A New Approach to Measure Gross CO₂ Fluxes in Leaves. Gross CO₂ Assimilation, Photorespiration, and Mitochondrial Respiration in the Light in Tomato under Drought Stress

Abstract

We developed a new method using ¹³CO₂ and mass spectrometry to elucidate the role of photorespiration as an alternative electron dissipating pathway under drought stress. This was achieved by experimentally distinguishing between the CO₂ fluxes into and out of the leaf. The method allows us to determine the rates of gross CO₂ assimilation and gross CO₂ evolution in addition to net CO₂ uptake by attached leaves during steady-state photosynthesis. Furthermore, a comparison between measurements under photorespiratory and non-photorespiratory conditions may give information about the contribution of photorespiration and mitochondrial respiration to the rate of gross CO₂ evolution at photosynthetic steady state. In tomato (Lycopersicon esculentum Mill. cv Moneymaker) leaves, drought stress decreases the rates of net and gross CO₂ uptake as well as CO₂ release from photorespiration and mitochondrial respiration in the light. However, the ratio of photorespiratory CO₂ evolution to gross CO₂ assimilation rises with water deficit. Also the contribution of re-assimilation of (photo) respiratory CO₂ to gross CO₂ assimilation increases under drought.

Water deficit limits plant growth and productivity because it decreases net CO₂ assimilation due to reduced stomatal conductance for CO₂ and/or because of non-stomatal effects like inhibition of enzymatic processes by changes in ionic or osmotic conditions (Lawlor, 1995). At high light intensities the lowered consumption of redox equivalents in the Calvin cycle makes it necessary to degrade photosynthetic electrons in processes other than CO₂ fixation to avoid photo-inhibition. Sharkey et al. (1988) showed that the activity of photosystem II can be regulated in a way that the rate of electron transport matches the capacity of the electron consuming reactions and that linear electron transport depends not only on light intensity and CO₂ concentration but also on the O₂ concentration. Oxygen can function as alternative electron acceptor directly in the Mehler reaction or indirectly in photorespiration (Badger, 1985).

By combined measurements of O₂ and CO₂ gas exchange it should be possible to investigate the distribution of photosynthetic electrons between the electron consuming reactions CO₂ assimilation, photorespiration, and Mehler reaction (Haupt-Herting, 2000). The influence of drought stress on photosystem II activity, gross O₂ evolution, and gross O₂ uptake in tomato (Lycopersicon esculentum Mill. cv Moneymaker) plants has been published elsewhere (Haupt-Herting and Fock, 2000). This paper deals with the corresponding carbon fluxes determined by a new CO₂ gas exchange method.

The magnitude of photorespiration and the role of mitochondrial respiration in the light under drought stress are still unclear. There are studies where photorespiration decreases under drought stress (Thomas and André, 1982; Biehler and Fock, 1995; Tourneux and Peltier, 1995) as well as studies where it increases (Renou et al., 1990; Biehler and Fock, 1996) or is not influenced at all (Stuhlfauth et al., 1990). According to Bradford and Hsiao (1982) respiration in the light declines with water deficit as dark respiration does. On the other hand, Lawlor (1995) assumes that dissimilation is stimulated under drought stress. However, the contribution of mitochondrial respiration to CO₂ release or O₂ uptake at photosynthetic steady state has not been resolved yet.

Photorespiratory CO₂ evolution is accompanied by CO₂ uptake in the Calvin cycle and CO₂ release from mitochondrial respiration in the light, whereas photorespiratory O₂ uptake is masked by O₂ evolution at photosystem II and O₂ consumption by Mehler reaction and mitochondrial respiration. Therefore, the determination of the rates of photorespiratory CO₂ evolution and O₂ uptake is difficult.

Rough calculations of photorespiration have been tried by different methods in the past (Jackson and Volk, 1970; Catzky et al., 1971). Progress in photorespiration research was made by the use of the ¹⁴CO₂ isotope to separate CO₂ fluxes into and out of leaves in an open gas exchange system under steady-state conditions (Ludwig and Canvin, 1971). In these experiments, a leaf is illuminated in ¹²CO₂ until steady state is reached, and then ¹⁴CO₂-labeled CO₂ is provided. From the uptake of ¹⁴CO₂ and the internal concentrations of ¹⁴CO₂ and ¹²CO₂ the rates of gross CO₂ assimilation, originating from external CO₂ and from re-assimilation, gross CO₂ evolution and re-assimilation were calculated (Gerbaud and André, 1987; Stuhlfauth et al., 1990).

