Protein sensors and transducers of cold and osmotic stress in cyanobacteria and plants

ISSN 0026-8933, Molecular Biology, 2007, Vol. 41, No. 3, pp. 427–437. © Pleiades Publishing, Inc., 2007.Original Russian Text © G.V. Novikova, I.E. Moshkov, D.A. Los, 2007, published in Molekulyarnaya Biologiya, 2007, Vol. 41, No. 3, pp. 478–490.

427

INTRODUCTION

Abiotic stress factors affecting photosynthesizingorganisms suppress many biochemical reactions andphysiological responses at various levels. When stressexposure is nonlethal, organisms activate the expres-sion of various genes and synthesize proteins and spe-cific metabolites essential for the survival under thenew conditions.

The first step of acclimation is the perception of thestress signal and its transmission to the regulatoryregions of relevant genes. Organisms and cells havespecific protein sensors to perceive environmentalchanges and proteins to transmit these signals.

As a result of their sessile life style, plants haveacquired multiple mechanisms allowing their adapta-tion and acclimation to particular climatic conditionsof the habitat. Studies of the molecular mechanisms ofplant adaptation to unfavorable factors is of both the-oretical and applied interest, because the capability towithstand unfavorable climatic conditions directlyaffects the diversity of plant species and the productiv-ity of crops, influencing the human food supply.

Modern biochemical and molecular-biologicaltechniques make it possible to study the mechanismsof plant adaptation and acclimation at the molecularlevel. To date, the genome has been sequenced com-pletely for

Arabidopsis

and rice and, partly, fortobacco, lotus, and about a dozen of other plants andcyanobacteria. DNA microarrays have been devel-oped to study the changes in genome expression inresponse to various stress factors. Many earlier

unknown genes have been found to respond tochanges in ambient temperature. Yet the functions arestill unknown for most of these genes, and the molec-ular mechanisms responsible for the perception oftemperature signals and the plant response to lowertemperatures are poorly understood. This review sum-marizes the data on sensor proteins of photosynthesiz-ing organisms. These proteins perceive environmentalchanges and trigger the expression of genes essentialfor adaptation to stress conditions.

TWO-COMPONENT SYSTEMS FOR SENSINGAND TRANSMISSION OF STRESS SIGNALS

A typical two-component system of signal percep-tion and transduction includes a His kinase (Hik) andthe corresponding response regulator (Rre). Pairs of

hik

and

rre

regulatory genes are close in the genomeand form operons in most eubacteria, such as

Esche-richia coli

and

Bacillus subtilis.

Hik perceives theenvironmental changes via its sensor domain, whichleads to autophosphorylation of His in the conservedHis kinase domain [1]. Then, the phosphoryl group istransferred from Hik to Asp in a conserved domain ofthe corresponding Rre. Phosphorylated Rre changesits conformation and binds to the promoter region oftarget genes, activating (positive regulation) orrepressing (negative regulation) their transcription.Two-component regulatory systems are found inprokaryotes (including cyanobacteria), fungi (includ-ing yeasts), plants, and primitive animals [2] but not inhigher animals and humans. Ser/Thr and Tyr protein

Protein Sensors and Transducers of Cold and Osmotic Stressin Cyanobacteria and Plants

G. V. Novikova, I. E. Moshkov, and D. A. Los

Timiryazev Institute of Plant Physiology, Russian Academy of Sciences, Moscow, 127276 Russia; e-mail: [email protected]

Received and accepted for publication December 26, 2006

Abstract

—Genome-wide analysis of gene expression at the transcriptional level with DNA microarrays iden-tified almost all genes induced by particular stress in cyanobacteria and plants. Adaptation to stress conditionsstarts with the perception and transduction of the stress signal. A combination of systematic mutagenesis ofpotential sensors and transducers with genome transcription profiling allowed significant progress in under-standing the mechanisms responsible for the perception of stress signals in photosynthesizing cells. The reviewconsiders the recent data on the cyanobacterial and plant signaling systems perceiving and transmitting the cold,hyperosmotic, and salt stress signals.

DOI:

10.1134/S0026893307030089

Key words

: cold stress, histidine kinases, hyperosmotic stress, response regulators, salt stress, cold stress, sen-sors

UDC 579.23''315

TO THE 40th ANNIVERSARY OF

MOLEKULYARNAYA BIOLOGIYA

MOLECULAR ADAPTATION

428

MOLECULAR BIOLOGY

Vol. 41

No. 3

2007

NOVIKOVA et al.

kinases mostly act as sensors in eukaryotes, in partic-ular, in higher animals and higher plants [3].

The chromosome of freshwater unicellular cyano-bacterium

Synechocystis

harbors 44

hik

genes and 42

rre

genes [4]. In addition, three

hik

and three

rre

genesare in

Synechocystis

plasmids [5]. Unlike in

E. coli

and

B. subtilis

, the

hik

and corresponding

rre

genes donot form operons in the

Synechocysts

genome. Hence,libraries of specific knockouts, including 42

hik

mutants and 41

rre

mutants, were obtained to studythe specific functions of individual regulatory genes(http://kazusa.or.jp/cyano/Synechocys-tis/mutants/index.html). Complete segregation ofmutant chromosomes was not achieved for

hik2

,

hik11, hik26, rre23, rre25

, and

rre26.

Recent studies revealed that the

Synechocystis

Hikproteins are involved in perceiving and transmittingthe signals of cold stress [6, 7], phosphorus deficiency[8], high ocmolarity [9, 10], nickel deficiency [11],manganese deficiency [12], and high NaCl concentra-tions [13, 14]. In this review, we focus on the geneexpression regulation by two-component systems incold and hyperosmotic stress.

Hik33–Rre26 Two-Component System Regulates the Expression of Genes Induced

at Low Temperatures in

Synechocystis

The cold stress sensor is Hik33 in

Synechocystis

[6] and its analog DesK in

B. subtilis

[15]. Gene

expression profiling with DNA microarrays in the

∆

hik33

mutant showed that Hik33 regulates 21 out of35 genes induced at low temperatures (Fig. 1). Theother 14 cold-inducible genes are probably controlledby some other regulatory systems. Screening of the

rre

mutants indicated that Rre26 mediates the effect ofHik33 on gene induction at lower temperatures. Muta-tions of the two corresponding genes block the induc-tion of the same gene sets at low temperatures, sug-gesting a Hik33–Rre26 two-component system forcold stress perception and cold signal transduction[16].

Two-Component Systems Involved in Perception and Signal Transduction in Hyperosmotic

and Salt Stress in

Synechocystis

Although hyperosmotic and salt stress exposuresare often considered the same, they are far from beingso. Hyperosmotic stress quickly expels water from thecytoplasm; as a result, the cytoplasm volume isreduced and the ion concentration within the cellincreases. Salt stress similarly reduces the cytoplasmvolume at the first step [17, 18], which is followed bya rapid increase in

Na

+

and

Cl

–

concentrations withinthe cell [19] owing to activation of

Na

+

/K

+

–Cl– chan-nels.

Systematic analysis of gene expression via

hik

and

rre

mutagenesis combined with transcription profilingwith DNA microarrays identified five two-componentHik–Rre systems involved in regulating gene expres-sion in response to hyperosmotic stress: Hik33–Rre31, Hik34–Rre1, Hik16–Hik41–Rre17, Hik10–Rre3, and Hik2–Rre1 [10]. The signal transductionpathways and target genes controlled by individualregulatory systems are shown in Fig. 2.

