FORUM REVIEW ARTICLE

Redox Regulation of Transcription Factorsin Plant Stress Acclimation and Development

Karl-Josef Dietz

Abstract

Significance: The redox regulatory signaling network of the plant cell controls and co-regulates transcriptionalactivities, thereby enabling adjustment of metabolism and development in response to environmental cues,including abiotic stress. Recent Advances: Our rapidly expanding knowledge on redox regulation of planttranscription is driven by methodological advancements such as sensitive redox proteomics and in silico pre-dictions in combination with classical targeted genetic and molecular approaches, often in Arabidopsis thaliana.Thus, transcription factors (TFs) are both direct and indirect targets of redox-dependent activity modulation.Redox control of TF activity involves conformational switching, nucleo-cytosolic partitioning, assembly withcoregulators, metal-S-cluster regulation, redox control of upstream signaling elements, and proteolysis. CriticalIssues: While the significance of redox regulation of transcription is well established for prokaryotes and non-plant eukaryotes, the momentousness of redox-dependent control of transcription in plants still receives in-sufficient awareness and, therefore, is discussed in detail in this review. Future Directions: Improved proteomesensitivity will enable characterization of low abundant proteins and to simultaneously address the various post-translational modifications such as nitrosylation, hydroxylation, and glutathionylation. Combining suchapproaches by gradually increasing biotic and abiotic stress strength is expected to result in a systematicunderstanding of redox regulation. In the end, only the combination of in vivo, ex vivo, and in vitro results willprovide conclusive pictures on the rather complex mechanism of redox regulation of transcription. Antioxid.Redox Signal. 21, 1356–1372.

The Redox Regulatory Network of the Cell

Plants are structured from modules such as leaves,branches, side roots, flowers, and fruits. The number of

modules developed during the plant life cycle varies. Thismodular characteristic enables plants to display an enormousmorphological plasticity in dependence on resource avail-ability and many other environmental parameters. Good en-vironmental conditions enable vigorous growth with manymodules, while adverse conditions delimit the number ofmodules. Complementary to the morphological plasticity,plants realize genetic and biochemical plasticity. By triggeringspecific gene expression programs and post-translationalstates, the genetic and biochemical plasticity enables an op-timization of acclimation responses to various types andcombinations of biotic and abiotic stress. Thus, plants accli-

mate to the environment at different molecular levels by (i)rapid biochemical modulation of metabolism, (ii) just-in-timeactivation of cell defense (iii) transcriptional and translationalrearrangement of metabolism, (iv) long-term control of sizeand number of organs (modules), and (v) the timing of de-velopmental transitions. Levels (ii)–(v) include gene expres-sion regulation. Optimized realization of metabolic anddevelopmental plasticity depends on a continuous sensing ofinternal and external parameters and appropriate adjustmentof the modular plant morphology within the preformed ge-netic and epigenetic program.

A huge body of evidence proves that reactive oxygenspecies (ROS), reactive nitrogen species (RNS), and other re-dox cues participate in most, if not all, of these acclimationsteps. A significant part of cellular redox regulation targets thethiolome. In simple terms, the dynamic thiol state of a cell can

Biochemistry and Physiology of Plants, Faculty of Biology, Bielefeld University, Bielefeld, Germany.

ANTIOXIDANTS & REDOX SIGNALINGVolume 21, Number 9, 2014ª Mary Ann Liebert, Inc.DOI: 10.1089/ars.2013.5672

1356

be defined as (i) the rate of sulfur assimilation; (ii) the levels ofthiols, including the small pool of free cysteine; the major low-molecular-mass thiol buffer glutathione (GSH) and the verylarge thiol pool in proteins, which together form the thiolome;(iii) the rate of sulfur diversion into other pathways; (iv) theelectron flux into the thiol network; and (v) the drainage ofelectrons from the network by oxidants such as ROS and RNS(Fig. 1). Thiols of the thiolome can undergo diverse modifi-cations, in particular conversion to the disulfide, sulfenic,sulfinic and sulfonic acid, nitrosothiol, and glutathionylatedforms (Fig. 1).

For a long time, the symptoms associated with ROS accu-mulation were solely considered indicators of severe ‘‘sick-ness’’ and termed oxidative stress, indicating a pathologicalmisbalance between oxidant generation and antioxidantprovision. Nowadays, it has become clear that this view is anobsolete simplification. The signaling function of ROS hasemerged as a fundamental principle in cellular communica-tion (1), and the concept of the redox regulatory network ofthe cell has been developed as a central element in acclimation(22, 45). Arguments that are based on their reactivity and thevery high superoxide dismutase (SOD) activity support theview that radicals such as O2

� - , RO�, Cl�, and, possibly inmost cases, NO� cannot function in specific signaling (8, 30).These reactive molecular species need to be detoxified, asotherwise they oxidize cell constituents in a rather non-specific manner. However, such highly reactive species mayact at a broader level as recognized by the identification ofS-nitrosylation as another commonly encountered post-translational modification (PTM) with yet insufficiently ex-plored functional implications (67, 96).

Dielectron oxidants such as H2O2 usually react at very lowrates with protein thiols in their protonated state, while thethiolate anion of cysteine acts as a strong nucleophile andreadily reacts with oxidants. In a simple view, reactivity isinversely correlated with pK-value. Thus, thiols can be acti-vated by lowering the pKa through neighboring effects withinthe peptide decreasing the activation barrier. However, inaddition, diffuse electric fields, dipole properties, solvent ac-cessibility, and the neighboring hydrogen bond network alsoaffect thiol reactivity (28). As a consequence, each protein thiol

and non-protein thiol may react specifically. In addition,many regulatory feedback loops control redox homeostasisand ROS generation of the cell (Fig. 2A) (101). All thesefunctional dependencies suggest that oxidative stress isreadout of deregulated signaling pathways (40), while thegeneration of ROS and RNS, first of all, is a mechanism that isused to adjust the redox state of the redox regulatory networkof the cell with all its signaling requirements. They also tunegene transcription.

Six functional elements co-operate in the redox regulatorynetwork, which is essentially conserved throughout all or-ganisms. These network elements are commonly found incytosol, nucleoplasm, mitochondrion, and plastids.

(i) Redox input elements link redox reactions of metabolismto redox regulation of protein thiols. They feed electrons intothe regulatory redox network. NADPH in combination withNADPH-dependent thioredoxin reductase, ferredoxin (Fd)along with Fd-dependent thioredoxin reductase, and gluta-thione reductase (GR)/GSH are the dominating redox inputelements in plants (Fig. 2B).

(ii) Redox transmitters transfer and distribute electrons fromthe input elements to downstream target proteins. Thesefamilies of proteins are particularly enlarged in plants withnearly 44 thioredoxins (Trx) and Trx-like proteins and 33glutaredoxins (Grxs) that are encoded in the Arabidopsisthaliana genome (45). Grx are grouped according to their se-quence similarities that correlate with the reaction mechanismin deglutathionylation by the monothiol- or dithiol mecha-nism or as disulfide reductase (111). They have specific sub-cellular localization, for example, Trx-h is also found in thenucleus of seeds (105).

(iii) Redox target proteins display redox-sensitive thiolswhose redox state is controlled by redox transmitters. Morethan 400 target proteins have tentatively been identified byvarious targeted and screening approaches, as summarized inrecent reviews (10, 66, 110). The drawback of the screeningapproaches (redox proteomics) is that mostly the abundantcellular thiol-disulfide redox proteins are identified. Thelimitation of the targeted approaches is their focus on usuallysingle proteins. However, it will be shown next that an in-creasing number of important players of less abundant

FIG. 1. The dynamic redox thiolome of thecell. The thiolome is defined by the rate of thiolsynthesis , the total redox-active thiol pool(gray box), including Cys, GSH, and proteinthiols , the withdrawal of sulfur from the pool

, the rate of electron influx , and the efflux tofinal electron acceptors such as ROS and RNS .It should be noted that only a part of the totalpool of theoretically available protein thiols is,indeed, redox active. The arrows mark relation-ships, but are far from complete; for example, amajor route to protein thiol reduction is realizedby the vast number of thioredoxins as redoxtransmitters. The list on the left hand sideindicates commonly encountered PTMs ofthiol groups, namely disulfide, sulfenic, sulfinic,and sulfonic acid, S-nitrosylation and S-glutathionylation. GSH, glutathione; PTM,post-translational modification; RNS, reactivenitrogen species; ROS, reactive oxygen species.

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1357

signaling elements have been identified and that the increasedsensitivity of redox proteomics opens the perspective to aglobal understanding of redox regulation. Redox targets havebeen assigned to most gene ontology groups, including met-abolic pathways, translation, and transcription. Interestingly,the energy-consuming process of translation is a prime targetof redox regulation (37).

(iv) Redox buffer proteins were recently suggested to play animportant role in maintaining redox homeostasis in the thiol-disulfide network. Cysteines of ribulose-1,5-bisphosphatecarboxylase oxygenase (RubisCO) were identified as beingsensitive to early oxidation in vitro and in vivo (74, 79). Takinginto consideration the concentration of thiols in the large andsmall RubisCO subunits, the RubisCO thiol pool dominatesthe thiol redox buffer capacity of the chloroplast, exceedingthe thiols in GSH by more than one order. Thus, redox tran-sitions of RubisCO thiols appear to be important in catalyticregulation, control of turnover (74), and redox homeostasis ofthe chloroplast. Due to its high concentration, RubisCO maybe considered a special example; however, the hypothesisshould be tested in general, namely that protein thiols exert animportant function in transiently buffering redox imbalances.

(v) Redox sensors deliver information on ROS/RNS levelsinto the redox regulatory network and realize cross-talk toother signaing pathways (Fig. 2C). One can distinguish be-tween kinetic and static redox sensors (21). Kinetic sensors area part of the electron flow within the signaling redox cascade,while the redox state of static sensors is adjusted in equilib-

rium with elements of the redox cascade but themselves donot take part in electron flow. Kinetic sensors, for example,have a high affinity to ROS or RNS and turn oxidized if thereaction rate with ROS and RNS exceeds the rate of reductiveregeneration of the sensor. Direct oxidation of redox targetsinstead of indirect electron drainage by redox sensors neces-sitates that the reaction constants compete with those of per-oxiredoxins and GSH peroxidases. Otherwise, they will not beable to function as reliable redox sensors (8).

(vi) Final electron acceptors are reactive low-molecular-massredox species (ROS, RNS, reactive sulfur species [RSS], andreactive carbonyl species [RCS]) (18). By abstracting electronsfrom appropriate donors, they finally reach the oxidationlevels of, for example, H2O, NO3

- , and SO42 - , which are

nonreactive. Superoxide and nitric oxide combine to formperoxinitrite, which is highly reactive. As rationalized byBrigelius-Flohe and Flohe (8), efficient final acceptors in ashort distance are NO and O2

- among the free radicals, andefficient signaling components are H2O2 and alkyl hydro-peroxides.

