Heterologous Expression of Nitrate Assimilation Related-Protein DsNAR2.1/NRT3.1 Affects Uptake of Nitrate and Ammonium in Nitrogen-Starved Arabidopsis

Nitrogen (N) is an essential macronutrient for plant growth. Plants absorb and utilize N mainly in the form of nitrate (NO3−) or ammonium (NH4+). In this study, the nitrate transporter DsNRT3.1 (also known as the nitrate assimilation-related protein DsNAR2.1) was characterized from Dianthus spiculifolius. A quantitative PCR (qPCR) analysis showed that the DsNRT3.1 expression was induced by NO3−. Under N-starvation conditions, the transformed Arabidopsis seedlings expressing DsNRT3.1 had longer roots and a greater fresh weight than the wild type. Subcellular localization showed that DsNRT3.1 was mainly localized to the plasma membrane in Arabidopsis root hair cells. Non-invasive micro-test (NMT) monitoring showed that the root hairs of N-starved transformed Arabidopsis seedlings had a stronger NO3− and NH4+ influx than the wild-type seedlings, using with NO3− or NH4+ as the sole N source; contrastingly, transformed seedlings only had a stronger NO3− influx when NO3− and NH4+ were present simultaneously. In addition, the qPCR analysis showed that the expression of AtNRT2 genes (AtNRT2.1–2.6), and particularly of AtNRT2.5, in the transformed Arabidopsis differed from that in the wild type. Overall, our results suggest that the heterologous expression of DsNRT3.1 affects seedlings’ growth by enhancing the NO3− and NH4+ uptake in N-starved Arabidopsis. This may be related to the differential expression of AtNRT2 genes.

OsNAR2.1 knockdown. Yeast two hybrid screening showed that OsNAR2.1 interacted with OsNRT2.1, OsNRT2.2, and OsNRT2.3a [18]. Furthermore, the arginine 100 and aspartic acid 109 of OsNAR2.1 were found to be key amino acids in their interaction with OsNRT2.3a, and their interaction occurs in the plasma membrane [19]. These studies demonstrate that NRT3 is essential for high-affinity NRT2 in NO 3 − uptake in Arabidopsis and rice.
Dianthus spiculifolius Schur (Caryophyllaceae), a perennial herbaceous flowering plant, is a high-economic value ornamental crop with a high economic value that has strong resistance to cold, drought, and barren conditions. Studying its tolerance toward barren soil contributes to expanding its application and production in barren soils. The NRT genes play key roles in N absorption and utilization by plants. In this study, the DsNRT3.1 gene was identified from the transcriptome of D. spiculifolius [20]. Its expression in response to different N sources was investigated using qPCR. The subcellular localization of DsNRT3.1 in plant cells was observed using green fluorescent protein (GFP) as a marker. We compared the phenotypes of the transformed Arabidopsis seedlings expressing DsNRT3.1 and wild-type seedlings with different N sources (NO 3 − or NH 4 + ). We monitored the net fluxes of NO 3 − and NH 4 + in transformed Arabidopsis root hairs with different N sources using non-invasive micro-test (NMT) technology.

Expression Analysis of DsNRT3.1
The complete cDNA sequence of DsNRT3.1 was isolated from the D. spiculifolius transcriptome, which contains a 585 bp open reading frame that encodes 194 amino acids and has a predicted molecular mass of 21.07 kDa. The sequence and phylogenetic tree analysis revealed that the DsNRT3.1 protein was similar to AtNRT3.1 (42.86% amino acid sequence identity) and NRT3.1 from Arabidopsis and other species ( Figure 1A,B). Two transmembrane domains of the DsNRT3.1 protein were predicted using HMMTOP software ( Figure 1C).

