Glutathione-dependent phytohormone responses

Abstract

Cellular redox state is regulated by numerous components. The thiol-disulfide compound, glutathione, is considered to be one of the most significant, owing to its antioxidant power and potential influence over protein structure and function. While signaling roles for glutathione in plants have been suggested for several years, hard proof is scarce. Recently, through an approach based on genetic manipulation of glutathione in an oxidative stress background, we reported evidence that glutathione status is important to allow intracellular oxidation to activate pathogenesis-related phytohormone signaling pathways. This effect does not seem to be caused by changes in glutathione antioxidant capacity, and appears to be distinct to regulation through known players in pathogenesis responses, such as NPR1. Our data therefore suggest that new glutathione-dependent components that link oxidative stress to response outputs await discovery.

Numerous studies that date back three decades have established that increased oxidation is an important factor in plant responses to stress. Within recent years, the concept of oxidative stress signaling has become firmly embedded in the literature. According to this concept, increased production of oxidizing molecules such as reactive oxygen species (ROS), decreased activity of ROS-metabolizing pathways, or both, can activate a wide range of cellular responses, as reflected, for example, by specific changes in transcriptomic signatures. However, transcriptomic responses to ROS are highly dependent on the nature and location of the ROS as well as the physiological and nutritional context in which ROS availability increases.^1,2 Interactions between ROS produced intracellularly, where high rates of ROS can be produced by basal metabolism but where redox buffering is high, and reactions in the apoplast, where antioxidative systems are less powerful, are likely to be of particular importance.^3,4

Concepts of Oxidative Stress and Cellular Redox Buffering

A key issue is that “oxidative stress” remains a poorly defined concept from both biochemical and physiological points of view. Among other things, the term can be understood to refer to increased ROS production or ROS concentrations (neither of which are easily quantifiable), vague notions of shifts in “cellular redox state,” or changes in readouts such as irreversible modifications of protein or nucleotides, lipid peroxidation, induction of genes encoding antioxidative enzymes, or accumulation of oxidized forms of major soluble antioxidant pools.

Roles of ROS in signaling likely involve protein oxidation as a key early event in the signal transduction chain. It remains unclear whether, within the highly reducing environment of the plant cell, modifications of specific proteins can occur readily through their direct reaction with ROS or rather whether adjustments in cellular redox buffering capacity are required to create a micro-environment conducive to such modifications.

Cellular redox buffering is itself an ill-defined concept because it is achieved by several components that have different midpoint redox potentials [eg, ferredoxin, thioredoxin, NADP(H), glutathione and ascorbate] that may not be at or close to thermodynamic equilibrium.³ Among these potential intracellular buffers, glutathione is an important component existing in two principal forms (GSH and GSSG) that can interact with proteins or signaling metabolites in various ways.^5-7 Several studies of plants with genetically modified glutathione synthesis or metabolism provide evidence that glutathione plays roles in regulating plant responses to pathogens.^8-11 However, the interactions underlying the effects of glutathione on pathogenesis-associated pathways remain to be eludicated. It is unclear whether effects of glutathione are related mainly to its well described antioxidant function in ROS responses, or whether this antioxidant function is coupled to a signaling role. According to the second view, changes in glutathione status driven by ROS-dependent oxidation could be sensed in some way by the cell.

Direct Analysis of The Roles of Glutathione in Oxidative Stress-Triggered Phytohormone Signaling

To explore these questions, we have generated a range of Arabidopsis lines in which glutathione status is modified, either through increased H₂O₂ availability (cat2), decreased glutathione reductase activity (gr1), impaired glutathione synthesis (cad2), or a combination of these mutations. To examine interactions between oxidative stress, glutathione, and salicylic acid (SA) synthesis and signaling, double and triple mutants carrying the sid2 and npr1 mutations have also been generated (Fig. 1).

Figure 1. Genetic tools for elucidating the relationship between H₂O₂, glutathione, and salicylic acid (SA). (A) Scheme summarizing effects of cat2 mutation on redox state (glutathione oxidation and accumulation) and the SA synthesis and signaling pathways. Mutants are indicated in red. (B) Leaf glutathione contents in a selection of single and multiple mutants affected in H₂O₂ metabolism (cat2), glutathione reduction (gr1), glutathione synthesis (cad2) or SA synthesis (sid2). The cat2 oxidative stress background is indicated red. White bars, GSH. Black bars, GSSG. For further information, see text.

When the cat2 mutant is grown in air, increased availability of photorespiratory H₂O₂ drives characteristic responses in glutathione and in signaling responses such as the SA-dependent pathway.^10,12 The cad2 and allelic mutations produce an effective block over the H₂O₂-triggered accumulation of GSSG that otherwise occurs in cat2 (Fig. 1B). While blocking upregulation of glutathione in this way does not affect key related redox pools in cat2, or the slowed growth which we take to be a phenotypic readout of stress intensity, it largely annuls several H₂O₂-triggered pathogenesis-related responses. First, the induction of lesion formation, pathogenesis-related gene expression and resistance to bacteria observed in cat2 are largely prevented in cat2 cad2, in which glutathione remains low and highly reduced.¹³ Direct comparison with the cat2 npr1 double mutant strongly suggests that this effect is not mediated via impaired function of the well-known thiol-regulated co-activator NPR1,¹⁴ but is rather linked to regulation of the SA synthesis pathway. Thus, whereas cat2-triggered SA accumulation is exacerbated in cat2 npr1, it is strongly damped by introducing the cad2 mutation into both cat2 and cat2 npr1 backgrounds.¹³

