annurev.arplant.59.032607

8/6/2019 annurev.arplant.59.032607

1/34

Mechanisms of Salinity ToleranceRana Munns1 and Mark Tester 21CSIRO Plant Industry, Canberra, ACT, Australia; email: [email protected] Australian Center for Plant Functional Genomics and University of Adelaide,SA, Australia; email: [email protected]

Annu. Rev. Plant Biol. 2008. 59:65181

The Annual Review of Plant Biologyis online atplant.annualreviews.org

This articles doi:10.1146/annurev.arplant.59.032607.092911

Copyright c 2008 by Annual Reviews. All rights reserved

1543-5008/08/0602-0651$20.00

Key Wordssalt tolerance, salinity stress, sodium toxicity, chloride, stresstolerance

Abstract The physiological and molecular mechanisms of tolerance to motic and ionic components of salinity stress are reviewed at the clular, organ, andwhole-plant level. Plant growth responds to salinin two phases: a rapid, osmotic phase that inhibits growth of yoleaves, and a slower, ionic phase that accelerates senescence ofture leaves. Plant adaptations to salinity are of three distinct tyosmotic stress tolerance, Na+ or Cl exclusion, and the tolerancetissue to accumulated Na+ or Cl . Our understanding of the role the HKT gene family in Na+ exclusion from leaves is increasing, athe understanding of the molecular bases for many other transpprocesses at the cellular level. However, we have a limited molecunderstanding of the overall control of Na+ accumulation and osmotic stress tolerance at the whole-plant level. Molecular genics and functional genomics provide a new opportunity to syntsize molecular and physiological knowledge to improve the salitolerance of plants relevant to food production and environmensustainability.

651

byWB6065TBUBHnnov

on

0Fop

onu

ony


2/34

ContentsINTRODUCTION... . . . . . . . . . . . . . . 652

Aim of This Review . . . . . . . . . . . . . . 653 THE BASES FOR PLANT

VARIATION IN TOLERANCE . . . . . . . . . . . . . . . . . . 653

Plants Vary in Tolerance . . . . . . . . . . 653Plant Responses Can Occur

in Two Distinct Phases Through Time . . . . . . . . . . . . . . . . 654

Three Distinct Types of PlantResponse or Tolerance. . . . . . . . . 656

Relative Importance of the Three Tolerance Mechanisms . . . . . . . . 657

Methods to Distinguish the Three Tolerance Mechanisms . . . . . . . . 657

OSMOTIC STRESS

TOLERANCE . . . . . . . . . . . . . . . . . . 658Growth . . . . . . . . . . . . . . . . . . . . . . . . . . 658Photosynthesis and Stomatal

Conductance. . . . . . . . . . . . . . . . . . 659Oxidative Stress . . . . . . . . . . . . . . . . . . 660Cellular Signaling . . . . . . . . . . . . . . . . 661

ACCUMULATION OF SODIUMIONS IN SHOOTS. . . . . . . . . . . . . . 662

Thermodynamics of Na+ Transport . . . . . . . . . . . . . . . . 663

Net Na + Inux Into the OuterHalf of Roots . . . . . . . . . . . . . . . . . 66

Na+ Loading Into and Retrieval

From the Xylem. . . . . . . . . . . . . . . 665 TISSUE TOLERANCE OF

SODIUM IONS . . . . . . . . . . . . . . . . . 666Intracellular Compartmentation

of Na+ . . . . . . . . . . . . . . . . . . . . . . . . 6 6 7Increased Accumulation

of Compatible Solutes . . . . . . . . . 66OBSERVATIONS IN WHICH

SALINITY TOLERANCE ISCLEARLY INDEPENDENTOF TISSUE SODIUM ION

CONCENTRATIONS . . . . . . . . . . 669 Mechanisms of Salinity Tolerance

Other than Na + Exclusion . . . . . 670 APPROACHES FOR FUTURE

STUDIES . . . . . . . . . . . . . . . . . . . . . . . 672 The Importance of Cell

TypeSpecic Processes . . . . . . . 67Relevant Growth Conditions and

Salinity Treatments. . . . . . . . . . . . 672

Stress: an adversecircumstance thatdisturbs, or is likely to disturb, thenormal physiologicalfunctioning of anindividual

INTRODUCTION Soil salinity stresses plants in two ways. Highconcentrations of salts in the soil make itharder for roots to extract water, and highconcentrations of salts within the plant can betoxic.Salts on the outside of roots have an im-mediate effect on cell growth and associatedmetabolism; toxic concentrations of salts taketime to accumulate inside plants before they affect plant function. We discuss the physi-ology and molecular biology of mechanismsthat allow plants to adapt to these stresses.

More than 800 million hectares of landthroughout the world are salt affected (31). This amount accounts for more than 6% of the worlds total land area. Most of this salt-affected land has arisen from natural causes,

from the accumulation of salts over long priods of time in arid and semiarid zones (10 Weathering of parental rocks releases soble salts of various types, mainly chloridesodium, calcium, and magnesium, and tolesser extent, sulfates and carbonates (12Sodiumchloride is themost solubleandabudant salt released. The other cause of accmulation is the deposition of oceanic sacarried in wind and rain. Rainwater conta

650 mg/kg of sodium chloride; the concetration decreaseswith distance from thecoaRain containing 10 mg/kg of sodium chride would deposit 10 kg/ha of salt for e100 mm of rainfall per year.

Apart from natural salinity, a signicproportion of recently cultivated agricultu

652 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


3/34

land has become saline owing to land clearingor irrigation, both of which cause water tablesto rise and concentrate the salts in the rootzone. Of the 1500 million ha of land farmedby dryland agriculture, 32 million ha (2%) areaffected by secondary salinity to varying de-grees. Of the current 230 million ha of irri-gated land, 45 million ha (20%) are salt af-fected (31). Irrigated land accounts for only 15% of total cultivated land, but because irri-gated land has at least twice the productivity of rainfed land, it produces one third of the worlds food.

Salinity is a soil condition characterized by a high concentration of soluble salts. Soils areclassied as saline when the ECe is 4 dS/mor more (131), which is equivalent to approxi-mately40mMNaClandgeneratesanosmoticpressure of approximately 0.2 MPa. This def-inition of salinity derives from the ECe thatsignicantly reduces the yield of most crops.

Because NaCl is the most soluble and widespread salt, it is not surprising that allplants have evolved mechanisms to regulateits accumulation and to select against it in fa- vor of other nutrients commonly present inlow concentrations, such as K + and NO 3 . Inmost plants, Na+ and Cl are effectively ex-cluded by roots while water is taken up fromthe soil (89). Halophytes, the natural ora of highly saline soils, are able to maintain this ex-clusion at higher salinities than glycophytes.For example, sea barleygrass,Hordeum mar-inum, excludes both Na+ and Cl until at least450 mM NaCl (44). It is also not surprisingthat because salinity is a common feature of arid and semiarid lands, plants have evolvedmechanisms to tolerate the low soil water po-tential causedbysalinity, aswell asbydrought,and so tolerance to osmotic stress is a featureof most glycophytes and halophytes.

Former reviews in this series on plantresponses to salinity were published either20 or more years ago (35, 53, 104) or muchmore recently (58, 145). The 20-year gap andthe recent revival in activity is indicative of the breakthroughs now emerging owing tothe application of molecular genetics to in-

ECe: the electricaconductivity of thesaturated pasteextract; equivalentthe concentration osalts in saturated sor in a hydroponicsolution

crease our understanding of the physiologi-cal and molecular mechanisms of salinity tol-erance in plants. This recent urry of activ-ity may also reect the current excitement inplant science for making practical contribu-tions to food production in the face of in-creasing salinization of agricultural regionsand global climate change (75).

Aim of This Review The focus of this review is mechanisms of salinity tolerance at the molecular, cellular,and whole plant levels. The aim is to providea fundamental biological understanding andknowledge to underpin future applications. The great opportunity for salinity toleranceresearch now is the ability to marry togethernewmoleculartechniqueswiththebodyoflit-erature on whole plant physiology. This newopportunityin salinity tolerance researchpro- vides exciting prospects for ameliorating theimpactof salinity stressonplants, andimprov-ing the performance of species important tohuman health and agricultural and environ-mental sustainability.

Ultimately, plant function is explained by the operation of genes in cells and tissues toregulate plant growth in coordination withenvironmental constraints. As such, gene andcell function must always be considered inthe context of the whole plant. This is espe-ciallyso in thecase of salinity tolerance,wherecell-specicprocesses areof particular impor-tance. A salt-tolerant cell does not necessarily make a salt-tolerant plant.

THE BASES FOR PLANT VARIATION IN TOLERANCE

Plants Vary in TolerancePlants differ greatly in their tolerance of salinity, as reected in their different growthresponses. Of the cereals, rice (Oryza sativa)is the most sensitive and barley ( Hordeumvulgare) is the most tolerant (Figure 1 ).Bread wheat (Triticum aestivum) is moderately

www.annualreviews.org Mechanisms of Salinity Tolerance 653

byWB6065TBUBHnnov

on

0Fop

onu

ony


4/34

120

100

80

60

40

20

0

NaCl (mM)

S a

l i n

i t y

t o l e r a n c e

( i n c r e a s e

i n

d r y m a

t t e r a s

% c

o n

t r o

l )

3000 100 200 400 500 600

Saltbushaltbush

Durumurumwheatheat

Breadreadwheatheat

Barleyarley

Alfalfalfalfa

Talall wheatgrassheatgrass

ArabidopsisrabidopsisRiceice

Saltbush

Durumwheat

Breadwheat

Barley

Alfalfa

Tall wheatgrass

ArabidopsisRice

Figure 1Diversity in the salt tolerance of various species, shown as increases inshoot dry matter after growth in solution or sand culture containing NaCl

for at least 3 weeks, relative to plant growth in the absence of NaCl. Dataare for rice (Oryza sativa) (6), durum wheat (Triticum turgidumssp durum)(19), bread wheat (Triticum aestivum) (19), barley ( Hordeum vulgare) (19),tall wheatgrass (Thinopyrum ponticum, syn. Agropyron elongatum) (19), Arabidopsis ( Arabidopsis thaliana) (21), alfalfa ( Medicago sativa) (70), andsaltbush ( Atriplex amnicola) (7).

