Nitrate photo-assimilation in tomato leaves under short-term exposure to elevated carbon dioxide and low oxygen


Arnold J. Bloom. Fax: +1 530 752 9659; e-mail:


The role of photorespiration in the foliar assimilation of nitrate (NO3) and carbon dioxide (CO2) was investigated by measuring net CO2 assimilation, net oxygen (O2) evolution, and chlorophyll fluorescence in tomato leaves (Lycopersicon esculentum). The plants were grown under ambient CO2 with ammonium nitrate (NH4NO3) as the nitrogen source, and then exposed to a CO2 concentration of either 360 or 700 µmol mol−1, an O2 concentration of 21 or 2%, and either NO3 or NH4+ as the sole nitrogen source. The elevated CO2 concentration stimulated net CO2 assimilation under 21% O2 for both nitrogen treatments, but not under 2% O2. Under ambient CO2 and O2 conditions (i.e. 360 µmol mol−1 CO2, 21% O2), plants that received NO3 had 11–13% higher rates of net O2 evolution and electron transport rate (estimated from chlorophyll fluorescence) than plants that received NH4+. Differences in net O2 evolution and electron transport rate due to the nitrogen source were not observed at the elevated CO2 concentration for the 21% O2 treatment or at either CO2 level for the 2% O2 treatment. The assimilatory quotient (AQ) from gas exchange, the ratio of net CO2 assimilation to net O2 evolution, indicated more NO3 assimilation under ambient CO2 and O2 conditions than under the other treatments. When the AQ was derived from gross O2 evolution rates estimated from chlorophyll fluorescence, no differences could be detected between the nitrogen treatments. The results suggest that short-term exposure to elevated atmospheric CO2 decreases NO3 assimilation in tomato, and that photorespiration may help to support NO3 assimilation.


ratio of net CO2 assimilation to the gross rate of O2 evolution estimated from chlorophyll fluorescence


ratio of net CO2 assimilation to net O2 evolution


difference in AQ between NO3 and NH4+ treatments


gross rate of O2 evolution estimated from chlorophyll fluorescence


rate of linear electron transport through photosystem II


non-photochemical quenching


photochemical quenching.


Foliar assimilation of nitrate (NO3) and carbon dioxide (CO2) interact through many complex pathways (Stitt & Krapp 1999; Paul & Foyer 2001). For example, the reduction of NO3 through nitrite (NO2) to ammonia (NH4+) and its subsequent assimilation to glutamate via the glutamine synthetase/glutamate oxyglutarate aminotransferase cycle (GS/GOGAT) is an energy-demanding process requiring the transfer of 10 electrons compared to four electrons for the assimilation of CO2 to carbohydrate (Turpin, Weger & Huppe 1997). Assimilation of NO3 and CO2 may compete for reductant such as ferredoxin that is produced during photosynthetic electron transport (Bloom et al. 2002) because NO2 reduction and the GS/GOGAT cycle both reside within the chloroplast where CO2 assimilation occurs. Additionally, the NH4+ used in leaf amino acid synthesis is derived both from the primary reduction of NO3 to NH4+ and from the NH4+ released during photorespiration (Keys et al. 1978; Novitskaya et al. 2002).

When C3 plants such as barley, pea, and wheat receive NO3 rather than NH4+ as the sole nitrogen source, the rate of photosynthetic electron transport often increases (Bloom et al. 1989; De la Torre, Delgado & Lara 1991; Bloom et al. 2002). Using net O2 evolution as the measure of electron transport, these increases tend to be light-dependent with the greatest differences found at high rather than low light. In tobacco, chlorophyll fluorescence measurements have indicated that the assimilation of NO3 can account for some percentage of electron transport even under low light conditions (Morcuende et al. 1998). Based on calculations from foliar C : N ratios, Foyer, Ferrario-Méry, & Noctor (2001) suggested that NO3 assimilation typically represents about 10% of photosynthetic electron flow although actual measurements of net O2 evolution can lead to higher estimates for the same species (De la Torre et al. 1991).

