungiella to salt did not result in any significant change in leaf chlorophyll content.
The effect of salt treatment on the total leaf chlorophyll content in Arabidopsis (white bars) and Thellungiella (black bars). Four-week-old Arabidopsis and 6-week-old Thellungiella were exposed to: 0, 100, 150, 250, and 500 mm NaCl. Leaves were collected 10 d after initiating salt treatment to determine chlorophyll concentration. Data represent the means ± se of at least five replicates.
Gas-Exchange Parameters
Gas exchange in Thellungiella and Arabidopsis was measured daily for 14 d, starting at the onset of salt treatment (Fig. 3, A and B ). Under control conditions, stomatal conductance in Thellungiella was lower than in Arabidopsis, resulting in a lower transpiration rate (data not shown); however, CO2 assimilation, measured under saturating light and atmospheric CO2, was higher. Exposure of Arabidopsis to salt induced stomatal closure, this being induced rapidly and then developing further over time. This decline was accompanied by a similar drop in CO2 assimilation. In Thellungiella there was no rapid stomatal closure observed, even upon exposure to 500 mm salt. By the end of the experiment, a small decline in stomatal conductance was seen, this resulting in only a slight drop in photosynthesis.
Gas exchange in Arabidopsis (closed symbols) and Thellungiella (open symbols) exposed to: 0 (diamonds), 100 (squares), 150 (triangles), 250 (circles), and 500 (stars) mm NaCl. Leaves were exposed to the white actinic light (PFD 850 μmol m−2 s−1) for 40 min at 25°C in the presence of 370 μL L−1 CO2. Stomatal conductance, gs (A), and CO2 assimilation rate, A (B), were measured. C, CO2 assimilation rate as a function of internal CO2 concentration determined 10 d after initiating salt treatment. Symbols as above. Data points represent the means ± se of at least five replicates. [See online article for color version of this figure.]
Measurements of the relationship between assimilation (A) and calculated internal CO2 concentrations (Ci) showed that this was unaffected in Thellungiella plants exposed to a wide range of salt concentrations (Fig. 3C). This relationship was also unchanged in Arabidopsis subjected to 100 mm NaCl over the first 10 d of salt treatment. In Arabidopsis treated with a high salt concentration (150 mm) the A/Ci relationship was found to be modified. An external CO2 concentration of 2,000 μmol m−2 s−1 was not sufficient to restore carbon fixation to the control level.
Chlorophyll Fluorescence Analysis
Measurements of chlorophyll fluorescence provide detailed information about PSII in intact leaves. In control plants of both species the ratio Fv/Fm, a measure of the maximum quantum yield of photosynthesis, was close to 0.8 (Fig. 4A ), consistent with measurements on a wide range of unstressed higher plants (Bjorkman and Demmig, 1987; Johnson et al., 1993). Exposure to salt did not have any immediate effect on Fv/Fm in either species; however, during the development of salt stress over a 14 d period, this parameter fell in Arabidopsis exposed to either 100 or 150 mm NaCl, consistent with a slow accumulation of photoinhibited PSII. In Thellungiella, no change in Fv/Fm occurred, even at the highest salt concentration.
Changes in the chlorophyll fluorescence parameters in Arabidopsis (closed symbols) and Thellungiella (open symbols) exposed to: 0 (diamonds), 100 (squares), 150 (triangles), 250 (circles), and 500 (stars) mm NaCl. The leaves of plants dark adapted overnight were illuminated with a 1.2-s pulse of white saturating light (PFD 8,000 μmol m−2 s−1) at 25°C in the presence of 370 μL L−1 CO2 to estimate the maximal fluorescence. Following this, the actinic light (PFD 850 μmol m−2 s−1) was switched on and the plant was left for 40 min to reach a steady state. The maximum quantum efficiency (A), photochemical efficiency (B), and NPQ (C) were calculated. Data points represent the means ± se of at least five replicates. [See online article for color version of this figure.]
