Guihong Ling,Yingpeng Hu,Hifei Chen,Jinsong Luo,Hongkun Xing,Hixing Song,Zhenhu Zhng,*

a College of Resources and Environmental Sciences,Hunan Agricultural University,Changsha 410128,Hunan,China

b School of Agricultural Sciences,Zhengzhou University,Zhengzhou 450000,Henan,China

Keywords:Nitrogen Amino acids Source-to-sink allocation Yield Genetic variation

ABSTRACT Nitrogen(N)is an essential plant growth nutrient whose coordinated distribution from source to sink organs is crucial for seed development and overall crop yield.We compared high and low N use efficiency(NUE)Brassica napus(rapeseed)genotypes.Metabonomics and transcriptomics revealed that leaf senescence induced by N deficiency promoted amino acid allocation from older to younger leaves in the high-NUE genotype at the vegetative growth stage.Efficient source to sink remobilization of amino acids elevated the numbers of branches and pods per plant under a N-deficiency treatment during the reproductive stage.A 15N tracer experiment confirmed that more amino acids were partitioned into seeds from the silique wall during the pod stage in the high-NUE genotype,owing mainly to variation in genes involved in organic N transport and metabolism.We suggest that the greater amino acid source-to-sink allocation efficiency during various growth stages in the high-NUE genotype resulted in higher yield and NUE under N deficiency.These findings support the hypothesis that strong amino acid remobilization in rapeseed leads to high yield,NUE,and harvest index.

1.Introduction

Nitrogen(N)is a macronutrient for plant growth and development that is needed for the synthesis of amino acids,proteins,and other metabolites[1].N supply is crucial for crop yield and fruit or seed development[2].The use of N fertilizer has increased more than 20-fold in recent decades,exceeding crop requirements by 35%[3],while causing environmental and human health damage.Reducing N fertilization inputs and cultivating genotypes with strong capacity for the absorption,utilization,and remobilization of N is desirable[4].

Brassica napus L.(rapeseed)is an oil crop representing the third major essential source of plant oil worldwide[5]and requires large N inputs(150–250 kg N ha-1year-1)to support growth and development.During vegetative growth,it requires strong N absorption efficiency[5,6].However,rapeseed has poor N use efficiency(NUE),with only 50% of assimilated N found in its seeds[6].This deficit may be due to the detachment of senescent leaves before N nutrients have been fully transferred to sink organs[7].In addition,rapeseed exhibits photosynthetic organ succession during growth[8].The silique wall becomes the major photosynthetic organ after final flowering,given that many leaves fall thereafter.The CO2fixation rate in the silique wall of oilseed rape is up to 35%of that in leaves[9],an amount higher than the photosynthetic capacity of other plants[10].Approximately 48%of total N cycling through rapeseed occurs in mature pods[11].Ultimately,the source-to-sink translocation of N,especially amino acids,affects sink development and seed yield.According to15N labeling experiments,20%of N stored in pea seeds is allocated from the adjacent pod walls during the seed-filling stage[12],contributing to increase the seed yield and protein content.N remobilized from vegetative tissue accounts for 50%–90% of the total N in mature seeds,suggesting that the re-transport of N stored in plants is a major N source for grain[6].Accordingly,manipulation of N allocation and translocation from senescent leaves and silique walls to younger leaves and grains is a target strategy for increasing rapeseed yield and NUE.

In most plant species,amino acids are the dominant organic N form translocated from source organs,such as roots and senescent leaves,to sink organs,such as flowers,fruits,and seeds[13].Great increases in NUE could be achieved by manipulation of amino acid allocation from vegetative to reproductive tissues[14–17].Transmembrane amino acids transport and source-sink translocation are regulated by membrane-localized amino acid transporters,and many of the specific roles of those transporters have been identified[18,19].In Arabidopsis,amino acid permease 1(AtAAP1)participates in the absorption and embryo loading of neutral amino acids,influencing the final yield[20].Transgenic AAP1 pea plants showed stronger phloem loading in addition to increased biomass and seed yield[21].AAP1 overexpression in rice also promoted tiller number and grain yield[2].AtAAP6 functions in the xylemphloem translocation of neutral and acidic amino acids[14],and has been reported to affect the grain protein content of rice[22]and wheat[23].AAP6-overexpressing soybean lines display high N storage under low N environments[24].AtAAP8 is expressed in source-leaf phloem and functions in the source-to-sink remobilization of amino acids,influencing source-leaf physiology and final yield[15].The cationic amino acid transporter(CAT)functions in amino acid transport in plants.AtCAT1 may be involved in long-distance transport of amino acids in the vascular sap and phloem loading of amino acids[25].AtCAT6 is expressed in plant sink tissues,such as lateral root primordia,flowers,and seeds,and participates in supplying amino acids to sink tissues[26].CAT members have been found in rice[27],soybean[28],potato[29],and other crops.Cultivating genotypes with a high capacity for N allocation mediated by amino acid transporters could be an effective strategy for increasing NUE and yield in rapeseed.

The aim of the present study was to investigate the effects of amino acid translocation from source to sink tissues on rapeseed yield and NUE using a high-and a low-NUE rapeseed genotype.Allocation of amino acids from senescent leaves and silique walls to younger leaves and seeds regulated by genetic variation influenced final yield and NUE,particularly under N-deficient treatment.These results lay a foundation for mining and functional verification of N-efficiency genes,with practical significance for improvement of rapeseed crop production and NUE.

2.Materials and methods

2.1.Plant materials and experimental design

A high-NUE(H73)and a low-NUE(L12)rapeseed cultivar were used for the experiment and cultivated as previously described[30]with some modifications.

