Long-distance signalling in plantdefenceMartin Heil1 and Jurriaan Ton2

1 Dpto. de Ingenierıa Genetica, CINVESTAV - Irapuato, Km. 9.6 Libramiento Norte, CP 36821, Irapuato, Guanajuato, Mexico2 Department of Biological Chemistry, Rothamsted Research, Harpenden, Hertfordshire, AL5 2JQ, UK

Review

Plants use inducible defence mechanisms to fend offharmful organisms. Resistance that is induced inresponse to local attack is often expressed systemically,that is, in organs that are not yet damaged. In the searchfor translocated defence signals, biochemical studiesfollow the physical movement of putative signals, andgrafting experiments use mutants that are impaired inthe production or perception of these signals. Long-distance signals can directly activate defence or canprime for the stronger and faster induction of defence.Historically, research has focused on the vascular trans-port of signalling metabolites, but volatiles can play acrucial role as well. We compare the advantages andconstraints of vascular and airborne signals for the plant,and discuss how they can act in synergy to achieveoptimised resistance in distal plant parts.

Dynamic and systemic plant defencesPlants possess chemical and physical defences to reducedamage by primary consumers: animals and microbes thatuse them as a food source. Because defence is costly, plantshave evolved inducible defence mechanisms that are acti-vated or amplified in response to attack. Particular signalscan prime plant tissues, that is, prepare them for anaugmented response to future attack without directly acti-vating costly defence mechanisms. Herbivorous animalsare highly mobile, however, and even pathogenic microbescan move within the plant. Hence, plants requireadditional countermeasures to halt the dispersal of theattacker and so prevent further damage. Systemicallyexpressed induced resistance is a response that specificallyaims to restrict the spread of the attacker: upon localattack, enhanced resistance spreads to distal organs thatare not yet affected.

Systemically induced resistance can be achieved by twomechanisms. First, it can result from systemic transport ofdefensive metabolites; and second, it can result from denovo expression of resistance mechanisms that are acti-vated by translocated signals from the stress-exposed tis-sues. Recent evidence supports a crucial role for bothmechanisms. This review focuses on the search for theselong-distance signals. We summarize phenotypic evidencefor systemic defence induction against pathogens and her-bivores in a historical context, discuss the traits thatqualify a factor as a long-distance signal, and presentcandidates that have been shown to act in this context.We compare airborne transport of signals with the more

Corresponding author: Heil, M. (mheil@ira.cinvestav.mx), (heil_martin@web.de).

264 1360-1385/$ – see front matter � 2008 Elsevie

traditionally studied translocation of signals through thevascular system, and outline the advantages and con-straints of these two long-distance transport mechanisms.We discuss the possibility that airborne signals reachsystemic tissues at concentrations that cause primingrather than the full induction of defence, and we proposea model that illustrates how vascular and airborne long-distance signals can act in synergy to induce a balancedresistance in distal plant parts.

Systemic induction of resistance at the phenotypiclevelThe first evidence for long-distance regulation of inducedresistance came from the work by Frank Ross in the1960s.In work published in 1961 [1], Ross demonstrated thatinoculation of the lower leaves of tobacco (Nicotiana taba-cum) plants with necrosis-inducing tobacco mosaic virus(TMV) induced a hypersensitive reaction that resulted inenhanced resistance to a second infection in the upperleaves [1]. This phenomenonwas termed systemic acquiredresistance (SAR). SAR was first detected 2–3 days afterprimary virus infection and reached a maximum after 7days, demonstrating that plants require time to generate,transport and deliver the long-distance signal that inducesresistance in the upper leaves. Ten years later, TerrenceGreen and Clarence Ryan [2] reported that tomato (Lyco-persicon esculentum) responds to insect feeding by produ-cing defensive proteins. Proteinase inhibitors (PIs; i.e.,compounds that reduce protein digestion by insects) wereinduced in both damaged and undamaged leaves. Supply-ing extracts from wounded leaves to excised, but otherwiseundamaged, plants could also trigger the production of PIs,a clear hint of the existence of a mobile signal [3]. Finally,systemically transported signals are also involved whentranslocation of defensive compounds contributes tosystemic resistance, for example, when nicotine is pro-duced in the roots of tobacco (Nicotiana sylvestris) uponleaf damage [4].

SAR, PIs and toxic secondary compounds directly targetthe plants’ enemies and are called ‘direct defences’, butplants can also defend themselves by releasing volatileorganic compounds or extrafloral nectar to attract naturalenemies of the herbivores [5,6]. Although some volatilecompounds have direct effects against pathogens and her-bivores, this strategy is generally referred to as ‘indirectdefence’. In many cases, induced volatile emission is asystemic response [7]. For instance, a water-soluble sub-stance from insect-infested lima bean (Phaseolus lunatus)

r Ltd. All rights reserved. doi:10.1016/j.tplants.2008.03.005 Available online 17 May 2008

Review Trends in Plant Science Vol.13 No.6

elicited volatile emissions in undamaged plants [8], andfeeding on roots can induce indirect defences in above-ground parts [9,10]. Thus, the activation of indirectdefences also involves regulation by long-distance signals.

Long-distance signals in the SAR responseThe initial discovery of SAR [1] initiated decades ofresearch into its underlying mechanisms. In the 1970s,Leendert C. van Loon found that both locally and systemi-cally induced resistance to pathogens are associated withthe induction of pathogenesis-related (PR) proteins [11]. Intobacco, injection with salicylic acid (SA) elicited enhancedresistance to TMV, and application of SA triggered theaccumulation of PR-proteins [12,13], leading to speculationthat SA might function as the endogenous signal. In manyplant species, SA undoubtedly plays an important role inboth locally expressed basal resistance and SAR [14]. SARis predominantly effective against pathogens that aresensitive to SA-dependent basal defence mechanisms[15], and so we can even state that SAR represents asystemic enhancement of SA-dependent basal resistance.Consequently, transportation of the elicitor through thewhole plant seems an obvious explanation for the systemicnature of SAR.