In addition to CO2 and NO3 assimilation, photorespiration expends a substantial amount of photosynthetic energy in C3 plants. As a consequence of the specificity of RuBP carboxylase/oxygenase (Rubisco) for both CO2 and O2, elevated CO2 or low O2 concentrations reduce photorespiration and typically lessen the electron requirement per CO2 assimilated (Stitt 1991). The substantial decrease in photorespiration under elevated CO2 or low O2 concentrations also removes the demand for photoreductant (Wingler et al. 2000). Thus, diminished rates of photorespiration may allow for more photosynthetic energy to be used in foliar NO3 assimilation (Matt et al. 2001).

The influence of elevated atmospheric CO2 concentration on interactions between NO3 and CO2 assimilation in wheat was recently examined in our laboratory (Bloom et al. 2002). Contrary to our expectation, short-term (i.e. hours) gas exchange measurements of shoots grown at ambient CO2 levels indicated that exposure to elevated CO2 decreased NO3 photo-assimilation. In longer-term experiments (i.e. days), wheat plants grown under elevated CO2 had less foliar NO3 reductase and NO2 reductase activities, and less shoot protein than plants grown under ambient CO2. Studies on Plantago major (Fonseca, Bowsher & Stulen 1997), Nicotiana tabacum (Geiger et al. 1999), Nicotiana plumbaginifolia (Ferrario-Mèry et al. 1997), and Spinacia oleracea (Kaiser et al. 2000) have also found that longer exposures (4 h to over 2 weeks) to elevated CO2 can inhibit NO3 reductase activity in shoots.

In the following study, we conducted short-term experiments to examine the influence of both low oxygen and elevated atmospheric CO2 concentrations on foliar NO3 photo-assimilation. Low oxygen conditions provide non-photorespiring conditions, and allow for a more direct assessment of photosynthetic energy demand by NO3 and CO2 assimilation, particularly at ambient CO2 concentrations, in which photorespiration is otherwise high. Nitrate assimilation was assessed using chlorophyll fluorescence measurements as well as measurements of photosynthetic gas exchange and changes in the assimilatory quotient (net CO2 assimilation/net O2 evolution). The assimilatory quotient, AQ, has been used successfully to assess NO3 assimilation in a number of species (Bloom et al. 1989; Cen, Turpin & Layzell 2001; Bloom et al. 2002).


Plant cultivation

Tomato (Lycopersicon esculentum cv. Ailsa Craig) seeds were surface sterilized for 10 min in a 25% bleach solution, washed thoroughly with water, and germinated on moist cheesecloth that was suspended over aerated nutrient solution in 3 L opaque plastic containers. Two sets of five seeds were germinated each week to provide plants of the same age and size for the gas exchange measurements. The 3 L containers were placed in a controlled environment cuvette (Conviron, Winnipeg, Canada) set at 25 °C day/18 °C night with a 16 h photoperiod and ambient CO2 concentration. The photosynthetic flux density (PFD) was 500–600 µmol m−2 s−1 at plant height. After 7 to 8 d, three to five seedlings were transferred to a larger, 19 L container. The aerated nutrient solution in both the 3 L and 19 L containers included 0.2 mm NH4NO3, 1 mm CaSO4, 0.65 mm K2HPO4, 0.35 mm KH2PO4, 1 mm MgSO4, 0.6 mm K2SO4, 0.01 g L−1 FeDPTA (sodium ferric diethylenetriaminepentaacet), and micronutrients (Epstein 1972). The nutrient solution in the 19 L containers was replenished after 7 d with a nutrient solution that was one-half strength of the original solution. Experiments were conducted on 17- to 19-day-old-plants that had two fully expanded leaves.

