ARTICLE IN PRESS

Journal of Plant Physiology 166 (2009) 507—520

0176-1617/$ - sdoi:10.1016/j.

Abbreviation�CorrespondE-mail addr

1These autho

www.elsevier.de/jplph

Physiological responses among Brassicaspecies under salinity stress show strongcorrelation with transcript abundancefor SOS pathway-related genes

Gautam Kumara,1, Ram Singh Purtya,1, Mahaveer P. Sharmac,Sneh L. Singla-Pareekb, Ashwani Pareeka,�

aStress Physiology and Molecular Biology Laboratory, School of Life Sciences, Jawaharlal Nehru University,New Delhi 110 067, IndiabPlant Molecular Biology, International Centre for Genetic Engineering and Biotechnology, New Delhi 110 067, IndiacMicrobiology Section, National Research Centre for Soybean (ICAR), Khandwa Road, Indore, MP, India

Received 18 June 2008; received in revised form 3 August 2008; accepted 3 August 2008

KEYWORDSBrassica;Electrolyte leakage;Proline;Salinity;Salt Overly Sensitive

ee front matter & 2008jplph.2008.08.001

s: DW, dry weight; FW,ing author. Tel./fax: +9ess: ashwanip@mail.jnurs contributed equally

SummarySignificant inter- and intra-specific variation for salt tolerance exists within thefamily Brassicaceae, which may be explored for dissecting genetic determinants ofthe salinity response in crops belonging to this family. Availability of contrastingcultivars for salinity response in crop species, such as Brassica, is highlyadvantageous for obvious reasons. Our analysis has indicated usefulness of availablelocal germplasm (diploid and amphidiploid) in this endeavor. Assessments carried outemploying suitable morphological, physiological and biochemical parameters inthese cultivars reconfirm established fact related to ‘in-general’ better adaptabilityof amphidiploid species over diploid ones. In our study, the salinity-tolerantamphidiploid Brassica juncea cv CS52 (AB genome) exhibited sharp contrast insalinity response as compared to the sensitive diploid species Brassica nigra(B genome). The differences included effects of salinity on overall growth,electrolyte leakage, proline accumulation and the K+/Na+ ratio (Pp0.01).Correlating well with relative stress tolerance of these Brassica cultivars, ourstudies on relative transcript abundance for salt overly sensitive (SOS) pathwayorthologues also exhibited contrasting patterns of transcript accumulation.Transcript accumulation pattern for various SOS members after 24 h of salinity

Elsevier GmbH. All rights reserved.

fresh weight; SOS, salt overly sensitive.1 11 26704504..ac.in (A. Pareek).to this work.

ARTICLE IN PRESS

G. Kumar et al.508

stress in various cultivars showed strong positive correlation with these parameters(rX0.4). Clearly, there is a need to carry out in-depth analysis to explore thesuitability of these contrasting cultivars to search for genetic determinant(s) of salttolerance among Brassica species. We propose that these contrasting Brassicacultivars can serve as suitable dicot crop models for elucidating stress-relevantgenetic determinants in genome-level analysis.& 2008 Elsevier GmbH. All rights reserved.

Introduction

Soil salinity is a major factor impairing agricul-tural productivity worldwide (Epstein et al., 1980).Efforts to improve crop performance under salinityhave been elusive owing to its multigenic andquantitative nature (Vinocur and Altman, 2005). Anapproach for studying and manipulating salt toler-ance in crop plants could be the analysis of naturalvariability existing amongst various cultivars of aparticular crop. Screening of available local/exoticcultivars of crop plants (along with their wildrelatives) for salinity tolerance has two majoradvantages, first the tolerant genotype thus madeavailable can be used in breeding programs andsecond, a comparative analysis at physiological/biochemical and/or molecular level of these con-trasting cultivars can pave the way in understand-ing and unraveling novel survival mechanisms(Amtmann et al., 2005; Bohnert et al., 2006).

The genus Brassica, which is closely related toArabidopsis thaliana (belongs to the same taxo-nomic family), is an economically important cropplant. It includes a number of crops with a widespectrum of adaptation for cultivation under variedagro-climatic conditions. Amongst the six culti-vated species of Brassica, Brassica campestris (AA,2n ¼ 20), Brassica nigra (BB, 2n ¼ 16) and Brassicaoleracea (CC, 2n ¼ 18) are diploids. The remainingthree, namely, Brassica juncea (AABB, 2n ¼ 36),Brassica napus (AACC, 2n ¼ 38) and Brassica car-inata (BBCC, 2n ¼ 34) are amphidiploids, whichhave evolved by hybridization between the diploids(Morinaga, 1934; UN, 1935). For salinity tolerance,significant inter- and intra-specific variation existswithin the Brassica genera, and thus these cultivarshave enticed researchers from all disciplines ofclassical and contemporary biology (Ashraf, 1992;Ashraf and Sharif, 1997, Ashraf and McNeilly, 2004;Li et al., 2004; Das et al., 2005). These tolerantcultivars can potentially be exploited for identifi-cation of useful candidate genes for enhancing salttolerance of the sensitive cultivars.

Investigations into wheat genetic system andsalinity tolerance have been fruitful since thegenetic determinant has also been identified

recently (Gorham et al., 1987; Lindsay et al.,2004). In contrast, in spite of well-reportedvariations in salinity tolerance among diploid andamphidiploid cultivars of Brassica (Ashraf et al.,2001), there has been a major lack of detailedinformation related to molecular analysis of theirresponse. Though multigenome sequencing pro-jects for Brassica are underway (http://brassica.bbsrc.ac.uk; Love et al., 2004), it may not bepossible to harvest the results of these efforts untilnext several years. In the past, investigations intohow Arabidopsis respond to stress such as salinityhave been boosted by several factors such asavailability of whole genome sequence, mutantsfor almost all traits, etc. (Rhee et al., 2003). ThusArabidopsis has been a preferred dicot modelsystem for analysis related to genetic architecture(Zhu, 2000; Quesada et al., 2002; Hannah et al.,2006). The information has got a further boost bythe recent findings that one of its halophyte wildrelative Thellungiella halophila shows a contrastingresponse when it comes to regulation of transcrip-tome (Kant et al., 2006). It has been hypothesizedthat the mechanism of salt tolerance in halophytesare primarily the same as those known in glyco-phyte, with subtle differences in ‘regulation’resulting in different responses towards salinity.Recent studies related to gene expression data incontrasting rice cultivars have further indicatedthat salinity tolerance of Pokkali may be due toconstitutive overexpression of many genes thatfunction in salinity tolerance and are stressinducible in IR64 (Kumari et al., 2008). Does itmean that the differential regulation of stresstranscriptome is the ‘key to survival’ under salinitystress in the tolerant cultivars?

