Hormonal control of nitrogen acquisition: roles of auxin, abscisic acid, and cytokinin

Abstract

Nitrogen is the mineral nutrient that often limits plant growth and development. In response to changes in nitrogen supply, plants display elaborate responses at both physiological and morphological levels to adjust their growth and development. Because higher plants consist of multiple organs with different functions and nutritional requirements, they rely on local and long-distance signalling pathways to coordinate the responses at the whole-plant level. Phytohormones have been considered as signalling substances of such pathways. Amongst phytohormones, abscisic acid, auxin, and cytokinins have been closely linked to nitrogen signalling. Recent evidence has provided some insights into how nitrogen and the phytohormone signals are integrated to bring about changes in physiology and morphology. In this review, the evidence is summarized, mostly focusing on examples related to nitrogen acquisition.

Abscisic acid, auxin, cytokinin, nitrate, nitrogen, NRT, phytohormone

Introduction

Nitrogen, an essential macronutrient for plants, is absorbed from the soil either in inorganic forms such as nitrate or ammonium, or in organic forms, mostly as free amino acids. In aerobic soils, nitrate is the major form, but its availability can fluctuate dramatically both spatially and temporally, depending on leaching and microbial activity (Vance, 2001; Miller and Cramer, 2004). Therefore, plants mount elaborate physiological and morphological responses to balance the amount of nitrogen acquired from soil with what is needed for growth and development. These responses include regulation of uptake capacity (Gazzarrini et al., 1999; Lejay et al., 1999; Gojon et al., 2009), adjustment of shoot/root growth balance (Walch-Liu et al., 2005), and changes in root architecture (Walch-Liu et al., 2006; Zhang et al., 2007; Forde and Walch-Liu, 2009). Because a higher plant body is made up of multiple organs with different functions and nutritional requirements, these responses must be coordinated at the whole-plant level, and therefore both local and long-distance signalling are required for the communication of nutrient status between organs. Nitrate, amino acids, sugars, and phytohormones have been implicated in this signalling (Forde, 2002; Oka-Kira and Kawaguchi, 2006; Sakakibara et al., 2006).

Phytohormones were originally defined as a group of naturally occurring organic substances which influence growth and development at low concentrations (Davies, 2004). In addition to their basic roles in growth and development, phytohormones have been linked to various aspects of environmental responses, such as light, temperature, salt, drought, pathogen, and nutrient responses. Recently, much progress has been made in our understanding of how plants use phytohormones to integrate these environmental signals with endogenous growth and developmental programmes (Zhu, 2002; Halliday et al., 2009; Kazan and Manners, 2009; Patel and Franklin, 2009). It has been proposed that abscisic acid (ABA), auxin, and cytokinins (CKs) act to coordinate demand and acquisition of nitrogen (Signora et al., 2001; Wilkinson and Davies, 2002; Walch-Liu et al., 2006; Argueso et al., 2009).

Regulation of nitrogen acquisition involves modulation of nitrate uptake systems, and proliferation of lateral roots (Zhang and Forde, 2000; Forde and Walch-Liu, 2009). In general, nitrate uptake systems consist of low- and high-affinity nitrate transporters encoded by NRT1 and NRT2 family genes, respectively (Miller et al., 2007; Tsay et al., 2007; Gojon et al., 2009). The expression of NRT genes is regulated by numerous signals. For instance, AtNRT2.1, which constitutes a major component of the high-affinity nitrate transport system, is induced by nitrate and sugars, and repressed by nitrogen assimilation products and CKs (Filleur and Daniel-Vedele, 1999; Lejay et al., 1999; Zhuo et al., 1999; Brenner et al., 2005). Lateral root outgrowth is a complex developmental process regulated by various signals, including nitrate, nitrogen assimilation products, ABA, auxin, and CKs (reviewed in Forde, 2002; Walch-Liu et al., 2006; Zhang et al., 2007; Fukaki and Tasaka, 2009). Here, the roles of phytohormones in the regulation of nitrate uptake systems and lateral root proliferation in response to changes in nitrogen availability are discussed, together with new data and recent progress.

