Emerging Complexity in Reactive Oxygen Species Production and Signaling during the Response of Plants to Pathogens

Plants have evolved a complex immune system to perceive microbial pathogens and respond by producing defense compounds preventing infection. Defense hormones such as salicylic acid (SA), jasmonates, and ethylene are key signals regulating the production of antimicrobial defenses. Moreover, other hormone pathways have critical actions by controlling responses to pathogen attack such as distribution of resources, cell death, water stress, or plant architecture. A fine-tuning regulation of these pathways through complex regulatory networks is necessary to achieve resistance against different pathogen classes (López et al., 2008; Grant and Jones, 2009).

Plants activate two forms of immunity by recognition of distinct pathogen molecules. A first and rapid response, known as basal resistance (microbe-associated molecular pattern-triggered immunity or MTI), is triggered after recognition of conserved microbial molecules (microbe-associated molecular patterns) by extracellular plant receptors (pattern recognition receptors; Boller and Felix, 2009). Second, effector-triggered immunity (ETI) is activated by resistance (R)-gene products (largely inside the cell) after recognition of specific effectors molecules (delivered into the plant cell by pathogens) and is commonly accompanied by a hypersensitive reaction (HR) involving localized host cell death at the point of infection (Jones and Dangl, 2006).

Oxidative burst involving production of reactive oxygen species (ROS) is a nearly ubiquitous response of plants to pathogen attack and has a key role in both MTI and ETI signaling (Torres et al., 2006). ROS activation is likely a primary consequence of the damage produced during the course of infection. However, whereas overaccumulation of ROS might enhance plant susceptibility (Govrin and Levine, 2000; Kariola et al., 2005) or cause an uncontrolled defense with spreading cell death lesions that can kill the plant (Lorrain et al., 2003; Moeder and Yoshioka, 2008), a tight regulation over ROS production and elimination through enzymatic and nonenzymatic antioxidants has allowed plants to use these reactive compounds as a critical feature of MTI and ETI (Torres et al., 2006). Reported defense responses associated with the production of ROS include direct killing of pathogens, activation of host cell death (HR), and contribution to cell wall strengthening (Bolwell and Daudi, 2009). Moreover, emerging data highlight the role of ROS as signals in MTI and ETI (Torres et al., 2006; Van Breusegem et al., 2008) as well as their contribution to provide an appropriate redox environment needed to activate defense (Tada et al., 2008). The importance of ROS in plant defense will be discussed here with a focus on the production of distinct types of ROS and on the cellular compartments involved in their production.

ESTABLISHED ROLE FOR SUPEROXIDE IONS AND HYDROGEN PEROXIDE IN RESPONSE TO PATHOGEN ATTACK

Two distinct reactions can convert ground state oxygen into different types of ROS during an oxidative burst (Fig. 1). Thus, dioxygen can be stepwise reduced by electron transfer to superoxide ion (O₂⁻) and hydrogen peroxide (H₂O₂), and the later can produce the hydroxyl radical (OH^•). Alternatively, dioxygen can be converted by energy transfer to singlet oxygen (¹O₂), a highly reactive short-lived product (half-life, approximately 200 ns) with a strong oxidizing potential (Apel and Hirt, 2004). Production of ROS occurs at different cellular locations in response to distinct environmental cues, and both the scavenging mechanisms and signaling events triggered by O₂⁻H₂O₂, and ¹O₂ have been investigated (op den Camp et al., 2003; Mittler et al., 2004; Gadjev et al., 2006). Results from transcriptomic analyses define common and specific responses toward different types of ROS as well as cross talk between distinct ROS signaling pathways, pointing to a complex scenario in which a fine regulation is critical for plant survival (op den Camp et al., 2003; Gadjev et al., 2006; Laloi et al., 2007).

Figure 1.

Production of ROS in plant defense. A, Generation of distinct ROS during oxidative burst. B, Tight regulation of ROS production at different cellular compartments and cell types is critical for plant survival. C, Callose accumulation after production of ¹O₂, O₂⁻H₂O₂, and specific ¹O₂-derived hydroxy fatty acids. Aniline blue staining in Arabidopsis leaves of 4-week-old plants (square-shaped sections) and in roots of in vitro-grown seedlings (rectangular sections). Leaves were infiltrated with Rose Bengal (1 μm as ¹O₂ producer), xanthine-xanthine oxidase (2 mm-0.1 units per mL as extracellular O₂⁻ generator), H₂O₂ (1 mm), 12-HOD (25 μm), 10-HOD (25 μm), and 10-HOT (25 μm). Roots are from 8-d-old seedlings grown in Murashige and Skoog medium and covered with a solution of the ROS inducers xanthine-xanthine oxidase and H₂O₂, at the above concentrations, or germinated 4 d in Murashige and Skoog medium and then transferred to a fresh medium containing Rose Bengal (100 nm), or 12-HOD, 10-HOD, and 10-HOT (10 μm). Representative examples of 24 h treated tissues are shown in all cases. Scale bars = 50 μm.

