Molecular fundamentals of nitrogen uptake and transport in trees

Abstract

Nitrogen (N) is frequently a limiting factor for tree growth and development. Because N availability is extremely low in forest soils, trees have evolved mechanisms to acquire and transport this essential nutrient along with biotic interactions to guarantee its strict economy. Here we review recent advances in the molecular basis of tree N nutrition. The molecular characteristics, regulation, and biological significance of membrane proteins involved in the uptake and transport of N are addressed. The regulation of N uptake and transport in mycorrhized roots and transcriptome-wide studies of N nutrition are also outlined. Finally, several areas of future research are suggested.

Ammonium transporters, forest trees, nitrogen economy, nitrate transporters, tree nutrition.

Introduction

Nitrogen (N) is a principal constituent of proteins, nucleic acids, and many other essential biomolecules of living organisms. The homeostasis of N at the whole-organism level requires the uptake, biosynthesis, degradation, and transport of nitrogenous compounds. A general characteristic of complex eukaryotic organisms is the differentiation of organs integrated by a variety of different tissues and cell types with specific functions. Moreover, at the cellular level, metabolic functions are organized and distributed in subcellular compartments. The transport of nitrogenous compounds between different organs, cell types, and subcellular compartments is of particular relevance for the growth and development of living organisms.

In plants, N is an important macronutrient and is frequently a limiting factor for growth and development. Different forms of N in the soil can contribute to plant nutrition including nitrate (NO₃^–), ammonium (NH₄⁺), amino acids, and peptides (Xu et al., 2012). High- and low-affinity nitrogen transporters guarantee the acquisition of N from the soil under a wide range of N availability.

The molecular basis for N acquisition and transport in plants has recently been reviewed (Xu et al., 2012; Wang et al., 2012; Krapp, 2015). Here, we review recent advances in tree N nutrition. The molecular characteristics, regulation, and biological significance of membrane proteins involved in the uptake and transport of N are addressed. The regulation of N uptake and transport in mycorrhized roots and transcriptome-wide studies of N nutrition are also discussed.

N availability, uptake, and transport in forest ecosystems

N availability is extremely low in forest soils, and forest trees have evolved mechanisms to acquire and transport different forms of N, along with biotic interactions in the rhizosphere to guarantee the strict economy of this essential nutrient. For example, mycorrhizal symbiosis, beneficial associations with free-living fungi and bacteria in the rhizosphere, and endophytic bacteria increase the efficiency of N acquisition and metabolic assimilation. In addition, efficient N storage and N recycling processes are of paramount importance for the N economy of forest and fruit trees.

In boreal forest ecosystems, numerous factors such as low temperatures, low soil pH, and high residual content of lignin and other secondary plant products limit nitrification rates in the soil. As a result, NH₄⁺ is the predominant source of inorganic N available for tree nutrition (Lipson and Näsholm, 2001; Rennenberg et al., 2009; Cañas et al., 2016). In contrast, nitrate nutrition is predominant in riparian forest ecosystems, particularly those located near areas of intensive agriculture where high levels of NO₃^– are present in surface and ground waters (Rennenberg et al., 2009). Physiological studies have demonstrated the presence of high-affinity transport systems (HATS) and low-affinity transport systems (LATS) for NO₃^– and NH₄⁺ in the roots of vascular plants (Forde, 2000; Howitt and Udvardi, 2000; von Wirén et al., 2000a; Glass et al., 2002). The HATS is a saturable system that operates under micromolar (<100 μM) concentrations of ammonium or nitrate supply and is regulated by the external concentration of these ions. Under millimolar concentrations of NO₃^– and NH₄⁺, LATS become operative. The dual kinetics of the HATS and LATS are very important in N assimilation in plants, as these systems ensure that root inorganic N acquisition from the soil is optimal.

Conifers from boreal forests have a marked preference for NH₄⁺ over NO₃^–, and most conifer species grow naturally on NH₄⁺-rich soils (Kronzucker et al., 1996; Cañas et al., 2016). Uptake of NH₄⁺ from soil is mediated by membrane proteins called ammonium transporters (AMTs) that are strategically located primarily in plant root cells. Plant AMTs are well-conserved proteins and exhibit a very similar tridimensional structure as homo- or heterotrimers that create a central pore allowing the transport either of NH₄⁺ or its neutral NH₃ form (Neuhäuser et al., 2014). In root cells, NH₄⁺ is assimilated into amino acids via the concerted action of the enzymes glutamine synthetase (GS) and glutamine oxoglutarate aminotransferase (GOGAT) (Cánovas et al., 2007; Masclaux-Daubresse et al., 2010).

