Response of leaf carbon metabolism and dry matter accumulation to density and row spacing in two rapeseed (Brassica napus L.) genotypes with differing plant architectures

Jie Kui,Xioyong Li,Jinli Ji,Zhen Li,c,Yn Xie,Bo Wng,Gungsheng Zhou,*

a MOA Key Laboratory of Crop Ecophysiology and Farming System in the Middle Reaches of the Yangtze River,College of Plant Science and Technology,Huazhong Agricultural University,Wuhan 430070,Hubei,China

b Oil Crops Research Institute of Chinese Academy of Agricultural Sciences,Wuhan 430070,Hubei,China

c College of Agriculture,Jinhua Polytechnic,Jinhua 321007,Zhejiang,China

d Haimen Municipal Guidance Station for Crop Cultivation and Technology,Haimen 226100,Jiangsu,China

Keywords:Rapeseed Density Row spacing Dry matter Leaf carbon metabolism

ABSTRACT Biological yield indicates the potential for increasing yield.Leaf carbon metabolism plays an important role in the biomass accumulation of rapeseed (Brassica napus L.).Field experiments with the hybrid HZ62(with a conventional plant architecture)grown in 2016–2017,and HZ62 and accession 1301(with a compact plant architecture)grown in 2017–2018 were conducted to characterize the physiological and proteomic responses of leaf photosynthetic carbon metabolism to density and row spacing configurations.The densities were set at 15×104 ha-1 (D1),30×104 ha-1(D2),and 45×104 ha-1 (D3)(main plot),with row spacings of 15 cm(R15),25 cm(R25),and 35 cm(R35)(subplot).Individual and plant population biomass accumulation was greatest at R25,R15,and R15 for D1,D2,and D3,respectively,for both genotypes.In comparison with D1R25,the individual aboveground biomass of HZ62 decreased by 60.2%,whereas the population biomass increased by 31.9%,and the individual biomass of genotype 1301 decreased by 54.0%and the population biomass increased by 53.9%at D3R15.Leaf carbon metabolic enzymes varied between genotypes at flowering stage.In contrast to D1R25,at D3R15 the activities of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) and sucrose phosphate synthase (SPS) and the contents of starch,sucrose and soluble sugars in leaves were significantly decreased in HZ62 and increased in genotype 1301.The activities of fructose-1,6-bisphosphatase (FBPase) decreased,in consistency with the abundance of fructose-bisphosphate aldolase in HZ62.In contrast,sucrose synthase(SuSy)activity appeared to decrease in both genotypes,but a significant increase in abundance of a protein with sucrose synthase was found in the 1301 genotype by proteomic analysis.With increased density and reduced row spacing,the expression of most key proteins involved in carbon metabolism was elevated,and enzyme activity and carbon assimilate content were increased in 1301,whereas HZ62 showed the opposite trend,indicating that the compact plant type can accumulate more population biomass with denser planting.

1.Introduction

Rapeseed (Brassica napus) is one of the most important oilseed crops worldwide.It is an excellent raw material for edible oil and biodiesel production [1].The rapeseed crop can also be used for forage before seed production and can replace winter cereals in crop rotations,allowing the incorporation of carbon into the soil owing to the abundance of its crop residues [2].The rising world population and renewable energy policies are driving a surge in oilseed demand which is predicted to continue,and production may need to double by 2050 to satisfy these projections[3].Winter rapeseed is widely cultivated along the Yangtze River in China.Rice–rapeseed rotation is one of the main two-crop-per-year cropping systems in this region.The prolonged growth duration of rice delays the sowing of rapeseed and is unfavorable for its growth[4].Rapeseed yield is affected by genotype,environment,agricultural practice,and their interactions [5,6].Advances in breeding and agronomy have driven yield improvements of 20–40 kg ha-1year-1in most growing areas[7].New cultivars and agricultural practices are urgently needed to promote the development of rapeseed production in this area.

Selecting plants with ideal architecture is crucial for crop domestication and improvement.A new plant ideotype of tomato with more spacious canopy architecture due to long internodes and long and narrow leaves led to an increase in crop photosynthesis of up to 10% [8].There have been several studies [9–11] in major crops to identify the mechanisms that control plant architecture.Plant height,branch length,branch angle,length of main inflorescences,leaf angle,and branch number per plant define plant architecture in rapeseed,and affect seed yield components such as pods and seeds per plant [12–14].Besides the genotype effect on plant architecture,several management factors can be identified,including rotation length,cultivation practice,fungicide use,fertilizer use,and sowing date[7,15].Plant density affects both quality and quantity of light penetration into the canopy.Plant density could be reduced with little penalty for seed production per area,given that short-cycle spring rapeseed genotypes expressed a strong vegetative and reproductive plasticity at individual level when plants grew under well-watered and fertilized conditions[16].However,with dense planting of winter rapeseed,plants begin to shade one another,affecting especially the older leaves lower in the canopy as plant growth progresses,leading to reduction in canopy light capture [17].Thus,one source of rapeseed yield variation may be associated with changes in the light environment experienced by plants that affect biomass production and its allocation to harvestable grain [17].Density and row spacing are effective agricultural practices for controlling individual and population growth of rapeseed via the improvement of plant architecture,effective use of light energy,and coordination of crop yield and lodging resistance [18,19].Research on other crops has also shown that optimized planting density and row spacing can reduce individual competition and improve population structure,leading to high yields of rice [20],wheat [21],corn [22–24],soybean [25],and cotton [26,27].

