The Arabidopsis thaliana ascorbate-deficient vtc-1 mutant has only 30% ascorbate contents of the wild type (WT). This ascorbate-deficient mutant was used here to study the physiological roles of ascorbate under salt stress in vivo. Salt stress resulted in a more significant decrease in CO2 assimilatory capacity in the vtc-1 mutant than in the WT. Photosystem II function in the Arabidopsis vtc-1 mutant also showed an increased sensitivity to salt stress. Oxidative stress, indicated by the hydrogen peroxide content, increased more dramatically in the vtc-1 mutant than in the WT under salt stress. To clarify the reason for the increased oxidative stress in the vtc-1 mutant, the contents of small antioxidant compounds and the activities of several antioxidant enzymes in the ascorbate–glutathione cycle were measured. Despite an elevated glutathione pool in the vtc-1 mutant, the ascorbate contents and the reduced form of ascorbate decreased very rapidly under salt stress. These results showed that the activities of MDAR and DHAR were lower in the vtc-1 mutant than in the WT under salt stress. Thus, low intrinsic ascorbate and an impaired ascorbate–glutathione cycle in the vtc-1 mutant under salt stress probably induced a dramatic decrease in the reduced form of ascorbate, which resulted in both enhanced ROS contents and decreased NPQ in the vtc-1 mutant.
Ascorbate is an abundant small molecule in plants. Ascorbate is a key substance in the network of antioxidants that include ascorbate, glutathione, α-tocopherol, and a series of antioxidant enzymes. Ascorbate has also been shown to play multiple roles in plant growth, such as in cell division, cell wall expansion, and other developmental processes (Arrigoni, 1994; Smirnoff, 1996; Asada, 1999; Conklin, 2001; Pignocchi and Foyer, 2003).
Many stress conditions, such as drought, salt, extreme temperatures, nutrient deprivation, UV-B radiation, and air pollutants can cause an increase of reactive oxygen species (ROS). Ascorbate, as an antioxidant, detoxifies H2O2, which is formed by the dismutation of
A mutant of Arabidopsis thaliana deficient in ascorbate (vtc-1 mutant) was isolated via its sensitivity to ozone exposure and has 30–60% ascorbate contents of the WT (Conklin et al., 1996). A point mutation occurs in vtc-1's gene of GDP-mannose pyrophosphorylase, a key enzyme in the ascorbate biosynthesis pathway. The mutation results in the considerable decrease in the activity of GDP-mannose pyrophosphorylase (Conklin et al., 1999), which controls ascorbate biosynthesis pathway flux (Keller et al., 1999). Previous work has shown that photosynthesis and the oxidative system are not perturbed in the vtc-1 mutant under normal growth conditions, although the vtc-1 mutant is smaller than the WT and shows retarded flowering and accelerated senescence (Veljovic-Jovanovic et al., 2001).
Salinity stress may result in the accumulation of ROS (Hernández et al., 1995). It has also been reported that exogenous ascorbate can increase resistance to salt stress and reduce oxidative stress (Shalata and Neumann, 2001). In response to a combination of high light intensity and salt stress, ascorbate-deficient mutants of Arabidopsis thaliana showed increased sensitivity to NaCl-induced photo-oxidation (Smirnoff, 2000). Arabidopsis ascorbate-deficient vtc-1 mutants were used to study the physiological roles of ascorbate under short-term salt stress in vivo.
Materials and methods
Plant materials and growth conditions
Arabidopsis thaliana ecotype Col-0, the background of the ascorbate-deficient mutant vtc-1, was used as the WT in this study. Seedlings of A. thaliana WT and the vtc-1 mutant were grown in a culture room at 22 °C with 12 h photoperiod under a photon flux density of 150 μmol photons m−2 s−1. Plastic pots containing 300 cm3 peat were used in the experiments. NaCl was dissolved in half-strength Hoagland nutrient solution and the plants were watered to drip with approximately 150 ml of 200 mM NaCl solution after the seedlings were grown for 5 weeks. The fully expanded leaves were collected for the experiments.