In evolutionary terms, the network components are likelyevolved as catalysts of redox reactions in metabolism such asthe reduction of desoxyribonucleotides to ribonucleotides,where redox co-substrates participate in the metabolic reac-tions (Fig. 3A) (33). Another redox pathway may have gainedfunction in feeding electrons in thiol antioxidants to decom-pose ROS and RNS when the atmospheric oxygen concen-tration was still low (Fig. 3B). It is proposed that subsequently,

FIG. 2. Important reactions in thiol-disulfide regulation and function of redox sensors. (A) General structure of asignaling cascade linking metabolism to response. The example shows how wound-induced lipid peroxidation increaseslevels of OPDA, which bind to cyclophilin Cyp20-3 (88). The OPDA-liganded Cyp20-3 binds to serine acetyltransferase,which activates the cysteine synthase complex and results in enhanced Cys synthesis and improved redox homeostasis. (B)Redox input element, redox transmitter, and redox sensor. Exemplary reactions given here are the NTR, the thioredoxin-mediated disulfide reduction of a target protein (middle), and the peroxiredoxin with its deprotonated peroxidatic cysteinereacting with a peroxide (ROOH, R stands for H, alkyl, or ON in peroxinitrite) (45). (C) Distinction between static and kineticsensors. The kinetic sensor is a component of the thiol-disulfide/ROS redox cascade, while the static sensor equilibrates withthe redox state of another redox component and itself does not participate in the electron flow (21). Cyp20-3, stromalcyclophilin 20-3; NTR, NADPH-dependent thioredoxin reductase; OPDA, oxophytodienoic acid.

1358 DIETZ

a salvage pathway for the regeneration of oxidized intra- orintermolecular disulfide bridges or other oxidized speciesdeveloped as the oxygen concentration increased in the at-mosphere. An example is methionine sulfoxide reductase,which reduces oxidized sulfur in methionine and is linked toredox transmitters (17). In the third step, redox switchingbetween dithiol- and disulfide states may have enabled reg-ulation of protein function, in particular of enzyme activity(Fig. 3C). Such new regulatory mechanisms by post-transla-tional redox modification tightly and immediately linkedcellular redox state to metabolic activity. In the fourthstep, diversification of redox regulation occurred by geneduplications of redox transmitters and incorporation ofredox-sensitive domains in target proteins. The number ofredox-sensitive proteins in plants appeared to have explodedon conquering land and radiation of angiosperms (10, 45).

According to this model, efficient ROS- and RNS-generatorsystems constitute an essential element of the cellular thiol-disulfide redox regulatory network. ROS and RNS generationoccurs as a part of chloroplast and mitochondrial energyconversion, in peroxisomal metabolism, in the cytosol and theplasmamembrane (Fig. 4). Oxygen activation in photosystemII releases singlet oxygen, which is quenched by tocopherol(57). Singlet oxygen causes oxidation of unsaturated fattyacids (UFA) and releases reactive electrophilic species that actas signaling molecules. The specificity of the oxidation patternof UFAs reveals that singlet oxygen-dependent UFA oxida-tion is usually less important than lipoxygenase (LOX)-cata-lyzed UFA oxidation (116). Superoxide is generated inphotosynthetic electron transport, respiratory electron trans-port (RET), and by plasmamembrane NADPH oxidase. SOD,which is present in most cellular compartments, dismutatestwo molecules O2

- to H2O2 and O2. H2O2 can be transportedbetween cell organelles (6) and has the longest lifetime amongthe various ROS and RNS species. In cytosol, mitochondrion

and plastid H2O2 is detoxified by the Foyer-Halliwell-Asadacycle or by thiol peroxidases (peroxiredoxins) (Fig. 4). TheFoyer-Halliwell-Asada cycle uses ascorbate as an electrondonor for H2O2 reduction and GSH as a primary regenerant.During episodes of acute H2O2 release, the transient increasesin dehydroascorbate (DHA) and oxidized glutathione (GSSG)levels may occur, both of which appear to act as signalingmolecules. Grx and protein disulfide isomerases can reduceDHA and turn oxidized (127). Thus, DHA drains electronsfrom the thiol disulfide redox regulatory network as well. Thecentral role of the ascorbate/DHA and GSH/GSSG pairs incell signaling has been reviewed repeatedly, and Foyer andNoctor (31) suggested that they constitute the ‘‘heart’’ of theredox hubs of the cell. Considering the whole redox regula-tory network as a hub may be misleading and not in line withpresent-day understanding of network theory (23), but thepronounced effects on multiple responses in developmentand stress acclimation assign outstanding roles within theredox regulatory network to the ascorbate and GSH systems,also in controlling gene expression as outlined next. An im-portant mechanism of signal amplification relies on theshutdown of antioxidant activity as described for chloroplastascorbate peroxidases, peroxiredoxins, and catalases (82, 94,108). Loss of antioxidant activity enables local accumulationof ROS and intensifies the ROS signal.

The Redox Proteome of Cell Signaling

Many proteins contain redox-sensitive sites. In particular,cysteine, methionine, and tyrosine undergo redox reactions asPTM and form disulfide, sulfenic, sulfinic, and sulfonic-acidderivatives, S-nitrosylated, O,- and S-nitrated forms. In ad-dition, metal-S-clusters containing proteins are susceptibleto redox modulation. These conformational changes altercatalytic activity or protein/protein interaction sites. An

FIG. 3. Hypothetical evolution of the thiol-disulfide redox regulatory network. By combining cosubstrate (A) and thiolantioxidant pathways (B), the redox regulation of targets was linked to ROS/RNS generation (C). By the conversion of thiolantioxidants into conformational sensors and the introduction of thiol buffers consisting of low-molecular-mass metabolitesand proteins, a flexible network emerged that enabled the adjustment of target proteins in dependence on environmentalconditions (D). The activity of ROS/RNS generator systems and the antioxidant capacity play decisive roles in adjusting thenetwork state.

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1359

exhaustive analysis of the redox protein inventory of the cellsis still not available (10, 66, 67, 110). However, the methodsthat have been developed for identifying additional membersof the redox network have reached unprecedented accuracyand sensitivity. In addition to redox proteomics methodologyusing affinity binding to mutated Trx or Grx, or differentiallabeling with thiol-reactive reagents such as N-ethylmalei-mide, biotinmaleimide, or isotope-coated affinity tags, newlydeveloped in silico tools enable prediction of potential targetsof redox regulation. Sanchez et al. (99) suggested a genome-wide selection of thiol-disulfide transition proteins based onaccessible surface area, pK-value of the cysteine thiol, anddistance between two cysteines. A dichotomous selection treewas implemented as an automated bioinformatics tool namedReversibly Oxidized Cysteine Detector and starts with a wholeset of predicted proteins of an organism or a subcellular site(60). However, this first implementation of an automated se-lection tree needs much refinement in order to reach sufficientreliability (Fig. 5). Such additional parameters concern prop-erties of the attacking and leaving thiol, steric and neighbor-ing effects (28, 81). Additional information, for example, fromtranscriptional co-expression may help narrow down inter-actions within the thiol-disulfide network that are possible inspace and time. Since intramolecular disulfide bridges onlyrepresent one possible redox mechanism, the tool predicts asmall subset of the potential targets of redox regulation.Abundant cell proteins such as enzymes and translation fac-tors dominate among the more than 400 thiol targets de-scribed so far (66). Redox regulation of metabolic pathways

enables short-term adjustment of metabolism to the prevail-ing environmental conditions. The available list of thiol pro-teins is far from being complete. This article focuses on redoxproteins controlling gene expression. Such proteins withregulatory function are often present in a low number in cells,and it has been difficult to reliably identify them by experi-mental proteomics. This limitation is overcome with new andhighly sensitive mass spectrometric methods that determinemillions of peptides in complex samples (89) and by betterin silico predictions which may also take into account thedegree of phylogenetic conservation.

ROS/RNS Generator Systems, Their Responseto Stress, and Transfer Across Membranes

ROS are generated in the metabolism of various subcellularcompartments (Fig. 4). ROS and RNS function in cell signalingand have been reviewed in detail, often in the context of an-tioxidants and their potential contribution to oxidative stress(31). The highest rate of H2O2 production occurs in the per-oxisomes of photosynthesizing cells. Mehler reaction pro-duces O2

- at photosystem I. The estimated O2� - release rates

range between < 5% of linear electron transport under normalconditions and about 30% when intercellular CO2 is low, forexample, under drought (5). Thus, both Mehler reaction andphotorespiration serve in transmitting information on envi-ronmental conditions to redox cues. Singlet oxygen is liber-ated in photosystem II in the absence of sufficient electronacceptors. Experimentally singlet oxygen may conditionally

FIG. 4. ROS and RNS generator systems and their response to stress. The chloroplast, peroxisome, mitochondrion,cytosol, and plasmamembrane are five cellular sites of major ROS and RNS production. The figure exemplarily depictspathways of ROS and RNS generation, their processing and detoxification, and potential signaling roles (thick arrows). AO,ascorbate oxidase; Asc, ascorbate; CAT, catalase; DHA, dehydroascorbate; DHAR, dehydroascorbate reductase; FNR, fer-redoxin-dependent NADP reductase; GR, glutathione reductase; GSNO, nitrosoglutathione; GSSG, oxidized glutathione;NiNOR, nitrite nitric oxide reductase; NR, nitrate reductase; PET, photosynthetic electron transport; Prx, peroxiredoxin;Rboh, NADPH oxidase; RET, respiratory electron transport; SOD, superoxide dismutase.

1360 DIETZ

be produced in excessive amounts in mutants such as flu,which accumulates protochlorophyllide in the dark and re-leases singlet oxygen on reillumination (70). The system wasused to search for components of the involved signalingpathways by screening for second site mutations in the ge-netic flu background using the AAA-ATPase-reporter asreadout (3). The nature of the identified genes suggests thatthe singlet oxygen pathway triggered in protochlorophyllideover-accumulating flu mutants is not uniform and involvescross-talk with many other signaling pathways (3). Thequestion remains unanswered as to whether the flu systemaddresses a physiologically relevant pathway or just repre-sents an elegant tool that is used to explore compensatorymechanisms on severe disturbance of redox homeostasis.Recent data suggest that the use of flu enables dissection of asinglet oxygen-triggered cell death program present in cellswith fully developed chloroplasts (51). The authors proposedthat the activation of the singlet oxygen-dependent cell deathprogram may participate in acclimation responses tostrengthen stress resistance. In addition to plastids, mito-chondria are conditional producers of ROS signals (95). In-ducers of mitochondrial retrograde signaling are not onlyinhibitors of RET (antimycin A), inhibitors of the citrate cycle(monofluoroacetate) but also environmental factors such asozone and UVB. They trigger changes in nuclear gene ex-

pression and aim at restoring mitochondrial functionality andantioxidant defense (95). Recently, transcriptome analysisprovided further evidence that in addition to mitochondrialROS signals, ROS-independent mitochondrial signaling exists(119). ROS produced in peroxisomes (120) or by NADPHoxidases at the plasma membrane and by peroxidases in thecell wall (7, 109) regulate nuclear gene expression. ROS sig-nals travel systemically through plants with about 1.4 mm/s(72) corresponding to a spreading speed of > 30 cells/s. Themechanism relies on self-propagating ROS, calcium, andNADPH oxidase circuitries. Nitric oxide is the most importantRNS. The subcellular sites and biochemical mechanisms ofits production in plants are still not entirely clear (Fig. 4).Cytosolic nitrate reductase, plasmamembrane-bound nitrite–nitric oxide oxidoreductase, NO synthase-like proteins in theperoxisome, and mitochondrion, plastid, cytosol, and non-enzymatic conversion of nitrite to NO are discussed as con-tributing to NO generation in plant cells (4). Two reactionproducts of NO, on the one hand with GSH to form ni-trosoglutathione (GSNO) and on the other hand with O2

� - toform peroxinitrite (ONOO - ), interfere with the thiol-disulfideredox regulatory network. GSNO is a donor for S-nitrosylation,and ONOO- is a donor for protein nitration. Both PTMsmodulate protein functions (67, 96).

