Iron toxicity in field-cultivated rice: contrasting tolerance mechanisms ...

Theor. Exp. Plant Physiol. (2014) 26:135–146 DOI 10.1007/s40626-014-0013-3

Iron toxicity in field-cultivated rice: contrasting tolerance mechanisms in distinct cultivars Ricardo Jose´ Stein • Se´rgio Iraçu Gindri Lopes Janette Palma Fett

•

Received: 25 March 2014 / Accepted: 28 May 2014 / Published online: 24 June 2014 Ó Brazilian Society of Plant Physiology 2014

Abstract Iron toxicity is a major nutritional disorder in irrigated and rainfed waterlogged rice. To elucidate mechanisms involved in tolerance to iron toxicity, plants from one cultivar susceptible to iron toxicity (BRIRGA 409) and two tolerant cultivars (EPAGRI 108 and EPAGRI 109) were grown in the field, at an iron-toxic site and at a control site in Southern Brazil. We evaluated chlorophyll concentrations, carbonyl concentrations, iron concentrations in leaves and roots, antioxidative enzyme activities (SOD, APX, CAT, GR and DHR), concentrations of reduced and oxidized forms of ascorbate and glutathione, and gene expression profile of four SOD genes in rice leaves. Only plants

R. J. Stein J. P. Fett Plant Physiology Laboratory, Centro de Biotecnologia & Departamento de Botaˆnica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil Present Address: R. J. Stein Department of Plant Physiology, Fakulta¨t fu¨r Biologie und Biotechnologie - Ruhr Universita¨t, Bochum, Germany S. I. G. Lopes Divisaõ de Pesquisa, Instituto Rio Grandense do Arroz (IRGA), Cachoeirinha, RS, Brazil J. P. Fett (&) Centro de Biotecnologia, Universidade Federal do Rio Grande do Sul, Caixa Postal 15005, Porto Alegre, RS CEP 91501-970, Brazil e-mail: [email protected]

from the susceptible cultivar showed symptoms of iron toxicity when grown at the iron-toxic site, accumulating high levels of iron in leaves. EPAGRI 108 plants had the lowest iron concentration in leaves and reached the highest iron concentration in the root symplast, suggesting that the capacity to safely store iron in root cells and to limit iron translocation to shoots could be a tolerance mechanism in this cultivar. Plants from the susceptible cultivar showed higher APX activity as well as higher DHA and GSSG concentrations. Plants from the EPAGRI 109 cultivar accumulated high iron levels in leaves, and showed the highest SOD, GR and DHR activities when grown in the iron-toxic site. The same cultivar also showed the highest expression of three out of four SOD genes tested. Therefore, the two tolerant cultivars seem to rely on different mechanisms to deal with iron toxicity in field conditions: limiting iron translocation to the shoot or inducing enzymes-dependent leaf tolerance. Keywords Abiotic stress Antioxidative enzymes Oxidative stress Superoxide dismutase

1 Introduction Iron toxicity is a major nutritional disorder in lowland and waterlogged rice. While it may occur in a wide range of soil types, general characteristics of most of the iron-toxic soils are high amounts of reducible iron, low pH, and low cation exchange capacity and

123

136

exchangeable K content (Ottow et al. 1982). Most importantly, iron toxicity is linked to waterlogging and only occurs under anoxic soil conditions, with the reduction of iron oxides and its solubilization in the soil solution (Ponnamperuma 1972). Losses in rice productivity caused by iron toxicity commonly vary between 15 and 20 %; however, in the most severe cases, complete crop failure can occur (Winslow et al. 1989, Audebert and Sahrawat 2000). Rice cultivars with variable degrees of resistance to iron toxicity have been obtained by breeding (Fageria and Rabelo 1987; Sahrawat et al. 1996), and some cultural practices have been used to minimize iron toxicity, such as alternative planting date, ridge planting, water management and the use of fertilizers (Benckiser et al. 1984; Winslow et al. 1989). However, due to the diversity in environmental conditions where iron toxicity occurs, none of those options is universally applicable or efficient (Becker and Asch 2005). Among the possible mechanisms used by distinct rice cultivars to resist high levels of iron, the involvement of the root exclusion power, oxidizing iron at the root surface and leading to the formation of the iron plaque, has been suggested as a mechanism used by resistant rice cultivars to exclude high amounts of iron in the soil solution from the plant body (Green and Etherington 1977; Ando et al. 1983), avoiding higher iron accumulation in leaf tissue. Another possible mechanism involved in the limitation of excessive uptake of iron could be the regulation of its uptake, as seen by Silveira et al. (2007) in rice plants grown in nutrient solutions. Development of leaf tissue tolerance to high levels of iron, with the induction of the leaf antioxidant system, has been suggested as another mechanism that can be used by rice cultivars to resist high levels of iron in shoots (Yamauchi and Peng 1995, Wu et al. 1998), as well as the capacity to keep lower proportion of iron present in low-molecular-mass fractions (Stein et al. 2009). Therefore, it has been accepted that resistance to iron toxicity may be achieved either by Fe exclusion from the symplast (avoidance mechanism, by accumulation in the apoplast) or from tolerance to high intracellular Fe levels (Sahrawat 2004). Those resistance mechanisms were independently identified by diverse research groups, using distinct rice cultivars and different methodological approaches, and based on plants grown in diverse laboratory settings or in field conditions with high soil heterogeneity. Also, physiological data concerning the mechanisms involved

123

Theor. Exp. Plant Physiol. (2014) 26:135–146

in varietal tolerance (or susceptibility) to iron toxicity and the comparison of different tolerant cultivars grown in field conditions are poorly available. This problem is even more relevant when considering that there are at least two distinct types of toxicity described in the literature: true (or real) iron toxicity (Sahrawat 2000; Olaleye et al. 2001; Stein et al. 2009) and indirect toxicity (Benckiser et al. 1984), related to other mineral imbalances. Although recent efforts have been made to understand the impact of iron excess in rice under laboratory conditions (Ricachenevsky et al. 2010; Quinet et al. 2012; Pereira et al. 2013), the interaction between iron excess and different rice genotypes in field conditions still needs to be clarified. Therefore, there is an urgent need to understand how different rice genotypes deal with iron toxicity in the field, as well as how some of them manage to tolerate high leaf iron concentrations. Based on this, this work aimed to identify and characterize resistance mechanisms to iron excess used by two closely related cultivars grown in the field at two different sites, an iron-toxic and a control site.

