ZmNRT1.1B (ZmNPF6.6) determines nitrogen use efficiency via regulation of nitrate transport and signalling in maize

Summary Nitrate (NO3 −) is crucial for optimal plant growth and development and often limits crop productivity under low availability. In comparison with model plant Arabidopsis, the molecular mechanisms underlying NO3 − acquisition and utilization remain largely unclear in maize. In particular, only a few genes have been exploited to improve nitrogen use efficiency (NUE). Here, we demonstrated that NO3 −‐inducible ZmNRT1.1B (ZmNPF6.6) positively regulated NO3 −‐dependent growth and NUE in maize. We showed that the tandem duplicated proteoform ZmNRT1.1C is irrelevant to maize seedling growth under NO3 − supply; however, the loss of function of ZmNRT1.1B significantly weakened plant growth under adequate NO3 − supply under both hydroponic and field conditions. The 15N‐labelled NO3 − absorption assay indicated that ZmNRT1.1B mediated the high‐affinity NO3 −‐transport and root‐to‐shoot NO3 − translocation. Transcriptome analysis further showed, upon NO3 − supply, ZmNRT1.1B promotes cytoplasmic‐to‐nuclear shuttling of ZmNLP3.1 (ZmNLP8), which co‐regulates the expression of genes involved in NO3 − response, cytokinin biosynthesis and carbon metabolism. Remarkably, overexpression of ZmNRT1.1B in modern maize hybrids improved grain yield under N‐limiting fields. Taken together, our study revealed a crucial role of ZmNRT1.1B in high‐affinity NO3 − transport and signalling and offers valuable genetic resource for breeding N use efficient high‐yield cultivars.


À
) is the predominant form of nitrogen (N) source for nutrition of most terrestrial plants (Crawford and Glass, 1998;Gao et al., 2022;Hirel et al., 2007).NO 3 À is readily dissolved and lost through leaching in most ecosystems and agricultural soils, leading to highly variable NO 3 À concentration in both time and space (Cameron et al., 2013;Gallardo et al., 2009;Miller et al., 2007).To acclimate to these fluctuations, plants evolved the spectacular physiological and developmental plasticity including modifying root architecture, adjusting NO 3 À uptake and transport, and optimizing NO 3 À metabolism (Giehl and von Wiren, 2014;Jia et al., 2022;Jia and von Wiren, 2020;Kiba and Krapp, 2016;Liu and von Wiren, 2022;Nacry et al., 2013).In addition to being the nutrition in plants, NO 3 À also acts as a potent signal molecule that modulates genome-wide gene expression, metabolism, physiology as well as growth and developmental processes (Bouguyon et al., 2015;Fredes et al., 2019;Poitout et al., 2018;Vidal et al., 2015).Identifying and genetically manipulating the key components of NO 3 À transport and signalling is crucial for genetic improvement of crop N use efficiency (NUE).
Extensive studies have uncovered a great array of regulators in the NO 3 À signalling in the past two decades (Vidal et al., 2020).
Their indispensable roles vary on N availability as the relevant function of OsNLP1 and OsNLP3 under N-sufficient conditions but OsNLP4 under N-deficient conditions.A similar function of AtNRT1.1 and AtNLP7 orthologous genes is supposed in maize, which is habitat-adapted to arable soil where NO 3 À is the major available N source.By in vitro functional characterization in the heterogeneous expression system X.laevis oocytes, ZmNPF6.6 (ZmNRT1.1B) and ZmNPF6.4(ZmNRT1.1A)have been shown to transport NO 3 À and chloride (Wen et al., 2017).In vitro assay also shows that ZmNLP5/6/8 activate the NO 3 À response by binding to NRE in the promoters of maize NO 3 À -responsive genes (Cao et al., 2017;Ge et al., 2020;Wang et al., 2018b).However, the in planta function of maize NRT1.1 s and NLPs and their contribution to NUE remain to be elucidated.
In this study, we showed the NO 3 À -inducible and plasmamembrane localized ZmNRT1.1Bmediates root NO 3 À uptake and signalling, which in turn modulates downstream N assimilation and plant growth through the regulation of nuclear-cytoplasmic shuttling of the transcription factor ZmNLP3.1.Loss of function of ZmNRT1.1B and ZmNLP3.1 inhibits growth of maize plants under NO 3 À nutrition.Notably, overexpressing ZmNRT1.1Bconfers significantly higher grain yield under low-to moderate-N supply in the fields.These findings extended our understanding of NO 3 À transport and signalling across plant species and provide valuable gene resources for breeding maize cultivars with improved NUE.