Photosynthate supply strongly influences pod and seed development [15].Leaves are initially the main photosynthetic plant structures fixing food for plant growth.The leaf area index of rapeseed starts to decrease shortly after first flower.At full flower,the stems become the major photosynthetic structures,although the leaves are still important.At the beginning of ripening,the pod walls and stems account for the majority of photosynthesis[28,29].The green pod walls and stems photosynthesize actively but not as efficiently as leaves,as their stomatal density is not as high [28].Removal of leaves reduced seed yield and oil content[30].

The anthesis period,which may last from two to six weeks in rapeseed,is considered [31] to be a critical period for yield determination,and leaves and stems are the principal sites of assimilation,taking up 46% and 41% of14CO2,while pods take up only about 5% at flowering [32].Major et al.[33] investigated the changes in photosynthate source–sink relationships during the development of rapeseed using14CO2.Rapeseed plants were allowed to assimilate14CO2through lower and upper stem internodes,leaves,and pods.Lower stem internodes exported few14Clabeled assimilates.Lower leaves exported assimilates to the roots,whereas upper stems and leaves exported primarily to seeds and pods.Pods were also sinks for labeled assimilates from upper stem internodes and leaves.These results suggest that the redistribution of carbon assimilates synthesized by and stored in the upper leaves of rapeseed at flowering stage influence pod and seed development in dense plantings,though they have lower LAI than lower leaves.

During photosynthesis,carbon dioxide is converted to sugars,or carbon assimilates.Most of these sugars are used to form the cell walls of the plant,such as lignin and cellulose,and used mainly for crop morphogenesis,which is closely associated with stem lodging resistance.Excess sugars are stored in the plant as nonstructural carbohydrates (NSCs),such as starch,sucrose,glucose,fructose,and fructan,participating in crop metabolism and yield formation [34].Increasing the output and reuse of NSCs in source organs is a vital measure for achieving high yield of crops [35].Enzyme activity is one of the major factors affecting photosynthetic rate,influencing the accumulation of carbon assimilates and the balance between source and sink.Ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) is a key plant enzyme controlling photosynthetic carbon metabolism and photorespiration [36].Fructose-1,6-bisphosphatase (FBPase) governs sucrose formation,and its activity directly influences photosynthetic efficiency and assimilate accumulation [37].Sucrose phosphate synthase (SPS) regulates the distribution and transformation of photosynthetic products between sucrose and starch,influences the balance between source and sink,and participates in carbon assimilation metabolism and cell differentiation.Highly expressed SPS elevates the distribution rate of sucrose and increased the formation of cotton yield and quality [38].Sucrose synthase (SuSy) is a type of glycosyltransferase widespread in plants,and involved in diverse metabolic pathways,primarily decomposing sucrose,a product of leaf photosynthesis,and promoting the formation of UDPG[39].In rapeseed,Pn(the photosynthetic rate),the activities of Rubisco and adenosine diphosphate glucose pyrophosphorylase (AGPase),SuSy,SPS,and the contents of sucrose and starch at the flowering stage increased with increasing N application,and the activities of AGPase,SuSy,and SPS at flowering were highly positively correlated with yield[40].

Agricultural practices are effective in regulating carbon assimilates and translocation at physiological level in crops.Proteins are among the fundamental functional components of a biological system.Metabolic processes and cellular signaling are triggered by protein–protein interactions,post-translational protein modifications,and enzymatic activities [41].Proteomic study is achieving recognition as a reliable and reproducible high-throughput approach to investigating biological processes under various environments,in particular under abiotic stress conditions in rapeseed[42–44].It is based on the systematic analysis and documentation of expressed proteins and their study at the functional level [45].Among the proteomics studies and reviews published on Brassica spp.to date [46–49],few differential proteomic studies have focused on the response of B.napus to agricultural practices [50].However,Regulations arising from plant interactions with their environment (such as nitrogen resources),final architecture,and therefore sink-source relations in planta,seem to be globally conserved between Arabidopsis and B.napus [51].

Some proteomic studies of oilseed rape roots suggest different proteomic profiles between genotypes[43].It seems,however,that leaf carbohydrate status and its role in the development of rapeseed plants under different densities and row spacings are currently poorly understood.We hypothesized that (1) the carbohydrate dynamics in leaf tissue of rapeseed are changed by density and row spacing throughout the ontogenetic stages of the plant;(2) differences in leaf carbohydrates depend on differences in activities of carbohydrate metabolic enzymes and protein abundance between two rapeseed genotypes with differing plant architectures;(3) the combination of physiological and proteomic changes associated with carbon assimilates and translocation in leaf tissue contributes to biomass accumulation.To test these hypotheses,we grew rapeseed under field conditions at several densities and row spacings.The aim of the experiment was to identify mechanisms by which density and row spacing arrangements regulate physiological,metabolic and proteomic changes in leaves of two rapeseed genotypes with differing plant architectures,and to correlate changes in photosynthetic activity with plant phenotypic response.