Laboratory protocol and experimental design

In the afternoon, approximately 15 h before an experiment, a plant was transferred from the growth chamber to the laboratory. The root system of this intact plant was sealed into an acrylic plastic cuvette in the laboratory by fitting a split rubber stopper around the stem. The root cuvette was filled with aerated nutrient solution and attached to the continuous flow nutrient system described by Nicoulaud & Bloom (1998). The nutrient solution contained either 0.2 mm KNO3 or 0.2 mm NH4Cl as the nitrogen source along with 1 mm CaSO4 and 0.5 µm K2HPO4. The following morning, the terminal leaflet and two distal leaflets of the most recently expanded leaf were sealed into a single leaf gas exchange cuvette that held the leaf perpendicular below a 1000 W metal halide lamp (Wide-Lite, San Marcos, TX, USA). Copper–constantan thermocouples were placed on the undersides of the terminal leaflet and one of the distal leaflets to monitor leaf temperature. The leaf was allowed to equilibrate for 2 h at a PFD of 300 µmol m−2 s−1 and at the CO2 and O2 concentrations of a specific treatment. The PFD was then reduced to 100 µmol m−2 s−1 for 30 min. A leaf was therefore exposed to at least 2.5 h of a particular CO2 and O2 concentration before gas-exchange and chlorophyll fluorescence measurements were made. As a consequence, we focused on longer-term responses rather than the transients in NO3 assimilation that may occur after a change in conditions (Kaiser et al. 2000).

In addition to the two nitrogen treatments (0.2 mm KNO3 or 0.2 mm NH4Cl), a leaf was exposed to one of four atmospheric mixtures in the leaf cuvette: a CO2 concentration of either 360 or 700 µmol mol−1 and an O2 concentration of either 20 (2% O2) or 210 mmol mol−1 (21% O2). For the gas exchange measurements, six to nine replicates for each of the eight treatment combinations (nitrogen form × CO2 × O2) were performed. Chlorophyll fluorescence was measured simultaneously with gas exchange, but there was slightly less replication (n = 5) for the leaves measured at 21% O2 due to equipment availability. Only one leaf was measured each day. The measurements were conducted at five different PFD levels (100, 300, 500, 800 and 1200 µmol m−2 s−1) from the lowest to the highest PFD to minimize the influence of the preceding PFD on subsequent chlorophyll fluorescence measurements. The leaf was at each PFD level for about 30 min to allow for an accurate net O2 evolution measurement using the custom O2 analyser described below.

Gas exchange measurements

An open gas exchange system previously described by Bloom et al. (1989) monitored net CO2 assimilation, net O2 evolution, and transpiration using, respectively, a commercial non-dispersive infrared CO2 analyser (Model VIA-500R; Horiba, Irvine, CA, USA), a custom-designed O2 analyser, and relative humidity sensors (Vaisala, Helsinki, Finland). The custom O2 analyser contains two cells of calcia-stabilized zirconium oxide ceramic similar to those found in an Applied Electrochemistry model N-37 M (Pittsburgh, PA, USA). Platinum electrodes are located on the inside and outside of each cell at one end. When heated to 752 ± 0.01 °C in an electric furnace, these cells become selectively permeable to O2, and a 106-nV Nernst potential per µmol difference in O2 concentration is generated between the two cells at a normal ambient O2 background of 209 700 µmol mol−1 (or 20.97% O2). As expected, the potential generated per oxygen concentration difference at 20 000 µmol mol−1 (or 2% O2) was about 10 times greater than at 209 700 µmol mol−1. In practice, this analyser can resolve O2 concentration differences to better than 2 µmol mol−1 at 21 or 2% O2 (Bloom et al. 1989).

Mass flow controllers (Tylan, Torrance, CA, USA) prepared the various gas mixtures. For the 21% O2 experiments, 2% CO2 in air from a compressed gas cylinder and CO2-free air from a 100 L storage tank were mixed to obtain the 360 and 700 µmol mol−1 CO2 concentrations. For the 2% O2 experiments, the controllers mixed 2% CO2 in nitrogen, pure oxygen, and pure nitrogen from three compressed gas cylinders. A pressure transducer (Validyne, North Ridge, CA, USA) monitored the gas flow through the leaf cuvette. The leaf cuvette was constructed from glass and Teflon-coated aluminium to minimize oxidative O2 exchange. The gas flow in the gas exchange system was humidified in a water bubbler filled with glass beads and then partially dehumidified in a condenser cooled to 6 °C before reaching the leaf cuvette. The leaf vapour pressure deficit was maintained at approximately 10 mbar. Leaf and root solution temperatures were maintained at 25 and 20 °C, respectively.