2 Materials and methods 2.1 Experimental site and plant material Rice plants (Oryza sativa ssp. indica) from cultivars EPAGRI 108, EPAGRI 109 (both tolerant to iron excess) and BR-IRGA 409 (susceptible) were grown in two distinct sites in Brazil: in an iron-toxic site with recognized history of iron toxicity in Camaqua˜, RS, Brazil (308 540 07.9600 S 518 510 26.2500 W), and a control site (without history of iron toxicity) in Cachoeirinha, RS, Brazil (298 560 51.9100 S 518 060 46.3600 W). Both sites are at the same altitude (about sea level) and have the same climate (humid subtropical Cfa, according to the Ko¨ppen–Geiger climate classification system—Peel et al. 2007), with temperatures ranging from 18 to 35 °C and long days (13:12 to 14:04 h of light) during the rice growing season (late Spring and Summer; sowing on December 18th). Soil characteristics of both sites are described in Stein et al. (2009), where pH values were low in both locations (5.2 and 4.8 in Camaqua˜ and Cachoeirinha, respectively). Both also had low cation exchangeable content, indicative of low soil fertility. Iron concentrations in the soil solution were much higher in Camaqua˜ (284 mg L-1) than in Cachoeirinha (29 mg L-1).


The experimental design was completely randomized. Plants were grown in consecutive rows of 1.0 m, alternating the cultivars, with spaces of 0.3 m between rows. Sowing was manual in fields mechanically prepared in the same way in both locations. No soil fertilizer was applied. Irrigation was done by flooding, 10 days after plant emergence. Samples were harvested randomly. Plants were collected at the V8 stage (Counce et al. 2000), which corresponded to about 70 days after emergence. Plants were separated in shoots and roots, immediately frozen in liquid nitrogen, transferred to the laboratory in containers with liquid N2 and stored at -80 °C until further analyses. Each individual sample used in chemical analyses combined plant material (roots or shoots) from at least four plants, depending on the biomass necessary for each analysis. The rice cultivars used in this study were characterized as susceptible and tolerant to iron toxicity by the Rice Breeding Group of the Instituto Rio Grandense do Arroz (IRGA, Brazil), following the methodology proposed by Bacha and Ishiy (1986). The susceptible cultivar was released to growers in 1979, and is still highly cultivated in Rio Grande do Sul state, the main rice producer state in Brazil. The two tolerant cultivars were released to growers in 1995 (EPAGRI 108) and 1996 (EPAGRI 109). They are derived from sister lines, originated from the same cross (CT 7347/IR 21015-72-3-3-3-1). Although tolerant to iron toxicity, these cultivars have not been highly planted in recent years.

137

derivatization with 2,4-dinitrophenyl-hydrazine. Fully expanded leaves were ground in cold extraction buffer [50 mM Tris (pH 8.0), 2 mM EDTA, 1 mM PMSF and 1 mM benzamidine], centrifuged at 12,0009g for 15 min at 4 °C, and the supernatants immediately treated with 10 % streptomycin sulphate to eliminate contaminant nucleic acids, and readily used for carbonyl determination according to Levine et al. (1990). The carbonyl concentration was normalized to the soluble protein concentration, determined by the dye-binding method (Bradford 1976), using BSA as standard. 2.4 Iron determination in plant material Iron concentration was determined in the plant material from the three cultivars (EPAGRI 108, EPAGRI 109 and BR-IRGA 409) grown at the two distinct sites. Roots were collected, thoroughly washed in abundant distilled water to remove excess soil and particulated material, and used for iron determination or immediately kept for 3 h in cold DCB (dithionite-citrate-bicarbonate) solution (Taylor and Crowder 1983) to remove the iron precipitated as iron plaque. Samples (fully expanded leaves, DCB-treated or non-treated roots) were dried at 60 °C and ashed at 500 °C for 3 h. The ashes were digested with concentrated HCl and iron was quantified by atomic absorption spectrophotometry (Varian-Model Spectra 10/20, Victoria, AU). 2.5 Ascorbate and glutathione determination

2.2 Chlorophyll determination Fully expanded leaves were ground in liquid nitrogen in a mortar and pestle and chlorophyll extracted in 85 % acetone. The extracts were centrifuged at 7,5009g for 3 min at 4 °C. Chlorophylls were quantified in the supernatants by measuring absorbance at 663 and 645 nm (spectrophotometer Cintra 5, GBC Scientific Equipment, Victoria, AU) and the concentrations of total chlorophyll (chlorophyll a ? chlorophyll b) calculated according to Ross (1974). 2.3 Oxidative damage to proteins The oxidative damage to proteins was determined through the quantification of carbonyl groups, by

Ascorbate (AA) and dehydroascorbate (DHA) were extracted from leaf tissue (0.5–0.6 g) with 1 ml of TCA 6 %, centrifuged at 12,0009g (10 min at 4 °C) and the supernatant collected and immediately used for AA and DHA determination, according to Okamura (1980). Glutathione (GSH) and glutathione disulfide (GSSG) were extracted with 0.8 ml of 10 % HClO4 and determined according to Griffith (1980) by the 5,50 -dithiobis-(2-nitrobenzoic acid)-GR recycling procedure. Changes in absorbance on the reaction mixture were followed at 412 nm and total GSH concentration was calculated from a standard curve with GSH. GSSG was determined after removal of GSH by 2-vinylpyridine derivatization. A specific standard curve with GSSG was used, and GSH

