Fine-tuning of root elongation by ethylene: a tool to study dynamic structure-function relationships between root architecture and nitrate absorption

REVIEW: PART OF A SPECIAL ISSUE ON ROOT BIOLOGY

Fine-tuning of root elongation by ethylene: a tool to study dynamic

structure–function relationships between root architecture and

nitrate absorption

Erwan Le Deunff1,2*, Julien Lecourt3and Philippe Malagoli4,5

1

Universite´ de Caen Basse-Normandie, UMR Ecophysiologie Ve´ge´tale & Agronomie, Nutritions NCS, F-14032 Caen, France, 2

INRA, UMR 950, Ecophysiologie Ve´ge´tale & Agronomie, Nutritions NCS, F-14032 Caen, France, 3East Malling Research, New Road, East Malling ME19 6BJ, Kent, UK,4Universite´ Blaise Pascal-INRA, 24, avenue des Landais, BP 80 006, F-63177 Aubie`re, France and 5INRA, UMR 547 PIAF, B^atiment Biologie Ve´ge´tale Recherche, BP 80 006, F-63177 Aubie`re, France *For correspondence. E-mail erwan.ledeunff@unicaen.fr. Current address: INRA, UMR 950, Laboratoire d’ Ecophysiologie

Ve´ge´tale, Agronomie & Nutritions N, C, S, Universite´ de Caen Basse-Normandie F-14000 Caen, France.

Received: 19 November 2015 Returned for revision: 26 February 2016 Accepted: 12 May 2016 Published electronically: 12 July 2016  Background Recently developed genetic and pharmacological approaches have been used to explore NO

3 /ene signalling interactions and how the modifications in root architecture by pharmacological modulation of ethyl-ene biosynthesis affect nitrate uptake.

 Key Results Structure–function studies combined with recent approaches to chemical genomics highlight the non-specificity of commonly used inhibitors of ethylene biosynthesis such as AVG (L-aminoethoxyvinylglycine). Indeed, AVG inhibits aminotransferases such as ACC synthase (ACS) and tryptophan aminotransferase (TAA) in-volved in ethylene and auxin biosynthesis but also some aminotransferases implied in nitrogen (N) metabolism. In this framework, it can be assumed that the products of nitrate assimilation and hormones may interact through a hub in carbon (C) and N metabolism to drive the root morphogenetic programme (RMP). Although ethylene/auxin interactions play a major role in cell division and elongation in root meristems, shaping of the root system depends also on energetic considerations. Based on this finding, the analysis is extended to nutrient ion–hormone interactions assuming a fractal or constructal model for root development.

 Conclusion Therefore, the tight control of root structure–function in the RMP may explain why over-expressing nitrate transporter genes to decouple structure–function relationships and improve nitrogen use efficiency (NUE) has been unsuccessful.

Key words: Nitrate uptake, root architecture, ethylene biosynthesis, auxin biosynthesis, 1-aminocyclopropane-1-carboxylic acid, l-aminoethoxyvinyl-glycine, aminotransferases, nitrogen use efficiency.

INTRODUCTION

The structural and functional plasticity of the root allow plants to adapt to their fluctuating hydro-mineral environment. Thus, in response to nutrient shortage, the root network comprising the exploratory system (primary root and lateral roots (LRs)) and the root hair system increases its uptake capacity and/or ab-sorbing surface. For instance, nitrate uptake by the root system can be modelled by the simple equation:

NO₃ absorption rate

¼ ðnumber of nitrate transporters  transporter activityÞ= root surface

with transporter activity expressed in terms of amounts of ni-trate taken up per unit of time and root surface is in cm2or ex-pressed as root length in cm or root fresh or dry weight in g. In the above equation, the root surface includes the epidermal cells

(root hair cells and non-hair cells) of the primary root and LRs. The number of nitrate transporters is dimensionless.

The two terms of the ratio reflect the nitrate absorption plas-ticity of the root and emphasize that this plasplas-ticity depends on functional and structural components that operate at two differ-ent timescales (Robinson, 1997, 2005; Lemaire et al., 2013). Transcriptomic and 15NO–₃ analyses have shown that the ex-pression and activity of nitrate transporters are quickly induced (minutes to hours) by nitrate supply (Lejay et al., 1999; Okamoto et al., 2003) whereas at the structural level, root growth and development in response to nitrate require long-term (days to weeks) compensatory mechanisms (Robinson, 1997,2005). Because nitrate is an essential growth-limiting nu-trient ion but also a signalling molecule involved in shoot/root growth through its involvement in nitrogen (N) and carbon (C) metabolism (Scheible et al., 1997a, b), recent emphasis has been placed on local and systemic nitrate signalling and their effects on growth and development of LRs (Scheible et al., 1997a,b;Zhanget al., 1999;Ruffelet al., 2011).

VC _{The Author 2016. Published by Oxford University Press on behalf of the Annals of Botany Company.} All rights reserved. For Permissions, please email: journals.permissions@oup.com

However, there are several lines of evidence indicating that nitrate is not a primary and direct signal involved in growth of exploratory roots and root hair systems:

1. Discrepancy, in terms of duration, between root growth re-sponses induced by nitrate (days to weeks) and hormones such as ethylene and auxin (minutes to hours) raises questions about the importance of nitrate as the primary and direct signal in root morphogenetic programming (Lemaire et al., 2013). For in-stance, auxin is a key signal of the root morphogenetic pro-gramme, orchestrating root meristem patterning and maintenance, the initiation, patterning and emergence of LRs, the building and patterning of the vascular system, the root gravitropic responses and the hydro-patterning of LRs (Turner and Sieburth, 2003;Woodward and Bartel, 2005; Tealeet al., 2006;Pe´retet al., 2009;Bandet al., 2012;Baoet al., 2014).

2. Compared with auxin, nitrate is not a major signal in acid growth theory although it is an essential molecule for growth (Rayle and Cleland, 1992;Hager, 2003;Staalet al., 2011).

3. The lag period (days to weeks) in the root architectural re-sponse to heterogeneous nitrate supply is inconsistent with the effect of a ligand on a receptor activating a morphological re-sponse in a few minutes or hours (Remanset al., 2006a,b). By contrast, ethylene and indole-3-acetic acid (IAA) modulate growth of exploratory roots and root hair systems more rapidly (minutes to hours) (Le et al., 2001; De Cnodderet al., 2007; Leblancet al., 2008;Fraaset al., 2014). In addition, the active concentrations of IAA and ethylene needed for root growth are lower by two or three order of magnitude (lM and nM) than those of nitrate in soil (mM).

4. In contrast to ethylene and IAA signalling mutants, no mu-tants of macronutrient transporters have demonstrated direct and rapid effects on short-term modification in elongation of the exploratory root and root hair systems (Casson and Lindsey, 2003; Remanset al., 2006a,b;Ivanchenkoet al., 2008; Negi et al., 2008;Pe´retet al., 2009).

5. The short-term primary effect of nitrate (minutes to hours) on gene expression obtained from transcriptomic studies has failed to uncover any nitrate receptors or sensors able to act quickly and directly on root growth (Krouket al., 2009;Nero et al., 2009; Vidal et al., 2010). Although nitrate transporter AtNRT1.1 (AtNPF6.3) and phospholipase C are required to trigger Ca2þsignalling involved in the expression changes of nitrate-responsive genes, the nature of this sensing mechanism remains elusive (Riveraset al., 2015).

6. Root structure is mainly composed of carbon skeletons and the pattern of root branching remains very robust even if root proliferation is spatially induced by a heterogeneous supply of a nutrient such as nitrate (X. Wanget al., 2002;Guoet al., 2005). Thus, under conditions of a heterogeneous supply of ni-trate (split root experiment), the total LR length within the root compartment supplied with nitrate after 4 d of treatment is the same as the sum of LR lengths found in both compartments provided with a homogeneous nitrate supply (Ruffel et al., 2011). Furthermore, in the long term, the foraging strategy em-ployed seems to have no effect on the general growth perfor-mances of different species (Johnson and Biondini, 2001; Kembel and Cahill, 2005;De Kroonet al., 2009).

Taken together, these results suggest that root foraging in ni-trate patches probably constitutes one of the environmental cues acting on a more general mechanism of root branching in

plants. It should be stressed that throughout the paper, root mor-phogenetic programme (RMP) means the establishment of a root branched structure under genetic and thermodynamic con-trol. However, the genetic and physical laws involved in root morphogenetic programming can be modulated by environmen-tal factors such as water and availability of nutrient ions. Such factors introduce real-time responses of the branched structure that explain why the branching process is less stereotyped and can adapt to fluctuating environmental cues (Cle´ment and Mauroy, 2014; Bao et al., 2014). These real-time adaptation processes can operate from the cellular level to the root organ level when plants are under steady growth conditions (Kaneko et al., 2015) without invoking the role of ‘developmental insta-bility’ in the shaping of a root system (Forde, 2009). The discrep-ancy between root structural and functional responses to nitrate in terms of duration (minutes vs. days) raises two questions. (1) How is the construction of the root coupled with nitrate absorp-tion during root branching organogenesis? (2) Is it possible to manipulate and decouple structure and function components over longer periods of time (rather than just temporary periods) to improve the capacity for N absorption in crop species?

