UniversidadPolitécnicadeMadrid

EscuelaTécnicaSuperiordeIngenieríaAgronómica,Alimentaria,ydeBiosistemas

“CaracterizaciónfuncionaldelasproteínasMTV9yMTV11ysuimplicacióneneltráficovacuolar”

TESISDOCTORAL

Doctorando:GuillermoRuanoBlanco

MásterenBiotecnologíaAgroforestal(UPM)

Madrid,2017

DepartamentodeGenéticaMoleculardePlantas(CentroNacionaldeBiotecnología,CSIC)

“CaracterizaciónfuncionaldelasproteínasMTV9yMTV11ysuimplicacióneneltráfico

vacuolar”

Doctorando:GuillermoRuanoBlanco

LicenciadoenFarmaciayenBioquímica

Directoresdetesis:Dr.DonEnriqueRojodelaViesca,Dr.DonJanZouhar

Madrid,2017

Quieroagradecera todos losmiembrosdeldepartamentodeGenéticaMoleculardePlantasdelCNB, a los compañeros y profesores del CBGP el tiempo que me han dedicado durante eldesarrollo delmáster y de esta tesis. Durante estos años vosotrosme habéis permitido darmecuenta, que la tarea ardua de investigar es emocionante y que la hora de salida demi jornadalaboralnomepreocupase.Entre todosellosquierohacermenciónespeciala JanZouharpor suamistadysudedicaciónhonestaencualquiermomentoquelerequiriese,haluchadomuchoporqueaumentasemisentidocríticoyporquemitrabajofuesemejorandocadadíaeneltranscursodeestatesis.QuieroagradeceraEnriqueconfiarenmíysuayudaalolargodetodosestosañosinclusoentiemposduros.Losproyectosenlosquehetrabajadoconélsiempremehaninteresadomucho,yhaconseguidoquenadamásverlequisiesepreguntarlesobreellos.Surigorcreoquehaconseguidoqueseaunmejorprofesionalyespero,mepermitaenel futuroseguiravanzandoanivelcientífico.Además,agradezcomuchoaJanyEnriquequeseanpersonascultasentiemposenlos que Vargas Llosa o Muñoz Molina son confundidos con nombres de bufetes de abogados.Quiero dar las gracias a Piluca, que era la primera persona que veía cuando llegaba al labo, suhumildad,susganasdeenseñarme,suagradablecharlaysucompañíaduranteestosaños.Quieroagradecer a Michi su pasión por la ciencia y haber compartido esto con él, su ayuda en ellaboratorio y su implicación y resultados en este proyecto. A Alfonso quiero agradecer sucompañía,suayudaenellaboratorioyesagraciaandaluzaqueamímeflipa.ASilvina,porhabercompartidorisasyserunaagradablecompañeraquesabedistinguirelcortedelacarneargentinoycuyalaborconMTV11esesencialparaestetrabajo.AMaite,porsupacienciayporhacermeverloimportantedelordenenestetrabajo.AMarco,porsuprofesionalidadysuexcepcionaltrabajoquemehaservidocomoreferenciaparalaredaccióndeestetrabajo.ARamón,porvenirdespuésque yo, entenderme y por lo mucho que nos hemos apoyado en el laboratorio. A Mary Paz,porqueaunquecoincidípocotiempoycompartímomentosenelcongresodePraga,estoysegurodequehubiesesidounaexcelentecompañera.AFelipe,porcontinuarunexperimentocuandoyoestuveausente.AmiscompañeroschilenosAlbertoyManuelPaneque,quehamerecidolapenapasar tiempo con ellos. A Liwen Jiang y a Jinbo Shen por su profesionalidad y porque han sidoesenciales en la progresión de este trabajo. También quiero agradecermi tiempo en el CBGP aJesúsVicenteyJoaquínMedinaduranteelmásteryalfinaldemitesis.QuieroagradeceraTomásCascón,sucapacidadparamotivarmeyapoyarmeyporhabersidocapazdeconocermemuybienyapreciarme.ACarmen,porabrirmelaspuertasdesulaboydesudespachosiemprequelohenecesitado,yporpreocuparsedequetrabajasetranquiloenmomentosdifíciles.ACarmenSimón,porsusimpatíaypornomolestarsenuncapormiscontinuasvisitasaEnrique.ARuthyYovanny,por sermaravillososcompañerosde tesisy sobretodoamigos.AGemma,por susonrisa,por sudisposición,ysucapacidadparaescucharmeyentendermisinseguridades.AAndrés,porserbuencompañeroyporsuscharlassobreciencia.AGabriel,porserungranamigoycompañeroyporhacerquemiestanciaenelCNBmerecieselapena.AMabel,porsuexcepcionalayudaconelY2H,yporsuexquisitotéquemetrajodeChina.ABárbara,porsuhumildad,subondad,suamistadyserunagrancompañeradetrabajo.ACarlosAlonsoBlanco,porsusganasdeenseñarmegenéticay suayuda con losexperimentosde floración.APilarCubas,por sudisposicióny ayuda con losensayosGUSdelaparatovasculardeestetrabajo.AVicenteyLauraporsuinestimableayudaparasabermásacercadelmutantealix queheutilizadoeneste trabajo.AMaríapor serunabuenaamiga y excepcional compañera. A Jon, por los buenos momentos durante las reunionestrasplanta.ASalomé,porsuejemploypornuncanegarseasolucionarmisdudas.AAbe,CristinaNieto,CristinaEspinosaCristinaMartínez,StellayCarlosporsergrandescompañerosydejarmeesterilizarlosmediossiemprequelonecesitaba.AAntonioLeyvaporsudisposiciónyayudaenel

desarrollodeestatesis.ÉljuntoaJavierPazAresylosmiembrosdel312mehicieronpasarmuybuenos ratos durante el almuerzo que fueronmuy necesarios para desconectar demi tarea. ACristina Navarro por su ejemplo y su amistad. A Elisa por ser una persona humilde y buenacompañera. A Marisa, Raquel y Beatriz por su contribución a este trabajo y su excelentedisposiciónhaciamis pedidospara cultivo in vitro.A Silvia por su simpatía y porque sin ella lasfotosdemicroscopíaconfocaldeestetrabajonohubieransidoposibles.AlserviciodeInvernaderoporayudarmeenelcuidadodemisplantasyenlospedidosquesolicitéduranteestatesis.Ahorametocacentrarmeenlagentedefueradelaciencia,misamigosdeFarmacia,quesiguensiendomis amigos después de tantos años, Dani, Carolina, Isa,Marta, Javi, Ana, Oihana, Irene,Natalia,Chus,Sergio,Michel,Lucía,Dave.Soisgeniales,nosayudamosmuchísimo,ycadavezquequedamosaunquenoseaconlafrecuenciaquemegustaríayconlapenadequealgunosestéislejos siempre demostramos que somos grandes compañeros de aventuras y excepcionalespersonasquecontribuísaquesealabuenapersonaquesoyyquequieroseguirsiendo.AAlbertoGorgojo,micompañerodecolegio,queaunquehayapasadomuchotiemposinvernos,hasidoungranapoyoysiemprehadichoqueteníaunamigocientífico.AJavierMuñoz,porquemesiguesqueriendoyporquenopuedepasarmástiemposinquetehagaunavisitaaHospitalet.ACristina,miamigaanestesista,quemehasapoyadocomofarmacéuticoymehasdadomuchoapoyoparacontinuarconmitesis.AClara,Óscar,MiriamÁvila,Nuria,Tamara,Coral,Mar,Miriam,Joséquemehabéis abierto losbrazos, habéis confiadoenmí yque vuestroejemplomeha servidoparacrecercomoprofesionalsanitarioymejorarlaempatíahacialospacientes.Ahoratocaescribirunaslíneasamifamilia,mispadresJesúsyPepa,porquemequeréismucho,porquenuncame faltadenada,porquemeseguísapoyandoapesarde todoparaqueseaunabuenapersona,sincerayhonesta.Porquemantenéislacabezafríaymeayudáismuchoaquenosea inmaduro ni quejica en ámbitos laborales. Ami hermano Jesús, porque sé queme quieresmucho y que aunque últimamente te vea menos y no perciba que confíes en mi desarrolloprofesionalséqueeresunexcepcionalpadreyqueRoser,mismaravillosossobrinosDaríoyElisaquehancambiadomivida, tussuegrosTerey Juan, tuscuñadosmehacensentirquesoydesufamiliaymehanacogidoconlosbrazosabiertos.TambiénamistíosGuillermoyAgueda,yamisprimasPilaryMarta,alosquequieromuchoysonmaravillosos.AmistíosPaco,MaxiySaturyamiabuelaMáxima.Porúltimo,amitíaDolores,quetesigorecordandoyhassidoungranejemplodeeleganciayvaloresqueexplicanloquesoyahoramismo.Yporúltimoati,TiffanyAnnFreda,queeltiempoquellevamosjuntosyeltiempoquenosesperaesyvaasermaravilloso.Noheconocidoanadiequemeescuche,meentiendayquemecuidecomo tú. Siempre tienesuna sonrisa, contigo siemprehayposibilidadde llegar a acuerdos, quehemoscrecidoy crecemos juntos,que ladistancianonoshaseparado,quenuestrosviajeshansidomágicos, que eres una gran profesional, que tienes el carisma que amíme falta, quemeaceptastalycómosoyyquetusamigossiemprequierensermisamigos,yquehasaceptadoamisamigosyhashechoquemequieranmás.Quequieresamispadres,yqueestánmuy felicesdesaberqueestásenmivida.Whereverweendup,wehavetocontinuefollowingourdreamsandgoalsandhelpingeachotheraswehavebeendoinguntilnow.Iwilltrymybesttoprovideyouabetterlifeandalwaysletyouknowthatthereisnothingyoucannotget.

Paraterminar,yparanoextendermemássóloquierodecir:

¡¡¡Muchísimasgracias!!!

Summary……………………………………………………………….1Resumen…………………………………………………………….…3Introduction………………………………………………………….51.Theendomembranesystemineukaryotes……………….......……………………………………………………62.Vesicletraffickingandthemachineryinvolved……………….……………………………………………………72.1.Vesicleformationmachinery.……………………………………..……………………………………………………72.2.Vesiclemovement……………………………………………………………………………………………………………142.3Vesiclefusion………………………………………………………………………………..…………………………………153.Thevacuoleinplants.Contentandphysiologicalfunctions………….……………………………………183.1.Lyticandstoragevacuoles…………...…………………………………………………………………………………193.2.Biotechnologicalpotentialofthevacuole:phytoremediationandstorage..……………………204.Vacuolartraffickinginplants.Routesandgenesinvolved…….……………………………………………214.1.Chemicalinhibitorsofvacuolartrafficking……..………………….……………………………………………214.2.Adaptorcomplexesdefinedifferentpathwaystothevacuole…………………………………………214.3.Geneexpansionforalternativepathways………..…………………………..…………………………………225.Geneticscreenstostudytraffickingtothevacuole………………………….…………………………………235.1.Geneticscreensfortraffickinggenesinyeast……….…………………………………………………………235.2.Geneticscreensfortraffickinggenesinplants…………………………………………………………………245.3.Geneticscreensfortraffickingtotheplantvacuole…………………………………………………………255.4.Screensusingchemicalgenomics………………………………………………….…………………………………27

Materialsandmethods……………………………………….291.Biologicalmaterial……………….………………………………………………………………………………………………301.1.Bacterialstrains……………….……………………………………………………………………………………………….301.2.Yeaststrains…………………….……………………………………………………………………………………………....301.3.Plantmaterial………………….………………………………………………………………………………………………..311.4Plasmids…………………………….………………………………………………………………………………………………322.Culturemethods………………….………………………………………………………………………………………………333.Methodsforbacterial,yeastandplanttransformation…….………………………………………………….374.Geneticinteractionsandphenotypicanalyses……………………………………………………………………..384.1.Arabidopsisthalianacrosses…………………………………….……………………………………………………….384.2.Floweringtimeanalyses………………………………………………….…………………………………………………384.Nucleicacidanalysisandextraction…………………………………..………………………………………………….395.Proteinanalyses……………………………………………………………….…………………………………………………..426.Microscopictechniques………………………………………………………………………………………………..………476.1.Confocalmicroscopy………………………………………………………..………………………………………………..476.2.Chemicaltreatments………………………………………………………..………………………………………………..47

Results………….……………………………………………………………481.Identificationofnovelmtvmutants…………………………………………………………….………………………...492.Mapbasedcloningofmtv9………………..……………………………………………………….………………….……..493.MTV9isaplantspecificgenewithaputativecoiled-coildomain………………….………………..……..524.MTV9expressioninRNA-seqdatasets…………………………………………………………………………………..565.MTV9promoteractivity………………………………………………………………………………………………..……….56

6.MTV9localizesprimarilyatthePVC…………………………………………………………….………………………..587.TheconservedC-terminaldomainofMTV9isresponsiblefortargetingtothePVC……..……….638.MTV9overexpressionperturbstransportofvacuolarcargobutnotPMproteins……….…………669.PVCaggregationbyMTV9overexpressionisnotaffectedbywortmannin………………….…………6910.CharacterizationofantibodiesagainstMTV9…………………………………………………………….…………7111.MTV9associateswithmembranesthatcorrespondtotheTGNandthePVC……………….………7312.MTV9andVTI11functioninseparatevacuolartraffickingpathways…………………………….……..7513.MTV11encodesahomologueofyeastVPS15…………………………………………………………….………..8014.MTV11/AtVPS15localizestoendosomalcompartments…………………………………………….………..8515.Themtv11-1mutanthasreducedPI3Plevels…………………………………………………………….…………8616.Themtv11-1mutantsshowincreasedgrowthinarsenatecontainingmedia…………….…………88

Discussion……………………………………………………………921.PhenotypicconsequencesofdisruptingMTV9activity…………………………………………………………..932.FunctionaldomainsinMTV9………………………………………………………………………………………………….943.MTV9localization…………………………………………………………………………………………………………………..964.MTV9mRNAexpression…………………………………………………………………………………………………………985.Ontheactivityofthemtv11-1allele………………………………………………………………………………………996.MTV9andMTV11,traffickingfactorsatthePVC………………………………………………………………….1017.Arsenatephytoremediation………………………………………………………………………………………………….102

Conclusions……….…………………………………………………103References…………………………………………………………..103

1

SummaryThis doctoral thesis adresses the characterization of MTV genes that are involved in vacuolar

traffickinginArabidopsis.Thevacuoleisanessentialorganelleforadaptativestrategiesofplants

andhas a great agronomic andbiotechnological importance for its storage capacityof essential

proteinsforhumannutritionorofrecombinantproteinsforbiotechnologicalpurposes.

Inthisthesis,twomutantsmodifiedintraffickingtovacuole(mtvmutants)havebeenstudiedand

pointmutationsintheMTV9/At1g24560andMTV11/At4g29380geneshavebeenidentifiedas

theresponsiblesforthedefectsinvacuolartrafficking.

The MTV9 gene is plant-specific and encodes a protein that is located at the prevacolar

compartment,possiblydirectedtothatcompartmentbythepalmitoylationofacysteinepresent

at itsC-terminalend.TheoverexpressionofMTV9provokestheaggregationof thepre-vacuolar

compartments and the delocalization of SNARE proteins, interfering with the trafficking of

proteins to vacuoles but not with the secretion of plasma membrane proteins. These results

indicatethatMTV9isessentialforthetraffickingtovacuolesandcouldbeinvolvedinanchoring

processesofvesiclesororganelleswiththeprevacuolarcompartmenttomediatetheirfusion.

The MTV11 gene encodes for the ortholog of the VPS15 protein, which is part of a

phosphatidylinositol-3-kinase (PI3K) complex, whose activity is required for all vacuolar related

traffickingroutes inyeastsandanimals.TheMTV11gene isessential inplantsand itsdisruption

causeslethality inpollen.Themutantmtv11-1 isolatedinthisthesis isahypomorphicallelethat

hasallowedtostudytheroleof thisgene inArabidopsisgrowthanddevelopment.Byusingthe

biomarker 2xFYVE we have obtained evidence that the synthesis of phosphatidylinositol-3-

phosphate issignificantlydecreased inthemutant, indicatingthat themutantallelereducesthe

activityofthePI3Kcomplex.Themutantmtv11-1hasaffectedthetraffickingofstorageproteinsin

seedsreservoirsandtherecyclingofplasmamembraneproteins.Thesedefectscauseadecrease

in growth and alterations in phyllotaxis under normal growth conditions. Moreover, different

assaystostudyresistancetoabioticstressesperformedinthemtv11-1mutant,haveshownthat

mtv11-1 plants are more tolerant to high concentrations of arsenate, a compound that is

2

sequestered in vacuoles once it is reduced to arsenite in the cytosol. This discoverymay have

important implications for the development of improved plants in phytoremediation of soils

contaminatedwitharsenic.

3

ResumenEstatesisdoctoralsecentraen lacaracterizacióndegenesMTVqueestán implicadosentráfico

vacuolarenArabidopsis.Lavacuolaesunorgánuloesencialen lasestrategiasadaptativasde las

plantas y tiene una gran importancia agronómica y biotecnológica por su capacidad de

almacenamientodeproteínasesencialesparalanutriciónhumanaodeproteínasrecombinantes

deusobiotecnológico.

Enestatesissehanestudiadodosmutantesmodificadosentráficoavacuolas(mutantesmtv)yse

hanidentificadomutacionespuntualesenlosgenesMTV9/At1g24560yMTV11/At4g29380como

lascausantesdelosdefectosentráficovacuolar.

ElgenMTV9esespecíficodeplantasycodificaunaproteínaqueselocalizaenelcompartimento

prevacuolar, posiblemente dirigida a ese compartimento por la palmitoilación de una cisteína

presente en su extremo C-terminal. La sobreexpresión deMTV9 provoca la agregación de los

compartimentosprevacuolaresyladeslocalizacióndeproteínasSNARE,interfiriendoconeltráfico

de proteínas a vacuolas pero no con la secreción de proteínas demembrana plasmática. Estos

resultados indican queMTV9 es esencial para el tráfico a vacuolas y podría estar implicado en

procesosdeanclajede vesículasuorgánulos conel compartimentoprevacuolarparamediar su

fusión.

El genMTV11 codifica para el ortólogo de la proteína VPS15, que forma parte de un complejo

fosfatidilinositol-3-kinasa (PI3K), cuya actividad es necesaria para todas las rutas de tráfico a

vacuola en levaduras y animales. El genMTV11 es esencial en plantas y su disrupción causa

letalidad en polen. El mutantemtv11-1 aislado en esta tesis es un alelo hipomórfico que ha

permitido estudiar el papel de este gen en el desarrollo de Arabidopsis. Mediante el uso del

biomarcador 2xFYVEhemosobtenido evidencias de que la síntesis de fosfatidilinositol-3-fosfato

está disminuida considerablemente en el mutante, indicando que el alelo mutante reduce la

actividaddelcomplejoPI3K.Elmutantemtv11-1tieneafectadoeltráficodeproteínasvacuolares

de reserva en semillas y el reciclaje de proteínas demembrana plasmática en la vacuola. Estos

defectosprovocanunadisminuciónenelcrecimientoyalteracionesenlafilotaxiaencondiciones

4

normales de crecimiento. Ensayos de resistencia a estreses abióticos han demostrado que las

plantasmtv11-1 sonmás tolerantes a altas concentraciones de arsenato, un compuesto que se

secuestra en vacuolas una vez es reducido a arsenito en el citosol. Este descubrimiento puede

tenerimportantesimplicacioneseneldesarrollodeplantasmejoradasparalafitoremediaciónde

sueloscontaminadosconarsénico.

5

Introduction

6

1.Theendomembranesystemineukaryotes.

ThepioneerstudiesbyGeorgeE.Paladeoninsulinsecretionwithinpancreaticcellsvisualizedthe

intricate network of endomembrane compartments comprising the secretory pathway (Palade,

1975). This network of intracellular compartments is referred to as the endomembrane system

andinplantsconsistsoftheendoplasmicreticulum(ER),theGolgiapparatus(GA),thetrans-Golgi

network/earlyendosome (TGN/EE), prevacuolar compartments (PVCs), also calledmultivesicular

bodies(MVBs),andthevacuole(Neumannetal.,2003).Theendomembranesystemofeukaryotic

cells allows spatial and temporal compartmentalization for the synthesis, sorting, delivery, and

degradation of cellular components. Possessing different compartments provides unique

environmentsforpost-translationalmodificationsandbiochemicalreactionsthatrequirespecific

conditions, such as a distinct pH. For example, the acidic pH of the vacuole/lysosome enables

degradationofproteins.

Most proteins that enter the endomembrane system do so in the ER, and then move

throughthedifferentcompartmentsuntilreachingtheirfinaldestination.Solubleproteinscontain

ahydrophobicsignalpeptideattheN-terminusthattargetsthemtothelumenoftheER(Blobel

andDobberstein,1975),andmay leavetheERoncetheyarequality-checkedforproper folding.

Transport between successive compartments can occur in three different ways: a) through a

vesicle trafficking step; b) through maturation of one compartment into another; c) through

heterotypic fusion between two consecutive compartments of the endomembrane system.

Proteins leave the ER from specific locations called ER exit sites (ERES) packed inside COPII

vesicles.COPIIvesiclesfuseatthecissideoftheGAreleasingtheircontents(BudnikandStephens,

2009). Transport along the GA occurs through cisternal maturation in a cis-to-trans direction,

coupledtobi-directionalvesicletraffickingbetweenthestacks(Orcietal.,1997).IntheGA/TGNa

major sorting event takes place that separates proteins targeted to the vacuolar pathway from

7

proteins targeted for secretion, packaging them in distinct vesicles. Secreted proteins follow a

defaultrouteandaretransportedfromtheGA/TGNinsecretoryvesiclesthatfusewiththeplasma

membrane. Soluble proteins targeted to the vacuolar pathwayhave specific sorting signals that

are recognized by vacuolar sorting receptors, which are transmembrane proteins that couple

cargorecognitioninthelumenofthecompartmenttorecruitmentofthemachineryrequiredfor

vesicle formation at the cytosolic side (Happel et al., 2004; Zouhar et al., 2010). Trafficking of

vacuolar cargo between the TGN and the PVC may occur through a vesicle trafficking step or

through maturation of the TGN into the PVC, without the proteins leaving the compartment

(Huotari andHelenius, 2011). ThePVCeventually fuseswith the vacuole, and vacuolar proteins

reach their final destination. In addition to the biosynthetic pathway to the vacuole, plasma

membrane proteins and secreted cargo are endocytosed to be recycled back to the plasma

membraneortobedegradedinthevacuole.Moreover,themachineryinvolvedinbiosyntheticor

endocytic trafficking may be recycled to prior compartments for further rounds of forward

transport.

2.Vesicletraffickingandthemachineryinvolved

Vesicle trafficking between compartments can be subdivided in three consecutive reactions: a)

cargorecruitmentandvesicleformationatthedonororganelle,mediatedbycargoreceptorsand

small GTPases that recruit coat proteins and effector proteins for vesicle budding; b) vesicle

movementalong thecytoskeletonmediatedbymotorproteins; c) fusionof thevesiclewith the

targetorganelleandcargodeliverymediatedbytetheringfactorsandSNAREproteins.

2.1.Vesicleformationmachinery

8

Foravesicletobeformed,themajorstepsinvolvedarecargorecognitionbythesortingreceptors,

coatrecruitment,membranedeformationandbuddingofthevesiclefromthedonormembrane.

• Sortingreceptors

Before anascent vesicle is formed, cargo selectionoccurs in the lumenof theorganelle,where

sorting receptors bind their cargo by recognizing sorting determinants. Sorting determinants in

plants are encoded in the amino acid sequence of the cargo proteins (Gershlick et al., 2014).

ReceptorsfunctioningintransportbetweenERandGolgiandthecorrespondingsortingmotifsare

conservedamongeukaryotes.TheCOPIIproteinsSEC23-24recognizesignalsforexportfromthe

ER (Yorimitsu et al., 2014)while the p24 cargo receptors and the ARF1GTPase are involved in

retrograde transport from the GA to the ER (Sun et al., 2007). In the TGN, vacuolar sorting

receptorsselectcargotobetransportedtothevacuole.Inplants,twofamiliesofvacuolarsorting

receptorshavebeendescribed,thevacuolarsortingreceptors(VSRs)andthereceptorhomology-

transmembrane-RINGH2 domain proteins (RMRs). Vacuolar sorting receptors (VSRs) are type 1

integralmembraneproteinsconsistingofalargelumenalN-terminaldomainresponsibleforcargo

binding, a transmembrane domain and a highly conserved cytosolic C-terminus that binds the

sorting machinery. Research performed with different vsr mutants in A. thaliana points to a

specializationwithinthisfamilyfortransportofstorageproteinsandsolubleproteinstothelytic

vacuole (LV) (Leeetal.,2013;Shimadaetal.,2003;Wangetal.,2011;Zouharetal.,2010).The

recognitionandbindingofVSRstocargoproteinsreliesonaminoacidmotifsinthecargoproteins,

whereas in mammals, cargo proteins contain mannose-6-phosphate modifications that are

recognizedbythesortingreceptors(Ghoshetal.,2003).RMRsarealsotype1integralmembrane

proteinswitha shorter lumenaldomainanda longC-terminal tail that contains aRINGdomain

(Shimada et al., 2003; Wang et al., 2011; Zouhar et al., 2010). RMRs have been localized to

9

endosomal compartments (Park et al., 2005a) but their physiological role is still controversial

(Bocock et al., 2009). AlthoughRMRs bind vacuolar cargo in vivo, no defects in trafficking have

beendescribedinmutantsofRMRgenes(Kimetal.,2005;Leeetal.,2013;Zouharetal.,2010).

• SmallGTPases

Cargo proteins bound to their sorting receptors accumulate in discrete siteswithin a particular

organelle, interacting with small GTPases that recruit cytosolic coat proteins to form a vesicle.

