Integrated comparative transcriptome and physiological analysis reveals the metabolic responses underlying genotype variations in NH4 + tolerance

Several mechanisms have been proposed to explain NH4 + toxicity. However, the core information about the biochemical regulation of plants in response to NH4 + toxicity is still lacking. In this study, the tissue NH4 + concentration is an important factor contributing to variations in plant growth even under nitrate nutrition and NH4 + tolerance under ammonium nutrition. Furthermore, NH4 + led to the reprogramming of the transcriptional profile, as genes related to trehalose-6-phosphate and zeatin biosynthesis were downregulated, whereas genes related to nitrogen metabolism, camalexin, stilbenoid and phenylpropanoid biosynthesis were upregulated. Further analysis revealed that a large number of genes, which enriched in phenylpropanoid and stilbenoid biosynthesis, were uniquely upregulated in the NH4 +- tolerant ecotype Or-1. These results suggested that the NH4 +-tolerant ecotype showed a more intense response to NH4 + by activating defense processes and pathways. Importantly, the tolerant ecotype had a higher 15NH4 + uptake and nitrogen utilization efficiency, but lower NH4 +, indicating the tolerant ecotype maintained a low NH4 + level, mainly by promoting NH4 + assimilation rather than inhibiting NH4 + uptake. The carbon and nitrogen metabolism analysis revealed that the tolerant ecotype had a stronger carbon skeleton production capacity with higher levels of hexokinase, pyruvate kinase, and glutamate dehydrogenase activity to assimilate free NH4 +, Taken together, the results revealed the core mechanisms utilized by plants in response to NH4 +, which are consequently of ecological and agricultural importance.


Introduction
Nitrogen (N) is an essential plant macronutrient that is required for growth and is thus important in agricultural production.NH 4 + based fertilizers contain commonly used N forms synthesized via the Haber-Bosch process (Halvorson et al., 2014).While NH 4 + is the preferred form of N for plants at low concentrations, it is toxic to plants at high concentrations (von Wireń et al., 2000;Britto and Kronzucker, 2002).As NH 4 + escapes little with water in soil and elevated CO 2 reduces nitrate reductions in C3 species, a "more NH 4 + solution" was considered to mitigate nitrogen pollution and improve crop yields (Rubio-Asensio and Bloom, 2016;Subbarao and Searchinger, 2021).However, excess NH 4 + /NH 3 atmospheric depositions have attracted attention in recent decades as they caused environmental problems related to species richness and composition (Stevens et al., 2004;Clark and Tilman, 2008;Duprè et al., 2010).The mechanisms underlying NH 4 + toxicity in plants are consequently of ecological and agricultural importance.NH 4 + toxicity causes a retardation in plant growth, shortened roots, reduced root gravitropism, and leaf chlorosis (Britto and Kronzucker, 2002;Li et al., 2014).Several mechanisms have been proposed to explain NH 4 + toxicity, including the depletion of organic acids (Hachiya et al., 2010), deficiency of cations (Hoopen et al., 2010), futile transmembrane NH 4 + cycling (Britto et al., 2001;Li et al., 2022), photodamage to photosystem II (Dai et al., 2014), reactive oxygen and nitrogen species (RONS) induced oxidative stress (Liu et al., 2022), and the disruption of hormonal homeostasis (Li et al., 2012).The reactions of NH 4 + conjugation to glutamic acid and the synthesis of glutamic acid are critical for the detoxification of NH 4 + , and are catalyzed by glutamine synthase (GS), glutamate synthase (GOGAT), and glutamate dehydrogenase (GDH) (Masclaux-Daubresse et al., 2006).Studies have shown that GS activity is upregulated by high external NH 4 + levels and that NH 4 + -tolerant plants have higher GS activity and lower levels of NH 4 + accumulation in their tissues (Cruz et al., 2006;Guan et al., 2016).In addition, NH 4 + uptake is tightly controlled through the NH 4 + -dependent inhibition of AMT1 by the phosphorylation of Thr-460 (Straub et al., 2017).
