iTRAQ方法杂交稻穗的蛋白质组学研究-J Proteomics-2014.7

Comparative proteomics analysis of superior and inferior spikelets in hybrid rice during grain filling and response of inferior spikelets to drought stress using isobaric tags for relative and absolute quantification

Minghui Dong a ,b ,?,Junrong Gu a ,b ,Li Zhang a ,Peifeng Chen a ,Tengfei Liu a ,Jinhua Deng a ,Haoqian Lu a ,Liyu Han b ,Buhong Zhao c

a

Suzhou Academy of Agricultural Science,Suzhou,215155,PR China

b

Key Laboratory of Crop Genetics and Physiology of Jiangsu Province,Yangzhou University,Yangzhou 225009,PR China c

Lixiahe Region Agricultural Research Institute of Jiangsu,Yangzhou,Jiangsu 225007,China

A R T I C L E I N F O

A B S T R A C T

Article history:

Received 27March 2014Accepted 4July 2014

The biological functions of the differentially abundant proteins between superior and inferior spikelet grains were investigated based on the isobaric tags for relative and absolute quantification to further clarify the mechanism of rice grain filling at the proteomic level,as well as the response of inferior spikelets to drought dress (?20kPa or ?40kPa).Compared with superior spikelets,inferior ones had lower sink strength due to the lower sink activities (lower

abundances of ADP-glucose pyrophosphorylase,granule-bound starch synthase,starch branching enzyme and pullulanase)and smaller sink sizes (lower abundances of structural proteins).The slower and later grain filling resulted from the weaker decomposition and conversion of photoassimilate and the slower cell division.Moderate drought stress (?20kPa)promoted the grain filling of inferior spikelets through regulating the proteins associated with photoassimilate supply and conversion.These proteins may be important targets for rice breeding programs that raise the rice yield under drought condition.The findings offer new insights into rice grain-filling and provide theoretical evidences for better quality control and scientific improvement of super rice in practice.Biological significance

Rice cultivars with large panicles do not always guarantee high yield and grain quality probably due to the slow grain filling and many unfilled grains of inferior spikelets.In general,earlier-flowering superior spikelets,which are usually located on apical primary branches,fill faster and produce larger and heavier grains.In contrast,later-flowering inferior spikelets located on proximal secondary branches are either sterile or fill slowly and poorly,and the differences are more significant in large panicle rice or super rice.The increase of rice yield has been limited by the unsatisfactory grain filling of inferior spikelets,

Keywords:Hybrid rice Grain filling Drought stress iTRAQ Proteome

J O U R N A L O F P R O T E O M I C S 109(2014)382–399

DOI of original article:https://www.360docs.net/doc/cc7001728.html,/10.1016/j.dib.2014.08.001.

?Corresponding author at:Suzhou Academy of Agricultural Science,Suzhou,215155,PR China.Tel.:+8651265382356;fax:+8651265381859.E-mail address:mhdong@https://www.360docs.net/doc/cc7001728.html, (M.

Dong).

https://www.360docs.net/doc/cc7001728.html,/10.1016/j.jprot.2014.07.001

1874-3919/?2014Elsevier B.V.All rights

reserved.

A v a i l a b l e o n l i n e a t w w w.s c i e n c e d i r e c t.c o m

ScienceDirect

w w w.e l s e v i e r.c o m /l o c a t e /j p r o t

and the inferior spikelets are more prone to environmental factors during grain filling. Thus,we herein investigated the biological functions of differently abundant proteins between superior and inferior spikelet grains by using iTRAQ to unravel the mechanism of rice grain filling and the response of inferior spikelets to drought stress at proteomic level. This study offers new insights into rice grain-filling and provides valuable evidences for better quality control and scientific improvement of super rice in practice.

?2014Elsevier B.V.All rights reserved.

1.Introduction

Rice sink capacity and grain filling efficiency significantly impact the yield that has been tentatively elevated by breeding of large panicle hybrid rice or super rice mainly through increasing the number of full-filling spikelets per panicle[1]. However,cultivars with large panicles do not always guarantee high yield and grain quality probably due to the slow grain filling and many unfilled grains of inferior spikelets[2,3].In general,earlier-flowering superior spikelets,which are usually located on apical primary branches,fill faster and produce larger and heavier grains.In contrast,later-flowering inferior spikelets located on proximal secondary branches are either sterile or fill slowly and poorly,and the differences are more significant in large panicle rice or super rice[4–6].The increase of rice yield has been limited by the unsatisfactory grain filling of inferior spikelets[7–10].Compared with numerous studies focused on sucrose-to-starch conversion[5,11–19],limitations in carbon supply and sink capacity[20–23],imbalance of phytohormones[7,10,24–27],enzyme activities[13,28–30],and gene expression[31–33]have seldom been referred.However, grain filling is a complex biological and molecular process,with the exact mechanism remaining elusive.

Rice quality is determined by both genotype and environment. The two peaks of grain filling and two environmental-sensitive periods,especially those of large panicle rice,are determined by the ear filling characteristics of different grains under normal conditions[34].The degree and the rate of grain filling as well as the grain weight of rice spikelets differ largely depending on their positions on a panicle,and the inferior spikelets are more prone to environmental factors during grain filling,indicating that effectively regulating cultivation can augment the yield by improving the grain filling of inferior spikelets[35].Soil water status,particularly that during grain filling,influences grain quality dramatically.Proper water stress can promote the remobilization of prestored carbon reserves to the grains[36], increase the ratio of abscisic acid(ABA)to ethylene[19],and enhance the activities of sucrose synthase(SuSase),soluble and insoluble invertase,starch branching enzyme(SBE)and soluble starch synthase(SSSase).It may also facilitate the starch accumulation during grain filling,especially that in inferior spikelets during the grain filling of wheat[37].

In the last decade,proteomics has become an indispensable complement of transcriptome in life science.Proteomics is able to analyze simultaneous changes and to classify the temporal patterns of protein accumulation in complex developmental processes[38].Rice is the main grain crop for human and one of the Poaceae plants with the smallest genome(about430MB)[39]. Besides,rice is an eligible model plant for molecular biology because its gene is easily operatable and similar to those of other monocot plants[40].Since the rice genome map has been finished in2002,the proteomics of rice has been highlighted [41–44],mainly on the protein abundance patterns of tissues/ organs and subcellular components.Zhang et al.[43]employed 2-D gel-based comparative proteomic and phosphoproteomic analyses to search the differentially abundant proteins in inferior spikelets under exogenous ABA treatment,and found that111 such proteins were related with defense response as well as carbohydrate,protein,amino acid,energy and secondary me-tabolisms,revealing that the grain filling of rice inferior spikelets was regulated by ABA through the proteins and phosphoproteins participating in carbon,nitrogen and energy metabolisms.Zi et al.[45]analyzed the stress responsive proteins during rice embryogenesis by using isobaric tags for relative and absolute quantification(iTRAQ)and shotgun techniques,and found that most of the up-regulated proteins,including heat shock-,lipid transfer-,and reactive oxygen species-related proteins,were functionally categorized as stress responsive.The proteomics of superior or inferior spikelets,especially that of large panicle rice or super rice,has scarcely been analyzed,and the differences of protein abundances and functions between superior and inferior spikelets remain unclear.Although2-DE provides a visual representation of the proteome in which distinct protein isoforms resulting from the changes in Mr and/or pI can be observed,it does not apply to detection of low-abundance proteins and more accurate quantification[46].iTRAQ,one of the mass-based quantitative approaches,has become prevalent in the field of crop proteomics[47]by simultaneously identifying and quantifying proteins from multiple samples with high coverage.Thus,we herein investigated the biological functions of the differentially abundant proteins between superior and inferior spikelet grains by using iTRAQ to unravel the mechanism of rice grain filling and the response of inferior spikelets to drought stress at proteomic level.This study offers new insights into rice grain-filling and provides valuable evidences for better quality control and scientific improvement of super rice in practice.

