![]()
|
|
||||||||
J. Biol. Chem., Vol. 281, Issue 36, 26280-26288, September 8, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1


¶||
From the
Plant Physiology Group, United States Department of Agriculture/Agricultural Research Service Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Texas 77030,
Plant Biology Division, Samuel Roberts Noble Foundation, Inc., Ardmore, Oklahoma 73401, the ¶Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas 77030, and the ||Vegetable and Fruit Improvement Center, Texas A & M University, College Station, Texas 77845
Received for publication, February 13, 2006 , and in revised form, June 16, 2006.
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Of those antioxidant systems, the physiological roles of thioredoxins have been intensively studied (10), whereas those of Grxs have not been fully defined (11, 12). Grxs are ubiquitous small heat-stable disulfide oxidoreductases, which are conserved in both prokaryotes and eukaryotes (11, 13). Through an active motif, namely the conserved CPYC sequence (a dithiol Grx), they catalyze the reduction of protein disulfides and of GSH-protein mixed disulfides via a dithiol or monothiol mechanism (14, 15). In bacterial, yeast, and mammalian cells, dithiol Grxs appear to be involved in many cellular processes and play an important role in protecting cells against oxidative stresses (16-18).
Besides the dithiol Grxs, a new group of monothiol Grxs has recently been identified in yeast (Grx3, -4, and -5) and bacteria (Grx4) that have a single cysteine residue in the putative active motif (19, 20). Yeast Grx5 encodes a mitochondrial monothiol Grx, which is required for biogenesis of iron-sulfur clusters, whereas Grx3 and Grx4 function in detoxification of cytotoxin and cell proliferation in yeast (21-23). Interestingly, bacterial Grx4, unlike other previously characterized Grxs, can serve as a substrate for thioredoxin reductase instead of NADPH/glutathione reductase (20), suggesting that those monothiol Grxs have distinct functions. This group of monothiol Grxs is also conserved across organisms and has now been identified in malarial parasites, plants, zebrafish, mice, and humans (24-27). Recent studies also indicate that those Grxs contain a newly identified protein kinase C-interacting cousin of thioredoxin homology domain (PICOT-HD) in their carboxyl-terminal regions (24, 28). However, ascertaining a unifying function of PICOT-HD Grxs has been problematic.
Photosynthetic organisms, particularly higher plants, have a large Grx gene family; however, until recently, only a few plant Grxs have been studied (12, 29-30). In the Arabidopsis genome, there are at least 31 open reading frames coding for putative Grxs, which can be divided into three major classes that include the aforementioned monothiol Grxs (12, 30). Genetic analysis of a CC type Grx, ROXY1, indicates an important role of this protein in floral petal development (31). In a previous study, we used a yeast functional screen to identify a PICOT-HD-containing protein that was able to activate Arabidopsis CAX1 Ca2+/H+ antiport activity in a yeast expression system; this gene was originally termed CXIP1 (26). Here we determine the subcellular localization of this first cloned plant PICOT-HD-containing protein and reclassify the gene as AtGRXcp. We functionally characterize the protein and perform initial structure-function studies using yeast expression assays. We go on to isolate AtGRXcp knock-out mutants and describe the phenotypes of these plants at the whole plant and biochemical levels. For the first time, we demonstrate a role for a plant monothiol Grx.
| EXPERIMENTAL PROCEDURES |
|---|
|
|
|---|
Plant Growth ConditionsArabidopsis wild type (ecotype Columbia, Col-0) and atgrxcp mutant seeds were surface-sterilized, germinated, and grown on one-half strength Murashige and Skoog medium (33) solidified with 0.8% agar and the same medium supplemented with various concentrations of H2O2. Iron-sufficient and -deficient media were made following a published protocol (34).
