|Home | About | Journals | Submit | Contact Us | Français|
Plant tissues are highly vulnerable to injury by herbivores, pathogens, mechanical stress, and other environmental insults. Optimal plant fitness in the face of these threats relies on complex signal transduction networks that link damage-associated signals to appropriate changes in metabolism, growth, and development. Many of these wound-induced adaptive responses are triggered by de novo synthesis of the plant hormone jasmonate (JA). Recent studies provide evidence that JA mediates systemic wound responses through distinct cell autonomous and nonautonomous pathways. In both pathways, bioactive JAs are recognized by an F-box protein-based receptor system that couples hormone binding to ubiquitin-dependent degradation of transcriptional repressor proteins. These results provide a new framework for understanding how plants recognize and respond to tissue injury.
Terrestrial plants encounter numerous environmental conditions that have the potential to injure or destroy above- and below-ground tissues. Arthropod herbivores and pathogens pose one of the most serious threats to plant survival. Tissue injury inflicted by plant-eating organisms does not go unnoticed; strong selection pressure imposed by biotic stress has fueled the evolution of innovative mechanisms by which plants perceive and respond to tissue damage (Howe and Jander, 2008). Plant hormones occupy a central role in regulating these highly dynamic and adaptive responses (Pieterse et al., 2009; Santner et al., 2009).
The idea that plants actively respond to tissue injury dates back to work by Wiesner in the 19th century (Wiesner, 1892). Bioassays were developed and used to purify the so-called wound hormone traumatin, which exhibits mitogenic properties related to wound healing (Wehnelt, 1927; English et al., 1939). Traumatin is a reactive fatty acid derivative (12-oxo-trans-10-dodecenoic acid) produced by the hydroperoxy lyase branch of the lipoxygenase pathway (Zimmerman and Coudron, 1979). Although the apparent limited activity of traumatin to specific bean cultivars dampened enthusiasm for the concept of a wound hormone, this early work helped to arouse interest in biochemical research on lipoxygenase-based pathways for oxidative metabolism of polyunsaturated fatty acids (Vick and Zimmerman, 1984; Hamberg and Gardner, 1992; Farmer, 1994; Matsui, 2006). Today, it is widely accepted that oxylipins (oxygenated fatty acid derivatives) produced by these pathways perform a variety of essential functions in the plant life cycle (Blee, 2002; Feussner and Wasternack, 2002; Howe and Schilmiller, 2002; Browse, 2009).
Wound signaling research in plants was revitalized in the 1970s following the discovery that leaf damage induces systemic accumulation of proteinase inhibitors (PIs) that impair insect digestive enzymes (Green and Ryan, 1972). This landmark paper introduced the idea that mobile signals generated at the site of injury orchestrate systemic protection against insect herbivores. During the past 35 years, systemic wound responses have been demonstrated in species throughout the plant kingdom, and intense research efforts have focused on the identification of systemic wound signals and the mechanisms by which they are generated, transported, and perceived (Karban and Baldwin, 1997; Ryan, 2000; Leon et al., 2001; Howe, 2004). The concept that biotic agents activate local and systemic host defense responses is a cornerstone of current theories of plant immunity (Jones and Dangl, 2006; Browse and Howe, 2008; Howe and Jander, 2008; Boller and He, 2009; Dicke et al., 2009; Wu and Baldwin, 2009). Oxylipins play a central role in many of these defense signaling pathways.
Tissue injury invariably results in perturbation or damage to cell membranes. In both plants and animals, oxylipins derived from membrane lipids are sentinels of wound stress. These signaling compounds are typically synthesized de novo in specific cell types upon activation of lipases that release fatty acids from membrane lipids. The biochemical logic underlying oxylipin synthesis in plants and animals is remarkably similar, with species in both kingdoms using lipoxygenases (or other oxygenases) and cytochromes P450 to oxidize fatty acids (Bergey et al., 1996; Blee, 2002; Lee et al., 2008). Several intrinsic features of oxylipins make them well suited for this purpose: (i) a large supply of precursors in the form of membrane lipids, (ii) rapid synthesis in response to cell damage or other inductive cues, (iii) structural diversity that permits functional specificity, (iv) capacity for intra- and intercellular transport, and (v) rapid turnover or inactivation. Knowledge gained from the study of plant oxylipins is providing insight into the underlying mechanisms and evolutionary origins of fatty acid-based signaling pathways in diverse biological systems (Lee et al., 2008).
