Enhancement of stress tolerance in transgenic tobacco plants overexpressing Chlamydomonas glutathione peroxidase in chloroplasts or cytosol

Authors


For correspondence (fax +81 742 43 2252; e-mail shigeoka@nara.kindai.ac.jp).

Summary

To evaluate the physiological potential of the defense system against hydroperoxidation of membrane-lipid components caused by environmental stresses in higher plants, we generated transgenic tobacco plants expressing a glutathione peroxidase (GPX)-like protein in the cytosol (TcGPX) or chloroplasts (TpGPX). The activities toward α-linolenic acid hydroperoxide in TcGPX and TpGPX plants were 47.5–75.3 and 32.7–42.1 nmol min−1 mg−1 protein, respectively, while no activity was detected in wild-type plants. The transgenic plants showed increased tolerance to oxidative stress caused by application of methylviologen (MV: 50 µm) under moderate light intensity (200 µE m−2 sec−1), chilling stress under high light intensity (4°C, 1000 µE m−2 sec−1), or salt stress (250 mm NaCl). Under these stresses, the lipid hydroperoxidation (the production of malondialdehyde (MDA)) of the leaves of TcGPX and TpGPX plants was clearly suppressed compared with that of wild-type plants. Furthermore, the capacity of the photosynthetic and antioxidative systems in the transgenic plants remained higher than those of wild-type plants under chilling or salt stress. These results clearly indicate that a high level of GPX-like protein in tobacco plants functions to remove unsaturated fatty acid hydroperoxides generated in cellular membranes under stress conditions, leading to the maintenance of membrane integrity and increased tolerance to oxidative stress caused by various stress conditions.

Introduction

In plant cells, many metabolic reactions, such as photosynthesis, photorespiration, and respiration, involve the formation of active oxygen species (AOS), such as superoxide radical (O2), H2O2, and hydroxyl radical (OH) (Asada, 1997; Foyer et al., 1994a). Furthermore, a wide range of environmental stresses such as drought, high salinity, and low temperature cause enhanced production of AOS. Thus, much of the injury to plants imposed by stress exposure is associated with oxidative damage at the cellular level, including lipid hydroperoxidation leading to membrane damage (Alscher et al., 1997; Bowler et al., 1992; Foyer et al., 1994b; Shigeoka et al., 2002). Therefore, higher plants have evolved antioxidative systems consisting of antioxidants and antioxidant enzymes such as superoxide dismutase (SOD), ascorbate peroxidase (APX; EC 1.11.1.11), catalase, glutathione peroxidase (GPX, EC 1.11.1.9), and glutathione S-transferase (GST) to interrupt the cascades of uncontrolled oxidation in some organelles (Noctor and Foyer, 1998).

GPX is the principal cellular enzyme capable of membrane lipid peroxidation repair and is generally considered to be the main line of enzymatic defense against oxidative membrane damage (Kühn and Borchert, 2002; Ursini et al., 1995). GPXs constitute a family of enzymes that are capable of reducing a variety of organic and inorganic hydroperoxides to the corresponding hydroxyl compounds by utilizing GSH and/or other reducing equivalents. At least four selenium-dependent forms of GPX have been identified so far and shown to differ with respect to structure, substrate specificity, and tissue distribution (Kühn and Borchert, 2002; Ursini et al., 1995). Among them, phospholipid hydroperoxide glutathione peroxidase (PHGPX) reduces the complex hydroperoxyester lipids incorporated in biomembranes and lipoproteins, suggesting that it may play a significant role in protecting membranes from oxidative damage (Brigelius-Flohéet al., 1994; Ursini and Bindoli, 1987). In higher plants, cDNAs with significant sequence homology to animal PHGPX have been isolated and characterized from photosynthetic organisms such as tobacco (Criqui et al., 1992), Citrus sinensis (Holland et al., 1993), Arabidopsis (Sugimoto and Sakamoto, 1997), sunflower (Roeckel-Drevet et al., 1998), tomato (Depège et al., 1998), and pea (Mullineaux et al., 1998). The mRNA and protein levels of GPXs were increased in response to many types of stress treatment (Agrawal et al., 2001; Avsian-Kretchmer et al., 1999; Depége et al., 2000; Roeckel-Drevet et al., 1998; Sugimoto and Sakamoto, 1997). However, GPXs in higher plants had lower activity than those in animals because the GPX genes of these plants encoded a cysteine (Cys) residue at the putative catalytic site instead of the selenocysteine (Sec) of animal GPXs. This low activity has made it difficult to clarify the potential physiological role of GPX in higher plants.

Recently, we used a simple screening method to isolate and characterize a cDNA encoding an antistress selenium-independent GPX-like protein from the halotolerant Chlamydomonas W80 strain (C. W80; Miyasaka et al., 2000; Takeda et al., 2003). The amino acid sequence deduced from the cDNA sequence showed 40–63 and 37–46% homology to the sequences of GPXs from higher plants and animals, respectively, and contained a Cys residue instead of a Sec at the catalytic site. Interestingly, the native and recombinant GPX-like proteins showed activity toward unsaturated fatty acid hydroperoxides, but toward neither H2O2 nor phospholipid hydroperoxide.