For the ascorbate feeding experiment, 20 mM ascorbate was dissolved in half-strength Hoagland solution to water Arabidopsis 1 week before salt treatment and during the salt treatment experiment.
Determination of H2O2 contents
H2O2 contents were determined by the peroxidase-coupled assay according to Veljovic-Jovanovic et al. (2002) modified from Okuda et al. (1991). 0.1 g of Arabidopsis leaves were ground in liquid nitrogen and the powder was extracted in 2 ml 1 M HClO4 in the presence of insoluble PVP (5%). The homogenate was centrifuged at 12 000 g for 10 min and the supernatant was neutralized with 5 M K2CO3 to pH 5.6 in the presence of 100 μl 0.3 M phosphate buffer (pH 5.6). The solution was centrifuged at 12 000 g for 1 min and the sample was incubated for 10 min with 1 U ascorbate oxidase (Sigma, St Louis, USA) to oxidize ascorbate prior to the assay. The reaction mixture consisted of 0.1 M phosphate buffer (pH 6.5), 3.3 mM DMAB (Sigma, St Louis, USA), 0.07 mM MBTH (Sigma, St Louis, USA), and 0.3 U POX (Sigma, St Louis, USA). The reaction was initiated by the addition of 200 μl of the sample. The absorbance change at 590 nm was monitored at 25 °C.
Ascorbate content determination
Ascorbate contents were determined according to the method as described previously (Foyer et al., 1983) with some modifications. 0.1 g of Arabidopsis leaves were ground in liquid nitrogen and 1 ml of 2.5 M perchloric acid was added. The crude extract was centrifuged at 2 °C for 10 min at 10 000 g, and the supernatant was neutralized with saturated Na2CO3 using methyl orange as the indicator. The reduced ascorbate was assayed spectrophotometrically at 265 nm in 1 M NaH2PO4 buffer, pH 5.6, with 1 U ascorbate oxidase. The total ascorbate was assayed after incubation in the presence of 30 mM DTT.
Glutathione content determination
Total glutathione contents were measured as described by Anderson et al. (1992) except that 7% sulphosalicylic acid was used as the extraction buffer. 0.2 g of Arabidopsis leaves were ground in liquid nitrogen and 1 ml of the extraction buffer was added. The crude extract was centrifuged at 2 °C for 10 min at 10 000 g. 400 μl reagent 1 (110 mM Na2HPO4, 40 mM NaH2PO4, 15 mM EDTA, 0.3 mM 5,5′-dithiobis(2-nitrobenzoic acid), and 0.04% BSA), 320 μl reagent 2 (1 mM EDTA, 50 mM imidazole solution, and 0.02% BSA), 320 μl 5% Na2HPO4 (pH 7.5), 1.5 U glutathione reductase, 80 μl extract, and 80 μl NADPHNa4 were mixed. The reaction mixture was measured at 412 nm. Oxidized glutathione was measured as total glutathione except that 1 ml extract was incubated with 40 μl 2-vinylpyridine for 1 h at 25 °C before mixing.
Ascorbate–glutathione cycle enzymes activity determination
The leaves were homogenized in 50 mM phosphate buffer (pH 7.0), 0.1 mM EDTA, 0.1% phenylmethylsulphonyl fluoride, 1% polyvinylpyrrolidone, 0.1% Triton X-100, and 1 mM ascorbate. The homogenate was centrifuged at 12 000 g for 15 min at 4 °C. The extract was stored at −70 °C or used in enzyme assays immediately. Glutathione reductase (GR) and dehydroascorbate reductase (DHAR) activities were determined according to the method of Foyer and Halliwell (1976). 860 μl 1 mM oxidized glutathione, 100 μl 2 mM NADPH, and 40 μl crude enzyme were mixed and GR activity was measured at 340 nm. 700 μl phosphate buffer (pH 7.0), 700 μl 20 mM reduced glutathione, 100 μl 2 mM dehydroascorbate, and 100 μl crude enzyme were mixed and DHAR activity was measured at 265 nm. Monodehydroascorbate reductase (MDAR) was measured according to Hossain and Asada (1984). 900 μl 2 mM ascorbate, 2 U ascorbate oxidase, 30 μl 2 mM NADPH, and 30 μl crude enzyme were mixed and measured at 340 nm. All the spectrophotometrical assays were determined with DU series 640 spectrophotometer (Beckman Coulter, Inc., Fullerton, CA, USA).
