'Living stones' reveal alternative petal identity programs within the core eudicots.

MISSING IMAGE

Material Information

Title:
'Living stones' reveal alternative petal identity programs within the core eudicots.
Series Title:
Plant J. 69: 193-203.
Physical Description:
Journal Article
Creator:
Oppenheimer, David
Publication Date:

Notes

Acquisition:
Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by David Oppenheimer.
Publication Status:
Published

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the submitter.
System ID:
IR00004157:00001


This item is only available as the following downloads:


Full Text

PAGE 1

FEATUREDARTICLE ‘Livingstones’revealalternativepetalidentityprograms withinthecoreeudicots SamuelF.Brockington 1,2,* ,PaulaJ.Rudall 3 ,MichaelW.Frohlich 3 ,DavidG.Oppenheimer 1 ,PamelaS.Soltis 2 andDouglasE.Soltis 1 1 DepartmentofBiology,UniversityofFlorida,Gainesville,FL32611,USA, 2 FloridaMuseumofNaturalHistory,UniversityofFlorida,Gainesville,FL32611,USA,and 3 JodrellLaboratory,RoyalBotanicGardens,Kew,Richmond,Surrey,TW93AB,UK Received8August2011;revised13September2011;accepted19September2011;publishedonline23November2011. Forcorrespondence(fax+44(0)1223333953;e-mailsb771@cam.ac.uk). SUMMARY Petals,denedastheshowylaminaroralorgansinthesecondoralwhorl,havebeenshowntobeunder similargeneticcontrolindistantlyrelatedcoreeudicotmodelorganisms.Onthebasisofthesendings,itis commonlyassumedthatthepetalidentityprogramregulatedbyB-classMADS-boxgenehomologsisinvariant acrossthecoreeudicotclade.However,thecoreeudicots,whichcomprise>70%ofangiospermspecies,exhibit numerousinstancesofpetalandsepalloss,transferenceofpetalfunctionbetweenoralwhorls,andrecurrent petalevolution.Inthefaceofthesecomplexpatternsofperianthevolution,theconceptofacoreeudicotpetal identityprogramhasnotbeentested.WethereforeexaminedthepetalidentityprogramintheCaryophyllales, acoreeudicotcladeinwhichperianthdifferentiationintosepalsandpetalshasevolvedmultipletimes. Specically,weanalyzedtheexpressionpatternsofB-andC-classMADS-boxhomologsforevidenceofa conservedpetalidentityprogrambetweensepal-derivedandstamen-derivedpetaloidorgansinthe‘living stone’familyAizoaceae.Wefoundthatneithersepal-derivednorstamen-derivedpetaloidorgansexhibitgene expressionpatternsconsistentwiththecoreeudicotpetalidentityprogram.B-classgenehomologsarenot expressedduringthedevelopmentofsepal-derivedpetalsandarenotimplicatedinpetalidentityinstamenderivedpetals,astheirtransientexpressioncoincideswithearlyexpressionoftheC-classhomolog.We thereforeprovideevidenceforpetaldevelopmentthatisindependentofB-classgenesandsuggestthat differentgeneticcontrolofpetalidentityhasevolvedwithinthislineageofcoreeudicots.Thesendingscallfor amorecomprehensiveunderstandingofperianthvariationanditsgeneticcauseswithinthecoreeudicots–an endeavorthatwillhavebroaderimplicationsfortheinterpretationofperianthevolutionacrossangiosperms. Keywords:MADS-box,evolutionofdevelopment,oraldevelopment,Caryophyllales,Aizoaceae,petal. INTRODUCTION Petalsmaybedenedasshowylaminaroralorgansthat arelocatedinthesecondwhorlofadifferentiatedperianth (Irish,2009),wheretheyfunctionintheattractionofpollinators.Similaritiesinthegeneticregulationofpetalsinthe coreeudicotclade( Gunneridae ;Cantino etal. ,2007),have beenrevealedbycomparisonbetweentwomodelcoreeudicottaxa,theasterid Antirrhinummajus andtherosid Arabidopsisthaliana (CoenandMeyerowitz,1991):(i)Inthe absenceoftheC-classMADS-boxgene AGAMOUS ( AG )or itsorthologs,activityofB-classMADS-boxgenesisnecessaryfordevelopmentofthepetalinthesecondwhorlofthe ower(Jack etal. ,1992;MizukamiandMa,1992;Bradley etal. ,1993;GotoandMeyerowitz,1994),(ii)persistent activityofB-classMADS-boxgenesthroughlatestagesof petaldevelopmentisnecessarytomaintaintheexpression ofcharacteristicpetalfeatures(Bowman etal. ,1991;Sommer etal. ,1991;Zachgo etal. ,1995),and(iii)heterotopic expressionofB-classMADS-boxgenesintherst-whorl sepalsoftheowerissufcienttoinduceectopicpetal morphology(CoenandMeyerowitz,1991;Krizekand Meyerowitz,1996).Basedontheseobservations,ithasbeen proposedthatapetalidentityprogramregulatedbyB-class 2011TheAuthors 193 ThePlantJournal 2011BlackwellPublishingLtd ThePlantJournal (2012) 69 ,193203doi:10.1111/j.1365-313X.2011.04797.x