Thiol Redox Control of Protein Phosphatasesand Kinases

Redox regulation of transcription often depends on up-stream signaling elements that are redox controlled. A com-mon mechanism employs redox regulation of proteinphosphorylation/dephosphorylation. The state of the cellularphosphoproteome is a prime target of thiol redox cross-talkand, in turn, controls many aspects of gene expression andtranscription factor (TF) activity. Several protein phospha-tases and protein kinases undergo thiol-dependent switch-ing as revealed in proteomic approaches and targetedapproaches in analyses of specific protein kinases andphosphatases (79, 97, 124). Protein phosphatases of the 2Ctype [PP2C, e.g., At5g53140 (97, 124)] function either asantagonists of mitogen-activated protein kinase (MAPK)pathways or as co-receptors for the phytohormone abscisicacid (ABA) (32, 43). PP2C mediates redox cross-talk into theABA signaling pathway (69).

The MAPKKK ANP1 is specifically activated by H2O2

and phosphorylates MAPK3 and MAPK6 in A. thaliana (56).The MAPK pathway activates typical ROS-induced tran-scripts such as glutathione-S-transferase 6, heat shock pro-tein HSP18.2, and RD29A and suppresses auxin-inducedgene expression (56). Tobacco ectopically expressing ANP1-homologue NPK1 was more resistant to freezing stress. Pro-tein tyrosine phosphatase 1 (At1g71860) dephosphorylatesMAPK. Its inactivation on oxidation may enable signalpropagation through the MAPK pathway under oxidativestress (38, 131).

The thylakoid state transition kinase STN7 (At1g68830)belongs to Ser/Thr protein kinases and co-ordinates short-term acclimation to light fluctuations by catalyzing rapidPTM of photosynthetic components and long-term acclima-tion by modifying nuclear gene expression through retro-grade signaling, leading to restructuring of the photosyntheticapparatus (90). STN7 contains conserved Cys residues that

FIG. 5. Decision tree for the bioinformatic prediction ofproteins that potentially undergo thiol-disulfide transi-tions. Candidates are selected starting from a genome-wideprotein set, followed by filtering for specific expression and/or subcellular localization, and then based on features suchas the thiol pK value, the accessible surface area, and thethiol distance (60, 99). Final support may be obtained from aphylogenetic analysis of conservation. Such tools can also beapplied to transcription factors.

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1361

undergo thiol-disulfide transitions (79). In contrast to oxida-tive inactivation of protein tyrosine phosphatase via the con-served Cys190 in the active center, Li et al. (61) reportedoxidative activation of a protein tyrosine phosphatase ZmRIP1in maize. On oxidation, probably via Cys181, ZmRIP1 istranslocated from the plastids to the nucleus in the heter-ologous system of onion epidermis cells. ZmRIP1 stimulatesexpression of a glutathione-S-transferase (At1g02950) (61).

Protein phosphorylation and redox-dependent mecha-nisms also control plastid gene expression. The chloroplastcasein kinase 2 (cpCK2) phosphorylates plastid Sigma factorsthat are required to initiate plastid gene expression. Thus,phosphorylated Sigma6 binds, for example, the atpB pro-moter (118). GSH inhibits cpCK2 by disassembling the activecpCK2-dimer. Thioredoxin z (Trx-z) interacts with fructoki-nase-like proteins 1 and 2 (2), which are a part of the plastidtranscription machinery of the plastid encoded polymerase(91). A. thaliana lacking Trx-z develop a severe albino phe-notype. Thus, Trx-z plays an important role in the regulation

of chloroplast transcription as catalyzed by the plastid en-coded DNA-dependent RNA polymerase (2, 14).

Mechanisms of Redox Control of TFs and Examplesfrom Non-Plants

Several redox-dependent mechanisms control the activityof TFs (Fig. 6) (8). Redox signals that control transcription canact via extended signaling pathways within cells and even atthe systemic level of organisms. Alternatively, redox-depen-dent PTMs directly control TF function. Currently, this is arapidly emerging field in plant science but has been investi-gated in detail in the bacterial and mammalian field.

(i) Redox-dependent conformational changes may exposethe buried nuclear localization sequence, which enablesnuclear import. Target genes then get transcribed aftertranslocation of the TF into the nucleus. In vertebrates, theglucocorticoid receptor functions as hormone-controlled TF,which translocates to the nucleus on hormone binding only

FIG. 6. Redox regulation of transcription factor activity. The figure depicts the seven principle mechanisms that have beendescribed for redox-dependent activity control of transcription factors and gives examples from vertebrates/humans (8) aswell as from plants. The dark gray loop marks the NLS, which may be either buried or exposed, and in the latter formmediates nuclear import. (A) Redox-dependent regulation of compartmentation, (B) redox regulation of conformational state,(C) redox control of assembly, (D) oxidative disassembly of metal-sulfur clusters, (E) indirect redox effects on post-transla-tional modifications, (F) redox-triggered proteolytic processing, and (G) direct redox control of DNA binding. It should benoted that redox control is only a part of the regulation of ABI5 activity. ABI, abscisic acid insensitive; ANAC, abscisic acidresponsive NAC transcription factor; AP1, activated protein 1; bZIP, basic leucine zipper; GR, glucocorticoid receptor; HIF,hypoxia-induced factor; HSF, heat shock factor; c-Jun, subunit of AP-1; Myb, myeloblastosis oncoprotein (MYB)-like tran-scription factor; NLS, nuclear localization sequence; Nrf, nuclear factor-E2-related factor-2; OxyR, repressor of H2O2 responsein bacteria; P53, tumor suppressor protein 53; RAP, related to apetala; SBP, SQUAMOSA promoter binding proteins (SBP)-transcription factor; SP1, Cys2His2-TF from vertebrates; TGA, transcription factor binding to a TGA-related motive.

1362 DIETZ

under reducing conditions, while oxidative effectors suppresstranslocation (85) (Fig. 6A).

(ii) Redox PTMs affect tertiary and quarternary structuressuch as mono-/dimer-/oligomerization state, thereby bind-ing to the target promoter (Fig. 6B). The global regulator OxyRcontrols the oxidative stress response in Escherichia coli. Itsactive form is the tetramer in its oxidized state that binds tofour consecutive major grooves and activates gene tran-scription by binding the a-subunit of RNA polymerase (59).Dithiol-disulfide transition occurs at concentrations as low as100 nM H2O2.

(iii) Interaction of some TFs with binding partners dependson the redox state of interacting proteins, which then functionas activators or suppressors (Fig. 6C). In human cells, the re-dox effector factor 1 (Ref1) controls the activity of TFs, such asthe activator protein 1 (AP1). This redox regulation dependson a direct interaction between Trx and Ref1 (41). Nuclearredox factor 2 (Nrf2) binds to Kelch domain-containing part-ner (Keap1) in the cytosol. Keap1 is anchored to the cyto-skeleton and is, thus, retained in the cytosol. On oxidationof Keap1, Nrf2 is released, translocates to the nucleus, andactivates the expression of target genes of the antioxidantdefense (123).

(iv) Several classes of TFs co-ordinate with metals in theirfunctional assembly. FeS clusters and Zn fingers are sensitiveto oxidative disassembly (Fig. 6D). Mammalian SP1 containsthree Zn fingers of the Cys2His2 type in the DNA bindingdomain. Cys exchange or oxidation releases Zn and inhibitsbinding to DNA (55). SP1 was identified from HeLa cell ex-tracts as TF specifically activating the viral SV40 promoter(27). With aging, SP1 adopts a more oxidized state and showsa lower Zn occupancy that could have physiological impli-cations (126). The human tumor suppressor P53 controls cellcycle and genome integrity. Cys176, 238, and 242 along withHis 179 co-ordinate a Zn2 + ion and establish the DNA bind-ing pocket needed for cis element binding and target geneexpression. On oxidation, P53 loses its DNA binding ability(15). The redox state of Cys277, which is not a part of themetal-sulfur cluster, affects cis element-binding specificity ofP53 (11).

(v) Redox-dependent regulation by PTMs such as phos-phorylation/dephosphorylation often controls import or ex-port, binding or unbinding of TFs (Fig. 6E). The mammaliannuclear factor kappa B translates xenobiotic, hormonal, andoxidative stimuli in gene expression control. This pathway isunder multiple redox control (8). For example, phosphoryla-tion of a particular subunit protein p65 is a prerequisite fortranslocation and expressional activation. Several phospha-tases such as PP1/PP2A counteract phosphorylation and areinactivated by oxidation of Cys residues (86).

(vi) Redox regulation of proteolytic processes controls therelease or stability of active TF or terminates its activity (Fig.6F). Nuclear factor E2-related factor 2 (Nrf2) controls a part ofthe antioxidant response in mammalian cells and is continu-ously degraded under non-stress conditions. On stress, Nrf2is stabilized, activates nuclear gene expression, and redoxhomeostasis is re-established. Oxidation of Cys136 in theubiquitin conjugating enzyme UbcM2 plays a role in the sta-bilization process (93).

(vii) Disulfide formation, S-glutathionylation, S-nitrosylation,or other Cys modifications control TF binding to the propercis-element (Fig. 6G). Along this line, several TFs have been

experimentally shown to be governed by redox regulation.The most straightforward mechanism involves Cys modifi-cation in the DNA binding domain. c-Jun, along with c-Fos,constitutes the transcriptional activator AP1 and is inactivatedby S-glutathionylation (53).

The Emerging Concept of Redox Controlof TFs in Plants

Significant progress is currently made in plants in identi-fying redox-dependent mechanisms that control TF activity.Increasing evidence suggests that the seven mechanisms de-scribed earlier for mammalia and bacteria are also present inplants. Heat shock transcription factors (Hsf) mediate therapid transcriptional readjustment on fast increases in tem-perature and on impact of other stressors. The A. thalianagenome codes for 21 Hsf proteins (84, 100). A subset of Hsfshas been shown to mediate the early response to high lightand one of these, HSFA1D fused to yellow fluorescent protein,was reported to be translocated from the cytosol to the nu-cleus in excess light (47). HsfA8 has been hypothesized tofunction as a redox sensor (71). Indeed, HsfA8 fused to fluo-rescent protein and transiently expressed in protoplaststranslocated from the cytosol to the nucleus on treatment withH2O2 (Fig. 7). This translocation depends on the presence ofspecific Cys residues (M. Giesguth and K.J. Dietz, Un-published). The functional implications of redox-dependentHsfA8 translocation still need to be explored. On the otherhand, the data show that scenario A depicted in Figure 6 isfound in plants.

Rap2.4a was identified in a yeast-1-hybrid screen using theredox-sensitive promoter region controlling expression of thechloroplast 2-cysteine peroxidoxin gene as a bait (106).Rap2.4a belongs to the TF family of AP2/ethylene responsefactor (ERF), which consists of 147 members in A. thaliana (24).Rap2.4a undergoes profound transitions of quaternarystructure in dependence on a redox environment in vitro.Binding to the cis element is impeded at highly reducing aswell as highly oxidizing conditions, with the latter fosteringoligomerization. At intermediate redox potentials, Rap2.4abinds to the promoter and activates transcription, at least inthe transient protoplast expression system (106). The redoxmidpoint for the monomer-dimer transition was almost- 270 mV. Unpublished data suggest that Rap2.4a co-ordinatessugar availability with developmental transitions such as seedgermination and time of flowering (D. Gomez-Perez, D.Staiger, and K.-J. Dietz, Unpublished). Rap2.4a represents anexample of scenario B of redox-controlled quaternary struc-ture (Fig. 6).

