Livecellimagingofmeiosisinanthers

ofArabidopsisthaliana

Dissertationwiththeaimofachievingadoctoraldegreeat

theFacultyofMathematics,InformaticsandNaturalSciences

DepartmentofBiology

ofUniversitätHamburg

Submittedby

MariaAdaPrusicki

2018inHamburg

Supervisor:Prof.Dr.ArpSchnittger

1stExaminer:Prof.Dr.ArpSchnittger

2ndExaminer:Jun.Prof.Dr.WimWalter

Date of oral defense: 22.02.2019

Index

I

Index

ABSTRACT...................................................................................................................V

ZUSAMMENFASSUNG................................................................................................VI

NOMENCLATURE........................................................................................................IX

1 INTRODUCTION.....................................................................................................1

MEIOSISINPLANTSANDITSREGULATION......................................................................11.11.1.1Meiosis,abriefintroduction.............................................................................1

1.1.2Cytologyofmeioticprogression.......................................................................3

1.1.3Chromosomedynamicsandtheroleofcohesioninmeiosis............................8

1.1.4Meiosisinpolyploids......................................................................................11

TIMECOURSESOFPLANTMEIOSIS..............................................................................141.2

1.2.1Temperatureandgenotypeeffectsonmeioticduration................................15

1.2.2Meioticdurationinpolyploids........................................................................16

1.2.3Experimentalproceduresoftimecourses.......................................................17

IMAGINGOFMEIOSIS...............................................................................................211.3

1.3.1Livecellimagingsetups..................................................................................21

1.3.2Livecellimagingofplantmeiosis...................................................................22

2 OBJECTIVES.........................................................................................................25

3 RESULTS..............................................................................................................27

TECHNIQUEESTABLISHMENT.....................................................................................273.13.1.1Sampleisolationandmounting......................................................................27

3.1.2Microscopesetup...........................................................................................29

SELECTIONOFREPORTERLINES..................................................................................313.2

3.2.1 Functionality of the PROREC8:REC8:mEGFP and the PRORPS5A:TagRFP:TUB4

reporter..........................................................................................................................33

3.2.2TheKINGBIRDline..........................................................................................34

ALANDMARKSYSTEMFORMALEMEIOSISOFARABIDOPSISTHALIANA..............................363.3

3.3.1Cellfeaturesdescriptionandco-occurrence...................................................36

3.3.2Assessmentofcellularstatesanddefinitionofneighboringscore.................39

3.3.3ALandmarksystem........................................................................................41

3.3.4ThecaseoftheNuclearEnvelopebreakdown................................................44

Index

II

TETRAPLOIDGENERATION.........................................................................................453.4

3.4.1VIGStreatmentandKINGBIRDtetraploids.....................................................45

3.4.2 Meiosis of the F2 tetraploid KINGBIRD line progresses through the same

landmarksasdiploidplants............................................................................................48

TIMECOURSEOFMEIOSISINDIPLOIDANDTETRAPLOIDKINGBIRDLINES.........................493.5

3.5.1Diploidtimecourse.........................................................................................49

3.5.2Tetraploidtimecourseandcomparison.........................................................53 TOWARDSNEWAPPLICATIONS..................................................................................553.6

3.6.1Screenofreportersandgenerationofnewcrossedlines...............................55

3.6.2KINGBIRD2anditsintrogressioninmutantbackgrounds..............................55

3.6.3GenerationofPROREC8:REC8:mNG...................................................................58

3.6.4PerformanceofPROREC8:REC8:mNG................................................................58

4 DISCUSSION........................................................................................................63

STRENGTHSANDLIMITATIONSOFTHEMICROSCOPESETUP............................................634.1

LIVECELLIMAGINGAPPLICATIONTOSTUDYMEIOSISINCROPS........................................654.2

THELANDMARKSYSTEM...........................................................................................664.3

TOWARDSANATLASOFMEIOSIS................................................................................674.4

SINGLE CELL IMAGING REVEALS NEW INSIGHTS INTO MEIOTIC TIMING AND TISSUES4.5

SYNCHRONIZATION..................................................................................................................69

TIMECOURSEIN2XAND4X.....................................................................................714.6 FUTUREPERSPECTIVES.............................................................................................724.7

5 MATERIALANDMETHODS..................................................................................75

PLANTMATERIALANDGROWTHCONDITIONS...............................................................755.1

GENOTYPING.........................................................................................................755.2

CLONINGOFPROREC8:REC8:MNG.........................................................................765.3

PLANTTRANSFORMATIONANDCROSSING...................................................................775.4

VIGS....................................................................................................................775.5

5.5.1VIGStreatment...............................................................................................77

5.5.2SelectionofVIGStreatedplants.....................................................................78

PHENOTYPEEVALUATION.........................................................................................795.6

5.6.1Evaluationofseedabortion............................................................................79

5.6.2Pollenviabilitytest.........................................................................................79

5.6.3Cellspreads.....................................................................................................79 CONFOCALMICROSCOPY.........................................................................................805.7

Index

III

5.7.1Stillpictures....................................................................................................80

5.7.2Livecellimaging.............................................................................................80

TIMELAPSESPROCESSINGANDANALYSIS.....................................................................815.8

QUANTITATIVEANALYSISOFLIVECELLIMAGINGDATA...................................................815.9

5.9.1Landmarkextraction......................................................................................81

5.9.2Meiotictimecoursecalculation......................................................................83

REFERENCES..............................................................................................................85

ANNEXES.................................................................................................................102

APPENDIX..................................................................................................................XI

INDEXOFFIGURES.............................................................................................................XI

INDEXOFTABLES..............................................................................................................XI

INDEXOFANNEXES...........................................................................................................XII

INDEXOFABBREVIATIONS.................................................................................................XIII

PUBLICATIONSANDPRESENTATIONS........................................................................XV

EIDESSTATTLICHEVESRICHERUNG/DECLARATIONONOATH...................................XVI

CONFIRMATIONOFCORRECTENGLISH...................................................................XVII

ACKNOWLEDGMENTS............................................................................................XVIII

Abstract

V

AbstractMeiosisisacrucialeventforsexualreproduction;duringitscoursethechromosome

numberishalved,andrecombinationbetweenhomologstakesplace.Understanding

howmeiosisisregulatedinplantshasadirectimpactonbreedingapplicationsand,

therefore, researchers invest constant effort in studying its fundamental aspects.

Extensive knowledge about the meiotic progression results from the cytological

analysisoffixedmaterial.Althoughhighlyinformative,thisapproachisnotsufficient

tounderstand the dynamics ofmeiosis; numerousworks have demonstrated that

keymeiotic events as homologs paring and segregation are heavily dependent on

chromosomemovementsandcytoskeletonrearrangements,underpinningtheneed

ofaspatiotemporaldescriptionofthecelldivision.

This dissertation introduces a live cell imaging technique, based on confocal

microscopy, which allows the observation of the entiremeiotic division of pollen

mothercellsofArabidopsisthaliana.Inthisstudy,thebehaviorofsinglemeiocytesis

monitoredthroughouttheprogressionofmeiosisbythesimultaneousvisualization

ofthemeioticsubunitofcohesin,RECOMBINATION8(REC8),andmicrotubules.The

doublereporterline,namedKINGBIRD(KleisinINGreen,tuBulinInReD),allowsthe

descriptionoffivecellularfeatures:cellshape,nucleusposition,nucleolusposition,

chromosome conformation, and microtubule array. These features combine in a

non-randommannertoformcellularstates;theanalysishereperformedledtothe

identificationof11principalstates,referredtoaslandmarks,whichareconvergent

pointsofthemeioticprogression.Usingthelandmarksystemasareference,itwas

possibletodescribeaprecisetimecourseofmeiosis,whichincludedthedurationof

shortandasynchronousphases,suchasmetaphasesandanaphases.Takentogether,

thehereestablishedmicroscopytechniqueandlandmarksystemconstituteanovel

approach,whichopensnewwaystothestudyofplantmeiosis.

Zusammenfassung

VI

ZusammenfassungDie Meiose ist ein essentieller Schritt der sexuellen Fortpflanzung;während ihres

Verlaufs wird die Chromosomenzahl halbiert und eine Rekombination zwischen

homologen Chromosomen ermöglicht. Unser Verständnis der Regulation der

Meiose in Pflanzen ist für die Pflanzenzüchtung von direktem Interesse, weshalb

große Anstrengungen unternommen werden, die grundlegenden Abläufe zu

verstehen.

Aus der zytologischen Analyse von fixiertem Material wurde bereits

umfangreiches Wissen über den grundsätzlichen Ablauf der Meiose gewonnen.

Obwohl sehr informativ, reicht dieser Ansatz aber nicht aus, um die Dynamik der

Meiose imDetail zuverstehen.ZahlreicheArbeitenhabengezeigt,dassmeiotische

Schlüsselereignissewie Paarung derHomologen und deren Segregation stark von

Chromosomenbewegungen und Zytoskelettumlagerungen abhängen, was die

Notwendigkeit einer genauen räumlich-zeitlichen Beschreibung der meiotischen

Zellteilung untermauert. Mit dieser Dissertation wird eine Technik zur

Lebendzellbeobachtung während der Meiose eingeführt, die auf konfokaler

Lasermikroskopie basiert und die Beobachtung des gesamten Ablaufs der

meiotischenTeilungderPollenmutterzellenvonArabidopsisthalianaermöglicht

Durch eine gleichzeitige Visualisierung der Mikrotubuli und der meiotischen

Untereinheit von Kohäsin, RECOMBINATION 8 (REC8), kann die Entwicklung

einzelnerMeiozytenwährend des Verlaufs derMeiose genaumitverfolgtwerden.

DiehierfürkonstruiertezweifacheReporterlinienamensKINGBIRD(KleisinINGreen,

tuBulin In ReD) ermöglicht die Beschreibung von fünf Zellmerkmalen: Zellform,

Position des Zellkerns, Position des Nucleolus im Zellkern,

Chromosomenkonformation und die Anordnung der Mikrotubuli. Die spezifische

KombinationdieserMerkmalecharakterisiertjeweilsbestimmtemeiotischeStadien.

Die hier durchgeführte Analyse führte zur Identifizierung von 11Hauptzuständen,

sogenannten Referenzpunkten, die konvergente Punkte des meiotischen Ablaufs

darstellen. Mit Hilfe des Referenzpunkt-Systems konnte ein genauer zeitlicher

Verlauf der Meiose beschrieben werden, der es nun ermöglicht, auch die Dauer

kurzerundasynchronerPhasen,wieMetaphaseundAnaphase,präzisezuerfassen.

