INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The uptake of NO₃ by roots cells is a key process for higher plants because it is the first step of the assimilatory pathway providing most of organic nitrogen required for synthesis of biomolecules, including proteins and nucleic acids (Beevers and Hageman, 1980). More than 30 years of physiological investigations have led to the conclusion that at least three uptake systems are responsible for the influx of NO₃ into the roots (for review, see Clarkson, 1986; Glass and Siddiqi, 1995; Crawford and Glass, 1998; Daniel-Vedele et al., 1998; Forde, 2000). Two high-affinity transport systems (HATS) are able to take up NO₃ at low concentrations in the external medium, and display saturable kinetics as a function of the external NO₃ concentration ([NO₃]_o), with saturation in the range of 0.2 to 0.5 mM [NO₃ ]_o. One of these systems appears to be present even in plants never supplied with NO₃, and thus is considered as constitutive (cHATS). The other HATS is specifically stimulated by NO₃, and is consequently assumed to be inducible (iHATS). The maximum activity (V_max) recorded for the iHATS is generally much larger than that of the cHATS, suggesting that the former system plays a key role in the root uptake of NO₃ from external media where [NO₃]_o does not exceed 1 mM. The iHATS and cHATS appear to be genetically distinct because a mutant defective in the cHATS, but not in the iHATS, has been isolated in Arabidopsis (Wang and Crawford, 1996). In addition to these systems, a low-affinity transport system (LATS) is present, with a linear activity as a function of [NO₃]_o. The activity of the LATS becomes significant, if not predominant, when [NO₃]_o increases above 1 mM. In barley (Hordeum vulgare), where pioneer studies were conducted at the influx level (Siddiqi et al., 1990), the LATS is constitutive and thus does not require prior supply of NO₃ to the plants for activity. All three transport systems may coexist on a single cell (Guy et al., 1988), and their activities are believed to be additive. This structure of the root NO₃ uptake system seems to be of general occurrence because all three systems have been reported in a wide range of plant species. The picture for the root NH₄⁺ uptake system is quite the same, with both HATS and LATS identified (Glass and Siddiqi 1995), but with only indirect evidence for the existence of an inducible HATS for NH₄⁺ (Clarkson and Lüttge, 1991; von Wirén et al., 2000b).

As compared with this large body of evidence from physiological studies, the molecular identity of the transport proteins participating to the activity of the three transport systems is far less understood. At least two gene families (Nrt1 and Nrt2) are thought to encode NO₃ transporters (Crawford and Glass, 1998; Daniel-Vedele et al., 1998; Forde, 2000). The Nrt1 family includes, among others, Nrt1.1 and Nrt1.2 genes, which were reported to encode a NO₃-inducible and a constitutive LATS, respectively (Tsay et al., 1993; Huang et al., 1996, 1999). The Nrt2 family consists of genes highly homologous to those encoding the HATS for NO₃ and/or NO₂ in Chlalmydomonas reinhardtii and Neurospora crassa. Following the first identification of Nrt2 genes in barley (Trueman et al., 1996; Vidmar et al., 2000), homologs were found in Nicotiana plumbaginifolia (Quesada et al., 1997), soybean (Glycine max; Amarasinghe et al., 1998), and Arabidopsis, where two genes, AtNrt2.1 and AtNrt2.2, have been characterized to date (Filleur and Daniel-Vedele, 1999; Zhuo et al., 1999). Due to its very low expression level, very little is known about AtNrt2.2 (Zhuo et al., 1999). However, AtNrt2.1 has been extensively studied. The expression of this gene is mainly located in the roots, inducible by NO₃ (Filleur and Daniel-Vedele, 1999), and under general feedback repression by nitrogen status of the plant, possibly mediated by NH₄⁺ and amino acids (Lejay et al., 1999; Zhuo et al., 1999). Furthermore, AtNrt2.1 transcript level is strongly increased by dark-to-light transition and Suc supply to the roots, suggesting that it is also under control by photosynthetic activity of the shoots (Lejay et al., 1999). Thus, regulation of AtNrt2.1 expression is very similar to that of the HATS for NO₃, which is inducible by NO₃ (Jackson et al., 1973; Siddiqi et al., 1989), repressed by high nitrogen status of the plant and nitrogen metabolites (Hole et al., 1990; Lee et al., 1992; Muller and Touraine, 1992), and stimulated by photosynthesis and sugars (Delhon et al., 1996; Lejay et al., 1999). The demonstration that plant NRT2 proteins do have a transport activity on their own is still lacking, possibly because the actual transporter requires the presence of another protein, encoded by the Nar2-like gene (Zhou et al., 2000). However, functional evidence supporting the role of Nrt2 genes in root NO₃ uptake has been recently obtained. Constitutive overexpression of the NpNrt2.1 gene in N. plumbaginifolia led, under some circumstances, to a marked stimulation of root NO₃ influx in the low [NO₃]_o range (Fraisier et al., 2000). In Arabidopsis, the atnrt2 mutant deleted in both AtNrt2.1 and AtNrt2.2 genes has been isolated, and shown to be specifically affected in the HATS, but not in the LATS (Filleur et al., 2001). Thus, all available data suggest that NRT2 proteins are major components of the HATS in higher plants (Crawford and Glass, 1998; Daniel-Vedele et al., 1998; Forde, 2000; Filleur et al., 2001). However, the specific transport system (cHATS or iHATS) encoded by Nrt2 genes is not known. More generally, how regulation of root NO₃ uptake is modified by the alteration of the expression of these genes is still poorly understood. In N. plumbaginifolia, constitutive overexpression of NpNrt2.1 does not suppress the inhibition of root NO₃ influx by NH₄⁺, suggesting posttranscriptional control, or involvement of other NO₃ transporters in this response (Fraisier et al., 2000).

