SEARCH

SEARCH BY CITATION

Keywords:

  • Day respiration;
  • isotope;
  • fractionation;
  • mesocosm;
  • sunflower

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

While there is currently intense effort to examine the 13C signal of CO2 evolved in the dark, less is known on the isotope composition of day-respired CO2. This lack of knowledge stems from technical difficulties to measure the pure respiratory isotopic signal: day respiration is mixed up with photorespiration, and there is no obvious way to separate photosynthetic fractionation (pure ci/ca effect) from respiratory effect (production of CO2 with a different δ13C value from that of net-fixed CO2) at the ecosystem level. Here, we took advantage of new simple equations, and applied them to sunflower canopies grown under low and high [CO2]. We show that whole mesocosm-respired CO2 is slightly 13C depleted in the light at the mesocosm level (by 0.2–0.8‰), while it is slightly 13C enriched in darkness (by 1.5–3.2‰). The turnover of the respiratory carbon pool after labelling appears similar in the light and in the dark, and accordingly, a hierarchical clustering analysis shows a close correlation between the 13C abundance in day- and night-evolved CO2. We conclude that the carbon source for respiration is similar in the dark and in the light, but the metabolic pathways associated with CO2 production may change, thereby explaining the different 12C/13C respiratory fractionations in the light and in the dark.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Isotopic fractionation against 13CO2 during photosynthesis (denoted as Δ) drives the isotopic signal of ecosystems, in which the carbon isotope composition of fixed CO2 follows the well-accepted relationship involving internal CO2 mole fraction and fractionations associated with diffusion and carboxylation (Farquhar, O’Leary & Berry 1982; Lloyd & Farquhar 1994). Nevertheless, at the ecosystem level, CO2 evolved by plant respiration is believed to account for 30–70% of the ecosystem CO2 exchange (Amthor 2000) so that the isotopic signal of respired CO2 has a major influence on the 12C/13C ecosystem mass balance (Lavigne et al. 1997). Therefore, current efforts are devoted to elucidating the isotopic signal of CO2 evolved by several ecosystem compartments such as trunks (Brandes et al. 2006; Gessler et al. 2007; Maunoury et al. 2007) and soil (Ekblad & Hogberg 2001; Bostrom, Comstedt & Ekblad 2007). At the leaf level, there is now compelling evidence that photorespiration fractionates against 13C, thereby liberating 13C-depleted CO2 as compared to photosynthates (Lanigan et al. 2008).

In contrast, little is known on the isotopic composition of day-respired CO2, although several authors have suggested that day respiration produces 13C-depleted CO2 (von Caemmerer 2000; Ghashghaie et al. 2003; Tcherkez et al. 2004; Lanigan et al. 2008). In the dark, leaf-respired CO2 has been shown to be 13C enriched, with some variations that depend on the respiratory rate, which is in turn influenced by leaf temperature or leaf respiratory substrates (Duranceau et al. 1999; Ghashghaie et al. 2001; Tcherkez et al. 2003). In fact, leaf-respired CO2 is considerably 13C enriched just after darkening (Mortazavi et al. 2005; Barbour et al. 2007; Werner et al. 2007; Gessler et al. 2009), and then reaches a steady value up to 6‰ enriched as compared to sucrose (Duranceau et al. 1999). The isotope composition of CO2 evolved by other organs is less well documented: while roots have been repeatedly shown to produce 13C-depleted CO2 (Badeck et al. 2005; Klumpp et al. 2005; Bathellier et al. 2009), twigs and trunks have a more variable pattern that depends upon environmental parameters (Damesin & Lelarge 2003). However, the respiratory contribution of all non-photosynthetizing organs when integrated as isotopic fractionations at the plant level is not well defined (for a recent review, see Bowling, Pataki & Randerson 2008).

In consequence, it has often been assumed that the respiratory isotopic signal of plants, mesocosm and ecosystems is similar in the light and in the dark (e.g. Schnyder et al. 2003, but see Kodama et al. 2008 and references therein). For isotope partitioning studies, the value of ecosystem-respired CO2 is typically measured at night, and then used to partition respiratory processes of the light period (for a specific discussion on this topic, see Zobitz et al. 2008). However, such an assumption does not seem consistent with published data of isotopic biochemistry of plants (see also Cernusak et al. 2009 for a review). Firstly, a positive relationship has been observed between the carbohydrate content and the respiration rate in many plant organs (see, e.g. Azcon Bieto & Osmond 1983; Tjoelker et al. 2008). In leaves, the availability of respiratory substrates has a clear impact on the δ13C value of dark-evolved CO2 (Tcherkez et al. 2003; Hymus et al. 2005; Gessler et al. 2009). For organs other than leaves, where circadian rhythms occur in sucrose content, there may be different δ13C values in respiratory CO2 in the light and in the dark. Secondly, sucrose molecules produced in the dark (from 13C-enriched, transitory starch) and in the light (from 13C-depleted, cytoplasmic triose phosphates) do not have the same isotope composition (Tcherkez et al. 2004; Gessler et al. 2008). This may drive a light/dark cycle of the δ13C value of CO2 evolved by source or sink organs depending on whether respiration is dominated by autotrophic or heterotrophic processes in those tissues. Thirdly, while some organs such as roots do not have any apparent growth circadian rhythm, others have (leaves and secondary meristems in stems) (Zweifel, Item & Hasler 2001; Walter & Schurr 2005; Deslauriers et al. 2007). Owing to the tight relationship between respiration and plant growth (for a review, see Amthor 2000), light/dark growth variations may have an effect on the δ13C of evolved CO2.

In the present paper, the isotopic signal of day-respired CO2 was determined (13C enrichment or depletion compared to net fixed CO2) as compared to dark-evolved CO2. We carried out mesocosm-level experiments with sunflower (Helianthus annuus) canopies, in which both the day and dark isotopic signals associated with respiration were investigated and compared with CO2 evolved by individual organs. For this purpose, we propose two techniques to measure the isotopic impact of CO2 evolved in the light. Furthermore, a 12C/13C labelling was conducted in order to see metabolic correlations between day-respired CO2, dark-respired CO2 and different plant components.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Experimental

Plant material and growth conditions

Sunflower (H. annuus L. cv. Sanluca) plants were sown individually in plastic pots (5 cm diameter – 35 depth) filled with washed quartz sand. The pots were distributed at a density of 118 plants m−2 in two growth chambers (E15, Conviron, Winnipeg, Canada). Modified Hoagland nutrient solution (7.5 mol N m−3) was supplied by an automatic irrigation system throughout the experiment. Irradiance during the 16 h photoperiod was supplied by cool white fluorescent tubes (16 × 160 W; Sylvania Germany GmbH, Erlangen, Germany) and incandescent lamps (12 × 100 W; General Electric Germany, München, Germany), and was maintained at 520 µmol m−2 s−1 photosynthetic photon flux density (PPFD) at the top of the canopy by adjusting the height of the lamps following plant development. Air temperature was controlled at 20/16 °C and relative humidity at 75/80% during the photo and dark periods, respectively. The CO2 mole fraction of the air in the chamber was 200 and 1000 µmol mol−1 (chambers 1 and 2, respectively; for details on gas mixing and controlling, see next paragraph).