Research examining the role of ROS in plant defense has been focused on the actions of O₂⁻H₂O₂, whereas other ROS such as OH^• and ¹O₂ have been far less examined (Apel and Hirt, 2004). Different enzymes have been implicated in the generation of apoplastic ROS in plant defense, among which NADPH oxidases (also known as respiratory burst oxidases or Rbohs), similar to those present in mammalian neutrophils, have received most attention. Plant NADPH oxidases catalyze the formation of superoxide by the following reaction: NADPH + 2O₂ = NADP⁺ + H⁺ + 2O₂⁻. Secondary spontaneous or superoxide dismutase-catalyzed conversion of superoxide provides H₂O₂, which in turn can afford OH^• in the presence of transition metal ions such as Fe²⁺ or Cu⁺. Genetic analysis demonstrated that reduction or lack of RbohD and RbohF leads to elimination of extracellular H₂O₂ (Torres et al., 2002). However, reduced production of O₂⁻of its dismutation product H₂O₂ exerts different effects in plant pathogen growth and HR cell death, which suggests that apoplastic ROS might interact with distinct signaling pathways to serve different purposes. Thus, the spreading lesion phenotype of the lsd1 mutants (for lesion stimulating disease) is enhanced in the triple mutant lsd1-rbohD-rbohF, which has led to propose the role of RbohD and RbohF in limiting SA-elicited cell death in cells surrounding an infection site (Torres et al., 2005).

In addition to the apoplast, evidence for a role of chloroplast, peroxisomes, or mitochondria in ROS production has been reported (Van Breusegem et al., 2008). Moreover, recent studies identified other cellular sites such as endoplasmic reticulum, endomembranes vesicles, and nuclei as producers of ROS during pathogen responses, although the actions of ROS from these cellular locations remain mostly unknown (Ashtamker et al., 2007). The participation of chloroplasts in pathogen responses is concluded by results showing that light is required to activate defense gene expression and HR (Karpinski et al., 2003) and that the light-growth conditions might affect the formation of infection-like lesions in a number of Arabidopsis (Arabidopsis thaliana) mutants (Lorrain et al., 2003; Moeder and Yoshioka, 2008). In many cases, these phenotypes correlate with a failure in the photosynthetic machinery or in the mechanisms protecting cells against oxidative damage, including the process of photorespiration that mitigates photooxidative damage and requires the participation of peroxisomes and mitochondria (Moreno et al., 2005; Queval et al., 2007).

The role of Enhanced Disease Susceptibility1 (EDS1) as a master regulatory protein that coordinates defense by processing chloroplastic ROS-derived signals has been shown (Straus et al., 2010). However, chloroplastic ROS production and plant defense can be uncoupled. A recent example is the demonstration that chloroplast-derived ROS are essential for the formation of HR cell death but not for the activation of other basal defense responses in tobacco (Nicotiana tabacum) transgenic plants (Zurbriggen et al., 2009). Also, mutation of the chloroplastic Resistance to Phytopthora1 protein in Arabidopsis led to reduced H₂O₂ and enhanced susceptibility to Phytopthora brassicae, but caused a rapid run away cell death that originated at the point of infection (Belhaj et al., 2009).

Like chloroplasts, the mitochondria can also be an important source of ROS during physiological or pathological conditions that possess an efficient antioxidant machinery to control their toxic effects (Apel and Hirt, 2004). Nevertheless, the role of these organelles in plant cell death and pathogen responses has received little attention. Several studies revealed that treatments with cell death inducers, such as bacterial elicitors or virulence effectors might disrupt the functionality of the mitochondria and increase basal levels of ROS (Balandin and Castresana, 2002; Yao et al., 2004; Block et al., 2010). These results suggest that mitochondrial disturbance is a broadly employed strategy by pathogens to suppress host immunity and that increased ROS may contribute to the protection of plants against pathogen damage.

Results described above indicate that plants have evolved sophisticated mechanisms to use the compartmentalized production of ROS in the modulation of the defense responses against pathogen attack.

EMERGING ROLES FOR ¹O₂ AND LIPID PEROXIDATION IN PLANT DEFENSE

Information on the role of ¹O₂ in plant defense is still very limited. However, direct and indirect evidence discussed below, is starting to disclose a signaling role of ¹O₂ and its participation in the response to pathogens, an area that can be expected to receive much attention in the near future. ¹O₂ is a highly reactive unstable molecule produced in plants under basal and light stress conditions (Triantaphylidès and Havaux, 2009). In the chloroplast, excited chlorophyll can act as a photosensitizer to produce ¹O₂ from ground state oxygen. In addition, secondary metabolites such as phenaleno-like phytoalexins and phytoanticipins might act as photosensitizers to generate ¹O₂ following absorption of light energy (Flors and Nonell, 2006). Increased levels of these metabolites after pathogen attack could thus contribute to generate ¹O₂ as a product of the plant defense machinery.

