Carbon isotope fractionation during dark respiration and ... - CiteSeerX

Phytochemistry Reviews 2: 145–161, 2003. © 2003 Kluwer Academic Publishers. Printed in the Netherlands.

145

Carbon isotope fractionation during dark respiration and photorespiration in C3 plants Jaleh Ghashghaie1,∗ , Franz-W. Badeck2 , Gary Lanigan3 , Salvador Nogués1 , Guillaume Tcherkez1 , Eliane Deléens4,† , Gabriel Cornic1 & Howard Griffiths3 1 D´ epartement

d’Ecophysiologie Végétale, ESE, CNRS-UMR 8079, Bât.362, Université de Paris XI, 91405Orsay Cedex, France; 2 Potsdam Institute for Climate Impact Research (PIK), P.O. Box 601203, 14412 Potsdam, Germany; 3 Department of Plant Sciences, Downing Street, University of Cambridge, Cambridge CB2 3EA, United Kingdom; 4 Laboratoire de Structure et Métabolisme des Plantes, IBP, Université de Paris XI, 91405Orsay Cedex, France; ∗ Author for correspondence (Tel: ++ 33 1 69 15 63 59; Fax: ++ 33 1 69 15 72 38; E-mail: [email protected])

Key words: carbon isotope, C3 plants, discrimination, fractionation, isotope effect, photorespiration, respiration

Abstract Carbon isotope discrimination during photosynthetic CO2 assimilation has been extensively studied and rigorous models have been developed, while the fractionations during photorespiratory and dark respiratory processes have been less well investigated. Whilst models of discrimination have included specific factors for fractionation during respiration (e) and photorespiration (f ), these effects have been considered to be very small, i.e. not significantly modifying the net discrimination expressed in organic material. On this paper we consider the fractionation effects associated with specific reactions set against the overall discrimination which occurs during source-product transformations. We review the studies which have recently shown that discrimination occurs during respiration at night in intact C3 leaves, leading to the production of CO2 enriched in 13 C (i.e., e = −6), and modifying the signature of the remaining plant material. Under photorespiratory conditions (i.e. increased oxygen concentration and high temperature), the photorespiratory fractionation factor may be high (with f around +10), and significantly alters the observed net photosynthetic discrimination measured during gas exchange. Fractionation factors for both respiration and photorespiration have been shown to be variable among species and with environmental conditions, and we suggest that the term ‘apparent fractionation’ be used to describe the net effect for each process. In this paper we review the fractionations during photorespiration and dark respiration and the metabolic origin of the CO2 released during these processes, and we discuss the ecological implications of such fractionations.

Introduction Carbon has two stable (non-radioactive) isotopes; the predominant one is 12 C (natural abundance is about 98.9%), and the minor one is 13 C (about 1.1%). During photosynthetic CO2 assimilation, a number of isotope effects (see Appendix for definitions) discriminate against the heavier stable isotope, leading to † This paper is dedicated to Eliane Deléens who sadly passed away in March 2003. She initiated work on discrimination during dark respiration at the University of Paris-XI (Orsay-France). She was a pioneer in the use of stable isotopes in plant ecophysiological studies.

photosynthetic products being depleted in 13 C compared to atmospheric CO2 . The carbon isotope composition (δ 13 C) of plant material varies from −7 to −35, largely depending on photosynthetic pathway (C3, C4 or CAM), with considerable variation in δ 13 C within each of these plant groups. This variation in C3 species is primarily dependent on plant species, anatomical characteristics and environmental conditions (Farquhar et al., 1989; Brugnoli and Farquhar, 2000). Further fractionations also occur during anabolic and catabolic metabolism, leading to different isotopic signatures for biochemical compounds. Thus,

146 for example, lipids are 13 C-depleted compared to carbohydrates (Abelson and Hoering, 1961; Park and Epstein, 1961; Deléens et al., 1984; Gleixner et al., 1993). Variations also occur within a single plant, such that leaves are generally 13 C-depleted compared to all other organs. This difference between organs could be attributed to a possible fractionation during export of assimilates (phloem loading) and/or to a possible fractionation during dark respiration, which may vary between organs. Carbon isotope discrimination () is used to define the net shift in carbon isotope composition during photosynthetic CO2 assimilation in C3 plants, from source CO2 to organic material, and has been extensively investigated and defined by mathematical models (see Appendix, also Vogel, 1980; O’Leary, 1981; Farquhar et al., 1982, 1989). This overall discrimination involves fractionation during both physical and biochemical processes. The complete theory for the net photosynthetic discrimination is described as follows (Farquhar et al., 1982): = ab b

Pa − Ps Ps − Pi Pi − Pc +a + (es + ai ) + Pa Pa Pa

Pc − Pa

eRd k

+ f ∗ , Pa

(1)

where, Pa , Ps , Pi and Pc are the CO2 partial pressures in the ambient air, at the leaf surface, in the intercellular air spaces (gas phase) and in the chloroplasts at the carboxylation sites (liquid phase), respectively. Accordingly, ab is the fractionation during CO2 diffusion in the boundary layer (= 2.9, Farquhar, 1980); a is the fractionation during CO2 diffusion in air through stomata into the leaf (= 4.4, Craig, 1954); es is the fractionation occurring when CO2 is dissolved in the cell solution (= 1.1 at 25 ◦ C, Vogel, 1980); ai is the fractionation during CO2 diffusion in the liquid phase (0.7, O’Leary, 1981); b is the net discrimination by the primary carboxylating enzymes ribulose1,5-bisphosphate carboxylase/oxygenase (Rubisco) and phosphoenolpyruvate carboxylase (PEPc) during carboxylation in C3 plants (b varies between 28.2 and 30, assuming that carboxylation by PEPc varies respectively between 5% and 0%, Brugnoli et al. 1988); e and f denote overall discriminations during day respiration (Rd ) and photorespiration, relative to photosynthetic products, respectively; k is the carboxylation efficiency and ∗ is the CO2 compensation point in the absence of day respiration (Brooks and Farquhar, 1985).

There is uncertainty regarding the magnitude of respiratory and photorespiratory fractionations (terms ‘e’ and ‘f ’, respectively) and, because of difficulties in experimentally determining these values, a simple model is usually used to approximate the entire theory. The simplified model (Farquhar et al., 1982) only includes two main discriminating steps; CO2 diffusion from the air, via stomata, into the leaf air spaces and net carboxylation, and allows us to predict instantaneous discrimination (i ) from measurements of Pi /Pa : i = a + (b − a)

Pi Pa

(2)

According to this model, all factors decreasing stomatal conductance and consequently decreasing Pi /Pa should also linearly decrease photosynthetic discrimination. The simple model is remarkably robust under a range of experimental conditions (see review of Brugnoli and Farquhar, 2000 and references therein), demonstrating a high correlation between Pi /Pa and δ 13 C of the photosynthetic products under varying light intensities, vapour pressure deficit or under water deficit conditions. When the values for ‘b’, determined by in vitro measurements are used, the net photosynthetic discrimination measured on-line in an open gas exchange system often deviates from the values predicted by Eq. 2 (Evans et al., 1986; von Caemmerer and Evans, 1991). This deviation results from (i) a mesophyll conductance component, leading to an additional drawdown in CO2 partial pressure from the substomatal cavity to Rubisco (Pi to Pc ); (ii) uncertainties in the estimation of Pi /Pa because of heterogeneous stomatal closure and/or a substantial cuticular conductance to water vapour under water deficit conditions (for a recent review see Evans and Loreto, 2000) and (iii) the effect of fractionation which may occur during photorespiratory and respiratory processes on the net photosynthetic discrimination, and extent of refixation of respiratory CO2 . During real-time measurements, Evans et al. (1986) showed that the difference between the observed net photosynthetic discrimination measured with an on-line system and the values predicted by the simple model was a linear function of A/Pi (A is the net photosynthetic CO2 assimilation), the slope being related to the mesophyll conductance and the intercept to the (photo)respiratory fractionation (the last term in Eq. 1). Their method is now routinely used for the estimation of leaf mesophyll conductance. Since the intercept is usually small, it is considered

