Transcription factor ZmEREB97 regulates nitrate uptake in maize (Zea mays) roots

Abstract Maize (Zea mays L.) has very strong requirements for nitrogen. However, the molecular mechanisms underlying the regulations of nitrogen uptake and translocation in this species are not fully understood. Here, we report that an APETALA2/ETHYLENE RESPONSE FACTOR (AP2/ERF) transcription factor ZmEREB97 functions as an important regulator in the N signaling network in maize. Predominantly expressed and accumulated in main root and lateral root primordia, ZmEREB97 rapidly responded to nitrate treatment. By overlapping the analyses of differentially expressed genes and conducting a DAP-seq assay, we identified 1,446 potential target genes of ZmEREB97. Among these, 764 genes were coregulated in 2 lines of zmereb97 mutants. Loss of function of ZmEREB97 substantially weakened plant growth under both hydroponic and soil conditions. Physiological characterization of zmereb97 mutant plants demonstrated that reduced biomass and grain yield were both associated with reduced nitrate influx, decreased nitrate content, and less N accumulation. We further demonstrated that ZmEREB97 directly targets and regulates the expression of 6 ZmNRT genes by binding to the GCC-box-related sequences in gene promoters. Collectively, these data suggest that ZmEREB97 is a major positive regulator of the nitrate response and that it plays an important role in optimizing nitrate uptake, offering a target for improvement of nitrogen use efficiency in crops.


Introduction
Nitrogen (N) is an essential macronutrient for plants, controlling almost every aspect of plant growth and development.Owing to the wide abundance of a synthetic N application in farmland, the yield of crops was dramatically increased in the past 50 yr (Han et al. 2015).This increase, however, comes with 2 caveats.The first one is that modern crop varieties become less and less responsive to additional N supply (Fischer et al. 2009).The second concern is that excessive use of nitrogenous fertilizers is expensive and leads to considerable negative impacts on the ecosystems due to eutrophication (Vitousek et al. 2009;Good and Beatty 2011).
New solutions are therefore needed to simultaneously increase yield while maintaining, or preferably decreasing, the amounts of applied N, to reduce the environmental footprint and maximize N use efficiency (NUE) of crops.
Maize (Zea mays L.) is the crop species with the highest global annual yield (Ort and Long 2014;Hochholdinger et al. 2017).N is a major contributor to the high yield of maize, prompting high demand for N fertilizers.Meanwhile, balancing sufficient N supply for maize yield and preventing leaching of N fertilizers in natural and agricultural ecosystems are highly challenging (Robertson and Vitousek 2009).An increased understanding of how maize plants respond to N supply and dissecting the downstream components of N signaling pathways is essential in the refinement of strategies to improve NUE.
Transcription factors (TFs) often act as master switches in plant regulatory networks (Xuan et al. 2017;Plett et al. 2018).ANR1 is an early discovered TF known to regulate systemic nitrate repression and participate in the signaling pathway of NRT1.1 in Arabidopsis (Zhang and Forde 1998;Gan et al. 2005).Other TFs were shown in controlling primary nitrate-responsive genes such as the NRT genes (NRT1.1,NRT2.1, NRT2.2), the nitrate reductase (NR) genes (NIA1, NIA2), and the nitrite reductase gene (NIR).This includes NLP6 (Konishi and Yanagisawa 2013), NLP7 (Castaings et al. 2009;Marchive et al. 2013), LBD37/38/39 (Rubin et al. 2009), TCP20 (Guan et al. 2014(Guan et al. , 2017)), SPL9 (Krouk et al. 2010), and NIGT1/HRS1s (Kiba et al. 2018).In Arabidopsis, these TFs bind to the promoter of NRT1.1, NRT2.1, NRT2.2, NIA1, or NIA2 genes and regulate their expression (Vidal et al. 2015).Recently, key TFs such as OsTCP19, OsNGR5, OsNAC42, and OsGRF4 were reported in rice.OsTCP19 can modulate the expression of N-responsive genes, and overexpression of OsTCP19 gene confers a higher N absorption ability (Liu et al. 2021).An increase in the transcription level of a TF gene, OsNGR5, results in an increase in rice yield and NUE for the same level of N supply (Wu et al. 2020).A rice N transporter OsNPF6.1 HapB , trans-activated by a NUE-related TF OsNAC42, enhances nitrate uptake (Tang et al. 2019).OsGRF4 promotes and integrates N assimilation, carbon fixation, and growth (Li et al. 2018).In maize, ZmNLP5 was found to be one of the key TFs in the molecular network for mediating N signaling and metabolism.A natural loss-of-function allele of ZmNLP5 conferred less N accumulation in the ear leaves and seed kernels resembling (Ge et al. 2020).However, the molecular mechanisms underlying transcriptional regulation of N uptake in maize remain largely unknown.
Here, using the combined approaches of gene coexpression network analysis, promoter sequence motif prediction, and yeast one hybrid (Y1H) screening, we identified a TF, ZmEREB97, that belongs to the AP2/ERF family.The AP2/ERF family is one of the largest plant TF families that consist of many subfamilies such as AP2, ERF (also known as ethylene-responsive element binding proteins, EREBPs), dehydration-responsive element binding protein (DREB), and RAV (related to ABI3/VP1) (Chandler 2018).Members of AP2/ERF TFs have emerged as regulators for plant developmental and abiotic stress responses (Licausi et al. 2013).The characteristic DNA-binding domain of AP2/ERF, which consists of a sheet with 3 β-strands followed by an α-helix, recognizes specific cis-regulatory elements, including the GCC-box (Shoji and Yuan 2021).The combined physiological, molecular, and genetic analyses indicate that ZmEREB97 concomitantly mediates N influx by regulating several NRT genes.The loss of ZmEREB97 gene resulted in downregulation of nitrate uptake and accumulation suggesting its critical role in optimizing N acquisition and utilization.

