Molecular regulation of seed and fruit set

This is an author-deposited version published in:

http://oatao.univ-toulouse.fr/

Eprints ID: 7967

To link to this article:

DOI:10.1016/j.tplants.2012.06.005

URL: http://dx.doi.org/10.1016/j.tplants.2012.06.005

To cite this version:

Ruan, Yong-Ling and Patrick, John William and

Bouzayen, Mondher and Osorio, Sonia and Fernie, Alisdair R. Molecular

regulation of seed and fruit set. (2012) Trends in Plant Science, vol. 17 (n°

11). pp. 1360-1385. ISSN 1360-1385!

O

pen

A

rchive

T

oulouse

A

rchive

O

uverte (

OATAO

)

OATAO is an open access repository that collects the work of Toulouse researchers and

makes it freely available over the web where possible.

Any correspondence concerning this service should be sent to the repository

administrator:

staff-oatao@listes.diff.inp-toulouse.fr

Molecular

regulation

of

seed

and

fruit

set

Yong-Ling

Ruan

,

John

W.

Patrick

,

Mondher

Bouzayen

,

Sonia

Osorio

,

and

Alisdair

R.

Fernie

1

Australia-ChinaResearchCentreforCropImprovement,TheUniversityofNewcastle,Callaghan,NSW2308,Australia

2

SchoolofEnvironmentalandLifeSciences,TheUniversityofNewcastle,Callaghan,NSW2308,Australia

3

Universite´deToulouse,InstitutNationalPolytechnique–EcoleNationaleSuperieurAgronomiqueToulouse,Ge´nomiqueet BiotechnologiedesFruits,F-31326,Castanet-Tolosan,France

4

InstitutNationaldelaRechercheAgronomique,Ge´nomiqueetBiotechnologiedesFruits,ChemindeBordeRouge,F-31326, Castanet-Tolosan,France

5

Max-Planck-Institutfu¨rMolekularePflanzenphysiologie,14476Potsdam-Golm,Germany

6

IHSM-UMA-CSIC,DepartamentodeBiologı´aMolecularyBioquı´mica,UniversidaddeMa´laga,29071Ma´laga,Spain

Seedandfruitsetareestablishedduringandsoonafter fertilizationanddetermineseedandfruitnumber,their finalsizeand,hence,yieldpotential.Theseprocessesare highly sensitive to biotic and abiotic stresses, which often leadto seedandfruitabortion.Here, wereview the regulation of assimilatepartitioning, including the potentialrolesofrecentlyidentifiedsucroseefflux trans-portersinseedandfruitsetandexaminethesimilarities of sucrose import and hydrolysis for both pollen and ovary sinks, and similar causes of abortion. We also discuss the molecular origins of parthenocarpy and the centralrolesof auxinsandgibberellinsin fruitset. Therecentlycompletedstrawberry(Fragariavesca)and tomato(Solanumlycopersicum)genomeshaveaddedto theexistingcropdatabases,andnewmodelsare start-ingtobeusedinfruitandseedsetstudies.

Seedandfruitset:attheheartoffoodsecurity

Seedandfruitarethekeyyieldcomponentsinmostcrop species. Assuch,theirdevelopmenthasbeenresearched extensively for decades. In broad terms, seed and fruit developmentcanbedividedintothreestages:set,growth and maturation. Seed and fruit set (see Glossary) are establishedduringandsoonafterfertilization.Thisstage features a transition from ovule and ovary to seed and fruit, respectively, andischaracterizedby extensivecell division and coordinated development of maternal and filialtissues[1,2].The newlyformed fruitandseedthen undergocellexpansionandaccumulation ofstorage pro-ducts,mainlyproteins,starchandoils,whicharetypical featuresofgrowthandmaturationstages[3].

Research on seed and fruit development has largely focused on thelatestagesof development.By contrast, much less is known about the mechanisms regulating their early development during the set phase. Under-standing thisearlyprocessisimportantforseveral rea-sons. First, molecular and biochemical pathways responsible for fruit and seed set are likely to have a profound impact on the later stages of development.

Second, the set stage determines the fruit and seed number and, to a largedegree, their finalsize through establishing cell numbers and, thus, yield potential. Third, similar to mammals, where early pregnancy is mostproneto abortion,thesetphaseishighlysensitive to internal and external stresses compared with later stages of fruit and seed development [4] or vegetative growth[5,6].Stressesincludeinsufficientsupplyof nutri-ents,drought,heatorcold,whichofteninducesubstantial floral, seed and fruit abortion and, hence, irreversible yieldlosses[4,7,8].Forexample,incereals,waterdeficit during floweringcanreduceyield byup to 60%,largely owingtoreductionsingrainset[1].Similarly,heatstress canresultin70%yieldlossintomato(Solanum lycoper-sicum)asaresultofflowerandfruitabortion[9].

Elucidating the mechanisms underpinning seed and fruitsetorthoseresponsiblefortheirabortionis funda-mental to our understanding of reproductive biology andisessentialfordesigningapproachestoreduce abor-tion, thereby improving crop yield. Indeed, in the next 50 years, crop yield per unit land area needs to be doubled to meet demand owing to projected global increases in the human population and of living stan-dards, whichislikelyto beexacerbatedbydecreasesin theavailabilityofarableland[10].Here,wereviewmajor advances in the regulation of seed and fruit set by focusing on: (i) source–sink interactions, (ii) sugar and hormonal signaling and (iii) transcriptional and meta-bolicpathways.Finally, weoutline futuredirectionsfor furtheringourunderstandingof themolecular mechan-ismsunderlyingseedandfruitsetandpotential applica-tionsinalleviating theirabortion.

Glossary

Fruitabortion:abscissionorstuntedgrowthoffruit. Fruitset:transitionofanovarytoagrowingyoungfruit.

Parthenocarpicfruit:fruitdevelopedwithoutfertilization,resultinginseedless fruit.

Seedabortion:abscissionorstuntedgrowthofseed. Seedset:transitionfromanovuletoaseeduponfertilization.

Correspondingauthors:Ruan,Y-L. (yong-ling.ruan@newcastle.edu.au); Fernie,A.R. (fernie@mpimp-golm.mpg.de).

Donutrienttransportandpartitioningconstrainfruit andseedset?

Sexual reproductive strategies influence the nature of nutrientpartitioningto,andbetween,reproductive struc-tures (Box 1). Nevertheless, some general principlesof resource allocation can beidentified. Thus,irrespective of inflorescence phenology and growth pattern (Box 1), manipulating source–sink ratios demonstrates that photoassimilatelimitation isaprimarydriverofflower, fruitandseedabortioningrain[11,12]andfruit[13]crops. Carbonlimitationatfruitandseedset[14]appliestoother essentialnutrients.Ofparticularsignificanceisthe deliv-eryofaminonitrogencompoundstosustainphysiological carbon:nitrogen ratios essential for early reproductive development [15]. This scenario is illustrated by an increased seed set arising from enhanced phloem sap concentrationsofS-methylmethionineinpea(Pisum sati-vum)transformantsexpressingyeastS-methylmethionine permease1underthecontrolofaphloem-specificpromoter [16].

Positiveresponsesofearlyreproductivedevelopmentto increased source–sink ratios demonstrate that breeding for higher reproductive potential, through increasing flower numbers per inflorescence [11], has not been matchedbyabsoluteincreasesinnutrientlevelsreaching these additional reproductive units required to support theirdevelopmentthroughtoseedandfruitset.Ingeneral, photosynthetic leaves are the primary nutrient source supportingearlyreproductivedevelopment(Box1).Thus, factorsconstrainingnutrientsupplyfromasourceleaftoa given sink, within an array of competing sinks, could includesourceleafphotosynthesis,hydraulicconductance ofphloempathwayslinkingsourceandsink(Lo)andsink

hydrostaticpressure(Ps)(seeEquationsI–IVandFigureI

inBox2).

