INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Grasses growing in terrestrial ecosystems receive the major part of their N as NH₄⁺ derived from mineralization of soil organic matter (Whitehead, 1995). However, some nitrification occurs even in grassland soils and may lead to significant supply of nitrate to the roots (Hatch et al., 2000). Once in the root, nitrate can be stored, assimilated, or transported to the shoot. Ammonium absorbed by the roots has previously been assumed to be almost completely assimilated in the roots, but recent studies have shown that significant amounts of NH₄⁺ can be transported in the xylem (Mattsson and Schjoerring, 1996; Finnemann and Schjoerring, 1999) and that plants can contain substantial concentrations of NH₄⁺ in their tissues (Wang et al., 1993; Kronzucker et al., 1995). These NH₄⁺ pools in leaf tissue and apoplast are important for regulating N utilization, including the exchange of gaseous NH₃ between plant leaves and the atmosphere.

Because plants can act as both a source of and a sink for atmospheric NH₃, it is important to know the physiological mechanisms that are involved in determining their NH₃ compensation point, i.e. the NH₃ concentration in the air within the sub-stomatal cavities at which no net exchange with the atmosphere takes place (Farquhar et al., 1980). The NH₃ compensation point varies with the level of N nutrition (Mattsson et al., 1998), the developmental stage of the plant (Husted et al., 1996), and with the activity of Gln synthetase (Mattsson et al., 1997). Compensation points for NH₃ are usually determined by gas exchange measurements in the field or in laboratory cuvettes. Alternatively, leaf apoplastic pH and NH₄⁺ concentrations have been used as bio-indicators for the NH₃ compensation point (Husted and Schjoerring, 1995; Mattsson et al., 1997; Hill et al., 2001). There exists, however, considerable discrepancies between results obtained by gas exchange measurements and apoplastic extracts (Hanstein et al., 1999; Sutton et al., 2001).

This investigation was conducted to study dynamic and steady-state effects of N supply to the roots (NO₃ or NH₄⁺) on plant NH₄⁺ pools and NH₃ exchange with the atmosphere. Two different grass species were compared: Lolium perenne, which is characteristic of N-fertilized grasslands, and Bromus erectus, characteristic of N-poor grasslands. A key question asked was: What are the dynamic responses of apoplastic pH and NH₄⁺ concentrations after changing the source of N supplied to the roots?

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Dynamic Response to Switches in N Source

When 3 mM NH₄⁺ was introduced to plants that had been growing in nutrient solution with 3 mM NO₃, the NH₄⁺ concentration in the apoplastic solution rapidly increased and became 3 to 6 times greater than the initial value within 3 h (Fig. 1A). After 3 to 9 h, the apoplastic NH₄⁺ concentration reached a maximum and after 24 h of NH₄⁺ supply it started to decrease again. Tissue NH₄⁺ concentrations did not increase as dramatically, but still showed a 2- to 3-fold increase after 3 h of exposure to NH₄⁺ and thereafter stayed at about the same level throughout the remaining 45 h of the experimental period (Fig. 1B). Concomitant with the change in NH₄⁺ concentration, apoplastic pH showed a transient decrease of about 0.4 pH units (Fig. 2). Switching back from NH₄⁺ to NO₃ nutrition after 24 h of NH₄⁺ treatment, i.e. when the NH₄⁺ concentration was still at the maximum level, resulted in a decrease in both apoplastic and bulk tissue NH₄⁺ concentrations (Fig. 1, C and D). After 24 h, the apoplastic NH₄⁺ concentration was down at the original level again, whereas the tissue NH₄⁺ concentration decreased more slowly and did not reach the initial level in 48 h. The two species showed similar responses to the changes in N source, although tissue NH₄⁺ concentrations and apoplastic pH values were generally higher for B. erectus than for L. perenne.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Apoplastic NH4+ concentrations (A and C) and leaf tissue NH4+ concentrations (B and D) of 11-week-old L. perenne and B. erectus over a 48-h period after a change in N source from 3 mM NO3 to 3 mM NH4+ (A and B) or from 3 mM NH4+ to 3 mM NO3 (C and D). Values are means ± SE of four replicate samples.

View larger version (15K): [in this window] [in a new window]   Figure 2.   Apoplastic pH of 11-week-old L. perenne and B. erectus during 48 h after a change in N source from 3 mM NO3 to 3 mM NH4+. Values are means ± SE of four replicate samples.