DNA Constructs and Site-directed MutagenesisYeast Grx5 was amplified by PCR using a forward primer (5'-GCC GGA TCC ATG TTT CTC CCA AAA TTC AAT-3') and a reverse primer (5'-CCG GAG CTC TCA ACG ATC TTT GGT TTC TTC-3'), and the PCR products were cloned into pGEM-T Easy (Promega, Madison, WI). The full-length cDNA of AtGRXcp was isolated through a yeast functional screen and originally termed CXIP1 (for CAX1-interacting protein 1) (26). AtGRXcp was predicted to have a 63-amino acid signal peptide by analysis with the Chloro P (version 1.1) program (available on the World Wide Web at www.cbs.dtu.dk/services/ChloroP/). To remove this N-terminal signal peptide, a truncated form of AtGRXcp was amplified by PCR using a forward primer (5'-GGG CTC GAG AGA TCT GCG ATG GCG TCG GCT CTT ACG CCG-3') and the AtGRXcp reverse primer. Site-directed mutagenesis was performed as described previously (35). A forward primer (5'-GAA TCC CGT CTC CCC ATG GCT GGA TTC TCC AAC ACT GTG GTT CAG ATT TTG-3') and a reverse primer (CGFS; 5'-GAA TTC CGT CTC CAT GGG GAA GTC TCT CGT TCC TTT C-3') were used for creating the C97A mutation. A forward primer (5'-GAA TCC CGT CTC CCG ATG TGT GGA GCA TCC AAC ACT GTG GTT CAG ATT TTG-3') and the reverse primer CGFS were used for creating the F99A mutation. The fidelity of all clones was confirmed by sequencing.
Yeast Strains, Expression Constructs, and Growth AssaysSaccharomyces cerevisiae wild type strain CML235 (MATa ura3-52 leu2
1 his3
200), grx5 (MATa ura3-52 leu2
1 his3
200 grx5::kanMX4) were provided by Dr. Enrique Herrero (Universitat de Lleida, Lleida, Spain) and used in all yeast experiments. Yeast Grx5 and AtGRXcp and its variants were cloned into piUGpd (36). Yeast cells were transformed by using the LiOAc method (37). All yeast strains were assayed on YPD medium (yeast peptone dextrose, rich medium), with or without various concentrations of H2O2, and SC medium plus six amino acids or five amino acids without lysine (21, 26).
Localization of AtGRXcp-Green Fluorescent Protein (GFP) Fusions in Yeast and Plant CellsFull-length and truncated AtGRXcp and its variants were fused to the N terminus of green fluorescent protein (GFP) using a procedure described previously (38). The GFP constructs were subcloned into yeast and plant expression vectors as described previously (38). The subcellular localization of the fused proteins was imaged in comparison with labeled organelle markers (chloroplasts, mitochondria, Golgi, and peroxisome) as described previously (38). A peroxisome-targeted DsRed (red fluorescent protein), DsRed-per, was constructed by adding the plant peroxisome-targeting signal, KSRM, to the end of DsRed (39). The fluorescence signals were detected at 510 nm (excitation at 488 nm) for GFP, at 582 nm (excitation at 543 nm) for DsRed, and at 660 nm (excitation at 633 nm) for chlorophyll using Leica TCS SP2 AOBS confocal laser-scanning microscope. The fluorescence intensities were quantified by using the LCS software.
AtGRXcp::GUS Transgenic PlantsA 397-bp DNA sequence upstream of ATG of AtGRXcp open reading frames was amplified from genomic DNA by using the following primer sets: forward primer (5'-GGC AAG CTT ATA AGT TTT AAT CGT TTA TGG GGT-3') and reverse primer (5'-GCC TCT AGA TTT TGA CGA CTT TTA GAT TTG GAA-3'). The PCR fragment was cloned into pBI121 to replace the 35S CaMV promoter, resulting in plasmid pAtGRXcp::GUS. Agrobacterium transformation of Arabidopsis plants was performed as described previously (40). More than 50 T2 generation plants were selected for Kan resistance.