Here, we discuss recent advances in our understanding of the role of the jasmonate (JA) family of oxylipins in plant responses to wound stress. First, we consider key features of JA as a wound hormone, including its mechanism of perception by plant cells. Second, we describe damage-associated signals that activate JA synthesis and discuss what these signals tell us about how plants perceive tissue injury. Finally, we review recent studies concerning the role of JA in regulating systemic wound responses. Because of the focus on JA, we do not discuss JA-independent wound responses or other oxylipins (e.g., green leaf volatiles produced by the hydroperoxide lyase pathway) that affect plant interactions with other organisms. Readers are referred to several review articles for additional information on these topics (Leon et al., 2001; Howe, 2004; Matsui, 2006; Felton and Tumlinson, 2008; Heil and Ton, 2008; Dicke et al., 2009).
The discovery of JA as a potent elicitor of proteinase inhibitor (PI) expression in tomato (Farmer and Ryan, 1990) and secondary metabolism in plant cell cultures (Gundlach et al., 1992) provided initial insight into the importance of JA as a defense hormone. Subsequent genetic studies established the indispensable role of JA in promoting resistance to herbivores and pathogens that betray their presence by activating JA synthesis in damaged or infected host tissue (Howe et al., 1996; McConn et al., 1997; Vijayan et al., 1998; Kessler et al., 2004; Browse and Howe, 2008; Howe and Jander, 2008). JA exerts its protective effects by regulating a wide range of defense-related processes, including the synthesis of toxic secondary metabolites (Pauwels et al., 2009), production of morphological barriers (Yoshida et al., 2009), and changes in the rate of vegetative growth (Yan et al., 2007; Balbi and Devoto, 2008; Zhang and Turner, 2008). These broad effects of JA imply a general role for the hormone in controlling tradeoffs between growth and defense (Herms and Mattson, 1992; Baldwin, 1998).
JA responses typically depend on large-scale changes in gene expression. In unstressed cells containing low JA levels, transcription factors that promote expression of JA-responsive genes are repressed by members of the JASMONATE ZIM-domain (JAZ) protein family (Chini et al., 2007; Thines et al., 2007; Yan et al., 2007). Increased JA levels stimulate JAZ binding to CORONATINE INSENSITIVE1 (COI1), which is the F-box protein component of the E3 ubiquitin ligase SCFCOI1 (Balbi and Devoto, 2008; Thines et al., 2007). Hormone-induced COI1-JAZ interaction triggers JAZ degradation via the ubiquitin/26S proteasome pathway, thereby releasing transcription factors from repression. Recent evidence indicates that the repressive action of JAZ proteins depends on their ability to dimerize via the conserved ZIM domain (Chini et al., 2009; Chung and Howe, 2009). Additional details about the SCFCOI1/JAZ pathway are described by Chung et al. (2009) in this issue.
A complete understanding of JA as a sentinel of tissue injury requires knowledge of how specific chemical forms of the hormone interact with cognate receptor(s). For this purpose, it is useful to distinguish bioactive JAs, which directly promote the formation of COI1-JAZ complexes, from non-bioactive JAs, which are either precursors or deactivated forms of the bioactive compounds (Katsir et al., 2008a). This distinction is important for defining the identity of specific components of the “signal signature” (Weber et al., 1997; de Vos et al., 2005) that evoke a particular response. Following its synthesis by the octadecanoid pathway in response to wounding or other inductive cues, the (3R,7S) stereoisomer of JA (also known as (+)-7-iso-JA) is metabolized to a wide range of derivatives (Fig. 1). The physiological function of the vast majority of these compounds is not known. It is well established, however, that many JA derivatives exert distinct biological activities when applied to plant tissues (Blechert et al., 1999; Miersch et al., 1999; Stintzi et al., 2001; Ribot et al., 2008; Wang et al., 2008). Because exogenous JAs do not target specific cell types, are often administered at non-physiological concentrations, and are subject to unknown metabolic fates, additional approaches are required to determine the in vivo role of specific JA derivatives.
A major breakthrough in identifying bioactive JAs came from the development of techniques to study hormone action with cell-free and yeast-based protein-protein interaction assays (Dharmasiri et al., 2003; Thines et al., 2007). These approaches were used to demonstrate that COI1-JAZ interactions are stimulated by the JA-amino acid conjugate jasmonoyl-L-isoleucine (JA-Ile), but not by JA, methyl-JA (MeJA), or the JA precursor 12-oxo-phytodienoic acid (OPDA) (Thines et al., 2007; Katsir et al., 2008b; Melotto et al., 2008). Identification of JA-Ile as a causal signal for COI1-JAZ binding extends previous work showing that conjugation of JA to Ile by the acyl-adenylate/thioester-forming enzyme (JASMONATE RESISTANT1) (JAR1) activates the JA signal (Staswick and Tiryaki, 2004; Staswick, 2008).