Gas exchange measurements and isotope analysis

Mesocosm level.  The two growth chambers formed part of the mesocosm 13CO2/12CO2 open gas exchange system described in detail by Schnyder et al. (2003). A screw compressor (S40, Boge, Bielefeld, Germany) and adsorption dryer (KEN3100, Zander, Essen, Germany) generated CO2-free, dry air. Each chamber had an individual gas-mixing system comprising two computer-operated mass flow controllers (FC-2925V for air and FC-2900 4S for CO2, Tylan General, San Diego, CA, USA) mixing dry air with CO2 of known isotopic composition (δ13C). The CO2 concentrations were held constant at the chamber outlets by adjusting the rate and CO2 concentration of the air supplied to the increasing rates of photosynthesis. For each CO2 concentration, the chambers were first supplied with CO2 coming from a mineral source (13C enriched, δ13C −3.5‰) (pre-labelling period) and then originating from a fossil organic source (13C depleted, δ13C −44.5‰; both CO2 from Linde AG, Munich, Germany) (labelling).

For measurements of CO2 fluxes and associated on-line 12C/13C discrimination of the canopies, sample air was collected at the inlet and outlet (vent) of each growth chamber, and continuously pumped to a computer-controlled sample air selector (SAS) at a rate of approx. 2 L min−1. During simultaneous operation of all chambers, the SAS sequentially sampled each sample air line (n = 8) at 2 min intervals. Sample air was split to serve the infrared gas analyser (IRGA; Li-6262; Li-Cor Inc., Lincoln, NE, USA) and isotope ratio mass spectrometer (IRMS; Delta Plus, Finnigan MAT, Bremen, Germany) in parallel. Gas lines between the SAS, IRMS and IRGA were flushed with sample air for 2 min before taking IRGA readings of absolute CO2 and H2O concentration and measurement of δ13C of CO2 by IRMS. The IRMS was interfaced with the SAS via a steel capillary tube (i.d. 0.1 mm), a six-port, two-position valve (Valco Instruments Co. Inc., Houston, TX, USA), dryer (Nafion), gas chromatograph (25 m × 0.32 mm Poraplot Q; Chrompack, Middelburg, the Netherlands) and open split. These components all formed part of a custom-made interface (GC–GP Interface; Finnigan MAT). Sample air was pumped continuously through the steel capillary feeding the Valco valve and a 300 µL sample loop attached to it. After a 2 min flushing period, shortly before the SAS switched to the next sample air line, the content of the sample loop was swept with helium carrier gas through the interface, where water vapour was removed by the Nafion membrane, and CO2 was separated from all other gaseous components of the air sample in the GC column. Finally, the CO2 was introduced directly into the ion source of the IRMS via the open split. Samples were compared with a V-PDB-gauged, working standard reference CO2 injected once at the start and once at the end of a measurement cycle. The overall precision of the measurement at the chamber inlets over a 24 h period was typically better than 0.15‰. A full measurement cycle, including one set of measurements (concentrations of CO2 and H2O, and δ13C of CO2) on the inlet and outlet of each growth chamber, was completed within 24 min.

The net CO2 exchange flux of the canopies (N, µmol CO2 m−2 s−1) was obtained as the balance of CO2 entering and leaving the chamber divided by the chamber ground area (s, m2):

  • image

with E and O the fluxes of CO2 (µmol s−1) entering and leaving the chamber. The difference E – O is denoted as P below (‘Calculations’ in Materials and methods). In the light, N is the net assimilation value at the mesocosm level, and as such, it is denoted as A (net assimilation) in the following. In the dark, N is the negative of the mesocosm-level respiration rate.

Mesocosm-scale ‘on-line’12C/13C discrimination (denoted as Δ) during photosynthesis (i.e. gas exchange in light) was obtained as given by Evans et al. (1986):

  • image

where ξ = ce/(ce − co), and ce and co are the CO2 mole fractions (µmol mol−1) of the air entering and leaving the chamber after correcting to standard humidity.

While N could be measured continuously from the first day of the experiment, the associated isotope analyses could be done with reasonable precision only after day 12, when rates of photosynthesis were high enough to yield significant differences in the C-isotopic composition of the CO2 entering and leaving the chambers. The δ13C of dark-respired CO2 was measured at days 30–34, that is, when canopies were closed (plants at the three fully expanded leaves stage) and CO2 exchange rates had reached a steady state on a day-to-day basis. At days 35 and 36, CO2 response curves of canopies were carried out by changing the CO2 concentration at the chamber inlet gradually from 100 to 1500 µmol mol−1 within 2 h. Again, measurements of net CO2 exchange fluxes were accompanied by the simultaneous determination of Δ. The corresponding results of the A/ca curve are shown in the Supporting Information Fig. S1. At day 37 after the start of the experiment, canopies were isotopically labelled by switching the source CO2 from mineral (−3.5‰) to fossil organic (−44.5‰) CO2. Gas exchange analyses continued for 5 more days with labelling CO2 and otherwise unchanged growth conditions.

Organ level.  One plant was removed from the growth chambers described above, and the attached organs were placed in a closed system for the on-line measurement of the δ13C of the respired CO2. The closed system which was described previously by Tcherkez et al. (2003) was directly coupled to the IRMS (Finnigan) as specified above. The procedure for accumulating respired CO2 in the dark and measure the isotope composition was identical. All experiments were carried out on the top expanding (EL) and mature (ML) leaves, and in the adjacent stem (ST). The contribution to the mesocosm exchange of the two cotyledons was considered negligible as they had died at the time the measurements were done. The photosynthetic fractionation at the leaf level was measured following the procedure of Nogués et al. (2004) and Tcherkez et al. (2005), that is, by coupling the portable gas exchange system (TPS-2, PP systems) to the IRMS through a three-way valve.

Metabolite extraction and quantification.  Plants were harvested and the different organs were frozen in liquid nitrogen, lyophilized and analysed as follows. The extraction–purification procedure for starch, sucrose, glucose and fructose was that described previously (Tcherkez et al. 2003). Purified metabolites were lyophilized, resuspended in 100 µL of distilled water and then transferred to tin capsules (Courtage Analyze Service, Mont-Saint-Aignan, France) and dried for isotope analysis. The isotope analysis of metabolites and total organic matter was carried out with the EA-IRMS (EA1500; Carlo-Erba, Milan, Italy, coupled to the Optima, GV Instruments (Villeurbanne, France) on the isotopic facility structure Plateforme Métabolisme-Métabolome, as already described (Tcherkez et al. 2003; Nogués et al. 2004).