With an objective to further test the abovehypothesis, and to make an attempt to bridge thegap in our understanding of the physiology andmolecular biology of salinity tolerance in Brassicacultivars, we have picked up six cultivatedspecies (with more than one variety for someof the species) along with two wild germplasms(Table 1) and analyzed their salinity-inducedvariations at the biochemical level. Further,we have also performed limited transcriptional

ARTICLE IN PRESS

Table 1. Description of various Brassica genotypes used in this study

Species Varieties Chromosomes (2n) Genome

B. carinata HC209 and HC210 34 BBCCB. juncea RH8813 and CS52 36 AABBB. napus HNS9605 38 AACCB. oleracea PT30 and PT303 18 CCB. campestris BSH1 20 AAB. nigra 16 BBE. sativa T27 22 Wild typeB. tournefortii 20 Wild type

B. ¼ Brassica, E. ¼ Eruca.

Salinity response in Brassica and SOS pathway 509

profiling (un-induced vs salinity induced) in theseBrassica species, for which we have chosen to work oncandidate genes belonging to a well-characterizedpathway in Arabidopsis related to ion homeostasis –

salt overly sensitive (SOS) pathway (Halfter et al.,2000; Ishitani et al., 2000; Guo et al., 2001; Qiuet al., 2002; Quintero et al., 2002; Gong et al., 2004).Though there may not be a direct relationshipestablished as yet between some of these physiologicalparameters (such as electrolyte leakage) and SOStranscript accumulation, the SOS pathway has beenchosen for analysis, because of its proven role insalinity-stress tolerance. For this purpose, we havecloned the orthologues of SOS1, SOS2 and SOS3 fromone of the salinity-tolerant Brassica genotype (Brassicajuncea var. CS52) and carried out RNA abundanceanalysis of root and shoot tissue. Tissue-specificexpressions for these SOS genes play a key role inNa+ regulation. We report the differential pattern ofregulation of these key genes in contrasting cultivarsof Brassica and hope that understanding of thesenatural variations, and the underlying mechanisms,may ultimately be useful for enlightening us and thesemay then be exploited for incorporating the ‘tolerancetrait’ into current economically important stress-sensitive cultivars (Bohnert et al., 2006).

Materials and methods

Plant materials and growth conditions

Seeds of various diploid and amphidiploid Brassicacultivars along with wild relatives (Table 1) were washedwith de-ionized water and allowed to germinate in ahydroponic system for 48 h in dark and then transferredto light for further growth under control conditions(2572 1C, 12 h light and dark cycle).

Stress treatments

For salinity-stress treatment, 6-d-old seedlings weretreated with 200mM NaCl for 24, 48 and 72 h using the

hydroponic system. Simultaneously, seedlings maintainedin de-ionized water were taken as control. For experi-ments related to transcript analysis, seedlings weretreated with 200mM NaCl for 5 h.

Analysis of stress response on growth of seedlings

To study the effect of salinity stress (200mM NaCl) onseedling growth, after 24, 48, 72 h of salt treatment (S),shoot and root length of the seedlings were measured bycomparing with the unstressed control seedlings (C).Since the various cultivars analyzed in this study haddifferent rates of growth under control conditions, wecompared these cultivars based on the relative percen-tage change, which was calculated by applying theformula (C�S)/C� 100. For each time point, a minimumof 30 seedlings were taken and the experiment wasrepeated at least three times. For each treatment,standard error was also calculated (n ¼ 90).

Electrolyte leakage

For measuring electrolyte leakage, seedlings wereharvested at 24, 48 and 72h of salt treatment (along withthe unstressed controls). Then they were quickly washedwith distilled water to remove adhering ions leachedduring the salt treatment, followed by making smallexcisions. Electrolyte leakage was measured as described(Bajji et al., 2004). Around 100mg of the tissue wasimmediately dipped in 20mL de-ionized water andincubated at 32 1C. After 2 h, the electrical conductivity(E1) of the immersion solution was measured using aconductivity meter (ELEINS Inc., India). To determine thetotal conductivity (E2), the seedlings with immersionsolution (effusate) were autoclaved for 15min at 121 1Cand the conductivity of the effusate was measured aftercooling it down to room temperature. Relative electricalconductivity was calculated by the formula E1/E2� 100.For each time point, the experiment was repeated at leastthree times (each time with three replicates). For eachtreatment the standard error was also calculated (n ¼ 9).

Proline content

Proline was extracted and estimated using the stan-dard protocol (Bates et al., 1973). Around 100mg

ARTICLE IN PRESS

G. Kumar et al.510

seedlings of both salt stressed as well as unstressedcontrols were used. Proline content was calculatedfrom the standard graph, prepared by using pure proline.The experiment was repeated thrice with three repli-cates each time and standard error was calculated(n ¼ 9).

Estimation of Na+ and K+ contents in various Brassicacultivars

For determination of endogenous Na+ and K+ contents,100mg of seedling tissue (unstressed or salinity stressed)of various Brassica cultivars was taken and digested in0.1% HNO3. Ions were extracted in distilled H2O by boilingfor 30min twice. The filtered extract thus obtained wasused to measure specific ions with a flame photometer(Corning EEL, UK). The experiment was repeated thriceand standard error was calculated (n ¼ 3).

RNA extraction and RNA gel blot analysis

Total RNA was extracted from tissue using the TRIzolmethod as per the manufacturer’s instructions (Invitro-gen, USA). RNA gel blots were prepared using 20 mg totalRNA as described (Kumari et al., 2008). RNA gel blotswere hybridized with various SOS probes isolated by ourgroup from Brassica juncea cv CS52 (sequence data fromthis article have been deposited at NCBI under accessionnumbers – SOS1: EF206779; SOS2: EF190471; SOS3:DQ248965) and scanned on Phosphorimager using thesoftware Fujifilm Image Reader. High stringency wasmaintained during hybridization as well as washing toensure specificity of signal on membranes. The relativetranscript abundance was calculated using the ImageGauge (Fuji Photofilm Co. Ltd., Japan). Deprobing ofblots was performed in 0.2� SSC, 0.1% SDS and blotswere re-used.

Data analysis and statistical relevance

Absolute values for transcript abundance from RNA gelblots were quantified using the Image Gauge software(Fuji Photofilm Co. Ltd., Japan). Analysis of variance ofdata and their correlation with transcript accumulationfor SOS genes were carried out using COSTAT computerpackage (CoHort Software, USA).