Although the major form of inorganic N available in most forest ecosystems is NH₄⁺, organic N in the form of amino acids and peptides is quantitatively important and may represent a major source of N for tree nutrition in some forest ecosystems (Lipson and Näsholm, 2001; Näsholm et al., 2009). Recently, using novel microdialysis techniques, it was found that amino acids may account for ~80% of total N available for tree and mycorrhizal uptake (Inselsbacher and Näsholm, 2012). Excess N fertilization or deposition of atmospheric N in coniferous forests greatly increases the content of arginine in needles and wood, and it has been proposed that arginine content may reflect the N status of conifers (Nordin et al., 2001). Consistent with these findings, conifers prefer NH₄⁺ and arginine instead of NO₃^–, as N sources (Gruffman et al., 2014).

Amino acid transporters are encoded by a large gene family in plants (for recent reviews, see Pratelli and Pilot, 2014; Tegeder, 2014). However, the uptake of external amino acids by plant roots is apparently mediated by a limited number of these family members (Lee et al., 2007; Svennerstam et al., 2008, 2011). Only a limited number of amino acid transporters have been functionally characterized, mainly in the model plant Arabidopsis (Pratelli and Pilot, 2014).

Symbiotic interactions between tree roots and ectomycorrhizal fungi enhance the acquisition of inorganic and organic N in exchange for photosynthetically fixed carbon (Näsholm et al., 2013; Franklin et al., 2014). These interactions are extremely beneficial in forest ecosystems, and it has been suggested that ectomycorrhizal fungi regulate N transfer to tree roots depending on the relative availability of N in the forest soil (Näsholm et al., 2013). Mycorrhizal fungi synthesize arginine which is translocated to the mycelium and degraded to ammonium that is assimilated by roots, incorporating N into plant metabolism (Chalot et al., 2006). Nitrogen transporters have been well characterized in ectomycorrhizal fungi, but much less is known about the transporters involved in the transfer of N from fungi to root cells.

Molecular characteristics and kinetics of ammonium transporters

The AMTs of higher plants are involved in the incorporation of external NH₄⁺ from the soil and internal transport of ammonium between different organs and cellular compartments (Ludewig et al., 2007). These proteins belong to the AMT/MEP/Rh protein superfamily including bacterial, fungal, and human homologs (Loqué and von Wirén, 2004). Ammonium transporters possess 10–12 transmembrane (TM) domains and are assembled into homo- or heterotrimers (Ludewig et al., 2003). The AMTs in plants are subdivided into two subfamilies: AMT1 and AMT2. AMT1s are related to prokaryotic AMTs and were presumably inherited vertically (McDonald et al., 2012), whereas AMT2s probably arose in plants through horizontal gene transfer from an Archaea, with the potential implication of a gamma-proteobacterial intermediate host (McDonald et al., 2012; von Wittgenstein et al., 2014). In Arabidopsis, members of the AMT1 subfamily include genes with no intronic sequences (Yuan et al., 2007, 2009), exhibiting specific NH₄⁺ affinities and patterns of expression (Yuan et al., 2007), suggesting a specialization of function that has not been tested for other plant species. In contrast, the AMT2 subfamily members are membrane proteins with amino acid sequences similar to those of the yeast MEP transporters (Loqué and von Wirén, 2004). The molecular and kinetic properties of angiosperm AMT2s have been studied, and their expression levels are markedly higher than those of the AMT1s in the shoots and roots (Sohlenkamp et al., 2002; Suenaga et al., 2003).

Both types of ammonium transporters are present in all plant genomes sequenced to date, suggesting that both types are required for significant and differentiated roles in plant N metabolism. Recently, electrophysiological analyses of AMTs in the basal plant Marchantia polymorpha led to the proposal that all AMT1 transporters in land plants are electrogenic (McDonald and Ward, 2016). In contrast, the AMT2 transporters in plants included in the MEP clade are proposed to be electroneutral (Neuhäuser et al., 2014). Accordingly, AMT1 transporters could be involved in net NH₄⁺ uptake driven by negative membrane potential, whereas electroneutral transporters could be driven by the TM NH₄⁺ gradient (McDonald and Ward, 2016).

Ammonium uptake and AMTs in trees

Ammonium transporters have been widely studied in Arabidopsis and other herbaceous species of agronomic interest. However, much less is known about AMTs in tree species, notwithstanding the large number of trees that live or prefer to live in ammonium-rich soils. The long life spans of trees determine shifts and cyclical responses in nutrient uptake, influenced by availability, environmental conditions, and plant metabolic rates. For instance, diurnal differences in root ammonium uptake have been found between tree species. Thus, in beech (Fagus sylvatica), ammonium uptake from the soil in the growing season was substantially lower during the night than during the day, with a minimum at midnight and a maximum at noon (Gessler et al., 2002). In contrast, far less pronounced changes were observed in NH₄⁺ uptake by the roots of spruce (Picea abies). The absence of diurnal changes in the content of carbohydrates, organic acids, and total soluble non-protein N may explain the observed differences in NH₄⁺ uptake (Gessler et al., 2002).

