The nitrate transporter OsNPF7.9 mediates nitrate allocation and the divergent nitrate use efficiency between indica and japonica rice

Abstract

Nitrate allocation in Arabidopsis (Arabidopsis thaliana) represents an important mechanism for mediating plant environmental adaptation. However, whether this mechanism occurs or has any physiological/agronomic importance in the ammoniphilic plant rice (Oriza sativa L.) remains unknown. Here, we address this question through functional characterization of the Nitrate transporter 1/Peptide transporter Family (NPF) transporter gene OsNPF7.9. Ectopic expression of OsNPF7.9 in Xenopus oocytes revealed that the gene encodes a low-affinity nitrate transporter. Histochemical and in-situ hybridization assays showed that OsNPF7.9 expresses preferentially in xylem parenchyma cells of vasculature tissues. Transient expression assays indicated that OsNPF7.9 localizes to the plasma membrane. Nitrate allocation from roots to shoots was essentially decreased in osnpf7.9 mutants. Biomass, grain yield, and nitrogen use efficiency (NUE) decreased in the mutant dependent on nitrate availability. Further analysis demonstrated that nitrate allocation mediated by OsNPF7.9 is essential for balancing rice growth and stress tolerance. Moreover, our research identified an indica–japonica divergent single-nucleotide polymorphism occurring in the coding region of OsNPF7.9, which correlates with enhanced nitrate allocation to shoots of indica rice, revealing that divergent nitrate allocation might represent an important component contributing to the divergent NUE between indica and japonica subspecies and was likely selected as a favorable trait during rice breeding.

Nitrate allocation between roots and shoots contributes to the divergent nitrogen use efficiency between indica and japonica subspecies.

Introduction

Nitrate and ammonium represent two major nitrogen nutrients essential for plant growth. For most terrestrial plants, nitrate is the predominant nitrogen source, while for aquatic plants such as rice, it is believed that ammonium is preferred (Wang et al., 1993; Arth et al., 1998; Kronzucker et al., 1998). However, nitrogen levels in the rhizosphere fluctuate widely and frequently and thus are not always readily available to plants. To cope with this special situation, plants have evolved various mechanisms, and one well-investigated example is the nitrate low-affinity transport system (LATS) and high-affinity transport system (HATS) uptake system (Crawford and Glass, 1998). When nitrate concentration is >1 mM, plants preferentially use the LATS system, whereas when nitrate concentration goes ˂1 mM, the HATS system is activated and takes over nitrate uptake (Crawford and Glass, 1998; Forde, 2000).

Arabidopsis (Arabidopsis thaliana) is a typical terrestrial plant and preferentially uses nitrate as the nitrogen source. Nitrate transporter 1 (NRT1) and NRT2 families are two major gene families identified for LATS and HATS nitrate uptake in plants, respectively, among which AtNRT1.1/AtNPF6.3 is the first cloned nitrate transporter, displaying both high- and low-affinity nitrate transport activity, as well as nitrate signal sensing (Tsay et al., 1993; Huang et al., 1996; Wang et al., 1998; Liu et al., 1999, Ho et al., 2009). When exposed to nitrate fluctuating in a wide range, CIPK23 dynamically modulates the phosphorylation status of T101 in AtNRT1.1/AtNPF6.3, which will then switch between high- and low-affinity modes. The phosphorylated and dephosphorylated forms of AtNRT1.1/AtNPF6.3 can also sense and translate the environmental nitrate signal into different primary nitrate response (Ho et al., 2009). Once entering plants, nitrate is translocated from roots to shoots, and AtNRT1.5/AtNPF7.3 and AtNRT1.8/AtNPF7.2 have been identified as important regulators involved in this process (Lin et al., 2008; Li et al., 2010). AtNRT1.5/AtNPF7.3 functions to load nitrate into xylem vessels, whereas AtNRT1.8/AtNPF7.2 mediates nitrate unloading from the xylem. Long distance nitrate transport not only provides plants with essential mineral nutrients in aerial parts, but is also involved in stress response. When under stressful conditions, ethylene and jasmonic acid signaling pathways are activated to coordinate the upregulation of AtNRT1.8/AtNPF7.2 and downregulation of AtNRT1.5/AtNPF7.3, thus initiating the stress induced-nitrate allocation to roots (SINARs) process to retain nitrate in roots and to rebalance plant growth and stress tolerance (Li et al., 2010; Chen et al., 2012; Zhang et al., 2014), yet it remains to be investigated how this rebalance is realized. Studies have shown that AtNRT1.9/AtNPF2.9 also participates in the circulation of nitrate to the underground parts by loading nitrate into the phloem (Wang and Tsay, 2011), but whether it contributes to the SINAR process is still unknown. Nitrate allocated to the shoot is distributed to leaves and seeds by the coordination of several nitrate transporters (Chiu et al., 2004; Chopin et al., 2007; Almagro et al., 2008; Fan et al., 2009; Hsu and Tsay, 2013).

In rice (Oriza sativa L.), more and more evidence shows that nitrate is essential to nitrogen nutrition, although rice has long been believed to use ammonium as the main nitrogen source. Studies showed that substantial nitrification occurs in rhizosphere and up to 40% of the total nitrogen might be taken up in the form of nitrate (Kronzucker et al., 1998, 1999; Britto and Kronzucker, 2004; Kirk and Kronzucker, 2005). Moreover, indica rice (IND) generally shows higher nitrogen use efficiency (NUE) than japonica, which was recently demonstrated to be attributable largely to a natural allele of OsNRT1.1B/OsNPF6.5 (Hu et al., 2015), whereas OsNRT1.1A/OsNPF6.3 plays a very important role in nitrogen utilization by altering OsNLP3/OsNLP4 localization and expression of key genes in nitrogen uptake, transport, and assimilation (Wang et al., 2018). A most recent study revealed that OsNPF6.1^HapB, a rare natural allele controlling nitrate uptake and NUE, has been lost in over 90% of rice varieties due to nitrogen overuse (Tang et al., 2019). In terms of long-distance nitrate transport from roots to shoots, only very few studies have been reported in rice (Tang et al., 2012; Li et al., 2015; Xia et al., 2015), among which the HATS family members OsNRT2.3a was proposed to mediate nitrate redistribution in a LATS manner, although OsNRT2.3a appeared to function more efficiently when external nitrate level is lower (Tang et al., 2012). Interestingly, all these three genes reported in long-distance nitrate transport were demonstrated to substantially affect rice development, yet the underlying mechanisms remain largely elusive.

