Hydroperoxide reduction by thioredoxin-specific glutathione peroxidase isoenzymes of Arabidopsis thaliana Aqib Iqbal 1 , Yukinori Yabuta 2 , Toru Takeda 2 , Yoshihisa Nakano 1 and Shigeru Shigeoka 2 1 Department of Applied Biological Chemistry, Osaka Prefecture University, Sakai, Japan 2 Department of Advanced Bioscience, Kinki University, Nara, Japan The generation of reactive oxygen species (ROS) is an inevitable process in living organisms and poses a hazard when present in high concentrations by irre- versibly damaging different macromolecules such as protein, lipid and DNA [1]. Plants, because of their sessile nature, are at greater risk of alterations in redox homeostasis resulting in oxidative stress [2]. However, at moderate concentrations ROS play an important role in signaling processes as the regulatory mediators. Thus, many ROS-mediated responses actually protect cells against oxidative stress and re-establish redox homeostasis [3]. In mammals, glutathione peroxidase (GPX) iso- enzymes (EC 1.11.1.9) play a key role in protecting cells against oxidative damage. At least five GPX iso- enzymes have been identified in mammals and these differ with respect to structure, substrate specificity and tissue distribution [4]. cDNA with homology to mammalian GPX isoenzymes has been cloned from Nicotiana sylvestris leaves [5] and since then more have been isolated from different plant species and photo- synthetic organisms [6–14]. Recently, we reported that GPX proteins (Gpx-1, Gpx-2) from Synechocystis PCC 6803 are able to util- ize NADPH, but not glutathione (GSH), as an elec- tron donor, and unsaturated fatty acid hydroperoxides or alkyl hydroperoxides as acceptors [15]. In addition, it has been reported that yeast GPX isoenzymes can utilize both GSH and thioredoxin (Trx) as a reducing agent [16]. This profile was also observed in GPX iso- enzymes from sunflower, tomato and Chinese cabbage [11,17]. Thus it is evident that in addition to GSH, GPX proteins can utilize other physiological substrates for hydroperoxide reduction. GSH and Trx from eukaryotic cells are two major reducing compounds that maintain cellular redox Keywords Arabidopsis; glutathione peroxidase; hydroperoxide; thioredoxin Correspondence T. Takeda, Department of Advanced Bioscience, Faculty of Agriculture, Kinki University, 3327-204 Nakamachi, Nara 631–8505, Japan Fax: +81 742 43 8976 Tel: +81 742 43 8179 E-mail: t_takeda@nara.kindai.ac.jp (Received 22 August 2006, revised 18 October 2006, accepted 20 October 2006) doi:10.1111/j.1742-4658.2006.05548.x Arabidopsis thaliana contains eight glutathione peroxidase (GPX) homologs (AtGPX1–8). Four mature GPX isoenzymes with different subcellular dis- tributions, AtGPX1, -2, -5 and -6, were overexpressed in Escherichia coli and characterized. Interestingly, these recombinant proteins were able to reduce H 2 O 2 , cumene hydroperoxide, phosphatidylcholine and linoleic acid hydroperoxides using thioredoxin but not glutathione or NADPH as an electron donor. The reduction activities of the recombinant proteins with H 2 O 2 were 2–7 times higher than those with cumene hydroperoxide. K m values for thioredoxin and H 2 O 2 were 2.2–4.0 and 14.0–25.4 lm, respect- ively. These finding suggest that GPX isoenzymes may function to detoxify H 2 O 2 and organic hydroperoxides using thioredoxin in vivo and may also be involved in regulation of the cellular redox homeostasis by maintaining the thiol ⁄ disulfide or NADPH ⁄ NADP balance. Abbreviations ABA, abscisic acid; AtGPX, Arabidopsis thaliana homolog of glutathione peroxidase; GPX, glutathione peroxidase; GSH, glutathione; Nbs 2 , 5,5¢-dithiobis(2-nitrobenzoic acid); PhGPX, phospholipid hydroperoxide glutathione peroxidase; ROS, reactive oxygen species; SeCys, selenocysteine; TPx, Trx-dependent peroxidase; Trx, thioredoxin. FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS 5589 balance and interact with various transducer and effec- ter molecules to bring about specific responses [18]. GSH not only acts as a redox sensor of environmental cues, but also forms part of multiple regulatory path- ways coordinating the expression of defense genes [19]. Trx is a ubiquitous protein that is present in prokaryotes and eukaryotes, and shows particularly wide diversity in photosynthetic organisms [20,21]. In higher plants, Trxs are distributed in the cytoplasm, plasma membrane, endoplasmic reticulum, nucleus, apoplast, mitochon- drion and chloroplast [18,22], and are involved in the regulation of different metabolic processes. Eight genes with homology to mammalian GPX iso- enzymes have been identified in Arabidopsis thaliana. Seven of these, AtGPX1–7, are differently expressed in various plant tissues. However, cDNAs encoding five GPX isoenzymes, AtGPX1, -2, -3, -5 and -6, were detected in the ESTs database and the expression of five genes was clearly