Many analyses of transgenic plants in which the levels of antioxidant enzymes have been manipulated through gene transfer technology have provided significant insights into the roles of these enzymes in higher plants. However, only a few studies of the scavenging systems of lipid hydroperoxide in higher plants have been reported. Transgenic tobacco plants expressing a GST with GPX activity had elevated levels of GSH and ascorbate (AsA) and monodehydroascorbate reductase (MDAR) activity compared to wild-type plants, as well as enhanced seedling growth under thermal and salt-stress conditions that caused an increased level of lipid hydroperoxidation (Roxas et al., 1997). Although preliminary, these results suggested that alterations in the expression levels of enzymes involved in lipid hydroperoxide-scavenging systems might provide a strategy for developing transgenic plants with increased tolerance to the oxidative membrane damage caused by a variety of stressful conditions.

In this study, we generated transgenic tobacco plants expressing the C. W80 GPX-like protein in chloroplasts (TpGPX) or cytosol (TcGPX), and demonstrated suppressed production of lipid hydroperoxide and increased tolerance to oxidative stress caused by methylviologen (MV) application or chilling in these plants. These results indicate the potential physiological role of C. W80 GPX-like proteins in higher plants.

Results

Expression of C. W80 GPX-like protein in cytosol or chloroplasts of tobacco plants

A transgene consisting of the cDNA encoding C. W80 GPX-like protein under the control of the cauliflower mosaic virus (CaMV) 35S promoter was constructed for cytosol-targeted expression (Figure 1a). In addition, for chloroplast-targeted expression, the transit peptide sequence of RuBisCO small subunit from tomato was translationally fused adjacent to the first methionine of the GPX cDNA in the transgene. Tobacco plants (Nicotiana tabacum cv. Xanthi) were used as hosts for the infection of Agrobacterium transformed with the transgenes. Five and four independent transformed lines for cytosol- and chloroplast-targeted expression, respectively, were produced thereby. The genomic DNAs were isolated from wild-type plants and all transgenic lines and analyzed by PCR using a C. W80 GPX-like protein-specific primer set. The expected 579-bp fragment was amplified from the DNAs of all transgenic lines tested, but not from those of wild-type plants (Figure 1b). No difference was observed in growth or morphology between wild-type and either type of transgenic plant. Some of the original transgenic lines were self-pollinated to produce homozygous T2 progeny (cytosol-targeted GPX expression lines: TcGPX-1, TcGPX-7, TcGPX-9, and TcGPX-13; chloroplast-targeted expression lines: TpGPX-1, TpGPX-6, and TpGPX-14).

Figure 1.

Creation of the transgenic plants.

(a) Construction of C. W80 GPX-like protein expression vector used to transform tobacco plants. The drawing is not to scale. Arrows indicate the locations of primers as used in (b).

(b) Detection of the transgene encoding C. W80 GPX-like protein in transformed tobacco plants. Genomic DNAs were isolated from leaves of wild-type (WT) and transgenic plants (100 ng each) and were used to amplify the transgene by PCR using primers as shown in (a). The amplified bands were photographed under UV light after electrophoresis.

(c) Detection of C. W80 GPX-like protein in transformed tobacco plants. Total proteins (25 µg each) from the leaves of WT and transgenic plants were subjected to SDS–PAGE followed by Western blot analysis with a mouse antiserum against C. W80 GPX-like protein. P, recombinant C. W80 GPX-like protein as a positive control.

(d) Localization of C. W80 GPX-like protein in transformed tobacco plants. Intact chloroplasts (36 µg protein each) were isolated as described in the Experimental procedures section from the leaves of WT and transgenic plants and then were subjected to SDS–PAGE followed by Western blot analysis as described in (c).

The levels of transcripts and protein and the activities of GPX-like protein were assayed using the third leaves from the top in these transgenic plants grown for 7 weeks under normal conditions. Northern blotting using 32P-labeled-GPX-like protein cDNA as a probe showed that transcripts derived from the transgene were highly expressed in some transgenic lines (TcGPX-7 and TcGPX-9; TpGPX-6 and TpGPX-14), but not in wild-type plants (data not shown). In Western blot analysis using anti-CWGPX, high levels of GPX-like protein derived from the transgene (molecular mass 18.5 kDa) were detected in the extracts prepared from the leaves of transgenic plants (Figure 1c). Furthermore, we isolated intact chloroplasts from leaves of wild-type and both types of the transgenic plants by a modification of the method described previously by Takeda et al. (1995). The C. W80 GPX-like protein was detected in the intact chloroplasts in TpGPX lines, but not in those from wild-type and TcGPX lines (Figure 1d). The detected molecular mass corresponded to that of the mature GPX-like protein from which the transit peptide derived from the RuBisCO small subunit had been removed. The specific activity (187.8 ± 22.5 nmol min−1 mg−1 protein) of C. W80 GPX-like protein toward α-linolenic acid hydroperoxide was detected in the stromal fraction from TpGPX-14 plant. These results clearly indicated that the C. W80 GPX-like protein was imported into chloroplasts in the TpGPX lines.

