Under temperate climates and in cultivated soils, nitrate is the most important source of nitrogen (N) available for crops and, before its reduction and assimilation into amino acids, must enter the root cells and then move in the whole plant. The aim of this review is to provide an overall picture of the numerous membrane proteins that achieve these processes by being localized in different compartments and in different tissues. Nitrate transporters (NRT) from the NRT1 and NRT2 families ensure the capacity of root cells to take up nitrate, through high- and low-affinity systems (HATS and LATS) depending on nitrate concentrations in the soil solution. Other members of the NRT1 family are involved subsequently in loading and unloading of nitrate to and from the xylem vessels, allowing its distribution to aerial organs or its remobilization from old leaves. Once in the cell, nitrate can be stored in the vacuole by passing through the tonoplast, a step that involves chloride channels (CLC) or a NRT2 member. Finally, with the exception of one NRT1 member, the transport of nitrite towards the chloroplast is still largely unknown. All these fluxes are controlled by key factors, the ‘major tour operators’ like the internal nutritional status of the plant but also by external abiotic factors.
Nitrogen is one of the most essential nutrients for plants and is involved in the building of the fundamental bricks of life: nucleotides, amino acids, and proteins. Only some plant species are able to use atmospheric nitrogen due to their capacity of a symbiotic relationship with specific microorganisms (Gordon et al., 2001). The other species find their resources in the soil where nitrogen is present in different forms. For example, the soil solution may contain different organic N forms such as soluble proteins or amino acids derived from proteolytic processes. In particular ecosystems, some plant species are able to use these organic N forms, as described in a recent review (Näsholm et al., 2009). In temperate climatic conditions, inorganic N forms are predominant and fertilizers are often supplied as nitrate, ammonium or urea (http://www.fertilizer.org/ifa/). In this review, the focus will be on nitrate as a main nitrogen source for Arabidopsis plants, and why and how it is present in soils, how it is taken up by plant roots, distributed in the whole plant or inside one cell, and, finally, what the ‘tour operators’ are that organize and regulate this complex journey.
Origin of nitrate
It is axiomatic that all the nitrate on earth today originates from atmospheric nitrogen gas (N2). However, this conversion to nitrate has involved both abiotic and biotic processes. In Fig. 1 a time-course of nitrate production on earth, from early pre-biotic times to the present day, is proposed. Originally, in an anoxic early-earth atmosphere, small amounts of NO (nitric oxide) would have been produced by lightning acting in the atmosphere (N2 and CO2) and this NO settled into the oceans where nitrate and nitrite would have formed (Canfield et al., 2006). The early organisms (facultative heterotrophs such as Archaea and Bacteria) evolved a process of using this nitrate as an electron acceptor for respiration in the absence of oxygen (Pina-Ochoa et al., 2010). Some of these bacteria can accumulate up to 800 mM nitrate in their vacuole (Schullz-Vogt, 2006). Recent genome sequencing of some of these bacteria (Mussmann et al., 2007) indicates that the transport of nitrate into the vacuole most probably involved an early version of chloride channels/transporters, of which one has recently been discovered to load nitrate into Arabidopsis vacuoles (see ‘Travelling within the cell’). A reduction of abiotic N fixation soon after the origin of life resulted in a possible N crisis, which triggered the development of biological nitrogen fixation (Navarro-Gonzalez et al., 2001). Once aerobic organisms (cyanobacteria) evolved on earth, the atmosphere was oxygenated and production of nitrate steadily increased, and as Falkowski points out, this production of nitrate must have evolved after the advent of oxygenic photosynthesis, as there can be no nitrate without oxygen (Falkowski, 1997). Roughly 42 million years ago, the Arabidopsis genus split from the Brassica complex (Beilstein et al., 2010), so clearly, the evolutionary ancestors of Arabidopsis were exposed to nitrate and therefore evolved mechanisms to absorb and transport nitrate. More recently, the introduction of the Haber–Bosch process dramatically increased the amount of ammonium-nitrogen produced and, when added as fertilizer to soil, has resulted in a large increase in nitrogen and nitrate in soils.
