Does a Low Nitrogen Supply Necessarily Lead to Acclimation of Photosynthesis to Elevated CO₂?

Abstract

Long-term exposure of plants to elevated partial pressures of CO₂ (pCO₂) often depresses photosynthetic capacity. The mechanistic basis for this photosynthetic acclimation may involve accumulation of carbohydrate and may be promoted by nutrient limitation. However, our current knowledge is inadequate for making reliable predictions concerning the onset and extent of acclimation. Many studies have sought to investigate the effects of N supply but the methodologies used generally do not allow separation of the direct effects of limited N availability from those caused by a N dilution effect due to accelerated growth at elevated pCO₂. To dissociate these interactions, wheat (Triticum aestivum L.) was grown hydroponically and N was added in direct proportion to plant growth. Photosynthesis did not acclimate to elevated pCO₂ even when growth was restricted by a low-N relative addition rate. Ribulose-1, 5-bisphosphate carboxylase/oxygenase activity and quantity were maintained, there was no evidence for triose phosphate limitation of photosynthesis, and tissue N content remained within the range recorded for healthy wheat plants. In contrast, wheat grown in sand culture with N supplied at a fixed concentration suffered photosynthetic acclimation at elevated pCO₂ in a low-N treatment. This was accompanied by a significant reduction in the quantity of active ribulose-1, 5-bisphosphate carboxylase/oxygenase and leaf N content.

Growth at elevated pCO₂ frequently brings about change in plant physiology that is commonly interpreted as acclimation (Drake et al., 1997). Photosynthesis is inextricably involved because CO₂ is the substrate in C₃ species that is limiting at the current atmospheric pCO₂. However, results from investigations on the effects of elevated pCO₂ on photosynthesis have been inconsistent. The stimulatory response brought about when pCO₂ is suddenly increased (Long, 1991) has often been found to decline with increasing duration of exposure (for review, see Gunderson and Wullschleger, 1994; Sage, 1994; Drake et al., 1997), but some experiments have failed to find any long-term effect, either in controlled environments (Radoglou and Jarvis, 1990; Wong, 1990) or in the field (Arp and Drake, 1991; Jones et al., 1995; Pinter et al., 1996). Why, then, is the acclimatory response so varied? Species differences can no doubt account for some of the variability, but often the same species in apparently similar conditions can yield different results with different investigators (Sage, 1994). This fact in itself suggests that there may be some uncontrolled factor(s) in the experimental design that may be crucial to the acclimatory response of photosynthesis.

Evidence that additional factors may be interacting with the CO₂ response was brought to prominence by Arp (1991), who, after reviewing the data from several investigations using a variety of experimental designs, suggested that root restriction by pot size had a significant effect on the acclimatory response. Limited rooting volume was suggested to create an imbalance in the supply and demand for carbohydrates and, consequently, would lead to carbohydrate feedback inhibition of photosynthesis (Stitt, 1991). However, further investigation has suggested that pot size and root restriction may only be involved partially in determining the degree of acclimation that occurs at elevated pCO₂; the supply of nutrients is also crucial (Pettersson et al., 1993). In particular, evidence has accumulated that N supply is of primary importance. This thesis is particularly attractive because by far the largest proportion of soluble N in the leaf is incorporated in Rubisco (Woodrow and Berry, 1988). At elevated pCO₂ carboxylation efficiency increases, enabling the photosynthetic rate to be maintained with less active Rubisco per unit leaf area. Release of N from excess Rubisco would then be advantageous if growth was limited by N supply. A number of experimental results now suggest that acclimation can be significantly slowed by high-N application (Webber et al., 1994; Drake et al., 1997).

Investigating the role that N supply has on photosynthesis at elevated pCO₂ is not easy. When plants are grown in pots and irrigated with a solution containing a fixed concentration of nutrients, the available N-to-plant mass ratio will decline with experimental duration. This occurs because there is a finite limit to the quantity of nutrient solution that can be applied to the pot and also because of the likely spatial constraint that the roots will progressively encounter within the container. If elevated pCO₂ increases growth, then the available N-to-plant mass ratio will decline more rapidly, with the danger of confounding the CO₂ treatment with earlier N deficiency (Pettersson and McDonald, 1994). The RAR method of Ingestad and Lund (1986) eliminates this problem by supplying N in direct proportion to the plant growth rate using a hydroponics-culture technique.