Coexpression network analysis with WGCNA
To predict the conserved motifs in nitrate-related genes, a gene coexpression network was performed using 2 representative inbred maize lines B73 (a stiff stalk [SS] inbred) and Mo17 (a non-stiff stalk [NSS] inbred; Mikel and Dudley 2006).To identify the nitrate sensitivity between B73 and Mo17, we designed a "recovery test" under 2 different N conditions.As shown in Supplementary Fig. S1, both B73 and Mo17 showed symptoms of nitrate starvation under deficient nitrate conditions, with no significant differences observed in Soil Plant Analysis Development (SPAD) values, nitrate content, and total N content.However, after being supplied with sufficient N (10 mM KNO 3 ) for 10 d, B73 exhibited a substantial increase in SPAD values (88.1%), nitrate content (72.2%), and total N content (90.4%), which were significantly higher than the corresponding values in Mo17 (53.7%, 65.4%, and 81.7%, respectively).These results suggest that B73 has a faster recovery rate from N deficiency as compared to Mo17, indicating that it is a more nitrate-sensitive inbred maize line.
Firstly, MapMan software was used to visualize the biological processes affected by nitrate treatment (Thimm et al. 2004).As shown in Supplementary Fig. S2 and Data Set 1, notable changes were observed in many metabolic processes in root tissues of B73 and Mo17 treated with 15 mM KNO 3 at 0.5-and 8-h timepoints, as compared with 15 mM KCl.These included minor CHO, light reactions, photorespiration, TCA and OPP cycles, and amino acid and N metabolism.In addition, carbon and carbohydrate metabolisms were also affected by nitrate treatment in this study, suggesting their close relationship with N metabolism.
We then applied the WGCNA on RNA sequencing (RNA-Seq) data of 36 root tissue samples to construct the consensus network of B73 and Mo17.As a result, 26 modules were identified and labeled by different colors in the hierarchical clustering dendrogram (Fig. 1A), with each module containing between 79 and 3,434 genes (Supplementary Data Set S1).By relating consensus modules to B73 set-specific modules, it was shown that the module structure of the B73 expression data was slightly different from the Mo17 expression data (Supplementary Figs.S3 and S4).To determine if any of the genes in the 26 modules was associated with nitrate treatment time, we tested the association of module eigengenes (the first principal component of a given module that can be considered as a representative of the gene expression profiles in a module) with the nitrate treatment time.First, the relationships between modules and nitrate treatment time were calculated in the B73 and Mo17 datasets, respectively.Then, the consensus eigengene networks in the B73 and Mo17 datasets were compared, and 4 consensus modules "dark orange," "gray," "blue," and "light cyan" were identified which showed high correlation to the time points after nitrate supply (corresponding correlation > 0.7, P < 0.01; Fig. 1B, Supplementary Data Set 1).The module membership (MM; the correlation of the module eigengene and the gene expression profile) and gene significance (GS; quantifies an association of individual genes with nitrate treatment time points) plots for the 4 modules showed that MM and GS are highly correlated, indicating that genes significantly associated with nitrate treatment time were also the most important elements in the modules (Supplementary Fig. S3 and Data Set 1).
Finally, gene ontology (GO) enrichment analysis on the 4 modules and the top 10 most enriched biological processes in "dark orange", "gray", "blue," and "light cyan" modules were conducted.The results showed that N metabolic processes were enriched in the "blue" and "light cyan" modules; photoperiodic processes were enriched in the "light cyan" module; carboxylic acid metabolic/biosynthetic processes and hormone biosynthetic process were enriched in the "gray" module; and amide metabolic/biosynthetic processes were enriched in the "blue" module (Fig. 1, C to F, Supplementary Data Set 1).Network visualization using the set of genes from these biological processes in the 4 modules showed that these genes were highly connected, and each gene had more than 2 edges to interact with other genes (Fig. 1G, Supplementary Data Set 1).
Collectively, these results show that 4 consensus modules that are conserved in 2 different inbred lines B73 and Mo17 are closely related to N. Exploration of the genes with similar expression patterns in the 4 modules and mining their conserved regulatory elements may shed light on more knowledge of maize response to N supplementation.

GCC-box is enriched in the promoter region of genes in the 4 N-related modules
The putative promoter sequences (1000-bp upstream and 100-bp downstream of the start codon) of the genes in each module were subjected to MEME Suite program for de novo motif discovery.A total of 6 motifs obtained from the 4 modules (E < 0.05) can be classified into 2 types according to their matrix (Fig. 2A, Supplementary Data Set 2), and 4 of them were significantly enriched in the 4 modules (P < 1e-4; Fig. 2B).The results showed that the 6 motifs exhibit a significant positional bias within the range of 200-to 0-bp upstream of the start codon in the promoter regions of genes from the 4 modules and 500 random genes.Furthermore, the frequency and density of the 6 motif occurrences were notably higher in the promoter regions of genes within the selected modules compared to those in the random genes (Fig. 2C).These findings suggested that these 6 motifs were unique to these 4 modules.To predict the potential TFs that could recognize the motifs, the 6 predicted motifs were compared with the known cis-elements in plant database, including AthaMap, AGRIS, and PLACE (Mahony and Benos 2007;Yilmaz et al. 2011).Several motifs shared high similarities (E < 0.05) with the binding sites of TF families including ARF, MYB, and SPL (Supplementary Data Set 2) that have been previously reported as key regulators in modulating N response in Arabidopsis (Gutierrez et al. 2008;Krouk et al. 2010).In addition, we found that Motif 1, the highest enriched motif within the 4 modules, contained a GCC-box (Fig. 2B, Supplementary Data Set 2) which is a core sequence in the binding site recognized by AP2/ERF TFs (Fujimoto et al. 2000).Plant and yeast promoter databases were used to further investigate Motif 1.The results showed that Motif 1 was significantly associated with the N metabolism processes in both databases (Supplementary Data Set 2).Thus, genes containing Motif 1 in their promoter regions may be regulated by an AP2/ERF family member.

An AP2/ERF TF was identified to interact with Motif 1
To explore the potential AP2/ERF family TFs binding to Motif 1, Y1H assay was performed.To select the sequence matching Motif 1 matrix for Y1H screening, we first chose the genes with more (>60; Supplementary Data Set 2) occurrences of Motif 1 in their promoter regions from the 4 modules; 63 genes were then retained.Considering the position bias and genes more related to N, we then chose the best-matching sequence to Motif 1 from a NRT (GRMZM2G124396, −55 to 20 bp from start codon, 10 occurrences in this region; Supplementary Data Set 2).As a result, a region from −55 to −1 bp of the start codon from this sequence, namely, Motif 1, was chosen for Y1H screening.
The 55-bp sequence was artificially synthesized 4 times and used for the Y1H screening.Among the 19 positive clones, 6 TFs belonging to the AP2/ERF family appeared 14 times during screening, and ZmEREB97 (GRMZM2G068967, which appears as high as 5-fold) was chosen for further analysis (Supplementary Data Set 2).The sequence of ZmEREB97 was located on maize B73_v3 genome chr2, 10794242-10793574, and the full-length protein contained 222 amino acids, with the AP2/ERF domain from residues 18 to 75.The theoretical molecular weight of ZmEREB97 was 22.9 kDa, and the pI was 9.73.To confirm the interaction in the yeast, we retransformed the 4×Motif 1 bait strain with a fulllength ZmEREB97-GAL4AD plasmid.As shown in Fig. 2D, ZmEREB97 can interact with the Motif 1 in vitro.Electrophoretic mobility shift assays (EMSAs) were then performed to investigate the binding specificity of ZmEREB97 using a sequence with Y1H assay.Yeast harboring the 4×Motif 1-HIS3 reporter gene was transformed with expression vectors for the GAL4 activation domain fused to the ZmEREB97 and ZmEREB83 (control) and grown on control (-Trp/-Leu) and selection medium (-His/-Trp/-Leu + 60 mM 3-AT).E) EMSA with 6×His-ZmEREB97 fusion protein using Motif 1 as a probe.Nonlabeled WT and mutated Motif 1 were used as competitor.F) Transcriptional activity assays of ZmEREB97.The 35S::REN-4×Motif 1::LUC reporter construct was transiently expressed in N. benthamiana leaves together with the negative control vector and the 35S::ZmEREB97 effector vector.The reporter construct contains 2 reporter genes, the LUC and the REN.4×Motif 1 activities are given as fLUC/rLUC activity ratios.The bars represent means ± SD (n = 3).Two-tailed Student's t test (n = 3 collective samples each containing 3 biological replicates) was used to test the significance.**P < 0.01.mutation in the GCC-like sites as the nonlabeled mutated probe.The EMSA result showed that ZmEREB97 can bind to the wild-type (WT) sequence of Motif 1 but not to the nonlabeled mutated sequence (Fig. 2E).
To test whether ZmEREB97 has a transcriptional activator or a repressor activity, we conducted a dual-luciferase transient transcriptional activity assay.The construct 35S::ZmEREB97, with ZmEREB97 ORF, was generated as an effector.The reporter construct contains 2 luciferase cassettes, the Renilla luciferase gene (REN) driven by the cauliflower mosaic virus (CaMV) with the 35S promoter (35S::REN) used as an internal control and the firefly luciferase gene (LUC) driven by the 4×motif that was used as a reporter (Fig. 2F).Coexpression of 4×motif1::LUC with 35S:: ZmEREB97 resulted in a 3.1-fold (t test, P ≤ 0.01) increase in LUC activity compared with the control (Supplementary Fig. S5, Fig. 2F), implying that ZmEREB97 acted as a transcriptional activator.