147 that the fractionation during photorespiration (term ‘f ’) and day respiration (term ‘e’) is negligible and does not significantly modify the net photosynthetic discrimination measured on-line, particularly since measurements are usually made under 1–2% O2 to inhibit photorespiration. However, we discuss the implications of photorespiration and day-time respiration and their associated fractionations below. The carbon isotope signature of plant dry matter integrates not only the discrimination during net CO2 assimilation in the light but also the discrimination that could occur during the night-time respiration. Therefore, any fractionation during the night and/or the use of heavy or light substrates for dark respiration (releasing 13 C-enriched or 13 C-depleted CO2 compared to leaf material) should change the isotopic signature of the remaining leaf material. Moreover, when non-photosynthesising organs are taken into consideration, release of 13 C-enriched or 13 C-depleted CO2 will further contribute to changes in the carbon isotopic signature of the whole plant. Henderson et al. (1992) observed on some C4 species that the discrimination determined on leaf dry matter was significantly greater than that measured on-line. Using a modelling approach, they proposed that at least a part of this difference could be explained by the fractionation during dark respiration releasing CO2 enriched in 13 C relative to the plant material. Any possible fractionation during export could also contribute to this difference, but to our knowledge, this has not been experimentally demonstrated. The potential changes of isotopic signatures in plant organic matter due to dark respiratory processes are of relevance for the use of 13 C in ecosystem studies. The models applied for these studies often build on the assumption that the signature of organic matter is determined by the signature of photosynthetic products, i.e. respiratory metabolism during the night does not alter the signature of organic matter left behind. Although Lin and Ehleringer (1997) found no apparent fractionation during dark respiration in mesophyll protoplasts (isolated from mature leaves of C3 and C4 plants) incubated with substrates of known δ 13 C (fructose, glucose or sucrose), Duranceau et al. (1999) Ghashghaie et al. (2001) and Tcherkez et al. (2003) recently showed a substantial 13 C-enrichment in dark respired CO2 compared to leaf plant material even compared to leaf sucrose, the most 13 C-enriched respiratory substrate, on intact leaves of some C3 plants. Sucrose being the potential substrate for dark respiration, they concluded that an apparent fraction-

ation during dark respiration occurs in the species studied. The discrimination during photosynthetic CO2 assimilation has already been recently reviewed (see Brugnoli and Farquhar 2000). In this paper, we focus on the discrimination which occurs during dark respiration and (photo)respiration (i.e., photorespiration and day respiration) in C3 plants. Fractionation during dark respiration What is the ‘fractionation factor’ during dark respiration? In the broadest sense, the discrimination which occurs during respiration can be determined from the release of respiratory CO2 , as compared to the isotopic signature of the reference (source) carbon pool, which we define as an ‘apparent fractionation’ (see Appendix). Individual fractionation factors will be associated with kinetic isotope effects, non-homogeneous isotope distribution in respiratory substrates (positional effects) or oxidation of different substrate pools (for definitions, see Appendix). The respiratory process will also lead to a change in the isotopic signature of the pool left behind. Depending on the question that is being addressed, the reference pool can be the total organic matter, recent photosynthetic products or the substrates actually used for respiration. Because there is currently no consensus defining the reference pool for respiratory processes, in the following, we explicitly define the reference pools we address. In particular, the fractionation factor used in Eq. 1 for respiration (e) relates to the offset in isotopic signal from the primary carboxylase (Rubisco) product during day respiration. While we use the term ‘e’ to indicate fractionation factors associated with both day and dark respiration, it should be noted that values of e quoted herein are for apparent fractionation associated with dark respiration, while the values associated with day respiration remain unknown. We now consider the magnitude of e and whether in practise it should be defined as an apparent fractionation. Carbon isotope signature of CO2 respired in the dark Only a few published studies have investigated the carbon isotope signature of CO2 respired in the dark by plants, and most of them were conducted on seedlings (see Table 1). To our knowledge, the earliest work was done by Baertschi (1953) who showed that CO2

148 Table 1. Carbon isotope signature of CO2 respired in the dark in different plant species compared to that of plant material. 13 C-enrichment (−) and 13 C-depletion (+) in respired CO2 compared to plant material is indicated by negative and positive signs, respectively, in the column ‘fractionation’. Respired CO2 Plant material Plant part δ 13 CR δ 13 CP analysed () () C3 species: Phaseolus sp. (bean) Lycopersicon esculentum (tomato) Curcubita moschata L. (Squash) Arachis hypogaea L. (peanut)

−24.3 −25.7 −23.8 −33.3 Helianthus annuus L. (sunflower) −30.2 −35.9 −26.0 Triticum aestivum L. (wheat) −25.4 −24.0 −25.5 Ricinus communis L. (Castor bean) −27.5 −28.4 Pisum sativum L. (pea) −25.8 −27.8 Raphanus sativus L. (radish) −30.1 Nicotiana tabacum (tobacco) −24.1 Pinus radiata −25.7 Phaseolus vulgaris L. (French bean) −20.2 Nicotiana sylvestris −23.5 Fagus sylvatica L. (beech) −22.1 −26.3 −22.1 −26.3 C4 species: Zea mays L. −13.5 −12.9 −12.0 −15.7 Paspalum dilatatum −14.9 Gomphrena globosa −17.5

Fractionation Reference δ 13 CP -δ 13 CR ()

−32.4 −26.7 −28.5 −27.9 −27.3 −25.6 −28.5 −23.7 −23.5 −30.4 −28.7 −28.7 −26.1 −25.9 −28.8 −25.5 −29.4 −26.0 −26.5 −24.8 −26.2 −22.8 −23.3

Germinating seeds 0 Whole plant −8.1 8-day seedling −1.0 Soaked seed −4.7 32-day seedling +5.4 Soaked seed +2.9 10-day seedling +10.3 Leaf sucrose pool −2.5 Soaked seed +1.7 7-day seedling +0.5 Leaf + ears −4.9 Soaked seed −1.2 12-day seedling −0.3 Soaked seed −0.3 24-day seedling +1.9 4-day seedling +1.3 Leaf starch −1.1 Needles + stem +3.7 Leaf sucrose pool −5.8 Leaf sucrose pool −3.0 Young twigs (May) −2.7 Young twigs (December) +0.1 Twig sucrose (May) −0.7 Twig sucrose (December) +3.0

−12.2 −11.6 −13.4 −13.9 −14.7 −15.7

Soaked seed 13-day seedling Leaf sucrose Leaf Leaf Leaf

respired by germinating bean had the same 13 C/12 C ratio as the whole seed. Later, Park and Epstein (1961) observed that CO2 respired by tomato plants was 13 Cenriched compared to whole plant material by about 8. They also observed, as did Abelson and Hoering (1961), that lipids were 13 C-depleted compared to other metabolites by about 7–8. In agreement with conservation laws, they emphasised that any depletion in 13 C in one chemical component of the plant, such as lipid, must be compensated by enrichment in

13 C

+1.3 +1.3 −1.4 +1.8 −0.2 +1.8

Baertschi (1953) Park and Epstein (1961) Smith (1971) Smith (1971) Smith (1971) Smith (1971) Smith (1971) Ghashghaie et al. (2001) Smith (1971) Smith (1971) Troughton et al. (1974) Smith (1971) Smith (1971) Smith (1971) Smith (1971) Smith (1971) Hsu and Smith (1972) Troughton et al. (1974) Duranceau et al. (1999) Ghashghaie et al. (2001) Damesin and Lelarge (2003) Damesin and Lelarge (2003) Damesin and Lelarge (2003) Damesin and Lelarge (2003) Smith (1971) Smith (1971) Ghashghaie (unpublished) Troughton et al. (1974) Troughton et al. (1974) Troughton et al. (1974)

in some other component. Since respiration and lipid formation are closely related biochemical systems, the 13 C-enrichment in respired CO2 could result from lipid formation. They also suggested, on the basis of their biochemical theory, that in the case of seeds germinating in the dark reported by Baertschi (1953), lipids are not being formed, but are being oxidised. Therefore, the isotopic signature of CO2 respired is expected to be that of seeds (sugars and lipids as a whole). However, if 13 C-depleted lipid is being syn-

149 thesised rather than oxidised, the respired CO2 would be expected to have a higher 13 C content than the whole plant. Similar experiments were conducted later (Smith, 1971; Hsu and Smith, 1972; Troughton et al., 1974) on different species, for which the CO2 respired in the dark was shown to be 13 C-enriched (1–8) or 13 C-depleted (1–10) compared to the source organic material (seed, seedling, leaf, whole plant material or carbohydrates) and varied between species (Table 1). In a review, O’Leary (1981) suggested that at least some of these variations may be due to the source of CO2 used for respiration which might have a carbon isotope composition different from that of whole plant material. One could also suggest that the enrichment or depletion of respired CO2 in 13 C compared to leaf dry matter or leaf carbohydrates may be due to a fractionation during dark respiration. Huge changes in respired CO2 signature were also reported in the above studies during seedling growth, ageing and with dark period duration. Moreover, in sliced potato tubers, the signature of respired CO2 varied with time (Jacobson et al., 1970). The observed changes do presumably reflect a switch in the substrate used for respiration and/or a possible change in the respiratory fractionation. Surprisingly, investigations on the isotopic signature of dark respired CO2 were halted for more than 20 years until Lin and Ehleringer (1997) showed that no apparent fractionation occurred during dark respiration measured in vitro on isolated protoplasts. This result reinforced the initial assumption made in the simple discrimination model, i.e. net fractionation during respiration (term ‘e’ in model, see Eq. 1) is negligible. However, Duranceau et al. (1999) recently observed in both control and dehydrated Phaseolus vulgaris plants (Figure 1A) that the respired CO2 was consistently 13 C-enriched compared with the leaf sucrose pool by about 6, whatever the leaf age and the leaf relative water content. They suggested that, if sucrose (or a closely linked metabolite) is used as the main substrate for dark respiration, a constant apparent fractionation of about −6 occurs in these plants. Ghashghaie et al. (2001) then showed that this varied between species and with drought. Respired CO2 was 13 C-enriched compared to leaf sucrose in average by about 3 in well-watered Nicotiana sylvestris and by 2–6 in control Helianthus annuus (Figure 1B and C, closed symbols). Using an on-line gas exchange system (in normal air), Duranceau et al. (2001) observed

Figure 1. Relationship between carbon isotope composition (δ 13 C) of CO2 respired in the dark and of leaf sucrose for well-watered (full symbols) and dehydrated (open symbols) P. vulgaris (A) N. sylvestris (B) and H. annuus (C) plants. Dashed line corresponds to 1:1 relationship. Data on A are redrawn from Duranceau et al. (1999) and those on B and C from Ghashghaie et al. (2001).

