ROS and Mitogen-activated protein kinase (MAPK)
MAPK signaling pathways have been well accepted as a general signal transduction mechanism in eukaryotes. These pathways consist of three functionally linked protein kinases: MAPK, MAPK kinase (MAPKK) and MAPK kinase kinase (MAPKKK). By protein phosphorylation cascades from MAPKKK to MAPK, this system links different receptors to their cellular and nuclear targets (Tena et al., 2001). MAPK signaling cascades in Arabidopsis are complex, involving multiple isoforms of each MAPK gene and more than 20 pathways (Wrzaczek and Hirt, 2001; Ichimura et al., 2002). Many environmental constraints were found to stimulate MAPK cascades, like cold, heat, drought, wounding, and pathogens, while these cascades are also involved in phytohormone signaling (Bowler and Fluhr, 2000). In plants and eukaryotes in general, it has been demonstrated that the transmission of oxidative signals is controlled by protein phosphorylation involving MAPKs (Kyriakis and Avruch, 1996; Gustin et al., 1998; Pitzschke and Hirt, 2006; Xing et al., 2008). It was reported that H2O2 initiated MAPK signaling by activating ANP1, an isoform of MAPKKK. This effect led to the phosphorylation of MPK3 and MPK6 (Kovtun et al., 2000). Oxidative signal inducible 1 (OXI1), a protein kinase, is also involved in the activation of MPK3 and MPK6. Infection by virulent fungal pathogens caused HR in oxi1 null mutant, concomitantly the activation of MPK3 and MPK6 by oxidative stress was compromised (Rentel et al., 2004). Protein phosphorylation was also demonstrated to trigger Ca2+-dependent ROS production, which was mediated by NADPH oxidases in Arabidopsis. This study also suggested the existence of a positive feedback regulation of Ca2+ and ROS (Kimura et al., 2012).
ROS and phytohormones
Plant stress responses are highly integrated with hormone signaling to regulate biological processes. Salicylic acid (SA) signaling is associated with resistance to biotrophic pathogens as well as abiotic stress (Vlot et al., 2009). Intracellular and extracellular ROS accumulation can induce SA production, pathogenesis-related (PR) gene expression, and cell death (Chamnongpol et al., 1998; Torres et al., 2005; Chaouch et al., 2010). All these effects can be reverted in the cat2 mutant by blocking SA synthesis, showing that H2O2-triggered cell death and related responses are not a direct consequence of damage (Chaouch et al., 2010). Intriguingly, the defence responses triggered by increased peroxisomal availability of H2O2 in cat2 occur in a photoperiod-dependent manner: the PR responses such as lesions and SA accumulation appear when plants are grown in long days (LD) but not short days (SD) (Queval et al., 2007; Chaouch et al., 2010). ROS-induced SA accumulation can have several effects, notably including enhanced resistance to pathogens through the induction of PR proteins and phytoalexins. In addition, it can also promote stomatal closure, which may also contribute to defence against pathogens that enter plants by this route (Khokon et al., 2011). The application of SA can also induce ROS production via a peroxidase-catalyzed reaction, which also favors stomatal closure (Melotto et al., 2006; Miura et al., 2013). Unlike SA, JA and the related compound methyl jasmonate (MeJA) are involved in the response to necrotrophic pathogens and wounding (Devoto and Turner, 2005). Accumulating evidence reveals a strong relationship between ROS and JA signaling. For instance, ROS derived from NADPH oxidases are critical for JA-induced gene expression regulated by MYC2, a transcription factor involved in JA-mediated response (Maruta et al., 2011). It was revealed that JA signaling in response to intracellular oxidative stress requires an accompanying accumulation of glutathione (Han et al., 2013b). A complex relationship between SA and JA in physiological processes has been elucidated. Many studies show the opposite effect between JA signaling and SA-dependent pathways (Dangl and Jones, 2001; Spoel et al., 2003; Takahashi et al., 2004; Koorneef et al., 2008). However, the two pathways may interact positively and can also be induced together. For example, it was demonstrated that increased H2O2 levels in cat2 induced both SA and JA signaling pathways, and that both pathways are less induced when glutathione accumulation is genetically blocked by the cad2 mutation (Han et al., 2013b). These observations point to some glutathione-dependent signaling process in the link between H2O2 and induction of phytohormone pathways, an issue that will receive attention in Chapters 4 and 5 of this thesis.