Analyses of nitrate transporter mutants and their effects on root architecture have revealed the existence of interactions be-tween nitrate, ethylene and IAA signalling and transport (Leblanc et al., 2008; Nero et al., 2009; Tian et al., 2009; Krouk et al., 2010; Zheng et al., 2013; Ma et al., 2014). However, the opposite strategy, in which the activity and ex-pression of nitrate transporters are studied after the induction of root structural changes by the modulation of ethylene or IAA biosynthesis and signalling, has little been used (Leblancet al., 2008; Lemaire et al., 2013). This approach can be used as a tool to explore how nitrate uptake and N metabolism (function) react or compensate for the forcing of root system architecture traits (structure) and shed light on the major mechanisms regu-lating the RMP.

Although the non-specificity of ethylene inhibitors as double edge tools to find new targets involved in the RMP has already been discussed elsewhere (Le Deunff and Lecourt, 2016), this review focuses on the modification of root growth by pharma-cological modulation of the ethylene biosynthesis pathway to evaluate the structure–function relationships between root ar-chitecture and nitrate absorption. It shows how this approach can revise our appreciation of structure–function relationships and can help to find major targets involved in responses of the RMP to nitrogen nutrition.

First consideration: need for a standard growth medium to compare the role of hormones and nutrient ions on root growth

Most of the genetic or pharmacological studies of the effects of ethylene and auxin on root and root hair growth are per-formed using vertical agar plates filled with full or half-strength Murashige and Skoog (MS) standard medium (Guzman and Ecker, 1990; Binder et al., 2004; Tromas et al., 2009; Bruex et al., 2012). Initially designed for in vitro tissue culture, this medium contains high concentrations of mineral nutrients in-cluding 61 mMof nitrogen with 40 mMKNO₃and 21 mMNHþ₄ (Murashige and Skoog, 1962). Such high concentrations of N exceed the maximum soil nitrate concentrations observed in

natural habitats (1 mM) or under field conditions (10 mM) (Le Deunff and Malagoli, 2014). The well-documented reduc-tion in root proliferareduc-tion in response to high concentrareduc-tions of nitrate (10 mM) raises questions about the significance of the data generated from plants growing on MS medium (Scheible et al., 1997a; Remans et al., 2006a, b; Le Ny et al., 2013). Moreover, as ethylene biosynthesis and signalling are involved in the primary response to nitrate supply in the short term (24 h), caution is required when interpreting studies with such high nitrate concentrations in the growth medium (Tianet al., 2009;Zheng et al., 2013). Likewise, long-term changes (days to weeks) of the root absorbing surface induced by the modula-tion of ethylene biosynthesis revealed a structural and func-tional relationship between K15NO3uptake rate and expression of NRT1.1 and NRT2.1 nitrate transporter genes (Leblanc et al., 2008;Lemaire et al., 2013). In addition, NRT1.1 trans-porter is also involved in auxin transport and LR development at low external nitrate concentrations (Krouk et al., 2010; Bouguyonet al., 2015). Again, this suggests that caution is re-quired when interpreting hormones studies with MS medium. Indeed, depending on the external nitrate concentration, nitro-gen effects on root development can be in synergy or antago-nism with the effects of ethylene or auxin treatments (Lemaire et al., 2013;Bouguyonet al., 2015). Accordingly, the definition of an adapted and standardized medium shared by the scientific community working respectively on the effects of the hormones and nutrient ions on root architecture is necessary to compare current and future available data.

BUILDING THE ROOT CATALYTIC FUNCTION FOR NITRATE UPTAKE: INVOLVEMENT OF A COMPLEX OF NITRATE TRANSPORTERS (CNT) Nitrate is the predominant soluble form of N in most high-input agricultural soils (Wolt, 1994;Milleret al., 2007). Because it is a highly mobile ion in the soil compared with NHþ₄ and PO2₄ , its convective and diffusive fluxes in soil are very sensitive to the transpiration stream (Barber, 1995; Tinker and Nye, 2000; Matimati et al., 2014; Hepworth et al., 2015). In the 1990s, based on theenzyme–substrate interpretation of nutrient ion iso-therms proposed by Epstein and co-workers, two distinct nitrate uptake systems corresponding to kinetic components of ion fluxes across the roots were defined at a functional level depending on the external nitrate concentrations (Epstein, 1966, 1972). A high-affinity transport system (HATS) involved at low external nitrate concentration (1 mM) and a low-affinity trans-port system (LATS) operating at high external concentration (1 mM) were defined (Siddiqi et al., 1989, 1990; Faure-Rabasseet al., 2002). However, in the last two decades, the iden-tification and characterization of different gene families of nitrate transporters have proliferated and analyses of transporter mutants in Arabidopsis have challenged this definition (Le Deunff and Malagoli, 2014;Le Deunffet al., 2016).

Families of nitrate transporters involved in the root catalytic function

Transporters belonging to theNRT2 family are mainly in-volved in nitrate uptake at low nitrate concentrations

(1 mM) (Krapp et al., 1998; Cerezo et al., 2001; Faure-Rabasseet al., 2002). InArabidopsis, the AtNRT2 family of nitrate transporter genes possesses seven members. Parallel studies with15N and13N isotope tracers, gene expression and mutant analyses have shown that AtNRT2.1, AtNRT2.2, AtNRT2.4 and AtNRT2.5 are mainly involved in nitrate up-take in mature plant roots whereas the other members of this family are expressed in the shoots (Filleuret al., 2001;Orsel et al., 2002;Okamotoet al., 2003;Liet al., 2007;Kotur and Glass, 2014). In situ histochemical GUS and LUC activities of pNRT2.1::GUS and pNRT2.1::LUC have shown that the AtNRT2.1 promoter is predominantly targeted to older root parts of the primary root and of all laterals. However, NRT2.1 expression was absent from the apices, cell division, transition, elongation and the beginning of maturation zones according to the definition of these zones proposed by Verbelenet al. (2006). In the mature roots, the expression of theAtNRT2.1 promoter was localized to outer layers of the mature root, namely epidermis, cortical paremchyma, endo-dermis and root hairs, but absent from the inner root tissues such as pericycle and stele (Nazoa et al., 2003; Remans et al., 2006b;Girinet al., 2007). A schematic view of the ex-pression location for the different nitrate transporters along the root from the root tip to the root mature zone is presented Fig. 1.

Under high nitrate concentration conditions (1 mM), some of the transporters belonging to the NRT1/Peptide TRansporter family (NRT1/PTR family, termed NPF6) par-ticipate in nitrate uptake (Liu and Tsay, 2003). In Arabidopsis, among the different NRT1 transporters identi-fied and characterized, AtNRT1.1 (NPF6.3) is nitrate-inducible and mainly involved in nitrate uptake whilst the other AtNRT1 genes show constitutive expression in both roots and shoots (Okamotoet al., 2003;Wang et al., 2012). Unlike AtNRT2.1 and AtNRT2.2, expression of AtNRT1.1 (NPF6.3) is mainly located in deeper layers of the mature root: endodermis and pericycle, with greater expression in the epidermis and cortex at the root tip level (Guo et al., 2001,2002). The complementary location of AtNRT2.1 and AtNRT1.1 in mature roots (Fig. 1) raises questions about a potential coupling between both AtNRT2.1 and AtNRT1.1 transporters (Krouk et al., 2006; Leblanc et al., 2013). The AtNRT1.1 transporter has a dual affinity for NO₃ controlled by phosphorylation/dephosphorylation of the transporter (Liu and Tsay, 2003;Glass and Kotur, 2013). Thus,AtNRT1.1 op-erates as a high-affinity transporter when it is phosphorylated and as a low-affinity transporter in the absence of phosphory-lation (Liu and Tsay, 2003;Hoet al., 2009). In essence, the root location and dual affinity for nitrate of NRT1.1 invali-dates the concept of HATS and LATS and the enzyme– substrate interpretation (Le Deunff and Malagoli, 2014; Le Deunffet al., 2016).

The influx of nitrate to roots is also regulated by nitrate import into the vacuole, efflux from the cell and loading into the xylem (Walker and Pitman, 1976;Britto and Kronzucker, 2001,2003). These mechanisms require other types of trans-porter that also contribute to the homeostasis of nitrate in root tissues. The complexity of the root catalytic structure for nitrate uptake has increased by the recent discovery of new gene families,CLC (ChLoride Channel) and NAXT (NitrAte

Excretion Transporter), encoding nitrate transporters in-volved in nitrate homeostasis and nitrate efflux, respectively (Segonzac et al., 2007; De Kroon et al., 2009; Monachello et al., 2009). This complexity will increase in the future with the identification of the genes encoding nitrate carriers in-volved in nitrate influx and efflux from the vacuole (Migocka et al., 2013) or nitrate xylem loading (Ko¨hler et al., 2002;Hanet al., 2016).

