Available online at www.sciencedirect.com
ScienceDirect Nitrogen signalling pathways shaping root system architecture: an update Brian G Forde Root system architecture is a fundamentally important trait for resource acquisition in both ecological and agronomic contexts. Because of the plasticity of root development and the almost infinite complexity of the soil, root system architecture is shaped by environmental factors to a much greater degree than shoot architecture. In attempting to understand how roots sense and respond to environmental cues, the striking effects of nitrate and other forms of nitrogen on root growth and branching have received particular attention. This minireview focuses on the latest advances in our understanding of the diverse nitrogen signalling pathways that are now known to act at multiple stages in the process of lateral root development, as well as on primary root growth. Addresses Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, UK Corresponding author: Forde, Brian G ([email protected])
Current Opinion in Plant Biology 2014, 21:30–36 This review comes from a themed issue on Cell signalling and gene regulation Edited by Xiangdong Fu and Junko Kyozuka
http://dx.doi.org/10.1016/j.pbi.2014.06.004 1369-5266/# 2014 Published by Elsevier Ltd. All right reserved.
Introduction How efficiently plants explore the soil for their essential supplies of water and nutrients is largely determined by the architecture of their root systems [1]. Root development has long been known to be highly plastic and subject to modification by a wide range of environmental factors [2], amongst the most intensively studied of which is the availability and distribution of different forms of environmental N [3]. Plant roots can absorb and assimilate N in a variety of different forms, both inorganic (nitrate and ammonium) and organic (amino acids and peptides) [4] and the intrinsic complexity and heterogeneity of soils means that there are huge variations in the concentration and distribution of these [5–7]. In terms of developmental plasticity, it is generally observed that lateral roots (LRs) are more sensitive to Current Opinion in Plant Biology 2014, 21:30–36
variations in the N supply (and other environmental signals) than are primary roots [8]. The post-embryonic and multistage nature of LR development provides multiple checkpoints at which root branching can be (and is) regulated: LR development begins with initiation of founder cells in the root pericycle just behind the primary root apex, continues with the formation of a cluster of cells that constitutes the LR primordium and is followed by the formation of a radially symmetrical meristem [9]. In Arabidopsis, activation of this meristem to produce a mature elongating LR only occurs after the LR has emerged from the primary root [9]. Even after activation of the meristem, the elongation rate of the LR may be regulated both by the local external conditions [10] and by endogenous factors [11]. Although the primary root is less sensitive than LRs to the N supply, there are cases where significant effects of nitrate [12,13,14] or amino acids [15,16] on primary root growth have been observed and the nature of the regulatory mechanisms investigated. This short review will focus on advances in the past two years in our understanding of the signal transduction pathways that shape root architecture in response to variations in the external N supply and the plant’s N status. For in-depth background information the reader is referred to a series of comprehensive reviews on this and related topics that have appeared in recent years [15,17– 22]. Also of relevance are recent reviews on nitrate signalling in the context of the regulation of gene expression [21,23,24]. The signalling pathways discussed here are depicted schematically in Figures 1 and 2.
Stimulation of lateral root elongation by external nitrate Early studies of the effect of nitrate on root branching in Arabidopsis were concerned with the ability of a localised nitrate treatment to stimulate LR elongation [25,26], the classic foraging-type response described in the 1970s by Drew and colleagues with barley [27]. Experimentally, the value of using a localised rather than a uniform nitrate treatment is that it allows the specific effect of external nitrate on LR growth to be studied under conditions where systemic effects due to changes in the N status of the plant can be largely discounted [10,25]. Using this approach, the positive effect of a localised nitrate treatment on LR proliferation in Arabidopsis was found to depend primarily on exposure of mature elongating LR tips to the elevated nitrate concentration [25,28]. A signalling pathway that www.sciencedirect.com
Nitrogen signalling and root architecture Forde 31
Figure 1
NO3NO3 NO 3 NO3-
NO3
NO3-
Low [NO3-]
NO3-
-
-
-
NO3
NPF6.3
Intracellular NO3-
NO3NO3NO3- NO3NO3 NO3-
NO3-
NPF6.3
NPF6.3
[auxin] in LR meristem
ANR1
N deficiency N status
N metabolites
systemic signals
TAR2
miR393 AFB3
CLE1, 3, 4 , 7
NAC3/OBP4
CLV1
LR initiation
Early LR development/emergence
LR elongation
Current Opinion in Plant Biology
Multiple pathways regulating the LR response to the N supply in Arabidopsis. Only those pathways discussed in the present review are depicted. Black arrows indicate nitrate transport or assimilatory pathways, green arrows indicate positive signalling steps and red lines indicate negative signalling steps, broken lines indicate systemic signals. See text for further explanation.