As well as SA and JA, other phytohormones are also involved in stress responses. ABA is known to be induced together with ROS production in different environmental stress responses. It has been reported that ABA-induced stomatal closure during stress is mediated by ROS which is derived from NADPH oxidases (Kwak et al., 2003). Besides, ABA is found to be required for H2O2 production in chloroplasts, mitochondria and peroxisomes under water stress (Hu et al., 2006). Moreover, ABA has been considered to play a negative role in biotic stress signaling orchestrated by SA, JA and ethylene (Coego et al., 2005). Ethylene is well documented to be a key player in programmed cell death during senescence, ozone stress or pathogen infection (Orzaez and Granell, 1997; Lund et al., 1998; Overmyer et al., 2000). Gibberellins (GA), a cyclic diterpene compound with multiple functions in the plant life cycle, are linked with ROS through DELLA proteins which modulate transcript levels of antioxidant enzymes (Achard et al., 2008).
Other components involved in ROS signaling
In addition to the aspects mentioned above, there are also some other second messengers such as GTP-binding proteins (G proteins) and Ca2+ which mediate the ROS signals. The role of G proteins in stress responses, especially in plant-pathogen interactions, has been extensively reported (Assmann, 2005; Trusov et al., 2009; Maruta et al., 2015). It was revealed that under ozone stress, the first biphasic oxidative burst is greatly attenuated or absent in mutants lacking Gα protein or Gβ protein (Joo et al., 2005). AtRbohD and AtRbohF were suggested to receive initial signals from G proteins to mediate ozone responses in guard cells (Suharsono et al., 2002). Calcium signaling is involved in many signal transduction pathways. Elevations in cytosolic Ca2+ represent an early response to many different biotic and abiotic stresses (McAinsh and Pittman, 2009; Dodd et al., 2010). ABA and H2O2 treatments are among those that can increase cytosolic Ca2+ concentrations. Moreover, Ca2+ has been reported to work both upstream and downstream of ROS production in signaling pathways (Bowler and Fluhr, 2000; Abuharbeid et al., 2004; Monshausen et al., 2009; Sewelam et al., 2013).
The above discussion underlines the close integration of ROS-dependent redox signaling in numerous physiological processes, while also revealing that our knowledge of the details of redox signaling remains incomplete. Although many downstream responses to ROS have been described, the key redox events that lie at the heart of ROS-triggered signaling remain to be elucidated. A key player in enabling and regulating ROS signals is the plant antioxidative system. This system is complex and includes efficient enzymatic and non-enzymatic mechanisms that have been developed during evolution. Enzymatic systems consist of CAT, APX, SOD, various types of peroxiredoxins (PRX), glutathione/thioredoxin peroxidases (GPX) and glutathione S-transferases (GST). Non-enzymatic components include glutathione and ascorbate, which can be considered major components of a redox hub, but also tocopherol, carotenoids and phenolic compounds. Through the network of these mechanisms, plants regulate ROS accumulation and maintain redox homeostasis (Figure 1.4；Noctor et al., 2017).
CAT was the first discovered antioxidative enzyme. All known CAT forms in eukaryotes are haem-dependent. Among them two main types have been characterized, which are monofunctional CATs (also known as typical CATs) and bifunctional CAT-peroxidases (Zamocky et al., 2008). They can be distinguished by the affinity for H2O2 and the sensitivity to the inhibitor 3-amino-1,2,4-triazole (3-AT) (Margoliash and Novogrodsky, 1960; Regelsberger et al., 2002). Unlike monofunctional CATs, which exist in diverse organisms, the CAT-peroxidases are only found in some fungi and prokaryotes (Mutsuda et al., 1996; Regelsberger et al., 2002).
