oxidative stress: molecular perception and transduction of ... · transduces environmental signals...

995

Braz J Med Biol Res 38(7) 2005

Antioxidant gene regulation and expression

Oxidative stress: molecular perceptionand transduction of signals triggeringantioxidant gene defenses

Department of Genetics, North Carolina State University, Raleigh, NC, USAJ.G. Scandalios

Abstract

Molecular oxygen (O2) is the premier biological electron acceptor thatserves vital roles in fundamental cellular functions. However, with thebeneficial properties of O2 comes the inadvertent formation of reactiveoxygen species (ROS) such as superoxide (O2

•-), hydrogen peroxide,and hydroxyl radical (OH•). If unabated, ROS pose a serious threat toor cause the death of aerobic cells. To minimize the damaging effectsof ROS, aerobic organisms evolved non-enzymatic and enzymaticantioxidant defenses. The latter include catalases, peroxidases, super-oxide dismutases, and glutathione S-transferases (GST). CellularROS-sensing mechanisms are not well understood, but a number oftranscription factors that regulate the expression of antioxidant genesare well characterized in prokaryotes and in yeast. In higher eukary-otes, oxidative stress responses are more complex and modulated byseveral regulators. In mammalian systems, two classes of transcrip-tion factors, nuclear factor κB and activator protein-1, are involved inthe oxidative stress response. Antioxidant-specific gene induction,involved in xenobiotic metabolism, is mediated by the “antioxidantresponsive element” (ARE) commonly found in the promoter regionof such genes. ARE is present in mammalian GST, metallothioneine-I and MnSod genes, but has not been found in plant Gst genes.However, ARE is present in the promoter region of the three maizecatalase (Cat) genes. In plants, ROS have been implicated in thedamaging effects of various environmental stress conditions. Manyplant defense genes are activated in response to these conditions,including the three maize Cat and some of the superoxide dismutase(Sod) genes.

CorrespondenceJ.G. Scandalios

Department of Genetics

North Carolina State University

Raleigh, NC 27695

USA

E-mail: [email protected]

Based on a Plenary Lecture presented

at the XXXIII Annual Meeting of the

Brazilian Society of Biochemistry

and Molecular Biology, Caxambu,MG, Brazil, May 15-18, 2004.

Received January 31, 2005

Accepted April 18, 2005

Key words• Catalase• Aging• Telomeres• Gene regulation• Superoxide dismutase• Genomics

Introduction

The environment in which they exist af-fects all living organisms. Whether internalor external to the organism, the environmentis continually changing, and the organismmust adapt if it is to survive. An organismapparently well adapted to its environmentat any one time may be poorly adapted only

a short time later if it cannot modify itsphysiology or behavior in response to chang-ing environmental or metabolic conditions.Organisms that can adjust to changes in theenvironment are likely to exhibit a greaterdegree of adaptiveness than those that can-not. Environmental changes, irrespective ofsource, cause a variety of “stresses” or“shocks” that a cell must face repeatedly,

Brazilian Journal of Medical and Biological Research (2005) 38: 995-1014ISSN 0100-879X Review

996


J.G. Scandalios

and to which its genome must respond in aprogrammed manner for the organism tosurvive. How the genome perceives andtransduces environmental signals to effectthe expression and/or repression of pertinentgenes in a selective manner remains a keyquestion. Examples are responses to light,oxidative stress, pathogenicity, wounding,anaerobiosis, thermal shock, and the “SOS”response in microorganisms (1). Some sens-ing mechanism(s) must be present to alertthe cell to imminent danger, and to triggerthe orderly sequence of events that will miti-gate this danger. In addition, there are ge-nomic responses to unanticipated, unpro-grammed challenges for which the genomeis unprepared, but to which it responds indiscernible though initially unforeseen andunpredictable ways (2). Many, though notall, signals are perceived at the cell surfaceby plasma membrane receptors. Activationof such receptors by mechanisms such asligand binding may lead to alterations inother cellular components, ultimately result-ing in alterations in cell shape, ion conduc-tivity, gene activity, and other cellular func-tions (3). Identification and isolation of mu-tants that are unable to respond, or that re-spond abnormally to a particular signal, mayprovide ways to decipher the mechanismsby which a particular signal is transducedinto a given response. Long before humansbegan manipulating and altering their envi-ronment, organisms from the simplest to themost complex began evolving methods tocope with stressful stimuli. Consequently,most living cells possess an amazing capac-ity to cope with a wide diversity of environ-mental challenges, including natural and syn-thetic toxins, pathogens, extreme tempera-tures, high metal levels, and radiation. Manystudies in the past have demonstrated clear“cause-effect” relationships upon exposureof a given organism or cell to a particularenvironmental factor or stressor. But onlyrecently have certain environmental insultsbeen shown to elicit specific genomic re-

sponses (4). At present, relatively little isknown of the underlying molecular mechan-isms by which the genome perceives envi-ronmental signals and mobilizes the organ-ism to respond. Such information is not onlyinteresting in and of itself, but is also essen-tial in any future attempts to engineer organ-isms for increased tolerance to environmen-tal adversity. Recent dramatic advances inmolecular biology and genomics have madeit possible to investigate the underlyingmechanisms utilized by organisms to copewith environmental stresses. Investigationsof genomic responses to challenge are shed-ding light on unique DNA sequences ca-pable of perceiving stress signals, thus al-lowing the cell or organism to mobilize itsdefenses (5). The general picture emergingfrom recent studies involves the sensing of asignal and the transduction of the signal tothe transcription apparatus to catalyze trans-cription initiation. A signal transduction path-way contains elements that enable a signal tobe transmitted within and between cells andto be translated into an appropriate response.Cells can respond to a variety of environ-mental, physical, and chemical stimuli usinga diverse range of transduction and responsemechanisms. The essential features of a sig-nal transduction pathway comprise a recep-tor (recognition element) capable of detect-ing a stimulus, second messengers (trans-mission elements) such as calcium or phos-phorylation cascades, and response elements(e.g., gene transcription). Such signalingnetworks are now amenable to study anddissection by biochemical and genetic ap-proaches that may elucidate the underlyingmechanisms and lead to the identification ofthe molecules responsible. A thorough under-standing of how organisms perceive, respond,and adapt to changing environmental and de-velopmental stimuli is now attainable.

Gene responses

Terminally differentiated cells express

997



an array of genes required for their stablefunctioning and precise metabolic roles. Agenome can respond in a rapid and specificmanner by selectively decreasing or increas-ing the expression of specific genes. Geneswhose expression is increased during timesof stress presumably are critical to theorganism’s survival under adverse condi-tions. Examination and study of such “stress-responsive” genes have implications for hu-man health and well being, for agriculturalproductivity, and for furthering basic bio-logical knowledge. In addition to aiding theorganism under stress, genomes that aremodified by stress can be utilized to studythe molecular events that occur during peri-ods of increased or decreased gene expres-sion. The mechanisms by which an organ-ism recognizes a signal to alter gene expres-sion and responds to fill that need are impor-tant physiologically and render possible theexamination of gene regulation under vari-ous environmental regimens. The mechan-isms of induction of stress response genesare similar among various organisms exam-ined. Similarities in stress-induced changesin gene expression have been observed for avariety of stressors (3,6-8). Some that havebeen studied in detail include radiation, ther-mal shock, pathogenic infections, anaero-biosis, trauma, photostress, physical wound-ing, oxidative stress, water stress, and heavymetals. In all cases, specific changes in tran-script and/or protein expression have beenobserved in various organisms subjected tosuch challenges.

Recent studies by several laboratories,using a variety of organisms, indicate thatoxidative stress is a common denominatorunderlying many diseases and environmen-tal insults which can lead to cell death invirtually all aerobes (9). It is also becomingclear that a variety of different biotic andabiotic stresses cause their deleterious ef-fects, directly or indirectly, via reactive oxy-gen species (ROS) generation.

