ORIGINAL ARTICLE

Differential antioxidative responses to dehydration-inducedoxidative stress in core set of foxtail millet cultivars[Setaria italica (L.)]

Charu Lata & Sarita Jha & Vivek Dixit &Nese Sreenivasulu & Manoj Prasad

Received: 15 October 2010 /Accepted: 15 December 2010 /Published online: 1 January 2011# Springer-Verlag 2010

Abstract Foxtail millet (Setaria italica L.) known as arelatively drought-tolerant crop across the world is grownin arid and semi-arid regions. To the best of our knowledge,no systematic study on drought tolerance screening offoxtail millet germplasm being a drought-tolerant crop hasbeen reported so far. To explore genetic diversity ofdrought-induced oxidative stress tolerance in foxtail millet,we employed lipid peroxidation measure to assess mem-brane integrity under stress as biochemical marker to screen107 cultivars and classified the genotypes as highlytolerant, tolerant, sensitive, and highly sensitive. From thiscomprehensive screening, four cultivars showing differen-tial response to dehydration tolerance were selected tounderstand the physiological and biochemical basis oftolerance mechanisms. The dehydration-tolerant cultivars(IC-403579 and Prasad) showed considerably lower levelsof lipid peroxidation and electrolyte leakage as compared withdehydration-sensitive cultivars (IC-480117 and Lepakshi),indicating better cell membrane integrity in tolerant cultivars.

Correspondingly, tolerant genotypes maintained higher activityof catalase (EC 1.11.1.6), ascorbate peroxidase (APX; EC1.11.1.11), and glutathione reductase (GR; EC 1.6.4.2) acrossdifferent time-course period of polyethylene glycol (PEG)treatments in comparison to sensitive ones. The abovebiochemical results were further validated through quantitativereal-time PCR analysis of APX and GR, whose transcriptswere substantially induced by PEG treatments in tolerantcultivars. These results suggest that tolerant cultivars possesswider array of antioxidant machinery with efficient ascorbate–glutathione pathway to cope with drought-induced oxidativestress.

Keywords Setaria italica . Dehydration . Relative watercontent (RWC) . Lipid peroxidation (LP) . Electrolyticleakage (EL) . Reactive oxygen species (ROS) .

Antioxidative enzymes . Quantitative real-time PCR(qRT-PCR)

Introduction

Abiotic stresses are major limiting factors that affect cropgrowth and productivity. Drought and salinity affect morethan 10% of arable land, which results in more than 50%decline in the average yields of major crops worldwide(Boyer 1982; Bray 1997). Dehydration is one of the mostimportant constraints for crop production, and its responseand adaptation in plants is generally a complex phenome-non. It involves an array of physiological and biochemicalmodifications in plants including changes in relative watercontent (RWC), electrolytic leakage (EL), generation ofreactive oxygen species (ROS) resulting in membranedamage, and inactivation of enzymes thus affecting cellviability (Bartels and Sunkar 2005). The relative water

Handling Editor: Peter Nick

The authors Charu Lata and Sarita Jha contributed equally to thisstudy.

Electronic supplementary material The online version of this article(doi:10.1007/s00709-010-0257-y) contains supplementary material,which is available to authorized users.

C. Lata : S. Jha :V. Dixit :M. Prasad (*)National Institute of Plant Genome Research (NIPGR),Aruna Asaf Ali Marg, JNU Campus,New Delhi 110067, Indiae-mail: manoj_prasad@nipgr.res.in

N. SreenivasuluLeibniz Institute of Plant Genetics and Crop Plant Research (IPK),Corrensstrasse 03,06466 Gatersleben, Germany

Protoplasma (2011) 248:817–828DOI 10.1007/s00709-010-0257-y

content determination has been widely accepted as a repro-ducible and meaningful index of plant water status (Smart andBingham 1974). Membrane lipids are primary targets of ROSand oxidizes fatty acid reaction products like conjugateddienes, hydroperoxides, and malondialdehyde (MDA), whichare usually used as markers of lipid peroxidation (LP) andalso as important indicators of stress susceptibility in plants(Varghese and Naithani 2008). Electrolytic leakage as anindicator of membrane damage has also been widely used tostudy oxidative stress tolerance in various crops (Guo et al.2006; Bhushan et al. 2007; Silva et al. 2010).

ROS are generated as natural products of plant cellularand aerobic metabolisms; however, its overproductionimparts a negative oxidative stress on cellular components,structures, and metabolism (Wang et al. 2009; Silva et al.2010). During drought-induced oxidative stress, ROS aregenerated in different organelles such as chloroplast,mitochondria, and peroxisomes due to altered process ofenergy dissipation during photosynthesis, respiration, andphotorespiration (Noctor and Foyer 1998). The generationof ROS such as singlet oxygen (O2

1), superoxide (O2-),

hydrogen peroxide (H2O2), and hydroxyl radicals (HO) indifferent compartments pose a threat to cells by unrestrictedoxidation of various cellular molecules such as nucleicacids, protein, and lipids, resulting in membrane damage oforganelles and eventual release of ROS into cytoplasm(Leprince et al. 2000). In order to cope with such damagingeffects, higher plants have evolved complex antioxidantsystems as well as low molecular weight quenchers likeascorbic acid and glutathione which are principally consti-tutive and vary in plants at cellular and subcellular level(Alscher et al. 1997; Foyer et al. 1997). O2

- is the primarysource of ROS molecules, which is converted into H2O2 bythe action of superoxide dismutase. The generated H2O2 isquenched by classical peroxidases, ascorbate peroxidases(APX), glutathione peroxidases (GPX), and catalase (CAT;Halliwell and Gutteridge 1989; Noctor and Foyer 1998).Quenching ROS in subcellular compartments is prettyimportant such as detoxifying generated H2O2 in glyoxysomesand peroxisomes through the action of CAT and in othersubcellular compartments such as chloroplast, mitochondria,and cytoplasm H2O2 is converted to H2O by APX and GPX.Thus, CAT eliminates hydrogen peroxide by breaking it downdirectly to form water and oxygen (Wang et al. 2009). APXand glutathione reductase (GR) are respectively the first andlast enzymes in the ascorbate–glutathione cycle responsiblefor H2O2 detoxification in green leaves (Foyer and Harbinson1994; Asada 1999). APX is considered to be a keyantioxidant enzyme in plants while GR plays a central rolein maintaining the reduced glutathione (GSH) pool duringstress (Orvar and Ellis 1997; Pastori et al. 1997).