Calculations

In the following, simple relationships are used to derive the isotopic signature of respired CO2 in the light. Firstly, the mesocosm system was considered as a big assimilation cuvette, and equations derived from gas exchange are used to calculate an estimate of the contribution (thereafter denoted as d*/ca) of day mesocosm-respired CO2 to net photosynthetic fractionation of the mesocosm. Secondly, the contribution of non-photosynthetic organs to the net photosynthetic fractionation of the mesocosm was estimated by taking advantage of the net photosynthetic fractionation of leaves that was measured separately. That is, the respiratory isotopic signal of non-photosynthetic organs (thereafter denoted as inline image, see the definition of symbols in Table 1 and just below) was determined by ‘substracting’ the effect of leaf net photosynthesis from the mesocosm-level photosynthetic fractionation. Thirdly, the respiratory isotopic signal of non-photosynthetic organs was followed after isotopic CO2 labelling to calculate the proportion (denoted as xday) of old, pre-labelling carbon to CO2 production in the light.

Table 1.  Summary of the main symbols used (upper panel) and main values examined (lower panel) in the present paper
Expression or symbol used (units)ConditionsDescription
  1. dl., dimensionless.

Δ (‰)LightIsotope fractionation associated with net photosynthesis (CO2 exchange) of the mesocosm
inline image (‰)LightAverage isotope fractionation associated with net photosynthesis (CO2 exchange) of leaves
Δ0 (‰)LightIsotope fractionation associated with mesocosm-level net photosynthetic assimilation, when net photosynthesis tends to zero
e (‰)LightIsotope fractionation associated with leaf day respiration
f (‰)LightIsotope fractionation associated with CO2 evolution by photorespiration at the leaf level
enp (‰)LightIsotope fractionation associated with respiration of non-photosynthetic organs in the light at the mesocosm level
eapp (‰)LightApparent isotope fractionation associated with respiration of non-photosynthetic organs in the light after isotopic labelling at the mesocosm level
N (µmol m−2 s−1)Light/NightNet mesocosm CO2 exchange
P (µmol s−1)Light/NightNet mesocosm CO2 exchange (not scaled to surface area)
xday (dl.)LightProportion of recent carbon in CO2 respired by non-photosynthetic organs in the light at the mesocosm level
xnight (dl.)NightProportion of recent carbon in CO2 respired by the mesocosm in darkness at the mesocosm level
d*/ca (‰)Light(Photo)respiratory isotopic contribution to the photosynthetic fractionation at the mesocosm level (Eqn 2)
inline image(‰)LightRespiratory isotopic contribution of non-photosynthetic organs to the photosynthetic fractionation at the mesocosm level (Eqn 3)
en (‰)NightIsotope fractionation of night respiration by the mesocosm (Eqn 7)

The main parameters used throughout the present paper and explained below are summarized in Table 1. The expression ‘(photo)respiratory’ CO2 used in the following stands for the sum of photorespiratory and day respiratory evolved CO2.

The isotopic (photo)respiratory component of Δ at the mesocosm level

The (photo)respiratory component of the net photosynthetic carbon isotope discrimination Δ (i.e. the (photo)respiratory term in Farquhar's expression giving Δ; see Eqn 1) was calculated using equations derived from Evans et al. (1986) as explained below. The present methods were developed to avoid the need of ci, the internal CO2 mole fraction, because this parameter could not be reached at the mesocosm level in the present study. The (photo)respiratory component is denoted here as d* as follows (Farquhar et al. 1982):

  • image(1)

where Δ is the net photosynthetic (observed) carbon isotope discrimination, and a, ai and b are the fractionations associated with diffusion in air, dissolution and diffusion in water, and carboxylation, respectively. ca, ci and cc are the CO2 mole fractions in the atmosphere, in the intercellular spaces and at the carboxylation sites, respectively. d* is equal to inline image where e and f are the carbon isotope fractionation associated with day respiration (the rate of which is Rd) and photorespiration. Γ* is the CO2 compensation point in the absence of day respiration, and k is the carboxylation efficiency.

In both methods used below, it is assumed that d* is constant with respect to ca. This hypothesis may not be verified when ca varies because k depends upon CO2 mole fraction. Here, we used the region of low assimilation values in which the relative change of k is rather small, so that the assumption of a constant d* is valid.

We also recognize that the value of d* obtained here includes both leaf day respiration and heterotrophic respiration (from roots and stems), and so is not strictly equal to that used in the equation of Farquhar et al. (1982) (Eqn 1). That is, the day respiratory component eRd/kca applies at the level of cc (i.e. day-respired CO2 is released into the intracellular CO2 pool), while CO2 evolved by heterotrophic organs is not released at cc but rather into surrouding air (at ca). Nevertheless, Eqn 1 is not altered. Let us denote as δh the carbon isotope composition of CO2 evolved by non-photosynthetic organs, and ε the CO2 amount (in µmol mol−1) produced by respiration of non-photosynthetic organs. ΔA is the net photosynthetic fractionation of photosynthetic organs (leaves). By mass balance, we have:

  • image

This re-arranges to:

  • image

where Δobs is the observed discrimination at the mesocosm level (including both leaves and heterotrophic organs), using the equation of Evans et al. (1986), as indicated above. Within the numerator, the term δhΔA may be neglected, giving δo − ΔA − δh, which is very close to the isotope fractionation associated with respiration of heterotrophic organs, with respect to carbon fixed by leaves. Let us denote this fractionation as eh. The denominator is very close to ce − co, and the ratio ε/(ce − co) is equal to Rh/A, where Rh denotes respiration by heterotrophic organs, and A is net mesocosm photosynthesis. This gives:

  • image

In other words, the apparent value of d* obtained from Δobs (as explained below) here includes an additional term that represents heterotrophic respiration. That is,

  • image

The heterotrophic term ehRhca/A, and the leaf term eRd/k are very similar because the carboxylation efficiency k is defined as vc/cc, where vc is the carboxylation rate and cc is internal CO2 mole fraction. Therefore, the input of heterotrophic respiration to mesocosm CO2 at the level of ca does not modify the general equations described here, and the d* value computed below simply includes heteretrophic respiration in the light. It should be noted that similarly, there would also be a contribution of heterotrophic respiration to the apparent d* value at the leaf level because of the contribution of non-photosynthetic leaf cells such as phloem tissue, epidermis, etc. In other words, the present equations may also apply to leaf-level gas exchange experiments, even though the heterotrophic term ehRhca/A is quantitatively modest in most leaves.