Results
The duplicated ZmNRT1.1B and ZmNRT1.1Cdiverge in expression pattern and protein localization Gene duplication is a major force of genetic innovation that acquires new genes or novel functions of retaining ones in the organisms (Magadum et al., 2013).To gain insight into the genetic evolution of NRT1.1 genes, we retrieved protein sequences of NRT1.1 s from plant species Arabidopsis thaliana, Zea mays, Oryza sativa, Sorghum bicolor and Brachypodium distachyon.Comparative genomic analysis showed that whereas only one NRT1.1 expresses in the Arabidopsis genome, the copy number of NRT1.1 increases to 3-4 in the monocot species, suggesting that gene expansion of NRT1.1 occurred after the dicot and monocot specification (Figure 1a).In support, the phylogenetic analysis clustered NRT1.1s into dicot-and monocot-specific clades.Collinearity analysis also showed that during the evolution, ZmNRT1.1A/OsNRT1.1Adiverged from NRT1.1s, and ZmNRT1.1B/OsNRT1.1B and ZmNRT1.1D/OsNRT1.1Csubsequently evolved through whole-genome duplication (WGD) (Figure 1b).Interestingly, ZmNRT1.1B and ZmNRT1.1Cencode 89% identical proteins and reside in close vicinity on the chromosome 1, probably deriving from a local tandem gene duplication (Figures 1b and S1).
The evolution of genome-wide duplicated transporter proteins, for instance in tandem, represents a likely driving force for gene sub-or neofunctionalization, either through changing the substrate selectivity, protein localization or spatiotemporal expression patterns of resulting proteoforms (Jia et al., 2021;Pottier et al., 2022).To gain insights into whether ZmNRT1.1B and ZmNRT1.1Cundergo sub-or neofunctionalization, we first compared the expression patterns of ZmNRT1.1B and ZmNRT1.1C in different tissues of field-grown maize inbred line B73 plants at the silking stage.The quantitative real-time (qRT)-PCR analyses showed that ZmNRT1.1Bexhibited 24-1600 times higher transcript levels than those of ZmNRT1.1C in diverse tissues analysed (Figure S2a,b).The AtNRT1.1 and OsNRT1.1Bare transcriptionally responsive to NO 3 À addition and crucial for NO 3 À uptake (Hu et al., 2015;Huang et al., 1996;Liu et al., 1999;Touraine and Glass, 1997;Tsay et al., 1993).We then investigated the transcriptional responses of ZmNRT1.1B and ZmNRT1.1C to NO 3 À and ammonium (NH 4 + ) at various concentrations.Despite not responsive to NH 4 + supply, transcript level of ZmNRT1.1Bincreased steadily with increasing NO 3 À supply (Figure 1c).Distinct to ZmNRT1.1B, the transcript level of ZmNRT1.1C was increased by both NO 3 À and NH 4 + at high concentrations (Figure 1d).
We then investigated the expression dynamics of ZmNRT1.1B and ZmNRT1.1C in response to N deprivation and resupply.Whereas N deprivation down-regulated transcript levels of ZmNRT1.1B and ZmNRT1.1C(Figure 1e,f), they were significantly up-regulated by addition of NO 3 À and reached maximum at 6 h (Figure 1g).In contrast to ZmNRT1.1C, the expression levels of ZmNRT1.1B were not responsive to NH 4 + resupply (Figure 1g,h).
These data collectively indicate that the duplicated ZmNRT1.1B and ZmNRT1.1Cdiverge in their expression pattern in response to N forms and doses.
Loss of function of ZmNRT1.1Bbut not ZmNRT1.1Cweakens plant growth under replete NO 3 À supply To assess the physiological significance of ZmNRT1.1B and ZmNRT1.1C,we then generated the loss-of-function mutants zmnrt1.1b and zmnrt1.1cby CRISPR/Cas9-mediated gene editing in the background of inbred line ND101 (Figure S4a,b).These mutants were hydroponically cultured under low-or high-NO 3 À conditions for 14 days to measure the dry biomass and N content (Figure 2a).In comparison with WT, both zmnrt1.1b-1and zmnrt1.1b-2reduced shoot dry weight by 10%-20% under low and high NO 3 À , and root dry weight by 20% under high NO 3 À (Figure 2b).As a consequence, the shoot and root N contents of two zmnrt1.1bmutants were significantly lower than those of WT, especially under high NO 3 À (Figure 2c).ZmNRT1.1Bmodulates high-affinity NO 3 À uptake and signalling in planta Retarded growth of the zmnrt1.1bmutants in the presence of NO 3 À implied that the zmnrt1.1bmight impair the NO 3 À transport and/or signalling.As ZmNRT1.1B is localized to the plasma membrane, we first characterized the high-and lowaffinity NO 3 À uptake by short-term 15 N-labelled NO 3 À influx of hydroponically grown plants (Figure 3a).When plants were precultured under N starvation for 4 days prior to influx assay, the high-affinity NO 3 À uptake capacity of zmnrt1.1bmutants was ~20% significantly lower than that of WT.For the nitrateinducible NO 3 À influx rates, zmnrt1.1bmutant plants showed 10%-15% reduction compared with WT after 3 h pretreatment with 4 mM NO 3 À .By contrast, no significant difference between WT and mutant plants was detected when they were supplied with 5 mM 15 NO 3 À to assess the low-affinity NO 3 À uptake capacity (Figure 3b).
To further determine whether ZmNRT1.1Bcontributes to the root-to-shoot NO 3 À translocation, we quantified the shoot/root 15 N ratio by extending the 15 NO 3 À feeding period to 0.25 h, 3 h and 24 h (Figure 3c).Whereas comparable to WT at 0.25 h and 3 h, we observed that shoot/root 15 N ratio in zmnrt1.1bmutants significantly reduced by 31.8% and 23% compared with WT after 24 h, suggesting that ZmNRT1.1B is also crucial for the rootto-shoot NO 3 À translocation.
We next examined the expression of N utilization-related genes in WT and zmnrt1.1bmutant plants cultivated under low-and high-NO 3 À regimes.The transcript levels of NO 3 À assimilation genes including ZmNR1.1 and ZmNR1.2 consistently and significantly decreased under high NO 3 À , suggesting that NO 3 À assimilation may be compromised (Figure 3d).We further assessed the expression of primary NO 3 À response (PNR) marker genes including ZmNR1.1, ZmNR1.2 and ZmNRT2.1 to investigate whether ZmNRT1.1B is required for NO 3 À signalling.After 3 h short-term NO 3 À treatments, the transcriptional induction of ZmNR1.1,ZmNR1.2 and ZmNRT2.1 in roots of both zmnrt1.1b-1and zmnrt1.1b-2mutants were largely reduced in comparison with WT, suggesting that ZmNRT1.1B is functionally relevant for NO 3 À signalling (Figure 3e).Taken together, these results indicate that ZmNRT1.1Bcontributes to the high-affinity NO 3 À uptake, root-to-shoot translocation and NO 3 À signalling in planta.
Given the large proportion of NO 3 À -regulated genes shared by ZmNRT1.1B and ZmNLP3.1,we then explored the genetic relationship of ZmNRT1.1B and ZmNLP3.1 in NO 3 À signalling.
We first investigated whether ZmNRT1.1B was sufficient to regulate ZmNLP3.1 expression.However, we observed that the transcript level of ZmNLP3.1 was not significantly different between WT and zmnrt1.1bmutants under neither low nor high NO 3 À condition (Figure S6).We then tested whether ZmNLP3.1 nuclear retention is regulated by NO 3 À and dependent on function of ZmNRT1.1B.Similar to AtNLP7 in Arabidopsis, we observed that ZmNLP3.1 dispersed in the cytoplasm in the absence of NO 3 À , and ZmNLP3.1-eGFPfluorescence was perfectly obviously co-localized with the nuclear dye DAPI in the presence of NO 3 À , suggesting NO 3 À addition promoted ZmNLP3.1 nucleus retention.Strikingly, the loss of function of ZmNRT1.1Bcompletely prevented ZmNLP3.1 cytoplasmic-nuclear shuttling (Figure 4e).Taken together, these results indicate that ZmNRT1.1Bregulates NO 3 À signalling through modulating cytoplasmic-nuclear shuttling of ZmNLP3.1.

