Where does it come from, where does it go? The role of the xylem for plant CO2 efflux

This article comments on: Stutz SS, Anderson J, Zulick R, Hanson DT. 2017. Inside out: efflux of carbon dioxide from leaves represents more than leaf metabolism. Journal of Experimental Botany 68, 2849–2857.

Not that long ago, the CO 2 efflux from a plant organ was attributed to metabolic reactions and activity only in that same organ. For tree stems, however, we know that CO 2 transported in the xylem and originating from other plant compartments compromises stem respiration measurements. Now, Stutz et al. (2017) have shown that leaf CO 2 efflux, in addition, does not solely represent leaf metabolic processes but is affected by xylem-borne CO 2 .
Knowledge of the processes of carbon (C) allocation and cycling in plants and ecosystems is important for understanding the terrestrial C cycle as a whole and changes as a result of anthropogenic impacts. The potential of ecosystems to sequester C from the atmosphere reflects the balance between C assimilation and the complex set of respiration processes (Trumbore et al., 2013). Even though respiration is likely to be the most important determinant of the C balance in terrestrial ecosystems (Valentini et al., 2000), understanding of the mechanistic background is far from complete. On the one hand, respiration is affected by direct environmental cues that regulate respiratory metabolism (Atkin et al., 2005). However, on the other hand, substrate supply for respiration, which is indirectly coupled to environmental drivers (e.g. via their effects on photosynthesis), controls respiratory CO 2 fluxes (Högberg and Read, 2006). Besides photosynthesis, phloem transport and the partitioning of assimilates among different C pools affect substrate availability for respiration. We need greater acknowledgement that our view of the controls of respiration fluxes is incomplete (e.g. Hagedorn et al., 2016) and that the 'system plant' might be more complex than previously thought.
CO 2 efflux from a given plant organ is not necessarily an indicator of its respiratory activity Only recently, a major new challenge became obvious: CO 2 efflux is not only affected by the catabolic metabolism of a given organ, but also by the respiration rate of plant compartments coupled via the transpiration stream to that organ. Thus, the CO 2 exchange of stems, for example, is influenced by downstream CO 2 production in the roots and the transpiration stream, adding within-plant spatial interconnectivities (Teskey et al., 2008).
While it was previously usually assumed that CO 2 produced in roots and stems escaped these organs more or less directly into the surrounding atmosphere, there is now strong evidence for trees that at least part of this CO 2 remains inside the plant, gets dissolved in the xylem sap, and is then transported acropetally (Teskey and McGuire, 2002;McGuire and Teskey, 2004;Teskey et al., 2008). Whilst parts of this xylem-transported CO 2 escape from stems and branches into the atmosphere, up to almost 20% can be re-assimilated, with the highest contribution made by woody branches (Bloemen et al., 2013).
Using a 13 C-labelling approach, Stutz et al. (2017) have now shown that CO 2 originating from the xylem also contributes to leaf CO 2 efflux, and they were also able to quantify the contribution of xylem-originating CO 2 to this flux (Box 1). Moreover, they determined the retention of CO 2 in the leaf as a result of anaplerotic processes in the dark. While there are now a number of papers reporting analyses of the effect of xylem transport of CO 2 on stem or branch efflux (e.g. Bowman et al., 2005;Gansert and Burgdorf, 2005;Teskey and McGuire, 2005;Maier and Clinton, 2006), there is a lack of information on its contribution to 'leaf respiration'. There is one observation by Stringer and Kimmerer (1993) showing that over 99% of xylem-transported CO 2 is fixed in the light, whereas 80% of the transpired label escaped the leaves in the dark. For a whole tree, based on a mass balance approach, Bloemen et al. (2013) calculated that about 90% of the xylem CO 2 left the tree via the stem and branches before reaching the leaves. Stutz et al. (2017) related leaf efflux of xylem CO 2 ( 13 C-labelled) to 'real' leaf respiration (non-labelled) and showed that flux can approach 50% of the latter depending on transpiration rate and xylem CO 2 concentrations. By taking the effects of light-enhanced dark respiration into account -an Box 1. CO 2 fluxes in plant stems and leaves Up to 50% of root-respired CO 2 is transferred via the xylem stream aboveground (Bloemen et al., 2013) and reaches the stem sapwood (in woody plants). Part of this CO 2 escapes from the stem to the atmosphere, whilst part is assimilated via Rubisco in the photosynthetically active bark and via phosphoenolpyruvate carboxylase. Moreover, (net) uptake of atmospheric CO 2 might occur in green stems. Living tissues in bark (e.g. phloem parenchymatic and companion cells) and wood (e.g. ray parenchyma) as well as the stem cambium produce CO 2 via respiration that might partially escape to the atmosphere, be partially re-fixed, but also dissolve in the xylem sap and thus contribute to xylem CO 2 . Depending on the ratio between the CO 2 transport rate from the roots, stem efflux, stem assimilation and contribution of stem respiration to xylem CO 2 transport, the xylem CO 2 concentration might increase or decrease along the plant axis, but there is no information on acropetal xylem CO 2 gradients available in the literature. The phloem might also transport CO 2 but similarly no published data are available. In the leaves the efflux of leaf-respired CO 2 plus the xylem-derived CO 2 contribute to the total efflux. Stutz et al. (2017) showed that the efflux of xylem CO 2 can approach 50% of the leaf respiration flux, but due to the difficulties of obtaining realistic twig-tip xylem concentrations and our lack of comprehensive data on night-time transpiration, generalizations might be difficult. Part of the xylem CO 2 is fixed in the leaves in the night for anaplerotic reactions. Estimations of day-time respiration fluxes are notoriously difficult. During the day, leaf respiration rates might decrease due to substantial changes in leaf metabolism (Tcherkez et al., 2009), but the contribution of xylem CO 2 efflux might increase as transpiration rates are higher. If considerable amounts of xylem CO 2 are fixed via photosynthesis, leaf-level photosynthetic measurements with classical gasexchange measurement devices might underestimate leaf photosynthetic capacities. Note that the size of arrows does not indicate any quantitative differences in fluxes. experimental artefact occurring when light-acclimated leaves are transferred into darkness and characterized by a burst of respiration exceeding ordinary dark respiration (Azcon-Bieto and Osmond, 1983) -the authors determined a realistic relationship between leaf respiratory CO 2 flux and the CO 2 efflux related to CO 2 transported in the xylem.