How well does SA fit the criteria that define a trans-ported long-distance signal (Box 1)? Intriguingly, SA levelsincrease in the petioles of pathogen-infected tobacco [16]and cucumber (Cucumis sativus) [17] leaves before theonset of SAR. Removal of pathogen-inoculated leavesbefore the SA levels increased in the petioles did not,however, impair the induction of SAR in upper leaves ofcucumber [18]. Furthermore, SA-deficient tobacco root-stocks expressing SA hydroxylase were still capable ofexporting the signal to wild-type scions [19]. Apparently,SAR does not require the accumulation of SA in theattacked tissue. This suggests that SA is not the cruciallong-distance signal in SAR, even though it is transportedthroughout the plant and activates SAR.

According to current views, SA is (mainly) transportedas its methyl ester, methyl salicylate (MeSA). This possib-ility was long overlooked, although originally suggested byVladimir Shulaev and co-workers in 1997 [20]. In tobacco,

Box 1. What qualifies a factor as a transported signal?

Since first discovering systemic plant resistance in response to

locally applied stress, scientists have searched for factors that serve

as long-distance signals. The topic is still a matter of intensive

discussion. Initially, many candidate molecules appeared promis-

ing, for example, because their exogenous application or over-

expression triggered plant-wide defence, or because mutants in

which they were not expressed failed to express resistance. To

qualify fully as a long-distance signal, however, a factor must

(i) induce a defensive response, (ii) be produced or released at the

site of attack, (iii) be translocated from the attacked to the systemic

tissue, and (iv) accumulate in the systemic tissue before resistance

expression takes place. Several factors failed to meet these criteria,

and comparisons among plant species have demonstrated that

different signals might occur in different species, or that different

signals orchestrate distinct defence responses depending on the

nature of the attacking organism. Factors might be derived from the

plant, the attacking agent, or both. For the majority of plant species,

therefore, the search for the translocated signal(s) is not yet

completed.

two enzymes control the balance between SA and MeSA:SA-binding protein2 (SABP2), which converts biologicallyinactive MeSA into active SA [21], and SA methyltrans-ferase1 (SAMT1), which catalyses the formation of MeSAfrom SA [22]. SABP2-silenced tobacco displays a reducedlevel of basal resistance and is unable to express SAR [23].Recently, this line of research reached a significant break-through, when Sang-Wook Park and co-workers [24]demonstrated that MeSA functions as the crucial long-distance SAR signal in tobacco (Box 2). It remains uncer-tain, however, whether MeSA plays a similar role in otherplant species. For instance, in Arabidopsis, lipid-derivedmolecules might function as crucial long-distance signalsin the activation of SAR [25,26].

Long-distance signals in the wound responseInsect feeding inevitably leads to mechanical wounding,and mechanical damage can induce defence responsessimilar to those stimulated by insect feeding [27,28], atleast when the spatio-temporal pattern of the mechanicaldamage resembles that caused by a consuming insect [29].Although the insect’s mode of feeding and elicitors in theinsect’s regurgitate clearly influence the nature and inten-sity of the response, the resulting changes in the plant’sphysiology are commonly termed ‘wound response’ [30,31].

A classic example of a systemic wound response is PIsynthesis in tomato [2]. Biochemical studies identified an18-amino-acid peptide, systemin, that is released at woundsites upon feeding by chewing herbivores. This peptide isprocessed from the 200-amino-acid precursor prosystemin.14C labelling studies showed that systemin is mobilewithin the plant [32]. Furthermore, tomato plants thatoverexpressed the prosystemin gene produced PIs consti-tutively [33]. Nevertheless, grafting experiments —particularly with spr1, a mutant that is defective in theperception of systemin [34] — have revealed that jasmo-nates, rather than systemin, are responsible for thesystemic wound response (Box 3). For a systemic responseto local damage, intact systemin signalling is required inthe damaged tissue but not in the distal responding tissue.It seems, therefore, that systemin locally induces jasmo-nates, which are subsequently transported throughout theplant to trigger the systemic response [30,31].

Thesynthesis of jasmonic acid (JA)and itsprecursorsandderivatives (collectively termed jasmonates) is a classicalwound response,which startswith the liberation of linolenicacid from membranes followed by a multi-step conversioninto JA by several enzymes [35]. Jasmonates induce a broadspectrum of defensive responses, such as the production ofPIs, the release of volatiles, alkaloid production, trichomeformationand the secretionof extrafloralnectar.Recently, itwas discovered that jasmonates (most likely the JA-aminoacid conjugate jasmonoyl–isoleucine) interactwith theCOI1(CORONATIN-INSENSITIVE 1) unit of an E3 ubiquitinligase complex termed SCFCOI1 (Skip/Cullin/F-box–COI1).This event promotes binding of the COI1-unit to JAZ (jas-monate ZIM-domain) proteins, thereby targeting the JAZproteins for ubiquitination and rapid degradation. BecauseJAZproteinsare repressors ofJA-inducible geneexpression,their degradation allows transcription factors to stimulateJA-inducible gene expression [36–38].

265

Box 2. MeSA as long-distance signal in systemic acquired resistance

Salicylic acid (SA) induces resistance to many pathogens [14,78] and

accumulates locally in pathogen-infected tissues of different species

such as Arabidopsis [79], tobacco [16], and cucumber [17]. SA bleeds

out of petioles of pathogen-infected leaves [16,17] and labelling

studies have confirmed systemic transport of SA from pathogen-

infected leaves [80]. However, SAR can develop before SA levels rise

in petioles of infected leaves [19], and accumulation and perception of

SA are only critical in the systemic organs [20]. Recently, Park and co-

workers provided evidence that MeSA, rather than SA, functions as

the critical mobile signal [24] (see Figure I).