PAGE 2

MADS-boxgenehomologshadarisenatleastbythetimeof originofthecoreeudicotclade(KramerandJaramillo,2005). Petalshavebeenlostandgainedmanytimes,andhave beenderivedfromeitherbracts(bracteopetals)orstamens (andropetals)(KramerandIrish,2000),andconsequently occurinahomoplasticpatternacrossangiosperms.Furthermore,althoughpetalsarecommonlyfoundinthe secondwhorloftheower,oralorgansoutsideofthe secondwhorlcanhaveapetal-like(orpetaloid)appearance i.e.petaltraitscanhaveheterotopicexpressioninadditional oralwhorls.Thepotentialofhomeoticpetalidentity programstoexplainthisheterotopicandhomoplasticpetal variationiswidelycited(Bowman,1997;BaumandWhitlock,1999;KramerandJaramillo,2005).Numerousstudies havelookedforageneticsignatureconsistentwiththis potentialaconservedpetalidentityprograminallpetals andpetaloidorgans,regardlessofphylogeneticlineage, derivation,orpositionwithintheower.Suchanalyseshave focusedlargelyonRanunculales(KramerandIrish,1999, 2000),themonocots(Park etal. ,2003,2004;Whipple etal. 2007), Magnoliidae ,andbasalangiospermlineages(Stellari etal. ,2004;Kim etal. ,2005).Incontrast,relativetospecies diversity,fewrepresentativesofcoreeudicotshavebeen studiedforvariationinpetalidentity(Irish,2003);onlyone studyincoreeudicotshascomparedorganidentityprogramsbetweentaxawithsignicantperianthvariation (Geuten etal. ,2006).Thisemphasisonlineagesoutside thecoreeudicotcladeisinpartaconsequenceofeffortsto determinethephylogeneticdepthofgeneticprograms,and inpartbecausepatternsoforalvariabilityintheselineages affordopportunitiestoexaminepetalvariation. Severalerroneousassumptionsfurtherencouragethis phylogeneticbias.Coreeudicotpetalsarecommonlyconsideredtobeandropetalsandaretreatedashomologous (Hiepko,1965;Takhtajan,1991;Endress,1994);conversely, bracteopetalsarethoughttobemorecommoninlineages outsidethecoreeudicots,suchasthe Magnoliidae (Walker andWalker,1984).However,theassumeduniversalityof andropetalsincoreeudicotsiscontradictedbyaestivation patternsandtheoccurrenceofsepaloustraitsinpetalsof manycoreeudicots(RonseDeCraene,2007,2008).Itis unclearwhethertherehasbeenasingleevolutionaryorigin ofpetalsatthebaseofthecoreeudicotsand/ormultiple originsofpetals(RonseDeCraene,2007,2008).Hence, historicalhomologyofthecoreeudicotpetalisuncertain. Furthermore,therateofchangeinperianthevolutionincore eudicotshasbeenignoredbycontemporaryevo-devo approaches(RonseDeCraene,2007,2008):therearewidespreadexamplesofpetaloidyinthesepals,lossofpetalsor sepals,andsubsequentiterativeregainofpetals(Ronse DeCraene,2007,2008;Brockington etal. ,2009).These complexpatternsofperianthvariationincoreeudicotshave beeninadequatelyexplored.Astablecoreeudicotpetal identityprogramhascommonlybeenassumed(e.g.Kim etal. ,2005),butrarelytested(seeGeuten etal. ,2006). Moreover,itislogicaltolookforevidenceofheterotopyin petalevolutionwithincoreeudicots,afterthelikelyxation ofthehomeoticpetalidentityprogram,i.e.,thedivergence ofthe Rosidae and Asteridae ;butasyetthereisnogenetic evidenceforhomeosisintheevolutionofthecoreeudicot perianth.Finally,thecoreeudicotsoffersomeoftheclearest instancesofderivedandropetals(RonseDeCraene,2008), butnogeneticanalysesoftheserecurrentandropetalorigins havebeenconducted. Giventheseuncertainties,thecoreeudicotcladeCaryophyllalesisanappropriategrouptoexaminethegenetics underlyingiterativepetalevolutionincoreeudicots.Previousanalysesshowthatonewhorlofperianthwaslostearly intheevolutionofCaryophyllales(Figure1;Brockington etal. ,2009).Theclassicalperspectiveisthatpetalswere ancestrallyabsentinthisclade(Figure1),andthatthe organsthatcomprisethesingle-whorledperianthofmany lineagesofCaryophyllalesarehomologoustocoreeudicot sepals(Hofmann,1994).Differentiationoftheperianthinto distinctpetalandsepalwhorlshasoccurredatleastnine timesinCaryophyllales,througheithertherecruitmentof subtendingbracts(e.g. Portulaca )orthegainofstaminodial petals(e.g.Aizoaceae)(Brockington etal. ,2009;RonseDe Craene,2007;Figure1).Thus,recurrentexamplesofa differentiatedperianthinCaryophyllalesfallintotwocategorieswithpetaloidorganseithersepal-derivedorstamenderived.Distinctevolutionaryoriginsofadifferentiated perianth,withcontrastingpetalderivations,providethe necessaryvariationandevolutionaryreplicatestoassessthe roleofthecanonicaleudicotpetalidentityprogramin recurrentpetalevolution. Herewefocusontheevolutionofdifferentiatedperianthin Aizoaceae,acladeoffoursubfamilies,inwhichSesuvioideae andAizooideaearesuccessivesisterstoMesembryanthemoideaeplusRuschoideae(Klak etal. ,2003;Figure1).Two oraltypesoccuracrossthesefoursubfamilies.SesuvioideaeandAizooideaeexhibitanundifferentiatedperianth comprisingasinglewhorlofvetepals(Figure1B),which arepetaloidontheiradaxialsurfacesandsepaloidontheir abaxialsurfaces;thisistheinferredancestralconditionfor Aizoaceae.MesembryanthemoideaeandRuschoideaedisplayadifferentiatedperianthwithanouterwhorlofsepals (homologoustothetepalsofSesuvioideae/Aizooideae)and aninnerwhorlofputativeandropetals(Hofmann,1994; RonseDeCraene,2007;Figure1A).Therefore,atransferof functionhasoccurredfromtheadaxialtepalsurfaceof SesuvioideaeandAizooideaetotheandropetalsofMesembryanthemoideaeandRuschoideae. Weexaminedthesedifferentlyderivedpetaloidorgansfor evidenceofasharedcoreeudicotpetalidentityprogram.We isolated AP3 ,PIand AG homologsfromallfoursubfamilies andobserved in-situ expressionpatternsofthesegenesin twospeciesrepresentingthetwodistinctoraltypes:194 SamuelF.Brockington etal. 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 3

Sesuviumportulacastrum (Sesuvioideae:petaloidtepals) and Delospermanapiforme (Ruschoideae:andropetals). Signicantly,neitherpetalclassexhibitsgeneexpression patternsconsistentwiththeclassiccoreeudicotpetal identityprogram.Thus,weprovideevidenceforcore eudicotpetaldevelopmentthatisindependentof AP3 -and PI -lineageMADS-boxgenes,andsuggestthatdifferent geneticcontrolofpetalidentityhasevolvedwithinthe Caryophyllalesinthecontextofitsunusualoralevolutionaryhistory. RESULTS IsolationandphylogeneticanalysisofMADS-box genesfromAizoaceae Orthologsof AGAMOUS(AG) PISTILLATA(PI) ,and APETALA3 ( AP3) wereisolatedfromrepresentativesofallfour subfamilieswithinAizoaceae; Portulacaoleracea (Portulacaceae)and Antigononleptopus (Polygonaceae)werealso sampledasrepresentativesofoutgroups.Singlecopiesof the AG PI AP3 wereisolatedfromallfoursubfamilies (Figure2).Thedegeneratenatureofthepolymerasechain reaction(PCR)alsoledtotheisolationofnumerousadditionalMADSlineages,includingrepresentativesofthe TM6 lineage(aparalogouscladetothe AP3 and PI lineages)from allfoursubfamilies(Figure2).Identitiesofallisolated AG PI AP3 ,and TM6 geneswereconrmedbytheirphylogeneticpositionswithinsubcladescontainingpreviously identiedgenesfrommodeleudicots.Asingle PI locusand twocopiesof AP3 wereisolatedfrom Antigonon ,andan AG orthologwasisolatedfrom Portulaca. Thedetectionoftwo copiesof AP3 in Antigonon isconsistentwiththetwocopies foundin Rumex ,suggestingapossiblePolygonaceae-specicduplicationinthe AP3 lineage(Figure2).Accession numbersforpublishedsequencesusedinthisphylogenetic analysistogetherwithsequencessubmittedbythisstudy arelistedinSupplementaryInformationonline(TableS1). Floraldevelopmentin Sesuviumportulacastrum and Delospermanapiforme In Sesuviumportulacastrum ,thevetepalprimordiaare initiatedina2/5sequence(Figure3C).Theprimordiaare crescent-shapedwithabroadbaseofinsertion(Figure3C). Androecialprimordiaareinitiatedinsomewhatchaotic fascicles(Figure3D).Thegynoeciumdevelopswiththree carpels(Figure3D).Outerstamensgrowandsubsequently hidetheinnerstamensanddevelopingcarpels(Figure3E). Thetepalprimordiadifferentiateintoupperandlower domainsearlyindevelopment(Figure3G).In S.portulacastrum ,anadaxialcross-zonegivesrisetoaligulethatmarks theboundarybetweentheupperandlowerdomains(digitallycoloredgreenandpurple,respectively,inFigure3GI), allowingthetwodomainstobetrackedthroughtheontogenyofthetepal.Earlyindevelopment,theunifacialupper domainrepresentsthebulkofthetepal(Figure3F).However,subsequentdifferentialdevelopmentensuresthatthe bifaciallowerdomainformsthebulkofthematuretepal (Figure3H,I).Thus,thepetaloidlaminaofthetepalis derivedfromthelowerdomainwhileadistalunifacialtip representstheupperdomain(Figure3J). Rhabdodendraceae Simmondsiaceae Asteropeiaceae Physenaceae Caryophyllaceae Achatocarpaceae Amaranthaceae Stegnospermataceae Limeum Aizoaceae Phytolaccoideae Sarcobatus Rivinioideae Nyctaginaceae Basellaceae Didiereaceae Halophytum Claytonia Molluginaceae Talinum Cactaceae Portulaca Gisekia Sesuvioideae Aizooideae Mesembryanthemoideae Ruschoideae Undifferentiated perianth Differentiated perianth Undifferentiated and differentiated perianth Inferred petal loss Gain of staminodial petals/loss of petaloidy in tepals (A) (B) Figure1. Reconstructionofperianthevolution illustrating:(i)theinferredlossofpetalsinthe coreCaryophyllales,(ii)thesubsequentiterative originsofdifferentiatedperianth(modiedin partfromBrockington etal. ,2009),(iii)the inferredgainofpetaloidstaminodesandlossof petaloidyinthetepalsinAizoaceae,and(iv)the twooraltypesinAizoaceae:Apetaloid staminodesin D.napiforme andBpetaloid tepalsin S.portulacastrum.Alternativepetalidentityin‘Livingstones’ 195 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 4

