-
Critical Reviews in Plant Sciences, 24:2358, 2005Copyright c
Taylor & Francis Inc.ISSN: 0735-2689 print / 1549-7836
onlineDOI: 10.1080/07352680590910410
Drought and Salt Tolerance in Plants
Dorothea Bartels and Ramanjulu SunkarInstitute of Molecular
Physiology and Biotechnology of Plants, University of Bonn,
Kirschallee 1,D-53115 Bonn, Germany
Table of Contents
I. INTRODUCTION
.............................................................................................................................................25
II. OSMOTIC STRESS
..........................................................................................................................................25A.
Osmotic Stress-Induced Growth Arrest
..........................................................................................................25B.
Osmotic Stress Affects Cell Division and Elongation
......................................................................................30
III. SIGNAL PERCEPTION
...................................................................................................................................30
IV. SIGNAL TRANSDUCTION
..............................................................................................................................31A.
MAPKinase Pathways
..................................................................................................................................31B.
SNF-1-Like Kinases Are Involved in Osmotic Stress Signalling
.......................................................................32C.
Phosphatases
...............................................................................................................................................32D.
Phospholipid Signalling
...............................................................................................................................33
1. Inositol 1,4,5-Triphosphate (IP3)
.........................................................................................................332.
Phosphatidic Acid (PA)
......................................................................................................................34
E. Other Signalling Molecules
..........................................................................................................................341.
Salicylic Acid
....................................................................................................................................342.
Nitric Oxide (NO)
.............................................................................................................................34
V. CALCIUM SIGNALLING DURING DEHYDRATION AND SALT STRESS
....................................................35A.
Calcium-Dependent Protein Kinases (CDPKs)
...............................................................................................35B.
Calcium-Binding Proteins
............................................................................................................................35C.
Ca2+-Mediated SOS Pathways Are Involved in Ion Homeostasis
.....................................................................35D.
Calcineurin B (CBLS) and Osmotic Stress Responses
.....................................................................................35E.
Ca2+ ATPases
.............................................................................................................................................36
VI. TRANSCRIPTIONAL REGULATION OF GENE EXPRESSION
....................................................................36A.
ABA Response Elements (ABREs)
...............................................................................................................36B.
The Dehydration Response Element (DRE)
...................................................................................................37C.
Transcription Factors Modulated by Osmotic Stress
........................................................................................37
1. Basic Region Leucine Zipper (bZIP) Proteins
.......................................................................................372.
Homeodomain-Leucine Zipper Proteins (HD-ZIP)
................................................................................373.
Zn-Finger Proteins
.............................................................................................................................374.
AP2/ERF-Type Transcription Factors
..................................................................................................385.
Myb-Like Proteins
.............................................................................................................................386.
Myc-Like Proteins
.............................................................................................................................397.
CDT-1
..............................................................................................................................................39
Address correspondence to Dorothea Bartels, Institute of
Molecular Physiology and Biotechnology of Plants, University of
Bonn, Kirschallee1, D-53115 Bonn, Germany. E-mail:
[email protected]
23
-
24 D. BARTELS AND R. SUNKAR
VII. ACCUMULATION OF SUGARS AND COMPATIBLE SOLUTES
...................................................................39A.
Sugars
........................................................................................................................................................39B.
Cyclitols
.....................................................................................................................................................40C.
Proline
........................................................................................................................................................40D.
Glycine Betaine
...........................................................................................................................................40
VIII. PROTECTIVE PROTEINS AND OTHER PATHWAYS INVOLVED IN STRESS
ADAPTATION .....................40A. Late Embryogenesis-Abundant
(LEA) Proteins
..............................................................................................40B.
Aquaporins
.................................................................................................................................................41C.
Heat Shock Proteins (Hsps)
..........................................................................................................................41D.
Proteases and Proteinase Inhibitors
...............................................................................................................42E.
Polyamines
.................................................................................................................................................42
IX. OXIDATIVE STRESS A CONSEQUENCE OF DEHYDRATION AND SALT
STRESS ....................................42A. Formation of
Reactive Molecules
..................................................................................................................42B.
Enzymes That Detoxify Aldehydes
...............................................................................................................43C.
Peroxiredoxins
............................................................................................................................................43D.
Thioredoxins
...............................................................................................................................................43E.
Protein Oxidation
........................................................................................................................................43
X. IONIC STRESS
................................................................................................................................................44A.
Na+ Toxicity and Homeostasis
.....................................................................................................................44B.
Na+ Exclusion
............................................................................................................................................44C.
Na+ Compartmentalization
...........................................................................................................................44D.
Proton Transporters and Salt Tolerance
..........................................................................................................45E.
SOS Pathway and Ion Homeostasis
...............................................................................................................45
XI. ABSCISIC ACID (ABA)
....................................................................................................................................46A.
Regulation of ABA Levels
............................................................................................................................46B.
The Role of ABA in Stomatal Closure
...........................................................................................................47C.
ABA Signalling Components
........................................................................................................................47
XII. MONITORING GLOBAL GENE EXPRESSION USING MICROARRAY ANALYSIS
.....................................48
XIII. CONCLUSIONS
...............................................................................................................................................48
ACKNOWLEDGMENTS
..................................................................................................................................49
REFERENCES
.................................................................................................................................................49
Agricultural productivity worldwide is subject to increasing
en-vironmental constraints, particularly to drought and salinity
dueto their high magnitude of impact and wide distribution.
Tradi-tional breeding programs trying to improve abiotic stress
toler-ance have had some success, but are limited by the
multigenicnature of the trait. Tolerant plants such as
Craterostigma planta-genium, Mesembryanthemum crystallinum,
Thellungiella halophilaand other hardy plants could be valuable
tools to dissect the ex-treme tolerance nature. In the last decade,
Arabidopsis thaliana, agenetic model plant, has been extensively
used for unravelling themolecular basis of stress tolerance.
Arabidopsis also proved to be
extremely important for assessing functions for individual
stress-associated genes due to the availability of knock-out
mutants andits amenability for genetic transformation. In this
review, the re-sponses of plants to salt and water stress are
described, the regula-tory circuits which allow plants to cope with
stress are presented,and how the present knowledge can be applied
to obtain tolerantplants is discussed.
Keywords dehydration, salinity, ABA, plant stress tolerance,
trans-genic plants
-
DROUGHT AND SALT TOLERANCE IN PLANTS 25
I. INTRODUCTIONDrought and salinity are two major environmental
factors
determining plant productivity and plant distribution.
Droughtand salinity affect more than 10 percent of arable land, and
de-sertification and salinization are rapidly increasing on a
globalscale declining average yields for most major crop plants
bymore than 50 percent (Bray et al., 2000). Understanding
planttolerance to drought and salinity is therefore of fundamental
im-portance and forms one of the major research topics. Plants
canperceive abiotic stresses and elicit appropriate responses
withaltered metabolism, growth and development. The
regulatorycircuits include stress sensors, signalling pathways
comprisinga network of protein-protein reactions, transcription
factors andpromoters, and finally the output proteins or
metabolites. Clas-sical breeding approaches revealed that stress
tolerance traitsare mainly quantitative trait loci (QTLs), which
make geneticselection of traits difficult. Nevertheless, very
respectable stresstolerant crops have been obtained, mainly by
introducing traitsfrom stress-adapted wild relatives.
As water and salt stresses occur frequently and can affect
mosthabitats, plants have developed several strategies to cope
withthese challenges: either adaptation mechanisms, which allowthem
to survive the adverse conditions, or specific growth habitsto
avoid stress conditions. Stress-tolerant plants have evolvedcertain
adaptive mechanisms to display different degrees of tol-erance,
which are largely determined by genetic plasticity. Dif-ferential
stress tolerance could be attributed to differences inplant
reactivity in terms of stress perception, signal transduc-tion and
appropriate gene expression programs, or other novelmetabolic
pathways that are restricted to tolerant plants. Thehypothesis that
the genetic program for tolerance is at least tosome extent also
present in nontolerant plants is supported by theobservation that
gradual acclimation of sensitive plants leads toacquisition of
tolerance to some degree. These plants may needgradual adaptation
for proper expression of genes responsiblefor acquisition of
tolerance (Zhu, 2001).
Our understanding of how plants respond to water and saltstress
has advanced by analyzing stress-tolerant species like
thedesiccation tolerant plant Cratesostigma plantagineum
(Bartelsand Salamini, 2001) or the salt-tolerant plant
Mesembryanthe-mum crystallinum (Bohnert and Cushman, 2000). Despite
thefact that research with the desiccation-tolerant plant C.
plan-tagineum revealed additional novel aspects such as specific
car-bohydrate metabolism and the existence of CDT-1 gene thatwere
unknown in other nontolerant plant species (Bartels andSalamini,
2001), the connection between these metabolites andtolerance is
still correlative. Molecular genetic studies have beenperformed
with Arabidopsis thaliana, which does not displayextreme stress
tolerance, but shows many stress responses atthe molecular level
and has therefore been successfully used fora genetic dissection of
stress response pathways (Zhu, 2002;Shinozaki et al., 2003). More
importantly, we learned that it isvery likely that the extreme
tolerant model plants did not acquireunique genes since
stress-relevant genes are ubiquitously present
in the plant kingdom. A particular gene expression pattern is
of-ten associated with the tolerant phenotype and it is unknownto
date how this is achieved. This may involve other molecu-lar
aspects, like chromatin organization, which have not beenwell
researched. Recently, the salt-tolerant plant
Thellungiellahalophila was introduced as an attractive model plant
to studymolecular genetics of salt tolerance, because it is a close
rel-ative to Arabidopsis and amenable for transformation unlikethat
of other tolerant model plants (Volkov et al., 2003; Bressanet al.,
2001; Zhu, 2001). Thus it may be possible to combineprofound
genetic knowledge with the expression of extremetolerance.
Exposure to drought or salt stress triggers many commonreactions
in plants. Both stresses lead to cellular dehydration,which causes
osmotic stress and removal of water from the cy-toplasm into the
extracellular space resulting in a reduction ofthe cytosolic and
vacuolar volumes. Another consequence is theproduction of reactive
oxygen species which then in turn affectscellular structures and
metabolism negatively. Early responsesto water and salt stress are
largely identical except for the ioniccomponent. These similarities
include metabolic processes suchas, for example, a decrease of
photosynthesis or hormonal pro-cesses like rising levels of the
plant hormone ABA. High in-tracellular concentrations of sodium and
chloride ions are anadditional problem of salinity stress.