Foxtail millet [Setaria italica (L.) P. Beauv.] is one of theworld's oldest crops (cultivated for hay, pasture, and food

grain), traditionally grown in arid and semi-arid regions ofIndia, Asia, North Africa, and South and North Americaknown for its drought tolerance. It is a diploid (2n=2×=18),self-pollinating, C4 panicoid crop species with a smallgenome of ∼490 Mb having low amount of repetitive DNAcontent (30%) and highly conserved genome structurerelative to the ancestral grass lineage making it a suitablespecies for genetic and molecular studies (Jayaraman et al.2008). It is also a close relative to several important biofuelcrops such as switchgrass, napiergrass, and pearl millet and,hence, would serve as an excellent surrogate genome toassist future studies of the related biofuel species. It isimportant to study dehydration stress responses in arelatively dehydration-tolerant crop to explore stress toler-ance mechanisms, which is expected to be different fromother cultivated crops grown normally in the irrigated lands.Though there have been a few studies on differentialresponses of antioxidant compounds in foxtail millet insalinity stress (Sreenivasulu et al. 1999, 2000), no system-atic study has been carried out to study dehydrationtolerance in foxtail millet.

Hence, we screened 107 foxtail millet cultivars for theirdehydration tolerance on the basis of LP to identify geneticdiversity. After primary screening, cv. IC-403579 and cv.Prasad were identified as highly dehydration-tolerant, whilecv. IC-480117 and cv. Lepakshi were found to be highlydehydration-sensitive among all the cultivars. We haveexamined the differences in the redox status and antioxidantdefense machinery (CAT, APX, and GR) between tolerantand sensitive cultivars. Effective upregulation of antioxidantmachinery in drought-tolerant cultivars to polyethyleneglycol (PEG) exposure positively correlates with bettermaintenance of RWC and higher membrane stability (withlower EL and LP values) under stress.

Materials and methods

Plant material, growth, and stress conditions

A primary screening experiment was carried out with 107foxtail millet accessions obtained from National Bureau ofPlant Genetic Resources (NBPGR), Hyderabad, India(Supplementary Table S1). Seeds were germinated, sownin composite soil (peat compost to vermiculite, 3:1), andgrown in growth chamber for 10 days after germination inpots at 28±1°C day/23±1°C night/70±5% relative humiditywith a photoperiod of 14 h and a photosynthetic photon fluxdensity of 500 μmol m−2 s-1. The plants were watered dailywith one-third-strength Hoagland solution. The other set ofplants were subjected to dehydration stress in the Hoaglandnutrient solution containing 20% polyethylene glycol(PEG-6000; Zhang et al. 2007). Stressed and control

818 C. Lata et al.

seedlings were sampled at the same time to avoid anydiurnal variation. Whole seedlings were collected and usedas such. For screening, all 107 foxtail millet accessions weregiven 20% polyethylene glycol (PEG-6000) treatments for24 h. From the selected subset of genotypes, dehydrationstress was imposed for 1, 3, 6, 12, 24, 48, and 72 h of 20%polyethylene glycol (PEG-6000) treatments. All experimentaldata are the means of at least three independent experiments,and for each experiment, ∼100 mg seedling samples werecollected by random sampling.

Physiological parameters

Leaf RWC was measured in control and stressed seedlingsas reported earlier (Lata et al. 2010). Fully expanded leaveswere excised and fresh weight was recorded immediately;then, leaves were soaked for 4 h in distilled water at roomtemperature under a constant light, and the turgid weightwas recorded. Total dry weight was recorded after dryingfor 24 h at 80°C in a hot-air oven. RWC was calculatedaccording to the formula given by Barrs and Weatherley(1962): RWC %ð Þ ¼ FW� DWð Þ= TW� DWð Þ½ � � 100

The lipid peroxidation level in plant tissues wasdetermined by measuring the MDA content via 2-thiobarbituric acid (TBA) reaction (Hodgson and Raison1991). Seedlings (100 mg) were homogenized in 1 ml of10 mM sodium phosphate buffer (pH 7.4) and centrifugedat 4,000×g for 5 min at room temperature. Two hundredmicroliters of the supernatant was added to a reactionmixture containing 100 μl of 8.1% (w/v) sodium dodecylsulfate (SDS), 750 μl of 20% (w/v) acetic acid (pH 3.5),750 μl of 0.8% (w/v) aqueous TBA, and 200 μl of Milli-Qwater. An identical reaction mixture in which 200 μl ofsupernatant that was substituted by an equal volume ofbuffer was simultaneously set up as blank. Both reactionmixtures were then incubated at 98°C for 1 h. After coolingto room temperature, the mixtures were centrifuged for5 min. Absorbance at 535 nm was measured and correctedfor nonspecific absorbance at 600 nm. The level of lipidperoxidation was expressed as micromoles of MDA formedderived from the difference in absorbance at 535 and600 nm using an extinction coefficient of 156 mM−1 cm−1.

EL was assessed according to Dionisio-Sese and Tobita(1998). The root tissues (50 mg) were placed in 20 ml ofdeionized water for 2 h at room temperature, and the initialconductivity (E1) was recorded using a microprocessor-based conductivity meter (Model 1601, ESICO). The tubescontaining root tissues were then placed in boiling water for0.5 h to release all the electrolytes into the solution, cooledto room temperature, and the final conductivity (E2) wasrecorded. The electrolytic leakage was calculated as theratio of conductivity before boiling to that of after boiling:E1/E2×100.

Antioxidative enzyme assay

Seedlings (100 mg; 2–3 plants) were ground in mortar andpestle under chilled condition in homogenization bufferspecific for each enzyme, and the homogenate wascentrifuged at 12,000×g for 10 min at 4°C. The supernatantwas desalted with Sephadex G-25 column equilibrated withenzyme-specific buffer. Protein estimation was carried outusing bovine serum albumin as standard (Bradford 1976).

For the estimation of catalase (EC 1.11.1.6), 100 mg oftissues was homogenized in 1 ml of 50 mM phosphate buffer(pH 7.0) and 1 mM dithiothreitol (DTT). CAT activity wasassayed spectrophometrically bymonitoring the production ofdioxygen from H2O2 (33.5 mM). The reaction was initiatedby adding enzyme extracts, and the decrease in absorbanceof H2O2 (extinction coefficient, 39.4 mM−1 cm−1) wasrecorded within 1 min at 240 nm (Abei 1984). One unit ofCAT activity was defined as the amount of enzyme requiredto oxidize 1 μmol of H2O2 per minute.