2.Materials and methods

2.1.Rice cultivation

Field experiments were carried out in an experimental farm of Taihu Area Institute of Agricultural Sciences,Su Zhou,Jiangsu province,China in2011,with large-panicle hybrid Japonica rice Yongyou8(according to the test for an average of181 grains with26.3grams weight per panicle)as the material. Seedlings were sown on20th May and transplanted on25th June at a hill spacing of0.3m×0.15m with1seedling per hill. The soil of the field was paddy soil that contained2.42% organic matter and158.4,8.4and127.0mg·kg?1available N–P–

383

J O U R N A L O F P R O T E O M I C S109(2014)382–399

K respectively.Field management was in accordance with the conventional technique for high-yield cultivation,N fertilizer (225kg/hm2),basal-tiller N fertilizer to ear-grain N fertilizer (6:4),the basal to ear-grain N was2:1,and ear-grain N fertilizer was used when the last fourth or fifth leaves came out.P fertilizer was converted into P2O5(70kg/hm2)as the basal fertilizer and K fertilizer was converted into K2O(150kg/hm2) according to the ratio of basal-tiller N fertilizer to ear-grain N fertilizer(5:5).

2.1.1.Collection of superior and inferior spikelets

Panicles that headed on the same day were chosen and tagged,and the flowering date of each spikelet on the tagged panicles was recorded.Two hundred panicles that headed on a same day were tagged.The flowering date and position of each spikelet on the tagged panicles were recorded.Fifteen tagged panicles were sampled at7days and14days after anthesis(DAA,the day was accounted from the first day after flowering).Superior spikelets(SS)and inferior spikelets(IS) were collected according to the previous report[4],then were frozen in liquid N2and then stored at?70°C for protein extraction.

2.1.2.Water stress treatment

Three irrigation patterns were set from c,i.e.shallow water irrigation(water status was controlled at0kPa),light wetting–drying irrigation(water status was controlled at?20kPa),and heavy wetting–drying irrigation(water status was controlled at?40kPa).The test base was covered by weather shed.The water status was determined at7:00–8:00and16:00–17:00 every day by a portable digital measuring instrument for soil water potential and temperature(TRS-II,Zhejiang Tuopu Equipment Co.,Ltd.).Shallow water irrigation was performed when the water status was lower than the set value.

2.2.Protein extraction and digestion

Frozen rice tissue was finely powdered in liquid nitrogen,and precipitated for1h with25mL TCA/acetone(1:9,containing 65mM DTT)at?20°C.The homogenate was centrifuged and the pellets were air-dried,dissolved in30μL STD buffer(4% SDS,150mM Tris–HCl,pH8.0),incubated with boiling water for5min,cooled to room temperature,and diluted with 200μL of UA buffer(8M urea,150mM Tris–HCl,pH8.0).The homogenate was centrifuged,the supernatants were collected and the protein content was determined by a BCA protein assay reagent.

The retained protein was washed with200μL of UA buffer, centrifuged,and added with100μL of UA buffer containing 0.05M iodoacetamide.The mix was incubated for20min in dark and then centrifuged under the above conditions.The filter was then washed three times with100μL of UA buffer, and100μL of DS buffer(50mM triethylammonium bicarbon-ate,pH8.5)was added.Then the solution was centrifuged for 10min in the same condition.This step was repeated twice. Finally,40μL of DS buffer containing3μg trypsin(Promega) was added to each filter.The samples were incubated overnight at37°C,and the resulting peptides were collected by centrifugation.The peptide content was estimated by UV density at280nm.2.3.iTRAQ reagent labeling and liquid chromatography(LC)

iTRAQ labeling was performed according to the manufacturer's instructions(Applied Biosystems).Briefly, the peptide mixtures were reconstituted with30μL of iTRAQ dissolution buffer.The label method of every sample(45μg) using iTRAQ Reagent-8plex Multiplex Kit(AB SCIEX)is shown in Table1,and every sample was labeled twice.The aliquots of iTRAQ were combined with peptide mixtures from5 different samples,respectively,and incubated at room temperature for1h.REF was a mixture containing same-amount proteins of the five samples.

Prior to LC-MS/MS analysis,the peptides were purified to eliminate excess labeling reagent by SCX chromatography using an AKTA Purifier system(GE Healthcare).A10μL solution from each peptide fraction was injected for nanoLC-MS/MS analysis using a Q-Exactive MS(Thermo Finnigan)equipped with Easy nLC(Proxeon Biosystems,now Thermo Fisher Scientific).The peptide mixture(5μg)was loaded onto a C18-reversed phase column packed in-house with RP-C18 resin(5μm)in buffer A(0.1%formic acid)and separated with a linear gradient of buffer B(0.1%formic acid in80%acetonitrile) at a flow rate of250nL/min controlled by IntelliFlow technology over140min.

2.4.Electrospray ionization(ESI)tandem MS(MS/MS) analysis by Q Exactive

MS data were acquired using a data-dependent top10method dynamically choosing the most abundant precursor ions from the survey scan(300–1800m/z)for the HCD fragmentation. The target value was determined based on predictive Auto-matic Gain Control(pAGC).The dynamic exclusion duration was60s.Survey scans were acquired at a resolution of70,000 at m/z200,and resolution for the HCD spectra was set to 17,500at m/z200.The normalized collision energy was30eV, and the underfill ratio,which specifies the minimum per-centage of the target value likely to be reached at maximum fill time,was defined as0.1%.The instrument was run with peptide recognition mode enabled.

Table1–Labeled method of every sample(45μg)using iTRAQ Reagent-8plex Multiplex Kit(AB SCIEX).Every sample was labeled twice(Labels1and2).

Sample IS A1C1E2B2D2 Label1113116117118119121 Sample IS B1E1A2D1C2 Label2113114115118119121

Note:

A:inferior spikelets on7DAA under0kPa(IS7DAA);

B:superior spikeletes on7DAA under0kPa(SS7DAA);

C:inferior spikelets on14DAA under0kPa(IS14DAA);

D:inferior spikelets on14DAA under?20kPa;

E:inferior spikelets on14DAA under?40kPa.

384J O U R N A L O F P R O T E O M I C S109(2014)382–399

2.5.Sequence database searching and data analysis

MS/MS spectra were searched using MASCOT engine (Matrix Science,London,UK;version 2.2)against a rice sequence database (uniprot_Oryza_sativa .fasta,released in February 2013,144512sequences).The MASCOT search results were further processed using Proteomics Tools (version 3.05).As-sembling protein identifications were qualitatively analyzed by Proteome Discoverer 1.4software.All data were reported based on 99%confidence for protein identification as determined by false discovery rate (FDR)≤1%.Isobaric Labeling Multiple File Distiller and Identified Protein iTRAQ Statistic Builder were used to calculate the ratios of protein,in which sample REF was used as the reference,based on the weighted average of the intensities of report ions in each identified peptide.The final ratios were then normalized with the median average protein ratio,assuming that most proteins remained unchanged in abundance.Only the protein identification that was inferred from the unique peptide identification in two independent experiments was considered.Statistical analysis was conduct-ed using a one-way ANOVA.P-values ≤0.05by Tukey's test were considered significant.Among the statistically significant proteins detected by the ANOVA test (p <0.05),protein abun-dances that changed less than 1.5-fold or 1.2-fold were discarded (Fig.2).

2.6.Bioinformatics analysis of the differentially abundant proteins

Sequence data of the selected the differentially abundant proteins were retrieved from UniProtKB database (Release 2013_07)in batches in FASTA format.The retrieved sequences were locally searched against Swiss-Prot database (plant)using the NCBI BLAST+client software (ncbi-blast-2.2.28+-win32.exe)to find homologue sequences from which the functional annotation was transferred to the studied sequences.In this study,the top 10blast hits with E-value less than 1e ?3for each query sequence were retrieved and loaded into Blast2GO (Version 2.6.6)for Gene Ontology (GO)mapping and annotation.

The sequences without BLAST hits and the un-annotated ones were then selected to go through InterProScan against EBI databases to retrieve the functional annotations.The GO project described the roles of proteins in three domains:biological process,molecular function and cellular component.Following annotation and annotation augmentation,enzyme codes were sequentially mapped to annotated sequences and metabolic pathways in Kyoto Encyclopedia of Genes and Genomes (KEGG,http://www.genome.jp/kegg/)[48].

3.Results

In the present study,4388proteins were not reproducibly identified in all experiments including developmental stages and replicates in iTRAQ proteomic analysis,in which any single analytical run may only identify a fraction of the relevant peptides in a highly complex mixture of peptides [49].1207proteins were identified for the two biological replicates and were subjected to comparative analysis.Among the statistically significant proteins detected by the ANOVA test (p <0.05),protein abundances that changed less than 1.5-fold or 1.2-fold were discarded.Following this criterion,we detected a total of 185proteins that are differentially abundant during rice grain filling.