Figure 2
Ca2+ Ca2+
Ca2+
MEKK1
2+ Ca2+ Ca
Glu N metabolites
Primary root elongation
NO3NO3-
NO3- NO3NO3NO
3
-
NPF6.3
miR393 NO3-
Intracellular NO3-
AFB3
Current Opinion in Plant Biology
Regulation of primary root growth in Arabidopsis by external nitrogen. The receptor for the external glutamate signal is shown as a glutamate-gated Ca2+ channel because these are known to be active in root tips [32], but their specific role in this pathway is unconfirmed. Note that nitrate has also been reported to have a stimulatory effect on primary root growth both indirectly, by antagonising the inhibitory effect of glutamate, and directly, via an unknown pathway [14]. See text for further information. www.sciencedirect.com
Current Opinion in Plant Biology 2014, 21:30–36
32 Cell signalling and gene regulation
involves the NPF6.3 (CHL1/NRT1.1) nitrate transceptor [29,30] and the ANR1 MADS-box transcription factor was identified that positively regulates meristematic activity at the LR tip in response to external nitrate [25,26]. More recently, the positive regulatory role of ANR1 in LR elongation was confirmed when it was shown that its overexpression in transgenic lines stimulated LR growth while having no direct effect on LR density or primary root growth [31]. This effect was strongest when the ANR1-overexpressing lines were grown in the presence of nitrate [31], indicating that there other nitrateregulated components of this signalling pathway or perhaps that ANR1 itself is post-translationally activated in a nitrate-dependent manner.
As discussed above, NPF6.3’s signalling role upstream of ANR1 in the positive regulation of LR elongation is already established. However, evidence of an additional role for NPF6.3 in repressing LR development on the ‘low nitrate’ part of a split root system has recently been reported [39]. The mechanism for this is identical to that previously proposed for NPF6.3 when acting as a negative regulator of LR growth under uniform N deficiency [40]. In this model, at low external nitrate concentrations (locally or uniformly) NPF6.3 serves as a basipetal auxin transporter in the developing LR tips, maintaining low auxin concentrations in the immature LR meristem and thereby inhibiting LR outgrowth in the N-depleted region of the root [39].