Osmotic stress:affects growthimmediately and iscaused by the saltoutside the roots

tolerant and durum wheat (Triticum turgidumssp.durum)islessso.Tallwheatgrass(Thinopy-rum ponticum, syn. Agropyron elongatum) is ahalophytic relative of wheat and is one of the most tolerant of the monocotyledonousspecies(Figure1 );itsgrowthproceedsatcon-centrations of salt as high as in seawater.

The variation in salinity tolerance in di-cotyledonous species is even greater than inmonocotyledonousspecies.Somelegumes are very sensitive, even more sensitive than rice(74); alfalfa or lucerne ( Medicago sativa) is very tolerant, and halophytes such as saltbush( Atriplexspp.) continue to grow well at salin-ities greater than that of seawater (Figure 1 ). Many dicotyledonous halophytes require a

Table 1 The effects of salinity stress on plants

Effect of stress Osmotic stressStress due to high leaf

Na + (ionic stress)Speed of onset Rapid SlowPrimary site of visible effect

Decreased new shootgrowth

Increased senescenceof older leaves

quite high concentration of NaCl (10200 mM) for optimum growth (35). Arbidopsis , when compared with other speciunder similar conditions of light and humiity (that is, at high transpiration rates), isalt-sensitive species (Figure 1 ). This senstive plant may provide limited insights imechanisms of salinity tolerance unless icompared with a tolerant relative such asThelungiella halophila. The differences betwethese two species are highlighted by their sponses to 100 mM NaCl under conditioof high transpiration. Continued exposure 100 mM does not allowArabidopsis to complete its life cycle (116), but has little effecthe growth rate of Thellungiella(69).

Plant Responses Can Occur in TwoDistinct Phases Through Time To understand the physiological mechanismresponsible for the salinity tolerance of thespecies, it is necessary to know whether thgrowth is being limited by the osmotic effof the salt in the soil, or the toxic effect ofsalt withinthe plant. In thesimplest analysisthe responseofa plant to salinitystress, the rduction in shoot growth occurs in two phasea rapid response to the increase in external omotic pressure, and a slower response duethe accumulation of Na+ in leaves ( Table 1

In the rst, osmotic phase,which starts immediately after the salt concentration aroutherootsincreasestoathresholdlevel,theraof shootgrowthfallssignicantly. Thethresold level is approximately 40 mM NaCl most plants (see denition of salinity abovor less for sensitive plants like rice andArbidopsis . This is largely (but not entirely) dto the osmotic effect of the salt outside troots. Figure 2 a shows the effect on the raof shoot growth, that is, the rate of increain shoot dry matter or in leaf area over tim The rate at which growing leaves expanreduced, new leaves emerge more slowly, alateral buds develop more slowly or remquiescent, so fewer branches or lateral shoform.

654 Munns Tester

byWB6065

TBUBHnnov

on

0Fop

onu

ony


5/34

In cereals, the major effect of salinity ontotal leaf area is a reduction in the numberof tillers; in dicotyledonous species, the majoreffect is the dramatic curtailing of the size of individual leaves or the numbers of branches.Curiously, shoot growthis more sensitive thanroot growth, a phenomenon that also occursin drying soils and for which there is as yetno mechanistic explanation (see the followingsection). The teleological explanation is thata reduction in leaf area development relativeto root growth would decrease the water useby the plant, thus allowing it to conserve soilmoisture and prevent an escalation in the saltconcentration in the soil.

The second, ion-specic, phase of plantresponse to salinity starts when salt accumu-lates to toxic concentrations in the old leaves(which are no longer expanding and so nolonger diluting the salt arriving in them as younger growing leaves do), and they die. If the rate at which they die is greater than therate at which new leaves are produced, thephotosynthetic capacity of the plant will nolonger be able to supply the carbohydrate re-

Ionic stress:develops over timeand is due to acombination of ionaccumulation in thshoot and an

inability to toleratethe ions that haveaccumulated

quirement of the young leaves, which furtherreduces their growth rate (Figure 2 a).

The osmotic stress not only has an imme-diate effect on growth, but also has a greatereffect on growth rates than the ionic stress.Ionic stress impacts on growth much later,and with less effect than the osmotic stress,especially at low to moderate salinity levels(Figure 2 a). Only at high salinity levels, or insensitive species that lack theability to controlNa+ transport, does the ionic effect dominatetheosmotic effect. The effect of increased tol-erance to the osmotic stress, with no changein ionic stress tolerance, is shown by the dot-ted line in Figure 2 a. A signicant genetic variation within species may exist in the os-motic response, but this has not yet been doc-umented. An increase in ionic tolerance takeslonger to appear (Figure 2 b). Within many species, documented genetic variation existsin the rate of accumulation of Na+ and Cl inleaves, as well as in the degree to which theseions can be tolerated. An increase in toleranceto both stresses would enable a plant to growat a reasonably rapid rate throughout its life

Shootgrowthrate

Time (weeks)

+NaCl

Osmoticphase

Ionicphase

Osmoticphase

Ionicphase

Osmoticphase

Ionicphase

lCaN+lCaN+

c Increase in bothb Increase in ionictolerance

a Increase in osmotictolerance

Figure 2 The growth response to salinity stress occurs in two phases: a rapid response to the increase in externalosmotic pressure (the osmotic phase), and a slower response due to the accumulation of Na+ in leaves(the ionic phase). The solid green line represents the change in the growth rate after the addition of NaCl. (a) The broken green line represents the hypothetical response of a plant with an increasedtolerance to the osmotic component of salinity stress. (b) The broken red line represents the response of aplant with an increased tolerance to the ionic component of salinity stress (based on Reference 93).(c ) The green-and-red line represents the response of a plant with increased tolerance to both theosmotic and ionic components of salinity stress.


byWB6065TBUBHnnov

on

0Fop

onu

ony


6/34

cycle. This combined tolerance is shown inFigure 2 c .

For most species, Na+ appears to reach atoxic concentration before Cl does, and somost studieshave concentratedon Na+ exclu-sion and the control of Na+ transport withinthe plant. However for some species, such assoybean, citrus, and grapevine, Cl is consid-ered to be the more toxic ion (74, 119). Theevidence for this is the association betweengenetic differences in the rate of Cl accumu-lation in leaves and the plants salinity toler-ance. This difference may arise because Na+

is withheld so effectively in the woody rootsand stems that little reaches the leaves,and K +

becomes the major cation. Thus Cl , whichcontinues to pass to the lamina, becomes themore signicant toxic componentof thesalinesolution.

Three Distinct Types of Plant Response or Tolerance The mechanisms of salinity tolerance fall ithree categories ( Table 2 ):

1. Tolerance to osmotic stress. The omotic stress immediately reduces cexpansion in root tips and young leavand causes stomatal closure. A reduresponse to the osmoticstress would rsult in greater leaf growth and stoatal conductance, but the resulting icreased leaf area would benet oplants that have sufcient soil watGreater leaf area expansion would productive when a supply of water is esured such as in irrigated food prodution systems, but could be undesirabin water-limited systems, and cause

Table 2 Mechanisms of salinity tolerance, organized by plant processes and their relevance to the three componentsof salinity tolerance

Osmotic stress Ionic stress

Process involvedCandidate

genes a Osmotic tolerance Na + exclusion Tissue toleranceSensing and signaling inroots

SOS3, SnRKs Modication of long-distancesignaling

Control of net iontransport to shoot

Control of vacuolarloading

Shoot growth ? Decreasedinhibition of cellexpansion andlateral buddevelopment

Not applicableb Delay in prematuresenescence of old(carbon source) leav

Photosynthesis ERA1, PP2C , AAPK , PKS3

Decreased stomatalclosure

Avoidance of ion toxicity inchloroplasts

Delay in ion toxicity chloroplasts

Accumulation of Na+ inshoots

HKT , SOS1 Increased osmoticadjustment

Reduced long distancetransport of Na+

Reduced energy spenon Na+ exclusion

Accumulation of Na+ in vacuoles

NHX , AVP Increased osmoticadjustment

Increased sequestration of Na+ into root vacuoles

Increasedsequestration of Na+

into leaf vacuoles Accumulation of organic

solutes P5CS , OTS , MT1D, M6PR,S6PDH , IMT1

Increased osmoticadjustment

Alteration of transportprocesses to reduce Na+

accumulation

Accumulation of highconcentrations of compatible solutes icytoplasm

a This list is not comprehensive, please see reviews such as Bartels & Sunkar (8), Munns (89), and Zhu (145), as well as the Clickable Guard Cavailable athttp://www-biology.ucsd.edu/labs/schroeder/clickablegc.htmlbIons do not accumulate to toxic levels in growing tissues.

656 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


7/34

soil water to be used up before the grainis fully matured.1

2. Na+ exclusionfrom leaf blades. Na+ ex-clusion by roots ensures that Na doesnot accumulate to toxic concentrations within leaves. A failure in Na+ exclu-sion manifests its toxic effect after daysor weeks, depending on the species, andcauses premature death of older leaves.

3. Tissue tolerance, i.e., tolerance of tis-sue to accumulated Na+ , or in somespecies,toCl . Tolerance requires com-partmentalizationof Na + andCl atthecellular and intracellular level to avoidtoxic concentrations within the cyto-plasm, especially in mesophyll cells inthe leaf. Toxicity occurs with time, afterleaf Na+ increases to high concentra-tions in the older leaves.

Table 2 summarizes some of the mechanismsrelevant to the three components of salin-ity tolerance, classied by various plant pro-cesses.