Some authors use the labeling of the CO₂ evolved after illumination in air containing ¹⁴CO₂ to calculate CO₂ evolution rates under photorespiratory and non-photorespiratory conditions (Bauwe et al., 1987). These measurements make it possible to separate the contribution of primary products and end products to the photorespiratory as well as to the respiratory CO₂ release (Pärnik and Keerberg, 1995).

The determination of carboxylation and oxygenation rates of Rubisco from fluorescence measurements, CO₂ gas exchange, and Rubisco kinetics has been described by Laisk and Sumberg (1994). This method can be used to calculate not only the rates of CO₂ assimilation and photorespiration but also the rate of mitochondrial respiration in the light (Laisk and Loreto, 1996). For this, the plastidic CO₂ concentration and the CO₂ transport resistance in mesophyll cells are required.

To address the problem of sources and sinks for CO₂ and O₂ in plants, we present a new method, based on considerations from Gerbaud and André (1987), using ¹³CO₂ and mass spectrometry to determine CO₂ fluxes under conditions of steady-state photosynthesis. The method was used to determine the rates of net CO₂ uptake, gross CO₂ assimilation, photorespiratory CO₂ release, and mitochondrial respiratory CO₂ evolution in the light by attached leaves of tomato under different light intensities and at varying drought stresses.

RESULTS

Signal Curve Characteristics

In Figure 1, the mass spectrometric signal curves for ¹²CO₂ and ¹³CO₂ after switching from the gas stream containing ¹²CO₂ to the gas stream containing ¹³CO₂ with an empty cuvette and with a leaf in the cuvette in the dark or in the light, respectively, are shown. The ¹³CO₂ signal rises while the ¹²CO₂ signal falls. The ¹³CO₂ concentration reached after switching to ¹³CO₂ with a leaf in the dark (II) is the same as without a leaf in the cuvette (I), because no CO₂ fixation occurs in the dark and no ¹³CO₂ is released from the plant.

Figure 1.

Mass-spectrometric signal curves of ¹²CO₂ and ¹³CO₂ after switching to the gas stream containing ¹³CO₂ and no ¹²CO₂ with an empty cuvette (I, IV) and with a leaf in the cuvette in the dark (II, V) or during photosynthetic steady state (III, VI) under a light intensity of 850 μmol photons m⁻² s⁻¹. The system was provided with a gas stream containing ¹²CO₂ prior to application of ¹³CO₂ (gas flow 50 L h⁻¹, 350 μL L⁻¹ CO₂, 210 mL L⁻¹ O₂, 70% relative humidity, 23°C). The assimilation of ¹³CO₂ by the leaf (A_13C) and the release of ¹²CO₂ from the leaf in the light (R_12C) are used to determine the intercellular ¹³CO₂ and ¹²CO₂ concentration, respectively, and to determine true CO₂ assimilation. The curves were smoothed with a quadratic Savitzky-Golay function using an appropriate software (HP ChemStationDataAnalysis, Hewlett-Packard) and transferred to the same time axis (switching to ¹³CO₂ at t = 0 including 2 s of response time). For further details see “Results” and “Discussion.”

The illuminated leaf (III) takes up ¹³CO₂ without releasing ¹³CO₂ for almost the first 20 s (Ludwig and Canvin, 1971) so that the ¹³CO₂ concentration in the gas stream with a leaf in the cuvette is lower than the maximal ¹³CO₂ concentration, which is reached when switching to ¹³CO₂ is repeated without a leaf in the cuvette. This maximal ¹³CO₂ concentration is reached in less than 20 s (12 s for a gas flow rate of 50 L h⁻¹) after switching to ¹³CO₂. The fact that the ¹³CO₂ curves with a darkened leaf, without a leaf, and with a filter paper in the cuvette (not shown) are identical confirms that the presence of a leaf in the cuvette does not affect the gas flow characteristics and that the signals without a leaf can serve as reference for the calculations of ¹³CO₂ uptake or ¹²CO₂ evolution.

The ¹²CO₂ concentration reached with a darkened leaf (V) is higher than without a leaf (IV) because ¹²CO₂ is generated from dark respiration. The ¹²CO₂ signal with an illuminated leaf (VI) is again higher because in the light ¹²CO₂ is evolved out of the photorespiratory pathway and from mitochondrial respiration.