Despite the evolutionary distance between Arabidopsis and cereals, there is recent evidence that ANR1-related genes may have a similar role in regulating the root’s developmental response to nitrate in rice (Oryza sativa L.) [32]. There are five genes in the ANR1-like clade in rice (OsMADS23, 25, 27, 57 and 61) [33], four of which are expressed in roots (OsMADS23, 25, 27 and 57) [34,35]; three of the latter group (OsMADS25, 27 and 57) are also to some extent nitrate-inducible [35]. Four of the ANR1-like genes (OsMADS23, 27, 57 and 61) are targets of miR444 [36,37], a microRNA that is conserved in monocots but absent in Arabidopsis [38]. Although ammonium is generally considered the preferred form of N for rice, LR growth in rice was strongly stimulated by a localised nitrate treatment but not by localised ammonium [32]. In miR444a-overexpressing rice lines, expression of the target MADS-box genes was down-regulated and LRs were no longer responsive to localised nitrate [32]. These data are consistent with a role for one or more of the three root-expressed miR444 targets (i.e. OsMADS23, 27 and 57) in regulating the root architectural response to localised nitrate. An additional role for one or more of the same genes in the root architectural response to P starvation (shorter primary and adventitious roots and longer LRs) was indicated by the finding that the miR444a-overexpressing lines were also defective in this response [32]. However, whether it is the same or different gene(s) from this group that are responsible for both the N and P responses cannot be distinguished from these results. Since miR444 expression was strongly up-regulated by P starvation but was not responsive to nitrate [32] this microRNA seems more likely to have a role in regulating the root’s response to P deficiency rather than its response to external nitrate. Although these results are intriguing in suggesting that the role of ANR1-like MADS-box genes in the nutritional regulation of root architecture has been evolutionarily conserved between dicots and monocots, a note of caution has to be introduced since there is evidence that miR444 has additional target genes in rice whose role in this response has not been formally excluded [36,37].
Stimulation of LR initiation by external nitrate
Current Opinion in Plant Biology 2014, 21:30–36
Most studies with Arabidopsis have found that external nitrate exerts its main effect on LR length and has either no effect [10] or a much smaller effect [39,41,42] on LR numbers or density. A recent study where nitrate’s effect (when applied uniformly to the whole root system) was specifically to stimulate an increase in LR density has identified a role for an auxin-mediated signalling pathway in this response [12]. Since LR initiation is known to be controlled by auxin acting as the signal for the acquisition of founder cell identity [43], an interaction between nitrate and auxin signalling in regulating this process should not be surprising. The AFB3 auxin receptor gene was previously reported to be strongly induced by NO3 , and stimulation of LR initiation by nitrate was markedly reduced in an afb3 mutant [13]. Use of a nitrate reductase (NR)-null mutant established that while nitrate itself was responsible for induction of AFB3, its expression was feedback regulated by products of nitrate assimilation via miR393, a microRNA that targets the AFB3 transcript for degradation [13]. This pathway has now been extended by the finding that nitrate-induction of the NAC4 and OBP4 transcription factor genes is dependent on AFB3, together with the demonstration that a nac4 mutant, like an afb3 mutant, shows markedly reduced stimulation of LR initiation by nitrate [12]. Extending the pathway in the other direction, nitrate regulation of AFB3 and NAC4 gene expression was later shown to be dependent on the NPF6.3 transceptor, but unexpectedly it was its transport function rather than its nitrate sensing function that was important [44]. Two NPF6.3-defective mutants were used in this study, one (chl1-5) is a deletion mutant, while the other (chl1-9) is defective in its activity as a nitrate transporter but not in its nitrate sensing role [29]. While nitrate induction of AFB3 was lost in chl1-5, as expected, it was also lost in chl1-9. This contrasts with NPF6.3’s sensing role in the induction of NRT2.1, in which its nitrate transport function was not required [29,44] and suggests that a different nitrate sensor may be responsible for induction of AFB3, one that is located intracellularly. Intracellular nitrate signalling dependent on a functional nitrate uptake system has also been www.sciencedirect.com
Nitrogen signalling and root architecture Forde 33
observed in the green alga Chlamydomonas reinhardtii [45] and in the fungus Aspergillus nidulans [46]. In these systems, the nitrate-inducible genes were over-expressed in NR-deficient mutants, due to the accumulation of intracellular nitrate in the absence of nitrate assimilation [45,46]. As a means of confirming the role of intracellular nitrate in regulating the expression of AFB3 it could be informative to test the prediction that induction of AFB3 expression (and the downstream effects on LR initiation) would be more sensitive to nitrate in an NR-null mutant than in wild-type Arabidopsis.
