The nitrate transporter NRT2.1 directly antagonizes PIN7-mediated auxin transport for root growth adaptation

Significance

Modulation of primary root growth is a foraging strategy adapting to NO₃⁻ fluctuation in soil. However, how that comes about remains unclear. Here, we identify a unique mechanism involving the interaction of two transporters specific to different substrates (NO₃⁻ or auxin). Our study demonstrates that the high-affinity NO₃⁻ transporter NRT2.1 directly associates with the auxin transporter PIN7, and antagonizes its activity depending on NO₃⁻ levels, thus regulating primary root elongation. Collectively, this work provides insight into the mechanisms regulating root growth adaptation in response to NO₃⁻ availability.

Abstract

As a crucial nitrogen source, nitrate (NO₃⁻) is a key nutrient for plants. Accordingly, root systems adapt to maximize NO₃⁻ availability, a developmental regulation also involving the phytohormone auxin. Nonetheless, the molecular mechanisms underlying this regulation remain poorly understood. Here, we identify low-nitrate-resistant mutant (lonr) in Arabidopsis (Arabidopsis thaliana), whose root growth fails to adapt to low-NO₃⁻ conditions. lonr2 is defective in the high-affinity NO₃⁻ transporter NRT2.1. lonr2 (nrt2.1) mutants exhibit defects in polar auxin transport, and their low-NO₃⁻-induced root phenotype depends on the PIN7 auxin exporter activity. NRT2.1 directly associates with PIN7 and antagonizes PIN7-mediated auxin efflux depending on NO₃⁻ levels. These results reveal a mechanism by which NRT2.1 in response to NO₃⁻ limitation directly regulates auxin transport activity and, thus, root growth. This adaptive mechanism contributes to the root developmental plasticity to help plants cope with changes in NO₃⁻ availability.

Plants are sessile and have evolved elaborated mechanisms to cope with unfavorable environmental conditions. Roots are vital organs responsible for the acquisition of resource from the soil and are thus responsive to a wide range of external stimuli. Changes in root architecture and development contribute to plant adaptation to environmental stresses. However, only a few mechanisms explaining root plasticity have been proposed.

Nitrogen is a vital macronutrient for plant growth. As the main form of inorganic nitrogen, nitrate (NO₃⁻) is not only the preferred nitrogen source but also acts as a signal molecule regulating many plant physiological processes. The NO₃⁻ signaling cascade, which comprises a transceptor (nitrate transporter1.1, NRT1.1), protein kinases (e.g., Ca²⁺-sensor protein kinases, CPKs), and transcription factors (e.g., NIN-like proteins, NLPs), contributes to the control of root development in response to NO₃⁻ availability (1–4). Furthermore, plant hormones (e.g., auxin; brassinosteroids) and signaling peptides (e.g., CLAVATA3/ESR-related, CLE; C-terminally encoded peptide, CEP) also mediate NO₃⁻-regulated root responses (1, 5–9).

Given the universal role of the phytohormone auxin in determining root architecture, it is not surprising that there are strong evidences supporting connections between NO₃⁻ and auxin. Indeed, auxin transport, signaling, and biosynthesis all participate in NO₃⁻-mediated root development. For instance, the NO₃⁻ transceptor NRT1.1 not only senses and transports NO₃⁻ in both high- and low-NO₃⁻ affinity states (switching between the two states by phosphorylation and dephosphorylation), but also facilitates auxin transport and thus inhibits lateral root development under low external NO₃⁻ concentrations (1, 10–12). Low NO₃⁻ increases auxin transport from shoots to roots and regulates plant root growth (13). NO₃⁻ availability transcriptionally regulates the expression of several auxin efflux carrier genes that are required for root development (14). Low NO₃⁻ represses auxin accumulation potentially by raising NRT1.1 protein abundance and ultimately stimulates distal stem cell differentiation in root tips (9). Dephosphorylation of the auxin efflux carrier PIN-FORMED2 (PIN2) is related to NO₃⁻-regulated root cell division and elongation (15). Expression of auxin response factor8 (ARF8) is responsive to the NO₃⁻ metabolite glutamine and regulates lateral root outgrowth (16). The auxin receptor auxin-signaling F-box protein3 (AFB3) specifically coordinates primary and lateral root growth in response to external and internal NO₃⁻ availability (17, 18). The MADS-box transcription factor AGAMOUS-like21 (AGL21) promotes lateral root initiation and elongation by enhancing local auxin biosynthesis when NO₃⁻ is limited (19). The auxin biosynthesis gene tryptophan amino-transferase related2 (TAR2) promotes lateral root formation under low-NO₃⁻ conditions (20). In summary, NO₃⁻ and auxin tightly interact to modulate root architecture; however, the underlying molecular mechanisms remain to be investigated.

