Chromium-induced physiological and proteomic alterations in roots of Miscanthus sinensis

Plant Science 187 (2012) 113–126 Contents lists available at SciVerse ScienceDirect Plant Science journal homepage: www.elsevier.com/locate/plantsci Chromium-induced physiological and proteomic alterations in roots of Miscanthus sinensis Shamima Akhtar Sharmin 1 , Iftekhar Alam 1 , Kyung-Hee Kim, Yong-Goo Kim, Pil Joo Kim, Jeong Dong Bahk, Byung-Hyun Lee ∗ Division of Applied Life Science (BK21 program), IALS, PMBBRC, Gyeongsang National University, Jinju 660-701, Republic of Korea a r t i c l e i n f o Article history: Received 13 December 2011 Received in revised form 31 January 2012 Accepted 2 February 2012 Available online 9 February 2012 Keywords: Abiotic stress Chromium Heavy metal Proteome Miscanthus sinensis a b s t r a c t Despite the widespread occurrence of chromium toxicity, its molecular mechanism is poorly documented in plants compared to other heavy metals. To investigate the molecular mechanisms that regulate the response of Miscanthus sinensis roots to elevated level of chromium, seedlings were grown for 4 weeks and exposed to potassium dichromate for 3 days. Physiological, biochemical and proteomic changes in roots were investigated. Lipid peroxidation and H2 O2 content in roots were significantly increased. Protein profiles analyzed by two-dimensional gel electrophoresis revealed that 36 protein spots were differentially expressed in chromium-treated root samples. Of these, 13 protein spots were up-regulated, 21 protein spots were down-regulated and 2 spots were newly induced. These differentially displayed proteins were identified by MALDI-TOF and MALDI-TOF/TOF mass spectrometry. The identified proteins included known heavy metal-inducible proteins such as carbohydrate and nitrogen metabolism, molecular chaperone proteins and novel proteins such as inositol monophosphatase, nitrate reductase, adenine phosphoribosyl transferase, formate dehydrogenase and a putative dihydrolipoamide dehydrogenase that were not known previously as chromium-responsive. Taken together, these results suggest that Cr toxicity is linked to heavy metal tolerance and senescence pathways, and associated with altered vacuole sequestration, nitrogen metabolism and lipid peroxidation in Miscanthus roots. © 2012 Elsevier Ireland Ltd. All rights reserved. 1. Introduction Heavy metal contamination is a cause of major environmental hazards worldwide, leading to losses in agricultural yields and harmfully affecting human health when contaminants enter the food chain. Chromium (Cr) is the seventh most abundant element on earth. It exists in nature in both trivalent (Cr III) and hexavalent (Cr VI) forms, of which the latter is more toxic [1]. Cr compounds cause environmental pollution as a result of a large number of industrial operations, including mining, pigment manufacturing, petroleum refining, leather tanning, wood preserving, textile manufacturing, pulp processing and fungicide development [2]. In India, about 2000–3200 tones of elemental Cr leak to the environment annually with a Cr concentration ranging between Abbreviations: 2-DE, two-dimensional gel electrophoresis; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; PMF, peptide mass fingerprinting; ROS, reactive oxygen species; SDS-PAGE, sodium dodecylsulfate polyacrylamide gel electrophoresis; TBARS, thiobarbituric acid reactive substance; V-ATPase, vacuolar-type H+ -ATPase; UDP-GlcDH, UDP-glucose dehydrogenase. ∗ Corresponding author. Tel.: +82 55 772 1882; fax: +82 55 772 1889. E-mail address: [email protected] (B.-H. Lee). 1 S.A. Sharmin and I. Alam contributed equally to this work. 0168-9452/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.plantsci.2012.02.002 2000 and 5000 mg L−1 [3]. Very high level of Cr(VI) contamination (14,600 mg kg−1 in ground water and 25,900 mg kg−1 in soil) has been reported in some sites of Oregon state, USA [4]. Generally, most Cr (VI) added to soil is promptly reduced to the inert form Cr (III) by several agents. However, re-oxidation of Cr (III) to Cr (VI) occur by microorganisms and, therefore, both states should be regarded hazardous for the environment and for humans [5]. Both forms cause serious damage to plant tissues and organs at differing concentrations. Cr phytotoxicity can result in the inhibition of seed germination, pigment degradation, disturbances in the nutrient balance and the generation of reactive oxygen species (ROS), which induces oxidative stress and alterations in antioxidant enzyme activities [6]. In the cell, free system reactivity of Cr is generally considered by its interaction with glutathione (GSH), NADH and H2 O2 -generating hydroxyl radicals (OH− ) [7]. Both Cr III and VI react with cellular H2 O2 , generating highly reactive hydroxyl radicals. Industrial chromium wastes are generally treated with physicochemical processes before they are released into the environment. Following primary treatments, the methods of removal of residual chromium (polishing) are expensive and the efforts are often insufficient [8]. Consequently, residual Cr are released to the environment and accumulated in agricultural products through water, 114 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 air and polluted soils [6]. Soils from numerous sites in the USA are contaminated with Cr at levels ranging from 1 to 1500 mg Cr kg−1 [9]. Soil pollution generates extra costs for soil management and pollution control. The uses of plants for soil phytoremediation by means of degradation (phytodegradation), adsorption (rhizofiltration) and absorption (phytoextraction) are efficient, renewable, and natural processes that are leading competitors in the search for solutions to these contamination issues. Unfortunately, most known hyperaccumulator plants have very low biomass and/or slow growth rates, are difficult to cultivate on a commercial scale and have very few commercial uses. Therefore, attention has been focused on several biomass crops that have fast growth rates and high biomass and are able to accumulate moderate to large amounts of heavy metals without sacrificing biomass gain. Most studies involving Cr overaccumulation have focused on extreme examples, representing plants native to highly Cr-rich environments [10,11]. Very little attention has been paid to commercially important biomass-producing crops. Miscanthus sinensis, a perennial rhizomatous C4 grass, is a potentially efficient, sustainable carbon-neutral producer of lignocellulosic biomass, making it very suitable and promising for the production of biofuels and fiber [12]. Miscanthus sequesters higher amount of Cr to the aerial part at extremely toxic levels, whereas the overall ability of this species to remove Cr from the solution is higher at moderate toxicities [13]. These suggest that M. sinensis is a potential bioaccumulator of Cr and other heavy meals. Heavy metal-accumulating plants have expansive, advanced antioxidant defense systems and other important features that enable them to acquire tolerance [14,15]. Unfortunately, little is known about the molecular basis of excess heavy metal tolerance. Unlike other heavy metals, such as As, Cu, Pb and Cd, the partitioning of Cr by phytochelatin synthesis has not been observed; therefore, the detoxification mechanism for this metal is poorly understood [6]. Molecular events underlying Cr toxicity and the defense-related signal transduction process have been only partially elucidated. A number of genes potentially involved in Cr tolerance and accumulation were assessed by cDNA-AFLP and reported [16]. Recently, the combination of genome-wide transcriptome profiling and metabolome analysis has been reported in Cr-stressed rice plant [17]. Proteomics, the comprehensive and quantitative analysis of proteins that are expressed in a given organ, tissue or cell line, provides unique insights into biological systems that cannot be acquired from genomic or transcriptomic approaches. Proteomics has been used extensively to investigate the protein expression pattern under abiotic stresses. Expression pattern of maize proteins in response to high concentrations of Cr (340–1019 ␮M) have been described for the first time by [18]. However, no proteomic study has been carried on M. sinensis in response to Cr stress. Therefore, we carried out a proteomic analysis of M. sinensis roots subjected to Cr stress to identify proteins or primary targets, hoping to gain a more thorough understanding of the molecular basis of heavy metal tolerance in this species. 2. Materials and methods 2.1. Plant growth and treatments M. sinensis (cv. Kosung) seeds were planted on commercial potting mix in plastic trays and allowed to germinate in a growth chamber. Three weeks after germination, the seedlings were transferred to hydroponic cultures supplied with half strength Hoagland nutrient solution (H2395, Sigma, USA). pH of the medium was adjusted to 5.8. To ensure proper growth, the solutions were aerated with aquarium aerators. Following a 1-week hydroponic adaptation, the seedlings were subjected to treatments of 0, 50, 100, 200, 300, 500, 750 and 1000 ␮M potassium dichromate (K2 Cr2 O7 ). After a 3-day treatment, the roots were excised from untreated (control) and treated seedlings and used for proteomic and physiological analyses. The entire experiment was conducted under light conditions (500 ␮mol m−2 s−1 , 16/8 h light/dark period) at 25 ◦ C and 65% humidity. 