In summary, the different types of nitrate transporter present in the mature root form a complex catalytic structure composed by multiple transporters (CNT) that is involved in nitrate ab-sorption over a large range of external nitrate concentrations (Fig. 1, inset). Because of correlative evidence for the func-tional role of NRT2.1 and NRT1.1 (NPF6.3) transporters in root nitrate uptake, the expression of both genes is often used as markers of nitrate uptake activity in plants.

BUILDING THE ROOT ABSORPTION SURFACE: THE IMPORTANCE OF ETHYLENE AND IAA INTERPLAY In the last three decades, pharmacological and genetic approaches have shown that ethylene and auxin are the major growth regulators involved in growth and development of the exploratory roots and root hair systems, resulting in an expan-sion of the root absorbing surface (Duckett et al., 1994; Tanimoto et al., 1995; Rahman et al., 2001; R˚uzicka et al., 2007;Swarupet al., 2007).

Elongation of the exploratory root system: primary root and LRs

Ethylene is one of the major factors controlling root growth and it is involved in IAA transport and partitioning along the primary root as well as in root growth (Stepanovaet al., 2005;

Meristematic zone Transition zone Fast elongation zone Growth terminating zone Mature root

NRT2.1/ Nrt2.2 High absorption zones

NRT1.1 (NPF6.3) Weak absorption zones NO3– NO3 -A Protoxylem/ metaxylem Pericycle/ xylem parenchyma

Endodermis Cortex Epidermis NO3

-Casparian band

Transcellular and symplastic Apoplastic Vacuole NRT2.1 and NRT2.2 CLC a, b NAXT1 NRT1.1 (NPF6.3) NRT1.5 (NPF7.3) NRT1.2 (NPF4.6) NRT1.8 (NPF7.2 ) NRT2.5 NRT2.4 NO3 -pathways B NO3– NO3–

FIG. 1. Location of the different root zones where nitrate is taken up by nitrate transporters of NRT family inArabidopsis. (A) Location of NRT1.1, NRT2.1 and NRT2.2 nitrate transporters in the different zones of the root tips and mature root parts. (B) Details of the complex of nitrate transporters (CNT) that form the root catalytic structure for nitrate uptake and translocation to the shoots in the mature plant roots (adapted fromGuoet al., 2001;Nazoaet al., 2003;Remanset al.,

2006a,b;Kotur and Glass, 2014; Lezhnevaet al., 2014). Red parts are indicative of weak nitrate absorption zones whereas blue parts represent high absorption zones

in high (10 mM) and low (01 mM) nitrate concentrations (fromLazofet al., 1992).

R˚uzickaet al., 2007;Swarupet al., 2007). Ethylene upregulates IAA biosynthesis through the activation of the tryptophan ami-notransferase TAA1 (also known as WEI8, SAV3 and TIR2) and tryptophan aminotransferase-related TAR1 and TAR2 that have overlapping roles in ethylene responses (Stepanovaet al., 2008). These enzymes are involved in the indole-3-pyruvic acid (IPyA) pathway (Stepanovaet al., 2005,2007;Maet al., 2014). Moreover, kinematic studies revealed that ethylene and auxin decrease the elongation rate in the elongation zone of the pri-mary root (Rahman et al., 2001; Swarup et al., 2007). Likewise, the use of microelectrode ion flux measurement ex-periments demonstrated that the control of root elongation by the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) is caused by a rapid alkalinization of apoplast in the root elongation zone through control of Hþ-ATPase activity (Staal et al., 2011). Analysis ofaux1, axr2 and axr4 mutants indicated that ethylene control of the Hþ-ATPases is caused by ethylene effects on both IAA biosynthesis and IAA influx into the root cells through the AUX1 transporter (Staal et al., 2011). However, this control is not the only mechanism by which eth-ylene affects final cell length in the root as etheth-ylene induced changes to the level of apoplast protein content and cross-linking as well as composition of the cellulose–hemicellulose and pectin network of the cell wall (De Cnodder et al., 2005, 2007).

The formation of LRs is also under the control of ethylene and auxin interactions (Ivanchenko et al., 2008; Negi et al., 2008; Lewiset al., 2012). However, contrary to their synergis-tic effect on root elongation, IAA and ethylene act antagonisti-cally on LR formation (Muday et al., 2012). Indeed, root treatment with IAA stimulates LR formation and growth whereas treatments increasing ethylene biosynthesis or its sig-nalling pathway negatively impact LR formation (Negiet al., 2008). Thus, ACC treatments or the use of aneto1 mutant that overproduces ethylene (Guzman and Ecker, 1990) andctr1.1 mutant that induces constitutive ethylene triple responses (Kieber et al., 1993) reduce LR formation. By contrast, treat-ments with 10 lM AgNO₃ [an ethylene receptor antagonist (McDaniel and Binder, 2012) and an efflux activator of IAA (Straderet al., 2009)] or the use of etr1-3 and ein2-5 mutants stimulate root formation by decreasing sensibility to ethylene (Negiet al., 2008). The inhibition of LR formation by ethylene depends also on the increase of IAA transport in both acropetal and basipetal directions as theaux1-7 mutant (IAA influx pro-tein) andpin3 and pin7 mutant (genes encoding IAA efflux pro-tein) are insensitive to ethylene inhibition of LR formation (Negi et al., 2008; Lewis et al., 2011). In fact, induction of AUX1, PIN3 and PIN7 gene expression by ACC treatment stim-ulates acropetal transport of auxin to the root tip and alleviates the local auxin accumulation that drives LR formation (Lewis et al., 2011).

Elongation of the root hair system

Ethylene and auxin act in overlapping and independent ways during the initiation and elongation of the root hair system, as demonstrated by pharmacological and mutant analyses (Masucci and Schiefelbein, 1994; Tanimoto et al., 1995; Rahman et al., 2001; Bruex et al., 2012; Balcerowicz et al.,

2015). Both hormones promote the process of root hair initia-tion and are involved in the molecular mechanisms that deter-mine the site on the epidermal cell where the root hair will be initiated: the planar polarity (Masucci and Schiefelbein, 1994; Fischer et al., 2007; Balcerowicz et al., 2015). For example, treatments with ACC and IAA restore root hair formation in the root hairless mutants rhd6 and theL-aminoethoxyvinylglycine (AVG)-induced defects in root-cell differentiation can be re-versed by ACC treatments (Masucci and Schiefelbein, 1994, 1996). The RHD6 gene encodes a basic helix–loop–helix (bHLH) transcription factor that directly targets the geneRSL4 (Root Hair Defective 6-Like 4), another bHLH transcription factor sufficient to promote root hair cell growth. Due to recip-rocal regulation of their biosynthesis pathways, auxin and ethyl-ene probably act upstream on the bHLH transcription factor family (Yi et al., 2010). However, these signalling pathways have not yet been deciphered (Balcerowiczet al., 2015). Auxin and ethylene are also involved in root hair cell elongation (Pitts et al., 1998; Schiefelbein, 2000; Bruex et al., 2012; Grierson et al., 2014). Auxin influx and efflux control root hair elonga-tion, as demonstrated by the use ofaux1, etr1 and axr1 mutants (Pittset al., 1998). Mutants of ethylene biosynthesis and signal-ling such aseto1, etr1 and ein2 have shown that ethylene is re-quired for root hair elongation. Compared with the wild type, the ethylene overproducereto1 induces longer root hairs, whilst the ethylene-insensitive mutants etr1 and ein2 present shorter root hairs (Pittset al., 1998;Seifertet al., 2004).

Because the biosynthesis pathway of ethylene is less com-plex than that of auxin (see below), modulation of ethylene bio-synthesis and signalling by using activators and inhibitors is an effective strategy to validate genetic approaches (Woodward and Bartel, 2005; Dugardeyn and Van Der Straeten, 2008). However, recent findings have challenged the effectiveness of conventional inhibitors of the ethylene biosynthesis pathway (Leblanc et al., 2008; Soeno et al., 2010; Le Deunff and Lecourt, 2016).

The most widely used inhibitor of ethylene biosynthesis AVG also inhibits IAA biosynthesis and N metabolism

Ethylene is produced from methionine by a three-step bio-chemical pathway.S-adenosylmethionine synthetase first con-verts methionine toS-adenosyl-L-methionine (S-AdoMet), then S-AdoMet is converted by ACC synthase (ACS) in ACC and 50-methylthioadenosine (MTA). Finally, in the presence of O2 and ascorbate, ACC oxidase (ACO) converts ACC to ethylene, hydrogen cyanide (HCN), CO2and dehydroascorbate (DHA).