Localization and expression of ZmEREB97 in maize roots
To clarify the tissue localization of ZmEREB97 in maize root, the spatial distribution of its mRNA in the roots was examined by RNA in situ hybridization assay (Fig. 3A).ZmEREB97 transcripts were mainly expressed in the main root, especially in cell elongation and division zones (Fig. 3A, II to V), and in the primordial lateral root (Fig. 3A, VI to VII).
The biological function of TFs depends highly on their subcellular localization.Therefore, the full-length CDS of ZmEREB97 driven by CaMV 35S promoter was fused with GFP at the C-terminal.The 35S:ZmEREB97-GFP construct was transiently transformed into maize mesophyll protoplast.As shown in Fig. 3B, the green fluorescence of ZmEREB97-GFP was detected in the nucleus.This localization was confirmed by staining nucleus using the red fluorescent probe and by merging both pictures (Fig. 3B), indicating that ZmEREB97 was located in the nucleus.Western blot assay of ZmEREB97 in roots of maize seedlings also showed that ZmEREB97 was mainly expressed in the nucleus (Fig. 3, C and D).
To test whether the expression of ZmEREB97 is involved in nitrate response, qPCR and western blot analysis were performed.The results of Fig. 4, A and B, showed that mRNA and newly synthesized protein of ZmEREB97 were rapidly accumulated under the nitrate treatment in 5 min and gradually decreased from 60 to 120 min.By contrast, when treating maize roots with cycloheximide (CHX), a chemical reagent that can inhibit protein synthesis (Gilkerson et al. 2016), the accumulation of ZmEREB97 mRNAs was increased, suggesting that de novo protein synthesis may not be required for ZmEREB97 expression under nitrate treatment.

ZmEREB97 binds to the promoter of genes related to N metabolism
To further elucidate genes directly targeted by ZmEREB97, a DAP-seq assay was performed using DNA from inbred maize lines B73 root tissues.ZmEREB97-binding sites were identified from consistent peaks generated by 2 biological replications.With a cutoff q-value of 0.05 and the genic regions defined as 1-kb upstream from the transcription start site (TSS) to 1-kb downstream from the transcription termination site (TTS), a total of 11,776 peaks distributed on 10,702 gene regions were identified.Among all the detected peaks, more than half (63.0%) of the peaks were located in the intergenic regions.Of the remaining genic region peaks, most of them were located within either 1-kb upstream regions (11.7% of all peaks) or 5′-untranslated regions (UTRs, 9.3%; Fig. 5A).
By overlapping the 6,069 differentially expressed genes (DEGs) from B73 (Supplementary Fig. S6 and Data Set 3) and the putative targets identified by DAP-seq assay, we found 1,446 DEGs that were directly targeted by ZmEREB97 (Fig. 5B, Supplementary Data Set 3), including 130 genes containing Motif 1 in the promoter regions from the 4 modules (Supplementary Figs.S7 and S8).In total, 2,555 peaks were detected locating in the genic regions of the 1,446 DEGs and of these peaks, of which 840 (32.9%) located in the 5′-UTR and 488 (19.1%) in the 1-kb region upstream from the TSS (Fig. 5C).In agreement with the predicted motif position bias (Fig. 2C), the binding sites are enriched in the vicinity of the start codon sites of nitrate-responsive genes (Fig. 5D).In addition, MEME-ChIP method was applied to analyze the enriched motifs, using sequences from 250-bp upstream to 250-bp downstream of the genomic regions where 2,555 peak points were located.Consequently, a GCC-box-like element as the topscoring motif (E = 2.2e-195, 52.5% enrichment) was identified (Fig. 5E).By examining the density plot of this motif with the 2,555 peaks, this motif was enriched around the peak point (Fig. 5F).