different in response to different abiotic stresses and plant hormones [23]. Two proteo- mic studies have identified AtGPX1 which is located in the thylakoid membrane of chloroplast [24,25]. In addition, AtGPX3 green fluorescent fusion protein has been localized in the cytoplasm [26]. Based on the deduced amino acid sequences of the other cDNAs, it seems likely that AtGPX2, -5 and -6 are putatively dis- tributed in the cytosol, endoplasmic reticulum and mitochondria ⁄ cytosol, respectively [23]. GPX proteins localized in the cytosol have been previously character- ized only from plants and photosynthetic organisms; thus there is no information available about the enzy- matic properties of the isoenzymes localized in other subcellular compartments. In this study, large amounts of four types of recombinant GPX proteins, AtGPX1, -2, -5 and -6 with different subcellular distribution were obtained. It was found that the recombinant pro- teins can reduce H 2 O 2 , cumene hydroperoxide, phos- phatidyl choline and linoleic acid hydroperoxides using Trx as an electron donor, but not GSH or NADPH. Their physiological roles in plants is also discussed. Results Expression of proteins of AtGPX1, -2, -5 and -6 and AtTrx h2 and h3 in E. coli Chimeric primers were designed to amplify AtGPX1, -2, -5 and -6 fragments comprising the amino acid residues encoding the mature proteins without transit peptides that target them to different subcellular compartments. Each recombinant protein was induced in E. coli BL21 (DE3) pLysS cells in an exponentially growing culture according to the procedure described in the Experimental procedures. SDS ⁄ PAGE analysis of the soluble proteins from the host cells showed that the proteins are successfully expressed using this sys- tem, resulting in the presence of a prominent band at  22.5 kDa, corresponding to the expected molecular mass of each AtGPX isoenzyme tagged with the histi- dine residues (Fig. 1A). The recombinant proteins pro- duced were 20–25% of the total soluble proteins. Recombinant AtGPX1, -2, -5 and -6 proteins of near electrophoretic homogeneity were obtained using a HiTrap TM chelating HP column. Recombinant thio- redoxin h2 (17.0 kDa) and h3 (15.5 kDa) were also purified using the same procedure (Fig. 1B). The pro- teins of recombinant AtGPX and Trx were used for the enzymatic analysis. A B Fig. 1. SDS ⁄ PAGE analysis of the expression of Arabidopsis recombinant proteins in E. coli using a pColdII vector system. (A) Recombinant AtGPX1, -2, -5 and -6 (GenBank TM accession numbers NP_180080, NP_180715, NP_191867 and AAK63967, respectively); (B) recombinant Txr h2 and h3 (GenBank TM accession numbers NP_198811 and NP_199112, respectively). The soluble crude extract (10 lg) and each puri- fied recombinant protein (1 lg) were separated by 15% SDS ⁄ PAGE and stained with Coomassie Brilliant Blue. The sizes of the molecular mass markers are shown on the left of the panel. Lane 1, empty pColdII vector. i, Total soluble proteins; ii, purified recombinant proteins. A. thaliana glutathione peroxidases A. Iqbal et al. 5590 FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS Detection of enzyme activities of GPX isoenzymes When the recombinant AtGPX isoenzymes were used with GSH or NADPH as a reducing agent and H 2 O 2 , cumene hydroperoxide, unsaturated fatty acids or lipid hydroperoxides as electron acceptor, no hydroperoxide reduction activities were detected (Table 1). However, it was found that AtGPX isoenzymes were able to reduce H 2 O 2 with an E. coli Trx as the reducing sub- strate. In addition, these isoenzymes were found to utilize Trx for the reduction of cumene hydroperoxide, unsaturated fatty acids hydroperoxides or lipid hydro- peroxides. The V max values for the purified recombinant AtGPX isoenzymes with cumene hydroperoxide and Trx were between 60 ± 16.2 and 233 ± 16.7 nmolÆ min )1 Æmg )1 protein. However, AtGPX5 was not able to reduce cumene hydroperoxide with either Trx or GSH. The V max values for AtGPX isoenzymes with H 2 O 2 and Trx were approximately two- to sevenfold higher than those of cumene hydroperoxide. Activities toward unsaturated fatty acid hydroperoxides and lipid hydroperoxides were very low compared with those toward cumene hydroperoxide (Table 1). The reduc- tion of hydroperoxides with two cytosolic Trx’s, h2 and h3 from Arabidopsis were also measured in an assay coupled with E. coli Trx-reductase. However, the AtGPX isoenzymes were not able to reduce hydroper- oxides with either Trx h2 or h3 (Table 1). Enzymatic properties of recombinant GPX isoenzymes Double reciprocal plots of 1 ⁄ [activity] against 1⁄ [Trx] for Trx and H 2 O 2 were linear and reproducible for each AtGPX isoenzyme (data not shown). The V max values for AtGPX1, -2, -5 and -6 with a fixed concentration of H 2 O 2 (0.1 mm) were 423 ± 22.5, Table 1. Reduction of alkyl hydroperoxide (cumene hydroperoxide), H 2 O 2 , polyunsaturated fatty acids hydroperoxide (PUFAOOH, linoleic acid) or phosphatidylcholine hydroperoxide (PCOOH) with GSH, NADPH or Trx by AtGPX isoenzymes and comparison with other plant GPXs. ER, endoplasmic reticulum; mit, mitochondria; ND, not determined. Species Name and location Substrate Activity (nmolÆmin )1 Æmg )1 protein) Arabidopsis Trx h2 Trx h3 GSH NADPH Trx (E. coli) Arabidopsis GPX1, Plastids CumOOH 0 0 233 ± 16.7 0 0 H 2 O 2 0 0 423 ± 22.5 0 0 PUFAOOH 0 0 35 ± 4.1 ND ND PCOOH 0 0 0 ND ND GPX2, Cytosol CumOOH 0 0 60 ± 16.7 0 0 H 2 O 2 0 0 346 ± 33.5 0 0 PUFAOOH 0 0 29 ± 0.5 ND ND PCOOH 0 0 12 ± 2.9 ND ND GPX5, ER CumOOH 0 0 0 0 0 H 2 O 2 0 0 409 ± 29.4 0 0 PUFAOOH 0 0 16 ± 3.3 ND ND PCOOH 0 0 8 ± 1.3 ND ND GPX6, Cytosol ⁄ mit CumOOH 0 0 65 ± 5.8 0 0 H 2 O 2 0 0 433 ± 64.2 0 0 PUFAOOH 0 0 38 ± 2.8 ND ND PCOOH 0 0 18 ± 2.8 ND ND Citrus a GPX1, Cytosol PCOOH 40 ND ND ND ND Tomato b GPXle1, Cytosol PUFAOOH 28 ± 0.04 ND 147 ± 1.11 ND ND AlkylOOH 58 ± 1.3 ND 147 ± 1.34 ND ND PCOOH – ND 108 ± 0.56 ND ND H 2 O 2 0 ND 153 ± 1.79 ND ND Sunflower b GPXha1, Cytosol PUFAOOH 42 ± 0.11 ND 147 ± 1.11 ND ND AlkylOOH 39 ± 0.61 ND 161 ± 0.93 ND ND PCOOH – ND 127 ± 0.05 ND ND H 2 O 2 0 ND 147 ± 2.12 ND ND Yeast c GPX2, Cytosol AlkylOOH 295 ND 1000 ND ND H2O2 270 ND 2600 ND ND Chinese cabbage d PHCC-TPx, Cytosol H 2 O 2 0 ND 17 200 ND ND CumOOH 0 ND 7600 ND ND a [6], b [17], c [16], d [11]. A. Iqbal et al. A. thaliana glutathione peroxidases FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS 5591 346 ± 33.5, 409 ± 29.4 and 433 ± 64.2 nmolÆmin )1 Æ mg )1 protein. The K m values for AtGPX1, -2, -5 and -6 with a fixed concentration of H 2 O 2 (0.1 mm) for Trx were 4.0, 2.2, 3.1 and 2.3 lm, respectively. Thus the K cat values for AtGPX1, -2, -5 and -6 for Trx were 13.4 · 10 )2 , 11.0 · 10 )2 , 13.0 · 10 )2 and 13.7 · 10 )2 Æs )1 and K cat ⁄ K m values were 3.35 · 10 4 , 5.0 · 10 4 , 4.19 · 10 4 and 5.96 · 10 4 m )1 Æs )1 , respect- ively (Table 2). V max and K m values for these isoenzymes for H 2 O 2 using a fixed concentration of 4 lm E. coli Trx were also calculated. The V max values for AtGPX1, -2, -5 and -6 were 262 ± 10.2, 218 ± 8.5, 247 ± 9.8 and 269 ± 9.1 nmolÆmin )1 Æmg )1 protein and the K m values were 17.1 ± 0.8, 15.3 ± 1.4, 25.4 ± 1.6 and 14.0 ± 1.2 lm, respectively. Accordingly, the K cat values for AtGPX1, -2, -5 and -6 calculated for H 2 O 2 were 8.30 · 10 )2 , 6.90 · 10 )2 , 7.80 · 10 )2 and 8.5 · 10 )2 Æs )1 and the K cat ⁄ K m values were 4.9 · 10 3 , 4.5 · 10 3 , 3.1 · 10 3 and 6.1 · 10 3 m )1 Æs )1 , respectively (Table 3). Discussion Based on amino acid sequence homology, GPX isoen- zymes from photosynthetic organisms, including higher plants, are closely related to mammalian GPX4 (phos- pholipid hydroperoxide GPX; PhGPX). However, these GPX isoenzymes contained a conserved Cys in their catalytic site, unlike mammalian PhGPX, which has a selenocysteine (SeCys) [4]. It has been shown that replacement of SeCys with a Cys via point mutation in pig heart GPX resulted in a drastic decrease in enzyme activity [27]. Furthermore, sequence alignments for plasma GPX (GPX3) and GPX4 from mammals showed that the amino acid residues necessary for GSH binding are not conserved. Consequently, GPX isoenzymes from photosynthetic organisms were not able to reduce sev- eral hydroperoxides utilizing GSH. However, even when they could reduce hydroperoxides with GSH as an elec- tron donor, they had a very low activity [15,28–30]. Similarly, the non-SeCys PhGPXs recently identified in mammals showed little GSH-dependent GPX activity in vitro [31]. These findings suggest that GSH is unlikely to be the sole physiological electron donor for GPX iso- enzymes under all circumstances and hence the reduc- tion activity of the AtGPX isoenzymes toward hydroperoxides with GSH is lost, which is in accordance with the results for non-SeCys GPX isoenzymes repor- ted previously [11,16,27,31]. Interestingly, AtGPX isoenzymes were able to utilize Trx as a sole electron donor for the reduction of H 2 O 2 or hydroperoxides (Table 1). GPX isoenzymes from Table 2. Comparison of kinetic characteristics of Arabidopsis, tomato and sunflower GPX isoenzymes towards thioredoxin using fixed con- centration of H 2 O 2 or t-butyl hydroperoxide (100 lM). Kinetic parameters of AtGPX isoenzymes were calculated using H 2 O 2 (100 lM) and that of GPXle1 and GPXha2 were calculated with t-butyl hydroperoxide (100 l M) as the substrates. Data for GPXle1 and GPXha2 were taken from Herbette et al. [17]. V max nmolÆmin )1 Æmg )1 protein K m lM K cat s )1 K cat ⁄ K m M )1 Æs )1 AtGPX1 423 ± 22.5 4.0 ± 0.4 13.4 · 10 )2 3.35 · 10 4 AtGPX2 346 ± 33.5 2.2 ± 0.1 11.0 · 10 )2 5.00 · 10 4 AtGPX5 409 ± 29.4 3.1 ± 0.2 13.0 · 10 )2 4.19 · 10 4 AtGPX6 433 ± 64.2 2.3 ± 0.1 13.7 · 10 )2 5.96 · 10 4 GPXle1 263 ± 0.4 2.2 ± 0.3 