Skip to main content

Specificity of H2O2 signaling in leaf senescence: is the ratio of H2O2 contents in different cellular compartments sensed in Arabidopsis plants?


Leaf senescence is an integral part of plant development and is driven by endogenous cues such as leaf or plant age. Developmental senescence aims to maximize the usage of carbon, nitrogen and mineral resources for growth and/or for the sake of the next generation. This requires efficient reallocation of the resources out of the senescing tissue into developing parts of the plant such as new leaves, fruits and seeds. However, premature senescence can be induced by severe and long-lasting biotic or abiotic stress conditions. It serves as an exit strategy to guarantee offspring in an unfavorable environment but is often combined with a trade-off in seed number and quality. In order to coordinate the very complex process of developmental senescence with environmental signals, highly organized networks and regulatory cues have to be in place. Reactive oxygen species, especially hydrogen peroxide (H2O2), are involved in senescence as well as in stress signaling. Here, we want to summarize the role of H2O2 as a signaling molecule in leaf senescence and shed more light on how specificity in signaling might be achieved. Altered hydrogen peroxide contents in specific compartments revealed a differential impact of H2O2 produced in different compartments. Arabidopsis lines with lower H2O2 levels in chloroplasts and cytoplasm point to the possibility that not the actual contents but the ratio between the two different compartments is sensed by the plant cells.


Senescence is a phase of plant development and becomes visible by the degreening of leaves, in which the photosynthetic apparatus is dismantled and chlorophyll is broken down, leading to light green and yellowish leaves [1]. However, when the light green color becomes apparent, the senescence program has been initiated long before, and changes at the molecular and physiological level have already been executed [2]. Plant cells must integrate a myriad of different signals driving onset and progression of senescence [3]. In the extreme scenario the already started program can be stopped or even reversed [1]. Even though senescence is often designated as the last step in the plant’s life history, it may occur from relatively early on throughout development. Older leaves in the lower canopy which are shaded by newly formed leaves or neighboring plants, or simply have reached a certain age, are sacrificed for the sake of newly developing organs, for example new leaves or stems. The aim of the senescence program is the remobilization of previously acquired resources such as nitrogen, carbon and mineral compounds out of the senescing tissues into developing parts of the plant. After remobilization has been executed, the senescent organs finally die and are shed. Before anthesis, this process is called sequential leaf senescence and aims at the repartitioning of nutrients from older leaves to newly developing, non-reproductive organs. In this case, the senescence program is restricted to a single organ. After anthesis, monocarpic senescence sets in and nutrients are now reallocated to the newly developing reproductive organs such as siliques and seeds. After this transition to reproductive growth during monocarpic senescence, potentially all leaves of a plant can now execute senescence, which finally leads to the death of the whole plant except for the newly formed seeds [4]. This systemic onset has to be coordinated between the different organs. However, how the local and systemic leaf senescence programs are discriminated is largely unknown. Even though a master regulator for leaf senescence was intensively searched for, it became clear that such a regulator does not exist. In contrast, multi-layered complex regulatory networks are in place to drive onset, progression and, in extreme cases, reversal of leaf senescence [3, 5]. For example, the transcription factor WRKY53 is just one of more than a hundred transcription factors driving the onset of senescence. Solely this transcription factor is tightly regulated at several levels. Not only is its expression responsive to ROS, jasmonic or salicylic acid (JA and SA) and influenced by more than 12 other transcription factors, also its activity is regulated by different molecular mechanisms, e.g. phosphorylation by MAPKKK1 or by changing interaction partners. On top of that, the degradation of the WRKY53 protein is controlled though a HECT ubiquitin ligase (UPL5) which exhibits an opposite expression pattern compared to WRKY53 itself. This ensures that even if the gene is mis-expressed for whatever reason, the protein content is still controlled by degradation and the protein will only be stable in the cells during onset of senescence [5]. To make the story even more complicated, WKRY53 also has a role in pathogen and wounding response. The overlap between senescence and pathogen response has been recognized in many instances due to similar signals which appear to be used in both processes. Recently, the cross-regulation network between leaf senescence and plant immunity, which is mediated by SA and ROS, has been nicely summarized [6].

Moreover, the senescence program adapts to diverse environmental conditions and requires very high plasticity. All kinds of stresses, abiotic as well as biotic, are sensed and constantly integrated into the regulatory networks, in which long-lasting unfavorable conditions have the potential to induce senescence prematurely [6, 7]. This serves as an exit strategy to guarantee viable offspring even under stressful conditions. As a tradeoff, seed quantity and quality are often diminished and can significantly lower productivity of plants. Consequently, in crop plants, stress-induced premature senescence can have a major negative impact on yield quantity and quality. Even though our knowledge on the physiological changes has increased enormously during the last decades and more and more signaling molecules acting in this process have been characterized, we are still far from understanding how this complex program is coordinated and how specificity is achieved in signaling. Almost all plant hormones are involved in regulation of this process as well as smaller signaling molecules such as calcium or ROS. All these components also have roles in other processes, so the question how specificity is achieved, especially in the case of ROS signaling, is not yet answered. How does the cell differentiate between an oxidative burst elicited by pathogen infection and the oxidative burst coinciding with the induction of senescence? In this review letter, we would like to shed light on the role of the ROS hydrogen peroxide in senescence signaling with a focus on the differences in signaling of H2O2 produced in different organelles. Even though hydrogen peroxide appears to be used as a signaling molecule in many different plant species, we focus on the findings in the model plant Arabidopsis thaliana.

General alterations in different organelles during senescence

In contrast to other cell death reactions, the long lasting and complex regulatory process to adequately execute leaf senescence requires continuous control by the nucleus as well as a constant energy supply. Consequently, not only is the nucleus kept intact until late stages of senescence, but also the mitochondria need to stay active to deliver the energy driving the process. Functionality of both organelles has to be guaranteed during onset and progression of senescence until finally both are also degraded in the terminal phase. In contrast, chloroplasts are massively restructured during progression of senescence and lose their photosynthetic function while peroxisomes are converted back to glyoxysomes with changed functional properties.


Chloroplasts are converted into “gerontoplasts” during the progression of leaf senescence. These gerontoplasts are morphologically characterized by an increased number of enlarged plastoglobuli, disorientation of the grana stacks and swelling of the thylakoids [1, 8]. These structural changes are combined with loss of function of the photosynthetic apparatus, thus allowing chlorophyll loss and photochemical capacity (FV/FM) to be used as parameters to describe senescence. As the majority of assimilated nutrients are stored in chloroplasts in the form of photosynthetic proteins, the remobilization of chloroplastic components is a central feature of senescence. It was estimated in the 1980s that more than 10 billion tonnes of Rubisco and 1 billion tonnes of chlorophyll are produced and degraded every year [9]. Even though removal of damaged proteins from chloroplasts takes place throughout plant development as part of the quality control of chloroplasts, chloroplast dismantling and degradation of chloroplastic proteins is strongly upregulated during senescence. Besides various types of chloroplastic proteases acting inside the chloroplasts, diverse extra-plastidic pathways mediate degradation of chloroplastic proteins. Intensive formation of different vesicles can be observed during senescence ranging from RCBs (Rubisco-containing bodies) and ATG8-interacting protein1-plasmid associated (ATI-PS) bodies, which depend on the autophagy machinery for further degradation, to SAVs (senescence-associated vacuoles) and CCVs (chloroplast vesiculation-containing vesicles) acting independently of autophagy [10, 11]. Moreover, whole chloroplasts can be degraded by “chlorophagy” [12] or by the 26S proteasome mediated by a cytosol-localized E3 ubiquitin ligase mainly targeting chloroplasts that over-accumulate singlet oxygen [13]. Degradation processes inside and outside the chloroplasts have to be coordinated and different pathways are induced at distinct time points during leaf senescence, e.g. chloroplast number is reduced at the later stages [11].