Systemic regulation of early LR development by N deprivation The effect of N deprivation on root branching can vary depending on the degree to which the plants are stressed, which may lead to confusion when results obtained in different studies are compared. In a comprehensive analysis of the effect of a diverse range of nutrient conditions on Arabidopsis root architecture it was found that while all root length traits increased under mild N deficiency they decreased under severe N deficiency [47]. A novel mechanism for the systemic inhibition of early LR development in response to N deficiency has been identified [48]. This regulatory pathway involves signalling via number of CLAVATA3 (CLV3)/ENDOSPERM SURROUNDING REGION (ESR)-related (CLE) peptides and the CLAVATA1 (CLV1) leucinerich repeat receptor-like kinase (LRR-RLK), best known for its role in controlling stem cell differentiation in the shoot apical meristem. A set of four CLE genes (CLE1, 3, 4 and 7) was found to be up-regulated when plants were grown for an extended period at low nitrate concentrations [48]. When these were constitutively overexpressed the effect was to delay the development and emergence of LRs, particularly at high nitrate concentrations. Of potential receptors for the CLE peptides, it was only CLV1 that when mutated caused an increase in LR growth. The authors proposed that CLV1 mediates a N-responsive CLE peptide signalling pathway that negatively regulates LR development under N deficiency. The localisation of CLV1 expression in phloem companion cells and expression of the CLE genes in the pericycle indicates a significant role for intercellular signalling and the potential for long-distance phloem-mediated signals to contribute information about the N status of the shoot through effects on CLV1. In a study where the effects of (apparently milder) N deprivation on Arabidopsis were investigated, a positive effect on early LR development was found to depend on the function of the auxin biosynthesis gene TAR2 (tryptophan aminotransferase related 2) [49]. TAR2 expression in the pericycle and the stele of the primary root was upregulated by N deficiency, which correlated with an increase in auxin levels in the developing LR. In a tar2 mutant both the increase in auxin levels and the www.sciencedirect.com
stimulation of LR development by N deprivation were impaired. These findings indicate the existence of a systemic regulatory pathway in which early LR development is stimulated by a reduction in the plant’s N status via induction of auxin biosynthesis in the vicinity of the developing LR. These systemic pathways would represent alternatives to the local, NPF6.3-mediated pathway for regulating early LR development in response to low N [40]. The value of having distinct systemic and local regulatory mechanisms would be that while the systemic pathways enable global LR development to be modulated according to the N status of the plant, it is only the local, NPF6.3-mediated pathway that would provide a mechanism for accentuating the differential in LR proliferation between nitrateenriched and nitrate-deficient zones of the root system [39]. The existence of two separate systemic pathways that act antagonistically on LR development in response to N deprivation would allow for opposite effects on root branching depending on the degree to which the plant is N-deficient, or on the interaction with other environmental factors.
Effects on primary root growth Although primary root growth in Arabidopsis is usually reported to be relatively insensitive to [8,14,25,41] or even stimulated by [14] the normal range of nitrate concentrations, a moderately high rate of nitrate supply can be inhibitory under some culture conditions [12,13]. A three day treatment with 5 mM KNO3 inhibited primary root growth by 30–100%, depending on the accession [13], an effect that was associated with an apparent increase in auxin concentration at the root tip. As was seen for the positive effect of nitrate on LR initiation, its negative effect on primary root growth was lost in an afb3 mutant and in an miR393-overexpressing line [13]. However, in contrast to nitrate’s effect of on LR initiation, its effect on primary root growth did not appear to involve NAC3, since primary root growth in a nac3 mutant was still inhibited by nitrate [12]. It was concluded that AFB3 regulates LR initiation and primary root growth by two distinct pathways, one NAC3-dependent, the other NAC3-independent. It is known that soluble forms of organic N, including amino acids and small peptides, can represent a significant potential source of N for plants, particularly in temperate soils where rates of mineralisation are low [4–7]. However relatively few studies have examined the effect of amino acids and peptides on root growth and branching [15]. The presence of even very low concentrations of glutamate (
ScienceDirect Nitrogen signalling pathways shaping root system architecture: an update Brian G Forde Root system architecture is a fundamentally important trait for resource acquisition in both ecological and agronomic contexts. Because of the plasticity of root development and the almost infinite complexity of the soil, root system architecture is shaped by environmental factors to a much greater degree than shoot architecture. In attempting to understand how roots sense and respond to environmental cues, the striking effects of nitrate and other forms of nitrogen on root growth and branching have received particular attention. This minireview focuses on the latest advances in our understanding of the diverse nitrogen signalling pathways that are now known to act at multiple stages in the process of lateral root development, as well as on primary root growth. Addresses Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, UK Corresponding author: Forde, Brian G ([email protected])
Current Opinion in Plant Biology 2014, 21:30–36 This review comes from a themed issue on Cell signalling and gene regulation Edited by Xiangdong Fu and Junko Kyozuka
http://dx.doi.org/10.1016/j.pbi.2014.06.004 1369-5266/# 2014 Published by Elsevier Ltd. All right reserved.