Results

Natural variability of salinity toleranceamong Brassica cultivars

Early seedling growth is differentially affected inBrassica cultivars under salinity

Within 24 h of stress treatment, seedlings of allcultivars started exhibiting the visual symptoms ofstress-induced injury. Seedlings started losing

‘greenness’, which is a reflection of the adverseaffects of stress on photosynthesis. However, theextent of loss in greenness was different indifferent cultivars. Seedlings of all the cultivarslost their turgidity within 48–72 h of salinitytreatment except Brassica nigra, where turgiditywas lost very early (within 2–3 h of stress). Theseedlings under stress showed arrest in both root(Figure 1A) and shoot growth (Figure 1B) within24 h. Among the diploids, Brassica nigra andBrassica oleracea (both var. PT30 and PT303)showed a 60% and 45% reduction in root length,respectively, over a period of 24–72 h of salinitystress, while Brassica campestris exhibited arelatively low impact of stress damage as it showedonly 20–30% reduction in root length. On theother hand, amphidiploids, Brassica napus andBrassica carinata (var. HC209 and HC210), exhib-ited an average 30% reduction in root lengthunder similar conditions. However, a unique dif-ferential response was observed between twocultivars of Brassica juncea (var. RH8813 andCS52), where the former behaved similar toBrassica carinata (�30% reduction in root length),and later showed very less (minimum among allcultivars) reduction in root length (�15%) over astress period of 24–72 h. Both the wild germplasms,i.e. Eruca sativa and Brassica tournefortii,showed around 40% decrease in root length. Theseobservations clearly indicate the better ability ofseedlings of Brassica juncea var. CS52 to be able toresist the salinity-induced reduction in root lengthas compared to all others cultivars tested here(Figure 1A).

To score the effect of salinity stress on reductionin shoot length of seedlings of Brassica species, wedecided to measure the relative percentage reduc-tion (similar to the parameter used for rootanalysis). At 24, 48 and 72 h, maximum percentagereduction was observed in case of Brassica tourne-fortii, Brassica oleracea (var. PT303) and Erucasativa (var. T27) respectively. Within diploids,Brassica nigra and Brassica oleracea showed almostsimilar pattern for salinity-induced percentagereduction in shoot length, which was higher thanthat of Brassica campestris (Figure 1B). Similarly,among the amphidiploids, Brassica napus andBrassica carinata exhibited similar salinity responsei.e. a progressive decrease in shoot length till 72 h.However, within Brassica juncea, there was a clearcontrast between the two cultivars with CS52showing less decrease in shoot length due tosalinity than RH8813. The wild germplasms, Erucasativa and Brassica tournefortii, exhibited a sensi-tivity pattern similar to the diploid species Brassicanigra and Brassica oleracea.

ARTICLE IN PRESS

Figure 1. Analysis of various morphological and biochemical parameters in seedlings of Brassica cultivars exposed to200mM NaCl for up to 72 h. (A) Relative percentage decrease in root length. (B) Relative percentage decrease in shootlength. (C) Percentage change in electrolyte leakage. (D) Endogenous proline content (mg g�1 DW). (E) Endogenous Na+

content (mg g�1 FW). (F) Ratio of K+/Na+. Data are means7SE. Each data set represents an average of minimum threeseparate experiments. The root length of seedlings (before stress treatment) of various cultivars was in the range of9–19 cm (24 h), 12–22 cm (48 h) and 13–28 cm (72 h), while shoot length varied from 1.6 to 4.3 cm (24 h), 1.5 to 4.6 cm(48 h) and 2.7 to 4.8 cm (72 h). Similarly, the proline concentration (before stress treatment) among the variouscultivars was found to be below 0.1mg g�1 DW. Concentration of Na+ was found to be in the range of 0.5–1.0mg g�1 FWin all the cultivars before stress treatment.

Salinity response in Brassica and SOS pathway 511

On an overall basis, the shoots were found to beprogressively more affected than roots (in terms ofreduction in length) with the increase in theduration of salinity stress. Cultivars differed sig-nificantly (Pp0.01) in growth (percent decrease inroot and shoot length) in response to 200mM NaCl.Taken together, it can be concluded that Brassica

campestris and Brassica juncea var. CS52 representthe cultivars that show least reduction in salinity-induced seedling growth among the various diploidsand amphidiploids, respectively. It can also beinferred that Brassica juncea var. CS52 and Brassicanigra represent the most salinity-tolerant andsensitive cultivars, respectively.

ARTICLE IN PRESS

G. Kumar et al.512

Electrolyte leakage to measure the injury duringsalinity stress

In our experiments, we observed an increase incell injury in terms of electrolyte leakage inseedlings under salinity stress (Figure 1C). Stabilityof cell membrane (as measured by electrolyteleakage) differed significantly (Pp0.01) amongstcultivars, in response to 200mM NaCl. Among allthe cultivars tested here, the seedlings of Brassicajuncea var. CS52 showed minimum electrolyteleakage (�50%) after 24 h of salinity stress, whichwas further not affected by extending the durationof stress (48–72 h). In contrast, electrolyte leakagein seedlings of Brassica nigra was found to bevery high (480%) throughout the experiment(Figure 1C). Similarly, the wild cultivars alsoexhibited high electrolyte leakage (70–90%) undersimilar conditions. This analysis clearly indicatesthat cell-membrane stability was least affectedunder salinity stress in seedlings of Brassica junceavar. CS52.

Differential accumulation of proline in Brassicaseedlings under salinity stress correlating withtheir stress tolerance

Accumulation of proline amongst cultivars dif-fered significantly (Pp0.01) in response to 200mMNaCl, indicating that this stress treatment hasosmotic factor along with ionic imbalance. As seenin Figure 1D, salinity-induced accumulation ofproline was noticed in seedlings of all the Brassicacultivars in a time-dependent manner. After 24 h ofsalinity treatment, not much difference was ob-served in the proline content with respect to un-stressed conditions in all the cultivars (Figure 1D; itwas below 0.5mg g�1 dry weight (DW) in all thespecies). However, the differential pattern ofstress-induced proline accumulation amongst thevarieties was apparent after 48 h of stress, where italmost doubled in some species (Brassica carinatavar. HC209; 1mg g�1 DW). The difference in prolinecontent between diploid and amphidiploid cultivarswidened further after 72 h of salinity treatmentwhere it was estimated to be between 0.8 and2.5mg g�1 DW in diploids, while in case ofamphidiploids, it increased up to 4.2mg g�1 DW(Brassica juncea var. CS52). In Brassica tournefortiiand Eruca sativa var. T27 proline concentration wasestimated to be similar to diploids (0.6 and1.3mg g�1 DW), after 72 h of salinity stress.

In our analysis, it was observed that in responseto salinity, Brassica nigra exhibits minimum per-centage change in endogenous proline level (up to20% of control), while Brassica juncea var. CS52exhibits maximum change (up to 80%), indicatingtowards its better capability to adjust the cellular

osmoticum under salinity stress. The salinity-induced proline accumulation pattern in variousBrassica cultivars is in good correlation with theirrelative stress tolerance, based on the parameteranalyzed in our experiments (Figure 1A–C). Thispossibly indicates towards direct or indirect adap-tive role for salinity-induced proline accumulation.