First method.  We used the following relationships: A = gs (ca − ci) = gm (ci − cc), where A is CO2 net assimilation, and gs and gm are the stomatal and internal conductance for CO2, respectively. Subtracting b from each side of Eqn 1, and re-arranging gives:

  • image(2)

In the following, we denote the left-hand side of Eqn 2 by 〈Δ〉. In the present study, experiments involved A/ca curves, that is, variation of the CO2 level (Supporting Information Fig. S1). Importantly, the use of Eqn 2 does not require ca to be constant as d* is assumed ca independent, and we used the intercept (and not the slope). When plotted against A, 〈Δ〉 shows a non-linear relationship, simply because both conductances vary with A. Stomatal conductance typically decreases at high CO2, and so the slope of the 〈Δ〉-versus-A relationship increases with A. Similarly, it has been recently shown that internal conductance responds to changes in ca (Flexas et al. 2007). Nevertheless, when A converges to zero, 〈Δ〉 converges to d*. In other words, with a set of Δ, ca and A values, extrapolating 〈Δ〉 at A = 0 gives a value of d*. This graphical method is depicted in Fig. 1, in which the linear regression used data with A less than 17 µmol m−2 s−1 (linear region of the plot).

image

Figure 1. Plot of 〈Δ〉 = ca (b − Δ) as a function of net CO2 assimilation A at the mesocosm level when assimilation is made to vary with CO2 mole fraction ca (the A/ca curve is given as a Supporting Information). The growth CO2 conditions were 200 µmol mol−1 (chamber 1, closed symbols) or 1000 µmol mol−1 (chamber 2, open symbols). The calculations use b = 29‰. The extrapolated values of 〈Δ〉 for A = 0 (intercepts) are 604‰ µmol mol−1 (chamber 1) and 1183‰ µmol mol−1 (chamber 2). Both regressions were significant with P < 0.008 (r2 = 0.54 and 0.83, and F = 11.4 and 38.4, respectively). Inset: mesocosm-level net photosynthetic fractionation Δ as a function of A. The extrapolated values of Δ at A = 0 (intercepts) are 18.89‰ (chamber 1, closed symbols) and 15.44‰ (chamber 2, open symbols).

Download figure to PowerPoint

Second method. Equation 2 is still valid at the CO2 compensation point, Γ, at which A = 0. This simply gives:

  • image(3)

where Δ0 is the net photosynthetic carbon isotope discrimination for A = 0. Clearly, such a value cannot be measured (no net CO2 exchange) and should be extrapolated with a plot representing Δ against A. This method is depicted in the inset of Fig. 1.

The respiratory contribution of non-photosynthetic organs to mesocosm-level isotopic gas exchange in the light

The respiratory component of isotopic exchange between the mesocosm and the atmosphere was calculated using classical mass balance equations and taking advantage of the photosynthetic carbon isotope fractionation measured at the leaf level and applied to the photosynthetizing leaves of the mesocosm. We used the following iso-flux conservation equation at the level of the mesocosm:

  • image(4)

where qA and qr are the total mass of leaves and non-photosynthetic organs (roots and stems), respectively; u is the air flow rate through the mesocosm chamber (in mol s−1); co and ce are the outlet and inlet CO2 mole fractions (in µmol mol−1) corrected to standard humidity, with the associated δ13C values δo and δe, respectively. inline image is the average net photosynthetic rate (in µmol g−1 s−1) of leaves. inline image and inline image are the mass average respiratory rate and the respiration average isotopic composition of CO2 respired by non-photosynthetic organs, respectively. inline image is the isotope composition of net fixed carbon in leaves.

If we denote as P the net CO2 exchange by the mesocosm (in µmol s−1) in the light, we have: inline image. Equation 4 may then be rewritten using P, which is the parameter measured by the mesocosm system. This gives:

  • image

With the proxies inline image where inline image is the average leaf net photosynthetic fractionation, and inline image where enp is the respiratory fractionation by non-photosynthetic organs, we have:

  • image(5)

It is not possible to separate the three variables in the left term of Eqn 5; it represents the non-photosynthetic respiratory contribution to the mesocosm-level fractionation. However, such a value is in ‰µmol s−1, which is not consistent with d*/ca value (see above), that is, the (photo)respiratory contribution to the mesocosm-level fractionation (in ‰). At the mesocosm level, we used here inline image, which is a scaled contribution of respiratory isotopic signal, and is in ‰. This may be calculated by dividing all the terms of Eqn 5 by P. It should be emphasized that the use of Eqn 5 does not require any assumption on the carbon isotope composition of CO2 evolved by respiration in the light. Other authors suppose a similar value in the light and in the dark. Nevertheless, this assumption is unlikely, simply because of the contribution of photosynthetizing organs to night respiration, and such organs have proven to be isotopically different (lighter) than other organs. It should be noted that the inline image value computed with Eqn 5 is relative because it depends upon the leaf photosynthetic fractionation inline image: any slight variation of this parameter at the mesocosm scale as compared to the leaf-level measurement causes variations in inline image. In practice, we used the average of pre-labelling inline image values (mature leaves). Any error in inline image only induces an offset in inline image and does not impact on the percentage of new carbon (see below Eqn 6) and the covariation analysis (Fig. 4).

image

Figure 4. Isotopomic array representation and hierachical clustering (left) of isotope ratios in glucose (Glc), fructose (Fru), total organic matter (TOM), starch and sucrose (Suc), and respiratory CO2 of mature leaves (ML), stems, roots and of the mesocosm. Data for chambers 1 (200 µmol mol−1 CO2) and 2 (1000 µmol mol−1 CO2) are indicated as C1 and C2, respectively, and the time after labelling (in days) is indicated as 0, 1, 2 and 3. The green and red colours mean 13C depletion and 13C enrichment, respectively, as indicated by the colour scale above (in which the values are multiplied by 100 for clarity).

Download figure to PowerPoint

The contribution of recent photosynthates to respiration of non-photosynthetic organs in the light

The term inline image of Eqn 5 is expected to vary after CO2 labelling. For example, if the labelling CO2 is 13C depleted, the CO2 produced by respiration is expected to be strongly 13C enriched relative to fixed CO2 because of the lag phase needed to renew respiratory pools. In other words, the use of ‘old’, 13C-enriched carbon by respiration will artificially lead to a large negative inline image term (this is illustrated in Fig. 3). That is, enp is artificially decreased by the decarboxylation of old, 13C-enriched carbon. In the following, we then denote it as eapp and use enp for the intrinsic respiratory fractionation (which is independent of the 13C abundance of the C source).

image

Figure 3. The relationship between the proportion (in %) of recent carbon in mesocosm night-respired CO2 (xnight, Eqn 7) and that in CO2 respired by non-photosynthetic organs in the light at the mesocosm level (xday, Eqn 6), under progressive labelling with 13C-depleted CO2. Closed symbols: chamber 1 (200 µmol mol−1 CO2); open symbols: chamber 2 (1000 µmol mol−1 CO2). Continuous line: 1:1 line. The (0,0) point on the left-hand side corresponds to the pre-labelling conditions (day 0).