Zusammenfassung

VII

Die hier etablierte mikroskopische Technik zur Lebendbeobachtung und das

Referenzpunkt-SystemstelleneineninnovativenAnsatzdar,deresermöglicht,neue

WegeinderErforschungderMeioseinPflanzenzugehen.

.

Nomenclature

IX

NomenclatureThe Nomenclature style used in this dissertation follows the nomenclature

guidelines of TAIR and refers to Meinke and Koornneef, 1997 (Meinke and

Koornneef,1997).Plantgenesareabbreviatedwithathree-lettersymbol,writtenin

uppercase italic letters (e.g. REC8), the respective protein is named by the same

abbreviation written in uppercase roman letters (e.g. REC8); mutant genes are

referredinlowercaseitalicletters(e.g.rec8),specificmutantallelesarespecifiedby

numbers after a slash sign (e.g. tam1-2), when it is relevant themutant name is

followedby+/-forheterozygous,andby-/-forhomozygousplants.

Transgeniclinesarenamedaftertheconstructandwritteninuppercaseitalic

(e.g. PROREC8:REC8:mEGFP), with the exception of the KINGBIRD1 and KINGBIRD2

lines,which are double constructs and therefore havebeen renamed for practical

reasons.KINGBIRD1couldbefollowedby2Xor4Xtoinformaboutitsploidystate.

The reporter gene is aswell indicated by the nameof the construct in uppercase

italics(e.g.PROREC8:REC8:mEGFP),whilethefusionprotein is indicatedbythesame

name written in uppercase roman letters (e.g. REC8:mEGFP). Plasmid names are

writteninuppercaseromanlettersprecededbylowercasep(pGWB501)eventually

followedbytheT-DNAinserted(e.g.pGWB501-REC8-mNG).

Organisms are indicated using the Linnean name written in italics (e.g.

Arabidopsis thaliana, Zea mays or Saccharomyces cerevisiae) or using the short

versionofit(e.g.C.elegans,S.pombe).Alternatively,thecommonlyusednamecan

befoundinthetext(e.g.maize).

Introduction

1

1 Introduction

Meiosisinplantsanditsregulation1.1

1.1.1Meiosis,abriefintroduction

Meiosisisaspecializedeukaryoticcelldivision,whichtakesplaceinthereproductive

tissues. In animals, meiosis gives rise to gametes whereas in plants spores are

generated that eventually form the actual gametes. Meiosis consists of only one

cycle of DNA replication followed by two consecutive chromosome segregation

events.Thus,meiosisallowstheformationofhaploidgametesinadiploidorganism.

This is crucial for sexual reproduction as it prevents the doubling of a genome in

everynewgeneration.Moreover,meiosisdrivesgeneticdiversityascrossovers(CO)

between homologous chromosomes (homologs) result in new assortments of

geneticalleles. Inaddition,homologouschromosomesare randomlysegregated to

completenewchromosomesets,furthercontributingtogeneticvariation.

Given its importance in the rearrangement of genetic information,

understandingmeiosis is of crucial interest forbreeding that largely relies on the

combination of favorable alleles (Crismani et al., 2013; Lambing and Heckmann,

2018; Hand and Koltunow, 2014). Thus, the molecular mechanisms underlying

recombination and chromosome segregation, as well as entry and progression of

meiosis, havebeenahot topic forplant investigationover the lastdecades.More

than80meioticgeneshaveuptonowbeen identified inArabidopsisthaliana,Zea

maysandOryzasativa(Mercieretal.,2015;WijnkerandSchnittger,2013;Ma,2006;

ZhouandPawlowski,2014;Lambingetal.,2017).Manipulationofthesegenesisalso

a cornerstone ofnewmolecular tools thatarebeingdeveloped for cropbreeding

(Barakateetal.,2014;Calvo-Baltanasetal.,2018;Dirksetal.,2009).

Meiosisishighlyconservedamongtaxa,anditscorefactorssuchaselements

involvedindoublestrandbreak(DSB) initiationandrecombination(SPO11,RAD51,

DMC1,etc.), COs formation (MHS4.MHS5,MLH1,etc.) (Figure1.1), and structural

proteinsofthesynaptonemalcomplex(ZYP1)havebeenfoundencodedingenomes

from protists to land plants and animals (Mercier et al., 2015; Loidl, 2016).

Introduction

2

Nonetheless,major differences exist among andwithin taxa, in processes such as

homologparing, recombination control or in the presence of developmental hold

and checkpoints (Loidl,2016). For example inyeast andplantsDSB formationand

repairarenecessaryforpairingandsynapsis(HendersonandKeeney,2004;Grelon,

2001), while in C. elegans and drosophila the two process are independent

(Dernburgetal.,1998;McKimetal.,1998).

In the course of this thesis, the flowering plant Arabidopsis thaliana was

chosenasamodelsystem,andthereforethedescriptionofmeiosis that follows is

referredtotheprogressioninthisorganism;anexplicitreferencewillbemadewhen

comparisonorknowledgederivedfromotherorganismsarepresented.

Figure1.1RecombinationpathwaysofArabidopsisthaliana

The scheme illustrates the differentmolecular pathways of recombination and crossover formation duringmeiosis ofArabidopsisthaliana.AtfirstDSBsareformedbySPO11andMTOPIV.DSBendsareprocessedtoobtainsinglestrandDNA.DMC1andRAD51bind the single strandDNAandmediate the strand invasion.The single strandDNA can invade theintactsisterchromatidoroneof thehomologouschromatids.The inter-homologous intermediatecanbe resolved intoClass ICO,mediatedbyZMMproteinsandMLH1-MLH3, intoClass IICO,mediatedbyMUS81 ,orcanresult inaNonCrossOver(NCO)event.

Introduction

3

1.1.2Cytologyofmeioticprogression

In floweringplants, the establishment of the germline occurs in late stages of

development,afterthetransitionfromavegetativetoafloralmeristem(Schmidtet

al.,2015)anditconsistsinafinereprogrammingofsomaticcellfateintoameiocyte

through genetic pathways. This involves the activity of factors known to regulate

plantdevelopmentandcellproliferationasRETINOBLASTOMARELATED1(RBR1)the

WUSCHEL (WUS), CYCLIN-DEPENDENT KINESES A;1 (CDKA;1) and its inhibitor KIP-

RELATEDPROTEINS(KRPs)(Zhaoetal.,2012,2017;WijnkerandSchnittger,2013).

The newly designated meiocytes adopt a characteristic shape that radically

changeswhileundergoingmeiosis, ultimately resulting in the formationof spores.

Thesechangeshavebeenclassifiedintoasetofphases,whichbecametheframeof

referencewhenanalyzingmeioticprogression.Theseconsecutivephasesarecalled

S-phase/G2,prophaseI,metaphaseI,anaphaseI,telophaseI/interkinesis,prophase

II,metaphaseII,anaphaseII,telophaseIIandcytokinesis.ProphaseIistraditionally

subdivided into severalsub-phases: leptotene,zygotene,pachytene, diploteneand

diakinesis(Figure1.2).Eachofthesemeioticstagesischaracterizedbyphase-specific

events,e.g.,DSBareformedinearlyleptotene,COsareresolvedatmetaphaseIand

only at anaphase II the sister chromatid segregate. The molecular network that

tightly regulates these events has been deeply explored in the past (for plant

meiosis,summarizedinthefollowingreviews:Hamantetal.,2006;Luoetal.,2014;

Mercieretal.,2015;WangandCopenhaver,2018)andiscurrentlybeingexpanded,

asforthecharacterizationofASYNAPTIC4(ASY4),involvedinchromosomesynapsis

(Chambonetal.,2018),orforthenewevidencedoftheroleofTOPOISOMERASEII

(TOPII), in the resolution of chromosome entanglements (Martinez-Garcia et al.,

2018).

Acytologicaldescriptionofeachphaseisofkeyimportanceinthecontextofthis

thesis,whichinvestigatesmeiosisbymicroscopy.Agraphicalrepresentationandcell

spreadsofeachmeioticphaseareillustratedinFigure1.2.

· S-Phase/G2

DuringS-phasetheDNAisduplicated;itis likelythatthecommitmenttomeiosisis

settled in this phase and the first steps for the subsequent events ofmeiosis are

prepared.Cellshaveahomogeneousinterphaseoutlook,exceptforthebiggersize

Introduction

4

of nuclei and nucleoli comparing to somatic interphases (Armstrong et al., 2003;

Ross et al., 1996). Its duration is estimated to last longer than amitotic S-phase,

between 5 and 9 hours in Arabidopsis thaliana (Armstrong et al., 2003), and it

coincides with the expression of meiotic-specific proteins such as the cohesin

subunitRECOMBINATIONDEFICIENT8(REC8)(Caietal.,2003).DuringG2phase,the

first stretchesof chromosomeaxesappear, revealingagradual transitionbetween

themeiotic interphase and the first phaseof the division (Armstronget al., 2003;

ArmstrongandJones,2003).

· Leptotene(ProphaseI)

Leptotene meiocytes are characterized by the presence of distinguishable thin,

unpaired chromosome threads, which, towards the end of the phase, become

unevenlydistributedwithinthenucleararea(ArmstrongandJones,2003;Rossetal.,

1996).Atthesametime,thenucleolusmovesontheoppositecornerofthenucleus

(Stronghilletal.,2014;Rossetal.,1996).

· Zygotene(ProphaseI)

Zygoteneisthephaseinwhichsynapsisbetweenhomologsstarts.DAPI-stainedcell

spreadsshowareasofthin(unsynapsed)andthick(synapsed)chromosomesinthe

same nucleus, revealing that synapsis progresses (Ross et al., 1996). MTs and

organellespolarizetowardasideofthecell,whilethenucleusmovesfromacentral

positiontothesideofthecell(Rossetal.,1996;Peirsonetal.,1997;Armstrongand

Jones, 2003; Stronghill et al., 2014) In zygotene, telomeres cluster at the nuclear

envelope (NE) and form a characteristic shape, called telomere bouquet. The

telomere bouquet has been observed inmany species includingmaize, barley, as

wellasfission,buddingyeast,andmice(amongmanyobservations:Golubovskayaet

al.,2002;Higginsetal.,2012;Yuetal.,2010;TomitaandCooper,2007;Leeetal.,

2012,2015).InArabidopsisthaliana,telomeresalsoclusterbutonlyverytransiently

(Hurel et al., 2018). In addition, telomeres were reported to aggregate at the

nucleolusduringG2,andtoloosethisassociationattheearlyleptotene(Armstrong

etal.,2001).