Nitrate in soils
The vast majority of the total nitrogen in soil (>98%) is in organic matter, which is mostly unavailable directly to plants. Only a very small fraction of this total N is in inorganic forms. This conversion to inorganic forms (ammonium, nitrite, and then nitrate) occurs via biological processes involving soil microorganisms. Generally in aerated soils, nitrate predominates and if it is not absorbed by plant roots or utilized by microorganisms, it is available for leaching. Nitrate leaching occurs because nitrate has a very weak affinity to form surface complexes with soil minerals and most soils more strongly adsorb cations compared with anions (Strahm and Harrison, 2006). Because of this, the diffusion coefficient for nitrate is very high, i.e. nitrate diffuses quickly to areas of depletion such as to the rhizosphere. The concentration of nitrate in soil solutions is very variable, both spatially and temporally (Miller et al., 2007), depending very much on soil type, fertilizer addition, microbial activity etc. Nitrate concentrations in soil solutions have been reported by a number of surveys and reviews (Reisenauer, 1966; Wolt, 1994). Concentrations generally range from very low levels of a few hundred micromolar to around 20 mM, the highest up to 70 mM. Consequently, plant roots are exposed to this wide range of concentrations, and it is then understandable that plants have developed a system of regulation of nitrate transporters to cope with this variability (see the next section, ‘Travelling from the soil to plant roots’). It is important to note that the plant's demand for N is the actual pacemaker for nitrate uptake by the roots from the soil solution.
Travelling from the soil to plant roots
The selective uptake of water and solutes to support plant growth and development is achieved by the root system. Its species-specific size and its architecture may vary depending on a wide variety of developmental and environmental factors (reviewed in Zhang et al., 2007). The main root, the lateral roots, and the root hairs are the major components of a root system that are organized in a simple or complex manner, such as in Arabidopsis or in cereal crops (Osmond et al., 2007). The nitrate fluxes through the root part can be considered in two dimensions: the radial and basipetal fluxes.
Transport across the root to the vascular tissue can occur through the inter-cellular space (apoplastic transport) or through the different cell layers (symplastic transport), namely the epidermis, cortex, and endodermis in Arabidopsis. Characterized by cell wall deposition that surrounds the cell like a belt, the endodermis layer separates the outer (peripheral) from the inner (central) root cell layers (Enstone and Peterson, 2002). Recently, Baxter and co-workers showed that suberin on the endodermis acts as a barrier to the apoplastic transport of both water and mineral ions (Baxter et al., 2009). Thus, whatever the nitrate routes from the soil to the stele, nitrate must enter either the epidermal, the cortex or the endodermal cell before being loaded into the xylem vessels.
Early studies indicated that ions enter the plant at the root tip and are translocated basipetally to the shoot (Cannings and Kramer, 1957). However, since this publication, many studies have shown that the root axis is characterized by regions that are heterogeneous in their uptake and transport capacities (Siebrecht et al., 1995).
Depending on the external nitrate concentrations, two different uptake systems occur within the plant: the so-called high- (HATS) and low-affinity (LATS) systems operate when nitrate is present in the soil at low (<1 mM) or high concentration (>1 mM), respectively. Physiological studies demonstrated that each system is composed of constitutive and nitrate inducible components (Glass and Siddiqi, 1995).
In the apical part of the Arabidopsis root, one master gene is expressed, NRT1.1, coding for a protein that, at the same time, plays the roles of nitrate transporter (Tsay et al., 1993) and nitrate sensor to activate the expression of nitrate-related genes (Ho et al., 2009). NRT1.1 is also involved in different signalling processes leading to changes in root development and seed germination (Ho and Tsay, 2010). Very recently, Krouk et al. showed that NRT1.1 is able to facilitate auxin transport in a nitrate repressible way (Krouk et al., 2010). NRT1.1 was thus proposed to repress lateral root growth at low nitrate availability by promoting basipetal auxin transport out of these roots. Discovered in the 1990s, the NRT1.1 gene, also named CHL1, belongs to the NRT1 family comprising 53 members in Arabidopsis (reviewed in Tsay et al., 2007). It was first described as a low-affinity nitrate transporter characterized by a Km of 5 mM, involved in the LATS. The expression of the NRT1.1 gene is inducible by nitrate, suggesting a role of the protein in the inducible component of the LATS. Near the root tip, it is found primarily in the epidermis where nitrate can enter directly by the symplastic route and, further from the root tip, its expression takes place in the cortex and endodermis (Huang et al., 1996). The same protein was also shown to be involved in HATS (Liu et al., 1999). Liu et al. demonstrated that the phosphorylation–dephosphorylation of threonine T101 was responsible for the shift from a low to a high affinity uptake characteristic (Liu and Tsay, 2003). The CBL-interacting protein kinase CIPK23 (SnRK3.23) is able to phosphorylate T101 under low nitrate conditions, allowing NRT1.1 to function as a high-affinity nitrate transporter (Ho et al., 2009).