ZmNLP3.1 positively regulate NO 3 À -mediated plant growth
To explore the role of ZmNLP3.1 in NO 3 À -regulated plant growth and development, we analysed the growth phenotypes of two zmnlp3.1 mutants.Although at low NO 3 À supply, growth of WT and zmnlp3.1 did not differ significantly, the growth promotion by high NO 3 À was obviously suppressed in each zmnlp3.1 mutant relative to WT (Figure 5a).Under replete NO 3 À , the root and shoot dry biomass was significantly reduced by ~35% and ~30%, respectively, compared with WT.Similarly, the total N accumulation in both root and shoot was also decreased by ~40% and ~43%, respectively (Figure 5b,c).We further assessed the transcript levels of NO 3 À uptake and assimilation genes to test whether compromised NO 3 À metabolism and/or acquisition was causal for the weak growth of zmnlp3.1 mutants.Whereas the transcript level of ZmNRT2.1 was similar as WT under replete NO 3 À condition, we observed that the expression of both ZmNR1.1 and ZmNR1.2 was significantly down-regulated (Figure S7), suggesting that NO 3 À assimilation is largely compromised, which in turn results in growth defects observed in zmnlp3.1 mutants.The zmnrt1.1b and zmnlp3.1 mutants reduce biomass and yield under field conditions We tested whether ZmNRT1.1B and ZmNLP3.1 determine plant growth and NUE in the field.In two environments (Beijing and Hainan), the field trials growing zmnrt1.1bmutants with sufficient N supply were then performed.Compared with WT plants, zmnrt1.1bmutants stably decreased the shoot biomass by ~20%-25% and N contents by ~15%-25%, resulting in a ~30%-40% reduction in grain yield (Figure S8).As a consequence, the NUE of zmnrt1.1bmutants reduced by ~44%-50% (Figure S8).
To further evaluate the potential contribution of ZmNRT1.1B and ZmNLP3.1 to N-dependent yield formation, we planted zmnrt1.1band zmnlp3.1 mutants under various N regimes.The aboveground biomass of zmnrt1.1bswas greatly reduced under high-N conditions, while no significant difference compared with WT under low-N during the main growth periods (Figure 6a).For yield formation, zmnrt1.1bstably reduced ear length by ~15% and ~12%, resulting in a decrease in yield per plant of ~12% and ~38% compared to WT under low-and high-N conditions, respectively (Figure 6b; Data S6).For zmnlp3.1 mutants, the aboveground biomass at the silking and maturation stages reduced by 26%-35% and 20%-23%, under low-and high-N, respectively (Figure 6c).Remarkably, zmnlp3.1 mutants showed significant decreases in ear length (average decrease ~14% and ~8.7%) and ear width (~9% and ~8.7%), resulting in ~29% and ~45% decrease of yield per plant than that of WT under lowand high-N conditions, respectively (Figure 6d; Data S6).These findings indicate that loss of function of ZmNRT1.1B and ZmNLP3.1 results in reduction in plant growth and grain productivity.