Although flower formation andfruit andseed set are carbon limited (see above), it does not follow that their carbondeficitsresultfrominadequateratesofsourceleaf photosynthesis. For instance, photoassimilate require-mentsofflowers,settingfruitandseedare,ataminimum, anorderof magnitudelessthanthe subsequentphaseof fillingfruitandseed,whichissink,notsource,limited(e.g. [12,13,17]).Indeed,thereareexamplesofsink-dependent repression of photosynthesis of source leaves supplying photoassimilatesfuelingseedandfruitset(e.g.[18]).Thus dampening sink repression of photosynthesis can be expectedtoresultinelevatedratesof seedandfruitset. Forinstance,decreasingsugar(glucose)repressionofleaf photosynthesis,bycompartmentingglucoseintomesophyll vacuolesthroughoverexpressingtonoplastmonosaccharide transporter 1(TPMT1), increased photoassimilate export andseedsize[19].Increasingthesinkcapacityofdeveloping wheat(Triticumaestivum)floretsbyoverexpressing ADP-glucose pyrophosphorylase under the control of a grain-specificpromoterledtohigherratesofseedsetsupported byincreasedflagleafphotosynthesis[20].Together,these findings indicate that photoassimilate limitation of early reproductive development is not imposed by leaf photo-synthesis(Re;seeEquationIinBox2).Aplausible

explana-tion for enhanced photosynthesis relieving C limitation duringearlyreproductivedevelopmentistoelevate hydro-staticpressureinleafphloem(Pe),whichinturnincreases

leaf–sinkhydrostaticpressuredifferentialstodrivehigher ratesofphloemimportintodevelopingreproductiveunits (seeEquationsII–IVinBox2).

There is a growing body of evidence that hydraulic conductancesof phloem(sieve tubes)adequatelyaccount for theobserved ratesof phloem transport(e.g. [21,22]), includingphloempathwaysservingdeveloping reproduc-tivestructures[23].Inaddition,thebufferingcapacityof thetransportphloem[24]canaugmentCflowstosupport seedand fruit set(e.g. [25]). Bycontrast,vascular path-ways within young fruit and seed contain xylem and phloemdifferentiating fromprovascular cells[26].Fluid mechanicspredictsthatthelowesthydraulicconductances areencounteredbyflowthroughplasmodesmata intercon-necting provascular cells and non-phloem cells forming symplasmicpathwaysdeliveringphloemsapinto develop-ingfruitletsandthematernaltissuesofseeds(e.g.[27]and Box2).Therelativelylowhydraulicconductancesof plas-modesmataaccountforthelargedifferencesinhydrostatic pressures(1MPa)detectedbetweensieveelementsand adjoiningvascularparenchymacellsofdevelopingwheat grainsand pointtowhere mostofthecontrol ofphloem transportresides[28](seeEquationIIinBox2).In addi-tion, these large hydrostatic pressure differentials are consistentwithbulkflowcontinuingintothenon-phloem symplasmicpathwayand,asaresult,Psislikelytoreside

innon-phloemcells(Box2).Consistentwiththis proposi-tion, overexpressing ADP-glucose pyrophosphorylase in developing wheat florets increased grain numbers set perspikelet [20],indicatingan enhancedrate ofphloem import by lowering Ps (see Equations II–IV in Box 2).

Nutrient transfer across the maternal–filial interface is Box1.Source–sinkframeworksaredefinedby

inflorescencephenologyandgrowthpatterns

Nutrientsourcesaredeterminedbyinflorescencephenology.Bud burstandearlyfloraldevelopmentindeciduouswoodyperennials largelydependsuponxylemdeliveryofnutrientsremobilizedfrom stemorrootstoragepoolsuntilemergingleavesfromvegetative budsreachphotosyntheticcompetence[115].Thereafter,fruitand seedsetareincreasinglysupportedbycurrentphotoassimilatesand other nutrientsexported fromphotosynthetic leavesthroughthe phloem(Box2).Bycontrast,floraldevelopmentandfruitandseed setofmanyherbaceouseudicots(e.g.[12])andgrasses(e.g.[17]) rely exclusivelyon nutrients deliveredthrough the phloem from sourceleaves.

Patternsofinflorescenceformationinfluencesource–sink relation-ships.Forexample,ingrassesatfloralevocationtheshootapical meristem of each branch (tiller) irreversibly converts to a floral meristem.Inordertoreachtheseapicallylocatedinflorescences, nutrientsaretransportedthroughaseriesofelongatinginternodes. Thedevelopingwheatspikeillustratesthecompetitivedisadvantage forfloretdevelopmentwithinthisconfiguration.Herebetween66% and75%oftheuppermostfloretswithineachspikeletatrophy[116]. However,followinganthesis,theintensecompetitionfornutrients abates as vegetative growthceases leavingdeveloping spikelets

[17]asthesolegrowthsinkwithexcessnutrientsflowingintostem storage[117].Thegrasspatternofinflorescenceformationcontrasts with one inwhichtheshoot apicalmeristemremainsvegetative whilstinflorescencesprogressivelyformfromlateralbudsinleaf axils.Asaresult,fromfloralbudinceptiononward,competitionfor nutrientsoccurswithinandbetweeninflorescencesaswellaswith ongoingvegetativegrowth[12,118].

transporter-mediated and turgor-regulated to link with phloemimport[29].Therefore,itisnosurprisethatseed setdependsuponadequatetransporteractivitiesatthese sites(e.g.[30–32]).Thesefindingshighlightopportunities, other than through manipulating photosynthesis, to increasenutrientavailabilityforfruitandseedset.These opportunities include exploiting mechanisms regulating hydraulicconductancesofplasmodesmata[33]toamplify ratesofphloemimportintofruitsandenhancingactivities oftransportersregulatingnutrientflowsintofilialtissues ofdevelopingseeds[29].

Regulationofseedandfruitsetthroughsugar metabolismandsignaling

Asdiscussedabove,plantreproductiondependsgreatlyon an adequate import of photoassimilates, whichfor most crop species is mainly in the form of sucrose. Efficient

utilizationof sucroseiscrucialforgametophyte develop-ment, fertilizationand coordinated development of filial andmaternaltissues,whichcollectivelydeterminesseed andfruitset.

Regulationofmalefertilitybysugars

Formostfloweringplants,pollinationisapre-requisitefor seedandfruitset.Aftergermination,pollentubeselongate throughthestyle andreleasetwosperm intotheovular embryosacfordoublefertilizationtoproducetheembryo andendosperm.Forpollentobeviable,itmustsynthesize sufficient starch as an energy source and cellulose and callosetobuildtheirinternalwall[1,34].Starch,cellulose andcalloseareallpolymerizedfromglucoseina-andb-1,4 andb-1,3linkages,respectively.Glucoseinpollencouldbe derivedfrom:(i)cellwallinvertase(CWIN)activity hydro-lyzing sucroseintoglucoseandfructose intheantheror Box2.Regulationofnutrienttransportandpartitioningtodevelopingfruitandseed

Nutrientsflowfromsourceleavestodevelopingfruitorseedsinks in thephloem andspecifically in sieve elements organized into longitudinalfilescalledsievetubes(ST,FigureI).Sugarsrepresent the major osmotica of sieve tube sap and phloem-loading capacitiesaccommodatethoseofsourceleafphotosynthesis.Thus ratesofsugarexport(Re,mols–1andseeFigureI)fromasource

leafcorrespondtonetratesofleafphotosynthesisandcollectively determineamountsofsugarsavailableforpartitioning(h)between sinksand,hence,theirratesofsinkimport(RiandseeEquationI).

A conclusion that equally applies to all phloem-transported nutrients.

Ri¼Reh [I]

Dependinguponplantspecies,phloemloadingmayfollowan apo-plasmicorsymplasmicpathway.Irrespective,nutrientsare accumu-latedtohighconcentrationsinleafphloemtoosmoticallycreatea hydrostaticpressureatthesiteofexport(Pe,MPa;FigureI).PhloemPis

dissipatedasflowproceedsalongSTconduitsbutprimarilyatthesink end(Ps)ofthetransportpathway(FigureI).Thisgeneratesa

hydro-staticpressuredifference(Pe–Ps)todriveabulk(volume)flowof

phloemsapfromsource tosinkatrates(J,m3_s–1_)modulatedby

hydraulicconductances(Lo,m3s–1MPa–1)ofthetransportpathway

(EquationII).