Plant chlorophyll fluorescence was determined by a portable fluorometer (PAM-2000, Walz, Effeltrich, Germany) connected with a leaf-clip holder (2030-B, Walz) and with a trifurcated fibre-optic (2010-F, Walz). Data acquisition software (DA-2000, Walz) was used in a notebook computer to dispose data on-line. The measurements were followed essentially according to Lu and Zhang (1998, 2000).
Before measurement, the leaves were dark-adapted for 30 min. The minimal fluorescence level in the dark-adapted state (F0) was measured using the measuring light which is sufficiently low (0.8 μmol m−2 s−1) so as not to induce notable variable fluorescence. Far-red light (5 μmol m−2 s−1) was adopted to oxidize the PSII fully before measurement of the minimal fluorescence during illumination (
The maximum photochemical efficiency of PSII was determined from the ratio of variable (Fv) to maximum (Fm) fluorescence (Fv/Fm=(Fm−Fo)/Fm) (Kitajima and Butler, 1975). The efficiency of excitation capture by open PSII centres was calculated as
The LI-6400 portable photosynthesis system with a LI-6250 CO2 analyser (Li-Cor, Inc., Lincoln, Nebraska, USA) was used in photosynthetic analysis. The apparent photosynthetic rate (Photo.), stomatal conductance (gs), and intercellular CO2 concentration (Ci) of the fully expanded leaves were measured according to the manual of the instrument.
Relative water contents (RWC)
RWC was measured as described in Munné-Bosch and Alegre (2002a). Dry material was obtained after being heated at 80 °C for 48 h. RWC was calculated using the formulae: RWC (%)=(FW–DW)/FW where FW is the fresh weight and DW is the dry weight.
Chlorophyll content determination
Chlorophyll was extracted and measured as described previously by Porra et al. (1989).
Effects of salt stress on relative water contents and chlorophyll contents
RWC is approximately the same in both the WT and the vtc-1 mutant. Under salt stress, RWC in both the WT and the vtc-1 mutant decreased progressively (Fig. 1A). After 48 h treatment, RWC was about 76% in the WT, and 63% in the mutant.
Figure 1 also shows the changes in chlorophyll contents on the basis of leaf area (B) and the ratio of chlorophyll a to b (C) in the WT and vtc-1 mutant under salt stress. Under normal conditions, chlorophyll contents and the ratio of chlorophyll a to b were slightly higher in the vtc-1 mutant than those in the WT, which is consistent with other reports (Veljovic-Jovanovic et al., 2001; Munné-Bosch and Alegre, 2002b). After NaCl treatment for 48 h, chlorophyll contents decreased by approximately 26% in the WT and by 53% in the vtc-1 mutant, and the ratio of chlorophyll a to b decreased by 8.5% in the WT and by 21.8% in the mutant. The chlorophyll contents remained almost constant in the WT and the vtc-1 mutant without salt treatment (Fig. 1D), which indicated that the vtc-1 mutant leaves used in the experiment were not at the senescent stages.
Effects of salt stress on CO2 assimilation and stomatal conductance
The photosynthesis rate, which was measured by the CO2 assimilatory capacity, changed from 9.1 to 7.0 μmol m−2 s−1 in the WT, and from 8.0 to 4.3 μmol m−2 s−1 in the vtc-1 mutant after NaCl treatment for 48 h (Fig. 2A).
The stomatal conductance (gs) in the vtc-1 mutant was slightly higher than that in the WT under normal growth condition (Fig. 2B). After the first 12 h NaCl treatment, it decreased by approximately 58% in the WT and by 60% in the vtc-1 mutant. After 48 h treatment, gs was about 12% of the control in both the WT and vtc-1 mutant (Fig. 2B).