Overexpressing ZmNRT1.1B improves grain yield under N-limiting field
To assess the breeding values of ZmNRT1.1B,we created three independent transgenic overexpression lines and introgressed them into the inbred PH6WC, that is female parent of commercial hybrid Xianyu335 (XY335) prevailing in the northeast of China (Figure 7a).We selected two sets of BC 4 F 2 lines, NIL WT and NIL OE .In these NIL OE lines, the transcript levels of ZmNRT1.1B were significantly increased up to 10-50 folds higher than those of corresponding NIL WT lines (Figure S9a).All these NIL OE lines exhibited significantly higher tolerance to low NO 3 À availability than NIL WT lines when they were grown in hydroponics, as exemplified by 18%-26% increases in shoot biomass under low NO 3 À (Figure S9b,c).
To further examine whether ZmNRT1.1Bcould improve grain yield in hybrid breeding practice, we first generated F 1 testcross, TC WT and TC OE , by crossing heterozygous BC 4 F 2 positive plants with the male parent PH4CV (Figure 7a).We then undertook the field trials in 2018 to evaluate the agronomic traits of the F 1 plants with overexpressed ZmNRT1.1B(TC OE ) and the corresponding control (TC WT ).The plants were grown in the fields fertilized with 0 kg N/ha (0% N, low), 158 kg N/ha (70% N, moderate) and 225 kg N/ha (100% N, high) of urea.Under the low-N conditions, TC OE plants increased ear length by ~5%-14.2%and hundred-kernel weight by ~2%-9%, resulting in an ~13%-22.5% increase in yield per plant compared with their WTs (Figure 7b; Data S7).Under the moderate-N condition, OE1 and OE13 plants increased hundred-kernel weight by ~3.5%-4.5%,resulting in 2.5%-7% increase of the yield per plant.However, yield improvement in OE1 and OE3 was not observed under the high-N condition (Figure 7b; Data S7).
We further repeated the back-crossing (BC 7 ) to develop ZmNRT1.1B-overexpressinginbred lines in PH6WC background, PH6WC OE (Figure 7a).By crossing them with the male parent PH4CV, the hybrid XY335 OE was created.We then undertook the field trials in 2020 by growing the transgenic version of XY335 (XY335 OE ) and non-transgenic controls (XY335) under various N fertilizer rates (Figure 7c).Overexpression of ZmNRT1.1Bimproved the ear length and hundred-kernel weight of transgenic XY335 hybrid under both low-and moderate-N conditions (Data S7).Consequently, grain yield per plant of the XY335 OE was increased about ~15% and ~11% over XY335 under lowand moderate-N condition, respectively (Figure 7c; Data S7).Under high-N conditions, these ZmNRT1.1B-OEplants did not show improved yield per plant, hundred-kernel weight and ear length (Figure 7; Data S7).These results collectively indicate that the enhanced expression of ZmNRT1.1Bincreases yield per se as well as at the hybrid levels, presenting great breeding values for developing N-efficient maize hybrids.

Discussion
To adapt to the fluctuant NO 3 À concentration in the environment, plants can activate complex regulatory networks for optimizing N uptake and utilization (Good et al., 2004;Li et al., 2017;Masclaux-Daubresse et al., 2010;Xu et al., 2012).In maize as an essential food and cash crop, however, the molecular mechanism of N acquisition and utilization is poorly understood.
In this study, we demonstrated that ZmNRT1.1Bmediates the high-affinity NO 3 À transport and regulates NO 3 À signalling to sustain maize growth and development under both laboratory and field conditions.These findings provide a valuable candidate gene ZmNRT1.1Bthat may innovate strategies to improve NUE in maize.