To test the hypothesis that acclimation to elevated pCO₂ is primarily a response to N availability, wheat (Triticum aestivum L.) was grown hydroponically at the current atmospheric [CO₂] or at 650 μmol mol⁻¹ and with either free access to N or at a relatively low-N RAR. The effects on photosynthesis were subsequently analyzed and the results were compared with those of a previous experiment in which wheat had been grown in sand culture at elevated pCO₂ and the two N treatments were applied in the traditional way by irrigating with solutions containing a fixed high and low N concentration.

MATERIALS AND METHODS

Plant Material

Winter wheat (Triticum aestivum L. cv Hereward, Plant Breeding International, Trumpington, UK) was germinated in 360 or 650 μmol mol⁻¹ CO₂ on moist filter paper. After 5 d, seedlings of equal size were transferred to hydroponics troughs.

Hydroponics System

Troughs with sectional covers were used for hydroponically growing the wheat. Each nutrient treatment comprised three troughs, with each trough holding 10 plants. The nutrient solution was circulated by a centrifugal pump (model 1060 11 993, Eheim, Deizisau, Germany) to a header tank, which fed the troughs by gravity, after which the nutrient solution was collected in a reservoir tank that resupplied the pump. A diaphragmmatic air pump was used to ensure that the solutions were continuously aerated. Roots were suspended in the flowing nutrient solution by inserting individual plants into holes in the trough covers and holding them in place with foam sleeves. The plant-culture solution was not changed throughout the course of the experiment and, consequently, care was taken to choose inert materials that were in contact with the solution (acrylonitrile butadiene styrene, polyethylene, and polypropylene). The system was kept scrupulously clean and light free to minimize growth of microbes and algae.

Nutrient Solutions

The nutrient solutions contained both macronutrients and micronutrients in the ratio that occurs in healthy wheat plants (Ingestad and Stoy, 1982). As the plants removed the nutrients, they were replaced at rates that provided either free access to all of the nutrients or at a strictly controlled RAR of N. Thus, the supply of nutrients was continually increased to match the rising demand of the growing plants. A detailed description of the principles and techniques for growing plants this way, with a controlled nutrient supply matching the rate of plant growth, has been described extensively by Ingestad and coworkers (Ingestad and Stoy, 1982; Ingestad and Lund, 1986). The culture solution adopted was based on the stock solutions described by Ingestad (1971), adjusted for cereals (Ingestad and Stoy, 1982) using both nitrate and ammonia as the N source. The “high-N” treatment provided the plants with free access to all nutrients and an optimal [N] of 14.3 mm (Ingestad and Stoy, 1982). Nutrients were replenished in proportion to plant uptake by daily titration with stock solutions in accordance with conductivity (CL91 Wissenschaftlich-Technische Werkstatten, Weilheim, Germany) and pH (digital pH meter, model CD 620, WPA Ltd., Linton, UK) measurements. The same techniques were used for the “low-N” treatment, except that nitrates were replaced by chlorides, together with additional adjustments to ensure that the other nutrients remained in correct proportion in the stock solutions. N (nitrate and ammonia) was added daily to the low-N treatment at a RAR of 0.07 mol N mol⁻¹ N d⁻¹. Frequent weighing of the plants allowed minor corrections to be made in the calculation of the N required by the plants. Solution conductivities were kept within 15% of the desired level for the “high-N-free access” treatment and within 5% of the “low-N-controlled RAR” treatment.

Sand Culture Experiment

Winter wheat seed was soaked for 24 h before sowing in washed, lime-free horticultural grit/sand (William Sinclair Horticultural Ltd., Lincoln, UK) using 0.6-L pots containing drainage holes. Plants were watered as required to drain through so as to avoid the rooting medium from drying, using a modified Shive's solution (Evans and Nason, 1953). The high- and low-N treatments received 10 and 4.5 mm nitrate, respectively. N was added as calcium nitrate and balanced by adding additional calcium sulfate in the low-N treatment.

