N Absorption, Transport, and Recycling in Nodulated Soybean Plants by Split-Root Experiment Using 15 N-Labeled Nitrate

: Nitrate concentration is variable in soils, so the absorbed N from roots in a high-nitrate site is recycled from shoots to the root parts in N-poor niche. In this report, the absorption, transport, and recycling of N derived from 15 N-labeled nitrate were investigated with split-root systems of nodulated soybean. The NO 3 − accumulated in the root in 5 mM NO 3 − solution; however, it was not detected in the roots and nodules in an N-free pot, indicating that NO 3 − itself is not recycled from leaves to underground parts. The total amount of 15 NO 3 − absorption from 2 to 4 days of the plant with the N-free opposite half-root accelerated by 40% compared with both half-roots that received NO 3 − . This result might be due to the compensation for the N demand under one half-root could absorb NO 3 − . About 2–3% of the absorbed 15 N was recycled to the opposite half-root, irrespective of N-free or NO 3 − solution, suggesting that N recycling from leaves to the roots was not affected by the presence or absence of NO 3 − . Concentrations of asparagine increased in the half-roots supplied with NO 3 − but not in N-free half-roots, suggesting that asparagine may not be a systemic signal for N status.


Introduction
Nitrogen (N) is one of the essential macronutrients required for plants, and the availability of N is often a limiting factor for plant growth and crop yield. Plant roots absorb N primarily in the forms of nitrate (NO 3 − ) and ammonium (NH 4 + ). Leguminous plants such as soybean incorporate atmospheric dinitrogen (N 2 ) in root nodules, forming a symbiosis with N 2 -fixing rhizobia. Nitrate absorbed by soybean roots can be directly translocated to the shoot or first assimilated then transported mainly as asparagine (Asn) via xylem vessels to the leaves. Some parts of N assimilated in the leaves may be re-translocated to the apical part of the roots or buds via the phloem to support the N requirement of the developing organs [1]. Therefore, recycling N from leaves to the roots is important for young roots and nodules, but the mechanisms of how to control the recycling of N and how the N conditions around the roots affect N recycling are not fully understood.
As nitrate concentrations in the field vary by several order of magnitude, both spatially and temporally [2,3], plant roots need to regulate NO 3 − absorption mediated by nitrate transporters [1]. The nitrate reductase (NR) activity is also finely regulated by various internal compounds and environmental conditions at the transcription, translation, protein modification, and degradation levels [4]. Nitrates and metabolites such as glutamine (Gln), sucrose, cytokinin, and light are regulatory factors for NR [4]. In addition, nitrate absorption is regulated systemically through root-shoot communications [2].
The fixed N in the nodule also recycles from the shoot to the roots. Oghoghorie and Pate [5] investigated the N transport of nodulated field pea (Pisum arvense L.) using 15 N-labeled N 2 and 15 NO 3 − . When the nodules at the upper roots were exposed to 15 N 2 , an appreciable amount of 15 N translocated to the lower part of the roots. However, 15 N Soybean seeds (Glycine max (L.) Merr., cultivar 'Williams') were surface-sterilized and sown in a vermiculite bed. At 7 days after planting (DAP), plants in a vermiculite bed were inoculated with a suspension of Bradyrhizobium diazoefficiens (USDA 110). At 10 DAP, the primary root below 5 cm long was cut off and the plant transplanted into an 800 mL of nitrogen-free nutrient solution [12] in a 900 mL glass bottle with continuous aeration. The bottle covered with aluminum foil for shading a culture solution. Plants were cultivated in a biophotochamber (LH-350S, Nippon Medical & Chemical instruments Co., Ltd., Osaka, Japan) under 28 • C day/18 • C night temperatures, 55% relative humidity, and under a photoperiod of 16 h light (228 µmol photons m −2 s −1 )/8 h darkness.
At 24 DAP, each plant transferred to a two-compartment container ( Figure 1). Two polyethylene terephthalate pots with a capacity of 1 L were attached. Each pot was filled with 0.8 L of N-free culture solution, covered with aluminum foil, and aerated continuously. Three treatments were imposed on 31 DAP using NaNO 3 (30.1 atom% 15 N). The N0-N0 treatment: [L]; N-free, [R]; N-free. 15 N5-N0 treatment: [L]; 15 N-labeled 5 mM NO 3 − , [R]; N-free; 15 N-N5 treatment: [L]; 15 N-labeled 5 mM NO 3 − , [R]; non-labeled 5 mM NO 3 − . The 800 mL of 5 mM NaNO 3 solution contained 56 mg N per pot. At 2 day and 4 day treatments, plants were harvested, washed, and separated into the lower half of the noduled roots, which were in direct contact with the solution, the upper half of the noduled roots, which were out of the solution, the basal part of roots, and the shoot. No nitrate ion was detected in the N-free solution after 4 days of cultivation in split-root culture, analyzed by the ion chromatograph (PIA-1000, Shimadzu Corporation, Kyoto, Japan), so mixing of the solution between two adjacent pots was excluded. The plants were dried using a freeze-dryer (VD-400F, TAITEC, Saitama, Japan) and separated into the leaves, stems plus petioles, basal roots, the upper part of nodules and roots in [L] or [R], and the lower part of nodules and roots in [L] or [R]. Then, the dry weight of them was measured and ground into a fine powder. About 2-5 mg of ground sample was packed in a tin capsule, and the N concentration and 15 N concentration were analyzed by the elemental analyzer (Flash EA1112; Thermo Electron, Milan, Italy) coupled to an isotope-ratio mass spectrometer (Delta Plus XP; Thermo Fisher Scientific, Bremen, Germany). The percentage of N derived from 15   Both pots were continuously aerated. Plants were harvested at 2 days (2 DAT) and 4 days (4 DAT) of treatments. Plant parts separated at the red line.