The Hik33–Rre31 system is responsible for theinduction of 11 genes. The Hik10–Rre3 system regu-lates the expression of only one gene,

htr

A, whichcodes for Ser protease. In the system of three proteins,Hik16 and Hik41 probably act as a sensor and anintermediate signal transducers and are His kinases,while Rre17 is responsible for the transcriptionalinduction of two genes with unknown functions,

sll0939

and

slr0967.

The Hik34–Rre1 system regu-lates 19 genes, including many genes for heat shockproteins. The Hik2–Rre1 system is responsible for theinduction of

sig

B, which codes for an alternative

σ

subunit of RNA polymerase, and four other geneswith unknown functions. In addition, 14 out of 48genes induced by hyperosmotic stress are not regu-lated by any of the five Hik–Rre systems identified.The mechanisms controlling stress-inducible tran-scription of these genes remain obscure [10].

Studies of the systems involved in the perceptionand signal transduction in salt stress (0.5 M NaCl)implicated the above five two-component systems in

Total

: 15

crhRrlpArbp1cbiOmutSdesBslr0082

slr1927sll1611slr0955slr0236sll0494slr1436slr1974

Total

: 21

ndhD2hliAhliBhliCfusycf39sigD

feoBcrtPslr1544sll1483sll1541sll0086slr0616

slr1747ssr2016slr0400sll1911slr0401sll0815sll1770

Rre26

Hik33

?

Cold stress

Plasma membrane

Fig. 1

Perception of cold stress by cyanobacterium

Synechocystis.

The relevant two-component regulatory sys-tem includes His kinase Hik33 and response regulatorRre26; its target genes are shown. ?, an unidentified systemof cold-dependent transcriptional induction.

MOLECULAR BIOLOGY

Vol. 41

No. 3

2007

SENSORS AND TRANSDUCERS OF STRESS SIGNALS 429

regulating the salt stress response [14]. The systemsmostly regulate the expression of the same genes, butto a different extent. However, there are exampleswhere individual Hik–Rre systems induce differentgenes in salt and hyperosmotic stress (Fig. 2), whichis difficult to explain in terms of the existing conceptof the function of two-component regulatory systems.For instance, Hik33 senses both salt and hyperosmoticstress and transmits the signal to Rre31 in either case.Yet Rre31 controls different gene sets in differentstress conditions; the mechanism of such differenti-ated regulation is unclear. Another open question isthe mechanism whereby one sensor kinase transmitsthe signal to different response regulators, as is thecase with Hik33, which interacts with Rre26 in coldstress and with Rre31 in salt and hyperosmotic stress.It is likely that some additional factors interact withthe identified regulatory components to determine sig-nal specificity. This assumption needs experimentalverification.

OSMOTIC STRESS IN PLANTS

Water deficiency (drought), extreme temperature,and/or high salinity of soil are the severe abiotic fac-tors that limit the growth and development of plantsand may cause a complete elimination of susceptiblespecies. The extraordinary capability of plants to sur-vive osmotic stress is related to their ability to altergrowth and development (duration of the life cycle,inhibited growth of above-ground organs, activatedgrowth of roots), the osmotic and turgor pressures(changes in transport of potentially toxic ions: uptake,secretion, and vacuolar compartmentalization), andmetabolism (synthesis of compatible osmolites).Some of these responses are directly induced by theprimary stress signal, while some others result fromsecondary signals, triggered by primary stress. Suchsecondary signals include phytohormones, abscisicacid (ABA), ethylene, reactive oxygen species (espe-cially

H

2

O

2

), and intracellular secondary messengers(Ca

2+

). Because of the complex nature of osmoticstress, studies of the perception and transduction ofthe primary stress signal are hindered by the fact thatplants are exposed to stress through several develop-mental stages. Stress factors, such as drought orhyperosmosis, do not arise abruptly. These factors actgradually, and experience in hormone studies showsthat the observed pattern depends on the rate and dura-tion of stress development rather than on the stressseverity. Hence, the question arises as to whether lab-oratory simulation of stress conditions adequatelyreflects the actual events. Since experiments are per-formed with intact plants or organs, it is clear that thestress response is determined by the programs realizedin cells of different types. For instance, cells of differ-ent types may have specific transcription factors ordiffer in the relative contents of particular sensors.

When considering the perception and transduction ofstress signals in plant cells, it is necessary to takeaccount of multiple cross-talks of signaling pathways[20].

With this in mind, below we discuss the data onsensors perceiving osmotic stress and componentsinvolved in the corresponding signal transductionpathways.

Plant Osmosensors

As all living organisms, plant cells perceive andprocess information with the use of various receptorsexposed on the cell surface. Membrane proteinkinases of two classes—receptor-like Ser/Thr kinases(RLKs) [21] and receptor His kinases (HKs) [22]—act as receptors in osmotic stress.

It became clear in the late 1990s that plant cellshave stress sensors.

Arabidopsis thaliana

was found

ÇÒ„Ó: 8 ÇÒ„Ó: 3 ÇÒ„Ó: 1

Total

: 11

fabGhliAhliBhliCgloAsigDsll1483slr1544ssr2016ssl3446sll1541

Total

: 19

hspAclpB1sodBhtpGdnak2spkHgroESgroEL1groEL2dnaJ

sll0846slr1963slr0959sll1884slr1603slr1945ssl2971slr1413

Total

: 5

sigBsll0528slr1119slr0852ssr3188

Total

: 2

sll0939slr0967

Total

: 1

htrA

Total

: 14

rlpArepAglpDsll1863sll1862sll1772slr0581

ssr1853slr0112sll0294slr0895slr1501sll0293sll0470

Rre31 Rre1 Rre1 Rre17 Rre3

?Hik34 Hik2 Hik41

Hik10Hik16Hik33

Hyperosmotic stress

Total

: 7

Salt stress

Hik33

Hik34 Hik2 Hik41

Hik16 Hik10

?

Rre31 Rre1 Rre1 Rre17 Rre3

hliAhliBhliCsigDsll1483slr1544ssr2016

hspAhik34clpB1pbpsodBdnaK2groEL2sll0846slr1603slr1963

slr1915

Total

: 19

slr1916sll1844ssl2971slr0095slr1192sll1022slr1413sll1107

sigBdnaJrimlsll0528slr0959slr1686slr0852ssr3188

sll0939slr0967sll0938

htrA

Total

: 26

ndhRglpDycf21ggtBggpSmenGsuhBsll1863sll1862

sll1722slr1687slr0895slr1704ssl3044sll1652slr0581slr1738sll1621

sll2194sll1620ssr2153slr1894slr1501slr0082slr1932sll1236

Plasma membrane

Plasma membrane

(‡)

(b)

Total

: 8

Total

: 1

Total

: 3

Fig. 2.

Synechocystis

two-component systems perceivingand transmitting the signals of (a) hyperosmotic and (b) saltstress.

430

MOLECULAR BIOLOGY

Vol. 41

No. 3

2007

NOVIKOVA et al.

to possess AtHK1, which is homologous to

Saccharo-myces cerevisiae

SLN1. In yeast cells, SLN1 sensesthe changes in osmolarity and triggers a two-compo-nent system, which activates MAPK (mitogen-acti-vated protein kinase) pathways.