Na+ and K+ accumulation patterns in response tosalinity

We used the plant’s ability to maintain low Na+

inside the cells – as a basis for screening variousbrassica cultivars for differential tolerance in ourexperiments. The analysis (Figure 1E and F)indicated towards time-dependent differentialNa+ accumulation amongst the various Brassicacultivars (Pp0.01). For example, during the first24 h, the three diploids accumulated relativelyhigher Na+ (2–4mg g�1 fresh weight (FW)) thanthe amphidiploids (o2mg g�1 FW) and wild germ-plasms (Figure 1E). Barring one genotype (Brassicacarinata var. HC210, which showed Na+ accumula-tion comparable with diploids), this trend contin-ued till 48 h of stress. However, after 72 h of stress,all the cultivars accumulated high Na+ (X6mg g�1

FW), indicating that by this time, seedlings losetheir ability to restrict the entry of Na+ ions and/orexclude them from the cytosol. Hence, they arefacing ion toxicity in addition to osmotic stress.

One of the key features of plant salt tolerance istheir ability to maintain optimal K+/Na+ ratio in thecytosol (Tester and Davenport, 2003; Singla-Pareeket al., 2003). With this idea, when Na+ and K+

amounts were calculated, Brassica juncea var. CS52exhibited the highest ratio of 4 (after 24 h ofstress), while Brassica carinata (HC209 and 210)exhibited a ratio of 2–2.5 (Figure 1F). The threediploids showed a sharp decline in this ratio (�0.5),while the wild species (Eruca sativa and Brassicatournefortii) showed an intermediate response(a ratio of 1–1.5). During the extended period ofstress (48 and 72 h), all the cultivars showedinability to maintain a ratio of X0.5. Takentogether, this study clearly established the abilityof amphidiploids to be able to maintain higherK+/Na+ ratio than their counterpart diploids at leastduring the initial period (24 h) of salinity stress,which is in conformity with previous reports (Ashrafet al., 2001). The inability of the seedlings tomaintain this favourable ratio indicated that seed-lings at this early stage are very sensitive to salinitystress (200mM NaCl). Though the different Brassicaspecies show different patterns for accumulation ofspecific ions, most have been reported to havemechanisms for exclusion of toxic ions (Ashraf andMcNeilly, 2004).

ARTICLE IN PRESS

Salinity response in Brassica and SOS pathway 513

Differential regulation of SOS transcripts invarious Brassica species during earlyexposure to salinity stress

Fine regulation of SOS pathway controls ionhomeostasis in Arabidopsis and mutation in any ofthese genes render plants more sensitive towardssalinity stress (Zhu, 2003). With this viewpoint, wedecided to isolate and characterize SOS pathwaymembers from Brassica juncea var. CS52. Constitu-tive (un-induced) and salinity-induced transcriptlevels for SOS genes, in various Brassica species,were analyzed using RNA gel blot analysis. Each ofthe SOS1, SOS2 and SOS3 probes hybridized with asingle band, corresponding to the respectivetranscript size in all the Brassica cultivars, indicat-ing the sequence conservation of SOS genes, as wellas conservation of this pathway in all Brassicacultivars studied (Figure 2; the RNA gel blots have

Figure 2. Transcript abundance for SOS pathway genes durintissues of various cultivars. RNA gel blot was sequentially hybrand ‘+’ on top of the lanes indicate the absence and presencThe ethidium bromide (EtBr)-stained gel has been shown asabundance in root and shoot tissues has been shown for BjSOSbars represent control samples and filled bars represent stre

been shown in the upper panel, while the bardiagrams in the lower panel depicts their relativetranscript abundance). Both root (Figure 2A) andshoot (Figure 2D) tissues, under non-stress condi-tions, showed constitutive expression of SOS3transcript, which was found to be further upregu-lated within 5 h of salinity stress in all the cultivarsstudied except in Brassica oleracea, which did notshow any change in shoots, while it showed a two-fold induction in roots. Barring Brassica campestris,all diploid species showed relatively low SOS3transcripts, than the amphidiploids, in both rootand shoot tissues. These results should be seen inlight of the observation that among the diploids,Brassica campestris exhibited a relatively lowerimpact of stress damage (Figure 1A and B). Takentogether, Brassica nigra showed significantlylow levels of SOS3 transcript in both root andshoot tissues (under both non-stress and stress

g ‘early phase’ of salinity stress in Brassica root and shootidized using BjSOS1, BjSOS2 or BjSOS3 cDNA as probe. ‘�’e of 200mM NaCl stress, respectively, for a period of 5 h.the loading control. Bar diagram for relative transcript3 (A and D), BjSOS2 (B and E) and BjSOS1 (C and F). Emptyss samples.

ARTICLE IN PRESS

G. Kumar et al.514

conditions), while Brassica juncea always exhibitedhigher SOS3 transcripts.

The expression pattern for SOS2 amongst Brassicaspecies was found to be very unique. In all theamphidiploid species, both root and shoot tissuesshowed constitutive presence of SOS2 trans-cript (Figure 2B and E). In contrast, the diploidsshowed undetectable or very marginal constitutivelevels for SOS2 transcripts in root (except forBrassica juncea) and shoot tissues (except Brassicacampestris). SOS2 was upregulated by salinitystress in both root and shoot tissues of all thespecies except in shoots of Brassica nigra (whichremain unchanged) and Brassica oleracea (whichshowed down-regulation). In roots, there was aconstitutive high expression of SOS2 exclusivelyobserved in Brassica juncea, which otherwise couldbe seen only upon salinity induction in all othercultivars. This indicates a very unique feature ofBrassica juncea (Figure 2B).

The transcript level for SOS1 was found to beconstitutively expressed in root and shoot tissues ofBrassica species (Figure 2C and F). All the cultivars,except Brassica oleracea, showed salinity-inducedupregulation of SOS1 transcript in shoots. It wasdown-regulated in roots of all the cultivars, exceptBrassica juncea var. CS52, where it increasedmarginally. The roots of Brassica nigra showed asharp decline in SOS1 transcripts upon salinitystress.

Differential transcript abundance under saltstress indicated a complex pattern of expressionfor SOS members, which may regulate Na+ accu-mulation in Brassica cultivars. Taken together, theaccumulation patterns for various SOS transcriptsmatched well with the relative tolerance pattern indiploid and amphidiploid species (Figure 1). Forexample, the two contrasting cultivars Brassicanigra and Brassica juncea var. CS52 exhibited agood contrast in SOS transcript abundance underboth un-induced and salinity-induced conditions(compare Figures 1 and 2).

Discussion

Soil salinity is one of the major threats toagriculture throughout the world. The problem ofsalinity is increasing, rendering more areas ofarable lands unproductive. Screening for salinity-tolerant cultivars within the genera has an addedobvious advantage over those among differentgenera. In this regard, polyploids and diploids fromthe same genera have been a preferred tool ofanalysis in most studies (Rawson et al., 1988;

Ashraf et al., 2001; Munns and James, 2003). This isso because generally polyploids can withstandsalinity better than diploids. For example, currentdurum wheat (tetraploid) cultivars are more sensi-tive to soil salinity than bread wheat (hexaploid)cultivars (Francois et al., 1986; Rawson et al.,1988; Shannon and Grieve, 1999). However, in spiteof several physiological studies in Brassica, thegenome-level molecular approaches are not asadvanced as research done using other geneticsystems that exhibit natural variation for salttolerance.

