Introduction

Structurally, cytokinins (CKs) are derivatives of adenine containing an isoprenoid or aromatic moiety at the N6 position. Typical representatives of isoprenoid CKs are isopentenyl adenine (iP) and its hydroxylated forms zeatin (Z) and dihydrozeatin (DHZ). Z occurs as two isomers, cis-Z (cZ) and trans-Z (tZ), where the cis and trans prefixes refer to the position of the terminal hydroxyl group on the isoprenoid side chain (Gajdosova et al. 2011). Transport of Z-type cytokinins via the transpiration stream and iP types via the phloem is considered to provide root-to-shoot-to-root communication. The bases tZ and iP are considered the active CK forms. However, the role of cZ in plants has long been questioned because of its high concentration but low attributed biological activity. Nevertheless, some recent studies have suggested that it may have unique physiological functions. Notably, it is heavily accumulated under growth-limiting conditions, especially under different plant stress conditions (Veach et al. 2003; Pertry et al. 2009; Schäfer et al. 2015).

In plants, CKs control several biological processes associated with active growth, metabolism and plant development. Among the most important events regulated by CKs are chloroplast biogenesis and chlorophyll synthesis, which profoundly affect the plant’s capacity for photosynthetic assimilation (Yaronskaya et al. 2006). This is important because the photosynthetic production of photoassimilates (i.e. carbohydrates) is essential for plant growth. Higher plants consist of a mosaic of photosynthetically active source tissues such as mature leaves, and photosynthetically less active or inactive sink tissues such as young leaves and roots (Kaschuk et al. 2010). Photosynthesis and carbohydrate utilization by sinks are tightly coordinated. Studies performed with CK-deficient plants have shown that CK deficiency conditions plant growth by reducing the strength of the shoot sink but not that of the sources, leading to dramatic reductions in carbohydrate content, vacuolar invertase activity, and ATP synthesis as consequence of less efficient fluorescence (Werner et al. 2008). However, the source-sink relationship is dynamic; it varies with the strengths of the sinks that compete for a common pool of carbohydrates, and in response to environmental cues (Osorio et al. 2014).

It is well established that endogenous CKs promote cell division in plants, and that their crosstalk with auxins regulates morphogenesis and organ formation (Zhao 2008; Šimura et al. 2016). Crosstalk between hormone signaling and gene expression in plant development is extremely complex, since they form a network in which relevant genes regulate hormone activities and hormones regulate gene expression (Liu et al. 2017). Therefore, the final effect of one particular hormone does not depend solely on its own concentration, but also on the synergistic and antagonistic interactions that arise between all of them (Šimura et al. 2016). Leaf development also requires of hormonal crosstalk to regulate sequential phases of cell proliferation and differentiation (Avramova et al. 2015). Together, these phases define the total number of cells and their sizes, which in turn determine the final leaf size. Several studies have suggested that cell division is also controlled by the levels of photoassimilates such as hexoses (i.e. cleavage products of CK-inducible invertase), which act as mitotic stimuli, whereas sucrose induces differentiation and leads to storage product synthesis (Roitsch and Ehneß 2000). This assumption was recently partially corroborated in pgi1_2, Arabidopsis mutants that present a draft phenotype with reduced starch, soluble carbohydrate, and CK levels (Bahaji et al. 2015).

Finally, CKs have been postulated to regulate nutrient uptake and mobilization by modifying source-sink relationships. For example, nitrogen supplementation increases the CK content of the xylem sap in maize and Arabidopsis (Takei et al. 2001, 2004), indicating that CKs play some role in long distance root-to-shoot signaling of increased nitrogen availability (Tanaka et al. 2006). Conversely, nitrogen deficiency reduces CK production and the rate of export from the roots to the shoot, causing changes in C metabolism. However, the CK levels recover when the original nitrogen supply is restored (Schluter et al. 2012). Furthermore, wheat plants treated with exogenous CKs exhibited modified N remobilization from sources to sinks, and were able to increase sink activity by improving their assimilation capacity and enlarging their chloroplasts (Criado et al. 2009). Finally, the relationship between CKs and nitrate (NO3−) application has been probed in roots. Tian et al. (2005) reported the possible involvement of CKs in nitrate-mediated root growth in maize. In addition, it has been reported that the relative abundance of different CK types changes during organ development, including in maize (Schäfer et al. 2015). Overall, the available data suggest that CKs play roles in N signaling, acting as local and long-distance messengers that regulate leaf morphogenesis based on nitrogen availability (Forde 2002).