Other members of the AP2/ERF TF family regulate geneexpression in response to anaerobiosis and submergence. Thesubfamily ERF VII TFs displays a conserved N-terminusstarting with MCGGAI/L. The Cys at position 2 functions as aredox sensor that terminates anaerobiosis-induced activationthrough the N-end rule pathway of protein degradation (35,65). Under aerobic conditions, Rap2.12 is bound to the plas-mamembrane-anchored acyl-CoA binding proteins 1 or 2(ACBP1/2; Fig. 8). In low oxygen, either the ACBP/Rap2.12complex is disrupted or the whole complex is released andtransferred to the nucleus (64). Met-1 is removed by me-thionine aminopeptidase (114). In the presence of O2, theterminal Cys is converted to the sulfenic acid derivative.

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1363

An O2-dependent arginine transferase conjugates an arginylresidue to the sulfenic acid. Subsequent ubiquitinylation ofRap2.12 triggers its degradation. This process does not takeplace in anaerobic conditions. Thus, on release from theplasmamembrane in anaerobic conditions, Rap2.12 moves tothe nucleus and activates gene transcription as a part of the

rapid anaerobiosis response (Fig. 8). Rap2.12 is conserved andconstitutively expressed in plants. It represents an examplecombining scenario C (redox regulation of assembly) andscenario F (redox-dependent proteolysis).

Rap2.2, which is highly similar to Rap2.12, binds theATCTA-cis element within the promoters of phytoene

FIG. 7. Translocation of HsfA8 fused toYFP into the nucleus of Arabidopsis thali-ana protoplasts on treatment with 5 mMH2O2. Leaf protoplasts were transfected withDNA containing the construct hsfA8:yfp un-der control of the 35S promoter (as in ref. 80).About 16 h after transfection, protoplastswere either treated with 5 mM H2O2 for60 min or maintained untreated as a control.Images were taken by confocal laser scan-ning microscopy. (A, C) Show fluorescencefrom the YFP channel, and (B, D) indicate theoverlay of YFP fluorescence with chlorophyllfluorescence to visualize the position of thechloroplasts. Translocation was not seen ontreatment with the reductant dithiotreitol(5 mM, not shown). The pictures are repre-sentative of many observations (M. Giesguthand K.-J. Dietz, Unpublished). YFP, yellowfluorescent protein. To see this illustration incolor, the reader is referred to the web ver-sion of this article at www.liebertpub.com/ars

FIG. 8. Activation and termination pathways of the hypoxia transcription factor ERFVII in A. thaliana (79). ERFVII isbound to the plasmamembrane-anchored ACBP and released at low oxygen. On translocation to the nucleus, ERFVIIactivates the expression of hypoxia response genes. In the presence of oxygen, the sensory Cys is oxidized to sulfenic acid,conjugated to Arg, and degraded on ubiquitination. ACBP, acyl-CoA binding proteins; ERF, ethylene response factor.

1364 DIETZ

synthase and possibly also of phytoene desaturase. Bothproteins are key enzymes in carotinoid synthesis. The pre-ferred model suggests that Rap2.2 interacts with the zinkfinger protein Seven in absentia in Arabidopsis2 (SINAT2,At3g58040), and together, they form a redox-controlled ma-chinery which regulates carotinoid synthesis (128).

The homeodomain (HD) is the DNA binding interface ofHD TFs, which often participate in developmental processes.Athb-9 controls shoot patterning of A. thaliana by affectingmeristem development, organ polarity, and vascular differ-entiation (16). Three conserved cysteines are arranged be-tween the a-helices that mediate the binding to the cis element(16). Two cysteinyl-residues, corresponding to Cys23 andCys38 in Athb-9, are conserved among all HD-Zip III classTFs. These Cys residues reside in the peptide loops linkinghelices I and II, and II and III of the DNA binding domain,respectively. DNA binding only occurs if these Cys exist in thethiol state. Oxidized inactive Athb-9 is activated by Trx. Afourth Cys at position 58 co-determines the specificity ofbinding to the DNA cis-motif (16). Sequence conservationsuggests that this type of redox-dependent control of targetgene expression during developmental processes is alsofound in other HD TFs (16).

Subgroups of the vast number of plant TFs also containsulfur-based metal-binding sites; for example, the SQUA-MOSA promoter binding proteins-TFs (132). Such TFs arelikely sensitive to oxidative modifications, as metal sulfurclusters often serve as redox sensory systems. A well-studiedexample is the bacterial oxidative stress sensor SoxR with its[2Fe–2S] cluster (125). More work should be performed on thisscenario D of redox regulation via assembly and disassembly

of metal-S clusters in plants. Such approaches should addressthe role of the clusters in expression control by site-directedmutagenesis and demonstrate the switch from bound to un-bound state in dependence on growth conditions ex vivo.

Basic leucine zipper (bZIP) TFs of the transcription factorbinding to a TGA-related motive (TGA) type are targets ofredox regulation. TGA1 was the founding member and itbinds to the cis element TGACGTGG (102). The A. thalianagenome codes for 10 TGA TF proteins (46). TGA1–7 are sortedinto three clades, regulate the expression of pathogenesis-re-lated (PR) genes, and co-ordinate plant immune system anddisease resistance (50, 133; for review, see 34). TGA8 (PERI-ANTHIA, PAN), TGA9, and 10 are involved in plant devel-opment. The phenotype of defective anther development in A.thaliana could be mapped to the Grxs ROXY1 and ROXY2(130). ROXY 1 and 2 are involved in the development of thecorrect petal number, anther, and microspore structure. Thesearch for interacting partners of ROXY identified TGA TFs.PAN regulates floral organ primordial. Using its C-terminallylocated domain at helix 5, ROXY1 binds TGA3 and PANat their second Gln-rich domain (62). In TGA1, Cys 172 andCys 287 form an intermolecular disulfide bridge. Two othercysteinyl residues, Cys 260 and Cys 266, can be glutathiony-lated or S-nitrosylated (68). Cys 340 is conserved betweenPAN and TGA1. Its site-directed mutation disables PAN tocomplement the PAN knockout phenotype in pan mutants(62). Complementation studies of roxy mutants with site-directed variants of ROXY proved the importance of thefirst C in the CCMC motif in redox regulation of PAN func-tion. All these data suggest, but still not finally resolve,the molecular mechanism by which the Grx ROXY controls

FIG. 9. Redox regulatory function of chloroplast cyclophilin Cyp20-3 as activator of Cys synthesis. After lipoxygenase-dependent synthesis, OPDA binds to Cyp20-3. Cyp20-3/OPDA associates with SAT1 and stimulates Cys synthesis (25, 88).The redox milieu of the cell is transformed into a more reducing state that triggers defense gene expression. The function ofCyp20-3 is also linked to 2-CysPrx, which forms complexes with Cyp20-3 (80). Within this redox regulatory scenario, oneshould also consider the role of redox buffer proteins such as ribulose-1,5-bisphosphate carboxylase/oxygenase, which likelybuffer transient redox disequilibria (79).

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1365

TGA-dependent gene expression of those clades with at leasttwo Cys residues (62).

Salicylic acid (SA) is a part of the molecular network thatmediates systemic acquired resistance (SAR) (48). SAR in-volves TGA2, 5, and 6, which have only one Cys residue (34).TGA TFs interact with Grx of the ROXY type, and all 21ROXY-like Grx interact with TGA2 (34). Thus, it is likely thatthere exists a link to thiol-based regulation that needs to beexplored. SA-dependent expression of PR genes involvesNPR1 (non-expressor of PR1), which may be considered oneof the best-studied redox-controlled signaling elements.NPR1 conveys the SA response in stress acclimation to abioticand some biotic stresses; for example, high light and suckinginsects. Intermolecular disulfide bridge formation stabilizesthe oligomeric homocomplex of NPR1 in the cytosol (75). SAtriggers monomerization of NPR1 by a reduction throughTrx-h5 (113). The NPR1 monomer is translocated into thenucleus and activates PR1-gene expression. Expression of SA-marker genes is enhanced in vtc1-mutants with low ascorbicacid levels and responds wild-type like in vtc1/npr1 doublemutant. The GSH-deficient mutants pad2 and cad2 resemblenpr1 mutants (9). NPR1 interacts with clade II TGA TF thatbinds to SA-responsive elements, for example, in the PR1promoter and, they, thus, activate PR1 gene expression. NPR1overexpressing plants are more resistant to pathogens (12).Redox regulation controls nuclear translocation of NPR1.Both TGA1 and NPR1 form a protein complex in their re-duced forms and activate PR gene expression (20). On oxi-dation to disulfide at Cys260 and Cys266, TGA1 is inactivatedand does not promote PR gene expression. This effect wasreported to be independent of promoter binding (20). Twoadditional Cys residues also participate in redox-dependentregulation of TGA1, as the TGA1 variant C172S/C287Sshowed constitutively enhanced PR gene expression (68).Both TGA1 and NPR1 react with GSNO. GSNO-treatedTGA1, probably due to S-nitrosylation, binds to the cis ele-ment in target gene promoters with a higher efficiency(68). NPR1 further stimulates the binding of GSNO-treatedTGA1 to the promoter. Occurrence of regulation of TGA1by S-nitrosylation still awaits evidence from in vivo studies.S-nitrosylation stimulates NPR1 oligomerization, whichwould be an inactive form in the cytosol (113); on the otherhand, NO treatment triggers NPR1 translocation into thenucleus (68). Oxidized NPR1 is reduced by redox transmitterTrx (113). Despite the significant progress made in under-standing the redox-dependent control of the TGA/NPR1system in SAR by thiol modifications, it is still not entirelyunderstood. It may be considered an illustrative example ofcomplex redox regulation that is representative of other reg-ulatory circuitries.

The discussed examples of redox-regulated TFs affectedbinding to cis elements, modulation of interacting partners, orshuttling between the cytoplasm and the nucleus. Abscisicacid-responsive NAC transcription factor (ANAC)089 wasidentified in a yeast-one-hybrid screen that was performed ina search for proteins binding to the nuclear encoded stromalascorbate peroxidase (sapx) gene (54). ANAC089 suppressedsapx activity in a transient transactivation assay. A trans-membrane domain tethers ANAC089 to the trans-Golgi net-work and the endoplasmic reticulum. The TF polypeptide isreleased from the endomembranes on treatment of the cellswith the non-physiological reductant dithiothreitol (54). A

reasonable working hypothesis is that a redox-sensitive pro-tease sheds the TF under reducing conditions when lessstromal ascorbate peroxidase (sAPX) activity is needed forantioxidant defense. Similarly, an envelope-anchored planthomeodomain (PHD) TF is proteolytically released from theenvelope membrane in response to retrograde signals fromthe chloroplast and stimulates expression of abscisic acid in-sensitive4 (ABI4) (112). ABI4 is an AP2/ERF TF that functionsas a versatile repressor and activator of diverse processesduring the complete life cycle of A. thaliana (129). Redox reg-ulation of ANAC089 and PHD describes scenario F of proteo-lytic processing for TF activation (Fig. 6). While the conditionalshedding appears convincingly proved, the precise redox-dependent mechanism triggering the release is unknown. Inaddition to the regulation of proteases, conformational changesof the tethered TF burying or exposing the cleavage site oradditional binding partners should be considered as well.