In the root tip and in the mature regions of roots, a second gene of the same family, AtNRT1.2, is expressed in root hairs and root epidermis (Huang et al., 1999). In contrast to the nitrate inducible expression of the NRT1.1 gene, NRT1.2 mRNA accumulates without prior nitrate exposure and was shown to participate to the constitutive component of the LATS.
Absent from the Arabidopsis root tip but expressed in the mature part of the root, the AtNRT2.1 gene belongs to another gene family, comprising seven members in Arabidopsis (Orsel et al., 2002). GUS expression, directed by the AtNRT2.1 promoter, is found in the epidermal, cortical, and endodermal cell layers of mature root parts (Nazoa et al., 2003). The NRT2.1 protein is localized on the plasma membrane (Chopin et al., 2007,a) and is the major actor of the HATS system, particularly when the external nitrate concentration is low (Filleur et al., 2001; Orsel et al., 2004). However, the NRT2.1 protein is not acting alone in the nitrate transport process but requires a second protein, named NAR2 (AtNRT3.1), as shown in barley (Tong et al., 2005) or in Arabidopsis (Okamoto et al., 2006; Orsel et al., 2006). In addition to possibly being involved in NRT2.1 stability (Wirth et al., 2007), NAR2 was recently found to be physically associated to NRT2.1 at the plasma membrane and to form the active nitrate transporter as a tetramer consisting of two subunits each of AtNRT2.1 and AtNAR2.1 (Yong et al., 2010). A second AtNRT2.2 gene is located near the AtNRT2.1 on chromosome I. Li and co-workers dissected the HATS activity in AtNRT2.1 and AtNRT2.2 single and double mutants and showed that AtNRT2.1 is responsible for 72% of HATS activity. This result suggests that AtNRT2.2 makes only a small contribution to the system, except when AtNRT2.1 is lost, its contribution increases and results in a partial compensation (Li et al., 2007).
Net root nitrate uptake results from the balance between nitrate influx, described above, and passive nitrate efflux whose physiological role still remains obscure. In response to biotic (Garcia-Brugger et al., 2006) or abiotic stresses (Macduff and Jackson, 1992; Aslam et al., 1995), this efflux increases and can even overcome the influx leading to net nitrate excretion. Using a functional biochemical approach, the NAXT1 protein from a sub-family of seven members from the large NRT1 family was identified and shown to be responsible for the prolonged root nitrate excretion occurring after acidification of the culture medium (Segonzac et al., 2007). The protein is localized on the plasma membrane of the cortical cells of mature roots but other members of this NAXT sub-family are expressed in different cellular types such as the root epidermis or root stele, possibly being involved in nitrate xylem loading.
Subsequently to this loading, nitrate is then translocated to the upper part of the plant where other NRT proteins allow its distribution in the whole plant.
Travelling within the plant
Nitrate concentrations vary to a large extent according to organs and nitrate supply. Figure 2 summarizes the nitrate contents of plants in the vegetative phase (Fig. 2A) grown under two nitrogen regimes, and plants in the reproductive stage (Fig. 2B) grown with full nutrient supply (J Dechorgnat, M Jossier, unpublished results). Nitrate is found in all organs and must therefore reach each of them by different routes.
The first step to transport nitrate towards the plant aerial part is the loading into the xylem vessels of the vascular stele and it has been shown that this step is a separate control point for transport of nutrient into the shoot (Herdel et al., 2001). The release of ions, such as potassium, across the plasma membrane of xylem-parenchyma cells into the xylem occurs through ion channels whose activity is regulated by membrane voltage (Gaymard et al., 1998). Kölher and co-workers used the patch-clamp technique on protoplasts isolated from xylem-parenchyma barley cells and showed that apoplastic nitrate itself controls the voltage-dependence of X-QUAC, the Quickly Activating Anion Conductance, exerting a positive feed-back on its loading into the xylem (Köhler et al., 2002). The potential of the plasma membrane is negative (around –100 mV), thus nitrate might only be passively transported from the parenchyma cells into the xylem. However, a nitrate transporter belonging to the NRT1 family, AtNRT1.5, was found to be expressed in the pericycle cells surrounding the protoxylem (Lin et al., 2008). Expression in Xenopus oocytes revealed that it is able to transport nitrate in the low affinity range but in both directions. In addition, nitrate uptake is not affected in atnrt1.5 knock-out mutants but translocation of nitrate, followed by the 15N tracer, is slower in the mutant compared with the wild type. Taken together, these results suggest that AtNRT1.5 plays a major role in the loading of root nitrate into the xylem from the pericycle cells but other mechanisms must co-exist within the root.