Introduction How efficiently plants explore the soil for their essential supplies of water and nutrients is largely determined by the architecture of their root systems [1]. Root development has long been known to be highly plastic and subject to modification by a wide range of environmental factors [2], amongst the most intensively studied of which is the availability and distribution of different forms of environmental N [3]. Plant roots can absorb and assimilate N in a variety of different forms, both inorganic (nitrate and ammonium) and organic (amino acids and peptides) [4] and the intrinsic complexity and heterogeneity of soils means that there are huge variations in the concentration and distribution of these [5–7]. In terms of developmental plasticity, it is generally observed that lateral roots (LRs) are more sensitive to Current Opinion in Plant Biology 2014, 21:30–36
variations in the N supply (and other environmental signals) than are primary roots [8]. The post-embryonic and multistage nature of LR development provides multiple checkpoints at which root branching can be (and is) regulated: LR development begins with initiation of founder cells in the root pericycle just behind the primary root apex, continues with the formation of a cluster of cells that constitutes the LR primordium and is followed by the formation of a radially symmetrical meristem [9]. In Arabidopsis, activation of this meristem to produce a mature elongating LR only occurs after the LR has emerged from the primary root [9]. Even after activation of the meristem, the elongation rate of the LR may be regulated both by the local external conditions [10] and by endogenous factors [11]. Although the primary root is less sensitive than LRs to the N supply, there are cases where significant effects of nitrate [12,13,14] or amino acids [15,16] on primary root growth have been observed and the nature of the regulatory mechanisms investigated. This short review will focus on advances in the past two years in our understanding of the signal transduction pathways that shape root architecture in response to variations in the external N supply and the plant’s N status. For in-depth background information the reader is referred to a series of comprehensive reviews on this and related topics that have appeared in recent years [15,17– 22]. Also of relevance are recent reviews on nitrate signalling in the context of the regulation of gene expression [21,23,24]. The signalling pathways discussed here are depicted schematically in Figures 1 and 2.
Stimulation of lateral root elongation by external nitrate Early studies of the effect of nitrate on root branching in Arabidopsis were concerned with the ability of a localised nitrate treatment to stimulate LR elongation [25,26], the classic foraging-type response described in the 1970s by Drew and colleagues with barley [27]. Experimentally, the value of using a localised rather than a uniform nitrate treatment is that it allows the specific effect of external nitrate on LR growth to be studied under conditions where systemic effects due to changes in the N status of the plant can be largely discounted [10,25]. Using this approach, the positive effect of a localised nitrate treatment on LR proliferation in Arabidopsis was found to depend primarily on exposure of mature elongating LR tips to the elevated nitrate concentration [25,28]. A signalling pathway that www.sciencedirect.com
Nitrogen signalling and root architecture Forde 31
Figure 1
NO3NO3 NO 3 NO3-
NO3
NO3-
Low [NO3-]
NO3-
-
-
-
NO3
NPF6.3
Intracellular NO3-
NO3NO3NO3- NO3NO3 NO3-
NO3-
NPF6.3
NPF6.3
[auxin] in LR meristem
ANR1
N deficiency N status
N metabolites
systemic signals
TAR2
miR393 AFB3
CLE1, 3, 4 , 7
NAC3/OBP4
CLV1
LR initiation
Early LR development/emergence
LR elongation
Current Opinion in Plant Biology
Multiple pathways regulating the LR response to the N supply in Arabidopsis. Only those pathways discussed in the present review are depicted. Black arrows indicate nitrate transport or assimilatory pathways, green arrows indicate positive signalling steps and red lines indicate negative signalling steps, broken lines indicate systemic signals. See text for further explanation.