Our objective in the present analysis has been tofirst draw a relative salinity-tolerance index foravailable local/exotic germplasm of Brassica basedon screening at morphological, physiological andbiochemical levels, followed by identification ofthe two most contrasting cultivars amongst them.Further, to have an insight into what might becontrolling this contrasting response, we haveperformed limited transcriptome analysis employ-ing SOS gene probes.

Exploring natural variation for salinitytolerance in cultivated and wild relatives ofBrassica and establishment of contrastingcultivars

In the present study, survey of 11 germplasms ofBrassica identified considerable natural variationfor salinity tolerance. In past, such screenings havebeen conducted by taking three most importantparameters viz. (a) growth/yield, (b) damage/tolerance to salinity and (c) physiological mechan-isms (Munns and James, 2003). In this work,detailed screening/comparative analysis has beencarried out at the seedling stage, employingphysiological (electrolyte leakage), biochemical(proline accumulation) and morphological (seedlinggrowth) investigations. As reported for similarscreenings in rice (Lee et al., 2003), we havecompared the varietal differences in salinitytolerance in terms of percentage growth reductionsfor all the parameters. The present study is thewidest survey of salinity response of Brassicagermplasms (including wild species) where differ-ent physio-morphological, biochemical and mole-cular responses have been co-analyzed.

In the present investigation, all the speciesshowed sensitivity towards salinity stress and adecrease in overall growth, though differential innature (Figure 1). Among the different speciestested, maximum percentage reduction in rootlength was observed in case of Brassica nigra,whereas Brassica juncea var. CS52 had the least

ARTICLE IN PRESS

Salinity response in Brassica and SOS pathway 515

penalty, indicating Brassica nigra to be moresensitive and Brassica juncea var. CS52 to berelatively more tolerant towards salinity stress,under the conditions employed in the present study(Figure 1A and B). Thus, genotype such as Brassicajuncea var. CS52, which is able to maintain growthof root even after 72 h exposure to salinity (whereother species could not do so), present before us aninteresting research material where detailed in-vestigation needs to be carried out to find thegenetic basis of this important trait. These ob-servations are significant in light of the findingsthat root growth and architecture are importantdeterminants in assessing the tolerance of plantstowards salinity (Gaxiola et al., 2001). However,Brassica carinata and Brassica napus, both amphi-diploids, also showed similar root growth patterns,indicating some correlation between salt tolerance(ability to grow roots under salinity) and chromo-some number. Previously, enhanced vegetativegrowth has been observed for amphidiploid culti-vars, as compared to diploids, in the presence ofsalinity (Ashraf and McNeilly, 1990; Ashraf et al.,2001).

It has now been well established that decrease inwater potential, caused by salinity stress, leads tocell membrane damage in almost all plant species(Chen et al., 1999; Sreenivasulu et al., 2000). Cellmembrane is one of the prime targets of manyplant stresses and its maintenance and integrityunder stress conditions is a major determinant oftolerance in plants. Our results indicate thatamphidiploids (especially Brassica juncea var.CS52) efficiently restrict the amount of electrolyteleakage as compared to diploids (Figure 1C). Inliterature, a broad range of compounds such asproline, glutamate, glycine betaine, etc. have beenidentified, which may contribute towards protect-ing the membrane proteins and other enzymesduring salinity stress (Chandler and Thorpe, 1987;Delauney and Verma, 1993).

The amphidiploid cultivars were able to maintainhigher levels of proline even under extendedduration of salinity stress, indicating their inherentability to maintain cellular osmoticum. Proline isbelieved to protect plant tissues against stress byacting as an osmo-regulator and as a protectant forenzymes and cellular structure (Kavi Kishor et al.,2005). Though proline accumulation under osmoticstresses such as salinity and drought is a muchwidely reported phenomenon in several biologicalsystems, its exact link with stress toleranceremains puzzling. Synthesis of compatible solutesin plants, such as proline, in response to salinitystress is a possible strategy to engineer salttolerance in plants and has been discussed as well

as debated several times (Apse and Blumwald,2002). Proline accumulation under stress conditionsmay either be caused by induction or activation ofenzymes of proline biosynthesis, a decreased pro-line oxidation to glutamate, decreased utilizationof proline in protein synthesis or enhanced proteinturnover (Delauney and Verma, 1993). However, adirect relationship between the increase in prolinelevels and increased expression of the key enzyme,P5CS, was found in sos1 mutants (Liu and Zhu,1997). In our work, Brassica juncea var. CS52 wasfound to have an advantage of being able toaccumulate the highest amounts of proline(4.5mg g�1 DW) under salinity stress. This may bedirectly or indirectly related to its tolerancetowards salinity, as observed in the analysis.However, which mechanism out of those listedabove contributes towards proline accumulation inthis genotype needs to be established.

Under high salt conditions, more Na+ enters thecell as the similarity in the hydrated ionic radiibetween Na+ and K+ makes it difficult for thetransporter to discriminate between the two ions(Blumwald et al., 2000). This ion homeostasis canactually be a reflection of several differentstrategies that the plant uses such as diminishingthe entry of Na+ ions into cells, extrusion of Na+

ions out of the cell or/and vacuolar compartmenta-tion of Na+ ions. The ability to maintain higherK+/Na+ ratio by amphidiploids over diploids mayindicate towards any or a combination of all ofthese strategies. It is clear that Brassica juncea var.CS52 exhibits the most effective combination ofthese strategies and is hence able to maintainfavourable K+/Na+ ratio, at least during the initial24 h of salinity stress (Figure 1F).

Taken together, our analysis clearly indicatesthat irrespective of the parameter employed forscreening, amphidiploids generally have a betterability to tolerate salinity stress as compared todiploids. We have analyzed the response of culti-vars only till 72 h, but keeping in mind that 6-d-oldseedlings are highly sensitive towards salinity, itwas not possible to extend the analysis period. Inliterature, there are numerous reports wheresuperiority of amphidiploids over diploids Brassicaspecies has been established. Studies have beenextended to as long as 51 d and similar responseshave been documented (Ashraf et al., 2001).Though we have employed seedlings of localcultivars in our analysis, our results are in con-formity with established reports. The wild speciesanalyzed here were picked up because they arereported for their expected ability to toleratestress (Ashraf and McNeilly, 2004). But our analysis,on the basis of comparison with other cultivars used

ARTICLE IN PRESS

G. Kumar et al.516

here, failed to establish their high tolerance underthe set of conditions we have used. However, wecannot rule out the possibility that under naturalenvironmental conditions, these wild relatives mayperform better. Additionally, the stress responseobserved at the seedling level may be differentfrom the one observed at the mature plant level.