· Pachytene(ProphaseI)

Pachytene is defined as the stage of full synapsis, the two paired arms of the

homologsarevisiblebycellspreadasadoublethreadalongthecomplete

Introduction

5

chromosome length.Nucleus and organelles can be either unevenly or evenly

distributed(Rossetal.,1996;Armstrongetal.,2003;ArmstrongandJones,2003).

· Diplotene(ProphaseI)

Also called diffused stage, diplotene is characterized by the gradual loss of

synapsis. Bivalents extend and become a mixture of paired and unpaired areas,

resembling the zygotene chromosome structure. Zygotene and diplotene cells can

bedistinguished fromeachotherbythenucleusposition:indiplotene,thenucleus

has returned to the center of the cell; therefore organelles are homogenously

distributed(Rossetal.,1996;Armstrongetal.,2003;ArmstrongandJones,2003).

· Diakinesis(ProphaseI)

Diakinesis is the last stage considered part of prophase; chromosome re-

condenseandthefivebivalentsofArabidopsisthalianacanbedetectedasseparate

entitiesinacharacteristicx-shape(Rossetal.,1996;Armstrongetal.,2003).

· MetaphaseI

At metaphase I the five bivalents reach the maximum level of condensation.

Theyalignat themetaphaseplateandchiasmataalongthechromosomearmscan

be counted (Armstrong and Jones, 2003). The spindle is formed (Peirson et al.,

1997).

· AnaphaseI

Homologs are segregated in two balanced pools and pulled towards the two

oppositepolesof thecell.Chromosomesarestillhighlycondensed,andunivalents

canbedistinguished(Rossetal.,1996)

· TelophaseI/Interkinesis

Afteranaphase I,meiocytespresent twodistinctnuclearareas, containing five

condensedunivalents.Thenuclearenvelope(NE)isre-formed.Arabidopsisthaliana

doesnotundergocytokinesisatthisstageofmeiosis,differentlyfromotherplants

suchasmaizeandrice (Zhangetal.,2018). Instead,anorganellarbandappears in

themiddle of thecell,which ismaintaineduntil theendofmeiosis II (Rossetal.,

1996; Armstrong and Jones, 2003). Most of the MTs are located between the

segregated chromosomes with a few radiating from each pole into the cortical

cytoplasm(Peirsonetal.,1997).

Introduction

6

· MeiosisII

Thesecondmeioticdivisionisthoughttolargelyresembleamitoticdivision.Its

primaryoutcome is thesegregationofsisterchromatidsandthe formationof four

spores.Itcanbedivided intothesub-phasesprophaseII,metaphaseII,anaphaseII

and finally telophase II, which is followed by cytokinesis and cell wall formation.

Sincemeiosis IIproceedsmuchfasterthanthefirstdivision, it ismorecomplicated

toobtaincellspreadsofmeiosisII,andhence, lessdetailedcytologicaldescriptions

havebeenpublished.Nonetheless, a few specific characteristics havebeennoted.

Prophase II cells can be recognized by the presence of two distinct nuclei and

diffusedchromosomes.Fivedensechromocentersarevisible ineachnucleus (Ross

etal.,1996).AttheonsetofmetaphaseII,twospindlesareformed.Theyaresmaller

than themetaphase I spindleandcomposedbya lowernumberofMTs. Theyare

parallel to the equatorial plane, but their reciprocal orientation can vary from

paralleltoperpendiculartoeachother.Chromosomesaligninthemetaphaseplane

as they would do in mitotic metaphase (Peirson et al., 1997). At anaphase II

individual chromatids are segregated, forming four groups of chromosomes. At

telophase II, the phragmoplast is formed, and the cytoplasm is finally partitioned

(Peirsonetal.,1997;Rossetal.,1996;ArmstrongandJones,2003).

Introduction

7

Figure1.2ProgressionofmalemeiosisinArabidopsisthaliana

A) Schematic representation of meiotic progression. Cytoplasm is green, nucleus is yellow and thehomologous chromosomes are depicted in blue and pink. During prophase I homologs pair andsynapse,COsaredetectableatmetaphaseI,whileexchangeofDNAbetweenhomologsisvisiblefromanaphaseI,whenhomologssegregationtakeplace.AftermetaphaseII,sisterchromatidsdividesandIIfourhaploidsporesareformed.ThefigureismodifiedfromMercieretal.,2015.

B) CellspreadsofWTCol-0.DNAwasstainedwithDAPItohighlightchromosomes(inlightgray).Scalebaris10μm.Amoredetaileddescriptionofcellspreadsinfoundinthemaintext,chapter1.1.2.

Introduction

8

1.1.3Chromosomedynamicsandtheroleofcohesioninmeiosis

As stated in the brief meiotic description in chapter 1.1, one of the most

important outcomes of meiosis is balanced chromosome segregation resulting in

fourhaploid cells.This isobtainedby a complex interactionofevents, including a

correctestablishmentofCOsandCOsresolution,aswellasanaccuratedeposition

and removalof cohesin, theprotein complex responsibleofestablishing cohesion

betweensisterchromatidsduringmitoticandmeioticdivisions.

Cohesin is formed by four conserved subunits: SMC1, SMC3, SCC3 and an α-

kleisinprotein (Figure1.3).Thesubunitsassemble inaring-likeshape,whichholds

the chromatids together,eitherembracingbothwithin the same ring (strong ring

model)orestablishingcohesindimers,eachonecontainingasinglechromatid(weak

ring model, reviewed in Nasmyth and Haering, 2009) (Figure 1.3). Arabidopsis

thaliana genome encodes for four α-kleisins: SYN1, SYN2, SYN3, and SYN4, also

known as REC8, RAD21.1, RAD21.2, and RAD21.3. Even though a certain level of

redundancyhasbeenobserved (Schubertet al.,2009), the four complexes,which

differsby the kleisin subunit,are involved indifferent functions.This is shownby

differences in mutant phenotypes (Schubert et al., 2009) as well as by their

expression in distinct tissues.REC8, in particular, is solely expressed inmeiocytes

(Cai et al., 2003) and its null mutation affects spores formation,with substantial

effectsonplantfertility,whilemitoticdivisionandplantdevelopmentdonotshow

deficiency(Peirsonetal.,1997;Baietal.,1999).

Cohesin isdepositedalongchromosomesduringS-phase(reviewed inPeterset

al., 2008), and it is maintained in position until the bipolar attachment of sister

chromatidswhen finally its complete cleavagepromotes their segregation.During

meiosis cohesin requires a stepwise removal: at the end of prophase I REC8 is

cleaved from the arm of the chromosomes allowing the resolution of COs, but

remainsloadedattheperi-centromericareas,ensuringthatsisterchromatidsdonot

segregate beforehand (Peters et al., 2008; Nasmyth and Haering, 2009). The

remainingcohesion isfinallyremovedattheonsetofanaphase II.Duetotechnical

difficultiesintheimmunolocalizationofREC8,whichdetectedastrongsignalfromS-

phase to metaphase I only (Cai et al., 2003), it was doubted that in Arabidopsis

thalianaREC8wasinvolvedinthecohesionmaintenanceaftertheonsetofanaphase

Introduction

9

I. Only recently, the presence of REC8 after metaphase I has been proved by

immunolocalization, confirming that the stepwisemodel could apply to plants as

well(Cromeretal.,2013;Yuanetal.,2018;Yuan,2018).

Both the cleavages are performed by the endopeptidase separase, which

recognizes phosphorylated REC8 as a target (Katis et al., 2010); thus the core

element of the stepwise removal is tight control of REC8 phosphorylation and

dephosphorylation (Figure 1.3). While there is no current evidence of the

phosphorylating factor of REC8, it is well known that in Arabidopsis thaliana the

protection of cohesion at anaphase I is performed by SHUGOSHIN 1 (SGO1)

(Zamariola et al., 2013; Cromer et al., 2013), which directs the PROTEIN

PHOSPHATASE PP2A to the centromeres, promoting dephosphorylation of REC8

(Yuanetal.,2018)inasimilarwaytowhatwasobservedinyeastandanimals(Clift

andMarston,2011);absenceofSGO1andPP2A results inearlydepletionofREC8

(Yuan,2018;Yuanetal.,2018),andconsequently inchromosomemis-segregation.

Additionally, another important protector of REC8, PATRONUS (PANS1) has been

identifiedinplants,havingaprominentroleduringinterkinesis(Cromeretal.,2013).

ThemeioticroleofREC8andofthecohesincomplex, ingeneral, isnotonly

restrictedtotheestablishmentchromatidcohesion.Ithasbeenprovedinsteadthat

itisinvolvedinhomologsrecognitionandpairing,inthedepositionofsynaptonemal

complex and in directing the kinetochore attachment (Bai et al., 1999; Cai et al.,

2003;Chelyshevaetal.,2005).

Introduction

10

Figure1.3Cohesinstructureandcleavageduringmeiosis

A) Schematicrepresentationofthecohesinringcomplex.B) Cohesincanconnectthetwosisterchromatids,eitherenclosingboththechromatidswithinthesame

ring(stringringmodel)orformingtworings,oneperchromatids,whichtheninteractandestablishtheconnection(weakringmodels).

C) Duringmeiosis,cohesiniscleavedinatwo-stepsmanner,whichismediatedbyREC8phosphorylation.Atfirstcohesiniscleavedfromthearmsofthechromosomes,andremainsatcentromeres,protectedbytheactionofSGO1.PP2AandPANS1.Onlyattheonsetofthesecondanaphasecohesinisentirelyremoved,andsisterchromatidsseparate.

α-kleisin

ModifiedfromNasmythandHaering2009

A B

C

ModifiedandextendedfromCli? andMarston2011

MEI

OSI

SI

MEI

OSI

SII

S-phase ProphaseI MetaphaseI AnaphaseI

AnaphaseIIMetaphaseIIInterkinesis

REC8(cohesin) Phosphate

CDK+cyclin(CDKA;1?)