123

138

determined by subtraction of GSSG from the total glutathione (GSH ? GSSG) concentration. 2.6 Activities of antioxidative enzymes For all enzymatic activity determinations, fully expanded leaves were ground in cold extraction buffer [50 mM HEPES (pH 7.4), 1 % PVP, 1 mM EDTA and 1 mM PMSF], centrifuged at 12,0009g (15 min at 4 °C) and the supernatants immediately used for enzymatic assays. The activity of ascorbate peroxidase (APX) was determined according to Klapheck et al. (1990), from the decrease in absorbance at 290 nm. Catalase (CAT) activity was determined following the decrease of absorbance at 240 nm due to H2O2 consumption (Cakmak and Marschner 1992). Superoxide dismutase (SOD) activity was measured as described by Beyer and Fridovich (1987), using 15 min of illumination and recording the absorbance at 560 nm. The activity of glutathione reductase (GR) was determined according to Sgherri et al. (1994), following the NADPH consumption at 340 nm, and the dehydroascorbate reductase (DHAR) activity determined according to Kato et al. (1997), following the reduction of DHA at 265 nm. Conditions for all assays were chosen so that the rate of reaction was constant during all the experimental period and proportional to the amount of protein extract added. All enzymatic activities were assayed in triplicate at 25 °C, with no lag period, and protein was quantified by the dye-binding method (Bradford 1976). 2.7 RNA extraction and cDNA synthesis Total RNA samples were extracted from fully expanded rice leaves using the Concert Plant RNA Purification Kit (Invitrogen, Carlsbad, CA, USA), according to the manufacturer instructions. RNA quality was evaluated by determining the ratio 260 nm/280 nm absorbance (selecting only samples that ranged from 1.9 to 2.1) and integrity by electrophoresis in 1.2 % agarose gels. RNA was quantified using the Quant-iT RNA Assay Kit and the Qubit Fluorometer (Invitrogen, Carlsbad, CA, USA). Two micrograms of total RNA were treated with DNAse I (Invitrogen, Carlsbad, CA, USA) to avoid genomic DNA contamination interference, and cDNA synthesized using M-MLV reverse transcriptase (Invitrogen, Carlsbad, CA, USA) and primer oligo-dT(30).

123


2.8 Semi-quantitative RT-PCR Semi-quantitative RT-PCR was performed using standard conditions and PCR products analyzed only in the linear phase of amplification. Diluted cDNA (1:10 in water) was amplified with specific primers for four different SOD isoforms (Kim et al. 2007): MnSOD (L34038, 50 -GGAAACAACTGCTAACCAGGAC-30 , 50 -GCAATGTACACAAGGTCCAGAA-30 ), FeSOD (AB014056, 50 -TGCACTTGGTGATATTCCACTC30 , 50 -CGAATCTCAGCATCAGGTATCA-30 ) and two Cu/ZnSOD (D852339, 50 -CAATGCTGAAGGTG TAGCTGAG-30 , 50 -GCGAAATCCATGTGATACAAGA-30 ; L19435, 50 -GGTTTTGGTGCTCTTTTAGG TG-30 , 50 -GCCACTCAGGTAAAGACGAAAC-30 ). The PCR products were resolved in 1.2 % agarose gels and stained with ethidium bromide. Expression of the ubiquitin gene (Miki et al. 2005) was used as the constitutive control (50 -AACCAGCTGAGGCCCAA GA-30 , 50 -ACGATTGATTTAACCAGTCCATGA-30 ). 2.9 Statistical analysis To achieve normality and homoscedasticity all data were log10 transformed, and compared using two-way analysis of variance (ANOVA) with cultivar (EPAGRI 109, EPAGRI 108, BR-IRGA 409) and site (Camaqua˜ and Cachoeirinha) as main effects, and the interaction between the two factors (cultivar x site). Pairwise T-test with Bonferroni correction and Tukey post hoc tests were conducted to detect differences between the groups of samples. Differences were considered significant when P B 0.05. All analyses were performed using the R statistical software (R Development Core Team, 2011).

3 Results 3.1 Physiological characterization of rice cultivars Plants from the susceptible cultivar (BR-IRGA 409) showed visible typical symptoms of iron toxicity when grown at the iron-toxic site (Camaqua˜), such as discoloration and necrosis in older leaves, which were not seen in the tolerant cultivars (EPAGRI 108 and EPAGRI 109, data not shown). Generally, plants grown at the iron-toxic site showed reduced levels of chlorophyll (with F = 12.7, P \ 0.05), and a


139

concentration was detected on leaves from the sensitive cultivar (M = 11.02, SD = 0.54) than in the tolerant cultivars EPAGRI 109 (M = 3.64, SD = 0.67) and EPAGRI 108 (M = 4.48, SD = 0.79). These results validated the previous classification from the Instituto Rio Grandense do Arroz, Brazil and allowed us to further dissect the physiological responses of the chosen cultivars. 3.2 Iron concentration in leaves and roots