Growth at Different Steady-State N Supplies

Both species remained in the vegetative stage of growth throughout the experimental period. After 6 weeks of growth at three different steady-state N treatments (3 mM KNO₃, 3 mM NH₄HCO₃, or 6 mM NH₄HCO₃), L. perenne generally had higher root and shoot fresh weights than B. erectus (Fig. 3). L. perenne plants growing at 3 mM NO₃ were the largest, followed by those growing on the same concentration of NH₄⁺. A doubling of the NH₄⁺ concentration retarded both shoot and root growth although no stress symptoms were visible. Fresh weights of B. erectus shoots and roots were similar across all the 3 N treatments. Leaf areas were 3 to 4 times higher for L. perenne (2,400-2,800 cm²) than for B. erectus (680-780 cm²).

View larger version (25K): [in this window] [in a new window]   Figure 3.   Shoot and root fresh weights of 11-week-old L. perenne and B. erectus plants grown for 6 weeks with different N treatments (3 mM KNO3, 3 mM NH4HCO3, or 6 mM NH4HCO3) in the nutrient solution. Values are means ± SE of three replicate samples.

Shoot dry matter N concentration of L. perenne was lowest (2.2%) in NO₃-grown plants and highest (4.9%) in the 6 mM NH₄⁺ treatment (Table I). B. erectus had higher shoot N concentration (4.4%-6%) and much lower shoot C to N ratio than L. perenne at all the N treatments. The total amount of N contained in the shoots at the end of the experimental period was twice as high in L. perenne as in B. erectus, whereas the N content did not differ between 3 mM NO₃- and NH₄⁺-fed plants in either species, amounting to 0.34 g N plant¹ for L. perenne and 0.13 g N plant¹ for B. erectus. Shoots of plants growing at double-strength NH₄⁺ contained 0.39 and 0.21 g N plant¹ for L. perenne and B. erectus, respectively.

Tissue Concentrations of Inorganic N at Different Steady-State N Supplies

Both roots and leaves of NO₃-grown plants had low bulk tissue NH₄⁺ concentrations (Fig. 4, B and C). When NH₄⁺ was the N source, roots had 4- to 7-fold higher NH₄⁺ concentration than leaves. Nitrate-grown B. erectus leaves contained considerable amounts of NO₃ (45 ± 4 and 68 ± 12 µmol g¹ fresh weight) in root and leaf tissue, respectively, whereas L. perenne had very low tissue NO₃ concentrations (5 ± 2 and 1 ± 0.4 µmol g¹ fresh weight). B. erectus showed higher tissue NH₄⁺ concentrations compared with L. perenne in all cases except in roots of plants grown with double-strength NH₄⁺ (Fig. 4, B and C). Leaf concentrations of free Asn and Gln increased 5- to 10-fold upon provision of NH₄⁺ instead of NO₃ (data not shown). Also Ser, Arg, and the nonvolatile amine -amino buturic acid were highest (P < 0.05) in NH₄⁺-grown plants, the latter only in B. erectus though (data not shown).

View larger version (25K): [in this window] [in a new window]   Figure 4.   Average daytime shoot NH3 emission (A), leaf (B), and root (C) tissue NH4+ concentrations of 11-week-old L. perenne and B. erectus grown for 6 weeks at different N treatments (3 mM KNO3, 3 mM NH4HCO3, or 6 mM NH4HCO3). Values are means ± SE of three replicate samples.

Leaf Apoplastic NO₃ and NH₄⁺ Concentrations at Different Steady-State N Supplies

The variation in apoplastic NH₄⁺ concentration between species and N treatments closely resembled that of bulk tissue NH₄⁺ (Table II). NO₃-grown plants had lower apoplastic NH₄⁺ concentrations than leaves from 3 mM NH₄⁺- grown plants and the 6 mM NH₄⁺ treatment resulted in further increased apoplastic NH₄⁺ concentrations. The apoplast of NO₃-grown B. erectus also had a high concentration of NO₃ (2.4 ± 0.7 mM), which was not the case for L. perenne (0.1 ± 0.05 mM).

Apoplastic pH values were lower in NH₄⁺-grown L. perenne plants (6.2-6.3 ± 0.1) compared with NO₃-grown plants (6.7 ± 0.1), whereas for B. erectus the pH values of the apoplastic solution were similar (6.4-6.6 ± 0.1) in all treatments (Table II).