The assimilatory quotient of gas exchange (AQG), the ratio of net CO2 assimilation to net O2 evolution, was used as a measure of foliar NO3 assimilation. Transfer of electrons to nitrate and to nitrite during NO3 assimilation increases O2 evolution from the light-dependent reactions of photosynthesis, while CO2 assimilation remains similar or decreases. Thus, leaves that are photo-assimilating NO3 should exhibit a lower AQ, and differences in the AQ between NO3 and NH4+ treatments (ΔAQ) should be correlated with NO3 assimilation.

Over half a century ago, Myers (1949) verified for algae that ΔAQ depended upon NO3 assimilation. We showed over a decade ago in a wild-type barley that ΔAQ reflected the difference between the NO3 absorbed and the NO3 accumulated (Bloom et al. 1989; Bloom, Sukrapanna & Warner 1992). Moreover, in barley mutants deficient in NO3 reductase, ΔAQs did not deviate from zero. More recently, Cen et al. (2001) documented that NO3 assimilation makes up about 74% of whole-plant reductant use in white lupin (Lupinus alba) and, thus, has a much greater effect upon ΔAQ than any other metabolic process. We have shown in wheat that ΔAQ correlates positively with nitrous oxide production (which depends on NO3 assimilation, Smart & Bloom 2001), leaf protein content, and nitrate reductase and nitrite reductase activities and that ΔAQ correlates negatively with accumulation of free NO3 (Bloom et al. 2002). In summary, all available data support that ΔAQ provides a real-time and continuous measure of NO3 assimilation.

Chlorophyll fluorescence measurements

A PAM 101 fluorometer (H. Walz GmbH, Effeltrich, Germany) equipped with a xenon-arc lamp to provide a saturating light pulse (10 000 µmol m−2 s−1 for 1 s) assessed chlorophyll fluorescence of the terminal leaflet. Steady-state fluorescence (Fs) and maximum fluorescence (Fm′) were recorded at each PFD level on a chart recorder. The quantum efficiency of linear electron transport through photosystem II (φPSII) was calculated as (Fm′ − Fs)/Fm′ according to the method of Genty, Briantais, & Baker (1989). The rate of linear electron transport through PSII (JPSII) was then estimated as (φPSII·α· 0.5), where the coefficient of leaf absorptance of PFD (α) was assumed to be 0.85 and the factor 0.5 was used to account for the partitioning of energy between PSII and PSI. The assumption that JPSII = φPSII · 0.85 · 0.5 is standard (Maxwell & Johnson 2000).

The JPSII is divided by 4, based on 4 e transported per O2 evolved, to estimate the gross rate of O2 evolution (JO2) (Edwards & Baker 1993). To calculate photochemical (qP) and non-photochemical (NPQ) quenching at each PFD level, the maximum quantum efficiency of PSII [FV/Fm = (Fm − Fo)/Fm] was measured in the dark before the experiment began. The qP and NPQ were calculated as (Fm′ − Fs)/(Fm′ − Fo) and (Fm − Fm′)/Fm, respectively.

Using the simultaneous measurements of gas exchange and chlorophyll fluorescence, an AQF was determined using the ratio of net CO2 assimilation to JO2. This AQ is similar to that described earlier except that the O2 evolution for AQF reflects a gross rate rather than a net rate.