Expansion and functional differentiation of NRT1.1 in grass species
Whole genome-wide and/or tandem duplications frequently occur in grass species during evolution, leading to functional diversification among the orthologs (Bolot et al., 2009).Distinct to Arabidopsis that retains only one NRT1.1 gene in the genome, the grass species evolved three to four NRT1.1 members falling into three subclades (Figure 1a; (Plett et al., 2010; Wang et al., 2020), suggesting that grass duplicated NRT1.1 may innovate their functions other than NO 3 À transport.In fact, accumulating evidences showed some of the NRT1.1 members retained their functions in the NO 3 À transport/signalling; however, some do evolve novel functions (Wang et al., 2020).For instance, OsNRT1.1Acould regulate flowering time and OsNRT1.1Bplayed an important role in determining the diversity of the root microbiome in rice (Wang et al., 2018a;Zhang et al., 2019a).Interestingly, in maize, the tandemly duplicated ZmNRT1.1B and ZmNRT1.1Clargely differentiated their expression pattern and protein localization (Figures 1 and S3).For example, expression of ZmNRT1.1B was 2-3 orders of magnitude higher than that of ZmNRT1.1C(Figures 1c-h and S2).Furthermore, similar to AtNRT1.1 and OsNRT1.1B,ZmNRT1.1B is NO 3 À -inducible and plasma membrane bound, while ZmNRT1.1C is localized in the endomembrane system (Figure S3).This suggests that ZmNRT1.1B,but not ZmNRT1.1C,retains the primary function in the NO 3 À uptake, transport and signalling.In support, we observed that the loss of function of ZmNRT1.1Bcauses significant reduction in the plant growth and NO 3 À uptake (Figures 2a-c and 3a).By contrast, the loss of function of ZmNRT1.1Cresulted in no apparent growth defects under various NO 3 À supply (Figure 2d-f).
Root NO 3 À absorption from soil determines plant growth and productivity for most crops (Dhugga et al., 1988).To counteract the changing NO 3 À condition in the environment, plants have evolved two NO 3 À uptake systems: the high-affinity transport system (HATS) and low-affinity transport system (LATS) (Crawford and Glass, 1998).
Within the NPF family, AtNRT1.1/OsNRT1.1Bwas demonstrated as the dual-affinity NO 3 À transporter (Hu et al., 2015;Liu et al., 1999).The ZmNRT1.1B functions as the dual-affinity NO 3 À transporter when it is heterogeneously expressed in Xenopus oocytes (Wen et al., 2017), while it displayed a high-affinity NO 3 À transport activity in planta (Figure 3a).Together with rapidly responds to exogenous NO 3 À supply in roots at transcript levels, ZmNRT1.1Bpresumably allows plants to efficiently absorb and utilize NO 3 À according to local N status (Krapp et al., 2014).
As the NO 3 À sensors, NRT1.1 and NLP7 act at the centre of the NO 3 À signalling network that is crucial for fast genome-wide transcriptional reprogramming and long-term developmental changes (Liu et al., 2017(Liu et al., , 2022;;Maghiaoui et al., 2020b).We found that more than 40% of the NO 3

Use of ZmNRT1.1B for maize breeding
Crop genetic improvement is the most cost-effective strategy to tackle the N dilemma for sustainable agriculture towards reducing N demand while pursuing high yield.Among cereals, maize has the highest yield potential that can be realized with more N fertilizer inputs (Wani et al., 2021).Our research shows application potential of ZmNRT1.1B for improving the NUE and grain yield in maize breeding.The zmnrt1.1b mutants showed growth retardation and yield loss in the field supplied with high N conditions (Figure 6a,b).By contrast, overexpressing of ZmNRT1.1Bsignificantly increases ~10% yield of maize testcross in the N-limiting field (Figure 7).In particular, overexpressing ZmNRT1.1B in the elite maize hybrid line XY335 performed well in a N-poor environment and maintained a normal yield even under field with 30% N cut (Figure 7c).Compared with known NO 3

À
-transporter genes such as OsNRT1.1Aand OsNRT2.3b,increased expression of ZmNRT1.1Bdid not alter grain yield under high N conditions, which may be related to the high-affinity N transporter activity (Figures 3a and 7; Fan et al., 2016;Wang et al., 2018a).Notably, introgression of ZmNRT1.1Binto the hybrid line XY335 barely impacts the elongation of plant height, minimizing the likelihood of plant lodging that may penalize the yield (Data S7).These suggest that ZmNRT1.1B is of great value to crop improvement with an aim of reducing N fertilizer use while maintaining grain yield.In addition, loss-of-function ZmNLP3.1 mutants showed corresponding reductions in biomass production and grain yield (Figure 6c,d).Along with this, field evaluation of breeding value of NO 3 À signalling gene ZmNLP3.1 will be worthwhile being undertaken in the future.Taken together, our results demonstrate that a plasma membrane-localized maize transporter ZmNRT1.1Bhas great potential for NUE and yield improvement via activating NO 3 À absorption and NO 3 À signalling.When supplied with sufficient NO 3 À , ZmNRT1.1B promotes ZmNLP3.1 nuclear retention, which activates the expression of NO 3 À assimilation genes and promotes plant growth.Upon limited NO 3 À , ZmNRT1.1B functions as a high-affinity NO 3 À transporter to acquire NO 3 À and enables higher tolerance of plants to limiting N. Given the significant contribution to NUE, resequencing of ZmNRT1.1B and consequent allelic mining in the germplasm pools to uncover relevant natural variants will be valuable in the further study.