Download figure to PowerPoint

We may use the apparent inline image value to calculate the proportion of ‘old’ and recent carbon in CO2 evolved by non-photosynthetic organs in the light. The proportion of recent carbon in CO2 is denoted as xday in the following. The isotope composition of respired CO2 coming from ‘old’ carbon is approx. inline image where, again, enp is here the ‘intrinsic’, δ13C-independent, respiratory fractionation, while that of new carbon is approx. inline image where δo′ is the new (labelled) isotope composition of outlet air. The resulting isotope composition of respiratory CO2 is then:

  • image

that simply re-arranges to: inline image, that is,

  • image

inline image in the light cannot be calculated nor measured. We take advantage here of the definitions (see the above section):

  • image

where the symbol ‘prime’ indicates values obtained after labelling. As inline image and inline image is assumed constant before and after labeling, we have:

  • image

In order to get an estimate of xday, we may multiply its numerator and denominator by inline image, giving:

  • image(6)

xday is then calculated using the inline image value calculated with Eqn 5 before and after (symbol ‘prime’) labelling, and the value of inline image measured at night, when only respiration occurs (it is equal to –P measured at night). Such a value of inline image is somewhat overestimated because photosynthetic organs respire at night; hence, the xday value obtained with Eqn 6 is slightly overestimated.

The contribution of recent photosynthates to mesocosm respiration in darkness

The carbon isotope composition of CO2 respired by the whole mesocosm (photosynthetic and non-photosynthetic organs) in darkness is denoted as δn. The isotope fractionation associated with night respiration of the mesocosm, with respect to CO2 fixed in the previous light period is then:

  • image(7)

where inline image is the isotope composition of mesocosm-level net fixed carbon, and Δ is the mesocosm-level net photosynthetic fractionation. The subscript ‘n’ is used here to distinguish the respiratory isotope fractionation in darkness from that in the light (denoted as e; see above).

After labelling, the proportion of recent carbon in evolved CO2 in darkness may be calculated. Such a proportion is denoted as xnight below, and is given by (Schnyder 1992; Schnyder et al. 2003; Nogués et al. 2004):

  • image(8)

where the superscripts ‘after’ and ‘before’ refer to after and before labelling, respectively.

Covariation analyses

The covariation analysis was done following the isotopomic array representation of Tcherkez, Ghashghaie & Griffiths (2007). In the present study, the isotope composition of the different plant fractions (starch, sucrose, etc.) and respired CO2 was recalculated to isotope ratios times 100. The intensity of the red or green colour represents the strength of the natural 13C enrichment or depletion, respectively. Both the drawing of the array and the clustering analysis were done with the MeV 4.1 software (Saeed et al. 2003). The clustering of Fig. 4 is based on the cosine correlation method. The results are similar to other correlation methods, such as the Euclidean distance. To introduce mesocosm day-respired CO2 into the correlation analysis, the estimated isotope composition (δ13C) of day-respired CO2 was calculated with the values of inline image obtained with Eqn 5, as follows: inline image, where δo is the isotope composition of mesocosm outlet air, and Δ is the mesocosm-level net photosynthetic fractionation (measured on-line). It should be noted that in Fig. 2, eapp is negative (favours 13C) because of the breakdown of ‘old’, 13C-enriched photosynthates. In other words, after the start of labelling, day-respired CO2 is enriched compared to fixed CO2.

image

Figure 2. Time-course of the respiratory contribution of non-photosynthetic organs to the mesocosm-level photosynthetic fractionation (Eqn 5) after labelling with 13C-depleted CO2. Day 0 is the value just before labelling, day 1 is the first day of labelling and so on. For convenience reasons, the opposite value of inline image (minus sign) was plotted here because it is a negative value (pre-labelling carbon is 13C enriched and so is evolved CO2 in the light). The values of P, co and δo used here were that in the photosynthetic steady state. Inset: the corresponding proportion (in %) of recent carbon in evolved CO2 (Eqn 6). Closed symbols: chamber 1 (200 µmol mol−1 CO2); open symbols: chamber 2 (1000 µmol mol−1 CO2).

Download figure to PowerPoint

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Day-respired CO2

The isotopic contribution of (photo)respiration to the net photosynthetic carbon isotope fractionation, usually written as inline image, is here abbreviated as d*/ca. At the mesocosm level, this term correponds to both leaf (photo) respired CO2 and CO2 evolved by other non-photosynthetic organs. It may be calculated using the intercept of the response curve of 〈Δ〉 = ca(b − Δ) to net assimilation A (Eqn 2). Such response curves are shown in Fig. 1. Both chambers showed a linear relationship for low values of A. For chamber 2 (high CO2 growth conditions), the 〈Δ〉 value did not increase linearly at higher A values, and the slope increased at lower stomatal conductances (Eqn 2). Accordingly, the plot of Δ versus A shows indeed that the carbon isotope discrimination decreased slightly at high A values, indicating a lower ci/ca (Fig. 1, inset). Otherwise, stomatal aperture increased slowly with A as evidenced by the general positive trend between Δ and A (Fig. 1, inset).

The isotopic contribution of (photo)respiration at both the leaf and mesocosm scale was then calculated with the data of Fig. 1; the results are shown in Table 2. Equations 2 and 3 gave very similar results. When expressed on the same scale (at near-ambient CO2 mole fraction, i.e. 400 µmol mol−1 as ca), the isotopic contribution was larger under high than low CO2, while both being in the range 1–3‰. The relative effect was higher when growth CO2 mole fraction was used to compute d*/ca (2–3‰ versus 1‰ at 200 and 1000 µmol mol−1, respectively). However, the d*/ca value included the photorespiration effect that was different at 200 and 1000 µmol mol−1 CO2. With typical values of * (f = 11‰, Tcherkez 2006; Lanigan et al. 2008; Γ* ∼ 40 µmol mol−1, Brooks & Farquhar 1985), the contribution of respiration inline image (Eqn 1) was very similar in both chambers, roughly around 0.5 ± 0.3‰ regardless of growth CO2 mole fraction (Table 2).