At the same time as damping the intensity of H₂O₂-triggered SA signaling, blocking glutathione accumulation strongly attenuates the expression of genes involved in jasmonic acid (JA) synthesis and signaling.¹⁵ Similar to its effects on SA signaling, the effect on JA-associated genes does not appear to be mediated through previously described interactions involving NPR1.¹⁶ Our observations implicate glutathione status as a potentially key factor in the regulation of early events that link oxidative stress to the activation of key defense phytohormone pathways.

Roles of glutathione status in the transmission of oxidative signals could involve changes brought about as glutathione is oxidized as part of its antioxidative function. If so, the close coupling of signaling and antioxidant functions is likely to make it difficult to separate the two effects. However, several of our observations suggest that the cad2 mutation exerts its effects on SA and JA signaling without strongly affecting the antioxidative function of glutathione.^13,15 Further, we reported that the phenotypic effects observed in cat2 cad2 are quite distinct from those observed in cat2 gr1,¹³ which shows a dramatic oxidative stress phenotype.¹⁰ Nevertheless, the two double mutants show some similarities in their transcriptome profiles. Transcript profiles triggered by the cat2 mutation are strongly modulated by growth day length.^2,10 Some genes are induced in cat2 in short days (SD) or in long days (LD), and may even show antagonistic responses in the two growth regimes.² When the effects of secondary cad2 or gr1 mutations on cat2 transcriptome signatures are examined, they both show a tendency to repress specific groups of genes. An example of one such gene cluster is shown in Figure 2A. Many of these genes are associated with phytohormone functions. JA-associated genes are the best represented, but other genes repressed by the secondary cad2 and gr1 mutants include several associated with abscisic acid. These repressive effects of modulating glutathione status are consistent with the analyses we recently reported of SA and JA marker genes in these genotypes grown in different conditions.^13,15

Figure 2. Gene expression profiles in cat2, cat2 gr1 and cat2 cad2. (A) Example of a day length-specific sub-cluster of cat2-sensitive genes. Genes show induction in short days (SD) and/or repression in long days (LD), and are enriched in jasmonic acid (JA)-associated genes (positions indicated beneath the heat map in blue). (B) Transcript levels of representative JA-associated genes in the different genotypes (relative to wild-type levels). *Indicates significant difference from wild-type at p < 0.05. Microarray analyses were performed as described in reference 10 and values are taken from a data set available at: http://urgv.evry.inra.fr/cgibin/projects/CATdb/consult_expce.pl?experiment_id=256.

It is interesting that cat2 gr1, in which GSSG accumulation is enhanced (Fig. 1B), shows similar effects on JA-associated gene expression to those observed in cat2 cad2, in which cat2-dependent accumulation of GSSG is abolished (Fig. 2). The common responses observed in the two double mutants may indicate that the key factor is glutathione redox potential rather than, for instance, the concentration of GSSG or GSH:GSSG ratios (for further discussion, see ref. 15). Indeed, an appropriate glutathione redox potential may be a requirement to achieve changes in thiol-disulfide status of proteins necessary to activate signaling. However, the details of the underlying mechanisms remain to be elucidated, and it is not yet clear whether similar responses in cat2 cad2 and cat2 gr1 are caused by common mechanisms or by different mechanisms that converge to produce the same effect.

A suite of genes associated with JA signaling were not only downregulated by cad2 in the cat2 background; they also showed lower basal expression in the cad2 single mutant, which shows no sign of oxidative stress.¹⁵ Some of these genes responded similarly in the gr1 single mutant which, like cad2, shows a wild-type phenotype,¹⁰ as well as the oxidative stress background, cat2. Since the glutathione redox potential is increased in both cad2 and gr1,^17,18 and may also be increased in the GSSG-accumulating cat2, this could be a further indication that this factor could be influential in the regulation of key signaling pathways within the plant. However, other possibilities remain to explain the effects of glutathione. Whatever the underlying mechanisms, it seems that glutathione status itself could form a modulating interface in the conversion of environmental signals into response outputs through phytohormone-dependent pathways.

Glossary

Abbreviations:

ABA

abscisic acid

GR

glutathione reductase

GSH

glutathione (reduced)

GSSG

glutathione disulfide

JA

jasmonic acid

NPR1

nonexpressor of pathogenesis related genes 1

ROS

reactive oxygen species

SA

salicylic acid

Han Y, Chaouch S, Mhamdi A, Queval G, Zechmann B, Noctor G. Han Y, et al Antioxid Redox Signal 2013; In press; PMID:23148658. Antioxid Redox Signal. 2013;18:2106–21. doi: 10.1089/ars.2012.5052.

Han Y, Mhamdi A, Chaouch S, Noctor G. Han Y, et al Plant Cell Environ 2012; In press; PMID:23210597. Plant Cell Environ. 2013;36:1135–46.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.