Measured and Calculated Compensation Points in Plants Growing at Different Steady-State N Supply

The exchange of ammonia between shoots and the atmosphere responded linearly to changes in NH₃ concentrations between 0 to 20 nmol mol¹ in both species (Fig. 5). Compensation points were 5.0 and 6.8 nmol mol¹ for NO₃-grown plants of L. perenne and B. erectus, respectively. The corresponding values for NH₄⁺-grown plants were 5.8 and 9.0 nmol mol¹. The 6 mM NH₄⁺ treatment resulted in very high NH₃ compensation points of 11.9 and 18.3 nmol mol¹ for L. perenne and B. erectus, respectively. B. erectus showed greater response to the NH₃ supply than L. perenne as evidenced by 2 to 3 times higher leaf NH₃ conductance (slope of the curve). This difference was mainly related to differences in the stomatal conductance, which also was 2 to 3 times higher in B. erectus (Table I).

View larger version (18K): [in this window] [in a new window]   Figure 5.   Ammonia fluxes at different external concentrations above the leaves of 11-week-old L. perenne and B. erectus grown for 6 weeks with 3 mM KNO3 (A), 3 mM NH4HCO3 (B), or 6 mM NH4HCO3 (C) in the nutrient solution. Positive fluxes denote NH3 emission and negative fluxes represent NH3 absorption. Curves are typical examples of two replicate measurements. Correlation coefficients were >0.99 for all curves.

Compensation points estimated on the basis of apoplastic pH and NH₄⁺ concentrations (Table II) showed the same pattern in relation to N supply as those measured in the fumigation experiments, but the calculated values were from 2- to 9-fold lower than the measured.

Gas-Exchange Measurements in Plants Growing at Different Steady-State N Supply

Ammonia emission, photosynthesis, and transpiration were measured after 1.5 h of acclimatization in the cuvette with no external NH₃ added. All plants emitted NH₃ in the light period. Emissions were lowest for NO₃-grown plants, tended to increase in 3 mM NH₄⁺-grown plants, and were particularly high in B. erectus growing at 6 mM NH₄⁺ (Fig. 4A).

Net photosynthesis and transpiration rates did not change with the N treatment but were on a leaf area basis more than two times higher for B. erectus compared with L. perenne (Table I).

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Ammonium concentrations in leaf tissue and apoplast increased rapidly in both L. perenne and B. erectus as a response to the switch from NO₃ to NH₄⁺ nutrition (Fig. 1). This demonstrates that NH₄⁺ is rapidly taken up from the nutrient solution, translocated in the xylem from root to shoot (Mattsson and Schjoerring, 1996; Finnemann and Schjoerring, 1999), further transferred to the leaf apoplast, and absorbed into the leaf cells. A transient decrease in apoplastic pH of 0.4 pH unit (Fig. 2) coincided with the increase in apoplastic NH₄⁺ concentration that followed upon the switch to NH₄⁺ nutrition. Thus, uptake of NH₄⁺ from the apoplast into the symplast of the cell was associated with a net release of H⁺ (see also Hoffmann et al., 1992; Mattsson et al., 1998). The oppositely directed responses of apoplastic NH₄⁺ and pH resulted in roughly unchanged NH₃ compensation points (not shown). Ammonium concentrations in the apoplast decreased again and pH values were back to normal after 48 h of NH₄⁺ treatment (Fig. 1), suggesting repression of NH₄⁺ uptake in the roots due to feedback regulation elicited by amino acids, especially Gln, or by NH₄⁺ itself (Kronzucker et al., 1998). Removal of the NH₄⁺ after 24 h of NH₄⁺ treatment resulted in a rapid decrease of NH₄⁺ concentrations in both leaf apoplast and tissue. Both the increase and decrease of NH₄⁺ concentration were more pronounced in the apoplast than in the leaf tissue, indicating that the apoplastic NH₄⁺ pool is a very dynamic N pool rapidly responding to changes in external N supply. This rapid response may constitute a signaling system coordinating leaf N metabolism with the actual N uptake by the roots and the external N availability.

L. perenne plants have been shown to prefer NH₄⁺ over NO₃, both when supplied individually (Griffith and Streeter, 1994) or in combination (Clarkson et al., 1992). With the N treatments used in the present study it was, however, clear that, although NO₃ and NH₄⁺ were taken up at similar rates, NO₃ produced the largest L. perenne plants (Fig. 3). The fact that only small amounts of NO₃ were present in the leaves of L. perenne (see "Results") indicates that all the NO₃ taken up was utilized for growth. B. erectus, on the other hand, utilized N similarly regardless of the form taken up. Part of the higher leaf N concentration in NH₄⁺- compared with NO₃-grown plants could be attributed to 10 to 20 times higher Gln and Asn concentrations, which are well known responses to NH₄⁺ nutrition (Goodchild and Givan, 1990; Clarkson et al., 1992). The reduced growth of L. perenne under NH₄⁺ nutrition compared with NO₃ nutrition may be due to factors such as intracellular pH disturbance, increased contents of polyamines, disturbance in osmoregulation, or lack of C skeletons in the root (Gerendás et al., 1997).