PlantgenomesencodesmallGTPasesfromtheRAB,RHO,ARF,andRANGTPasesubfamilies,but

noRASGTPaseshavebeen identified(Vernoudetal.,2003).TheRABandARFGTPasesregulate

the formation of vesicles on donor membranes and direct fusion with the target membrane

(Figure 1). These proteins associate with membranes in the GTP-bound form, which prevents

recognitionbytheRabchaperoneGDP-displacementinhibitor(GDI).SmallGTPasescyclebetween

active(GTP-bound)orinactive(GDP-bound)conformations,acycleregulatedbyGTPaseactivating

proteins (GAPs) and guanine nucleotide exchange factor proteins (GEFs). GAPs inactivate small

GTPasespromotingthehydrolysisofGTPandGEFsexchangeGDPbyGTPinordertorenderthem

active again.Although this basic cycle iswell-studied, it is unclearhow theseproteins associate

withaparticularmembranetoperformtheirspecificfunction(Nielsenetal.,2008).

10

Figure 1. Schematic representation of a role of small GTPases in vesicular fusion. The green objects represent two

different conformations of a small GTPase. When it is bound to GTP (green squares) the protein promotes vesicle

formation from the donormembrane andmay also direct vesicle fusionwith the acceptormembrane.GAP andGEF

proteins associate with small GTPases producing GTP binding and GTP hydrolysis respectively; and GDI, acts as a

chaperonethatpreventstheassociationofsmallGTPasesbindingtheGDPformofsmallGTPases.AdaptedfromDavid

Lambrightwebsite).

• Coatproteins

Fourtypesofcoatedvesicleshavebeendescribedinplants:clathrin-coatedvesicles(CCVs),coat

protein I (COPI)-coated vesicles, COPII-coated vesicles, and retromer coated vesicles (Figure 2).

Thesecoatsaremadeupofcytosolicproteins thatarerecruitedto thedonormembraneof the

11

nascentvesicle.Inadditiontotheirroleinshapingthevesicles,coatscontainsequencemotifsthat

recognizesortingreceptorsthatensurethatbothreceptorsandtheircargoeswillbe includedin

thenascentvesiclebeforeitsscission.Oncevesiclesarereleasedandtransportedbycytoskeletal

elements,vesiclesshedtheircoattofusewiththesubsequentcompartmentandeventuallythey

fuse with their target compartment to release their content. Interestingly, vesicles without

apparent coats forming from the Golgi or the ER and transporting vacuolar cargo have been

describedinplants(Fujietal.,2016;Wangetal.,2010),andmayrepresentalternativeroutesto

reachthevacuole.

Figure 2. Coat recruitment structure of different trafficking vesicles. The model shows adaptor proteins that are

responsible of cargo internalization; the GTPase enzymatic core, responsible for the identity and formation of the

vesicles;andtheouterlayerofthevesiclethatisdecoratedbycoatproteins.Adaptedfrom(Gurkanetal.,2006)

Adaptor protein complex!

GTPase!

b-propeller structure!

b-propeller and! a-solenoid structure!

a) COPII!

b) COPI!

c) Clathrin!

12

COPIIvesiclesare involved inanterogradeERtoGolgitrafficking.TheCOPIIcoatcomprisesfour

proteinswhicharearrangedasaninternalreceptor/cargo-bindingdimerofSEC23andSEC24and

anoutercagedimerofSEC31andSEC13(Staggetal.,2007).Theseheterodimercomplexesbind

activatedSAR1GTPase,andtogetherwithSEC23-24recruitthecargoprotein intheERexitsites

whereERmembraneisdeformedtoproducethenascentvesicle.SAR1GTPhydrolysisandrelease

ofCOPIIcoatsprecedesfusionwiththeGolgi(Yorimitsuetal.,2014).

CCVs are formed at the TGN and at the plasma membrane. Clathrin uses adaptor protein

complexes(APcomplexes)toconnectwithsortingreceptorsandtheirboundcargo.Therearefive

APcomplexesinanimalandplantcells(RobinsonandPimpl,2014).AP-1andAP-2recognizethe

conserved tyrosine motif present in sorting receptors and aid in the formation of CCVs at the

plasma membrane (AP-2) and the TGN (AP-1). AP-3 also interacts with clathrin to transport

membraneproteins to thePVCordirectly to the vacuole (Chenet al., 2011;Doreset al., 2012;

Saueretal.,2013;Zwiewkaetal.,2011).

COPI vesicles in plants are involved in retrograde transport from the Golgi to the ER and in

intercisternal trafficking between the Golgi stacks (Hwang and Robinson, 2009). The COPI coat

consistsoftwosubunits:F-COPsandB-COPs.COPIvesiclesinteractwithp24proteinstotransport

proteinssuchastheERD2receptorbacktotheER(Montesinosetal.,2014).

Retromer coated vesicles in plants are involved in retrograde transport of traffickingmachinery

(i.e: vacuolar sorting receptors) from the PVC to the TGN to perform further rounds of

anterograde transport. The retromer complex recognizes sorting receptors through the VPS26-

VPS35-VPS29 heterotrimer and together with sorting nexins bind phosphatidyl-inositol-3-

13

phosphate (PI3P), deforms the target membrane and performs the scission to form the

correspondingvesiclesforreceptorrecycling(McGoughandCullen,2011).

Uncoated vesicleswithout apparent protein cover havebeen reported tomediate trafficking of

storageproteins fromtheTGNto thevacuole inplantsand inotherorganisms (Gershlicketal.,

2014).

• Proteinsinducingmembranecurvature

Once the coat proteins have been recruited with the aid of small GTPases to the membrane

surface, themembrane needs to be physically deformed for formation and scission of vesicles.

Eachcellcompartmentpossessesaspecificmembranecomposition,mainlydifferentiatedbytheir

phosphoinositide (PtdIns) composition, which allows specific binding of proteins to deform the

membrane.PtdInsaremodifiedthroughphosphorylationatdifferentpositionsoftheinositolring

and specific forms label the different membrane compartments of the cell (Di Paolo and De

Camilli, 2006; Simonet al., 2014). Phosphatidylinositol-3-phosphate (PI3P), phosphatidylinositol-

4,5-phosphate (PI4,5P), and phosphatidylinositol-4-phosphate (PI4P) are the most abundant

PtdIns inplants (Simonetal.,2014)andtheir levelsare tightly regulatedbyvariouskinasesand

phosphatases (Balla et al., 2009; De Matteis and Godi, 2004). Analysis of overexpression and

knockoutmutants in PIP kinase/phosphatase genes have shown that some trafficking steps are

significantlyaffectedwhen thebalanceof thesePtdIns isnotproperlymaintained (Novakovaet

al.,2014;Tejosetal.,2014).SeveralproteindomainsthatbindPtdInswereidentifiedwithincoat

proteincomplexes,adaptorproteinsorotherproteins(e.g,EPSINs).Theinteractionbetweenthem

andthemembrane lipidsaid intheefficientmembranebendingandsubsequentvesiclescission

(Brett and Traub, 2006). Domains responsible for PtIns binding and for inducing membrane

curvature are the ENTH domain, ANTH domain, and BAR-domains (Zouhar and Sauer, 2014).

14

Proteinswith thesedomainsact incombinationwithcoatproteinsanddynamins todeformthe

membraneforbuddingandscissionofthevesicles(BrettandTraub,2006).PI4,5P isenrichedat

theneckofCCVwhereENTHandANTHdomainsbind,andPI3PbindingproteinssuchasFREE1or

FYVE1areresponsibleofcargorecruitmentanditsdeliveryintotheMVBs(Gaoetal.,2014).

2.2.Vesiclemovement

Thecytoskeletonservesasahighlydynamicplatformtoregulatethemovementandtheposition

ofthedifferentcellcompartments. Inplants,therearetwotypesofcytoskeletalfilaments,actin

filaments(F-actin)formedbypolymerizationofactinandmicrotubulesformedbypolymerization

of tubulin subunits. These filaments are dynamic structures capable of growth or disassembly

(Nogales, 2010). Motor proteins mediate the movement of vesicles along these cytoskeleton

tracks.Themotorproteinscontainaheaddomainresponsible forATPaseactivity thatenergizes

themovement,aneckthataidsthemovementandregulatetheATPaseactivity,andataildomain

thatbindstovesiclesororganelles(LeeandLiu,2004).Themotorproteinsattachedtotheactin

polymersarecalledmyosinsandtheonesattachedtomicrotubulesaredividedintokinesinsand

dyneins.Kinesinsmovetowardtheplusendofmicrotubulesanddyneinsmovetotheminusend.

Microtubuleandactin filamentsdisplaydifferent functionswithin thecell. In thecaseofplants,

where cells are larger and highly vacuolated, the use ofmyosinmotors for vesicle trafficking is

preferredsincetheyaremoresuitabletoreachlongerdistances.Infact,onlykinesin14familyhas

been associatedwith long distance transport in a retrogrademanner (likemammalian dyneins)

and the function of this protein depends on its interaction with actin (Jonsson et al., 2015;

Lawrenceetal.,2001;Reddy,2001;WicksteadandGull,2007;WilhelmJ.Walter,2015).Although

experimental evidence of myosin association with vesicles is lacking, the localization data of

certainplantmyosinsandtheeffectofactindisassemblysuggeststhatmyosinsareresponsiblefor

15

the movement of post-Golgi organelles (Avisar et al., 2012; Kim et al., 2005). In contrast, the

deliveryandorganizationofcellulosesynthasecomplexesintheplasmamembraneisgovernedby

corticalmicrotubulefilaments(Gutierrezetal.,2009).

2.3.Vesiclefusion

Once vesicles reach the target compartment, their membranes fuse to complete the transport

process.Thefusionmaybetransient(kissandrun)orpermanent(Giraudoetal.,2005),inwhich

case the vesicle membrane is incorporated into the target membrane. The major factors to

accomplishmembranefusionareSNAREproteins(solubleNSFAttachmentproteinreceptor),but

manyothercellularproteinsarerequiredtoensureproperfusionreactionsinvivo.

Multisubunittetheringcomplexesandhomodimericcoiledcoiltethers

Tethering factors play an essential function in membrane fusion reactions by establishing the

specificcontactsbetweenthedonorandtargetmembranecompartments(DubukeandMunson,

2016;YuandHughson,2010).ThesecontactsrelyonthefunctionalcycleofRabGTPasesandthe

shedding of the protein coat that allows the formation of a pre-fusogenic complex. The

multisubunit tethering complexes (MTCs) involved in the different trafficking steps of the

biosyntheticpathwayare:DSL1,forERtoGolgitrafficking;TRAPPIforGolgitoERtrafficking;COG,

for intra Golgi trafficking; TRAPPII and GARP for anterograde and retrograde TGN to Golgi

trafficking;CORVETforTGNtoPVCtraffickingandHOPSforPVCtovacuoletrafficking.Inaddition

toMTCs, several coiled coil proteins function as homodimeric tethering factors. In contrast to

MTCs, these homodimeric tethering factors show little sequence conservation across different

kingdoms (Kim and Bassham, 2011; Takahashi et al., 2010; Vukasinovic and Zarsky, 2016).

Unraveling the mechanisms governing the interaction between these tethering factors, the

16

arriving vesicles and the SNAREs involved in membrane fusion is key to understand how

competentfusogeniccomplexesareformedinthecell(ChiaandGleeson,2014).

SNAREandSMproteins

SNAREproteinsmediatethefusionbetweenthetargetmembraneandthevesicle,providingboth

specificityandenergytodrivethisprocess.SNAREscontainacoiled-coilregionreferredtoasthe

SNARE motif domain, which is essential for binding to other SNAREs and is widely conserved

among species (Weimbs et al., 1997).When a vesicle is tethered to a target compartment for

fusion,SNAREproteinspresentinthetwomembranesformatetramericbundleofcoiledhelices

thatbringsthemembranesclosetogether,eliminatingthewaterinterface,initiatingthemixingof

the lipids and eventually provoking the fusion between themembranes (Figure 3). It has been

shownthat theenergyreleased in the formationof thetetramericcomplex issufficient todrive

liposomal fusion (Wickner and Schekman, 2008). In vitro experiments have demonstrated that

there aremultiple combinations of SNAREs that are able to produce tetrameric bundles, but in

vivo only a few combinations lead to a successfulmembrane fusion (Varlamov et al., 2004). In

addition,ithasbeenproposedthattherearesomemembersoftheSNAREfamilythatcanactively

inhibitmembranefusionbyinteractingwithotherSNAREproteins(Bielopolskietal.,2014)orthat

a non-functional SNARE complex could assemble in vivo and inhibit the fusion process (Di

Sansebastiano,2013;Varlamovetal.,2004)

ThelocalizationofSNAREproteinsisusuallypredictedbysequenceconservationtocharacterized

SNAREsfromotherorganisms,asnoaminoacidsequencehasbeenidentifiedtolinkaparticular

SNAREprotein to its correspondingmembrane (Sanderfootetal., 2000; Scalesetal., 2000). For

instance,Qa (syntaxins) SNAREs arewell conservedbetween yeast andplants and thedifferent

typeslocalizeinanalogouscompartmentsinbothspecies.OnlyfortheSYP1SNAREclass,itcould

17

be assumed that its long transmembrane region targets this protein family to the plasma

membrane(Brandizzietal.,2002;Sanderfootetal.,2000).

SM(Sec1-Munc18)proteinsaresolubleperipheralmembraneproteinsthat interactwith

the syntaxin class of SNAREs to regulate the membrane fusion reaction (Hong and Lev, 2014).

Syntaxinsshiftbetweenopenandclosedconformationsthatarerespectively,competentandnot

competenttoformtetramericcomplexeswithotherSNARESanddrivemembranefusion.Current

evidence suggests that SM proteins may play a dual regulatory role in membrane fusion: by

bindingSNAREcomplexandcompletingfusionorbyinteractingwithsyntaxinsandstabilizingthe

closed conformation to prevent SNARE complex formation and membrane fusion. Examples of

both types of interactions have been described but a structural data about conformational

changes that allow the fusion is still missing (Archbold et al., 2014). Four major classes of SM

proteinsarepresentineukaryotes:SLY1,VPS45,VPS33andSEC1.Thesefourclassesintervenein

different trafficking stepsand togetherwithSNAREsprovide the specificity tomembrane fusion

reactionsinthecell.

Figure 3. Schematic diagram of a SNARE competent fusion complex. The R-SNARE protein necessary to form a

tetramericbundlewiththeQ-SNAREsresidingatthetargetmembraneisshowninblue.TheinteractionofSMproteins

withsyntaxinsand theotherSNAREs isalsoshown.Thesyntaxin-SM interaction isessential for thedifferentstepsof

membranefusionandthisassociationisregulatedbycalciumions.

18

3.Thevacuoleinplants.Contentandphysiologicalfunctions.

There are particularities of the plant endomembrane system that distinguish it from the

mammalianandyeastsystems,suchasthelackofanintermediatecompartmentbetweenthethe

ERandtheGolgi,thehighmotilityoftheGolgistacks,theprocessofcellplateformationandthe

presenceoflargecentralvacuoles(ContentoandBassham,2012;Dettmeretal.,2006;Kimetal.,

2005;Robinson,2014;ZouharandRojo,2009).Thepresenceoftheselargevacuolesinplantsisan

essential adaptation to the unique life style of these organisms (Rojo et al., 2001). These large

vacuolesprovideahighbufferingcapacitythatmaintainscytoplasmichomeostasisinthedifferent

environments where plants may happen to germinate. Moreover, large vacuoles allow for

energetically cheap growth which is essential for these autotrophic organisms to explore the

surroundingsfornutrients,waterandlight(ZouharandRojo,2009).Looseningofthecellwallby

enzymessuchasexpansinscoupled to thehigh turgorpressureprovidedbyvacuolesdrivescell

expansion(Cosgrove,2000;WangandRuan,2010)inaprocessthatisregulatedbyauxins(Lofke

etal.,2015).Thevacuoleisareservoirof ionsandmetabolitesandit iscrucialfordetoxification

andgeneralcellhomeostasis.Italsostoresproteinsandsolublecarbohydratesasreservesinseed

and vegetative tissues (Marty, 1999) and hydrolytic enzymes that function in recycling and

degradationofcellularcontents.Manyplantspeciescontainspecializedproteinstoragevacuoles

inseedtissues,whichaccumulatehighamountsofreserveproteinstobeusedduringgermination

and seedling establishment (Muntz, 2007). The proteins stored in seeds are an essential

agronomicalcommodityobtainedfromcropsandconstitutesthemainproteinsourceforhuman

andanimalnutrition(HermanandLarkins,1999).

19

3.1.Lyticandstoragevacuoles

It has been shown that plantsmay contain separate types of vacuoles within a single cell. For

instance, seed cells have been shown to contain both lytic vacuoles (LVs) and protein storage

vacuoles(PSVs)(Bolteetal.,2011;Frigerioetal.,2008).ThePSVisuniquetoplantsandisactually

a compoundorganellemadeupof three independentcompartments: thecrystalloidandmatrix

are separate structures that are responsible of storage protein accumulation,while the globoid

containsphyticandoxalatecrystalsassociatedwithmetalsthatprovideastableenvironmentfor

theaccumulationofcertainenzymes(Baudetal.,2008).Ithasbeensuggestedthatthegloboidis

alyticcompartmentinsidethePSV(JiangandSun,2002).Aquaporinmarkers(tonoplastproteins)

have been used successfully to show the existence of separate compartments (LV and PSV) at

some stages of development or as fused compartments (central vacuole) in the majority of

vegetative tissues (Frigerio et al., 2008). It has been proposed that the protein storage vacuole

originates from ER-derived compartments during seed maturation (Viotti, 2014). Throughout

embryo development the LV shrinks and appears as amembrane enclosed structure inside the

PSV(Bolteetal.,2011;Frigerioetal.,2008).DuringseedgerminationthePSVistransformedinto

a LV through acidification and progressive degradation of the stored proteins (Zheng and

Staehelin,2011).IthasalsobeensuggestedthattheERisthemembranesourceforLVformation

(Viotti et al., 2013). In vegetative tissues, where only a central vacuole with lytic features is

present,thepathwaysthattransportstorageproteinstothevacuolearestillfunctional(Sanmartin

et al., 2007). However, the physiological role of the PSV pathway in vegetative tissues remains

obscure.Interestingly,somepathogenesis-relatedproteins,e.g.thepathogenrelatedproteinPR5

and various lectins expressed in vegetative cells share the same sorting determinants as seed

vacuolarstorageproteinsandarelikelysortedbythesamepathwaysinvegetativetissues(Carter

etal.,2004).

20

3.2.Biotechnologicalpotentialofthevacuole:phytoremediationandstorage

Theplantvacuole is the largest compartment inplant cells,whichmakes ita suitable target for

biotechnological strategies aimed at improving the storage and the detoxification capacity of

plants.Forinstance,invegetativetissues,theoverexpressionoftonoplasttransporterstoincrease

the concentration gradient between the vacuole and the cytoplasm is one of the common

strategies to engineer plantswith properties such as heavymetal tolerance (jan Stomph et al.,

2009), salt tolerance, drought resistance (Park et al., 2005b), and accumulation of secondary

metabolites (Butelli et al., 2008). The storage capacity of PSVs, which accumulate enormous

amounts of proteins during seed maturation, makes them valuable biotechnological

compartments to store recombinant proteins. One of the major drawbacks for recombinant

protein expression in plant vacuoles is the saturation of the vacuolar trafficking capacity when

cargoproteinsareoverexpressed,whichleadstosecretionofthecargototheapoplasmbybulk-

flowmechanisms and limits the amount of protein produced (Denecke et al., 1990). A possible

solution to this problem would be to increase expression of the trafficking machinery that is

limiting for transport to the vacuole. However, the transcriptional regulation of the trafficking

genes in plants is not yet well characterized and the transcription factors involved in their

regulation remain mostly unknown. Finding those transcription factors would be crucial for

activating coordinately the expression of trafficking genes and increasing the transport capacity

(PizarroandNorambuena,2014).Inmammals,suchtranscriptionfactorsinvolvedincoordinated

activation of the trafficking machinery have been already identified (Sardiello et al., 2009),

supporting the idea that the capacity of trafficking pathways can be modulated through

transcriptionalregulation.

21

4.Vacuolartraffickinginplants.Routesandgenesinvolved

Genetic and pharmacological analysis has revealed the existence of parallel pathways for

traffickingtothevacuoleinplants.

4.1.Chemicalinhibitorsofvacuolartrafficking

Pharmacological inhibition has provided evidence for alternative trafficking pathways to the

vacuole being functional in plants. It was first shown that treatment with wortmannin, a

phosphatidylinositol 3-kinase and 4-kinase inhibitor, selectively blocked vacuolar trafficking of

barley lectin but not of sweet potato sporamin (Matsuoka et al., 1995). Wortmannin induces

homotypic fusion of the PVC, resulting in its enlargement and malfunction, which leads to

mistargeting of vacuolar cargo proteins to the apoplast (Jia et al., 2013). Brefeldin A (BFA) is a

fungaltoxinthatinhibitsBFA-sensitiveARF-GEFsandpreventsvesicleformationatdifferentsteps

of the intracellular trafficking pathways (Geldner et al., 2003). BFA treatment blocks vacuolar

traffickingpathwaysbutitdoesnotaffecttheAP-3dependentpathway,suggestingthattheAP-3

derivedvesiclesformedattheGolgibypassthePVConitswaytothevacuole(Feraruetal.,2010;

Wolfenstetteretal.,2012).

4.2.Adaptorcomplexesdefinedifferentpathwaystothevacuole

The adaptor protein complexes (AP-complexes) mediate sorting of receptors and associated

solublecargointospecificvesiclesfortargetingtotheirparticulardestination.Plantscontainfive

adaptor complexes (AP-1 to AP-5). The AP-2 complex is involved in endocytosis from the PM,

(Gadeyneetal.,2014)whileAP-1,AP-3andAP-4are involved intraffickingtothevacuole.AP-1

localizes at the TGN where it binds VSR1. Moreover, soluble vacuolar cargo sorted by VSRs is

22

abnormally secreted in ap-1 mutants, demonstrating that AP-1 is involved in their proper

transport tothevacuole (Happeletal.,2004;Parketal.,2013).Similarly, tonoplastproteinsare

mislocalizedandthevacuolebiogenesisiscompromisedinap-3mutants,suggestingthattheAP-3

complexalsofunctionsinvacuolartransport(Feraruetal.,2010;Zwiewkaetal.,2011).Recently,it

wasshownthattheAP-4complexresidesontheTGNseparatelyfromtheAP-1complex,whereit

alsobindsVSRreceptorsandisrequiredforvacuolartransportofseedglobulins(Fujietal.,2015),

suggestingthat,asisthecaseinanimalsandyeast,theAP-4pathwayisanalternativeroutetothe

canonicalAP-1dependentpathwayfortransportofvacuolarsortingreceptorsandtheircargoesto

thevacuole(Gershlicketal.,2014).

4.3.Geneexpansionforalternativepathways

Thespecialcharacteristicsoftheplantvacuolemayhaverequiredanexpansionintherepertoire

of genes involved in their biogenesis and function, including in the machinery for vacuolar

trafficking.Forinstance,ArabidopsisencodesfourQbVTISNAREs(VTI11,VTI12,VTI13andVTI14)

incontrasttothesingleVti1pproteinencodedbytheyeastgenome.Moreover,thereisevidence

thatthisexpansionhasresultedinfunctionalspecialization.VTI11residesatthePVCandformsa

SNARE complexwith SYP2 and SYP5 syntaxins,whereasVTI12 is located at the TGN in complex

with SYP4 and SYP6 syntaxins (Sanderfoot et al., 2001). Mutations in VTI11 and VTI12 affect

traffickingofdistinctvacuolarcargoes, supporting that theyact inparallel routes to thevacuole

(Sanmartinetal.,2007).

Another example of gene expansion and specialization in the vacuolar trafficking

machinery is the Rab5 GTPase subfamily. Arabidopsis encodes two conventional Rab5-type

GTPasesARA7andRHA1andaplantspecificisoform,ARA6(Becketal.,2012;Ebineetal.,2011).

In pollen tubes ARA7 and ARA6 containing vesicles are affected differently by the actin

23

depolymerizingdruglatrunculinB(LatB)(Zhangetal.,2010).ARA7andRHA1localizetothePVC

andareinvolvedinvacuolartrafficking.ARA6localizestothePVCandalsotheplasmamembrane

inacomplexwithSYP121andVAMP727,consistentwithadualfunctioninvacuolartraffickingand

in secretion. The common activator of these RAB GTPases is the guanine nucleotide exchange

factor VPS9. Interestingly, overexpression of constitutively active ARA7 but not ARA6

complementstherootgrowthdefectsofavps9mutant(Gohetal.,2007).Moreover,theara7and

therha1knockoutmutantsexacerbatethedevelopmentaldefectsofasyp22mutantwhereasthe

ara6mutantrescuesthem(Ebineetal.,2011),supportingthatARA6hasdifferentfunctionsfrom

the conventional Rab5 GTPases in plants. Moreover, by studying the MON1-CCZ complex that

functionsasaGEF forRab7GTPases,evidence for three independent routes to thevacuolehas

been provided: 1) a route that involves the maturation of RAB5 endosomes into RAB7 late

endosomes; 2) an AP-3 dependent route that does not involve RAB5 nor RAB7; 3) a RAB5-

dependent andAP-3 independent route (Cui et al., 2014; Ebine et al., 2011; Ebine et al., 2014;

Singhetal.,2014).

5.Geneticscreenstostudytraffickingtothevacuole

5.1.Geneticscreensfortraffickinggenesinyeast

ThefirstgenesencodingtraffickingmachinerywereisolatedingeneticscreensinSaccharomyces

cerevisiaeinthelate1970s.Thesecmutantsthatwereisolatedinatemperaturesensitivescreen,

showedatrestrictivetemperature(37°C)aberrantintracellularmembranousorganellesfilledwith

proteins that normally destined to the vacuole or plasmamembrane.Moreover, cell expansion

anddivisionwereaffectedbecauseofthedefectsinsecretionofcellwallremodelingenzymes.By

measuringtheactivityofapoplasticenzymessuchasinvertaseandacidphosphatase,thesec1and

24

sec2mutantswithcompromisedsecretionof theseenzymeswere isolated.Theaccumulationof

secretoryproteinsinendomembranestructuresinthesesecmutantsledtochangesincelldensity

inthemutants,whichwassubsequentlyusedtoisolatenovelmutants(Novicketal.,1980;Novick

and Schekman, 1979). In this way, sec mutants belonging to 21 complementation groupswere

found(Novicketal.,1981;Novicketal.,1980;Sataetal.,1998;SchekmanandNovick,2004).