Effect of Heterologous Expression of DsNRT3.1 on Arabidopsis Seedings Growth
Transformed Arabidopsis seedlings expressing DsNRT3.1, generated using the 35S promoter, were used to investigate the role of DsNRT3.1 in plant growth. The expression of DsNRT3.1 in the transformed Arabidopsis lines was confirmed using a semi-quantitative RT-PCR ( Figure 3A). After 10 days on N-free 1/2 strength Murashige and Skoog (MS) medium, wild-type and transformed seedlings showed the N-starvation phenotype, but there was no significant difference in the primary root length. On N-free 1/2 MS medium supplemented with NO 3 − or NH 4 + (0.01 or 0.02 mM) as the sole N source, the primary roots of the transformed seedlings were significantly longer than those of the wild-type seedlings. At NO 3 − or NH 4 + concentrations over 1 mM, there was no significant difference in primary root length between the wild-type and transformed seedlings ( Figure 3B-D). Furthermore, the fresh weight of the transformed seedlings was generally higher than that of the wild-type seedlings after 20 days on N-free 1/2 MS medium supplemented with NO 3 − (0.05 to 0.5 mM) or NH 4 + (0.1 mM) ( Figure 4A-C).
These results show that with low-concentration NO 3 − or NH 4 + as the sole N source, the transformed seedlings grew better than the wild-type seedlings.

Effect of Heterologous Expression of DsNRT3.1 on Arabidopsis Seedings Growth
Transformed Arabidopsis seedlings expressing DsNRT3.1, generated using the 35S promoter, were used to investigate the role of DsNRT3.1 in plant growth. The expression of DsNRT3.1 in the transformed Arabidopsis lines was confirmed using a semi-quantitative RT-PCR ( Figure 3A). After 10 days on N-free 1/2 strength Murashige and Skoog (MS) medium, wild-type and transformed seedlings showed the N-starvation phenotype, but there was no significant difference in the primary root length. On N-free 1/2 MS medium supplemented with NO3 − or NH4 + (0.01 or 0.02 mM) as the sole N source, the primary roots of the transformed seedlings were significantly longer than those of the wild-type seedlings. At NO3 − or NH4 + concentrations over 1 mM, there was no significant difference in primary root length between the wild-type and transformed seedlings ( Figure 3B-D). Furthermore, the fresh weight of the transformed seedlings was generally higher than that of the wild-type seedlings after 20 days on N-free 1/2 MS medium supplemented with NO3 − (0.05 to 0.5 mM) or NH4 + (0.1 mM) ( Figure 4A-C). These results show that with low-concentration NO3 − or NH4 + as the sole N source, the transformed seedlings grew better than the wild-type seedlings.  Different letters indicate significant differences (Student's t-test; p < 0.05). Error bars represent the standard error (n = 6).

Effect of Heterologous Expression of DsNRT3.1 on the NO 3 − and NH 4 + Fluxes in Arabidopsis Root Hairs
It is thought that DsNRT3.1 is a membrane protein ( Figure 1C); therefore, we examined the DsNRT3.1 localization in Arabidopsis using GFP as a marker. In the root hair cells of Arabidopsis stably expressing DsNRT3.1-GFP, the DsNRT3.1-GFP signals were localized mainly to the cell periphery, which is similar to the plasma membrane. However, as a control, the GFP signal was generally distributed throughout the root hair cells of Arabidopsis stably expressing GFP ( Figure 5A). Root hairs are a primary site for nutrient uptake in plants [21]. The net NO 3 − and NH 4 + fluxes in the root hairs of wild-type and transformed Arabidopsis were monitored and compared using NMT ( Figure 5B).