In this study, we identified OsNPF7.9 as an ortholog of Arabidopsis NRT1.5/AtNPF7.3, the major player in mediating nitrate loading into xylem vessels and long-distance transport to shoots. Our research demonstrated that nitrate allocation to shoots is essential for rice growth and impacts greatly on grain yield, and further revealed that in addition to nitrate uptake (Hu et al., 2015), the OsNPF7.9-mediated nitrate allocation to shoots also contributes to the long observed divergent NUE between indica and japonica subspecies (Koutroubas and Ntanos, 2003).

Results

Isolation and characterization of OsNPF7.9

To investigate whether SINAR might function in the ammoniphilic rice, we performed a phylogenetic analysis and found that one NRT1 family gene model Os02g0689900 is highly similar to Arabidopsis NRT1.8/AtNPF7.2 and NRT1.5/AtNPF7.3 (Figure 1A), which have been established to mediate SINAR in Arabidopsis (Li et al., 2010; Chen et al., 2012; Zhang et al., 2014). The predicted protein sequence of Os02g0689900 contains 610 amino acids and shows 55.6% and 55.7% identity to AtNRT1.5/AtNPF7.3 and AtNRT1.8/AtNPF7.2, respectively (Figure 1B). Os02g0689900 has four exons and three introns, encoding a predicted protein with 12 transmembrane (TM) domains as observed in all other NRT1 family members. Further analysis showed that when exposed to increased nitrate, Os02g0689900 expression was significantly elevated in roots, while no apparent induction was observed in shoots (Figure 1C). Moreover, Os02g0689900 was expressed most strongly in roots, whereas in above ground parts, the expression level was very low in all the tested tissues except in the culm (Supplemental Figure S1), suggesting that Os02g0689900 might be the AtNRT1.5/AtNPF7.3 ortholog in rice, and OsNFP7.9 was assigned to this gene according to the unified nomenclature system (Léran et al., 2014).

Figure 1.

Sequence alignment and expression analysis of OsNPF7.9. A, Phylogenetic tree of functionally characterized Arabidopsis NPF genes and OsNPF7.9/Os02g0689900. The CDSs were downloaded from NCBI. Mega 6.06 was used to perform the multiple alignment and to draw the phylogenetic tree, and the tree was drawn using iTOL (http://itol.embl.de/). B, Alignment of OsNPF7.9 with NRT1.5 and NRT1.8. Amino acid sequence alignment was performed using the GeneDoc. Black and shaded regions indicate identical residues and conservative substitutions, respectively. The putative TM domains are underlined and numbered. C, Two-week-old seedlings were subjected to nitrogen starvation for 4 d, then nitrate induction was performed by incubating the plants with 5 mM NH₄Cl (−${\text{NO}}_3^{-}$) or with NH₄NO₃ (+${\text{NO}}_3^{-}$) for 2 h before RNA extraction and RT-qPCR determination of OsNPF7.9 expression. Data were normalized to those of OsActin1. Values are mean ± standard deviation (sd), n = 6 plants. Asterisk above the column indicates significant differences by Student’s t test, ^**P < 0.01

Both the microarray data from Rice Expression Database (http://expression.ic4r.org/) and our results (Supplemental Figure S1) showed that OsNPF7.9 expresses mainly in roots. To further investigate the expression pattern of OsNPF7.9, we generated rice transgenic lines expressing β-glucuronidase (GUS) driven by the OsNPF7.9 promoter. As shown in Figure 2, A–D, GUS activity was detected in vascular tissues of rice roots, leaves, and young panicles. Root cross-sectioning analysis further revealed that OsNPF7.9 was exclusively expressed in xylem parenchyma cells of the vasculature (Figure 2B). mRNA in-situ hybridization also revealed that OsNPF7.9 transcripts accumulated in xylem parenchyma cells (Figure 2, E and F), consistent with the GUS staining analysis.

Figure 2.

OsNPF7.9 expression in xylem parenchyma cells within vasculature. A–D, Histochemical localization of GUS gene activity in transgenic rice expressing the GUS reporter gene driven by the OsNPF7.9 promoter. GUS staining of root (A), a root cross section (B), leaf blade (C), young panicle (D). E and F, In situ hybridization using the sense or antisense OsNPF7.9 probe hybridized to a cross section of rice root. Scale bars are 0.5 mm (A), 100 µm (B), 5 mm (C and D), 50 μm (E and F).

To determine the subcellular localization, we fused OsNPF7.9 in frame with the enhanced yellow fluorescent protein (eYFP) tag and transiently expressed the construct in rice mesophyll protoplasts and onion epidermal cells driven by the cauliflower mosaic virus 35S promoter. The results showed that in contrast to the diffuse nucleocytoplasmic localization of eYFP (Figure 3, A and D), OsNPF7.9:eYFP was consistently observed in the plasma membrane (Figure 3, B and C), indicating that OsNPF7.9 is plasma membrane-localized.

Figure 3.

Subcellular localization of OsNPF7.9. The eYFP control and fusion protein OsNPF7.9-eYFP were expressed in rice mesophyll protoplasts (A and B) or onion epidermal cells (C and D). Green color represents eYFP and red represents chlorophyll fluorescence. Fluorescence imaging was performed using confocal microscope. A, eYFP control, a-1: chlorophyll, a-2: bright field image, a-3: merged image. B, OsNPF7.9-eYFP. b-1: chlorophyll, b-2: bright field image, b-3: merged image. C and D, Onion epidermal cells were incubated in 0.8 M mannitol to induce plasmolysis, and fluorescence was imaged for OsNPF7.9-eYFP (C) and eYFP control (D).

OsNPF7.9 encodes a low-affinity nitrate transporter

To characterize the biochemical feature of OsNPF7.9, we then injected the OsNPF7.9 capped RNA (cRNA) into oocyte cells. As shown in Figure 4A, when incubated with 10 mM ¹⁵NO₃⁻ at pH 5.5, significantly more nitrate accumulated in the cRNA-injected oocytes than in the water-injected cells, while no difference was observed between them when incubated with 0.25 mM ¹⁵NO₃⁻ at pH 5.5 or 10 mM ¹⁵NO₃⁻ at pH 7.4. To further determine the nitrate uptake affinity of OsNPF7.9, nitrate uptake activity was measured under different ¹⁵NO₃⁻ concentration ranging from 0.25 to 30 mM at pH 5.5, and a K_m value of 7.59 ± 1.46 was estimated by fitting to the Michaelis–Menten equation (Figure 4B). Oocytes were further voltage clamped and inward currents were recorded in the OsNPF7.9 cRNA-injected oocytes which were incubated with 10-mM nitrate at pH 5.5 (Figure 4C). All these results indicated that OsNPF7.9 mediates nitrate transport in a pH-dependent and low affinity manner.

Figure 4.