Peroxisomes consist of an amorphous matrix surrounded by a single membrane. Notably, the term “peroxisomes” originates from their high H2O2 content. A significant number of enzymatic systems capable of generating H2O2 are present in the peroxisomes, e.g. acyl CoA oxidase, glycolate oxidase, superoxide dismutase, urate oxidase and many others. The extensive production of H2O2 is counterbalance by the presence of large amounts of the H2O2 scavenging enzyme catalase. Besides being generated as a side product of various reactions, H2O2 itself has a critical role as a signal molecule in the crosstalk between peroxisomes and other organelles to coordinate cellular functions. Apart from the peroxisomal metabolism of ROS, reactive nitrogen species and reactive sulfur species with signaling functions are produced in peroxisomes [14]. Moreover, peroxisomes also have important roles in the biosynthesis of plant hormones, such as JA, indole-3-acetic acid (IAA) and SA, all of which influence leaf senescence in one way or another. Therefore, peroxisomes are important signal generators in the cell. Not only do peroxisomes undergo dynamic changes in metabolism and morphology, but also peroxisome abundance can be adjusted to environmental conditions. Moreover, the primary functions of plant peroxisomes vary with different developmental stages. Whereas peroxisomes in green leaves are mainly involved in photorespiration, peroxisomes in seeds harbor the glyoxylate cycle together with β-oxidation and are therefore called glyoxysomes. In senescent tissues, peroxisomes are converted into “gerontosomes” which resemble glyoxysomes in terms of their metabolic functions [14, 15].


Even though mitochondria exhibit a driving force in programmed cell death (PCD) in animal systems, in which the loss of the mitochondrial membrane integrity leads to the release of PCD elicitors such as cytochrome C, plant mitochondrial membranes are maintained until very late stages of senescence. Thus, it is unlikely that plant mitochondria will trigger plant senescence in a similar way as they do during animal PCD. In Arabidopsis, functionality of the mitochondria is preserved until the latest stages of leaf senescence, even though their number drops by 30%. In contrast to structural changes observed in chloroplasts, the ultrastructure of the remaining mitochondria appears to be unchanged, retaining clearly demarcated cristae and membrane boundaries throughout the process of senescence. Integrity of mitochondria as well as their ability to provide energy are maintained during leaf senescence to drive the energetically demanding reallocation of nutrients. Transcriptome analysis combined with metabolomic approaches revealed that mitochondrial metabolism is partially reorganized to support the selective catabolism of both amino acids and fatty acids. Thereby, mitochondria provide the carbon backbone essential for nitrogen remobilization, which is of utmost importance during senescence [16]. In parallel, the alternative respiration pathway is activated during senescence [17]. An important function of the alternative oxidase (AOX) is to prevent the formation of excess ROS. Hence, a low reduction status of the ubiquinone pool by oxidizing ubiquinol is ensured, and thus the electron flow is guaranteed [18].

The nucleus

The nucleus appears to be the command center for senescence coordination. Therefore, its overall structure and integrity are maintained to guarantee its function. However, onset and progression of this developmental process is driven by massive changes in gene expression; e.g. in Arabidopsis thaliana almost one quarter of all genes are differentially regulated during senescence [2]. While genes involved in degradation and mobilization of macromolecules are switched on, genes related to photosynthesis are turned off. A chronology of events has been determined by high-resolution temporal transcript profiling covering 22 time points of single leaves of a defined position within Arabidopsis rosettes [2]. In this chronology, activation of autophagy and transport as well as the response to ROS are prominent early events of leaf senescence. These massive changes in the transcriptome imply a central and important role for transcriptional regulators [19, 20]. Two transcription factor (TF) families, the WRKY and the NAC factors, which greatly expanded in the plant kingdom, appear to play outstanding roles in senescence regulation in Arabidopsis but also in other plant species [21, 22]. NAC proteins form one of the largest plant-specific TF super families, and its members control transcriptional reprogramming associated with many developmental processes including senescence [22, 23]. Recently, genome-wide dynamic modelling of transcriptional gene regulatory networks uncovered new and distinct pathways during the onset of Arabidopsis leaf senescence including miRNA action [24]. Besides the expansion in number, the WRKY family adopted a specific functional strategy of networking. Almost all Arabidopsis WRKY transcription factor genes are characterized by one or several binding motifs for WRKY factors, so called W-boxes, in their promoters, implying a complex WRKY-driven transcriptional network [25]. Moreover, complexity is increased even more by the fact that WRKY factors are also able to form homo- and heterodimers with different functional properties [26,27,28].

A long-lasting H2O2 increase acts as a signal in leaf senescence of Arabidopsis

In the absence of stress, developmental senescence is mainly driven by the actual age of the individual leaves and the age and developmental stage of the whole plant [29]; however, the mechanisms by which plants sense these parameters is still not well understood. The senescence program is synergistically as well as antagonistically influenced by almost all plant hormones [3, 20]. Moreover, during the last two decades, it became evident that ROS, and especially hydrogen peroxide, can function as signaling molecules in developmental as well as stress-induced senescence. On the one hand, ROS can have a direct impact on gene expression by altering the expression as well as the activity of transcription factors [2, 30,31,32,33], thereby modulating regulatory processes. On the other hand, ROS can oxidize more or less all kinds of macromolecules, influencing a variety of physiological changes. Less well-known signaling compounds such as melatonin are involved in postharvest senescence regulation by modulating the ROS homeostasis [34].

ROS are inevitable by-products of aerobic metabolism in all organisms. They are formed by energy transfer to or partial reduction of molecular oxygen (O2) producing either free radicals such as superoxide (O·2) and hydroxyl radicals (OH·) or non-radical but highly reactive molecules such as hydrogen peroxide (H2O2) and singlet oxygen (1O2), of which the latter is generated exclusively in the chloroplasts. Here, excess light energy results in an electron spin shift in the molecular orbitals, which renders 1O2 very unstable and highly reactive. Except H2O2, which has a half-life of milliseconds to seconds [35] and can traverse longer distances, including across plant cell membranes via aquaporins [36], all ROS have very short half-lives and relatively high oxidation potentials, and thus act close to their site of production. All ROS can act as signaling molecules but, when present in excess, are toxic for the cell. Therefore, a delicate balance between ROS production and scavenging needs to be established in all cellular compartments. OH· is the most reactive and thus most toxic ROS. Despite the apparent lack of an established system for the tight control of hydroxyl radicals, its function in multiple processes has been documented as e.g. ion flux in roots, germination, stomatal closure, or post-harvest senescence of fruits [37,38,39,40]. Most recently, Hancock and Russel [41] suggested a H2-dependent scavenging mechanism for OH·. However, the reactive and short-lived nature of OH· renders its study challenging. In any case, the cell should avoid the Fenton reaction between H2O2 and O·2 which produces OH· in the presence of metal, especially iron, ions [42]. The inherent reactivity of different ROS determines their reactivity towards specific sites in all kinds of proteins. Whereas due to its high oxidative potential OH··does not pick and choose, the superoxide anion (O·2) predominately attacks iron-sulfur ([Fe-S]) clusters while H2O2 targets Cys residues. Therefore, the subcellular colocalization of short-lived ROS and their respective targets might contribute to the ROS signaling specificity [43].

Similar to plant hormones, ROS participate in a wide range of developmental and stress-related signaling processes, leaving the question largely open, how specificity can be achieved, how the plant cells can differentiate signaling cues and elicit an appropriate response. So far, no clear signature, e.g. the amplitude, frequency or duration of Ca2+ spikes [44], can be linked to ROS or more specifically to H2O2 signaling. In contrast to calcium signaling, which is executed by storage and release of Ca2+ [45], ROS signaling is controlled by production and scavenging. In Arabidopsis, a network of at least 152 genes is involved in managing the ROS levels [46]. However, in contrast to e.g. the oxidative burst after pathogen infection or wounding lasting for minutes to hours [42], we observed a clear difference in the duration of the senescence-related H2O2 signal, lasting for more than a week during induction of monocarpic senescence, at least in Arabidopsis or oil-seed rape [47, 48]. Measurement of intracellular H2O2 contents in different leaves of a single Arabidopsis rosette indicated that the increase during onset of monocarpic senescence can be observed more prominently in the younger leaves (Fig. 1) [48], suggesting that sequential senescence might use different signals.