Cytokinins

CKs are a class of phytohormones implicated in many aspects of plant growth and development (Mok and Mok, 2001; Sakakibara, 2006), including nitrogen signalling. A CK–nitrogen link has been indicated by findings that nitrogen supply and CK content are closely correlated in barley (Samuelson and Larsson, 1993), tobacco (Singh et al., 1992), and Urtica dioica (Wagner and Beck, 1993), and that exogenous treatment with CKs can partially overcome growth-limiting effects caused by low nitrogen supply in Plantago major (Kuiper, 1988). Nitrogen supplementation causes an increase in CK content in xylem sap (also in roots and shoots) of maize (Takei et al., 2001), indicating that CKs function as a root-to-shoot long-distance signal of nitrogen supplement (Sakakibara et al., 2006). A similar association has been reported in Arabidopsis (Takei et al., 2004). Furthermore, it was detected that Arabidopsis seedlings grown on high concentrations of nitrate (HN, 10 mM) contain higher levels of CKs than those grown on low nitrate (LN, 0.1 mM; Table 1, and Supplementary Table S1 available at JXB online), implying that CKs are not just a nitrogen supplement signal but are also a nitrogen status signal. Recent evidence consistently indicates that CKs also act as a local signal or as a shoot-to-root long-distance signal (Miyawaki et al., 2004; Matsumoto-Kitano et al., 2008). Because the role of CKs as a root-to-shoot long-distance signal has been well reviewed (Sakakibara et al., 2006; Hirose et al., 2008; Kudo et al., 2010), this section focuses mainly on CK function as a local signal or as a shoot-to-root long-distance signal.

Table 1.

Auxin, ABA, and CK contentsa in shoots and roots of seedlings grown in HN and LNb

Phytohormones (pmol gFW−1)	Shoot			Root
	HN	LN		HN	LN
tZ-type CKc	17.13±3.46	2.78±0.56	dec***g	45.79±3.33	10.38±1.59	dec***
iP-type CKd	15.59±0.51	5.57±1.68	dec***	10.27±0.8	6.67±0.7	dec***
IAAe	2359±149	2327±240		3466±936	4911±585	inc*
ABAf	8.14±1.67	7.83±2.21		16.26±7.28	12.04±2.37

a

Mean value from four replicate samples with SD.

b

Wild-type seedlings were grown on high nitrate plates (HN, 10 mM) or low nitrate plates (LN, 0.1 mM) for 3 d as described in the legend of Fig. 2. Shoots and roots were harvested separately, and phytohormone analyses were conducted as described by Kojima et al. (2009). The full list of phytohormonal compounds and the quantification results are provided in Supplementary Table S1 at JXB online.

c

Sum of trans-zeatin (tZ), tZ riboside (tZR), and tZR 5'-phosphates. See Supplementary Table S1 at JXB online for the quantification result of each compound.

d

Sum of N⁶-(Δ²-isopentenyl)adenine (iP), iP riboside (iPR), and iPR 5'-phosphates. See Supplementary Table S1 at JXB online for the quantification result of each compound.

e

Indole-3-acetic acid.

f

Abscisic acid.

g

Asterisks indicate that the contents of the compound were significantly increased (inc) or decreased (dec) in LN compared with HN (*P <0.05, ***P <0.001, Student's t-test).