Table 2.  Values of the (photo)respiratory contribution d*/ca to net photosynthetic fractionation at the mesocosm and leaf scale
MethodΓd*/400d*/cainline image
  1. In this table, ca stands for atmospheric CO2 mole fraction under growth conditions, that is, 200 (chamber 1) or 1000 (chamber 2) µmol mol−1. Γ is the CO2 compensation point of net CO2 assimilation (in µmol mol−1); it was obtained with A/CO2 curves at the mesocosm level. On the right-hand side, the pure respiratory value was calculated by substracting the photorespiratory term *, using arbitrary typical values of 40 µmol mol−1 for Γ* and 11‰ for f. Between brackets: sensitivity coefficients of the calculated pure respiratory value with respect to Γ* (in ‰ mol µmol−1) and f (in ‰‰−1), respectively. The sign of the values given above follows the convention of eqn 1 of Farquhar et al. (1982), that is, positive values mean fractionations against 13C.

With b = 29‰
Chamber 1 (ca = 200 µmol mol−1)
Eqn 2 and Fig. 11.513.020.410.82 (−0.06, −0.2)
Eqn 3 and Fig. 1, inset591.503.010.400.81 (−0.06, −0.2)
Chamber 2 (ca = 1000 µmol mol−1)
Eqn 2 and Fig. 13.091.231.990.76 (−0.01, −0.04)
Eqn 3 and Fig. 1, inset792.681.071.580.63 (−0.01, −0.04)
With b = 27‰
Chamber 1 (ca = 200 µmol mol−1)
Eqn 2 and Fig. 11.442.880.340.68 (−0.06, −0.2)
Eqn 3 and Fig. 1, inset591.202.410.110.21 (−0.06, −0.2)
Chamber 2 (ca = 1000 µmol mol−1)
Eqn 2 and Fig. 12.971.191.870.75 (−0.01, −0.04)
Eqn 3 and Fig. 1, inset792.290.911.190.47 (−0.01, −0.04)

Because ca was larger in chamber 2, the similar inline image value probably came from a higher Rday value. In fact, respiration rates in the dark were more than 80% higher at 1000 µmol mol−1 CO2 (6.3 µmol m−2 s−1 compared to 3.4 µmol m−2 s−1 at 200 µmol mol−1 CO2). This arose from both a higher specific respiration rate (6.8 versus 5.4 nmol g−1 DW s−1) and higher plant biomass (5.6 versus 3.9 g DW per plant) at 1000 compared to 200 µmol mol−1 CO2. Still, both d* values were (slightly) positive, showing that evolved CO2 was 13C depleted in the light (e > 0) with an average inline image value within the 0.6–0.8‰ range (Table 2, right column). These results are clearly sensitive to the 12C/13C fractionation associated with carboxylation (b value: see Eqns 2 & 3). While it is currently believed that b = 29‰, we also calculated the d* with b = 27‰ (Table 2, lower): the inline image value was then lower, but fractionation against 13C remained positive, within the 0.2–0.7‰ range.

Night-respired CO2

The night respiration rate at the mesocosm level (Nvalue in the dark, see Materials and methods) was 3.4 and 6.3 µmol m−2 s−1 in chambers 1 (200 µmol mol−1 CO2) and 2 (1000 µmol mol−1 CO2), respectively. The carbon isotope composition of CO2 within the mesocosm or evolved by individual organs in darkness was measured, and the results are shown in Table 3, where the δ13C values are also expressed as apparent fractionations with respect to net fixed CO2 (between parentheses). At the mesocosm level, night-respired CO2 was slightly 13C enriched, so was leaf-respired CO2, while roots produced 13C-depleted CO2, and stem respiration did not seem associated with an apparent fractionation. Such a pattern was very similar in both chambers. While there are not enough data to close the isotopic mass balance in our study, it is likely that 13C-enriched, dark-respired CO2 of the mesocosm was dominated by leaf respiration, which was the only 13C-enriched signal (Table 3).

Table 3.  Carbon isotope composition of CO2 evolved by the mesocosm or intact individual organs
δ13C of respired CO2 in darkness (‰)
MesocosmRootsStemMature leavesYoung leaves
  1. The isotope composition of inlet CO2 during growth was −3.5‰, and so in the steady state, the δ13C value of atmospheric CO2 (δo) was on average +0.5‰ (chamber 1, 200 µmol mol−1 CO2) and +0.7‰ (chamber 2, 1000 µmol mol−1 CO2). The values are average obtained at days 32–34 when stands were closed and plants were at the three fully expanded leaves stages. Between brackets: calculated respiratory fractionation with respect to mesocosm-level net fixed CO2 (Eqn 7). Positive values indicate fractionations against 13C.

Chamber 1 (ca = 200 µmol mol−1):
−19.6 ± 2.1−26.6 ± 2.0−23.0 ± 0.2−13.8 ± 1.0−15.7 ± 1.0
[−2.7][+3.2][−0.5][−9.9][−8.0]
Chamber 2 (ca = 1000 µmol mol−1):
−24.4 ± 2.1−28.5 ± 1.0−26.1 ± 0.6−19.0 ± 1.2−20.5 ± 1.0
[−0.5][+3.4][+0.9][−6.4][−4.8]

Day-respired CO2 after labelling

The isotopic gas exchange of the mesocosm was followed during a labelling experiment that used 13C-depleted CO2 (−44.5‰) while maintaining the CO2 mole fraction (200 and 1000 µmol mol−1). The isotopic contribution of respiration of non-photosynthetic organs to the isotopic CO2 exchange in the light (denoted as inline image, Table 1) was calculated using the ‘deviation’ from the expected Δ (Eqn 5). The results are shown in Fig. 2 in which, for convenience, inline image (positive value) was represented. All the values are associated with a large standard error, simply because very slight variations in P (the net mesocosm gas exchange, in µmol s−1) or inline image propagate into large uncertainties in inline image. In other words, it is not possible to compare directly the initial value at day 0 and the isotopic contribution d*/ca computed above (see also Materials and methods).

That said, it is apparent that CO2 evolved in the light was 13C enriched, that is, respiration used carbon atoms that were fixed before labelling. The minimal proportion of ‘new’ carbon in day-respired CO2 was calculated (Eqn 6; Fig. 2, inset). Both chambers behaved similarly with a progressive turnover of day-respired CO2 which reached a maximum near 40% (chamber 2, at 1000 µmol mol−1) and 60% (chamber 1, at 200 µmol mol−1). That is, mesocosm respiration was fed by: (1) current photosynthates via a pool that had a half-life time of several hours (accounting for 40–60% of total respiration); and (2) stored carbon with a half-life time in the order of several days (accounting, respectively, for the remaining 40–60% of total respiration).