Effect of Heterologous Expression of DsNRT3.1 on the NO3 − and NH4 + Fluxes in Arabidopsis Root Hairs
It is thought that DsNRT3.1 is a membrane protein ( Figure 1C); therefore, we examined the DsNRT3.1 localization in Arabidopsis using GFP as a marker. In the root hair cells of Arabidopsis stably expressing DsNRT3.1-GFP, the DsNRT3.1-GFP signals were localized mainly to the cell periphery, which is similar to the plasma membrane. However, as a control, the GFP signal was generally distributed throughout the root hair cells of Arabidopsis stably expressing GFP ( Figure 5A). Root hairs are a primary site for nutrient uptake in plants [21]. The net NO3 − and NH4 + fluxes in the root hairs of wild-type and transformed Arabidopsis were monitored and compared using NMT ( Figure 5B). Seedlings grown on the 1/2 MS medium or N-free 1/2 MS medium (N-starvation treatment) for 10 days were transferred to a test solution containing KNO3 (0.1 mM), NH4Cl (0.1 mM), or NO3NH4 (0.5 mM) as the sole N source. With KNO3 or NH4Cl as the sole N source, the root hairs of the wildtype and transformed seedlings cultured on the 1/2 MS medium displayed a marked NO3 − or NH4 + efflux, with no significant difference in the efflux rates between the transformed and wild-type seedlings ( Figure 6A-D). However, there was marked NO3 − or NH4 + influx displayed in the root hairs Seedlings grown on the 1/2 MS medium or N-free 1/2 MS medium (N-starvation treatment) for 10 days were transferred to a test solution containing KNO 3 (0.1 mM), NH 4 Cl (0.1 mM), or NO 3 NH 4 (0.5 mM) as the sole N source. With KNO 3 or NH 4 Cl as the sole N source, the root hairs of the wild-type and transformed seedlings cultured on the 1/2 MS medium displayed a marked NO 3 − or NH 4 + efflux, with no significant difference in the efflux rates between the transformed and wild-type seedlings ( Figure 6A-D). However, there was marked NO 3 − or NH 4 + influx displayed in the root hairs of the N-starved wild-type and transformed seedlings ( Figure 6E,G); the mean influx rate in the transformed seedlings was significantly higher than in the wild-type seedlings ( Figure 6F,H). With NO3NH4 as the sole N source, the root hairs of N-starved transformed seedlings showed a higher NO3 − influx rate than those of the wild type ( Figure 7E,F), whereas the NH4 + influx rate was similar to that of the wild type ( Figure 7G,H). Moreover, the NO3 − and NH4 + efflux from the root hairs of the wild-type or transformed seedlings cultured on 1/2 MS medium was not marked and showed no significant difference among the lines ( Figure 7A-D). This indicates that the root hairs of N-starved transformed seedlings have a stronger NO3 − uptake ability than those of wild-type seedlings; they also have a stronger NH4 + uptake ability in the absence of NO3 − , suggesting that the heterologous expression of DsNRT3.1 affects the NO3 − and NH4 + uptake in N-starved Arabidopsis seedlings. With NO 3 NH 4 as the sole N source, the root hairs of N-starved transformed seedlings showed a higher NO 3 − influx rate than those of the wild type ( Figure 7E,F), whereas the NH 4 + influx rate was similar to that of the wild type ( Figure 7G,H). Moreover, the NO 3 − and NH 4 + efflux from the root hairs of the wild-type or transformed seedlings cultured on 1/2 MS medium was not marked and showed no significant difference among the lines ( Figure 7A-D). This indicates that the root hairs of N-starved transformed seedlings have a stronger NO 3 − uptake ability than those of wild-type seedlings; they also have a stronger NH 4 + uptake ability in the absence of NO

Effect of Heterologous Expression of DsNRT3.1 on Arabidopsis NRT2 Genes Expression
NRT3, as a partner protein to the NTR2 family, is important for NRT2 function [15]. The expression levels of AtNRT2 genes in the wild-type and transformed Arabidopsis seedlings were compared. The qPCR analysis revealed that the expression levels of six of the seven NRT2 family members differed between the transformed Arabidopsis and wild-type seedlings; NRT2.1, 2.2, 2.4, and 2.6 were down-regulated and NRT2.5 was up-regulated. However, the expression of AtNRT3. 1 and AtNRT2.7 in transformed Arabidopsis was similar to that in the wild type, with no significant difference (Figure 8). This suggests that the heterologous expression of DsNRT3.1 affects the expression of members of the AtNRT2 gene family in transformed Arabidopsis seedlings.