OsNPF7.9 mediates low-affinity nitrate transport. A, Nitrate uptake assays were performed after 3-h incubation with 0.25 mM or 10 mM K¹⁵NO₃, at pH 5.5 or 7.4 as displayed. B, Nitrate uptake kinetics assay. Oocytes injected with OsNPF7.9 cRNA were incubated with different concentrations of K¹⁵NO₃ at pH 5.5 for 3 h, and the Km was calculated by fitting to the Michaelis–Menten equation. C, Current–voltage curves recorded for OsNPF7.9 cRNA-injected oocytes. Oocytes were voltage clamped at −60 mV and stepped into a test voltage between 0 and −140 mV for 300 ms. Currents (I) shown here are the difference between OsNPF7.9 cRNA or water-injected oocytes in the 10-mM nitrate at pH 5.5. D, Nitrate efflux assay. K¹⁵NO₃ was injected into oocytes that were pre-injected with OsNPF7.9 cRNA or CHL1/AtNPF6.3 cRNA, or water. ¹⁵${\text{NO}}_3^{-}$ level in the oocytes was determined right after injection (time = 0) or after 3-h incubation in nitrate-free ND96 buffer, at pH 5.5 or 7.4. Values are mean ± sd (n = 6–10). Different letters above the column indicate significant differences by ANOVA Tukey's multiple comparison, P-value < 0.05. Vm represents the TM potential.

Considering that nitrate efflux activity was observed for the Arabidopsis NRT1.5/AtNPF7.3 (Lin et al., 2008), we then injected 5 nmol K¹⁵NO₃ into the OsNPF7.9 cRNA, water or CHL1/AtNPF6.3 cRNA-injected oocytes, the latter two serving as a negative and a positive control, respectively. As shown in Figure 4D, no obvious difference was observed in the amount of nitrate retained in water-injected oocytes, while in the CHL1/AtNPF6.3 cRNA-injected oocytes, nitrate level decreased to 25% or 12% of the original amount after 3-h incubation in nitrate-free ND96 solution at pH 5.5 or 7.4, indicating that CHL1/AtNPF6.3 has significant nitrate efflux activity at pH 5.5 and 7.4. In the OsNPF7.9 cRNA-injected oocytes, nitrate retained in cells decreased to 70% or 83% of the original level after 3-h incubation in nitrate-free ND96 solution at pH 5.5 or 7.4. This result suggests that although OsNPF7.9 was not as efficient as CHL1/AtNPF6.3, it did show substantial nitrate efflux activity at PH 5.5 or 7.4, therefore, it is a bi-directional nitrate transporter, as was reported for the Arabidopsis NRT1.5/AtNPF7.3.

Isolation and characterization of osnpf7.9 mutants

To functionally characterize OsNPF7.9, two independent mutants osnpf7.9-1 and osnpf7.9-2 were obtained by screening the Targeting Induced Local Lesions IN Genomes (TILLING) library. Single-nucleotide C-to-T transitions occurred in osnpf7.9-1 (C1301T) and osnpf7.9-2 (C1289T), which resulted in the substitution of Thr-434 with an Ile-434 residue and Thr-430 with a Met-430 residue, respectively (Supplemental Figure S2A). Reverse transcription quantitative PCR (RT-qPCR) indicated that OsNPF7.9 expression levels were comparable between the mutant and the wild-type (WT) plants (Supplemental Figure S2B). Further nitrate uptake assay showed that injection of the osnpf7.9-1 and osnpf7.9-2 cRNA into oocytes did not increase nitrate uptake, in contrast to what was observed in the OsNPF7.9-injected oocytes (Supplemental Figure S2C). Given that OsNPF7.9 also mediates nitrate efflux, we then determined nitrate efflux activity in both the OsNPF7.9 and osnpf7.9 cRNA-injected oocytes. The results showed that nitrate retained in the OsNPF7.9-injected oocytes was lower than those in the water-injected and the osnpf7.9-injected, and no significance was observed between the latter two (Supplemental Figure S2D). Moreover, subcellular localization assay indicated that both osnpf7.9-1 and osnpf7.9-2 were localized to plasma membrane, comparable to the WT OsNPF7.9 (Supplemental Figure S3). All these data together demonstrate that the OsNPF7.9^T434I and OsNPF7.9^T430M mutant proteins were disrupted in nitrate transport function.

OsNPF7.9 mediates root-to-shoot nitrate transport

Considering that OsNPF7.9 is a low-affinity nitrate transporter and exclusively expressed in xylem parenchyma cells of vascular tissues, we suspected that OsNPF7.9 might mediate long-distance nitrate transport in rice. When supplied with 0.25 mM ${\text{NH}}_4^{15}$NO₃, the ratio of shoot/root ¹⁵N concentration was comparable in mutants and WT, with ratios of 0.18, 0.18, and 0.19, for WT, osnpf7.9-1 and osnrt1 mutant plants, respectively (Figure 5A). However, when treated with 5-mM ${\text{NH}}_4^{15}$NO₃, the ratio in the mutants was much lower (0.20 in osnpf7.9-1 or 0.23 in osnpf7.9-2) than in WT 0.32 in the WT (Figure 5B), reflecting increased and decreased concentrations of nitrate in the mutant roots and shoots, respectively. To determine if the altered nitrate distribution in the mutants was caused by altered nitrate uptake activity, we compared nitrate uptake between WT and the mutants and found that despite the high nitrate concentrations used, no difference was observed (Figure 5C). Moreover, nitrate concentration in xylem sap collected from osnpf7.9-1 and osnpf7.9-2 was substantially decreased compared to the WT (Figure 5D). These findings suggest that OsNPF7.9 mediates nitrate loading into xylem vessels and hence the long-distance transport of nitrate, and is the functional ortholog of Arabidopsis NRT1.5/AtNPF7.3.

Figure 5.

OsNPF7.9 mediates nitrate translocation from root to shoot in rice. A and B, ¹⁵N contents in root and shoot of WT plants and osnpf7.9 mutants. Seedlings were grown for 3  weeks in Yoshida’s solution containing 1-mM NH₄NO₃, and then treated with 0.25-mM ${\text{NH}}_4^{15}$NO₃ (A) or 5-mM ${\text{NH}}_4^{15}$NO₃ (B) for 30 min. Numbers above the dark bars represent the ratio of shoot/root ¹⁵${\text{NO}}_3^{-}$ content. C, Nitrate uptake activity measured in WT and osnpf7.9 mutants treated with 5-mM ${\text{NH}}_4^{15}$NO₃ for 30  min. D, Nitrate concentration in xylem sap. Seedlings were grown hydroponically for 4 weeks, then xylem sap was sampled and analyzed by high performance liquid chromatography (HPLC). Values are mean ± sd, n = 3 in (A–C), and n = 4 in (D). Different letters above the columns indicate significant differences by one-way ANOVA (Analysis of Varianace) with Tukey‘s test, P=value < 0.05. WT represents the WT rice variety ZH11. DW: dry weight.