Ascreenforvacuolarproteinsorting(vps)mutantswassetuptoidentifygenesspecifically

involved in trafficking to the vacuole. The screen searched for vps mutants that abnormally

secretedthevacuolarcarboxypeptidaseY,usingapep4leu2doublemutantbackground(Stevens

etal.,1982).Thepep4mutationpreventsprocessingandactivationofproCPYinsidethevacuole,

while the leu2mutation renders the strain auxotrophic for leucine. In the screening, amedium

containing theN-CBZ-L-phenylalanyl-L-leucine dipeptidewas used. In the vpsmutants, secreted

proCPYisprocessedbyperiplasmicpeptidasesandthencleavesthedipeptide,allowinggrowthin

Leu-media (Rothman and Stevens, 1986). 41 vps mutants were isolated that classified into six

differentphenotypicclasses(AtoF),dependingonthealterationsinmorphologyofthedifferent

endomembranecompartments(Bantaetal.,1988;Raymondetal.,1992).Anindependentscreen

forvacuolarmorphology(vam)mutantsresistanttocloroquine,whichaccumulatesinthevacuole

and provokes toxicity, yielded nine vam mutants (Wada et al., 1992). In 2002, a genome-wide

reversegeneticscreenusingacollectionof4653homozygousdiploidgenedeletionyeaststrains

andscoringforCPYsecretionbywestenblotyielded93newVPSgenes(Bonangelinoetal.,2002).

5.2.Geneticscreensfortraffickinggenesinplants

Plasma membrane transporters such as PIN1 (auxin efflux membrane protein) and tonoplast

transporterssuchasTIP2;1(vacuolaraquaporin)fusedtofluorescentproteinshavebeenusedas

markersto identifymutantswithalteredsubcellulardistribution(Avilaetal.,2003;Feraruetal.,

25

2010).PIN1isendocytosedandtheneitherrecycledbacktotheplasmamembraneortargetedto

the vacuole for degradation. In a screen of amutagenized population ofM2 plants expressing

PIN1-GFP,twopatmutants(proteinalteredtrafficking)showingPIN1accumulationinintracellular

endosomeswereisolated.Theircharacterizationrevealedthatthedefectivetraffickingwasdueto

mutations inAP-3 components, demonstrating thatAP-3was required for vacuolar targeting of

PIN1andfor lyticvacuolebiogenesis (Feraruetal.,2010;Zwiewkaetal.,2011). Inascreenofa

mutagenized population of M2 plants expressing GFP-TIP2;1 (Avila et al., 2003) , two of them

werecharacterized:themvp1mutantthatshowsalteredERmorphologyandretentionofvacuolar

and secretory proteins in the ER (Agee et al., 2010) and the csp-1 mutant that affects vacuole

morphology and cell shape (Chary et al., 2008). A screen ofmutagenized plants containing the

Golgi associated protein sialyl transferase fused toGFP (ST-GFP) led to the identification of the

gom8mutantshowinganaberrantERmorphologyduetoamissensemutationinRHD3,aGTPase

involvedinmorphogenesisoftheER(Stefanoetal.,2012).

5.3.Geneticscreensfortraffickingtotheplantvacuole

Several genetic screenshavebeendeveloped inplants specifically aimedat identifying vacuolar

traffickinggenes.Basedon theobservation thatamutant in thevacuolar sorting receptorVSR1

secreted storage proteins (12S globulins and 2S albumins) that accumulated as unprocessed

precursors(Shimadaetal.,2003),acollectionof28000T-DNAmutantswasscreenedtoidentify

mutants defective in processing of storage proteins. This rendered eight mutants, referred as

maigomutants (Lietal.,2013;Shimadaetal.,2006;Takagietal.,2013;Takahashietal.,2010),

fourofwhichhavebeenalreadycharacterized.ThefirstMAIGOgeneidentifiedwasVPS29,coding

forasubunitof the retromercomplex.ThismutantaffectsVSRrecycling fromthePVCandasa

consequencestorageproteinsaresecreted(Shimadaetal.,2006).Inadditiontothis,thegroupof

26

Dr.Hara-Nishimura developed another screen based on the increased fluorescence observed in

mutantsthatsecretetheartificialvacuolarcargoGFP-CT24,whichconsistsofGFPfusedtotheC-

terminal sortingsignalof thesoybeanstorageproteinconglycinin (Nishizawaetal.,2003).From

thisscreening,morethan100mutantsshowingagreenfluorescentseedphenotype(gfsmutants),

indicative of missorting of the transgene GFP-CT24 to the apoplast, were isolated. GFS1 was

shown to encodeVSR1,GFS2 endodedKAM2 (gfs2), amembraneprotein involved in endocytic

processes, andGFS9 encoded a protein required for anthocyanin accumulation in the vacuoles.

Thegfs9mutantshowsthatalterationsintraffickingofvacuolarproteinscanaffectthedeposition

ofothervacuolarcontent(Fujietal.,2007;Ichinoetal.,2014).

Barley lectin is a vacuolar protein that contains a C-terminal vacuolar sorting signal (Ct-

VSS) that is necessary and sufficient to target soluble endomembrane cargo into the vacuole

(Bednarek andRaikhel, 1991;Rojoet al., 2001). CLAVATA3 (CLV3) is anextracellular ligand that

activates the CLAVATA signaling pathway to restrict the size of the stem cell pool in the shoot

apicaland flowermeristemsofArabidopsis (Fletcheretal.,1999).Agenetic screen forvacuolar

traffickingmutantswassetupusingachimericfusionproteinbetweenCLV3andthebarleylectin

Ct-VSS expressed under a constitutive 35S promoter (VAC2 construct). In wild type plants the

VAC2protein (CLV3-CtVSS)accumulates in thevacuolewhere it is inactive (Rojoetal.,2001). In

mutantswheretraffickingto thevacuole is impaired,VAC2 issecretedto theapoplast,where it

reduces the pool of stem cells in the shoot apical meristem and may eventually terminate it

(Sanmartinet al., 2007).With this assay,modified trafficking to the vacuole (mtv)mutantswith

terminatedmeristemswere isolated. In apreliminary screen fourmutants (mtv1 tomtv4)were

found.MTV1encodesanepsinrelatedprotein;MTV2encodesthevacuolarsortingreceptorVSR4;

MTV3encodesthephosphoinositidephosphatasePTEN2bandMTV4encodestheARF-GAPAGD5

(Saueretal.,2013).Mutantsinsomeoftheparaloguesofthesegenesalsohadmtvphenotypes,

27

namelythevsr1,vsr3,agd12andpten2bamutants. Inaddition,reversegeneticanalysisshowed

thatmutants in the SNAREVTI12and in its SM-regulatorVPS45also secreted theVAC2protein

resultinginterminatedmeristems(Sanmartinetal.,2007;Zouharetal.,2009).

5.4.Screensusingchemicalgenomics

Thetechniquesthatcombinetheuseofchemicallibrariesofcompoundswiththeeffectofthese

compounds in vivoare referredas chemical genomics, and represent apowerful tool todissect

essentialand/orredundantcellmechanisms.Asmallmoleculecantargetallmembersofthesame

proteinfamilyandrevealaneffectthatwouldnotbedetectedifonlyonememberwasdisrupted.

Furthermore,smallmoleculescouldtargetessentialgenesthatarelethalwhenknockedout(Hicks

andRaikhel,2012).Sortinsareinhibitorsofvacuolartraffickingdiscoveredinachemicalgenomics

screenbymeasuringCPYsecretioninyeast.Someoftheisolatedsortinsweresubsequentlyshown

to alter vacuolar trafficking and/or vacuolarmorphology inplants (Zouhar et al., 2004).Genetic

screenshaveyieldedmutantshypersensitivetocertainsortinsinyeastandplants(Chandaetal.,

2009;Norambuenaetal.,2008;Rosadoetal.,2011).Inascreenforcompoundscausingdefectsin

gravitropism in Arabidopsis, 34 molecules causing altered gravitropic response and aberrant

morphologyoftheendomembranesystemwereisolated(Surpinetal.,2005;Surpinetal.,2003;

Yano et al., 2003). Another example of a small molecule derived from chemical genomics that

affects trafficking is endosidin1. Using pollen expressing GFP-RIP1 fluorescent marker and

analyzingitsgerminationinvitroledtothediscoveryofafamilyofdrugscalledendosidins.Some

ofthemcausedmislocalizationofGFP-RIP1andresultedinincreasedcytosoliclocalizationofGFP-

RIP1, inappropriate localizationat the tip,and lossofpolargrowth.RIP1 isaplasmamembrane

proteinthat interactswithROPandlocalizesatthetipofpollentubes.Theactivityofthesetwo

proteins is essential for polar growth that is necessary for pollen germination. Indeed, pollen

28

grains treated with endosidin1 are not able to grow in a polar manner, and RIP1-GFP is

mislocalizedtothecytoplasm(Robertetal.,2008).

29

MaterialsandMethods

30

1.Biologicalmaterial

1.1.Bacterialstrains

• ForpropagationofvariousplasmidstheDH5αstrainofEscherichiacoli(E.coli)wasused.

• For recombinantproteinexpression inbacterial cultures theBL21 (DE3)RosettaTM E.coli

strainwasused.

• ForpropagationandpurificationofGatewayTMplasmidscarryingtheccdBgenetheDB3.1

E.colistrainwasused.

• For transient expression in Nicotiana benthamiana and stable transformation of

Arabidopsis thaliana the C58C1 strain of Agrobacterium tumefaciens harbouring the

pGV2260plasmidwasused.

1.2.Yeaststrains

• Saccharomyces cerevisae AH109: MATa, trp1-901, leu2-3, 112, ura3-52, his3-200, gal4Δ,

gal80Δ(Clontech TM).Thisstrain isused forauxotrophydependentyeast tohybrid (Y2H)

experimentsandmatingpurposeswithY187strain.

• SaccharomycescerevisaeY187:MATα,ura3-52,his3-200,ade2-101,trp1-901,leu2-3,112,

gal4Δ, met–, gal80Δ (Clontech TM). This strain is used for auxotrophy dependent Y2H

experimentsandmatingpurposeswithAH109strain.

• For yeast two-hybrid screening, we used a Y2H librarymade from cDNAs derived from

RNAisolatedfromplantsstarvedforphosphate(Pugaetal.,2014)clonedinthepGADT7-

Rec vector (Clontech TM) following theMatchmakerprotocol PT3529-1 (Clontech TM) and

tranformedintotheY187strain.

31

1.3.Plantmaterial

• Arabidopsisthaliana(L.Heynh),accessionsColumbia-0(Col-0)andLandsbergerecta(Ler).

• Nicotianabenthamianafortransientexpressionassays(Table1)

• ArabidopsisthalianatransgeniclinesinCol-0background(Table1).

• ArabidopsisthalianaT-DNAmutantallelesinCol-0background(Table2).

• Themtv9-1 andmtv11-1mutantswere isolated from an ethylmethanesulfonate (EMS)

mutagenized population derived from the L1 transgenic line expressing the VAC2

transgene, which consists of the CLV3 protein (At2g27250) fused to the barley lectin

vacuolar sorting signal expressed under 35S constitutive promoter (Rojo et al., 2002),

stably transformed in a clv3-2 mutant in Ler background (Sanmartin et al., 2007).

Therefore, the original mtv mutants obtained by forward genetics are in Ler clv3-2

background.

• Arabidopsis thaliana subcellular marker lines: wave2R (mCherry-RabF2b/ARA7), wave2Y

(YFP-ARA7), wave13R (mCherry-VTI12), wave22R (mCherry-SYP32), wave7R (mCherry-

RHA1)(Geldneretal.,2009);SYP61-CFP(Drakakakietal.,2012);VHA1-RFP(Dettmeretal.,

2006),YFP-2xFYVE(Vermeeretal.,2006).

• Subcellularmarkersusedforprotoplasttransfectionandmicroscopyanalyses:SYP41-CFP,

mRFP-VSR2, ARA7Q69L-RFP, ARA7-RFP, mRFP-SYP61, ManI-RFP (Golgi Marker), mRFP-

SYP61, Aleurain-RFP, RFP-SCAMP1 and RFP-VIT1 were used when cargo sorting was

analyzed. These subcellularmarkerswere generated at Liwen Jiang´s lab using pBI221

gatewayTMcompatibledestinationvector.

32

1.4Plasmids

-pDONR207andpDONR221GatewayTMplasmidswereusedtogenerateentryclones(Invitrogen).

An integration reaction is carried out by homologous recombination between the PCR product

carrying attB flanks (recombination site for bacteriophage λ), attB1 (5’-

GGGGACAAGTTTGTACAAAAAAGCAGGCTNN-(sequence of interest)-3’) and attB2 (5’-

GGGGACCACTTTGTACAAGAAAGCTGGGTN-(sequence of interest)-3’) and the pDONR207 or

pDONR221containing theattP sites. This results in the formationof anentry clone that canbe

recombinedintodestinationvectorscarryingattRsites.Thepositiveclonesareselectedbasedon

itsresistancetotherespectiveantibioticmarkerpresentintheplasmidused(gentamycinincase

ofpDONR207andkanamycinincaseofpDONR221).

- pGADT7: GatewayTM-compatible destination vector used in the yeast two-hybrid (Y2H)

experiments(Chinietal.,2007).Theproteinof interest isfusedupstreamtoanHAepitopeand

downstreamoftheGAL4activationdomain(AD)undercontroloftheT7promoter.

- pGBKT7: GatewayTM-compatible destination vector used in the yeast two-hybrid (Y2H)

experiments(Chinietal.,2007).TheproteinofinterestisfuseddownstreamoftheGAL4binding

domain(BD)andupstreamtoac-mycepitopeunderthecontroloftheT7promoter.

- Binary plasmids (destination vectors): pGWB3, pGWB5, pGWB6, pGWB14 (Nakagawa et al.,

2007), andp-UBQ10-N-RFP (Grefenet al., 2010).pGWBsareGatewayTM-compatible vectors and

presenthygromycinandkanamycinresistantgenes.pGWB5andpGWB6allowtranslationalfusion

of a protein of interest with GFP at the C-terminus and the N-terminus, respectively. The

recombinantproteinexpressionisdrivenbythe35Sconstitutivepromoter.IncaseofpGWB3the

promoterofthegeneofinterestisfusedtotheβ-glucuronidasegene(GUSgene).Thedestination

vectorpGWB14carries the35Sconstitutivepromoter in framewith thegeneof interest tagged

withthreeconstitutiveHApolypeptides.

33

-pBI221modifiedplasmidwasusedforprotoplasttransformation(Gaoetal.,2014).Thisplasmid

carries an Ampicillin resistant gene and allows LR GatewayTM recombination reaction using

Gatewaycompatibleentryvectors(collaborationwithJinboShenatLiwenJiang´slab,CUHK).

- Plasmid for recombinant protein expression: pDEST17 (GatewayTM plasmid, Invitrogen). This

plasmid carries a T7 based promoter, recognized by the T7 polymerase. The T7 polymerase is

encoded by the BL21 (DE3) bacterial genome and its expression is inducible by IPTG. The BL21

(DE3)rosettastrainusedforexpressionisoptimizedforthecodonusageofeukaryoticgenes.The

resultingrecombinantprotein is taggedwithsixhistidinesat itsC-terminalpart toallowprotein

detectionandpurification.

-Plasmid for inducibleexpression inplants:pMDC7,aGatewayTM-compatibledestinationvector

for estradiol-induced expression of the protein of interest in planta (Curtis and Grossniklaus,

2003).

2.Culturemethods

• Bacterialculturemethods

ThedifferentEscherichiacolistrainswereincubatedovernightat37°CinLuriaBertani(LB)media

with appropriate antibiotics. LBmedia ismade of bacto-tryptone (10 g/l), yeast extract (5 g/l),

sodiumchloride(10g/l)andforcultureinsolidmedia,agarwasadded(3%,18g/l).Theworking

concentrations for thedifferentantibioticswere:ampicillin (100μg/mL),gentamicin (50μg/mL),

hygromycin(40μg/mL),kanamycin(50μg/mL),rifampicin(50μg/mL),spectinomycin(50μg/mL),

streptomycin(10μg/mL)andtetracycline(5μg/mL).

ForproteinexpressioninE.coli,coloniesweregrowninliquidLBuntilreachinganopticaldensity

of 0.6, induced with 0.4 mM IPTG, and aliquots were taken for protein extraction at different

times.

34

The strains ofAgrobacterium tumefaciens C58C1 were incubated during 2 days at 28°C in YEP

media (10 g bacto-peptone, 5 g of NaCl, and 10 g of yeast extract per liter) with appropriate

antibiotics.

In order to preserve E.coli andA. tumefaciens transformant lines during short periods of time,

strainswerekeptat4°CinsolidLBmediawithadequateantibiotics.Forlongerperiodsoftimethe

cellswereresuspendedina20%glycerolsolutionandkeptat–80°C.

• Yeastculturemethods

TheAH109andY187strainsofS.cerevisiaeweregrowninYPADandYPDmedia,respectively

(providedbyClontechTM).TheliquidYPADmediumisastandardrichYPDmedium(20g/L

peptone/tryptone,10g/Lyeastextract,20g/LglucosewithpHadjustedto5.8)supplemented

with40mg/Ladeninetoallowgrowthofstrainscarryingtheade2mutation.Thecorresponding

solidmediawassupplementedwith20g/Lofbactoagar(DuchefaTM).Transformationofboth

AH109andY187strainswithpGADT7andpGBKT7plasmidsledtonormalgrowthofyeastcellson

SDmedium.TheSDmedium (syntheticallydefinedmedium)isminimalmediathatisroutinely

usedforculturingS.cerevisiae.SDbasesupplieseverythingthatayeastcellneedstosurvive

35

(including20g/Lofglucoseand6,7g/Lofyeastextractascarbonsourcesandnitrogensources,

respectively)anditissupplementedwiththeaminoaciddrop-out(DO)mixanditspHadjustedto

5.8withKOH.Inaddition,20g/LbactoagarwereaddedfortheSDsolidmedia.

SD-WL: The SD media without tryptophan and leucine, prepared with the -Trp/-Leu DO

Supplement(Clontech,No.630417).

SD-WLH:TheSDmediawithouttryptophan,leucineandhistidine,preparedwiththe-Leu/-Trp/-

His/DOSupplement(Clontech,No.630419).

SD-WLA:TheSDmediawithout tryptophan, leucineandadenine,preparedwith the -Ade/-His/-

Leu/-TrpDOSupplement(Clontech,No.630428)andsupplementedwith0.002%histidine.

SD-WLHA:TheSDmediawithouttryptophan, leucine,histidineandadenine,preparedwiththe-

Ade/-His/-Leu/-TrpDOSupplement(Clontech,No.630428).

3-AT (3-amino-1,2,4-triazol): this compound is a competitive inhibitor for histidine biosynthesis

and it is added in case of bait autoactivation or to assess the strength of protein-protein

interactions in the yeast two hybrid assays, which use histidine auxotrophy as a marker for

protein-proteininteractions.

• Arabidopsisinvitroculture

Arabidopsis seeds were surface sterilized with a mixture of 70% ethanol and 0,05% Tween-20

during10minutes.Afterthistime,seedswerewashed3timeswithsterilizedwaterandwerecold

treatedforuniformgerminationduring3daysat4°Cindarkness.Seedsweregerminatedon½MS

media(MurashigeandSkoogmedia,DuchefaTM)supplementedwith1%sucrose,and0,7%agaror

1,2% for horizontal or vertical growth, respectively. The growth chamberswere set to long day

photoperiodconditions(16hoursoflightand8hoursofdarkness)witha60%relativehumidity.

36

• Assaysinarseniclimitingconditions

Seedsweresurfacesterilizedandcoldtreatedduring2daysat4°Cindarkness.Seedsweresown

onto 1% agar vertical plates containing low phosphate Johnson media (12,5 µM phosphate)

(Johnson CM, 1957) and grown for five days. Subsequently, seedlings were transferred to low

phosphateJohnsonmediasupplementedwithorwithout10or20µMarsenateandrootgrowth

inthenewplateswasmeasured3daysaftertransferringtheplants.

• CultivationofArabidopsisthalianainsoil

Tendaysoldseedlingsweretransplantedintoamixofsoilandvermiculite(ratio3:1).Plantswere

growninthegreenhouseat22°Cwithaphotoperiodof16hoursoflightand8hoursofdarkness

forlong-dayconditions,and8hoursoflightand16hoursofdarknessforshort-dayconditions.

3.Methodsforbacterial,yeastandplanttransformation

• Bacterialtransformation

Transformation of competent DB3.1 and DH5α E. coli strains was carried out by heat-shock

protocolsaspreviouslydescribed(Sambrook,1989).CompetentE.colicellswerepreparedusinga

calciumchloridetreatment(Hanahan,1985).TransformationofcompetentA.tumefaciensC58C1

strain was carried out as described in (Weigel, 2002). Competent Agrobacterium cells were

generatedusingafreezingmethodandcalciumchloride(Holstersetal.,1978).TransformedE.coli

andA.tumefaciensstrainswereplatedonselectivemedia(LBwithcorrespondingantibiotics)and

incubatedovernightat37°Cor48hoursat28°C,respectively.

37

• TransformationofArabidopsisthaliana

A. thaliana plants were grown in soil during 20-25 days in long-day conditions before their

transformation. Young inflorescences were immersed in a suspension of A. tumefaciens

transformedwith the construct of interest resuspended inMurashige and Skoog (MS)medium

supplemented with 5 % of sucrose, 0.02 % of Silwet L77 (a surfactant agent) and 50 mM of

benzylaminopurine(BAP,acytokinin)for10minutes(Bechtold,1993).Afterinmersionplantswere

setinatraycoveredwithplasticfor2daysandthentheplasticwasremovedandplantswerekept

inthegreenhouseuntilseedswereharvested.ThecollectedseedsweresowninMSmediumwith

thecorrespondingantibioticsforselectionoftheresistanttransformants.

• AgroinfiltrationofNicotianabenthamianaleaves

Three to four week old N. benthamiana plants were used for transient transformation

experiments. Bacterial suspensions of A. tumefaciens transformed with the corresponding

constructswere used for infiltration of the abaxial side ofN. benthamiana leaves as previously

described (Sparkes et al., 2006). To enhance the expression of a protein of interest, the leaves

wereco-transformedwithAgrobacteriumcarryingthepBin61-P19plasmidthatencodesthep19

proteinoftomatobushystuntvirus,asuppressorofpost-transcriptionalgenesilencing(Voinnet,

2003).After2or3daysofinfiltration,leaveswereanalyzedbyconfocalmicroscopy.

• Saccharomycescerevisiaetransformation

S. cerevisiae cellswere transformed as described inMatchmakerGal4 Two-HybridUserManual

(ClontechTM).Briefly,culturedyeastcellsareincubatedovernightandthenrefreshedbydilutionin

freshmediaandsubsequentgrowthuntilreachinga0.6-0.7ODvalue,followedbyapolyethylene

38

glycol(PEG)/LiAc-basedpreparationofcompetentyeastcells.Forplasmidtransformationweused

aheatshockbasedprotocolwhereyeastcellsareincubatedat30°Cfor30min,mixedwithDMSO

(5%volume)andincubatedfor15minat42°Cbeforeplatingontoselectivemedia.

4.Geneticinteractionsandphenotypicanalyses

• Arabidopsisthalianacrosses

Recipientflowersatstage13(whenthepetalswerebarelyvisible)wereemasculatedwithfine

forcepsandpollinizedbygentlybrushingthematureanthersofdonorplantontothestigma.

• Floweringtimeanalyses

To quantify the differences in flowering time between mtv9-2 and Col-0, three independent

experimentswerecarriedout.Rosetteandcaulineleaveswerecountedatthetimeofflowering,

consideredas the timewhenthe first floweredopenedandpetalswerevisible.A t-student test

wasusedforstatisticalcomparisonofthemeanfloweringtime(countedasnumberofleaves)in

mtv9-2andCol-0.

4.Nucleicacidanalysisandextraction

• ExtractionofplasmidDNAfrombacteria

Toisolateplasmids,weusedtheWizardPlusSVMiniprepsDNAPurificationSystem(PromegaTM).

• PlantDNAisolation

For standard PCR analyses, A. thaliana genomic DNA was extracted following an isopropanol-

based method described previously (Doyle, 1990). DNA samples intended for whole genome

sequencingwere prepared using the plant DNeasy kit (QIAGENTM). Thismethod ensures longer

39

DNA fragments and allows successful library preparation for the next generation sequencing

(NGS).Incaseofmtv9-1,onegramofyoungleavescollectedfromtwentyfiveplantsshowingthe

mtvphenotype(meristemterminated)wereusedforDNAisolationandtheNGS.

• Gel/PCRDNAfragmentsextractionkit

DNAfragmentsseparatedbyagarosegelelectrophoresisorDNAamplifiedinapolymerasechain

reactions(PCR),werepurifiedusingQIAquickGelExtractionKit(QIAGEN).

• ExtractionofplasmidDNAfromyeast

To isolate plasmids from yeast, we utilized previously described methodology (Hoffman and

Winston,1987).Forsequencingtheplasmids,wereretransformedthemintothestandardDH5α

E.colistrainandsubsequentlyisolatedbyamethoddescribedinthesection4.1.

• AmplificationofDNAfragments

Taq polymerase (Roche)was used for standardPCR amplifications and thehigh-fidelity Phusion

DNA polymerase (Roche) for cloning purposes. Primers used for PCR amplifications throughout

thisworkarelistedintheTables2and3.

40

41

42

• Gelelectrophoresisofnucleicacids

DNA fragments obtained by PCR or the fragment analysis derived from plasmid digestionwere

visualisedbyagarosegelelectrophoresisusingethidiumbromidein1xTBEbuffer(50mMTris,1

mMEDTA,pH8).

• SequenceanalysisofDNA

LibrarypreparationandtheNGSsequencingoftheDNAfromthepoolofmtv9-1plantswasdone

by the Genome Center at the Max Planck Institute for Developmental Biology. Reads were

mappedontotheTAIR10genomeandaregionthathadonlyreadsfromLandsbergaccesionwas

identified.Within this region, EMS type polymorphisms (G/A or C/T) producing changes in the

open reading framewere selected.For sequencealignmentofMTV9andSTV9 theGENOMATIX

programalignmenttoolwasused(http://www.genomatix.de/cgi-bin/dialign/dialign.pl).

