Mapping the redox regulatory landscape: a bit of history and a look to the future
Amna Mhamdi, Graham Noctor · Journal of Experimental Botany · 2024
Redox exchanges are the beating heart of plant growth. Through photosynthetic and respiratory electron transfer, as well as numerous other oxidoreduction processes, plants generate reducing compounds and energy essential to their development and survival. Thus, it is hardly surprising that systems have evolved to control and monitor the status of key redox components, and to use this information as an indicator of environmental conditions. Such information is crucial in plant adaptation and acclimation to ongoing modifications in atmospheric composition and associated changes in climate, which are predicted to exacerbate the intensity of stress conditions. Redox regulation of plant function has been known for over half a century. The seminal report that chloroplastic CO2 fixation is light regulated through factors other than ATP and NADPH supply (Buchanan et al., 1967) was followed some years later by evidence that plastoquinone, or a functionally adjacent component of the photosynthetic electron transport chain, modulated the phosphorylation status of thylakoid complexes and, thereby, their light-harvesting functions (Horton et al., 1981). Subsequent studies elucidated the details of the mechanisms underlying these phenomena (Wolosiuk and Buchanan, 1977; Droux et al., 1987; Bellafiore et al., 2005), leading to a conceptual shift in our understanding of plant bioenergetics: not only is electron exchange essential for energy conversion but it is also co-opted to regulate key physiological processes (Fig. 1). The last 30 years have witnessed an explosion in this area, as researchers have built on these early observations to explore a hinterland of redox regulation that is turning out to be surprisingly complex and intricate. Thanks to these efforts, we now know that redox regulation is not confined to photosynthesis, or even to the chloroplast. It extends to multiple processes occurring in subcellular compartments such as the apoplast, the cytosol, the mitochondria, and the nucleus. Some of the many processes regulated by redox in plants. The figure places emphasis on processes covered in this Special Issue and is not meant to be an exhaustive depiction of all players or processes that are under redox control. Created with BioRender.com. A key development that occurred alongside the elucidation of redox regulation of plant enzymes was the increasing acceptance that reactive oxygen species (ROS), and notably superoxide and hydrogen peroxide (H2O2), are integrated into cell function and that they are not purely damaging molecules whose production must be avoided at all costs. In plants, the generation of superoxide and H2O2 was studied in terms of their roles in regulating photosynthetic electron transport and the associated yield of ATP (Allen and Hall, 1974; Asada et al., 1974). A paradigm shift was initiated by the discovery that these reactive molecules are also generated at the cell surface in response to pathogen attack (Doke, 1985). The characterization of this so-called ‘oxidative burst’ marked the beginning of the view that ROS can act as signalling molecules in plants. It subsequently became evident that ROS functions are tightly entwined with phytohormone signalling, as evidenced by the similarity between hormonal responses to pathogen attack and to the oxidizing pollutant, ozone (Kangasjärvi et al., 1994). As the roles of oxidants as regulators became increasingly accepted, nitric oxide (NO) also emerged as an important player in plant physiology. Here again, breakthrough findings were reported for plant systems interacting with pathogenic invaders (Delledonne et al., 1998). Shortly afterwards, key roles in plant defence were reported for S-nitrosoglutathione (GSNO) and GSNO reductase (Feechan et al., 2005). The redox regulatory landscape has grown to be increasingly complex and has been the subject of numerous papers and special issues in plant-focused journals over recent years. In this Special Issue, we have tried to place emphasis on the redox regulation of gene expression, bearing in mind that the circuitry involved can be relatively direct or quite complex (Fig. 2). Within this overarching area, key themes that emerge repeatedly in this issue are regulation of factors involved in chromatin remodelling, the roles of redox in plant responses to pathogens, and the biochemical modifications that underlie such processes. Nine reviews written by researchers active in the field discuss the latest knowledge in these and related areas, complemented by three original papers covering the effect of alternative chloroplast electron transfer pathways on plant development (Arce et al., 2024), the roles of NADPH oxidase-sourced ROS in cysteine modification (Hino et al., 2024), and the functions of specific transcription factors that intervene in redox-related retrograde signalling from the chloroplast to the nucleus (Luo et al., 2024). Overview of known or potential sites in the redox regulation of gene expression. Three levels of control are (1) chromatin remodelling through redox control of histone-modifying enzymes or histone themselves (a and b) (2) redox regulation of transcription factors and associated co-regulators, and (3) redox modification of the core