The Effects of Split-Root Treatments on the Principal N Metabolites in Each Part of Soybean Plants
Freeze-dried plant powder was extracted with 80% ethanol containing 0.2 mM MES as an internal standard for capillary electrophoresis. The extracts were dried in vacuo and redissolved in water. Then, the concentrations of nitrate, Glu, Asp, Gln, Asn, allantoin, and allantoate were analyzed by capillary electrophoresis (7100, Agilent Technologies, Inc., Santa Clara, CA, USA) using a fused silica tube (inner diameter (id); 50 µm, 104 cm long) and a commercial buffer solution (α-AFQ109, Ohtsuka Electronics Co., Ltd., Osaka, Japan), with an applied voltage of −25 kV. Peaks detected with a signal wavelength of 400 nm and a reference wavelength of 265 nm. As half-roots in the right pot [R] and the left pot [L] were equivalent in N0-N0 and 15 N5-N5 treatments, data of the concentrations for these treatments were combined (N = 8).  30.1 mg N and 50.5 mg N at 2 DAT and 4 DAT, respectively. By the N0-N0 treatment, plants depended on sole nitrogen fixation, and the 20.4 mg increase of total N from 2 DAT to 4 DAT was derived from N 2 fixed by the root nodules. The increase in N content of the leaves from 2 DAT to 4 DAT was about 15 mg N, and the N contents were statistically significant (p < 0.01) between 2 DAT and 4 DAT based on Student's T-test. The N contents in the stems and basal roots tended to increase from 2 DAT to 4 DAT but were not significant. The N contents in both the upper and lower roots and nodules did not change significantly. The total amount of N in the plants with 15 N5-N0 treatment was 29.5 mg N and 48.8 mg N at 2 DAT and 4 DAT, respectively. The N increase from 2 to 4 days was 19.3 mg N, almost the same as that in the N0-N0 treatment. The increase in the N content of leaves was 12 mg N, most remarkable among organs, and the N contents in the basal roots and the lower roots in contact with 15 NO 3 − were also significantly increased from 2 DAT to 4 DAT. Total amount of N in the plants with 15 N5-N5 treatment was 32.5 mg N, and 53.5 mg N, at 2DAT and 4 DAT, respectively. Total amounts in 15 N5-N5 treatment were slightly higher than those of N0-N0, and 15 N5-N0 treatments, although statistically not significant. In this treatment, the increase in the total N content from 2 DAT to 4 DAT was 21 mg N, almost the same as those in N0-N0 (20.4 mg N) and 15 N5-N0 ( 19.3 mg N) treatments. The N contents in the leaves, basal roots, and lower roots were significant between 2 DAT and 4 DAT. In all treatments, the increase of N in leaves was remarkable from 2 DAT to 4 DAT because the leaves grew faster in this stage (about V4), and the N fixation activity in this stage might become higher.