Fig. 1 Physiological characterization of tolerance to iron toxicity in three rice cultivars (EPAGRI 109, EPAGRI 108 and BR-IRGA 409). Chlorophyll concentration (A) and oxidative stress in proteins, indicated by the carbonyl concentration (B). Plants were grown in an iron-toxic site (Camaqua˜-RS, Brazil) or in a control site (Cachoeirinha-RS, Brazil). Values represent the means of six replicates ± standard error. Data was analyzed by two-way ANOVA. Significant interaction between site and cultivar were detected, and distinct letters indicate statistical difference by the Tukey test (P B 0.05)

interaction between site and cultivar could be detected (F = 87.5, P \ 0.05), with the cultivars EPAGRI 109 (M = 11.12, SD = 0.51) and EPAGRI 108 (M = 11.22, SD = 0.64) showing higher chlorophyll concentration than the sensitive cultivar (BR-IRGA 409; M = 6.03, SD = 0.37) when grown at the iron-toxic site (Fig. 1a). The oxidative damage to proteins, as indicated by high carbonyl concentrations, was significantly different depending on cultivar (F = 6.6, P \ 0.05), site (F = 46.6, P \ 0.05) and their interaction (F = 31.9, P \ 0.05). Similarly to the leaf chlorophyll concentration, the leaf carbonyl concentration was higher on plants grown at the iron-toxic site (Fig. 1b), and a much higher leaf carbonyl

The leaf iron concentrations varied between samples, with significant main effects being detect for cultivar (F = 6.2, P \ 0.05), site (F = 197.1, P \ 0.05) and also for the interaction between the two factors (F = 5.6, P \ 0.05). All cultivars showed higher leaf iron concentration when grown at the iron-toxic site, with the tolerant cultivar EPAGRI 109 (M = 3.41, SD = 0.46) and the sensitive cultivar (BR-IRGA 409; M = 3.05, SD = 0.60) showing higher leaf iron concentrations than the tolerant cultivar EPAGRI 108 (M = 1.66, SD = 0.53). To detail the root iron accumulation in the studied cultivars, we collected roots from the two distinct sites and quantified the concentrations of total iron (DCBuntreated roots), intracellular iron (DCB-treated fraction) and apoplastic iron (total root iron concentration minus the DCB-treated root iron concentration) (Fig. 2b–d). Treatment with DCB solution resulted in a remarkable decrease in iron concentrations (86.7– 94.9 % reduction, depending on cultivar and site respectively), confirming the effectiveness of the DCB solution in solubilizing the iron precipitated in the root extraplasmatic space. Total iron concentration in roots was significantly influenced by the site (F = 9.7, P = 0.006), with plants accumulating higher concentration of iron in Camaqua˜ (the iron-toxic site) than in Cachoeirinha (the control site) (Fig. 2b). The same pattern was seen for the apoplastic fraction, being significantly affected only by the site (F = 6.8, P = 0.017). A different pattern was detected for the intracellular fraction, with cultivar (F = 12.1, P \ 0.05), site (F = 146.6, P \ 0.05), and the interaction (F = 76.8, P \ 0.05) showing significant effects. Along with the other fractions (seen in Fig. 2c, d), the root intracellular iron concentration was higher in plants growing at the iron-toxic site, with roots from the tolerant cultivar EPAGRI 108 accumulating higher intracellular iron concentrations (M = 7.10, SD = 0.23) than

123

140


Fig. 2 Iron accumulation and distribution in leaves (A), total Fe in roots (B), root intracellular fraction (C), and root apoplastic fraction (D) of rice plants from cultivars EPAGRI 109, EPAGRI 108 and BR-IRGA 409 grown in an iron-toxic site (Camaqua˜-RS, Brazil) or in a control site (Cachoeirinha-RS, Brazil). Values represent the mean ± standard error (n = 6 for leaves; n = 4 for

roots). Data was analyzed by two-way ANOVA. In cases when significant interaction between site and cultivar was detected, distinct letters indicate statistical difference by the Tukey test (P B 0.05). When ANOVA indicated only significant effect for the cultivation site, *** indicates significant statistical difference between sites by pairwise T-test

roots from the tolerant cultivar EPAGRI 109 (M = 2.52, SD = 0.21) and the sensitive cultivar BR-IRGA 409 (M = 1.93, SD = 0.26). The results indicate that plants grown at the irontoxic site accumulated higher iron concentrations on both organs than at the control site, and that the two tolerant cultivars differ in the iron accumulation pattern. While EPAGRI 109 accumulated iron at levels similar to the susceptible BR-IRGA 409 plants, EPAGRI 108 accumulated lower levels of iron in leaves but higher levels in its roots.

oxidized forms of ascorbate and glutathione (Table 1). The ascorbate (AA) concentration in leaves was significantly different between cultivars (F = 4.95, P = 0.014) and on the interaction between site and cultivar (F = 5.24, P \ 0.011), with leaves from the sensitive cultivar (BR-IRGA 409) showing the lowest concentration of AA when grown at the iron-toxic site (M = 0.84, SD = 0.07). A very similar pattern was observed for the leaf reduced glutathione (GSH) concentration, with a significant effect observed on cultivar (F = 60.42, P \ 0.05) and on the interaction between cultivar and site (F = 30.98, P \ 0.05). Similarly to DHA, plants from the sensitive cultivar showed the highest leaf glutathione disulfide (the oxidized form of GSH) concentration (M = 2.85, SD = 0.55), with site (F = 10.53, P \ 0.05) and the interaction between site and cultivar (F = 36.04,

3.3 Leaf antioxidative status (ascorbate and glutathione) To access the leaf antioxidative status of the plants we quantified the concentrations of the reduced and

123


141

Table 1 Accumulation of reduced (AA and GSH) and oxidized (DHA and GSSG) forms of ascorbate and glutathione in fully expanded leaves of three rice cultivars (EPAGRI 108, EPAGRI 108

EPAGRI 109 and BR-IRGA 409) grown in an iron-toxic (Camaqua˜-RS, Brazil) and a control site (Cachoeirinha-RS, Brazil) EPAGRI 109

BR-IRGA 409

Camaqua˜

Cachoeirinha

Camaqua˜

Cachoeirinha

Camaqua˜

Cachoeirinha

AA (lmol mg-1 DW)

1.14 ± 0.13b

1.18 ± 0.05b

1.10 ± 0.03b

1.73 ± 0.06ab

0.84 ± 0.07c

1.42 ± 0.16ab

DHA (lmol mg-1 DW)

N.D.