In recent years, numerous genetic loci controlling NH 4 + toxicity have been identified by screening for mutants.The Arabidopsis mutant hsn1-1 and its alelic mutant vtc1 are hypersensitive to NH 4 + because of point mutations in the gene encoding GDP-mannose pyrophosphorylase (GMPase).Defects in protein N-glycosylation of hsn1-1 have been linked to the root hypersensitivity phenotype (Qin et al., 2008).Based on leaf hypersensitivity to NH 4 + , ammonium overly sensitive 1 (amos1) and amos2 were identified.amos1 was found to be an allelic mutation of EGY1 encoding a plastid metalloprotease.AMOS1/EGY1 mediated plastid retrograde signaling regulates NH 4 + -responsive genes to maintain chloroplast functionality (Li et al., 2012).An ammonium tolerance mutant (amot1) was identified as allelic to EIN3, which positively regulates ROS production and induces oxidative stress under NH 4 + stress (Li et al., 2019a).Moreover, other genetic loci that control root development and gravitropism have been identified in Arabidopsis, including auxin resistant 1 (aux1), tiny root hair 1 (trh1), dolichol phosphate mannose synthase1 (dpms1), and gravitropism sensitive to ammonium 1 (gsa1) (Li et al., 2014).
Although the numerous genetic loci were found to associate with the toxicity of NH  et al., 2016;Li et al., 2021).In this study, we explore the cellular threshold of NH 4 + concentration by investigated the A. thaliana natural accessions.More importantly, the core information about the biochemical regulation of plants in response to NH 4 + toxicity was identified by comparative assessments of the sensitive and tolerant species.

Plant material and culture conditions
A natural accessions of Arabidopsis thaliana were used in this study (Table S1).Surface-sterilized seeds were sown onto nutritional soil in a greenhouse (300 mmol photons m -2 s -1 , 16 h photoperiod, 22°C) for 7 d.Then, eight uniformly growing plants of each ecotype were grown hydroponically with 4 L of nutritional media per pot, containing 1 mM Ca(NO 3 ) 2 or 1 mM (NH 4 ) 2 SO 4 .The other essential nutrients in the two different nitrogen nutrient solutions were the same as our previous study (Chen et al., 2021), specially 1.25 mM KCl, 0.625 mM KH 2 PO 4 , 0.5 mM MgSO 4 , 25 mM Fe-EDTA, 17.5 mM H 3 BO 3 , 3.5 mM MnCl 2 , 0.25 mM ZnSO 4 , 0.05 mM NaMoO 4 , and 0.125 mM CuSO 4 .It is of note that the concentration of Ca 2+ in the solution was uniformly set to 1.5 mM, and The pH was buffered to 6.0 using 2.5 mM MES.The thaliana natural accessions was grown under the NH 4 + and NO 3 -.
The seedlings were first grown in 1 mM Ca(NO 3 ) 2 nutrient solution for 5 d and then transferred to nitrogen-free nutrient solution for 3 d.Subsequently, the thaliana natural accessions were grown under the NH 4 + and NO 3 -treatments for 8 days to investigate the NH 4 + concentration and fresh weight.The culture solution was refreshed every 4 days.
Measurement of the NH 4 + and total nitrogen concentrations + was extracted with deionized water using fresh rosette leaves.The supernatant was used to determine the NH 4 + concentration after the samples were centrifuged at 12,000 × g.Briefly, 20 mL supernatant was mixed with 0.5 mL phenol solution (10 g/L Phenol, 100 mg/L sodium nitrosoferricyanide) and 5 mL sodium hypochlorite alkaline solution (Sodium hydroxide 5 g, sodium hydrogen phosphate 3.53 g, sodium phosphate 15.9 g, sodium hypochlorite solution (w= 5.25%) 5 mL, dissolved in 500 ml deionized water).The NH 4 + concentrations were measured colorimetrically using phenol hypochlorite (Berthelot reaction) colorimetry at 630 nm, and (NH 4 ) 2 SO 4 was used as a standard.For total nitrogen (TN) measurement, the plants were sampled and then dried at 105°C for half an hour and further dried at 65°C until a constant weight.The samples were weighed 0.100g using a thousand balance, packed in 150 mL narrow-mouth triangular flasks, boiled with H 2 SO 4 -H 2 O 2 until transparent and clear.The solution was transferred to a 50 mL volumetric flask and made up to volume.The solution was then filtered and analyzed using a continuous-flow analyzer AA3 (Autoanalyzer 3; SEAL, Germany).NUtE (Nitrogen utilization efficiency) was calculated as dry shoot biomass/shoot N.