N Content from 15 N-Labeled Nitrate in Each Part of Soybean Plants
To see how much 15 NO 3 − was absorbed and how the N derived from 15 NO 3 − was distributed among organs from 2 to 4 DAT, the N content from 15 NO 3 − in each part of plants with 15 N5-N0 and 15 N5-N5 treatments is shown in Figure 3. The total amounts of N derived from 15 NO 3 − were 2.51 mg N at 2 DAT and 10.4 mg N at 4 DAT in 15 N5-N0 treatment, while those were 3.04 mg N at 2 DAT and 8.67 mg N at 4 DAT in 15 N5-N5 treatment. The amounts of N derived from 15 NO 3 − in 15 N5-N0 treatment at 4 DAT were significantly (p < 0.05) higher than that of 15 N5-N5 treatment, suggesting that when N was applied only from one side of the split-roots, the N absorption might be accelerated compared with when both sides of the roots were supplied with NO 3 − . In both cases, 15 NO 3 − absorption was higher during the second 2 days period from 2 DAT to 4 DAT compared with the first 2 days from 0 DAT to 2 DAT. Total amount of N from 15 NO 3 − from 2 DAT to 4 DAT was 7.89 mg N and 5.63 mg N in 15 N5-N0 and 15 N5-N5 treatments, respectively. These accounted for a 40% increase in one-side NO 3 − supply ( 15 N5-N0) compared with both sides supplied with NO 3 − ( 15 N5-N5). About half of 15 N was distributed in leaves at 4 DAT with both 15 N5-N0 (6.16 mg N) and 15 N5-N5 (4.79 mg N) treatments. The accumulation of labeled-N in the leaves of the plants with 15 N5-N0 treatment was significantly higher than that with 15 N5-N5 treatment at 4 DAT (p < 0.05). The distribution of 15 N-labeled N was also high in the lower roots, which were in direct contact with 15

N Content from 15 N-Labeled Nitrate in Each Part of Soybean Plants
To see how much 15 NO3 − was absorbed and how the N derived from 15 NO3 − was distributed among organs from 2 to 4 DAT, the N content from 15 NO3 − in each part of plants with 15 N5-N0 and 15 N5-N5 treatments is shown in Figure 3. The total amounts of N derived from 15 NO3 − were 2.51 mg N at 2 DAT and 10.4 mg N at 4 DAT in 15 N5-N0 treatment, while those were 3.04 mg N at 2 DAT and 8.67 mg N at 4 DAT in 15 N5-N5 treatment. The amounts of N derived from 15 NO3 − in 15 N5-N0 treatment at 4 DAT were significantly (p < 0.05) higher than that of 15 N5-N5 treatment, suggesting that when N was applied only high in the lower roots, which were in direct contact with 15 NO3 − . Those were 2.36 mg N with 15 N5-N0 treatment and 1.96 mg N with 15 N5-N5 treatment. At the beginning of treatments, the 15 N-labeled solution contained 56 mg N of nitrate per pot, and the total N absorption from 15 NO3 − did not exceed 12 mg N at 4 DAT, so nitrate in the solution has not been depleted during the 4 days of 15 N treatment. The N derived from atmospheric N (Ndfa) and N derived from nitrate (Ndfn) were estimated by subtracting the amount of N derived from 15 NO3 − from the increase in total N from 2 DAT to 4 DAT. The Ndfa from 2 DAT to 4 DAT in 15 N5-N0 treatment was 13.14 (19.30 − 6.16) mg N. While the amount of Ndfn from 15 N from 2 DAT to 4 DAT in 15 N5-N5 treatment was 5.63 (8.67 − 3.04) mg N from the half-root with 15 NO3 − , both sides of the halfroots were in the same N conditions except for labeling, so the same amount of N was absorbed from non-labeled half-roots. Therefore, a total of 11.26 mg (5.63 + 5.63) mg N should derive from both the 15 N-labeled and non-labeled nitrate. The total N increase from 2 DAT to 4 DAT was 21.0 mg N, so the amount of Ndfa was 9.7 (21.0 − 11.3) mg N in the 15 N5-N5 treatment. The percentage dependence on Ndfa and Ndfn was 100% Ndfa and 0% Ndfn in N0-N0 treatment, 59% Ndfa and 41% Ndfn in 15 N5-N0 treatment, and 46% Ndfa and 54% Ndfn in 15 N5-N5 treatment.