A hydroponic experiment was conducted in a phytotron(MGC-800HP-2;Bluepard Test Equipment Company,Shanghai,China)at Hunan Agricultural University,Changsha,China.The culture conditions were as follows:daytime temperature,24°C;nighttime temperature,22 °C night;photoperiod,16 h light/8 h dark;light intensity 300–320 μmol m-2s-1;relative humidity,70%.Seeds were germinated on gauze soaked with deionized water,grown for 7 d,transplanted into black plastic containers.Seedlings were hydroponically grown in 6 mmol L-1NO3-solution for 10 d and then transferred to either fresh 6 mmol L-1NO3-(moderate N)or 0.1 mmol L-1NO3-(low N)solutions for another 5 d.The moderate N nutrient solution used for both genotypes consisted of 2.0 mmol L-1KNO3,2.0 mmol L-1Ca(NO3)2·4H2O,1.0 mmol L-1KH2PO4,2.0 mmol L-1MgSO4,0.05 mmol L-1Fe-EDTA,and micronutrients(stock solution concentrations:50 mmol L-1H3BO3,9 mmol L-1MnCl2,0.8 mmol L-1ZnSO4,0.3 mmol L-1CuSO4,and 0.1 mmol L-1Na2MoO4).When plants were treated with 0.1 mmol L-1NO3-,the nutrient solution contained 0.1 mmol L-1KNO3,1.9 mmol L-1KCl,and 2 mmol L-1CaCl2,and the concentrations of KH2PO4,MgSO4,Fe-EDTA,and micronutrients in the nutrient solution remained constant.The nutrient solution(pH 5.8)provided to plants was replaced every five days and the pots were arranged in a randomized complete block design.

A sand culture experiment was conducted in a greenhouse at Hunan Agricultural University.Plants were grown in a plastic barrel filled with perlite and washed in distilled water.To eliminate edge effects,the barrel position was changed when new solution was added every 5 d.After seedlings were grown in perlite with 15 mmol L-1NO3-solution for 10 d,Ca(15NO3)2(5.13% atom abundance)was used to replace Ca(NO3)2in solution,and the culture was continued for 30 d.Whole plants were washed with doubly deionized water,transferred to clean perlite,and grown in 15 mmol L-1NO3-solution until flowering.When one fourth of the rapeseed plants were in bloom,the plants were grown in either 15 mmol L-1NO3-(moderate N)or 1 mmol L-1NO3-(low N)until maturity.The concentrations of KH2PO4,MgSO4,Fe-EDTA,and micronutrients in the nutrient solution for plant growth described as above.Plant height,dry biomass per plant,leaf soil plant analyzer development(SPAD)values,and15N contents were recorded.

A randomized block design approach was applied to analyze agronomic traits.The field experiment was conducted in the 2019–2020 season.Five N supply levels(0,60,120,180,and 240 kg N ha-1)were applied,and each treatment was replicated three times.Urea(46% N)was used for N fertilization,and a basal fertilizer,overwintering fertilizer,and bolting fertilizer were applied in a ratio of 6:2:2.Calcium magnesium phosphate fertilizer(12% P2O5),potassium chloride(60% K2O),and borax(95%H3BO3)provided phosphorus,potassium,and boron nutrition,respectively,and they were applied once before transplanting the seedlings.The applications were 90 kg P2O5ha-1,120 kg K2O ha-1and 1.5 kg B ha-1,respectively.Soil samples for analysis were taken from the research field at a depth of 0–20 cm before sowing,including organic matter(34.57 g kg-1),total N(1.72 g kg-1),total P(2.85 g kg-1),total K(12.60 g kg-1),alkali-N(156.26 mg kg-1),Olsen P(58.62 mg kg-1),available K(112.38 mg kg-1),and pH(5.69).Each plot(5×3 m)contained 120 plants(60 L12 and 60 H73).Dry biomass per plant,seed yield,total siliques per plant,secondary branches per plant,total seeds per silique,and total N content were recorded at harvest.Plant NUE was estimated based on seed yield relative to total shoot N[2].

2.2.15N tracing

Tracing was performed in hydroponically grown seedlings and sand-cultured pods.Sand-cultured plants were cultured to the middle pod stage without15N.Forty similar pods were selected for each plant and evenly coated with 1.5%(15NH4)2SO4(99.9%atom abundance)at 10:00 AM.After 5 and 25 d,the samples were divided into pericarp,grain,and septum.In hydroponically grown seedlings 5.13%atom abundance of Ca(15NO3)2was used to replace Ca(NO3)2for labeling prior to N stress treatment.After 10 d,the roots were washed with 0.1 mmol L-1CaSO4for 1 min with deionized water.Seedlings were transferred to either fresh 6 mmol L-1NO3-(moderate N)or 0.1 mmol L-1NO3-(low N)for 5 d.Old leaves,young leaves,stems,and roots were harvested after washing with distilled water.

Samples were dried at 105 °C for 30 min and at 60 °C to constant weight.They were then ground to homogenous fine powder with a TissueLyser-48(Shanghai Jingxin Industrial Development Co.,Ltd.,Shanghai,Hunan,China).A 2.0-mg subsample was used to measure15N abundance with aMAT253 isotope ratio mass spectrometer(Thermo-Fisher Scientific,Waltham,MA,USA)coupled with a FLASH 2000 elemental analyzer(Thermo-Fisher Scientific).

2.3.Nitrate,ammonium,and amino acid soluble protein content

Old and young leaves were sampled as previously described[30].The samples were soaked in distilled water and heated in a boiling water bath for 30 min.NO3-content was measured at 410 nm as previously described[32].For determination of NH4+content in leaves,samples were extracted with deionized water for 30 min.The NH4+content was estimated using indophenol blue colorimetry at an absorbance wavelength of 630 nm[31].

Total amino acid content was determined using a micro amino acid content assay kit(Beijing Solarbio Technology Co.,Ltd.,Beijing,China),as previously described[33].A 0.1 g leaf sample was ground to homogeneity with 1 mL of extraction buffer using a TissueLyser-48 and incubated at 100 °C for 15 min with shaking(1000 r min-1).Following centrifugation at 10,000×g for 10 min at 4°C,the absorbance of the supernatant at 570 nm was measured following the manufacturer’s instructions.Soluble protein content in phloem exudates was quantified according to the instructions of the NanoOrange kit(Invitrogen,Carlsbad,CA,USA).