Regulation of CK biosynthesis by nitrogen

Until the identification of genes encoding adenosine phosphate-isopentenyltransferase (IPT), which catalyses the initial step of CK biosynthesis, it was believed that CKs are synthesized in roots (Letham, 1994). In Arabidopsis, IPT is encoded by seven genes that are differentially expressed in various tissues, indicating that CK production is not confined to roots (Miyawaki et al., 2004; Takei et al., 2004). Among these seven genes, AtIPT3 is nitrate inducible. Accumulation of CKs was greatly attenuated in an atipt3 mutant, indicating that AtIPT3 is a key determinant of nitrate-dependent CK biosynthesis (Miyawaki et al., 2004; Takei et al., 2004). Interestingly, nitrate-inducible expression of AtIPT3 was also observed in detached shoots (Miyawaki et al., 2004). Similarly, nitrogen supplementation induces CK accumulation in detached sunflower and tobacco leaves (Salama and Wareing, 1979; Singh et al., 1992). Microarray analyses have shown that the nitrate-inducible expression of AtIPT3 is partly mediated by NRT1.1/CHL1 (NRT1.1), a protein which functions as a dual-affinity nitrate transporter and nitrate sensor (Liu et al., 1999; Ho et al., 2009; Wang et al., 2009). AtIPT3 is expressed in phloem throughout the plant (Miyawaki et al., 2004; Takei et al., 2004), and this expression pattern overlaps with that of NRT1.1 in roots (Guo et al., 2001). However, in shoots, NRT1.1 expression is detected only in young leaves (Guo et al., 2001). Therefore, whether or not NRT1.1 mediates nitrate-inducible AtIPT3 expression in shoots is an open question. Given that AtIPT3 is expressed in phloem, it is likely that CKs synthesized by AtIPT3 in shoots function as a shoot-to-root long-distance signal of shoot nitrate availability. In this context, it has been shown that xylem sap predominantly contains trans-zeatin (tZ)-type CKs, and phloem sap mostly contains N⁶-(Δ²-isopentenyl)adenine (iP)-type and cis-zeatin (cZ)-type CKs (Hirose et al., 2008). Thus, either the iP- or cZ-type CKs, or possibly both, could be the shoot-to-root long-distance signal in Arabidopsis. Grafting experiments using a higher order atipt mutant (atipt1;3;5;7) have provided unequivocal evidence that iP-type CKs are translocated from the shoot to the root (Matsumoto-Kitano et al., 2008). The atipt1;3;5;7 mutant is characterized by extremely low iP- and tZ-type cytokinin levels, retarded shoot growth, and enhanced lateral root outgrowth. When a wild-type shoot was grafted onto the atipt1;3;5;7 mutant root, the normal growth phenotype and levels of iP-type CKs were restored in the mutant root, indicating that iP-type CKs translocated from the shoot are biologically functional (Matsumoto-Kitano et al., 2008). Notably, the expression of AtIPT3 is also regulated by iron, phosphate, and sulphate availability, both in shoots and in roots (Hirose et al., 2008; Seguela et al., 2008). It could be that AtIPT3 functions as an integrator of nutrient availability signals.

Regulation of nitrogen uptake by CKs

CKs produced locally within the root, or translocated from the shoot may signal that there is sufficient nitrogen present. In this regard, one of the proposed roles of CKs is negative regulation of nitrogen uptake-related genes. Microarray analyses have shown that exogenous application of CK represses two AtNRT2 genes (AtNRT2.1 and AtNRT2.3), three ammonium transporter genes, three amino acid transporter genes, and a urea transporter gene in Arabidopsis (Fig. 1; Brenner et al., 2005; Kiba et al., 2005; Sakakibara et al., 2006; Yokoyama et al., 2007). Because the expression levels of AtNRT1 and AtNRT2 correlate well with low- and high-affinity nitrate transport activity, respectively (Forde, 2000; Okamoto et al., 2003), it is conceivable that CK repression of the AtNRT genes results in a reduction in nitrate uptake activity. To form a more complete picture of CK regulation, the effect of CKs, along with other phytohormones, on each of the AtNRT genes (AtNRT2.1–AtNRT2.7, AtNRT1.1–AtNRT1.7, and AtNAR2.1) was analysed by quantitative PCR (qPCR). Under the experimental conditions used, it was found that CKs repress all the root-type AtNRT genes (R/L ratio in log₂ >1; Fig. 2). The repression was observed under both HN (Fig. 2A) and LN conditions (Fig. 2B), indicating that the CK effect is independent of the nitrogen status of plants. On the other hand, no such consistent repression was observed with other phytohormones, implying that the effect is CK specific (Fig. 2). Amongst the root-type genes, AtNRT2.1, AtNRT2.2, AtNRT1.1, and AtNRT1.5 have been characterized and shown to act as major components of the nitrate uptake and xylem loading system (Liu et al., 1999; Orsel et al., 2004; Li et al., 2007; Lin et al., 2008). These results support the hypothesis that CKs act as a satiety signal of nitrogen to inhibit nitrate uptake in the root. Similarly, it has been reported that CKs negatively regulate other nutrient acquisition-related genes in Arabidopsis, such as high-affinity phosphate transporter genes (Pht1;2 and Pht1;4; Martin et al., 2000; Sakakibara et al., 2006), high-affinity sulphate transporter genes (SULTR1;2 and SULTR1;4; Maruyama-Nakashita et al., 2004), and iron deprivation-inducible genes (IRT1, FRO2, and FIT; Seguela et al., 2008). Similar regulation was also reported in rice (Hirose et al., 2007), indicating that CK regulation of nutrient acquisition-related genes is a common mechanism in both dicots and monocots. In Arabidopsis, these repression effects have been shown to depend on the CK receptors AHK4/CRE1/WOL (CRE1) and/or AHK3, suggesting the involvement of CK signal transduction mediated by the His–Asp phosphorelay (Franco-Zorrilla et al., 2002; Maruyama-Nakashita et al., 2004; Seguela et al., 2008). As expected, any CK effect on the AtNRT genes is also dependent on the canonical CK signal transduction system (Fig. 3). The cre1 ahk3 double mutant (Higuchi et al., 2004) showed reduced sensitivity in the CK repression (Fig. 3A) and CK induction (Fig. 3B) of AtNRT genes. In contrast, the hextuple mutant (arr3,4,5,6,8,9; To et al., 2004) of negative regulators of the CK signalling system displayed hypersensitivity (Fig. 3A, B). It has been reported that CK-repressive nutrient acquisition-related genes are largely expressed in root epidermal or cortical cells. Thus, it is possible that CKs specifically target those cell types. In the case of iron starvation-inducible genes, CKs repress IRT1, FRO2, and FIT, all of which are expressed in epidermal cells. In contrast, vascular tissue-expressed AtNRAMP3 and AtNRAMP4 are not affected by CKs (Seguela et al., 2008), supporting the idea that the site of CK action is epidermal or cortical. However, this assumption may be an oversimplification, because it was found that AtNRT1.5, which is expressed in pericycle cells, is repressed by CKs (Fig. 3A; Lin et al., 2008), whereas epidermal cell-expressed AtNRT1.2 is not (Fig. 3C; Huang et al., 1999).