Percentage of N Derived from 15 N-Labeled Nitrate in Each Part of Soybean Plants
The percentage of N derived from 15 NO3 − ( 15 N%) in total N in each part of the soybean plant with split-root systems is shown in Figure 4. The 15 N% of leaves, stems, and basal roots were about 8.6, 8.0, and 5.8, respectively at 2 DAT, and there were no significant differences between 15 N5-N0 and 15 N5-N5 treatments. However, the 15 N% of these organs The N derived from atmospheric N (Ndfa) and N derived from nitrate (Ndfn) were estimated by subtracting the amount of N derived from 15 NO 3 − from the increase in total N from 2 DAT to 4 DAT. The Ndfa from 2 DAT to 4 DAT in 15 N5-N0 treatment was 13.14 (19.30 − 6.16) mg N. While the amount of Ndfn from 15 N from 2 DAT to 4 DAT in 15 N5-N5 treatment was 5.63 (8.67 − 3.04) mg N from the half-root with 15 NO 3 − , both sides of the half-roots were in the same N conditions except for labeling, so the same amount of N was absorbed from non-labeled half-roots. Therefore, a total of 11.26 mg (5.63 + 5.63) mg N should derive from both the 15 N-labeled and non-labeled nitrate. The total N increase from 2 DAT to 4 DAT was 21.0 mg N, so the amount of Ndfa was 9.7 (21.0 − 11.3) mg N in the 15 N5-N5 treatment. The percentage dependence on Ndfa and Ndfn was 100% Ndfa and 0% Ndfn in N0-N0 treatment, 59% Ndfa and 41% Ndfn in 15 N5-N0 treatment, and 46% Ndfa and 54% Ndfn in 15 N5-N5 treatment.

Percentage of N Derived from 15 N-Labeled Nitrate in Each Part of Soybean Plants
The percentage of N derived from 15 NO 3 − ( 15 N%) in total N in each part of the soybean plant with split-root systems is shown in Figure 4. The 15 N% of leaves, stems, and basal roots were about 8.6, 8.0, and 5.8, respectively at 2 DAT, and there were no significant differences between 15 N5-N0 and 15 N5-N5 treatments. However, the 15 N% of these organs at 4 DAT were significantly higher in the 15 N5-N0 treatment, compared with the 15 N5-N5 treatment ( Figure 4A-C). It is the same for the 15

Percentage Distribution of N Derived from 15 N-Labeled Nitrate in Each Part of Soybean Plants
To clarify the effects of the presence or absence of NO3 − in the opposite side of nonlabeled pot on the distribution of 15 N absorbed from one half-root, the percentage distribution of the 15 N-labeled N in each organ in total 15 N was compared between 15 N5-N0 and 15 N5-N5 treatments ( Table 1). The 15 N distribution was the highest in leaves at 2 DAT (ca. 46%) and 4 DAT (ca. 56%) and not significantly different between the two treatments (Table 1 a. ). The 15 N distribution in the stems was relatively constant about 10% irrespective of DAT or treatments. Those in the basal roots were 3% at 2 DAT and 4% at 4 DAT in both treatments.

Percentage Distribution of N Derived from 15 N-Labeled Nitrate in Each Part of Soybean Plants
To clarify the effects of the presence or absence of NO 3 − in the opposite side of non-labeled pot on the distribution of 15 N absorbed from one half-root, the percentage distribution of the 15 N-labeled N in each organ in total 15 N was compared between 15 N5-N0 and 15 N5-N5 treatments ( Table 1). The 15 N distribution was the highest in leaves at 2 DAT (ca. 46%) and 4 DAT (ca. 56%) and not significantly different between the two treatments (Table 1a). The 15 N distribution in the stems was relatively constant about 10% irrespective of DAT or treatments. Those in the basal roots were 3% at 2 DAT and 4% at 4 DAT in both treatments.  − can be transported from the shoot to the roots and nodules, nitrate concentration in each organ was determined ( Figure 5). Nitrate was not detected in the plant parts with N0-N0 treatment where both half-roots were not supplied with NO 3 − , although small peaks less than 1 µmol g −1 DW were occasionally detected, possibly due to the contamination from the culture solution or environment. When both half-roots were in 5 mM