J¼ðPe-PsÞLo [II]

Forwallsandcoatsofsettingfruitsandseedsrespectively,unloading fromtheirSTsintosurroundingnon-vascularcellsoccursasabulk flowthroughinterconnectingplasmodesmata(FigureI).HencePsis

locatedinnon-vasculartissuesofthesesinksandsourcetosinkLois

largelydeterminedbyplasmodesmalLoformingthephloem

unload-ingpathway.Ratesofimport(Ri)ofanutrientintoaspecifiedfruitor

seedsinkdependupontheconcentration(C,molm–3_{)ofthenutrientin}

thephloemsapandJthroughphloemterminiandunloadingpathway (EquationIII).

Ri¼JC [III]

Giventheseconsiderations,partitioning(h)ofnutrientsbetweenan array(n)ofsinks,fromacommonpoolofSTnutrients,isafunctionof eachindividualbulkflow rate(Ri)intoa specifiedfruitor seed(j)

expressedasaproportionofthecombinedbulkflowratesintoall[(1) to(n)]fruitorseedwithinaninflorescence(EquationIV).

h¼ RiðjÞ SRið1Þþ





þRiðnÞ [IV] FW SC E C End SL ST Ps Pe Ps Ri Re Re Ri

Figure I.Nutrienttransport todeveloping fruitand seed.Following phloem loading(curvedarrows)intosievetubes(ST)ofsourceleaves(SL),nutrientsare exportedatgivenrates(Re)andconcentrations(C)bybulkflow(straightarrows)

drivenbydifferencesinhydrostaticpressuresgeneratedbysourceleaves(Pe)and

sinks(Ps).AproportionofSTsapisimported,atgivenrates(Ri),intowalls(FW)and

coats (SC) of setting fruit and seed respectively by bulk flow through plasmodesmata(curvedarrowsthroughgaps)linkingSTswithsurrounding non-vascularcells.Membranetransporters(circles)facilitatenutrienttransport(arrows throughcircles)fromtheSCandsubsequentuptakeintotheendosperm(End)and embryo(E)acrossthesymplasmicdiscontinuityatthematernal–filialinterface.

pollengrainapoplasm;(ii)degradationofstarchinanthers by a-amylase [35]; and (iii) sucrose cleavage by sucrose synthase(Sus)toproducefructoseandUDP-glucose (fruc-tose is convertible to glucose and UDP-glucose is an immediatesubstrateforcelluloseandcallosebiosynthesis) [34]. Transgenic suppression of a tapetum- and pollen-specificCWINgeneintobacco(Nicotianatabacum)results inunviablepollen,characterizedbylossofstarchandcell wallintegrity[36],demonstratingthecrucialroleofCWIN inpollendevelopment.Consistentwiththis,malesterility andhencegrainorfruitabortion,isattributableto reduc-tionofCWINandvacuolarinvertase(VIN)expressionin wheatunderdrought[37],ofCWINandhexose transpor-tersinrice(Oryza sativa)under coldstress[38]and dis-ruption of sucrose metabolism in tomato under heat stress[39].Indeed,geneticvariationindroughttolerance correlates withtheexpression levelof CWINandstarch abundanceinthepollen andovaries inwheat[40].Itis clearthatmaintainingthesucrosesupplyandits degrada-tion into hexoses is key to male fertility and seed and fruitset.

Invertasemodulatesseedandfruitset

Oneofthemostimportantrecentfindingsinreproductive biologyresearchisthesimilaritybetweenthebiochemical andmoleculareventsleadingtopollensterilityandovary abortion[2].Inmaize(Zeamays),thefertilityoftheovaries has a greater influence on kernel number than that of pollen,whichisdifferenttothescenarioinwheat,riceand barley (Hordeum vulgare) where pollen development is morepronetostressthanovaries[1,4].

In maize ovaries, phloem-imported sucrose supplies carbonforstarchaccumulationinovarywallsandpedicels andto generatehigh glucose concentrationsthatflow to embryo sacs following sucrose hydrolysis by CWIN in pedicelsandbyVINinnucellartissues[41].Upon impos-ingawaterdeficitfivedaysbeforeanthesis,sucroseimport isblockedowingtoleafphotosynthesisbeinginhibitedand ovarywallstarchreservesareremobilized,butthesesoon becomedepletedifdroughtpersistsforseveraldays. Con-comitantly,CWINandVINactivitiesandglucose concen-trations decrease, leading to severe ovary abortion and yield loss [42]. Abortion is triggered by expression of programmed cell death (PCD) genes encoding a ribo-some-inactivating protein (RIP2) and phospholipase D (PLD1). Feeding sucrose to water-stressed maize plants reverses the decline of CWIN (Incw2) and VIN (Ivr2) expression levels and glucose concentrations as well as blocking expression of PCD genes, RIP2 and PLD1, in ovaries to restore kernel number by about 70% [42]. Remarkably, these are the only sugar-responsive genes amongmanycandidategenessurveyed[4],demonstrating that an invertase-mediated glucose signaling pathway regulatingPCDistheprimarybiochemicalandmolecular mechanism controlling maize ovary abortion under droughtconditions.

Significantly,thismechanismappearstobeconserved between monocot and eudicot herbaceous species. For example,RNAi-mediatedsilencingof aCWIN geneLin5

aggravatesfruitabortionintomato[43],whereaselevation of CWIN activity enhances fruit and seed development

[44].Furthermore,incomparisonwithaheat-susceptible genotype, heat-tolerant tomato exhibits greater sucrose importinto,and invertaseactivities within,young fruit, accompaniedbyalowerlevelofexpressionofaPCDgene,

LePLDa1[7].Theheat-inducedor-enhancedexpressionof heat-shock proteins (HSPs) is hypothesized to protect CWIN and VIN from misfolding for correct targeting andfunction[7].

Together, the above analyses support the following model of seed and fruit set regulated by sugars (Figure1).Phloem-importedsucroseservesasaprimary signal sensed by invertase that generates glucose to repressthePCDpathwayontheonehandandtopromote celldivisionoffilialandfruittissuesontheother(see[2,3]), therebyallowingseedandfruitsettoproceed.Themodel changes under stress where a decreased glucose level activatesaPCDpathwayleadingtoseedandfruit abor-tion.Thewholeprocessisconditionedbystarch degrada-tionandHSPexpression(seeabove)andprobablythrough interactionwithhormones(seebelow).

Hormonalregulationoffruitandseedset

Uponflowerfertilization,fruitandseedundergo concomi-tantdevelopment;however,incontrasttofruit,whichcan developintheabsenceofpollination,seeddevelopmentis more strictly dependent on successful fertilization. Seed development comprises endosperm proliferation and embryo growthand both processes show multihormonal regulation by auxins, cytokinins, gibberellins (GAs) and brassinolides [45]. Mutations affecting auxin perception [46] or transport [47] resulted in abnormal embryo morphologies whereasmutantsalteredincomponents of the cytokinin signaling pathway produced seeds of enlargedsize[48].Likewise,alterationofgibberellicacid responsegenes[49]resultsinincreasedseedsizeandalow level of active brassinolides induces altered seed shape [50].Accordingly,defectsinthebiosynthesisofrice bras-sinolide or perception genes also result in reduced seed length [51]. However, determining how these hormones

INV PCD

Sucrose Glucose Set

Cell division Starch

INV PCD

Sucrose Glucose Abort

Cell division Starch

(a)

(b)

Optimal

Stressful

Figure1.Amodelforregulationofseedandfruitsetthroughsugarsignaling.(a) Underoptimalconditions,phloemunloadedsucroseishydrolyzedbyinvertase (INV)inovariesandovules.Theresultantglucosefunctionsasasignaltorepress programmedcelldeath(PCD)genesandtopromotecelldivision,whichtogether leadtoseedandfruitset.Starchreservemaybe remobilizedtosupplement glucoseproduction,particularlyundermildstressconditions.(b)Undersevere stress,phloemimportofsucroseisblocked,whichinturndecreasesINVactivity. This,togetherwithdepletionofstarchreserve,reducesglucoselevelsthatactivate thePCDpathwayandinhibitcelldivision.Consequently,seedandfruitabort.

interactto regulateseedsetanddevelopmenthas sofar remainedelusive.