Moving up to the aerial part of the plant, the petiole is an important organ, ensuring nitrate distribution towards the leaves. It is generally characterized by a high nitrate content (Raab and Terry, 1995). One member of the NRT1 family, AtNRT1.4, encodes a low-affinity nitrate transporter that is constitutively and specifically expressed in petioles and mid-ribs of Arabidopsis leaves. In addition, atnrt1.4 knock-out mutants showed a reduced nitrate content in the petiole and the mid-rib (45–50% of wild-type levels), while that of the lamina was slightly increased (Chiu et al., 2004). These results suggest that this NRT1 protein plays a role in the nitrate distribution to the lamina of the leaves where nitrate reduction takes place. However, the AtNRT1.4 mRNA, detected by whole-mount in situ hybridization, does not seem to be located in the vascular tissues. The mild phenotype of the mutants indicates that AtNRT1.4 is not the only actor of nitrate distribution to the lamina. Another gene of the same family, the AtNRT1.8, is in fact expressed predominantly in parenchyma cells within the vasculature in both roots and shoots of seedlings, suggesting a role in xylem loading and/or unloading (Li et al., 2010). Further experiments showed that xylem sap collected from mutants accumulate more nitrate than the wild-type, allowing the authors to demonstrate that the protein, localized in the plasma membrane, is a low-affinity nitrate transporter, the function of which is to transport nitrate across parenchyma cells to unload nitrate from the xylem sap. Thus, two proteins, AtNRT1.5 and AtNRT1.8, closely related phylogenetically, are, respectively, involved in nitrate loading into the root stele and nitrate unloading from the root stele or from the shoot vasculature, ensuring the fine-tuning of long-distance transport from the roots to the shoots.
Once inside the cell, nitrate can follow different routes. It can be stored as ‘temporary parking’, in the vacuole, where it can serve as an osmoticum or as a stock to be mobilized when the plant is growing under unfavourable conditions (see ‘Travelling within the cell’). Similarly to amino acids remobilized from proteins during seed filling (Masclaux-Daubresse et al., 2010), nitrate can be exported from the vacuole and transported towards growing organs, such as young leaves, when the external resource availability is decreasing, or when the mature leaf is ageing and becomes a source leaf. Thus, membrane(s) protein(s) must transport nitrate through the plasma membrane towards the vascular tissues of source organs. Looking at this specific expression, Fan et al. showed recently that the AtNRT1.7 was a good candidate to achieve this function (Fan et al., 2009). The protein is localized in the plasma membrane of companion cells and sieve element complex of mature leaves and characterization of null mutants revealed that they accumulate higher amounts of nitrate in older leaves and are defective in remobilizing nitrate from older leaves to young leaves. The AtNRT1.7 gene expression is also induced by nitrogen starvation and, as a consequence, null mutants showed growth retardation under these stress conditions (Fan et al., 2009). Two genes of the NRT2 family, AtNRT2.4 and AtNRT2.5, are also regulated by nitrogen availability, being induced by nitrogen starvation and repressed by nitrogen re-supply (Orsel et al., 2002; Okamoto et al., 2003). They could represent complementary actors for nitrate remobilization and the double or triple mutants’ sensitivity to nitrogen starvation would be interesting to study.