N.D.

N.D.

N.D.

0.48 ± 0.03

N.D.

GSH (lmol mg-1 DW)

11.22 ± 0.67bc

12.29 ± 0.48bc

10.58 ± 0.68c

13.79 ± 1.46b

11.4 ± 1.10c

19.75 ± 1.01a

GSSG (lmol mg-1 DW)

0.28 ± 0.02c

0.31 ± 0.02c

0.38 ± 0.07c

0.26 ± 0.04c

2.85 ± 0.55a

0.50 ± 0.03b

Values represent the mean of six biological replicates ± standard error. Data was analyzed by two-way ANOVA. Significant interaction between site and cultivar was detected, and distinct letters indicate statistical difference by the Tukey test (P B 0.05) N.D. below the detection limit

P \ 0.05) being significantly different between samples. These results indicate an altered leaf oxidative status in the sensitive cultivar (BR-IRGA 409) when grown in Camaqua˜. 3.4 Activities of leaf antioxidative enzymes Leaf SOD activity was significantly affected by the site (F = 987.35, P \ 0.05) and by the interaction between site and cultivar (F = 49.05, P \ 0.05), with all cultivars showing higher SOD activity levels in Camaqua˜. Interestingly, the highest SOD activity was observed in the tolerant cultivar EPAGRI 109 (M = 82.23, SD = 24.47) when grown in Camaqua˜. Comparatively lower SOD activities were detected in the tolerant cultivar EPAGRI 108 (M = 13.19, SD = 3.82) and the sensitive cultivar BR-IRGA 409 (M = 20.71, SD = 3.42) when grown at the irontoxic site (Fig. 3a). Leaf APX activity was significantly affected by cultivar (F = 34.46, P \ 0.05), site (F = 73.24, P \ 0.05), and by the interaction between site and cultivar (F = 191.62, P \ 0.05). Overall, plants showed higher leaf APX activities in Camaqua˜ (Fig. 3b), with the highest values being detected in the sensitive BR-IRGA 409 (M = 50.19, SD = 27.14). Leaf CAT activity was significantly affected by the site (F = 97.69, P \ 0.05) and by the interaction between site and cultivar (F = 22.09, P \ 0.05), with the highest activities being observed in EPAGRI 109 (M = 4.01, SD = 1.22) and BR-IRGA 409 (M = 3.94, SD = 0.68) when grown at the iron-toxic site (Camaqua˜; Fig. 3c). Leaf GR activity was significantly affected by cultivar (F = 22.38, P \ 0.05), site (F = 455.17,

P \ 0.05), and by the interaction between the two factors (F = 28.33, P \ 0.05). Higher leaf GR activities were observed in plants grown in the iron-toxic site (Camaqua˜, Fig. 3d), with the highest GR activity being detected in the tolerant cultivar EPAGRI 109 (M = 51.95, SD = 7.77). The same pattern was observed for leaf DHAR activities, with significant effects being observed for cultivar (F = 14.15, P \ 0.05), site (F = 1607.63, P \ 0.05), and the interaction (F = 175.44, P \ 0.05). 3.5 SOD gene expression in leaves To further detail the high SOD activity seen in EPAGRI 109 leaves, we analyzed the expression profile of four distinct SOD genes—one MnSOD (L34038), one FeSOD (AB014056) and two Cu/ ZnSODs (D85239 and L19435). Expression of the MnSOD and the two Cu/ZnSOD tested were higher in EPAGRI 109 leaves from the iron-toxic soil than in the other two cultivars analyzed (Fig. 4a), while no difference was seen in plants cultivated in the control site (Fig. 4b).

4 Discussion Iron toxicity differently affected the three cultivars used in this field study. To characterize the response to iron toxicity, we quantified the chlorophyll and carbonyl concentrations in leaves from plants grown in two distinct sites. Both parameters have been linked to the toxic effects of iron excess in plants (Gallego et al. 1996; Fang et al. 2001; Stein et al. 2009). Our results (Fig. 1a, b) validated the use of the chosen

123

142


Fig. 3 Activities of antioxidant enzymes (SOD, APX, CAT, GR and DHR, shown in A, B, C, D and E, respectively) in fully expanded leaves from the rice cultivars EPAGRI 109, EPAGRI 108 and BR-IRGA 409 grown in an iron-toxic site (Camaqua˜RS, Brazil) or in a control site (Cachoeirinha-RS, Brazil).

Values represent the means of six replicates ± standard error. Data was analyzed by two-way ANOVA. Significant interaction between site and cultivar was detected, and distinct letters indicate statistical difference by the Tukey test (P B 0.05)

cultivars, suggesting that only the susceptible cultivar (BR-IRGA 409) suffered from photo-oxidative and oxidative damage. The impairment on light utilization caused by excessive leaf Fe concentrations was also seen under hydroponics conditions, with rice plants

showing lower photosynthetic rates and decreased chlorophyll concentrations, both due to stomatal and non-stomatal limitations (Pereira et al. 2013). Iron accumulation in leaves and roots varied widely between the two cultivars and growth sites (Fig. 2). As

123


Fig. 4 Expression profile of SOD isoforms in fully expanded leaves from three rice cultivars (EPAGRI 109, EPAGRI 108 and BR-IRGA 409) grown in two distinct sites—an iron-toxic site (A) located in Camaqua˜-RS, Brazil; and a control site (B) in Cachoeirinha-RS, Brazil. Total RNA was isolated from fully expanded leaves and used for cDNA synthesis and PCR amplification using specific primers for four rice SOD isoforms (one MnSOD, one FeSOD and two Cu/ZnSOD). The expression level of ubiquitin (OsUbq) was used as control. The experiments were repeated twice with independent RNA samples, and similar results were obtained

expected, plants grown in the iron-toxic site accumulated higher levels of iron than plants from the control site. The iron concentrations in the soil solution from both sites were previously reported to be above 280 mg L-1 in the iron-toxic site (Camaqua˜) and 29 mg L-1 in the control site, Cachoeirinha (Stein et al. 2009). The difference between the two sites clearly impacted on the iron accumulation in the plant body. Interestingly, the two selected tolerant cultivars differed in their iron accumulation in leaves, suggesting that the two cultivars make use of distinct tolerance mechanisms. The iron accumulation in roots