Our preliminary results demonstrated that the maize exposed to high nitrate concentrations exhibited reduced aerial and root growth (Saiz-Fernández et al. 2015). High NO3− treatment also conditioned the development of the new leaves by reducing the total number of cells in developing leaves and their sizes without changing the cells’ ontogeny. In the roots, high NO3− levels also reduced root growth. Although these processes were induced by changes in the levels of various phytohormones, including CKs and the ethylene precursor ACC in both shoots and roots (Saiz-Fernández et al. 2015), we hypothesized that the changes in cell division and size, which determine the tissue final growth, might also be mediated by the action of specific CK forms. We therefore studied changes in the relative and absolute abundance of different CK forms in selected source and sink tissues, and their potential interactions with growth inhibition induced by high nitrate concentrations in maize plants.

Materials and methods

Plant material

Plant culture

Maize seeds were placed in moistened paper and kept at room temperature in the dark until the radicle of the seedlings reached 2 cm. Seedlings were then transplanted into 9 L pots (25 cm ∅) filled with perlite-vermiculite (1/1, v/v) and placed in a custom walk-in growth room for the duration of the experiment. Growth conditions were 14-h photoperiod, 500 µmol photons m−2 s−1 light intensity at the canopy, 25/20 °C temperature and 70/80% relative humidity (day/night).

Nitrate treatment

Maize seedlings were irrigated with 500 mL of distilled water three times per week. After 5 days of germination, irrigation was performed using a modified Hoagland solution with an NO3− concentration of 5 or 15 mM, as described by Saiz-Fernández et al. (2015). Any leaked watering solution was discarded from the plates after each watering to avoid the accumulation of nitrate in the pots. The 5 mM of NO3− was chosen as the control treatment based on previous studies that showed it as the optimal NO3− concentration in terms of biomass production (Saiz-Fernandez et al. 2015) and photosynthetic capacity (Saiz-Fernández et al. 2017).

Analysis of phytohormones

Material for phytohormone analysis was collected from 22-day-old maize plants. Samples were divided into different fractions: sink tissues consisted of developing 5th leaves and roots, and source tissues were fully expanded 4th leaves. For each extraction, a tissue sample (5 mg dry weight) was ground and extracted using ice-cold modified Bieleski buffer (methanol/water/formic acid, 15/4/1, v/v/v) containing isotope-labeled CK internal standards (0.5 pmol of CK bases, ribosides, N-glucosides, 1 pmol of CK O-glucosides and nucleotides). The extract was then passed through two SPE columns (C18 and MCX). Endogenous CK levels were determined by ultra-high performance liquid chromatography-electrospray tandem mass spectrometry (UHLC-MS/MS) using a minor modification of the method described by Svačinová et al. (2012).

Abscisic acid (ABA) and indole-3-acetic acid (IAA) levels were determined using the method described by Floková et al. (2014). Briefly, both ABA and IAA were extracted in phosphate buffer with stable isotope-labelled internal standards (10 pmol of each labelled metabolite). After centrifugation, the supernatant was immediately purified on a MAX column and evaporated to dryness prior to chromatographic analysis.

All purified samples were analysed using an ACQUITY UPLC® System and a XevoTM 122 TQ-S triple quadrupole mass spectrometer (Waters). CKs and free IAA and ABA were quantified using multiple monitoring mode (MRM), targeting selected precursor ions and the corresponding product ions. The stable isotope-labelled internal standards were used as references.

