Introduction

Plant root architecture displays a considerable degree of plasticity in response to the heterogeneous distribution of soil resources (Hodge 2004; Malamy 2005; Nibau et al. 2008). In particular, the availability and distribution of nitrate, the main source of nitrogen (N) for plants growing in aerobic soils, influences both the carbon investment in, and the architecture of, plant root systems (Lynch et al. 2012). In recent decades, researchers have developed various laboratory-based technologies for studying the root systems of seedlings grown in non-soil or soil media (Armengaud et al. 2009; Iyer-Pascuzzi et al. 2010; Clark et al. 2011, 2013; Downie et al. 2012; Gruber et al. 2013; Shi et al. 2013; Adu et al. 2014). These techniques have revealed how root architectures of different plants respond to the heterogeneous distribution of nutrients in the rooting medium (Drew 1975; Robinson 1994; Hodge 2004, 2006; Hodge et al. 2009; Li et al. 2014) and provided some knowledge of the molecular mechanisms coordinating these processes (Zhang and Forde 1998, 2000; Forde and Walch-Liu 2009; Ruzicka et al. 2010; Ruffel et al. 2011; Giehl et al. 2013; Kellermeier et al. 2013; Mounier et al. 2014). However, although breeding for root traits could play an unprecedented role in improving crop establishment and increasing yield (Lynch 2007, 2013; Hammer et al. 2009; Gewin 2010; White et al. 2013a, b; York et al. 2013), limited attention has been paid to genotypic responses of root architecture to the heterogeneous availability of soil resources in crop breeding programs. One reason for this might be that roots are entangled belowground and studying the architectures of root systems is very time consuming, which limit both the accuracy and speed of measuring root traits in the field (Waines and Ehdaie 2007; Gewin 2010; Herder et al. 2010). Irrespective of the great progress in documenting and understanding the phenotypic plasticity of plant roots in the laboratory, there are still various aspects of root growth and nutrient acquisition, especially of cereal crops, that must be addressed in the field.

An understanding of genotypic variation in the responses of root system architecture to the heterogeneous distribution of nutrients in the field will determine the potential for the exploitation of such traits to increase crop yields and maximize nutrient-use efficiency. Maize, which is a classic model cereal plant and a staple crop in many areas of the world, forms a complex root system composed of several different root types (Hochholdinger et al. 2004; Hochholdinger and Tuberosa 2009). This paper reviews the aspects of phenotypic plasticity of maize root system architecture with emphasis on describing (1) the development of its complex root system; (2) phenotypic responses in its root system architecture to heterogeneous N availability; (3) the importance of phenotypic plasticity in root system architecture for N acquisition; (4) different regulation of root growth and nutrient uptake by shoot; and (5) root traits in maize breeding.

Development of the maize root system

Maize plants possess an embryonic root system consisting of a single primary root and a variable number of seminal roots and a post-embryonic root system consisting of several whorls of shoot-borne roots (Hochholdinger et al. 2004). The root axes of all seminal roots and shoot-borne roots are called axial roots (ARs), and the roots initiated from these ARs are called lateral roots (LRs). There are also higher-order LRs initialed from lower-order LRs. The maize root system develops in an orderly manner that is coordinated with the development of the shoot, starting with the emergence of the primary root, then seminal and shoot-borne roots that subsequently branch to form different orders of LRs (Foth 1962; Hanway 1963; Hoppe et al. 1986; Feldman 1994; Fig. 1). The LRs are initiated from pericycle and endodermal cells close to the phloem poles in the differentiation zone of roots (Bell and McCully 1970; Esau 1977; Fahn 1990; Casero et al. 1995; Jansen et al. 2012). In most cases, some of the first-order LRs and ARs become leaders that elongate rapidly, persist for a long time, and thicken with time (Varney et al. 1991). These roots comprise most of the biomass of the root system and form the long-distance transport pathways and branch-root junctions (McCully and Canny 1988; Varney et al. 1991; Shane et al. 2000). The higher-order LRs are usually finer roots that make up most of the length and surface area of the root system, and are essential for water and nutrient uptake (Varney et al. 1991; McCully 1995, 1999; Sullivan et al. 2000). The diameter of individual roots varies widely. The ARs are thicker roots, which are able to exert greater force to penetrate compacted soil (McCully and Canny 1988; McCully 1995; Waisel and Eshel 2002; Hund et al. 2009a) and determine both the direction of growth of the root system and its spatial distribution in the soil (Eissenstat 1997).