Protein Oxidation AnalysisCarbonyl assays for analysis of oxidized proteins in both yeast and plant cells were performed as previously described (19, 41, 42). Yeast cultures of the strains (CML235 and grx5) expressing vector, Grx5, and AtGRXcp were grown in YPD media overnight at 30 °C. Half of the culture of each strain was subjected to treatment with 5 mM H2O2 for 1 h. Total proteins were extracted from cells with and without treatment. Western blot analysis was used to determine carbonyl group content. Arabidopsis chloroplasts were isolated from photosynthetic tissues of 6-week-old flowering wild type-, atgrxcp-, and AtGRXcp-overexpressing plants (43). The oxidized proteins were detected by protein gel blotting using anti-dinitrophenylhydrazone antibody (42) (Bethyl Laboratory, Montgomery, TX).
|
| RESULTS |
|---|
|
|
|---|
AtGRXcp Is a Chloroplast-localized Monothiol GrxMonothiol Grxs have a diverse subcellular distribution in multiple organisms. For example, yeast Grx5 is mitochondria-localized (21), and Grx3 targets to nuclei (45), whereas a human PICOT is located in the cytosol (24). In order to gain insight into the function of AtGRXcp, we fused it with GFP at its C terminus and transiently expressed this fusion protein in tobacco leaf cells. Using various organelle markers, AtGRXcp-GFP was shown to clearly target to chloroplasts rather than mitochondria, Golgi, or peroxisomes in mesophyll cells (Fig. 2A). Analysis of the AtGRXcp sequence predicts that a 63-amino acid signal peptide is present at the N terminus (data not shown). To experimentally verify this, we removed this putative signal peptide and fused the truncated AtGRXcp (AtGRXcp
63) with GFP for transient expression in tobacco cells. As shown in Fig. 2B, AtGRXcp
63-GFP no longer targeted to chloroplasts and instead was dispersed throughout the cytosol and nuclei, similar to observations with free GFP (Fig. 2B, data not shown).
AtGRXcp Suppresses the Sensitivity of a Yeast grx5 Mutant to H2O2Yeast grx5 cells are growth-impaired in minimal medium and sensitive to oxidative stresses (19, 21). To examine if AtGRXcp could restore Grx5 function and suppress the sensitivity of grx5 cells to an external oxidant, H2O2, we expressed vector control, AtGRXcp, and yeast endogenous Grx5 in grx5 cells. All yeast strains grew normally in YPD liquid media (rich media) after 48 h of growth (Fig. 3A). Whereas vector-expressing grx5 cells were growth-impaired in the medium with 3 mM H2O2, both AtGRXcp- and Grx5-expressing grx5 cells grew in a similar manner to wild type cells (Fig. 3A). These observations suggest that AtGRXcp is able to suppress the sensitivity of grx5 cells to oxidative stress.
|
63-GFP was unable to target to this organelle (Fig. 3C). Targeting of AtGRXcp to mitochondria was essential for the function of this protein in yeast, since both AtGRXcp
63 and AtGRXcp
63-GFP were unable to suppress the sensitivity of grx5 cells to H2O2 (Fig. 3, A and B). AtGRXcp Is Able to Protect Cells against Protein Oxidation and Rescue the Lysine Auxotrophy of a Yeast grx5 MutantIn grx5 cells, total protein carbonyl content is significantly increased under oxidative stress (Fig. 4A), suggesting that yeast Grx5 plays a vital role in directly protecting enzymes from oxidative damages (19, 21). In order to determine if AtGRXcp could directly reduce protein carbonylation in the grx5 cells, we performed Western blot analysis of total proteins isolated from vector-, AtGRXcp- and Grx5-expressing grx5 cells grown in H2O2-containing YPD media. Oxidized protein content in AtGRXcp-expressing cells was reduced compared with the vector-expressing cells but similar to Grx5-expressing cells (Fig. 4A). These results demonstrate that AtGRXcp can protect cells against protein oxidative damages.