2.2. Determination of Cr accumulation in roots After 3 days of treatment, root samples were washed five times with deionized water to remove surface Cr salts. The samples were dried in an incubator at 60 ◦ C for 72 h, weighed, and then ground to a fine powder. Approximately 1 g of fine powder from each treatment group was digested, using a ternary solution (HNO3 /H2 SO4 /HClO4 , 10:1:4 v/v), and the total Cr in the digestion solution was determined with a graphite furnace atomic absorption spectrophotometer (GFAAS) (PerkinElmer SIMAA 6000, Norwalk, CT, USA) [19]. Three different biological replicate root samples were used for the analysis. 2.3. Measurement of lipid peroxidation and hydrogen peroxide Lipid peroxidation was estimated by measuring the concentrations of 2-thiobarbituric acid-reactive substances (TBARS) as described previously [20]. Briefly, 300 mg of powdered tissue were homogenized in 20% trichloroacetic acid (TCA), containing 0.5% 2thiobarbituric acid, and heated at 95 ◦ C for 30 min [21]. The TBARS concentrations were measured as the malondialdehyde (MDA; ε = 155 mM−1 cm−1 ) concentrations, which were determined at A532 and corrected for nonspecific turbidity at A600 . The hydrogen peroxide (H2 O2 ) concentrations were measured spectrophotometrically as described by [22]. Briefly, H2 O2 was extracted by homogenizing 300 mg of tissue samples with 3 mL of phosphate buffer (50 mM, pH 6.8), containing the catalase inhibitor hydroxylamine (1 mM). The homogenate was centrifuged at 6000 × g for 25 min. A mixture comprised of 3 mL of extracted solution and 1 mL of 0.1% titanium sulfate in 20% (v/v) H2 SO4 was centrifuged at 6000 × g for 15 min. The intensity of the yellow color of the supernatant was measured at 410 nm. The H2 O2 level was calculated, using the extinction coefficient 0.28 ␮mol−1 cm−1 . 2.4. Protein extraction and 2-D electrophoresis Proteins were extracted from the root sample using a phenol extraction method according to our previous paper [23]. Briefly, 750 mg of tissue was homogenized with a Mg/NP-40 extraction buffer [0.5 M Tris–HCl, pH 8.3, 2% (v/v) NP-40, 20 mM MgCl2 , 1 mM phenyl methyl sulfonyl fluoride, 2% (v/v) ␤-mercaptoethanol and 1% (w/v) polyvinyl polypyrrolidone] and fractionated with water-saturated phenol, followed by centrifugation at 12,000 × g for 15 min. The proteins were recovered from the supernatant by precipitation with ammonium acetate in methanol. The protein samples were then quantified using the Lowry method [24] and subjected to two-dimensional gel electrophoresis (2-DE) using a standard procedure. The protein samples were dissolved in a reswelling buffer [8 M urea, 1% CHAPS, 0.5% (v/v) IPG buffer pH 4–7, 20 mM dithiothreitol (DTT), and a trace of bromophenol blue]. A total of 500 ␮g of dissolved protein sample was applied to the immobilized pH gradient (IPG) dry strip (pH 4–7, 18 cm) for 13–14 h, followed by focusing for 47,500 V-h using an IPGphor (Amersham Bioscience, Uppsala, Sweden). After isoelectric focusing (IEF), the IPG strips were equilibrated for 15 min in an equilibration buffer [50 mM Tris–HCl, pH 8.8, 6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS, and a trace of bromophenol blue] containing 10 mg/mL DTT, followed by 15 min in an equilibration buffer S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 containing 25 mg/mL iodoacetamide. Second dimension SDS-PAGE was carried out using a 12% polyacrylamide gel, and the gels were stained with colloidal Coomassie brilliant blue (CBB). 2.5. Gel documentation and analysis Images of CBB-stained gels, which were acquired using a high-resolution scanner (GS-800 Calibrated Imaging Densitometer; Bio-Rad, Hercules, CA, USA), were used for analysis. Spots were detected, quantified and then matched using the Bio-Rad PDQuest software (Version 7.2; Bio-Rad). To compensate for the variability in gel staining, the volume of each spot (spot abundance) was normalized as a relative volume. After automated detection and matching, manual editing was performed. A minimum of three gels were generated for each sample. Only spots that showed significant and reproducible changes of at least 1.5-fold were considered to be differentially expressed proteins. The standard error (SE) was calculated from three spots in replicated gels. 2.6. In-gel digestion, MALDI-TOF MS and database search Selected protein spots were excised manually from the CBBstained gels, washed with 50% (v/v) acetonitrile (ACN) in a 0.1 M NH4 HCO3 solution and then vacuum-dried. The gel fragments were reduced for 45 min at 55 ◦ C in a solution of 10 mM DTT in 0.1 M NH4 HCO3 . After cooling, the DTT solution was immediately replaced with 55 mM of iodoacetamide in 0.1 M NH4 HCO3 . After washing with 50% ACN in 0.1 M NH4 HCO3 , the dried gel pieces were left to swell in a minimum volume of 10 ␮L of digestion buffer (25 mM NH4 HCO3 and 12.5 ng/␮L trypsin, Promega, WI, USA). Following overnight digestion at 37 ◦ C, the peptides were dried. The samples were analyzed using a Voyager-DE STR MALDITOF mass spectrometer (PerSeptive Biosystems, Framingham, MA, USA). Parent ion masses were measured in the reflectron/delayed extraction mode with an accelerating voltage of 20 kV, a grid voltage of 76.000%, a guide wire voltage of 0.01%, and a delay time of 150 ns. A two-point internal standard for calibration was used with des-Arg1-Bradykinin (m/z 904.4681) and neurotensin (m/z 1672.9175). The software Data Explorer® (PerSeptive Biosystems, Inc.,USA) was used to view and process data files. The peptide mass fingerprintings (PMFs) obtained from each digested protein were compared with PMFs in the non-redundant National Center for Biotechnology Information database (NCBInr, 2011/01/01) using the ProFound program (http://prowl.rockefeller.edu/prowlcgi/profound.exe). The search was performed within all green plants (Viridiplantae) using the following parameters: the maximum number of missed cleavages was set at one, the complete carbamidomethylation of cysteines and variable oxidation of methionines was assumed, monoisotopic masses were used and a mass tolerance of 100 ppm was allowed. Only significant hits, as defined by the ProFound ‘expectation value’ of <5e−2 (i.e. p < 0.05) were chosen. The estimated experimental Mr /pI was applied to increase the confidence of identification (Table 1). 2.7. MS/MS analysis MS and MS/MS analyses were performed as described earlier [25]. Mass spectra were acquired with an ABI 4800 Plus TOF–TOF Mass Spectrometer (Applied Biosystems, Framingham, MA, USA), which uses a 200 Hz ND:YAG laser operating at 355 nm. The ten most and least intense ions per MALDI spot, with signal/noise ratios >25, were selected for subsequent MS/MS analysis in 1 kV mode and 800–1000 consecutive laser shots. During MS/MS analysis, air was used as the collision gas. Data were subjected to a Mass Standard Kit for the 4700 Proteomics Analyzer (Calibration mixture 1). MS/MS spectra were searched against the NCBInr database by ProteinPilot 115 v.3.0 (with MASCOT as the database search engine) with peptide and fragment ion mass tolerance of 50 ppm. Carbamidomethylation of cysteines and oxidation of methionines were allowed during the search of the peptides. One missing trypsin cleavage was allowed. Peptide mass tolerance and fragment mass tolerance of the selected 95 proteins were set to 50 ppm. High confidence identifications had statistically significant search scores (greater than 95% confidence, equivalent to MASCOT expect value p < 0.05), were consistent with the protein’s experimental pI and Mr , and accounted for the majority of ions present in the mass spectra. 2.8. Statistical analysis Results of the physiological parameters and spot intensity were statistically analyzed by using analysis of variance (ANOVA) and Duncan’s multiple range test (DMRT) to determine significant differences among group means. Significant differences from control values were determined at p < 0.05 levels. All the results are represented as means ± SE of at least three independent replicates. The statistical program SAS, version 9.1 (SAS Institute, Cary, NC, USA) was used for the statistical analyses. 2.9. Multivariate analysis To visualize patterns in abundance, multivariate analysis was performed to the data from the differentially expressed proteins. The 36 protein spots that had a 1.5-fold or greater variation (p ≤ 0.05) in at least one point were used for principal component analysis (PCA) and cluster analysis. The datasets from the three replicates were grouped with components showing differences between the individual treatments and the differentially expressed spots. The PCA analysis was performed using XLSTAT software (Addinsoft SARL). Clustering of the dataset was performed using MeV software. For the calculation settings, six clusters were defined and Euclidean distance was selected as the similarity measure. 