Growth Conditions

Plants were grown under artificially lit, controlled-environment conditions (HPS 1500, Heraeus Vötsch GmbH, Balingen, Germany). Day and night temperatures were 20°C/15°C, the water vapor pressure deficit was <0.7 kPa, and the photoperiod was 14 h, with a PPFD at leaf height of approximately 750 μmol m⁻² s⁻¹. The [CO₂] was controlled at 360 or 650 ± 15 μmol mol⁻¹ CO₂ using a combined IR gas analyzer and microprocessor unit (model WMA-2, PP Systems, Hitchin, UK). CO₂ was supplied from a compressed gas cylinder (Linde Gas UK Ltd., Stoke on Trent, UK) certified as 999.95 mmol mol⁻¹ CO₂, <0.5 μmol mol⁻¹ C₂H₄. Before entering the controlled environment chamber the gas was passed through a potassium permanganate column as a further precaution against contamination by hydrocarbons.

Wet Weight, Dry Weight, and Leaf-Area Measurement

To ensure that plant growth in the hydroponics experiment was increasing in accordance with the rate of N addition in the low-N treatment and that growth of the control plants was uninhibited, the fresh weight of 10 plants was measured every 2 to 3 d. During this procedure the roots were kept submerged, only removed from solution for the short time it took to dab off surplus water with absorbent paper and weighed on a top pan balance (model HC22, Oertling, Smethwick, UK).

At the end of the growing period (ligule emergence of the sixth leaf) plants were divided into their component parts (roots, pseudo stems, leaf laminae, and tillers). Leaf area was measured with a leaf-area meter (Delta T Devices, Burwell, UK). The plant tissue was dried to constant mass in a fan-assisted oven at 80°C before weighing on an analytical balance (model 2006 MP, Sartorius, Göttingen, Germany), which was self-calibrating and cross-checked annually by the manufacturer.

Gas-Exchange Measurements

Leaf gas-exchange measurements were made using a portable IR gas analysis system (CIRAS-1, PP Systems) and a narrow leaf cuvette with a quartz-iodide light source (PLC, PP Systems). The CO₂ and water analyzers were routinely calibrated against a CO₂ standard (Linde Gas UK Ltd.) and water vapor generator (model WG600, ADC Ltd., Hoddesdon, UK). The PPFD at the level of the leaf was 1400 μmol m⁻² s⁻¹, whereas the rest of the plant remained at the controlled-environment growth conditions. Responses of photosynthetic CO₂ uptake to changes in pCO₂ over the range of 50 to 150 μmol mol⁻¹ and 1200 to 1600 μmol mol⁻¹, at pO₂ of 210 mmol mol⁻¹, were made to calculate the V_c,max and the A_max. Calculation of V_c,max followed the procedure of McKee et al. (1995). The effect of inhibiting photorespiration was investigated by reducing the pO₂ from 210 to 21 mmol mol⁻¹.

Tissue Analysis of Rubisco and N

Samples for the Rubisco assays were taken from the central portion of the sixth leaf at ligule emergence, i.e. identical sections to those used for gas-exchange measurements. The sections were collected halfway through the photoperiod and immediately immersed in liquid N₂. Extraction and assay of Rubisco activity, activation, and content were as described in McKee et al. (1995).

Total leaf N content was measured by GC using an elemental analyzer (model PE 2400 series II CHNS/O Analyser, Perkin-Elmer Cetus). Samples were first ground to a fine powder and the instrument was calibrated with acetanilide standards (Perkin-Elmer Cetus).

Statistical Analysis

The data were analyzed using two-way analysis of variance (Systat Inc., Evanston, IL) with pCO₂ and N as independent factors. Post-hoc pairwise comparisons were made using Scheffé's probability. Growth rates of the hydroponically grown plants were transformed and analyzed by a regressions comparison (Sokal and Rholf, 1995).