Fig. 1
figure 1

Development of the maize root system from embryonic roots in seedlings to crown and brace roots initiated from shoot nodes in adult plants. a The root system of a 35-day-old maize plant with the major root types indicated. b The root system of a 49-day-old maize plant with its shoot-borne roots from the fifth node. c Roots emerging from shoot nodes on a 49-day-old maize plant. d Phenotypic responses of the shoot-borne roots from the fifth node of a 49-day-old maize plant to a non-uniform N environment (LoHN, local high nitrate supply: HN 4 mM, Control 0.5 mM). e Phenotypic responses of the shoot-borne roots from the seventh node of a 70-day-old maize plant in response to non-uniform N environments. f All root types of a 70-day-old maize plant grown with a sufficient N supply

Full size image

Embryonic roots of maize are functionally important during early growth, but their function is soon replaced by post-embryonic shoot-borne roots (Varney and McCully 1991; Feldman 1994; Feix et al. 2002; Fig. 1). Nevertheless, if the development and growth of the shoot-borne roots of maize plants is restricted, then the remaining seminal roots will show compensatory growth and increased nutrient uptake compared to those of control plants (Jeschke et al. 1997; Xu et al. 2009; Yan et al. 2011a), suggesting that the longevity and importance of different root classes can alter depending on the integrated functioning of the entire root system.

Post-embryonic shoot-borne roots include crown roots (CRs; Fig. 1a, b, c; Hochholdinger et al. 2004) and brace roots (BRs; Fig. 1d, e; Feldman 1994). The CRs start to emerge from below-ground stem nodes 5–10 days after germination and BRs emerge from above-ground stem nodes approximately 6 weeks after germination (Hochholdinger et al. 2004). The number and diameter of CRs are greater on higher shoot nodes (Fig. 1b, c). The growth of the BRs results in dramatic increases in both the total length of the root system (TRL) and its dry weight (DW), and is contemporaneous with a rapid increase of shoot growth and demand for nutrients (Hoppe et al. 1986; Niu et al. 2010; Peng et al. 2012a). Although 35–55 % of the N in maize plants at harvest is taken up by the root during the reproductive stage (Hirel et al. 2007), the TRL of the maize root system reaches its highest value at silking (Durieux et al. 1994; Peng et al. 2012a). This is consistent with studies on many cereals in which it has been observed that the DW of the whole root system does not increase after flowering and in some studies a substantial decrease has been observed during grain filling (reviewed by Gregory 2006). Although the TRL of the maize root system decreases markedly after silking, there is only a small change in the DW of the root system (Peng et al. 2012a). This is because LRs comprise only a small portion of the total root DW, but constitute the majority of the TRL (Walk et al. 2005; Hund et al. 2009b; Peng et al. 2010).

The decline in total length of LRs after silking is primarily due to root mortality. Root mortality during the reproductive stage has been associated with the translocation of N and carbon from roots to the developing ear of plants grown in the field (Wiesler and Horst 1993). The rapid reduction in TRL of maize plants grown in the field might be explained by the influence of soil biota. The presence of microorganisms, decomposers, herbivores and parasites in the rhizosphere could accelerate root mortality, especially when roots receive less carbon from the shoot during the reproductive stage (Eissenstat and Yanai 1997; Eissenstat et al. 2000; Watt et al. 2006a, b). Interestingly, when maize plants are grown in quartz sand or nutrient solution their root DW and TRL increase continuously until maturity (Warncke and Barber 1974) even when they suffer from N deficiency, which indicates that their roots do not exhibit ontogenetically programmed mortality during the reproductive stage (Niu et al. 2010; Yu et al. 2014). Overall, the developmental and environmental signaling that co-regulates root growth and development allows plants to optimize the placement of roots within root system adapted to complex and flexibly changing soil environments (Malamy 2005).