For APX (EC 1.11.1.11) activity determination, 100 mgof tissues was extracted in 1 ml extraction mediumcontaining 100 mM phosphate buffer, pH 7.0; 0.1 mMethylenediaminetetraacetic acid (EDTA); 1.0 mM ascor-bate; and 1 mM DTT. APX activity was determined bymonitoring the rate of hydrogen peroxide dependentoxidation of ascorbic acid, as an electron donor, in areaction mixture containing 50 mM phosphate buffer(pH 7.0), 0.6 mM ascorbate, and enzyme extract, in a totalvolume of 1 ml (Chen and Asada 1989). The reaction wasstarted by adding 10 μl of 10% (v/v) H2O2. The oxidationrate of ascorbate was estimated by monitoring the decreasein absorbance at 290 nm (ε=2.8 mM−1 cm−1) for 1 min.One unit of enzyme activity is defined as 1 μ mol ofascorbate oxidized min.

−1

For GR (EC 1.6.4.2) estimation, 100 mg of tissues washomogenized in 1 ml extraction buffer containing 100 mMphosphate, pH 7.5 and 0.5 mM EDTA. GR was assayed byfollowing the increase in absorbance at 412 nm (ε=6.22 mM−1 cm−1) when 5,5′-dithiobis (2-nitrobenzoic acid)(DTNB) was reduced by glutathione (GSH) to form TNB(Smith et al. 1988). The reaction mixture contained 100 mMphosphate buffer, pH 7.5; 0.5 mM EDTA; 0.75 mM DTNB;0.1 mM NADPH; enzyme extract, and 1.0 mM oxidizedglutathione (GSSG) which was added last to initiate thereaction. One unit of enzyme activity defined as 1 μ mol ofGS-TNB formed per minute due to reduction of DTNB.

Partial cDNA cloning and real-time quantitative PCRanalysis of APX and GR from foxtail millet

Total RNA was isolated from 10-day-old foxtail seedlingsexposed to short-term PEG-induced dehydration stress (0, 1,3, 6, 12, 24, 48, and 72 h) using modified hot phenol method

Differential antioxidative responses 819

(Lata et al. 2010). DNA contamination was removed fromthe RNA samples using DNaseI RNase-free (50 U/μl,Fermentas, USA). RNA concentrations were measured at260 nm by a spectrophotometer (Nanodrop, USA). Thequality and purity of the preparations were determined bymeasuring the OD260/OD280 nm absorption ratio (1.9–2.0),and the integrity of the preparations was ascertained byelectrophoresis in a 1.2% agarose gel containing formal-dehyde. About 2 μg of total RNA was used to synthesizefirst strand cDNA primed with OligodT in a 20-μlreaction mix using Protoscript M-MuLV reverse tran-scriptase (New England Biolabs, USA) following manu-facturer's instructions. For PCR amplification of thedesired cDNA fragments, degenerate primers were used(Table 1). The PCR cycling conditions were 94°C for3 min, 94°C for 30 s, 55°C for 30 s, and 72°C for 1 minfor 35 cycles and 72°C for 10 min. The desired sizedbands were excised and eluted from gel and cloned intopGEM-T Easy vector (Promega, USA). The plasmidsisolated were sequenced with ABI Sequencer, VersionNo.3770 using M13 forward and reverse primers.Quantitative real-time PCR (qRT-PCR) was performedusing Power SYBER-Green dye (Applied Biosystems) onstep 1 real-time PCR system of Applied Biosysytems intriplicates. A constitutive gene actin of rice was used asendogenous control. The amount of transcript accumu-lated for each gene normalized to the internal controlactin was analyzed using 2-ΔΔCt method (Livak andSchmittgen 2001). The PCR efficiency is dependent on theassay, the master mix performance, and sample quality, andwas calculated as: Efficiency=10 (−1/slope)−1 (3.6≥slope≥3.1)by the software itself (Applied Biosystems). The primersused for real-time PCR analysis were designed from theselected sequences of the respective clones using the PrimerExpress Version 3.0 (Table 1). The PCR cycling conditionswere: initial denaturation at 95°C for 10 min, 95°C for 15 s,and 60°C for 1 min for 40 cycles and then 95°C for 15 s, 60°Cfor 1 min, and 95°C for 15 s.

Statistical analysis

All experimental data are the means of at least threeindependent experiments, and the results are presented asthe mean values±SD. The significance of differencesbetween mean values of control and each dehydrationstressed samples were statistically performed using one-way analysis of variance, and comparison among meanswas carried out using Tukey–Kramer multiple comparisonstest. The differences in the effects of PEG-induceddehydration stress on various parameters in 107 foxtailmillet accessions were considered statistically significant at“a” P<0.05, “b” P<0.01, “c” P<0.001.

Results and discussion

Screening of foxtail millet cultivars for dehydrationtolerance

In this study, 107 Indian foxtail millet (S. italica L.) cultivarswere studied for their dehydration tolerance. Primaryscreening on the basis of lipid peroxidation was carried outon cultivars subjected to 24 h of dehydration stress ascompared with respective controls (Fig. 1, SupplementaryTable ESM1, Supplementary Figure ESM1). Based on lipidperoxidation, the 107 cultivars were grouped into highlytolerant, tolerant, sensitive, and highly sensitive to PEG-induced dehydration stress. Five cultivars namely those withAccession numbers IC-403579, 403438, 403476, 403480,and Prasad that showed >30% decrease in MDA levels in24 h of dehydration stress samples in comparison to controlwere grouped as highly tolerant, whereas 15 cultivarsmaintain MDA levels <30% under stress were grouped astolerant; 49 cultivars depicting elevated MDA levels (>20%)under stress as sensitive and approximately 38 cultivarsshowed >100% increase in MDA content under stress weregrouped as highly sensitive (Supplementary Table S1).