Theseontogeneticdataillustratingderivationofthe petaloidportion(lamina)ofthetepalfromthelower domaincanbecomparedwithdatafromthevegetative leavesofthesamespecies,inwhichtheleafsheathis derivedfromthelowerdomain,belowthecross-zone,and thelaminaisderivedfromtheupperdomain(Figure3L). Thus,thetepallaminacorrespondstotheleafsheathand thereducedunifacialdistaltipofthetepaltotheleaf blade(Figure3J).Thisclosehomologybetweenthe laminaofthetepalandtheleafsheathisstrongly reectedinthegrossmorphology(compareimagesin Figure3J,K,M,N). In D.napiforme ,thevesepalsalsoariseina2/5 arrangement(datanotshown).However,removalofthe sepalsrevealsthatvecarpelprimordiacomprisingthe gynoeciumarevisiblepriortotheemergenceoftheve primaryandroecialprimordia,whichalternatewiththe carpelprimordia(Figure3P).Furtherprimordiaariseoutside oftheseinnerandroecialprimordiainacentrifugaldirection (Figure2Q).Onlytheinnermostprimordiadevelopinto fertilestamens(Figure3R);inamatureowerapproximately10fertilestamensformasingleringaroundthe gynoecium(Figure3T).Outerprimordiadevelopintosterile staminodesandbecomeincreasinglypetaloidinacentrifugaldirection(Figure3R),resultinginmanywhite,showy petalsinthematureowerthatcontrastwiththegreen sepalsinthematureower(Figure3O).Theinitiationof severalwhorlsofpetalprimordiainacentrifugalpattern providesanopportunitytosimultaneouslyexaminegene expressioninseveralpetalsatdifferentstagesofdevelopment(Figure3R,S). Expressionpatternsof AGAMOUS PISTILLATA APETALA3 orthologs In S.portulacastrum SpPI expressionisrestrictedtothe androeciumandisabsentfromthedevelopingcarpelsand tepals(Figure4AD).Expressionintheandroeciumpersists throughoutlaterstagesofstamendifferentiationandis presentinbothlamentanddifferentiatedanther (Figure4G,H). SpPI expressionisabsentfromdeveloping ovules.Similarly, SpAP3 expressionin S.portulacastrum is restrictedtotheandroeciumandisabsentfromthedevelopingcarpelsandtepals(Figure4EH)butisstrongin developingovules(Figure4I).Expressionintheandroecium persiststhroughoutlaterstagesofstamendifferentiation andispresentinbothlamentanddifferentiatedanther (Figure4G,H). SpAG isabsentforthetepalsexpressedearly inthedevelopmentoftheandroeciumandgynoecium (Figure4J)andpersistsuntillatestagesofstamenand carpeldifferentiation(Figure4K,L). SpAG isalsostrongly expressedindevelopingovules(Figure4L). In D.napiforme DnPI expressionisrestrictedtothe androecialprimordiathatgiverisetofertilestamensand staminodes(Figure5AF).Expressionisinitiallystrongin Arabidopsis AP3Spinacia AP3 Silene SLM3 Antigonon AP3-1 Sesuvium AP3 (SpAP3) Tetragonia AP3 Aptenia AP3 Delosperma AP3 (DnAP3) Antigonon AP3-2 Rumex D1 Rumex RAD2 Lycopersicon AP3Petunia GPNicotiana MADS-boxGerbera MADS-boxRosa MASAKO B3Dianthus CMB2Malus MdTM6Lycopersicon TDR6Silene SdAP3YPetunia TM6Sesuvium TM6Tetragonia TM6Aptenia TM6Delosperma TM6TM6 AGArabidopsis AGMalus MADS15Fragaria STAGRumex RAP1Portulaca AGSilene SLM1Spinacia AGBeta M341Sesuvium AG (SpAG) Tetragonia AG Aptenia AGDelosperma AG (DnAG)AP3Malus MdMADS14Rosa MASAKOD1Vitis VvMADS1Lycopersicon AGAntirrhinum PLENicotiana PLE36Petunia FBP6PLEAntigonon PITetragonia PISesuvium PI (SpPI)Delosperma PI (DnPI)Aptenia PIArabidopsis PIPIPetunia FBP1Nicotiana GLOPetunia PRMADS2Gerbera MADS-boxMalus PIRosa MASAKO BPGerbera MADS62 90 88 63 90 84 99 99 99 89 56 99 100 100 100 82 100 99 100 99 98 70 85 99 99 71 100 75 96 98 63 100 8998 70 85 9695 55 90 Figure2. PhylogeneticanalysisofB-andC-class genesisolatedfromAizoaceaetogetherwith previouslypublishedeudicotsequences.Treeis a50%majorityruleconsensusandphylogenetic analyseswereperformedasdescribedin Experimentalprocedures .Maximumlikelihoodbootstrapvaluesareindicatedabovethebranches.196 SamuelF.Brockington etal. 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 5