Because ACS is the rate-limiting step of the ethylene biosyn-thesis pathway, this step was used as a preferential target for in-hibition of ethylene production (Yang and Hoffman, 1984; Wang et al., 2004). Since the 1970s, several bioactive small molecules have been sought to inhibit ACS and ACO and then used to modulate ethylene biosynthesis in fruit slices, flowers and vegetative tissues such as root (Lieberman, 1979;Yang and Hoffman, 1984). However, recent results in root growth and de-velopment have highlighted the non-specificity of commonly used inhibitors of ACS enzyme activity such as AVG and 2-(aminooxy)acetic acid (AOA) (Leblanc et al., 2008; Soeno et al., 2010; Le Deunff and Lecourt, 2016). These two

compounds are Kcat-type irreversible inhibitors of pyridoxal phosphate (PLP)-linked enzymes (Rando, 1974). AVG belongs to the family of olefinic glycine analogues that act as terminal inhibitors of many PLP-dependent enzymes of the subgroup I belonging to the alpha (a) family of aminotransferases (Lieberman, 1979;Satoh and Yang, 1989;Mehtaet al., 1993). This subgroup I contains ACC, aspartate, alanine, tyrosine, histidine-phosphate and phenylalanine aminotransferases (Mehtaet al., 1993;Liepman and Olsen, 2004). AOA belongs to the family of hydroxylamine analogues that react with PLP coenzyme to form stable oximes (Amrhein and Wenker, 1979; Yu et al., 1979; Broun and Mayak, 1981). Recent studies based on chemical genomic or chemical genetic strategies have re-evaluated the effects of these drugs on auxin and ethylene biosynthesis (Fig. 2A) and signalling pathways (Soeno et al.,

2010;Ma and Robert, 2014;Carlandet al., 2016). Chemical ge-nomics and genetics can be defined as the ability of small bio-active molecules to modify protein activity and gene transcription, overcoming the limitations of the mutational approaches such as classical reverse and forward genetics (Zheng and Chan, 2002; Blackwell and Zao, 2003; Robert et al., 2009). Recently, one of these approaches has shown that AOA and AVG are also potent inhibitors of tryptophan amino-transferase enzymes (TAA and TAR) involved in IAA biosyn-thesis (Fig. 2A) through the IPyA pathway (Stepanova et al., 2008;Soenoet al., 2010).

Due to the frequent use of PLP-enzyme inhibitors to block the aminotransferases involved in N metabolism in plants and animals (Miflin and Lea, 1977;Johnet al., 1978), doubts arise regarding the specificity of these families of compounds. In this

9393, 9370, 7303 compounds Indole-3-acetic acid (IAA) L-Tryptophan + 2-OG (Pyruvate)

Indole-3-pyruvic acid (IPyA) + Glutamate (alanine) S-Adenosyl-L-methionine ACC Ethylene Tryptophan aminotransferase PLP-dependent

TAA1 and TARs

Flavine mono-oxygenase YUCCA ACC synthetase PLP-dependent ACS ACC oxidase ACO Ethylene pathway AIA (IPyA) pathway

Methionine SAM synthetase AVG, AOA N metabolism A B

Known effects after mutation or over-expression on the root morphogenetic programme

Ethylene and auxins biosynthesis

Histidine homeostasis and root meristem maintenance

Imidazoleacetol phosphate + Glutamate 2-OG + Histidinol Phosphate IAP PLP-dependent Tyrosine + 2-OG Glutamate + 4-hydrophenylpyruvate TAT PLP-dependent

?

Main effects on the root morphogenetic programme

Phenylpyruvic acid (PPA) +

Glutamate (alanine)

Tryptophan aminotransferase

PLP-dependent

TAA1 and TARs

Flavine mono-oxygenase

YUCCA

PAA (PPA) pathway

AVG, AOA L-Kynurenine L-Phenylalanine + 2-OG (Pyruvate) Phenylacetic acid (PAA)

?

?

?

?

FIG. 2. Some aminotransferase targets of AVG (L-a-(2-aminoethoxyvinyl)glycine) inhibitor involved in the metabolic pathways of auxins, ethylene and nitrogen. (A) Targets of AVG inhibitor on PLP-dependent enzymes involved in AIA, PAA and ethylene biosynthesis pathways. Inhibitors in green are more specific to the enzyme target (Linet al., 2010;Heet al., 2011,Sugawaraet al., 2015). (B) Aminotransferases of the subgroup I belonging to the alpha family of aminotransferases inhibited

by AVG (Miesak and Coruzzi, 2002;Moet al., 2006;McAllister and Good, 2015).

regard, it has been shown in Brassica napus that AVG treat-ment significantly increased the root and shoot concentrations of free amino acids such as Asn, Asp, Gln and Glu (Leblanc et al., 2008;Lemaireet al., 2013). This demonstrates that AVG not only inhibits the synthesis of ethylene and IAA, but at the same time irreversibly blocks several aminotransferases in-volved in N and C assimilation and shuttling through the inter-conversions between organic and amino acids in N metabolism (Fig. 2B). Furthermore, it is well known that AVG-induced de-fects in primary root elongation cannot be restored by ACC treatment (Leblanc et al., 2008; Soeno et al., 2010). Because root treatment with 1 mM glutamate can restore more effec-tively the root elongation of seedlings treated with 10 lMAVG rather than 10 lM ACC, it can be assumed that some amino-transferases of N metabolism targeted by AVG could also be involved in the RMP (Fig. 2B). In the future, identification of small bioactive molecules involved specifically in the inhibition of some aminotransferases involved in N metabolism will allow a better characterization of the loss-of-function phenotypes when all redundant proteins are inhibited (Fig. 2). For example, the use ofL-kynurenine (an analogue ofL-tryptophan) as a com-petitive inhibitor of tryptophan aminotransferase activities of TAA and TAR proteins has allowed the identification of a posi-tive feedback loop between auxin biosynthesis and ethylene signalling through the ethylene-insensitive (EIN3) transcription factor during primary root growth (Heet al., 2011).

GENETIC AND PHARMACOLOGICAL APPROACHES USED TO DECIPHER ETHYLENE AND NITRATE SIGNALLING INTERACTIONS ON

ROOT GROWTH

Genetic and pharmacological approaches in Arabidopsis and B. napus seedlings have recently been used to explore NO₃/ ethylene signalling interactions. Genetic approaches in Arabidopsis have focused on the responses of AtNRT2.1 and AtNRT1.1 nitrate transporters under deficiency and excess of nitrate (Tianet al., 2009;Zhenget al., 2013). Pharmacological approaches inB. napus have focused on structure–function rela-tionships, especially the compensatory mechanisms of nitrate uptake and expression ofBnNRT2.1 and BnNRT1.1 genes when root architecture components are strongly modified by the mod-ulation of ethylene biosynthesis (Leblancet al., 2009;Lemaire et al., 2013).

Genetic approaches reveal an interplay between expression of nitrate transporter and ethylene signalling components

Several studies have focused on expression ofAtNRT nitrate transporter genes in short-term ethylene responses to rapid changes in external nitrate availability (24 h). One of these explored the NO₃/ethylene interaction with pre-grown seed-lings in low nitrate (01 mM) for 5 d followed by transfer to high external concentration (10 mM) for 6–24 h (Tian et al., 2009). Another study explored this interaction when seedlings were grown under high nitrate concentration (10 mM) for 1 week and then transferred to low external nitrate concentration (02 mM) for 24 h (Zhenget al., 2013). Both nitrate nutritional

stresses induced a burst (05–1 h) of ethylene production in the roots, followed by a gradual decrease in ethylene concentration. In seedlings transferred from low to high nitrate concentra-tions,AtNRT1.1 and AtNRT2.2 expression was respectively up-and down-regulated. These opposite regulations were con-firmed in ACC- and AVG-treated seedlings subjected to low and high external nitrate concentration (Tian et al., 2009). BecauseNRT gene expression was no longer responsive to high nitrate concentration in etr1-3 and ein2-1 ethylene-insensitive mutants, the results led to the conclusion that the regulation of both AtNRT genes depends on the ethylene biosynthesis and signalling pathway (Fig. 3A).

In seedlings transferred from high to low nitrate concentra-tions,AtNRT2.1 rather than AtNRT1.1 expression played a posi-tive role in the ethylene biosynthesis and signalling response to nitrate deficiency (Zheng et al., 2013). This result was con-firmed by monitoring EBS:GUS activity [synthetic EIN3(ethylene receptor)-responsive promoter coupled to b-glu-curonidase] in nrt2.1/EBS:GUS and nrt1.1/EBS:GUS lines treated with ACC and AVG (10 lM). In addition, comparison ofAtNRT2.1 expression in control and ctr1-1, ein3-1 and eil-1 mutants indicated that ethylene down-regulatedAtNRT2.1 ex-pression and nitrate uptake via one component of the ethylene-signalling cascade (Zhenget al., 2013). This leads to the con-clusion that a feedback loop should exist between the expres-sion of AtNRT2.1 and ethylene biosynthesis and signalling under nitrate deficiency (Fig. 3B).

Taken together, these results demonstrate that ethylene bio-synthesis and signalling are involved in the short-term primary response to nitrate deficiency or excess by a fine-tuning of AtNRT2.1 and AtNRT1.1 gene expression. However, they do not show how the long-term morphological effects of ethylene on the root system can influence nitrate uptake.

Pharmacological approaches shed light on temporal and dynamic relationships between root architecture and nitrate absorption

InB. napus, a week-long chronic treatment of seedling roots with ACC, AVG and AIB (a-aminoisobutyric acid) under a ho-mogenous supply of nitrate (1 mM) revealed different types of compensatory responses to ethylene-induced changes in explor-atory root and root hair systems. Three distinct groups, pre-sented inFig. 4, can be distinguished.