Although direct evidence for the osmosensor roleof AtHK1 in

A. thaliana

is still lacking,

AtHK1

expression is known to increase in salt stress (250 mMNaCl) or a decrease in temperature to

4°ë

[23]. In the

sln1

∆

sho1

∆

double mutant (the two affected proteinsact as osmosensors in yeasts),

AtHK1

expressioninhibited the salt-sensitive

sln1

∆

sho1

∆

phenotype[23]. It is important to note that AtHK1 did not inter-act with any of the five

A. thaliana

response regulatorsexamined [22]. Yeast SHO1, which acts as an osmo-sensor at a high osmolarity, contains four transmem-brane domains and the SH3 domain exposed into thecytoplasm and lacks enzymatic activity. SHO1 ini-tiates a signaling pathway typical of higher eukaryoticcells [24, 25].

At a high osmolarity, HK CRE1, which acts as amembrane cytokinin receptor in

A. thaliana

, substi-tutes HK SLN1 in

sln1

∆

yeast cells when activatedwith its ligand zeatin [26]. Such CRE1 activity wasalso observed when osmotic stress was simulated withsorbitol and, consequently, the turgor pressure andcell volume decreased [27]. CRE1 and SLN1 are sim-ilar in domain organization but are highly homologousonly in the cytoplasmic kinase and sensor domains.The questions arise as to how CRE1 perceives thechanges in turgor pressure and which domains areindispensable for the osmosensor function of acti-vated CRE1. Reiser et al. [27] assumed that CRE1mediates a physical contact of the cell wall with theplasma membrane. The osmosensor function requiresa unique combination of the periplasmatic and trans-membrane domains of CRE1, as well as the integrityof the total periplasmatic domain. The above propertyof CRE1 is of immense importance, suggesting thatplants have an osmosensing system that is structurallyand functionally similar to its yeast analog. Thisassumption needs verification with plant cells.

Tamura et al. [28] recently cloned

NtC7

(

Nicotianatabacum

), whose expression increased rapidly (10 min)and transiently (within 60 min) when tobacco leaveswere treated with 200 mM NaCl and/or 500 mM man-nitol. Transgenic tobacco plants overexpressing

NtC7

were tolerant of osmotic stress regardless of their age,while germination of transgenic seeds and the growthof transgenic seedlings were inhibited by NaCl andLiCl. This finding indicates that the NtC7 function inosmotic stress is not related to the maintenance of ionhomeostasis.

NtC7

codes for a transmembrane protein of 308amino acid residues (33.9 kDa). Its N-terminal regionis similar to the RLK receptor domain, but NtC7 lacksthe kinase domain exposed into the cytoplasm. How

does NtC7 transmit the signal to the downstream sig-naling components? Note that, apart from the kinasedomain, NtC7 is homologous to tomato leaf Cf-9, atransmembrane receptor of the

Cladosporium fulvum

avr-9 elicitor. It is commonly accepted that the Cf-9C-terminal region, which is exposed into the cyto-plasm, is responsible for pathogen-induced signaltransduction. In the cytoplasm, Cf-9 probably inter-acts with Ser/Thr kinase Pto, which is involved in sig-nal transduction upon a pathogen attack. It is possiblethat NtC7 similarly interacts with its partners in thecytoplasm to transmit the signal about changes inosmolarity. Further studies will identify the NtC7partners.

It is important that the primary receptors of abioticsignals be searched among RLKs, which are charac-teristic of higher eukaryotic cells. The RLK familyharbors about 60% of all kinases and is encoded by2.5% of genes in

Arabidopsis

, while there are only11 HKs in this plant. The RLK-family proteins vary indomain structure and have highly diverse sequencesof the extracellular domains [29], suggesting percep-tion of various signals. In planta studies of the RLKfunction will elucidate whether RLK signaling is spe-cific for particular signals or RLKs are involved in thecross-talk of signaling pathways triggered by manydifferent signals.

Plant Protein Kinases and Protein Phosphatases

Reversible phosphorylation of proteins is animportant mechanism regulating the cell response, inparticular, to osmotic stress. Among all relevantenzymes, mitogen-activated protein kinases (MAPKs)are a subject of intense studies. MAPKs are of specialinterest partly because the well-studied

S. cerevisiae

osmotic stress signaling pathway involves the HOG1(high osmolarity glycerol response 1) MAPK cascade,which mediates signal transduction from the receptorto transcription factors triggering the stress responsegenetic program.

There are only six different MAPK genes in the

S.cerevisiae

genome and 20 MAPK genes in the A.thaliana genome [30]. Some of these genes werefunctionally characterized not only in A. thaliana, butalso in alfalfa Medicago sativa and tobacco N.tabacum [31], although studies of A. thaliana MAPKsare most successful.

MAPKs MPK3, MPK4, and MPK6 are activated inresponse to low temperatures and osmotic and saltstress in A. thaliana [32, 33]. The kinetic and extent ofactivation vary among different MAPKs, suggestingtheir independent roles in the stress response. Forinstance, mutant A. thaliana seedlings with inacti-vated MPK4 are tolerant of hyperosmotic stress [34];i.e., AtMPK4 can negatively regulate the response to

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

SENSORS AND TRANSDUCERS OF STRESS SIGNALS 431

hyperosmicity, in contrast to the positive regulatorsMPK3 and MPK6.

Of about 1000 A. thaliana protein kinase genes,more than 100 genes code for kinases involved inMAPK cascades (modules), including 60 MAPKkinase kinases (MAPKKKs), 10 MAPK kinases(MAPKKs, or MKKs), and 20 MAPKs (or MPKs)[30]. This proportion suggests that MAPKKs act bothas a main multifunctional component integrating thesignals from upstream MAPKKKs and as a branchingpoint for activation of downstream MPKs.

It was demonstrated that AtMPK4/6 are involvedin transmitting the osmotic signal and MAPKKAtMKK1 is activated in osmotic stress [35] and phos-phorylates and activates AtMPK4 in vitro [36]. Thesefindings gave grounds to assume that A. thaliana has aMAPK cascade that is activated in osmotic stress andincludes AtMPK4, AtMKK1, and AtMEKK1 (MAP-KKK).

Recent studies confirmed this assumption [37]. Itwas shown using transient expression in A. thalianaleaf mesophyll protoplasts and the yeast two-hybridsystem that MAPKK AtMKK2 is phosphorylated inresponse to salt and cold stress and that its activatedphosphorylated form phosphorylates AtMPK4 andAtMPK6. MAPKKK MEKK1 was identified as apotential activator of AtMKK2. Upon simultaneousexpression of AtMKK2, AtMPK4, and AtMPK6, con-stitutively active AtMEKK1 significantly activatedboth MAPKs in protoplasts.

The identification of the first MAPK module (Fig. 3)activated in response to salt and cold stress is an

important event, which would meet great enthusiasmif not for the following. First, it is unclear which sen-sor perceives the stress signal and how this informa-tion is translated into a “language” understandable tothe MAPK cascade. Second, what is the mechanismensuring the specificity of signal transduction fromAtMEKK1 to downstream AtMKK2 and then toAtMPK4/6? This is important to establish, becauseAtMEKK1 was earlier implicated in the MAPK cas-cade involved in the A. thaliana response to flagellin[38]. A well-known mechanism ensuring transductionof individual signals involves scaffold proteins, whichare found in plants as well as in other organisms. Yetit is unclear whether their function is similar to that ofscaffold proteins of other eukaryotes. Third, A. thalianaplants overexpressing activated MKK2 display con-siderable changes in expression of 152 genes, includ-ing the marker genes of salt and cold stress [39, 40].Hence, it is important to determine whether the tran-scription factors that regulate the expression of genesinvolved in the response to abiotic stress act as MAPKsubstrates in planta. Analysis of 1690 A. thaliana pro-teins with the use of protein microarrays identified39 proteins as AtMPK6 substrates and 48 as AtMPK3substrates, including some transcription factors [41].To study the physiological significance of the sub-strates identified, it is necessary to map the phospho-rylation sites and to verify the results in vivo.