allandroecialprimordial(Figure5A,B).Infertilestamens, expressionismaintainedthroughoutdifferentiationofthe lamentandanther(Figure5C,D).Expressionremains stronginthedevelopinglamentsandanthers,particularly intheloculewall(Figure5E).Sectionsthatsimultaneously capturepetalsatvaryingstagesofdevelopmentreveal temporalvariationinexpressionof DnPI. Inprimordiathat willgiverisetoouterpetals(theandropetals), DnPI expressionisinitiallystrong(Figure5C),butexpression becomesrestrictedtothedistaltipofthepetalsduring earlydevelopment(compareoldestpetalP1withyounger outerpetalP3inFigure5C,E,F).Expressionislostorvery weakbythetimethestaminodesreachapproximately 300400 lMinlength. DnAP3 isverysimilarto DnPI expressionitisalsomainlyrestrictedtoandroecial primordiagivingrisetofertilestamensandpetals.Expressionissimilarlystronginallandroecialprimordia(Figure5G,H).Infertilestamens,expressionismaintained throughoutdifferentiationofthelamentsandanthers (Figure5H,I).Incontrast,inprimordiagivingrisetopetals, DnAP3 expressionisinitiallystrongbutweakensindevelopmentofpetals(Figure5I)andbecomesrestrictedina distaldirection(Figure5H,I),inacomparablemannerto DnPI. Expressionislostbythetimethepetalsreach approximately300400 lMinlength(Figure5I). DnAP3 is stronglyexpressedintheovules(Figure5J). DnAG is expressedinthemeristempriortotheemergenceof androecialandgynoecialprimordia(Figure5K).Expression Br Br T4 T1 T2 T3 T5 A G G1 2 2 3G Ca St Pe Ca St Pe 50 m 5 mm 50 m 50 m 100 m 250 m 500 m 25 m 750 m 2.5 mm 2.5 mm 2.5 mm 2.5 mm 5 mm 2.5 mm 400 m 200 m 50 m 60 m 40 m(A) (F) (K) (P) (Q)(R) (S)(T) (L)(M)(N)(O) (G)(H)(I)(J) (B) (C) (D) (E) Figure3. Floraldevelopmentin Sesuviumportulacastrum ,Sesuvioideae(AN)and Delospermanapiforme Ruschoideae(OT).Tepal(Te),Androecium(A), Gynoecium(G),Carpel(Ca),Stamen(St),Petal(Pe),Bracts(Br).(A)Matureowerof S.portulacastrum. (B)Bractsenclosingtheoralmeristem(C)Arrangementof tepalsinorderofinitiation(Te1-Te5)(D)Developingtricarpellategynoeciumandfasciculatedandroecium.(E)Stamensenclosingdevelopinggyn oecium.(F)Early indevelopmenttheupperleafzone(digitallycoloredgreen)formsthebulkofthetepal.(GI)Expansionoflowerleafzone(digitallycoloredpink). (H,I)Lowerleaf zoneformsthebulkofthetepalandupperleafzoneformsdistaltipofthetepal.(J)Adaxialsurfaceofamaturetepalwithupperleafzoneformingdista ltip(arrow). (K)Abaxialsurfaceofamaturetepal.(L)Vegetativeleafshowingleafsheathattachedtostem.(M)Adaxialsurfaceoftheleafsheath.(N)Abaxialsur faceoftheleaf sheath.(O)Matureowerof D.napiforme. (P)Reproductivestructureswithnumberedprimordiaindicatingsequenceofcentrifugalstameninitiation.(Q)Arrows markcentrifugaldirectionofstameninitiation.(R)Differentiationofinnerandroecialpartsintostamenswithouterwhorlssterileandlaminar. (S)Petalsenclosing developingstamensandcarpels.(T)Openowershowingsingleinnerwhorloffertilestamenssurroundedbymultiplewhorlsofandropetals.Alternativepetalidentityin‘Livingstones’ 197 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 6

isstronginthedevelopingcarpelandinallorgansofthe androecium:itisstronglyexpressedintheprimordiathat eventuallyformsterilepetalsaswellasinprimordiathat giverisetofertilestamens(Figure5L,M).Inpetals, DnAG is expressedthroughouttheorgansuntilaftertheyhave reachedapproximately100 lMinlength(Figure5N,O). Afterthispointofpetaldevelopment,expressionof DnAG appearstobecomerestrictedinadistaldirectionuntil expressionisrestrictedtothetipofthematuringsterile staminode(Figure5N,O).Laterindevelopment,thisrestrictedexpressionislost(Figure5N,O). DnAG isstrongly expressedintheovarywallandthedevelopinganthers andovules(Figure5P). DISCUSSION Homeotictransformationoforganidentity,mediatedbythe expressionofthecanonicalpetalidentityprogram,hasthe potentialtoexplainthehomoplasticoccurrenceofpetaloid organs,andvariationinthespatialexpressionofpetaloid traits(Bowman,1997;BaumandWhitlock,1999;Kramerand Jaramillo,2005).However,thechallengeofdeningperianthhomologyatmultiplehierarchicallevels(Jaramilloand Kramer,2007)(i.e.,positional,historical,morphological,and geneticcorrespondence)makessuchpotentialdifcultto evaluate.Caryophyllalesareausefulcladeinwhichto assesstheevolutionaryroleofahomeoticpetalidentity program,preciselybecausewecancondentlyarticulate homologyatseverallevels.ThisisexempliedinAizoaceae, whichexhibitanunequivocaloriginofandropetalsanda cleartransferenceoffunctionbetweenpetaloidorgansof distincthistoricalderivation. ThepetaloidtepalsofSesuvioideaeandAizooideaefulll criteriaofsepal-derivedpetals(RonseDeCraene,2008).The tepalprimordiaarecrescent-shaped,haveabroadbaseat AA G Te Te Te TeTeTeCa St St Te Te Te Ca St St Fl At A A G Te Te At At Ov Ov 100 m 100 m 100 m 100 m100 m 100 m100 m 100 m 100 m 100 m 100 m 100 mA G A Te Te TeTe Te G A A StSt Ca Te Te Te Ca St St TeTe St St Te Te Te Ca Ca St St Ov Ca St StTe Te (A) (E) (I) (J) (K) (L) (F) (G) (H) (B) (C) (D) Figure4. ExpressionpatternsofMADS-boxhomologsin Sesuviumportulacastrum (AL).Tepal(Te),Androecium(A),Gynoecium(G),Stamen(St),Filament(Fl), Anther(At),Petal(P),Ovary(Ov).(A)Strong SpPI expressionmarksthesiteoforiginofstamens;noexpressiondevelopingtepals.(B,C) SpPI expression indevelopingandroecium.(D) SpPI expressionindifferentiatingstamens.(E)Strong SpAP3 expressionmarksthesiteoforiginofstamens;noexpressionin developingtepals.(F) SpAP3 expressionindevelopingandroecium.(G,H) SpAP3 expressionindifferentiatingstamens;absentinthecarpel.(I) SpAP3 expressionin ovulesandanthers.(J) SpAG expressioningynoeciumandandroecium.(K) SpAG expressionincarpelsandstamens.(L)Strongexpressionof SpAG inthe developingovules.198 SamuelF.Brockington etal. 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 7