The typical CAT reaction is the dismutation of H2O2. The reaction is initiated by splitting the O-O bond of H2O2 to produce a molecule of H2O, as well as an oxy-ferryl enzyme intermediate (compound I) and a porphyrin cation radical. Then a second H2O2 is oxidized to O2. Simultaneously, compound I is reduced back to the initial state by releasing the bound O which is involved in the formation of the second molecule of water (Regelsberger et al., 2001; Alfonso-Prieto et al., 2009). Intriguingly, 15 monofunctional CATs can also catalyze some H2O2-dependent peroxidation of reducing substrates, making the functional division between the two types of CAT indistinct. During this process, the reduction of compound I is performed by the interaction with small compounds like ethanol instead of the second H2O2 (Zamocky et al., 2008). CAT-associated peroxidation has been reported in both mammalian and plants (Havir and McHale, 1989; Kirkman and Gaetani, 2007), but its physiological significance remains as yet unclear.
CATs are notably distinguished from other H2O2-metabolizing enzymes based on two factors. Firstly, when functioning as a dismutase, CAT function does not require reductant other than H2O2. Secondly, 16 CAT has high specificity for H2O2. So far three CAT genes have been identified in angiosperm species such as Arabidopsis, rice, tobacco, maize and pumpkin (Willekens et al., 1995; Frugoli et al., 1996; Guan and Scandalios, 1996; Esaka et al., 1997; Iwamoto et al., 2000). In Arabidopsis, CAT1 and CAT3 are located contiguously on chromosome 1 while CAT2 is located on chromosome 4 (Frugoli et al., 1996). According to the classification system based on tobacco genes and first proposed by Willekens et al. (1995), Class I CATs are mainly expressed in photosynthetic tissues, Class II CATs are associated with vascular tissues, while Class III CATs are notably expressed in seeds and reproductive tissues. For Arabidopsis, CAT1, CAT2 and CAT3 correspond to Class III, Class I and Class II, respectively (Table 1.1). The CAT1 gene is mainly expressed in pollens and seeds and at very low levels in leaves, CAT2 is expressed in photosynthetic tissues but also in roots and seeds, while CAT3 is associated with vascular tissues. It has been reported that compared to CAT1, the transcript levels of CAT2 and CAT3 are much higher in mature Arabidopsis rosettes (Frugoli et al., 1996; McClung, 1997). In addition, CAT2 transcripts show a day-night rhythm with a peak expression at the night/day transition, which is opposite to that of CAT3 (Zhong et al., 1994; Mhamdi et al., 2010a). Studies on subcellular localization showed that CATs exist mainly in peroxisomes in plants. Evidence supporting this conclusion includes the high CAT activity in peroxisomes and the identification of an import mechanism due to a Peroxisomal Targeting Sequence 1 (PTS1) pathway (Mullen et al., 1997).
Table 1.1. Probable classification of the three CATs found in different plant species (taken from Mhamdi et al., 2010a)
Previous studies on gene-specific T-DNA insertion mutants showed that CAT2 encodes the major leaf CAT isoform and this enzyme makes the major contribution to leaf CAT activity (Queval et al., 2007; Mhamdi et al., 2010a). In cat2 knock-out mutants, the CAT activity in rosettes is decreased by around 90% compared with that in wild type, and root CAT activity is also attenuated, though less severely (Bueso et al., 2007). In cat3, there is around 20% activity lost. In contrast, the effect of the cat1 mutation on leaf CAT activity is negligible. This is consistent with its relatively low transcript levels in leaves. To date, therefore, it seems that CAT2 and CAT3 may play the major roles in Arabidopsis rosette tissue. However, the relative contribution of these CAT isoforms changes according to the developmental stage of the plants (Zimmermann et al., 2006). For instance, as a senescence-associated gene, CAT3 transcripts increase with leaf age.
Because of the close association with photorespiration, Arabidopsis CAT2-deficient mutants (cat2) have been extensively used in various studies of plant development and physiology. Under conditions where photorespiratory H2O2 production is highly active, CAT function deficiency leads to increased availability of endogenous H2O2, with effects on cell redox state. The intracellular H2O2 signal can be modulated by changing photorespiration-related growth conditions, such as CO2 concentration and irradiance (Queval et al., 2007). When grown in air (400 µL CO2 L-1) with a moderate irradiance of 200 μmol m-2 s-1 at the surface of leaves, cat2 shows a dwarf phenotype accompanied by redox perturbation, evidenced by the decreases in the GSH:GSSG ratio and increases in total glutathione. By contrast, it has a wild-type phenotype when grown in high CO2 (hCO2; 3000 µL L-1) or if the irradiance is lower than 50 μmol m-2 s-1, both conditions in which photorespiration is rather inactive.