The intercellular CO2 concentrations (Ci) were slightly higher in the vtc-1 mutant than in the WT under normal conditions and during NaCl treatment (Fig. 2C). A similar decreasing trend was found in both the WT and the vtc-1 mutant during NaCl treatment, from about 250 to 100 μmol CO2 mol−1 and from 285 to 125 μmol CO2 mol−1 in the WT and the vtc-1 mutant, respectively.
Effects of salt stress on PSII photochemical activities
Under normal conditions, the maximal efficiency of PSII photochemistry (Fv/Fm) is approximately similar (0.84) in the WT and the vtc-1 mutant. The maximal efficiency of PSII decreased in both the WT and the vtc-1 mutant for the duration of treatment, and the decreasing extent was greater in the vtc-1 mutant than in the WT. After 48 h treatment, Fv/Fm decreased from 0.83 to 0.79 in the WT, and from 0.84 to 0.52 in the vtc-1 mutant (Fig. 3A).
The responses of the efficiency of excitation energy capture by open PSII reaction centres (
In the vtc-1 mutant, mannose metabolism may also be affected (Lukowitz et al., 2001). Conklin et al. reported that ascorbate feeding did elevate ascorbate in the vtc-1 mutant (Conklin et al., 1996). To confirm that the decreased photochemical activities in the vtc-1 mutant were due to ascorbate deficiency, ascorbate feeding experiments were performed. When the leaves were treated with ascorbate, the decreased Fv/Fm ratio in the vtc-1 mutant after treatment with NaCl for 48 h was restored to the control value (Fig. 4).
Changes of H2O2 contents and small antioxidant molecules under salt stress
The H2O2 contents were similar both in the WT and the vtc-1 mutant without salt treatment, while the H2O2 contents increased under salt stress, especially in the vtc-1 mutant. The H2O2 contents increased by approximately 18% in the WT, while it increased by about 150% in the vtc-1 mutant after 48 h salt treatment (Fig. 5A).
As an ascorbate-deficiency mutant, the ascorbate contents in the vtc-1 mutant were about 30% of the WT in 5-week-old fully extended leaves under normal conditions in this study (about 17.3 μmol g−1 DW in the WT and 5.0 μmol g−1 DW in the vtc-1 mutant). After salt stress for 12 h, ascorbate contents increased by about 54% and 27% in the WT and the vtc-1 mutant, respectively. However, they decreased afterwards and after 48 h treatment they were approximately 77% and 15% of the control in the WT and the vtc-1 mutant, respectively (Fig. 5B). The ratio of reduced to total ascorbate decreased progressively under salt stress, from 95% to 45% in the WT, and from 95% to 9% in the vtc-1 mutant (Fig. 5C).
The total glutathione contents were higher in the vtc-1 mutant under normal conditions, which is consistent with previous reports (Veljovic-Jovanovic et al., 2001). During the NaCl treatment, the glutathione contents increased during the 24 h salt treatment by about 54% in the WT and 73% in the vtc-1 mutant, respectively. After 48 h treatment, it showed a greater increase in the WT than in the vtc-1 mutant (Fig. 5D). The ratio of reduced to total glutathione remained almost constant, with about 95% in the WT and the vtc-1 mutant.
Changes of ascorbate–glutathione cycle reducing enzymes activity under salt stress
To elucidate the mechanism of redox balance maintenance, the activities of enzymes in the ascorbate–glutathione cycle were measured. GR is a thiol enzyme which uses NADPH as an electron donor and reduces oxidized glutathione (GSSG) to the reduced form. GR activity increased by 70% and 50% in the WT and the vtc-1 mutant during the 48 h salt treatment, respectively (Fig. 6A).
Using NADPH as an electron donor, MDAR catalyses monodehydroascorbate to dehydroascorbate and ascorbate. A slight increase was observed during the first 12 h treatment in the WT, and the activity reached the control level after 48 h treatment, whereas the activity of MDAR decreased during the salt treatment in the vtc-1 mutant (Fig. 6B).