The first response group, induced by increased external ACC concentrations (from 01 to 10 lM), revealed a dramatic reduc-tion in15NO₃ accumulation and elongation of the exploratory root system (Fig. 4A). This was partly compensated for by an increase in 15NO₃ uptake per root length unit (Fig. 4B), this compensation being unable to restore normal growth and 15

NO₃ accumulation to control levels. The increase in nitrate uptake rate was explained by a strong up-regulation in BnNRT2.1 rather than BnNRT1.1 expression, mainly in root hairs (Leblancet al., 2008;Lemaireet al., 2013).

The second response group was revealed by the inhibition of ACO by AIB treatments (05 and 1 lM). After 5 d of treatment, the total root elongation and15NO₃ accumulation were signifi-cantly increased (Fig. 4A). However, the nitrate uptake rate per root length unit and transcript levels of the BnNRT2.1 and

BnNRT1.1 genes showed no significant difference (Fig. 4B), suggesting that the gain in15NO₃ accumulation was mainly ex-plained by an increase in root length and a conservation of the rate of nitrate uptake (Lemaireet al., 2013). The fine-tuning of ethylene signalling on nitrate uptake in AIB- and ACC-treated seedlings seems mainly exercised via theNRT2.1 gene expres-sion and its transport activity, in accordance with the specific location of NRT2.1 transporter in root hairs and the epidermal and cortical parenchyma cell layers of the mature roots in many species (Nazoaet al., 2003;Tianet al., 2009;Fenget al., 2011; Zhenget al., 2013). This represents the first evidence that ethyl-ene signalling is involved in this compensatory phenomenon of nitrate uptake (Lemaire et al., 2013). Such a view is strongly supported by the up-regulation of transcript levels ofACO and ESR (ethylene responsive sensor) genes in the roots portion fed with nitrate during a split root experiment in rice (Wanget al., 2002).

The third response group was identified by AVG treatment (Fig. 4A). The inhibition of ethylene biosynthesis by 10 lM

AVG treatment significantly reduced the exploratory root and root hair systems of the seedlings by 225–50 % (Leblancet al., 2008; E. Le Deunff et al., unpubl. res.). Surprisingly, AVG-treated seedlings accumulated as much15N as the control seed-lings (Fig. 4A) while the AVG treatment inhibited the synthesis of IAA and ethylene and some other aminotransferases implied in N metabolism (Leblanc et al., 2008; Soeno et al., 2010; Lemaire et al., 2013). This result was mainly explained by a compensatory increase of BnNRT2.1 transcript levels. The ex-pression ofBnNRT2.1 rather than BnNRT1.1 was again linearly correlated with the reduction of the root length and root hair pro-duction (Leblancet al., 2008). In addition, the results also dem-onstrated that root hair cells (trichoblasts) are not the only root location for nitrate absorption, demonstrating that non-hair cells (atrichoblasts) are also actively involved in nitrate uptake to compensate for the depleted number of root hair cells. This result validates the fact that NRT2.1 is targeted to the root plasma membrane of epidermal cells independently of their atrichoblast or trichoblast nature (Chopinet al., 2007;Orselet al., 2007).

A B TAA1/ TARs Ethylene biosynthesis and signalling NRT1·1 transcription ?

Low [NO3–]ext = 0·1 mM

High [NO3–]ext = 10 mM

NRT2·1 transcription

Number and length of lateral roots

Number and length of lateral roots Stress N Increase in NO₃– _uptake EIN3 IAA Ethylene biosynthesis and signalling NRT2·1 transcription Increase in NO₃– _uptake ? Stress N EIN3 TAR2 IAA NRT1·1 transcription NRT1·1 IAA transport High [NO₃–]_ext

= 10 mM

Low [NO3–]ext = 0·2 mM

FIG. 3. Effect of short-term ethylene biosynthesis induction after nitrate nutritional stresses on expression of the NRT nitrate transporter genes and root system archi-tecture inArabidopsis seedlings during homogeneous supply of nitrate. (A) Ethylene burst induced by nitrate provision down-regulates the AtNRT2.1 gene and up-regulates theAtNRT1.1 gene (adapted from Guo et al., 2003;Tianet al. 2009;Heet al., 2011). (B) Ethylene burst induced by deprivation of external nitrate concen-tration in the medium down-regulatesAtNRT2.1 transcription that is normally up-regulated by low nitrate availability. In this model, the ethylene signalling compo-nent will be involved in de-induction of theAtNRT2 gene (adapted fromKrouket al., 2010;Maet al., 2014; Zhenget al., 2014). Arrows are indicative of positive control of gene expression whereas blunted lines are indicative of negative control of gene expression or physiological responses. The dotted arrows indicate

tempo-rary regulations (24–48 h). The ? symbol indicates an unknown signalling cascade.

Taken together, these groups of response demonstrate that dynamic changes of total root length and nitrate uptake are finely adjusted and coordinated by ethylene signalling (Wang et al., 2002; Leblanc et al., 2008; Lemaire et al., 2013). Moreover, the expression and the activity ofNRT2.1 rather than NRT1.1 (NPF6.3) adapts to the ethylene-induced changes of the roots’ absorbing surface. Therefore, the NRT2.1 transporter is probably the main target of ethylene signalling in this struc-ture–function coupling mechanism.

How is the construction of the root coupled with nitrate absorption and assimilation during the root morphogenetic programme?

Importance of aminotransferases in N and C shuttling and hormone biosynthesisThe ethylene modulation of root architec-ture highlighted root strucarchitec-ture–function relationships and shed new perspective on already published studies. AVG, AIB and ACC treatments confirmed that rather than nitrate, ethylene, auxin and their interactions are the primary signals involved in the modulation of the RMP (Ortega-Martinez et al., 2007; Leblancet al., 2008;Thomannet al., 2009;Soenoet al., 2010;

Lemaireet al., 2013;Maet al., 2014). However, the speed of root morphological changes in response to the modulation of ethylene biosynthesis (minutes to hours) is inconsistent with the slow process of root morphogenesis observed under optimal ni-trate supply (days to weeks). This implies that the coordination of absorption and assimilation of N and C fixation slow down the RMP by adjusting the concentrations of endogenous hor-mones. Because ethylene and auxins are synthesized from me-thionine, phenylalanine and tryptophan amino acids, the hormonal control of RMP is dependent on both N and C assimi-lation and shuttling through production of amino and organic acids by aminotransferase activities (Sugawara et al., 2015;Le Deunff and Lecourt, 2016). To avoid falling into a circular ar-gument between nitrogen and hormones, all these results sug-gest that the RMP requires metabolic hubs in the primary metabolism – downstream nitrogen uptake and reduction – able to control both the flows of C and N assimilates in primary me-tabolism and hormone biosynthesis (Fig. 2).

The structural and functional relationships of the root system are also based on energetic considerationsAlthough at the cellular level ethylene and auxin transport and synthesis both modulate cell division and elongation in the primary root ( Ortega-Martinez et al., 2007; Stepanova et al., 2008; Tromas et al., 2009;Thomannet al., 2009) and participate in LR development (Negiet al., 2008;Lewiset al., 2011), the shaping of the root system under steady growth conditions on agar plates also needed an additive assumption based on energetic consideration (Lambers et al., 1996; Cle´ment and Mauroy, 2014; Kaneko et al., 2015). This complementary hypothesis is also consistent with the building of a fractal or constructal root network (West et al., 1997, 1999; Bejan, 2005; Miguel, 2006; Bejan and Lorente, 2010). Briefly, building of a fractal root network sup-poses the involvement of an iterative mathematical algorithm mainly driven by a genetic programme whereas the constructal root network supposes that thermodynamic laws are mainly in-volved in root shaping (Bejan, 2005;Miguel, 2006;Bejan and Zane, 2012). Indeed, in both approaches the filling volume of soil by a branched structure such as roots is designed to opti-mize the rates at which energy, materials and wastes may be supplied or removed by the circulation of xylem and phloem fluids. In other words, the shaping of a root system under ho-mogenous or heterogeneous nitrate supply requires adopting a thermodynamic viewpoint during steady-state growth. Furthermore, the relationship found between the decrease in ni-trate uptake per root length during root ageing presented in Fig. 5Areveals one of these thermodynamic aspects involved in RMP (Lemaireet al., 2013). A similar observation has been made inB. napus subjected to flowing solutions of nitrate at different concentrations from 10 lM to 10 mM (Fig. 5B). Indeed, the increase in length of the root network was accompa-nied by a reduction in nitrate uptake per root length for all the tested nitrate concentrations (Bhatet al., 1979a,b; Le Deunff and Malagoli, 2014). Although it is likely that this weighted be-haviour does not reflect the heterogeneity of nitrate uptake along the root caused by root ageing (Robinsonet al., 1991;Le Deunff and Malagoli, 2014;Vetterlein and Doussan, 2016), we can assume that such behaviour minimizes the energy costs re-quired for the structure building and the root functional proper-ties (Fig. 5). In addition, it allows the roots to adopt a

y = 0.42503e0.05413x r = 0.98 20 0 40 60 80 100 120 140 60 70 80 90 100 110 120 130 140 1 mMKNO₃ 10µM ACC 0.5-1µM AIB 1µM ACC 0.1µM ACC 10µM AVG