SHUGOSHIN1PP2A SEPARASE

PATRONUS1PP2A SEPARASE

PATRONUS1PP2A

APC/C

Strongringmodel Weakringmodels

Introduction

11

1.1.4Meiosisinpolyploids

Polyploidyisacommonconditionamongplants;itsroleofpromoterofevolutionary

flexibility and speciation is supported by evidence of whole genome duplication

(WGD)inancestorsofmonocots(Yuetal.,2005;Jiaoetal.,2014)andangiosperms

ingeneral(Soltisetal.,2007).Advantagesofbeingpolyploidlieingeneredundancy,

whichmasks recessiveallelesandhas thepotential todevelop intogeneparalogy

(Comai,2005),aswellasinhigherheterosisandvigor,whichmightfosterpolyploid

survivalinstressconditions(Comai,2005;Sattleretal.,2016;Peeretal.,2017).The

latter characteristic, together with the increased cell size, made polyploids

interestingforbreeding.

Even though polyploidization is a frequent event, its stabilization over

generations is not (Peer et al., 2017). More commonly, low ploidy levels are

preferred, as exemplified by the experiment of Wang et al. where they proved

genome instability in octaploids of Arabidopsis thaliana mutated for TARDY

ASYNCHRONOUSMEIOSIS(TAM).Themutationcausesprematureexitaftermeiosis

I,andconsequentgenerationof2ngametes.Thefourthgenerationofoctaploidtam,

didnotshowfurthergenomedoubling,onthecontrary itsprogenywentbacktoa

hexaploid state in 32% of the cases, while the remaining were distributed from

diploids(2.4%)tooctaploids(22%)andnohigherploidylevelwereobserved(Wang

et al., 2010).Moreover,we know from numerous studies that neopolyploids can

suffer from severe infertility,primarily causedbymeioticaberrations (reviewed in

RamseyandSchemske,2002;JenczewskiandAlix,2004;Comai,2005;Zielinskiand

Mittelsten Scheid, 2012). From this observation raised the interest in studying

meioticregulationinpolyploidsorganisms.

Themajor disturbance ofmeiosis in polyploids is caused by the formation of

multivalents(associationofmorethantwohomologs)whichresultsindifficultiesin

disentanglement of chromosomes association, errors in COs resolution,

chromosome fragmentation, and ultimately in mis-segregation which induces

aneuploidyandunviablegametes(Figure1.4).Notably,thesolutionofthesedefects

isfundamentaltoestablishpolyploidyovergenerations,anditistestifiedbythefact

thatneopolyploidsare themostaffected,whileestablished linespresenta regular

Introduction

12

diploid-likemeioticdivision;epigenetics,aswellasgenetic,seemstoplayarole in

thisadaptation(Comai,2005;Bombliesetal.,2015;Peléetal.,2018)

Autopolyploids,whichderivefromaWGDeventwithinasinglespecies,andasa

consequencehaveadoublenumberofhomologouschromosomes(Figure1.4),are

the most subjected to multivalents formation. It has been hypothesized that an

increaseinCOsinterference,andthereforeareductioninCOsnumbersbutnottheir

complete disappearance, would promote the formation of bivalents over

multivalents (Bomblies et al., 2015). This concept is supported by studies in

Arabidopsis arenosa,whichexists asdiploid and asestablishednatural tetraploid.

Comparisonbetweenthegenomesofthetwopopulationsrevealed39differentiated

regions, which encodes among others for eight meiotic genes involved in

homologous recombination and synapsis: PRD3, ASY3, ASY1, REC8, ZYP1a, ZYP1b,

SMC3andPDS5(Yantetal.,2013).

Conversely,asecondgroupofpolyploidsnamedallopolyploidshasdifferent

solutions to the problem. Allopolyploids generate from hybridization followed by

WGD; theseconcomitantevents result in twoormorediploid setsofhomologous

chromosomes,whichareconsideredtobehomoeologousonetotheother (Figure

1.4).Homoeologpairing isstronglyrestrictedtherefore inallopolyploids,compared

toautopolyploids, the formationofmultivalents is rare (Comai,2005;Bomblieset

al., 2015; Pelé et al., 2018). This phenomenon is regulated at a genetic level as

demonstrated by the cases of the Pairing homoeologous 1 locus (Ph1) in the

allohexaploidTriticumaestivum (breadwheat,AABBDD;2n=6x=42).ThePh1 locus

containsaclusterofCYCLIN-DEPENDENTKINASES (CDKs)thatcontrolchromosome

arrangement at premeiotic phases, as well as chromosome synapsis and COs

formation. The absence of the Ph1 locus induces homoeologous pairing and

recombination. Similar evidence comes from the allotetraploid Brassica napus

(AACC;2n=4x=38), inwhichasinglegenewas identifiedasprimaryresponsible for

the constraint:PAIRING REGULATOR IN B. NAPUS (PRBN) (reviewed in Jenczewski

andAlix,2004;Cifuentesetal.,2010;Grandontetal.,2013;Bombliesetal.,2015).

Curiously,sincethelownumberofmultivalentsformedduringallopolyploidmeiosis,

thestrengtheningofCOsinterferenceisnotrequired,tothepointthatthenumber

Introduction

13

of COs between homologs in allopolyploids seems to increase (Zielinski and

MittelstenScheid,2012;Grandontetal.,2013)

Figure1.4Meioticdefectsandadaptationinpolyploids

The figure illustrates themeiotic defects and the possible outcome ofmeiosis in diploid, autopolyploid andallopolyploid. Homologous chromosomes are presented in the same color (magenta and dark green) whilehomeologous chromosomes are colored with different tones (magenta/purple an dark/light green). Duringdiploidmeiosis,bivalentsareformedinaringshape(indicationoftwoormoreCOsonthechromosomearms)orin a rod shape (only one CO is formed on one chromosome arm). Multivalents are easily formed duringneopolyploidmeiosis, and it results in the generationofunbalanced spores, chromosome fragmentation andaneuploidy.Establishedpolyploids insteadpresentacorrectmeioticoutcome,andbalancedspores.Thismightbeachievedwitha reductionofCOnumber inautopolyploid (only rodbivalents)orwitha strong restrictionagainsthomeologspairing,inallopolyploids.

Introduction

14

Timecoursesofplantmeiosis1.2

The course of meiosis and its consequent outcome (recombination, duration,

gameteviability)canvarysignificantlydependingon intrinsiccharacteristicsof the

organism,e.g.,chromosomenumber,ploidylevel,andgender;aswellasdepending

on environmental factors, e.g. temperature and exposure to chemicals, both in

plants(reviewedinBennett,1971,1977;Bombliesetal.,2015)andinanimals(Allard

andColaiácovo,2010;Zenzesetal.,2001).

Amongall,durationhasbeenproven tobeoneof themost variableaspects:

within the Plantae kingdom alone, meiosis can last from 16 hours in anthers of

Petunia(IzharandFrankel,1973)upto16.5daysinFritillariameleagris(reviewedby

Bennett,1977andTable1.1),orpresentanincreaseof130hoursinfemalemeiosis

of Lilium hybrids compared with the male division (Bennett, 1977; Bennett and

Stern, 1975). The main investigations about meiotic duration are dated back

between 1950 and 1980. Themajority of theseworks have been summarized by

M.D. Bennett in his thorough review “The time and duration of meiosis” (1977),

whereafewcommontraitsofthemeioticdurationwerestatedforthefirsttime:

1) Meiosisisalwaysslowerthanthemitoticdivisionofthesameorganism.

2) ProphaseIisalwaysthemostextendedphase.

3) The overall duration ofmeiosis is determined by the interaction of four

mainfactors:

· Environment,inparticulartemperature.

· Genotype.

· NuclearDNAcontent.

· Ploidylevel.

4) Theincreaseindurationofmeiosisisduetoaproportionalincreaseofthe

length of every single phase, except for organisms that present

developmentalholdor,asmorerecentdataprove,ofsinglemutationsof

meiotic genes which are responsible for specific transitions during the

division. For example in the case of Atmlh3 and Atmsh4, involved in

recombination, which cause a delay of prophase I (Higgins et al., 2004;

Jacksonetal.,2006).

Introduction

15

Bennett’s main conclusions have been re-proven and expanded by later

experiments as illustrated in the following chapter. In particular, the influence of

temperatureandgenotypehavebeenassessedinthelastdecade.

1.2.1Temperatureandgenotypeeffectsonmeioticduration

Temperature effects can be quite drastic and influencemany aspects ofmeiosis.

Heatandcoldshockscanalter the recombination rate (Bombliesetal.,2015)and

cause the arrest of meiotic progression (Draeger and Moore, 2017). Detailed

experimentsonthemeioticbehavioratdifferenttemperaturehavebeenconducted

onDasypyrumvillosum(StefaniandColonna,1996),Endymionnonscriptus(Wilson,

1959)SecalecerealeandTriticumaestivum(Bennettetal.,1971)bringingallatthe

conclusionthatmeiosisproceedsfasterathighertemperatures(Table1.1and1.2).

Implicationsofthis informationforbreedinghavebeendiscussedbyStefani

andColonna,whohypothesized that incompatiblybetweenhybridsofDasypyrum

andTriticumdependsondifferentdurationsofprophase(9hourslongerinTriticum

turgidum),andsuggestedthatalteringthetemperatureduringmeiosiscouldreduce

thetimingdifferencesand increasethehybridfertility(StefaniandColonna,1996).

Another example is brought by Higgins et al., where they correlate in Hordeum

vulgaris an altered spatiotemporal distribution of chiasmata with changes in

temperature. In particular, higher temperature synchronizes the onset of

recombination foci among the entire chromosome lengths, with no distinction

between regions distal and proximal to centromeres. This phenomenon leads to

rearrangement in thedistributionofCOs towardsotherwise cold-spots,modifying

therecombinationlandscapeofbarley(Higginsetal.,2012).

The genotypic effects on meiotic duration can be analyzed under different

aspects. At first, Bennett compares different varieties of the same species,

concluding that plants sharing a large part of the genotype proceed at similar or

identical speed, e.g. meiosis of the two cultivars of Triticum aestivum ‘Chinese

Spring' and ‘Holdfast' lasts 24-25 hours when plants are grown at the same

conditions. (Table 1.1, reviewed by (Bennett, 1977). On the other end, Bennett

hypothesizesthatmutationsinsinglemeioticgenescouldinfluencemeiosis,butthe

experimentalmeansofthetimewerenotadvancedenoughtobringclearevidence

Introduction

16

toconfirmthishypothesis.WiththedevelopmentofgenetictoolssuchasT-DNA,it

hasbeenpossible lateron toverify it.Adelayedandprolongeddivisionhasbeen

found for example in themaizemutant pam1, (plural abnormalities inmeiosis 1)

(Golubovskayaetal.,2002)andtheArabidopsisthalianamutantstam1(Magnardet

al., 2001),msh4 (MutS homolog 4) (delayof 8 hours in prophase) (Higgins et al.,

2004),andmlh3(MutLhomolog3)whichshowsadelayof25hoursinfirstmeiotic

division,foranoveralldurationofmeiosisofalmost60hours(Jacksonetal.,2006).