Fig. 1.

The effect of phytohormones on the expression of nitrogen uptake-related genes in Arabidopsis. Microarray data were obtained from the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/), and fold change values were calculated between phytohormone-treated and control data. Microarray data for abscisic acid (GSE7112), auxin (GSE1110), brassinosteroid (GSE862), cytokinins (GSE5698, GSE1766, and GSE20232), ethylene (GSE5174), gibberellin (GSE20223), and jasmonic acid (GSE21762) treatment were analysed. Among nitrogen uptake-related genes listed in Supplementary Table S3 at JXB online, genes with a >2-fold difference with statistical significance (P <0.05) are presented. Up-regulated (fold change >2) genes and down-regulated (fold change <0.5) genes are indicated with ‘U’ in orange cells and ‘D’ in blue cells, respectively. The Arabidopsis ATH1 Genome Array is unable to distinguish AtAMT1;3 from AtAMT1;5 (indicated by an asterisk). Numbers in parentheses indicate GEO data sets: 1, GSE5698 (Kiba et al., 2005); 2, GSE1766 (Brenner et al., 2005); 3, GSE20232 (Yokoyama et al., 2007); 4, GSE1110 (Redman et al., 2004); 5, GSE3350 (Vanneste et al., 2005); 6, GSE863 (Nemhauser et al., 2004); 7, GSE5174 (Olmedo et al., 2006).

Fig. 2.

The effect of phytohormones on the expression profiles of AtNRT genes in roots. (A) Transcript level of each AtNRT in phytohormone-treated seedling roots grown on high nitrate plates (HN, 10 mM) relative to that in non-treated seedlings. (B) Transcript level of each AtNRT in phytohormone-treated seedling roots grown on low nitrate plates (LN, 0.1 mM) relative to that in non-treated seedlings. Wild-type seedlings were grown vertically on HN plates (modified from Fujiwara et al., 1992) for 8 d under the following conditions: 12 h light/12 h dark cycles (80 m⁻² s⁻¹) at 22 °C. Then, the seedlings were treated with 10⁻⁶ M phytohormones for 3 d on HN or LN plates. Roots were harvested and the expression pattern of each AtNRT gene was analysed by qPCR using specific primer sets (Supplementary Table S2 at JXB online). A root/shoot expression ratio (R/L ratio) was calculated using the expression level of each gene in shoots and roots of non-treated seedlings grown under HN or LN conditions. Based on the R/L ratio, all AtNRT genes were grouped into four categories: root-type genes (R/L ratio in log₂ >1), shoot-type genes (R/L ratio in log₂ < –1), whole plant-type genes (R/L ratio in log₂ between –1 and 1), and genes that were not detected in roots. Plant hormones tested were trans-zeatin (tZ), indole-3-acetic acid (IAA), abscisic acid (ABA), gibberellin 3 (GA), 1-amino-1-cyclopropane-carboxylic acid (ACC), brassinolide (BL), salicylic acid (SA), and jasmonic acid (JA). Each experiment was performed twice and mean values are shown on a log₂ scale. Asterisks indicate that the values are of no statistical significance because the expression level in one of the samples was below the detection limit. A colour code is used to visualize data, and samples that were expressed below the detection limit are shown in grey (ND).