In Group ?20kPa/0kPa,fifty proteins changed over 1.2-fold and only 7proteins showed more than 1.5-fold changes.In Group ?40kPa/0kPa,61proteins were subject to more than 1.2-fold changes,and 30proteins changed more than 1.5-fold.Considering that fewer proteins changed more than 1.5-fold,we listed the proteins more than 1.2-fold to provide more informa-tion of the differentially abundant proteins (Table 2).

In the SDS-PAGE maps with high resolution (Fig.1)(loaded with 20μg,10μg BSA as control),the protein bands differ https://www.360docs.net/doc/cc7001728.html,pared with inferior spikelets 7days after anthe-sis (A,IS7DAA),protein band nos.2,3,5and 7of sample B differed significantly,while protein band nos.1,7and 10of sample C showed down-regulated abundances,and band nos.3,4,6,8and 9exhibited up-regulated ones.Hence,the grain proteins in inferior spikelets changed remarkably compared with superior spikelets during grain filling (7–14days after anthesis).

3.1.Differences of numbers and types of proteins among superior and inferior spikelets during grain filling

Table 2lists the protein names and sequences in UniProt and Swiss-Prot databases,and the fold changes of the protein abundances.In Group SS7DAA/IS7DAA,106proteins underwent more than 1.5-fold changes,with 30of them up-regulated and 76down-regulated.Of the 106differentially abundant proteins,19were subject to more than 2-fold changes,i.e.4were up-regulated and 15were down-regulated.As evidenced by the function analysis,the proteins were related to sugar metabolism (e.g.SuSase 3,trehalose-phosphate synthase,6-phosphogluconate dehydrogenase),adenylic acid,nucleotide,amino acid metabo-lisms (e.g.nucleoside diphosphate kinase 1,adenylate kinase B,putative alanine aminotransferase),protein metabolism (e.g.small ubiquitin-related modifier,60S ribosomal protein),starch biosynthesis (e.g.pyruvate phosphate dikinase 1,

chloroplastic,

BSA M A B C D E

10

97kDa 66kDa 43kDa

31kDa

20kDa

Fig.1–SDS-PAGE maps of inferior spikelets 7DAA under 0kPa (A),superior spikeletes on 7DAA (B)under 0kPa,and inferior spikelets on 14DAA under 0kPa (C),?20kPa (D)or ?40kPa (E).

385

J O U R N A L O F P R O T E O M I C S 109(2014)382–399

glucose-1-phosphate adenylyltransferase),cell structure (e.g.actin-7,actin-97,tubulin beta-1chain),glycolysis and pentose phosphate pathway (glucose-6-phosphate isomerase,cytosolic 1,pyruvate dehydrogenase)(Table 3).

In Group IS14DAA/IS7DAA,127proteins showed more than 1.5-fold changes,with 56of them up-regulated and 71down-regulated.Of the 127differentially abundant proteins,44underwent more than 2-fold changes,i.e.35were up-regulated and 9were down-regulated.The proteins with differential abundances were involved in starch biosynthesis and degrada-tion (e.g.granule-bound starch synthase,alpha-amylase tryp-sin),cell structure (e.g.glycine-rich RNA-binding protein),respiratory metabolism (e.g.glucose-6-phosphate isomerase,cytosolic 1),sugar metabolism (e.g.endo-beta-mannosidase),stress response and defense (e.g.chitinase 7,ferredoxin,chloroplastic),etc.Particularly,ubiquitin thioesterase showed the largest changes (6.98-fold),and many glutelins showed more than 2-fold abundances.Moreover,many uncharacterized proteins with differential abundances were also identified.

3.2.Differential protein abundances of inferior spikelets 14days after anthesis at 0kPa (C),?20kPa (D)and ?40kPa (E)

In Group ?20kPa/0kPa,fifty proteins changed over 1.2-fold,with 22up-regulated and 28down-regulated,and only 7proteins showed more than 1.5-fold changes.Among the 50proteins,two down-regulated proteins underwent more than 2-fold changes.The proteins participated in starch biosynthe-sis (e.g.granule-bound starch synthase 1,SBE),starch degradation (e.g.alpha-amylase trypsin),cell division (e.g.Ras-related protein,actin-7),nitrogen metabolism (e.g.ala-nine aminotransferase),stress response and defense (e.g.

seed allergenic protein,S-adenosylmethionine synthase 2),and storage (e.g.glutelin,prolamin,globulin).

In Group ?40kPa/0kPa,61proteins were subject to more than 1.2-fold changes,including 6up-regulated and 55down-regulated ones.Among the proteins,8down-regulated proteins changed more than 2-fold,and 30down-regulated proteins changed more than 1.5-fold.The proteins were involved in respiratory metabolism (e.g.formate dehydroge-nase,mitochondrial,glucose-6-phosphate isomerase,cyto-solic 1),water transportation (e.g.probable aquaporin),stress response and defense (e.g.cytochrome b5)and storage.

3.3.GO annotation of the differentially abundant proteins

To represent the overall trends of the specific functional categories that are enriched in rice grains,a Gene Ontology (GO)category enrichment analysis was conducted using all 4388identified proteins.The 4388proteins were categorized according to GO Slim classification for plants.Among 4388proteins,1207were annotated by this analysis.The GO annotation of proteins with differential abundances in bio-logical process,molecular function and cellular component is exhibited in Fig.3.

In biological process,the differentially abundant proteins mainly experienced metabolic process,cellular process,biolog-ical regulation,response to stimulation,cellular component biogenesis and establishment of localization.In cellular com-ponent,the differentially abundant proteins of superior and inferior spikelets in different water statuses were mainly distributed in cytoplasm,organelle and cell membrane.In molecular function,the differentially abundant proteins were mostly related to binding and catalysis,and also

as

Fig.2–Representative MS/MS spectrum showing the peptides from glycine-rich RNA-binding protein (No.44,peptide sequence:GFGFVTFDEK)were labeled with iTRAQ (A-116,B-119,C-117,D-121,E-118)tags.

386

J O U R N A L O F P R O T E O M I C S 109(2014)382–399

antioxidants,enzyme regulators,molecular transducers,struc-tural molecules,transporters and electron carriers.

Following annotation and annotation augmentation,en-zyme codes were sequentially mapped to annotated se-quences and metabolic pathways.Fig.4presents the representative metabolic pathway maps of enzymes with differential abundances involved in sucrose and starch metabo-lism in KEGG.UDP-glucose6-dehydrogenase(E.C.1.1.1.22),SuSase (E.C.2.4.1.13),trehalose6-phosphate synthase(E.C.2.4.1.15), glucose-1-phosphate adenylyltransferase(E.C.2.7.7.27)and tre-halose6-phosphate phosphatase(E.C.3.1.3.12)that changed considerably in superior and inferior spikelets during grain filling were colored.

Compared with Group SS7DAA/IS7DAA,more proteins in Group IS14DAA/IS7DAA participated in cellular component biogenesis and developmental process,and less protein was related with metabolic process.Besides,more differentially abundant proteins were distributed in cytoplasm and organ-elle in Group IS14DAA/IS7DAA,accompanied by more pro-teins possessing enzyme-regulating and catalytic activities as well as less structural molecule-related proteins.

Compared with Group?40kPa/0kPa,significantly more proteins in Group?20kPa/0kPa were involved in metabolic and cellular processes,less proteins were distributed in cytoplasm and organelle,and there were more proteins with binding functions and catalytic activities.

4.Discussion

Recently,researchers have devoted to clarifying the differ-ences between the superior and inferior spikelets during grain filling on the levels of photosynthetic assimilate supply,grain hormone,enzyme activity and gene expression.The changes of protein abundances in superior and inferior spikelets during grain filling have been qualitatively and quantitatively analyzed by using2-DE to elucidate the molecular mechanism [43,45,50].However,the protein abundance change patterns of superior and inferior spikelets in super hybrid rice and the response of inferior spikelets to water stress remain un-known.In this study,by using the large spikelet hybrid rice samples under normal condition or drought stress,iTRAQ method was applied to get more differentially abundant proteins with more accurate quantitative outcomes and to detect some unknown proteins with significant functions through bioinformatic analysis.