N Concentrations of Major N Metabolites in Leaves, Stems, and Basal Roots
The concentrations of Glu, Asp, Asn, and allantoate in the leaves tended to be high with 15 N5-N5 > 15 N5-N0 > N0-N0 treatment, although the concentrations of Gln and allantoin were higher in N0-N0 treatment at 4 DAT than the other treatments ( Figure 6A,D). The Asn concentration in the stems was significantly higher in the 15 N5-N5 than N0-N0 treatment at 4 DAT ( Figure 6B,E). In the basal roots, the concentrations of Glu, Asp, Gln, and Asn were the highest in 15 N5-N5, compared with the 15 N5-N0, and N0-N0 treatments at 4 DAT ( Figure 6C,F). The concentrations of allantoin and allantoate were not different among treatments at 4 DAT, although these were higher in the 15 N5-N5 treatment at 2 DAT.
The concentrations of Glu, Asp, Asn, and allantoate in the leaves tended to be high with 15 N5-N5 > 15 N5-N0> N0-N0 treatment, although the concentrations of Gln and allantoin were higher in N0-N0 treatment at 4 DAT than the other treatments ( Figure 6A,D). The Asn concentration in the stems was significantly higher in the 15 N5-N5 than N0-N0 treatment at 4 DAT ( Figure 6B,E). In the basal roots, the concentrations of Glu, Asp, Gln, and Asn were the highest in 15 N5-N5, compared with the 15 N5-N0, and N0-N0 treatments at 4 DAT ( Figure 6C,F). The concentrations of allantoin and allantoate were not different among treatments at 4 DAT, although these were higher in the 15 N5-N5 treatment at 2 DAT.

N Concentrations of Principal N Metabolites in the Upper and Lower Roots
The concentrations of N metabolites were compared between roots and nodules of the plants with N0-N0 treatment and 15 N5-N5 treatment ( Figures 7A,B and 8A,B). In addition, these were compared in the single plant where one half-root was supplied with 5 mM 15 NO 3 − and another half was supplied N-free solution ( Figures 7C,D and 8C,D). Figure 7A,B shows the concentrations of major N metabolites in the upper roots and the lower roots comparing N0-N0 treatment which depends on only N 2 fixation, and 15 N5-N5 treatment in which both sides of half-roots were supplied with 5 mM NO 3 − . The concentrations of Gln, Asp, and Asn significantly increased by supplying NO 3 − for 2 days, while the concentrations of allantoin and allantoate decreased at 4 DAT. mM 15 NO3 − and another half was supplied N-free solution ( Figure 7C,D and Figure 8C,D Figure 7A,B shows the concentrations of major N metabolites in the upper roots and th lower roots comparing N0-N0 treatment which depends on only N2 fixation, and 15 N5-N treatment in which both sides of half-roots were supplied with 5 mM NO3 − . The concen trations of Gln, Asp, and Asn significantly increased by supplying NO3 − for 2 days, whil the concentrations of allantoin and allantoate decreased at 4 DAT.   Figure 8A,B shows the concentrations of principal N compounds in the upper nod ules and the lower nodules comparing N0-N0 and 15 N5-N5 treatments. The concentratio of Glu in the nodules was 20-40 µmol g −1 DW and higher than those in the roots, which i   Figure 8A,B shows the concentrations of principal N compounds in the upper nodules and the lower nodules comparing N0-N0 and 15 N5-N5 treatments. The concentration of Glu in the nodules was 20-40 µmol g −1 DW and higher than those in the roots, which is less than 10 µmol g −1 DW, and the concentration was not affected by the treatments. The concentration of Asp tended to increase in the nodules by supplying NO 3 − , whereas the concentrations of allantoate decreased by NO 3 − supply. Figure 8C,D shows the concentrations of principal N compounds in the upper and lower nodules attached to each side of the half-roots of a single plant treated with 15 N5-N0 treatment. The concentrations and the trends by 5 mM NO 3 − supply were similar to the upper ( Figure 8A) and lower ( Figure 8B) nodules with N0-N0 and 15 N5-N5 treatments, and the increase in Asn concentration at 4 DAT was remarkable only in the half-roots with 15 NO 3 − . The concentration of allantoate tended to decrease at 4 DAT in the half-roots with 5 mM NO 3 − solution. These results indicate that the concentrations of principal N compounds in nodules were not affected compared with those in the roots.

N Concentrations of Principal N Compounds in the Upper and Lower Nodules
Nitrogen 2022, 3, FOR PEER REVIEW 14 less than 10 µmol g −1 DW, and the concentration was not affected by the treatments. The concentration of Asp tended to increase in the nodules by supplying NO3 − , whereas the concentrations of allantoate decreased by NO3 − supply.