The GPX activities in the transgenic plants were assayed using α-linolenic acid hydroperoxide, cumene hydroperoxide, and phospholipid hydroperoxide as substrates (Table 1). GPX activities toward cumene hydroperoxide in TcGPX and TpGPX lines were 2.4–3.3- and 2.4–2.7-fold, respectively, higher than those of wild-type plants, while the total activities toward phospholipid hydroperoxide, including the activity derived from endogenous plant GPX, were almost equal in all of the transgenic and wild-type plants. No differences in the activity of GST, which has GPX activity toward cumene hydroperoxide as a substrate (Roxas et al., 2000), were observed between the transgenic and wild-type plants. The activities toward α-linolenic acid hydroperoxide in TcGPX and TpGPX plants were 47.5–75.3 and 32.7–42.1 nmol min−1 mg−1 protein, respectively. These results clearly indicate that the C. W80 GPX-like protein is expressed in TcGPX and TpGPX lines and functions as the native form.

Table 1.  Activities of GPX and GST of wild-type, TcGPX, and TpGPX plants
 GPXGST
α-Linolenic acid hydroperoxideCumene hydroperoxidePhospholipid hydroperoxide
  • *

    Indicates that the mean values are significantly different from those of wild-type plants (P < 0.05).

  • All values were expressed in nmol min−1 mg−1 protein. Third leaf of wild-type and transgenic plants was analyzed. Data are mean ± SD (n = 3).

Wild type023.1 ± 2.99.5 ± 1.2230.4 ± 14.3
TcGPX-175.2 ± 8.1*52.6 ± 5.0*9.5 ± 1.1246.8 ± 15.1
TcGPX-747.5 ± 5.1*40.6 ± 4.6*9.4 ± 1.1243.3 ± 16.0
TcGPX-950.2 ± 6.3*41.8 ± 4.6*10.2 ± 1.2237.1 ± 15.0
TcGPX-1375.3 ± 7.9*54.0 ± 6.2*9.8 ± 1.0252.3 ± 14.5
TpGPX-132.7 ± 4.0*34.5 ± 3.7*9.2 ± 0.9236.7 ± 13.8
TpGPX-633.4 ± 3.9*35.0 ± 3.6*9.4 ± 1.1243.0 ± 14.1
TpGPX-1442.1 ± 4.3*39.1 ± 4.1*9.5 ± 1.1245.1 ± 13.2

The CO2 fixation, photosystem II (PSII) activity (variable fluorescence (Fv)/maximal fluorescence (Fm)), antioxidant levels (AsA, dehydroascorbate (DAsA), GSH, and oxidized GSH (GSSG)), and activities of SOD, APX, and the AsA-regenerating enzymes (MDAR, dehydroascorbate reductase (DHAR), and GSH reductase (GR)) in TcGPX-13 and TpGPX-14 plants, which had the highest levels of GPX activity toward α-linolenic acid hydroperoxide of all the lines, were compared to those of wild-type plants (Table 2). No significant differences in any of these parameters were detected between wild-type and transgenic plants. The parameters in the other lines of TcGPX and TpGPX plants also showed similar values (data not shown). Therefore, we concluded that the expression of the exogenous C. W80 GPX-like protein did not affect the overall plant metabolism.

Table 2.  Photosynthetic activities, activities of antioxidative enzymes, and levels of antioxidants in leaves of wild-type, TcGPX-13, and TpGPX-14 plants
 Wild typeTcGPXTpGPX
  • *

    Value expressed in U mg−1 protein.

  • Third leaf of wild-type and transgenic plants was analyzed. The activities of APX isoenzymes (soluble and insoluble forms) were separately assayed by the method reported by Yoshimura et al. (2000). Data are mean ± SD (n = 3).

CO2 fixation (µmol CO2 m−2 sec−1)13.8 ± 1.214.5 ± 1.213.4 ± 1.1
PSII activity (Fv/Fm)0.76 ± 0.080.76 ± 0.090.77 ± 0.08
Activities of antioxidative enzymes (nmol min−1 mg−1 protein)
 APX (soluble forms)678.4 ± 60.8670.8 ± 62.1673.4 ± 57.4
 APX (insoluble forms)175.4 ± 20.2176.3 ± 20.8163.2 ± 17.8
 GR32.8 ± 3.832.8 ± 3.430.9 ± 3.2
 SOD*3.8 ± 0.44.0 ± 0.33.8 ± 0.4
 Catalase128.5 ± 10.8125.4 ± 9.5126.7 ± 10.6
Amounts of antioxidants (µmol g−1 FW)
 AsA1.95 ± 0.111.97 ± 0.121.95 ± 0.11
 DAsA0.25 ± 0.010.25 ± 0.010.21 ± 0.01
 GSH0.23 ± 0.010.24 ± 0.010.23 ± 0.01
 GSSG0.015 ± 0.0010.015 ± 0.0010.015 ± 0.001

Effect of oxidative stress caused by MV application on membrane damage

To assess the tolerance to oxidative stress, the leaf discs from wild-type, TcGPX, and TpGPX plants were incubated with MV (5 µm) under moderate light intensity (200 µE m−2 sec−1) (Figure 2). While the bleaching of the chlorophyll in the leaf discs of wild-type plants was observed after 9 h of incubation, the discs from TcGPX and TpGPX plants clearly remained green. The chlorophyll content in both types of transgenic plant (211–253 mg m−2) was higher than that in wild-type plants (21 mg m−2) at 9 h of incubation. Neither wild-type nor transgenic plants incubated with MV in the dark showed degradation of chlorophyll (data not shown).

