Unravelling mechanisms and impacts of day respiration in plant leaves: an introduction to a Virtual Issue

Abstract

Leaf respiration plays a key role in plant primary production due to its roles in determining rates of daily net carbon gain, nutrient acquisition, and growth. It is now well-accepted that leaf respiration differs considerably between darkness and light, not only in rate but also in its metabolic pathways and biological functions (Tcherkez et al., 2012). As such, it is common practice to use different terms: light (day) and dark (night) respiration, with associated variables describing their rates, Rlight (Rd) and Rdark (Rn). While leaf dark respiration has been documented for a long time, including its well-known relationship with growth, substrate supply and energy demand (O’Leary et al., 2019), there is still considerable uncertainty about many aspects of day respiration including: (1) how to measure the rate Rlight; (2) metabolic mechanisms underlying carbon dioxide (CO2) release and oxygen (O2) consumption; and (3) its impact on photosynthesis. This persistent uncertainty is highly problematic because it affects calculations of photosynthetic parameters, our understanding of nitrogen (N) assimilation and how it interacts with photorespiration, and calculations of carbon use efficiency or gross primary production at the canopy and/or ecosystem scale. This Virtual Issue compiles 33 articles that provide significant advances in addressing these challenges, propose new techniques to measure Rlight, quantitatively assess the impact of Rlight on calculated rates of carbon exchange in leaves, or enhance our understanding about day respiration through the documentation of diel patterns in Rdark. A specific review on day respiration can be found in Tcherkez et al. (2017b). Key aspects related to day respiration that are discussed in this Editorial are summarized in Fig. 1. Given that the Kok effect is unlikely to be due solely to irradiance-dependent changes in respiratory metabolism, the question arises of what might be a better method for measuring Rlight. Three new methods are proposed in this Virtual Issue. Gong et al. (2018) took advantage of the slow turnover of respiratory substrates to calculate Rlight using an isotopic disequilibrium approach. That is, when the isotope composition of CO2 of the atmosphere is changed abruptly, day respired-CO2 has the isotope composition of ‘old’ substrates for some time and this represents an excellent opportunity to extract the value of Rlight. One advantage of this technique is that it is independent of gm, thus providing a way of estimating Rlight, and in turn, assisting in calculating values of gm and cc. Berghuijs et al. (2019) elaborated on the 2D model of photosynthetic cells described in Berghuijs et al. (2017) and used observed values of A to predict maximal carboxylation (Vcmax) and Rlight by fitting. This technique assumes a certain geometry of the cells to incorporate gm implicitly (in practice, it is modelled with the diffusion coefficient of CO2 in water, potentially corrected for porosity and viscosity) thereby allowing the calculation of Rlight. This technique is interesting because it is more easily implementable than 3D-based models (lowering the demand for computing power). However, as always with geometric models, users must remain conscious that Rlight estimates are highly sensitive to calculated electron transport (J) and the chosen value of Γ*. The method proposed by Moualeu-Ngangue et al. (2017) takes advantage of an A–ci curve fitting where fluorescence and gas exchange measurements are made in parallel. Here, the input parameters are A, Φ, incident light and ci, while Rlight is a fitted parameter, as well as Vcmax, J and photosystem photon absorption parameters combined in a unique variable, denoted as τ. This study is very helpful because it clearly demonstrates that while estimating Rlight is technically possible using photosynthetic response curves, there are many instances where fitted Rlight values equal zero. In addition, Rlight values are generally lower than that those found using fitting methods that assume a constant gm. In other words, the persisting problem of mathematical interdependency between Rlight and cc (and thus gm) mentioned above (Eqn 1) cannot be easily overcome: alternative techniques independent of J or cc are required to estimate Rlight, like the isotope disequilibrium technique proposed by Gong et al. (2018). A further question that needs to