Synthetic preparations of JA-Ile used for initial COI1-JAZ interaction studies (Thines et al., 2007; Katsir et al., 2008b; Melotto et al., 2008) contained a mixture of the two naturally occurring stereoisomers, namely (3R,7S)-JA-Ile (also known as (+)-7-iso-JA-Ile) and (3R,7R)-JA-Ile (also referred to as (-)-JA-Ile) (Fig. 1). In plant tissues and in solution, these two isomers are in equilibrium, with the thermodynamically more stable (3R,7R)-JA-Ile in excess (~94:6) over the less stable (3R,7S)-JA-Ile isomer (Miersch et al., 1986; Creelman and Mullet, 1997; Wasternack, 2007; Fonseca et al., 2009). Preparations of JA-Ile synthesized in this manner have been shown to activate JA-responsive gene expression and JA-dependent physiological processes (Kramell et al., 1997, 2000; Wasternack et al., 1998; Staswick and Tiryaki, 2004; Wang et al., 2008). Initial COI1-JAZ interaction studies (Thines et al., 2007; Katsir et al., 2008b; Melotto et al., 2008) used these preparations to demonstrate that JA-Ile is the relevant signal for COI1-JAZ partnering but, because of the mixed composition of isomers, did not draw conclusions about the relative activity of (3R,7S)-JA-Ile and (3R,7R)-JA-Ile. Recent experiments performed with purified isomers showed that (3R,7S)-JA-Ile is much more active than stereochemically pure (3R,7R)-JA-Ile in promoting JAZ9 interaction with COI1, as well as JA-dependent phenotypic responses in Arabidopsis (Fonseca et al., 2009). These findings demonstrate that (3R,7S)-JA-Ile is an endogenous bioactive JA, whereas the (3R,7R)-JA-Ile isomer is largely inactive.
The relative potency of (3R,7S)-JA-Ile in promoting COI1-JAZ interaction is consistent with the fact that this compound is stereochemically similar to the bacterial toxin coronatine that acts as a potent agonist of the JA receptor (Katsir et al., 2008b; Melotto et al., 2008; Staswick, 2008). The ability of radiolabeled coronatine to bind COI1-JAZ complexes with high affinity (Kd ~20 nM) in a manner that depends on COI1 indicates that COI1 or a COI1-JAZ complex is a component of the coronatine receptor, and most likely the receptor for active stereoisomers of JA-Ile as well (Katsir et al., 2008b; Melotto et al., 2008). The conclusion that COI1 is part of a receptor for JA-Ile is supported by extensive genetic evidence showing that loss of COI1 function abolishes JA responses in planta (Feys et al., 1994; Li et al., 2004; Paschold et al., 2007). Formal biochemical proof that COI1 or a COI1-JAZ complex is a JA receptor will require demonstration that specific isomers of JA-Ile bind specifically, saturably, and reversibly to the purified receptor.
Current models indicate that JA signaling is initiated upon accumulation of JA-Ile to levels that are sufficient to promote COI1-JAZ interactions. A detailed understanding of JA/JA-Ile homeostasis in healthy and distressed tissues is therefore paramount to understanding how the pathway is controlled. It is well established that JA levels vary widely depending on developmental stage and environmental conditions (Creelman and Mullet, 1997; Wasternack, 2007). Leaf damage inflicted by mechanical wounding and herbivory are highly effective triggers for de novo JA synthesis (Blechert et al., 1995; McCloud and Baldwin, 1997; Weber et al., 1997; Strassner et al., 2002; Koo et al., 2009). Mechanical damage results in large increases (~25-fold) in JA and JA-Ile accumulation within 5 min of tissue injury (Chung et al., 2008; Glauser et al., 2008; Koo et al., 2009). The remarkable speed of this response indicates that all biosynthetic enzymes involved in production of JA/JA-Ile are present in resting cells. Wounded leaves also accumulate other JA-amino acid conjugates, including JA-Val and JA-Leu (Wang et al., 2007; Suza and Staswick, 2008; Koo et al., 2009). The ability of JA-Val and JA-Leu to promote interaction of tomato COI1 and JAZ1/3 in vitro (Katsir et al., 2008b) suggests that these compounds may be active signals in the wound response. However, because the wound-induced level of JA-Ile is at least 10-fold higher than that of JA-Val and JA-Leu, it seems likely that JA-Ile is the major signal for COI1-dependent wound responses in Arabidopsis (Koo et al., 2009).