Relative total root length

compared to control (1 m M KNO 3 ) Relative 15N accumulation compared to control (1 mM KNO₃)

B 0 0·001 0·002 0·003 0·004 0·005 0·006 15 N uptake rate (µ mo l. h –1 . root cm –1 ) b c a a a a A 1mM KNO₃ Root treatments Control ACC 10µM ACC 1µM ACC 0.1µM AIB 0.5µM AIB 1µM

FIG. 4. Structure–function relationship between15_{N accumulation and total root} length inB. napus seedlings (‘Capitol’) growing in agarose gel for 5 d fed with 1 mMK15_NO

3and treated with activator (ACC) or inhibitors (AIB and AVG) of ethylene biosynthesis pathway during 5 d. (A) Data come from two independent experiments and represent the relative evolution compared to control of total root length (cm) and15N accumulation (lg15N per plant). (B) Effect of ACC and AIB on the15_NO

3 uptake rate per cm of root length after 120 h of treat-ment. Values are mean 6 s.e. ofn¼ 4–5 repeats (Petri dishes) of four seedlings

each (adapted fromLeblancet al., 2008;Lemaireet al., 2013).

prospective and active strategy to adapt to a heterogeneous sup-ply of nutrient ions and water in the soil (Drew and Saker, 1975; Robinson, 1997, 2005; Bao et al., 2014). When roots meet a nitrate-rich patch, they proliferate and increase locally and temporarily their energy cost for nitrate absorption and root structure formation (Bloomet al., 1992;Cannell and Thornley, 2000;Lamberset al., 1996). This is accompanied by a drop in energy demand in the rest of the unsupplied root system. Indeed, nitrate uptake represents 20–40 % of the respiratory en-ergy cost whereas the cost of biomass maintenance increases from 25 to 70 % with plant age at the expense of nitrate uptake and growth (van der Werf et al., 1988; Lambers et al., 1996). The fundamental role played by the circulation of xylem and phloem fluids associated with energetic considerations involved in RMP is emphasized by two major results. First, the root hy-draulic properties and total volume flow in xylem vessels are dependent on nitrate availability (Hoarau et al., 1996; Gorska et al., 2008; Schulze-Till et al., 2009). Second, it has been shown that a part of the water used for xylem flow (19–54 %) is replenishing water provided to the phloem via Mu¨nch’s coun-terflow (Tanner and Beevers, 2001;Windtet al., 2006). Hence, a preferential circulation of nutrients, amino acids, carbohy-drates, signalling molecules and hormones such IAA is induced in roots fed with nitrate, causing root proliferation without alter-ing substantially the RMP but maximizalter-ing energy costs

(Robinson, 1997,2005;Wanget al., 2002;Guoet al., 2005;De Kroon et al., 2009). These results are also in agreement with the hydro-patterning mechanism inArabidopsis that determines the position of lateral roots through local regulation of TAA1 and PIN3 (PINE-FORMED 3) genes and the involvement of the OsNPF2.2 nitrate transporter in nitrate xylem loading and vasculature formation of the roots and shoots in rice (Baoet al., 2014; Li et al., 2015). Accordingly, the LR proliferation in-duced by the spatial heterogeneity of available water and nitrate in the soil raises the question of a possible overlap in the signal-ling pathways between the root mineral- and hydro-patterning (Drew and Saker, 1975;Baoet al., 2014;Maet al., 2014).

Is it possible to manipulate and to decouple nitrate uptake from the root morphogenetic programme? The tight control of root structure–function in RMP may explain why attempts to im-prove nitrate absorption by over-expressing NRT2.1 and NRT1.1 genes in order to decouple structure–function relation-ships were unsuccessful. Indeed,Nicotiana plumbaginifolia and Oriza sativa transgenic plants constitutively over-expressing NpNRT2.1 and OsNRT2.1 revealed that the NRT2.1 transpor-ter is also regulated at post-transcriptional level by reduced N sources or metabolites (Fraisieret al., 2000; Katayama et al., 2009), while over-expressing AtNRT1.1 in Arabidopsis in-creased nitrate uptake in constitutive but not in inductive nitrate conditions (Liu et al., 1999). Moreover, the dependence of NRT1.1 to Ca2þsignalling in relation to its phosphorylation/de-phosphorylation state demonstrates that NRT1.1 activity is also strongly regulated at the post-transcriptional level (Riveras et al., 2015). Furthermore, recent discovery of alternative mRNA splicing product of the NRT1.1 gene in rice encoding a low-affinity nitrate transporter with six transmembrane domains increases the complexity of NRT1.1 regulation in nitrate uptake in plants (Fanet al., 2015). Although NRT1.1 and NRT2.1 are induced in the nitrate priming effect (Krouket al., 2009;Nero et al., 2009), AtNRT1.1 and AtNRT2.1 transporter genes do not belong to the metabolic hub in primary N and C metabolism involved in the root branching programme, as revealed by the nrt1.1 and nrt2.1 mutants in 6-d and older seedlings (Guoet al., 2001;Remanset al., 2006a,b;Hoet al., 2009).

As demonstrated with tobacco Nia30(145) transformants of nitrate reductase (NR), changes in RMP under homogeneous and heterogeneous supply of nitrate depends more on the shoots accumulation of nitrate rather than changes in root nitrate up-take (Dorbeet al., 1992;Scheible et al., 1997a; Gojonet al., 1998;Stitt and Feil, 1999). In fact, nitrate-limited Nia30(145) transformants treated with 12 mMnitrate lead to wide changes in the expression of genes involved in the pathways of C and N metabolism (Scheible et al., 1997a, b; Stitt and Feil, 1999). Accumulation of nitrate in the shoots leads to a strong inhibi-tion of starch synthesis and turnover and to a decrease of sugar allocation to and concentration in the roots. Furthermore, it has been well demonstrated in Arabidopsis that the primary root growth rate depends on diurnal carbon allocation from an ap-propriate rate of starch degradation orchestrated by clock genes ELF3 and CCA1/LHY (Yazdanbakhshet al., 2011). Although changes in root growth under homogeneous and heterogeneous nitrate supply can be also explained by energetic consider-ations, the mechanisms that connect starch, hormone and nitro-gen metabolism are still missing. However, the fundamental

Total root length (cm)

15 N u ptake rate (µ mo l. h –1 . cm –1 root length) 0 50 100 150 200 0 5 10 15 20 25 0 0·001 0·002 0·003 0·004 0·005 0·006 0·007 0·008 0·009 A y = 0·0068 e–0·0116x r = 0·951

Time after transplanting (days) B 0 0·005 0·01 0·015 0·02 0·025 0·03 0·035 0·04 1mM KNO₃ 10µM KNO₃ 100µM KNO₃ 10mM KNO₃ Nitrate uptake rate (µ mo l. h –1 . cm –1 root length)

FIG. 5. Structure–function relationships between15N uptake rate and total root length inB. napus seedlings. (A) Relationship between15_NO

3 uptake and total root length during the time course of the experiment from 0 to 120 h inB. napus seedlings (‘Capitol’) (fromLemaireet al., 2013). Values are mean 6 s.e. ofn¼ 4–5 repeats (Petri dishes) of four seedlings each. (B) Long-term down-regulation of the nitrate uptake rate inB. napus plants (‘Emerald’). The plants were grown in a continuous flow culture system at 25_{C and 32 kLux and supplied with} con-stant 10 mM, 100 mM, 1 mMor 10 mMof external nitrate concentrations (from

Bhatet al., 1979a).

role of some aminotransferases in N and C assimilation and shuttling and the biosynthesis of ethylene and auxin could be the missing and centrepiece in the puzzle.

CONCLUDING REMARKS

Although systems biology approaches have allowed an holistic view on the connections between N metabolites and compo-nents of hormonal pathways and have indicated that hormones are crucial elements in plant developmental responses to N (Krouket al., 2009;Neroet al., 2009;Vidalet al., 2010), this systems approach does not take account of the thermodynamic considerations involved in structure–function relationships dur-ing the builddur-ing of a fractal or constructal root network durdur-ing steady-growth state (West et al., 1997, 1999; Bejan, 2005; Bejan and Lorente, 2010). Chemical genetics can provide a complementary tool to pinpoint the central hub(s) in primary metabolism involved in the regulation of the root morphoge-netic programme (Le Deunff and Lecourt, 2016).

AQ1: Please confirm that Heads 1 and 2 are used correctly throughout

LITERATURE CITED

Amrhein N, Weker D. 1979. Novel inhibitors of ethylene production in higher plants.Plant, Cell and Physiology 20: 1635–1642.

Balcerowicz D, Schoenaers S, Vissenberg K. 2015. Cell fate determination and the switch from diffuse growth to planar polarity inArabidopsis root epi-derlmal cells.Frontiers in Plant Science 6: 1163.