be considered is what effect variations in Rlight have on calculated photosynthetic parameters? This question is of importance because photosynthetic capacity is essential for photosynthetic modelling (when scaling up from leaves to whole ecosystems) and perhaps, calculated photosynthetic parameters are not very sensitive to Rlight values. A comprehensive analysis of internal conductance showed that across plant lineages, gm estimates are not very sensitive to the chosen value of Rlight between 0.5 and 1.5 µmol m–2 s–1, with variations in Rlight leading to less than 10% changes in gm (Xiong & Flexas, 2020). By contrast, estimating Vcmax appears to be very sensitive to the chosen Rlight value, particularly when using the one-point Asat method (De Kauwe et al., 2016). Because of this, there is a much better agreement between Vcmax values calculated from A–ci curves and Asat, when observed values of Rlight are used instead of assumed values (e.g. Rlight = 1.5% of Vcmax) (De Kauwe et al., 2016). A critical question is the origin of CO2 produced by day respiration, not only because it may impact on the 12C/13C natural isotope composition but also because it reflects metabolic pathways mobilized by respiratory metabolism and the use of stored metabolites (reserves). In fact, it is important to remember that many pathways can generate CO2 such as plastidial and mitochondrial pyruvate dehydrogenase, the TCAP, the γ-aminobutyrate (GABA) shunt, the malic enzyme or cytosolic pentose phosphates (reviewed in Tcherkez et al., 2017b). Articles in this Virtual Issue provide very useful information on the origin of CO2 generated by day respiration. Gauthier et al. (2020) show that under low irradiance, the carbon flux through malic enzyme contributes to Rlight. Under moderate irradiance, 13C labelling and NMR analyses show that the TCAP makes a rather small contribution to day respiratory CO2 efflux and is linked to N assimilation via glutamate synthesis (Abadie et al., 2017), while the contribution of pyruvate dehydrogenase prevails (Lothier et al., 2019), as found previously (Tcherkez et al., 2008). The relatively low contribution of the TCAP to CO2 evolution is not only linked to the capture of TCAP intermediates to feed N assimilation but also related to the down-regulation of TCAP enzyme activity in the light (Tcherkez et al., 2012). This effect is nicely illustrated in Ahmad Rashid et al. (2020) where metabolomics were used to show the opposite behavior of citrate plus aconitate (which is more abundant at night) and other TCAP intermediates (more abundant in the light). Also, in mitochondrial mutants affected in Complex I activity, rates of day respiration are higher, due to pyruvate dehydrogenase activity being higher while anaplerotic bicarbonate fixation by phosphoenolpyruvate carboxylase (PEPc) is lower (Lothier et al., 2019). This raises the question of the influence of PEPc on observed rates of Rlight. That is, Rlight can be viewed as a net efflux resulting from CO2 production by catabolism, refixation (accounted for in calculations reviewed above) and PEPc activity. Direct assessment of PEPc fixation by labelling shows that the flux is moderate under ordinary conditions but can reach 1 µmol m–2 s–1 or more at high photorespiration (Abadie & Tcherkez, 2019). Conversely, the bicarbonate leaf pool could reform CO2 by acid-base equilibration and this could contribute to CO2 efflux. Feeding cut poplar (Populus deltoides) leaves with 13C-bicarbonate shows that the contribution of bicarbonate to Rlight is very small, only up to 20% when bicarbonate is fed with concentrations as high as 11 mM (Stutz & Hanson, 2019). Processes in which respiration plays a role were summarized and hierarchized in ‘core principles’ by O’Leary et al. (2019). In brief, these principles relate to: (1) metabolic functions (ATP and NADH energy generation, carbon skeleton synthesis, redox balancing); (2) metabolic flexibility (alternative routes); (3) regulation with supply (sugars, intracellular ADP regeneration) and demand (such as ATP utilization); (4) effects of plant life cycle (such as carbon investment in growth at different developmental stages); and (5) its roles in acclimation to environmental parameters such as temperature, or tolerance to salinity (Munns et al., 2020). These five principles are relevant to day respiration (as they are to dark respiration). Here, special emphasis can be given to aspect (1) because in leaves, nutrient assimilation takes place in the light and