For example, numerous toxic environ-

mental chemicals such as xenobiotics, pesti-cides, herbicides, fungicides, ozone, ciga-rette smoke, and radiation cause their harm-ful environmental effects via generation offree radicals and other ROS.

Herein, I will focus on oxidative stress,its causes and consequences, and mechan-isms employed by organisms to cope with it.

Oxygenation of the Earth

Earth is the only planet in our solar sys-tem known to contain molecular oxygen(O2) in its atmosphere and to support aerobiclife. However, when Earth was formed about4.5 billion years ago, its atmosphere wasunlike the present, being primarily reducingand essentially free of oxygen. Most likely,the earliest living organisms were anaerobicheterotrophs living in the primitive oceandepths, shielded from the damaging effectsof solar ionizing radiation. The earliest rela-tively low levels of oxygen were probablythe result of photolytic dissociation of waterby the sun’s ionizing radiation. The bulk ofEarth’s present oxygen concentration (21%O2) is derived from the photosynthetic ac-tivities of cyanobacteria and plants. It hasbeen estimated that Earth contains about 410x 103 Erda moles (Emol = 108 mol) of oxy-gen, and of this, 38.4 x 103 Emol is in thehydrosphere as water. Molecular oxygen ispresent in the atmosphere (37 Emol) and inthe hydrosphere (0.4 Emol) and undergoescontinuous turnover, with the total oxygenexchange estimated at ~15 x 103 Emol/106

years. Aerobic life is responsible for themajor portion of oxygen turnover, with pho-tosynthesis being the main input into theoxygen reservoir, and respiration the mainoutput. The two processes are in approxi-mate equilibrium, and fossil fuel combus-tion is the major source of oxygen loss fromthe reservoir (10,11).

The accumulation of dioxygen (O2) inEarth’s atmosphere permitted the evolutionof the enormous variety of aerobic organ-

998


J.G. Scandalios

isms that use O2 as the electron acceptor,thus providing a higher yield of energy com-pared with fermentation and anaerobic res-piration.

O2, ROS, and oxygen toxicity

In its ground state (its normal configura-tion, O2) molecular oxygen is relativelyunreactive. However, during normal meta-bolic activity, and as a consequence of vari-ous environmental perturbations (e.g., ex-treme temperatures, radiation, xenobiotics,toxins, air pollutants, various biotic and abi-otic stresses, and diseases) O2 is capable ofgiving rise to frightfully reactive excitedstates such as free radicals and derivatives(9,12).

The oxidation powers of O2 are restrictedbecause electrons can only be absorbed fromanother species whose electron spin is anti-parallel to the two unpaired, parallel-spinelectrons in diatomic oxygen. This spin re-striction renders ground state molecular oxy-gen sufficiently unreactive, so that it cannotabstract electrons from other species. How-ever, removal of the spin restriction by add-ing a single e-, or upon transfer of energy tooxygen from a photosensitizer (e.g., chloro-phyll, flavin-containing compounds), in-creases the reactivity of oxygen. Photosensi-tizers can harvest light and energize O2 toform singlet oxygen (1∆gO2), which can in-teract directly with another molecule, trans-ferring the additional energy to the target

molecule.The complete reduction of O2 to water

requires four electrons. O2 has a preferencefor a stepwise univalent pathway of reduc-tion resulting in partially reduced intermedi-ates (Figure 1). The reactive species of re-duced dioxygen include the superoxide radi-cal (O2

•−), hydrogen peroxide (H2O2), andthe hydroxyl radical (OH•). The latter canalso be generated by the interaction of O2

•−

and H2O2 in the presence of metal ions. BothO2

•− and OH• are extremely reactive and cancause molecular damage, leading to celldeath. The hydroxyl radical reacts with vir-tually anything, inflicting indiscriminate andextensive intracellular damage. The O2

•− isthe conjugate base of a weak acid, theperhydroxyl radical (HO2), whose pKa is4.69 ± 0.08. Thus, under acidic conditions,the very reactive perhydroxyl radical maypredominate following a one electron reduc-tion of dioxygen, while at higher pH valuesthe O2

•− is predominant. These and the physi-cally energized form of dioxygen, singletoxygen (1O2), are the biologically most im-portant ROS (Table 1). An activation energyof ~22 kcal/mol is required to raise molecu-lar oxygen (O2) from its ground state to itsfirst singlet state. In higher plants, this en-ergy is readily obtained from light quantavia such transfer molecules as chlorophyll.Unless abated, all of these intermediate oxy-gen species are extremely reactive and cyto-toxic in all organisms (9,13). ROS can inter-act with proteins, lipids, and nucleic acids tocause severe molecular damage (Table 2).Thus, oxygen provides a paradox, in that it isessential for aerobic life, yet in its reducedforms is one of the most toxic substanceswith which life on Earth must cope. ROS arefound in virtually all intracellular organellesor compartments as a consequence of nor-mal metabolic activity. Each organelle orcompartment has potential targets for oxida-tive damage, as well as mechanisms for theelimination of excess ROS accumulation(Figure 2).

Figure 1. Pathways in the univalent reduction of oxygen to water leading to generation ofvarious intermediate reactive oxygen species (ROS).

999



Table 1. Reactive oxygen species of interest in oxidative stress.

Name Notation Some comments and basic sources

Molecular oxygen (triplet ground state) O2; 3Σ Common form of dioxygen gasSinglet oxygen (1st excited singlet state) 1O2; 1∆ Photoinhibition; UV irradiation; PS II e- transfer

reactions (chloroplasts)Superoxide anion O2

•− Formed in many photooxidation reactions(flavoprotein, redox cycling); Mehler reaction inchloroplasts; mitochondrial e- transfer reactions;glyoxysomal photorespiration; peroxisomal activity;nitrogen fixation; reactions of O3 and OH• in apoplasticspace; defense against pathogens; oxidation ofxenobiotics

Hydrogen peroxide H2O2 Formed from O2•− by dismutation; photorespiration;

ß-oxidation; proton-induced decomposition of O2•−;

defense against pathogensHydroxyl radical OH• Decomposition of O3 in apoplastic space; defense

against pathogens; reactions of H2O2 with O2•−

(Haber-Weiss); reactions of H2O2 with Fe2+ (Fenton);highly reactive with all macromolecules

Perhydroxyl radical O2H• Protonated form of O2•−; reactions of O3 and OH• in

apoplastic spaceOzone O3 UV radiation or electrical discharge in stratosphere;

reactions involving combustion products of fossilfuels and UV radiation in troposphere

Table 2. Examples of reactive oxygen species (ROS) damage to lipids, proteins and DNA.

Oxidative damage to lipids• Occurs via several mechanisms of ROS reacting with fatty acids in the membrane lipid bilayer, leading

to membrane leakage and cell death.• In foods, lipid peroxidation causes rancidity and development of undesirable odors and flavors.

Oxidative damage to proteins• Site-specific amino acid modifications (specific amino acids differ in their susceptibility to ROS attack)• Fragmentation of the peptide chain• Aggregation of cross-linked reaction products• Altered electrical charge• Increased susceptibility to proteolysis• Oxidation of Fe-S centers by O2

•− destroys enzymatic function• Oxidation of specific amino acids “marks” proteins for degradation by specific proteases• Oxidation of specific amino acids (e.g., Try) leads to cross-linking

Oxidative damage to DNA• DNA deletions, mutations, translocations• Base degradation, single-strand breakage• Cross-linking of DNA to proteins

Defenses against reactive oxygen

To minimize the damaging effects ofROS, aerobic organisms evolved both non-enzymatic and enzymatic antioxidant de-

fenses (Table 3). Non-enzymatic defensesinclude compounds of intrinsic antioxidantproperties, such as vitamins C and E, gluta-thione, and ß-carotene. Purely enzymaticdefenses, such as superoxide dismutases

1000


J.G. Scandalios

Table 3. Some natural antioxidants.