insertion,aresuppliedbythreevasculartraces(compared withonetraceinstamens),andhaveaspiralsequenceof initiation(Figure2C).Theyalsodifferentiateintoupperand lowerdomains(Figure2FI),whichrepresentsacommon morphogeneticfeatureintheleavesofmanyangiosperm species(Kaplan,1973).Thegreen,chlorophylloussepalsof AA G Se Se100 m SeSe St P1 OvP3 At P2 At At Ov P1 P2 P3 StCa Se Ca Ov100 mP1 P2 P3 St AA G A A GSe Se SeA A GSeSeSe St Se Se OvAt100 mP1 P2 P3 At Se P1 P3 P2 CaOvSe SeSt P 100 m St P1 P2 Ca P3 Se Se Se100 m A A Ca P St A Ca A 100 mSe Se 100 m 100 mSeP1 P3 St P2 St 100 m 100 m 100 m 100 m 100 m100 m100 m 100 m(A) (E) (I) (M)(N)(O)(P) (J)(K)(L) (F)(G)(H) (B)(C)(D) Figure5. ExpressionpatternsofMADS-boxhomologsin Delospermanapiforme (AP).Sepal(Se),Androecium(A),Gynoecium(G),Stamen(St),Filament(Fl), Anther(At),Petal(P),Ovary(Ov).(A) DnPI expressioninemerginganddevelopingandroecialprimordial,absentfromgynoecium.(C)enlargedfrom(B)Expression maintainedinanthersandlament;expressionininnermostpetals(P1andP2)butweakerthaninoutermostpetal(P3).(E)enlargedfrom(D)Expressio nof DnPI weakinP1;restrictedtothedistaltipofP2(blackarrow)andstronginP3; DnPI expressionabsentindevelopingovules.(F)Expressionof DnPI lostinP1andP2and restrictedtothetipoftheoutermostpetalP3(blackarrow).(G) DnAP3 expressioninemerginganddevelopingandroecialprimordia.(H) DnAP3 expressionin developingplacenta,expressionmaintainedinanthersandlament;expressioninpetal1weakerthaninpetals2and3.(I)Expressionstronginlocul ewalls,absent inpetal1,expression(blackarrows)restrictedtodistaltipofpetal2,strongerinpetal3.(J)Strongexpressionof DnAP3 intheovules.(K) DnAG expressioninthe gynoeciumandandroeciumprimordia.(L,M) DnAG expressioningynoeciumandallfertileandsterilemembersoftheandroecium.(N) DnAG expressionstrongin anthersandlamentsofinnerstamens,andovules;absentfromdevelopingpetal1;expressionrestrictedtothetipofpetal2(blackarrow),strongex pressionin emergingpetal3(blackarrow).(O)Expressionof DnAG inanthers;absentnowinpetals1and2;expressionrestrictedtodistaltipofpetal3(blackarrow).(P)Strong expressionof DnAG indevelopingovules.Alternativepetalidentityin‘Livingstones’ 199 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 8

MesembryanthemoideaeandRuschoideaeareclearly homologouswiththepetaloidtepalsandexhibitsimilar patternsofinitiation,insertion,vasculature,anddifferentiationintoupperandlowerdomains(Payer,1857).Thepetals ofMesembryanthemoideaeandRuschoideaefulllthe criteriaofstamen-derivedpetals.Theypossessasingular vasculartrace,anarrowpointofinsertion,andbothpetals andstamensdevelopfromprimordiathatareinitiated centrifugally(Hofmann,1994).MembersofAizooideae (e.g., Gunniopsis )exhibitcentrifugalandroecialinitiation butallprimordiadevelopintofertilestamens(Hofmann, 1994);positionallyhomologousprimordiacantherefore developintoeitherpetalsorstamens.Finally,organsare presentthatareintermediatebetweenstamensandpetals, asoralorgansdevelopingclosesttothefertilestamensare increasinglylamentouswhileoutermostorgansare increasinglylaminar. Ourmorphologicaldatasuggestaninterestingcorrespondencebetweenthepetaloidlaminaandleafsheath.The tepalsdifferentiateintoanupperandlowerzoneearlyin development:differentiationisawellestablishedrststage inthemorphogenesisoftheleavesofmanyspeciesof angiosperms(Kaplan,1973).Trackingtheexpansionof thesetwozonesduringdevelopmentrevealsthatthe petaloidlaminaofthetepalisderivedfromthelowerzone andthuscorrespondstotheleafsheathinthevegetative leavesof S.portulacastrum. Thiscorrespondencecanbe clearlyseenwhencomparingthematureleafsheathandthe petaloidtepal(compareFigure3JN).Interestingly,inthis contextthereisalsocorrespondencebetweenthedelicate petal-liketissueandhyalinemarginsoftheleafsheath (compareFigure3K,N).Thisobservationsuggeststhat petaloidtissuecanhaveunexpecteddevelopmentaland evolutionaryorigins.Wespeculatethathyalinesepalmarginsmaybepre-adaptedforco-optionaspigmentedpetal tissueastheyarebothdelicateand,perhapsmoresignicantly,achlorophyllous.Theroleofhyalinemarginsinthe perianthislargelyunstudied(butseeHofmann,1994and referencestherein);howevernumeroustaxawithinthe Caryophyllalesalsoexhibitasepal-likeperianthwithdelicatepetaloidmargins(e.g. Hypertelis ).Thecorrespondence betweenhyalinemarginsandpetaltissueimpliesthatpetal tissuesmaybeformedbydifferentdevelopmentalmechanismsdependingonthehistoricalderivationofthepetaloid organ. However,giventheconceptofaconservedpetalidentity programwithincoreeudicots,andahypothesisofheterotopyinpetalevolution,weexpectedsimilaritiesinMADSboxgeneexpressionbetweenpetaloidtepalsandpetaloid stamensinmembersofAizoaceae.Contrarytothisexpectation, SpAP3 and SpPI arenotexpressedatanypointinthe developmentofthepetaloidtepal(in S.portulacastrum ) withexpressionofthesegenesrestrictedtothestamens while DnAP3 and DnPI aretransientlyexpressedinthe petaloidstaminodes(in D.napiforme ).Similarly, SpAG is notexpressedinpetaloidtepalswhile DnAG istransiently expressedinpetaloidstaminodes.Therefore,atthelevelof AP3 PIandAG homologgeneexpression,wendno evidenceforhomologybetweenthepetaloidtepalsand petaloidstaminodesof S.portulacastrum and D.napiforme (seeFigure6)andbyinferencetoothermembersof Aizoaceaewithsimilaroraltypes. Thesedatademonstrateanabsenceof AP3 and PI homologsatthetranscriptionallevelinapetaloidorgan withinacoreeudicotlineage.Petalidentityandexpression of AP3 / PI homologshavebeensimilarlydecoupledin angiospermtaxaoutsideofthecoreeudicots.Forexample, inthemagnoliid Aristolochia AP3 and PI homologsare expressedonlylateinthedifferentiationoftheperianth (JaramilloandKramer,2004); AP3 and PI homologsarenot expressedintheoutermostpetaloidperianthwhorlin owersofthemonocot Asparagus (Park etal. ,2003,2004); and AqvPI isrequiredforpetalidentityinthesecondwhorl butnotthepetaloidrstwhorlinthebasaleudicot Aquilegia (Kramer etal. ,2007).Theabsenceof SpAP3 and SpPI activity inthepetaloidtepalsof Sesuviumportulacastrum extends theseobservationstothecoreeudicots,andcontradictsearlyhypotheses(KramerandIrish,1999,2000)thatall petaloidorgansincoreeudicotsmightshareacommon AP3and PIdependentpetalidentitypathway. Se Pe St Ca C B A SePeStCa Se Pe St Ca A B* B C*C SePeStCa Te St Ca A B C TeStCaCore eudicot (e.g. Arabidopsis )Aizoaceae (e.g. Sesuvium )Aizoaceae (e.g. Delosperma ) Figure6. SummaryofgeneexpressionpatternsfoundinacoreeudicotowersuchasArabidopsisversusthetwooraltypesofAizoaceae;petaloidtepaloral type(e.g. Sesuvium )andandropetaloraltype(e.g. Delosperma ).Se,Sepal;Pe,Petal;Te,Tepal;St,Stamen;Ca,Carpel.A:A-class,B:B-class,C:C-class,Asterix*: transientearlygeneexpression.200 SamuelF.Brockington etal. 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 9