Adaptation to salinity and drought is undoubtedly one ofthe
complex processes, involving numerous changes includ-ing attenuated
growth, the activation/increased expression orinduction of genes,
transient increases in ABA levels, accumu-lation of compatible
solutes and protective proteins, increasedlevels of antioxidants
and suppression of energy-consumingpathways. However, no consensus
has been reached in defin-ing the key processes determining
tolerance and the secondaryfollow-up processes. With the
advancement of high throughputDNA technologies, several hundred
stress-induced or upregu-lated genes have been identified. The
search for stress-associatedgenes may have been saturated at least
in Arabidopsis. How-ever, the function of only a limited number of
gene products areknown (Ingram and Bartels, 1996; Bray, 1997;
Shinozaki andYamaguchi-Shinozaki, 1997; Hasegawa et al., 2000;
Ramanjuluand Bartels, 2002). Several stress-associated genes have
beenevaluated or studies are in progress for their contribution
todrought or salt tolerance in laboratory studies (Table 1).
Morerigorous and perhaps field studies are required for possible
uti-lization of these genes for improving stress tolerance in
agricul-tural plants through biotechnological approaches.
II. OSMOTIC STRESS
A. Osmotic Stress-Induced Growth ArrestArrest of plant growth
during stress conditions largely de-
pends on the severity of the stress. Mild osmotic stress
leadsrapidly to growth inhibition of leaves and stems, whereas
rootsmay continue to elongate (Westgate and Boyer, 1985; Sharp
-
26 D. BARTELS AND R. SUNKAR
TABLE 1Stress-responsive genes contributing to drought or salt
tolerance in transgenic plants
Gene Plant species Parameters evaluated Remarks Reference
DREB1A (AP2Transcription factor)
OsDREB1A
Arabidopsis
Arabidopsis
survived 2 weeks withoutwatering; activated theexpression of
genesinvolved in stress tolerance(rd29A)
improved drought, salt,and cold tolerance
improved drought, salt,and freezingtolerance
Kasuga et al., 1999;Liu et al., 1998
Dobouzet et al., 2003
Alfin1 (Transcriptionfactor)
Alfalfa increase in root growth undernormal and
salineconditions
improved salt tolerance Winicov, 2000
Tsi1 (EREBP/AP2DNA binding motif)
Tobacco activated the expression ofseveral PR genes, PR1,PR2,
PR3, osmotin andSAR8.2 under unstressedcondition
improved salt andpathogen tolerance
Park et al., 2001
CBF1 (DREB1B) Tomato activated gene expression,catalase1 coupled
withdecreased accumulation ofH2O2
improved droughttolerance
Hsieh et al., 2002
CBF4 Arabidopsis activated the expression ofstress responsive
genes
improved droughttolerance
Haake et al., 2002
ABF3/ABF4 Arabidopsis reduced transpiration andbetter survival
underdrought conditions
improved droughttolerance
Kang et al., 2002
AtMYC2/AtMYB2 Arabidopsis less electrolyte leakage intransgenic
plants
improved droughttolerance
Abe et al., 2003
ZPT2-3(Cys2/His2-typeZincfinger protein)
Petunia better survival rate duringdrought stress
improved droughttolerance
Sugano et al., 2003
CpMYB10 Arabidopsis better germination abilityunder saline or
mannitoltreatments
improved drought andsalt tolerance
Villalobos et al., 2004
mt1D (Mannitol-1-phosphatedehydrogenase)
Tobacco
targeted to chloroplastsof tobacco
Wheat
increased root biomassaccumulation
increased retention ofchlorophyll against methylviologen
treatment
better biomass production
improved salt tolerance
increased oxidativestress tolerance
improved salt andosmotic stresstolerance
Tarczynski et al., 1993
Shen et al., 1997
Abede et al., 2003
P5CS (Pyrroline-5-carboxylatesynthase)
Tobacco
Rice
enhanced root biomass andflower development
increase in biomassaccumulation
improved salt tolerance
improved drought andsalt tolerance
KaviKishor et al., 1995
Zhu et al., 1998
(Continued on next page)
-
DROUGHT AND SALT TOLERANCE IN PLANTS 27
TABLE 1Stress-responsive genes contributing to drought or salt
tolerance in transgenic plants (Continued)
Gene Plant species Parameters evaluated Remarks Reference
SacB Tobacco
Beta vulgaris
increased biomassproduction
better dry weightaccumulation
improved tolerance toPEG treatmentimproved droughttolerance
Pilon-Smits et al., 1995
Pilon-Smits et al.,1999
TPS1 (yeast)(Trehalose-6-phosphatesynthetase) (subunit)
Tobacco delay in withering orenhanced moistureretention
capacity
improved droughttolerance
Holmstrom et al., 1996
IMT1(myo-Inositol-O-methyltransferase)
Tobacco less inhibition inphotosynthetic rate; betterrecovery
from stress
improved drought andsalt tolerance
Sheveleva et al., 1997
CodA (Cholineoxidase)
Arabidopsis
Rice
better germination andphotosynthetic activity
higher photosyntheticactivity; faster recoveryfrom stress
improved salt and coldtolerance
improved salt and coldtolerance
Hayashi et al., 1997
Sakamoto et al., 1998
ProDH (Prolinedehydrogenase)
antisense-Arabidopsis antisense plants took longerduration to
lodge theinflourescence aftersubjecting to salinity
improved salt tolerance Nanjo et al., 1999
OtsA (E. coli)(Trehalose-6-phosphatesynthase)
Tobacco increased leaf area, betterphotosynthetic activity
andbetter water retainingcapacity
improved droughttolerance
Pilon-Smits et al., 1998
OtsB (E. coli)(Trehalose-6-phosphatesynthase)
Tobacco increased leaf area, betterphotosynthetic activity
andbetter water retainingcapacity
improved droughttolerance
Pilon-Smits et al., 1998
OtsA and OtsB Rice better plant growth and lessphotooxidative
damage
improved drought, saltand low-temperaturetolerance
Garg et al., 2002
AtOAT (Ornithineamino transferase)
Tobacco accumulated more proline;higher biomass and
highergermination rate underosmotic stress conditions
improved NaCl ormannitol tolerance
Roosens et al., 2002
BADH1 (Betainealdehydedehydrogenase)
Tomato better root development andless leakage of
electrolytes
improved salt tolerance Jia et al., 2002
TPS and TPP(trehalose-6-phosphatephosphatase)(E. coli)
Rice better growth performanceand photosyntheticcapacity
improved drought andsalt tolerance
Jang et al., 2003
Cu/ZnSOD Tobacco better photosyntheticcapacity under
chillingstress
improved cold andoxidative stresstolerance
Gupta et al., 1993
MnSOD (superoxidedismutase)
Alfalfa better photosyntheticefficiency, yield, andsurvival
rate
improved water stresstolerance
McKersie et al., 1996
(Continued on next page)
-
28 D. BARTELS AND R. SUNKAR
TABLE 1Stress-responsive genes contributing to drought or salt
tolerance in transgenic plants (Continued)
Gene Plant species Parameters evaluated Remarks Reference
FeSOD (superoxidedismutase)
Tobacco protected the plasmalemmaand PSII against thedamaging
effects ofsuperoxide
improved salt andoxidative stresstolerance
Van Camp et al., 1996
Glutathione-S-transferase/Glutathioneperoxidase
Tobacco decreased oxidative damage improved salt and
coldtolerance
Roxas et al., 1997;2000
MsALR(Aldose/aldehydereductase)
Alfalfa decreased lipid peroxidationand better
photosyntheticcapacity
improved drought andheavy metaltolerance
Oberschall et al., 2000
AtALDH3 (Aldehydedehydrogenase)
Arabidopsis decreased lipid peroxidation improved drought,
saltand oxidative stresstolerance
Sunkar et al., 2003
Ascorbate peroxidase Tobacco better photosyntheticcapacity under
stressconditions
improved drought andsalt tolerance
Badawi et al., 2004
OsCDPK(Calcium-dependentprotein kinase)
Rice enhanced levels ofstress-responsive genes,rab16A, SalT, and
wsi18
improved drought andsalt tolerance
Saijo et al., 2000
AtGSK1 Arabidopsis better root growth,expression
ofstress-responsive genes inthe absence of NaCl stress;AtCP1,
RD29A, AtCBL1
improved salt anddrought tolerance
Piao et al., 2001
AtNDPK2 (Nucleotidediphosphate kinase)
Arabidopsisoverexpressing andknock-out mutant
decreased ROS accumulationin overexpressing plants,while
increased ROSaccumulation in knock-outmutants
enhanced tolerance tosalt, cold and methylviologen
treatments
Moon et al., 2002
AtNHX1 (VacuolarNa+/H+ antiporter)
Arabidopsis
Tomato
Brassica napusRice
Na+ compartmentation inthe vacuole
improved salt tolerance
Apse et al., 1999
Zhang and Blumwald,2001
Zhang et al., 2001Ohta et al., 2002
SOS1 (PlasmamembraneNa+/H+antiporter)
Arabidopsis better root growth, PSIIactivity and survival
undersalt stress conditions
improved salt tolerance Shi et al., 2003
AtHAL3a Arabidopsis transgenic seedlingsdeveloped roots and
trueleaves but not the WTseedlings
improved tolerance tosalt and osmoticstresses
Espinosa-Ruiz et al.,1999
HAL1 (Yeast) Watermelon
Arabidopsis
better growth performance oftransgenic plants;transgenic
plantsaccumulated less Na+
improved salt tolerance Yang et al., 2001
Ellul et al., 2003
(Continued on next page)
-
DROUGHT AND SALT TOLERANCE IN PLANTS 29
TABLE 1Stress-responsive genes contributing to drought or salt
tolerance in transgenic plants (Continued)
Gene Plant species Parameters evaluated Remarks Reference
CDT1 Craterostigmaplantagenium
desiccation survival of calluswithout ABA treatment;constitutive
expression ofstress-induced genes,CdeT-27-45,
CdeT-6-19,CdeT-11-24
improved dehydrationtolerance
Furini et al., 1996
GlyoxylaseI Brassica juncea better retention ofchlorophyll
improved salt tolerance Veena et al., 1999
Glyoxylase I andGlyoxylase II
Tobacco grew, flower and set viableseeds under salinity
improved salt tolerance Singla-Pareek et al.,2003
GS2 (Chloroplasticglutaminesynthetase)
Rice PSII activity retained for twoweeks of NaCl salinity,while
the control plant lostPSII activity within oneweek
improved salt tolerance Hoshida et al., 2000
AtHsp17.6A (Smallheat shock protein)
Arabidopsis control plants witheredearlier than transgenicplants
under droughtconditions; better survivalrate and fresh
weightaccumulation under salineconditions
improved drought andsalt tolerance
Sun et al., 2001
()Phospholipase D antisense-Arabidopsis stomatal closure
impaired;more water loss
decreased tolerance towater stress
Sang et al., 2001
AtNCED3 Arabidopsis-sense andantisense andknock-out plants
transpiration rate reduced insense plants and enhancedin
antisense and knock-outmutants, rab18, kin1, andrd29B gene
induction insense plants
improved droughttolerance in senseand decreaseddrought tolerance
inantisense andknock-out plants
Iuchi et al., 2001
AVP1(H+Pyrophosphatase)
Arabidopsis reduced stomatal openingincreased saltaccumulation
in thevacuoles
improved salt anddrought tolerance
Gaxiola et al., 2001
BiP Tobacco-sense andantisense plants
sense plants displayed bettersurvival whereas theantisense
plants arehypersensitive to droughtstress
improved droughttolerance
Alvim et al., 2001
Invertase (yeast) Tobacco better photosyntheticefficiency
improved salt tolerance Fukushima et al., 2001.