It can also be safely concluded that among thecultivars being studied here Brassica juncea var.CS52 is most tolerant towards salinity. Brassicajuncea var. CS52 is a local variety developed at theCentral Soil Salinity Research Institute, Karnal,India. Brassica juncea var. CS52 (genome AABB;2n ¼ 36) has been recommended for cultivationin saline lands [having EC up to 7–8 dsm�1 and pHup to 9.2–9.3 and average yield 1.1 t ha�1 (www.plantstress.com/files/salt-karnal.htm)]. On theother hand, Brassica nigra is diploid having genomeBB, 2n ¼ 16 chromosomes. Previous studies havesuggested that the tolerance in amphidiploids hasbeen acquired from the A (Brassica campestris)and C (Brassica oleracea) genomes. However, this isstill being debated in the literature (Ashraf andMcNeilly, 2004). Nonetheless, the availability ofthese two contrasting cultivars may serve as aplatform where future investigations employing allpossible tools may be attempted to understand theresponse of this crop plant towards salinity stress.

SOS pathway genes are differentiallyregulated in contrasting Brassica cultivars atthe seedling stage

Salinity tolerance is a very complex trait in plantspecies as there are numerous mechanisms operat-ing at cell, tissue, organ or whole plant level (Yeo,1998). Plants adapt to these stress conditions bycoordinated and orchestrated functioning of var-ious complex mechanisms. Most of the adaptiveresponses of plant towards salinity stress are alsocontrolled by their developmental status (DeRocherand Bohnert, 1993). For example, it is establishednow that the seedling stage as well as thereproductive stage represents the two most sensi-tive stages in the life cycle of plants (Drake andDrake, 1998; Houle et al., 2001). The SOS pathwayis important for ion homeostasis and salt tolerancein plants, for which fine details related to itsfunctioning have been fairly established (Hasegawaet al., 2000; Sanders, 2000; Zhu, 2000, 2001a, b,2002, 2003; Shi et al., 2003). SOS1, SOS2 and SOS3are the genes whose mutation causes a saltresponse different from that of the wild-typeancestors during vegetative growth and have beenthe candidate genes of choice in studies related to

scoring of genetic variability that occurs naturallyin Arabidopsis genotypes (Quesada et al., 2002).The SOS pathway has recently been shown to beconserved in rice as well (Martınez-Atienza et al.,2006).

All the cultivars (diploids and amphidiploids)show a clear single transcript on blots for all theSOS genes, indicating that the SOS pathway genesare conserved among Brassica species as well. Onan overall basis, the expression pattern of SOS1,SOS2 and SOS3 was found to be different (Figure 2).Analysis of SOS transcripts at the seedling stageindicated that barring Brassica campestris, thetranscript for calcium sensor component (SOS3) ismore abundant in both roots and shoots (Figure 2Aand D) of the amphidiploids (both constitutiveas well as stress inducible). SOS2 is a major salt-tolerance locus in Arabidopsis thaliana and muta-tion in SOS2 gene drastically reduces planttolerance to high Na+ stress (Liu et al., 2000).Previous reports have documented the presence ofSOS2 mRNA in both root and shoot tissues ofArabidopsis thaliana. SOS2 mRNA has beenreported to be inducible in both root and shoottissues within 3–6 h of salinity stress (Qiu et al.,2002). In our analysis, we found that this regulatorygene, i.e., SOS2, showed highly inducible nature inall cultivars studied except Brassica juncea, whichalready maintains high SOS2 transcripts even undernon-stress conditions. We are further working onthe isolation of SOS2 promoter from both Brassicanigra and Brassica juncea var. CS52 and efforts arebeing made to find the key cis-regulatory elementsresponsible for its differential expression in the twocontrasting cultivars in a tissue-dependent manner.

SOS1, the final element of the SOS salt-tolerancepathway, has been found predominantly in the roottips (Shi and Zhu, 2002). Consequently, it wasassumed that this is the region where the SOS1 isfunctionally expressed (Zhu, 2003). However, thefunction of the entire root has been documented tobe altered in the sos1 mutants, suggesting thatlocation of the gene expression does not alwaysportray the location of the gene function (Shabalaet al., 2005). In our study, this effector gene, i.e.,SOS1, did not show much change in transcriptaccumulation in various Brassica species (exceptroots of Brassica nigra). But it has to be notedclearly that only Brassica juncea var. CS52 showedinduction for SOS1 transcript in response to stressin roots, while other showed a down-regulation(Figure 2C). SOS1 expression has been reported tobe upregulated in roots as well as shoots ofArabidopsis thaliana within 5 h by 300mM NaClstress (Shi et al., 2000). In our analysis, we havefound an upregulation of SOS1 only in shoots in

ARTICLE IN PRESS

Salinity response in Brassica and SOS pathway 517

response to salinity stress. Whether this differencein tissue-specific expression is due to the differencein stress dose or difference in cultivars needs to beanalyzed further. In rice also, OsSOS1 has beenfound to be regulated by salinity stress in a time-dependent manner (Martınez-Atienza et al., 2006).It has been reported earlier that salt stress causes apost-transcriptional stabilization of SOS1 tran-script, which seem to be an important mechanismfor gene regulation and has been seen previouslyfor several salt-regulated genes (Shi et al., 2000,Cushman et al., 1990, Hua et al., 2001). A note ofcaution is thus required when interpreting tran-scriptomics data. Sometimes the transcript andprotein levels may not be correlated with each

Table 2. Relationship of SOS1, SOS2 and SOS3 transcript accgenotypes of Brassica in response to salinity stress

Category Parameter

Growth Percent decrease in root lengthPercent decrease in shoot length

Injury Electrolyte leakageNa+ content

Altered physiology Proline content

*Values shown in bold are highly and positively related.

Figure 3. Cartoon depicting comparative account for stress rvar. CS52. Comparison has been made taking growth (rootbiochemical (proline) and molecular (transcript accumulatispecific differences could be noted in transcript accumulationspecific manner, for simplicity purpose, only one representativgiven response is indicated in the form of an empty bar, whilebars. Note the differences between the two cultivars withRelationship between molecular and rest of the parameters isin Brassica.

other, as was exemplified for developmental andenvironmental regulation of aquaporins in radish(Suga et al., 2001) and maize (Aroca et al., 2005).In addition, mechanisms that couple stimuli tochanges in protein sub-cellular localization canentirely determine their expression properties.

Thus, our study could establish a strong relation-ship between SOS transcript accumulation andphysiological parameters across different cultivarsof Brassica in response to salinity stress (Table 2and Figure 3). Our present observations providefurther support to the hypothesis that, similar tothe situation observed in Arabidopsis and rice,differential gene regulation is the key to survivalfor salinity-tolerant cultivars of Brassica. Finally, we

umulation with physiological parameters across different

Correlation coefficient (r) Inference

SOS1 SOS2 SOS3

0.041 �0.33 �0.66 Strong negative�0.02 �0.28 �0.62 Strong negative

0.059 �0.26 �0.55 Strong negative0.40 0.64 0.89 Strong positive

0.91 0.85 0.50 Strong positive

esponse in seedlings of Brassica nigra and Brassica junceaand shoot length), damage (cell injury), K+/Na+ ratio,on for various genes belonging SOS pathway). Though,of various SOS members within these cultivars in a tissue-e bar has been shown for all SOS members. The value of athe changes in response to salinity are reflected as filledrespect to various parameters analyzed in this study.evident, indicating the genetic basis of salinity response

ARTICLE IN PRESS

G. Kumar et al.518

propose that identification and availability of thesecontrasting cultivars within the Brassicaceae familywill be very useful for future studies targeting thegenome-level molecular investigations.