The comparison of day- and night-respired CO2 by the mesocosm

The δ13C value of CO2 evolved by the mesocosm in the dark was also measured after labelling. As night-respired CO2 is intrinsically 13C enriched (see above and Table 3) while day-respired CO2 was intrinsically 13C depleted (see above and Table 2), it is more convenient to compare the proportions of ‘new’ carbon in respired CO2 (Eqns 6 & 8) instead of δ13C values. Such a representation is shown in Fig. 3. While being somewhat noisy, day- and night-respired CO2 followed similar carbon sources (close to the 1:1 line). This agrees with the covariation analysis carried out with the whole set of δ13C values in CO2 and metabolites, as shown in Fig. 4. When represented as an isotopomic array, it appears that the closest relative of day-respired CO2 (for the mesocosm) was night-respired CO2 of both the mesocosm and individual organs (Fig. 4, bottom). Within the same cluster (Fig. 4, left), the second-order relative of respired CO2 was stem and leaf sucrose, suggesting that, unsurprisingly, sucrose was a major source of carbon for respiration. Fructose, glucose, starch and organic matter belong to a different cluster.

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

While there is currently a growing literature on the carbon isotope composition of CO2 evolved in the dark by either leaves, roots or ecosystems, less is known on the isotope composition of day-respired CO2 (Ghashghaie et al. 2003; Bowling et al. 2008). This lack of knowledge stems from the technical difficulties associated with the measurement of the pure, respiratory isotopic signal. Such difficulties are independent of the scale of interest (leaf to ecosystem). Disentangling the (photo)respiratory contribution to the net isotopic fractionation usually requires assumptions on either its magnitude or associated fractionation. Here, we manipulated both the atmospheric CO2 mole fraction (two growth conditions) and the isotope composition of inlet CO2 (labelling), and took advantage of the mesocosm gas exchange facility (Schnyder et al. 2003). We used new simple equations (Materials and methods; Table 1) and Δ versus A curves to gain information of the carbon isotope composition of respiratory CO2 evolved in the light, without prerequisite assumptions.

Is day-respired CO213C depleted or 13C enriched?

When plotted against A, the ca(b − Δ) intercept directly gave the d* value. Scaled to atmospheric CO2 mole fraction of interest (d*/ca), this gave the (photo)respiratory component of mesocosm photosynthetic fractionation (Eqn 2; Fig. 1). Our results show that day-(photo)respired CO2 was 13C depleted as compared to net fixed CO2, and this difference depended on growth CO2 conditions (Table 2). However, the contribution of photorespiration was dissimilar under different CO2 conditions, and the estimated pure respiratory signal was close to +0.7‰ in both chambers (Table 2, with b = 29‰), that is, day-respired CO2 was slightly 13C depleted. We, nevertheless, recognize that such a value depended upon the chosen values of f (photorespiratory fractionation) and Γ*, that is, they were assumed to be independent of growth CO2 mole fraction. Both assumption are nevertheless reasonable (Thomas et al. 1993; Lanigan et al. 2008), and our sensitivity analysis shows that variations in Γ* and f variations have little effect on the pure respiratory signal (Table 2). It may be argued that day-respired CO2 is 13C enriched at the leaf level, as shown by isotopic analyses of CO2 evolved from darkened leaves during the light period (see, e.g. Hymus et al. 2005). It is believed that such a 13C enrichment comes from the metabolism associated with light-enhanced dark respiration (Barbour et al. 2007; Gessler et al. 2009). In addition, our value (depletion by 0.7‰) agrees with the data from previous investigations: in Senecio, Ghashghaie et al. (2003) found an intercept (Δ versus A/ca curve) of around 1.5‰, which corresponds to a pure respiratory component of +0.31‰, with their values of f = 11‰ and Γ* = 39 µmol mol−1. This is also in agreement with the theoretical study of Tcherkez et al. (2004), in which the commitment of 13C-depleted triose phosphates to mitochondrial respiration in the light led to 13C-depleted CO2. That said, measurements under 2% O2 (non-photorespiratory conditions) gave an intercept of about 1‰, and such a value indicates a much larger respiratory component (Ghashghaie et al. 2003) than the value of 0.31‰ quoted above. In fact, there is now a body of evidence that day respiratory metabolism is affected by the photorespiration rate (Tcherkez et al. 2008 and for a review see Noguchi and Yoshida 2007) and likely therefore, so is the isotopic contribution of evolved CO2 to net photosynthetic fractionation.

The similar values of 0.3–0.8‰ associated with the respiratory contribution to photosynthetic fractionation at both the leaf and mesocosm level show that presumably, CO2 evolved by leaves and non-photosynthetic organs was isotopically similar. With typical mesocosm values of Rday = 3 µmol m−2 s−1 and k = 0.1 mol m−2 s−1, the respiratory fractionation (usually denoted as e) is then about 5 ± 1‰ at the leaf level under a CO2 mole fraction of 400 µmol mol−1. This value agrees with the theoretical estimation of Tcherkez et al. (2004) (see also the discussion below).

Such a pattern is in clear contrast with night-respired CO2 which is 13C enriched by several per mil in leaves (Ghashghaie et al. 2003 and references therein, and Table 3) and ca. 0.6‰ (Schnyder et al. 2003) to 3‰ (Table 3) in the mesocosm. At the forest level, night-time respired CO2 appears slightly 13C enriched compared to photosynthates with, however, considerable variations (Bowling, Baldocchi & Monson 1999; Cai et al. 2008). Under our conditions, the 13C enrichment of mesocosm dark-evolved CO2 came from leaves that were indeed the sole organs to produce 13C-enriched CO2. As such, leaves probably dominated the CO2 production in the night in the present study. This scenario agrees with another study on sunflower (Klumpp et al. 2005) in which night-time-respired CO2 was 13C enriched in shoots and 13C depleted in roots, suggesting that the slight 13C enrichment of night-respired CO2 at the stand scale originated from shoots.

The day/night oscillation of the isotopic composition of CO2 respired by the mesocosm may come from: (1) the decrease of the contribution of leaves because of the inhibition of leaf respiration by light (Atkin et al. 2000); (2) the decarboxylation of 13C-enriched malate in the dark in leaves (Gessler et al. 2009); (3) the change of respiratory carbon source (13C-enriched starch at night and 13C-depleted sucrose in the light) (Gessler et al. 2008); and (4) a change of metabolic pathways or commitments so that 12C/13C fractionations of respiratory enzymes vary. While assumptions (1) and (2) are likely and supported by experimental results, assumptions (3) and (4) are not demonstrated yet and are addressed below.

Are day- and night-respired CO2 correlated?