Figure 2
Ca2+ Ca2+
Ca2+
MEKK1
2+ Ca2+ Ca
Glu N metabolites
Primary root elongation
NO3NO3-
NO3- NO3NO3NO
3
-
NPF6.3
miR393 NO3-
Intracellular NO3-
AFB3
Current Opinion in Plant Biology
Regulation of primary root growth in Arabidopsis by external nitrogen. The receptor for the external glutamate signal is shown as a glutamate-gated Ca2+ channel because these are known to be active in root tips [32], but their specific role in this pathway is unconfirmed. Note that nitrate has also been reported to have a stimulatory effect on primary root growth both indirectly, by antagonising the inhibitory effect of glutamate, and directly, via an unknown pathway [14]. See text for further information. www.sciencedirect.com
Current Opinion in Plant Biology 2014, 21:30–36
32 Cell signalling and gene regulation
involves the NPF6.3 (CHL1/NRT1.1) nitrate transceptor [29,30] and the ANR1 MADS-box transcription factor was identified that positively regulates meristematic activity at the LR tip in response to external nitrate [25,26]. More recently, the positive regulatory role of ANR1 in LR elongation was confirmed when it was shown that its overexpression in transgenic lines stimulated LR growth while having no direct effect on LR density or primary root growth [31]. This effect was strongest when the ANR1-overexpressing lines were grown in the presence of nitrate [31], indicating that there other nitrateregulated components of this signalling pathway or perhaps that ANR1 itself is post-translationally activated in a nitrate-dependent manner.
As discussed above, NPF6.3’s signalling role upstream of ANR1 in the positive regulation of LR elongation is already established. However, evidence of an additional role for NPF6.3 in repressing LR development on the ‘low nitrate’ part of a split root system has recently been reported [39]. The mechanism for this is identical to that previously proposed for NPF6.3 when acting as a negative regulator of LR growth under uniform N deficiency [40]. In this model, at low external nitrate concentrations (locally or uniformly) NPF6.3 serves as a basipetal auxin transporter in the developing LR tips, maintaining low auxin concentrations in the immature LR meristem and thereby inhibiting LR outgrowth in the N-depleted region of the root [39].