Figure I. The role of MeSA as a transported signal in SAR. (a) Grafting experiments with plants in which the SA-binding protein 2 (SABP2) gene is silenced

demonstrated that a SABP2–silenced rootstock, which cannot convert MeSA into SA, was still capable of generating a long-distance signal to activate SAR in an upper

wild-type scion. (b) Conversely, a SA methyltransferase1 (SAMT1)-silenced rootstock, in which MeSA production is impaired, was not capable of generating this long-

distance signal. (c) Furthermore, a recombinant SABP2 protein was constructed that is affected in SA-induced feedback inhibition of MeSA esterase activity. When this

so-called A13L protein was expressed in SABP2-silenced rootstocks, no SAR signal was transported to the upper scion. Hence, SA-induced feedback inhibition of SABP2

is necessary to generate enough MeSA for long-distance transport to distal plant parts, where it is converted by SABP2 into active SA that triggers SAR. Together with

the observation that exposure of lower plant parts to MeSA in gas-tight chambers can trigger SAR in non-exposed, upper leaves, these findings demonstrate that MeSA

functions as the crucial long-distance signal in tobacco [24]. +, expression of SAR; �, no expression of SAR. (d) The interaction of SAMT1 with SABP2 (which is inhibited

by free SA) regulates the synthesis of MeSA from SA in the infected tissue, and the release of SA from the transported MeSA in the systemic tissue. Although MeSA can

be transported within the plant, it is possible that under natural circumstances a combination of airborne and vascular transport of MeSA is responsible for long-

distance regulation of SAR [20]. (e) Structures of SA and methyl salicylate (MeSA).

Review Trends in Plant Science Vol.13 No.6

Biosynthesis of JA is required at the site of damage,whereas its perception is required in the distal plant partsfor a systemic induction of PIs [30]. This shows thatjasmonates function as long-distance signals (Box 3). Inter-estingly, jasmonate signalling is also required for rhizo-bacteria-induced systemic resistance (ISR) in Arabidopsis,even though ISR is not marked by systemic induction ofPIs [39].

Several characteristics make jasmonates optimal can-didates for long-distance signals. First, they transientlyinduce the expression of genes that are required for theirown synthesis, several of which are expressed in thephloem of tomato, which leads to the amplification of thesignal during its transport [40]. Second, jasmonates havebeen shown to augment their own transport [41]. Together,

266

these mechanisms counteract dilution of the signal duringlong-distance transport. Labelling studies [41,42], how-ever, found that a surprisingly low percentage of exogen-ously applied jasmonates arrive at the target organs. It istempting to speculate, therefore, that jasmonate signallingin plants functions similarly to the propagation of actionpotentials in neurons: a locally initiated signal triggers JAsynthesis in the neighbouring sector of the phloem, and themovement of this induction event functions, at least partly,as the moving signal.

Volatiles in long-distance signallingTraditionally, most studies have focused on chemical or, toa lesser extent, electrical [43] signals that travel throughthe vascular system. Damage to a single leaf usually elicits

Box 3. Jasmonates and long-distance signalling in the wound response

Hundreds of studies have demonstrated that exogenous application of

JA induces defensive responses [5,35,81]. Concentrations of JA

increase locally in response to tissue damage [4,28]. In tobacco, these

locally increased JA levels are followed by enhanced synthesis of

nicotine in the roots [4]. Labelling studies using tobacco confirmed that14C-JA was transported from leaves to roots, and spatially distributed

according to a strictly phloem-based transportation pathway [42]. Also

the methyl ester of JA, MeJA, can be transported systemically.

However, 11C-imaging of MeJA in tobacco revealed distribution

patterns of MeJA that could only be explained by transportation

through both phloem and xylem accompanied by exchange between

non-orthostichous vascular pathways [41]. The high vapour pressure of

MeJA as compared to JA might account for the higher mobility of

MeJA and opens the possibility of airborne defence signalling. Grafting

experiments have confirmed that jasmonates function as crucial mobile

signals in the wound response of tomato (see Figure I).

Figure I. The role of jasmonates in the systemic wound response. (a) The importance of JA perception in distal organs was demonstrated by the finding that wild-type

scions of tomato that had been grafted onto damaged rootstocks of the jasmonate-insensitive 1 mutant ( jai1; impaired in the COI1 protein) were still capable of

expressing PIs. Conversely, jai1 scions grafted onto a damaged wild-type rootstock failed to activate the defence response. (b) That systemin is not the systemically

transported signal becomes obvious from the observation that spr1 rootstocks, which are defective in systemin perception, cannot generate the transported signal,

whereas spr1 scions can perceive the signal from wildtype rootstocks and express PIs. (c) Correspondingly, grafts with JA-biosynthetic mutants, such as the suppressor

of prosystemin 2 (spr2) mutants [82] and the acyl-coA oxidase1 (acx1) mutant [83], revealed that intact JA synthesis in the rootstock, and not in the scion, is required for

a functioning signal transmission [30], which can be amplified by active JA synthesis in the vascular bundles [31] +, expression of PIs; �, no expression of PIs. (d) GLVs

(green leaf volatiles) have a double role as local and (in some, but not all plants) systemic products of the wound response and as elicitors of further systemic responses,

which are mediated by vascular and airborne signalling. Starting from membrane-derived linolenic acid, JA is synthesized locally in the damaged tissue via the

octadecanoid pathway, which is activated by systemin or other wounding-associated elicitors. JA can be converted to the more volatile MeJA, and both compounds

have been reported to function as transported signals that induce systemic wound responses such as the synthesis of PIs, GLVs and other volatiles. During wound-

induced synthesis of JA, GLVs are synthesized and subsequently released from the damaged tissue [63]. At relatively low concentrations, GLVs prime plants for

enhanced induction of (Me)JA-dependent defences [57,61], thus they can synergize the systemic wound response to jasmonates. (e) Although structurally not related to

JA and MeJA, GLVs such as (E)-2-hexenal, (Z)-3-hexenol and (Z)-3-hexenyl acetate can induce or prime wound-responsive genes and defence responses, pointing to a

yet-unidentified mechanism of signal perception and transduction.