The loss of function of ZmEREB97 regulates nitrate uptake and accumulation in maize seedlings
To better characterize the function of ZmEREB97, we obtained 2 different kinds of mutants for further analysis.A Mu insertion was found in the promoter of the ZmEREB97 gene (Supplementary Fig. S9, A and B).A knockout mutant using the CRISPR-Cas9 system in the genetic background of KN5585 was generated (Supplementary Fig. S9, C and D).qPCR revealed that the expression of ZmEREB97 was significantly reduced in zmereb97-MU and zmereb97-CRI mutant seedlings (Supplementary Fig. S9E).
We first performed RNA-seq with the WT and zmereb97-MU and zmereb97-CRI roots in 5-mM KNO 3 treatment and analyzed 1,446 DEGs targeted by ZmEREB97.Among these 1,446 DEGs, 586 and 603 genes were found to be downregulated in zmereb97-MU and zmereb97-CRI , respectively, as compared with WTs.At the same time, 733 and 708 genes were found to be upregulated.An important observation was made regarding the overlap of downregulated and upregulated genes in both zmereb97-MU and zmereb97-CRI .Specifically, 359 genes were found to be downregulated, while 405 genes were upregulated (Supplementary Fig. S10).This finding suggests that the defect in ZmEREB97 had a significant impact on nitrate metabolism.
The mutants and WT were then hydroponically cultured under high-NO 3 − (5-mM KNO 3 ) and low-NO 3 − (0.5-mM KNO 3 ) conditions for 20 d to measure the morphological and physiological phenotypes.
As in Fig. 6, A and B, ZmERER97 mutation caused a notable shoot and root growth inhibition regardless of high or low KNO 3 .Consistently, the shoot and root dry weights of zmereb97-MU and zmereb97-CRI mutants were significantly lower than those of the WTs (Fig. 6, C and D).The root/shoot ratio exhibited similar responses in WT and mutants under high-or low-KNO 3 conditions.In the presence of 5-mM KNO 3 or 0.5-mM KNO 3 condition, the loss of ZmEREB97 function all led to an increase in the root/shoot ratio (Fig. 6E).Compared with WTs, the SPAD value of ZmEREB97 mutants was significantly decreased regardless of the 5-mM KNO 3 or 0.5-mM KNO 3 condition (Fig. 6F).In addition, we tested NO 3 − content and total N accumulation in these lines.The zmereb97-MU and zmereb97-CRI mutant plants took up less NO 3 − and showed lower total N accumulation in the roots and shoots as compared with their responsive WT (Fig. 6, G to J). Together, these data confirm that loss of function of ZmEREB97 impairs N uptake and accumulation and subsequently affects plant growth.
To further evaluate the potential contribution of ZmEREB97 to N-dependent yield formation, we planted WT and zmereb97-MU and zmereb97-CRI mutants under soil condition.As shown in Fig. 6, L and M, the aboveground biomass of zmereb97-MU and zmereb97-CRI was greatly reduced as compared with the responsive  WT.For yield formation, zmereb97-MU and zmereb97-CRI stably reduced total N accumulation in leaves by ∼15.1% and ∼11.9%, resulting in a decrease in grain yield per plant of ∼13.6% and ∼15.4% compared to responsive WT (Fig. 6, K, N, and O).These findings indicate that loss of function of ZmEREB97 results in reduction in plant growth and grain productivity.
We then studied the difference in kinetics of NO 3 − uptake of roots between WT and zmereb97 mutants.Addition of 10 or 1 mM KNO 3 to the bath solution resulted in a massive and transient net NO 3 − uptake, with net NO 3 − influx being lower in zmereb97-MU and zmereb97-CRI root compared with that in WT (Fig. 7, A and B).When roots were incubated in working solution with 10 mM KNO 3 for ZmEREB97 regulates nitrate uptake | 541 40 min, the substantially lower initial NO 3 − influx was also observed in the root of zmereb97 mutants (inserts in Fig. 7A); the same trends were observed for 1-mM KNO 3 treatment (inserts in Fig. 7B).Using a 15 N-abundance assay, nitrate influx rates in root and shoot of zmereb97-MU and zmereb97-CRI lines were shown to be lower than in the WT-MU and WT-CRI (Fig. 7, C and D), further indicating that ZmEREB97 may be beneficial for enhancing NO 3 − uptake.
NR is the first enzyme of the N reduction pathway in plants, serving as a bridge connecting nitrate uptake and assimilation.NR is usually considered a marker gene, and its expression and activity can reflect the nitrate level and metabolism in plants (Berger et al. 2020;Cao et al. 2023).Here, we found that the expression of ZmNR1.1 and ZmNR1.2 were greatly repressed in zmereb97-MU and zmereb97-CRI under low-and high-N conditions, respectively (Supplementary Fig. S11, A and B).Consistent with the qPCR results, NR enzyme activity analysis confirmed that the activity of NR was significantly inhibited in zmereb97-MU and zmereb97-CRI mutants compared with respective WT (Supplementary Fig. S11C).We further assessed the expression of another 2 NO 3 − response marker genes including ZmGS1 and ZmGS2 to investigate whether ZmEREB97 is required for NO 3 − signaling.Under 5-mM or 0.5-mM KNO 3 treatments, the transcriptional induction of ZmGS1 and ZmGS2 in roots of both zmereb97-MU and zmereb97-CRI mutants were largely reduced in comparison with their respective WT (Supplementary Fig. S11, D and E), suggesting that ZmEREB97 is functionally relevant for NO 3 − signaling.In support, we also observed that the loss of function of ZmEREB97 causes significant reduction in the GS activity (Supplementary Fig. S11F).These results indicate that nitrate reduction could be significantly influenced by the impact of ZmEREB97 on nitrate uptake.
To verify the interaction determined by Y1H, we performed transcript-level analyses in maize roots supplied with 5 or 0.5 mM KNO 3 .Compared with the WT, ZmNRT1.1Atranscripts were significantly decreased in the zmereb97-MU and zmereb97-CRI mutants regardless of nitrate concentration (Fig. 9A).Decreases in transcript levels were also found for 2 additional ZmNRT1 genes (ZmNRT1.1B,ZmNRT1.2) in the mutants (Fig. 9, B and C).Among the 3 genes belonging to the ZmNRT2 family, the expressions of ZmNRT2.1 and ZmNRT2.5 in roots were significantly inhibited in the zmereb97 mutants under the 0.5-mM KNO 3 condition, whereas the ZmNRT2.5 was not obviously affected (Fig. 9, E to G). Besides, we observed decreased expression of ZmNRT3.1A in zmereb97 mutants in comparison with WT (Fig. 9H).Collectively, 6 genes of ZmNRT1.1A,ZmNRT1.1B,ZmNRT1.2,ZmNRT2.1,ZmNRT2.5, and ZmNRT3.1Awere shown to be strongly affected when the function of ZmERER97 was lost.
To further confirm the above interaction, a transient transcription dual-luciferase assay was performed in Nicotiana benthamiana leaves, using ZmEREB97 driven by the CaMV 35S promoter as an effector and LUC (the coding region of firefly luciferase) driven by the different gene promoters as reporters (Fig. 10A).As shown in Fig. 10, B to G, coexpression of NRT gene promoters::LUC with − influx of roots and shoots were then measured using 0.5 mM K 15 NO 3 for 60 and 120 min.The bars represent means ± SD (n = 6).Lowercase letters indicate significant differences at P < 0.01 according to Student's t test.
The family of AP2/ERF TFs has been shown to recognize and bind to GCC-box element (Shoji and Yuan 2021).Since the GCC-box-like element was identified as the top-scoring motif in our experiments (Fig. 5E), it prompted us to examine whether ZmEREB97 binds to the GCC-box in the promoters of NRT genes.The EMSA showed ZmERER97 could bind in vitro to probes containing GCC-box in each promoter of these genes, and the binding activity was partially abolished by the addition of 50-fold cold competitor probes (Fig. 10, H to M).

Discussion
TFs control diverse biological processes by binding to the promoters of target genes, and a number of TFs have been identified that control N assimilation (Vidal et al. 2015).For example, the overexpression of a member of the AP2/ERF, OsDREB1C, not only boosts grain yields but also confers high NUE in rice (Wei et al. 2022).Maize is an essential dual-use food and energy crop predominantly cultivated in aerobic soils where nitrate is often the primary source of N available for growth (Wen et al. 2017).A better understanding of the regulatory system of maize to maximize nitrate uptake and utilization is therefore a critical step toward deciphering the molecular mechanism of nitrate use in maize.Here, we also identified an AP2/ERF TF, ZmEREB97, and reported its functional characterization in modulating the nitrate response and uptake, suggesting that it was a potential candidate for improving NUE in maize.