Woody angiosperms

Ammonium transport systems have been characterized in a variety of fruit and forest trees including pear (Mota et al., 2011; Li et al., 2015), citrus (Cerezo et al., 2001; Camañes et al., 2009), and poplar (Min et al., 2000; Couturier et al., 2007).

The existence of LATS and HATS ammonium transporters has been reported in Citrus sinensis (Cerezo et al., 2001). The HATS were down-regulated when plants were supplied with N and up-regulated under N starvation. The activity of the LATS increased in response to NH₄⁺ availability. A cDNA encoding an AMT1 protein mediating HATS was characterized in Citrus sinensis, CitAMT1 (Camañes et al., 2007). Both HATS activity and CitAMT1 expression in roots were stimulated by light, and the authors proposed that the supply of photosynthates had an important role in the regulation of external NH₄⁺ import (Camañes et al., 2007). Citrus plants exhibited a higher capacity for NH₄⁺ uptake than NO₃^– uptake. HATS activity and the expression of CitAMT1 were also enhanced by NH₄⁺ (Camañes et al., 2009). Using ¹⁵N-labeled inorganic N, pear trees were also recently shown to absorb N preferentially as NH₄⁺ over NO₃^– (Mota et al., 2011). Expression analysis of five members of the AMT1 subfamily indicated that NH₄⁺ uptake was mediated by several AMT1s simultaneously. In Pyrus betulaefolia, a high-affinity transporter gene (PbAMT1.3) is specifically expressed in roots, where it is strongly influenced by NH₄⁺ availability, suggesting the high-affinity transporter could be involved in the uptake of external NH₄⁺ (Li et al., 2016). In contrast, PbAMT1.5 was preferentially expressed in leaves, where it exhibits diurnal rithms and is potentially associated with leaf NH₄⁺ metabolism and senescence (Li et al., 2016). An additional member of the pear AMT gene family (PbAMT1.1) is expressed and differentially regulated in roots and leaves, where it possibly plays a complementary role to PbAMT1.3 and PbAMT1.5 (Li et al., 2015).

The gene family of Populus trichocarpa AMTs consists of 16 members (Fig. 1), 7 of the AMT1 subfamily and 9 of the AMT2 subfamily (Couturier et al., 2007; Wu et al., 2015). The intron–exon organization is generally conserved in members of the AMT2 subfamily, while no introns have been found in the structure of poplar AMT1 genes, with the single exception of PtrAMT1.7, which has two exons and a single intron (Couturier et al., 2007; Wu et al., 2015). The family of AMT genes is much larger in poplar than in Arabidopsis, due to a recent whole-genome duplication in Salicaceae (Sterck et al., 2005). After this genome duplication, six pairs of duplicated AMT genes were retained (PtAMT1.1/1.3, PtAMT1.4/1.5, PtAMT1.2/1.7, PtAMT2.1/2.2, PtAMT3.2/3.1, and PtAMT4.1/4.3). The remaining members of the family lack their corresponding duplicates. Whether a higher number of sequences in this species, compared with herbaceous species, may be related to functional redundancy, or have special/additional functions, remains poorly understood. However, specific transcription profiles detected for duplicated genes point to putative specialized roles in planta (Wu et al., 2015).

Fig. 1.

Phylogenetic analysis of AMT1 genes in Populus trichocarpa and Pinus pinaster. Phylogenetic trees were constructed using the maximum parsimony (MP) method. The bootstrap consensus tree inferred from 1000 replicates was taken to represent the evolutionary history of the taxa analyzed (Felsenstein, 1985). Branches corresponding to partitions reproduced in <50% bootstrap replicates are collapsed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches (Felsenstein, 1985). The MP tree was obtained using the Subtree-Pruning-Regrafting (SPR) algorithm (Nei and Kumar, 2000) with search level 1 in which the initial trees were obtained by the random addition of sequences (100 replicates). For the AMT family, the analysis involved 33 amino acid sequences from Arabidopsis thaliana (At), Populus trichocarpa (Pt), and Pinus pinaster (Pp) including Escherichia coli (Ec) as an outgroup. The initial amino acid sequence alignment was conducted in MAFFT (Katoh et al., 2002); the remaining evolutionary analyses were conducted in MEGA7 (Kumar et al., 2016). (This figure is available in colour at JXB online.)