4.1.Proteins involved in starch biosynthesis

The activities of ADP-glucose pyrophosphorylase(AGPase,No. 148),granule-bound starch synthase(GBSS,No.108),SBE(No. 137),SuSase(No.18),pullulanase(No.150)and isoamylase(No. 146)that determine the synthesis of grain starch and amylopec-tin[5]are positively related to the accumulation rates of total starch and amylopectin.Since the enzyme activity of GBSS is

Table3–Functional classification of differentially abundant proteins in inferior and superior spikelets on7DAA(B/A, SS7DAA/IS7DAA)and on14DAA(C/A,IS14DAA/IS7DAA),and inferior spikelets on14DAA in group(D/C,?20kPa/0kPa) and(E/C,?40kPa/0kPa).

Function catalogues Protein no.(total,percentage)

Cell division2,34,82,83,88,89,90,97,115(9,5.69%)

Starch biosynthesis and metabolism19,93,108,117,132,137,145,148(8,5.06%)

Sugar metabolism3,18,29,77,84,91,98(7,4.43%)

Other macromolecule metabolism5,7,15,22,28,31,32,35,36,37,38,39,42,45,46,50,53,56,58,61,63,67,68,69,70,73,74,75,86,96,

106,107,117,128,130,133,138,140,150(39,24.68%)

Respiration30,54,85,92,94,95,99,127(8,5.06%)

Stress response and defense4,6,8,9,16,17,20,21,23,24,25,26,33,40,43,48,49,57,59,60,72,78,79,81,87,123,143,147,151,

152,153(31,19.62%)

Material transport and signal transduction1,10,11,12,13,14,16,18,27,44,65,66,71,80(14,8.86%)

Photosynthesis41,104(2,1.26%)

Storage proteins109,110,111,112,113,114,118,119,120,129,131,135,139,141,144,149,154,155(18,11.39%) Unknown47,51,52,55,62,64,76,93,100,101,102,103,105,121,122,124,125,126,134,136,142,146,156,157,

158(25,15.82%)

Both in B/A and C/A

Both in E/C and D/C

Fig.3–1.Distribution of158differentially abundant proteins in

inferior and superior spikelets on7DAA(B/A,SS7DAA/IS7DAA),

and in inferior spikelets on7DAA and14DAA(C/A,IS14DAA/

IS7DAA).2.Distribution of84differently abundant proteins

in inferior spikelets on14DAA group(D/C,?20kPa/0kPa)

and(E/C,?40kPa/0kPa).

392J O U R N A L O F P R O T E O M I C S109(2014)382–399

positively related to the amylose accumulation rate,it is the key enzyme that controls amylose synthesis [24].SBE plays an important role in amylopectin cluster formation or tree struc-ture [51].In this study,the enzymes involved in starch biosynthesis,such as AGPase,GBSS,SBE and starch debranching enzyme,were of significant protein abundance differences in Groups SS7DAA/IS7DAA and IS14DAA/IS7DAA.AGPase is the first enzyme involved in the pathway of starch synthesis,and up-regulated abundance of cytoplasmic AGPase gene during grain filling contributes to a higher yield [52].In this study,AGPase,GBSS,SBE and pullulanase were subject to 1.88-fold,2.38-fold,1.81-fold and 1.80-fold up-abundances in IS14DAA respectively compared with those in IS8DAA.At the proteomic level,the grain filling rate was dominantly hindered by the efficiency of starch biosynthesis in inferior spikelets,and this result was in accordance with Zhang et al.'s proteomic research

Biological Process (B/A)

response to

cellular component

8080

organismal process, 16

extracellular 16

r complex, 27

lumen, 13

Cellular Component(B/A)transporter activity, 8

activity, 17

catalytic nucleic acid

binding transcription 2enzyme Molecular Function(B/A)

cellular component process, 241

response to organization or organismal

process, 24

74

Biological Process (C/A)

ar complex,

23

Cellular Component (C/A)

nucleic acid nutrient reservoir transporter activity, 6

Molecular Function (C/A)

organismal process, 4

process, 416

process, 6cellular component

Biological Process (D/C)

Cellular Component(D/C)

16

nutrient reservoir activity, 12

activity, 3

activity, 1

Molecular Function(D/C)

15

cellular component

Biological Process (E/C)

complex, 5

cell junction, 3

Cellular Component(E/C)nutrient reservoir activity, 22

activity, 14

Molecular Function(E/C)

Fig.4–Bioinformatics analysis of the above mentioned differentially abundant proteins through Gene Ontology (GO)in three domains:biological process,molecular function and cellular component.The statistics at GO level 2is shown in this figure.

393

J O U R N A L O F P R O T E O M I C S 109(2014)382–399

[53].Fu has stated that the low activities of SuSase and AGPase,which are extremely significantly positively correlated with the grain filling rate,may be mainly responsible for the slow grain filling and light grain weight of inferior spikelets [7].Li indicated that the activities of AGPase,SuSase and SBE gradually increased and then declined as a single-peak curve with the process of grain filling in the progenies with high and low amylose contents in grains [13].

Above all,the present results combined with previous biological studies proved that the low activities of enzymes involved in starch biosynthesis (AGPase,SuSase,SBE,etc.)in inferior spikelets were the main reason of the slow grain filling fate,and further research could be focused on regulation of the enzymes by means of molecular biology to gain higher grain filling rate of inferior spikelets.Meanwhile,α-amylase/trypsin inhibitor (No.145)was 2.17-fold up-regulation in Group IS14DAA/IS7DAA.Alpha-amylase/trypsin inhibitor is synthe-sized during grain filling and is an abundant protein of the endosperm and the aleurone layers of the mature seed [54].The inhibitor can inhibit the endogenous a-amylase activity and in defense against pathogens and pests [55].The present results indicated that starch degradation had been depressed during grain filling of inferior spikelets,due to grain filling of inferior spikelets that was initiated after 7DAA,later than superior spikelets,and the up-regulation of α-amylase/trypsin inhibitor could promote the starch accumulation.

Under water stress,GBSS and SBE were up-regulated 1.45-fold and 1.27-fold respectively in Group ?20kPa/0kPa,but there were no significant differences in Group ?40kPa/0kPa.Moderate drought stress induced the abundances of these two enzymes,facilitated the synthesis of starch in inferior spikelets,and accelerated grain filling [35].Water deficit enhanced the activity of APGase in the initial stage of

grain filling [11],but it remained unchanged in inferior spikelets either under ?20kPa or ?40kPa water status,probably owing to the water stress intensity or the different sampling period.

Meanwhile,α-amylase/trypsin inhibitor (No.145)was 0.74-fold down-regulated under ?20kPa water stress,while alpha-amylase trypsin (No.132)was 0.63-fold down-regulated under ?40kPa water stress.In other words,water stress affected starch degradation by regulating the protein abun-dance of α-amylase.Moderate water stress (?20kPa water status)augmented the α-amylase activity by increasing the abundance of alpha-amylase inhibitor,whereas excessive water stress (?40kPa water status)down-regulated the abundance of α-amylase.Given that this enzyme activity and starch degradation rate were significantly positively correlated,water deficit was conducive to starch hydrolysis and stem carbohydrate export mainly through regulating the α-amylase activity [56].In the meantime,water stress contributed to grain weight [57]by benefiting the transport of storage substance to the ear and the output of carbohydrate in stem and sheath.Ergo,moderate drought stress boosted the transport of storage materials from vegetative organs to grain,thus reducing the low photosynthetic rate-induced yield loss.In practice,grain filling rate and grain yield can be raised by moderate water stress and by choosing the cultivars with highly active metabolic enzymes.

4.2.Proteins involved in sucrose synthesis and metabolism

The proteins with differential abundances involved in sucrose metabolism were searched in KEGG (http://www.genome.jp/kegg/)and colored in Fig.5.Although catalyzing sucrose synthesis and sucrose decomposition,SuSase (SuS,No.

18)

Fig.5–Representative metabolic pathway maps of differentially abundant enzymes involved in sucrose and starch

metabolism in KEGG (http://www.genome.jp/kegg/).And the colored enzyme codes were noted as follows:UDP-glucose 6-dehydrogenase (E.C.1.1.1.22);sucrose synthase (E.C.2.4.1.13);trehalose 6-phosphate synthase (E.C.2.4.1.15);glucose-1-phosphate adenylyltransferase (E.C.2.7.7.27);trehalose 6-phosphate phosphatase (E.C.3.1.3.12).