ZmEREB97 is a unique AP2/ERF gene modulating nitrate uptake
The AP2/ERF superfamily is one of the largest groups of TFs in plants and plays important roles in the abiotic stress response and developmental processes (Licausi et al. 2013).Up to now, this family was also mainly implicated in maize developmental processes and response to stress conditions (Zhuang et al. 2010).Here, we show that a TF from AP2/ERF family, ZmEREB97, plays a critical role in mediating plant responses to nitrate availability.Several lines of evidence support this claim.First, the expression of ZmEREB97 can be rapidly induced by nitrate in root (Fig. 4), confirming its potential role in the nitrate response.Second, RNA in situ hybridization assay indicated that ZmEREB97 was preferentially expressed in the main root, particularly in cell elongation and division zones, as well as in the primordial lateral root (Fig. 3A), suggesting that ZmEREB97 may exert function in nitrate uptake.Third, DAP-seq assay showed that ZmEREB97 could bind to 130 genes containing Motif 1 in the promoter regions from the 4 N response modules (Fig. 2, D and E, Supplementary Fig. S8).Motif 1, which contains the GCC element, has been identified as the most enriched motif among the 4 N response modules (Fig. 2, A to C).These data strongly suggest that ZmEREB97 is an essential regulator in nitrate-responsive network in maize.
As N is a key factor limiting plant growth and crop productivity, the loss of function of N regulatory TFs is often accompanied by plant phenotype changes.Some established regulators of N status such as OsDREB1C, OsTCP19, AtNLP7, AtTCP20, ZmNLP5, and ZmM28 were shown to exert limitation to plant growth (Marchive et al. 2013;Guan et al. 2017;Wen et al. 2017;Wu et al. 2019;Ge et al. 2020;Liu et al. 2021).Consistent with previous reports, zmereb97 mutants exhibited a weakened phenotype under both sufficient and limiting nitrate conditions (Fig. 6, A, B, and  L), highlighting the crucial role of ZmEREB97 as a key regulator of N response network.In the course of this work, ZmEREB97 mutation in maize plants resulted in substantial shoot biomass reductions under both hydroponic (Fig. 6C) and soil conditions (Fig. 6M), and these biomass decreases were accompanied by a less grain yield (Fig. 6O).Plants respond to N availability by changing their root to shoot ratios.One hypothesis used to explain this allocation is that plants optimize their behavior by maximizing their relative growth rate (Agren and Franklin 2003).The zmereb97 mutants displayed higher root to shoot ratio under sufficient and limiting nitrate conditions (Fig. 6E).Given the significant biomass reduction caused by ZmEREB97 mutation, we propose that ZmEREB97 may act as a positive regulator in influencing the efficiency of nitrate acquisition from maize root.
We next present physiological evidence in support of the assumption that ZmEREB97 confers promoted nitrate acquisition.Analysis of nitrate and total N content in both roots and shoots revealed a significant decrease in NO 3 − and total N accumulation in 2 zmereb97 mutant lines compared to WT plants (Fig. 6, G to J and N).The lower nitrate and N accumulation in mutation lines may be attributed to the inhibited influx of NO 3 − in root tissues.The following results demonstrated that the net NO 3 − influx and steady NO 3 − influx in mature root surface were remarkably reduced in zmereb97 mutants (Fig. 7, A and B).Likewise, the lower 15 NO 3 − influx was also observed in root and shoot of zmerer97-MU and zmerer97-CRI lines (Fig. 7, C and D).Here, we established that the function of ZmEREB97, which operates maize growth, is significantly associated with its ability to regulate nitrate uptake and accumulation.

ZmEREB97 is involved in nitrate transport through regulating ZmNRT gene expression
Changes in nitrate assimilation are accompanied by changes in transcription (Gaudinier et al. 2018).Given that the ZmEREB97 has a positive effect on nitrate uptake, it is expected that it may have multiple downstream targets and distinct translational regulation.NRTs play a crucial and extensive role in taking up nitrate from the soil and transporting it to different organs for plant growth and development (Glass 2003).In maize, a total of 78 NRT1, 7 NRT2, and 2 NRT3 genes were identified (Jia et al. 2023).
In this study, 17 subgroups of key ZmNRT genes including 10 ZmNRT1, 5 ZmNRT2, and 2 ZmNRT3 genes were selected to verify their potential interaction with ZmEREB97.Based on the results of Y1H assay, 9 ZmNRT genes associated with nitrate transport were identified to interact with ZmEREB97 (Fig. 8).qPCR analysis of these 9 genes demonstrated that the expression of 6 genes of ZmNRT1.1A,ZmNRT1.1B,ZmNRT1.2,ZmNRT2.1,ZmNRT2.5, and ZmNRT3.1Awere significantly downregulated in zmereb97 mutants as compared with WT (Fig. 9).Within the ZmNRT gene family, the function and nitrate expression patterns were different.
ZmNRT1.1A has been shown to enhance expression in shoots and improve growth under N deficiency stress (Sakuraba et al. 2021).Very recently, overexpressing ZmNRT1.1B was proven to determine N use efficiency, and its overexpression confers significantly higher grain yield under low to moderate N supply in the fields (Cao et al. 2023).ZmNRT2.1 has improved functional interaction in nitrate uptake along the root axis of maize, as evidenced by gene expression results (Lupini et al. 2016).The expression of ZmNRT2.5 was upregulated with prolonged N starvation and reduced after nitrate resupply, indicating it may play an important role in mediating nitrate uptake by plants under long-term low-N stress.On the contrary, ZmNRT3.1Adisplayed decreased expression levels during prolonged N starvation but exhibited a rapid increase in expression after nitrate resupply (Jia et al. 2023).Because all 6 genes targeted by ZmEREB97 are closely associated with different nitrate translocations, it is reasonable to speculate that ZmEREB97 serves as a hub and a positive regulator in mediating NRTs.AP2/ERF TF family members contain motifs and domains involved in transcription activation, repression, or protein-protein interactions, indicating diverse roles in gene regulation (Chandler 2018).However, much less is known about its function in regulating ZmNRT genes in maize.Luciferase-based transient transactivation assays verified that ZmEREB97 activates the expression of 6 interacted ZmNRT genes involved in nitrate transport (Fig. 10, B to G).The activating role of ZmEREB97 is also in accordance with the decreased expression levels of these ZmNRT genes in the zmereb97 mutants in our study (Fig. 8).These results suggest that ZmEREB97 functioned as a transcription activator by binding to the promoters of ZmNRT genes.
TFs induce or repress gene expression by binding to specific motifs present in the promoters of target genes (Yamasaki et al. 2012).Within the AP2/ERF TF family, DREB subfamily can bind to a consensus A/TGCCGCC sequence and GCC-box-related sequences in the promoter of many target genes (Mizoi et al. 2012).The sequence analysis of the promoters of 6 ZmNRT target genes shows that all contain the GCC-box-related sequences.EMSA confirmed that GCC-box elements are necessary for ZmERER97 binding (Fig. 10, H to M).This finding is similar to the previous studies that 2 AP2/ERF TFs, ERF1B and ERF104, can bind to the GCC-boxes in the NRT1.8 promoter region, thus upregulating NRT1.8 in Arabidopsis (Zhang et al. 2014).Recently, another AP2/ERF TF in rice, OsDREB1C, was found to bind to the promoters of OsNRT1.1B and OsNRT2.4 and activate their expression in roots (Wei et al. 2022), indicating that the regulation of NRT by AP2/ERF TFs may be common in plants.
In conclusion, we demonstrate that the TF ZmEREB97 may serve as a positive regulator of ZmNRT gene expression, thereby conferring NUE through the activation of nitrate uptake in maize.The discovery of the role of ZmEREB97 in regulating nitrate response advances our understanding of the regulatory networks and a mechanistic basis of plant responses to N availability.Manipulating ZmEREB97 expression may be a promising strategy for future biotechnologies to breed high NUE maize to adapt to fluctuating N supply.