The large NRT transporter family in Arabidopsis (Arabidopsis thaliana) comprises 53 NRT1 and seven NRT2 members (2). NRT1 family members appear to be critical for low-affinity NO₃⁻ transport (21), with the exception of NRT1.1, which not only acts as a dual-affinity NO₃⁻ transporter and NO₃⁻ sensor, but also functions in auxin translocation (1, 10–12). NRT2-type proteins have been characterized as specific high-affinity NO₃⁻ transporters, and to date, no other substrate has been identified (21, 22). The transporter activity of most NRT2 members depends on the partner protein nitrate assimilation–related protein (NAR2.1, also named NRT3.1), which is required for the plasma membrane targeting and protein stabilization of NRT2 members (21, 23). Among others, NRT2.1 serves as the predominant high-affinity NO₃⁻ transporter, as NO₃⁻ uptake capacity dropped by about 56% in nrt2.1 knockouts (24). NRT2.1 transcription is differentially regulated as a function of external NO₃⁻ levels. NO₃⁻ at low concentrations effectively induces NRT2.1 expression, but higher concentrations repress it (25). Previous studies have reported a role for NRT2.1 in the response of lateral root initiation to low NO₃⁻ (26, 27). While NRT1.1 is known to use both NO₃⁻ and auxin as possible substrates to regulate lateral root development (1, 10), the mechanism by which NRT2.1 regulates root architecture is still unknown.

In this study, we used forward genetics to identify Arabidopsis low-nitrate-resistant mutant2 (lonr2) whose primary root elongation showed reduced sensitivity to low NO₃⁻. We discovered that lonr2 mutants were defective in NRT2.1. lonr2 (nrt2.1) mutants displayed impaired polar auxin transport and their low-NO₃⁻-induced root phenotype required auxin transport activity of PIN7. We established that NRT2.1 directly associated with PIN7 and antagonized PIN7-mediated auxin efflux in Xenopus (Xenopus laevis) oocytes and tobacco (Nicotiana tabacum) Bright Yellow-2 (BY-2) cells depending on NO₃⁻ levels. These observations extend the spectrum of biological function of NRT2.1, highlighting an important role in the adaptation of root growth to NO₃⁻ limitation by directly counteracting auxin transport activity.

Results

Identification of lonr2 Mutants.

Modulation of plant root architecture, including primary root growth, is an essential plant adaptive response to dynamically changing NO₃⁻ concentrations in the soil (28, 29). However, the molecular mechanisms underlying this adaptation remain largely unknown.

Different NO₃⁻ concentrations can either inhibit or stimulate primary root elongation (3, 5, 30–33). To further identify NO₃⁻-related regulators of root growth, we performed a forward genetic screen to look for Arabidopsis mutants that were less sensitive to low NO₃⁻ when grown on medium containing 0.05 mM NO₃⁻. We identified three mutant lines, which we named low-nitrate-resistant mutant2 (lonr2-1 through lonr2-3). These lines harbored mutations that largely restored primary root growth under low-NO₃⁻ conditions (Fig. 1 A and B). We did not observe phenotypic differences between the roots of wild-type and lonr2 seedlings when grown on 10 mM NO₃⁻ medium, but lonr2 primary roots were much longer than those of wild-type seedlings when grown on 0.05 mM NO₃⁻ (Fig. 1 A and B). Genetic analysis indicated that the lonr2-1 through lonr2-3 mutations occur in a single gene (SI Appendix, Fig. S1 A–D).

Fig. 1.

lonr2 mutants are less sensitive to low NO₃⁻ in terms of primary root elongation. (A and B) Root phenotypes (A) and primary root length (B) of lonr2 mutants grown under 10 mM NO₃⁻ (full NO₃⁻; FN) or 0.05 mM NO₃⁻ (low NO₃⁻; LN) conditions for 9 to 10 d. (Scale bars, 1 cm.) Data are shown as means ± SE (n ≥ 17 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (C) Confocal images of the root meristematic (white line), elongation (blue line), and differentiation (green line) zones of lonr2 mutants. Each image is composed of several joined photographs of the same root. MZ, meristematic zone; EZ, elongation zone; DZ, differentiation zone. (Scale bars, 100 µm.) (D) Cell division activity in root meristems of lonr2 mutants as visualized by CYCB1;1:DB-GUS. (Scale bars, 50 µm.)