Relative Importance of the Three Tolerance Mechanisms The relative importance of these various pro-cesses clearly varies with the species (i.e., the

strategy a particular plant species has evolvedforresponding to thesalinity stress), butprob-ably also varies with the length of exposureto the salinity, the concentration of the salt,and possibly the local environmental condi-tions, notably soil water supplyandairhumid-ity, and thus transpiration rate and leaf waterpotential.

For example, in some conditions a highshoot Na+ may be benecial by helping the

1 The focus of this review is on tolerance in agricul-tural systems, where growth and productivity of annualcrops is more important than survival per se. Thus, tol-erance to osmotic stress is considered in this review tobe the ability to maintain growth. However, in an eco-logical context, especially for perennial species, survivalis often more important than growth, so the emphasis ongrowth maintenance as an adaptive (benecial) response islesspronounced.

plant maintain turgor. This may become par-ticularly important in drying soils, whereaccess by the plant to other benecial nutri-ents (such as N, P, and K) becomes increas-ingly difcult. A balance probably needs to bestruck between the use of Na+ and Cl by theplant to maintain turgor and the need to avoidchemical toxicity. Where that balance lieswilldepend on the species and conditions. Thisdilemma has been likened to that of Ulysses who hadto steer a course through treacherous waters between the twin perils of Scylla andCharybdis (53).

Methods to Distinguish the Three Tolerance Mechanisms The two-phase effects of salinity on plants arenot obvious if the salinity is high, or if thespecies is particularly sensitive to Na+ . Theroots of some species, such as rice, are leaky and Na+ may be taken up apoplastically (48). Then, the rst phase, or osmoticeffect, mightlastonlyafewhoursordaysatthemostbeforethe Na + levels build up to toxic levels withinthe leaves (142). However, for most plants inmost conditions, the two phases are clearly separated in time (93), which facilitates theexperimental separation of the three tolerancemechanisms.

Distinguishing theosmotic effect from theion-specic effect requires observations overtimeof the rate of new leafproduction and therate of increase in injuryof old leaves. The ef-fectoftheosmoticstressisseenasarapidinhi-bition of the rate of expansion of young leavesand reduced stomatal conductance of matureleaves. Daily measurements of the length of agrowing leaf, or spot measurements of stom-atal conductance with a porometer, are goodindicators of growth rate.

Ion-specic toxicity is seen as an increasein the rate of senescence of older leaves, dueto either high leaf Na+ concentrations or tolow tolerance of the accumulated Na+ . Leaf Na+ concentration is best measured in a de-ned leaf of a dened age if the plant wasexposed to Na+ at around the time of the


byWB6065TBUBHnnov

on

0Fop

onu

ony


8/34

emergence of that leaf (91, 141). Leaf senes-cence can be measured nondestructively witha chlorophyll meter or image analysis. Com-bining ratesof senescence in older leaves withmeasures of leaf Na+ concentration providesan estimate of tolerance to Na+ that has ac-cumulated (tissue tolerance). The use of non-destructive assays, exploiting image analysis,thermography, and hyperspectral reectancetechniques, greatly facilitates the separationof these different types of Na+ tolerance.

Increased osmotic tolerance and increasedtissue tolerance will both lead to an increasedabilityto maintaingrowthfora givenaccumu-lation of Na+ in the leaf tissue. However, they can be distinguished because of their differ-ential effects on younger versus older tissue.Increasedosmotic tolerancewill be mainlyev-ident by an increased ability to continue pro-duction of new leaves, whereas tissue toler-ance will be primarilyevidentby the increasedsurvival of older leaves ( Table 1 ).

Interestingly, the sos ( salt overly sensitive)mutants of Arabidopsis were identied froma screen based on the maintenance of rootgrowth in nontranspiring conditions, wherethe delivery of Na+ to the shoot in the tran-spiration stream would be low. The sos mutantscreenmightdetectmutants that arerelated tothe osmotic component of salinity stress be-cause in the nontranspiring conditions usedfor the initial screen, the primary effect of salinity would be osmotic. In nontranspiringconditions, salinity tolerance inArabidopsis isunrelated to the extent of shoot Na+ accu-mulation; however, in transpiring conditionssalinity tolerance is related to the extent of shoot Na+ accumulation (86).

In the following three sections, each of the three tolerance mechanisms is discussedin more detail.

OSMOTIC STRESS TOLERANCE

Growth

The decreased rate of leaf growth after an in-crease in soil salinity is primarily due to the

osmotic effect of the salt around the roo A sudden increase in soil salinity causes cells to lose water, but this loss of cell voluand turgor is transient. Within hours, cells regain their original volume and turgor owito osmotic adjustment, but despite this, celongation rates are reduced (21, 42, 97, 14Over days, reductions in cell elongation aalso celldivision lead toslower leaf appearaandsmaller nal size.Cell dimensions chang with more reduction in area than depth, leaves are smaller and thicker.

Fora moderate salinity stress, an inhibitioof lateral shoot development becomes appent over weeks, and over months there aeffects on reproductive development, suchearlyoweringorareducednumberoforetDuring this time, a number of older leavmay die. However, production of youngleaves continues. All these changes in plgrowth are responses to the osmotic effectthe salt, and are similar to drought respons

The reduction in leaf development is dto the salt outside the roots. That this rduction is largely due to the osmotic effof the salt is supported by experiments usmixed salts such as concentrated Hoaglansolution (125), other single salts such as K(142), and nonionic solutes such as mnitol or polyethylene glycol (PEG) (1142). These different osmotica all havesimilar qualitative effect as NaCl on lexpansion.

However, the salt outside the roots maffect plant growth not only through its fect on osmotic pressure. S umer and coworers (121) found evidence for Na+ but nCl toxicity during the rst phase of sstress in maize in innovativeexperiments wdifferent salts and PEG, via the use of ditional PEG to adjust the equimolar soltions to equivalent osmotic pressures. Futher,Cramer(20)foundevidencefortheeffeof supplemental Ca2+ in the rooting soltion affecting rapid responses of leaf elontion rate from working with two maize cu vars of different salinity tolerance. A possNa+ -specic effect associatedwith thegrow

658 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


9/34

response is discussed below in the section onsignaling.

The mechanism that downregulates leaf growth and shoot development under stressis not precisely known. The reduction in leaf growthmustberegulatedbylongdistancesig-nals in the form of hormones or their precur-sors, because the reduced leaf growth rate isindependent of carbohydrate supply (90) and water status (42, 90). The reduction occursin the absence of nutrient deciency (61) andion toxicity, as evidenced by very low concen-trations of Na+ and Cl in expanding cellsor tissues that do not correlate with growthrates (38, 61, 62, 94). Changes in wall prop-erties must occur (22), but their exact natureremains unknown. The long distance and lo-calsignals regulating these wall properties andexpansion rates are still obscure.

Abscisic acid (ABA) plays a central rolein root-to-shoot and cellular signaling indrought stressand in the regulation of growthandstomatal conductance (26,145). However,measurements of ABA in growing zones of barley and maize leaves in saline soil do notsupport a simple ABA control theory. ABA concentrations in the growing zone of salt-treated barley increase transiently but returnto the original low value after 24 h, whereasleaf growth rate is still reduced (39). ABA-decient mutants in maize and tomato gen-erally have the same leaf growth rates as wild-type in drying soil and saline soil (80, 132),indicating that there is another limiting fac-tor. Gibberellins (GAs) are a good candidate. ABA can inhibit leaf elongation in barley by lowering the content of active GA, as indi-cated by exogenous treatments with ABA andGA and measurements of endogenous GAsin the elongating zone (P.M. Chandler, M. Maheswari & R. Munns, unpublished data). Accumulating evidence shows that membersof a class of negative regulators of growth,the DELLA proteins, mediate the growth-promoting effects of gibberellins in a num-ber of species, and integrate signals from arange of hormones and abiotic stress condi-tions, including salinity (2). DELLA proteins

may be the central coordinators that adaptplant growth to different environments (2).

Root growth is usually less affected thanleaf growth, and root elongation rate recov-ers remarkably well after exposure to NaCl orother osmotica (88). Recovery from a moder-ate stress of up to 0.4 MPa of mannitol, KCl,or NaCl (i.e., an osmotic shock that does notcause plasmolysis) is complete within an hour(37). Even so, recovery from NaCl concen-trations as high as 150 mM can occur within aday (88). In contrast to leaves, these recover-ies take place despite turgornot being fully re-stored(37).This indicatesdifferentchanges incell wall properties compared with leaves, butthe mechanism is unknown. With time, re-duced initiationof newseminalor lateralrootsprobably occurs, but little is known about this.

Photosynthesis and StomatalConductance The most dramatic and readily measurable whole plant response to salinity is a decreasein stomatal aperture. Stomatal responses areundoubtedly induced by the osmotic effectof the salt outside the roots. Salinity affectsstomatal conductance immediately, rstlyandtransiently owing to perturbedwater relationsand shortly afterward owing to the local syn-thesis of ABA (39). A short-lived increase in ABA is detected in the photosynthetic tissues within 10 minutes of the addition of 100 mMNaCl to barley (39, 40); the rapidity of theincrease suggesting in situ synthesis of ABA rather than transport from the roots. How-ever, a new reduced rate of transpiration sta-bilizes within hours (40) whileABA tissue lev-els return to control concentrations (39, 40). This stomatal response is probably regulatedby root signals in common with plants in adrying soil (26), as evidenced by stomatal clo-sure in salt-treated plants whose water statusis kept high by applying a balance pressure(126).

Ratesofphotosynthesisperunitleafareainsalt-treated plants are often unchanged, eventhough stomatal conductance is reduced (68).


byWB6065TBUBHnnov

on

0Fop

onu

ony


10/34

This paradox is explained by the changes incell anatomy described above that give rise tosmaller, thicker leaves and result in a higherchloroplast density per unit leaf area. Whenphotosynthesis is expressed on a unit chloro-phyll basis, rather than a leaf area basis, a re-duction due to salinity can usually be mea-sured. In any case, the reduction in leaf areadue to salinity means that photosynthesis perplant is always reduced.