The aim of our work was to clarify the role of AtNrt2.1 and AtNrt2.2 genes in the regulation of NO₃ uptake in Arabidopsis. Therefore, the atnrt2 mutant has been further investigated to determine whether deletion of these two genes results in a modification of the response of root NO₃ influx to various treatments, including induction by NO₃, nitrogen starvation, change in external NO₃ availability, and repression by NH₄⁺. In wild-type (WT) Arabidopsis plants, these treatments are known to strongly affect NO₃ influx (Lejay et al., 1999; Zhuo et al., 1999), and reveal key regulatory processes allowing the adaptation of plants to both spatial and temporal changes in the availability of nitrogen in their environment.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The NO₃ Inducibility of the HATS Is Lost in the atnrt2 Mutant

The induction by NO₃ is one of the major regulations affecting the activity of the HATS for NO₃, leading to very pronounced changes in the uptake rate of NO₃ by the roots. To investigate this regulation in the two genotypes, the NH₄NO₃-grown plants were first subjected to nitrogen deprivation for 1 week, to ensure de-induction of the transport systems, and then resupplied with NO₃. At the end of the 7-d nitrogen starvation, both WT and mutant plants showed a similar reduced activity of the HATS, with ¹⁵NO₃ influx measured at 0.2 mM around 30 µmol h¹ g¹ root dry weight for both groups of plants (Fig. 1A). In WT plants, the supply of 4 mM NO₃ resulted in a dramatic increase in ¹⁵NO₃ influx, which reached nearly 100 µmol h¹ g¹ root dry weight after 12 h of treatment (Fig. 1A). This was associated with a strong increase in the steady-state AtNrt2.1 transcript level (Fig. 1B). This very classical response of the HATS was lost in the mutant, in which AtNrt2.1 mRNA was not detected (Fig. 1B), and which displayed only a 30% increase in ¹⁵NO₃ influx 12 h after exposure to NO₃ (Fig. 1A). These results indicate that the atnrt2 mutant is drastically deficient in a HATS component inducible by NO₃ (iHATS).

View larger version (37K): [in this window] [in a new window]   Figure 1.   Induction of 15NO3 influx (A) and AtNrt2.1 expression (B) by NO3 in roots of Wamilewskija WT (WS) and atnrt2 mutant (M) plants. The plants were grown on 1 mM NH4NO3 until the age of 5 weeks, and were transferred to nitrogen-free solution for 1 week to ensure de-induction of NO3 transport systems. Thereafter, the plants were supplied with 4 mM NO3. At the times indicated in the figure, 15NO3 influx was measured at 0.2 mM [15NO3]o, and root samples were harvested for northern-blot analysis of AtNrt2.1 mRNA accumulation. Root 15NO3 influx data are the means of 12 replicates ± SE.