Figure 2.

Effect of MV application on leaf discs of wild-type (WT), TcGPX, and TpGPX tobacco plants.

The leaf discs from WT and transgenic plants were floated on a solution containing 5 µm MV and 0.1% Tween-20, placed in the dark for 1 h, and then illuminated at modulated light intensity (200 µE m−2 sec−1) for 9 h at 25°C.

(a) Photograph of the leaf discs after MV treatment.

(b) The chlorophyll contents in the leaf discs before and after the treatment. The data are the mean ± SD of three individual experiments.

To analyze the effect of C. W80 GPX-like protein expression on the membrane damage caused by oxidative stress, the rates of electrolyte leakage from the leaf discs were measured after MV application. The rates of electrolyte leakage, which is caused by lipid hydroperoxidation of the cell membranes, of the leaf discs from both TcGPX (55.7–68.0%) and TpGPX (42.3–52.6%) plants were lower than the rates from wild-type plants (100%) at 9 h of exposure to the stress condition (Figure 3).

Figure 3.

Effect of MV application on electrolyte leakage in wild-type, TcGPX, and TpGPX plants.

The conductivity in MV-treated discs were determined as described in the Experimental procedures section and were compared with that of autoclaved discs. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

The changes in the rates of production of lipid hydroperoxide induced by oxidative stress were measured by determining the MDA content in the leaf discs. Lipid hydroperoxidation is an effective indicator of cellular oxidative damage (Roxas et al., 2000). The production of MDA in wild-type plants was significantly increased (4.1-fold) at 9 h under the stress condition (Figure 4). In contrast, the production in the TcGPX and TpGPX plants was only slightly increased (1.3–2-fold). These results indicate that the expression of C. W80 GPX-like protein in tobacco plants provides increased tolerance to oxidative stress related to membrane-lipid hydroperoxidation.

Figure 4.

Effect of MV application on lipid hydroperoxide contents in wild-type (WT), TcGPX, and TpGPX plants.

Lipid hydroperoxide contents were determined by measurement of MDA in leaf discs using TBA assay as described in the Experimental procedures section. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

Effect of oxidative stress caused by chilling on membrane damage

To evaluate the effect of the C. W80 GPX-like protein expression on the antioxidative systems in chloroplasts under environmental stress conditions, whole plants of wild-type and transgenic lines (TcGPX-1, TcGPX-7, TcGPX-9 and TcGPX-13, and TpGPX-1, TpGPX-6 and TpGPX-14) were exposed to chilling stress at high light intensity (4°C, 1000 µE m−2 sec−1) for 6 h and then were transferred to normal conditions (25°C, 300 µE m−2 sec−1). The third leaves of both plants were used to analyze several physiological parameters. The leaves of both types of plants wilted at 3 h after the chilling-stress treatment (data not shown). CO2 fixation could not be detected in wild-type plants at 6 h after the chilling stress (Figure 5a). However, CO2 fixation in the TcGPX and TpGPX plants remained high (7.4–21.0 and 17.2–47.1%, respectively, compared with the values before the stress). Furthermore, at 3 h of recovery following the chilling stress, CO2 fixation in the TcGPX and TpGPX plants recovered to 53.0–56.6 and 64.2–76.6%, respectively, while that in wild-type plants remained low (21.2%; Figure 5b). The PSII activity of wild-type plants decreased to 80.2% at 6 h of the stress, while the activity in the TcGPX and TpGPX plants remained high (90.7–94.0 and 92.3–94.4%, respectively).

Figure 5.

Effects of chilling on the CO2 fixation and PSII activity (Fv/Fm) in wild-type, TcGPX, and TpGPX plants.

(a) The CO2 fixation in the third leaf in wild-type and transgenic plants was measured at 1000 µmol CO2 m−2 sec−1, 400 µE m−2 sec−1, 25°C, and 60% relative humidity.

(b) The PSII activity was determined at 25°C after dark adaptation for 30 min. The data are the mean value ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

The production of MDA in the wild-type plants was significantly increased (1.43-fold) at 6 h of chilling stress (Figure 6). In contrast, the production in the TcGPX and TpGPX plants was slightly increased (1.17–1.37- and 1.16–1.23-fold, respectively).

Figure 6.

Effect of chilling on lipid hydroperoxide contents in wild-type (WT), TcGPX, and TpGPX plants.

Lipid hydroperoxide contents were determined as described in the Experimental procedures section. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

The activities of antioxidative enzymes and the amounts of antioxidants during the chilling stress and upon recovery from the stress were compared between wild-type and transgenic plants (TcGPX-13 and TpGPX-14; Figures 7 and 8). The GSH content was increased significantly after the stress in both types of plant. The AsA content in both types of plant remained unchanged in the stress and recovery periods. However, the APX activities in the soluble and insoluble forms of wild-type plants were remarkably decreased at 3 h of the stress, while those of the TcGPX-13 and TpGPX-14 plants remained high and reached somewhat less than 90% within 3 h of recovery. The activity of GR decreased after the stress in wild-type plants, while that of transgenic plants did not change.