The parameter ΦPSII provides an estimate of PSII quantum efficiency (Genty et al., 1989). With increasing salt stress and decreasing CO2 fixation, PSII electron transport was inhibited in Arabidopsis, as indicated by the fall in ΦPSII; however, this inhibition was less marked than that of CO2 assimilation (Fig. 4B), probably reflecting a role of photorespiration in consuming excess reducing power (Cornic and Briantais, 1990; Tourneux and Peltier, 1995). In salt-treated Thellungiella, ΦPSII did not fall below the control under any treatment. For plants irrigated with 100 or 150 mm NaCl, there was little change in ΦPSII. In plants exposed to the highest salt concentrations, a considerable increase in ΦPSII was observed, with this increase developing progressively through the experiment.
It is widely observed that plants are able to protect themselves from high light by increasing the dissipation of light energy as heat, measured as nonphotochemical quenching of chlorophyll fluorescence (NPQ). Increasing salinity resulted in a substantial increase in NPQ in Arabidopsis, while in Thellungiella, NPQ remained close to control levels at all salt concentrations (Fig. 4C). The increase in Arabidopsis might result from changes in one of at least two processes: protective high-energy-state quenching or photoinhibition. These processes can be partially distinguished by the kinetics with which they relax following the end of illumination (Maxwell and Johnson, 2000). Measurements of the fast and slow relaxing components of quenching showed that the majority of quenching relaxed rapidly in the dark (NPQf), indicating that it was high-energy-state quenching (Fig. 5 ). However, a part of the quenching was more persistent (NPQs), suggesting the occurrence of photoinhibition in leaves of Arabidopsis due to high NaCl. Both forms of quenching increased in response to salt treatment, each contributing to a similar extent to the overall increase. The increase in total NPQ in Thellungiella was small and was mainly due to an increase in NPQf.
Fast- and slow-relaxing components of NPQ (NPQf and NPQs, respectively) in the leaves of Arabidopsis and Thellungiella subjected to: 0 and 150, and 0 and 500 mm NaCl, respectively. Measurements were carried out 14 d after initiating salt treatment at 25°C in the presence of 370 μL L−1 CO2. Leaves were illuminated with 850 μmol m−2 s−1 actinic light. Data points represent the means ± se of at least three replicates.
PSI Photochemistry
In addition to chlorophyll fluorescence, simultaneous measurements were made of the redox state and turnover of the PSI primary electron donor, P700. With increasing NaCl and time of treatment, P700 became more oxidized in Arabidopsis, rising at maximum of 72% oxidized (Fig. 6A ). No significant changes in P700 redox state were seen in Thellungiella subjected to salt. The conductance of the electron transport chain (gETC) was also estimated (Fig. 6B). This parameter provides information about the extent to which the electron transport chain is being regulated (Golding and Johnson, 2003). In Arabidopsis, there was a progressive decline in gETC in response to salt, implying a down-regulation of the electron transport chain. In Thellungiella no change in gETC was observed.
Changes in PSI parameters in leaves of Arabidopsis (closed symbols) and Thellungiella (open symbols) exposed to: 0 (diamonds), 100 (squares), 150 (triangles), 250 (circles), and 500 (stars) mm NaCl. The redox state of the PSI primary donor (A), the conductance of electron transport chain, gETC (B), the proportion of active PSI centers, P700Act (C), and the PSI ETR (D) were estimated as described in “Materials and Methods.” Measurements were carried out on the same leaves used in Figure 4. Data points represent the means ± se of at least five replicates. [See online article for color version of this figure.]
Golding and Johnson (2003) observed that the proportion of PSI reaction centers that were active (i.e. where P700 could be oxidized by light and then rapidly rereduced in darkness) increased in plants of barley (Hordeum vulgare) exposed to drought stress. This was suggested to represent the activation of a pool of PSI centers involved in cyclic electron flow. Subjecting Arabidopsis plants to salt stress induced a considerable increase in the proportion of active PSI at either salt concentration used (Fig. 6C). In Thellungiella, regardless of salt treatment, the active PSI pool was unaltered. Interestingly, although the proportion of P700 oxidized was measured to be similar in control plants of both species, the proportion of the active P700 in the same plants was notably higher for Thellungiella.