3. Results 3.1. Cr-induced morphological changes Our objective was to understand the possible mechanisms that might be activated by the M. sinensis cell in detoxification during Cr stress using proteomic approach. M. sinensis is a heavy metal tolerant species [26,27]. However, a preliminary study was necessary to define the concentrations of Cr that induce cellular response without leading to immediate cell death. A short exposure to low concentration of Cr (50–300 ␮M) did not exhibit any sign of growth reduction (Supplementary Fig. 1A). In low level of Cr, Miscanthus plants continued to grow for several weeks. MDA content in roots were almost similar to non-treated plants (Supplementary Fig. 1B), suggesting a short exposure to low level Cr cause negligible oxidative damage. However, growth suppression was observed starting from 500 ␮M of Cr with the highest inhibition occurring at 1000 ␮M. Under control condition, new roots were being developed, while their formation was suppressed over 500 ␮M. At 1000 ␮M, new root formation was severely affected. These observations suggest that M. sinensis is relatively tolerant to Cr like maize. [28] reported that low concentration of Cr have positive effect on root growth of Miscanthus. Thus, based on the growth pattern and earlier reports, 500–750 ␮M could be considered moderate to toxic, while 1000 ␮M or higher concentrations are acutely toxic. Labra et al. [18] also carried out a proteomic analysis of maize seedling subjected to 340 and 1019 ␮M Cr based on growth suppression. A short exposure to moderate to acute toxic Cr may reveal the proteins involved in altered metabolic homeostasis. 116 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 Table 1 Chromium stress-responsive differentially expressed proteins in Miscanthus sinensis roots identified by MALDI-TOF MS. Spot no. Protein Organism Accession no.a Theoretical Observed 1 2 5 7 9 10 11 12 13 Predicted protein Vacuolar H+-ATPase subunit B Os06g0136600 Hypothetical protein SORBIDRAFT Vacuolar ATP synthase catalytic subunit A Unknown Inositol monophosphatase Vacuolar ATP synthase catalytic subunit A SAM-2(S-adenosylmethionine synthetase 2); copper ion binding S-adenosyl methionine synthetase Cell division control protein 2 homolog C Tetratricopeptide-like helical PsHSP71.2 Hypothetical protein OsJ 02626 CPK31; ATP binding/calcium ion binding/calmodulin-dependent protein kinase Predicted protein Glutamine synthetase Os11g0229200 Hypothetical protein Unknown Hypothetical protein Hypothetical protein SORBIDRAFT 02g044060 Hypothetical protein SORBIDRAFT 10g016920 Hypothetical protein Glyceraldehyde-3-phosphate dehydrogenase C subunit (GapC) Hypothetical protein Unknown protein Hypothetical protein SORBIDRAFT 01g043060 Hypothetical protein ATP synthase F0 subunit 1 Unknown protein Hypothetical protein UDP-glucose 6-dehydrogenase Hypothetical Protein SORBIDRAFT 03g013290 Micromonas pusilla Zostera marina Oryza sativa Sorghum bicolor Zea mays Zea mays Ostreococcus tauri Zea mays Arabidopsis thaliana 226456463 118721470 115466256 242054033 195658441 223973319 116055491 195658441 15234354 90.50/4.9 54.47/5.2 48.15/5.4 58.42/5.9 68.69/5.3 72.82/5.5 77.03/5.6 68.69/5.3 43.64/5.7 Oryza rufipogon Antirrhinum majus Medicago truncatula Pisum sativum Oryza sativa Arabidopsis thaliana 100801628 5921446 92870988 562006 125571194 42570056 Populus trichocarpa Saccharum officinarum Oryza sativa Arabidopsis thaliana Zea mays Zea mays Sorghum bicolor Sorghum bicolor Oryza sativa Arabidopsis thaliana Vitis vinifera Arabidopsis thaliana Sorghum bicolor Oryza sativa Oryza sativa Arabidopsis thaliana Oryza sativa Zea mays Sorghum bicolor 14 18 21 22 26 27 30 32 36 37 39 48 50 55 56 57 62 67 68 69 70 72 73 75 78 a b c d SC (%)b PMc Expectd 54/5.0 56/5.2 54/5.1 55/5.7 70/5.6 70/5.6 70/5.7 70/5.7 45/6.1 14 24 21 35 25 22 21 21 23 8 9 7 13 13 12 10 9 7 2.0e−2 2.8e−5 7.2e−3 6.5e−5 8.9e−4 2.1e−3 1.5e−2 1.0e−3 4.2e−3 42.99/5.7 34.51/6.8 87.67/6.9 71.55/5.2 27.50/5.5 55.08/6.0 45/5.9 47/6.1 80/6.8 70/5.2 35/5.7 34/5.7 25 28 13 28 17 25 8 6 5 11 4 7 2.0e−3 2.0e−2 2.2e−2 9.6e−3 8.7e−3 4.9 e−2 222834292 56681315 115484821 7268210 194701624 226507242 242051414 242095836 47900451 21593240 58.00/5.8 39.57/5.5 38.89/5.9 23.86/5.5 19.50/5.1 38.78/6.3 27.26/5.2 41.64/6.2 40.92/6.5 37.09/6.6 40/6.2 42/5.7 30/5.2 30/5.4 25/5.1 39/6.8 27/5.7 43/6.5 43/6.6 40/6.8 27 18 22 26 44 22 49 21 24 35 8 8 7 5 7 5 8 5 7 7 2.0e−2 1.6e−2 3.7e−2 2.6 e−2 1.1 e−2 1.9 e−2 4.8 e−4 2.7 e−3 1.2 e−2 9.9 e−3 225448323 15222614 242041787 222631421 194033257 79474381 35215055 195623986 242057247 41.80/5.3 44.84/5.4 58.41/6.1 55.45/7.3 55.64/5.8 58.34/8.4 34.34/5.5 53.51/5.7 53.39/6.5 44/5.5 55/5.5 56/6.1 55/6.2 55/6.2 35/6.3 36/5.8 55/6.4 58/6.7 44 19 35 15 36 23 33 22 25 12 6 13 6 18 6 6 7 7 4.4 e−4 1.1e−2 1.9 e−4 9.9 e−3 8.6 e−7 2.6 e−4 3.7 e−2 2.7 e−2 5.3 e−3 Mr /pI Accession number in NCBI database. SC, sequence coverage by PMF using MALDI-TOF MS. PM, number of peptides matched. ProFound expectation value, a value of <5e−2 indicates p < 0.05. Fig. 1. Effect of Cr treatment on morphology (A), root growth (B) and Cr content (C) in M. sinensis roots after treated with indicated concentrations of K2 Cr2 O7 for 3 days in hydroponics in a Hoagland medium. Dry weights were calculated from pooled root samples of 12 plants. The root length and Cr amounts represent the mean values and SE. Different letters above the bars indicate statistically significant differences (p < 0.05). S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 When M. sinensis seedlings were subjected to 500–1000 ␮M of Cr for 3 days, a considerable reduction in root growth was observed in parallel with the doses of Cr in the medium. The decreases in root growth following Cr treatment were characterized by reductions in root lengths and dry weight with increasing Cr concentrations (Fig. 1A and B). Chromium accumulation in M. sinensis roots increased with increasing concentrations of potassium dichromate added to the solution (Fig. 1C). After 3 days of exposure to 1000 ␮M, the root tissue accumulated 1308 mg kg−1 on a dry weight basis. No Cr accumulation could be detected in the control plants. 3.2. Accumulation of H2 O2 and lipid peroxidation To investigate whether growth inhibition was associated with oxidative stress, the amount of H2 O2 and malondialdehyde (MDA) were examined. As shown in Fig. 2, H2 O2 accumulation was much higher in Cr-treated root samples compared to the control. Although physiological concentrations of ROS have important functions in stress signaling, excess amount can cause oxidative stress, leading to cell death, if they are not detoxified. Membrane lipids are the main cellular targets that are susceptible to damage, and lipid peroxidation is believed to be a free radical-mediated process [29]. Thus, we estimated lipid peroxidation in roots by the thiobarbituric acid (TBA) method, in which the quantified TBA-reactive substance was malondialdehyde, an end product of lipid peroxidation. MDA concentrations were increased markedly in 750 and 1000 ␮M of Cr, indicating increased lipid peroxidation. The occurrence of lipid peroxidation induced by Cr was further validated by a histochemical assay, using Schiff’s reagent. As shown in Fig. 3A, an intense coloration was detected by Schiff’s stain with increasing concentrations of Cr. By contrast, control roots had very small stain. Together with the quantitative estimation, the histochemical detection provides additional advantages for localizing TBA-reactive products in situ in roots with high sensitivity [30]. These results suggest that like other heavy metals, Cr toxicity also generated ROS, which resulted in oxidative stress in the Miscanthus roots. Increased lipid peroxidation induced by heavy metals such as aluminum [31], lead [32] and arsenic toxicity [23] have been reported in various plants. Evans blue staining indicated that cell death occurred earlier and more robustly with increasing Cr concentrations (Fig. 3B). These results are consistent with those from the lipid peroxidation assay, indicating that Miscanthus suffers from Cr-induced oxidative stress at high concentrations. Cr (VI)-mediated • OH radical generation in cells has been reported [33]. Taken together, these results indicated that plants exposed to Cr treatment generate ROS, which resulted oxidative stress and cell death in roots. 3.3. Proteomic alteration of Miscanthus roots under Cr stress To investigate differentially expressed proteins from the Miscanthus root in response to excess levels of Cr, proteins were extracted from control and Cr-treated roots and separated by 2DE. A high resolution of 2-DE gel pattern with a pI range of 4–7 was detected by CBB staining (Fig. 4). More than 1150 protein spots were reproducibly detected in each CBB-stained gel by 2-DE analysis. Among the well-resolved spots, a densitometric analysis of the replicated gels revealed 36 proteins showed at least 1.5-fold increase or decrease in expression in at least one treatment (Figs. 4 and 5 ). Several regions of the gels are enlarged in Supplementary Fig. 2. Average spot volumes were compared for the individual spots across the three treatments. The relative abundance of protein spots on the gel is shown in Fig. 5. Two spots (spots 11 and 12) were hardly detectable in the control sample and 117 were induced after treatment, while three spots nearly disappeared (spots 21, 67 and 72) due to Cr treatments. 