Auxinandgibberellinarethekeyplayersinfruit initia-tionfollowingfertilization[52–55].Thisviewissupported bythefactthatexogenousapplicationortransgenic eleva-tionofthesehormonesleadtotheuncouplingoffruitset fromfertilizationresultinginthedevelopmentof parthe-nocarpicfruit[56,57].Transcriptionalregulatorsfromthe ARF(AuxinResponseFactors)andAux/IAA(Auxin/indole aceticacid)typeoftranscriptionfactorareencodedbytwo largegenefamilies[58,59]thatareknowntochannelauxin signaling to specific physiological responses [60–63]. A significant advance in our understanding of the auxin-dependent mechanism underlying fruit set came from mutations or transgenic manipulation of specific ARF

andAux/IAAgenesleadingtothedevelopmentof parthe-nocarpicfruitsinbothtomatoandArabidopsis

(Arabidop-sis thaliana) [61,64,65]. These studies identified auxin signalingasoneoftheearlyeventsinthefruitinitiation cascade. Downregulation of thetomato IAA9 and ARF7 resultsinuncouplingfruit setfrompollinationand ferti-lization,givingrisetoparthenocarpicfruit,thus suggest-ingthatbothgenesencodenegativeregulatorsoffruitset [61,65].Althoughinteraction betweenARFand Aux/IAA proteinsplaysapivotalroleinregulatingauxinresponses [60,66], it is still not known whether IAA9, ARF8 and ARF7controlfruit setthroughcommonordistinct path-ways.Furtherinvestigationshoulduncoverwhether het-erodimerizationbetweenARF8/ARF7andIAA9ispartof the controlmechanism regulating fruit initiation. More-over,giventhatARFgenescanbepotentiallyregulatedby siRNAs at both the transcriptional and post-transcrip-tional levels [67,68], it would be important to know whether the auxin-dependent fruit set is impacted by epigenetic regulation. If proved, this level of regulation mayexplaintheenvironmentallyinducedvariationinfruit setandparthenocarpyobservedinentire,anaturaltomato mutant impaired in Sl-IAA9 function [61,69], and the initiation of fruit developmentbeforeand independently offertilization.

Insupportofmultihormonalcontroloffruitset,thereis strongevidencesuggestingthattheroleofauxinis facili-tated by synergistic activity with gibberellins [70,71]. Notably, GA treatmentof unpollinated ovaries triggers fruitinitiationwithoutimpactingtheexpressionofauxin signalinggenes [72],whereasauxin-inducedfruit devel-opmentissignificantlyreducedbysimultaneous applica-tion of GAbiosynthesis inhibitors [70].Together, these datasuggestthatauxinmayactbeforeGAandthatthe effectofauxinismediatedatleastpartlybyGA(Figure2). In linewith thishypothesis,GAbiosynthesis genesare upregulatedafterpollinationanduponauxintreatment of emasculated flowers [70]. However, in tomato, each hormone seems to play aspecific rolegiven that auxin application results in alarge number of pericarp cells, whereasGAtreatmentresultsinfewerpericarpcellsbut ofalargersize[73].Interestingly,concomitanttreatment with both hormones results in the formation of fruits similartopollination-inducedseededfruits[73], suggest-ingthatGAandauxinarebothrequiredfornormalfruit development.

Besides the established role of auxin and GA, an increasingnumberofstudies,buildingonglobal transcrip-tomicprofiling (seebelow),pointtotheputative involve-mentofotherhormonessuchasethyleneandabscisicacid (ABA)inregulatingfruitformation[72,74,75].Ithasbeen reportedthatethylenebiosynthesisandethylenesignaling genesdecreaseafterpollination.Atthesametime,genes related to ABA biosynthesis decreased andthe opposite behavior was found for ABA degradation-related genes [76].Thesefindingssuggestthatnormalfruitdevelopment dependsoninductionofGAandauxinresponses,whereas ethylene and ABA responses are attenuated (Figure 2). Nevertheless, thedirect contribution ofethylene to fruit set per se has remained largely unstudied. By contrast, exogenous application of cytokinincan induce partheno-carpicfruitformationinarangeofagriculturalspeciesbut the mechanism by which these hormones may interact withauxinorGAduringfruitsetremainstobeelucidated. Transcriptionalandmetabolicregulationoffruitand seedset

Inadditiontothemorehypothesis-drivenstudiesoffruit andseedsetdetailedabove,recentyearshavealsoseenthe adoption of broad technologies to assess transcriptional andmetabolicprogramsoffruitdevelopmentandripening [77–81],whichhaverevealedbothconservedand species-specific changes that underpin both processes. Here we focus onchangesduringearlyfruit development– speci-ficallyfruitandseedset[82–85]–usingtomatoasamodel becausethisspecieshasthemostpertinentdata.

Consistent with analyses in the previous sections, recentstudieshaveidentifiednotonlyauxinandethylene

Ethylene ABA

Gibberellins Pollination and fertilization

Fruit set Auxins Cytokinin ? ? ? ?

Figure 2.Amodel formultihormonalregulation offruit set. Pollination and fertilizationresultinincreasedlevelsofbothauxinandgibberellins(GA),which triggersfruitgrowththroughstimulationofcelldivisionandexpansion.Auxincan stimulatefruitseteitherdirectlyorviainducingGAbiosynthesis.Eachhormone seemstoplayaspecificrolegiventhatauxinapplicationresultsinahighnumber ofpericarpcells,whereasGAtreatmentresultsinfewerbutlargerpericarpcells. Naturalfruitsetseemstorequirebothhormonesgiventhatonlyparthenocarpic fruitsinducedbyconcomitantauxinandGAtreatmentaresimilarto pollination-inducedseededfruits.Putativeinvolvementofethyleneandabscisicacid(ABA)in regulatingfruitformationhasbeenmainlysuggestedbytranscriptomicstudies.In particular,duringthetransitionfromanthesistopost-anthesis,ethylene-related genes,alongwiththoserelatedtoauxin,accountformostofthechangesamong allphytohormone-relatedgenes,thusindicatingthatethylenemustplayanactive, butyetnotunderstood,roleinfruitset.Exogenousapplicationofcytokinincan induceparthenocarpicfruityettheunderlyingmodelofactionremainsunknown. Brokenarrowsrepresenteffectsthatarestillnotsustainedbysolidandmultiple experimentaldata.

signaling,butalsophotosynthesisandsugarmetabolism, as major events of the fruit-set program (Box 3) [75]. Transcriptome analyseshaverevealed thatduring polli-nated fruit setalmost5%of thedifferentially expressed genesarerelatedtohormoneresponseandmetabolism.As mentionedabove,Aux-IAAandauxintranscriptionfactor (ARF) genes show a dramatic shift in their expression levels across natural fruit development, suggestingthat thesetranscriptionalregulatorsplayanactiveroleinthis developmentalprocess.Moreover,downregulationofIAA9

resultsinfeedbackregulationofseveralARFandAux-IAA

genes[75].Inkeepingwiththeimportanceofauxin,recent transgenic workhas implicated both the auxin receptor homologTIRandARF7infruitset[86,87].Moreover,the highnumberofethylene-relatedgenesobservedtochange suggests that, inadditionto the well-established role of auxinandGAs[53,73],ethylene isalsolikelyto playan activeroleinfruitset,albeitonethatisantagonistictothat ofauxinandGA[72].Downregulationoftheexpressionof auxin-relatednegativeregulatorsoffruit-setgenesoccurs earlierthanthatofethylene-relatedgenes,suggestingthe

existenceofatemporaldependenceonhormoneactionfor theflower-to-fruittransitiontoproceed.