5.Proteinanalyses

• SDS-PAGEproteinanalyses

ArabidopsisthalianaandNicotianabenthamianaplantsampleswerefrozeninliquidnitrogenand

homogenized in Laemmli loading buffer (Laemmli, 1970). The samples were boiled during ten

minutes and after a short centrifugation to pellet cell debris theywere loaded on the gels. To

visualizeseedstorageproteins,gelswerestainedwithCoomassieBrilliantBlue.Forwesternblot

analysis, proteins were transferred onto ECL-nitrocellulose membranes (GE Healthcare Life

Sciences).Themembraneswerefirstincubatedinblockingsolution(5%non-fatmilkinTBS+0.1%

Tween-20)toavoidunspecificproteinbinding,andthenincubatedwithprimaryantibodiesdiluted

in 1% non-fat milk in TBS+0.1% Tween-20, washed five times in TBS+0.1% Tween-20 buffer,

incubatedwithsecondaryantibodiesfusedtohorseradishperoxidasedilutedin1%non-fatmilkin

43

TBS+0.1% Tween-20, washed five times in TBS+0.1% Tween-20 buffer, incubated with ECL

reagentsandexposedonfilmsfordetectionoftheproteins.

• ExpressionofaMTV9fragmentforantigenproduction

Weexpressed aN-terminal fragment of theMTV9protein (N-terCORE) inE.coli. Analysis of the

distribution in soluble and insoluble fractions showed that the protein was mostly found in

inclusion bodies. 50 mL of a bacterial culture from a colony that overexpressed the MTV9

fragmentwereresuspendedin5mL50mMTRISbuffersolutioncontaining2mMEDTAatpH=8.2

andlysedbyultrasoundwithasonicatingprobe,centrifugedat13,000rpmfor15minutesat4°C

and the resultingpelletwas resuspended inLaemmli loadingbufferandsubjected toSDS-PAGE.

ThegelbandcontainingtherecombinantMTV9fragmentwascutandusedtoimmunizerabbits.

• Antibodyproductionandpurification

OncewewereabletoinduceasufficientamountoftheN-terCOREMTV9fragment,wesentthe

gel band corresponding to this fragment for antibody production. Five different rabbits from

PinedaAntikörperService(http://www.pineda-abservice.de/main.php)werepretestedforlackof

cross-reaction with Arabidopsis proteins. After initial immunization and three boost shots we

obtained a serum that recognized the endogenousMTV9 protein and was utilized for western

blotting analyses. We further purified the serum by binding to MTV9 protein transferred into

nitrocellulosemembranes and used it for immunoelectronmicroscopy analyses. Briefly, the N-

terCORE fragmentwas run and transferred to anECLnitrocellulosemembraneand thebandof

interestwascutout.A0.1MglycinepH=2,5solutionwasusedtoremovepoorlyboundproteins

during 5 min and the membrane was washed two times with 1xTBS 0,1% Tween buffer. The

membranewasblockedwith3%BSA1xTBSsolutionduring1hour,andafterwashing500µLof

44

MTV9 antibodywere allowed to bind themembrane during 3 hours at room temperature. The

supernatantwassavedandthemembranewas incubatedwith500µLof theglycinesolutionto

elutetheMTV9purifiedantibody.After10mintheglycinesolutionwasneutralizedwith23µLof

TRIS1MpH=8andthiswasthesampleusedforMTV9immunoelectronmicroscopyanalyses.

• AnalysesofArabidopsisapoplasticfluid

ToisolatetheproteincontentoftheArabidopsisapoplast,onegramof5weekoldrosetteleaves

grown in short day conditions was used. These leaves were vacuum infiltrated with 50 mL of

sodiumphosphatebuffer,pH8.0,containing150mMNaClfor15min,breakingdownthevacuum

every3min.After infiltration, leaveswerecarefullydriedwithatissuepaperandtheapoplastic

fluidwascollectedbylow-speedcentrifugation(900xg).Theisolatedapoplasticfluidwasmixed

withLaemmliloadingbufferandprocessedasdescribedinthesection5.1.

• Sucrosedensitygradients

Toanalyzelocalizationofvariousproteinmarkerstodistinctendomembranepopulations,protein

samplesfromflowertissuesor10daysoldseedlingswereseparatedonsucrosedensitygradients.

The samples (≈1-1.5 g) were lysed in 8 ml of 50 mM Hepes-KOH, pH 6.5/5 mM EDTA/13.7%

(wt/vol) sucrose/0.1 mM phenylmethylsulfonyl fluoride/1 mM DTT/ 1x Complete Protease

Inhibitor Cocktail (Roche), centrifuged (1,000g and 4 °C for 10 min) and the supernatant was

collected. 1.5 mL of the supernatant were layered on top of the 9-mL line step gradient, as

previouslydescribed(Sanderfootetal.,1998).Gradientswerecentrifugedfor2hat100,000xgin

aBeckmanSW40Tirotorat4°C.Subsequently,0.5mLfractionswerecollectedfromthetopofthe

tubeandtheproteinswereprecipitatedin10%trichloroaceticacid(TCA),washedin90%acetone

45

and dried. The protein pellets were resuspended in the Laemmli loading buffer, processed as

described in the SDS-PAGE analysis section, subjected to electrophoresis followed by

immunoblotting.

• DifferentialcentrifugationofArabidopsismicrosomalfractions

Arabidopsis seedlings were grown in the liquidMS for 10 days under long-day conditions. The

plants (≈1-1.5 g)were carefully driedwith a tissuepaper andhomogenized in extractionbuffer

(100mMTris-HCl, pH7.5, 400mMsucrose, 1mMEDTA, 0.1mMPMSF, 1XCompleteProtease

InhibitorCocktail) usingamortar andpestle, followedby centrifugationat1,000xg for15min

resulting inapellet(P1)andacorrespondingsupernatant(S1)(BasshamandRaikhel,1998).The

S1supernatantwascentrifugedat3,000xgfor15mintoproduceamembranepellet(P3).TheS3

supernantant was subsequently centrifuged at 16,000 x g for 15 min to produce a membrane

pellet (P16). 100,000 x g for 30 min to yield a membrane pellet (P100) and a soluble fraction

(S100).PelletsandsupernatantswereresuspendedintheLaemmliextractionbufferandanalyzed

bySDS-PAGEandimmunoblotting.

• ExtractionofMTV9frommembranes

1-1.5gramsofArabidopsisseedlingswereprocessedasdescribedintheprevioussectiontoyield

theS1fraction,whichwasdirectlysubjectedto100,000xgultracentrifugation.TheisolatedP100

membrane pellets were resuspended in 200 µL of the HE extraction buffer containing 0.1 M

Na2CO3,1MNaCl,2Murea,or1%(v/v)TritonX-100,respectively,andincubatedfor2honice.

Sampleswere subjected toa secondultracentrifugationat100,000xgand the resultingpellets

wereresuspendedintheLaemmli loadingbuffer.Supernatantswereprecipitatedusing10%TCA

andalsoresuspendedintheLaemmliloadingbufferandprocessedasinthe5.1.section.Samples

46

weresubsequentlyanalyzedbySDS-PAGEandimmunoblotting.

• Co-immunoprecipitationanalyses

MicrosomesobtainedfromArabidopsisseedlingswerehomogenizedin500μLofextractionbuffer

(100mMTris-HCl, pH7.5, 400mMsucrose, 1mMEDTA, 0.1mMPMSF, 1XCompleteProtease

Inhibitor Cocktail), 0.2%NP-40,and 1% bovine serum albumin (BSA). Extractswere centrifuged

twiceduring1minuteat13,000 rpmat4°C. Supernatantsweremixedwith15μLofanti-MTV9

antibodyandincubatedfortwohoursat4°C.Inanewtesttubeinwhich150μLofproteinAresin

(Roche)wereaddedandtheproteinextractwasgentlyshakenfor2hoursat4°C.Subsequently,

thetubeswerecentrifugedat13,000rpmfor30seconds.Theresinwaswashedthreetimeswith

theextractionbuffersupplementedwith1%BSA,andalsowithHEbufferlackingBSA.Afterwards,

theresinwasresuspendedintheLaemmliloadingbufferandboiledfor10minutes.Sampleswere

analyzedbySDS-PAGEandimmunoblotting.

• Antibodies

Different primary antibodies were used and these were diluted 1:1000 but express indication:

anti-GFP,anti-MTV9,anti-Aleurain(Ahmedetal.,2000);anti-CPY(Rojoetal.,2003);anti-SYP21(da

Silva Conceicao et al., 1997), anti-VSR(Sohn et al., 2003), anti-VTI11, anti-VTI12 (Surpin et al.,

2003), anti-SEC12 (Bar-Peled and Raikhel, 1997) , anti-PR5 (Palacin et al., 2010), anti-VacPerox

(Sanmartin et al., 2007), anti-TGG1 (Ueda et al., 2006), anti-BiP (Santa Cruz Biotechnology sc-

33757). Secondary antibodies fused to horseradish peroxidase were used at 1:10000 dilutions

(anti-rabbit,anti-goat,anti-mouse).

47

6.Microscopictechniques

6.1.Confocalmicroscopy

Confocal images were acquired by Leica TCS SP2 and Leica TCS SP5 multispectral confocal

microscopes (LeicaMicrosystems)witha63xwater-immersionobjective, recording images from

laser lines of 405, 488, and 561 nmwavelengths. LAS AF v.2.3.6 software was used for image

compilation.Imageprocessingincludingcolocalizationofsinglecolorprotoplastimagesweredone

with ImageJ and Photoshop (http://www.macbiophotonics.ca/imagej/). A. thaliana seedlings

visualized inconfocal imagingweregrownfor3-6daysat22°Cwitha60%ofhumidityand16-

hours photoperiod with a fluorescent light of a 100 μmol m-2s-1 intensity. We visualized root

epidermalcellsinA.thalianaandepidermalcellsoftheabaxialsideofleavesinN.benthamiana

plants.

6.2.Chemicaltreatments

A.thalianaseedlingsandcultureprotoplastsweretreatedchemicallywithBFAandwortmanninto

checkpossibledefectsintheendomembranesystemandincargotrafficking.BFAandwortmannin

weredilutedinDMSOat50µMand33µM,respectively.Seedlingswereincubatedfor2hoursin

theseconcentrationsbeforemicroscopeexamination.

48

Results

49

1.Identificationofnovelmtvmutants

SeedsfromtheL1transgenic linecontainingasingleVAC2insertioninchromosome3(17.1Mb)

weremutagenizedwithEMSandsownin48traysthatwerecollectedseparatelyin48poolsofM2

seeds.M2plantswere screened formutants showing prematurely terminated shoot apical and

flowermeristems. Inapartialscreenofsomeofthepools,eightmutants(mtv1-mtv4andmtv8-

mtv11) with prematurely terminated vegetative, inflorescence and flower meristems (mtv

phenotype)wereisolated.Usingthissamestrategy,threeothermutants(mtv5-mtv7)havebeen

identifiedandcharacterizedinthelaboratoryofDr.NatashaRaikhel(Rosadoetal.,2011;Sohnet

al., 2007). The mutations responsible for the mtv phenotype in four of these mutants were

previously identified: mtv1/At3g16270 (pool 40); mtv2/vsr4/At2g14720 (pool 43);

mtv3/pten2a/At3g19420(pool37)andmtv4/agd5/At5g54310(pool46)(Saueretal.,2013;Zouhar

et al., 2010); PhD Thesis of María Otilia Delgadillo). In this work, we characterized two other

mutants from that initial screen, mtv9 (pool 31) and mtv11 (pool 32) and identified the

corresponding genes through map-based cloning. Recently, in the lab we have carried out a

saturatingscreenofapproximately2000M2plantsfromeachofthepoolsandisolated121new

putativemutantswithastrongmtvphenotype.Wehaveidentifiedbynextgenerationsequencing

thecausativemutationof20ofthesemtvmutants,andfoundanadditionalalleleofmtv9 (pool

48).

2.MAP-basedcloningofmtv9

Themtv9-1mutantwas isolated frompool 31. The F1plants froma cross between themtv9-1

mutantandCol-0hadawild typephenotype.Moreover, theproportionofplantswitha strong

mtv phenotype in the F2mapping population (157 out of 2011 plants)was consistentwith the

expected ratio fora recessivemutationunlinked to theVAC2 transgeneproducinga strongmtv

50

phenotypewhen both themutation and the VAC2 transgene were in homozygosis. Hence, for

rough mapping of themtv9-1 mutation, we performed bulk-segregant analysis with F2 plants

displayingthemtvphenotype.WeisolatedDNAfromapoolof49plantswithmtvphenotypeand

scanned the five chromosomes for regions showinganenrichment in Ler SSLPs. In thisway,we

found that the AthZFPG SSLP marker (chromosome I) was highly enriched for the Ler

polymorphisminthepoolofmutantDNA.Genotypingofthe49individualmutantplantsrevealed

that 97of the98 chromosomeshad theAthZFPG Ler SSLP, indicating close linkage to themtv9

mutation.Moreover,theanalysisoftheseplantswithmarkersaroundAthZFPGdelimiteda1Mb

region containing the mutation. We sequenced the candidate genes in that region

VTI12/At1g26670, ISTL2/At1g25420,andTNO1/At1g24460 thatencodeproteinswithpreviously

reportedrolesinvesiculartrafficking(KimandBassham,2011),butnomutationswereidentified

in those genes. Finemapping restricted thepositionofMTV9 to 0.1Mb interval containing 34-

genes (At1g24320–At1g24640) between the BOF316 marker and the AthZPFG SSLP markers

(Figure 1A). However, no obvious candidates were apparent in that interval, so we performed

deep sequencing of a DNA sample from a pool of 30 plants withmtv phenotype from the F2

mappingpopulation(Figure1B). Inthe intervaldefinedthroughfine-mapping,wefoundasingle

EMS(G/A)mutationthatgeneratedaprematurestopcodoninthefirstexonofAt1g24560,agene

ofunknownfunction(Figure5A).ThisEMSmutantallelewasnamedmtv9-1.

51

Figure1.mtv9 identificationthroughmap-basedcloning. (A)mtv9mappedtochromosome1 inan intervalbetween

BOF316(8,63Mb;3recombinantchromosomes)andAthZPG(8,73Mb;1recombinantchromosome)SSLPmarkers.(B)

TheDNAfromapoolof30plantswithmtvphenotypewasdeepsequencedandthepercentageofreadswithCol-0or

LerSNPsalongchromosome1 isshownastheblack line.Thered linemarksequalpercentageofLerandCol-0SNPs,

abovetheredlinemarksenrichmentinCol-0SNPsandbelowthelineenrichmentinLerSNPs.Theregionencircledhas

readswith100%LerSNPs.

In order to confirm that the mutation of At1g24560 is responsible for the observed mtv

phenotype,twoadditionalT-DNAmutantalleleswereanalyzed.Themtv9-2allele(SAIL_24_C10)

has a T-DNA insertion in the first exon of the At1g24560 gene and the mtv9-3 mutant

(SAIL_670_H06) has a T-DNA insertion in the fourth intron (Figure 5A). Plants homozygous for

these T-DNA insertions were isolated and crossed with VAC2 line. The analysis of the two F2

populations showed thatplantshomozygous formtv9-2 ormtv9-3 andat leastone copyof the

VAC2 transgenehadterminatedmeristems, reproducingthephenotypeofmtv9-1plants (Figure

5B).Moreover,we identified an additionalmtv9 allele in anmtvmutant isolated frompool 48,

whichcontainedanEMSmutationthatdisruptedthespliceacceptorsiteofthethirdintronofthe

MTV9 gene (mtv9-4 allele). Altogether, the analysis of these four independent alleles strongly

supports thatMTV9 is required forvacuolar traffickingofVAC2andthateliminating its function

resultsinVAC2secretionandconsequentprematuremeristemtermination.

52

Figure5. Independentmutantalleles forAt1g24560showmtvphenotype in thepresenceofVAC2. (A)Aschematic

representationoftheMTV9genelocus,showingthepositionsofthefourisolatedmtv9mutantalleles.Themtv9-1and

mtv9-4areEMS-inducedpointmutantswhilethemtv9-2andmtv9-3areT-DNAinsertionalmutants.(B)Confirmationof

meristemterminationinmtv9-2andmtv9-3mutantsexpressingtheVAC2transgene.

3.MTV9isaplantspecificgenewithaputativecoiled-coildomain.

Asearchforhomologousproteinsthroughsequencealignmentsearchtools(Altschuletal.,1990;

GishandStates,1993)showedthattherearenohomologuesofMTV9presentoutsidetheplant

kingdom. In contrast, close homologues of MTV9 were found in all land plants, including in

Selaginella moellendorffii, Physcomitrella patens andMarchantia polymorpha. The Arabidopsis

genomehasalsoaparalogousgene(At3g49055)thatwenamedSTV9. Interestingly,seedplants

contain bothMTV9 and STV9 paralogues whereas non-seed plants like S. moellendorffii and P.

patensonlycontaintheMTV9paralogue(Figure6).

53

Figure6.PhylogeneticanalysisofMTV9homologuesinArabidopsis,rice,PhyscomitrellaandSelaginella.

The Arabidopsis STV9 protein has 27% overall identity (46% similarity) withMTV9, and

homology is maximal in the C-terminal half (Figure 7). We analyzed whether STV9 was also

involved in traffickingofVAC2bycrossing the transgene intoT-DNAknockoutmutants forSTV9

gene (SAIL_193_D05 and SALK_029261).However, thehomozygous stv9mutant did not display

meristemterminationinthepresenceoftheVAC2transgene,indicatingthatSTV9hasadifferent

functionthanMTV9.

54

Figure7.AlignmentofArabidopsisMTV9andSTV9proteinsequences.Redcolorindicatessequenceidentityandblue

colorindicatessequencesimilarity.

Structural sequence analysis showed thatMTV9 has no previously described functional

domain (Wang et al., 2007). However, the Pfam server predicted MTV9 to contain coiled-coil

55

domainsalongitssequence(Finnetal.,2016)(Figure8).Thisisconsistentwithapriorprediction

ofMTV9beingacoiled-coilprotein(Gardineretal.,2011).

Figure8.MTV9ispredictedtocontaincoiled-coildomains.Predictedcoiled-coildomainsinMTV9aremarkedasgreen

rectangles.

Coiled-coil domains are present in homodimeric membrane tethering factors (Chia and

Gleeson, 2014), involved in intracellular trafficking throughout eukaryotes. To analyze whether

MTV9 could also homodimerize, we tested for interaction by the yeast two-hybrid assay. This

assayshowedthatMTV9caninteractwithitself(Figure9)andmaythushomodimerizeinvivo.

Figure 9. MTV9 interacts with itself in yeast two hybrid assays. pGKB7-MTV9 was cotransformed into AH109

S.cerevisaestrainwithpGADemptyvectorandpGAD-MTV9.Thecontransformantswereplatedonsyntheticselection

mediatoanalyzetheproteininteractionanditsstrength.

56

4.MTV9expressioninRNA-seqdatasets

TheATH1microarrayplatformcommonlyusedfortranscriptomicanalysisinArabidopsisdoesnot

containaprobe forMTV9,butwecouldanalyze itsexpression inpublicRNA-seqdatasetsusing

the GENEVESTIGATOR server. MTV9 is a medium to highly expressed gene in most tissues.

Interestingly, gene co-expression analysis showed thatMTV9 is highly co-expressedwith genes

involved or potentially related with intracellular trafficking: among the 25 genes most co-

expressed withMTV9 we found VCL1/VPS16 (Rojo et al., 2001), VPS39, the golgin GC5, ALIX

(Cardona-Lopez et al., 2015), Dynamin2B, and four proteins with phosphatidylinositol binding

domains.

5.MTV9promoteractivity

To study theMTV9 expression profile with cellular resolution, we studied the activity of two

promoter constructs (500 bp or 1200 bp fragments upstream from the start codon) driving

expression of the GUS reporter gene (Figure 10). We analyzed several independent transgenic

linesforeachpromoterconstructandtheyshowedsimilarGUSexpressionpatterns.GUSstaining

wasfirstdetectableinfloraltissues,suggestingthatMTV9expressionismaximalinthosetissues,

inparticularinstamensandinthepetalabscissionzone.Inyoungseedlings,weobservedstrong

GUS expression in the hypocotyl. Transversal cross section of the stem showed that MTV9 is

expressedintheepidermis,thecortexandthevascularbundles(Figure10D).Inrootsandmature

leaves,MTV9expressionisalsodetectableinthevasculatureandin leaftrichomes(Fig.7G),but

not inroothairs.Wealsoobservedexpression in theshootapicalmeristem,which isconsistent

withthemutationaffectingtraffickingofVAC2thereandresultinginmeristemtermination(Figure

10I). In contrast, the expression in root apicalmeristemswas very low. Intriguingly,MTV9was

57

expressed specifically in the quiescent center (QC) (Figure 10L), suggesting that those cellsmay

have special requirements for this gene. We also investigated if treatments with hormones

(auxins, cytokinins, brassinosteroids, gibberellins and jasmonates) or with specific inhibitors of

trafficking (wortmanninandBFA)wouldalterMTV9expression.However,wecouldnotobserve

anysignificantchangescomparedtomocktreatedplants.

Figure10.TissueexpressionoftheProMTV9:GUStransgene.(A)flowermeristem.(B)flowertissue.(C)Gynoeciumwith

pollengrainsonitspistil. (D)Transversalcutofthestem.(E)10days-oldseedlingshowingMTV9expressionmainly in

thehypocotyl.(F)CotyledonshowingMTV9expressioninthevasculartissue.(G)TrichomesinadultleafexpressMTV9.

(H) Root-hypocotyl junction showing expression at the hypocotyl and the vascular tissue of the root. (I) Shoot apical

meristem ina10days-oldseedling. (J)RootshowingMTV9expression in thevascular tissue. (K)15days-oldseedling

showingexpressionintheolderpartsoftheroot.(L)MTV9expressionatthequiescentcenter(QC,arrowhead).

58

6.MTV9localizesprimarilyatthePVC

MTV9may be recruited to the one of the compartments of the vacuolar pathway to exert its

function in vesicular trafficking. Todetermine the subcellular localizationofMTV9,weanalyzed

various versions of the protein fused to theGFP.We cloned theMTV9 coding sequence in the

pGWB5andpGWB6vectorstoproduceinframefusionswithGFPsequenceattheN-terminusand

the C-terminus ofMTV9 expressed under the control of the constitutive 35S promoter (Figure

11A).TransientexpressioninNicotianabenthamianaleavesshowedthatbothversionslocalizedin

a punctate pattern throughout the cytosol. Moreover, transfection of Arabidopsis protoplasts

rendereda similar localizationpattern (Figure11B). These findings suggest that, firstly,MTV9 is

associated with an endomembrane compartment, and secondly, that the localization is not

affectedbyatranslationalfusiontotheGFP,asbothN-terminalandC-terminalversionsshowed

thesameapparentdistribution.

59

Figure11.MTV9showsapunctatepatternintransientlytransformedArabidopsisprotoplastsandNicotianacells.(A)

Nicotiana benthamiana pavement cells agroinfiltrated with 35S-GFP-MTV9 (left) and 35S:MTV9-GFP (right). (B)

Arabidopsisleafcellprotoplaststransfectedwith35S:MTV9-YFP(left)and35S:GFP-MTV9(right).Scalebars:10µm.

Todeterminewhatcompartment(s)correspondedto theobservedpunctatepattern,we

studiedtheco-localizationofGFP-MTV9withmarkersofdifferentendomembranecompartments,

incollaborationwiththegroupofDr.LiwenJiangfromtheChineseUniversityofHongKong.We

co-transformed well-established markers developed in Dr. Jiang’s lab with GFP-MTV9 in

Arabidopsisprotoplasts.Weobservedcompleteco-localizationofGFP-MTV9withthePVC-marker

VSR2-RFP (Figure 12A), indicating that MTV9 is localized at steady state primarily there. In

protoplastswithmoderateexpressionofGFP-MTV9,theGFPsignalwascompletelyseparatefrom

the punctate signal observed for the TGNmarkers SYP41-CFP, and theGolgimarkersManI-RFP

(Figure12Aand13),consistentwithMTV9beinglocalizedinthePVC.Interestingly,weobserved

that in protoplasts overexpressing GFP-MTV9, the RFP-VSR2 signal concentrated in larger dots

(Figure 12B). In order to discard possible effects of the GFP fusion and to confirm that MTV9

overexpression is responsible for thisapparentPVCenlargement,MTV9 fused toa smallHA-tag

was co-transformedwith RFP-VSR2. This experiment also showed accumulation of RFP-VSR2 in

enlargeddots,confirmingthattheeffectoftheMTV9overexpression is independentoftheGFP

fusionattheN-terminus.

60

-

Figure12.MTV9localizesatthePVC.(A)Protoplastswereco-transformedwithGFP-MTV9andRFP-VSR2(PVCmarker),

ManI-RFP(Golgimarker)orSYP41-CFP(TGNmarker).(B)AprotoplastshowinghighlevelsofGFP-MTV9expressionand

PVCaggregation.Scalebars,10µm.

To complement the data from the transient expression systems, we generated stably

transformedArabidopsis plants expressing35S-GFP-MTV9 (pGWB6plasmid). In these lines,GFP

61

fluorescencewasdistributed in a punctate pattern, similar towhatweobserved in protoplasts.

Many of the lines showed transgene silencing and unfortunately, lines showing high expression

were not identified, so the effect of overexpression could not be assessed. For subcellular

localizationstudieswecrossedinmarkersfordifferentcompartments(Geldneretal.,2009).The

PVC markers RHA1-RFP (wave7 line) and ARA7-RFP (wave2 line) co-localized largely with GFP-

MTV9 (Figure 13B,C).Moreover, no co-localizationwas obviouswith the TGNmarker Vha1-RFP

(Figure13A).

62

Figure13.MTV9overexpressionaggregatesthePVCinArabidopsisstabletransformedplants.(A)F1plantsexpressing

GFP-MTV9 and the TGNmarker Vha1-RFP. (B) F1 plants expressingGFP-MTV9 and the PVCmarker Rha1-RFP. (C) F1

plantsexpressingPVCmarkersARA7-YFPandRHA1-RFP.(D),F1plantsexpressingGFP-MTV9andthePVCmarkerARA7-

RFP.Scalebarscorrespondto2,5µmand10µminthecaseofDpanel.