Band LR, Wells DM, Larrieu A, et al. 2012. Root gravitropism is regulated by a transient lateral auxin gradient controlled by a tipping-point mechanism. Proceedings of the National Academy of Sciences USA 109: 4668–4673. Bao Y, Aggarwal P, Robbins NE, et al. 2014. Plant roots use a patterning

mech-anism to position lateral root branches toward available water.Proceedings of the National Academy of Sciences USA 111: 9319–9324.

Barber SA. 1995. Nutrient absorption by plant roots. In: Barber SA, ed. Soil nu-trient bioavailability: a mechanistic approach. New York: John Wiley & Sons, 49–84.

Bejan A (2005) The constructal law of organization in nature: tree-shaped flows and body size.The Journal of Experimental Botany 208: 1677–1686. Bejan A, Lorente S. 2010. The constructal law of design and evolution in nature.

Philosophical Transactions of the Royal Society B: Biological Sciences 365: 1335–1347.

Bejan A, Zane JP. 2012. Design in nature: how the constructal law governs evo-lution in biology, physics, technology and social organization. Doubleday: New York.

Bhat K, Nye P, Brereton A. 1979a. The possibility of predicting solute uptake and plant growth response from independently measured soil and plant char-acteristics VI. The growth and uptake of rape in solutions of constant nitrate concentration.Plant and Soil 53: 137–167.

Bhat K, Nye P, Brereton A. 1979b. The possibility of predicting solute uptake and plant growth response from independently measured soil and plant char-acteristics VII. The growth and uptake of rape in soil with two nitrate con-centrations and a comparaison of the results with model predictions.Plant and Soil 53: 169–191.

Binder BM, O’Malley RC, Wang W, et al. 2004. Arabidopsis seedling growth response and recovery to ethylene. A kinetic analysis.Plant Physiology 136: 2913–2920.

Blackwell HE, Zhao Y. 2003. Chemical genetic approaches to plant biology. Plant Physiology 133: 448–455.

Bloom AJ, Sukrapanna SS, Warner RL. 1992. Root respiration associated with ammonium and nitrate absorption and assimilation by barley.Plant Physiology 99: 1294–1301.

Bouguyon E, Brun F, Meynard D, et al. 2015. Multiple mechanisms of nitrate sensing by Arabidopsis nitrate transceptor NRT1.Nature Plants 1: 15015.

Britto DT, Kronzucker HJ. 2001. Constancy of nitrogen turnover kinetics in the plant cell: insights into the integration of subcellular N fluxes.Planta 213: 175–181.

Britto DT, Kronzucker HJ. 2003. Ion fluxes and cytosolic pool sizes: examin-ing fundamental relationships in transmembrane flux regulation. Planta 217: 490–497.

Broun R, Mayak S. 1981. Aminooxyacetic acid as an inhibitor of ethylenesyn-thesis and senescence in carnation flowers. Scientia Horticulturae 15: 277–282.

Bruex A, Kainkaryam RM, Wieckowski Y, et al. 2012. Agene regulatory net-work for root epidermis cell differentiation inArabidopsis. PLOS Genetics 8: 1–20, e1002446.

Cannell M, Thornley J. 2000. Modelling the components of plant respiration: some guiding principles.Annals of Botany 85: 45–54.

Carland F, Defries A, Cutler S, Nelson T. 2016. Novel vein patterns in Arabidopsis induced by small molecules.Plant Physiology 170: 339–353 Casson SA, Lindsey K. 2003. Genes and signalling in root development. New

Phytologist 158: 11–38.

Cerezo M, Tillard P, Filleur S, Munos S, Daniel-Vedele F, Gojon A. 2001. Major alterations of the regulation of root NO₃ upatake are associated with the mutation ofNrt2.1 and Nrt2.2 genes in Arabidopsis. Plant Physiology 127: 262–271.

Chopin F, Wirth J, Dorbe M-F, et al. 2007. The Arabidopsis nitrate transporter AtNRT2. 1 is targeted to the root plasma membrane.Plant Physiology and Biochemistry 45: 630–635.

Cle´ment R, Mauroy B. 2014. An archetypal mechanism for branching organo-genesis.Physical Biology 11: 016003.

De Cnodder T, Vissenberg K, Van Der Straeten D, Verbelen JP. 2005. Regulation of cell length in theArabidopsis thaliana root by the ethylene precursor 1-aminocyclopropane-1-carboxylic acid: a matter of apoplastic re-actions.New Phytologist 168: 541–550.

De Cnodder T, Verbelen J-P, Vissenberg K. 2007. The control of cell size and rate of elongation in theArabidopsis root. In: Verbelen J-P, Vissenberg K, eds. The expanding cell. Vol. 5. Berlin: Springer Science & Business Media, 249–269.

De Kroon H, Visser EJ, Huber H, Mommer L, Hutchings MJ. 2009. A modu-lar concept of plant foraging behaviour: the interplay between local re-sponses and systemic control.Plant, Cell & Environment 32: 704–712. Dorbe MF, Caboche M, Daniel-Vedele F. 1992. The tomato nia gene

comple-ments aNicotiana plumbaginifolia nitrate reductase-deficient mutant and is properly regulated.Plant Molecular Biology 18: 363–375.

Drew MC, Saker LR. 1975. Nutrient supply and the growth of the seminal root system in Barley.Journal of Experimental Botany 26: 79–90.

Duckett C, Grierson C, Linstead P, et al. 1994. Clonal relationships and cell patterning in the root epidermis of Arabidopsis. Development 120: 2465–2474.

Dugardeyn J, Van Der Straeten D. 2008. Ethylene: fine-tuning plant growth and development by stimulation and inhibition of elongation.Plant Science 175: 59–70.

Epstein E. 1966. Dual pattern of ion absorption by plant cells and by plants. Nature 212: 1324–1327.

Epstein E. 1972. Nutrient absorption by plants. In: Epstein E, ed. Mineral nutri-tion of plants: principles and perspectives, 2nd edn. New York: John Wiley & Sons, 69–95.

Fan X, Feng H, Tan Y, Xu Y, Miao Q, Xu G. 2015. A putative 6 trans-membrane transporter OsNRT1.1b plays a key role in rice under low nitro-gen.Journal of Integrative Plant Biology. doi:10.1111/jipb.12382. Faure-Rabasse S, Le Deunff E, Laine P, Macduff JH, Ourry A. 2002. Effects

of nitrate pulses onBnNRT1 and BnNRT2 genes: mRNA levels and nitrate influx rates in relation to the duration of N deprivation inBrassica napus L. Journal of Experimental Botany 53: 1711–1721.

Feng H, Yan M, Fan X, et al. 2011. Spatial expression and regulation of rice high-affinity nitrate transporters by nitrogen and carbon status.Journal of Experimental Botany 62: 2319–2332.

Fischer U, Ikeda Y, Grebe M. 2007. Planar polarity of root hair positioning in Arabidopsis.Intercellular Signalling in Plants 35: 149–151.

Filleur S, Dorbe M-F, Cerezo M, et al. 2001. An Arabidopsis T-DNA mutant af-fected in Nrt2 genes is impaired in nitrate uptake.FEBS Letters 489: 220–224. Forde BG. 2009. Is it good noise? The role of developmental instability in the

shaping of the root system.Journal of Experimental Botany 60: 3989–4002. Fraas S, Niehoff V, Lu¨then H. 2014. A high-throughput imaging auxanometer for roots and hypocotyls ofArabidopsis using a 2D skeletonizing algorithm. Physiologia Plantarum 151: 112–118.

Fraisier V, Gojon A, Tillard P, Daniel-Vedele F. 2000. Constitutive expression of a putative high-affinity nitrate transporter in Nicotiana plumbaginifolia: evidence for post-transcriptional regulation by a reduced nitrogen source. The Plant Journal 23: 489–496.

Girin T, Lejay L, Wirth J, et al. 2007. Identification of a 150 bp cis-acting ele-ment of the AtNRT2. 1 promoter involved in the regulation of gene expres-sion by the N and C status of the plant.Plant, Cell & Environment 30: 1366–1380.

Glass ADM, Kotur Z. 2013. A re-evaluation of the role of NRT1.1 in high-affinity nitrate transport?Plant Physiology 163: 1103–110.

Gojon A, Dapoigny L, Lejay L, Tillard P, Rufty TW. 1998. Effects of genetic modification of nitrate reductase expression on15_NO

3 uptake and reduction inNicotiana plants. Plant, Cell & Environment 21: 21–53.

Gorska A, Ye Q, Holbrook NM, Zwieniecki MA. 2008. Nitrate control of root hydraulic properties in plants: translating local information to whole plant response.Plant Physiology 148: 1159–1167.

Grierson C, Nielsen E, Ketelaarc T, Schiefelbein J. 2014. Root hairs. In The Arabidopsis book. American Society of Plant Biologists.

Guo F-Q, Wang R, Chen M, Crawford NM. 2001. The Arabidopsis dual-affinity nitrate transporter gene AtNRT1. 1 .CHL1. is activated and func-tions in nascent organ development during vegetative and reproductive growth.The Plant Cell 13: 1761–1777.