Among the most obvious phenotypes of cat2 grown in air at moderate light intensity are necrotic lesions. This phenotype is photoperiod-dependent. When grown in a 16 h photoperiod (LD), cat2 shows spreading necrotic lesions. By contrast, no lesions are observed in plants grown in an 8 h photoperiod (SD), even though the dwarf phenotype and redox perturbation are as marked as in LD (Queval et al., 2007, 2009). The lesion appearance in cat2 bears a striking resemblance to the HR, and includes SA accumulation, PR genes induction (e.g. PR1 and PR2), activation of camalexin and its synthesis pathway, and induced resistance to bacterial challenge. As well as necrotic lesions, all these effects are absent in SD. By the combination of pharmacological and genetic approaches, it is revealed that exogenous SA treatment could induce these responses in SD and revert it in LD when SA synthesis is genetically blocked. All these observations show that SA is involved in the peroxisomal H2O2-triggered HR-like lesion formation in cat2 grown in LD air condition (Chaouch et al., 2010).
Because of their influence on cell redox state, peroxisomal CATs act as regulators to fine-tune redox signaling, which is also a function of the carbon flux through photorespiration. This makes cat2 mutants an interesting model system to investigate the possible relationship between photorespiration and other metabolic, transcriptional and physiological processes, especially those possessing a redox component. Reductive pathways for H2O2 processing appear to compensate quite rapidly when CAT is deficient. Even though glutathione redox status is perturbed within hours after the onset of photorespiratory H2O2 production in cat2, little or no change is found in the redox states of glutathione-associated redox compounds, like ascorbate/dehydroascorbate or NADPH/NADP+ (Queval et al., 2007; Mhamdi et al., 2010b,c). Glutathione status appears to play an active part in the H2O2-triggered signal transduction provoked by CAT deficiency. For example, the GR knockout line gr1, which shows qualitatively similar changes in glutathione to those observed in cat2, has a wild-type phenotype, but nevertheless shows gene expression patterns that partly recapitulate those observed in cat2. Further, when the gr1 mutation is introduced into the cat2 background, H2O2-associated transcript profiles are significantly affected, suggesting that glutathione plays a role in transmitting H2O2-induced signals (Mhamdi et al., 2010b). Further evidence for this conclusion was obtained by analysis of cat2 lines in which glutathione accumulation was genetically impaired (Han et al., 2013a,b).
Extensive evidence reveals that phytohormones are important in determining cat2 phenotypes. Transcriptome analysis of cat2 roots points to the modulation of ethylene and auxin signaling (Bueso et al., 2007). Another study showed that the decreased CAT activity in cat2 induced effects on ABA signaling (Jannat et al., 2011). Glutathione was shown to be a key player in linking H2O2 to SA signaling (Han et al., 2013a). It was also reported that CAT2 participates in SA-mediated repression of auxin accumulation and JA biosynthesis during pathogen infection (Yuan et al., 2017). Another study reported that CAT2 and APX1 work in a coordinated way during the nuclear DNA damage response (DDR). Compared with the cat2 single mutant, the apx1 cat2 double mutant is more tolerant to oxidative stress imposed by high light, heat and paraquat application, and genome-wide transcriptome analysis shows an activation of typical DDR hallmarks (Vanderauwera et al., 2011).