Root length is coordinately determined by the activity of the root stem cell niche and the differentiation of the cell progenies. Along the longitudinal axis of the root tip, cellular activities are separated spatially into the meristematic zone (MZ), elongation zone (EZ), and differentiation zone (DZ) (34, 35). To pinpoint the underlying cause of the altered growth of lonr2 roots, we measured the cell numbers, cell lengths, and zone lengths of wild-type and lonr2 roots grown at different NO₃⁻ concentrations. On medium with 10 mM NO₃⁻, the lengths of the MZ, EZ, and DZ of lonr2 seedlings were similar to those of wild-type seedlings; the seedlings had similar cell numbers and average cell lengths across the three zones (Fig. 1C and SI Appendix, Fig. S1E). Low-NO₃⁻ treatment decreased cell number and/or cell length in each zone to a greater extent in wild-type than in lonr2 seedlings, leading to longer root zones in lonr2 seedlings under low NO₃⁻ (Fig. 1C and SI Appendix, Fig. S1E). We also examined the expression of the mitotic marker CYCB1;1:DB-GUS in wild-type and lonr2 seedlings. We observed a much higher cell division activity in lonr2 root meristems than that in the wild type under low-NO₃⁻ conditions (Fig. 1D), suggesting that loss of LONR2 function promotes mitosis.

We next examined the effects of various concentrations of NO₃⁻ on root growth and determined that lonr2 mutants had longer primary roots than those of wild-type seedlings when exposed to lower concentrations of NO₃⁻ (SI Appendix, Fig. S1F). However, there was no difference in root length between wild-type and lonr2 seedlings at higher NO₃⁻ concentrations (SI Appendix, Fig. S1F). We also investigated the effect of ammonium (NH₄⁺) on the root growth of lonr2 mutants and observed that the mutants exhibited similar sensitivity as wild-type seedlings to all NH₄⁺ concentrations tested (SI Appendix, Fig. S1G). Together, these results indicated that LONR2 is involved in the regulation of root development, specifically under low-NO₃⁻ conditions.

lonr2 Is Defective in the High-Affinity NO₃⁻ Transporter NRT2.1.

Using whole-genome sequencing, we identified the high-affinity NO₃⁻ transporter gene NRT2.1 (At1g08090) as LONR2. One of the alleles introduced a premature stop codon (C to T at nucleotide 1405, R406STOP), and the other two alleles created different amino acid substitutions (G to A at nucleotide 247, A83T; G to A at nucleotide 592, G198S) (Fig. 2A). The T-DNA insertion mutant SALK_141712 (nrt2.1 hereafter) had approximately 10% of the NRT2.1 transcript level seen in the wild type (24) (SI Appendix, Fig. S2A). This mutant showed a phenotype similar to the other lonr2 alleles grown on medium containing 0.05 mM NO₃⁻; the roots of the nrt2.1 mutant were much longer than those of the wild type (Fig. 2B). We next evaluated NO₃⁻ uptake and accumulation in the lonr2 mutants. We observed that lonr2-1 through lonr2-3 seedlings exposed to ¹⁵NO₃⁻ had significantly lower rates of NO₃⁻ influx compared to wild-type seedlings, as did the nrt2.1 T-DNA insertion mutant (SI Appendix, Fig. S2B). Moreover, lonr2-1 through lonr2-3 alleles also had lower NO₃⁻ contents than wild type, resembling the nrt2.1 T-DNA insertion mutant (SI Appendix, Fig. S2C). These results suggest that either the premature stop codon (R406STOP) or the altered amino acids (A83T, G198S) compromise the NRT2.1 function in NO₃⁻ uptake activity and thus reduce NO₃⁻ accumulation. Several previous studies have indicated that NRT2.1 has a regulatory role in the lateral root formation in response to NO₃⁻ limitation (26, 27). We therefore carefully analyzed lateral root initiation and development in lonr2 mutants under our experimental conditions and found that mutations identified in lonr2-1 or lonr2-2 did strongly reduce the density of lateral root primordia and emerged lateral roots under low-NO₃⁻ conditions (SI Appendix, Fig. S2D), which is consistent with the previous observations in other mutant alleles of NRT2.1 (27). Thus, it is tempting to suggest that low-NO₃⁻-repressed root branching phenotype in lonr2 was specifically due to NRT2.1 mutation. The distinct growth behaviors of lateral roots and primary roots (i.e., an increase in root length) of lonr2 mutants are likely caused by the different mechanisms or associated key regulators in response to exogenous cues.

Fig. 2.

LONR2 encodes the NO₃⁻ transporter NRT2.1. (A) Genomic structure of the LONR2/NRT2.1 mutant locus. Light gray box, promoter; dark gray boxes, UTRs; black boxes, exons; lines, introns. The relative positions of the lonr2-1/nrt2.1-5, lonr2-2/nrt2.1-6, and lonr2-3/nrt2.1-7 mutations are indicated. T-DNA insertion for nrt2.1 (SALK_141712) is 236 bp upstream of the start codon. (B and C) Primary root length of the T-DNA insertion mutant nrt2.1 (B) and the lonr2-1 NRT2.1:NRT2.1-GFP 1# complementation line (C). Data are shown as means ± SE (n ≥ 20 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (D and E) Expression pattern of the promoter fusions NRT2.1:GUS (D) and NRT2.1:GFP (E) in root meristematic, elongation, and differentiation zones (from Left to Right). Spatial patterns of GFP fluorescence are shown by two joined photographs of the same root. (Scale bars, 100 µm.)