Phenotypic plasticity of maize root responses to nitrogen supply

Root architecture is controlled both by intrinsic developmental programs and by complex interactions between the plant physiological status and exogenous biotic and abiotic stimuli in the rhizosphere (McCully 1995, 1999; Rich and Watt 2013). The phenotypic plasticity of plant root systems is manifest in the ability of plants to optimize root system architecture for resource acquisition under diverse environmental conditions in response to the identity, availability and distribution of soil resources (Fitter 1994; Robinson 1994; Hodge et al. 1999; Fitter 2002; Hodge 2004, 2006; Shemesh et al. 2010; Croft et al. 2012; Gruber et al. 2013). The stimulation of LR development by nitrate-rich patches is a classic example of a nutrient-induced alteration in root system architecture (Drew et al. 1973; Drew and Saker 1975; Granato and Raper 1989; Schortemeyer et al. 1993; Forde and Lorenzo 2001; López-Bucio et al. 2003; Forde and Walch-Liu 2009; Giehl et al. 2013). In the short-term, systemic N deficiency in maize seedlings increases carbon partitioning to roots, accelerates root growth, and results in fewer, longer ARs with longer LRs (Pan et al. 1985; Feil et al. 1990; Eghball and Maranville 1993; Barber 1995; Gaudin et al. 2011) as is also observed in other plant species (Hermans et al. 2006; Giehl et al. 2013; Gruber et al. 2013). However, the lengths of ARs of the 60-day-old (adult) maize plants respond less dramatically than those of seedlings to systemic N deficiency (Gaudin et al. 2011; Orman-Ligeza et al. 2013; Yu et al. 2014). By contrast, although nitrate-rich patches only increase the length of LRs from the treated roots of maize seedlings, they increase both the length and density of LRs from shoot-borne roots of adult maize plants (Peng et al. 2012a; Yu et al. 2014). Thus, the mechanisms controlling the responses of embryonic roots and shoot-borne roots to nitrate-rich patches appear to differ (Hochholdinger and Tuberosa 2009; Zhu et al. 2011) and merit further investigation. In seedlings, a transient IAA increase in root tips of nitrate-fed root segments is observed 2 days after the treatment (Sattelmacher and Thoms 1995). Auxin and auxin transport inhibitor application revealed the pivotal roles of auxin transported from shoot to roots in response of LR growth to localized supply of nitrate in maize (Wang et al. 2004; Guo et al. 2005; Liu et al. 2008, 2010).

Another general response of roots of maize to systemic N deficiency is to increase rooting depth (Gastal and Lemaire 2002; Dunbabin et al. 2003, 2004). Root growth angles of maize plants become steeper when they are grown with a reduced N supply (Gaudin et al. 2011; Lynch 2013; Trachsel et al. 2013). Conversely, excessive N supply inhibits root growth and produces a shallower root system (Durieux et al. 1994; Walch-Liu et al. 2006; Gaudin et al. 2011). The N concentration in shoot tissues affects the root responses to the availability and distribution of nitrate in the soil, implicating long-distance signals from the shoot to the root in these responses (Zhang and Forde 1998, 2000; Zhang et al. 1999; Hermans et al. 2006; Desnos 2008; Ruffel et al. 2008; Alvarez et al. 2012).