Changes in Total N and N Derived from 15 N-Labeled Nitrate with Split-Root Systems
Soybean plants utilize the N fixed by root nodules and the N absorbed from the roots. However, the nodulation, nodule growth, and N2 fixation activity are inhibited when the nodulated roots are in direct contact with high concentrations of combined N, especially nitrate [13][14][15][16][17]. The effects of nitrate on nodule growth vary by nitrate concentrations, treatment period, placement in the medium [18,19], and legume species [20,21]. In addition, local and systemic effects of NO3 − are distinguished; local (direct) effects where

Changes in Total N and N Derived from 15 N-Labeled Nitrate with Split-Root Systems
Soybean plants utilize the N fixed by root nodules and the N absorbed from the roots. However, the nodulation, nodule growth, and N 2 fixation activity are inhibited when the nodulated roots are in direct contact with high concentrations of combined N, especially nitrate [13][14][15][16][17]. The effects of nitrate on nodule growth vary by nitrate concentrations, treatment period, placement in the medium [18,19], and legume species [20,21]. In addition, local and systemic effects of NO 3 − are distinguished; local (direct) effects where nitrate is supplied directly to the nodulated roots and systemic (indirect) effects where nitrate applied from the distal part of the roots [18,19]. Fijikake et al. [12,22] found that the direct effect of 5 mM nitrate on the soybean nodule growth and nitrogen fixation activity was quick and reversible in soybean plants. Nodule growth was suppressed at several hours after addition of 5 mM of nitrate [23,24]. Transcriptome and metabolome analysis after one day of 5 mM NO 3 − treatment supported that nitrate promotes nitrogen and carbon metabolism in the roots but repressed it in the nodules [25]. The quick and reversible inhibition of nodule growth and nitrogen fixation activity was also observed by ammonium, urea, and glutamine, although the inhibitory effects were less than nitrate [26].
In this experiment, the total amounts of N were not significantly different among N0-N0, 15 N5-N0, and 15 N5-N5 treatments at 2 DAT and 4 DAT, indicating that the amount of N from absorbed NO 3 − was almost equivalent to the decrease in the fixed N due to repression of nitrogen fixation by NO 3 − under this experiment conditions. Based on the total N accumulation and 15 N absorption from 2 DAT to 4 DAT, the percentage dependences on Ndfa and Ndfn were estimated. The %Ndfa decreased to 59% and 46% when one half-root or both half-roots supplied with NO 3 − . The amount of 15 N absorption was higher in the second period of 2 DAT to 4 DAT than the first period from 0 DAT to 2 DAT in both 15 N-N0 and 15 N-N5 treatments. These results might be due to the induction period of nitrate transporters in the roots or the accelerating growth of the plants. In all treatments, the increase of N in leaves from 2 DAT to 4 DAT was remarkable for about 3/4 of the total N increase, because the leaf growth was vigorous in this stage (about V4). In addition, N fixation activity in this stage might become increasing.
The 40% higher 15 N absorption in the second 2 days from 2 DAT to 4 DAT with 15 N5-N0 than that with 15 N5-N5 ( Figure 3) and the higher 15 N% in the roots and nodules in [L] pot with N-free solution ( Figure 4) might be due to the compensation of nitrate absorption due to only half-roots absorbing NO 3 − . Similar upregulation of NO 3 − absorption was reported in Brassica napus L. using split-root systems, and the uptake rates by half-roots with NO 3 − increased by about two-fold when the opposite side of half-root was in N-free solution within 24 h [27]. Forde [2] suggested that NO 3 − -fed plants can compensate for a local N limitation and maintain their N status by stimulating NO 3 − acquisition by roots exposed to mineral N. A short-term adaptation response may be mediated by a rapid induction of the nitrate uptake systems in these roots. High-affinity nitrate transporters involved in these responses were identified in Arabidopsis [28,29]. From the results by Laine [27], the nitrate absorption rates correlated with total N and NO 3 − concentrations in the shoot, but not with concentrations of each amino acid. The author suggested that total N and NO 3 − concentrations in the shoot may regulate the nitrate absorption rate in Brassica. Alternatively, stimulation of the NO 3 − absorption in the NO 3 − supplied halfroots with N-free solution in the opposite half-roots might be related to the promotion of the photoassimilate supply to the half-roots with 5 mM NO 3 − compared with the N-free solution [12]. When plant leaves were exposed to 11 C-labeled CO 2 to the soybean plants with split-root systems, of which one half-root was supplied with 5 mM NO 3 − and the opposite half-roots received an N-free solution, the translocation of 11 C labeled-C in the nitrate-fed half-root was stimulated compared with N-free half-roots [12].
Although 15 N absorption by the half-roots in the 15  This result indicated that nitrate accumulation in the lower roots is independent of the N conditions of the other half-roots. The lower nodules, which are also in direct contact with 5 mM NO 3 − accumulated less than 1/5 of the lower roots ( Figure 5A,B). The NO 3 − absorption in nodules was reported through a nodule surface, and not transported via the xylem or phloem in the roots [31]. The NO 3 − accumulation in the basal roots, stems, and leaves was observed even at 2 DAT, but NO 3 − was not detected in the other half-roots with N-free pot in 15 N5-N0 as same as N0-N0 treatment. This result clearly shows that NO 3 − itself did not readily recycle from leaves to the roots. Li et al. [32] reported that NO 3 − absorbed from the N-supply side of the roots of the dual-root system of soybean might be transported via the basal root pealed skin (via phloem) and woody part of the roots (via xylem). The discrepancy between our results and theirs may be the split-root system by cutting a primary root or the dual-root system made by grafting two seedlings. Most of the recycled N from the shoot may be transported through the phloem. [33].