NO treatment induces profound changes in the planttranscriptome. S-nitrosylation has emerged as a redox regu-latory mechanism that controls TF activity. S-nitrosylation ofCys53 by NO released from sodium nitroprusside and GSNOinhibits binding of the TF AtMYB2 to the core DNA bindingsite 5¢-AAACCA-3¢ in vitro (104). Recently, AtZIP16, AtZIP68,and AtGBF1, members of bZIP TF group G, were shown to beunder redox control as well. These bZIPs bind to G boxes thatare enriched in high light-regulated genes; for example, in thegene coding for light harvesting chlorophyll a/b bindingprotein Lhcb2.4. Cys330 is conserved among the G-groupbZIPs and should be reduced for optimal target gene binding(107). High light-responsive genes belong to a regulon ofgenes that is under the control of retrograde signals which arereleased from the organelles in order to control nuclear geneexpression. Several types of signals and signaling pathwayscontribute to the operation control of nuclear gene expres-sion (92). Redox- and ROS-related cues participate in the ret-rograde control network (101). TFs belonging to the TCP(teosinte branched1-cycloidea-PCF) control developmentalprocesses such as cell division and morphology. The con-served Cys20 present in the DNA binding and dimerizationdomain conveys redox sensitivity of TCP activity (122). Theseexamples show that redox PTMs constitute a commonly en-countered mechanism which controls the binding of TFs totheir respective cis element. This can be achieved by redoxmodifications in the DNA binding domain, by allosteric ef-fects, or by inhibition of TF activation. Many minor variationsof the described mechanisms of redox regulation appearpossible; for example, burying the nuclear localization se-quence by redox-dependent binding of interactors or byPTMs, expressional regulation by monomerization instead ofdimerization, and redox-dependent effects on nuclear exportsequence.

Redox Regulation in Stress Acclimation

Redox imbalances and oxidative stress appear to be in-separably connected to any kind of biotic and abiotic stress,particularly at an elevated stress strength (36). Nevertheless,many studies have aimed at dissecting oxidative stress effectsfrom specific responses to the particular stress type such asdrought, salinity, cold, heavy metals, or herbivory (73, 103).Genetic approaches and effector studies provided some clueson redox, ROS, and RNS-related signaling. This is an

1366 DIETZ

interesting and expanding topic that cannot be discussed here,and the reader is referred to expert reviews (1). Stress accli-mation decisively depends on plant hormones that controlspecific sets of defense genes. There exists tight and mutualcross-talk between redox signaling and hormone signaling(87), for example, by ABA (39), auxin (115), SA (48), jasmo-nates (78), and brassinosteroids (83). Ascorbic acid and GSHare major determinants of plant cell redox state and stressresponses. vtc1 mutant with low ascorbic acid levels accu-mulates high amounts of SA and SA conjugates. Mukherjeeet al. (77) suggested that the high SA levels in vtc1 prime themutant to rapidly respond to stress. Recently, a new cross-talkwas established between the thiol-disulfide redox regulatorynetwork of the chloroplast, Cys synthesis and defense geneexpression (Fig. 9). The stromal cyclophilin 20-3 (Cyp20-3)activates plastid Cys synthase by binding to serine acetyl-transferase (SAT1) (25). Cyp20-3 interacts with the abundantchloroplast 2-cysteine peroxiredoxin and contains four Cysresidues that modulate its function (58, 79, 80). Park et al. (88)established that this mechanism is tied into oxylipin signaling.Oxophytodienoic acid (OPDA) is produced in chloroplasts bythe action of LOX, allene oxide synthase, and alleneoxidecyclase. OPDA binds to SAT1 with a very high affinity of< 200 nM (Abb. 9). Stimulation of Cys synthesis causes non-protein thiols to accumulate, leading to an increased reduc-tion potential and activation of defense gene expressionindependent of SA (88). It is tempting to speculate that redoxregulation of TFs participates in this novel retrograde sig-naling pathway that relies on redox signaling from the chlo-roplast to the nucleus.

Redox Control of Development

Redox cues control developmental transition and differ-entiation steps independent of the environment. In roots,maintenance of the oxidizing environment in the quiescentcenter, control of cell division, transition to cell differentiation,and root hair expansion are under redox control. Redox ho-meostasis affects signaling pathways independent of hor-mones. The involved regulatory pathways were geneticallyaddressed by identifying the mutant root meristemless (RML1)with a defect in GSH synthesis (13, 121). Lack of GSH arreststhe cell cycle in the G1 state (121).

An A. thaliana mutant defective in NADPH oxidase showsstunted root growth and short root hairs (29). This mutantphenotype could partly be rescued by exogenous addition ofROS or mimicked by addition of the NADPH oxidase inhib-itor diphenylene iodonium. Changes in redox state of the rootapical meristem determine whether the root continues orceases to grow (19). The accumulation pattern of O2

� - andH2O2 determines root development (26). The TF UPBEAT(bHLH transcription factor UPBEAT [UPB1]), which belongsto the bHLH family, contributes to the switch from cell divi-sion to differentiation (117). UPB1 suppresses the transcrip-tion of peroxidases such as Per39, 40, and 57, which produceH2O2 in the apoplast. The application of H2O2 stimulates theexpression of UPB1, suggesting the existence of a feedbackloop that controls proper accumulation of H2O2 by peroxi-dases (117). UPB1 transcript accumulation is unaffected byauxin and cytokinin. The authors propose a model in whichO2� - is required for the maintenance of cell division while

H2O2 regulates the elongation and differentiation process.

Cell redox state also controls root hair development. Re-cently, the late embryogenesis-abundant protein SAG21 wasdescribed as a novel component in redox-dependent controlof root hair development. SAG21 transcript accumulates inresponse to treatments with ROS-generating reagents (76),under biotic and abiotic stress, and in response to hormonessuch as SA, jasmonic acid, and ethylene (98). SAG21 is local-ized in the mitochondrion. Its expression level in over-expressing and antisense plants affects lateral root formationand root hair elongation (98). SAG21 transcript amountscorrelate with root hair length. Primary root length and thenumber of lateral roots decreased with decreased SAG21mRNA levels. The authors hypothesize a role of SAG21 inmitochondrial redox homeostasis that could interfere withNADPH oxidases. However, this hypothesis needs validationby additional experiments.

Redox cues affect flower development. The maize mutantmale sterile converted anther 1 (msca1) and the rice mutant mi-crosporeless 1 (mil1) are defective in primordial formation forfurther anther development. The msca1 and mil1 loci code forCC-type Grxs (42, 49) that interact with TGA TFs. Redoxcontrol of the activity of PAN by ROXY1 and ROXY 2 (63),HD-Zip III class TFs by Cys oxidation (16), or AtMYB2 by S-nitrosylation (104) demonstrates that redox regulation of TFsis an important regulatory mechanism in the control of cellgrowth, pattern formation, and developmental switching. Allthese examples represent single puzzle pieces that underpinthe fundamental significance of redox-regulated TFs in de-velopment, but many more pieces are needed to complete thepicture.

Outlook

The redox network of the cell involves final redox acceptorsin addition to ROS and RNS. RSS and RCS have been intro-duced, and De Tullio and Asard (18) coined the expressionreactive oxygen, nitrogen, carbonyl, and sulfur species. Thebouquet of final redox acceptors, that is, type and relativereactivity, likely encodes specificity that has been littleaddressed.

Phylogenetic comparisons suggest that redox cues from thechloroplast decisively contribute toward realizing phenotypicplasticity (44). These authors elaborate on the example of coldstress acclimation in cyanobacteria, green algae, crop plants,and A. thaliana mutants. Excess excitation controls plant size,pigmentation, and phenotype by using information from theredox network (44). Redox regulation involves inputs fromvarious redox cues and should not be considered a primitiveswitching mechanism. The bacterial redox-sensitive TF OxyRadopts distinct states with different properties for DNAbinding that are dependent on the redox environment and theinputs received from diverse redox-linked signaling path-ways (52). Low GSNO levels enable S-nitrosylation, H2O2

leads to S-hydroxylation (sulfenic acid form) and increasingGSSG accumulation under severe oxidative stress glutathio-nylation of OxyR. Each of these forms depicts a differentability to bind and transactivate target genes (52). Ourmethods that analyze redox regulation of plant transcriptionhave only partly reached the necessary sophistication to ad-dress this level of complexity. Thus, there is an urgent need toimprove proteome sensitivity to routinely characterize lowabundant proteins and to simultaneously address the various

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1367

PTMs such as nitrosylation, hydroxylation, and glutathiony-lation. Combining such approaches with gradually increasingbiotic and abiotic stress strength is expected to pave the wayfor a systematic understanding of redox regulation.

In the end, only the combination of in vivo, ex vivo, andin vitro results will provide conclusive pictures on the rathercomplex mechanism of redox regulation of transcription.

Acknowledgment

This article, which is the original cited work of its author, wassupported by the Deutsche Forschungsgemeinschaft (Di 346).

References

1. Apel K and Hirt H. Reactive oxygen species: metabolism,oxidative stress, and signal transduction. Annu Rev PlantBiol 55: 373–399, 2004.

2. Arsova B, Hoja U, Wimmelbacher M, Greiner E, Ustun S,Melzer M, Petersen K, Lein W, and Bornke F. Plastidialthioredoxin z interacts with two fructokinase-like proteinsin a thiol-dependent manner: evidence for an essential rolein chloroplast development in Arabidopsis and Nicotianabenthamiana. Plant Cell 22: 1498–1515, 2010.

3. Baruah A, Simkova K, Apel K, and Laloi C. Arabidopsismutants reveal multiple singlet oxygen signaling pathwaysinvolved in stress response and development. Plant MolBiol 70: 547–563, 2009.

4. Baudouin E. The language of nitric oxide signalling. PlantBiol 13: 233–242, 2011.

5. Biehler K and Fock H. Evidence for the contribution of theMehler-peroxidase reaction in dissipating excess electronsin drought-stressed wheat. Plant Physiol 112: 265–272, 1996.

6. Bienert GP, Møller AL, Kristiansen KA, Schulz A, MøllerIM, Schjoerring JK, and Jahn TP. Specific aquaporins facil-itate the diffusion of hydrogen peroxide across membranes.J Biol Chem 282: 1183–1192, 2006.

7. Bindschedler LV, Dewdney J, Blee KA, Stone JM, Asai T,Plotnikov J, Denoux C, Hayes T, Gerrish C, Davies DR,Ausubel FM, and Bolwell GP. Peroxidase-dependent apo-plastic oxidative burst in Arabidopsis required for pathogenresistance. Plant J 47: 851–863, 2006.

8. Brigelius-Flohe R and Flohe L. Basic principles andemerging concepts in the redox control of transcriptionfactors. Antioxid Redox Signal 15: 2335–2381, 2011.

9. Brosche M and Kangasjarvi J. Low antioxidant concentra-tions impact on multiple signalling pathways in Arabidopsisthaliana partly through NPR1. J Exp Bot 63: 1849–1861, 2012.

10. Buchanan BB and Balmer Y. Redox regulation: a broaden-ing horizon. Annu Rev Plant Biol 56: 187–220, 2005.

11. Buzek J, Latonen L, Kurki S, Peltonen K, and Laiho M.Redox state of tumor suppressor p53 regulates its se-quence-specific DNA binding in DNA-damaged cells bycysteine 277. Nucleic Acids Res 30: 2340–2348, 2002.