394

J O U R N A L O F P R O T E O M I C S 109(2014)382–399

mainly decomposes sucrose[15]into fructose and UDPG as the first step of catalytic starch synthesis in grain,the protein abundance level of which reflects the degradation ability[58]. In this study,the abundances of SuS in SS7DAA and IS14DAA were 1.67-fold and 1.58-fold higher than those in IS7DAA respectively.Our previous studies also showed that the earlier-flowered spikelets had greater contents of the soluble sugar content(SSC)than later-flowered,and SSC in the spikelets on the secondary rachis branch was more than that on the primary rachis branches at the initial and mid-filling stages[4].

Trehalose-6-phosphate synthase catalyzes the first step in trehalose synthesis,playing a role in stress protection and carbohydrate storage[59].The biosynthesis of trehalose com-prises the formation of trehalose-6-phosphate out of UDP-glucose and glucose-6-phosphate by the enzyme trehalose-6-phosphate synthase[60].Results provide evidence that trehalose-6-phosphate regulates utilization of sugars for storage starch synthesis by promoting reductive activation of AGPase in the plastid[61].In this study,the abundances of trehalose-phosphate synthase(TPS,No.84)in SS7DAA and IS14DAA were 6.80-fold and5.12-fold higher than that in IS7DAA,respectively. The results showed that trehalose–phosphate synthase plays important roles in starch synthesis during grain filling.

The present studies suggested that SS7DAA had stronger capacity against sucrose decomposition and conversion to starch,and inferior spikelets had this capacity after14DAA, which may lead to the low grain filling rate and plumpness of them.

4.3.Involved in protein synthesis and metabolism

More than95%of inorganic nitrogen in higher plants is assimilated to glutamate and glutamine via the glutamine synthetase pathway,and then transformed into other amino acids when catalyzing aspartate aminotransferase and alanine aminotransferase,providing various amino acid donors for the synthesis and metabolism of grain protein[62].In this study, putative alanine aminotransferase(No.86)in SS7DAA was

1.80-fold higher than that in IS7DAA,and that in IS14DAA was

2.30-fold higher.Previous proteomics research carried out by Zhang showed that the abundances of alanine aminotransfer-ase were down-regulated on inferior spikelets during grain filling,possibly due to an insufficient N supply and an accelerated aging of the rice plants at the later stages[53].The alanine aminotransferase activity was more prone to changes in early grain filling and remained steady thereafter[63].As suggested by the similar results herein,alanine aminotransfer-ase played a key role in the nitrogen metabolism during grain filling,and the capacity of photosynthetic assimilation in superior spikelets exceeded that of inferior spikelets.The protein experienced 1.26-fold higher abundance under?20kPa water stress,while there was no difference under?40kPa water stress.Previous studies have verified that moder-ate drought boosted the activity of alanine aminotransferase and the nitrogen metabolism of grain[64],while this study demonstrated that water stress regulated the photosyn-thetic assimilation capacity of inferior spikelets by elevat-ing the abundance of alanine aminotransferase on the protein level.

Some proteins involved in protein synthesis,such as60S ribosomal proteins(Nos.32,33,36,46,50,53,63,67,68,70,138, 162and185)and40S ribosomal proteins(Nos.37,56,74,96 and176),were also identified.They were differentially down-regulated in SS7DAA and IS14DAA,except for the up-regulated abundances of ribosomal proteins(Nos.32,36, 63,46and162)probably owing to their different roles during grain filling.Many ribosome proteins have been linked with cell structure,protein translation,protein biosynthesis and plant development[65].

Under?20kPa water stress,two60S ribosomal proteins (Nos.46and162)were1.27-fold and1.23-fold up-regulated respectively,whereas other two ribosomal proteins(Nos.50 and53)were0.77-fold and0.75-fold down-regulated.Under?40kPa water stress,60S ribosomal proteins(Nos.176,185,53 and50)and40S ribosomal protein No.176were all down-regulated.The ribosomal proteins showed complex protein abundance change patterns in grain filling under water stress,which thus need further studies.

4.4.Proteins involved in respiration

In this study,some proteins involved in respiration(glycolysis, tricarboxylic acid cycle and pentose phosphate pathway)were identified,such as UDP-glucose6-dehydrogenase(No.77), 6-phosphogluconate dehydrogenase(No.95),pyruvate dehy-drogenase(No.94),glucose-6-phosphate isomerase(No.91), ATP synthase subunit(Nos.85and99)and malate dehydroge-nase(No.127).These proteins,which showed complex protein abundance change patterns during grain filling under water stress,participated in respiratory metabolism synergistically to maintain higher respiration and to meet the demands for ATP and the synthesis of cellular components.In Zhang's study,the abundances of most of the identified proteins relating to the glycolysis and TCA cycle were found to be lower in the grains on inferior spikelets than on superior ones[53].

It is worth noting that compared with IS7DAA,glucose-6-phosphate isomerase(No.91)showed2.53-fold(IS14DAA)and 4.17-fold(SS7DAA)up-regulations,while6-phosphogluconate dehydrogenase(No.95)also underwent1.50-fold(IS14DAA)and 1.21-fold(SS7DAA)up-regulations.Moreover,glucose-6-phosphate isomerase increased1.45-fold under?20kPa water stress and reduced0.66-fold under?40kPa water stress. Furthermore,as suggested by the1.20-fold increase of malate dehydrogenase under?20kPa water stress,moderate drought (?20kPa water stress)provided sufficient metabolic substrate and energy during grain filling by boosting the respiratory metabolism through elevating the abundances of these two enzymes,while excessive drought(?40kPa water stress)might suppress the respiratory metabolism by increasing the abun-dance of glucose-6-phosphate isomerase,thereby reducing the grain filling rate.

4.5.Proteins involved in cell structure

RAS-related proteins play a crucial role in mitogenic signal transduction.In this study,Ras-related proteins(Nos.10and 11)in both SS7DAA and IS14DAA were0.54-fold and0.62-fold down-regulated compared with those in IS7DAA,respective-ly.Two structural proteins,glycine-rich RNA-binding protein

395

J O U R N A L O F P R O T E O M I C S109(2014)382–399

(No.44)and tubulin(No.35),also varied following the same trend.

Plant tubulin(Nos.35and97)and actin(Nos.88,89and 90),which are important components of the cytoskeleton, support cells as the basis of cell surface morphology, tissue and normal growth[66].In this study,the lower abundances of tubulin and actin97in SS7DAA might be ascribed to the completion of mitosis on7DAA in superior spikelets,while the inferior spikelets were still in the peak of division.Therefore,the low abundance of structural proteins may result in the slow cell division and small sink size,being one of the reasons for low grain filling rate of inferior spikelets.

Under?20kPa water stress,actin-7increased1.25-fold and Ras-related protein decreased0.79-fold in IS14DAA,while cell structure-related proteins changed moderately under?40kPa water stress,revealing that water stress regulated cell division and the protein abundances of structural proteins by a compli-cated paradigm.

4.6.Proteins involved in material transport and signal transduction

Calmodulin(CaM,No.1),an important protein activated by Ca2+ in plant signal transduction,mediates many physiological and biochemical processes in plants[67].With rising CaM content, the Ca2+-ATPase activity becomes stronger and the material transport becomes more exuberant[68].Aquaporin(AQP,Nos.19 and28),which predominantly sits in the protein storage vacuoles,is also transiently accumulated at the plasma mem-brane during the early stages of seed germination and matura-tion,can prevent water deficit by preventing the moisture lost in cells with appropriate osmotic substances[69].CaM and AQP in SS7DAA and IS14DAA were down-regulated compared with those in IS7DAA,indicating that material synthesis and accu-mulation were intensified during grain filling and material transport may be suppressed.Given that AQP was0.77-fold down-regulated under?40kPa water stress,excessive water stress might destroy the dynamic balance of cell moisture.To date,clear genetic evidence for a role of aquaporins in seed germination has only been provided in rice,using transgenic plants with loss-and gain of function of AQP[70].In particular, previous studies show seed-specific AQP protein abundance and their abundance markedly decreased during germination,ac-companied with the massive deposition of storage proteins, oligosaccharides and phytins in protein storage vacuoles during late seed development[71].Our present studies showed that AQP plays important roles in seed germination and the response to water stress at proteomic level.