Sl no. Primer ID Forward sequence (5′-3′) Amplicon length, bpReverse sequence (5′-3′)

1 SiAPX_F CCTCATCGCCGAGAAGARCTG 336SiAPX_R GVAGCTGAAGAAGGCCYTCCTT

2 SiGR_F GACTACGACCTCTTCACCAT 546SiGR_R TCCCACTATTGCAATYTTCTCAG

3 SiAPX RT_F GAGCCCGTGAAGGAGGAGTT 60SiAPX RT_R TCCCGCAAGCTGGTACAGAT

4 SiGR RT_F TGTGTGCTTCGTGGATGTGTT 120SiGR RT_R CCAGTCATGCTTCGGATCAGT

5 Actin RT_F CTGACGCCGAGGATATCCA 58Actin RT_R GCCTTGACCATACCAGTTCCA

Table 1 Primer sequencesused for cDNA cloning andqRT-PCR analysis of APXand GR

820 C. Lata et al.

Based on primary screening, we selected five relativelyhighly tolerant (IC-403579, 403438, 403476, 403480, andPrasad) and five relatively highly sensitive cultivars (IC-480117, 403521, 480554, 404178, and Lepakshi) forassessing cellular defense mechanisms against drought-induced oxidative stress. From these ten contrastingcultivars, total catalase activity was measured at 1, 6, 12,and 24 h of dehydration stress in comparison to theirrespective controls (Fig. 2). All the selected tolerantcultivars as a whole showed increase in total catalaseactivity up to 12 h and a slight decline at 24 h. The catalaseactivity is much more pronounced in two cultivars cv. IC-403579 and cv. Prasad. In contrast, all the sensitivecultivars maintained low absolute levels of catalase activitywith slight initial increase in catalase activity up to 6 h andsubsequent decline at 12 and 24 h. Among them, cv. IC-480117 and cv. Lepakshi showed drastic decline in CATactivity at 12 and 24 h of dehydration stress. CAT is one ofthe ubiquitous enzymes in aerobic organism and plays akey role in cellular defense mechanisms against hydrogenperoxide (H2O2; Azevedo-Neto et al. 2006). Hence, noticeddifferences in the efficiency of maintaining higher catalaseactivity could reflect the better adaptability of genotypes todrought-induced oxidative stress. Considering the lipid

peroxidation rate (primary screening) and assessing theefficiency of CAT activity (secondary screening), werefined the list by identifying the most obvious contrastinglines, IC-403579 and cv. Prasad as highly drought-tolerant,and cv. IC-480117 and cv. Lepakshi as highly drought-sensitive for detailed biochemical analysis to study thetemporal kinetics of dehydration responses.

Estimation of RWC and membrane integrity by assessingLP and EL

RWC represents a useful indicator of the state of waterbalance of a plant (González and González-vilar 2001). Thechoice of RWC as the best representation of plant waterstatus for assessing genetic differences in dehydrationtolerance is supported by genetic association betweenRWC and plant biomass under dehydration (Blum 1996;Bhushan et al. 2007). Leaf RWC was determined in fourfoxtail millet cultivars exposed to different durations ofPEG-induced dehydration stress. Stress treatments signifi-cantly decreased the RWC in all four cultivars (Fig. 3);however both tolerant cultivars, cv. IC-403579 and cv.Prasad, showed the capacity to maintain relatively highRWC, whereas cv. IC-480117 and cv. Lepakshi showed

0.0

20.0

40.0

60.0

80.0

100.0

120.0

IC-4

0357

9

Prasad

IC-4

0343

8

IC-4

0347

6

IC-4

0348

0

IC-5

1964

2

IC-4

3872

5-1

IC-3

4021

6

IC-4

7961

5

IC-2

7659

8

IC-4

7958

2

IC-4

8041

5

IC-4

0383

5

IC-4

8037

5

IC-4

8018

5

IC-4

0352

1

IC-4

8055

4

IC-4

0417

8

Lepak

shi

IC-4

8011

7

Cultivars

C 24 hr

HT T S HS

MD

A-µ

mol

e(gF

W)-

1

Fig. 1 Levels of lipid peroxida-tion expressed in terms ofMDA concentration in 20 foxtailmillet cultivars exposed to 0and 24 h dehydration treatment.Five cultivars from each groupnamely, highly tolerant (HT),tolerant (T), sensitive (S), andhighly sensitive (HS) have beenrepresented in the figure. Datarepresent the means±SD of threeindependent experiments (n=3)

020406080

100120140160180200

IC-4

0343

8

IC-4

0347

6

IC-4

0348

0

IC-4

0357

9

Prasa

d

IC-4

0417

8

IC-4

0352

1

IC-4

8055

4

IC-4

8011

7

Lepak

shi

Cultivars

CA

T U

nits

(m

g pr

otei

n)-1

C 1 hr 6 hr 12 hr 24 hrFig. 2 Catalase activity in tencultivars exposed to 0, 1, 6, 12,and 24 h dehydration treatment.One unit of enzyme activitydefined as 1 μmol H2O2 oxi-dized per minute. Data representthe means±SD of three inde-pendent experiments (n=3)

Differential antioxidative responses 821

maximum decline in leaf RWC at all stress durations (up to90% in both cultivars at 72 h, P<0.001). Several water-stressed crop plants like Cenchrus sp., chickpea, groundnut,wheat, sorghum, and maize have also shown genotypicvariations in RWC as evident from several previous studies(Nagy et al. 1995; El Hafid et al. 1998; Madhusudan et al.2002; Bhushan et al. 2007; Chandra and Dubey 2010).

Total amount of thiobarbituric acid-reacting substancesreflect the level of lipid peroxidation resulting from oxidativestress (Guo et al. 2006; Fazeli et al. 2007). The increasedaccumulation of lipid peroxides upon oxidative stress indi-cates enhanced production of toxic ROS. ROS are responsiblefor stress-induced peroxidation of membrane lipids and areoften used as an indicator of increased oxidative damage(Smirnoff 1995). The level of MDA increased significantly inthe sensitive cultivars at all stress durations with respect tocontrol (P<0.001). The level of lipid peroxidation in sensitive

cv. IC-480117was almost 289% at 24 h which declined to 202and 118% at 48 and 72 h, respectively. In tolerant cvs. IC-403579 and Prasad, the absolute MDA levels were quite lowin comparison to sensitive cultivars. throughout the dehydra-tion period exhibiting maximum membrane integrity understress as evident from low levels of MDA accumulation in latedurations of stress (Fig. 4). Sreenivasulu et al. (1999) alsoobserved an increase in MDA content in sensitive foxtailmillet cv. Lepakshi as compared with tolerant cv. Prasad inresponse to 150 mM salinity stress in 5-day-old seedlings.Lipid peroxidation was demonstrated as a function ofmembrane integrity in chickpea plants exposed to dehydra-tion stress for 7 days, and, together with electrolytic leakage,has been shown to be a direct indicator of dehydration stresstolerance (Bhushan et al. 2007).