RESULTS

The two N treatments used in the hydroponics-culture technique produced wheat plants with significantly different rates of growth (P > 0.01), but pCO₂ had no significant effect on growth rate (P > 0.05; Fig. 1a). The elevatedpCO₂-grown plants did, however, exhibit an initial slight growth advantage and so were always bigger than their ambient-pCO₂-grown counterparts. Growth was exponential at 0.18 and 0.20 d⁻¹ for the control and elevated-pCO₂-grown plants with free access to N and was 0.09 and 0.10 d⁻¹ for the control and elevated-pCO₂ plants grown at the restricted rate of N supply (Fig. 1b). There was no significant effect of CO₂ treatment (F = 1.609, P <0.2) or N supply (F = 0.021, P <0.8) on A_sat, so there was no evidence that photosynthesis was down-regulated by elevated pCO₂ (Fig. 2). An identical pattern of results was obtained when the measurements were made at 650 μmol mol⁻¹ CO₂ (data not shown), which is commensurate with this apparent lack of acclimation. The A_sat of the elevated pCO₂-grown plants measured at their growth pCO₂ at the sixth-leaf stage was significantly higher (56%) than the rate obtained for plants grown and measured at 360 μmol mol⁻¹ (P <0.01).

Figure 1.

The effects of pCO₂ and N supply on the increase in total wet weight (a) and ln wet weight (b) of wheat. Plants were grown hydroponically with day/night temperatures of 20°C/15°C and a photosynthetically active photon flux density at leaf height of approximately 750 μmol m⁻² s⁻¹. Treatments were: ▪, 650 μmol mol⁻¹ CO₂, free access to N; □, 360 μmol mol⁻¹ CO₂, free access to N; •, 650 μmol mol⁻¹ CO₂, RAR of 0.07 mol N mol⁻¹ N d⁻¹; and ○, 360 μmol mol⁻¹ CO₂, RAR of 0.07 mol N mol⁻¹ N d⁻¹. Vertical bars represent se; n = 10.

Figure 2.

Rates of A_sat and V_c,max of the sixth leaf at ligule emergence of wheat grown hydroponically (hy), and the V_c,max of the sixth leaf at ligule emergence of wheat grown in sand culture (sd). Measurements were made with a leaf temperature of 23°C with a photosynthetically active photon flux density of 1400 μmol m⁻² s⁻¹ and pO₂ of 210 mmol mol⁻¹. The A_sat was measured at 360 μmol mol⁻¹ CO₂, and V_c,max was obtained from the initial slope of the CO₂ response curve. Treatments consisted of the following: Hydroponics: ▪, 650 μmol mol⁻¹ CO₂, free access to N; □, 360 μmol mol⁻¹ CO₂, free access to N; ▩, 650 μmol mol⁻¹ CO₂, N RAR of 0.07 mol N mol⁻¹ N d⁻¹; ▨, 360 μmol mol⁻¹ CO₂, N RAR of 0.07 mol N mol⁻¹ N d⁻¹. Sand culture: ▪, 650 μmol mol⁻¹ CO₂, 10 mmol nitrate; □, 360 μmol mol⁻¹ CO₂, 10 mmol nitrate; ▩, 650 μmol mol⁻¹ CO₂, 4.5 mmol nitrate; ▨, 360 μmol mol⁻¹ CO₂, 4.5 mmol nitrate. Vertical bars represent se; n = 4.

Analysis of tissue N showed that the low-N hydroponically grown plants exhibited a significant reduction (P <0.001) in the N content of each plant organ (roots, stems, and leaves) compared with controls, although the leaves suffered the smallest decrease (Table I). This change in N content was accompanied by a significant increase in the C-to-N ratio of each organ (Table I). However, when N content was expressed on a leaf-area basis, the effect of N treatment was greatly reduced at the sixth-leaf stage and was only significantly different between N treatments of the elevated pCO₂ leaves (Table I; F = 14.037; P > 0.01). This result is apparently due to the LAR, which showed the greatest change in response to CO₂ treatment (Table II), brought about by a decrease in SLA (P <0.001) rather than a decrease in LWR (P = 0.1).

Table I.