1.2.2Meioticdurationinpolyploids

Another central point of Bennett’s work was the study of polyploidy effects on

meiosis (Bennettet al.,1971;Bennettand Smith,1972; Finch andBennett,1972;

Bennett andKaltsikes,1973).Makinguseof cereal systems,which arepresent as

diploid progenitors (Triticum monococcum, Secale cereale and Hordeum vulgare),

and as tetraploids (Triticum dicoccum, Secale cereale and Hordeum vulgare),

hexaploids or octaploids (different varieties of Triticum aestivum and the hybrid

Triticale) Bennett and his collaborators observed that the duration of meiosis is

shortened when the ploidy level becomes higher, e.g., meiosis of Triticum

monococcumlasts42hours,whileinitshexaploidrelativeTriticumaestivummeiosis

lastsonly24hourswhengrownatthesametemperature(BennettandSmith,1972)

(Table 1.1). This phenomenonwas recorded in autopolyploids (Hordeum vulgaris,

FinchandBennett,1972)aswellasinallopolypolids(TriticaleandTriticumaestivum;

Bennett et al., 1971; Bennett and Smith, 1972) and could be considered

counterintuitivesincepreviousfindings,aswellasacomparisonamongthedifferent

speciesused in thisstudy,describedan increase inmeioticdurationparallel toan

increment in DNA content (Bennett, 1971; Bennett and Smith, 1972). The

controversy of the data did not find an exhaustive explanation in the work of

Bennett, and only recently a study conducted in Triticum aestivum proposed a

regulatory mechanism based on observations of pre-meiotic association of

centromeresinthehexaploidwheat,whichwerenotrecordedinthecorresponding

diploid(Martinez-Perezetal.,2000).

It has to be noted that the work on cereals is focused exclusively on well-

establishedpolyploids,andthereforetheshorteningofmeiosiscouldbeasecondary

Introduction

17

effectofthepreviouslymentionedmeioticadaptation(chapter1.1.4).Moreoveran

oppositetrendwassuggestedbyobservationsofmeioticprogressionofArabidopsis

arenosa;cellspreadsof thediploidand tetraploidpopulations revealed thatwhile

theonsetofmeiosiswasfoundinbudsofthesamesizeforboth,thetetraploidbuds

hosting pollen were bigger than the diploid, showing a possible delay in meiosis

(Higginsetal.,2014).

1.2.3Experimentalproceduresoftimecourses

Lookingatthetables1.1and1.2,andattheyearsofpublication,togetherwiththe

amountofinformationgiven,itbecomesclearthatafterthereviewofBennettand

untilearly2000,onlyafewworksfocusedonthedurationofmeiosis.

The reason for this gap likely lies in technical issues. The basic experimental

procedurereliedonsynchronicityofmeiosiswithinthesameflowerbudorspikelet,

and on tediousDNA labelingwith radioactive compounds such as [3H]-thymidine,

followed by autoradiography. The first attempts were based on relative timing,

expressedinthefrequencyofcellsfoundatacertainstage,withinaspecificposition

in the inflorescence or spike (Lindgren et al., 1969). These types of experiments

highlighted the differences of duration among the stages (e.g. In barley pre-

pachytene and pachytene were the longest phases, followed by diplotene and

telophaseIIandthepositionofthestageswithinthespikelet(EkbergandEriksson,

1965;Lindgrenetal.,1969),butoftenledtoanimpreciseorquitebroadcalculation

oftime.

Anewriseoftimecoursesastoolstostudymeiosiscameaftertheintroduction

of immunolabelling techniquesbasedon themodified thymineanalog5-bromo-2’-

deoxyuridine (BrdU) (Gratzner,1982)or5-ethynyl-2’-deoxyuridine (EdU) (Salicand

Mitchison,2008).Theusageofantibodiesagainstthesubstitutiveformofthymine,

introgressed inDNAduringreplication, ismuch fasterand lessdangerousthanthe

utilization of autoradiographic procedures. Armstrong was the first to describe a

time course of plant meiosis using BrdU in Arabidopsis thaliana, being able to

quantify the duration of each stage until diplotene (Armstrong et al., 2003). This

methodallowstheconcomitantimmunostainingofmeioticproteinsandwasapplied

as a tool to study mutant phenotypes and protein expression patterns in later

Introduction

18

studies such as (Higgins et al., 2004; Jackson et al., 2006; Sanchez-Moran et al.,

2007).EdUwasinsteadintroducedafewyearslatereitherincombinationwithBrdU

aspresentedbyHigginsetal.intheirtimecourseofHordeumvulgaris(Higginsetal.,

2012),or to introduce anewcytological technique tostudymeiosis inArabidopsis

thaliana as in Stronghill et al. (Stronghill et al., 2014) where they maintain the

tridimensionalstructureofthepollensacs,allowingtheevaluationofmorecellular

features. In this study, they re-confirmed the duration of meiosis in WT male

Arabidopsisthalianatobearound22-24hoursaspreviouslyobtainedbyArmstrong

andSanchez-Moran(Table1.2).

All the methods described above, independently on the labeling system, are

basedon fixationofmeiocytes,and theactual timeofeachmeiotic step is retro-

calculated as an estimation of the distribution in the percentage of the meiotic

stagesover the samples,after acertain intervalof times.Thiscalculationhas two

major drawbacks: on one side it flattens the small asynchrony, which is present

within the samepollen sac.Thisasynchronywasestimated tobe0.5 to1hour in

PetuniabyIzharandFrankel,(IzharandFrankel,1973),anditislikelythereasonwhy

manytimecoursesarenotabletodescribedistinctphasesdurationfromdiplotene

onwards (Armstrong et al., 2003; Higgins et al., 2012; Pacini and Cresti, 1978;

Sanchez-Moranetal.,2007;StefaniandColonna,1996;Stronghilletal.,2014and

Table1.2).Theseconddrawbackistheimpossibilitytodisambiguatecasesinwhich

themeioticdivisionproceedatadifferentspeedthaninWTfromcasesinwhichthe

progression isarrestedas in themutantdescriptionofpam1 (Golubovskayaetal.,

2002).Bothproblemscanbesolvedapplyinglivecellimagingtechniques,asproven

bydatafromYuetal.,Nannasetal.onmaizemeiosis(Yuetal.,2009;Nannasetal.,

2016).YuandNannaswereabletodefinepreciselythedurationandprogressionof

thesecondmeioticdivision,withaparticularfocusonanaphase II,which lasts less

than15minutes(Table1.1)andhavebeenrarelyrecordedbyfixedmaterial(Nannas

et al., 2016; Yu et al., 1997). Moreover, live cell imaging permits direct time

calculationand,whenperformedforsufficienttime,wouldbringnewinsightsabout

meiosis of mutants, starting with the distinction between arrested and delayed

progression, to the identification of their specific malfunctions and cytological

effects.

Table1.1Du

ratio

nofm

eiosisinM

onocotyled

ons

MONOCO

Tplan

tspe

cie

publication

tempe

rature

overalldurationofm

alemeiosisprem

eiosis

leptoten

ezygo

tene

pachyten

ediploten

ediakinesis

metap

haseIan

apha

seI

teloph

aseI

interkinesis

prop

haseII

metap

haseIIan

apha

seII

teloph

aseII

tetrad

sAlliumcepa

Vasil195

9NOTGIVE

N96h

ours

////

////

////

////

////

////

////

//Co

nvallaria

majalis

Benn

ett1

973

20°C

72h

ours

////

////

////

////

////

////

////

//Da

sypyrumvillosum

Stefaniand

Colon

na199

6fie

ldinM

ay35

±1.7hou

rs//

8ho

urs

//fie

ldinJu

ly22

±2hou

rs//

3.5ho

urs

//5°

C13

6±14

.4hou

rs//

22h

ours

//10

°C88

±5.3hou

rs//

20h

ours

//20

°C29h

ours

//5ho

urs

//28

°C21

±0.7ho

urs

//4.5ho

urs

//35

°C17

±0.7ho

urs

//3ho

urs

//0°

C864ho

urs

////

////

////

////

////

////

////

//5°

C360ho

urs

////

////

////

////

////

////

////

//10

°C168ho

urs

////

////

////

////

////

////

////

//15

°C84h

ours

////

////

////

////

////

////

////

//20

°C48h

ours

////

////

////

////

////

////

////

//25

°C30h

ours

////

////

////

////

////

////

////

//30

°C20h

ours

////

////

////

////

////

////

////

//15

-21°

C66h

ours

////

////

////

////

////

////

////

//Fritillaria

meleagris

Barber194

212

-15°

C40

0ho

ursA

PPRO

XIMAT

E//

////

////

////

////

////

////

////

Gasteriatrigon

aSt

raub

,193

7NOTGIVE

N96h

ours

////

////

////

////

////

////

////

////

32.6

0%23

.70%

////

////

//19

.50%

8%24

.80%

2.40

%4.

60%

13.0

0%1.

50%

7.20

%5.

40%

13.6

0%//

////

////

17.9

0%7.

50%

24.1

0%2.

50%

4.70

%13

.40%

1.60

%7.