In some species, nitrate uptake seems to be completely repressed by flowering and no longer operates during seed filling (Malagoli et al., 2004) whereas in Arabidopsis this process is lower than during the vegetative stage but still occurs (Masclaux-Daubresse et al., 2010). In addition, as shown in Fig. 2, not only the vegetative organs contain nitrate but also reproductive organs such as stems, siliques, and even seeds (Chopin et al., 2007,b). Although nitrate is not the major nitrogen form in seeds, it plays an important role in the regulation and breaking of seed dormancy (Alboresi et al., 2005). How does nitrate reach this particular organ? One good candidate is the AtNRT1.6, a low-affinity nitrate transporter localized on the plasma membrane (Almagro et al., 2008). The gene is expressed in the vascular tissue of the silique and in the funiculus, which ensures the link between the developing seed and the sustaining silique. In agreement with the expected decrease in nitrate supply to developing embryos, atnrt1.6 seed nitrate content was found to be reduced. Interestingly, the mutant also showed an increased seed abortion due to embryo abnormalities at the globular stage of development (Almagro et al., 2008). This phenotype was found only under normal nitrate supply but not under deficiency conditions, suggesting that the plant can sense the nitrogen availability and modulate its nutrition strategy for grain development. In summary, up to now, 12 genes are shown to encode transport proteins involved in nitrate fluxes through the whole plant and they are represented in a schematic diagram in Fig. 3.
Travelling within the cell
Techniques for measuring intracellular nitrate concentrations include the compartmental tracer efflux method, nitrate selective-microelectrodes, cellular nitrate reductase assays, and 133Cs nuclear magnetic resonance (NMR) (Miller and Smith, 2008). The nitrate concentrations measured depend very much on the technique used, cell type, and growing conditions, but a few generalizations can be made. Cytosolic nitrate concentrations (1–6 mM) are more stable than vacuolar concentrations (5–75 mM) when the external nitrate concentrations vary 10 000-fold (van der Leij et al., 1998; Cookson et al., 2005; Miller and Smith, 2008). This result is consistent with the idea that the vacuolar nitrate pool provides a reservoir of stored nitrate that can be used to maintain the cytosolic nitrate concentration. This was confirmed by the fact that the nitrate pool is not maintained at a steady-state in the cytosol of the root tip cells where the vacuoles are not fully developed (Radcliffe et al., 2005; Miller and Smith, 2008). Nitrate can also be found in chloroplasts. For example, in isolated chloroplasts from spinach leaves, the concentration of this anion is around 5 mM and stays constant independently of the external nitrate conditions (Schroppelmeier and Kaiser, 1988). Finally, it is assumed that nitrate might be present in other organelles, mainly ER (Siddiqi and Glass, 2002). The future development of fluorescent probes that report nitrate concentrations would be the ultimate tool for measuring more precisely intracellular nitrate concentrations and following possible changes in different nitrate pools according to environmental conditions. Recent advances in FRET (Fluorescence Resonance Energy Transfer) and FP (Fluorescent Protein) technologies have allowed the production of fluorescent probes for calcium, glucose, sucrose, and phosphate (Okumoto, 2010). Similar probes for nitrate have yet to be developed (Okunola et al., 2008). However, the discovery of proteins that transport nitrate and their crystallization provides future hope for developing such nitrate-specific fluorescent probes. The challenge will be to ensure a nitrate probe highly selective for nitrate (over nitrite and chloride) and relatively pH-insensitive for use in different cellular compartments.
It is now well established that nitrate in the vacuole is not only important for storage but also for turgor maintenance (McIntyre, 2001; Miller and Smith, 2008). Very early studies suggested the presence of a /H+ transporter in the tonoplast (Schumaker and Sze, 1987; Miller and Smith, 1992). Recent evidence shows that the activities of these nitrate transporters are connected to the activity of the vacuolar H+-ATPase, confirming the key role of the pH gradient as a driving force of fluxes through the tonoplast (Krebs et al., 2010). AtCLCa was the first /2H+ exchanger discovered, responsible for nitrate loading into the vacuole in Arabidopsis (Fig. 4) (De Angeli et al., 2006). This protein belongs to a family of chloride channel/transporters. The difference of selectivity between AtCLCa and the other AtCLCs is mainly due to the presence of a proline residue in the selectivity motif (Bergsdorf et al., 2009; Wege et al., 2010; Zifarelli and Pusch, 2010).
Knock-out plants for AtCLCa have only a 50% reduction in nitrate accumulation compared to the wild-type (Geelen et al., 2000; Monachello et al., 2009). This indicates that there are additional genes that are also responsible for nitrate loading into the vacuole. Recently, AtCLCb, the AtCLCa closest related protein, has also been shown to be a nitrate transporter in the tonoplast (von der Fecht-Bartenbach et al., 2010). Nevertheless, it is still unclear if AtCLCb is involved in nitrate loading or release from the vacuole. In contrast to the CLC family, only one member of the NRT families has been shown to be involved in nitrate flux within the cell, the AtNRT2.7 protein. The gene is specifically expressed at the final stage of seed maturation and the protein is localized on the tonoplast (Chopin et al., 2007,b). Characterization of null mutants, showing reduced nitrate contents in mature seeds, suggests that the AtNRT2.7 protein is involved in the loading of nitrate into the embryo vacuoles, and thus complements the role of the AtNRT1.6 in seed nitrate feeding (Almagro et al., 2008, see previous section ‘Travelling within the plant’).