S-adenosylmethionine synthetase(SAMS,No.14),a key protein in the methionine cycle,mainly catalyzes the synthesis of S-adenosylmethionine(SAM)[72],the precursor for synthe-sizing ethylene and polyamines that are involved in the response of plants to multiple stresses[73].SAMS is closely related to cell wall physicochemical properties,lignin biosyn-thesis,and cell wall lignification,and also has many biological stress resistances[74].SAMS was up-regulated by?20kPa water stress and1.79-fold in SS7DAA as well.The up-regulation of SAMS might raise the concentrations of hormones such as ethylene and polyamine in cells.Ethylene affects grain filling rate through regulating the activities of starch synthesis-related enzymes[7],but the mechanism has not been unraveled.

In this study,non-specific lipid transfer proteins(nsLTP, Nos.6,8,9,59,60,123and175)were also identified.nsLTP participates in intracellular phospholipid transfer,synthesis of biological membrane and plant stress and defense response [75].Drought or heat shock,which dehydrates cells,increases the nsLTP gene expression[76]Similarly,nsLTP(No.175)herein was subject to1.27-fold up-regulation under?40kPa water stress.Various environmental stresses,such as drought[77], cold[78]and high-salt level[79],induced the expression of nsLTP gene and significantly enhanced the plant resistance or tolerance to stress[80].Therefore,the over-abundance of nsLTP induced by water stress(?40kPa)might contribute to the self-protection mechanism.Nevertheless,nsLTP was only up-regulated under suitable conditions,as confirmed by the absence of changes under?20kPa water stress.The exact functions of nsLTP during grain filling still need further studies.

4.7.Proteins involved in stress response

Thioredoxin H-type(No.4),glutaredoxin-C6(No.12),superoxide dismutase(No.17),tricin synthase2(No.23),peroxiredoxin-2C (No.27),ferredoxin(No.43)and cytochrome b5(No.87),which were identified as involved in stress response,have antioxidant effects on cells and can scavenge reactive oxygen species. Germin-like proteins(GLPs,No.159)are also involved in various stress responses,the protein abundances of which can be induced by fungal pathogens,bacteria,viruses,salt,heavy metal and drought[81,82].Zhang et al.reported that GLPs were up-regulated by abscisic acid(ABA),predicted that ABA could enhance resistance to adversities that ensured the success of grain-filling of the inferior spikelets[43],and then they reported that GLPs could be part of the auxin signal network that mediates the cell division and expansion,and the lower protein abundance of inferior spikelets may lead to their lowered cell division than superior spikelets[53].In the present study,GLPs were depressed by both moderate and excessive water stresses, while no significant difference was found in Groups SS7DAA/ IS7DAA and IS14DAA/IS7DAA,the results indicated that the abundances of GLPs were regulated complicatedly during grain filling under normal condition or water stress.

4.8.Storage protein

Most storage proteins exist in endosperm cells as proteasomes.Hence,grain quality was determined by the contents and ratios of these proteins.A large number of storage proteins,such as glutelin(Nos.11,109,110,111,113, 114,129,135,139,161,etc.),alcohol soluble protein(Nos.155, 158,170,etc.),globulin(Nos.115,141,etc.)and seed allergenic protein(Nos.118,119,131,144,etc.),were identified.Previous research showed that the abundance of glutelin-related proteins in rice caryopsis increased significantly under high temperature stress during the grain-filling stage[83].The present study provide new evidences that these proteins with differential abundances during grain filling can be affected not only by high temperatures but also drought stress in different cultivars,however,the detailed regulation mecha-nism still need further analysis.

396J O U R N A L O F P R O T E O M I C S109(2014)382–399

4.9.Unknown proteins

A series of unknown proteins,such as Nos.20,55,102,105, 126,134,136,142and158,were also discovered by biological information comparisons.The proteins,as indicated by the over two-fold up-regulation in superior or inferior spikelets, exerted crucial physiological effects during grain filling. However,even with the state-of-the-art iTRAQ technology and bioinformatics protocols,these proteins could not be identified and their functions were still unknown.

5.Conclusion

A proteomics approach based on iTRAQ was applied to analyze the differences between superior and inferior spikelets as well as the effects of drought stress on the proteomic patterns of inferior spikelets during grain filling.Inferior spikelets were of lower sink strength than superior ones due to the lower sink activities(lower abundances of AGPase,GBSS,SBE and pullulanase)and smaller sink sizes(lower abundances of structural proteins).Moreover, the milder decomposition and conversion of photoassimilate and the slower cell division decelerated and undermined grain filling. In addition,moderate drought stress(?20kPa)facilitated the synthesis of inferior spikelets,starch hydrolysis and stem carbohydrate export by inducing the accumulation of GBSS,SBE and alanine aminotransferase,thus regulating the capacity of photosynthetic assimilation and intensifying the respiratory metabolism for catering the needs of metabolic substrate and energy.Nevertheless,excessive drought stress(?40kPa)might destroy the dynamic balance of cell moisture and enhance the resistance or tolerance to stress as a kind of self-protection.In summary,this study provides innovative results for elucidating the differences between superior and inferior spikelets during grain filling and the tolerance of rice to drought stress through the regulation of proteins associated with photoassimilate supply and conversion.These proteins may be essential targets for rice breeding programs aiming to elevate the rice yield under drought condition.Furthers studies are needed to determine whether the drought responsive proteins are genetically fixed and eligible markers for breeding.Moreover,it is reasonable to increase grain filling rate and grain yield by moderate water stress and by choosing the cultivars with highly active metabolic enzymes in practice.

Conflict of interest

All authors have read and approve this version of the article, and due care has been taken to ensure the integrity of the work.No part of this paper has published or submitted elsewhere.No conflict of interest exists in the submission of this manuscript.

Acknowledgment

This research was financially supported by the grants from the National Natural Science Foundation of China(Grant No.31171490),the Natural Science Foundation of Jiangsu Prov-ince,China(Grant No.BK2011269),Jiangsu cultivation projects of“333high-level talent”and Open project of key laboratory of crop genetics and physiology of Jiangsu Province(K13013).

The mass spectrometry proteomics data have been depos-ited to the Proteome Xchange Consortium via the PRIDE partner repository with the dataset identifier PXD001046.

R E F E R E N C E S

[1]Fu J,Yang J-c.Research advances in high-yielding

cultivation and physiology of super rice.Rice Science2012;

19:177–84.

[2]Peng S,Khush GS,Virk P,Tang Q,Zou Y.Progress in ideotype

breeding to increase rice yield potential.Field Crop Res2008;

108:32–8.

[3]Yang W,Peng S,Dionisio-Sese ML,Laza RC,Visperas RM.

Grain filling duration,a crucial determinant of genotypic

variation of grain yield in field-grown tropical irrigated rice.

Field Crop Res2008;105:221–7.

[4]Dong M-h,Chen P-f,Xie Y-l,Qiao Z-y,Yang J-c.Variations in

carbohydrate and protein accumulation among spikelets at different positions within a panicle during rice grain filling.

Rice Science2012;19:223–32.

[5]Mohapatra PK,Sarkar RK,Kuanar SR.Starch synthesizing

enzymes and sink strength of grains of contrasting rice

cultivars.Plant Sci2009;176:256–63.

[6]Zhang H,Tan G-L,Sun X-L,Liu L-J,Yang J-C.Changes of grain

quality during evolution of mid-season indica rice cultivars in Jiangsu Province.Acta Agron Sin2009;35:2037–44.

[7]Fu J,Xu Y-j,Chen L,Yuan L-m,Wang Z-q,Yang J-c.Changes

in enzyme activities involved in starch synthesis and

hormone concentrations in superior and inferior spikelets

and their association with grain filling of super rice.Rice

Science2013;20:120–8.

[8]Fu J,Huang Z,Wang Z,Yang J,Zhang J.Pre-anthesis

non-structural carbohydrate reserve in the stem enhances

the sink strength of inferior spikelets during grain filling of

rice.Field Crop Res2011;123:170–82.

[9]Zhang C,Jiang D,Liu F,Cai J,Dai T,Cao W.Starch granules

size distribution in superior and inferior grains of wheat is

related to enzyme activities and their gene expressions

during grain filling.J Cereal Sci2010;51:226–33.

[10]Tan G-L,Zhang H,Fu J,Wang Z-Q,Liu L-J,Yang J-C.

Post-anthesis changes in concentration of polyamine in

superior and inferior spikelets and their relationship with grain filling of super rice.Acta Agron Sin2009;35:2225–33.