Effect of Heterologous Expression of DsNRT3.1 on Arabidopsis NRT2 Genes Expression
NRT3, as a partner protein to the NTR2 family, is important for NRT2 function [15]. The expression levels of AtNRT2 genes in the wild-type and transformed Arabidopsis seedlings were compared. The qPCR analysis revealed that the expression levels of six of the seven NRT2 family members differed between the transformed Arabidopsis and wild-type seedlings; NRT2.1, 2.2, 2.4, and 2.6 were down-regulated and NRT2.5 was up-regulated. However, the expression of AtNRT3.1 and AtNRT2.7 in transformed Arabidopsis was similar to that in the wild type, with no significant difference (Figure 8).
The roots of N-starved transformed DsNRT3.1 seedlings were longer than those of the wild type under low N conditions, with NO3 − or NH4 + as the sole N source (Figure 3). After longer periods of culture, the fresh weight of the transformed DsNRT3.1 seedlings was generally higher than that of the wild type ( Figure 4). With KNO3 as the sole N source, the transformed seedlings had longer roots and higher root NO3 − influx rates than the wild type ( Figure 3C and 6E,F). Likewise, with NH4NO3 as the sole N source, the transformed seedling roots had higher NO3 − influx rates than the wild type ( Figure 7E,F). These results suggest that longer root growth in transformed seedlings may be associated with a stronger NO3 − uptake. A confocal observation using GFP as a marker revealed that DsNRT3.1 was mainly localized to the plasma membrane in Arabidopsis root hair cells ( Figure 5A). Furthermore, NMT monitoring showed that the root hairs of N-starved transformed seedlings had a stronger NO3 − and NH4 + influx than those of the wild-type seedlings ( Figure 6A-D). Several studies have shown that NRT3.1 regulates NO3 − uptake via interaction with NRT2 members on the plasma membrane [15,18,19,23]. Thus, we hypothesized that the heterologous expression of DsNRT3.1 might affect the NO3 − uptake in N-starved seedlings via interaction with AtNRT2 members on the plasma membrane. We compared the expression levels of AtNRT2 genes in the transformed and wild-type seedlings. The qPCR analysis revealed that the expression of multiple NRT2 members (AtNRT2.1-2.6), and particularly of AtNRT2.5, was altered in transformed Arabidopsis-expressing DsNRT3.1 (Figure 8). Arabidopsis AtNRT2.5 is a high-affinity NO3 − transporter localized on the plasma membrane and is expressed in the root hair zone [25]. Under long-term N-starvation, AtNRT2.5