Reduced glutathione and dehydroascorbate are the substrates of DHAR. It is an important enzyme to reduce ascorbate. DHAR activities were almost the same in the WT and vtc-1 mutant under normal conditions. In WT, it increased by 70% during 12 h treatment, and decreased after 24 h treatment. In the vtc-1 mutant, DHAR activity levels were almost constant during the 24 h salt treatment. After 48 h treatment, the DHAR activity decreased by 60% (Fig. 6C).
The Arabidopsis vtc-1 mutant was used to study the physiological roles of ascorbate under salt stress in vivo. CO2 assimilation rate and PSII function in the Arabidopsis vtc-1 mutant showed increased sensitivity to salt stress. More oxidative stress occurred in the vtc-1 mutant under salt stress than in the WT. Under salt stress, the metabolism of the ascorbate–glutathione cycle seems to be impaired in the vtc-1 mutant, most probably due to the reduced ascorbate contents.
Decrease of PSII function exacerbated under salt stress in the vtc-1 mutant
Photosynthetic CO2 exchange and chlorophyll a fluorescence in vivo were measured to investigate the effects of salt stress on photosynthetic functions in the ascorbate-deficient vtc-1 mutant. The photosynthesis rate, reflected by the CO2 assimilation capacity, decreased much more in the vtc-1 mutant than in the WT under salt stress (Fig. 2A). The slightly higher values of stomatal conductance and intercellular CO2 in the vtc-1 mutant than that in WT under salt stress (Fig. 2B, C) indicated that the decrease of CO2 assimilation was not caused by CO2 limitation in plant cells. Photosynthetic CO2 assimilation is considered to be a major sink for reducing equivalents (ATP and NADPH) generated by the primary photochemical reaction. The normal recycling of NADPH is especially important for the function of NADP+ as the terminal electron acceptor and maintains the photochemical de-excitation of reaction centres in a steady-state. In addition, the proton motive force through the coupling of electron transport to ATP synthesis regulates the amplitude of the photosynthetic electron transport by feedback inhibition (Krause, 1994). This serves as a dissipation mechanism for excess excitation energy when the rate of ATP and NADPH synthesis exceeds the demand for CO2 fixation. Thus, the dramatic decrease in CO2 assimilation would result in the decrease of PSII function in the vtc-1 mutant under salt stress through a feedback system.
The PSII activity, reflected by the maximum efficiency of PSII photochemistry measured as Fv/Fm, decreased much more in the vtc-1 mutant than that in WT under salt stress (Fig. 3A). In order to examine the possible changes in PSII photochemistry under normal light irradiation, the fluorescence characteristics during the steady-state of photosynthesis was investigated. Photochemical quenching (qP) decreased much more in the vtc-1 mutant than in the WT (Fig. 3C), which indicates a significant increase in the proportion of closed PSII reaction centres or the proportion of the reduced state of QA (Dietz et al., 1985; Genty et al., 1989). An increase in the fraction of QA in the reduced state estimated from the decreased qP in the vtc-1 mutant under salt stress suggests a decrease in the proportion of available excitation energy used for photochemistry (Havaux et al., 1991) and an increase in the excitation pressure on PSII under the steady-state of photosynthesis (Öquist and Huner, 1993). Such an increase in excitation pressure would result in further damage to PSII if excess excitation pressure was not dissipated safely since the excitation pressure on PSII has been shown to be a determining factor for photodamage on PSII (Demmig-Adams and Adams, 1992; Chow, 1994). In this study, NPQ decreased dramatically in the vtc-1 mutant under salt stress (Fig. 3E), whereas it remained relatively constant in the WT. This result suggested defective thermal dissipation in the vtc-1 mutant. As a result, the efficiency of excitation energy capture by open PSII reaction centres (
Increased oxidative stress due to impaired ascorbate–glutathione cycle in the vtc-1 mutant under salt stress
Under 200 mM NaCl treatment, the decrease in chlorophyll contents and the ratio of chlorophyll a to b accelerated in the vtc-1 mutant compared with those in the WT (Fig. 1B, C). This result indicates that the vtc-1 mutant suffers more damage from salt stress than the WT.