Acknowledgements

The work was supported by research grants to APand SLS-P from the Department of Biotechnology,Ministry of Science and Technology, Government ofIndia.

References

Amtmann A, Bohnert HJ, Bressan RA. Abiotic stress andplant genome evolution: search for new models. PlantPhysiol 2005;138:127–30.

Apse MP, Blumwald E. Engineering salt tolerance inplants. Curr Opin Biotechnol 2002;13:146–50.

Aroca R, Amodeo G, Fernadez-Illescas S, Herman EH,Chaumont F, Chrispeels MJ. The role of aquaporins andmembrane damage in chilling and hydrogen peroxideinduced changes in the hydraulic conductance ofmaize roots. Plant Physiol 2005;137:341–53.

Ashraf M. Effect of NaCl on growth, seed protein and oilcontents of some cultivars/lines of brown mustard(Brassica juncea (L.) Czern. and Coss.). Agrochimica1992;36:137–47.

Ashraf M, McNeilly T. Responses of four Brassica species tosodium chloride. Environ Exp Bot 1990;30:475–87.

Ashraf M, McNeilly T. Salinity tolerance in Brassicaoilseeds. Crit Rev Plant Sci 2004;23:157–74.

Ashraf M, Sharif R. Inter-cultivar variation for salt (NaCl)tolerance in a potential oil-seed crop Ethiopianmustard (Brassica carinata A.Br.). Arch Acker PflBoden 1997;42:129–36.

Ashraf M, McNeilly T, Nazir M. Comparative salt toleranceof amphidiploid and diploid Brassica species. Plant Sci2001;160:683–9.

Bajji M, Bertin P, Lutts S, Kinet JM. Evaluation of droughtresistance-related traits in durum wheat somaclonallines selected in vitro. Aust J Exp Agric 2004;44:27–35.

Bates LS, Waldren RP, Teare ID. Rapid determination offree proline for water stress studies. Plant Soil 1973;39:205–7.

Blumwald E, Aharon GS, Apse MP. Sodium transport inplant cells. Biochim Biophys Acta 2000;1465:140–51.

Bohnert HJ, Gong Q, Li P, Ma S. Unraveling abiotic stresstolerance mechanisms – getting genomics going. CurrOpin Plant Biol 2006;9:180–8.

Chandler SF, Thorpe TA. Proline accumulation and sodiumsulfate tolerance in callus cultures of Brassica napusL.cv. Westar. Plant Cell Rep 1987;6:176–9.

Chen Q, Zhang WH, Liu YL. Effect of NaCl, glutathioneand ascorbic acid on function of tonoplast vesicles

isolated from barley leaves. J Plant Physiol 1999;155:685–90.

Cushman JC, Michalowski CB, Bohnert HJ. Developmentalcontrol of Crassulacean acid metabolism inducibilityby salt stress in the common ice plant. Plant Physiol1990;94:1137–42.

Das S, Hussain A, Bock C, Keller WA, Georges F. Cloning ofBrassica napus phospholipase C2 (BnPLC2), phospha-tidylinositol 3-kinase (BnVPS34) and phosphatidylino-sitol synthase1 (BnPtdIns S1)-comparative analysis ofthe effect of abiotic stresses on the expression ofphosphatidylinositol signal transduction-related genesin B. napus. Planta 2005;220:777–84.

Delauney AJ, Verma DPS. Proline biosynthesis andosmoregulation in plants. Plant J 1993;4:215–23.

DeRocher EJ, Bohnert HJ. Development and environmentalstress employ different mechanisms in the expression ofa plant gene family. Plant Cell 1993;5:1611–25.

Drake FT, Drake AK. The agricultural potential ofEstuarine waters. J Mississippi Acad Sci 1998;43:152–6.

Epstein E, Norlyn JD, Rush DW, Kingsbury RW. Salineculture of crops: a genetic approach. Science1980;210:399–404.

Francois LE, Maas EV, Donovan TJ, Youngs VL. Effect ofsalinity on grain yield and quality, vegetable growth,and germination of semi-dwarf and durum wheat.Agron J 1986;78:1053–8.

Gaxiola RA, Li J, Undurraga S, Dang LM, Allen GJ, AlperSL, et al. Drought- and salt-tolerant plants result fromoverexpression of the AVP1 H+-pump. Proc Natl AcadSci USA 2001;98:11444–9.

Gong D, Guo Y, Schumaker KS, Zhu JK. The SOS3 family ofcalcium sensors and SOS2 family of protein kinases inArabidopsis. Plant Physiol 2004;134:919–26.

Gorham J, Hardy C, Wyn Jones RG, Joppa LR Law CN.Chromosomal location of a K+/Na+ discriminationcharacter in the D genome of wheat. Theor ApplGenet 1987;74:584–8.

Guo Y, Halfter U, Ishitani M, Zhu JK. Molecular char-acterization of functional domains in the proteinkinase SOS2 that is required for plant salt tolerance.Plant Cell 2001;13:1383–400.

Halfter U, Ishitani M, Zhu JK. The Arabidopsis SOS2protein kinase physically interacts with and is acti-vated by the calcium-binding protein SOS3. Proc NatlAcad Sci USA 2000;97:3735–40.

Hannah MA, Wiese D, Freund S, Fiehn O, Heyer AG,Hincha DK. Natural genetic variation of freezingtolerance in Arabidopsis. Plant Physiol 2006;142:98–112.

Hasegawa PM, Bressan RA, Zhu JK, Bohnert HJ. Plantcellular and molecular responses to high salinity. AnnuRev Plant Phys 2000;51:463–99.

Houle G, Morel L, Reynolds C, Siegel J. The effectof salinity on different developmental stages ofan endemic annual plant, Aster laurentianus(Asteraceae). Am J Bot 2001;88:62–7.

Hua XJ, Van De Cotte B, Van Montagu M, Verbruggen N.The 50 untranslated region of the At-P5R gene is

ARTICLE IN PRESS

Salinity response in Brassica and SOS pathway 519

involved in both transcriptional and post-transcrip-tional regulation. Plant J 2001;26:157–69.

Ishitani M, Liu J, Halfter U, Kim CS, Shi W, Zhu JK.SOS3 function in plant salt tolerance requiresN-myristoylation and calcium-binding. Plant Cell 2000;12:1667–77.