Discussion
The first NRT3.1 was identified from Chlamydomonas [13], and its orthologous proteins were subsequently identified in barley (HvNRT3.1/HvNAR2.1) [22], Arabidopsis (AtNRT3.1) [16], rice (OsNRT3.1/OsNAR2.1) [18], and chrysanthemums (CmNRT3/CmNAR2) [23]. A sequence analysis showed that DsNRT3.1 was highly similar to AtNRT3.1, HvNRT3.1, and OsNRT3.1/OsNAR2.1 ( Figure 1A,B). NRT3.1 was considered to be a high-affinity transporter; it was induced at low concentrations of N sources, whereas its expression was largely unaffected at high concentrations or using saturated N sources. Therefore, we performed a gene expression analysis after N-starvation. In N-starved seedlings, DsNRT3.1 was also induced by NO 3 − (Figure 2). Similarly, the expression of AtNRT3.1, OsNRT3.1/OsNAR2.1, and CmNRT3/CmNAR2 was also induced by NO 3 − in N-starved seedlings [16,18,23]. However, the expression of OsNRT3.1/OsNAR2.1 was almost unaffected by NH 4 + [18]. We found that DsNRT3.1 expression was not induced by NH 4 + ( Figure 2B). In addition, K + affected the DsNRT3.1 expression (Figure 2A). NRT1.5 from the NRT family serves as a proton-coupled H + /K + antiporter for K + loading in Arabidopsis [24]. The roots of N-starved transformed DsNRT3.1 seedlings were longer than those of the wild type under low N conditions, with NO 3 − or NH 4 + as the sole N source (Figure 3). After longer periods of culture, the fresh weight of the transformed DsNRT3.1 seedlings was generally higher than that of the wild type ( Figure 4). With KNO 3 as the sole N source, the transformed seedlings had longer roots and higher root NO 3 − influx rates than the wild type ( Figures 3C and 6E,F). Likewise, with NH 4 NO 3 as the sole N source, the transformed seedling roots had higher NO 3 − influx rates than the wild type ( Figure 7E,F). These results suggest that longer root growth in transformed seedlings may be associated with a stronger NO 3 − uptake. A confocal observation using GFP as a marker revealed that DsNRT3.1 was mainly localized to the plasma membrane in Arabidopsis root hair cells ( Figure 5A). Furthermore, NMT monitoring showed that the root hairs of N-starved transformed seedlings had a stronger NO 3 − and NH 4 + influx than those of the wild-type seedlings ( Figure 6A-D).
Several studies have shown that NRT3.1 regulates NO 3 − uptake via interaction with NRT2 members on the plasma membrane [15,18,19,23]. Thus, we hypothesized that the heterologous expression of DsNRT3.1 might affect the NO 3 − uptake in N-starved seedlings via interaction with AtNRT2 members on the plasma membrane. We compared the expression levels of AtNRT2 genes in the transformed and wild-type seedlings. The qPCR analysis revealed that the expression of multiple NRT2 members (AtNRT2.1-2.6), and particularly of AtNRT2.5, was altered in transformed Arabidopsis-expressing DsNRT3.1 (Figure 8). Arabidopsis AtNRT2.5 is a high-affinity NO 3 − transporter localized on the plasma membrane and is expressed in the root hair zone [25]. Under long-term N-starvation, AtNRT2.5 becomes the most abundant transcript amongst the seven AtNRT2 members in Arabidopsis shoots and roots [25]. Subsequent studies have revealed that AtNRT3.1 and AtNRT2.5 form a complex on the plasma membrane; this complex is a major contributor to high-affinity NO 3 − influx in Arabidopsis [17].
However, in our study, the AtNRT3.1 expression in transformed seedlings was not altered (Figure 8). Recent studies have shown that AtNRT1.1 participates in the NH 4 + uptake by affecting the NH 4 + transporter expression [26].

Plant Materials and Growth Conditions
Seeds of D. spiculifolius and A. thaliana (Columbia-0) were surface sterilized using 30% (v/v) bleach solution for 6 min and rinsed five times with sterile water. The sterilized seeds were grown on 1/2 strength Murashige and Skoog (MS) medium ( For the different N-source treatments, D. spiculifolius seedings were N-starved for 1 week, then transferred to N-free 1/2 MS medium (3% sucrose, 1% agar, pH 5.8) supplemented with 1 mM KNO 3 , 1 mM KCl, 1 mM NH 4 NO 3 , or 0.5 mM (NH 4 ) 2 SO 4 as the sole N source. At least 10 seedlings from each treatment were harvested and pooled at different time points (0, 3, 6, 12, or 24 h after treatments), frozen immediately in liquid N, and stored at −80 • C for RNA preparation.

RNA Preparation and Expression Analysis
The total RNA (1.2 µg) was extracted using an RNeasy ® Mini Kit (Qiagen, Valencia, CA, USA), and cDNA (20 µL) was synthesized using an M-MLV RTase cDNA Synthesis Kit (TaKaRa, Shiga, Japan), according to the manufacturers' instructions. The qPCR analysis was performed using Bio-Rad CFX Manager Software Version 3.1 (Bio-Rad, Hercules, CA, USA) using the appropriate pairs of species-specific primers (Supplementary Materials Table S1). The reaction components per 20 µL were as follows: 7 µL H 2 O, 10 µL Hieff qPCR SYBR Green Master Mix (YEASEN, Shanghai, China), 1 µL 10 µM of each primer and 1 µL cDNA. The thermal cycling program was as follows: initial denaturation at 95 • C for 60 s, and 40 cycles at 95 • C for 10 s, 60 • C for 30 s, and 72 • C for 30 s. DsActin or AtActin was used as an internal reference gene. The relative quantification of gene expression was evaluated using the delta-delta-Ct method. Three biological replicates and three technical replicates were performed for each analysis.