Functional disruption of OsNPF7.9 decreased plant biomass and grain yield in rice

Previous studies proposed that nitrate allocation to shoots is essential to plant growth (Zhang et al., 2014; Han et al., 2016). To determine if this is true in the aquatic plant rice, we first measured biomass gain in osnpf7.9 mutants and their WT seedlings under different NH₄NO₃ concentrations. When grown under low (0.25 mM) NH₄NO₃, all the plants had similar biomass (Figure 6, A and C), however, when under 2.5 mM NH₄NO₃, significant biomass gain was only observed in the WT (Figure 6, B and C). Similarly, total nitrogen was lower in the mutants when exposed to the higher level of nitrogen (Figure 6D). To determine if these phenotypes could be attributed to the nitrate transport function of OsNPF7.9, we measured the biomass of the osnpf7.9 mutant and the WT plants supplied with nitrate or ammonium nitrogen. The results showed that when external nitrogen level was low, all the plants had similar biomass regardless of the nitrogen source (Figure 6, E and F). In contrast, when supplied with 2.5-mM nitrate, osnpf7.9 mutants gained significantly less biomass in both shoots and roots than the WT, while no difference was observed when nitrate was replaced with same level of ammonium (Figure 6, E and F). We then grew plants with a fixed level of ammonium but with steadily increasing nitrate levels, and the result indicated that once above 1 mM, nitrate significantly increased biomass of the WT shoots, but had no apparent effect on shoots of the mutant (Supplemental Figure S4A). A comparable growth pattern was also observed in root tissue (Supplemental Figure S4B). These results suggest that nitrate allocation to shoots mediated by OsNPF7.9 is essential for promotion of plant growth.

Figure 6.

OsNPF7.9 and nitrate-dependent effects on biomass. A and B, Both WT plants and osnpf7.9 mutants were grown for 3 weeks in the Yoshida’s solution containing (A) 0.25 mM NH₄NO₃ or (B) 2.5-mM NH₄NO₃. Bars = 5 cm. C and D, DW of (C) and total N in (D) the WT and osnpf7.9 mutant seedlings in different concentrations of NH₄NO₃ solution in (A and B). (E) Shoot DW and (F) Root DW when grown in 0.25 or 2.5-mM nitrogen (NO₃ or NH₄) solution for 3 weeks. Values are mean ± sd (n = 8–12). Different letters above the column indicate significant differences by one-way ANOVA with Tukey‘s test, P-value < 0.05.

At the ripening stage, even more obvious phenotypes were observed. As shown in Figure 7A, when grown in paddy fields, osnpf7.9 mutants grew generally weaker and grain yield essentially decreased. When compared with the WT, all major yield parameters measured were decreased in the mutants. Specifically, tiller numbers in the mutants decreased by 17%–23% (Supplemental Figure S5A), biomass by 24%–29% (Supplemental Figure S5B), grain yield by 26%–49% (Supplemental Figure S5C), and NUE by 26%–54% (Supplemental Figure S5D). Subtle but significant differences were also observed for other yield-related agronomic traits between the mutants and the WT (Supplemental Figure S6). These results indicate that the osnpf7.9 mutations affected a set of parameters that influence grain yield, and we speculated that the drop in NUE might represent the primary contribution. To test this postulation, we further measured biomass, grain yield, and tiller number of plants grown in growth pools supplemented with high and low levels of nitrogen. The results indicated that when under low nitrogen (LN) supply, no significant difference was observed between the mutants and the WT; however, under high nitrogen (HN) supply significant increases in biomass, tiller number, and grain yield were detected in the WT rice, while the enhancing effect was much weaker in the mutant (Figure 7, B–D), supporting our hypothesis that OsNPF7.9 mediates NUE to modulate yield-related agronomic traits and hence grain yield.

Figure 7.

Major agronomic traits regulated by OsNPF7.9. Rice plants grown in paddy field (A) or growth pool (B–D). Biomass and grain yield of osnpf7.9 decreased compared with the WT in field. A, Phenotype of WT and osnpf7.9 mutants at ripening stage. Bar = 20 cm. B, plant biomass, (C) grain yield, and (D) number of tilling of WT and osnpf7.9 mutants. Values are mean ± sd, n = 30 plants in (B–D). Different letters above the column indicate significant differences by one-way ANOVA with Tukey‘s test, P-value < 0.05. WT, ZH11. HN supply (300 kg/ha). LN (0 kg/ha).

Nitrogen assimilation affected in osnpf7.9 mutants

Given nitrate uptake ability was not affected despite decreased NUE (Figure 5C;  Supplemental Figure S5D), we then determined the expression of key genes involved in nitrogen assimilation. The results showed that in the mutant the nitrate reductase genes OsNIA1 and OsNIA2 were substantially decreased compared with the WT (Figure 8, A and B), while only OsNIR2 among the two nitrite reductase genes OsNIR1 and OsNIR2 was significantly decreased (Figure 8, C and D). The glutamine synthase genes OsGS1.1 and OsGS1.2 decreased only in osnpf7.9-2 mutant (Figure 8, E and F). The expression of glutamate synthase gene OsGOGAT1 was unchanged in the mutant, while OsGOGAT2 expression decreased in both mutants (Figure 8, G and H). These results indicated that nitrate assimilation was decreased by the osnpf7.9 mutation.

Figure 8.

Expression of key genes in nitrate assimilation pathway affected in osnpf7.9 mutants. RNA extracted from roots of 2-week-old seedlings were used to determine expression levels. Data were normalized to those of OsActin1.Values are mean ± sd, n = 6 plants. Asterisk above the column indicates significant differences by Student’s t test, ^*P < 0.05, ^**P < 0.01.

Given nitrogen metabolism and photosynthesis are highly associated (Bloom et al., 2010; Nunes-Nesi et al., 2010), we then determined if photosynthesis was affected in mutant seedlings. Surprisingly, no matter what source or level of nitrogen was applied, all the key photosynthetic parameters measured, including net photosynthetic rate (Pn), stomatal conductance (Gs), intercellular CO₂ concentration (Ci), and transpiration rate (Tr), were not altered in the mutants relative to WT (Supplemental Figure S7). Similar results were observed in paddy field-grown plants (Supplemental Figure S8), where the light condition is more optimal for rice growth and agronomic phenotypes are more apparent.