2.4.Collection of leaf phloem exudates

Phloem exudates were collected from 15-day-old plants,as previously described[34].The exudates were collected from a pair of cotyledons per plant,and the leaf was transferred into a tube containing 360 μL of 20 mmol L-1EDTA(pH 7.5)for 2 h.Following centrifugation at 13,000×g at 4 °C for 10 min,the supernatant was used for amino acid analysis.

2.5.Metabonomic analysis

Fresh plants were sampled,washed with distilled water,blotted dry,snap-frozen in liquid N,and stored at-80 °C.The young and old leaves of each plant were taken as a sample,and five or six samples were collected.A 25-mg sample was weighed into a tube and extracted in 500 μL of acetonitrile:methanol:water(2:2:1)containing an internal standard.After 30 s of vortexing,the samples were homogenized at 35 Hz for 4 min and sonicated in an ice-water bath for 5 min before incubation at-40 °C for 1 h,and centrifuged at 12,000×g at 4 °C for 15 min.The supernatant(250 μL)was transferred to a tube and dried in a vacuum concentrator at 37 °C.The dried samples were mixed in 200 μL of 50% acetonitrile by sonication in an ice-water bath for 10 min before centrifugation at 13,000×g at 4 °C for 15 min.The supernatant(100 μL)was used for liquid chromatography–mass spectrometry analysis.Ultra-high-performance liquid chromatographic separation was performed using an Agilent 1290 Infinity series UHPLC System(Agilent Technologies,Santa Clara,CA,USA)equipped with a UPLC BEH Amide column(2.1×100 mm,1.7 μm,Waters).

2.6.Leaf area,root configuration,and chlorophyll content determination

Fresh plants were divided into shoots and roots,as described by Han et al.[30].Leaf area was measured using the CI-202 Laser Area Meter(CID Bio-Science,Camas,WA,USA).The root surface was washed and spread completely in transparent plastic 1000-cm3containers filled with deionized water.The root configuration was determined with WinRHIZO(EPSON Expression 11000XL,Nagano,Japan).

Old and young leaves were collected at 15 d for chlorophyll measurement.Chlorophyll was extracted in 10 mL of absolute ethanol:acetone(1:1)at 4 °C for 48 h in the dark.The absorbance of extracted chlorophyll was estimated at 652 nm using a UV–VIS spectrophotometer(UV-2600,Shimadzu,Kyoto,Japan)[31].Leaf SPAD values were determined with a chlorophyll meter(SPAD-502 plus,Kyoto,Japan).

2.7.Whole-genome and transcriptome analysis

2.7.1.Whole-genome resequencing

Fresh leaves were sampled from 10-day-old plants to isolate genomic DNA(gDNA).Variations in gDNA were distinguished using an Illumina HiSeq 4000 system by Novogene Biotechnology Company(Beijing,China).Genome-wide single-nucleotide polymorphisms(SNPs)and insertion/deletions(InDels)distinguishing the two genotypes were identified and characterized following Hua et al.[35].The Damor-bzh gene sequence was used as a reference.

2.7.2.RNA extraction and quantification

Fresh plants were harvested,snap-frozen in liquid N,and stored at-80 °C.Total RNA was extracted with 1 mL TRIzol reagent(Invitrogen),precipitated with an equal volume of isopropanol,washed with 75% ethanol,and dissolved in RNAasefree water.First-strand cDNA was synthesized from total RNA using a HiScript II 1st Strand cDNA Synthesis Kit(+gDNA wiper)(Vazyme,Nanjing,Jiangsu,China)according to the manufacturer’s instructions.Relative gene expression was determined using the ChamQ Universal SYBR qPCR Master Mix(Vazyme)with a pair of gene-specific primers.Quantitative reverse transcription PCR analysis was performed using the StepOnePlus Real-Time PCR Instrument(Life Technologies Holdings Pte Ltd.,Singapore)following the manufacturer’s instructions:95 °C for 30 s,followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s.A melt curve analysis was performed to ensure the primer gene specificity,as follows:95 °C for 15 s,60 °C for 1 min,and 95 °C for 15 s.The BnaEF1-A gene[36]was used as an internal reference,and gene expression levels were calculated using a slightly modified version of the 2-△△CTmethod[37].The gene-specific primers are listed in Table S1.

2.7.3.Transcriptome sequencing

Ten-day-old seedlings were grown hydroponically in 6 mmol L-1NO3-.Half of the seedlings were transferred to 6 mmol L-1NO3-and the other half to 0.1 mmol L-1NO3-solution,where they were allowed to grow for 72 h.A mixture of three plant cotyledons was used as an RNA sample,and three biological replicates were collected from nine plants.The samples were snap-frozen in liquid N for 30 min and stored at-80 °C for high-throughput mRNA transcriptome sequencing using an Illumina Hiseq X Ten platform(Shanghai,China),which generated～6.0 Gb of sequence with 150 bp paired-end reads for each sample.

2.8.Quantification of medium-and long-chain fatty acids and glucosinolate content analysis

Fresh seeds were harvested and dried at 50 °C to constant weight for quantification of medium-and long-chain fatty acids.An approximately 50-mg sample was added to a 2 mL glass centrifuge tube and mixed with 5 mL of methylene dichloride:methanol(1:1).The homogenate was incubated in a water bath at 80 °C for 30 min for methyl esterification.Then 200 μL of the internal standard,1 mL of n-hexane,and 5 mL of water were added and vortexed.The medium-and long-chain fatty acid contents of the supernatant were estimated in an Agilent model 7890-5977 GC–MS system.

Fresh seeds were harvested and dried at 50 °C to a constant weight for glucosinolate measurement.An approximately 100-mg sample was added to a 2-mL glass centrifuge tube,mixed with 1 mL of distilled water,and heated in a boiling water bath for 60 min.Following centrifugation at 10,000×g for 10 min at 4 °C,the absorbance at 505 nm of the supernatant was recorded according to the manufacturer’s instructions.

2.9.Statistical analysis

SPSS 17.0(IBM Corp.,Chicago,IL,USA)was used for one-way analysis of variance and Tukey’s honestly significant difference multiple comparison tests.The results presented here are from one set of plants but are representative of at least two independently grown sets of plants.