The analyses of these vacuolar transporters revealed that co-operation between the tonoplast and plasma membrane exists at the molecular level to regulate the fluxes to maintain the cytosolic nitrate homeostasis. For example, in atclca mutants, both expression of NRT1.1 and NRT2.1 are reduced (Monachello et al., 2009; Wang et al., 2009). Further analyses will be needed in the future to understand the integrated network responsible for this process.
Nitrate homeostasis is buffered as well by the cytosolic nitrate reductase (NR), the first enzyme in the nitrate assimilation pathway, reducing nitrate into nitrite (Fig. 4) in an NAD(P)H-dependent manner (Cookson et al., 2005; Fan et al., 2006). Once nitrite is formed in the cytosol, it is translocated into the chloroplast to be reduced into ammonium by the nitrite reductase (NiR). The transport of nitrite from the cytosol to the chloroplast remains largely unknown in higher plants. Although a free diffusion has been proposed, the physiological nitrite concentration in the cytosol (in the range of micromolar) is too low to allow this mechanism (Kawamura et al., 1996) and requires the presence of transporters. A nitrite transporter has been isolated in cucumber: CsNitr1 (Cucumis sativus nitrite transporter) (Sugiura et al., 2007). This transporter belongs to the proton-dependent oligopeptide transporter (POT) family and is able to export nitrite when expressed in yeast. CsNitr1 is localized in the inner membrane of the chloroplast envelope and Arabidopsis mutants of the orthologous gene (At1g68570), which belongs to the NRT1 family (Tsay et al., 2007) over-accumulates compared with the wild type. Another transporter which could participate to nitrite transport in the chloroplast is the AtCLCe member of CLC family (Fig. 4), targeted to thylakoid membranes (Marmagne et al., 2007). Knock-out mutants of AtCLCe present an altered photosynthetic activity, an under-accumulation of nitrate associated with an over-accumulation of nitrite (Monachello et al., 2009). Nevertheless, it is difficult to conclude as to the exact function of AtCLCe as the expression of several genes implicated in nitrate fluxes (NRT2.1 or NRT1.1) are modified in atclce mutants (Monachello et al., 2009). Another actor of fluxes from the cytosol to the chloroplast is the highly conserved PII protein, a sensor of carbon/nitrogen balance and energy status in bacteria, cyanobacteria, and plants (Lillo, 2008). In Arabidopsis, one PII homologue (GLB1) has been cloned (Hsieh et al., 1998). Arabidopsis PII mutant seedlings present an increased sensitivity to toxicity (Ferrario-Mery et al., 2005). This sensitivity can be explained by an increase by 30–40% of the light-dependent uptake observed on isolated chloroplasts of the PII mutant compared with the wild type (Ferrario-Mery et al., 2008). Thus, PII appears to be involved in the down-regulation of uptake into the chloroplast; however, its partners in this mechanism remain to be identified.
The tour operators
The importance of nitrate for growth and development lead to the existence of tightly regulating systems to co-ordinate nitrate transport and assimilation with developmental needs and environment cues. Globally, operators of nitrate trips can be summarized as (i) nitrate itself, (ii) the plant N status (amino acids) exerting a negative regulation on transporters, and (iii) the plant C status (light, carbohydrate) or reduced nitrogen forms (ammonium or amino acids) stimulating or inhibiting, respectively, nitrate transport and assimilation (see Gojon et al., 2009, for a review).