Review Trends in Plant Science Vol.13 No.6

the strongest responses in orthostichous leaves [44], andresistance expression to herbivores in poplar (Populustrichocarpa x deltoides) [45], tomato [46], wild tobacco[47] and the clonal plant Trifolium repens [48] followsthe spatial patterns of assimilate transport. Furthermore,girdling of the petioles of pathogen-infected leaves blocksSAR development, indicating the involvement of phloem

transport [49]. For these reasons, research on long-dis-tance signals usually focused on the vascular system(Boxes 2 and 3). In some cases, however, phenotypic resist-ance is expressed just as quickly and strongly in distalleaves that lack a direct vascular connection to theattacked leaf as it is in leaves that have such a connection[50,51]. This suggests that additional routes or mechan-

267

Review Trends in Plant Science Vol.13 No.6

isms are behind the induction of systemic resistance inphotosynthesising source leaves or in leaves that are non-orthostichous to the stimulated ones [51,52].

What means of transportation could exist if the signal isnot only transmitted through the vascular system? JA andSA can be methylated and then become volatile. Further-more, both MeJA and MeSA are potent inducers of defenceresponses [20,24,53,54]. It is possible, therefore, thatMeSA and MeJA travel by both airborne transport andvascular transport to mediate long-distance induction ofresistance. In fact, volatile-mediated induction of resist-ance in neighbouring plants is a well-documented phenom-enon [55–57]. Undoubtedly, ‘long-distance signalling’ doesnot stop at a plant’s surface, and it has been suggested thatvolatile compounds also act as within-plant signals [44,58].Indeed, studies using mechanically damaged sagebrush(Artemisia tridentata) revealed that airflow was necessaryfor systemic induction of resistance against herbivores,even among branches of the same individual [59]. Sim-ilarly, systemic induction of extrafloral nectar secretion byleaves of wild lima bean in response to beetle feedingoccurred only when air was moving freely between leaves[57], and volatiles fromherbivore-damaged leaves of poplar(Populus deltoides x nigra) augmented defence responsesin adjacent leaves. Unfortunately, similar studies thatexplicitly demonstrate the involvement of volatiles inthe within-plant regulation of SAR are lacking. This func-tion is extremely likely, however, as demonstrated byIngrid Kiefer and Alan Slusarenko’s observation thatSAR induction in the Arabidopsis rosette extends beyondthe route of assimilate movement along an orthostichy[51], and by the recent finding that MeSA functions as acrucial SAR signal in tobacco [24]. Further studies arerequired to determine how common the phenomenon ofairborne within-plant signalling really is. It seems evident,however, that within-plant signalling by volatiles cansynergize the response to vascular signalling moleculesto regulate systemically expressed resistance (Figure 1).

Likely candidates in airborne long-distance signallingWhich volatiles are responsible for airborne signalling?Green-leaf volatiles (GLVs) have been associated withinduced resistance in intact plants [60–62]. GLVs are C6-compounds that are rapidly released upon tissue damage[63]. Though mainly discussed in the context of insect feed-ing, GLVs are also active against pathogen infection [64],and this type of direct defencemight, in fact, represent theiroriginal function [5]. Another airborne candidate is thegaseous hormone ethylene. Ethylene plays a modulatingrole in plant defensive reactions to pathogens [65]. Forexpression of SAR, ethylene perception is required locallyin the pathogen-infected leaf [66], suggesting that ethylenedoes not function as a long-distance signal. Ethylene alsoaugmented induced volatile production by maize uponexposure to theGLV (Z)-3-hexenol, but exposure to ethylenealone had no effect [67]. This indicates that ethyleneincreases the plant’s response to GLVs but is unlikely toserve as a primary signal. In addition to MeSA, MeJA,ethylene and GLVs, (Z)-jasmone can trigger defensiveresponses having undergone airborne transport. This her-bivore-induced volatile activates different sets of genes than

268

those induced by MeJA [68,69], however, suggesting adifferent mode of action. Correspondingly, (Z)-jasmonefailed to induce extrafloral nectar secretion in lima bean,a typical JA-responsive trait that can be boosted by theGLV(Z)-3-hexenyl acetate [62].

Elucidating the mechanisms by which plants perceivevolatile signals is a major challenge for future research.MeSA andMeJA can be reverted back into the active planthormones jasmonic acid and salicylic acid, respectively. Bycontrast, little is known about the signalling response toGLVs. Details about the molecular perception of GLVsremain unknown. The structural diversity of resistance-inducing GLVsmakes a specific receptor-protein-mediatedmechanism of perception unlikely. It has been suggestedthat GLVs that have an a,b-unsaturated carbonyl groupcan trigger defence through their activity as reactive elec-trophile species [70], but other GLVs that have beenreported to be biologically active lack this motif[62,67,71]. Changes in transmembrane potentials areinvolved in early signalling events in the cellular responseto stress [43], and exposure to several GLVs did changemembrane potentials in intact lima bean leaves (MassimoMaffei, pers. comm.). It is therefore tempting to speculatethat the dissolving of volatiles in the membranes combinedwith interactions with membrane proteins, which aresimilar to the odorant-binding proteins of insects [72],leads to changes in transmembrane potentials and therebyinduces gene activity [71].