Cause-effect relationships between photo-synthesis and growth rate can be difcult tountangle.Itisalwaysdifculttoknowwhethera reduced rate of photosynthesis is the causeof a growth reduction, or the result. With theonset of salinity stress, a reduced rate of pho-tosynthesis is certainly not the sole cause of a growth reduction because of the rapidity of the change in leaf expansion rates describedearlier (22, 39, 97), and also because of the in-crease in storedcarbohydrate,which indicatesunused assimilate (90). However, with time,feedback inhibition from sink to source may ne tune the rate of photosynthesis to matchthe reduced demand arising from growth in-hibition (98). Reduced leaf expansion result-ing in a buildup of unused photosynthate ingrowing tissuesmaygenerate feedback signalsto downregulate photosynthesis.

At high salinity, salts can build up in leavesto excessive levels. Exactly how the salts ex-ert their toxicity remains unknown. Salts may build up in the apoplast and dehydrate thecell, they may build up in the cytoplasmand inhibit enzymes involved in carbohydratemetabolism, or they may build up in thechloroplast and exert a direct toxic effect onphotosynthetic processes.

Oxidative Stress The reduced rate of photosynthesis increasesthe formation of reactive oxygen species(ROS), and increases the activity of enzymesthat detoxify these species (4, 36, 78). Whenplants acclimate to a changed environment,they undergo adjustments in leaf morphol-ogy, chloroplast pigment composition, and in

the activity of biochemical processes that p vent oxidative damage to photosystems. Ttwo processes that avoid photoinhibition oing to excess light are heat dissipation by xanthophyll pigments andelectron transferoxygen acceptors other than water. The later response necessitates the upregulationkeyenzymes for regulating ROSlevels suchsuperoxide dismutase, ascorbate peroxidacatalase, and the various peroxidases (4, Thecoordinatedactivityofthemultipleformof these enzymes in the different cell copartments achieves a balance between the raof formation and removal of ROS, and matains hydrogen peroxide (H2O2) at the levrequired for cell signaling.

All these ROSdetoxifying mechanisms apresent naturally in surfeit (4, 36, 78), aare woven into the regulatory regimes the chloroplast (78), to protect the photsystems from photoinhibition that might otherwise occur from the rapidly increasing liloads experienced by leaves under natura variable situations. If a plant has sufcientpacity to adjust to the instant, large changin light intensity as the sun emerges from bhinda cloud, it has more thanenoughcapacito adjust to the slower changes in the ratephotosynthesis induced by a saline soil. Tonly situation in which antioxidants appearbe insufcient is when an oxidative bursinduced. However, this does not occur undabiotic stress, but is conned to pathogen tack, when a massive rise in ROS triggers prgrammed cell death (4).

Therefore, genetic differences in salintolerance are not necessarily due to diffences in the ability to detoxify ROS. Mastudies have found differences in levels ofpression or activity of antioxidant enzymthese differences are sometimes associa with the more tolerant genotype, and somtimes with the more sensitive genotype. Wsuggest that differences in antioxidant actity between genotypes may be due to gentypic differences in degrees of stomatal csure or in other responses that alter the raof CO2 xation, differences that bring in

660 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


11/34

play the processes that avoid photoinhibi-tion and for which the plant has abundantcapacity. For such basic and important de-fense mechanisms, the biochemical pathwaysare complex, interactive, and have built-inredundancy. More than 150 genes make upthe complex ROS network inArabidopsis (84).Knowledge of the many possible functions of these genes, and the coordination, degree of redundancy, and cross talk between differentbranches of the ROS network, is still incom-plete (84). Doubt has been expressed that themanipulationof a singlegene related to oxida-tivestress tolerance canenhance the toleranceto any abiotic stress (78). Recently Arabidop- sis mutants lacking either or both a cytosolicandchloroplastic ascorbate peroxidase (H2O2removal enzymes) were found to be actually more tolerant of salinity stress (83), illustrat-ing theplasticityof ROSregulatory pathways,and the redundancy of pathways for ROS reg-ulation and protection.

Cellular Signaling Long-distance signaling of salinity stress tothe shoot from the roots, mediated at least inpart by ABA, is discussed above in the contextof the rapid inhibition of growth upon addi-tion of NaCl. Although this initial responseappears similar at the whole plant level withaddition of NaCl or isosmotic concentrationsof PEG or mannitol (see Growth, above),comparison of cytosolic Ca2+ responses insolutions with physiologically realistic ioniccomposition revealed that responses of rootsto addition of NaCl and sorbitol differ (129). Thus, cells in the roots initially must senseboth the ionic and osmoticcomponents of theaddition of Na+ and then respond rapidly tochanges in its external concentration. The re-sponses root cells need to make are necessary not only to maintain their own correct func-tion in the face of the new elevated externalNa+ , but also for them to signal to the shootthat shoot function must also be altered. Inthis section, we focus on signaling within rootcells,whichislikelytobeindependentofABA.

Plants respond directly and specically tothe addition of Na+ within seconds (73, 129), yet the mechanism by which plants sense theaddition of Na+ and the change in osmoticpressure remains obscure. The extracellularNa+ is either sensed at the plasma mem-brane, or if it is sensed intracellularly, then itmust rst cross the plasma membrane. Thus,a plasma membrane protein must either bethe sensor or be immediately upstream of thesensor. This gap in our knowledge is surpris-ing given the importance of this rst step inthe response by a plant to changes in its envi-ronment. A similar notable absence of knowl-edge exists about the molecular basis for tur-gor sensing.

The rst recorded response to an increasein Na+ around roots is an increase in cy-tosolic free Ca2+ ([Ca2+ ]cyt); the extracellu-lar addition of Na+ is apparently able toactivate the ux of Ca2+ into the cytosolacross the plasma membrane and also, in-terestingly, the tonoplast (7173, 87, 129). The changes in [Ca2+ ]cyt are complex, andare modulated by differences in extracellularcomposition, including Na+ concentration,providing opportunities for information to beencodedbythe[Ca2+ ]cyt changes (129).An ad-ditional level of complexity in NaCl-induced[Ca2+ ]cyt increases has been demonstrated by root cell typespecic expression of aequorinin Arabidopsis (71). In response to 220 mMNaCl, the increase in [Ca2+ ]cyt is lower inthe pericycle than in the other cell types(71).

The best-characterized signaling pathway specic to salinity stress likely involves theseincreases in [Ca2+ ]cyt (145). In this pathway,the Na + -induced increase in [Ca2+ ]cyt may besensedbya calcineurinB-likeprotein (CBL4),originally identied as SOS3. Although theafnity for Ca2+ binding of this protein is un-known, physiologically realistic increases incytosolic Ca2+ likely facilitate the dimeriza-tion of CBL4/SOS3and thesubsequent inter-action with a CBL-interacting protein kinase(CIPK24, originally identied as SOS2) (55). The CBL4/CIPK24 (SOS3/SOS2) complex


byWB6065TBUBHnnov

on

0Fop

onu

ony


12/34

is targeted to the plasma membrane via amyristoyl fatty acid chain covalently bound toCBL4/SOS3 (65), enabling the phosphoryla-tion and thus the activation of the membrane-bound Na + /H + antiporter, SOS1 (102, 103,115).

However, therole of SOS1 in plant salinity tolerance remains uncertain, because recon-ciliation of its pattern of expression with itsfunction remains incomplete. Measurementof the effects of SOS1 knockout on long-distance transport of Na+ is confounded be-causemostexperimentsareperformedinnon-transpiring conditions (86).

Althoughwhichaspect of salinity toleranceis contributed to by this pathway remains un-certain, this pathway is likely important forsome aspects of salinity tolerance, becausesos mutantsof Arabidopsis thalianaareless tolerantto salinity stress than wild-type plants (146).

Manyother components of signalingpath- ways have also been implicated in plant re-sponses to salinity, inferred by a range of ap-proaches such as transcriptomics and reversegenetics. These are reviewed extensively elsewhere (e.g., 18, 137, 145). However, invok-ing the adaptive relevance of a particular re-sponse to Na+ ina plant that is poorlyadaptedto salinity ( Arabidopsis ) is risky. These ap-proaches could be strengthenedbycomparingresponses in salt-tolerant and salt-sensitivelinesif the response is greater in the tolerantline, this suggests a role in the tolerance, butif the response is smaller, this may indicatethe response is not related to the toleranceper se, but is a downstream response to thestress.

Genetic approaches, such as the screen-ing of mutant populations of Arabidopsis foraltered salinity tolerance (115, 145) and theidentication of the genetic alteration caus-ing observed differences in tolerance (12, 64),are essential for identifying signicant genesfor tolerance. More work is necessary to dis-entangle the complexities of the myriad signaltransduction networks in plants. It is essentialthat these experiments areperformedin phys-iologically relevant conditions. Future work

may also be able to allow the identicatof the different processes that are relevantparticularaspects of salinity tolerance (assumarized in this review).

Signaling pathways identied in satolerant species (e.g., Thellungiella haloph(50, 133, 135) are more likely to deliver sults relevant to adaptive, ratherthan dysfuntional, responses to salinity, than those in tsalt-sensitive Arabidopsis unless, of courscreens of Arabidopsis are designed to identisalt-tolerant, rather than salt-sensitive, mtants. This is reected in two componentsionic stress toleranceion exclusion and sue tolerance.

Overall, cells respond to the perceived dference in extracellular Na+ with changin diverse aspects of functionfrom bchemistry and gene transcription to phyology, growth, and development. Transcrition factors and small RNAs are centralcontrolling the core aspects of the longeterm plant transcriptional responses, as r viewed in this series and elsewhere; readare referred to these detailed overviews (1139).