Intracellular Ca2+ concentration changes inresponse to osmotic, salt, and cold stress in plant cells.Calcium-dependent protein kinases (or calmodulin-like domain protein kinases (CDPKs)) were foundonly in plants and green algae. In A. thaliana, the

Signal

Sensor

åÄêäää

åÄêää

åÄêä

Targetgenes

Negativeregulator

Hyperosmosis

AtMEKK1

AtMKK1

AtMPK3/6

NaCl

AtHK1

AtMEKK1

AtMKK2

AtMPK4/6

Drought Cold

AtMEKK1

AtMKK1

AtMPK4/6

AtMKK1

AtMPK4/6

AtMKP1AtPTP1

AtMKP1AtPTP1

Hyperosmosis Hyposmosis

NtC7

NtNPK1

NtMEK2

NtSIPK

NtMEK2

NtWIPK

Hyperosmosis

NaCl

ColdDrought

MsSIMKK

MsSIMK

MsNPK1

MsPRKK

MsSAMK

MsMP2C

M. sativaN. tabacumA. thaliana

Fig. 3. MAPK involved in osmotic stress signal transduction in plant cells. A scheme of the signaling pathway is shown on the left.AtHK1 and NtC7 presumably act as osmosensors. MAPK modules whose function in stress was demonstrated experimentally areframed. Species: At, Arabidopsis thaliana; Nt, Nicotiana tabacum; Ms, Medicago sativa. Dual-specificity (AtMKP1), phosphoty-rosine (AtPTP1), and PP2C (MsMP2C) MAPK-specific protein phosphatases act as negative regulators.

432

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

NOVIKOVA et al.

Ser/Thr-CDPKs family is one of the largest familiesand includes 34 unique CDPKs and CDPK-relatedprotein kinases. It should be emphasized that CDPK isa Ca2+ sensor possessing an enzymatic activity. Recentstudies focused on the specific functions of individualCDPKs in signal transduction. Such works seem log-ical in view of the subcellular location of CDPKs [42]and the diversity of their potential substrates(enzymes of carbohydrate and nitrogen metabolism,stress proteins, membrane transporters, ion channelproteins, cytoskeletal proteins, and transcription fac-tors).

A role of CDPKs as positive regulators of signaltransduction in osmotic stress can be inferred from theinduction of individual CDPK genes. It was found,indeed, that cold and salt tolerance of rice increaseswith the increasing expression of OsDCPK7, an

ortholog of AtCDPK1 [43]. Plants overexpressingOsCDPK7 displayed induction of some genesinvolved in the response to cold, but not to salt, stress.Thus, these findings did not confirm the aboveassumption. Moreover, the same cytosolic OsCDPK7proved to play a role in two different pathways, sug-gesting an unknown mechanism of signaling specific-ity.

In leaves of facultative halophyte Mesembryanthe-mum crystallinum, McCPK1 expression increasesquickly and transiently in salt and osmotic stress, thelatter being more effective [44]. McCPK1 phosphory-lates CSP1 (CDPK substrate protein 1) in the presenceof Ca2+ in vitro and, probably, in vivo. Note that CSP1expression is not regulated in osmotic stress. CSP1 isconstitutively located in the nucleus and acts as a tran-scription factors, belonging to the family of two-com-ponent response pseudoregulators. McCPK1 is foundin the nucleus of cells exposed to salt stress and dehy-dration and is associated with the plasmalemma inintact leaf cells. These findings indicate that, in hyper-osmotic stress, McCPK1 phosphorylates CSP1 in thenucleus, which changes the ability of CSP1 to regulatethe expression of genes involved in the hyperosmoticstress response. When stress is eliminated, McCPK1returns to the plasmalemma, allowing the cell torespond to new stress. Thus, McCPK1 perceives thesignal when on the plasmalemma and transmits infor-mation into the nucleus to activate stress-induciblegenes.

Regulation of ion homeostasis under stress condi-tions is important for the formation of salt tolerance inplants. Molecular-genetic analysis of the A. thalianasos mutants (salt-overlay-sensitive) identified thecomponents (SOS1, SOS2, and SOS3) involved in thesignal transduction pathway that transmits informa-tion about a stress-induced increase in intracellularCa2+ and functions to restore ion homeostasis in thecell (Fig. 4) [45]. Studies of the Ca2+ sensor SOS3 andits partner protein kinase SOS2 allowed the identifica-tion of the A. thaliana genes coding for eight SOS3-like Ca2+ sensor/binding proteins (SCaBPs) and 23SOS2-like protein kinases (PKSs). Like SOS3, allother SCaBPs lack enzymatic activity and bind Ca2+,utilizing their so-called EF hands. It is of particularinterest that SOS3 may serve as a Na+ sensor, sinceNa+ binding was demonstrated for EF-hand proteins[46].

SCaBPs and their partner PKSs are capable ofnumerous interactions in vitro [47]. Phosphorylationis probably a key event in the regulation of PKS activ-ity, because PKS is enzymatically inactive in theabsence of SCaBP. PKS-phosphorylating kinases areas yet unidentified, but it is known that protein phos-phatase 2C (PP2C) can dephosphorylate PKS [48].

The physiological substrates of PKSs are unknownin most cases. SOS1 (plasmalemma Na+/H+ anti-

SCaBP (SOS3)

PPI

ABI2

PPI

FISL

SCaBP (SOS3)

Ca2+

FISLCa2+

Saltstress

ëa2+

Kinase domain

Kinase domain

InactivePKS (SOS2)

ActivatedPKS (SOS2)

Transporterphosphorylation

Ion homeostasis

Salt tolerance

Fig. 4. Role of SOS2, SOS3, and ABI2 in the A. thalianacell response to salt stress. Salt stress generates a Ca2+ sig-nal, which is perceived by the Ca2+ sensor SCaBP (SOS3).In the absence of stress, protein kinase S (PKS, SOS2) isinactive owing to the intramolecular interaction between theN-terminal kinase and C-terminal receptor domains. Theregulatory domain includes the FISL motif, which isinvolved in SCaBP (SOS3) binding to PKS (SOS2), and thePPI motif, involved in the interaction with protein phos-phatase 2C (ABI2). The activation loop (black bar) is withinthe kinase domain of PKS (SOS2). The Ca2+–SKBP(SOS3) complex binds to the FISL domain and thereby acti-vates PKS (SOS2). Activated PKS (SOS2) phosphorylatesion transporters responsible for ion homeostasis. ABI2either inactivates PKS (SOS2) or dephosphorylates thetransporters.

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

SENSORS AND TRANSDUCERS OF STRESS SIGNALS 433

porter) is an exception; its function prevents the cellfrom accumulating excess Na+ in salt stress. On recentevidence, PKS (SOS2) can regulate vacuolar trans-porters such as CAX1 (H+/Ca2+ antiporter) [49] andAtNHX1 (Na+/H+ antiporter) [50], the changes inantiporter activity being independent of the presenceor activity of SCaBP (SOS3).