Exposing the illuminated leaf to ¹³CO₂ up to 20 min results in a slight but continuous increase of the ¹³CO₂ signal (data not shown). Because of the ongoing labeling of photosynthetic and photorespiratory intermediates ¹³CO₂ is released from the leaf and the visible uptake of ¹³CO₂ gets smaller. After 20 min no apparent increase in the ¹³CO₂ signal occurs any longer, and the rate of net ¹³CO₂ uptake calculated at this point of time is higher than the net ¹²CO₂-uptake rate (A) measured before switching to ¹³CO₂. A reason for this might be an incomplete labeling of intermediates of the glycolate pathway, as the evolved CO₂ does not become labeled completely, although the signal for ¹²CO₂ release decreases because part of the evolved CO₂ is released as ¹³CO₂.

Figure 2 shows the original mass spectrometric signal curves for ¹²CO₂ with and without a leaf in the cuvette after providing 3,000 μL L^{−1 13}CO₂ at photosynthetic steady state. Flushing the gas exchange system with ¹³CO₂ leads to a decrease of the ¹²CO₂ concentration from 3,000 μL L⁻¹ to nearly 25 μL L⁻¹. Within 12 s a difference between the ¹²CO₂ concentrations with and without a leaf in the cuvette can be observed. The ¹²CO₂ concentration is higher with a leaf because of mitochondrial ¹²CO₂ release. From this gross CO₂ evolution out of the leaf mitochondrial respiration can be calculated.

Figure 2.

Mass spectrometric signal curves for ¹²CO₂ after switching to the gas stream containing 3,000 μL L^{−1 13}CO₂ and no ¹²CO₂ without and with an illuminated leaf (850 μmol m⁻² s⁻¹) in the cuvette. The system was provided with a gas stream containing ¹²CO₂ prior to application of ¹³CO₂ (gas flow 50 L h⁻¹, 3,000 μL L⁻¹ CO₂, 210 mL L⁻¹ O₂, 70% relative humidity, 23°C). The curves were not smoothed but transferred to the same time axis (switching to ¹³CO₂ at t = 0 including 2 s of response time). The difference between the curves at t = 12 s was used to calculate the evolution of CO₂ by mitochondrial respiration in the light.

The rates of net CO₂ uptake, gross CO₂ assimilation, and gross CO₂ evolution measured with the new mass spectrometric isotope technique change typically in relation to the ambient CO₂ and O₂ concentration (data not shown). This shows the validity of the new method.

Effect of Light Intensity and Water Deficit on Steady-State Net CO₂ Uptake, Gross CO₂ Assimilation, and Gross CO₂ Evolution

The rates of net CO₂ uptake (A), gross CO₂ assimilation (TPS), gross CO₂ evolution (R_C), mitochondrial respiration (Resp), and photorespiration (PR) were measured at photosynthetic steady state on control and drought-stressed tomato leaves under different light intensities.

When the light intensity is increased from 90 to 850 μmol photons m⁻² s⁻¹ A and TPS rise 2.5-fold in control and weakly stressed plants (Fig. 3) because more light-generated ATP and NADPH are available for CO₂ fixation in the Calvin cycle. Lowering the leaf water potential from −0.6 MPa in controls to −1.8 MPa in severely stressed plants leads to a decrease of transpiration and leaf conductance (Haupt-Herting, 2000). This results in an internal CO₂ concentration of 112 μL L⁻¹ in severely stressed plants compared with 227 μL L⁻¹ in controls under saturating light intensities and A and TPS decrease by 82% and 72%, respectively. Severely stressed plants seem to be widely unaffected by light intensity. This means that CO₂ fixation under drought stress is not limited by light absorption but by internal CO₂ deficiency because of stomatal closure or by non-stomatal effects like inhibition of ATP synthase, photosystem II efficiency, or Rubisco activity (Cornic, 1994; Lawlor, 1995). In severely stressed tomato plants the specific activity of Rubisco, measured as ¹⁴C-incorporation into acid-stabile compounds, decreases to less than one-half of the activity measured in controls (Haupt-Herting, 2000).

Figure 3.

Rates of steady-state net CO₂ uptake (A, □), gross CO₂ assimilation (TPS, ▪), and gross CO₂ evolution (R_C, ●) of attached leaves of tomato at three different light intensities in relation to leaf water potential. Measurements were taken in an open gas exchange system using a mass spectrometric ¹³CO₂ isotope technique. Leaves were provided with air (gas flow 50 L h⁻¹) containing 210 mL L⁻¹ O₂, and 350 μL L^{−1 12}CO₂ or ¹³CO₂, respectively, at 70% relative humidity and 23°C. Points are means of at least six replicates; se ≤ 10%.