3.Results

3.1.Leaf senescence occurred faster but led to higher seed yield in H73 rapeseed during N limitation

During the late growth stage,leaf senescence appeared earlier in old H73 leaves than in L12 leaves,especially under low N(Fig.1A).However,comparison of seed yield and N input revealed that the seed yield of H73 was higher than that of L12 under low N(＜113 kg ha-1).H73 yield increased by 29.1%compared to L12 yield under low N,with more seeds per silique and more secondary branches and total siliques per plant(Figs.1B,S1A–C).No differences in main branch numbers per plant or 1000-seed weight were found(Fig.S1D,E).The grain yield NUE was higher in H73 than in L12 under low N(Fig.1C).H73 outperformed L12 in terms of oil content and quality,showing higher total fatty acid and total polyunsaturated fatty acid contents and lower glucosinolate contents(Fig.S2).Thus,in H73,earlier leaf senescence in the late growth stage resulted in higher seed yield and NUE under low N than L12,without a reduction in edible oil quality.

3.2.Old-leaf senescence in H73 promoted N remobilization from older to younger leaves during the vegetative growth stage during N limitation

The hydroponic experiment showed that no significant differences in growth were observed between the two varieties under moderate N,including biomass,leaf area,total root length,total root volume,total root surface area,total root tips,and forks(Figs.2A,S3A–G).However,compared to L12,old leaves of H73 displayed severe leaf senescence(Fig.2A)and reduction in chlorophyll content(Fig.S3H)under N-deficiency.Similar total N accumulation and N allocation in shoots and roots following both N treatments prompted us to investigate N allocation in older and younger leaves(Fig.S4A,B).After exposure to N deficiency for 5 d,N content in older leaves of H73 was lower than that of L12(Fig.S4C).15N content of older H73 leaves gradually decreased with the continuation of low-N stress and was much lower than that of leaves of L12 plants(Fig.2B).Consequently,the total15N loss rate and total N reduction in older H73 leaves increased relative to that in L12(Fig.S5A,B).Relative to its values at the start of the experiment,the15N content in younger H73 and L12 leaves decreased by 69.6% and 79.2%,respectively,whereas it decreased by 64.0% and 69.8% in H73 and L12 stems,respectively(Fig.S5C,D).No clear differences in15N content were detected in the roots between genotypes(Fig.S5E).Thus,accelerated leaf senescence in H73 promoted the remobilization of N from older to younger leaves under low N.

3.3.Amino acids were the main form of transport from older to younger leaves during the vegetative growth stage during N limitation

Fig.1.Seed yield and N use efficiency in L12 and H73 under five N applications in the field.(A)Morphological analysis of L12 and H73 under 0,60,120,180,and 240 kg N ha-1 at pod stage.(B)Nonlinear regression model between seed yield and nitrogen input in L12 and H73 under five N applications.(C)Plant NUE,calculated based on seed yield relative to total shoot N(n≥3).Values are means±SE.Significant differences at the same N supply level are indicated by asterisks(ANOVA;*,P≤0.05;**,P≤0.01).DW,dry weight.

Fig.2.Source-to-sink N allocation in L12 and H73 rapeseed plants.Fifteen-day-old plants grown under moderate N(6.0 mmol L-1 N)and low N(0.3 mmol L-1 N)were examined.(A)Phenotype of L12 and H73 at 15 days after transplanting seedlings.(B)15N content in old leaf during treatment periods(n≥3).Heatmap of hierarchical clustering analysis of metabolomic differential metabolites for young leaf(C)(n≥6)and old leaf(D)(n≥5)under N starvation.Red represents higher and blue lower content of differential metabolites.(E)Heat map of expression of genes involved in amino acid transport from source to sink tissues,including genes encoding cationic amino acid transporter(CATs)and amino acid permeases(AAPs)(n≥3).Red represents up-regulated and blue down-regulated gene expression.Content of soluble protein(F)and amino acids(G)in phloem exudates of old leaves treated with low N(n≥5).Amino acid content(H)and amino acid accumulation(I)in old leaves(O)and young leaves(Y)under several N supply levels(n≥5).Values are means±SD.Significant differences at the same N supply level are indicated by letters or asterisks(ANOVA;*,P≤0.05;**,P≤0.01).

Metabonomics was used to evaluate which forms of N contributed to remobilization from older to younger leaves under low-N.Most organic N in younger leaves,including most amino acids and their derivatives,was lower in H73 than in L12 under moderate N(Fig.S6A);however,the contents of most differential metabolites behaved oppositely under N starvation(Fig.2C).Under N shortage for 3 d,most differential metabolites in older leaves,including most of the amino acids and their derivatives,were more abundant in H73 than in L12(Fig.2D).Relatively few differential metabolite concentrations were observed in older leaves under sufficient N(Fig.S6B).Transcriptome sequencing of older H73 and L12 leaves showed that more differentially expressed genes(DEGs)were induced under low-N than under N sufficiency,with totals of respectively 11,821 and 4546 DEGs identified(Fig.S7A).KEGG pathway analysis showed that the DEGs between Ndeficient and N-sufficient conditions in L12 were enriched in pathways involved in autophagy(Fig.S7B),and the expression of AuTophaGy(ATG)genes was upregulated under low N(Fig.S7C).However,the N-supply-associated DEGs in H73 were enriched in pathways involved in amino acid metabolism(Fig.S7D)and their transfer from source to sink organs,which included upregulation under low N of the amino acid transporter genes BnaCAT1,Bna-CAT4,BnaCAT6,BnaAAP1,and BnaAAP6(Fig.2E).

In phloem exudates from older leaves,the contents of soluble protein and free amino acids were higher in H73 than in L12 under low N(Fig.2F,G).As expected,compared with that in L12,the total free amino acid content of H73 was reduced in older leaves,as they accumulated instead in younger leaves(Fig.2H,I).To confirm that amino acids were the main form of N transport from older to younger leaves,the nitrate and ammonium contents in both older and younger leaves were also measured.No differences in nitrate content were observed between H73 and L12 in either older or younger leaves,irrespective of N treatment(Fig.S8A).However,the content of ammonium was higher in H73 than in L12 older leaves after 5 d of low N(Fig.S8B).Thus,organic N in amino acids,rather than inorganic N,such as nitrate and ammonium,was the primary substrate for N translocation from older to younger leaves in H73 under low N.