Tissue N and C-to-N ratio of hydroponically grown wheat

Plant Nitrogen	Treatment				ANOVAa
	LCHN	HCHN	LCLN	HCLN	CO2	N
Tissue N concentration
All leaves (mg g−1)	66.1  ± 0.7	60.3  ± 3.9	55.6  ± 0.7	52.2  ± 0.7		***
Leaf 6 (g m−2)	2.7  ± 0.1	3.0  ± 0.1	2.3  ± 0.21	2.0  ± 0.4		**
Shoot (mg g−1)	59.1  ± 1.1	54.4  ± 2.6	39.3  ± 1.6	39.7  ± 4.1		***
Root (mg g−1)	50.1  ± 1.4	48.2  ± 0.6	35.7  ± 1.2	36.6  ± 1.3		***
C-to-N ratio
Leaf	6.4  ± 0.1	6.9  ± 0.4	8.1  ± 0.1	8.5  ± 0.1	*	***
Shoot	6.9  ± 0.1	7.2  ± 0.2	10.6  ± 0.3	10.6  ± 1.0		***
Root	8.0  ± 0.0	7.8  ± 0.2	11.7  ± 0.4	11.6  ± 0.3		***

Tissue N and C-to-N ratio of hydroponically grown wheat in 360 μmol mol⁻¹ CO₂ (LC) or 650 μmol mol⁻¹ CO₂ (HC) with either free access to N (HN) or at a low rate of N supply (LN). Plants were harvested when the sixth leaf had fully expanded. Values shown are the means ± se per unit dry weight; n = 4 to 5.

^aANOVA, Analysis of variance: *, P < 0.05; **, P < 0.01; ***, P < 0.001. There were no significant interactions between CO₂ and N. 

Table II.

Leaf growth analysis of hydroponically grown wheat

Leaf Growth Parameters	Treatment				ANOVAa
	LCHN	HCHN	LCLN	HCLN	CO2	N
LAR (m2 g−1)	0.017  ± 0.0005	0.013  ± 0.0003	0.016  ± 0.0008	0.011  ± 0.0002	***	**
SLA (m2 g−1)	0.034  ± 0.0008	0.026  ± 0.0006	0.030  ± 0.0006	0.021  ± 0.0004	***	***
LWR (g g−1)	0.51  ± 0.007	0.48  ± 0.005	0.52  ± 0.025	0.51  ± 0.004

LAR, SLA, and LWR of hydroponically grown wheat in 360 μmol mol⁻¹ CO₂ (LC) or 650 μmol mol⁻¹ CO₂ (HC) with either free access to N (HN) or at a low rate of N supply (LN). Plants were harvested when the sixth leaf had fully expanded. Values are the means ± se.

^aANOVA, Analysis of variance: **, P < 0.01; ***, P < 0.001. There were no significant interactions between CO₂ and N. 

A comparison with results from the sand-culture experiment shows that leaf N content on a leaf-weight basis was not substantially lower than that of the hydroponically grown plants, but only when plants were irrigated with the high-N treatment (Table III). Those plants given the low-N solution suffered a large, significant reduction in leaf N content even when expressed on a leaf-area basis (Table III; F = 130.503; P > 0.001). Elevated pCO₂ exacerbated this reduction in [N] (F = 13.650; P > 0.001). This result occurred in spite of the low-N sand-culture plants receiving a higher total N dose over the course of the experiment than their hydroponically grown counterparts. Thus, growing wheat with fixed [N] dramatically reduced leaf N content; an effect that was augmented by elevated pCO₂, whereas growth in hydroponic culture had very little effect on leaf N.

Table III.

Leaf tissue N content of sand-grown wheat

Tissue N Concentration	Treatment				ANOVAa
	LCHN	HCHN	LCLN	HCLN	CO2	N
Leaf (mg g−1)	57.1  ± 2.5	54.4  ± 1.5	28.8  ± 1.2	17.2  ± 0.8		***
Leaf (g m−2)	1.7  ± 0.1	1.9  ± 0.1	1.1  ± 0.1	0.7  ± 0.0	***	***

Leaf tissue N content of sand-culture-grown wheat at 360 μmol mol⁻¹ CO₂ (LC) or 650 μmol mol⁻¹ CO₂ (HC) with either 10 mmol of nitrate (HN) or 4.5 mmol of nitrate (LN). Analyses are for the third leaf at ligule emergence. Values shown are the mean ± se per unit dry weight; n = 10.