OsNPF7.9 conditions stress responses in rice

Arabidopsis NRT1.5/AtNPF7.3 and NRT1.8/AtNPF7.2 were identified as key regulators involved in SINAR (Li et al., 2010; Chen et al., 2012; Zhang et al., 2014), and, therefore, mediate the tradeoff between plant growth and stress tolerance (Zhang et al., 2014). We then tested whether OsNPF7.9, like AtNRT1.5/AtNPF7.3 (Li et al., 2010; Zhang et al., 2014), is also inhibited by stress treatments. As shown in Supplemental Figure S9A, both Cd²⁺ and Na⁺ stresses significantly decreased OsNPF7.9 expression. Moreover, JA treatments also strongly inhibited OsNPF7.9 expression (Supplemental Figure S9B), but ACC had no effect. When exposed to 15 μM Cd²⁺ or 10 mM Na⁺, it appeared that osnpf7.9-1 and osnpf7.9-2 grew better than the WT as determined by shoot elongation (Supplemental Figure S9, C and D). Under control conditions, shoot elongation was slower in the mutants than the WT at Day 6 (Supplemental Figure S9C) and was comparable at Day 10 (Supplemental Figure S9D), indicating that functional disruption in OsNPF7.9 led to a growth advantage in rice under stress conditions. Taking into account that more nitrate was retained in osnpf7.9 roots (Figure 5, B and D), these data suggest that nitrate allocation to roots contributes to promoting rice growth under different stresses, fitting the SINAR model proposed in Arabidopsis (Zhang et al., 2014).

OsNPF7.9 affects long-distance K transport

Functional disruption of Arabidopsis NRT1.5/AtNPF7.3 decreased K accumulation in shoots (Lin et al., 2008), and a recent study proposed that NRT1.5/AtNPF7.3 transports K (Li et al., 2017). To test if OsNPF7.9 also mediates K transport, we first analyzed OsNPF7.9 expression in response to low K. It appeared that short term K starvation significantly decreased OsNPF7.9 expression (Figure 9A). In osnpf7.9 mutants, less K accumulated in mutant shoots than the WT when supplied with 1 mM NH₄NO₃ (Figure 9B), but not with 0.25-mM NH₄NO₃ (Figure 9C), indicating that K translocation to shoots decreased in the mutant plants. However, when ectopically expressed in oocyte cells, it seemed that OsNPF7.9 did not significantly change K efflux activity (Figure 9D), suggesting that the effects of OsNPF7.9 on long distance K transport are probably indirect, in contrast to what was observed in the Arabidopsis mutant nrt1.5/atnpf7.3 (Li et al., 2017). Furthermore, when exposed to nitrate starvation, earlier leaf chlorosis was not observed in the rice mutant (Supplemental Figure S10), while in Arabidopsis, this phenotype became very obvious when exposed to nitrogen starvation (Drechsler et al., 2015; Meng et al., 2016; Li et al., 2017). These observations suggest that OsNPF7.9 does influence K translocation in rice, but that this may occur through a different mechanism than reported for NRT1.5/AtNPF7.3 in Arabidopsis.

Figure 9.

OsNPF7.9 and K transport. A Two-week-old seedlings grown in Yoshida solution were exposed to K starvation for 6 h (LK). Relative expression of OsNPF7.9 under K starvation, and the data were normalized to those of OsActin1. B and C, K contents in shoots or roots of WT and mutant plants supplied with 1-mM NH₄NO₃ (B) or 0.25-mM NH₄NO₃ (C). D, K efflux activity was measured in oocytes injected with water (H₂O) or OsNPF7.9 cRNA. Values are mean ± sd, n = 5–6. Different letters above the column indicate significant differences by one-way ANOVA with Tukey‘s test, P-value < 0.05. WT rice, ZH11. CK represents control condition (without K deficiency treatment).

OsNPF7.9 allelic divergence between indica and japonica alters nitrate transport

Comparative analysis of OsNPF7.9 CDS identified a single-nucleotide polymorphism (SNP) at site 1298, which is predominantly “A” in japonica rice but “G” in IND and leads to an H433R substitution (Figure 10A, in red). We further examined the allele frequency of this SNP (located at Chr2: 28,313,521; MSU version 7.0) in different rice subgroups based on a collection of 2,452 accessions that include 169 O. rufipogon accessions (WLD), 79 Aus varieties (AUS), 888 indica varieties (IND), 22 basmati varieties (BAS), 135 tropical japonica varieties (TRJ), and 1,159 temperate japonica varieties (TEJ) (Supplemental Dataset S1). The results showed that in WLD, the frequency of the “G” allele was 51.7% while that of the “A” allele was 48.3%, indicating an even distribution in wild rice (Figure 10B). However, the allele frequency diverged largely between indica and japonica subspecies. For indica subspecies, the frequency of the “G” allele was 62.7% in AUS but 93.2% in IND. On the contrary, for the japonica subspecies, the frequency of the alternative allele “A” was almost fixed, reaching 92.9%, 97.6%, 99.9% in BAS, TRJ, and TEJ, respectively (Figure 10B).

Figure 10.

A conserved SNP in OsNPF7.9 alters nitrate transport between indica and japonica rice. A, The SNP (A/G) at 1,298 bp in OsNPF7.9 CDS between indica and japonica rice. B, Allele frequency of the SNP in different rice subgroups, AUS: Aus rice; BAS: basmati rice; TRJ: tropical japonica rice; TEJ: temperate japonica rice. C, Nitrate uptake activity of oocytes injected with cRNA of SNP-G (OsNPF7.9-ind) or SNP-A (OsNPF7.9-jap). Oocytes were incubated with 10-mM K¹⁵NO₃ buffer at pH 5.5 for 3 h. CHL1/AtNPF6.3 and water were used as positive and negative control respectively. D and E, Twenty randomly picked indica and japonica rice varieties (D), or transgenic plants harboring SNP-G or SNP-A (E), were grown in Yoshida solution for 2 weeks, then treated with 5-mM ${\text{NH}}_4^{15}$NO₃ for 30 min before determination of the ratio of shoot/root ¹⁵N contents. Values are mean ± sd, n =  6 in (C and E) and 10 in (D). Different letters above the column indicate significant differences by one-way ANOVA with Tukey‘s test, asterisk indicates significant differences by Student’s t test, ^**P < 0.01.

We then determined the nitrate uptake activity of both alleles in Xenopus oocytes and found that the indica type (SNP-G) showed a significantly stronger nitrate uptake ability than the japonica type SNP-A (Figure 10C;  Supplemental Figure S11A), which further raised a question of whether the SNP (G/A) leads to differential long-distance nitrate transport between japonica and indica varieties. To answer this question, we used 20 randomly chosen indica and japonica rice varieties, and more nitrate was transported to IND shoots compared to those japonica (Figure 10D), as would be expected. We further tested this hypothesis using transgenic plants harboring SNP-G or SNP-A of OsNPF7.9 driven by the ubiquitin promoter, where gene expression levels were considerably higher than the WT control ZH11 but comparable between transgenic lines (Supplemental Figure S11B). Consistently, more nitrate was translocated to shoots of transgenic plants than ZH11, and transgenic lines harboring the indica allele SNP-G performed better than the japonica SNP-A (Figure 10E), suggesting that the A/G SNP in OsNPF7.9 contribute to differential nitrate allocation between indica and japonica rice. Interestingly, the transgenic plants generally grew better than the WT ZH11 under control condition, but decreased stress tolerance was not as obvious as the SINAR model would expect (Supplemental Figure S11, C and D), and no further difference was observed between transgenic SNP-G and SNP-A.