Root NO₃ Uptake Is No More Regulated by the Nitrogen Status of the Plant in the atnrt2 Mutant

Another major regulation of the HATS for NO₃ is its derepression by nitrogen starvation, which is thought to illustrate feedback control by the nitrogen status of the plant. The transfer of WT plants from 10 mM NO₃ to nitrogen-free solution for 1 or 2 d prior to the uptake measurements resulted in a marked stimulation of both root ¹⁵NO₃ influx and AtNrt2.1 expression (Fig. 2). The same treatment had no effect on the HATS activity in the atnrt2 plants (Fig. 2A). Again, the AtNrt2.1 transcript could not be detected in the roots of the mutant (Fig. 2B). Another spectacular illustration of the lack of response of NO₃ uptake to nitrogen starvation in the mutant is given by the results of localized deprivation experiments with split-root plants. In WT plants, the transfer of one side of the split-root system to nitrogen-free solution for 3 d led to a compensatory increase in ¹⁵NO₃ influx in the other part of the root system still fed with NO₃ (Fig. 3). This compensatory response was never seen in the mutant, where ¹⁵NO₃ influx remained unchanged in the NO₃-fed roots after the initiation of the localized deprivation treatment (Fig. 3).

View larger version (32K): [in this window] [in a new window]   Figure 2.   Response of 15NO3 influx (A) and AtNrt2.1 expression (B) to nitrogen starvation in roots of WS and atnrt2 plants. The plants were grown on 10 mM NO3 until the age of 6 weeks, and were transferred to nitrogen-free solution for 24 or 48 h. Root 15NO3 influx was measured at 0.2 mM [15NO3]o, and root samples were harvested for northern-blot analysis of AtNrt2.1 mRNA accumulation. Root 15NO3 influx data are the means of 12 replicates ± SE.

View larger version (26K): [in this window] [in a new window]   Figure 3.   Response of 15NO3 influx to localized nitrogen deprivation in split-root WS and atnrt2 plants. The plants were grown on 1 mM NH4NO3 until the age of 5 weeks, and were transferred on 1 mM NO3 1 week before the experiments. The principle of the experiments (A) was to separate the root system of the plants in two parts. Localized nitrogen deprivation was initiated by the transfer of one side of the split-root system to nitrogen-free solution. Control plants remained supplied with 1 mM NO3 on both sides of the split-root system. Root 15NO3 influx (B) was measured at 0.2 mM [15NO3]o, in either the NO3-fed side, or both sides of the split root system, of plants under localized nitrogen starvation or control plants, respectively. The data are the means of six replicates ± SE.

The removal of NO₃ from the nutrient solution is not required to trigger derepression of NO₃ uptake by plants. This derepression also is observed under limiting supply of NO₃. In accordance, we investigated in both genotypes the response of the HATS-mediated NO₃ influx to the level of prior NO₃ provision to the plants. Therefore, plants grown on 1 mM NH₄NO₃ were transferred to either 1 or 4 mM NO₃ solution during the week preceding the measurements. The V_max of the HATS in the WT was strongly stimulated in plants supplied with 1 mM NO₃, as compared with those receiving 4 mM NO₃ (107 instead of 37 µmol h¹ g¹ root dry weight, Fig. 4, A and B, respectively). However, the difference in V_max between the two groups of plants was much lower for the mutant (Fig. 4). As a result, although V_max values did not differ markedly between the two genotypes, when supplied with 4 mM NO₃ (Fig. 4B), V_max of the HATS in WT plants on 1 mM NO₃ was more than twice as high as in the mutant (Fig. 4A).

View larger version (20K): [in this window] [in a new window]   Figure 4.   Kinetics of 15NO3 influx in function of [15NO3]o in WS and atnrt2 plants supplied either with 1 (A) or 4 (B) mM NO3 for 1 week before the measurements. The plants were grown on 1 mM NH4NO3 before the treatment at different NO3 concentrations in the medium. Root 15NO3 influx was measured at [15NO3]o ranging from 0.01 to 0.5 mM. The data are the means of 12 replicates ± SE. The calculated Vmax values are 107 and 37 µmol h1 g1 root dry weight for WS plants on 1 and 4 mM NO3, respectively. The corresponding values for atnrt2 plants are 44 and 33 µmol h1 g1 root dry weight on 1 and 4 mM NO3, respectively.

These data collectively suggest that the mutant is impaired in the uptake system component under feedback control by the nitrogen status, and thus is unable to modulate the activity of the HATS as a function of the nitrogen demand of the plant.

Root NO₃ Uptake Is Not Repressed by NH₄⁺ in the atnrt2 Mutant

To further investigate the alterations in the feedback control of the HATS in the mutant, experiments were performed to determine whether NO₃ uptake was still repressed by NH₄⁺ in the atnrt2 plants. In the experiment presented, the initial HATS-mediated ¹⁵NO₃ influx in plants supplied with 1 mM NO₃ for 1 week after growth on NH₄NO₃ medium was more than three times lower in the mutant than in the WT (Fig. 5). The addition of 1 mM NH₄⁺ to the nutrient solution containing 1 mM NO₃ led in the WT to a fast and strong decline in both root ¹⁵NO₃ influx and AtNrt2.1 expression, as expected (Fig. 5). At the opposite, this supply of a reduced nitrogen source did not affect the residual HATS activity in the mutant (Fig. 5A). Forty-height hours after the NH₄⁺ supply, the HATS-mediated ¹⁵NO₃ influx was very similar between the two genotypes.