Fig. 3.

CKs mediate both induction and repression of AtNRT genes via AHK4/CRE1/WOL- and/or AHK3-dependent His–Asp phosphorelay. Dose-dependent effect of CK on (A) CK-repressive AtNRT genes, (B) CK-inducible AtNRT genes, and (C) control genes in wild-type, cre1-12 ahk3-3, and arr3,4,5,6,8,9 roots. Seedlings were grown for 8 d on HN plates, and then treated on LN plates supplemented with trans-zeatin for 3 d, as indicated. Roots were harvested and transcript levels were determined by qPCR using specific primer sets (Supplementary Data at JXB online). All quantifications were standardized with AtACT8 transcript levels as an internal standard. Relative values indicate comparison with the transcript level of dimethylsulphoxide (DMSO)-treated roots. Error bars represent standard deviations of three technical replicates. The experiment was performed twice with similar results.

Fig. 4.

Schematic representation of the interaction between nitrogen and phytohormones (abscisic acid, auxin, and cytokinin) in the regulation of nitrogen acquisition. Abscisic acid (ABA), auxin (AUX), and cytokinin (CK) produced locally, or translocated from shoots, regulate nitrogen (N) acquisition events. N availability and/or nitrate (NO₃^–) signal interacts with CK biosynthesis (AtIPT3), cell-to-cell auxin transport (NRT1.1), auxin perception (AFB3), and auxin signal transduction (ARF8/miR167a) to regulate N acquisition. NRT1.1/CHL1 has dual activity, namely as an AUX transporter and NO₃^– sensor. Arrows and blunted lines designate positive and inhibitory interactions, respectively. Solid lines represent defined interactions; dashed lines indicate presumed interactions.

Besides this negative effect, CKs also have a positive regulatory effect on some AtNRT genes. It was found that most shoot-expressed AtNRTs (R/L ratio in log₂ less than –1; Fig. 2) are up-regulated by CKs under both HN and LN conditions (Fig. 2). Recent studies have assigned physiological functions to the shoot-type genes AtNRT2.7, AtNRT1.4, and AtNRT1.7 in aerial organs (Chiu et al., 2004; Chopin et al., 2007; Fan et al., 2009). Considering the evidence that transcript abundance of these shoot-type genes in roots is much lower than that of root-type genes (data not shown), it is unlikely that shoot-type AtNRT genes play a major role in nitrate uptake. Although clarification of the function of shoot-type AtNRT genes in roots awaits further analysis, a plausible explanation is that shoot-type AtNRT genes are similarly induced in shoots by CKs and enhance nitrate distribution and translocation.

A possible role for CKs in the regulation of root architecture in response to nitrogen

CKs also have a regulatory role in root architecture development. Many reports describe the inhibitory effects of exogenous CKs on lateral root formation (Wightman et al., 1980; Laplaze et al., 2007). In contrast, mutants and transgenic plants with reduced CK levels, such as higher order atipt mutants (Miyawaki et al., 2006) and transgenic plants overexpressing the gene encoding the CK-degrading enzyme CKX (Werner et al., 2003), or decreased CK sensitivity, such as the higher order signalling mutants ahk (Higuchi et al., 2004), exhibit enhanced root growth and branching. CKs act at both initiation and organization of the lateral root primordium (LRP), most probably through perturbation of the auxin gradient, to inhibit LRP formation (Laplaze et al., 2007). In general, a uniformly high nitrogen supply suppresses root branching, while nitrogen limitation accelerates root growth and root branching (Linkohr et al., 2002; Tranbarger et al., 2003). Given that nitrogen status and CK content are closely correlated (Table 1, and Supplementary Table S1 at JXB online), it is very likely that CKs have a role in regulating root architecture in response to nitrogen availability. However, there is as yet no solid evidence to support this. Now that a number of CK signalling and biosynthesis mutants are available, questions about the CK regulation of root architecture development can be answered.