60%

5.80

%14

.90%

//Ho

rdeumvulga

re:S

ulta

nBe

nnetta

ndFinch197

120

°C39h

ours

////

////

////

////

////

////

////

//Ho

rdeumvulga

re:Y

mer

FinchandBe

nnett1

972

20°C

39h

ours

////

////

////

////

////

////

////

//Ho

rdeumvulga

re:Y

mer4

XFinchandBe

nnett1

972

20°C

31h

ours

////

////

////

////

////

////

////

//

22°C

43h

ours

13h

ours

30°C

43h

ours

9ho

urs

Liliumcan

didu

mSaue

rland

195

6NOTGIVE

N168ho

urs

////

////

////

////

////

////

////

//Liliumhenryi

Pereira

and

Linksins1

963

NOTGIVE

N170ho

urs

////

////

////

////

////

////

////

//Liliumhybrid

:BlackBeuty

Benn

etta

ndStern197

520

°C264ho

urs

////

////

////

////

////

////

////

//Liliumhybrid

:Son

ata

Benn

etta

ndStern197

520

°C180ho

urs

////

////

////

////

////

////

////

//Liliumlong

iflorum

:varietyunspecified

Marqu

ardt193

7NOTGIVE

N

96h

ours

////

////

////

////

////

////

////

//Liliumlong

iflorum

:NellieW

hite

Itoand

Stern196

722

°Cca19

2ho

urs

////

////

////

////

////

////

////

//Liliumlong

iflorum

:Cro

ftTaylorand

McM

aster1

954

23°C

ca19

2ho

urs

////

////

////

////

////

////

////

//Liliumlong

iflorum

:Flo

ridii

Eric

kson

194

8NOTGIVE

Nca24

0hou

rs//

////

////

////

////

////

////

////

Ornith

ogalum

vire

nsCh

urchand

Wim

ber1

969

18°C

72hou

rs-A

PPRO

XIMAT

E//

////

////

////

////

////

////

////

Rhoeodiscolor

Vasil195

9NOTGIVE

N48h

ours

////

////

////

////

////

////

////

//15

°C88h

ours

////

////

////

////

////

////

////

//20

°C51h

ours

////

////

////

////

////

////

////

//25

°C39h

ours

////

////

////

////

////

////

////

//Secalecereale4X

20°C

38h

ours

////

////

////

////

////

////

////

//Steinitz194

4;Taylor1

949,

1950

;BeattyandBe

atty

1953

NOTGIVE

N126ho

urs

////

////

////

////

////

////

////

//Steinitz194

4NOTGIVE

N52h

ours

////

////

////

////

////

////

////

//

SaxandEdmon

ds,195

318

°to23

°48h

ours

////

////

////

////

////

////

////

//SaxandEdmon

ds,193

3NOTGIVE

N144ho

urs

////

////

////

////

////

////

////

//Ho

ttaandStern19

631°

C2160h

ours

////

////

////

////

////

////

////

//Ho

ttaandStern19

632°

C1680h

ours

////

////

////

////

////

////

////

//Ke

mp

1964

5°C

960ho

urs

////

////

////

////

////

////

////

//Ito

and

Stern196

715

°C288ho

urs

////

////

////

////

////

////

////

//Triticaletu

rgidum

:durum

Benn

etta

ndKaltsikes197

320

°C31h

ours

////

////

////

////

////

////

////

//Triticale:genotypeA(CS/K-TA

)Be

nnetta

ndSmith

197

220

°C21h

ours

////

////

////

////

////

////

////

//Triticale:genotypeB(CS/Pet-TA

)Be

nnetta

ndSmith

197

220

°C22h

ours

////

////

////

////

////

////

////

//Triticale:R

osner

Benn

etta

ndSmith

197

220

°C34or3

5ho

urs

////

////

////

////

////

////

////

//Triticumdiococcum

4X

Benn

etta

ndSmith

197

220

°C30h

ours

////

////

////

////

////

////

////

//

Triticumeastivum

xAe.M

utica

Benn

ett,Do

vera

ndRiley

1974

20°C

31h

ours

////

////

////

////

////

////

////

//Triticumeastivum

xSecalecereale

Benn

ett1

973

20°C

35h

ours

////

////

////

////

////

////

////

//15

°C43h

ours

////

////

////

////

////

////

////

//20

°C24h

ours

////

////

////

////

////

////

////

//25

°C18h

ours

15°C

45h

ours

////

////

////

////

////

////

////

//20

°C24or2

5ho

urs

////

////

////

////

////

////

////

//Triticumm

onococcum

Benn

etta

ndSmith

197

220

°C42h

ours

////

////

////

////

////

////

////

//Tulbag

hiaviolacea

Taylor195

320

°C130ho

urs

////

////

////

////

////

////

////

//Hs

uetal.19

88NOTGIVE

N11

9.1ho

urs

//43h

ours

31h

ours

12.2h

ours

7.1ho

urs

7.2ho

urs

4.4ho

urs

1.6ho

urs

1.6ho

urs

1.8ho

urs

0.4ho

urs

3.9ho

urs

2.1ho

urs

2.8ho

urs

//Yueta

l.19

9725

±1°C

meiosisII:5hou

rs//

////

////

////

////

2.5ho

urs

//Nen

nasa

tal201

6NOTGIVE

Nanaphases:12min

////

////

////

//12

.7±3.2m

in//

////

//11

±3.7m

in//

//

Zeamays

Benn

ette

tal197

1

Triticumeastivum

:Holdfast

Benn

ette

tal197

2

Triticumeastivum

:ChineseSpring

Secalecereale

Benn

ette

tal197

1

Trad

escantiapalud

osa

Trilliumerectum

Trad

escantiare

flexa

Endymionno

nscriptus

Wils

on1

959

Hordeumvulga

re:unspe

cifie

dvarie

ty

Lind

gren

eta

l.19

69NOTGIVEN

Hordeumvulga

re:M

orex

Higginse

tal.20

12

12.5h

ours

10h

ours

10.5h

ours

6ho

urs

46h

ours

20.5h

ours

10h

ours

5.5ho

urs

5ho

urs

15.5h

ours

69.5h

ours

12h

ours

48h

ours

1.5ho

urs

1ho

ur

5.30

%19

.90%

3daysafterthe

firsta

naylsedmaterialalltheanthersh

adm

icrospores-->allm

eiocyteste

rminated

meiosis.One

"spikeletu

nit"se

eEriksson

and

Ekberg19

65,islessthan16

hou

rs-->shorterstageslessthan1ho

ur

14h

ours

43h

ours

43h

ours

Table 1.1

19

Table1.2Du

ratio

nofm

eisoisinDycotiledo

nsand

gym

nosperms

DICO

TSplan

tspe

cie

publication

tempe

rature

meiosisdurationoverall

prem

eiosis

leptoten

ezygo

tene

pachyten

ediploten

ediakinesis

metap

haseIan

apha

seI

teloph

aseI

interkinesis

prop

haseII

metap

haseIIan

apha

seII

teloph

aseII

tetrad

sAlliaria

petiolata

Benn

ett1

973

NOTGIVE

N24h

ours

////

////

////

////

////

////

////

//An

thirriumm

ajus

Ernst1

938

NOTGIVE

N24

to34ho

urs

////

////

////

////

////

////

////

//Arab

idop

sisth

aliana

:Ws

Armstrongeta

l.20

0318

.5°-

20°C

33h

ours(2

4)9

6//

Arab

idop

sisth

aliana

:Col

-0Sanche

z-Moraneta

l.20

07NOTGIVE

N32h

ours(2

2)10

7//

//Arab

idop

sisth

aliana

:Ler

Stronghilletal.20

1421

°C29h

ours

(22)

75

610

1//

////

////

////

////

//Arab

idop

sisth

aliana

:Col-02X

thisthesis

21°C

35h

ours(2

6.5)

8.5

1.5

610

//Arab

idop

sisth

aliana

:Col-04X

thisthesis

21°C

51h

ours(3

2.5)

191.

57.

511

//Be

taVulga

risBe

nnett1

973

20°C

24h

ours

////

////

////

////

////

////

////

//Ca

psellabursa-pastoris

Benn

ett1

973

NOTGIVE

N18h

ours

////

////

////

////

////

////

////

//Ha

plop

appu

sgracilis

Marith

amu&Threlked

NOTGIVE

N24-36ho

urs

////

////

////

////

////

////

////

//Lycopersicum

esculentum

(Solan

umlycopersicum

)Be

nnett1

973

20°C

24-30ho

urs

////

////

////

////

////

////

////

//

Lycopersicum

peruvianu

mPaciniand

Cresti197

8NOTGIVE

Nprop

hase12ho

urs

////

////

////

////

////

Petuniahybrida

Izhara

ndFrankel197

315

-17°Cnight

/25-30

°Cday

16h

ours

22

13

12

Pisumsa

tivum

Benn

ett1

976

20°C

30h

ours

////

////

////

////

////

////

////

//Ve

ronicacha

maedrys

Benn

ett1

973

NOTGIVE

N20h

ours

////

////

////

////

////

////

////

//Viciafaba

Marqu

ardt195

1NOTGIVE

N72

to96ho

urs

////

////

////

////

////

////

////

//Viciasativa

Benn

ett1

976

20°C

24h

ours

////

////

////

////

////

////

////

//

GYM

NOSPER

Mplan

tspe

cie

publication

tempe

rature

meiosisdurationoverall

prem

eiosis

leptoten

ezygo

tene

pachyten

ediploten

ediakinesis

metap

haseIan

apha

seI

teloph

aseI

interkinesis

prop

haseII

metap

haseIIan

apha

seII

teloph

aseII

tetrad

sPinu

slaricio

Cham

berla

in193

5(rep

orted

inIzhare

ta.l19

73)

NOTGIVE

N

3mon

ths

////

////

////

////

////

////

////

//

1.5

6

15.3

2.7

12h

ours

41

21

3

31

14

12

31

Table 1.2

20

Introduction

21

Imagingofmeiosis1.3

1.3.1Livecellimagingsetups

Threemainsetupshavebeenusedsofartofollowmeiosislive:wide-field,confocal

andmulti-photonmicroscopy.

Wide-fieldmicroscopy,oftensupportedbydeconvolution,providesaneasy-to-

handlesystemtoobtaintime-lapsesandz-stacksofthedivision.Thepairingofwide-

fieldwithafluorescentlightsource(e.g.,UV-lamp)allowedtheemploymentofdyes

andfluorophoresfusedtoreporterstovisualizecellularandnuclearelements,e.g.,

telomeres and centromeres (Tomita and Cooper, 2007) or synaptonemal proteins

(Lee et al., 2015). As an example, the functions of telomere bouquet in budding

yeastS.pombehavebeenanalyzedbywide-fieldmicroscopy.Theseworksrevealed

its involvement in controlling the behavior of the microtubule-organizing center

(TomitaandCooper,2007),aswellas in creating a specialized sub-nuclearmicro-

environment that directs the assembly of meiotic centromeres (Klutstein et al.,

2015). Other works conducted both in yeast (Lee et al., 2012) and in isolated

mammaloocytes(Leeetal.,2015;Shibuyaetal.,2014)dissectedtherapidprophase

movementsofchromosomes,showingthattheyfollowdifferentdynamicsoverthe

meioticdivisionandthattheyareresponsiblefortheformationofcorrectsynapsis

andrecombinationevents.