Enzyme activity
Fresh samples (100 mg) were homogenized in 3 mL of 50 mM Tris-HCl buffer (pH 8.0) containing 2 mM Mg 2+ , 2 mM DTT, and 0.4 M sucrose.After centrifuged at 10000 × g and 4°C for 10 min, the supernatant was then used to determine the activity of glutamine synthase.The activities of GS were measured according to the reference (Chen et al., 2018).Briefly, 0.5 mL supernatant was mixed with 1.6 mL reagent ① , which is composed of 100 mM Tris-HCl (pH 7.4), 80 mM MgSO4, 20 mM glutamate, 20 mM cysteic acid, 2 mM EGTA and 80 mM hydroxyl-amine hydrochloride, and 0.7 mL reagent ② , which is composed of 40 mM ATP.The 3 mL reaction mixture was incubated for 30 min at 37°C and was terminated by adding 1 mL FeCl 3 reagent (88 mM FeCl 3 , 670 mM HCl and 200 mM TCA).After 10 min, the mixture was centrifuged at 4000 × g for 10 min, and the absorption value of supernatanta solution was determined at 540 nm wavelength.
NADH-dependent glutamate dehydrogenase was measured according to the reference (Groat and Vance, 1981).Briefly, fresh samples (100 mg) were homogenized in 3 mL of 50 mM Tris-HCl buffer (pH 8.0).50 mL of the supernatant was mixed with the 950 mL reagent, which contained 100 mM NADH, 2.5 mM 2-oxoglutarate (2-OG), 200 mM NH 4 Cl.The enzyme activity was defined as the reduction of absorbance due to NADH at 340 nm.The assay of enzyme activity was performed according the instruction of commercial kit (NADH-GDH kit, BC1465, Solarbio).
Pyruvate (Pyr) concentrations were measured according to the reference with some modifications (Kachmar and Boyer, 1953).Pyr reacts with 2, 4-dinitrophenylhydrazine to form pyruvate-2, 4dinitrophenylhydrazone, which could be determined by calorimetry.Briefly, 0.1 g Fresh samples (100 mg) were homogenized in 1 mL trichloroacetic acid (8%) and placed on ice for 3 min.After centrifuged at 8000 × g and 4°C for 10 min, 75 mL the supernatant was added to the enzyme label plate and added 25 mL 2, 4-dinitrophenylhydrazine (0.05%), to react 2 min.Finally, 125 mL 1.5 M NaOH was added, and the absorption value of the tube was determined at 520 nm wavelength.The assay was performed using the Micro Pyruvate Assay Kit (BC2205; Solarbio).
Hexose kinase (HXK) activity was measured according to the reference (Pancera et al., 2006).The assay was performed according to the instructions of the Hexose Kinase Assay Kit (BC0740; Solarbio).Briefly, 0.1 g Fresh samples (100 mg) were homogenized in 1 mL extracting solution.After centrifuged at 8000 × g and 4°C for 10 min, 10 mL the supernatant was added to the enzyme label plate, followed by 10 mL G-6-PDH solution (0.12 g/L) and 180 mL reagent, which 5 ml of a 2.0 mM NADP + solution, 15 ml of a 0.1 M ATP solution, 50 ml of a 1 M glucose solution, and 110 mL of a 100 mM Tris-HCl buffer (pH 7.5).The absorption values at 340 nm wavelength were immediately recorded at 20s and 320s.

N-NH 4 + isotope tracing
The objective of 15 N-labeling experiment after 3d of N starvation is investigate the transport capacity of both genotypes.Seedlings were first grown in 1 mM Ca(NO 3 ) 2 nutrient solution for 5 d, and then transferred to a nitrogen-free nutrient solution for 3 d.Subsequently, the plants were transplanted into 1 mM ( 15 NH 4 ) 2 SO 4 with a 15 N abundance of 5%.Samples were taken at 3, 6, and 24 h to determine the 15 N content.

