Elucidating the Possible Involvement of Maize Aquaporins and Arbuscular Mycorrhizal Symbiosis in the Plant Ammonium and Urea Transport under Drought Stress Conditions

This study investigates the possible involvement of maize aquaporins which are regulated by arbuscular mycorrhizae (AM) in the transport in planta of ammonium and/or urea under well-watered and drought stress conditions. The study also aims to better understand the implication of the AM symbiosis in the uptake of urea and ammonium and its effect on plant physiology and performance under drought stress conditions. AM and non-AM maize plants were cultivated under three levels of urea or ammonium fertilization (0, 3 µM or 10 mM) and subjected or not to drought stress. Plant aquaporins and physiological responses to these treatments were analyzed. AM increased plant biomass in absence of N fertilization or under low urea/ ammonium fertilization, but no effect of the AM symbiosis was observed under high N supply. This effect was associated with reduced oxidative damage to lipids and increased N accumulation in plant tissues. High N fertilization with either ammonium or urea enhanced net photosynthesis (AN) and stomatal conductance (gs) in plants maintained under well-watered conditions, but 14 days after drought stress imposition these parameters declined in AM plants fertilized with high N doses. The aquaporin ZmTIP1;1 was up-regulated by both urea and ammonium and could be transporting these two N forms in planta. The differential regulation of ZmTIP4;1 and ZmPIP2;4 with urea fertilization and of ZmPIP2;4 with NH4+ supply suggests that these two aquaporins may also play a role in N mobilization in planta. At the same time, these aquaporins were also differentially regulated by the AM symbiosis, suggesting a possible role in the AM-mediated plant N homeostasis that deserves future studies.


Introduction
Nitrogen (N) is one of the most important macronutrients for plants, being required in significant amounts as is a constituent of nucleic acids, amino acids, proteins, lipids, co-enzymes, chlorophyll, phytohormones, and secondary metabolites [1]. However, the availability of this nutrient is very variable in soils, and losses in chemical fixation and leaching lead to deficiency in many agricultural lands, which at the end limit plant growth. Plants absorb N from soils mainly as nitrate (NO 3 − ) or ammonium (NH 4 + ). Nitrate is predominant in aerobic soils, while ammonium dominates in high-affinity urea transporters, like ZmDUR3, confirmed the hypothesis of active urea uptake and transport in plants [8] Nevertheless, urea toxicity is still under debate, and similar effects to NH 4 + were also reported [9]. Arbuscular mycorrhizal symbiosis plays important roles in nutrient acquisition by the majority of land plants. This symbiosis is mainly known by the enhanced uptake of phosphorous (P), but these fungi can also transfer N to their hosts, although their contribution is still under debate [10]. However, several studies suggest that P and N uptake and transport are tightly related in the AM symbiosis, and they could be controlling mycorrhizal functioning [11]. There are mainly evidence of ammonium and nitrate being taken up by the AM fungi, although they can also increase access to organic N forms from the soil [12,13]. However, the flow of N within the fungal hyphae would also involve the urea cycle, which produces urea and is further transformed in ammonium which would be transferred to the host plant [14]. LjAMT2;2, a plant high-affinity ammonium transporter was the first characterized to be involved in the uptake of N during arbuscular mycorrhizal symbiosis. It is localized in the periarbuscular membrane, and was only expressed in arbusculated cells of mycorrhizal roots, releasing ammonia in the cytoplasm of the root cells [15]. Several other ammonium transporter genes were identified in different plant species [16].
Recent evidence suggests that the affinity of the AM fungi for NH 4 + uptake is five times higher than NH 4 + uptake by plant systems, which means that the fungus may take this nutrient even under low N supply [11]. Moreover, it was demonstrated in mycorrhizal maize plants that the ability of the AM fungus to transfer N to the plant differs depending on the N source. In fact, ammonium was transferred at an equivalent rate as P, while nitrate seemed to not be transported [17]. AM fungi prefer the direct uptake of ammonium instead of nitrate due to the extra energy needed for the reduction of nitrate, required for the incorporation of N into organic compounds [16]. Apart from specific ammonium or urea transporters, plants contain an array of aquaporin isoforms (AQPs), which are ubiquitous membrane channels facilitating the passive flux of water and a range of small solutes [18]. Among the diversity of plant aquaporin substrates, urea and ammonium were found to be transported in heterologous systems [5,19,20]. Thus, it can be speculated that these proteins have a role in nitrogen mobilization in plants. In fact, differences in AQP gene expression or posttranslational modifications were found in different studies depending on N availability and form [21], being also suggested that the response is dependent on the plant variety [5]. In the case of maize, different aquaporins regulated by the AM symbiosis under drought stress were demonstrated to transport both N forms in yeast assays [22,23].
Plants 2020, 9, 148 3 of 17 Herein, this study aims to investigate the possible involvement of AM-regulated aquaporins in the transport in planta of ammonium and/or urea under well-watered and drought stress conditions. Additionally, the study also attempted to better understand the implication of the AM symbiosis in the uptake of different N forms (urea and ammonium) and its effect on plant physiology and performance under drought stress conditions.