While for single cell imaging (unicellularorganisms,or isolatedmeiocytes) the

wide-fieldmicroscope is a good option, formore complex samples confocal laser

scanningmicroscopy(CLSM)ismoreadequate.Bysettingupapinholeinfrontofthe

detector,thesignalfromtheoff-focalplanecanbefiltered,restitutingimageswitha

highsignal/noiseratio.Thisallowstheobservationofthickerspecimenthatcouldbe

scanned a series of optical sections. Consequently, confocalmicroscopy has been

successfullyapplied to studyhomologpairing inS.pombe (Chacónet al.,2016),C.

elegans (Rog and Dernburg, 2015; Wynne et al., 2012), Drosophila melanogaster

(Christophorouetal.,2015),andmammalianoocytesandspermatocytes,whichcan

be visualized ex vivo within cultured embryonic ovaries and tubules (Enguita-

Marruedoetal.,2018).Likewise,chromosomesegregation inmammaloocyteshas

been analyzed by confocal microscopy: kinetochores could be tracked for over 8

Introduction

22

hours,revealingthatthebi-orientedattachmentofhomologs isestablishedaftera

lengthytry-and-errorprocess(Kitajimaetal.,2011);microtubulesorganizingcenters

andactinelementsofthecytoskeletonhavebeenshowntoberelevantforspindle

formation and correct segregation (Schuh and Ellenberg, 2007; Holubcová et al.,

2013;MogessieandSchuh,2017),aswellasitwasconfirmedbylivecellimagingof

fetalmouse oocytes that cohesin establishment ismaintainedwithout detectable

turnoverandthatitslossinolderoocytesremainsuncorrected,leadingtoformation

ofaneuploidandnon-viablegametes(Burkhardtetal.,2016).

A further advantage of confocal microscopy is the usage of lasers as a light

source,allowingthepreciseselectionofexcitationwavelength.Thisopenedtheway

toproceduressuchasFRAP,aspresented in thestudyofGigantetal.Byapplying

photobleachingtothecytoskeletonreporterGFP:NMY2,theywereabletodetecta

change inthespindledynamicsofoocyteswhichcarriedamutation inthekinesis-

13,provingitsinvolvementintheformationofmeioticspindlesofC.elegans(Gigant

etal.,2017).

At last, two-photonmicroscopyhasbeenused to imagemeiosis inC. elegans.

Two-photonmicrosocpyuses infrared lightasexcitationsource,whichallowsdeep

penetration in the tissues. Coupling two-photon technology with the FLIM/FRET

techniqueLlèresetal.,wereabletovisualizeatananoscalelevelthecompactionof

prophasechromosomeswithinC. elegansovaries,and to link its regulation to the

actionofcondensinIandII(Llèresetal.,2017).

1.3.2Livecellimagingofplantmeiosis

Incontrasttothestudyofmeiosisinotherorganisms,researchinplantsisonlyinits

infancytoexplorethepowerof live imaging.Sofar,onlyfivestudiesarepublished

that employ two different approaches to describe chromosome movements and

MTsrearrangementsinmaizemeiocytes(Yuetal.,1997;Nannasetal.,2016;Higgins

etal.,2016;SheehanandPawlowski,2009)aswellaschromatinreprogramming in

Arabidopsisthaliana(Ingouffetal.,2017).

ThesystemdevelopedbyYuetal.1997,re-adaptedintheworksfromNannaset

al.,2016andHigginsetal.,2016, isbasedonwide-field fluorescentmicroscopyof

isolatedmaizemeiocytes;thecellsarecultured in liquidmedium,whichoffersthe

Introduction

23

advantageofeasy treatmentwithdyesas Syto12 tomark chromosomeswhereas

other cellularelements couldbe visualizedwith fluorescent reporters such as the

fusionproteinCFP:β-TUB1formicrotubules.

Whilethissetupiseasilyapplicable,itsusageisrestrictedtothestudyofshort

meioticphases such asmetaphase and anaphase:meiocytes couldbemaintained

aliveforamaximumof9hours(Yuetal.,1997)andwereimagedoverperiodsof80

minutes or less (Nannas et al., 2016), failing to restitute information about the

longer prophase. Nonetheless, important knowledge could be gained about the

regulationofmeioticspindles,whichcouldnotberevealedby fixedspecimen.For

example, theworkofNannasdescribed theexistenceofasymmetricalanaphases,

which correct off-center positioning of the spindles in anaphase I and II, and the

appearanceofphragmoplastequidistant from the chromosomes insteadof in the

spindlemid-zone,providingabackupsystemforfailureincompletingchromosomes

segregation(Nannasetal.,2016).

The secondapproach isbasedonmultiphotonmicroscopy.Exploiting itsgreat

focusdepth,whichreaches200µm,meiocytescanbe imagedwithouttheneedof

isolation. This set up has been successfully applied to maize anthers cultured in

liquid medium (Sheehan and Pawlowski, 2009) and on Arabidopsis thaliana

inflorescences,embedded insolidmediumanddissectedwithavibratome(Ingouff

etal.,2017).Samplescouldbemaintainedaliveforperiodslongerthan30hoursand

imaged for 24 hours (maize anthers in Sheenan and Pawlowski, 2009, no time

indications for Ingouffet al.,2017).SheenanandPawlowskiwereable toobserve

and analyze chromosome movements similar to the one described for yeast, C.

elegans,andmammals,revealingthepresenceofdifferentdynamicscharacterizing

zygotene and pachytene stages (Sheehan and Pawlowski, 2009). Ingouff et al.,

instead, were interested in investigating chromatin reprogramming during

Arabidopsis thaliana reproduction, and revealed that methylation levels are very

stableexceptforasignificantdecreaseofthesignaluponeggcellmaturation.Since

intheirstudy Ingouffetal.,aimtofollowthecompletesexualdevelopmentofthe

plant,meiosiswasconsidered asingleunit,withoutdistinctionamongsub-phases,

andthereforetheirresolutionofthecelldivisionwasminimal.

Objectives

25

2 ObjectivesOverthelastyears,thestudyofbiologicalprocesseshasbeenincrediblyfosteredby

livecell imaging,whichdisclosed thecomplexdynamicsunderlyingevents suchas

cellproliferation,patternformation,andcelldeath.Differentlythaninothertopics,

inthefieldofmeiosisplantshaslackedbehindyeastorotheranimals,countingonly

ahandfulnumberofpublicationsusinglivecellimagingapproaches(Yuetal.,1997;

SheehanandPawlowski,2009;Nannasetal.,2016;Higginsetal.,2016; Ingouffet

al.,2017).

As aconsequence, thedescriptionof thedynamicsof themeioticdivisionhas

beenrestrictedtotheapplicationofcytochemicalmethodssuchascellspreadsand

immunolocalization,whicharebasedonfixedmaterial.Whilethesetechniqueshave

been and continue tobe, very informative, theydidnot allow fully capturing the

natureofmeiosis,characterizedbyspecificchromosomemovementsduringparing

andsegregation,orbythefastdynamicsofproteinre-location.

Thefirstaimofthisstudywas,therefore,theestablishmentofalivecellimaging

technique to follow theentiremeioticdivision in anthersofArabidopsis thaliana.

The technique should fulfill three main requisites: long-time imaging (and hence

maintenance of sample viability for a long time) to follow the complete division,

chromosomal resolution in imaging to distinguish chromosomes and cellular

structures,and finally simplicity in itsexecution tomake themethodavailable for

otherresearchers.

Secondly, a system to unequivocally describe the images and allow for a

quantitativedescriptionoftheobtaineddatashouldbedeveloped.

Finally, Iwas interested in the application of the newmethod, paired to the

analysissetup,toperformacomparisonbetweenthetimecourseofmalemeiosisin

wildtypeArabidopsisthaliana,indiploidandtetraploidpopulations.

Results

27

3 Results

Techniqueestablishment3.1

Live cell imaging of plants benefited greatly from CLSM application; for example

confocalmicroscopyhasbeenusedtostudymitosisandcelldifferentiation inroot

apicalmeristem (RAM) (e.g., in Komaki and Schnittger, 2017) and in shoot apical

meristem (SAM) (e.g., in Hamant et al., 2014; Gruel et al., 2016) of Arabidopsis

thaliana.Conversely, ithasnotbeenappliedsofartoobserveplantmeiosis.Inthe

firstchapterofthe“Results”sectionanewmethodisintroduced.

3.1.1Sampleisolationandmounting

Theselectionandpreparationofoptimalmaterialareofkeyimportancetoperform

live imaging. To facilitate the handling of the sample the whole procedure was

performed under a dissection microscope with a magnification of 4X. An

inflorescencewascutfromafivetosixweeksoldplantandlaiddownonasupport

of1%agarosedissolvedinMilliQwater.Underourgrowthconditions(Materialand

Methodssection6.1),wildtype-likeflowerbudsundergoingmeiosisare0.3-0.5mm

longandpresentaroundshape(Figure3.1);thereforeallflowerslargerthan0.5mm

wereremovedatthepedicelwiththeuseoftweezers(Figure3.1A).

CLSM has a typical penetration depth of 70-100 μm, which allows imaging

throughthe fourcell layersthatenwrapthepollenmothercells (PMCs)withinthe

anther, but not to penentrate the sepals. Thus, to obtain clear images of male

meiocytes,itwasnecessarytoremovetheuppermostsepaloftheflowerbud;inthis

way, two of the six anthers are exposed and directly accessible to the objective

(Figure3.1C1andC2).After the sepal removal, the innerorganizationof the floral

organs is disclosed, giving a further hint about the staging: in flower primordia

undergoing meiosis, petals are visible, but they do not overlap with the anthers

which in turn have the same length of the gynoecium (Figure 3.1C2). This

developmental stage corresponds to stage 9 in the description from Smyth et al.

1990(Smythetal.,1990).

Results

28

Conversely,allthesmallflowers,presumablynotcontaininganymeiocytesyet,

wereclippedaway(Figure3.1B)toobtainasinglebudattachedtoafewmillimeters

of its stem. The sample was transferred and anchored onto a small petri dish

(diameterof35mm)filledwithApexCultureMedium(ACM)(Hamantetal.,2014).

Further stabilization was obtained using a drop of 2% agarose in MilliQ water

distributedaroundtheflowerhead.