At the high NH₄⁺ treatment (6 mM), growth was retarded in L. perenne and total N concentration was further increased. Growth inhibition by high external NH₄⁺ concentrations has often been demonstrated (Shelp, 1987; Magalhaes and Huber, 1989; Cramer and Lewis, 1993; Raab and Terry, 1995). Ammonium supplied at a high concentration (6 mM) to B. erectus produced plants of the same size as the 3 mM treatments but with increased total N concentration (Fig. 3, Table I), indicating that B. erectus was not able to utilize the extra NH₄⁺ taken up for increasing growth.

The 6-week NH₄⁺ treatment increased leaf tissue and apoplastic NH₄⁺ concentrations only slightly compared with NO₃-grown plants (Fig. 4, Table II). Seen together with the short-term responses to NH₄⁺ nutrition (Fig. 1), this indicates that at the 3 mM level of supply, NH₄⁺ was efficiently removed from the apoplast into the symplast. An increase in NH₄⁺ supply from 3 to 6 mM resulted, however, in further increased tissue NH₄⁺ concentrations, particularly in the roots, and in dramatically increased NH₄⁺ concentrations of the apoplast, indicating intracellular overloading.

It has been suggested (Sutton et al., 2001) that the tissue NH₄⁺ concentration can be used to estimate the NH₃ compensation point because it often shows a proportional increase to apoplastic NH₄⁺. Linear relationships between the external NH₄⁺ concentration and that in leaf tissue, xylem sap, or leaf apoplastic solution have been shown for barley (Hordeum vulgare) and oilseed rape (Brassica napus; Mattsson et al., 1998; Finnemann and Schjoerring, 1998, 1999). The correlation between leaf tissue NH₄⁺ concentrations and leaf apoplastic NH₄⁺ concentrations in the present experiment was not close (Fig. 6A, r² = 0.5), but still significant (P = 0.02). Emission of NH₃ for both L. perenne and B. erectus followed the same pattern as the apoplastic NH₄⁺ concentration and the leaf tissue NH₄⁺ concentration (Fig. 4); therefore, NH₃ emission was closely correlated (r² = 0.94, P = 0.001) with the NH₄⁺ concentration in the leaf tissue (Fig. 6B).

View larger version (16K): [in this window] [in a new window]   Figure 6.   Correlation between leaf tissue NH4+ concentrations and apoplastic NH4+ (A) and NH3 emission (B) of 11-week-old L. perenne and B. erectus grown for 6 weeks at different N treatments (3 mM KNO3, 3 mM NH4HCO3, or 6 mM NH4HCO3).

The plants supplied with 6 mM NH₄⁺ experienced pH fluctuations from 6 to 8.5 in the nutrient solution. The very high concentrations of NH₄⁺ in the root tissue and in the leaf apoplast of these plants (Fig. 4, Table II) may at least partly be due to NH₃ influx (Gerendás et al., 1997) because at high external pH the NH₃/NH₄⁺ ratio increases and NH₃ may diffuse uncontrolled over the plasma membrane (Walch-Liu et al., 2000).

The compensation points measured by gas exchange (Fig. 5) were higher for B. erectus than for L. perenne in all the treatments. B. erectus also showed a 3-fold higher leaf NH₃ conductance (slope of the curves in Fig. 5) and higher stomatal conductances (Table I) than L. perenne. A relatively high NH₃ conductance for B. erectus was also observed by Hanstein et al. (1999), who in addition found a high correlation between stomatal conductance and total leaf NH₃ conductance. The NH₃ compensation points reported by Hanstein et al. (1999) were, however, very low compared with the NH₃ compensation points obtained in the present study. The differences between the two experiments likely reflect the fact that in the experiment of Hanstein et al. (1999), the leaf N concentrations were considerably lower (3.3% on a dry matter basis) than in the present work (4.4% for 3 mM NH₄⁺-grown plants; Table I).