Non-enzymatic antioxidant molecules

Antioxidant molecule Subcellular location

Ascorbate (vitamin C) Plastid; apoplast; cytosol; vacuoleß-Carotene PlastidGlutathione, reduced (GSH) Plastid; mitochondrion; cytosolPolyamines (e.g., putrescine, spermine) Nucleus; plastid; mitochondrion; cytosolα-Tocopherol (vitamin E) Cell and plastid membranesZeaxanthin Chloroplast

Antioxidant enzymes

Enzyme EC number Subcellular location

Ascorbate peroxidase 1.11.1.11 Plastid stroma and membranesPeroxidases (non-specific) 1.11.1.7 Cytosol; cell wall-boundCatalase 1.11.1.6 Glyoxysome; peroxisome; cytosol;

mitochondriaSuperoxide dismutase (SOD) 1.15.1.1 Cytosol (Cu/ZnSOD); plastid (Cu/ZnSOD;

FeSOD); mitochondrion (MnSOD); peroxisomeDehydroascorbate reductase 1.8.5.1 Cytosol; plastidGlutathione reductase 1.6.4.2 Mitochondrion; cytosol; plastidMonodehydroascorbate reductase 1.6.5.4 Plastid stromaGlutathione S-transferases 2.5.1.18 Cytosol; microsomal

Figure 2. Intracellular antioxidantresources in plant cells. SOD =superoxide dismutase.

(SOD), catalases (CAT) and peroxidases,protect by directly scavenging superoxideradicals and hydrogen peroxide, convertingthem to less reactive species. SODs catalyze

the dismutation of O2•− to H2O2, and CAT

and peroxidases reduce H2O2 to 2H2O. Thesimilarity between the SOD and CAT reac-tions is that each is an oxidation-reduction in

1001



dria; FeSODs are generally found in prokary-otes, in algae and in some higher plant chlo-roplasts; NiSODs have been found in Strep-tomyces. Unlike most other organisms thathave only one of each type of SOD in thevarious cellular compartments, plants havemultiple forms of each type encoded bymore than one gene (Figure 3), indicativethat plants have far more complex antioxi-dant defenses (14,15). Plants also produce alarge variety of small non-enzymatic anti-oxidant compounds as second tier defenses,such as glutathione, ascorbate, tocopherols,flavonoids, alkaloids, and carotenoids in highconcentrations that are capable of quench-ing ROS. The dismutation of O2

•− to O2 +H2O2 by SOD is hardly a bargain, as theresulting H2O2 can react with metal ions,giving rise to the highly toxic OH•. Fortu-nately, CAT come to the rescue by degrad-ing H2O2 to O2 and H2O. Most aerobes,including mammals, possess at least oneform of homotetrameric CAT with ferrihemeat the active sites.

Catalase

CAT is largely, but not exclusively, lo-calized in peroxisomes, wherein many H2O2-producing enzymes reside. Thus CAT, which

Figure 3. Zymograms showing multiplicity of superoxide dismutase (SOD, left) and catalase (CAT, right) in maize.The cytosolic CuZnSOD-4 and SOD-4A co-migrate to the same position, as do all members of the Sod3 multigenefamily (14). Intracellular location is indicated in parentheses. Catalases are tissue-specific in their expression: LE =milky endosperm; COL = coleoptile; SC = scutellum; ALEU = aleurone; PER = pericarp; LF = green leaf. WhenCAT-1 and CAT-2 are co-expressed (ALEU) their subunits interact to generate intergenic heterotetramers. CAT-3does not form heterotetramers in vivo when co-expressed with either CAT-1 or CAT-2 (16). Arrow shows directionof migration in gel.

which the substrate, O2•− for SOD and H2O2

for CAT, is both reductant and oxidant,whereas different reductants are required forthe peroxidases, depending upon their speci-ficities. Under some conditions CAT can actas an efficient peroxidase. SODs deal withthe first product of the univalent reduction ofO2, converting it to H2O2, which must thenbe destroyed by CAT and/or peroxidases.Thus, the SOD and CAT serve, in tandem, asfront-line antioxidant defenses:

O2•− + O2

•− + 2H+ SOD O2 + H2O2 (K2 = 2.4 x 109 M-1 s-1)

H2O2 + H2O2 CAT 2 H2O + O2 (K1 = 1.7 x 107 M-1 s-1)

H2O2 + R(OH)2 Px 2H2O + R(O)2 (K4 = 0.2-1 x 103 M-1 s-1)

Superoxide dismutase

SODs have been isolated and character-ized from a wide variety of organisms. Oneclass consists of SODs with Cu(II) plus Zn(II)at the active site (Cu/ZnSOD), another withMn(III) (MnSOD), a third with Fe(III)(FeSOD), and a fourth with Ni(II/III)(NiSOD). Cu/ZnSODs are generally foundin the cytosol of eukaryotic cells, in chloro-plasts, and in some prokaryotes; MnSODsare found in prokaryotes and in mitochon-

CAT

CuZnSOD-1 (Chl)

CuZnSOD-2 (Cyt)

MnSOD-3 (Mit)

CuZnSOD-4/4A (Cyt)

CuZnSOD-5 (Cyt)

CAT 3

CAT 1

CAT 2

SOD

CAT 1

CAT 2

CAT 3

CAT 1

CAT 2

LE COL SC10 ALEU PER LF

+

1002


J.G. Scandalios

exhibits a high Km for H2O2, can act upon theH2O2 produced before it diffuses to otherparts of the cell. CAT is a tetrameric heme-containing enzyme that is found in all aero-bic organisms. Because of its wide distribu-tion, evolutionary conservation, and capac-ity to rapidly degrade hydrogen peroxide, ithas been proposed that CAT plays an impor-tant role in systems which have evolved toallow organisms to live in aerobic environ-ments.

CAT is one of the most active catalystsproduced by nature. It decomposes hydro-gen peroxide at an extremely rapid rate,corresponding to a catalytic center activityof about 107 min-1. Depending upon the con-centration of H2O2, it exerts a dual function.At low concentrations (<1 µM) of H2O2, itacts “peroxidatically”, i.e., a variety of hy-drogen donors (e.g., ethanol, ascorbic acid)can be oxidized in the following manner:RH2 + H2O2 → R + 2H2O.

At high concentrations of substrate, CATdecomposes toxic hydrogen peroxide at anextremely rapid rate using the “catalatic”reaction in which H2O2 acts as both an ac-ceptor and donor of hydrogen molecules:2H2O2 → 2H2O + O2.

CAT is unique among H2O2 degradingenzymes in that it degrades H2O2 withoutconsuming cellular reducing equivalents.Hence, CAT provides the cell with a veryenergy efficient mechanism to remove hy-drogen peroxide. Therefore, when cells arestressed for energy and are rapidly generat-ing H2O2 through “emergency” catabolic pro-cesses, H2O2 is degraded by CAT in anenergy-efficient manner. This results in a netgain of reducing equivalents and, therefore,cellular energy. It has been proposed thatCAT may be uniquely suited to regulate thehomeostasis of H2O2 in the cell. In the cata-latic mode, CAT has a very high apparentMichaelis constant and, therefore, is not eas-ily saturated with substrate. Thus, the en-zyme activity increases linearly over a widerange of H2O2 concentrations, thereby main-

taining a controlled intracellular H2O2 con-centration. In mammalian systems, organswith high concentrations of CAT (i.e., liverand kidney) have low levels of endogenousH2O2, and organs with low concentrations ofCAT (i.e., lung and heart) have high endog-enous levels of H2O2. Further, if CAT activ-ity is inhibited, H2O2 concentrations rise inthe liver. As in the case of SOD, multipleCATs (isozymes; Figure 3) encoded by spe-cific genes are found in plants, whereas ani-mals exhibit one form of CAT (16,17). BothCat and Sod genes respond differentially tovarious stresses known to generate ROS (Fig-ure 4).

ROS, telomeres, and aging

Many hypotheses have been proposed toexplain the root cause of aging. One broad-based hypothesis is that generalized homeo-static failure leads to age-related decline.Another is the “free-radical theory of aging”suggesting that endogenous ROS continu-ally damage cellular macromolecules, in-cluding DNA (18). Incomplete repair of suchdamage would lead to its accumulation overtime resulting in age-related deterioration.