3.4. Identification of M. sinensis root proteins induced by Cr stress To identify differentially expressed proteins, spots were excised from the preparative gels, in-gel digested by trypsin and analyzed using MALDI-TOF or MALDI-TOF/TOF MS. The identity of 34 differentially expressed protein spots was obtained by PMF of MALDI-TOF MS (Table 1). Two additional proteins not recognized by MALDI-TOF MS were identified by MALDI-TOF/TOF MS and the sequences were determined (Table 2). Relatively small differences were observed between the theoretical and the predicted molecular masses. The molecular mass is robust toward amino acid changes. The pI values, however, vary more substantially, probably due to occurrence of isoforms and amino acid changes between species. Some of the identified proteins were annotated either as unknown and hypothetical proteins or as proteins without a specific function in the database. To gain functional information about these proteins, we searched them against their known homologs with BLASTP algorithm (www.ncbi.nlm.nih.gov/BLAST/) using their amino acid sequences as queries. Thirteen corresponding homologues with the highest homology are shown in Table 3. Most spots except spot 26, 30 and 48 shared more than 95% positives with homologues at the amino acid level, indicating that they might have similar function. Differential expression levels of the protein spots revealed that 13 proteins were up-regulated, 21 were downregulated and two were newly induced (Fig. 5, Supplementary Fig 2). Despite the progress being made in plant proteomics, the power of proteomics in non-model species has not been assessed thoroughly. As few nucleotide sequences are available from Miscanthus, cross-species protein identification is used. In the present experiment, more than half of the proteins were matched with rice, maize and sorghum sequences. Compared to nucleotides, proteins are generally better-conserved, making the identification of nonmodel gene products quite efficient when they are compared to well-known orthologous proteins. Therefore, cross-species identification is the only option for studying gene expression when analyzing poorly characterized genomes. Our results indicate that proteomic techniques can be successfully applied to plant systems that are not well-represented in nucleic acid and protein databases. Among the identified proteins, two enzymes involved in glycolysis pathway, enolase (spot 5) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, spot 57), were significantly downregulated, except for GAPDH, which was up-regulated under 1000 ␮M Cr. This could be due to post-translational modification. On the other hand, proteins involved in mitochondrial respiration such as ATP synthase (spot 70) and dihydrolipoamide dehydrogenase (spot 78) were increased. A putative mitochondrial processing peptidase (MPP; spot 68) is up-regulated; the protein plays an essential role in mitochondrial protein import. Novel accumulation of inositol monophosphatase (IMPase; spot 11) was observed in the treated plants. Several spots representing components of vacuolar transporters (spots 2, 9 and 12) were up-regulated by 2- to 25fold. Among the differentially accumulated nitrogen metabolism related proteins, both glutamine synthetase (GS; spot 32) and nitrate reductase (spot 30) were down-regulated. Defense-related proteins, such as chitinase (spots 34 and 35) and the NB-LRR protein (spot 36) were highly increased. Adenine phosphoribosyltransferase (APRT; spot 39), formate dehydrogenase (spot 55) and two spots representing S-adenosyl-l-methionine synthetase (SAMS; spots 13 and 14) were down-regulated; these are involved in biosynthesis of mugineic acid (MA), an iron chelating components exclusively in grasses. Multiple spots could be isoforms or same proteins with different post-translational modifications. The cell wall polysaccharide biosynthesis-related enzyme UDP-glucose 118 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 Fig. 2. Physiological responses of M. sinensis roots subjected to treatment. H2 O2 (A) and MDA (B) concentration in control and Cr-treated roots. The data represent the mean values and SE of three independent experiments. Different letters above the bars indicate statistically significant differences (p < 0.05). Fig. 3. Histochemical localization of lipid peroxidation and loss of plasma membrane integrity. (A) Differentially stained M. sinensis roots by Schiff’s reagent under different concentrations of Cr. A more intense pink color indicates more TBA-reactive products. (B) Loss of plasma membrane integrity detected by Evan’s blue staining. Higher concentration of Cr accumulates more frequent and intense pigmentation as a result of greater damage compared to control. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) Table 2 Chromium stress-responsive differentially expressed proteins in Miscanthus sinensis roots identified by MS/MS analysis. Spot no. Protein (organism) Accession no.a SC (%)b Scorec Peptide hitd Sequence identified 34 Chitinase II (Hordeum vulgare subsp. vulgare) 9501334 25 210 3 R.ELAAFFGQTSHETTGGTR.G R.GAADQFQWGYCFK.E K.ATSPPYYGR.G 35 Chitinase II (Hordeum vulgare subsp. vulgare) 563487 25 118 3 R.ELAAFFGQTSHETTGGTR.G R.GAADQFQWGYCFK.E K.ATSPPYYGR.G a b c d NCBI accession number. SC, sequence coverage. Score is the protein score based on combined MS and MS/MS spectra. Peptide hit is the unique number of MS/MS spectra matched to the trypsin peptide. S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 119 Fig. 4. A 2-DE analysis of M. sinensis root proteins under 500 (B), 750 (C) and 1000 ␮M (D) of Cr compared to control (A). The arrows indicate differentially expressed proteins in response to the Cr stress. A total of 500 ␮g of protein was separated by 2-DE as described in Section 2 and visualized with colloidal CBB staining. dehydrogenase (UDP-GlcDH, spot 75) was up-regulated. In addition, several proteins were identified as unknown/hypothetical proteins. Overall, the proteins can be broadly classified into several groups according to their putative physiological functions: (1) energy- and metabolism-related proteins (2) vacuolar ATPases (3) defense related proteins such as heat shock proteins (HSPs) (4) nitrogen metabolism proteins, (5) cell division and (6) stress signaling proteins associated with metal detoxification. Ion transporters, HSPs and energy metabolism-related proteins were the largest functional categories, suggesting that energy metabolism pathways are disrupted, and ion transporters and HSPs may play important roles in protecting the cells from Table 3 The homologs of unknown proteins. BLASTP (http://www.ncbi.nlm.nih.gov/BLAST/) was used to search for homologs of the unknown proteins. Spot no. Accessiona Homolog protein Organism Accessionb Identities Positives Expect 5 7 10 26 30 36 39 48 50 55 62 67 68 69 72 78 115466256 242054033 223973319 125571194 222834292 115484821 194701624 226507242 242051414 242095836 225448323 15222614 242041787 222631421 79474381 242057247 Enolase1 Mitochondrial F1-ATPase beta subunit Heat shock 70 kDa protein Protein phosphatase 2C Assimilatory nitrate reductase (NADH) small subunit NBS-LRR type resistance protein - barley (fragment) Adenine phosphoribosyl transferase 1 Putative r40c1 protein Cytosolic Ascorbate Peroxidase Formate dehydrogenase 1 Actin Heat shock protein 70 Mitochondrial-processing peptidase beta subunit Putative cytochrome P450 Salt-inducible protein homolog Putative dihydrolipoamide dehydrogenase precursor Zea mays Oryza sativa Zea mays Oryza sativa Cupriavidus metallidurans Oryza sativa Zea mays Oryza sativa Zea mays Zea mays Persea americana Planctomycete str. 140 Zea mays Oryza sativa Arabidopsis thaliana Oryza sativa CAA3944 218147 226500540 4339763 94313742 62732752 226498252 AAN64997.1 195643366 195640660 281485191 2644998 195628546 47777427 2244775 13365781 96 93 98 79 88 100 99 89 98 98 99 100 97 95 97 93 99 96 99 91 90 100 99 94 99 99 99 100 99 95 97 97 0.0 0.0 0.0 3e−115 1e−154 0.0 4e−96 5e−173 7e−142 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a b The accession number of the unknown proteins in Table 1. The accession number of the homologs identified by BLAST. 120 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 30 Spot 1 a 14 a Spot 2 a 25 45 Spot 5 b b 10 c 20 c 5 0 500 45 750 a Spot 10 40 b 35 0 1000 b C 500 750 35 c 25 500 750 a 20 10 10 10 5 5 0 5 b 0 C 500 750 1000 90 Spot 18 a 80 C 500 750 b 0 1000 C 500 750 Spot 21 a 140 8 100 6 80 40 180 750 a 140 b b c 80 80 60 60 40 40 20 0 20 0 500 750 Spot 26 bc 10 0 c C 500 C 1000 120 250 a Spot 30 100 b b 500 750 b 40 150 b b 500 750 1000 0 0 500 750 1000 b 10 5 0 750 60 1000 Spot 56 a 50 b 40 500 750 a 100 150 60 100 40 b c c 0 10 0 0 0 500 500 750 1000 Spot 48 a d 0 1000 Spot 35 a a 0 250 500 a 200 750 1000 Spot 36 b b b 750 1000 0 0 500 120 750 1000 Spot 50 a 120 120 100 c 60 c 20 0 0 0 0 500 750 1000 500 1000 Spot 55 a 45 750 40 35 60 25 b b c b 20 b b 750 1000 15 10 5 0 500 750 1000 Spot 62 a a a b 40 40 0 50 30 b 80 b 0 80 20 Spot 57 a 140 c b 100 140 160 20 1000 0 b 10 750 c 20 50 20 c 40 10 60 20 b 30 c 20 80 500 C 40 20 100 30 0 0 30 b b 500 b 30 15 C b b 50 40 25 a 80 10 Spot 34 60 30 b 1000 80 Spot 39 50 20 750 200 70 a Spot 37 a 35 500 a 50 60 40 1000 Spot 27 100 50 20 750 90 150 b c 0 b 100 c 500 b 50 120 250 b 40 C Spot 32 80 b 40 10 300 a 200 60 0 1000 b 70 c 20 0 750 a C 1000 a 20 40 2 1000 60 4 b 750 Spot 14 120 100 30 20 500 160 100 C C 180 140 b 50 60 30 c 0 1000 a Spot 13 160 60 b 50 500 Spot 22 120 60 C 70 a 10 70 0 1000 160 12 30 20 120 15 15 10 1000 a Spot 12 a 25 15 b 50 10 C 30 a b b 5 35 20 20 b 40 a Spot 11 a 30 60 b c 5 0 1000 25 30 b 10 2 0 Spot 9 a 40 15 4 a 80 70 15 25 c 6 Spot 7 a 30 8 15 25 20 35 10 20 0 a 40 12 0 0 500 750 1000 750 90 1000 Spot 67 a 80 0 60 60 50 50 40 40 30 30 20 20 0 b b b 500 750 1000 500 Spot 68 a 80 70 0 0 90 70 10 0 500 b c d 10 0 0 500 750 1000 Fig. 5. The expression levels of the identified proteins compared to those of control. Bars indicate the expression level of control, 500, 750 and 1000 ␮M of Cr consecutively. Spot intensities were measured using a densitometer and compared to those of the control. The average values of the relative increase levels of three replicate samples are shown in the histograms. The data represent the mean values and SE of three independent experiments. Different letters above the bars indicate statistically significant differences (p < 0.05). S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 90 50 Spot 69 a 45 a b 40 70 35 60 30 50 160 a Spot 70 a 80 a a Spot 72 140 120 100 80 25 40 20 c b 60 30 15 10 5 0 121 20 40 10 20 0 0 0 500 40 750 1000 0 120 Spot 73 a a 35 500 a 100 750 1000 Spot 75 a a 0 b b b 500 750 1000 120 a 100 a Spot 78 a 30 80 25 20 15 10 60 60 40 40 20 20 b c 5 0 80 b b 0 500 750 1000 0 0 500 750 1000 0 0 500 750 1000 Fig. 5. (Continued). damage following Cr toxicity. The broad category view is shown in Fig. 6. 