143

also varied between the studied cultivars, but in an opposite way. EPAGRI 108 plants (which accumulated lower iron levels in the leaf tissue) showed the highest iron accumulation in roots. Iron oxidation at the root surface has been proposed as a potential mechanism used by plants to tolerate high levels of iron in the soil solution (Becker and Asch 2005), as well as avoidance of iron uptake into the plant through other mechanisms (Silveira et al. 2007). Using the DCB solution, we were able to evaluate the iron partitioning in roots, and found that, compared to the other tolerant cultivar, EPAGRI 108 accumulated more iron in the DCB-treated roots, which remained only with the intracellular iron. Iron concentration in the extraplasmatic spaces (possibly forming the iron plaque) was not significantly higher in this cultivar than in the other two. Therefore, tolerance to iron excess in this cultivar does not seem to rely on increased oxidation of iron at the root surface resulting in higher iron levels in the apoplast. Despite the high iron concentration in the root symplast, EPAGRI 108 plants did not suffer from iron toxicity, indicating that the capacity to keep higher intracellular iron concentrations at the root zone played a significant role in the tolerant character of these plants, helping to protect the shoots from high iron concentrations. It would be interesting to investigate, in future studies, the location of root intracellular iron in EPAGRI 108 plants cultivated under high-iron conditions. It is possible that iron is stored in vacuoles and also inside the protein Ferritin, located in plastids or mitochondria. However, the proportion of iron in each cellular compartment still needs to be determined. The capacity to tolerate high levels of iron, found in some genotypes, could also depend on the induction of antioxidant defenses in the leaf tissues (Wu et al. 1998). We analyzed the activity of five antioxidant enzymes involved in the Ascorbate–Glutathione cycle in the three cultivars, grown in the two distinct sites (Fig. 3). Interestingly, EPAGRI 109, the tolerant cultivar that accumulated iron at levels similar to the ones found in the susceptible cultivar (BR-IRGA 409), showed higher activities of SOD, GR and DHAR, enzymes known to be involved in this cycle. Thereby, the tolerance to high levels of iron found in this cultivar could be directly related to the capacity to scavenge iron-mediated oxygen free radicals. Plants from the susceptible cultivar (BR-IRGA 409) showed higher concentrations of DHA and GSSG

123

144

(Table 1), indicating a clear disturbance in the GSH/AA ratio, when grown in the iron-toxic site. The GSH and AA redox state is maintained through GR, MDAR and DHAR, and they have a pivotal role in the defense against ROS-induced oxidative damage (Noctor and Foyer 1998). The higher amounts of DHA found in BRIRGA 409 leaves probably have a direct relationship with the higher APX activity observed in the same plants, since AA is used as a reducing agent by APX to catalyze the reduction of H2O2 to H2O (Shigeoka et al. 2002). While higher APX activity observed in BR-IRGA 409 plants could be directly related to the oxidative stress driven by higher levels of iron accumulated in the leaf tissue, the higher CAT activity observed in the same cultivar and in EPAGRI 109 could be related to common effects of iron excess in both cultivars, such as increased photorespiration. CAT is an enzyme known to be mainly involved in the photorespiration process (Mhamdi et al. 2012), being used by plants to deviate the energy received through the photosystems (Noctor et al. 2002). Indeed, photorespiration is considered to contribute as an important sink for electrons in rice plants exposed to high iron concentrations (Pereira et al. 2013), but this observation still needs to be confirmed on field grown plants. Besides that, plants from the cultivar EPAGRI 109 are tolerant to iron toxicity (exhibiting higher levels of chlorophyll and lower levels of carbonyl), and BR-IRGA 409 are susceptible. A plausible explanation for the high CAT activity seen in both cultivars grown only in the irontoxic site is that both cultivars deal with high levels of iron in their leaves. Apart from the higher activity of GR and DHAR, the capacity to tolerate high levels of iron in the leaf tissue (even higher levels than the susceptible cultivar—BR-IRGA 409) observed in EPAGRI 109 could rely on the remarkably higher SOD activity. SOD is responsible for the dismutation of the superoxide anion (O2-), and constitutes the first line of defense against ROS (Alscher et al. 2002). SODs have been linked to diverse stressful conditions (Bowler et al. 1994) and are found in different sub-cellular compartments. Besides different localization, the classification of SODs is dependent on the metal co-factor used by the enzyme (Alscher et al. 2002). Since EPAGRI 109 plants showed higher SOD activities, we further investigated the SOD gene expression by evaluating the specific expression of the genes encoding four SOD isoforms (one MnSOD, one