3.4.Old-leaf senescence in H73 promoted N remobilization from vegetative organs to pods during the reproductive stage during N limitation

To investigae the response of the two genotypes to N deficiency at the reproductive stage,plant height,biomass,and leaf SPAD values were investigated in greenhouse pot experiments.The plant height of H73 was lower than that of L12 during all growth stages(Fig.S9A).Similarly,plant biomass at the pod and mature stages was lower in H73 than in L12(Fig.S9B).No differences were observed in younger-leaf SPAD values between genotypes at different growth stages for either N treatment(Fig.S9C).Slight differences in older leaves were observed between genotypes under moderate N(Figs.3A,S9D).Senescence(Fig.3B)and a reduction,relative to L12,in the SPAD value(Fig.S9D)appeared at the pod stage in the older leaves of H73 under low N.We assessed the dynamic changes in N content(traced by15N)for both genotypes and observed that15N content increased over time in older H73 leaves(0–10 d)and then decreased dramatically when plants were grown under N-deficiency for another 20 d(Fig.3C).In plants grown under low N for 30 d,the15N contents of pods,stems,and younger leaves were higher in H73 than in L12,especially in pods and younger leaves(by 55.2% and 28.9%,respectively)(Fig.3D–F).Thus,senescence of older H73 leaves contributed to the remobilization of N from older leaves to younger leaves and pods during the reproductive stage under low N.

3.5.Amino acids were the main form of transport from vegetative organs to pods during the reproductive stage during N limitation

To identify the primary form of N in its transport from vegetative tissues to pods at the reproductive growth stage,we measured amino acid content and transporter activity.Compared to that in L12,BnaA9.AAP1 and BnaC9.AAP1 expression in older H73 leaves was lower following low-N treatment for 3 d.However,under low N for 10 or 30 d,BnaA9.AAP1 and BnaC9.AAP1 expression were significantly upregulated in H73,especially under N deficiency(Fig.3G,H).Consequently,amino acid contents were higher at 10 d,but reduced when plants were exposed to low N for 30 d(Fig.3I).The level of expression of BnaA9.AAP1 in younger leaves showed a trend similar to that in old leaves(Fig.S10A).The levels of expression of BnaC9.AAP1 in younger H73 leaves were significantly lower than those in L12 under moderate N conditions for 3 d or 10 d;however,opposite results were observed under Ndeficiency.After treatment for 30 d,compared to L12,the expression of BnaC9.AAP1 in H73 was up-regulated under moderate N and down-regulated under N-deficiency(Fig.S10B).After 35 d of low-N treatment,the expression of BnaA9.AAP1 in silique wall of H73 was significantly down-regulated by 42.2% under moderate N in comparison with those in L12.In contrast,the levels of expression were significantly upregulated by 79.2% under N deficiency(Fig.S10C).The levels of expression of BnaC9.AAP1 in the silique wall of H73 showed trends similar to those of BnaA9.AAP1(Fig.S10D).However,no significant differences between the two genotypes were detected in pods,stems,or younger leaves when they were grown under N deficiency for 30 d(Fig.S11A–C).The nitrate contents of stems,younger leaves,and pods under N limitation showed no differences after 30 d of low N(Fig.S12B–D);however,the nitrate contents of older leaves were higher in H73 than in L12(Fig.S12A).Thus,senescence of vegetative organs in H73 contributed to accelerated N remobilization to plant pods under low N and confirmed that amino acids,carried by amino acid transporters,represent the main N transport form and that the contribution of N allocation in old leaves is greater.

3.6.Higher proportion of N is allocated to seeds from the silique wall,increasing N harvest index in H73 during the pod stage

To investigate the effects of N partitioning from the silique wall to seeds at the pod stage,N allocation was examined in the seeds,silique wall,and diaphragm.Total15N accumulation in seeds was higher in H73 than in L12 after 25 d of tracing under moderate and low N(Fig.4A).15N abundance was higher in the pod wall of H73 than in that of L12 after 5 d;however,after 25 d,15N accumulation was lower in H73 than in L12(Fig.4B).No differences between H73 and L12 in diaphragm15N accumulation were detected(Fig.4C).The proportion of total15N in seeds was higher in H73 than in L12,whereas the inverse was found in the silique wall after15N tracing for 25 d under both N treatments(Fig.4D).The field experiment showed that the proportion of total N in the seeds increased gradually with pod maturity and was higher in H73 than in L12,whereas the reverse was observed in the silique wall and diaphragm(Fig.4E).As a consequence,a higher N harvest index was obtained for H73 than for L12(Fig.S13).Dynamic N-content analysis of the silique wall and diaphragm revealed that N content in H73 was higher than that in L12 at 35–42 d after flowering;however,the inverse was observed at 49–63 d after flowering(Fig.S14B,C).Unexpectedly,N content in grain was not affected by genotype at 63 d after flowering(Fig.S14A).The proportion of grain biomass exhibited similar results to that of total N and was significantly higher in H73 than in L12(Fig.S15),which facilitated a higher harvest index in H73(Fig.S16).Thus,ample N was allocated from the silique wall to the seeds in H73,generally increasing N harvest index and harvest index.

Fig.3.Morphological analysis and 15N allocation from vegetative organs to pods at the reproductive stage.(A)and(B),Morphological analysis of L12 and H73 rapeseed plants under moderate N treatment(A)and low N treatment(B)for 30 d.(C–F)15N content in old leaf(C),young leaf(D),stem(E),and pod(F)during different N deficiency stress time(n≥3).(G)and(H),Relative mRNA expression level of BnaA9.AAP1 and BnaC9.AAP1 in source leaf.(I)Content of amino acids in old leaf during N deficiency treatment periods(n≥4).Values are means±SD.Significant differences at the same N supply level are indicated by different letters(ANOVA;*,P≤0.05;**,P≤0.01).DW,dry weight.