The available evidence is consistent with a model in which wound-induced production of (3R,7S)-JA-Ile promotes SCFCOI1-dependent degradation of JAZs, thereby allowing downstream transcription factors to engage early response genes. The transient nature of JA-Ile accumulation and JA responses in wounded leaves (Wang et al., 2007; Chung et al., 2008; Suza and Staswick, 2008; Koo et al., 2009) implies the existence of efficient mechanisms to metabolize or inactivate the signal. Signal attenuation may be accomplished by hydroxylation of JA/JA-Ile (Glauser et al., 2008; Miersch et al., 2008) or by epimerization of (3R,7S)-JA-Ile to (3R,7R)-JA-Ile (Fig. 1; Farmer, 1994; Fonseca et al., 2009). Rigorous testing of these ideas will require identification and transgenic manipulation of enzymes involved in JA-Ile metabolism. Plant cells likely use additional mechanisms to attenuate JA-mediated wound responses. In Arabidopsis leaves, for example, wound-induced expression of JA response genes declines sharply during a period when JA-Ile levels remain high (Koo et al., 2009). This observation suggests that damaged leaves become desensitized to the JA-Ile signal. Wound-induced production of JAZ repressors that are stabilized against hormone-induced degradation may be involved in this form of negative feedback control (Chung and Howe, 2009).
JA synthesis is controlled at the level of substrate availability (Stenzel et al., 2003b; Wasternack, 2007). Lipases that release JA precursors (e.g., linolenic acid) from plastid lipids generate substrates for JA synthesis (Ishiguro et al., 2001; Hyun et al., 2008). Expression of these lipases in reproductive tissues appears to be regulated by developmental cues (Ito et al., 2007). However, the manner in which lipase or other rate-limiting enzymes in the JA biosynthetic pathway are controlled in response to wounding and herbivory is not known. Elucidation of these mechanisms will significantly advance our understanding of how JA synthesis is regulated and, more broadly, how plants recognize tissue injury.
Several observations are pertinent to the question of how wounding controls de novo synthesis of JA. First, mechanical tissue damage is sufficient to trigger a rapid and transient JA/JA-Ile burst. The timing and amplitude of JA accumulation is affected by the temporal and spatial patterns of leaf damage (McCloud and Baldwin, 1997; Mithöfer et al., 2005; Stork et al., 2009). Second, many wound-induced defense responses are stimulated by compounds in insect oral secretions and by host-derived elicitors. Numerous studies have shown, for example, that insect feeding or application of oral secretions to wound sites elicits a different volatile response than mechanical damage alone (Turlings et al., 1990; De Moraes et al., 2001; Arimura et al., 2004a, b; Schmelz et al., 2006). This and related phenomena suggest that plants have evolved recognition systems to perceive compounds of insect origin (Truitt et al., 2004; Schmelz et al., 2006; Felton and Tumlinson, 2008; Mithöfer and Boland, 2008; Wu and Baldwin, 2009). Finally, de novo JA synthesis can be activated in the absence of wound stress. Signals produced during reproductive development, for example, promote JA synthesis in specific cell types (Hause et al., 2000; Itoh et al., 2007). Several chemical elicitors, including the peptide systemin and cell wall-derived oligosaccharides, also induce synthesis of the hormone in undamaged leaves (Mueller et al., 1993; Doares et al., 1995; Lee and Howe, 2003). As discussed in greater detail below, wounding can also activate de novo JA synthesis in undamaged parts of the plant.
These collective observations imply the existence of cell surface receptors or other sensory systems that, upon recognition of non-self or damage-associated (self) signals, activate the octadecanoid pathway for JA synthesis (Fig. 2). This concept is consistent with models of plant innate immunity in which pathogen- or microbe-associated molecular patterns (PAMPs/MAMPs) trigger host defense responses by activating pattern recognition receptors (PRRs) at the plant cell surface (Boller and He, 2009). The term herbivore-associated molecular pattern (HAMP) has been used to describe herbivore-derived compounds that elicit host defense responses (Mithöfer and Boland, 2008). Several compounds that modulate the host response have been identified in insect oral secretions, saliva, and oviposition fluids (Felton and Tumlinson, 2008; Schmelz et al., 2009; Wu and Baldwin, 2009). The mechanisms by which these compounds are perceived by plant cells, however, remain elusive.