Plant materials
Maize (Zea mays L.) inbred line B73 was used for gene cloning and expression analyses., seedlings of maize inbred line (B73) were grown in N-free nutrient solution supplied either with 0.04/0.4/4/10mM NH 4 Cl or KNO 3 for 10 days.To investigate the dynamic transcriptional responses to N starvation or resupply, B73 seedlings were hydroponically pre-cultured under supply of 2 mM NH 4 NO 3 for 10 days and transferred to N-free nutrient solution for 24, 48 or 96 h.Afterwards, N was added in the form of 4 mM KNO 3 or 4 mM NH 4 Cl for 1, 3, 6, 18 or 72 h.Similarly, the NO 3 À induction assay and RNA-seq analysis were assessed by supplying 5 mM KCl or 5 mM KNO 3 for 2 h to seedlings starved of N for 4 days after 10-day pre-culture under the full-nutrient solution.Root tissues of maize plants in all experiments were collected for RNA-seq or qPCR.For phenotypic analysis at the seedling stage, maize seedlings were hydroponically cultivated under the modified Hoagland solution supplied either with 0.04 mM KNO 3 (LN) or with 4 mM KNO 3 (HN).Under LN, additional 3.96 mM KCl was provided to maintain the same level of potassium as that in HN.After 14 days, shoots and roots were sampled to assess the biomass accumulation and gene expression.
For organ-specific expression pattern analysis of ZmNRT1.1B and ZmNRT1.1C,0-20 cm underground roots, ear leaf (whole), ear stem (whole), tassels (whole), silk and husk were collected from adult field-grown B73 plants at silking stage.Total RNA was extracted with RNAiso Plus (Takara Biomedical Technology Co., Ltd., Beijing).Approximately 1.5 lg total RNA was cleaned and reversely transcribed into cDNA with gDNA Eraser of PrimeScript TM RT reagent Kit (Takara Biomedical Technology Co., Ltd., Beijing).Then qRT-PCR assays were performed with TB Greenâ Premix Ex Taq TM II (Takara Biomedical Technology Co., Ltd., Beijing) using CFX96 Real-Time PCR System (Bio-rad).The expression of ZmTUB4 in maize was used for normalizing respective gene expression.The gene-specific primers used in the qPCR analysis are listed in Data S1.

RNA-seq assays
Total RNA was extracted with RNAiso Plus (Takara Biomedical Technology Co., Ltd., Beijing) and RNA integrity was assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA).The mRNA libraries were sequenced on an Illumina Novaseq platform at Novogene (Beijing) and 150 bp paired-end reads were generated.The reads were mapped to the maize B73 reference genome (B73 RefGen_v4, AGPv4) using HISAT2 (v.2.0.5) with default parameters.Differential gene expression analysis of KNO 3 /KCl was performed using the DESeq2 R package (1.20.0).Corrected P-value ≤0.05 and log2 (fold-change) ≥1 were set to clarify genes that were significantly differentially expressed.The web server rKOBAS (http://kobas.cbi.pku.edu.cn/) was used to perform KEGG pathway analysis of DEGs in different clusters.
To generate transgenic maize lines overexpressing ZmNRT1.1B, the open reading frame (ORF) of ZmNRT1.1B was amplified using I-5 TM DNA polymerase (Tsingke Biotechnology Co., Ltd., Beijing).The amplified fragment was subsequently cloned into vector pCloneEZ-Blunt-Kana/HC cloning kit and sub-cloned into p1301 vector through the BamHI site.All primes used are listed in Data S1.

Protein subcellular localization assay
To investigate the subcellular localization of ZmNRT1.1B and ZmNRT1.1Cproteins, the coding sequences of ZmNRT1.1B and ZmNRT1.1Cwere cloned into pEZS-NL vector.Plasmids were extracted and purified using the Plasmid Midi Kit (Qiagen; No. 12143) following the manufacturer's manual.Leaves of etiolated B73 seedlings grown in dark were digested by Cellulase R10 (Yakult Honsha, Tokyo, Japan) and Macerozyme R10 (Yakult Honsha, Tokyo, Japan) for the preparation of mesophyll cell protoplasts.The empty vector and fusion constructs of ZmNRT1.1B-eGFP and ZmNRT1.1C-eGFPwere transformed into maize mesophyll cell protoplasts by the polyethylene glycol (PEG)-induced method.After 12 h incubation in the dark, protoplasts were harvested and eGFP fluorescence was inspected using a ZEISS710 confocal microscope.FM4-64 was used to counterstain the plasma membrane.Fluorescence signals for eGFP (excitation 488 nm, emission 505-550 nm) and FM4-64 (excitation 561 nm, emission >575 nm) were detected.
For the analysis of the nuclear-cytoplasmic shuttling of ZmNLP3.1 by NO 3 À and their dependence on ZmNRT1.1B, the coding region of ZmNLP3.1 was cloned into pEZS-NL vector and then transformed into mesophyll protoplasts of 13-day-old WT and zmnrt1.1b-1mutant.Transformed maize protoplasts were incubated for 12 h in the dark.For nuclear-cytoplasmic shuttling of ZmNLP3.1, the transformed maize protoplasts were treated with 5 mM KCl or 5 mM KNO 3 for 2 h.The nuclear dye DAPI (4 0 ,6diamidino-2-phenylin-dole) was used to counterstain the nucleus.
The GFP fluorescence was then detected using a confocal microscope (ZEISS710, Carl Zeiss).