2.Materials and methods

2.1.Experiment design

The experiment was conducted with two genotypes of winter oilseed rape at Huazhong Agricultural University,Wuhan,China during 2016–2018.The preceding crop on the trial site was rice.In 2016 and 2017,the respective soil nutrient concentrations were as follows:84.26 mg kg-1and 86.34 mg kg-1of available nitrogen,14.25 mg kg-1and 16.25 mg kg-1of available phosphorus,and 148.36 mg kg-1and 154.51 mg kg-1of available potassium.

In 2016–2017,the rapeseed hybrid HZ62 (with high yield and moderate lodging resistance) was selected as a test genotype.In 2017–2018,HZ 62 and rapeseed accession 1301 (with low yield and strong lodging resistance) were selected as test genotypes.The two genotypes were similar in growth duration,but differed in plant height,branch height main inflorescence length,branch angle,and branch number [52].Whereas the rapeseed ideotype is a semi-dwarf plant with a height of 120–140 cm,narrow(＜30°) branch angles and a high density of upward-standing,medium-length siliques that was suitable for dense planting,the conventional plant type has a branch angle ＞50° and is suitable for planting at densities below 45×104plants ha-1[53–56],HZ 62 was defined in the study as a conventional plant type and 1301 as a compact type.

A split-plot design was adopted.The main plot consisted of three levels of density:15×104plants ha-1(D1),30×104plants ha-1(D2),and 45×104plants ha-1(D3),and the sub-plots consisted of three row spacings:15 cm (R15),25 cm (R25),and 35 cm (R35).Each plot was 10 m in length and 2 m in width and was replicated three times.The plant spacings of D1R15,D1R25,D1R35,D2R15,D2R25,D2R35,D3R15,D3R25 and D3R35 were respectively 44.5,26.7,19.1,22.2,13.3,9.5,14.8,8.9,and 6.4 cm.The plots were rainfed.

Six hundred kg ha-1of compound fertilizer(N:P2O5:K2O,15%:15%:15%) and 15 kg ha-1of borax were applied as basal fertilizer before seedbed preparation,and 120 kg ha-1of urea was applied at the 8-leaf stage.Trials were seeded on September 26 each year.After seedling emergence,thinning was applied from the 2-leaf to 4-leaf stages,and final thinning was applied from the 4-leaf to 5-leaf stage.

2.2.Measured indicators and methods

2.2.1.Acquisition of meteorological data

Meteorological data were obtained from the meteorological test station of Huazhong Agricultural University.

2.2.2.Accumulation of dry matter

Ten plants were randomly sampled in each plot at the seedling,bolting,flowering,pod,and maturity stages.The corresponding growth stages (GS) were respectively 19,34,65,75,and 87 on the BBCH scale following Lancashire et al.[57].After measurement of agronomic characteristics,the samples were heated at 105°C for 30 min and then dried at 75°C to constant weight,and the weight of dry matter at each stage was recorded.

2.2.3.Physiological indicators associated with carbon metabolism in leaves

As Müller et al.[58]reported that the fourth leaf from the top of the plant was the first fully developed leaf (in terms of size) in which the source–sink relations were nearly balanced,the upper leaves (with main vein and leaf edge removed) of five plants from each plot were collected at the seedling,bolting and flowering stages.The leaf tissues were snap-frozen in liquid nitrogen and stored at–80 °C for determining physiological and biochemical indictors.

Enzymatic activities including that of Rubisco,FBPase,SuSy,and SPS,were measured using enzyme immunoassay kits purchased from Jiangsu Jingmei Industrial Co.,Ltd.,Dafeng,Jiangsu,China.Samples (0.1 g of fresh tissue) were homogenized in 0.9 mL of phosphate buffer.The prepared sample and the standard substrate were then mixed and allowed to react for 30 min at 37°C.The plate was washed five times.The enzyme reagent was then added and allowed to react for 30 min at 37 °C;the plate was then rinsed another five times,followed by adding the stain for 10 min at 37 °C,after which the stop solution was added.Finally,the OD value was read at 450 nm [59].

Leaf tissues were dried in an oven for 30 min at 105°C to deactivate enzymes and then dried at 80 °C to constant weight.The samples were ground and stored for analyses.The contents of starch,sucrose and soluble sugar were determined as described[60].

2.2.4.Proteomic analysis

At the flowering stage,the leaves in upper canopy layers (with main vein and leaf edge removed) from the conventional plant density and row spacing arrangement D1R25,and the optimized plant density and row spacing D3R15 of the two genotypes were sampled with three replications,and then frozen in liquid nitrogen and stored at–80 °C for proteomic analysis.

Total protein was extracted from leaf tissue with a urea lysis buffer(7 mol L-1urea,2 mol L-1thiourea,and 1%SDS)with a protease inhibitor.Protein concentrations were detected with a BCA Protein Assay Kit (Pierce,Thermo Fisher Scientific,Waltham,MA,USA).Following reduction,cysteine alkylation,and digestion,samples were labeled with isobaric-tags for relative and absolute quantitation (iTRAQ) Reagents (Applied Biosystems,Foster City,CA,USA) according to the manufacturer’s instructions.After being desalted with a C18 solid-phase extraction,peptides were used for nano liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) analysis.