Plant materials and growth conditions
Maize (Z.mays L.) zmereb97-MU mutant in B73 background was obtained from a Mu insertional library (ChinaMu, http://chinamu.jaas.ac.cn/;Liang et al. 2019).Homozygous mutant plants were identified by PCR using the specific primers listed in Supplementary Table S1.The mutant were backcrossed to the B73 genetic background for 3 generations, and then, the WT and positive plants from 1 maize cob were self-pollinated for 2 generations.
Maize ZmEREB97 CRISPR-Cas9 transgenic lines named zmereb97-CRI (in the KN5585 background) were generated by the WIMI Biotechnology Company.The 170-bp guide RNA target editing sequence was selected.Eleven CRISPR/Cas9-edited plants were crossed with KN5585 inbred lines to obtain construct-free materials.The mutants were backcrossed to the KN5585 genetic background for 2 generations, and then, the heterozygote plants were self-pollinated for 1 generation.The WT-CRI and zmereb97-CRI mutant was identified and obtained from the same corn cob of the self-pollinated plants.
The pH was adjusted to 5.8.The concentration of KNO 3 was determined based on the experimental requirement.Normally, the seedlings of WT and ZmEREB97 mutants were grown in Hoagland solution with 5 mM KNO 3 for 1 wk and then transferred to 5-mM KNO 3 (HN) and 0.5-mM KNO 3 (LN) Hoagland liquid solution for 2 wk.
For soil-based experiments, the soil collected from the field was thoroughly mixed with a cement mixer and then evenly divided into uniformly sized cultivation pots.The soil used possessed 1.66 g N kg −1 of extracted mineral N, together with 23.1 mg kg −1 of available phosphorus (P), 129.6 mg kg −1 of available potassium (K), and 32.4 g kg −1 of organic matter.To ensure sufficient N and other nutrient elements in the soil, the maize seedlings were watered with Hoagland solution containing 5 mM KNO 3 3 times during the entire growth period.

Transcriptome analysis
Total RNA from 36 samples per treatment was used for transcriptome analysis.Library construction and sequencing were performed according to Illumina instructions by the Berry Genomics Company (Beijing, China), after quality control process.Top Hat was used to align the reads to the maize B73_v3 genome (Trapnell et al. 2009).Fragments per kilobase of exon per million fragments mapped (FPKM) were calculated, representing the expression level.DEGs between control and nitrate-treated samples were calculated.The predicted gene function in this study is based on the annotation provided in MaizeGDB.

De novo motif discovery and analysis
The MEME Suit (http://meme-suite.org/)was used to identify motifs in the promoter regions of genes from the N-related modules.We defined the promoter regions as 1 kb upstream and 100-bp downstream of the start codon sites.For each module, 50 motifs were generated by MEME with the length from 10 to 25 bp, and the motif with E < 0.05 was retained.MAST was used to identify the motif similarities (P < 1e-4), and motifs with the similarities >0.6 were merged by STAMP.FIMO was used to identify the enrichment (occurrences) of the motifs in the promoter regions of genes (P < 1e-4).
For Y1H assay, the promoter region (−1 bp to approximately −1,200 bp) of NRT gene was fused into the pHIS2.1 to generate the bait construct and then transformed into Y187 yeast strain.The coding region of ZmEREB97 was linked with linearized pGADT7-DEST vector as a prey construct.The transformed yeast cells were grown at 30 °C for 4 d on Synthetic Drop-out (SD) medium plates lacking leucine (Leu), tryptophan (Trp), and histidine (His) in the presence of different concentrations of 3-amino-1,2,4triazole.The prey fragments from the positive colonies were identified by DNA sequencing.

EMSA
Full-length ZmEREB97 cDNA was subcloned into pCold (Takara).His-tagged recombinant protein was expressed in E. coli host strain DE3 cells.The culture solution was added with IPTG (final concentration 0.4 mM) and kept on culturing at 15 °C, for 24 h.His-tagged recombinant protein was purified using His-tag Protein Purification Kit (catalog no.P2226, Beyotime, China).Oligonucleotide probes were synthesized and labeled with biotin at the 5′-end.About 80 ng of purified proteins was mixed with the 2.5 ng of probes at 25 °C, for 20 min in an EMSA/Gel Shift Binding Buffer (Beyotime).The DNA in the gel was then transferred to N + nylon membranes (0.2 μm, Millipore).The DNA on the membrane was detected using Chemiluminescent EMSA Kit (Beyotime).

Transient transcriptional activity assay
For the transient transcriptional activity assays of the promoters, the 4×motif1 were cloned into vector pGreen II-0800-LUC to generate reporters for the dual-luciferase assays.The full-length ZmEREB97 cDNA was inserted into vector pMDC83-35S to generate a 35S promoter-driven ZmEREB97 effector.The plasmids carrying effectors and reporters were transformed into and Agrobacterium tumefaciens GV3101 (pSoup-P19), respectively.Reporter and reference values were measured using Dual-Luciferase Reporter Gene Assay Kit (Beyotime) and BioTek Cytation 5 (BioTek) Microplate Reader.Promoter activity was calculated as the firefly luciferase (fLUC)/Renilla luciferase (rLUC) ratio.

qPCR analysis
Total RNA was extracted using the SV Total RNA Isolation System Kit (Promega, USA).Prime Script RT Reagent Kit (Takara) was used to reverse-transcribe RNA to cDNA.qPCR was performed in 96-well plates using a Bio-Rad CFX96 system with SYBR Premix Ex Taq II (Takara).The housekeeping genes ZmUPF1 and ZmDUF were used as internal control genes.Three independent biological replicates were used for qPCR analysis.

Western blot assay
For western blot analyses, total proteins were extracted from B73 roots.Nucleus proteins and cytosolic proteins were isolated by CelLytic PN Isolation/Extraction Kit (Sigma).The protein concentration was quantified by BCA Protein Assay Kit (Beyotime) and resolved on SDS-PAGE.The resolved protein was transferred to the PVDF membrane (0.2 μm, Beyotime), and the membrane with protein samples attached was incubated with 5% nonfat milk, primary antibodies, and secondary antibodies.The antibody against ZmEREB97 was used at 1:1,000; the antibody against UDPGP and H3 was used at 1:3,000.