Fluorescence parameters

Chlorophyll fluorescence parameters were measured using a Hansatech Fluorescence Monitoring System FMS2 (Hansatech Instruments Ltd., Thain Wildbur, England). To determine the maximum quantum yield of photosystem II (PSII) photochemistry (ΦPo or Fv/Fm), the leaves were dark adapted for 30 min.

$${\Phi _{{\text{Po}}}}={\text{ }}\left( {{F_{\text{m}}} - {F_{\text{o}}}} \right)/{F_{\text{m}}}.$$
(1)

After determining the minimal fluorescence (Fo), a saturating pulse lasting 800 ms was applied to calculate the maximal fluorescence level (Fm). The leaves were then adapted to the light conditions for a further 30 min to calculate the maximal quantum yield of PSII photochemistry for a light-adapted state (ΦPSII):

$$\Phi _{{{\text{PSII}}}} \;{\text{ = }}\;{{\left( {F_{{\text{m}}} ^{\prime } - F_{{\text{o}}} ^{\prime } } \right)} \mathord{\left/ {\vphantom {{\left( {F_{{\text{m}}} ^{\prime } - F_{{\text{o}}} ^{\prime } } \right)} {F_{{\text{m}}} ^{\prime } }}} \right. \kern-\nulldelimiterspace} {F_{{\text{m}}} ^{\prime } }}.$$
(2)

Fo′ is the minimal fluorescence level for the light-adapted state and Fm′ is the maximal fluorescence in the light. These quantities were estimated as described by Humplík et al. (2015). The photochemical quenching (qP) was computed using the following equation:

$$q{\text{P}}\;{\text{ = }}\;{{\left( {F_{{\text{m}}} ^{\prime } - F_{{\text{t}}} } \right)} \mathord{\left/ {\vphantom {{\left( {F_{{\text{m}}} ^{\prime } - F_{{\text{t}}} } \right)} {\left( {F_{{\text{m}}} ^{\prime } - F_{{\text{o}}} ^{\prime } } \right)}}} \right. \kern-\nulldelimiterspace} {\left( {F_{{\text{m}}} ^{\prime } - F_{{\text{o}}} ^{\prime } } \right)}},$$
(3)

Here, Ft is the steady-state fluorescence level immediately before the flash.

The difference between Fm and Fm′ in the irradiance-adapted state permits us to determine the quantum yield of regulatory light-induced heat dissipation using three non-photochemical quenching coefficients; the effective quantum yield of non-photochemical processes in PSII (qN), the complete non-photochemical quenching of chlorophyll fluorescence (qCN), and the non-photochemical chlorophyll fluorescence quenching (NPQ) (Lichtenthaler et al. 2005). Finally, the non-cyclic electron transport rate in the thylakoid membrane (ETR) was calculated as follows:

$${\text{ETR}} = {\text{ PPFD}} \times \alpha _{{\text{L}}} \times f_{{{\text{II}}}} \times \Phi _{{{\text{PSII}}}} .$$
(4)

Here, PPFD is the photosynthetic photon flux density, αL is the fraction of the incident PPFD absorbed by the leaf [a value of 0.845 was assumed for maize during vegetative growth, as calculated by Echarte et al. (2008)], and fII is the fraction of the PPFD absorbed by the light harvesting complex of photosystem II [a value of 0.4 was assumed for C4 plants based on the work of O’Neill et al. (2006) and the previously determined value of ФPSII].

Statistical analysis

For each studied parameter, the differences between treatments in SOURCE (fully expanded 4th leaves) and SINK (developing 5th leaves and roots) maize tissues were compared by non-parametric analysis of variance (Kruskal–Wallis one-way analysis of variance) using version 3.01 of the open source R software package (http://cran.r-project.org/).

The effect of high nitrate levels was also evaluated using multivariate analysis techniques. Principal component analysis (PCA) was performed using data on the studied quantitative variables (the applied nitrate dose, phytomer length, fluorescence parameters, and metabolite levels) in 4th (SOURCE) and 5th (SINK) leaves under the 5 and 15 mM NO3− treatments. We analysed standardized data using the stats and gclus packages of R. Any principal component (PC) that explained less than 10% of the total variation was excluded from the model.