Relationships between respiration and environmental drivers
Release of CO 2 in plant tissues away from the original source of respiration does not necessarily affect the measurements of net ecosystem exchange or its separation into photosynthetic and respiratory fluxes. However, it might compromise a proper parameterization of respiratory CO 2 fluxes from plant and ecosystem compartments which is needed to obtain a mechanism-based projection of ecosystem CO 2 fluxes and thus the C balance of ecosystems under changing environmental conditions. For example, it is assumed that the temperature dependency of root and leaf respiration can differ (Atkin et al., 2005). If, however, root-respired CO 2 mixes with leaf-respired CO 2 to form a combined efflux, which is affected by temperature and transpiration rate, such temperature dependencies cannot easily be separated. Moreover, estimates of assimilate distribution within the plant, and above-vs belowground C partitioning, will be wrong if we assume that up to almost 50% of root-respired CO 2 is transferred via the xylem stream aboveground (Bloemen et al., 2013), leading to a strong underestimation of the belowground C demand. Stutz et al. (2017) conducted their measurements in the dark and concluded that mainly due to the low rates of nighttime transpiration the efflux from xylem CO 2 would on average amount to only about 1.5% of leaf respiration. There are, however, different uncertainties concerning this estimate. On the one hand, Resco de Dios et al. (2015) reported that nocturnal transpiration might be generally underestimated as it can contribute to >20% of day-time transpiration and thus the 1.5% might be a rather conservative estimate. On the other hand, there is no reliable information on the CO 2 concentration in the xylem of twig tips or on concentration gradients from the main stem or branches, where a number of measurements exist (see Table 1 in Teskey et al., 2008), to the twigs and petioles. If the assumption of Bloemen et al. (2013) that most of the xylem CO 2 reaches the atmosphere via trunks and stems is correct, the CO 2 concentrations reaching the leaves might be lower than the ones taken into account by Stutz et al. (2017), and thus they might overestimate the effective night-time flux. During the day, transpiration will be higher, but xylem CO 2 concentrations vary inversely with xylem flow rates (Teskey et al., 2008) and at least part of the CO 2 approaching the leaf will be assimilated. Thus, we are far from being able to quantify the contribution of CO 2 produced in roots and stems but released by the leaves. More research is needed to quantify xylem CO 2 concentrations along the plant axis and changing with time, as are direct measurements of the efflux of xylem-borne CO 2 via the leaf over diel timecourses.
Leaf respiration in the light -old issues, new problems During the light period, not only anaplerotic CO 2 fixation, but also photosynthetic assimilation of xylem CO 2 , is likely to occur. Moreover, leaf respiration in the light is notoriously difficult to study. Only recently, Farquhar and Busch (2017) postulated that the Kok effect and the Laisk approach, which are usually applied, are error-prone and not straightforwardly applicable to estimate day respiration. Even though from a biochemical point of view it is wellknown that the TCA cycle is not closed down in the light, and changes in the commitment of major biochemical pathways in the light and during light-to-dark transitions point to a reduction of CO 2 release from glycolysis and the TCA cycle (Tcherkez et al., 2009;Werner et al., 2011), Farquhar andBusch (2017) suggest that it is currently best to assume that dark respiration in the light equals the respiration rate in the dark at the same temperature. If we now take into account that various amounts of xylem CO 2 -depending on the variation of transpiration rate and of CO 2 production, release and re-fixation in heterotrophic tissues -contribute to day-time CO 2 efflux from the leaf, the situation gets even more complicated. Not only the complex and still poorly understood changes in leaf metabolism in the light, but also xylem-mediated teleconnections, will affect the measured CO 2 fluxes. Moreover, net CO 2 exchange measurement at the leaf level -an indispensable tool for plant physiology and ecology -will be incorrect if considerable amounts of the xylem CO 2 are fixed, leading to underestimations of the real CO 2 -fixing capacity of a leaf.

New challenges for understanding respiration -spatial and temporal influences on respiration
There is growing evidence that the CO 2 efflux from aboveground tissues is affected by CO 2 production from distant tissues connected via the xylem, and thus does not solely represent the local metabolism but rather the integrated plant activity weighted by the transpiration rate and thestill not sufficiently understood -assimilation in, and CO 2 efflux from, different plant organs. And as if this situation were not complex enough in spatial terms, recent research indicates that temporal interconnections also affect leaf and canopy respiratory CO 2 fluxes. Gessler et al. (2017) showed that in addition to the current environmental drivers affecting respiration, the antecedent conditions as mediated by the circadian clock were involved. This circadian control is assumed to act as an adaptive memory to adjust plant metabolism based on environmental conditions from previous days and thus adds a temporal component to respiration and its control. All these recent findings indicate that at least parts of our understanding of plant respiration have been too simplistic, but at least we do now have the tools in hand to fully account for spatial and temporal controls.