View larger version (31K): [in this window] [in a new window]   Figure 5.   Repression of 15NO3 influx (A) and AtNrt2.1 expression (B) by NH4+ in WS and atnrt2 plants. The plants were grown on 1 mM NH4NO3 for 5 weeks, transferred for 1 additional week to 1 mM NO3, and supplied again with 1 mM NH4NO3. At the times indicated in the figure, 15NO3 influx was measured at 0.2 mM [15NO3]o, and root samples were harvested for northern-blot analysis of AtNrt2.1 mRNA accumulation in WS plants. Root 15NO3 influx data are the means of 12 replicates ± SE.

HATS and LATS for NH₄⁺ did not show a reduced activity in atnrt2 plants, as compared with WS plants (Fig. 6). Thus, lack of repression of NO₃ influx by NH₄⁺ in the mutant was not due to impaired NH₄⁺ uptake. It is interesting that ¹⁵NH₄⁺ influx at 0.2 mM was even slightly higher in the mutant than in the WT, suggesting compensation for the reduced NO₃ uptake rate in the mutant. This shows that the NH₄⁺ uptake systems are not inhibited in the atnrt2 plants, and thus that the nitrogen acquisition in the mutant is affected only due to the alteration of the HATS for NO₃.

View larger version (24K): [in this window] [in a new window]   Figure 6.   Root 15NH4+ influx in WS and atnrt2 mutant plants. The plants were grown either on 1 or 5 mM NH4NO3. Root 15NH4+ influx was measured at both 0.2 (A) and 10 (B) mM [15NH4+]o to provide estimation of HATS and HATS + LATS activity for NH4+, respectively. The data are the means of 12 replicates ± SE.

The atnrt2 Mutant Retains a Significant Uptake Capacity at NO₃ Concentrations in the Micromolar Range

The data obtained in the various ¹⁵NO₃ influx kinetics experiments (such as those in Fig. 4) were used to determine the concentration range of NO₃ for which uptake in the mutant was most affected. Therefore, the relative reduction of ¹⁵NO₃ influx in the atnrt2 mutant as compared with the WT was calculated for each external ¹⁵NO₃ concentration ([¹⁵NO₃]₀). To allow some statistical analysis, the data obtained in three independent experiments with NH₄NO₃-grown plants supplied for 1 week with 1 mM NO₃ were used as replicates. The maximum reduction of ¹⁵NO₃ influx in the mutant as compared with the WT was observed at 25 µM [¹⁵NO₃]₀ (Fig. 7), where ¹⁵NO₃ influx in the mutant represented only 20% of that in the WT. When [¹⁵NO₃]₀ increased above 25 µM, the relative reduction of influx in the mutant decreased, as expected from a similar activity of the LATS for NO₃ in both genotypes. At 100 µM [¹⁵NO₃]₀, the influx in the mutant represented 40% of that in the WT (Fig. 7). This value raised to 70% at 20 mM [¹⁵NO₃]₀ (data not shown). It is surprising that ¹⁵NO₃ influx in the mutant was also less affected when [¹⁵NO₃]₀ decreased from 25 to 10 µM (Fig. 7). At 10 µM [¹⁵NO₃]₀, ¹⁵NO₃ influx in the mutant represented almost one-half of that in the WT (Fig. 7). Thus, the atnrt2 plants have retained a significant ability to take up NO₃ in the micromolar range, although they lack a major component of the HATS.

View larger version (21K): [in this window] [in a new window]   Figure 7.   Reduction of 15NO3 influx in the atnrt2 plants, as compared with WS plants, as a function of [15NO3]o. The plants were grown on 1 mM NH4NO3 until the age of 5 weeks, and were supplied with 1 mM NO3 for 1 additional week before the measurements. The experiments were performed as described in Figure 4. The data are the means of those from three independent experiments ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The atnrt2 Mutant Is Deficient in the iHATS for NO₃