Rapid accumulation of JA in response to mechanical stress or wounding indicates that de novo JA synthesis can be efficiently triggered by plant-derived signals. The danger model of vertebrate immunity has used the term damage-associated molecular patterns (DAMPs) to describe host-derived signals that trigger immune responses (Matzinger, 2002; Seong and Matzinger, 2004). In plants, these signals may include peptide or cell wall-derived oligosaccharide fragments (Narvaez-Vasquez et al., 2007), as well as wound-induced hydraulic or electrical signals (Malone, 1993; Zimmermann et al., 2009). Recognition of the latter physical signals may involve sensors of mechanical stress or cell wall integrity. Genetic and pharmacological disruption of cell wall integrity has been shown to constitutively activate the JA pathway (Ellis et al., 2002; Hamann et al., 2009). There is also evidence to indicate that mitogen-activated protein (MAP) kinase cascades are involved in the activation of JA synthesis in response to wound damage and wound-associated elicitors, including fatty-acid amino acid conjugates and systemin (Seo et al., 1999; Kandoth et al., 2007; Wu et al., 2007). Wound-induced ion fluxes have been implicated in the early steps of wound recognition, but the underlying mechanisms are still vague (Schaller and Oecking, 1999; Bonaventure et al., 2007).
Biotic stress often results in systemic expression or priming of host defense responses that serve to protect healthy tissues from secondary attack (Conrath et al., 2006; Frost et al., 2007; Ton et al., 2007; Erb et al., 2008; Heil and Ton, 2008; Vlot et al., 2008). A definitive role for JA in long-distance responses to wounding and herbivory was established through the use of mutants that are defective in JA synthesis or perception (Leon et al., 2001; Schilmiller and Howe, 2005; Wasternack et al., 2006). JA-dependent systemic responses are highly diverse and are expressed over various temporal and spatial scales. Rapid systemic changes in gene expression and hormone levels occur within minutes of tissue injury. Other wound responses, including the production of toxins, feeding deterrents (e.g., PIs), and volatiles that mediate tritrophic interactions, are slower (e.g., measured in hours) (Green and Ryan, 1972; Engelberth, 2004; Mithöfer et al., 2005; Baldwin et al., 2006; Matsui, 2006; Heil and Silva Bueno, 2007; Pauwels et al., 2009). Wound-induced developmental responses such as trichome initiation are manifested over the course of days (Yoshida et al., 2009).
There are two general models to explain how the JA pathway components are spatially organized between damaged and undamaged leaves: (i) JA produced at the wound site serves as a mobile signal to activate responses in systemic tissues, and (ii) wound-induced production of a mobile signal other than JA activates synthesis of the hormone in systemic tissues. We refer to these two scenarios as the cell non-autonomous and autonomous pathways, respectively, for what they imply about sites of JA synthesis and action. Both models predict wound-induced systemic increases in JA/JA-Ile levels and subsequent perception of JA-Ile by the SCFCOI1/JAZ receptor complex (Fig. 2). The models differ in several key respects, including the source of systemic JA/JA-Ile (i.e., transported or de novo synthesized), the requirement for JA synthesis in the production of the mobile signal and, of course, the identity of the mobile signal.
There is considerable evidence to support the hypothesis that JA (or a JA derivative) acts cell nonautonomously in the systemic wound response. This topic has been extensively reviewed (Stratmann, 2003; Howe, 2004; Schilmiller and Howe, 2005; Wasternack et al., 2006; Heil and Ton, 2008; Howe and Jander, 2008) and thus we summarize only the main points here. Exogenous JA exerts systemic responses solanaceous plants (Farmer et al., 1992; Halitschke and Baldwin, 2003; Zhao et al., 2003; Pluskota et al., 2007). Moreover, isotope labeling experiments indicate that JA/MeJA can be transported in the phloem and xylem (Zhang and Baldwin, 1997; Thorpe et al., 2007). In Arabidopsis, exogenous JA did not elicit systemic expression of a reporter gene that is expressed in distal leaves of wounded plants (Kubigsteltig et al., 1999). Results from these and other JA application experiments are thus inconsistent and, as mentioned above, are inadequate to infer a role for endogenous JA in the wound response.