^aANOVA, Analysis of variance: ***, P < 0.001. There were no significant interactions between CO₂ and N. 

The failure of elevated CO₂ to bring about a reduction of N content in hydroponically grown wheat may be a crucial factor in the ability of these plants to avoid photosynthetic acclimation as determined by A_sat. Central to this effect is likely to be the response of Rubisco. Figure 3 shows A/c_i curves for the sixth leaves once they had reached full expansion. Plants grown and measured at the current ambient pCO₂ at high and low rates of N supply had values of c_i on the initial phase of the curve, inferring Rubisco limitation. Growth at elevated pCO₂ shifted the operating point to the inflection of the curve (Fig. 3), suggesting increased control by ribulose-1,5-bisphosphate regeneration.

Figure 3.

CO₂-response curves for the sixth leaf at ligule emergence of wheat grown hydroponically with free access to N or with a RAR of 0.07 mol N mol⁻¹ N d⁻¹ in atmospheres of 360 or 650 μmol mol⁻¹ CO₂. Symbols are data points from two representative leaves used in the calculation of V_c,max and A_max, whereas the curves are fitted by a maximum likelihood regression using the equations of Farquhar et al. (1980) to all of the leaves measured. The supply functions are indicated by dotted lines. a, 360 μmol mol⁻¹ CO₂, free access to N; b, 650 μmol mol⁻¹ CO₂, free access to N; c, 360 μmol mol⁻¹ CO₂, RAR of 0.07 mol N mol⁻¹ N d⁻¹; and d, 650 μmol mol⁻¹ CO₂; RAR of 0.07 mol N mol⁻¹ N d⁻¹. Measurement conditions are described in Figure 2.

In vivo measurements of Rubisco activity modeled from the A/c_i curves show that for the sixth leaf at ligule emergence there was no significant difference in V_c,max between the CO₂ or N treatments (Fig. 2; F = 0.170, 0.170; P > 0.65). This was confirmed by in vitro measurements, which also found no significant difference in the initial activity (Fig. 4; F = 0.032, 2.218; P > 0.1) or the activated activity of Rubisco (Fig. 4; F = 0.510, 1.449; P > 0.25). There was also no change in the quantity of this enzyme (Fig. 4; F = 1.205, 2.651; P > 0.1).

Figure 4.

V_c,max of Rubisco measured in vitro together with the concentration of Rubisco protein. Activities are for the initial activity upon extraction and for the V_c,max following incubation with CO₂ and Mg²⁺. Legend for bars is in Figure 2; vertical bars represent se; n = 4.

When in vivo Rubisco activity was estimated for wheat grown in the sand-culture experiment, it was found that for the fourth and sixth leaves at ligule emergence there was a significant effect of both pCO₂ and N on V_c,max (Fig. 2; F = 13.326, 13.469; P <0.01). The V_c,max of the elevated pCO₂, low-N-grown plants was significantly less than for those grown at control pCO₂ and low N (Fig. 2; P <0.01), indicating a significant decrease in the amount of active Rubisco.

The operating point on the A/c_i curve of the hydroponically grown plants at elevated pCO₂ was at the inflection of the curve and so would be partially influenced by the rate of regeneration of ribulose-1,5-bisphosphate and ultimately on the rate of electron transport. The rates of A_max for the sixth leaves show that although a depression was indicated for leaves grown at elevated pCO₂ with low N, there was no significant effect of pCO₂ or N treatment (Fig. 5; F = 0.379, 4.592; P > 0.05). To verify that triose-3 phosphate export from the chloroplasts was not limiting to leaf gas exchange the pO₂ response of photosynthesis was investigated. Plants responded positively to a reduction in the pO₂ from 21 to 2.1 kPa and showed no pCO₂ or N treatment effect (Fig. 5; F = 3.333, 0.146; P > 0.05). The ability of all plants to respond similarly to the removal of photorespiration consequently supports the A_max results.