Figure 7.

Effects of chilling on the activities of AOS-scavenging enzymes in wild-type, TcGPX, and TpGPX plants.

The activities of the APX isoenzymes (soluble and insoluble forms) were separately assayed by the method reported by Yoshimura et al. (2000). Detailed procedures are described in the Experimental procedures section. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

Figure 8.

Effect of chilling on the levels of antioxidants in wild-type, TcGPX, and TpGPX plants. Detailed procedures are described in the Experimental procedures section. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

Effect of oxidative stress caused by salt stress

Whole plants of the wild-type and transgenic lines (TcGPX-1, TcGPX-7, TcGPX-9, and TcGPX-13, and TpGPX-1, TpGPX-6, and TpGPX-14) were exposed to salt stress by treatment with 250 mm NaCl under normal light intensity (300 µE m−2 sec−1). The leaves of both types of plants wilted at 12 h after the salt-stress treatment (data not shown). CO2 fixation could not be detected in the third leaves of wild-type plants at 24 h after the salt stress (Figure 9). However, CO2 fixation remained high in the TcGPX and TpGPX plants (24.7–35.7 and 47.0–56.5%, respectively). The PSII activity of wild-type plants decreased to 78.5% at 24 h of the stress, while the activity in the TcGPX and TpGPX plants remained high (90.3–92.6 and 91.4–93.1%, respectively). The production of MDA in wild-type plants was significantly increased (1.58-fold) at 24 h after salt stress (Figure 10). In contrast, the production in the TcGPX and TpGPX plants was increased to a lower extent (1.36–1.43- and 1.30–1.35-fold, respectively).

Figure 9.

Effects of salinity on the CO2 fixation in wild-type, TcGPX, and TpGPX plants.

The CO2 fixation in the third leaf in wild-type and transgenic plants was measured at 1000 µmol CO2 m−2 sec−1, 400 µE m−2 sec−1, 25°C, and 60% relative humidity. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

Figure 10.

Effect of salinity on lipid hydroperoxide contents in wild-type (WT), TcGPX, and TpGPX plants.

Lipid hydroperoxide contents were determined as described in the Experimental procedures section. The data are the mean ± SD of three individual experiments. Different letters indicate significant differences (P < 0.05).

Discussion

Lipid hydroperoxidation is commonly regarded as a deleterious process that leads to structural modification of complex lipid protein assemblies such as biomembranes and lipoproteins, and is usually associated with disruption of cellular functions (Munnik et al., 1998). AOS are introduced into the hydrophobic tails of unsaturated fatty acids to produce lipid hydroperoxides. Lipid hydroperoxide-scavenging systems appear to be necessary for regulation of the lipid hydroperoxidation of biomembranes (Munnik et al., 1998). It is likely that the unsaturated fatty acid hydroperoxides in oxidized biomembranes are released by the activity of phospholipase A2 (EC 3.1.1.4), which hydrolyzes phospholipids at the sn-2 position, generating lyso-phospholipids and free fatty acids. Furthermore, drought, chilling, and salt stress rapidly stimulate phospholipase A2 activity (Meijer et al., 2001; Munnik and Meijer, 2001). On the other hand, GPXs in higher plants have lower activity toward phospholipid hydroperoxide than those in animals, although a large number of genes encoding GPX (-like) proteins exist in higher plants and constitute a multigene family (Criqui et al., 1992; Depège et al., 1998; Holland et al., 1993; Mullineux et al., 1998; Roeckel-Drevet et al., 1998; Sugimoto and Sakamoto, 1997). Furthermore, GPXs cannot reduce unsaturated fatty acid hydroperoxides, unlike enzymes such as C. W80 GPX-like protein and bovine cytosolic GPX.

It has been reported that transgenic tobacco plants overexpressing GST with GPX activity display enhanced seedling growth under thermal- and salt-stress conditions that cause an increase in lipid hydroperoxidation (Roxas et al., 1997). However, no other studies concerning the scavenging systems of lipid hydroperoxide in higher plants have been reported. Palmitate (16 : 0), stearate (18 : 0), oleate (18 : 1), linoleate (18 : 2), and linolenic acid (18 : 3) are ubiquitously present in the tissues of higher plants (Griffiths et al., 2000; Hitchcock and Nichols 1971). Among them, linolenic acid is the predominant polyunsaturated fatty acid. The C. W80 enzyme reduces neither H2O2 nor lipid hydroperoxide, but reduces various unsaturated fatty acid hydroperoxides as well as alkyl hydroperoxides using GSH as an electron donor (Takeda et al., 2003). Especially, the enzyme shows its highest activity and high affinity (Km = 94 ± 16 µm) toward α- and γ-linolenic acid hydroperoxide.

To investigate the activity of the unsaturated fatty acid hydroperoxide-scavenging system in the TcGPX and TpGPX plants under oxidative stress conditions, we analyzed the effect of oxidative stress caused by MV application on membrane damage. The chlorophyll degradation in the leaf discs from both the TcGPX and TpGPX plants was lower than that in discs from wild-type plants after incubation with MV (5 µm) under moderate light intensity (200 µE m−2 sec−1; Figure 2). The rate of electrolyte leakage, which is because of lipid hydroperoxidation of the cell membranes, in the leaf discs from both types of transgenic plant remained lower than that from wild-type plants (Figure 3). The production of lipid hydroperoxide in the TcGPX and TpGPX plants was significantly suppressed compared with that in wild-type plants. These results indicate that the expression of C. W80 GPX-like protein in tobacco plants provides increased tolerance to oxidative stress related to suppression of membrane-lipid hydroperoxidation.