AtRab7 (Vesicletrafficking protein)
Arabidopsis decreased accumulation ofreactive oxygen
speciesunder salinity
improved salt andosmotic stresstolerance
Mazel et al., 2004
ADR1 (CC-NBS-LRRprotein)
Arabidopsis increased expression ofdehydration responsivegenes
such as ERD11,GST, etc.
improved droughttolerance but not tosalinity or heat stress
Chini et al., 2004
-
30 D. BARTELS AND R. SUNKAR
et al., 1988; Nonami and Boyer, 1990; Spollen et al., 1993).
Thedegree of growth inhibition due to osmotic stress depends on
thetime scale of the response, the particular tissue and species
inquestion, and how the stress treatment was given (rapid or
grad-ual). Growth arrest can be considered as a possibility to
preservecarbohydrates for sustained metabolism, prolonged energy
sup-ply, and for better recovery after stress relief. The
inhibitionof shoot growth during water deficit is thought to
contribute tosolute accumulation and thus eventually to osmotic
adjustment(Osorio et al., 1998). For instance, hexose accumulation
ac-counts for a large proportion of the osmotic potential in the
cellelongation zone in cells of the maize root tip (Sharp et al.,
1990).On the other hand, continuation of root growth under
droughtstress is an adaptive mechanism that facilitates water
uptakefrom deeper soil layers. Similarly, continued root growth
undersalt stress may provide additional surfaces for sequestration
oftoxic ions, leading to lower salt concentration. For example,
salttolerance of barley was correlated with the better root
growthrates coupled with fast development and early flowering
(Munnset al., 2000).
B. Osmotic Stress Affects Cell Division and ElongationPlants
have the unique attribute of modulating their devel-
opment with the prevailing environmental conditions involv-ing
plant hormones, which in turn influence gene expressionprograms.
Cell division is the principal determinant of meris-tem activity
and determines the overall plant growth rate. It hasbeen proposed
that environmental and developmental controlsof growth rate act by
regulating cyclin-dependent kinase (CDK)activity and cell division
(Cockcroft et al., 2000; West et al.,2004). CDKs are a family of
protein kinases, each with a posi-tive regulatory subunit termed a
cyclin and the catalytic subunitCDK (den Boer and Murray, 2000).
CDKs are emerging as keyplayers in regulation of cell division and
are likely to be regulatedat both transcriptional and
post-translational levels in responseto stress. For example, maize
ZmCdc2 (a member of the CDKfamily) was shown to be downregulated by
water stress lead-ing to a decrease in mitotic cell cycling (Setter
and Flanningan,2001). The decrease in cell division in response to
water stressis characterized by lower CDK activity, which is
correlated withtyrosine phosphorylation (Schuppler et al., 1998).
The recentfinding that ABA induces expression of an inhibitor of
CDK(ICK1) links cell division and ABA (Wang et al., 1998). This
isfurther supported by studies of Kang et al. (2002) who
reportedthat expression of the ABA inducible cell cycle regulator
ICK1was increased in transgenic plants overexpressing ABF3 andABF4
resulting in dwarf phenotypes. These mechanisms couldbe responsible
for ABA-dependent cell cycle arrest during os-motic stress in
plants.
Cell expansion is a coordinately regulated process at thewhole
plant level and is influenced by external stimuli includingwater
availability. The rate of cell expansion is mainly deter-mined by
two parameters, cell wall extensibility and cellular
osmotic potential. The enlargement of plant cells involves
con-trol of wall synthesis and expansion, solute and water
transport,membrane synthesis, Golgi secretion, ion transport and
otherprocesses (Cosgrove, 1997). Expansins are a family of plant
pro-teins essential for acid-induced cell wall loosening
(Cosgrove,1997).
Expression of three expansin genes Exp1, Exp5, and ExpB8was
upregulated in the apical region of roots after growth atlow water
potential leading to higher amounts of expansin pro-tein, which is
closely correlated with the root elongation (Wuet al., 2001). These
results are consistent with the hypothesisthat the adaptive wall
loosening and growth maintenance in theapical region of maize roots
are partly due to altered expansingene expression in the root tip
at low water potentials (Wu et al.,2001). Expansin genes in
Craterostigma plantagineum, CpExp1and CpExp3, are upregulated to
different degrees in response todehydration. In addition CpExp1 but
not CpExp3, is also upreg-ulated in response to rehydration (Jones
and McQueen-Mason,2004). These results implicate a role for
expansins during de-hydration and rehydration, particularly in
increasing wall flexi-bility. However, the effect of osmotic stress
on cell enlargementis still not clear and may also involve other
hormones such asauxin, cytokinin, or gibberellins.
III. SIGNAL PERCEPTIONIt is still an open question how plants
sense osmotic stress.
Because no plant molecule has truely been identified as
os-mosensor, scientists oriented themselves to study how yeast
andmicroorganisms sense osmotic stress. In yeast,
hyperosmoticstress is sensed by two types of osmosensors, SLN1 and
SHO1,that feed finally into HOG (high-osmolarity glycerol)
MAPKpathway. High osmolarity induces loss of turgor that leads
con-comitantly to shrinkage of cell volume and an increase in
thedistance between plasma membrane and cell wall. SLN1 is likelyto
sense the change in turgor pressure (Reiser et al., 2003).SLN1 is a
two-component regulatory system, which is also awell characterized
signal transduction element in prokaryotes.The two-component
regulatory system consists of a phospho-relay between three
proteins. The yeast osmosensor SLN1 is afused two-component system
that autophosphorylates a histi-dine residue in the N-terminal
sensor domain and then transfersthe phosphate group to an aspartate
residue in the C-terminal-located response-regulator domain. The
phosphate is transferredto YPD1, which functions as a second
histidine phosphorelayintermediate between SLN1 and the response
regulator SSK1.SSK1 finally feeds into the HOG pathway, which
responds toincreased extracellular osmolarity and is responsible
for os-molyte (glycerol) accumulation (Posas et al., 1996;
Wurgler-Murphy and Saito, 1997). Another yeast membrane protein
in-volved in osmosensing is SHO1, which also converges withthe MAPK
pathway. Which pathway is activated depends onthe osmotic stress
level. In Arabidopsis an SLN1 homologue,AtHK1, was identified; it
can function as an osmosensor in yeast
-
DROUGHT AND SALT TOLERANCE IN PLANTS 31
and complements SLN1-deficient yeast mutants (Urao et
al.,1999).
Recently, NtC7, which is originally identified as a gene thatis
responsive to wounding, has been suggested as another proba-ble
candidate for sensing osmotic stress in plants (Tamura et
al.,2003). NtC7 transcript accumulates rapidly and transiently
notonly in response to wounding but also to salt and osmotic
stress.The NtC7 gene encodes a receptor-like membrane protein
andoverexpression improved osmotic stress tolerance induced
bymannitol but not by NaCl (Tamura et al., 2003). These
resultssuggest that NtC7 may play a role in sensing specifically
os-motic stress. Cre1 (cytokinin response 1) is a cytokinin
receptorin Arabidopsis. Recent experiments implicate Cre1 as
anotherlikely candidate for sensing osmotic stress in plants
(Reiseret al., 2003). Cre1 and Sln1 have similar organizations of
thecytoplasmic histidine kinase and receiver domains. These
ob-servations suggest that several candidates for osmosensors
havebeen proposed, although their role as osmosensors has yet to
bedemonstrated.
IV. SIGNAL TRANSDUCTIONPlants react to external stimuli by
initiating signalling cas-
cade which activate the expression of appropriate responses.
Incontrast to signal perception various components of the
signaltransduction have been identified, although it is largely
unknownhow the different molecules interact with each other and
wherethey are positioned in the complex signalling network.
Thesesignalling pathways comprise a network of protein-protein
re-actions and signalling molecules (for example, ROS, Ca2+
etc.).Reversible phosphorylation of proteins is an important
mecha-nism, by which organisms regulate cellular processes in
responseto environmental cues. In this review, we will consider
severalclasses of protein kinases and phosphatases as signal
transduc-ers that were shown to be involved in osmotic stress
signalling.This will be followed by a description of the role of
calcium assecond messenger molecules.
A. MAPKinase PathwaysProtein phosphorylation is one of the major
mechanisms for
controlling cellular functions in response to external signals.
Themitogen-activated protein kinase (MAPK) cascades are
commonsignalling modules in eukaryotic cells including plants. A
gen-eral feature of MAPK cascades is their composition of
threefunctionally linked protein kinases. MAP kinase activation
re-quires the phosphorylation of conserved threonine and
tyrosineresidues in the so-called TEY (Thr, Glu, Tyr) activation
loop bya specific MAPK kinase (MAPKK). A MAPKK kinase (MAP-KKK)
activates through phosphorylation of conserved threonineand/or
serine residues. At the downstream end of the cascade, ac-tivation
of the cytoplasmic MAPK module often induces translo-cation of the
MAPK into the nucleus where the kinase is ableto activate genes
through phosphorylation of transcription fac-tors (Triesmann,
1996). In some other cases, a given MAPK
may translocate to other sites in the cytoplasm to
phosphory-late specific enzymes or cytoskeletol components
(Robinson andCobb, 1997). By tight regulation of the MAPK
localization andthrough expression of signalling components and
substrates intarget cells, tissues or organs, MAPK pathways can
mediatesignalling of an extracellular stimulus and bring about
specificresponses.