3.5. Multivariate analysis Principal component analysis was conducted to statistically classify protein spots with differential expression patterns and exhibit the difference in proteomes across the treatments. The twodimensional PCA plots show that samples are positioned differently (Fig. 7). The variation in expression pattern appears to be correlated to Cr concentrations. Energy metabolism and ion transporters are closely grouped. We also applied hierarchical clustering to the proteome dataset. Hierarchical clustering method uses pair wise average-linkage algorithm and constructs a dendrogram by which all expression patterns assemble in a single tree whose branch length reflects the degree of similarity (Fig. 8A). Pearson correlation coefficient was applied to define the similarity and the averagelinkage to assemble the items. K-mean cluster analysis was used to place the protein spots with differing abundance for Cr treatments into six clusters. K-means clustering was also applied to Miscellaneous and unknown Energy metabolism Ion transportation Stress signalling and metal detoxification Defense and detoxification Cell division Nitrogen metabolism Fig. 6. Pie chart illustrating the assignment of the identified proteins to functional categories. categorize the differentially expressed proteins and showed more clearly the abundance relationship with Cr treatments. Thirty-six Cr-responsive proteins were categorized in six expression groups (Fig. 8B). 4. Discussion Cr has a complex chemistry and hence the detailed mechanism of toxicity of Cr is yet to be clearly explained in higher plants. Cr (VI) is a strong oxidant with a high redox potential. Higher H2 O2 production and lipid peroxidation observed in the present study indicates that extensive oxidative damage could have occurred to the root cells under Cr stress. To investigate molecular mechanism behind the stress response, protein- and RNA-based measurements are complementary, because each technique focuses on a subset of genes/proteins. Each technique has its advantages and disadvantages. However, a 2-DE approach will result in a better characterization when a species is poorly represented in sequence databases [34]. We used multivariate approach to express of expression patterns of the protein spots. The PCA reduces the dimensionality of the multidimensional analysis to display the two principal components that distinguish between two largest sources of variation within the dataset. When the abundance patterns for the protein species of different Cr treatments were analyzed by PCA, the spots are positioned differently (Fig. 7). However, several proteins with similar functions tend to group together. Such application of PCA had been successfully used before by other authors [35]. The cluster analysis summarizes major protein expression patterns, possible resistance, adaptation and sustained tolerance, following exposure to Cr (Fig. 8A and B). For instance, cluster-2 proteins were upregulated following Cr treatment, and maintained their level at higher concentrations. These proteins are involved in vacuolar transportation, ATP production and protein stabilization during stress condition. Regardless of Cr concentrations, its increased level might indicates their primary role in Cr tolerance. On the other hand, some other proteins increased (cluster 4) or decreased (cluster 3) at moderate Cr concentration but maintained similar level at higher concentrations indicate their role in Cr adaptation. Our results suggest that the application of hierarchical and 122 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 Fig. 7. Representation of the samples by PCA. 2-D plot of main principal components (PC1 and PC2) of: (A) all spots and (B) differentially expressed spots. Each data point in PCA plots (B) describes the expression values for the subset of proteins whose ratios varied 1.5-fold or more. nonhierarchical clustering methods is useful in presenting proteomic data as shown by others also [36]. In the following sections, we discussed the possible role of the Cr-induced proteins involved in a wide range of plant processes. 4.1. Energy metabolism Our proteomic data showed that the two key enzymes of the glycolysis pathway were strongly affected following Cr treatment. Fig. 8. Clustering analysis of the differentially expressed proteins under Cr treatments. (A) Dendrogram of the spots clustering is showed in the left. Relative expression values of individual proteins displayed as heat map. All quantitative information is transmitted using a color scale in which the color ranges from green for the highest down-regulation (−1.5) to red for the highest up-regulation (1.5). Black boxes indicate no changes in expression pattern compared to control condition (0 ␮M Cr). Each row of colored boxes is representative of a single spot and each treatment is represented using a single column (indicated). (B) K-means clustering showing the expression patterns for individual protein spots in the six main Cr-responsive clusters. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 Enolase (spot 5) and GAPDH (spot 57) were significantly downregulated by Cr. Interestingly, GAPDH was up-regulated at 1000 ␮M Cr. Photosynthesis and respiration are negatively affected by Cr due to damage of the photosynthetic apparatus, inhibited redox reactions and oxidative stress damage [37–39]. Several glycolytic enzymes and heat-shock proteins are specifically prone to Crdependent oxidative damage in Cr-treated yeast cells [40]. Thus, along with the evidence of oxidative stress, it could be speculated that Cr treatment may inhibit carbon flux in glycolysis in M. sinensis plants. Consequently, a lower reducing power (NADH) and reductions in ATP, carbon skeletons and pyruvate could be expected. Growth inhibition may lead to an accumulation of carbohydrates or direct inclusion into the mitochondrial respiration process (simultaneously) to compensate for the higher energy demands. In addition, Cr-induced oxidative stress requires a high reducing power to cope with the stress. These could be provided by the up-regulation of mitochondrial respiration. The higher abundance of a mitochondrial ATP synthase (spot 70), which catalyzes the formation of ATP from ADP in the membranes of mitochondria, indicated that mitochondrial respiration may be increased under Cr stress. In addition, dihydrolipoamide dehydrogenase (E-3 component of pyruvate dehydrogenase; spot 78) is up-regulated, which is essential for ATP production. The up-regulation of these enzymes has been reported under arsenic [41] or aluminum stress [42], but not in Cr. Because of its prime role in energy transduction, increased abundance under stress presumably reflects altered patterns of carbon flux in response to reduced photosynthesis and increased need for energy. Our proteomic data are in agreement with metabolite analysis showing inhibition of glycolysis enzymes by heavy metal in poplar [43]. Less usage of photoassimilates due to growth suppression and a decreased breakdown of carbohydrates may trigger an increased accumulation of stored carbohydrates, as observed in Cr-treated bush bean [38] and Cd-treated poplar [44]. These molecules may play roles in osmotic adjustments or protection of cell constituents. Cr has been shown to induce osmotic stress in bush bean [45]. We identified inositol monophosphatase (IMPase; spot 11) that catalyzes de novo inositol synthesis from glucose-6-phosphate and is required for the breakdown of inositol trisphosphate. Free inositol can act as an osmolyte. Accumulation of free inositol was reported in Cd-treated poplar plants [44]. Thus, in our proteomics experiment, novel accumulation of an IMPase could further strengthen the previous hypothesis proposed from sugar analysis [38,44]. 