Fig. 1
figure 1

A schematic drawing of the development of the model plant Arabidopsis thaliana is shown at the top. At the transition from growth to reproduction (dotted grey line), senescence switches from sequential senescence which remobilizes nutrients from old to young leaves to monocarpic senescence in plants which started to bolt and flower. Now nutrients are remobilized from all rosette leaves to the now developing reproductive organs. During the time interval of bolting and flowering, intracellular hydrogen peroxide contents were measured and are indicated in arbitrary units of DCFDA fluorescence. H2O2 levels increased during transition to reproductive growth and the long-term increase at the transition point is indicated by the orange bar. H2O2 concentration showed more pronounced elevations in young (dark red; pos. 8–12 within the rosette) and middle-aged leaves (red, pos.4–7 within the rosette) than in old leaves (orange, pos. 1–3 within the rosette). The age of the plants is indicated in weeks after sowing (w)

This long-term intracellular H2O2 increase at the onset of monocarpic senescence in Arabidopsis and oilseed rape is predominately due to sophisticated regulation of the activities of the H2O2 scavenging enzymes catalase (CAT) and ascorbate peroxidase (APX) [48, 49]. All subcellular compartments are equipped with their own enzymatic and non-enzymatic scavenging systems. While catalases are predominantly located in the peroxisomes, different APX isoforms are found in chloroplasts, mitochondria, the cytosol and the outer membrane of the peroxisomes. Catalases have a very low affinity for H2O2, with a Km of approx. 43 mM, but have a high reaction rate and are highly abundant, which compensates for this low affinity [50]. Under oxidative stress conditions, CAT can be retained in the cytosol to protect this compartment against H2O2-mediated redox changes and intensify defenses against oxidative damage [51]. In Arabidopsis, APX3 and 5 isoforms are associated with the outer peroxisomal membrane. This localization and the much higher affinity of APXs to H2O2 as compared to CAT (100 μM), could be responsible for the tight control of H2O2 leakage from peroxisomes to the cytosol [52,53,54,55]. Furthermore, APX1, 2 and 6 are present in the cytoplasm and two additional APX isoforms are associated with the stroma (sAPX) and the thylakoid membranes (tAPX) of the chloroplasts. Therefore, CAT and APX are positioned strategically to tightly control H2O2 concentrations and to ensure that H2O2 can act as a signaling molecule [56].

To allow for a long-term intracellular H2O2 increase at the onset of monocarpic senescence, in the first place CAT2 gene expression starts to decline at bolting time through inhibition of transcription by the bZIP transcription factor G-Box binding factor 1 (GBF1). This factor is most likely activated by phosphorylation through casein kinase II and translocation to the nucleus [48, 57,58,59]. As CAT2 protein has a high turn-over rate and accounts for approximately 80% of the total CAT activity in leaves [60], this rapidly leads to a decline in CAT2 enzyme activity and an increase in H2O2 concentration. In addition, APX1 activity also declines at the same time during bolting and flowering, but its activity is restored at later time points. In this case, gene expression is not downregulated but inhibition of activity appears to be carried out post-transcriptionally [61]. All in all, this leads to low protection against H2O2 in the cytosol as well as in the peroxisomes. Miyake and Asada [62] found that under low concentrations of ascorbate, hydrogen peroxide damages compound I of ascorbate peroxidase and inactivates enzyme activity. Furthermore, an experiment with suspension culture cells also showed that, paradoxically, APX activity can be inhibited by its own substrate H2O2. H2O2 treatment of whole plants revealed that this inhibition can only be induced during the time window of bolting and flowering, and exogenous application of H2O2 is only effective during this period [48]. Moreover, removal of the stem leads to restoration of APX activity, and subsequent re-bolting induces a second period of APX inactivation [63], clearly indicating a connection of the inhibitory mechanism to the bolting and flowering state. This means that APX is rendered sensitive towards hydrogen peroxide by a so far unknown mechanism during this specific period. Thus, at this time point, H2O2 concentrations further increase and create a positive feedback loop. After a certain time interval, inhibition of APX is released again and, in addition, CAT3 expression and enzyme activity start to increase. This regulatory cue results in a long-term hydrogen peroxide elevation as illustrated in Fig. 1. If downregulation of CAT2 expression and activity is abolished in gbf1 knock-out plants, APX activity is not inhibited by increasing H2O2 amounts, no long-term H2O2 increase is created and the onset of senescence is delayed [57]. This clearly indicates that this long-lasting H2O2 elevation acts as a signal to induce monocarpic senescence.

Furthermore, so far only scant attention has been dedicated to the fact that the circadian clock also triggers transcriptional regulation of ROS-responsive genes, ROS homeostasis, tolerance to oxidative stress [64,65,66] as well as senescence [67]. Not only do hydrogen peroxide production and scavenging display time-of-day phases but also mutations in the core-clock regulator, CIRCADIAN CLOCK ASSOCIATED 1, as well as mis-expression of components of the evening complex, EARLY FLOWERING 3, LUX ARRHYTHMO, and TIMING OF CAB EXPRESSION, affect ROS production [67]. This clearly indicates a global effect of the clock on the ROS network, and hence Sanchez and Kay [68] have suggested calling the circadian clock the “mastermind” of plant life. Interestingly, the period of the diurnal rhythm also changes during senescence, decreasing from 24 h in young plants to 22–23 h in old plants [69]. Therefore, time of the day should always be considered in future research.

Impact of H2O2 generated in different subcellular compartments in Arabidopsis leaves

In the last two decades it has become clear that the ROS concentrations in the cytoplasm and in different organelles have a different impact on senescence. Changing the H2O2 levels in different subcellular compartments has different effects on the senescence program ([49, 70], unpublished observations). Increasing H2O2 production in peroxisomes or in chloroplasts induced two different types of responses, one of which is independent and the other is dependent on the site of H2O2 production. While the independent response comprised the activation of more general oxidative stress response genes, an H2O2 increase in chloroplasts specifically induced early senescence signaling components, including transcription factors and biosynthetic genes involved in production of secondary signaling molecules [70]. In 2013, Rosenwasser et al. established a bioinformatic tool called “ROSMETER” for the identification of transcriptomic signatures related to ROS type and production site [71]. Interestingly, an unexpected but highly significant ROS transcriptome signature of mitochondrial stress was detected during the early stages of leaf senescence by the ROSMETER when the high-resolution transcriptome dataset of leaf senescence in Arabidopsis of Breeze et al. [2] was analyzed. This signature was defined by different reference plants: (i) rotenone treated plants and (ii) AOX1-T-DNA insertion lines under mild drought and light stress and (iii) antisense AOX1a plants. Rotenone treatment inhibits complex I of the respiratory electron transport chain and leads to induction of the AOX pathway and thereby to the reduction of ROS production [72, 73], while AOX1-T-DNA insertion as well as antisense AOX1a expression abolishes the alternative respiration pathway and would therefore increase mitochondrial ROS production, at least under stress conditions. However, there was a strong positive correlation of the early senescence expression profiles with rotenone treatment starting at 23 days after sowing and almost at the same time, from day 25 on, a negative correlation with AOX1-T-DNA insertion. This would mean that the transcriptome signature at early stages of leaf senescence would point to low mitochondrial ROS production. However, long-term antimycin A treatment which inhibits complex III of the respiratory chain and induces the AOX pathway or overexpression of AOX1a diminished ROS production but did not significantly alter senescence [74]. Therefore, the role of mitochondrial ROS in leaf senescence is still debatable. However, antimycin A also has effects on the cyclic electron flow in the chloroplasts [75].

Peroxisomes appear to be the most oxidized cellular organelles, with a redox potential of approximately − 360 mV [76]. Peroxisomes are one of the main sources of intracellular ROS generation. In photosynthetic tissue, photorespiration is the main source of H2O2 production in peroxisomes, contributing up to 70% to the total H2O2 production in plant cells [42]. In addition, photorespiration needs coordination of the chloroplasts, peroxisomes, and mitochondria in the cytosol, and often close contacts of these organelles can be observed, most likely to directly exchange substrates. Interestingly, at these contact sites H2O2 accumulation was observed inside the peroxisomes, also pointing to a role of H2O2 in organelle communication [77]. Using in vivo imaging with fluorescent proteins targeted to peroxisomes, rapid formation of tubular extensions called peroxules was observed in response to changes in ROS levels. Not only could these peroxules participate in the transfer of metabolites to mitochondria and chloroplasts, they could also be involved in ROS flux between these compartments, even though so far there is no direct evidence for this assumption [56]. In contrast, similar dynamic structures in chloroplasts called stromules can transfer H2O2 from chloroplasts to nuclei [78, 79]; however, to date no contact of peroxules to nuclei has been described. To counterbalance the large amounts of H2O2 generated in peroxisomes, 10–25% of the total peroxisomal proteins are catalases [80], in which CAT2 accounts for approximately 80% of the total catalase activity. As already mentioned above, CAT2 regulation is a central element of creating the long-lasting increase of intracellular H2O2 at bolting and flowering time, which is used as a senescence signal [48, 57]. Altering peroxisomal H2O2 by different means induces changes in gene expression [65, 66, 81, 82] that differ from those induced by chloroplast-derived H2O2 [70]. Thus, peroxisomal H2O2 clearly participates in retrograde signaling to the nucleus. However, the underlying molecular mechanisms and crosstalk with ROS from other compartments still have to be elucidated [56, 83]. How this specificity is produced downstream of the production is unclear. Thiol-based antioxidants such as peroxiredoxins, thioredoxins, and glutaredoxins are likely to be involved in ROS signaling cascades [84]. However, H2O2 can leak out of all compartments to the cytosol, so the cytosol most likely acts as a key site of redox signal integration. Thus, MAPK pathways in the cytoplasm are proposed to integrate ROS signals from different organelles to regulate the gene expression in the nucleus [85].