8.4 · 10 )2 3.8 · 10 4 GPXha2 244 ± 0.4 1.5 ± 0.06 7.8 · 10 )2 5.2 · 10 4 Table 3. Comparison of kinetic characteristics of Arabidopsis, tomato, sunflower and yeast GPX isoenzymes toward H 2 O 2 using fixed con- centration of thioredoxin (4 l M). Data for GPXle1 and GPXha2 were taken from Herbette et al. [17], data for yeast GPX2 were taken from Tanaka et al. [16]. V max nmolÆmin )1 Æmg )1 protein K m lM K cat s )1 K cat ⁄ K m M )1 Æs )1 AtGPX1 262 ± 10.2 17.1 ± 0.8 8.3 · 10 )2 4.9 · 10 4 AtGPX2 218 ± 8.5 15.3 ± 1.4 6.9 · 10 )2 4.5 · 10 4 AtGPX5 247 ± 9.8 25.4 ± 1.6 7.8 · 10 )2 3.1 · 10 4 AtGPX6 269 ± 9.1 14.0 ± 1.2 8.5 · 10 )2 6.1 · 10 4 GPXle1 154 ± 1.79 13.7 ± 0.02 4.9 · 10 )2 2.2 · 10 4 GPXha2 147 ± 1.34 13.9 ± 0.20 4.7 · 10 )2 3.1 · 10 4 Yeast GPX2 2.60 · 10 3 20.0 9.6 · 10 )2 4.8 · 10 3 A. thaliana glutathione peroxidases A. Iqbal et al. 5592 FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS several other plants, such as tomato, sunflower and Chinese cabbage, and microorganisms such as Synecho- cystis PCC 6803, Saccharomyces cerevisiae and Plasmo- dium falciparum have also been found to utilize other physiological electron donors, such as NADPH or Trx, for the reduction of hydroperoxides [11,15–17]. GPX isoenzymes from photosynthetic organisms contain three conserved Cys; the third Cys is outside the classical GPX catalytic domains (Fig. 2). The first and third Cys residues conserved in yeast GPX2 and Chinese cabbage Trx-dependent peroxidase (TPx) are needed to form a disulfide bond within the GPX monomer in vivo, suggesting that the first Cys func- tions as a peroxidatic site attacked by hydroperoxides to produce H 2 O and Cys-SOH which then form an in- tramolecular disulfide bond with the third Cys [an atypical 2 Cys peroxiredoxin (Prx)-type reaction] [11,16]. The process for reduction of the disulfide bond Fig. 2. Alignment of the amino acid sequence of the GPX proteins. The deduced amino acid sequences aligned for comparison are from A. thaliana (AtGPX1–8), Helianthus annuus (HaPhGPX), Lycopersicon esculentum (LePhGPX), Chinese cabbage (PHCC-TPx), Setaria italica (SiPhGPX), Raphanus sativus (RsPhGPX), S. cerevisciae (yeast GPX2) and non-SeCys type GPX from human and mouse. The three con- served Cys of these GPX isoenzymes are shown by inverted triangles. The GPX conserved regions are shown by a horizontal line below the amino acid residues and represented by I, II and III. GenBank TM accession numbers for AtGPX1, -2, -5 and -6 are shown in Fig. 1. The acces- sion numbers of other isoenzymes are AtGPX3 (NP_181863), AtGPX4 (NP_566128), AtGPX7 (NP_194915), AtGPX8 (NP_176531), HaPhGPX (CAA75009), LePhGPX (CAA75054), PHCC-TPx (AF411209), RsPhGPX (AF322903), SiPhGPX (AAS47590), yeast GPX2 (NP_009803), human NPGPx (BC032788) and mouse NPGPx (BC003228). A. Iqbal et al. A. thaliana glutathione peroxidases FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS 5593 in target proteins by the reduced Trx has already been elucidated [32,33]. The Arabidopsis and other plant GPX isoenzymes also contained the same Cys arrange- ment and thus may possess a similar mode of action for the reduction of hydroperoxide utilizing Trx (Fig. 2). K m values for both Trx and H 2 O 2 of AtGPX isoen- zymes were nearly similar to those of GPXs from sev- eral plants and yeast (Tables 2 and 3). K m values for AtGPXs for Trx were in lm, which were compatible with Trx levels in vivo and thus within physiological limits. However, the TPx activities with H 2 O 2 or alkyl hydroperoxides of Arabidopsis, sunflower and tomato GPXs were significantly lower than those of yeast GPX2 and Chinese cabbage TPx (Tables 2 and 3). There are no significant differences between these enzymes with respect to the arrangement of the well- conserved secondary structural elements, that is, the existence of three Cys residues and deletion of the region for tetramerization (Fig. 2). Accordingly, it is necessary to undertake a more detailed investigation to clarify why Arabidopsis GPX isoenzyme has low TPx activity with H 2 O 2 or alkyl hydroperoxides. Tomato and sunflower GPX isoenzymes had approximately similar reduction activity with H 2 O 2 , alkyl and lipids hydroperoxide in the presence of Trx. In contrast, those from Arabidopsis and Chinese cab- bage, and yeast GPX2 had higher activity with H 2 O 2 than with the other hydroperoxides (Table 1), indica- ting a gradual evolution toward a more specialized function of H 2 O 2 reduction [11,16,17]. Even though the GPX isoenzymes from photosynthetic organisms and Prx are weakly homologous at the amino acid level, substrate specificity, reaction mechanism, affinity and catalytic efficiency toward various hydroperoxides are very similar [36]. As the three conserved Cys resi- dues appear to be essential for the thioredoxin-depend- ent reduction activity [11,16], it is suggested that GPX isoenzymes