We attempted to complement the lonr2-1 allele with a C-terminal GFP fusion of NRT2.1 under the control of its endogenous promoter (~1.9 kb). This transgene (NRT2.1:NRT2.1-GFP) only partially rescued the long root phenotype of lonr2-1 under low-NO₃⁻ conditions (Fig. 2C and SI Appendix, Fig. S2E). We speculated that the partial rescue was likely due to impaired NRT2.1 function by the fused GFP. Therefore, we tested the previously reported transgenic line nrt2.1 NRT2.1:NRT2.1 expressing the NRT2.1 genomic fragment without GFP, which was shown to completely rescue the shoot growth–restricted phenotype of the nrt2.1 T-DNA insertion allele (36). Importantly, even under the control of the shorter endogenous promoter (1.2 kb), the NRT2.1:NRT2.1 transgene was able to fully complement the low-NO₃^–-insensitive root growth phenotype of the nrt2.1 mutant (SI Appendix, Fig. S2F).

Monitoring NRT2.1 promoter activity in NRT2.1:GUS transgenic lines suggested that the gene has broad expression in cotyledons, floral organs (SI Appendix, Fig. S2G), and throughout the entire primary roots, including the MZ, EZ, and DZ (Fig. 2D and SI Appendix, Fig. S2H). In addition to epidermal and cortical cells, as previously reported (27, 37, 38), we also detected NRT2.1 in the columella, lateral root caps, and vasculature of roots (Fig. 2D and SI Appendix, Fig. S2H). The same overall expression pattern in the roots was also detected in transgenic lines harboring an NRT2.1:GFP transgene (Fig. 2E and SI Appendix, Fig. S2I). In agreement with previous studies (25, 36), low NO₃⁻ promoted expression of NRT2.1 as visualized by NRT2.1:GUS and NRT2.1:GFP (Fig. 2 D and E and SI Appendix, Fig. S2 H and I). These observations were further confirmed by reverse transcription-quantitative PCR (RT-qPCR) (SI Appendix, Fig. S2J).

The independent alleles and mutant complementation experiments above demonstrated that mutations in the NRT2.1 gene are responsible for the low-NO₃⁻-hyposensitive root growth phenotype of lonr2 mutants. Hence, we renamed the lonr2-1, lonr2-2, and lonr2-3 mutants as nrt2.1-5, nrt2.1-6, and nrt2.1-7, respectively. NRT2.1 encodes a major nitrogen starvation–inducible component of the NO₃⁻ high-affinity transport system (2, 25, 27). In addition to NO₃⁻ uptake, NRT2.1 also plays a crucial role in NO₃⁻ signaling, regulating lateral root initiation (26, 27). As such, NRT2.1 function would be required for root development, generally, with primary root elongation being only a subset of them.

nrt2.1 Seedlings Are Defective in Auxin Transport.

Auxin is the main regulator of root system architecture (39, 40). To explore how NO₃⁻ regulates root development, we analyzed auxin accumulation in the nrt2.1 alleles. We established that nrt2.1 mutants always showed higher auxin signaling in their root tips under low-NO₃⁻ supply compared to wild-type seedlings, as visualized using the auxin response reporters DR5rev:GFP and DII-VENUS (Fig. 3 A–D and SI Appendix, Fig. S3 A–D). Consistent with these observations, the nrt2.1 alleles grown on low NO₃⁻ had more endogenous indole-3-acetic acid (IAA) contents in their roots than those of wild-type seedlings (SI Appendix, Fig. S3E). These results indicate that NRT2.1 function leads to a reduction in auxin levels in root tips under low-NO₃⁻ conditions. The quantitative changes in auxin signaling activity and auxin contents thus suggested that NRT2.1 has a direct effect on auxin distribution.

Fig. 3.

nrt2.1 is defective in auxin transport. (A–D) Effects of low NO₃⁻ on auxin signaling in the root tips of the nrt2.1 T-DNA insertion mutant as revealed by expression analysis of DR5rev:GFP and DII-VENUS. The fluorescence intensity of DR5rev:GFP (A) and DII-VENUS (C) reporters is quantified in (B) and (D), respectively. (Scale bars, 50 µm.) Data are shown as means ± SE (n ≥ 15 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (E) Measurements of acropetal auxin transport in roots of nrt2.1 mutants. Data are shown as means ± SE (n ≥ 14 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (F) Primary root length in the nrt2.1 pin7-2 double mutant. Data are shown as means ± SE (n ≥ 18 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (G) Measurements of acropetal auxin transport in roots of the nrt2.1 pin7-2 double mutant. Data are shown as means ± SE (n ≥ 16 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (H–J) Primary root length in Super:PIN7-Flag seedlings (H), the F1 progeny from a cross between Super:NRT2.1-Flag and Super:PIN7-Flag seedlings (I), and nrt2.1 Super:PIN7-Flag lines (J). Data are shown as means ± SE (n ≥ 14 roots). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test).