Ascorbate plays a central role in photosynthetic protection, for it functions in ROS scavenging in the Mehler peroxidase reaction as a reductant and in excess light energy dissipation in NPQ (Eskling et al., 1997; Noctor and Foyer, 1998; Asada, 1999; Müller-Moule et al., 2002, 2003, 2004). The accumulation of more H2O2 in the vtc-1 mutant than in the WT during salt treatment indicated that more oxidative stress takes place in the vtc-1 mutant under salt stress (Fig. 5A). There was about 30% ascorbate compared with the WT in 5-week-old fully extended leaves of the vtc-1 mutant under normal conditions without an increase of ROS and NPQ, which is consistent with other reports (Veljovic-Jovanovic et al., 2001; Munné-Bosch and Alegre, 2002b). This result suggests that 30% ascorbate was sufficient to maintain normal plant physiological activity under normal conditions. However, under salt stress the contents of ascorbate and the ratio of reduced to total ascorbate declined more dramatically in the vtc-1 mutant than that in the WT (Fig. 5B, C), which indicates that the regeneration of reduced ascorbate is impaired in the vtc-1 mutant.
The ascorbate–glutathione cycle is very efficient in regenerating the reduced forms of ascorbate. In this cycle monodehyroascorbate radical and DHA are reduced to ascorbate by NAD(P)-dependent MDAR and glutathione-dependent DHAR, respectively. The oxidative form of glutathione (GSSG) is reduced by GR (Noctor and Foyer, 1998). Under salt stress the activity of GR increased and the glutathione contents in the vtc-1 mutant accumulated even more than in the WT, while the redox states of glutathione remained almost constant in both the WT and the vtc-1 mutant (Fig. 5D, 6A). These results are consistent with previous reports that glutathione synthesis was elevated in plants under environmental stress (Foyer et al., 1997; Ruiz and Blumwald, 2002). Thus, the changes in glutathione metabolism obviously may not account for the ascorbate decrease in the vtc-1 mutant.
The activities of MDAR and DHAR, two ascorbate reducing enzymes in the ascorbate–glutathione cycle, are crucial to regenerate ascorbate. Overexpression of the DHAR gene increased the ascorbate contents in transgenic plants, which demonstrates that the ascorbate contents of plants can be elevated through enhanced ascorbate recycling (Chen et al., 2003). DHAR overproducing transgenic plants also showed enhanced oxidative resistant ability (Kwon et al., 2003). The transcripts of the MDAR and DHAR genes are induced by oxidative stress so as to meet the requirement for the regeneration of reduced ascorbate upon increased oxidative stress (Grantz et al., 1995; Chew et al., 2003). These results have shown that the activities of MDAR and DHAR were lower in the vtc-1 mutant than in the WT under salt stress. They do not meet the requirement for a higher ascorbate reducing capability in the vtc-1 mutant, which induced the inhibition of the ascorbate–glutathione cycle under salt stress. That both the total ascorbate contents and the ratio of reduced to total ascorbate decreased dramatically in the vtc-1 mutant under salt stress may be partially due to the blocked ascorbate recycling.
Thus, the ascorbate-deficiency in the vtc-1 mutant seems to be the main reason for the sensitivity of the vtc-1 mutant to salt stress. Low intrinsic ascorbate and an impaired ascorbate–glutathione cycle in the vtc-1 mutant under salt stress probably induced an excessive decrease of the reduced form of ascorbate, which induced both enhanced ROS contents and decreased NPQ in the vtc-1 mutant. The dramatic decrease of NPQ in the vtc-1 mutant, a process requiring ascorbate as a cofactor of violaxanthin de-epoxidase in the xanthophyll cycle, could also provide an explanation. The aggravated oxidative stress and deficient excess energy dissipation are the considerable sources of the exacerbated decrease of PSII function in the vtc-1 mutant. The results above demonstrate the important role of ascorbate in a physiologically protective function under salt stress in vivo.
This work was supported by the One Hundred Talent Project, Excellent PhD Thesis Foundation (199924), Key Natural Science Foundation of Gansu Province (ZS031-A25-034-D). We are grateful to Nottingham Stock Centre for the Arabidopsis seeds.