A function for JA as a mobile signal in the wound response is supported by immunocytochemical studies showing that tomato JA biosynthetic enzymes are located in the companion cell-sieve element complex of vascular tissue (Hause et al., 2000, 2003). The preferential accumulation of JA and OPDA in vascular bundles further indicates that the JA synthetic pathway is operational in these tissues (Stenzel et al., 2003a). In Arabidopsis, various JA derivatives are enriched in the midveins of wounded leaves (Glauser et al., 2008), and it was also shown that petiole exudates contain JA (Truman et al., 2007; Chaturvedi et al., 2008). Additional support for a phloem-mobile wound signal comes from studies showing that spatial patterns of wound-induced systemic gene expression are influenced by the strength of vascular connections between damaged and responding leaves (Davis et al., 1991; Orians et al., 2000; Schittko and Baldwin, 2003). Estimates of the speed (1 to 5 cm/h) of the systemic signal for wound-induced PI expression in tomato are consistent with the involvement of the phloem in signal transport (Nelson et al., 1983; Ryan, 2000). Taken together, these observations support the hypothesis that JA or one of its derivatives is a mobile signal in the systemic wound response (Fig. 2; cell nonautonomous pathway).
Grafting experiments have provided functional evidence that JA is a component of the long-distance signal for PI expression in tomato. Grafts between wild-type and COI1-deficient plants showed that JA responsiveness is not required for production of the systemic signal in damaged leaves, but rather is necessary for recognition of the signal in undamaged systemic leaves (Li et al., 2002). Experiments performed with JA synthesis mutants showed that wound-induced systemic expression of PIs depends on JA production in rootstock tissues but not in distal leaves of the scion (Li et al., 2002, 2005; Lee and Howe, 2003). The most straightforward interpretation of these results is that JA or another product of the octadecanoid pathway is a component of the transmissible wound signal for systemic PI expression (Schilmiller and Howe, 2005; Wasternack et al., 2006).
In theory, wound-induced production of volatile MeJA could provide a cell nonautonomous, airborne route for chemical communication between damaged and undamaged leaves of the same plant (Heil and Ton, 2008). However, experiments conducted with sentinel plants located in close proximity to the wounded plant showed that systemic expression of PIs in tomato is mediated by a signal traveling within the plant rather than a factor diffusing through the atmosphere (Farmer et al., 1992). In addition, wounding of Nicotiana attenuata leaves did not result in the release of significant quantities of MeJA (Von Dahl and Baldwin, 2004).
Recent studies in Arabidopsis implicate JA as a cell autonomous wound signal for rapid systemic changes in gene expression. Mechanical tissue damage inflicted on rosette leaves resulted in a large (~10-fold) and rapid (within 5 min) increase in JA and JA-Ile levels in undamaged leaves of the rosette (Glauser et al., 2008; Koo et al., 2009). The systemic increase in JA-Ile preceded the onset of expression of early JA-response genes, which was detected in systemic leaves within 15 min of wounding. As predicted by current models of JA signaling, these rapid changes in gene expression were spatially and temporally correlated with degradation of JAZ proteins (Koo et al., 2009). Wound-induced systemic increases in JA-Ile levels and turnover of JAZ proteins have also been reported in other recent studies (Suza and Staswick, 2008; Wang et al., 2008; Zhang and Turner, 2008).
Systemic accumulation of JA-Ile could result either from its transport between wounded to unwounded leaves or from de novo synthesis in undamaged leaves in response to a long-distance signal. A transgenic system in which conversion of OPDA to JA can be spatially manipulated by dexamethasone-inducible expression of the JA synthetic enzyme OPR3 was used to distinguish these possibilities (Koo et al., 2009). Wounding did not elicit systemic JA/JA-Ile accumulation in plants that had the ability to produce the hormone in damaged (local) leaves only. However, wounding did induce systemic production of JA-Ile in plants that were complemented for OPR3 function in the systemic undamaged leaves. Significantly, production of this systemic pool of JA-Ile occurred even in the absence of JA synthesis in the local damaged leaves. It was concluded that wound-induced systemic JA/JA-Ile accumulation in Arabidopsis results, at least in part, from de novo synthesis of these compounds in systemic leaves rather than transport from the site of wounding (Koo et al., 2009). Metabolic labeling experiments in N. attenuata have also provided evidence that JA-Ile is synthesized de novo in leaves distal to the wound site (Wang et al., 2008). These new findings are consistent with a model in which a rapidly transmitted wound signal activates systemic synthesis of JA which, upon conversion to JA-Ile, mediates expression of early response genes by the SCFCOI1/JAZ pathway (Fig. 2).