ACCUMULATION OF SODIUMIONS IN SHOOTS The main site of Na+ toxicity for most planis the leaf blade, where Na+ accumulates aftbeing deposited in the transpiration streamratherthan in theroots (88).A plant transpire50 times more water than it retains in leav(92), so excluding Na+ from the leaf bladesimportant, evenmoreso forperennial than foannual species, because the leaves of perenalsliveandtranspireforlonger.MostNa + this delivered to the shoot remains in the shobecauseformostplants,themovementofNafrom the shoot to the roots in the phloem clikely recirculate only a small proportiontheNa + that is delivered to the shoot. Assucthe processes determining Na+ accumulatioin the shoot are primarily the processes cotrolling the net delivery of Na+ into the ro xylem.

662 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


13/34

The net delivery of Na+ to the xylemcan be divided into four distinct components(127):

1. Inux into cells in the outer half of theroot;

2. Efux back out from these cells to thesoil solution;

3. Efux from cells in the inner half of theroot to the xylem; and4. Inux back into these cells from the

xylem before the transpiration streamdelivers the Na+ to the leaf blade.

Thermodynamics of Na + Transport The thermodynamics of each of these pro-cesses for Na+ are illustrated in Figure 3 a,and the likely molecular mechanisms are

shown in Figure 3 b. The thermodynamicanalysis assumes cytosolic Na+ concentra-tions of 30 mM and an electrical potential of 120 mM, but even if values differ by a factorof two, the principles remain unchanged.For example, at the xylem parenchyma,the efux of Na+ from the cells would beactive even if the xylem Na+ concentrations were nearly ten times lower than cytosolic

Figure 3 The thermodynamics and mechanisms of Na+ andCl transport at the soil-root and stelar cellxylem vessel interfaces in roots. Indicative cytosolic pH,ion concentrations, and voltages are derived fromthe literature (127, 134). (a) Longitudinal sectionof wheat root (provided by Dr. Michelle Watt). The cells between the endodermis and the xylem vessel are not labeled, but include pericycle cellsand xylem parenchyma (darker blue) as well asphloem parenchyma. The stele of dicotyledonousplants is more complex because it includes cambial vascular elements. The thermodynamics of ionmovements are indicated by the arrow colors: Active transport is shown as a red arrow, passivetransport is shown as a blue arrow. (b) Theproposed mechanisms of passive and active Na+

and Cl transport at the two interfaces, mediatedby ion channels and carriers (uniporters andH + -coupled antiporters and symporters). Abbreviations: SOS1, salt overly sensitive mutant1; HKT, high-afnity K + transporter.

Na+ concentrations (owing to the xylemparenchyma cytoplasm potential being60 mV negative of the potential in the xylemapoplast). With a xylem Na+ free concen-tration of 10 mM and a potential differencebetweenthexylemparenchymacellcytoplasmandxylem apoplast of 60 mV, active inux of Na+ into the xylem parenchyma cells would

Soioil solutionolution100 mM00 mM

0 mVmVpH 6H 6

Soil solution100 mM

0 mVpH 6

Root cell cytosol30 mM

120 mVpH 7.2

Xylem apoplast10 mM

60 mVpH 6

Naa +Na +Naa +Na +Naa +Na +

Na +

CllCl

CllCl Cll

Cl

a Thermodynamics of Na + and Cl transport

b Proposed mechanisms of Na+

and Cl

transport

Soil CortexEpidermis Endodermis Stele

Na +

Na +

Cl

Cl

H+

H+

2H +

Na +

Na +

Cl

2H +

Cl

Soilcellinterface

Nonselectivecation

channels

SOS1

Cellxylem apoplastinterface

Cl /2H +symporter

Anionchannel

SOS1

HKT

Anionchannel

Cl /2H +symporte


byWB6065

TBUBHnnov

on

0Fop

onu

ony


14/34

Net Na + inux: theresult of unidirectional inuxand unidirectionalefux; Na+ inux ispassive, as opposedto efux, whichrequires energy

Unidirectional Na +

inux into roots: very rapid, requireshigh rates of efux tocontrol net Na +

accumulation

only be necessary with cytoplasmic free Na+

concentrations greater than approximately 100 mM (which, with an activity coefcientof 0.7, is a total concentration of around 140mM). Another way to look at this is if thecytoplasmic free Na+ were 30 mM and themembrane potential difference were 60 mV,active inux would only be necessary with xylemapoplasticconcentrations below 3 mM.

Consideration of the thermodynamics of a Na+ /H + antiporter is simpler, because theelectroneutral exchange this antiporter cat-alyzes is unaffected by membrane potential. Thus, the direction of Na+ movement is de-termined simply by the differences in freeconcentrations of Na+ and H+ . A Na+ /H +

antiporter could only work in the opposite di-rection to that indicated (i.e., it could only pump Na+ into cells) if, for a pH difference of one unit (xylem more acidic), the xylem con-centration increased to 10 times that found inthe cytoplasm(i.e., to over 300 mMfor a cyto-plasmic Na+ concentration of 30 mM). Alter-natively, if the pH became more alkaline thanpH 7.7, then the Na + /H + antiporter couldpump Na+ into xylem parenchyma cells froma free concentration of 10 mM. These condi-tions would rarely, if ever, occur, and thus, the

Na + concentration(mM)

100

20

0

120

40

140

60

160

80

E p i d e

r m i s

S u b e

p i d e r m

i s

C o r t e

x 1

C o r t e

x 2

C o r t e

x 3

E n d o

d e r m

i s

P e r i c

y c l e

X y l e m

p a r e n

c h y m

a

X y l e m

v e s s

e l

STELECORTEX

Figure 4Differences in vacuolar concentrations of Na+ across roots of transpiring wheat plants growing in 150 mM NaCl. Concentrations were measuredby quantitative and calibrated X-ray microanalysis of snap-frozen sectionsusing a cryo-SEM (scanning electron microscope) method on root tissues10 cm from the tip (A. L auchli, R.A. James, R. Munns, C.X. Huang & M. McCully, unpublished data).

Na+ /H + antiporter will mostly act to pumNa+ out of cells.

The various processes ofNa+ transportareach briey considered here, but the readerreferredtothemoreextensiveanalysisoftheprocesses in Tester & Davenport (127).

Net Na + Inux Into the Outer Half of RootsNa+ enters roots passively, via voltaindependent (or weakly voltage-dependenonselective cation channels (3,127) andposibly via other Na+ transporters such as sommembers of the high-afnity K + transporte(HKT) family (57, 76). High afnity Na+ iux is also mediated by some members ofHKT transporter family in low salt roots (60but this is repressed by moderate concenttions of Na+ and so is unlikely to be re vant to salinity tolerance. The identities of thgenes encoding nonselective cation channremain uncertain, although there are sevecandidates, including cyclic nucleotidegachannels and ionotropic glutamate receptolike channels (27).

The main site of Na+ entry in roots uncertain, although it seems intuitively likthat as water moves across the root ctex toward the stele, ions are removed frthis stream into cells, where they are thsequestered in the vacuoles of these ce This is supported by X-ray microanalysisroots from rapidly transpiring wheat plan(A. L auchli, R.A. James, R. Munns, CHuang, & M. McCully, unpublished data which shows that vacuolar Na+ and Cl cocentrations decrease across the cortex; vacular Na+ and Cl concentrations are highestthe epidermis and subepidermis and lowestthe endodermis (Figure 4 ).

Most of the Na+ that enters root cells the outer part of the root is likely pumpback out again via plasmamembrane Na+ /Hantiporters (127), a process that likely cosumessignicant energy, giventhelargeuxthat have been measured. The identities the genes encoding these Na+ efux protei

664 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


15/34

are uncertainin Arabidopsis , only two mem-bers of the SOS1 gene family exist,SOS1and a currently uncharacterized gene at locus At1g14660 (95). Plasma membrane Na+ /H +

antiporter activity has been demonstrated forthe Arabidopsis protein SOS1 (101); although,as discussed above, the levels of expression inthe outerhalf of the mature root are currently uncertain.

Given that active Na+ efux is required inall cells throughout the plant, it is likely thatother genes encoding Na+ /H + antiportersalso exist. Many efux proteins may be en-coded in Arabidopsis by the gene at locus At1g14660 (95), but members of other genefamilies, particularly the large family of CHX genes, may also be important (95). The pos-sibility of other mechanisms for Na+ ef-ux, such as primary pumping by Na+ -translocating ATPases, also needs to be keptin mind (14, 82).

Na+ remaining in the root can be se-questered in vacuoles or transported tothe shoot. Compartmentation in vacuoles isachieved by tonoplast Na+ /H + antiporterssuch as those belonging to the Na+ /H + ex-changer (NHX) family in Arabidopsis (95). There is passive leakage of Na+ back tothe cytosol from vacuoles (possibly via tono-plast nonselective cation channels), requir-ing constant resequestration of Na+ into vac-uoles. Constitutive overexpression of NHX1or the gene encoding the Arabidopsis thaliana vacuolar H+ -translocating pyrophosphatase(AVP1), which contributes to the electro-chemical potential difference for H+ , whichenergizes the pumping of Na+ into the vac-uole, increases both Na+ accumulation andNa+ tolerance in Arabidopsis , suggesting thatmore efcient sequestration may improve tis-sue tolerance, perhaps by reducing cytosolicNa+ concentrations (5, 45).

Na + Loading Into and RetrievalFrom the Xylem Na+ moves in the symplast across the endo-dermis, is released from stelar cells into the

stelar apoplast, and then moves to the xylemin the transpiration stream. The plasmamem-brane Na+ /H + antiporter, SOS1, is expressedin stelar cells and could be involved in the ef-ux of Na+ from stelar cells into the xylem.However, this statement needs to be recon-ciled with the observation that the knock-out of this gene causes elevated, not reduced,shoot Na+ levels. The effect of the knockouton Na+ efux in the outer half of the rootmay possibly be greater than the effect of theknockout on loading in the inner half of theroot.

In another attempt to reconcile the observations, SOS1 has also been implicated in re-trieval of Na+ from the xylem (115). How-ever, given the likelydifference inpHbetweenthe stelar cytosol and apoplast (Figure 3 ), thiselectroneutral exchange would only be possi-ble with a large (at least an approximately 50-fold) difference in Na+ activity (the apoplast with higher activity), which is extremely un-likely (see section above, Thermodynamics of Na+ Transport).