an apparent fractionation value in N. sylvestris which was nearly identical to the values obtained by Ghashghaie et al. (2001) on the same species using a CO2 -free closed system. This indicates that the use of a CO2 free system does not affect the signature of the dark respired CO2 . Contrary to what had been observed in P. vulgaris, the apparent respiratory fractionation increased in dehydrated N. sylvestris and decreased in dehydrated H. annuus compared to control plants (Figure 1, open symbols). Ghashghaie et al. (2001) concluded that (i) carbon isotope fractionation during dark respiratory process is a widespread phenomenon occurring in C3 plants, (ii) this apparent fractionation is not constant and varies among species and also varies with drought, but differently among species

150 (constant in P. vulgaris, increased in N. sylvestris and decreased in H. annuus under drought compared to control conditions). Such a variable apparent fractionation observed during dark respiration is consistent with data in the older literature (see Table 1). Possible causes of the observed overall fractionation during dark respiration Apparent fractionation may be expected during dark respiration because of (i) positional effects such as non-uniform 13 C-distribution within the hexose molecules (Rossmann et al., 1991; Gleixner and Schmidt, 1997) and (ii) isotope effects during the pyruvate dehydrogenase (PDH) reaction (De Niro and Epstein, 1977; Jordan et al., 1978; Melzer and Schmidt, 1987). Effect (i) can lead to 13 C-enriched respiratory CO2 , effect (ii) can result in 13 C-depleted respiratory CO2 . The overall effect will depend, as we are going to discuss below, on the relative activities of different metabolic pathways. (i) Non-uniform 13 C-distribution within hexose molecules Abelson and Hoering (1961) initially suggested that there was a homogeneous 13 C-distribution in hexose molecules. Rossmann et al. (1991) experimentally demonstrated that the C-3 and C-4 of glucose molecules (commercial glucose extracted from sugar beet syrup and hydrolysed from maize flour) were enriched in 13 C compared to other carbon positions which they related to an isotope effect during aldolase reactions. In the course of lipid biosynthesis, the pyruvate issued from glucose by the glycolytic pathway produces 13 Cenriched CO2 by decarboxylation of the 13 C-enriched carbons (C-3 and C-4 coming from glucose). The more depleted sites (i.e., C-1, C-2, C-5 and C-6) form acetyl-CoA which is subsequently oxidised in the Krebs cycle or diverted to biosynthesis of other metabolites (e.g., lipids) (Rossmann et al., 1991). De Niro and Epstein (1977) have already emphasised that, assuming such a heterogeneous 13 C-distribution in glucose molecules, the well-known 13 C-depletion in lipids compared to carbohydrates reported by Abelson and Hoering (1961) and Park and Epstein (1961) thus the 13 C-enrichment in CO2 produced during pyruvate decarboxylation could be explained without recourse to isotope effects during lipid metabolism. By analogy, a constant shift of 1.3 to 1.7 between δ 13 C of ethanol in wine and sugars relative to the ‘must’ has been observed in grape-wine. This fractionation has also

been attributed to decarboxylation of the 13 C-enriched C3 and C4 positions during fermentation (Rossmann et al., 1996). In agreement with Park and Epstein (1961), we suggest that the magnitude of the expected effect will depend on the relative importance of the metabolic pathways in plants. If a high fraction of catabolised carbon is used for lipid biosynthesis (or other substances issued from acetyl-CoA), this should lead to a high enrichment in 13 C of released CO2, conversely if catabolised carbon is completely respired, all ‘light’ as well as ‘heavy’ carbon atoms of sugars will be decarboxylated. This pattern may explain the variability in the values of δ 13 C of respired CO2 reported in the literature (Table 1). This may also explain the results of Lin and Ehleringer (1997): if, for isolated protoplasts, all the carbon atoms (heavy and light) from sugars are consumed in the Krebs Cycle and no carbon is directed towards the biosynthesis of metabolites, such as lipids, the signature of overall dark respired CO2 would then be the same as the signature of the sugars fed to the protoplasts. (ii) Isotope effects during the pyruvate dehydrogenase (PDH) reaction 13 C-depletion in lipids and subsequent 13 C content in respired CO2 may also result from a kinetic isotope effect during decarboxylation of pyruvate. This possibility has already been discussed by Park and Epstein (1961), De Niro and Epstein (1977), Rossmann et al. (1991), Gleixner et al. (1993) and Schmidt and Gleixner (1998). Microorganisms provide a simple system to identify the biochemical steps during which fractionation occurs, leading to 13 C-depletion in lipids (Abelson and Hoering, 1961; De Niro and Epstein, 1977; Melzer and Schmidt, 1987). Using glucose, pyruvate or acetate as growth substrate for Escherichia coli, De Niro and Epstein (1977) showed, as did Abelson and Hoering (1961), little or no fractionation during glucose metabolism to pyruvate (Embden Meyerhof pathway) but found a substantial fractionation by about 7–8 during lipid formation from pyruvate. In addition, positional isotope effects of 1.021 and 1.003 on the C-2 and C-3 of pyruvate, corresponding to differences in δ 13 C of 18, have been demonstrated using E. coli PDH enzyme in vitro (Melzer and Schmidt, 1987). As a consequence, acetyl-CoA and hence fatty acids are 13 C-depleted compared to carbohydrates, as reported above. Similarly, using yeast pyruvate decarboxylase, De Niro and Epstein (1977)

151 observed a 13 C-depletion in acetaldehyde compared to supplied pyruvate, which was largely (7.5) confined to the carbonyl carbon atom of pyruvate (C-2) and only slightly (1) to the methyl group (C-3). An additional isotope effect of about 1.009 was also reported on C-1 of pyruvate (the carbon atom producing CO2 during PDH reaction) using E. coli PDH (Melzer and Schmidt, 1987). Therefore, CO2 released during PDH reaction is expected to be 13 C-depleted. This effect could, however, be masked by the abovementioned 13 C-pattern in the glucose molecules. De Niro and Epstein (1977) showed experimentally, that the magnitude of the expected isotope effect depended on the proportion of pyruvate decarboxylated by yeast enzyme in vitro. Since PDH reaction is one of the several reactions at a branching point, they suggested that other reactions involving pyruvate as reactant modulate the fraction of pyruvate oxidised to acetyl-CoA and consequently the isotope effect during PDH reaction. A similar pattern was also recently shown by Schmidt’s group (for a review see Schmidt and Gleixner, 1998). Park and Epstein (1961) observed a negative correlation between 13 C-depletion in the lipid fraction and the amount of lipid across several plant species and algae which they regarded as consistent with this feature. Besides, the isotope effect of PDH reaction will depend on the relative contribution of the enzymesubstrate binding step (i.e., equilibrium isotope effect thus 13 C accumulation in the enzyme-substrate complex compared to the substrate) and the decarboxylation step itself (i.e., kinetic isotope effect thus 13 Cdepletion in the products). Therefore, the overall isotope effect and thus the released CO2 will have the signature of the limiting step. A similar pattern (including both rate limitation and corresponding isotope effects, leading to either enriched or depleted signals) was previously demonstrated for yeast pyruvate decarboxylase (Jordan et al., 1978). Accordingly, 13 C-enrichment (or 13 C-depletion) in respiratory CO2 relative to carbohydrates is expected to be variable among different species, to change with plant development and with environmental conditions as well as relative activities of different metabolic pathways. Metabolic origin of δ 13 C of dark respired CO2 and its variability In order to confirm the hypothesis that changes in metabolic rates may induce changes in signature of

dark respired CO2 , Tcherkez et al. (2003 and unpublished data) conducted experiments on intact leaves of C3 species under varying leaf temperatures. They argued, based on the metabolic theory of Park and Epstein (1961), that increasing leaf temperature should increase the rate of oxidation of lipids rather than their biosynthesis and, consequently, the light carbons will be consumed in the Krebs Cycle, so decreasing the δ 13 C of overall respired CO2 . By connecting a CO2 -free, closed system, to the GC column of the elemental analyser, and using a small sample loop, on-line measurements of δ 13 C of dark respired CO2 were made during changes of temperature on an intact leaf. Since the specific substrate used for respiration under varying temperatures was unknown, the overall respiratory fractionation in the dark was calculated, as did Farquhar et al. (1989) for the term ‘e’, relative to the isotopic signature of leaf total organic matter (see the legend of Figure 2). As expected, for all the species studied dark respired CO2 was 13 C-enriched compared to leaf material (i.e., ‘e’ is negative) while the 13 C content linearly decreased with increasing leaf temperature, thus with increasing respiration rate (Figure 2). The slope was almost the same for P. vulgaris and N. sylvestris but different and again more variable for H. annuus. One can hypothesise that the changes in δ 13 C of respired CO2 with increasing leaf temperature observed by Tcherkez et al. (2003) is a direct effect of the temperature on the temperature-dependent kinetic isotope effect of the PDH reaction. Indeed, a substantial decrease in the isotope effect with increasing temperature has already been reported during decarboxylation reactions, e.g., a decrease in the isotope effect by about 4 over the range 15–35 ◦ C during pyruvate decarboxylation by yeast PDC (De Niro and Epstein, 1977) and by about 10 over the range 25–37 ◦ C during decarboxylation of arginine by bacterial ADC (O’Leary, 1980). In fact, as temperature is lowered, the decarboxylation step becomes more and more rate limiting and thus the isotope effect of the overall reaction increases (O’Leary, 1980). In order to avoid any temperature effect, Tcherkez et al. (2003) conducted experiments at constant leaf temperature but under a continuous dark period on intact P. vulgaris leaves. For a given temperature, the 13 C-content in respired CO decreased with the dur2 ation of the dark period, together with a decrease in leaf carbohydrate content, suggesting that carbohydrate starvation under continuous darkness induced a switch in the substrate used, from carbohydrates to

152

Figure 3. Relationship between δ 13 C of CO2 respired in the dark and respiratory quotient (RQ) redrawn from Tcherkez et al. (2003). Closed symbols were obtained on P. vulgaris intact leaves at either different leaf temperatures (10, 20 or 30 ◦ C) or at varying dark period length for a given temperature. Open symbols are from the literature on different plant species (James, 1953; Park and Epstein, 1961; Smith, 1971). The linear regression does not take into account data from the literature. The regression equation is: y = 16.57 x − 37.62 (r2 = 0.87).