R_C, which consists of the CO₂ released from photorespiration and mitochondrial respiration, is stimulated by increasing light intensities from 2.0 μmol CO₂ m⁻² s⁻¹ under low light to 3.7 μmol CO₂ m⁻² s⁻¹ under saturating light in control plants and from 1.1 μmol CO₂ m⁻² s⁻¹ to 2.0 μmol CO₂ m⁻² s⁻¹ in severely stressed plants (Fig. 4). Severe drought stress (−1.8 MPa) results in a decrease of R_C of approximately 47% under high light, which is smaller than the relative decrease of TPS (72%).

Figure 4.

Rates of steady-state gross CO₂ evolution (R_C, ●), photorespiration (PR, ○), and mitochondrial respiration (Resp, X) of attached leaves of tomato at three different light intensities in relation to leaf water potential. Measurements were taken in an open gas exchange system using a mass spectrometric ¹³CO₂ isotope technique. Leaves were provided with air (gas flow 50 L h⁻¹) containing 210 mL L⁻¹ O₂ at 70% relative humidity and 23°C. The CO₂ concentration was 350 μL L^{−1 12}CO₂ or ¹³CO₂, respectively, for determination of gross CO₂ evolution and 3,000 μL L^{−1 12}CO₂ or ¹³CO₂, respectively, for measurement of mitochondrial respiration. Photorespiration is calculated from R_C and Resp. Points are means of at least six replicates; se ≤ 10%.

Effect of Light Intensity and Water Deficit on Mitochondrial Respiration in the Light, Photorespiration, and Re-Assimilation

The rates of mitochondrial respiration in the dark in 350 and 3,000 μL L⁻¹ CO₂ are the same in tomato leaves (Haupt-Herting, 2000). Provided that mitochondrial respiration in the light is not affected by CO₂ partial pressure between 350 and 3,000 μL L⁻¹ CO₂, Figure 4 shows the rates of mitochondrial respiration in the light at 3,000 μL L⁻¹ CO₂ and the contribution of mitochondrial respiration and photorespiration to R_C. Mitochondrial respiration in the light depends on the incident light intensity (0.82 μmol CO₂ m⁻² s⁻¹ under low and 0.93 μmol CO₂ m⁻² s⁻¹ under high light in control plants) and is in the same range as respiration in the dark (0.85 μmol CO₂ m⁻² s⁻¹). Also, mitochondrial respiration in the dark is lower in stressed plants than in controls (data not shown). Mitochondrial respiration in the light responds to drought stress with a decrease from 0.93 μmol CO₂ m⁻² s⁻¹ in controls to 0.17 μmol CO₂ m⁻² s⁻¹ under severe stress (Fig. 4).

The rate of photorespiration is higher than the rate of mitochondrial respiration and represents the main part of gross CO₂ evolution (Fig. 4). It depends on light intensity (1.2 μmol CO₂ m⁻² s⁻¹ under low light and 2.8 μmol CO₂ m⁻² s⁻¹ under saturating light for control plants) and is decreased by drought stress from 2.8 μmol CO₂ m⁻² s⁻¹ in control to 1.8 μmol CO₂ m⁻² s⁻¹ in severely stressed plants under high light (Fig. 4A). As PR is less inhibited than TPS the ratio of PR to TPS rises with increasing drought stress in tomato under all light regimes from 22% in control plants to 39% in severely stressed plants (Fig. 5A), which shows that the oxygenation reaction of Rubisco is stimulated under drought stress relative to the carboxylation reaction.

Figure 5.

A, Ratio of photorespiration (PR) to gross CO₂ assimilation (TPS); B, ratio of re-assimilation (A_R) to TPS in relation to leaf water potential at 90 (□), 400 (○), and 850 μmol photons m⁻² s⁻¹ (▵). Measurements were taken in an open gas exchange system using a mass spectrometric ¹³CO₂ isotope technique. Leaves were provided with air (gas flow 50 L h⁻¹) containing 210 mL L⁻¹ O₂ and 350 μL L^{−1 12}CO₂ or ¹³CO₂, respectively, at 70% relative humidity and 23°C.

The CO₂ evolved by the glycolate pathway is available for CO₂ assimilation in addition to the CO₂ in the atmosphere and is partly re-assimilated before leaving the leaf. The relative contribution of re-assimilation of (photo) respiratory ¹²CO₂ (A_R) to gross CO₂ assimilation is only slightly affected by light intensity (Fig. 5B). In control plants, 23% of TPS are due to A_R under low or moderate light and 29% under saturating light. This can be explained by higher rates of photorespiratory CO₂ evolution under high light conditions. As PR is less decreased under drought than TPS, the contribution of A_R to TPS rises to over 40% in weakly stressed plants, 50% in moderately stressed plants, and nearly 60% in severely stressed plants.