While many studies have focused on the morphological responses of roots within nitrate-rich patches, less attention has been paid to morphological responses of roots outside these patches. However, recent studies have demonstrated that, whilst the root length density (RLD; cm cm−3) and specific root length (SRL; m g−1) of the root system within nitrate-rich patches are increased, the development and growth of LRs outside the nitrate-rich patches is reduced, which suggests a systemic carbon partitioning strategy within a whole root system both in Arabidopsis and maize plants (Mounier et al. 2014; Yu et al. 2014). It is known that CHL1 functions upstream of ANR1 in regulating the proliferation of LRs in response to a locally high nitrate supply; there are also some additional components involved in ANR1 function regulated by high nitrate supply (Zhang and Forde 1998; Remans et al. 2006). It is hypothesized that, at low-nitrate concentrations, CHL1 represses the growth of LRs by promoting basipetal auxin transport (Krouk et al. 2010) and acts locally to modulate auxin levels and meristematic activity in Arabidopsis plants (Mounier et al. 2014). Alternatively, or additionally, the increased N concentration in shoots resulting from a localized high N supply might initiate a ‘systematic inhibitory effect’ on the development and growth of LRs outside nitrate-rich patches (Mounier et al. 2014; Yu et al. 2014). The NO-signaling pathway also plays a role in the responses of maize roots to nitrate availability (Trevisan et al. 2011; Manoli et al. 2014), and it is thought that peroxisomal NO modulates auxin-induced lateral root formation both in Arabidopsis and maize (Schlicht et al. 2013). Microarray analyses revealed that early-responsive genes related to cell division and expansion such as α-expansin, cellulose synthase, kinesin, plasma membrane and tonoplast aquaporins are possibly involved in localized nitrate stimulation of lateral root development in maize (Liu et al. 2008). But so far, it remains unclear what triggers LR elongation and initiation to deploy nitrogen in the heterogeneous nutrients medium in maize.

The importance of root phenotypic plasticity in N capture

Our knowledge of both the physiology and molecular biology of root N acquisition has improved greatly over the past few decades (Crawford 1995; Crawford and Glass 1998; Miller et al. 2007; Gojon et al. 2009; Gojon 2013; Nacry et al. 2013). The role of root architecture in N acquisition has also been studied widely. The development of a plant root system continuously extends the rhizosphere periphery into the soil, and the resulting root architecture largely determines the efficiency of nutrient acquisition (Lynch 1995; Linkohr et al. 2002; Zhu et al. 2005). Root length per unit of soil volume (RLD) is an important factor for nutrient acquisition (Robinson and Rorison 1983; Wiesler and Horst 1994; Garnett et al. 2009; White et al. 2013a). Increasing the total length of LRs results in greater N acquisition (Linkohr et al. 2002; Wang et al. 2004, 2006) and fine roots take up more N than thicker roots on a dry weight (DW) basis (Sullivan et al. 2000). As mentioned in the previous section, LRs proliferate in regions of the soil with higher N availability.

In addition to root architecture, the characteristics of nitrate uptake by root cells (i.e., their uptake capacity V max and affinity for nitrate K m) also play an important role in N acquisition. When roots encounter a local region of high nitrate supply, their N uptake capacity increases transiently by up to 75 % compared to regions of the root outside the nitrate-rich patch (Jackson et al. 1990; Robinson 1994; van Vuuren et al. 1996; Hodge 2004). In maize, three nitrate transporters have been characterized: ZmNrt1.1 encodes a low-affinity nitrate transporter expressed in both root and shoot tissues (Quaggiotti et al. 2004), ZmNrt2.1 encodes a high-affinity nitrate transporter that is predominantly expressed in the epidermis and cortex of roots and is implicated in high-affinity NO3 − uptake, and ZmNrt2.2 encodes a constitutive and inducible high-affinity nitrate transporter that is expressed in the cortex and central cylinder of the root (Santi et al. 2003; Trevisan et al. 2008). Transcription of ZmNrt2.1 and ZmNrt2.2 is 1000-fold higher than that of ZmNrt1.1A, ZmNrt1.1B and ZmNrt1.2; even when plants are grown with an optimal N supply, and basal transcription of ZmNRT2.1 and ZmNRT2.2 is much higher than other ZmNRT2 and ZmNRT1 genes over a plant’s life cycle (Garnett et al. 2013). Curiously, however, the expression of ZmNRT2.1 and ZmNRT2.2 in roots of maize plants within nitrate-rich patches is suppressed, and the nitrate uptake capacity of roots within these patches is lower than in those outside the patches (Quaggiotti et al. 2003). Increased N uptake by maize roots in nitrate-rich patches is achieved mainly by changes in root system architecture, which is observed not only in seedlings, but also in adult plants (Yu et al. 2014). Because nitrate is highly mobile in the soil profile, due to its high water solubility, morphological plasticity of high-order laterals of adult maize plants is beneficial for taking up nitrate leached to deeper soil layers and, thereby, improving crop N use efficiency and preventing environmental pollution (Lynch and Brown 2001; Peng et al. 2012a; White et al. 2013a, b). Further studies demonstrated that deep root genotypes are important for deep soil exploration and that low crown root numbers in maize tend to deserve consideration as a potential trait for genetic improvement in low N soil (Lynch et al. 2012; York et al. 2013; Saengwilai et al. 2014).