Results

High nitrate levels alter leaf growth and condition the development of later phytomers in maize

The high NO3− concentration significantly reduced the biomass of the aerial parts of maize plants on the 22nd day after germination (Saiz-Fernández et al. 2015), but root growth was not modified until later in the developmental process. The application of high NO3− doses also altered the phytomers’ development, reducing their final size and conditioning the growth of later ones. Based on these results, we sought to determine whether high nitrate levels affected the distribution of biomass between fully expanded 4th phytomer (SOURCE) leaves and developing 5th phytomer (SINK) leaves (Fig. 1). At 22 days after germination, plants treated with 15 mM NO3− exhibited significantly shorter 4th and 5th phytomers than those treated with 5 mM NO3−. This difference was especially pronounced for the 5th phytomer leaves, which were 50% smaller under the 15 mM NO3− treatment (Fig. 1).

Fig. 1
figure 1

The lengths of the 4th and 5th phytomers in maize plants treated with 5 and 15 mM NO3− at 22 days after germination. Statistical significance was evaluated using the Kruskal–Wallis test. ***p < 0.001

Full size image

High nitrate levels affect the tZ- and iP-type CKs more strongly than the cZ-type

To determine how the abundance of different CK metabolites varies with the NO3− concentration, we analyzed several CKs and their derivatives, including precursor compounds and degradation products, in maize plants treated with 5 and 15 mM NO3−. Aromatic CKs were not detected in any tissue, but the levels of isoprenoid types varied among the maize organs. Under both treatments, the most abundant CKs in the leaves were of the cZ-type, followed by the tZ-type, with smaller quantities of DHZ and the iP-type (Fig. 2). Treatment with the high concentration of NO3− did not affect the levels of cZ-type CKs but significantly increased those of tZ- and iP-types in both source (4th leaf) and sink (5th leaf) leaves. In the roots, iP-type CKs were most abundant under both treatments, followed by cZ-type, while tZ-type CKs were less abundant (Fig. 2). The total CK levels in the roots of plants treated with 15 mM NO3− were approximately 4 times greater than those in the roots of plants treated with 5 mM NO3−.

Fig. 2
figure 2

The concentrations of the different CK types (pmol g− 1 DW) in source (4th leaf) and sink (5th leaf) leaves, and roots, of maize plants treated with 5 mM and 15 mM NO3− at 22 days after germination. Statistical significance was evaluated using the Kruskal–Wallis test; results labeled with different letters differ significantly. For the root measurements, asterisks indicate significant differences between the two NO3− treatments. **p ≤ 0.01; ***p ≤ 0.001

Full size image

Differences were also observed among the different groups of CK metabolites. Under the high NO3− treatment, O-glucosides were the most abundant CK form, and accumulated in every tested tissue (i.e. both sink and source leaves and roots), with the highest increment occurring in the roots (Fig. 3a, b). The concentrations of the CK bases also increased in every tested tissue (relative to the concentrations seen under the low NO3− treatment), but those of the ribosides did not change (Fig. 3a, b). However, whereas the levels of N-glucosides increased in the roots under the high NO3− treatment, their concentrations in the leaves remained unchanged or decreased slightly.

Fig. 3
figure 3

The content of the different CK groups (pmol g−1 DW) in source (4th leaf) and sink (5th leaf) tissues (a) and roots (b) of maize treated with 5 mM and 15 mM NO3− at 22 days after germination

Full size image

While the high NO3− treatment did not change the total riboside concentration (Fig. 3), it did cause significant increases in the levels of tZR and iPR in both source and sink leaves—by as much as a factor of 3 in fully expanded 4th leaves (Table 1). The concentrations of the corresponding bases also increased in source and sink leaves. Notably, the concentration of iP in plants treated with 15 mM NO3− was four times that in plants treated with 5 mM NO3−. The levels of the O-glucoside tZOG fell significantly in the 4th leaves, while those of cZROG doubled in both types. The various N-glucosides also exhibited divergent responses to the high NO3− treatment: tZ9G accumulated in both source and sink leaves, whereas iP9G appeared only in sink tissues and the levels of cZ9G were below the limit of detection in both source and sink leaves of 15 mM NO3− plants (Table 1).

Table 1 The concentrations (pmol g−1 DW) of CK derivatives, IAA, and ABA in source (4th leaf) and sink (5th leaf) tissues of maize plants treated with 5 mM and 15 mM NO3− at 22 days after germination
Full size table

In the roots, the concentrations of every tested metabolite increased significantly in the high NO3−-treated plants, except those of iPR, cZR and cZ, which did not change. The highest increase was observed for the cytokinin precursors tZRMP and iPRMP, which were four times more abundant than in maize plants supplied with 5 mM NO3−. The concentrations of these metabolites in the leaves were below the limit of detection (Table 2).