thus the generation of carbon skeletons needed by N and S assimilation is essential (electron utilization for N and S assimilation is further discussed in Tcherkez & Limami, 2019). Mitochondrial metabolism is also important for redox homeostasis since excess redox power and photorespiratory-derived NADH can be consumed by the mitochondria via NADH re-oxidation through the respiratory electron chain (via the alternative and cytochrome pathways). In Dahal et al. (2015) and Dahal & Vanlerberghe (2017), Rlight was estimated using the Kok method during progressive drought in either wild-type plants or alternative oxidase (AOX) knock-down mutants or over-expressors. Interestingly, Rlight was found to decrease under drought but this decline is not observed in over-expressors. In addition, there was a negative relationship between electron pressure on PSII (represented by 1 – qp where qp is photochemical quenching) and Rlight. While the link between these results and the regulation of catabolism (decarboxylations) remains to be elucidated, these studies suggest that there is a beneficial effect of the electron flow via AOX to sustain PSII function, i.e. mitigation of electron pressure in the chloroplast. Similarly, at high irradiance, the engagement of the AOX (measured with 16O/18O respiratory discrimination in the dark) seems to be beneficial for PSII photochemistry across four species (Florez-Sarasa et al., 2016). There is relatively little practical data on how day respiration varies with environmental conditions. In the recent past, the relationship between dark or day respiration, temperature and CO2 has been examined (e.g. Ayub et al., 2014) and surveys across different biomes have been carried out (e.g. Heskel et al., 2016). However, the response of Rlight to temperature, growth CO2 mole fraction and nutrients is not very well known, and furthermore, it is possible that the response to such environmental parameters varies when different techniques are used to measure Rlight. A number of studies on Rdark are included in this Virtual Issue since they provide information on how Rlight may respond to environmental parameters. For instance, Vcmax calculated from Asat at different temperatures requires the knowledge of Rlight; in cases were Rlight estimates are not available, one can use Rdark as a surrogate. Possible uncertainties in doing so are discussed in Atkin et al. (2015). The response of photosynthesis and Rdark to key parameters involved in climate change (temperature and CO2) is thoroughly reviewed in Dusenge et al. (2019). The impact of temperature was also explored in Aspinwall et al. (2016) where daily average air temperature and prevailing leaf temperature were found to have contrasting effects on Rdark. In addition, Aspinwall et al. (2016) provides the surprising result that the Rdark : Asat ratio (at prevailing leaf temperature in each growth treatment) was negatively related to total nonstructural carbohydrates, suggesting that carbon partitioning to respiration changed as A increased. It is worth noting that in the short-term (i.e. minutes to hours), it has also been shown with isotopic labelling that the carbon allocation to TCAP intermediates in illuminated leaves is not proportional to the rate of assimilation, with allocation to the TCAP being lower under conditions that increase assimilation rates (i.e. conditions that lower photorespiration) (Abadie et al., 2018). This Virtual Issue also comprises highly valuable articles that directly address day respiration and include measurements of Rlight under different environmental conditions. Way et al. (2019) looked at the response of day respiration to temperature in forest red gum (Eucalyptus tereticornis) both in the short-term (via rapid changes in leaf measurement temperature) and in the long-term (via a comparison of plants grown at different temperatures). Due to acclimation, there was little effect of growth temperature on Rlight (and Rdark) taking place at the prevailing temperature of each environment, regardless of whether the Laisk or the Kok method was used. There was also little effect of growth temperature on Rlight calculated from A–ci curve fitting (Kumarathunge et al., 2019). In contrast to earlier studies pointing to leaf respiration being less temperature sensitive in the light than in darkness (Atkin et al., 2005, and references therein), recent work reported in this Virtual Issue found that Rlight was more sensitive than Rdark to measured temperature, with higher