from photosynthetic organisms may be classified as a distinct group of 3-Cys peroxiredoxins (Fig. 2). Although plant GPX isoenzymes were able to utilize E. coli Trx as a substrate in vitro and have been identi- fied as potential Trx targets in different proteomic approaches [34], AtGPX isoenzymes were not able to reduce hydroperoxides with the two cytosolic Trxs h2 and h3 from Arabidopsis [21]. The two mitochondrial Trxs, poplar PtTrxh2 and Arabidopsis AtTrxo1, could not serve as electron donors to either the mitochond- rial PtGPX3 or cytosolic PtGPX1 and -5 of poplar, whereas the cytosolic Trx isoforms, PtTrxh1 and PtTrxh3, could reduce them [22]. It has been reported that various types of Trx isoforms show considerable differences in structure and electrostatic potentials around the redox active site [35]. These findings sug- gest that plant Trxs show functional redundancy and a high degree of specificity toward target proteins [35]. Although we checked hydroperoxide reduction activity in only two cytosolic Trx isoforms from Arabidopsis,it is possible that AtGPX isoenzymes could utilize a par- ticular Trx with the respective subcellular distribution or some other unidentified reductant, as reported pre- viously [22]. Accordingly, it is imperative to identify the in vivo reductant for each GPX isoenzyme because the various thiol buffers, including Trx, can affect a number of redox reactions in the cells. Thiol-dependent redox regulation is more diverse in plants than in animals, bacteria or fungi [36]. Thiol– disulfide exchange reactions, which are rapid and read- ily reversible, are ideally suited to controlling protein function via the redox state of structural or catalytic SH groups [37–39]. The occurrence of thiol-dependent activities in the GPX isoenzymes suggests that they may be involved in redox modification and signal transduction in plant cells. Anti-apoptotic activities have been reported for mitochondrial PhGPX [40] and tomato GPX1 [41]. Similarly, it has been reported that tomato plants overexpressing an eukaryotic selenium- independent GPX (GPX5) maintained a significantly higher photosynthesis rate and fructose-1,6-bisphos- phatase activity under chilling stress, because of the sustenance of the cellular redox homeostasis [42]. Fur- thermore, plant GPX isoenzymes are induced to remarkable levels by various stress conditions surmi- sing involvement in defense [6,23,43]. Overexpression of Chlamydomonas GPX in tobacco plants either in the cytosol (TcGPX) or chloroplasts (TpGPX) also resulted in the maintenance of a higher photosynthetic capacity and increased tolerance to various abiotic stresses [44]. Recently, it has been reported that the atgpx3 mutation in Arabidopsis disrupted abscisic acid (ABA) activation of calcium channels and the expres- sion of ABA and stress-responsive genes [26]. GPXs may thus work in tandem with peroxiredoxins, the other antioxidant enzymes utilizing Trx, to detoxify H 2 O 2 and organic hydroperoxides and also be involved in the regulation of the redox homeostasis by main- taining the thiol ⁄ disulfide or NADPH ⁄ NADP balance. Experimental methods Plant materials and chemicals NADPH, H 2 O 2 , GSH, E. coli Trx and Trx reductase were purchased from Sigma Aldrich (St Louis, MO). Hydroper- oxides of unsaturated fatty acids were prepared by oxygen- A. thaliana glutathione peroxidases A. Iqbal et al. 5594 FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS ation with 2,2¢-azobis (2-amidinopropane) dehydrochloride and purified using RP-HPLC in the mobile phase as repor- ted previously [45]. The amount of hydroperoxide was calculated from the UV absorption at 234 nm (e ¼ 27400 m )1 Æcm )1 ). Arabidopsis plants (ecotype Columbia) were grown for 15 days in MS medium containing 3% sucrose under a white light of  75–100 lEÆm )2 Æs )1 at 25 °C with a 16 h light period. Cloning of AtGPX isoenzymes Total RNA was extracted from 15-day-old Arabidopsis seed- lings as described previously [46]. First-strand cDNA was synthesized using oligo(dT 20 ) primer and RevTra Ace (reverse transcriptase; Toyobo, Osaka, Japan) with 5 lg total RNA as a template according to the manufacturer’s proto- col. Full-length cDNAs encoding mature AtGPX1 (CAT ATGGCTGCAGAAAAAACCG ⁄ GGATCCTTTAAGCG GCAAGCAA), AtGPX2 (CATATGGCGGATGAATC TCCAA ⁄ GGATCCTCCTCCTCCTGGTGAT), AtGPX5 (CATATGGGTGCTTCATCATCAT ⁄ GG ATCCCTGTG CGTGTTCACAA) and AtGPX6 (CATATGGC AGC AGAGAAGTCTG ⁄ GGATCCAGTTATCCAGATTGAA) without transit peptides or Trx h2 (CATATGGGA GGAGCTTTATC ⁄ GGATCCGCGTTAACAATGCTCA) and Trx h3 (CATATGGAAGAGAAGCCGCA ⁄ GGATCC AAATCAAGCAGCAGC) proteins were amplified by RT- PCR with chimeric primers (in parenthesis) introducing the NdeI ⁄ BamHI sites into the respective forward and reverse primers (bold sequences). The amplification conditions were 30 cycles at 94, 58 and 72 °C for 1 min each and final exten- sion at 