Table 2 The concentrations (pmol g-1 DW) of CK derivatives, IAA, and ABA in roots of maize plants treated with 5 mM and 15 mM NO3− at 22 days after germination
Full size table

High nitrate levels reduce the IAA content of maize leaves and increased that of roots

There is extensive and complex crosstalk between the phytohormones that regulate plant growth and development. To better understand this crosstalk, we analysed the content of IAA in the studied maize tissues. In the leaves, the high NO3− treatment halved the concentration of IAA in both source and sink leaves relative to that in plants treated with 5 mM NO3− (Table 1). However, the levels of IAA in the roots increased significantly under the high NO3− treatment (Table 2).

High nitrate only alters ABA content in maize roots

The growth reduction induced by high NO3− levels suggests that nutrient imbalances impose stress on plants. In addition, it has been reported that CKs play important roles in regulating environmental stress responses through a mechanism that involves extensive interactions and crosstalk with ABA (Ha et al. 2012). We therefore expected that the high NO3− treatment would induce ABA accumulation in all maize tissues. Interestingly, no significant changes in ABA levels were observed in either source or sink maize leaves under the high NO3− treatment (Table 1). However, significant ABA accumulation was observed in the roots of 15 mM NO3−-treated plants (Table 2).

Fluorescence parameters are affected more strongly in source than sink leaves of high nitrate-treated plants

CKs have been reported to stimulate chloroplast biogenesis and chlorophyll synthesis, and to control carbon assimilation in source organs. Moreover, their effects on shoot phenotypes suggest that they might also regulate shoot sink strength (Werner et al. 2008). We thus hypothesized that the smallness of the phytomers in maize plants treated with 15 mM NO3− was due to the effect of CKs on the efficiency of the plants’ photochemical machinery. The high NO3− treatment reduced the fluorescence parameters of the source leaves substantially but had much weaker effects on the sink leaves (Table 3). Every parameter related to the fluorescence efficiency of both dark- and light-adapted leaves (Fm, ΦPo, Fo′, Fm′, ΦPSII), and the electron transport rate (ETR), was significantly reduced in the source leaves of 15 mM NO3−-treated maize. However, in the sink leaves only Fm′ and ΦPSII were reduced; Fo, ΦPo and Fo′ did not change, and Fm increased significantly. Conversely, the values of parameters relating to non-photochemical quenching increased in both source and sink leaves under the high NO3− treatment. The sole exception was qP, which did not vary in source leaves and declined in sink leaves (Table 3).

Table 3 Photosynthetic parameters in source (4th leaf) and sink (5th leaf) tissues of maize treated with 5 and 15 mM NO3− at 22 days after germination
Full size table

High nitrate concentrations change fluorescence parameters in source and sink tissues

To determine whether the different CK metabolites affect the growth inhibition induced by high NO3− concentrations, affect the source-sink relationship, or modulate fluorescence parameters in source and sink leaves, we performed a principal component (PC) analysis (Fig. 4). To facilitate visualization, the results were projected onto the first two principal components (PC1 and PC2), which together captured 85% of the variance and thus almost completely explained the experimental results (supplemental Table 1). PC1 accounted for 55% of the total variation. The variables that loaded on this component included the “nitrate” concentration, the concentrations of the CKs iP and tZ and their ribosides as well as iPR and tZR, and the non-photochemical parameters, all of which correlated directly to the source (4L) and sink (5L) leaves of the plants treated with 15 mM NO3−. These traits were inversely related to the “length” of the leaves and some fluorescence parameters for light-adapted conditions such as the maximal fluorescence level (Fm′ = Fm) and the quantum yield of the PSII photochemistry for the light-adapted state (ΦPSII = Fv..Fm.), the photochemical quenching (qP), the linear ETR, and the 4L and 5L score for 5 mM NO3− treated plants. PC2 captured an additional 30% of the total variance (supplemental Table 1), and was dominated (negatively) by cZ and the minimal fluorescence levels in the dark- and light-adapted states (Fo and Fo′= Fo) as well as the 5L score for 15 mM NO3− treated plants. This PC also included positive coefficients for the concentration of cZR and the score (4L) for both types of maize plants under the 5 and 15 mM NO3− treatments (Fig. 4). These results suggest that the CK metabolites control different biological process in maize plants that condition the sink-source relationship and plant growth.