Comparative transcriptome analysis
The seedlings were grown in a NO 3 -nutrient solution for 5 d, and then transferred to a nitrogen-free nutrient solution for 3 d to deplete the stored NO 3 -.Subsequently, the plants were transferred to 1 mM Ca(NO 3 ) 2 or 1 mM (NH 4 ) 2 SO 4 for 1 d.The total RNA was then extracted from the roots of the NH 4 + -tolerant and -sensitive ecotypes.Three biological replicates were used for each ecotype under each treatment.The sequencing library was generated using the NEBNext Ultra ™ RNA Library Prep Kit from Illumina (New York, NEB, USA), followed by sequencing on an Illumina HiSeq 2500 platform (San Diego, CA, USA).Gene expression levels were calculated using the FPKM method (Fragments Per Kilobase per Million mapped reads.Differentially expressed genes (DEGs) were defined as genes with |log 2 (fold change) | >1 and a false discovery rate (FDR) < 0.05.The comparative transcriptome between NH 4 + and NO 3 -was identified through comparisons of the FPKM values for each gene between NH 4 + and NO 3 -.The comparative transcriptome between the two genotypes was identified through comparisons of the FPKM values for each gene between Or-1 and the Rak-2.GO enrichment analysis was performed by the agriGo program (http:// bioinfo.cau.edu.cn/agriGO/).The significantly enriched GO terms were defined with corrected P < 0.05.

Statistical analysis
Graphical values represent the mean ± SD.Statistical analyses were conducted using two-tailed Student's t-tests or one-way analysis of variance (ANOVA).Least significant difference (LSD) tests for different treatments were used for multiple comparisons with P < 0.05.The significance level was set at P < 0.05 (*) and P < 0.01 (**).

Results
The tissue NH 4 + concentration was negatively linked to Arabidopsis growth A panel of A. thaliana natural accessions was grown under identical concentrations of NH 4 + and NO 3 -.The growth of each ecotype was significantly inhibited by NH 4 + (Figures 1A, B).
Conversely, nitrogen concentration was significantly higher in plants grown under NH 4 + than under NO 3 -conditions (Figure 1C).However, the free NH 4 + concentration in the shoots The tissue NH 4 + concentration is negatively linked to the growth of Arabidopsis.After 8 d cultivation with NO 3 -or NH 4 + nutrition, the physiological parameters including the fresh weight, rosette length, nitrogen concentration and free NH 4 + were investigated.Frontiers in Plant Science frontiersin.orgwas significantly higher than that under NO 3 -, and its concentration varied widely among ecotypes (Figure 1D).The two-segment linear model accurately simulated the relationship between fresh weight and tissue NH 4 + concentration under different nitrogen sources (Figure 1E, Table S1).The fresh weight sharply decreased with the tissue NH 4 + concentration (<50 mg/g) under NO 3 -conditions, and the pace of decline slowed when the NH 4 + concentration was > 50 g/g under NH 4 + conditions (Figure 1E).Our data revealed that the tissue NH 4 + concentration was negatively linked to Arabidopsis growth, even under NO 3 -culture conditions.
Sensitivity to NH 4 + between Or-1 and the Rak-2 We comprehensively considered the biomass under ammonium nitrogen and the biomass under normal conditions, and excluded the ecotypes with poor growth under nitrate nitrogen.Additionally, the Or-1 and Rak-2 showed similar growth period.Thus, ecotypes Or-1 and Rak-2 were selected for further investigation.Or-1 grew better under both NO 3 -and NH 4 + nutrition (Figure 2A), which was also supported by its biomass (Figure 2B).With NO 3 -nutrition, the biomass of Rak-2 was 18.4% smaller than that of Or-1.However, the difference in biomass between the two ecotypes was 53% under NH 4 + nutrition (Figure 2B).The free NH 4 + concentration in Rak-2 was significantly higher than that in Or-1 under both nitrate and NH 4 + nutrition conditions (Figure 2C).Thus, Or-1 was characterized as an NH 4 + -tolerant ecotype with high biomass and low NH 4 + concentration, while Rak-2 was characterized as an NH 4 + -sensitive ecotype.
Core biological processes and pathways involved in the responses to NH 4
By analyzing the higher expression genes in the Rak-2 compared with the Or-1, The most enriched pathways were found to be thiamine metabolism, pentose phosphate pathway (Figure 4D).Analysis of genes with higher expression in Or-1 than in Rak-2 under NH 4 + -rich conditions.The most enriched pathways included stilbenoid, phenylpropanoid, glutathione, and sucrose metabolism (Figure 4C).Importantly, these pathways enriched in the Or-1 were up-regulated by NH 4 + (Figure S4).