Statistical analysis

A repeated measures analysis of variance was performed using the mixed procedure in SAS (PROC MIXED; SAS Institute, Cary, NC, USA) to investigate the effects of nitrogen form (N), CO2 treatment, O2 treatment, and PFD on the gas exchange and fluorescence parameters. The PFD was considered to be a repeated factor since each leaf was measured at all five levels of PFD. Natural log or square root transformations were used where appropriate to normalize the data for a given dependent variable. Effects of the treatments and their interactions were considered significant when P = 0.05 and are presented in Table 1.

Table 1.  Analyses of variance for gas exchange and chlorophyll fluorescence parameters of tomato leaves under two nitrogen sources (0.2 mm KNO3, 0.2 mm NH4Cl), two CO2 concentrations (360 µmol mol−1, 700 µmol mol−1), two O2 concentrations (2%, 21%), and at five different irradiance (PFD) levels
Source of variationNet CO2 uptakeNet O2 evolutionAQG (CO2/O2)JPSIIAQF (CO2/JO2)NPQqP
  1. Statistically significant main effects and interactions are shown as: *P < 0.05, **P < 0.01.

O2 **** ** **
N × O2       
CO2***  **** 
N × CO2       
O2 × CO2*   *** 
N × O2 × CO2  *    
N × PFD       
O2 × PFD**************
N × O2 × PFD       
CO2 × PFD****    
N × CO2 × PFD  **    
O2 × CO2 × PFD* **    
N × O2 × CO2 × PFD     * 


Net CO2 assimilation at the higher PFD levels was greater under elevated CO2 (700 µmol mol−1) than ambient CO2 (360 µmol mol−1) for both N treatments under normal atmospheric O2 (21%) (Fig. 1a). In contrast, net CO2 assimilation was not enhanced by elevated CO2 under 2% O2 (Fig. 1b). Relative to normal atmospheric O2, 2% O2 stimulated net CO2 assimilation under ambient CO2 at low PFD, but no difference was apparent at high PFD. All of the above responses contributed to the three-way O2 × CO2 × PFD interaction (P < 0.05) in Table 1.

Figure 1.

Net CO2 assimilation and net O2 evolution in tomato leaves as a function of photosynthetic photon flux density (PFD). Plants were grown in controlled environmental chambers under ambient CO2 (360 µmol mol−1) and O2 (21%) conditions, and then measured at 360 (light symbols) or 700 (dark symbols) µmol mol−1 CO2 and under 2% (panels b & d) or 21% (panels a & c) O2. Either NH4+ (circles) or NO3 (triangles) was the sole nitrogen source during measurements. The means are shown ± SE for six to nine replicates per treatment. Some symbols are hidden by those of other treatments.

At the higher PFD levels, net O2 evolution was greater under elevated CO2 than ambient CO2 across both O2 levels (CO2 × PFD; P = 0.05), although the difference was more pronounced at 21% O2 than 2% O2 (Fig. 1c & d). The greater net O2 evolution under elevated CO2 versus ambient CO2 at 21% O2 in part reflects the greater rate of net CO2 assimilation under elevated CO2.

Net O2 evolution and JPSII estimated from chlorophyll fluorescence were both used to assess electron transport rate. Nitrogen form by itself (i.e. at all PFD levels and under all CO2 and O2 treatments) did not have a statistically significant effect on either net O2 evolution or JPSII (Table 1). Net O2 evolution at the highest PFD level under ambient CO2 and 21% O2, however, was slightly higher (11%) for the NO3 than the NH4+ treatment (Fig. 1c). Similarly, JPSII at the two highest light levels under ambient CO2 and 21% O2 was slightly higher (11–13%) for the NO3 than the NH4+ treatment (Fig. 2a). Under ambient CO2 and 2% O2, neither net O2 evolution (Fig. 1d) nor JPSII (Fig. 2b) differed significantly between the N treatments.

Figure 2.

Estimated rate of linear electron transport through photosystem II (JPSII) in tomato leaves as a function of photosynthetic photon flux density (PFD). Plant growth and measurements conditions are the same as Fig. 1. The means are shown ± SE for five to six replicates per treatment. Some symbols are hidden by those of other treatments.