Figure 2. Relationship between carbon isotope fractionation (e) during dark respiration and respiration rate in P. vulgaris (A) N. sylvestris (B) and H. annuus (C) intact leaves under varying leaf temperature (10, 20 and 30 ◦ C) (Tcherkez and Ghashghaie, unpublished data). ‘e’ is calculated according to Farquhar et al. (1998) as follows: e = (δ 13 C leaf organic matter – δ 13 C respired CO2 )/(1 + δ 13 C respired CO2 ).

more 13 C-depleted substrates such as lipids or proteins (Tcherkez et al., 2003). As it has been earlier proposed by Smith (1971), this should be confirmed experimentally by measuring the respiratory quotient (RQ). Since RQ is the ratio of CO2 production to O2 consumption, it is dependent on the state of oxidation of the substrate used for respiration. Thus from different types of metabolic oxidation (e.g., lipid or carbohydrate oxidation, or gluconeogenesis) emerge different RQ values (around 0.6, 1, 0.4, respectively) and a change in RQ may indicate a switch in respiratory substrate. Only about 30 years after Smith’s recommendation, Tcherkez et al. (2003) measured simultaneously isotopic signature of respired CO2 and RQ on P. vulgaris leaves under varying leaf temperatures and under

continuous darkness. Interestingly, they obtained a linear relationship between the two parameters for both temperature and continuous dark experiments (Figure 3) confirming that changes in the isotopic signature of respired CO2 originate from substrate switching. Indeed, RQ values close to 1 indicating highly oxygenated substrates (e.g. carbohydrates) are observed for a high 13 C content in respired CO2 and RQ values around 0.6 indicating weakly oxygenated substrates (e.g., fatty acids) are observed for a low 13 C content in respired CO2 (Figure 3). In agreement with the earlier statements of Park and Epstein (1961) and De Niro and Epstein (1977), Tcherkez et al. (2003) proposed that there are two main origins of metabolic CO2 sources: one 13 Cenriched from pyruvate decarboxylation and the other 13 C-depleted from acetyl-CoA degradation through Krebs cycle. The imbalance between these two sources may be responsible for the prevalence of 13 C in overall respired CO2 . Indeed, when carbohydrates are degraded (RQ around 1) and acetyl-CoA (light carbons) are used for anabolic pathways (e.g., lipid biosynthesis) the isotopic composition of respired CO2 should be close to the mean value of C-3 and C-4 in glucose molecules (e.g. for the glucose values reported by Rossmann et al., 1999, this would correspond to −21). By contrast, when lipids are degraded (RQ around 0.6) acetyl-CoA is produced by oxidation of fatty acids, thus the isotopic signature of respired CO2 should be close to the mean value of C-1, C-

153

Figure 4. Predicted δ 13 C of assimilates remaining in the leaf calculated using a net photosynthetic discrimination () of 18 and a dark respiratory fractionation (e) of 4 for R/A ratios of 0.025 (continuous line), 0.05 (dotted line), 0.1 (dashed line) as a function of night-length, where R is average night-time respiration rate and A is average assimilation rate per unit time.

2, C-5 and C-6 of glucose about 6 more negative than the C-3, C-4 positions. Using a metabolic model, based only on the 13 C pattern in hexose molecules (Rossmann et al., 1999) and on the isotope effects of PDH measured in vitro (Melzer and Schmidt, 1987), Tcherkez et al. (2003) showed that the observed variation range of δ 13 C of respired CO2 (between −20 and −30) fitted well with the predicted interval. They concluded that the isotopic signature of dark respired CO2 in C3 plants is not constant and is determined by (i) the carbon source used for respiration, i.e., the relative metabolic activities in the cell, (ii) the non-statistical carbon isotope distribution in glucose molecules and (iii) by possible isotope effects of respiratory enzymes. Respiratory fractionation in the wider sense of the definition can be the result of any of the three processes discussed above. The studies with concurrent measurement of the isotopic signature of respired CO2 and the RQ (Tcherkez et al., 2003) showed that a change in respiratory substrates contributes to the overall dark respiratory fractionation of leaves under changing temperature regimes and under carbohydrate starvation. However, the δ 13 C of CO2 respired by leaves in the dark was also enriched relative to δ 13 C of the main potential respiratory substrates when respiratory CO2 was sampled in the course of a night of normal length and at temperatures close to the day-time temperatures (Duranceau et al., 1999; Ghashghaie et al., 2001; Tcherkez et al., 2003). These results indicate that apparent frac-

tionation relative to the respiratory substrates occurs frequently in leaves of C3 plants. In conclusion, it appears that night-time leaf respiratory fractionation will often lead to the release of carbon that is isotopically heavier than day-time assimilates and shift the overall isotopic signature of the assimilates remaining in the plant towards more negative values, in accordance with the model proposed by Henderson et al. (1992). The expected ranges of this effect are illustrated in Figure 4. On longer time scales, variation of night-time leaf respiratory δ 13 C of several per mil can be expected when changes in environmental conditions or plant ontogeny induce switching between substrate classes. However, mitochondrial respiration is inhibited in the light even at light levels as low as 3 µmol Photon m−2 s−1 ). This light inhibition of mitochondrial respiration was shown to vary between 16% and 77% (Atkin et al., 1998, 2000). Therefore the relative effect of respiratory fractionation will differ in the light. Ultimately, these changes in apparent fractionation will reflect systematic rules associated with positional effects and rate-limiting steps provided that the biosynthetic pathway is under steady state conditions. Kinetic isotope effects contribute primarily to isotopic patterns of natural compounds, particularly for irreversible enzyme steps connected to metabolic branching events, substrate effects and variations in metabolic pathway. Increased understanding of the detailed processes involved, and the contributory isotope signals, will allow prediction and modelling to show how individual isotope effects are integrated into the apparent fractionation factor, as measured, and contribute to the overall discrimination expressed during photosynthetic and respiratory metabolism.

Fractionation during photorespiration Carbon isotope signature of CO2 produced during photorespiration and causes of fractionation Discrimination during photorespiration is primarily associated with enzymatic fractionation. In the case of oxygen isotopes, discrimination against the heavier isotope (18O) can occur during oxygenation by Rubisco, by glycolate oxidase or the Mehler reaction (Guy et al., 1993). However, the specific effects of carbon isotope discrimination are found mainly during decarboxylation processes (Jordan et al., 1978; Ivlev, 2001). Unlike dark respiration, however, there

154 13 C

Figure 5. Schematic diagram of the effects of carbon isotope discrimination during assimilation and photorespiration in vivo.

are no positional or branch-specific effects, as glycine substrate pools have a high turnover rate (Parnik et al., 1972) and in vitro studies have demonstrated that fractionation effects are primarily associated with glycine decarboxylation (Jordan et al., 1978; Ivlev, 2001; Ivlev et al., 1996, 1999). Therefore, photorespiration will discriminate against 13 C and photorespired CO2 should, in theory, be depleted in 13 C, leaving the substrate pools enriched. This is analogous to Jordan et al. (1987), where CO2 liberated during decarboxylation of pyruvate was depleted in 13 C, which is typical for such a rate-limiting step, as discussed above. Glycine decarboxylations in isolated mitochondria from several plant species showed large shifts in the isotopic composition of CO2 evolved (Ivlev et al., 1996, 1999) with photorespired CO2 fluctuating between enrichment in 13 C (by up to 8) to being 13 C-depleted by as much as 16. Again, the extent and direction of the fractionation effect was highly dependent on reaction conditions, with the variations in fractionation due to alterations in pH and enzyme co-factors which impacted on the rate-limiting stage of the reaction (Ivlev et al., 1999; Igamberdiev et al., 2001). The measurement of leaf 13 C in a glycine decarboxylase (GDC)-deficient barley (Hordeum vulgare L.) mutant has also indicated that photorespiratory fractionation takes place in vivo (Igamberdiev et al., 2001). That photorespiratory carbon isotope fractionation leads to production of CO2 depleted in