The investigation on temporal and spatial dynamics of root distribution and nutrient depletion in the soil is imperative (Kuchenbuch et al. 2009; Buczko and Kuchenbuch 2013). Roots are generally more abundant in the topsoil than in deeper soil horizons, and this distribution corresponds primarily to the availability of phosphorus and other immobile soil nutrients (Lynch and Brown 2001; Fitter 2002; Zhu et al. 2005; White et al. 2013a). The relationship between root distribution and soil depth can be described by a simple asymptotic function in most cases (Jackson et al. 1996) and the length and mass of roots decreases exponentially with depth for many crops (Gerwitz and Page 1974; Robertson et al. 1993; Oikeh et al. 1999; Zhuang et al. 2001). A close correlation of RLD of catch crops in the sub-soil layer with nitrate depletion is reported in the field (Thorup-Kristensen 2001, 2006; Kristensen and Thorup-Kristensen 2004) and similar trends with soil residual mineral nitrogen (Nmin) in the whole soil profile in maize (Wiesler and Horst 1994; Peng et al. 2010). In the upper 0.1 m of soil, typical values of RLD are about 20 cm cm−3 in grasses, 5–10 cm cm−3 in temperate cereal crops, and 1–2 cm cm−3 in other crops (reviewed by Jackson et al. 1996). It is clear that root systems with greater RLD in the topsoil horizon can acquire more of the N fertilizer applied and reduce the movement of water and nitrate toward deeper soil layers (Dunbabin et al. 2002; Lynch 2013; White et al. 2013b).

Different regulation of root growth and nutrients uptake by shoot

The balance between root growth and shoot photosynthesis accumulation should be coordinated in crop production and vary in relation to factors such as genotypic differences and resource supply. In annual crops, the allocation of dry matter to roots changes during their life cycle. Typically, proportionally more assimilates are allocated to roots during early stages of plant growth. After flowering, a large proportion of recent assimilates are translocated to the growing reproductive structures and the proportion of assimilates translocated to roots decreases (Brown and Biscoe 1985; Snapp and Shennan 1992; Wells and Eissenstat 2003). Data from many studies indicate that the root: shoot DW ratio (R/S) of maize decreases from about 0.68 at emergence (9 days after planting) to 0.16 at physiological maturity (137 days after planting) (Amos and Walters 2006). During the vegetative growth period, the increase in RLD is synchronized with shoot growth and demand for nutrients, and is correlated with fluctuations in carbon flux from shoot to roots (Rajcan and Tollenaar 1999; Gallais and Coque 2005; Hirel et al. 2007). During reproductive growth, reduced carbon flux from shoot to roots, and accelerated root mortality, results in a rapid decrease in R/S DW ratio (Wiesler and Horst 1993; Wells and Eissenstat 2003; Ogawa et al. 2005).