DISCUSSION

Critical Assessment of the New Method

The new mass spectrometric ¹²CO₂/¹³CO₂ isotope technique for the determination of accurate CO₂ flux rates into and out of the leaf is derived from ¹⁴CO₂ measurements of total stomatal CO₂ uptake by Ludwig and Canvin (1971). The method has been expanded for re-assimilation calculations by Gerbaud and André (1987) and Stuhlfauth et al. (1990). The substitution of ¹⁴CO₂ used in earlier studies by ¹³CO₂ used here has some important advantages: The discrimination of Rubisco against ¹³CO₂ (approximately 27‰) is smaller than against ¹⁴CO₂ (approximately 60‰; Farquhar et al., 1982). The radioactive isotope ¹⁴CO₂ can only be applied in tracer concentrations of approximately 0.3% of whole CO₂ content of the gas mixture (Stuhlfauth et al., 1990). So the rates for ¹⁴CO₂ uptake during photosynthesis are quite small and the rates of gross CO₂ assimilation and refixation as well as internal ¹⁴CO₂ concentration calculated from this might have large errors. The stable isotope ¹³CO₂, however, can be used in any concentration necessary, e.g. 3,000 μL L^{−1 13}CO₂ (and no ¹²CO₂) to suppress photorespiration. The high signals for ¹³CO₂ uptake facilitate correct determination of internal ¹³CO₂ concentration, refixation, and gross CO₂ assimilation. In addition, the evolution of ¹²CO₂ through stomata (Fig. 2) and the influence of external conditions on it can directly be observed, which is impossible at ¹⁴CO₂ measurements where ¹²CO₂ uptake accompanies ¹²CO₂ evolution at high external ¹²CO₂ concentrations.

In contrast to Loreto et al. (1999), who observed the emission of ¹²CO₂ in a ¹³CO₂-atmosphere with a ¹³CO₂-insensitive infrared gas-analyzer, the mass spectrometric method allows the monitoring of the ¹³CO₂ signal in addition to the ¹²CO₂ signal. Therefore, the determination of gross CO₂ evolution, gross CO₂ assimilation, and re-assimilation of (photo) respiratory CO₂ out of gas exchange data is possible and it is not necessary to calculate carboxylation and oxygenation from the electron transport rate, often detected by fluorescence measurements, and theoretical considerations of Rubisco kinetics (Di Marco et al., 1994; Laisk and Sumberg, 1994; Loreto et al., 1994).

For the calculation of re-assimilation rates the determination of internal CO₂ concentrations is necessary. Therefore, it is essential to measure leaf conductance carefully, especially under drought stress conditions where transpiration rates are small. Also, nonuniform stomatal closure, which does not occur in tomato leaves (not shown), would lead to a failure in c_i calculations (Terashima, 1992) and, therefore, result in false estimations of refixation.

In addition to intercellular refixation, intracellular refixation may occur. Gerbaud and André (1987) assume the intracellular refixation of ¹²CO₂ to be negligible because the carboxylation resistance would be dominant over the resistance for diffusion to the air space. On the other hand, high internal or stomatal resistances may favor intracellular re-assimilation (Laisk and Loreto, 1996), especially under drought stress. Intracellular re-assimilation, which is not considered by the method described here, may lead to an underestimation of gross CO₂ assimilation and CO₂ evolving reactions the extent of which is unknown.

Mitochondrial Respiration in the Light

It is widely accepted that oxidative phosphorylation occurs in the light (Sharp et al., 1984; Gerbaud and André, 1987). However, the magnitude of mitochondrial respiration in the light is still unclear (Krömer, 1995). In our experiments mitochondrial respiration in the light, which was determined at high CO₂ concentrations, is smaller than photorespiration and almost light independent (Fig. 4). Light affects the activity of the pyruvate dehydrogenase complex by a light activated protein kinase. This kinase depends on the NH₃ formed in the glycolate pathway (Randall et al., 1996). The inhibition of photorespiration by high CO₂ during the measurement of mitochondrial respiration and, as a consequence, the deficiency in photorespiratory NH₃ might result in less protein kinase activity and, therefore, in the pyruvate dehydrogenase complex being no longer inactivated in the light. This effect of non-photorespiratory conditions on mitochondrial respiration would also occur if 20 mL L⁻¹ O₂ are used to suppress photorespiration, but under these conditions light inhibition of mitochondrial respiration has often been observed (Krömer, 1995).