Thedistalrestrictionandsubsequentlossof DnAP3 and DnPI expressionwithinthepetaloidstaminodesissimilar tothatreportedforsomepetaloidorgansinRanunculales (KramerandIrish,1999).Constantexpressionuntillate stagesofpetaldevelopmentisnecessaryforthemaintenanceofpetalidentityin A.thaliana and A.majus (Bowman etal. ,1991;Sommer etal. ,1991;Zachgo etal. ,1995). GiveninconstantexpressionofB-classMADS-boxhomologs,itwassuggestedthatthepetalidentityprogram differswithinRanunculales,eitherbecauseofalternative petalevolutioninsometaxa,orbecauseRanunculales arosepriortothexationofapetalidentityprogram. Followingthislogic,thedetectionofasimilarinconstant patternofgeneexpressionwithinthecoreeudicots arguesagainstanancestralpetalidentityprogramthat hassimplybeenturnedbackoninthepetaloidstaminodesofAizoaceae. Inmodelorganisms(e.g.Arabidopsis)theabilityof AP3 and PI homologstospecifypetalidentityiscontingenton theabsenceoftheC-classMADS-boxgenehomologs(Jack etal. ,1992;MizukamiandMa,1992;Bradley etal. ,1993; GotoandMeyerowitz,1994).Here, DnAG isexpressed concurrentlywiththeB-classMADS-boxhomologsand exhibitsidenticalearlyexpression,distalrestriction,and lossinthepetaloidstaminodes.Thisco-expressionof DnAP3 and DnPI with DnAG suggeststhat DnAP3 and DnPI expressionisnotimplicatedinpetalidentity.Rather,coincidentearlylossofB-andC-classMADS-boxgeneexpressioninthedevelopmentofpetaloidstaminodesmorelikely signiesdeactivationofthestamendevelopmentalprogram.Thesedataemphasizetheimportanceofobserving AG homologexpressionwheninterpretingvariablepatterns ofB-classgeneexpression. Thesetwouniquepatternsofgeneexpressioninpetaloid organsofcoreeudicottaxa(seeFigure6)canbesimply explainedinanevolutionarycontext.Earlyintheevolution oftheCaryophyllales,onewhorlofperianthwaslostsuch thattheuniseriateperianthoftheSesuvioideae/Aizooideae typepredominates.Assuminglossofthepetalwhorl,the tepalwhorlinasimpleperianthofCaryophyllales(as occursin Sesuviumportulacastrum )ishistoricallyhomologouswiththesepalwhorlofothercoreeudicots(Hofmann,1994).Wesuggestthattheabsenceof SpAP3 and SpPI inthepetaloidtepalsof Sesuviumportulacastrum isa resultoftheirderivationfromcoreeudicotsepals,which alsodonotrequireB-classMADS-boxgeneactivity. Thesubsequentreturntothepetaloidconditionwas consequentlyachievedindependentlyofthe AP3 -and PI -dependentpathway.Numerouslineagesacrossthe Caryophyllalespossesspetaloidorgansthatarehistorically homologouswiththetepalsofSesuvioideaeandAizooideae.Consequently,thesepal-derivedinterpretationof thesegeneexpressionpatternspredictsfurthercasesof petaldevelopmentindependentof AP3 and PI homologs; forexample,inthetepalsofMolluginaceae,Nyctaginaceae, Hypertelis ,andtheinnerperianthwhorlsofthePortulacineae. AcrosstheCaryophyllales,multipleoriginsofadifferentiatedperianthentailtherecruitmentofstaminodesto functionaspetals(Brockington etal. ,2009).Althoughan ancestralpetalidentityprogrammightbeiteratively recruitedinthispatternofperianthevolution,ourdatado notsupportthismodelfortherecurrentevolutionofpetals intheCaryophyllales.Theabsenceof AP3 and PI homolog expressioninpetaloidtepalsandsepalsinAizoaceae impliesthatan AP3 -and PI -dependentpetalidentitypathwaywasnotretainedbydeploymenttotheuniseriate perianth.Withoutsuchspatialredeployment,itisunclear howan AP3 -and PI -dependentpetalidentitypathwaycould havebeenmaintainedfollowingpetalloss.Itismorelikely thatlossofpetalsresultedinlossofaformerpetalidentity pathwaydependenton AP3 andPI homologs;thislossis thenreectedinthegeneexpressionpatternsofandropetals inAizoaceae,whicharenoveloralorgansexhibitinga uniquedevelopmentalgeneticprogramcomparedwiththe petalsofothercoreeudicots. Considerabledatahavenowbeencollatedontheexpressionpatternsof AP3 and PI homologsacrossangiosperms, butacollectiveinterpretationofthesedataisnoteasy.Some authorshaveproposedacorrelationbetweentheexpression of AP3 and PI homologsandpetaloidythroughsliding' (Bowman,1997;Kramer etal. ,2003)orfadingborders' (Soltis etal. ,2007)whileothershaveexploredthelink betweenvariationingeneexpressionpatternsandcomplexity'inpetalmorphology(KramerandJaramillo,2005). Morerecentlyithasbeensuggestedthat AP3 and PI homologsareconservedprimarilytospecifyregional domainsintheowerratherthanorganidentity perse (Drea etal. ,2007;Whipple etal. ,2007).Finally,inouranalysiswe proposethatobservedgeneexpressionpatternsinAizoaceaearebestexplainedthroughthedistincthistorical derivationofthepetaloidorgansinthecontextofan unusualevolutionaryhistory.Thesemodelsarenotmutuallyexclusiveaseachmayholdfordifferentoralstructures, taxonomicgroups,phylogeneticlevels,andepisodesof evolutionaryhistory.Nevertheless,intheabsenceof evidenceforheterotopyatthelevelofMADS-boxgenes, onefuturechallengeliesindeterminingwhichgenesare responsibleforthespecicationofnovelpetalsinthe Caryophyllales. EXPERIMENTALPROCEDURES PlantmaterialApteniacordifolia Delospermanapiforme ,and Tetragoniatetragonoides weregrowninagreenhouseattheUniversityofFlorida, USA.Floweringmaterialof Sesuviumportulacastrum wascollected fromCedarKey,Florida,USA.Voucheredspecimensaredeposited attheherbariumattheFloridaMuseumofNaturalHistory(FLAS).Alternativepetalidentityin‘Livingstones’ 201 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 10