The amounts of mineral nutrients taken up by the root system are coordinated with shoot growth and demand for nutrients. Although covering or removing ears of maize leads to greater allocation of carbon and nutrients to the root, and results in a larger root system and a higher R/S ratio, less nutrients are taken up after these treatments compared with control plants (Christensen et al. 1981; Ceppi et al. 1987; Yan et al. 2011b; Ning et al. 2012). These results demonstrate that the growth and absolute size of the maize root system is not necessarily correlated with the amount of nutrients it acquires. Root growth is often determined by the amount of carbon translocated from the shoot, whilst the amounts of mineral nutrients taken up is often determined by shoot demand or shoot growth potential (Ma and Dwyer 1998; Wang et al. 2006; Coque et al. 2008; Yan et al. 2011b; Ning et al. 2012).

It is interesting to observe that post-silking uptake accounts for 16–43 % of the total N and 16–55 % of the total P in maize plants at maturity, whilst there is little accumulation of potassium (K) post-silking. The ratio of grain K content to whole plant K content was the lowest compared with the other two macro-elements N and P at maturity. Indeed, there can even be a net loss of K in the post-silking period (Peng et al. 2012b; Ciampitti et al. 2013; Ning et al. 2013). It is believed that the asynchronous uptake of N, P and K is closely related to their functions in plants. A requirement for K to maintain enzyme activities and cell turgor underlies vigorous vegetative growth (Hawkesford et al. 2012), which might explain the preferential accumulation of K pre-silking, whereas N and P are required in large amounts for the formation of new tissues and the accumulation of seed reserves during post-silking grain development. Also K is important for the loading of sucrose and the rate of the mass flow-driven solute transport in the sieve tubes of the phloem (Hawkesford et al. 2012). There is a net remobilization of N and P from vegetative to reproductive tissues post-silking (Ning et al. 2012, 2013).

Plasticity in the partitioning of carbon between shoot and roots also depends upon external environmental factors. In the short-term, plants experiencing nutrient deficiency can increase resource acquisition by increasing their R/S DW ratio (Ågren and Ingestad 1987; Aikio and Markkola 2002; Hermans et al. 2006). This plasticity of carbon allocation to compensate for limitations in nutrient uptake is modulated by species-specific developmental patterns (Siddique et al. 1990; Gedroc et al. 1996), and is negatively correlated with the concentrations of mineral nutrients in leaves (Scheible et al. 1997; de Groot et al. 2003; Hermans et al. 2006). However, greater carbon allocation to roots will restrict shoot growth, which can ultimately result in reduced grain yield. Optimized nutrient supply in crop production is therefore important for maximizing grain yield and, since agronomic nutrient-use efficiency is the quotient of grain yield and nutrient supply, improving nutrient-use efficiency.

Root traits in maize breeding

Although roots play a significant role in increasing nutrient acquisition, limited attention has been paid to root traits in crop breeding during the past decades, possibly because roots grow belowground and are difficult to study (Gewin 2010; White et al. 2013b). Plant breeders usually make considerable gains on ‘aboveground’ traits and tremendous genetic variations trapped in roots have been neglected (Duvick 2005). The development of new crop varieties with enhanced soil resource acquisition is an important strategy for global grain security (Lynch 1998, 2007; Vance et al. 2003; White et al. 2013b). Improvement in the ability of plants to tolerate stress conditions, rather than increases in primary productivity perse, has been the primary driving force for increasing grain yield in maize breeding in recent decades (Duvick 2005; Fischer and Edmeades 2010). The ‘Green Revolution’ succeeded in increasing maize yields using conventional germplasm to develop disease-resistant F1-hybrid varieties, whose yields responded positively to irrigation and fertilizer applications without lodging (Evenson and Gollen 2003; Godfray et al. 2010). Increased root DW in new maize varieties plays an important role in providing resistance to lodging at high plant densities (Duvick and Cassman 1999; Duvick 2005; Echarte et al. 2008). In addition, the deeper roots of the new varieties enable maize plants to access more water and soluble nutrients, such as nitrate, in deep soil horizons (Dunbabin et al. 2003; Hammer et al. 2009; Lynch 2013). Hammer et al. (2009) modeled the relationship between root architecture and grain yield for maize varieties released in the US over the last century and concluded that improvements in root system architectures and, in particular, water capture through breeding have increased both biomass accumulation and crop yields. Study on the three-dimensional distributions of roots from maize varieties released in different eras have demonstrated that new maize varieties have larger root DW, but similar TRLs and vertical root distributions, to older varieties at silking, when the root system of maize attains its largest size, but new maize varieties have greater TRL and deeper roots than older varieties at maturity. The latter phenomenon was the result of more growth of new roots and/or less root mortality after silking in the new varieties, especially of roots in the topsoil (Ning et al. 2014). Irrespective of similar TRLs in both new and older maize varieties at silking, new varieties accumulate more mineral nutrients than older varieties. This is likely to be partly a consequence of greater shoot demand, rather than simply an effect of root size per se, as discussed in the preceding section (Clarkson et al. 1988; Imsande and Touraine 1994; Peng et al. 2010; Yan et al. 2011b). Fortunately, some ongoing breeding and genetic improvement programs have been implemented gradually by the geneticists and agronomists in characterizing promising root traits for enhancing nutrient acquisition and increasing the ability of stresses tolerance (Tuberosa et al. 2003; Hund et al. 2004, 2011; Zhu et al. 2005, Trachsel et al. 2009; Ruta et al. 2010; Lynch 2013; Rose et al. 2013; White et al. 2013a, b; Postma et al. 2014; Saengwilai et al. 2014).