Like peroxisomes, plastids form stroma-filled tubular extensions surrounded by the envelope membrane called stromules. Stromules are dynamic structures that extend along actin microfilaments and the ER. In chloroplasts, stromule frequency increases during the day or in response to light-sensitive redox signals [86]. Interestingly, stromules can still form on isolated chloroplasts after extraction from the cytoplasm, pointing to the fact that stromule formation is plastid autonomous. In addition, plastid-derived vesicles are suggested to bud from stromules through tip shedding or simple breakage, either to recycle plastid content during nutrient stress, to remove toxic molecules, or for intracellular communication [87]. These stromules allow the plastid compartment to make direct contacts to other subcellular compartments such as other plastids, mitochondria or nuclei. Metabolites as well as proteins can flow through these connective tunnels. Moreover, using a nuclear-targeted H2O2 sensor protein (HyPER) [88], the translocation of ROS generated in chloroplasts into adjacent nuclei could be observed [89]. This suggests that stromules have a function in retrograde signaling in channeling and speeding up signal transduction directly to the nucleus, avoiding the scavenging systems of the cytoplasm [89]. Additionally, organellar H2O2 can oxidize thiol groups of specific proteins, thereby converting the ROS signal into thiol redox signals [90, 91].

In addition, changes in the apoplastic space appear to be involved in signaling and molecular trafficking during senescence [92]. Along with variations in the volume and pH of the apoplastic fluid, leaf senescence-related changes in the secretome included many proteins involved in stress responses and ROS metabolism, indicating that leaves of different ages control stress responses differentially to balance growth or survival at the whole plant level [93]. Moreover, production of apoplastic ROS can influence ROS generation in the chloroplast and both together can have an impact on gene expression in the nucleus [94].

On top of that, all these forms of retrograde signaling between the different organelles and the nucleus have to be integrated to finetune the final outcome of gene expression. We are far from understanding which molecular mechanisms underly this crosstalk. Recently, the protein RADICAL-INDUCED CELL DEATH1 (RCD1) has been characterized as a molecular component that allows a dialog between the retrograde signals of both mitochondria and chloroplasts. On one hand, RCD1 directly interacts and inhibits ANAC013 and ANAC017, two regulatory proteins that usually activate a set of genes involved in repair of mitochondria. On the other hand, the nuclear-localized RCD1 changes abundance, redox state and oligomerization in response to ROS generated in the chloroplasts. Interestingly, the interaction domain with the NAC factors was also necessary for the sensitivity to chloroplastic ROS [95].

The search for ROS receptors is far from complete and still has to be continued. It can be speculated that a plethora of ROS sensors is likely to exist in numerous compartments including the apoplast, cytosol, and organelles. In addition, the respective down-stream signaling components are yet to be identified. Plant cells may be able to decode specific ROS-dependent signatures by monitoring the relative intensity of redox stimuli at different locations. Such signatures could be part of the activation of appropriate responses [85].

In vivo imaging using fluorescent dyes and sensor proteins

Colorimetric methods using nitro blue tetrazolium (NBT) and/or diaminobenzidine (DAB) were initially used to detect ROS in cells, tissues, and organs. However, these methods were not suitable to analyze dynamic ROS levels in vivo. To solve this problem, a collection of different fluorescent dyes was established, in which different dyes exhibit different properties with regard to specificity and cell permeability, allowing in vivo imaging in real time. From among different fluorescent probes which were tested for in vivo imaging (DHE, H2DCFDA, H2HFF-OxyBURST, Amplex red, SOSG, and PO1), H2DCFDA displayed the highest signal-to-noise ratio [96]. One of the drawbacks of these dyes is that often the reactions between ROS and dye are not reversible and the dyes have to be reapplied. In parallel, transgenic approaches were developed using ROS-responsive promoters coupled to reporter genes [97, 98], which also allow one to measure ROS contents at the whole plant level and to follow up systemic ROS signaling. Moreover, protein-based fluorescent sensors have been developed and by now a variety of genetically encoded fluorescent redox probes are available [99]. For example, the redox-sensitive roGFP [100] or the H2O2 sensitive HyPER [88] can image more precisely the in vivo redox conditions and/or ROS concentrations in cells utilizing ratiometric measurements with two different wavelengths of the excitation laser light. These protein-based approaches also opened up the possibility to direct the proteinaceous sensors to different subcellular compartments via signal peptides or other localization signals. In vivo imaging using these sensor proteins can now be employed to resolve the spatiotemporal dynamics inside the cells and even in different subcellular compartments. Very recently the application of this technique could demonstrate that ROS locally generated in chloroplasts of intact Arabidopsis seedlings by methyl viologen treatment cause changes in H2O2 and glutathione redox potential in other subcellular compartments such as the cytosol and mitochondria [101]. However, given the nature of these sensors, one has to keep in mind that the expression of such sensor proteins cannot only be used to measure in vivo ROS concentrations but at the same time scavenge ROS and thereby also influence intracellular ROS concentrations. When using these sensor lines for long-term studies of ROS levels during senescence, we observed differences in production in different compartments but encountered the problem that all these lines showed altered senescence phenotypes compared to non-transgenic lines ([49], unpublished observations). Therefore, the use of these sensor proteins for in vivo imaging during senescence has its limitations.

However, these lines were still useful to characterize the role of ROS, especially H2O2 during senescence. For sensing especially H2O2 the fluorescent protein HyPER was constructed, which combines the regulatory domain of the Escherichia coli OxyR transcription factor for sensing H2O2 in bacteria with a circular permutated YFP [88]. The E. coli OxyR transcription factor contains an H2O2-sensitive regulatory domain (amino acids 80–310, OxyR-RD), and a DNA-binding domain (amino acids 1–79). The reduced form of OxyR-RD is oxidized preferentially by H2O2. [102]. Aslund et al. suggested that OxyR acts as a peroxidase as OxyR returned to its reduced form in a time course experiment and a significant amount of NADPH was consumed in in vitro experiments [103]. Moreover, canonical H2O2 measurements using H2DCFDA [49] as well as H2O2 treatments of HyPER expressing leaf material in a perfusion chamber directly under the confocal LSM revealed that these sensor proteins scavenge H2O2 (unpublished observations). This means that even though the reaction is reversible, so these proteins can function as sensors, HyPER expression in plants will change the amplitude of intracellular H2O2 signals by consuming H2O2. This is exactly what has been observed during onset of monocarpic senescence; the H2O2 signal in the HyPER expressing lines was dampened and senescence was affected [49]. Having shown this, we now used these lines no longer as sensor lines but as lines with altered intracellular ROS levels as described in the next paragraph.

Lowering H2O2 contents specifically in different subcellular compartments

Costa and coworkers created transgenic lines which express this sensor protein in defined cellular compartments [104]. Yet, by directing HyPER to different compartments and thereby scavenging H2O2 specifically in these compartments, we could observe that the senescence process is differentially affected in these transgenic lines: While a reduction in peroxisomal and cytosolic H2O2 levels led to a delay in the onset of leaf senescence, low chloroplastic H2O2 retarded progression of leaf senescence (Fig. 2) ([49], unpublished observations). Moreover, genome wide gene expression analyses of these plants uncovered severe differences in gene expression, which were also dependent on the developmental stage. Here, plants with low chloroplastic H2O2 show more severe and often even opposite changes in gene expression compared to plants with lower cytoplasmic H2O2, which appear to be more similar in gene expression to wild type plants. Strikingly, higher expression of CAT and APX genes in plants with low chloroplastic H2O2 was observed. Moreover, expression of defensin genes is increased under low chloroplastic H2O2 but decreased under low cytoplasmic H2O2 (unpublished observation). In wild type plants, the expression of the pathogen-related defensin genes and genes encoding H2O2 scavenging proteins CAT and APX is triggered by oxidative stress but not by low H2O2. In the same chain of evidence, formation of stromules clearly increased under low chloroplastic H2O2 (Fig. 3; unpublished observation), which, again, should be opposite and rise after SA or H2O2 treatment [80]. These opposing expression patterns together with the unexpected increase in stromule formation prompted us to speculate that not the actual H2O2 concentration is sensed by the plants but more likely the ratio between the cytoplasmic and the chloroplastic H2O2. This would mean that low chloroplastic H2O2 resembles “cytoplasmic” oxidative stress (Fig. 4). If our hypothesis of sensing the H2O2 ratio is correct, crossings between the two HyPer transgenic lines should be similar to wild type plants as now H2O2 would be low in both compartments. Presumably, defensin and CAT/APX gene expression, stromule formation, and senescence onset and progression will then be more reminiscent of wild type plants. However, how this ratio is sensed by the plants still has to be elucidated.