The NH 4 + -tolerant ecotype Or-1 exhibits a more intensive response to NH 4 + by activating defense processes and pathways As shown in the VENN diagram, 148 DEGs between the two ecotypes were simultaneously responsive to NH 4 + (Figure 5A).In  addition, 391 DEGs between the two ecotypes specifically responded to NH 4 + in the tolerant ecotype Or-1 (Figure 5A).
The core 148 genes were classified into upregulated and downregulated genes following NH 4 + treatment (Figures 5B, C; Table S2).Among the core 148 genes, a higher proportion of genes were upregulated by NH 4 + , especially Or-1 (Figure 5C).The cytochrome P450 genes CYP86A4, CYP71B22, CYP706A2, and CYP709B3 were strongly induced by NH 4 + and showed higher Or-1 expression (Figure 5C).The transcription factors WRKY55, WRKY41, bZIP8, and ZAT8 were more strongly induced in the NH 4 + -tolerant ecotype Or-1 than that in the NH 4 + -sensitive ecotype Rak-2 (Figure 5C).Surprisingly, 194 DEGs between the two ecotypes that specifically responded to NH 4 + in Rak-2 were mostly   S2).These genes were mostly upregulated by NH 4 + and showed higher expression levels in Or-1 cells (Figures 5D, F).These results suggest that a large number of genes were uniquely upregulated in the NH 4 + -tolerant ecotype Or-1.The function of these unique NH 4 + -responsive genes in the Or-1 were predicted using GO enrichment analysis and KEGG pathways.The most highly enriched biological processes were defense responses, including responses to ABA, water deprivation, chitin, wounding, and oxidative stress (Figure 5G).Similarly, the KEGG pathway analysis indicated that nitrogen metabolism, phenylpropanoid biosynthesis, glutathione metabolism, and plant hormone signal transduction were the most highly enriched pathways (Figure 5H).Collectively, these results suggest that the NH 4 + -tolerant ecotype Or-1 exhibits a more intensive response to NH 4 + by activating defense processes and pathways, including phenylpropanoid biosynthesis, nitrogen metabolism, and glutathione metabolism.
The tolerant ecotype maintained a low NH 4 + level, mainly by promoting NH 4 + assimilation rather than inhibiting NH 4 + uptake As the NH 4 + -sensitive ecotype, Rak-2 accumulated more NH 4 + than the NH 4 + -tolerant ecotype Or-1 (Figure 2C), their uptake and assimilation of NH 4 + were investigated.Both the concentration and accumulation of 15 N were significantly higher in the NH 4 + -tolerant ecotype Or-1 (Figures 6A, B).The higher nitrogen utilization efficiency (NUtE) in the tolerant ecotype indicated the higher efficiency of the conversion of shoot N into shoot biomass (Figure 6C).The expression of AMT2 and AMT1.2 were also higher in the NH 4 + -tolerant ecotype (Figure 6D), which supports the results for 15 NH 4 + assimilation.Although the GS activity of both ecotypes increased with increasing NH 4 + concentration, no difference was observed between the two (Figure 6E).The expression of Gln1s was significantly induced by NH 4 + , and only the expression of Gln1.4 slightly higher in the NH 4 + -sensitive ecotype (Figure 6D).However, GDH activity was strongly induced by NH 4 + nutrition and was significantly higher in the tolerant ecotype when compared with that in the sensitive ecotype (Figure 6F).Accordingly, the expression of GDH1 and GDH2 was more significantly induced in the roots of the tolerant ecotype than in those of the sensitive ecotype (Figure 6D).The vital nodes of C and N metabolism were also investigated.The expression of hexokinase gene (HXK2) and pyruvate kinase gene (PK4) were much higher in the roots of the tolerant ecotype than the sensitive ecotype with NH 4 + nutrition (Figure 6G).These results are consistent with the enzymatic activities of HXK and PK (Figure 6H; Figure S5).Importantly, pyruvate content, which is essential for N metabolism, was reduced by NH 4 + nutrition, whereas its content was significantly higher in the roots of the tolerant ecotype than in the sensitive ecotype under NH 4 + nutrition (Figure 6I).These results indicated that the tolerant ecotype maintained a low NH 4 + level, mainly by promoting NH 4 + assimilation rather than inhibiting NH 4 + uptake.