Nitrate itself, either as an external nutrient or as an intracellular ion induces specific sets of genes to shape plant shoot and root morphology, influence flowering time, and relieve seed dormancy (Bernier et al., 1993; Alboresi et al., 2005; Zhang et al., 2007). The typical primary nitrate response, observed after re-providing nitrate to short-term starved plants, is characterized by the induction of nitrate transporters (NRT2.1; NAR2; NRT1.1) and nitrogen assimilation genes (nitrate reductase, glutamine synthetase). Nitrate uptake is decreased if internal N status is high or when other nitrogen sources are provided. Glutamine was shown to be the most powerful repressor of NRT2.1 transcript and represents the main shoot-to-root N status signal to regulate N uptake (Vidmar et al., 2000; Gansel et al., 2001; Nazoa et al., 2003). In contrast to the amino acid negative regulation, induction of nitrate transporters by photosynthesis products involves metabolites downstream of the hexokinase activity and requires the oxidative pentose phosphate pathway (Lejay et al., 2008). A putative signalling candidate would be glucose-6-P, already known to be an allosteric regulator of the key carbohydrate metabolic steps phosphoenolpyruvate carboxylase and sucrose phosphate synthase (Matsumura et al., 2002; Takahashi-Terada et al., 2005). The glucose-6-P level was found to be well correlated with the transcript levels of NRT1.1 and NRT2.1 in roots, but further work is needed to assign a regulatory function of this metabolite for nitrate uptake. Interestingly, this new regulation mode also concerns members of the and SO42- transporter family, thus illustrating a co-ordinated regulation of different nutrient uptake (Lejay et al., 2008).
Next to these operators, means of regulating N nutrition acting either at the transcriptional or post-translational levels of target gene products have been characterized. They include protein kinases, transcription factors, RING-type ubiquitin ligase or microRNAs (for reviews see Gojon et al., 2009; Miller et al., 2009; Castaings et al., 2011). Ordering this complex web of regulating components into a primary master regulator or secondary developmental regulatory loops around nitrate pools is a major challenge.
Other parameters of the environment also affect the nitrate journey into a plant. For example, cadmium stress was shown to induce the expression of the xylem nitrate unloading transporter AtNRT1.8 (Li et al., 2010) and to repress the xylem nitrate loading system AtNRT1.5 (Lin et al., 2008). This response illustrates a strategy of nitrate reallocation under Cd2+ stress aiming at maintaining nitrate in roots. In plants experiencing a prolonged potassium deficiency, nitrate absorption is reduced by reversibly repressing the expression of AtNRT2.1, AtNRT2.3, and AtNRT2.6 (Armengaud et al., 2004). In these conditions, the concomitant depletion of root nitrate content, together with the stimulation of nitrogen assimilatory pathway (NR, GS, GOGAT), represents an original situation where nitrogen absorption and assimilation are uncoupled (Armengaud et al., 2009). The interaction between water and nitrate are already known and molecular (Wang et al., 2001) and genetic interactions (Loudet et al., 2003) have been investigated. Interestingly, AtNRT1.1 is expressed and functions in Arabidopsis guard cells. In the absence of NRT1.1 protein, stomatal opening and transpiration rates are reduced, leading to enhanced drought tolerance when compared with the wild type (Guo et al., 2003).
During the last decade, huge efforts have been devoted to the molecular dissection of nitrate fluxes through the whole plant. A very complex picture is now emerging (Figs 3, 4), combining the action of members of at least four gene families, namely NRT1, NRT2, NAR, and CLC. Given that the specific functions of 12 NRT1 and NRT2 genes are at present described in Arabidopsis (Tsay et al., 2007; this review), there are still 48 genes whose roles in planta remain to be elucidated. In addition, we are just at the beginning of understanding how these different proteins act in a co-ordinated manner, particularly when fluxes through the plasma membrane, tonoplast or chloroplast envelopes take place in the same cell. Finally, even if Arabidopsis constitutes a very good model to study this complex biological process, the function of orthologous genes in crops is yet to be elucidated. However, efforts must be emphasized particularly in species characterized, for example, by different nitrogen metabolisms such as rice (paddy environment). For example, in rice, one NRT1 gene has been functionally characterized (Lin et al., 2000) while the functions of two NAR2 like (Araki and Hasgawa, 2006) and four NRT2 like genes (Cai et al., 2008) are still under investigation. Symbiotic interactions, bacterial or mycorrhizal, may also have pronounced effects on the functions and regulations of nitrogen transport. Recent studies on mycorrhizal interactions have highlighted the roles of newly identified key transporters, in the fungi or in the host plants as well (Chalot et al., 2006, Javot et al., 2007).
P Armengaud and M Jossier are supported by a French ANR program (Nitrapool, ANR08-Blan-008). We thank A Krapp for critical reading of our manuscript.