Changes in Amides, Amino Acids, and Ureides in Each Part of the Plants with Split-Root Systems
Phloem-borne amino acids have been prominent candidates for the role of shootderived signal for feedback regulation of NO 3 uptake [34,35]. Rapid cycling of amino acids between the shoot and the root may occur, so the changes in the N status in the shoot will rapidly reflect the delivery rate of amino acids to the roots [36,37]. However, contrary to the requirements of this model, N deficiency can sometimes lead to an increase rather than a decline in amino acid cycling [38]. Furthermore, split-root studies on mung bean (Ricinus communis) found no correlation between NO 3 − uptake rate and the amino acid content of the phloem [39].
There are many hypotheses for the causes of nitrate inhibition of nodulation and nitrogen fixation, carbohydrate deprivation in nodules, feedback inhibition by a product of nitrate metabolism such as asparagine or ureides (allantoate and allantoin) [31,32,40,41], and the decrease in oxygen diffusion into the nodules which restricts the respiration of bacteroid, a symbiotic state of rhizobia in nodules [36,[42][43][44]. The concentrations of Asn in the lower roots supplied with 5 mM nitrate at 4 DAT were significantly higher at 35 µmol g −1 DW in 15 N5-N5 treatment than those in N0-N0 treatment about 13 µmol g −1 DW ( Figure 6B). The same was true for the half-roots with 15 N5 [L] pot at 35 µmol g −1 DW and N0 [R] pot at 15 µmol g −1 DW ( Figure 6D). These results indicate that some part of nitrate absorbed by the half-roots in direct contact with 5 mM NO 3 − was assimilated there, and the metabolite Asn accumulated. The stimulation of Asn accumulation in the opposite side of N-free roots was not observed, so the acceleration of Asn transport from the 15 NO 3 −fed half-roots via shoot to the N-free roots did not occur in these experimental conditions. The changes in the concentrations of N compounds in nodules were not apparent either in the +N or −N site (Figure 7), suggesting that NO 3 − absorption by root nodules from culture solution is much less than that from the roots.

Conclusions
Characteristics of absorption, transport, and recycling of nitrogen investigated using split-root systems of nodulated soybean plants, supplying 15 NO 3 − to one half-root, and another half-root with N-free solution or non-labeled nitrate. A high accumulation of nitrate was detected in the roots part in direct contact with 5 mM NO 3 − , but nitrate was absent in the roots and nodules in the N-free pot, suggesting that nitrate itself did not recycle as it was from the shoot to the roots. The 15 NO 3 − absorption in the half-roots with N-free solutions in opposite half-roots was 40% accelerated from 2 to 4 days of nitrate treatment, compared with those with non-labeled NO 3 − . The recycling percentages of 15 N in the roots plus nodules in the opposite N-free and non-labeled NO 3 − half-root were the same at 2-3%. The concentrations of amino acids, especially Asn, increased in the half-roots in direct contact with NO 3 − as well as basal roots, stems, and leaves but Asn did not accumulate in the opposite half-roots with N-free solution. Funding: This research received no external funding.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.