4.2. Ion transporters for excess metal management A number of up-regulated spots were identified as components of vacuolar (spots 2, 9 and 12) and mitochondrial (spot 7) transporters. Vacuolar-type H+ -ATPase (V-ATPase) energizes plant endomembranes. The differential resistances of a single species, such as barley, to various heavy metals involve unique capabilities for vacuolar compartmentation [46]. Metal tolerances versus metal sensitivities of closely related species or genotypes also appear to depend on additional membranes, including the tonoplast, and transporters [47,48]. Thus, in absence of phytochelatin synthesis [6,49], a very important role could be reserved for the transmembrane transport of toxic Cr ions by V-ATPases. Therefore, it is reasonable to assume that V-ATPase is affected by Cr treatment. Little information is available on the influence of heavy metals on either the structure or the activity of V-ATPase. The antiporter activity depends on the presence of a proton gradient across the vacuolar membrane and thus, indirectly, on the V-ATPase. Metal–proton antiport activity has been reported for several metals, such as Cd, Zn, and Mn in oat roots [50,51] and Zn in Silene vulgaris [48]. By contrast, other authors could not detect such activity for Cd, either in oat or Silene [52]. Until now, Cr-specific transporters were not 123 known. Therefore, the large differential expression of several vacuolar and mitochondrial ATPases found in our proteomic study may provide important clues for further investigations. 4.3. Proteins associated with defense and detoxification mechanisms Biotic and abiotic stresses often share multiple nodes of the same response signaling pathways, and their outputs may have significant functional overlaps [53]. Cr treatment resulted in highly up-regulated chitinase (spots 34 and 35) and NB-LRR protein (spot 36). Chitinases are components of plant defense against pathogens and heavy metals as well at toxic level [54]. The chitinases activity was reported to increase in barley and rape by Cr(VI) but not by Cr(III) [55]. Increased chitinase activities in plants subjected to abiotic stresses may result from the induction of cross-tolerance via cross-talking signaling pathways. Chitinase profiling in pea, bean, soybean, barley and maize under As, Pb or Cd stress [56] suggests isoform/specific expression. Chitinase appears to counteract oxidative stress as shown in transgenic tobacco plants expressing a fungal chitinase [57]. The plants were not only more resistant to fungal infection but also to salt and metal ion stress. Although elucidating the relationships between chitinase isoforms and metal tolerances would be speculative at this moment, the study of metalspecific chitinases may provide very promising insight into the mechanisms of Cr detoxification processes. NB-LRR proteins are key molecules in signaling cascades that often culminate in the activation of programmed cell death (PCD) [58]. Here, large increases in one of these proteins (spot 36) may be associated with PCD signaling under toxic concentrations of Cr. We also identified three spots as heat shock proteins (HSP70 family; spots 10, 22 and 67) and another chaperone protein (spot 21). Members of the HSP70 family are up-regulated as a result of thermal and oxidative stress, including exposure to Cd, As and other heavy metals [59]. However, HSP has not been reported to be up-regulated following Cr exposure in the published proteomics reports. HSPs have a broad range of functions, including protein folding, assembly, translocation and degradation [59]. In Arabidopsis, Cd treatment induces the up-regulation of genes involved in protein folding [60], which demonstrates that Cd toxicity is in part due to the induction of protein denaturation, probably by oxidative modifications [61]. Thus, HSPs and chaperone like proteins were up-regulated probably to protect cells against damages induced by Cr. Cytochrome P450s are one of the bioindicator of altered cellular metabolism caused by environmental pollution. Cytochrome P450 (spot 69), was previously reported as responsive to Cd and other stresses [62] and are probably functioning in detoxification of cytotoxic product. In addition, one of the key ROS scavenging enzymes, ascorbate peroxidase (APX) is down-regulated (spot 50). Antioxidant enzymes are generally increased in activity during metal toxicity. Indeed, decreased APX activity has been observed at very high heavy metal toxicities [63]. 4.4. Nitrogen metabolism Glutamine synthetase (GS) is the key enzyme in ammonia assimilation and catalyzes the ATP-dependent condensation of ammonium ions with glutamate to produce glutamine. The GS is prone to degradation under oxidative stress conditions [64]. Oxidation of GS was reported in GS-enriched soybean root extract by metal-catalyzed oxidation system (producing hydroxyl radicals). The oxidized GS is inactive and more susceptible to degradation than the non-oxidized form [65]. Cr stress causes free radical generation via the Fenton reaction. Thus, the co-occurrence of increased oxidative stress and a down-regulated GS (spot 32) in our experiment is reasonable. Aside from transcriptional regulation, GS in 124 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 plants may also be regulated at the level of protein turnover. Decreased GS activity has been shown in leaves subjected to water stress and exposed to excess Cu [66]. A decline in GS activity may result, at least in part, in an accumulation of ammonium ions. Under physiological conditions, nitrate reductase [NAD(P)H:nitrate oxidoreductase EC 1.6.6.2] is reversibly converted into an inactive enzyme upon the addition of ammonia. In the presence of methionine sulfoximine, nitrate reductase is no longer inactivated by ammonia when GS activity is lost. The addition of ammonia to cell suspensions drastically prevents the reduction of nitrate to nitrite in the blue-green alga Agmenellum [67] and in the red alga Cyanidium [68]. Here, we identified a nitrate reductase protein (spot 30) that was down-regulated by Cr stress for the first time. Thus, a down-regulated nitrate reductase protein is consistent with previous physiological data. 4.5. Cell division related proteins Cr-induced disruption in microtubule organization and subsequent micronuclei formation has been reported [69]. However, proteins involved with these processes are not well described. Microtubules play key roles in both nuclear division and cytokinesis in plants and are one of the main subcellular targets of Cr (VI) toxicity. We identified increased levels of an actin protein (spot 62), which is major cytoskeletal components that play roles in gene transcription and signal transduction events in plants [70]. The functional complexity of actins makes them likely targets of oxidative stress. Al-induced damaging effect of F-actin in root stele cells and subsequent inhibition of root elongation has been shown in maize [71]. In addition, cell division control protein 2 homologue C (CDC2; spot 18), a protein kinase, is conserved throughout eukaryotes and acts as a key regulator of the cell cycle, acting through cyclin-dependent phosphorylation. Mutations in cdc2/cdc28 result in arrest at the G1/S or G2/M phase of the cell cycle in fission and budding yeast, respectively [72]. Cr treatment perturbed the alignment of microtubules in a concentration-dependent manner in onion roots [69]. Thus, proteomic identification of cell divisionassociated proteins may provide the biochemical basis of the cytotoxicity and/or stress signaling. 4.6. Stress signaling and metal detoxification SAMS synthesizes SAM from l-methionine, which is the major methyl group donor in the transmethylation of proteins and many other substances. The existence and differential expression of different SAM homologues has been generally connected to the metabolic importance of SAM. SAM is the key precursor in the ethylene biosynthesis pathway. Downregulation of two isoforms of SAMS imply a decrease in SAM and ACC levels. It has been demonstrated that Cr inhibits ethylene biosynthesis by reducing availability of ACC for ethylene synthesis in plants and delays senescence [73]. By contrast, other heavy metals such as Cd were shown to induce senescence by accelerated ethylene production in root tissues [74]. The Ca2+ and calmodulin messenger systems have been recognized to be involved in plant-environment interactions [75]. Calmodulin-dependent protein kinases (CPK31; spot 27) are primarily regulated by the Ca2+ /calmodulin complex. Some members of this protein family could act as metal transporters. A calmodulinbinding tobacco plasma membrane protein (NtCBP4) was identified to be similar in structure to cyclic-nucleotide-gated non-selective cation channels. A 2–3-fold overexpression of NtCBP4 enhanced Pb2+ uptake and improved hypersensitivity to Pb2+ , attenuated Ni2+ uptake and improved Ni2+ tolerance in transgenic plants [75]. Thus, the precise roles of specific CPK proteins under Cr toxicity need to be further investigated. Adenine phosphoribosyltransferase (APRT) is a ubiquitous enzyme that functions to specifically salvage adenine by converting it to AMP. The balance of ATP in the cell is maintained by de novo adenine biosynthesis and purine salvage. This protein was downregulated in our experiment (spot 39). Earlier studies demonstrated that the single step APRT pathway is the predominant pathway for the salvaging of adenine in plants under stress conditions [76]. However, APRT has never been reported before to be involved in Cr stress. The increased level of transcript (HvAPT1) and enzyme activity was reported in barley roots following iron deficiency [77]. The authors explained possible role of APRT, formate dehydrogenase (spot 55) and SAMS (spots 13 and 14) in biosynthesis of mugineic acid that chelate iron exclusively in grasses. The biosynthesis of MA begins with the activation of methionine molecules by ATP to form SAM [78]. APRT may function to salvage large amount of adenine during biosynthesis of MA [77]. AMP synthesized by APRT would finally be converted to ATP, which again takes part in the methionine cycle through S-adenylation of methionine. MA chelates both iron and other heavy metals, such as copper, zinc and cobalt [79]. The overall decreases in the expressions of these proteins may be associated with reduced ability of Cr chelation. The inability of plants to sequester Cr by phytochelation has been reported earlier [6,49] and need to be further investigated. 