Interestingly, all duplicated genes of the GS family have been retained in the poplar genome (Castro-Rodríguez et al., 2011). The duplicated GS genes are expressed in the same cellular types and display similar seasonal and spatial gene expression patterns (Castro-Rodríguez et al., 2011, 2015). The co-ordination of NH₄⁺ uptake and its subsequent assimilation are critical for plant growth and development, and therefore it is interesting to explore the existence of potential mechanisms of co-ordination between these two processes. Putative co-expression profiles of AMTs with GS genes have been examined using the http://popgenie.org tool (Sjödin et al., 2009). These analyses have shown the existence of significant co-expression profiles between PtAMT1.3 and PtGS1.1 in mature leaves, between PtAMT3.1 and PtGS1.3 in stems, and between PtAMT3.2 and PtGS2 preferentially in leaves. Further analysis is necessary to determine whether these preliminary results have physiological relevance. Transgenic hybrid poplars overexpressing cytosolic GS have enhanced nitrogen use efficiency (NUE) at low and high levels of nitrate availability (Man et al., 2005; Castro-Rodríguez et al., 2016a). Increased N availability represses the expression of PtAMT3.1 (Potri.019G000800.1) that is highly expressed in senescent leaves (Couturier et al., 2007).

Gymnosperms

The kinetics of NH₄⁺ influx have been studied in conifers, and HATS and LATS for NH₄⁺ have been identified (Kronzucker et al., 1996). NH₄⁺ and NO₃^– uptake, and the expression of genes encoding NH₄⁺ and NO₃^– transporters, have been examined in roots of white spruce, larch, and Chinese fir (Alber et al., 2012; Meng et al., 2016a).

Recent developments in next-generation sequencing have improved the genomic resources available in conifers (Mackay et al., 2012; Canales et al., 2014). Taking advantage of these new resources, the molecular basis of NH₄⁺ uptake and transport in this important group of gymnosperms has recently been elucidated by the identification of 10 AMT family members in the transcriptome database of maritime pine (Pinus pinaster) (Castro-Rodríguez et al., 2016b; this work). The phylogenetic analysis of pine AMT sequences revealed that they encode AMT1 and AMT2 homologs of the membrane proteins found in angiosperms (Fig. 1). Members of the AMT1 subfamily in pine and poplar probably evolved from a common ancestor that is related to the ancient PpAMT1.2 and PtAMT1.6 genes. Interestingly, PpAMT1.2 displays biphasic kinetics, suggesting that this transporter may have a role in high- and low-affinity NH₄⁺ uptake, as proposed for the Arabidopsis AtAMT1.2 transporter that shows similar biphasic kinetics (Shelden et al., 2001). These data suggest that ancestral AMTs exhibited biphasic kinetics, and HATS and LATS evolved later in land plants. The existence of two AMT1 clades as well as several AMT2 clades was recently reported by von Wittgenstein et al. (2014).

Functional expression in yeast and Xenopus oocytes reveals that the family of AMT genes includes HATS (PpAMT1.1, PpAMT1.2, PpAMT1.3, and PpAMT2.3) and LATS (PpAMT2.1) transporters with different kinetics and with different capacities for NH₄⁺ transport. The expression of AMT genes is temporally and spatially regulated in pine seedlings and in 25-year-old trees showing differentiated patterns of transcript distribution and relative abundance (Castro-Rodríguez et al., 2016b). These findings suggest that family members play different roles in NH₄⁺ uptake and transport. This assumption has been further supported by expression analyses in specific tissue types from P. pinaster (Fig. 2) that were isolated by laser capture microdissection (Cañas et al., 2014; R.A. Cañas et al., unpublished results). Genes for the HATS PpAMT1.3 and PpAMT2.3 were preferentially expressed in roots and in response to N availability, with differential and complementary patterns. Interestingly, AMT2 expression was particularly high in the root tips (Alber et al., 2012; Castro-Rodríguez et al., 2016b) where NH₄⁺ uptake occurs at higher levels (Canales, 2011; Alber et al., 2012). Both expression analyses and immunolocalization studies conducted in pine roots suggest that AMT1 and AMT2 proteins play complementary roles in the uptake and transport of external NH₄⁺ (Castro-Rodríguez et al., 2016b). Taken together, these results improved our understanding of N acquisition and transport in coniferous trees.

Fig. 2.

Expression of AMT genes in different pine tissues. Distribution of transcripts for PpAMT1.1, PpAMT1.2, PpAMT1.3, and PpAMT2.3 in different tissues from 1-month-old seedlings of Pinus pinaster (R.A. Cañas et al., unpublished results).

Molecular characteristics and kinetics of nitrate transporters

The molecular basis of LATS and HATS comprises nitrate transporters of two major families: the NITRATE TRANSPORTER 1/PEPTIDE TRANSPORTER family (NPF) (Léran et al., 2014) and the NITRATE TRANSPORTER 2 family (NRT2/NRT3) (von Wittgenstein et al., 2014). However, some members of the chloride channel (CLC) and slow anion channel-associated homolog (SLAC/SLAH) families are also involved in the NO₃^– transport in plants (Krapp, 2015).