The NH₃ compensation points determined on the basis of NH₃ flux measurements in the present experiment were severalfold higher than the NH₃ compensation points derived from apoplastic NH₄⁺ concentrations and pH (Fig. 5, Table I). The general pattern showing the lowest compensation points for NO₃-grown plants, somewhat higher values for NH₄⁺-grown plants, and much higher NH₃ compensation points for plants growing at double-strength NH₄⁺ was, however, the same for both measured and calculated values (Fig. 5, Table I). In vegetative growth stages of oilseed rape, a good agreement was found between measured and calculated NH₃ compensation points, whereas during later stages several cases of discrepancy were observed (Husted and Schjoerring, 1996). In Luzula sylvatica, the NH₃ compensation points determined by NH₃ fumigation in cuvettes were 2 to 30 times higher than those calculated on the basis of apoplast extracts (Hill et al., 2001). Also, compared with NH₃ compensation points derived from micrometeorological measurements in the field, the apoplastic bio-assay has in several cases resulted in lower NH₃ compensation points (Sutton et al., 2001). One reason for the discrepancy may be temporal and spatial variability in apoplastic pH (Hanstein and Felle, 1999; Yu et al., 2000) and NH₄⁺ concentration. The latter may be due to convection of NH₄⁺ in the transpiration stream of water, which would lead to an enrichment of NH₄⁺ at the sites of evaporation and cause NH₃ compensation points to be greater in practice (gas exchange experiments). Despite uncertainties about the actual level of the NH₃ compensation point, the apoplastic bio-assay was still capable of predicting relative effects of N treatment and species differences. In general, NH₃ compensation points increased with NH₄⁺ nutrition as compared with NO₃ nutrition and also increased with increasing levels of N fertilization.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

The apoplast is a highly dynamic NH₄⁺ pool that rapidly responds to changes in external N. Both the N source (NO₃ or NH₄⁺) and amount of N supplied to the plants largely influence tissue and apoplastic NH₄⁺ concentrations and NH₃ emission to the atmosphere.

Increasing N availability leads to higher steady-state levels of inorganic N compounds and higher NH₃ compensations points in leaves of a grass species adapted to growth under N-poor conditions as compared with a species naturally occurring in N-rich ecosystems.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Plant Culturing

Seeds of Lolium perenne L. cv Bastion and Bromus erectus Huds. were germinated on wet filter paper in the dark for 7 d at 20°C. Three seedlings were transferred to hydroponics in 4-L high-density polyethylene containers and grown in a greenhouse with a day/night period of 16/8 h. To keep the day light intensity above 400 µmol m² s¹, supplementary light was given by HQI lamps (Power Star 400W, Osram, Munich). Day/night temperatures were 20°C ± 3°C/15°C ± 2°C. The nutrient solution consisted of: H_xPO₄^(3  x) (0.2 mM), K⁺ (1.2 mM), Mg²⁺ (0.6 mM), Ca²⁺ (0.3-0.9 mM), SO₄² (0.5 mM), Na⁺ (0.1 mM), Cl (0.1 mM), Fe-EDTA (50 µM), Mn (7 µM), B (2 µM), Zn (0.7 µM), Cu (0.8 µM), and Mo (0.8 µM). All plants were supplied with 3 mM KNO₃ during the first 4 weeks of the experiment. After this, the plants were grown for 6 weeks at three different N regimes: 3 mM KNO₃, 3 mM NH₄HCO₃, and 6 mM NH₄HCO₃. The solutions were renewed once a week. N was supplied and pH was adjusted three times a week.

Gas-Exchange Measurements

Gas-exchange measurements were performed during the 5th week of growth with different N treatments at a total plant age of 10 to 11 weeks. Uptake and emission of gaseous NH₃ in the shoots was monitored in NH₃ concentrations ranging from 0 to 20 nmol mol¹ air. A computerized cuvette system designed for simultaneous measurements of NH₃ exchange, photosynthesis, and transpiration was used (Mattsson and Schjoerring, 1996). Into this cuvette, 50 L min¹ of pressurized filtered air with the stated NH₃ concentration was led. The plant cuvette was installed in a growth chamber in which air temperature, light, and relative humidity of the air were controlled. The temperature was 20°C, the relative humidity was 65%, and the photon flux density was 300 µmol m² s¹. Ammonia was measured continuously by sampling of the air stream leaving the cuvette in a rotating denuder and analysis of the collected NH₃ by conductometry as described by Wyers et al. (1993). CO₂ and water were measured using a combined infrared gas analyser (Ciras-1, PP-Systems, Herts, UK). All data were logged once a minute and NH₃ flux, photosynthesis, and transpiration were calculated using the air flow rates through the cuvette and leaf areas of the plants. Estimates of stomatal conductances were based on transpiration data and a leaf temperature of 20°C.