was supported by the analysis of oxalates, formed via photorespiration, which were enriched in 13 C (Raven et al., 1982). Further studies on photorespiratory effects have revealed that net instantaneous discrimination at leaf-level, measured in real time during gas exchange, is reduced under conditions promoting increased relative rates of photorespiration (Gillon and Griffiths, 1997; Gillon, 1997; Lanigan and Griffiths, unpublished data). Therefore, the final isotopic composition of CO2 retro-diffusing during gas exchange will be determined by the mixing of both 13 C-depleted CO2 produced from photorespiration with 13 C-enriched CO2 left from carboxylation. In parallel with this, during carboxylation by Rubisco, 13 C-depleted carbon will contribute to carbohydrate pools, whereas during oxygenation, i.e. the glycolate cycle, and subsequent decarboxylations will result in the formation of 13 C-enriched carbon compounds (Ivlev, 2002, see Figure 5). Ultimately, if the serine returning as phosphoglycerate feeds back into the regeneration of RuBP, then any enrichment in 13 C will be incorporated into Calvin cycle intermediates. What is the ‘fractionation factor’ during photorespiration? As stated above, net photosynthetic discrimination has been robustly modelled and takes into account associated kinetic and enzymatic fractionations, including photorespiratory fractionation (Eq. 1). Thus we can use the simplified model of Farquhar et al., 1982 (see Eq. 2) to predict photosynthetic discrimination (i ) as compared to that measured in real-time during photosynthesis (obs), in CO2 leaving a gas exchange cuvette. The extent that photorespiratory fractionation will alter the final net discrimination value is the product of both the rate of photorespiration (defined by ∗ /Pc ) and the discrimination during photorespiration (represented by f, the photorespiratory fractionation factor). However, the size of this fractionation has been the cause for some debate (see Table 2). Troughton et al. (1974) first directly estimated fractionation associated with photorespiration (f ) to be between −1.6 and −0.2 by collecting photorespired air from a stream of CO2 -free air. However, these values for photorespiratory fractionation were later questioned (Farquhar et al., 1982), since CO2 leaving the leaf may have been subjected to partial re-assimilation by Rubisco, which would have led to 13 C enrichment of CO2 .

155 Table 2. Variation in the estimated values of the photorespiratory fractionation factor (f ). Species

Estimate of f Reference

P. sativum/S. oleracea −8 to (mitochondria) +16 T. aestivum +2 (minus refixation) P. vulgaris +0.5 (minus refixation) T. aestivum/P. vulgaris +8 (including refixation) Glycine max +7 S. cineraria/S.squalidis +9 S. greyii

+11

Ivlev et al. (1996) Gillon and Griffiths (1997) Gillon and Griffiths (1997) Gillon (1997) Rooney (1988) Lanigan and Griffiths (unpublished data) Lanigan and Griffiths (unpublished data)

Whether the extent that day respiratory and photorespiratory (i.e. (photo)respiratory) effects are manifested in the long-term plant biomass isotopic composition is debatable, these fractionations have been shown to comprise a significant component of net observed instantaneous discrimination (Gillon and Griffiths, 1997; Gillon et al., 1998). This was first revealed when i was compared with obs for Piper aduncum, under high-temperature conditions in the field in Trinidad. There was a breakdown in the normal positive correlation between i − obs (the difference between theoretical photosynthetic discrimination and overall observed discrimination in a gas exchange system) and increasing A/Pa (assimilation rate normalised to ambient CO2 partial pressure) (Gillon, 1997; Gillon et al., 1998; Harwood et al. 1998; Griffiths et al., 1999). This was a direct relationship between ambient temperature and obs during the measurement periods due to an increase in the rates of (photo)respiration relative to assimilation, with decreases in obs independent of changes in A/Pa . This temperature effect was later confirmed for Phaselous vulgaris; an increase in temperature from 22 ◦ C to 31 ◦ C was observed to elicit a breakdown in the i − obs versus A/Pa relationship (Lanigan and Griffiths, unpublished data). Subsequent studies have attempted to quantify the effects of photorespiratory fractionation during online discrimination, by estimating the value of f from the effects on the observed discrimination (obs ) of retro-diffused CO2 from leaves exposed to different

O2 partial pressures (Gillon and Griffiths, 1997; Gillon, 1997; Lanigan and Griffiths, unpublished data). In these studies, the discrimination effects of photorespiration (and hence estimation of f ) were assessed from the effects on the relationship between i − obs and A/Pa elicited by changing O2 partial pressure (pO2 ). Results are shown for Senecio vulgaris in Figure 6. Increasing pO2 from 20 mbar to 300 mbar increased photorespiration, as scaled by ∗ (Figure 6A). When compared to measurements made under nonphotorespiratory conditions (i.e., low O2 partial pressure), i − obs for a given value of A/Pa increased at higher pO2 (see Figure 6B). This is a result of decreasing obs at higher relative rates of photorespiration, caused by an increased proportion of 13 C-depleted CO2 in the total CO2 retro-diffused from the leaf. Net discrimination (obs ) is the sum of gross carboxylation, day-time respiration and photorespiration effects. The contribution from day-time respiratory discrimination (the product of appropriate apparent fractionation factor and rate of day-time respiration -eRd /V c ) and photorespiratory discrimination (the product of apparent photorespiratory fractionation and rate, - f ∗ /Pc ) can be corrected. This results in the convergence of the i − obs versus A/Pa relationships for all pO2 treatments, as discrimination now solely represents photosynthetic fractionation. However, day-time respiratory rate is thought to be relatively low, so respiratory discrimination makes only a small contribution to the overall fractionation effects, especially at high assimilation rates (A/Pa > 1.5 mol m−2 s−1 bar−1 ). Hence, observed non-photosynthetic fractionation effects are almost exclusively due to photorespiratory discrimination, which remains a constant proportion of assimilation. Therefore, convergence will occur for a given value of f, since photorespiratory discrimination is defined as the product of photorespiratory fractionation and flux (f ∗ /Pa ). Convergence of i − pr versus A/Pa (where pr is defined as gross photosynthetic discrimination accounting for refixation, see below) is illustrated for Senecio vulgaris in Figure 6C. The equations used to account for these (photo)respiratory effects and estimate f are based on the instantaneous discrimination equations of Evans et al. (1986). Gillon and Griffiths (1997) modified these to include the above terms for respiratory and photorespiratory discrimination in order to estimate f for Phaselous vulgaris and Triticum aestivum. This value was observed to be species-specific (see Table 2). However, the equation of Gillon and Griffiths (1997) assumed

156 for refixation of a portion of the photorespired CO2 resulted in an increase in the estimates of f for both species as only a portion of photorespired CO2 was assumed to leak out of the leaf (Gillon, 1997; Griffiths et al., 1999, see Table 2). With these assumptions, values of f were consistent between species and also agreed with previous estimates made in Glycine max (Rooney, 1988). Current studies on both shrubby and herbaceous species of Senecio also observed a 7–9 shift in the estimate of f, depending on how the CO2 fluxes were modelled. Estimates accounting for refixation were similar to those of Gillon (1997) (see Figure 6C for estimates for S. vulgaris), although values of f were not consistent across species, being higher for the shrubby species Senecio greyii (Lanigan and Griffiths, unpublished data).

Ecological implications of fractionation during dark respiration and photorespiration

Figure 6. The effect of changes in pO2 (oxygen partial pressure) on gas exchange and isotope discrimination characteristics for Senecio vulgaris (Lanigan and Griffiths, unpublished data). (A), The compensation point in the absence of respiration ( ∗ ); (B), the relationship between i − obs (the difference between calculated and measured discrimination) and A/pa (the ratio of assimilation/ambient CO2 partial pressure); (C), the relationship between i − pr (the difference between calculated and net photosynthetic discrimination including refixation) and A/pa (the ratio of assimilation/ambient CO2 partial pressure). Symbols denote pO2 treatments at 20 mbar (grey symbols), 210 mbar (white symbols) and 300 mbar (black symbols).

total retro-diffusion of all photorespiratory CO2 , so the fractionation effects observed were assumed to represent those for the total photorespiratory rate. However, refixation of respiratory CO2 has been demonstrated to occur in several studies and functions as a means to maintain an electron sink under adverse conditions (Gerbaud and Andre, 1987; Bort et al., 1996; Loreto et al., 2001). Hence, further modifications to account

Respiration is a key component of net ecosystem and global gas exchange, with much effort currently being directed towards resolving CO2 fluxes, and specifically regional carbon sinks (Ciais et al., 1995a; Yakir and Wang, 1996; Buchmann et al., 1998; Ciais and Meijer, 1998; Ciais et al., 1999; Bowling et al., 2001). These estimates are dependent on measurements of ecosystem discrimination (E ) as well as regional [CO2 ] in order to determine the magnitude of the terrestrial carbon sinks. In turn, ecosystem discrimination (E ) represents the net effect of both vegetation and soil processes over an integrated period of time and is defined (Lloyd and Farquhar 1994) as: E =

δ 13 CT − δ 13 CR , 1 + δ 13 CR

(3)

where δ 13 CT and δ 13 CR are the isotopic composition of tropospheric and ecosystem-respired CO2 . Keeling (1958) predicted that the integrated δ 13 CR of all respiring components could be estimated via the intercept of the regression of δ 13 CR and 1/[CO2 ]. However, the extent to which δ 13 CR varies both temporally and spatially is poorly understood and this could alter conclusions about the nature of the terrestrial carbon sink (Bowling et al., 2002). In particular, the soil respiratory component of δ 13 CR requires further study. While the largest variations in δ 13 CR are mainly due to the dominant vegetation type (Buchmann et al., 1998), other causal factors can induce spatial and temporal variations in δ 13 CR of up to 10 within