In crop plants, cell membrane integrity has been widelyused as a criterion to differentiate stress-tolerant and

cc

c

c

cc

c

cc

c

c

cc

c

c

c

c

c

c

cc

c

c

c

c

cc

c

Cultivars

Fig. 3 Percentage relative watercontent estimates in four foxtailmillet cultivars exposed to 0, 1,3, 6, 12, 24, 48, and 72 hdehydration treatment. Datarepresent the means±SD of threeindependent experiments (n=3);c, P<0.001

CultivarsCultivars

Fig. 4 Levels of lipid peroxida-tion expressed in terms of MDAconcentration in four foxtailmillet cultivars exposed to 0, 1,3, 6, 12, 24, 48 and 72 hdehydration treatment. Datarepresent the means ± SD ofthree independent experiments(n=3); a, P<0.05; b, P<0.01; c,P<0.001

822 C. Lata et al.

sensitive cultivars. Also, in some cases, capacity to avoid orrepair membrane damage has been directly correlated withabiotic stress tolerance, e.g., dehydration and salinity(Sreenivasulu et al. 2000; Masood et al. 2006; Bhushan etal. 2007). Significant and time-dependent increase in mem-brane permeability was observed in all the four cultivars.However, in cv. IC-403579 and cv. Prasad, membrane wasrelatively less permeable except at higher durations ofdehydration stress treatment (Fig. 5). Among all the cultivars,maximum ion leakage was found in cv. Lepakshi followed bycv. IC-480117 at all durations of dehydration stress. Lowlevels of electrolyte leakage suggest better maintenance ofcell membrane integrity in the tolerant cultivars. The valueswere significantly higher than the control (P<0.001) at alldurations in all the cultivars. The electrolytic leakage was alsofound to be maintained at constant low level in chickpeatolerant cultivar JG-62 during 96 h dehydration stresswhereas sensitive cv. ICCV2 showed sharp rise in electrolyticleakage accompanied by maximal lipid peroxidation(Bhushan et al. 2007). A similar result was also observed intwo drought-tolerant cultivars of rice namely Xiangzhongxianno. 2 and IR-50 in comparison to sensitive Xiangnou no.1and Zimanuo (Guo et al. 2006). The above observation hasalso been supported by several previous studies on variouscrop plant species (Sreenivasulu et al. 2000; Masood et al.2006; Kumutha et al. 2009).

Antioxidative enzymes assay

Abiotic stresses like drought and salinity cause oxidativestress via the production and accumulation of ROS. H2O2 isa ROS more toxic to a cell's internal milieu than O2

- , which

is needed to be adequately scavenged by low molecularweight antioxidants and antioxidant enzymes includingCAT, APX, and GR (Wang et al. 2009). The levels ofantioxidant enzymes are reported to be higher in tolerantthan sensitive species under various environmental stresses(Bor et al. 2003; Demiral and Turkan 2005; Guo et al.2006; Jogeshwar et al. 2006). Accordingly, we also gothigher CAT activity in tolerant cvs. IC-403579 and Prasadthan the sensitive ones at all durations of PEG-induceddehydration stress that suggests tolerant cultivars possessbetter ROS scavenging ability (Fig. 6a).

Plants have evolved a unique peroxidase, utilizing ascor-bate as electron donor, namely, APX. Unlike classical plantperoxidases, APX has a remarkably high preference forascorbate as electron donor and, in combination with GR(AsA-GSH cycle), functions to prevent accumulation of toxiclevels of H2O2 in photosynthetic organisms (Asada 1997,1999). All the foxtail millet cultivars exhibited differentialAPX activity in response to dehydration stress. IncreasedAPX activity was observed in tolerant cultivars comparedwith sensitive ones. Significant induction (P<0.001) in APXactivity was observed in the tolerant cultivars up to 12 h oftreatment which, however, declined at later durations oftreatment. The APX activity was slightly increased in thesensitive cultivars at early stages of stress with successivedecline at later durations (Fig. 6b). Increased APX activitywas also reported in tolerant cultivars of other crop species(Shigeoka et al. 2002; Demiral and Turkan 2005; Kumuthaet al. 2009; Garg and Noor 2009).

GR is one of the key enzymes that help in maintainingthe cellular redox status of the plants by convertingoxidized glutathione (GSSG) to reduced form (GSH) using

c

bb

c

cc

c

c c

c

c

c

cc

c

cc

c

c

c

c

cc

cc

c

c

c

Cultivars

Fig. 5 Percentage electrolyticleakage estimates in four culti-vars of foxtail millet exposed to0, 1, 3, 6, 12, 24, 48, and 72 hdehydration treatment. Datarepresent the means±SD of threeindependent experiments (n=3);b, P<0.01; c, P<0.001

Differential antioxidative responses 823

c

a

Cultivars

c

a

c

ab

Cultivars

a

b

a

a

aa

c

b

a

a

a

a

aa

a

a

b

b

c

b

b

b

a

aa

a

a

aa

Cultivars

a

b

c

Fig. 6 Changes in antioxidativeenzyme activity in stress-tolerant and sensitive foxtailmillet cultivars exposed to 20%PEG 6000 for 0, 1, 3, 6, 12, 24,48, and 72 h. a CAT activity.One unit of enzyme activitydefined as 1 μmol H2O2

oxidized min−1; b APX activity.One unit of enzyme activitydefined as 1 μ mol of ascorbateoxidized min−1; and c GRactivity. One unit of enzymeactivity defined as 1 μ mol ofGS-TNB formed per minute dueto reduction of DTNB. Datarepresent the means±SD of threeindependent experiments(n=3); a, P<0.05; b, P<0.01;c, P<0.001