Abscisic acid

ABA is generally known as a stress hormone involved in abiotic and biotic stress responses. Although there is considerable evidence linking ABA levels and nitrogen status in several plant species (Radin et al., 1982; Peuke et al., 1994; Brewitz et al., 1995; Wilkinson and Davies, 2002), there is not necessarily a consistent correlation between the two. For example, in Arabidopsis, there is no statistically significant difference in ABA contents between seedlings grown under the HN and LN conditions used here (Table 1, and Supplementary Table S1 at JXB online). Thus, whether changes in ABA content are relevant to nitrogen signalling is still unclear, but involvement of ABA in nitrogen signalling is becoming increasingly evident.

Several reports provide genetic evidence for the involvement of ABA in lateral root development in response to high nitrate supply in Arabidopsis. Signora et al. (2001) clearly showed that ABA-insensitive mutants (abi4-1, abi4-2, and abi5-1) and ABA-deficient mutants (aba1-1, aba2-3, aba2-4, and aba3-2) are less sensitive to the inhibitory effects of high nitrate. A set of mutants identified based on their ability to produce lateral roots in the presence of ABA (labi mutants) shows reduced sensitivity to the inhibitory effects of high nitrate (Zhang et al., 2007). Identification of the labi genes would provide a breakthrough toward understanding the mechanisms underlying this inhibition effect. A recent study in a Medicago truncatula latd mutant provides another line of evidence for a link between ABA and nitrogen signalling (Yendrek et al., 2010). The latd mutant is characterized by severe defects in root meristem maintenance and root growth, which is rescued by exogenous ABA application (Bright et al., 2005; Liang et al., 2007). Interestingly, the primary root growth of the latd mutant is insensitive to nitrate, and the LATD gene encodes a transporter belonging to the NRT1 (PTR) family (Yendrek et al., 2010). However, whether LATD acts as a nitrate transporter/sensor remains to be shown.

Auxin

Auxin has long been a candidate for mediating nitrogen signals from shoot to root because auxin is transported basipetally, and enhances lateral root initiation and development (Forde, 2002; Fukaki and Tasaka, 2009). Auxin contents in phloem sap and roots are lower in maize supplied with high doses of nitrate, and this reduction is correlated with reduced root growth (Tian et al., 2008). Similarly, transfer from high nitrate media to low nitrate media has been shown to increase auxin contents in roots, which is followed by lateral root outgrowth in Arabidopsis (Walch-Liu et al., 2006). Furthermore, it has been confirmed that Arabidopsis seedlings grown in LN conditions contain higher levels of root auxin than seedlings grown in HN conditions (Table 1, and Supplementary Table S1 at JXB online), indicating that dicots and monocots share a common system to regulate auxin levels in the root depending on the nitrogen status of the plant. However, it was also reported that the inhibitory effect of high nitrate on lateral root growth is not alleviated by exogenous application of auxin, indicating that the auxin content is not the only factor regulating lateral root development (Zhang et al., 2007).