Increasing evidence exists for the role of some members of theHKT gene family in re-trieval of Na+ from the xylem. In the Ara-bidopsis root, AtHKT1;1 is involved in theretrieval of Na+ from the xylem before itreaches the shoot (25, 122). Good evidence isaccumulating for a similar function for mem-bers of the closely relatedHKT1;5 gene fam-ily in rice (106) and wheat (12, 24, 66). Thecandidate gene for the classic K + /Na + dis-crimination ( Kna1) locus on the long armof chromosome 4D, described more than 20 years ago by Gorham and colleagues (51)and mapped by Dubcovsky and coworkers(28) and Luo and coworkers (79), is likely an HKT1;5 gene (12). Kna1 was associated with a higher leaf K + /Na + ratio (mainly de-termined by the variation in Na+ concentra-tion), andwasattributed with providingbread wheat with its superior salinity tolerance overtetraploid wheats (51).

Furthermore, good evidence exists that aclosely related gene, TmHKT1;4-A2, is thecandidate gene for the Na+ exclusion ( Nax1)


byWB6065TBUBHnnov

on

0Fop

onu

ony


16/34

AtHKT1;1 , A CASE STUDY OF CONFUSION

The HKT family of proteins comprises a structurally diversegroup that separates naturally into two distinct subfamilies(99). This diversity led to early reports of apparently contra-dictory properties, because the same name used for the rst

two genes studied, HKT1, from wheat and Arabidopsis , sug-gested similar function. Members of the HKT family func-tion as Na+ /K + symporters and as Na+ -selective transportersof both high and low afnity. Subfamily 1 contains low afnity Na+ uniporters.

Different patterns of expressionwithin the plant also affectthe role of these transporters in net cation uptake to the shoot;expression of a protein that catalyzes inux in the outerhalf of the root (epidermis and cortex) increases inux into the plant,but an inuxer in the stele reduces net inux into the plant(Figure 3 ).

Althoughthe rst HKT gene identiedwas from subfamily 2 (114), this group is less well characterized than subfamily 1. The wheat TaHKT2;1 protein functions as a Na + /K + sym-porter when expressed in Xenopus oocytes (109), and down-regulation of expression in planta reduces root Na+ accumu-lation and improves growth in saline conditions (76). In rice,OsHKT2;1 catalyzeshigh afnity Na+ inux in low salt roots,conditionswhere Na+ inux is benecial (60). At higherexter-nalNa+ concentrations,OsHKT2;1 is rapidly downregulated,to reduce potentially toxic Na+ inux.

The most studied member of the HKT1 subfamily is in Arabidopsis , whichcontainsa single HKT homolog, AtHKT1;1. AtHKT1;1 functions as a Na+ -selective uniporter when ex-pressed inXenopus oocytes and yeast, but it also complementsan E. coli K + uptake decient mutant and increases its K +

accumulation, suggesting some role in K + transport (130). Athkt1;1 mutants are salt-sensitive compared with wild-

type and hyperaccumulate Na+ in the shoot but show re-duced accumulation of Na+ in the root (10, 81, 110). Severalhypotheses have been advanced concerning the function of AtHKT1;1 in Arabidopsis .

Because hkt1;1 mutations ameliorated the sos phenotypesand reduced whole seedling Na+ in the sos3background, Rusand coworkers (111) proposed that AtHKT1;1 is an inuxpathway for Na+ uptake into the root. However, Berthomieuand colleagues (10) and Essah and coworkers (30) showedthat hkt1;1 mutants do not have lower root Na+ inux andBerthomieu and coworkers proposed instead that AtHKT1;1

(Continued )

locus in durum wheat (64), which is associa with Na+ exclusion and a high leaf K + /Naratio. The protein encoded by this gene rtrieves Na+ from the xylem, and has actity in the leaf sheaths as well as in the r(66).

In Arabidopsis , the importance of retrievofNa+ from the xylem as a primary controllof shoot Na+ concentration and plant saliity tolerance is suggested by forward genestudies that have revealedAtHKT1;1 as a pmary determinant of these parameters (10, 4111). It is noteworthy that, to date, no othgenes have been revealed from forward netic screens for altered shoot Na+ concetration. A suppressorscreenof Athkt1;1planmay usefully reveal other steps in the Ntransport process.

It should be noted that the HKT gene family is quite diverse, which has confused isin the past (seeAtHKT1;1, A Case StudyConfusion), and this sequence diversity likreects a diversity of function. As such,creased clarity has been provided by diving the HKT gene family into two distinsubfamilies (99). These subfamilies largalthough not exclusively, reect differenin a likely selectivity-determining amino aresidue in therst so-called pore loop regiof the protein, and differences in the catiselectivity. Subfamily 1 members containimportantserineresidue, andare largely Na+

selective; subfamily 2 members have the ine replaced by a glycine, and can catalyzetransportof K + andprobably also cancatalyhigh afnity Na+ inux (60).

TISSUE TOLERANCEOF SODIUM IONS At the cellular level, high amounts of Nand Cl arriving in leaves can be toleraby anatomical adaptations and intracellupartitioning. Dicotyledonous halophytexemplify two types of anatomical adaptions: salt-induced increase in cell size dueincreases in vacuole volume (succulence), the excretion of Na+ and Cl by salt glan

666 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


17/34

(modied trichomes) or bladders (modiedepidermal cells) (34). Succulence is extremely rare in monocotyledonous species, and saltglands occur in only approximately 15% of monocotyledonous halophytes (T.J. Flowers,personal communication), but in all species,intercellular transport processes can promotepartitioning across the leaf.

The effectof salinity on intercellular parti-tioningof ions hasbeen particularly studied inbarley, a cereal known for its ability to toleratehigh leaf tissueconcentrations ofNa+ andCl

(19, 67), by measurement of vacuolar concen-trations by scanning electron microscope X-ray microanalysis, either in situ (67) or in saptaken from single cells using a microcapillary (41). In salt-treated barley, there is a greateraccumulation of Cl in epidermal compared with mesophyll cells (41,63, 67, 77).The con- verse is true for K + , that is, there is a greateraccumulation of K + in mesophyll compared with epidermal cells (23, 41, 67), but there isno evidence of partitioning of Na+ betweendifferent cell types (67).

Intracellular Compartmentation of Na +

Na+ must be partitioned within cells so thatconcentrations in the cytoplasm are keptlow, possibly as low as 1030 mM. No directmeasurements of cytosolic concentrations inleaves have been reported, but in roots, directmeasurementsof cytosolicNa+ in salt-treatedplants via the use of ion-sensitive microelec-trodes indicate cytosolic Na+ concentrationsrange from 10 to 30 mM (13). In animal cells,cytosolic concentrations are also of this order(9). However, theconcentration at which Na +

becomes toxic is not well dened. In vitrostudies showed Na+ starts to inhibit most en-zymes at concentrations approaching100mM(54), although some enzymes are sensitive tolower concentrations (33).The concentrationat which Cl becomes toxic is even less welldened, but is probably similar to that forNa+ (33). Even K + starts to inhibit enzymesat concentrations above 100 mM (33, 54).

(Continued )

functionsinNa+ recirculationfromshoots to roots,by loadingNa+ from the shoot into phloem and then unloading it intothe roots for efux.

However, Sunarpi and colleagues (122) demonstrated

that AtHKT1;1 localizes to the plasma membrane of xylemparenchyma cells in the shoot. They found both reducedphloem Na+ and elevated xylem Na+ in the shoot of hkt1;1mutants and proposed that AtHKT1;1 functions primarily toretrieve Na+ from the xylem, at least in the shoot, and thatretrieval of Na+ into the symplast has a secondary effect onphloem Na+ levels.

Mostrecently, Davenportandcoworkers (25)usedradioac-tive tracers to dissect the individual transport processes con-tributing to Na + and K + accumulation in intact, transpir-ing plants to provide the most direct evidence to date that AtHKT1;1 is involved in Na+ retrieval from the xylem.

Results from closely related members of the HKT1 sub-family in rice and wheat are also consistent with a functionof AtHKT1;1 in retrieval of Na+ from the xylem. Thus, eventhough AtHKT1;1 catalyzes Na+ inux into cells, its effect atthe level of the whole plant is to reduce net Na+ inux intothe shoot.

Hypothesesregardingthe role of AtHKT1;1 in Na + trans-port have relied mainly on measurements of tissue ion con-tents, which are the net result of a number of different trans-port processes, or on disruptive measurements of phloem and xylemcontents.These measurementscanoften be interpretedin many ways.

In addition, many of the experiments have been conductedin plants grown on agar plates (where transpiration is ex-tremely limited). Transpiring conditions have a major inu-ence on Na+ transport and tolerance (86). This is especiallyimportant when studying a gene whose function appears to beto remove Na+ from the transpiration stream.

In Arabidopsis , although AtHKT1;1 function is now welldened in roots, its function in the shoot remains obscure,and the hypotheses of Berthomieu and colleagues (10) requirecareful consideration. In rice, functions for the nineHKT -like genes identied thus far remain largely unknown. Although OsHKT1;5 appears to have a similar role to that ofthe Arabidopsis gene, the functions of other members of thegene family may well be quite distinct, as indicated by Horiand coworkers (60). Much more work is required to properlyelucidate the functions of this important gene family.


byWB6065TBUBHnnov

on

0Fop

onu

ony


18/34

Na + activity: thetotal amount of freely diffusing Na+

available fortransport per unit volume of solution

Ideally, Na+ and Cl should be largely se-questered in the vacuole of the cell. That thissequestering occurs is indicated by the highconcentrations of Na+ found in leaves thatare still functioning normally. Concentrations well over 200 mM on a tissue basis are com-mon, yet these same concentrations will com-pletely repress enzyme activity in vitro andare beyond all known direct measurements of cytosolic Na+ inboth eukaryoticand prokary-otic cells, other than the extremely halophilicprokaryotes (127). Importantly, enzymes inhalophytes are not more tolerant of salt in vitro than the corresponding enzymes in non-halophytes, suggesting compartmentation of Na+ is an essential mechanism in all plants,rather than a result of the evolution of tol-erance of enzymatic functions in plants fromsaline environments.