The evidence implicating the generationof ROS as key factors in determining lon-gevity is accumulating (19). Much of theearlier evidence was correlative. However,recent evidence has identified longevity-in-fluencing genes responsive to ROS. It hasfurther been shown that continual inductionof DNA damage during normal aging resultsin genomic instability due to persistent DNAlesions, mutations, stalled repair, and trans-cription interference. The involvement ofROS in limiting lifespan has been suggestedby analyses of transgenic Drosophila, whichsystematically overexpress both Sod and Catgenes. Some strains of these transgenic flieslive up to 30% longer than their naturalcounterparts, whereas flies carrying only oneof these constitutively expressed transgenesdo not live longer, suggesting that ROS tox-

1003



icity plays an important role in longevity(20,21).

Much attention has also recently beengiven to the role of telomere shortening as acausative factor in aging/senescence (22,23).Telomeres are stretches of repetitive DNAof tandem short sequence repeats (TTAGGG)that “cap” the ends of eukaryotic chromo-somes to protect against degradation. Te-lomeres in most human cells shorten witheach round of DNA replication, becausethey lack the enzyme telomerase. Telomer-ase is a specialized reverse transcriptase,which helps to replicate the telomere ends ofchromosomes. However, telomerase is nor-mally expressed only in germ-line cells andis derepressed in tumor cells in which telom-eres are stabilized. It has recently been dem-

onstrated that oxidative stress is the mainreason for telomere shortening (24). It hasbeen observed that H2O2 plus Cu(II) induced8-oxo-7,8-dihydro-2'-deoxyguanosine for-mation in the telomere sequences more effi-ciently than in non-telomere sequences. Oxi-dative damage is not repaired as well intelomeric DNA as elsewhere in the chromo-some. Oxidative stress accelerates telomereloss, while antioxidants decelerate it. Thus,oxidative stress is a critical modulator oftelomere loss and telomere-driven replica-tive senescence is primarily a stress response.In a recent study, individuals under general-ized stress were found to have shorter telom-eres and less telomerase activity, and moreoxidative stress (25). Given the availableinformation, aging is likely to be a multifac-

Figure 4. Response to acuteozone exposure of the Sod (top-left) and Cat (top-right) genes inmaize leaves after 6 h of O3 fu-migation. As noted, some of thetranscripts are upregulated (e.g.,Sod3, Sod4, Cat1, Cat3), whileothers are downregulated (e.g.,Sod1, Cat2 ). The 18S rRNA is aloading control. The productionof H2O2 in maize leaves (bot-tom-left) in response to wound-ing. Seven-day-old plants wereexcised at the base of the stemsand supplied with diaminobenzi-dine (DAB) for 6 h. The plantsthen were wounded and continu-ously supplied with DAB for 4 h.The production of H2O2 can bevisualized by the deposition ofthe brown-red (dark areas) colorproducts in the leaves. A, Con-trol and wounded leaves(wounding was conducted nearthe main vein). B, Control andwounded leaves (wounding wasconducted at the main vein andcut at the leaf edges). Cat tran-script accumulation in responseto wounding (bottom-right). RNAwas isolated from similarlytreated leaves, indicative thatdifferential transcript accumula-tion is effected by induction ofH2O2.

Sod1

Sod3

Sod4

Sod4A

18S

Cat1

Cat2

Cat3

Gst1

18S

0 100 200 300 500 1000Ozone (ppb) Ozone (ppb)

0 100 200 300 500 1000

Wounded PunctureControl Control Cut

A

B

Cat1

Cat2

Cat3

SOD CAT

1004


J.G. Scandalios

torial process; whether ROS are peripheraltargets that correlate with longevity or cen-tral regulators of human aging remains to beresolved.

Oxidative stress

ROS such as O2•−, H2O2, and OH• are

produced in all aerobic organisms and nor-mally exist in the cell in balance with anti-

oxidant molecules. Oxidative stress occurswhen this critical balance is disrupted due todepletion of antioxidants or excess accumu-lation of ROS, or both (Figure 5). That is,when antioxidants are depleted and/or if theformation of ROS increases beyond the abil-ity of the defenses to cope, then oxidativestress and its detrimental consequences en-sue. Such stress occurs when severely ad-verse environments or physiologic condi-tions overwhelm biological systems. Onerapid and clear indicator of oxidative stressis the induction of antioxidant defenses and/or increases in endogenous ROS levels. Theformation of ROS can be accelerated as aconsequence of various environmental stressconditions, including UV-radiation, highlight intensities, exposure to herbicides, ex-treme temperatures, toxins such as cerco-sporin and aflatoxin, air pollutants, metals,wounding, and xenobiotics. Many inducersof oxidative stress are known carcinogens,mutagens, and toxins. ROS production andaccumulation is a common denominator inmany diseases and environmental insults andcan lead to severe cellular damage leading tophysiological dysfunction and cell death invirtually all aerobes (Figure 6).

When oxidative stress occurs, cells func-tion to counteract the oxidant effects and torestore redox balance by resetting criticalhomeostatic parameters. Such cellular activ-ity leads to activation or silencing of genesencoding defensive enzymes, transcriptionfactors, and structural proteins (26,27).

ROS perform essential cellularfunctions

It has recently become apparent that ROSare not always harmful metabolic byprod-ucts as generally believed (27). When tightlyregulated, ROS perform critical functions inthe cell. In fact, a significant body of evi-dence indicates that ROS, particularly O2

•−

and H2O2, act as intracellular signaling mol-ecules. In bacteria, the transcription factor

Figure 5. Oxidative stress resultsfrom imbalance between the lev-els of antioxidants (AOX) and re-active oxygen species (ROS).Cells are normally able to bal-ance the production of oxidantsand antioxidants to maintain re-dox equilibrium. Oxidative stressoccurs when this equilibrium isupset by excess levels of ROS,or depletion of antioxidant de-fenses.

Figure 6. Scheme showing some of the initiators (stressors) of reactive oxygen species(ROS) and the biological consequences leading to a variety of physiological dysfunctionsthat can lead to cell death.

Equilibrium(AOX = ROS)

Oxidative stress(excess ROS)

Oxidative stress(depleted AOX)

Antioxidants Oxidants

#

#

#

AOX

AOX

AOX

ROS

ROS

ROS

Aging/senescenceWounding

XenobioticsRadiation/light

Heat and ColdPathogensBiotoxinsDrought

Heavy metalsAir pollutants

(O3; SO2 )

Hormones

ROS

Lipids and fatty acidsAmino acids

ProteinsNucleic acids

Pigments

Cellular effects

Molecular damage

Cell

DEATH

Stressors

H2O2

O2•−

•OH

OxidativeSTRESS

Membrane damageLoss of organelle functions

Reduction in metabolic efficiencyReduced carbon fixation

Electrolyte leakageChromatid breaks

Mutations

1005



OxyR activates a number of genes inducibleby H2O2, while the transcription factors SoxR/SoxS mediate responses to O2

•− (28,29). Yeastalso has two distinct adaptive stress re-sponses, one directed towards H2O2 and onetowards O2

•− (30-32). In higher eukaryotes,both animals and plants, oxidative stressresponses are more complex and are modu-lated by several regulators (9). ROS-depend-ent redox cycling of cysteinyl thiols is criti-cal for establishing protein-protein and pro-tein-DNA interactions that determine manyaspects of signal transduction pathways byregulating the activity of many transcriptionfactors (26,32). For example, activation ofnuclear factor κB (NF-κB) and activatorprotein-1 (AP-1), known to have critical rolesin proliferation, differentiation and morpho-genesis, can result from stimulation by di-verse agents proceeding through a commonpathway involving ROS generation. Thepathway leading to H2O2 production and sub-sequent redox activation of NF-κB has beenshown to involve the Rho family of smallGTP-binding proteins (33).