The assimilatory quotients of gas exchange (AQG) were calculated at each PFD level using simultaneous measurements of CO2 and O2 exchange to assess NO3 assimilation. The AQG is the ratio of net CO2 assimilation to net O2 evolution, and differences in AQG under the two nitrogen sources should reflect the amount of NO3 assimilation. Under high PFD and ambient atmospheric conditions (21% O2, 360 µmol mol−1 CO2), the AQG was lower in the NO3 than the NH4+ treatment (Fig. 3a). No response to the N treatments was apparent under elevated CO2 at 21% O2. This indicates greater NO3 assimilation under ambient CO2 than under elevated CO2 at high PFD. Under 2% O2, the AQG was not affected by N form (Fig. 3b). These results are highlighted in Table 1 by the N × O2 × CO2 (P < 0.05) and the N × CO2 × PFD (P < 0.01) interactions. In addition to these interactions, the AQG was significantly greater under 21% O2 than 2% O2 (P < 0.01).

Figure 3.

The assimilatory quotient (AQ) in tomato leaves as a function of photosynthetic photon flux density (PFD). The AQ is calculated as the ratio of net CO2 assimilation to either net O2 evolution from gas exchange (AQG) or estimated gross O2 evolution (JO2) from chlorophyll fluorescence (AQF). Plant growth and measurement conditions are the same as Fig. 1. The means are shown ± SE for five to nine replicates per treatment. Some symbols are hidden by those of other treatments.

An assimilatory quotient (AQF) was also calculated as the ratio of net CO2 assimilation to the estimated gross rate of O2 evolution (JO2) from chlorophyll fluorescence. In contrast to AQG, no differences in AQF were apparent due to the N treatments (see Discussion). The AQF was influenced by CO2 and O2 concentration (CO2 × O2; P < 0.01). The AQF values at 21% O2 were consistently lower under ambient CO2 than under elevated CO2, whereas the AQF values at 2% O2 were fairly similar to elevated CO2 at 21% O2 (Figs 3c & d). This indicates a greater number of electrons per CO2 fixed under ambient CO2 and 21% O2 than under diminished (elevated CO2, 21% O2) or non-photorespiring conditions (2% O2).

The results for non-photochemical quenching (NPQ) included a complex four-way N × O2 × CO2 × PFD interaction (Table 1, P < 0.05). The NO3 treatment had lower values of NPQ than the NH4+ treatment at 21% O2 with the response varying by the level of CO2 and PFD (Fig. 4a). Lower NPQ in the NO3 treatment was apparent only at high PFD under ambient CO2, whereas NPQ was lower at all PFD levels under elevated CO2. No differences in NPQ occurred at 2% O2. Photochemical quenching (qP) was not affected by the N treatments (Table 1; data not shown).

Figure 4.

Non-photochemical quenching (NPQ) of chlorophyll fluorescence as a function of photosynthetic photon flux density (PFD). Plant growth and measurement conditions are the same as Fig. 1. The means are shown ± SE for five to nine replicates per treatment. Some symbols are hidden by those of other treatments.


Photorespiration expends a considerable amount of reductant and ATP from photosynthetic electron transport to re-assimilate NH4+ and to refix CO2 (Leegood et al. 1995; Wingler et al. 2000). Similarly, photo-assimilation of NO3 to NH4+ oxidizes NAD(P)H and reduced ferredoxin. Thus, if a leaf has limited amounts of these reductants, one might expect the highest rates of NO3 photo-assimilation to occur under high light and non-photorespiring conditions. As expected, we observed NO3 photo-assimilation in tomato at high, but not at low PFD, based on differences in AQG between NO3 and NH4+ treatments (ΔAQ). However, NO3 assimilation was apparent only under ambient CO2 (360 µmol mol−1) and O2 (21%) conditions, which promote photorespiration, and not under conditions where photorespiration was diminished (700 µmol mol−1 CO2, 21% O2) or negligible (2% O2). These results are consistent with those of our short-term gas exchange experiments with wheat in which elevated CO2 concentrations inhibited the photo-assimilation of NO3 at 21% O2 (Bloom et al. 2002).