Thus, differences in the expression levelsof AtNHX1 or AtAVP1 may affect the po-tential to sequester Na+ in vacuoles of theleaves. Increased salinity tolerance of a rangeof plant species overexpressingNHX genes(5, 11, 15, 59, 138, 143, 144) orAtAVP1 (45)indicates the feasibility of such a mechanismand suggests that this process is important forNa+ tolerance not only in Arabidopsis but alsoacross plant species.

Increased efciency of intracellular com-partmentation may explain differences insalinity tolerance between closely relatedspecies. This hypothesis is supported by ndings of a much greater salt stressinduced Na+ /H + antiporter activity in thesalt-tolerant species Plantago maritima thanin the salt-sensitive speciesPlantago media(118).

Increased vacuolar Na+ concentrations would require a coordinated increase in theosmotic pressure of theothersubcellular com-partments, including the cytosol, to maintaintheir volume. This can be achieved by an in-crease in the concentration of K + to sub-toxiclevels, as well as the concentration of compat-ible solutes.

Increased Accumulation of Compatible SolutesIf Na+ and Cl are sequestered in the vacuoof a cell, organic solutes that are compati with metabolic activity even at high conctrations (hence compatible solutes) must cumulate in the cytosol and organelles to b

ance the osmotic pressure of the ions in t vacuole (35, 136). The compounds that accmulate most commonly are sucrose, proliandglycine betaine, although other moleculcan accumulate to high concentrations in ctain species (35, 58, 89).

In many halophytes, proline or glycine btaine occur at sufciently high concentratioin leaves (over 40 mM on a tissue water bato contribute to the osmotic pressure (ov0.1 MPa) in the cell as a whole (35). In g

cophytes, the concentrations of compatibsolutes that accumulate are not so high, the order of 10 mM, but if partitioned eclusively to the cytoplasm, they could genate a signicant osmotic pressure and fution as an osmolyte. At low concentratiothese solutes presumably have another roperhaps in stabilizing the tertiary structuof proteins, and function as osmoprotectan(108).Anosmolyterolehasbeensuggestedfglycine betaine accumulation in maize; co

parison of near-isogenic maize lines with cotrasting glycine betaine accumulation showthat lines that were homozygous for theBe(glycine betaine accumulation) gene had10%20% higher biomass under saline coditions (113).

Accumulation of these compatible solutsuch as proline and mannitol, also occurs uderdrought stress andsometimes under othestresses that reduce growth, such as low teperature. Many studies of genes controlli

the synthesis or metabolism of these soluhave indicated their essential role in toleranto abiotic stresses (16, 56, 108). For examthe lower expression of a gene encoding pline dehydrogenase ( PDH ) may contributo the higher salt tolerance of Thellungie

668 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


19/34

halophilacompared with its salt-sensitive rel-ative Arabidopsis thaliana(69). Enhancementof mannitol accumulation in Arabidopsis by overexpression of a mannose-6-phosphate re-ductase from celery caused substantial andsustained increases in growth rate and photo-synthesis in saline treatment but not drought,suggesting that mannitol protects the chloro-plasts against salt (116).The transgenehadnoeffect on growth in control conditions (116). This is noteworthy, because most reports of transgenic alterations in levelsofenzymes thatcatalyze rate-limiting steps describe plants whose growth is signicantly reduced. Thismay be because uncontrolled accumulationof the solutes perturbs other metabolic pathways, diverting substrates from essential pro-cesses such as protein synthesis and cell wallsynthesis.

Compatible solute synthesis comes withanenergy cost and hence involves a potentialgrowth penalty. In leaf cells, approximately seven moles of ATP are needed to accumulateone mole of NaCl as an osmoticum, whereastheamountof ATPrequired to synthesize onemole of an organic compatible solute is anorder of magnitude higher (105). The ATPrequirement for the synthesis or accumula-tion of solutes has been estimated as 3.5 forNa+ , 34 for mannitol, 41 for proline, 50 forglycine betaine, and approximately 52 for su-crose (105).These valuesassumea productionof 0.5 mole of ATP per photon and nitrate asthe source of N. The synthesis of these com-pounds occurs at the expense of plant growth,but may allow the plant to survive and recoverfrom the presence of high external concentra-tions of salt.

Tolerance of leaf tissue to high Na+ con-centrations is clearly an adaptive mechanism,as exemplied by most halophytes (34) andglycophytes such as barley, which can toler-ate at least 400 mM Na+ in leaf blades (67). The high Na + and the accompanying Cl al-lows barley to osmotically adapt and to main-tain turgor in the face of high soil salinities. This is the cheapest form of osmotic adap-tation. The mechanism of Na+ exclusion en-

ables the plant to avoid or postpone the prob-lem of ion toxicity, but unless the exclusionof Na+ is compensated for by the uptake of K + , it creates a greater demand for organicsolutes for osmotic adjustment. The synthesisof organic solutes jeopardizes the energy bal-ance of the plant. Thus, the plant must steer acourse through ion toxicity on the one hand,and turgor loss on the other, in analogy tothe Scylla versus Charybdis dilemma faced by Ulysses.

OBSERVATIONS IN WHICHSALINITY TOLERANCE ISCLEARLY INDEPENDENT OF TISSUE SODIUM ION CONCENTRATIONS

A negative correlation between salinity toler-ance and Na+ accumulation in leaves is of-ten seen when comparing different genotypes within a species (88, 127), but this is not thecase when comparingdifferentspecies, such as wheat and barley.Figure 5 illustrates the re-lationship between salinity tolerance and leaf Na+ concentration found within a species, in

Salinity tolerance

Leaf Na + concentration

IncreasedNa + exclusion

Increased osmoticand tissue tolerance

a b c

Figure 5Hypothetical relationships between salinity tolerance and leaf Na+

concentration for three different species, denoted by a, b, and c for rice,durum wheat, and barley. Within most species, there is a negativecorrelation between salinity tolerance and shoot Na+ concentration, as inrice (141) and durum wheat (91) and, with less conviction, in barley (19larger intercept on the x-axis indicates an increased tolerance to the osmopressure of the soil solution or to high internal concentrations of Na+ orCl .


byWB6065TBUBHnnov

on

0Fop

onu

ony


20/34

cba

0 100 200 300 4000 500 1000 1500

Leaf Na + concentration(mol g 1 DW)



S

a l i n

i t y

t o l e r a n c e

( % c

o n

t r o

l )

2000 2500 3000

T. turgidumssp. durum T. aestivum

0 0

10 10

20 20

30 60

40 80

50

0

10

20

30

40

50 100

0 500 1000 1500 2000

T. turgidumssp. turanicumssp. turgidumssp. carthlicumssp. polonicum

Figure 6Relationships measured between salinity tolerance (biomass in salt as a % of biomass in controlconditions) and leaf Na+ concentration in different wheat species. (a) Negative relationship for durum wheat (91). (b) Lack of relationship for other tetraploid wheats (91). (c ) Lack of relationship for bread wheat (46).

this case rice (141), durum wheat (91), andthe Hordeumgenus, including barley (44); thegure also shows that this relationship shiftsfor different species. This shift can reect dif-ferences in tissue tolerance of Na+ betweendifferent species, differences in tolerance of Cl , or differences in tolerance of the osmoticpressure of the soil solution.

Although Na+ exclusion is often a pri-mary determinant of variability in salinity tol-erance within a species (Figure 6 a), many ex-ceptions to this generalization exist, such as within certain subspecies of tetraploid wheat,Triticum turgidum (Figure 6 b), in which dif-ferences in salt tolerance do notcorrelatewithdifferences in Na+ exclusion. One study ob-serveda lack of correlationwithinbreadwheat(Triticum aestivum) (Figure 6 c ), possibly be-cause at the moderate salinity of 100 mMNaCl, the leaf Na+ concentration was belowthe toxic level. The genetic variation in Na+

exclusion may contribute to greater salinity tolerance only in highly saline soil that causeshigher leaf Na+ accumulation. We concludethat Na + exclusion remains an important fac-tor, and increasing Na+ exclusion by conven-tional or transgenic methods could increasesalinity tolerance, but these results indicateother mechanisms may be important in many species, especially at high salinity.

Mechanisms of Salinity ToleranceOther than Na + Exclusion In addition to tissue tolerance mechanismdiscussed above, other mechanisms of saity toleranceunrelated to Na + exclusioncoualso be important in these plants.

Osmotic tolerance. The relative impotance of variation in osmotic tolerance mains unknown for most species, which likreects the relative difculty of quantify

this parameter. A close association likely ists between osmotic tolerance and tissue terance of Na+ , because genotypes that toerate high internal Na+ concentrations leavesby compartmentalizingit in thevacuoshould also be more tolerant of the osmostress owing to their elevated osmotic adjument. However, this speculation remains be tested.

K + accumulation in cytoplasm. The con

centration of K +

in the cytoplasm relativethat of Na+ may be a contributing factor salinity tolerance. In Arabidopsis,an additionsupply of K + alleviated the phenotype of t sos mutants (145), which may be due to an crease in cytoplasmic K + concentrations. barley, Shabala and colleagues (17) foun

670 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


21/34

negative correlation between Na+ -activatednet K + efux in 3d-old seedlings and salinity tolerance of mature barley plants. This phe-nomenon may be related to root K + status,although a strong relationship between leaf K + concentrations and salinity tolerance hasnot been found.