MAPK pathways may integrate a variety of upstream signalsthrough
interaction with other kinases or G proteins (Robinsonand Cobb,
1997). The G proteins often directly serve as couplingagent between
a plasma membranelocated receptor protein thatsenses an
extracellular stimulus and a cytoplasmic module. Gproteins and
kinases in yeast and mammals have been shown toregulate MAPKKKs
(Kyriakis and Avruch, 2001). The actualmechanisms of MAPKKK
activation by osmotic stress in mam-malian cells remain largely
uncharacterized. Mammalian cellsactivate three different MAPKs in
response to osmotic stress:P38, JNK, and ERK5 (extracellular signal
regulated proteinkinases) (de Nadal et al., 2002). When compared
with mam-malian MAPKs, all plant MAPKs have highest homology to
theERK subfamily. Control of gene expression is a major outcomeof
stress-activated MAPK pathways. In mammalian cells, P38(MAPK)
controls the expression of >100 genes, while in
yeast,genome-wide transcription studies revealed that a large
numberof genes (7%) show transient changes in their expression
levelsafter a mild osmotic shock and that the HOG1 MAPK
pathwayplays a key role (de Nadal et al., 2002).
At least 20 MAPK, 10 MAPKK and 60 MAPKKK geneshave been
identified in Arabidopsis on the basis of sequence sim-ilarities
(Riechmann et al., 2000; Ichimura et al., 2002). Giventhe imbalance
in numbers it is likely that the pathways are notbeing linear and
convergence of pathways is expected. We arealready aware that some
MAPKinase-activated pathways appar-ently overlap, i.e., several
genes are induced by more than onestressor (Knight and Knight,
2001). For instance, AtMPK6 andAtMPK3 are activated by osmotic
stress in Arabidopsis, and thetobacco orthologs SIPK (salicylic
acidinducible protein kinase)and WIPK (wound-inducible protein
kinase) are also activatedby biotic stresses indicating the point
of convergence of differ-ent signalling cascades and yet leading to
appropriate responses(Singh et al., 2002). Recent studies suggest
that different stimuliactivate MAPKs to different levels and with
different kinetics,which may encode signal specificity. Thus, the
same MAPKsmay participate in different signalling events (reviewed
in Tenaet al., 2001).
Moderate and extreme hyperosmotic stress activated two dis-tinct
kinases in alfalfa cells, a 46 kDa MAP kinase identified tobe SIMK
under moderate osmotic stress conditions, whereas a38 kDa protein
kinase becomes activated under extreme hyper-osmotic stress
conditions (Munnik et al., 1999). This situationresembles the
operation of the osmosensing SLN1and SHO1pathways in yeast, and
suggests the possible existence of dis-tinct sensors for moderate
and extreme hyperosmotic stress inplants. To determine the upstream
activator of SIMK, Kiegerl
-
32 D. BARTELS AND R. SUNKAR
et al. (2000) used SIMK as bait in yeast two-hybrid screeningand
isolated MAPKK (SIMK kinase). SIMKK encodes a func-tional protein
kinase that specifically activates SIMK in vitro andin vivo. SIMKK
phosphorylates SIMK on the threonine and ty-rosine residues of the
activation loop, establishing that SIMKKis a specific
dual-specificity protein kinase of SIMK (Kiegerlet al., 2000). The
interaction of SIMK with SIMKK was furtherconfirmed in an in vivo
system. SIMK is only partially activatedby NaCl in transfected
parsley protoplasts, whereas it can befully activated by
co-transfection of SIMKK.
In Arabidopsis, AtMEKK1 (a MAPKinase-kinase-kinase)and AtMPK3 (a
MAPKinase) are activated by dehydration, touchand cold (Mizoguchi
et al., 1996). In accordance with a role forAtMPK3 during
dehydration, the closely related alfalfa homo-logue SAMK is also
transcriptionally upregulated upon droughtstress (Jonak et al.,
1996). In Arabidopsis, AtMPK4 and AtMPK6are post-translationally
activated by cold, osmotic stress, andwounding (Ichimura et al.,
2000). Recently, OXI1, a serine/threonine kinase was identified as
a downstream componentof ROS signalling, possibly acting upstream
to the AtMPK3and AtMPK6. OXI1 is transcriptionally upregulated by
H2O2,wounding pathogen attack and osmotic stress (Rentel et
al.,2004). Downstream targets of OXI1 could be AtMPK3 andAtMPK6,
since the OXI1 is required for the activation of MPK3and MPK6.
ADR1, a CC-NBS-LRR gene, that shows homology withthe
serine/threonine protein kinases and that has been
originallyimplicated as a component of disease resistance turns out
to beimportant also for abiotic stress responses (Chini et al.,
2004).The ADR1 transgenic plants display dual but opposite
responsesto dehydration and other abiotic stresses. These plants
exhibitimproved dehydration tolerance but are hypersensitive to
salinityand high temperature stresses. The improved tolerance
appearsto be correlated with the dehydration-responsive gene
expression(Chini et al., 2004). These observations implicate
overlappingbiotic and abiotic stress signalling pathways using
kinases asconverging points of stress signalling pathway and yet
capableof eliciting stress-specific responses.
B. SNF-1-Like Kinases Are Involved in OsmoticStress
Signalling
Another family of protein kinases are the SNF1/AMP-activated
protein kinases, which were first analysed in yeast fromwhere the
name originated (SNF = sucrose-nonfermenting).These kinases may
sense the ATP/AMP ratio and thus controlfluxes between anabolism
and catabolism via transcription ofgenes encoding enzymes related
to carbohydrate metabolism.In plants some members of this group of
kinases are expressedin response to dehydration or ABA. The kinases
range between40 and 50 kDa, and are activated by phosphorylation of
serineor threonine. SNF-1-related protein kinases (SnRKs) have
beenclassified into three families, SnRK1, SnRK2, and SnRK3
withunknown function (Halford and Hardie, 1998). PKABA1 from
wheat (Anderberg and Walker-Simmons, 1992), ARSK1 (Ara-bidopsis
root specific kinase 1) (Hwang and Goodman, 1995),two kinases in
Dunaliella (Yuasa and Muto, 1996), a maize45 kDa kinase (Conley et
al., 1997), SPK3 and SPK4-kinasesfrom soybean (Yoon et al., 1997),
a 42 kDa kinase from to-bacco (Mikolayczyk et al., 2000), a 38 kDa
kinase in alfalfa(Munnik et al., 1999; Munnik and Meijer, 2001), a
guard cell-localized ABA-activated protein kinase, AAPK from Vicia
faba(Li et al., 2000), Arabidopsis OST1 protein kinase which is
re-lated to AAPK of Vicia faba (Mustilli et al., 2002) are
predictedto belong to this group of kinases and are activated in
responseto osmotic stress. The rice genome appears to encode 10
pro-tein kinases belonging to this class and surprisingly all are
acti-vated by osmotic stress, and three of them (SAPK8, SAPK9,and
SAPK10) are also activated by ABA (Kobayashi et al.,2004). Both
SAPK1 and SAPK2 from rice are highly homol-ogous to PKABA1.
Additionally, these kinases seem to be tran-scriptionally either
upregulated or downregulated by osmoticstress and ABA. The
biological significance of the osmotic- orABA-activation of these
kinases is largely unknown. However,their activation in response to
different abiotic stresses to differ-ent levels implicates a major
role for these kinases.
C. PhosphatasesThe action of the protein kinases is counteracted
by phos-
phatases providing modulation and reversibility of the
phos-phoregulatory mechanism. Phosphatases are classified
accord-ing to their substrate specificity. There are two major
groupsof phosphatases: phosphoprotein (serine/threonine)
phophatases(or PPases) and phosphotyrosine (protein tyrosine
phosphatasesor PTPases). PPases are classified into four groups
(PP1, PP2A,PP2B, and PP2C) based on their biochemical and
pharmaco-logical properties (Cohen, 1989). The PTPases form three
sub-groups: receptor-like PTPases, intracellular PTPases, and
dual-specific PTPases.
Two major families of phosphatases interact with and inac-tivate
HOG1 in yeast: the serine/threonine protein phosphatasetype 2C
(PP2C) and the protein tyrosine phosphatases (PTPases)(reviewed in
de Nadal et al., 2002). Tyrosine specific phos-phatases (PTPases)
play a major role in the regulation of MAPKpathway in yeast
(Shinozaki and Russel, 1995). PTP2 and PTP3are major tyrosine
specific PTPases responsible for dephos-phorylation and
inactivation of HOG1 (Wurgler-Murphy et al.,1997). Expression of
genes encoding PTPases is often upregu-lated by the MAPK pathway,
forming a negative feedback loopfor MAPK regulation (Jacoby et al.,
1997; Wurgler-Murphyet al., 1997). Extrapolating from mammals,
transient and low-level MAPK activation may contribute to stress
tolerance inplants, whereas prolonged and high level activation may
bedetrimental to the organism. Salt stress upregulated
transientlythe AtPTP1 gene in Arabidopsis (Xu et al., 1998), which
isdownregulated by cold stress. This implicates a unique mecha-nism
for AtPTP1 in response to environmental stresses. AtPTP1
-
DROUGHT AND SALT TOLERANCE IN PLANTS 33
dephosphorylates AtMPK4 resulting in a complete loss of en-zyme
activity (Huang et al., 2000). This is important, becauseboth
AtPTP1 and ATMKP4 respond to salt stress (Xu et al.,1998).
The Arabidopsis mutant Atmkp1 is hypersensitive to geno-toxic
stress conditions such as UV-C and MMS (Ulm et al.,2001). AtMKP1
codes for a dual-specificity kinase phosphatase.Disruption of
Atmkp1 has no effect on the plant phenotype undernormal conditions,
however, the mutant exhibits hypersensitiv-ity to stress caused by
xenotoxic stresses but not to other abioticstresses. In wild-type
plants, AtMKP1 is specifically activatedin response to UV and MMS.
This is the only in vivo analysisof a phosphatase available up to
now that has a role in stressconditions.