The speed of the wound signal responsible for triggering de novo JA/JA-Ile synthesis in systemic leaves was estimated to be less than 2 cm/min (Glauser et al., 2008; Koo et al., 2009). This rate of movement is at least 20-fold faster than the estimated speed of the phloem-borne signal for PI expression in tomato. A significant body of literature implicates electrical potentials and/or hydraulic forces as signals for a variety of wound-induced rapid systemic responses (Wildon et al., 1992; Malone, 1993, 1996; Seo et al., 1995; Herde et al., 1996; Stankovic and Davies, 1996, 1998; Davies et al., 1997; Coker et al., 2005; Beaubois et al., 2007; Zimmermann et al., 2009). In maize, hydrostatic pressure changes induced by wounding affect stomatal aperture in systemic leaves within seconds (Raschke, 1970). An abscisic acid-independent, rapid hydraulic signal plays a role in communicating root-to-shoot water status in tomato and Arabidopsis (Holbrook et al., 2002; Christmann et al., 2007). Based on studies performed in the tomato system, Malone proposed a hydraulic dispersal mechanism in which rapid wound-induced changes in hydraulic pressure generate a basipetal mass flow that delivers chemical elicitors to distal leaves via the xylem, in a time frame that can account for rapid (e.g., within 20 min of wounding) activation of gene expression (Malone, 1993; Malone and Alarcon, 1995). In accordance with the involvement of a rapid physical signal in systemic wound responses in Arabidopsis, simple excision of petioles of undamaged leaves was sufficient to cause rapid systemic changes in JA/JA-Ile levels and gene expression (Koo et al., 2009).
These recent studies indicate that all components of the JA pathway, ranging from the production of JA precursors to activation of JA-response genes, are actively engaged in systemic undamaged leaves of the Arabidopsis rosette. In referring to this mode of JA action as the cell autonomous pathway (Fig. 2), we assume for simplicity that systemic synthesis and subsequent action of JA-Ile occurs in the same responding cell. A potential clue to the mechanism by which de novo JA synthesis is activated in distal responding cells came from experiments showing that the steep rise in systemic JA-Ile levels correlated with a rapid decline in levels of free OPDA (Koo et al., 2009). Although a precursor-product relationship was not established, the magnitude of the decline in OPDA content during the first 5 min after wounding was sufficient to account for the systemic increase in JA-Ile. These findings suggest that, in Arabidopsis, OPDA metabolism may be a control point for JA/JA-Ile production in distal leaves.
The speed with which wounding triggers systemic changes in OPDA and JA levels suggests rapid mobilization of a JA precursor or a post-translational event that activates a pre-existing biosynthetic enzyme. Previous studies in tomato and tobacco have shown that leaf wounding produces a rapid xylem-borne signal that activates MAP kinases in systemic leaves (Seo et al., 1995; Stratmann and Ryan, 1997). Reversible phosphorylation is also implicated in the control of JA-dependent wound responses (Rojo et al., 1998; Seo et al., 1999; Kandoth et al., 2007; Wu et al., 2007). Potential points of post-translational regulation of OPDA metabolism include transport of OPDA into peroxisomes via the ATP-binding cassette transporter COMATOSE (Theodoulou et al., 2005) or peroxisomal matrix enzymes involved in the conversion of OPDA to JA. Structural studies showing that OPR3 forms a self-inhibited dimer provide evidence that the in vivo activity of OPR3 may be regulated by phosphorylation (Breithaupt et al., 2006).
The occurrence of distinct JA-dependent wound response pathways (Fig. 2), together with the well established but poorly understood phenomenon of JA-independent wound signaling (Leon et al., 2001; LeBrasseur et al., 2002; Howe, 2004), indicates that the mechanisms by which plants perceive and respond to tissue injury are highly complex. It seems clear that the final outcome of the plant wound response varies depending on the developmental stage of the plant, the specific nature of the threat, and the environmental conditions under which the threat is encountered. Multiple systemic signaling pathways may interact functionally with one another to optimize the plant’s response to different types of wound stress or during different stages of plant development. It is possible, for example, that rapid elicitation of systemic responses via the cell autonomous pathway primes undamaged tissues for enhanced resistance to secondary attacks. A similar model was recently proposed for priming of rhizobacteria-induced systemic resistance in Arabidopsis (Pozo et al., 2008). Rapidly induced systemic responses may also prime undamaged tissues for increased responsiveness to a slower moving, transmissible JA signal. Such a two-step mechanism may allow plants to better discriminate between real (e.g., insect attack) threats and false alarms, and to fine tune the systemic response to a particular type of aggressor. This view is consistent with recent studies (Wu et al., 2007) indicating that wound-induced systemic responses in N. attenuata involve multiple long-distance signaling systems that operate at different temporal and spatial scales. The involvement of multiple long-distance signals in controlling systemic PI expression in tomato may help to integrate opposing theories about the chemical versus physical nature of the systemic signal in this classic model system (Wildon et al., 1992; Malone and Alarcon, 1995; Ryan, 2000).