Protein digestion was performed by the standard procedure and the resulting peptide mixture was labeled using the 8-plex iTRAQ reagent (Applied Biosystems) according to the manufacturer’s instructions.Briefly,100 μg protein from each sample was mixed with 100 μL of the lysate.Tris (2-carboxyethyl) phosphine(10 mmol L-1)was added and the mixture was held at 37°C.After 1 h,iodoacetamide(40 mmol L-1)was added and the mixture was held in the dark at room temperature for 40 min.Six volumes of cold acetone were added to precipitate protein at–20 °C for 4 h.After centrifugation at 10,000×g at 4 °C for 20 min,the pellet was re-suspended with 100 μL 50 mmol L-1triethylammonium bicarbonate buffer.Trypsin was added at 1:50 trypsin-to-protein mass ratio and incubated at 37 °C overnight.Then,one unit of iTRAQ reagent was thawed and reconstituted in 50 μL acetonitrile.After tagging for 2 h at room temperature,hydroxylamine was added to react for 15 min at room temperature.Finally,all samples were pooled,desalted,and vacuum-dried.

The pooled samples were fractionated by ACQUITY ultraperformance liquid chromatography (Waters,Milford,MA,USA)with an ACQUITY UPLC BEH C18 Column (1.7 μm,2.1 mm ×150 mm,Waters) to increase proteomic depth.Peptides were first separated with an elution gradient (phase B:5 mmol L-1ammonium hydroxide solution containing 80% acetonitrile,pH 10) over 48 min at a flow rate of 200 μL min-1.Twenty fractions were collected from each sample and pooled into ten fractions per sample.

Labeled peptides were separated by online nanoflow liquid chromatography tandem mass spectrometry performed on an 9RKFSG2_NCS-3500R system (Thermo,USA) connected to a Q Exactive Plus quadrupole orbitrap mass spectrometer (Thermo)through a nano electrospray ion source.The C18 reversed-phase column (75 μm × 25 cm,Thermo) was equilibrated with solvent A (2% formic acid with 0.1% formic acid) and solvent B (80% acetonitrile with 0.1% formic acid).The peptides were eluted using the following gradient:0–4 min,0–5% B;4–66 min,5%-23% B;66–80 min,23%-29% B;80–89 min,29%-38% B;89–91 min,38%–48% B;91–92 min,48%–100% B;92–105 min,100% B;105–106 min,100%–0%B)at a flow rate of 300 nL min-1.The mass spectrometer was operated in data-dependent acquisition mode to switch automatically between full-scan MS and MS/MS acquisition.The survey of full-scan MS spectra (m/z 350–1300) was acquired in the orbitrap with 70,000 resolution.The automatic gain control target wasset to 3e6 and the maximum fill time was set to 20 ms.The 20 most intense precursor ions were selected into a collision cell for fragmentation by higher-energy collision dissociation.The MS/MS resolution was set at 35,000 (at m/z 100),the automatic gain control target at 1e5,the maximum fill time at 50 ms,and dynamic exclusion at 18 s.

The raw data files were analyzed using Proteome Discoverer(Thermo Scientific,Version 2.2).The rapeseed databases from https://www.uniprot.org/(Uniprot.Rape.20181026.fasta,60,180 protein sequences) were used for peptide identifications.The MS/MS search criteria were as follows:mass tolerance of 10 mg L-1for MS and 0.02 Da for MS/MS tolerance,trypsin as the enzyme with 2 missed cleavages allowed,carbamido methylation of cysteine and the iTRAQ of N-terminus and lysine side chains of peptides as fixed modification,and methionine oxidation as dynamic modifications.The false discovery rate for peptide identification was set as FDR 0.01.A minimum of one unique peptide identification was used to support protein identification.

2.3.Data analysis

The data were statistically analyzed using SPSS 21.0 (SPSS Inc.,Chicago,IL,USA),and Origin 9.0 (OriginLab Corporation,Northampton,MA,USA) was used for graphical presentation of data.The LSD test was applied to compare treatment means.The thresholds of fold change (FC 2.0 or 0.5) for two groups(D3R15–D1R25 of HZ 62 and D3R15–D1R25 of 1301) were used to identify differentially expressed proteins (DEPs).The screening criteria for differences in protein expression were as follows:FC 2.0 represented up-regulated and FC 0.5 down-regulated proteins.P ＜0.05 and FC 2.0 represented significantly upregulated and P ＜0.05 and FC 0.5 significantly down-regulated proteins.Annotation of identified proteins was performed using Gene Ontology (GO) (http://www.blast2go.com/b2ghome;http://geneontology.org/) and Kyoto Encyclopedia of Genes and Genomes(KEGG) pathway (http://www.genome.jp/kegg/).DEPs were further subjected to GO and KEGG enrichment analysis.

3.Results

3.1.Meteorological and canopy light conditions

Compared with 2016–2017,the rapeseed growing season(October–May) during 2017–2018 had less average monthly rainfall and lower temperature (Fig.S1).

There were significant differences in the canopy interception rate under different density and row spacing configurations,and the effect of density on the interception rate was greater than that of row spacing (Table 1).The canopy light interception rate increased gradually with increasing density at each row spacing.With increasing row spacing,the interception rate increased and then decreased,with the highest interception rate being found at R25 for D1.The interception rate decreased with increasing row spacing under D2 and D3,with the highest rate being found at the R15 row spacing.HZ62 intercepted more light than 1301.