63

Together, theresults fromtransientexpression inArabidopsisprotoplastsor fromstable

expressionintransgenicplantsarecoincidentandsupportthatMTV9associateswiththePVC.

7.TheconservedC-terminaldomainofMTV9isresponsiblefortargetingtothePVC

Analysis of theMTV9 primary sequence in the TMHMM and TMPRED servers for prediction of

transmembranedomainsandintheSignalP4.1Serverforpredictionofsignalpeptides(Petersen

etal.,2011) suggest thatMTV9proteindoesnotcontaina signalpeptidenora transmembrane

domainforassociationwithmembranes.To identifythesequencesresponsible forthetargeting

to the PVCwedissected theMTV9 coding sequence in three non-overlapping constructs, fused

them toGFP (Figure14A) andanalyzed their localization inN.benthamiana leaves (Figure14B)

andinArabidopsisprotoplasts(Figure15).ThefragmentsanalyzedwereNT(aminoacids1-199),

CORE(aminoacids200-472)andCT(aminoacids473-678).ThelocalizationofNT-GFP,CORE-GFP

andNT-CORE-GFPwascytosolic,andnoassociationwithmembranecompartmentswasapparent.

Incontrast,theCORE-CT-GFPandtheCT-GFPconstructspresentedapunctatepatternsimilarto

the full length MTV9 (Figure 14B). Similar results were obtained in Arabidopsis protoplasts

transformedwiththeseconstructs(Figure15).TheseresultssuggestthattheCTfragmentcontains

thenecessary information for targetingMTV9 to the PVC. TheCT is themost conserved region

among MTV9 homologues and the STV9 paralogue, which also localizes to endosomes (Figure

14C),indicatingthattheCTisinvolvedinmembranerecruitmentandthatthisisessentialforthe

functionoftheseproteins.

64

Figure 14. The CT domain is necessary for correct localization ofMTV9. (A), different deletion constructs ofMTV9

clonedintothepGWB6vector.(B),GFP-localizationofthedifferentdeletionformsofMTV9inN.benthamianaleaves.

(C), i) STV9 construct that carriesRFP fused to STV9 cDNAunder the control of the constitutivepromoterUBQ10, ii)

localizationofRFP-STV9inN.benthamianaleafpavementcells.Scalebarscorrespondto10µmand2,5µminthecase

ofSTV9picturemagnification.

65

Figure15.TheCTdomaindirectslocalizationatthePVC.(A),N-terminalpartofMTV9co-transformedwithRFP-VSR2.

(B), N-terminal part and CORE domain of MTV9 co-transformed with RFP-VSR2. (C), CORE domain of MTV9 co-

transformed with RFP-VSR2. (D), CORE domain and C-terminal part of MTV9 co-transformed with RFP-VSR2. (E), C-

terminalpartofMTV9co-transformedwithRFP-VSR2.Scalebars,10µm.

66

8.MTV9overexpressionperturbstransportofvacuolarcargobutnotofPMproteins

Asnotedabove,MTV9overexpressioninprotoplastscausedtheRFP-VSR2signaltoconcentratein

largerdots,whereasmoderateexpressiondidnot cause this apparentPVCenlargement (Figure

12). Co-transformation experiments with the TGN syntaxin markers RFP-SYP61 and SYP41-CFP

showed that in protoplast expressingmoderate amounts ofGFP-MTV9, theGFP-MTV9 signal at

the PVC was separated from RFP-SYP61 and SYP41-CFP signal at the TGN (Figure 16A).

Interestinglyhowever,whenGFP-MTV9wasstronglyoverexpressed,SYP61butnotSYP41wasre-

localizedintheenlargedGFP-MTV9positivecompartments(Figure16B).

67

-

Figure 16.MTV9overexpression leads to SYP61mislocalization. (A). Protoplasts expressingmoderate levels ofGFP-

MTV9co-transformedwithSYP41-CFPorRFP-SYP61.(B),ProtoplastsexpressinghighlevelsofGFP-MTV9co-transformed

withSYP41-CFPorRFP-SYP61.Scalebars,10µm.

We reasoned that PVC enlargement and SNARE mislocalization caused by MTV9

overexpression should interfere with the functionality of the compartment and with protein

traffickingtothevacuole.Totestthis,weco-transformedArabidopsisprotoplastswithGFP-MTV9

68

andasolublevacuolarcargo(Aleurain-RFP),atonoplastmembranevacuolarcargo(RFP-VIT1)and

aplasmamembranecargo(RFP-SCAMP1).AleurainisathiolproteasewithanN-terminalvacuolar

sorting signal that has beenwidely used as amodel soluble vacuolar cargo.When transformed

into Arabidopsis protoplast, it labels the lumen of the vacuole.When it is co-transformedwith

GFP-MTV9,Aleurain-RFPalsoreachesthevacuolebutthereisalsosomeretentionintheenlarged

PVCs together with MTV9-GFP (Figure 17), suggesting that trafficking to the vacuole is partly

disrupted.Solubleandmembranecargoesmayusedifferent routes to reach thevacuole, sowe

analyzedhow transport of the tonoplastmembranemarkerwas affectedby co-expressionwith

MTV9-GFP. Interestingly, RFP-VIT1 was mostly found in the enlarged PVCs together with GFP-

MTV9(Figure17)andalmostnoneoftheproteinreachedthetonoplast.Hence,theinterference

with targeting of thismembrane cargowas higher thanwith aleurain, possibly because soluble

cargocanentervesiclesfromalternativevacuolartransportpathways.Totestiftheeffectswere

specific to vacuolar proteins, we co-transformed the PMmarker RFP-SCAMP1 withMTV9-GFP.

Importantly,SCAMP1-RFPwasproperlytargetedtothePM,indicatingthatonlytraffickingtothe

vacuolewasdisrupted.ThisisconsistentwithMTV9overexpressionaffectingthefunctionalityof

thePVC,alateendosomespecificforthevacuolartransportroute.

69

Figure17.TraffickingtothevacuoleisdisruptedwhenMTV9isoverexpressed.ArabidopsisprotoplastsexpressingGFP-

MTV9andthevacuolarsolublecargoAleurain-RFP,theplasmamembraneproteinRFP-SCAMPorthetonoplastprotein

RFP-VIT1.Scalebars,10µm.

9.PVCaggregationbyMTV9overexpressionisnotaffectedbywortmannin

Wortmannin,aPI3kinaseinhibitor,causeshomotypicfusionofPVCsleadingtotheformationof

greatlyenlargedPVCsthatappearassphericalcompartmentswithalimitingmembranethatcan

beresolvedfromthelumeninalightmicroscope(Figure18A).Thisisdifferentfromtheenlarged

PVCsformedinMTV9-GFPoverexpressingprotoplasts,whichhaveamoreamorphousshapeand

thelumenisnotdistinguished.ThissuggeststhatthemechanismcausingPVCenlargementinthe

MTV9-GFP overexpressing protoplasts is different from that of wortmannin and may instead

reflect an aggregation of non-fusing PVCs. This process of aggregation of intracellular

70

compartments in certain cytosolic domains is also observed in plants treated with BFA, which

induces the formation of the so-called BFA bodies that are large aggregates of Golgi and TGN

cisternaeinthecytosol.GiventhatwortmanninandMTV9overexpressionwerebothaffectingthe

PVC,weanalyzedtheeffectoftheircombinedaction.Interestingly,wortmannintreatmenthadno

effectonPVCmorphologyinprotoplasts(Figure18A)orintransgeniclinesoverexpressingMTV9-

GFP(Figure19),indicatingthattheMTV9-inducedaggregationofthePVCpreventedwortmannin-

induced homotypic fusion. To confirm this, we co-transformed protoplast with a constitutively

active formof ARA7(Q69L) that also causes homotypic fusion of PVCs andmimics the effect of

wortmannin treatment (Jia et al., 2013). Protoplasts transformedwith ARA7(Q69L)-RFP induced

the formationof large ring–likePVCs,but thiseffectwas lostwhenco-transformedwithMTV9-

GFP,whichledtoPVCaggregation(Figure18B).

Figure 18. PVC aggregates caused byGFP-MTV9 overexpression are not affected bywortmannin treatments or by

constitutivelyactiveARA7.(A)Arabidopsisprotoplastsco-transformedwithGFP-MTV9andRFP-VSR2(leftpanels)and

treatedwithwortmannin do not produce ring-like structures observed in RFP-VSR2 transformed protoplasts treated

with wortmannin (right panels). (B) GFP-MTV9 overexpression changes the ring-like structures of ARA7 (Q69L)

71

transformedprotoplasts(leftpanels),andaggregatesthePVCinARA7-RFPtransformedprotoplasts(rightpanels).Scale

bars,10µm.

Together, these experiments suggest thatMTV9 overexpression causes a rearrangement of the

PVCsintoaggregates,whichimpedesnormalfunctionalityofthePVCandblockswortmannin-and

ARA7(Q69L)-inducedhomotypicfusionofthecompartment.

Figure19.Wortmannindidnotproduceenlargedring-likePVCsinMTV9oeplants. InArabidopsisplantstransformed

withARA7-YFPthePVCsshowthetypicalringshape(arrowhead)aftertreatmentwith33µMwortmannintreatment.In

contrast, in plants expressing GFP-MTV9 the aggregated PVCs do not respond to wortmannin treatment. Scale bars

correspondto10µm(control:mocktreatedplants)and2,5µm(Wm:wortmannintreatedplants).

10.CharacterizationofantibodiesagainstMTV9

To characterize the endogenousMTV9protein,we raisedpolyclonal antibodies against the first

472aminoacids (NT-CORE fragment)ofMTV9.Wecloned the fragment in thepDEST17plasmid

andexpresseditinEscherichiacoli.Figure20showsthatwithinthreehoursofinductiontherewas

highexpressionoftherecombinantproteinthataccumulatedalmostentirelyininclusionbodies.

72

Figure20.RecombinantexpressionoftheNT-COREMTV9fragment(aminoacids1-472)inE.coli.Proteinextractsof

theE.coliculturesinducedwithIPTGfor0,1.5and3hwereanalyzedbySDS-pageandcoomasiestaining.i:insoluble

fraction;s:solublefraction.

We purified inclusion bodies containing high concentration of the recombinant protein

andusedthemtoimmunizerabbits.Theobtainedserashowedaspecificrecognitionofaprotein

band of the expected size (90 kDa) in samples fromWild-type (Wt) plants that was absent in

extracts from mtv9-2 mutant plants (Figure 21), indicating that it corresponds to the MTV9

protein.

Figure21.TheMTV9antibodyspecifically recognizesaproteinbandabsent inmtv9-2proteinextracts.Theasterisk

indicatestheproteinbandspecificallyrecognizedbytheMTV9antibodyinstemandflowerproteinextracts.

73

11.MTV9associateswithmembranesthatcorrespondtotheTGNandthePVC

Togain insight intothe localizationof theendogenousMTV9protein,wecarriedoutsubcellular

fractionation studies using the antibody for detection. First we performed differential

centrifugationexperimentswith samples fromWt seedlings.As shown in Figure22A, theMTV9

protein was found in the pellet fractions, consistent with association with endomembrane

compartments.ThefractionationpatternwasverysimilartothatofSYP21,aQa-SNAREresidingat

the PVC. To determine how MTV9 associates with membranes, we isolated microsomal

membranes and tested for extractionwithdifferent agents, using SYP21, an integralmembrane

protein, and VPS45, a peripheral membrane protein, as controls. Interestingly, MTV9 behaved

similarly toSYP21andcouldonlybeefficientlyextracted frommembraneswithTritonX-100. In

contrast,VPS45wasalsoextracted in thepresenceofurea. These results indicate thatMTV9 is

likelyinsertedintothemembrane(Figure22B).However,MTV9doesnotcontaintransmembrane

domains, so the insertion should be through post-translationalmodificationwith lipid anchors..

Interestingly,ithasbeenrecentlyreportedthatMTV9maybepalmitoylated(Hemsleyetal.,2013)

whichcouldservetoattachtheproteintomembranes.Palmitoylationnormallyoccursatcysteine

residues andMTV9 contains a single cysteine that is present in the CT domain, and may thus

promotePVCtargetingoftheprotein.

74

Figure22.MTV9istightlyassociatedwithmembranes(A)Differentialcentrifugationofproteinsamplesfrom10days-

oldCol-0ormtv9-2seedlings.Thenumbersindicatecentrifugalforceappliedtotheproteinsample(inthousandsg),the

“S”indicatessoluble,and“P”indicatesthepelletfraction.(B)100,000gmembranepelletfromanArabidopsis10days-

oldseedlingsextractwasresuspendedinHEbufferalone,orin2Murea,1MNaCl,0.1MNa2CO3,or1%TritonX-100,

respectively.Insolublematerialwasre-pelleted,andsolubleandpelletfractionswereanalyzedbySDS-PAGEwithanti-

MTV9, anti-SYP21 (integral membrane protein localized to the PVC) and anti-VPS45 (peripheral membrane protein

localizedtotheTGN).

In sucrose gradients, themajority ofMTV9was found in fractions containing the PVC resident

proteins VTI11 and SYP21, while VSR proteins that cycle between the TGN and the PVC only

partiallyoverlappedwithMTV9(Figure23).ThesubcellularfractionationpatternofMTV9(Figures

22and23)isconsistentwiththelocalizationoftheGFPfusionconstructsinthePVCandsupports

thattheendogenousMTV9localizestothatcompartment.

75

Figure 23. MTV9 associates with membrane fractions that correspond to the prevacuolar compartment. Sucrose

gradientoftendaysoldsseedlingsfromCol-0wereanalyzedbywesternblotswithantibodiesagainstMTV9andVTI11,

VSR,andSYP21,thepreviouslycharacterizedorganellemarkers.

12.MTV9andVTI11functioninseparatevacuolartraffickingpathways

Noneof the four isolatedmtv9mutant allelespresentedobviousphenotypes in the absenceof

VAC2,exceptforaslightbutsignificantdelay infloweringtime(Figure24).Totest if this lackof

phenotypewasduetothegeneredundancyoftheSTV9gene,weisolatedandcharacterizedtwo

T-DNAinsertionsinthatgene.Thestv9mutantallelesdidnotcausesecretionofVAC2whenthe

transgenewascrossedin,arguingagainstSTV9havingthesamefunctionsasMTV9.Moreover,an

mtv9-2stv9-1doublemutant,whichispredictedtocarrynullallelesinbothgenes,didnotdisplay

anyotherdevelopmentalphenotypethanthedelayedfloweringalreadyobservedinsinglemtv9-2

mutants.Hence,itappearsthatMTV9andSTV9arenotessentialforplantviabilityorfornormal

development.

76

Figure24.mtv9-2floweringtimeisslightlydelayedunderlongdayconditions.(A)4weekoldplantsfrommtv9-2and

Col-0backgroundsshowingthemtv9-2floweringdelay(B)Theaveragenumberofrosetteleavesorrosettepluscauline

leavesatthetimeofpetalemergenceinthefirstflower(54plantspergenotype).Theerrorbarsrepresentthestandard

deviation.AnunpairedStudent`st-testwasusedtocomparethemeans.Twoasterisks:p-value<0.001;threeasterisks:

p-value<0.0001.

The lack of any obvious phenotype in the mtv9-2 stv9-1 double mutants is striking

considering the conservation of theMTV9 gene family in plants and the drastic effect ofMTV9

overexpressiononPVCmorphologyandfunctionality.Apossibleexplanationwouldbethatthere

are MTV9-independent pathways to the vacuole that mask the effects of blocking the MTV9-

dependent pathway. To gain genetic evidence for this, we crossed mtv9 with previously

77

characterizedvacuolar traffickingmutants (mtv1,mtv2/vsr4, vsr1,mtv3,mtv4,mtv11, vti12 and

vti11), with the rationale that alterations in parallel pathways should have a synergistic effect

whencombinedwiththemtv9mutations.Thedoublemutantsfromallthecrossesdidnotshow

any increased phenotypic changes in standard growth conditions relative to the respective

parental single mutants, except for themtv9 vti11 double mutant that showed more severe

dwarfismthanthesinglevti11mutant(Figure25and26).

78

79

Figure 25. Growth phenotypes of single and double mutant combinations. Only the vti11mtv9 mutant shows

synergisticeffectsongrowthinhibitionrelativetothesinglemutantparents.

Figure26.mtv9enhancesthephenotypeofvti11mutant.Thepicturescorrespondto6weekoldplantsgrownunder

longdayconditions.

Analysisofseedstorageprecursoraccumulationinthedoublemutantsgenerateddidnot

reveal defects in processing in any of themutants (Figure 27), indicating that at least in seeds,

vacuolartraffickingwasstillproceedingefficiently.Incontrast,inleaveswecoulddetectabnormal

secretionofavacuolarperoxidaseinthemtv9-2vti11mutant,consistentwithvacuolartrafficking

beingsynergisticallyperturbedinvegetativetissuesofthedoublemutant.Consideringthatmtv9-

80

2 and vti11 are likely null alleles, these results suggest that MTV9 and VTI11 function in

independentvacuolartraffickingpathways.

Figure27.mtv9doesnotalterthetraffickingofstorageproteinsinseeds.ThelanescontainCol-0seedproteinextract

asacontrolandseedextractsfrommtv9-2mutantandindicateddoublemutantsofmtv9-2.

13.MTV11encodesahomologueofyeastVPS15.

Themtv11-1mutantwasisolatedfrompool32.TheF1plantsfromacrossbetweenmtv11-1and

Col-0 had a wild type phenotype, and in the F2 population, the proportion of plants withmtv

phenotypewasconsistentwitharecessivemutation.Weperformedbulk-segregantanalysiswith

DNAfromapoolof50F2mtvplantsandfoundaregionofchromosomeIVshowingenrichmentin

Ler SSLPs. Fine mapping restricted the interval containing themtv11-1mutation to a 0.26Mb

region(between14.40and14.66Mb)(Figure28).

81

Figure28.mtv11mappedtoa0.26MbregiononchromosomeIV.Thenumberofrecombinantplantsforthedifferent

molecularmarkers is indicated. 5 recombinant plants delimited a region between 14.4 and 14.66Mb containing the

mtv11mutation.

TheregiondelimitedthroughmappingcontainedtheAt4g29380genethatencodesahomologue

ofVPS15, anessentialgene forvacuolar traffickingandautophagy inyeastandanimals (Anding

andBaehrecke,2015;HermanandEmr,1990;Lindmoetal.,2008).SequencingofAt4g29380 in

mtv11-1plantsrevealedanEMS(G/A)mutationthatdisruptedthespliceacceptorsiteofthe7th

intron(Figure29C).

82

Figure29.mtv11encodesVPS15protein.A)DomainspresentinMTV11/AtVPS15B)Terminatedshootmeristem

phenotypeofthemtv11-1mutant,C)SchematicrepresentationoftheMTV11/AtVPS15genelocus,showingthemtv11-

1EMSmutationatthespliceacceptorsiteofthe7thintronofAtVPS15andthemtv11-2T-DNAinsertion.

Analysis by RT-PCR revealed that most of the At4g29380/AtVPS15 transcripts in the

mtv11-1 plants retained the 7th intron (Figure 30), which introduces an early stop codon that

deletes most of the coding sequence of the protein. Moreover, we also detected a significant

amount of mRNAmolecules in the mutant with the 7th intron spliced out using an alternative

spliceacceptorsite4nucleotidesdownstreamoftheoriginalone,whichcausesaframeshiftand

also leads toanearly stopcodon.These twosplicevariants (7th intronretentionandalternative

spliceacceptorsite inexon8)wouldproduceatruncatedMTV11/VPS15proteincontainingonly

thekinasedomain.Interestingly,wealsoobservedinmtv11-1skippingoftheeighthexonbyusing

anintronacceptorsiteinthe8thintron.Thisexonskippingeventmaintainsthereadingframeand

the resultant transcript would produce a truncated protein containing the kinase domain and

threeofthefourWD.

A

B

83

Figure30.Themtv11-1mutantaccumulatesdifferentially spliced transcripts. (A)RT-PCRanalysisof the retentionof

the7thintronofMTV11inWtandmtv11-1mutantplantswithflankingprimers.(B)SchematicrepresentationoftheWt

MTV11/AtVPS15protein and the two truncated versions encodedby thedifferentially spliced transcripts detected in

mtv11-1plants.

Intriguingly, null mutations in AtVPS15 show pollen lethality and thus homozygous

mutants cannot be recovered (Wang et al., 2012; Xu et al., 2011), indicating that themtv11-1

alleleretainspartialactivityofthegene.ToconfirmthatthemutationinAt4g29380/AtVPS15was

the cause of the mtv phenotype, we crossed a heterozygous T-DNA null mutant (atvps15-

2/Salk_004719)withmtv11-1homozygousforVAC2.IntheF1,weobservedplantseitherwithWt

orwithmtvphenotype.GenotypingusingaCAPSmarkerformtv11-1andflankingprimersforthe

T-DNA insertion showed that plants withmtv phenotype corresponded tomtv11-1/atvps15-2

plants andplantswithWtphenotype tomtv11-1/MTV11 plants.Hence,mtv11-1 andatvps15-2

areallelic,whichconfirmsthattheactivityofMTV11/AtVPS15isnecessaryfortransportofVAC2

tothevacuole.

Toanalyze thephenotypeofmtv11-1plants in theabsenceofVAC2orother secondary

EMSmutations,webackcrossedtheplant5timeswiththeparentalLergenotypeandsegregated

out theVAC2 transgene. Themtv11-1 showed a reduced growth rate in soil (Figure 31) and an

altered phyllotaxis, possibly due to alterations in vacuolar turnover of PIN auxin-efflux

transporters.

84

Figure31.GrowthphenotypeofWt(Ler)andmtv11-1plants.

We analyzed if transport of endogenous seed storage proteinswas also affected in the

mutant. As shown in Figure 32, themtv11-1mutant accumulated precursors of 12S globulins,

indicatingthattransportofthesestorageproteinstothevacuolewashindered(Fig32A).Protein

storage vacuoles (PSV) in seeds are autofluorescent and can thus be visualized by fluorescence

microscopy. Interestingly, PSVs inmtv11-1 embryos had lower fluorescence levels than in Wt

embryos(Figure32B),consistentwithlowerlevelsofproteinaccumulation.Weconcludeforthese

analyses that MTV11/AtVPS15 is required for vacuolar trafficking of both exogenous and

endogenousvacuolarcargo.

85

Figure32.MTV11 is involved in transportof 12S globulins. (A) Total seedprotein extracts showing accumulationof

precursor form of 12S globulin (p12S) inmtv11 (left panel). (B) Autofluorescence of protein storage vacuoles in

hypocotylcellsfrommatureWtandmtv11-1embryos.Scalebar10µm.

14.MTV11/AtVPS15localizestoendosomalcompartments

To study the localization ofMTV11/AtVPS15,we fusedGFP at theN andC-terminal end of the

proteinandexpressedtheseconstructsinNicotianabenthamianaleaves.GFP-AtVPS15showeda

cytosolic localizationwhereasAtVPS15-GFP showedboth cytosolic andpunctate pattern (Figure

33), suggesting that there are soluble and membrane bound pools of the fusion protein.

Interestingly,MTV11 has an N-terminal glycine residue that is conserved in VPS15 homologues

from plants, animals and fungi. Moreover, it has been reported that the N-terminal glycine is

myristoylated in yeast VPS15 (Herman and Emr, 1990). This myristoylation motif would be

maintained in theAtVPS15-GFPbutnot inGFP-AtVPS15, explaining thedifference in subcellular

localizationandsuggestingthatadditionofmyristicacidisnecessaryforreversibleattachmentof

MTV11 to membranes. When expressed in Arabidopsis protoplasts, GFP-AtVPS15 was also

cytosolic. Inprotoplasts,AtVPS15-GFPshowedmainlya cytosolicpatternandonly ina fewcells

86

some punctate structures could be observed. Stable 35S-AtVPS15-GFP transgenic lines were

generated but they had very low levels of transgene expression and could not be used for co-

localizationexperiments.Hence,wehaveyet to identify the compartmentswhereAtVPS15-GFP

localizes.

Figure33. Subcellular localizationofGFP-taggedMTV11/AtVPS15 in tobacco leaf cells.MTV11/AtVPS15was tagged

with GFP at the N-teminal (GFP-AtVPS15) or C-terminal (AtVPS15-GFP) of the protein and expressed in Nicotiana

benthamianaleafepidermiscells.Scalebar=100µm.

15.Themtv11-1mutanthasreducedPI3Plevels

VPS15 is a subunit of phosphatidylinositol 3-kinase from class III (PIK3C3) that is essential for

membranerecruitmentofthecomplexandforactivationofthecatalyticVPS34subunit(Stackand

Emr,1994;Stacketal.,1993).TheactivatedPIK3C3phosphorylatesphosphatidylinositol(PI)atD-3

positions to form PI3P. The localized synthesis of PI3P functions as a signal for membrane

recruitment of effector proteins involved in vesicle trafficking and autophagy. The specificity of

87

ArabidopsisthalianaPIK3C3isstillunknown.TodetermineifMTV11isinvolvedinregulatingPI3P

levels in vivo, wemade use of the PI3P biosensor line YFP-2xFYVE, consisting on YFP fused to

tandemdimeroftheFYVEdomainofmouseHsr,whichspecificallybindsPI3P,underthecontrol

of the constitutive35Spromoter (Vermeeret al., 2006).Wecrossed transgenic linesexpressing

2xFYVE (inCol-0background)withmtv11-1mutantsbackcrossed4 times intoCol-0background

andobtained in theF2mtv11-1plantshomozygous for2xFYVE.Comparisonof the fluorescence

pattern between these plants and the parental YFP-2xFYVE line showed clearly reduced

fluorescence in the mtv11-1 mutants. Whereas in Wt plants, YFP-2xFYVE strongly labelled

numerous endosomes, in many cases forming large aggregates, in mtv11-1 plants overall

fluorescencewaslowerandthe2xFYVElabeledendosomesremainedmostlyisolated(Figure34).

TheseresultssupportthatMTV11/AtVPS15functionsasanactivatoroftheVPS34PI3Pkinasein

plants.Consistentwiththis,AtVPS15interactswithAtVPS34inyeasttwohybridassays(Wanget

al., 2012) and defects in pollen germination of vps15 mutants are rescued with external

applicationof PI3P (Xu et al., 2011). The reduction in PI3P levels in endosomeswould interfere

with recruitment of trafficking effectors, explaining why vacuolar transport is perturbed in the

mtv11-1mutant.