157 pure C3 stands (Buchmann et al., 1997). Site history must be taken into account if there has been a transition between C3 and C4 species growing on the area in question (Buchmann et al., 1998). Changing environmental conditions have been shown to cause large variations in δ 13 CR via effects on assimilation rates and stomatal conductance. Ekblad and Hogberg (2001) demonstrated that alterations in assimilation rate caused by changing air humidity were linked to changes in δ 13 CR that occurred one to four days later and comprised up to 65% of the soil respiration signal, while Bowling et al. (2002) have also found a strong link between δ 13 CR and changes in vapour pressure deficit. However, respiratory fractionation effects between canopy and soil are debatable. Fractionation due to microbial respiration was initially believed to be significant (Blair et al., 1985), although recent studies have estimated rhizosphere respiratory fractionation at less than 1 (Ekblad and Hogberg, 2000; Ekblad et al., 2002). It is not clear whether the respiratory fractionations described in the previous sections translate into a significant component at canopy and ecosystem level, but distinguishing between autotrophic and heterotrophic contributions to soil respiration is a key aim at present. Current approaches using flux partitioning of 13 CO2 have discounted any fractionation associated with respiration per se (Bowling et al., 2001). Additional measurements are required to determine whether the fractionations described for P. vulgaris, N. tabacum and H. annuus (Duranceau et al., 1999; Ghashghaie et al., 2001, Tcherkez et al., 2003) alters soil respiratory efflux, particularly those associated with changing environmental conditions and whether this affects NEE partitioning. Finally, we note that whilst the 18 O signal in CO2 does provide a more definitive means to separate soil from canopy-based gas exchange processes (due to the evaporative enrichment in leaves), this technique has not yet been widely applied at canopy and ecosystem scales, in contrast to global modelling (Ciais et al., 1995a, b, 1999; Buchmann et al., 1997; Brooks et al., 1997). The effect of photorespiratory discrimination on the overall carbon isotope signature of plant biomass is unclear. Ivlev et al. (1999) has suggested that isotopic differences in δ 13 C of leaves, stems and seeds of Triticum aestivum are due to differences in the relative contribution of photorespiration at the stages of organ formation. Also, leaves of GDC-deficient Hordeum vulgare have been observed to be enriched

in 13 C relative to wild-type plants (Igamberdiev et al., 2001). However, photorespiratory processes are generally transient and during leaf expansion very little structural carbon will be photorespiratory in origin. As a result, it is unlikely that photorespiratory discrimination contributes significantly to variation in the carbon isotopic composition of plant biomass. Finally the fractionation associated with respiration in the light (Rd ) is unknown. This light-mediated inhibition is thought to be due to rapid light inactivation of key enzymes such as pyruvate dehydrogenase complex and NAD+ -malic enzyme (Hill and Bryce, 1992). While fractionations associated with Rd will have a negligible effect on organic matter or carbohydrate carbon isotope composition, there may be detectable effects in retro-diffused CO2 , which will vary depending on species and environmental conditions. In studies on snow gum, for instance, temperature has been shown to have a large effect, invoking high levels of inhibition (97%) at 30 ◦ C and high irradiance, but with inhibition being almost totally alleviated under low (6 ◦ C) temperatures (Atkin et al., 2000). Teasing apart the competing fractionations associated with photorespiration and light respiration and, hence, obtaining a value for e in the light is technically extremely difficult however, and little work has been done to address this problem. Photorespiratory fractionation does make a significant contribution to net instantaneous discrimination. Therefore, in order to interpret day-time Keeling plots, and thus deconvolute day-time carbon fluxes, the contribution of the photorespiratory component will need to be assessed. This would then enable researchers to scale up changes in atmospheric CO2 temporally from the current daily assessments to changes occurring across a whole season.

Conclusions Whilst the more complicated model of discrimination (Eq. 1) accounts for the effects of fractionation during (photo)respiration on resultant plant organic material, there are several important points to be considered. Firstly, there is uncertainty as to the magnitude and definition of the specific fractionation factors e and f ; secondly, any apparent fractionation expressed during photosynthesis or dark respiration may vary with environmental conditions, and thirdly, the impact of such fractionations may be significant in real time as we try

158 to assess (photo)respiratory exchanges in the global context of climatic change. We have used a variety of techniques to resolve the magnitude of fractionation processes, from direct isotopic analysis of released CO2 (and associated mass balance effects on residual C pools) to modelling the discrimination effects instantaneously during gas exchange. In general terms, fractionation during consumption of sucrose produces CO2 which is enriched in 13 C by some 6, such that the fractionation factor e = −6, meanwhile, fractionation expressed during photorespiration tends to oppose this process, with CO2 released being depleted in 13 C (i.e., f = +8 to +11). However, overall fractionation during dark respiration is dependent on species and environmental conditions. The value expressed reflects a combination of three processes, namely the carbon source being utilised, specific positional effects as well as additional enzymatic fractionation. For photorespiration, the fractionation factor is to some degree speciesspecific, but the value expressed is also dependent on stomatal conductance and the magnitude of refixation of photorespired CO2 . Indeed, we should conclude that in contrast to specific fractionation factors which can be allocated to diffusion and dissolution of CO2 , or carboxylation by Rubisco or PEPC (see discussion of Eqs 1 and 2 above), the apparent fractionation for respiration and photorespiration are variable. Thus it is intriguing that temperature does not alter the fractionation expressed by Rubisco (O’Leary, 1981), allowing one fairly consistent value to be applied throughout. During dark respiration, there are interactions between specific enzymatic isotope effects for multiple reactions and source-specific substrate effects, which are also dependent on temperature and other environmental constraints. Thus, we now define the resultant ‘apparent’ fractionation to represent a combination of these interacting processes. In a similar way, the apparent fractionation expressed during photorespiration in not simply a product of the isotope effect for glycine decarboxylase, determined in vitro. How important are these fractionations quantitatively? There is often an excellent agreement between gas exchange-derived measurements of Pi /Pa and the predicted or measured organic material carbon isotope composition (using Eq. 2). Indeed, this may in part reflect the way that respiratory and photorespiratory fractionations would effectively cancel each other out, provided that the respective isofluxes (i.e. sum of fractionation and

associated gaseous flux) were quantitatively similar. However, the use of the isoflux terminology brings us to one of the most important current applications of stable isotopes, namely in trying to resolve the contrasting interplay between photosynthetic and respiratory fluxes at the global level. Here, understanding CO2 exchanges at leaf, soil and canopy level and their relationship with the inter-annual changes seen seasonally at the global level (particularly in the northern hemisphere), are dependent on resolving respiratory inputs across each of these levels. We now need to conduct additional measurements to determine the way that respiratory and photorespiratory overall fractionations translate into altered organic carbon isotope signals in plant material, particularly under elevated CO2 for the future, or the high photorespiratory conditions pertaining at the last glacial maximum, when CO2 levels were much lower than at present. Additionally, the way that fractionation during dark respiration could alter carbon isotope signal of carbon exported from the canopy, when released some four or five days later by roots, is another fundamental application requiring additional experimentation. In conclusion, the provision of more definitive values for fractionation factors associated with (photo)respiration and dark respiration, as described in this review, will allow us to explore their implications at global levels, whilst additional work is required to resolve the interplay between compound-specific positional effects and fractionation at the enzymatic, biochemical and molecular levels. Acknowledgements Gary Lanigan and Salvador Nogués acknowledge the financial support provided through the European Community’s Human Potential Program under contract HPRN-CT-1999-00059, NETCARB. We also acknowledge the referees for their valuable contribution to the drafting of this paper. Appendix Definitions Isotope ratio: R is defined as the molar ratio of the heavy to light isotope e.g. for carbon, R = 13 C/12 C. Isotope effects: When physical and chemical processes modify the isotope ratio of product compared to substrate (source), there is either an enrichment or depletion of the heavier isotope. Kinetic isotope effect: For irreversible reactions the kinetic isotope effect, αk , is defined as the ratio of the

159 rate constants for the molecules containing (k12 and k13 , respectively) as follows:

12 C

and

13 C

αk = k 12 /k 13

(A1)

For CO2 diffusing in air, α is 1.0044. When the source is large enough to be not appreciably affected by product formation, then the kinetic isotope effect is: αk, = Rs /Rp

(A2)

where, Rs and Rp are the isotope ratios of the source and of the product. Equilibrium isotope effect: For reversible reactions, an equilibrium isotope effect, (αeq ) is the ratio of equilibrium constants for the compounds containing 12 C and 13 C (K12 and K13 , respectively) as follows: αeq = K 12 /K 13

(A3)

Fractionation factor: Because isotope effects are very small, it is easier to describe each transformation involving isotopes as a fractionation factor, (1 − α), and express in parts per thousand (). For example, the fractionation factor associated with CO2 diffusing in air, a, which depletes the heavy isotope, is +4.4. Limiting step: When enzymatic reactions proceed by multi-step mechanisms the overall isotope effect depends on both the ‘intrinsic isotope effect’ of the isotopically sensitive step and on the extent to which this step is rate-limiting (O’Leary et al., 1992). Branching point: When a substrate contributes to other competing reactions the overall isotope effect will depend on the relative contribution of the substrate to the enzymatic reaction and to the other reactions. Positional effect: Carbon atoms in a particular molecule can show large differences in isotope ratio, dependent on the isotope effects associated with each biosynthetic process. Temperature dependence of isotope effects: Isotope effects of enzymatic reactions depend on physical and chemical conditions with some decarboxylation steps less rate-limiting with increasing temperature or because of pH modification with temperature (O’Leary, 1980). Isotope discrimination: We define the overall shift in isotope composition from source to product as discrimination (), expressed as = α − 1 = (Rs /Rp ) − 1.