Both intracellular and extracellularsources of ROS are capable of modulatinggene expression. Low doses of H2O2 (<20µM) can elicit changes in phosphorylation ofspecific regulatory proteins including pro-tein kinase B. Direct signaling action of

H2O2 in the differential regulation of anti-oxidant genes in plants likely occurs viaprotein-DNA interactions in the region ofthe antioxidant-responsive element (ARE;TGACTCA), NF-κB, and abscisic acid-re-sponsive element 2 (ACGT) in the promot-ers of these genes. In addition to induction ofdefense gene expression, other roles of ROSin plants include direct killing of pathogens,involvement in cell wall structure, and pro-motion of programmed cell death (9). Inyeast and animal cells, ROS have been shownto arrest cell division, and cell cycle progres-sion is under negative ROS control (34). Aclear and powerful example of how ROS areput to constructive uses was the observationthat O2

•− plays an important role againstinvading microbes, in effect serving as abroad-spectrum antibiotic (35). In responseto invasion by pathogens, plants also mounta broad range of defense responses, includ-ing a rapid and transient production of largeamounts of ROS (“oxidative burst”) (36).

Thus, ROS play different, or even oppos-ing roles, during different cellular processes(Figure 7). For example, under physiologicconditions, H2O2 may play an important rolein signal transduction pathways and in acti-vation of the transcription factor NF-κB,while under pathologic (stress) conditions,H2O2 can lead to apoptosis or necrosis. Con-

Figure 7. Scheme showingsome of the useful pleiotropicroles of reactive oxygen species(ROS) known to occur in mosthigher organisms, indicative ofthe fact that ROS are not alwaysharmful to cells. MAPK = mito-gen-activated protein kinase.

1006


J.G. Scandalios

sequently, the steady-state level of ROSwithin cells is critical and is determined bythe interplay between ROS-generatingmechanisms and the subtle modulating rolesplayed by cellular antioxidants. In plants,H2O2 is generated under a diverse range ofbiotic and abiotic conditions, and its accu-mulation in specific tissues at specific devel-opmental times, in the appropriate quanti-ties, benefits plants and can mediate cross-tolerance to other stresses (37). H2O2 affectsgene expression and activates MAP kinases(MAPK), which in turn function as regula-tors of transcription (3).

ROS and gene expression

Numerous studies indicate that cells havethe means to sense ROS and to induce spe-cific responses, but the underlying mechan-isms are not fully understood (9,26). Thetranscriptional network that responds to ROSin eukaryotes is currently being deciphered,whereas the prokaryotic system is better un-derstood.

Nearly three decades ago, it was shownthat the expression of ~30 proteins was in-duced by H2O2 in bacteria (38). Of these 30proteins, 12 were maximally induced within10 min and 18 between 10-30 min. TheOxyR regulatory protein was subsequentlyshown to regulate expression of 9 of the 12rapidly induced proteins. The tetramericOxyR protein is a member of the LysR fam-ily of transcription activators and exists intwo forms, reduced and oxidized; only theoxidized form is able to activate transcrip-tion. Further studies led to the identificationof a number of OxyR-activated genes (39).Similarly, the SoxRS regulatory proteinswere found to regulate expression of O2

•−-responsive proteins in bacteria (40). Regula-tion of the SoxRS regulon occurs by conver-sion of SoxR to an active form that enhancessoxS transcription. The enhanced levels ofSoxS in turn activate expression of theregulon (40). In addition to SoxR and OxyR,

several other transcriptional regulators modu-late the expression of antioxidant genes inbacteria, indicative of the complexity andconnectivity of overlapping regulatory net-works. No apparent homologs of OxyR,SoxR, or SoxS have been found in eukary-otes (28), but a number of other transcriptionfactors have been found to play a role inregulating the expression of antioxidant genesin eukaryotes. In yeast, transcription regula-tors of antioxidant genes include ACE1,MAC1, YAP1, YAP2, HAP1, and HAP2/3/4 (30). In higher eukaryotes, oxidative stressresponses are more complex and are modu-lated by several different regulators. In mam-malian systems, NF-κB and AP-1 are in-volved in regulating the oxidative stress re-sponse. The ARE, present in the promoterregion of mammalian glutathione S-trans-ferase, metallothionein-I, and MnSod genes,causes induction of these genes in responseto oxidants (41). NF-κB, AP-1, and AREhave also been found in promoters of anti-oxidant genes in higher plants (Figure 8) (9).The role of these factors is not unique toactivation of antioxidant genes, as they areknown, particularly NF-κB, to play centralroles in regulating cellular responses to otherstresses as well as regulating normal growthand metabolism.

There is substantial evidence that a vari-ety of biotic and abiotic stresses induce ROS,which serve as a common factor in regulat-ing various signaling pathways (Figure 9)(9,42). Similar stresses also activate MAPKswith kinetics that either precede or parallelH2O2 production, indicating that MAPKsmay be one of several converging points inthe defense-signaling network (43). In addi-tion, exogenous application of several planthormones and toxins has been shown toinduce O2

•− and H2O2 synthesis, leading todifferential induction of some antioxidantgenes and isozymes (Figure 10) (9,37,41,44).Thus, the identification of all genes andproteins regulated by H2O2 is an importantstep toward treatments that might confer

1007



Figure 8. Schematic representation of the promoter region of each of the three catalase (Cat ) genes of maize indicating the locations of the NF-κB,ARE, AP-1, and ABRE motifs relative to the transcription start site (+1) of each gene. NF-κB = nuclear factor κB; ARE = antioxidant-responsiveelement: ABRE = abscisic acid-responsive element; AP-1 = activator protein-1.

Figure 9. Convergence of vari-ous biotic and abiotic stressstimuli onto mitogen-activatedprotein kinase (MAPK) path-ways via reactive oxygen spe-cies (ROS) as a common factor,leading to activation of antioxi-dant defense genes. Alternativepathways have also been impli-cated. MAPKKK = MAPK kinasekinases.

1008


J.G. Scandalios

tolerance to multiple, but interrelated,stresses. In addition to induction/repressionof antioxidant defense genes, ROS are knownto similarly affect expression of a variety ofother genes involved in different signalingpathways in microbes (28), yeast (45), plants(46), and animals (19).

Genomic scale ROS-responsive geneexpression

The advent of microarray expression anal-ysis makes possible the assessment of geneexpression on a genomic scale, renderingtens of thousands of genes assayable in asingle experiment (47). Thus, identificationof ROS-responsive genes on a global scale isnow tenable. DNA microarrays can compre-hensively examine gene expression networksduring oxidative stress. There is now signifi-cant progress being made in surveying geneexpression in response to H2O2 in Esche-richia coli (48), yeast (4,45), animals (19),and higher plants (46).

A genome-wide transcription profile ofE. coli cells exposed to H2O2 was examinedwith a DNA microarray composed of 4169E. coli open reading frames (48). Gene ex-

pression was measured in isogenic wild-type and oxyR deletion mutants (∆oxyR) toconfirm that the H2O2-response regulatorOxyR activates most of the H2O2-induciblegenes. A very rapid and strong inductionwas observed of a set of OxyR-regulatedgenes in the wild type but not in the ∆oxyR,providing internal validation of the experi-ment and confirmation for the induction ofthe oxidative stress genes identified earlierby other means (38). Several new H2O2-inducible genes were also identified: somewere members of the OxyR regulon andsome induced by an OxyR-independentmechanism suggestive of other H2O2 sen-sors and regulators in E. coli (38). Severalgenes repressed by OxyR were highly ex-pressed in the ∆oxyR mutant. Overall, themRNA of 140 genes in the wild type, and167 genes in the ∆oxyR were significantlyinduced after H2O2 treatment. It was alsofound that soxS was induced by H2O2, indi-cating an overlap with other regulatory path-ways. Two genes, Fpr and sodA, known tobe members of the SoxRS regulon were alsohighly induced by H2O2 in both wild typeand ∆oxyR. The microarray data also showedan overlap between oxidative stress, heat

Figure 10. Schematic represen-tation of the possible signaltransduction pathways of Cat1gene expression in response tothe plant hormone abscisic acid(ABA) and osmotic stress inmaize. ARE = antioxidant-re-sponsive element; ABRE-2 =ABA responsive element 2;CBF1 = Cat1 binding factor-1;CBF2 = Cat1 binding factor-2.