824 C. Lata et al.

NADPH as reducing power. GSH is also consumed anddegraded in order to protect cellular membranes from lipidperoxidation (Wise and Naylor 1987). A significant andtime-dependent increase in GR activity has been found inall the cultivars exposed up to 12 h of dehydration stress ascompared with respective controls, and then, the activitydeclined at later stages (P<0.01, P<0.001). The maximumincrease in GR activity was observed in cv. IC-403579, andthe minimum was in cv. Lepakshi at 12 h of dehydrationstress with respect to control. Though GR activitydecreased in all the cultivars at 72 h of stress, withmaximum decline in IC-480117 and minimum in cv.Prasad, the values were largely significant with respect tocontrol (P<0.05, P<0.01; Fig. 6c). Similar observationswere reported in wheat, rice, sorghum, and pigeon pea(Loggini et al. 1999; Jogeshwar et al. 2006; Guo et al.2006; Kumutha et al. 2009) under various abiotic stresses.Adaptation to drought has been reported to involvedifferent protective mechanisms including the capacity tomaintain high levels of antioxidants (ascorbate and GSH)and to regenerate them through the induction of GR(Loggini et al. 1999). Activation of ascorbate–glutathionecycle has earlier been suggested as essential in stressedplants to combat oxidative damage (Alscher et al. 1997).

Partial cDNA cloning and expression analysis of APX and GR

To validate our physiological and biochemical analysis,qRT-PCR analysis was performed to evaluate the mRNA

accumulation of APX and GR in all the four cultivars. Toclone APX cDNA fragment, cytosolic APX cDNA sequen-ces from rice (Os03g0285700), maize (LOC100283822),and sorghum (XM_002468008.1) were aligned usingmultiple sequence alignment program CLUSTAL W(1.81). The regions which showed maximum homologyand minimum degeneracy were used to design forwardand reverse primers of product size 336 bp. Similarly, GRcDNA fragment was amplified using degenerate primersdesigned by aligning GR cDNA sequences of rice(Os03g0163300), Hordeum (AB277096.1), and sorghum(XM_002468412.1) with a product size of 546 bp. TheAPX and GR sequences have been submitted to GenBankwith accession numbers HQ327492 and HQ327493,respectively.

The amplified APX and GR cDNA fragments were usedto design qRT-PCR primers (Table 1). IC-403579 showednearly twofold APX gene expression at 12 h followed bycv. Prasad that showed maximum APX expression up to 6 hof stress (Fig. 7a; Supplementary Table ESM2), whereasIC-480117 and cv. Lepakshi showed decreased mRNAaccumulation at all durations of dehydration stress incomparison to tolerant cultivars. Increase in transcriptlevels of APX is positively correlated to increased APXactivity. Similar study has also been done in pigeon peawhere partial cDNA of APX gene was cloned and RT-PCRwas carried out. APX gene expression was higher in thetolerant in comparison to the sensitive Cajanus cajangenotype (Sairam et al. 2009).

0

0.5

1

1.5

2

2.5

IC-403579 Prasad IC-480117 Lepakshi

Cultivars

AP

X R

elat

ive

fold

exp

ress

ion C 1 hr 3 hr 6 hr 12 hr 24 hr 48 hr 72 hr

0

0.5

1

1.5

2

2.5

IC-403579 Prasad IC-480117 Lepakshi

Cultivars

GR

Rel

ativ

e fo

ld e

xpre

ssio

n C 1 hr 3 hr 6 hr 12 hr 24 hr 48 hr 72 hr

a

b

Fig. 7 qRT-PCR analysis of aAPX and b GR in four cultivarsof foxtail millet exposed to 0, 1,3, 6, 12, 24, 48, and 72 hdehydration treatment. A con-stitutive gene actin of rice wasused as endogenous control tonormalize the expression ofAPX and GR. Bars indicatethe standard error (±SE) calcu-lated from three independentexperiments

Differential antioxidative responses 825

The relative expression of GR constantly increased in allthe four cultivars up to 6 h of dehydration stress followed bygradual decrease except the tolerant cultivars which showedtwofold induction up to 12 h of stress (Fig. 7b; Supplemen-tary Table ESM3). Hence, both genes showed differentialexpression in all the four cultivars exposed to dehydrationstress, and their expression was confirmatory to that of ourbiochemical analyses. The PCR efficiency for all thereactions falls between the acceptable range of 90–110%.

Conclusion

In the present work, we show the existence of wide array ofgenetic diversity in a wider panel of 107 foxtail milletcultivars showing contrasting differences in the biochemicaland physiological adaptations between the dehydration-tolerant and dehydration-sensitive foxtail millet cultivars.The increased activities of CAT, APX, and GR in tolerantcultivars indicate that the oxidative damage imposed bydehydration stress is avoided in these plants whereassensitive cultivars suffered maximum oxidative damage.The overall study concludes that cv. IC-403579 and cv.Prasad are relatively more tolerant to dehydration-inducedoxidative stress in comparison to the other cultivars studied.Further qRT-PCR analysis of APX and GR in response todehydration stress also validated our physiological andbiochemical studies. Based on the differential responses of

contrasting foxtail millet cultivars subjected to dehydrationstress along with the published literature covering cellularstress tolerance in other plant species, we have elaborated aworking hypothesis for the mechanism of dehydrationtolerance in foxtail millet (Fig. 8). Increased CAT, APX,and GR activities, low levels of LP and EL, and maintenanceof membrane integrity as well as RWC in tolerant cultivarsprovide resistance against dehydration-induced oxidativestress and are therefore likely to represent components of thegeneral response to dehydration stress in foxtail millet. Inconclusion, our results comprehensively suggest that tolerantcultivars IC-403579 and Prasad possess constitutively higherlevels of H2O2 scavenging capacity and also possesseffective activation mechanisms, particularly during earlystress responses of up to 12 h of stress, in comparison to thesensitive cultivars, thus providing crucial protection againstdrought-induced oxidative damage. Thus, the identifiedpanel of dehydration-tolerant cultivars of foxtail milletmight be used for plant breeding activities.

Acknowledgments We are grateful to the Director, National Instituteof Plant Genome Research (NIPGR) for providing facilities and toUniversity Grants Commission (U.G.C.), Govt. of India for providingUGC-SRF to Ms Charu Lata. We gratefully acknowledge the financialsupport from the Department of Biotechnology, Government of India(BT/IN/FRG/04/MP/2008), and IB-BMBF, Germany (IND 07/037), forcarrying out the present study under the Indo-German bilateralprogram. We are also thankful to National Bureau of Plant GeneticResources, Hyderabad, India, for providing the seed material.