Providing P700 is >20% oxidized, PSI turnover can be estimated from the reduction kinetics of P700+, following transition from actinic light to dark (Clarke and Johnson, 2001; Golding et al., 2005). The electron transport rate through PSI (PSI ETR) estimated in Arabidopsis increased by up to 23% in response to salt treatment (Fig. 6D), in spite of the greatly inhibited rate of PSII electron transport. This implies an increase in cyclic electron flow around PSI in Arabidopsis exposed to salt stress. In salt-treated Thellungiella, PSI ETR was unaltered.
Electron Transport to Oxygen under Salt Stress
Reducing potential produced by photosynthetic electron transport can be consumed by a number of alternative pathways. Of particular note are reactions involving oxygen. Two major oxygen-using pathways have been described: photorespiration and the Mehler reaction. To evaluate the importance of these pathways, the oxygen dependence of electron transport was examined. Arabidopsis and Thellungiella were exposed to a range of irradiances in the presence of 2,000 μL L−1 CO2 and either 21% or 2% O2. In control plants of both species, PSII ETR, calculated as the product of PSII photochemical efficiency (ΦPSII) and photon flux density (PFD), exhibited a standard light response reaching a maximum at around 600 μmol m−2 s−1 of light (Fig. 7, A and B ). Subjecting control plants to low (2%) oxygen resulted in a decrease of PSII ETR at saturating irradiances. NPQ was unaffected by low O2 at all irradiances (Fig. 7, C and D).
Oxygen dependence of electron transport: PSII ETR (A and B), NPQ (C and D), P700+ (E and F), gETC (G and H), and PSI ETR (I and J), measured in leaves of Arabidopsis and Thellungiella subjected to NaCl (circles): 150 mm for Arabidopsis and 250 mm for Thellungiella. Control plants (squares) were maintained in a NaCl-free soil. The measurements were carried out 10 d after initiating salt treatment under saturating CO2 (2000 μL L−1), at 25°C and under 21% (closed symbols) or 2% (open symbols) oxygen. Data points represent the means ± se of at least five replicates. [See online article for color version of this figure.]
Simultaneous measurements of the redox state and turnover of the PSI primary electron donor revealed slight effects of oxygen concentration in control plants. With increasing irradiance, P700 became progressively more oxidized in both species (Fig. 7, E and F). Although the proportion of P700 oxidized was insensitive to oxygen in control plants, the conductance of the electron transport chain (gETC) decreased and the PSI ETR was lowered by a similar amount (Fig. 7, G–J).
In Arabidopsis exposed to 150 mm salt, maximum PSII ETR at high CO2 was lower than in control plants, consistent with observations at ambient CO2. As in the control, electron transport through PSII fell slightly in response to low O2 (Fig. 7A). In contrast, salt treatment resulted in an increase in PSII ETR in Thellungiella relative to plants not exposed to salt; however, this increase in electron transport was entirely abolished when the O2 concentration was lowered (Fig. 7B). In salt-treated Arabidopsis, lowering the O2 concentration resulted in a decrease in NPQ (Fig. 7C). In Thellungiella, NPQ was insensitive to O2 concentration (Fig. 7D).
The proportion of P700 oxidized in salt-stressed Arabidopsis was significantly higher under low O2 conditions (Fig. 7E). This, in spite of the small reduction in gETC, resulted in the electron transport through PSI being maintained at the same level (Fig. 7I). In contrast, PSI ETR in Thellungiella subjected to NaCl, like the control, decreased at low O2 (Fig. 7J). This resulted from an increase in P700 oxidation and a drop in gETC.