88

Figure 34. Altered distribution and reduced levels of PI3P in endosomes ofmtv11-1 revealed by the YFP-2xFYVE

biosensor.Scalebar25µm.

16.Themtv11-1mutantsshowincreasedgrowthinarsenatecontainingmedia.

Manytoxiccompoundsaresequesteredintothevacuolesfordetoxifixation.Wethuswonderedif

defective vacuolar trafficking in mtv9 and mtv11 mutants would affect vacuolar-mediated

detoxificationcapacity.Totest this,we focusedonarsenic,oneof themost toxiccompounds in

soil.Arsenicisfoundinsoilsmainlyasarsenate[As(V)]thatistransportedintocellsbyphosphate

transporterspresent in thePM. In the cytosol arsenate is rapidly reduced into arsenite [As(III)],

complexedtosolublethiols,andtransportedintothevacuolefordetoxification(Catarechaetal.,

2007; Sanchez-Bermejoet al., 2014). Todetermine if arsenic tolerancewasaltered inmtv9 and

mtv11mutants,wemeasuredincollaborationwiththegroupofDr.AntonioLeyvaattheCNBthe

root growth dynamics of Wt and mutant plants under different concentrations of As(V). We

comparedmtv9-2withitsWtbackground(Col-0)andmtv11-1(backcrossed5timesintoLer)with

89

itsWtbackground(Ler).Wedidnot findanydifferences inAs(V)-inducedrootgrowth inhibition

betweenWtandmtv9-2plants,but strongdifferenceswereobservedbetweenmtv11-1 and its

WtcontrolinmediaccontaininghighAs(V)concentration(Figure35).Themtv11-1plantsshowed

significantlowerlevelsofrootgrowthinhibition,indicatingthattheyhaveincreasedtoleranceto

As(V)eventhoughtraffickingtovacuolesisdisturbed.

Figure35.mtv11-1showsenhancedgrowthinarseniccontainingmedia.Rootgrowthofthegenotypesindicatedinthe

graphswasmeasured inmediawith lowphosphate(-Pi:12,5µMphosphate)or inthesamemediawith10or20µM

As(V).Theaveragegrowth(relativetogrowthoftheWtin–Pimedia)isshown.Errorbars:standarddeviation.Asterisks:

significantdifferencerelativetotheWtinthatmedia(p<0.001;Student´st-test).

90

Apossibleexplanationtothissurprisingresult,consideringthatperturbationstovacuoles

were expected to interfere with detoxification, is that MTV11 participates in other trafficking

routes that require PI3P andwould be necessary for As(V) uptake into the plant. For instance,

enhanced tolerance would be expected if the mtv11-1 mutation interfered with targeting of

phosphatetransporterstothePM.Totestthis,wecrossedthemtv11-1mutantwithamarkerline

expressingthehighaffinityphosphatetransporterPHT1;1fusedtoGFPunderthecontrolofthe

35Spromoter(Gonzalezetal.,2005).Withlowphosphateinthemedia(-Pi:12,5µMphosphate),

PHT1;1-GFP was found primarily in the PM, both in Wt plants and inmtv11-1 mutants. The

presenceof30µMAs(V)in-PimediainducedturnoverofPHT1;1-GFPfromthePMinWtplants

(Figure36)aspreviouslyreported(Castrilloetal.,2013).Interestingly,themtv11-1plantsshowed

higherlevelsofPHT1;1-GFPinthePMafterexposuretoAs(V)andinadditiontheyshowedstrong

accumulationofPHT1;1-GFPinendocyticcompartments.Similarly,inphosphaterichmedia(+Pi:1

mMphosphate),mtv11-1plantsshowedhigherlevelsofPHT1;1-GFPinthePMthanWtplantsand

also significant accumulation in endocytic compartments. The detection of PHT1;1-GFP in

endocytic compartments is likely due to the perturbations in vacuolar trafficking, which would

delay targeting of the protein to the vacuole for degradation (Cardona-Lopez et al., 2015).

Importantly, these results indicate that the higher As(V) tolerance ofmtv11-1 plants cannot be

ascribedto lower levelsofPHTtransporters inthePM,butmay insteadbeduetoalterations in

the vacuole than somehow favor As(III) uptake into that compartment. A possibility is that the

concentration of arsenite conjugate transporters in the tonoplast is relatively increased in the

mutantorthatlowerlevelsofproteinaccumulationpermithigherarseniteuptake.

91

Figure36.PHT1;1isretainedinendosomalcompartmentsinmtv11-1background.SubcellulardistributionofPHT1;1-

GFPinWtandmtv11-1roots.ThedistributionofPHT1;1-GFPinlowphosphatemedium(-Pi),inhighphosphatemedium

(+Pi)andinlowphosphatemedium24hafteradding30µMAs(V)(-Pi+As(V)24h)isshown.Scalebar2,5µm.

92

Discussion

93

1.PhenotypicconsequencesofdisruptingMTV9activityMap-basedcloningofthemtv9mutantrevealedthatdefectivetraffickingoftheVAC2proteinwas

due to a point mutation introducing an early stop codon in At1g24560, a gene of unknown

function.HomologuesofMTV9/At1g24560arenotpresentoutsidetheplantkingdomorinalgae.

In contrast, all land plants contain homologues of MTV9, including Physcomitrella patens and

Marchantia polymorpha. This could indicate that MTV9 may have functions related with the

specificneedsoflandplants,suchasresistancetosalinityordrought,asdescribedforARA6(Ebine

et al., 2011). Vacuolar trafficking is an essential cellular process, and disruption of vacuolar

traffickingmachineryoftenleadstoplantlethalityorseveregrowthdefects(Niihamaetal.,2005;

Rojo et al., 2001; Sanmartin et al., 2007; Sauer et al., 2013). The mtv9 mutants have weak

phenotypes, but there is aparalogueofMTV9 inArabidopsis, theSTV9/At3g49055 gene,which

couldbegeneticallyredundantwithMTV9.However,thestv9mutantdoesnothaveterminated

meristems in thepresenceof theVAC2 transgene, indicating that its activity isnotessential for

trafficking of this cargo, possibly due to redundancy fromMTV9. Moreover,mtv9 stv9 double

mutants do not display any additional phenotypes in normal growth conditions relative to the

mtv9singlemutants,supportingthatthegenesdonothaveanessentialfunctionunderstandard

growth conditions.However, it is possible that phenotypesmaybecomeapparent in conditions

requiring optimal vacuolar trafficking capacity or under stress conditions. Moreover, it is also

possible that the alleles analyzed may still retain some gene activity, although this is unlikely

because the T-DNAs are inserted in exons and should eliminatemost of the coding sequences.

AnotherpossibilityisthatthecargotransportedthroughtheMTV9/STV9-dependentpathwaymay

bereroutedtoanMTV9/STV9-independentpathwayandstillreachthevacuole inthemtv9stv9

mutant,whichwouldexplainwhyphenotypesarenotclearlyobservable.Furthermore,MTV9and

STV9maycooperatewithotherproteins toachieveaparticular trafficking reaction (seebelow),

94

and strong defectsmay only become evident if both factors are disrupted. An example of this

would be theMTV1 andMTV4 genes, which belong to distinct gene families butmay both be

involved in clathrin coated vesicle formation at the TGN. Only when both genes are mutated,

phenotypicconsequencesatthewholeplantscaleareevident(Saueretal.,2013).

2.FunctionaldomainsinMTV9

AsearchforknowndomainsinthePfamwebsitesuggeststhattheMTV9proteincontainscoiled-

coil domains along its sequence. Coiled-coiled domains may be involved in homo or hetero

oligomerization and may also function as molecular spacers (Truebestein and Leonard, 2016).

Coiled-coildomainsare found inhomodimericmembranetethering factorssuchasgolgins (Chia

andGleeson,2014).MTV9homodimerizesandcouldalsofunctionasamembranetetheringfactor

at the PVC. Indeed, the PVC aggregation caused byMTV9 overexpression supports that itmay

tether incoming vesicles or endosomal compartments for homotypic or heterotypic fusion.

Interestingly,MTV9 is highly co-expressedwith VCL1 and VPS39,which are components of the

CORVET and HOPS multisubunit tethering complexes that regulate homotypic fusion of

endosomes to formthePVCand theheterotypic fusionof thePVCwith thevacuole in the final

step of the vacuolar trafficking pathway (Chia andGleeson, 2014; Rojo et al., 2001; Rojo et al.,

2003).Homodimerizingmembranetetheringfactorsworkatlongranges(ChiaandGleeson,2014),

whilemultisubunittetheringcomplexesfunctionatshorterdistances.ItisthuspossiblethatMTV9

cooperateswiththeCORVETand/ortheHOPScomplexestomediatemembranefusionatthePVC.

Then,anexplanationtothelackofphenotypeofthemtv9mutantscouldbethatMTV9improves

theefficiencyorspeedsuptethering,butisnotessentialfortheprocess.Incontrast,theCORVET

and HOPS complexes appear to be essential for tethering, and knockout mutations abrogate

vacuoleformationandimpairplantviability(Rojoetal.,2001).However,wehaverecentlyisolated

95

a novelmtv mutant with a hypomorphic mutation in aVPS39-like gene from the TRAP1 clade

(Klinger et al., 2013). Yeast VPS39 is a subunit of theHOPS complex that is distantly related to

yeastVPS3,asubunitoftheclassCcorevacuole/endosometethering(CORVET)complex(Klinger

etal.,2013).Interestingly,animalandplantgenomescontainseveralclosehomologuesofVPS39,

but not of VPS3. It has been proposed that the human VPS39-like homologue TRAP1 could

substituteforVPS3intheCORVETcomplex(Lachmannetal.,2014,Klingeretal.,2013).Itisthus

possiblethattheArabidopsisVPS39-likegenealsoassumesthefunctionoftheVPS3subunitinthe

CORVET complex, which functions as a Rab5 effector and coordinates membrane fusion at

early/lateendosomes.Knockoutmutations in thisVPS39-likegeneareembryonic lethal,but the

viablehypomorphicallelethatwehave isolatedwillallowustotestwhether indeedit ispartof

the CORVET complex, andwhether this complex cooperates withMTV9 for tethering incoming

vesiclesat thePVC. If thatwere the case,we should find stronggenetic interaction inadouble

mtv9vps39-Lmutant.

The fact that MTV9 and STV9 genes are specific for land plants suggests that these

organisms may have exclusive needs that require the function of these genes. Because MTV9

homologues are present in basal plants like Physcomitrella andMarchantia, it implies that the

plantspecificneedspredatetheappearanceofseedplants.InthatregarditistellingthatMTV9is

not required for vacuolar trafficking of storageproteins in seeds, but it is required for vacuolar

trafficking in vegetative cells. Clearly, vacuoles in seed cells and in vegetative cells are strikingly

different(ZouharandRojo,2009)andsoistheorganizationofintracellularcompartments,sothe

trafficking requirements may be different. Indeed, a key adaptation of land plants was the

acquisitioninvegetativecellsoflargevacuolesthatoccupymostofthecellularvolume,incontrast

toalgal,yeastandmammaliancellsthatcontainsmallvacuoles.Thelargevacuolesinlandplants

allowforenergeticallycheapgrowththat isessential fortheseautotrophicorganismstoexplore

96

the surrounding environment for light,water and nutrients. In addition, the large vacuoles also

serve tomaintain cellular homeostasis by acting as a buffering compartment (Zouhar andRojo,

2009). It is evident that having such large vacuoles alters considerably the organization of the

cytosol and the organelles within it, and therefore the trafficking between the different

compartments. It is then foreseeable the need for novel tethering factors that help vesicles to

navigate this new cellular landscape and promote efficient membrane fusion reactions. In this

regard,MTV9 (and STV9)may have been selected to perform an activity essential for effective

vacuolar trafficking in thesenovel cellular settings.However, in standardgrowth conditions,we

havenotdetectedanydeleteriousconsequencesofknockingoutMTV9andSTV9,arguingthatthe

activityofthesegenesislargelyirrelevantfornormalfunctioningoftheplant.Thechallengethen

istofindifthereareconditionsinwhichalteringthistheiractivityhasphenotypicconsequences

fortheplant.Apossibilityisthattheiractivityisespeciallyimportantfortheresponsetostresses

specifically encountered by land plants, such as salt and drought stress, or for processes that

involveactivevacuoledynamics,suchasstomatalmovements.Therefore,itwillthenbeimportant

to test all the doublemutants generated, and especially themtv9stv9 doublemutant, under a

battery of stress conditions to find conditions in which the activity of this gene family has an

essential role. Asmentioned above,Marchantia polymorpha has a singleMTV9 gene and gene

disruptionissimpleinthatorganism,soitwouldbeagoodalternativemodeltotesthowMTV9

knockoutaffectsplantdevelopmentandphysiologyinthisbasalplant.

3.MTV9localization

Subcellular fractionation of endogenous MTV9 protein and confocal analysis of GFP fusion

constructs suggest that MTV9 localizes primarily at the PVC. MTV9 does not contain a signal

peptideortransmembranedomains, indicatingthatitassociateswiththePVCfromthecytosolic

97

side, possibly through post-translationalmodifications or through interactionwith PVC resident

proteins. Recently, a proteomic study identified MTV9 as a putative palmitoylated protein

(Hemsley et al., 2013), suggesting that palmyotylation could mediate its targeting to the PVC.

Palmitoylationisareversiblepost-translationalmodificationthatcovalentlyattachespalmiticacid,

a saturated 16-carbon fatty acid, inmost cases to a cysteine residue. Thismodification confers

hydrophobicitytotheproteinfavoringtheassociationwithmembranes(Leventaletal.,2010).Our

analysis revealedthatMTV9couldonlybeextractedefficiently frommembraneswithdetergent

but not with buffers used to extract peripheralmembrane associated proteins. This pattern of

extraction resembles that of ERBIN, a humanprotein that is recruited to plasmamembrane via

palmitoylationoftwocysteines(Izawaetal.,2008),whichsupportsthatmembraneassociationof

MTV9maybemediatedthroughpalmitoylation.Inthisregard,wehavedeterminedthatthePVC-

targetingmotifofMTV9residesintheC-terminalCTdomain(aminoacids473-678),whichcontains

theonlycysteineresiduepresentinMTV9.WethusproposethatrecruitmentofMTV9tothePVC

maybemediatedbyreversiblepalmitoylationofthatcysteineresidue. Importantly, thecysteine

residueandthenexttwoaminoacidsdefineamotif (CWPmotif) that isstrictlyconserved inall

theMTV9orthologuespresentinplants,suggestingthatitiskeyfortheirfunction.TheCWPmotif

is also present in the STV9 paralogues, which are also localized in endosomal compartments.

Further characterization of the CT domain, including site directed mutagenesis of the CWP

domain,willclarifywhatmotifsarenecessaryforMTV9andSTV9localizationandfunction.Afull

characterizationofthephenotypescausedbyMTV9andSTV9knockoutwillbenecessarytothen

analyzewhatfunctionscanbecomplementedbythedifferentmutantversionsoftheproteins.

98

4.MTV9mRNAexpression

Expression studies can provide important hints to uncover gene function. For example,

MTV4/AGD5 is expressed specifically in the petal abscission zone of flowers, and, accordingly,

mtv4 mutants have a nevershed phenotype of delayed petal abscission (Liljegren et al., 2009).

Interestingly, co-expression analysis shows that MTV9 is highly co-expressed with several

traffickinggenes,includingVCL1,VPS39andALIX,whichareinvolvedinPVCtovacuoletransport.

This suggests that coordinated expression of these genes in particular conditions and tissues is

important forproper regulationofvacuolar trafficking.Promoteranalysis supports thatMTV9 is

not expressed ubiquitously throughout the plant, but has a rather specific spatial expression

pattern.Inthematurerootit isexpressedinthevasculatureandinroottipsit isspecificforthe

quiescentcentercells.Inleaves,maximalexpressionisfoundinthevasculatureandintrichomes.

MTV9 is also expressed in the hypocotyl, in the shoots and in the SAM, consistent with the

mutationofMTV9provokingVAC2 secretion in theSAMand leading topremature termination.

Cross sections of shoots reveal highest expression also in the vasculature, indicating thatMTV9

mayhaveafunctionrelatedwithconductivetissues.ThespecificexpressionofMTV9inthesecell

typesandtissueswillinformthesearchforphenotypesinthemtv9knockoutplants.Forinstance,

thevasculatureplaysakeyroleinthetransitiontofloweringbytransmittinginformationfromthe

leaves (the florigen signal, nutritional status, etc.) to the SAM to promote the switch from

vegetativetoreproductivegrowth,soitisinterestingthatmtv9mutantsshowdelayedflowering.

In contrast, no expression ofMTV9was observed in seeds,which is consistentwith the lack of

effectofthemtv9mutationinthosetissues.

99

5.Ontheactivityofthemtv11-1allele

Phosphatidylinositol 3-kinases (PI3Ks) are key enzymes involved inmany aspects of cell biology

andphysiology inorganisms fromalleukaryotickingdoms.ThreeclassesofPI3Ksarepresent in

animalcells,andeachclasshasacharacteristicsubstratespecificityandasetofcellularfunctions.

OnlythePI3KsfromclassIII(PIK3C3class)areconservedacrosskingdoms,andinyeastandplant

genomes they are the sole PI3Ks present. PIK3C3 have an invariant core formed by a catalytic

subunit(VPS34)andaregulatorysubunit(VPS15).ThiscoreinteractswithVPS30,Atg14andAtg38

toformthePIK3C3-C1complexinvolvedinautophagy,withVPS30andVPS38toformthePIK3C3-

C2 complex involved in all the vacuolar trafficking pathways in yeast (biosynthetic, endocytic,

autophagicandthecytoplasm-tovacuolepathways)andwithGPA1andAtg18toformthePIK3C3-

C3 complex involved in pheromone signaling (Reidick et al., 2017). The PIK3C3-C2 complex

synthesizes phosphatidylinositol 3-phosphate (PI3P) at specific domains of the membrane to

recruit effector proteins containing PI3P binding domains, most often PX or FYVE domains,

involvedindifferentstepsofthevacuolartraffickingpathway.UsingthesyntheticbiosensorYFP-

2XFYVEitwasreportedthatPI3PinArabidopsiscellsisgraduallydistributedintheTGN,thePVC

and the tonoplast (Vermeer et al., 2006) consistent with VPS15 and the PI3K functioning

specifically in the vacuolar trafficking pathway. Importantly, themtv11-1 mutation reduces but

does not abolish the accumulation of PI3P in endosomes, indicating that the PI3K activity is

affected but not blocked. Indeed, Arabidopsis null mutants in AtVPS34 or AtVPS15 are

gametophyticlethal(Leeetal.,2008;Wangetal.,2012;Xuetal.,2011).Inthemtv11-1mutantwe

detected abnormally splicedAtVPS15 transcripts encoding two typesof truncatedproteins, one

containing only the N-terminal kinase domain and another containing the N-terminal kinase

domain fused to threeof theC-terminalWDdomains.Most likely, it is this latter fusionprotein

that retains sufficient activity to allow plant viability. Recently, a structure of the PIK3C3-C2

100

complexat4.4Åresolutionwasreported(Rostislavlevaetal.,2015).VPS15andVPS34intertwine

in an antiparallel fashion, with all three domains of VPS15 (Kinase, HEAT and WD domains)

establishing contacts with VPS34. The catalytic domains of both proteins contact each other,

whereas the WD domain of VPS15 bridges the VPS34C2 domain and the VPS30/VPS38 BARA

domains.ThecentralregionofVPS15,whichismissinginthetruncatedmtv11-1protein,isalinker

domainthatestablishesfurthercontactswithVPS34,VPS30andVPS38.Thislinkerdomainadopts

a folded V shape so that the N-terminal kinase domain and the C-terminal WD domains are

actually nearby in the tertiary structure. Hence, the truncated protein could still maintain the

interactionswiththeVPS34kinasedomainatoneendandbridgetheVPS34C2domainandthe

VPS30/VPS38BARAdomainsattheotherend.Thiscouldexplainwhythistruncatedfusionprotein

lackingtheentirecentralregionwouldstillmaintainactivityofthePIK3C3-C2complex.Thestudy

ofthemtv11-1mutantallelemaythenreveal importantinformationonthestructural-functional

relationship of the PIK3C3-C2 complex. The complex containing the truncated AtVPS15 protein

encoded by the mutant allele is probably still correctly targeted to membranes, since the

myristoylationmotif ofAtVPS15 is intact.However, the complexmaybe less stable,becauseof

reduced interaction strength between the subunits. Alternatively, the activity of the catalytic

VPS34 subunit may be compromised. Several catalytic elements of VPS34 contact the VPS15

kinasedomain(Rostislavlevaetal.,2015)sothedeletionofthelinkerdomainmayperturbthese

contactsandaffecttheactivityofVPS34.Analyzinghowthetruncatedproteinreducesbutdoes

notabolishactivityofthePIK3C3-C2complexwilladvanceourunderstandingonhowthiscomplex

synthesizes PI3P on targetmembranes to direct effector recruitment andmembrane trafficking

reactions.

101

6.MTV9andMTV11,traffickingfactorsatthePVC

AlltheMTVgenesisolateduntilnowencodeproteinslocalizedinpost-Golgicompartmentsofthe

biosyntheticpathway.VTI12,AtVPS45,MTV1andMTV4are localizedat theTGNand inclathrin

coatedvesicles (Sanmartinetal.,2007;Saueretal.,2013;Zouharetal.,2009)andMTV2/VSR4,

VSR1andVSR3cyclebetweentheTGNandthePVC(Zouharetal.,2010).MTV9isalsolocalizedat

the PVC,where itmay function in tetheringmembranes for fusion. The compartment(s)where

MTV11 is localized is yet unidentified, but considering that PIP3, the product of the PIK3C3-C2

complex,isenrichedinmembranesoftheTGN,PVCandvacuole,andthatMTV11-GFPdisplaysa

punctatepattern,wecanassumethatMTV11isassociatedmainlywiththeTGNandthePVC.Why

doesthemtvscreenonlyunveilpost-Golgistepsofvacuolartrafficking?Thesestepsarespecific

fortransporttothevacuoleandmutationsaffectingthemshouldnotinterferewithtraffickingto

thePM.Moreover,thepresenceofindependentroutestothevacuoleinplantsmeansthatwhen

one of the pathways is blocked, transport may still proceed through an alternate pathway. In

contrast,mutationsinearlierstepsofthesecretorypathway,attheERortheGolgi,wouldaffect

transportbothtothevacuoleandthePM,andwouldbemoredetrimentaltoplantviability.Early

plant lethality precludes isolation of the mtv mutants, which are detected relatively late in

development. Indeed, some of themtv mutants isolated correspond to hypomorphic alleles of

essential genes that when knocked out cause pollen or embryo lethality. Moreover, the mtv

screenreliesontheabnormalsecretionofVAC2andifbothtransporttothevacuoleandthePM

would be affected, thenVAC2would be retained intracellularly andwould not produce anmtv

phenotype.Thesefeaturesfavortheidentificationofmutantsinthelatesecretorypathway,and

explainswhyallthemutantsisolatedthusfarareingenesactinginpost-Golgicompartments.

102

7.Arsenatephytoremediation

Although in standard growth media or in soil the mtv11-1 shows a growth delay, in media

containing high As(V) concentration,mtv11-1 has significantly increased growth relative to Wt

plants,suggestingthatitismoretoleranttothistoxicmetalloid.Themtv11-1mutantshowsclear

accumulation of the PM PHT1;1 high affinity phosphate (and As(V)) transporter in endocytic

organelles.This retention inendosomes isconsistentwithvacuolar transportbeingperturbed in

themutant.Thislikelyimpairsproperturnoverofthetransporterinthevacuoleandmayincrease

their recycling into the PM. Indeed, we found increased levels of PHT1;1 in the PM in media

containing As(V), which normally induces the internalization and vacuolar degradation of these

phosphate transporters (Castrilloetal.,2013).The failure to recycleproperly these transporters

into thevacuoleand theiraccumulation in thePM implies that themtv11-1mutantmay in fact

uptakemoreAs(V)thanWtplants,butstillbemoretoleranttothistoxicmetalloid.Ifindeedthis

were the case, understanding how themtv11-1mutant achieves As(V) hyper-accumulation and

higher tolerancewouldbeofgreat interest for thebioremediation field.Apossibility is that the

alterationsinthevacuolarpathwayallowforhigherlevelsofAs(III)sequestrationinthevacuole.

Measuring As(V) and As(III) levels in whole tissues and in isolated vacuoles (Zouhar, 2017) will

clarify if themtv11-1mutanthyper-accumulatesAs(V)andhas increasedvacuolar sequestration

capacity forAs(III). Itwill also be interesting to analyze if the reduced activity of thePIK3C3-C2

complexdirectly results inhigherAs(V) tolerance. In thatcase thiscomplexcouldbea target to

breedplantswithimprovedphytoremediationproperties.

103

Conclusions

1) Theanalysisoffourindependentmtv9allelesindicatesthatMTV9isessentialforvacuolartraffickingofVAC2invegetativetissues.

2) MTV9encodesacoiled-coilproteinthathomodimerizes.

3) MTV9islocalizedprimarilyattheprevacuolarcompartment.

4) MTV9overexpressionaltersPVCmorphologyandblocksvacuolartraffickingbutnottransporttotheplasmamembrane.

5) TheCTdomainofMTV9containsthePVCtargetinginformation.

6) Themtv11mutantphenotypeiscausedbyapointmutationinaspliceacceptorsiteofthe7thintronoftheAtVPS15gene.

7) Thehypomorphicmtv11-1alleleaccumulatesalternativelysplicedAtVPS15transcripts,includingatranscriptthatskipsexon8andretainstheopenreadingframe.

8) TheactivityoftheAtVPS15isrequiredfortraffickingofendogenousvacuolarstorageproteinsinseedtissues.

9) Themtv11-1mutationcausesreducedPIP3levelsinendosomes.

10) Themtv11-1mutantretainsmorePHT1;1intheplasmamembranebutismoretoleranttoAs(V)inthemedia,suggestingthatitcombinestolerancewithhyperaccumulation.