IsolationofMADS-boxgenesfromAizoaceaeTotalRNAwasextractedfromoralbudsofvariousstagesof developmentusingtheRNeasyextractionkit(Qiagen,http:// www.qiagen.com/).cDNAwassynthesizedusingSuperscriptII (Invitrogen,http://www.invitrogen.com/)accordingtothemanufacturer'sinstructions.Amplicationoftargetgeneswascarriedout usingdegenerateprimersanddegeneratePCRthermocyclingconditionsfrompreviouslypublishedstudies,(B-classgenes;Kramer etal. ,1998)and,C-class;Stellari,JaramilloandKramer,2004).PCR bandsover500bpinsizewereexcisedfromtheagarosegeland puriedusingtheGenecleanIIKit(QBioGene,http://www.qbiogene.com/).PuriedDNAswereclonedusingtheTOPOTACloning Kit(Invitrogen).Intotal,50100cloneswerescreenedfromeach PCRband.PlasmidDNAswereampliedusingtheTempliphi Cycle(Amersham,http://www5.amershambiosciences.com/),and sequencesweregeneratedonanABI3730XLDNAsequencer (AppliedBiosystemsInc.,http://www.appliedbiosystems.com/).PhylogeneticanalysesForMLanalysesweemployedtheprogramGARLI(GeneticAlgorithmforRapidLikelihoodInference;version0.942,https:// www.nescent.org/wg_garli/Main_Page)(Zwickl,2000).Modelsof nucleotidesubstitutionweredeterminedusingMrModeltest(Nylander,2004).TheAkaikeinformationcriterion(AIC)wasusedto selectGTR+I+Gasanappropriatemodelbasedontherelative informationaldistancebetweentherankedmodels.Fivereplicates wereperformedandthetreewiththehighestlikelihoodscoreis representedhere.Analyseswererunwithdefaultoptions,and bootstrapanalyseswereconductedfor1000replicates.ScanningelectronmicroscopyDevelopinginorescencesweredissectedandxedinfreshly preparedFAA(3.7%formaldehyde,50%ethanol,5%aceticacid). SamplesweredehydratedthroughabsoluteethanolandcriticalpointdriedusinganAutosamdri-815BCPD(TousimisResearch, http://tousimis.com/),thencoatedwithplatinumusinganEmitech K550sputtercoater(http://www.quorumtech.com/)andexamined usingaHitachicold-eldemissionSEMS4700II(http://www.hht-eu. com/cms/142.html).SomeimageswerecoloredusingAdobe Photoshop(http://www.adobe.com).RNA–RNA in-situ hybridizationDevelopinginorescencesweredissectedandxedinfreshly preparedFAA(3.7%formaldehyde,50%ethanol,5%aceticacid) overnightat4 C.Anti-senseandsenseprobesweredesignedtothe C-terminusand3 ¢ UTRregionofthehomologsof AP3 ( SpAP3 and DnAP3 ), PI ( SpPI and DnPI ),and AG ( SpAG and DnAG )andsynthesizedusingaDIGRNAlabelingkitwithaT3polymerase(Roche, http://www.roche-applied-science.com/).Probeswerehydrolyzed toapproximately150bp.Probehybridization,post-hybridization treatment,andimmuno-localizationwereperformedaspreviously described(MalcomberandKellogg,2004).SlideswerephotographedusingaZeissAxiocamMRc5digitalcameramountedona ZeissAxioskop2Plusmicroscope(CarlZeissMicroImaging,http:// www.zeiss.de/)withbrighteldillumination.ACKNOWLEDGEMENTSWethankChryssiePyrchidandParisGrayfortechnicalassistance, andmembersoftheSoltislabforadviceanddiscussion.Weare gratefultomembersoftheGloverlaboratoryfordiscussionand editstoearlierversionsofthismanuscript.ThisstudywassupportedbyaNationalScienceFoundationGrantPRG-0115684for theFloralGenomeProject,aDDIGgrant(DEB-0808342)toDESand SFB,andanEUMarieCurieIRGFellowshiptoSFB.SUPPORTINGINFORMATIONAdditionalSupportingInformationmaybefoundintheonline versionofthisarticle: TableS1. Accessionnumbersforsubmittedandpreviouslypublishedsequencesincludedinthephylogeneticanalyses. Pleasenote:Asaservicetoourauthorsandreaders,thisjournal providessupportinginformationsuppliedbytheauthors.Such materialsarepeer-reviewedandmaybere-organizedforonline delivery,butarenotcopy-editedortypeset.Technicalsupport issuesarisingfromsupportinginformation(otherthanmissing les)shouldbeaddressedtotheauthors.REFERENCESBaum,D.A.andWhitlock,B.A. (1999)Plantdevelopment:geneticcluesto petalevolution. Curr.Biol. 9 ,R525R527. Bowman,J.L. (1997)Evolutionaryconservationofangiospermowerdevelopmentatthemolecularandgeneticlevels. J.Biosci. 22 ,515527. Bowman,J.L.,Smyth,D.R.andMeyerowitz,E.M. (1991)Geneticinteractions amongoralhomeoticgenesofArabidopsis. Development 112 ,120. Bradley,D.,Carpenter,R.,Sommer,H.,Hartley,N.andCoen,E. (1993) ComplementaryFloralhomeoticphenotypesresultfromoppositeorientationsofatransposonattheplena-locusof Antirrhinum Cell 72 ,8595. Brockington,S.F.,Alexandre,R.,Ramdial,J.,Moore,M.J.,Crawley,S., Dhingra,A.,Hilu,K.,Soltis,D.E.andSoltis,P.S. (2009)Phylogenyofthe Caryophyllalessensulato:revisitinghypothesesonpollinationbiologyand perianthdifferentiationinthecoreCaryophyllales. Int.J.PlantSci. 170 627643. Cantino,P.D.,Doyle,J.A.,Graham,S.W.,Judd,W.S.,Olmstead,R.G.,Soltis, D.E.,Soltis,P.S.andDonoghue,M.J. (2007)Towardsaphylogenetic nomenclatureofTracheophyta. Taxon 56 ,822846. Coen,E.S.andMeyerowitz,E.M. (1991)Thewarofthewhorlsgenetic interactionscontrollingowerdevelopment. Nature 353 ,3137. Drea,S.,Hileman,L.C.,DeMartino,G.andIrish,V.F. (2007)Functionalanalysesofgeneticpathwayscontrollingpetalspecicationinpoppy. Development 134 ,41574166. Endress,P.K. (1994) DiversityandEvolutionaryBiologyofTropicalFlowers Cambridge,UK:CambridgeUniversityPress. Geuten,K.,Becker,A.,Kaufmann,K.,Caris,P.,Janssens,S.,Viaene,T., Theissen,G.andSmets,E. (2006)PetaloidyandpetalidentityMADS-box genesinthebalsaminoidgeneraImpatiensandMarcgravia. PlantJ. 47 501518. Goto,K.andMeyerowitz,E.M. (1994)Functionandregulationofthe Arabidopsisoralhomeoticgene Pistillata GenesDev. 8 ,15481560. Hiepko,P. (1965)Vergleichend-morphologischeundentwicklungsge-schichtlicheUntersuchungenu berdasPerianthbeidenPolycarpicae. Bot.Jahrb. Syst. 84 ,359508. Hofmann,U. (1994)Flowermorphologyandontogeny.In Caryophyllales: EvolutionandSystematics (Behnke,H.-D.andMabry,T.J.,eds).Berlin: Springer,pp.123166. Irish,V.F. (2003)Theevolutionoforalhomeoticgenefunction. Bioessays 25 ,637646.Irish,V.F. (2009)Evolutionofpetalidentity. J.Exp.Bot. 60 ,25172527. Jack,T.,Brockman,L.L.andMeyerowitz,E.M. (1992)Thehomeoticgene Apetala3of Arabidopsisthaliana encodesaMADSboxandisexpressedin petalsandstamens. Cell 68 ,683697. Jaramillo,M.A.andKramer,E.M. (2004)APETALA3andPISTILLATAhomologsexhibitnovelexpressionpatternsintheuniqueperianthof Aristolochia (Aristolochiaceae). Evol.Dev. 6 ,449458. Jaramillo,M.A.andKramer,E.M. (2007)Theroleofdevelopmentalgeneticsin understandinghomologyandmorphologicalevolutioninplants. Int.J. PlantSci. 168 ,6172. Kaplan,D.R. (1973)Theproblemofleafmorphologyandevolutioninthe monocotyledons. Q.Rev.Biol. 48 ,437457. Kim,S.,Koh,J.,Yoo,M.J.,Kong,H.Z.,Hu,Y.,Ma,H.,Soltis,P.S.andSoltis, D.E. (2005)ExpressionoforalMADS-boxgenesinbasalangiosperms: implicationsfortheevolutionoforalregulators. PlantJ. 43 ,724744.202 SamuelF.Brockington etal. 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203