Despite the evolutionary distance between Arabidopsis and cereals, there is recent evidence that ANR1-related genes may have a similar role in regulating the root’s developmental response to nitrate in rice (Oryza sativa L.) [32]. There are five genes in the ANR1-like clade in rice (OsMADS23, 25, 27, 57 and 61) [33], four of which are expressed in roots (OsMADS23, 25, 27 and 57) [34,35]; three of the latter group (OsMADS25, 27 and 57) are also to some extent nitrate-inducible [35]. Four of the ANR1-like genes (OsMADS23, 27, 57 and 61) are targets of miR444 [36,37], a microRNA that is conserved in monocots but absent in Arabidopsis [38]. Although ammonium is generally considered the preferred form of N for rice, LR growth in rice was strongly stimulated by a localised nitrate treatment but not by localised ammonium [32]. In miR444a-overexpressing rice lines, expression of the target MADS-box genes was down-regulated and LRs were no longer responsive to localised nitrate [32]. These data are consistent with a role for one or more of the three root-expressed miR444 targets (i.e. OsMADS23, 27 and 57) in regulating the root architectural response to localised nitrate. An additional role for one or more of the same genes in the root architectural response to P starvation (shorter primary and adventitious roots and longer LRs) was indicated by the finding that the miR444a-overexpressing lines were also defective in this response [32]. However, whether it is the same or different gene(s) from this group that are responsible for both the N and P responses cannot be distinguished from these results. Since miR444 expression was strongly up-regulated by P starvation but was not responsive to nitrate [32] this microRNA seems more likely to have a role in regulating the root’s response to P deficiency rather than its response to external nitrate. Although these results are intriguing in suggesting that the role of ANR1-like MADS-box genes in the nutritional regulation of root architecture has been evolutionarily conserved between dicots and monocots, a note of caution has to be introduced since there is evidence that miR444 has additional target genes in rice whose role in this response has not been formally excluded [36,37].
Stimulation of LR initiation by external nitrate
Current Opinion in Plant Biology 2014, 21:30–36
Most studies with Arabidopsis have found that external nitrate exerts its main effect on LR length and has either no effect [10] or a much smaller effect [39,41,42] on LR numbers or density. A recent study where nitrate’s effect (when applied uniformly to the whole root system) was specifically to stimulate an increase in LR density has identified a role for an auxin-mediated signalling pathway in this response [12]. Since LR initiation is known to be controlled by auxin acting as the signal for the acquisition of founder cell identity [43], an interaction between nitrate and auxin signalling in regulating this process should not be surprising. The AFB3 auxin receptor gene was previously reported to be strongly induced by NO3 , and stimulation of LR initiation by nitrate was markedly reduced in an afb3 mutant [13]. Use of a nitrate reductase (NR)-null mutant established that while nitrate itself was responsible for induction of AFB3, its expression was feedback regulated by products of nitrate assimilation via miR393, a microRNA that targets the AFB3 transcript for degradation [13]. This pathway has now been extended by the finding that nitrate-induction of the NAC4 and OBP4 transcription factor genes is dependent on AFB3, together with the demonstration that a nac4 mutant, like an afb3 mutant, shows markedly reduced stimulation of LR initiation by nitrate [12]. Extending the pathway in the other direction, nitrate regulation of AFB3 and NAC4 gene expression was later shown to be dependent on the NPF6.3 transceptor, but unexpectedly it was its transport function rather than its nitrate sensing function that was important [44]. Two NPF6.3-defective mutants were used in this study, one (chl1-5) is a deletion mutant, while the other (chl1-9) is defective in its activity as a nitrate transporter but not in its nitrate sensing role [29]. While nitrate induction of AFB3 was lost in chl1-5, as expected, it was also lost in chl1-9. This contrasts with NPF6.3’s sensing role in the induction of NRT2.1, in which its nitrate transport function was not required [29,44] and suggests that a different nitrate sensor may be responsible for induction of AFB3, one that is located intracellularly. Intracellular nitrate signalling dependent on a functional nitrate uptake system has also been www.sciencedirect.com
Nitrogen signalling and root architecture Forde 33
observed in the green alga Chlamydomonas reinhardtii [45] and in the fungus Aspergillus nidulans [46]. In these systems, the nitrate-inducible genes were over-expressed in NR-deficient mutants, due to the accumulation of intracellular nitrate in the absence of nitrate assimilation [45,46]. As a means of confirming the role of intracellular nitrate in regulating the expression of AFB3 it could be informative to test the prediction that induction of AFB3 expression (and the downstream effects on LR initiation) would be more sensitive to nitrate in an NR-null mutant than in wild-type Arabidopsis.