Recent progress in understanding auxin action indicates that, in addition to auxin levels in the tissue, a concentration gradient and the differential sensitivity of various cell types are the driving force of auxin-regulated growth and development (Overvoorde et al., 2010; Zazimalova et al., 2010). The auxin gradient is established by cell-to-cell polar transport, and differential sensitivity is accomplished by modulation of signalling components (Overvoorde et al., 2010; Zazimalova et al., 2010). Several lines of recent evidence suggest that nitrogen signalling is mediated by these same, or similar mechanisms (Gifford et al., 2008; Krouk et al., 2010; Vidal et al., 2010). Using a cell sorting technique, Gifford et al. (2008) identified a nitrogen-inducible Auxin Response Factor (ARF8) which is expressed in pericycle cells, and whose transcript level is repressed by microRNA167a (miR167a). Although transcription of ARF8 is not regulated by nitrogen, miR167a levels are under the control of glutamine or some downstream metabolite (Gifford et al., 2008). Hence, in the presence of glutamine or a downstream metabolite, miR167a levels in pericycle cells are down-regulated, thus permitting ARF8 transcript to accumulate in the cell. The arf8 mutant and MIR167a-overexpressing seedlings fail to respond to nitrogen treatment by increasing the ratio of initiating/emerging lateral roots, providing genetic evidence that the nitrogen signal regulates auxin signalling through the action of miR167a to control lateral root initiation (Gifford et al., 2008). Recently, the auxin receptor gene AFB3 was found to be nitrate inducible (Vidal et al., 2010). The induction was also observed in a nitrate reductase (NR)-null mutant, suggesting that it is triggered by nitrate itself. Analysis of the afb3 mutant revealed that the mutant is insensitive to nitrate in the regulation of primary root growth and lateral root density, indicating that nitrate signal regulates root architecture through AFB3, possibly by modulating auxin signalling (Vidal et al., 2010). Surprisingly, Krouk et al. (2010) provided another link between nitrate and auxin by showing that NRT1.1/CHL1, which is known as a dual-affinity nitrate transporter and nitrate sensor (Liu et al., 1999; Ho et al., 2009), also facilitates cell-to-cell auxin transport. Detailed analyses of an nrt1.1 (chl1) mutant revealed that chl1 accumulates auxin in lateral root primordia, which led the authors to hypothesize that NRT1.1 transports auxin. Data obtained from transport assays in heterologous systems and in planta were consistent with this hypothesis. Furthermore, nitrate was shown to be an inhibitor of the auxin transport activity (Krouk et al., 2010). Localization studies demonstrated that NRT1.1 is present within the anticlinal membranes of the outermost cell layer of lateral roots, indicating that NRT1.1 facilitates auxin movement out of the root tip. Thus, apparently under low nitrate conditions, NRT1.1 prevents auxin accumulation in the root tip, resulting in inhibition of lateral root growth. In contrast, under high nitrate conditions the auxin transport activity of NRT1.1 is inhibited by nitrate, resulting in an accumulation of auxin within the root tip, leading to lateral root outgrowth (Krouk et al., 2010).

Conclusions and perspectives

It has long been proposed that there is an interaction between nitrogen signalling and phytohormone activity. Recent genetic studies have provided compelling evidence that ABA, auxin, and CKs are involved in nitrogen signalling (Fig. 4). In addition, it was found that brassinosteroid (BR) up-regulates a large number of AtNRT genes (Fig. 2). Although it has been reported that there is a shared signalling pathway for BR and auxin (Goda et al., 2004; Nemhauser et al., 2004), auxin or other phytohormones did not induce AtNRT genes in any way similar to BR, implying that the effect is BR specific. Because BR promotes lateral root growth and development (Mori et al., 2002; Bao et al., 2004), it is tempting to speculate that BR enhances nitrogen acquisition, in opposition to CK action. However, both the mechanism and physiological relevance of this regulation remain to be elucidated. Given the apparent involvement of multiple phytohormones in nitrogen signalling, one future challenge will be to understand how phytohormones interact to convey the nitrogen signal.

A growing body of evidence shows that phytohormones interact not only with nitrogen but also with other nutrients. For instance, CKs interact with iron (Seguela et al., 2008), sulphur (Maruyama-Nakashita et al., 2004), and phosphorus signalling (Wagner and Beck, 1993; Franco-Zorrilla et al., 2002), and auxin interacts with phosphorus signalling (Nacry et al., 2005), indicating that phytohormones are not the specific (direct) signal of a certain nutrient itself. Rather, they might be the translated (indirect) signals to coordinate growth and development with nutritional status within an organ and/or between organs, and act concurrently with the other nutrient-specific signals to trigger a response characteristic to each nutrient. Although this is an intriguing hypothesis, many questions remain to be answered. How is the nutritional status sensed and translated into phytohormone signals? In which cell, tissue, or organ does the sensing and translation occur? What is the nature of the nutrient-specific signal? How are the phytohormone and nutrient-specific signals transported to target sites? Where and how are the phytohormone signal and nutrient-specific signal perceived, and integrated to bring about a characteristic response to each nutrient? Now that classic biochemical, physiological, and genetic approaches can be combined with state-of-the-art ‘omics’ (e.g. highly sensitive and high-throughput phytohormonome analysis; Kojima et al., 2009) and systems approaches, it will not take long until the answers to these questions are known.

We would like to thank Dr T. Kakimoto for cre1-12 ahk3-3, and Dr J. J. Kieber for arr3,4,5,6,8,9 mutant seeds. Research in our laboratory is supported by the Ministry of Education, Culture, Sports, Science, and Technology and the Ministry of Agriculture, Forestry and Fisheries, Japan.