Vector Construction and Plant Transformation
The open reading frame of DsNRT3.1 was constructed into the pBI121 vector driven by 35S promoter, using the following restriction sites: 121DsNRT3.1(X)-F (TCTAGAATGGCGGTGCGAGG ATTAAC; the XbaI site is underlined) and 121DsNRT3.1(S)-R (GAGCTCGTAGCTAGCTTCGACTT CTTC; the SacI site is underlined). To construct the DsNRT3.1-GFP fusion gene, the open reading frame of DsNRT3.1 without a stop codon was constructed into a pBI121-GFP vector using the 121DsNRT3.1(X)-F and 121DsNRT3.1(K)-R (GGTACCGTAGCTAGCTTCGACTTCTTC; the KpnI site is underlined) restriction sites. These pBI121-GFP constructs have been described elsewhere [27]. These constructs were transformed into the Agrobacterium tumefaciens strain EHA105 for plant transformation; Arabidopsis was transformed using the floral dip method [28]. The transformed plants were placed on 1/2 MS medium containing 30 µg mL −1 kanamycin. The primers used in this study are shown in Supplementary Materials Table S1.

Phenotypic Analysis of Transformed Arabidopsis
The sterilized wild-type and transformed Arabidopsis seeds were stratified for 3 days at 4 • C. The seedlings were then transferred to N-free 1/2 MS medium (3% sucrose, 1% agar, pH 5.8) supplemented with different concentrations of KNO 3 or NH 4 Cl (0, 0.01, 0.02, 0.05, 0.1, 0.3, 0.5, 1, or 2.5 mM) for 10 days before measuring root length, or 20 days before measuring the seedlings fresh weight. After examining seedling growth phenotypes, the plants were photographed using a Canon EOS 80D camera (Canon, Tokyo, Japan). Images were processed using Adobe Photoshop CS. Seedling root length was measured using Image Pro Plus 6.0 software (Media Cybernetics, Silver Spring, MD, USA). The fresh weight of the aerial parts of the seedlings was measured using an analytical balance with an accuracy of ± 0.1 mg.

NMT Measurement of Net NO 3 − and NH 4 + Fluxes in Arabidopsis Root Hairs
The net fluxes of NO 3 − and NH 4 + in the root hairs of the wild-type and transformed Arabidopsis seedlings were measured using NMT (NMT100 Series, YoungerUSA LLC, Amherst, MA, USA) as previously described [29]. The wild-type and transformed Arabidopsis seeds were surface sterilized and kept at 4 • C for 3 days in the dark. After germination, the seedlings were transferred into N-free 1/2 MS medium for 7 days, after which the roots of the seedlings were immediately equilibrated in a measuring solution containing 0.1 mM KNO 3 , 0.1 mM NH 4 Cl, or 0.1 mM NH 4 NO 3 for 10 min. The roots were fixed to the bottom of the plate using resin blocks and filter paper strips. To take flux measurements, the ion-selective electrodes were calibrated using NO 3 −

Statistical Analysis
The data were analyzed using a one-way analysis of variance in SPSS (SPSS, Inc., Chicago, IL, USA), and statistically significant differences were calculated using the Student's t-test, with p < 0.05 as the threshold for significance.

Conclusions
In this study, the nitrate assimilation-related protein DsNAR2.1/DsNRT3.1 was characterized from D. spiculifolius. DsNRT3.1 was localized mainly in the plasma membrane of Arabidopsis root hair cells. The findings reveal that the heterologous expression of DsNRT3.1 in Arabidopsis affects seedling root growth and NO 3 − and NH 4 + uptake under N-starvation, as well as the expression of AtNRT2 family members. Overall, our results suggest that DsNRT3.1 may affect the seedling growth in N-starved Arabidopsis by enhancing the NO 3 − and NH 4 + uptake, which may be associated with altered AtNRT2 family expression. Our study also suggests that DsNRT3.1 functions as a positive regulator of plant N uptake and could be utilized for the genetic improvement of N-efficient plants.