A wealth of information establishes JA as a wound hormone in land plants. Nevertheless, several major gaps in our understanding of how JA exerts its local and systemic effects in injured plants remain to be filled. Further characterization of rapid wound signals and elucidation of the mechanisms by which they are perceived in responding cells is centrally important. A related challenge is to determine how early signaling events at the cell surface are linked to activation of the JA biosynthetic pathway. Among the earliest biochemical events implicated in response to tissue damage are changes in plasma transmembrane potential and Ca+2 concentration (Maffei et al., 2007; Zimmermann et al., 2009; Knight, 1991; Lecourieux et al., 2006; Bonaventure et al., 2007), generation of reactive oxygen species (Orozco-Cardenas et al., 2001; Foyer and Noctor, 2005; Almagro et al., 2008), activation of mechanosensitive ion channels and cytoskeletal-linked stretch sensors (Kung, 2006; Nakagawa et al., 2007; Haswell et al., 2008; Chehab et al., 2009; Na et al., 2008), and post-translational protein modification (Seo et al., 1995; Stratmann and Ryan, 1997). It remains to be determined which of these processes are directly involved in activating JA synthesis in response to wounding. It will be equally important to determine the biochemical mechanisms involved in signal inactivation. Additional work is needed, for example, to determine whether enzymatic modification of JA-Ile (Miersch et al., 2008; Glauser et al., 2009), epimerization of (3R,7S)-JA-Ile (Farmer, 1994; Fonseca et al, 2009), production of stabilized JAZ proteins (Chung and Howe, 2009), or other mechanisms play a role in attenuating of wound responses.
Continuing research to discern the inner workings of the SCFCOI1/JAZ complex (Chung et al., 2009, this issue) will enhance our understanding of F-box proteins as intracellular sensors of wound stress. The identification of bioactive forms of the hormone (Thines et al., 2009; Fonseca et al., 2009) places increasing importance on efforts to quantify the endogenous level of these compounds, including specific stereoisomers, in healthy and stressed tissues. Because wound-induced JA signal output is determined largely by the intracellular concentration of JA-Ile (Koo et al., 2009), there is a clear need to increase our understanding of the cellular mechanisms of JA/JA-Ile homeostasis.
These mechanistic questions concerning JA-dependent wound responses should be complemented by parallel studies at the whole plant and ecological levels. Although it is firmly established that JA-mediated responses to tissue injury are critical for plant fitness in natural environments, the relevance of the cell autonomous and non-autonomous pathways of JA signaling in plant interactions with insects and pathogens is largely unknown. Identification of receptors for herbivore- and pathogen-derived signals that modulate the JA response will be a key part of this effort, as will elucidation of the molecular basis of cross talk between JA and other hormone signaling pathways. Undoubtedly, progress in all of these research areas will help to address the fundamental question of how plants recognize and respond to tissue injury.
Jasmonate research in our lab is supported by the National Institutes of Health (grant R01GM57795) and the Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, Office of Science, U.S. Department of Energy (grant DE—FG02–91ER20021).
Abraham J.K. Koo did his undergraduate studies in the Department of Biology of Korea University in South Korea. He received an MS degree in Molecular Biology and Genetics from the same department in 1999. His master’s thesis involved cloning and characterization of a maize mitochondrial translational elongation factor gene, tufM. He then moved to Dr. John Ohlrogge’s lab at Michigan State University to study lipid metabolism in higher plants, with a focus in the area of lipid trafficking. He earned his PhD degree in Plant Biology in 2004, and then joined Dr. Gregg Howe’s lab in the DOE-Plant Research Laboratory to study the fatty acid-derived hormone jasmonate (JA). He has identified OPCL1, which encodes a peroxisomal acyl activating enzyme involved in JA biosynthesis. His current research is focused on understanding how JA synthesis and signaling are regulated during biotic- and abiotic stress conditions.
Gregg Howe received his B.A. and M.S. degrees in Biology from East Carolina State University. After working in the plant biotechnology industry for two years, he pursued Ph.D. studies at the University of California (Los Angeles) on the role of metals in the assembly of the photosynthetic apparatus in Chlamydomonas reinhardtii. He then worked as an NIH Postdoctoral Fellow in Clarence Ryan’s lab at Washington State University. A major focus of this work was the identification and characterization of wound response mutants of tomato. In 1997, he joined the faculty of the DOE-Plant Research Lab and Department of Biochemistry and Molecular Biology at Michigan State University. Research in the Howe lab is currently focused on understanding the mechanisms of JA synthesis and perception, as well as the role of JA in wound responses and plant-insect interactions.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.