Fig.4.N transport from silique wall to seeds during pod stage.(A–C)Total 15N content in seed(A),silique wall(B),and diaphragm(C)for 5 d and 25 d after pericarp was coated with 15N in greenhouse(n≥3).(D)Proportion of total N in seed,silique wall,and diaphragm after flowering for 35,42,49,56,and 63 d after flowering in the field(n≥3).Values are means±SE.Significant differences at the same N supply level are indicated by asterisks(ANOVA;*,P≤0.05;**,P≤0.01).

3.7.Amino acids were the main form of transport from silique wall to seeds during the pod stage

To evaluate the effects of amino acids on the transport of N from the silique wall to grains,the amino acid content in the grains,silique wall,and diaphragm was measured during dynamic pod growth.Under several N fertilization supply levels,amino acid content in the three tissues decreased 49 d after flowering(Fig.5A–C).The amino acid content in the silique wall and diaphragm was much higher in H73 than in L12 after flowering for 35–42 d,whereas the inverse was observed after flowering for 49–63 d(Fig.5B,C),in agreement with the variation in N content(Fig.S14B,C).Measurement of nitrate content confirmed that it was not the major form of N allocation from the silique wall to the seeds(Fig.S17).These results showed that amino acids are the primary form of N involved in N translocation from the silique wall to the seeds.

3.8.Genomic variation may contribute to the allocation of amino acids from the silique wall to seeds in H73 during the pod stage

To determine whether observed differences were caused by genomic variation,we performed whole-genomic resequencing of H73 and L12,generating 75.85 Gb of clean reads with a mean depth of 30×.Genome-wide genetic variants,including SNPs and InDels,were detected by comparing the genomes of H73 and L12 with a reference genome(Damor-bzh).Based on the sequencing data,a total of 1,994,598 SNPs and 412,402 InDels were found between H73 and Damor-bzh,whereas 1,809,220 SNPs and 382,330 InDels were found between L12 and Damor-bzh(Tables S2,S3).KEGG pathway enrichment analysis of the genomic variations showed that the DEGs in the two genotypes were enriched mainly in amino acid metabolism,protein processing in endoplasmic reticulum,and ABC transporters(Fig.6A).The polymorphic genes,BnaC1.CAT4(BnaC01g36510D),BnaC6.AVT6E(BnaC06g40470D),BnaC7.BAT1(BnaC07g48770D),BnaC5.LHT2(BnaC05g21130D),BnaC7.CAT6(BnaC07g34120D),BnaC6.AVT6B(BnaC06g12470D),BnaA6.BAT1(BnaA06g34530D),BnaA9.AAP1(BnaA09g14700D),BnaC3.PUT4(BnaC03g38040D),and BnaC9.PUT3(BnaC09g49830D)were involved in the process of organic N transport,which included transporter activity,organic substance transport,and N compound transport(Fig.6B;Table S4).The process of organic N metabolism included protein metabolism,cellular N compound biosynthesis,and cellular amino acid metabolism(Fig.6C).The coding sequences of BnaC1.CAT4,BnaA6.BAT1,and BnaC3.PUT4 contained SNP variations between the two genotypes of rapeseed(Table S5).In addition,InDel variations were detected in BnaC6.AVT6E,BnaC7.BAT1,BnaC5.LHT2,BnaC7.CAT6,BnaC6.AVT6B,BnaA6.BAT1,BnaA9.AAP1,and BnaC9.PUT3(Table S6).BnaA6.BAT1 and BnaA9.AAP1 were found to harbor SNP and InDel variations in the promoter region that might account for the differences in organic nitrogen transport between the two genotypes rapeseed.To further investigate whether genomic differences in promoter region affected gene expression at the transcriptional level,transcriptome data from old leaves at the seedling stage showed that the fragments per kilobase of transcript per million mapped reads(FPKM)values of BnaA6.BAT1 and BnaA9.AAP1 in H73 were higher than those in L12,especially under low N(Fig.S18).Thus,the H73 genome harbored more genomic variations and the genomic variation could affect the gene expression levels of H73 and L12,thereby contributing to N remobilization from source to sink organs during the pod stage.

Fig.5.Amino acid remobilization from pod shell to seeds during pod stage in the field.(A–C)The content of amino acids in seed(A),silique wall(B),and diaphragm(C)after flowering for 35,42,49,56,and 63 d in the field(n≥3).Data are means±SE.Significant differences at the same N supply level are indicated by asterisks(ANOVA;*,P≤0.05;**,P≤0.01).DW,dry weight.

Fig.6.KEGG pathway enrichment analysis of genomic variations and molecular functions in H73 and L12.(A)KEGG pathway enrichment analysis of the genomic variations distinguishing H73 and L12.(B)Functional categories of polymorphic genes potentially involved in the process of organic nitrogen transport.(C)Functional categories of polymorphic genes potentially involved in the process of organic nitrogen metabolism.The size of the symbol indicates the number of genes with genomic variants distinguishing H73 and L12,and the rich factor indicates the degree of enrichment of the KEGG pathways.

4.Discussion and conclusions

4.1.Source leaf senescence promotes the allocation of amino acidsfrom source to sink organs in both vegetative and reproductive stages under N deficiency

Plants require large amounts of N for their growth and development.The amount of N that is assimilated and allocated from source to sink tissues is positively correlated with the rate of fruit or seed development and final crop yield[1,38].In this study,we compared high-NUE(H73)and low-NUE(L12)genotypes of rapeseed.No differences were detected in N uptake or allocation between the shoots and roots of the two rapeseed genotypes at the vegetative stage(Fig.S4A,B),whereas a clean early leaf senescence phenotype appeared in H73 under N deficiency,accompanied by a reduction in N content(Figs.2A,B,S4C).Indeed,N depletion is known[39]to induce leaf senescence and accelerate N recycling and remobilization[39].At the vegetative stage,N reallocation in rapeseed occurs primarily between leaves according to their unique source and sink balance,through a regulated sequential senescence process,and is part of a complex proteolytic degradation system[40].This strategy of recycling endogenous nutrients from senescing leaves is used by plants to support the growth of younger leaves and reproductive organs under N starvation[7].