Protein phosphatases 2Cs are serine/threonine phosphatasesand
their involvement in stress is well studied in fungal mod-els. In
yeast PP2C interacts with a MAP kinase cascade thatcontrols
osmolyte biosynthesis (HOG pathway). Several PP2Cs(Mpcs) from M.
crystallinum have been isolated and studied fortheir tissue,
developmental stage- and stress-specific responses.Out of ten
PP2Cs, four transcripts, MPC2, MPC3, MPC5, andMPC8, are transiently
increased by salt and dehydration, sug-gesting a role for these
PP2Cs during stress (Miyazaki et al.,1999). Arabidopsis PP2Cs
appear to be the largest protein phos-phatase family with 76 genes
(reviewed in Schweighofer et al.,2004). Most of the research is
focused on PP2C action in rela-tion to ABA signalling. PP2C has
been described as abi1 andabi2 Arabidopsis mutants defective in a
PP2C isoform (Leunget al., 1994), which seems to function as a
negative regulatorin a pathway that mediates responses to
environmental stressesinvolving ABA. However, the substrates of
these PP2C in Ara-bidopsis are still unknown. Additional studies
have shown thatguard cells of abi1-1 and abi2-1 plants are
disrupted in ABA acti-vation of hyperpolarization-activated
Ca2+(ICa) channels (Allenet al., 1999; Murata et al., 2001).
Further, experiments on abi1-1 and abi2-1 mutants revealed both
PP2Cs are acting at differ-ent levels of the same pathway, i.e.,
abi1-1 acts upstream andabi2-1 downstream of ABA-induced ROS
production in guardcells (Murata et al., 2001). It was found that
ABI1 can interactwith the ABA-inducible transcription factor ATHB6.
ATHB6promoter-reporter expression was abrogated in abi1-1
mutant,suggesting that ABI1 acts upstream of the transcription
factorATHB6 (Himmelbach et al., 2002).
Recently, the involvement of protein phosphatases in stom-atal
regulation has been demonstrated using pharmacologicalapproaches.
Application of phenylarsine oxide, a specific in-hibitor of protein
tyrosine phosphatase prevented stomatal clo-sure in response to
four stomatal closing signals such as ABA,H2O2, Ca2+ and dark
(MacRobbie, 2002). This suggests thatprotein tyrosine
dephosphorylation is involved mostly down-stream of the Ca2+
signalling which is responsible for stomatalclosure. Stomatal
aperture regulation is dependent on ion ef-flux from guard cells.
Identification of the target protein whosedephosphorylation results
in activation of ion release from the
vacuole will allow us to connect the signalling events in
responseto osmotic stress and ABA.
D. Phospholipid SignallingThe plasma membrane must play an
important role in per-
ceiving and transmitting environmental signals. Osmotic
stressoften leads to altered membrane fluidity and changes in
phos-pholipids have recently been recognized as important
eventsmediating osmotic stress signals in plants (Munnik and
Meijer,2001). The current hypothesis is that phopspholipids are
cleavedby phospholipases, which produce phospholipid-derived
secondmessengers. In plants, like in other organisms four major
classesof phospholipases are distinguished based on their cleavage
site:phospholipase C (PLC), phospholipase D (PLD), and
phospho-lipase A1 and A2 (PL A1 and PL A2) (Wang, 2002).
Phospho-lipid signalling may be regulated through G-proteins and
maybe tightly linked with calcium. The major
phospholipid-derivedsignalling molecules which will be considered
in the context ofosmotic stress are inositol 1,4,5-triphosphate
(IP3 ), diacylglyc-erol (DAG) and phosphatidic acid (PA).
1. Inositol 1,4,5-Triphosphate (IP3)PLC cleaves the phospholipid
phosphatidylinositol 4,5-
bisphosphate (PIP2) into the soluble IP3 and the membranebound
DAG. Osmotic stress rapidly increases PIP2 synthesisin Arabidopsis
(Pical et al., 1999; De Wald et al., 2001) and con-comitantly the
transcript levels increase of PIP5K, aphosphatidylinositol-kinase,
that synthesizes PIP2 (Mikamiet al., 1998). PIP2 is a signal
molecule which in animal cellsleads to K+- desensitization
(Kobrinsky et al., 2000). If thiscould also be confirmed for plant
cells, then ion flux and osmoticstress can be linked via PIP2.
There are reports from several plantspecies that osmotic stress
leads to increased PLC transcript lev-els. One of the Arabidopsis
phospholipase C genes (AtPLC1) isinduced by dehydration, salinity
and low temperature (Hirayamaet al., 1995). Takahashi et al. (2001)
have shown that the hyper-osmotic stress induces a rapid and
transient elevation in IP3levels due to activation of PI-PLC in
Arabidopsis cell cultures.Two genes, VuPLC1 and VuPLC2, were
isolated from drought-tolerant and -sensitive varieties of cowpea.
Whereas VuPLC1is constitutively expressed and decreases under
drought stress,VuPLC2 transcript levels accumulate under water
stress in thetolerant plant and progressively declines in the
sensitive plant(El-Maarouf et al., 2001).
Involvement of PLC-genes in dehydration has also been ob-served
in potato (Kopka et al., 1998). PLC activation leads tothe
synthesis of IP3 and DAG. IP3 then releases Ca2+ from in-ternal
stores (Sanders et al., 1999; Schroeder et al., 2001), whileDAG may
be converted to PA or activates a protein kinase C,which, however,
has not been isolated yet in plants. Hyperos-motic stress increased
IP3 levels were reported in several stud-ies on osmotic stress
(Takahashi et al., 2001; De Wald et al.,2001; Drbak and Watkins,
2000). PI-PLC was also shown to
-
34 D. BARTELS AND R. SUNKAR
be involved in the regulation of stomatal movements.
Drought-induced activation of PI-PLC led probably via IP3 to an
increasein cytosolic Ca2+ in guard cells, which triggers stomatal
closureand thus presents a drought avoidance mechanism (Staxen et
al.,1999).
2. Phosphatidic Acid (PA)Phosphatidic acid (PA) is another
second messenger in animal
cells that can activate PLC and protein kinase C (Munnik
andMeijer, 2001). PA is synthesized through cleavage of
membranephospholipids by PLD. The predicted structure of plant
PLDsdiffer from that of yeast and animals in that most plant
PLDscontain not only the conserved catalytic motifs, but also a
Ca2+-binding domain (a C2 domain) which is not found in the PLDsof
other organisms (Wang, 2001). This feature indicates a
directregulation of PLD by calcium, in addition to changes in
PLDgene expression.
Elevated PLD activity was found to be correlated with
drought-stress in a drought-sensitive cowpea strain, when it was
com-pared with a drought-tolerant strain (El-Maarouf et al.,
1999).Further evidence was provided by Frank et al. (2000) by
isolatingtwo phospholipase D cDNAs from C. plantagineum: CpPLD1is
constitutively expressed and is likely to be involved in
earlyresponses to dehydration, producing the second messenger
phos-photidic acid to amplify the signal after its perception,
while thedehydration-responsive CpPLD2 may be involved in
phospho-lipid metabolism and membrane rearrangements at later
stagesof dehydration. Four PLD genes were analyzed for their
re-sponse to osmotic stress conditions in Arabidopsis (Katagiriet
al., 2001).
Only the transcript levels of PLD were transiently elevatedin
response to dehydration, whereas the expression was rapidlyinduced
by high salt stress. Plants transformed with a PLDpromoter fused to
a reporter gene showed strong induction un-der dehydration, mainly
in vascular tissues of leaves, cotyle-dons and roots. Increased PLD
activity coincides with the el-evation in phosphatidic acid levels
in response to dehydration.Evidence to support that the elevated
phosphatidic acid levelswere linked with stress-increased PLD
activity was providedby antisense PLD transgenic plants, in which
the dehydration-induced production of phosphatidic acid
accumulation was sub-stantially reduced (Katagiri et al., 2001). In
castor bean leavesPLD mRNA and enzymatic activity increased by ABA
treat-ment. Further, an increase was observed in the proportion
ofenzyme activity associated with microsomal membranes (Ryuand
Wang, 1995) implying both enzyme level and its locationare
important factors in the ABA response. Like PLC, PLD hasa role in
the regulation of the stomatal aperture during osmoticstress.
ABA-promoted stomatal closure was shown to be me-diated by guard
cell PLD activity in response to water stress(Jacob et al., 1999).
Antisense suppression of PLD impairsstomatal closure mediated by
ABA or water stress and increaseswater loss in Arabidopsis, whereas
the overexpression has theopposite effect and leads to a decreased
water loss by enhancing
the sensitivity to ABA (Sang et al., 2001). These
experimentsprovided direct evidence for PLD in regulating water
loss. Theinvolvement of PLD in ABA responses obtained further
supportfrom experiments using rice protoplasts. The expression of
sev-eral ABA-related genes was repressed when PLD-derived PAwas
blocked, which indicates that PLD activity is important
forABA-inducible gene expression (Gampala et al., 2001;
2002).Osmotic stress activated PLDs implies a role for PA that is
pro-duced by PLD. The targets of PA in plants are unknown,
however,PIP kinase, PDK (phosphoinositide dependent kinase),
MAPKpathway, MGDG synthetase, K+ channel are possible
targets(Munnik, 2001).
E. Other Signalling Molecules1. Salicylic Acid
It has been established for quite some time that salicylic
acid(SA) plays an important role in the defense response in
manyplant species to pathogen attack. SA mediates the oxidative
burstthat leads to cell death in the hypersensitive response, and
actsas a signal for the development of the systemic acquired
resis-tance (Shirasu et al., 1997). Recently the involvement of SA
inosmotic stress was demonstrated by using an SA-deficient
trans-genic line expressing a salicylate hydroxylase (NahG). Wild
typeseeds germinated in the presence of NaCl or mannitol
showedextensive necrosis in the shoot, but not in the NahG mutant.
Wildtype and NahG behaved similarly during germination at
variousLi+ concentrations, excluding the possible involvement of
ioniccomponents. Greater oxidative damage occurred in
wild-typeseedlings compared with NahG seedlings under NaCl
stress.Methyl viologen treatment resulted in necrotic phenotypes
onlyin wild-type plants. These different observations lend support
tothe hypothesis that SA potentiates the effects of salt and
osmoticstress by enhancing the generation of ROS during
photosynthe-sis (Borsani et al., 2001).
2. Nitric Oxide (NO)Recent studies revealed that NO is an
important signalling
molecule involved in several physiological functions rangingfrom
plant development to defense responses (Wendehenne et al.,2001).