3.2.Accumulation of aboveground dry matter

The response of dry matter to density and row spacing was similar for the two genotypes.Increased density was associated with significantly decreased dry matter accumulation per plant,but the population biomass was significantly increased.The effect of row spacing on individual and population dry matter accumulation depended on density.With increased row spacing,the dry matter of individual plants and the population rose first and then dropped.They were greatest when the row spacing was R25 at D1,but were markedly reduced in D2 and D3,where they reached their maximum at R15.The analysis of variance revealed that the interaction between density and row spacing significantly influenced dry matter accumulation.Compared with D1R25,the individual aboveground biomass of HZ62 decreased by respectively 49.1% and 60.2% whereas the population biomass increased by 27.2% and 31.9%,in 2016–2017 and 2017–2018 in D3R15 at maturity.The individual biomass of genotype 1301 declined by 54.0% and the population biomass rose by 53.9% at D3R15 in comparison with D1R25 at maturity (Table 2).

Table 1 Effect of density and row spacing on interception rate of canopy light in rapeseed (%).

Table 2 Effect of density and row spacing on aboveground dry matter accumulation at five growth stages of rapeseed.

3.3.Enzymatic activity associated with photosynthetic carbon metabolism

3.3.1.Enzymatic activity associated with photosynthetic carbon fixation

3.3.1.1.Rubisco.Rubisco activity in HZ62 leaves decreased with an increase in density at each stage.With an increase in row spacing,the enzyme activity rose first and dropped at D1,but gradually declined at D2 and D3,reaching its peak at D1R25,D2R15,and D3R15.Rubisco activity of accession 1301 increased with density at flowering,in contrast to the activity in HZ62 leaves,and the trend in other treatments was consistent with that of HZ62.Comparison with D1R25,Rubisco activities in HZ62 leaves declined by 36.2% and 27.2% at D3R15 in 2016–2017 and 2017–2018,respectively,while that of 1301 in the D3R15 treatment rose by 37.0%in 2017–2018 at the flowering stage (Fig.1A).

3.3.1.2.Fbpase.The trends of variation of FBPase activity in leaves were similar for the two genotypes as affected by density and row spacing configuration.The activity dropped with an increase in density.With increased row spacing,enzyme activity first increased and then decreased at D1,but gradually decreased at D2 and D3.In contrast to the D1R25 treatment,FBPase activity in HZ62 decreased by 48.6% and 21.0% in 2016–2017 and 2017–2018,respectively,while that of 1301 decreased by 39.3% in 2017–2018 in the D3R15 treatment at the flowering stage(Fig.1B).

3.3.2.Enzymatic activity associated with photosynthetic carbon translocation

3.3.2.1.SuSy.With increased density,SuSy activity in HZ62 leaves in the seedling stage first increased and then decreased.It first increased and then decreased at R15,and gradually decreased at R25 and R35 at bolting stage,but gradually decreased at flowering stage.SuSy activity in 1301 leaves declined with an increase in density at each stage.With increased row spacing,the enzyme activity first increased and then decreased at D1,while it decreased gradually at D2 and D3 for the two genotypes.In contrast to the D1R25 treatment,SuSy activity in HZ62 decreased by 13.1% and31.0%in 2016–2017 and 2017–2018,respectively,and in genotype 1301 decreased by 39.3%in the D3R15 treatment in 2017–2018 at the flowering stage (Fig.2A).

3.3.2.2.SPs.With increased density,SPS activity in HZ62 leaves decreased gradually at the seedling and flowering stages for R25 and R35,while it rose first and then fell at R15,and gradually decreased at R25 and R35 at the bolting stage.The SPS activity in 1301 leaves declined at the seedling stage,but rose at the bolting and flowering stages with increased density.With an increase in row spacing,the SPS activities in HZ62 and 1301 leaves increased,rising first and then falling at D1,but gradually declined at D2 and D3.In comparison with D1R25,SPS activity in HZ62 leaves decreased by 22.1% and 29.4% in 2016–2017 and 2017–2018,respectively,and that in genotype 1301 increased by 42.6% in 2017–2018 in the D3R15 treatment at the flowering stage(Fig.2B).

3.4.Leaf proteomics at flowering stage

3.4.1.Proteome iTRAQ and differential protein analysis

By iTRAQ identification,4461 proteins were identified in the leaf tissue.According to the screening criteria (FC 2.0 or 0.5),the number of DEPs is shown in Fig.3A.A total of 193 differential proteins were detected in the D3R15–D1R25 group for HZ62.Of those,141 were down-regulated and 52 up-regulated.A total of 216 differentially expressed proteins were detected in the D3R15–D1R25 group for 1301.Of those 67 were down-regulated and 149 up-regulated.The Venn diagram in Fig.3B shows 42 proteins DEPs in leaf samples from the two groups.