transcriptional machinery. Ac, acetylation; CHR, chromatin-remodelling protein; GTF, general transcription factors; HAT, histone acetyltransferase; HDA, histone deacetylase; JMJ, Jumonji domain-containing protein; Me, methylation; NAC, NAM/ATAF/CUC; NPR1, NONEXPRESSOR OF PATHOGENESIS-RELATED GENE 1; PAF1C, polymerase-associated factor 1; CPSF, cleavage and polyadenylation specificity factor; TGA, TGACG SEQUENCE-SPECIFIC BINDING PROTEIN. Created with BioRender.com. A Community Resource article reports an updated version of a tool focused on analysis of covalent post-translational changes to proteins (Willems et al., 2024), emphasizing the potential importance of cysteine status within the complex and diverse network of amino acid modifications. The importance of this theme echoes through several of the review papers, including redox control of epigenetic processes and gene transcription in general (Auverlot et al., 2024; Dard et al., 2024). The essential roles of redox changes in plant responses to stress are discussed by Denjalli et al. (2024), while several reviews consider aspects of redox-linked orchestration of plant immunity, with a focus on the influence of factors such as ROS, NO, and glutathione (Giulietti et al., 2024; Liu et al., 2024; Noctor et al., 2024). NO is also central to the discussion of plant responses to hypoxia (Samant et al., 2024), and the enigmatic but established roles of ROS in root hair development also receive attention (Lopez et al., 2024). Last, but not least, an overview focusing specifically on superoxide within the nuclear signalling network provides a timely reminder that H2O2 is not the only reduced form of oxygen that can act as a signal (Karpinska and Foyer, 2024). Today, the key roles of redox factors in regulating a diverse set of plant developmental processes and environmental responses are no longer in doubt. The question is: how does this regulation occur? Within the context of a longstanding search for H2O2 sensors, it has been reported that a leucine-rich repeat (LLR) receptor kinase acts in such a role (Wu et al., 2020). However, this probably explains only part of the response to H2O2, which is only one of many other redox-active factors that can influence protein function. While various mechanisms are possible, most attention has continued to focus on modification of protein cysteine groups (Fig. 3). Starting with fluorescent labelling of cysteine residues, through columns carrying modified thioredoxins, YAP1-based technology, and chemical probes such as biotin, inventories of proteins with modifiable cysteines have grown increasingly long. Technological developments in mass spectrometry have been key to identifying the cysteines that are susceptible to modification. Nevertheless, many of these potential targets await functional evaluation, and a number of questions often remain unresolved. For example, does the modification occur in vivo (or in the absence of the probe)? If so, under which physiological conditions? How specific is the modification to a particular cysteine residue? What proportion of cysteines at any given position is likely to undergo the modification as a function of changes in external conditions, developmental stage, or internal status? What effects does the modification have on protein function? Some of the many modifications of cysteine groups that can affect protein structure and/or function. Most of these functions are reversible through a variety of pathways and mechanisms. H2O2 triggered changes in cysteine residues and their recycling pathways. GSH, reduced glutathione; GSNO, S-nitrosoglutathione; NO, nitric oxide. Created with BioRender.com. One redox target that has been extensively studied over the last 20 years is NPR1 (NONEXPRESSOR OF PATHOGENESIS RELATED GENE 1), a cytosolic/nuclear transcription co-regulator that interacts with TGA transcription factors to activate salicylic acid-dependent signalling. Cysteines on NPR1 can be subjected to several types of redox modification involving not only thioredoxins but also NO and/or GSNO, and the TGA factors themselves also contain sensitive cysteines (Després et al., 2003; Mou et al., 2003; Tada et al., 2008; Lindermayr et al., 2010). However, some of the key details of the model developed have recently been questioned (Ishihama et al., 2021), emphasizing the technical difficulties in arriving at definitive concepts of how redox regulation occurs, even for a relatively simple well-defined system. As mentioned above, light regulation of photosynthetic metabolism by the thioredoxin system was one of the first redox regulatory systems identified in plants. This entails reduction of a disulfide bond on several target enzymes, producing a clear change in conformation and activity, with the latter being readily measurable in extracts (Wolosiuk and Buchanan, 1977; Droux et al., 1987). Today, the list of possible thioredoxin targets grows ever longer, within the context of our increasing knowledge of modifications of numerous proteins not only by disulfide oxidoreduction but also by S-nitrosylation, S-glutathionylation, sulfenylation, etc. (Fig. 3). The impact of many of these changes on protein function remains less well established compared with the thioredoxin targets characterized in the 1970s and 1980s. Even in these more well-established cases, where the effect of oxidoreduction on a relatively