The NPF family consists of proton-driven symporters typically with 12 TM domains (Sun et al., 2014), including the essential motif E1X1X2E2[K/R] for proton coupling and active transport (Jørgensen et al., 2015). In plants, a variety of molecules can be transported by members of the NPF family, including nitrates, oligopeptides (di- and tri-peptides), amino acids, and hormones (auxins, abscisic acid, glucosinolates, and gibberellins). The different subfamilies of plant NPF proteins have been identified based on their sequence relationships rather than by the molecules that they transport because no correlation has been found between substrate selectivity and sequence homologies (Léran et al., 2014). Among the NPF transporters that can transport NO₃^–, only a few are involved in NO₃^– uptake from the soil. In Arabidopsis, nitrate uptake at high external concentration is mainly mediated by two NPF proteins, AtNPF6.2 and AtNPF6.3 (Krapp, 2015). AtNPF6.3 (NRT1.1) has a dual function, serving as a nitrate transporter and as a nitrate sensor to modulate the expression of nitrate-regulated genes (Ho et al., 2009; Krouk et al., 2010). The NRT2 transporters have no sequence homology with the NPF proteins, but they do have 12 TM domains with a proton-driven symport mechanism (Tsay et al., 2007). The NRT2 transporters function as a part of a two-component high-affinity nitrate uptake system (Okamoto et al., 2006). A second component identified is NRT3.1 (NAR2.1), which contains a single TM domain. The molecular role of NRT3.1 in the complex remains unknown.

Nitrate uptake and NRT in trees

Most research on NO₃^– uptake by trees has been done at the physiological level (Gruffman et al., 2014; Rennenberg and Dannenmann, 2015). Most reports have focused on the N source preferred by trees, in relation to biomass and xylem production (Omena-Garcia et al., 2015), or drought stress (Faustino et al., 2015), which are important topics in forestry. Trembling aspen (Populus tremuloides) exhibited a higher capacity for NO₃^– acquisition than for NH₄⁺, and the ability to acquire NO₃^– was mediated by the activity of saturable high affinity (HATS) and constitutive non-saturable low affinity (LATS) transporters (Min et al., 2000). A high-affinity nitrate transporter (similar to Arabidopsis AtNRT2.7, Potri.001G348300.1) was up-regulated in hybrid poplar overexpressing GS when grown under high NO₃^– conditions (Castro-Rodríguez et al., 2016a). Furthermore, the amino acid transporter PtAAP11 (Potri.008G036300.1) has an important role in N transfer during xylem differentiation (Couturier et al., 2010) and was also up-regulated in GS transgenics and wild-type poplar plants growing under high NO₃^– conditions (Castro-Rodríguez et al., 2016a). These findings strongly suggest that PtAAP11 could mediate the response to N availability and modify wood formation. Consistently, in silver birch, application of NO₃^– activates sucrose synthase and increases sucrose utilization, resulting in an increase in wood formation (Galibina et al., 2016). N uptake, assimilation, and accumulation of N compounds were enhanced in Grey poplar (Populus tremula×Populus alba) exposed to saline conditions when the plants were supplied with NO₃^– instead of NH₄⁺ (Ehlting et al., 2007). In contrast, salt-treated roots of P. simonii took up more NH₄⁺ than NO₃^– (Zhang et al., 2014). Consistently, most AMT genes were up-regulated under moderate salinity, particularly AMT1.2 and AMT1.4, whereas genes encoding nitrate transporters were down-regulated (Zhang et al., 2014). Taken together, these results reflect variability in the source of inorganic N taken up by trees, depending on the tree species and the environmental growth conditions; in contrast, most crop species prefer nitrate.

Woody angiosperms

The large gene family of P. trichocarpa nitrate transporters consists of 68 PtNPF and 11 PtNRT2/NRT3 genes (Bai et al., 2013). Expression analyses reveal tissue-specific expression patterns of PtNRT genes, suggesting different roles in poplar tree biology. For example, a member of the NRT2 subfamily (PtNRT2.4a) encoding HATS was induced in poplar roots following the application of low NO₃^– concentrations (Willman et al., 2014). In peach, two NRT2 genes (PpNrt2.1 and PpNrt2.2) specifically expressed in roots were identified and their expression was found to be up-regulated by NO₃^– and repressed by NH₄⁺ (Nakamura et al., 2007). A high-affinity nitrate transporter (MdNRT2.4) was expressed at high levels in the roots of apple trees under drought stress (Bassett et al., 2014). These results highlight the relevance of members of the NRT2 subfamily in the uptake of external NO₃^– from the soil.