Next, we checked the effect of the expression of C. W80 GPX-like protein on the tolerance to oxidative stress caused by chilling at high light intensity (4°C, 1000 µE m−2 sec−1) or by salt stress (250 mm NaCl). The photosynthetic capacities of the TcGPX and TpGPX plants remained high compared with those of wild-type plants under both stress conditions (Figures 5 and 9). Furthermore, the production of lipid hydroperoxide and the decrease in antioxidative enzymes in the TcGPX and TpGPX plants were greatly suppressed compared with those in the wild-type plants (Figures 6, 7, and 10). It has been reported that chilling stress or salt stress enhances the AOS production by the electron transport system because of the slowdown of cellular metabolism, and that lipid hydroperoxidation of the cellular membrane is consequently induced (Foyer et al., 2002; Lalk and Dorffing, 1985; Lang and Palva, 1992; Prasad, 1996; Prasad et al., 1994). The reduction of membrane phospholipid peroxidation by GLK-8903 has been found to alleviate chilling injury in Phaseolus vulgaris L. (Zhang et al., 1994). Furthermore, Escherichia coli cells expressing the C. W80 GPX-like protein gene showed increased resistance against the oxidative stress caused by M▿ or NaCl treatment (Takeda et al., 2003). Similar results were observed in E. coli cells expressing GPX-like protein gene of C. sinensis (Holland et al., 1993). Therefore, it is likely that enhancement of the lipid hydroperoxide-scavenging capacity is essential for tolerating chilling and salt stresses.

Interestingly, judging from all of the parameters related to tolerance to several stresses, the plastid-targeted TpGPX lines seemed to be more tolerant compared with the cytosol-targeted TcGPX lines. The fatty acid profile of total lipids is known to be almost the same in leaves, roots, and isolated chloroplasts (Griffiths et al., 2000; Hitchcock and Nichols, 1971). Under several stress conditions, the main sites of AOS production in the plant cell during abiotic stress are the organelles with highly oxidizing metabolic activities or with sustained electron flows: chloroplasts, mitochondria, and microbodies (Dat et al., 2000). This may explain why the expression of GPX-like protein in chloroplasts was more effective at providing stress tolerance than the expression in the cytosol. It is also well known that the lipid hydroperoxide-scavenging system is associated with signal transduction for gene expression of various proteins, as unsaturated fatty acids act as a signal or are metabolized to produce signals (Munnik et al., 1998). Therefore, it is conceivable that the overexpression of GPX-like protein in chloroplasts may affect the signal transduction system by modulating lipid hydroperoxide levels under stress conditions. We are now attempting to analyze the metabolism of unsaturated fatty acids from oxidized biomembranes by phospholipase A2 in higher plants.

Experimental procedures

Materials

The cDNA encoding the C. W80 GPX-like protein was originally cloned into plasmid pBluescript SK(+) (Takeda et al., 2003). Hydroperoxides of unsaturated fatty acids including α-linolenic acid and linoleic acid were prepared by the method described previously by Takeda et al. (2003). All other chemicals were of the highest grade commercially available.

Vector construction and transformation of tobacco

For cytosol-targeted expression, the cDNA encoding the C. W80 GPX-like protein was ligated into plant binary vector pBI121 containing linker derived from Ti-plasmid, in which the cDNA was placed under the control of the CaMV 35S promoter in the sense orientation. For plastid-targeted expression, the transit peptide sequence of RuBisCO small subunit from tomato was translationally fused with the first methionine of the GPX cDNA in the transgene. The recombinant plasmid was introduced into Agrobacterium tumefaciens strain LBA 4404, and transformation of tobacco plants (N. tabacum cv. Xanthi) by A. tumefaciens was performed as described previously by Miyagawa et al. (2000) and Shikanai et al. (1998). Transgenic tobacco and wild-type plants were cultured for 7 weeks in a growth chamber under a 12-h photoperiod with moderate light intensity (300 µE m−2 sec−1), 60% relative humidity, and day/night temperature of 25/20°C.

Detection of the transgene by PCR

A total of 100 mg of fresh tissue from wild-type and transgenic plants (T1 generation) was ground in liquid N2 into fine powder and then homogenized in 400 µl of grinding buffer (50 mm Tris–HCl (pH 7.5), 20 mm EDTA, 0.3 m NaCl, 0.5% (w/v) SDS, 5 m urea, and 5% (w/v) phenol). Four hundred microliters of PCI (phenol:chloroform:isoamylalcohol = 25 : 24 : 1) was added to the homogenate. After centrifugation, the supernatant was transferred to a new microfuge tube and two volumes of ethanol were added. DNA was precipitated by centrifugation for 15 min, and the pellet was washed with 70% (v/v) ethanol, dried, and suspended in 10 µl. About 100 ng of genomic DNA was used to amplify the transgene by PCR using as primers oligonucleotides C. W80 GPX-f: 5′- CAGGTCTTGTTAATGCCCTGA-3′ and C. W80 GPX-r: 5′-ACTGCCAAAACTCCAATCACAATC-3′. The amplified bands were photographed under UV light after electrophoresis.