Figure 5.

The rate of CO₂ uptake at light and CO₂ saturation (A_max) and the relative stimulation (stim.) of CO₂ uptake by inhibition of photorespiration following a reduction in pO₂ to 21 mmol mol⁻¹. Measurement conditions for leaf gas exchange and legends for bars are described in Figure 2; the [CO₂] was 650 μmol mol⁻¹. Vertical bars represent se; n = 4.

DISCUSSION

Acclimation of photosynthesis to elevated pCO₂ was accentuated by low-N supply when wheat was grown in pots with a fixed [N]. However, when N was supplied in direct proportion to plant growth, elevated pCO₂ did not produce acclimation of photosynthesis regardless of whether the N supply was strongly limiting growth or optimal. Pettersson et al. (1993), who have previously used the RAR for applying N, have obtained similar findings with birch. The data support the hypothesis that acclimation of photosynthesis to elevated pCO₂ results from a greater dilution of plant N content rather than from a low availability of N. Our results suggest that if a plant commences development with a low availability of N, paralleling a plant germinating on a N-deficient soil, the major effect will be on the rate of leaf-area development rather than on leaf N content (Scott et al., 1994). As the root system expands, further N may become available, a situation that may be simulated by the RAR method. When a plant germinates within a pot with N supplied at a fixed concentration, as simulated by our sand-culture experiment, initially there is a high availability of N relative to plant mass, allowing rapid growth. However, with further growth the relative amount of N will decline and trigger acclimation of photosynthetic capacity. A corresponding situation may occur in the field, when over-winter mineralization or fertilizer application at sowing creates a flush of available N followed by a depletion of the N reserve through the season.

The dilution effect on plant N status produced by growth in elevated pCO₂ can result either from increasing dilution of a given N supply (Coleman et al., 1993) or by increased carbohydrate accumulation diluting the [N] within the plant (Wong, 1990; Kuehny et al., 1991). Complications can arise if changes in SLA are not taken into account. This is because elevated pCO₂ frequently alters leaf morphology and, consequently, effects of pCO₂ on [N] are decreased (Norby et al., 1992; Rogers et al., 1996a) or absent (Rowland-Bamford et al., 1991) when results are expressed per leaf area rather than leaf weight. Our hydroponically grown wheat did show an increase in the C-to-N ratio, but the effect was relatively small and leaf N content remained above 1.8 g m⁻². It is well established that rates of photosynthesis are correlated to N content, but for wheat, this relationship begins to plateau at leaf concentrations above 1.75 g m⁻² (Evans, 1989). In another investigation, when growth techniques were compared the C-to-N ratio of tobacco plants was markedly increased at elevated pCO₂ when they were grown in pots but was largely unchanged following growth in hydroponic culture (Ferrario-Méry et al., 1997).

Availability of N is also a crucial factor in sink development. Rogers et al. (1996b) have shown that the degree of N fertilization is an important contributor to sink strength, demonstrating that it can prevent acclimation at all but the lowest rates of N application. In relation to wheat, Rogers et al. (1996a) have demonstrated the requirement of N for tiller and leaf production for avoidance of acclimation. Similar conclusions were obtained by Ryle et al. (1992) and by Newbery and Wolfenden (1996). However, in our hydroponics experiment both leaf area and the number of tillers were significantly reduced by the low-N RAR at both control and elevated pCO₂, but acclimation was still avoided. This demonstrates that production of a large sink capacity may not be necessary to avoid acclimation, rather, the balance between source and sink at the whole-plant level is the key factor, as has been proposed by Pettersson and McDonald (1994). Thus, at elevated pCO₂, although our hydroponically grown, low-N wheat plants maintained their photosynthetic rates, the absolute amount of photosynthate produced was lessened because of their smaller leaf area (data not shown) compared with that of their high-N-grown counterparts. In addition, the ability of the hydroponically grown wheat to respond to a lowering of the pO₂ demonstrates that the requirement for Pi by ATP phosphorylase was not greater than the rate of sugar phosphate use (Sharkey, 1985). An interesting observation, however, was that the hydroponically grown wheat had a greater mass than those plants grown in ambient air, despite there being no significant effect of pCO₂ on relative growth rate. This difference must have been initiated from a very early and transient stimulation of growth, and consequently appears similar to observations made in other elevated-CO₂ studies where differences in biomass between pCO₂ treatments are reported to have arisen from brief alterations in relative growth rate (Poorter, 1993).