Fig. 4
figure 4

Scores plot for the PC analysis of the fluorescence parameters for source (4th leaf-4L) and sink (5th leaf-5L) leaves of maize plants treated with 5 mM and 15 mM NO3− at 22 days after germination

Full size image

Discussion

In maize, cZ has been reported to be an active CK (Yonekura-Sakakibara et al. 2004), and its concentration greatly exceeds the combined concentrations of iP-, tZ and DHZ-type CKs in both roots and leaves of seedlings (Vyroubalova et al. 2009; Schäfer et al. 2015). Its most abundant metabolites are cZ-O-glucoside (cZOG) and cZ riboside-O-glucoside (cZROG), which accounted for around 80% of the total CKs in both sink and source leaves (Tables 1, 2). These findings are consistent with previous works analysing CK content in maize (Saleem et al. 2010; Gajdosova et al. 2011). However, we observed that the root samples contained higher levels of iP- than cZ- type CKs (Fig. 2).

Despite cZ type metabolites being the main CKs in maize plants, the levels of their active form did not change appreciably in leaves or roots under the high NO3− doses. However, there were changes in its relative abundance in source (4th leaf) and sink (5th leaf) tissues (Tables 1, 2). Mok et al. (2000) reported that cZ levels in sink leaves are probably maintained by reducing the level of cZROG via the action of the enzyme β-glucosidase, explaining the reduced levels of cis-O-glucosides observed in the sink compared to the source. These differences may have occurred because the 5th leaf was in the linear phase of expansion according to the growth model described by Hillier et al. (2005), whereas the 4th leaf was fully expanded at 22 days after germination (Saiz-Fernández et al. 2015). We therefore hypothesized that the levels of cZ determine the leaf’s developmental stage. This suggestion was corroborated by the PC analysis (Fig. 4). A similar association between cZ and fast-growth developmental stages was reported for Tagetes minuta, suggesting that this metabolite is the main CK controlling plant growth in some developmental stages (Stirk et al. 2005). The low levels of cZ in the source (fully expanded) leaves also supported this result: leaves at this stage of development start undergoing senescence, and a previous study suggested that cZ suppresses the induction of senescence in maize leaves (Behr et al. 2012).

In roots, although cZ and its riboside did not varied in roots under high NO3− doses, there was a high accumulation of the storage forms, cZOR and cZROG, and the catalytic metabolite cZ9G. It would allow maintaining the levels of the base form cZ and, therefore, the normal growth of maize roots. It has been reported that the content of cZ-type CKs changes rapidly during maize seedling growth, and that both cZ catabolism and glycosylation by cZ O-glucosyl transferases work synergistically to fine tune cZ levels during plant development (Zalabák et al. 2014). Thus, this CK and its endogenous levels controlled by both cZ catabolism and glycosylation help regulate development in maize roots and leaves under different nitrogen regimes.

It has been reported that the levels of iP- and tZ-type CKs do not change much during maize development (Zalabák et al. 2014), but they are also reported to be highly sensitive to NO3− availability in Arabidopsis (Kiba et al. 2013) and pumpkin (Nawaz et al. 2017). We observed similar results in maize; treatment with a high concentration of NO3− increased the levels of iP- and tZ and their ribosides in all leaf types, and the levels of their precursors in the roots (Tables 1, 2). This finding was corroborated by the PC analysis, which showed that these metabolites correlated strongly with the nitrate treatment and the fluorescence parameters, especially in source leaves (Fig. 4). These results contradict those of Werner et al. (2008), who found that CK-deficient tobacco plants (35S:CKX1) exhibited reduced photosynthesis-related parameters and altered source-sink relationships. However, a different tobacco line that overproduces CKs (PSAG12-IPT) has also been reported to have a modified source-sink relationship in which the delivery of photoassimilates to sinks is delayed (Cowan et al. 2005). This inconsistency could be explained by the effects of the high NO3− concentration, which has been suggested to modify CK biosynthesis (Sakakibara et al. 2006), and especially the levels of iP- and tZ-type CKs (Takei et al. 2004; Kiba et al. 2011). In addition, in barley low phosphorous conditions can also alter the sink-source relationship by modifying the transport of nitrogen related compounds and enhancing the expression of CKX2 genes (Criado et al. 2017). In our recently published work (Saiz-Fernández et al. 2017), we also reported that 15 NO3− maize plants modified the sink-source by reducing the transport of photoassimilates from leaves to roots. Thus, how the CK level is modified and whether IPT or CKX is targeted may depend on whether the plant is in a source- or sink-limiting environment (Jameson and Song 2016).