Perhapssurprisingwasthefindingthattranscriptomic profiling across the flower-to-fruit transition identified manygenesthatarecommontobothpollination-induced and pollination-independent fruit set, suggesting thata relatively smallnumber of genesareresponsible forthe differencesintheseprocesses(Box3).Inthesamestudy, gaschromatography–massspectrometry-basedmetabolite profilingwasappliedtoextendpreviouscharacterization ofmetabolicshiftsoccurringduringfruit ripening[78]to encompasstheearlyeventsoftheprocess[75].Therewere proportionally far more changes in metabolite than in transcriptlevels;however,(i)thosemetabolitesand tran-scripts associated withphotoassimilatemetabolism, and (ii)thosesugarsandtranscriptsassociatedwithboththe photosyntheticapparatusandenzymesoftheCalvincycle, revealed common regulation at both levels. In addition, statisticalanalyseswereabletohighlighttranscriptional regulationofputrescineandspermidinelevelsinkeeping with previousdescriptions of the kineticsof their levels duringearlyfruitdevelopment[88],andsimilarlyforthe regulationofthepathwaysofascorbatebiosynthesis[78]. Given the widely documentedcrosstalk between hor-moneandsugarsignaling,itishighlyinterestingthatthe

IAA9 antisense plants were also characterized as being upregulatedintheSuspathwayofsucrosedegradationin youngtomatofruit.Thispathwayismoreenergyefficient than that mediated by INV [89] and is likely to bethe prominent pathway of sucrose degradation in the later stages of development of many heterotrophic tissues, including tomato [90]. Future experiments, examining therolesofasubsettherecentlyclonedSWEET transpor-ters–someofwhichactassucroseeffluxtransporters[91], mayfurtherilluminatethefruit-setprocess regulatedby sugars.ReturningtotheIAA9transgenics,elevationofthe transcriptsofthispathwaycould,atleastpartially,explain theprecociousnatureoffruitdevelopment.Theelevation ofsugarlevelsintheantisenselinescouldbeduetoeithera moreefficientunloading of photoassimilateinto fruit,as observed on the introgression of awild speciesallele of CWIN[92],ortoanincreaseinphotoassimilateproduction by fruits. Interestingly, increasing auxin sensitivity via downregulationoftomatoIAA9hasbeenreportedto pro-motevascularbundlesdevelopment[61].Themore abun-dant vasculature may enhance sink strength and sugar supplytothefruit andhencedirectlylinks theeffects of auxin andphotoassimilates on theinitiation of fruit set andearly development.Strongactivation of photosynth-esis-relatedgenesduringfruitsetinIAA9-downregulated linesisamajorphenomenon[75].Strikingly,activationof photosynthesis-related genes is delayed to the post-anthesis stage in the wild type but takes place at the bud stage in AS-IAA9 [75]. Many recent studies have endorsed the prevailing opinion that fruit growth and metabolismarepredominantlysupportedby photoassimi-late supply from source tissues [93,94]. However, it is importanttonotethatthecarpelofthefruitisessentially a modified leaf thathas folded into a tubularstructure enclosingtheovules[95]andthatcellsindevelopingfruit containphotosyntheticallyactivechloroplastsandexpress Box3.Pollination-dependentand-independentfruitset

A detailed comparison of transgenic plants in which IAA9 was antisense inhibited, provoking pollination-independent fruit set, with the wild-typepollination-dependentcontrol atthe transcrip-tomicandprimarymetabolitelevelsprovedhighlyrevealing[75]. Manygeneswerecommonlyregulatedinbothfruit-settypes,with onlyasmallsubsetbeingIAA9dependent.Thusminimalchangesin gene expression clearly have dramatic consequences on the developmentalfateoftheflower.Indeedonlyatthebud-to-anthesis transition weretheremajor changesintranscripts,with anthesis being the phase at which the most genes were differentially expressedbetweengenotypes,whichpotentiallyexplainswhy AS-IAA9entersthefruitdifferentiationprocessearlierthanthewildtype and withno requirement fora pollinationand fertilizationsignal

[75].

Aside from the comparative aspect of this study, several interestingfeaturesofwild-typedevelopmentwerecharacterized, whichexpandeduponotherearlyfruit-developmentstudies[119]. Intriguingly,manytranscription factorswererecruitedduringthis period with the MADS-box genes being particularly strikingly downregulated inyoungfruitcomparedwith atthe ovarystage. When taken together with observations from transgenic experi-mentation[120–122],theseobservationsstronglyimplicateTomato Agamous1 (TAG1), Tomato Agamous-like6 (TAGL6) and Tomato MADS-box29(TM29)genesasactivatorsofthefruitsetprocess. WhetherIAA9and/orauxinperseactinconcertorindependentlyof theseMADS-boxgenesincontrollingfruitdevelopmentremainsto beclarified.Inadditiontothesetranscriptionfactors,alargenumber of cell division, protein biosynthesis and cellwall-related genes wereupregulatedpost-anthesisinthewildtype(butearlyduring pollination-independent fruit set). However, the links between transcriptionfactorexpressionanddownstreamprocessesremain tobeclarifiedinmostinstancesandarenotaswellcharacterizedas thoseorchestratinglaterstagesoffruitdevelopment[102].

Duringpollination-inducedfruitset,photosynthesis-relatedgenes were downregulated during the transitionfrom bud to anthesis, whereas they were generally upregulated in the period from anthesistopost-anthesis.Bycontrast,photosynthesis-relatedgenes werestronglyactivatedinAS-IAA9throughoutfruitsetandnotably, allthesegeneswereupregulated inthe AS-IAA9lines compared with the wild type. Inlinewith the elevation of photosynthesis-related genes, downregulation of IAA9 was associated with a dramatic upregulation of genes involved in sucrose catabolism, whichalsosuggeststhatthesegenesplayanimportantroleinthe flower-to-fruittransitionprocesses[75].

bothnuclear-encodedandplastid-encodedgenesfor photo-synthetic proteins [96]. Furthermore, a combination of indirectevidenceprovidessupportforthehypothesisthat duringtheearlystageoffruitdevelopment,photosynthesis itself may provide a considerable contribution to both metabolismandgrowthoftheorgan[97].Recentstudies haveindicatedthatintomatopericarpcells,theinduction ofgenesrelatedtophotosynthesisandchloroplast biogen-esispositivelycorrelatewithchloroplastnumbersandcell size[98].However,amorerecentdirectstudyoftheroleof fruit photosynthesis by downregulating chlorophyll bio-synthesis inafruit-specificmanner revealedthat photo-synthesiswasunimportantforfruitgrowthbutratherwas essentialforseedset[99].Therefore,itwouldappearthat fruitphotosynthesismaycontributetothecorrectspatial distributionofsugarsrequiredtotriggerthecorrect hor-monal signaling events underlying seed set. That said, unlikethesituationdescribedaboveforfruitset,detailed transcriptomic and metabolomic analysis of seed set is lacking.

Futureperspectives

Inrecentyearsourlevelofunderstandingofthemolecular eventsatthetranscriptional,biochemical, hormonaland metabolite levels underlying fruit and seed set has increased considerably. Although to date cereal grains andtomatofruithavepredominantlybeenusedasmodels, thereisagrowingbodyofknowledgeaboutotherseedand fruitsystems.Thelowhydraulicconductanceencountered byplasmodesmataconnectingprovascularcellswith non-vascularcellsinfruitletandseedmaternaltissues repre-sentsabottleneckforassimilateimportintotheseyoung sinks. Thus, enhancingtheir phloemdifferentiation and plasmodesmalconductancecouldenhancerateofphloem importand seedandfruit set. Inv-mediatedglucose sig-naling regulating PCD and IAA-signaling controlling parthenocarpy and vascular development are two new avenuesforimprovingseedandfruitsetand,hence,crop yield, and present opportunities to dissect interactions between sugar-and hormone-signaling pathways under-pinningseedandfruitset.

Todate,mostpublishedstudiesoftranscriptionaland metabolic regulation areof relatively low resolution at both spatial and temporal levels and are furthermore restricted incoverageof thevariouscellmolecular enti-ties.However,arangeofnascentandemerging technol-ogiesnowallowsustorefineouranalyticalabilityfurther tocopewithissuessuchassubcellularcompartmentation and the behavior of different cell types, which should increaseourunderstandingintheyearsahead.For exam-ple, laser capture microdissection has already been employed to gather tissue-specific transcriptomes in tomato fruit [100] and rice seed [101], and the use of RNAseqtechnologiesandhighlysensitivereal-time quan-titative PCR methods [80]have dramatically enhanced transcriptomecoverage.Theirapplicationtothe floral-to-fruit or floral-to-seed transition should allow us to enhanceourunderstandingofgeneregulatorynetworks to at least the level of those characterized for later stages of fruit development (for a recent review see [102]).Similarly,methodsfortemporallyspecificgenetic

perturbationsoffruitgeneticsusingvirus-induced gene-silencingprotocolshavebeengenerated[103]andrecently havealsobeenoptimizedforyoungfruit[104].Although changesintheprimarymetabolismofyoungfruitarenow well documented it will be important to look also at specializedmetabolism,particularlyinlightoftherecent report that the absence of certain flavonoids leads to parthenocarpic fruit [105]. Several methods have been establishedthatarecapableof this[106–108].However, althoughthesehavebeenusedinthestudyoffruit devel-opment, to datetheir use hasfocusedon laterripening stages.Alongthesamelines,methodscapableofdetecting all major classes of phytohormones have been recently established[109].Theearlydevelopmentoftomatofruit wouldappeartobeaperfectbiologicalsituationfortheir application.