3.4.2.GO annotation analysis of DEPs

The GO database provides a useful tool to annotate and analyze the functions of a large number of proteins.GO annotation analysis was performed to identify functional annotations of genes whose expression was altered in the two genotypes by density and row spacing.Proteins were grouped according to the Plant GOslim categories obtained for molecular functions.Based on the GO annotation analysis,identified proteins in leaf protein extracts were classified into three groups:biological process (BP),molecular function (MF),and cellular component (CC) according to their molecular function(Fig.3C).MF proteins accounted for the largest proportion.In the two genotypes,one carbon metabolic process and microtubule-based process in BP-related proteins,microtubule in CC-related proteins,and the structural constitution of cytoskeleton in MF were common DEPs.

3.4.3.Identification of proteins involved in leaf carbon metabolism

DEPsassociated with carbon metabolism were identified in both genotypes.In comparison with D1R25,seven down-regulated proteins involved in photosynthesis and sugar metabolism were identified in D3R15 for HZ62.Those proteins included a protein fructose-bisphosphate aldolase(A0A078JFE6),a protein with phosphopyruvate hydratase activity (A0A078H5M7),a protein with 6-phosphofructokinase activity (A0A078I890),a 2-phytyl-1,4-naphthoquinone methyltransferase (A0A078JNR1),a protein with electron carrier activity (A0A078GVQ6) involved in catalysis of photosynthetic carbon fixation,and two proteins with kinase or phosphotransferase activity(A0A078IJ16,A0A078GF28)associated with sucrose and starch metabolism.Among them,A0A078JFE6,A0A078H5M7 and A0A078IJ16 showed significant differential expression (P ＜0.05).

For 1301,six proteins involved in photosynthesis and sucrose metabolism were identified in D3R15 relative to D1R25.For of those proteins were up-regulated and two down-regulated.Among the up-regulated proteins,one with monooxygenase activity(A5GZU7) and one with oxidoreductase activity (A0A078HCB3),one with sucrose synthase activity(A0A078FUA8),and one participating in photosystem II (PS II) and involved in sucrose metabolism (A0A078G019) were identified.Among the down-regulated proteins,one associated with starch synthesis (A0A078J344),and another involved in glycolysis,having 6-phosphate fructokinase activity (A0A078JNQ8),were identified.Among the six proteins,A0A078HCB3 and A0A078FUA8 showed significant differential expression (P ＜0.05) (Table 3).

Table 3 Identification of DEPs involved in carbon metabolism in leaf tissue at flowering stage of rapeseed for D3R15–D1R25.

3.5.Carbohydrate content at flowering stage

The 1301 plants had much lower starch and sugar levels overall in leaf than HZ62.The carbohydrate responses of the two genotypes to the density and row spacing were very different.With an increase in density,the starch content in HZ62 leaves fell gradually,and the decrease was more marked when density increased from D2 to D3.The trend of starch content in leaves of genotype 1301 depended on the row spacing.With an increase in density,it was elevated significantly at R15,while no significant difference was observed at R25 and R35.With increasing row spacing,the starch content in leaves of the two genotypes rose first and then fell at D1,reaching a maximum at R25,whereas it gradually declined and reached its maximum at R15 for D2 and D3.In comparison with D1R25,the starch content in HZ62 leaves decreased by 16.7% and 21.9% in 2016–2017 and 2017–2018,respectively,while that of 1301 increased by 36.8% in 2017–2018 (Fig.4).

With an increase in density,the sucrose content in HZ62 leaves dropped gradually,especially when density increased from D2 to D3,and the trend was consistent in the two years,while it rose significantly for genotype 1301.With increased row spacing,the sucrose content in leaves of the two genotypes rose first and then fell,and it reached the highest level at R25 for D1,while it decreased gradually and reached the highest level at R15 for D2 and D3,and the amplitude of variation in HZ62 was greater than that in genotype 1301.In comparison with the D1R25 treatment,the sucrose content in HZ62 leaves decreased by 36.9% and 29.6%in 2016–2017 and 2017–2018,respectively,and that in genotype 1301 increased by 33.7% in 2017–2018.The changes in soluble sugar content in leaves of the two genotypes in response to density and row spacing configurations were consistent with those in sucrose content (Fig.4).

Fig.1.Enzymatic activity associated with photosynthetic carbon fixation(A,Rubisco;B,FBPase)in rapeseed leaves of hybrid HZ62(a conventional plant type)and accession 1301(a compact plant type)under differing density and row spacing arrangements.D1,D2,and D3 represent planting densities of 15×104,30×104,and 45×104 plants ha-1 respectively.R15,R2,5,and R35 represent row spacings of 15,25,and 35 cm respectively.Vertical bars in each line represents standard error of the mean of three replicates.*and ** indicate significance at the 0.05 and 0.01 probability levels,respectively,while NS means the difference is not significant by ANOVA.

4.Discussion

As a common agricultural practice,plant density can affect crop yield by regulating root proliferation,canopy structure,and the efficiency of use of light energy,water,and nutrients[4].In spring rapeseed genotypes,the reduction of plant density had no significant negative effects on seed yield per unit area,as was observed for densities from 100 to 40 plants m-2[61] and 40 to 20 plants m-2[62].However,lower plant density from 15 to 5 plant m-2reduced crop productivity in winter rapeseed genotypes [18,62–63].These results suggest that there is a physiological limit to the capacity of plant plasticity to compensate seed yield at low densities,and winter rapeseed will achieve higher yield under dense planting compared to that of low density,as reported by Li et al.[4].However,in winter rapeseed genotypes,at an intermediate plant density of 50–60 plant m-2,an increase in canopy density caused a reduction in the number of rapeseed adult plants due to a self-thinning effect,limiting maximizing yield[64].Dense planting with optimal plant spacing synergistically improved population growth as well as yield in rapeseed[18,19],corn[22–24]and soybean[25].Coordinating the conflict between individual and population growth is an effective way to obtain higher yield of densely planted rapeseed.