Discussion

Nitrate allocation has been proposed to play an important role in mediating terrestrial plant adaptation to changing environments (Smirnoff and Stewart, 1985), and Arabidopsis NRT1.8/AtNPF7.2 and NRT1.5/AtNPF7.3 were identified as two key players in this process (Li et al., 2010; Chen et al., 2012). In this study, we demonstrate that OsNPF7.9 is an ortholog of Arabidopsis NRT1.5/AtNPF7.3, and functions to mediate nitrate allocation and divergent NUE between rice subspecies.

OsNPF7.9 mediates nitrate allocation and SINAR in rice

It has been demonstrated that nitrate allocation between roots and shoots plays an important role in balancing plant growth and stress tolerance in Arabidopsis (Li et al., 2010; Chen et al., 2012), and this process is dynamically mediated by ethylene and jasmonic acid signaling pathways through the coordinated regulation of two nitrate transporter genes NRT1.8/AtNPF7.2 and NRT1.5/AtNPF7.3 (Zhang et al., 2014, 2018). Although ammonium is the main nitrogen source for rice due to its water-logged habitat, nitrate long-distance transport to shoots has been reported (Tang et al., 2012; Li et al., 2015; Xia et al., 2015), thus SINAR may also occur in rice. OsNRT2.3a was a promising candidate for this role because it expresses exclusively in xylem parenchyma cells and nitrate allocation to shoots was substantially reduced in its loss-of-function mutant (Tang et al., 2012). However, it appeared that OsNRT2.3a functions only as a HATS uptake transporter, hence it can hardly mediate direct nitrate loading into xylem vessels. OsNPF2.2 and OsNPF2.4 also affect nitrate long-distance transport, but they both express in several different tissues and led to pleotropic effects when disrupted (Li et al., 2015; Xia et al., 2015). From these aspects, these genes are not likely functional orthologs of either NRT1.8/AtNPF7.2 or NRT1.5/AtNPF7.3.

However, this study demonstrated that OsNPF7.9 is found exclusively in the plasma membrane of xylem parenchyma cells of the root vasculature, and that it shows both nitrate uptake and efflux activities (Figures 2–4). Functional disruption of OsNPF7.9 decreased nitrate in xylem sap and nitrate translocation from roots to shoots (Figure 5). A recent study found that overexpression of OsNPF7.9 enhanced nitrate accumulation in shoots (Feng et al., 2017). Moreover, OsNPF7.9 was downregulated by Cd/Na stress (Supplemental Figure S9A), which would help to retain nitrate in roots when under stress conditions. Accordingly, stress tolerance was enhanced in the osnpf7.9 mutants (Supplemental Figure S9, C and D). All these observations suggest that SINAR does occur in rice and OsNPF7.9 is an important mediator in this process. Further support came from the enhanced growth of transgenic plant expressing indica or japonica alleles of OsNPF7.9 (Supplemental Figure S11, B and C), although decreased stress tolerance was not as significant as expected in the transgenic lines, which might be attributable to the fact that nitrate translocation was high enough to mask allelic effect in transgenic plants. It is also possible that overexpression of transporter might have resulted in more effects than nitrate transport.

Note although OsNPF7.9 and NRT1.5/AtNPF7.3 are highly functionally orthologous, they do show some functional divergence between rice and Arabidopsis. Arabidopsis NRT1.5/AtNPF7.3 responds to both ethylene and jasmonic acid treatments (Zhang et al., 2014), but OsNPF7.9 only responds to jasmonic acid (Supplemental Figure S9B). In both osnpf7.9 and nrt1.5/atnpf7.3 mutants, K transport to shoots was affected (Figure 9B;  Lin et al., 2008; Drechsler et al., 2015; Li et al., 2017). However, nitrate starvation induced earlier leaf senescence in Arabidopsis mutant and this phenotype could be rescued by K application (Drechsler et al., 2015; Meng et al., 2016), but rice mutant did not show any premature chlorosis under low nitrate (Supplemental Figure S10). Moreover, OsNPF7.9 does not appear to transport K directly (Figure 9D), in contrast to NRT1.5/AtNPF7.3 in Arabidopsis (Li et al., 2017). These observations indicate that OsNPF7.9 and NRT1.5/AtNPF7.3 do diverge functionally, thus SINAR in rice might be regulated in a way likely different from that in Arabidopsis, at least not mediated by ethylene signaling pathway.

OsNPF7.9 mediates biomass gain and increased grain yield in rice

Nitrogen is one of the macronutrients that is essential to plant growth. This study demonstrated that the osnpf7.9 mutant seedlings had less biomass and total nitrogen than the WT, and the differential biomass gaining between the mutant and WT plants was exclusively correlated to the nitrogen form and level (Figure 6; Supplemental Figure S4). Consistently, the WT produced more grain and biomass than osnpf7.9 mutants only when high levels of nitrogen was applied (Figure 7, A–C), demonstrating that the OsNPF7.9-mediated nitrate transport plays a pivotal role in rice growth and grain yield, possibly by affecting nitrogen metabolism (Figure 8) or other secondary effects yet to be identified. Furthermore, our data indicate that OsNPF7.9 mediates grain yield by modulating multiple agronomic traits, such as tiller number, panicle branches, and grain number of each panicle, but not grain filling (Figure 7D;  Supplemental Figures S5 and S6). It is worth mentioning that other genes involved in long-distance nitrate transport to shoots also and significantly affect plant growth in rice (Tang et al., 2012; Li et al., 2015), consistent with the hypothesis of higher energy efficiency and evolutionary advantage for nitrate assimilation in aerial parts (Smirnoff and Stewart, 1985; Zhang et al., 2014). Thus, allocating more nitrates to shoots might be an efficient way to improve crop growth with better nitrogen economy. This strategy seems to have been realized in the terrestrial crop Brassica napus, as one cultivar with high NUE was recently proven to be attributable to enhanced nitrate allocation to shoots that is driven by root vacuolar nitrate sequestration (Han et al., 2016).