Proteins are the primary source of N recycling in senescing leaves,contributing to 50%of leaf N translocation[41,42].In plants,N can be redistributed in both inorganic and organic forms.Nitrate,as the major type of inorganic N absorbed by rapeseed,can be translocated after being taken up from the soil[43].Unused nitrate is remobilized throughout the plant via short-distance distribution and long-distance transportation[44].Short-distance distribution of the nitrate process is mediated by members of the chloride channel protein family and the activity of V-ATPase and Vpyrophosphatase in the tonoplast[30].Long-distance transportation relies mainly on the coordination of nitrate transporters(NRTs),including NRT1.5 and NRT1.8 proteins[45,46].In our previous study,we identified two genotypes,one with high NUE(Xiangyou 15)and one with low NUE(814),which support the notion that the efficiency of short-distance nitrate distribution and long-distance nitrate transportation increases NUE[30].However,in most plant species,organic N,particularly in amino acids,is the primary form of N translocation and allocation[13].Interestingly,in contrast to the results of previous studies,we found that N reduction in old leaves was not due to differences in nitrate and ammonium(Fig.S8),which are the major transport forms of inorganic N.Instead,metabonomics revealed that organic N content was higher in old H73 leaves than in old L12 leaves after 3 d of low-N stress(Fig.2D),possibly because the early stages of protein degradation in the old leaves released abundant amino acids and their derivatives for transport to younger leaves and reproductive organs,thereby increasing organic N content in newly developing leaves(Fig.2C,H,and I).

Protein degradation-derived amino acids are a dominant N source for source-to-sink translocation[47].In a previous study[48],50–70% of seed proteins were derived from amino acids transported from source organs in wheat.Amino acid remobilization from source to sink relies on amino acid transporters.Transcriptome analysis revealed that N-deficiency-induced genes of H73 were involved in the metabolism of amino acids and their transport from source to sink,CAT1,CAT4,CAT6,AAP1,and APP6,which increased the remobilization of amino acids to younger leaves and exacerbated the depletion of N in old leaves under N deficiency(Figs.2E,S7D).This was confirmed by demonstrating that the higher soluble protein and amino acid export ability in phloem sap resulted in reduced amino acid content in old H73 leaves under low-N stress(Fig.2F–I).N translocation may already be activated in source leaves(mature but not senescent)of rapeseed during the vegetative stage[7].The expression of autophagy-related genes(ATG8E,ATG8F,and ATG8I)was upregulated in old(not senescent)L12 leaves under low N(Fig.S7B,C),implying that other processes,such as autophagy,could also be involved in N translocation under N insufficiency[49].

During the reproductive growth stage,the absorption of N by plant roots is partially or completely inhibited,and the N absorbed and stored in plants at the vegetative stage is redistributed from older leaves to younger leaves and seeds,thereby meeting the N nutrient requirements for plant growth during the reproductive stage[39,50,51].In the present study,the senescence of old leaves appeared in H73 during the reproductive growth stage in an Npoor environment,an observation consistent with the phenotype of the vegetative stage and the field experiment(Figs.1A,2A,and 3A,B).Consequently,15N abundance in old leaves of H73 rapidly decreased under N limitation,whereas it increased in younger leaves and pods(Fig.3C–F),confirming that N stored in older leaves is translocated via the phloem to developing flowers,pods,and seeds[12].Amino acids,rather than nitrate,were the primary form of N allocated from older leaves to younger leaves and pods during reproductive growth(Figs.3I,S11,and S12).A recent study[52]revealed that the expression of BnaA9.AAP1 and BnaC9.AAP1 was induced under nitrate limitation conditions and that their expression was upregulated in a high-NUE genotype,particularly under N deficiency(Fig.3G,H).In short,N reallocation in plants is generally associated with senescence,and the N in a high-NUE rapeseed genotype,especially the amino acids obtained from old leaves,showed efficient source-to-sink allocation in both the vegetative and reproductive stages under N deficiency.

4.2.Genomic variation may influence high amino acid remobilization from silique wall to seeds during the pod stage

The rapeseed silique is a composite organ,including the silique wall,diaphragm,and seeds,which have physiological,biochemical,and functional differentiations[9].Rapeseed pods have dual source and sink functions for supporting N recycling[8],with approximately 48% of total N recycling detected in mature pods[53],that may contribute to seed yield and plant NUE.N is translocated from all vegetative organs once seed development occurs in pea:30%of the total N stored in seeds is allocated from leaves,20% from pod walls,10% from stems,and 11% from roots[12].Thus,N efficient remobilization from the adjacent pod wall to the seeds is crucial during the seed-filling stage.In this study,the15N tracer experiment and field experiment confirmed that a greater proportion of N was reallocated to seeds from the silique wall in H73 than in L12 during the pod stage(Fig.4).Amino acids rather than nitrate were the primary form of N allocated from the silique wall to seeds during the pod stage(Figs.5,S17).In contrast to the source-to-sink allocation of N during the vegetative and reproductive stages,the remobilization of N from the silique wall to the seeds during the pod stage was determined by genotype and remained unaffected by N levels.Whole-genome resequencing showed that compared to Damor-bzh gene sequence,more genomic variations,including SNPs and InDels,were present in H73 than in L12(Tables S2,S3),and the polymorphic genes,including BnaA9.AAP1,BnaC1.CAT4,BnaC7.CAT6,BnaC6.AVT6E,and BnaC6.AVT6B,were involved in organic N metabolism and transport(Fig.6;Table S4).In Arabidopsis,AtAAP1 functioned in embryo loading of neutral amino acids,and transgenic pea plants exhibited stronger phloem loading and higher yield[20,21].AtCAT4 was located on the vacuolar membrane and participated in the transport of amino acids across the vacuolar membrane[54].AtCAT6 is expressed in sink tissues such as lateral root primordia,flowers,and seeds,and functions in the supply of amino acids to sink tissues of plants[26].ScAVT6 encodes vacuolar acidic amino acid exporter,which is a direct target of the GATA transcription factor,and is involved in amino acid efflux from vacuoles in Saccharomyces cerevisiae,especially under low-N conditions[55,56].Previous study[57]showed that a 29-bp insertion and/or deletion in the OsTCP19 promoter resulted in a differential transcriptional response and variation in tillering response to N among rice varieties,further influencing final grain yield and N use efficiency under low or moderate levels of N,indicating that genetic variation in promoter regions influenced transcriptional response.The present study showed that BnaA6.BAT1 and BnaA9.AAP1 harbored SNP and InDel variations in their promoter regions(Tables S5,S6),and transcriptome data of old leaves at the seedling stage showed that the FPKM values of the two genes in H73 were higher than those in L12(Fig.S18),suggesting that the genomic variations might affect gene expression in H73 and L12,potentially contributing to amino acid remobilization from the silique wall to the seed during growth.