Gymnosperms

Recent developments in conifer genomics have allowed the identification of at least 40 PpNPF and 8 PpNRT2/NTR3 genes in maritime pine (P. pinaster) (Fig. 3; Supplementary Table S1 at JXB online), one of the best characterized gymnosperm transcriptomes (Canales et al., 2014; Cañas et al., 2014). Phylogenetic analyses demonstrate the presence of all NPF subfamilies in maritime pine compared with poplar and Arabidopsis, supporting an early diversification of this transporter family (Bai et al., 2013; Léran et al., 2014; von Wittgenstein et al., 2014; Fig. 3A). The canonical number of TM domains for the NPF family members is 12; however, TM domain prediction analyses (TMHMM Server v. 2.0) showed a variable number of TM domains in the members of the NPF family, regardless of species (Supplementary Table S1). Interestingly, the modifications in the essential domain for the proton-driven symport E1X1X2E2[K/R] (Jørgensen et al., 2015) are conserved between the NPF subfamily members (Supplementary Table S1). This domain can be considered as one of the determinants for subfamily group. An excellent example is the NPF7 subfamily, where in all their members, regardless of species, the E1X1X2E2[K/R] domain has been replaced by the QGLAT domain and other derivatives. This is suggestive, since AtNPF7.3/AtNRT1.5 in Arabidopsis is involved in shoot K homeostasis; the action of this transporter under limited nitrate supply promotes the expression of K transporters (Drechsler et al., 2015; Meng et al., 2016b). It should be highlighted that the loss of this domain in several NPFs of poplar might imply the neo-functionalization of the duplicated genes.

Fig. 3.

Phylogenetic analysis of NPF and NRT2/NRT3 genes in Pinus pinaster. Phylogenic trees for NPF and NRT2 protein families are presented in (A) and (B), respectively. The phylogenetic trees were constructed using the maximum parsimony (MP) method. The bootstrap consensus tree inferred from 1000 replicates was taken to represent the evolutionary history of the taxa analyzed (Felsenstein, 1985). Branches corresponding to partitions reproduced in <50% bootstrap replicates are collapsed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches (Felsenstein, 1985). The MP tree was obtained using the Subtree-Pruning-Regrafting (SPR) algorithm (Nei and Kumar, 2000) with search level 1, in which the initial trees were obtained by the random addition of sequences (100 replicates). (A) For the NPF family, the analysis involved 178 amino acid sequences from Arabidopsis thaliana (At), Populus trichocarpa (Pt), and Pinus pinaster (Pp), including Gallus gallus (Gg) and Bos taurus (Bt) NPG SOCA sequences as outgroups (Supplementary Table S1). There were a total of 190 positions in the final data set. All positions containing gaps and missing data were eliminated. The different NPF subfamilies are highlighted in the tree by colored lines. The P. pinaster sequences are in bold. (B) For the NRT2/NRT3 family, the analysis involved 29 amino acid sequences from Arabidopsis thaliana (At), Populus trichocarpa (Pt), and Pinus pinaster (Pp) including a Cytophaga hutchinsonii (Ch) NNTA sequence as an outgroup (Supplementary Table S2). There were a total of 54 positions in the final data set. All positions containing gaps and missing data were eliminated. The different NRT2 subfamily is highlighted with blue lines and the NRT3 subfamily with red lines. The P. pinaster sequences are in bold. The initial amino acid sequence alignment was conducted in MAFFT (Katoh et al., 2002) and the remaining evolutionary analyses were conducted in MEGA7 (Kumar et al., 2016).

It is worth mentioning the relatively high number of NRT3 members in the maritime pine NRT2/NRT3 family (Fig. 3B; Supplementary Table S2). These findings suggest the possible involvement of these genes in the regulation of as yet unknown processes. However, potential roles can be proposed on the basis of gene expression (Fig. 4). PpNRT2.1 and PpNRT3.4, mainly expressed in the root meristem and in the developing root cortex, may function as two-component high-affinity nitrate uptake systems, as described for NRT2 and NRT3 proteins in other plants (Okamoto et al., 2006). These data also suggest new regulatory roles for some pine NRT3s (PpNRT3.1, PpNRT3.3, and PpNRT3.6). These genes have a characteristic expression pattern in the vascular regions of the aerial part of the seedlings shared with different NPF transporters (Fig. 4). This could imply unknown NPF and NRT3 association, but a two-component transporter system with still unknown NRT2 cannot be ruled out. Whatever the case, these NRT3 proteins in maritime pine must have an important role in the transport of metabolites across the aerial part of the plant and even in the signaling of the plant metabolic status.

Fig. 4.

Expression heatmap of NPF and NRT2/NRT3 genes in different pine tissues. A heatmap showing the expression of NPF and NRT2 genes in different tissues from 1-month-old seedlings of Pinus pinaster (R.A. Cañas et al., unpublished results). The data are the logarithm of the normalized values with respect to the total mean expression. White boxes correspond to undetected expression. Higher levels of expression correspond to deeper red. Lower levels of expression correspond to lighter yellow.