Our previous results with the atnrt2 mutant indicated that root NO₃ uptake was reduced in this genotype due to a specific inhibition of the saturable component of the NO₃ influx kinetics (Filleur et al., 2001), which is classically attributed to the activity of the HATS (Clarkson, 1986; Glass and Siddiqi, 1995; Crawford and Glass, 1998; Daniel-Vedele et al., 1998). However, the deletion of AtNrt2.1 and AtNrt2.2 genes in this mutant did not result in the complete disappearance of the NO₃ uptake capacity in the low [NO₃]_o range, as indicated by the reproducible observations that the atnrt2 plants kept a significant residual HATS activity (Figs. 1-5; see also Filleur et al., 2001). Moreover, the linear component of NO₃ influx, corresponding to the LATS working in the high [NO₃]_o range, was not affected by the mutation (Filleur et al., 2001). We have shown that neither HATS nor LATS for NH₄⁺ are altered inthe mutant as compared with the WT (Fig. 6). At the opposite, ¹⁵NH₄⁺ influx measured at 0.2 mM [¹⁵NH₄⁺]_o, was slightly higher in the atnrt2 plants than in WS plants, possibly indicating compensation for decreased NO₃ uptake by increased activity of the HATS for NH₄⁺. These observations collectively rule out the hypothesis that atnrt2 plants are deficient in NO₃ uptake due to a general detrimental effect on root ion transport. It is clear that the phenotype of atnrt2 plants is due specifically to the alteration of the HATS for NO₃, and is most probably directly related to the absence of the putative NO₃ transporters (or transporter components) encoded by either AtNrt2.1 or AtNrt2.2 genes. In accordance, complementation of the mutant with NpNrt2.1 of N. plumbaginifolia successfully restored a HATS-mediated ¹⁵NO₃ influx similar to that measured in the WT (Filleur et al., 2001).

Furthermore, our results suggest that the HATS component that is affected in the atnrt2 mutant corresponds to the iHATS, identified in many species on the basis of physiological approaches (Behl et al., 1988; Clarkson and Lüttge, 1991; Glass and Siddiqi, 1995). The fact that first provision of NO₃ to plants results in an accelerated rate of NO₃ uptake by roots in known for nearly 30 years (Jackson et al., 1973). Several lines of evidence suggest that this accelerated rate is due to de novo synthesis or activation of specific transport proteins, representing iHATS, as opposed to the constitutive component cHATS, which is present even in the absence of NO₃ (Jackson et al., 1973; Behl et al., 1988; Clarkson and Lüttge, 1991). The molecular identity of the iHATS for NO₃ in plants has been unknown until now. The fact that Nrt2.1 genes are inducible by NO₃ in various plant species (Trueman et al., 1996; Amarasinghe et al., 1998; Filleur and Daniel-Vedele, 1999; Forde, 2000), and highly homologous to CrNrt2 genes, encoding HATS for NO₃ and/or NO₂ in C. reinhardtii (Forde, 2000), led to the strong suspicion that they may encode transporters belonging to the iHATS. Our data provide the first functional evidence that AtNRT2.1 and/or AtNRT2.2 proteins are strictly required for the activity of the iHATS. This is shown by the very limited response of the HATS-mediated ¹⁵NO₃ influx to NO₃ supply to noninduced atnrt2 plants, whereas the same treatment resulted in a marked induction of both AtNrt2.1 expression and ¹⁵NO₃ influx in the WT (Fig. 1). On this basis, it is postulated that the residual HATS activity detectable in the mutant is mainly due to the cHATS. However, we noticed that a slight but reproducible stimulation of root ¹⁵NO₃ influx occurred in the mutant in response to the induction treatment (Fig. 1). This may indicate that other NO₃-inducible transporters participate in the HATS activity in the mutant. One obvious candidate would be AtNRT1.1, whose expression is inducible by NO₃ (Tsay et al., 1993), and which displays a dual activity of both high and low affinity (Wang et al., 1998; Liu et al., 1999). In an alternate manner, stimulation of the cHATS activity by NO₃ has also been proposed (Crawford and Glass, 1998).