Concluding remarks and future perspectives

Much is known about how root architecture responds to heterogeneous nutrient availabilities in plants of different species and of contrasting nutritional status under controlled environmental conditions, and the molecular mechanisms coordinating these responses are beginning to be elucidated in model plants, such as Arabidopsis (reviewed by Casimiro et al. 2003; López-Bucio et al. 2003; Hardtke 2006; Desnos 2008; Benková and Bielach 2010; De Smet 2012; Ubeda-Tomás et al. 2012). However, little is known about the responses’ root architecture to heterogeneous nutrient availability under field conditions (reviewed by Hochholdinger et al. 2004; Hochholdinger and Tuberosa 2009; Smith and De Smet 2012; Orman-Ligeza et al. 2013).

While nitrate-rich patches increase the length of LRs from the treated roots of maize seedlings, they increase both the length and density of LRs of shoot-borne roots that form adult maize plants at reproductive growth stage, when TRL begins to decrease dramatically under field conditions (Peng et al. 2012a; Yu et al. 2014). The possible regulatory mechanism underlying the markedly increase in both the length and density of LRs initiated from shoot-borne roots is unknown. In Arabidopsis, systemic LP supply also causes an increase in length and density of LRs. Expressional induction of the auxin receptor TRANSPORT INHIBITOR RESPONSE1 (TIR1) in pericycle cells promotes LRs initiation and emergence; as a result, root density is significantly enhanced in P-deficient roots (Pérez-Torres et al. 2008). Since modern maize varieties take up more post-silking nutrients than old ones, more attention should be paid to the genetic improvements and novel adaptive mechanism of maize shoot-borne roots, especially the later initiated ones to resource deployment (Gaudin et al. 2011; Saengwilai et al. 2014; Yu et al. 2014).

Although the amount of nutrients taken up is decided by shoot demand, a larger and deeper root system is apparently beneficial for increasing the contact with the soil and thus nutrients uptake; and intercept/take up the nitrate leached to deeper soil layers. Knowledge of root development and the phenotypic plasticity of root system architecture to nutrient availability will help harnessing the potential for manipulating root growth in the field to exploit the biological potential of root system on one hand, and breeding roots with more efficient acquisition of soil resources on the other hand, and thereby maximizing the nutrient-use efficiency and reducing N losses in maize production.

Approaches toward the understanding of the changing demands of crops for nutrients over the growing season, and the temporal and spatial heterogeneities in the availability of mineral nutrients in the field, enable us optimizing root zone management to synchronize crop growth demand, root resource acquisition and fertilizer application throughout the crop growing season (Peng et al. 2012a; Shen et al. 2013).