Therecentcompletionofthegrape(Vitisvinifera) straw-berry (Fragaria vesca) and tomato genomes [110–112] rendersthemevenbettermodelsystemsinwhichtostudy thefruitsetprocess.Itislikelythatnotonlytransgenicor mutational studies but also theincreasingly widespread adoptionofnaturalgeneticdiversity[113]willultimately allowustotakesuchbroadstudiesfromthedescriptiveto thefunctionallevel[114]andmayultimatelyaidin breed-ing strategies intent on increasing crop yield and food security.

Acknowledgments

WethanktheAustralianResearchCouncil(DP110104931;DP120104148) andFrance’sLaboratoired’Excellence(LABEX)entitledTULIP (ANR-10-LABX-41)forfundingsupportandCOSTActionFA1106forprovidingthe networkingframework.

References

1 Barnabas,B.etal.(2008)Theeffectofdroughtandheatstresson reproductiveprocessesincereals.PlantCellEnviron.31,11–38 2 Ruan, Y.L. et al. (2010) Sugar input, metabolism, and signaling

mediatedbyinvertase:roles indevelopment,yield potential,and responsetodroughtandheat.Mol.Plant3,942–955

3 Weber, H. et al. (2005) Molecular physiology of legume seed development.Annu.Rev.PlantBiol.56,253–279

4 Boyer, J.S. and McLaughlin,J.E. (2007) Functional reversion to identifycontrollinggenesinmultigenicresponses:analysisoffloral abortion.J.Exp.Bot.58,267–277

5 Suwa,R.etal.(2010)Hightemperatureeffectson photosynthate partitioningandsugarmetabolismduringearexpansion inmaize (ZeamaysL.)genotypes.PlantPhysiol.Biochem.48,124–130 6 Zinn, K.E. et al. (2010) Temperature stress and plant sexual

reproduction: uncovering the weakest links. J. Exp. Bot. 61, 1959–1968

7 Li,Z.M.etal.(2012)Highinvertaseactivityintomatoreproductive organs correlates with enhanced sucrose import into, and heat toleranceof,youngfruit.J.Exp.Bot.63,1155–1166

8 Xu,S-M.etal.(2012)Overexpressionofapotatosucrosesynthase geneincottonacceleratesleafexpansion,reducesseedabortion,and enhancesfiberproduction.Mol.Plant5,430–441

9 Bita, C.et al.(2011)Temperature stress differentiallymodulates transcriptioninmeioticanthersofheat-tolerantandheat-sensitive tomatoplants.BMCGenomics12,384

10Eckardt,N.A.etal.(2009)Thefutureofscience:foodandwaterforlife. PlantCell21,368–372

11Reynolds,M.etal.(2009)Raisingyieldpotentialinwheat.J.Exp.Bot. 60,1899–1918

12Egli,D.B.(2010)SOYPOD:Amodeloffruitsetinsoybean.Agron.J. 102,39–47

13Wubs,A.M.etal.(2011)Quantifyingabortionratesofreproductive organs and effects of contributing factors using time-to-event analysis.Funct.PlantBiol.38,431–440

14 Jablonski,L.M.etal.(2002)PlantreproductionunderelevatedCO2

conditions:ameta-analysisofreportson79cropandwildspecies.New Phytol.156,9–26

15 Lawlor,D.W.(2002)Carbonandnitrogenassimilationinrelationto yield:mechanismsarethekeytounderstandingproductionsystems. J.Exp.Bot.53,773–787

16 Tan,Q.etal.(2010)IncreasedphloemtransportofS-methylmethionine positivelyaffectssulfurandnitrogenmetabolismandseeddevelopment inpeaplants.PlantPhysiol.154,1886–1896

17 Foulkes, M.J. et al. (2011) Raising yield potential of wheat. III. Optimizing partitioning to grain while maintaining lodging resistance.J.Exp.Bot.62,469–486

18 Urban,L.etal.(2008)Interpretingthedecreaseinleafphotosynthesis duringfloweringinmango.TreePhysiol.28,1025–1036

19 Wingenter, K. et al. (2010) Increased activity of the vacuolar monosaccharidetransporterTMT1alterscellularsugarpartitioning, sugarsignaling,andseedyieldinArabidopsis.PlantPhysiol.154, 665–677

20 Smidansky,E.D.etal.(2007)ExpressionofamodifiedADP-glucose pyrophosphorylase large subunit in wheat seeds stimulates photosynthesisandcarbonmetabolism.Planta225,965–976 21 Jensen,K.H.etal.(2012)Universalityofphloemtransportinseed

plants.PlantCellEnviron.35,1065–1076

22 Froelich,D.R.etal.(2011)Phloemultrastructureandpressureflow: sieve-element-occlusion-related agglomerations do not affect translocation.PlantCell23,4428–4445

23 Zhang,C.etal.(2005)13_{C-photosynthateaccumulationinJapanese}

pearfruitduringtheperiodofrapidfruitgrowthislimitedbythesink strengthoffruitratherthanbythetransportcapacityofthepedicel.J. Exp.Bot.56,2713–2719

24 Gould, N. et al. (2004) Direct measurements of sieve element hydrostatic pressure reveal strong regulation after pathway blockage.Funct.PlantBiol.31,987–993

25 Hiyane,R.etal.(2010)Sucrosefeedingreversesshade-inducedkernel lossesinmaize.Ann.Bot.106,395–403

26 Esau,K.(1965)VascularDifferentiationinPlants,Holt,Rinehartand Winston

27 Werner,D.etal.(2011)Adualswitchinphloemunloadingduring ovuledevelopmentinArabidopsis.Protoplasma248,225–235 28 Fisher,D.B.andCash-Clark,C.E.(2000)Sievetubeunloadingand

post-phloemtransportoffluorescenttracersandproteinsinjected into sieve tubes via severed aphid stylets. Plant Physiol. 123, 125–137

29 Zhang,W.H.etal.(2007)Nutrientloadingofdevelopingseeds.Funct. PlantBiol.34,314–331

30 Schmidt,R.etal.(2007)TheaminoacidpermeaseAAP8isimportant forearlyseeddevelopmentinArabidopsisthaliana.Planta226,805– 813

31 Sanders,A.etal.(2009)AAP1regulatesimportofaminoacidsinto developingArabidopsisembryos.PlantJ.59,540–552

32 Li,F.etal.(2011)Characterizationofsucrosetransporterallelesand theirassociationwithseedyield-relatedtraitsinBrassicanapusL. BMCPlantBiol.11,168

33 Burch-Smith, T.M. and Zambryski, P.C. (2012) Plasmodesmata paradigmshift:regulationfromwithoutversuswithin.Annu.Rev. PlantBiol.63,239–260

34 Cai, G. et al. (2011) Distribution of callose synthase, cellulose synthase,andsucrosesynthaseintobaccopollentubeiscontrolled indissimilarwaysbyactinfilamentsandmicrotubules.PlantPhysiol. 155,1169–1190

35 Castro,A.andCle´ment,C.(2007)Sucroseandstarchcatabolismin theantherofLiliumduringitsdevelopment:acomparativestudy amongtheantherwall,locularfluidandmicrospore/pollenfractions. Planta225,1573–1582

36 Goetz, M. et al. (2001) Induction of male sterility in plants by metabolicengineeringofthecarbohydratesupply.Proc.Natl.Acad. Sci.U.S.A.98,6522–6527

37 Koonjul,P.K.etal.(2005)Selectivetranscriptionaldown-regulationof antherinvertasesprecedesthefailureofpollendevelopmentin water-stressedwheat.J.Exp.Bot.56,179–190