N content assays
Maize shoot and root samples were collected from hydroponic and field experiments, respectively.They were then dried at 65 °C for 7 days and ground with ball mill.Approximately 0.3 g material was taken for N concentration determination with a modified Kjeldahl acid-digestion method (Nelson and Sommers, 1973).

NO 3
À uptake and translocation assay Maize seedlings were hydroponically grown under Hoagland nutrient solution supplied with 2 mM NH 4 NO 3 for 10 days and then subsequently deprived of N for 4 days, followed by resupply of 4 mM NO 3 À for 3 h.The plant roots were then rinsed in 1 mM CaSO 4 for 1 min, followed by transferring to a solution containing 0.2 mM or 5 mM 15 NO 3 À (20 atom% 15 N, Shanghai Research Institute of Chemical Industry) for 6 min 15 N influx.The roots were finally rinsed in 1 mM CaSO 4 for 1 min and then separated from the shoots immediately after the final wash in CaSO 4 .
For 15 N accumulation determination, maize seedlings were hydroponically cultivated inHoagland solution containing 4 mM KNO 3 for 6 days, and then transferred to a solution containing 4 mM 15 NO 3 À (20% atom 15 N, Shanghai Research Institute of Chemical Industry) for 0.25, 3 and 24 h, respectively.The plant roots were then rinsed in 1 mM CaSO 4 for 1 min.Roots and shoots were dried and then collected immediately.Samples were dried at 65 °C for 7 days.Measurement of 15 N content was performed by Delta V plus Isotope Mass Spectrometry (DELTA Plus XP, Thermo-Finnigan, Germany).Uptake activity was calculated as the amount of 15 N taken up per unit weight of roots per unit time.The ratio between 15 N in shoot and root was calculated as an indicator of the root-to-shoot NO 3 À translocation as described by Hu et al. (2015).

Field trials of maize
Field experiments growing zmnrt1.1bmutants were carried out in a completely randomized block design with a single row plot at two different locations in Beijing (N40°13 0 54 00 , E116°33 0 50 00 ) and Sanya (N18°23 0 23 00 , E109°11 For determining the potential contribution of ZmNRT1.1B and ZmNLP3.1 to N-dependent grain yield, maize plants were grown in two independent field blocks supplied with 0 or 120 kg N/ha.The study was performed in the CAU experimental station (Sanya, Hainan, China) from November 2022 to February -2023.The N was applied in the form of urea and fertilized with a proportion of 40% at the silking stage and 60% at the heading stage.The plants were grown in 32 rows917 plants for each plot with a 0.5 m distance between rows and replicated three times.Shoots of plants were harvested at silking and maturation stages, respectively.All plants were open-pollinated, three plants were measured in each row and taken the average of three as a plot replicate.
To evaluate the breeding value of ZmNRT1.1B,large-scale field tests for ZmNRT1.1B-OEplants in the Xianyu335 background were performed in the field under three nitrogen conditions: 0 kg N/ha, 158 kg N/ha, 225 kg N/ha from May 2018 to October 2020 at Gongzhuling (N43°31 0 11 00 , E124°47 0 41 00 ).In 2018, plants were grown in 4 rows 920 plants, and 20 plants were selected to analysis agronomic traits for each genotype.Another field test was also conducted using transgenic plants and Xianyu335 in 2020, with a planting density of 5 rows 920 plants.For each transformation event and Xianyu335, 50 plants were selected to analyse agronomic traits.The spacing between plants at 2 years was 25 cm (row space) 930 cm (plant space).For testing yield-related traits, mature ears were harvested and dried.Yield-related traits included ear length, ear diameter, hundred-kernel weight and grain yield per ear.

Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article.Figure S8 The phenotype of the WT and zmnrt1.1bmutants in the field.ZmNRT1.1Bimproves N use efficiency and yield in maize 329

Figure 1
Figure 1 Phylogenetic and collinearity analyses of NRT1.1s and N-dependent expression pattern of ZmNRT1.1B and ZmNRT1.1C.(a) Phylogenetic relationship of NRT1.1s in Zea mays, Oryza sativa, Sorghum bicolor, Brachypodium distachyon and Arabidopsis thaliana.The phylogenetic tree was constructed using MEGA7.(b) Collinearity analysis of NRT1.1 s in maize and rice.WGD, whole-genome duplication; TD, tandem duplication.(c), (d) Transcriptional responses of ZmNRT1.1B(c) and ZmNRT1.1C(d) to different N availabilities.(e), (f) Transcript levels of ZmNRT1.1B(e) and ZmNRT1.1C(f) in response to nitrogen deprivation.Maize seedlings were pre-cultured with 2 mM NH 4 NO 3 (NN) for 10 days and then transferred to N-free nutrient solution (-N) for 24, 48 or 96 h.(g), (h) Transcript levels of ZmNRT1.1B(g) and ZmNRT1.1C(h) in response to resupply of nitrate or ammonium for 1, 3, 6, 18 or 72 h of B73 roots.After N starvation for 4 days, maize seedlings were resupplied with either 4 mM KNO 3 or NH 4 Cl.Bars represent means AE SD (n = 3 independent biological replicates).ZmTUB4 was used as the internal reference.Different letters indicate significant differences at P < 0.05 according to one-way ANOVA followed by Duncan's multiple comparison test.