Yeast Grx5 is a mitochondrial Grx required for the maturation of Fe-S clusters (21, 46). Deletion of Grx5 results in inactivation of the mitochondrial Fe-S enzyme homoaconitase, which is involved in lysine synthesis (21, 27). Previous work details that grx5 cells fail to grow on lysine-deficient medium (21) (Fig. 4B). Expression of AtGRXcp rescued the lysine auxotrophy of grx5 cells, although AtGRXcp suppression was less efficient in comparison with Grx5-expressing grx5 cells (Fig. 4B). These results again indicate that AtGRXcp can partially restore Grx5 function in yeast cells.
|
|
Ala change was termed AtGRXcp-AGFS, and Phe99
Ala change was termed AtGRXcp-CGAS). We performed the aforementioned liquid growth assays to examine if these mutations in the conserved motif would alter AtGRXcp function. Yeast cells (grx5) expressing AtGRXcp-AGFS (AtGRXcp-AGFS-GFP) or AtGRXcp-CGAS (AtGRXcp-CGAS-GFP) grew similarly in YPD liquid media in comparison with vector- and AtGRXcp-expressing grx5 cells (Fig. 5A, data not shown). However, under oxidative stress, AtGRXcp-CGAS-(AtGRXcp-CGAS-GFP-) and AtGRXcp-expressing grx5 cells grew significantly better than vector-expressing grx5 cells, whereas AtGRXcp-AGFS-(AtGRXcp-AGFS-GFP-) expressing grx5 cells grew at a slower rate than vector-expressing cells (Fig. 5A, data not shown). To determine if these amino acid changes (Cys97
Ala and Phe99
Ala) affect protein expression and stability, we performed Western blots to detect AtGRXcp-GFP and the variants among yeast total proteins. Indeed, AtGRXcp-AGFS-GFP was rarely detectable, whereas AtGRXcp-CGAS-GFP levels were similar to wild type AtGRXcp (Fig. 5B). Those observations suggest that the Cys97
Ala change may affect protein expression, which results in the inability of AtGRXcp-AGFS to suppress the sensitivity of grx5 cells to H2O2. We can infer from these yeast assays that the Phe99 in the CGFS motif is not essential for AtGRXcp function.
|
AtGRXcp Is Expressed in Cotyledon, Leaves, Vascular Tissues, and FlowersPreviously, RNA blot analysis indicated that AtGRXcp is ubiquitously expressed in Arabidopsis plants with high levels of AtGRXcp mRNA accumulating in green leaves (26). To further determine spatial and temporal AtGRXcp expression, the 397-bp AtGRXcp promoter was cloned and transcriptionally fused with the
-glucuronidase gene (GUS) and then transformed into Arabidopsis plants. More than 50 independent transgenic lines were generated. Preliminary GUS staining indicated that all transgenic lines harboring AtGRXcp::GUS had similar expression patterns (data not shown). AtGRXcp::GUS was highly expressed in the young cotyledons at 3 days after germination (Fig. 6A). In addition, AtGRXcp::GUS was detected in the green tissues (leaves), vascular bundles, roots, stems, and flowers (Fig. 6, B-E).
AtGRXcp Is Critical for Early Seedling GrowthTo gain insight into the biological function of AtGRXcp in planta, we analyzed two Salk T-DNA insertional lines (Salk_125903 and Salk_056587) (32) carrying a T-DNA in AtGRXcp (Fig. 7A). To confirm the presence of the T-DNA in atgrxcp plants, we used AtGRXcp-specific and T-DNA-specific primers to PCR-screen T3 progeny and obtained two homozygous lines, termed atgrxcp1 and atgrxcp2 (Fig. 7A). Sequence analysis of the T-DNA flanking regions revealed that in atgrxcp1 the T-DNA is located in the middle of AtGRXcp at position Met71, and in atgrxcp 2 the T-DNA is inserted at position Ile12 (Fig. 7A). Both T-DNA insertions disrupted AtGRXcp expression as determined by RNA gel blot analysis (Fig. 7B).
Both atgrxcp1 and atgrxcp2 failed to display visible defects in seed germination and early growth in normal nutrient medium in comparison with wild type (Fig. 8). Given that AtGRXcp is able to restore yeast Grx5 function (Figs. 3, 4, 5), whose mutation causes iron accumulation (21), we were interested in determining if an AtGRXcp deletion would affect the iron sensitivity of seedlings. Both wild type and atgrxcp seeds were germinated and tested on both iron-sufficient and iron-deficient media. No significant difference was seen between wild type and atgrxcp mutants. However, when wild type and atgrxcp seeds were germinated and grown on one-half strength Murashige and Skoog medium containing 3 mM H2O2, young seedlings of atgrxcp grew at a reduced rate as measured by alterations in primary root growth (Fig. 8).