123


FeSOD and two Cu/ZnSODs, Fig. 4). EPAGRI 109 plants showed higher mRNA abundance of three out of the four SOD genes tested. Interestingly, the same result observed in the pattern of SOD expression was seen by Kim et al. (2007), with the induction of the expression of the MnSOD and the two CuSODs by exposure to a high salt concentration, indicating common responses between both stresses. The subcellular localization of the three genes was predicted using PSORT (http:// psort.ims.u-tokyo.ac.jp) and TargetP (http://www.cbs. dtu.dk/services/TargetP). The results obtained (data not shown) suggest different localizations, in mitochondria (MnSOd—L34038), plastids (Cu/ZnSOD—D85239) and cytoplasm (Cu/ZnSOD—L19435). Whether this prediction indicates that the capacity to tolerate iron excess in EPAGRI 109 relies on increased accumulation of SOD proteins in these sub-cellular compartments remains to be tested. Nonetheless, our data suggest that iron toxicity results in differential regulation of the SOD genes in the cultivars tested, and the tolerance mechanism could rely on higher SOD activity and differential SOD gene expression. Plants overexpressing SOD isoforms have been shown to be tolerant to several different types of stress, such as salt stress in rice (Tanaka et al. 1999), drought in alfalfa (McKersie et al. 1996), methyl viologen, H2O2, and heavy metals in tall fescue (Lee et al. 2007). Besides conferring tolerance to salt, the overexpression of a Mn-SOD isoform resulted in higher activity of not only the transgenic enzyme, but also of other antioxidative enzymes such as Cu/Zn-SOD and Fe-SOD in Arabidopsis (Wang et al. 2004). It is possible that the co-expression of SOD isoforms seen in our results derives from a complex regulation of different SOD isoforms, involving different sub-cellular compartments. Becker and Asch (2005) categorized the iron-toxic environments into three distinct clusters, according to differences in soil types, in soil iron content and in the rice growth stage showing most symptoms and yield losses. Possibly, the variability observed and classified into these three clusters had a major influence in the distinct tolerance mechanisms used by rice cultivars to tolerate excess levels of iron in the soil solution. Thereby, each mechanism may be more suitable for the plant to cope with specific adverse soil conditions. Our results showing that two related cultivars (EPAGRI 109 and EPAGRI 108) displayed distinct mechanisms to tolerate iron toxicity, even when grown in the same sites, indicate that not only the


conditions where iron toxicity is expressed, but also the capacity to respond to and to tolerate this stress are variable and influenced by the genotype. Experiments focusing on the genetic basis of tolerance to iron toxicity, looking at differentially expressed genes between rice cultivars with distinct tolerance levels are also being pursued by our group. We hope our studies will provide breeders with useful genetic markers of resistance to iron toxicity or new candidate genes destined to plant transformation. Acknowledgments This work was supported by CNPq (Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico— Brazil), CAPES (Coordenaçaõ de Aperfeiçoamento de Pessoal de Nı´vel Superior—Brazil) and FAPERGS (Fundaçaõ de Apoio a` Pesquisa do Estado do Rio Grande do Sul—Brazil).

References Alscher RG, Erturk N, Heath LS (2002) Role of superoxide dismutases (SODs) in controlling oxidative stress in plants. J Exp Bot 53:1331–1341 Ando T, Yoshida S, Nishiyama I (1983) Nature of oxidizing power of rice roots. Plant Soil 72:57–71 Audebert A, Sahrawat KL (2000) Mechanisms for iron toxicity tolerance in lowland rice. J Plant Nutr 23:1877–1885 Bacha RE, Ishiy T (1986) Toxicidad por hierro em arroz: metodologia para seleccionar geno´tipos resistentes en Brasil. Bol Prog CIAT 7:1–4 Becker M, Asch F (2005) Iron toxicity - conditions and management concepts. J Plant Nutr Soil Sci 168:558–573 Benckiser G, Santiago S, Neue HU, Watanabe I, Ottow JCG (1984) Effect of iron fertilization on exudation, dehydrogenase activity, iron-reducing populations and Fe?? formation in the rhizosphere of rice (Oryza sativa L.) in relation to iron toxicity. Plant Soil 79:305–316 Beyer WF, Fridovich I (1987) Assaying of superoxide dismutase activity: some large consequences of minor changes in conditions. Anal Biochem 161:559–566 Bowler C, van Camp W, van Montagu M, Inze´ D (1994) Superoxide dismutases in plants. Crit Rev Plant Sci 13:199–218 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 Cakmak I, Marschner H (1992) Magnesium deficiency and high light intensity enhance activities of superoxide dismutase, ascorbate peroxidase and glutathione reductase in bean leaves. Plant Physiol 98:1222–1227 Counce PA, Keisling TC, Mitchell AJ (2000) A Uniform, objective, and adaptive system for expressing rice development. Crop Sci 40:436–443 Fageria NK, Rabelo NA (1987) Tolerance of rice cultivars to iron toxicity. J Plant Nutr 10:653–661 Fang WC, Wang JW, Lin CC, Kao CH (2001) Iron induction of lipid peroxidation and effects on antioxidative enzyme activities in rice leaves. Plant Growth Regul 35:75–80