[11]Jeng TL,Tseng TH,Wang CS,Chen CL,Sung JM.Starch

biosynthesizing enzymes in developing grains of rice cultivar Tainung67and its sodium azide-induced rice mutant.Field Crop Res2003;84:261–9.

[12]Kaur A,Kaur P,Singh N,Virdi AS,Singh P,Rana JC.Grains,

starch and protein characteristics of rice bean(Vigna

umbellata)grown in Indian Himalaya regions.Food Res Int

2013;54:102–10.

[13]Li X-g,Liu H-y,Jin Z-x,Liu H-l,Huang X,Xu M-l,et al.Changes

in activities of key enzymes for starch synthesis and

glutamine synthetase in grains of progenies from a rice cross during grain filling.Rice Science2010;17:243–6.

[14]Schulman AH,Runeberg-Roos P,J??skel?inen M.Grain filling

and starch synthesis in barley.In:Anil Kumar G,Narinder K,

editors.Developments in crop science.Elsevier;2000.p.147–67.

[15]Su J-C.Starch synthesis and grain filling in rice.In:Anil

Kumar G,Narinder K,editors.Developments in crop science.

Elsevier;2000.p.107–24.

397

J O U R N A L O F P R O T E O M I C S109(2014)382–399

[16]Devi TA,Sarla N,Siddiq EA,Sirdeshmukh R.Activity and

expression of adenosine diphosphate glucose

pyrophosphorylase in developing rice grains:varietal

differences and implications on grain filling.Plant Sci2010;

178:123–9.

[17]Teramura H,Oshima T,Matsuda F,Sasaki K,Ogino C,

Yamasaki M,et al.Glucose content in the liquid hydrolysate after dilute acid pretreatment is affected by the starch

content in rice straw.Bioresour Technol2013;149:520–4. [18]Tuncel A,Okita TW.Improving starch yield in cereals by

over-expression of ADPglucose pyrophosphorylase:

expectations and unanticipated outcomes.Plant Sci2013;

211:52–60.

[19]Yang J,Zhang J,Wang Z,Zhu Q,Wang W.Hormonal changes

in the grains of rice subjected to water stress during grain

filling.Plant Physiol2001;127:315–23.

[20]Albacete AA,Martínez-Andújar C,Pérez-Alfocea F.Hormonal

and metabolic regulation of source–sink relations under

salinity and drought:From plant survival to crop yield

stability.Biotechnology Advances2014;32:12–30.

[21]Mae T,Inaba A,Kaneta Y,Masaki S,Sasaki M,Aizawa M,et al.

A large-grain rice cultivar,Akita63,exhibits high yields with

high physiological N-use efficiency.Field Crop Res2006;

97:227–37.

[22]Xiong F,Wang Z,Gu Y-j,Chen G,Zhou P.Effects of nitrogen

application time on caryopsis development and grain quality of rice variety Yangdao6.Rice Science2008;15:57–62.

[23]Dong G-C,Wang Y-L,Zhou J,Zhang B,Zhang C-S,Zhang Y-F,

et al.Characteristics of nitrogen distribution and translocation in conventional indica rice varieties with different nitrogen use efficiency for grain output.Acta Agronomica Sinica2009;

35:149–55.

[24]Peng D-l,Cai T,Yin Y-p,Yang W-b,Ni Y-l,Yang D-q,et al.

Exogenous application of abscisic acid or gibberellin acid has different effects on starch granule size distribution in grains of wheat.J Integr Agric2013;12:1551–9.

[25]Yang J-c,Chang E-h,Tang C,Zhang H,Wang Z-q.Relationships of

ethylene evolution rate and1-aminocylopropane-1-carboxylic acid concentration in grains during filling period with

appearance quality of rice.Rice Science2007;14:33–41.

[26]Zhao B-h,Liu K,Zhang H-x,Zhu Q-s,Yang J-c.Causes of poor

grain plumpness of two-line hybrids and their relationships to the contents of hormones in the rice grain.Agric Sci China 2007;6:930–40.

[27]Zhang H,Tan G,Yang L,Yang J,Zhang J,Zhao B.Hormones in

the grains and roots in relation to post-anthesis development of inferior and superior spikelets in japonica/indica hybrid

rice.Plant Physiol Biochem2009;47:195–204.

[28]Huang X,Jin Z-x,Li X-g,Liu H-l,Xu M-l,Zhang F-z,et al.

Effects of grain protein content selection on protein content and key enzyme activity involved in nitrogen metabolism in progenies derived from a rice cross.Rice Sci2010;17:156–60.

[29]Krishnan P,Ramakrishnan B,Reddy KR,Reddy VR.Chapter

three—high-temperature effects on rice growth,yield,and grain quality.In:Donald LS,editor.Advances in Agronomy.

Academic Press;2011.p.87–206.

[30]Liang C-g,Chen L-p,Wang Y,Liu J,Xu G-l,Li T.High

temperature at grain-filling stage affects nitrogen

metabolism enzyme activities in grains and grain nutritional quality in rice.Rice Science2011;18:210–6.

[31]Galléá,Csiszár J,Secenji M,Guóth A,Cseuz L,Tari I,et al.

Glutathione transferase activity and expression patterns

during grain filling in flag leaves of wheat genotypes differing in drought tolerance:response to water deficit.J Plant Physiol 2009;166:1878–91.

[32]Huh SM,Y-s Hwang,Shin YS,Nam MH,Kim DY,Yoon IS.

Comparative transcriptome profiling of developing caryopses from two rice cultivars with differential dormancy.J Plant

Physiol2013;170:1090–100.[33]Wei K-S,Zhang Q-F,Cheng F-M,Zhong L-J,Chen N.

Expression profiles of rice soluble starch synthase isoform

genes in response to high temperature.Acta Agronomica

Sinica2009;35:18–24.

[34]Ji K,Wang Y,Sun W,Lou Q,Mei H,Shen S,et al.

Drought-responsive mechanisms in rice genotypes with

contrasting drought tolerance during reproductive stage.J

Plant Physiol2012;169:336–44.

[35]Cattivelli L,Rizza F,Badeck F-W,Mazzucotelli E,Mastrangelo

AM,Francia E,et al.Drought tolerance improvement in crop plants:an integrated view from breeding to genomics.Field Crop Res2008;105:1–14.

[36]Yang J,Zhang J,Wang Z,Zhu Q,Wang W.Remobilization of

carbon reserves in response to water deficit during grain

filling of rice.Field Crop Res2001;71:47–55.

[37]Yang J,Zhang J,Wang Z,Zhu Q,Liu L.Activities of enzymes

involved in sucrose-to-starch metabolism in rice grains

subjected to water stress during filling.Field Crop Res2003;

81:69–81.

[38]Jorrín-Novo JV,Maldonado AM,Echevarría-Zome?o S,

Valledor L,Castillejo MA,Curto M,et al.Plant proteomics

update(2007–2008):second-generation proteomic

techniques,an appropriate experimental design,and data

analysis to fulfill MIAPE standards,increase plant proteome coverage and expand biological knowledge.J Proteomics

2009;72:285–314.

[39]Sasaki T,Burr B.International Rice Genome Sequencing

Project:the effort to completely sequence the rice genome.

Curr Opin Plant Biol2000;3:138–42.

[40]Goff SA.Rice as a model for cereal genomics.Curr Opin Plant

Biol1999;2:86–9.

[41]Khan MMK,Komatsu S.Rice proteomics:recent

developments and analysis of nuclear proteins.

Phytochemistry2004;65:1671–81.

[42]Li Z-W,Xiong J,Li Z-F,Qi X-H,Chen H-F,Shao C-H,et al.

Analysis of differential expression of proteins in rice leaf

sheath during grain filling.Acta Agronomica Sinica2008;

34:619–26.

[43]Zhang Z,Chen J,Lin S,Li Z,Cheng R,Fang C,et al.Proteomic

and phosphoproteomic determination of ABA's effects on

grain-filling of Oryza sativa L.inferior spikelets.Plant Sci2012;

185–186:259–73.

[44]Zhang A,Lu Q,Yin Y,Ding S,Wen X,Lu https://www.360docs.net/doc/cc7001728.html,parative

proteomic analysis provides new insights into the regulation of carbon metabolism during leaf senescence of rice grown under field conditions.J Plant Physiol2010;167:1380–9. [45]Zi J,Zhang J,Wang Q,Zhou B,Zhong J,Zhang C,et al.Stress

responsive proteins are actively regulated during rice(Oryza sativa)embryogenesis as indicated by quantitative

proteomics analysis.PLoS One2013;8:e74229.