It is worth mentioning that the transport of Z-type cytokinins via the transpiration stream and iP types via the phloem is considered to provide root-to-shoot-to-root communication (Jameson and Song 2016). However, in our work, the high concentration of the iP precursors iPRMP detected in the roots and not in the leaves suggested that, at least in this maize variety, the synthesis of iP type CKs occurs mainly in roots.

It has also been reported that auxins can regulate plant growth and development in a tissue-specific and stage-dependent manner (Wu et al. 2016). Their biosynthesis and transport are also altered by changes in the availability of NO3− (Krouk 2016). It was previously proposed that IAA levels in the roots of Zea mays are inversely correlated with the external nitrate concentration (Tian et al. 2008); the authors of that study concluded that the inhibitory effect of high NO3− concentrations on root growth is partly due to a decrease in the auxin concentration in the roots. However, our results indicated an accumulation of IAA in the roots upon treatment with high NO3− concentrations, but without root growth inhibition. In the source and sink leaves, IAA levels decreased in maize plants treated with high NO3− whereas the levels of tZ and iP-type CKs increased. In addition, there were significant differences in IAA levels between mature and developing leaves, suggesting a possible role in regulating leaf growth. In keeping with this hypothesis, LaMotte et al. (2002) observed that the levels of IAA and ABA differed in the leaves of Coleus blumei, with lower levels in old leaves than in younger ones. Furthermore, it is well-known that the CK:auxin ratio controls shoot and root meristem development (Su et al. 2011). Altogether, these results suggest that in maize leaf development depends on the homeostasis between ABA, IAA and the different CK-types, and their interaction with nitrate doses.

Finally, as described by LaMotte et al. (2002), there is an interesting parallelism between the behaviours of ABA and IAA in the leaves of maize plants (Table 1). However, this parallelism was absent in plants treated with high NO3− concentrations, which exhibited no significant differences in the ABA concentrations of young and old leaves. This was another unexpected result because high NO3− induced a stress growth condition that limited leaf development (Fig. 1; Table 1). However, it is consistent with earlier work performed in Arabidopsis, where there was no significant difference in the ABA contents of seedlings grown on low- and high-nitrogen media (Kiba et al. 2011). Interestingly, the ABA level in the roots of plants treated with 15 mM NO3− was three times higher than in the controls (Table 2). Maybe, the high ABA levels reduced the production of 1-aminocyclopropane-1-carboxylic acid (ACC) and, therefore, ethylene in plants with high NO3− supply (Saiz-Fernandez et al. 2015), as suggested by (Sharp 2002). This accumulation may inhibit root growth if the high NO3− regime is sustained, as described by Saiz- Fernández et al. (2015). An inhibitory effect of NO3− in roots has also been reported for Arabidopsis (Signora et al. 2001).

In summary, we propose for the first time that under high NO3− conditions, changes in the levels of specific metabolites induce distinct effects. In maize, cZ and IAA control the transition between different leaf developmental stages, but the changes in the rates of expansion induced by the high NO3− treatment seem to be attributable to the influence of IAA together with iP and tZ-type CKs (Fig. 5). The cZ content determined via catabolism and glycosylation also define root growth under different nitrate conditions, along with IAA and ABA. Further studies should be conducted on a wider range of species and developmental stages to determine whether this behavior is general or species-specific.

Fig. 5
figure 5

Proposed model describing the regulation of development in maize tissues by phytohormones and the response to high NO3− concentrations

Full size image