Nitric oxide is a labile free radical that is produced
fromL-arginine by NO synthase in various mammalian cells, whereit
has been shown to be protective against damages caused byoxidative
stress conditions. Exogenous application of NO im-proved water
stress tolerance both in wheat seedlings and indetached leaves
(Mata and Lamattina, 2001). Detached leavespretreated with the NO
releasing agent SNP withstand the im-posed stress by retaining a
higher water content, lower ion leak-age, and greater accumulation
of LEA 3 transcripts compared tocontrol leaves. The drought
tolerance of NO could be attributedto its ability to maintain
higher RWC, decreasing the rate oftranspiration by closing stomata
and the ability to induce geneexpression involved in stress
tolerance.
-
DROUGHT AND SALT TOLERANCE IN PLANTS 35
V. CALCIUM SIGNALLING DURING DEHYDRATIONAND SALT STRESS
In plant cells, calcium functions as a second messenger
cou-pling a wide range of extracellular stimuli to
intracellularresponses (Snedden and Fromm, 1998, 2001). Different
ex-tracellular stimuli elicit specific calcium signatures:
kinetics,amplitude and duration of Ca2+ transients specify the
natureand the intensity of stimulus. To date, three major classes
ofCa2+ sensors have been characterized in plants. These classesare
calmodulin, CDPKs (calcium-dependent protein kinase) andCBLs
(calcineurin B-like proteins) (Yang and Poovaiah, 2003).Several
lines of evidence suggest that all these three classes ofCa2+
sensors are involved in stress signal transduction (Sneddenand
Fromm, 2001; Luan et al., 2002; Zhu, 2000).
The involvement of Ca2+ signalling in response to osmoticand
ionic stress is well documented. NaCl causes a rapid andtransient
increase in cytosolic calcium, that in turn triggers manysignal
transduction pathways, including the regulation of enzy-matic
activity, ion channel activity, and gene expression whichresults in
diverse cellular responses (Snedden and Fromm, 1998,2001) and
mediates salt adaptation (Bressan et al., 1998; Liuand Zhu, 1998;
Serrano et al., 1999). Recent progress providedinsights into how
the response to different osmotic stresses is en-coded in the
spatial and temporal dynamics of the Ca2+ signal(Knight and Knight,
2001). The role of calcium and its dynam-ics were investigated
using either pharmacological approachesor using transgenic plants
which express the calcium reporterprotein aequorin in different
cellular compartments or under thecontrol of promoters with
different responsiveness to environ-mental stimuli. This has
allowed determining calcium fluxeswithin one cell and in different
tissues. For instance, mannnitoland sodium chloride-induced
increased cytosolic [Ca2+] is dueto release of calcium from the
vacuole (Knight et al., 1997).Cell-type specific changes in
cytosolic calcium levels were ob-served in Arabidopsis root cells
in response to drought, salinity,and low temperature (Kiegle et
al., 2000).
A. Calcium-Dependent Protein Kinases (CDPKs)Osmotic
stress-induced CDPKs have been reported from sev-
eral plants (Kawasaki et al., 2001; Seki et al., 2002; Ozturk et
al.,2002). Recent experiments on salt-tolerant and salt-sensitive
ricevarieties have strengthened the importance of CDPKs in
osmoticstress responses. A specific CDPK is induced earlier and its
ex-pression is sustained for longer duration in the tolerant
varietycompared to the sensitive variety (Kawasaki et al., 2001).
Furtherevidence for the involvement of CDPKs in stress was
obtainedfrom studying Arabidopsis CDPKs (Sheen, 1996). Out of
sev-eral CDPKs tested, only AtCDPK1 and AtCDPK1a were ableto
transcriptionally activate selected reporter genes indicatingthat
specific CDPK isoforms mediate the effects of stress.
Inter-estingly, in the salt-tolerant Mesembryanthemum
crystallinumosmotic stress-induced CDPK (McCDPK1) was shown to
inter-act with CSP1 (calcium-dependent protein kinase Substarte
Pro-
tein1), which is a transcription factor belonging to a class of
two-component pseudoresponse regulators (Patharkar and
Cushman,2000). These results establish a role for specific CDPKs in
stress-induced gene expression.
B. Calcium-Binding ProteinsModulation of intracellular calcium
levels is partly regulated
by calcium-binding proteins such as calmodulin, which is
ac-tivated by increased calcium concentrations and then
inducesspecific kinases. The importance of the calcium-binding
pro-teins has first been derived from yeast mutant analysis.
Muta-tions in calmodulin genes of yeast render the yeast cells
sensi-tive to high NaCl concentrations and NaCl-induced genes areno
longer induced, suggesting that calmodulin is involved inthe
NaCl-stress signal transduction pathway (Cunningham andFink, 1996).
The regulation of calcium-binding proteins can oc-cur in a cell- or
tissue-specific manner as the following examplesillustrate.
Ca2+/calmodulin dependent kinases have been shownto be regulated by
salt stress, e.g., PsCCaMK, a Ca2+/calmodulindependent protein
kinase in pea was specifically upregulated byNaCl in roots, while
the shoot kinase was not affected (Pandeyet al., 2002). A family of
calmodulin binding transcription acti-vators were first discovered
in drought-stressed Brassica napusand were subsequently shown to be
only present in multicellularorganisms (Bouche et al., 2002).
Further studies are needed todetermine the importance of this
family during osmotic stress.
Other examples for osmotic stress-activated
calcium-bindingproteins are the Arabidopsis protein AtCP1, the
membrane-associated rice protein OsEFA27, and the Arabidopsis
counter-part RD20 (Frandsen et al., 1996; Jang et al., 1998;
Takahashiet al., 2001). A recently identified calmodulin binding
proteinin Arabidopsis, AtCaMBP25, has been implicated as a
negativeregulator of osmotic stress. AtCaMBP25 is upregulated by
os-motic stress and its overexpression leads to sensitivity to
osmoticor salt stresses, whereas the suppression by antisense
technologyresulted in improved tolerance (Perruc et al., 2004)
suggestingthat AtCaMBP25 is a negative regulator of osmotic and
saltstresses. Further, microarray analysis in these plants will
revealif there is a correlation between the observed phenotypes
andaltered gene expression.
C. Ca2+-Mediated SOS Pathways Are Involvedin Ion Homeostasis
Insight into the underlying mechanisms of external calciumon
cellular responses to salt has been provided by the isolation
ofsalt hypersensitive mutants sos1, sos2 and sos3 from
Arabidopsisthaliana (Zhu, 2003). Cloning and characterization of
SOS geneshas led to the discovery of a novel Ca-dependent pathway
forthe regulation of ion homeostasis and plant salt tolerance
(seelater in Na+ extrusion).
D. Calcineurin B (CBLS) and Osmotic Stress ResponsesDespite the
genetic analysis of the interaction of the SOS 1
to 3 genes, the complexity of the calcium signals requires a
lot
-
36 D. BARTELS AND R. SUNKAR
more mutant analysis which becomes apparent by the fact
thatArabidopsis thaliana has at least 10 calcineuron B-like
(CBL)genes encoding highly similar but functionally distinct
Ca2+-binding proteins (Kudla et al., 1999). Drought, high salt,
coldand wound signals induce AtCBL1 gene transcripts (Kudla et
al.,1999). Both CBL1 and CBL2 respond to light, but CBL2 lacksthe
other responses of CBL1 (Nozawa et al., 2001). Knockoutmutant
plants of SCaBP5/CBL1 displayed hypersensitivity todrought and
salinity, in contrast to CBL1 overexpressing plantsthat showed
drought tolerance (Albrecht et al., 2003). Such ex-pression
patterns suggest that CBL1 and CBL2 have both over-lapping and
specific functions.
E. Ca2+ ATPasesThe major physiological role of Ca2+-ATPases is
to restore
and maintain homeostasis by pumping Ca2+ out of the cytosolto
terminate a signalling event, which is critical in all
eukaryoticcells (Sze et al., 2000). Plant and animal cells utilize
two distincttypes of Ca2+ pumps, identified as type IIA and type
IIB, basedon their protein sequences. NaCl stress has been reported
to in-duce the expression of genes encoding the type IIA Ca2+
pumpin tobacco, tomato, and soybean (Wimmers et al., 1992; Chunget
al., 2000). The consequence of upregulating the Ca2+ pumpin
response to NaCl is not known, but it was speculated that itis
likely to provide an adaptive response such that a stimulatedcell
would acquire an enhanced efflux capacity capable of sub-sequently
decreasing the magnitude or duration of a calcium re-lease to
further exposure to a given stimulus (Chung et al., 2000).The
soybean Ca2+-ATPase1 was induced by NaCl but not byKCl and
mannitol, indicating that specific calcium signals trig-ger an
increase in transcription (Chung et al., 2000). Geisler et
al.(2000a) have cloned and characterized Ca2+-ATPase isoform
4(ACA4), a calmodulin-regulated Ca2+-ATPase from Arabidop-sis. Two
lines of evidence suggest that ACA4 might be part ofthe
Ca2+-dependent signal transduction pathway linked to saltstress:
(1) Arabidopsis seedlings treated with different concen-trations of
NaCl for 24 h showed a dose-dependent increase inACA4 transcript
levels, (2) when N-terminal truncated ACA4was expressed in yeast,
it conferred increased NaCl tolerance toits host (Geisler et al.,
2000b).
VI. TRANSCRIPTIONAL REGULATIONOF GENE EXPRESSION
Stress responses primarily include transcriptional regulationof
gene expression and this depends on the interaction of
tran-scription factors with cis-regulatory sequences. In several
in-stances, the quantity and availability of regulatory proteins
maydepend on their own expression patterns. Such autocatalytic
con-trols may be exerted on the transcriptional,
post-transcriptionalor translational level. Phosphorylation of
regulatory proteins isa major event in controlling the gene
expression in eukary-otes. Therefore, multiple protein-protein
and/or protein-DNAinteractions frequently determine the rate of
transcription by
activation/repression of a promoter under given
environmentalconditions.
Although the cis-elements of dehydration and ABA-responsivegenes
have been intensively studied, our understanding is lim-ited. More
efforts are still needed to identify additional novelcis- and
trans-acting elements that function in ABA-dependentas well as
ABA-independent gene expression. Presently, twoclasses of
drought-responsive DNA elements have been wellcharacterized: the
ABA-responsive element (ABRE) and thedehydration responsive element
(DRE, also referred to as C-repeat; Baker et al., 1994). It is
highly likely that yet unidentifiedosmotic-stress responsive
cis-elements exist in plants.