Plastid Terminal Oxidase as a Sink for Electrons
Results in Figure 7 strongly indicate that the additional turnover of PSII seen in salt-treated leaves is due to electron transport to oxygen. Given that sensitivity to oxygen is seen at high CO2, we can exclude a contribution of photorespiration to this effect. Photoreduction of oxygen may occur at the acceptor side of PSI, via the Mehler reaction; however, the absence of a sensitivity of PSI parameters to oxygen tends to speak against this. It has been observed that higher plant chloroplasts contain a putative quinone-oxygen oxidoreductase, the plastid terminal oxidase (PTOX) or IMMUTANS protein. To determine whether the PTOX plays a role in electron transport from PSII to oxygen, measurements of PSII ETR were repeated in control and salt-treated leaves vacuum infiltrated with either water or a solution of the PTOX inhibitor n-propyl gallate (nPG; 3,4,5-trihydroxy-benzoic acid-n-propyl ester; Josse et al., 2003). In Arabidopsis, PSII ETR was insensitive to nPG, regardless of whether plants had been exposed to salt treatment (Fig. 8A ). This was also the case in control Thellungiella. In Thellungiella exposed to 250 mm NaCl, PSII ETR was sensitive to nPG. PSII ETR, measured 10 d after initiating salt treatment, was reduced by 35% in leaves infiltrated with to 1 mm nPG, falling to the control level.
The effect of nPG (A) and DNP-INT (B) on PSII photochemical efficiency measured in the leaves of Arabidopsis and Thellungiella subjected to: 0 and 100, and 0 and 250 mm NaCl, respectively. Measurements were carried out 10 d after initiating salt treatment at 25°C in the presence of 370 μL L−1 CO2. Leaves were illuminated with 850 μmol m−2 s−1 red light. Leaves were vacuum infiltrated with water (white bars) or with 1 mm nPG (A) or 35 μm DNP-INT (B) in the presence of 21% (gray bars) and 2% (black bars) oxygen. Data points represent the means ± se of at least five replicates.
The effect of nPG suggests the activity of a PTOX protein in Thellungiella leaves; however, it does not rule out a contribution of the Mehler reaction to electron transport. To measure electron transport to oxygen in the absence of that reaction, leaves were infiltrated with the cytochrome b6f inhibitor 2′iodo-6-isopropyl-3-methyl-2′,4,4′-trinitrodiphenylether (DNP-INT), a specific inhibitor of the Qo-binding site (Trebst et al., 1978). In Arabidopsis, this almost completely blocked PSII electron transport, regardless of salt treatment (Fig. 8B). In control Thellungiella leaves, DNP-INT also largely blocked PSII electron transport, although a significant residual level of ΦPSII remained. In salt-treated Thellungiella leaves, DNP-INT only partially inhibited electron transport. Lowering O2 further inhibited electron transport. The extent of DNP-INT-insensitive, oxygen-sensitive electron transport was close to that of nPG-sensitive electron transport in the same leaves.
Immunoblot analyses of thylakoid membrane extracts using antibodies raised against the PTOX from Arabidopsis revealed the presence of a 40-kD band in both Arabidopsis and Thellungiella leaves (Fig. 9 ). The band detected in control plants, loaded on the basis of equal protein content, was more prominent in Thellungiella than in Arabidopsis. In Arabidopsis, subjecting plants to salt did not result in any change in the estimated PTOX content. Thellungiella responded to 250 mm NaCl by increasing polypeptide content nearly 4-fold, relative to the plants maintained in a NaCl-free medium.
The effect of salt treatment on PTOX protein expression in the leaves of Arabidopsis and Thellungiella subjected to: 0 and 100, and 0 and 250 mm NaCl, respectively. Leaves from control and salt-treated plants were collected 10 d after initiating salt treatment for immunodetection after SDS-PAGE, separation of 35 μg protein from the thylakoid membrane samples, and electrophoretic transfer to nitrocellulose membrane. Immunoblot was quantified by the optical densitometry. Data points represent the means ± se of six blots from three separate membrane preparations. Insert shows typical bands from an original blot, loaded on an equal protein basis.
DISCUSSION
A considerable effort has been made, over many years, to understand the fundamental basis of salt-tolerance physiology. A number of studies have examined the responses of Arabidopsis to salt (Zhu, 2001). This has the advantage of being a well-studied model system with a wealth of molecular and genetic information; however, as is illustrated here, its usefulness is limited by the fact that it is a glycophyte. Thellungiella in contrast has a high degree of salt tolerance, as well as tolerance of other environmental stresses, but, as a close Arabidopsis relative, shares a high degree of homology and is amenable to many of the same experimental approaches. In particular, a comparative approach studying both species has the potential to identify key characteristics that might be transferred between species to enhance stress tolerance.