12. Cao H, Glazebrook J, Clarke JD, Volko S, and Dong X. TheArabidopsis NPR1 gene that controls systemic acquired re-sistance encodes a novel protein containing ankyrin re-peats. Cell 88: 57–63, 1997.

13. Cheng JC, Seeley KA, and Sung ZR. RML1 and RML2,Arabidopsis genes required for cell proliferation at the roottip. Plant Physiol 107: 365–376, 1995.

14. Chibani K, Tarrago L, Schurmann P, Jacquot JP, andRouhier N. Biochemical properties of poplar thioredoxin z.FEBS Lett 585: 1077–1081, 2011.

15. Cho Y, Gorina S, Jeffrey PD, and Pavletich NP. Crystalstructure of a p53 tumor suppressor-DNA complex: un-derstanding tumorigenic mutations. Science 265: 346–355,1994.

16. Comelli RN and Gonzalez DH. Conserved homeodomaincysteines confer redox sensitivity and influence the DNAbinding properties of plant class III HD-Zip proteins. ArchBiochem Biophys 467: 41–47, 2007.

17. Couturier J, Vignols F, Jacquot JP, and Rouhier N. Glu-tathione- and glutaredoxin-dependent reduction of methi-onine sulfoxide reductase A. FEBS Lett 586: 3894–3899,2012.

18. De Tullio MC and Asard H. Molecules tell stories: reactiveoxygen, nitrogen, carbonyl and sulphur species. PlantPhysiol Biochem 59: 1–2, 2012.

19. De Tullio MC, Jiang K, and Feldman LJ. Redox regulationof root apical meristem organization: connecting root de-velopment to its environment. Plant Physiol Biochem 48:328–336, 2010.

20. Despres C, Chubak C, Rochon A, Clark R, Bethune T,Desveaux D, and Fobert PR. The Arabidopsis NPR1 diseaseresistance protein is a novel cofactor that confers redoxregulation of DNA binding activity to the basic domain/leucine zipper transcription factor TGA1. Plant Cell 15:2181–2191, 2003.

21. Dietz KJ. Plant thiol enzymes and thiol homeostasis in re-lation to thiol-dependent redox regulation and oxidativestress. In: Antioxidants and Reactive Oxygen Species in Plants,edited by Smirnoff N. Oxford, United Kingdom: BlackwellPubl., 2005, pp. 25–52.

22. Dietz KJ. Redox signal integration: from stimulus to net-works and genes. Physiol Plant 133: 459–468, 2008.

23. Dietz KJ, Jacquot JP, and Harris G. Hubs and bottlenecks inplant molecular signalling networks. New Phytol 188: 919–938, 2010.

24. Dietz KJ, Vogel MO, and Viehhauser A. AP2/EREBPtranscription factors are part of gene regulatory networksand integrate metabolic, hormonal and environmentalsignals in stress acclimation and retrograde signalling.Protoplasma 245: 3–14, 2010.

25. Dominguez-Solis JR, He Z, Lima A, Ting J, Buchanan BB,and Luan S. A cyclophilin links redox and light signals tocysteine biosynthesis and stress responses in chloroplasts.Proc Natl Acad Sci USA 105: 16386–16391, 2008.

26. Dunand C, Crevecoeur M, and Penel C. Distribution ofsuperoxide and hydrogen peroxide in Arabidopsis root andtheir influence on root development: possible interactionwith peroxidases. New Phytol 174, 332–341, 2007.

27. Dynan WS and Tjian R. Isolation of transcription factorsthat discriminate between different promoters recognizedby RNA polymerase II. Cell 32: 669–680, 1983.

28. Ferrer-Sueta G, Manta B, Botti H, Radi R, Trujillo M, andDenicola A. Factors affecting protein thiol reactivity andspecificity in peroxide reduction. Chem Res Toxicol 24, 434–450, 2011.

29. Foreman J, Demidchik V, Bothwell JH, Mylona P, MiedemaH, Torres MA, Linstead P, Costa S, Brownlee C, Jones JD,Davies JM, and Dolan L. Reactive oxygen species producedby NADPH oxidase regulate plant cell growth. Nature 422:442–446, 2003.

30. Forman HJ, Fukuto JM, and Torres M. Redox signaling:thiol chemistry defines which reactive oxygen and nitrogenspecies can act as second messengers. Am J Physiol CellPhysiol 287: C246–C256, 2004.

1368 DIETZ

31. Foyer CH and Noctor G. Ascorbate and glutathione: theheart of the redox hub. Plant Physiol 155: 2–18, 2011.

32. Fuchs S, Grill E, Meskiene I, and Schweighofer A. Type 2Cprotein phosphatases in plants. FEBS J 280: 681–693, 2013.

33. Gallogly MM, Starke DW, and Mieyal JJ. Mechanistic andkinetic details of catalysis of thiol-disulfide exchange byglutaredoxins and potential mechanisms of regulation.Antioxid Redox Signal 11: 1059–1081, 2009.

34. Gatz C. From pioneers to team players: TGA transcriptionfactors provide a molecular link between different stresspathways. Mol Plant Microbe Interact 26: 151–159, 2013.

35. Gibbs DJ, Lee SC, Isa NM, Gramuglia S, Fukao T, BasselGW, Correia CS, Corbineau F, Theodoulou FL, Bailey-Serres J, and Holdsworth MJ. Homeostatic response tohypoxia is regulated by the N-end rule pathway in plants.Nature 479: 415–418, 2011.

36. Gill SS and Tuteja N. Reactive oxygen species and antiox-idant machinery in abiotic stress tolerance in crop plants.Plant Physiol Biochem 48: 909–930, 2010.

37. Grant CM. Regulation of translation by hydrogen peroxide.Antioxid Redox Signal 15: 191–203, 2011.

38. Gupta R and Luan S. Redox control of protein tyrosinephosphatases and mitogen-activated protein kinases inplants. Plant Physiol 132: 1149–1152, 2003.

39. He J, Duan Y, Hua D, Fan G, Wang L, Liu Y, Chen Z, HanL, Qu LJ, and Gong Z. DEXH box RNA helicase-mediatedmitochondrial reactive oxygen species production in Ara-bidopsis mediates crosstalk between abscisic acid and auxinsignaling. Plant Cell 24: 1815–1833, 2012.

40. Herrmann JM and Dick TP. Redox biology on the rise. BiolChem 393: 999–1004, 2012.

41. Hirota K, Matsui M, Iwata S, Nishiyama A, Mori K, andYodoi J. AP-1 transcriptional activity is regulated by a di-rect association between thioredoxin and Ref-1. Proc NatlAcad Sci U S A 94: 3633–3538, 1997.

42. Hong L, Tang D, Zhu K, Wang K, Li M, and Cheng Z.Somatic and reproductive cell development in rice anther isregulated by a putative glutaredoxin. Plant Cell 24: 577–588,2012.

43. Hubbard KE, Nishimura N, Hitomi K, Getzoff ED, andSchroeder JI. Early abscisic acid signal transduction mech-anisms: newly discovered components and newly emerg-ing questions. Genes Dev 24: 1695–1708, 2010.

44. Huner NP, Bode R, Dahal K, Hollis L, Rosso D, Krol M, andIvanov AG. Chloroplast redox imbalance governs pheno-typic plasticity: the ‘‘grand design of photosynthesis’’ re-visited. Front Plant Sci 3: 255, 2012.

45. Jacquot JP, Dietz KJ, Rouhier N, Meux E, Lallement PA,Selles B, and Hecker A. Redox regulation in plants: gluta-thione and ‘‘redoxin’’-related families. In: Oxidative Stressand Redox Regulation, edited by Jakob U and Reichmann D.Dordrecht, Netherlands: Springer Science and BusinessMedia, 2013, pp. 213–231.

46. Jakoby M, Weisshaar B, Droge-Laser W, Vicente-CarbajosaJ, Tiedemann J, Kroj T, and Parcy F. bZIP research group.bZIP transcription factors in Arabidopsis. Trends Plant Sci 7:106–111, 2002.

47. Jung HS, Crisp PA, Estavillo GM, Cole B, Hing AX,Mockler TC, Pogson BJ, and Chory J. Subset of heat-shocktranscription factors required for the early response ofArabidopsis to excess light. Proc Natl Acad Sci U S A 110:14474–14479, 2013.

48. Karpinski S, Szechynska-Hebda M, Wituszynska W, andBurdiak P. Light acclimation, retrograde signalling, cell

death and immune defences in plants. Plant Cell Environ 36:736–744, 2013.

49. Kelliher T and Walbot V. Hypoxia triggers meiotic fateacquisition in maize. Science 337: 345–348, 2012.

50. Kesarwani M, Yoo J, and Dong X. Genetic interactions ofTGA transcription factors in the regulation of pathogenesis-related genes and disease resistance in Arabidopsis. PlantPhysiol 144: 336–346, 2007.

51. Kim C, Meskauskiene R, Zhang S, Lee KP, LakshmananAshok M, Blajecka K, Herrfurth C, Feussner I, and Apel K.Chloroplasts of Arabidopsis are the source and a primarytarget of a plant-specific programmed cell death signalingpathway. Plant Cell 24: 3026–3039, 2012.

52. Kim SO, Merchant K, Nudelman R, Beyer WF, Jr., Keng T,DeAngelo J, Hausladen A, and Stamler JS. OxyR: a molec-ular code for redox-related signaling. Cell 109: 383–396, 2002.

53. Klatt P, Molina EP, De Lacoba MG, Padilla CA, Martinez-Galesteo E, Barcena JA, and Lamas S. Redox regulation ofc-Jun DNA binding by reversible S-glutathiolation. FASEB J13: 1481–1490, 1999.

54. Klein P, Seidel T, Stocker B, and Dietz KJ. The membrane-tethered transcription factor ANAC089 serves as redox-dependent suppressor of stromal ascorbate peroxidasegene expression. Front Plant Sci 3: 247, 2012.

55. Knoepfel L, Steinkuhler C, Carrı MT, and Rotilio G. Role ofzinc-coordination and of the glutathione redox couple inthe redox susceptibility of human transcription factor Sp1.Biochem Biophys Res Commun 201: 871–877, 1994.

56. Kovtun Y, Chiu WL, Tena G, and Sheen J. Functionalanalysis of oxidative stress-activated mitogen-activatedprotein kinase cascade in plants. Proc Natl Acad Sci U S A97: 2940–2945, 2000.

57. Krieger-Liszkay A, Fufezan C, and Trebst A. Singlet oxy-gen production in photosystem II and related protectionmechanism. Photosynth Res 98: 551–564, 2008.

58. Laxa M, Konig J, Dietz KJ, and Kandlbinder A. Role of thecysteine residues in Arabidopsis thaliana cyclophilin CYP20-3 in peptidyl-prolyl cis-trans isomerase and redox-relatedfunctions. Biochem J 401: 287–297, 2007.

59. Lee C, Lee SM, Mukhopadhyay P, Kim SJ, Lee SC, Ahn WS,Yu MH, Storz G, and Ryu SE. Redox regulation of OxyRrequires specific disulfide bond formation involving a rapidkinetic reaction path. Nat Struct Mol Biol 11: 1179–1185,2004.