Fig. 2
figure 2

Changes of intracellular H2O2 contents in different compartments have different impacts on onset and progression of monocarpic senescence compared to wildtype Arabidopsis plants Columbia-0 (Col-0). The expression of the H2O2 sensor HyPER under the control of the 35S-CaMV promoter leads to constant and high expression of the HyPER protein and thereby scavenges H2O2. By directing HyPER to specific compartments, the H2O2 levels were specifically reduced there. While cytoplasmic and peroxisomal HyPER protein expression caused a delay in the onset of monocarpic senescence, chloroplastic HyPER expression retarded the progression of senescence

Fig. 3
figure 3

A An increase in the number of stromules formed by chloroplasts harboring the HyPER protein which scavenges H2O2 or a GFP which does not scavenge H2O2 (control plants) was observed. HyPER fluorescence (green) and autofluorescence of the chloroplasts (red) and a merge of these are presented. B The highest average number of stromules per chloroplasts in leaves of the chloro-35S:HyPER line and the control plants, which was counted over different developmental stages, is indicated. C Stromules are able to contact the nucleus shown in a merged microscopic image of transmitted light, HyPER fluorescence (green) and autofluorescence of the chloroplasts (red)

Fig. 4
figure 4

Model for H2O2 sensing. A Balanced H2O2 contents in wild type plants (WT) in normal situations without exogenous stress. B Imbalanced H2O2 contents between chloroplast and cytoplasm in two different situations: (1) wild type versus HyPER expression in chloroplasts (left) or (2) wild type under control conditions versus wild type under stress conditions (right). Total intracellular H2O2 contents were measured by H2DCFDA and expressed relative to wild type or wild type control conditions. In both cases, H2O2 content is imbalanced in the same way and high stromule numbers as well as expression of genes induced by oxidative stress are observed, indicating that the ratio between the contents of the cytoplasm and the chloroplasts might be the parameter which is sensed by the cells. This is still a hypothesis and has to be further analyzed, e.g. in crossings between plants expressing HyPER in the cytoplasm and in the chloroplasts

Future perspectives

Reactive oxygen species (ROS) are inevitable by-products of many reactions of aerobic organisms, are produced in all subcellular compartments and play pivotal roles in the communication between all plant organelles [35, 105,106,107]. However, how specificity is achieved in H2O2 or ROS signaling in general is still an open question. Direct connections between different compartments such as stromules or peroxules enable small signaling molecules as well proteins or metabolites to be directly exchanged without passing through the cytoplasm. ROS, especially H2O2, can also pass through these tunnels to avoid the scavenging system of the cytoplasm imposing directionality in signaling. However, how the cells differentiate between stress-induced ROS signals and developmental ROS signals, e.g. the senescence-inducing H2O2 increase, is still not well understood. Localization of ROS production, especially for short-lived ROS, as well as duration of the production, might be part of this differentiation. However, as no “master receptors” for ROS are known so far, and a plethora of possible targets for each ROS is available, the complex sensing process still has to be elucidated. Improvement of the sensor proteins [108] as well as of the microscopic technologies, e.g. single molecule or super resolution microscopy, will increase resolution and will help to understand the signaling processes in more detail. Single cell transcriptomics and metabolomics will also contribute to a more holistic view of the outcome of ROS signaling. In addition, artificial intelligence highly improved our understanding of protein structures [109] and could help in the future to predict structural changes under different redox conditions. So far, an enormous experimental effort has been necessary to unravel the structures of approximately 100,000 unique proteins, which still is only a tiny fraction of the billions of existing proteins. In contrast to the explosion in available genomic sequences and gene expression data produced by the technical advances in sequencing methods, a similar expansion in knowledge on available protein structures has so far been prevented by the intrinsic challenge of experimental structure determination. AlphaFold2 is a novel machine learning approach that incorporates physical and biological knowledge about protein structure which is now capable of predicting protein structures to near experimental accuracy in a majority of cases and greatly outperforms other methods [109]. Changes of protein structures by different redox environments await integration in the next step. Therefore, it can be expected that these technological advances will foster our understanding of intracellular redox signaling during development and stress and will help to determine how specificity in signaling is achieved.

Availability of data and materials

Not applicable.



Reactive oxygen species


Photochemical capacity


Rubisco-containing bodies

ATI-PS bodies:

ATG8-interacting-protein1-plastid-associated bodies


Senescence-associated vacuoles


Chloroplast vesiculation-containing vesicles


Transcription factor


Ascorbate peroxidase




G-Box binding factor 1


Alternative oxidase


Hydrogen peroxide sensor protein


Redox-sensitive green fluorescent protein




Nitroblue tetrazolium




2′,7′-Dichlorofluorescin diacetate


OxyR transcription factor regulatory domain of Escherichia coli


  1. Noodén LD, Leopold AC. Senescence and aging in plants. San Diego: Academic Press; 1988.

    Google Scholar 

  2. Breeze E, Harrison E, McHattie S, Hughes L, Hickman R, Hill C, et al. High-resolution temporal profiling of transcripts during Arabidopsis leaf senescence reveals a distinct chronology of processes and regulation. Plant Cell. 2011;23:873–94.

    CAS  PubMed  PubMed Central  Google Scholar 

  3. Ahmad S, Guo Y. Signal transduction in leaf senescence: progress and perspective. Plants. 2019;8:405.

    CAS  PubMed Central  Google Scholar 

  4. Thomas H. Senescence, ageing and death of the whole plant. New Phytol. 2013;197(3):696–711.

    PubMed  Google Scholar 

  5. Zentgraf U, Doll J. Arabidopsis WRKY53, a node of multi-layer regulation in the network of senescence. Plants. 2019;8(12):578.

    CAS  PubMed Central  Google Scholar 

  6. Zhang Y, Wang HL, Li Z, Guo H. Genetic network between leaf senescence and plant immunity: crucial regulatory nodes and new insights. Plants. 2020;9(4):495.

    CAS  PubMed Central  Google Scholar 

  7. Sade N, Del Mar R-W, Umnajkitikorn K, Blumwald E. Stress-induced senescence and plant tolerance to abiotic stress. J Exp Bot. 2018;69(4):845–53.

    CAS  PubMed  Google Scholar 

  8. Biswal UC, Biswal B, Raval MK. Transformation of chloroplast to gerontoplast. In: Chloroplast biogenesis. Springer, Dordrecht; 2003. p. 155–242.

  9. Hendry GAF, Houghton JD, Brown SB. The degradation of chlorophyll-a biological enigma. New Phytol. 1987;107:255–302.

    CAS  PubMed  Google Scholar 

  10. Domínguez F, Cejudo FJ. Chloroplast dismantling in leaf senescence. J Exp Bot. 2021;72(16):5905–18.

    PubMed  Google Scholar 

  11. Izumi M, Nakamura S. Chloroplast protein turnover: the influence of extraplastidic processes, including autophagy. Int J Mol Sci. 2018;19(3):828.

    PubMed Central  Google Scholar 

  12. Zhuang X, Jiang L. Chloroplast degradation: multiple routes into the vacuole. Front Plant Sci. 2019;10:359.

    PubMed  PubMed Central  Google Scholar 

  13. Woodson JD, Joens MS, Sinson AB, Gilkerson J, Salom PA, Weigel D, et al. Ubiquitin facilitates a quality-control pathway that removes damaged chloroplasts. Science. 2015;350:450–4.

    CAS  PubMed  PubMed Central  Google Scholar 

  14. Corpas FJ, González-Gordo S, Palma JM. Plant peroxisomes: a factory of reactive species. Front Plant Sci. 2020;11:853.