Systemic regulation of early LR development by N deprivation The effect of N deprivation on root branching can vary depending on the degree to which the plants are stressed, which may lead to confusion when results obtained in different studies are compared. In a comprehensive analysis of the effect of a diverse range of nutrient conditions on Arabidopsis root architecture it was found that while all root length traits increased under mild N deficiency they decreased under severe N deficiency [47]. A novel mechanism for the systemic inhibition of early LR development in response to N deficiency has been identified [48]. This regulatory pathway involves signalling via number of CLAVATA3 (CLV3)/ENDOSPERM SURROUNDING REGION (ESR)-related (CLE) peptides and the CLAVATA1 (CLV1) leucinerich repeat receptor-like kinase (LRR-RLK), best known for its role in controlling stem cell differentiation in the shoot apical meristem. A set of four CLE genes (CLE1, 3, 4 and 7) was found to be up-regulated when plants were grown for an extended period at low nitrate concentrations [48]. When these were constitutively overexpressed the effect was to delay the development and emergence of LRs, particularly at high nitrate concentrations. Of potential receptors for the CLE peptides, it was only CLV1 that when mutated caused an increase in LR growth. The authors proposed that CLV1 mediates a N-responsive CLE peptide signalling pathway that negatively regulates LR development under N deficiency. The localisation of CLV1 expression in phloem companion cells and expression of the CLE genes in the pericycle indicates a significant role for intercellular signalling and the potential for long-distance phloem-mediated signals to contribute information about the N status of the shoot through effects on CLV1. In a study where the effects of (apparently milder) N deprivation on Arabidopsis were investigated, a positive effect on early LR development was found to depend on the function of the auxin biosynthesis gene TAR2 (tryptophan aminotransferase related 2) [49]. TAR2 expression in the pericycle and the stele of the primary root was upregulated by N deficiency, which correlated with an increase in auxin levels in the developing LR. In a tar2 mutant both the increase in auxin levels and the www.sciencedirect.com
stimulation of LR development by N deprivation were impaired. These findings indicate the existence of a systemic regulatory pathway in which early LR development is stimulated by a reduction in the plant’s N status via induction of auxin biosynthesis in the vicinity of the developing LR. These systemic pathways would represent alternatives to the local, NPF6.3-mediated pathway for regulating early LR development in response to low N [40]. The value of having distinct systemic and local regulatory mechanisms would be that while the systemic pathways enable global LR development to be modulated according to the N status of the plant, it is only the local, NPF6.3-mediated pathway that would provide a mechanism for accentuating the differential in LR proliferation between nitrateenriched and nitrate-deficient zones of the root system [39]. The existence of two separate systemic pathways that act antagonistically on LR development in response to N deprivation would allow for opposite effects on root branching depending on the degree to which the plant is N-deficient, or on the interaction with other environmental factors.
Effects on primary root growth Although primary root growth in Arabidopsis is usually reported to be relatively insensitive to [8,14,25,41] or even stimulated by [14] the normal range of nitrate concentrations, a moderately high rate of nitrate supply can be inhibitory under some culture conditions [12,13]. A three day treatment with 5 mM KNO3 inhibited primary root growth by 30–100%, depending on the accession [13], an effect that was associated with an apparent increase in auxin concentration at the root tip. As was seen for the positive effect of nitrate on LR initiation, its negative effect on primary root growth was lost in an afb3 mutant and in an miR393-overexpressing line [13]. However, in contrast to nitrate’s effect of on LR initiation, its effect on primary root growth did not appear to involve NAC3, since primary root growth in a nac3 mutant was still inhibited by nitrate [12]. It was concluded that AFB3 regulates LR initiation and primary root growth by two distinct pathways, one NAC3-dependent, the other NAC3-independent. It is known that soluble forms of organic N, including amino acids and small peptides, can represent a significant potential source of N for plants, particularly in temperate soils where rates of mineralisation are low [4–7]. However relatively few studies have examined the effect of amino acids and peptides on root growth and branching [15]. The presence of even very low concentrations of glutamate (