60. Lee HM, Dietz KJ, and Hofestadt R. Prediction of thior-edoxin and glutaredoxin target proteins by identifying re-versibly oxidized cysteinyl residues. J Integr Bioinform 7: 3,2010.

61. Li B, Zhao Y, Liang L, Ren H, Xing Y, Chen L, Sun M, WangY, Han Y, Jia H, Huang C, Wu Z, and Jia W. Purification andcharacterization of ZmRIP1, a novel reductant-inhibitedprotein tyrosine phosphatase from maize. Plant Physiol 159:671–681, 2012.

62. Li S, Gutsche N, and Zachgo S. The ROXY1 C-terminalL**LL motif is essential for the interaction with TGA tran-scription factors. Plant Physiol 157: 2056–2068, 2011.

63. Li S, Lauri A, Ziemann M, Busch A, Bhave M, and ZachgoS. Nuclear activity of ROXY1, a glutaredoxin interactingwith TGA factors, is required for petal development inArabidopsis thaliana. Plant Cell 21: 429–441, 2009.

64. Licausi F. Molecular elements of low-oxygen signaling inplants. Physiol Plant 148: 1–8, 2013.

65. Licausi F, Kosmacz M, Weits DA, Giuntoli B, Giorgi FM,Voesenek LA, Perata P, van and Dongen JT. Oxygen

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1369

sensing in plants is mediated by an N-end rule pathway forprotein destabilization. Nature 479: 419–422, 2011.

66. Lindahl M, Mata-Cabana A, and Kieselbach T. The dis-ulfide proteome and other reactive cysteine proteomes:analysis and functional significance. Antioxid Redox Signal14: 2581–2642, 2011.

67. Lindermayr C, Saalbach G, and Durner J. Proteomic iden-tification of S-nitrosylated proteins in Arabidopsis. PlantPhysiol 137: 921–930, 2005.

68. Lindermayr C, Sell S, Muller B, Leister D, and Durner J.Redox regulation of the NPR1-TGA1 system of Arabidopsisthaliana by nitric oxide. Plant Cell 22: 2894–2907, 2010.

69. Meinhard M and Grill E. Hydrogen peroxide is a regulatorof ABI1, a protein phosphatase 2C from Arabidopsis. FEBSLett 508: 443–446, 2001.

70. Meskauskiene R, Nater M, Goslings D, Kessler F, op denCamp R, and Apel K. FLU: a negative regulator of chlo-rophyll biosynthesis in Arabidopsis thaliana. Proc Natl AcadSci U S A 98: 12826–12831, 2001.

71. Miller G and Mittler R. Could heat shock transcriptionfactors function as hydrogen peroxide sensors in plants?Ann Bot 98: 279–288, 2006.

72. Miller G, Schlauch K, Tam R, Cortes D, Torres MA, ShulaevV, Dangl JL, and Mittler R. The plant NADPH oxidaseRbohD mediates rapid systemic signaling in response todiverse stimuli. Sci Signal 2: 84, 2009.

73. Miller G, Suzuki N, Ciftci-Yilmaz S, and Mittler R. Reactiveoxygen species homeostasis and signalling during droughtand salinity stresses. Plant Cell Environ 33: 453–467, 2010.

74. Moreno J, Garcıa-Murria MJ, and Marın-Navarro J. Redoxmodulation of Rubisco conformation and activity throughits cysteine residues. J Exp Bot 59: 1605–1614, 2008.

75. Mou Z, Fan WH, and Dong XN. Inducers of plant systemicacquired resistance regulate NPR1 function through redoxchanges. Cell 113: 935–944, 2003.

76. Mowla SB, Cuypers A, Driscoll SP, Kiddle G, Thomson J,Foyer CH, and Theodoulou FL. Yeast complementationreveals a role for an Arabidopsis thaliana late embryogenesisabundant (LEA)-like protein in oxidative stress tolerance.Plant J 48: 743–756, 2006.

77. Mukherjee M, Larrimore KE, Ahmed NJ, Bedick TS, Bar-ghouthi NT, Traw MB, and Barth C. Ascorbic acid defi-ciency in Arabidopsis thaliana induce constitutive primingthat is dependent on hydrogen peroxide, salicylic acid, andthe NPR1 gene. Mol Plant Microbe Interact 23: 340–351, 2010.

78. Mur LA, Kenton P, Atzorn R, Miersch O, and WasternackC. The outcomes of concentration-specific interactions be-tween salicylate and jasmonate signaling include synergy,antagonism, and oxidative stress leading to cell death. PlantPhysiol 140: 249–262, 2006.

79. Muthuramalingam M, Matros A, Scheibe R, Mock HP, andDietz KJ. The hydrogen peroxide-sensitive proteome of thechloroplast in vitro and in vivo. Front Plant Sci 4: 54, 2013.

80. Muthuramalingam M, Seidel T, Laxa M, Nunes de MirandaSM, Gartner F, Stroher E, Kandlbinder A, and Dietz KJ.Multiple redox and non-redox interactions define 2-Cysperoxiredoxin as a regulatory hub in the chloroplast. MolPlant 2: 1273–1288, 2009.

81. Nagi P. Kinetics and mechanisms of thiol-disulfide ex-change covering direct substitution and thiol oxidation-mediated pathways. Antioxid Redox Signal 18: 1623–1641,2013.

82. Nakano Y and Asada K. Purification of ascorbate peroxidasein spinach chloroplasts—its inactivation in ascorbate-

depleted medium and reactivation by monodehydro-ascorbate radical. Plant Cell Physiol 28: 131–140, 1987.

83. Nie WF, Wang MM, Xia XJ, Zhou YH, Shi K, Chen Z, andYu JQ. Silencing of tomato RBOH1 and MPK2 abolishesbrassinosteroid-induced H2O2 generation and stress toler-ance. Plant Cell Environ 36: 789–803, 2013.

84. Nover L, Bharti K, Doring P, Mishra SK, Ganguli A, andScharf KD. Arabidopsis and the heat stress transcriptionfactor world: how many heat stress transcription factors dowe need? Cell Stress Chaperones 6: 177–189, 2001.

85. Okamoto K, Tanaka H, Ogawa H, Makino Y, Eguchi H,Hayashi S, Yoshikawa N, Poellinger L, Umesono K, andMakino I. Redox-dependent regulation of nuclear import ofthe glucocorticoid receptor. J Biol Chem 274: 10363–10371,1999.

86. O’Loghlen A, Perez-Morgado MI, Salinas M, and MartınME. Reversible inhibition of the protein phosphatase 1 byhydrogen peroxide. Potential regulation of eIF2 alphaphosphorylation in differentiated PC12 cells. Arch BiochemBiophys 417: 194–202, 2003.

87. Overmyer K, Brosche M, and Kangasjarvi J. Reactive oxy-gen species and hormonal control of cell death. Trends PlantSci 8: 335–342, 2003.

88. Park SW, Li W, Viehhauser A, Kim S, Nilsson AK, An-dersson MX, Kittle JD, Ambavaram MMR, Luan S, EskerAR, Tholl D, Ellerstr}om M, Coaker G, Mitchell TK, PereiraA, Dietz KJ, and Lawrence CB. Cyclophilin 20-3 relays a 12-oxo-phytodienoic acid signal during stress responsive reg-ulation of cellular redox homeostasis. Proc Natl Acad Sci U SA 110: 9559–9564, 2013.

89. Paulech J, Solis N, Edwards AV, Puckeridge M, White MY,and Cordwell SJ. Large-scale capture of peptides contain-ing reversibly oxidized cysteines by thiol-disulfide ex-change applied to the myocardial redox proteome. AnalChem 85: 3774–3780, 2013.

90. Pesaresi P, Hertle A, Pribil M, Kleine T, Wagner R, StrisselH, Ihnatowicz A, Bonardi V, Scharfenberg M, Schneider A,Pfannschmidt T, and Leister D. Arabidopsis STN7 kinaseprovides a link between short- and long-term photosyn-thetic acclimation. Plant Cell 21: 2402–2423, 2009.

91. Pfalz J, Liere K, Kandlbinder A, Dietz KJ, and Oelmuller R.pTAC2, - 6, and - 12 are components of the transcrip-tionally active plastid chromosome that are required forplastid gene expression. Plant Cell 18: 176–197, 2006.

92. Pfannschmidt T. Plastidial retrograde signalling - a true‘‘plastid factor’’ or just metabolite signatures? Trends PlantSci 15: 427–435, 2010.

93. Plafker KS, Nguyen L, Barneche M, Mirza S, Crawford D,and Plafker SM. The ubiquitin-conjugating enzyme UbcM2can regulate the stability and activity of the antioxidanttranscription factor Nrf2. J Biol Chem 285: 23064–23074,2010.

94. Rabilloud T, Heller M, Gasnier F, Luche S, Rey C, Ae-bersold R, Benahmed M, Louisot P, and Lunardi J. Pro-teomics analysis of cellular response to oxidative stress.Evidence for in vivo overoxidation of peroxiredoxins attheir active site. J Biol Chem 277: 19396–19401, 2002.

95. Rhoads DM and Subbaiah CC. Mitochondrial retrograderegulation in plants. Mitochondrion 7: 177–194, 2007.

96. Romero-Puertas MC, Laxa M, Matte A, Zaninotto F, Fin-kemeier I, Jones AM, Perazzolli M, Vandelle E, Dietz KJ,and Delledonne M. S-nitrosylation of peroxiredoxin II Epromotes peroxynitrite-mediated tyrosine nitration. PlantCell 19: 4120–4130, 2007.

1370 DIETZ

97. Rouhier N, Villarejo A, Srivastava M, Gelhaye E, Keech O,Droux M, Finkemeier I, Samuelsson G, Dietz KJ, Jacquot JP,and Wingsle G. Identification of plant glutaredoxin targets.Antioxid Redox Signal 7: 919–929, 2005.

98. Salleh FM, Evans K, Goodall B, Machin H, Mowla SB, MurLA, Runions J, Theodoulou FL, Foyer CH, and Rogers HJ.A novel function for a redox-related LEA protein (SAG21/AtLEA5) in root development and biotic stress responses.Plant Cell Environ 35: 418–429, 2012.

99. Sanchez R, Riddle M, Woo J, and Momand J. Prediction ofreversibly oxidized protein cysteine thiols using proteinstructure properties. Protein Sci 17: 473–481, 2008.

100. Scharf KD, Berberich T, Ebersberger I, and Nover L. Theplant heat stress transcription factor (Hsf) family: structure,function and evolution. Biochim Biophys Acta 1819: 104–119,2012.

101. Scheibe R and Dietz KJ. Reduction-oxidation network forflexible adjustment of cellular metabolism in photoauto-trophic cells. Plant Cell Environ 35: 202–216, 2012.

102. Schindler U, Beckmann H, and Cashmore AR. TGA1 andG-box binding factors: two distinct classes of Arabidopsisleucine zipper proteins compete for the G-box-like elementTGACGTGG. Plant Cell 4: 1309–1319, 1992.

103. Schutzendubel A and Polle A. Plant responses to abioticstresses: heavy metal-induced oxidative stress and protec-tion by mycorrhization. J Exp Bot 53: 1351–1365, 2002.

104. Serpa V, Vernal J, Lamattina L, Grotewold E, Cassia R, andTerenzi H. Inhibition of AtMYB2 DNA-binding by nitricoxide involves cysteine S-nitrosylation. Biochem Biophys ResCommun 361: 1048–1053, 2007.