Dehydration Stress

Perception Plant response

ROS(O2

.-, H2O., H2O2, OH-)

Lipid peroxidation

MDA accumulation

Antioxidative enzymes(CAT, APX, GR, GPX, SOD )

Cellular damage

Proteins and Nucleic acids damage

Membrane and Lipid damage

Cell death

ROS Scavenging

APX

2

GSH

Electron transport

O2-

H2O2

SODCAT2 2 + ½ 2

2+

GR

Asa

NAD(P)H

NADP+

GPX

Stress responsive gene expression

Signal transduction

Synthesis of specific proteins like LEA, RAB, Dehydrins

Dehydration stress tolerance

Decreased RWCIncreased ELDecline in net photosynthesis

Accumulation of stress metabolites like MDHA, proline, Glybet, polyamines

Dehydration Stress

Perception Plant response

ROS(O2

.-, H2O., H2O2, OH-)

Lipid peroxidation

MDA accumulation

Antioxidative enzymes(CAT, APX, GR,GPX, SOD)

Cellular damage

Proteins and Nucleic acids damage

Membrane and Lipid damage

Cell death

APX

2

GSH

O2-

H2O2

SODCAT2 2 + ½ 2

2+

GR

Asa

NAD(P)H

NADP+

GPX

APX

H2O +MDAsA

AsADAsAMDAsA

GSSG

GSH

O2-

H2O2

SODCATH2O2 + ½ O2

H2O +

GR

Asa

NAD(P)H

NADP+

GPX

Stress responsive gene expression

Synthesis of specific proteins like LEA, RAB, Dehydrins

Dehydration stress tolerance

Decreased RWCIncreased ELDecline in net photosynthesis

Fig. 8 A model of the physiological, biochemical, and molecularbases of dehydration stress tolerance operating in foxtail millet iscreated based on the differential response of the four contrasting

cultivars. The enzyme assays and physiological parameters estimatedin this study are indicated in normal font and well-known conceptsreported in model species are shown in italics

826 C. Lata et al.

Conflict of interest The authors declare that they have no conflict ofinterest.

References

Abei M (1984) Catalase in vitro. Meth Enzymol 105:121–126Alscher RG, Donahue JL, Cramer CL (1997) Reactive oxygen species

and antioxidants: relationships in green cells. Physiol Plant100:224–233

Asada K (1997) The role of ascorbate peroxidase and monodehy-droascorbate reductase in H2O2 scavenging in plants. In:Scandalios JG (ed) Oxidative stress and the molecular biologyof antioxidant defenses. Cold Spring Harbor Laboratory Press,Woodbury, NY, pp 715–735

Asada K (1999) The water-water cycle in chloroplast: scavenging ofactive oxygens and dissipation of excess photons. Ann Rev PlantPhysiol Plant Mol Biol 50:601–639

Azevedo-Neto ADD, Prisco JT, Enéas-Filho J, De Abreu CEB, Gomes-Filho E (2006) Effect 446 of salt stress on antioxidative enzymesand lipid peroxidation in leaves and 447 roots of salt-tolerant andsalt-sensitive maize genotypes. Env Exp Bot 56:87–94

Barrs HD, Weatherley PE (1962) A re-examination of the relativeturgidity technique for estimating water deficit in leaves. Aus JBiol Sci 15:413–428

Bartels D, Sunkar R (2005) Drought and salt tolerance in plants. CritRev Plant Sci 24:23–58

Bhushan D, Pandey A, Choudhary MK, Datta A, Chakraborty S,Chakraborty N (2007) Comparative proteomics analysis ofdifferentially expressed proteins in chickpea extracellular matrixduring dehydration stress. Mol Cell Proteom 6:1868–1884

Blum A (1996) In: Edmeades GO, Banziger M, Mickelson HR, Pena-Valdivia CB (eds) Developing drought and low N-tolerant maizein proceedings of a symposium at the international maize andwheat improvement center. CIMMYT, El-Batan, pp 131–135, El-Batan, March 25–29, 1996

Bor M, Ozdemir F, Turkan I (2003) The effect of salt stress on lipidPeroxidation and antioxidants in leaves of sugar beet Betavulgaris L. and wild beet Beta maritime L. Plant Sci 164:7–84

Boyer JS (1982) Plant productivity and environment. Science218:443–448

Bradford MM (1976) A rapid and sensitive method for quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding. Ann Biochem 72:248–254

Bray EA (1997) Plant responses to water deficit. Trends Plant Sci 2:48–54Chandra A, Dubey A (2010) Effect of ploidy levels on the activities of

Δ-pyrroline-5-carboxylate synthetase, superoxide dismutase andperoxidase in Cenchrus species grown under water stress. PlantPhysiol Biochem 48:27–34

Chen GX, Asada K (1989) Ascorbate peroxidase in tea leaves:occurrence of two isozymes and the differences in their enzymaticand molecular properties. Plant Cell Physiol 30:987–998

Demiral T, Turkan I (2005) Comparative lipid peroxidation, antioxidantsystems and proline content in roots of two rice cultivars differingin salt tolerance. Env Exp Bot 53:247–257

Dionisio-Sese ML, Tobita S (1998) Antioxidative responses of riceseedlings to salinity stress. Plant Sci 135:1–9

El Hafid R, Smith DH, Karrou M, Samir K (1998) Physiologicalattributes associated with early-season drought resistance inspring durum wheat cultivars. Can J Plant Sci 78:227–237

Fazeli F, Ghorbanli M, Niknam V (2007) Effect of drought onbiomass, protein content, lipid peroxidation and antioxidantenzymes in two sesame cultivars. Biol Plant 51:98–107

Foyer CH, Harbinson J (1994) Oxygen metabolism and the regulationof photosynthetic electron transport. In: Foyer CH, Mullineaux

PM (eds) Causes of photo-oxidative stress and amelioration ofdefense systems in plants. CRC Press, Boca Raton, pp 1–4

Foyer CH, Lopez-Delgado H, Dat JF, Scott IM (1997) Hydrogenperoxide and glutathione associated mechanisms of acclamatorystress tolerance and signaling. Physiol Plant 100:241–254

Garg N, Noor Z (2009) Genotypic differences in plant growth,osmotic and antioxidative defence of Cajanus cajan (L.) Millsp.modulated by salt stress. Arch Agron Soil Sci 55:3–33