Several protein phosphatases interact with PKSs(Fig. 4). The PKS C-terminal region harbors the con-served PPI motif, which is necessary and sufficient forthe interaction with PP2C [48]. Mutations of the PPImotif prevent the PKS (SOS2)–PP2C interaction.PP2C has the PKI domain and utilizes it to interactwith PKSs. The PP2C–PKS interaction is impossiblewhen the PKI domain is altered by a mutation. ThePKS (SOS2)–PP2C (ABI2) interaction is distorted inthe abi2-1 A. thaliana mutant (ABI codes for PP2C),but the mutant is resistant to salt stress and is insensi-tive to ABA [48]. It is possible that SOS2 and ABI2control each other’s phosphorylation or regulate phos-phorylation of a common substrate. Another possibil-ity is that the ABA and SOS signaling pathways cross-talk at the ABI2 level. If so, SOS2 acts as a scaffoldprotein, rendering SOS3 and ABI2 in a complexwhose specific function is associated with salt resis-tance in response to the Ca2+ signals arising in saltstress. Yet this is only one of all possible scenarios.

Considerable progress has recently been made instudies of SCaBP (SOS3) and PKS (SOS2). However,such studies are far from being completed and manyquestions are still open. What are the functions ofother members of their families and which substratesdo they affect? Which protein kinases phosphorylateSer/ Thr, or Tyr in PKS? What is the mechanism sus-taining the function of the SCaBP–PKS signalingmodule in organisms other than A. thaliana?

Increasing attention in studies of the signal trans-duction pathways is attracted by protein phosphatasesresponsible for only transient activation of proteinkinase-involving pathways. In contrast to proteinkinases with their highly similar structures, proteinphosphatases form a less abundant family and consid-erably differ from each other in structure, biochemicalproperties, and subcellular location. In the context ofthis review, it is of special interest to consider Tyrphosphatases (PTPs), dual specificity protein phos-phatases (DsPTPs), and PP2C, since these enzymesdephosphorylate MAPKs, including those involved insignal transduction in osmotic stress. An A. thalianamutant with an increased sensitivity to short-wave UVirradiation (UV-C) and methyl methanesulfonate(MMS) made it possible to identify MKP1 (MAPKphosphatase 1), which codes for DsPTP regulating thesensitivity to DNA-damaging agents. The recessivemkp1 mutants were hypersensitive to UV-C and MMSbut tolerant of salt stress [51]. The potential partner ofMKP1 is MPK6, while the interaction with MPK3 and

MPK4 is weaker [51]. These findings indicate thatMKP1 controls the NaCl stress signal transductionpathway, which involves MPK6, but it is still unclearwhether MPK6 plays a central role in A. thalianaadaptation to salt stress and is the only target ofMKP1. On the other hand, NaCl-induced activation ofAtMPK6 and AtMPK4 does not change in the mkp1mutant, while the mpk4 mutant displays a higher resis-tance to hyperosmotic stress [34]. These data make itpossible to assume that MKP1 is involved in osmoticsignal transduction, functioning in the MAPK cascadewith MAPKs other than MPK4/6. In addition toMKP1, AtPTP1 dephosphorylates Tyr residues inAtMPK4 and AtMPK6 in vitro [52]. It is possible thatMKP1 and/or AtPTP1 dephosphorylate AtMPK4/6 instress, but the conditions allowing protein phos-phatases from different groups inactivate the samesubstrates are a subject of further studies.

In mammalian and yeast cells, Ser/Thr proteinphosphatase PP2C inactivates MAPK. Interestingly,alfalfa PP2C (MP2C) produced in yeast cells provedto suppress the lethal phenotype caused by expressionof constitutively active MAPKKK STE11, whichinduces the HOG1 MAPK cascade. This givesgrounds to think that MP2C inactivates the MAPKcascade involved in signaling in osmotic stress inalfalfa.

It was found, indeed, that PP2C (MP2C) dephos-phorylates and inactivates stress-inducible MAPKSIMK (a homolog of AtMPK6) upon coexpression oftheir genes in Petroselinum crispum protoplasts [53].This finding is of fundamental importance, suggestivea selective interaction of SIMK and MP2C in vivo.Hence, it is reasonable to assume that MP2C nega-tively regulates the SIMK signaling pathway involvedin the salt stress response in alfalfa. The A. thalianagenome harbors 69 PP2C genes. Among all PP2Cenzymes, ABI1 and ABI2 deserve special attention.The role of these protein phosphatases as negative reg-ulators of the ABA-dependent signaling pathwaysleading to stomatal closure in drought has recentlybeen reviewed [54].

The nature of the abi1-1 mutation, identified genet-ically, is unclear. The abi1-1 mutant is ABA-insensi-tive and displays a lower activity of AtPP2C. Identifi-cation of a potential substrate of alfalfa MP2C eluci-dated, to a certain extent, the mechanism whereby theabi1-1 mutation affects the PP2C function. An MP2Cmutation equivalent to abi1-1 reduced the phos-phatase activity of MP2C without affecting its interac-tion with SIMK. Mutant MP2C was capable of inacti-vating SIMK [53]. It was assumed that a decrease inphosphatase activity does not influence the PM2C–SIMK interaction, which is mediated by the MP2CKIM motif. In mammals and yeasts, this conservedmotif is found in proteins interacting with MAPKsand protein phosphatases are incapable of MAPK

434

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

NOVIKOVA et al.

dephosphorylation in the absence of KIM. Indeed,SIMK is not inactivated by PP2C ABI2, which lacksKIM in contrast to ABI1 [53]. Although this assump-tion is appealing, it cannot be excluded that the cata-lytic domain of PP2C is involved in determining itssubstrate specificity.

Transcription Factors

As plant genomes are sequenced and studies withDNA microarrays performed, data are accumulatingconcerning the changes in plant transcriptomes inresponse to osmotic stress. A fundamental conclusionbased on these data is that genes induced in cold andsalt stress are similarly induced in response to waterdeficiency, suggesting a cross-talk of the relevant sig-naling pathways, as is the case in cyanobacteria [16,55].

In particular, signals interact and are integrated atthe level of the target gene promoters, which containthe cis-acting dehydration-responsive element (DRE),cold-responsive element (CRT), and ABA-responsiveelement (ABRE) (Fig. 5). These molecular switchesinteract with transcription factors, which ensureexpression of both genes involved in the early

response to osmotic stress and genes responsible forstress adaptation.

A NAC-domain transcription factor family is spe-cific to plants. Proteins of the family play diverseroles, being involved in the responses to various fac-tors from pathogens to abiotic stress [56]. In total, 26NAC genes were identified in sugar cane Saccharumofficinarum. Of these, SsNAC23 is induced in coldstress, while its expression in water deficiency is tran-sient [57]. Rice OsNAC6 (an ortholog of SsNAC23) isinduced both in response to osmotic stress and treat-ment with exogenous ABA [58].

As all eukaryotes, plants have transcription factorsof the bZIP family. Rice low temperature-inducibleprotein 19 (LIP19), a bZIP factor, does not bind toDNA [58]. Like animal Fos, LIP19 binds to OsOBF1,a potential new transcription factor of the bZIP family[59]. In contrast to LIP19, OsOBF1 is synthesized atan optimal temperature and binds to specific DNAsites. When temperature decreases, OsOBF1 binds toLIP19, and the resulting heterodimer interacts withunidentified promoters. Since LIP19 is unstable atnormal temperatures, genes activated by the het-erodimer are inactivated when plants are exposed athigher temperatures. It is unknown whether otherbZIP factors act similarly. In A. thaliana, bZIP ABRE-binding factor (ABF)/ABRE-binding protein (AREB)activate the RD29A promoter by binding to its ABRE(Fig. 5) and this activation is enhanced by ABA-dependent phosphorylation of the transcription factor.