Exposure of Arabidopsis to salt resulted in substantial Na+ uptake (Fig. 1) and loss of chlorophyll (Fig. 2) and induced stomatal closure (Fig. 3A). The resulting limitation on CO2 entry and any direct toxic effect of salt accumulation resulted in inhibition of CO2 assimilation. Consistent with previous findings (Allakhverdiev et al., 2000; Stepien and Klobus, 2005; Stepien and Klobus, 2006), there were indications that the highest salinity caused nonstomatal limitation of photosynthesis. High CO2 (2,000 μL L−1), the highest experimentally available in our gas-exchange system, failed to reverse the inhibition of assimilation, although internal Ci is liable to be overestimated under these conditions.
The inhibition of assimilation in salt-stressed Arabidopsis is accompanied by a decrease in electron transport through PSII, indicated by the decline in ΦPSII, and cumulative damage to PSII, indicated by the progressive drop in Fv/Fm. This was not, however, mirrored in the responses of PSI electron transport. Measurements of PSI electron transport have been the subject of some controversy recently (see Johnson, 2005 for discussion). We have based our estimates on the kinetics of rereduction of P700+ following a light to dark transition (Clarke and Johnson, 2001); however, using estimates based on the proportion of centers that are reduced but can be oxidized (P700Act-P700+; see Klughammer and Schreiber, 1994), we reach the same conclusions. PSI turnover does not drop under salt stress in line with assimilation and PSII turnover and indeed we argue that it rises under certain experimental conditions. From this we conclude that cyclic electron flow around PSI increases under salt stress. This is in line with studies of the responses of a number of plant species to a variety of stresses, including for example studies of high light, chilling, and low CO2 or drought stress (Clarke and Johnson, 2001; Golding and Johnson, 2003; Miyake et al., 2005a, 2005b). The increase in cyclic electron transport was accompanied by an increase in NPQ. NPQ is made up of both reversible, ΔpH-dependent quenching, which protects PSII from light damage, and irreversible NPQ, which reflects photoinhibition. Although we did not attempt to resolve fast and slowly relaxing across all experimental conditions, measurements under selected conditions showed that the additional quenching resulted from both these processes increasing (Fig. 5).
The inhibition of linear electron flow in Arabidopsis under salt stress was accompanied by a down-regulation of electron flow through the cytochrome b6f complex, as indicated by the decline in the conductance of the electron transport chain, gETC. A similar response has been seen previously in response to low CO2 or drought (Harbinson, 1994; Golding and Johnson, 2003). This down-regulation has recently been shown to occur in response to changes in the redox state of the NADP/NADPH pool and serves to limit oxidative stress (Hald et al., 2008).
Overall, the response of Arabidopsis to salt stress is very much in line with that expected from the responses of a variety of other species to a variety of stresses. Down-regulation of linear electron transport limits oxidative stress and increased cyclic flow enhances photoprotective energy dissipation.
In contrast, the response of the halophyte Thellungiella was rather different. Our results confirmed the salt-tolerant nature of Thellungiella (Inan et al., 2004; Taji et al., 2004). Plants grew rapidly at moderate salinity and survived exposure to concentrations of up to 500 mm NaCl without significant mortality. In terms of the responses of photosynthesis, even the highest salt concentration used did not represent a substantial stress. The chlorophyll content of leaves did not drop significantly in response to salt (Fig. 2) and both stomatal conductance and assimilation at atmospheric CO2 concentrations were maintained (Fig. 3).
Given the ability of Thellungiella to maintain stomatal conductance and assimilation even under severe salt stress, it might be expected that the photosynthetic apparatus would show little or no response to salt. This is however not the case. Even at the highest salt concentration, there is no sign that the photosynthetic apparatus is stressed, there is no decline in Fv/Fm, which would indicate photoinhibition, and no increase in NPQ (Fig. 4). There was, however, a substantial effect of salt treatment on PSII photochemistry. With prolonged exposure to salt, the quantum efficiency of PSII, ΦPSII, increased.