Biological yield is the product of growth rate and duration of the growing period,both of which indicate the potential for improvement in yield [65].In the present study,density and row spacing configurations influenced the aboveground dry matter accumulation of individual plants as well as that of the population at maturity.At the same row spacing,when density increased from D1 to D3,the dry matter accumulation per plant decreased,but the population biomass increased.Dry matter accumulation per plant and population of the two genotypes reached the maximum at R25,R15,and R15 for D1,D2,and D3,respectively,with the corresponding plant spacings of 26.7,22.2,and 14.8 cm,respectively,when the difference between plant spacing and row spacing was smallest.In comparison with the conventional plant density and row spacing configuration (D1R25) in rapeseed production,the optimized density and row spacing (D3R15) reduced the biomass per plant of the two genotypes,but increased the population biomass.The seed yield response to population biomass was similar to that reported by Kuai et al.[52].When a mean rice row spacing of 30 cm was arranged at a density of 26.88×104ha-1,the dry matter accumulation from the heading to the maturity stages was promoted,thus achieving the highest yield and improving grain quality[20].In maize,a more uniform plant distribution increased light attenuation only when crop canopy did not reach the critical LAI,which affected yield formation [66].

Fig.2.Enzymatic activity associated with photosynthetic carbon translocation(A,SuSy;B,SPS)in rapeseed leaves of hybrid HZ62(a conventional plant type)and accession 1301(a compact plant type) under several density and row spacing arrangements.D1,D2,and D3 represent planting densities of 15×104,30×104,and 45×104 plants ha-1 respectively.R15,R25,and R35 represent row spacings of 15,25,and 35 cm respectively.Vertical bars in each line represents standard error of the mean of three replicates.*and ** indicate significance at the 0.05 and 0.01 probability levels,respectively,while NS means the difference is not significant by ANOVA.

Manipulation of plant structure can strongly influence light distribution in the canopy and photosynthesis.Plant architectural characteristics (such as the number and geometry of organs,describing their shape and position within the plant and the canopy),are genotype-specific,while at the same time highly dependent on the climatic conditions at the time of their initiation and development [67].The importance of leaf elevation angles for an improved light-absorption strategy at the level of the whole plant has been shown in previous study [68].Both leaf shape and size are important aspects of leaf morphology affecting mutual shading of leaves and light absorption of the canopy [69].Crops with upright and compact plant type and large population leaf area index can make more effective use of light energy to accumulate more dry matter [20].In the present study,the reduction (60.2%)in the single-plant biomass of the conventional plant type HZ62 at D3R15 was greater than that (54.0%) of the compact plant type 1301,whereas the increase (53.9%) in the population biomass of the compact plant type 1301 was greater than that of the conventional plant type HZ62 (31.9%) in comparison with D1R25.The interception rate of canopy light showed a response similar to that of population biomass to density and row spacing.Sunlight is a crucial environmental factor for photosynthesis.The capacity of the crop to capture radiation throughout the crop cycle is closely associated with biomass production at harvest,and thus with the magnitude of seed yield[16].These results indicate that narrowing row spacing under dense planting can optimize light absorption and favor photosynthesis.The combination of plant architecture with agricultural practices led to an increase in photosynthesis which,in turn,could potentially result in a biomass and yield increase.

Photosynthate supply influences pod and seed development[15].Though the leaf area index of rapeseed starts to decrease shortly after first flower,the leaves are still important at full flower,as they photosynthesize more efficiently owing to their higher stomatal density than that of the pod and stem [28].The accumulation and distribution of dry matter in leaves depend on the key enzymes of photosynthetic carbon metabolism,including Rubisco,SuSy,and SPS.The overexpression of cytosolic FBPase can promote sucrose synthesis and growth by increasing available energy reserves or influencing sugar signaling in a transgenic leaf oil crop[70].Enzyme activity is also strongly affected by light environment.Difference in canopy light in response to density and row spacing can also lead to differences in enzyme activity,affecting the accumulation and distribution of dry matter [66,71].In contrast to D1R25,at D3R15 the activities of Rubisco and SPS and the contents of starch,sucrose and soluble sugar decreased in HZ62 leaves and increased in 1301 leaves at flowering stage.The activities of FBPase and SuSy were reduced in HZ62 and 1301.Carbon content in leaves is the key indicator of their physiological status[72].In leaves,soluble sugar is the main carbohydrate that can be transported to seeds [70] and is the main carbon source of oil metabolism in seeds.Increased sugar content in leaves of maize was closely associated with increased vegetative growth [73].In comparison with 1301,lower leaf soluble sugar for HZ 62 may have triggered the signal transduction pathway to slow plant growth,as reported by Mitchell et al.[70].