The A. thaliana MYC- and MYB-family transcrip-tion factors also function as activators of the ABA-dependent pathway leading to expression of dehydra-tion-inducible (RD) genes such as RD22 and AtADH1[60]. Transgenic plants overexpressing AtMYC andAtMYB are hypersensitive to exogenous ABA. Suchplants display a higher tolerance of osmotic stress, buttheir growth is somewhat slower than in wild-typeplants possibly because of a higher ABA content.

Apart from the transcription factors binding to thecis elements in the promoters of stress response genes,activation of transcription requires some extra cofac-tors, which determine the expression level. CRT-bind-ing factor 1 (CBF)/DRE-binding factor 1 (DREB1)act as transcription factors to ensure the expression ofgenes responding to osmotic stress in an ABA-inde-pendent manner. A specific feature of CBF is its rapid(15 min) and transient accumulation in plants exposedto low temperatures. This is quite conceivable, assum-ing that a constitutive transcription factor is activatedat low temperatures to induce the expression of theCBF gene [61]. Such a new regulator was recentlyidentified and termed the inducer of CBF expression(ICE1). ICE1 is a MYC-like transcription factor thatis phosphorylated in plants exposed to a low positivetemperature and activates transcription of the CBF3gene [60, 61]. CBF1/2 activate transcription of a set of

ICEr2

NaCl, droughtCold

ICE1

ICE1P

ICEr1 DREB1/CBF

CBF/DREB1 DREB2 CBF4 MYC/MYB ABF/AREB/bZIPP?

ÄBA

MYCR/MYBR

DRE/CRT ABRE TATATATA

Fig. 5. Regulation of the expression of ABA-dependent andABA-independent genes involved in the cell response toabiotic stress in A. thaliana. Abiotic factors (cold, drought,and NaCl) activate the expression of the stress responsegenes via stress-inducible transcription factorsCBF/DREB1 and DREB2 and ABA-inducible bZIP tran-scription factors ABF/AREB and CBF4. Transcriptionalactivator ICE1 binds to cis element ICEr1, which regulatesthe expression of CBF/DREB1 together with ICEr2. ABA-independent transcription factors CBF/DREB1 bind to thecis elements of DRE/CRT to ensure the expression of thecold stress response genes. Transcription factors are shownwith ovals; cis elements are shown with black bars. Tran-scription factors whose activity is regulated via phosphory-lation are indicated with encircled P. Unknown posttransla-tional modification is indicated with encircled ?.

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

SENSORS AND TRANSDUCERS OF STRESS SIGNALS 435

genes whose promoters contain CRT/DRE. The roleof transcription factors CBF and ICE1 has beenintensely studied in silico [62]. Although the ICE1-dependent pathway is certainly important for theinduction of cold-responsive (COR) genes [62], itshould be noted that a substantial role can also beplayed by other transcription factors. For instance, theA. thaliana hos9-1 mutant (high expression of osmot-ically responsive genes) is hypersensitive to freezing.Although its CBF expression is intact, the hos9-1mutant is incapable of cold adaptation. It seems that A.thaliana HOS9 controls constitutive tolerance of lowtemperatures. Since the expression of CBF-controlledgenes is not altered in the hos9-1 mutant, it is possibleto assume that HOS9 is necessary for cold toleranceand functions in a CBF-independent pathway [63].

Returning to Fig. 5, it appears that each stress sig-nal is transmitted via a specific pathway. Yet experi-ments with transient expression showed that CBF andDREB function together with ABF to increase theRD29A expression. Hence, it is impossible to excludeinterplay of ABA-dependent and ABA-independentpathways. Moreover, overexpression of the transcrip-tion factor genes, e.g., CBF, confers resistance notonly to low temperatures but also to changes in osmo-larity, testifying again to interplay of signaling path-ways.

PROSPECTS OF FURTHER STUDIES

By definition, a primary receptor recognizes itsown ligand with high specificity. A noncovalentreversible binding of the ligand to its receptor modu-lates the function of a primary signal transductionpathway, inducing a primary response. This is true forsome ligands such as phytohormones. However, theligand can hardly be identified in the case of abioticstress (cold, drought). Hence, the main problem is thatplant primary sensors of osmotic signals are as yetunidentified. The Ca2+ signal induced in osmoticstress is decoded and transmitted by the Ca2+ sensors,which await further investigation. Although the role ofprotein kinases and, in particular, MAPKs has beendemonstrated, in particular, for osmotic stress signaltransduction, it is still poorly understood, first, howputative primary sensors transmit information to pro-tein kinases. A second question is how protein kinasesare connected with transcription factors. As for pro-tein kinases, questions again prevail over answers. Inparticular, little is known on their substrates in planta.Studies with protein microarrays have yielded data onthe potential MAPK substrates [41], but the physio-logical significance of these data should be verified invivo.

Phosphoproteomics will certainly develop as a glo-bal approach, although this needs genome sequencing

in many plants, expensive high-tech equipment, andexpert evaluation.

Studies with DNA microarrays have identifiedhundreds of genes whose expression changes inresponse to osmotic stress. The main current problemsare to determine their regulons, identify the specifictranscription factors, and elucidate the role of the reg-ulons in sustaining plant growth and developmentunder adverse conditions.

As -omics techniques will be used to study theplant tolerance of abiotic stress in the nearest future,important data will be obtained for the perception andtransduction of signals determining stress resistance.

It should be emphasized that an important contri-bution to further progress can be made by studies ofextremophilic plants such as Thellungiella halophila,whose genome is about twice as large as theA. thaliana genome. In contrast to A. thaliana,T. halophila is capable of surviving extreme osmoticstress. Studies involving T. halophila as a model in theproject “Integrating International Research of PlantAbiotic Stress Tolerance Using Arabidopsis RelativeModel Systems (ARMS): Thellungiella halophila”will substantially improve the understanding of themechanisms responsible for plant adaptation to con-tinuously changing environmental conditions.

ACKNOWLEDGMENTS

This work was supported by the Russian Founda-tion for Basic Research (project nos. 06-04-48581,05-04-50883, 05-04-49643) and the program Molecu-lar and Cell Biology of the Presidium of the RussianAcademy of Sciences.

REFERENCES

1. Stock J., Da R.S. 2000. Signal transduction: Responseregulators on and off. Curr. Biol. 10, R420–R424.

2. Koretke K.K., Lupas A.N., Warren P.V., Rosenberg M.,Brown J.R. 2000. Evolution of two-component signaltransduction. Mol. Biol. Evol. 17, 1956–1970.

3. Widmann C., Gibson S., Jarpe M.B., Johnson G.L. 1999.Mitogen-activated protein kinase: conservation of athree-kinase module from yeast to human. Physiol. Rev.79, 143–180.

4. Mizuno T., Kaneko T., Tabata S. 1996. Compilation ofall genes encoding bacterial two-component signaltransducers in the genome of the cyanobacterium,Synechocystis sp. strain PCC 6803. DNA Res. 3, 407–414.

5. Kaneko T., Nakamura Y., Sasamoto S., Watanabe A.,Kohara M., Matsumoto M., Shimpo S., Yamada M.,Tabata S. 2003. Structural analysis of four large plasmidsharboring in a unicellular cyanobacterium, Synechocys-tis sp. PCC 6803. DNA Res. 10, 221–228.

6. Suzuki I., Los D.A., Kanesaki Y., Mikami K., Murata N.2000. The pathway for perception and transduction of

436

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

NOVIKOVA et al.

low-temperature signals in Synechocystis. EMBO J. 19,1327–1334.

7. Suzuki I., Kanesaki Y., Mikami K., Kanehisa M.,Murata N. 2001. Cold-regulated genes under control ofthe cold sensor Hik33 in Synechocystis. Mol. Microbiol.40, 235–244.