Results
In this study, the principal components analysis (PCA) done for data obtained in the ammonium and urea experiments (Figures 1 and 2) showed that axes PC1 and PC2 explained 51.9% of data variability in the ammonium experiment and 60.6% of data variability in the urea experiment. In both cases, the main factor affecting the data distribution is the N level applied. Thus, data of high N treatments distribute along the right side of vertical axis and data of no N and low N treatments distribute along the left side of this axis. Moreover, ANOVA results were included as supplementary material in Tables S1 and S2. Plants 2020, 9,148 3 of 19 AM symbiosis in the uptake of different N forms (urea and ammonium) and its effect on plant physiology and performance under drought stress conditions.

Results
In this study, the principal components analysis (PCA) done for data obtained in the ammonium and urea experiments ( Figure 1 and Figure 2) showed that axes PC1 and PC2 explained 51.9% of data variability in the ammonium experiment and 60.6% of data variability in the urea experiment. In both cases, the main factor affecting the data distribution is the N level applied. Thus, data of high N treatments distribute along the right side of vertical axis and data of no N and low N treatments distribute along the left side of this axis. Moreover, ANOVA results were included as supplementary material in Tables S1 and S2.

Plant Growth and AM Fungal Colonization
In absence of N fertilization and under low N supply, mycorrhization enhanced plant fresh weight regardless of N form, both under well-watered (WW) and under drought stress (DS) conditions. However, under high N supply, mycorrhization did not have a significant effect on plant growth, with the exception of WW plants with high NH4 + supply, which slightly decreased plant fresh weight (Table 1).
At each watering regime, levels of root AM colonization were not significantly affected by the different N forms and concentrations applied (Table 1).

Plant Growth and AM Fungal Colonization
In absence of N fertilization and under low N supply, mycorrhization enhanced plant fresh weight regardless of N form, both under well-watered (WW) and under drought stress (DS) conditions. However, under high N supply, mycorrhization did not have a significant effect on plant growth, with the exception of WW plants with high NH 4 + supply, which slightly decreased plant fresh weight (Table 1). At each watering regime, levels of root AM colonization were not significantly affected by the different N forms and concentrations applied (Table 1). However, under such conditions, the plants that were not fertilized with any N presented higher MDA levels as compared to those fertilized with both urea levels ( Table 1). Under DS, mycorrhization decreased MDA production in urea-fertilized plants and in those that did not receive any N. Moreover, MDA accumulation was also reduced with the highest urea concentration ( Table 1). Relative chlorophyll content was measured with SPAD by measuring dual optical absorbances. Values increased with high N-fertilization under both WW and DS conditions. In urea-fertilized plants under WW, mycorrhization slightly decreased chlorophyll levels compared to non-AM plants, while under drought stress it increased slightly the chlorophyll levels under low urea fertilization (Table 1).

Nitrogen and Carbon Concentrations in Tissues
Under WW conditions, total nitrogen levels in roots and leaves were differently affected by N form and concentration. In pattern was observed in C/N ratio in roots of urea-treated plants (Figure 3ii(E)). In leaves under WW conditions, the ratio decreased also with high NH4 + application both in non-AM and AM plants. Under DS, the C/N ratio followed a similar trend with NH4 + and urea fertilization. Thus, it decreased in non-AM plants compared to AM plants when plants were not fertilized with N. The ratio increased in both inoculation treatments under low-N fertilization and it decreased in both inoculation treatments under high NH4 + fertilization (Figure 3i(F) and Figure 3ii(F)).