72 °C for 10 min. The PCR product was gel purified using GFX TM PCR Gel Band Purification kit (GE Health- care, Chalfont St Giles, UK), ligated into pT7 vector (Nov- agen, EMD Bioscience Inc., La Jolla, CA) and sequenced using the dideoxy chain terminator method with an automa- tic DNA sequencer (ABI PRISM TM 310; Applied Biosys- tems). The NdeI ⁄ BamHI fragment was obtained and subsequently ligated into pColdII vector (Takara, Kyoto, Japan) giving an in-frame fusion with His 6 tag. After con- firming sequence using the same procedure mentioned above, E. coli BL21 (DE3) pLysS (Promega Corp., Madison, WI) cells were transformed and used for the protein expression. Heterologous expression and purification of A. thaliana GPX1, -2, -5, -6 and AtTrx h2 and h3 Escherichia coli BL21 (DE3) pLysS cells containing the pos- itive clones for AtGPX1, -2, -5 and -6, AtTrx h2 and h3 were grown in 50 mL LB medium, containing 50 lgÆmL )1 of ampicillin and 34 lgÆmL )1 of chloramphenicol at 37 °C until the D 600 of the culture is between 0.4 and 0.6. Cells were kept at 15 °C for 30 min without shaking and the fusion protein expression was induced by the addition of 0.4 mm isopropyl-thio-b-d-galactoside. Cells were further grown for 24 h at 15 °C with shaking, collected by centrifu- gation at 3000 g for 10 min with a himac CR21 centrifuge (Hitachi, Tokyo, Japan; rotor type R20A2), resuspended in 100 mm Tris ⁄ HCl buffer, pH 8.0, and disrupted by sonica- tion at 10 kHz for a total of 2 min (five intervals of 40 s each). Soluble proteins were collected by centrifugation at 15 000 g for 15 min at 4 °C (himac CR21 centrifuge, rotor type R20A2) and purified using a HiTrap TM chelating HP column (Amersham Biosciences, Uppsala, Sweden) accord- ing to the manufacturer’s protocol. Briefly, the column was washed with 10 mL of distilled water, charged with 0.5 mL of 0.1 m NiCl 2 and washed again with 5 mL of distilled water. After preparation, the column was equilibrated with 10 vol of binding buffer (20 mm Tris ⁄ HCl, pH 8.0, 0.5 m NaCl, 5 mm imidazole and 1 mm 2-mercaptoethanol). The sample was applied using a syringe and the column was washed with 10 vol of binding buffer containing 20 mm im- idazole. The recombinant protein was eluted with 5–10 vol of the binding buffer containing 500 mm imidazole, and collected in a 1-mL fraction. Protein content was deter- mined according to the method of Bradford [47]. The reduced forms of recombinant Trxs were confirmed by determining the reduction of 5,5¢-dithiobis(2-nitrobenzoic acid) (Nbs 2 ) in the presence of E. coli Trx reductase. One millilitre of assay mixture contained 100 mm Tris ⁄ HCl, pH 7.5, 5 mm EDTA, 0.2 mm NADPH, 0.6 mm Nbs 2 and the recombinant proteins. The reaction was started by the addition of 0.5 U E. coli Trx reductase and the reduction of Nbs 2 was monitored at 412 nm [e 412 (Nbs 2 ) ¼ 13 600 m )1 Æcm )1 ] in a spectrophotometer (data not shown) [48]. Enzymatic assays GPX activity with GSH and H 2 O 2 or hydroperoxides was assayed spectrophotometrically at 340 nm by the decrease in the absorbance due to the conversion of NADPH in the presence of GSH reductase, which catalyzes the reduc- tion of oxidized GSH formed by GPX [14]. The reaction mixture contained 100 mm Tris ⁄ HCl, pH 7.5, 1 mm GSH, 0.2 mm NADPH, 0.1 mm H 2 O 2 or hydroperoxides, 5 mm EDTA, 1 U GSH reductase and the enzyme in a total vol- ume of 0.5 mL. Reduction of hydroperoxides was meas- ured in the same assay mixture, except that H 2 O 2 was replaced by 0.1 mm cumene, unsaturated fatty acid or lipid hydroperoxides. Enzymes activities were calculated using an e-value of 6220 m )1 Æcm )1 . Trx- and NADPH- dependent reduction activities were measured in a manner similar to that described previously [15,17]. The GSH and GSH reductase in the reaction mixture mentioned above were replaced with E. coli Trx and Trx reductase (0.3 UÆmL )1 )orArabidopsis Trx h2 and h3 (4 lm). NADPH-dependent Trx peroxidase activity was measured in the same way as the GSH-dependent peroxidase activity mentioned above. A. Iqbal et al. A. thaliana glutathione peroxidases FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS 5595 Acknowledgements We thank Miss Aya Muramoto, Mr Takahisa Ogawa and Mr Noriaki Tanabe for their excellent technical assistance and expert advice during the course of this experiment. This work was supported by CREST, JST (S.S: 2005-10) and by the ‘Academic Frontier’ Project for Private Universities: matching fund subsidy from MEXT (S.S: 2004-08). References 1 Imlay JA (1988) DNA damage and oxygen radical tox- icity. Science 240, 1302–1309. 2 Foyer CH, Valadier MH & Ferrario S (1994) Co-regula- tion of nitrogen and carbon assimilation in leaves. In Environment and Plant Metabolism, Flexibility and Accli- mation (Smirnoff N, ed.), pp. 17–33. Bios Scientific, Guildford. 