105. Serrato AJ, Perez-Ruiz JM, and Cejudo FJ. Cloning ofthioredoxin h reductase and characterization of the thior-edoxin reductase-thioredoxin h system from wheat. Bio-chem J 367: 491–497, 2002.

106. Shaikhali J, Heiber I, Seidel T, Stroher E, Hiltscher H,Birkmann S, Dietz KJ, and Baier M. The redox-sensitivetranscription factor Rap2.4a controls nuclear expression of2-Cys peroxiredoxin A and other chloroplast antioxidantenzymes. BMC Plant Biol 8: 48, 2008.

107. Shaikhali J, Noren L, de Dios Barajas-Lopez J, Srivastava V,Konig J, Sauer UH, Wingsle G, Dietz KJ, and Strand A.Redox-mediated mechanisms regulate DNA binding ac-tivity of the G-group of basic region leucine zipper (bZIP)transcription factors in Arabidopsis. J Biol Chem 287: 27510–27525, 2012.

108. Shao N, Beck CF, Lemaire SD, and Krieger-Liszkay A.Photosynthetic electron flow affects H2O2 signaling by in-activation of catalase in Chlamydomonas reinhardtii. Planta228: 1055–1066, 2008.

109. Shapiguzov A, Vainonen JP, Wrzaczek M, and KangasjarviJ. ROS-talk—how the apoplast, the chloroplast, and the nu-cleus get the message through. Front Plant Sci 3: 292, 2012.

110. Stroher E and Dietz KJ. The dynamic thiol-disulphide re-dox proteome of the Arabidopsis thaliana chloroplast as re-vealed by differential electrophoretic mobility. Physiol Plant133: 566–583, 2008.

111. Stroher E and Millar AH. The biological roles of glutar-edoxins. Biochem J 446: 333–348, 2012.

112. Sun X, Feng P, Xu X, Guo H, Ma J, Chi W, Lin R, Lu C, andZhang L. A chloroplast envelope-bound PHD transcriptionfactor mediates chloroplast signals to the nucleus. NatCommun 2: 477, 2011.

113. Tada Y, Spoel SH, Pajerowska-Mukhtar K, Mou ZI, SongJQ, Wang C, Zuo JR, and Dong XN. Plant immunity re-

quires conformational changes of NPR1 via S-nitrosylationand thioredoxin. Science 321: 952–956, 2008.

114. Tasaki T, Sriram SM, Park KS, and Kwon YT. The N-endrule pathway. Annu Rev Biochem 81: 261–289, 2012.

115. Tognetti VB, Muhlenbock P, and Van Breusegem F. Stresshomeostasis - the redox and auxin perspective. Plant CellEnviron 35: 321–333, 2012.

116. Triantaphylides C, Krischke M, Hoeberichts FA, Ksas B,Gresser G, Havaux M, Van Breusegem F, and Mueller MJ.Singlet oxygen is the major reactive oxygen species in-volved in photooxidative damage to plants. Plant Physiol148: 960–968, 2008.

117. Tsukagoshi H, Busch W, and Benfey PN. Transcriptionalregulation of ROS controls transition from proliferation todifferentiation in the root. Cell 143: 606–616, 2010.

118. Turkeri H, Schweer J, and Link G. Phylogenetic and func-tional features of the plastid transcription kinase cpCK2from Arabidopsis signify a role of cysteinyl SH-groups inregulatory phosphorylation of plastid sigma factors. FEBS J279: 395–409, 2012.

119. Umbach AL, Zarkovic J, Yu J, Ruckle ME, McIntosh L,Hock JJ, Bingham S, White SJ, George RM, Subbaiah CC,and Rhoads DM. Comparison of intact Arabidopsis thalianaleaf transcript profiles during treatment with inhibitors ofmitochondrial electron transport and TCA cycle. PLoS One7: e44339, 2012.

120. Vanderauwera S, Zimmermann P, Rombauts S, Vandena-beele S, Langebartels C, Gruissem W, Inze D, and VanBreusegem F. Genome-wide analysis of hydrogen perox-ide-regulated gene expression in Arabidopsis reveals a highlight-induced transcriptional cluster involved in anthocya-nin biosynthesis. Plant Physiol 139: 806–821, 2005.

121. Vernoux T, Wilson RC, Seeley KA, Reichheld JP, Muroy S,Brown S, Maughan SC, Cobbett CS, Van Montagu M, InzeD, May MJ, and Sung ZR. The ROOT MERISTEMLESS1/CADMIUM SENSITIVE2 gene defines a glutathione-de-pendent pathway involved in initiation and maintenance ofcell division during postembryonic root development. PlantCell 12: 97–110, 2000.

122. Viola IL, Guttlein LN, and Gonzalez DH. Redox modula-tion of plant developmental regulators from the class I TCPtranscription factor family. Plant Physiol 162: 1434–1447,2013.

123. Wakabayashi N, Dinkova-Kostova AT, Holtzclaw WD,Kang MI, Kobayashi A, Yamamoto M, Kensler TW, andTalalay P. Protection against electrophile and oxidant stressby induction of the phase 2 response: fate of cysteines of theKeap1 sensor modified by inducers. Proc Natl Acad Sci U SA 101: 2040–2045, 2004.

124. Wang H, Wang S, Lu Y, Alvarez S, Hicks LM, Ge X, andXia Y. Proteomic analysis of early-responsive redox-sensitiveproteins in Arabidopsis. J Proteome Res 11: 412–424, 2011.

125. Watanabe S, Kita A, Kobayashi K, and Miki K. Crystalstructure of the [2Fe-2S] oxidative-stress sensor SoxR boundto DNA. Proc Natl Acad Sci U S A 105: 4121–4126, 2008.

126. Webster KA, Prentice H, and Bishopric NH. Oxidation ofzinc finger transcription factors: physiological conse-quences. Antioxid Redox Signal 3: 535–548, 2001.

127. Wells WW, Xu DP, Yang YF, and Rocque PA. Mammalianthioltransferase (glutaredoxin) and protein disulfide isom-erase have dehydroascorbate reductase activity. J Biol Chem265: 15361–15364, 1990.

128. Welsch R, Maass D, Voegel T, Dellapenna D, and Beyer P.Transcription factor RAP2.2 and its interacting partner

REDOX SIGNALING IN TRANSCRIPTIONAL CONTROL 1371

SINAT2: stable elements in the carotenogenesis of Arabi-dopsis leaves. Plant Physiol 145: 1073–1085, 2007.

129. Wind JJ, Peviani A, Snel B, Hanson J, and Smeekens SC.ABI4: versatile activator and repressor. Trends Plant Sci 18:125–132, 2013.

130. Xing S, Rosso MG, and Zachgo S. ROXY1, a member of theplant glutaredoxin family, is required for petal develop-ment in Arabidopsis thaliana. Development 132: 1555–1565,2005.

131. Xu Q, Fu HH, Gupta R, and Luan S. Molecular character-ization of a tyrosine-specific protein phosphatase encodedby a stress-responsive gene in Arabidopsis. Plant Cell 10:849–857, 1998.

132. Yamasaki K, Kigawa T, Inoue M, Tateno M, Yamasaki T,Yabuki T, Aoki M, Seki E, Matsuda T, Nunokawa E, Ishi-zuka Y, Terada T, Shirouzu M, Osanai T, Tanaka A, Seki M,Shinozaki K, and Yokoyama S. A novel zinc-binding motifrevealed by solution structures of DNA-binding domainsof Arabidopsis SBP-family transcription factors. J Mol Biol337: 49–63, 2004.

133. Zhang Y, Tessaro MJ, Lassner M, and Li X. Knockoutanalysis of Arabidopsis transcription factors TGA2, TGA5,and TGA6 reveals their redundant and essential roles insystemic acquired resistance. Plant Cell 15: 2647–2653, 2003.

Address correspondence to:Prof. Karl-Josef Dietz

Biochemistry and Physiology of PlantsFaculty of Biology—W5-134

Bielefeld UniversityBielefeld 33501

Germany

E-mail: karl-josef.dietz@uni-bielefeld.de

Date of first submission to ARS Central, October 14, 2013; dateof acceptance, November 1, 2013.

Abbreviations Used

ABA¼ abscisic acidABI¼ abscisic acid insensitive

ACBP¼ acyl-CoA binding proteinsANAC¼ abscisic acid responsive NAC transcription

factorAO¼ ascorbate oxidase

AP1¼ activator protein 1Asc¼ ascorbate

CAT¼ catalasec-Jun¼ subunit of AP-1

cpCK2¼ chloroplast casein kinase 2Cyp20-3¼ stromal cyclophilin 20-3

DHA¼dehydroascorbateDHAR¼dehydroascorbate reductase

ERF¼ ethylene response factorFd¼ ferredoxin

FNR¼ ferredoxin-dependent NADP reductaseFTR¼ ferredoxin-dependent thioredoxin reductaseGR¼ glucocorticoid receptorGR¼ glutathione reductaseGrx¼ glutaredoxin

GSH¼ glutathioneGSNO¼nitrosoglutathioneGSSG¼ oxidized glutathione

HD¼homeodomainHIF¼hypoxia-induced factorHsf¼heat shock transcription factorsJA¼ jasmonic acid

Keap1¼Kelch domain-containing partnerLOX¼ lipoxygenase

MAPK¼mitogen-activated protein kinaseMyb¼myeloblastosis oncoprotein (MYB)-like

transcription factorNERP¼N-end rule pathway of protein degradation

NiNOR¼nitrite nitric oxide reductaseNLS¼nuclear localization sequenceNR¼nitrate reductase

Nrf2¼nuclear factor E2-related factor 2Nrf2¼nuclear redox factor 2NTR¼NADPH-dependent thioredoxin reductase

OPDA¼ oxophytodienoic acidOxyR¼ repressor of H2O2 response in bacteria

P53¼ tumor suppressor protein 53PDI¼protein disulfide isomerasesPET¼photosynthetic electron transport

PHD¼plant homeodomainPrx¼peroxiredoxin

PTM¼post-translational modificationRAP¼ related to apetala

Rboh¼NADPH oxidaseRCS¼ reactive carbonyl speciesRef1¼ redox effector factor 1RES¼ reactive electrophilic speciesRET¼ respiratory electron transportRNS¼ reactive nitrogen speciesROS¼ reactive oxygen speciesRSS¼ reactive sulfur species

RubisCO¼ ribulose-1,5-bisphosphate carboxylaseoxygenase

SA¼ salicylic acidSAR¼ systemic-acquired resistanceSBP¼ SQUAMOSA promoter binding proteins

(SBP)-transcription factorSOD¼ superoxide dismutase

TF¼ transcription factorTGA¼ transcription factor binding to a TGA related

motiveTrx¼ thioredoxin

Trx-z¼ thioredoxin zUFA¼unsaturated fatty acid

UPB1¼ bHLH transcription factor UPBEATYFP¼ yellow fluorescent protein

1372 DIETZ

This article has been cited by:

1. RACHEL H. V. SOUSA, FABRICIO E. L. CARVALHO, CAROL W. RIBEIRO, GISELE PASSAIA, JULIANA R. CUNHA,YUGO LIMA-MELO, MÁRCIA MARGIS-PINHEIRO, JOAQUIM A. G. SILVEIRA. 2014. Peroxisomal APX knockdowntriggers antioxidant mechanisms favourable for coping with high photorespiratory H 2 O 2 induced by CAT deficiency in rice.Plant, Cell & Environment n/a-n/a. [CrossRef]