Cl tolerance. The question is often asked:Why focus only on Na+ , why not also con-sider Cl ? This question relates particularly to species that accumulate high concentra-tions of Cl and not Na+ in leaves, suchas soybean, woody perennials such as avo-cado, and those species that are routinely grown on Cl -excluding rootstocks such asgrapevines and citrus. For these species, Cl

toxicity is more important than Na+ toxic-ity. However, this statement does not imply thatCl is more metabolically toxic than Na+ ,rather these species are better at excludingNa+ from the leaf blades than Cl . For exam-ple, Na+ does not increase in the leaf blade of grapevines until after severalyearsofexposureto saline soil, then the exclusion within theroot, stem, and petiole breaks down, and Na+

starts to accumulate in the leaf blade, whereasleaf blade Cl concentrations increase pro-gressively (100). Thus, Na+ may be a moretoxic solute, but because the plant is manag-ing the Na+ transport better than Cl trans-port, Cl becomes the potentially more toxiccomponent.

Many studies have been undertaken to de-termine whether Na + is more or less toxicthan Cl . The use of different salts has pro-duced only equivocal results, because of thedifculty in changing the external concentra-tion of one ion versus another without chang-ing the osmotic pressure of the external solu-tion or the rate of uptake of other ions. Themost convincing approaches to test the toxic-ity of Na+ versus Cl are genetic approaches.Between different species of wheat, genetic variation in salinity tolerance correlates withleaf Na+ accumulation but not Cl accumu-lation (51, 52). However, genetic variation insalinity tolerance correlates with leaf Cl ac-

THE THERMODYNAMICS AND MECHANISMS CONTROLLING CL

TRANSPORT

Mechanisms of Cl transport are shown in Figure 3 . In mostsituations, Cl inux requires energy and is probably cat-

alyzed by a Cl

/2H+

symporter (32, 112), although Skerrett& Tyerman (117) have shown that passive uptake could occuin saline conditions if the membrane potential is depolarizedand cytosolic Cl is low (less than 20 mM). The cytosolicCl concentration is likely in the range of 10 to 20 mM, butmay be higher in saline conditions. Felle (32) showed thathe cytoplasmic concentration doubled (from 15 to 33 mM) within minutes of increasing the external Cl concentrationfrom 0 to 20 mM. Given the uncertainties surrounding thethermodynamics, useful speculation on the role of net inuxprocesses in salinity tolerance is difcult. Nevertheless, if Cl

inux is active, and thus efux is passive, the opening ofCl -permeable channel in nonsaline conditions would favorthe passive efux of Cl . Thus, activation of a Cl -permeablechannel in saline conditions could be useful for reducing thenet inux of Cl . Yamashita and coworkers (140) observedan increase in Cl permeability of protoplasts isolated frombarley roots after plants were pretreated with 200 mM NaCl,supporting such a role for Cl channels. Comparisons of Cl

transport in lines with different levels of Cl accumulation inthe shoot would reveal the signicance of different transportprocesses in whole plant accumulation.

cumulation in citrus (119). Cl inux is likely active. See The Thermodynamics and Mech-anisms Controlling Cl Transport.

Cl loading into the xylem is most likely apassive mechanism mediated by anion chan-nels such as those characterized by Gilliham& Tester (47). These channels are downreg-ulated by ABA, which may serve to limit Cl

transfer to the shoot in saline conditions. Ra-dioactive tracer studies have shown that netCl loading into the root xylem is lower ingrapevine genotypes that have lower shootCl accumulation ( J. Tregeagle, M. Tester &R. Walker, unpublished results). The controlof Cl transport to shoots may be due to re-duced loading of Cl via anion channels, butmay also be due to increased active retrieval


byWB6065TBUBHnnov

on

0Fop

onu

ony


22/34

of Cl from the xylem stream. Sites of tis-sue Cl accumulation indicate that Cl canbe retrievedfrom thexyleminpetioles,woody stems, and roots.

Results from biochemical approaches(study of the effects of different salts on pro-tein synthesis or enzyme activity) have beenequivocal, as have attempts to estimate Cl

concentrations in the cytoplasm or organellessuch as chloroplast and mitochondria. Yet tis-sue concentrations as high as 400 mM are tol-erated by most species, and even the sensitivespecies like citrus can tolerate tissue concen-trations of 250 mM, so Cl must be com-partmentalized in the vacuole. The thermo-dynamics and mechanisms of Cl transport atthe tonoplast are largely unknown, and dif-ferences in properties between tolerant andsensitive lines are regrettably obscure.

APPROACHES FOR FUTURESTUDIES

The Importance of Cell TypeSpecic Processes

Gene expression studies using constitutivepromoters provide limited biological infor-mation compared with the use of induciblepromoters (120) or cell typespecic pro-moters (127, 128). The choice of promot-ers can signicantly affect the results froma transgenic manipulation. The constitutiveexpression of genes encoding compatible so-lutes often inhibits plant growth, as shownby stunted growth and sterility of lines withhigher concentrations of mannitol (1). Ex-pression that is inducible upon plant stressshould have little effect on growth in controlconditions, but can increase tolerance to theappliedstress,as shown for trehalose accumu-lation in rice (43). In a similar vein, constitu-tive expression of AtHKT1;1 causes increasedshoot accumulationof Na+ andreduced salin-ity tolerance, whereas expression specically in the stele of mature roots has the oppositeeffect (85).

Relevant Growth Conditionsand Salinity Treatments There are several easily adopted methodsgrowing plants that could greatly facilitthe interpretation of results, comparisons btween experiments in different laboratoriand the relevance of experiments to eld

uations. The time of exposure to salinity and

severity of the salt treatment determine tphysiological and molecular changes that observed. Metabolomics and transcriptomstudies produce different answers dependion the tissue examined and whether the plais growing or dying. Whether the plant transpiring or not is also important, as showfor the HKT gene family (SeeAtHKT1;1, Case Study of Confusion). A high-salt tre

ment for a sensitive plant likeArabidop will induce changes predominantly associa with senescence; however, a low-salt trement may not result in discernable changesgene expression and metabolite levels. Fiing the right balance can be difcult. Fapplication to the agricultural context, expiments should focus on growth and reprodutive yield, rather than survival. Toleranceextreme stress is of ecological relevanceperennial species, but is generallynot releva

to annual species.Osmotic effects could be distinguishfrom ionic effects by analyzing growing sues for the osmotic effect, and analyzolder transpiring leaf blades for the ionic fect( Table 1 ).Shorttimesofexposuremayuseful for signaling studies; however, it is portant to recognize that transientcell shrinkage and recovery of volume occur after a shock, and to relate measurements to the nsteady state reached.

Addition of Na+

or any electrolyreduces Ca2+ activity in solutions. If Caactivity is not maintained by addition of C with the Na+ , uncertainty remains abo whether the effects of Na+ addition are duto the increase in Na+ or the decrease

672 Munns Tester

byWB6065TBUBHnnov

on

0Fop

onu

ony


23/34

available Ca2+ . Thus, salt treatments needto include supplemental Ca2+ to maintainstable Na+ /Ca2+ ratios, or constant Ca2+

activity (as calculated using programs such asGeochem (96) or MINTEQ ( http://www.lwr.kth.se/English/OurSoftware/vminteq/ ).Consideration should also be given to the

addition of silicon to solutions (29, 48) assodium silicate (or liquid glass, mainly Na2SiO3), taking care to adjust the pH afterits addition.

Without a good understanding of thephysiology involved and the phenotype tomeasure, complemented by the discovery of key genes in model systems, the recent fastprogress on control of shoot Na+ in rice and wheatwouldnothave been possible.Elucidat-ing more basic physiology and the moleculargenetics of other aspects of salinity responses(notably osmotic tolerance) will facilitate thegeneration of further applications in majorcrops.

SUMMARY POINTS

1. Plant responses to salinity occur in two phases: a rapid, osmotic phase that inhibitsgrowthof young leaves,anda slower, ionicphase that accelerates senescence of matureleaves.

2. Plant adaptations to salinity are of three distinct types: osmotic stress tolerance; Na+

exclusion; and tissue tolerance, i.e., tolerance of tissue to accumulated Na+ , and pos-sibly Cl .

3. Our understanding of Na + exclusion fromleaves and the role of the HKT gene family is increasing, although the molecularbases formany other transport processes remainobscure.

4. The salt overly sensitive (SOS) signal transduction pathway is clearly important insalinity tolerance, although the mechanism of action at the whole plant level remainsto be established.

5. Osmotic tolerance and tissue tolerance both increase the ability to maintain growthfor a given accumulation of Na+ in the leaf tissue. Increased osmotic tolerance isevident mainly by the increased ability to continue production of new leaves, whereastissue tolerance is evident primarily by the increased survival of older leaves.

6. Na+ sequestration and compatible solute synthesis are important processes for tissuetolerance. Mechanisms of osmotic tolerance remain unknown.

7. To benet more from the new genomics approaches, molecular studies with plantsgrown in physiologically realistic conditions are needed.

FUTURE ISSUES

1. Signicant breakthroughs have been made on the mechanisms and control of Na+ ac-cumulation by thehigh-afnityK + transporter( HKT )genefamilyandtheimportanceof the intraplant management of Na+ . Nevertheless, large gaps remain in our knowl-edge of Na+ transport, notably the control of phloem transport, the identity of thegenes encoding nonselective cation channels responsible for the initial entry of Na+

into plants, and the role of other solutes in salinity tolerance, including K + and Cl .


byWB6065TBUBHnnov

on

0Fop

onu

ony


24/34

2. Molecularprocesses that control Na+ compartmentalization in vacuoles have receivedmuch attention, but other essential processes in tissue tolerance of Na+ and Cl andosmotic adjustment remain relatively unknown.

3. Signaling pathways at the intracellular level have been well described, but longdistance signaling requires more attention. How do the leaves know the roots arein saline soil, when so little salt is delivered in the xylem to the leaves? Yet, the legrowth rate andstomatal conductance are reduced in proportion to the concentrationof salt in the soil solution, and not in proportion to the salt concentration in the xylemor the leaves.

4. Forward genetic approaches wil

Documents

annurev.arplant.59.032607