Q10 values. Interestingly, Rlight obtained with the Laisk method was sometimes negative (net CO2 fixation), perhaps showing the prevalence of PEPc fixation (mentioned above) over catabolic decarboxylation at low temperature (Way et al., 2019). This quite complicated effect of temperature is important to keep in mind, particularly when attempting to predict how rates of nonphotorespiratory CO2 release in the light respond to short- and long-term changes in temperature. When the temperature varies, Rlight is the result of both changes in the degree of inhibition of respiration by light (i.e. the extent to which Rlight differs from Rdark) and thermal acclimation, and thus both phenomena may affect computations of plant, ecosystem or global CO2 net exchange (Heskel et al., 2013; Vanderwel et al., 2015). The effect of neglecting the inhibition of respiration by light has been examined precisely using whole canopy CO2 exchange in forest gum trees by Drake et al. (2016). They found that while autotrophic respiration (Ra) is visibly lower (on average by 21%) when light inhibition is accounted for, gross primary production (GPP) and the Ra : GPP ratio are only impacted by a few percent (and the dependence on temperature is also minimally affected). In addition to temperature, nutrient availability is also a fundamental parameter that dictates respiration rate, and this not only includes N but also phosphorus (P), sulphur (S) or potassium (Tcherkez, 2017). Using a survey across many species of different life forms grown under contrasted N/P conditions, Crous et al. (2017) demonstrated the relationship between Rlight (measured with the Kok method and expressed per leaf surface area) and N and P elemental content (per leaf surface area). It is worth noting that Rdark has also been found to be related to N and P content (per leaf surface area) in a survey of tropical trees, also included in this Virtual Issue (Rowland et al., 2017). The relationship between Rlight and N and P seems to be specific to these nutrients since there is no correlation between Rlight and leaf mass per surface area (Crous et al., 2017), unlike Rdark (Rowland et al., 2017). Interestingly, the dataset of Crous et al. (2017) did not reveal a close relationship between Rlight and Rdark, with Rlight < Rdark in 84% of cases and Rlight > Rdark in 16% of cases. When the dataset is restricted to measurements where Rlight < Rdark, the average Rlight : Rdark ratio is about 0.64 (i.e. 36% inhibition) which falls within the range of values commonly observed. Furthermore, the rather unusual situation where Rlight > Rdark is associated with significantly larger Amax (per unit of mass), lower N elemental content (per leaf surface area) and concerns more herbs than woody species. Collectively, these observations suggest a possible link between nitrogen use efficiency (and/or C : N ratio) and day respiration – a topic that requires further work in the future. In addition, Clemente-Moreno et al. (2020) present an interesting correlation analysis between mitochondrial electron chain capacity via the cytochrome pathway (assayed in the dark using 16O/18O fractionation) and some metabolites (quantified in leaves sampled in the light), using Antarctic pearlwort (Colobathus quitensis) and a temperate counterpart of the same family, rainbow pink (Dianthus chinensis). That study suggests a possible relationship between mitochondrial electron transfer capacity and sulphur (S) assimilation, which is in turn beneficial to antioxidant metabolism (glutathione synthesis) and thus growth at low temperature. Taken as a whole, papers of this Virtual Issue show the critical importance of day respiration for leaf photosynthesis, since it interacts with quantum yield, correlates to N and S elemental content and/or assimilation, and participates in tolerance to oxidative stress. Of course, measuring Rlight remains challenging and most methods have their drawbacks. Importantly, the papers presented in this Virtual Issue have helped in identifying and addressing such problems. Moreover, this Virtual Issue combines papers that deal with modelling, biochemistry and physiological experiments. Such a combination reflects the fact that: (1) coupling different approaches (trans-disciplinarity) is critical to the provision of significant advances; and (2) classical physiology such as gas exchange remains essential and must not be overlooked despite the increasing role played by next-generation omics in plant science.

Topics & Concepts