Acclimation of photosynthesis to growth in elevated pCO₂ is most frequently accompanied by a reduction in carboxylation capacity (for review, see Bowes, 1991). It is commonly suggested that a decrease in Rubisco activity is ultimately responsible for acclimation of photosynthesis to elevated pCO₂ (Bowes, 1991). Estimations of V_c,max in vivo and direct measurement of Rubisco activity and content in vitro confirmed that the pCO₂ and N treatments had no significant effect on the hydroponically grown wheat, and this is the most likely reason that net photosynthesis was unaffected. The lack of effect on Rubisco most likely stems from the absence of either any major reduction in leaf N content or of any end-product inhibition of photosynthesis. Increase of leaf carbohydrate has been correlated with a decrease in V_c,max (McKee and Woodward, 1994), and it is this increase in photosynthate that is suggested to be the signal that brings about a decrease in Rubisco levels (Stitt, 1991). The underlying mechanism is believed to operate via the repression of photosynthetic gene expression (Webber et al., 1994; Koch, 1996; Drake et al., 1997). Whether the quantity of Rubisco is decreased at elevated pCO₂ in response to buildup in leaf carbohydrate or not, less active enzyme is required because it is not saturated at the current atmospheric pCO₂ (Long, 1991). Consequently, N may be reallocated to proteins of the electron-transport chain or photosynthetic carbon-reduction-cycle enzymes (Woodrow, 1994). Wheat grown under true field conditions (Free Air CO₂ Enrichment) showed no loss of Rubisco activity or quantity in unshaded leaves, and this was accompanied by a complete absence of any down-regulation of photosynthesis (Nie et al., 1995; Drake et al., 1997).

In conclusion, the results from our investigation have shown that low rates of N supply need not cause acclimation of photosynthesis to elevated pCO₂. Development of large sinks, e.g. tillers and leaves, was not necessary for the avoidance of acclimation; rather, an adjustment of plant growth rate to match the N supply appears to be the decisive factor. This agrees with the hypothesis by Pettersson and McDonald (1994) that acclimation to elevated pCO₂ is dependent upon whether the whole-plant growth response has acclimated to elevated pCO₂, together with any other resource limitations in the immediate environment. The majority of experimental methodologies used in elevated pCO₂ investigations by necessity grow plants in artificial conditions, but the outcome may be that resources do not keep pace with demand and, therefore, the plants will be prone to acclimation. The experimental approach that has least often found acclimation is the field system (Gunderson and Wullschleger, 1994; Sage, 1994; Drake et al., 1997). Plants growing in the natural environment are more likely to be adjusted to surrounding environmental influences. However, this does not preclude acclimation from occurring; natural environments are often limiting in nutrients, especially N (Eamus and Jarvis, 1989; Gifford, 1992), and, consequently, there is ample opportunity for acclimation of photosynthesis to occur. Plants can experience periods of source:sink imbalance, especially late in the growing season when reproductive sinks are developing. For example, translocation of N during grain filling was assumed to produce a decrease in Rubisco concentration of wheat exposed to elevated CO₂ in a free air CO₂ enrichment system (Nie et al., 1995), and changes in developmental state have been shown to initiate reversible acclimation in beet (Ziska et al., 1995). We might therefore expect phenology, together with specific environmental conditions, to be instrumental in determining the response of plants to elevated pCO₂ in the natural environment.

Abbreviations:

A_max

rate of CO₂ uptake at light and CO₂ saturation

A_sat

rate of CO₂ uptake at light saturation

c_i

intercellular pCO₂

LAR

leaf area ratio

LWR

leaf weight ratio

pCO₂

partial pressure of CO₂

RAR

relative addition rate

SLA

specific leaf areaV_c,

_max

maximum velocity for carboxylation