The atnrt2 Mutant Lacks the HATS Component under Feedback Control by Nitrogen Status of the Plant

Feedback regulation of NO₃ or NH₄⁺ uptake by nitrogen status of the whole plant is a major feature of the overall control of root mineral nitrogen acquisition (Grignon, 1990; Imsande and Touraine, 1994; Glass and Siddiqi, 1995; von Wirén et al., 2000a). Nitrogen starvation or nitrogen-limiting conditions lead to a marked increase in the root capacity to take up NO₃ or NH₄⁺ (Lee and Rudge 1986), a response that is mostly due to the stimulation of the HATS-mediated influx of the two ions (Morgan and Jackson, 1988; Hole et al., 1990; Lee, 1993; Wang et al., 1993), and is associated in Arabidopsis with a strong increase in the expression of the NO₃ and NH₄⁺ transporter genes AtNrt2.1 and AtAmt1.1 (Gazzarrini et al., 1999; Lejay et al., 1999). This regulation is thought to be due to repression of NO₃ and NH₄⁺ transporters by reduced nitrogen metabolites accumulating in the tissues under satiety conditions (Jackson et al., 1986; Clarkson and Lüttge, 1991; Lee et al., 1992; Muller and Touraine, 1992). Both NH₄⁺ and amino acids were reported to exert this repression, also at the molecular level through inhibition of the expression of the Nrt2.1 and Amt1.1 genes (Krapp et al., 1998; Rawat et al., 1999; Zhuo et al., 1999; von Wirén et al., 2000b).

One of the major outcomes of our analysis of the atnrt2 mutant is that the HATS-mediated NO₃ uptake in this genotype is independent of the nitrogen status of the plant. Unlike what was observed in WS plants, root ¹⁵NO₃ influx in atnrt2 plants was no more up-regulated by nitrogen starvation (Figs. 2 and 3), and was almost insensitive to the decrease in external NO₃ availability (Fig. 4). Moreover, a major change in the regulation of the high-affinity NO₃ uptake in the mutant is also indicated by the absence of repression of NO₃ influx by exogenous NH₄⁺ supply (Fig. 5). This deregulation of root NO₃ uptake in atnrt2 plants does not result from the absence of the regulatory mechanisms responsible for repression of NO₃ and NH₄⁺ transport systems. This is shown by the fact that root ¹⁵NH₄⁺ influx in the low [¹⁵NH₄⁺]_o range was up-regulated in the mutant as in the WT by the decrease of the prior nitrogen provision to the plants (compare plants supplied with 5 or 1 mM NH₄NO₃ in Fig. 6). Thus, the atnrt2 mutant apparently is not a regulatory mutant. Rather, our result strongly support the hypothesis that root NO₃ uptake is deregulated in the mutant because the HATS component under feedback regulation by the nitrogen status of the plant is the iHATS, which is absent in the atnrt2 plants. A particularly important observation is that deletion of AtNrt2.1 and AtNrt2.2 genes resulted in the complete loss of the ability of NO₃-fed roots to develop the compensatory increase in NO₃ uptake in response to the nitrogen deprivation of another portion of the root system (Fig. 3). In WT plants, this up-regulation of NO₃ uptake in roots under localized supply of NO₃ is associated with a strong increase in the steady-state AtNrt2.1 mRNA level (Gansel et al., 2001). This demonstrates that AtNrt2.1 and/or AtNrt2.2 play a critical role in the adaptation of the plant to the spatial heterogeneity of NO₃ availability to the root system.

iHATS: A Role for AtNrt2.1 or AtNrt2.2?

One limitation in the conclusions of our work is related to the fact that the T-DNA insertion in the atnrt2 mutant unusually resulted in a quite large deletion, which affects both AtNrt2.1 and AtNrt2.2 genes, as well as possibly other unknown genes (Filleur et al., 2001). As a consequence, there is no definite proof that the phenotype of the mutant is specifically associated with the absence of AtNRT2.1 and/or AtNRT2.2 proteins. To unambiguously answer this question, complementation of the mutant with either AtNrt2.1 or AtNrt2.2 gene (or both) is required. However, the available information is consistent with the hypothesis that deregulation of root NO₃ uptake in the mutant results from the deletion of AtNrt2.1. First, the deletion in the atnrt2 mutant does not include any other identified gene known to play a role in ion transport or nitrogen nutrition (data not shown). Second, the regulation of NH₄⁺ uptake is not altered in the mutant, and no growth defect is observed when NH₄⁺ is provided to the plants. Third, complementation of the mutant with a constitutively expressed NpNrt2.1 gene from N. plumbaginifolia succeeded in restoring root NO₃ influx at the WT level (Filleur et al., 2001). Although we cannot rule out alternative hypotheses, these three observations indicate that the most straightforward explanation for the phenotype of the mutant is the lack of NO₃ transporters encoded by AtNrt2.1 and/or AtNrt2.2 genes.