Figure 3
Figure 3 ZmNRT1.1Bmodulates NO 3 À uptake, root-to-shoot translocation and expression of NO 3 À -responsive genes.(a) and (b) Root influx of 15 Nlabelled NO 3 À supplied at the concentration of 0.2 mM to assess the high-affinity transport system (HATS) (a) or at 5 mM for the low-affinity transport system (LATS) (b) of WT and zmnrt1.1bmutants.Maize plants were pre-cultured hydroponically under 2 mM NH 4 NO 3 for 10 days before transferring to N-free nutrient solution for 4 days (-N).N-deficient plants were then resupplied with 4 mM KNO 3 (+NO 3 À ) for 3 h.(c) Nitrate translocation in WT and zmnrt1.1bplants.Ratio of shoot/root 15 NO 3 À contents of different plants in the WT and zmnrt1.1bmutants assessed at 15 N-nitrate addition for 0.25, 3 or 24 h.(d) Transcript levels of nitrate assimilation genes ZmNR1.1 and ZmNR1.2 in WT and zmnrt1.1bmutants under 0.04 (LN) or 4 mM KNO 3 (HN) for 14 days.Relative gene expression of WT grown under low NO 3 À condition was normalized to 1. (e) Nitrate induction levels of ZmNR1.1,ZmNR1.2 andZmNRT2.1 transcripts in WT and zmnrt1.1bmutants.Ten-day-old hydroponically grown maize seedlings were subjected to N starvation for 4 days, and then resupplied with 5 mM NO 3 À for 3 h.Bars represent mean AE SD (n ≥ 3 independent biological replicates).Different letters indicate significant differences at P < 0.05 according to one-way ANOVA followed by Duncan's multiple comparison test.

Figure 5
Figure 5 Loss-of-function of ZmNLP3.1 weakens plant growth.(a) Growth performance of WT and zmnlp3.1 mutants under low-or high-NO 3 À conditions.Scale bar, 20 cm.(b), (c) Dry biomass (b) and N contents (c) of WT and zmnlp3.1 mutants under 0.04 mM (LN) or 4 mM KNO 3 (HN) for 14 days.Data represent mean AE SD (n = 6 independent biological replicates).Different letters indicate significant differences at P < 0.05 according to one-way ANOVA followed by Duncan's multiple comparison test.

Figure 6
Figure 6 Loss of function of ZmNRT1.1B and ZmNLP3.1 decreases the grain yield in the field.(a), (b) Aboveground biomass at the different developmental stages (a) and yield per plant (b) of the WT and zmnrt1.1bsgrown under LN or HN conditions.(c), (d) Aboveground biomass at the different developmental stages (c) and yield per plant (d) of the WT and zmnlp3.1sgrown under LN or HN conditions.The plants were grown in the field plot in Sanya experimental station (Hainan) in 2022.Data represent mean AE SD (n = 3 independent biological replicates).Different letters indicate significant differences at P < 0.05 according to one-way ANOVA followed by Duncan's multiple comparison test.

Figure 7
Figure 7 Overexpressing ZmNRT1.1Bimproves grain yield in maize.(a) Genetic construction of transgenic materials used for hydroponic trials, 2018 field trials and 2020 field trials.(b) Yield per plant of TC WT and TC OE lines under low nitrogen (LN), moderate nitrogen (MN) or high nitrogen (HN) condition in the field trial in Changchun (2018).Data represent mean AE SD (n = 20 independent biological replicates).TC OE and TC WT represent ZmNRT1.1Boverexpression lines and corresponding control lines segregating from the same transgenic events.(b) Yield per plant of ZmNRT1.1Boverexpressed in the XY335 commercial hybrid background under LN, MN or HN condition in the field trial in Changchun (2020).Data represent mean AE SD (n = 50 independent biological replicates).Statistical significance was determined by a two-sided t-test.XY335 OE represents homozygous XY335-overexpressing ZmNRT1.1Blines.
Figure S9 Growth phenotypes of ZmNRT1.1Boverexpression lines under different NO 3 À concentrations.ª 2023 The Authors.Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 22, 316-329 Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 22, 316-329 ª 2023 The Authors.
The maize loss-of-function mutants in the background of ND101 were obtained from the Center for Crop Functional Genomics and Molecular Breeding of China Agricultural University, with the accession number CAUC0713, CAUC0714 and CAUC1783.The construct was transformed into inbred line ND101.The ZmNRT1.1B overexpression lines were selected and provided by the Institute of Agricultural Biotechnology, Jilin Academy of Agricultural Sciences.Primers used for isolating CRISPR/CAS9 mutants and overexpression lines are listed in Data S1.