Kant S, Kant P, Raveh E, Barak S. Evidence thatdifferential gene expression between the halophyte,Thellungiella halophila, and Arabidopsis thaliana isresponsible for higher levels of the compatibleosmolyte proline and tight control of Na+ uptake inT. halophila. Plant Cell Environ 2006;29:1220–34.

Kavi Kishor PB, Sangam S, Amrutha RN, Sri Laxmi P, NaiduKR, Rao KRSS, et al. Regulation of proline biosynthesis,degradation, uptake and transport in higher plants: itsimplications in plant growth and abiotic stresstolerance. Curr Sci 2005;88:424–38.

Kumari S, Sabharwal VP, Kushwaha HR, Sopory SK, Singla-Pareek SL, Pareek A. Transcriptome map for seedlingstage specific salinity stress response indicates aspecific set of genes as candidate for saline tolerancein Oryza sativa L. Funct Int Genomics 2008,doi:10.1007/s10142-008-0088-5.

Lee S, Kim J, Son JS, Nam J, Jeung DH, Lee K, et al.Systematic reverse genetic screening of t-DNA taggedgenes in rice for functional genomic analysis: MADS-box genes as a test case. Plant Cell Physiol 2003;44:1403–11.

Li Z, Zhao L, Kai G, Yu S, Cao Y, Pang Y, et al. Cloning andexpression analysis of a water stress-induced genefrom Brassica oleracea. Plant Physiol Biochem2004;42:789–94.

Lindsay MP, Lagudah ES, Hare RA, Munns R. A locus forsodium exclusion (Nax1) a trait for salt tolerance,mapped in durum wheat. Funct Plant Biol 2004;31:1105–14.

Liu J, Zhu JK. Proline accumulation and salt-stress-induced gene expression in a salt-hypersensitivemutant of Arabidopsis. Plant Physiol 1997;114:591–6.

Liu J, Ishitani M, Halfter U, Kim CS, Zhu JK. TheArabidopsis thaliana SOS2 gene encodes a proteinkinase that is required for salt tolerance. Proc NatlAcad Sci USA 2000;97:3730–4.

Love CG, Batley J, Lim G, Robinson AJ, Savage D, Singh D,et al. New computational tools for Brassica genomeresearch. Comp Funct Genomics 2004;5:276–80.

Martınez-Atienza J, Jiang X, Garciadeblas B, Mandoza I,Zhu JK, Pardo JM, et al. Conservation of the SOS salttolerance pathway in rice. Plant Physiol 2006;143:1001–12.

Morinaga T. Interspecific hybridization in Brassica. VI. Thecytology of F1 hybrids of Brassica juncea and B. nigra.Cytologia 1934;6:62–7.

Munns R, James RA. Screening methods for salt toler-ance: a case study with tetraploid wheat. Plant Soil2003;253:201–18.

Qiu QS, Guo Y, Dietrich MA, Schumaker KS, Zhu JK.Regulation of SOS1, a plasma membrane Na+/H+

exchanger in Arabidopsis thaliana, by SOS2 andSOS3. Proc Natl Acad Sci USA 2002;99:8436–41.

Quesada V, Garcıa-Martınez S, Piqueras P, Ponce MR,Micol JL. Genetic architecture of NaCl tolerance inArabidopsis thaliana. Plant Physiol 2002;130:951–63.

Quintero FJ, Ohta M, Shi H, Zhu J-K, Pardo JM.Reconstitution in yeast of the Arabidopsis SOS signal-ing pathway for Na+ homeostasis. Proc Natl Acad SciUSA 2002;99:9061–6.

Rawson HM, Richards RA, Munns R. An examination ofselection criteria for salt tolerance in wheat, barleyand triticale. Aust J Exp Agric 1988;39:759–72.

Rhee SY, Beavis W, Berardini TZ, Chen G, Dixon D, Doyle A,et al. The Arabidopsis Information Resource (TAIR):a model organism database providing a centralized,curated gateway to Arabidopsis biology, researchmaterials and community. Nucleic Acids Res 2003;31:224–8.

Sanders D. Plant biology: the salty tale of Arabidopsis.Curr Biol 2000;10:R486–8.

Shabala L, Cuin TA, Newman IA, Shabala S. Salinity-induced ion flux patterns from the excised roots ofArabidopsis sos mutants. Planta 2005;222:1041–50.

Shannon MC, Grieve CM. Tolerance of vegetable crops tosalinity. Sci Hortic 1999;78:5–38.

Shi H, Wu SJ, Zhu JK. Overexpression of a plasmamembrane Na+/H+ antiporter improves salt tolerancein Arabidopsis. Nat Biotechnol 2003;21:81–5.

Shi HZ, Zhu JK. Regulation of expression of the vacuolarNa+/H+ antiporter gene AtNHX1 by salt stress and ABA.Plant Mol Biol 2002;50:543–50.

Shi HZ, Ishitani M, Kim CS, Zhu JK. The Arabidopsisthaliana salt tolerance gene SOS1 encodes a putativeNa+/H+ antiporter. Proc Natl Acad Sci USA 2000;97:6896–901.

Singla-Pareek SL, Reddy MK, Sopory SK. Genetic en-gineering of the glyoxalase pathway in tobacco leadsto enhanced salinity tolerance. Proc Natl Acad Sci USA2003;100:14672–7.

Sreenivasulu N, Grimm B, Wobus U, Weshke W. Differ-ential response of antioxidant compounds to salinitystress in salt-tolerant and salt-sensitive seedlings offoxtail millet (Setaria italica). Plant Physiol 2000;109:435–42.

Suga S, Imagawa S, Maeshima M. Specificity of theaccumulation of mRNAs and proteins of the plasmaand tonoplast aquaporins in radish organs. Planta2001;212:294–304.

Tester M, Davenport RJ. Na+ transport and Na+ tolerancein higher plants. Ann Bot (London) 2003;91:503–27.

UN. Genome analysis of Brassica with special referenceto the experimental formation of Brassica napus andpeculiar mode of fertilization. Jpn J Bot 1935;7:389–452.

Vinocur B, Altman A. Recent advances in engineeringplant tolerance to abiotic stress: achievements andlimitations. Curr Opin Biotechol 2005;16:123–32.

Yeo AR. Molecular biology of salt tolerance in the con-text of whole-plant physiology. J Exp Bot 1998;49:915–29.

Zhu JK. Genetic analysis of plant salt tolerance usingArabidopsis thaliana. Plant Physiol 2000;124:941–8.

ARTICLE IN PRESS

G. Kumar et al.520

Zhu JK. Plant salt tolerance. Trends Plant Sci 2001a;6:66–71.

Zhu JK. Cell signaling under salt, water and cold stresses.Curr Opin Plant Biol 2001b;4:401–6.

Zhu JK. Salt and drought stress signal transduction inplants. Annu Rev Plant Phys 2002;53:247–73.

Zhu JK. Regulation of ion homeostasis under salt stress.Curr Opin Plant Biol 2003;6:441–5.