3 Mittler R (2002) Oxidative stress, antioxidants, and stress tolerance. Trends Plant Sci 9, 405–410. 4 Ursini F, Maiorino M, Brigelius-Flohe ´ R, Aumann KD, Roveri A, Schomburg D & Flohe ´ L (1995) Diversity of glutathione peroxidases. Methods Enzymol 252, 38–53. 5 Criqui MC, Plesse B, Durr A, Marbach J, Parmentier Y, Jamet E & Fleck J (1992) Characterization of genes expressed in mesophyll protoplasts of Nicotiana sylves- tris before the re-initiation of the DNA replicational activity. Mech Dev 38, 121–132. 6 Holland D, Ben-Hayyim G, Faltin Z, Camoin L, Stros- berg AD & Eshdat Y (1993) Molecular characterization of salt-stress-associated protein in citrus: protein and cDNA sequence homology to mammalian glutathione peroxidase. Plant Mol Biol 21, 923–927. 7 Sugimoto M & Sakamoto W (1997) Putative phospholi- pids hydroperoxide glutathione peroxidase gene from Arabidopsis thaliana induced by oxidative stress. Genes Genet Syst 72, 311–316. 8 Roeckel-Drevet P, Gango G, deLabrouhe T, Dufaure JP, Nicolas P & Drevet JR (1998) Molecular characteri- zation, organ distribution and stress-mediated induction of two glutathione peroxidase-encoding mRNAs in sun- flower (Helianthus annuus). Physiol Plant 103, 385–394. 9 Depege N, Drevet J & Boyer N (1998) Molecular clon- ing and characterization of tomato cDNAs encoding glutathione peroxidase-like proteins. Eur J Biochem 253, 445–451. 10 Mullineaux PM, Karpinski S, Jimenez A, Cleary SP, Robinson C & Creissen GP (1998) Identification of cDNAS encoding plastid-targeted glutathione peroxi- dase. Plant J 13, 375–379. 11 Jung BG, Lee KO, Lee SS, Chi YH, Jang HH, Kang SS, Lee K, Lim D, Yoon SC & Yun DJ (2002) A Chinese cabbage cDNA with high sequence identity to phospholipids hydroperoxide glutathione peroxidase encodes a novel isoform of thioredoxin-dependent per- oxidase. J Biol Chem 277, 12572–12578. 12 Xiao-Dong Y, Jun LW & Yuan LJ (2005) Isolation and characterization of a novel PhGPx gene in Raphanus sativus. Biochem Biophys Acta 1728, 199–205. 13 Leisinger U, Ru ¨ fenacht K, Zehnder AJB & Eggen RIL (1999) Structure of a glutathione peroxidase homolo- gous gene involved in the oxidative stress response in Chlamydomonas reinhardtii . Plant Sci 149, 139–149. 14 Takeda T, Nakano Y & Shigeoka S (1993) Effects of selenite, CO 2 and illumination on the induction of sele- nite-dependent glutathione peroxidase in Chlamydomo- nas reinhardtii. Plant Sci 94, 81–88. 15 Gaber A, Tamoi M, Takeda T, Nakano Y & Shigeoka S (2001) NADPH-dependent glutathione peroxidase-like proteins (Gpx-1, Gpx-2) reduce unsaturated fatty acid hydroperoxides in Synechocystis PCC 6803. FEBS Lett 499, 32–36. 16 Tanaka T, Izawa S & Inoue Y (2005) GPX2, encoding a phospholipid hydroperoxide glutathione peroxidase homologue, codes for an atypical 2-Cys peroxiredoxin in Saccharomyces cerevisiae. J Biol Chem 280, 42078–42087. 17 Herbette S, Lenne C, Leblanc N, Julien JL, Drevert JR & Drevert PR (2002) Two GPX-like proteins from Lycopersicon esculentum and Helianthus annuus are anti- oxidant enzymes with phospholipid hydroperoxide glu- tathione peroxidase and thioredoxin peroxidase activities. Eur J Biochem 269, 2414–2420. 18 Das KC & White CW (2002) Redox systems of the cell: possible links and implications. Proc Natl Acad Sci USA 99, 9617–9618. 19 Grene R (2002) Oxidative stress and acclimation mecha- nisms in plants. In The Arabidopsis Book (Somerville CK & Meyerwitz EM, eds), pp. 1–19. American Society of Plant Biologists, Rockville, MD. 20 Buchanan BB, Schu ¨ rmann P, Wolosiuk RA & Jacquot JP (2002) The ferredoxin ⁄ thioredoxin system: from dis- covery to molecular structures and beyond. Photosynth Res 73, 215–222. 21 Schu ¨ rmann P & Jacquot JP (2000) Plant thioredoxin system revisited. Annu Rev Plant Physiol Plant Mol Biol 51, 371–400. 22 Gelhyde E, Rouhier N, Ge ´ rard J, Jolivet Y, Gualberto J, Navrot N, Ohlsson PI, Wingsle G, Hirasawa M, Knaff DB et al. (2004) A specific form of thioredoxin h occurs in plant mitochondria and regulates alternative oxidase. Proc Natl Acad Sci USA 101, 14545–14550. 23 Rodriguez Milla MA, Maurer A, Rodriguez Huete A & Gustafson JP (2003) Glutathione peroxidase genes in Arabidopsis are ubiquitous and regulated by abiotic stresses through diverse signaling pathways. Plant J 36, 602–615. 24 Jean-Benoit P, Ytterberg AJ, Sun Q & Klaas JW (2004) New functions of the thylakoid membrane proteome of A. thaliana glutathione peroxidases A. Iqbal et al. 5596 FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS Arabidopsis thaliana revealed by a simple, fast, and ver- satile fractionation strategy. J Biol Chem 279, 49367– 49383. 25 Myriam F, Salvi D, Brugie ´ re S, Miras S, Kowalski S, Louwagie M, Garin J, Joyard J & Rolland N (2003) Proteomics of the chloroplast envelope membranes from Arabidopsis thaliana. Mol Cell Proteomics 2, 325–345. 