Accumulation of New Nitrogen in Tissues
The calculation of the percentage of new 15 N nitrogen in the different analyzed tissues showed that the higher proportion of newly incorporated 15 N was found in non-AM plants fertilized with low N concentration (independently of the N form and of the water regime). Thus, under well-watered conditions higher proportion of new 15 N was found in leaves and stems of non-AM plants fertilized with low NH 4 + ( Table 2). Under drought stress, higher values were found in leaves, stems and roots of non-AM plants fertilized with low NH 4 + . In low urea-fertilized plants, a higher proportion of new 15 N was found in leaves of non-AM plants under well-watered conditions and in leaves and stems of non-AM plants under drought stress conditions ( Table 2).

Discussion
Although many aquaporins are highly selective for water, the selectivity filters of plant aquaporins show a high divergence, suggesting wide functional diversity for these proteins [25]. Thus, it became increasingly clear that apart from water, urea, and ammonia, aquaporins can also transport other small neutral molecules such as non-electrolyte acetamide, long polyols, short polyols including glycerol, CO 2 , O 2 , purines and pyrimidines, silicic acid, boric, arsenic and germanic acids, lactic acid or hydrogen peroxide, and related oxy-radicals [26]. This fact highlights the paramount relevance of aquaporins for plant physiology. For instance, the ability of aquaporins to transport N compounds pointed to important roles for aquaporins in N mobilization and metabolism. The diffusion of CO 2 through aquaporins suggests a role in carbon fixation. Silicon uptake and metabolism seem to be crucial for responses to biotic and abiotic stresses and boron is closely related with nutrition and structural development. The ability of aquaporins to transport H 2 O 2 points to potential involvement in stress signaling and responses [22].
Under drought stress conditions, the uptake of soil nutrients by plant roots and subsequent mobilization to aerial parts is hampered by low soil water availability. The AM symbiosis was shown to alleviate limited soil nutrients availability, including N, under different environmental conditions [10,11] and to improve plant performance under both biotic and abiotic stresses [27]. In the case of drought, several mechanisms affecting plant physiology were proposed to explain the enhanced drought tolerance of AM plants [28,29]. However, given the fact that the AM symbiosis regulates a wide number of plant aquaporins under drought stress [22] and that besides water, some of these aquaporins can transport ammonium or urea, it is plausible that these AM-regulated aquaporins may be involved in the plant mobilization of N forms, contributing in this way to the enhanced drought tolerance of AM plants. This study aims to shed light on this hypothesis. Moreover, this study also aims to further understand how the AM symbiosis affects N balance in plants depending on N form and supply and how this combination of treatments affects plant physiology and performance under drought stress. The importance of plant-microbe interactions for reducing chemical fertilizers and inputs while maintaining normal development and plant growth is attracting more attention [30]. Furthermore, N fertilization strongly affects crop productivity, and the increase in plant N-uptake efficiency and its mobilization capacity could reduce the consumption of N fertilizers.