In addition, our current knowledge suggests that AtNrt2.1 plays a much more important role than AtNrt2.2 in root NO₃ uptake. Unlike AtNrt2.1, AtNrt2.2 expression appears to be very restricted and is only detectable using reverse transcriptase-PCR (Crawford and Glass, 1998; Zhuo et al., 1999). Moreover, it is noteworthy that all aspects of the regulation of the HATS were observed at the molecular level for expression of AtNrt2.1 gene. This includes induction by NO₃ (Fig. 1; see also Filleur and Daniel-Vedele, 1999), derepression by nitrogen starvation (Fig. 2; see also Lejay et al., 1999), repression by high NO₃ provision to the plant (Fig. 4; see also Zhuo et al., 1999; Filleur et al., 2001), repression by reduced nitrogen metabolites such as NH₄⁺ and amino acids (Fig. 5; see also Zhuo et al., 1999), and stimulation by light and sugars (Lejay et al., 1999). At the exception of the regulation by light and sugars, which was not investigated here, the above-listed responses were all suppressed or markedly attenuated for the residual HATS in the atnrt2 mutant. As a consequence, the phenotype of the mutant concerning ¹⁵NO₃ influx was strictly related to the AtNrt2.1 expression level in the WT. The difference of HATS-mediated ¹⁵NO₃ influx between WS and atnrt2 plants was large only when AtNrt2.1 mRNA accumulation was high in the WS plants. This includes situations where AtNrt2.1 was strongly induced after NO₃ addition (Fig. 1), strongly derepressed in response to nitrogen starvation (Fig. 2), and strongly expressed due to low NO₃ concentration and absence of NH₄⁺ in the nutrient solution (Figs. 4 and 5). At the opposite, repressive conditions for AtNrt2.1 expression, such as a noninduced state of the plants (time zero in Fig. 1), high exogenous NO₃ supply (Fig. 4), or NH₄⁺ addition in the nutrient solution (Fig. 5), resulted in an almost identical root ¹⁵NO₃ influx in both WS and atnrt2 genotypes. Although nothing is known about the regulation of AtNrt2.2 expression, these observations strongly favor the hypothesis that the alterations in the activity and regulation of the HATS in the mutant as compared with the WT are due to the deletion of AtNrt2.1. As stated above, only the complementation of the atnrt2 mutant with either each one of the AtNrt2 genes (or both) will allow to assign a precise role to each protein in the global HATS for NO₃.

Root NO₃ Uptake in the Very Low Concentration Range

Due to the unaffected activity of the LATS for NO₃ in the atnrt2 mutant (Filleur et al., 2001), we expected to find that the relative reduction of root ¹⁵NO₃ influx in the mutant decreased with increasing [¹⁵NO₃]_o (Fig. 7). The observation that this relative reduction also decreased with decreasing [¹⁵NO₃]_o below 25 µM was much more surprising. This suggests that the atnrt2 plants are still able to take up NO₃ at a quite significant rate in the micromolar concentration range, and thus that both AtNRT2.1 and AtNRT2.2 proteins may not play a crucial role in the NO₃ acquisition from very diluted nutrient media. The explanation for this observation is not straightforward, and we can only speculate about the reasons for the occurrence of a very high-affinity NO₃ uptake in the mutant. Because our measurements were based on ¹⁵N accumulation into the roots, and not on NO₃ disappearance from the medium, this makes it unlikely that this apparent uptake of NO₃ was an artifact due to bacterial activity. Efficient NO₃ uptake from solutions containing micromolar concentrations of NO₃ may result from the activity of the cHATS, which is thought to mediate most of the residual ¹⁵NO₃ influx of the mutant in the low concentration range. However, kinetics analysis of ¹⁵NO₃ influx in the whole low [¹⁵NO₃]_o range (10-500 µM) did not show that the apparent K_m of the HATS was lower in the mutant than in the WT (Filleur et al., 2001), suggesting that the cHATS remaining in the mutant does not have a significantly higher affinity for NO₃ than the combined cHATS + iHATS in the WT. In an alternate manner, a relatively unaffected ¹⁵NO₃ influx in the very low [¹⁵NO₃]_o range in the mutant may indicate the presence of another, yet unknown, NO₃ transport system with a very high affinity for NO₃, which is not encoded by AtNrt2.1 or AtNrt2.2 genes.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Material and Growth Conditions