26 Miao Y, Lv D, Wang P, Wang XC, Chen J, Miao C & Song CP (2006) An Arabidopsis glutathione peroxidase functions as both a redox transducer and a scavenger in abscisic acid and drought stress responses. Plant Cell, doi:10.1105 ⁄ tpc.106.044230. 27 Mairino M, Aummann KD, Brigelius-Fohle ´ R, Doria D, Van den Heuvel J, McCarthy J, Roveri A, Ursini F & Fohle ´ L (1995) Probing the presumed catalytic triade of selenium-containing peroxidases by mutational analy- sis of phospholipid hydroperoxide glutathione peroxi- dase (PhGPx). Biol Chem Hoppe-Seyler 376, 651–660. 28 Wilkinson SR, Meyers DJ & Kelly JM (2000) Biochem- ical characterization of trypanosome enzyme with glu- tathione dependent peroxidase activity. Biochem J 352, 755–761. 29 Avery AM & Avery SV (2001) Saccharomyces cerevisiae expresses three phospholipid hydroperoxide glutathione peroxidases. J Biol Chem 276, 33730–33735. 30 Sztajer H, Gamain B, Aumann KD, Slomainny C, Becker K, Brigelius-Folhe R & Folhe L (2001) The putative glutathione peroxidase gene of Plasmodium fal- ciparum codes for a thioredoxin peroxidase. J Biol Chem 276, 7397–7403. 31 Ahmad U, Xianzhi J, Saori F, Jeanho Y, David LS, Yi-Chun JW, Kartiki VD, Jeffrey EJ, Phang-Lang C & Wen-Hwa L (2004) Identification of a novel putative non-selenocysteine containing phospholipid hydroperoxide glutathione peroxidase (NPGPx) essential for alleviating oxidative stress generated from polyunsa- turated fatty acids in breast cancer cells. J Biol Chem 279, 43522–43529. 32 Holmgren A (1985) Thioredoxin. Annu Rev Biochem 54, 37–271. 33 Holmgren A (1989) Thioredoxin and glutaredoxin sys- tems. J Biol Chem 264, 13963–13966. 34 Laloi C, Rayapuram N, Chartier Y, Grienberger JM, Bonnard G & Meyers Y (2001) Identification and char- acterization of a mitochondrial thioredoxin system in plants. Proc Natl Acad Sci USA 98, 14144–14149. 35 Collin V, Bourguet EI, Marchand C, Hirasawa M, Lancelin JM, Knaff DB & Maslow MM (2003) Arabi- dopsis plastidial thioredoxins; new function and new sight into specificity. J Biol Chem 278, 23747–23752. 36 Dietz KJ, Jacob S, Oelze ML, Laxa M, Tognetti V, de Miranda SM, Baier M & Finkemeier I (2006) The function of peroxiredoxins in plant organelle redox metabolism. J Exp Bot 57, 1697–1709. 37 Laughner BJ, Sehnke PC & Ferl RJ (1998) A novel nuclear member of the thioredoxin superfamily. Plant Physiol 118, 987–996. 38 Motohashi K, Kondoh A, Stumpp MT & Hisabori T (2001) Comprehensive survey of proteins targeted by chloroplast thioredoxin. Proc Natl Acad Sci USA 98, 11224–11229. 39 Laloi C, Dominique MO, Marco Y, Meyers Y & Reichheld JP (2004) The Arabidopsis h5 gene induction by oxidative stress and its W-box mediated response to pathogen elicitor. Plant Physiol 134, 1006–1016. 40 Nomura K, Imai H, Koumura T, Arai M & Nakagawa Y (1999) Mitochondrial phospholipid hydroperoxide glutathione peroxidase suppresses apoptosis mediated by a mitochondrial death pathway. J Biol Chem 274, 29294–29302. 41 Chen S, Vaghchhipawala Z, Li W, Asard H & Dickman MB (2004) Tomato phospholipid hydroperoxide glu- tathione peroxidase inhibits cell death induced by Bax and oxidative stresses in yeast and plants. Plant Physiol 135, 1630–1641. 42 Herbette S, Menn AL, Rousselle P, Ameglio T, Faltind Z, Branlard G, Eshdat Y, Julien JL, Drevet JR & Dre- vet PR (2005) Modification of photosynthetic regulation in tomato overexpressing glutathione peroxidase. Biochim Biophys Acta 1724, 108–118. 43 Churin Y, Schilling S & Borner T (1999) A gene family encoding glutathione peroxidase homologues in Hor- dium vulgare. FEBS Lett 459, 33–38. 44 Yoshimura K, Miyao K, Gaber A, Takeda T, Kanabo- shi H, Miyasaka H & Shigeoka S (2004) Enhancement of stress tolerance in transgenic tobacco plants over expressing Chlamydomonas glutathione peroxidase in cytosol or chloroplast. Plant J 37, 21–33. 45 Funk CD, Gunne H, Steiner H, Izumi T & Samuelsson B (1989) Native and mutant 5-lipoxygenase expression in baculovirus ⁄ insect cell system. Proc Natl Acad Sci USA 86, 2592–2596. 46 Ogawa T, Ueda Y, Yoshimura K & Shigeoka S (2005) Comprehensive analysis of cytosolic Nudix hydrolases from Arabidopsis thaliana. J Biol Chem 280, 25277– 25283. 47 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein util- izing the principle of protein–dye binding method. Anal Biochem 717, 1448–1454. 48 Pasternak C, Haberzettl K & Klug G (1999) Thiore- doxin is involved in oxygen-regulated formation of the photosynthetic apparatus of Rhodobacter sphaeroides. J Bacteriol 181, 100–106. A. Iqbal et al. A. thaliana glutathione peroxidases FEBS Journal 273 (2006) 5589–5597 ª 2006 The Authors Journal compilation ª 2006 FEBS 5597 . Hydroperoxide reduction by thioredoxin-specific glutathione peroxidase isoenzymes of Arabidopsis thaliana Aqib Iqbal 1 , Yukinori. homeostasis by maintaining the thiol ⁄ disulfide or NADPH ⁄ NADP balance. Abbreviations ABA, abscisic acid; AtGPX, Arabidopsis thaliana homolog of glutathione peroxidase;