An increase in ΦPSII might be explained in a number of ways. Most simply, this could reflect salt-induced changes in PSII concentration and/or antenna complex size. If the concentration or antenna size of PSII decreased, then the efficiency of each reaction center would have to increase to maintain the same overall electron flux. However, given that there is no significant change in either chlorophyll content (Fig. 2) or chlorophyll a/b ratio (data not shown), this is unlikely to play a major role. The fact that there is no concomitant increase in the CO2 assimilation rate measured in the same plants, strongly suggests that there must be an alternative pathway for use of electrons coming from PSII in Thellungiella exposed to severe salinity. A number of pathways in addition to CO2 fixation are known to derive their reducing potential from the photosynthetic electron transport, including nitrogen and sulfur metabolism, and the demand for these certainly might increase in Thellungiella under salt stress to meet the needs, for example, for Pro accumulation as a compatible solute. However, the increase in ΦPSII we observe is seen only at high light—at growth light intensities there was no significant change in ΦPSII. This implies that it is not simply an increase in the competition from an alternative electron sink that has increased but the overall capacity of that alternative sink. Crucially, the additional electron transport through PSII was sensitive to low O2 concentrations (Fig. 7B). This indicates that it is oxygen acting as the additional sink for electrons.
There are a number of different ways in which oxygen might act as a sink for reducing equivalents from PSII. The most widely known is photorespiration, when the carbon-fixing enzyme Rubisco reacts ribulose bisphosphate with oxygen rather than CO2. This occurs in competition with CO2 fixation and can be suppressed when leaves are supplied with high CO2. In our experiments however the oxygen sensitivity of PSII was also seen at high CO2 concentrations, so photorespiration cannot explain the data. Photoreduction of oxygen can occur at the acceptor side of PSI to produce superoxide—the Mehler reaction. The protective potential of this reaction has been widely discussed. It will provide a sink for reductant, taking electrons away from the electron transport chain and so tend to protect PSII from photoinhibition. It is also believed to generate a ΔpH across the thylakoid membrane, supporting protective, NPQ of energy (Asada, 2000; Chen et al., 2004); however, it is liable to produce ROS. These are tightly controlled by coupling the Mehler reaction to detoxifying processes (Asada, 2000), but will still impose a metabolic burden on the plant. Also, given that the primary reaction involved is the essentially spontaneous, uncatalyzed reaction between iron sulfur centers and molecular oxygen, it is not clear how the capacity of this would increase. The Mehler reaction is limited by control of electron flux through the cytochrome b6f complex (Hald et al., 2008); however, there is no indication that this is altered in Thellungiella under salt stress.
The Mehler reaction involves consumption of reducing potential after PSI. Therefore, this should be reflected in an increased PSI turnover, in addition to the enhanced PSII photochemistry observed. This we do not observe. The PSI ETR (Fig. 6D) remained unaltered in response to salt. Furthermore, PSI electron transport was seen to be relatively insensitive to altering oxygen concentration; indeed, PSI turnover tended to increase at low oxygen. This speaks against the Mehler reaction being involved in acting as a sink for electrons from PSII, although a contribution of this pathway cannot be totally excluded.
Various authors have discussed the possibility of oxygen acting as an electron sink prior to PSI. Reduction of oxygen might occur at PSII itself, either directly at the QB site or via cytochrome b559 (Cleland and Grace, 1999; Bondarava et al., 2003). It has also been discussed that oxygen might oxidize plastoquinol directly, although the extent to which this can occur spontaneously is uncertain (Khorobrykh and Ivanov, 2002). The reaction with PQH2 can however occur in an enzyme-catalyzed reaction in the chloroplast, mediated by the PTOX (Aluru et al., 2006). The PTOX protein was first detected as the protein responsible for the variegated or bleached phenotype of various mutants (immutans or ghost mutants; Carol et al., 1999; Josse et al., 2000). PTOX shares sequence similarity with the alternative oxidase in the mitochondrion and is able to divert the electron flow from PSII via PQ to O2, producing water rather than superoxide (Josse et al., 2003). The variegated phenotype seen in the immutans mutants arises because PTOX participates in carotenoid biosynthesis. It is also suggested to participate in a pathway of chlororespiration, together with an NADPH plastoquinone oxidoreductase complex, ndh (Cournac et al., 2002; Joet et al., 2002; Aluru et al., 2006). The exact role of chlororespiration remains obscure; however, it is suggested to function in regulating cyclic electron flow around PSI (Joet et al., 2002).