References

Agee,A.E.,Surpin,M.,Sohn,E.J.,Girke,T.,Rosado,A.,Kram,B.W.,Carter,C.,Wentzell,A.M.,Kliebenstein,D.J.,Jin,H.C.,etal.(2010).MODIFIEDVACUOLEPHENOTYPE1isanArabidopsismyrosinase-associatedproteininvolvedinendomembraneproteintrafficking.Plantphysiology152,120-132.Ahmed,S.U.,Rojo,E.,Kovaleva,V.,Venkataraman,S.,Dombrowski,J.E.,Matsuoka,K.,andRaikhel,N.V.(2000).TheplantvacuolarsortingreceptorAtELPisinvolvedintransportofNH(2)-terminalpropeptide-containingvacuolarproteinsinArabidopsisthaliana.TheJournalofcellbiology149,1335-1344.Altschul,S.F.,Gish,W.,Miller,W.,Myers,E.W.,andLipman,D.J.(1990).Basiclocalalignmentsearchtool.Journalofmolecularbiology215,403-410.

104

Anding,A.L.,andBaehrecke,E.H.(2015).Vps15isrequiredforstressinducedanddevelopmentallytriggeredautophagyandsalivaryglandproteinsecretioninDrosophila.Celldeathanddifferentiation22,457-464.Archbold,J.K.,Whitten,A.E.,Hu,S.H.,Collins,B.M.,andMartin,J.L.(2014).SNARE-ingthestructuresofSec1/Munc18proteins.Currentopinioninstructuralbiology29C,44-51.Avila,E.L.,Zouhar,J.,Agee,A.E.,Carter,D.G.,Chary,S.N.,andRaikhel,N.V.(2003).Toolstostudyplantorganellebiogenesis.Pointmutationlineswithdisruptedvacuolesandhigh-speedconfocalscreeningofgreenfluorescentprotein-taggedorganelles.Plantphysiology133,1673-1676.Avisar,D.,Abu-Abied,M.,Belausov,E.,andSadot,E.(2012).MyosinXIKisamajorplayerincytoplasmdynamicsandisregulatedbytwoaminoacidsinitstail.Journalofexperimentalbotany63,241-249.Balla,T.,Szentpetery,Z.,andKim,Y.J.(2009).Phosphoinositidesignaling:newtoolsandinsights.Physiology24,231-244.Banta,L.M.,Robinson,J.S.,Klionsky,D.J.,andEmr,S.D.(1988).Organelleassemblyinyeast:characterizationofyeastmutantsdefectiveinvacuolarbiogenesisandproteinsorting.TheJournalofcellbiology107,1369-1383.Bar-Peled,M.,andRaikhel,N.V.(1997).CharacterizationofAtSEC12andAtSAR1.ProteinslikelyinvolvedinendoplasmicreticulumandGolgitransport.Plantphysiology114,315-324.Bassham,D.C.,andRaikhel,N.V.(1998).AnArabidopsisVPS45phomologimplicatedinproteintransporttothevacuole.Plantphysiology117,407-415.Baud,S.,Dubreucq,B.,Miquel,M.,Rochat,C.,andLepiniec,L.(2008).StoragereserveaccumulationinArabidopsis:metabolicanddevelopmentalcontrolofseedfilling.TheArabidopsisbook/AmericanSocietyofPlantBiologists6,e0113.Bechtold,N.,Ellis,J.,andPelletier,G.(1993).InplantaAgrobacteriummediatedgenetransferbyinfiltrationofadultArabidopsisthalianaplants.CRAcadSciParisLifeSciences,15-18.Beck,M.,Zhou,J.,Faulkner,C.,MacLean,D.,andRobatzek,S.(2012).Spatio-temporalcellulardynamicsoftheArabidopsisflagellinreceptorrevealactivationstatus-dependentendosomalsorting.ThePlantcell24,4205-4219.Bednarek,S.Y.,andRaikhel,N.V.(1991).Thebarleylectincarboxyl-terminalpropeptideisavacuolarproteinsortingdeterminantinplants.ThePlantcell3,1195-1206.Bielopolski,N.,Lam,A.D.,Bar-On,D.,Sauer,M.,Stuenkel,E.L.,andAshery,U.(2014).DifferentialinteractionoftomosynwithsyntaxinandSNAP25dependsondomainsintheWD40beta-propellercoreanddeterminesitsinhibitoryactivity.TheJournalofbiologicalchemistry289,17087-17099.Blobel,G.,andDobberstein,B.(1975).Transferofproteinsacrossmembranes.I.Presenceofproteolyticallyprocessedandunprocessednascentimmunoglobulinlightchainsonmembrane-boundribosomesofmurinemyeloma.TheJournalofcellbiology67,835-851.Bocock,J.P.,Carmicle,S.,Chhotani,S.,Ruffolo,M.R.,Chu,H.,andErickson,A.H.(2009).ThePA-TM-RINGproteinRINGfingerprotein13isanendosomalintegralmembrane

105

E3ubiquitinligasewhoseRINGfingerdomainisreleasedtothecytoplasmbyproteolysis.TheFEBSjournal276,1860-1877.Bolte,S.,Lanquar,V.,Soler,M.N.,Beebo,A.,Satiat-Jeunemaitre,B.,Bouhidel,K.,andThomine,S.(2011).DistinctlyticvacuolarcompartmentsareembeddedinsidetheproteinstoragevacuoleofdryandgerminatingArabidopsisthalianaseeds.Plant&cellphysiology52,1142-1152.Bonangelino,C.J.,Chavez,E.M.,andBonifacino,J.S.(2002).GenomicscreenforvacuolarproteinsortinggenesinSaccharomycescerevisiae.Molecularbiologyofthecell13,2486-2501.Brandizzi,F.,Frangne,N.,Marc-Martin,S.,Hawes,C.,Neuhaus,J.M.,andParis,N.(2002).Thedestinationforsingle-passmembraneproteinsisinfluencedmarkedlybythelengthofthehydrophobicdomain.ThePlantcell14,1077-1092.Brett,T.J.,andTraub,L.M.(2006).Molecularstructuresofcoatandcoat-associatedproteins:functionfollowsform.Currentopinionincellbiology18,395-406.Budnik,A.,andStephens,D.J.(2009).ERexitsites--localizationandcontrolofCOPIIvesicleformation.FEBSletters583,3796-3803.Butelli,E.,Titta,L.,Giorgio,M.,Mock,H.P.,Matros,A.,Peterek,S.,Schijlen,E.G.,Hall,R.D.,Bovy,A.G.,Luo,J.,etal.(2008).Enrichmentoftomatofruitwithhealth-promotinganthocyaninsbyexpressionofselecttranscriptionfactors.Naturebiotechnology26,1301-1308.Cardona-Lopez,X.,Cuyas,L.,Marin,E.,Rajulu,C.,Irigoyen,M.L.,Gil,E.,Puga,M.I.,Bligny,R.,Nussaume,L.,Geldner,N.,etal.(2015).ESCRT-III-AssociatedProteinALIXMediatesHigh-AffinityPhosphateTransporterTraffickingtoMaintainPhosphateHomeostasisinArabidopsis.ThePlantcell27,2560-2581.Carter,C.,Pan,S.,Zouhar,J.,Avila,E.L.,Girke,T.,andRaikhel,N.V.(2004).ThevegetativevacuoleproteomeofArabidopsisthalianarevealspredictedandunexpectedproteins.ThePlantcell16,3285-3303.Castrillo,G.,Sanchez-Bermejo,E.,deLorenzo,L.,Crevillen,P.,Fraile-Escanciano,A.,Tc,M.,Mouriz,A.,Catarecha,P.,Sobrino-Plata,J.,Olsson,S.,etal.(2013).WRKY6transcriptionfactorrestrictsarsenateuptakeandtransposonactivationinArabidopsis.ThePlantcell25,2944-2957.Catarecha,P.,Segura,M.D.,Franco-Zorrilla,J.M.,Garcia-Ponce,B.,Lanza,M.,Solano,R.,Paz-Ares,J.,andLeyva,A.(2007).AmutantoftheArabidopsisphosphatetransporterPHT1;1displaysenhancedarsenicaccumulation.ThePlantcell19,1123-1133.Chanda,A.,Roze,L.V.,Kang,S.,Artymovich,K.A.,Hicks,G.R.,Raikhel,N.V.,Calvo,A.M.,andLinz,J.E.(2009).Akeyroleforvesiclesinfungalsecondarymetabolism.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica106,19533-19538.Chary,S.N.,Hicks,G.R.,Choi,Y.G.,Carter,D.,andRaikhel,N.V.(2008).Trehalose-6-phosphatesynthase/phosphataseregulatescellshapeandplantarchitectureinArabidopsis.Plantphysiology146,97-107.Chen,X.,Irani,N.G.,andFriml,J.(2011).Clathrin-mediatedendocytosis:thegatewayintoplantcells.Currentopinioninplantbiology14,674-682.Chia,P.Z.,andGleeson,P.A.(2014).Membranetethering.F1000PrimeRep6,74.

106

Chini,A.,Fonseca,S.,Fernandez,G.,Adie,B.,Chico,J.M.,Lorenzo,O.,Garcia-Casado,G.,Lopez-Vidriero,I.,Lozano,F.M.,Ponce,M.R.,etal.(2007).TheJAZfamilyofrepressorsisthemissinglinkinjasmonatesignalling.Nature448,666-671.Contento,A.L.,andBassham,D.C.(2012).Structureandfunctionofendosomesinplantcells.Journalofcellscience125,3511-3518.Cosgrove,D.J.(2000).Looseningofplantcellwallsbyexpansins.Nature407,321-326.Cui,Y.,Zhao,Q.,Gao,C.,Ding,Y.,Zeng,Y.,Ueda,T.,Nakano,A.,andJiang,L.(2014).ActivationoftheRab7GTPasebytheMON1-CCZ1ComplexIsEssentialforPVC-to-VacuoleTraffickingandPlantGrowthinArabidopsis.ThePlantcell26,2080-2097.Curtis,M.D.,andGrossniklaus,U.(2003).Agatewaycloningvectorsetforhigh-throughputfunctionalanalysisofgenesinplanta.Plantphysiology133,462-469.daSilvaConceicao,A.,Marty-Mazars,D.,Bassham,D.C.,Sanderfoot,A.A.,Marty,F.,andRaikhel,N.V.(1997).ThesyntaxinhomologAtPEP12presidesonalatepost-Golgicompartmentinplants.ThePlantcell9,571-582.DeMatteis,M.A.,andGodi,A.(2004).PI-lotingmembranetraffic.Naturecellbiology6,487-492.Denecke,J.,Botterman,J.,andDeblaere,R.(1990).Proteinsecretioninplantcellscanoccurviaadefaultpathway.ThePlantcell2,51-59.Dettmer,J.,Hong-Hermesdorf,A.,Stierhof,Y.D.,andSchumacher,K.(2006).VacuolarH+-ATPaseactivityisrequiredforendocyticandsecretorytraffickinginArabidopsis.ThePlantcell18,715-730.DiPaolo,G.,andDeCamilli,P.(2006).Phosphoinositidesincellregulationandmembranedynamics.Nature443,651-657.DiSansebastiano,G.P.(2013).DefiningnewSNAREfunctions:thei-SNARE.Frontiersinplantscience4,99.Dores,M.R.,Paing,M.M.,Lin,H.,Montagne,W.A.,Marchese,A.,andTrejo,J.(2012).AP-3regulatesPAR1ubiquitin-independentMVB/lysosomalsortingviaanALIX-mediatedpathway.Molecularbiologyofthecell23,3612-3623.Doyle,J.J.,andDoyle,J.L.(1990).IsolationofplantDNAfromfreshtissue.Focus,13-15.Drakakaki,G.,vandeVen,W.,Pan,S.,Miao,Y.,Wang,J.,Keinath,N.F.,Weatherly,B.,Jiang,L.,Schumacher,K.,Hicks,G.,etal.(2012).IsolationandproteomicanalysisoftheSYP61compartmentrevealitsroleinexocytictraffickinginArabidopsis.Cellresearch22,413-424.Dubuke,M.L.,andMunson,M.(2016).TheSecretLifeofTethers:TheRoleofTetheringFactorsinSNAREComplexRegulation.FrontCellDevBiol4,42.Ebine,K.,Fujimoto,M.,Okatani,Y.,Nishiyama,T.,Goh,T.,Ito,E.,Dainobu,T.,Nishitani,A.,Uemura,T.,Sato,M.H.,etal.(2011).Amembranetraffickingpathwayregulatedbytheplant-specificRABGTPaseARA6.Naturecellbiology13,853-859.Ebine,K.,Inoue,T.,Ito,J.,Ito,E.,Uemura,T.,Goh,T.,Abe,H.,Sato,K.,Nakano,A.,andUeda,T.(2014).Plantvacuolartraffickingoccursthroughdistinctlyregulatedpathways.Currentbiology:CB24,1375-1382.Feraru,E.,Paciorek,T.,Feraru,M.I.,Zwiewka,M.,DeGroodt,R.,DeRycke,R.,Kleine-Vehn,J.,andFriml,J.(2010).TheAP-3betaadaptinmediatesthebiogenesisandfunctionoflyticvacuolesinArabidopsis.ThePlantcell22,2812-2824.

107

Finn,R.D.,Coggill,P.,Eberhardt,R.Y.,Eddy,S.R.,Mistry,J.,Mitchell,A.L.,Potter,S.C.,Punta,M.,Qureshi,M.,Sangrador-Vegas,A.,etal.(2016).ThePfamproteinfamiliesdatabase:towardsamoresustainablefuture.NucleicAcidsRes44,D279-285.Fletcher,J.C.,Brand,U.,Running,M.P.,Simon,R.,andMeyerowitz,E.M.(1999).SignalingofcellfatedecisionsbyCLAVATA3inArabidopsisshootmeristems.Science283,1911-1914.Frigerio,L.,Hinz,G.,andRobinson,D.G.(2008).Multiplevacuolesinplantcells:ruleorexception?Traffic9,1564-1570.Fuji,K.,Shimada,T.,Takahashi,H.,Tamura,K.,Koumoto,Y.,Utsumi,S.,Nishizawa,K.,Maruyama,N.,andHara-Nishimura,I.(2007).Arabidopsisvacuolarsortingmutants(greenfluorescentseed)canbeidentifiedefficientlybysecretionofvacuole-targetedgreenfluorescentproteinintheirseeds.ThePlantcell19,597-609.Fuji,K.,Shirakawa,M.,Shimono,Y.,Kunieda,T.,Fukao,Y.,Koumoto,Y.,Takahashi,H.,Hara-Nishimura,I.,andShimada,T.(2015).TheAdaptorComplexAP-4RegulatesVacuolarProteinSortingattrans-GolgiNetworkbyInteractingwithVACUOLARSORTINGRECEPTOR1.Plantphysiology.Fuji,K.,Shirakawa,M.,Shimono,Y.,Kunieda,T.,Fukao,Y.,Koumoto,Y.,Takahashi,H.,Hara-Nishimura,I.,andShimada,T.(2016).TheAdaptorComplexAP-4RegulatesVacuolarProteinSortingatthetrans-GolgiNetworkbyInteractingwithVACUOLARSORTINGRECEPTOR1.Plantphysiology170,211-219.Gadeyne,A.,Sanchez-Rodriguez,C.,Vanneste,S.,DiRubbo,S.,Zauber,H.,Vanneste,K.,VanLeene,J.,DeWinne,N.,Eeckhout,D.,Persiau,G.,etal.(2014).TheTPLATEadaptorcomplexdrivesclathrin-mediatedendocytosisinplants.Cell156,691-704.Gao,C.,Luo,M.,Zhao,Q.,Yang,R.,Cui,Y.,Zeng,Y.,Xia,J.,andJiang,L.(2014).AuniqueplantESCRTcomponent,FREE1,regulatesmultivesicularbodyproteinsortingandplantgrowth.Currentbiology:CB24,2556-2563.Gardiner,J.,Overall,R.,andMarc,J.(2011).PutativeArabidopsishomologuesofmetazoancoiled-coilcytoskeletalproteins.Cellbiologyinternational35,767-774.Geldner,N.,Anders,N.,Wolters,H.,Keicher,J.,Kornberger,W.,Muller,P.,Delbarre,A.,Ueda,T.,Nakano,A.,andJurgens,G.(2003).TheArabidopsisGNOMARF-GEFmediatesendosomalrecycling,auxintransport,andauxin-dependentplantgrowth.Cell112,219-230.Geldner,N.,Denervaud-Tendon,V.,Hyman,D.L.,Mayer,U.,Stierhof,Y.D.,andChory,J.(2009).Rapid,combinatorialanalysisofmembranecompartmentsinintactplantswithamulticolormarkerset.ThePlantjournal:forcellandmolecularbiology59,169-178.Gershlick,D.C.,LousaCde,M.,Foresti,O.,Lee,A.J.,Pereira,E.A.,daSilva,L.L.,Bottanelli,F.,andDenecke,J.(2014).Golgi-dependenttransportofvacuolarsortingreceptorsisregulatedbyCOPII,AP1,andAP4proteincomplexesintobacco.ThePlantcell26,1308-1329.Ghosh,P.,Dahms,N.M.,andKornfeld,S.(2003).Mannose6-phosphatereceptors:newtwistsinthetale.NaturereviewsMolecularcellbiology4,202-212.Giraudo,C.G.,Hu,C.,You,D.,Slovic,A.M.,Mosharov,E.V.,Sulzer,D.,Melia,T.J.,andRothman,J.E.(2005).SNAREscanpromotecompletefusionandhemifusionasalternativeoutcomes.TheJournalofcellbiology170,249-260.

108

Gish,W.,andStates,D.J.(1993).Identificationofproteincodingregionsbydatabasesimilaritysearch.Naturegenetics3,266-272.Goh,T.,Uchida,W.,Arakawa,S.,Ito,E.,Dainobu,T.,Ebine,K.,Takeuchi,M.,Sato,K.,Ueda,T.,andNakano,A.(2007).VPS9a,thecommonactivatorfortwodistincttypesofRab5GTPases,isessentialforthedevelopmentofArabidopsisthaliana.ThePlantcell19,3504-3515.Gonzalez,E.,Solano,R.,Rubio,V.,Leyva,A.,andPaz-Ares,J.(2005).PHOSPHATETRANSPORTERTRAFFICFACILITATOR1isaplant-specificSEC12-relatedproteinthatenablestheendoplasmicreticulumexitofahigh-affinityphosphatetransporterinArabidopsis.ThePlantcell17,3500-3512.Grefen,C.,Donald,N.,Hashimoto,K.,Kudla,J.,Schumacher,K.,andBlatt,M.R.(2010).Aubiquitin-10promoter-basedvectorsetforfluorescentproteintaggingfacilitatestemporalstabilityandnativeproteindistributionintransientandstableexpressionstudies.ThePlantjournal:forcellandmolecularbiology64,355-365.Gurkan,C.,Stagg,S.M.,Lapointe,P.,andBalch,W.E.(2006).TheCOPIIcage:unifyingprinciplesofvesiclecoatassembly.NaturereviewsMolecularcellbiology7,727-738.Gutierrez,R.,Lindeboom,J.J.,Paredez,A.R.,Emons,A.M.,andEhrhardt,D.W.(2009).Arabidopsiscorticalmicrotubulespositioncellulosesynthasedeliverytotheplasmamembraneandinteractwithcellulosesynthasetraffickingcompartments.Naturecellbiology11,797-806.Hanahan,D.(1985).TechniquesfortransformationofE.coli.In:DNAcloning.(edGlover,DM)Oxford,WashingtonDC,1,109-136.Happel,N.,Honing,S.,Neuhaus,J.M.,Paris,N.,Robinson,D.G.,andHolstein,S.E.(2004).ArabidopsismuA-adaptininteractswiththetyrosinemotifofthevacuolarsortingreceptorVSR-PS1.ThePlantjournal:forcellandmolecularbiology37,678-693.Hemsley,P.A.,Weimar,T.,Lilley,K.S.,Dupree,P.,andGrierson,C.S.(2013).AproteomicapproachidentifiesmanynovelpalmitoylatedproteinsinArabidopsis.TheNewphytologist197,805-814.Herman,E.M.,andLarkins,B.A.(1999).Proteinstoragebodiesandvacuoles.ThePlantcell11,601-614.Herman,P.K.,andEmr,S.D.(1990).CharacterizationofVPS34,agenerequiredforvacuolarproteinsortingandvacuolesegregationinSaccharomycescerevisiae.Molecularandcellularbiology10,6742-6754.Hicks,G.R.,andRaikhel,N.V.(2012).Smallmoleculespresentlargeopportunitiesinplantbiology.Annualreviewofplantbiology63,261-282.Hoffman,C.S.,andWinston,F.(1987).Aten-minuteDNApreparationfromyeastefficientlyreleasesautonomousplasmidsfortransformationofEscherichiacoli.Gene57,267-272.Holsters,M.,deWaele,D.,Depicker,A.,Messens,E.,vanMontagu,M.,andSchell,J.(1978).TransfectionandtransformationofAgrobacteriumtumefaciens.Molecular&generalgenetics:MGG163,181-187.Hong,W.,andLev,S.(2014).TetheringtheassemblyofSNAREcomplexes.Trendsincellbiology24,35-43.Huotari,J.,andHelenius,A.(2011).Endosomematuration.EMBOJ30,3481-3500.Hwang,I.,andRobinson,D.G.(2009).Transportvesicleformationinplantcells.Currentopinioninplantbiology12,660-669.

109

Ichino,T.,Fuji,K.,Ueda,H.,Takahashi,H.,Koumoto,Y.,Takagi,J.,Tamura,K.,Sasaki,R.,Aoki,K.,Shimada,T.,etal.(2014).GFS9/TT9contributestointracellularmembranetraffickingandflavonoidaccumulationinArabidopsisthaliana.ThePlantjournal:forcellandmolecularbiology80,410-423.Izawa,I.,Nishizawa,M.,Hayashi,Y.,andInagaki,M.(2008).PalmitoylationofERBINisrequiredforitsplasmamembranelocalization.Genestocells:devotedtomolecular&cellularmechanisms13,691-701.janStomph,T.,Jiang,W.,andStruik,P.C.(2009).Zincbiofortificationofcereals:ricediffersfromwheatandbarley.Trendsinplantscience14,123-124.Jia,T.,Gao,C.,Cui,Y.,Wang,J.,Ding,Y.,Cai,Y.,Ueda,T.,Nakano,A.,andJiang,L.(2013).ARA7(Q69L)expressionintransgenicArabidopsiscellsinducestheformationofenlargedmultivesicularbodies.Journalofexperimentalbotany64,2817-2829.Jiang,L.,andSun,S.S.(2002).Membraneanchorsforvacuolartargeting:applicationinplantbioreactors.Trendsinbiotechnology20,99-102.JohnsonCM,S.P.,BroyerTC,CarltonAB(1957).Comparativechlorinerequirementsofdifferentplantsspecies.PlantSoil8,337-353.Jonsson,E.,Yamada,M.,Vale,R.D.,andGoshima,G.(2015).Clusteringofakinesin-14motorenablesprocessiveretrogrademicrotubule-basedtransportinplants.Natureplants1.Kim,H.,Park,M.,Kim,S.J.,andHwang,I.(2005).ActinfilamentsplayacriticalroleinvacuolartraffickingattheGolgicomplexinplantcells.ThePlantcell17,888-902.Kim,S.J.,andBassham,D.C.(2011).TNO1isinvolvedinsalttoleranceandvacuolartraffickinginArabidopsis.Plantphysiology156,514-526.Laemmli,U.K.(1970).CleavageofstructuralproteinsduringtheassemblyoftheheadofbacteriophageT4.Nature227,680-685.Lawrence,C.J.,Morris,N.R.,Meagher,R.B.,andDawe,R.K.(2001).Dyneinshaveruntheircourseinplantlineage.Traffic2,362-363.Lee,Y.,Jang,M.,Song,K.,Kang,H.,Lee,M.H.,Lee,D.W.,Zouhar,J.,Rojo,E.,Sohn,E.J.,andHwang,I.(2013).FunctionalidentificationofsortingreceptorsinvolvedintraffickingofsolublelyticvacuolarproteinsinvegetativecellsofArabidopsis.Plantphysiology161,121-133.Lee,Y.,Kim,E.S.,Choi,Y.,Hwang,I.,Staiger,C.J.,Chung,Y.Y.,andLee,Y.(2008).TheArabidopsisphosphatidylinositol3-kinaseisimportantforpollendevelopment.Plantphysiology147,1886-1897.Lee,Y.R.,andLiu,B.(2004).CytoskeletalmotorsinArabidopsis.Sixty-onekinesinsandseventeenmyosins.Plantphysiology136,3877-3883.Levental,I.,Grzybek,M.,andSimons,K.(2010).Greasingtheirway:lipidmodificationsdetermineproteinassociationwithmembranerafts.Biochemistry49,6305-6316.Li,L.,Shimada,T.,Takahashi,H.,Koumoto,Y.,Shirakawa,M.,Takagi,J.,Zhao,X.,Tu,B.,Jin,H.,Shen,Z.,etal.(2013).MAG2andthreeMAG2-INTERACTINGPROTEINsformanER-localizedcomplextofacilitatestorageproteintransportinArabidopsisthaliana.ThePlantjournal:forcellandmolecularbiology76,781-791.Liljegren,S.J.,Leslie,M.E.,Darnielle,L.,Lewis,M.W.,Taylor,S.M.,Luo,R.,Geldner,N.,Chory,J.,Randazzo,P.A.,Yanofsky,M.F.,etal.(2009).RegulationofmembranetraffickingandorganseparationbytheNEVERSHEDARF-GAPprotein.Development136,1909-1918.