1009



shock and SOS responses (38). The resultsfrom the E. coli microarrays clearly indicatethat the activities of transcription factors inaddition to OxyR and SoxRS are likely modu-lated by oxidative stress.

Whole-genome expression patterns inSaccharomyces cerevisiae cells exposed toH2O2, in addition to other stresses, indicatedthat ~2/3 of the genome is involved in theresponse to environmental changes, and theglobal set of genes induced/repressed byeach environmental signal were identified(4,45). The response to oxidative stress in-volves ~1/3 of the yeast genome and themaximal effects on gene expression occurslightly later relative to other stresses exam-ined during similar time-courses, with mostof the transcriptome returning to prestresslevels within 2 h following exposure to H2O2

(4). Genes that are repressed for ~60 minafter exposure to H2O2 are only transientlyrepressed in other stress time courses. Thus,genes encoding the translation apparatus andits regulators are remarkably coordinated foreach environmental change, although thedynamics of each response are different.The expression programs following H2O2 orO2

•− treatment were essentially identicaldespite the fact that different ROS are in-volved. There was strong induction of genesknown to be involved in detoxification ofboth H2O2 and O2

•−, such as CAT, SOD, andglutathione peroxidase, as well as genes in-volved in oxidative and reductive reactions(e.g., thioredoxin, glutathione reductase,glutaredoxin). The genes most strongly in-duced in response to H2O2 and O2

•− weredependent on the transcription factor Yap1pfor their induction. Genes moderately in-duced by ROS and other signals are regu-lated by different transcription factors, de-pending on the conditions, and different up-stream signaling pathways may govern theirresponse. It has also been demonstrated thatin Schizosaccharomyces pombe H2O2 acti-vates the Sty1 (stress-activated MAPK) path-way in a dose-dependent manner via two

sensing mechanisms (49). At low H2O2 lev-els, a two-component signaling pathway,which feeds into either of the two (Wak1 orWin1) stress-activated MAPK kinase ki-nases, regulates Sty1. At high H2O2 levels,however, Sty1 activation is controlled mainlyby an independent two-component mechan-ism, which requires the function of bothWak1 and Win1. In addition, the individualbZip transcription factors, Pap1 and Atf1,were found to function within a limited rangeof [H2O2]: Pap1 activates target genes at low[H2O2], whereas Atf1 controls transcriptionalresponses to high [H2O2], with some minoroverlap. Some apparent cross talk amongSty1, Atf1, and Pap1 has been detected (50).Thus, S. pombe deploys a combination ofstress-responsive regulatory proteins togauge and trigger the appropriate transcrip-tional response to increasing H2O2 concen-trations (49). This organism mounts two sepa-rate responses to oxidative stress: an adap-tive response to low-level H2O2 exposurethat protects it from subsequent exposures tohigher [H2O2], and an acute response thatallows the cell to survive a sudden, poten-tially lethal dose of H2O2.

Large-scale cDNA microarray analysisof the Arabidopsis transcriptome during oxi-dative stress identified 175 non-redundantexpressed sequence tags from a sample of11,000, which are regulated by H2O2. Ofthese, 62 are repressed and 113 are induced.In addition, RNA blots showed that some ofthe H2O2-regulated genes are also modu-lated by other signals known to involve oxi-dative stress (51). Furthermore, a substantialportion of these genes have predicted func-tions in defense response, cell rescue andsignaling, and transcription, underscoringthe pleiotropic effects of H2O2 in the re-sponse of plants to stress. Overall, the mi-croarray used was estimated to representonly ~30% of the Arabidopsis genome, de-pending on redundancy, and 1 to 2% of thegenes represented in the array are affectedby H2O2-imposed oxidative stress, being

1010


J.G. Scandalios

comparable to the situation in yeast (4). Ofthe 175 genes identified as H2O2-responsive,most have no obvious direct role in oxida-tive stress but may be linked to oxidativestress indirectly, as a consequence of otherbiotic and abiotic stresses, explaining theirsensitivity to H2O2. Among the genes in-duced by H2O2 were genes encoding trans-cription factors, suggesting that they maymediate downstream H2O2 responses. As inother organisms, expression of the MAPKsin Arabidopsis is induced by oxidative stress,which in turn can mediate the induction ofoxidative stress-responsive genes (52).

Toward an integrated view ofoxy-stress responses

The traditional view of ROS as mereindiscriminate reactive byproducts of cellu-lar metabolism has recently undergone a

metamorphosis. This view came about bythe discovery that ROS may act as signal-transducing molecules and that activation ofintracellular transcription factors such asOxyR, SoxRS, NF-κB, and AP-1 occur viainteraction with ROS, leading to gene trans-cription (Figure 11).

Genome sequencing and expression pro-filing using DNA or oligonucleotide mi-croarrays, and related technologies, havebeen used effectively in the study of globalgene expression patterns in response to dif-ferent growth and environmental conditionsto which organisms are exposed. Subsequenthierarchical clustering methods allow forthe allocation of genes, coregulated tempo-rally or in response to a given signal, intospecific expression groups or regulons (48,53). The numbers of genes that can be de-tected by these methods in response to anygiven environmental or developmental sig-

Figure 11. The major signalingpathways activated in responseto oxidative stress. Reactiveoxygen species (ROS) originat-ing from environmental signalsor from metabolic activity aremodulated by antioxidants tonontoxic levels, at which pointthey serve as signaling mol-ecules. ROS can activate genetranscription in two ways: a) viatranscription factors, such asNF-κB, AP-1, and ARE-bindingproteins (ARE-BP) that can in-teract directly with specific DNAmotifs on promoters of targetgenes, or b) via activation ofMAPK cascades, which in turnactivate transcription factors thattrigger target gene transcription.The degree to which a givenpathway is activated depends onthe nature and duration of thestress, as well as on cell typeand developmental stage. NF-κB = nuclear factor κB; AP-1 =activator protein-1; ARE-BP =antioxidant-responsive elementbinding proteins; HSF1 = heatshock transcription factor 1.

1011



nal far exceed the limited number that couldhave been detected only a few years ago.With the current methods the expression oftens of thousands of genes can be detected ina single experiment in response to ROS, orto any given signal. Since transcription ofgenes into mRNA is governed by transcrip-tion factors which bind to cis-regulatory re-gions of the DNA in the vicinity of the targetgene, the question arises as to whether largeco-regulated groups of genes share cis-regu-latory elements that bind to common trans-cription factors. The data available with re-spect to oxidative stress seem to suggest this.Cis-acting elements within the promoters ofROS-activated genes are being defined aswell as their cognate trans-acting factors. Acomparative analysis of promoter sequencesof genes with similar expression profilesshould provide a basis for unraveling com-mon regulatory sequences and overlappinggene expression networks modulating ROS-responsive genes. The antioxidant enzymesCAT and SOD play key roles in modulatingthe levels of endogenous H2O2 and O2

•−,which in turn, at specific concentrations, actto modulate the expression of other ROS-responsive genes.

The use of microarrays and future deri-vations thereof, to examine global gene re-sponses to ROS is assured. It has becomeclear that there are far more genes and geneclusters responding to ROS than previouslythought, and that ROS likely play far morekey roles in cellular activities than antici-pated. As global gene responses to ROS areexamined temporally and spatially along withdifferent time-courses, and varied oxidantconcentrations, even more complex regulonsand regulatory cascades will emerge. It willalso be instructive to examine gene expres-sion profiles in specific mutants to help de-lineate the roles of specific regulators and ofROS. The identification of alternate path-ways of ROS-dependent gene expressionand characterization of the redox sensingmechanisms involved should point to new

insights and directions. Some genes havebeen identified whose transcription is re-sponsive to a variety of stresses in additionto oxystress, while others appear to be re-sponsive only to ROS or other specific sig-nals. Although some regulatory systems havebeen implicated in modulating such re-sponses, the complete network of regulatorsof ROS responses that activate such genesremains unclear.