Discussion
The tissue NH 4 + concentration is an important factor contributing to variations in plant growth and NH 4 + tolerance NH 4 + is toxic to plants, even at micromolar concentrations (von Wireń et al., 2000).Many hypotheses have been proposed to explain NH 4 + toxicity, such as the depletion of organic acids, deficiency of cations, futile transmembrane NH 4 + cycling and so on (Li et al., 2014).Previous studies have explored the medium NH 4 + concentration toxic to different crops (Britto and Kronzucker, 2002).A previous study has expounded the critical role of NH 4 + in Arabidopsis natural variability in NH 4 + tolerance (Sarasketa et al., 2014).In this study, we found the two-segment linear model accurately simulated the relationship between fresh weight and tissue NH 4 + concentration under different nitrogen sources (Figure 1E).When the tissue NH 4 + concentration was below 50 mg/g under nitrate conditions, the fresh weight rapidly decreased with the tissue NH 4 + concentration.Whereafter, the pace of declines slowed, when the tissue NH 4 + concentration was over 50 mg/g under NH 4 + conditions, (Figure 1E).This phenomenon possibly enlightened that low concentrations of NH 4 + do not cause obvious symptoms of NH 4 + toxicity, such as leaf chlorosis, but affect plant growth in an imperceptible way.However, the visible symptoms of NH 4 + toxicity appeared over 50 mg/g under NH 4 + conditions.A previous study revealed that the shoot K + was positively correlated with the growth and NH 4 + tolerance (Chen et al., 2021), and this correlation may be indirectly caused by the inhibition of K + transport.Thus, the tissue NH 4 + concentration is an important factor contributing to variations in plant growth and NH 4 + tolerance.(Hachiya et al., 2021).Indeed, the expression of Gln1s was significantly induced by NH 4 + , whereas the expression of Gln2 was repressed (Figure 6D), suggesting an interesting regulation of GS to adapt NH 4 + stress in plants.Although the GS activity of both ecotypes increased with increasing NH 4 + concentration, no difference was observed between the two ecotypes (Figure 6E).Another study found no correlation between GS activity and shoot biomass in Arabidopsis under NH 4 + nutrition (Sarasketa et al., 2014).Thus, variations in GS activity may not be crucial for NH 4 + tolerance in different Arabidopsis ecotypes.GDH can incorporate NH 4 + independently of the GS/GOGAT cycle and plays a critical role in the detoxification of NH 4 + in stress conditions (Xian et al., 2020).
Although the capacity of GDH to synthesize Glu in vivo has not been clearly demonstrated, heterologous expression of fungal GDH in plants could alleviates NH 4 + toxicity and improve nitrogen assimilation (Tang et al., 2018;Yan et al., 2021).In this study, GDH activity and its encoding genes in both ecotypes were induced

by NH 4
+ nutrition, and the activity was much higher in the tolerant ecotype than in the sensitive ecotype (Figure 6F).Moreover, GDH1 and GDH2 were more strongly induced in the roots of the tolerant ecotype than in those of the sensitive ecotype (Figure 6D).Overall, these results possibly indicate the critical role of GDH in Arabidopsis natural variation in NH 4 + tolerance.
The sucrose metabolism pathway was enriched in the tolerant ecotype, indicating that this process was more active in the NH 4 + tolerant ecotype (Figure 4C; Figure S4).Hexokinase (HXK) and Pyruvate kinase (PK) function crucial roles in Glycolysis and TCA cycle, providing 2-OG and energy for N metabolism (Stitt, 1999).In the present work, the expression of HXK2 and PK4 were much higher in the root of tolerant ecotype than the sensitive ecotype under NH 4 + nutrition (Figure 6G).These results are consistent with the enzymatic activities of HXK and PK (Figure 6H; Figure S5).Importantly, the crucial carbon metabolite pyruvate was reduced by NH 4 + nutrition, whereas its content was significantly higher in the roots of the tolerant ecotype than in the sensitive ecotype under NH 4 + nutrition (Figure 6I).These results suggest that the tolerant ecotype had a stronger carbon skeleton (2-OG) production capacity for NH 4 + assimilation (Figure 6J).
Coordination of carbon (C) and nitrogen (N) metabolism is essential for plant growth and stress tolerance (Reguera et al., 2013;Liang et al., 2020).The higher nitrogen utilization efficiency (NUtE) in the tolerant ecotype indicated the higher efficiency of the conversion of shoot N into shoot biomass (Figure 6C).The NUtE was similar among the accessions with different nitrogen use efficiency and uniformly decreased with high N supply, but the accessions differed in their NUtE under N restriction (Menz et al., 2018).The nitrogen was not limited and the concentration was even higher under NH 4 + nutrition than that under nitrate nutrition.