The Arabidopsis plants of the atnrt2 mutant (Filleur et al., 2001) and of the corresponding WT (WS) were grown hydroponically as described by Lejay et al. (1999). The seeds were germinated directly on top of modified Eppendorf tubes filled with prewetted sand. The tubes were then positioned on floating rafts transferred on tap water in a growth chamber under the following environmental conditions: light/dark cycle 8 h/16 h, light intensity 200 µmol s¹ m², temperature 22°C/20°C, and hygrometry 85%. After 1 week, the tap water was replaced by complete nutrient solution. Most of the experiments involved culture of the plants on 1 mM NH₄NO₃, which prevented any growth difference between the two genotypes (data not shown). However, nitrogen was sometimes supplied during the experiments as 5 mM NH₄NO₃ or 1, 4, or 10 mM NO₃ [mixture of KNO₃ plus Ca(NO₃)₂; see Lejay et al., 1999]. The other nutrients were added as described by Lejay et al. (1999). After 1 additional week, the plants were transferred to another growth room with similar environmental conditions except higher light intensity (300 µmol s¹ m²) and lower hygrometry (70%), and were allowed to grow for 3 to 4 additional weeks before the experiments. Nutrient solutions were aerated vigorously, renewed weekly, and the day before the experiments the pH was adjusted at 6.0.

All experiments were repeated two or three times, with typical results shown, except when results from all replicate experiments are presented.

Root Influxes of ¹⁵NO₃ and ¹⁵NH₄⁺

Root influxes of ¹⁵NO₃ and ¹⁵NH₄⁺ were assayed as described by Delhon et al. (1995) and Gazzarrini et al. (1999), respectively. The plants were sequentially transferred to 0.1 mM CaSO₄ for 1 min and to complete nutrient solution (pH 6.0) containing either ¹⁵NO₃ or ¹⁵NH₄⁺ (99% atom excess ¹⁵N) for 5 min, at the concentrations indicated in the figures. At the end of the ¹⁵N labeling, roots were washed for 1 min in 0.1 mM CaSO₄ and were separated from shoots. The organs were dried at 70°C for 48 h, weighed, crushed in a hammer mill, and analyzed for total ¹⁵N content using a continuous-flow isotope ratio mass spectrometer coupled with a carbon/nitrogen elemental analyzer (model ANCA-MS, PDZ Europa, Crewe, UK), as described by Clarkson et al. (1996). Root influx of ¹⁵NO₃ or ¹⁵NH₄⁺ is expressed in µmol h¹ g¹ root dry weight.

Kinetics of ¹⁵NO₃ Influx

Plants growing on 1 mM NH₄NO₃ were transferred to solutions containing either 1 or 4 mM NO₃ for 1 week before kinetics experiments were done. The kinetics of ¹⁵NO₃ influx as a function of external ¹⁵NO₃ concentration ([¹⁵NO₃]₀) was measured with [¹⁵NO₃]₀ ranging from 0.01 to 0.5 mM. Data transformation method based on the Michaelis-Menten formalism was used to obtain V_max and K_m estimates. Based on the kinetics studies (see "Results"), the influx by the HATS saturated between 200 and 500 µM [¹⁵NO₃]₀. Thus, influx at 200 µM [¹⁵NO₃]₀ was selected to assay the activity of the HATS in all other experiments.

Split-Root Experiments

Split-root experiments were performed as described by Gansel et al. (2001). In brief, after growth on 1 mM NH₄NO₃, the plantlet lawn was cleared to leave only one plant per tube. At the age of 5 weeks, the plants were transferred for 1 week to 1 mM NO₃ [0.5 mM KNO₃ plus 0.25 mM Ca(NO₃)₂]. The root system of each plant was then gently separated into two approximately equal parts, each transferred to a separate container (3 L), with the tube supporting the plant fixed on top of the edge between the two containers. The plants were then allowed to recover from mechanical shock for 3 d, with both sides of the root system supplied with 1 mM NO₃ solution. The localized starvation treatment was initiated by the transfer of one side of the split-root system to nitrogen-free solution.

Northern Blots

RNA extraction was performed as described previously (Lobreaux et al., 1992). Total RNA (10 µg) were separated by electrophoresis on MOPS [3-(N-morpholino)-propanesulfonic acid]-formaldehyde agarose gel and blotted on nylon membrane (Hybond N+, Amersham, Freiburg, Germany). Membranes were prehybridized for 2h at 65°C in 0.5 M NaHPO₄, 1% (w/v) bovine serum albumin, and 7% (w/v) SDS (pH 7.2 with H₃PO₄). Hybridizations were performed overnight at 65°C after addition of the randomly primed ³²P-labeled cDNA probe in the prehybridization buffer. Membranes were washed twice at room temperature for 2 min and twice at 65°C for 15 min with 0.5× SSC and 0.1% (w/v) SDS. The cDNA probe used in this work corresponded to the full-length cDNA of AtNrt2.1 (Filleur and Daniel-Vedele, 1999). A 25S rRNA probe was used as a reference.