Microarray results from different organ-isms clearly underscore the fluidity of ge-nomes to reorganize and respond to changesin the cellular and extracellular environment.However, characterization of global geneexpression programs at the transcript level isonly the first step toward defining the role orfunction of each ROS-responsive gene. It isnot unlikely that small changes in gene ex-pression could lead to large alterations inprotein levels. Consequently, proteomicanalyses are essential to correlate mRNAchanges with protein levels. There are in factample data demonstrating that ROS can alterthe activity of cellular proteins.

A large and growing number of sequencedgenomes and emerging technologies presentus with enormous opportunities to advancebiological science, and our knowledge ofhow genomes perceive signals to respond tovariable environments. Oxidative stress andthe responses to it seem to be central to manyof the key biological questions. However, asthe euphoria of new sequencing and techno-logical advances begins to fade, we mustrecognize the magnitude of the problemsthat still need to be solved. Knowing thesequences of tens of thousands of ROS-responsive genes only reminds us that westill do not know the many proteins theyencode, nor the biochemical or biologicalfunction of the great majority of such pro-teins. How such proteins interact with ROSto drive the various physiological processesin aerobic organisms remains a great puzzle.For the future, the fundamental challengewill be to integrate the information now

1012


J.G. Scandalios

being obtained on gene expression patternswith structural, topological, and functionalparameters and interactions of the variousproteins encoded by ROS-responsive regu-lons, and to view the cell in which theyfunction holistically.

The paradox

The oxygen paradox is indeed the para-dox of evolution itself. Evolutionary pres-sures have made the best of a bad situationby generating mechanisms to curtail the un-desirable toxic effects of ROS, which are anunavoidable consequence of the aerobiclifestyle, and to put them to constructiveuses. Indeed evolution has co-opted ROS toserve necessary and useful purposes in themaintenance of cellular homeostasis and inthe communication of cells with the externalenvironment. The focus must now be placedon a more thorough understanding of howROS-mediated signals are perceived, trans-duced, and interpreted by the cell’s geneticmachinery. Perhaps the most noteworthyobservation to date concerning oxidativestress and the negative and positive roles ofROS is their universality among aerobic or-ganisms and the similarities emerging in theregulatory mechanisms underlying their rolesin all species.

The classical view of ROS as villains thatindiscriminately destroy biomolecules hasundergone a shift, in which positive biologi-cal roles are considered as well. It is nowaccepted that ROS, particularly H2O2 andO2

•−, are carefully regulated metabolites ca-pable of signaling and communicating criti-cal information to the cell’s genetic machin-ery. Redox regulation of gene expression byoxidants and antioxidants is emerging as avital mechanism in the health of all eukary-otes, including man.

It would indeed be interesting and chal-lenging to identify all the changes in geneexpression regulated by oxidative stress, andto determine their commonality among di-

verse species. Such a global analysis of theeffects of H2O2 and O2

•− on the transcriptomeof eukaryotes has not yet been attained, butwith the emergence of post-genomic tech-nologies it will likely be not far off.

Conclusions

The survival of organisms on earth de-pends upon the interactions of their genomeswith the environments in which they exist.In the course of evolution, organisms evolveda complex array of mechanisms for adaptingto both minor and major fluctuations in theenvironment. The emergence of oxygenicphotosynthesis presented early life formswith the greatest environmental challengeand an opportunity. The challenge was todevelop antioxidant defenses in order to sur-vive; the opportunity was to exploit the reac-tivity of oxygen for energy yielding andbiosynthetic reactions. The opportunity ledto the highly diversified life forms thatevolved sufficient defenses and managed toexploit the aerobic lifestyle. Oxygen toxic-ity likely led to massive extinctions of thoseorganisms unable to cope with it, unless theytook refuge in isolated anaerobic niches.Thus, oxygen is a “double-edged sword” inthat it makes life on earth possible, but in itsreduced forms (ROS), it is highly toxic andlethal. Oxidative stress can arise from animbalance between generation and elimina-tion of ROS leading to excess ROS levelsinflicting indiscriminate damage to virtuallyall biomolecules, leading in turn to variousdiseases and cell death. The notion that ROSare merely toxic byproducts of O2 metabo-lism has recently been altered by experimen-tal evidence indicating that ROS are care-fully regulated metabolites capable of sig-naling and communicating information tothe cell’s genetic machinery. Redox regula-tion of gene expression by oxidants andantioxidants is emerging as a vital mechan-ism in the health and disease of all organ-isms. Irrespective of how or when ROS are

1013



generated, an increase in intracellular oxi-dants results in two critical effects: damageto various cell components and activation ofspecific signaling pathways, influencing vari-ous cellular processes leading to proper cellfunctions or to cell death. Genomic tools areaccelerating the discovery of ROS-respon-sive genes on a global scale and are expand-ing our understanding of the oxidative stressresponse and the pleiotropic roles of ROS insignaling and gene expression.

Acknowledgments

I gratefully acknowledge the generous,and continuous, support of my research overmany years by the U.S. National Institutes ofHealth, Department of Energy, Environmen-tal Protection Agency, National ScienceFoundation, and Department of Agriculture.

References

1. Scandalios JG (2002). The rise of ROS. Trends in BiochemicalSciences, 27: 483-486.

2. McClintock B (1984). The significance of responses of the genometo challenge. Science, 226: 792-801.

3. Scandalios JG (2002). Oxidative stress responses: what have ge-nome-scale studies taught us? Genome Biology, 3: 1019.1-1019.6.

4. Gasch A, Spellman P, Kao C, Harel O, Eisen M, Storz G, Botstein D& Brown P (2000). Genomic expression programs in the response ofyeast cells to environmental changes. Molecular Biology of the Cell,11: 4241-4257.

5. Scandalios JG & Guan LM (2001). Transcription factors regulatingantioxidant gene expression in response to biotic and abiotic sig-nals. In: Gene Families: Studies of DNA, RNA, Enzymes and Pro-teins. World Scientific, Singapore, 287-301.

6. Laloi C, Apel K & Danon A (2004). Reactive oxygen signaling: thelatest news. Current Opinion in Plant Biology, 7: 323-328.

7. Thorpe GW, Fong CS, Alic N, Higgins VJ & Dawes IW (2004). Cellshave distinct mechanisms to maintain protection against differentreactive oxygen species: Oxidative stress-response genes. Pro-ceedings of the National Academy of Sciences, USA, 101: 6564-6569.

8. Hancock JT, Desikan R & Neill SJ (2001). Role of reactive oxygenspecies in cell signaling pathways. Biochemical Society Transac-tions, 29: 345-350.

9. Scandalios JG (1997). Oxidative Stress and the Molecular Biologyof Antioxidant Defenses. Cold Spring Harbor Laboratory Press,Plainview, NY, USA.

10. Elstner EF (1987). Metabolism of activated oxygen species. In:Davies DD (Editor), The Biochemistry of Plants. Vol. 11. AcademicPress, New York, 253-315.

11. Dismukes GC, Klimov VV, Baranov SV, Kozlov YN, DasGupta J &Tyryshkin A (2001). The origin of atmospheric oxygen on Earth: Theinnovation of oxygenic photosynthesis. Proceedings of the NationalAcademy Sciences, USA, 98: 2170-2175.

12. Fridovich I (1995). Superoxide radical and superoxide dismutases.Annual Review of Biochemistry, 64: 87-112.

13. Halliwell B (1996). Free radicals in biochemistry and medicine. In:Meyers RA (Editor), Encyclopedia of Molecular Biology and Molecu-lar Medicine. Vol. 2. Wiley-VCH, Weinheim, Germany, 330-337.