Auxin accumulation at the root tip is the result of auxin biosynthesis and polar transport and determines auxin-mediated root growth (39, 41). We thus measured the expression levels of the auxin biosynthesis–related genes tryptophan aminotransferase of Arabidopsis1 (42), anthranilate synthase alpha subunit1 (43), and AMIDASE1 (44) by RT-qPCR, and found that they were expressed at similar levels in wild-type and nrt2.1 mutant seedlings grown under low-NO₃⁻ conditions (SI Appendix, Fig. S3F). Thus, these data suggest that NRT2.1 may not regulate auxin biosynthesis. We next quantified auxin transport in the roots of nrt2.1 mutants and observed higher acropetal (toward the root tip) IAA transport under low-NO₃⁻ conditions compared to that in the wild type (Fig. 3E). This result is consistent with the low-NO₃⁻-induced auxin response and auxin levels in roots of nrt2.1 mutants (Fig. 3 A–D and SI Appendix, Fig. S3 A–E). In summary, nrt2.1 loss-of-function phenotypes hint at a role for NRT2.1 in NO₃⁻-regulated root development, presumably as a result of the misregulated auxin transport.

Directional cell-to-cell auxin transport is mediated by, among other proteins, members of the canonical PIN auxin exporter family, which exhibit polar localization at the plasma membrane (41, 45, 46). To investigate the genetic interaction between NRT2.1 and PIN proteins, we crossed the nrt2.1 T-DNA insertion mutant with various pin loss-of-function mutants. The pin mutants displayed tolerance to low NO₃⁻ comparable to that of wild-type seedlings (Fig. 3F and SI Appendix, Fig. S3 G–I). Under low-NO₃⁻ conditions, the double mutants nrt2.1 eir1-1, nrt2.1 pin3-4, and nrt2.1 pin4-3 had the same enhanced root elongation as the nrt2.1 single mutant (SI Appendix, Fig. S3 G–I). By contrast, pin7-2 substantially suppressed the lower sensitivity to low NO₃⁻ of nrt2.1 in terms of root elongation (Fig. 3F). Moreover, under 0.05 mM NO₃⁻ conditions, pin7-2 also restored the normal endogenous IAA levels and acropetal IAA transport in nrt2.1 roots, as inferred from direct measurements of IAA contents and ³H-IAA accumulation (Fig. 3G and SI Appendix, Fig. S3J). Thus, we conclude that pin7 suppresses various aspects of the nrt2.1 phenotype.

We also generated transgenic plants carrying the Super:PIN7-Flag transgene in the wild-type background. Overexpressing PIN7 caused a low-NO₃⁻-hyposensitive phenotype affecting root elongation (Fig. 3H), similar to that of the nrt2.1 mutants. To further explore the genetic interaction between NRT2.1 and PIN7, we crossed Super:PIN7-Flag with Super:NRT2.1-Flag. In the F1 generation, we observed that NRT2.1 overexpression fully blocked the long-root phenotype of the PIN7 gain-of-function line when grown on medium containing 0.05 mM NO₃⁻ (Fig. 3I). Moreover, transgenic seedlings stably overexpressing PIN7 in the nrt2.1 mutant background were even more resistant than nrt2.1 mutants to low NO₃⁻ in terms of root growth (Fig. 3J). Collectively, these results reveal that the auxin efflux carrier PIN7 acts specifically downstream of NRT2.1 to positively regulate root growth in response to low-NO₃⁻ availability.

NRT2.1 Interacts Directly with PIN7 and Interferes with PIN7 Auxin Transport Activity.

NRT2.1 and PIN7 are expressed in largely overlapping patterns in columella and stele cells of the roots (47), and both are NO₃⁻-regulated genes (25, 38) (Fig. 2 D and E and SI Appendix, Figs. S2 H and I and S4 A and B). Low concentration of NO₃⁻ is most effective in promoting NRT2.1, but inhibiting PIN7 expression (Fig. 2 D and E and SI Appendix, Figs. S2 H and I and S4B).

We next assessed PIN7 expression pattern in the nrt2.1-6 mutant background. Notably, NRT2.1 mutation did not visibly affect PIN7 localization, reflected by the preferential occurrence of PIN7 at the basal side of stele cells and nonpolar plasma membrane distribution in the columella cells (48) (SI Appendix, Fig. S4C). Meanwhile, we also noticed no change of PIN7:PIN7-GFP signal intensity in nrt2.1-6 in response to NO₃⁻ availability when compared to that in the wild-type roots (SI Appendix, Fig. S4 C and D). Hence, these results suggest that NRT2.1 does not modulate PIN7 expression.