Thus, the higher NUtE in the tolerant ecotype may suggest bettercoordinated C/N metabolism compared with the sensitive ecotype under NH 4 + stress.
The tolerant ecotype displayed stronger defense responses by activating phenylpropanoids and the derived stilbenoids NH 4 + induces ROS formation, leading to oxidative damage in plants (Podgorska et al., 2013;Liu et al., 2022).Here, stilbenoid biosynthesis, phenylpropanoid biosynthesis, and glutathione metabolism pathways were commonly activated by NH 4 + in all ecotypes (Figure 3D).Stilbenoids, which are hydroxylated derivatives of stilbene belonging to the phenylpropanoid family, have been shown to have a wide spectrum of biological functions, such as antioxidant and antimicrobial activities (Dong and Lin, 2021).Many other secondary metabolites, such as anthocyanins and flavonols, share a common origin in the phenylpropanoid biosynthetic pathway and functions as an ROS scavenger induced by abiotic stress (Dong and Lin, 2021).Treatment with flavonoids, such as naringenin, reduces oxidative damage under Cd stress in rice (Chen et al., 2022).Moreover, phenylpropanoid metabolism is associated with the reinforcement of cell walls under NH 4 + nutrition, which was considered as an important NH 4 + tolerance mechanism (Royo et al., 2019;Yang et al., 2022).Importantly, the NH 4 + -tolerant ecotype Or-1 showed a more intense response to NH 4 + by activating pathways, including phenylpropanoid and stilbenoid biosynthesis (Figure 4C).Therefore, the biosynthesis of phenylpropanoids and stilbenoids derived from NH 4 + may play a positive role in the defense response to NH 4 + toxicity, and this thus warrants further investigation.
The tolerant ecotype was more responsive to NH 4 + stress than the sensitive ecotype.These genes were highly enriched in defense responses, including their responses to ABA, water deprivation, chitin, wounding, and oxidative stress (Figure 5G).ABA signaling plays a key role in oxidative stress responses and is associated with the induction of antioxidant defense systems (Yang et al., 2015;Li et al., 2019b).Interestingly, the NH 4 + -tolerant ecotype has stronger ABA responses and phenylpropanoid metabolism for ROS scavenging (Figures 5G, H).Recently, a research investigated the genetic variation underlying differential ammonium and nitrate responses in Arabidopsis thaliana, and found the preferring ammonium or nitrate, appeared to be generated by combinations of loci rather than a few large-effect loci, which most are specific to a developmental or defense trait under specific nitrogen source (Katz et al., 2022).In our results, we also found only few genes were simultaneously responsive to NH 4 + between the two genotypes (Figure 5A).This suggested that ammonium tolerance genotypes are a combination of multiple mini-effect genes.

Conclusion
In this study, the tissue content of NH 4 + was found the main cause for NH 4 + toxicity, and the two-segment linear model accurately simulated the relationship between fresh weight and tissue NH 4 + concentration under different nitrogen sources.we revealed that the tolerant ecotype maintained a low NH 4 + level, mainly by promoting NH 4 + assimilation rather than inhibiting NH 4 + uptake.The carbon and nitrogen metabolism analysis revealed that the tolerant ecotype had a stronger carbon skeleton (2-OG) production capacity with higher levels of hexokinase (HXK), pyruvate kinase (PK), and GDH activity to assimilate free NH 4

+
. Furthermore, the core information about the biochemical regulation of plants in response to NH 4 + toxicity was identified.The most enriched pathways included nitrogen metabolism, camalexin, stilbenoid and phenylpropanoid biosynthesis were upregulated by NH 4 + .Interestingly, a large number of genes, which enriched in phenylpropanoid and stilbenoid biosynthesis, were uniquely upregulated in the NH 4 + -tolerant ecotype.These results suggested that the NH 4 + -tolerant ecotype showed a more intense response to NH 4 + by activating defense processes and pathways.