The TM domain prediction analyses determined the presence of two TM domains for some maritime pine NRT3, as occurs in poplar (Supplementary Table S2). For the NRT2 family, the canonical number of TM domains is 12, but 11 can be found in Arabidopsis (with the exception of 10 for AtNRT2.7), between 8 and 11 in poplar, and 10 for PpNRT2.1 and PpNRT2.2 in pine (Supplementary Table S2).

The expression of NPF and NRT2 genes has been detected in the roots of several conifer species such as white spruce, larch, and Chinese fir (Alber et al., 2012; Meng et al., 2016a). In the present work, taking advantage of the available transcriptome resources, the relative expression levels of NPF and NRT2 genes have been determined in maritime pine seedlings (Fig. 4). However, further studies are needed to determine the kinetics and precisely clarify the specific roles of NPF and NRT2 genes in NO₃^– uptake and transport in gymnosperms.

N uptake and transport in mycorrhized roots

It is widely accepted that the association of forest tree species with ectomycorrhizal (ECM) fungi greatly enhances the ammonium and amino acid uptake capacities of their roots and therefore contributes to increased NUE (Chalot et al., 2002). ECM fungi in symbiosis with poplar roots are able to incorporate NO₃^–, NH₄⁺, urea, amino acids, and peptides, and specific transporters for such N forms have been characterized (Guidot et al., 2005; Näsholm et al., 2009; García et al., 2016). Differences have been found among ECM fungi species in their capacity to improve N uptake in forest trees (Danielsen and Polle, 2014), and some fungal species are particularly effective at enhancing N acquisition in poplars under stressful conditions such as drought (Pena and Polle, 2014). Interestingly, ECM colonization is positively correlated with increases in wood production (Danielsen et al., 2013). In regard to N transfer from ECM fungi to plant roots, NH₄⁺ has been proposed to be a good candidate, although the transfer of organic N in the form of amino acids and peptides has not been excluded (Chalot et al., 2006). Furthermore, it was suggested that the expanded AMT family in poplar could be related to the physiology of mycorrhized trees (Couturier et al., 2007). During mycorrhizal colonization, a highly regulated root-specific AMT has been identified in P. trichocarpa (AMT1.2). This high-affinity AMT is potentially involved in NH₄⁺ transport at the mycelial interface. In conifers, PpAMT1.3 is hightly expressed in roots, and is also induced during mycorrhizal interaction (Castro-Rodriguez et al., 2016b). The participation of specific amino acid and peptide transporters has also been suggested (Muller et al., 2007). However, much less is known about how NH₄⁺ transfer is regulated. In conifers, the small peptide AMP1 is a potential regulator of NH₄⁺ uptake in P. pinaster roots, possibly by interaction with AMT proteins (Canales et al., 2011). AMP1 is up-regulated at the early stages of ectomycorriza formation and therefore might be involved in the regulation of N transfer from the fungi to the plant host (Flores Monterroso et al., 2013). Further research efforts are necessary to elucidate the regulatory mechanisms underlying symbiotic N acquisition of forest trees.

Transcriptomic analysis of N nutrition in trees

Transcriptomic analysis has been performed in poplar to study differential responses to varied sources of inorganic N and also to N stress. Global transcriptomic analyses revealed specific responses to N starvation and N excess in roots and leaves (Luo et al., 2015). Ion transport and auxin-related regulation were enhanced in roots, whereas response to abscisic acid was more important in leaves. Regulatory networks associated with N starvation and N excess have been identified and potential key regulators deserve a detailed functional analysis (Luo et al., 2015). Differentially expressed miRNA and mRNA have been identified in hybrid poplars (Populus deltoides×P. euramericana) in response to N stress (Wang et al., 2016). Interestingly, differences in the expression of miRNAs were observed between hybrid poplar clones with contrasting responses to N stress. Transgenic poplars (Populus tremula×P. alba) overexpressing a cytosolic glutamine synthetase (GS1) and grown under high N levels exhibited increased biomass compared with untransformed controls (Castro-Rodríguez et al., 2016a). The transcriptome of these hybrid poplars was greatly affected by N availability, with >1000 genes differentially regulated under high N, many of which are essential in adaptation and response to N excess. In the same hybrid poplar, exposure to low N levels activates extensive root growth and significant reprogramming of the transcriptome. Transcriptomic analysis revealed the activation of genetic networks related to signaling transduction pathways in response to low N abundance (Wei et al., 2013a, b). A NAC transcription factor was identified and its functional role in greater root biomass was confirmed in transgenic poplar (Wei et al., 2013a). These results demonstrate how the response of poplar roots to limited N involves hierarchical genetic networks controlled by key transcription factors. Furthermore, other transcription factors have been identified that could be involved in the regulation of morphological changes observed in poplar roots grown under different nitrogen sources (Qu et al., 2016). Future research efforts on this topic are necessary to unravel the molecular and regulatory mechanisms underlying the morphological modifications and biochemical events associated with changes in N availability.