Iftheflowerbudwasnotatthecorrectstage,eitheroneoftwostrategieswas

followed:when the flowerpresented amoreadvanceddevelopmental stage (e.g.,

long petals), it was clipped, and the very next flower of the inflorescence was

isolated;when the flowerwas tooyoung itwaspossible tomount itonACM,seal

thepetridishandletitgrowatthesamegrowthconditionsofthemotherplantuntil

thecorrectdevelopmentalstagewasreached.

ToassessthesampleviabilityonACM,flowerbudswereisolatedandmounted

on themediumas inpreparation for imaging,with the removalof theuppermost

sepal. The petri dish containing the sampleswere sealed and repositioned in the

same growth chamber as the mother plant, to assure the maintenance of the

environmentalconditions.Flowerbudgrowthwasmonitoredthroughoutoneweek

(Figure3.1D).

Theorgansoftheflowerspresentedadevelopmentfromstage9,(Figure3.1D,

DAY0)tostage15(Figure3.1D,DAY7),withinasimilartimeframe(7daysversus

6.5days)aspreviouslypublishedinSmythetal.1990(Smythetal.,1990).

Results

29

3.1.2Microscopesetup

UprightCLSMZEISS780and880wereequippedwith40Xwaterdippingobjectives,

whichcanbesubmergeddirectlyintowater;thusbypassingtheuseofcoverglasses.

Thissetupallowsmountingthesampleonsolidmediuminasmallpetridish,as

described in 3.1.1, leaving it free to grow and expand without constraint in the

vertical direction, but at the same time offering an anchoring system to avoid

flotation.

Othermagnificationlensessuchas20Xand63Xweretestedforthepurpose,but

the 40X objective was finally selected since I could visualize the entire flower

structurewithoutsacrificingresolutionwhenzoomingontosinglemeiocytes(Figure

3.2 B1). Once ready, the petri dish containing the sample was positioned on the

microscope stage and stabilized with double-sided tape. The petri dish was then

Figure3.1Sampleisolationandviability

Thethreemainstepsofsamplepreparationareillustrated:A)Atfirst,an inflorescence isdissected.Alltheolderflowersareremoved,whiletheyoungerflowerbudsarekeptattachedtothestem.Thewhitearrowheadindicatesaflowerbudatameioticstage.B) After the removal of the uppermost sepal and the validation of the correct developmental stage of theselectedflower,allthesmallerflowerbudsareclippedaway.C1)ThesampleisanchoredintoACMwiththeexposedanthersfacingup.C2)Magnificationof the flower bud inC2); thedifferent floralorgansarehighlighted: inblue sepals, inwhitepetals,inyellowanthersandinpinkthegynoecium.D)MonitoringofflowergrowthanddevelopmentonACM.The3flowerspresentanormalgrowthoveraperiodof7days:thefloralorgansandthestemelongate inaconstantmannerfromday0today7,confirmingtheirviabilityonthemedium.

Results

30

filledwithautoclavedwater,and left inposition for aperiodofminimumhalfan

hour,toensurethatthemediumwouldabsorbthewater,andavoidlateronfurther

movementsduetomediumswallowing.

Such amicroscope set up applies to several investigations, and therefore the

settings for imageacquisition couldbeadjustedaccording towith thepurpose. In

thecontextofthisdissertation, Iappliedthetechniquetoperformtimecoursesof

the entire meiotic division, on flower buds expressing GFP and TagRFP fusion

proteins; thus themain concernwas to balance image resolution,with temporal

resolutionandsampleupkeep.

Imageswereacquiredasaseriesofz-stack.The intervaltimewassetbetween

minimum3andmaximum15minutes,dependingon theobservedmeioticphase.

Thez-stackswerecomposedbysixfocalplanes,with50μmdistance.Thisintervalin

the z-dimension allowed the buffering of small verticalmovementof the sample,

assuring that the same meiocytes would be captured during the whole data

acquisition.Whenmorethanflowerbudwasmountedonthesamepetridish,their

differentpositionsweresavedusingthemulti-positionfunctionofZENsoftware,and

eachofthemwasautomaticallyre-focusedateachtimepoint,usingtheauto-focus

functionbasedonfluorescence.

Argon laser (λ488)andDPSS561-10 laser (λ561)wereusedas the sourceof

excitationwavelengthsforGFPandTagRFPrespectively;theirintensitywasadjusted

accordingtowiththesampleemission,butingeneral,wasneverexceeding10%for

theArgonlaserand4%fortheDPSS561-10laser.Itwasessentialtokeepitaslowas

possible to not compromise the sample viability. The emitted signalswere firstly

filtered through theBeamsplitterMBS488/561.Greenandred fluorescencewere

recorded intwochannelsbysequential linemode filteredrespectively for498-550

nm and 578-649 nm; an additional third channel was used to collect the auto-

fluorescenceofchloroplasts,andfilteredfor680-750nm.Thepinholewassetat1

Airy Unit, and scan time did not exceed 0.7 μsec pixel dwell. The bidirectional

function was on, and averaging was performed on two lines. Images were

1024x1024pixels.

Results

31

Selectionofreporterlines3.2

Atypicalsetupfor livecell imagingofcelldivisions isconcomitantuseofamarker

for chromatin, such as histone fusion protein, with a marker that highlights

microtubulestomonitorchromosomeandspindlebehavior(twoexamplestofollow

plantmeiosiscanbefoundinPeirsonetal.,1997;Nannasetal.,2016).Additionally,

Iwas interested in ameiotic-specificmarker, to unequivocally identifymeiocytes

evenatearlystagesofmeiosiswhentheirsizeandshapearenotyetverydistinct.

Iidentifiedasagoodcandidateamongthereportersavailableinourlaboratory

theGFP fusion to REC8 PROREC8:REC8:mEGFP, generated byDr. Shinichiro Komaki

(Prusickietal.,2018).

REC8 accumulation has been detected by immunolocalization studies starting

frompre-meioticS-phaseuntiltheonsetofanaphaseI(Caietal.,2003).Moreover,

seenitsroleasα-kleisinsubunitinthemeioticcohesincomplex,itlocalizesalongthe

entire length of the chromosomes during prophase I, allowing the detection of

chromosomedynamicsandsynapsisstateuntiltheendofmetaphaseI.

Figure3.2MicroscopesetupExampleofmicroscopesetupandsamplevisualization.

A) Thesample(ingreen)ismountedinasmallpetridishcontainingACM.Thepetridishisfilledupwithdistilledwaterandtheobjectiveisimmerseddirectlyintothewater.

B1)OverviewofaflowerbudcarryingTUB4-RFPreporter,underthe40Xwaterdippingobjective.B2)IdentificationoffloralorgansoftheflowerbudinB1:inbluesepals,inwhiteapetal,inpinkthetipofthegynoeciumandinyellow2anthers.Withintheantherspollensacsarehighlightedinred.C)ZoominontheanthersofB,onadifferentfocalplane.Meiocytesoccupytheinnerpartoftheantherandarerecognizablebytheirregularshapeandthebigcentralnucleus.

Results

32

Therefore I decided to utilize PROREC8:REC8:mEGFP as amarker for chromatin

duringmeiosis.Ifoundthatthereporterlineonlyaccumulatesinmeiocytesandits

localization pattern is consistent with the previous description (Cai et al.,

2003)(Figure3.3A).

Moreover, the REC8 reporter allowed us to estimate the sensitivity of the

imagingprocedure. WhileREC8 isremoved fromchromosomesarmsattheendof

meiosis I to allow the resolution of cross-overs, a small fraction remains at the

centromerestomaintainsisterchromatidcohesion(Yuanetal.,2018).Thedetection

of the centromeric fraction of REC8 has been challenging by immunolocalization

studies(Caietal.,2003;Yuanetal.,2018).WhenIfollowedthefirstmeioticdivision,

I observed the remaining REC8:GFP at centromeres indicating that the here

presentedlivecellimagingsystemishighlysensitive(Figure3.3B).

Ascytoskeletonmarker,IselectedafusionproteinoftheTUBULINβSUBUNIT4

(TUB4)withTagRFPexpressedundertheRPS5Apromoter (PRORPS5A:TagRFP:TUB4),

generatedandkindlyprovidedbyDr.Takashi Ishida(KumamotoUniversity,Japan).

TheexpressionofPRORPS5A:TagRFP:TUB4 isnotcell-specific,andthereforenotonly

allows a straightforward recognition of meiotic phases such as metaphase and

anaphase, but permits as well the observation of the complete structure of the

anther,andconsequentlythestudyofthebehaviorofothercell layersconstituting

thepollensac,e.g.,thetapetum.

Results

33

3.2.1 Functionality of the PROREC8:REC8:mEGFP and the

PRORPS5A:TagRFP:TUB4reporter

ToassurethatthefusionproteinPROREC8:REC8:mEGFPwasfunctional,Iperformeda

rescue assay.Homozygous plants of the REC8 T-DNA insertion line SAIL_807_B08

(rec8-/-)presentanormalvegetativegrowth,butarecompletelysterile,withshort

siliquesandpollenandseedabortionratesatalmost100%(Figure3.4A-E).Aclose

Figure3.3ExpressionpatternofPROREC8:REC8:mEGFPandPRORPSA5:TagRFP:TUB4intheKINGBIRDline

A) The figuredepicts the localizationofPROREC8:REC8:mEGFPandPRORPSA5:TagRFP:TUB4over thecourseofentiremeiosisinafloweroftheKINGBIRDline.A1:premeiosis;A2:leptotene;A3:zygotene;A4:diploteneinthe lower anther and metaphase I in the upper anther; A5: telophase I in the lower anther and lateprophaseII-metaphaseIItransitionintheupperanther;A6:tetrads.

B) REC8-GFP localizationaftermetaphaseI(B1) inaPROREC8:REC8:mEGFP-onlyplant.ThewhitearrowheadsinB2andB3indicatecentromericREC8.

Results

34

look tomeioticprogression revealsall the typicalphenotypespreviouslydescribed

for rec8 homozygous plants: irregular chromosome condensation and pairing at

prophase I, chromosome fragmentation,presenceofunivalents,and chromosome

mis-segregationresulting intheformationofunbalancedgametes,micronucleiand

polyads (Figure 3.4 F, Bai et al., 1999). The introgression of PROREC8:REC8:mEGFP

construct into rec8 -/- background fully restored fertility, with the plant growing

elongatedsiliques,andhavingasimilarlevelofpollenandseedproductionaswild-

typeplants(figure3.4A-E).CellspreadsofPROREC8:REC8:mEGFP