abundant enzyme is clearly detectable, the impact on the relevant wider physiological process may not necessarily be marked. Rather, it is often conditional, depending on the relative importance of other regulatory mechanisms such as enzyme control by pH and other ions. Such conditionality underlines the tight integration of redox regulation with numerous other mechanisms that can influence protein function. Almost any protein modification that affects structure and function will have a hierarchy of impact at different biological levels. In many cases, the impact is likely to decrease as the complexity of the system under consideration increases; that is, a large effect on a protein’s functions is likely to have a smaller effect on the cellular processes in which the protein is involved, and the effect on the whole organism is likely to be smaller still (Fig. 4). This notion is well established in the area of metabolic control analysis. Other properties characteristic of plants can contribute to apparently explicit effects on protein function becoming increasingly cryptic at higher organizational levels. One is a strong influence of a fluctuating environment, so that many phenotypes are likely to be conditional. Another is that plants show considerable developmental plasticity and can, if necessary, sacrifice individual tissues or organs that may be functionally compromised. Hierarchy of impact: to what extent do modifications of specific proteins feed through to modulate plant function, survival, and yield? Modification of a protein might substantially affect its structure or, in the case of enzymes, their in vitro activity. However, any such effects may, in many cases, become gradually attenuated as the system becomes increasingly more complex (inverted triangle). In the end, the impact on outcomes at the whole-plant level may be minor, or even undetectable. Created with BioRender.com. The resilience inherent in complex systems almost certainly contributes to their stability, but also makes it difficult to establish the true biological importance of any single modification as each encounters the inertia of the system. Indeed, the multiplicity of redox modifications at several levels of control could be an important point of this stability, as decisive modulation of overall biological processes may require the simultaneous activation of multiple switches. Within this context, one of the advantages of redox control is that many are inherently transient and dynamic, allowing easy reversal, sometimes via non-enzymatic reactions. Unquestionably, research risk significantly impedes progress in assessing the importance of specific redox modifications. Studying the functional importance of any given cysteine within the context of the whole organism is not only laborious and expensive in terms of time and other resources: because of the issues described above, it is also likely in many cases to yield negative results. Ongoing technological advances are likely to be crucial to overcoming this bottleneck. Promising developments include refinements in wet lab technology (click chemistry, etc.) and gene editing, while artificial intelligence (AI) is also bringing powerful new tools to the game that are likely to become increasingly indispensable in predicting functional impact. Tools such as AlphaFold are already providing informative data on three-dimensional protein structures (Jumper et al., 2021), and the accuracy and resolving power of such resources will almost certainly improve over the coming years. AI also promises to greatly enhance our ability to predict protein–protein and protein–nucleic acid interactions, both spatially and temporally (in terms of which physical interactions occur where and when), and to allow information from large analytical datasets obtained at different levels to be more easily integrated. Such developments will make it all the more important to define experimental conditions and avoid other potential pitfalls that might inadvertently skew the interpretation. Even here, however, AI is likely to have a potentially beneficial impact by helping to define appropriate experimental contexts and by shaping the technological development necessary to improve quantification of the interactions and/or modifications. Within the complexity of the subcellular and cellular network, this step-change in processing power is likely to enable us to take greater strides forward over the still poorly mapped redox landscape and to define its topography and geography in ever closer detail. In this way, we will be in a position to better understand how redox modifications functionally interact, not only with each other but also with non-redox regulatory mechanisms. The authors wish to take this opportunity to pay tribute to the outstanding contributions of Professors Jaakko Kangasjärvi and Gary Loake—both of whom recently and prematurely passed away—to advancing this area of plant research. The authors declare no conflict of interest. The Paris-Saclay laboratory receives financial support from the French Agence Nationale de la Recherche HIPATH project (ANR-17-CE20- 0025) and the Institut Universitaire de France (IUF). Work in this area in the Ghent laboratory is supported in part by the Research Foundation Flanders (FWO) [The Excellence of Science (EOS) Research project 869 30829584] and NUCLEOX (grant no. G007723N).