DAP-seq
For DAP-seq, the genomic DNA was extracted from 2-wk-old maize inbred line B73 roots, and the genomic library was constructed according to the protocol of Bartlett (Bartlett et al. 2017).Full-length ZmEREB97 cDNA was subcloned into pFN19K expression vector, which contains an N-terminal HaloTag affinity tag and then expressed in an in vitro wheat germ system (Los et al. 2008).The Halo-ZmEREB97 was immobilized on Magen HaloTag beads, washed, and then incubated with the genome DNA library.DNA was eluted and amplified with indexed TruSeq primers after bead washing.Sequencing was performed on an Illumina HiSeq 2500 with 100 bp SR reads.Input DNA and GFP were used as control, and 2 technical replicates were used for each sample (Ning et al. 2019).Reads were mapped to the maize B73_v3 genome, and DAP-seq peaks were called by MASC2 (Zhang et al. 2008).Peaks were identified with the cutoff q-value of 0.05.MEME-ChIP was used for de novo motif discovery (Machanick and Bailey 2011).

Subcellular localization of ZmEREB97
A full-length ZmEREB97 cDNA, without termination codon, was cloned into vector pAN580 and fused to the N-terminal of GFP.The fusion construct pAN580-ZmEREB97-GFP and the control pAN580-GFP were transformed into maize mesophyll protoplasts independently.Maize mesophyll protoplast isolation and transformation methods were performed according to the protocol described by Tu et al. (2020).The fluorescence signals were captured by a confocal laser scanning microscope (UltraView VOX, PerkinElmer).The settings used for confocal microscopy were as follows (in nanometers: excitation [ex] and emission [em]): for sGFP, ex 488, em 500 to 550; for RFP, ex 552, em 575 to 625; fluorescence intensity, 5%; and gains value, 500 to 800.

RNA in situ hybridization
A 291 bp specific fragment of ZmEREB97 was amplified by PCR with the primers (Supplementary Table S1) and inserted into pClone007 Blunt Vector (TsingKe Biotech) for sequencing.Sense probe was generated by primers T7-F and ZmEREB97-R and the antisense probe by primers ZmEREB97-F and T7-R.Sense and antisense probes were transcribed in vitro from the T7 promoter with T7 RNA polymerases using the digoxigenin RNA Labeling Kit (Roche).Two-week-old maize inbred line B73 root tissues were fixed overnight in 4% (w/v) paraformaldehyde in phosphate buffer, pH 7.0, and embedded in Paraplast Plus (Sigma-Aldrich) for in situ hybridization.As described by Coen et al. (1990), nonradioactive RNA in situ hybridization with digoxigenin-labeled sense and antisense probes were performed on 8-mm sections of different root parts.

Measurement of net NO 3 − fluxes and nitrate content
Net NO 3 − fluxes were measured noninvasively with NO 3 − -selective microelectrodes using the NMT system (BIO-003A system; Younger USA Science and Technology).The working principle and measurement procedure were described in detail by Tang et al. (2012).Maize seedlings were grown in half-strength Hoagland nutrient solution for 2 wk and then transferred to half-strength Hoagland containing 1 mM KNO 3 for 1 d.Prior to the flux measurements, the ion-selective electrodes were calibrated using NO 3 − concentrations of 0.05 and 0.5 mM.During the entire measurement process, the shoot was not in contact with the measuring solution.The net fluxes of NO 3 − at the meristem were measured individually.Each plant was measured once.The final flux values were the means of more than 5 individual plants.The measuring solution was composed of 0.2 mM CaCl 2 , 0.1 mM NaCl, 0.1 mM MgSO 4 , 1 mM KNO 3 , and 0.3 mM MES (pH 6.0, adjusted with 1 M NaOH).
The influx rate of 15 NO 3 − was assayed as already described (Tang et al. 2012) on plants grown hydroponically.Maize seedlings were grown in Hoagland solution for 2 wk and then deprived of N for 4 d.The plants were transferred first to 0.1 mM CaSO 4 for 1 min, then to a complete nutrient solution containing 0.5 mM K 15 NO 3 for 10 and 30 min, and finally to 0.1 mM CaSO4 for 1 min.After grinding in liquid N, an aliquot of the powder was dried to a constant weight at 70 °C.About 10 mg of powder of each sample was analyzed using the isotope ratio mass spectrometer system (Elementar IsoPrime100).Influx of 15 NO 3 − was calculated from the 15 N concentrations of the roots and shoots.
For plant nitrate measurement, maize seedlings of the WT and zmereb97 mutant were grown in half-strength Hoagland solution for 2 wk and then N starvation for 5 d before measurement.Nitrate concentration was determined as described previously (Leleu and Vuylsteker 2004).Total N concentration in plants was determined by the Kjeldahl method (Li et al. 2006).

Statistical analysis
All experiments were carried out with at least 3 independent biological replicates.Each measurement was carried out in triplicate.Data represent the mean ± SD of 3 biological replicates and each containing 5 plant samples.Data were statistically analyzed by ANOVA performed using SPSS Statistics 20.

Figure 1 .
Figure 1.Network and GO enrichment analysis of maize roots' response to nitrate supply.A) Hierarchical cluster tree showing coexpression modules identified using WGCNA.Modules correspond to branches and are labeled by colors.B) Matrix with the modules and time points after nitrate supply relationships.Each row corresponds to a module eigengene and column to a time point.Each cell contains the corresponding correlation coefficient (up) and P-value (down).The table is color-coded by correlation according to the color legend.C to F) GO enrichment analyses showing N-related processes are enriched in four coexpression modules.G) A schematic diagram of representative highly connected genes and network of the four coexpression modules, showing high correlation with N, carbon, carbohydrate metabolism, signaling pathway, and hormone.

Figure 2 .
Figure 2. Identification of ZmEREB97.A to C) De novo motif discovery of 6 motifs and their enrichment in the 4 N-related modules.A) Relationship tree and sequence logos for 6 sequence motifs.B) The enrichment of 6 motifs in the promoter of genes from the 4 N-related modules.C) Frequency distribution histogram and density map showing the position bias of 6 motifs in the promoter of genes from 4 modules and 500 random genes.D)Y1H assay.Yeast harboring the 4×Motif 1-HIS3 reporter gene was transformed with expression vectors for the GAL4 activation domain fused to the ZmEREB97 and ZmEREB83 (control) and grown on control (-Trp/-Leu) and selection medium (-His/-Trp/-Leu + 60 mM 3-AT).E) EMSA with 6×His-ZmEREB97 fusion protein using Motif 1 as a probe.Nonlabeled WT and mutated Motif 1 were used as competitor.F) Transcriptional activity assays of ZmEREB97.The 35S::REN-4×Motif 1::LUC reporter construct was transiently expressed in N. benthamiana leaves together with the negative control vector and the 35S::ZmEREB97 effector vector.The reporter construct contains 2 reporter genes, the LUC and the REN.4×Motif 1 activities are given as fLUC/rLUC activity ratios.The bars represent means ± SD (n = 3).Two-tailed Student's t test (n = 3 collective samples each containing 3 biological replicates) was used to test the significance.**P < 0.01.