González L, González-vilar M (2001) Determination of relative watercontent. Handbook of plant ecophysiology techniques Reigosa.Kluwer, Dordrecht, pp 207–212

Guo Z, Ou W, Lu S, Zhong Q (2006) Differential responses ofantioxidative system to chilling and drought in four rice cultivarsdiffering in sensitivity. Plant Physiol Biochem 44:828–836

Halliwell B, Gutteridge JMC (1989) Free radicals in biology andmedicine, vol 2. Oxford University Press, Oxford

Hodgson RAJ, Raison JK (1991) Lipid peroxidation and superoxidedismutase activity in relation to photoinhibition induced bychilling in moderate light. Planta 185:215–219

Jayaraman A, Puranik S, Rai NK, Vidapu S, Sahu PP, Lata C, PrasadM (2008) cDNA-AFLP analysis reveals differential gene expres-sion in response to salt stress in foxtail millet (Setaria italica L.).Mol Biotechnol 40:241–251

Jogeshwar G, Pallela R, Jakka NM, Reddy PS, Rao JV, SreenivasuluN, Kishor PBK (2006) Antioxidative response in differentsorghum species under short-term salinity stress. Acta PhysiolPlant 28:465–475

Kumutha D, Ezhilmathi K, Sairam RK, Srivastava GC, Deshmukh PS,Meena RC (2009) Waterlogging induced oxidative stress andantioxidant activity in pigeonpea genotypes. Biol Plant 53:75–84

Lata C, Sahu PP, Prasad M (2010) Comparative transcriptome analysis ofdifferentially expressed genes in foxtail millet (Setaria italica L.)during dehydration stress. Biochem Biophys Res Commun393:720–727

Leprince O, Harren FJM, Buitink J, Alberda M, Hoekstra FA (2000)Metabolic dysfunction and unabated respiration precede the lossof membrane integrity during dehydration of germinatingradicles. Plant Physiol 122:597–608

Livak KJ, Schmittgen TD (2001) Analysis of relative gene expressiondata using real-time quantitative PCR and the 2-ΔΔCt method.Methods 25:402–408

Loggini B, Scartazza A, Brugnoli E, Navari-Izzo F (1999) Antiox-idative defense system, pigment composition, and photosyntheticefficiency in two wheat cultivars subjected to drought. PlantPhysiol 119:1091–1099

Madhusudan KV, Giridarakumar S, Ranganayakulu GS, Reddy PC,Sudhakar C (2002) Effect of water stress on some physiologicalresponses in two groundnut (Arachis hypogea L.) cultivars withcontrasting drought tolerance. J Plant Biol 29:199–202

Masood A, Shah NA, Zeeshan M, Abraham G (2006) Differentialresponse of antioxidant enzymes to salinity stress in two varietiesof Azolla (Azolla pinnata and Azolla filiculoides). Env Exp Bot58:216–222

Nagy Z, Tuba Z, Zsoldos F, Erdei L (1995) CO2 exchange and waterrelation responses of sorghum and maize during water stress. JPlant Physiol 145:539–544

Noctor G, Foyer CH (1998) Ascorbate and glutathione: keeping activeoxygen under control. Annu Rev Plant Physiol Plant Mol Biol49:249–279

Orvar BL, Ellis BE (1997) Transgenic tobacco plants expressingantisense RNA for cytosolic ascorbate peroxidase show increasedsusceptibility to ozone injury. Plant J 11:1297–1305

Pastori G, Foyer CH, Mullineaux P (1997) Low temperature-inducedchanges in the distribution of H2O2 and antioxidants between thebundle sheath and mesophyll cells of maize leaves. J Exp Bot51:107–113

Differential antioxidative responses 827

Sairam RK, Kumutha D, Aezhilmathi K, Chinnusamy V, Meena RC(2009) Waterlogging induced oxidative stress and antioxidantenzyme activities in pigeon pea. Biol Plant 53:493–504

Shigeoka S, Ishikawa T, Tamoi M, Miyagawa Y, Takeda T, Yabuta Y,Yoshimura K (2002) Regulation and function of ascorbateperoxidase isoenzymes. J Exp Bot 53:1305–1319

Silva EN, Ferreira-Silva SL, Fontenele AV, Ribeiro RV, Viégas RA,Silveira JAG (2010) Photosynthetic changes and protectivemechanisms against oxidative damage subjected to isolated andcombined drought and heat stresses in Jatropha curcas plants. JPlant Physiol. doi:10.1016/j.jplph.2010.03.005

Smart RE, Bingham GE (1974) Rapid estimates of relative watercontent. Plant Physiol 53:258–260

Smirnoff N (1995) Antioxidant systems and plant response to theenvironment. In: Smirnoff N (ed) Environment and plantmetabolism. Bios Scientific Publishers, Oxford, pp 217–243

Smith IK, Vierheller TL, Thorne CA (1988) Assay of glutathionereductase in crude tissue homogenates using 5, 5′-dithiobis(2-nitrobenzoic acid). Anal Biochem 175:408–413

Sreenivasulu N, Ramanjulu S, Ramachandra-Kini K, Prakash HS, Shekar-Shetty H, Savithri HS, Sudhakar C (1999) Total peroxidase activityand peroxidase isoforms as modified by salt stress in two cultivars offoxtail millet with differential salt tolerance. Plant Sci 141:1–9

Sreenivasulu N, Grimm B, Wobus U, Weschke W (2000) Differentialresponse of antioxidant compounds to salinity stress in salt-tolerant and salt-sensitive seedlings of foxtail millet (Setariaitalica). Physiol Plant 109:435–442

Varghese B, Naithani SC (2008) Oxidative metabolism-relatedchanges in cryogenically stored neem (Azadirachta indica A.Juss) seeds. J Plant Physiol 165:755–756

Wang WB, Kim YH, Lee HS, Kim KY, Deng XP, Kwak SS (2009)Analysis of antioxidant enzyme activity during germination of alfalfaunder salt and drought stresses. Plant Physiol Biochem 47:570–577

Wise RR, Naylor AW (1987) Chilling-enhanced photooxidation. PlantPhysiol 83:278–282

Zhang J, Liu T, Fu J, Zhu Y, Jia J, Zheng J, Zhao Y, Zhang WG (2007)Construction and application of EST library from Setaria italicain response to dehydration stress. Genomics 90:121–131

828 C. Lata et al.