4.7. Miscellaneous and unknown proteins The enzyme UDP-GlcDH converts UDP-glucose into UDPglucuronic acid (UDP-GlcA). Roughly 50% of the cell wall biomass is metabolically derived from UDP-GlcA in Arabidopsis [80]. An up-regulation of UDPGlcDH (spot 75) may be involved in modification of cell wall to prevent metal uptake or block plasmodesmata, inhibiting the symplastic transfer of solutes, increase root apoplastic barriers under Cr toxicity. Increased cell wall elasticity was observed during Cr (VI) treatment [45]. An R40c1 protein (spot 48) was down-regulated, which plays a role in the adaptative response of roots to an hyper-osmotic environment [81]. Transcript level of Osr40cl was increased in response to ABA, conversely negatively regulated by JA and SA. This protein was also down-regulated in rice roots following exposure to As [23]. Since, metal toxicity induces an increased amount of jasmonic acid (JA) and salicylic acid (SA) production in plants [74]. Thus, decreased expression of the Osr40c1 protein in response to Cr may be involved in metal-induced JA production in Miscanthus roots. A putative protein phosphatase 2C (spot 26) is involved in stress signal transduction. A salt-inducible protein homolog (spot 72) was down-regulated. However, the precise role of these proteins is not clear under Cr stress. In addition to the proteins described here, several proteins (spots 1, 37, 56 and 73) were identified as protein of unknown functions. We were unable to correlate their activities with Cr stress. Further studies are needed to address their possible roles in relation to this heavy metal stress. 5. Conclusions In this study, for the first time we investigated chromium stressinduced physiological and biochemical responses, and proteomic changes in M. sinensis roots. A total of 36 proteins were identified that were differentially expressed in chromium-treated root samples. The majority of these proteins were ion transporters, energy and nitrogen metabolism-related proteins, oxidative stress-related regulatory proteins that might work together to establish a new homeostasis in response to chromium stress. The identification of some chromium-responsive proteins might provide new insights to the heavy metal homeostasis as well as helpful to improve this commercially important biomass crop. S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 Acknowledgments This work was supported by the National Research Foundation of Korea (NRF) Grant (NRF-2011-F00013). SA Sharmin, KH Kim and YG Kim are supported by a scholarship; I Alam is supported by a postdoctoral grant from BK21 program at Gyeongsang National University, Republic of Korea. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.plantsci.2012.02.002. References [1] S.K. Panda, H.K. Patra, Physiology of chromium toxicity in plants—a review, Plant Physiol. Biochem. 24 (1997) 10–17. [2] Y.T. Wang, C. Xiao, Factors affecting hexavalent chromium reduction in pure cultures of bacteria, Water Res. 29 (1995) 2467–2474. [3] P. Chandra, S. Sinha, U.N. Rai, Bioremediation of chromium from water and soil by vascular aquatic plants, in: Phytoremediation of Soil and Water Contaminants, American Chemical Society, 1997, pp. 274–282. [4] KrishnamurthyF S., M. Wilkens, Environmental chemistry of Cr, Northeast. Geol. 16 (1994) 14–17. [5] N. Sethunathan, M. Megharaj, L. Smith, S.P.B. Kamaludeen, S. Avudainayagam, R. Naidu, Microbial role in the failure of natural attenuation of chromium(VI) in long-term tannery waste contaminated soil, Agric. Ecosyst. Environ. 105 (2005) 657–661. [6] A.K. Shanker, M. Djanaguiraman, B. Venkateswarlu, Chromium interactions in plants: current status and future strategies, Metallomics 1 (2009) 375–383. [7] J. Aiyar, H.J. Berkovits, R.A. Floyd, K.E. Wetterhahn, Reaction of chromium(VI) with glutathione or with hydrogen peroxide: Identification of reactive intermediates and their role in chromium(VI)-induced DNA damage, Environ. Health Perspect. 92 (1991) 53–62. [8] C. Namasivayam, W.H. Höll, Chromium(III) removal in tannery waste waters using Chinese Reed (Miscanthus sinensis), a fast growing plant, Eur. J. Wood Prod. 62 (2004) 74–80. [9] C.D. Palmer, P.R. Wittbrodt, Processes affecting the remediation of chromiumcontaminated sites, Environ. Health Perspect. 92 (1991) 25–40. [10] Y. Su, F.X. Han, B.B.M. Sridhar, D.L. Monts, Phytotoxicity and phytoaccumulation of trivalent and hexavalent chromium in brake fern, Environ. Toxicol. Chem. 24 (2005) 2019–2026. [11] X.-H. Zhang, J. Liu, H.-T. Huang, J. Chen, Y.-N. Zhu, D.-Q. Wang, Chromium accumulation by the hyperaccumulator plant Leersia hexandra Swartz, Chemosphere 67 (2007) 1138–1143. [12] I. Lewandowski, U. Schmidt, Nitrogen, energy and land use efficiencies of miscanthus, reed canary grass and triticale as determined by the boundary line approach, Agric. Ecosyst. Environ. 112 (2006) 335–346. [13] I. Arduini, A. Masoni, L. Ercoli, Effects of high chromium applications on miscanthus during the period of maximum growth, Environ. Exp. Bot. 58 (2006) 234–243. [14] N. Rascio, F. Navari-Izzo, Heavy metal hyperaccumulating plants: how and why do they do it? And what makes them so interesting? Plant Sci. 180 (2011) 169–181. [15] P. Pongrac, F.-J. Zhao, J. Razinger, A. Zrimec, M. Regvar, Physiological responses to Cd and Zn in two Cd/Zn hyperaccumulating Thlaspi species, Environ. Exp. Bot. 66 (2009) 479–486. [16] S. Quaggiotti, G. Barcaccia, M. Schiavon, S. Nicolé, G. Galla, V. Rossignolo, M. Soattin, M. Malagoli, Phytoremediation of chromium using Salix species: cloning ESTs and candidate genes involved in the Cr response, Gene 402 (2007) 68–80. [17] S. Dubey, P. Misra, S. Dwivedi, S. Chatterjee, S. Bag, S. Mantri, M. Asif, A. Rai, S. Kumar, M. Shri, P. Tripathi, R. Tripathi, P. Trivedi, D. Chakrabarty, R. Tuli, Transcriptomic and metabolomic shifts in rice roots in response to Cr (VI) stress, BMC Genomics 11 (2010) 648. [18] M. Labra, E. Gianazza, R. Waitt, I. Eberini, A. Sozzi, S. Regondi, F. Grassi, E. Agradi, Zea mays L. protein changes in response to potassium dichromate treatments, Chemosphere 62 (2006) 1234–1244. [19] J.B. Jones Jr., B. Wolf, H.A. Mills, Plant Analysis Handbook. A Practical Sampling, Preparation, Analysis, and Interpretation Guide, Micro-Macro Publishing, Inc., Athens, Georgia, 1991. [20] I. Alam, S.A. Sharmin, K.-H. Kim, J. Yang, M. Choi, B.-H. Lee, Proteome analysis of soybean roots subjected to short-term drought stress, Plant Soil 333 (2010) 491–505. [21] R.L. Heath, L. Packer, Photoperoxidation in isolated chloroplasts: I. Kinetics and stoichiometry of fatty acid peroxidation, Arch. Biochem. Biophys. 125 (1968) 189–198. [22] I. Alam, D.-G. Lee, K.-H. Kim, C.-H. Park, S. Sharmin, H. Lee, K.-W. Oh, B.-W. Yun, B.-H. Lee, Proteome analysis of soybean roots under waterlogging stress at an early vegetative stage, J. Biosci. 35 (2010) 49–62. 125 [23] N. Ahsan, D.G. Lee, I. Alam, P.J. Kim, J.J. Lee, Y.O. Ahn, S.S. Kwak, I.J. Lee, J.D. Bahk, K.Y. Kang, J. Renaut, S. Komatsu, B.H. Lee, Comparative proteomic study of arsenic-induced differentially expressed proteins in rice roots reveals glutathione plays a central role during As stress, Proteomics 8 (2008) 3561–3576. [24] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275. [25] I. Alam, S.A. Sharmin, K.-H. Kim, Y.-G. Kim, J.J. Lee, J.D. Bahk, B.-H. Lee, Comparative proteomic approach to identify proteins involved in flooding combined with salinity stress in soybean, Plant Soil 346 (2011) 45–62. [26] J.R. Stewart, Y.O. Toma, F.G. FernÁNdez, A.Y.A. Nishiwaki, T. Yamada, G. Bollero, The ecology and agronomy of Miscanthus sinensis, a species important to bioenergy crop development, in its native range in Japan: a review, GCB Bioenergy 1 (2009) 126–153. [27] B. Ezaki, E. Nagao, Y. Yamamoto, S. Nakashima, T. Enomoto, Wild plants, Andropogon virginicus L. and Miscanthus sinensis Anders, are tolerant to multiple stresses including aluminum, heavy metals and oxidative stresses, Plant Cell Rep. 27 (2008) 951–961. [28] I. Arduini, L. Ercoli, A. Masoni, Response of Miscanthus sinensis to long term application of low Cr levels in hydroponics, Agrochimica 50 (2006) 187–199. [29] M. Thompson, T. Thorpe, Metabolic and non metabolic role of carbohydrates, in: J. Bonga, D. Durzan (Eds.), Cell and Tissue Culture in Forestry, Martinus Nijhoff, Dordrecht, 1987, pp. 89–112. [30] R. Sperotto, T. Boff, G. Duarte, J. Fett, Increased senescence-associated gene expression and lipid peroxidation induced by iron deficiency in rice roots, Plant Cell Rep. 27 (2008) 183–195. [31] Y. Yamamoto, Y. Kobayashi, H. Matsumoto, Lipid peroxidation is an early symptom triggered by aluminum, but not the primary cause of elongation inhibition in pea roots, Plant Physiol. 125 (2001) 199–208. [32] S. Verma, R.S. Dubey, Lead toxicity induces lipid peroxidation and alters the activities of antioxidant enzymes in growing rice plants, Plant Sci. 164 (2003) 645–655. [33] S.K. Panda, Chromium-mediated oxidative stress and ultrastructural changes in root cells of developing rice seedlings, J. Plant Physiol. 