Benefits of airborne and vascular signallingAlthough volatiles have emerged as significant signals ininduced resistance, it seems puzzling that plants use air-borne regulation if they are already equipped with asophisticated vascular transportation system. Allowing‘eavesdropping’ by neighbouring plants might have eco-logical costs because neighbours usually compete for light,water and soil nutrients. However, airborne signalling hascrucial advantages over vascular signalling.

First, airborne signalling overcomes restrictions result-ing from the plant’s orthostichy [44,73]. Plant enemiesoften move from one leaf to the adjacent one, but directlyadjacent leaves usually lack direct vascular connections.Vascular restrictions become even more problematic inlarge plants, such as trees, shrubs and vines, in whichspatially neighbouring leaves often originate from separ-ate branches. Furthermore, phloem fluxes are usuallydirected towards non-photosynthesizing tissues, whereastransport through xylem vessels follows the upwards tran-spiration flow. Even signals that are transported throughboth xylem and phloem hardly reach mature leaves thatare older than the signal-producing leaf. By contrast,volatile-mediated induction of extrafloral nectar wasobserved in mature lima bean leaves that inserted belowthe induced leaves [57], and hence, airborne signals appearto have important functions in smaller plants too.

Second, volatile signal molecules can reach distalplant parts faster than compounds that are transportedthrough vascular tissues. It is probably no coincidencethat, of all classes of herbivore-induced volatiles,GLVs are most frequently described as active airbornesignals. GLVs are in part released from pre-

Figure 1. Model of a two-step regulation system of systemic resistance by airborne and vascular long-distance signals. (a) Upon local attack by insects (left) or pathogens

(right), plants rapidly emit airborne signals, such as green leaf volatiles, methyl jasmonate or methylsalicylic acid, which can sensitize distal plant parts for a second

vascular signal, such as jasmonic acid or salicylic acid. (b) Kinetics of systemically expressed induced resistance. Rapid airborne signals (minutes – hours) trigger priming

(priming phase – green symbols) and relatively little defence expression. These signals are followed by a second vascular signal (days – weeks; defence phase, red symbols)

that boosts the induced defence expression. (c) Relationship between induced defence expression (defence amplitude) and dosage of airborne (green arrow) and vascular

(red arrow) long-distance signals. Airborne signals can trigger induced defence at relatively high concentrations, but at lower concentrations, they prime for an enhanced

defence induction by the vascular signal.

Review Trends in Plant Science Vol.13 No.6

existing pools [63] and are generally synthesized fromearly precursors of JA extremely early during the woundresponse (Box 3). Fast signalling is an importantadvantage, as speed of defence activation is cruciallyimportant for the effectiveness of resistance to pathogensand insects.

The velocity and direction of within-plant signalling byvolatiles depends on different biotic and abiotic factors,such as wind speed and direction, and on various plantcharacteristics, such as the presence of leaf hairs or a waxlayer and probably many more factors. Further studieswill be required to determine the extent by which locally

269

Review Trends in Plant Science Vol.13 No.6

released volatiles affect systemic defence expression,and how the chemical properties of volatile signalsaffect the spatiotemporal pattern of systemic inductionof resistance.

Vascular and airborne signalling interact for optimalresistance expressionThe combination of internal and external pathways toachieve systemic defence activation provides plants withan interesting additional option for fine-tuning systemicresistance in yet-undamaged tissues. Relatively highconcentrations of volatiles can trigger defences directly[74], but exposure to lower concentrations of volatiles,which fail to activate defences to their full extent, canstill affect resistance. Such an alternative mode of actioncan be explained by sensitization of the plant’s defencearsenal [57,61,75]. This so-called priming for defence hasbeen associated with various induced resistancephenomena [76]. Priming prepares the plant to respondmore rapidly and/or effectively to subsequent attack, andit can be activated by much lower concentrations of theresistance-inducing signal than those required to induceactive defences fully [69,77]. Hence, primed plants showno enhanced defence activity, but they respond muchfaster or more strongly to wounding or infection than un-primed plants. Although the priming and induction ofdirect defence constitute different mechanisms of protec-tion, both result in a phenotypically similar resistance topathogens and herbivores. Furthermore, some induced-resistance phenomena are based on a combination ofboth mechanisms. For instance, the induction ofpathogen-induced SAR activates some defence-relatedgenes, whereas other defensive genes become primedfor augmented induction to subsequent pathogen attack[76].

Given the fact that volatiles diffuse in the air and aredispersed by eddy currents, it is plausible that resist-ance-inducing volatiles are often diluted to priming con-centrations. In fact, self-priming by herbivore-inducedvolatiles has been described in the context of airbornewithin-plant signalling [57,73]. This suggests a two-stepregulatory system in which airborne and vascular sig-nals interact, probably to varying extents, to achieve anoptimized orchestration of the systemic defence response(Figure 1). In this model, airborne signals prime distaltissues to respond more efficiently to vascular signals ordirect attack. Such a two-step regulation providesadditional fine-tuning of the systemic resistanceresponse, while preventing defence reactions to falsealarm signals. Self-priming by airborne signals preparesthe systemic tissues for a rapid response, but full acti-vation of costly defence mechanisms is allowed only afterconfirmation by the vascular long-distance signal(Figure 1).