Plant Harvest

Some containers with NO₃-grown plants were utilized for studying effects over time of switching the N source from NO₃ to NH₄⁺ and back again. Plant material from these containers was harvested for apoplastic extractions and tissue NH₄⁺ determinations at 0, 3, 6, 9, 24, and 48 h after start of the NH₄⁺ treatments, or following a 24-h NH₄⁺ treatment at 0, 3, 8, 24, and 48 h after switching back to NO₃ treatment. The other containers were all harvested after 11 weeks of growth and the fresh weights and leaf areas of the plants were monitored and roots were rinsed in deionized water. Some of the leaf material was used for apoplastic extractions, whereas other tissue material was immediately frozen in liquid N₂ and then stored at 80°C until extraction. The rest of the plant material was dried at 70°C, ground to a fine powder, and used for total C/N analysis.

Extraction of Leaf and Root Tissue

The plant tissue was homogenized in 10 mM formic acid in a cooled mortar with a little sand. The homogenate was centrifuged at 25,000g (2°C) for 10 min and the supernatant was transferred to 500-µL 0.45-µm polysulphone centrifugation filters (Micro VectraSpin, Whatman Ltd, Maidstone, UK) and spun at 5,000g (2°C) for 5 min. The filtered solution was used for analysis of NO₃ and NH₄⁺ concentrations.

Extraction of Apoplastic Fluid

Apoplastic solution was extracted with a vacuum infiltration technique slightly modified from the method described by Husted and Schjoerring (1995). The technique is based on vacuum infiltration of leaves with isotonic sorbitol solutions (490 mOsm for L. perenne and 530 mOsm for B. erectus corresponding to 392 and 424 mM sorbitol, respectively). Fifty-milliliter plastic syringes were mounted on a hydraulic arm that automatically moved the plunger up and down to infiltrate the leaf. The infiltrator exposed the leaves to a pressure of 4 atm and vacuum for 10 s, and repeated the procedure five times, thereby ensuring full infiltration. Infiltrated leaves were blotted dry and left in sealed plastic bags for 15 min at 18°C to allow full equilibration between apoplast and symplasm. After this time, the leaf apoplast was extracted by centrifugation at 2,000g for 10 min at 4°C and the apoplastic solution (20-100 µL) was collected in small vials (Eppendorf Scientific, Westbury, NY). pH of the apoplast was determined with a microelectrode (Metrohm, Herisau, Switzerland) and then the apoplast extracts were stabilized with ice-cold 20 mM HCOOH in a 1:1 (v/v) ratio. Apoplastic extracts were used for determination of NH₄⁺ concentrations.

Analysis of N Compounds

Ammonium was determined by fluorimetry on an HPLC system (Waters Corp., Milford, MA) equipped with a pump, a column oven with a 3.3-m stainless steel reaction coil, an autosampler cooled to 2°C, and a scanning fluorescence detector. The reaction between NH₄⁺ and o-phthaldehyde to form an alkylthioisoindole fluorocrome was performed at neutral pH with 2-mercaptoethanol as reducing agent. This fluorochrome was detected at an excitation wavelength of 410 nm and an emission wavelength of 470 nm (Husted et al., 2000).

Nitrate concentration in tissue and apoplast was determined according to Cataldo et al. (1975), where nitration of salicylic acid is recorded at 410 nm.

Amino acids were analyzed by the AccQ-Tag method developed by Cohen and Michaud (1993), following derivatization with 6-N-aminoquinolyl-N-hydroxysuccinimidyl carbamate and quantification in a HPLC system (method described in detail by Husted et al., 2000).

Total C and N concentrations were determined on an elemental analyzer (20-20, Europa Scientific, Crewe, UK) according to the Dumas method.

Compensation Point Determination

Ammonia compensation points were determined by two different procedures: (a) on the basis of NH₃ fumigation experiments in which the NH₃ exchange fluxes with the atmosphere were measured, and (b) by using the apoplastic NH₄⁺ and H⁺ concentrations (Husted and Schjoerring, 1995) according to the equation:
where _NH3 is the NH₃ compensation point, C_tot is the NH₄⁺/NH₃ concentration of the apoplast, and (H⁺) is the H⁺ activity of the apoplast. K_d and K_H are thermodynamic constants at 20°C.