Considering that NRT2.1 and PIN7 are plasma membrane–localized transporters (2, 47), we next wondered whether NRT2.1 may directly interact with PIN7. To this end, we conducted yeast two-hybrid (Y2H) assays by cotransforming constructs encoding the N-terminal domain of NRT2.1 fused to the yeast GAL4 activation domain (AD-NRT2.1N) and the PIN7 hydrophilic loop fused with the GAL4 DNA-binding domain (BD-PIN7HL) into yeast strain AH109. We observed that NRT2.1N directly interacted with PIN7HL in yeast cells (Fig. 4A). We also performed coimmunoprecipitation (Co-IP) assays using total proteins extracted from Arabidopsis wild-type protoplasts transiently expressing NRT2.1N-GFP and PIN7HL-Flag constructs. We established that NRT2.1N coimmunoprecipitated with PIN7HL in vivo (Fig. 4B). The interaction between NRT2.1N and PIN7HL was further confirmed using a luciferase complementation imaging (LCI) assay by Agrobacterium tumefaciens–mediated transient coexpression in Nicotiana benthamiana leaves (SI Appendix, Fig. S4 E-G). Notably, the interaction of NRT2.1N and PIN7HL was attenuated by spraying the leaves with 10 mM NO₃⁻ (Fig. 4 C and D). These observations reveal that NRT2.1 physically interacts with PIN7 in vitro and in vivo, and that this interaction is sensitive to NO₃⁻.

Fig. 4.

NRT2.1 physically interacts with PIN7 and interferes with PIN7-mediated auxin transport activity. (A) Y2H assay demonstrating that NRT2.1N interacts with PIN7HL. As a negative control, NRT2.1N does not interact with PIN2HL. -LW, synthetic defined (SD) medium lacking Leu and Trp; -LWHA, SD medium lacking Leu, Trp, His, and Ade. (B) Co-IP assay demonstrating that NRT2.1N interacts with PIN7HL in Arabidopsis protoplasts. As a negative control, NRT2.1N does not coimmunoprecipitate with PIN2HL. (C and D) LCI assay demonstrating that interaction between NRT2.1N and PIN7HL is attenuated by application of 10 mM NO₃⁻ in N. benthamiana leaves. The luciferase signal intensity (C) is quantified in (D). Data are shown as means ± SE (n = 24 replicates). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (E) Auxin efflux assays in Xenopus oocytes injected with different cRNAs as specified. Data are shown as means ± SE (n = 4 replicates; each replicate contains six oocytes). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test). (F) Auxin efflux assays in BY-2 cells expressing the indicated constructs under different external NO₃⁻ conditions. Data are shown as means ± SE (n ≥ 3 replicates; each replicate contains 0.5 mL suspension cells). Different lowercase letters indicate significant differences (P < 0.05; one-way ANOVA followed by Fisher’s LSD test).

To investigate the physiological relevance of NRT2.1 and the NRT2.1–PIN7 interaction in NO₃⁻-regulated root development, we measured the auxin transport activity of NRT2.1 in a heterologous test system using Xenopus oocytes. As reported previously (22, 38), bimolecular fluorescence complementation signal for the interaction between full-length NRT2.1 and NRT3.1 was detected at the plasma membrane of the oocytes (SI Appendix, Fig. S5A). The proper accumulation of PIN7 at the plasma membrane in oocytes was also confirmed by immunoblot analysis (SI Appendix, Fig. S5B). Consistent with a previous study (49), we observed that the oocytes coexpressing PIN7 and serine/threonine protein kinase PINOID (PID) accumulated less ³H-IAA than that of the background control (Fig. 4E). Auxin efflux in oocytes coexpressing NRT2.1 and NRT3.1 did not differ from that in the background control, suggesting that NRT2.1 cannot directly transport auxin. However, when we coinjected complementary RNA (cRNA) of NRT2.1 and NRT3.1 with PIN7 and PID into oocytes, we measured significantly higher ³H-IAA accumulation compared to that in oocytes expressing PIN7 and PID only (Fig. 4E). Because PID was coexpressed with NRT2.1 in these experiments, it is unclear whether the differences in intracellular auxin accumulation observed in oocytes are a result of modulated NRT2.1 activity by PID kinase through direct phosphorylation. Hence, we first investigated the possibility of physical interaction between PID and NRT2.1. By preforming Y2H and Co-IP assays, we observed that PID did not associate with NRT2.1N in vitro and in vivo (SI Appendix, Fig. S5 C and D). To establish whether PID kinase could phosphorylate NRT2.1, we next performed in vitro phosphorylation assays with recombinant proteins. Although we demonstrated that PID underwent autophosphorylation and mediated the phosphorylation of PIN2HL, no phosphorylated band of NRT2.1N was observed (SI Appendix, Fig. S5E). Moreover, we coinjected cRNA of NRT2.1 and NRT3.1 with PID into oocytes, and we observed that the amount of NO₃⁻ retained was essentially unchanged compared to that in oocytes expressing NRT2.1 and NRT3.1 (SI Appendix, Fig. S5F), indicating that NO₃⁻ uptake activity of NRT2.1 is not dependent on PID kinase. We also tested the potential PID effect on NRT2.1 in ³H-IAA efflux assays and found that NRT2.1 alone or NRT2.1 and NRT3.1 together could not directly transport auxin regardless of the presence or absence of PID (Fig. 4E). Together, these observations exclude the involvement of PID in regulating NRT2.1 activity. Furthermore, to rule out the possibility of artifacts induced by overloading oocytes with multiple samples, we also included several negative controls among the transport assays. Importantly, we observed that PIN7-mediated ³H-IAA efflux was not influenced by the NRT2.1^R406STOP variant (the mutation identified in lonr2-3) with impaired NO₃⁻ uptake (SI Appendix, Fig. S5F), the plasma membrane intrinsic protein PIP2;1, or the NO₃⁻ transporter NRT2.5 (Fig. 4E and SI Appendix, Fig. S5G). In summary, these data show that although NRT2.1 is not an auxin transporter, it specifically antagonizes PIN7-mediated auxin efflux in the oocyte expression system.