Differences in carbohydrate metabolic enzymes were found between genotypes at flowering stage.The responses to density and row spacing of carbon metabolism-associated proteins in leaves of differing plant architectures were markedly different.In contrast to D1R25,there were seven down-regulated proteins(A0A078JFE6,A0A078H5M7,A0A078I890,A0A078JNR1,A0A078GVQ6,A0A078IJ16 and A0A078GF28) involved in photosynthesis and sugar metabolism in the conventional plant type HZ62 at D3R15.Among them,A0A078JFE6,A0A078H5M7,involved in catalysis of photosynthetic carbon fixation,and A0A078IJ16,involved in sucrose and starch metabolism,were statistically significant(P ＜0.05).Six DEPs associated with photosynthesis and sucrose metabolism were identified in the compact plant type 1301,of which four (A5GZU7,A0A078HCB3,A0A078FUA8 and A0A078G019) were up-regulated and two(A0A078J344 and A0A078JNQ8) down-regulated.A0A078HCB3,associated with cytochrome 450 and A0A078FUA8,with sucrose synthase activity,were statistically significant (P ＜ 0.05).The expression of several key proteins involved in carbon metabolism,enzyme activity,and content of carbohydrate assimilates in leaves of the genotype with compact plant architecture increased,whereas the opposite trends were observed in those of the genotype with conventional plant architecture in response to increased density and reduced row spacing (Fig.5).

Fig.3.Protein identification (A),illustration of DEPs by Venn map (B),and GO annotation enrichment analysis (C) in leaves at flowering stage of rapeseed hybrid HZ62 (a conventional plant type) and accession 1301 (a compact plant type) between D1R25 and D3R15.The horizontal axis shows the number of proteins enriched by molecular function.Down-regulated and up-regulated mean that a protein was respectively less or more abundant in D3R15.D1R25 represents a planting density of 15×104 plants ha-1 and row spacing of 25 cm;D3R15 represents a planting density of 30×104 plants ha-1 and row spacing of 15 cm.

Fig.4.Starch,sucrose,and soluble sugar content in leaves of the rapeseed hybrid HZ62(a conventional plant type) and accession 1301(a compact plant type)at flowering stage under several density and row spacing arrangements.D1,D2,and D3 represent planting densities of 15×104,30×104,and 45×104 plants ha-1 respectively.R15,R25,and R35 represent row spacings of 15,25,and 35 cm respectively.Vertical bars in each line represent standard error of the mean of three replicates.* and ** indicate significance at the 0.05 and 0.01 probability levels,respectively,while NS means the difference is not significant according to ANOVA.

Fig.5.A suggested model depicting the leaf response for D3R15 in comparison with D125 at flowering stage and its relationship with biomass of different rapeseed genotypes at the physiological and proteomic levels.The red arrow indicates promotion and the blue arrow indicates inhibition.Rubisco,ribulose-1,5-bisphosphate carboxylase/oxygenase SPS,sucrose phosphate synthase;FBPase,fructose-1,6-bisphosphatase;SuSy,sucrose synthase.D1R25 represents a planting density of 15×104 plants ha-1 and row spacing of 25 cm.D3R15 represents a planting density of 30×104 plants ha-1 and row spacing of 15 cm.

These results of the present study illustrate the different changing amplitude of individual and population biomass of the two genotypes to increased density and reduced row spacing.The individual plant biomass in both genotypes was reduced at maturity,possibly owing to reduced capacity of the two plant types to remobilize to pods the photosynthetic products stored in leaves,and a larger decline in this remobilization capacity was observed in the conventional plant type.Pods are the main source organ providing carbon assimilates for seed development once they start developing.In previous studies [28,29],with leaf aging,the pod wall became the main photosynthetic organ for supporting grain development.Investigating differences in pod development among different plant architectures may identify the physiological regulatory mechanisms by which density and row spacing influence rapeseed biomass and yield.

5.Conclusions

Leaf carbohydrate metabolism in two genotypes showed different responses to varied density and row spacing,especially at flowering stage.In contrast to D1R25,the activities of Rubisco and SPS and the contents of starch,sucrose,and soluble sugars in leaves were decreased in HZ62 and increased in genotype 1301 at D3R15.The activities of FBPase decreased,in agreement with the proteomic analysis of HZ62.In contrast,SuSy activity appeared to decrease in both genotypes but increased in abundance in the 1301 genotype ain the proteomic analysis.The accumulation of individual leaf carbohydrates increased in the compact and inhibited in the conventional plant type,but that in the population increased in both genotypes.We conclude that the compact plant type can accumulate more population biomass with denser planting and that the appropriate density can be further increased for achieving higher population biomass and yield.

CRediT authorship contribution statement

Jie Kuai:Writing– original draft,Investigation.Xiaoyong Li:Conceptualization,Data curation,Writing–review&editing.Jianli Ji:Investigation.Zhen Li:Investigation.Yan Xie:Investigation.Bo Wang:Conceptualization,Data curation,Writing– review &editing.Guangsheng Zhou:Conceptualization,Data curation,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

The study was supported by the National Natural Science Foundation of China(31671616),the China Agriculture Research System(CARS-12),and the Fundamental Research Funds for the Central Universities (2662019PY076).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.10.006.