RNA extraction and Northern blot analysis

Total RNA was isolated from tobacco leaves (1.0 g FW) as described previously by Yabuta et al. (2002). Total RNA (30 µg each) was subjected to electrophoresis on 1.2% (w/v) agarose gels containing 2.2 m formaldehyde and transferred to a Hybond N membrane (Amersham Biosciences, NY, USA). The membrane was pre-hybridized at 55°C for 3 h in a buffer containing 6× SSC, 5× Denhardt's solution, 1% (w/v) SDS, and 100 µg ml−1 denatured salmon sperm DNA. The membrane was hybridized at 55°C for 12 h in the presence of the 32P-random-primed cDNA encoding C. W80 GPX-like protein and washed two times at room temperature in 2× SSC, 0.1% SDS for 10 min each, and in 0.1× SSC, 0.1% SDS at 60°C for 1 h. The membrane was then exposed to an imaging plate, and the relative expression ratio of C. W80 GPX-like protein transcript was calculated using a Mac BAS 2000 (Fuji Photofilm, Tokyo, Japan) and expressed as the mean values from three individual experiments.

Western blot analysis

Polyclonal antibody (anti-CWGPX) raised against C. W80 GPX-like protein was prepared using the purified recombinant protein (Takeda et al., 2003). Mice were injected with 50 µg of the purified recombinant protein emulsified with Freund's complete adjuvant followed by three subcutaneous injections of the same amount of the protein. After bleedings, the antiserum was separated from the blood and purified by saturation with a crude extract from the E. coli strain BL21 (DE3) pLysS.

Discs (1.1 cm2) of leaf tissues (30 discs) were ground to a fine powder in liquid N2 and then homogenized with 1 ml of SDS-loading buffer (150 mm Tris–HCl (pH 6.8), 4% (w/v) SDS, and 10% (v/v) 2-mercaptoethanol). The homogenates were boiled for 5 min and centrifuged at 10 000 g for 10 min. The proteins in the supernatant (30 µl) were separated by electrophoresis in a 12.5% slab gel according to the method by Laemmli (1970) and stained with Coomassie Brilliant Blue R-250. For immunoblotting, the proteins were transferred to poly-vinylla-dene-fluoride (PVDF) membranes using an electroblot apparatus (Model 200/2.0, Bio-Rad Laboratories, CA, USA) at 13 V for 1 h. The membranes were treated with anti-CWGPX, which specifically reacted with C. W80 GPX-like protein. Signals on the membranes were visualized with alkaline phosphatase-conjugated goat antimouse IgG (Bio-Rad Laboratories).

Isolation of chloroplasts

Intact chloroplasts were isolated from leaves of wild-type, TcGPX, and TpGPX plants by a modification of the method described previously by Takeda et al. (1995). To rid most of starch, the 7-week-old plants were grown in the dark for 48 h prior to the isolation of chloroplasts. The leaves (30 g FW) were homogenized in buffer A (100 ml of 50 mm HEPES buffer adjusted to pH 7.6 with 1 m KOH, containing 0.33 m sorbitol, 2 mm EDTA, 1 mm MgCl2, 1 mm MnCl2, 5 mm pyrophosphate, and 0.5 mm KH2PO4) with polytron mixer. The homogenate was squeezed through four layers of gauze, and the filtrate was centrifuged at 2000 g for 30 sec. The pellets were suspended in buffer A and centrifuged at 2000 g for 1 min. The pellets were re-suspended in buffer A to give a chlorophyll concentration of approximately 1 mg ml−1 and subjected to discontinuous density gradient centrifugation using Percoll (Amersham Biosciences, NJ, USA) gradients in steps of 10, 40, 70, and 90%. Centrifugation was carried out at 4000 g for 15 min, and the intact chloroplast layer between the 40 and 70% Percoll fractions was removed. Intactness of chloroplasts was measured by the ferricyanide method (Takeda et al., 1995), indicating that between 85 and 95% of the chloroplasts were intact.

Enzyme extraction and assays

The GPX activity was assayed spectrophotometrically according to Takeda et al. (1993) with GSH and H2O2 or hydroperoxides in the presence of GR, which catalyzed the reduction of GSSG formed by GPX. Leaf tissues (10 discs of 1.1 cm2 each) were ground to a fine powder in liquid N2 and then homogenized in 1 ml of 100 mm Tris–HCl buffer (pH 8.2), 10% (w/v) sorbitol, and 5 mm GSH using a mortar and pestle. The homogenate was centrifuged for 15 min at 100 000 g. The supernatant was used for the assay of GPX activity. The reaction mixture contained 100 mm Tris–HCl (pH 8.2), 1 mm GSH, 0.4 mm NADPH, 0.2 mm H2O2, 1 unit of GR, and the enzyme in a total volume of 1 ml. The reduction of hydroperoxides was measured in the same assay mixture, except for the replacement of H2O2 with 0.2 mm unsaturated fatty acid or lipid hydroperoxide and 0.1% Triton X-100.