We used here isotopic labelling to make the respiratory isotopic signal to vary: the inlet CO2 source to the mesocosm was changed from −3.5‰ to −44.5‰. The day respiratory contribution to the photosynthetic fractionation was then calculated with Eqn 3. It should be emphasized that such a value included the production of respired CO2 by non-photosynthetic organs, as we use the net photosynthetic fractionation of leaves in Eqn 3. Unsurprisingly, the respiratory component was associated with 13C-enriched CO2 (Fig. 2), clearly showing the contribution of 13C-enriched, old carbon sources to day respiration. The calculated proportion of ‘new’ carbon was indeed 40–60% (Fig. 2, inset). Similar values have already been obtained at both the leaf (Nogués et al. 2004), plant (Lehmeier et al. 2008), mesocosm (Schnyder et al. 2003) and ecosystem scale (Gamnitzer, Schäufele & Schnyder 2009) in the dark. Our data show that: (1) day- and night-respired CO2 had a very similar turnover pattern (Fig. 3); and (2) the covariation analysis indicated that the closest relative of day-respired CO2 was night-respired CO2 (Fig. 4). We therefore conclude that both day and night respiratory substrate pools were likely to be fed by the same carbon source. The latter may comprise several components: the kinetics of CO2 turnover indeed suggest that at least two carbon sources provide substrates to respiratory metabolism. It has been argued that one of them arises from carbohydrates because of the respiratory quotient of 1 (Nogués et al. 2004). Our data support this assumption, because the closest relative to evolved CO2 was stem (i.e. mainly phloem) and leaf sucrose (Fig. 4).

While both day- and dark-respired CO2 comprised always a lower proportion of ‘new’ carbon under high CO2 (chamber 2) than under low CO2 (chamber 1) (Fig. 3), the absolute decarboxylation rate of ‘new’carbon was similar under both conditions: it was 60% × 3.4 µmol m−2 s−1 = 2.04 µmol m−2 s−1 at low CO2, and 40% × 6.3 µmol m−2 s−1 = 2.52 µmol m−2 s−1 at high CO2. Therefore, the rate at which recently fixed carbon fed respiration did not vary with CO2 conditions, while the rate of remobilization increased as CO2 (and carbon availability) increased. Consistent with this are observations that when pools of respiratory intermediates (such as citrate) and storage molecules (such as starch) increase under high CO2 conditions at fixed N supply, the specific N content (Bernacchi et al. 2007) and amino acid pools were identical or even smaller (Geiger et al. 1999; Li et al. 2008). Such a metabolic effect tends to impede the turnover of respiratory metabolites, as evidenced here (Fig. 2).

Why is day-respired CO213C depleted?

While assumption (2) above (decarboxylation of 13C-enriched material in the dark) has received strong support in the literature, it is likely that both day- and night-evolved CO2 originate from similar substrates (see just above). In fact, the involvement of the decarboxylation of 13C-enriched malate in leaves has been shown to last less than half an hour (Barbour et al. 2007). Therefore, different metabolic processes probably explain why the natural 13CO2 abundance is dissimilar in the light and in the dark. One of them is the inhibition of leaf respiration that is accompanied by the decrease of both the pyruvate dehydrogenase and TCA cycle activity (Randall et al. 1990; Hanning and Heldt 1993; Tcherkez et al. 2005, 2008, and for a review, see Hurry et al. 2005). These two metabolic steps fractionate against 13C (Melzer and Schmidt 1987; Tcherkez & Farquhar 2005), and the associated isotope effects are expected to increase as the metabolic commitment decreases (O’Leary 1980), thereby depleting evolved CO2 in 13C. By contrast, the δ13C value of CO2 evolved by roots under continuous darkness has been shown to be independent of the carbon source availability (Bathellier et al. 2009). This suggests that the day/night transition does not induce major changes in the root-respired isotopic signal, unless the δ13C of root-imported sucrose (carbon input) varies. Because leaf respiration seemed to be a major component of mesocosm respiration in our study, it is therefore likely that the inhibition of leaf respiration by light contributed to the 13C depletion of mesocosm CO2.

We nevertheless recognize that day/night cycles are accompanied by circadian variations of the 13C abundance in sucrose (Tcherkez et al. 2004; Gessler et al. 2008), with 13C-enriched values at night and 13C-depleted values in the light. Such variations are caused by the 12C/13C isotope effect of aldolases (Gleixner & Schmidt 1997), that catalyse the production of fructose-1,6-bisphosphate from triose phosphates, thereby depleting day sucrose and enriching transitory starch (and night sucrose). Such a circadian variation in sucrose that feeds non-photosynthetic organs is likely to contribute to further deplete mesocosm day-respired CO2 in 13C, because non-photosynthetic organs were taken into account in our estimate of day respiratory fractionation.

Perspectives

Our results indicate that the respiratory isotopic signal is very dynamic and follows day/night variations, which correlate predominantly with leaf-level 13C signals, and this appears to be unaffected by growth CO2 conditions. In that sense, the mesocosm can be considered as a big leaf under our experimental conditions. We, nevertheless, gave little attention to other parameters that may change this pattern (such as temperature, nitrogen availability, etc.) through an effect on the root/shoot ratio, the growth rate and the δ13C value of evolved CO2. In fact, the isotope composition of night-time-respired CO2 is sensitive to temperature, vapour pressure deficit or light level (Cai et al. 2008). In addition, the results can certainly not be extrapolated to natural forest ecosystems because trees comprise woody organs (branches and trunks) that have particular isotopic signals. In natural conditions, CO2 respired by heterotrophic organs of trees is ordinarily 13C enriched (Brandes et al. 2006; Gessler et al. 2007; Maunoury et al. 2007) with noticeable diel variations (Kodama et al. 2008). Such a pattern is in clear contrast to herbaceous plants in which heterotrophic organs produce 13C-depleted CO2 (see references above and the present study). Therefore, further studies are needed to determine whether day-respired CO2 is similarly 13C depleted in natural ecosystems.

On-line carbon isotope discrimination measured at the leaf level is often used to gain information on, for example, mesophyll conductance, and quite frequently, the respiratory contribution to the net photosynthetic carbon isotope fractionation is thought to be negligible (Warren 2006; Flexas et al. 2007). Our results suggest that this very contribution may not be negligible at the mesocosm level (Table 2) particularly when the δ13C value of CO2 used for isotopic measurements strongly differs from that of growth CO2.

ACKNOWLEDGMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

S.N., C.P., A.B., G.L., C.B. and C.M. acknowledge the financial support of the European Community's Human Potential Program NETCARB through postdoctoral grants. The authors thank Max Hill for high-performance liquid chromatography purification of sugars.

The present work was supported by the European Community's Human Potential Program under contract HPRN-CT-1999-00059, NETCARB, coordinated by Jaleh Ghashghaie.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Supporting Information

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Figure S1. Typical A/ca curves at the mesocosm level. The growth CO2 conditions were 200 μmol mol−1 (chamber 1, closed symbols) or 1000 μmol mol−1 (chamber 2, open symbols).

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
PCE_2115_sm_f1.doc55KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.