145 Gallego SM, Benavides MP, Tomaro ML (1996) Effect of heavy metal ion excess on sunflower leaves: evidence for involvement of oxidative stress. Plant Sci 121:151–159 Green MS, Etherington JR (1977) Oxidation of ferrous iron by rice (Oryza sativa L.) roots: a mechanism for waterlogging tolerance. J Exp Bot 28:678–690 Griffith OW (1980) Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Anal Biochem 106:207–212 Kato Y, Urano J, Maki Y, Ushimaru T (1997) Purification and characterization of dehydroascorbate reductase from rice. Plant Cell Physiol 38:173–178 Kim D, Shibato J, Agrawal GK, Fujihara S, Iwahashi H, Kim DH, Shim I, Rakwal R (2007) Gene transcription in the leaves of rice undergoing salt-induced morphological changes (Oryza sativa L.). Mol Cell 24:45–59 Klapheck S, Zimmer Z, Cosse H (1990) Scavenging of hydrogen peroxide in the endosperm of Ricinus communis by ascorbate peroxidase. Plant Cell Physiol 31:1005–1013 Lee SH, Ahsan N, Lee KW, Kim DH, Lee DG, Kwak SS, Kwon SY, Kim TH, Lee BH (2007) Simultaneous overexpression of both CuZn superoxide dismutase and ascorbate peroxidase in transgenic tall fescue plants confers increased tolerance to a wide range of abiotic stresses. J Plant Physiol 164:1626–1638 Levine RL, Garland D, Oliver CN, Amici A, Climent I, Lenz A, Ahn B, Shaltiel S, Stadtman ER (1990) Determination of carbonyl content in oxidatively modified proteins. Method Enzymol 186:464–478 McKersie BD, Bowley SR, Harjanto E, Leprince O (1996) Water-deficit tolerance and field performance of trangenic Alfafa overexpressing superoxide dismutase. Plant Physiol 111:1177–1181 Mhamdi A, Noctor G, Baker A (2012) Plant catalases: peroxisomal redox guardians. Arch Biochem Biophys 525:181–194 Miki D, Itoh R, Shimamoto K (2005) RNA silencing of single and multiple members in a gene family in rice. Plant Physiol 138:1903–1913 Noctor G, Foyer CH (1998) Ascorbate and glutathione: keeping active oxygen under control. Annu Rev Plant Physiol Mol Biol 49:249–279 Noctor G, Veljovic-Jovanovic S, Driscoll S, Novitskaya L, Foyer CH (2002) Drought and oxidative load in the leaves of C3 plants: a predominant role for photorespiration? Ann Bot 89:841–850 Okamura M (1980) An improved method for the determination of L-ascorbic acid and L-dehydroascorbic acid in blood plasma. Clin Chim Acta 103:259–268 Olaleye AO, Tabi FO, Ogunkule AO, Singh BN, Sahrawat KL (2001) Effect of toxic iron concentration on the growth of lowland rice. J Plant Nutr 24:441–457 Ottow JCG, Benckiser GH, Watanabe IJ, Santiago S (1982) A multiple nutritional soil stress as the prerequisite for iron toxicity of wetland rice (Oryza sativa L.). Trop Agr 60:102–106 Peel MC, Finlayson BL, McMahon TA (2007) Updated world map of the Ko¨ppen-Geiger climate classification. Hydrol Earth Syst Sci 11:1633–1644 Pereira EG, Oliva MA, Rosado-Souza L, Mendes GC, Colares DS, Stopato CH, Almeida AM (2013) Iron excess affects

123

146 rice photosynthesis through stomatal and non-stomatal limitations. Plant Sci 201:81–92 Ponnamperuma FN (1972) The chemistry of submerged soils. Adv Agron 24:29–96 Quinet M, Vromman D, Clippe A, Bertin P, Lequeux H, Dufey I, Lutts S, Lefe`vre I (2012) Combined transcriptomic and physiological approaches reveal strong differences between short- and long-term response of rice (Oryza sativa) to iron toxicity. Plant Cell Environ 35:1837–1859 R Development Core Team (2011) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. http://www.R-project.org/ Ricachenevsky FK, Sperotto RA, Menguer PK, Fett JP (2010) Identification of Fe-excess-induced genes in rice shoots reveals a WRKY transcription factor responsive to Fe, drought and senescence. Mol Biol Rep 37:3735–3745 Ross CW (1974) Plant physiology laboratory manual. Wadsworth Publishing Company, Belmont Sahrawat KL (2000) Elemental composition of rice plants as affected by iron toxicity under field conditions. Comm Soil Sci Plant Anal 31:2819–2827 Sahrawat KL (2004) Iron toxicity in wetland rice and the role of other nutrients. J Plant Nutr 27:1471–1504 Sahrawat KL, Mulbah CK, Diatta K, Delaune RD, Patrick WH, Singh BN, Jones MP (1996) The role of tolerant genotypes and plant nutrients in the management of iron toxicity in lowland rice. J Agr Sci 126:143–149 Sgherri CLM, Loggini B, Puliga S, Navari-Izzo F (1994) Antioxidant system in Sporobolus stapfianus: changes in response to desiccation and rehydration. Phytochem 33:561–565

123

Theor. Exp. Plant Physiol. (2014) 26:135–146 Shigeoka S, Ishikawa T, Tamoi M, Miyagawa Y, Takeda T, Yabuta Y, Yoshimura K (2002) Regulation and function of ascorbate peroxidase isoenzymes. J Exp Bot 53:1305–1319 Silveira VC, Oliveira AP, Sperotto RA, Espindola LS, Amaral L, Dias JF, Cunha JB, Fett JP (2007) Influence of iron on mineral status of two rice (Oryza sativa L.) cultivars. Braz J Plant Physiol 12:127–139 Stein RJ, Duarte GL, Spohr MG, Lopes SIG, Fett JP (2009) Distinct physiological responses of two rice cultivars subjected to iron toxicity in the field. Ann Appl Biol 154:269–277 Tanaka Y, Hibino T, Hayashi Y, Tanaka A, Kishitani S, Takabe T, Yokota S, Takabe T (1999) Salt tolerance of transgenic rice overexpressing yeast mitochondrial Mn-SOD in chloroplasts. Plant Sci 148:131–138 Taylor G, Crowder AA (1983) Use of the DCB technique for extraction of hydrous iron oxides from roots of wetland plants. Am J Bot 70:1254–1257 Wang Y, Ying Y, Chen J, Wang X (2004) Transgenic Arabidopsis overexpressing Mn-SOD enhanced salt-tolerance. Plant Sci 167:671–677 Winslow MD, Yamauchi M, Alluri K, Masajo TM (1989) Reducing iron toxicity in rice with resistant genotype and ridge planting. Agron J 81:458–460 Wu P, Hu B, Liao CY, Zhu JM, Wu YR, Senadhira D, Paterson AH (1998) Characterization of tissue tolerance to iron by molecular markers in different lines of rice. Plant Soil 203:217–226 Yamauchi M, Peng XX (1995) Iron toxicity and stress-induced ethylene production in rice leaves. Plant Soil 173:21–28