Physiological Effects of AM Symbiosis under Different N Supply and Watering Conditions
As reported in numerous studies with arbuscular mycorrhizal symbioses and in previous experiments with similar conditions [31,32], the total biomass (represented by plant FW) was generally enhanced by the AM symbiosis under both watering conditions. This effect, however, was not observed under high N supply (regardless of the N form, Table 1). This is understandable, as when N is in excess in the medium, the plant can directly take it, and AM symbiosis does not suppose an additional benefit for nutrient acquisition. Consistent with this, a recent study also found that the growth-promoting effect of the AM fungus was only observed under low N conditions [33]. High nitrogen supply increased leaf N content, especially during drought stress, and it was positively correlated with plant growth (Figure 3i(B) and Figure 3ii(B)). AM plants did not show higher N levels in leaves. However, AM plants increased N levels in roots compared to non-AM plants in urea-fed plants under either WW or DS and under DS in plants irrigated with high NH 4 + (Figure 3i(A) and Figure 3ii(A)).
Low MDA levels, produced as the result of lipid peroxidation are usually related to higher membrane stability [34]. Our study showed that without N supply under WW conditions, the levels of MDA were higher compared to the other treatments, regardless of the N form (Table 1). This fact suggests that nutrient status affects cell membrane stability. Under drought stress conditions, the AM fungus generally decreased MDA levels regardless of N concentration or form, which is in accordance with previous studies [24,35].
SPAD values estimate chlorophyll content of the leaves and can be correlated with N concentration per leaf area, although this relationship is very variable depending on environmental factors and plant species [36]. Moreover, leaf chlorophyll content is a determinant factor in plant photosynthesis [37]. In our study, SPAD values were enhanced with the increase in N concentration with both urea and NH 4 + and in both well-watered and drought stressed plants. In some cases, when the concentration of N was low, mycorrhization was found to increase SPAD values, but once more, under high N supply, the symbiosis did not enhance this parameter (Table 1). Altogether, physiological data showed that maize plants grew well under the highest N concentration and grew with restriction under the two low N levels. This may be related to the plant species, or variety used [38].
Under well-watered conditions, A N and gs were enhanced by high N concentrations (regardless of N form) at the two measured time points (Figure 4). This effect was also shown in other studies [39,40]. During water deprivation, stomatal closure and increase in the resistance of CO 2 diffusion commonly lead to a decrease in photosynthetic rate [41]. In our study, drought stress negatively impacted A N and gs under both high NH 4 + and urea supply, especially in AM plants after 14 days of drought ( Figure 4).
In fact, after 14 days of drought, AM plants fertilized with high NH 4 + or urea doses had even lower A N and gs levels than non-AM plants. As explained above, when plant growth is not limited by soil N availability, mycorrhizal inoculation might not suppose an advantage and the plant may reduce its dependence on the AM fungus.

N Acquisition by the Plant
In this study, the use of 15 N-labelled urea and ammonium showed that non-AM plants receiving low N supply enhanced most the uptake of new N under both well-watered or drought stress conditions, being reflected by the measured percentage of 15 N in leaves, stems and roots. This may be due to a higher N demand in plants fertilized with low N levels, as comparted to those receiving high N levels or to AM plants. Indeed, it is likely that the high nutrient concentration under high urea or ammonium supply inhibited the uptake of new N. The effect was observed with both ammonium and urea, highlighting that the two nutrients are very similar in their assimilation pathways, as previously suggested [4]. However, with urea nutrition, these differences were only observed in leaves or in stem in the case of drought stressed plants. In the case of AM plants, the transport of labelled 15 N was generally low, which may reflect a lower N demand in these plants during the short time of 15 N labelling (3h) due to a general better N nutrition in AM plants. In other study, the transport of 15 N-NH 4 + to the AM plant was lower compared to NO 3 − nutrition in wheat [42].

AQP Relative Gene Expression Levels Are Affected by N Supply and Form
Generally, the patterns of root AQPs gene expression were different with the application of NH 4 + or urea and also differed in response to N levels in our experiment ( Figure 5). This fact suggests that the plant discriminated between the two N forms and that at least not all the applied urea was transformed into NH 4 + in the soil. N acquisition is commonly affected by water flow from the soil and through the plants. In this sense, aquaporins were related to N balance in several independent studies, and their overexpression positively affected total N uptake and transpiration rate [21]. Two main families of AQPs are potentially involved in the mobilization of N: tonoplast intrinsic proteins (TIPs) and nodulin 26-like intrinsic proteins (NIPs) [41]. This may explain why ZmPIP1;1, ZmPIP1;3, ZmPIP2;2 and also ZmTIP2;3 did not show significant differences among treatments, regardless of NH 4 + or urea application. In contrast, ZmTIP1;1 and ZmNIP2;1 proteins were shown to transport urea when expressed in yeast. In addition, ZmTIP1;1 was found to transport NH 4 + in the same heterologous system in previous studies [22,23]. Our results indicate that ZmTIP1;1 could be transporting in planta both urea and ammonium, as the gene was up-regulated under high NH 4 + concentrations (especially under DS conditions) and under low or high urea concentrations under WW or DS conditions (Figure 5i(B) and Figure 5ii(B)). The induction of ZmTIP1;1 by the AM symbiosis under high N conditions may be related to the storage in the vacuole of excess N taken up by the plants under such conditions of abundant N, as this aquaporins is from tonoplast and could be involved in cellular N homeostasis [43]. In the case of ZmNIP2;1, our data suggests that together with urea it could be transporting NH 4 + , as the gene was up-regulated with low NH 4 + supply under WW and with high NH 4 + supply under DS (Figure 5i(D)). The differential regulation of ZmTIP4;1 and ZmPIP2;4 with urea fertilization and ZmPIP2;4 with NH 4 + supply made us hypothesize that these two AQPs may also play a role in N homeostasis in planta. Besides, the genes found to be regulated by N status were at the same time differentially regulated by the AM symbiosis. For instance, ZmNIP2;1 was strongly up-regulated in AM plants with high NH 4 + under DS but not in non-AM plants, and in non-AM plants with low NH 4 + but not in AM plants (Figure 5i(D)). On the contrary, ZmTIP4;1 was up-regulated by low urea supply under both water treatments in non-AM plants but unaffected in AM plants (Figure 5ii(C)). These results support a differential regulation of some plant aquaporins when the plant is colonized by the AM fungus.