Conclusions and outlookJA and SA are crucial signals in the regulation of inducedresistance to herbivores and pathogens, and have beenshown to move in the vascular system. Their volatilederivatives, MeJA and MeSA, however, have been putforward asmore likely candidates to mediate long-distance

270

regulation of induced resistance. MeJA and MeSA, as wellas GLVs, are released from damaged leaves and aremobilein the gas phase. The major benefit of a within-plantsignalling by volatiles is that they overcome the spatialand temporal restrictions of the vascular system. Particu-larly at lower doses, airborne signals can prime for defence,thus preparing the distal organs for a more effectivedefence response rather than fully inducing them. Primingprovides an additional dimension to the delicate balancebetween defence expression and competing processes, suchas growth and reproduction.

Dosage-dependent effects also strongly reduce aputative ecological cost of external signalling: ‘eavesdrop-ping’ by neighbours. Volatile-mediated signalling is mosteffective over relatively short distances, and the prob-ability that the leaf nearest to an attacked one belongsto the same plant is relatively high. As a result, the chancethat eavesdropping by competing plants becomes a quan-titatively relevant problem remains low.

Plants combine vascular and airborne transport of thelong-distance signals, providing a carefully balancedsystemic defence reaction to local attack by herbivoresand pathogens. Although both transportation pathwayscan co-exist within the same plant, their relative contri-bution to the overall response is likely to depend on theplant’s life history and anatomy, and on the nature of theattacking organism. Therefore, different plant species withdifferent life styles will have to be considered if we are togain a better understanding of the strategies by whichplants optimize the induction of systemic resistance. Thesame holds true for the nature of locally and systemicallyfunctioning signals, which might differ amongst plantspecies.

AcknowledgementsWe thank Christian Kost, Peter Bakker, Leendert C. (Kees) van Loon andthree anonymous referees for valuable comments on earlier versions ofthis manuscript. Research activities of J.T. are supported by a BBSRCInstitute Career Path Fellowship (no. BB/E023959/1), M.H. gratefullyacknowledges financial support from the German Research Foundation(DFG-grant He 3169/4–2) and from Consejo Nacional de Ciencia yTecnologıa (CONACYT, Mexico).

References1 Ross, A.F. (1961) Systemic acquired resistance induced by localized

virus infection in plants. Virology 14, 340–3582 Green, T.R. and Ryan, C.A. (1972) Wound-induced proteinase inhibitor

in plant leaves: a possible defense mechanism against insects. Science175, 776–777

3 Ryan, C.A. (1974) Assay and biochemical properties of the proteinaseinhibitor inducing factor, a wound hormone. Plant Physiol. 54, 328–332

4 Baldwin, I.T. et al. (1997) Quantification, correlations andmanipulation of wound-induced changes in jasmonic acid andnicotine in Nicotiana sylvestris. Planta 201, 397–404

5 Heil, M. (2008) Indirect defence via tritrophic interactions.New Phytol.178, 41–61

6 Pieterse, C.M. and Dicke, M. (2007) Plant interactions with microbesand insects: from molecular mechanisms to ecology. Trends Plant Sci.12, 564–569

7 Turlings, T.C.J. and Tumlinson, J.H. (1992) Systemic release ofchemical signals by herbivore-injured corn. Proc. Natl. Acad. Sci.U. S. A. 89, 8399–8402

8 Dicke, M. and Dijkman, H. (2001) Within-plant circulation of systemicelicitor of induced defence and release from roots of elicitor that affectsneighbouring plants. Biochem. Syst. Ecol. 29, 1075–1087

Review Trends in Plant Science Vol.13 No.6

9 Bezemer, T.M. and van Dam, N.M. (2005) Linking aboveground andbelowground interactions via induced plant defenses. Trends Ecol.Evol. 20, 617–624

10 Bruce, T.J. and Pickett, J.A. (2007) Plant defence signalling induced bybiotic attacks. Curr. Opin. Plant Biol. 10, 387–392

11 van Loon, L.C. (1997) Induced resistance in plants and the role ofpathogenesis-related proteins. Eur. J. Plant Pathol. 103, 753–765

12 van Loon, L.C. and Antoniw, J.F. (1982) Comparison of the effects ofsalicylic acid and ethephon with virus-induced hypersensitivity andacquired resistance in tobacco. Eur. J. Plant Pathol. 88, 237–256

13 White, R.F. (1979) Acetylsalicylic-acid (aspirin) induces resistance totobacco mosaic virus in tobacco. Virology 99, 410–412

14 Durrant, W.E. and Dong, X. (2004) Systemic acquired resistance.Annu. Rev. Phytopathol. 42, 185–209

15 Ton, J. et al. (2002) Differential effectiveness of salicylate-dependentand jasmonate/ethylene-dependent induced resistance in Arabidopsis.Mol. Plant Microbe Interact. 15, 27–34

16 Malamy, J. et al. (1990) Salicylic acid: a likely endogenous signal in theresistance response of tobacco to viral infection. Science 250, 1002–1004

17 Metraux, J-P. et al. (1990) Increase in salicylic acid at the onset ofsystemic acquired resistance. Science 250, 1004–1006

18 Rasmussen, J.B. et al. (1991) Systemic induction of salicylic acidaccumulation in cucumber after inoculation with Pseudomonassyringae pv. syringae. Plant Physiol. 97, 1342–1347

19 Vernooij, B. et al. (1994) Salicylic acid is not the translocated signalresponsible for inducing systemic acquired resistance but is required insignal transduction. Plant Cell 6, 959–965

20 Shulaev, V. et al. (1997) Airborne signalling by methyl salicylate inplant pathogen resistance. Nature 385, 718–721

21 Forouhar, F. et al. (2005) Structural and biochemical studies identifytobacco SABP2 as a methyl salicylate esterase and implicate it in plantinnate immunity. Proc. Natl. Acad. Sci. U. S. A. 102, 1773–1778

22 Ross, J.R. et al. (1999) S-adenosyl-L-methionine: salicylic acid carboxylmethyltransferase, an enzyme involved in floral scent production andplant defense, represents a new class of plant methyltransferases.Arch. Biochem. Biophys. 367, 9–16