A. ABA Response Elements (ABREs)Most, but not all of the
dehydration-induced genes are also in-
duced by the application of the phytohormone ABA(Chandler and
Robertson, 1994; Leung and Giraudat, 1998) (fordetails see section
XI). Until now, two ABA-dependent pathwaysare known to mediate gene
expression in plants during osmoticstress. The distinction is
largely based on cis-elements that existin the promoters of
ABA-inducible genes. The ABA-dependentpathways are thought to
mediate the gene expression throughan ABRE-element and b-ZIP
transcription factors (Busk andPages, 1998), while the other
pathway is through a MYC andMYB elements and transcription factors
(Yamaguchi-Shinozakiand Shinozaki, 1993).
Many cis-regulatory elements known as ABA-responsive ele-ments
(ABREs) have been identified. Among them, (C/T)ACGTG(G/T)C motifs
have been reported to function as ABREsin many genes (Guiltinan et
al., 1990; Mundy et al., 1990; In-gram and Bartels, 1996; Busk and
Pages, 1998). The core el-ement of these ABREs is the CACGTG motif
also known asG-box motif, which functions in the regulation of
plant genesstimulated by a variety of environmental signals.
SystematicDNA-binding studies have shown that nucleotides flanking
theACGT core specify the DNA-binding interactions and subse-quent
gene activation (Williams et al., 1992; Izawa et al.,
1993).Nevertheless, in promoters such as CDeT27-45 and
CDeT6-19,isolated from C. plantagineum, G-box related ABREs do not
ap-pear to be major determinants of the ABA or drought
response(Michel et al., 1993, 1994).
Ho and coworkers identified cis-elements called couplingelements
which are active in combination with an ABRE butnot alone (Shen et
al., 1996). In the promoters of the barleygenes HVA22 and HVA1 the
coupling elements CE1 and CE3(ACGCGTGTCCTC) are necessary for
activation by ABA. Dis-section of these promoters defined ABA
response complexes(ABRC) consisting of a coupling element and an
ABRE capa-ble of conferring ABA-inducible transcription. Other
ABREsthat do not belong to the above groups have also been
reported:a Sph element-containing sequence (CGTGTCGTCCATGCAT)of the
maize C1gene (Kao et al., 1996), the MYB and the MYCbinding sites
of the Arabidopsis rd22 gene (Abe et al., 1997),
-
DROUGHT AND SALT TOLERANCE IN PLANTS 37
and a novel element present in the CDeT27-45 gene of C.
plan-tagineum (Michel et al., 1993).
B. The Dehydration Response Element (DRE)The investigation of
dehydration-induced genes in Arabidop-
sis has also revealed ABA-independent signal transduction
path-ways (Shinozaki and Yamaguchi-Shinozaki, 2000). In aba
(ABAdeficient) and abi (ABA insensitive) mutants, several genes
areinduced by dehydration indicating that these genes do not
re-quire ABA for their expression under drought conditions. TheA.
thaliana genes rd29A (also described as COR 78 or LTI78) andrd29B
are differentially induced under conditions of dehydra-tion, salt,
cold stress, or ABA treatment. This multiple expressionpattern
requires at least two cis-acting elements for the rd29Agene. The
9-base pair direct repeat (TACCGACAT), termed thedehydration
responsive element (DRE), functions in the initialrapid response of
rd29A to dehydration, salt, or low temper-ature. The slower ABA
response is mediated by another pro-moter fragment that contains an
ABRE (Yamaguchi-Shinozakiand Shinozaki, 1993, 1994). The DRE is an
essential cis-actingelement for the regulation of rd29A induction
in the ABA-independent response to dehydration in Arabidopsis.
DRE-related motifs have been reported in promoters of several
genesregulated by osmotic and low temperature stress, including
kin1,cor6.6/kin2, and rd17/cor47 in Arabidopsis (Wang et al.,
1995;Iwasaki et al., 1997). A similar motif was also reported in
the pro-moter region of cold and dehydration-inducible cor15A
(Bakeret al., 1994). In a recent study 16 genes containing DRE
(TAC-CGACAT) or DRE-related core motif (CCGAC) were identifiedin
the promoters of drought stress inducible Arabidopsis genes,which
are likely targets of DREB1 or DREB2 transcription fac-tors (see
below) (Seki et al., 2002). In another microarray ex-periment which
analysed DREB1A overexpressing plants Sekiet al. (2001) have
identified 12 target genes of DREB. Amongthem, 11 promoters contain
the CCGAC sequence motif, and6 promoters contain ABRE elements
(Seki et al., 2001), whichimplies additional unidentified stress
responsive cis-elements.
C. Transcription Factors Modulated by Osmotic StressPlants need
a large number of transcription factors governing
proper and strict transcriptional regulation in response to
devel-opmental and environmental cues. Indeed, over 5 percent of
theArabidopsis genome is devoted to encoding more than
1,500transcription factors (Riechmann et al., 2000). Despite the
factthat not many stress-specific consensus sequences were
identi-fied in promoters of stress specific genes transcription
factorsbelonging to different families have been shown to be
modi-fied by stress. To activate or repress transcription,
transcriptionfactors must be located in the nucleus, bind DNA, and
interactwith the basal transcription apparatus. Therefore,
environmentalsignals that regulate transcription factor activity
may affect anyone or a combination of these processes. Regulation
of a tran-scription factor is achieved by reversible
phosphorylation or by
de novo synthesis of transcription factors. The current
analysisof identified transcription factors suggests that different
stresssignalling pathways may overlap or converge at specific
points.
1. Basic Region Leucine Zipper (bZIP) ProteinsBasic region
leucine zipper (bZIP) proteins contain a DNA-
binding domain rich in basic residues and adjacent to a
leucinezipper dimerization domain. bZIPs are a large family of
tran-scription factors in plants and are represented by 75
membersin Arabidopsis (Jacoby et al., 2002). All plant bZIP
proteinsbind to an ACGT core sequence, but the sequences flanking
thecore sequence affect the precise DNA binding. Several
bZIPfactors that bind ABREs have been cloned as candidates
forABA-responsive transcription factors that induce gene
expres-sion by osmotic stress and/or ABA. One bZIP subfamily
hasbeen linked genetically to an ABA response: ABI5 and its
ho-mologs, the ABRE binding factors (ABFs/AREBs). ABRE bind-ing
factors (ABFs)/ABA-responsive element binding (AREBs)proteins
respond at the transcriptional and post-transcriptionallevel to
dehydration and salt stress (Choi et al., 2000; Uno et al.,2000).
ABF3 and ABF4 overexpression in transgenic Arabidop-sis plants
resulted in enhanced drought tolerance accompaniedby decreased
transpiration, suggesting that ABF3 and ABF4 areinvolved in
stomatal closure mediated by ABA (Kang et al.,2002). Promoters of
both ABF3 and ABF4 were found to bemost active in roots and guard
cells, consistent with their rolesin stomatal regulation and water
stress response.
2. Homeodomain-Leucine Zipper Proteins (HD-ZIP)HD-ZIP genes
encode proteins that have only been identi-
fied in plants so far and are thought to regulate
developmentalprocesses and responses to environmental cues. HD-ZIP
pro-teins are characterized by the presence of a DNA-binding
home-odomain with a closely linked leucine zipper motif. The
activityof HD-ZIP resides primarily in the specific DNA-binding
prop-erty of the homeodomain and the ability of the leucine
zipperto mediate protein-protein interaction with other HD-ZIPs.
HD-ZIPs have been described for several plant species such as
Ara-bidopsis, carrot, tomato, rice, sunflower, and C.
plantagineum.Seven HD-ZIP are affected by dehydration in C.
plantagineum,which are placed in different branches of the
signalling network(Deng et al., 2002).
In Arabidopsis three HD-ZIP genes (ATHB-6, ATHB-7, andATHB-12)
are shown to be responsive to dehydration (Sodermanet al., 1996,
1999; Lee and Chun, 1998). Hahb-4 is upregu-lated by drought and
ABA in sunflower roots, stems, and leaves(Gago et al., 2002).
HD-ZIP proteins can form both homo- andheterodimers, which gives
many degrees of freedom in regula-tion processes. Direct targets of
HD-ZIP genes have still to beidentified.
3. Zn-Finger ProteinsA zinc-finger motif represents the sequence
in which cys-
teines and/or histidines coordinate a zinc atom(s) to form
local
-
38 D. BARTELS AND R. SUNKAR
peptide structures that are required for their specific
functions.The zinc-finger motifs are classified based on the
arrangementof the zinc-binding amino acids. They play critical
roles in in-teractions with other molecules. The cDNA clone for
Alfin1 wasisolated by differential screening of salt-tolerant
alfalfa cellsgrown on NaCl (Winicov, 1993). Alfin1 cDNA encodes a
novelmember of putative Zn-finger proteins, which binds to
promotersequences of MsPRP2 in vitro (Bastola et al., 1998).
Overex-pression of the Alfin1 gene in transgenic plants increased
theMsPRP2 transcript levels in root and enhanced NaCl
tolerance,indicating that the Alfin1 gene product regulates MsPRP2
ex-pression in vivo (Winicov and Bastola, 1999). The function
ofMSPRP2 is yet unknown, however, it was predicted that it islikely
to be a cell wall protein with potential as an anchor to
themembrane.
4. AP2/ERF-Type Transcription FactorsThe group of AP2/ERF DNA
binding proteins were iden-
tified in different scientific contexts and the first proteins
de-scribed gave this group its name. AP2/ERF domain proteinsinclude
the DREB or CBF proteins binding to dehydration re-sponse elements
(DRE) or C-repeats. The Arabidopsis genomeencodes 145
DREB/ERF-related proteins (Sakuma et al., 2002).The Arabidopsis AP2
proteins have been classified into fiveclasses based on
similarities in their DNA-binding domains:AP2 subfamily (14 genes),
RAV (related to ABI3/VP1) sub-family (6 genes), DREB subfamily (55
genes; Group A), ERFsubfamily (65 genes, Group B), and others (4
genes) (Sakumaet al., 2002). A major transcriptional regulatory
system is repre-sented by DRE/C-repeat promoter sequences in
stress-activatedgenes and DREBs/CBF factors that control stress
gene expres-sion (Stockinger et al., 1997; Liu et al., 1998). A
C-repeatbinding factor, CBF1, was isolated by screening an
ArabidopsiscDNA expression library using the C-repeat of the cold
stressCOR15a gene as target (Stockinger et al., 1997). The
samefactor and other CBF family members, referred to as
DREBs(Dehydration Responsive Element Binding proteins) have
beenreported from studies of dehydration stress genes (Liu et
al.,1998). A detailed expression analysis showed that these
factorsmay be associated with different physiological conditions.