(A) The fresh weight, and (B) rosette length under nitrate or NH 4 + nutrition.(C) Total nitrogen concentration, and (D) free NH 4 + concentration in the plants.(E) The correlation analysis of free NH 4 + with fresh weight.Dashed lines indicated two linear models.Pearson R 2 values are given when p <.05.Student's t test (**p <.01) was used to analyze statistical significance.For fresh weight and rosette length, results of each ecotype are means of eight biological replicates, and for TN and free NH 4 + concentration, results of each ecotype are means of three biological replicates.Chen et al.  10.3389/fpls.2023.1286174 (A) Photos of the two ecotypes under NO 3 -or NH 4 + nutrition.(B) The dry weight of whole plants, and (C) free NH 4 + concentration of the two ecotypes under NO 3 -or NH 4 + nutrition.For dry weight, results of each ecotype are means ± SD of eight biological replicates, and for free NH 4 + concentration, results of each ecotype are means ± SD of three biological replicates.**p < .01.

+.
FIGURE 3 Overview of the transcriptome and the DEGs that were highly enriched in the core biological process and pathways in response to NH 4 + .(A) Gene ontology (GO) enrichment analysis of the DEGs under NH 4 + stress.(B) The expression profiling of the genes involving in camalexin biosynthesis, H 2 O 2 transmembrane transport and trehalose biosynthesis.(C) The enrichment pathways of the down-regulated genes by NH 4 + .(D) The enrichment pathways of the up-regulated genes by NH 4 + .(E) The effects of NH 4 + nutrition on the pathway of nitrogen assimilation.(F) The effects of NH 4 + nutrition on the pathway of zeatin biosynthesis.The red lines indicated the process was facilitated and the blue lines indicated the process was inhibited by NH 4 + .

+
Stilbenoid, phenylpropanoid, and sucrose metabolism enriched in the tolerant ecotype were up-regulated by NH 4To explore the molecular mechanism of NH 4 + sensitivity in Or-1 and Rak-2, we compared the transcriptomes of the NH FIGURE 4 Transcriptional characterization of the DEGs between the NH 4 + -tolerant ecotype and the NH 4 + -sensitive ecotype.(A) The expression profiling of DGEs between the NH 4 + -tolerant ecotype and the NH 4 + -sensitive ecotype.(B) GO enrichment analysis of the DGEs.(C) The enrichment pathways of the DGEs with higher expression in the NH 4 + -tolerant ecotype Or-1.(D) The enrichment pathways of the DGEs with higher expression in the NH 4 + -sensitive ecotype Rak-2.The circle size indicates the number of DEGs, and the rich factor indicates the degree of enrichment of the KEGG pathways.
FIGURE 5 Analysis of the common and specific NH 4 + -responsive genes between the two ecotypes.(A) The VENN diagram of the DEGs.(B) The commonly down-regulated DEGs between the two ecotypes.(C) The commonly down-regulated DEGs between the two ecotypes.(D) The DEGs between the two ecotypes specifically responded to NH 4 + only in one ecotype.(E) The expression profiling of the DEGs specifically responded to NH 4 + in the sensitive ecotype Rak-2.(F) The expression profiling of the DEGs specifically responded to NH 4 + in the tolerant ecotype Or-1.(G) GO enrichment analysis of the DGEs specifically responded to NH 4 + in the tolerant ecotype Or-1.(H) The enrichment pathways of the DGEs specifically responded to NH 4 + in the tolerant ecotype Or-1.
FIGURE 6 Analysis of NH 4 + assimilation capacity between the two varieties.(A, B) The short-term of 15 NH 4 + uptake for 3 h, 6 h and 24 h.(C) The nitrogen utilization efficiency (NUtE) between the two ecotypes under NH 4 + nutrition.NUtE is calculated by biomass/nitrogen.(D) Differential expression profiling of the NH 4 + uptake and assimilation genes, including the NH 4 + transporter genes AMTs, glutamine synthesis genes GLN and glutamate dehydrogenase genes GDH. (E) The activity of glutamine synthesis in the roots of two ecotypes.(F) The activity of GDH in the roots of two ecotypes.(G) Differential expression profiling of the carbohydrate metabolism genes, including the cell wall-bound invertase gene CWINV1, hexokinase gene HXK2 and pyruvate kinase gene PK4.(H) The activity of hexokinase in the roots of two ecotypes.(I) The content of pyruvate in the roots of two ecotypes.(J) The schematic diagram of carbon and nitrogen metabolism.The enzymes and metabolite highlighted in red font represent higher activity or content in the root of tolerant ecotype.Results are means ± SD of three biological replicates.*p < .05 and **p < .01.