4.3.Amino acid remobilization from source to sink contributes to higher yield and NUE under N deficiency

Efficient amino acid translocation from source to sink is crucial for the growth and development of both vegetative and reproductive organs[58].Biotechnological approaches of regulating whole plant distribution of organic N metabolites,such as amino acids and ureides,have proven successful in increasing yield in several crop species[19].Alteration of amino acid transport increased the number of pods and seeds per plant at the reproductive stage in Arabidopsis[20]and pea[15,16].OsAAP1 overexpression increased tillering and grain production in rice[2].Seed yield of a rapeseed plant is influenced directly by three components:number of siliques per plant,number of seeds per silique,and seed weight[9].Our study also confirmed higher seed yield and NUE in H73 than in L12 owing to the higher number of secondary branches per plant,total siliques,and seeds per plant,under N deficiency(Figs.1B,C,7,and S1A–C).We demonstrated that senescence induced by protein degradation into amino acids in source leaves would be activated by an N-deficiency signal.A large number of amino acids were transported from source leaves to sink organs(including younger leaves,flowers,and pods)in phloem sap,mediated by amino acid transporters,resulting in depletion of N in source leaves and deposition in sink organs.Compared to that in L12,efficient amino acid translocation from source to sink increased N remobilization efficiency in H73 and led to a higher number of secondary branches and total pods per plant under N deficiency,and in turn,increased seed yield and NUE(Fig.7A).

Rapeseed pods can remobilize N from the adjacent silique wall to the seeds during the seed-filling stage[8].Under several N supply levels,15N labeling and field experiments showed that a greater proportion of N was partitioned to seeds from the silique wall in H73 than in L12(Fig.4),possibly increasing the number of seeds per silique rather than the weight per seed and N content,and finally increasing seed yield and NUE under N-deficiency treatment(Figs.1B,C,7B,and S1B).We observed a higher harvest index and N harvest index in H73 than in L12(Figs.S13,S16).This finding was attributed to the fact that,compared to L12,H73 exhibited a similar biomass but higher yield under N limitation and a similar yield but lower biomass under moderate and high N levels.These findings differed from our previous investigation of the Xiangyou 15 and 814 genotypes,which revealed a similar harvest index for both genotypes owing to the simultaneously higher seed yield and higher plant biomass in Xiangyou 15[59].

This study provides evidence for the importance of amino acid remobilization from source organs to sink organs in rapeseed,which would be an effective strategy for producing high seed yields without increasing N fertilizer use.The high-NUE genotype,H73,exhibited superior phenotypes,particularly under N-deficient treatments,such as increased yield and harvest index,owing to the effective allocation of amino acids from senescent leaves and silique walls to younger leaves and seeds,and the process was mediated by variation in genes involved in organic N transport and organic N metabolism.The high-NUE genotype,H73,had superior edible oil quality.The results of the present study lay a foundation for the mining and functional verification of N-efficient genes,with implications for increasing NUE and yield by manipulation of amino acid remobilization.

Fig.7.A model in which increasing amino acid allocation from source organs to pods and seeds suggests an effective strategy for increasing seed yield and plant NUE under low N.(A)The translocation of amino acids from source to sink organs was compared between H73 and.L12.In H73 source leaves,the senescence induced by degrading protein into amino acids was activated when the signal of N deficiency stimulated the plant.Large number of amino acids were transported from old leaves to the sink organs(such as younger leaves,flowers,and pods)via phloem sap,with the assistance of amino acid transporters,resulting in depletion of N in the source leaves and further accelerating senescence.Compared with that in the low-NUE genotype,efficient amino acid translocation from source to sink organs increased N remobilization efficiency(NRE)and contributed to a higher number of secondary branches and total pods per plant in the high-NUE genotype under N deficiency,ultimately increasing seed yield and NUE.(B)Translocation amino acids from silique wall to seeds was compared between H73 and L12.Gene variations influencing amino acid transport and metabolism supported the finding that a greater proportion of N was partitioned to seeds from silique walls in the high-NUE genotype than in the low-NUE genotype,possibly increased the number of seeds per silique without reducing N content and ultimately increasing seed yield and NUE under N deficiency.

This study supports the notion that increasing amino acid allocation from source organs to pods and seeds offers an effective strategy for increasing seed yield and plant NUE.Red arrows indicate increasing effects and blue arrows decreasing effects.Gray arrows represent the processes of metabolism and transport.Blue dotted lines represent the transport route of amino acids from source organs(such as older leaves and silique walls)to sink organs(such as younger leaves,pods,and seeds).

CRediT authorship contribution statement

Guihong Liang:provided data curation,formal analysis,visualization,validation,investigation,and writing-original draft.Yingpeng Hua:made the software and supervision.Zhenhua Zhang:made the conceptualization and provided funding acquisition,project administration and resources.Haifei Chen,Jinsong Luo,Haixing Song,and Hongkun Xiang:provided methodology assistance and writing-review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We thank Prof.Chun-Yun Guan(Hunan Agricultural University,Changsha)for providing rapeseed materials.We thank Shanghai Applied Protein Technology(Shanghai,China),Biotree(Shanghai,China),and Novogene Biotechnology Company(Beijing,China)for valuable technical help.This study was supported by the National Natural Science Foundation of China(U21A20236,32072664),Natural Science Foundation of Hunan Province(2021JJ0004),China Agriculture Research System,and the Hunan Postgraduate Scientific Research Innovation Project(CX20190505).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2022.05.011.