Previous studies proposed that nitrate allocation between roots and shoots dynamically modulates plant growth and environmental adaptation, since disrupted nitrate allocation to roots substantially decreased stress tolerance (Li et al., 2010), and when nitrate allocation to roots was enhanced, plants tend to show higher stress tolerance but worse plant growth (Zhang et al., 2014). Studies from tobacco (Nicotiana tabacum) led to the proposal that nitrate allocation might serve as a signal to coordinately reprogram nitrogen and carbon metabolism to regulate plant growth (Scheible et al., 1997a, 1997b). Interestingly, our data showed that several key genes in the nitrogen assimilation pathway were downregulated in osnpf7.9 mutants (Figure 8), indicating that nitrogen assimilation was decreased when more nitrate accumulated in roots, but photosynthesis was not affected although the mutant plants grew much worse than the WT (Supplemental Figures S7 and S8; Figures 6 and 7). These observations support the previous hypothesis that nitrate allocation might function to partition energy within plants in response to environmental signals, thus allowing balance between growth and stress tolerance (Zhang et al., 2014). Indeed, enhanced stress tolerance was observed in osnpf7.9 mutants (Supplemental Figure S9) at the expense of plant growth (Figures 6 and 7), which indicates that SINAR might represent a common mechanism regulating environmental adaptation for plants across terrestrial and aquatic species that are either nitriphilic or ammoniphilic.

Nitrate allocation mediated by OsNPF7.9 contributes to the NUE divergence between indica and japonica

Indica and japonica represent two major distinctive rice subspecies. Japonica is widely cultivated in East Asia and has lower NUE than IND (Koutroubas and Ntanos, 2003). A recent study showed that nitrate uptake mediated by OsNRT1.1B/OsNPF6.5 might contribute to the divergent NUE between indica and japonica (Hu et al., 2015). Our study showed that long distance nitrate transport to shoots might also contribute to the overall higher NUE in IND, as the SNP detected in OsNPF7.9 is highly divergent between indica and japonica subspecies (Figure 10, A and B), and the more efficient allele SNP-G was present in >93% indica, in contrast, the japonica rice almost exclusively contained the less efficient SNP-A allele (Figure 10, B and C). Consistent with this hypothesis, more nitrate was allocated to shoots (Figure 10D) of transgenic lines expressing the indica SNP-G allele. Moreover, generally higher nitrate translocation occurs in indica than in japonica (Figure 10E). In contrast to the highly divergent allele frequency between indica and japonica subspecies, the two alleles distribute evenly in WLD (Figure 10B), implying that different alleles of the OsNPF7.9 were favorably selected and fixed during improvement of different rice groups. Interestingly, the natural single amino acid (aa) change between indica and japonica is adjacent to the loss-of-function substitution of osnpf7.9-1 and just 3 aa away from that of osnpf7.9-2 (Supplemental Figure S2A; Figure 10A). Moreover, all 3 aa reside within the same cytoplasmic loop region between TM domains 8 and 9 (Figure 1B), indicating that this region might subject to protein level regulation in response to environmental cues.

In conclusion, we revealed that SINAR does occur in rice and OsNPF7.9 plays an important role in this process, though the regulation of SINAR might differ between rice and Arabidopsis. We further propose that the OsNPF7.9-mediated nitrate translocation to shoots contributes essentially to NUE divergence between indica and japonica subspecies. Our findings suggest that the region containing the indica–japonica divergent SNP might be a key marker for improving NUE in most japonica varieties.

Materials and methods

Plant materials and growth conditions

The japonica rice (O. sativa) Zhonghua 11 (ZH11) was mutagenized by ethyl methane sulfonate and the derived population was screened by TILLING to identify the osnpf7.9-1 and osnpf7.9-2 mutants as described (Liu et al., 2018). Rice seedlings were grown in a phytotron with a 14-h light (30°C)/10-h dark (25°C) photoperiod with ∼300-μM m⁻²s⁻² photon density and 60% humidity. Yoshida solution (1-mM NH₄NO₃, 0.6-mM NaH₂PO₄, 0.3-mM K₂SO₄, 0.3-mM CaCl₂, 0.6-mM MgSO₄, 45.1-µM EDTA-Fe, 4.87-µM H₃BO₃, 0.9-µM MnSO₄·5H₂O, 0.03-µM CuSO₄, 0.07-µM ZnSO₄, 0.01-µM Na₂MnO₄, pH =  5.8–6.2) was used for hydroponic culture and refreshed every 2 d. Seeds were germinated under 28°C and those uniformly germinated were then transferred to the hydroponic box containing Yoshida solution to grow for the indicated time. Alternatively, grain husks were stripped and then surface sterilized by 20% (v/v) NaClO before gemination on half-strength Murashige and Skoog medium containing 0.8% (W/V) agar in a growth chamber with 14-h light (30°C)/10-h dark (25°C) for the indicated time. Rice seedlings at the indicated age were treated by nitrate, Cd²⁺ or Na⁺ at indicated concentrations for the indicated time, then the seedlings were subjected to RNA extraction or phenotypic assays.

Phylogenetic tree construction and sequence alignment

All amino acid sequences of NPFs were downloaded from National Center for Biotechnology Information (NCBI) and were used to construct the phylogenetic tree, which was further modified on the iTOL website (http://itol.embl.de/). Sequence alignment was performed using GeneDoc.

Nitrate uptake and translocation activity assay using ¹⁵NO₃⁻

Rice seedlings grown in Yoshida solution for 3 weeks were washed with 0.1-mM CaSO₄ for 1 min before transferring to fresh Yoshida solution containing 5-mM ${\text{NH}}_4^{15}$NO₃ with 99% atom excess of ¹⁵N, and washed again with 0.1-mM CaSO₄ for another 1 min after 30 min incubation with the 5-mM ${\text{NH}}_4^{15}$NO₃ solution, which was further washed with ddH₂O five times before the shoot and root were sampled and dried at 80°C for determination of ¹⁵N contents. ¹⁵N was measured using Elemental Analysis-Stable Isotope Ratio Mass Spectrometer (Vario EL III/Isoprime, Elementar Germany) and the data indicating nitrate uptake and translocation were calculated as described (Lin et al., 2008). Alternatively, rice seedlings were grown in modified Yoshida solution containing 2.5-mM NH₄NO₃ for 4 weeks, then shoots were removed at about 1 cm above the root–shoot junction and xylem sap was collected for 2 h. Nitrate content in xylem sap was determined as described previously (Chiu et al., 2004; Chen et al., 2012) using HPLC (Agilent 1200 series).

GUS staining, in situ hybridization, and subcellular localization

A 2,649-bp fragment of OsNPF7.9 promoter was polymerase chain reaction (PCR) amplified and ligated into the modified pCambia1300 to generate the construct pOsNPF7.9::GUS, which was then transformed to ZH11 by Agrobacterium tumefaciens-mediated transformation. Positive transgenic lines were subjected to GUS staining and tissue sectioning as described previously (Li et al., 2010). GUS imaging was performed using a microscope (Leica DM6000B), and the Leica DFC495 camera was used for capturing images.