38 Oliver,S.N.etal.(2007)ABAregulatesapoplasticsugartransport andisapotentialsignalforcold-inducedpollensterilityinrice.Plant CellPhysiol.48,1319–1330

39Pressman,E.etal.(2002)Theeffectofheatstressontomatopollen characteristics is associated with changes in carbohydrate concentrationinthedevelopinganthers.Ann.Bot.90,631–636 40Ji,X.etal.(2010)Importanceofpre-anthesisanthersinkstrengthfor

maintenanceofgrainnumberduringreproductivestagewaterstress inwheat.PlantCellEnviron.33,926–942

41McLaughlin,J.E.andBoyer,J.S.(2004)Glucoselocalizationinmaize ovarieswhenkernelnumberdecreasesatlowwaterpotentialand sucroseisfedtothestems.Ann.Bot.94,75–86

42McLaughlin, J.E. and Boyer, J.S. (2004) Sugar-responsive gene expression, invertase activity, and senescence in aborting maize ovariesatlowwaterpotentials.Ann.Bot.94,675–689

43Zanor,M.I.etal.(2009)RNAinterferenceofLIN5intomatoconfirms itsroleincontrollingbrixcontent,uncoverstheinfluenceofsugarson thelevels offruit hormones,anddemonstratestheimportance of sucrosecleavagefornormalfruit developmentandfertility.Plant Physiol.150,1204–1218

44Jin,Y.etal.(2009)Posttranslationalelevationofcellwallinvertase activitybysilencingitsinhibitorintomatodelaysleafsenescenceand increasesseedweightandfruithexoselevel.PlantCell21,2072–2089 45Sun,X.etal.(2010)Transcriptionalandhormonalsignalingcontrolof Arabidopsisseeddevelopment.Curr.Opin.PlantBiol.13,611–620 46Shen,W.H.etal.(2002)NullmutationofAtCUL1causesarrestin

earlyembryogenesisinArabidopsis.Mol.Biol.Cell13,1916–1928 47Tanaka, H. et al. (2006) Spatiotemporal asymmetric auxin

distribution:a meanstocoordinateplant development.Cell.Mol. LifeSci.63,2738–2754

48Riefler,M.etal.(2006)Arabidopsiscytokininreceptormutantsreveal functionsinshootgrowth,leafsenescence,seedsize,germination, rootdevelopment,andcytokininmetabolism.PlantCell18,40–54 49Roxrud,I.etal.(2007)GASA4,oneofthe14-memberArabidopsis

GASAfamily ofsmall polypeptides,regulates flowering andseed development.PlantCellPhysiol.48,471–483

50Takahashi, N. et al.(2005) shk1-D, a dwarfArabidopsis mutant caused by activation of the CYP72C1 gene, has altered brassinosteroidlevels.PlantJ.42,13–22

51Tanabe,S.etal.(2005)AnovelcytochromeP450isimplicatedin brassinosteroidbiosynthesisviathecharacterizationofaricedwarf mutant,dwarf11,withreducedseedlength.PlantCell17,776–790 52Olimpieri,I.etal.(2007)Tomatofruitsetdrivenbypollinationorby

the parthenocarpic fruit allele are mediated by transcriptionally regulatedgibberellinbiosynthesis.Planta226,877–888

53Pandolfini,T.etal.(2007)Moleculardissectionoftheroleofauxinin fruitinitiation.TrendsPlantSci.12,327–329

54Serrani,J.C.etal.(2010)Inhibitionofauxintransportfromtheovary orfromtheapicalshootinducesparthenocarpicfruit-setintomato mediatedbygibberellins.PlantPhysiol.153,851–862

55Hazra,P.etal.(2010)Alteredgibberellinandauxin levelsinthe ovaries in the manifestation of genetic parthenocarpy in tomato (Solanumlycopersicum).Curr.Sci.99,1439–1443

56Rotino, G.L. et al. (1997) Genetic engineering of parthenocarpic plants.Nat.Biotechnol.15,1398–1401

57Carmi,N.et al.(2003) Inductionof parthenocarpyintomato via specificexpressionoftherolBgeneintheovary.Planta217,726–735 58Wu,J.etal.(2011)Identification,isolationandexpressionanalysisof auxinresponsefactor(ARF)genesinSolanumlycopersicum.Plant CellRep.30,2059–2073

59Audran-Delalande, C. et al. (2012) Genome-wide identification, functionalanalysisandexpressionprofilingofAux/IAAgenefamily intomato.PlantCellPhysiol.53,659–672

60Guilfoyle,T.J.andHagen,G.(2007)Auxinresponsefactors.Curr. Opin.PlantBiol.10,453–460

61Wang,H.etal.(2005)ThetomatoAux/IAAtranscriptionfactorIAA9 isinvolvedinfruitdevelopmentandleafmorphogenesis.PlantCell 17,2676–2692

62Chaabouni, S. et al. (2009) Sl-IAA3, a tomato Aux/IAA at the crossroadsofauxinandethylenesignallinginvolvedindifferential growth.J.Exp.Bot.60,1349–1362

63Deng,W.etal.(2012)ThetomatoSlIAA15isinvolvedintrichome formationandaxillaryshootdevelopment.NewPhytol.194,379–390 64Goetz, M. et al. (2007) Expression of aberrant forms of AUXIN RESPONSE FACTOR8 stimulates parthenocarpy in Arabidopsis andtomato.PlantPhysiol.145,351–366

65deJong,M.etal.(2009)TheSolanumlycopersicumauxinresponse factor7(SlARF7)regulatesauxinsignalingduringtomatofruitset anddevelopment.PlantJ.57,160–170

66Dharmasiri,N.andEstelle,M.(2004)Auxinsignalingandregulated proteindegradation.TrendsPlantSci.9,302–308

67Wang,J.W.etal.(2005)Controlofrootcapformationby microRNA-targetedauxinresponsefactorsinArabidopsis.PlantCell17,2204– 2216

68Ru,P.etal.(2006)Plantfertilitydefectsinducedbytheenhanced expressionofmicroRNA167.CellRes.16,457–465

69Zhang,J.etal.(2007)Asingle-basedeletionmutationinSlIAA9gene causestomato(Solanumlycopersicum)entiremutant.J.PlantRes. 120,671–678

70Serrani, J.C. et al. (2008) Auxin-induced fruit-set in tomato is mediatedinpartbygibberellins.PlantJ.56,922–934

71deJong,M.etal.(2009)Theroleofauxinandgibberellinintomato fruitset.J.Exp.Bot.60,1523–1532

72Vriezen,W.H.etal.(2008)Changesintomatoovarytranscriptome demonstratecomplexhormonalregulationoffruitset.NewPhytol. 177,60–76

73Serrani, J.C.et al. (2007) Gibberellin regulationof fruit setand growthintomato.PlantPhysiol.145,246–257

74Pascual,L. etal.(2009) Transcriptomicanalysisoftomato carpel developmentrevealsalterationsinethyleneandgibberellinsynthesis duringpat3/pat4parthenocarpicfruitset.BMCPlantBiol.9,67 75Wang,H.etal.(2009)Regulatoryfeaturesunderlying

pollination-dependentand-independenttomatofruitsetrevealedbytranscript andprimarymetaboliteprofiling.PlantCell21,1428–1452 76Nitsch,L. et al.(2009)Abscisic acidlevelsin tomatoovariesare

regulatedbyLeNCED1andSlCYP707A1.Planta229,1335–1346 77Alba,R.etal.(2005)Transcriptomeandselectedmetaboliteanalyses

reveal multiple points of ethylene control during tomato fruit development.PlantCell17,2954–2965

78Carrari, F. et al. (2006) Integrated analysis of metabolite and transcriptlevelsrevealsthemetabolicshiftsthatunderlietomato fruit development and highlight regulatory aspects of metabolic networkbehavior.PlantPhysiol.142,1380–1396

79Csukasi, F. et al. (2011) Gibberellin biosynthesis and signalling duringdevelopmentofthestrawberryreceptacle.NewPhytol.191, 376–390