Experimental Design
The experiment consisted of a randomized block design with (1) two inoculation treatments: non-inoculated plants (Non-AM) and plants inoculated with an arbuscular mycorrhizal fungus (AM plants); (2) two water regimes: WW plants and DS plants; and (3) two sources of nitrogen: urea and ammonium applied at three different concentrations: No N added, low concentration (3 µM) and high concentration (10 mM). The experiment consisted of six replicates per treatment and plants receiving no N fertilization were common for the urea and ammonium experiments, resulting in a total of 120 plants.
Seeds of Zea mays L. cultivar PR34B39 were provided by Pioneer Hi-Bred (DuPont, Sevilla, Spain), and were pre-germinated in sand for five days and then transferred to 1.5 L pots containing 1250 g of the above described substrate. At planting time, half of the plants were inoculated with ten grams of AM inoculum with Rhizophagus irregularis (Schenck and Smith), strain EEZ 58. The inoculum consisted of soil, spores, mycelia, and infected root fragments. Non-inoculated plants received a 10 mL aliquot of an inoculum filtrate (<20 µm), in order to provide the microbial population accompanying the AM inoculum but free of AM propagules.

Growing Conditions
Plants were cultivated for eight weeks under greenhouse conditions (photosynthetic photon flux density of ca. 800 µmol m −2 s −1 , 25/20°C, 16/8 light dark period and 50-60% RH). Each plant received 50 mL of Hoagland nutrient solution [44] three times per week. The solution was modified to contain only 25% of P, in order to avoid inhibition of mycorrhization. The Hoagland solution was also modified to provide the different nitrogen forms and concentrations as explained above. Ammonium was applied in the nutrient solution as (NH 4 ) 2 S 2 O 8 . On alternate days plants received the same amount of water. The drought stress treatment was imposed the last two weeks to half of the plants by irrigating with half of the water/Hoagland volume of well-watered plants (25 mL vs. 50 mL). To avoid nutrient deficiency, droughted treatments received 2X Hoagland nutrient solution. The water stress imposed was similar to previous studies and considered to be a severe stress [24,31,32,45].
A pulse with labelled 10% 15 N was applied with the nutrient solution three hours before harvest to four plants per treatment in two different forms: Urea-15 N 2 (98 atom % 15 N), and ammonium-15 N 2 sulphate (60 atom % 15 N), Sigma Aldrich.

Biomass Production and Symbiotic Development
The roots and aerial parts of six plants per treatment were used for fresh weight measurement at harvest (8 weeks after sowing).
Aliquots of three maize roots per treatment were used for staining with Trypan blue following the procedure described by Phillips and Hayman [46], in order to differentiate AM fungal structures. The gridline intersect method [47] was used in those roots to calculate the extent of mycorrhizal colonization.

Oxidative Damage to Lipids
Lipid peroxides were extracted from three fresh leaves per treatment by grinding 500 mg of tissue with 6 mL of 100 mM potassium phosphate buffer (pH 7) and quantified as described in Quiroga et al. [24]. Lipid peroxidation was estimated as the content of 2-thiobarbituric acid-reactive substances (TBARS) and expressed as equivalents of malondialdehyde (MDA) according to Halliwell and Gutteridge [48].