In Thellugiella, we observe a substantial up-regulation of PTOX protein in response to salt stress (Fig. 9). In Arabidopsis, PTOX is only ever expressed at very low concentrations, approximately 1% of PSII, and is therefore thought unlikely to ever act as a significant electron sink (Lennon et al., 2003). Based on immunoblot analysis, it seems possible that even in control conditions, the PTOX content of Thellungiella leaves is substantially higher and this is increased significantly in response to salt stress (Fig. 9). Furthermore, in salt-treated leaves of Thellungiella, electron transport was observed to be sensitive to the known PTOX inhibitor nPG. No such sensitivity was seen in either control Thellungiella or control or salt-treated Arabidopsis. In salt-treated Thellungiella, in the presence of the cytochrome b6f inhibitor DNP-INT, there was a substantial residual electron transport activity, indicated by ΦPSII measurements, which was sensitive to oxygen (Fig. 8). These observations provide strong evidence that PTOX is acting as the additional electron sink in salt-treated Thellungiella.
High expression of PTOX has been observed previously. The montane plant Ranunculus glacialis acclimated to high-light and low-temperature conditions was demonstrated to possess an alternative electron sink and to contain high PTOX protein abundance (Streb et al., 2005). These authors suggested this to serve as a mechanism to consume electrons; however, apart from protein expression, no practical test for the PTOX activity was shown. This lead Rosso et al. (2006) to question whether PTOX was in fact the electron sink in R. glacialis. They examined plants of Arabidopsis in which the native PTOX was overexpressed and were unable to detect any effect on the redox poising of the PSII acceptor pool or any protection from photoinhibition. Rosso et al. (2006) did not however demonstrate that the overexpressed PTOX was catalytically active. Joet et al. (2002) examined the effect of overexpressing Arabidopsis PTOX in tobacco (Nicotiana tabacum) leaves and were able to detect activity; however, the rate of turnover shown was low, and no data was presented to indicate that PTOX significantly affected steady-state electron transport.
The contradictions in the above studies might be explained in a number of ways. Diffusion of plastoquinol in the thylakoid membrane is heavily restricted, due to the high protein concentration (Kirchhoff et al., 2008). So, to act as an efficient electron sink, PTOX would probably need to be colocated with PSII. In spinach (Spinacia oleracea), PTOX is localized to the stromal lamellae (Lennon et al., 2003) whereas PSII is concentrated in the granal stacks (Albertsson, 2001). This would make it difficult for PTOX to act as a significant direct electron sink from PSII. It may be, therefore, that the targeting of PTOX within the thylakoid membrane is different in stress-tolerant plants such as Thellungiella and R. glacialis. Alternatively, additional regulation of PTOX may be required. The mitochondrial alternative oxidase possesses a regulatory disulphide that is not conserved in the chloroplast homolog (Berthold and Stenmark, 2003); however, some alternative form of regulation can be envisaged.
In conclusion, we have shown substantial differences in the responses of the photosynthetic apparatus of Arabidopsis and Thellungiella to salt stress. In particular, Thellungiella induces an alternative pathway for electron transport that may protect the leaf under stress. This pathway is absent in Arabidopsis; however, a better understanding of how this pathway operates in Thellungiella may open the possibility of engineering this into more stress-sensitive plants.
MATERIALS AND METHODS
Plant Material and NaCl Treatment
Seeds of Arabidopsis (Arabidopsis thaliana; ecotype Columbia 0) and Thellungiella (Thellungiella halophila; ecotype Shandong wild type) were stratified at 4°C for 5 d and then germinated in a controlled-environment cabinet (E.J. Stiell) in an 8-h photoperiod (photosynthetic photon flux of 120 μmol m−2 s−1 provided from cool-white fluorescent bulb), at 23°C/15°C (day/night). A short day length was used to delay flowering in Arabidopsis. One-week-old seedlings were transferred to 7.5-cm pots filled with Viking MM peat-based compost. Four-week-old Arabidopsis and 6-week-old Thellungiella plants, similar in size, were irrigated with 0, 50, 100, 150, 250, and 500 mm NaCl for up to 14 d.