Ascorbate and glutathione in ROS metabolism
Many compounds function as effective antioxidants in cells by regulating ROS accumulation. Among the best studied players are the cellular redox buffers, ascorbate and glutathione. They are well accepted to be key in controlling concentrations of ROS, although both metabolites are involved in multiple physiological processes (Cobbett et al., 1998; Vernoux et al., 2000; Dowdle et al., 2007). Ascorbate and glutathione are distinguished from most other antioxidant small molecules by (1) specific enzyme systems (peroxidases) that couple them to H2O2 metabolism; (2) the relative stability of their oxidized forms; (3) recycling of oxidized forms to reduced compounds by high-capacity reductases and associated systems that depend on the key electron carrier, NAD(P)H. Based on these features, ascorbate and glutathione can effectively regulate cellular redox state by repeated redox cycling (Foyer and Noctor, 2011). The pools of these two metabolites are generally highly reduced (over 95%) within the cytosol, chloroplasts and mitochondria, while the oxidized forms accumulate in compartments lacking efficient redox-recycling mechanisms like the vacuole and the apoplast (Schwarzländer et al., 2008; Queval et al., 2011; Noctor et al., 2016).
Glutathione in plants
Glutathione (γ-glutamylcysteinylglycine) is the principal low-molecular-weight thiol in most cells. It is an essential metabolite with multiple functions in plant development, biosynthetic pathways, detoxification, antioxidant biochemistry, and redox homeostasis. The fundamental function of glutathione is in thiol-disulfide interactions, in which the interconversion of reduced glutathione (GSH) and GSSG allows an appropriate cell redox state to be achieved. The concentration of cellular glutathione is high, as is its reduction state in the absence of stress. Under optimal physiological conditions the average ratio of GSH:GSSG in tissues such as leaves is at least 20:1 (Mhamdi et al., 2010a; Han et al., 2013a). However, this value may be higher (e.g. cytosol) or lower (e.g. vacuole) in specific subcellular compartments (Meyer et al., 2007; Queval et al., 2011).
In plants glutathione is synthesized by two ATP-dependent steps which rely on the activity of γ-EC synthetase (γ-ECS) and glutathione synthetase (GSH-S) (Rennenberg, 1980; Meister, 1988; Noctor et al., 2002; Mullineaux and Rausch, 2005). Each synthetic enzyme is encoded by a single gene (May and Leaver, 1994; Ullman et al., 1996), and both are indispensable to plant growth and development. Knocking out GSH1, the gene encoding γ-ECS, leads to lethality at the embryo stage (Cairns et al., 2006) while knocking out GSH2, the gene encoding GSH-S, causes a seedling-lethal phenotype (Pasternak et al., 2008). In Arabidopsis γ-ECS is located in plastids (Wachter et al., 2005), while GSH-S is found in both chloroplasts and cytosol, with the latter compartment considered to contain the major form of GSH-S because of the higher abundance of the corresponding transcript (Wachter et al., 2005). Glutathione synthesis, which responds to factors such as oxidative stress or heavy metals, can be modulated by regulating γ-ECS expression and activity, or by changing cysteine availability (Strohm et al., 1995; Noctor et al., 1998; Creissen et al., 1999; Harms et al., 2000; Noji and Saito, 2002; Wirtz and Hell, 2007). Both GSH1 and GSH2 transcripts respond to phytohormone and heavy metals (Xiang and Oliver, 1998; Sung et al., 2009), as well as to light, drought and pathogens.
While glutathione synthesis is stress-responsive, degradation is also important to achieve appropriate levels. In Arabidopsis there are four different types of enzyme which have been implicated in glutathione degradation. These are phytochelatin synthase (PCS), γ-glutamyl transpeptidase (GGT), γ-glutamyl cyclotransferase (GGC), and 5-oxoprolinase (5-OPase). Most attention has thus far focused on GGT, which is considered to degrade mainly GSSG located in the vacuole or the apoplast (Masi et al., 2007; Ferretti et al., 2009; Ohkama-Ohtsu et al., 2011; Su et al., 2011).