Leaf Chlorophyll Content
Chlorophyll content was estimated in the second fully expanded youngest leaf of six plants per treatment four hours after the onset of photoperiod and one day before harvest by using a Chlorophyll Content Measurement System CL-01 (SPAD, Hansatech Instruments Ltd., Norfolk, UK). This instrument determines relative chlorophyll content in leaf samples by measuring dual optical absorbances (620 and 940 nm wavelengths).

Mineral Analysis
Mineral analysis was performed in the INAN-EEZ Institute in Armilla, Granada, Spain. The equipment used was an elemental analyzer Leco TruSpec CN, which detects Nitrogen in the sample by Dumas method, including a complete combustion of the sample at 950°C with high pressure oxygen, chemical reduction of nitrogen oxides to molecular nitrogen and analysis of this molecular nitrogen with a thermal conductivity detector (TCD). Carbon was detected as carbon dioxide (from the sample combustion) by an infrared detector. Quantification was performed with certified standards from Leco.

Net Photosynthesis and Stomatal Conductance
Stomatal conductance (gs) and net photosynthesis (A N ) were measured in the second fully expanded youngest leaf of five plants per treatment at three time points: one week before starting the drought stress treatment (6 weeks of growing), after 7 days of drought stress and after 14 days of drought treatment. Both parameters were measured by using a portable photosystem system LI-6400 (LICOR Biosciences, Lincoln, NE, USA) two hours after sunrise. Measurements were performed at an ambient CO 2 concentration of 390 µmol mol −1 , temperature of 25/30 0 C, 50 ± 5% relative humidity and a photosynthetic photon flux density (PPFD) of 1000 µmol m −2 s −1 .

RT-qPCR
Total RNA of three biological replicates from maize roots was extracted as described in Quiroga et al. [24]. Maxima H Minus first strand cDNA kit (Thermo Scientific TM ) was used for the synthesis of first strand cDNA by using 1 µg of purified total RNA, according to the manufacturer's instructions.

Estimation of 15 N Isotopic Abundances
Sample analysis for 15 N composition was carried out at the Stable Isotope Laboratory of the Instituto Andaluz de Ciencias de la Tierra in Granada (CSIC-UGR), Spain. Stable isotope (δ 15 N) analysis of leaf, stem and root samples were performed on a Carlo Elba NC1500 (Milan, Italy) elemental analyzer coupled on-line via a ConFlow III with a Delta Plus XP (Thermo-Finnigan, Bremen, Germany) mass spectrometer (EA-IRMS). Commercial N 2 was used as the internal standard for the N isotopic analyses.

Statistical Analysis
The SPSS Statistics (Version 26, IBM Analytics) was used to perform data analysis. The normality and homogeneity of variances were tested. Some parameters were logarithmically transformed to achieve homogeneity of variance. Data obtained for ammonium and urea were analyzed separately by using one-way ANOVA in order to facilitate interpretation of results in this experiment that contains up to four factors. Duncan's multiple range test was used to pairwise multiple comparisons between treatments at α = 0.05.

Conclusions
AM symbiosis improved plant biomass in absence of N fertilization or under low urea or ammonium fertilization, regardless of the watering regime. In contrast, under high N supply, no effect of AM symbiosis on plant development was observed. This effect was associated with reduced oxidative damage to lipids and increased N accumulation in plant tissues. High N fertilization with either ammonium or urea enhanced net photosynthesis (A N ) and stomatal conductance (gs) values in plants maintained under well-watered conditions, but 14 days after drought stress imposition these parameters declined in AM plants fertilized with high N doses. The aquaporin ZmTIP1;1 could be transporting in planta both urea and ammonium, as the gene was up-regulated under high NH 4 + concentrations and under low and high urea concentrations. The differential regulation of ZmTIP4;1 and ZmPIP2;4 with urea fertilization and of ZmPIP2;4 with NH 4 + supply suggests that these two aquaporins may also play a role in N mobilization in planta. At the same time, the aquaporin genes found to be regulated by N status were also differentially regulated by the AM symbiosis, suggesting a possible role of these aquaporins in the AM-mediated plant N homeostasis that deserves future studies.