(A4)

According to Eq. A4, to determine the carbon isotope discrimination by plants, one should measure the carbon isotope ratios in the air (source) and in the plant

(product), Ra and Rp , but in practice each is measured against a defined standard, with an isotope ratio arbitrarily set to 0. The carbon isotope composition of plant material, δ 13 Cp (expressed in ), is defined as δ 13 Cp = (Rp − Rstd )/Rstd = (Rp /Rstd ) − 1, (A5) where, Rstd is the (molar) isotope ratio, 13 C/12 C, of the standard. Using Eq. A4 and Eq. A5, the overall carbon isotope discrimination, , could be calculated as follows: = (δ 13 Ca − δ 13 Cp )/(1 + δ 13 Cp )

(A6)

δ 13 Ca

where, is the carbon isotope composition of air. Atmospheric CO2 has an isotope ratio of approximately −8, and typical C3 plant material −28, and using Eq. A6, an overall discrimination () value of 20.6 (Farquhar et al., 1989). Discrimination can be approximated as the difference between the carbon isotope composition of the source and that of the product: = δ 13 Ca − δ 13 Cp

(A7)

In this case, the discrimination value for C3 plants on the basis of plant organic matter will be approximated to = (−0.008) − (−0.028) = 20. Apparent fractionation: The fractionation effects for respiration or photorespiration (respectively termed e and f ) represent many biochemical reactions. Whilst these can be measured as the difference between organic material and CO2 released (as an overall discrimination: see Eq. A7), we have adopted the term ‘apparent fractionation’ to distinguish between individual fractionation factors and the cumulative fractionation which is measured for a given process. References Abelson H & Hoering TC (1961) Carbon isotope fractionation in formation of aminoacids by photosynthetic organisms. PNAS 47: 623–632. Atkin OK, Evans JR & Siebke K (1998) Relationship between the inhibition of leaf respiration by light and enhancement of leaf dark respiration following light treatment. Aust. J. Plant. Physiol. 25: 437–443 Atkin OK, Evans JR, Ball MC, Lambers H & Pons TL (2000) Leaf respiration of snow gum in the light and dark. Interactions between temperature and irradiance. Plant Physiol. 122: 915–924. Baertschi P (1953) Die Fraktionierung der natürlichen Kohlenstoffisotopen im Kohlendioxydstoffwechsel grüner Planzen. Helvetica Chim. Acta 95: 773–781. Blair N, Leu A, Munoz E, Olsen J, Kwong E & Desmarais D (1985) Carbon isotopic fractionation in heterotrophic microbial metabolism. App. Env. Microbiol. 50 (4): 996–1001. Bort J, Brown RH & Araus JL (1996) Refixation of respiratory CO2 in the ears of C-3 cereals. J. Exp. Bot. 47: 1567–1575.

160 Brooks A & Farquhar GD (1985) Effect of temperature on the CO2 /O2 specificity of ribulose-1,5-bisphosphate carboxylase/oxygenase and the rate of respiration in light. Planta 165: 397–406. Brooks JR, Flanagan LB, Varney GT & Ehleringer JR (1997) Vertical gradients in photosynthetic gas exchange characteristics and refixation of respired CO2 within boreal forest canopies. Tree Physiol. 17: 1–12. Bowling DR, McDowell NG, Bond BJ, Law BE & Ehleringer JR (2002) C-13 content of ecosystem respiration is linked to precipitation and vapor pressure deficit. Oecologia 131: 113–124. Bowling DR, Tans PP & Monson RK (2001) Partitioning net ecosystem carbon exchange with isotopic fluxes of CO2 . Global Change Biol. 7: 127–145. Brugnoli E, Hubick KT, von Caemmerer S, Wong SC & Farquhar GD (1988) Correlation between the carbon isotope discrimination in leaf starch and sugars of C3 plants and the ratio of intercellular and atmospheric partial pressures of carbon dioxide. Plant Physiol. 88: 1418–1424. Brugnoli E & Farquhar GD (2000) Photosynthetic fractionation of carbon isotopes. In: Leegood RC, Sharkey TD & von Caemmerer (eds) Photosynthesis: Physiology and Metabolism (pp. 399– 434). Kluwer Academic Publishers, the Netherlands. Buchmann N, Brooks JR, Flanagan LB & Ehleringer JR (1998) Carbon isotope discrimination of terrestrial ecosystems. In: Griffiths H (ed) Stable Isotopes and the Integration of Biological, Ecological and Geochemical Cycles (pp. 203–221). BIOS Scientific, Oxford. Buchmann N, Kao WY & Ehleringer J (1997) Influence of stand structure on carbon-13 of vegetation, soils, and canopy air within deciduous and evergreen forests in Utah, United States. Oecologia 110: 109–119. Ciais P & Meijer HAL (1998) The 18 O/16 O isotope ratio of atmospheric CO2 and its role in global carbon cycle research ecosystems. In: Griffiths H (ed) Stable Isotopes: Integration of Biological, Ecological and Geochemical Processes (pp. 409–431). Environmental plant biology series, Bios Scientific Publishers, Oxford, UK. Ciais P, Friedlingstein P, Schimel DS & Tans PP (1999) A global calculation of the delta C-13 of soil respired carbon: Implications for the biospheric uptake of anthropogenic CO2 . Global Biogeochem. 13: 519–530. Ciais P, Tans PP, White JWC, Trolier M, Francey RJ, Berry JA, Randall DR, Sellers PJ, Collatz JG & Schimel DS (1995a) Partitioning of Ocean and Land Uptake of CO2 as Inferred by Delta-C-13 Measurements from the Noaa Climate Monitoring and Diagnostics Laboratory Global Air Sampling Network. J. Geophys. Res.-Atmos. 100: 5051–5070. Ciais P, Tans PP, Trolier M, White JWC & Francey RJ (1995b) A Large Northern-Hemisphere Terrestrial CO2 Sink Indicated by the C-13/C-12 Ratio of Atmospheric CO2 . Science 269: 1098– 1102. Craig H (1954) Carbon 13 in plants and the relationship between carbon 13 and carbon 14 variation in nature. J. Geol. 62: 115– 149. Damesin C & Lelarge C (2003) Carbon isotope composition of current-year shoots from Fagus sylvatica in relation to growth, respiration and use of reserves. Plant Cell Environ. 26: 207–219. Deléens E, Schwebel-Dugué N & Trémolières A (1984) Carbon isotope composition of lipidic classes isolated from tobacco leaves. FEBS Lett. 178: 55–58. DeNiro MJ & Epstein S (1977) Mechanism of carbon isotope fractionation associated with lipid synthesis. Science 197: 261–263.

Duranceau M, Ghashghaie J, Badeck F, Deleens E & Cornic G (1999) δ 13 C of CO2 respired in the dark in relation to δ 13 C of leaf carbohydrates in Phaseolus vulgaris L. under progressive drought. Plant Cell Environ. 22: 515–523. Duranceau M, Ghashghaie J & Brugnoli E (2001) Carbon Isotope discrimination during photosynthesis and dark respiration in intact leaves of Nicotiana sylvestris: comparison between wild type and mitochondrial mutant plants. Aust. J. Plant. Physiol. 28(1): 65–71. Ekblad A & Hogberg P (2000) Analysis of delta C-13 of CO2 distinguishes between microbial respiration of added C-4-sucrose and other soil respiration in a C-3-ecosystem. Plant and Soil 219: 197–209. Ekblad A & Hogberg P (2001) Natural abundance of C-13 in CO2 respired from forest soils reveals speed of link between tree photosynthesis and root respiration. Oecologia 127: 305–308. Ekblad A, Nyberg G & Hogberg P (2002) C-13-discrimination during microbial respiration of added C-3-, C-4- and C-13-labelled sugars to a C-3-forest soil. Oecologia 131: 245–249. Evans JR, Sharkey TD, Berry JA & Farquhar GD (1986) Carbon isotope discrimination measured concurrently with gas exchange to investigate CO2 diffusion in leaves of higher plants. Austral. J. Plant Physiol. 13: 281–292. Evans & Loreto (2000) Acquisition and diffusion of CO2 in higher plant leaves. In: Leegood RC, Sharkey TD & von Caemmerer (eds) Photosynthesis: Physiology and Metabolism (pp. 321– 351). Kluwer Academic Publishers, the Netherlands. Farquhar GD (1980) Carbon isotope discrimination by plants and the ratio of intercellular and atmospheric CO2 concentrations. In: Pearman GI (ed) Carbon Dioxide and Climate: Australian Research (pp. 105–110). Australian Academy of Science, Canberra. Farquhar GD, O’Leary MH & Berry JA (1982) On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Austr. J. Plant Physiol. 9: 121–137. Farquhar GD, Hubick KT, Condon AG & Richards RA (1989) Carbon isotope fractionation and water-use efficiency. In: Rundel PW, Ehleringer JR & Nagy KA (eds), Satble Isotopes in Ecological Research. Ecological Studies, Vol. 68 (pp. 21–40). Springer-Verlag, New York. Gerbaud A & Andre M (1987) An evaluation of the recycling in measurements of photorespiration. Plant Physiol. 83: 933–937. Ghashghaie J, Duranceau M, Badeck F, Cornic G, Adeline MT & Deleens E (2001) δ 13 C of CO2 respired in the dark in relation to leaf metabolites: comparisons between Nicotiana sylvestris and Helianthus annuus under drought. Plant Cell Environ. 24: 505– 515. Gillon JS (1997) Carbon isotope discrimination: interactions between respiration, leaf conductance and photosynthetic capacity. Ph.D. Thesis. University of Newcastle upon Tyne, Newcastle. Gillon JS & Griffiths H (1997) The influence of (photo)respiration on carbon isotope discrimination in plants. Plant Cell Environ. 20: 1217–1230. Gillon JS, Borland AM, Harwood KG, Roberts A, Broadmeadow MSJ & Griffiths H (1998) Carbon isotope discrimination in terrestrial plants: carboxylations and decarboxylations. In: Griffiths H (ed) Stable Isotopes: Integration of Biological, Ecological and Geochemical Processes (pp. 203–221). Environmental plant biology series, Bios scientific publishers, Oxford, UK. Gleixner G, Danier HJ, Werner RA & Schmidt HL (1993) Correlation between the 13 C content of primary and secondary plant