¹O₂ has a crucial role during acclimation of plants to high light intensity and photooxidative stress (Triantaphylidès et al., 2008), a response that shows strong similarities to plant defense, including the functional integration of SA and of defense regulatory proteins such as LSD1, EDS1, and Phytoalexin Deficient4 (Mühlenbock et al., 2008). Of great interest, studies with the conditional flu mutant that generates ¹O₂ upon light illumination (op den Camp et al., 2003) allow to distinguish two modes of ¹O₂ activity. Thus, whereas high ¹O₂ production leads to photooxidative damage, decreased levels mediate a signaling activity, two responses that could be executed by ¹O₂ or by more stable ¹O₂-dependent products (Przybyla et al., 2008).

A universal response of plants to pathogen attack is the generation of a host of active lipid derivatives, collectively known as oxylipins (Andreou et al., 2009; Mosblech et al., 2009). Such compounds can be formed either by enzymatic or nonenzymatic peroxidation of fatty acids, however, certain of the hydroxy oxylipins, i.e. linoleic acid-derived 10-hydroxy-octadecadienoic acid (10-HOD) and 12-HOD and linolenic acid-derived 10-hydroxy-octadecatrienoic acid (10-HOT) and 15-HOT, can only be formed by ¹O₂-dependent nonenzymatic oxygenation and can therefore be used as in vivo markers of ¹O₂ generation (Przybyla et al., 2008). Whereas oxylipins formed by both specific enzymatical pathways (Hamberg et al., 2005; Kachroo and Kachroo, 2009) and by nonenzymatical free-radical reactions (Loeffler et al., 2005) play important roles in plant defense, no function has yet been assigned to the ¹O₂-derived hydroxy fatty acids. Of interest, these latter compounds accumulate in etiolated flu seedlings following illumination (Przybyla et al., 2008) and in leaves of Arabidopsis responding to Pseudomonas syringae pv tomato inoculation (Grun et al., 2007), reflecting the generation of ¹O₂ during stress responses, photooxidation, and pathogen attack.

Deposition of callose is a frequent response of cells to pathogen assault (Hématy et al., 2009). Importantly, production of ¹O₂ (triggered by Rose Bengal) and application of ¹O₂-formed hydroxy acids, induce a strong accumulation of callose in leaves (¹O₂ and 12-HOT) and roots (¹O₂, 10-HOD, and 12-HOD) of Arabidopsis (Fig. 1). Callose deposition was also observed (preferentially in roots) after application of O₂⁻ (generated by xanthine-xanthine oxidase) and was only weakly detected in roots of seedlings responding to H₂O₂ (Fig. 1). These results suggest that ROS such as ¹O₂ and O₂⁻ might contribute to the accumulation of callose during the response of plants to pathogen attack. Of interest, the differences in the pattern of callose deposition observed after generation of ¹O₂ and O₂⁻ or application of distinct ¹O₂-derived hydroxy fatty acids point to tissue-specific variations in the mode of action to these compounds. Further support of the participation of ¹O₂ in plant defense comes from results showing an overrepresentation of biotic stress-related genes during the transcriptomic reprogramming activated after generation of ¹O₂ (M. Martínez and C. Castresana, unpublished data).

Although these new observations deserve further investigation, these results are indicative of an active response of plants toward specific ¹O₂-derived hydroxy fatty acids. Related to this, we note that the nonenzymatic oxidation of linolenic acid contributes to limit pathogen infection and spreading cell death and that the action of linolenic acid as a sink for ROS has been suggested (Mène-Saffrané et al., 2009). Also, in line with our discussion, we speculate that the ¹O₂-derived hydroxy fatty acids could play a role in oxidative stress signaling and actively contribute to protect plant tissues against pathogen attack.

CONCLUDING REMARKS

Results described above reveal that plants have evolved unique defense responses that depend on ROS production and redox signals generated by different mechanisms at specific cellular locations. Tight control over production and accumulation of ROS is likely to be crucial to plants grown in natural environments where survival to pathogen attack and acclimation to prevailing abiotic stress factors (like high light) have to be integrated. However, key aspect of ROS production and signaling remain still poorly understood. Plants generate chemically distinct oxygen derivatives, which may be selectively produced at specific cellular locations in response to different environmental stresses. Studies to investigate the actions of ROS have revealed common and specific responses toward different types of ROS as well as cross talk between distinct ROS signaling pathways, thus showing a complex scenario hampering the investigation of selective actions by a given ROS. Major efforts in examining the actions of O₂⁻H₂O₂ might have oversimplified the analyses of ROS in plant defense. Compared to other ROS, ¹O₂ has received little attention, and recent studies indicating its participation in plant defense are emerging. Polyunsaturated fatty acids are a preferred target of ¹O₂ attack and several of its oxidation products could act as secondary messengers to trigger defense responses. Studies on the actions of ROS will benefit from newly developed tools helping to monitor in a noninvasive manner, the generation and signaling events of distinct types of ROS in the activation of resistance after pathogen attack.