Mitochondrial respiration in the light is inhibited by water deficit in tomato plants (Fig. 4). According to Laisk and Sumberg (1994), the part of CO₂ evolution in the light that cannot be attributed to the oxygenation reaction is influenced by the internal CO₂ concentration. In mitochondrial respiration dissimilation of not only end products but also primary products occurs (Pärnik and Keerberg, 1995). Therefore, respiration in the light may depend on the amount of primary products, which are expected to be smaller in drought-stressed plants than in controls because of a decrease in CO₂ assimilation. This could be a reason for lower rates of mitochondrial respiration in the light under drought stress. Functions of mitochondrial respiration in the light might be the supply of ATP and carbon skeletons for synthesis reactions in the cytosol and chloroplast or the oxidation of excess redox equivalents under light or drought stress (Krömer, 1995).

In this study, high CO₂ concentrations were used to determine mitochondrial respiration in the light under non-photorespiratory conditions (Fig. 4). However, elevated CO₂ may influence mitochondrial respiration and the rates at 3,000 μL L⁻¹ CO₂ may differ from those at 350 μL L⁻¹ CO₂. The effect of high CO₂ concentrations on the rate of dark respiration seems to depend on growth conditions and varies in different plant species between 60% of inhibition and 30% of stimulation (Gonzàles-Meler et al., 1996). The inhibition of dark respiration could be the result of a direct effect on cytochrome c oxidase or succinat dehydrogenase (Gonzàles-Meler and Siedow, 1999). In tomato plants, mitochondrial respiration in the dark is not affected by 3,000 μL L⁻¹ CO₂ compared with 350 μL L⁻¹ CO₂ (Haupt-Herting, 2000). This leads to the conclusion that no inhibition of cytochrome c oxidase or other enzymes occurs. However, it is not yet fully understood how changes of CO₂ assimilation and inhibition of photorespiration may influence dissimilatory processes.

At high CO₂ concentrations, Laisk and Sumberg (1994) detected carboxylation of a substrate other than RuBP, in addition to RuBP carboxylation, that may be caused by phosphoenolpyruvate carboxylase activity. In our respiration measurements this non-RuBP carboxylation would be included in TPS and the rate of respiration in the light calculated from this TPS value is independent from the type of CO₂ assimilation. But mitochondrial respiration in the light determined by the ¹²CO₂/¹³CO₂ technique could be accompanied by CO₂ evolution from the decarboxylation of malate or pyruvate (Laisk and Sumberg, 1994).

Photorespiration and Re-Assimilation of (Photo) Respiratory CO₂

Most studies dealing with the effects of light intensity or drought stress on photorespiration used ¹⁶O₂/¹⁸O₂ mass spectrometry to determine gross O₂ uptake, which is often related to photorespiration without taking mitochondrial respiration or Mehler reaction into account (Renou et al., 1990; Tourneux and Peltier, 1995). In our investigations, however, oxygen as well as carbon fluxes have carefully been determined (Haupt-Herting, 2000; Haupt-Herting and Fock, 2000). In control plants of tomato the rate of photorespiration is 22% of the rate of gross CO₂ assimilation (Figs. 3 and 4). This matches data of ¹⁸O- or ¹⁴C-labeling of intermediates of the glycolate pathway, which proved that photorespiration is 27% of net photosynthesis at ambient CO₂ concentration in wheat (de Veau and Burris, 1989).

In the experiments presented here, photorespiratory CO₂ release is stimulated by light and rises relatively to CO₂ uptake under drought stress (Figs. 4 and 5). This is in accordance with earlier results (Thomas and André, 1982; Tourneux and Peltier, 1995; Biehler and Fock, 1996).

Studies on tomato plants showed that in addition to A and TPS the activity of photosystem II as well as gross O₂ uptake decrease in relation to water deficit (Haupt-Herting and Fock, 2000) and that the reduction of A cannot solely be caused by reduced internal CO₂ concentration. In stressed tomato plants a greater part of photosynthetic electrons flows to oxygen rather than to CO₂ than in the controls. It appears that these electrons feed Mehler reaction and the photosynthetic oxidation cycle.