Table of contents :
CHAPTER 1 GENERAL INTRODUCTION
General overview of energy conversion in plants
1.1 ROS in plants
1.1.1 ROS: definition
1.1.2 ROS generation
18.104.22.168 ROS generation and compartmentation
22.214.171.124 Stress and ROS
1.1.3 ROS signaling
126.96.36.199 ROS and photosynthesis
188.8.131.52 ROS and redox homeostasis
184.108.40.206 ROS and Mitogen activated protein kinase (MAPK) 12
220.127.116.11 ROS and photohormones
18.104.22.168 Other components involved in ROS signaling
1.2 ROS processing
1.2.2 Ascorbate and glutathione in ROS metabolism
22.214.171.124 Glutathione in plants
126.96.36.199 Ascorbate in plants
188.8.131.52 Ascorbate glutathione pathway
184.108.40.206 Other pathways of ROS processing
220.127.116.11 NADPH linked reaction in plants
18.104.22.168.1 Glucose 6 phosphate dehydrogenase
22.214.171.124.2 Isocitrate dehydrogenase
126.96.36.199.3 Nonphosphorylating glyceraldehyde 3 phosphate dehydrogenase
188.8.131.52.4 NADP malic enzyme
1.3 Lesion mimic mutants in plants
1.4 Arabidopsis: a model to aid quick progress in understanding plant function
1.5 Aims of the project
CHAPTER 2 THE FUNCTION OF SPECIFIC CATALASES IN PLANTS
2.2 Material and methods
2.2.1 Plant material
2.2.2 Growth conditions and sampling
2.2.3 Enzyme activities
2.2.4 Measurements of transcript abundance
2.2.5 Ascorbate, glutathione and ROS assays
2.3.1 CAT expression and activity in roots
2.3.2 Root and seed phenotypes in CAT mutants
2.3.3 CAT2 and CAT3 function in leaves
2.4.1 Decreased root growth is specific to cat2 and is a secondary effect
2.4.2 The enigmatic roles of non–photorespiratory CATs in Arabidopsis
2.4.3 Growth day length affects oxidative signaling independent of oxidative stress duration
CHAPTER 3 EFFECT OF A SPECIFIC ISOFORM OF GLUCOSE–6–PHOSPHATE DEHYDROGENASE ON H2O2–INDUCED SA SIGNALING: A GENETIC SCREEN FOR REVERTANT LINES
3.2.1 The evaluation of mutagenesis efficiency
3.2.2 The revertant screen
3.2.3 Backcross to cat2 g6pd5 and segregation analysis
3.2.4 Identification of causal mutations
CHAPTER 4 ANALYSIS OF THE ROLES OF MONODEHYDROASCORBATE REDUCTASES IN H2O2 METABOLISM USING GENE–SPECIFIC MUTANTS
4.2.1 Identification of mdhar single mutants
4.2.2 MDHAR transcripts in response to intracellular oxidative stress
4.2.3 Impact of mdhar mutations on cat2–triggered lesion formation and phytohormone signaling
4.2.4 Impact of the loss of MDHAR functions on leaf redox status
CHAPTER 5 GENERAL CONCLUSION AND PERSPECTIVES
5.1.1 The function of specific CAT isoforms
5.1.2 Effect of G6PD5 on the H2O2–induced SA signaling pathway
5.1.3 Functions of specific MDHAR isoforms in response to oxidative stress
5.2.1 Specificity of CAT functions and interactions between oxidative signaling and day length
5.2.2 Further analysis of revertant mutations that allow lesion formation in cat2 g6pd5
5.2.3 MDHAR isoforms in responses to H2O2
5.2.4 A functional link between G6PD5 and MDHAR2 in H2O2 signaling?
CHAPTER 6 MATERIALS AND METHODS
6.1 Plant materials and growth conditions
6.1.1 Plant materials
6.1.2 Growth and sampling
6.2.1 Phenotypic analysis and lesion quantification
6.2.2 DNA extraction and plant genotyping
6.2.3 RNA extraction and transcripts analysis
6.2.4 Antioxidative enzyme activity measurements
184.108.40.206 Activity assay
6.2.5 Metabolite analysis
220.127.116.11 Glutathione and ascorbate assay by plate reader
18.104.22.168.2 Glutathione analysis
22.214.171.124.3 Ascorbate analysis
126.96.36.199 Total SA assay by High Performance Liquid Chromatography (HPLC)
6.2.6 ROS visualization in roots
6.2.7 EMS screen
188.8.131.52 EMS mutagenesis
184.108.40.206 Phenotype screen
220.127.116.11 Backcross with cat2 g6pd5
18.104.22.168 Sample collection for sequencing
6.2.8 Nuclear DNA isolation for sequencing
6.2.9 Statistical analysis