Under ambient CO2 and O2 conditions, the net O2 evolution due to NO3 assimilation at the highest PFD level was about 10% based on the AQ values from gas exchange (AQG). Nitrate assimilation was not clearly reflected in the AQ values calculated from net CO2 assimilation and gross O2 evolution from chlorophyll fluorescence (AQF) (Table 1). Whereas net CO2 and O2 exchange were measured in parallel over the same leaf area, chlorophyll fluorescence was measured in the centre of only one of the three tomato leaflets. This and other factors such as slight differences in irradiance over the surface of the leaflets may complicate comparisons of chlorophyll fluorescence measured on a small area of the leaf with the whole-leaf gas exchange measurements (Haupt-Herting & Fock 2000; Ruuska et al. 2000). Simultaneous measurements of gas exchange and chlorophyll fluorescence in our laboratory on maize leaves (Zea mays) show better agreement between AQG and AQF (Asaph Cousins & Arnold Bloom, unpublished results). It is unlikely that estimates of NO3 assimilation were biased due to the use of net O2 exchange rather than gross O2 evolution through photosystem II since both measures of O2 evolution were 11–13% higher under the NO3 than the NH4+ treatment under ambient CO2 and O2 concentrations.

Values of AQ from gas exchange measurements of leaves and entire shoots (AQG) commonly range between 0.8 and 1.2 (e.g. Kaplan & Bjorkman 1980; Cen et al. 2001; Smart & Bloom 2001; Bloom et al. 2002). Theoretically, AQG can reach as high as 1.33 if a large proportion of the carbon assimilated enters into organic acids, free amino acids, or nucleic acids, whereas AQG drops below 1.0 if the shoots are assimilating NO3 or if a large proportion of the carbon assimilated enters protein, lipids, or lignin (Kaplan & Bjorkman 1980; Cen et al. 2001). Our AQG values for tomato were 0.88–1.12 under 21% O2 and slightly lower, 0.76–0.99, under 2% O2. Previous studies have also reported that AQG declines under 2% O2 (Fock, Hilgenberg & Egle 1972; Kaplan & Bjorkman 1980).

For many of our treatments, AQG and AQF declined as PFD increased from 100 to 300 µmol m−2 s−1 (Fig. 3). This might reflect a shift in the balance between respiratory O2 consumption and CO2 production as light levels increased from just above the compensation point (Hoefnagel, Atkin & Wiskich 1998). Another possibility is that plants might allocate a greater proportion of their carbon to nucleic or amino acids when operating near the light compensation point and might generate relatively more lignin or lipids under increasing light levels. Clearly, changes in AQ that are independent of nitrogen source deserve further examination.

In C3 plants, a doubling of CO2 concentration typically increases carbon fixation by 25% or more in short-term studies under 21% O2 (Stitt 1991; Curtis 1996). Net CO2 assimilation in our study increased by about 20% at the higher PFD levels when CO2 was doubled under 21% O2 (Fig. 1a). No stimulation of net CO2 assimilation occurred under 2% O2 at high PFD (Fig. 1b) possibly due to a limitation on the rate of photosynthesis by end product formation, primarily starch and sucrose synthesis. End product limitation appears to be common in tomato (Sage & Sharkey 1987; Micallef et al. 1995) and occurs under low O2, high CO2, and high PFD conditions that favour high rates of triose phosphate production by the chloroplast (Häusler, Schlieben & Flügge 2000).