110

Lindmo,K.,Brech,A.,Finley,K.D.,Gaumer,S.,Contamine,D.,Rusten,T.E.,andStenmark,H.(2008).ThePI3-kinaseregulatorVps15isrequiredforautophagicclearanceofproteinaggregates.Autophagy4,500-506.Lofke,C.,Dunser,K.,Scheuring,D.,andKleine-Vehn,J.(2015).AuxinregulatesSNARE-dependentvacuolarmorphologyrestrictingcellsize.eLife4.Marty,F.(1999).Plantvacuoles.ThePlantcell11,587-600.Matsuoka,K.,Bassham,D.C.,Raikhel,N.V.,andNakamura,K.(1995).Differentsensitivitytowortmanninoftwovacuolarsortingsignalsindicatesthepresenceofdistinctsortingmachineriesintobaccocells.TheJournalofcellbiology130,1307-1318.McGough,I.J.,andCullen,P.J.(2011).Recentadvancesinretromerbiology.Traffic12,963-971.Montesinos,J.C.,Pastor-Cantizano,N.,Robinson,D.G.,Marcote,M.J.,andAniento,F.(2014).Arabidopsisp24delta5andp24delta9facilitateCoatProteinI-dependenttransportoftheK/HDELreceptorERD2fromtheGolgitotheendoplasmicreticulum.ThePlantjournal:forcellandmolecularbiology80,1014-1030.Muntz,K.(2007).Proteindynamicsandproteolysisinplantvacuoles.Journalofexperimentalbotany58,2391-2407.Nakagawa,T.,Kurose,T.,Hino,T.,Tanaka,K.,Kawamukai,M.,Niwa,Y.,Toyooka,K.,Matsuoka,K.,Jinbo,T.,andKimura,T.(2007).Developmentofseriesofgatewaybinaryvectors,pGWBs,forrealizingefficientconstructionoffusiongenesforplanttransformation.Journalofbioscienceandbioengineering104,34-41.Neumann,U.,Brandizzi,F.,andHawes,C.(2003).Proteintransportinplantcells:inandoutoftheGolgi.Annalsofbotany92,167-180.Nielsen,E.,Cheung,A.Y.,andUeda,T.(2008).TheregulatoryRABandARFGTPasesforvesiculartrafficking.Plantphysiology147,1516-1526.Niihama,M.,Uemura,T.,Saito,C.,Nakano,A.,Sato,M.H.,Tasaka,M.,andMorita,M.T.(2005).ConversionoffunctionalspecificityinQb-SNAREVTI1homologuesofArabidopsis.Currentbiology:CB15,555-560.Nishizawa,K.,Maruyama,N.,Satoh,R.,Fuchikami,Y.,Higasa,T.,andUtsumi,S.(2003).AC-terminalsequenceofsoybeanbeta-conglycininalpha'subunitactsasavacuolarsortingdeterminantinseedcells.ThePlantjournal:forcellandmolecularbiology34,647-659.Nogales,E.(2010).Whencytoskeletalworldscollide.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica107,19609-19610.Norambuena,L.,Zouhar,J.,Hicks,G.R.,andRaikhel,N.V.(2008).IdentificationofcellularpathwaysaffectedbySortin2,asyntheticcompoundthataffectsproteintargetingtothevacuoleinSaccharomycescerevisiae.BMCchemicalbiology8,1.Novakova,P.,Hirsch,S.,Feraru,E.,Tejos,R.,vanWijk,R.,Viaene,T.,Heilmann,M.,Lerche,J.,DeRycke,R.,Feraru,M.I.,etal.(2014).SACphosphoinositidephosphatasesatthetonoplastmediatevacuolarfunctioninArabidopsis.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica111,2818-2823.Novick,P.,Ferro,S.,andSchekman,R.(1981).Orderofeventsintheyeastsecretorypathway.Cell25,461-469.

111

Novick,P.,Field,C.,andSchekman,R.(1980).Identificationof23complementationgroupsrequiredforpost-translationaleventsintheyeastsecretorypathway.Cell21,205-215.Novick,P.,andSchekman,R.(1979).Secretionandcell-surfacegrowthareblockedinatemperature-sensitivemutantofSaccharomycescerevisiae.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica76,1858-1862.Orci,L.,Stamnes,M.,Ravazzola,M.,Amherdt,M.,Perrelet,A.,Sollner,T.H.,andRothman,J.E.(1997).BidirectionaltransportbydistinctpopulationsofCOPI-coatedvesicles.Cell90,335-349.Palacin,A.,Tordesillas,L.,Gamboa,P.,Sanchez-Monge,R.,Cuesta-Herranz,J.,Sanz,M.L.,Barber,D.,Salcedo,G.,andDiaz-Perales,A.(2010).Characterizationofpeachthaumatin-likeproteinsandtheiridentificationasmajorpeachallergens.Clinicalandexperimentalallergy:journaloftheBritishSocietyforAllergyandClinicalImmunology40,1422-1430.Palade,G.(1975).Intracellularaspectsoftheprocessofproteinsynthesis.Science189,347-358.Park,M.,Lee,D.,Lee,G.J.,andHwang,I.(2005a).AtRMR1functionsasacargoreceptorforproteintraffickingtotheproteinstoragevacuole.TheJournalofcellbiology170,757-767.Park,M.,Song,K.,Reichardt,I.,Kim,H.,Mayer,U.,Stierhof,Y.D.,Hwang,I.,andJurgens,G.(2013).Arabidopsismu-adaptinsubunitAP1Mofadaptorproteincomplex1mediateslatesecretoryandvacuolartrafficandisrequiredforgrowth.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica110,10318-10323.Park,S.,Li,J.,Pittman,J.K.,Berkowitz,G.A.,Yang,H.,Undurraga,S.,Morris,J.,Hirschi,K.D.,andGaxiola,R.A.(2005b).Up-regulationofaH+-pyrophosphatase(H+-PPase)asastrategytoengineerdrought-resistantcropplants.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica102,18830-18835.Petersen,T.N.,Brunak,S.,vonHeijne,G.,andNielsen,H.(2011).SignalP4.0:discriminatingsignalpeptidesfromtransmembraneregions.NatMethods8,785-786.Pizarro,L.,andNorambuena,L.(2014).Regulationofproteintrafficking:posttranslationalmechanismsandtheunexploredtranscriptionalcontrol.Plantscience:aninternationaljournalofexperimentalplantbiology225,24-33.Puga,M.I.,Mateos,I.,Charukesi,R.,Wang,Z.,Franco-Zorrilla,J.M.,deLorenzo,L.,Irigoyen,M.L.,Masiero,S.,Bustos,R.,Rodriguez,J.,etal.(2014).SPX1isaphosphate-dependentinhibitorofPhosphateStarvationResponse1inArabidopsis.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica111,14947-14952.Raymond,C.K.,Howald-Stevenson,I.,Vater,C.A.,andStevens,T.H.(1992).Morphologicalclassificationoftheyeastvacuolarproteinsortingmutants:evidenceforaprevacuolarcompartmentinclassEvpsmutants.Molecularbiologyofthecell3,1389-1402.Reddy,A.S.(2001).Molecularmotorsandtheirfunctionsinplants.Internationalreviewofcytology204,97-178.Reidick,C.,Boutouja,F.,andPlatta,H.W.(2017).TheclassIIIphosphatidylinositol3-kinaseVps34inSaccharomycescerevisiae.Biologicalchemistry398,677-685.

112

Robert,S.,Chary,S.N.,Drakakaki,G.,Li,S.,Yang,Z.,Raikhel,N.V.,andHicks,G.R.(2008).Endosidin1definesacompartmentinvolvedinendocytosisofthebrassinosteroidreceptorBRI1andtheauxintransportersPIN2andAUX1.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica105,8464-8469.Robinson,D.G.(2014).TraffickingofVacuolarSortingReceptors:NewDataandNewProblems.Plantphysiology165,1417-1423.Robinson,D.G.,andPimpl,P.(2014).Clathrinandpost-Golgitrafficking:averycomplicatedissue.Trendsinplantscience19,134-139.Rojo,E.,Gillmor,C.S.,Kovaleva,V.,Somerville,C.R.,andRaikhel,N.V.(2001).VACUOLELESS1isanessentialgenerequiredforvacuoleformationandmorphogenesisinArabidopsis.Developmentalcell1,303-310.Rojo,E.,Sharma,V.K.,Kovaleva,V.,Raikhel,N.V.,andFletcher,J.C.(2002).CLV3islocalizedtotheextracellularspace,whereitactivatestheArabidopsisCLAVATAstemcellsignalingpathway.ThePlantcell14,969-977.Rojo,E.,Zouhar,J.,Kovaleva,V.,Hong,S.,andRaikhel,N.V.(2003).TheAtC-VPSproteincomplexislocalizedtothetonoplastandtheprevacuolarcompartmentinarabidopsis.Molecularbiologyofthecell14,361-369.Rosado,A.,Hicks,G.R.,Norambuena,L.,Rogachev,I.,Meir,S.,Pourcel,L.,Zouhar,J.,Brown,M.Q.,Boirsdore,M.P.,Puckrin,R.S.,etal.(2011).Sortin1-hypersensitivemutantslinkvacuolar-traffickingdefectsandflavonoidmetabolisminArabidopsisvegetativetissues.Chemistry&biology18,187-197.Rostislavleva,K.,Soler,N.,Ohashi,Y.,Zhang,L.,Pardon,E.,Burke,J.E.,Masson,G.R.,Johnson,C.,Steyaert,J.,Ktistakis,N.T.,etal.(2015).StructureandflexibilityoftheendosomalVps34complexrevealsthebasisofitsfunctiononmembranes.Science350,aac7365.Rothman,J.H.,andStevens,T.H.(1986).Proteinsortinginyeast:mutantsdefectiveinvacuolebiogenesismislocalizevacuolarproteinsintothelatesecretorypathway.Cell47,1041-1051.Sambrook,J.,Fritsch,E.F.,andManiatis,T.(1989).MolecularCloning:alaboratorymanual.ColdSpringLaboratoryPress.Sanchez-Bermejo,E.,Castrillo,G.,delLlano,B.,Navarro,C.,Zarco-Fernandez,S.,Martinez-Herrera,D.J.,Leo-delPuerto,Y.,Munoz,R.,Camara,C.,Paz-Ares,J.,etal.(2014).NaturalvariationinarsenatetoleranceidentifiesanarsenatereductaseinArabidopsisthaliana.Naturecommunications5,4617.Sanderfoot,A.A.,Ahmed,S.U.,Marty-Mazars,D.,Rapoport,I.,Kirchhausen,T.,Marty,F.,andRaikhel,N.V.(1998).AputativevacuolarcargoreceptorpartiallycolocalizeswithAtPEP12ponaprevacuolarcompartmentinArabidopsisroots.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica95,9920-9925.Sanderfoot,A.A.,Assaad,F.F.,andRaikhel,N.V.(2000).TheArabidopsisgenome.AnabundanceofsolubleN-ethylmaleimide-sensitivefactoradaptorproteinreceptors.Plantphysiology124,1558-1569.Sanderfoot,A.A.,Kovaleva,V.,Bassham,D.C.,andRaikhel,N.V.(2001).InteractionsbetweensyntaxinsidentifyatleastfiveSNAREcomplexeswithintheGolgi/prevacuolarsystemoftheArabidopsiscell.Molecularbiologyofthecell12,3733-3743.

113

Sanmartin,M.,Ordonez,A.,Sohn,E.J.,Robert,S.,Sanchez-Serrano,J.J.,Surpin,M.A.,Raikhel,N.V.,andRojo,E.(2007).DivergentfunctionsofVTI12andVTI11intraffickingtostorageandlyticvacuolesinArabidopsis.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica104,3645-3650.Sardiello,M.,Palmieri,M.,diRonza,A.,Medina,D.L.,Valenza,M.,Gennarino,V.A.,DiMalta,C.,Donaudy,F.,Embrione,V.,Polishchuk,R.S.,etal.(2009).Agenenetworkregulatinglysosomalbiogenesisandfunction.Science325,473-477.Sata,M.,Donaldson,J.G.,Moss,J.,andVaughan,M.(1998).BrefeldinA-inhibitedguaninenucleotide-exchangeactivityofSec7domainfromyeastSec7withyeastandmammalianADPribosylationfactors.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica95,4204-4208.Sauer,M.,Delgadillo,M.O.,Zouhar,J.,Reynolds,G.D.,Pennington,J.G.,Jiang,L.,Liljegren,S.J.,Stierhof,Y.D.,DeJaeger,G.,Otegui,M.S.,etal.(2013).MTV1andMTV4encodeplant-specificENTHandARFGAPproteinsthatmediateclathrin-dependenttraffickingofvacuolarcargofromthetrans-Golginetwork.ThePlantcell25,2217-2235.Scales,S.J.,Chen,Y.A.,Yoo,B.Y.,Patel,S.M.,Doung,Y.C.,andScheller,R.H.(2000).SNAREscontributetothespecificityofmembranefusion.Neuron26,457-464.Schekman,R.,andNovick,P.(2004).23genes,23yearslater.Cell116,S13-15,11pfollowingS19.Shimada,T.,Fuji,K.,Tamura,K.,Kondo,M.,Nishimura,M.,andHara-Nishimura,I.(2003).VacuolarsortingreceptorforseedstorageproteinsinArabidopsisthaliana.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica100,16095-16100.Shimada,T.,Koumoto,Y.,Li,L.,Yamazaki,M.,Kondo,M.,Nishimura,M.,andHara-Nishimura,I.(2006).AtVPS29,aputativecomponentofaretromercomplex,isrequiredfortheefficientsortingofseedstorageproteins.Plant&cellphysiology47,1187-1194.Simon,M.L.,Platre,M.P.,Assil,S.,vanWijk,R.,Chen,W.Y.,Chory,J.,Dreux,M.,Munnik,T.,andJaillais,Y.(2014).Amulti-colour/multi-affinitymarkersettovisualizephosphoinositidedynamicsinArabidopsis.ThePlantjournal:forcellandmolecularbiology77,322-337.Singh,M.K.,Kruger,F.,Beckmann,H.,Brumm,S.,Vermeer,J.E.,Munnik,T.,Mayer,U.,Stierhof,Y.D.,Grefen,C.,Schumacher,K.,etal.(2014).ProteindeliverytovacuolerequiresSANDprotein-dependentRabGTPaseconversionforMVB-vacuolefusion.Currentbiology:CB24,1383-1389.Sohn,E.J.,Kim,E.S.,Zhao,M.,Kim,S.J.,Kim,H.,Kim,Y.W.,Lee,Y.J.,Hillmer,S.,Sohn,U.,Jiang,L.,etal.(2003).Rha1,anArabidopsisRab5homolog,playsacriticalroleinthevacuolartraffickingofsolublecargoproteins.ThePlantcell15,1057-1070.Sohn,E.J.,Rojas-Pierce,M.,Pan,S.,Carter,C.,Serrano-Mislata,A.,Madueno,F.,Rojo,E.,Surpin,M.,andRaikhel,N.V.(2007).TheshootmeristemidentitygeneTFL1isinvolvedinflowerdevelopmentandtraffickingtotheproteinstoragevacuole.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica104,18801-18806.

114

Sparkes,I.A.,Runions,J.,Kearns,A.,andHawes,C.(2006).Rapid,transientexpressionoffluorescentfusionproteinsintobaccoplantsandgenerationofstablytransformedplants.Natureprotocols1,2019-2025.Stack,J.H.,andEmr,S.D.(1994).Vps34prequiredforyeastvacuolarproteinsortingisamultiplespecificitykinasethatexhibitsbothproteinkinaseandphosphatidylinositol-specificPI3-kinaseactivities.TheJournalofbiologicalchemistry269,31552-31562.Stack,J.H.,Herman,P.K.,Schu,P.V.,andEmr,S.D.(1993).Amembrane-associatedcomplexcontainingtheVps15proteinkinaseandtheVps34PI3-kinaseisessentialforproteinsortingtotheyeastlysosome-likevacuole.TheEMBOjournal12,2195-2204.Stagg,S.M.,LaPointe,P.,andBalch,W.E.(2007).Structuraldesignofcageandcoatscaffoldsthatdirectmembranetraffic.Currentopinioninstructuralbiology17,221-228.Stefano,G.,Renna,L.,Moss,T.,McNew,J.A.,andBrandizzi,F.(2012).InArabidopsis,thespatialanddynamicorganizationoftheendoplasmicreticulumandGolgiapparatusisinfluencedbytheintegrityoftheC-terminaldomainofRHD3,anon-essentialGTPase.ThePlantjournal:forcellandmolecularbiology69,957-966.Stevens,T.,Esmon,B.,andSchekman,R.(1982).EarlystagesintheyeastsecretorypathwayarerequiredfortransportofcarboxypeptidaseYtothevacuole.Cell30,439-448.Sun,Z.,Anderl,F.,Frohlich,K.,Zhao,L.,Hanke,S.,Brugger,B.,Wieland,F.,andBethune,J.(2007).MultipleandstepwiseinteractionsbetweencoatomerandADP-ribosylationfactor-1(Arf1)-GTP.Traffic8,582-593.Surpin,M.,Rojas-Pierce,M.,Carter,C.,Hicks,G.R.,Vasquez,J.,andRaikhel,N.V.(2005).Thepowerofchemicalgenomicstostudythelinkbetweenendomembranesystemcomponentsandthegravitropicresponse.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica102,4902-4907.Surpin,M.,Zheng,H.,Morita,M.T.,Saito,C.,Avila,E.,Blakeslee,J.J.,Bandyopadhyay,A.,Kovaleva,V.,Carter,D.,Murphy,A.,etal.(2003).TheVTIfamilyofSNAREproteinsisnecessaryforplantviabilityandmediatesdifferentproteintransportpathways.ThePlantcell15,2885-2899.Takagi,J.,Renna,L.,Takahashi,H.,Koumoto,Y.,Tamura,K.,Stefano,G.,Fukao,Y.,Kondo,M.,Nishimura,M.,Shimada,T.,etal.(2013).MAIGO5functionsinproteinexportfromGolgi-associatedendoplasmicreticulumexitsitesinArabidopsis.ThePlantcell25,4658-4675.Takahashi,H.,Tamura,K.,Takagi,J.,Koumoto,Y.,Hara-Nishimura,I.,andShimada,T.(2010).MAG4/Atp115isagolgi-localizedtetheringfactorthatmediatesefficientanterogradetransportinArabidopsis.Plant&cellphysiology51,1777-1787.Tejos,R.,Sauer,M.,Vanneste,S.,Palacios-Gomez,M.,Li,H.,Heilmann,M.,vanWijk,R.,Vermeer,J.E.,Heilmann,I.,Munnik,T.,etal.(2014).BipolarPlasmaMembraneDistributionofPhosphoinositidesandTheirRequirementforAuxin-MediatedCellPolarityandPatterninginArabidopsis.ThePlantcell26,2114-2128.Truebestein,L.,andLeonard,T.A.(2016).Coiled-coils:Thelongandshortofit.BioEssays:newsandreviewsinmolecular,cellularanddevelopmentalbiology38,903-916.

115

Ueda,H.,Nishiyama,C.,Shimada,T.,Koumoto,Y.,Hayashi,Y.,Kondo,M.,Takahashi,T.,Ohtomo,I.,Nishimura,M.,andHara-Nishimura,I.(2006).AtVAM3isrequiredfornormalspecificationofidioblasts,myrosincells.Plant&cellphysiology47,164-175.Varlamov,O.,Volchuk,A.,Rahimian,V.,Doege,C.A.,Paumet,F.,Eng,W.S.,Arango,N.,Parlati,F.,Ravazzola,M.,Orci,L.,etal.(2004).i-SNAREs:inhibitorySNAREsthatfine-tunethespecificityofmembranefusion.TheJournalofcellbiology164,79-88.Vermeer,J.E.,vanLeeuwen,W.,Tobena-Santamaria,R.,Laxalt,A.M.,Jones,D.R.,Divecha,N.,Gadella,T.W.,Jr.,andMunnik,T.(2006).VisualizationofPtdIns3Pdynamicsinlivingplantcells.ThePlantjournal:forcellandmolecularbiology47,687-700.Vernoud,V.,Horton,A.C.,Yang,Z.,andNielsen,E.(2003).AnalysisofthesmallGTPasegenesuperfamilyofArabidopsis.Plantphysiology131,1191-1208.Viotti,C.(2014).ERandvacuoles:neverbeencloser.Frontiersinplantscience5,20.Viotti,C.,Kruger,F.,Krebs,M.,Neubert,C.,Fink,F.,Lupanga,U.,Scheuring,D.,Boutte,Y.,Frescatada-Rosa,M.,Wolfenstetter,S.,etal.(2013).TheendoplasmicreticulumisthemainmembranesourceforbiogenesisofthelyticvacuoleinArabidopsis.ThePlantcell25,3434-3449.Voinnet,O.(2003).RNAsilencingbridgingthegapsinwheatextracts.Trendsinplantscience8,307-309.Vukasinovic,N.,andZarsky,V.(2016).TetheringComplexesintheArabidopsisEndomembraneSystem.FrontCellDevBiol4,46.Wada,Y.,Ohsumi,Y.,andAnraku,Y.(1992).GenesfordirectingvacuolarmorphogenesisinSaccharomycescerevisiae.I.Isolationandcharacterizationoftwoclassesofvammutants.TheJournalofbiologicalchemistry267,18665-18670.Wang,H.,Rogers,J.C.,andJiang,L.(2011).PlantRMRproteins:uniquevacuolarsortingreceptorsthatcoupleligandsortingwithmembraneinternalization.TheFEBSjournal278,59-68.Wang,J.,Ding,Y.,Wang,J.,Hillmer,S.,Miao,Y.,Lo,S.W.,Wang,X.,Robinson,D.G.,andJiang,L.(2010).EXPO,anexocyst-positiveorganelledistinctfrommultivesicularendosomesandautophagosomes,mediatescytosoltocellwallexocytosisinArabidopsisandtobaccocells.ThePlantcell22,4009-4030.Wang,L.,andRuan,Y.L.(2010).Unravelingmechanismsofcellexpansionlinkingsolutetransport,metabolism,plasmodesmtalgatingandcellwalldynamics.Plantsignaling&behavior5,1561-1564.Wang,W.Y.,Zhang,L.,Xing,S.,Ma,Z.,Liu,J.,Gu,H.,Qin,G.,andQu,L.J.(2012).ArabidopsisAtVPS15playsessentialrolesinpollengerminationpossiblybyinteractingwithAtVPS34.Journalofgeneticsandgenomics=Yichuanxuebao39,81-92.Wang,Y.,Addess,K.J.,Chen,J.,Geer,L.Y.,He,J.,He,S.,Lu,S.,Madej,T.,Marchler-Bauer,A.,Thiessen,P.A.,etal.(2007).MMDB:annotatingproteinsequenceswithEntrez's3D-structuredatabase.Nucleicacidsresearch35,D298-300.Weigel,D.,andGlazebrook,J.(2002).Arabidopsis:ALaboratoryManual.CSHLPress.Weimbs,T.,Low,S.H.,Chapin,S.J.,Mostov,K.E.,Bucher,P.,andHofmann,K.(1997).Aconserveddomainispresentindifferentfamiliesofvesicularfusionproteins:anewsuperfamily.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica94,3046-3051.

116

Wickner,W.,andSchekman,R.(2008).Membranefusion.Naturestructural&molecularbiology15,658-664.Wickstead,B.,andGull,K.(2007).Dyneinsacrosseukaryotes:acomparativegenomicanalysis.Traffic8,1708-1721.WilhelmJ.Walter,I.M.,FereshtehRafieian&StefanDiez(2015).Thenon-processivericekinesin-14OsKCH1transportsactinfilamentsalongmicrotubuleswithtwodistinctvelocities.Natureplants.Wolfenstetter,S.,Wirsching,P.,Dotzauer,D.,Schneider,S.,andSauer,N.(2012).Routestothetonoplast:thesortingoftonoplasttransportersinArabidopsismesophyllprotoplasts.ThePlantcell24,215-232.Xu,N.,Gao,X.Q.,Zhao,X.Y.,Zhu,D.Z.,Zhou,L.Z.,andZhang,X.S.(2011).ArabidopsisAtVPS15isessentialforpollendevelopmentandgerminationthroughmodulatingphosphatidylinositol3-phosphateformation.Plantmolecularbiology77,251-260.Yano,D.,Sato,M.,Saito,C.,Sato,M.H.,Morita,M.T.,andTasaka,M.(2003).ASNAREcomplexcontainingSGR3/AtVAM3andZIG/VTI11ingravity-sensingcellsisimportantforArabidopsisshootgravitropism.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica100,8589-8594.Yorimitsu,T.,Sato,K.,andTakeuchi,M.(2014).MolecularmechanismsofSar/ArfGTPasesinvesiculartraffickinginyeastandplants.Frontiersinplantscience5,411.Yu,I.M.,andHughson,F.M.(2010).Tetheringfactorsasorganizersofintracellularvesiculartraffic.Annualreviewofcellanddevelopmentalbiology26,137-156.Zhang,Y.,He,J.,Lee,D.,andMcCormick,S.(2010).InterdependenceofendomembranetraffickingandactindynamicsduringpolarizedgrowthofArabidopsispollentubes.Plantphysiology152,2200-2210.Zheng,H.,andStaehelin,L.A.(2011).Proteinstoragevacuolesaretransformedintolyticvacuolesinrootmeristematiccellsofgerminatingseedlingsbymultiple,celltype-specificmechanisms.Plantphysiology155,2023-2035.Zouhar,J.(2017).IsolationofVacuolesandtheTonoplast.Methodsinmolecularbiology1511,113-118.Zouhar,J.,Hicks,G.R.,andRaikhel,N.V.(2004).Sortinginhibitors(Sortins):ChemicalcompoundstostudyvacuolarsortinginArabidopsis.ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica101,9497-9501.Zouhar,J.,Munoz,A.,andRojo,E.(2010).Functionalspecializationwithinthevacuolarsortingreceptorfamily:VSR1,VSR3andVSR4sortvacuolarstoragecargoinseedsandvegetativetissues.ThePlantjournal:forcellandmolecularbiology64,577-588.Zouhar,J.,andRojo,E.(2009).Plantvacuoles:wheredidtheycomefromandwherearetheyheading?Currentopinioninplantbiology12,677-684.Zouhar,J.,Rojo,E.,andBassham,D.C.(2009).AtVPS45isapositiveregulatoroftheSYP41/SYP61/VTI12SNAREcomplexinvolvedintraffickingofvacuolarcargo.Plantphysiology149,1668-1678.Zouhar,J.,andSauer,M.(2014).Helpinghandsforbuddingprospects:ENTH/ANTH/VHSaccessoryproteinsinendocytosis,vacuolartransport,andsecretion.ThePlantcell26,4232-4244.Zwiewka,M.,Feraru,E.,Moller,B.,Hwang,I.,Feraru,M.I.,Kleine-Vehn,J.,Weijers,D.,andFriml,J.(2011).TheAP-3adaptorcomplexisrequiredforvacuolarfunctioninArabidopsis.Cellresearch21,1711-1722.

117