GST was assayed using 1-chloro-2,4-dinitrobenzene (CDNB) as the substrate (Takeda et al., 1992). Leaf tissues (10 discs of 1.1 cm2 each) were ground to a fine powder in liquid N2 and then homogenized in 1 ml of 50 mm potassium phosphate buffer (pH 7.2), 20% (w/v) sorbitol, and 2% (w/v) polyvinylpyrrolidone using a mortar and pestle. The homogenate was centrifuged for 15 min at 100 000 g. The supernatant was used for the assay of GST activity. The reaction mixture contained 50 mm potassium phosphate buffer (pH 7.0), 1 mm GSH, 1 mm CDNB, and the enzyme in a total volume of 1 ml. One unit of activity of each enzyme except for SOD is defined as the amount required to reduce 1 µmol of substrate per min. The activities of the APX isoenzymes were separately assayed by the method reported by Yoshimura et al. (2000). Leaf tissues (10 discs of 1.1 cm2 each) were ground to a fine powder in liquid N2 and then homogenized in 1 ml of 100 mm potassium phosphate buffer (pH 7.6), 20% (w/v) sorbitol, 1 mm EDTA, 5 mm AsA, and 2% (w/v) polyvinylpyrrolidone using a mortar and pestle. The homogenate was centrifuged for 15 min at 100 000 g. The soluble fraction contained the activities of the stromal and cytosolic APX isoenzymes, while the membrane fraction had the activities of the thylakoid membrane- and microbody membrane-bound APX isoenzymes. Preparation and assay of GR were carried out according to Yabuta et al. (2002). Protein was determined by the method by Bradford (1976), with bovine serum albumin as a standard. Chlorophyll was measured by the method by Arnon (1949).

Gas exchange and chlorophyll fluorescence measurements

CO2 fixation and chlorophyll fluorescence were measured according to Miyagawa et al. (2000). CO2 fixation in the third leaf of the tobacco plants was measured with a portable photosynthesis system LI-6400 (Li-Cor, Lincoln, NE, USA). Net CO2 assimilation rates were measured using fully expanded leaves under the following conditions: 400 µE m−2 sec−1, 1000 µmol CO2 m−2 sec−1, 25°C, and 60% relative humidity. Chlorophyll fluorescence in the third leaf of the tobacco plants was measured at 25°C with a Mini PAM Chlorophyll Fluorometer (Waltz, Efeltrich, Germany).

MV application to leaf discs

Leaf discs (1-cm diameter) punched out from the third leaves of 7-week-old plants using a cork borer were subjected to MV application. The leaf discs from wild-type, TcGPX, and TpGPX plants were floated on a solution containing 5 µm MV and 0.1% Tween-20, placed in the dark for 1 h, and then illuminated at moderate light intensity (200 µE m−2 sec−1) for 9 h at 25°C.

Chilling stress

Chilling stress was imposed by transferring 7 weeks-old plants to 4°C at 1000 µE m−2 sec−1 illumination supplied by a fluorescent lamp (FPL36EX, 36 W, National, Osaka, Japan) for 6 h in the large-size refrigerator. To minimize any heat effects, the light illumination was filtered through a glass tray with a frosted bottom containing 5 cm of cold water placed over the plants. The treated plants were transferred to normal conditions (25°C, 300 µE m−2 sec−1) for recovery from the stress. For the enzyme assays and the measurement of metabolites, discs (1.1 cm2) from the third leaves of four plants of each line were harvested at the indicated times.

Salt stress

Salt stress was imposed by applying 15 ml of a 250 mm NaCl solution dissolved in water to 7-week-old plants two times at 0 and 12 h. The treated plants were then cultured under 300 µE m−2 sec−1 illumination and were harvested at the indicated times.

Measurement of electrolyte leakage

The electrolyte leakage into the solutions used for floating the leaf discs was determined after MV application by using a HORIBA (Kyoto, Japan) ES-12 conductivity meter to measure ion leakage from the leaf discs because of the lipid hydroperoxidation of the cell membranes. The conductivity of the solution was determined, and the percentage of electrolyte leakage attributable to the MV treatment was determined by dividing the conductivity value of the test sample by the conductivity of the sample after autoclaving (100% electrolyte leakage).

Measurement of lipid hydroperoxide contents

Lipid hydroperoxide contents were determined by measuring MDA using the 2-thiobarbituric acid (TBA) assay as described by Roxas et al. (1997). Leaf tissues (10 discs of 1.1 cm2 each) were extracted in a solution of 5% trichloroacetic acid and 0.5 g l−1 methanolic butylated hydroxytoluene. The extracts were boiled for 30 min and reacted with TBA. MDA contents were estimated by measuring A532 to A600 and using a molar absorption coefficient of 1.56 × 105 (Gueta-Dahan et al., 1997).

Acknowledgements

This work was supported by the New Energy and Industrial Technology Development Organization (NEDO) subsidized by the Ministry of Economy, Trade and Industry of Japan (S.S. and K.Y.) and by the Research for the Future Program from the Japan Society for the Promotion of Science (JSPS-RFTF00L01604) (S.S.). This work was in part supported by the Ministry of Agriculture, Forestry, and Fisheries, Japan (S.S.).

Ancillary

Advertisement