164 (2007) 1419–1428. [34] S.C. Carpentier, B. Panis, A. Vertommen, R. Swennen, K. Sergeant, J. Renaut, K. Laukens, E. Witters, B. Samyn, B. Devreese, Proteome analysis of non-model plants: a challenging but powerful approach, Mass Spectrom. Rev. 27 (2008) 354–377. [35] B. Sghaier-Hammami, L. Valledor, N. Drira, J.V. Jorrin-Novo, Proteomic analysis of the development and germination of date palm (Phoenix dactylifera L.) zygotic embryos, Proteomics 9 (2009) 2543–2554. [36] P.M. Lonosky, X. Zhang, V.G. Honavar, D.L. Dobbs, A. Fu, S.R. Rodermel, A proteomic analysis of maize chloroplast biogenesis, Plant Physiol. 134 (2004) 560–574. [37] J.L. Hall, Cellular mechanisms for heavy metal detoxification and tolerance, J. Exp. Bot. 53 (2002) 1–11. [38] M.D. Vázquez, C. Poschenrieder, J. Barceló, Chromium VI induced structural and ultrastructural changes in bush bean plants (Phaseolus vulgaris L.), Ann. Bot. 59 (1987) 427–438. [39] F.A. Austenfeld, The effect of Ni, Co and Cr on net photosynthesis of primary and secondary leaves of PhaseoIus vulgaris L., Photosynthetica 13 (1979) 434–438. [40] E.R. Sumner, A. Shanmuganathan, T.C. Sideri, S.A. Willetts, J.E. Houghton, S.V. Avery, Oxidative protein damage causes chromium toxicity in yeast, Microbiology 151 (2005) 1939–1948. [41] N. Ahsan, D.G. Lee, K.H. Kim, I. Alam, S.H. Lee, K.W. Lee, H. Lee, B.H. Lee, Analysis of arsenic stress-induced differentially expressed proteins in rice leaves by two-dimensional gel electrophoresis coupled with mass spectrometry, Chemosphere 78 (2010) 224–231. [42] T. Fukuda, A. Saito, J. Wasaki, T. Shinano, M. Osaki, Metabolic alterations proposed by proteome in rice roots grown under low P and high Al concentration under low pH, Plant Sci. 172 (2007) 1157–1165. [43] K. Stobrawa, G. Lorenc-Plucinska, Changes in carbohydrate metabolism in fine roots of the native European black poplar (Populus nigra L.) in a heavy-metalpolluted environment, Sci. Total Environ. 373 (2007) 157–165. [44] P. Kieffer, S.b. Planchon, M. Oufir, J. Ziebel, J. Dommes, L. Hoffmann, J.-F. Hausman, J. Renaut, Combining proteomics and metabolite analyses to unravel cadmium stress-response in poplar leaves, J. Proteome Res. 8 (2008) 400–417. [45] J. Barceló, C. Poschenrieder, B. Gunsé, Water relations of chromium VI treated bush bean plants (Phaseolus vulgaris L. cv. contender) under both normal and water stress conditions, J. Exp. Bot. 37 (1986) 178–187. [46] A. Brune, W. Urbach, K.-J. Dietz, Differential toxicity of heavy metals is partly related to a loss of preferential extraplasmic compartmentation: a comparison of Cd-, Mo-, Ni- and Zn-stress, New Phytol. 129 (1995) 403–409. [47] J. Verkleij, P. Koevoets, M. Blake-Kalff, A. Chardonnens, Evidence for an important role of the tonoplast in the mechanism of naturally selected zinc tolerance in Silene vulgaris, J. Plant Physiol. 153 (1998) 188–191. [48] A.N. Chardonnens, P.L.M. Koevoets, A. van Zanten, H. Schat, J.A.C. Verkleij, Properties of enhanced tonoplast zinc transport in naturally selected zinc-tolerant Silene vulgaris, Plant Physiol. 120 (1999) 779–786. [49] S.K. Panda, S. Choudhury, Chromium stress in plants, Braz. J. Plant Physiol. 17 (2005) 95–102. [50] D.E. Salt, G.J. Wagner, Cadmium transport across tonoplast of vesicles from oat roots. Evidence for a Cd2+ /H+ antiport activity, J. Biol. Chem. 268 (1993) 12297–12302. [51] A. Gonzalez, V. Koren’Kov, G.J. Wagner, A comparison of Zn, Mn, Cd, and Ca transport mechanisms in oat root tonoplast vesicles, Physiol. Plant 106 (1999) 203–209. 126 S.A. Sharmin et al. / Plant Science 187 (2012) 113–126 [52] K.S. Schumaker, H. Sze, Calcium transport into the vacuole of oat roots. Characterization of H+ /Ca2+ exchange activity, J. Biol. Chem. 261 (1986) 12172–12178. [53] S. AbuQamar, H. Luo, K. Laluk, M.V. Mickelbart, T. Mengiste, Crosstalk between biotic and abiotic stress responses in tomato is mediated by the AIM1 transcription factor, Plant J. 58 (2009) 347–360. [54] L.S. Graham, M.B. Sticklen, Plant chitinases, Can. J. Bot. 72 (1994) 1057–1083. [55] S. Jacobsen, M.Z. Hauschild, U. Rasmussen, Induction by chromium ions of chitinases and polyamines in barley (Hordeum vulgare L.) and rape (Brassica napus L. ssp. oleifera), Plant Sci. 84 (1992) 119–128. [56] B. Békésiová, Š. Hraška, J. Libantová, J. Moravčíková, I. Matušíková, Heavy-metal stress induced accumulation of chitinase isoforms in plants, Mol. Biol. Rep. 35 (2008) 579–588. [57] M.d.l.M. Dana, J.A. Pintor-Toro, B. Cubero, Transgenic tobacco plants overexpressing chitinases of fungal origin show enhanced resistance to biotic and abiotic stress agents, Plant Physiol. 142 (2006) 722–730. [58] J.D.G. Jones, J.L. Dangl, The plant immune system, Nature 444 (2006) 323–329. [59] W. Wang, B. Vinocur, O. Shoseyov, A. Altman, Role of plant heat-shock proteins and molecular chaperones in the abiotic stress response, Trends Plant Sci. 9 (2004) 244–252. [60] N. Suzuki, N. Koizumi, H. Sano, Screening of cadmium-responsive genes in Arabidopsis thaliana, Plant Cell Environ. 24 (2001) 1177–1188. [61] M.C. Romero-Puertas, J.M. Palma, M. Gómez, L.A. Del Río, L.M. Sandalio, Cadmium causes the oxidative modification of proteins in pea plants, Plant Cell Environ. 25 (2002) 677–686. [62] M. González-Fernández, T. García-Barrera, A. Arias-Borrego, J. Jurado, C. Pueyo, J. López-Barea, J.L. Gómez-Ariza, Metallomics integrated with proteomics in deciphering metal-related environmental issues, Biochimie 91 (2009) 1311–1317. [63] L.M. Sandalio, H.C. Dalurzo, M. Gómez, M.C. Romero-Puertas, L.A. del Río, Cadmium-induced changes in the growth and oxidative metabolism of pea plants, J. Exp. Bot. 52 (2001) 2115–2126. [64] J.F. Palatnik, N. Carrillo, E.M. Valle, The role of photosynthetic electron transport in the oxidative degradation of chloroplastic glutamine synthetase, Plant Physiol. 121 (1999) 471–478. [65] J.L. Ortega, D. Roche, C. Sengupta-Gopalan, Oxidative turnover of soybean root glutamine synthetase. In vitro and in vivo studies, Plant Physiol. 119 (1999) 1483–1496. [66] L.-M. Chen, C. Huei Kao, Relationship between ammonium accumulation and senescence of detached rice leaves caused by excess copper, Plant Soil 200 (1998) 169–173. [67] S.E. Stevens, C. Van Baalen, Control of nitrate reductase in a blue-green alga: the effects of inhibitors, blue light, and ammonia, Arch. Biochem. Biophys. 161 (1974) 146–152. [68] C. Rigano, V. Di Martino Rigano, V. Vona, A. Fuggi, G. Aliotta, Studies in vivo on the control by ammonia of nitrate reduction to nitrite in the unicellular alga Cyanidium caldarium, Plant Sci. Lett. 13 (1978) 301–307. [69] E. Eleftheriou, I.-D. Adamakis, P. Melissa, Effects of hexavalent chromium on microtubule organization, ER distribution and callose deposition in root tip cells of Allium cepa L., Protoplasma (2011) 1–16. [70] P. Percipalle, N. Visa, Molecular functions of nuclear actin in transcription, J. Cell Biol. 172 (2006) 967–971. [71] M. Amenós, I. Corrales, C. Poschenrieder, P. Illéš, F. Baluška, J. Barceló, Different effects of aluminum on the actin cytoskeleton and brefeldin A-sensitive vesicle recycling in root apex cells of two maize varieties differing in root elongation rate and aluminum tolerance, Plant Cell Physiol. 50 (2009) 528–540. [72] J. Hindley, G.A. Phear, Sequence of the cell division gene CDC2 from Schizosaccharomyces pombe; patterns of splicing and homology to protein kinases, Gene 31 (1984) 129–134. [73] C. Poschenrieder, B. Gunsé, J. Barceló, Chromium-induced inhibition of ethylene evolution in bean (Phaseolus vulgaris) leaves, Physiol. Plant 89 (1993) 404–408. [74] M. Rodríguez-Serrano, M.C. Romero-Puertas, A. Zabalza, F.J. Corpas, M. Gómez, L.A. Del Río, L.M. Sandalio, Cadmium effect on oxidative metabolism of pea (Pisum sativum L.) roots. Imaging of reactive oxygen species and nitric oxide accumulation in vivo, Plant Cell Environ. 29 (2006) 1532–1544. [75] T. Arazi, B. Kaplan, R. Sunkar, H. Fromm, Cyclic-nucleotide-and Ca2+ /calmodulin-regulated channels in plants: targets for manipulating heavy-metal tolerance, and possible physiological roles, Biochem. Soc. Trans. 28 (2000) 471. [76] B. Moffatt, C. Pethe, M. Laloue, Metabolism of benzyladenine is impaired in a mutant of Arabidopsis thaliana lacking adenine phosphoribosyltransferase activity, Plant Physiol. 95 (1991) 900–908. [77] R. Itai, K. Suzuki, H. Yamaguchi, H. Nakanishi, N.K. Nishizawa, E. Yoshimura, S. Mori, Induced activity of adenine phosphoribosyltransferase (APRT) in irondeficient barley roots: a possible role for phytosiderophore production, J. Exp. Bot. 51 (2000) 1179–1188. [78] J.F. Ma, G. Kusano, S. Kimura, K. Nomoto, Specific recognition of mugineic acid–ferric complex by barley roots, Phytochemistry 34 (1993) 599–603. [79] T. Iwashita, Y. Mino, H. Naoki, Y. Sugiura, K. Nomoto, High-resolution proton nuclear magnetic resonance analysis of solution structures and conformational properties of mugineic acid and its metal complexes, Biochemistry 22 (1983) 4842–4845. [80] E. Zablackis, J. Huang, B. Muller, A.G. Darvill, P. Albersheim, Characterization of the cell-wall polysaccharides of Arabidopsis thaliana leaves, Plant Physiol. 107 (1995) 1129–1138. [81] A. Moons, J. Gielen, J. Vandekerckhove, D.V.D. Straeten, G. Gheysen, M.V. Montagu, An abscisic-acid- and salt-stress-responsive rice cDNA from a novel plant gene family, Planta 202 (1997) 443–454.