PAGE 11

Klak,C.,Khunou,A.,Reeves,G.andHedderson,T. (2003)Aphylogenetic hypothesisfortheAizoaceae(Caryophyllales)basedonfourplastidDNA regions. Am.J.Bot. 90 ,14331445. Kramer,E.M.andIrish,V.F. (1999)Evolutionofgeneticmechanisms controllingpetaldevelopment. Nature 399 ,144148. Kramer,E.M.andIrish,V.F. (2000)Evolutionofthepetalandstamen developmentalprograms:evidencefromcomparativestudiesofthelower eudicotsandbasalangiosperms. Int.J.PlantSci. 161 ,S29S40. Kramer,E.M.andJaramillo,M.A. (2005)Geneticbasisforinnovationsinoral organidentity. J.Exp.Zool.BMol.Dev.Evol. 304B ,526535. Kramer,E.M.,Dorit,R.L.andIrish,V.F. (1998)Molecularevolutionofgenes controllingpetalandstamendevelopment:duplicationanddivergence withintheAPETALA3andPISTILLATAMADS-boxgenelineages. Genetics 149 ,765783. Kramer,E.M.,DiStilio,V.S.andSchluter,P.M. (2003)Complexpatternsof geneduplicationintheAPETALA3andPISTILLATAlineagesofthe Ranunculaceae. Int.J.PlantSci. 164 ,111. Kramer,E.M.,Holappa,L.,Gould,B.,Jaramillo,M.A.,Setnikov,D.andSantiago,P.M. (2007)ElaborationofBgenefunctiontoincludetheidentityof noveloralorgansinthelowereudicotAquilegia. PlantCell 19 ,750766. Krizek,B.A.andMeyerowitz,E.M. (1996)TheArabidopsishomeoticgenes APETALA3andPISTILLATAaresufcienttoprovidetheBclassorgan identityfunction. Development 122 ,1122. Malcomber,S.T.andKellogg,E.A. (2004)Heterogeneousexpressionpatterns andseparaterolesoftheSEPALLATAgeneLEAFYHULLSTERILE1in grasses. PlantCell 16 ,16921706. Mizukami,Y.andMa,H. (1992)Ectopicexpressionoftheoralhomeoticgene agamousintransgenicArabidopsisplantsaltersoralorganidentity. Cell 71 ,119131. Nylander,J. (2004) MrModeltest,version2 .Uppsala:EvolutionaryBiology Centre,UppsalaUniversity,http://www.abc.se/~nylander/mrmodeltest2/ mrmodeltest2.html Park,J.H.,Ishikawa,Y.,Yoshida,R.,Kanno,A.andKameya,T. (2003) ExpressionofAODEF,aB-functionalMADS-boxgene,instamensandinnertepalsofthedioeciousspeciesAsparagusofcinalisL. PlantMol.Biol. 51 ,867875. Park,J.H.,Ishikawa,Y.,Ochiai,T.,Kanno,A.andKameya,T. (2004)Two GLOBOSA-likegenesareexpressedinsecondandthirdwhorlsofhomochlamydeousowersin Asparagusofcinalis L. PlantCellPhysiol. 45 325332. Payer,J.-B. (1857) Traited’OrganogenieCompareedelaFleur .Paris:Masson. RonseDeCraene,L.P. (2007)Arepetalssterilestamensorbracts?Theorigin andevolutionofpetalsinthecoreeudicots Ann.Bot. 100 ,621630. RonseDeCraene,L. (2008)Homologyandevolutionofpetalsinthecore eudicots. Syst.Bot.33 ,301325. Soltis,D.E.,Ma,H.,Frohlich,M.W.,Soltis,P.S.,Albert,V.A.,Oppenheimer, D.G.,Altman,N.S.,dePamphilis,C.W.andLeebens-Mack,J. (2007)The oralgenome:anevolutionaryhistoryofgeneduplicationandshifting patternsofgeneexpression. TrendsPlantSci. 12 ,358367. Sommer,H.,Nacken,W.,Beltran,P.,Huijser,P.,Pape,H.,Hansen,R.,Flor,P., Saedler,H.andSchwarzsommer,Z. (1991)Propertiesofdeciens,a homeoticgeneinvolvedinthecontrolofowermorphogenesisin Antirrhinummajus Development 9 ,169175. Stellari,G.M.,Jaramillo,M.A.andKramer,E.M. (2004)Evolutionofthe APETALA3andPISTILLATAlineagesofMADS-box-containinggenesinthe basalangiosperms. Mol.Biol.Evol. 21 ,506519. Takhtajan,A.L. (1991) EvolutionaryTrendsinFloweringPlants .NewYork: ColumbiaUniversityPress. Walker,J.W.andWalker,A.G. (1984)UltrastructureoflowerCretaceous angiospermpollenandtheoriginandearlyevolutionofoweringplants. Ann.Mo.Bot.Gard. 71 ,464521. Whipple,C.J.,Zanis,M.J.,Kellogg,E.A.andSchmidt,R.J. (2007)ConservationofBclassgeneexpressioninthesecondwhorlofabasalgrassand outgroupslinkstheoriginoflodiculesandpetals. Proc.NatlAcad.Sci. USA 104 ,10811086. Zachgo,S.,Silva,E.D.,Motte,P.,Trobner,W.,Saedler,H.andSchwarzsommer,Z. (1995)Functional-analysisofthe Antirrhinum oralhomeotic deciensgenein-vivoandin-vitrobyusingatemperature-sensitive mutant. Development 121 ,28612875. Zwickl,D. (2000) Geneticalgorithmapproachesforthephylogeneticanalysis oflargebiologicalsequencedatasetsunderthemaximumlikelihoodcriterion ,PhDThesis.Austin:UniversityofTexas.Alternativepetalidentityin‘Livingstones’ 203 2011TheAuthors ThePlantJournal 2011BlackwellPublishingLtd, ThePlantJournal ,(2012), 69 ,193203