About 0.5- to 1-cm root segments without root tips were sampled from 5-d-old seedlings grown in hydroponic culture. These segments and corresponding RNA probes were prepared for in situ hybridization as described (Long and Barton, 1998) with minor modification: A 464-bp OsNPF7.9 gene-specific fragment spanning the coding region and 3′-untranslated region (UTR) was amplified by PCR and cloned into pGEM-T Easy vector (Promega, Madison, WI, USA). Sense and antisense probes were synthesized in vitro using T7 and SP6 RNA polymerase according to the manufacturer’s instructions.

For subcellular localization, coding sequences (CDS) of OsNPF7.9 and the mutant allele osnpf7.9 were amplified by PCR to generate constructs p35S::OsNPF7.9-eYFP/PA7, p35S::osnpf7.9-1-eYFP/PA7 and p35S::osnpf7.9-2-eYFP/PA7. Rice protoplasts were isolated by cutting stem and sheaths of 50 rice seedlings into approximately 0.5-mm strips using sharp razor blades. The constructs were then transiently expressed in rice protoplasts (Zhang et al., 2011). The construct p35S::OsNPF7.9-eYFP/PA7 was transiently expressed in onion epidermal cells using a particle gun system (PDS-1000/He; Bio-Rad, Hercules, CA, USA). The transformed rice protoplasts and bombarded cells were held in the dark for 12 h at 22 °C. After overnight cultivation, YFP fluorescence was excited at 514 nm line ray of the argon laser and observed between 520 and 580 nm, set detector gain at 1 using a confocal laser scanning microscope (Olympus FV1000, Tokyo, Japan).

RNA extraction, cDNA preparation, and RT-qPCR

Total RNA was extracted from rice plants grown under the indicated conditions using TRIzol reagent (Ambion, Austin, TX, USA). First-strand cDNA was synthesized using PrimeScript RT reagent Kit with gDNA Eraser (Takara, Shiga, Japan) following the manufacturer’s instructions. RT-qPCR was performed as described with minor modifications (Li et al., 2010). Expression levels were normalized to those of the OsActin 1 control. RT-qPCR primer sequences are listed in Supplemental Table S1.

Functional analyses of OsNPF7.9 in Xenopus laevis oocytes

The 1,833-bp CDS of OsNPF7.9 and osnpf7.9 were amplified by PCR and cloned into the oocyte expression vector pOO2 between restriction sites EcoRI and SacI. cRNA was synthesized in vitro using MMESSAGE MMACHINE SP6 KIT 25RXNS (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instruction. Oocytes were isolated and injected with 50 nL (1 ng/nL) cRNA or water, and then incubated for 44–48 h at 16°C in ND96 (96-mM NaCl, 2-mM KCl, 1-mM MgCl₂, 1.8-mM CaCl₂, 5-mM HEPES, 10-mg/L streptomycin sulfate, and 50-mg/L gentamycin sulfate, pH 7.4) containing 60 mg/L penicillin and 100-mg/L streptomycin. Current recording and ¹⁵NO₃⁻ uptake/efflux analyses were performed as described with minor modifications (Lin et al., 2008).

Field trials

Field trials were performed in the years 2016, 2017, and 2018 at the experimental station of the Institute of Plant Physiology and Ecology (Shanghai), where the rice mutants and their WT parent ZH11 were grown and subjected to measurements of important agronomic traits. Rice was grown in a paddy field supplied with urea nitrogen at 300 kg/ha. Other plants were grown in cement growth pools next to the paddy field in the open air. The pools were filled with infertile paddy soil and a total of 300 kg or 0 kg/ha of urea, 75 kg/ha of Ca(H₂PO₄)₂ and 75 kg/ha of K₂SO₄ was supplied, and the N was supplied over three treatments: the first N treatment was given 2 weeks after transplanting and 20% of the total N fertilizer was used, the second treatment of 40% of the total N was made at the tillering stage, and the remaining 40% was applied 1 week later. The spacing distance between rice plants was 20 cm.

Agronomic traits including tiller numbers, biomass, grain yield, NUE, and yield-related characters were measured on a single plant basis. All filled grains were collected and dried at 40°C before measuring the grain yield. Total grain numbers include all the filled and unfilled grains of a single plant, and setting percentage was calculated using the formula: filled grains/total grain numbers × 100%. Grain width and 1,000-grain weight were measured by an SC-G automatic seed scorer. NUE was defined as grain yield per fertilizer N applied in the field.

Determination of photosynthetic characters and light response curve

Three-week-old rice seedlings grown in Yoshida solution or rice leaves from field-grown plants (in the cement pool) during mature period were used. Photosynthetic characteristics or light response curve were measured according to manufacturer’s instructions, using the portable photosynthetic apparatus LI-6400XT (LI-COR). Measurement was performed between 9:00 a.m. and 15:00 p.m. Light intensity is 1,200 µm/m²/s for seedling stage and 1,500 µm/m²/s for maturation stage.

Accession numbers

Sequence data from this article can be found in the Arabidopsis TAIR database (https://www.arabidopsis.org), Rice Genome Annotation Project (https://rice.plantbiology.msu.edu/) or GenBank/EMBL data libraries under accession numbers (Supplemental Table S2).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. OsNPF7.9 expression in different tissues and at different ages.

Supplemental Figure S2. Identification of the loss-of-function mutant osnpf7.9.

Supplemental Figure S3. Subcellular localization of OsNPF7.9 and osnpf7.9 in rice mesophyll protoplasts.

Supplemental Figure S4. Plant growth with increasing nitrate supply.

Supplemental Figure S5. Major agronomic traits affected in osnpf7.9.

Supplemental Figure S6. Field trials for agronomic traits of WT and npf7.9 mutants.

Supplemental Figure S7. Measurement of photosynthetic parameters in WT and osnpf7.9 seedlings.

Supplemental Figure S8. Measurement of photosynthetic parameters in field-grown WT and osnpf7.9.

Supplemental Figure S9. OsNPF7.9 expression level and phenotype of mutants involvement in SINAR.

Supplemental Figure S10. osnpf7.9 and WT growth under nitrate starvation condition.

Supplemental Figure S11. Characterization of SNP-A and SNP-G in oocyte cells and transgenic plants.

Supplemental Table S1. Primers sequences used in this study.

Supplemental Table S2. Accession numbers.

Supplemental Dataset S1. Rice germplasms used in this study.

 

These authors contributed equally (Y.G. and D.L.).

J.G. conceived and supervised the project. J.G., Y.G., and D.L. designed the experiments. Y.G. and D.L. performed most of the experiments. Z.L., J.Q., Y.H., and Z.F. performed some of the experiments. J.G., X. H., D.L., and Y. G. wrote the manuscript.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is: Ji-Ming Gong (jmgong@cemps.ac.cn).