Forexample, expression of DREB1A, B, C/CBF1, 2, 3 is inducedby low
temperature, DREB1A, DREB1D/CBF4, DREB2A andDREB2B are induced by
salt and dehydration, DREB1F is in-duced by salt, whereas, DREB1E
is induced by ABA (Liu et al.,1998; Shinwari et al., 1998;
Nakashima et al., 2000; Haakeet al., 2002; Sakuma et al., 2002).
Several stress-inducible genessuch as rd29A, Cor6.6, Cor15a and
Kin1 are target genes ofDREBs/CBFs in Arabidopsis and contain
DRE/C-repeat se-quences in their promoters. Seki et al. (2001) have
identified sixnew genes containing both DRE/C-repeat and ABRE
motifs intheir promoters, implying complex regulation of
stress-inducedgenes by ABA-dependent and ABA-independent pathways.
Theemerging picture from functional studies of CBFs/DREBs
indi-cates that CBF genes coordinate both, activation or
repression
of stress responsive genes. Ectopic overexpression of
CBF1/DREB1B and CBF3/DREB1A in Arabidopsis results in the
con-stitutive expression of downstream stress-inducible genes
andenhances freezing tolerance and dehydration/salt tolerances,
re-spectively (Kasuga et al., 1999; Gilmour et al., 2000). On
theother hand, the functional analysis of cbf2/dreb1c knock-out
mu-tant plants revealed somewhat surprising results. Cbf2
mutantplants displayed tolerant phenotype to dehydration and
salin-ity (Novillo et al., 2004). Another interesting feature of
thesemutant plants is enhanced expression of CBF1/DREB1B
andCBF3/DREB1A and the downstream genes such as LTI78, KIN1,COR47,
and COR15A which are known to impart stress toler-ance. These
results imply that CBF2 is a negative regulator ofCBF1 and CBF3
transcription factors and of the correspondingdownstream genes.
Rice AP2 transcription activators, OsDREB1A, OsDREB1B,OsDREB1C,
OsDREB1D and OsDREB2A have been isolated.OsDREB2A was induced by
dehydration and high salinity stress.Overexpression of OsDREB1A in
transgenic Arabidopsis im-proved stress tolerance (Dubouzet et al.,
2003). Two novel DREbinding factors (DBF1 and DBF2) were isolated
from maize(Kizis and Pages, 2002). DBF1 and DBF2 are capable of
bind-ing DRE motifs and are not homologues of DREBs from
Ara-bidopsis. DBF1 is induced by dehydration, salinity, and
ABAwhereas DBF2 is constitutively expressed at low levels and
notresponsive to stress.
The Tsi1 (tobacco stress-inducible gene 1) from tobacco be-longs
to the ERF subfamily (Sakuma et al., 2002) and is inducedby NaCl,
salicyclic acid, and ethylene. Tsi1 was able to bind GCCmotifs in
both ethylene response elements and DRE/CRT se-quences. Tobacco
plants overexpressing Tsi1 showed enhancedresistance to osmotic
stress and pathogen attack (Tsugane et al.,1999). This demonstrates
that overexpression of a single tran-scription factor effects both
biotic and abiotic stress responsessuggesting convergence of both
signalling cascades.
5. Myb-Like ProteinsThe Myb-motif comprises three imperfect
repeats forming a
helix-turn-helixrelated motif. Each repeat contains three
con-served tryptophan residues every 18 to 19 amino acids,
whichpromotes a secondary structure with a functional Myb-domain.In
plants, the first tryptophan of R3 is substituted by phenylala-nine
or isoleucine. This latter amino acid is present in Atmyb2
inArabidopsis and myb-related genes from C. plantagineum cp-MYB7
and cpMYB10 whose expression is upregulated by dehy-dration and ABA
(Urao et al., 1993; Iturriga et al., 1996). Ectopicexpression of
CpMYB10 in transgenic Arabidopsis plants re-sulted in drought and
salinity tolerance (Villalobos et al., 2004).The Myb gene family is
represented by 190 genes in Arabidop-sis (Riechmann et al., 2000).
Atmyb-2 has been identified as apositive regulator of
stress-induced rd22 gene expression (Abeet al., 1997). Consistent
with the role, AtMYB-2 overexpress-ing transgenic plants exhibited
hypersensitivity whereas knock-out mutants showed insensitivity to
ABA (Abe et al., 2003). In
-
DROUGHT AND SALT TOLERANCE IN PLANTS 39
addition, three additional Arabidopsis Myb-like
transcriptionfactors were identified as upregulated by dehydration,
high salin-ity, or cold stress in a microarray study (Seki et al.,
2002).
6. Myc-Like ProteinsMyc-like proteins contain the basic
helix-loop-helix (bHLH)
domain, which is composed of two subdomains: the basic re-gion
(as found in bZIPs) responsible for DNA binding and theHLH
(helix-loop-helix) region for dimerization with
interactingproteins. The Arabidopsis rd22BP1 gene encodes a
Myc-liketranscription factor and is induced by dehydration, high
saltconditions, and ABA. It activates the downstream rd22 geneby
binding to the myc promoter motif (Abe et al., 1997). Theexpression
of rd22BP1 (AtMYC-2) precedes the rd22 gene ex-pression, which
underlines the hierarchic relationship. Overex-pression and
knock-out mutants displayed contrasting pheno-types with respect to
ABA sensitivity, similar to AtMYb-2 (Abeet al., 2003).
Hypersensitivity to ABA was enhanced in trans-genics when both
genes are overexpressed together (Abe et al.,2003), confirming the
cooperation of AtMYB-2 and AtMYC-2interaction in vivo in
stress-activated gene expression.
7. CDT-1A novel gene CDT-1 was isolated from C. plantagineum
us-
ing a T-DNA activation tagging approach. A
gain-of-functionphenotype in C. plantagineum calli was due to the
activationof the CDT-1 gene that confers desiccation tolerance to
calluscells and activates stress genes independent of exogenous
ABA(Furini et al., 1997). However, it is not known whether
CDT-1functions as RNA or as a peptide; recent experiments favor
RNAas functional molecule (Furini and Bartels, unpublished).
Thereis no direct sequence homologue to CDT-1 in Arabidopsis.
VII. ACCUMULATION OF SUGARSAND COMPATIBLE SOLUTES
Almost all organisms, ranging from microbes to animalsand
plants, synthesize compatible solutes in response to os-motic
stress (Burg et al., 1996). Compatible solutes are non-toxic
molecules such as amino acids, glycine betaine, sugars, orsugar
alcohols. They do not interfere with normal metabolismand
accumulate predominantly in the cytoplasm at high con-centrations
under osmotic stress (reviewed in Chen and Murata,2002). These
molecules may have a primary role of turgor main-tenance but they
may also be involved in stabilizing proteinsand cell structures
(Yancey et al., 1982). Initially it was thoughtthat compatible
solutes have their main role in osmotic adjust-ment, but there is
increasing discussion of other roles (Serraj andSinclair, 2002).
The accumulation of these solutes per se maynot be important for
osmotic stress tolerance but the metabolicpathways may have
adaptive value (Hasegawa et al., 2000). Afurther hypothesis is that
compatible solutes are also involved inscavenging reactive oxygen
species (Shen et al., 1997a, b; Honget al., 2000; Akashi et al.,
2001; Chen and Murata, 2002). Engi-
neering the synthesis of compatible solutes has been a
relativelysuccessful approach to obtain stress tolerant plants (see
Table 1).
A. SugarsSeveral physiological studies suggested that under
stress con-
ditions nonstructural carbohydrates (sucrose, hexoses, and
sugaralcohols) accumulate although to varying degree in
differentplant species. A strong correlation between sugar
accumulationand osmotic stress tolerance has been widely reported,
includingtransgenic experiments (Ramanjulu et al., 1994a; Abd-El
Bakiet al., 2000; Gilmour et al., 2000; Streeter et al., 2001; Taji
et al.,2002).
The increase in sugars mostly results in increased starch
hy-drolysis, which requires activities of hydrolytic enzymes.
Resur-rection plants and seeds of many higher plants are good
examplesfor a link of carbohydrate (in particular sucrose)
accumulationand the acquisition of stress tolerance (Hoekstra et
al., 2001;Phillips et al., 2002). This is illustrated with the
example ofthe resurrection plant C. plantagineum, which contains a
highamount of the unusual sugar octulose (an 8-carbon sugar),
whichis rapidly converted into sucrose during dehydration
(Bianchiet al., 1991). This sugar conversion is coupled with the
in-creased expression of sucrose synthase and sucrose
phosphatesynthase genes (Ingram et al., 1997). The currant
hypothesis isthat sugars either act as osmotica and/or protect
specific macro-molecules and contribute to the stabilization of
membrane struc-tures. Sugars may protect cells during desiccation
by formingglasses (Black and Pritchard, 2002). Sugars are also
thought tointeract with polar headgroups of phospholipids in
membranesso that membrane fusion is prevented. It is unknown
whethersugars fulfil this function on their own or in conjunction
withother molecules such as LEA proteins. Many seeds
accumulateconsiderable amounts of raffinoase-type oligosaccarides
(RFOs)such as raffinose and stachyose which are thought to play a
rolein the acquisition of desiccation tolerance. Despite many
studiesthe link between the presence of these carbohydrates and
des-iccation tolerance has not always been confirmed.
Galactinolsynthase catalyzes the first committed step in the
biosynthesisof RFOs. Overexpression of galactinol synthese resulted
in accu-mulation of galactinol and raffinose under controlled
conditionsand improved drought tolerance (Taji et al., 2002),
confirmingthe importance of these sugars under stress
conditions.
A sugar which has been shown to contribute to
desiccationtolerance in yeast and some nematodes is trehalose. In
higherplants substantial amounts of trehalose were identified in
tworesurrection plants Myrothamnus flabellifolia and
Sporobolusstapfianus (Phillips et al., 2002). It has been reported
that manyhigher plants possess trehalase activity, which is perhaps
re-sponsible for rapid degradation of any trehalose
synthesized.Arabidopsis thaliana has at least one gene which
encodestrehalose-6-phosphat