The CO₂ evolved by (photo) respiration is partly re-assimilated into the Calvin cycle. Drought stress results in a remarkable increase of the contribution of re-assimilation to gross CO₂ assimilation in tomato plants (Fig. 5). Corresponding data have been found earlier in Digitalis lanata (Stuhlfauth et al., 1990). Re-assimilation of CO₂ consumes ATP and reducing equivalents, and higher rates of re-assimilation under water deficit were interpreted as contribution to the degradation of excess electrons (Fock et al., 1992). Also, re-assimilation maintains carbon flux and enzyme substrate turnover, which helps the plant to recover after rewatering (Stuhlfauth et al., 1990). Thus, photorespiration plays an important role in protecting plants from photoinhibition by using up excessive photosynthetic electrons in the glycolate pathway and by re-assimilation of (photo) respiratory CO₂.

MATERIALS AND METHODS

Plant Growth and Stress Application

Tomato (Lycopersicon esculentum Mill. cv Moneymaker; Hild, Marbach, Germany) seeds were sown individually in small pots of compost (ED 73, Einheitserdenwerk, Hameln, Germany) and then transferred to 2.5-L pots with a mixture of 10% sand in potting compost 7 d after germination. Plants were grown in a growth chamber under weak light (200 μmol photons m⁻² s⁻¹) during a 16-h-light period with 23°C in the light and 17°C in the dark with a constant relative air humidity of 70%. Plants were watered daily and regularly supplied with a commercial nutrient solution (Flori 3, Planta Düngemittel, Regenstauf, Germany). The youngest, fully expanded leaf (normally the fifth leaf from the top) of 5-week-old plants was used. Leaves of well watered plants then showed a leaf water potential of −0.6 MPa measured according to Scholander et al. (1965) with a pressure bomb (self constructed, Metallwerkstätten der Universität, Kaiserslautern, Germany). To induce an almost natural, reversible drought stress allowing the plant enough time to acclimate, irrigation was stopped 2, 5, or 8 d before measurements were taken. These treatments resulted in weak (leaf water potential −0.9 MPa), moderate (−1.3 MPa), or severe water stress (−1.8 MPa). Even severely stressed plants showed complete recovery of leaf water potential, transpiration, and net photosynthesis after rewatering.

The CO₂ Isotope Fluxes in Illuminated Leaves

To determine true CO₂ assimilation, photorespiration, and mitochondrial respiration in the light in attached leaves, we use ¹³CO₂ and mass spectrometry to measure the ¹³CO₂ flux into and the ¹²CO₂ flux out of an illuminated leaf.

Figure 6 shows the fluxes of ¹³CO₂ and ¹²CO₂ into and out of an illuminated leaf, respectively. The ¹³CO₂ isotope provided in the atmosphere (e.g. 350 μL L⁻¹ pure ¹³CO₂ and no ¹²CO₂) is taken up into the intercellular space and the mesophyll cells to be assimilated in the Calvin cycle, which is located in the chloroplasts. The ¹³CO₂ isotope is not evolved by photorespiration or mitochondrial respiration after internal cycling through primary products for the first 20 to 30 s (Ludwig and Canvin, 1971; Gerbaud and André, 1987; Pärnik and Keerberg, 1995). Therefore, net ¹³CO₂ uptake equals gross ¹³CO₂ uptake (A_13C) for the first 20 s after switching to ¹³CO₂. Photorespiration and mitochondrial respiration release ¹²CO₂ into the intercellular space (R_C). A part of this ¹²CO₂ is re-assimilated in the Calvin cycle (A_R), whereas the other part is evolved into the atmosphere. Because the ¹²CO₂ concentration is in the atmosphere and, therefore, ¹²CO₂ uptake is very small, net ¹²CO₂ evolution (R_12C) can be registered outside the leaf. Refixation of ¹²CO₂ occurs corresponding to the fixation rate of ¹³CO₂ and depends on the ratio of the internal concentrations of ¹²CO₂ to ¹³CO₂. The discrimination of ¹³CO₂ is small (27‰; Farquhar et al., 1982) and needs not be taken into account.

Figure 6.

Scheme of CO₂ fluxes into and out of an illuminated leaf provided with ¹³CO₂ in the atmosphere. The fluxes of ¹³CO₂ and ¹²CO₂ measurable outside the leaf (gross ¹³CO₂ uptake A_13C; ¹²CO₂ release R_12C) and the assumed fluxes inside the leaf (gross ¹²CO₂ release R_C; ¹²CO₂ re-assimilation A_R) are shown. For further details see text in “Materials and Methods.”