Determination of Na+ and K+ Content
Leaves of control and treated plants were harvested and washed with deionized water. The leaf samples were dried at 105°C for 1 h, subsequently at 60°C for 48 h, and then weighed for determination of dry weight. Lyophilized leaves were milled to powder for mineral nutrient analyses. Powdered samples (0.5 g) were then extracted with 10 mL of HNO3 for 60 min at 95°C. The resulting solutions were filtered through Whatman filter paper, diluted appropriately, and analyzed for Na+ and K+. Cation concentrations were determined with a UNICAM 929 atomic absorption spectrophotometer (Unicam Ltd.).
Chlorophyll Measurements
Leaves from control and salt-treated plants were collected, weighed fresh, washed in distilled water, and extracted in 80% (v/v) acetone. Chlorophyll content was measured according to Porra et al. (1989).
Photosynthetic Parameters
Photosynthetic parameters were measured as described previously (Golding and Johnson, 2003; Hald et al., 2008). Briefly, gas exchange was monitored using a CIRAS1 infrared gas analyzer (PP Systems). Leaves illuminated for 40 min, to achieve steady-state conditions. Leaf temperature (25°C ± 1°C) and CO2 concentration within the chamber were controlled by the gas analyzer. Chlorophyll fluorescence emission was measured using a PAM-101 chlorophyll fluorimeter (Walz). Fluorescence parameters were calculated as described by Maxwell and Johnson (2000). The redox state of the PSI primary donor, P700, was determined using a Walz PAM 101 fluorometer in combination with an ED-P700DW-E emitter-detector unit (Walz). Light was provided by Volpi Intralux lamps, except for measurements in Figure 7 that used a Luxeon LXHL-PD09 LED (Phillips-Lumiled) in a laboratory-built lamp. Calibration of maximum P700 signal size was conducted using a high-power LED array (LED 735-66-60 Roithner Laser). Two percent and 21% O2 gas were supplied by mixing compressed oxygen and nitrogen from cylinders (BOC Gases) using an MKS controller (MKS Instruments Inc.).
Measurements of PTOX Activity and Protein
To estimate the contribution of the PTOX to overall PSII electron transport, the leaves of control and salt-treated plants were vacuum infiltrated with either water or with 1 mm nPG (3,4,5-trihydroxy-benzoic acid-n-propyl ester; Sigma) or 35 μm DNP-INT (kindly provided by Dr. Anja Krieger-Liszkay, CE-Saclay, France).
For immunoblot analysis, thylakoids were isolated as described by Cerovic and Plesnicar (1984). Thylakoid proteins were extracted from membranes in 125 mm TRIS-HCl, pH 6.8, 20% glycerol, 4% (w/v) SDS, 5% (v/v) β-mercaptoethanol, 0.1% (w/v) bromphenol blue. Protein concentration was estimated using a Bio-Rad protein assay kit (Bio-Rad Laboratories). Immunoblotting was carried out as described by Mudd et al. (2008). Polyclonal antibodies against PTOX were kindly provided by Dr. M. Kuntz (Université Joseph Fourier, Grenoble, France).
Acknowledgments
We would like to thank to Dr. M. Kuntz for kindly providing us with polyclonal antibodies against PTOX and Dr. Anja Krieger-Liszkay for DNP-INT. We would like to thanks Drs. Simon Hald, Rachel Webster, and Panagiotis Madesis for help and advice with experimental procedures, and Mr. John Simpson for help building instrumentation.
Footnotes
-
The author responsible for the distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Giles N. Johnson (giles.johnson{at}manchester.ac.uk).
-
↵1 This work was supported by a Marie-Curie Fellowship of the European Commission (grant no. MEIF–CT–2006–040053 to P.S.).
-
↵[C] Some figures in this article are displayed in color online but in black and white in the print edition.
-
↵[OA] Open access articles can be viewed online without a subscription.
- Received November 10, 2008.
- Accepted November 26, 2008.
- Published December 3, 2008.