There are several factors that might help to explain why NO3 assimilation was only apparent under ambient CO2 and O2 conditions:

  • 1The re-assimilation of NH3 produced by the photorespiratory nitrogen cycle is essential for maintaining nitrogen status (for a review, see Wingler et al. 2000). If the recycling of NH3 is inefficient and leaf emissions of NH3 are high as a consequence, then an acceleration of primary NO3 assimilation would be needed to balance or even overcompensate for this loss. Leaf NH3 emissions, however, are considered to be negligible relative to photorespiratory NH3 production (Mattsson et al. 1997; Schjoerring et al. 2000). Thus, it is unlikely that an acceleration of NO3 assimilation occurred to account for NH3 loss.
  • 2A lowering of photorespiratory capacity in barley mutants deficient in glycine decarboxylase leads to an enhanced reduction state and over-energization of chloroplasts (Igamberdiev et al. 2001). In contrast, photorespiration in wild-type plants serves as an important redox transfer mechanism that increases the cytosolic NADH/NAD ratio via the export of malate from the chloroplast as described by Backhausen, Kitzmann & Scheibe (1994). Because the first step of NO3 assimilation (i.e. the reduction of NO3 to NO2) occurs in the cytosol and uses NADH from the malate shuttle, this may explain why we observed NO3 assimilation to be greater when photorespiration was highest, that is, under ambient CO2 and 21% O2.
  • 3In addition to the malate shuttle, reductant for the NO3 to NO2 reaction can be provided by the conversion of triose phosphate to organic acids in the cytosol (Noctor & Foyer 1998). Triose phosphate is also an intermediate in the formation of sucrose and starch. Given the apparent end product limitation in our study on the rate of photosynthesis by starch and sugar formation under diminished photorespiratory conditions, one might expect that more triose phosphate would have been diverted from sucrose production to the production of organic acids and NADH for NO3 assimilation under 2% O2. However, this did not seem to occur based on the lack of apparent NO3 assimilation under 2% O2 in our study.
  • 4Nitrate photo-assimilation may be a means of photoprotection from high irradiance (Zhu et al. 2000). In our study, non-photochemical quenching (NPQ) at high irradiance was indeed lower when plants received NO3 rather than NH4+ under ambient CO2 and O2 conditions. However, NPQ was also lower at elevated CO2 and 21% O2.

Of these possibilities, increased cytosolic NADH due to photorespiration (hypothesis B) appears to best explain the higher rates of NO3 assimilation under ambient CO2 and O2 conditions. As indicated above, inefficiencies in photorespiratory NH3 recycling, conversion of triose phosphate to organic acids and NADH, and nitrate assimilation as a means of photoprotection seem unlikely to explain the results based on previous studies or the results presented in this study.

In addition to tomato, NO3 assimilation has previously been shown to decrease under elevated CO2 in wheat (Bloom et al. 2002). Gas exchange measurements, NO3 reductase activity, shoot protein, and shoot biomass of wheat plants grown under either ambient or elevated CO2 and either NH4+ or NO3 as a nitrogen source all indicated an inhibition of NO3 assimilation under elevated CO2. In another recent study on wheat, Novitskaya et al. (2002) found no consistent trends in leaf amino acid levels with short-term exposure to various CO2 concentrations, but suggested that NO3 assimilation might increase under negligible photorespiration (2% O2) based on trends in leaf malate levels. Surprisingly, net CO2 assimilation in this second study did not increase under elevated CO2 even though the plants were grown at ambient CO2 levels. Distinct approaches as well as differences in conditions, such as plant age and nutrient levels that can influence NO3 assimilation (Geiger et al. 1998; Geiger et al. 1999), may help to explain some of the discrepancies among these studies on wheat.

In conclusion, conditions that diminish photorespiration, either elevated CO2 or low O2 limited leaf NO3 photo-assimilation in short-term experiments on tomato. Consequently, complex interactions between photorespiratory metabolism and NO3 assimilation may be more important than previously recognized in plant leaves.


We thank Carrie Louie and Alan Tan for their technical assistance; Robert Pearcy for use of his chlorophyll fluorometer; and Asaph Cousins, Werner Kaiser, Shimon Rachmilevitch, and David Smart for their critical review of the manuscript. This work was supported by the National Science Foundation under Grant IBN-99–74927.