Gas Exchange Measurements

The Open Gas-Exchange System

The rates of net CO₂ uptake (A), true photosynthetic CO₂ assimilation (TPS), and gross CO₂ release (R_C) by attached leaves were determined at photosynthetic steady state in an open gas exchange system coupled to a mass spectrometer (Fig. 7). The continuous gas stream (50 L h⁻¹) passed through a humidifier and a condenser to achieve a relative air humidity of 70%. A three-way valve allows the gas stream to be switched between ¹²CO₂ and another gas stream containing the same concentration of pure ¹³CO₂. The system contained a thermostated aluminum leaf cuvette illuminated by a halogen lamp and a thermostated flow chamber (self constructed, Metallwerkstätten der Universität, Kaiserslautern, Germany) with a teflon-membrane inlet into a quadrupole mass spectrometer (5970 Series Mass Selective Detector, Hewlett-Packard, Waldbronn, Germany) where the concentrations of ¹³CO₂ (m/z = 45) and ¹²CO₂ (m/z = 44) in the gas stream were detected simultaneously and continuously. The diffusion through the Teflon-membrane and the sensitivity of the mass spectrometer were equal for both isotopes. No significant drifts in the CO₂ signals occur during the measuring time. The partial pressure of water vapor in the air was measured with a humidity sensor (HMP 233, Vaisala, Hamburg, Germany). Leaf temperature was determined with a copper-constantan-thermocouple in contact with the lower side of the leaf.

Figure 7.

Diagram of the open gas exchange system used for ¹²CO₂/¹³CO₂ measurements.

Proceeding of the Gas-Exchange Measurement and CO₂ Flux Calculations

At the beginning of an experiment the mass spectrometric signals for ¹²CO₂ (350 μL L⁻¹) without a leaf in the cuvette are registered. Then an attached leaf is placed into the cuvette and illuminated in a continuous gas stream (50 L h⁻¹) containing ¹²CO₂ until photosynthetic steady state is reached. The rates of net photosynthetic CO₂ uptake (A), transpiration (E), and leaf conductance (gs) are calculated as previously described (Biehler and Fock, 1996).

At steady-state photosynthesis, a gas mixture containing no ¹²CO₂ but ¹³CO₂ in air is suddenly supplied to the leaf for 1 min. The rate of ¹³CO₂ assimilation (A_13C) can be calculated from the gas flow rate (F, [μmol s⁻¹]), the difference in ¹³CO₂ concentration with (¹³c_o) and without a leaf (¹³c_a, [μL L⁻¹]) in the cuvette and the illuminated leaf area (a, [m⁻²]):

Re-assimilation of released ¹²CO₂ (A_R) must be taken into account when calculating the rate of true CO₂ assimilation (Fig. 6). The ¹²CO₂ isotope will be re-assimilated according to ¹³CO₂ assimilation and the ratio of internal concentrations of ¹²CO₂ and ¹³CO₂. The rate of re-assimilation of ¹²CO₂ (A_R) is then:

The internal concentrations of CO₂ can be calculated from the fluxes of CO₂ into or out of the leaf, the external CO₂ concentration, and the leaf conductance (gs) determined from transpiration measurements:

The internal ¹³CO₂ concentration is calculated from the ¹³CO₂ flux into the leaf, whereas the internal ¹²CO₂ concentration is calculated from the ¹²CO₂ flux out of the leaf as determined by the mass spectrometer.

The rate of true CO₂ assimilation (TPS), which is the sum of ¹³CO₂ assimilation (A_13C) and ¹²CO₂ re-assimilation (A_R), is given by:

and the rate of gross CO₂ release (R_C), which is the sum of photorespiration and mitochondrial respiration, can be written as:

For measurements of mitochondrial respiration in the light the leaf was provided with 3,000 μL L^{−1 12}CO₂ (to inhibit photorespiration) until steady state was reached, and then ¹²CO₂ was replaced by the same concentration of ¹³CO₂. R_C, calculated as described above, is then a measure for the rate of mitochondrial respiration, which is the only reaction pathway releasing CO₂ under these conditions. It is assumed that the rate of mitochondrial respiration in the light is not affected by CO₂ partial pressures between 350 and 3,000 μL L⁻¹ CO₂.

Measurements at different light intensities were done on the same leaf one after the other beginning with the lowest intensity. It was carefully checked that no ¹³CO₂ taken up in the previous measurement was released in the subsequent run. In these experiments ¹³CO₂ was offered for only 1 min before the gas mixture containing ¹²CO₂ was applied again. The ground signal for ¹³CO₂ was then reached within 2 min and ¹³CO₂ was not evolved from the leaf.