14. Scandalios JG (1997). Molecular genetics of superoxide dismutases.

In: Scandalios JG (Editor), Oxidative Stress and the Molecular Biol-ogy of Antioxidant Defenses. Cold Spring Harbor Laboratory Press,Plainview, NY, USA, 527-568.

15. Fink RC & Scandalios JG (2002). Molecular evolution and structurefunction relationships of the superoxide dismutase gene families inangiosperms, and their relationship to other eukaryotic and prokary-otic superoxide dismutases. Archives of Biochemistry and Biophys-ics, 399: 19-35.

16. Scandalios JG, Guan L & Polidoros AN (1997). Catalases in plants:Gene structure, properties, regulation, and expression. In: Scanda-lios JG (Editor), Oxidative Stress and the Molecular Biology ofAntioxidant Defenses. Cold Spring Harbor Laboratory Press,Plainview, NY, USA, 343-406.

17. Guan L & Scandalios JG (1996). Molecular evolution of maizecatalases and their relationship to other eukaryotic and prokaryoticcatalases. Journal of Molecular Evolution, 42: 570-579.

18. Beckman KB & Ames BN (1998). The free radical theory of agingmatures. Physiological Reviews, 78: 547-581.

19. Finkel T & Holbrook NJ (2000). Oxidants, oxidative stress and thebiology of aging. Nature, 408: 239-247.

20. Beckman KB & Ames BN (1997). Oxidants, antioxidants, and aging.In: Scandalios JG (Editor), Oxidative Stress and the Molecular Biol-ogy of Antioxidant Defenses. Cold Spring Harbor Laboratory Press,Plainview, NY, USA, 201-246.

21. Helfand SL & Rogina B (2003). Molecular genetics of aging in the fly:is this the end of the beginning? BioEssays, 25: 134-141.

22. Chan SRWL & Blackburn EH (2004). Telomeres and telomerases.Philosophical Transactions of the Royal Society of London, SeriesB. Biological Sciences, 359: 109-121.

23. McNight TD & Shippen DE (2004). Plant telomere biology. PlantCell, 16: 794-803.

24. Von Zglincki T (2002). Oxidative stress shortens telomeres. Trendsin Biochemical Sciences, 27: 339-344.

25. Sapolsky RM (2004). Organismal stress and telomeric aging: Anunexpected connection. Proceedings of the National Academy ofSciences, USA, 101: 17323-17324.

26. Dalton TP, Shertzer HG & Puga A (1999). Regulation of geneexpression by reactive oxygen. Annual Review of Pharmacologyand Toxicology, 39: 67-101.

27. Scandalios JG (2004). Genomic responses to oxidative stress. In:

1014


J.G. Scandalios

Meyers RA (Editor), Encyclopedia of Molecular Cell Biology andMolecular Medicine. Vol. 5. 2nd edn. Wiley-VCH, Weinheim, Ger-many, 489-512.

28. Storz G & Imlay JA (1999). Oxidative stress. Current Opinion inMicrobiology, 2: 188-194.

29. Kiley PJ & Storz G (2004). Exploiting thiol modifications. PLoSBiology, 2: 1714-1717.

30. Ruis H & Schuller C (1995). Stress signaling in yeast. BioEssays,17: 959-965.

31. Moradas-Ferreira P & Costa V (2000). Adaptive response of theyeast Saccharomyces cerevisiae to reactive oxygen species: de-fenses, damage and death. Redox Report, 5: 277-285.

32. Delaunay A, Isnard A & Toledano MB (2000). H2O2 sensing throughoxidation of the Yap1 transcription factor. EMBO Journal, 19: 5157-5166.

33. Gabbita SP, Robinson K, Stewart C, Floyd R & Hensley K (2000).Redox regulatory mechanisms of cellular signal transduction. Ar-chives of Biochemistry and Biophysics, 376: 1-13.

34. Paulovich AG, Toczysky DP & Hartwell LH (1997). When check-points fail. Cell, 88: 315-321.

35. Babior BM (1984). The respiratory burst of phagocytes. Journal ofClinical Investigation, 73: 599-601.

36. Doke N (1997). The oxidative burst: roles in signal transduction andplant stress. In: Scandalios JG (Editor), Oxidative Stress and theMolecular Biology of Antioxidant Defenses. Cold Spring HarborLaboratory Press, Plainview, NY, USA, 785-813.

37. Guan LM, Zhao J & Scandalios JG (2000). Cis-elements and trans-acting factors that regulate expression of the maize Cat1 antioxidantgene in response to ABA and osmotic stress: H2O2 is the likelyintermediary signaling molecule for the response. Plant Journal, 22:87-95.

38. Christman MF, Morgan RW, Jacobson FS & Ames BN (1985).Positive control of a regulon for defenses against oxidative stressand some heat-shock proteins in Salmonella typhimurium. Cell, 41:753-762.

39. Storz G & Zheng M (2000). Oxidative stress. In: Storz G & Henge-Aronis R (Editors), Bacterial Stress Responses. ASM Press, Wash-ington, DC, USA, 47-59.

40. Wu J & Weiss B (1991). Two divergently transcribed genes, soxRand SoxS, control a superoxide response regulon of Escherichiacoli. Journal of Bacteriology, 173: 2864-2871.

41. Scandalios JG (2001). Molecular responses to oxidative stress. In:Hawkesford MJ & Buchner P (Editors), Molecular Analysis of Plant

Adaptation to the Environment. Kluwer Academic Publishers, Dor-drecht, The Netherlands, 181-208.

42. Somssich IE & Halbrock K (1998). Pathogen defence in plants - aparadigm of biological complexity. Trends in Plant Sciences, 3: 86-90.

43. Zhang S & Klessig D (2001). MAPK cascades in plant defensesignaling. Trends in Plant Sciences, 6: 520-527.

44. Williamson J & Scandalios JG (1992). Differential response of maizecatalases and superoxide dismutases to the photoactivated fungaltoxin cercosporin. Plant Journal, 2: 351-358.

45. Causton H, Bing R, Koh S, Harbison C, Kanin E, Jennings E, Lee T,True H, Lander E & Young R (2001). Remodeling of yeast genomeexpression in response to environmental changes. Molecular Biol-ogy of the Cell, 12: 323-337.

46. Desikan R, Neill SJ & Hancock JT (2000). Hydrogen peroxide-induced gene expression in Arabidopsis thaliana. Free Radical Biol-ogy and Medicine, 28: 773-778.

47. Schena M, Shalon D, Davis R & Brown P (1995). Quantitativemonitoring of gene expression patterns with complementary DNAmicroarray. Science, 270: 467-470.

48. Zheng M, Wang X, Templeton L, Smulski D, LaRossa R & Storz G(2001). DNA microarray-mediated transcriptional profiling of theEscherichia coli response to hydrogen peroxide. Journal of Bacteri-ology, 183: 4562-4570.

49. Quin J, Findlay VJ, Dawson K, Millar JB, Jones N, Morgan BA &Toone WM (2002). Distinct regulatory proteins control the gradedtranscriptional response to increasing H2O2 levels in fission yeastSchizosaccharomyces pombe. Molecular Biology of the Cell, 13:805-816.

50. Nguyen AN, Lee A, Place W & Shiozaki K (2000). Multistep phos-phorelay proteins transmit oxidative stress signals to the fissionyeast stress-activated protein kinase. Molecular Biology of the Cell,14: 1169-1181.

51. Desikan R, MacKerness A, Hancock JT & Neill SJ (2001). Regula-tion of the Arabidopsis transcriptome by oxidative stress. PlantPhysiology, 127: 159-172.

52. Kovtun Y, Chiu WL, Tena G & Sheen J (2000). Functional analysisof oxidative stress-activated mitogen-activated protein kinases cas-cade in plants. Proceedings of the National Academy of Sciences,USA, 97: 2940-2945.

53. Frances M & Rao S (2002). Of chips and ChIPs. Science, 296: 666-669.

oxidative stress: molecular perception and transduction of ... · transduces environmental signals...

Documents