Figure 3 .
Figure 3. Localization of ZmEREB97 in maize.A) RNA in situ hybridization assay to detect transcripts of ZmEREB97 in maize root.(I) Sense probe (negative control).(II to VII) Antisense probe.(II) Main root.(III) Zone of cell differentiation.(IV) Zone of cell elongation.(V) Zone of cell division.(VI) Longitudinal section of lateral root.(VII) Transverse section of lateral root.Hybridization signals detected by labeled antisense probes in lateral root are shown by stars.Bar = 100 μm.B) The subcellular localization of ZmEREB97 in maize mesophyll protoplasts.Bar = 20 μm.C) Determination of ZmEREB97 antibody specificity in the input sample (input), elution sample (elution), and postbind input sample (post bind) of the immunoprecipitation experiment by western blot analyses.Arrows indicate ZmEREB97specific band.D) Western blot assay of ZmEREB97 in roots of maize seedlings.Total, total proteins; cytosolic, cytosolic proteins; nucleus, nucleus proteins.UDPGP protein was used as a cytosolic marker, and histone H3 was used as a nucleus marker, respectively.Red arrows indicate ZmEREB97-specific band.

Figure 4 .
Figure 4. Time course analysis of expression of ZmEREB97 in maize roots with or without CHX under nitrate treatment.A) Expression levels were analyzed by qPCR and normalized using ZmUPF1 as an internal control.Three independent biological replicates were performed with similar results.The bars represent means ± SD (n = 3).ANOVA was used to analyze the expression levels of ZmEREB97 under different treatments.**P < 0.01.B) Western blot analysis of protein levels of ZmEREB97 with or without CHX.Anti-UDPGP was used as a sample loading control, red arrows indicate the ZmEREB97-specific band.

Figure 5 .
Figure 5. Genome-wide binding profiles from DAP-seq analysis of ZmEREB97.A) Distribution of ZmEREB97-binding regions in the maize genome.B) Venn diagram showing the overlap of 10,702 genes bound by the ZmEREB97 in the promoter regions to the 6,069 DEGs.Genes in the overlapping regions were identified as potential ZmEREB97-targeted genes.C) Distribution of ZmEREB97-binding sites in 1,446 potential ZmEREB97-targeted genes.D) Frequency distribution histogram and density map showing the distribution of ZmEREB97-binding peaks in the 1,446 potential ZmEREB97-targeted genes from −1-kb upstream to 4-kb regions.E and F) ZmEREB97-binding motifs identified by MEME-ChIP in the 500-bp flanking sequences around the summits of peaks associated with ZmEREB97 potential targets and the density plot of this motif around the summits of the peaks.

Figure 6 .
Figure 6.Morphological and physiological phenotypes of zmereb97 mutants in different nitrate conditions.The plants were grown in 5 mM KNO 3 for 5 d and then transferred into 5 mM KNO 3 and 0.5 mM KNO 3 , respectively, for 15 d.A and B) The phenotype of WT (WT-MU and WT-CRI ) and zmereb97 mutants (zmereb97-MU and zmereb97-CRI ) grown in 5-mM and 0.5-mM KNO 3 conditions.C to F) The biomass, SPAD value, and root/shoot ratio of WT and zmereb97 mutants under 5-mM and 0.5-mM KNO 3 treatments.G to J) The nitrate and total N content in the root and shoot of WT and zmereb97 mutants.K and L) Phenotype of the ear and aboveground in WT and zmereb97 mutants on soil condition.M to O) The aboveground dry weight, ear length, and grain yield in WT and zmereb97 mutants in soil-based experiments.Bars = 30 cm in A) and B), 6 cm in K), and 35 cm in L).The error bars in C to J) and M to O) represent means ± SD (n = 6).Lowercase letters indicate significant differences at P < 0.01 according to Student's t test.

Figure 7 .
Figure 7. Analysis of net NO 3 − flux, steady NO 3 − flux, and 15 N-NO 3 − flux in WT and zmereb97 seedlings.A and B) Net NO 3 − flux in root of 7-d-old maize seedlings.Seedlings were transferred into measuring solution with 0.5 mM KNO 3 for 40 min before measurement.After 5 min of flux recordings, 10 mM KNO 3 or 1 mM KNO 3 was added into the measuring solution.Inserts in A) and B) denote the steady NO 3 − influx after 30-min KNO 3 treatment.The sign convention is "influx negative."C and D) Maize seedlings of the WT (WT-MU and WT-CRI ) and zmereb97 mutants (zmereb97-MU and zmereb97-CRI ) were grown in Hoagland solution for 2 wk and then deprived of N for 4 d. 15N-NO 3 − influx of roots and shoots were then measured using 0.5 mM K 15 NO 3 for 60 and 120 min.The bars represent means ± SD (n = 6).Lowercase letters indicate significant differences at P < 0.01 according to Student's t test.

Figure 8 .
Figure8.Validation of the interaction of ZmEREB97 on ZmNRT gene promoters in maize.A) Y1H assay.The interaction was determined on the SD medium lacking Leu, Trp, and His in the presence of different concentrations of 3-amino-1,2,4-triazole (-Leu/-Trp/-His + X mM 3-AT).B) Display of potential interaction between ZmEREB97 and 17 ZmNRT genes in maize.The genes labeled indicate interaction between ZmEREB97 and 9 ZmNRT gene promoters.

Figure 9 .
Figure9.Transcription of transporter genes in the maize root of WT and zmereb97 mutant seedlings on 5 and 0.5 mM KNO 3 treatments.Maize seedlings of the WT (WT-MU and WT-CRI ) and zmereb97 mutants (zmereb97-MU and zmereb97-CRI ) were grown in Hoagland solution with 5 mM KNO 3 for 1 wk and then transferred to 5 and 0.5 mM KNO 3 treatments for 3 d.All transcript levels were quantified by RT-qPCR and normalized to housekeeping genes (ZmUPF1).The bars represent means ± SD (n = 3).Lowercase letters indicate significant differences at P < 0.01 according to Student's t test (n = 3 collective samples each containing 3 biological replicates).

Figure 10 .
Figure 10.The transcriptional regulation of ZmEREB97 on 6 ZmNRT genes.A) Schematic diagrams of the effector and reporter constructs."REN" represents Renilla luciferase gene; "LUC" firefly luciferase gene; "Term" terminator; "Pro35S" the promoter of CaMV 35S; "TF" the cDNA of ZmEREB97; and "Promoter" the promoter of NRT gene.B to G) The promoter activities of ZmNRT1.1A,ZmNRT1.1B,ZmNRT1.2,ZmNRT2.1,ZmNRT2.5, and ZmNRT3.1Aactivated by ZmEREB97 LUC assays were expressed as the ratio of LUC to REN in N. benthamiana leaves cotransformed with the effector and the reporter combinations.The bars represent means ± SD (n = 3).Two-tailed Student's t test (n = 3 collective samples each containing 3 biological replicates) was used to test the significance.**P < 0.01.H to M) EMSA using probes derived from the ZmEREB97 target gene promoters containing GCC-box-binding sites.+ and − indicate the presence and absence of the indicated probe or protein, respectively.