Guo FQ, Wang R, Crawford NM. 2002. The Arabidopsis dual-affinity nitrate transporter gene AtNRT1.1.CHL1. is regulated by auxin in both shoots and roots.Journal of Experimental Botany 53: 835–844.

Guo Y, Chen F, Zhang F, Mi G. 2005. Auxin transport from shoot to root is in-volved in the response of lateral root growth to localized supply of nitrate in maize.Plant Science 169: 894–900.

Guzman P, Ecker JR. 1990. Exploiting the triple response of Arabidopsis to identify ethylene-related mutants.The Plant Cell 2: 513–523.

Hager A. 2003. Role of the plasma membrane Hþ-ATPase in auxin-induced elongation growth: historical and new aspects.Journal of Plant Research 116: 483–505.

Han Y-L, Song H-X, Liao Q, Yu Y, Jian S-F, Lepo JE, Liu Q, Rong X-M, Tian C, Guan C-Y, Ismail AM, Zhang Z-H. 2016. Nitrogen use efficiency is mediated by vacuolar nitrate sequestration capacity in roots ofBrassica napus. Plant Physiology.170: 1684–1698.

He W, Brumos J, Li H, et al. 2011. A small-molecule screen identifies l-lynure-nine as a competitive inhibitor of TAA1/TAR activity in ethylene-directed auxin biosynthesis and root growth inArabidopsis. The Plant Cell 23: 3944–3960.

Hepworth C, Doheny-Adams T, Hunt L, Cameron D, Gray J. 2015. Manipulating stomatal density enhances drought tolerance without deleteri-ous effect on nutrient uptake.New Phytologist 2: 336–341.

Ho C-H, Lin S-H, Hu H-C, Tsay Y-F. 2009. CHL1 functions as a nitrate sensor in plants.Cell 138: 1184–1194.

Hoarau J, Barthes L, Bousser A, Dele´ens E, Prioul J-L. 1996. Effect of nitrate on water transfer across roots of nitrogen pre-starved maize seedlings. Planta 200: 405–415.

Ivanchenko GM, Muday G, Dubrovsky JG. 2008. Ethylene-auxin interactions regulate lateral root initiation and emergence inArabidopsis thaliana. The Plant Journal 55: 335–3347.

John RA, Charteris A, Fowler LJ. 1978. The reaction of amino-oxyacetate with pyridoxal phosphate-dependent enzymes.Biochemistry Journal 171: 771–779.

Johnson HA, Biondini ME. 2001. Root morphological plasticity and nitrogen uptake of 59 plant species from the Great Plains grasslands, USA.Basic and Applied Ecology 2: 127–143.

Kaneko K, Furusawa S, Yomo T. 2015. Universal relationship in gene express-sion changes for cells in steady-growth state.Physical Review 5: 1–8. Katayama H, Mori M, Kawamura Y, Tanaka T, Mori M, Hasegawa H.

2009. Production and characterization of transgenic rice plants carrying a high-affinity nitrate transporter gene OsNRT2.1. Breeding Science 59: 237–243.

Kembel SW, Cahill JFJ. 2005. Plant phenotypic plasticity belowground: a phy-logenetic perspective on root foraging trade-offs.The American Naturalist 166: 216–230.

Kieber JJ, Rothenberg M, Roman G, Feldmann KA, Ecker JR. 1993. CTR1, a negative regulator of the ethylene response pathway in Arabidopsis, encodes a member of the raf family of protein kinases. Cell 72: 427–441.

Ko¨hler B, Wegner LH, Osipov V, Raschke K. 2002. Loading of nitrate into the xylem: Apoplastic nitrate controls the voltage dependence of X-QUAC, the

main anion conductance in xylem-parenchyma cells of barley roots.The Plant Journal 30: 133–142.

Kotur Z, Glass AD. 2014. A 150 kDa plasma membrane complex of AtNRT2.5 and AtNAR2.1 is the major contributor to constitutive high-affinity nitrate influx inArabidopsis thaliana. Plant, Cell & Environment 38: 1490–1502. Krapp A, Fraisier V, Scheible WR, et al. 1998. Expression studies of

Nrt2;1Np, a putative high-affinity nitrate transporter: evidence for its role in nitrate uptake.Plant Journal 14: 723–731.

Krouk G, Tillard P, Gojon A. 2006. Regulation of the high-affinity NO3 up-take system by NRT1. 1-mediated NO₃ demand signaling inArabidopsis. Plant Physiology 142: 1075–1086.

Krouk G, Tranchina D, Lejay L, et al. 2009. A systems approach uncovers re-strictions for signal interactions regulating genome-wide responses to nutri-tional cues inArabidopsis. PLoS Computational Biology 5: e1000326. Krouk G, Lacombe B, Bielach A, et al. 2010. Nitrate-regulated auxin transport

by NRT1. 1 defines a mechanism for nutrient sensing in plants. Developmental Cell 18: 927–937.

Lazof DB, Rufty TW, Redinbaugh MG. 1992. Localization of nitrate absorp-tion and translocaabsorp-tion within morphological regions of the corn root.Plant Physiology 100: 1251–1258.

Lambers H, Atkin OK, Millenaar FF. 1996. Respiratory patterns in roots in re-lation to their functioning. In: Waisel Y, Eshel A, Kafkaki K, eds.Plant roots. The hidden half. New York: Marcel Dekker, Inc., 323–362. Le J, Vandenbussche F, Van Der Straeten D, Verbelen J-P. 2001. In the early

response ofArabidopsis roots to ethylene, cell elongation is up-and down-regulated and uncoupled from differentiation. Plant Physiology 125: 519–522.

Le Deunff E, Lecourt J. 2016. Non-specificity of ethylene inhibitors: ‘double-edged’ tools to find out new targets involved in the root morphogenetic pro-gramme.Plant Biology 18: 353–361.

Le Deunff E, Malagoli P. 2014. Breaking conceptual locks in modelling root ab-sorption of nutrients: reopening the thermodynamic viewpoint of ion trans-port across the root.Annals of Botany 114: 1555–1570.

Le Deunff E, Tournier P-H, Malagoli P. 2016. The thermodynamic flow-force interpretation of root nutrient uptake kinetics: A powerful formalism for agronomic and phytoplanktonic models.Frontiers in Physiologie 7: 243 Le Ny F, Leblanc A, Beauclair P, Deleu C, Le Deunff E. 2013. In low

transpir-ing conditions, nitrate and water fluxes for growth ofB. napus plantlets cor-relate with changes in BnNrt2.1 and BnNrt1.1 nitrate transporters expression.Plant Signalling and Behaviour 8: e22902.

Leblanc A, Renault H, Lecourt J, Etienne P, Deleu C, Le Deunff E. 2008. Elongation changes of exploratory and root hair systems induced by amino-cyclopropane carboxylic acid and aminoethoxyvinylglycine affect nitrate uptake and BnNrt2.1 and BnNrt1.1 transporter gene expression in oilseed rape.Plant Physiology 146: 1928–1940.

Leblanc A, Segura R, Deleu C, Le Deunff E. 2013. In low transpiring condi-tions, uncoupling the BnNrt2. 1 and BnNrt1. 1 NO3-transporters by gluta-mate treatment reveals the essential role of BnNRT2. 1 for nitrate uptake and the nitrate-signaling cascade during growth.Plant Signaling & Behavior 8. Lejay L, Tillard P, Lepetit M, et al. 1999. Molecular and functional regulation

of two NO3–uptake systems by N-and C-status ofArabidopsis plants. The Plant Journal 18: 509–519.

Lemaire L, Deleu C, Le Deunff E. 2013. Modulation of ethylene biosynthesis by ACC and AIB reveals a structural and functional relationship between the K15NO3 uptake rate and root absorbing surfaces. Journal of Experimental Botany 64: 2725–2737.

Lewis DR, Negi S, Sukumar P, Muday GK (2011) Ethylene inhibits lateral root development, increases IAA transport and expression of PIN3 and PIN7 auxin efflux carriers.Development 138: 3485–3495.

Li W, Wang Y, Okamoto M, Crawford NM, Siddiqi MY, Glass AD. 2007. Dissection of the AtNRT2. 1: AtNRT2. 2 inducible high-affinity nitrate transporter gene cluster.Plant Physiology 143: 425–433.

Li Y, Ouyang J, Wang Y-Y, Hu R, Xia K, Duan J, Wang Y, Tsay Y-F, Zhang M (2015) Disruption of the rice nitrate transporter OsNPF2.2 hinders root-to-shoot nitrate transport and vascular development.Science Report 29; 5: 9635. doi:10.1038/srep09635.

Lieberman M. 1979. Biosynthesis and action of ethylene. Annual Review of Plant Physiology 30: 533–591.

Liepman AH, Olsen LJ. 2004. Genomic analysis of aminotransferases in Arabidopsis thaliana. Critical Reviews in Plant Sciences 23: 73–89. Lin L-C, Hsu J-H, Wan, L-C. 2010. Identification of novel inhibitors of

1-ami-nocyclopropane-1-carboxylic acid synthase by chemical screening in Arabidopsis thaliana. Journal of Biological Chemistry 285: 33445–33456.