23 Kumar, D. and Klessig, D.F. (2003) High-affinity salicylic acid-bindingprotein 2 is required for plant innate immunity and has salicylic acid-stimulated lipase activity. Proc. Natl. Acad. Sci. U. S. A. 100, 16101–16106

24 Park, S.W. et al. (2007) Methyl salicylate is a critical mobile signal forplant systemic acquired resistance. Science 318, 113–116

25 Maldonado, A.M. et al. (2002) A putative lipid transfer proteininvolved in systemic resistance signalling in Arabidopsis. Nature419, 399–403

26 Truman, W. et al. (2007) Arabidopsis systemic immunity usesconserved defense signaling pathways and is mediated byjasmonates. Proc. Natl. Acad. Sci. U. S. A. 104, 1075–1080

27 Major, I.T. and Constabel, C.P. (2006) Molecular analysis of poplardefense against herbivory: comparison of wound- and insect elicitor-induced gene expression. New Phytol. 172, 617–635

28 Heil, M. et al. (2001) Extrafloral nectar production of the ant-associated plant, Macaranga tanarius, is an induced, indirect,defensive response elicited by jasmonic acid. Proc. Natl. Acad. Sci.U. S. A. 98, 1083–1088

29 Mithofer, A. et al. (2005) Effects of feeding Spodoptera littoralis onLima bean leaves. II. Continuous mechanical wounding resemblinginsect feeding is sufficient to elicit herbivory-related volatile emission.Plant Physiol. 137, 1160–1168

30 Schilmiller, A.L. and Howe, G.A. (2005) Systemic signaling in thewound response. Curr. Opin. Plant Biol. 8, 369–377

31 Wasternack, C. et al. (2006) The wound response in tomato - role ofjasmonic acid. J. Plant Physiol. 163, 297–306

32 Narvaez-Vasquez, J. et al. (1995) Autoradiographic and biochemicalevidence for the systemic translocation of systemin in tomato plants.Planta 195, 593–600

33 McGurl, B. et al. (1994) Overexpression of the prosystemin gene intransgenic tomato plants generates a systemic signal thatconstitutively induces proteinase inhibitor synthesis. Proc. Natl.Acad. Sci. U. S. A. 91, 9799–9802

34 Lee, G.I. and Howe, G.A. (2003) The tomato mutant spr1 is defective insystemin perception and the production of a systemic wound signal fordefense gene expression. Plant J. 33, 567–576

35 Wasternack, C. (2007) Jasmonates: An update on biosynthesis, signaltransduction and action in plant stress response, growth anddevelopment. Ann. Bot. (Lond.) 100, 681–697

36 Thines, B. et al. (2007) JAZ repressor proteins are targets of theSCFCOI1 complex during jasmonate signalling. Nature 448, 661–666

37 Chini, A. et al. (2007) The JAZ family of repressors is themissing link injasmonate signalling. Nature 448, 666–672

38 Staswick, P.E. (2008) JAZing up jasmonate signaling.Trends Plant Sci.13, 66–71

39 van Loon, L.C. et al. (1998) Systemic resistance induced by rhizospherebacteria. Annu. Rev. Phytopathol. 36, 453–483

40 Stenzel, I. et al. (2003) Allene oxide cyclase dependence of the woundresponse and vascular bundle-specific generation of jasmonates intomato — amplification in wound-signalling. Plant J. 33, 577–589

41 Thorpe, M.R. et al. (2007) 11C-imaging: methyl jasmonate moves inboth phloem and xylem, promotes transport of jasmonate, and ofphotoassimilate even after proton transport is decoupled. Planta226, 541–551

42 Zhang, Z-P. and Baldwin, I.T. (1997) Transport of [2-14C]jasmonic acidfrom leaves to roots mimics wound-induced changes in endogenousjasmonic acid pools in Nicotiana sylvestris. Planta 203, 436–441

43 Maffei, M.E. et al. (2007) Before gene expression: early events in plant–insect interaction. Trends Plant Sci. 12, 310–316

44 Orians, C. (2005) Herbivores, vascular pathways, and systemicinduction: facts and artifacts. J. Chem. Ecol. 31, 2231–2242

45 Davis, J.M. et al. (1991) Assimilate movement dictates remote sites ofwound-induced gene expression in poplar leaves. Proc. Natl. Acad. Sci.U. S. A. 88, 2393–2396

46 Orians, C.M. et al. (2000) Vascular architecture generates fine-scalevariation in the systemic induction of proteinase inhibitors in tomato.J. Chem. Ecol. 26, 471–485

47 Schittko, U. and Baldwin, I.T. (2003) Constraints to herbivore-inducedsystemic responses: bidirectional signaling along orthostichies inNicotiana attenuata. J. Chem. Ecol. 29, 763–770

48 Gomez, S. and Stuefer, J.F. (2006) Members only: induced systemicresistance toherbivory ina clonal plantnetwork.Oecologia147, 461–468

49 Guedes, M.E.M. et al. (1980) Induced systemic resistance toanthracnose in cucumber as influenced by the location of theinducer inoculation with Colletotrichum lagenarium and the onset offlowering and fruiting. Physiol. Plant Pathol. 17, 229–233

50 Mutikainen, P. et al. (1996) Herbivore-induced resistance in Betulapendula: the role of plant vascular architecture. Oecologia 108, 723–727

51 Kiefer, I.W. and Slusarenko, A.J. (2003) The pattern of systemicacquired resistance induction within the Arabidopsis rosette inrelation to the pattern of translocation. Plant Physiol. 132, 840–847