    PubMed  PubMed Central  Google Scholar 

  15. Pan R, Reumann S, Lisik P, Tietz S, Olsen LJ, Hu J. Proteome analysis of peroxisomes from dark-treated senescent Arabidopsis leaves. J Integr Plant Biol. 2018;60:1028–50.

    CAS  PubMed  Google Scholar 

  16. Chrobok D, Law SR, Brouwer B, Lindén P, Ziolkowska A, Liebsch D, et al. Dissecting the metabolic role of mitochondria during developmental leaf senescence. Plant Physiol. 2012;172(4):2132–53.

    Google Scholar 

  17. Maxwell DP, Nickels R, McIntosh L. Evidence of mitochondrial involvement in the transduction of signals required for the induction of genes associated with pathogen attack and senescence. Plant J. 2002;29:269–79.

    CAS  PubMed  Google Scholar 

  18. Millenaar FF, Lambers H. The alternative oxidase: in vivo regulation and function. Plant Biol. 2003;5:2–15.

    CAS  Google Scholar 

  19. Bengoa Luoni S, Astigueta FH, Nicosia S, Moschen S, Fernandez P, Heinz R. Transcription factors associated with leaf senescence in crops. Plants. 2019;8:411.

    PubMed Central  Google Scholar 

  20. Guo Y, Ren G, Zhang K, Li Z, Miao Y, Guo H. Leaf senescence: progression, regulation, and application. Mol Hortic. 2021;1:5.

  21. Guo Y, Cai Z, Gan S. Transcriptome of Arabidopsis leaf senescence. Plant Cell Environ. 2004;27:521–49.

    CAS  Google Scholar 

  22. Balazadeh S, Kwasniewski M, Caldana C, Mehrnia M, Zanor MI, Xue GP, Mueller-Roeber B. ORS1, an H2O2-responsive NAC transcription factor, controls senescence in Arabidopsis thaliana. Mol Plant. 2011;4:346–60.

  23. Kim HJ, Nam HG, Lim PO. Regulatory network of NAC transcription factors in leaf senescence. Curr Opin Plant Biol. 2016;33:48–56.

    PubMed  Google Scholar 

  24. Mishra B, Sun Y, Howton T, Kumar NM, Mukhtar S. Dynamic modeling of transcriptional gene regulatory network uncovers distinct pathways during the onset of Arabidopsis leaf senescence. npj Syst Biol Appl. 2018;4:35.

  25. Dong J, Chen C, Chen Z. Expression profiles of the Arabidopsis WRKY gene superfamily during plant defense response. Plant Mol Biol. 2003;51:21–37.

    CAS  Google Scholar 

  26. Xu X, Chen C, Fan B, Chen Z. Physical and functional interactions between pathogen-induced Arabidopsis WRKY18, WRKY40, and WRKY60 transcription factors. Plant Cell. 2006;18(5):1310–26.

    CAS  PubMed  PubMed Central  Google Scholar 

  27. Potschin M, Schlienger S, Bieker S, Zentgraf U. Senescence networking: WRKY18 is an upstream regulator, a downstream target gene, and a protein interaction partner of WRKY53. J Plant Growth Regul. 2014;33:106–18.

    CAS  Google Scholar 

  28. Llorca CM, Potschin M, Zentgraf U. bZIPs and WRKYs: two large transcription factor families executing two different functional strategies. Front Plant Sci. 2014;5:169.

    PubMed  PubMed Central  Google Scholar 

  29. Zentgraf U, Jobst J, Kolb D, Rentsch D. Senescence-related gene expression profiles of rosette leaves of Arabidopsis thaliana: leaf age versus plant age. Plant Biol. 2004;6(2):178–83.

    CAS  PubMed  Google Scholar 

  30. Xie Y, Huhn K, Brandt R, Potschin M, Bieker, S, Straub D, Doll J, Drechsler T, Zentgraf U, Wenkel S. REVOLUTA and WRKY53 connect early and late leaf development in Arabidopsis. Development. 2914;141:4772–83.

  31. Doll J, Muth M, Riester L, Nebel S, Bresson J, Lee HC, Zentgraf U. Arabidopsis thaliana WRKY25 transcription factor mediates oxidative stress tolerance and regulates senescence in a redox-dependent manner. Front Plant Sci. 2020;10:1734.

    PubMed  PubMed Central  Google Scholar 

  32. Dietz KJ. Redox regulation of transcription factors in plant stress acclimation and development. Antioxid Redox Signal. 2014;21:1356–72.

    CAS  PubMed  Google Scholar 

  33. Li Y, Liu W, Zhong H, Zhang HL, Xia Y. Redox-sensitive bZIP68 plays a role in balancing stress tolerance with growth in Arabidopsis. Plant J. 2019;100:768–83.

    CAS  PubMed  Google Scholar 

  34. Tan XL, Zhao YT, Shan W, Kuang JF, Lu WJ, Su XG, Tao NG, Lakshmanan P, Chen JY. Melatonin delays leaf senescence of postharvest Chinese flowering cabbage through ROS homeostasis. Food Res Int. 2020;138:109790.

  35. Waszczak C, Carmody M, Kangasjärvi J. Reactive oxygen species in plant signaling. Annu Rev Plant Biol. 2018;69:209–36.

    CAS  PubMed  Google Scholar 

  36. Bienert GP, Schjoerring JK, Jahn TP. Membrane transport of hydrogen peroxide. BBA. 2006;1758(8):994–1003.

    CAS  PubMed  Google Scholar 

  37. Duan X, Zhang H, Zhang D, Sheng J, Lin H, Jiang Y. Role of hydroxyl radical in modification of cell wall polysaccharides and aril breakdown during senescence of harvested longan fruit. Food Chem. 2011;128(1):203–7.

    CAS  PubMed  Google Scholar 

  38. Pottosin I, Zepeda-Jazo I, Bose J, Shabala S. An anion conductance, the essential component of the hydroxyl-radical-induced ion current in plant roots. Int J Mol Sci. 2018;19(3):897.

    PubMed Central  Google Scholar 

  39. Richards SL, Wilkins KA, Swarbreck SM, Anderson AA, Habib N, Smith AG, McAinsh M, Davies JM. The hydroxyl radical in plants: from seed to seed. J Exp Bot. 2015;66(1):37–46.

    CAS  PubMed  Google Scholar 

  40. Demidchik V, Cuin TA, Svistunenko D, Smith SJ, Miller AJ, Shabala S, Sokolik A, Yurin V. Arabidopsis root K+-efflux conductance activated by hydroxyl radicals: single-channel properties, genetic basis and involvement in stress-induced cell death. J Cell Sci. 2010;123(Pt 9):1468–79.

    CAS  PubMed  Google Scholar 

  41. Hancock JT, Russell G. Downstream signalling from molecular hydrogen. Plants. 2021;10(2):367.

    CAS  PubMed  PubMed Central  Google Scholar 

  42. Foyer CH, Bloom AJ, Queval G, Noctor G. Photorespiratory metabolism: genes, mutants, energetics and redox signaling. Ann Rev Plant Biol. 2009;60:455–84.

    CAS  Google Scholar 

  43. D’Autréaux B, Toledano M. ROS as signalling molecules: mechanisms that generate specificity in ROS homeostasis. Nat Rev Mol Cell Biol. 2007;8:813–24.

    PubMed  Google Scholar 

  44. Sathyanarayanan PV, Poovaiah BW. Decoding Ca2+ signals in plants. CRC Crit Rev Plant Sci. 2004;23(1):1–11.

    CAS  PubMed  Google Scholar 

  45. Schönknecht G. Calcium signals from the vacuole. Plants. 2013;2(4):589–614.

    PubMed  PubMed Central  Google Scholar 

  46. Mittler R, Vanderauwera S, Gollery M, Van Breusegem F. Reactive oxygen gene network of plants. Trends Plant Sci. 2004;9:490–8.

    CAS  PubMed  PubMed Central  Google Scholar 

  47. Zandalinas SI, Mittler R. Vascular and nonvascular transmission of systemic reactive oxygen signals during wounding and heat stress. Plant Physiol. 2021;186(3):1721–33.

    CAS  PubMed  Google Scholar 

  48. Zimmermann P, Heinlein C, Orendi G, Zentgraf U. Senescence specific regulation of catalases in Arabidopsis thaliana (L.) Heynh. Plant Cell Environ. 2006;29:1049–60.