Tobacco BY-2 suspension cells provide a well-established and tractable model for measuring auxin transport. Therefore, we determined the effect of NRT2.1 on auxin transport activity of PIN7 in BY-2 cells. First, we verified the normal expression levels of NRT2.1 and PIN7 in BY-2 cell lines by RT-qPCR (SI Appendix, Fig. S5H). Without NO₃⁻ treatment, cells stably expressing PIN7 accumulated less ³H-IAA than that of control cells transfected with the empty vector (Fig. 4F). Coexpression of NRT2.1 and NRT3.1 had no effect on ³H-IAA accumulation but clearly interfered with PIN7-mediated auxin efflux (Fig. 4F). In agreement with the oocyte data, when NRT2.1 was replaced with the truncated variant NRT2.1^R406STOP, or the plasma membrane–localized PIP2;1, or the NO₃⁻ transporter NRT2.5, PIN7-dependent auxin transport was no longer influenced (Fig. 4F). However, when BY-2 cells were treated with NO₃⁻ (5 mM), PIN7-mediated IAA efflux was no longer affected by the presence of NRT2.1 (Fig. 4F). This observation is consistent with the nrt2.1 defects specifically seen under low-NO₃⁻ conditions (SI Appendix, Fig. S1F). To further determine whether the effect on auxin transport is due to indirect factors associated with low-NO₃⁻ conditions, we measured ATP levels and apoplastic pH in BY-2 cells under different external NO₃⁻ concentrations. Notably, changes in NO₃⁻ availability did not affect either the ATP concentration or the apoplastic pH (SI Appendix, Fig. S5 I and J). Together, these results suggest that NRT2.1 negatively and specifically affects PIN7-induced auxin transport in BY-2 cells, and this regulation is dependent on NO₃⁻ levels.

Taken together, these data suggest that the high-affinity NO₃⁻ transporter NRT2.1, which responds to NO₃⁻ availability, specifically interferes with PIN7-mediated auxin transport under limited NO₃⁻ conditions and thereby regulates root growth.

Discussion

Modulation of primary root growth is a prominent foraging strategy for adaptation to NO₃⁻ fluctuation (28, 29); however, the underlying molecular mechanisms remain to be investigated. In this study, we identified lonr2 mutants with a defect in low-NO₃⁻-mediated suppression of primary root growth. We determined that LONR2 was the previously characterized high-affinity NO₃⁻ transporter gene NRT2.1. Our results indicated that LONR2/NRT2.1 plays an essential role in regulating primary root elongation in response to low-NO₃⁻ stress. We uncovered a functional interaction between NRT2.1 and PIN7, finding that NRT2.1 interfered with PIN7-mediated auxin efflux under low-NO₃⁻ conditions (Fig. 5). This study reveals a previously unanticipated mechanism behind adaptive root development.

Fig. 5.

Schematic conceptual model for NRT2.1 control of primary root growth in response to NO₃⁻ availability. Two situations are shown to illustrate the specific effect of low NO₃⁻ on primary root growth of the wild type and the nrt2.1 mutant. The model postulates that under NO₃⁻-limited conditions, expression of the high-affinity NO₃⁻ transporter NRT2.1 is induced and accumulated NRT2.1 directly interacts with the auxin efflux carrier PIN7, which in turn suppresses PIN7-mediated acropetal auxin efflux. This ultimately slows down primary root elongation. Conversely, lack of NRT2.1 relieves auxin export activity of PIN7, eventually facilitating the rootward polar auxin transport by PIN7 and accelerating adaptive root growth in response to low NO₃⁻.