|
|
|
| DISCUSSION |
|---|
|
|
|---|
A large number of Grxs have been identified in various species based solely on genome analysis (Fig. 1). In plants, several members of this protein family are predicted to target to and function in plastids/chloroplasts (30); however, little is known about their physiological roles. Both plant AtGRXcp and zebrafish Grx5 are able to suppress yeast grx5 mutant phenotypes (Figs. 3, 4, 5) (27), suggesting that the biological function of this group of monothiol glutaredoxins is evolutionarily conserved. Interestingly, AtGRXcp-GFP, like Grx5, localized to the mitochondria when expressed in yeast cells (Fig. 3C); however, AtGRXcp localized to chloroplasts in plant cells (Fig. 2). The dual targets of AtGRXcp have been observed for other plant and yeast proteins (45, 48). For example, the plant phosphate transporter, Pht2,1, localizes to the chloroplast envelope in plants but to mitochondria when expressed in yeast (48). The nuclear localized yeast Grx3 has been shown to be able to compensate for Grx5 when localized to the mitochondria (45). Mitochondrial localization of AtGRXcp in yeast is necessary for its function, as evidenced by the fact that when the 63-amino acid signal peptide of AtGRXcp was removed, the truncated form of AtGRXcp was unable to target mitochondria and suppress the sensitivity of grx5 cells to H2O2 (Fig. 3).
The ability of AtGRXcp to partially restore the Fe-S enzyme activities and suppress iron accumulation in yeast grx5 cells suggests that AtGRXcp may be required for biogenesis of iron-sulfur (Fe-S) clusters and/or involved in the regulation of iron homeostasis in chloroplasts. In plants, Fe-S clusters have an important role in the light-harvesting photosystem I and the cytochrome b6/f complex for electron transport (49). Chloroplasts/plastids also contain many Fe-S proteins, such as a 2Fe-2S ferredoxin and 4Fe-4S ferredoxin-thioredoxin reductase (50). Recent studies indicate that Arabidopsis proteins directly and indirectly linked to chloroplast activities exhibit more sensitivity to oxidative damage (9). There is a possibility that AtGRXcp protects protein oxidative damage via modulating iron homeostasis to control iron-generated oxygen radicals within chloroplasts. Support for this model was observed in AtGRXcp-expressing yeast cells. Apparently, AtGRXcp attenuates protein oxidation by reducing intracellular iron levels (Fig. 4, C and D). Furthermore, in plants, proteins from chloroplasts of atgrxcp plants were subjected to increased carbonylation, an indicator of increased oxidative damage to proteins (Fig. 7C). It is also possible that AtGRXcp suppression of yeast grx5 mutant phenotypes was due to activation of the ROS scavenging system (Figs. 3 and 4). In support of this hypothesis, recent studies indicate that monothiol Grxs can modulate cellular signaling events through protein-protein interactions mediated by PICOT-HD (23, 26, 51). Future work will be necessary to determine the target(s) of AtGRXcp and understand the complexity of ROS regulation in planta.
AtGRXcp contains the putative active motif, CGFS (Fig. 1), and substitution of the conserved cysteine residue (Cys97) with alanine caused a deceased expression and shorter half-life of AtGRXcp in both yeast and plant cells (Fig. 5, B-D). In contrast, mutation of both the conserved Cys60 and the nonconserved Cys117 in yeast Grx5 and the conserved Cys30 in bacterial Grx4 did not significantly affect protein stability (20, 47, 52). These findings suggest that this particular cysteine residue (Cys97) may play a unique role in the plant AtGRXcp. In addition, our results suggest that the conserved phenylalanine residue (Phe99) is not essential for the function of AtGRXcp (Fig. 6). In this case, our findings are similar to that observed with the Phe62 variants in yeast Grx5 (47). Recently, the resolved three-dimensional solution structure of the bacterial Grx4 reveals that this monothiol Grx has a unique structure, which is significantly different from the dithiol Grxs (53). AtGRXcp consists of four cysteine residues: the conserved Cys97 and the nonconserved Cys151, which are located in the conserved GRX region (Fig. 1), and two additional cysteine residues (Cys62 and Cys172) that are not conserved in the monothiol Grxs (Fig. 1) (26). Future work will need to be directed at the various roles these cysteine residues may have in disulfide formation, electron donation, and redox-catalysis.