  49. Bieker S, Riester L, Stahl M, Franzaring J, Zentgraf U. Senescence-specific alteration of hydrogen peroxide levels in Arabidopsis thaliana and oilseed rape spring variety Brassica napus L. cv. Mozart. J Integr Plant Biol. 2012;54(8):540–54

  50. Foyer CH, Noctor G. Stress-triggered redox signalling: what’s in pROSpect? Front Plant Sci. 2016;39:951–64.

    CAS  Google Scholar 

  51. Walton PA, Brees C, Lismont C, Apanasets O, Fransen M. The peroxisomal import receptor PEX5 functions as a stress sensor, retaining catalase in the cytosol in times of oxidative stress. Biochim Biophys Acta. 2017;1864:1833–43.

    CAS  Google Scholar 

  52. Del Río LA, Corpas FJ, Sandalio LM, Palma JM, Barroso JB. Plant peroxisomes, reactive oxygen metabolism and nitric oxide. IUBMB Life. 2003;55:71–81.

    PubMed  Google Scholar 

  53. Kaur N, Reumann S, Hu J. Peroxisome biogenesis and function. Arabidopsis Book. 2009;7:e0123.

  54. Eastmond PJ. MONODEHYROASCORBATE REDUCTASE4 is required for seed storage oil hydrolysis and postgerminative growth in Arabidopsis. Plant Cell. 2007;19:1376–87.

    CAS  PubMed  PubMed Central  Google Scholar 

  55. Fransen M, Lismont C. Redox signaling from and to peroxisomes: progress, challenges, and prospects. Antioxid Redox Signal. 2019;30:95–112.

    CAS  PubMed  Google Scholar 

  56. Sandalio LM, Peláez-Vico MA, Molina-Moya E, Romero-Puertas MC. Peroxisomes as redox-signaling nodes in intracellular communication and stress responses. Plant Physiol. 2021;186(1):22–35.

    CAS  PubMed  PubMed Central  Google Scholar 

  57. Smykowski A, Zimmermann P, Zentgraf U. G-Box binding factor1 reduces CATALASE2 expression and regulates the onset of leaf senescence in Arabidopsis. Plant Physiol. 2010;153:1321–31.

    CAS  PubMed  PubMed Central  Google Scholar 

  58. Smykowski A, Fischer SM, Zentgraf U. Phosphorylation affects DNA-binding of the senescence-regulating bZIP transcription factor GBF1. Plants. 2015;4(3):691–709.

    CAS  PubMed  PubMed Central  Google Scholar 

  59. Sibéril Y, Doireau P, Gantet P. Plant bZIP G-box binding factors. Modular structure and activation mechanisms. Eur J Biochem. 2001;268(22):5655–66.

  60. Feierabend J, Schaan C, Hertwig B. Photoinactivation of catalase occurs under both high and low temperature stress conditions and accompanies photoinhibition of photosystem II. Plant Physiol. 1992;100:1554–61.

    CAS  PubMed  PubMed Central  Google Scholar 

  61. Panchuk II, Zentgraf U, Volkov RA. Expression of the APX gene family during leaf senescence of Arabidopsis thaliana. Planta. 2005;222:926–32.

    CAS  PubMed  Google Scholar 

  62. Miyake C, Asada K. Inactivation mechanism of ascorbate peroxidase at low concentrations of ascorbate: hydrogen peroxide decomposes compound I of ascorbate peroxidase. Plant Cell Physiol. 1986;37:423–30.

    Google Scholar 

  63. Ye ZZ, Rodriguez R, Tran A, Hoang H, de los Santos D, Brown S, Vellanoweth RL. The developmental transition to flowering represses ascorbate peroxidase activity and induces enzymatic lipid peroxidation in leaf tissue in Arabidopsis thaliana. Plant Sci. 2000;158:115–27.

  64. Lai AG, Doherty CJ, Mueller-Roeber B, Kay SA, Schippers JH, Dijkwel PP. CIRCADIAN CLOCK-ASSOCIATED 1 regulates ROS homeostasis and oxidative stress responses. Proc Natl Acad Sci USA. 2012;109(42):17129–34.

  65. Chaouch S, Queval G, Vanderauwera S, Mhamdi A, Vandorpe M, Langlois-Meurinne M, et al. Peroxisomal hydrogen peroxide is coupled to biotic defense responses by ISOCHORISMATE SYNTHASE1 in a daylength-related manner. Plant Physiol. 2010;153:1692–705.

    CAS  PubMed  PubMed Central  Google Scholar 

  66. Queval G, Neukermans J, Vanderauwera S, Van Breusegem F, Noctor G. Day length is a key regulator of transcriptomic responses to both CO2 and H2O2 in Arabidopsis. Plant Cell Environ. 2012;35:374–87.

    CAS  PubMed  Google Scholar 

  67. Zhang Y, Wang Y, Wei H, Li N, Tian W, Chong K, Wang L. Circadian evening complex represses jasmonate-induced leaf senescence in Arabidopsis. Mol Plant. 2018;11(2):326–37.

    CAS  PubMed  Google Scholar 

  68. Sanchez SE, Kay SA. The plant circadian clock: from a simple timekeeper to a complex developmental manager. Cold Spring Harb Perspect Biol. 2016;8:a027748.

  69. Kim H, Kim Y, Yeom M, Lim J, Nam HG. Age-associated circadian period changes in Arabidopsis leaves. J Exp Bot. 2016;67(9):2665–73.

    CAS  PubMed  PubMed Central  Google Scholar 

  70. Sewelam N, Jaspert N, Van Der Kelen K, Tognetti NB, Schmitz J, Frerigmann H, et al. Spatial H2O2 signaling specificity: H2O2 from chloroplasts and peroxisomes modulates the plant transcriptome differentially. Mol Plant. 2014;7:1191–210.

    CAS  PubMed  Google Scholar 

  71. Rosenwasser S, Fluhr R, Joshi JR, Leviatan N, Sela N, Hetzroni A, Friedman H. ROSMETER: a bioinformatic tool for the identification of transcriptomic imprints related to reactive oxygen species type and origin provides new insights into stress responses. Plant Physiol. 2013;163:1071–83.

    CAS  PubMed  PubMed Central  Google Scholar 

  72. Maxwell DP, Wang Y, McIntosh L. The alternative oxidase lowers mitochondrial reactive oxygen production in plant cells. Proc Natl Acad Sci USA. 1999;96(14):8271–6.

    CAS  PubMed  PubMed Central  Google Scholar 

  73. Sako K, Futamura Y, Shimizu T, Matsui A, Hirano H, Kondoh Y, et al. Inhibition of mitochondrial complex I by the novel compound FSL0260 enhances high salinity-stress tolerance in Arabidopsis thaliana. Sci Rep. 2020;10(1):8691.

    CAS  PubMed  PubMed Central  Google Scholar 

  74. Zentgraf U, Zimmermann P, Smykowski A. Role of intracellular hydrogen peroxide as signaling molecule for plant senescence. In: Nagata T, editor. Senescence. In-Tech Open Access Publishing; 2012. ISBN: 978-953-51-0144-4.

  75. Labs M, Rühle T, Leister D. The antimycin A-sensitive pathway of cyclic electron flow: from 1963 to 2015. Photosynth Res. 2016;2016(129):231–8.

    Google Scholar 

  76. Smirnoff N, Arnaud D. Hydrogen peroxide metabolism and functions in plants. New Phytol. 2019;221:1197–214.

    CAS  PubMed  Google Scholar 

  77. Romero-Puertas MC, Rodríguez-Serrano M, Corpas FJ, Gómez M, del Río LA, Sandalio LM. Cadmium-induced subcellular accumulation of O·2 and H2O2 in pea leaves. Plant Cell Environ. 2004;27:1122–34.

    CAS  Google Scholar 

  78. Caplan JL, Kumar AS, Park E, Padmanabhan MS, Hoban K, Modla S, et al. Chloroplast stromules function during innate immunity. Dev Cell. 2015;34:45–57.

    CAS  PubMed  PubMed Central  Google Scholar 

  79. Kumar AS, Park E, Nedo A, Alqarni A, Ren L, Hoban K, et al. Stromule extension along microtubules coordinated with actin-mediated anchoring guides perinuclear chloroplast movement during innate immunity. eLife. 2018;7:1–33.

    Google Scholar 

  80. Reumann S, Ma C, Lemke S, Babujee L. AraPerox. A database of putative Arabidopsis proteins. Plan Physiol. 2004;136:2587–608.

    CAS  Google Scholar 

  81. Vandenabeele S, Van Der Kelen K, Dat J, Gadjev I, Boonefaes T, Morsa S, et al. A comprehensive analysis of hydrogen peroxide-induced gene expression in tobacco. Proc Natl Acad Sci USA. 2003;100:16113–8.