8. Suzuki S., Ferjani A., Suzuki I., Murata N. 2004. TheSphS–SphR two component system is the exclusive sen-sor for the induction of gene expression in response tophosphate limitation in Synechocystis. J. Biol. Chem.279, 13234–13240.

9. Mikami K., Kanesaki Y., Suzuki I., Murata N. 2002. Thehistidine kinase Hik33 perceives osmotic stress and coldstress in Synechocystis sp. PCC 6803. Mol. Microbiol.46, 905–915.

10. Paithoonrangsarid K., Shoumskaya M.A., Kanesaki Y.,Satoh S., Tabata S., Los D.A., Zinchenko V.V., Hayashi H.,Tanticharoen M., Suzuki I., Murata N. 2004. Five histi-dine kinases perceive osmotic stress and regulate distinctsets of genes in Synechocystis. J. Biol. Chem. 279,53078–53086.

11. Lopez-Maury L., Garcia-Dominguez M., Florencio F.J.,Reyes J.C. 2002. A two-component signal transductionsystem involved in nickel sensing in the cyanobacteriumSynechocystis sp. PCC 6803. Mol. Microbiol. 43, 247–256.

12. Yamaguchi K., Suzuki I., Yamamoto H., Lyukevich A.,Bodrova I., Los D.A., Piven I., Zinchenko V., Kane-hisa M., Murata N. 2002. A two-component Mn2+-sens-ing system negatively regulates expression of the mnt-CAB operon in Synechocystis. Plant Cell. 14, 2901–2913.

13. Marin K., Suzuki I., Yamaguchi K., Ribbeck K., Yama-moto H., Kanesaki Y., Hagemann M., Murata N. 2003.Identification of histidine kinases that act as sensors inthe perception of salt stress in Synechocystis sp. PCC6803. Proc. Natl. Acad. Sci. USA. 100, 9061–9066.

14. Shoumskaya M.A., Paithoonrangsarid K., Kanesaki Y.,Los D.A., Zinchenko V.V., Tanticharoen M., Suzuki I.,Murata N. 2005. Identical Hik–Rre systems are involvedin perception and transduction of salt signals and hyper-osmotic signals but regulate the expression of individualgenes to different extents in Synechocystis. J. Biol.Chem. 80, 21531–21538.

15. Aguilar P.S., Hernandez-Arriaga A.M., Cybulski L.E.,Erazo A.C., Mendoza D. 2001. Molecular basis of ther-mosensing: A two-component signal transduction ther-mometer in Bacillus subtilis. EMBO J. 20, 1681–1691.

16. Murata N., Los D.A. 2006. Histidine kinase Hik33 is animportant participant in cold signal transduction incyanobacteria. Physiol. Plant. 126, 17–27.

17. Kanesaki Y., Suzuki I., Allakhverdiev S.I., Mikami K.,Murata N. 2002. Salt stress and hyperosmotic stress reg-ulate the expression of different sets of genes in Syne-chocystis sp. PCC 6803. Biochem. Biophys. Res. Com-mun. 290, 339–348.

18. Shapiguzov A., Lyukevich A.A., Allakhverdiev S.I.Sergeyenko T.V., Suzuki I., Murata N., Los D.A. 2005.Osmotic shrinkage of cells of Synechocystis sp. PCC6803 by water efflux via aquaporins regulates osmo-stress-inducible gene expression. Microbiology SGM.151, 447–455.

19. Allakhverdiev S.I., Nishiyama Y., Miyairi S., Yama-moto H., Inagaki N., Kanesaki Y., Murata N. 2002. Salt

stress inhibits the repair of photodamaged photosystemII by suppressing the transcription and translation ofpsbA genes in Synechocystis. Plant Physiol. 130, 1443–1453.

20. Perk S.H., Zarrinpar A., Lin W.A. 2003. Rewiring MAPkinase pathways using alternative scaffold assemblymechanisms. Science. 299, 679–691.

21. Shiu S.-H., Bleecker A.B. 2001. Plant receptor-likekinase gene family: Diversity, function, and signaling.Sci. Signal Transduction Knowledge Environ. 113,RE22.

22. Urao T., Yamaguchi-Shinozaki K., Shinozaki K. 2001.Plant histidine kinases: An emerging picture of two-component signal transduction in hormone and environ-mental responses. Sci. Signal Transduction KnowledgeEnviron. 113, RE18.

23. Urao T., Yakubov B., Satoh R., Yamaguchi-Shinozaki K.,Seki M., Hirayama T., Shinozaki K. 1999. A transmem-brane hybrid-type histidine kinase in Arabidopsis func-tions as an osmosensor. Plant Cell. 11, 1743–1754.

24. Raitt D.C., Posas F., Saito H. 2000. Yeast Cdc42 GTPaseand Ste20 PAK-like kinase regulate Sho1-dependentactivation of Hog1 MAPK pathway. EMBO J. 19, 4623–4631.

25. Reiser V., Salah S.M., Ammerer G. 2000. Polarizedlocalization of yeast Pbs2 depends on osmostress, themembrane protein Sho1 and Cdc42. Nature Cell Biol. 2,620–627.

26. Inoue T., Higuchi M., Hashimoto Y., Seki M., Kobayashi T.,Takara S., Shinozaki K., Kakimoto T. 2001. Identifica-tion of CRE1 as a cytokinin receptor from Arabidopsis.Nature. 409, 1060–1063.

27. Reiser V., Raitt D.C., Saito H. 2003. Yeast osmosensorSln1 and plant cytokinin receptor Cre1 respond tochanges in turgor pressure. J. Cell Biol. 161, 1035–1040.

28. Tamura T., Hara K., Yamaguchi Y., Koizumi N., Sano H.2003. Osmotic stress tolerance of transgenic tobaccoexpressing a gene encoding a membrane-located recep-tor-like protein from tobacco plants. Plant Physiol. 131,454–462.

29. Shiu S.-H., Bleecker A.B. 2003. Expansion of the recep-tor-like kinase/Pelle gene family and receptor-like pro-teins in Arabidopsis. Plant Physiol. 132, 530–543.

30. MAPK Group. 2002. Mitogen-activated protein kinasecascades in plants: A new nomenclature. Trends PlantSci. 7, 301–308.

31. Jonak C., Okresz L., Bogre L., Hirt H. 2002. Complex-ity, cross talk and integration of plant MAP kinase sig-nalling. Curr. Opin. Plant Biol. 5, 415–424.

32. Ichimura K., Mizoguchi T., Yoshida R., Yuasa T., Shi-nozaki K. 2000. Various abiotic stresses rapidly activateArabidopsis MAP kinases ATMPK4 and ATMPK6.Plant J. 24, 655–665.

33. Droillard M.-J., Boudsocq M., Barbier-Brygoo H., Lau-riere C. 2002. Different protein kinase families are acti-vated by osmotic stress in Arabidopsis thaliana cell sus-pension. Involvement of the MAP kinases AtMPK3 andAtMPK6. FEBS Lett. 527, 43–50.

34. Droillard M.-J., Boudsocq M., Barbier-Brygoo H., Lau-riere C. 2004. Involvement of MPK4 in osmotic stressresponse pathways in cell suspension and plantlets ofArabidopsis thaliana: Activation by hypoosmolarity andnegative role in hyperosmolarity tolerance. FEBS Lett.574, 42–48.