We have already pursued several avenues to further discern the biochemical properties of AtGRXcp. Initial analysis indicated that AtGRXcp, like yeast Grx5 and bacterial Grx4, was not active in either the insulin or the
-hydroxyethyl disulfide assays (data not shown) (20, 52). We certainly envision AtGRXcp playing a myriad of roles in planta; however, studies directed at clarifying other AtGRXcp functions require additional inquiry.
Recent reports have shown that Arabidopsis has high steady-state levels of protein carbonylation, particularly chloroplastic proteins, during vegetative growth (9, 54). AtGRXcp is highly expressed in young cotyledons, green leaves, and the vasculature of roots (Fig. 6, A-C), implicating that AtGRXcp may be critical for protecting chloroplasts/plastids against oxidative damage during early growth and development. In agreement with this, atgrxcp mutant roots are more sensitive to external oxidants (Fig. 8). It is possible that plastidic AtGRXcp could be involved in the glutathiolation/deglutathiolation of proteins through coupling with GSH and GSH reductase in chloroplasts/plastids, which is required for early seedling growth (55). This hypothesis is reinforced by results suggesting that changes in glutathione redox state controlled via GSH reductase activities have an important role in cell differentiation, root growth, and plant development (56, 57). In addition, a CC type Grx, ROXY1, appears to play a role in post-translational regulation of floral identity gene products that are required for floral petal development (31).
Although atgrxcp plants did not display any altered sensitivity to iron imbalance at the whole plant level (data not shown), this may be due to tight regulation of iron uptake (34) or iron-related phenotypes being masked by functional redundancy among Grxs (12, 30). The interplay among Grxs and various antioxidant systems will also be addressed in future studies.
The characterization of AtGRXcp reported here is particularly noteworthy in that Grxs have not been previously functionally characterized. AtGRXcp appears to be evolutionarily conserved across taxa, and the capability of AtGRXcp to rescue yeast Grx5 deficiency phenotypes suggests a conserved biochemical mechanism among monothiol Grxs. The AtGRXcp-deficient plant lines demonstrate that this protein plays a critical role in protecting protein oxidation during stress conditions. Given that there are at least 31 genes in the Arabidopsis genome coding glutaredoxins, the characterization of AtGRXcp is an important first step toward understanding how this large gene family functions in adapting plants to external stresses.
| FOOTNOTES |
|---|
1 To whom correspondence should be addressed: Plant Physiology Group, USDA/ARS Children's Nutrition Research Center, Dept. of Pediatrics, Baylor College of Medicine, 1100 Bates St., Houston, TX 77030. Tel.: 713-798-7012; Fax: 713-798-7078; E-mail: ncheng{at}bcm.tmc.edu.
2 The abbreviations used are: ROS, reactive oxygen species; PICOT-HD, protein kinase C-interacting cousin of thioredoxin homology domain; Grx, glutaredoxin; Trx, thioredoxin; GFP, green fluorescent protein; GUS,
-glucuronidase gene; CaMV, cauliflower mosaic virus promoter. ![]()
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Zaffagnini, L. Michelet, V. Massot, P. Trost, and S. D. Lemaire Biochemical Characterization of Glutaredoxins from Chlamydomonas reinhardtii Reveals the Unique Properties of a Chloroplastic CGFS-type Glutaredoxin J. Biol. Chem., April 4, 2008; 283(14): 8868 - 8876. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Sundaram, B. Rathinasabapathi, L. Q. Ma, and B. P. Rosen An Arsenate-activated Glutaredoxin from the Arsenic Hyperaccumulator Fern Pteris vittata L. Regulates Intracellular Arsenite J. Biol. Chem., March 7, 2008; 283(10): 6095 - 6101. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||