[46]Abreu IA,Farinha AP,Negr?o S,Gon?alves N,Fonseca C,

Rodrigues M,et al.Coping with abiotic stress:Proteome

changes for crop improvement.Journal of Proteomics2013;

93:145–68.

[47]Robbins ML,Roy A,Wang P-H,Gaffoor I,Sekhon RS,de O.

BuanafinaMM MM,et https://www.360docs.net/doc/cc7001728.html,parative proteomics analysis by DIGE and iTRAQ provides insight into the regulation of

phenylpropanoids in maize.Journal of Proteomics2013;

20(93):254–75.

[48]Kanehisa M,Goto S,Sato Y,Furumichi M,Tanabe M.KEGG for

integration and interpretation of large-scale molecular data sets.Nucleic Acids Res2012;40:D109–14.

[49]Lee J,Koh H-J.A label-free quantitative shotgun proteomics

analysis of rice grain development.Proc Natl Acad Sci U S A 2011;9:1–10.

[50]Zhang C,Yin Y,Zhang A,Lu Q,Wen X,Zhu Z,et https://www.360docs.net/doc/cc7001728.html,parative

proteomic study reveals dynamic proteome changes between superhybrid rice LYP9and its parents at different

developmental stages.J Plant Physiol2012;169:387–98.

398J O U R N A L O F P R O T E O M I C S109(2014)382–399

[51]Wei K-s,Cheng F-m,Zhang Q-f,Liu K-g.Temperature stress

at grain filling stage mediates expression of three isoform

genes encoding starch branching enzymes in rice

endosperm.Rice Science2009;16:187–93.

[52]Kang G,Liu G,Peng X,Wei L,Wang C,Zhu Y,et al.Increasing

the starch content and grain weight of common wheat by

overexpression of the cytosolic AGPase large subunit gene.

Plant Physiol Biochem2013;73:93–8.

[53]Zhang Z,Zhao H,Tang J,Li Z,Li Z,Chen D,et al.A proteomic

study on molecular mechanism of poor grain-filling of rice (Oryza sativa L.)inferior spikelets.PLoS One2014;9(2):e89140.

https://www.360docs.net/doc/cc7001728.html,/10.1371/journal.pone.0089140.

[54]Nielsen PK,B?nsager BC,Fukuda K,Svensson B.Barley

α-amylase/subtilisin inhibitor:structure,biophysics and protein engineering.Biochim Biophys Acta Protein Proteomics2004;

1696:157–64.

[55]Svensson B,Fukuda K,Nielsen PK,B?nsager BC.

Proteinaceousα-amylase inhibitors.Biochim Biophys Acta

Protein Proteomics2004;1696:145–56.

[56]Wei W,Yi Xia C,Kun Zheng C,Jian Hua Z,Jian Chang Y,Qing

Sen Z.Regulation of soil water deficits on stem stored

carbonhydrate remobilization to grain of rice.Acta

Phytoecologica Sin2005;29:819–28[in Chinese].

[57]Wang H-z,Zhang L-h,Ma J,Li X-y,Li Y,Zhang R-p,et al.

Effects of water stress on reactive oxygen species generation and protection system in rice during grain-filling stage.Agric Sci China2010;9:633–41.

[58]Hirose T,Scofield GN,Terao T.An expression analysis profile

for the entire sucrose synthase gene family in rice.Plant Sci 2008;174:534–43.

[59]Goddijn O,Smeekens S.Sensing trehalose biosynthesis in

plants.Plant J1998;14:143–6.

[60]Goddijn OJM,van Dun K.Trehalose metabolism in plants.

Trends Plant Sci1999;4:315–9.

[61]Wingler A.The function of trehalose biosynthesis in plants.

Phytochemistry2002;60:437–40.

[62]Zhao X-Q,Shi W-M.Expression analysis of the glutamine

synthetase and glutamate synthase gene families in young rice(Oryza sativa)seedlings.Plant Sci2006;170:748–54. [63]Foulkes MJ,Hawkesford MJ,Barraclough PB,Holdsworth MJ,

Kerr S,Kightley S,et al.Identifying traits to improve the

nitrogen economy of wheat:recent advances and future

prospects.Field Crop Res2009;114:329–42.

[64]Ramanjulu S,Sudhakar C.Drought tolerance is partly related

to amino acid accumulation and ammonia assimilation:a

comparative study in two mulberry genotypes differing in

drought sensitivity.J Plant Physiol1997;150:345–50.

[65]Yao Y,Ni Z,Du J,Wang X,Wu H,Sun Q.Isolation and

characterization of15genes encoding ribosomal proteins in wheat(Triticum aestivum L.).Plant Sci2006;170:579–86. [66]Sadiq I,Fanucchi F,Paparelli E,Alpi E,Bachi A,Alpi A,et al.

Proteomic identification of differentially expressed proteins in the anoxic rice coleoptile.J Plant Physiol2011;168:2234–43.

[67]Perochon A,Aldon D,Galaud J-P,Ranty B.Calmodulin and

calmodulin-like proteins in plant calcium signaling.

Biochimie2011;93:2048–53.[68]Cheval C,Aldon D,Galaud J-P,Ranty B.

Calcium/calmodulin-mediated regulation of plant immunity.

Biochim Biophys Acta Mol Cell Res2013;1833:1766–71. [69]Yu Q,Hu Y,Li J,Wu Q,Lin Z.Sense and antisense expression

of plasma membrane aquaporin BnPIP1from Brassica napus in tobacco and its effects on plant drought resistance.Plant Sci2005;169:647–56.

[70]Li G,Santoni V,Maurel C.Plant aquaporins:roles in plant

physiology.Biochim Biophys Acta Gen Subj2014;

1840:1574–82.

[71]Maurel C.Plant aquaporins:novel functions and regulation

properties.FEBS Lett2007;581:2227–36.

[72]Mato J,Alvarez L,Ortiz P,Pajares MA.S-adenosylmethionine

synthesis:molecular mechanisms and clinical implications.

Pharmacol Ther1997;73:265–80.

[73]Roy M,Wu R.Overexpression of S-adenosylmethionine

decarboxylase gene in rice increases polyamine level and

enhances sodium chloride-stress tolerance.Plant Sci2002;

163:987–92.

[74]CvikrováM,GemperlováL,MartincováO,VankováR.Effect of

drought and combined drought and heat stress on polyamine metabolism in proline-over-producing tobacco plants.Plant Physiol Biochem2013;73:7–15.

[75]Salcedo G,Sánchez-Monge R,Barber D,Díaz-Perales A.Plant

non-specific lipid transfer proteins:an interface between

plant defence and human allergy.Biochim Biophys Acta Mol Cell Biol Lipids2007;1771:781–91.

[76]Altenbach SB.New insights into the effects of high

temperature,drought and post-anthesis fertilizer on wheat grain development.J Cereal Sci2012;56:39–50.

[77]Romo S,Labrador E,Dopico B.Water stress-regulated gene

expression in Cicer arietinum seedlings and plants.Plant

Physiol Biochem2001;39:1017–26.

[78]Gaudet DA,Laroche A,Frick M,Huel R,Puchalski B.Plant

development affects the cold-induced expression of plant

defence-related transcripts in winter wheat.Physiol Mol

Plant Pathol2003;62:175–84.

[79]Jang CS,Lee HJ,Chang SJ,Seo YW.Expression and promoter

analysis of the TaLTP1gene induced by drought and salt

stress in wheat(Triticum aestivum L.).Plant Sci2004;

167:995–1001.

[80]Carvalho AdO,Gomes VM.Role of plant lipid transfer

proteins in plant cell physiology—a concise review.Peptides 2007;28:1144–53.

[81]Bernier F,Berna A.Germins and germin-like proteins:plant

do-all proteins.But what do they do exactly?Plant Physiol Biochem2001;39:545–54.

[82]Davidson RM,Reeves PA,Manosalva PM,Leach JE.Germins:a

diverse protein family important for crop improvement.Plant Sci2009;177:499–510.

[83]Liao J-L,Zhou H-W,Zhang H-Y,Zhong P-A,Huang Y-J.

Comparative proteomic analysis of differentially expressed proteins in the early milky stage of rice grains during high

temperature stress.J Exp Bot2014;65:655–71.

399

J O U R N A L O F P R O T E O M I C S109(2014)382–399