    CAS  PubMed  PubMed Central  Google Scholar 

  82. Vandenabeele S, Vanderauwera S, Vuylsteke M, Rombauts S, Langebartels C, Seidlitz HK, et al. Catalase deficiency drastically affects gene expression induced by high light in Arabidopsis thaliana. Plant J. 2004;39:45–58.

    CAS  PubMed  Google Scholar 

  83. Su T, Li W, Wang P, Ma C. Dynamics of peroxisome homeostasis and its role in stress response and signaling in plants. Front Plant Sci. 2019;10:705.

    PubMed  PubMed Central  Google Scholar 

  84. Dietz KJ, Hell R. Thiol switches in redox regulation of chloroplasts: balancing redox state, metabolism and oxidative stress. Biol Chem. 2015;396:483–94.

    CAS  PubMed  Google Scholar 

  85. Noctor G, Foyer CH. Intracellular redox compartmentation and ROS-related communication in regulation and signaling. Plant Physiol. 2016;171:1581–92.

    CAS  PubMed  PubMed Central  Google Scholar 

  86. Brunkard JO, Runkel AM, Zambryski PC. Chloroplasts extend stromules independently and in response to internal redox signals. Proc Natl Acad Sci USA. 2015;112(32):10044–9.

    CAS  PubMed  PubMed Central  Google Scholar 

  87. Hanson MR, Hines KM. Stromules: probing formation and function. Plant Physiol. 2018;176(1):128–37.

    CAS  PubMed  Google Scholar 

  88. Belousov VV, Fradkov AF, Lukyanov KA, Staroverov DB, Shakhbazov KS, Terskikh AV, Lukyanov S. Genetically encoded fluorescent indicator for intracellular hydrogen peroxide. Nat Methods. 2006;3:281–6.

    CAS  PubMed  Google Scholar 

  89. Exposito-Rodriguez M, Laissue PP, Yvon-Durocher G, Smirnoff N, Mullineaux PM. Photosynthesis-dependent H2O2 transfer from chloroplasts to nuclei provides a high-light signalling mechanism. Nat Commun. 2017;8(1):49.

    PubMed  PubMed Central  Google Scholar 

  90. Møller IM, Kristensen BK. Protein oxidation in plant mitochondria as a stress indicator. Photochem Photobiol Sci. 2004;3:730–5.

    PubMed  Google Scholar 

  91. Nietzel T, Mostertz J, Hochgräfe F, Schwarzländer M. Redox regulation of mitochondrial proteins and proteomes by cysteine thiol switches. Mitochondrion. 2017;33:72–83.

    CAS  PubMed  Google Scholar 

  92. Borniego ML, Molina MC, Guiamét JJ, Martinez DE. Physiological and proteomic changes in the apoplast accompany leaf senescence in Arabidopsis. Front Plant Sci. 2020;10:1635.

    PubMed  PubMed Central  Google Scholar 

  93. Berens ML, Wolinska KW, Spaepen S, Ziegler J, Nobori T, Nair A, et al. Balancing trade-offs between biotic and abiotic stress responses through leaf age-dependent variation in stress hormone cross-talk. Proc Natl Acad Sci USA. 2019;116(6):2364–73.

    CAS  PubMed  PubMed Central  Google Scholar 

  94. Shapiguzov A, Vainonen JP, Wrzaczek M, Kangasjärvi J. ROS-talk—how the apoplast, the chloroplast, and the nucleus get the message through. Front Plant Sci. 2012;212(3):292.

    Google Scholar 

  95. Shapiguzov A, Vainonen JP, Hunter K, Tossavainen H, Tiwari A, Järvi S, et al. Arabidopsis RCD1 coordinates chloroplast and mitochondrial functions through interaction with ANAC transcription factors. eLife. 2019;8:e43284.

    PubMed  PubMed Central  Google Scholar 

  96. Fichman Y, Miller G, Mittler R. Whole-plant live imaging of reactive oxygen species. Mol Plant. 2019;12:1203–10.

    CAS  PubMed  Google Scholar 

  97. Davletova S, Schlauch K, Coutu J, Mittler R. The zinc-finger protein Zat12 plays a central role in reactive oxygen and abiotic stress signaling in Arabidopsis. Plant Physiol. 2005;139(2):847–56.

    CAS  PubMed  PubMed Central  Google Scholar 

  98. Miller G, Schlauch K, Tam R, Cortes D, Torres MA, Shulaev V, et al. The plant NADPH oxidase RBOHD mediates rapid systemic signaling in response to diverse stimuli. Sci Signal. 2009;2(84):ra45.

  99. Ren W, Ai HW. Genetically encoded fluorescent redox probes. Sensors. 2013;13(11):15422–33.

    PubMed  PubMed Central  Google Scholar 

  100. Meyer AJ, Brach T, Marty L, Kreye S, Rouhier N, Jacquot JP, Hell R. Redox-sensitive GFP in Arabidopsis thaliana is a quantitative biosensor for the redox potential of the cellular glutathione redox buffer. Plant J. 2007;52:973–86.

    CAS  PubMed  Google Scholar 

  101. Ugalde JM, Fuchs P, Nietzel T, Cutolo EA, Homagk M, Vothknecht UC, et al. Chloroplast-derived photo-oxidative stress causes changes in H2O2 and EGSH in other subcellular compartments. Plant Physiol. 2021;186(1):125–41.

    CAS  PubMed  PubMed Central  Google Scholar 

  102. Zheng M, Aslund F, Storz G. Activation of the OxyR transcription factor by reversible disulfide bond formation. Science. 1998;279:1718–21.

    CAS  PubMed  Google Scholar 

  103. Aslund F, Zheng M, Beckwith J, Storz G. Regulation of the OxyR transcription factor by hydrogen peroxide and the cellular thiol-disulfide status. Proc Natl Acad Sci USA. 1999;96:6161–5.

    CAS  PubMed  PubMed Central  Google Scholar 

  104. Costa A, Drago I, Behera S, Zottini M, Pizzo P, Schroeder JI, et al. H2O2 in plant peroxisomes: an in vivo analysis uncovers a Ca(2+)-dependent scavenging system. Plant J. 2010;62:760–72.

    CAS  PubMed  PubMed Central  Google Scholar 

  105. Zimmermann P, Zentgraf U. The correlation between oxidative stress and leaf senescence during plant development. Cell Mol Biol Lett. 2005;10:515–34.

    CAS  PubMed  Google Scholar 

  106. Dietz KJ, Turkan I, Krieger-Liszkay A. Redox- and reactive oxygen species-dependent signaling into and out of the photosynthesizing chloroplast. Plant Physiol. 2016;171(3):1541–50.

    CAS  PubMed  PubMed Central  Google Scholar 

  107. Noctor G, Reichheld JP, Foyer CH. ROS-related redox regulation and signaling in plants. Semin Cell Dev Biol. 2018;80:3–12.

    CAS  PubMed  Google Scholar 

  108. Schwarzländer M, Dick TP, Meyer AJ, Morgan B. Dissecting redox biology using fluorescent protein sensors. Antioxid Redox Signal. 2016;24(13):680–712.

    PubMed  Google Scholar 

  109. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583–9.

    CAS  PubMed  PubMed Central  Google Scholar 

Download references


We apologize for not being able to cite all relevant work. We thank Prof. Dr. Christopher Grefen, University of Bochum, Germany for providing the plant line expressing GFP in chloroplasts to measure stromule formation in the wild type background. This article was specially invited by the editors and represents work by leading researchers.


Open Access funding enabled and organized by Projekt DEAL. We acknowledge sponsorship by Open Access Publishing Fund of University of Tübingen. This work was supported by the Deutsche Forschungsgemeinschaft (DFG), CRC1101, B06.

Author information

Authors and Affiliations



UZ wrote the first draft, SB provided the confocal pictures, SB and AA-G have corrected and edited the text and the figures; all authors contributed to the final manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Ulrike Zentgraf.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zentgraf, U., Andrade-Galan, A.G. & Bieker, S. Specificity of H2O2 signaling in leaf senescence: is the ratio of H2O2 contents in different cellular compartments sensed in Arabidopsis plants?. Cell Mol Biol Lett 27, 4 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Leaf senescence
  • Free oxygen radicals
  • ROS
  • Hydrogen peroxide
  • Stromules
  • Senescence regulation
  • Intracellular compartments