Although NO₃⁻ uptake rates are compromised in nrt2.1 mutants (24, 26, 27) (SI Appendix, Fig. S2B), our results support the idea that the loss of NRT2.1 function promotes primary root elongation under low-NO₃⁻ conditions independently of this reduced NO₃⁻ uptake. When NO₃⁻ was almost depleted from the medium, wild-type seedlings did not recapitulate the phenotype of nrt2.1 mutants (SI Appendix, Fig. S1F). Conversely, when exogenous NO₃⁻ concentrations gradually increased, root growth of nrt2.1 mutants was not repressed (SI Appendix, Fig. S1F). Therefore, we propose that NRT2.1 plays a role in developmental signaling in addition to its function in high-affinity NO₃⁻ uptake.

Due to its role in NO₃⁻ uptake, we expected NRT2.1 expression to be confined to the outer cell layers of mature roots (27, 37, 38), as this is where the bulk of nutrient uptake usually occurs. However, according to our observations of promoter and functional fusions in this study, NRT2.1 was expressed throughout the entire root. Indeed, we detected NRT2.1 in the columella, lateral root caps, epidermis, cortex, and vasculature of roots (Fig. 2 D and E and SI Appendix, Figs. S2 H and I and S4A). The discrepancy between previous studies and our results likely reflects the use of shorter NRT2.1 promoter fragments in the previous reports (27, 37, 38). Although unexpected at first glance, this widespread expression pattern in roots hints that NRT2.1 must have a second function in addition to its role in NO₃⁻ uptake. The overlapping expression patterns of NRT2.1 and the auxin efflux carriers (e.g., PIN7) in root columella and vascular cells raise the possibility that these proteins have at least partially overlapping functions, which are probably related to auxin transport.

As demonstrated by the Y2H, Co-IP, and LCI assays, NRT2.1 physically interacted with PIN7 (Fig. 4 A and B and SI Appendix, Fig. S4 E–G). Unlike NRT1.1, which directly transports auxin, NRT2.1 did not; instead, we showed that it antagonized the auxin efflux activity of PIN7 in Xenopus oocytes and tobacco BY-2 cells depending on NO₃⁻ levels (Fig. 4 E and F). Given that auxin is structurally similar to one of the important nitrogen source amino acids, such as tryptophan (50, 51), it is not surprising that the NO₃⁻ transporter NRT2.1 might also participate in auxin transport. Indeed, several lines of evidence support the notion that NRT1.1 also directly transports auxin (1). Here, the observations that NRT2.1 expression inhibits PIN7-mediated auxin efflux in oocytes or BY-2 cells whereas the mutated variant NRT2.1^R406STOP, the plasma membrane–localized PIP2;1, or the NO₃⁻ transporter NRT2.5 did not (Fig. 4 E and F and SI Appendix, Fig. S5G), rule out the possibility of artifacts caused by the heterologous expression of transporters. The elevated acropetal auxin transport seen in nrt2.1 mutants was clearly suppressed in the nrt2.1 pin7-2 double mutant (Fig. 3G), confirming the idea that NRT2.1 regulates auxin transport in planta. Thus, in this study, we established a regulatory mechanism, whereby NRT2.1 plays a critical role in the integration of environmental and physiological information in response to nutrient availability. However, further research is needed to determine whether other factors contribute to the interaction between NRT2.1 and PIN7 and whether this interaction is regulated by protein modification events, such as phosphorylation.

In summary, we demonstrated that NRT2.1 modulates root growth in response to fluctuating NO₃⁻ availability in the soil, interacts with PIN7, and interferes with PIN7-mediated auxin transport activity depending on NO₃⁻ levels. Manipulation of the evolutionarily conserved gene NRT2.1 (52, 53) may enable targeted engineering of auxin flux to achieve desired changes in root architecture and development, especially in food crops.

Materials and Methods

Plant Materials and Growth Conditions.

The following previously described mutant and transgenic Arabidopsis plants were used: nrt2.1 (SALK_141712) (24); nrt2.1 NRT2.1:NRT2.1 (36); eir1-1 (54); pin4-3 (55); pin3-4, pin7-2, and DR5rev:GFP (56); DII-VENUS (57); CYCB1;1:DB-GUS (58); and PIN7:PIN7-GFP (59). The primers used for genotyping are listed in SI Appendix, Table S1.

Arabidopsis seedlings were grown on full-strength NO₃⁻ (10 mM) and low-NO₃⁻ medium (0.05 mM) as described (9). Arabidopsis ecotype Columbia (Col-0) was used as the wild type in this study.

Detailed Materials and Methods can be found in SI Appendix.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.