52 van Bel, A.J.E. and Gaupels, F. (2004) Pathogen-induced resistanceand alarm signals in the phloem. Mol. Plant Pathol. 5, 495–504

53 Karban, R. et al. (2000) Communication between plants: inducedresistance in wild tobacco plants following clipping of neighboringsagebrush. Oecologia 125, 66–71

54 Farmer, E.E. and Ryan, C.A. (1990) Interplant communication:airborne methyl jasmonate induces synthesis of proteinaseinhibitors in plant leaves. Proc. Natl. Acad. Sci. U. S. A. 87, 7713–7716

55 Baldwin, I.T. and Schultz, J.C. (1983) Rapid changes in tree leafchemistry induced by damage: evidence for communication betweenplants. Science 221, 277–279

56 Rhoades, D.F. (1983) Responses of alder and willow to attack by tentcaterpillars and webworms: evidence for pheromonal sensitivity ofwillows. In Plant Resistance to Insects (Hedin, P.A., ed.), pp. 55–68,American Chemical Society

57 Heil, M. and Silva Bueno, J.C. (2007) Within-plant signaling byvolatiles leads to induction and priming of an indirect plant defensein nature. Proc. Natl. Acad. Sci. U. S. A. 104, 5467–5472

58 Farmer, E.E. (2001) Surface-to-air signals. Nature 411, 854–85659 Karban, R. et al. (2006) Damage-induced resistance in sagebrush:

volatiles are key to intra- and interplant communication. Ecology87, 922–930

60 Arimura, G. et al. (2000) Herbivory-induced volatiles elicit defencegenes in lima bean leaves. Nature 406, 512–515

61 Engelberth, J. et al. (2004) Airborne signals prime plants against insectherbivore attack. Proc. Natl. Acad. Sci. U. S. A. 101, 1781–1785

271

Review Trends in Plant Science Vol.13 No.6

62 Kost, C. and Heil, M. (2006) Herbivore-induced plant volatiles inducean indirect defence in neighbouring plants. J. Ecol. 94, 619–628

63 Turlings, T.C.J. and Wackers, F.L. (2004) Recruitment of predatorsand parasitoids by herbivore-injured plants. In Advances in InsectChemical Ecology (Cardes, R.T. and Millar, J.G., eds), pp. 21–75,Cambridge University Press

64 Shiojiri, K. et al. (2006) Changing green leaf volatile biosynthesis inplants: an approach for improving plant resistance against bothherbivores and pathogens. Proc. Natl. Acad. Sci. U. S. A. 103,16672–16676

65 van Loon, L.C. et al. (2006) Ethylene as a modulator of diseaseresistance in plants. Trends Plant Sci. 11, 184–191

66 Verberne, M.C. et al. (2003) Signaling of systemic acquired resistancein tobacco depends on ethylene perception. Plant J. 35, 27–32

67 Ruther, J. and Kleier, S. (2005) Plant–plant signaling: ethylenesynergizes volatile emission in Zea mays induced by exposure to (Z)-3-hexen-1-ol. J. Chem. Ecol. 31, 2217–2222

68 Birkett, M.A. et al. (2000) New roles for cis-jasmone as an insectsemiochemical and in plant defense. Proc. Natl. Acad. Sci. U. S. A.97, 9329–9334

69 Bruce, T.J.A. et al. (2007) Stressful ‘memories’ of plants: evidence andpossible mechanisms. Plant Sci. 173, 603–608

70 Almeras, E. et al. (2003) Reactive electrophile species activate defensegene expression in Arabidopsis. Plant J. 34, 205–216

71 Heil, M. et al. (2008) Defence-inducing volatiles: in search for the activemotif. J. Chem. Ecol. 34, 601–604

72 Campanacci, V. et al. (2001) Revisiting the specificity of Mamestrabrassicae and Antheraea polyphemus pheromone-binding proteins

272

with a fluorescence binding assay. J. Biol. Chem. 276, 20078–20084

73 Frost, C.J. et al. (2007) Within-plant signalling by volatiles overcomesvascular constraints on systemic signalling and primes responsesagainst herbivores. Ecol. Lett. 10, 490–498

74 Baldwin, I.T. et al. (2006) Volatile signaling in plant–plantinteractions: ‘talking trees’ in the genomics era. Science 311, 812–815

75 Ton, J. et al. (2007) Priming by airborne signals boosts direct andindirect resistance in maize. Plant J. 49, 16–26

76 Conrath, U. et al. (2006) Priming: getting ready for battle. Mol. PlantMicrobe Interact. 19, 1062–1071

77 van Hulten, M. et al. (2006) Costs and benefits of priming for defense inArabidopsis. Proc. Natl. Acad. Sci. U. S. A. 103, 5602–5607

78 Metraux, J-P. (2001) Systemic acquired resistance and salicylic acid:current state of knowledge. Eur. J. Plant Pathol. 107, 13–18

79 Cameron, R.K. et al. (1999) Accumulation of salicylic acid and PR-1gene transcripts in relation to the systemic acquired resistance (SAR)response induced by Pseudomonas syringae pv. tomato in Arabidopsis.Physiol. Mol. Plant Pathol. 55, 121–130

80 Molders, W. et al. (1996) Transport of salicylic acid in tobacconecrosis virus-infected cucumber plants. Plant Physiol. 112, 787–792

81 Farmer, E.E. et al. (2003) Jasmonates and related oxylipins in plantresponses to pathogenesis and herbivory. Curr. Opin. Plant Biol. 6,372–378

82 Stratmann, J.W. (2003) Long-distance run in the wound response —jasmonic acid is pulling ahead. Trends Plant Sci. 8, 247–250

83 Li, C. et al. (2005) Role of b-oxidation in jasmonate biosynthesis andsystemic wound signaling in tomato. Plant Cell 17, 971–986