MOLECULAR BIOLOGY Vol. 41 No. 3 2007

SENSORS AND TRANSDUCERS OF STRESS SIGNALS 437

35. Matsuoka D., Nanmori T., Sato K., Fukami Y., Kik-kawa U., Yasuda T. 2002. Activation of AtMEK1, anArabidopsis mitogen-activated protein kinase kinase,in vitro and in vivo: analysis of active mutants expressedin E. coli and generation of the active form in stressresponse in seedlings. Plant J. 29, 637–647.

36. Huang Y., Li H., Gupta P., Morris P., Luan S., Kieber J.J.2000. AtMPK4, an Arabidopsis homolog of mitogen-activated protein kinase, is activated in vitro by AtMEK1through threonine phosphorylation. Plant Physiol. 122,1301–1310.

37. Teige M., Scheikl E., Eulgem T., Doczi R., Ichimura K.,Shinozaki K., Dangl J.L., Hirt H. 2004. The MKK2 path-way mediates cold and salt stress signaling in Arabidop-sis. Mol. Cell. 15, 141–152.

38. Asai T., Tena G., Plotnikova J., Willmann M.R., Chiu W.-L.,Gomez-Gomez L., Boller T., Ausubel F.M., Sheen J.2002. MAP kinase signalling cascade in Arabidopsisinnate immunity. Nature. 415, 977–983.

39. Kreps J.A., Wu Y.J., Chang H.S., Zhu T., Wang X.,Harper J.F. 2002. Transcriptome changes for Arabidop-sis in response to salt, osmotic, and cold stresses. PlantPhysiol. 130, 2129–2141.

40. Lee B.H., Henderson D.A., Zhu J.K. 2005. The Arabi-dopsis cold-responsive transcriptome and its regulationby ICE1. Plant Cell. 17, 3155–3175.

41. Feilner T., Hultschig C., Lee J., Meyer S., Immink R.G.H.,Kenig A., Possling A., Seitz H., Beveridge A., Scheel D.,Cahill D.J., Lehrach H., Kreutzberger J., Kersten B.2005. High throughput identification of potential Arabi-dopsis mitogen-activated protein kinases substrates.Moll. Cell. Proteom. 4, 1558–1568.

42. Cheng S.H., Willmann M.R., Chen H.C., Sheen J. 2002.Calcium signalling through protein kinases: The Arabi-dopsis calcium-dependent protein kinase gene family.Plant Physiol. 129, 469–485.

43. Saijo Y., Hata S., Kyozuka J., Shimamoto K., Izui K.2000. Over-expression of a single Ca2+-dependent pro-tein kinase confers both cold and salt/drought toleranceon rice plants. Plant J. 23, 319–327.

44. Chehab E.W., Patharker O.R., Hegeman A.D., Taybi T.,Cushman J.C. 2004. Autophosphorylation and subcellu-lar localization dynamics of a salt- and water deficit-induced calcium-dependent protein kinase from iceplant. Plant Physiol. 135, 1430–1446.

45. Zhu J.-K. 2002. Salt and drought stress signal transduc-tion in plants. Annu. Rev. Plant Biol. 53, 247–273.

46. Ward J.M., Hirschi K.D., Sze H. 2003. Plants pass thesalt. Trends Plant Sci. 8, 200–201.

47. Gong D., Guo Y., Schumaker R.S., Zhu J.-K. 2004. TheSOS3 family of calcium sensors and SOS2 family ofprotein kinases in Arabidopsis. Plant Physiol. 134, 919–926.

48. Ohta M., Guo Y., Halfter U., Zhu J.-K. 2003. A noveldomain in the protein kinase SOS2 mediates interactionwith the protein phosphatase 2C ABI2. Proc. Natl. Acad.Sci. USA. 100, 11771–11776.

49. Cheng N.-H., Pittman J.K., Zhu J.-K., Hirschi K.D.2004. The protein kinase SOS2 activates the ArabidopsisH+/Ca2+ antiporter CAX1 to integrate calcium transportand salt tolerance. J. Biol. Chem. 279, 2922–2926.

50. Qiu Q.-S., Guo Y., Quintero F., Pardo J.M., Schu-maker K.S., Zhu J.K. 2004. Regulation of vacuolar

Na+/H+ exchange in Arabidopsis thaliana by salt-overly-sensitive (SOS) pathway. J. Biol. Chem. 279, 207–215.

51. Ulm R., Ichimura K., Mizoguchi T., Peck S.C., Zhu T.,Wang X., Shinozaki K., Paszkowski J. 2002. Distinctregulation of salinity and genotoxic stress responses byArabidopsis MAP kinase phosphatase 1. EMBO J. 21,6483–6493.

52. Gupta R., Luan S. 2003. Redox control of proteintyrosine phosphatases and mitogen-activated proteinkinases in plants. Plant Physiol. 132, 1149–1152.

53. Meskiene I., Baudouin E., Schweighofer A., Liwosz A.,Jonak C., Rodriguez P., Jelinek H., Hirt H. 2003. Stress-induced protein phosphatase 2C is a negative regulatorof a mitogen-activated protein kinase. J. Biol. Chem.278, 18945–18952.

54. Schweighofer A., Hirt H., Meskiene I. 2004. Plant PP2Cphosphatases: Emerging functions in stress signaling.Trends Plant Sci. 9, 236–243.

55. Shinozaki K., Yamaguchi-Shinozaki K., Seki M. 2003.Regulatory network of gene expression in the droughtand cold stress response. Curr. Opin. Plant Biol. 6, 410–417.

56. Olsen A.N., Ernst H.A., Lo Leggio L., Skiver K. 2005.NAC transcription factors: Structurally distinct, func-tionally diverse. Trends Plant Sci. 10, 79–87.

57. Nogueira F.T.S., Scholgl P.S., Camargo S.R., Fernan-dez J.H., De Rosa V.E., Pompermayer P., Arruda P. 2005.SsNAC23, a member of the NAC domain protein family,is associated with cold, herbivory and water stress insugarcane. Plant Sci. 169, 93–106.

58. Rabbani M.A., Maruyama K., Abe H., Khan M.A., Kat-sura K., Ito Y., Yoshiwara K., Seki K., Shinozaki K.,Yamaguchi-Shinozaki K. 2003. Monitoring expressionprofiles of rice genes under cold, drought, and high-salinity stresses and abscisic acid application usingcDNA microarray and RNA gel-blot analyses. PlantPhysiol. 133, 1755–1767.

59. Shimizu H., Sato K., Berberich T., Miyazaki A., Ozaki R.,Imai R., Kusano T. 2005. LIP19, a basic region leucinezipper protein, is a Fos-like molecular switch in the coldsignaling in rice plants. Plant Cell Physiol. 46, 1623–1634.

60. Nakashima K., Yamaguchi-Shinozaki K. 2006. Regulonsinvolved in osmotic stress-responsive and cold stress-responsive gene expression in plants. Physiol. Plant.126, 62–71.

61. Chinnusamy V., Zhu J., Zhu J.K. 2006. Gene regulationduring cold acclimation in plants. Physiol. Plant. 126,62–71.

62. Benedict C., Geisler M., Trygg J., Huner N., Hurruy V.2006. Consensus by democracy. Using meta-analyses ofmicroarray and genomic data to model the cold acclima-tion signaling pathway in Arabidopsis. Plant Physiol.141, 1219–1232.

63. Zhu J., Shi H., Lee B.H., Damsz B., Cheng S., Stirm V.,Zhu J.-K., Hassegawa P.M., Bressan R.A. 2004. An Ara-bidopsis homeodomain transcription factor gene, HOS9,mediates cold tolerance through a CBF-independentpathway. Proc. Natl. Acad. Sci. USA. 101, 9873–9878.