161 products in different cell compartments and that in decomposing basidiomycetes. Plant. Physiol. 102: 1287–1290. Gleixner G & Schmidt HL (1997) Carbon isotope effects on the fructose-1,6-bisphosphate aldolase reaction, origin for nonstatistical 13 C distribution in carbohydrates. J. Biol. Chem. 272(9): 5382–5387. Gleixner G, Scrimgeour C, Schmidt HL & Viola R (1998) Stable Isotope distribution in the major metabolites of source and sink organs of Solanum tuberosum L.: A powerful tool in study of metabolic partitioning in intact plants. Planta 207: 241–245. Griffiths H, Borland A, Gillon JS, Harwood K, Maxwell K & Wilson J (1999) Stable isotopes reveal exchanges between soil, plants and the atmosphere. In: Press MC, Scholes JD & Barker MG (eds) Physiological Plant Ecology. Blackwell Science. Guy RD, Fogel ML & Berry JA (1993) Photosynthetic Fractionation of the Stable Isotopes of Oxygen and Carbon. Plant Physiol. 101: 37–47. Harwood KG, Gillon JS, Griffiths H & Broadmeadow MSJ (1998) Diurnal variation of CO2 δ 13 C and δ 18 O and evaporative site enrichment of H2 O δ 18 O in Piper aduncum under field conditions in Trinidad. Plant Cell Environ. 21: 269–283. Henderson SA, von Caemmerer S & Farquhar GD (1992) Shortterm measurements of carbon isotope discrimination in several C4 species. Aust. J. Plant Physiol. 19: 263–285. Hill SA & Bryce JH (1992) Malate metabolism and light-enhanced dark respiration in barley mesophyll protoplasts. In: Lambers H, van der Plas LHW (eds) Molecular, Biochemical and Physiological Aspects of Plant Respiration (pp. 221–230). SPB Academic Publishing, The Hague, The Netherlands. Hsu JC & Smith BN (1972) 13 C/12 C ratios of carbon dioxide from peanut and sunflower seedlings and tobacco leaves in light and in darkness. Plant Cell Physiol. 13: 689–694 Igamberdiev AU, Ivlev AA, Bykova NV, Threlkeld CN, Lea PJ & Gardestrom P (2001) Decarboxylation of glycine contributes to carbon isotope fractionation in photosynthetic organisms. Photosynth. Res. 67: 177–184. Ivlev AA, Bykova NV & Igamberdiev AU (1996) Fractionation of carbon (C-13/C-12) isotopes in glycine decarboxylase reaction. FEBS Letts. 386: 174–176. Ivlev AA, Igamberdiev AU, Threlkeld C & Bykova NV (1999) Carbon isotope effects in the glycine decarboxylase reaction in vitro on mitochondria from pea and spinach. Russ. J. Plant Physiol. 46: 653–660. Ivlev AA (2001) Carbon isotope effects (C-13/C-12) in biological systems. Separ. Sci. Tech. 36: 1819–1914. Ivlev AA (2002) Carbon isotope (C-13/C-12) effect of photorespiration in photosynthesizing organisms. Evidence in favor of the existence and a possible mechanism. Biofizika 47: 55–70. James WO (1953) Plant Respiration. Oxford University Press, Amen House, London. Jordan F, Kuo DJ, & Monse EU (1978) Carbon kinetic isotope effects on pyruvate decarboxylation catalysed by yeast pyruvate decarboxylase and models. J. Amer. Chem. Soc. 100 (9): 2872–2878. Jacobson BS, Smith BN, Epstein S & Laties GG (1970) The prevalence of carbon 13 in respiratory carbon dioxide as an indicator of the type of endogenous substrate: The change from lipid to carbohydrate during the respiratory rise in potato slices. J. Gen. Physiol. 55: 1–17. Keeling CD (1958) The concentration and isotopic abundances of carbon dioxide in rural areas. Geochim. Cosmochim. Acta 13: 322–334.

Lin G & Ehleringer JR (1997) Carbon isotopic fractionation does not occur during dark respiration of C3 and C4 plants. Plant Physiol. 114: 191–194. Lloyd J & Farquhar GD (1994) C-13 Discrimination During CO2 Assimilation by the Terrestrial Biosphere. Oecologia 99: 201– 215. Loreto F, Velikova V & Di Marco G (2001) Respiration in the light measured by (CO2 )-C-12 emission in (CO2 )-C-13 atmosphere in maize leaves. Aus. J. Plant Physiol. 28: 1103–1108. Melzer E & Schmidt HL (1987) Carbon isotope effects on the pyruvate dehydrogenase reaction and their importance for relative carbon-13 depletion in lipids. J. Biol. Chem. 262: 8159–8164. O’Leary MH (1980) Determination of heavy-atom isotope effects on enzyme-catalyzed reactions. Meth. Enzymol. 64: 83–107. O’Leary MH (1981) Carbon Isotope Fractionation in plants. Phytochemistry 20: 553–567. O’Leary MH, Madhavan S & Paneth P (1992) Physical and chemical basis of carbon isotope fractionation in plants. Plant Cell Environ. 15: 1099–1104. Park R & Epstein S (1961) Metabolic fractionation of 13 C and 12 C in plants. Plant Physiol. 36: 133–138. Parnik T, Keerberg O & Viil J (1972) Decarboxylation of primary and end-products of photosynthesis at different oxygen concentrations. J. Exp. Bot. 46: 1439–1447. Raven JA, Griffiths H, Glidewell SM & Preston T (1982) The Mechanism of Oxalate Biosynthesis in Higher-Plants – Investigations with the Stable Isotopes O-18 and C-13. Proc. Royal Soc. B. 216: 87–101. Rooney MA (1988) Short-term carbon isotope fractionation by plants. Ph.D Thesis. University of Wisconsin, Wisconsin. Rossmann A, Butzenlechner M & Schmidt HL (1991) Evidence for a non-statistical carbon isotope distribution in natural glucose. Plant Physiol. 96: 609–614. Rossmann A, Schmidt HL, Reniero F, Versini G, Moussa I & Merle MH (1996) Stable carbon isotope content in ethanol of EC data bank wines from Italy, France and Germany. Zeitschr. Lebensmittel-Untersuch. Forsch. 203: 293–301. Schmidt HL & Gleixner G (1998) Carbon isotope effects on key reactions in plant metabolism and 13 C-patterns in natural compounds. In: Griffiths H (ed) Stable Isotopes: Integration of biological, ecological and geochemical processes (pp. 13–25). Environmental plant biology series. Bios Scientific Publishers, Oxford, UK. Smith BN (1971) Carbon isotope ratios of respired CO2 from castor bean, peanut, pea, radish, squash, sunflower and wheat seedlings. Plant Cell Physiol. 12: 451–455. Tcherkez G, Nogués S, Bleton J, Cornic G., Badeck F.-W & Ghashghaie J (2003) Metabolic origin of carbon isotope composition of leaf dark-respired CO2 in Phaseolus vulgaris L. Plant Physiol. 131: 237–244. Troughton JH, Card KA, & Hendy CH (1974) Photosynthetic pathways and carbon isotope discrimination by plants. Carnegie Inst. Washington Yearb. 73: 768–780. Vogel JC (1980) Fractionation of the carbon isotopes during photosynthesis. In: Sitzungsberichte der Heidelberger Akademie der Wissenschaften, matematisch-naturwissen-schaftlichs Klasse Jahrgang, 3 Abhandlung (pp. 111–135). Springer-Verlag, Berlin/New York. von Caemmerer S & Evans JR (1991) Determination of the average partial pressure of CO2 in chloroplasts from leaves of several C3 plants. Aust. J. Plant Physiol. 18: 287–305. Yakir D & Wang XF (1996) Fluxes of CO2 and water between terrestrial vegetation and the atmosphere estimated from isotope measurements. Nature 380: 515–517.