80Rohrmann,J.etal.(2011)Combinedtranscriptionfactorprofiling, microarray analysis and metabolite profiling reveals the transcriptionalcontrolofmetabolicshiftsoccurringduringtomato fruitdevelopment.PlantJ.68,999–1013

81Karlova,R.etal.(2011)Transcriptomeandmetaboliteprofilingshow thatAPETALA2aisamajorregulatoroftomatofruitripening.Plant Cell23,923–941

82Fait,A.et al.(2008)Reconfigurationoftheacheneandreceptacle metabolic networks during strawberry fruit development. Plant Physiol.148,730–750

83Lombardo, V.A. et al. (2011) Metabolic profiling during peach fruitdevelopmentandripeningrevealsthemetabolicnetworksthat underpineachdevelopmentalstage.PlantPhysiol.157,1696–1710 84Osorio,S.etal.(2011)Systemsbiologyoftomatofruitdevelopment:

combined transcript, protein, and metabolite analysis of tomato transcription factor(nor,rin) andethylenereceptor(Nr)mutants revealsnovelregulatoryinteractions.PlantPhysiol.157,405–425 85Zamboni, A. et al. (2010) Identification of putative stage-specific

grapevine berry biomarkers and omics data integration into networks.PlantPhysiol.154,1439–1459

86Ren,Z.X.et al.(2011)The auxinreceptorhomologueinSolanum lycopersicumstimulatestomatofruitsetandleafmorphogenesis.J. Exp.Bot.62,2815–2826

87de Jong, M. et al. (2011) The Solanum lycopersicum AUXIN RESPONSE FACTOR 7 (SlARF7) mediates cross-talk between auxin and gibberellin signalling during tomato fruit set and development.J.Exp.Bot.62,617–626

88Egea-Cortines,M.etal.(1993)Polyaminelevels inpollinatedand auxin-induced fruit of tomato (Lycopersicon esculentum) during development.Physiol.Plant.87,14–20

89Bologa,K.L.etal.(2003)Abypassofsucrosesynthaseleadstolow internal oxygen and impaired metabolic performance in growing potatotubers.PlantPhysiol.132,2058–2072

90Hackel,A.etal.(2006)SucrosetransporterLeSUT1andLeSUT2 inhibitionaffectstomatofruitdevelopmentindifferentways.PlantJ. 45,180–192

91Chen,L.Q.etal.(2012)SucroseeffluxmediatedbySWEETproteins asakeystepforphloemtransport.Science335,207–211

92Fridman,E.etal.(2004)Zoominginonaquantitativetraitfortomato yieldusinginterspecificintrogressions.Science305,1786–1789 93Schauer, N. et al.(2006) Comprehensive metabolic profiling and

phenotyping of interspecific introgression lines for tomato improvement.Nat.Biotechnol.24,447–454

94Phuc, T.D.et al.(2010) Theinfluenceoffruitloadonthetomato pericarpmetabolomeinaSolanumchmielewskiiintrogressionline population.PlantPhysiol.154,1128–1142

95Gillaspy,G.etal.(1993)Fruits-adevelopmentalperspective.Plant Cell5,1439–1451

96Piechulla,B.et al.(1987)Changesinphotosyntheticcapacityand photosyntheticproteinpatternduringtomatofruitripening.Plant Physiol.84,911–917

97Obiadalla-Ali,H.etal.(2004)Inhibitionofchloroplastic fructose 1,6-bisphosphataseintomatofruitsleadstodecreasedfruitsize, butonlysmallchangesincarbohydratemetabolism.Planta219, 533–540

98Kolotilin,I.etal.(2007)Transcriptionalprofilingofhigh pigment-2(dg) tomato mutant linksearly fruit plastid biogenesiswith its overproductionofphytonutrients.PlantPhysiol.145,389–401 99 Lytovchenko, A. et al. (2011) Tomato fruit photosynthesis is

seeminglyunimportantin primarymetabolismandripeningbut playsaconsiderableroleinseeddevelopment.PlantPhysiol.157, 1650–1663

100 Matas,A.J.etal.(2011)Tissue-andcell-typespecifictranscriptome profilingofexpandingtomatofruitprovidesinsightsintometabolic andregulatoryspecializationandcuticleformation.PlantCell23, 3893–3910

101 Yin,L-L.andXue,H-W.(2012)TheMADS29transcription factor regulatesthedegradationofthenucellusandthenucellarprojection duringriceseeddevelopment.PlantCell24,1049–1065

102 Klee,H.J.andGiovannoni,J.J.(2011)Geneticsandcontroloftomato fruitripeningandqualityattributes.Ann.Rev.Gen.45,41–59 103 Orzaez,D.etal.(2009)AVisualreportersystemforvirus-induced

genesilencingintomatofruitbasedonanthocyaninaccumulation. PlantPhysiol.150,1122–1134

104 Quadrana,L. etal.(2011)Couplingvirus-inducedgenesilencing to exogenous green fluorescence protein expression provides a highlyefficientsystemforfunctionalgenomicsinArabidopsisand acrossallstagesoftomatofruitdevelopment.PlantPhysiol.156, 1278–1291

105 Schijlen, E. et al. (2006) Pathway engineering for healthy phytochemicals leading to the production of novel flavonoids in tomatofruit.PlantBiotechnol.J.4,433–444

106 Tohge, T. and Fernie, A.R. (2010) Combining genetic diversity, informaticsandmetabolomicstofacilitateannotationofplantgene function.Nat.Protoc.5,1210–1227

107 Iijima,Y.etal.(2008)Metaboliteannotationsbasedontheintegration ofmassspectralinformation.PlantJ.54,949–962

108 Giavalisco, P. et al. (2009) (13)C Isotope-labeled metabolomes allowing for improved compound annotation and relative quantificationinLiquidChromatography-MassSpectrometry-based metabolomicresearch.Anal.Chem.81,6546–6551

109 Kojima, M. et al. (2009) Highly sensitive and high-throughput analysis of plant hormones using MS-probe modification and liquidchromatographytandemmassspectrometry:anapplication for hormone profiling in Oryza sativa. Plant Cell Physiol. 50, 1201–1214

110 Jaillon,O. etal.(2007)The grapevine genomesequencesuggests ancestralhexaploidizationinmajorangiospermphyla.Nature449, U463–U465

111 Shulaev, V. et al. (2011) The genome of woodland strawberry (Fragariavesca).Nat.Genet.43,109–151

112 Sato,S.etal.(2012)Thetomatogenomesequenceprovidesinsights intofleshyfruitevolution.Nature485,635–641

113 Schauer,N.etal.(2005)Metabolicprofilingofleavesandfruitofwild speciestomato:asurveyoftheSolanumlycopersicumcomplex.J. Exp.Bot.56,297–307

114Fernie,A.R.andSchauer,N.(2009)Metabolomics-assistedbreeding: aviableoptionforcropimprovement?TrendsGenet.25,39–48 115Lebon,G.etal.(2008)Sugarsandfloweringinthegrapevine(Vitis

viniferaL.).J.Exp.Bot.59,2565–2578

116Ghiglione, H.O. et al.(2008) Autophagyregulated byday length determines the number of fertile florets in wheat. Plant J. 55, 1010–1024

117Dreccer,M.F.etal.(2009)Grainnumberandgrainweightinwheat linescontrastingforstemwatersolublecarbohydrateconcentration. FieldCropsRes.112,43–54

118Marcelis,L.F.M.etal.(2004)Flowerandfruitabortioninsweetpepper inrelationtosourceandsinkstrength.J.Exp.Bot.55,2261–2268

119 Lemaire-Chamley,M.etal.(2005)Changesintranscriptionalprofiles areassociatedwithearlyfruittissuespecializationintomato.Plant Physiol.139,750–769

120 Vrebalov,J.et al.(2009)Fleshyfruitexpansion andripeningare regulatedbythetomatoSHATTERPROOFgeneTAGL1.PlantCell 21,3041–3062

121 Ampomah-Dwamena,C.etal.(2002)Down-regulationofTM29,a tomato SEPALLATA homolog, causes parthenocarpic fruit developmentandfloralreversion.PlantPhysiol.130,605–617 122Pnueli, L.et al.(1994)TheTM5 MADSboxgene mediatesorgan

differentiationinthethreeinnerwhorlsoftomatoflowers.PlantCell 6,175–186