Only a few transcriptomic studies have examined the responses of conifers to changes in N nutrition, mainly because the available genomic resources were quite limited until recently. Microarray analysis and suppressive subtraction hybridization were used to identify differentially expressed genes in the roots of maritime pine (P. pinaster) exposed to changes in NH₄⁺ availability (Canales et al., 2010). The expression analyses of selected candidate genes suggest the existence of potential links between unknown ammonium-responsive genes and genes involved in amino acid metabolism, particularly in asparagine biosynthesis and utilization. One such ammonium-responsive gene encodes AMP1, a member of the β-barrelin family previously described in response to pathogen attack, which inhibited NH₄⁺ uptake, indicating that it is involved in the regulation of NH₄⁺ ion flux into pine roots (Canales 2011; Canales et al., 2011). A possible mechanism for this regulation could involve the interaction with root-specific AMT proteins (Fig. 5; Canales, 2011). These findings suggest cross-talk between NH₄⁺ uptake and transport, and the response to pathogen attack, and deserve further investigation. Consistently, the disruption of AMT1.1 in Arabidopsis also generates resistance to pathogen attack (Pastor et al., 2014). In addition to triggering changes in the root transcriptome, it was found that the apex of maritime pine trees is extremely sensitive to conditions of NH₄⁺ excess or deficiency, as revealed by changes in the expression of stress-responsive genes (Canales et al., 2012). This new knowledge may be used to detect early symptoms of N nutritional stresses, thereby increasing survival and growth of young plantlets (Canales et al., 2012).

Fig. 5.

AMP1 expression in response to ammonium nutrition in pine. Proposed model for PpAMP1-mediated regulation of ammonium influx in pine (Pinus pinaster) roots. (This figure is available in colour at JXB online.)

Concluding remarks

In the last few years, considerable advances have been made in the molecular understanding of how N is acquired and transported in trees. Gene families involved in NH₄⁺ and NO₃^– transport have been characterized in woody angiosperms and in gymnosperms. Particularly in poplar and pine, genes involved in the uptake of N from the soil have been identified (Fig. 6). However, the specific roles of other individual genes involved in intracellular ammonium and nitrate transport remain to be determined. Additional gene expression studies and transgenic approaches in poplar, a model tree species, will be required to accomplish this task. In conifers, recent advances have been made in the generation of transgenic trees via somatic embryogenesis, and the function of N transporters in conifers could also be addressed using this approach. The increase in genome resources in poplar and the sequencing of several conifer genomes will permit genome editing using the powerful CRISPR/Cas9 technology. Few amino acid and peptide transporters in trees have been studied in detail and, considering the quantitative importance of organic compounds as a source of N in the forest soils, research efforts on this topic are urgently needed. Several areas of research require special attention. For example, elucidating the regulation of N transport during remobilization of N reserves and the regulation of N transfer in mycorrhized roots are of particular interest. Another area deserving further research effort is the interaction of N nutrition and the response to biotic stresses in trees. Finally, the identification of regulatory networks involved in the response of trees to N nutrition and the functional analysis of key transcription factors would be meaningful additions. The new knowledge derived from such studies will be useful for biotechnological applications aimed to increase NUE in trees.

Fig. 6.

Ammonium and nitrate transporters of poplar and pine involved in the uptake of external nitrogen from the soil. On the left, an illustration of nitrogen uptake and movement in poplar roots. AMT1 subfamily members are mainly involved in ammonium movement from soil to vascular tissues (A). NRT2 subfamily members are specifically involved in nitrate uptake (B). On the right, an illustration of nitrogen uptake and movement in pine roots. Ammonium is taken up through AMT1.3 and AMT2.3. AMT1.3 is mainly distributed in the root surface and endodermis, and AMT2.3 in the root epidermis and cortex (C). Nitrate uptake mediated by the NPF family (D). Putative nitrate transporters could be NPF5.4, NPF7.2, NPF7.5, and NPF8.3, which were detected in the root cortex, and NPF7.2, NPF7.8, and NPF8.3 expressed in the root vasculature.

Supplementary data

Supplementary data are available at JXB online.

Table S1. Protein sequences of the NPF family members used for the phylogenetic tree (Fig. 3A).

Table S2. Protein sequences of the NRT2/NRT3 family members used for the phylogenetic tree (Fig. 3B).

Acknowledgements

Research work in the authors’ laboratory was supported by a grant from the Plant KBBE program, Scientific and Technological Cooperation in Plant Genome Research (PLE2009-0016), and by grants from the Spanish Ministerio de Economía y Competitividad (BIO2015-69285-R) and Junta de Andalucia (BIO-474).

Author notes

*

Present address: Department of Plant Biology, Carnegie Institution for Science, 260 Panama Street, Stanford, CA 94306, USA.

Editor: Alain Gojon, INRA